Draft Regulatory Impact Analysis:

           Proposed Rulemaking for 2017-2025
           Light-Duty Vehicle Greenhouse Gas
           Emission Standards and Corporate
           Average Fuel Economy Standards
&EPA
United States
Environmental Protection
Agency

-------
                Draft Regulatory Impact Analysis:

                Proposed Rulemaking for 2017-2025
                Light-Duty Vehicle Greenhouse Gas
                 Emission Standards and Corporate
                 Average Fuel Economy Standards
                         Assessment and Standards Division
                        Office of Transportation and Air Quality
                        U.S. Environmental Protection Agency
SEPA
United States
Environmental Protection
Agency
EPA-420-D-11-004
November 2011

-------
                            TABLE OF CONTENTS
EXECUTIVE SUMMARY	V
1    TECHNOLOGY PACKAGES, COST AND EFFECTIVENESS	1-1
1.1   Overview of Technology	1-1
1.2   Technology Cost and Effectiveness	1-4
1.3   Vehicle Package Cost and Effectiveness	1-15
1.4   Use of the Lumped Parameter Approach in Determining Package Effectiveness...1-
45
  1.4.1     Background	1-45
  1.4.2     Role of the model	1-46
  1.4.3     Overview of the lumped parameter model	1-46
1.5   Lumped Parameter Model Methodology	1-49
  1.5.1     Changes to the LP model for the proposed rulemaking	1-49
  1.5.2     Development of the  model	1-50
  1.5.3     Baseline loss categories	1-50
  1.5.4     Baseline fuel efficiency by vehicle class	1-52
  1.5.5     Identification and calibration of individual technologies	1-54
  1.5.6     Example build-up of LP package	1-56
  1.5.7     Calibration of LP results to vehicle simulation results	1-59
  1.5.8     Notable differences  between LP model and Ricardo results	1-63
  1.5.9     Comparison of results to real-world examples	1-65
2    EPA'S VEHICLE SIMULATION TOOL	2-1
2.1   Introduction	2-1
  2.1.1     Background	2-1
2.2   Descriptions of EPA's Vehicle Simulation Tool	2-2
  2.2.1     Overall Architecture	2-2
  2.2.2     System Models	2-5
2.3   Applications of Simulation Tool for the Proposed Rule	2-7
  2.3.2     Off-Cycle Credit Calculation	2-11
2.4   On-Going and Future Work	2-14

-------
   2.4.1    Simulation Tool Validation	2-14
   2.4.2    Simulation Tool Upgrade	2-15
3    RESULTS OF PROPOSED AND ALTERNATIVE STANDARDS	3-1
3.1  Introduction	3-1
3.2  OMEGA model overview	3-1
3.3  OMEGA Model Structure	3-4
3.4  Model Inputs	3-5
   3.4.1    Market  Data	3-5
   3.4.2    Technology Data	3-11
3.5  The Scenario File	3-14
   3.5.1    Reference Scenario	3-14
   3.5.2    Control Scenarios	3-21
3.6  Fuels and reference data	3-22
3.7  OMEGA model calculations	3-22
3.8  Analysis Results	3-24
   3.8.1    Targets  and Achieved Values	3-24
   3.8.1    Penetration of Selected Technologies	3-36
   3.8.2    Projected Technology Penetrations in Reference Case	3-38
   3.8.3    Projected Technology Penetrations in Proposal case	3-44
   3.8.4    Projected Technology Penetrations in Alternative Cases	3-50
   3.8.5    Additional Detail on Mass Reduction Technology	3-74
   3.8.6    Air Conditioning Cost	3-74
   3.8.7    Stranded Capital	3-75
3.9  Per Vehicle Costs 2021 and 2025	3-78
3.10   Alternative Program Stringencies	3-81
3.11   Comparative cost of advanced technologies under credit scenarios	3-84
3.12   How Many of Today's Vehicles Can Meet or Surpass the Proposed MY 2017-
2025 CO2 Footprint-based Targets with Current Powertrain Designs?	3-85
3.13   Analysis of Ferrari & Chrysler/Fiat	3-91
3.14   Cost Sensitivities	3-91
   3.14.1   Overview	3-91
   3.14.2   Mass Sensitivity	3-93
   3.14.3   Battery  Sensitivity	3-94
                                        11

-------
  3.14.4  ICM Sensitivity	3-95
  3.14.5  Learning Rate Sensitivity	3-97
  3.14.6  Summary of Sensitivity Impacts	3-97
  3.14.7  NAS report	3-98
4    PROJECTED IMPACTS ON EMISSIONS, FUEL CONSUMPTION, AND
SAFETY	4-1
4.1   Introduction	4-1
4.2   Analytic Tools Used	4-2
4.3   Inputs to the emissions analysis	4-2
  4.3.1    Methods	4-2
  4.3.2    Activity	4-4
  4.3.3    Scenarios	4-12
  4.3.4    Emission Results	4-21
  4.3.5    Fuel Consumption Impacts	4-25
  4.3.6    GHG and Fuel Consumption Impacts from Alternatives	4-26
4.4   Safety Analysis	4-27
4.5   Sensitivity Cases	4-28
  4.5.1    Rebound	4-28
  4.5.2    EV impacts	4-30
4.6   Calculation of Impacts from An Electric Vehicle	4-31
5    VEHICLE PROGRAM COSTS AND FUEL SAVINGS	5-1
5.1   Costs per Vehicle	5-1
5.2   Annual  Costs of the Proposed National Program	5-8
5.3   Cost per Ton  of Emissions Reduced	5-9
5.4   Reduction in Fuel Consumption and its Impacts	5-10
  5.4.1    What Are the Projected Changes in Fuel Consumption?	5-10
  5.4.2    What are the Fuel Savings to the Consumer?	5-11
5.5   Consumer Payback Period and Lifetime Savings on New Vehicle Purchases	5-13
6    HEALTH AND ENVIRONMENTAL IMPACTS	6-1
6.1   Health and Environmental Impacts of Non-GHG Pollutants	6-1
  6.1.1    Health Effects Associated with Exposure to Non-GHG Pollutants	6-1
  6.1.2    Environmental Effects Associated with Exposure to Non-GHG Pollutants ..6-
  15
                                       in

-------
6.2   Air Quality Impacts of Non-GHG Pollutants	6-27
  6.2.1    Current Levels of Non-GHG Pollutants	6-27
  6.2.2    Impacts on Future Air Quality	6-29
6.3   Quantified and Monetized Non-GHG Health and Environmental Impacts	6-32
  6.3.1    Economic Value of Reductions in Criteria Pollutants	6-33
  6.3.2    Human Health and Environmental Benefits for the Final Rule	6-37
6.4   Changes in Atmospheric CO2 Concentrations, Global Mean Temperature, Sea
Level Rise, and Ocean pH Associated with the Proposed Rule's GHG Emissions
Reductions	6-43
  6.4.1    Introduction	6-43
  6.4.2    Projected Change in Atmospheric CO2 Concentrations, Global Mean
  Surface Temperature and Sea Level Rise	6-45
  6.4.3    Projected Change in Ocean pH	6-50
  6.4.4    Summary of Climate Analyses	6-52
7    OTHER ECONOMIC AND SOCIAL IMPACTS	7-1
7.1   Monetized CO2 Estimates	7-3
7.2   Summary of Costs and Benefits	7-10
8    VEHICLE SALES AND EMPLOYMENT IMPACTS	8-1
8.1   Vehicle Sales Impacts	8-1
  8.1.1    Vehicle Sales Impacts and Payback Period	8-1
  8.1.2    Consumer Vehicle Choice Modeling	8-4
8.2   Employment Impacts	8-17
  8.2.1    Introduction	8-17
  8.2.2    Approaches to Quantitative Employment Analysis	8-19
  8.2.3    Employment analysis of this proposal	8-23
  8.2.4    Effects on Employment for Auto Dealers	8-29
  8.2.5    Effects on Employment in the  Auto Parts Sector	8-30
  8.2.6    Effects on Employment for Fuel Suppliers	8-30
  8.2.7    Effects on Employment due to Impacts on Consumer Expenditures	8-30
  8.2.8    Summary	8-31
9    SMALL BUSINESS FLEXIBILITY ANALYSIS	9-1
                                      IV

-------
                                 Executive Summary
       The Environmental Protection Agency (EPA) and the National Highway Traffic
Safety Administration (NHTSA) are issuing a joint notice of proposed rulemaking (NPRM) to
establish standards for light-duty highway vehicles that will reduce greenhouse gas emissions
(GHG) and improve fuel economy. EPA is proposing greenhouse gas emissions standards
under the Clean Air Act, and NHTSA is proposing Corporate Average Fuel Economy
standards under the Energy Policy and Conservation Act (EPCA), as amended.  These
proposed standards apply to passenger cars, light-duty trucks, and medium-duty passenger
vehicles, covering model years (MY) 2017 through 2025. The proposed standards will require
these vehicles to meet an estimated combined average emissions level of 163 grams of CO2
per mile in MY 2025 under EPA's proposed GHG program. These proposed standards are
designed such that  compliance can be achieved with a single national vehicle fleet whose
emissions and fuel  economy performance improves year over year.  The proposed National
Program will result in approximately 1,967 million metric tons of CC>2 equivalent emission
reductions and approximately 3.9 billion barrels of oil savings over the lifetime of vehicles
sold  in model years 2017 through 2025.

       Mobile sources are significant contributors to air pollutant emissions (both GHG and
non-GHG) across the country, internationally, and into the future. The Agency has
determined that these emissions cause or contribute to air pollution which may reasonably be
anticipated to endanger public health or welfare, and is therefore establishing standards to
control these emissions as required by section 202 (a) of the Clean Air Act.A The health- and
environmentally-related effects associated with these emissions are a classic example of an
externality-related market failure.  An externality occurs when one party's actions impose
uncompensated costs on another party. EPA's NPRM rule will deliver additional
environmental and  energy benefits, as well as cost savings, on a nationwide basis that would
likely not be available if the rule were not in place.

       Table 1 shows EPA's estimated lifetime discounted cost, benefits and net benefits  for
all vehicles projected to be sold in model years 2017-2025.  It is important to note that there is
significant overlap  in costs and benefits for NHTSA's CAFE program and EPA's GHG
program  and therefore combined program costs and benefits are not a sum of the individual
programs.
A "Technical Support Document for Endangerment and Cause or Contribute Findings for Greenhouse Gases
Under Section 202(a) of the Clean Air Act" Docket: EPA-HQ-OAR-2010-0799,
http://epa.gov/climatechange/endangerment.html. See also State of Massachusetts v. EPA. 549 U.S. 497. 533
("If EPA makes a finding of endangerment, the Clean Air Act requires the agency to regulate emissions of the
deleterious pollutant from new motor vehicles").

-------
Table 1 EPA's Estimated 2017-2025 Model Year Lifetime Discounted Costs,
       Benefits, and Net Benefits assuming the $22/ton SCC Valuea'b'c'd
                             (Billions of 2009 dollars)
Lifetime Present Value0 - 3% Discount Rate
Program Costs
Fuel Savings
Benefits
Net Benefits'1
$140
$444
$117
$421
Annualized Value6 - 3% Discount Rate
Annualized costs
Annualized fuel savings
Annualized benefits
Net benefits
$6.43
$20.3
$5.36
$19.3
Lifetime Present Value0 - 7% Discount Rate
Program Costs
Fuel Savings
Benefits
Net Benefits'1
$138
$347
$101
$311
Annualized Value6 - 7% Discount Rate
Annualized costs
Annualized fuel savings
Annualized benefits
Net benefits
$10.64
$26.7
$6.35
$22.4
      Notes:
      " The agencies estimated the benefits associated with four different values
      of a one ton CO2 reduction (model average at 2.5% discount rate, 3%, and
      5%; 95th percentile at 3%), which each increase over time.  For the
      purposes of this overview presentation of estimated costs and benefits,
      however, we are showing the benefits associated with the marginal value
      deemed to be central by the interagency working group on this topic: the
      model average at 3% discount rate, in 2009 dollars.  Section III.H provides
      a complete list of values for the 4 estimates.
      * Note that net present value of reduced GHG emissions is calculated
      differently than other benefits. The same discount rate used to discount the
      value of damages from future emissions (SCC at 5, 3, and 2.5 percent) is
      used to calculate net present value of SCC for internal consistency. Refer
      to Section III.H for more detail
      c Present value is the total, aggregated amount that a series of monetized
      costs or benefits that occur over time is worth in a given year.  For this
      analysis, lifetime present values are calculated for the first year of each
      model year for MYs 2017-2025 (in year 2009 dollar terms). The lifetime
      present values shown here are the present values of each MY in its first
      year summed across MYs.
      d Net benefits reflect the fuel savings plus benefits minus costs.
      e The annualized value is the constant annual value through a given time period (the
      lifetime of each MY in this analysis) whose summed present value equals the
      present value from which it was derived. Annualized SCC values are calculated
      using the same rate as that used to determine the  SCC value while all other costs
      and benefits are annualized at either 3% or 7%.
                                         VI

-------
        This draft Regulatory Impact Analysis (DRIA) contains supporting documentation to
the EPA rulemaking. NHTSA has prepared their own preliminary RIA (PRIA) in support of
their rulemaking (this can be found in NHTS A's docket for the rulemaking, NHTSA-2010-
0131). While the two rulemakings are similar, there are also differences in the analyses that
require separate discussion.  This is largely because EPA and NHTSA act under different
statutes. EPA's authority comes under the Clean Air Act, and NHTS A's authority comes
under EPCA and EISA, and each statute has somewhat different requirements and
flexibilities. As a result, each  agency has followed a unique approach where warranted by
these differences. Where each agency has followed the same approach—e.g., development of
technology costs and effectiveness—the supporting documentation is contained in the draft
joint Technical Support Document (draft joint TSD can be found in EPA's docket EPA-HQ-
OAR-2010-0799).  Therefore,  this DRIA should be viewed as a companion document to the
draft Joint TSD and the two documents together provide the details of EPA's technical
analysis in support of its rulemaking.
This document contains the following;

Chapter 1: Technology Packages, Cost and Effectiveness.  The details of the vehicle
technology costs and packages used as inputs to EPA's Optimization Model for Emissions of
Greenhouse gases from Automobiles (OMEGA) are presented. These vehicle packages
represent potential ways of meeting the CO2 stringency established by this rule and are based
on the technology costs and effectiveness analyses discussed in Chapter 3 of the draft Joint
TSD. This chapter also contains details on the lumped parameter model, which is a major
part of EPA's determination of the effectiveness of these packages. More detail on the
effectiveness of technologies and the Lumped Parameter model can be found in Chapter 3 of
the draft Joint TSD.

Chapter 2: The development and application of the EPA vehicle simulation tool are
discussed. This chapter first provides a detailed description of the simulation  tool including
overall architecture, systems, and components of the vehicle simulation model. The chapter
also describes applications and results of the vehicle simulation runs for estimating impact of
A/C usage on fuel consumption and calculating off-cycle credits particularly for active
aerodynamic technologies. For the result of the A/C study, the impact of A/C usage was
estimated at 11.9 CO2 g/mile for cars and 17.2 CO2 g/mile for trucks.  This  corresponds to an
impact of approximately 14.0 CO2 g/mile for the (2012) fleet, which is comparable to the
2012-2016 final rule result. For the off-cycle credits, EPA based its analysis on manufacturer
data, where active grill shutters (one of the active aerodynamic technologies considered)
provide a reduction of 0-5% in aerodynamic drag (Cd) when deployed.  EPA expects that
most other active aerodynamic technologies will provide a reduction of drag in the same
range as active grill  shutters.  Based on this analysis, EPA will provide a credit for active
aerodynamic technologies that can demonstrate a reduction in aerodynamic drag  of 3% or
more. The credit will be 0.6 g/mile for cars and 1.0 g/mile for trucks when  the reduction in
aerodynamic drag is around 3%.
                                         vn

-------
       ChapterS: This chapter provides the methodology from and results of the technical
assessment of the future vehicle scenarios presented in this proposal.  As in the analysis of
the MY 2012-2016 rulemaking, evaluating these scenarios included identifying potentially
available technologies and assessing their effectiveness, cost, and impact on relevant aspects
of vehicle performance and utility. The wide number of technologies which are available and
likely to be used in combination required a method to account for their combined cost and
effectiveness, as well as estimates of their availability to be applied to vehicles. These topics
are discussed.

    Chapter 4: This chapter documents EPA's analysis of the emission, fuel consumption
and safety impacts of the proposed emission standards for light duty vehicles. This proposal,
if finalized, will significantly decrease the magnitude of greenhouse gas emissions from light
duty vehicles.  Because of anticipated changes to driving behavior, fuel production, and
electricity generation, a number of co-pollutants would also be affected by this  proposed rule.
This analysis quantifies the proposed program's impacts on the greenhouse gases (GHGs)
carbon dioxide (CO2), methane (CH4), nitrous oxide (N2O) and hydrofluorocarbons (HFC-
134a); program impacts on "criteria" air pollutants, including carbon monoxide (CO), fine
particulate matter (PM2.5) and sulfur dioxide (SOx) and the ozone precursors hydrocarbons
(VOC) and oxides of nitrogen (NOx); and impacts on several air toxics including benzene,
1,3-butadiene, formaldehyde, acetaldehyde, and acrolein.

       CO2 emissions from automobiles are largely the product of fuel combustion, and
consequently, reducing CO2 emissions will also produce a significant reduction in projected
fuel consumption. EPA's projections of these impacts (in terms of gallons saved) are also
shown in this chapter. DRIA Chapter 5 presents the monetized fuel savings.

       In addition to the intended effects of reducing CO2 emission, the agencies also
consider the potential of the standards to affect vehicle safety. This topic is introduced in
Preamble Section II. G.  EPA's analysis of the change in fatalities due to projected usage of
mass reduction technology is shown in this chapter.

       Chapter 5: Vehicle Program Costs Including Fuel Consumption Impacts. The
program costs and fuel savings associated with EPA's proposed rulemaking. In Chapter 5, we
present briefly some of the outputs of the OMEGA model (costs per vehicle) and how we use
those  outputs to estimate the annual program costs (and fuel savings) of the proposal through
2050 and for each of the model years 2017 through 2025 that are effected by the proposal.
We also present our cost per ton analysis showing the cost incurred for each ton of GHG
reduced by the program.

       Also presented in Chapter 5 is what we call our "payback analysis" which looks at
how quickly the improved fuel efficiency of new vehicles provides savings to buyers despite
the vehicles having new technology (and new costs).  The consumer payback analysis shows
that fuel savings will  outweigh up-front costs in less than four years for people  purchasing
new vehicles with cash. For those purchasing new vehicles with a typical five-year car loan,
the fuel savings will outweigh increased costs in the first month of ownership.
                                         Vlll

-------
Chapter 6: Environmental and Health Impacts. This Chapter provides details on both the
climate impacts associated with changes in atmospheric CC>2 concentrations and the non-GHG
health and environmental impacts associated with criteria pollutants and air toxics.

       Based on modeling analysis performed by the EPA, reductions in CO2 and other GHG
emissions associated with this proposed rule will affect future  climate change. Since GHGs
are well-mixed in the atmosphere and have long atmospheric lifetimes, changes in GHG
emissions will affect atmospheric concentrations of greenhouse gases and future climate for
decades to millennia, depending on the gas. This section provides estimates of the projected
change in atmospheric CC>2 concentrations based on the emission reductions estimated for this
proposed rule, compared to the reference case. In addition, this section analyzes the response
to the changes in GHG concentrations of the following climate-related variables:  global mean
temperature, sea level rise, and ocean pH. See Chapter 4 in this DRIA for the estimated net
reductions in global emissions over time by GHG.

       There are also health and environmental impacts associated with the non-GHG
emissions projected to change as a result  of the proposed standards. To adequately assess
these impacts, full-scale photochemical air quality modeling is necessary to project changes in
atmospheric concentrations of PM2.5, ozone and air toxics.  The length of time needed to
prepare the necessary emissions inventories, in addition to the processing time associated with
the modeling itself, has precluded us from performing air quality modeling for this proposal.
However, for the final rule, a national-scale air quality modeling analysis will be  performed to
analyze the impacts of the vehicle standards on PM2.5, ozone, and selected air toxics (i.e.,
benzene, formaldehyde, acetaldehyde, acrolein and 1,3-butadiene).

       The atmospheric chemistry related to ambient concentrations of PM2.5, ozone and air
toxics is very complex, and making predictions based solely on emissions changes is
extremely difficult.  However, based on the magnitude of the emissions changes predicted to
result from the proposed vehicle standards (as shown in Chapter 4), we expect that there will
be an improvement in ambient air quality, pending a more comprehensive analysis for the
final rule.

       Chapter 7: Other Economic and Social Impacts.  This Chapter outlines a number of
additional impacts that contribute to the overall costs and benefits associated with the
proposed GHG standards. These impacts affect people outside the markets for vehicles and
their use; these effects are termed "external" and include the climate impacts,  energy security
impacts, and the effects on traffic, accidents, and noise due to  additional driving.

       Energy Security Impacts:  A reduction of U.S. petroleum imports reduces both
financial and strategic risks associated with a potential disruption in supply or a spike in cost
of a particular energy source.  This reduction in risk is a measure of improved U. S. energy
security.
                                          IX

-------
       SCC and GHG Benefits: EPA uses four estimates of the dollar value of marginal
reductions in CC>2 emissions—known as the social cost of carbon—to calculate total
monetized CC>2 benefits. Specifically, total monetized benefits in each year are calculated by
multiplying the SCC by the reductions in CC>2 for that year. EPA uses four different SCC
values to generate different estimates of total CO2 benefits and capture some of the
uncertainties involved in regulatory impact analysis.  The central value is the average SCC
across models at the 3 percent discount rate. For purposes of capturing the uncertainties
involved in regulatory impact analysis, we emphasize the importance and value of considering
the full range. Chapter 7 also presents an analysis of the CO2 benefits over the model year
lifetimes of the 2017 through 2025 model year vehicles.

       Other Impacts: There are other impacts associated with the GHG emissions standards
and associated reduced fuel consumption.  Lower fuel consumption would, presumably, result
in fewer trips to the filling station to refuel and, thus, time saved. The rebound effect,
discussed in detail in  Chapter 4 of the draft joint TSD, produces additional benefits to vehicle
owners in the form of consumer surplus from the increase in vehicle-miles driven, but may
also increase the societal costs associated with traffic congestion, motor vehicle crashes, and
noise. These  effects are likely to be relatively small in comparison to the value of fuel saved
as a result of the standards, but they are nevertheless important to include.

       Chapter 7 also presents a summary of the total costs, total benefits,  and net benefits
expected under the proposed  rule.  Table 2 presents these economics impacts. We note that
several of the cost and benefit categories we would typically discuss in an RIA are considered
joint economic assumptions shared between EPA and NHTSA and are therefore discussed in
more detail in EPA and NHTSA's draft Joint TSD Chapter 4.
  Table 2 Monetized Benefits Associated with the Proposed Program (Millions, 2009$)

Technology Costs
Fuel Savings
2017
$2,300
$570
2020
$8,470
$7,060
2030
$35,700
$85,800
2040
$39,800
$144,000
2050
$44,600
$187,000
NPV, 3%a
$551,000
$1,510,000
NPV, 7%a
$243,000
$579,000
Total Annual Benefits at each assumed SCC value b
5% (avg SCC)
3% (avg SCC)
2.5% (avg SCC)
3%(95th%ile)
$101
$141
$173
$250
$1,240
$1,730
$2,120
$3,100
$15,600
$22,000
$26,700
$40,500
$29,000
$40,400
$48,700
$75,100
$40,700
$55,600
$65,900
$102,000
$275,000
$413,000
$534,000
$764,000
$124,000
$263,000
$384,000
$614,000
Monetized Net Benefits at each assumed SCC value °
5% (avg SCC)
3% (avg SCC)
2.5% (avg SCC)
3%(95th%ile)
-$1,630
-$1,590
-$1,560
-$1,480
-$166
$325
$712
$1,690
$65,600
$72,000
$76,800
$90,500
$133,000
$144,000
$153,000
$179,000
$183,000
$198,000
$208,000
$244,000
$1,230,000
$1,370,000
$1,490,000
$1,720,000
$460,000
$599,000
$719,000
$950,000
Notes:
a Net present value of reduced CO2 emissions is calculated differently than other benefits. The same discount
rate used to discount the value of damages from future emissions (SCC at 5, 3, 2.5 percent) is used to calculate
net present value of SCC for internal consistency. Refer to the SCC TSD for more detail. Annual costs shown
are undiscounted values.
* DRIA Chapter 7.2 notes that SCC increases over time. For the years 2012-2050, the SCC estimates range as
follows: for Average SCC at 5%: $5-$16; for Average SCC at 3%: $23-$46; for Average SCC at 2.5%: $38-
$67; and for 95th percentile SCC at 3%:  $70-$ 140. DRIA Chapter 7.2 also presents these SCC estimates.
0 Net Benefits equal Fuel Savings minus Technology Costs plus Benefits.

-------
Chapter 8: Vehicle Sales and Employment.  Chapter 8 provides background on analyses of
the impacts of this rule on vehicle sales and employment in the auto industry and closely
related sectors.  We discuss how the payback period expected for the proposed standards (less
than 3 years in 2021, less than 4 years in 2025) is expected to affect vehicle sales.
Employment effects due to the rule depend in part on the state of the economy when the rule
becomes effective.  The auto industry (the directly regulated sector) is expected to require
additional labor due to increased production of fuel-saving technologies; employment in the
auto industry will also be affected by changes in vehicle sales and by the labor intensity of the
new technologies relative to the old technologies. Effects  on other  sectors vary.  Employment
for auto dealers as well as auto parts manufacturing will be affected by any changes in vehicle
sales.  Parts manufacturers may face increased labor demand due to production of the new
technologies.  Employment is expected to be reduced in petroleum-related sectors due to
reduced fuel demand. Finally, consumer spending is expected to affect employment through
changes in expenditures in general retail sectors; net fuel savings by consumers  are expected
to increase demand (and therefore employment) in other sectors. It should be noted that none
of these analyses was used in the benefit-cost analysis of the proposed standards, but they
provide a fuller picture of the impacts of this rule.

       Chapter 9:  Small Business Flexibility Analysis. EPA's analysis of the small
business impacts due to EPA's proposed rulemaking. EPA is proposing to exempt domestic
and foreign businesses that meet small business size definitions established by the Small
Business Administration.
                                         XI

-------

-------
                                               2017 Draft Regulatory Impact Analysis
1      Technology Packages, Cost and Effectiveness

1.1 Overview of Technology

       The proposed program is based on the need to obtain significant GHG emissions
reductions from the transportation sector, and the recognition that there are technically
feasible, cost effective technologies to achieve such reductions in the 2017-2025 timeframe at
reasonable per vehicle cost and short consumer payback periods, with no compromise to
vehicle utility or safety. As in many prior mobile source rulemakings,  the decision on what
standard to set is largely based on the effectiveness of the emissions control technology, the
cost (both per manufacturer and per vehicle) and other impacts of implementing the
technology, and the lead time needed for manufacturers to employ the  control technology.
EPA also considers the need for reductions of greenhouse gases, the degree of reductions
achieved by the standards, and the impacts of the standards in terms of costs, quantified and
unquantified benefits, safety, and other impacts. The availability of technology to achieve
reductions and the cost and other aspects of this technology are therefore a central focus of
this rulemaking.

       It is well known that CO2 is a stable compound produced by the complete combustion
of the fuel.  Vehicles combust fuel to perform two basic functions:  1) transport the vehicle, its
passengers and its contents, and 2) operate various accessories during the operation of the
vehicle such as the air conditioner.  Technology can reduce CC>2 emissions by either making
more efficient use of the energy that is produced through combustion of the fuel or by
reducing the energy needed to perform either of these functions.

       This focus on efficiency involves a major change in focus and calls for looking at the
vehicle as an entire system.  In addition to fuel delivery, combustion, and aftertreatment
technology, any aspect of the vehicle that affects the need to produce energy must also be
considered. For example, the efficiency of the transmission system, which takes the energy
produced by the engine and transmits it to the wheels, and the resistance of the tires to rolling
both have major impacts on the amount of fuel that is combusted while operating the vehicle.
Braking system drag, the aerodynamic drag of the vehicle, and the efficiency of accessories
(such as the air conditioner) all  affect how much fuel is combusted.

       This need to focus on the efficient use of energy by the vehicle  as a system leads to a
broad focus on a wide variety of technologies that affect almost all  the  systems in the design
of a vehicle. As discussed below, there are many technologies that are currently available
which can reduce vehicle energy consumption.  These technologies are already being
commercially utilized to a limited degree in the current light-duty fleet. These technologies
include hybrid technologies that use higher efficiency electric motors as the power source in
combination with or instead of internal combustion engines.  While already commercialized,
hybrid technology continues to be developed and offers the potential for even greater
efficiency improvements.  There are a number of technologies that  were described in the
2012-2016 rule (TSD and RIA) that are also common to this rule. While we expect
significant penetration of these technologies within the 2016 timeframe, there will be some
technologies that will have continued improvement, and others that are only partially
implemented into the fleet by 2016.  We describe those technologies for which  we expect to
                                         1-1

-------
                                               2017 Draft Regulatory Impact Analysis
see continued improvement—engine friction reduction, lower rolling resistance tires—in
Chapter 3 of the draft joint TSD and generally denote them as "level 2" versions of each
technology.  The primary examples of those technologies that we expect to be only partially
implemented into the fleet by 2016 would be weight reduction greater than 5-10% and
electrification of powertrains to hybrid, plug-in electric and full electric which we do not
project manufacturers as needing to utilize to meet their MY 2012-2016 standards . There are
also other advanced technologies under development (that were not projected to be available
to meet 2012-2016 standards), such as turbocharged engines with increasingly high levels of
boost and lean burn gasoline engines, both of which offer the potential of improved energy
generation through enhancements to the basic combustion process.  Finally, there may be
technologies not considered for this rule that, given the long lead time, can be developed and
introduced into the market.  These currently unknown technologies (or enhancements of
known technologies) could be more cost effective than those included in this analysis.  The
more cost-effective a new technology is, the more it is likely that an auto manufacturer will
implement it.

       The large number of possible technologies to consider and the breadth of vehicle
systems that are affected mean that consideration of the manufacturer's design and production
process plays a major role in developing the standards.  Vehicle manufacturers typically
develop their many different models by basing them on a limited number of vehicle platforms.
Several different models of vehicles are produced using a common platform, allowing for
efficient use of design and manufacturing resources. The platform typically consists of
common vehicle architecture and structural components. Given the very large investment put
into designing and  producing each vehicle model, manufacturers cannot reasonably redesign
any given vehicle every year or even every other year, let alone redesign all of their vehicles
every year or every other year.  At the redesign stage, the manufacturer will upgrade  or add all
of the technology and make all of the other changes needed so the vehicle model will meet the
manufacturer's plans for the next several  years.  This includes meeting all of the emissions
and other requirements that would apply during the years before  the next major redesign of
the vehicle.

       This redesign often involves a package of changes, designed to work together to meet
the various requirements and plans for the model for several model years after the redesign.
This typically involves significant engineering, development, manufacturing, and marketing
resources to create  a new product with multiple new features.  In order to leverage this
significant upfront  investment, manufacturers plan vehicle redesigns with several model years
of production in mind. That said, vehicle  models are not completely static between redesigns
as limited changes  are often incorporated for each model year. This interim process is called
a refresh of the vehicle and generally does not allow for major technology changes although
more minor ones can be done (e.g., aerodynamic improvements,  valve timing improvements).
More major technology upgrades that affect multiple systems of the vehicle (e.g.,
hybridization) thus occur at the vehicle redesign stage and not between redesigns.

       Given that the regulatory timeframe of the GHG program is nine years (2017 through
2025), and given EPA's belief that full line manufacturers (i.e., those making small cars
through large cars,  minivans, small trucks and large trucks) cannot redesign, on average, their
entire product line more than twice during that timeframe, we have assumed two full redesign
                                         1-2

-------
                                                2017 Draft Regulatory Impact Analysis
cycles in the 2017-2025 timeframe  This means that the analysis assumes that each vehicle
platform in the US fleet can undergo at least two full redesigns during our regulatory
timeframe.

       As discussed below, there are a wide variety of emissions control technologies
involving several different systems in the vehicle that are available for consideration. Many
can involve major changes to the vehicle, such as changes to the engine block and heads, or
redesign of the transmission and its packaging in the vehicle. This calls for tying the
incorporation of the emissions control technology into the periodic redesign process.  This
approach would allow manufacturers to develop appropriate packages of technology upgrades
that combine technologies in ways that work together and fit with the overall goals of the
redesign. It also allows the manufacturer to fit the process of upgrading emissions control
technology into its multi-year planning process, and it avoids the large increase in resources
and costs that would occur if technology had to be added outside of the redesign process.

       Over the nine model years at issue in this rulemaking, 2017-2025, EPA projects that
almost the entire fleet of light-duty vehicles will have gone through two redesign cycles.  If
the technology to control greenhouse gas emissions is  efficiently folded into this redesign
process, then by 2025 the entire light-duty fleet could be designed to employ upgraded
packages of technology to reduce emissions of CO2, and as discussed below, to reduce
emissions of harmful refrigerants from the air conditioner.

       In determining the projected technology needed to meet the standards, and the cost of
those technologies, EPA is using an approach that accounts for and builds on this redesign
process.  This provides the opportunity for several control technologies to be incorporated
into the vehicle during redesign, achieving significant  emissions reductions from the model at
one time. This is in contrast to what would be a much more costly approach of trying to
achieve small increments of reductions over multiple years by adding technology to the
vehicle piece by piece outside of the redesign process.

       As described below, the vast majority of technology we project as being utilized to
meet the GHG standards is commercially available and already being used to a limited
extent across the fleet, although far greater penetration of these technologies into the fleet is
projected as a result of both the 2012-2016 final rule and this proposal. The vast majority of
the emission reductions associated with this proposal would result from the increased use  of
these technologies. EPA also believes the proposal would  encourage the development and
limited use of more advanced technologies, such as plug-in hybrid electric vehicles (PHEVs)
and full electric vehicles (EVs), and is structuring the proposal to encourage these
technologies' use.

       In section 1.2 below, a summary of technology costs and effectiveness is presented. In
section 1.3, the process of combining technologies into packages is described along with
package costs and effectiveness. Sections 1.4 and 1.5  discuss the lumped parameter approach
which provides background and support for determining technology and package
effectiveness.
                                         1-3

-------
                                               2017 Draft Regulatory Impact Analysis
1.2 Technology Cost and Effectiveness

       EPA collected information on the cost and effectiveness of CC>2 emission reducing
technologies from a wide range of sources.  The primary sources of information were the
2012-2016 FRM, the 2010 Technical Assessment Report (TAR), tear-down analysis done by
FEV and the 2008 as well as 2010 Ricardo studies. In addition, we considered confidential
data submitted by vehicle manufacturers in response to NHTSA's request for product plans ,
along with confidential information shared by automotive industry component suppliers in
meetings with EPA and NHTSA staff. These confidential data sources were used primarily as
a validation of the estimates since EPA prefers to rely on public data rather than confidential
data wherever possible.

       Since publication of the 2012-2016 FRM, EPA has continued the work with FEV that
consists of complete system tear-downs to evaluate technologies down to the nuts and bolts—
i.e., a "bill of materials"—to arrive at very detailed estimates of the costs associated with
manufacturing them.  Also, cost and effectiveness estimates were adjusted as a result of
further meetings between EPA and NHTSA staffs following publication of the 2010 TAR and
into the first half of 2011 where both piece costs and fuel consumption efficiencies were
discussed in detail. EPA and NHTSA also met with Department of Energy (DOE) along with
scientists and engineers from a number of national laboratories to discuss vehicle
electrification. EPA also reviewed the published technical literature which addressed the
issue of CC>2 emission control, such as papers published by the Society of Automotive
Engineers and the American Society of Mechanical Engineers.1  The results of these efforts
especially the results  of the FEV tear-down and Ricardo studies were used extensively in this
proposal  as described in  detail in Chapter 3 of the draft joint TSD.

       For all of the details behind the cost and effectiveness values used in this analysis the
reader is referred to Chapter 3 of the draft joint TSD.  There we present direct manufacturing
costs, indirect costs and total costs for each technology in each MY 2017 through 2025.  We
also describe the source for each direct manufacturing cost and how those costs change over
time due to learning,  and the indirect costs and how they change over time.  Note that all costs
presented in the tables that follow are total costs and include both direct manufacturing and
indirect costs.

       For direct manufacturing costs (DMC) related to turbocharging, downsizing,  gasoline
direct injection, transmissions, as well as non-battery-related costs on hybrid, plug-in hybrid
and electric vehicles,  the agencies have relied on costs derived from teardown studies. For
battery related DMC  for HEVs, PHEVs and EVs, the agencies have relied on the BatPaC
model developed by Argonne National Laboratory for the Department of Energy. For mass
reduction DMC, the agencies have relied on several studies as  described in detail in the draft
Joint TSD. For the majority of the other technologies considered in this proposal, the
agencies have relied on the 2012-2016 final rule and sources described there for estimates of
DMC.

       For this analysis, indirect costs are estimated by applying indirect cost multipliers
(ICM) to direct cost estimates. ICMs were derived by EPA as a basis for estimating the
impact on indirect costs of individual vehicle technology changes that would result from
                                         1-4

-------
                                               2017 Draft Regulatory Impact Analysis
regulatory actions.  Separate ICMs were derived for low, medium, and high complexity
technologies, thus enabling estimates of indirect costs that reflect the variation in research,
overhead, and other indirect costs that can occur among different technologies.  ICMs were
also applied in the MYs 2012-2016 rulemaking. We have also included an estimate of
stranded capital that could result due to introduction of technology on a more rapid pace than
the industry norm.  We describe our ICMs and the method by which they are applied to direct
costs and our stranded capital estimates in the draft Joint TSD Chapter 3.2.2.  Stranded capital
is also discussed in this draft RIA at Chapter 3.8.7 and Chapter 5.1.

       Regarding learning effects, we continue to apply learning effects in the same way as
we did in both the MYs 2012-2016 final rule and in the 2010 TAR. However, we have
employed some new terminology in an effort to eliminate some confusion that existed with
our old terminology.  This new terminology was described in the recent heavy-duty GHG
final rule (see 76 FR 57320). Our old terminology suggested we were accounting for two
completely different learning effects—one based on volume production and the other based
on time.  This was not the case since, in fact, we were actually relying on just one learning
phenomenon, that being the learning-by-doing phenomenon that results from cumulative
production volumes.

       As a result, we have considered the impacts of manufacturer learning on the
technology cost estimates by reflecting the phenomenon of volume-based learning curve cost
reductions in our modeling using two algorithms depending  on where in the learning cycle
(i.e., on what portion of the learning curve) we consider a technology to be - "steep" portion
of the curve for newer technologies and "flat" portion of the curve for more mature
technologies. The observed phenomenon in the economic literature which supports
manufacturer learning cost reductions are based on reductions in costs as production volumes
increase with the highest absolute cost reduction occurring with the first doubling of
production.  The agencies use the terminology "steep" and "flat" portion of the curve to
distinguish among newer technologies and more mature technologies, respectively, and how
learning cost reductions are applied in cost analyses.

       Learning impacts have been considered on most but not all of the technologies
expected to be used because some of the expected technologies  are already used rather widely
in  the industry and, presumably, quantifiable learning impacts have already occurred. We
have applied the steep learning algorithm for only a handful  of technologies considered to be
new or emerging technologies such as PHEV and EV batteries which are experiencing heavy
development and, presumably, rapid cost declines in coming years. For most technologies,
we have considered them to be more established and, hence, we have applied the lower flat
learning algorithm. For more discussion of the learning approach and the technologies to
which each type of learning has been applied the reader is directed to Chapter 3.2.3 of the
draft Joint TSD.

       Fuel consumption reductions are possible from a variety of technologies whether they
be engine-related (e.g., turbocharging), transmission-related (e.g., six forward gears in place
of four), accessory-related (e.g., electric power steering), or vehicle-related (e.g., lower rolling
resistance tires). Table 1.2-1 through Table 1.2-14 present the costs associated with the
technologies we believe would be the enabling technologies for compliance with the proposed
                                         1-5

-------
                                               2017 Draft Regulatory Impact Analysis
standards. Note that many of these technologies are expected to have penetrated the fleet as
much as 85 to 100 percent by the 2016 MY and, as such, would represent reference case
technologies in this proposal.  That is, technologies such as lower rolling resistance tires and
level 1 aerodynamic treatments are expected to exceed 85 percent penetration by 2016 so they
cannot be added "again" to comply with the 2017-2025 proposed standards. However, we list
all such technologies in the tables that follow for completeness and comparison to earlier
analyses.
       One thing that is immediately clear from the cost tables that follow is that we have
updated our costing approach for some technologies in an effort to provide better granularity
in our estimates.  This is easily seen in Table 1.2-1 and Table 1.2-2 where we list costs for
technologies by engine configuration—in-line or "I" versus "V"—and/or by number of
cylinders. In the 2012-2016 final rule, we showed costs for a small car, large car, large truck,
etc. The limitation of that approach was that different vehicle classes can have many different
sized engines. This is exacerbated when estimating costs for turbocharged and downsized
engines.  For example, we project that many vehicles in the large car class which, today, have
V8 engines would have highly turbocharged 14 engines under the proposal.  As such, we
would not want to estimate the large car costs of engine friction reduction (EFR)—which
have always and continue to be based on the number of cylinders—assuming that all large
cars have V8 engines.  With our new approach, the large cars that remain V8 would carry
EFR costs for a V8, one downsized to a V6 would carry EFR costs for a V6 and one
downsized further to an 14 would carry EFR costs for an 14.  Our old approach would have
applied the EFR cost for a V8 to each.
       Note that Table  1.2-14 presents costs for mass reduction technology on each of the 19
vehicle types used in OMEGA. We present costs for only a 10% and a 20% applied weight
reduction. We use the term "applied" weight reduction to reflect the amount of weight
reduction technology—or weight reduction cost—applied to the package.  We also use the
term "net" weight reduction when determining costs for hybrid, plug-in hybrid, and full
electric vehicles (see Table 1.2-8 through Table  1.2-12). The net weight reduction is the
applied weight reduction less the added weight of the hybrid and/or electric vehicle
technologies. Table 1.2-7 shows costs for P2 hybrids.  For the subcompact P2 FIEV with an
applied weight reduction of 10%, the net weight reduction is shown as 5%. Therefore, our
cost analysis would add the costs for 10% weight reduction for such a P2 FIEV even though
the net weight reduction was  only 5%. Likewise, we would add the cost of P2 HEV
technology  for only a 5% weight reduction  since that is the net weight reduction of the
vehicle. Note that the higher the net weight reduction the lower the cost for HEV and/or EV
technologies since smaller batteries and motors can be used as the vehicle gets lighter). How
we determined the necessary  battery pack sizes and the resultant net weight impacts is
described in Chapter 3 of the draft joint TSD.  We note that the approach described there is a
departure from our earlier efforts where the weight increase of the electrification components
was not fully recognized. Importantly, that had little impact on the analysis used to support
the 2012-2016 rule since that rule projected very low penetration of HEVs and no PHEV or
EV penetrations.
       All costs continue to be relative to a baseline vehicle powertrain system (unless
otherwise noted) consisting of a multi-point, port fuel injected, naturally aspirated gasoline
engine operating at a stoichiometric air-fuel ratio with fixed valve timing and lift paired with a
                                         1-6

-------
                                                   2017 Draft Regulatory Impact Analysis
4-speed automatic transmission.  This configuration was chosen as the baseline vehicle
because it was the predominant technology package sold in the United States in the baseline
model year 2008. Costs are presented in terms of their hardware incremental compliance
cost. This means that they include all potential product development costs associated with
their application on vehicles, not just the cost of their physical parts. A more detailed
description of these and the following estimates of cost and effectiveness of CO2 reducing
technologies can be found in Chapter 3 of the draft joint TSD, along with a more detailed
description of the comprehensive technical evaluation underlying the estimates.
                    Table 1.2-1 Costs for Engine Technologies (2009$)
Technology
Conversion to Atkinson
CCP-OHC-I
CCP-OHC-V
CCP-OHV-V
CVVL-OHC-I4
CVVL-OHC-V6
CVVL-OHC-V8
DCP-OHC-I
DCP-OHC-V
DCP-OHV-V
Deac-V6
Deac-V8
DVVL-OHC-I4
DVVL-OHC-V6
DVVL-OHC-V8
EFR1-I3
EFR1-I4
EFR1-V6
EFR1-V8
EFR2-I3
EFR2-I4
EFR2-V6
EFR2-V8
EGR-I
EGR-V
LUB
Stoich GDI-I4
Stoich GDI-I4>I3
Stoich GDI-V6
Stoich GDI-V8
V6 OHV to V6 DOHC
V6 SOHC to V6 DOHC
V8 OHV to V8 DOHC
V8 SOHC 3V to V8
DOHC
V8 SOHC to V8 DOHC
WTI-OHC-I
wn-OHC-v
wn-OHV-v
2017
$0
$46
$91
$46
$240
$440
$480
$94
$201
$102
$192
$216
$160
$232
$332
$44
$58
$87
$116
$95
$124
$182
$240
$303
$303
$4
$274
$274
$413
$497
$676
$212
$740
$153
$245
$46
$91
$46
2018
$0
$45
$90
$45
$237
$434
$473
$92
$198
$101
$189
$213
$158
$229
$327
$44
$58
$87
$116
$95
$124
$182
$240
$298
$298
$4
$270
$270
$407
$490
$661
$209
$724
$151
$241
$45
$90
$45
2019
$0
$42
$84
$42
$216
$396
$432
$84
$181
$92
$173
$194
$144
$209
$298
$42
$56
$84
$111
$95
$124
$182
$240
$294
$294
$4
$246
$246
$370
$445
$599
$190
$656
$137
$219
$42
$84
$42
2020
$0
$42
$83
$42
$212
$389
$425
$83
$178
$90
$170
$191
$142
$205
$293
$42
$56
$84
$111
$95
$124
$182
$240
$290
$290
$4
$242
$242
$364
$438
$585
$187
$640
$135
$216
$42
$83
$42
2021
$0
$41
$82
$41
$209
$383
$418
$81
$175
$89
$167
$188
$139
$202
$289
$42
$56
$84
$111
$95
$124
$182
$240
$285
$285
$4
$238
$238
$359
$431
$571
$184
$625
$133
$212
$41
$82
$41
2022
$0
$40
$80
$40
$206
$377
$412
$80
$173
$87
$165
$185
$137
$199
$284
$42
$56
$84
$111
$95
$124
$182
$240
$281
$281
$4
$234
$234
$353
$425
$558
$181
$611
$131
$209
$40
$80
$40
2023
$0
$40
$79
$40
$203
$372
$405
$79
$170
$86
$162
$182
$135
$196
$280
$42
$56
$84
$111
$95
$124
$182
$240
$277
$277
$4
$231
$231
$348
$418
$549
$179
$601
$129
$206
$40
$79
$40
2024
$0
$39
$78
$39
$200
$366
$399
$78
$167
$85
$160
$180
$133
$193
$276
$42
$56
$84
$111
$95
$124
$182
$240
$274
$274
$4
$227
$227
$343
$412
$540
$176
$592
$127
$203
$39
$78
$39
2025
$0
$38
$76
$38
$197
$360
$393
$77
$165
$84
$157
$177
$131
$190
$271
$42
$56
$84
$111
$91
$119
$175
$230
$247
$247
$4
$224
$224
$338
$406
$532
$173
$583
$125
$200
$38
$76
$38
CCP=coupled cam phasing; CVVL=continuous variable valve lift; DCP=dual cam phasing; Deac=cylinder deactivation;
DOHC=dual overhead cam; DVVL=discrete variable valve lift; EFRl=engine friction reduction level 1; EFR2=EFR level 2;
EGR=exhaust gas recirculation; GDI=gasoline direct injection; I=inline engine; I3=inline 3 cylinder; I4=inline 4 cylinder;
LUB=low friction lube; OHC=overhead cam; OHV=overhead valve; SOHC=single overhead cam; Stoic=stoichiometric
                                             1-7

-------
                                                       2017 Draft Regulatory Impact Analysis
air/fuel; V=V-configuration engine; V6=V-configuration 6 cylinder; V8=V-configuration 8 cylinder; VVTI=intake variable
valve timing; 3V=3 valves per cylinder.
All costs are incremental to the baseline case.
                 Table 1.2-2 Costs for Turbocharging & Downsizing (2009$)
Technology
14 to 13 wT
14 to 13 wT
14 to 13 wT
14 DOHC to 14
DOHC wT
14 DOHC to 14
DOHC wT
14 DOHC to 14
DOHC wT
V6 DOHC to 14
wT
V6 DOHC to 14
wT
V6 DOHC to 14
wT
V6 SOHC to 14
wT
V6 SOHC to 14
wT
V6 SOHC to 14
wT
V6 OHV to 14
DOHC wT
V6 OHV to 14
DOHC wT
V6 OHV to 14
DOHC wT
V8 DOHC to V6
DOHC wT
V8 DOHC to V6
DOHC wT
V8 DOHC to 14
DOHC wT
V8 SOHC to V6
DOHC wT
V8 SOHC to V6
DOHC wT
V8 SOHC to 14
DOHC wT
V8 SOHC 3V to
V6 DOHC wT
V8 SOHC 3V to
V6 DOHC wT
V8 SOHC 3V to
14 DOHC wT
V8 OHV to V6
DOHC wT
V8 OHV to V6
DOHC wT
V8 OHV to 14
DOHC wT
BMEP
18 bar
24 bar
27 bar
18 bar
24 bar
27 bar
18 bar
24 bar
27 bar
18 bar
24 bar
27 bar
18 bar
24 bar
27 bar
18 bar
24 bar
27 bar
18 bar
24 bar
27 bar
18 bar
24 bar
27 bar
18 bar
24 bar
27 bar
2017
$424
$685
$1,205
$478
$738
$1,259
$248
$509
$1,029
$330
$591
$1,111
$903
$1,163
$1,683
$741
$1,180
$787
$835
$1,274
$906
$800
$1,238
$861
$1,323
$1,762
$1,155
2018
$420
$677
$1,189
$472
$728
$1,241
$250
$507
$1,019
$329
$586
$1,098
$887
$1,143
$1,656
$733
$1,165
$792
$824
$1,256
$905
$790
$1,222
$863
$1,301
$1,733
$1,143
2019
$357
$650
$1,155
$418
$710
$1,216
$159
$451
$957
$251
$544
$1,049
$805
$1,097
$1,602
$631
$1,124
$716
$738
$1,231
$842
$698
$1,191
$795
$1,180
$1,673
$1,108
2020
$353
$642
$1,140
$412
$701
$1,199
$161
$449
$948
$250
$539
$1,037
$789
$1,078
$1,576
$624
$1,110
$720
$727
$1,214
$842
$688
$1,175
$796
$1,159
$1,646
$1,097
2021
$350
$635
$1,126
$407
$692
$1,183
$163
$448
$939
$250
$535
$1,026
$775
$1,060
$1,551
$616
$1,097
$725
$717
$1,197
$841
$679
$1,160
$797
$1,138
$1,618
$1,085
2022
$346
$627
$1,111
$401
$683
$1,167
$164
$446
$930
$249
$530
$1,014
$760
$1,042
$1,526
$609
$1,084
$729
$707
$1,181
$840
$670
$1,144
$799
$1,118
$1,592
$1,074
2023
$342
$620
$1,097
$396
$674
$1,151
$166
$444
$921
$248
$526
$1,003
$749
$1,026
$1,504
$602
$1,071
$726
$697
$1,165
$834
$661
$1,130
$793
$1,101
$1,569
$1,061
2024
$338
$613
$1,083
$390
$665
$1,135
$168
$442
$913
$247
$522
$992
$737
$1,012
$1,482
$595
$1,058
$723
$687
$1,149
$828
$652
$1,115
$788
$1,084
$1,547
$1,048
2025
$335
$547
$972
$385
$598
$1,022
$170
$382
$807
$246
$459
$884
$726
$938
$1,363
$588
$946
$622
$677
$1,035
$724
$644
$1,002
$686
$1,067
$1,426
$938
DOHC=dual overhead cam; I3=inline 3 cylinder; I4=inline 4 cylinder; OHV=overhead valve; SOHC=single overhead cam;

-------
                                                        2017 Draft Regulatory Impact Analysis
V6=V-configuration 6 cylinder; V8=V-configuration 8 cylinder; 3V=3 valves per cylinder; wT=with turbo.
All costs are incremental to the baseline case.
                  Table 1.2-3 Costs for Transmission Technologies (2009$)
Technology
ASL
ASL2
5spAT
6sp AT
6sp DCT-dry
6sp DCT-wet
6spMT
SspAT
8sp DCT-dry
8sp DCT-wet
HEG
TORQ
2017
$32
$33
$103
-$9
-$115
-$81
-$168
$61
-$16
$47
$248
$29
2018
$32
$33
$101
-$9
-$111
-$78
-$163
$60
-$15
$46
$242
$29
2019
$30
$32
$95
-$9
-$130
-$91
-$170
$55
-$14
$46
$236
$27
2020
$29
$31
$94
-$9
-$126
-$89
-$166
$54
-$14
$45
$231
$27
2021
$29
$30
$92
-$9
-$122
-$86
-$162
$53
-$13
$44
$225
$26
2022
$28
$30
$91
-$9
-$118
-$84
-$158
$52
-$13
$44
$220
$26
2023
$28
$29
$89
-$8
-$115
-$81
-$154
$52
-$12
$43
$216
$25
2024
$27
$29
$88
-$8
-$111
-$79
-$150
$51
-$12
$43
$213
$25
2025
$27
$27
$86
-$8
-$108
-$76
-$146
$50
-$15
$38
$200
$24
ASL=aggressive shift logic; ASL2=aggressive shift logic level 2
clutch transmission; HEG=high efficiency gearbox; MT=manual
lockup.
All costs are incremental to the baseline case.
(shift optimizer); AT=automatic transmission; DCT=dual
transmission; sp=speed; TORQ=early torque converter
        Table 1.2-4 Costs for Electrification & Improvement of Accessories (2009$)
Technology
EPS/EHPS
IACC
IACC2
Stop-start (12V)
for Subcompact,
Small car
Stop-start (12V)
for Large car,
Minivan, Small
truck
Stop-start (12V)
for Large truck
2017
$108
$87
$141
$394
$446
$490
2018
$106
$86
$139
$385
$436
$479
2019
$100
$81
$131
$348
$395
$433
2020
$98
$80
$129
$340
$385
$423
2021
$96
$78
$127
$332
$376
$413
2022
$95
$77
$124
$324
$368
$403
2023
$93
$76
$122
$317
$359
$394
2024
$92
$75
$120
$310
$351
$385
2025
$90
$73
$118
$303
$343
$376
EPS=electric power steering; EHPS=electro-hydraulic power steering; IACC=improved accessories level 1; IACC2=IACC
level 2; 12V=12 volts.
All costs are incremental to the baseline case.
                      Table 1.2-5 Costs for Vehicle Technologies (2009$)
Technology
Aerol
Aero2
LDB
LRRT1
LRRT2
SAX
2017
$48
$210
$73
$7
$72
$96
2018
$47
$207
$73
$7
$72
$94
2019
$45
$201
$70
$6
$60
$89
2020
$44
$198
$70
$6
$60
$88
2021
$43
$195
$70
$6
$50
$86
2022
$42
$192
$70
$6
$48
$85
2023
$42
$190
$70
$6
$47
$83
2024
$41
$187
$70
$6
$46
$82
2025
$40
$173
$70
$6
$43
$81
Aerol=aerodynamic treatments level 1; Aero2=aero level 2; LDB=low drag brakes; LRRTl=lower rolling resistance tires
                                                 1-9

-------
                                                  2017 Draft Regulatory Impact Analysis
level 1; LRRT2=LRRT level 2; SAX=secondary axle disconnect.
All costs are incremental to the baseline case.
                Table 1.2-6 Costs for Advanced Diesel Technology (2009$)
Vehicle Class
Subcompact/
Small car
Large car
Minivan
Small truck
Large truck
2017
$2,936
$3,595
$2,942
$2,967
$4,114
2018
$2,893
$3,543
$2,900
$2,924
$4,054
2019
$2,627
$3,218
$2,633
$2,656
$3,682
2020
$2,587
$3,168
$2,593
$2,615
$3,625
2021
$2,547
$3,120
$2,553
$2,575
$3,570
2022
$2,509
$3,072
$2,514
$2,536
$3,515
2023
$2,471
$3,026
$2,476
$2,497
$3,462
2024
$2,433
$2,980
$2,439
$2,460
$3,410
2025
$2,397
$2,936
$2,402
$2,423
$3,359
All costs are incremental to the baseline case.
                   Table 1.2-7 Costs for P2-Hybird Technology (2009$)
Vehicle Class
Subcompact
Subcompact
Subcompact
Small car
Small car
Small car
Large car
Large car
Large car
Minivan
Minivan
Minivan
Small truck
Small truck
Small truck
Minivan-towing
Minivan-towing
Minivan-towing
Large truck
Large truck
Large truck
Applied
WR
10%
15%
20%
10%
15%
20%
10%
15%
20%
10%
15%
20%
10%
15%
20%
10%
15%
20%
10%
15%
20%
Net
WR
5%
10%
15%
5%
10%
15%
5%
10%
15%
5%
10%
15%
5%
10%
15%
4%
9%
14%
4%
9%
14%
2017
$3,489
$3,452
$3,415
$3,665
$3,625
$3,585
$4,196
$4,131
$4,067
$4,148
$4,092
$4,036
$4,005
$3,953
$3,900
$4,286
$4,228
$4,170
$4,417
$4,358
$4,298
2018
$3,435
$3,399
$3,363
$3,609
$3,569
$3,530
$4,132
$4,068
$4,004
$4,084
$4,029
$3,974
$3,943
$3,892
$3,841
$4,219
$4,162
$4,105
$4,349
$4,290
$4,232
2019
$3,027
$2,995
$2,962
$3,180
$3,145
$3,110
$3,640
$3,583
$3,527
$3,597
$3,548
$3,498
$3,473
$3,428
$3,382
$3,728
$3,677
$3,626
$3,844
$3,792
$3,739
2020
$2,977
$2,945
$2,913
$3,128
$3,093
$3,058
$3,580
$3,524
$3,469
$3,538
$3,489
$3,440
$3,416
$3,371
$3,326
$3,666
$3,616
$3,566
$3,780
$3,728
$3,677
2021
$2,928
$2,897
$2,866
$3,076
$3,042
$3,008
$3,521
$3,466
$3,412
$3,480
$3,432
$3,384
$3,360
$3,316
$3,271
$3,606
$3,556
$3,507
$3,717
$3,667
$3,616
2022
$2,880
$2,850
$2,819
$3,026
$2,993
$2,959
$3,464
$3,410
$3,356
$3,423
$3,376
$3,329
$3,305
$3,262
$3,218
$3,547
$3,498
$3,450
$3,657
$3,607
$3,557
2023
$2,834
$2,804
$2,774
$2,977
$2,944
$2,912
$3,408
$3,355
$3,302
$3,368
$3,322
$3,275
$3,252
$3,209
$3,166
$3,489
$3,441
$3,394
$3,597
$3,548
$3,499
2024
$2,788
$2,759
$2,729
$2,930
$2,897
$2,865
$3,353
$3,301
$3,249
$3,314
$3,268
$3,223
$3,200
$3,158
$3,116
$3,433
$3,386
$3,339
$3,539
$3,491
$3,443
2025
$2,595
$2,567
$2,540
$2,725
$2,696
$2,666
$3,121
$3,073
$3,025
$3,085
$3,044
$3,002
$2,979
$2,940
$2,901
$3,185
$3,142
$3,099
$3,282
$3,238
$3,194
WR=weight reduction.
All costs are incremental to the baseline case.
  Table 1.2-8 Costs for Plug-in Hybrid Technology with 20 Mile EV Range, or PHEV20
                                         (2009$)
Vehicle Class
Subcompact
Subcompact
Subcompact
Small car
Small car
Small car
Large car
Large car
Applied
WR
10%
15%
20%
10%
15%
20%
10%
15%
Net
WR
3%
8%
13%
3%
8%
13%
3%
8%
2017
$11,325
$11,099
$10,874
$12,143
$11,890
$11,637
$15,448
$15,020
2018
$10,191
$9,985
$9,779
$10,945
$10,714
$10,483
$13,989
$13,597
2019
$9,521
$9,332
$9,143
$10,206
$9,993
$9,781
$12,971
$12,613
2020
$8,607
$8,433
$8,260
$9,240
$9,045
$8,851
$11,793
$11,464
2021
$8,558
$8,386
$8,214
$9,186
$8,993
$8,800
$11,719
$11,392
2022
$8,510
$8,339
$8,168
$9,133
$8,941
$8,749
$11,646
$11,322
2023
$8,463
$8,293
$8,124
$9,081
$8,890
$8,700
$11,575
$11,253
2024
$8,417
$8,248
$8,080
$9,030
$8,841
$8,652
$11,505
$11,185
2025
$6,976
$6,832
$6,688
$7,509
$7,348
$7,187
$9,657
$9,382
                                           1-10

-------
                                                 2017 Draft Regulatory Impact Analysis
Large car
Minivan
Minivan
Minivan
Small truck
Small truck
Small truck
20%
10%
15%
20%
10%
15%
20%
13%
3%
8%
13%
3%
8%
13%
$14,593
$14,951
$14,577
$14,203
$14,138
$13,796
$13,453
$13,205
$13,511
$13,169
$12,827
$12,766
$12,454
$12,141
$12,255
$12,559
$12,245
$11,932
$11,878
$11,591
$11,304
$11,135
$11,397
$11,110
$10,822
$10,771
$10,508
$10,246
$11,065
$11,327
$11,042
$10,757
$10,706
$10,445
$10,185
$10,997
$11,259
$10,976
$10,693
$10,643
$10,384
$10,125
$10,931
$11,193
$10,911
$10,630
$10,581
$10,323
$10,066
$10,866
$11,127
$10,848
$10,568
$10,520
$10,264
$10,008
$9,107
$9,301
$9,062
$8,824
$8,779
$8,561
$8,344
WR=weight reduction.
All costs are incremental to the baseline case.
  Table 1.2-9 Costs for Plug-in Hybrid Technology with 40 Mile EV Range, or PHEV40
                                         (2009$)
Vehicle Class
Subcompact
Subcompact
Small car
Small car
Large car
Large car
Minivan
Minivan
Small truck
Small truck
Applied
WR
15%
20%
15%
20%
15%
20%
15%
20%
15%
20%
Net
WR
2%
7%
3%
8%
2%
7%
2%
7%
3%
8%
2017
$14,293
$14,067
$15,655
$15,248
$20,644
$20,000
$19,848
$19,382
$18,637
$18,139
2018
$12,717
$12,510
$13,929
$13,568
$18,411
$17,836
$17,677
$17,257
$16,589
$16,145
2019
$12,043
$11,854
$13,190
$12,847
$17,386
$16,844
$16,720
$16,328
$15,702
$15,282
2020
$10,775
$10,601
$11,802
$11,495
$15,589
$15,102
$14,973
$14,619
$14,054
$13,678
2021
$10,726
$10,553
$11,748
$11,443
$15,514
$15,030
$14,903
$14,551
$13,989
$13,615
2022
$10,677
$10,506
$11,695
$11,391
$15,441
$14,959
$14,835
$14,485
$13,926
$13,553
2023
$10,630
$10,460
$11,643
$11,341
$15,369
$14,889
$14,768
$14,419
$13,864
$13,493
2024
$10,584
$10,415
$11,593
$11,291
$15,298
$14,821
$14,702
$14,356
$13,803
$13,433
2025
$8,571
$8,427
$9,390
$9,146
$12,450
$12,061
$11,932
$11,644
$11,189
$10,888
WR=weight reduction.
All costs are incremental to the baseline case.
  Table 1.2-10 Costs for Full Electric Vehicle Technology with 75 Mile Range, or EV75
                                         (2009$)
Vehicle Class
Subcompact
Subcompact
Subcompact
Small car
Small car
Small car
Large car
Large car
Large car
Minivan
Minivan
Minivan
Small truck
Small truck
Small truck
Applied
WR
10%
15%
20%
10%
15%
20%
10%
15%
20%
10%
15%
20%
10%
15%
20%
Net
WR
10%
15%
20%
10%
15%
20%
10%
15%
20%
10%
15%
20%
10%
15%
20%
2017
$15,236
$14,914
$14,592
$16,886
$16,442
$15,999
$20,727
$19,962
$19,198
$20,440
$19,716
$18,993
$19,874
$19,223
$18,572
2018
$13,009
$12,718
$12,427
$14,484
$14,086
$13,688
$17,834
$17,147
$16,460
$17,543
$16,896
$16,249
$17,112
$16,530
$15,949
2019
$13,001
$12,712
$12,424
$14,466
$14,071
$13,676
$17,805
$17,123
$16,441
$17,520
$16,877
$16,235
$17,082
$16,504
$15,927
2020
$11,219
$10,955
$10,691
$12,541
$12,183
$11,825
$15,486
$14,867
$14,248
$15,198
$14,618
$14,038
$14,868
$14,347
$13,826
2021
$11,211
$10,949
$10,688
$12,525
$12,169
$11,813
$15,458
$14,844
$14,229
$15,177
$14,601
$14,025
$14,840
$14,322
$13,805
2022
$11,203
$10,944
$10,685
$12,508
$12,155
$11,802
$15,432
$14,822
$14,212
$15,156
$14,584
$14,012
$14,813
$14,299
$13,785
2023
$11,199
$10,941
$10,683
$12,498
$12,146
$11,795
$15,414
$14,807
$14,200
$15,142
$14,573
$14,003
$14,795
$14,283
$13,772
2024
$11,194
$10,937
$10,681
$12,488
$12,138
$11,788
$15,397
$14,793
$14,189
$15,129
$14,562
$13,995
$14,778
$14,268
$13,759
2025
$8,253
$8,055
$7,857
$9,245
$8,976
$8,707
$11,430
$10,965
$10,501
$11,206
$10,772
$10,337
$10,977
$10,587
$10,197
WR=weight reduction.
All costs are incremental to the baseline case.
 Table 1.2-11 Costs for Full Electric Vehicle Technology with 100 Mile Range, or EV100
                                         (2009$)
Vehicle Class
Subcompact
Applied
WR
10%
Net
WR
4%
2017
$18,157
2018
$15,513
2019
$15,502
2020
$13,385
2021
$13,375
2022
$13,365
2023
$13,359
2024
$13,352
2025
$9,850
                                          1-11

-------
                                                    2017 Draft Regulatory Impact Analysis
Subcompact
Subcompact
Small car
Small car
Small car
Large car
Large car
Large car
Minivan
Minivan
Minivan
Small truck
Small truck
Small truck
15%
20%
10%
15%
20%
10%
15%
20%
10%
15%
20%
10%
15%
20%
9%
14%
5%
10%
15%
5%
10%
15%
4%
9%
14%
2%
7%
12%
$17,733
$17,309
$20,332
$19,775
$19,219
$24,009
$23,170
$22,331
$24,740
$24,070
$23,400
$24,324
$23,698
$23,072
$15,135
$14,757
$17,433
$16,939
$16,446
$20,660
$19,911
$19,161
$21,235
$20,634
$20,032
$20,938
$20,378
$19,818
$15,126
$14,750
$17,412
$16,922
$16,431
$20,626
$19,881
$19,136
$21,207
$20,610
$20,013
$20,902
$20,346
$19,790
$13,046
$12,707
$15,090
$14,650
$14,210
$17,942
$17,269
$16,597
$18,398
$17,857
$17,316
$18,189
$17,686
$17,183
$13,038
$12,702
$15,071
$14,633
$14,196
$17,910
$17,242
$16,574
$18,372
$17,835
$17,298
$18,155
$17,656
$17,157
$13,031
$12,696
$15,052
$14,617
$14,182
$17,879
$17,215
$16,552
$18,346
$17,813
$17,280
$18,123
$17,627
$17,131
$13,025
$12,692
$15,040
$14,607
$14,173
$17,858
$17,198
$16,537
$18,330
$17,799
$17,269
$18,102
$17,608
$17,115
$13,021
$12,689
$15,028
$14,596
$14,165
$17,839
$17,181
$16,523
$18,314
$17,785
$17,257
$18,081
$17,590
$17,098
$9,596
$9,343
$11,122
$10,793
$10,464
$13,244
$12,740
$12,236
$13,566
$13,160
$12,754
$13,428
$13,051
$12,674
WR=weight reduction.
All costs are incremental to the baseline case.
 Table 1.2-12 Costs for Full Electric Vehicle Technology with 150 Mile Range, or EV150
                                           (2009$)
Vehicle Class
Subcompact
Small car
Large car
Minivan
Small truck
Applied
WR
20%
20%
20%
20%
20%
Net
WR
2%
3%
2%
2%
0%
2017
$23,714
$26,621
$32,589
$34,229
$33,589
2018
$20,242
$22,785
$27,974
$29,311
$28,831
2019
$20,230
$22,763
$27,936
$29,281
$28,793
2020
$17,450
$19,691
$24,239
$25,342
$24,980
2021
$17,439
$19,671
$24,204
$25,315
$24,944
2022
$17,428
$19,651
$24,170
$25,288
$24,910
2023
$17,421
$19,639
$24,148
$25,270
$24,887
2024
$17,414
$19,626
$24,127
$25,253
$24,865
2025
$12,836
$14,502
$17,873
$18,668
$18,419
WR=weight reduction.
All costs are incremental to the baseline case.
               Tablel.2-13 Costs for EV/PHEV In-home Chargers (2009$)
Technology
PHEV20
Charger
PHEV40
Charger
EV Charger
Charger
labor
Vehicle
Class
All
Subcompact
Small car
Larger car
Minivan
Small truck
All
All
2017
$78
$410
$476
$521
$521
$1,009
2018
$65
$344
$399
$437
$437
$1,009
2019
$65
$344
$399
$437
$437
$1,009
2020
$55
$291
$338
$369
$369
$1,009
2021
$55
$291
$338
$369
$369
$1,009
2022
$55
$291
$338
$369
$369
$1,009
2023
$55
$291
$338
$369
$369
$1,009
2024
$55
$291
$338
$369
$369
$1,009
2025
$41
$214
$249
$272
$272
$1,009
EV=electric vehicle; PHEV=plug-in electric vehicle; PHEV20=PHEV with 20 mile range; PHEV40=PHEV with 40
mile range.
All costs are incremental to the baseline case.
                                            1-12

-------
                                                 2017 Draft Regulatory Impact Analysis
    Table 1.2-14 Costs for 10% and 20% Weight Reduction for the 19 Vehicle Types"
                                         (2009$)
Vehicle
Type
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
Base
Weight
2615
2907
3316
3357
3711
4007
3535
3845
4398
4550
4784
4162
5169
5020
3598
4389
5271
4251
5269
Applied
WR
10%
20%
10%
20%
10%
20%
10%
20%
10%
20%
10%
20%
10%
20%
10%
20%
10%
20%
10%
20%
10%
20%
10%
20%
10%
20%
10%
20%
10%
20%
10%
20%
10%
20%
10%
20%
10%
20%
2017
$141
$628
$156
$698
$178
$797
$181
$807
$200
$892
$215
$963
$190
$849
$207
$924
$237
$1,057
$245
$1,093
$257
$1,149
$224
$1,000
$278
$1,242
$270
$1,206
$193
$865
$236
$1,055
$283
$1,267
$229
$1,021
$283
$1,266
2018
$137
$614
$153
$683
$174
$779
$176
$788
$195
$872
$210
$941
$185
$830
$202
$903
$231
$1,033
$239
$1,069
$251
$1,123
$218
$977
$271
$1,214
$263
$1,179
$189
$845
$230
$1,031
$277
$1,238
$223
$998
$276
$1,237
2019
$128
$600
$143
$667
$163
$761
$165
$771
$182
$852
$197
$920
$173
$812
$189
$883
$216
$1,010
$223
$1,045
$235
$1,098
$204
$955
$254
$1,187
$246
$1,153
$177
$826
$215
$1,008
$259
$1,210
$209
$976
$258
$1,210
2020
$125
$587
$139
$653
$159
$744
$161
$754
$178
$833
$192
$899
$169
$793
$184
$863
$210
$987
$218
$1,021
$229
$1,074
$199
$934
$247
$1,160
$240
$1,127
$172
$808
$210
$985
$252
$1,183
$203
$954
$252
$1,183
2021
$122
$574
$136
$638
$155
$728
$157
$737
$173
$815
$187
$880
$165
$776
$179
$844
$205
$965
$212
$999
$223
$1,050
$194
$914
$241
$1,135
$234
$1,102
$168
$790
$205
$964
$246
$1,157
$198
$933
$246
$1,157
2022
$119
$561
$132
$624
$151
$712
$153
$721
$169
$797
$182
$860
$161
$759
$175
$826
$200
$944
$207
$977
$218
$1,027
$189
$894
$235
$1,110
$228
$1,078
$164
$773
$200
$942
$240
$1,132
$193
$913
$240
$1,131
2023
$117
$553
$130
$615
$148
$702
$150
$710
$166
$785
$179
$848
$158
$748
$172
$814
$197
$931
$204
$963
$214
$1,012
$186
$881
$231
$1,094
$225
$1,062
$161
$761
$196
$929
$236
$1,115
$190
$900
$236
$1,115
2024
$115
$545
$128
$606
$146
$692
$148
$700
$163
$774
$176
$836
$156
$737
$169
$802
$194
$917
$200
$949
$211
$998
$183
$868
$227
$1,078
$221
$1,047
$158
$750
$193
$915
$232
$1,099
$187
$887
$232
$1,099
2025
$113
$495
$126
$550
$144
$627
$145
$635
$161
$702
$173
$758
$153
$669
$166
$727
$190
$832
$197
$861
$207
$905
$180
$787
$224
$978
$217
$949
$156
$680
$190
$830
$228
$997
$184
$804
$228
$996
    a See section 1.3 for details on the 19 vehicle types—what they are and how they are used.
    WR=weight reduction.
    All costs are incremental to the baseline case.
       Table 1.2-15 through Table 1.2-19 summarize the CC>2 reduction estimates of various
technologies which can be applied to cars and light-duty trucks. A more detailed discussion
of effectiveness is provided in Chapter 3 of the joint TSD.
                                           1-13

-------
                                                        2017 Draft Regulatory Impact Analysis
                        Table 1.2-15 Engine Technology Effectiveness
Technology
Low friction lubricants
Engine friction reduction level 1
Engine friction reduction level 2
Cylinder deactivation (includes imp. oil pump, if
applicable)
VVT - intake cam phasing
VVT - coupled cam phasing
VVT - dual cam phasing
Discrete VVLT
Continuous VVLT
Stoichiometric Gasoline Direct Injection
Turbo+downsize (incremental to GDI-S) (18-27 bar)*
Cooled Exhaust Gas Recirculation (incremental to 24
bar TRBDS+SGDI)
Advanced diesel engine (T2B2 emissions level)
Absolute CO2 Reduction (% from baseline vehicle)
Small Car
0.6
2.0
3.5
n.a.
2.1
4.1
4.1
4.1
5.1
1.5
10.8-16.6
3.6
19.5
Large Car
0.8
2.7
4.8
6.5
2.7
5.5
5.5
5.6
7.0
1.5
13.6-20.6
3.6
22.1
Minivan
0.7
2.6
4.5
6.0
2.5
5.1
5.1
5.2
6.5
1.5
12.9-19.6
3.6
21.5
Small
Truck
0.6
2.0
3.4
4.7
2.1
4.1
4.1
4.0
5.1
1.5
10.7-16.4
3.5
19.1
Large
Truck
0.7
2.4
4.2
5.7
2.4
4.9
4.9
4.9
6.1
1.5
12.3-18.8
3.6
21.3
        * Note: turbo downsize engine effectiveness does not include effectiveness of valvetrain improvements

                    Table 1.2-16 Transmission Technology Effectiveness
Technology
5-speed automatic (from 4-speed auto)
Aggressive shift logic 1
Aggressive shift logic 2
Early torque converter lockup
High Efficiency Gearbox
6-speed automatic (from 4-speed auto)
6-speed dry DCT (from 4-speed auto)
Absolute CO2 Reduction (% from baseline vehicle)
Small
Car
1.1
2.0
5.2
0.4
4.8
1.8
6.4
Large
Car
1.6
2.7
7.0
0.4
5.3
2.3
7.6
Minivan
1.4
2.5
6.6
0.4
5.1
2.2
7.2
Small
Truck
1.1
1.9
5.1
0.5
5.4
1.7
7.1
Large
Truck
1.4
2.4
6.2
0.5
4.3
2.1
8.1
                        Table 1.2-17 Hybrid Technology Effectiveness
Technology
12V Start-Stop
HV Mild Hybrid*
P2 Hybrid drivetrain**
Plug-in hybrid electric vehicle - 20 mile range***
Plug-in hybrid electric vehicle - 40 mile range***
Full electric vehicle (EV)
Absolute CO2 Reduction (% from baseline vehicle)
Small
Car
1.8
7.4
15.5
40
63
100
Large
Car
2.4
7.2
15.4
40
63
100
Minivan
2.2
6.9
14.6
40
63
n.a.
Small
Truck
1.8
6.8
13.4
40
63
n.a.
Large
Truck
2.2
8.0
15.7
n.a.
n.a.
n.a.
         ' Only includes the effectiveness related to the hybridized drivetrain (battery and electric motor) and supported
accessones.
        ** Only includes the effectiveness related to the hybridized drivetrain (battery and electric motor) and supported
accessories. Does not include advanced engine technologies. Will vary based on electric motor size; table values are based
on motor sizes in Ricardo vehicle simulation results (ref Joint TSD, Section 3.3.1)
                                                1-14

-------
                                               2017 Draft Regulatory Impact Analysis
         *Based on utility factors used for 20-mile (40%) and 40-mile (63%) range PHEV
                   Table 1.2-18 Accessory Technology Effectiveness
Technology
Improved high efficiency alternator & electrification of
accessories (12 volt)
Electric power steering
Improved high efficiency alternator & electrification of
accessories (42 volt)
Absolute CO2 Reduction (% from baseline vehicle)
Small
Car
1.7
1.5
3.3
Large
Car
1.3
1.1
2.5
Minivan
1.2
1.0
2.4
Small
Truck
1.3
1.2
2.6
Large
Truck
1.8
0.8
3.5
                 Table 1.2-19 Other Vehicle Technology Effectiveness
Technology
Aero drag reduction (20% on cars, 10% on trucks)
Low rolling resistance tires (20% on cars, 10% on trucks)
Low drag brakes
Secondary axle disconnect (unibody only)
Absolute CO2 Reduction (% from baseline vehicle)
Small
Car
4.7
3.9
0.8
1.3
Large
Car
4.7
3.9
0.8
1.3
Minivan
4.7
3.9
0.8
1.3
Small
Truck
4.7
3.9
0.8
1.3
Large
Truck
2.3
1.9
0.8
1.3
1.3 Vehicle Package Cost and Effectiveness

       Individual technologies can be used by manufacturers to achieve incremental CC>2
reductions. However, EPA believes that manufacturers are more likely to bundle
technologies into "packages" to capture synergistic aspects and reflect progressively larger
CC>2 reductions with additions or changes to any given package. In addition, manufacturers
typically apply new technologies in packages during model redesigns that occur
approximately once every five years, rather than adding new technologies one at a time on an
annual or biennial basis. This way, manufacturers can more efficiently make use of their
redesign  resources and more effectively plan for changes necessary to meet future standards.

       Therefore, the approach taken by EPA is to group technologies into packages of
increasing cost and effectiveness.  Costs for the packages are a sum total of the costs for the
technologies included. Effectiveness is somewhat more complex, as the effectiveness of
individual technologies cannot often be simply summed. To quantify the CC>2 (or fuel
consumption) effectiveness, EPA relies on its Lumped Parameter Model, which is described
in greater detail in the following section as well as in Chapter 3 of the draft joint TSD.

       As was done in the 2012-2016 rule and then updated in the 2010 TAR, EPA uses 19
different  vehicle types to represent the entire fleet in the OMEGA model. This was the result
of analyzing the existing light duty fleet with  respect to vehicle size and powertrain
configurations. All vehicles, including cars and trucks, were first distributed based on their
relative size, starting  from  compact cars and working upward to large trucks. Next, each
                                         1-15

-------
                                                2017 Draft Regulatory Impact Analysis
vehicle was evaluated for powertrain, specifically the engine size, 14, V6, and V8, then by
valvetrain configuration (DOHC, SOHC, OHV), and finally by the number of valves per
cylinder. For this analysis, EPA has used the same 19 vehicle types that were used in the 2010
TAR. EPA believes (at this time) that these 19 vehicle types broadly encompass the diversity
in the fleet as the analysis is appropriate for "average" vehicles.  EPA believes that modeling
each and every vehicle in the fleet individually is cumbersome and can even give a false sense
of accuracy in the analysis of a future fleet.  Each of these 19 vehicle types is mapped into one
of seven vehicle classes:  Subcompact, Small car, Large car, Minivan, Minivan with towing,
Small truck, and Large truck.  Note that, for the current assessment and representing an
update since the 2010 TAR, EPA has created a new vehicle class called "minivan-towing" or
a minivan (or MPV or Multi-Purpose Vehicle) with towing capability (as defined below)
which allows for greater differentiation of costs for this popular class of vehicles (such as the
Ford Edge, Honda Odyssey, Jeep Grand Cherokee).8 Note also that our seven vehicle classes
are not meant to correlate one-to-one with consumer-level vehicle classes.  For example, we
have many sport utility and cross-over utility vehicles  (SUVs and CUVs) in our "Minivan"
and "Minivan-towing" vehicle classes. We are not attempting to inappropriately mix
minivans with SUVs or equating the two in terms of features and utility, but we are  grouping
them with respect to technology effectiveness and some technology costs. The seven
OMEGA vehicle classes serve primarily to determine the effectiveness levels of new
technologies by determining which vehicle class is chosen within the lumped parameter
model (see sections 1.4 and 1.5 below).  So, any vehicle models mapped into a minivan-
towing vehicle type will get technology-specific effectiveness results for that vehicle class.
The same is true for vehicles mapped into the other vehicle classes.  Similarly, any vehicle
models mapped into a minivan-towing vehicle type will get technology-specific cost results
for that vehicle class.  The same is true for vehicles mapped into the other vehicle classes.
This is true only for applicable technologies, i.e., those costs developed on a vehicle class
basis such as advanced diesel, hybrid and other electrified powertrains (see Table 1.2-6
through Tablel.2-13 which show costs by vehicle class). Note that most technology costs are
not developed according to vehicle classes but are instead developed according to engine size,
valvetrain configuration, etc. (see Table 1.2-1 through Table 1.2-5 which show costs by
specific technology).  Lastly, note that these 19 vehicle types span the range of vehicle
footprints—smaller footprints for smaller vehicles and larger footprints for larger vehicles—
which served as the basis for the 2012-2016 GHG standards and the standards in this
proposal.  A detailed table showing the 19 vehicle types, their baseline engines, their
descriptions and some example models for each is  contained in Table 1.3-1.

         Table  1.3-1 List of 19 Vehicle Types  used to Model  the light-duty Fleet
Vehicle
Type*
1
2
Base Engine
1.5L4V
DOHC 14
2.4L 4V
DOHC 14
Base
Trans
4spAT
4spAT
Vehicle
Class
Subcompact
Small car
Description
Subcompact car 14
Compact car 14
Example Models
Ford Focus, Chevy
Aveo, Honda Fit
Chevy Cobalt,
Honda Civic, Mazda
Towing?
No
No
 Note the distinction between "vehicle type" and "vehicle class." We have the same 19 vehicle types as were
used in the 2010 TAR but have added a 7th vehicle class where the TAR used six.
                                         1-16

-------
                                               2017 Draft Regulatory Impact Analysis

3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19

2.4L 4V
DOHC 14
3.0L 4V
DOHC V6
3.3L4V
DOHC V6
4.5L 4V
DOHC V8
2.6L 4V
DOHC 14
(15)
3.7L 2V
SOHC V6
4.0L 2V
SOHC V6
4.7L 2V
SOHC V8
4.2L 2V
SOHC V6
3.8L2V
OHVV6
5.7L 2V
OHVV8
5.4L 3V
SOHC V8
5.7L 2V
OHVV8
3.5L4V
DOHC V6
4.6L 4V
DOHC V8
4.0L 4V
DOHC V6
5.6L 4V
DOHC V8

4spAT
4spAT
4spAT
4spAT
4spAT
4spAT
4spAT
4spAT
4spAT
4spAT
4spAT
4spAT
4spAT
4spAT
4spAT
4spAT
4spAT

Small car
Minivan
Large car
Large car
Minivan
Small truck
Minivan-
towing
Minivan-
towing
Large truck
Large truck
Large truck
Large truck
Large car
Minivan-
towing
Minivan-
towing
Large truck
Large truck

Midsize car/Small
MPVI4
Compact car/Small
MPVV6
Midsize/Large car V6
Midsize car/Large car
V8
Midsize MPV/Small
truck 14
Midsize MPV/Small
truck V6
Large MPV V6
Large MPV V8
Large truck/van V6
Large truck/MPV V6
Large truck/van V8
Large truck/van V8
Large car V8
Large MPV V6
Large MPV V8
Large truck/van V6
Large truck/van V8
3
Ford Fusion, Honda
Accord, Toyota
Camry
Dodge Caliber,
Subaru Impreza,
VWJetta
Dodge Avenger,
Ford Fusion, Honda
Accord
BMW 750, Ford
Mustang, Cadillac
STS
Jeep Compass, Ford
Escape, Nissan
Rogue
Jeep Liberty, Ford
Ranger
Dodge Durango,
Jeep Commander,
Ford Explorer
Dodge Durango,
Jeep Grand
Cherokee, Ford
F150
Dodge Ram 1500,
Ford F 150
Chrysler Town &
Country, Chevy
Silverado
Dodge Ram 1500,
Chevy Silverado
Ford Explorer, Ford
F150
Chrysler 300, Chevy
Corvette
Ford Edge, Chevy
Equinox, Honda
Odyssey
Jeep Grand
Cherokee, Nissan
Armada, VW
Touareg
Ford F 150, Nissan
Frontier, Toyota
Tacoma
Nissan Titan, Toyota
Tundra

No
No
No
No
No
No
Yes
Yes
Yes
Yes
Yes
Yes
No
Yes
Yes
Yes
Yes
       Note that we refer throughout this discussion of package building to a "baseline"
vehicle or a "baseline" package. This should not be confused with the baseline fleet, which is
the fleet of roughly 16 million 2008MY individual vehicles comprised of over 1,100 vehicle
models as described in Chapter 1 of the joint TSD. In this discussion, when we refer to
"baseline" vehicle we refer to the "baseline" configuration of the given vehicle type.  So, we
                                         1-17

-------
                                                2017 Draft Regulatory Impact Analysis
have 19 baseline vehicles in the context of building packages. Each of those 19 baseline
vehicles is equipped with a port fuel injected engine and a 4 speed automatic transmission.
The valvetrain configuration and the number of cylinders changes for each vehicle type in an
effort to cover the diversity in the 2008 baseline fleet as discussed above.  When we apply a
package of technologies to an individual vehicle model in the baseline fleet, we must first
determine which package-technologies are already present on the individual vehicle model.
From this information, we can determine the effectiveness and cost of the individual vehicle
model in the baseline fleet relative to the baseline vehicle that defines the vehicle type. Once
we have that, we can determine the incremental increase in effectiveness and cost for each
individual vehicle model in the baseline fleet once it has added the package of interest. This
process is known as the TEB-CEB process, which is short for Technology Effective Basis -
Cost Effective Basis. This process  allows us to accurately reflect the level of technology
already in the 2008 baseline fleet as well as the level of technology expected in the 2017-2025
reference case (i.e., the fleet as it is expected to exist as a result of the 2012-2016 final rule
which serves as the starting point for the larger analysis supporting this proposal).  But again,
the discussion here is focused solely on building packages. Therefore, while the baseline
vehicle that defines the vehicle type is relevant, the baseline and reference case fleets of real
vehicles are relevant to the discussion presented later in Chapter 3 of this draft RIA.

       Importantly, the effort in creating the packages attempts to maintain a constant utility
and acceleration performance for each package as compared to the baseline package.  As
such, each package is meant to provide equivalent driver-perceived performance to the
baseline package. There are two possible exceptions.  The first is the towing capability of
vehicle types which we have designated "non-towing." This requires a brief definition of
what we consider to be a towing vehicle versus a non-towing vehicle. Nearly all vehicles sold
today, with the exception of the smaller subcompact and compact cars, are able to tow up to
1,500 pounds provided the vehicle is equipped with a towing hitch. These vehicles require no
special OEM "towing package" of add-ons which typically include a set of more robust
brakes and some additional transmission cooling.  We do not consider such vehicles to be
towing vehicles. We reserve that term for those vehicles  capable of towing significantly more
than 1,500 Ibs.  For example, a base model Ford Escape can tow 1,500 pounds while the V6
equipped towing version can tow up to 3,500 pounds.  The former would not be considered a
true towing vehicle while the latter would. Note that all large trucks and most minivan
vehicle classes are considered towing vehicles in our analysis.

       The importance of this distinction can be found in the types of hybrid and plug-in
hybrid technologies we apply to towing versus non-towing vehicle types.0 For the towing
vehicle types, we apply a P2 hybrid technology with a turbocharged and downsized gasoline
direct injected engine.  These packages are expected to maintain equivalent towing capacity to
the baseline engine they replace.  For the non-towing vehicle types, we apply a P2 hybrid
technology with an Atkinson engine that has not been downsized relative to the baseline
engine.  The Atkinson engine, more correctly called the "Atkinson-cycle" engine, is used in
c This towing/non towing distinction is not an issue for non-HEVs, EPA maintains whatever towing capability
existed in the baseline when adding/substituting technology.


                                         1-18

-------
                                               2017 Draft Regulatory Impact Analysis
the current Toyota Prius and Ford Escape hybrid. We have maintained the original engine
size (i.e., no downsizing) to maintain utility as best as possible, but EPA acknowledges that
due to its lower power output, an Atkinson cycle engine cannot tow loads as well as a
standard Otto-cycle engine of the  same size. However, the presence of the hybrid powertrain
would be expected to maintain towing utility for these vehicle types in all but the most severe
operating extremes.  Such extremes would include towing in the Rocky Mountains (i.e, up
very long duration grades) or towing up Pike's Peak (i.e., up a shorter but very steep grade).
Under these extreme towing conditions, the battery on a hybrid powertrain would eventually
cease to provide sufficient supplemental power and the vehicle would be left with the
Atkinson engine doing all the work. A loss in utility would result (note that the loss in utility
should not result in breakdown or safety concerns, but rather loss in top speed and/or
acceleration capability). Importantly, those towing situations involving driving outside
mountainous regions  would not be affected.

      We do not address towing at the vehicle level.  Instead, we deal with towing at the
vehicle type level. As a result of the discretization of our vehicle types, we believe that some
towing vehicle models have been  mapped into non-towing vehicle types while some non-
towing vehicle models may have been mapped into towing vehicle types.  One prime example
is the Ford Escape mentioned above.  We have mapped all Escapes into non-towing vehicle
types. This is done because the primary driver behind the vehicle type into which a vehicle is
mapped is the engine technology in the base engine (number of cylinders, valvetrain
configuration, etc.).  Towing  capacity was not an original driver in the decision. Because of
this, our model outputs would put Atkinson-HEVs on some vehicle models that are more
properly treated as towing vehicles0, and would put turbocharged/downsized HEVs on some
vehicle models that are more properly treated as non-towing vehicles. Table 1.3-2 shows
some of these vehicle models that have been mapped into a non-towing vehicle type even
though they may be towing vehicles (the right column).  The table also shows some vehicle
models that have been mapped into a towing vehicle type even though they may not be
towing vehicles (the left column). The vehicles in the right column may experience some loss
of towing on a long grade for any that have been converted to Atkinson-HEV although they
would not have a lower tow rating. The vehicles in the left column may, when converted to
HEV, be costlier and slightly less  effective (less CO2 reduction) since they would be
converted to turbocharged/downsized HEVs rather than Atkinson-HEVs.  Accurate data on
towing specification is difficult to find, we hope to have better data on towing capacity for the
final rule analysis and may create new vehicle types to more properly model towing and non-
towing vehicles.
D The Ford Escape HEV does utilize an Atkinson engine and has a tow rating of 1,500 pounds which is identical
to the base 14 (non-HEV) Ford Escape.
                                        1-19

-------
                                              2017 Draft Regulatory Impact Analysis
    Table 1.3-2 Potential Inconsistencies in our Treatment of Towing & Non-towing
                                      Vehicles
Non-towing vehicles mapped into towing
vehicle types	
Towing vehicles mapped into non-towing
vehicle types	
Mercedes-Benz SLR
Ford Mustang
Buick Lacrosse/Lucerne
Chevrolet Impala
Pontiac G6/Grand Prix
Dodge Magnum V8
Ford Escape AWD V6
Jeep Liberty V6
Mercury Mariner AWD V6
Saturn Vue AWD V6
Honda Ridgeline 4WD V6
Hyundai Tuscon 4WD V6
Mazda Tribute AWD V6
Mitsubishi Outlander 4WD V6
Nissan Xterra V6
Subaru Forester AWD V6
Subaru Outback Wagon AWD V6
Suzuki Grand Vitara 4WD V6
Land Rover LR2 V6
Toyota Rav4 4WD V6	
       The second possible exception to our attempt at maintaining utility is the electric
vehicle range. We have built electric vehicle packages with ranges of 75, 100 and 150 miles.
Clearly these vehicles would not provide the same utility as a gasoline vehicle which typically
has a range of over 300 miles. However, from an acceleration performance standpoint, the
utility would be equal if not perhaps better. We believe that buyers of electric vehicles in the
2017-2025 timeframe will be purchasing the vehicles with a full understanding of the range
limitations and will not attempt to use their EVs for long duration towing trips.  As such, we
believe that the buyers will experience no practical loss of utility.

       To prepare inputs for the OMEGA model, EPA builds a "master-set" of technology
packages.  The master-set of packages for each vehicle type are meant to reflect the most
likely technology packages manufacturers would consider when determining their plans for
complying with future standards.  In other words, they are meant to reflect the most cost
effective groups of technologies—those that provide the best trade-off of costs versus fuel
consumption improvements.  This is done by grouping reasonable technologies in all possible
permutations and ranking those groupings based on the Technology Application Ranking
Factor (TARF). The TARF is the factor used by the OMEGA model to rank packages and
determine which are the most cost effective to  apply. The TARF is calculated as the net
incremental cost (or savings) of a package per kilogram of CO2 reduced by the package
relative to the previous package.  The net incremental cost is calculated as the incremental
cost of the technology package less the incremental discounted fuel savings of the package
over 5 years. The incremental CO2 reduction is calculated as the incremental CO2/mile
emission level of the package relative to the prior package multiplied by the lifetime miles
travelled. More detail on the TARF can be found in the OMEGA model supporting
                                        1-20

-------
                                                 2017 Draft Regulatory Impact Analysis
documentation (see EPA-420-B-10-042).  We also describe the TARF ranking process in
more detail below.  Grouping "reasonable technologies" simply means grouping those
technologies that are complementary (e.g., turbocharging plus downsizing) and not grouping
technologies that are not complementary (e.g., dual cam phasing and coupled cam phasing).

       To generate the master-set of packages for each of the vehicle types, EPA has built
packages in a step-wise fashion looking first at "simpler" conventional gasoline and vehicle
technologies, then more advanced gasoline technologies such as turbocharged (with very high
levels of boost) and downsized engines with gasoline direct injection and then hybrid and
other electrified vehicle technologies. This was done by presuming that auto makers would
first concentrate efforts on conventional gasoline engine and transmission technologies paired
with some level of mass reduction to improve fuel consumption. Mass reduction varied  from
no mass reduction up to 20 percent as the maximum considered in this analysis.E

       Once the conventional gasoline engine and transmission technologies have been fully
implemented, we expect that auto makers would apply more complex (and costly)
technologies such as the highly boosted (i.e. 24 bar and 27 bar brake mean effective pressure,
BMEP) gasoline engines and/or converting conventional gasoline engines to advanced diesel
engines in the next redesign cycle. The projected penetrations of these more advanced
technologies are presented in Chapter 3.8 of this draft RIA and the OMEGA model phase-in
caps are shown in Chapter 3.5 of the joint TSD.

       From there,  auto makers needing further technology penetration to meet their
individual standards would most likely move to hybridization. For this analysis, we have
built all of our hybrid packages using the newly emerging P2 technology. This technology
and why we believe it will be the predominant hybrid technology used in the 2017-2025
timeframe is described in Chapter 3 of the draft joint TSD.  As noted above, we have built
two types of P2 hybrid packages for analysis. The first type is for non-towing vehicle types
and uses an Atkinson-cycle engine with no downsizing relative  to the baseline engine. The
second P2 hybrid type is for towing vehicle types and uses a turbocharged and downsized
engine (rather than an Atkinson-cycle engine) to ensure no loss  of towing capacity.F
E Importantly, the mass reduction associated for each of the 19 vehicle types was based on the vehicle-type sales
weighted average curb weight. Although considerations of vehicle safety are an important part of EPA's
consideration in establishing the proposed standards, note that allowable weight reductions giving consideration
to safety is not part of the package building process so we have built packages for the full range of 0-20% weight
reduction considered in this analysis. Weight consideration for safety is handled within OMEGA as described in
Chapter 3 of this draft RIA.
F This is a departure from the 2010 TAR where we built several flavors of P2 HEV packages in the same manner
for each of the 19 vehicle types. We built P2 HEV packages with downsized engines, some with turbocharged
and downsized engines, some with cooled EGR, etc. We then used the TARF ranking process (described below)
to determine which packages were  most cost effective. We also did not, in the 2010 TAR, consider the weight
impacts of the hybrid powertrain, which we have done in this analysis. The effect of the changes used in this
analysis has been to decrease the effectiveness of HEV packages and to increase their costs since heavier
batteries and motors are now part of the packages.
                                          1-21

-------
                                                  2017 Draft Regulatory Impact Analysis
       Lastly, for some vehicle types (i.e., the non-towing vehicle types), we anticipate that
auto makers would move to more advanced electrification in the form of both plug-in hybrid
(PHEV, sometimes referred to as range extended electric vehicles (REEV)) and full battery
electric vehicles (EV).

       Importantly, the HEV, PHEV and EV (called collectively P/H/EV) packages take into
consideration the impact of the weight of the electrified components, primarily the battery
packs. Because these battery packs can be quite heavy, if one removes 20 percent of the mass
from a gasoline vehicle then converts it to an electric vehicle, the resultant net weight
reduction will be less than 20 percent. We discuss this in more below where we provide
additional discussion regarding the P/H/EV packages.

       Focusing first on the conventional and more  advanced (higher boost, cooled EGR)
gasoline packages, the first step in creating these packages was to consider the following 14
primary categories of conventional gasoline engine technologies. These are:

       1.   Our "anytime technologies" (ATT).G  These consist of low friction lubes, engine
           friction reduction, aggressive shift logic,  early torque converter lock-up (automatic
           transmission only).  ATT is broken into two levels:

              ATT, which consists of our first level of low friction lubes, engine friction
              reduction, aggressive shift logic, early torque converter lock-up (automatic
              transmission only) and low drag brakes.

              ATT2, which consists of our second level of low friction lubes and engine
              friction reduction (collectively referred to as EFR2), aggressive shift logic,
              along with the same early torque converter lock-up (automatic transmission
              only) and low drag brakes that are part of ATT.

       2.   Our "Other" conventional technologies.  These consist of improved accessories,
           electric power steering (EPS) or electrohydraulic power steering (EHPS, used for
           large trucks), aerodynamic improvements and lower rolling resistance tires.  The
           "other" technology category is broken into  two levels:
G Note that the term "anytime technology," is a carryover term from the 2012-2016 rule. At this point, we
continue to use the term, but it has become merely convenient nomenclature to denote very cost effective
technologies that are relatively easy to implement and would likely be implemented very early by auto makers
when considering compliance with CO2 standards. This is true also of the term "other" technologies.  We group
these technologies largely because they are very cost effective so will likely be implemented early in some form
and combination.  This grouping also serves to minimize the number of packages to be considered in our
modeling process. As explained in the text, we have built roughly 40,000 packages. Grouping the anytime and
other technologies together and treating them, essentially, as four technologies (ATT, ATT2, Otherl, Other2)
when building packages helps to keep the number of packages lower.  If we considered each "anytime" and
"other" technology separately, we would have had to build well over 200,000 packages which becomes
unworkable given the analytical tools at our disposal.
                                           1-22

-------
                                              2017 Draft Regulatory Impact Analysis
             Other 1, which consists of our first level of improved accessories, our first level
             of aerodynamic improvements and our first level of lower rolling resistance
             tires along with EPS/EHPS. This category also includes the high efficiency
             gearbox technology (HEG).

             Other2, which consists of our second levels of improved accessories,
             aerodynamic improvements and lower rolling resistance tires along with the
             same EPS/EHPS and HEG that are part of Otherl.

       3.  Variable valve timing (VVT) consisting of coupled cam phasing (CCP, for OHV
          and SOHC engines) and dual cam phasing (DCP, for DOHC engines)

       4.  Variable valve lift (VVL) consisting of discrete variable valve lift (DVVL, for
          DOHC engines)

       5.  Cylinder deactivation (Deac, considered for OHV and SOHC engines)

       6.  Gasoline direct injection (GDI)

       7.  Turbocharging and downsizing (TDS, which always includes a conversion to GDI)
          with and without cooled EGR.  Note that 27 bar BMEP engines must include the
          addition of cooled EGR in our  analysis.

       8.  Stop-start

       9.  Secondary axle disconnect (SAX)

       10. Conversion to advance diesel, which includes removal of the gasoline engine and
          gasoline fuel system and aftertreatment, and replacement by a diesel engine with
          diesel fuel system, a selective catalytic reduction (SCR) system  and advanced fuel
          and SCR controls.

       11. Mass reduction consisting of 0%, 5%, 10%, 15% and 20%.
       In this first step, we also considered the 6 primary transmission technologies. These
       are:

       12. 6 and 8 speed automatic transmissions (6sp AT/8sp AT)

       13.6 and 8 speed dual clutch transmissions with wet clutch (6sp wet-DCT/8sp wet-
          DCT)

       14. 6 and 8 speed dual clutch transmission with dry clutch (6sp dry-DCT/8sp wet-
          DCT)

In considering the transmissions, we had to first determine how each transmission could
reasonably be applied. DCTs, especially dry-DCTs, cannot be applied to every vehicle type
                                        1-23

-------
                                              2017 Draft Regulatory Impact Analysis
due to low end torque demands at launch (another example of how the proposed standards are
developed to preserve all vehicle utility). In addition, dry-DCTs tend to be more efficient
than wet-DCTs, which are more efficient than 6sp ATs primarily due to the elimination of wet
clutches and torque converter in the dry-DCT.  Further, each transmission has progressively
lower costs.  Therefore, moving from wet-DCT to dry-DCT will result in lower costs and
increased effectiveness.  Unlike the TAR analysis, we have limited towing vehicle types to
use of automatic transmissions (both 6 and 8 speed).  Like the TAR, we have added dry-DCTs
to vehicle types in baseline 14 engines and wet-DCTs to vehicle types with baseline V8
engines. This was done to ensure no loss of launch performance. For the V6 baseline vehicle
types, and again as was done in the 2010 TAR, we have added dry versus wet DCTs
depending on the baseline weight of the vehicle type. If the vehicle type were below 2,800
pounds curb weight, or removed enough weight in the package such that the package weight
would be below 2,800 pounds, we added a dry-DCT. Otherwise, we added a wet-DCT. In the
end, this allowed change from wet- to dry-DCT impacted only vehicle type 4 and only in
packages with 20% weight reduction applied.  Only then was this vehicle type light enough to
add the dry-DCT.
                                       1-24

-------
                                                 2017 Draft Regulatory Impact Analysis
       Table 1.3-3 shows the vehicle types, baseline curb weights and transmissions added in
this analysis.

       It is important to note that these heavier towing vehicles (including pickup trucks)
have no access to the more effective technologies such as Atkinson engine, dry-DCT
transmission, PHEV, or EV (as we describe below). Together these result in a decrease in
effectiveness potential for the heavier towing vehicle types compared to the non-towing
vehicle types. This discrepancy is one justification for the adjustment to the 2017-2025 truck
curves (in comparison to the 2012-2016 curves) as described in Chapter 2 of the draft joint
TSD and in preamble Section III.D.6.b.H
H While it's also offset by more mass reduction capability on these vehicles, the curve analysis did not assume an
uneven distribution of mass reduction throughout the fleet.


                                          1-25

-------
                                              2017 Draft Regulatory Impact Analysis
  Table 1.3-3 Application of Transmission Technologies in Building OMEGA Packages
Vehicle
Type
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
Vehicle
class
Subcompact
Small car
Small car
Minivan
Large car
Large car
Minivan
Small truck
Minivan-towing
Minivan-towing
Large truck
Large truck
Large truck
Large truck
Large car
Minivan-towing
Minivan-towing
Large truck
Large truck
Base
engine
14
14
14
V6
V6
V8
14
V6
V6
V8
V6
V6
V8
V8
V8
V6
V8
V6
V8
Base
weight
2615
2907
3316
3357
3711
4007
3535
3845
4398
4550
4784
4162
5169
5020
3598
4389
5271
4251
5269
Mass Reduction
0% 5% 10% 15% 20%
6/8 speed dry-DCT
6/8 speed dry-DCT
6/8 speed dry-DCT
6/8 speed wet-DCT 6/8 speed dry-DCT
6/8 speed wet-DCT
6/8 speed wet-DCT
6/8 speed dry-DCT
6/8 speed wet-DCT
6/8 speed AT
6/8 speed AT
6/8 speed AT
6/8 speed AT
6/8 speed AT
6/8 speed AT
6/8 speed wet-DCT
6/8 speed AT
6/8 speed AT
6/8 speed AT
6/8 speed AT
       For example, vehicle type 4 is equipped with a 4 speed automatic transmission in the
baseline. In a package consisting of a 0% to 15% mass reduction, we believe this vehicle type
could convert to a wet-DCT because the lighter weight results in reduced low end torque
demand thus making the wet-DCT feasible. Upon reaching 20% mass reduction, the vehicle
type could employ a dry-DCT because the even lighter weight (3357 less 20% equals 2686
pounds, which is less than our 2800 pound cutoff) results in further reduction in low end
torque demand.

       We start by first building a "preliminary-set" of non-electrified (i.e., gasoline and
diesel) packages for each vehicle type consisting of nearly every combination of each of the
14 primary engine technologies listed above.  The initial package for each vehicle type
represents what we expect a manufacturer will most likely implement as a first step on all
vehicles because the technologies included are so attractive from a cost effectiveness
standpoint. This package consists of ATT but no weight reduction or transmission changes.
We then add Otherl (less HEG), again with no weight reduction or transmission changes (we
do not consider the addition of HEG without a simultaneous improvement in the transmission
itself). The next package would add HEG and a transmission improvement. The subsequent
packages would iterate on nearly all possible combinations with the result being just under
2000 packages per vehicle type. Table 1.3-4 shows a subset of packages built for vehicle type
5, a midsized/large car with a 3.3L 4 valve DOHC V6 in the baseline.  These are package
built for the 2025 MY, so costs shown represent 2025 MY costs. Shown in this table are
packages built with 5% weight reduction only, and excluded are packages with an 8 speed
                                        1-26

-------
                                              2017 Draft Regulatory Impact Analysis
transmission.  So this table represents roughly one-tenth of the non-electrified packages built
for vehicle type 5.

 Table 1.3-4 A Subset of 2025 MY Non-HEV/PHEV/EV Packages Built for Vehicle Type
                       5 (Midsize/Large car 3.3L DOHC V6)a
Prelim
Pkg#
50000
50001
50002
50395b
50396
50397
50398
50399
50400
50401
50402
50403
50404
50405
50406
50407
50408
50409
50410
50411
50412
50413
50414
Weight
rdxn
base
base
base
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
Package contents
3.3L4VDOHCV6
4V DOHC V6 +LUB+EFR1+LDB+ASL
4V DOHC V6 +LUB+EFRl+LDB+ASL+IACC+EPS+Aerol+LRRTl
4V DOHC V6
+LUB+EFR 1 +LDB+ASL+IACC+EPS+Aero 1 +LRRT1 +HEG
4V DOHC V6
+LUB+EFR 1 +LDB+ASL+IACC+EPS+Aero 1 +LRRT1 +HEG+ DCP
4V DOHC V6
+LUB+EFR 1 +LDB+ASL+IACC+EPS+Aero 1 +LRRT1 +HEG+
DCP+DVVL
4V DOHC V6
+LUB+EFR 1 +LDB+ASL+IACC+EPS+Aero 1 +LRRT1 +HEG+
DCP+GDI
4V DOHC V6
+LUB+EFR 1 +LDB+ASL+IACC+EPS+Aero 1 +LRRT1 +HEG+
DCP+DVVL+GDI
4V DOHC V6
+LUB+EFR 1 +LDB+ASL+IACC+EPS+Aero 1 +LRRT1 +HEG+ DCP+S S
4V DOHC V6
+LUB+EFR 1 +LDB+ASL+IACC+EPS+Aero 1 +LRRT1 +HEG+
DCP+DVVL+SS
4V DOHC V6
+LUB+EFR 1 +LDB+ASL+IACC+EPS+Aero 1 +LRRT1 +HEG+
DCP+GDI+SS
4V DOHC V6
+LUB+EFR 1 +LDB+ASL+IACC+EPS+Aero 1 +LRRT1 +HEG+
DCP+DVVL+GDI+SS
4V DOHC V6
+LUB+EFR 1 +LDB+ASL+IACC+EPS+Aero 1 +LRRT1 +HEG+
DCP+SAX
4V DOHC V6
+LUB+EFR 1 +LDB+ASL+IACC+EPS+Aero 1 +LRRT1 +HEG+
DCP+DVVL+SAX
4V DOHC V6
+LUB+EFR 1 +LDB+ASL+IACC+EPS+Aero 1 +LRRT1 +HEG+
DCP+GDI+SAX
4V DOHC V6
+LUB+EFR 1 +LDB+ASL+IACC+EPS+Aero 1 +LRRT1 +HEG+
DCP+DVVL+GDI+SAX
4V DOHC V6
+LUB+EFR 1 +LDB+ASL+IACC+EPS+Aero 1 +LRRT1 +HEG+
DCP+SS+SAX
4V DOHC V6
+LUB+EFR 1 +LDB+ASL+IACC+EPS+Aero 1 +LRRT1 +HEG+
DCP+DVVL+SS+SAX
4V DOHC V6
+LUB+EFR 1 +LDB+ASL+IACC+EPS+Aero 1 +LRRT1 +HEG+
DCP+GDI+SS+SAX
4V DOHC V6
+LUB+EFR 1 +LDB+ASL+IACC+EPS+Aero 1 +LRRT1 +HEG+
DCP+DVVL+GDI+SS+SAX
4V DOHC 14
+LUB+EFR 1 +LDB+ASL+IACC+EPS+Aero 1 +LRRT1 +HEG+
DCP+GDI+TDS18
4V DOHC 14
+LUB+EFR 1 +LDB+ASL+IACC+EPS+Aero 1 +LRRT1 +HEG+
DCP+DVVL+GDI+TDS 1 8
4V DOHC 14
Transmission
4sp AT
4sp AT
4sp AT
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
2025MY
Cost
$0
$184
$394
$558
$723
$913
$1,060
$1,250
$1,066
$1,256
$1,404
$1,593
$804
$993
$1,141
$1,331
$1,147
$1,337
$1,484
$1,674
$998
$1,129
$1,341
CO2%
0%
7%
13%
24%
27%
29%
28%
30%
28%
30%
29%
31%
27%
29%
28%
30%
28%
30%
30%
31%
34%
35%
35%
                                       1-27

-------
      2017 Draft Regulatory Impact Analysis

50415
50416
50417
50418
50419
50420
50421
50422
50423
50424
50425
50426
50427
50428
50429
50430
50431
50432
50433
50434
50435
50436
50437
50438
50439
50440
50441

5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
+LUB+EFR 1 +LDB+ASL+IACC+EPS+Aero 1 +LRRT1 +HEG+
DCP+GDI+SS+TDS18
4V DOHC 14
+LUB+EFR 1 +LDB+ASL+IACC+EPS+Aero 1 +LRRT1 +HEG+
DCP+DVVL+GDI+S S+TDS 1 8
4V DOHC 14
+LUB+EFR 1 +LDB+ASL+IACC+EPS+Aero 1 +LRRT1 +HEG+
DCP+GDI+S AX+TDS 1 8
4V DOHC 14
+LUB+EFR 1 +LDB+ASL+IACC+EPS+Aero 1 +LRRT1 +HEG+
DCP+DVVL+GDI+S AX+TDS 1 8
4V DOHC 14
+LUB+EFR 1 +LDB+ASL+IACC+EPS+Aero 1 +LRRT1 +HEG+
DCP+GDI+S S+S AX+TDS 1 8
4V DOHC 14
+LUB+EFR 1 +LDB+ASL+IACC+EPS+Aero 1 +LRRT1 +HEG+
DCP+DVVL+GDI+SS+SAX+TDS18
4V DOHC 14
+LUB+EFR 1 +LDB+ASL+IACC+EPS+Aero 1 +LRRT1 +HEG+
DCP+GDI+TDS24
4V DOHC 14
+LUB+EFR 1 +LDB+ASL+IACC+EPS+Aero 1 +LRRT1 +HEG+
DCP+DVVL+GDI+TDS24
4V DOHC 14
+LUB+EFR 1 +LDB+ASL+IACC+EPS+Aero 1 +LRRT1 +HEG+
DCP+GDI+SS+TDS24
4V DOHC 14
+LUB+EFR 1 +LDB+ASL+IACC+EPS+Aero 1 +LRRT1 +HEG+
DCP+DVVL+GDI+S S+TDS24
4V DOHC 14
+LUB+EFR 1 +LDB+ASL+IACC+EPS+Aero 1 +LRRT1 +HEG+
DCP+GDI+SAX+TDS24
4V DOHC 14
+LUB+EFR 1 +LDB+ASL+IACC+EPS+Aero 1 +LRRT1 +HEG+
DCP+DVVL+GDI+SAX+TDS24
4V DOHC 14
+LUB+EFR 1 +LDB+ASL+IACC+EPS+Aero 1 +LRRT1 +HEG+
DCP+GDI+SS+SAX+TDS24
4V DOHC 14
+LUB+EFR 1 +LDB+ASL+IACC+EPS+Aero 1 +LRRT1 +HEG+
DCP+DVVL+GDI+SS+SAX+TDS24
4V DOHC V6 +EFR2+LDB+ASL2+IACC+EPS+Aerol+LRRTl+HEG
4V DOHC V6 +EFR2+LDB+ASL2+IACC+EPS+Aerol+LRRTl+HEG+
DCP
4V DOHC V6 +EFR2+LDB+ASL2+IACC+EPS+Aerol+LRRTl+HEG+
DCP+DVVL
4V DOHC V6 +EFR2+LDB+ASL2+IACC+EPS+Aerol+LRRTl+HEG+
DCP+GDI
4V DOHC V6 +EFR2+LDB+ASL2+IACC+EPS+Aerol+LRRTl+HEG+
DCP+DVVL+GDI
4V DOHC V6 +EFR2+LDB+ASL2+IACC+EPS+Aerol+LRRTl+HEG+
DCP+SS
4V DOHC V6 +EFR2+LDB+ASL2+IACC+EPS+Aerol+LRRTl+HEG+
DCP+DVVL+SS
4V DOHC V6 +EFR2+LDB+ASL2+IACC+EPS+Aerol+LRRTl+HEG+
DCP+GDI+S S
4V DOHC V6 +EFR2+LDB+ASL2+IACC+EPS+Aerol+LRRTl+HEG+
DCP+DVVL+GDI+S S
4V DOHC V6 +EFR2+LDB+ASL2+IACC+EPS+Aerol+LRRTl+HEG+
DCP+SAX
4V DOHC V6 +EFR2+LDB+ASL2+IACC+EPS+Aerol+LRRTl+HEG+
DCP+DVVL+SAX
4V DOHC V6 +EFR2+LDB+ASL2+IACC+EPS+Aerol+LRRTl+HEG+
DCP+GDI+SAX
4V DOHC V6 +EFR2+LDB+ASL2+IACC+EPS+Aerol+LRRTl+HEG+
DCP+DVVL+GDI+SAX
4V DOHC V6 +EFR2+LDB+ASL2+IACC+EPS+Aerol+LRRTl+HEG+
DCP+SS+SAX

6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet

$1,472
$1,079
$1,210
$1,422
$1,553
$1,210
$1,341
$1,554
$1,685
$1,291
$1,422
$1,634
$1,765
$646
$810
$1,000
$1,148
$1,338
$1,153
$1,343
$1,491
$1,681
$891
$1,081
$1,228
$1,418
$1,234

36%
35%
36%
36%
36%
37%
37%
38%
38%
38%
38%
38%
38%
27%
30%
32%
31%
33%
31%
33%
32%
34%
31%
33%
32%
34%
32%
1-28

-------
      2017 Draft Regulatory Impact Analysis
50442
50443
50444
50445
50446
50447
50448
50449
50450
50451
50452
50453
50454
50455
50456
50457
50458
50459
50460
50461
50462
50463
50464
50465
50466
50467
50468
50469
50470
50471
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
4V DOHC V6 +EFR2+LDB+ASL2+IACC+EPS+Aerol+LRRTl+HEG+
DCP+DVVL+SS+SAX
4V DOHC V6 +EFR2+LDB+ASL2+IACC+EPS+Aerol+LRRTl+HEG+
DCP+GDI+SS+SAX
4V DOHC V6 +EFR2+LDB+ASL2+IACC+EPS+Aerol+LRRTl+HEG+
DCP+DVVL+GDI+SS+SAX
4V DOHC 14 +EFR2+LDB+ASL2+IACC+EPS+Aerol+LRRTl+HEG+
DCP+GDI+TDS18
4V DOHC 14 +EFR2+LDB+ASL2+IACC+EPS+Aerol+LRRTl+HEG+
DCP+DVVL+GDI+TDS 1 8
4V DOHC 14 +EFR2+LDB+ASL2+IACC+EPS+Aerol+LRRTl+HEG+
DCP+GDI+SS+TDS18
4V DOHC 14 +EFR2+LDB+ASL2+IACC+EPS+Aerol+LRRTl+HEG+
DCP+DVVL+GDI+S S+TDS 1 8
4V DOHC 14 +EFR2+LDB+ASL2+IACC+EPS+Aerol+LRRTl+HEG+
DCP+GDI+S AX+TDS 1 8
4V DOHC 14 +EFR2+LDB+ASL2+IACC+EPS+Aerol+LRRTl+HEG+
DCP+DVVL+GDI+S AX+TDS 1 8
4V DOHC 14 +EFR2+LDB+ASL2+IACC+EPS+Aerol+LRRTl+HEG+
DCP+GDI+S S+S AX+TDS 1 8
4V DOHC 14 +EFR2+LDB+ASL2+IACC+EPS+Aerol+LRRTl+HEG+
DCP+DVVL+GDI+SS+SAX+TDS18
4V DOHC 14 +EFR2+LDB+ASL2+IACC+EPS+Aerol+LRRTl+HEG+
DCP+GDI+TDS24
4V DOHC 14 +EFR2+LDB+ASL2+IACC+EPS+Aerol+LRRTl+HEG+
DCP+DVVL+GDI+TDS24
4V DOHC 14 +EFR2+LDB+ASL2+IACC+EPS+Aerol+LRRTl+HEG+
DCP+GDI+SS+TDS24
4V DOHC 14 +EFR2+LDB+ASL2+IACC+EPS+Aerol+LRRTl+HEG+
DCP+DVVL+GDI+S S+TDS24
4V DOHC 14 +EFR2+LDB+ASL2+IACC+EPS+Aerol+LRRTl+HEG+
DCP+GDI+SAX+TDS24
4V DOHC 14 +EFR2+LDB+ASL2+IACC+EPS+Aerol+LRRTl+HEG+
DCP+DVVL+GDI+SAX+TDS24
4V DOHC 14 +EFR2+LDB+ASL2+IACC+EPS+Aerol+LRRTl+HEG+
DCP+GDI+SS+SAX+TDS24
4V DOHC 14 +EFR2+LDB+ASL2+IACC+EPS+Aerol+LRRTl+HEG+
DCP+DVVL+GDI+SS+SAX+TDS24
4V DOHC V6
+LUB+EFRl+LDB+ASL+IACC2+EPS+Aero2+LRRT2+HEG
4V DOHC V6
+LUB+EFRl+LDB+ASL+IACC2+EPS+Aero2+LRRT2+HEG+DCP
4V DOHC V6
+LUB+EFRl+LDB+ASL+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+DVVL
4V DOHC V6
+LUB+EFRl+LDB+ASL+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+GDI
4V DOHC V6
+LUB+EFRl+LDB+ASL+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+DVVL+GDI
4V DOHC V6
+LUB+EFRl+LDB+ASL+IACC2+EPS+Aero2+LRRT2+HEG+DCP+SS
4V DOHC V6
+LUB+EFRl+LDB+ASL+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+DVVL+SS
4V DOHC V6
+LUB+EFRl+LDB+ASL+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+GDI+S S
4V DOHC V6
+LUB+EFRl+LDB+ASL+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+DVVL+GDI+S S
4V DOHC V6
+LUB+EFRl+LDB+ASL+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+SAX
4V DOHC V6
+LUB+EFRl+LDB+ASL+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+DVVL+SAX
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
$1,424
$1,571
$1,761
$1,058
$1,189
$1,401
$1,532
$1,138
$1,269
$1,481
$1,612
$1,270
$1,401
$1,613
$1,744
$1,351
$1,482
$1,694
$1,825
$772
$937
$1,127
$1,274
$1,464
$1,280
$1,470
$1,618
$1,807
$1,017
$1,207
34%
33%
35%
37%
38%
38%
39%
38%
39%
39%
39%
40%
40%
40%
41%
40%
40%
41%
41%
29%
31%
33%
32%
34%
32%
34%
33%
35%
32%
33%
1-29

-------
      2017 Draft Regulatory Impact Analysis
50472
50473
50474
50475
50476
50477
50478
50479
50480
50481
50482
50483
50484
50485
50486
50487
50488
50489
50490
50491
50492
50493
50494
50495
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
4V DOHC V6
+LUB+EFRl+LDB+ASL+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+GDI+SAX
4V DOHC V6
+LUB+EFRl+LDB+ASL+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+DVVL+GDI+SAX
4V DOHC V6
+LUB+EFRl+LDB+ASL+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+SS+SAX
4V DOHC V6
+LUB+EFRl+LDB+ASL+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+DVVL+SS+SAX
4V DOHC V6
+LUB+EFRl+LDB+ASL+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+GDI+SS+SAX
4V DOHC V6
+LUB+EFRl+LDB+ASL+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+DVVL+GDI+SS+SAX
4V DOHC 14
+LUB+EFRl+LDB+ASL+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+GDI+TDS18
4V DOHC 14
+LUB+EFRl+LDB+ASL+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+DVVL+GDI+TDS 1 8
4V DOHC 14
+LUB+EFRl+LDB+ASL+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+GDI+SS+TDS18
4V DOHC 14
+LUB+EFRl+LDB+ASL+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+DVVL+GDI+S S+TDS 1 8
4V DOHC 14
+LUB+EFRl+LDB+ASL+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+GDI+S AX+TDS 1 8
4V DOHC 14
+LUB+EFRl+LDB+ASL+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+DVVL+GDI+S AX+TDS 1 8
4V DOHC 14
+LUB+EFRl+LDB+ASL+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+GDI+S S+S AX+TDS 1 8
4V DOHC 14
+LUB+EFRl+LDB+ASL+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+DVVL+GDI+SS+SAX+TDS18
4V DOHC 14
+LUB+EFRl+LDB+ASL+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+GDI+TDS24
4V DOHC 14
+LUB+EFRl+LDB+ASL+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+DVVL+GDI+TDS24
4V DOHC 14
+LUB+EFRl+LDB+ASL+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+GDI+SS+TDS24
4V DOHC 14
+LUB+EFRl+LDB+ASL+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+DVVL+GDI+S S+TDS24
4V DOHC 14
+LUB+EFRl+LDB+ASL+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+GDI+SAX+TDS24
4V DOHC 14
+LUB+EFRl+LDB+ASL+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+DVVL+GDI+SAX+TDS24
4V DOHC 14
+LUB+EFRl+LDB+ASL+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+GDI+SS+SAX+TDS24
4V DOHC 14
+LUB+EFRl+LDB+ASL+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+DVVL+GDI+SS+SAX+TDS24
4V DOHC V6 +EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG
4V DOHC V6
+EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+DCP
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
$1,355
$1,545
$1,361
$1,550
$1,698
$1,888
$1,212
$1,343
$1,555
$1,686
$1,293
$1,424
$1,636
$1,767
$1,424
$1,555
$1,768
$1,899
$1,505
$1,636
$1,848
$1,979
$860
$1,024
33%
34%
33%
34%
34%
35%
38%
39%
39%
39%
38%
39%
39%
40%
40%
40%
41%
41%
41%
41%
41%
41%
32%
35%
1-30

-------
      2017 Draft Regulatory Impact Analysis
50496
50497
50498
50499
50500
50501
50502
50503
50504
50505
50506
50507
50508
50509
50510
50511
50512
50513
50514
50515
50516
50517
50518
50519
50520
50521
50522
50523
50524
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
4V DOHC V6
+EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+DCP+DVVL
4V DOHC V6
+EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+DCP+GDI
4V DOHC V6
+EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+DVVL+GDI
4V DOHC V6
+EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+DCP+SS
4V DOHC V6
+EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+DVVL+SS
4V DOHC V6
+EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+GDI+SS
4V DOHC V6
+EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+DVVL+GDI+S S
4V DOHC V6
+EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+DCP+SAX
4V DOHC V6
+EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+DVVL+SAX
4V DOHC V6
+EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+GDI+SAX
4V DOHC V6
+EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+DVVL+GDI+SAX
4V DOHC V6
+EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+SS+SAX
4V DOHC V6
+EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+DVVL+SS+SAX
4V DOHC V6
+EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+GDI+SS+SAX
4V DOHC V6
+EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+DVVL+GDI+SS+SAX
4V DOHC 14 +EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+GDI+TDS18
4V DOHC 14 +EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+DVVL+GDI+TDS 1 8
4V DOHC 14 +EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+GDI+SS+TDS18
4V DOHC 14 +EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+DVVL+GDI+S S+TDS 1 8
4V DOHC 14 +EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+GDI+S AX+TDS 1 8
4V DOHC 14 +EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+DVVL+GDI+S AX+TDS 1 8
4V DOHC 14 +EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+GDI+S S+S AX+TDS 1 8
4V DOHC 14 +EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+DVVL+GDI+SS+SAX+TDS18
4V DOHC 14 +EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+GDI+TDS24
4V DOHC 14 +EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+DVVL+GDI+TDS24
4V DOHC 14 +EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+GDI+SS+TDS24
4V DOHC 14 +EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+DVVL+GDI+S S+TDS24
4V DOHC 14 +EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+GDI+SAX+TDS24
4V DOHC 14 +EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
$1,214
$1,362
$1,552
$1,367
$1,557
$1,705
$1,895
$1,105
$1,295
$1,442
$1,632
$1,448
$1,638
$1,785
$1,975
$1,272
$1,403
$1,615
$1,746
$1,352
$1,483
$1,695
$1,826
$1,484
$1,615
$1,827
$1,958
$1,565
$1,696
36%
35%
37%
35%
37%
36%
38%
35%
37%
36%
38%
36%
37%
37%
38%
41%
41%
41%
42%
41%
42%
42%
42%
43%
43%
43%
44%
43%
43%
1-31

-------
      2017 Draft Regulatory Impact Analysis

50525
50526
50527
50528
50529
50530
50531
50532
50533
50534
50535
50536
50537
50538
50539
50540
50541
50542
50543
50544
50545
50546
50547
50548
50549
50550

5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
DCP+DVVL+GDI+SAX+TDS24
4V DOHC 14 +EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+GDI+SS+SAX+TDS24
4V DOHC 14 +EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+DVVL+GDI+SS+SAX+TDS24
4V DOHC 14
+LUB+EFR 1 +LDB+ASL+IACC+EPS+Aero 1 +LRRT1 +HEG+
DCP+GDI+TDS24+EGR
4V DOHC 14
+LUB+EFR 1 +LDB+ASL+IACC+EPS+Aero 1 +LRRT1 +HEG+
DCP+DVVL+GDI+TDS24+EGR
4V DOHC 14
+LUB+EFR 1 +LDB+ASL+IACC+EPS+Aero 1 +LRRT1 +HEG+
DCP+GDI+SS+TDS24+EGR
4V DOHC 14
+LUB+EFR 1 +LDB+ASL+IACC+EPS+Aero 1 +LRRT1 +HEG+
DCP+DVVL+GDI+S S+TDS24+EGR
4V DOHC 14
+LUB+EFR 1 +LDB+ASL+IACC+EPS+Aero 1 +LRRT1 +HEG+
DCP+GDI+SAX+TDS24+EGR
4V DOHC 14
+LUB+EFR 1 +LDB+ASL+IACC+EPS+Aero 1 +LRRT1 +HEG+
DCP+DVVL+GDI+SAX+TDS24+EGR
4V DOHC 14
+LUB+EFR 1 +LDB+ASL+IACC+EPS+Aero 1 +LRRT1 +HEG+
DCP+GDI+SS+SAX+TDS24+EGR
4V DOHC 14
+LUB+EFR 1 +LDB+ASL+IACC+EPS+Aero 1 +LRRT1 +HEG+
DCP+DVVL+GDI+SS+SAX+TDS24+EGR
4V DOHC 14 +EFR2+LDB+ASL2+IACC+EPS+Aerol+LRRTl+HEG+
DCP+GDI+TDS24+EGR
4V DOHC 14 +EFR2+LDB+ASL2+IACC+EPS+Aerol+LRRTl+HEG+
DCP+DVVL+GDI+TDS24+EGR
4V DOHC 14 +EFR2+LDB+ASL2+IACC+EPS+Aerol+LRRTl+HEG+
DCP+GDI+SS+TDS24+EGR
4V DOHC 14 +EFR2+LDB+ASL2+IACC+EPS+Aerol+LRRTl+HEG+
DCP+DVVL+GDI+S S+TDS24+EGR
4V DOHC 14 +EFR2+LDB+ASL2+IACC+EPS+Aerol+LRRTl+HEG+
DCP+GDI+SAX+TDS24+EGR
4V DOHC 14 +EFR2+LDB+ASL2+IACC+EPS+Aerol+LRRTl+HEG+
DCP+DVVL+GDI+SAX+TDS24+EGR
4V DOHC 14 +EFR2+LDB+ASL2+IACC+EPS+Aerol+LRRTl+HEG+
DCP+GDI+SS+SAX+TDS24+EGR
4V DOHC 14 +EFR2+LDB+ASL2+IACC+EPS+Aerol+LRRTl+HEG+
DCP+DVVL+GDI+SS+SAX+TDS24+EGR
4V DOHC 14
+LUB+EFRl+LDB+ASL+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+GDI+TDS24+EGR
4V DOHC 14
+LUB+EFRl+LDB+ASL+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+DVVL+GDI+TDS24+EGR
4V DOHC 14
+LUB+EFRl+LDB+ASL+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+GDI+SS+TDS24+EGR
4V DOHC 14
+LUB+EFRl+LDB+ASL+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+DVVL+GDI+S S+TDS24+EGR
4V DOHC 14
+LUB+EFRl+LDB+ASL+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+GDI+SAX+TDS24+EGR
4V DOHC 14
+LUB+EFRl+LDB+ASL+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+DVVL+GDI+SAX+TDS24+EGR
4V DOHC 14
+LUB+EFRl+LDB+ASL+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+GDI+SS+SAX+TDS24+EGR
4V DOHC 14
+LUB+EFRl+LDB+ASL+IACC2+EPS+Aero2+LRRT2+HEG+

6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet

$1,908
$2,039
$1,457
$1,588
$1,801
$1,932
$1,538
$1,669
$1,881
$2,012
$1,517
$1,648
$1,860
$1,991
$1,598
$1,729
$1,941
$2,072
$1,671
$1,802
$2,015
$2,146
$1,752
$1,883
$2,095
$2,226

44%
44%
39%
39%
40%
40%
40%
40%
40%
41%
42%
42%
43%
43%
42%
43%
43%
43%
42%
42%
43%
43%
43%
43%
43%
44%
1-32

-------
      2017 Draft Regulatory Impact Analysis

50551
50552
50553
50554
50555
50556
50557
50558
50559
50560
50561
50562
50563
50564
50565
50566
50567
50568
50569
50570
50571
50572
50573
50574
50575
50576
50577
50578

5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
DCP+DVVL+GDI+SS+SAX+TDS24+EGR
4V DOHC 14 +EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+GDI+TDS24+EGR
4V DOHC 14 +EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+DVVL+GDI+TDS24+EGR
4V DOHC 14 +EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+GDI+SS+TDS24+EGR
4V DOHC 14 +EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+DVVL+GDI+S S+TDS24+EGR
4V DOHC 14 +EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+GDI+SAX+TDS24+EGR
4V DOHC 14 +EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+DVVL+GDI+SAX+TDS24+EGR
4V DOHC 14 +EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+GDI+SS+SAX+TDS24+EGR
4V DOHC 14 +EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+DVVL+GDI+SS+SAX+TDS24+EGR
4V DOHC 14
+LUB+EFR 1 +LDB+ASL+IACC+EPS+Aero 1 +LRRT1 +HEG+
DCP+GDI+TDS27+EGR
4V DOHC 14
+LUB+EFR 1 +LDB+ASL+IACC+EPS+Aero 1 +LRRT1 +HEG+
DCP+DVVL+GDI+TDS27+EGR
4V DOHC 14
+LUB+EFR 1 +LDB+ASL+IACC+EPS+Aero 1 +LRRT1 +HEG+
DCP+GDI+SS+TDS27+EGR
4V DOHC 14
+LUB+EFR 1 +LDB+ASL+IACC+EPS+Aero 1 +LRRT1 +HEG+
DCP+DVVL+GDI+S S+TDS27+EGR
4V DOHC 14
+LUB+EFR 1 +LDB+ASL+IACC+EPS+Aero 1 +LRRT1 +HEG+
DCP+GDI+SAX+TDS27+EGR
4V DOHC 14
+LUB+EFR 1 +LDB+ASL+IACC+EPS+Aero 1 +LRRT1 +HEG+
DCP+DVVL+GDI+SAX+TDS27+EGR
4V DOHC 14
+LUB+EFR 1 +LDB+ASL+IACC+EPS+Aero 1 +LRRT1 +HEG+
DCP+GDI+SS+SAX+TDS27+EGR
4V DOHC 14
+LUB+EFR 1 +LDB+ASL+IACC+EPS+Aero 1 +LRRT1 +HEG+
DCP+DVVL+GDI+SS+SAX+TDS27+EGR
4V DOHC 14 +EFR2+LDB+ASL2+IACC+EPS+Aerol+LRRTl+HEG+
DCP+GDI+TDS27+EGR
4V DOHC 14 +EFR2+LDB+ASL2+IACC+EPS+Aerol+LRRTl+HEG+
DCP+DVVL+GDI+TDS27+EGR
4V DOHC 14 +EFR2+LDB+ASL2+IACC+EPS+Aerol+LRRTl+HEG+
DCP+GDI+SS+TDS27+EGR
4V DOHC 14 +EFR2+LDB+ASL2+IACC+EPS+Aerol+LRRTl+HEG+
DCP+DVVL+GDI+S S+TDS27+EGR
4V DOHC 14 +EFR2+LDB+ASL2+IACC+EPS+Aerol+LRRTl+HEG+
DCP+GDI+SAX+TDS27+EGR
4V DOHC 14 +EFR2+LDB+ASL2+IACC+EPS+Aerol+LRRTl+HEG+
DCP+DVVL+GDI+SAX+TDS27+EGR
4V DOHC 14 +EFR2+LDB+ASL2+IACC+EPS+Aerol+LRRTl+HEG+
DCP+GDI+SS+SAX+TDS27+EGR
4V DOHC 14 +EFR2+LDB+ASL2+IACC+EPS+Aerol+LRRTl+HEG+
DCP+DVVL+GDI+SS+SAX+TDS27+EGR
4V DOHC 14
+LUB+EFRl+LDB+ASL+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+GDI+TDS27+EGR
4V DOHC 14
+LUB+EFRl+LDB+ASL+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+DVVL+GDI+TDS27+EGR
4V DOHC 14
+LUB+EFRl+LDB+ASL+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+GDI+SS+TDS27+EGR
4V DOHC 14
+LUB+EFRl+LDB+ASL+IACC2+EPS+Aero2+LRRT2+HEG+

6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet

$1,731
$1,862
$2,074
$2,205
$1,811
$1,942
$2,155
$2,286
$1,882
$2,013
$2,225
$2,356
$1,963
$2,094
$2,306
$2,437
$1,942
$2,073
$2,285
$2,416
$2,022
$2,153
$2,365
$2,496
$2,096
$2,227
$2,439
$2,570

45%
45%
46%
46%
45%
45%
46%
46%
40%
40%
41%
41%
40%
40%
41%
41%
43%
43%
43%
43%
43%
43%
44%
44%
43%
43%
44%
44%
1-33

-------
                                                 2017 Draft Regulatory Impact Analysis

50579
50580
50581
50582
50583
50584
50585
50586
50587
50588
50589
50590
51963
51964
51965
51966

5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
5%
DCP+DVVL+GDI+S S+TDS27+EGR
4V DOHC 14
+LUB+EFRl+LDB+ASL+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+GDI+SAX+TDS27+EGR
4V DOHC 14
+LUB+EFRl+LDB+ASL+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+DVVL+GDI+SAX+TDS27+EGR
4V DOHC 14
+LUB+EFRl+LDB+ASL+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+GDI+SS+SAX+TDS27+EGR
4V DOHC 14
+LUB+EFRl+LDB+ASL+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+DVVL+GDI+SS+SAX+TDS27+EGR
4V DOHC 14 +EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+GDI+TDS27+EGR
4V DOHC 14 +EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+DVVL+GDI+TDS27+EGR
4V DOHC 14 +EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+GDI+SS+TDS27+EGR
4V DOHC 14 +EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+DVVL+GDI+S S+TDS27+EGR
4V DOHC 14 +EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+GDI+SAX+TDS27+EGR
4V DOHC 14 +EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+DVVL+GDI+SAX+TDS27+EGR
4V DOHC 14 +EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+GDI+SS+SAX+TDS27+EGR
4V DOHC 14 +EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+DVVL+GDI+SS+SAX+TDS27+EGR
4V DOHC 14 +EFR2+LDB+ASL2+IACC+EPS+Aerol+LRRTl+HEG+
DCP+DSL-Adv
4V DOHC 14 +EFR2+LDB+ASL2+IACC+EPS+Aerol+LRRTl+HEG+
DCP+DSL-Adv+SAX
4V DOHC 14 +EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+DSL-Adv
4V DOHC 14 +EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+DSL-Adv+SAX

6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet

$2,177
$2,308
$2,520
$2,651
$2,156
$2,287
$2,499
$2,630
$2,236
$2,367
$2,579
$2,710
$3,602
$3,683
$3,816
$3,896

43%
43%
44%
44%
45%
45%
46%
46%
46%
46%
46%
46%
40%
40%
43%
43%
a Excludes packages with weight reduction of 0%, 10%, 15%, 20% and Sspeed DCT.
b The jump from package # 50002 to 50395 represents the packages built with 0% weight reduction
which are intentionally not included in the table.

       As stated, this preliminary-set of packages is meant to maintain utility relative to the
baseline vehicle. Having built nearly 2000 packages for each vehicle type suggests question
"how can EPA know that each has the same utility as the baseline vehicle for a given vehicle
type?" We believe that this is inherent in the effectiveness values used, given that they are
based on the recent Ricardo work which had maintenance of baseline performance as a
constraint in estimating effectiveness values. Maintaining utility is also included in the cost
of the technologies with proper consideration of engine sizing (number of cylinders), motor
and battery sizing, etc. This is discussed in more detail in Section 3.3.1.11 of the draftjoint
TSD. Therefore, with the possible exception of the towing issue raised above—maintenance
of towing capacity over operating extremes for "non-towing" vehicles—we are confident that
the packages we have built for OMEGA modeling maintain utility relative to the baseline for
the "average" vehicles represented by our 19 vehicle types.

       This preliminary-set of conventional gasoline packages (roughly 2000 packages per
vehicle type) was then ranked within vehicle type by TARF. This is done by first calculating
the TARF of each package relative to the baseline package. The package with the best TARF
is selected as OMEGA package #1 (or, more accurately, #501 for vehicle type 5, #101 for
vehicle type 1, etc.). The remaining packages for vehicle type 5 are then ranked again by
                                          1-34

-------
                                              2017 Draft Regulatory Impact Analysis
TARF, this time relative to OMEGA package #501.  The best package is selected as OMEGA
package #502, etc.  Table 1.3-6 illustrates this process, while Table 1.3-5 presents 2008
baseline data used in the TARF ranking process.

   Table 1.3-5 Lifetime VMT & Baseline CO2 used for TARF Ranking in the Package
                                  Building Process
Vehicle
Type
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
Vehicle
class
Subcompact
Small car
Small car
Minivan
Large car
Large car
Minivan
Small truck
Minivan-towing
Minivan-towing
Large truck
Large truck
Large truck
Large truck
Large car
Minivan-towing
Minivan-towing
Large truck
Large truck
Base
engine
14
14
14
V6
V6
V8
14
V6
V6
V8
V6
V6
V8
V8
V8
V6
V8
V6
V8
Car/
Truck3
C
C
C
C
C
C
T
C
T
T
T
T
T
T
C
T
T
T
T
Lifetime
VMT
190,971
190,971
190,971
190,971
190,971
190,971
221,199
190,971
221,199
221,199
221,199
221,199
221,199
221,199
190,971
221,199
221,199
221,199
221,199
Base CO2
(g/mi)b
241.0
236.8
274.2
310.5
335.5
387.7
309.9
385.1
421.7
437.5
422.5
357.6
447.7
480.0
386.7
375.6
463.3
403.0
477.6
      a Designation here matters only for lifetime VMT determination in the package building and ranking process.
      b Sales weighted CO2 within vehicle type.
1 Woldring, D., Landenfeld, T., Christie, M.J., 2007, "DI Boost: Application of a High
Performance Gasoline Direct Injection Concept." SAE Technical Paper Series No. 2007-01-
1410; Kapus, P.E., Fraidl, G.K., Prevedel, K., Fuerhapter, A., 2007, "GDI Turbo - The Next
Steps." JSAE Technical Paper No. 20075355; Hancock, D., Fraser, N., Jeremy, M., Sykes, R.,
Blaxill, H., 2008, "A New 3 Cylinder 1.21 Advanced Downsizing Technology Demonstrator
Engine." SAE Technical Paper Series No. 2008-01-0611; Lumsden, G., OudeNijeweme, D.,
Fraser, N. Blaxill, H., 2009, "Development of a Turbocharged Direct Injection Downsizing
Demonstrator Engine." SAE Technical Paper Series No. 2009-01-1503; Cruff, L., Kaiser, M.,
Krause,  S., Harris, R., Krueger, U., Williams, M., 2010, "EBDI® - Application of a Fully
Flexible High Bmep Downsized Spark Ignited Engine." SAE Technical Paper Series No.
2010-01-0587; Taylor, J., Fraser, N., Wieske, P., 2010, "Water Cooled Exhaust Manifold and
Full Load EGR Technology Applied to a Downsized Direct Injection Spark Ignition Engine."
SAE Technical Paper Series No. 2010-01-0356; Roth, D.B., Keller, P, Becker, M., 2010,
"Requirements of External EGRSystems for Dual Cam Phaser Turbo GDI Engines." SAE
Technical Paper Series No. 2010-01-0588.
                                        1-35

-------
                                                         2017 Draft Regulatory Impact Analysis
    Table 1.3-6 Illustration of the TARF Ranking Process, Vehicle Type 5, 2025MY Costs
Preli
m
Pkg#
Weig
ht
rdxn
Package contents
Tran
s
Cost
Fuel
Saving
sa
Net
Cost
CO2
Rdx
n
(%)
CO2
Rdxn
(gram
s)
TARF
b
Round 1 (determine TARF relative to baseline package #50000)
5000
0
5000
1
5000
2
5039
5
5039
6
base
base
base
5%
5%
3.3L4VDOHCV6
4V DOHC V6 +LUB+EFR1+LDB+ASL
4V DOHC V6
+LUB+EFRl+LDB+ASL+IACC+EPS+Aerol+LRRTl
4V DOHC V6
+LUB+EFRl+LDB+ASL+IACC+EPS+Aerol+LRRTl+H
EG
4V DOHC V6
+LUB+EFRl+LDB+ASL+IACC+EPS+Aerol+LRRTl+H
EG+ DCP
4sp
AT
4sp
AT
4sp
AT
6sp
DCT
-wet
6sp
DCT
-wet
$0
$184
$394
$558
$723
$0
$841
$1,567
$2,946
$3,349
$0
-$657
$1,17
3
$2,38
8
$2,62
6
0%
7%
13%
24%
27%
0
23
42
79
90
-
0.152
8
0.146
4
0.158
5
0.153
3
Packages not shown for ease of presentation
5042
7
5042
8
5042
9
5%
5%
5%
4V DOHC 14
+LUB+EFRl+LDB+ASL+IACC+EPS+Aerol+LRRTl+H
EG+ DCP+DVVL+GDI+SS+SAX+TDS24
4V DOHC V6
+EFR2+LDB+ASL2+IACC+EPS+Aerol+LRRTl+HEG
4V DOHC V6
+EFR2+LDB+ASL2+IACC+EPS+Aerol+LRRTl+HEG+
DCP
6sp
DCT
-wet
6sp
DCT
-wet
6sp
DCT
-wet
$1,76
5
$646
$810
$4,808
$3,419
$3,795
$3,04
3
$2,77
4
$2,98
5
38%
27%
30%
128.7
91.6
101.6
0.123
8
0.158
6
0.153
8
Etc... remaining packages have larger TARFs so are not shown; #50428 becomes the new base; all packages with lower effectiveness than 50428
are eliminated from further consideration
Round 2 (determine net cost, CO2 reduction & TARF relative to new base |
5042
8
5042
7
5042
9
5%
5%
5%
4V DOHC V6
+EFR2+LDB+ASL2+IACC+EPS+Aerol+LRRTl+HEG
4V DOHC 14
+LUB+EFRl+LDB+ASL+IACC+EPS+Aerol+LRRTl+H
EG+ DCP+DVVL+GDI+SS+SAX+TDS24
4V DOHC V6
+EFR2+LDB+ASL2+IACC+EPS+Aerol+LRRTl+HEG+
DCP
6sp
DCT
-wet
6sp
DCT
-wet
6sp
DCT
-wet
$0
$1,12
0
$165
jackage #50428)
$0
$1,388
$376
$0
-$269
-$211
0%
11%
3%
0.0
37.2
10.1
-
0.037
9
0.109
9
Packages not shown for ease of presentation
5062
3
5062
4
5062
5
5%
5%
5%
4V DOHC
I4+LUB+EFRl+LDB+ASL+IACC+EPS+Aerol+LRRTl
+HEG+DCP+DVVL+GDI+SS+SAX+TDS24
4V DOHC
V6+EFR2+LDB+ASL2+IACC+EPS+Aerol+LRRTl+HE
G
4V DOHC
V6+EFR2+LDB+ASL2+IACC+EPS+Aerol+LRRTl+HE
G+DCP
8sp
DCT
-wet
8sp
DCT
-wet
8sp
DCT
-wet
$1,23
4
$115
$279
$1,598
$365
$695
-$364
-$251
-$416
13%
3%
6%
42.8
9.8
18.6
0.044
6
0.134
3
0.117
0
Etc... remaining packages have larger TARFs so are not shown; #50624 becomes the new base; all packages with lower effectiveness than 50624
are eliminated from further consideration
Round 3 (determine net cost, CO2 reduction & TARF relative to new base j
5062
4
5%
4V DOHC
V6+EFR2+LDB+ASL2+IACC+EPS+Aerol+LRRTl+HE
G
8sp
DCT
-wet
$0
jackage #50624)
$0
$0
0%
0.0
-
Etc. Further ranking rounds not shown for ease of presentation
1 Fuel savings calculated based on the effectiveness of the package, the energy content of the fuel and AEO 2011 reference case fuel prices
(gasoline, diesel, electric). Fuel savings are considered for the first 5 years of life assuming VMT consistent with our car/truck VMT estimates
excluding any rebound driving and are discounted at 3%.
b TARF units are $/kg, so a multiplicative factor of 1000 is included to convert g/mile to kg/mile.

  As illustrated in Table 1.3-6, the TARF ranking process eliminates most packages in favor of
  more cost effective packages. The packages that remain after the TARF ranking process are
                                                 1-36

-------
                                              2017 Draft Regulatory Impact Analysis
then included in the master-set of packages for each vehicle type. These packages are shown
for vehicle type 5 in Table 1.3-7, along with their new OMEGA package # identifier.

  Table 1.3-7 Master-set of 2025 MY Non-HEV/PHEV/EV Packages for Vehicle Type 5
                        (Midsize/Large car 3.3L DOHC V6)
Prelim
Pkg#
50000
50428
50624
50445
50641
50707
51099
51107
51139
51491
51531
51883
51923
51887
51927
51888
51890
51929
51961
51994
OMEGA
Pkg#
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
Weight
Rdxn
base
5%
5%
5%
5%
5%
10%
10%
10%
15%
15%
20%
20%
20%
20%
20%
20%
20%
20%
20%
Package contents
3.3L4VDOHC V6
4V DOHC V6
+EFR2+LDB+ASL2+IACC+EPS+Aerol+LRRTl+HEG
4V DOHC V6
+EFR2+LDB+ASL2+IACC+EPS+Aerol+LRRTl+HEG
4V DOHC 14
+EFR2+LDB+ASL2+IACC+EPS+Aerol+LRRTl+HEG+
DCP+GDI+TDS18
4V DOHC 14
+EFR2+LDB+ASL2+IACC+EPS+Aerol+LRRTl+HEG+
DCP+GDI+TDS18
4V DOHC 14
+EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+GDI+TDS18
4V DOHC 14
+EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+GDI+TDS18
4V DOHC 14
+EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+GDI+TDS24
4V DOHC 14
+EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+GDI+TDS24+EGR
4V DOHC 14
+EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+GDI+TDS18
4V DOHC 14
+EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+GDI+TDS24+EGR
4V DOHC 14
+EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+GDI+TDS18
4V DOHC 14
+EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+GDI+TDS24+EGR
4V DOHC 14
+EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+GDI+S AX+TDS 1 8
4V DOHC 14
+EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+GDI+SAX+TDS24+EGR
4V DOHC 14
+EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+DVVL+GDI+S AX+TDS 1 8
4V DOHC 14
+EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+DVVL+GDI+SS+SAX+TDS18
4V DOHC 14
+EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+GDI+SS+SAX+TDS24+EGR
4V DOHC 14
+EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+GDI+SS+SAX+TDS27+EGR
4V DOHC 14
+EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+DSL-Adv+SAX
Transmission
4sp AT
6sp DCT-wet
8sp DCT-wet
6sp DCT-wet
8sp DCT-wet
8sp DCT-wet
8sp DCT-wet
8sp DCT-wet
8sp DCT-wet
8sp DCT-wet
8sp DCT-wet
8sp DCT-wet
8sp DCT-wet
8sp DCT-wet
8sp DCT-wet
8sp DCT-wet
8sp DCT-wet
8sp DCT-wet
8sp DCT-wet
8sp DCT-wet
Cost
$0
$646
$760
$1,058
$1,172
$1,386
$1,507
$1,719
$1,966
$1,741
$2,200
$2,048
$2,507
$2,128
$2,588
$2,259
$2,602
$2,931
$3,356
$4,673
C02%
0%
27%
30%
37%
39%
43%
44%
46%
48%
46%
49%
47%
51%
48%
51%
48%
49%
52%
52%
49%
                                       1-37

-------
                                              2017 Draft Regulatory Impact Analysis
       The next packages after the advanced gasoline and diesel packages are the HEVs. We
noted above that we have considered applying only the P2 HEV for this analysis.  As done
with non-electrified packages, we began with a preliminary-set of HEV packages that paired
the HEV powertrain with increasing levels of engine technologies.  For non-towing vehicle
types we have paired the hybrid powertrain with an Atkinson engine. With each Atkinson
engine, we include dual cam phasing, discrete variable valve lift and stoichiometric gasoline
direct injection. Since most non-towing vehicle types are DOHC engines in the baseline,
these costs were simply added to the baseline engine to ensure that the Atkinson engine is
consistent with those modeled by Ricardo to ensure that our effectiveness values are
consistent.  But vehicle types 8 and 15 are SOHC and OHV, respectively,. Therefore, the
package by definition included costs associated with converting the valvetrain to a DOHC
configuration.  For towing vehicle types, we have paired the hybrid powertrain with a
turbocharged and downsized engine. By definition,  such engines include both dual cam
phasing and stoichiometric gasoline direct injection. Further, such engines might be 18/24/27
bar BMEP and the 24 bar BMEP engines may or may not include cooled EGR while the 27
bar BMEP engines must include cooled EGR as explained in Chapter 3.4.1 of the draft Joint
TSD. As a result, we have built more HEV packages for towing vehicle types than for non-
towing types.  Lastly, we built HEV packages with a constant weight reduction across the
board in the year of interest. For example, in building packages for a 2016MY OMEGA run,
we built HEV packages with 10% weight reduction as this was the maximum weight
reduction in 2016  allowed in the analysis.  This maximum allowed weight reduction was 15%
for the 2021MY and 20% for 2025 based on the technology penetration caps set forth and
explained i  n Chapter 3 of the joint TSD. Table 1.3-8 shows the HEV packages built for
vehicle types 5 and, for comparison, 10 which is a towing vehicle type.
   Table 1.3-8 HEV Packages Built for Vehicle Types 5 (3.3L DOHC V6) and 10 (4.
                                     SOHC V8)
7L
Prelim
Pkg#
500
501
502
503
504
505
506
507
508
509
510
Weight
Rdxn
base
20.0%
20.0%
20.0%
20.0%
20.0%
20.0%
20.0%
20.0%
20.0%
20.0%
Package contents
3.3L4VDOHC V6
4V DOHC V6
+LUB+EFRl+LDB+ASL+IACC+EPS+Aerol+LRRTl+HEG+
DCP+DVVL+GDI+ATKCS+HEV
4V DOHC V6
+LUB+EFRl+LDB+ASL+IACC+EPS+Aerol+LRRTl+HEG+
DCP+DVVL+GDI+ATKCS+HEV+SAX
4V DOHC V6 +EFR2+LDB+ASL2+IACC+EPS+Aerol+LRRTl+HEG+
DCP+DVVL+GDI+ATKCS+HEV
4V DOHC V6 +EFR2+LDB+ASL2+IACC+EPS+Aerol+LRRTl+HEG+
DCP+DVVL+GDI+ATKCS+HEV+SAX
4V DOHC V6
+LUB+EFRl+LDB+ASL+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+DVVL+GDI+ATKCS+HEV
4V DOHC V6
+LUB+EFRl+LDB+ASL+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+DVVL+GDI+ATKCS+HEV+SAX
4V DOHC V6 +EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+DVVL+GDI+ATKCS+HEV
4V DOHC V6 +EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+DVVL+GDI+ATKCS+HEV+SAX
4V DOHC V6
+LUB+EFRl+LDB+ASL+IACC+EPS+Aerol+LRRTl+HEG+
DCP+DVVL+GDI+ATKCS+HEV
4V DOHC V6
+LUB+EFRl+LDB+ASL+IACC+EPS+Aerol+LRRTl+HEG+
Transmission
4sp AT
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
6sp DCT-wet
8sp DCT-wet
8sp DCT-wet
2025
Cost
$0
$4,937
$5,018
$5,024
$5,105
$5,151
$5,231
$5,238
$5,319
$5,051
$5,132
C02%
0.0%
51.9%
52.3%
54.3%
54.7%
55.6%
55.9%
57.7%
58.1%
53.7%
54. 1%
                                        1-38

-------
      2017 Draft Regulatory Impact Analysis

511
512
513
514
515
516
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022

20.0%
20.0%
20.0%
20.0%
20.0%
20.0%
base
20.0%
20.0%
20.0%
20.0%
20.0%
20.0%
20.0%
20.0%
20.0%
20.0%
20.0%
20.0%
20.0%
20.0%
20.0%
20.0%
20.0%
20.0%
20.0%
20.0%
20.0%
20.0%
DCP+DVVL+GDI+ATKCS+HEV+SAX
4V DOHC V6 +EFR2+LDB+ASL2+IACC+EPS+Aerol+LRRTl+HEG+
DCP+DVVL+GDI+ATKCS+HEV
4V DOHC V6 +EFR2+LDB+ASL2+IACC+EPS+Aerol+LRRTl+HEG+
DCP+DVVL+GDI+ATKCS+HEV+SAX
4V DOHC V6
+LUB+EFRl+LDB+ASL+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+DVVL+GDI+ATKCS+HEV
4V DOHC V6
+LUB+EFRl+LDB+ASL+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+DVVL+GDI+ATKCS+HEV+SAX
4V DOHC V6 +EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+DVVL+GDI+ATKCS+HEV
4V DOHC V6 +EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+DVVL+GDI+ATKCS+HEV+SAX
4.7L 2V SOHC V8
4V DOHC V6
+LUB+EFRl+LDB+ASL+IACC+EPS+Aerol+LRRTl+HEG+
DCP+GDI+TDS 1 8+HE V
4V DOHC V6
+LUB+EFRl+LDB+ASL+IACC+EPS+Aerol+LRRTl+HEG+
DCP+GDI+S AX+TDS 1 8+HEV
4V DOHC V6
+LUB+EFRl+LDB+ASL+IACC+EPS+Aerol+LRRTl+HEG+
DCP+GDI+TDS24+HEV
4V DOHC V6
+LUB+EFRl+LDB+ASL+IACC+EPS+Aerol+LRRTl+HEG+
DCP+GDI+SAX+TDS24+HEV
4V DOHC V6 +EFR2+LDB+ASL2+IACC+EPS+Aerol+LRRTl+HEG+
DCP+GDI+TDS 1 8+HEV
4V DOHC V6 +EFR2+LDB+ASL2+IACC+EPS+Aerol+LRRTl+HEG+
DCP+GDI+S AX+TDS 1 8+HEV
4V DOHC V6 +EFR2+LDB+ASL2+IACC+EPS+Aerol+LRRTl+HEG+
DCP+GDI+TDS24+HEV
4V DOHC V6 +EFR2+LDB+ASL2+IACC+EPS+Aerol+LRRTl+HEG+
DCP+GDI+SAX+TDS24+HEV
4V DOHC V6
+LUB+EFRl+LDB+ASL+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+GDI+TDS 1 8+HEV
4V DOHC V6
+LUB+EFRl+LDB+ASL+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+GDI+S AX+TDS 1 8+HEV
4V DOHC V6
+LUB+EFRl+LDB+ASL+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+GDI+TDS24+HEV
4V DOHC V6
+LUB+EFRl+LDB+ASL+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+GDI+SAX+TDS24+HEV
4V DOHC V6 +EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+GDI+TDS 1 8+HEV
4V DOHC V6 +EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+GDI+S AX+TDS 1 8+HEV
4V DOHC V6 +EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+GDI+TDS24+HEV
4V DOHC V6 +EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+GDI+SAX+TDS24+HEV
4V DOHC V6 +EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+GDI+TDS24+EGR+HEV
4V DOHC V6 +EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+GDI+SAX+TDS24+EGR+HEV
4V DOHC 14 +EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+GDI+TDS27+EGR+HEV
4V DOHC 14 +EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+GDI+SAX+TDS27+EGR+HEV
4V DOHC V6
+LUB+EFRl+LDB+ASL+IACC+EPS+Aerol+LRRTl+HEG+
DCP+GDI+TDS 1 8+HEV
4V DOHC V6

8sp DCT-wet
8sp DCT-wet
8sp DCT-wet
8sp DCT-wet
8sp DCT-wet
8sp DCT-wet
4sp AT
6sp AT
6sp AT
6sp AT
6sp AT
6sp AT
6sp AT
6sp AT
6sp AT
6sp AT
6sp AT
6sp AT
6sp AT
6sp AT
6sp AT
6sp AT
6sp AT
6sp AT
6sp AT
6sp AT
6sp AT
8sp AT
8sp AT

$5,139
$5,219
$5,265
$5,346
$5,353
$5,433
$0
$5,749
$5,829
$6,107
$6,187
$5,836
$5,916
$6,194
$6,274
$5,963
$6,043
$6,321
$6,401
$6,050
$6,130
$6,408
$6,488
$6,655
$6,735
$6,084
$6,164
$5,807
$5,887

56.0%
56.3%
57.2%
57.5%
59.3%
59.6%
0.0%
45.7%
46.3%
47.8%
48.3%
48.0%
48.5%
49.9%
50.4%
49.5%
50.0%
51.2%
51.7%
51.6%
52.1%
53.3%
53.7%
54.9%
55.4%
55.4%
55.8%
47.7%
48.2%
1-39

-------
                                                   2017 Draft Regulatory Impact Analysis

1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040

20.0%
20.0%
20.0%
20.0%
20.0%
20.0%
20.0%
20.0%
20.0%
20.0%
20.0%
20.0%
20.0%
20.0%
20.0%
20.0%
20.0%
20.0%
+LUB+EFRl+LDB+ASL+IACC+EPS+Aerol+LRRTl+HEG+
DCP+GDI+S AX+TDS 1 8+HEV
4V DOHC V6
+LUB+EFRl+LDB+ASL+IACC+EPS+Aerol+LRRTl+HEG+
DCP+GDI+TDS24+HEV
4V DOHC V6
+LUB+EFRl+LDB+ASL+IACC+EPS+Aerol+LRRTl+HEG+
DCP+GDI+SAX+TDS24+HEV
4V DOHC V6 +EFR2+LDB+ASL2+IACC+EPS+Aerol+LRRTl+HEG+
DCP+GDI+TDS 1 8+HEV
4V DOHC V6 +EFR2+LDB+ASL2+IACC+EPS+Aerol+LRRTl+HEG+
DCP+GDI+S AX+TDS 1 8+HEV
4V DOHC V6 +EFR2+LDB+ASL2+IACC+EPS+Aerol+LRRTl+HEG+
DCP+GDI+TDS24+HEV
4V DOHC V6 +EFR2+LDB+ASL2+IACC+EPS+Aerol+LRRTl+HEG+
DCP+GDI+SAX+TDS24+HEV
4V DOHC V6
+LUB+EFRl+LDB+ASL+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+GDI+TDS 1 8+HEV
4V DOHC V6
+LUB+EFRl+LDB+ASL+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+GDI+S AX+TDS 1 8+HEV
4V DOHC V6
+LUB+EFRl+LDB+ASL+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+GDI+TDS24+HEV
4V DOHC V6
+LUB+EFRl+LDB+ASL+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+GDI+SAX+TDS24+HEV
4V DOHC V6 +EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+GDI+TDS 1 8+HEV
4V DOHC V6 +EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+GDI+S AX+TDS 1 8+HEV
4V DOHC V6 +EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+GDI+TDS24+HEV
4V DOHC V6 +EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+GDI+SAX+TDS24+HEV
4V DOHC V6 +EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+GDI+TDS24+EGR+HEV
4V DOHC V6 +EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+GDI+SAX+TDS24+EGR+HEV
4V DOHC 14 +EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+GDI+TDS27+EGR+HEV
4V DOHC 14 +EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+GDI+SAX+TDS27+EGR+HEV

8sp AT
8sp AT
8sp AT
8sp AT
8sp AT
8sp AT
8sp AT
8sp AT
8sp AT
8sp AT
8sp AT
8sp AT
8sp AT
8sp AT
8sp AT
8sp AT
8sp AT
8sp AT

$6,165
$6,245
$5,894
$5,975
$6,252
$6,333
$6,021
$6,101
$6,379
$6,459
$6,108
$6,188
$6,466
$6,547
$6,713
$6,794
$6,142
$6,222

49.6%
50.0%
49.9%
50.4%
51.7%
52. 1%
51.3%
51.7%
52.9%
53.3%
53.4%
53.8%
54.9%
55.2%
56.5%
56.8%
56.9%
57.2%
Note: Prelim Pkg #s 500-516 are for vehicle type 5, #s 1000-1040 are for vehicle type 10.
Note also that any automatic transmission that has been improved from the base 4sp AT also includes early torque converter
lockup even though that technology is not specifically listed in the package contents. This is the only technology that does
not appear in the package content descriptions.

We then ranked the preliminary-set of HEV packages according to TARF as described above
to generate the most cost effective set of HEV packages for each vehicle type that would then
be included in the master-set of packages. The TARF ranking process eliminated most
packages in favor of more cost effective packages.  These packages are shown for vehicle
types 5 and 10 (as examples) in
                                            1-40

-------
                                              2017 Draft Regulatory Impact Analysis
Table 1.3-9, along with their new OMEGA package # identifier.
                                       1-41

-------
                                              2017 Draft Regulatory Impact Analysis
Table 1.3-9 Master-set of 2025 MY HEV Packages for Vehicle Types 5 (3.3L DOHC V6)
                               & 10 (4.7L SOHC V8)
Prelim
Pkg#
515
516
1033
1037
1039
1034
1040
OMEGA
Pkg#
520
521
1018
1019
1020
1021
1022
Weight
rdxn
20%
20%
20%
20%
20%
20%
20%
Package contents
4V DOHC V6
+EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+DVVL+GDI+ATKCS+HEV
4V DOHC V6
+EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+DVVL+GDI+ATKCS+HEV+SAX
4V DOHC V6
+EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+GDI+TDS 1 8+HEV
4V DOHC V6
+EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+GDI+TDS24+EGR+HEV
4V DOHC 14
+EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+GDI+TDS27+EGR+HEV
4V DOHC V6
+EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+GDI+S AX+TDS 1 8+HEV
4V DOHC 14
+EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+
DCP+GDI+SAX+TDS27+EGR+HEV
Transmission
8sp DCT-wet
8sp DCT-wet
8sp AT
8sp AT
8sp AT
8sp AT
8sp AT
2025MY
Cost
$5,353
$5,433
$6,108
$6,713
$6,142
$6,188
$6,222
C02%
59.3%
59.6%
53.4%
56.5%
56.9%
53.8%
57.2%
       The last step was to build the PHEVs (also known as REEVs) and EVs for vehicle
types 1 through 8 and 15.  The other vehicle types were not considered for electrification
beyond HEVs for purposes of the current analysis, either because of their expected towing
demands or because of their high vehicle weight which would make the electrification of the
vehicle prohibitively costly. We have developed 2 primary types of PHEV packages and 3
primary types of EV packages all of which are included in the master-set of packages. The
PHEVs consist of packages with battery packs capable of 20 miles of all electric operation
(REEV20) and packages with battery packs capable of 40 miles of all electric operation
(REEV40). For EVs, we have built packages capable of 75, 100 and 150 miles of all electric
operation, EV75, EV100 and EV150, respectively. These ranges were selected to represent
an increasing selection of ranges (and costs) that consumers would likely require and that we
believe will be available in the 2017-2025 timeframe. For each of these packages, we have
estimated specific battery-pack costs based on the net weight reduction of the vehicle where
the net weight reduction is the difference between the weight reduction technology applied to
the "glider" (i.e., the vehicle less any powertrain elements) and the weight increase that results
from the inclusion of the electrification components (batteries, motors, etc.). The applied and
net weight reductions for HEVs, PHEVs and EVs are presented in Chapter 3 of the draft joint
TSD, and full system costs for each depending on the net weight reduction are presented there
and are also presented in Table 1.2-7 through Table 1.2-12. Table 1.3-10 shows the PHEV
and EV packages built for the 2025MY in this proposal (note that PHEVs are shown as
REEVs in the table). Note that the PHEV and EV packages are included directly in the
master-set of packages for a 2025MY OMEGA run.  We have not built a long preliminary-set
of PHEVs  and EVs and ranked them based on TARF to determine which packages to include
in the master-set. This is because for each MYof interest we built the
REEV20/REEV40/EV75/EV100/EV150 with the maximum allowed weight reduction
                                        1-42

-------
                                                2017 Draft Regulatory Impact Analysis
(applied weight reduction) of 20% even though for MYs 2016 and 2021 the maximum
allowed weight reduction under our phase-in caps was 10% and 15% for those MYs.1 We
have done this for two reasons.  First, some PHEV and EV packages cannot be built unless a
20% applied weight reduction is available because the weight of the electrification
components is such that the net weight reduction would be less than zero without the ability to
apply a 20% reduction (i.e., the vehicle would increase in weight).  We did not want to build
packages with net weight increases and we did not have the ability to properly determine their
effectiveness values even if we wanted to build them.  Second, we believe it is reasonable that
auto makers would be more aggressive with respect to weight reduction on PHEVs and EVs
(so as to be able to utilize lower weight, and hence less expensive batteries) and that it is
reasonable to believe that PHEVs and EVs could achieve higher levels of weight reduction in
the 2016 and 2021  MYs than we have considered likely for other vehicle technologies.

   Table 1.3-10 Master-set of 2025 MY PHEV (REEV) & EV Packages for all Vehicle
                                        Types
OMEG
A
Pkg#
120
121
122
123
124
223
224
225
226
111
323
324
325
326
327
421
422
Appli
ed
Weig
ht
Rdxn
20.0%
20.0%
20.0%
20.0%
20.0%
20.0%
20.0%
20.0%
20.0%
20.0%
20.0%
20.0%
20.0%
20.0%
20.0%
20.0%
20.0%
Package contents
4V DOHC
I4+EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+DCP+DVVL+G
DI+ATKCS+REEV20
4V DOHC
I4+EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+DCP+DVVL+G
DI+ATKCS+REEV40
EV75 mile+IACC2+Aero2+LRRT2+EPS
EV100 mile+IACC2+Aero2+LRRT2+EPS
EV150 mile+IACC2+Aero2+LRRT2+EPS
4V DOHC
I4+EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+DCP+DVVL+G
DI+ATKCS+REEV20
4V DOHC
I4+EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+DCP+DVVL+G
DI+ATKCS+REEV40
EV75 mile+IACC2+Aero2+LRRT2+EPS
EV100 mile+IACC2+Aero2+LRRT2+EPS
EV150 mile+IACC2+Aero2+LRRT2+EPS
4V DOHC
I4+EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+DCP+DVVL+G
DI+ATKCS+REEV20
4V DOHC
I4+EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+DCP+DVVL+G
DI+ATKCS+REEV40
EV75 mile+IACC2+Aero2+LRRT2+EPS
EV100 mile+IACC2+Aero2+LRRT2+EPS
EV150 mile+IACC2+Aero2+LRRT2+EPS
4V DOHC
V6+EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+DCP+DVVL+
GDI+ATKCS+REEV20
4V DOHC
Trans
8sp
DCT-
dry
8sp
DCT-
dry
N/A
N/A
N/A
8sp
DCT-
dry
8sp
DCT-
dry
N/A
N/A
N/A
8sp
DCT-
dry
8sp
DCT-
dry
N/A
N/A
N/A
8sp
DCT-
wet
8sp
2025
Cost
$9,489
$11,402
$10,056
$11,542
$15,036
$10,044
$12,211
$10,962
$12,719
$16,757
$10,121
$12,288
$11,039
$12,796
$16,834
$12,135
$15,186
CO2%
74.8%
84.5%
100.0%
100.0%
100.0%
75.6%
85.0%
100.0%
100.0%
100.0%
75.6%
85.0%
100.0%
100.0%
100.0%
74.8%
84.5%
1 Note, as noted above, the weight reduction of a technology package has no impact on the weight reduction
allowed under our safety analysis, with the exception that it serves as an upper bound . The safety aspect to
weight reduction is not dealt with in the package building process and is instead dealt with in the TEB-CEB
process and OMEGA model itself. This is described in Chapter 3 of this draft RIA.
                                         1-43

-------
                                                  2017 Draft Regulatory Impact Analysis

423
424
425
522
523
524
525
526
621
622
623
624
625
723
724
725
726
727
822
823
824
825
826
1521
1522
1523
1524
1525

20.0%
20.0%
20.0%
20.0%
20.0%
20.0%
20.0%
20.0%
20.0%
20.0%
20.0%
20.0%
20.0%
20.0%
20.0%
20.0%
20.0%
20.0%
20.0%
20.0%
20.0%
20.0%
20.0%
20.0%
20.0%
20.0%
20.0%
20.0%
V6+EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+DCP+DVVL+
GDI+ATKCS+REEV40
EV75 mile+IACC2+Aero2+LRRT2+EPS
EV100 mile+IACC2+Aero2+LRRT2+EPS
EV150 mile+IACC2+Aero2+LRRT2+EPS
4V DOHC
V6+EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+DCP+DVVL+
GDI+ATKCS+REEV20
4V DOHC
V6+EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+DCP+DVVL+
GDI+ATKCS+REEV40
EV75 mile+IACC2+Aero2+LRRT2+EPS
EV100 mile+IACC2+Aero2+LRRT2+EPS
EV150 mile+IACC2+Aero2+LRRT2+EPS
4V DOHC
V8+EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+DCP+DVVL+
GDI+ATKCS+REEV20
4V DOHC
V8+EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+DCP+DVVL+
GDI+ATKCS+REEV40
EV75 mile+IACC2+Aero2+LRRT2+EPS
EV100 mile+IACC2+Aero2+LRRT2+EPS
EV150 mile+IACC2+Aero2+LRRT2+EPS
4V DOHC
I4+EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+DCP+DVVL+G
DI+ATKCS+REEV20
4V DOHC
I4+EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+DCP+DVVL+G
DI+ATKCS+REEV40
EV75 mile+IACC2+Aero2+LRRT2+EPS
EV100 mile+IACC2+Aero2+LRRT2+EPS
EV150 mile+IACC2+Aero2+LRRT2+EPS
4V DOHC
V6+EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+DCP+DVVL+
GDI+ATKCS+REEV20
4V DOHC
V6+EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+DCP+DVVL+
GDI+ATKCS+REEV40
EV75 mile+IACC2+Aero2+LRRT2+EPS
EV100 mile+IACC2+Aero2+LRRT2+EPS
EV150 mile+IACC2+Aero2+LRRT2+EPS
4V DOHC
V8+EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+DCP+DVVL+
GDI+ATKCS+REEV20
4V DOHC
V8+EFR2+LDB+ASL2+IACC2+EPS+Aero2+LRRT2+HEG+DCP+DVVL+
GDI+ATKCS+REEV40
EV75 mile+IACC2+Aero2+LRRT2+EPS
EV100 mile+IACC2+Aero2+LRRT2+EPS
EV150 mile+IACC2+Aero2+LRRT2+EPS
DCT-
wet
N/A
N/A
N/A
8sp
DCT-
wet
8sp
DCT-
wet
N/A
N/A
N/A
8sp
DCT-
wet
8sp
DCT-
wet
N/A
N/A
N/A
8sp
DCT-
dry
8sp
DCT-
dry
N/A
N/A
N/A
8sp
DCT-
wet
8sp
DCT-
wet
N/A
N/A
N/A
8sp
DCT-
wet
8sp
DCT-
wet
N/A
N/A
N/A

$12,677
$15,094
$21,008
$12,485
$15,670
$12,908
$14,643
$20,280
$12,747
$15,931
$12,964
$14,699
$20,336
$11,799
$14,851
$12,711
$15,128
$21,041
$11,747
$14,522
$12,864
$15,107
$20,852
$12,670
$15,854
$12,886
$14,622
$20,259

100.0%
100.0%
100.0%
75.2%
84.7%
100.0%
100.0%
100.0%
75.2%
84.7%
100.0%
100.0%
100.0%
74.9%
84.5%
100.0%
100.0%
100.0%
74.0%
84.0%
100.0%
100.0%
100.0%
75.2%
84.7%
100.0%
100.0%
100.0%
Note that the net weight reduction of these packages as a percent can be determined by cross-referencing the applied weight reduction shown
here with the proper cost table (PHEV20/40, EV75/100/150) shown in Section 1.2 and the vehicle class information shown in Table 1.3-1.
       The end result is a master-set of roughly 25 packages for each vehicle type. Because
of the large number of total packages and the difficulty of presenting them all here, we have
placed in the docket (EPA-HQ-OAR-2010-0799) a memorandum that contains the master-set
of packages used for our 2016MY, 2021MY and 2025MY OMEGA runs.2

       The remaining package building step in developing a set of OMEGA inputs is to rank
the master-set of packages according to TARF.  The end result of this ranking is a ranked-set
of OMEGA packages that includes the package progression that OMEGA must follow when
determining which package to employ next. The package progression is key because
                                           1-44

-------
                                               2017 Draft Regulatory Impact Analysis
OMEGA evaluates each package in a one-by-one, or linear progression. The packages must
be ordered correctly so that no single package will prevent the evaluation of the other
packages. For example, if we simply listed packages according to increasing effectiveness,
there could well be a situation where an HEV with higher effectiveness and a better TARF
than a turbocharged and downsized package with a poor TARF could never be chosen
because the turbocharged and downsized package, having a poor TARF, would never get
chosen and would effectively block the HEV from consideration. For that reason, it is
important to first rank by TARF so that the proper package progression can be determined.
The docket memorandum mentioned earlier also contains a ranked-set of packages for each of
the master-sets we have created.3  The ranked-set also includes the package progression
information. These ranked-sets of packages are reformatted and used as Technology Input
Files for the  OMEGA model.

1.4 Use of the Lumped Parameter Approach in Determining Package Effectiveness

           1.4.1     Background

       While estimating the GHG and fuel consumption reduction effectiveness of individual
vehicle technologies can often be confirmed with existing experimental and field data, it is
more challenging to predict the combined effectiveness of multiple technologies for a future
vehicle.  In 2002 the National Research Council published "Effectiveness and Impact of
Corporate Average Fuel Economy (CAFE) Standards4." It was one of the first and most
authoritative analyses of potential fuel consumption-reducing technologies available to future
light-duty vehicles, and is still widely referenced to this day.  However, it was criticized for
not fully accounting for system interactions ("synergies") between combinations of multiple
engine, transmission and vehicle technologies that could reduce the overall package
effectiveness.

       Comments to the 2002 NRC report recommended the use of a more sophisticated
method to account for vehicle technology package synergies - that of detailed, physics-based
vehicle simulation modeling. This method simulates the function of a vehicle by physically
modeling and linking all of the key components in a vehicle (engine, transmission, accessory
drive, road loads, test cycle speed schedule, etc) and requires an intricate knowledge of the
inputs that define those components. If the inputs are well-defined and plausible, it is
generally accepted  as the most accurate method for estimating future vehicle fuel efficiency.

       In one of the most thorough technical responses to the NRC report, Patton et al5
critiqued the overestimation of potential benefits of NRC's "Path 2" and "Path 3" technology
packages. They presented a vehicle energy balance analysis to highlight the synergies that
arise with the combination of multiple vehicle technologies. The report then demonstrated an
alternative methodology (to vehicle simulation) to estimate these synergies, by means of a
"lumped parameter" approach. This approach served as the basis for EPA's lumped
parameter model. The lumped parameter model was created for the 2012-2016 light duty
vehicle GHG and CAFE standards, and has been improved to reflect updates required for the
proposed 2017-2025 light duty GHG rule.
                                        1-45

-------
                                               2017 Draft Regulatory Impact Analysis
           1.4.2     Role of the model

       It is widely acknowledged that full-scale physics-based vehicle simulation modeling is
the most thorough approach for estimating future benefits of a package of new technologies.
This is especially important for quantifying the efficiency of technologies and groupings (or
packages) of technologies that do not currently exist in the fleet or as prototypes. However,
developing and running detailed vehicle simulations is very time and resource-intensive, and
generally not practical to implement over a large number of vehicle technology packages (in
our case, hundreds). As part of rulemakings EPA analyzes a wide array of potential
technology options rather than attempt to pre-select the "best" solutions.  For example, in
analysis for the 2012-2016 Light Duty Vehicle GHG rule6, EPA built over 140 packages for
use in its OMEGA compliance model, which spanned 19 vehicle classes and over 1100
vehicle models; for this rulemaking the number of packages has increased by another order of
magnitude over the previous rule.  The lumped parameter approach was chosen as the most
practical surrogate to estimate the package effectiveness (including synergies) of many
technology combinations. However, vehicle simulation modeling was a key part of the
process to ensure that the lumped parameter model was thoroughly validated. An overview of
the vehicle simulation study (conducted by Ricardo, PLC) for this rulemaking is provided in
Section 3.3.1 of the Joint TSD.  Additional details can be found in the project report7.

           1.4.3     Overview of the lumped parameter model

       The basis for EPA's lumped parameter analysis is a first-principles energy balance
that estimates the manner in which the chemical energy of the fuel is converted into various
forms of thermal and mechanical energy on the vehicle. The analysis accounts for the
dissipation of energy into the different categories of energy losses, including each of the
following:

       •  Second law losses (thermodynamic losses inherent in the combustion of fuel),

       •  Heat lost from the combustion process to the exhaust and coolant,

       •  Pumping losses, i.e., work performed by the engine during the intake and exhaust
          strokes,

       •  Friction losses in the engine,

       •  Transmission losses, associated with friction and  other parasitic losses of the
          gearbox, torque converter (when applicable) and driveline

       •  Accessory losses, related directly to the parasitics associated with the engine
          accessories,

       •  Vehicle road load (tire and aerodynamic) losses;

       •  Inertial losses (energy dissipated as heat in the brakes)
                                         1-46

-------
                                                2017 Draft Regulatory Impact Analysis
       The remaining energy is available to propel the vehicle. It is assumed that the baseline
vehicle has a fixed percentage of fuel lost to each category. Each technology is grouped into
the major types of engine loss categories it reduces. In this way, interactions between
multiple technologies that are applied to the vehicle may be determined. When a technology is
applied, the lumped parameter model estimates its effects by modifying the appropriate loss
categories by a given percentage. Then, each subsequent technology that reduces the losses in
an already improved category has less of a potential impact than it would if applied on its
own.

       Using a lumped parameter approach for calculating package effectiveness provides
necessary grounding to physical principles. Due to the mathematical structure of the model, it
naturally limits the maximum effectiveness achievable for a family of similar technologies^
This can prove useful when computer-simulated packages are compared to a "theoretical
limit" as a plausibility check. Additionally, the reduction of certain energy loss categories
directly impacts the effects on others.  For example, as mass is reduced the benefits of brake
energy recovery decreases because there is not as much inertia energy to recapture.

       Figure 1.4-1 is an example spreadsheet used by EPA to estimate the package
effectiveness and the synergistic impacts of a technology package for a standard-size car.
1 For example, if only 4% of fuel energy is lost (in a baseline engine) to pumping work, leveraging multiple
technologies to theoretically eliminate all pumping losses would yield an aggregate reduction of no more than
15% in fuel consumption.
                                         1-47

-------
                                                                  2017 Draft Regulatory Impact Analysis
                                         EPA Staff Deliberative Materials-Do Not Quote or Cite
                    Vehicle Energy Effects Estimator
                                                                                                          Evaluate New
                                                                                                            Package
Vehicle Type
Standard car
% of tractive energy
Baseline %
offuel
Reduction
% of NEW fuel
2008 Baseline
New
Road
Indicated
Efficiency
Rated Power
' 158
0
hP
Rated Torque | ETW
161
ft-lb ' 3625 Ib
0
SOmphRL
11.3 hp
0.0
Q-oss Indicated Energy
Brake Energy
load kWh
Mech
Efficiency
Road Loads
Mass
Braking /
Inertia
4.0%
0%
4.0%
0.47
Brake
Efficiency
36.0% 59.6% 21.5%
38.0% 76.5% 29.0%
Drag
Aero
Load
6.4%
8%
5.9%
0.71
Drivetrain
Efficiency
80.6%
84.9%
Tires
Rolling
Load
Geaibox,
T.C.
Trans
Losses
Total Engine Friction

Access
Losses
40%
6.9% 4.2% 1.3%
0.77
Cycle
Efficiency
Fuel
Efficiency
Road
Loads
100.0% 17.3% 100.0%
100.0% 24.7% 94.2%
Friction Pumping
Losses Losses
7.9% 5.3%
15.4% 81.2%
7.1% 1.0%
| Package Notes

12V Stop-Start
Reset LP Model
Stoich GDI Turbo
Heat
Lost To
Exhaust &
Coolant
IndEff
Losses
34.0%
Irreversibilities,
etc.
Second
Law
30.0%
Check
100.0%
32.0% | 30%
2008 Ricardo baseline ralues includes some techs
Fuel Economy 32.0 mpg (combined)
Fuel Consumption 0.031 gal/mi
GHG emissions 284 g/niCO2E
       Current Results
    66.1%    Fuel Consumption (GGE/nile)
   33.9%    FC Reduction w no-techs
   51.2%    FE Improvement (mpgge)
    51.2%    FE Improvement (mpg)
 Tractive
  1.95
Original friction/brake ratio
Based on PMEP/IMEP »»
(GM study)
                                                         PMEP   Brake
                                                        Losses [Efficiency
|   30.5%   |GHGreductionvs2008Ricardo baseline
   33.9%    GHG reduction vs no-techs
                          Independent
Technology                 FC Estimate*
                                         Loss Category
                      =71.1% mech efficiency
                                                       Implementation into estimator
Vehicle mass reduction
Aero Drag Reduction
Rolling Resistance Reduction
Low Fric Lubes
EF Reduction
4V on 2V Bas eline
ICP
DCP
CCP
Deac
DWL
CWL
Turbo/Downsi^ (gas engines only)
5-spd gearbox
6-spd gearbox
8-spd gearbox
CVT
DCTWet
DCTDry
Early upshift (formerly ASL)
CptimzEd shift strategy
Agg TC Lockup
High efficiency gearbox(auto)
,»,,„„ ,,,^ „. .
High voltage SS, with launch (BAS)
Alternator regen on braking
EPS
Electric access ("l2V)
Electric access (liigh V)
High efficiency alternator (70%)
GDI (stoich)
GDI (stoich)w/ cooled EGR
GDI (lean)
Diesel - LNT (2008)
Diesel -SCR (2008)
5-6% per 10%
2.1% per 10%
1.5%
0.5%

3.0%
2.0%
4.0% total WT
4.0% total WT
6.0%
4.0%
5.0%

2.5%
5.5%

6.0% '
6.7%
10.0%
2.0% '
5.5%
0.5%

2.0%
7 5% ^
2.0%
2.0%
1.5%
3.0% "•

1.5%


30.0%
35.0%
Hybrid drivetrain (need to select transmission style!)
Secondary axle disconnect
Low drag brakes
Atkinson cycle engine
Advanced Diesel (2020)
1.3%
0.8%


Braking/stopped, inertia, rolling resistance 0%
Aero
Rolling
Friction
Friction
Pumping, friction
Pumping
Pumping
Pumping
Pumping, friction
Pumping
Pumping
Pumping
Pumping
Pumping
Pumping
Trans, pump ing
Trans
Trans
Pumping
Pumping, IE, friction
Trans
Trans
P.F.trans
B/I, P, F, trans
Access
Access
Access
Access
Access
IndEff

Ind. Eff, pumping
Ind Eff, P, F, trans
Ind Eff, P, F, trans
14.4% aero (cars), 9.5% aero (true ' 10%
9.5% rolling ' 10%
2% friction
rariable% friction 1
20.5% pumping, -2.5% trie
13.5% pumping, +0.2% IE, -3.5% fric
23.5% pumping, +0.2% IE, -2.5% fric
23.5% pumping, +0.2% IE, -2.5% fric
30% pumping, -2.5% frict
27% pumping, -3% friction 0%
33% pumping, -3% friction
rariable IE ratio, P,F 35%
6% pumping
pumping, +0.1% IE
15% putr: ::_:'- trans, +0.5% IE
41 % pumping, -5% trans
21% trans (increment)
25% trans (increment)
10.5% pumping
11% pumping, 11% fiict, +0.1% IE
2% trans
variable % Trans 7%
3% pumping, 3% friction, 2% trans
11% B/I, 3% P, 3% F, 2% trans
10% pumping
22% access 100%
12% access
42% acces^s
15% access
+ 0.55% IE
+1.9% IE, 41% pumping
+1.3% IE, 41% pumping
see comment
see comment victor kW
Inertia, trans, ace IE, F, P 0
Trans
Braking/inertia
Ind. Eff, -pumping
Ind Eff, P, F, trans
6% trans
3.5% B/I
+6% IE, -30% pumping
see comment
0
1
1
0
1
0
0
1
0
0
1
0
1
0
0
1
0
0
0
0
1
1

1
0
1
1
1
0
1
1
0
0
0
0
0
0
0
0
0
Regressed baseline ralues    assumes no techs
   req'd fuel energy   11.95  kWh
     fuel economy   30.4  mpg (unadj)
   fuel consumption   0.033  gal/mi
   GHG emissions    299   g/ni CO2E
 Current package \a ues
     fuel economy   46.03  mpg (unadj)
   fuel consumption   0.022  gal/mi
   GHG emissions    197  1g/niCO2E
or    Lser Picklist
     Include? (0/1) Devstatus
                                                                                                          Pick one max
                                                                                                          Pick one max
                                                                                                         Additive to trans;
                                                                                                          Included in P2
Plug-In
                    Figure 1.4-1 Sample lumped parameter model spreadsheet
                                                         1-48

-------
                                               2017 Draft Regulatory Impact Analysis
       The LP model has been updated from the MYs 2012-2016 final rule to support the
MYs 2017-2025 proposed standards. Changes were made to include new technologies for
2017 and beyond, improve fidelity for baseline attributes and technologies, and better
represent hybrids based on more comprehensive vehicle simulation modeling. Section 1.5
provides details of the methodology used to update and refine the model.
1.5 Lumped Parameter Model Methodology

           1.5.1     Changes to the LP model for the proposed rulemaking

       The LP model was updated in conjunction with this rulemaking to provide more
flexibility in assessment of package effectiveness, to incorporate new technologies not
previously analyzed, and to improve the calculation methodology in an effort to increase
calibration accuracy with respect to the supporting vehicle simulation data.

       Flexibility was added in several ways. First, the model now provides the user with the
capability of estimating package effectiveness for multiple vehicle classes. Second, several
compound technologies in the 2012-2016 rulemaking version have been "deconstructed" into
separate components so that there is more flexibility in adding different technology
combinations. The most visible example of that is in the new model's treatment of hybrids.
In the last generation LP model, a hybrid vehicle package served as a technology in and of
itself- irrespective of engine type,  ancillary technologies or road load reductions. In the
latest version the LP model offers a "hybrid drivetrain" technology which can be combined
with any engine technology and subset of road load reductions (e.g., mass reduction, rolling
resistance and aerodynamic drag reductions) and other technologies.  In this way, there is
more resolution and effectiveness distinction between the many combinations of technologies
on hybrids.

       The LP model also added new technologies, most stemming from the 2011 Ricardo
simulation project, which included multiple steps of transmission shift logic, more
mechanically efficient transmissions ("gearboxes"), alternator technologies, an Atkinson-
cycle engine for hybrids, highly downsized and  turbocharged engines including lean-burn and
cooled EGR options, and stop-start (idle-off without launch assist). The effectiveness of some
of these technologies vary based on additional required user inputs. For example,
turbocharging and downsizing effectiveness is now based on a percentage of displacement
reduction, and hybrid effectiveness is tied to electric motor size.

       EPA revisited the calculation methodology of the model with more rigor.  Through
more detailed analysis of simulation data, physical trends became more apparent, such as:

       •  the relationship between mass reduction and rolling resistance - naturally, as
          vehicle weight decreases, the normal force on the tires decreases, and  should
          reduce rolling resistance

       •  Reduced road  loads (with other variables held constant) changed the required
          tractive forces and usually resulted in reduced engine efficiency.


                                         1-49

-------
                                                 2017 Draft Regulatory Impact Analysis
       •  For hybrids, mass reduction was synergistic with the hybrid drivetrain, as there
          was less recoverable braking energy with a lighter vehicle.

       All of these trends were identified through the analysis of the simulation data and
performance metrics (detailed further in the Joint TSD, Section 3.3.1), and were incorporated
during the development of the model.

            1.5.2     Development of the model

       The LP model must be flexible in accommodating a wide variety of possible vehicle
and technology package combinations and also must reasonably reflect the physical system
effects of each technology added to a vehicle.  Finally, its outputs must be well calibrated to
the existing vehicle simulation results for it to  serve as a reliable tool for use in generating
OMEGA model  inputs. To properly  build the  LP model with all of these requirements in
mind, several steps were needed:

       •  Develop a baseline energy loss distribution for each vehicle class

       •  Calibrate baseline fuel economy for each vehicle class based on simulation and
          vehicle certification data

       •  Add technologies to the model  and  identify the significant loss categories that each
          applied technology affects, and

       •  Assign numerical loss category modifiers for each individual technology to
          achieve the estimated independent effectiveness

       •  Calibrate LP technology package effectiveness with simulation results
            1.5.3     Baseline loss categories

       In 2007, EPA contracted with PQA, who subcontracted Ricardo, LLC to conduct a
vehicle simulation modeling project in support of the 2012-2016 light-duty vehicle GHG rule.
Further simulation work was conducted by Ricardo from 2010-2011 to support EPA's
analysis for the 2017-2025 vehicle GHG rule. In both projects, Ricardo built versions of its
EASY5 and WAVE models to generate overall vehicle package GHG reduction effectiveness
results and corresponding 10-hz output files of the intermediate data. EPA's detailed analysis
of the Ricardo 2008 and 2010 baselineK vehicle simulation output files for the FTP and
HWFE test cycles helped quantify the distribution of fuel energy losses in the baseline LP
K The 2008 baseline vehicles are those originally used in the 2008 Ricardo simulation project and represent
actual vehicles in production.  The 2010 "baseline" vehicles (from the 2011 Ricardo report) have additional
content including stop-start, improved alternator with regenerative capability, and a six-speed automatic
transmission. For more information reference the Joint TSD, Section 3.3.1.8.
                                          1-50

-------
                                                 2017 Draft Regulatory Impact Analysis
model.  City/highway combined cycle average data were obtained for brake efficiency, torque
converter and driveline efficiencies, accessory losses, and wheel (tractive) energy.  These
values were regressed against basic vehicle parameters (power, weight, etc) to generate curve
fits for the baseline vehicle category attributes.
       The distribution of energy loss categories in the baseline vehicle were estimated as
follows:
       •  Indicated efficiency was assumed at a combined test cycle average of 36% for all
          vehiclesL

       •  Baseline engine brake efficiency was estimated as a function of (ETW, road load,
          engine torque, and alternator regeneration or "regen"). These inputs were used in
          a linear regression, shown in Figure 1.5-1, which fits the 2008 and 2010 Ricardo
          baseline data from the output summaries.

        Regression data used - net engine brake efficiency
Vehicle
Camry
Vue
Caravan
300
F-150
Yaris
Camry
Vue
Caravan
300
F-150

Power
154
169
205
250
300
106
158
169
205
250
300

Torque
160
161
240
250
365
103
161
161
240
250
365

ETW
3625
4000
4500
4000
6000
2625
3625
4000
4500
4000
6000

SOmph RL Alt regen Net BE%
11.33
15.08
15.84
14.78
22.86
10.82
11.33
15.08
15.84
14.78
22.86

0
0
0
0
0
1
1
1
1
1
1

21.5%
24.0%
21.2%
21.3%
21.8%
25.0%
23.8%
25.8%
23.1%
23.2%
24.0%

predicted
21.5%
23.7%
21.7%
21.0%
21.9%
25.3%
23.5%
25.7%
23.7%
23.0%
23.9%
avg error
% error
0.1%
1.3%
2.3%
1.3%
0.5%
1.3%
1.3%
0.5%
2.3%
0.9%
0.8%
1.1%
Coefficients
Intercept
Torque
ETW
SOmph RL
Alt regen







0.207831
-0.00028
-6.2E-06
0.006531
0.019809







     Figure 1.5-1 Regression data used to establish engine brake efficiency formula

       •  Pumping and friction losses are scaled based on the difference between (brake
          efficiency + accessory losses) and indicated efficiency.  The distribution of
          pumping and friction losses was based on a combination of literature (Patton,
          Heywood8 ) and prior success with values used in the LP model for the 2012-2016
          rule.  It is assumed that pumping and friction losses for fixed valve, naturally
          aspirated engines, distributed over the test cycles, average roughly 60% and 40%
          of total friction, respectively.

       •  Accessory loss (as % of total fuel) is based on a regression of engine torque and
          ETW, and comes directly from Ricardo output file data.

       •  Baseline driveline losses are estimated in the following manner:
L Indicated efficiency data was not included as an output in the Ricardo model. Very little data on indicated
efficiency exists in the literature.  The value of 36% was assumed because it fits fairly well within the LP model,
and it is comparable to the few values presented in the Patton paper.
                                          1-51

-------
                                                 2017 Draft Regulatory Impact Analysis
              a)  Torque converter efficiency, which is a function of (engine torque/power
                  ratio, RL and ETW)

              b)  Transmission efficiency, which is calculated at 87% for 2008 vehicles
                  (based on the average gear efficiency values used by Ricardo in the
                  baseline models) For 4WD vehicles a multiplier of 96.2% is applied to
                  represent the rear axle efficiency

              c)  Losses through the TC and transmission are then determined and added to
                  represent driveline losses as the total % of fuel energy lost.

       •  Baseline tractive wheel energy (the energy delivered to the wheels to actually
          move the vehicle) is a simple relationship of ETW and road load.

       •  The remaining terms (braking losses, inertia load, aero load, and rolling load)
          make up the remainder of the losses and are proportioned similarly to the original
          LP model.

       Reference the "input page" tab in the LP model to see the breakdown for each
predefined vehicle classM.

            1.5.4     Baseline fuel efficiency by vehicle class

       The new LP model estimates the basic fuel energy consumption, Efoei, for an
"unimproved" vehicle (naturally aspirated fixed valve engine with 4 speed automatic
transmission). It is calculated for each vehicle class with Equation 1.5-1:
                                              'wheel
                                         ^engine

                                     Equation 1.5-1
       To estimate the terms in the above equation, EPA regressed several known vehicle
parameters (rated engine power, rated engine torque, ETW, RL (chassis dyno road load at 50
mph)) against simulation output data. Definitions for each term and the relevant parameters
are listed below:
M For the "custom" vehicle class, values were regressed based on the following inputs: rated engine power,
torque, vehicle weight (ETW) and road load, in hp, at 50 mph (from certification data). Note that the defined
vehicle classes were validated by simulation work, while the custom vehicle data was not validated - it is for
illustrative purposes and represents a rougher estimate
                                          1-52

-------
                                                    2017 Draft Regulatory Impact Analysis
           1)  Ewheei: required wheel (or tractive) energy over the city/HW test cycle =
               f(ETW, RL)
           2)  Tlengine: net engine brake efficiency = f(torque, ETW, RL, alternator regenN )
           3)  T|D/L:  driveline efficiency is derived from the losses associated with the torque
               converter,  transmission,  and final drive, where TC losses = f(torque, power,
               RL, ETW) and transmission efficiency is based on vintage of the baseline0
            (kWh) was then converted to fuel economy in mpg by applying the energy
content of gasoline (assumed at 33.7 kWh/gallon - for diesel it is 37.6 kWh/gallon) and
factoring in the distance traveled (10.64 miles) over the combined FTP/HWFE test cycle.

       The LP model predicted baseline fuel economy for each class was then validated to
2008 baseline vehicle simulation results. Baseline unimproved vehicle FE values were first
estimated with the regression as mentioned above.  From there, all other technologies
consistent with the 2008 Ricardo modeled baseline packages were added.  Similarly, the
following technologies were added to the 2008 vehicles for comparison to the 2010 Ricardo
"baseline" packages:  6-speed automatic transmission, higher efficiency gearbox, 12V SS,
alternator regeneration during coastdowns, and 70% efficient alternator.  The predicted LP
fuel economy values of both the 2008 baseline and 2010 vehicles all fall within roughly 2% of
the modeled data, as shown in Figure  1.5-2 below.
          Vehicle
           Class    Trans    EPS   Valvetrain
         Small car  4spdauto    Y     ICP
        Standard car 5 spd auto    N     DCP
         Large car  5 spd auto    N     fixed
         Small MPV  4 spd auto    Y     DCP
         Large MPV  4spdauto    N     fixed
           Truck   4 spd auto    N     CCP
  2008    2008
simulated LP model
 comb.   comb.
  mpg     mpg
  41.5     41.3
  32.0     32.3
  25.5     25.2
  28.8     29.1
  23.1     23.7
  17.6     17.4
%FE
error
-0.5%
0.9%
-1.0%
1.1%
2.4%
-1.1%
  2010    2010
simulated LP model
 comb.   comb.    % FE
 mpg    mpg    error
 43.4    44.1    1.7%
 34.9    34.7    -0.6%
 27.4    27.3    -0.4%
 30.5    31.1    2.0%
 25.2    25.9    2.6%
 18.6    18.6    -0.1%
        2010 packages add 6spd auto trans, higher efficiency gearbox, 12V SS, alternator regen on decel, 70% efficient alternator
 Figure 1.5-2 Comparison of LP model to Ricardo simulation results for 2008 and 2010
                                      baseline vehicles
N When the alternator regeneration technology is included, it changes the efficiency of the engine by moving the
average speed and load to a more efficient operating region. It was included in the definition of the 2010
baseline vehicle models.
0 Two levels of baseline transmission efficiency were included in the simulation work, for 2008 baselines and
2010 baselines ("vintage"). Refer to the Input Page tab in the LP model for more detail.
                                            1-53

-------
                                                2017 Draft Regulatory Impact Analysis
           1.5.5     Identification and calibration of individual technologies

       The next step was to identify the individual technologies of interest and categorize
how they affect the physical system of the vehicle. Engineering judgment was used in
identifying the major loss categories that each individual LP model technology affected. In
some cases two or even three, loss categories were defined that were deemed significant.  Not
all categories were a reduction in losses - some increased the amount of losses (for example,
increased frictional losses for various valvetrain technologies). A list of the technologies and
the categories they affect is shown in Figure 1.5-3 below. The technologies added for this
rule's version of the LP model are highlighted in bold.  For a more detailed description of
each technology, refer to Section 3.4 of the Joint TSD.
                                         1-54

-------
                                                        2017 Draft Regulatory Impact Analysis
         Technology

         Vehicle mass reduction
         Aero Drag Reduction
         Rolling Resistance Reduction
         Low Fric Lubes
         EF Reduction
         4V on 2V Baseline
         ICP
         DCP
         CCP
         Deac
         DVVL
         CVVL
         Turbo/Downsize (gas engines only)
         5-spd gearbox
         6-spd gearbox
         8-spd gearbox
         CVT
         DCTWet
         DCTDry
         Early upshift (formerly ASL)
         Optimized shift strategy
         AggTC Lockup
         High efficiency gearbox (auto)
         12V SS (idle off only)
         High voltage SS, with launch (BAS)
         Alternator regen on braking
         EPS
         Electric access (12V)
         Electric access (high V)
         High efficiency alternator (70%)
         GDI (stoich)
         GDI (stoich)w/cooled EGR
         GDI (lean)
         Diesel-LNT(2008)
         Diesel-SCR(2008)
         Hybrid drivetrain
         Secondary axle disconnect
         Low drag brakes
         Atkinson cycle engine
         Advanced Diesel (2020)
Braking/
 Inertia
Access
Losses
Friction  Pumping    Ind
 Losses    Losses  Efficiency
                                                                             Code:
                                                                                     Major
                                                                                     Minor
                                                                                     Negative
                   Figure 1.5-3 Loss categories affected by each technology
        After losses were identified, EPA calibrated the loss modifiers so that each individual
technology would achieve a nominal effectiveness independent of other technologies and
consistent with the values given in Section 1.2.  For example, discrete variable valve lift
(DVVL) can achieve roughly a 4-5% decrease in GHG emissions.  It is coded in the LP model
                                                1-55

-------
                                                  2017 Draft Regulatory Impact Analysis
as a 27% reduction in pumping losses and a 3% increase (penalty) in friction losses.
Depending on the vehicle class, it reflects an effectiveness ranging from 4.1-5.6% reduction
in the LP model. Other technologies were coded in the LP model in similar fashion. In cases
where more than one loss category was affected, the majority of the effectiveness was linked
to the primary loss category, with the remainder of the effectiveness coded via the other
secondary loss categories.  In some cases the LP model also reflects loss categories that are
penalized with certain technologies - for example, the increased mechanical friction
associated with advanced variable valvetrains (coded as a negative reduction in the LP
model). All technologies were calibrated on an "unimproved" vehicle (without any other
technologies present ) to avoid any synergies from being accidentally incorporated.  Once the
entire list of line-item technologies was coded, the next step was to compare the effectiveness
of actual (Ricardo-modeled) vehicle simulation packages to the LP model results.
            1.5.6     Example build-up of LP package

       The following example package for a Large Car demonstrates how synergies build as
content is added to a vehicle technology package.
505
4V DOHC 14 +EFR2 +LDB +ASL2 +IACC2 +EPS +Aero2 +LRRT2 +HEG +DCP
+GDI+TDS18
8sp DCT-
wet
12V
5%
$1,386
42.6%
       •   Add anytime technologies (EFR2, LDB, ASL2, IACC2, EPS)

       These technologies primarily reduce accessory loads, mechanical engine friction and
pumping losses.  The sum of these technologies is reflected below in Table 1.5-lp and
provides a total of 14.9% reduction in GHG.
                                       Table 1.5-1
        % of tractive energy
        Baseline % of fuel
          Reduction
         % of NEW fuel
Braking / Aero Rolling
Inertia Load Load
Trans
Losses
Access Friction Pumping
Losses Losses Losses
IndEff
Losses
Second
Law
23%    37%
3.9%    6.4%
4%    0%
3.8%    6.4%
3.9%   1.1%    8.3%   5.6%    34.0%
 0%    42%    22%   20%
4.5%   0.6%    6.5%   4.5%    33.9%
       2008 Baseline
         New
Indicated
Efficiency
Mech
Efficiency
Brake
Efficiency
Drivetrain
Efficiency
Cycle
Efficiency
Fuel
Efficiency
Road
Loads
36.0% 58.4% 21.0% 81.6% 100.0% 17.1% 100.0%
36. 1 % 67.9% 24. 5% 8 1 .6% 1 00. 0% 20. 0% 99.2%
30.0%
 n/a
 30%
                                    85.1% Fuel Consumption
                                    14.9% GHGreduction
p For this table and similar subsequent tables, the "Reduction" row refers to the percentage reduction in fuel
energy for each particular loss category. Each values in that row does not translate into an absolute percentage
GHG savings, but are listed as indices between 0% (no reduction) and 100% (maximum theoretical reduction)
for each loss category. For example, in Table 1.5-1, roughly 42% of theoretical accessory losses have been
eliminated associated with the applied anytime technologies.
                                           1-56

-------
                                                2017 Draft Regulatory Impact Analysis
       •  Add road load reductions (Aero2, LRRT2) and 5% mass reduction

       These technologies reduce braking/inertia, aerodynamic and rolling resistance loads,
with a minor degradation in indicated efficiency (because the engine is running at lower
overall loads). Combined with the technologies previously added in 1), the sum of these
technologies is shown below in Table 1.5-2 and provides a total of 24.5% reduction in GHG
compared to an unimproved vehicle.

                                      Table 1.5-2
        % of tractive energy
Braking / Aero Rolling
Inertia Load Load
Trans
Losses
Access Friction Pumping
Losses Losses Losses
Ind Eff
Losses
Second
Law
        Baseline % of fuel
          Reduction
         % of NEW fuel
                                             3.9%
                                                                    34.0%
      2008 Baseline
         New
Indicated
Efficiency
Mech
Efficiency
Brake
Efficiency
Drivetrain
Efficiency
Cycle
Efficiency
Fuel
Efficiency
Road
Loads
36.0% 58.4% 21.0% 81.6% 100.0% 17.1% 100.0%
34.7% 67.9% 23.6% 81.6% 100.0% 19.2% 84.8%
75.5%  Fuel Consumption
24.5%  GHGreduction
       •  Add high efficiency gearbox

       The high efficiency gearbox reduces transmission (driveline) losses due to the
mechanical improvements as described in Section 3.4.2.4 of the Joint TSD. Combined with
the technologies previously added, the sum of these technologies is shown below in Table
1.5-3 and provides a total of 28.5% reduction in GHG compared to an unimproved vehicle.

                                      Table 1.5-3
        % of tractive energy
        Baseline % of fuel
          Reduction
         % of NEW fuel
Braking /
Inertia
23%
3.9%
8%
3.6%
Aero
Load
37%
6.4%
17%
5.3%
Rolling
Load
40%
6.9%
18%
5.6%
Trans
Losses

3.9%
25%
3.3%
Access
Losses

1.1%
42%
0.6%
Friction
Losses

8.3%
22%
6.2%
Pumping
Losses

5.6%
20%
4.3%
Ind Eff
Losses

34.0%

35.3%
Second
Law

30.0%
n/a
30%
      2008 Baseline
         New
Indicated
Efficiency
Mech
Efficiency
Brake
Efficiency
Drivetrain
Efficiency
Cycle
Efficiency
Fuel
Efficiency
Road
Loads
36.0% 58.4% 21.0% 81.6% 100.0% 17.1% 100.0%
34.7% 67.9% 23.6% 86.2% 100.0% 20.3% 84.8%
71.5%  Fuel Consumption
28.5%  GHGreduction
       •  Add dual cam phasing

       Dual cam phasing provides significant pumping loss reductions at the expense of
increased mechanical friction due to the more complex valvetrain demands (as a result, the
"friction loss" reduction value below is actually reduced). Combined with the technologies
previously added, the sum of these technologies is shown below in Table 1.5-4 and provides a
total of 31.4% reduction in GHG compared to an unimproved vehicle.
                                          1-57

-------
                                                 2017 Draft Regulatory Impact Analysis
        o of tractive energy
         laseline % of fuel
          Reduction
         % of NEW fuel
                                      Table 1.5-4
Braking /
Inertia
23%
3.9%
8%
3.6%
Aero
Load
37%
6.4%
17%
5.3%
Rolling
Load
40%
6.9%
18%
5.6%
Trans
Losses

3.9%
25%
3.4%
Access
Losses

1.1%
42%
0.6%
Friction
Losses

8.3%
20%
6.4%
Pumping
Losses

5.6%
39%
3.3%
Ind Eff
Losses

34.0%

35.1%
Second
Law
30.0%
n/a
30%
      2008 Baseline
         New
Indicated
Efficiency
Mech
Efficiency
Brake
Efficiency
Drivetrain
Efficiency
Cycle
Efficiency
Fuel
Efficiency
Road
Loads
36.0% 58.4% 21.0% 81.6% 100.0% 17.1% 100.0%
34.9% 70.4% 24.6% 86.2% 100.0% 21.2% 84.8%
68.6%  Fuel Consumption
31.4%  GHGreduction
       •  Add stoichiometric GDI, downsized, turbocharged engine (18-bar)

       An 18-bar downsized and turbocharged engine, combined with stoichiometric gasoline
direct injection increases an engine's indicated efficiency, and drastically reduces pumping
losses. Combined with the technologies previously added, the sum of these technologies is
shown below in Table  1.5-5 and provides a total of 38.3% reduction in GHG compared to an
unimproved vehicle.

                                      Table 1.5-5
Braking / Aero Rolling
Inertia Load Load
Trans
Losses
Access Friction Pumping
Losses Losses Losses
Ind Eff
Losses
Second
Law
        % of tractive energy
         laseline % of fuel
          Reduction
         % of NEW fuel
2008 Baseline
New
Indicated
Efficiency
Mech
Efficiency
Brake
Efficiency
Drivetrain
Efficiency
Cycle
Efficiency
Fuel
Efficiency
Road
Loads
36.0% 58.4% 21.0% 81.6% 100.0% 17.1% 100.0%
36.6% 74.7% 27.3% 86.2% 100.0% 23.6% 84.8%
                                                              61.7% Fuel Consumption
                                                              38.3% GHGreduction
       •  Add 8-speed wet clutch DCT

       An 8-speed wet clutch DCT reduces losses in several ways. The elimination of the
planetary gearset and torque converter increases the reduction in transmission losses, while
engine pumping losses are further reduced with the addition of more fixed gears (allowing for
more efficient engine operation). Combined with the technologies previously added, the sum
of these technologies is shown below in Table 1.5-6 and provides a total of 42.6% reduction
in GHG compared to an unimproved vehicle.
                                          1-58

-------
                                                2017 Draft Regulatory Impact Analysis
        o of tractive energy
         laseline % of fuel
          Reduction
         % of NEW fuel
                                     Table 1.5-6
Braking /
Inertia
23%
3.9%
8%
3.6%
Aero
Load
37%
6.4%
17%
5.3%
Rolling
Load
40%
6.9%
18%
5.6%
Trans
Losses

3.9%
48%
2.7%
Access
Losses

1.1%
42%
0.6%
Friction
Losses

8.3%
20%
6.8%
Pumping
Losses

5.6%
72%
1.6%
Ind Eff
Losses

34.0%

32.9%
Second
Law
30.0%
n/a
30%
2008 Baseline
New
Indicated
Efficiency
Mech
Efficiency
Brake
Efficiency
Drivetrain
Efficiency
Cycle
Efficiency
Fuel
Efficiency
Road
Loads
36.0% 58.4% 21.0% 81.6% 100.0% 17.1% 100.0%
37.1% 75.5% 28.0% 90.5% 100.0% 25.3% 84.8%
                                                             57.4%  Fuel Consumption
                                                             42.6%  GHGreduction
       In summary, for this technology package, the mathematical combination of individual
effectiveness values (added without synergies) would yield a GHG reduction value of about
50%. Based on the lumped parameter model - which is calibrated to vehicle simulation
results that include synergies - this technology package would provide a GHG reduction of
42.6%.  In most cases negative synergies develop between technologies addressing the same
losses, and with increasing magnitude as the level of applied technology grows. This
increasing disparity is shown below in Table 1.5-7.
       Table 1.5-7: Comparison of LP-predicted to gross aggregate effectiveness
Technologies
Added

EFR2, LDB, ASL2, IACC2, EPS
Aero 2, LRRT2, MRS
HEG
DCP
GDI,TDS18
8spDCT-wet
Individual
Effectiveness
(for step)
16.4%
10.8%
5.3%
5.5%
14.9%
11.9%
Combined
Effectiveness
LP total
14.9%
24.5%
28.5%
31.4%
38.3%
42.6%
Gross
Effectiveness
total
16.4%
25.5%
29.4%
33.3%
43.2%
50.0%
           1.5.7
Calibration of LP results to vehicle simulation results
       The LP model includes a majority of the new technologies being considered as part of
this proposed rulemaking.  The results from the 2011 Ricardo vehicle simulation project
(Joint TSD, Section 3.3-1) were used to successfully calibrate the predictive accuracy and the
synergy calculations that occur within the LP model. When the vehicle packages Ricardo
modeled are estimated in the lumped parameter model, the results are comparable.  All of the
baselines for each vehicle class, as predicted by the LP model, fall within 3% of the Ricardo-
modeled baseline results.  With a few exceptions (discussed in 1.5.8), the lumped parameter
results for the 2020-2025 "nominal" technology packages are within 5% of the vehicle
simulation results.  Shown below in Figure 1.5-4 through Figure 1.5-9 are Ricardo's vehicle
                                         1-59

-------
                                               2017 Draft Regulatory Impact Analysis
simulation package results (for conventional stop-start and P2 hybrid packages'2) compared to
the lumped parameter estimates.
                            Small Car Nominal Results
        Figure 1.5-4 Comparison of LP to simulation results for Small Car class
Q Refer to Joint TSD, Section 3.3-1 for definitions of the baselines, "conventional stop-start" and "P2 hybrid"
vehicle architectures.
                                         1-60

-------
                                       2017 Draft Regulatory Impact Analysis
                   Standard Car Nominal Results
                                                             I Ricardo
                                                             I LP results
                         ConventionaISS
                                      P2 Hybrid
Figure 1.5-5 Comparison of LP to simulation results for Standard Car class
                     Large Car Nominal Results
     60

     50

     40

     30

     20

     10
I   I   I    I
I   I   I    I
III
III
III
III
I Ricardo
I LP results

                         ConventionaISS
                                       P2 Hybrid
  Figure 1.5-6 Comparison of LP to simulation results for Large Car class
                                 1-61

-------
                                      2017 Draft Regulatory Impact Analysis
                   Small MPV Nominal Results
    60






    50






    40






    30






    20






    10
n    •   •   •   •    •    •
     •    •   •   •   •    •    •
     •    •   •   •   •    •    •
I Ricardo



I LP results
                           ^    ,
-------
                                        2017 Draft Regulatory Impact Analysis
                       Truck Nominal Results
                                                                Ricardo

                                                                LP results
   Figure 1.5-9 Comparison of LP to simulation results for Truck class



    1.5.8    Notable differences between LP model and Ricardo results

1.5.8.1 Small car

       At first glance, it would appear that the results for small cars predicted by the
lumped parameter model- (especially hybrids) are too high when compared to the
Ricardo vehicle simulation results. However, further investigation of the simulation
results showed that the applied road load coefficients for the small car, as modeled by
Ricardo, may have been higher than they should have been.  Figure 1.5-10, below,
shows road load power (in units of horsepower, or RLHP) plotted as a function of
vehicle speed for the simulated vehicles. As expected, road load curves decrease as
the vehicle class (weight and size) decreases.  The road load coefficients used by
Ricardo were all taken from certification test data. As shown, the modeled Yaris
(small car) road load curve, in purple, is actually comparable to that for a Camry (the
standard car exemplar vehicle), shown in green. By investigating the certification test
data, EPA identified a second (alternate) road load curve for an alternative Yaris
vehicle configuration, shown as a dashed line. Applying the mathematical  equivalent
of this alternate road load curve to the small car in the vehicle simulation Complex
Systems tool (described in the Joint TSD, Section 3.3.1) achieved results much closer
to those predicted by the LP model. While both Yaris road load curves are based on
actual certification coefficients, it would make sense that the small car class should
exhibit lower road loads than a standard car class.
                                  1-63

-------
                                        2017 Draft Regulatory Impact Analysis
                                  RLHP comparison
           Figure 1.5-10 Road load power for modeled vehicles

       The LP results for the small car P2 hybrids appear to deviate further.
However, the deviation can be explained due to two main factors. Aside from the
higher road load curve employed by Ricardo, the small car P2 hybrid effectiveness
was understated due to a relatively undersized nominal motor/generator (30% smaller
than the optimal motor size of 21 kW). The percentage of available braking energy
did not match levels seen with the other vehicle classes, and fuel economy suffered
slightly as a result.

       For these reasons, EPA finds the LP model estimate for the small car class to
be more appropriate for package effectiveness estimates.

1.5.8.2 Diesels

       Detailed analysis of the diesel vehicle simulation results showed that the
vehicles did not operate in the most efficient operating region, either due to a potential
inconsistency in the application of the optimized shift strategy and/or due to the
apparent oversizing of the nominal diesel engines. Diesel engines appeared to have
been initially sized for rated power, not torque, which led to oversized displacement.
This conversely reduced the average transmission efficiency realized in the model test
runs.   Plotting the average engine speed and load operating points for the diesel
simulation data on top of the diesel engine maps showed that there was room for
improvement in choice of selected gear, for example.  EPA's LP estimate for the
Ricardo diesel packages compare well with the simulation results when optimized
shifting and early torque converter lockup  (for automatic transmissions) are excluded
                                  1-64

-------
                                                 2017 Draft Regulatory Impact Analysis
         from the LP model. Based on this comparison which is more consistent with the
         technology that appeared to be modeled, EPA is more comfortable with the LP diesel
         estimates which have slightly higher effectiveness estimates than the diesel package
         vehicle simulation results.

             1.5.9     Comparison of results to real-world examples

         To validate the lumped parameter model, representations of actual late-model
  production vehicles exhibiting advanced technologies were created. Shown below in Table
  1.5-8 are a set of select vehicle models containing a diverse array of technologies: included
  are the pertinent technologies and vehicle specifications, along with actual vehicle
  certification fuel  economy test data compared to the lumped parameter fuel economy
  estimates. For the vehicles and technologies shown, the predicted fuel economy is within
  about 3% of the actual data.
Table 1.5-8  Production vehicle certification data compared to lumped parameter predictions
Vehicle
Vehicle class
Engine
Transmission
HEV motor (kW)
ETW(lbs)
City/HW FE (mpg)
LP estimate (mpg)
Key technologies applied
in LP model
2011ChevyCruzeECO
Small Car
1.4LI4
turbo GDI
6 speed auto
n/a
3375
40.3
40.2
GDI (stoich.)
turbo (30%downsize)
ultra low R tires
active grill shutters
2011 Sonata Hybrid
Standard Car
2.4LI4
Atkinson
6 speed DCT
30
3750
52.2
51.7
P2 hybrid
aero improvements
2011 Escape Hybrid
Small MPV
2.5LI4
Atkinson
CVT
67
4000
43.9
44.0
Powersplit hybrid
2011 F-150 Ecoboost
Truck
3.5LV6
turbo GDI
6 speed auto
n/a
6000
22.6
21.9
GDI (stoich)
turbo (37%downsize)
                                          1-65

-------
                                            2017 Draft Regulatory Impact Analysis
                                      References

2 "OMEGA Master-sets and Ranked-sets of Packages," Memorandum to Air Docket EPA-
HQ-OAR-2010-0799 from Todd Sherwood, November 10, 2011.

3 "OMEGA Master-sets and Ranked-sets of Packages," Memorandum to Air Docket EPA-
HQ-OAR-2010-0799 from Todd Sherwood, November 10, 2011.

4 National Research Council, "Effectiveness and Impact of Corporate Average Fuel Economy
(CAFE) Standards", National Academy Press, 2002.

5 Patton, et al. "Aggregating Technologies for reduced Fuel Consumption: A Review of the
Technical Content in the 2002 National Research Council Report on CAFE".  SAE 2002-01-
0628. Society of Automotive Engineers, 2002.

6 Ref OMEGA description in RIA of 2012-2016 final rule

7 U.S. EPA, "Project Report: Computer Simulation of Light-Duty Vehicle Technologies for
Greenhouse Gas Emission Reduction in the 2020-2025 Timeframe", Contract No. EP-C-11-
007, Work Assignment 0-12, Date TBD, Report# TBD, Docket EPA-HQ-OAR-2010-0799.

8 Heywood, J.  Internal Combustion Engine Fundamentals. Figures 13-9 and  13-10, p. 723.
McGraw-Hill,  1988.
                                      1-66

-------
                                                     2017 Regulatory Impact Analysis
2      EPA's Vehicle Simulation Tool

2.1 Introduction

       2.1.1  Background

       It is well known that full-scale physics-based vehicle simulation modeling is the most
sophisticated method for estimating fuel saving benefits by a package of advanced new
technologies (short of actually building an actual prototype). For this reason, EPA has used
full vehicle simulation results generated by Ricardo, Inc. to calibrate and validate the lumped
parameter model to estimate technology effectiveness of many combinations of different
technologies. However, EPA only has limited access to the Ricardo's model and proprietary
data, so there has been a growing need for developing and running detailed vehicle
simulations in-house for GHG regulatory and compliance purposes (notwithstanding that it
this is a very time-consuming and resource-intensive task). As a result, over the past year,
EPA has begun to develop full vehicle simulation capabilities in order to support regulations
and vehicle compliance by quantifying the effectiveness of different technologies with
scientific rigor over a wide range of engine and vehicle operating conditions. This in-house
vehicle simulation tool has been developed for modeling a wide variety of light, medium, and
heavy-duty vehicle applications over various driving cycles. The first application of this
vehicle simulation tool was intended for medium and heavy-duty vehicle compliance and
certification. This simulation tool, the "Greenhouse gas Emissions Model" (GEM), has been
peer-reviewed9 and has also recently been published.10 For the model years 2014 to 2017
final rule for medium and heavy-duty trucks,) GEM is used both to assess Class 2b-8
vocational vehicle and Class 7/8 combination tractor GHG emissions and to demonstrate
compliance with the vocational vehicle and combination tractor standards.  See 40 CFR
sections 1037.520 and 1037.810 (c)(l).Objective and Scope

       Unlike in the heavy-duty program , where the vehicle simulation tool is used for GHG
certification since chassis-based certifications are not yet practical or feasible for most HD
vehicles, we intend to use the light duty  simulation tool to develop the light duty regulatory
program but not for certification  since it is not only feasible but common practice to certify
light duty vehicles based on chassis-based vehicle testing. For light-duty vehicles, EPA has
been developing this simulation tool for non-hybrid, hybrid, and electric vehicles, which is
capable of simulating a wide range of conventional  and advanced engines, transmissions, and
vehicle technologies over various driving cycles. The tool evaluates technology package
effectiveness while taking into account synergy effects among vehicle components and
estimates GHG emissions for various combinations of future technologies.  This LD vehicle
simulation tool is capable of providing reasonably (though not absolutely) certain predictions
of the fuel economy and GHG emissions of specific vehicles to be produced in  the future, It
is also capable of simulating non-hybrid vehicles with a Dual-Clutch Transmission (DCT),
under warmed-up conditions only. Additional simulation capabilities such as automatic
transmissions, cold-start conditions, engine start-stops, and hybrid/electric vehicles are being
developed by EPA for the final rule.  In this proposal, we are using the current simulation tool
in  a more limited manner: to determine the maximum credit potential for A/C efficiency and

                                         2-1

-------
Chapter 2	

to determine the default credit value for the pre-defined active aerodynamic and electrical
load off-cycle technologies. See section 2.3 below.

       The simulation tool is a full vehicle simulator that uses the same physical principles as
commercially available vehicle simulation tools (such as Autonomie, AVL-CRUISE, GT-
Drive, Ricardo-EASYS, etc.).  In order to ensure transparency of the models and free public
access, EPA has developed this tool in MATLAB/Simulink environment with a completely
open source code. For the 2017 to 2025 GHG proposal, EPA used the simulation tool to
quantify the amount of GHG emissions reduced by improvements in A/C systems and off-
cycle technologies, as explained in Chapter 5 of the Joint TSD and Section III.C of the
Preamble.

2.2 Descriptions of EPA's Vehicle Simulation Tool

       2.2.1   Overall Architecture

       Table 2.2-1 provides a high-level architecture of the light-duty (LD) vehicle
simulation model, which consists of six systems: Ambient, Driver, Electric, Engine,
Transmission,  and Vehicle.  With the exception of "Ambient" and "Driver" systems, each
system consists of one or more component models which represent physical elements within
the corresponding system.  The definition and function of each system and their respective
component models are discussed in the next section.

                Table 2.2-1  High-Level Structure of Vehicle Simulator
System
Ambient
Driver
Electric
Engine
Transmission
Vehicle
Component Models
n/a
n/a
Accessory (electrical)
Accessory (mechanical), Cylinder
Clutch, Gear
Final Drive, Differential, Axle, Tire,
Chassis
       Figure 2.2-1 illustrates the overall streamline process of the vehicle simulation and
how the current tool is designed for a user to run desired vehicle simulations. Upon execution
of the main MATLAB script, it prompts the user to enter desired inputs such as vehicle type,
engine technology type, driving cycle, etc. Then, it initializes all necessary vehicle model
parameters including engine maps, transmission gear ratios, and vehicle road load parameters.
After the initialization, the script runs the Simulink vehicle model over the desired driving
cycles. Upon completing the simulation, it automatically displays the simulation outputs in
terms of fuel economy and GHG emissions.  It also displays a plot of the simulated vehicle
speed trace, showing how closely the simulation vehicle followed the desired speed trace.

       Although this version of the vehicle simulation tool is still in an early development
stage and provides  only a handful of simulation capabilities in terms of vehicle types, engine
and transmission technologies, and driving cycles, it is undergoing constant upgrades and
improvements to include more technology choices and simulation flexibilities.  In fact, the
                                         2-2

-------
                                                   2017 Regulatory Impact Analysis
first official version of the tool will have a Graphical User Interface (GUI) which will allow
the user to choose from different technologies and other simulation options while making the
use of the tool much easier and straightforward. The Section 2.4.2 will discuss and address
these additional choices and simulation capabilities that are being planned for the improved
version of the tool.

9 "Peer Review of the Greenhouse gas Emissions Model (GEM) and EPA's Response to
Comments," Docket EPA-HQ-OAR-2010-0162-3418, Publication Number: EPA-420-R-11-
007, July 2011.
10 Lee, S., Lee, B., Zheng, H., Sze, C., Quinones, L., and Sanchez, J., "Development of
Greenhouse Gas Emissions Model for 2014-2017 Heavy- and Medium-Duty Vehicle
Compliance," SAE 2011 Commercial Vehicle Engineering Congress, Chicago, September
2011, SAE Paper 2011-01-2188.
                                        2-3

-------
Chapter 2
           Pre-Processing
 *    LD Vehicle Jlwiel for Study of Jt/C Lopd Effect    $
                         L JIPG £ GHG
   Dsftt Selections '
   » Vehicle Type:  Toyota Y«rt« • 0
               To ata Camry • 1
               Ch ysler 300 - 2
               Fo a F1SO  • 3
    Enter Youc Select! n	
   » Driving Cycle:  FTP  - t>
               wt ir - i
               3C 3 - 2
    Entec Tout Select!
    - Engine Typ<:   f--~. eline Engine * 0
               EG  Boost Engine = 1
    Enter Youc Select! n 	
   » A/C Usage:    A/  O±£ - 0
               A/  On =1
    Enter Youc Selecti n 	
                                                          Simulation  Run
              Post-Processing
* SC03 Cycle  Simulation  *
  Percent Time Hissed by 2roph -   1.32 %
  Fuel Consumption (Total)     -  31.86 mpg
  C02 Emission                 - 278.95 g/mile
                                                                                                                      SC03 Cycle Simulation


'



J |Mt.m>«4 J> 	 fe

^

l'"^-""'-'""* ^— ^
iS"**">_t»JS tr*na_bus



t»»n«_«pd_oiJ t--»n> _spd_tfi
_^<^ II.JMI j | t**t.i*_h«*l ^> 	 ft,



"»< l'-'"--<'-'"l | |-£t— 	 >•
iy3**'I»_E!^II VtH_t>U»



v*.»d_«« ^_,M_.n
                                                   Figure 2.2-1  LD Vehicle Simulation Tool
                                                                          2-4

-------
                                                     2017 Regulatory Impact Analysis
       2.2.2   System Models

       In this section, detailed descriptions of the system models (Ambient, Driver, Electric,
Engine, Transmission, and Vehicle) are provided. For Electric, Engine, Transmission, and
Vehicle systems, the components within each of the systems will be described as well. These
system models remain consistent regardless of vehicle types, engine or transmission
technologies, and driving cycles.

       2.2.2.1  Ambient System

       This system defines surrounding environment conditions, such as pressure,
temperature, and road gradient, where vehicle operations are simulated. By default, the
environmental conditions defined in this system are in accordance with the standard SAE
practices - air temperature of 25°C, air pressure of 101.325 kPa, and air density based on the
                                                3
                                                                            I OX
Ideal Gas law which results in a density of 1.20 kg/m .  The road gradient is set to 0
indicating a vehicle moving on a flat surface.  However, these conditions are easily
reconfigurable by the user.

       2.2.2.2 Driver System

       The driver model utilizes two control schemes to keep the simulated vehicle speed at
the desired values: feedforward and feedback.  It uses the targeted vehicle speed defined by a
desired driving cycle to first estimate vehicle's torque requirement at the wheel at any given
time.  The engine power demand is then calculated based on the required wheel torque. And,
the required accelerator and braking pedal positions are determined to deliver the demanded
engine power which will drive the vehicle at the desired speed. If the simulated vehicle speed
deviates the desired target, a speed correction logic is applied via a classical proportional-
integral-derivative (PID) controller to adjust the accelerator and braking pedal positions by
necessary  amount in order to maintain the targeted vehicle speed at every simulation time
step.

       2.2.2.3 Electric System

       The electric system was originally modeled as a system which consists of four
individual electrical components - starter, electrical energy storage  such as battery, alternator,
and electrical accessory. However, for the purpose of calculating A/C credits as well as off-
cycle credits, the simulation tool has modeled the electrical system as a constant power
consumption devise as a function of the vehicle category. It basically represents the power
loss associated with the starter, alternator, and other electrical accessories.  This type of
simplification was made since the purpose of the simulation was A-B comparisons only, i.e.
relative difference between case A and case B on GHG emissions.

       2.2.2.4 Engine System

       The engine system mainly consists of two components: Mechanical Accessory and
Cylinder, which represent torque loss and torque production by an engine, respectively.

                                         2-5

-------
Chapter 2	

       2.2.2.4.1 Mechanical Accessory

       This component is modeled as a simple power consumption source. Most vehicles run
a number of accessories that are driven by mechanical power generated from the engine
crankshaft rotation.  Some of these accessories are necessary for the vehicle to run, like the
coolant pump, while others are only used occasionally at the operator's discretion, such as the
air conditioning compressor. For estimating the impact of A/C usage on fuel consumption,
the mechanical accessory is modeled as a power consumption devise which varies with engine
speed.  More detailed description of the A/C compressor model  is provided in the next
section.

       2.2.2.4.2 Cylinder

       The cylinder component is modeled based on engine torque curves at wide open
throttle (maximum torque) and closed throttle (minimum torque) as well as a steady-state fuel
map covering a wide range of engine speed and torque conditions. The engine fuel map is
represented as fueling rates pre-defined in engine speed and load conditions. This part of the
model is not physics-based, therefore does not attempt to model the in-cylinder combustion
and the corresponding torque production process. During the vehicle simulation, the
instantaneous engine torque and speed are monitored and used to select an appropriate fueling
rate based on the fuel map. This map is adjusted automatically by taking into account three
different driving modes:  acceleration, brake, and coast. The fuel map, torque curves, and the
different driving modes are pre-programmed into the model for  several different engine
technologies.

       2.2.2.5 Transmission System

       The transmission system consists of two components: Clutch and Gear.  The current
version of the transmission system only models a DCT.

       2.2.2.5.1 Clutch

       This component represents a mechanical clutch in either a manual transmission or a
DCT. For an  automatic transmission, it can be replaced by a torque converter component. It
is modeled as an ideal clutch, where no dynamics during clutch  slip is considered during
clutch engaging and disengaging process.

       2.2.2.5.2 Gear

       This component is modeled as a simple gearbox. The number of gears and
corresponding gear ratios are predefined during the preprocessing of simulation runs. Also,
torque transmitting efficiency is defined for each gear to represent the losses that occur in the
physical system. Like the clutch component, the gear is modeled as an ideal gear, where no
dynamics is considered during gear engaging and disengaging process.
                                         2-6

-------
                                                     2017 Regulatory Impact Analysis
       2.2.2.6 Vehicle System

       The vehicle system consists of five components: Final Drive, Differential, Axle, Tire,
Chassis.  It basically models all components after transmission in a vehicle.

       2.2.2.6.1 Final Drive and Differential

       Both final drive and differential components are modeled as mechanical systems
which transmit inertia and toque from an upstream component to a downstream component
with a certain gear ratio and efficiency.  The gear ratios for both components can be specified
by the user according to the simulated vehicle. The torque transmitting efficiencies are
defined by maps based on input speed and torque to the modeled component.

       2.2.2.6.2 Axle

       Typically, all axles are lumped together, and one axle model represents the overall
behavior of vehicle axles during vehicle simulations.  In the LD vehicle simulation tool,
however, the axle component is modeled to simulate the behavior of each individual axle used
by the simulated vehicle. The axle is treated individually in order to properly simulate all
wheel drive vehicle types.

       2.2.2.6.3 Tire and Chassis

       This part of the vehicle system models the body of the vehicle including tires.  For the
chassis component, the coefficient of aerodynamic drag, mass of vehicle, and vehicle frontal
area are the key model parameters. For tire component, the user specifies the configuration of
each axle on the vehicle, including the tire diameter and its rolling resistance coefficient.
However, these components will have a capability to use typical coast-down coefficients to
calculate road load, instead of tire rolling resistance and aerodynamic drag.

2.3 Applications of Simulation Tool for the Proposed Rule

       As mentioned previously, EPA used the vehicle simulation tool for the proposed rule
to quantify the amount of GHG emissions reduced by improvements in A/C system efficiency
(thus fixing the maximum credit potential) and to determine the default credit value for active
aerodynamics — one of the listed off-cycle technologies (off-cycle technologies for which a
credit of pre-determined amount may be obtained).  . In this section, we discuss the specifics
of these applications of the simulation tool.  Impact of A/C on Fuel Consumption

       Among the simulation model systems  described in the previous section, there are four
key system elements in the light-duty vehicle  simulation tool which describe the overall
vehicle dynamics behavior and the corresponding fuel efficiency: electric, engine,
transmission, and vehicle.  The electric system model consists of parasitic electrical load and
A/C blower fan, both of which were assumed  to be  constant. The engine system model is
comprised of engine torque and fueling maps.  For estimating indirect A/C impact on fuel
consumption increase, two engine maps were  used: baseline and EGR boost engines.  These
engine maps were obtained by reverse-engineering the vehicle simulation results provided by
                                         2-7

-------
Chapter 2	

Ricardo Inc. For the transmission system, a Dual-Clutch Transmission (DCT) model was
used along with the gear ratios and shifting schedules used for the earlier Ricardo simulation
work. For the vehicle system, four vehicles were modeled: small, medium, large size
passenger vehicles, and a light-duty pick-up truck.  The transient behavior and
thermodynamic properties of the A/C system was not explicitly simulated, in favor of a
simpler approach of capturing the compressor load based on national average ambient
conditions.  We believe this simplification is justified since the goal is to capture the behavior
on the average of a fleet of vehicles (not an individual make or model).

       In order to properly represent average load values to the engine caused by various A/C
compressors in various vehicle types, EPA has adopted  the power consumption curves of A/C
systems, published by an A/C equipment supplier, Delphi.11'12 Also, in an effort to
characterize an average A/C compressor load in the presence of widely varying environmental
conditions in the United States, EPA has adopted data from the National Renewable Energy
Laboratory (NREL) to estimate environmental conditions associated with typical vehicle A/C
usage.13'14'15 Based on the NREL data, EPA selected an A/C power consumption curve as a
function of engine speed that was acquired by Delphi at 27°C and 60% relative humidity as a
representative average condition. This power consumption data was taken from a fixed
displacement compressor with a displacement volume of 210 cc. The curve includes the
effect of compressor cycling as well as non-summer defrost/defog usage. In order to associate
each vehicle type with appropriate A/C compressor displacement, EPA scaled the curve based
on the displacement volume ratio.  For determining indirect A/C impact on fuel consumption
increase for various vehicle types, EPA estimated A/C compressor sizes of 120 cc, 140 cc,
160 cc, and 190 cc for small, medium, large passenger cars, and light-duty pick-up truck,
respectively. By applying these ratios to the 210 cc power consumption curve, EPA created
A/C load curves for four vehicle types, as shown in Figure 2.3-1.
                               A/C Load Demand
        6
        0-
••-210 cc

	190 cc

	160 cc

	140 cc

—-120 cc
             500 1000 1500 2000 2500 3000 3500 4000 4500 5000 5500 6000 6500

                                 Engine Speed (RPM)
              Figure 2.3-1 Representative A/C Compressor Load Curves
                                        2-8

-------
                                                      2017 Regulatory Impact Analysis
        With these A/C compressor load curves, EPA ran full vehicle simulations based on the
  following matrix shown below. In this matrix, the baseline engine represents a typical Spark-
  Ignition (SI), Port-Fuel Injection (PFI), Naturally-Aspirated (NA) engine equipped with a
  Variable Value Actuation (VVA) technology. In this technology, the valve timing (both
  intake and exhaust) is continuously varied over a wide range of engine operating conditions in
  order to result in optimal engine breathing efficiency.  On the other hand, the EGR boost
  engine uses turbocharging and cooled EGR to increase engine's Brake Mean Effective
  Pressure (BMEP) level while managing combustion and exhaust temperatures.  This engine
  usually has a peak BMEP of 25 to 30 bars, which supports significant downsizing (e.g. about
  50%) compared to the baseline engines. Table 2.3-1 provides simulation results over SC03
  driving cycle with an EGR boost engine for various vehicle classes.

               •  Small, medium, large cars, and pick-up truck

               •  FTP, Highway, and SC03  cycles

               •  Baseline and EGR boost engines

               •  A/C off and A/C on

Table 2.3-1 Vehicle Simulation Results on CO2 Emissions over SC03 Cycle with EGR Boost
                                        Engine
SC03 Cycle
CO2 with A/C off
CO2 Increase with A/C on
Total CO2 with A/C
Indirect A/C Fuel Use
[g/mi]
[g/mi]
[g/mi]
[%]
Small Car
196.4
11.7
208.1
5.6
Medium Car
235.7
12.0
247.7
4.8
Large Car
293.7
13.8
307.5
4.5
Truck
472.4
17.2
489.6
3.5
        EPA ran the SC03 cycle simulations instead of the FTP/Highway combined cycle
  simulations so that the simulation results would represent the actual A/C cycle test. EPA also
  assumed the EGR boost engine during vehicle simulations because the EGR boost engine
  better represents an engine technology more likely to be implemented in model years 2017 to
  2025 and because the A/C impact on CC>2 increase in the EGR boost engine is similar to that
  in the baseline engine as shown in Table 2.3-1 and Table 2.3-2. Details of this analysis which
  showed impact of A/C usage on fuel consumption is relatively independent  of engine
  technology are provided in the next section. Moreover, EPA assumed 62%  and 38% of
  market penetrations for manual and automatic climate control systems, respectively.  EPA
  also assumed 23.9% and 35.0% of A/C on-time for manual and automatic climate control
  systems, respectively.  These are the same assumptions made for the 2012-2016 rule.16 In
  order to come up with the overall impact of A/C usage on CC>2 emissions for passenger cars,
  the simulation results for cars shown in Table 2.3-1 were sales-weighted for each year from
  2017 to 2025. For the end result, the impact of A/C usage was estimated at  11.9 CC>2 g/mile
  for cars and 17.2 CC>2 g/mile for trucks.  This corresponds to an impact of approximately 14.0
                                          2-9

-------
    Chapter 2	

    CO2 g/mile for the (2012) fleet, which is comparable to the 2012-2016 final rule result, but
    still lower than the two studies by NREL14 and NESCCAF15 cited above.

           2.3.1.1 Effect of Engine Technology on Fuel Consumption by A/C System

           In order to continue to maintain the credit levels from the 2012-2016 rule, EPA had to
    first demonstrate that the fuel economy and CC>2 emissions due to A/C was relatively
    insensitive to the engine technologies that may be expected to be used in 2012-2016 light duty
    vehicles .  If, for example, more efficient engines are able to run the A/C system more
    efficiently such that the incremental increase in emissions due to A/C decreased compared to
    the base engines, then credits for the same A/C technologies must decrease over time as
    engines become more efficient. This would correspond to a decrease in credits proportional
    (or multiplicative) to the increase in efficiency of the engine. Conversely, if the incremental
    increase in emissions due to A/C remained relatively constant, then the credits available for
    A/C efficiency should also remain stable. This would correspond to  the credits (A/C impact)
    being additive to the base emissions rate, thus being independent of engine efficiency.  The
    EPA based the hypothesis on the latter assumption.

           In order to prove out this hypothesis, EPA carried out vehicle simulations for several
    cases, including two engine technologies: baseline and EGR boost engines (a surrogate for a
    future advanced efficient engine). Table 2.3-2 shows the vehicle simulation results of CO2
    emissions over the SC03 driving cycle when baseline engines are used, as opposed to the
    advanced EGR boost engines. By comparing the values of CC>2 increase with A/C on in Table
    2.3-1 and Table 2.3-2, it is evident that the impact of A/C usage on fuel consumption is not
    very dependent on the engine technologies.  In fact, the difference in the CO2 increase with
    A/C on (2nd row in table) between the  emissions  from the baseline and EGR boost engines is
    less than 10% for all vehicle classes.

Table 2.3-2 Vehicle Simulation Results on CO2 Emissions over SC03 Cycle with Baseline Engine
SC03 Cycle
CO2 with A/C off
CO2 Increase with A/C on
Total CO2 with A/C
Indirect A/C Fuel Use
[g/mi]
[g/mi]
[g/mi]
[%]
Small Car
259.3
11.3
270.6
4.2
Medium Car
348.0
11.1
359.1
3.1
Large Car
425.4
12.5
437.9
2.9
Truck
628.1
16.2
644.3
2.5
           Figure 2.3-2 depicts zoomed-in BSFC maps for baseline and EGR boost engines. The
    circles on these maps represent average operating conditions of the engines over the FTP
    (city) drive cycle. The blue circle represents a simulated average operating condition without
    A/C while the red circle represents an average operating condition with A/C. As can be seen
    in the figure, the engines operate at higher load levels when the A/C is on.

           For the baseline engine case, the engine efficiency improves significantly (375 g/kW-h
    to almost 330 g/kW-h) as it moves along the BSFC surface, whereas the improvement is
    much less for the EGR boost engine as it moves from approximately 250 g/kW-h to 240
    g/kW-h.  However, the large improvement in engine  efficiency for the baseline engine is
                                            2-10

-------
                                                      2017 Regulatory Impact Analysis
offset by the fact that the engine itself is less efficient than the EGR boost engine.
Conversely, the small efficiency improvement for the EGR boost engine is compensated by
the fact that the engine is much more efficient than the baseline engine.  As a result, the CC>2
increase seen by both engines due to A/C usage becomes similar in two different
technologies. This result allows us to approximate the A/C impact on vehicle fuel
consumption as an additive effect rather than a multiplicative effect since it is independent of
engine technologies.  For the same reason, it also means that A/C credits for a given
technology can remain constant over time, which will greatly simplify the progression of
future credits.R
100
 80
 GO
 40
      100
       80
       60
	   40
   Figure 2.3-2  Average Engine Operating Conditions with A/C Off and A/C On over
                  Fueling Maps for Baseline and EGR Boost Engines

       2.3.2   Off-Cycle Credit Calculation

       The aerodynamics of a vehicle plays an important role in determining fuel economy.
Improving the aerodynamics of a vehicle reduces drag forces that the engine must overcome
to propel the vehicle, resulting in lower fuel consumption. The aerodynamic efficiency of a
vehicle is usually captured in a coast-down test that is used to determine the dynamometer
parameters used  during both  the two-cycle and five-cycle tests. This section discusses active
aerodynamic technologies that are activated only at certain speeds to improve aerodynamic
efficiency while  preserving other vehicle attributes or functions. Two examples of active
aerodynamic technologies are active grill shutters and active ride height control. Active
 It also means that the last row in the above two tables are somewhat misleading as A/C impact should not be
quantified as a fraction of the total emissions, but rather an additive increment. The numbers are left onto the
tables only for comparison purposes to studies in the literature that use this convention.
                                         2-11

-------
Chapter 2	

aerodynamic features can change the aerodynamics of the vehicle according to how the
vehicle is operating, and the benefit of these vehicle attributes may not be fully captured
during the EPA test cycles.

       EPA is proposing to limit credits to active aerodynamic systems only (not passive).
The aerodynamic drag on the vehicle is highly dependent on the vehicle shape, and the
vehicle shape is (in turn) highly dependent on the design characteristics for that brand and
model. EPA feels that it would be inappropriate to grant off-cycle credits for vehicle aesthetic
and design qualities that are passive and fundamentally inherent to the vehicle.

       2.3.2.1 Performance-Based Metrics

       To evaluate technologies that reduce aerodynamic drag, the EPA conducted an
analysis of the reduction in emissions corresponding to a general reduction of aerodynamic
drag on a vehicle. Using the EPA's full vehicle simulation tool described in the previous
section, the agency evaluated the change in fuel consumption for increasing reductions in
aerodynamic drag for a typically configured vehicle.  The results of this analysis form the
basis for  a consistent methodology that the EPA applied to technologies that provide active
aerodynamic improvements.

       Vehicle aerodynamic properties impact both the combined FTP/Highway and 5-cycle
tests. However, these impacts are larger at higher speeds and have a larger impact on the 5-
cycle tests. By their nature of being "active" technologies, EPA understands that active
aerodynamic technologies will not be in use at all times.  While deployment strategies for
different  active aerodynamic technologies will undoubtedly vary by individual technology,
the impact of these technologies will mostly be realized at high speeds. EPA expects that the
5-cycle tests will capture the additional real-world benefits not quantifiable with the
FTP/Highway test cycles due to the higher speed in the US06 cycle. Active aero may also
depend on weather conditions. For example, active aerodynamics may operate less in hot
weather when  air cooling is required to exchange heat at the condenser. Also, active grill
shutters may need to stay open during snowy conditions in order to prevent them from
freezing shut (potentially causing component failure).

       Using the EPA's full vehicle simulation tool, the impact of reducing aerodynamic drag
was simulated on both the combined FTP/Highway cycle and the 5-cycle drive tests. In order
to determine the fuel savings per amount of aerodynamic drag reduction, the  fuel  savings on
the FTP/Highway test cycle was subtracted from the fuel savings on the 5-cycle test. This is
consistent with the approach taken for other technologies. Table 2.3-3  shows the results of the
vehicle simulation. Also, Figure 2.3-3 represents this GHG reduction metrics in a graphical
form. These results assume that the active aerodynamics affects the coefficient of drag only,
which is  currently assumed to be constant over a wide range of vehicle operating speed.
However, if the coefficient of aerodynamic drag is assumed to be vehicle speed dependent,
then a different relationship could result.
                                        2-12

-------
                                                    2017 Regulatory Impact Analysis
 Table 2.3-3 Simulated GHG Reduction Benefits of Active Aerodynamic Improvements
Reduction in Aerodynamic Drag
(Cd)
1%
2%
3%
4%
5%
10%
GHG Reduction in Cars
[g/mile]
0.2
0.4
0.6
0.8
0.9
1.9
GHG Reduction in Trucks
[g/mile]
0.3
0.6
1.0
1.3
1.6
3.2



6.0 -
E 5.0
.a
I
1!
tf 20 .




0

Performance Metrics for Active Aero Technology


	 1 ;_,/ 	 '
y= 33.159X /'"
; ^,,'" _.„..»-
,.gf' .--•-'"""""
/' : ^-''"^"
/'' ! .---'"*""""
/"*" ^**^~ v : — : i y » 3 w jt
=-^ „--'-"" :
^^ES j*-*^ '-
f 1 !
96 5% 10% 15% 2E
AeroDyrtarnk Improvement








S Truck




%

Figure 2.3-3 Simulated GHG Reduction Benefits of Active Aerodynamic Improvements

       2.3.2.2 Active Aerodynamics

       One of the active aerodynamic technologies is active grill shutters.  This technology is
a new innovation that is beginning to be installed on vehicles to improve aerodynamics at
higher speeds. Nearly all vehicles allow air to pass through the front grill of the vehicle to
flow over the radiator and into the engine compartment.  This flow of air is important to
prevent overheating of the engine (and for proper functioning of the A/C system), but it
creates a significant drag on the vehicle and is not always necessary. Active grill shutters
close off the area behind the front grill so that air does not pass into the engine compartment
                                        2-13

-------
   Chapter 2	

   when additional cooling is not required by the engine. This reduces the drag of the vehicle,
   reduces CO2 emissions, and increases fuel economy. When additional cooling is needed by
   the engine, the shutters open until the engine is sufficiency cooled.

          Based on manufacturer data, active grill shutters provide a reduction in aerodynamic
   drag (Cd) from 0 to 5% when deployed. EPA expects that most other active aerodynamic
   technologies, such as active suspension lowering will provide a reduction of drag in the same
   range as active grill shutters.  EPA also expects that active aerodynamic technologies may not
   always be available during all operating conditions. Active grill shutters, for example, may
   not be usable in very cold temperatures due to concerns that they could freeze in place and
   cause overheating.  Control and calibration issues, temperature limitations, air conditioning
   usage, and other factors may limit the usage of grill shutters and other active aerodynamic
   technologies. Therefore, EPA is proposing to provide a credit for active aerodynamic
   technologies assuming that any of these technologies will achieve an aerodynamic drag of at
   least 3% improvement.  The proposed default value for the credit will be 0.6 g/mile for cars
   and 1.0  g/mile for trucks, in accordance with the simulation results in
          Table 2.3-3.  It is conceivable that some systems can achieve better performance.
   Manufacturers may  apply for greater credit for better performing systems through the normal
   application process described in Section IILC.S.b of the preamble to the proposed rule..

   2.4 On-Going and Future Work

          2.4.1   Simulation Tool Validation

          Since the EPA's full vehicle simulation tool is still in the development phase, it has
   not been fully validated against actual vehicle test data yet.  However, EPA has attempted to
   compare the EPA's simulation results to those of Ricardo's. Unfortunately, none of the
   Ricardo's vehicle simulation metrics exactly matched with the simulation runs performed by
   the EPA's simulation tool.  For this reason, EPA used the lumped parameter model which had
   been calibrated and tuned with Ricardo's simulation results for a comparison.

Table 2.4-1 Comparison between EPA's Full Vehicle Simulation Tool and Lumped Parameter
                                       Model Runs
Simulation Tool
Vehicle Simulation
Lumped Parameter Model
Percent Difference
Small-Size Car
[g/mile]
211.7
220
3.8%
Mid-Size Car
[g/mile]
273.8
280
2.2%
Large-Size Car
[g/mile]
350.2
359
2.5%
Pick-up Truck
[g/mile]
532.7
520
2.4%
          Using the same simulation metrics (e.g. baseline engine, DCT transmission, vehicle
   types) for both the EPA's full vehicle simulation tool and the lumped parameter model, the
   results were obtained as shown in Table 2.4-1.  As shown in Table 2.4-1, it is evident that the
   EPA vehicle simulation tool provides GHG estimations which are very comparable with
                                           2-14

-------
                                                     2017 Regulatory Impact Analysis
lumped parameter model results, and therefore with Ricardo's simulation results for various
vehicle types.  The differences are all within ±5% between the two simulations. Although this
benchmarking result against the Ricardo's simulation does provide a certain level of
confidence in the EPA's simulation tool, a full validation of the tool will be performed using
actual vehicle test data before the final rule.

       2.4.2   Simulation Tool Upgrade

       As mentioned previously, the EPA's full light-duty vehicle simulation tool is still in
the development phase. There are a number of improvements and new additions being
planned for the simulation tool so that it will be capable of performing various different types
of simulations for a number of vehicle technologies. EPA expects that the upgraded vehicle
simulation tool can provide assistance in further analysis for the final rule.

       First, an automatic transmission model will be added for the conventional (non-
hybrid) vehicle simulation tool. Although EPA expects that DCT will be a dominant
technology in transmissions in 2017 to 2025, EPA must be able to simulate vehicles with
automatic transmissions which give baseline vehicle performances.  Also, 8-speed automatic
transmissions with lock up will also require this model as a basis. Along with the automatic
transmission, a transmission shifting algorithm will be developed, which will help us avoid
requiring transmission shifting maps.  This algorithm will automatically optimize the shifting
strategy based on torque required by the vehicle and torque produced by the engine during
simulation. Therefore, it will  eliminate the need for having shifting maps for different
combinations of powertrains and vehicles.

       In addition to upgrading the non-hybrid vehicle simulation tool, EPA is planning to
add hybrid  electric vehicle (HEV) simulation capabilities.  The HEV simulation tool is being
currently developed within the EPA for power-split and P2 configurations.  For both non-
hybrid and  hybrid simulation tools, EPA is also planning to design a Graphical User Interface
(GUI) and integrate it with the vehicle simulation tool. This GUI will allow the user to
choose from different technologies and simulation options while making the use of the tool
much easier and straightforward.  These tools are  expected to assist in further analysis for the
final rule as necessary.
                                         2-15

-------
Chapter 2	

                                        References

11 Forrest, W.O., "Air Conditioning and Gas Guzzler Tax Credits," Society of Automotive
Engineers, International Congress & Exposition, Detroit, Michigan, March 2002, SAE Paper
2002-01-1958.
12 Forrest, W.O. and Bhatti, M.S., "Energy Efficient Automotive Air Conditioning System,"
Society of Automotive Engineers, International Congress & Exposition, Detroit, Michigan,
March 2002, SAE Paper 2002-01-0229.
13 Johnson, V.H., "Fuel Used for Vehicle Air Condition: A State-by-State Thermal Comfort-
Based Approach," Society of Automotive Engineers, International Congress & Exposition,
Detroit, Michigan, March 2002, SAE Paper 2002-01-1957.
14 Rugh, J.P., Hovland, V., Andersen, S.O., "Significant Fuel Savings and Emissions
Reductions by Improving Vehicle Air Conditioning," Presentation at the 15th Annual Earth
Technologies Forum and Mobile Air Conditioning Summit, April  15, 2004.
15 Northeast States Center for a Clean Air Future, "Reducing Greenhouse Gas Emissions from
Light-Duty Motor Vehicles," September, 2004.
16 EPA and DOT, "Light-Duty Vehicle Greenhouse Gas Emission Standards and Corporate
Average Fuel Economy Standards: Final Rule," May 7, 2010.
                                       2-16

-------
                                                2017 Draft Regulatory Impact Analysis
3      Results of Proposed and Alternative Standards

3.1 Introduction

       This chapter provides the methodology from and results of the technical assessment of
the future vehicle scenarios presented in this proposal.  As in the analysis of the MY 2012-
2016 rulemaking, evaluating these scenarios included identifying potentially available
technologies and assessing their effectiveness, cost, and impact on relevant aspects of vehicle
performance and utility. The wide number of technologies which are available and likely to be
used in combination required a method to account for their combined cost and effectiveness,
as well as estimates of their availability to be applied to vehicles.

       Applying these technologies efficiently to the wide range of vehicles produced by
various manufacturers is a challenging task. In order to assist in this task, EPA is again using
a computerized program called the Optimization Model for reducing Emissions of
Greenhouse gases from Automobiles (OMEGA). Broadly, OMEGA starts with a description
of the future vehicle fleet, including manufacturer, sales, base CO2 emissions, footprint and
the extent to which emission control technologies are already employed. For the purpose of
this analysis, EPA uses OMEGA to analyze over 200 vehicle platforms which encompass
approximately 1300 vehicle models in order to capture the important differences in vehicle
and engine design and utility of future vehicle sales of roughly 16-18 million units annually in
the 2017-2025 timeframe.  The model is then provided with a list of technologies which are
applicable to various types of vehicles, along with the technologies' cost and effectiveness
and the percentage of vehicle sales which can receive each technology during the redesign
cycle of interest. The model combines this information with economic parameters, such as
fuel prices and a discount rate, to project how various manufacturers would  apply the
available technology in order to meet increasing levels  of emission control.  The result is a
description of which technologies are added to each vehicle platform, along with the resulting
cost.  The model can also be set to account for various types of compliance flexibilities.8

       EPA has described OMEGA's specific methodologies and algorithms previously in
the model documentation,17 the model is publically available on the EPA website,18 and it has
recently been peer reviewed.19

3.2 OMEGA model overview

       The OMEGA model evaluates the relative cost  and effectiveness of available
technologies and applies them to a defined vehicle fleet in order to meet a specified
GHG emission target. Once the target has been met,  OMEGA reports out the cost and
societal benefits of doing so. OMEGA is capable of modeling two GHGs; carbon
dioxide (CO2) from fuel use and HFC refrigerant emissions from the air conditioning
s While OMEGA can apply technologies which reduce CO2 efficiency related emissions and refrigerant leakage
emissions associated with air conditioner use, this task is currently handled outside of the OMEGA model. A/C
improvements are relatively cost-effective, and would always be added to vehicles by the model, thus they are
simply added into the results at the projected penetration levels.
                                          5-1

-------
Chapter 3

(A/C) system. The model is written in the C# programming language, however both
inputs to and outputs from the model are provided using spreadsheet and text files.
The spreadsheet output files also facilitate additional manipulation of the results, as
discussed in the next section.

       OMEGA is primarily an accounting model.  It is not a vehicle simulation
model, where basic information about a vehicle, such as its mass, aerodynamic drag,
an engine map, etc. are used to predict fuel consumption or CO2 emissions over a
defined driving cycle.1 While OMEGA incorporates functions which generally
minimize the cost of meeting a specified CO2 target, it is not an economic simulation
model which adjusts vehicle sales in response to the cost of the technology added to
each vehicle.u

       OMEGA can be used to model either a single vehicle model or any number
vehicle models.  Vehicles can be those of specific manufacturers as in this analysis or
generic fleet-average vehicles as in the 2010 Technical Assessment Report supporting
the MY 2017-2025 NOT. Because OMEGA is an accounting model, the vehicles can
be described using only a relatively few number of terms.  The most important of
these terms are the vehicle's baseline emission level, the level of CO2 reducing
technology already present, and the vehicle's "type," which indicates the technology
available for addition to that vehicle. Information required determining the applicable
CO2 emission target for the vehicle must also be provided.  This may simply be
vehicle class (car or truck) or it may also include other vehicle attributes, such as
footprint.v In the case of this rulemaking, footprint and vehicle class are the relevant
attributes.

       Emission control  technology can be applied individually or in groups, often
called technology "packages."  The user specifies the cost and effectiveness of each
technology or package for a specific "vehicle type," such as midsize cars with V6
engines or minivans. The user can limit the application of a specific technology to a
specified percentage of each vehicle's sales (i.e., a "cap").  The effectiveness, cost,
application limits of each technology package can also vary over time.w A list of
technologies or packages is provided for each vehicle type, providing the connection
to the specific vehicles being modeled and a description of these packages can be
found in Chapter 1  of this draft RIA (DRIA)
T Vehicle simulation models may be used in creating the inputs to OMEGA as discussed in Draft Joint TSD
Chapter 3 as well as Chapter 1 of the Draft RIA.
u While OMEGA does not model changes in vehicle sales, Draft RIA Chapter 8 discusses this topic.
v A vehicle's footprint is the product of its track width and wheelbase, usually specified in terms of square feet.
w "Learning" is the process whereby the cost of manufacturing a certain item tends to decrease with increased
production volumes or over time due to experience. While OMEGA does not explicitly incorporate "learning"
into the technology cost estimation procedure, the user can currently simulate learning by inputting lower
technology costs in each subsequent redesign cycle based on anticipated production volumes or on the elapsed
time.
                                           5-2

-------
                                                2017 Draft Regulatory Impact Analysis
       OMEGA is designed to apply technology in a manner similar to the way that a
vehicle manufacturer might make such decisions.  In general, the model considers
three factors which EPA believes are important to the manufacturer: 1) the cost of the
technology, 2) the value which the consumer is likely to place on improved fuel
economy and 3) the degree to which the technology moves the manufacturer towards
its fleetwide CO2 emission target.

       Technology can be added to individual vehicles using one of three distinct
ranking approaches. Within a vehicle type, the order of technology packages is set by
the user.  The model then applies technology to the vehicle with the lowest
Technology Application Ranking Factor (hereafter referred to as the TARF).
OMEGA offers several different options for calculating TARF values.  One TARF
equation considers only the cost of the technology and the value of any reduced fuel
consumption considered by the vehicle purchaser.  The other two TARF equations
consider these two factors in addition to the mass of GHG emissions reduced over the
life of the vehicle. Fuel prices by calendar year, vehicle survival rates and annual
vehicle miles travelled with age are provided by the user to facilitate these
calculations.

       For each manufacturer,  OMEGA applies technology (subject to constraints, as
discussed in Draft Joint TSD 3) to vehicles until the sales-weighted emission average
complies with the specified standard or until all the available technologies have been
applied. The standard can be a flat standard applicable to all vehicles within a vehicle
class (e.g., cars, trucks or both cars and trucks). Alternatively the GHG standard can
also be in the form of a linear or constrained logistic function, which sets each
vehicle's target as a function of vehicle footprint (vehicle track width times
wheelbase).  When the linear form of footprint-based standard is used, the "line" can
be converted to a flat standard for footprints either above or below specified levels.
This is referred to as a piece-wise linear standard,  and was used in modeling the
standards in this analysis.

       The emission target can vary over time, but not on an individual model year
basis.  One of the fundamental features of the OMEGA model is that it applies
technology to a manufacturer's fleet over a specified vehicle redesign cycle.  OMEGA
assumes that a manufacturer has the capability to redesign any or all of its vehicles
within this redesign cycle. OMEGA does not attempt to determine exactly which
vehicles will be redesigned by each manufacturer in any given model year. Instead, it
focuses on a GHG emission goal several model years in the future, reflecting the
manufacturers' capability to plan several model years in advance when determining
the technical designs of their vehicles.  Any need to further restrict the application of
technology can be effected through the caps  on the application of technology to each
vehicle type mentioned above.

       Once technology has been added so that every manufacturer meets the
specified targets (or exhausts all of the available technologies), the model  produces a
variety of output files. These files include specific information about the technology
added to each vehicle and the resulting costs and emissions.  Average costs and
                                          5-3

-------
Chapter 3

emissions per vehicle by manufacturer and industry-wide are also determined for each
vehicle class.

3.3 OMEGA Model Structure

       OMEGA includes several components, including a number of pre-processors that
assist users in preparing a baseline vehicle forecast,x creating and ranking technology
packages/ and calculating the degree to which technology is present on baseline vehicles.
The OMEGA core model collates this information and produces estimates of increases in
vehicle cost and CO2 reduction.  Based on the OMEGA core model output, the technology
penetration of the new vehicle mix and the scenario impacts (fuel savings, emission impacts,
and other monetized benefits) are calculated by post-processors. The pre- and post-
processors are Microsoft Excel spreadsheets and visual basic programs, while the OMEGA
core model is an executable program written in the C# language.

       OMEGA is designed to be flexible in a number of ways. Very few numerical values
are hard-coded in the model, and consequently, the model relies heavily on its input files. The
model utilizes five input files: Market, Technology, Fuels, Scenario, and Reference. Figure
3.3-1  shows the (simplified) information flow through OMEGA, and how these files interact.

                 Figure 3.3-1 Information Flow in the OMEGA Model

  Pre-Processors
                                  Existing Technology
                                    Calculation
                                                                       Reference File
                                                                       Scenario File
                                                                        Fuels File
                                                                Core Model
                                                                           v
Post-Processors
                                                                         OMEGA
                                                                   Technology Penetration
                                                                         Impacts
       OMEGA utilizes four basic sets of input data. The first, the market file, is a
description of the vehicle fleet.  The key pieces of data required for each vehicle are its
manufacturer, CO2 emission level, fuel type, projected sales and footprint.  The model also
requires that each vehicle be assigned to one of the 19 vehicle types, which tells the model
 ' Joint Draft TSD Chapter 1
 DRIA Chapter 1
                                          5-4

-------
                                               2017 Draft Regulatory Impact Analysis
which set of technologies can be applied to that vehicle.  Chapter 1 of the Joint TSD contains
a description of how the vehicle reference fleets were created for modeling purposes, and
includes a discussion on how EPA defined the 19 vehicle types.  In addition, the degree to
which each vehicle already reflects the effectiveness and cost of each available technology in
the 2008  baseline fleet must be input. This prevents the model from adding technologies to
vehicles already having these technologies in the baseline.  It also avoids the situation, for
example, where the model might try to add a basic engine improvement to a current hybrid
vehicle.  Section 3.4.1.2 of this Draft Regulatory Impact Analysis (RIA) contains a detailed
discussion of how EPA accounts for technology present in the baseline fleet in OMEGA.

       The second type of input data, the technology file is a description of the technologies
available to manufacturers, primarily their cost, effectiveness, and electricity consumption.
This information was described in Chapter 1 of this Draft RIA and Chapter 3 of the Draft
Joint TSD. In all cases, the order of the technologies or technology packages for a particular
vehicle type is designated by the model user in the input files prior to running the model. The
ranking of the packages is described in Chapter 1 of the DRIA.

       The third type of input data describes vehicle operational data, such as annual scrap
rates and mileage accumulation rates, and economic data, such as fuel prices and discount
rates. These estimates are described in chapter 4 of the Draft Joint TSD.

       The fourth type of data describes the CO2 emission standards being modeled. These
include the MY 2016 standards and the proposed standards. As described in more detail in
Chapter 5 of the Draft Joint TSD and briefly in section 3.8.5 below, the application of A/C
technology is evaluated in a separate analysis from those technologies which impact CO2
emissions over the 2-cycle test procedure. For modeling purposes, EPA applies this AC
credit by  adjusting manufacturers' car and truck CO2 targets by an amount associated with
EPA's projected use of improved A/C systems, as discussed in Section 3.8.5, below.

       The input files used in this analysis, as well as the current version of the OMEGA core
model, are available in the docket (EPA-HQ-OAR-2010-0799).  The following sections
describe creation of each of the input files from the data and parameters discussed in the Draft
Joint.TSD and in this RIA.

3.4 Model Inputs

       3.4.1  Market Data

       3.4.1.1 Vehicle platforms

       As discussed in Draft Joint TSD Chapter 3 and in Chapter 1 of the DRIA, vehicle
manufacturers typically develop many  different models by basing them on a smaller number
of vehicle platforms.  The platform typically consists of a common set of vehicle architecture
and structural components. This allows for efficient use of design and manufacturing
resources. In this analysis, EPA created over 200 vehicle platforms which were used to
capture the important differences in vehicle and engine design and utility of future vehicle
sales. The approximately sixty vehicle platforms are a result of mapping the vehicle fleet into
the 19 engine based vehicle types (Table 3.4.1) and the 10 body size and structure based
                                          5-5

-------
Chapter 3
utility classes (Table 3.4-2) by manufacturer. As not all vehicle types match to all utility
types, and not all manufacturers make all vehicle and utility types, the number of vehicles is
less than the multiplicative maximum of the two tables.

                   Table 3.4-1 Vehicle Types in the MY 2017-2025 Analysis
Vehicle Type #
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
Name
Subcompact Car
Compact Car 14
Midsize Car/Small MPV (unibody)
Compact Car/Small MPV (unibody)
Midsize/Large Car
Midsize Car/Large Car
Mid-sized MPV (unibody )/Small Truck
Midsize MPV (unibody)/Small Truck
Large MPV (unibody)
Large MPV (unibody)
Large Truck (+ Van)
Large Truck + Large MPV
Large Truck (+ Van)
Large Truck (+Van)
Large Car
Large MPV (unibody)
Large MPV (unibody)
Large Truck (+ Van)
Large Truck (+ Van)
Cam
DOHC
DOHC
DOHC
DOHC
DOHC
DOHC
DOHC
SOHC
SOHC
SOHC
SOHC
OHV
OHV
SOHC3V
OHV
DOHC
DOHC
DOHC
DOHC
Engine
14
14
14
V6
V6
V8
14
V6
V8
V8
V6
V6
V8
V8
V8
V6
V8
V6
V8
               Table 3.4-2 Vehicle Types in the Technical Assessment Analysis
Utility
Class #
1
2
3
4
5
6
7
8
9
10
Utility Class
Subcompact Auto
Compact Auto
Mid Size Auto
Large Auto
Small SUV
Large SUV
Small Pickup
Large Pickup
Cargo Van
Mini van
Vehicle Use *
Car
Car
Car
Car
SUV
SUV
Pickup
Pickup
Van
Van
Footprint Criteria
Footprint <43
43<=Footprint<46
46<=Footprint<53
56<=Footprint
43<=Footprint<46
46<=Footprint
Footprint < 50
50<=Footprint
—
—
Structure Criteria
—
—
—
—
—
—
—
—
Ladder Frame
Unibody
       1. Vehicle use type is based upon analysis of EPA certification data.

-------
                                               2017 Draft Regulatory Impact Analysis
       3.4.1.2 Accounting for technology already on vehicles

       As mentioned above for the market data input file utilized by OMEGA, which
characterizes the vehicle fleet, our modeling accounts for the fact that many 2008 MY
vehicles are already equipped with one or more of the technologies discussed in Draft Joint
TSD 3. Because of the choice to apply technologies in packages, and because 2008 vehicles
are equipped with individual technologies in a wide variety of combinations, accounting for
the presence of specific technologies in terms of their proportion of package cost and CC>2
effectiveness requires careful, detailed analysis.

       Thus, EPA developed a method to account for the presence of the combinations of
applied technologies in terms of their proportion of the technology packages. This analysis
can be broken down into four steps

       The first step in the process is to break down the available GHG control technologies
into five groups:  1) engine-related, 2) transmission-related, 3) hybridization, 4) weight
reduction and 5)  other. Within each group we gave each individual technology a ranking
which generally followed the degree of complexity, cost and effectiveness of the technologies
within each group. More specifically, the ranking is based on the premise that a technology
on a 2008 baseline vehicle with a lower ranking would be replaced by one with a higher
ranking which was contained in  one of the technology packages which we included in our
OMEGA modeling. The corollary of this premise is that a technology on a 2008 baseline
vehicle with a higher ranking would be not be replaced by one with an equal or lower ranking
which was contained in one of the technology packages which we chose to include in our
OMEGA modeling. This ranking scheme can be seen in an  OMEGA pre-processor (the
TEB/CEB calculation macro), available in the docket (EPA-HQ-OAR-2010-0799).

       In the second step of the  process, we used these rankings to estimate the complete list
of technologies which would be  present on each vehicle after the application of a technology
package.  In other words, this step indicates the specific technology on each vehicle after a
package has been applied to it. We then used the EPA lumped parameter model to estimate
the total percentage CO2 emission reduction associated with the technology present on the
baseline vehicle (termed package 0), as well as the total percentage reduction after application
of each package. We used a similar approach to determine the total cost of all of the
technology present on the baseline vehicle and after the application of each applicable
technology package.

       The third step in this process is to account for the degree of each technology
package's incremental effectiveness and incremental cost is  affected by the technology
already present on the baseline vehicle.  In this step  , we calculate the degree to which a
technology package's effectiveness is already present on the baseline vehicle, and produces a
value for each package termed the technology effectiveness  basis, or TEB.  The degree to
which a technology package's incremental cost is reduced by technology already present on
the baseline vehicle is termed the cost effectiveness basis, or CEB, in the OMEGA model.

       The value of each vehicle's TEB for each applicable technology package is
determined as follows:
                                         5-7

-------
Chapter 3
f TotalEffect , ] f \-TotalEffect
1 ' \x p'
\ l^l
( 1 - TotalEffect 1 11 — TotalEffect ,
jng^ _ v . / v P> y
r I -TotalEffect^ }
[ I -TotalEffect
P,i-l )

       Where
       TotalEffectv! =   Total effectiveness of all of the technologies present on the baseline vehicle after
                      application of technology package i
       TotalEffectv u =  Total effectiveness of all of the technologies present on the baseline vehicle after
                      application of technology package i-1
       TotalEffectp! =  Total effectiveness of all of the technologies included in technology package i
       TotalEffectp)1_i  = Total effectiveness of all of the technologies included in technology package i-1
                            Equation 3.4-1 - TEB calculation

       The degree to which a technology package's incremental cost is reduced by
technology already present on the baseline vehicle is termed the cost effectiveness basis, or
CEB, in the OMEGA model.  The value of each vehicle's CEB for each applicable
technology package is determined as follows:

       CEB; = 1 - (TotalCostV;i - TotalCosVi) / (TotalCostp,; - TotalCostp;i.i)

       Where
       TotalCostv =     total cost of all of the technology present on the vehicle after addition
                      of package i or i-1 to baseline vehicle v
       TotalCostp =     total cost of all of the technology included in package i or i-1
       i = the technology package being evaluated
       i-1 = the previous technology package
                            Equation 3.4-2 - CEB calculation

       As described above, technology packages are applied to groups of vehicles which
generally represent a single vehicle platform and which are equipped with a single engine size
(e.g., compact cars with four cylinder engine produced by Ford).  Thus, the fourth step is to
combine the fractions of the CEB and TEB of each technology package already present on the
individual  MY 2008 vehicle models for each vehicle grouping. For cost, percentages of each
package already present are combined using a simple sales-weighting procedure, since the
cost of each package is the same for each vehicle in a grouping.  For effectiveness, the
individual  percentages are combined by weighting them by both sales and base CC>2 emission
level.  This appropriately weights vehicle models with either higher sales or CC>2 emissions
within a grouping.  Once  again, this process prevents the model from adding technology
which is already present on vehicles, and thus ensures that the model does not double count
technology effectiveness  and cost associated with complying with the modeled standards.2
z The OMEGA TEB/CEB calculator used in the analysis of the proposal did not properly calculate CEBs for
vehicles where a more efficient and less expensive engine was placed in a vehicle. We estimate that this issue

-------
                                                 2017 Draft Regulatory Impact Analysis
       3.4.1.3 Accounting for Net Mass Reduction and Safety related Mass reduction

      For this analysis, EPA modified its application of mass reduction to be similar to that
used by NHTSA in the CAFE model analysis. In this methodology, and in contrast to the
approach taken in the MY 2012-2016 rule, more mass is taken out of heavier vehicles, and
less mass is taken out of lighter vehicles.  This approach allows the agency to provide costs
for a technology assessment that includes no net additional fatalities to the fleet.
Manufacturers may not necessarily apply mass reduction in this manner, but EPA
demonstrates that a technically feasible and economically practicable path exists for
manufacturers to meet their fleet standards without compromising safety. The limits on mass
reduction, as applied in the OMEGA model, are dependent upon both the technology inputs
discussed in TSD Chapter 3, as well as on the fatality coefficients from the 2011 Kahane
report and the related adjustments for improvements in federal motor vehicle safety standards
(FMVSS) as discussed in Section II.G of the Preamble to the proposed rule, and are  subject to
the same caveats.^  Changes to these coefficients would change the projected amount of
mass  reduction projected for the fleet.

      Using a spreadsheet scoping tool, EPA projected the maximum amount of mass
reduction on a vehicle by vehicle basis that would result in a net fatality neutral result. Based
on the coefficients used in the analysis, reducing weight from trucks above 4,594 pounds and
from minivans, reduces fatalities. By contrast, this analysis implies that removing weight
from the other vehicle categories increases fatalities.  The inputs used in the OMEGA analysis
are shown below (Table 3.4-3 Fatality coefficients used in OMEGA analysis
                   Table 3.4-3 Fatality coefficients used in OMEGA analysis
Vehicle Category
by class and
weight
PC below 31 06
PC above 3106
LT below 4594
LT above 4594
Minivan
Kahane
Coefficients :
1 .44%
0.47%
0.52%
-0.39%
-0.46%
Base
fatalities
per billion
miles
12.38
10.33
14.77
14.43
8.30
adjustment for
new FMVSS
0.884
0.884
0.884
0.884
0.884
Change in Fatalities
per pound per mile2
1.58E-12
4.29E-13
6.79E-13
-4.97E-13
-3.38E-13
 Expressed as percent change in base fatalities per 100 pound change in vehicle weight
Calculated as coefficients x base fatalities x adjustment x one billion miles /100
causes an overestimate of compliance costs by approximately $25 across the fleet in MY 2025, and will update
the model appropriately in the final rulemaking.
^ Please note that the OMEGA safety assessment was performed with a draft version of the FMVSS adjustment
, that raises the impact of the coefficients by approximately 1% relative to the analysis conducted by NHTSA,
which uses an FMVSS adjustment of 0.874
                                           5-9

-------
Chapter 3

       The mass reduction scoping tool contains the entire fleet discussed in TSD 1, along
with their curb weight, and their passenger car, light truck, minivan classification according to
the criteria in the 2011 Kahane report. Using this tool, EPA determined that a simulation of
fatality neutrality could result by assuming that no passenger car was light-weighted below
3,000 pounds, and no light trucks were reduced below 4,594 pounds. These values were
determined iteratively, with the end product a fatality neutral analysis.  Vehicles above these
weight could have their weight reduced through mass reduction technology in the OMEGA
model.  The per vehicle limit on weight reduction for these vehicles was therefore determined
by these specific weight cut points, or by the maximum phase-in caps for mass reduction, of
15% in 2021, 20% in 2025.. Vehicles below these weights had no net mass reduction applied.

       The term "net mass reduction" is used because EPA explicitly accounted for the mass
impacts (generally increases) from converting a vehicle into a hybrid-electric, plug-in hybrid
electric, or battery electric vehicle. This was not done in the MYs 2012-2016 analysis or in
the technical assessment report.  A table of these weight impacts is presented in Draft Joint
TSD Chapter 3. EPA did not include a weight penalty for dieselization, but will consider
including such impacts in the final rulemaking.  The per-vehicle limit on weight reduction
determined above is for net mass reduction, not for the application of total mass reduction
technology.

       Because the limits  on net mass reduction are at the individual vehicle level, they are
reflected through modifications to the individual TEB and CEB values  rather than the "caps"
in the technology file  (which are discussed in the next section). EPA assumed that there was
no mass reduction technology being utilized in the 2008 fleet.

       To implement this  schema, each vehicle in the 2008 baseline was assigned the
following parameters:

          •  Amount of mass reduction already present in baseline vehicle (assumed to be zero in
             this analysis)
          •   Maximum amount of mass reduction allowed
          •   Mass penalty for adding various technologies to that vehicle

       Some examples:

          •  A baseline vehicle is defined with a 10 percent maximum mass reduction. A vehicle
              package is applied containing a 15 percent mass reduction. The package mass
              reduction will be overridden  resulting in a 10 percent cost and effectiveness applied
             to the vehicle.
          •  A baseline vehicle has a 5 percent penalty for P2HEV conversion. A vehicle package
              is applied containing a 10 percent mass reduction and a conversion to P2 hybrid.
              Due to the 5 percent penalty for conversion, the baseline vehicle will incur a cost of
              15 percent mass reduction to result in an overall 10 percent reduction. The resulting
             effectiveness due to the mass reduction will be 10 percent.
                                         3-10

-------
                                                2017 Draft Regulatory Impact Analysis
       Under this system, any amount of mass reduction already in the baseline vehicle will
be subtracted from the maximum amount of mass reduction allowed.  All vehicles in the
baseline fleet are assumed to have no mass reduction technology applied.
       3.4.2      Technology Data

       Consistent with OMEGA's redesign cycle approach, the technology input file defines
the technology packages which the model can add to the vehicle fleet. In brief, each of the 19
vehicle types have an associated list of technology packages, costs and effectivenesses.BB
Each of the 19 lists was then ordered by how OMEGA should add them to that specific
vehicle type.  The order of this list is influenced by the relative cost and effectiveness of
technologies as well as their market penetration cap (or maximum penetration rate). Market
penetration caps of less than 100% restrict the model to that fraction of a vehicle platform.cc
The processes to build and rank technology packages for the technology file are described in
detail in Chapter 1 of the DRIA.

       For this analysis, a separate technology file was developed for each model year (2021
and 2025) for which OMEGA was run. The MY 2021 and MY 2025 costs differ due to the
learning effects discussed in the Draft Joint TSD Chapter 3, and also differ due to the different
limits on maximum penetrations of technologies.

       OMEGA adds technology effectiveness according to the following equation in which
the subscripts t and t-1 represent the times before and after technology addition, respectively.
The numerator is the effectiveness of the current technology package and the denominator
serves to "back out" any effectiveness that is present in the baseline. AIE is the "average
incremental effectiveness" of the technology package on a vehicle type, and TEB is the
"technology effectiveness  basis", which denotes the fraction of the technology present in the
baseline.

                       Equation 3.4-3 - Calculation of New

                                          CO2t_lx(l-AIE)
                                   CO2=-
                                             1-AIExTEB
       OMEGA then adds technology cost according to the equations below, where CEB
refers to the "cost effectiveness basis", or in other words, the technology cost that is present in
BB Given that effectiveness is expressed in percentage terms, the absolute effectiveness differs even among
vehicles of the same vehicle type, but the relative effectiveness is the same.
cc Penetration caps may reflect technical judgments about technology feasibility and availability, consumer
acceptance, lead time, and other reasons as detailed in Chapter 3 of the Draft Joint TSD.


                                          3-11

-------
Chapter 3

the baseline. Cost can be calculated for the application of the a package, or eventually, for the
average cost of a manufacturers fleet (Equation 3.4-4,Equation 3.4-5).
          Equation 3.4-4 - Calculation of New Cost after applying a package
                            Cost =Cos^_1 + TechCosf(l-CEB)
           Equation 3.4-5 - Calculation of Average Cost for a manufacturer

                                           TechCost * ModelSales
                      AvgVehicleCostMPR =
                                              TotalFleetSales
                                                                MFR
       EPA's OMEGA model calculates the new CO2 and average vehicle cost after each
technology package has been added.

       In light of the complex set of technology caps used in this analysis, EPA modified the
methodology used to generate the OMEGA technology input file relative to previous
analyses. As background, for both the MY 2012-2016 rulemaking analysis and the Technical
Assessment Report supporting the 2017-2025 NOT, the technology caps generally fell into a
few broad numeric categories. As an example, in the analysis supporting the 2012-2016 final
rulemaking, most technologies were capped at one of three levels (15%, 85%, 100%). The
small number of technology caps made it relatively simple to build packages around
technologies which had a shared cap.  By contrast, and as discussed in chapter 3 of the joint
draft TSD, there are both more technologies and more technology cap levels considered in
this proposal. Thus, it was more difficult to construct packages with uniform sets of caps. As
a means of doing so in this proposal, these caps were incorporated into the OMEGA modeling
in one of two ways.  Major engine technologies such as turbo-charging  and downsizing,
hybridization, electrification and dieselization were directly controlled through caps in the
technology file. Maximum penetration rates of other technology were managed through
multiple runs of the TEB-CEB computation algorithm and modifications to the cost,
effectiveness, and  electric conversion values in the technology file.

       For reference case runs, EPA used three sets of TEB/CEB files in order to model the
input caps.

       - Set A is a normally Ranked Master-set (30%)
       - Set B removes 8sp trans, IACC2, Aero2 from Set A (55%)
                                       3-12

-------
                                               2017 Draft Regulatory Impact Analysis
       - Set C removes CCP, DCP, Deac, DVVL, CCC, GDI, IDS 18, TDS24, 6sp trans,
       EPS, Stop-start, SAX, Aerol, DSL-Adv from Set B (15%)

       For proposal and alternative runs in MY 2021, EPA used four sets of TEB/CEB files
in order to model the input caps.

       - Set A is a normally Ranked Master-set (60%)
       - Set B removes HEG & EFR2 from Set A (15%)
       - Set C removes LRRT2 from Set B (5%)
       - Set D removes 8sp trans, Aero2, IACC2 from Set C (20%)

       In creating the OMEGA input market file and technology file, these sets were then
weighted together according to the fractions listed next to each set above. As an example,
eight speed transmissions are capped at 80% in 2021 (see  TSD 3).  When weighted together,
set D, which removes eight speed transmissions only gets 20% of the weighting in the cost,
effectiveness and electric conversion fraction. Using this  method, in the OMEGA input file,
the cost, effectiveness and electricity consumption of each package was calculated to reflect
the weighted cost and effectiveness of each package after  accounting for the weighting of the
sets.00 The technology penetrations are also calculated using the weighting of each set.
Using the combination of the set weighting, and the technology cap feature in the technology
input file, EPA reflects the analytic constraints. For the final rulemaking, EPA intends to
simplify this process. When a technology package is applied to fewer than 100% of the sales
of a vehicle model due to the market penetration cap, OMEGA tracks the sales volume of
vehicles with each technology package applied.

       OMEGA also tracks electrical consumption in kWh per mile. Each technology
package is associated with an "electricity conversion percentage" which refers to the increase
in the energy consumed by the electric drivetrain relative to  reduction in the consumption of
energy from liquid fuel.  Electricity is a highly refined form  of energy which can be used
quite efficiently to create kinetic energy.  Thus, electric  motors are much more efficient than
liquid fuel engines.  Consequently, the electric consumption percentage input in the
Technology File for plug-in vehicles is generally well below than 100%. It may be possible
that this percentage could exceed 100% under certain circumstances, for example when one
type of plug-in vehicle is being converted into another plug-in vehicle and electricity
consumption per mile is increasing due to larger and heavier batteries, etc. However, that was
not the case for any of the technologies evaluated in this analysis.

       The electric consumption for each vehicle as entered into the OMEGA technology file
(in this analysis) in the on-road energy consumption, calculated as

           Equation 3.4-6 - Electricity Consumption considered in OMEGA
DD Please note that incremental effectiveness values were not simply weighted together, as the resulting rates
would not be correct. Therefore, EPA calculated the accurate CO2 and backcalculated the appropriate
incremental effectiveness values.
                                         3-13

-------
Chapter 3

Electricity Consumption =

       2 cycle energy consumption from the battery / (1-on road gap)/ (1-charging losses)

       Where:
       2 cycle energy consumption      = Based on vehicle type as documented in TSD 3
       On road gap for electricity        = 30%
       Charging losses                = 10%

       The actual input to the model is the "electric conversion percentage," which is
computed as a single fraction for each vehicle type.   Thus, in OMEGA's calculations, the
resulting electricity consumption differs based on the starting CO2 of the vehicle.

                   Equation 3.4-7 - Electrical Conversion Percentage

Electric Conversion Percentage =

                        Electricity consumption
~~~12 gram C       i gallon fuel           3409 btuper kwh     .
^°             44 Grams CO2 Carbon content of fuel   Energy content of gasoline (btu)'

       Where:
       Electricity consumption = values from TSD 3 or RIA 1
       Carbon content of fuel = 2433 for gasoline
       Energy content of fuel = 115,000btu/gallon
3.5 The Scenario File

       3.5.1   Reference Scenario

       In order to determine the technology costs associated with this NPRM, EPA
performed three separate modeling exercises.  The first was to determine the costs associated
with meeting the MY 2016 CC>2 regulations.  EPA considers the MY 2016 CC>2 regulations
to constitute the "reference case" for calculating the costs and benefits of this GHG rule.  In
other words, absent any further rulemaking, this is the vehicle fleet EPA would expect to see
through 2016 — the "status quo".  In order to calculate the costs and benefits of this NPRM
alone, EPA seeks to subtract out any costs associated with meeting any existing standards
related to GHG emissions.

       EPA assumes that in the absence of the proposed GHG and CAFE standards, the
reference case fleet in MY 2017-2025 would have fleetwide GHG emissions performance no
better than that projected to be necessary to meet the MY 2016 standards. While it is not
possible to know with certainty the future fleetwide GHG emissions performance in the
absence of more stringent standards, EPA believes that this approach is the most reasonable
assumption for developing the reference case  fleet for MY 2017-2025.  A discussion of this
topic is presented in section HID  of the preamble, and is presented below with additional
figures and tables.
                                          3-14

-------
                                                2017 Draft Regulatory Impact Analysis
       While it is not possible to know with certainty the future fleetwide GHG emissions
performance in the absence of more stringent standards, EPA believes that this approach is the
most reasonable assumption for developing the reference case fleet for MY 2017-2025. One
important element supporting the proposed approach is that AEO2011 projects relatively
stable gasoline prices over the next 15 years.  The average actual gasoline price in the U.S. for
the first nine months of 2011 of $3.57 per gallon ($3.38 in 2009 dollars)EE. However, the
AEO2011 reference case projects a 2011 price of $2.80 per gallon (in 2009 dollars), well
below actual prices. AEO2011 projects prices to be $3.25 in 2017, rising slightly to $3.54 per
gallon in 2025 (which is less than a 4 cent/year increase on average). Based on these fuel
price projections, the reference fleet for MYs 2017-2025 should correspond to a time period
where there is a stable, unchanging GHG standard, and essentially stable gasoline prices.

       EPA reviewed the historical record for similar periods when we had stable fuel
economy standards and stable gasoline prices. EPA maintains, and publishes every year, the
seminal reference on new light-duty vehicle CC>2 emissions and fuel economy.FF This report
contains very detailed data from MYs 1975-2010.  There was an extended 18-year period
from 1986 through 2003 during which CAFE standards were essentially unchanged,00 and
gasoline prices were relatively stable and remained below $1.50 per gallon for almost the
entire period. The 1975-1985  and 2004-2010 timeframes are not relevant in this regard due to
either rising gasoline prices, rising CAFE standards, or both. Thus, the 1986-2003 time frame
is an excellent analogue to the period out to MY  2025 during which AEO projects relatively
stable gasoline prices. EPA analyzed the Fuel Economy Trends data from the 1986-2003
timeframe (during which CAFE standards were universal rather than attribute-based ) and
have drawn three conclusions: 1) there was a small, industry-wide, average over-compliance
with CAFE on the order of 1-2 mpg or 3-4%, 2) almost all of this industry-wide over-
compliance was from 3 companies (Toyota, Honda, and Nissan) that routinely over-complied
with the universal CAFE standards simply because they produced  smaller and lighter vehicles
relative to the industry average, and 3) full line car and truck manufacturers, such as General
Motors, Ford, and Chrysler, which produced larger and heavier vehicles relative to the
industry average and which were constrained by  the universal CAFE standards, rarely over-
complied during the entire 18-year period.


17 Previous OMEGA documentation for versions used in MYs 2012-2016 Final Rule (EPA-
420-B-09-035), Interim Joint TAR (EPA-420-B-10-042)

18 http://www.epa.gov/oms/climate/models.htm

19 EPA-420-R-09-016, September 2009.
EE The Energy Information Administration estimated the average regular unleaded gasoline price in the U.S. for
the first nine months of 2011 was $3.57 per gallon.
FF Light-Duty Automotive Technology, Carbon Dioxide Emissions, and Fuel Economy Trends: 1975 through
2010, November 2010, available at www.epa.gov/otaq/fetrends.htm.
GG There are no EPA LD GHG emissions regulations prior to MY 2012.
                                         3-15

-------
Chapter 3
                                                   Table 3.5-1
                              Fuel Economy Data for Selected Manufacturers, 1986-2003—Cars
Year
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
Standard
27.5
27.5
27.5
27.5
27.5
27.5
27.5
27.5
27.5
27.5
27.5
27.5
27.5
27.5
27.5
27.5
27.5
27.5
Average 1986-2003
GM
27.0
27.2
28.1
27.4
27.3
27.2
26.7
27.3
27.5
27.3
27.9
28.2
27.6
27.4
27.6
28.1
28.5
28.6

Ford
26.7
26.5
27.0
26.9
26.3
27.2
26.7
27.8
27.1
27.6
26.3
26.9
27.3
27.2
27.1
26.8
27.1
26.7

Chrysler
28.6
Til
28.5
28.0
27.4
27.5
27.7
27.9
26.2
28.2
27.2
27.2
28.3
27.0
27.6
27.6
27.0
28.5

Sales-
Weighted
average
27.1
27.1
27.8
27.3
27.0
27.2
26.8
27.6
27.2
27.6
27.3
27.6
27.6
27.3
27.4
27.6
27.8
27.9

Delta
-0.4
-0.4
0.3
-0.2
-0.5
-0.3
-0.7
0.1
-0.3
0.1
-0.2
0.1
0.1
-0.2
-0.1
0.1
0.3
0.4
-0.1
Vehicle
Weight
3145
3149
3157
3207
3298
3252
3329
3269
3334
3330
3388
3353
3347
3429
3448
3463
3442
3506





















Toyota
32.3
32.9
32.7
31.8
30.4
30.6
28.9
29.0
29.1
30.0
29.5
29.8
30.2
30.4
30.5
31.3
30.7
32.4

Honda
33.6
32.8
31.8
31.3
30.4
30.3
30.9
32.2
32.1
32.8
31.8
32.1
32.0
30.9
31.0
32.2
32.0
32.7

Nissan
29.9
29.3
30.6
30.2
28.4
29.0
29.9
29.1
29.8
29.2
30.2
29.6
30.2
29.6
28.0
28.3
28.9
27.9

Sales-
Weighted
average
32.0
31.5
31.8
31.2
29.9
30.1
29.9
30.1
30.3
30.8
30.5
30.6
30.9
30.4
30.2
31.0
30.8
31.5

Delta
4.5
4.0
4.3
3.7
2.4
2.6
2.4
2.6
2.8
3.3
3.0
3.1
3.4
2.9
2.7
3.5
3.3
4.0
3.3
Vehicle
Weight
2706
2782
2779
2822
2943
2950
3051
3071
3084
3102
3126
3122
3249
3280
3258
3233
3303
3276





















Vehicle
Weight
delta
439
368
378
385
355
303
279
198
250
228
262
230
98
148
190
230
140
230
262
                                                      5-16

-------
                                                    2017 Draft Regulatory Impact Analysis
                          Table 3.5-2
Fuel Economy Data for Selected Manufacturers, 1986-2003—Trucks
Year
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
Standard
20.0
20.5
20.5
20.5
20.0
20.2
20.2
20.4
20.5
20.6
20.7
20.7
20.7
20.7
20.7
20.7
20.7
20.7
Average 1986-2003
GM
20.2
20.5
20.2
20.4
19.8
21.2
20.3
20.3
20.2
20.1
20.8
20.4
21.2
20.3
20.7
20.4
19.8
20.2

Ford
20.3
20.5
20.6
20.1
20.2
20.5
20.2
20.8
20.8
20.6
20.8
20.2
20.2
19.8
20.0
20.1
20.2
20.0

Chrysler
20.7
21.3
21.4
21.0
21.4
21.1
21.3
21.2
20.5
20.1
20.2
20.2
20.0
19.9
20.4
19.5
20.0
20.9

Sales-
Weighted
average
20.3
20.7
20.6
20.5
20.3
20.9
20.5
20.7
20.5
20.3
20.6
20.3
20.5
20.0
20.4
20.0
20.0
20.3

Delta
0.3
0.2
0.1
0.0
0.3
0.7
0.3
0.3
0.0
-0.3
-0.1
-0.4
-0.2
-0.7
-0.3
-0.7
-0.7
-0.4
-0.1
Vehicle
Weight
3917
3876
3961
4016
4102
4026
4132
4141
4204
4248
4295
4445
4376
4508
4456
4591
4686
4738





















Toyota
26.1
25.9
24.4
23.2
21.8
22.4
21.9
22.1
22.0
21.2
23.1
22.6
23.4
23.0
22.0
22.3
22.2
22.0

Honda








20.2
25.5
22.2
24.7
25.5
25.2
25.0
24.7
25.3
24.8

Nissan
24.7
23.5
22.7
23.7
25.3
24.8
24.0
23.7
22.9
22.4
22.9
22.3
22.3
21.2
20.8
20.7
20.7
21.9

Sales-
Weighted
average
25.5
24.9
23.8
23.3
23.2
23.1
22.5
111
22.3
22.0
23.0
22.8
23.5
23.1
22.2
22.3
22.5
22.9

Delta
5.5
4.4
3.3
2.8
3.2
2.9
2.3
2.3
1.8
1.4
2.3
2.1
2.8
2.4
1.5
1.6
1.8
2.2
2.6
Vehicle
Weight
3240
3259
3352
3420
3528
3628
3620
3637
3711
3797
3678
3734
3762
3943
4098
4125
4149
4195





















Vehicle
Weight
delta
677
617
609
596
574
397
512
505
494
452
617
711
614
564
359
465
537
544
547
                             5-17

-------
Chapter 3
       Since the MY 2012-2016 standards are footprint-based, every major manufacturer is
expected to be constrained by the new standards in 2016 and manufacturers of small vehicles
will not routinely over-comply as they had with the past universal standards.HH  Thus, the
historical evidence and the footprint-based design of the 2016 GHG emissions and CAFE
standards strongly support the use of a reference case fleet where there are no further fuel
economy improvements beyond those required by the MY 2016  standards.  There are
additional factors that reinforce the historical evidence.  While it is possible that one or two
companies may over-comply, any voluntary over-compliance by one company would
generate credits that could be sold to other companies to substitute for their more expensive
compliance technologies; this ability to buy and sell credits could eliminate any over-
compliance for the overall fleet.20

       Figure 3.5-2  shows that,  over the 1986-2003 period discussed above, overall average
fleetwide fuel economy decreased by about 3 mpg, even with stable car CAFE standards and
very slightly increasing truck CAFE standards, as the market shifted from a market dominated
by cars in the 1980s to  one split between cars and trucks in 2003.n All projections of actual
GHG emissions and fuel economy performance in 2016 or any other future year are
projections, of course, and it is plausible that actual GHG emissions and fuel economy
performance in 2017-2025, absent more stringent standards, could be lower than projected if
there are shifts from car market share to truck market share, or to higher footprint  levels.

  Figure 3.5-2 Average Fleetwide Light-Duty Vehicle Fuel Economy, Horsepower, and
                                  Weight, 1975-2010
        (fuel economy data is consumer label values, about 20% lower than compliance values)
          24.0
          23.0
          22.0
          21 .O
          2O.O
          19.O
                                                                        8O%
                                                                        6O%
                                                                        4O%
                                                                        2O%
                                                                        O%
                                                                        -2O%
          12.O
                                                                        -4O%
             1975
                    198O
                           1985
                                  199O
                                         1995
                                                2OOO
                                                       2OO5
                                                              2O1O
                                                                     2O15
101 With the notable exception of manufacturers who only market electric vehicles or other limited product lines.
11 Note that the mpg values in this one figure are consumer label values, not the CAFE/compliance values shown
throughout this preamble.  Consumer label values are typically about 20% lower than compliance values. The
trends are the same.
                                         3-18

-------
                                                2017 Draft Regulatory Impact Analysis
       Consistent with this discussion, for the reference case, EPA configured the OMEGA
model to determine the cost to comply with the MY 2016 standards and did not allow access
to the post-MY 2016 technology caps. This reflects the belief that manufacturers will (a)
need to comply in MY 2016, and so will not add additional technology to their vehicles
afterwards to comply with GHG standards (b) will use that new technology for attributes
other than fuel economy, since their vehicles are already compliant,  (c) in the absence of
additional regulation beyond the MYs 2012-2016 rule would not develop many of the
technologies become available under the control case runs. Similarly, the air conditioning
technology usage was capped at the MY 2016 projections, as manufacturers that were already
compliant would have no need to add additional air conditioning technology (especially as the
alternative refrigerant cost is significantly higher than the present refrigerant).

       EPA ran the OMEGA model three times with the same MY 2016 technology input but
with the market data file configured to MY 2016, MY 2021, and MY 2025 sales.  The model
was run three times because car/truck sales mix shifts between 2016 and 2025 require some
manufacturers to add minimal additional technology to their vehicles in order to remain in
compliance. While slight additional amounts of technology are added or removed, the
compliance cost for the MY 2016 rule decline over time as a result of the learning effects
discussed in the RIA Chapter 1. To reflect this learning progression,  but also that the
technology choices were made during MY 2016, OMEGA was run with MY 2016 costs,
which were then post-processed to the proper cost-year.

       Consistent with the MYs 2012-2016 rule analysis, EPA did not allow EVs and PHEVs
(maximum penetration caps of zero) in the reference case. While the penetration of EVs and
PHEVs in MY 2016 will like be non-zero, as they are being sold in MY 2011, EPA chose not
to include these technologies in the reference case assessment due to their cost-distorting
effects on the smallest companies (Table 3.5-3 ).  In the OMEGA projections, the vast
majority of companies do not use EVs or PHEVs to comply with the MY 2016 standards.
Five smaller companies under the technology restrictions set forth in this  analysis, cannot
comply with the MY 2016 standards." This finding is consistent with the MY 2012-2016 rule
analysis, and are Daimler, Geely-Volvo, Volkswagen, Porsche and Tata (which is comprised
of Jaguar and Land Rover vehicles in the U.S. fleet).21

       As shown below, these manufacturers (other than Porsche) could comply with the MY
2016 standards by including electric vehicles and plug-in hybrids in their fleet. As reflected
in the MY 2012-2016 rule, EPA believes that it is unlikely that these manufacturers will make
8%-10% of their fleet EVs and PHEVs by MY 2016.  As an alternative to this choice, these
11 While OMEGA model results are presented assuming that all manufacturers must comply with the program as
proposed (to the extent that they can), some manufacturers, such as small volume manufacturers may be eligible
for additional options (and alternative standards) which have not been considered here. As described in the
preamble, small volume manufacturers with U.S. sales of less than 5,000 vehicles would be able to petition EPA
for an alternative standard for MY 2017 and later. Manufacturers currently meeting the 5,000 vehicle cut point
include Lotus, Aston Martin, and McLaren.
                                         3-19

-------
Chapter 3

companies could exceed the technology caps on other technologies (such as mass reduction),
make use of carry-forward credits, carry-back credits, or purchase credits from another
manufacturer. Alternatively, they could use a vehicle compliance strategy not considered
here, as discussed in section HID of the MY 2012-2016 rule.  Thus the compliance cost for
these vehicles for the 2016 rule could potentially be greater than presented in this analysis,
which would decrease the incremental cost of the proposed later MY standards.

       For these manufacturers, the MY 2016 reference case results presented are those with
the fully allowable application of technology available in EPA's OMEGA modeling analysis
and not for the technology projected to enable compliance with the final MY 2016  standards.
KK Again, this analytic choice increases the incremental costs of the MY 2017-MY 2025
program for these companies.
                    Table 3.5-3 - MY 2016 EV+PHEV Penetrations
Manufacturer






Daimler
Geely-Volvo
Porsche
Tata
Volkswagen
MY 2016
Shortfall
without
EV/PHEV
(g/mile)


17
18
46
20
10
MY 2016
Shortfall
with
EV/PHEV
(g/mile)


-
-
23
-
-
Reference
Cost
Delta
added by
including
EVs
($)
$1,506
$1,869
$2,570
$1,826
$645
EV+PHEV
(% of MY
2016 Sales
if added)



8%
9%
11%
10%
5%
       The MY 2016 coefficients are found in 75 FR at 25409. When input to OMEGA,
these coefficients were adjusted vertically upward by 10.2 grams (cars) and 11.4 grams
(trucks) to account for external calculations relating to air conditioning costs.

       No additional compliance flexibilities were  explicitly modeled for the MY 2016
standards. The EPA flexible fueled vehicle credit expires before MY 2016.LL The Temporary
Leadtime Allowance Alternative Standards (TLAAS), as analyzed in RIA chapter 5 of the
KK In the OMEGA analysis, only BMW's MY 2016 compliance costs increase (by ~$350) because EV and
PHEV technology was made unavailable.
LL The credit available for producing FFVs will have expired, although the real world usage credits will be
available.
                                         3-20

-------
                                                2017 Draft Regulatory Impact Analysis
MY 2012-2016 rule, is projected have an impact of approximately 0.1 g/mile in MY 2016,
and expire afterwards. Therefore, no incentive credits are projected to be available to the
reference case.  Off-cycle credits, which are designed to be environmentally neutral, would
only lower costs.  These credits are not modeled here due to the difficult in predicting
manufacturers use of these credits under the MY 2016 program.

       With respect to car-truck trading, the OMEGA model facilitates the trading of car-
truck credits on a total lifetime CO2 emission basis, consistent with the provisions of the
proposal and the MY 2016 rule. For example, if a manufacturer over-complies with its
applicable CO2  standard for cars by 10 g/mi, sells 1,000,000 cars, and cars have a lifetime
VMT of 195,264 miles, it generates 1,952,640 metric tons  of CO2 credits. If these credits are
used to compensate for under-compliance towards the truck CO2 standard and truck sales are
500,000, with a lifetime truck VMT of 225,865  miles, the manufacturer's truck CO2 emission
level could be as much as 17.3 g/mi CO2 above the standard.  Car truck trading was allowed
in the OMEGA runs without limit consistent with the trading provisions of the MY 2012-
2016 and proposed MY 2017-2025  GHG rules.

       3.5.2  Control Scenarios

       Similar to the reference scenario, OMEGA runs were conducted in 2021 and 2025 for
the proposal and alternative scenarios. The standards for these scenarios were derived from
the coefficients  discussed in Section III.B of the preamble. The joint EPA/NHTSA
development of these target curve coefficients is discussed in Draft Joint TSD Chapter 2. As
in MYs 2012-2016, the curves from that joint fitting process were adjusted for air
conditioning through a negative additive offset based on the estimated year over year
penetrations of air conditioning shown in preamble III.C and below.  For the OMEGA cost
analysis, as we analyzed air conditioning costs outside of the model, we re-adjusted the model
input curves to remove this projected penetration of air conditioning technology. For the MY
2021 and MY 2025 OMEGA runs,  air conditioning credits were projected at 18.8 for cars and
24.4 for light trucks..

       EPA's NPRM incorporates  several additional compliance flexibilities.  See generally
Preamble section III.C for an extended discussion of these  credits. EVs and PHEVs were
modeled with zero g/mile in all cases. As discussed in Section III.B  of the preamble, the cap
for EVs and PHEVs at zero g/mile is related to the standard level proposed.  For purposes of
this cost modeling, we assume that  this cap is never reached. This does not imply that EPA
has proposed a cap based on this criteria. The proposed PH/EV multipliers were not modeled
in this analysis,  but may be considered in the final rule analysis. A discussion of the potential
impacts of these credits can be found in preamble section III.B and DRIA chapter 4. Costs
beyond MY 2025 assume no technology changes on the vehicles, and implicitly assume that
the compliance  values for EVs remains at zero gram/mile.MM
MM The costs for PHEVs and EVs in this rule reflect those costs discussed in Draft Joint TSD Chapter 3, and do
not reflect any tax incentives, as the availability of those tax incentives in this time frame is uncertain.


                                         3-21

-------
Chapter 3

       The proposed credit for HEV and performance based pickups was also not modeled in
this proposal analysis of costs. Off-cycle credits, which are not modeled here, could only
reduce costs. A discussion of the potential impacts on a g/mile and total tons basis can be
found in DRIA chapter 4.

       Like the reference case, car truck trading was allowed without limit. Depending on
comment and other new input, these proposed flexibilities may be modeled differently for the
final rule.

3.6 Fuels and reference data

       Fuels data was based on AEO fuel prices, as documented in Chapter 4 of the Draft
Joint TSD. Estimates of carbon and energy content per gallon of liquid fuel are consistent
with the MY 2012-2016 rule analysis.

       VMT used in the payback analysis, which is used for calculating TARFs, was
determined using the EPA benefits post-processor. As the general VMT formula used in this
proposal is dependent on a vehicle's fuel cost per mile (see Draft Joint TSD Chapter 4), this
was determined in an iterative process.

3.7 OMEGA model calculations

       Using the data and equations discussed above, the OMEGA model begins by
determining the specific CO2 emission standard applicable for each manufacturer and  its
vehicle class (i.e., car or truck). As the reference case, the proposal, and all alternatives allow
for averaging across a manufacturer's cars and trucks, the model  determines the CO2 emission
standard applicable to each manufacturer's car and truck sales from the two sets of
coefficients describing the piecewise linear standard functions for cars and trucks (i.e. the
respective car  and truck curves) in the inputs, and creates a combined car-truck standard. This
combined standard considers the difference in lifetime VMT of cars and trucks, as indicated
in the proposed regulations which govern credit trading between these two vehicle classes.NN

       The model then works with one manufacturer at  a time to add technologies until that
manufacturer meets its applicable proposed standard. The OMEGA model can utilize several
approaches to  determining the order in which vehicles receive technologies. For this analysis,
EPA used a "manufacturer-based net cost-effectiveness factor" to rank the technology
packages in the order in which a manufacturer is likely to apply them.  Conceptually, this
approach estimates the cost of adding the technology from the manufacturer's perspective and
divides it by the mass of CO2 the technology will reduce. One component of the cost of
adding a technology is its production cost, as discussed above.  However, it is expected that
new vehicle purchasers value improved fuel economy since it reduces the cost of operating
the vehicle.  Typical vehicle purchasers are assumed to value the fuel savings accrued over
the period of time which they will own the vehicle, which is estimated to be approximately
^ The analysis for the control cases in this proposal was run with slightly different lifetime VMT estimates than
those proposed in the regulation. The impact is on the cost estimates is small and varies by manufacturer.


                                         3-22

-------
                                                  2017 Draft Regulatory Impact Analysis
five years.00 It is also assumed that consumers discount these savings at the same rate as that
used in the rest of the analysis (3 or 7 percent).pp Any residual value of the additional
technology which might remain when the vehicle is sold is not considered for this analysis.
The CC>2 emission reduction is the change in CC>2 emissions multiplied by the percentage of
vehicles surviving after each year of use multiplied by the annual miles travelled by age.

       Given this definition, the higher priority technologies are those with the lowest
manufacturer-based net cost-effectiveness value (relatively low technology cost or high fuel
savings leads to lower values). Because the order of technology application is set for each
vehicle, the model uses the manufacturer-based net cost-effectiveness primarily to decide
which vehicle receives the next technology addition. Initially, technology package #1 is the
only one available to any particular vehicle. However, as soon as a vehicle receives
technology package #1, the model considers the manufacturer-based net cost-effectiveness of
technology package #2 for that vehicle and so on.  In general terms, the equation describing
the calculation of manufacturer-based cost effectiveness is as follows:

        Equation 3.7-1 - Calculation of Manufacturer-Based Cost Effectiveness

                                              ATechCost  - AFS
                        CostEffManuft =
                                            &C02xVMTregulatory

       Where:

       CostEffManuft= Manufacturer-Based Cost Effectiveness (in dollars per kilogram CO2),
       TechCost = Marked up cost of the technology (dollars),
       FS = Difference in fuel consumption due to the addition of technology times fuel price and discounted
       over the payback period, or the number of years of vehicle use over which consumers value fuel savings
       when evaluating the value of a new vehicle at time of purchase
       dCO2 = Difference in CO2 emissions (g/mile) due to the addition of technology
       VMTregulatoiy = the statutorily defined VMT

       EPA describes the technology ranking methodology and manufacturer-based cost
effectiveness metric in greater detail in the OMEGA documentation.22  Please note that the
TARF equation does not consider attributes other than cost effectiveness and relative fuel
savings.  This distinction  is significant when considering the technology penetrations
presented later in this chapter.  An electric vehicle, which is approximately the same cost as a
plug-hybrid but is significantly more effective over the certification cycles, will generally be
chosen before the plug-in hybrid.  The current TARF does not reflect potential consumer
concerns with the range limits of the electric vehicle, although it could be modified to do so.
As a result of EVs greater cost-effectiveness, relatively more (although still few in an absolute
sense) are shown in the projected technology penetrations. When calculating the fuel savings
00 For a fuller discussion of this topic see Section III.H
pp While our costs and benefits are discounted at 3% or 7%, the decision algorithm (TARF) used in OMEGA
was run at a discount rate of 3%. Given that manufacturers must comply with the standard regardless of the
discount rate used in the TARF, this has little impact on the technology projections shown here.  Further, the fuel
savings aspect of the TARF are only directly relevant when two different fuels are being compared, because the
fuel saving/delta CO2 ratio is a constant for any given vehicle on a single fuel in a single model year.
                                           3-23

-------
Chapter 3

in the TARF equation, the full retail price of fuel, including taxes is used. While taxes are not
generally included when calculating the cost or benefits of a regulation, the net cost
component of the manufacturer-based net cost-effectiveness equation is not a measure of the
social cost of this proposed rule, but a measure of the private cost, (i.e., a measure of the
vehicle purchaser's willingness to pay more for a vehicle with higher fuel efficiency).  Since
vehicle operators pay the full price of fuel, including taxes, they value fuel costs or savings at
this level, and the manufacturers will consider this when choosing among the technology
options. QQ

        The values of manufacturer-based net cost-effectiveness for specific technologies will
vary from vehicle to vehicle, often substantially. This occurs for three reasons.  First, both the
cost and fuel-saving component cost, ownership fuel-savings, and lifetime CC>2 effectiveness
of a specific technology all vary by the  type of vehicle or engine to which it is being applied
(e.g., small car versus large truck, or 4-cylinder versus 8-cylinder engine).  Second,  the
effectiveness  of a specific technology often depends on the presence of other technologies
already being used on the vehicle (i.e., the dis-synergies).  Third, the absolute fuel savings and
CC>2 reduction of a percentage an incremental reduction in fuel consumption depends on the
CC>2 level of the  vehicle prior to adding the technology.  Chapter 1 of EPA's draft RIA
contains further detail on the values  of manufacturer-based net cost-effectiveness for the
various technology packages.

3.8 Analysis Results

        3.8.1  Targets and Achieved Values

        3.8.1.1 Reference Case
oo
   This definition of manufacturer-based net cost-effectiveness ignores any change in the residual value of the
vehicle due to the additional technology when the vehicle is five years old. Based on historic used car pricing,
applicable sales taxes, and insurance, vehicles are worth roughly 23% of their original cost after five years,
discounted to year of vehicle purchase at 7% per annum.  It is reasonable to estimate that the added technology
to improve CO2 level and fuel economy will retain this same percentage of value when the vehicle is five years
old. However, it is less clear whether first purchasers, and thus, manufacturers consider this residual value when
ranking technologies and making vehicle purchases, respectively.  For this proposal, this factor was not included
in our determination of manufacturer-based net cost-effectiveness in the analyses.
                                            3-24

-------
                                     2017 Draft Regulatory Impact Analysis
Table 3.8-1 Reference Case Targets and Projected Shortfall in MY 2021
Manufacturer
Aston Martin
BMW
Chrysler/Fiat
Daimler
Ferrari
Ford
Geely
General Motors
Honda
Hyundai
Kia
Lotus
Mazda
Mitsubishi
Nissan
Porsche
Spyker
Subaru
Suzuki
Tata
Tesla
Toyota
Volkswagen
Fleet
Car Target
222
228
230
239
235
230
232
229
222
223
225
206
223
219
227
206
222
216
208
250
206
221
219
225
Truck Target
-
285
295
301
-
306
280
308
283
280
291
-
276
270
294
287
280
267
272
273
-
293
296
297
Fleet Target
(VMT and
Sales Weighted)
222
245
261
256
235
259
248
271
243
236
241
206
233
238
249
227
231
229
221
262
206
251
236
253
Fleet Target (Sales
weighted)
222
243
259
254
235
256
247
268
241
235
239
206
232
237
247
225
230
228
219
261
206
249
234
250
Car
Achieved
342
234
224
251
386
232
246
227
214
224
222
241
228
222
222
251
249
229
209
244
-
209
222
222
Truck
Achieved
-
284
300
329
-
304
302
310
297
277
299
-
253
265
303
333
325
230
268
322
-
309
326
304
Shortfall
119
3
-
17
152
-
17
-
-
-
-
35
-
-
-
46
30
0
-
24
-
-
9
1
                               3-25

-------
Chapter 3
      Table 3.8-2  Reference Case Targets and Projected Shortfall in MY 2025
Manufacturer
Aston Martin
BMW
Chrysler/Fiat
Daimler
Ferrari
Ford
Geely
General Motors
Honda
Hyundai
Kia
Lotus
Mazda
Mitsubishi
Nissan
Porsche
Spyker
Subaru
Suzuki
Tata
Tesla
Toyota
Volkswagen
Fleet
Car Target
222
228
229
239
235
230
232
229
222
223
224
206
223
219
227
206
222
216
208
250
206
221
219
225
Truck Target
-
286
294
302
-
304
280
307
282
280
292
-
277
270
292
287
280
267
272
273
-
292
296
295
Fleet Target
(VMT and
Sales Weighted)
222
245
259
255
235
255
248
269
242
236
240
206
233
238
248
226
230
229
220
261
206
249
236
251
Fleet Target (Sales
weighted)
222
243
257
253
235
253
246
266
240
234
239
206
232
236
246
224
229
227
219
260
206
247
234
248
Car
Achieved
342
234
225
253
386
232
245
226
214
224
221
241
228
222
222
251
249
228
209
244
-
209
222
222
Truck
Achieved
-
287
299
329
-
301
302
307
299
277
303
-
255
265
302
333
325
231
268
322
-
309
326
303
Shortfall
119
4
-
17
152
-
16
-
-
-
-
35
-
-
-
46
30
0
-
22
-
-
9
1
                                      3-26

-------
                                              2017 Draft Regulatory Impact Analysis
     3.8.1.1 Proposal and Alternatives




             Table 3.8-3 Proposal Targets and Projected Shortfall in MY 2021
Manufacturer
Aston Martin
BMW
Chrysler/Fiat
Daimler
Ferrari
Ford
Geely
General Motors
Honda
Hyundai
Kia
Lotus
Mazda
Mitsubishi
Nissan
Porsche
Spyker
Subaru
Suzuki
Tata
Tesla
Toyota
Volkswagen
Fleet
Car Target
171
175
176
184
181
177
178
176
170
171
172
157
171
168
174
157
170
165
158
193
157
169
167
173
Truck Target
-
236
246
252
-
261
230
261
233
230
242
-
226
220
247
237
230
217
222
223
-
245
247
249
Fleet Target
(VMT and
Sales Weighted)
171
193
211
203
181
209
196
222
192
185
190
157
182
188
199
179
180
179
171
209
157
202
186
203
Fleet Target (Sales
weighted)
171
191
208
201
181
205
194
218
190
183
188
157
181
186
197
176
179
177
170
208
157
199
184
199RR
Car
Achieved
193
181
182
180
220
188
176
188
174
177
180
157
178
178
175
149
164
180
162
169
-
175
163
178
Truck
Achieved
-
222
239
263
-
243
234
250
225
210
218
-
197
204
243
260
258
177
207
243
-
238
258
239
Shortfall
22
-
-
-
39
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
While OMEGA does not model changes in vehicle sales, Draft RIA Chapter 8 discusses this topic.
                                       3-27

-------
Chapter 3
              Table 3.8-4 Proposal Targets and Projected Shortfall in MY 2025
Manufacturer
Aston Martin
BMW
Chrysler/Fiat
Daimler
Ferrari
Ford
Geely
General Motors
Honda
Hyundai
Kia
Lotus
Mazda
Mitsubishi
Nissan
Porsche
Spyker
Subaru
Suzuki
Tata
Tesla
Toyota
Volkswagen
Fleet
Car Target
142
146
146
153
150
147
148
146
142
142
143
131
142
139
145
131
141
137
132
161
131
141
139
144
Truck Target
-
194
201
208
-
213
189
213
191
188
199
-
186
180
202
195
188
177
181
182
-
200
203
203
Fleet Target
(VMT and
Sales Weighted)
142
160
172
167
150
170
162
181
158
153
157
131
150
154
164
146
149
147
141
172
131
165
154
166
Fleet Target (Sales
weighted)
142
159
170
166
150
167
160
178
156
151
155
131
149
153
162
144
148
146
140
171
131
163
152
163
Car
Achieved
142
145
148
146
159
153
141
146
143
145
146
131
145
144
143
119
133
147
132
134
-
140
133
144
Truck
Achieved
-
196
199
230
-
200
204
212
186
178
189
-
172
171
204
231
231
149
179
208
-
201
225
202
Shortfall
-
-
-
-
9
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
                                      3-28

-------
                                     2017 Draft Regulatory Impact Analysis
Table 3.8-5 Alternative 1- (Trucks +20) Targets and Projected Shortfall in MY
                               2021
Manufacturer
Aston Martin
BMW
Chrysler/Fiat
Daimler
Ferrari
Ford
Geely
General Motors
Honda
Hyundai
Kia
Lotus
Mazda
Mitsubishi
Nissan
Porsche
Spyker
Subaru
Suzuki
Tata
Tesla
Toyota
Volkswagen
Fleet
Car Target
171
175
176
184
181
177
178
176
170
171
172
157
171
168
174
157
170
165
158
193
157
169
167
173
Truck Target
-
256
267
273
-
284
250
283
253
250
263
-
245
238
267
258
249
235
241
242
-
266
268
270
Fleet Target
(VMT and
Sales Weighted)
171
199
221
209
181
217
203
234
199
189
195
157
186
195
206
184
183
184
175
219
157
211
191
211
Fleet Target (Sales
weighted)
171
197
217
206
181
213
201
229
196
187
193
157
184
192
203
181
181
182
173
217
157
207
188
207
Car
Achieved
193
186
195
188
220
196
186
200
182
180
185
157
181
184
185
154
168
184
167
190
0
180
170
186
Truck
Achieved
0
230
248
263
0
253
235
262
229
216
224
0
204
212
245
264
259
182
207
244
0
251
258
248
Shortfall
22
-
-
-
39
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
                               3-29

-------
Chapter 3
        Table 3.8-6 Alternative 2- (Trucks -20) Targets and Projected Shortfall in MY
                                      2021
Manufacturer
Aston Martin
BMW
Chrysler/Fiat
Daimler
Ferrari
Ford
Geely
General Motors
Honda
Hyundai
Kia
Lotus
Mazda
Mitsubishi
Nissan
Porsche
Spyker
Subaru
Suzuki
Tata
Tesla
Toyota
Volkswagen
Fleet
Car Target
171
175
176
184
181
177
178
176
170
171
172
157
171
168
174
157
170
165
158
193
157
169
167
173
Truck Target
0
217
227
232
0
241
212
241
215
212
223
0
208
202
228
219
212
199
204
205
0
226
228
230
Fleet Target
(VMT and
Sales Weighted)
171
188
201
198
181
201
190
211
186
181
185
157
178
181
193
174
177
174
168
199
157
194
181
195
Fleet Target (Sales
weighted)
171
186
199
196
181
199
189
208
184
179
184
157
177
180
191
172
176
173
167
199
157
192
180
193
Car
Achieved
193
173
176
172
220
181
167
175
169
171
174
157
174
170
167
143
161
173
159
162
0
167
158
171
Truck
Achieved
0
222
227
263
0
234
234
242
216
210
217
0
195
198
240
259
258
177
204
231
0
228
257
232
Shortfall
22
-
-
-
39
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
                                       3-30

-------
                                          2017 Draft Regulatory Impact Analysis
Table 3.8-7 Alternative 3- (Cars +20) Targets and Projected Shortfall in MY 2021
Manufacturer
Aston Martin
BMW
Chrysler/Fiat
Daimler
Ferrari
Ford
Geely
General Motors
Honda
Hyundai
Kia
Lotus
Mazda
Mitsubishi
Nissan
Porsche
Spyker
Subaru
Suzuki
Tata
Tesla
Toyota
Volkswagen
Fleet
Car Target
190
196
197
205
202
197
199
196
190
191
192
176
191
187
194
176
190
184
177
215
176
189
187
193
Truck Target
0
236
247
253
0
262
231
262
233
231
243
0
226
220
248
238
230
217
222
223
0
246
248
250
Fleet Target
(VMTand
Sales Weighted)
190
208
221
218
202
222
210
231
205
200
205
176
198
200
213
192
196
193
186
219
176
214
201
215
Fleet Target (Sales
weighted)
190
206
219
217
202
219
209
229
203
199
204
176
197
199
211
190
196
192
185
219
176
211
199
213
Car
Achieved
193
197
195
198
220
199
191
199
187
191
195
176
195
189
188
166
184
191
177
190
0
184
184
191
Truck
Achieved
0
233
249
267
0
259
244
259
239
231
235
0
210
215
259
264
260
196
217
244
0
252
258
251
Shortfall
2
-
-
-
18
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
                                   3-31

-------
Chapter 3
   Table 3.8-8  Alternative 4- (Cars -20) Targets and Projected Shortfall in MY 2021
Manufacturer
Aston Martin
BMW
Chrysler/Fiat
Daimler
Ferrari
Ford
Geely
General Motors
Honda
Hyundai
Kia
Lotus
Mazda
Mitsubishi
Nissan
Porsche
Spyker
Subaru
Suzuki
Tata
Tesla
Toyota
Volkswagen
Fleet
Car Target
151
155
156
163
160
157
158
156
150
151
152
139
151
148
154
139
150
146
140
171
139
150
148
153
Truck Target
0
236
247
253
0
262
231
262
233
231
243
0
226
220
248
238
230
217
222
223
0
246
248
250
Fleet Target
(VMT and
Sales Weighted)
151
179
201
188
160
196
183
213
179
170
175
139
166
176
186
165
164
165
157
199
139
191
171
191
Fleet Target (Sales
weighted)
151
177
197
185
160
192
180
208
176
167
173
139
164
173
183
162
162
163
155
197
139
187
168
187
Car Achieved
0
222
227
262
0
228
234
243
212
201
213
0
190
197
232
257
258
167
201
231
0
225
254
229
Truck Achieved
193
178
201
188
220
196
183
212
179
170
175
139
166
175
186
172
164
165
157
199
0
191
171
190
Shortfall
42
-
-
-
60
-
-
-
-
-
-
-
-
-
-
7
-
-
-
-
-
-
-
-
                                       3-32

-------
                                           2017 Draft Regulatory Impact Analysis
Table 3.8-9 Alternative 1- (Trucks +20) Targets and Projected Shortfall in MY 2025
Manufacturer
Aston Martin
BMW
Chrysler/Fiat
Daimler
Ferrari
Ford
Geely
General Motors
Honda
Hyundai
Kia
Lotus
Mazda
Mitsubishi
Nissan
Porsche
Spyker
Subaru
Suzuki
Tata
Tesla
Toyota
Volkswagen
Fleet
Car Target
142
146
146
153
150
147
148
146
142
142
143
131
142
139
145
131
141
137
132
161
131
141
139
144
Truck Target
0
213
221
228
0
234
207
234
210
207
218
0
204
198
221
214
207
195
200
200
0
220
223
223
Fleet Target
(VMT and
Sales Weighted)
142
166
182
173
150
177
168
192
164
157
161
131
154
161
170
151
151
152
145
181
131
173
158
173
Fleet Target (Sales
weighted)
142
164
178
170
150
174
166
188
162
155
159
131
152
159
168
149
150
150
143
179
131
170
156
170
Car
Achieved
0
202
207
230
0
204
204
219
194
184
194
0
172
176
213
231
232
157
183
215
0
206
225
208
Truck
Achieved
142
165
182
173
159
177
167
192
163
157
161
131
154
161
170
151
151
152
144
181
0
173
158
173
Shortfall
-
-
-
-
9
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
                                    3-33

-------
Chapter 3
  Table 3.8-10 Alternative 2- (Trucks -20) Targets and Projected Shortfall in MY 2025
Manufacturer
Aston Martin
BMW
Chrysler/Fiat
Daimler
Ferrari
Ford
Geely
General Motors
Honda
Hyundai
Kia
Lotus
Mazda
Mitsubishi
Nissan
Porsche
Spyker
Subaru
Suzuki
Tata
Tesla
Toyota
Volkswagen
Fleet
Car Target
142
146
146
153
150
147
148
146
142
142
143
131
142
139
145
131
141
137
132
161
131
141
139
144
Truck Target
0
174
182
187
0
192
170
192
172
170
179
0
167
162
182
175
170
159
163
164
0
181
183
183
Fleet Target
(VMT and
Sales Weighted)
142
154
163
162
150
163
156
170
152
148
152
131
147
148
157
142
146
143
138
162
131
157
149
158
Fleet Target (Sales
weighted)
142
153
161
161
150
161
155
168
150
148
151
131
146
147
156
140
145
142
137
162
131
155
148
157
Car
Achieved
142
136
136
138
159
146
130
138
137
140
142
131
141
134
134
113
130
141
128
126
0
130
127
136
Truck
Achieved
0
196
191
230
0
194
204
200
180
178
183
0
170
171
203
231
231
149
179
198
0
196
225
195
Shortfall
-
-
-
-
9
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
                                       3-34

-------
                                          2017 Draft Regulatory Impact Analysis
Table 3.8-11 Alternative 3- (Cars +20) Targets and Projected Shortfall in MY 2025
Manufacturer
Aston Martin
BMW
Chrysler/Fiat
Daimler
Ferrari
Ford
Geely
General Motors
Honda
Hyundai
Kia
Lotus
Mazda
Mitsubishi
Nissan
Porsche
Spyker
Subaru
Suzuki
Tata
Tesla
Toyota
Volkswagen
Fleet
Car Target
162
166
166
174
171
168
169
167
161
162
163
149
162
159
165
149
161
156
150
183
149
160
159
164
Truck Target
0
194
201
208
0
213
189
213
191
188
199
0
186
180
202
195
188
177
181
182
0
200
203
203
Fleet Target
(VMT and
Sales Weighted)
162
174
183
183
171
183
175
191
171
168
172
149
166
167
177
160
165
162
156
182
149
177
169
178
Fleet Target (Sales
weighted)
162
173
181
182
171
182
175
189
170
167
171
149
166
166
176
159
165
161
155
182
149
175
167
177
Car
Achieved
162
162
159
165
171
165
158
161
157
160
164
149
162
158
155
137
153
163
150
149
0
155
152
159
Truck
Achieved
0
202
208
233
0
217
210
218
196
193
195
0
183
182
221
231
232
157
183
215
0
207
225
211
Shortfall
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
                                    3-35

-------
Chapter 3
   Table 3.8-12  Alternative 4- (Cars -20) Targets and Projected Shortfall in MY 2025
Manufacturer
Aston Martin
BMW
Chrysler/Fiat
Daimler
Ferrari
Ford
Geely
General Motors
Honda
Hyundai
Kia
Lotus
Mazda
Mitsubishi
Nissan
Porsche
Spyker
Subaru
Suzuki
Tata
Tesla
Toyota
Volkswagen
Fleet
Car Target
122
126
126
132
130
127
128
126
122
122
123
112
122
120
125
112
122
118
113
139
112
121
120
124
Truck Target
0
194
201
208
0
213
189
213
191
188
199
0
186
180
202
195
188
177
181
182
0
200
203
203
Fleet Target
(VMT and
Sales Weighted)
122
146
162
152
130
157
148
171
145
137
142
112
134
142
150
133
132
133
127
161
112
153
138
153
Fleet Target (Sales
weighted)
122
144
158
149
130
153
146
167
142
135
139
112
133
140
147
130
130
131
125
159
112
150
136
150
Car
Achieved
137
123
135
125
159
137
120
139
130
126
129
112
128
127
128
101
114
132
116
122
0
127
114
129
Truck
Achieved
0
196
191
229
0
191
204
200
173
175
182
0
158
165
195
229
230
138
168
198
0
190
224
192
Shortfall
15
-
-
-
30
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
       3.8.1   Penetration of Selected Technologies

       OMEGA model projected penetrations of selected technologies by manufacturer,
model year, and car/truck class are presented on the following pages.  These tables show
results of both the reference case, the proposed standards as well as the four alternatives.
While OMEGA model results are presented assuming that all manufacturers must comply
with the program as proposed (to the extent that they  can), some manufacturers, such as small
volume manufacturers may be eligible for additional options (and alternative standards) which
have not been considered here. As described in the preamble, small volume manufacturers
with U.S. sales of less than 5,000 vehicles would be able to petition EPA for an alternative
standard for MY 2017 and later. Manufacturers currently meeting the 5,000 vehicle cut point
include Lotus, Aston Martin, and McLaren.

       Most obviously, while no manufacturer is actually restricted by the technology caps
modeled in this analysis, a smaller manufacturer with only a few vehicle platforms may
pursue a single technology path.
                                         3-36

-------
                                              2017 Draft Regulatory Impact Analysis
       The technology penetrations presented here are absolute, and include baseline
technologies.  The analyses shown here represent a single path towards compliance, of which
there are many.  The breadth of technology options in the Technical Assessment Report
analysis reflected these opportunities.

                       Table 3.8-13 Technology abbreviations
Abbreviation
Mass Tech Applied
True Mass
Mass Penalty
TDS 18/24/27
AT6/8
DCT6/8
MT
HEG
EGR
HEV
EV
PHEV
SS
LRRT2
IACC2
EFR2
DI
DSL
Meaning
Mass Technology Applied, expressed as a negative
number
Net Mass Reduced
Mass increase due to technology
turbocharged & downsized at 18/24/27 bar BMEP
Automatic transmission
Dual Clutch Transmission
Manual transmission
High Efficiency Gearbox
Cooled exhaust gas recirculation
Hybrid electric vehicle
Full electric vehicle
Plug-in HEV
12V stop-start
Lower rolling resistance tires level 2
Improved Accesssories level 2
Engine friction reduction level 2
Stoichiometric gasoline direct injection
Advanced diesel
20 Gates, Wallace E., Paul R. Portney, and Albert M. McGartland. "The Net Benefits of
Incentive-Based Regulation: A Case Study of Environmental Standard Setting." American
Economic Review 79(5) (December 1989): 1233-1242.
21
22
See 75 FR at 25457.

See OMEGA documentation at http://www.epa.gov/otaq/climate/models.htm.
                                        3-37

-------
Chapter 3
       3.8.2  Projected Technology Penetrations in Reference Case




Table 3.8-14  Reference Car Technology Penetrations in MY 2021

Aston Martin
BMW
Chrysler/Fiat
Daimler
Ferrari
Ford
Geely
General Motors
Honda
Hyundai
Kia
Lotus
Mazda
Mitsubishi
Nissan
Porsche
Spyker
Subaru
Suzuki
Tata
Tesla
Toyota
Volkswagen
Fleet
Mass Tech
Applied
-10%
-9%
-7%
-9%
-8%
-8%
-9%
-7%
-2%
-3%
-2%
-1%
-4%
-6%
-5%
-4%
-10%
-6%
0%
-10%
0%
-2%
-8%
-5%

w
15%
15%
0%
15%
15%
2%
15%
0%
3%
0%
0%
15%
0%
2%
1%
15%
15%
0%
2%
15%
0%
15%
15%
6%
w
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
100%
0%
0%
0%
w
PH
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
X
VI
57%
57%
0%
57%
57%
0%
57%
0%
0%
0%
0%
57%
0%
0%
0%
57%
57%
0%
0%
57%
0%
0%
57%
8%
LRRT2
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
IACC2
30%
30%
10%
30%
30%
20%
30%
16%
5%
5%
6%
30%
26%
26%
21%
30%
30%
20%
25%
30%
0%
6%
30%
15%

-------
                                                                              2017 Draft Regulatory Impact Analysis
Table 3.8-15 Reference Truck Technology Penetrations in MY 2021

Aston Martin
BMW
Chrysler/Fiat
Daimler
Ferrari
Ford
Geely
General Motors
Honda
Hyundai
Kia
Lotus
Mazda
Mitsubishi
Nissan
Porsche
Spyker
Subaru
Suzuki
Tata
Tesla
Toyota
Volkswagen
Fleet
Mass Tech
Applied
NA
-9%
-7%
-10%
NA
-7%
-10%
-8%
-4%
-5%
-4%
NA
-10%
-10%
-5%
-10%
-3%
-10%
-8%
-7%
NA
-3%
-10%
-6%
s|
H S
NA
-9%
-7%
-9%
NA
-7%
-9%
-8%
-4%
-5%
-4%
NA
-10%
-10%
-5%
-9%
-2%
-10%
-8%
-6%
NA
-3%
-9%
-6%
r-^i
M -T3
Gj %
~-* S
S£
NA
1%
0%
1%
NA
0%
1%
0%
0%
0%
0%
NA
0%
0%
0%
1%
1%
0%
0%
1%
NA
0%
1%
0%
TDS18
NA
60%
25%
64%
NA
56%
57%
33%
19%
85%
39%
NA
73%
68%
57%
59%
57%
70%
67%
55%
NA
13%
57%
37%
TDS24
NA
15%
7%
13%
NA
15%
15%
9%
0%
0%
0%
NA
15%
15%
15%
15%
15%
15%
15%
15%
NA
0%
15%
8%
TDS27
NA
0%
0%
0%
NA
0%
0%
0%
0%
0%
0%
NA
0%
0%
0%
0%
0%
0%
0%
0%
NA
0%
0%
0%
§
NA
70%
56%
55%
NA
42%
70%
57%
49%
61%
66%
NA
49%
39%
54%
69%
55%
18%
58%
51%
NA
56%
70%
54%
oo
S
NA
30%
26%
45%
NA
21%
30%
28%
19%
27%
30%
NA
19%
19%
23%
30%
30%
7%
24%
20%
NA
24%
30%
25%
\D
§
NA
0%
6%
0%
NA
14%
0%
4%
21%
5%
0%
NA
19%
19%
12%
0%
0%
37%
11%
19%
NA
7%
0%
9%
00
§
NA
0%
3%
0%
NA
8%
0%
2%
11%
3%
0%
NA
10%
11%
7%
0%
0%
20%
6%
10%
NA
4%
0%
5%
H
S
NA
0%
3%
0%
NA
3%
0%
0%
0%
0%
1%
NA
2%
0%
2%
1%
0%
8%
0%
0%
NA
3%
0%
2%
O
w
NA
0%
0%
0%
NA
0%
0%
0%
0%
0%
0%
NA
0%
0%
0%
0%
0%
0%
0%
0%
NA
0%
0%
0%
Pi
o
w
NA
15%
7%
13%
NA
15%
15%
9%
0%
0%
0%
NA
10%
15%
10%
15%
15%
5%
15%
15%
NA
0%
15%
7%
>
w
NA
15%
0%
15%
NA
2%
15%
0%
0%
0%
0%
NA
0%
2%
0%
15%
15%
0%
3%
15%
NA
5%
15%
3%
>
w
NA
0%
0%
0%
NA
0%
0%
0%
0%
0%
0%
NA
0%
0%
0%
0%
0%
0%
0%
0%
NA
0%
0%
0%
>
w
PH
NA
0%
0%
0%
NA
0%
0%
0%
0%
0%
0%
NA
0%
0%
0%
0%
0%
0%
0%
0%
NA
0%
0%
0%
C/3
C/3
NA
57%
0%
57%
NA
0%
57%
0%
0%
0%
0%
NA
0%
0%
0%
57%
57%
0%
0%
57%
NA
0%
57%
5%
LRRT2
NA
0%
0%
0%
NA
0%
0%
0%
0%
0%
0%
NA
0%
0%
0%
0%
0%
0%
0%
0%
NA
0%
0%
0%
IACC2
NA
30%
27%
30%
NA
24%
30%
27%
14%
0%
16%
NA
17%
30%
22%
30%
30%
8%
30%
30%
NA
21%
30%
23%
CN
Pi
CM
w
NA
0%
0%
0%
NA
0%
0%
0%
0%
0%
0%
NA
0%
0%
0%
0%
0%
0%
0%
0%
NA
0%
0%
0%
Q
NA
75%
32%
64%
NA
71%
72%
42%
19%
85%
39%
NA
88%
85%
72%
87%
72%
85%
85%
72%
NA
19%
87%
46%
_1
C/3
Q
NA
0
0
0
NA
-
0
-
-
-
-
NA
-
-
-
0
0
-
-
0
NA
-
0
0
                                                       J-J

-------
Chapter 3
Table 3.8-16 Reference Fleet (Sales-Weighted) Technology Penetration in MY 2021

Aston Martin
BMW
Chrysler/Fiat
Daimler
Ferrari
Ford
Geely
General Motors
Honda
Hyundai
Kia
Lotus
Mazda
Mitsubishi
Nissan
Porsche
Spyker
Subaru
Suzuki
Tata
Tesla
Toyota
Volkswagen
Fleet
Mass Tech
Applied
-10%
-9%
-7%
-9%
-8%
-7%
-9%
-7%
-3%
-3%
-3%
-1%
-5%
-8%
-5%
-6%
-9%
-7%
-2%
-8%
0%
-2%
-8%
-5%
S %
5 a
F S
-9%
-8%
-7%
-8%
-8%
-7%
-8%
-7%
-3%
-3%
-3%
0%
-5%
-7%
-5%
-5%
-8%
-7%
-1%
-8%
0%
-2%
-7%
-5%
1 |
>--H S
S£
1%
1%
0%
1%
1%
0%
1%
0%
0%
0%
0%
1%
0%
0%
0%
1%
1%
0%
0%
1%
0%
0%
1%
0%
TDS18
42%
51%
45%
48%
42%
52%
54%
37%
6%
28%
9%
54%
26%
68%
33%
48%
57%
46%
67%
48%
0%
5%
50%
30%
TDS24
15%
15%
11%
14%
15%
12%
15%
8%
0%
0%
0%
15%
11%
15%
8%
15%
15%
5%
15%
15%
0%
0%
15%
7%
TDS27
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
§
0%
28%
27%
16%
14%
30%
37%
38%
24%
24%
22%
0%
22%
17%
20%
20%
19%
6%
11%
33%
0%
26%
22%
27%
oo
S
0%
8%
12%
19%
0%
12%
12%
18%
6%
11%
9%
0%
7%
7%
8%
7%
4%
2%
4%
10%
0%
10%
6%
11%
\D
§
59%
36%
31%
39%
52%
29%
32%
20%
41%
32%
35%
10%
34%
39%
39%
19%
41%
39%
42%
37%
0%
28%
40%
31%
00
§
25%
19%
17%
21%
28%
16%
17%
11%
19%
17%
19%
5%
19%
22%
21%
10%
23%
22%
23%
20%
0%
15%
21%
16%
H
S
16%
9%
3%
5%
5%
6%
2%
3%
8%
6%
7%
85%
14%
6%
4%
43%
11%
22%
9%
0%
0%
5%
11%
6%
O
w
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
Pi
O
w
15%
15%
4%
14%
15%
10%
15%
6%
0%
0%
0%
15%
4%
15%
3%
15%
15%
3%
15%
15%
0%
0%
15%
5%
>
w
15%
15%
0%
15%
15%
2%
15%
0%
2%
0%
0%
15%
0%
2%
1%
15%
15%
0%
3%
15%
0%
12%
15%
5%
>
w
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
100%
0%
0%
0%
>
w
PH
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
C/3
C/3
57%
57%
0%
57%
57%
0%
57%
0%
0%
0%
0%
57%
0%
0%
0%
57%
57%
0%
0%
57%
0%
0%
57%
7%
LRRT2
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%

-------
                                                                              2017 Draft Regulatory Impact Analysis
Table 3.8-17 Reference Car Technology Penetrations in MY 2025

Aston
Martin
BMW
Chrysler/Fiat
Daimler
Ferrari
Ford
Geely
General
Motors
Honda
Hyundai
Kia
Lotus
Mazda
Mitsubishi
Nissan
Porsche
Spyker
Subaru
Suzuki
Tata
Tesla
Toyota
Volkswagen
Fleet
Mass Tech
Applied
-10%
-9%
-7%
-9%
-8%
-8%
-9%
-7%
-2%
-2%
-2%
-1%
-4%
-6%
-5%
-4%
-10%
-6%
0%
-10%
0%
-2%
-8%
-5%
s 1
F S
-9%
-8%
-7%
-8%
-8%
-8%
-8%
-7%
-2%
-2%
-2%
0%
-4%
-6%
-5%
-4%
-9%
-6%
0%
-9%
0%
-2%
-7%
-5%
>:
31
s£
1%
1%
0%
1%
1%
0%
1%
0%
0%
0%
0%
1%
0%
0%
0%
1%
1%
0%
0%
1%
0%
0%
1%
0%
TDS18
42%
47%
48%
43%
42%
51%
52%
43%
0%
13%
0%
54%
15%
68%
23%
45%
57%
47%
68%
42%
0%
0%
48%
27%
TDS24
15%
15%
13%
15%
15%
10%
15%
7%
0%
0%
0%
15%
10%
15%
4%
15%
15%
2%
15%
15%
0%
0%
15%
6%
TDS27
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
\D
%
0%
12%
3%
4%
14%
24%
22%
18%
13%
14%
9%
0%
16%
4%
5%
5%
13%
2%
1%
15%
0%
7%
9%
12%
oo
S
0%
0%
1%
11%
0%
8%
4%
8%
0%
7%
2%
0%
4%
0%
1%
0%
0%
0%
0%
0%
0%
1%
0%
4%
£
8
59%
48%
53%
52%
52%
35%
46%
37%
50%
39%
46%
10%
38%
50%
50%
25%
48%
40%
49%
55%
0%
52%
50%
45%
&
8
25%
26%
29%
28%
28%
19%
25%
20%
22%
21%
25%
5%
20%
27%
27%
14%
26%
22%
27%
30%
0%
12%
26%
21%
H
16%
13%
3%
6%
5%
7%
3%
6%
12%
7%
9%
85%
18%
9%
5%
56%
13%
27%
11%
0%
0%
7%
14%
8%
O
w
a
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
Pi
O
w
15%
15%
1%
15%
15%
8%
15%
2%
0%
0%
0%
15%
3%
15%
0%
15%
15%
2%
15%
15%
0%
0%
15%
4%
>
w
a
15%
15%
0%
15%
15%
1%
15%
0%
3%
0%
0%
15%
0%
2%
1%
15%
15%
0%
2%
15%
0%
16%
15%
6%
w
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
100%
0%
0%
0%
>
w
PH
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
VI
VI
57%
57%
0%
57%
57%
0%
57%
0%
0%
0%
0%
57%
0%
0%
0%
57%
57%
0%
0%
57%
0%
0%
57%
8%
LRRT2
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
IACC2
30%
30%
15%
30%
30%
19%
30%
15%
5%
5%
6%
30%
26%
26%
20%
30%
30%
17%
25%
30%
0%
6%
30%
15%

-------
Chapter 3
Table 3.8-18 Reference Truck Technology Penetrations in MY 2025

Aston Martin
BMW
Chrysler/Fiat
Daimler
Ferrari
Ford
Geely
General Motors
Honda
Hyundai
Kia
Lotus
Mazda
Mitsubishi
Nissan
Porsche
Spyker
Subaru
Suzuki
Tata
Tesla
Toyota
Volkswagen
Fleet
Mass Tech
Applied
NA
-9%
-7%
-10%
NA
-7%
-10%
-8%
-4%
-5%
-5%
NA
-10%
-10%
-5%
-10%
-3%
-10%
-8%
-7%
NA
-3%
-10%
-6%
S %
£s
NA
-8%
-7%
-9%
NA
-7%
-9%
-8%
-4%
-5%
-5%
NA
-10%
-10%
-5%
-9%
-2%
-10%
-8%
-6%
NA
-3%
-9%
-6%
>1
11
S£
NA
1%
0%
1%
NA
0%
1%
0%
0%
0%
0%
NA
0%
0%
0%
1%
1%
0%
0%
1%
NA
0%
1%
0%
TDS18
NA
60%
26%
64%
NA
57%
57%
36%
15%
85%
28%
NA
72%
68%
57%
59%
57%
70%
67%
55%
NA
9%
57%
36%
TDS24
NA
15%
8%
13%
NA
15%
15%
10%
0%
0%
0%
NA
15%
15%
12%
15%
15%
15%
15%
15%
NA
0%
15%
8%
TDS27
NA
0%
0%
0%
NA
0%
0%
0%
0%
0%
0%
NA
0%
0%
0%
0%
0%
0%
0%
0%
NA
0%
0%
0%
§
NA
70%
56%
55%
NA
41%
70%
58%
49%
60%
66%
NA
50%
39%
54%
69%
55%
18%
58%
51%
NA
55%
70%
54%
oo
S
NA
30%
26%
45%
NA
20%
30%
28%
18%
27%
30%
NA
20%
19%
23%
30%
30%
7%
24%
20%
NA
23%
30%
25%
\D
§
NA
0%
6%
0%
NA
15%
0%
4%
21%
5%
0%
NA
17%
19%
12%
0%
0%
37%
11%
19%
NA
7%
0%
9%
00
§
NA
0%
3%
0%
NA
8%
0%
2%
11%
3%
0%
NA
9%
11%
7%
0%
0%
20%
6%
10%
NA
4%
0%
5%
H
s
NA
0%
3%
0%
NA
3%
0%
0%
0%
0%
1%
NA
2%
0%
2%
1%
0%
8%
0%
0%
NA
3%
0%
2%
O
w
NA
0%
0%
0%
NA
0%
0%
0%
0%
0%
0%
NA
0%
0%
0%
0%
0%
0%
0%
0%
NA
0%
0%
0%
Pi
O
w
NA
15%
8%
13%
NA
15%
15%
10%
0%
0%
0%
NA
11%
15%
11%
15%
15%
5%
15%
15%
NA
0%
15%
7%
>
w
NA
15%
0%
15%
NA
3%
15%
0%
0%
0%
0%
NA
0%
2%
0%
15%
15%
0%
3%
15%
NA
6%
15%
3%
>
w
NA
0%
0%
0%
NA
0%
0%
0%
0%
0%
0%
NA
0%
0%
0%
0%
0%
0%
0%
0%
NA
0%
0%
0%
>
w
PH
NA
0%
0%
0%
NA
0%
0%
0%
0%
0%
0%
NA
0%
0%
0%
0%
0%
0%
0%
0%
NA
0%
0%
0%
C/3
C/3
NA
57%
0%
57%
NA
0%
57%
0%
0%
0%
0%
NA
0%
0%
0%
57%
57%
0%
0%
57%
NA
0%
57%
5%
LRRT2
NA
0%
0%
0%
NA
0%
0%
0%
0%
0%
0%
NA
0%
0%
0%
0%
0%
0%
0%
0%
NA
0%
0%
0%

-------
                                                                                2017 Draft Regulatory Impact Analysis
Table 3.8-19 Reference Fleet (Sales-Weighted) Technology Penetration in MY 2025

Aston Martin
BMW
Chrysler/Fiat
Daimler
Ferrari
Ford
Geely
General Motors
Honda
Hyundai
Kia
Lotus
Mazda
Mitsubishi
Nissan
Porsche
Spyker
Subaru
Suzuki
Tata
Tesla
Toyota
Volkswagen
Fleet
Mass Tech
Applied
-10%
-9%
-7%
-9%
-8%
-7%
-9%
-7%
-3%
-3%
-3%
-1%
-5%
-7%
-5%
-6%
-9%
-7%
-1%
-9%
0%
-2%
-8%
-5%
11
-9%
-8%
-7%
-8%
-8%
-7%
-8%
-7%
-3%
-3%
-3%
0%
-5%
-7%
-5%
-5%
-8%
-7%
-1%
-8%
0%
-2%
-7%
-5%
>1
jl
S£
1%
1%
0%
1%
1%
0%
1%
0%
0%
0%
0%
1%
0%
0%
0%
1%
1%
0%
0%
1%
0%
0%
1%
0%
TDS18
42%
51%
38%
48%
42%
53%
53%
40%
4%
27%
6%
54%
25%
68%
33%
48%
57%
52%
67%
48%
0%
3%
50%
30%
TDS24
15%
15%
11%
14%
15%
12%
15%
9%
0%
0%
0%
15%
11%
15%
7%
15%
15%
5%
15%
15%
0%
0%
15%
6%
TDS27
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
§
0%
28%
26%
15%
14%
29%
36%
37%
23%
24%
21%
0%
22%
16%
20%
19%
18%
6%
11%
32%
0%
25%
21%
26%
oo
S
0%
8%
12%
18%
0%
12%
12%
18%
5%
11%
8%
0%
7%
6%
8%
6%
4%
2%
4%
9%
0%
9%
6%
11%
\D
§
59%
36%
33%
40%
52%
29%
33%
21%
42%
32%
36%
10%
34%
40%
39%
20%
42%
39%
42%
38%
0%
36%
40%
33%
00
§
25%
19%
18%
22%
28%
16%
18%
11%
19%
18%
20%
5%
19%
22%
21%
11%
23%
21%
23%
21%
0%
9%
21%
15%
H
s
16%
9%
3%
5%
5%
6%
2%
3%
8%
6%
7%
85%
15%
6%
4%
44%
11%
23%
9%
0%
0%
5%
11%
6%
O
w
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
Pi
O
w
15%
15%
4%
14%
15%
10%
15%
6%
0%
0%
0%
15%
4%
15%
3%
15%
15%
3%
15%
15%
0%
0%
15%
5%
>
w
15%
15%
0%
15%
15%
2%
15%
0%
2%
0%
0%
15%
0%
2%
1%
15%
15%
0%
3%
15%
0%
12%
15%
5%
w
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
100%
0%
0%
0%
>
w
PH
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
C/3
C/3
57%
57%
0%
57%
57%
0%
57%
0%
0%
0%
0%
57%
0%
0%
0%
57%
57%
0%
0%
57%
0%
0%
57%
7%
LRRT2
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%

-------
Chapter 3
       3.8.3  Projected Technology Penetrations in Proposal case




Table 3.8-20  Proposal Car Technology Penetrations in MY 2021

Aston
Martin
BMW
Chrysler/Fiat
Daimler
Ferrari
Ford
Geely
General
Motors
Honda
Hyundai
Kia
Lotus
Mazda
Mitsubishi
Nissan
Porsche
Spyker
Subaru
Suzuki
Tata
Tesla
Toyota
Volkswagen
Fleet
Mass Tech
Applied
-16%
-12%
-8%
-13%
-13%
-8%
-13%
-7%
-4%
-5%
-3%
-3%
-5%
-9%
-5%
-8%
-15%
-8%
-1%
-15%
0%
-3%
-10%
-6%
0> 02
£ rt
F S
-7%
-8%
-8%
-8%
-7%
-8%
-9%
-7%
-4%
-5%
-3%
0%
-5%
-8%
-5%
-3%
-9%
-8%
0%
-10%
0%
-3%
-7%
-6%
«> £
J a
Sl
9%
4%
0%
5%
6%
0%
4%
0%
0%
0%
0%
3%
0%
0%
0%
5%
7%
0%
1%
6%
0%
0%
3%
1%
TDS18
0%
40%
89%
37%
0%
76%
43%
48%
15%
38%
20%
16%
80%
64%
65%
14%
27%
68%
46%
42%
0%
23%
34%
45%
TDS24
0%
26%
7%
21%
0%
15%
15%
9%
0%
0%
0%
30%
20%
30%
9%
25%
25%
30%
30%
7%
0%
0%
29%
10%
TDS27
15%
2%
1%
4%
15%
1%
8%
1%
0%
0%
0%
0%
0%
0%
0%
3%
3%
0%
0%
11%
0%
0%
1%
1%
\D
%
0%
0%
1%
0%
0%
5%
3%
6%
0%
4%
2%
0%
3%
0%
1%
0%
0%
0%
0%
0%
0%
1%
0%
2%
oo
S
0%
0%
2%
0%
0%
21%
11%
23%
0%
18%
7%
0%
12%
0%
3%
0%
0%
0%
0%
0%
0%
3%
0%
9%
£
O
Q
15%
18%
19%
17%
16%
13%
15%
13%
17%
14%
17%
9%
14%
18%
18%
12%
16%
15%
18%
18%
0%
15%
17%
15%
00
U
Q
61%
70%
76%
69%
65%
52%
61%
52%
68%
56%
66%
35%
54%
74%
73%
47%
64%
59%
73%
72%
0%
60%
67%
61%
H
s
7%
10%
3%
5%
2%
7%
2%
6%
12%
7%
9%
46%
17%
8%
5%
27%
8%
26%
9%
0%
0%
7%
10%
8%
O
w
50%
59%
60%
55%
50%
59%
55%
60%
58%
60%
60%
54%
60%
60%
59%
52%
53%
60%
60%
54%
0%
51%
57%
57%
&
O
w
15%
28%
3%
25%
15%
16%
22%
5%
0%
0%
0%
30%
7%
30%
10%
27%
27%
30%
30%
19%
0%
0%
29%
9%
w
30%
30%
0%
30%
30%
2%
30%
0%
3%
0%
0%
30%
0%
6%
1%
30%
30%
2%
24%
30%
0%
15%
30%
8%
>
w
16%
2%
0%
8%
16%
0%
8%
0%
0%
0%
0%
11%
0%
0%
0%
14%
12%
0%
0%
10%
100%
0%
6%
1%
>
w
PH
15%
0%
0%
0%
15%
0%
0%
0%
0%
0%
0%
13%
0%
0%
0%
15%
4%
0%
0%
0%
0%
0%
1%
0%
X
VI
15%
0%
0%
0%
15%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
4%
0%
0%
0%
0%
0%
0%
0%
0%
LRRT2
75%
75%
75%
75%
75%
74%
75%
75%
73%
75%
31%
75%
75%
75%
74%
75%
75%
75%
75%
75%
0%
17%
75%
62%

-------
                                                                              2017 Draft Regulatory Impact Analysis
Table 3.8-21 Proposal Truck Technology Penetrations in MY 2021

Aston
Martin
BMW
Chrysler/Fiat
Daimler
Ferrari
Ford
Geely
General
Motors
Honda
Hyundai
Kia
Lotus
Mazda
Mitsubishi
Nissan
Porsche
Spyker
Subaru
Suzuki
Tata
Tesla
Toyota
Volkswagen
Fleet
Mass Tech
Applied
NA
-14%
-7%
-15%
NA
-8%
-15%
-8%
-8%
-10%
-9%
NA
-14%
-15%
-5%
-15%
-4%
-15%
-12%
-11%
NA
-4%
-14%
-8%
11
NA
-12%
-7%
-13%
NA
-8%
-13%
-8%
-8%
-10%
-9%
NA
-14%
-14%
-5%
-13%
-2%
-15%
-12%
-9%
NA
-4%
-12%
-8%
>i
11
S£
NA
2%
0%
2%
NA
0%
2%
0%
0%
0%
0%
NA
0%
1%
0%
2%
2%
0%
0%
2%
NA
0%
2%
0%
TDS18
NA
73%
35%
80%
NA
75%
72%
33%
72%
70%
99%
NA
70%
66%
73%
55%
76%
70%
64%
66%
NA
66%
73%
59%
TDS24
NA
24%
25%
14%
NA
17%
27%
19%
0%
30%
0%
NA
30%
30%
24%
24%
19%
30%
30%
10%
NA
0%
23%
14%
TDS27
NA
3%
3%
6%
NA
6%
2%
5%
0%
0%
0%
NA
0%
0%
3%
8%
6%
0%
0%
10%
NA
3%
3%
4%
§
NA
20%
17%
20%
NA
14%
20%
19%
12%
18%
20%
NA
13%
13%
15%
20%
20%
5%
16%
13%
NA
16%
20%
16%
oo
S
NA
80%
69%
80%
NA
55%
80%
75%
50%
72%
79%
NA
51%
52%
61%
79%
80%
20%
63%
53%
NA
63%
80%
65%
\D
§
NA
0%
2%
0%
NA
5%
0%
1%
8%
2%
0%
NA
7%
7%
5%
0%
0%
13%
4%
6%
NA
3%
0%
3%
00
§
NA
0%
9%
0%
NA
21%
0%
5%
30%
8%
0%
NA
27%
28%
18%
0%
0%
54%
16%
24%
NA
11%
0%
13%
H
S
NA
0%
3%
0%
NA
3%
0%
0%
0%
0%
1%
NA
2%
0%
2%
1%
0%
8%
0%
0%
NA
3%
0%
2%
O
w
NA
60%
60%
60%
NA
59%
60%
60%
60%
60%
60%
NA
60%
60%
60%
60%
60%
60%
60%
58%
NA
57%
60%
59%
Pi
O
w
NA
27%
27%
20%
NA
23%
28%
24%
0%
30%
0%
NA
30%
30%
27%
32%
24%
30%
30%
20%
NA
3%
27%
17%
>
w
NA
30%
0%
30%
NA
2%
30%
0%
0%
0%
0%
NA
0%
4%
0%
30%
30%
0%
6%
30%
NA
5%
30%
4%
>
w
NA
0%
0%
0%
NA
0%
0%
0%
0%
0%
0%
NA
0%
0%
0%
0%
0%
0%
0%
4%
NA
0%
0%
0%
>
w
PH
NA
0%
0%
0%
NA
0%
0%
0%
0%
0%
0%
NA
0%
0%
0%
0%
0%
0%
0%
0%
NA
0%
0%
0%
C/3
C/3
NA
0%
0%
0%
NA
0%
35%
0%
0%
0%
0%
NA
0%
0%
0%
55%
70%
0%
0%
9%
NA
0%
0%
1%
LRRT2
NA
75%
75%
75%
NA
73%
75%
75%
75%
75%
75%
NA
75%
75%
75%
75%
75%
75%
75%
75%
NA
71%
75%
74%

-------
Chapter 3
Table 3.8-22 Proposal Fleet Technology Penetration in MY 2021

Aston
Martin
BMW
Chrysler/Fiat
Daimler
Ferrari
Ford
Geely
General
Motors
Honda
Hyundai
Kia
Lotus
Mazda
Mitsubishi
Nissan
Porsche
Spyker
Subaru
Suzuki
Tata
Tesla
Toyota
Volkswagen
Fleet
Mass Tech
Applied
-16%
-13%
-7%
-14%
-13%
-8%
-13%
-8%
-5%
-6%
-4%
-3%
-6%
-11%
-5%
-10%
-14%
-10%
-3%
-13%
0%
-3%
-11%
-7%
11
-7%
-9%
-7%
-10%
-7%
-8%
-10%
-8%
-5%
-6%
-4%
0%
-6%
-10%
-5%
-5%
-8%
-9%
-2%
-9%
0%
-3%
-8%
-6%
>i
11
S£
9%
4%
0%
4%
6%
0%
3%
0%
0%
0%
0%
3%
0%
1%
0%
4%
6%
0%
1%
4%
0%
0%
3%
0%
TDS18
0%
49%
64%
48%
0%
75%
52%
41%
33%
45%
37%
16%
78%
64%
68%
24%
34%
69%
49%
54%
0%
40%
42%
50%
TDS24
0%
25%
15%
20%
0%
16%
18%
14%
0%
6%
0%
30%
22%
30%
14%
25%
24%
30%
30%
9%
0%
0%
27%
11%
TDS27
15%
2%
2%
4%
15%
3%
6%
3%
0%
0%
0%
0%
0%
0%
1%
4%
3%
0%
0%
11%
0%
1%
1%
2%
§
0%
5%
8%
5%
0%
8%
8%
12%
4%
7%
6%
0%
5%
5%
5%
5%
3%
1%
3%
7%
0%
7%
4%
7%
oo
S
0%
21%
32%
20%
0%
32%
32%
49%
15%
29%
23%
0%
19%
18%
21%
19%
11%
5%
11%
26%
0%
26%
16%
28%
\D
§
15%
13%
11%
13%
16%
10%
10%
7%
14%
12%
13%
9%
12%
14%
14%
9%
14%
14%
16%
12%
0%
10%
13%
11%
00
§
61%
52%
45%
52%
65%
42%
42%
29%
56%
46%
51%
35%
50%
58%
56%
36%
55%
58%
63%
48%
0%
41%
54%
44%
H
s
7%
8%
3%
4%
2%
6%
1%
3%
8%
6%
7%
46%
14%
5%
4%
21%
7%
22%
7%
0%
0%
5%
8%
6%
O
w
50%
59%
60%
57%
50%
59%
57%
60%
59%
60%
60%
54%
60%
60%
60%
54%
54%
60%
60%
56%
0%
53%
57%
58%
Pi
O
w
15%
28%
14%
24%
15%
18%
24%
14%
0%
6%
0%
30%
11%
30%
15%
29%
27%
30%
30%
19%
0%
1%
29%
12%
>
w
30%
30%
0%
30%
30%
2%
30%
0%
2%
0%
0%
30%
0%
6%
1%
30%
30%
1%
21%
30%
0%
12%
30%
7%
>
w
16%
1%
0%
6%
16%
0%
6%
0%
0%
0%
0%
11%
0%
0%
0%
11%
10%
0%
0%
7%
100%
0%
5%
1%
>
w
PH
15%
0%
0%
0%
15%
0%
0%
0%
0%
0%
0%
13%
0%
0%
0%
11%
3%
0%
0%
0%
0%
0%
0%
0%
C/3
C/3
15%
0%
0%
0%
15%
0%
11%
0%
0%
0%
0%
0%
0%
0%
0%
16%
10%
0%
0%
4%
0%
0%
0%
0%
LRRT2
75%
75%
75%
75%
75%
74%
75%
75%
73%
75%
41%
75%
75%
75%
75%
75%
75%
75%
75%
75%
0%
38%
75%
66%
IACC2
80%
80%
80%
80%
80%
79%
80%
80%
78%
80%
44%
80%
80%
80%
80%
80%
80%
80%
80%
80%
0%
40%
80%
71%
CN
Pi
CM
w
50%
59%
60%
57%
50%
59%
57%
60%
59%
60%
60%
54%
60%
60%
60%
54%
54%
60%
60%
56%
0%
53%
57%
58%
Q
60%
99%
81%
91%
60%
94%
94%
58%
33%
51%
37%
89%
100%
100%
83%
86%
90%
100%
100%
93%
0%
41%
95%
65%
_1
C/3
Q
24%
0%
0%
3%
24%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
3%
0%
0%
0%
0%
0%
0%
0%
0%
                                                        5-46

-------
                                                                              2017 Draft Regulatory Impact Analysis
Table 3.8-23 Proposal Car Technology Penetrations in MY 2025

Aston
Martin
BMW
Chrysler/Fiat
Daimler
Ferrari
Ford
Geely
General
Motors
Honda
Hyundai
Kia
Lotus
Mazda
Mitsubishi
Nissan
Porsche
Spyker
Subaru
Suzuki
Tata
Tesla
Toyota
Volkswagen
Fleet
Mass Tech
Applied
-20%
-15%
-12%
-17%
-15%
-13%
-16%
-12%
-5%
-8%
-5%
-4%
-8%
-11%
-8%
-9%
-19%
-10%
-1%
-20%
0%
-6%
-11%
-9%
11
-6%
-9%
-9%
-10%
-6%
-11%
-9%
-11%
-5%
-8%
-5%
0%
-6%
-9%
-7%
-3%
-11%
-9%
0%
-11%
0%
-5%
-8%
-8%
>i
11
S£
14%
5%
2%
7%
9%
2%
7%
2%
0%
0%
0%
4%
2%
2%
1%
6%
8%
2%
1%
9%
0%
0%
4%
2%
TDS18
0%
0%
8%
1%
0%
15%
3%
11%
24%
25%
43%
0%
4%
4%
8%
0%
0%
5%
0%
0%
0%
20%
0%
14%
TDS24
0%
56%
70%
43%
0%
64%
31%
68%
73%
75%
57%
44%
73%
69%
73%
35%
47%
70%
66%
14%
0%
61%
60%
65%
TDS27
9%
5%
2%
9%
5%
4%
16%
2%
0%
0%
0%
0%
0%
0%
0%
1%
1%
0%
0%
26%
0%
1%
1%
2%
§
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
oo
s
0%
0%
2%
0%
0%
27%
13%
28%
0%
22%
7%
0%
14%
0%
4%
0%
0%
0%
0%
0%
0%
3%
0%
11%
\D
§
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
00
§
75%
81%
95%
80%
77%
62%
71%
65%
85%
71%
84%
36%
71%
88%
90%
53%
76%
73%
83%
85%
0%
75%
78%
75%
H
S
2%
9%
2%
5%
0%
6%
1%
4%
12%
7%
9%
44%
12%
6%
4%
24%
6%
22%
8%
0%
0%
7%
10%
7%
O
w
77%
90%
99%
85%
77%
95%
85%
97%
97%
100%
100%
80%
98%
94%
98%
78%
82%
95%
91%
85%
0%
84%
88%
93%
Pi
O
w
9%
61%
72%
52%
5%
68%
47%
70%
73%
75%
33%
44%
73%
69%
73%
36%
48%
70%
66%
41%
0%
62%
60%
66%
>
w
50%
28%
19%
32%
50%
13%
41%
15%
3%
0%
0%
25%
21%
21%
18%
29%
29%
20%
25%
44%
0%
17%
26%
15%
w
23%
10%
1%
15%
23%
4%
15%
3%
0%
0%
0%
20%
2%
6%
1%
22%
18%
5%
9%
15%
100%
0%
12%
4%
>
w
PH
18%
0%
0%
0%
22%
0%
0%
0%
0%
0%
0%
11%
0%
0%
0%
13%
5%
0%
0%
0%
0%
0%
1%
0%
C/3
C/3
0%
0%
0%
0%
5%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
LRRT2
100%
100%
100%
100%
100%
99%
100%
100%
97%
100%
100%
100%
100%
100%
99%
100%
100%
100%
100%
100%
0%
84%
100%
96%
IACC2
100%
100%
100%
100%
100%
99%
100%
100%
97%
100%
100%
100%
100%
100%
99%
100%
100%
100%
100%
100%
0%
84%
100%
96%
3
CM
w
77%
90%
99%
85%
77%
95%
85%
97%
97%
100%
100%
80%
98%
94%
98%
78%
82%
95%
91%
85%
0%
84%
88%
93%
Q
77%
90%
99%
84%
77%
95%
85%
97%
97%
100%
100%
80%
98%
94%
98%
78%
82%
95%
91%
85%
0%
84%
88%
93%
_1
C/3
Q
0%
0%
0%
1%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
                                                       5-47

-------
Chapter 3
Table 3.8-24 Proposal Truck Technology Penetrations in MY 2025

Aston
Martin
BMW
Chrysler/Fiat
Daimler
Ferrari
Ford
Geely
General
Motors
Honda
Hyundai
Kia
Lotus
Mazda
Mitsubishi
Nissan
Porsche
Spyker
Subaru
Suzuki
Tata
Tesla
Toyota
Volkswagen
Fleet
Mass Tech
Applied
NA
-18%
-13%
-20%
NA
-12%
-20%
-12%
-15%
-20%
-14%
NA
-20%
-20%
-9%
-20%
-5%
-20%
-16%
-14%
NA
-10%
-19%
-13%
11
NA
-11%
-12%
-12%
NA
-9%
-12%
-12%
-15%
-19%
-14%
NA
-18%
-18%
-6%
-12%
-2%
-17%
-16%
-7%
NA
-10%
-11%
-11%
>i
11
S£
NA
7%
1%
8%
NA
3%
8%
0%
0%
1%
0%
NA
1%
2%
3%
8%
3%
3%
0%
6%
NA
0%
7%
1%
TDS18
NA
30%
26%
46%
NA
27%
28%
31%
23%
23%
25%
NA
17%
16%
24%
39%
35%
6%
20%
33%
NA
27%
31%
27%
TDS24
NA
61%
61%
35%
NA
40%
66%
51%
75%
74%
75%
NA
74%
71%
59%
34%
46%
74%
72%
18%
NA
59%
58%
57%
TDS27
NA
10%
8%
19%
NA
20%
6%
15%
0%
0%
0%
NA
0%
0%
9%
28%
19%
0%
0%
33%
NA
8%
11%
11%
§
NA
0%
0%
0%
NA
0%
0%
0%
0%
0%
0%
NA
0%
0%
0%
0%
0%
0%
0%
0%
NA
0%
0%
0%
oo
S
NA
100%
86%
100%
NA
67%
100%
93%
62%
90%
99%
NA
66%
65%
77%
99%
100%
25%
79%
66%
NA
78%
100%
81%
\D
§
NA
0%
0%
0%
NA
0%
0%
0%
0%
0%
0%
NA
0%
0%
0%
0%
0%
0%
0%
0%
NA
0%
0%
0%
00
§
NA
0%
10%
0%
NA
27%
0%
6%
38%
9%
0%
NA
32%
31%
20%
0%
0%
69%
17%
26%
NA
14%
0%
15%
H
s
NA
0%
2%
0%
NA
2%
0%
0%
0%
0%
1%
NA
1%
0%
1%
1%
0%
6%
0%
0%
NA
2%
0%
1%
O
w
NA
100%
99%
100%
NA
96%
100%
99%
100%
99%
100%
NA
100%
96%
98%
100%
100%
99%
97%
92%
NA
94%
100%
98%
Pi
O
w
NA
70%
70%
54%
NA
60%
72%
66%
75%
74%
75%
NA
74%
71%
68%
61%
65%
74%
72%
50%
NA
67%
69%
67%
>
w
NA
50%
10%
50%
NA
31%
50%
3%
2%
2%
0%
NA
8%
9%
24%
50%
50%
19%
5%
41%
NA
7%
50%
13%
>
w
NA
0%
1%
0%
NA
3%
0%
1%
0%
1%
0%
NA
0%
4%
2%
0%
0%
1%
3%
8%
NA
0%
0%
1%
>
w
PH
NA
0%
0%
0%
NA
1%
0%
0%
0%
0%
0%
NA
0%
0%
0%
0%
0%
0%
0%
0%
NA
0%
0%
0%
C/3
C/3
NA
0%
0%
0%
NA
0%
0%
0%
0%
0%
0%
NA
0%
0%
0%
6%
19%
0%
0%
0%
NA
0%
0%
0%
LRRT2
NA
100%
100%
100%
NA
98%
100%
100%
100%
100%
100%
NA
100%
100%
100%
100%
100%
100%
100%
100%
NA
94%
100%
99%

-------
                                                                               2017 Draft Regulatory Impact Analysis
Table 3.8-25 Proposal Fleet Technology Penetration in MY 2025

Aston
Martin
BMW
Chrysler/Fiat
Daimler
Ferrari
Ford
Geely
General
Motors
Honda
Hyundai
Kia
Lotus
Mazda
Mitsubishi
Nissan
Porsche
Spyker
Subaru
Suzuki
Tata
Tesla
Toyota
Volkswagen
Fleet
Mass Tech
Applied
-20%
-15%
-12%
-18%
-15%
-13%
-17%
-12%
-8%
-10%
-7%
-4%
-10%
-14%
-8%
-11%
-17%
-13%
-4%
-17%
0%
-7%
-13%
-11%
11
-6%
-10%
-11%
-11%
-6%
-11%
-10%
-11%
-8%
-10%
-7%
0%
-8%
-12%
-7%
-5%
-10%
-10%
-3%
-9%
0%
-7%
-8%
-9%
>i
11
S£
14%
6%
2%
7%
9%
2%
7%
1%
0%
0%
0%
4%
2%
2%
1%
7%
7%
2%
1%
8%
0%
0%
5%
2%
TDS18
0%
8%
16%
11%
0%
18%
11%
21%
24%
25%
39%
0%
6%
8%
13%
8%
5%
6%
3%
15%
0%
23%
6%
19%
TDS24
0%
58%
66%
41%
0%
56%
42%
60%
73%
75%
61%
44%
73%
70%
69%
35%
47%
71%
67%
16%
0%
60%
60%
62%
TDS27
9%
6%
5%
11%
5%
9%
13%
8%
0%
0%
0%
0%
0%
0%
3%
7%
3%
0%
0%
29%
0%
4%
3%
5%
§
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
oo
s
0%
26%
38%
23%
0%
39%
39%
59%
18%
35%
27%
0%
23%
21%
25%
21%
13%
6%
14%
31%
0%
30%
20%
34%
\D
§
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
00
§
75%
59%
58%
62%
77%
51%
50%
37%
71%
58%
66%
36%
65%
69%
69%
42%
66%
72%
72%
58%
0%
53%
63%
55%
H
S
2%
7%
2%
4%
0%
4%
1%
2%
8%
6%
7%
44%
11%
4%
3%
19%
6%
19%
6%
0%
0%
5%
8%
5%
O
w
77%
92%
99%
88%
77%
95%
89%
98%
98%
100%
100%
80%
98%
95%
98%
83%
84%
96%
92%
88%
0%
88%
90%
94%
Pi
O
w
9%
64%
71%
53%
5%
65%
54%
68%
73%
75%
42%
44%
73%
70%
72%
41%
50%
71%
67%
45%
0%
64%
62%
66%
>
w
50%
34%
15%
36%
50%
19%
44%
10%
3%
0%
0%
25%
19%
17%
20%
34%
32%
19%
22%
43%
0%
13%
31%
15%
w
23%
8%
1%
12%
23%
4%
11%
2%
0%
0%
0%
20%
2%
5%
2%
17%
16%
4%
8%
12%
100%
0%
10%
3%
>
w
PH
18%
0%
0%
0%
22%
0%
0%
0%
0%
0%
0%
11%
0%
0%
0%
10%
5%
0%
0%
0%
0%
0%
1%
0%
C/3
C/3
0%
0%
0%
0%
5%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
1%
2%
0%
0%
0%
0%
0%
0%
0%
LRRT2
100%
100%
100%
100%
100%
99%
100%
100%
98%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
0%
88%
100%
97%
IACC2
100%
100%
100%
100%
100%
99%
100%
100%
98%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
0%
88%
100%
97%
3
CM
w
77%
92%
99%
88%
77%
95%
89%
98%
98%
100%
100%
80%
98%
95%
98%
83%
84%
96%
92%
88%
0%
88%
90%
94%
Q
77%
92%
99%
86%
77%
95%
89%
98%
98%
100%
100%
80%
98%
95%
98%
83%
84%
96%
92%
88%
0%
88%
90%
94%
_1
C/3
Q
0%
0%
0%
2%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
                                                        5-49

-------
Chapter 3
       3.8.4  Projected Technology Penetrations in Alternative Cases
Table 3.8-26  Alternative 1- (Trucks +20) Car Technology Penetrations in MY 2021

Aston Martin
BMW
Chrysler/Fiat
Daimler
Ferrari
Ford
Geely
General
Motors
Honda
Hyundai
Kia
Lotus
Mazda
Mitsubishi
Nissan
Porsche
Spyker
Subaru
Suzuki
Tata
Tesla
Toyota
Volkswagen
Fleet
Mass Tech
Applied
-16%
-12%
-7%
-13%
-13%
-8%
-13%
-7%
-2%
-4%
-2%
-3%
-4%
-7%
-5%
-8%
-15%
-7%
-1%
-15%
0%
-2%
-10%
-6%
1

w
30%
22%
0%
28%
30%
2%
26%
0%
3%
0%
0%
30%
0%
0%
1%
30%
30%
0%
13%
23%
0%
15%
30%
7%
w
16%
1%
0%
4%
16%
0%
5%
0%
0%
0%
0%
11%
0%
0%
0%
12%
11%
0%
0%
2%
100%
0%
3%
1%
w
PH
15%
0%
0%
0%
15%
0%
0%
0%
0%
0%
0%
13%
0%
0%
0%
15%
3%
0%
0%
0%
0%
0%
1%
0%
X
VI
15%
0%
0%
0%
15%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
LRRT2
75%
75%
75%
75%
75%
74%
75%
75%
25%
75%
6%
75%
75%
75%
74%
75%
75%
75%
75%
75%
0%
2%
75%
53%

-------
                                                                              2017 Draft Regulatory Impact Analysis
Table 3.8-27 Alternative 1- (Trucks +20) Truck Technology Penetrations in MY 2021

Aston
Martin
BMW
Chrysler/Fiat
Daimler
Ferrari
Ford
Geely
General
Motors
Honda
Hyundai
Kia
Lotus
Mazda
Mitsubishi
Nissan
Porsche
Spyker
Subaru
Suzuki
Tata
Tesla
Toyota
Volkswagen
Fleet
Mass Tech
Applied
NA
-14%
-7%
-15%
NA
-7%
-15%
-8%
-6%
-10%
-5%
NA
-10%
-11%
-5%
-15%
-4%
-11%
-12%
-11%
NA
-3%
-14%
-7%
11
NA
-13%
-7%
-13%
NA
-7%
-13%
-8%
-6%
-10%
-5%
NA
-10%
-11%
-5%
-13%
-2%
-11%
-12%
-9%
NA
-3%
-12%
-7%
>i
11
S£
NA
0%
0%
2%
NA
0%
2%
0%
0%
0%
0%
NA
0%
0%
0%
2%
2%
0%
0%
2%
NA
0%
2%
0%
TDS18
NA
73%
28%
80%
NA
40%
72%
40%
66%
100%
99%
NA
79%
70%
82%
78%
76%
70%
64%
66%
NA
66%
73%
57%
TDS24
NA
24%
3%
14%
NA
16%
27%
0%
0%
0%
0%
NA
21%
30%
15%
13%
19%
30%
30%
10%
NA
0%
23%
6%
TDS27
NA
3%
3%
6%
NA
6%
2%
0%
0%
0%
0%
NA
0%
0%
3%
8%
6%
0%
0%
10%
NA
0%
3%
2%
§
NA
20%
17%
20%
NA
14%
20%
19%
12%
18%
20%
NA
13%
13%
15%
20%
20%
5%
16%
13%
NA
16%
20%
16%
oo
S
NA
80%
69%
80%
NA
55%
80%
75%
50%
72%
79%
NA
51%
52%
61%
79%
80%
20%
63%
53%
NA
63%
80%
65%
\D
§
NA
0%
2%
0%
NA
5%
0%
1%
8%
2%
0%
NA
7%
7%
5%
0%
0%
13%
4%
6%
NA
6%
0%
4%
00
§
NA
0%
9%
0%
NA
21%
0%
5%
30%
8%
0%
NA
27%
28%
18%
0%
0%
54%
16%
24%
NA
7%
0%
12%
H
S
NA
1%
3%
0%
NA
3%
0%
0%
0%
0%
1%
NA
2%
0%
2%
1%
0%
8%
0%
0%
NA
3%
0%
2%
O
w
NA
60%
60%
60%
NA
59%
60%
60%
60%
60%
60%
NA
60%
60%
60%
60%
60%
60%
60%
58%
NA
57%
60%
59%
Pi
O
w
NA
27%
6%
20%
NA
22%
28%
0%
0%
0%
0%
NA
21%
30%
18%
22%
24%
30%
30%
20%
NA
0%
27%
7%
>
w
NA
6%
0%
30%
NA
2%
30%
0%
0%
0%
0%
NA
0%
0%
0%
30%
30%
0%
6%
30%
NA
5%
30%
4%
>
w
NA
0%
0%
0%
NA
0%
0%
0%
0%
0%
0%
NA
0%
0%
0%
0%
0%
0%
0%
4%
NA
0%
0%
0%
>
w
PH
NA
0%
0%
0%
NA
0%
0%
0%
0%
0%
0%
NA
0%
0%
0%
0%
0%
0%
0%
0%
NA
0%
0%
0%
C/3
C/3
NA
0%
0%
0%
NA
0%
0%
0%
0%
0%
0%
NA
0%
0%
0%
25%
46%
0%
0%
0%
NA
0%
0%
0%
LRRT2
NA
75%
75%
75%
NA
73%
75%
75%
75%
75%
75%
NA
75%
75%
75%
75%
75%
75%
75%
75%
NA
13%
75%
62%

-------
Chapter 3
Table 3.8-28 Alternative 1- (Trucks +20) Fleet Technology Penetration in MY 2021

Aston
Martin
BMW
Chrysler/Fiat
Daimler
Ferrari
Ford
Geely
General
Motors
Honda
Hyundai
Kia
Lotus
Mazda
Mitsubishi
Nissan
Porsche
Spyker
Subaru
Suzuki
Tata
Tesla
Toyota
Volkswagen
Fleet
Mass Tech
Applied
-16%
-12%
-7%
-13%
-13%
-7%
-13%
-7%
-4%
-5%
-3%
-3%
-5%
-8%
-5%
-9%
-14%
-8%
-3%
-13%
0%
-2%
-11%
-6%
11
-7%
-10%
-7%
-10%
-7%
-7%
-10%
-7%
-4%
-5%
-3%
0%
-5%
-8%
-5%
-5%
-8%
-8%
-2%
-10%
0%
-2%
-8%
-6%
>i
11
S£
9%
2%
0%
4%
6%
0%
3%
0%
0%
0%
0%
3%
0%
0%
0%
4%
6%
0%
1%
3%
0%
0%
3%
0%
TDS18
0%
55%
31%
52%
0%
43%
55%
26%
31%
47%
37%
16%
70%
73%
47%
31%
36%
89%
58%
61%
0%
40%
45%
40%
TDS24
0%
25%
1%
20%
0%
10%
18%
0%
0%
0%
0%
30%
19%
27%
5%
22%
24%
11%
30%
9%
0%
0%
27%
5%
TDS27
15%
2%
1%
4%
15%
3%
6%
0%
0%
0%
0%
0%
0%
0%
1%
4%
3%
0%
0%
11%
0%
0%
1%
1%
§
0%
5%
8%
5%
0%
8%
8%
12%
4%
7%
6%
0%
5%
5%
5%
5%
3%
1%
3%
7%
0%
7%
4%
7%
oo
S
0%
21%
32%
20%
0%
32%
32%
49%
15%
29%
23%
0%
19%
18%
21%
19%
11%
5%
11%
26%
0%
26%
16%
28%
\D
§
15%
13%
11%
13%
16%
10%
11%
7%
14%
12%
13%
9%
12%
14%
14%
9%
14%
14%
16%
13%
0%
39%
14%
17%
00
§
61%
51%
45%
53%
65%
42%
44%
29%
56%
46%
51%
35%
50%
57%
56%
36%
55%
57%
62%
52%
0%
12%
55%
39%
H
s
7%
9%
3%
6%
2%
6%
1%
3%
8%
6%
7%
46%
14%
6%
4%
22%
7%
22%
8%
0%
0%
5%
9%
6%
O
w
50%
59%
60%
58%
50%
59%
58%
60%
59%
60%
60%
54%
60%
60%
60%
55%
54%
60%
60%
58%
0%
53%
59%
58%
Pi
O
w
15%
28%
3%
24%
15%
13%
24%
0%
0%
0%
0%
30%
8%
27%
6%
26%
27%
11%
30%
19%
0%
0%
29%
6%
>
w
30%
18%
0%
28%
30%
2%
27%
0%
2%
0%
0%
30%
0%
0%
1%
30%
30%
0%
12%
26%
0%
12%
30%
6%
>
w
16%
1%
0%
3%
16%
0%
3%
0%
0%
0%
0%
11%
0%
0%
0%
9%
9%
0%
0%
3%
100%
0%
2%
0%
>
w
PH
15%
0%
0%
0%
15%
0%
0%
0%
0%
0%
0%
13%
0%
0%
0%
11%
2%
0%
0%
0%
0%
0%
0%
0%
C/3
C/3
15%
0%
0%
0%
15%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
6%
7%
0%
0%
0%
0%
0%
0%
0%
LRRT2
75%
75%
75%
75%
75%
74%
75%
75%
40%
75%
22%
75%
75%
75%
75%
75%
75%
75%
75%
75%
0%
6%
75%
56%
IACC2
80%
80%
80%
80%
80%
79%
80%
80%
43%
80%
23%
80%
80%
80%
80%
80%
80%
80%
80%
80%
0%
7%
80%
60%
CN
Pi
CM
w
50%
59%
60%
58%
50%
59%
58%
60%
59%
60%
60%
54%
60%
60%
60%
55%
54%
60%
60%
58%
0%
53%
59%
58%
Q
60%
99%
34%
93%
60%
55%
97%
26%
31%
47%
37%
89%
89%
100%
53%
91%
91%
100%
100%
97%
0%
40%
98%
49%
_1
C/3
Q
24%
0%
0%
3%
24%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
                                                        5-52

-------
                                                                               2017 Draft Regulatory Impact Analysis
Table 3.8-29 Alternative 2- (Trucks -20) Car Technology Penetrations in MY 2021

Aston
Martin
BMW
Chrysler/Fiat
Daimler
Ferrari
Ford
Geely
General
Motors
Honda
Hyundai
Kia
Lotus
Mazda
Mitsubishi
Nissan
Porsche
Spyker
Subaru
Suzuki
Tata
Tesla
Toyota
Volkswagen
Fleet
Mass Tech
Applied
-16%
-12%
-10%
-13%
-13%
-10%
-13%
-10%
-4%
-5%
-4%
-3%
-6%
-9%
-6%
-8%
-16%
-8%
-1%
-15%
0%
-4%
-10%
-7%
11
-7%
-8%
-9%
-8%
-7%
-10%
-8%
-10%
-4%
-5%
-4%
0%
-6%
-7%
-6%
-3%
-8%
-7%
0%
-9%
0%
-4%
-7%
-7%
>i
11
S£
9%
5%
0%
5%
6%
0%
5%
0%
0%
0%
0%
3%
0%
2%
0%
5%
7%
2%
1%
6%
0%
0%
4%
1%
TDS18
0%
37%
67%
33%
0%
68%
37%
68%
48%
53%
20%
16%
69%
45%
69%
11%
24%
53%
40%
38%
0%
23%
32%
50%
TDS24
0%
26%
28%
21%
0%
27%
15%
28%
0%
7%
0%
30%
30%
30%
30%
25%
25%
30%
30%
7%
0%
0%
29%
17%
TDS27
15%
2%
1%
4%
15%
1%
8%
1%
0%
0%
0%
0%
0%
0%
0%
3%
3%
0%
0%
11%
0%
0%
1%
1%
§
0%
0%
1%
0%
0%
5%
3%
6%
0%
4%
2%
0%
3%
0%
1%
0%
0%
0%
0%
0%
0%
1%
0%
2%
oo
S
0%
0%
2%
0%
0%
21%
11%
23%
0%
18%
7%
0%
12%
0%
3%
0%
0%
0%
0%
0%
0%
3%
0%
9%
\D
§
15%
17%
19%
17%
16%
13%
15%
13%
17%
14%
17%
9%
14%
19%
18%
12%
16%
15%
18%
18%
0%
15%
16%
15%
00
§
61%
68%
76%
69%
65%
53%
60%
52%
68%
56%
66%
35%
55%
75%
73%
46%
65%
61%
73%
71%
0%
60%
66%
61%
H
s
7%
10%
2%
5%
2%
7%
2%
5%
12%
7%
9%
46%
16%
6%
5%
26%
7%
23%
8%
0%
0%
7%
10%
8%
O
w
50%
57%
60%
55%
50%
59%
54%
60%
58%
60%
60%
54%
60%
60%
59%
50%
53%
60%
60%
53%
0%
51%
55%
57%
Pi
O
w
15%
28%
29%
25%
15%
28%
22%
29%
0%
7%
0%
30%
30%
30%
30%
27%
27%
30%
30%
19%
0%
0%
29%
18%
>
w
30%
30%
1%
30%
30%
4%
30%
1%
3%
0%
0%
30%
1%
25%
1%
30%
30%
17%
30%
30%
0%
15%
30%
9%
w
16%
6%
0%
8%
16%
0%
9%
0%
0%
0%
0%
11%
0%
0%
0%
16%
12%
0%
0%
11%
100%
0%
8%
1%
>
w
PH
15%
0%
0%
4%
15%
0%
6%
0%
0%
0%
0%
13%
0%
0%
0%
15%
7%
0%
0%
3%
0%
0%
1%
0%
C/3
C/3
15%
0%
0%
0%
15%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
14%
0%
0%
0%
0%
0%
0%
0%
0%
LRRT2
75%
75%
75%
75%
75%
74%
75%
75%
73%
75%
75%
75%
75%
75%
74%
75%
75%
75%
75%
75%
0%
63%
75%
72%

-------
Chapter 3
Table 3.8-30 Alternative 2- (Trucks -20) Truck Technology Penetrations in MY 2021

Aston
Martin
BMW
Chrysler/Fiat
Daimler
Ferrari
Ford
Geely
General
Motors
Honda
Hyundai
Kia
Lotus
Mazda
Mitsubishi
Nissan
Porsche
Spyker
Subaru
Suzuki
Tata
Tesla
Toyota
Volkswagen
Fleet
Mass Tech
Applied
NA
-14%
-9%
-15%
NA
-9%
-15%
-10%
-8%
-10%
-9%
NA
-15%
-15%
-6%
-15%
-4%
-15%
-12%
-11%
NA
-5%
-14%
-9%
11
NA
-12%
-9%
-13%
NA
-8%
-13%
-10%
-8%
-10%
-9%
NA
-15%
-13%
-6%
-13%
-2%
-15%
-12%
-9%
NA
-5%
-12%
-9%
>i
11
S£
NA
2%
0%
2%
NA
1%
2%
0%
0%
0%
0%
NA
0%
2%
0%
2%
2%
0%
0%
3%
NA
0%
2%
0%
TDS18
NA
73%
71%
80%
NA
66%
72%
51%
78%
70%
91%
NA
69%
58%
73%
48%
76%
70%
63%
59%
NA
73%
73%
66%
TDS24
NA
24%
25%
14%
NA
17%
27%
20%
22%
30%
9%
NA
30%
30%
24%
27%
19%
30%
30%
10%
NA
13%
23%
19%
TDS27
NA
3%
3%
6%
NA
6%
2%
5%
0%
0%
0%
NA
0%
0%
3%
8%
6%
0%
0%
10%
NA
3%
3%
4%
§
NA
20%
17%
20%
NA
14%
20%
19%
12%
18%
20%
NA
13%
13%
15%
20%
20%
5%
16%
13%
NA
16%
20%
16%
oo
S
NA
80%
69%
80%
NA
55%
80%
75%
50%
72%
79%
NA
51%
52%
61%
79%
80%
20%
63%
53%
NA
63%
80%
65%
\D
§
NA
0%
2%
0%
NA
5%
0%
1%
8%
2%
0%
NA
7%
7%
5%
0%
0%
13%
4%
6%
NA
3%
0%
3%
00
§
NA
0%
9%
0%
NA
22%
0%
5%
30%
8%
0%
NA
27%
27%
18%
0%
0%
54%
16%
23%
NA
11%
0%
13%
H
S
NA
0%
3%
0%
NA
2%
0%
0%
0%
0%
1%
NA
2%
0%
1%
1%
0%
8%
0%
0%
NA
3%
0%
1%
O
w
NA
60%
60%
60%
NA
59%
60%
60%
60%
60%
60%
NA
60%
59%
60%
60%
60%
60%
59%
57%
NA
57%
60%
59%
Pi
O
w
NA
27%
27%
20%
NA
23%
28%
25%
22%
30%
9%
NA
30%
30%
27%
36%
24%
30%
30%
20%
NA
15%
27%
23%
>
w
NA
30%
2%
30%
NA
16%
30%
2%
0%
0%
0%
NA
1%
11%
0%
30%
30%
0%
6%
30%
NA
5%
30%
7%
>
w
NA
0%
0%
0%
NA
0%
0%
0%
0%
0%
0%
NA
0%
1%
0%
0%
0%
0%
1%
6%
NA
0%
0%
0%
>
w
PH
NA
0%
0%
0%
NA
0%
0%
0%
0%
0%
0%
NA
0%
0%
0%
0%
0%
0%
0%
5%
NA
0%
0%
0%
C/3
C/3
NA
0%
0%
0%
NA
0%
35%
0%
0%
0%
0%
NA
0%
0%
0%
53%
70%
0%
0%
9%
NA
0%
31%
1%
LRRT2
NA
75%
75%
75%
NA
74%
75%
75%
75%
75%
75%
NA
75%
75%
75%
75%
75%
75%
75%
75%
NA
71%
75%
74%

-------
                                                                                2017 Draft Regulatory Impact Analysis
Table 3.8-31 Alternative 2- (Trucks -20) Fleet Technology Penetration in MY 2021

Aston
Martin
BMW
Chrysler/Fiat
Daimler
Ferrari
Ford
Geely
General
Motors
Honda
Hyundai
Kia
Lotus
Mazda
Mitsubishi
Nissan
Porsche
Spyker
Subaru
Suzuki
Tata
Tesla
Toyota
Volkswagen
Fleet
Mass Tech
Applied
-16%
-13%
-9%
-14%
-13%
-10%
-14%
-10%
-5%
-6%
-5%
-3%
-7%
-11%
-6%
-10%
-14%
-10%
-3%
-13%
0%
-5%
-11%
-8%
11
-7%
-9%
-9%
-9%
-7%
-9%
-10%
-10%
-5%
-6%
-5%
0%
-7%
-9%
-6%
-5%
-8%
-9%
-2%
-9%
0%
-5%
-8%
-7%
>i
11
S£
9%
4%
0%
5%
6%
1%
4%
0%
0%
0%
0%
3%
0%
2%
0%
5%
6%
1%
1%
4%
0%
0%
3%
1%
TDS18
0%
46%
69%
45%
0%
67%
48%
59%
58%
57%
35%
16%
69%
50%
70%
20%
31%
57%
44%
48%
0%
42%
41%
55%
TDS24
0%
25%
27%
20%
0%
23%
18%
24%
7%
12%
2%
30%
30%
30%
28%
25%
24%
30%
30%
9%
0%
5%
27%
18%
TDS27
15%
2%
2%
4%
15%
3%
6%
3%
0%
0%
0%
0%
0%
0%
1%
4%
3%
0%
0%
11%
0%
1%
1%
2%
§
0%
5%
8%
5%
0%
8%
8%
12%
4%
7%
6%
0%
5%
5%
5%
5%
3%
1%
3%
7%
0%
7%
4%
7%
oo
S
0%
21%
32%
20%
0%
32%
32%
49%
15%
29%
23%
0%
19%
18%
21%
19%
11%
5%
11%
26%
0%
26%
16%
28%
\D
§
15%
12%
11%
13%
16%
11%
10%
7%
14%
12%
13%
9%
13%
15%
14%
9%
14%
15%
16%
12%
0%
10%
13%
11%
00
§
61%
50%
45%
52%
65%
42%
41%
29%
56%
46%
51%
35%
50%
58%
56%
35%
55%
60%
63%
47%
0%
41%
52%
44%
H
s
7%
7%
3%
4%
2%
5%
1%
3%
8%
6%
7%
46%
13%
4%
4%
20%
6%
20%
7%
0%
0%
5%
8%
5%
O
w
50%
58%
60%
56%
50%
59%
56%
60%
59%
60%
60%
54%
60%
60%
60%
52%
54%
60%
60%
55%
0%
53%
56%
58%
Pi
O
w
15%
28%
28%
24%
15%
26%
24%
27%
7%
12%
2%
30%
30%
30%
29%
29%
27%
30%
30%
19%
0%
6%
29%
20%
>
w
30%
30%
2%
30%
30%
8%
30%
2%
2%
0%
0%
30%
1%
20%
1%
30%
30%
13%
26%
30%
0%
12%
30%
8%
>
w
16%
4%
0%
6%
16%
0%
6%
0%
0%
0%
0%
11%
0%
0%
0%
13%
10%
0%
0%
8%
100%
0%
6%
1%
>
w
PH
15%
0%
0%
3%
15%
0%
4%
0%
0%
0%
0%
13%
0%
0%
0%
11%
6%
0%
0%
4%
0%
0%
0%
0%
C/3
C/3
15%
0%
0%
0%
15%
0%
11%
0%
0%
0%
0%
0%
0%
0%
0%
23%
10%
0%
0%
4%
0%
0%
6%
1%
LRRT2
75%
75%
75%
75%
75%
74%
75%
75%
73%
75%
75%
75%
75%
75%
75%
75%
75%
75%
75%
75%
0%
66%
75%
73%
IACC2
80%
80%
80%
80%
80%
79%
80%
80%
78%
80%
80%
80%
80%
80%
80%
80%
80%
80%
80%
80%
0%
71%
80%
78%
3
CM
w
50%
58%
60%
56%
50%
59%
56%
60%
59%
60%
60%
54%
60%
60%
60%
52%
54%
60%
60%
55%
0%
53%
56%
58%
Q
60%
96%
99%
90%
60%
98%
94%
88%
64%
68%
38%
89%
100%
100%
99%
84%
90%
100%
100%
92%
0%
49%
94%
79%
_1
C/3
Q
24%
0%
0%
3%
24%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
4%
0%
0%
0%
0%
0%
0%
0%
0%
                                                        5-55

-------
Chapter 3
Table 3.8-32 Alternative 3- (Cars +20) Car Technology Penetrations in MY 2021

Aston
Martin
BMW
Chrysler/Fiat
Daimler
Ferrari
Ford
Geely
General
Motors
Honda
Hyundai
Kia
Lotus
Mazda
Mitsubishi
Nissan
Porsche
Spyker
Subaru
Suzuki
Tata
Tesla
Toyota
Volkswagen
Fleet
Mass Tech
Applied
-16%
-12%
-7%
-13%
-13%
-8%
-13%
-7%
-2%
-3%
-2%
-2%
-4%
-6%
-5%
-7%
-15%
-6%
0%
-15%
0%
-2%
-9%
-6%
11
-7%
-11%
-7%
-10%
-7%
-8%
-9%
-7%
-2%
-3%
-2%
0%
-4%
-6%
-5%
-3%
-10%
-6%
0%
-11%
0%
-2%
-7%
-5%
>i
11
S£
9%
1%
0%
2%
6%
0%
3%
0%
0%
0%
0%
2%
0%
0%
0%
4%
5%
0%
0%
4%
0%
0%
2%
0%
TDS18
0%
67%
33%
57%
0%
51%
49%
12%
15%
33%
12%
27%
19%
68%
32%
23%
36%
68%
100%
57%
0%
19%
58%
32%
TDS24
0%
26%
0%
21%
0%
0%
15%
0%
0%
0%
0%
30%
1%
15%
0%
25%
25%
3%
0%
7%
0%
0%
29%
4%
TDS27
15%
2%
1%
4%
15%
0%
8%
0%
0%
0%
0%
0%
0%
0%
0%
3%
3%
0%
0%
11%
0%
0%
1%
0%
§
0%
0%
1%
0%
0%
5%
3%
6%
0%
4%
2%
0%
3%
0%
1%
0%
0%
0%
0%
0%
0%
1%
0%
2%
oo
S
0%
0%
2%
0%
0%
21%
11%
23%
0%
18%
7%
0%
12%
0%
3%
0%
0%
0%
0%
0%
0%
3%
0%
9%
\D
§
15%
17%
19%
18%
16%
13%
16%
13%
38%
20%
83%
7%
14%
18%
18%
11%
17%
15%
18%
20%
0%
73%
17%
31%
00
§
61%
68%
76%
71%
65%
52%
65%
52%
47%
50%
0%
26%
54%
73%
73%
43%
67%
59%
71%
79%
0%
2%
68%
46%
H
s
7%
14%
3%
8%
2%
7%
2%
6%
12%
7%
9%
57%
17%
9%
5%
34%
9%
27%
12%
0%
0%
7%
15%
8%
O
w
50%
60%
60%
58%
50%
59%
58%
60%
58%
60%
60%
54%
60%
60%
59%
53%
56%
60%
60%
59%
0%
51%
60%
58%
Pi
O
w
15%
28%
1%
25%
15%
0%
22%
0%
0%
0%
0%
30%
1%
0%
0%
27%
27%
3%
0%
19%
0%
0%
29%
4%
>
w
30%
4%
0%
14%
30%
2%
26%
0%
3%
0%
0%
30%
0%
0%
1%
30%
30%
0%
0%
23%
0%
15%
12%
5%
>
w
16%
1%
0%
3%
16%
0%
3%
0%
0%
0%
0%
11%
0%
0%
0%
12%
6%
0%
0%
2%
100%
0%
1%
1%
>
w
PH
15%
0%
0%
0%
15%
0%
0%
0%
0%
0%
0%
3%
0%
0%
0%
8%
0%
0%
0%
0%
0%
0%
0%
0%
C/3
C/3
15%
0%
0%
0%
15%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
LRRT2
75%
75%
75%
75%
75%
74%
75%
75%
0%
0%
0%
75%
75%
75%
45%
75%
75%
75%
75%
75%
0%
0%
75%
43%
IACC2
80%
80%
80%
80%
80%
79%
80%
80%
0%
0%
0%
80%
80%
80%
48%
80%
80%
80%
80%
80%
0%
0%
80%
46%
CN
Pi
CM
w
50%
60%
60%
58%
50%
59%
58%
60%
58%
60%
60%
54%
60%
60%
59%
53%
56%
60%
60%
59%
0%
51%
60%
58%
Q
60%
99%
33%
96%
60%
50%
97%
13%
15%
33%
12%
89%
21%
83%
32%
88%
94%
71%
100%
98%
0%
24%
99%
38%
_1
C/3
Q
24%
0%
0%
1%
24%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
                                                       5-56

-------
                                                                               2017 Draft Regulatory Impact Analysis
Table 3.8-33 Alternative 3- (Cars +20) Truck Technology Penetrations in MY 2021

Aston
Martin
BMW
Chrysler/Fiat
Daimler
Ferrari
Ford
Geely
General
Motors
Honda
Hyundai
Kia
Lotus
Mazda
Mitsubishi
Nissan
Porsche
Spyker
Subaru
Suzuki
Tata
Tesla
Toyota
Volkswagen
Fleet
Mass Tech
Applied
NA
-13%
-7%
-15%
NA
-7%
-15%
-8%
-4%
-5%
-4%
NA
-10%
-10%
-4%
-15%
-4%
-10%
-8%
-11%
NA
-3%
-14%
-6%
11
NA
-13%
-7%
-14%
NA
-7%
-15%
-8%
-4%
-5%
-4%
NA
-10%
-10%
-4%
-13%
-2%
-10%
-8%
-9%
NA
-3%
-12%
-6%
>i
11
S£
NA
0%
0%
1%
NA
0%
0%
0%
0%
0%
0%
NA
0%
0%
0%
2%
2%
0%
0%
2%
NA
0%
2%
0%
TDS18
NA
73%
28%
80%
NA
44%
72%
42%
66%
100%
99%
NA
53%
81%
59%
78%
76%
23%
70%
66%
NA
66%
73%
55%
TDS24
NA
24%
3%
14%
NA
0%
27%
0%
0%
0%
0%
NA
19%
19%
0%
13%
19%
10%
30%
10%
NA
0%
23%
2%
TDS27
NA
3%
1%
6%
NA
4%
2%
0%
0%
0%
0%
NA
0%
0%
3%
8%
6%
0%
0%
10%
NA
0%
3%
1%
§
NA
20%
17%
20%
NA
14%
20%
19%
12%
18%
20%
NA
13%
13%
15%
20%
20%
5%
16%
13%
NA
16%
20%
16%
oo
S
NA
80%
69%
80%
NA
55%
80%
75%
50%
72%
79%
NA
51%
52%
61%
79%
80%
20%
63%
53%
NA
63%
80%
65%
\D
§
NA
0%
2%
0%
NA
5%
0%
1%
8%
2%
0%
NA
7%
7%
5%
0%
0%
13%
4%
6%
NA
11%
0%
5%
00
§
NA
0%
9%
0%
NA
21%
0%
5%
30%
8%
0%
NA
27%
28%
18%
0%
0%
54%
16%
24%
NA
2%
0%
11%
H
S
NA
1%
3%
0%
NA
3%
0%
0%
0%
0%
1%
NA
2%
0%
2%
1%
0%
8%
0%
0%
NA
3%
0%
2%
O
w
NA
60%
60%
60%
NA
59%
60%
60%
60%
60%
60%
NA
60%
60%
60%
60%
60%
60%
60%
58%
NA
57%
60%
59%
Pi
O
w
NA
27%
4%
20%
NA
4%
28%
0%
0%
0%
0%
NA
19%
19%
3%
22%
24%
10%
30%
20%
NA
0%
27%
4%
>
w
NA
0%
0%
16%
NA
2%
3%
0%
0%
0%
0%
NA
0%
0%
0%
30%
30%
0%
0%
30%
NA
5%
30%
3%
>
w
NA
0%
0%
0%
NA
0%
0%
0%
0%
0%
0%
NA
0%
0%
0%
0%
0%
0%
0%
4%
NA
0%
0%
0%
>
w
PH
NA
0%
0%
0%
NA
0%
0%
0%
0%
0%
0%
NA
0%
0%
0%
0%
0%
0%
0%
0%
NA
0%
0%
0%
C/3
C/3
NA
0%
0%
0%
NA
0%
0%
0%
0%
0%
0%
NA
0%
0%
0%
25%
0%
0%
0%
0%
NA
0%
0%
0%
LRRT2
NA
75%
75%
75%
NA
73%
75%
75%
26%
15%
0%
NA
75%
75%
75%
75%
75%
75%
75%
75%
NA
13%
75%
54%

-------
Chapter 3
Table 3.8-34 Alternative 3- (Cars +20) Fleet Technology Penetration in MY 2021

Aston
Martin
BMW
Chrysler/Fiat
Daimler
Ferrari
Ford
Geely
General
Motors
Honda
Hyundai
Kia
Lotus
Mazda
Mitsubishi
Nissan
Porsche
Spyker
Subaru
Suzuki
Tata
Tesla
Toyota
Volkswagen
Fleet
Mass Tech
Applied
-16%
-12%
-7%
-13%
-13%
-7%
-13%
-8%
-3%
-3%
-3%
-2%
-5%
-8%
-5%
-9%
-13%
-7%
-1%
-13%
0%
-2%
-10%
-6%
11
-7%
-12%
-7%
-11%
-7%
-7%
-11%
-8%
-3%
-3%
-3%
0%
-5%
-8%
-5%
-6%
-9%
-7%
-1%
-10%
0%
-2%
-8%
-6%
>i
11
S£
9%
0%
0%
2%
6%
0%
2%
0%
0%
0%
0%
2%
0%
0%
0%
3%
5%
0%
0%
3%
0%
0%
2%
0%
TDS18
0%
69%
31%
63%
0%
49%
56%
27%
31%
47%
31%
27%
25%
73%
40%
36%
42%
57%
95%
61%
0%
37%
61%
40%
TDS24
0%
25%
1%
20%
0%
0%
18%
0%
0%
0%
0%
30%
5%
17%
0%
22%
24%
5%
5%
9%
0%
0%
27%
3%
TDS27
15%
2%
1%
4%
15%
2%
6%
0%
0%
0%
0%
0%
0%
0%
1%
4%
3%
0%
0%
11%
0%
0%
1%
1%
§
0%
5%
8%
5%
0%
8%
8%
12%
4%
7%
6%
0%
5%
5%
5%
5%
3%
1%
3%
7%
0%
7%
4%
7%
oo
S
0%
21%
32%
20%
0%
32%
32%
49%
15%
29%
23%
0%
19%
18%
21%
19%
11%
5%
11%
26%
0%
26%
16%
28%
\D
§
15%
13%
11%
13%
16%
10%
11%
7%
29%
16%
64%
7%
12%
14%
14%
8%
14%
14%
15%
13%
0%
49%
14%
22%
00
§
61%
50%
45%
53%
65%
42%
45%
29%
42%
42%
0%
26%
50%
57%
56%
33%
58%
57%
61%
52%
0%
2%
54%
34%
H
s
7%
10%
3%
6%
2%
6%
2%
3%
8%
6%
7%
57%
14%
6%
4%
26%
8%
22%
10%
0%
0%
5%
12%
6%
O
w
50%
60%
60%
59%
50%
59%
59%
60%
59%
60%
60%
54%
60%
60%
60%
55%
57%
60%
60%
58%
0%
53%
60%
58%
Pi
O
w
15%
28%
2%
24%
15%
2%
24%
0%
0%
0%
0%
30%
5%
7%
1%
26%
27%
5%
5%
19%
0%
0%
29%
4%
>
w
30%
3%
0%
15%
30%
2%
19%
0%
2%
0%
0%
30%
0%
0%
1%
30%
30%
0%
0%
26%
0%
12%
16%
4%
>
w
16%
0%
0%
2%
16%
0%
2%
0%
0%
0%
0%
11%
0%
0%
0%
9%
5%
0%
0%
3%
100%
0%
0%
0%
>
w
PH
15%
0%
0%
0%
15%
0%
0%
0%
0%
0%
0%
3%
0%
0%
0%
6%
0%
0%
0%
0%
0%
0%
0%
0%
C/3
C/3
15%
0%
0%
0%
15%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
6%
0%
0%
0%
0%
0%
0%
0%
0%
LRRT2
75%
75%
75%
75%
75%
74%
75%
75%
8%
3%
0%
75%
75%
75%
54%
75%
75%
75%
75%
75%
0%
5%
75%
47%
IACC2
80%
80%
80%
80%
80%
79%
80%
80%
9%
3%
0%
80%
80%
80%
58%
80%
80%
80%
80%
80%
0%
6%
80%
50%
CN
Pi
CM
w
50%
60%
60%
59%
50%
59%
59%
60%
59%
60%
60%
54%
60%
60%
60%
55%
57%
60%
60%
58%
0%
53%
60%
58%
Q
60%
100%
33%
94%
60%
50%
98%
27%
31%
47%
31%
89%
30%
89%
41%
91%
95%
62%
100%
97%
0%
40%
100%
46%
_1
C/3
Q
24%
0%
0%
3%
24%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
                                                        5-58

-------
                                                                               2017 Draft Regulatory Impact Analysis
Table 3.8-35 Alternative 4- (Cars -20) Car Technology Penetrations in MY 2021

Aston
Martin
BMW
Chrysler/Fiat
Daimler
Ferrari
Ford
Geely
General
Motors
Honda
Hyundai
Kia
Lotus
Mazda
Mitsubishi
Nissan
Porsche
Spyker
Subaru
Suzuki
Tata
Tesla
Toyota
Volkswagen
Fleet
Mass Tech
Applied
-16%
-12%
-10%
-14%
-13%
-11%
-14%
-10%
-4%
-6%
-4%
-4%
-7%
-9%
-7%
-8%
-16%
-9%
-2%
-15%
0%
-4%
-11%
-8%
11
-7%
-7%
-9%
-7%
-7%
-10%
-8%
-10%
-4%
-6%
-4%
0%
-5%
-6%
-6%
-3%
-7%
-6%
0%
-9%
0%
-4%
-6%
-7%
>i
11
S£
9%
5%
0%
6%
6%
1%
6%
0%
0%
0%
0%
4%
2%
3%
1%
5%
9%
2%
2%
6%
0%
0%
4%
1%
TDS18
0%
30%
67%
26%
0%
58%
31%
68%
70%
67%
69%
0%
43%
37%
56%
0%
14%
42%
33%
37%
0%
28%
25%
51%
TDS24
0%
26%
28%
21%
0%
27%
15%
28%
14%
30%
12%
15%
30%
30%
30%
12%
25%
30%
30%
7%
0%
1%
29%
21%
TDS27
15%
2%
1%
4%
15%
1%
8%
1%
0%
0%
0%
15%
0%
0%
0%
15%
3%
0%
0%
11%
0%
0%
1%
1%
§
0%
0%
1%
0%
0%
5%
3%
6%
0%
4%
2%
0%
3%
0%
1%
0%
0%
0%
0%
0%
0%
1%
0%
2%
oo
S
0%
0%
2%
0%
0%
21%
11%
23%
0%
18%
7%
0%
12%
0%
3%
0%
0%
0%
0%
0%
0%
3%
0%
9%
\D
§
15%
16%
19%
17%
16%
13%
15%
13%
17%
14%
17%
9%
14%
18%
18%
12%
16%
16%
17%
18%
0%
15%
16%
15%
00
§
61%
65%
76%
66%
65%
53%
59%
52%
68%
57%
66%
36%
58%
73%
73%
46%
64%
64%
69%
71%
0%
60%
64%
61%
H
s
7%
9%
2%
5%
2%
6%
1%
6%
12%
7%
9%
39%
11%
6%
4%
26%
6%
18%
7%
0%
0%
7%
9%
7%
O
w
50%
54%
60%
53%
50%
59%
53%
60%
58%
60%
60%
50%
59%
58%
60%
50%
52%
59%
56%
53%
0%
51%
53%
57%
Pi
O
w
15%
28%
29%
25%
15%
28%
22%
29%
5%
30%
2%
30%
30%
30%
30%
27%
27%
30%
30%
19%
0%
1%
29%
20%
>
w
30%
30%
1%
30%
30%
14%
30%
0%
3%
3%
0%
30%
26%
30%
14%
30%
30%
27%
30%
30%
0%
15%
30%
12%
w
16%
10%
0%
12%
16%
0%
11%
0%
0%
0%
0%
16%
1%
3%
0%
16%
13%
1%
7%
11%
100%
0%
11%
2%
>
w
PH
15%
1%
0%
6%
15%
0%
10%
0%
0%
0%
0%
15%
0%
0%
0%
15%
15%
0%
0%
3%
0%
0%
5%
1%
C/3
C/3
15%
0%
0%
0%
15%
0%
0%
0%
0%
0%
0%
30%
0%
0%
0%
27%
4%
0%
0%
0%
0%
0%
1%
0%
LRRT2
75%
75%
75%
75%
75%
74%
75%
75%
73%
75%
75%
75%
75%
75%
75%
75%
75%
75%
75%
75%
0%
63%
75%
72%

-------
Chapter 3
Table 3.8-36 Alternative 4- (Cars -20) Truck Technology Penetrations in MY 2021

Aston
Martin
BMW
Chrysler/Fiat
Daimler
Ferrari
Ford
Geely
General
Motors
Honda
Hyundai
Kia
Lotus
Mazda
Mitsubishi
Nissan
Porsche
Spyker
Subaru
Suzuki
Tata
Tesla
Toyota
Volkswagen
Fleet
Mass Tech
Applied
NA
-14%
-9%
-15%
NA
-10%
-15%
-10%
-9%
-15%
-9%
NA
-15%
-15%
-7%
-15%
-4%
-15%
-12%
-11%
NA
-5%
-14%
-9%
11
NA
-12%
-9%
-13%
NA
-8%
-13%
-10%
-9%
-15%
-9%
NA
-13%
-13%
-6%
-13%
-2%
-12%
-12%
-9%
NA
-5%
-12%
-9%
>i
11
S£
NA
2%
0%
2%
NA
2%
2%
0%
0%
0%
0%
NA
2%
2%
1%
2%
2%
3%
0%
3%
NA
0%
2%
1%
TDS18
NA
73%
71%
80%
NA
65%
72%
47%
70%
67%
70%
NA
59%
57%
66%
30%
76%
47%
61%
59%
NA
71%
61%
62%
TDS24
NA
24%
25%
14%
NA
17%
27%
20%
30%
30%
30%
NA
30%
30%
24%
25%
19%
30%
30%
10%
NA
21%
29%
22%
TDS27
NA
3%
3%
6%
NA
6%
2%
5%
0%
0%
0%
NA
0%
0%
3%
15%
6%
0%
0%
10%
NA
3%
3%
4%
§
NA
20%
17%
20%
NA
14%
20%
19%
12%
18%
20%
NA
13%
13%
15%
20%
20%
5%
16%
13%
NA
16%
20%
16%
oo
S
NA
80%
69%
80%
NA
55%
80%
75%
50%
72%
79%
NA
51%
52%
61%
79%
80%
20%
63%
53%
NA
63%
80%
65%
\D
§
NA
0%
2%
0%
NA
5%
0%
1%
8%
2%
0%
NA
7%
7%
4%
0%
0%
14%
4%
6%
NA
3%
0%
3%
00
§
NA
0%
9%
0%
NA
22%
0%
5%
30%
8%
0%
NA
27%
27%
18%
0%
0%
56%
15%
23%
NA
11%
0%
13%
H
S
NA
0%
3%
0%
NA
2%
0%
0%
0%
0%
1%
NA
1%
0%
1%
1%
0%
6%
0%
0%
NA
3%
0%
1%
O
w
NA
60%
60%
60%
NA
59%
60%
60%
60%
60%
60%
NA
60%
59%
60%
60%
60%
60%
59%
57%
NA
57%
60%
59%
Pi
O
w
NA
27%
27%
20%
NA
23%
28%
25%
30%
30%
30%
NA
30%
30%
27%
40%
24%
30%
30%
20%
NA
24%
32%
26%
>
w
NA
30%
2%
30%
NA
25%
30%
2%
0%
3%
0%
NA
11%
11%
13%
30%
30%
23%
6%
30%
NA
5%
30%
9%
>
w
NA
0%
0%
0%
NA
1%
0%
0%
0%
0%
0%
NA
0%
2%
0%
0%
0%
0%
2%
6%
NA
0%
0%
0%
>
w
PH
NA
0%
0%
0%
NA
0%
0%
0%
0%
0%
0%
NA
0%
0%
0%
0%
0%
0%
0%
5%
NA
0%
0%
0%
C/3
C/3
NA
0%
0%
19%
NA
0%
35%
0%
0%
0%
0%
NA
0%
0%
0%
40%
70%
0%
0%
9%
NA
0%
63%
2%
LRRT2
NA
75%
75%
75%
NA
74%
75%
75%
75%
75%
75%
NA
75%
75%
75%
75%
75%
75%
75%
75%
NA
71%
75%
74%

-------
                                                                                2017 Draft Regulatory Impact Analysis
Table 3.8-37 Alternative 4- (Cars -20) Fleet Technology Penetration in MY 2021

Aston
Martin
BMW
Chrysler/Fiat
Daimler
Ferrari
Ford
Geely
General
Motors
Honda
Hyundai
Kia
Lotus
Mazda
Mitsubishi
Nissan
Porsche
Spyker
Subaru
Suzuki
Tata
Tesla
Toyota
Volkswagen
Fleet
Mass Tech
Applied
-16%
-13%
-9%
-14%
-13%
-10%
-14%
-10%
-6%
-8%
-5%
-4%
-8%
-11%
-7%
-10%
-14%
-10%
-3%
-13%
0%
-5%
-11%
-8%
11
-7%
-8%
-9%
-9%
-7%
-9%
-9%
-10%
-6%
-8%
-5%
0%
-6%
-9%
-6%
-5%
-7%
-8%
-2%
-9%
0%
-5%
-7%
-7%
>i
11
S£
9%
4%
0%
5%
6%
1%
5%
0%
0%
0%
0%
4%
2%
3%
1%
5%
8%
3%
1%
5%
0%
0%
4%
1%
TDS18
0%
42%
69%
40%
0%
60%
43%
58%
70%
67%
69%
0%
46%
44%
59%
7%
23%
43%
38%
48%
0%
45%
32%
55%
TDS24
0%
25%
27%
20%
0%
23%
18%
24%
19%
30%
16%
15%
30%
30%
28%
15%
24%
30%
30%
9%
0%
9%
29%
21%
TDS27
15%
2%
2%
4%
15%
3%
6%
3%
0%
0%
0%
15%
0%
0%
1%
15%
3%
0%
0%
11%
0%
1%
1%
2%
§
0%
5%
8%
5%
0%
8%
8%
12%
4%
7%
6%
0%
5%
5%
5%
5%
3%
1%
3%
7%
0%
7%
4%
7%
oo
S
0%
21%
32%
20%
0%
32%
32%
49%
15%
29%
23%
0%
19%
18%
21%
19%
11%
5%
11%
26%
0%
26%
16%
28%
\D
§
15%
12%
11%
12%
16%
11%
10%
7%
14%
12%
13%
9%
13%
14%
14%
9%
14%
16%
15%
12%
0%
10%
13%
11%
00
§
61%
48%
45%
50%
65%
42%
41%
29%
56%
47%
51%
36%
52%
57%
56%
35%
55%
62%
59%
47%
0%
41%
51%
44%
H
s
7%
6%
3%
3%
2%
5%
1%
3%
8%
5%
7%
39%
10%
4%
3%
20%
5%
15%
6%
0%
0%
5%
7%
5%
O
w
50%
55%
60%
54%
50%
59%
55%
60%
59%
60%
60%
50%
59%
59%
60%
52%
53%
59%
56%
55%
0%
53%
55%
58%
Pi
O
w
15%
28%
28%
24%
15%
26%
24%
27%
12%
30%
9%
30%
30%
30%
29%
30%
27%
30%
30%
19%
0%
10%
30%
22%
>
w
30%
30%
2%
30%
30%
18%
30%
1%
2%
3%
0%
30%
23%
23%
14%
30%
30%
26%
26%
30%
0%
12%
30%
11%
>
w
16%
8%
0%
9%
16%
0%
8%
0%
0%
0%
0%
16%
1%
2%
0%
13%
12%
1%
6%
8%
100%
0%
9%
1%
>
w
PH
15%
1%
0%
5%
15%
0%
7%
0%
0%
0%
0%
15%
0%
0%
0%
11%
13%
0%
0%
4%
0%
0%
4%
0%
C/3
C/3
15%
0%
0%
5%
15%
0%
11%
0%
0%
0%
0%
30%
0%
0%
0%
30%
13%
0%
0%
4%
0%
0%
14%
1%
LRRT2
75%
75%
75%
75%
75%
74%
75%
75%
73%
75%
75%
75%
75%
75%
75%
75%
75%
75%
75%
75%
0%
66%
75%
73%
IACC2
80%
80%
80%
80%
80%
79%
80%
80%
78%
80%
80%
80%
80%
80%
80%
80%
80%
80%
80%
80%
0%
71%
80%
78%
CN
Pi
CM
w
50%
55%
60%
54%
50%
59%
55%
60%
59%
60%
60%
50%
59%
59%
60%
52%
53%
59%
56%
55%
0%
53%
55%
58%
Q
60%
92%
98%
88%
60%
98%
92%
85%
89%
100%
85%
75%
99%
98%
99%
72%
88%
99%
94%
92%
0%
55%
90%
85%
_1
C/3
Q
24%
0%
0%
3%
24%
0%
0%
0%
0%
0%
0%
9%
0%
0%
0%
16%
0%
0%
0%
0%
0%
0%
1%
0%
                                                        5-61

-------
Chapter 3
Table 3.8-38 Alternative 1- (Trucks +20) Car Technology Penetrations in MY 2025

Aston
Martin
BMW
Chrysler/Fiat
Daimler
Ferrari
Ford
Geely
General
Motors
Honda
Hyundai
Kia
Lotus
Mazda
Mitsubishi
Nissan
Porsche
Spyker
Subaru
Suzuki
Tata
Tesla
Toyota
Volkswagen
Fleet
Mass Tech
Applied
-20%
-14%
-10%
-17%
-15%
-12%
-16%
-10%
-4%
-5%
-4%
-4%
-8%
-11%
-8%
-9%
-19%
-10%
-1%
-20%
0%
-4%
-11%
-8%
11
-6%
-10%
-10%
-11%
-6%
-12%
-10%
-10%
-4%
-5%
-4%
0%
-7%
-9%
-7%
-3%
-12%
-9%
0%
-12%
0%
-4%
-8%
-7%
>i
11
S£
14%
5%
0%
6%
9%
0%
6%
0%
0%
0%
0%
4%
1%
2%
0%
6%
7%
2%
1%
8%
0%
0%
3%
1%
TDS18
0%
3%
24%
1%
0%
25%
3%
18%
28%
25%
28%
0%
13%
5%
20%
0%
0%
10%
0%
0%
0%
46%
0%
23%
TDS24
0%
58%
71%
46%
0%
68%
32%
70%
60%
75%
45%
44%
73%
73%
73%
38%
49%
71%
67%
16%
0%
20%
63%
57%
TDS27
9%
5%
3%
9%
5%
4%
18%
3%
0%
0%
0%
0%
0%
0%
0%
1%
1%
0%
0%
31%
0%
1%
1%
2%
§
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
oo
s
0%
0%
2%
0%
0%
27%
13%
28%
0%
22%
7%
0%
14%
0%
4%
0%
0%
0%
0%
0%
0%
3%
0%
11%
\D
§
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
00
§
75%
81%
95%
83%
77%
65%
74%
66%
85%
71%
84%
36%
71%
91%
90%
54%
76%
72%
84%
91%
0%
75%
81%
76%
H
S
2%
10%
2%
5%
0%
6%
1%
6%
12%
7%
9%
44%
14%
7%
4%
27%
7%
23%
8%
0%
0%
7%
10%
8%
O
w
77%
91%
100%
88%
77%
98%
88%
100%
97%
100%
100%
80%
98%
98%
98%
81%
83%
96%
92%
91%
0%
84%
91%
94%
Pi
O
w
9%
63%
74%
56%
5%
71%
50%
73%
36%
75%
21%
44%
73%
73%
73%
39%
50%
71%
67%
47%
0%
15%
64%
54%
>
w
50%
25%
2%
32%
50%
3%
35%
0%
3%
0%
0%
25%
13%
20%
6%
29%
29%
15%
25%
44%
0%
16%
26%
9%
w
23%
9%
0%
12%
23%
1%
12%
0%
0%
0%
0%
20%
2%
2%
1%
19%
17%
4%
8%
9%
100%
0%
9%
2%
>
w
PH
18%
0%
0%
0%
22%
0%
0%
0%
0%
0%
0%
11%
0%
0%
0%
13%
4%
0%
0%
0%
0%
0%
1%
0%
C/3
C/3
0%
0%
0%
0%
5%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
LRRT2
100%
100%
100%
100%
100%
99%
100%
100%
97%
100%
100%
100%
100%
100%
99%
100%
100%
100%
100%
100%
0%
84%
100%
96%
IACC2
100%
100%
100%
100%
100%
99%
100%
100%
97%
100%
100%
100%
100%
100%
99%
100%
100%
100%
100%
100%
0%
84%
100%
96%
3
CM
w
77%
91%
100%
88%
77%
98%
88%
100%
97%
100%
100%
80%
98%
98%
98%
81%
83%
96%
92%
91%
0%
84%
91%
94%
Q
77%
91%
99%
88%
77%
98%
88%
91%
89%
100%
73%
80%
98%
98%
98%
81%
83%
96%
92%
91%
0%
68%
91%
88%
_1
C/3
Q
0%
0%
0%
1%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
                                                       5-62

-------
                                                                              2017 Draft Regulatory Impact Analysis
Table 3.8-39 Alternative 1- (Trucks +20) Truck Technology Penetrations in MY 2025

Aston
Martin
BMW
Chrysler/Fiat
Daimler
Ferrari
Ford
Geely
General
Motors
Honda
Hyundai
Kia
Lotus
Mazda
Mitsubishi
Nissan
Porsche
Spyker
Subaru
Suzuki
Tata
Tesla
Toyota
Volkswagen
Fleet
Mass Tech
Applied
NA
-17%
-12%
-20%
NA
-12%
-20%
-12%
-10%
-17%
-10%
NA
-20%
-20%
-8%
-20%
-5%
-20%
-16%
-13%
NA
-6%
-19%
-11%
11
NA
-13%
-12%
-12%
NA
-10%
-12%
-12%
-10%
-17%
-10%
NA
-18%
-18%
-8%
-12%
-2%
-20%
-16%
-7%
NA
-6%
-11%
-10%
>i
11
S£
NA
4%
0%
8%
NA
2%
8%
0%
0%
0%
0%
NA
1%
2%
0%
8%
3%
0%
0%
6%
NA
0%
7%
1%
TDS18
NA
30%
24%
46%
NA
32%
28%
15%
25%
25%
25%
NA
17%
16%
24%
39%
35%
25%
20%
33%
NA
28%
31%
24%
TDS24
NA
61%
62%
35%
NA
41%
66%
53%
75%
75%
75%
NA
75%
73%
60%
34%
46%
75%
73%
20%
NA
59%
58%
57%
TDS27
NA
10%
8%
19%
NA
20%
6%
15%
0%
0%
0%
NA
0%
0%
9%
28%
19%
0%
0%
33%
NA
8%
11%
11%
§
NA
0%
0%
0%
NA
0%
0%
0%
0%
0%
0%
NA
0%
0%
0%
0%
0%
0%
0%
0%
NA
0%
0%
0%
oo
S
NA
100%
86%
100%
NA
67%
100%
93%
62%
90%
99%
NA
66%
65%
77%
99%
100%
25%
79%
66%
NA
78%
100%
81%
\D
§
NA
0%
0%
0%
NA
0%
0%
0%
0%
0%
0%
NA
0%
0%
0%
0%
0%
0%
0%
0%
NA
0%
0%
0%
00
§
NA
0%
11%
0%
NA
27%
0%
7%
38%
10%
0%
NA
32%
34%
22%
0%
0%
67%
19%
29%
NA
14%
0%
16%
H
s
NA
0%
3%
0%
NA
2%
0%
0%
0%
0%
1%
NA
1%
0%
1%
1%
0%
8%
0%
0%
NA
3%
0%
1%
O
w
NA
100%
99%
100%
NA
96%
100%
100%
100%
100%
100%
NA
100%
98%
99%
100%
100%
100%
98%
95%
NA
94%
100%
98%
Pi
O
w
NA
70%
70%
54%
NA
61%
72%
67%
75%
75%
75%
NA
75%
73%
70%
61%
65%
75%
73%
53%
NA
67%
69%
68%
>
w
NA
30%
0%
50%
NA
26%
50%
2%
0%
0%
0%
NA
8%
9%
8%
50%
50%
0%
5%
41%
NA
6%
50%
9%
>
w
NA
0%
1%
0%
NA
2%
0%
0%
0%
0%
0%
NA
0%
2%
1%
0%
0%
0%
2%
5%
NA
0%
0%
0%
>
w
PH
NA
0%
0%
0%
NA
0%
0%
0%
0%
0%
0%
NA
0%
0%
0%
0%
0%
0%
0%
0%
NA
0%
0%
0%
C/3
C/3
NA
0%
0%
0%
NA
0%
0%
0%
0%
0%
0%
NA
0%
0%
0%
6%
0%
0%
0%
0%
NA
0%
0%
0%
LRRT2
NA
100%
100%
100%
NA
98%
100%
100%
100%
100%
100%
NA
100%
100%
100%
100%
100%
100%
100%
100%
NA
94%
100%
99%
IACC2
NA
100%
100%
100%
NA
98%
100%
100%
100%
100%
100%
NA
100%
100%
100%
100%
100%
100%
100%
100%
NA
94%
100%
99%
3
CM
w
NA
100%
99%
100%
NA
96%
100%
100%
100%
100%
100%
NA
100%
98%
99%
100%
100%
100%
98%
95%
NA
94%
100%
98%
Q
NA
100%
94%
92%
NA
96%
100%
84%
100%
100%
100%
NA
100%
98%
99%
100%
100%
100%
98%
95%
NA
94%
100%
94%
_1
C/3
Q
NA
0%
0%
8%
NA
0%
0%
0%
0%
0%
0%
NA
0%
0%
0%
0%
0%
0%
0%
0%
NA
0%
0%
0%
                                                       5-63

-------
Chapter 3
Table 3.8-40 Alternative 1- (Trucks +20) Fleet Technology Penetration in MY 2025

Aston
Martin
BMW
Chrysler/Fiat
Daimler
Ferrari
Ford
Geely
General
Motors
Honda
Hyundai
Kia
Lotus
Mazda
Mitsubishi
Nissan
Porsche
Spyker
Subaru
Suzuki
Tata
Tesla
Toyota
Volkswagen
Fleet
Mass Tech
Applied
-20%
-15%
-11%
-17%
-15%
-12%
-17%
-11%
-6%
-8%
-6%
-4%
-10%
-14%
-8%
-11%
-17%
-12%
-4%
-17%
0%
-5%
-13%
-9%
11
-6%
-11%
-11%
-11%
-6%
-11%
-11%
-11%
-6%
-8%
-6%
0%
-9%
-12%
-7%
-5%
-11%
-11%
-3%
-10%
0%
-5%
-9%
-8%
>i
11
S£
14%
5%
0%
7%
9%
1%
6%
0%
0%
0%
0%
4%
1%
2%
0%
6%
7%
1%
1%
7%
0%
0%
4%
1%
TDS18
0%
10%
24%
11%
0%
27%
11%
17%
27%
25%
28%
0%
13%
9%
21%
8%
5%
13%
3%
15%
0%
40%
6%
24%
TDS24
0%
59%
67%
44%
0%
59%
43%
62%
65%
75%
51%
44%
73%
73%
69%
37%
49%
72%
68%
18%
0%
34%
62%
57%
TDS27
9%
6%
5%
12%
5%
9%
14%
9%
0%
0%
0%
0%
0%
0%
3%
7%
3%
0%
0%
32%
0%
4%
3%
5%
§
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
oo
s
0%
26%
38%
23%
0%
39%
39%
59%
18%
35%
27%
0%
23%
21%
25%
21%
13%
6%
14%
31%
0%
30%
20%
34%
\D
§
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
00
§
75%
59%
59%
64%
77%
53%
52%
38%
71%
59%
66%
36%
64%
72%
70%
43%
66%
71%
73%
62%
0%
52%
65%
56%
H
S
2%
8%
3%
4%
0%
5%
1%
3%
8%
6%
7%
44%
11%
5%
3%
21%
6%
20%
7%
0%
0%
5%
8%
6%
O
w
77%
93%
100%
91%
77%
97%
92%
100%
98%
100%
100%
80%
98%
98%
98%
85%
85%
97%
93%
93%
0%
88%
93%
96%
Pi
O
w
9%
65%
72%
55%
5%
68%
57%
71%
48%
75%
32%
44%
73%
73%
72%
44%
52%
72%
68%
50%
0%
34%
65%
59%
>
w
50%
26%
1%
36%
50%
10%
39%
1%
2%
0%
0%
25%
12%
16%
7%
34%
32%
12%
22%
43%
0%
12%
31%
9%
w
23%
7%
0%
9%
23%
1%
8%
0%
0%
0%
0%
20%
2%
2%
1%
15%
15%
3%
7%
7%
100%
0%
7%
2%
>
w
PH
18%
0%
0%
0%
22%
0%
0%
0%
0%
0%
0%
11%
0%
0%
0%
10%
3%
0%
0%
0%
0%
0%
1%
0%
C/3
C/3
0%
0%
0%
0%
5%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
1%
0%
0%
0%
0%
0%
0%
0%
0%
LRRT2
100%
100%
100%
100%
100%
98%
100%
100%
98%
100%
100%
100%
100%
100%
99%
100%
100%
100%
100%
100%
0%
88%
100%
97%
IACC2
100%
100%
100%
100%
100%
98%
100%
100%
98%
100%
100%
100%
100%
100%
99%
100%
100%
100%
100%
100%
0%
88%
100%
97%
3
CM
w
77%
93%
100%
91%
77%
97%
92%
100%
98%
100%
100%
80%
98%
98%
98%
85%
85%
97%
93%
93%
0%
88%
93%
96%
Q
77%
93%
97%
89%
77%
97%
92%
88%
92%
100%
79%
80%
98%
98%
98%
85%
85%
97%
93%
93%
0%
77%
93%
90%
_1
C/3
Q
0%
0%
0%
2%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
                                                        5-64

-------
                                                                               2017 Draft Regulatory Impact Analysis
Table 3.8-41 Alternative 2- (Trucks -20) Car Technology Penetrations in MY 2025

Aston
Martin
BMW
Chrysler/Fiat
Daimler
Ferrari
Ford
Geely
General
Motors
Honda
Hyundai
Kia
Lotus
Mazda
Mitsubishi
Nissan
Porsche
Spyker
Subaru
Suzuki
Tata
Tesla
Toyota
Volkswagen
Fleet
Mass Tech
Applied
-20%
-15%
-12%
-17%
-15%
-13%
-17%
-13%
-6%
-8%
-6%
-4%
-8%
-11%
-8%
-9%
-19%
-11%
-2%
-20%
0%
-7%
-12%
-10%
11
-6%
-9%
-9%
-10%
-6%
-11%
-8%
-10%
-6%
-8%
-6%
0%
-6%
-8%
-6%
-3%
-11%
-8%
0%
-10%
0%
-6%
-8%
-8%
>i
11
S£
14%
6%
3%
7%
9%
3%
8%
3%
1%
0%
0%
4%
2%
3%
2%
7%
8%
2%
2%
10%
0%
1%
4%
2%
TDS18
0%
0%
6%
1%
0%
11%
3%
7%
14%
16%
25%
0%
4%
0%
2%
0%
0%
3%
0%
0%
0%
8%
0%
8%
TDS24
0%
52%
64%
42%
0%
61%
30%
64%
71%
73%
75%
44%
70%
64%
69%
30%
45%
68%
63%
14%
0%
58%
57%
63%
TDS27
9%
4%
2%
6%
5%
4%
11%
2%
0%
0%
0%
0%
0%
0%
0%
1%
1%
0%
0%
22%
0%
1%
0%
2%
§
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
oo
s
0%
0%
2%
0%
0%
29%
13%
28%
0%
22%
7%
0%
14%
0%
4%
0%
0%
0%
0%
0%
0%
3%
0%
11%
\D
§
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
00
§
75%
77%
88%
78%
77%
59%
68%
62%
85%
71%
84%
36%
69%
84%
87%
56%
76%
74%
81%
81%
0%
74%
75%
73%
H
S
2%
8%
2%
4%
0%
4%
1%
3%
10%
6%
9%
44%
12%
6%
4%
21%
6%
19%
7%
0%
0%
5%
9%
6%
O
w
77%
85%
92%
83%
77%
92%
81%
93%
96%
98%
100%
80%
95%
89%
94%
77%
82%
93%
88%
81%
0%
82%
84%
90%
Pi
O
w
9%
57%
66%
48%
5%
65%
41%
66%
71%
73%
75%
44%
70%
64%
69%
31%
46%
68%
63%
37%
0%
59%
57%
64%
>
w
50%
28%
20%
32%
50%
19%
41%
20%
13%
9%
0%
25%
21%
25%
24%
29%
29%
22%
25%
44%
0%
30%
26%
21%
w
23%
15%
8%
17%
23%
7%
19%
7%
1%
2%
0%
20%
5%
11%
5%
23%
18%
7%
12%
19%
100%
3%
16%
7%
>
w
PH
18%
0%
0%
2%
22%
0%
3%
0%
0%
0%
0%
11%
0%
0%
0%
17%
7%
0%
0%
0%
0%
0%
1%
0%
C/3
C/3
0%
0%
0%
0%
5%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
LRRT2
100%
100%
100%
100%
100%
99%
100%
100%
97%
100%
100%
100%
100%
100%
99%
100%
100%
100%
100%
100%
0%
85%
100%
96%
IACC2
100%
100%
100%
100%
100%
99%
100%
100%
97%
100%
100%
100%
100%
100%
99%
100%
100%
100%
100%
100%
0%
85%
100%
96%
3
CM
w
77%
85%
92%
83%
77%
92%
81%
93%
96%
98%
100%
80%
95%
89%
94%
77%
82%
93%
88%
81%
0%
82%
84%
90%
Q
77%
85%
92%
82%
77%
92%
81%
93%
96%
98%
100%
80%
95%
89%
94%
77%
82%
93%
88%
81%
0%
82%
84%
90%
_1
C/3
Q
0%
0%
0%
1%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
                                                        5-65

-------
Chapter 3
Table 3.8-42 Alternative 2- (Trucks -20) Truck Technology Penetrations in MY 2025

Aston
Martin
BMW
Chrysler/Fiat
Daimler
Ferrari
Ford
Geely
General
Motors
Honda
Hyundai
Kia
Lotus
Mazda
Mitsubishi
Nissan
Porsche
Spyker
Subaru
Suzuki
Tata
Tesla
Toyota
Volkswagen
Fleet
Mass Tech
Applied
NA
-18%
-13%
-20%
NA
-12%
-20%
-14%
-15%
-20%
-18%
NA
-20%
-20%
-9%
-20%
-5%
-20%
-16%
-14%
NA
-10%
-19%
-13%
11
NA
-11%
-12%
-12%
NA
-8%
-12%
-9%
-14%
-19%
-18%
NA
-18%
-18%
-6%
-12%
-2%
-17%
-16%
-7%
NA
-9%
-11%
-10%
>i
11
S£
NA
7%
2%
8%
NA
4%
8%
5%
1%
1%
0%
NA
1%
2%
3%
8%
3%
3%
0%
7%
NA
1%
7%
3%
TDS18
NA
30%
26%
46%
NA
27%
28%
31%
15%
23%
25%
NA
17%
16%
24%
39%
35%
6%
20%
33%
NA
25%
31%
26%
TDS24
NA
61%
59%
35%
NA
38%
66%
51%
73%
74%
75%
NA
74%
71%
59%
34%
46%
74%
72%
10%
NA
58%
58%
56%
TDS27
NA
10%
8%
19%
NA
20%
6%
15%
0%
0%
0%
NA
0%
0%
9%
28%
19%
0%
0%
33%
NA
8%
11%
11%
§
NA
0%
0%
0%
NA
0%
0%
0%
0%
0%
0%
NA
0%
0%
0%
0%
0%
0%
0%
0%
NA
0%
0%
0%
oo
S
NA
100%
87%
100%
NA
67%
100%
93%
62%
90%
99%
NA
66%
65%
77%
99%
100%
25%
79%
66%
NA
78%
100%
81%
\D
§
NA
0%
0%
0%
NA
0%
0%
0%
0%
0%
0%
NA
0%
0%
0%
0%
0%
0%
0%
0%
NA
0%
0%
0%
00
§
NA
0%
10%
0%
NA
25%
0%
6%
37%
9%
0%
NA
31%
31%
20%
0%
0%
69%
17%
26%
NA
14%
0%
15%
H
s
NA
0%
2%
0%
NA
2%
0%
0%
0%
0%
1%
NA
1%
0%
1%
1%
0%
6%
0%
0%
NA
2%
0%
1%
O
w
NA
100%
99%
100%
NA
94%
100%
99%
98%
99%
100%
NA
99%
96%
98%
100%
100%
99%
97%
92%
NA
94%
100%
97%
Pi
O
w
NA
70%
68%
54%
NA
58%
72%
66%
73%
74%
75%
NA
74%
71%
68%
61%
65%
74%
72%
43%
NA
66%
69%
67%
>
w
NA
50%
21%
50%
NA
38%
50%
39%
10%
2%
0%
NA
8%
9%
27%
50%
50%
19%
5%
41%
NA
17%
50%
27%
>
w
NA
0%
1%
0%
NA
5%
0%
1%
2%
1%
0%
NA
1%
4%
2%
0%
0%
1%
3%
8%
NA
0%
0%
1%
>
w
PH
NA
0%
2%
0%
NA
1%
0%
0%
0%
0%
0%
NA
0%
0%
0%
0%
0%
0%
0%
7%
NA
0%
0%
0%
C/3
C/3
NA
0%
0%
0%
NA
0%
0%
0%
0%
0%
0%
NA
0%
0%
0%
6%
19%
0%
0%
8%
NA
0%
0%
0%
LRRT2
NA
100%
100%
100%
NA
98%
100%
100%
100%
100%
100%
NA
100%
100%
100%
100%
100%
100%
100%
100%
NA
94%
100%
99%

-------
                                                                                2017 Draft Regulatory Impact Analysis
Table 3.8-43 Alternative 2- (Trucks -20) Fleet Technology Penetration in MY 2025

Aston
Martin
BMW
Chrysler/Fiat
Daimler
Ferrari
Ford
Geely
General
Motors
Honda
Hyundai
Kia
Lotus
Mazda
Mitsubishi
Nissan
Porsche
Spyker
Subaru
Suzuki
Tata
Tesla
Toyota
Volkswagen
Fleet
Mass Tech
Applied
-20%
-15%
-13%
-18%
-15%
-13%
-18%
-13%
-9%
-11%
-8%
-4%
-10%
-14%
-9%
-12%
-17%
-13%
-4%
-17%
0%
-8%
-13%
-11%
11
-6%
-9%
-10%
-10%
-6%
-10%
-10%
-9%
-8%
-10%
-8%
0%
-8%
-11%
-6%
-5%
-10%
-10%
-3%
-9%
0%
-7%
-8%
-8%
>i
jl
S£
14%
6%
3%
7%
9%
3%
8%
4%
1%
0%
0%
4%
2%
3%
2%
7%
7%
3%
1%
9%
0%
1%
5%
3%
TDS18
0%
8%
15%
11%
0%
16%
11%
18%
14%
18%
25%
0%
6%
5%
8%
8%
5%
3%
3%
15%
0%
14%
6%
14%
TDS24
0%
55%
62%
41%
0%
54%
41%
58%
72%
74%
75%
44%
71%
66%
66%
31%
45%
69%
65%
13%
0%
58%
57%
60%
TDS27
9%
6%
5%
9%
5%
9%
10%
8%
0%
0%
0%
0%
0%
0%
3%
7%
3%
0%
0%
27%
0%
4%
3%
5%
§
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
oo
s
0%
26%
39%
23%
0%
40%
39%
59%
18%
35%
27%
0%
23%
21%
25%
21%
13%
6%
14%
31%
0%
30%
20%
34%
\D
§
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
00
§
75%
57%
54%
60%
77%
49%
48%
35%
71%
58%
66%
36%
63%
66%
67%
44%
66%
73%
70%
55%
0%
52%
60%
54%
H
s
2%
6%
2%
3%
0%
3%
1%
2%
7%
5%
7%
44%
10%
4%
3%
17%
5%
16%
6%
0%
0%
4%
7%
4%
O
w
77%
89%
95%
87%
77%
92%
87%
96%
96%
99%
100%
80%
96%
91%
95%
82%
84%
94%
90%
86%
0%
86%
87%
92%
Pi
O
w
9%
60%
67%
50%
5%
63%
50%
66%
72%
74%
75%
44%
71%
66%
69%
38%
48%
69%
65%
40%
0%
62%
60%
65%
>
w
50%
34%
20%
36%
50%
25%
44%
29%
12%
7%
0%
25%
19%
20%
25%
34%
32%
22%
22%
43%
0%
25%
31%
23%
>
w
23%
11%
5%
13%
23%
6%
13%
4%
1%
1%
0%
20%
4%
9%
4%
18%
16%
6%
10%
14%
100%
2%
13%
5%
>
w
PH
18%
0%
1%
1%
22%
0%
2%
0%
0%
0%
0%
11%
0%
0%
0%
13%
6%
0%
0%
3%
0%
0%
1%
0%
C/3
C/3
0%
0%
0%
0%
5%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
1%
2%
0%
0%
4%
0%
0%
0%
0%
LRRT2
100%
100%
100%
100%
100%
99%
100%
100%
98%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
0%
88%
100%
97%

-------
Chapter 3
Table 3.8-44 Alternative 3- (Cars +20) Car Technology Penetrations in MY 2025

Aston
Martin
BMW
Chrysler/Fiat
Daimler
Ferrari
Ford
Geely
General
Motors
Honda
Hyundai
Kia
Lotus
Mazda
Mitsubishi
Nissan
Porsche
Spyker
Subaru
Suzuki
Tata
Tesla
Toyota
Volkswagen
Fleet
Mass Tech
Applied
-20%
-14%
-10%
-16%
-14%
-12%
-16%
-11%
-4%
-5%
-4%
-3%
-5%
-11%
-6%
-8%
-19%
-10%
0%
-20%
0%
-4%
-11%
-8%
11
-8%
-11%
-9%
-12%
-6%
-12%
-11%
-11%
-4%
-5%
-4%
0%
-5%
-10%
-6%
-3%
-13%
-10%
0%
-12%
0%
-4%
-8%
-7%
>i
11
S£
12%
3%
0%
4%
8%
0%
5%
0%
0%
0%
0%
3%
0%
1%
0%
5%
6%
0%
0%
8%
0%
0%
2%
1%
TDS18
0%
14%
24%
12%
0%
26%
3%
18%
12%
31%
14%
0%
25%
12%
25%
0%
0%
25%
25%
0%
0%
11%
15%
18%
TDS24
0%
62%
71%
50%
0%
68%
35%
70%
36%
28%
5%
53%
75%
75%
74%
45%
53%
74%
69%
16%
0%
17%
66%
50%
TDS27
22%
5%
3%
9%
13%
4%
20%
3%
0%
0%
0%
0%
0%
0%
0%
1%
6%
0%
0%
32%
0%
1%
2%
2%
§
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
oo
s
0%
0%
2%
0%
0%
27%
13%
28%
0%
22%
7%
0%
14%
0%
4%
0%
0%
0%
0%
0%
0%
3%
0%
11%
\D
§
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
00
§
73%
83%
95%
85%
76%
65%
79%
66%
85%
71%
84%
31%
68%
93%
91%
52%
80%
72%
83%
92%
0%
75%
80%
76%
H
S
4%
12%
2%
7%
1%
7%
1%
6%
12%
7%
9%
53%
18%
7%
5%
31%
8%
27%
11%
0%
0%
7%
14%
8%
O
w
77%
95%
100%
92%
77%
98%
93%
100%
97%
100%
100%
85%
100%
100%
99%
83%
88%
99%
94%
92%
0%
84%
93%
95%
Pi
O
w
22%
67%
74%
60%
13%
72%
55%
73%
11%
24%
5%
53%
47%
75%
56%
46%
59%
74%
69%
48%
0%
14%
67%
46%
>
w
50%
15%
2%
21%
50%
2%
35%
0%
3%
0%
0%
25%
0%
13%
1%
29%
29%
0%
0%
44%
0%
16%
11%
6%
w
23%
5%
0%
8%
23%
0%
7%
0%
0%
0%
0%
15%
0%
0%
0%
17%
12%
1%
6%
8%
100%
0%
7%
1%
>
w
PH
5%
0%
0%
0%
14%
0%
0%
0%
0%
0%
0%
6%
0%
0%
0%
8%
0%
0%
0%
0%
0%
0%
0%
0%
C/3
C/3
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
LRRT2
100%
100%
100%
100%
100%
99%
100%
100%
97%
100%
100%
100%
100%
100%
99%
100%
100%
100%
100%
100%
0%
84%
100%
96%
IACC2
100%
100%
100%
100%
100%
99%
100%
100%
97%
100%
100%
100%
100%
100%
99%
100%
100%
100%
100%
100%
0%
84%
100%
96%
3
CM
w
77%
95%
100%
92%
77%
98%
93%
100%
97%
100%
100%
85%
100%
100%
99%
83%
88%
99%
94%
92%
0%
84%
93%
95%
Q
77%
95%
99%
92%
77%
98%
93%
91%
48%
59%
19%
85%
100%
100%
99%
83%
88%
99%
94%
92%
0%
29%
93%
73%
_1
C/3
Q
0%
0%
0%
1%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
                                                       5-68

-------
                                                                               2017 Draft Regulatory Impact Analysis
Table 3.8-45 Alternative 3- (Cars +20) Truck Technology Penetrations in MY 2025

Aston
Martin
BMW
Chrysler/Fiat
Daimler
Ferrari
Ford
Geely
General
Motors
Honda
Hyundai
Kia
Lotus
Mazda
Mitsubishi
Nissan
Porsche
Spyker
Subaru
Suzuki
Tata
Tesla
Toyota
Volkswagen
Fleet
Mass Tech
Applied
NA
-17%
-12%
-20%
NA
-11%
-20%
-12%
-9%
-10%
-9%
NA
-15%
-20%
-7%
-20%
-5%
-20%
-16%
-13%
NA
-6%
-19%
-11%
11
NA
-13%
-12%
-14%
NA
-10%
-15%
-12%
-9%
-10%
-9%
NA
-15%
-19%
-7%
-12%
-2%
-20%
-16%
-7%
NA
-6%
-11%
-10%
>i
11
S£
NA
4%
0%
6%
NA
0%
5%
0%
0%
0%
0%
NA
0%
1%
0%
8%
3%
0%
0%
6%
NA
0%
7%
1%
TDS18
NA
30%
24%
46%
NA
32%
28%
16%
25%
25%
25%
NA
25%
22%
30%
39%
35%
25%
20%
33%
NA
28%
31%
25%
TDS24
NA
61%
62%
35%
NA
43%
66%
52%
75%
75%
75%
NA
75%
75%
61%
34%
46%
75%
73%
20%
NA
59%
58%
58%
TDS27
NA
10%
8%
19%
NA
20%
6%
15%
0%
0%
0%
NA
0%
0%
9%
28%
19%
0%
0%
33%
NA
8%
11%
11%
§
NA
0%
0%
0%
NA
0%
0%
0%
0%
0%
0%
NA
0%
0%
0%
0%
0%
0%
0%
0%
NA
0%
0%
0%
oo
S
NA
100%
86%
100%
NA
67%
100%
93%
62%
90%
99%
NA
66%
65%
77%
99%
100%
25%
79%
66%
NA
78%
100%
81%
\D
§
NA
0%
0%
0%
NA
0%
0%
0%
0%
0%
0%
NA
0%
0%
0%
0%
0%
0%
0%
0%
NA
0%
0%
0%
00
§
NA
0%
11%
0%
NA
28%
0%
6%
38%
10%
0%
NA
32%
35%
22%
0%
0%
67%
19%
29%
NA
14%
0%
16%
H
s
NA
0%
3%
0%
NA
2%
0%
0%
0%
0%
1%
NA
2%
0%
2%
1%
0%
8%
0%
0%
NA
3%
0%
1%
O
w
NA
100%
100%
100%
NA
97%
100%
100%
100%
100%
100%
NA
100%
100%
100%
100%
100%
100%
98%
95%
NA
94%
100%
98%
Pi
O
w
NA
70%
71%
54%
NA
63%
72%
67%
75%
75%
74%
NA
75%
75%
70%
61%
65%
75%
73%
53%
NA
67%
69%
68%
>
w
NA
30%
0%
38%
NA
5%
30%
2%
0%
0%
0%
NA
0%
3%
0%
50%
50%
0%
5%
41%
NA
6%
50%
6%
>
w
NA
0%
0%
0%
NA
0%
0%
0%
0%
0%
0%
NA
0%
0%
0%
0%
0%
0%
2%
5%
NA
0%
0%
0%
>
w
PH
NA
0%
0%
0%
NA
0%
0%
0%
0%
0%
0%
NA
0%
0%
0%
0%
0%
0%
0%
0%
NA
0%
0%
0%
C/3
C/3
NA
0%
0%
0%
NA
0%
0%
0%
0%
0%
0%
NA
0%
0%
0%
6%
0%
0%
0%
0%
NA
0%
0%
0%
LRRT2
NA
100%
100%
100%
NA
97%
100%
100%
100%
100%
100%
NA
100%
100%
100%
100%
100%
100%
100%
100%
NA
94%
100%
98%
IACC2
NA
100%
100%
100%
NA
97%
100%
100%
100%
100%
100%
NA
100%
100%
100%
100%
100%
100%
100%
100%
NA
94%
100%
98%
3
CM
w
NA
100%
100%
100%
NA
97%
100%
100%
100%
100%
100%
NA
100%
100%
100%
100%
100%
100%
98%
95%
NA
94%
100%
98%
Q
NA
100%
94%
92%
NA
97%
100%
85%
100%
100%
100%
NA
100%
100%
100%
100%
100%
100%
98%
95%
NA
94%
100%
94%
_1
C/3
Q
NA
0%
0%
8%
NA
0%
0%
0%
0%
0%
0%
NA
0%
0%
0%
0%
0%
0%
0%
0%
NA
0%
0%
0%
                                                       5-69

-------
Chapter 3
Table 3.8-46 Alternative 3- (Cars +20) Fleet Technology Penetration in MY 2025

Aston
Martin
BMW
Chrysler/Fiat
Daimler
Ferrari
Ford
Geely
General
Motors
Honda
Hyundai
Kia
Lotus
Mazda
Mitsubishi
Nissan
Porsche
Spyker
Subaru
Suzuki
Tata
Tesla
Toyota
Volkswagen
Fleet
Mass Tech
Applied
-20%
-15%
-11%
-17%
-14%
-11%
-17%
-11%
-5%
-6%
-5%
-3%
-7%
-14%
-7%
-10%
-17%
-12%
-3%
-17%
0%
-5%
-12%
-9%
§|
H S
-8%
-12%
-11%
-12%
-6%
-11%
-12%
-11%
-5%
-6%
-5%
0%
-7%
-13%
-7%
-5%
-12%
-12%
-3%
-10%
0%
-5%
-9%
-8%
t/3 &
VI -T3
rf %
M C3
s£
12%
3%
0%
5%
8%
0%
5%
0%
0%
0%
0%
3%
0%
1%
0%
5%
6%
0%
0%
7%
0%
0%
3%
1%
TDS18
0%
18%
24%
20%
0%
28%
11%
17%
16%
30%
16%
0%
25%
15%
26%
8%
5%
25%
24%
15%
0%
17%
18%
20%
TDS24
0%
62%
67%
47%
0%
60%
44%
62%
48%
38%
20%
53%
75%
75%
70%
43%
52%
75%
70%
18%
0%
32%
64%
53%
TDS27
22%
6%
5%
12%
13%
9%
16%
9%
0%
0%
0%
0%
0%
0%
3%
7%
8%
0%
0%
32%
0%
4%
3%
5%
§
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
oo
s
0%
26%
38%
23%
0%
39%
39%
59%
18%
35%
27%
0%
23%
21%
25%
21%
13%
6%
14%
31%
0%
30%
20%
34%
\D
§
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
00
§
73%
61%
59%
66%
76%
53%
55%
38%
71%
59%
66%
31%
62%
74%
70%
41%
70%
71%
72%
63%
0%
52%
64%
56%
H
s
4%
9%
3%
5%
1%
5%
1%
3%
8%
6%
7%
53%
15%
5%
4%
25%
7%
23%
9%
0%
0%
5%
11%
6%
O
w
77%
97%
100%
94%
77%
98%
95%
100%
98%
100%
100%
85%
100%
100%
99%
87%
90%
100%
95%
93%
0%
88%
95%
96%
Pi
O
w
22%
68%
72%
58%
13%
69%
60%
70%
30%
35%
20%
53%
52%
75%
60%
49%
60%
75%
70%
51%
0%
33%
68%
54%
>
w
50%
19%
1%
25%
50%
3%
33%
1%
2%
0%
0%
25%
0%
10%
1%
34%
32%
0%
1%
43%
0%
12%
19%
6%
>
w
23%
3%
0%
6%
23%
0%
5%
0%
0%
0%
0%
15%
0%
0%
0%
13%
10%
0%
5%
7%
100%
0%
5%
1%
>
w
PH
5%
0%
0%
0%
14%
0%
0%
0%
0%
0%
0%
6%
0%
0%
0%
6%
0%
0%
0%
0%
0%
0%
0%
0%
C/3
C/3
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
1%
0%
0%
0%
0%
0%
0%
0%
0%
LRRT2
100%
100%
100%
100%
100%
98%
100%
100%
98%
100%
100%
100%
100%
100%
99%
100%
100%
100%
100%
100%
0%
88%
100%
97%
IACC2
100%
100%
100%
100%
100%
98%
100%
100%
98%
100%
100%
100%
100%
100%
99%
100%
100%
100%
100%
100%
0%
88%
100%
97%
3
CM
w
77%
97%
100%
94%
77%
98%
95%
100%
98%
100%
100%
85%
100%
100%
99%
87%
90%
100%
95%
93%
0%
88%
95%
96%
Q
77%
97%
97%
92%
77%
98%
95%
89%
63%
68%
36%
85%
100%
100%
99%
87%
90%
100%
95%
93%
0%
53%
95%
80%
_1
C/3
Q
0%
0%
0%
2%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
                                                        5-70

-------
                                                                               2017 Draft Regulatory Impact Analysis
Table 3.8-47 Alternative 4- (Cars -20) Car Technology Penetrations in MY 2025

Aston
Martin
BMW
Chrysler/Fiat
Daimler
Ferrari
Ford
Geely
General
Motors
Honda
Hyundai
Kia
Lotus
Mazda
Mitsubishi
Nissan
Porsche
Spyker
Subaru
Suzuki
Tata
Tesla
Toyota
Volkswagen
Fleet
Mass Tech
Applied
-20%
-15%
-12%
-17%
-15%
-13%
-17%
-13%
-6%
-9%
-6%
-5%
-8%
-11%
-8%
-10%
-19%
-11%
-2%
-20%
0%
-7%
-12%
-10%
§|
H S
-5%
-8%
-8%
-8%
-6%
-9%
-7%
-10%
-5%
-7%
-5%
0%
-6%
-7%
-6%
-3%
-10%
-7%
0%
-10%
0%
-6%
-6%
-7%
t/3 &
VI -T3
rf %
M C3
s£
15%
7%
4%
9%
9%
5%
9%
3%
1%
1%
1%
5%
2%
4%
2%
8%
10%
3%
2%
10%
0%
1%
6%
3%
TDS18
0%
0%
2%
0%
0%
9%
3%
7%
12%
5%
5%
0%
4%
0%
1%
0%
0%
0%
0%
0%
0%
8%
0%
6%
TDS24
0%
48%
64%
38%
0%
58%
26%
64%
68%
67%
71%
32%
62%
61%
66%
0%
34%
63%
55%
14%
0%
57%
47%
59%
TDS27
5%
1%
2%
2%
5%
4%
6%
2%
0%
0%
0%
0%
0%
0%
0%
14%
1%
0%
0%
20%
0%
1%
0%
1%
§
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
oo
s
0%
0%
2%
0%
0%
29%
13%
28%
0%
22%
7%
0%
14%
0%
4%
0%
0%
0%
0%
0%
0%
3%
0%
11%
\D
§
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
00
§
76%
74%
88%
76%
77%
57%
68%
62%
83%
66%
83%
45%
62%
80%
84%
67%
74%
71%
78%
79%
0%
73%
73%
71%
H
s
1%
7%
2%
4%
0%
4%
1%
3%
10%
5%
6%
32%
10%
6%
3%
10%
5%
17%
6%
0%
0%
5%
8%
5%
O
w
77%
81%
92%
81%
77%
89%
81%
93%
92%
92%
96%
77%
87%
86%
91%
77%
78%
88%
85%
79%
0%
81%
82%
88%
Pi
O
w
5%
48%
66%
39%
5%
61%
32%
67%
68%
67%
71%
32%
62%
61%
66%
14%
35%
63%
55%
35%
0%
58%
47%
61%
>
w
50%
28%
24%
32%
50%
28%
41%
20%
16%
20%
20%
25%
22%
25%
25%
41%
29%
25%
25%
44%
0%
30%
26%
24%
>
w
23%
19%
8%
19%
23%
10%
19%
7%
4%
8%
4%
23%
13%
14%
9%
23%
22%
12%
15%
21%
100%
5%
18%
9%
>
w
PH
22%
4%
0%
9%
22%
0%
12%
0%
0%
0%
0%
20%
0%
0%
0%
22%
14%
0%
4%
0%
0%
0%
8%
1%
C/3
C/3
5%
0%
0%
0%
5%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
14%
0%
0%
0%
0%
0%
0%
0%
0%
LRRT2
100%
100%
100%
100%
100%
99%
100%
100%
97%
100%
100%
100%
100%
100%
99%
100%
100%
100%
100%
100%
0%
85%
100%
96%
IACC2
100%
100%
100%
100%
100%
99%
100%
100%
97%
100%
100%
100%
100%
100%
99%
100%
100%
100%
100%
100%
0%
85%
100%
96%
3
CM
w
77%
81%
92%
81%
77%
89%
81%
93%
92%
92%
96%
77%
87%
86%
91%
77%
78%
88%
85%
79%
0%
81%
82%
88%
Q
77%
81%
92%
80%
77%
89%
81%
93%
92%
92%
96%
77%
87%
86%
91%
77%
78%
88%
85%
79%
0%
81%
82%
88%
_1
C/3
Q
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
                                                        5-71

-------
Chapter 3
Table 3.8-48 Alternative 4- (Cars -20) Truck Technology Penetrations in MY 2025

Aston
Martin
BMW
Chrysler/Fiat
Daimler
Ferrari
Ford
Geely
General
Motors
Honda
Hyundai
Kia
Lotus
Mazda
Mitsubishi
Nissan
Porsche
Spyker
Subaru
Suzuki
Tata
Tesla
Toyota
Volkswagen
Fleet
Mass Tech
Applied
NA
-18%
-13%
-20%
NA
-13%
-20%
-14%
-16%
-20%
-18%
NA
-20%
-20%
-10%
-20%
-5%
-20%
-16%
-14%
NA
-11%
-19%
-14%
§|
H S
NA
-11%
-12%
-12%
NA
-8%
-12%
-9%
-13%
-19%
-18%
NA
-13%
-15%
-6%
-12%
-2%
-15%
-9%
-7%
NA
-8%
-11%
-10%
t/3 &
VI -T3
rf %
M C3
s£
NA
7%
2%
8%
NA
5%
8%
5%
3%
1%
0%
NA
7%
5%
4%
8%
3%
5%
7%
7%
NA
2%
7%
4%
TDS18
NA
30%
26%
46%
NA
27%
28%
31%
15%
23%
25%
NA
17%
16%
24%
39%
35%
6%
20%
33%
NA
25%
31%
26%
TDS24
NA
61%
59%
35%
NA
38%
66%
51%
71%
74%
75%
NA
72%
70%
57%
11%
46%
69%
72%
10%
NA
57%
58%
55%
TDS27
NA
10%
8%
19%
NA
20%
6%
15%
0%
0%
0%
NA
0%
0%
9%
50%
19%
0%
0%
33%
NA
8%
11%
11%
§
NA
0%
0%
0%
NA
0%
0%
0%
0%
0%
0%
NA
0%
0%
0%
0%
0%
0%
0%
0%
NA
0%
0%
0%
oo
S
NA
100%
87%
100%
NA
67%
100%
93%
62%
90%
99%
NA
66%
65%
77%
99%
100%
25%
79%
66%
NA
79%
100%
81%
\D
§
NA
0%
0%
0%
NA
0%
0%
0%
0%
0%
0%
NA
0%
0%
0%
0%
0%
0%
0%
0%
NA
0%
0%
0%
00
§
NA
0%
10%
0%
NA
24%
0%
6%
34%
8%
0%
NA
29%
30%
19%
0%
0%
64%
17%
26%
NA
13%
0%
14%
H
S
NA
0%
2%
0%
NA
2%
0%
0%
0%
0%
1%
NA
1%
0%
1%
1%
0%
5%
0%
0%
NA
2%
0%
1%
O
w
NA
100%
99%
100%
NA
93%
100%
99%
96%
99%
100%
NA
97%
95%
96%
100%
100%
94%
97%
92%
NA
94%
100%
97%
Pi
O
w
NA
70%
68%
54%
NA
58%
72%
66%
71%
74%
75%
NA
72%
70%
67%
61%
65%
69%
72%
43%
NA
65%
69%
66%
>
w
NA
50%
21%
50%
NA
44%
50%
39%
17%
7%
0%
NA
42%
25%
39%
50%
50%
25%
45%
41%
NA
26%
50%
33%
>
w
NA
0%
1%
0%
NA
5%
0%
1%
4%
1%
0%
NA
3%
5%
4%
0%
0%
6%
3%
8%
NA
1%
0%
2%
>
w
PH
NA
0%
2%
0%
NA
1%
0%
0%
0%
0%
0%
NA
0%
0%
0%
0%
0%
0%
0%
7%
NA
0%
0%
0%
C/3
C/3
NA
0%
0%
27%
NA
0%
0%
0%
0%
0%
0%
NA
0%
0%
0%
50%
50%
0%
0%
8%
NA
0%
50%
2%
LRRT2
NA
100%
100%
100%
NA
98%
100%
100%
100%
100%
100%
NA
100%
100%
100%
100%
100%
100%
100%
100%
NA
96%
100%
99%

-------
                                                                                2017 Draft Regulatory Impact Analysis
Table 3.8-49 Alternative 4- (Cars -20) Fleet Technology Penetration in MY 2025

Aston
Martin
BMW
Chrysler/Fiat
Daimler
Ferrari
Ford
Geely
General
Motors
Honda
Hyundai
Kia
Lotus
Mazda
Mitsubishi
Nissan
Porsche
Spyker
Subaru
Suzuki
Tata
Tesla
Toyota
Volkswagen
Fleet
Mass Tech
Applied
-20%
-16%
-13%
-18%
-15%
-13%
-18%
-13%
-9%
-11%
-9%
-5%
-10%
-14%
-9%
-13%
-18%
-13%
-5%
-17%
0%
-8%
-13%
-11%
§|
H S
-5%
-9%
-10%
-9%
-6%
-8%
-9%
-9%
-7%
-10%
-8%
0%
-7%
-10%
-6%
-5%
-9%
-9%
-2%
-9%
0%
-7%
-7%
-8%
t/3 &
VI -T3
rf %
M C3
s£
15%
7%
3%
8%
9%
5%
9%
4%
2%
1%
1%
5%
3%
4%
3%
8%
9%
4%
3%
9%
0%
1%
6%
3%
TDS18
0%
8%
13%
11%
0%
15%
11%
18%
13%
9%
9%
0%
6%
5%
8%
8%
5%
1%
3%
15%
0%
14%
6%
12%
TDS24
0%
51%
62%
37%
0%
52%
38%
58%
69%
69%
72%
32%
64%
64%
63%
2%
36%
65%
58%
13%
0%
57%
49%
58%
TDS27
5%
3%
5%
6%
5%
9%
6%
8%
0%
0%
0%
0%
0%
0%
3%
22%
3%
0%
0%
26%
0%
4%
2%
4%
§
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
oo
s
0%
26%
39%
23%
0%
41%
39%
59%
18%
35%
27%
0%
23%
21%
25%
21%
13%
6%
14%
31%
0%
31%
20%
34%
\D
§
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
00
§
76%
55%
54%
59%
77%
47%
48%
35%
68%
54%
65%
45%
57%
63%
65%
53%
64%
69%
68%
54%
0%
51%
59%
53%
H
s
1%
5%
2%
3%
0%
3%
0%
2%
7%
4%
5%
32%
9%
4%
2%
8%
4%
15%
5%
0%
0%
4%
7%
4%
O
w
77%
86%
95%
85%
77%
90%
87%
96%
94%
94%
97%
77%
89%
89%
92%
82%
81%
90%
87%
85%
0%
86%
85%
91%
Pi
O
w
5%
54%
67%
43%
5%
60%
44%
67%
69%
69%
72%
32%
64%
64%
66%
24%
39%
65%
58%
39%
0%
61%
51%
63%
>
w
50%
34%
23%
36%
50%
33%
44%
29%
16%
17%
16%
25%
25%
25%
29%
43%
32%
25%
28%
43%
0%
28%
31%
27%
>
w
23%
14%
5%
15%
23%
9%
13%
4%
4%
6%
3%
23%
11%
11%
7%
18%
19%
10%
13%
15%
100%
3%
15%
7%
>
w
PH
22%
3%
1%
7%
22%
0%
8%
0%
0%
0%
0%
20%
0%
0%
0%
17%
12%
0%
3%
3%
0%
0%
7%
1%
C/3
C/3
5%
0%
0%
6%
5%
0%
0%
0%
0%
0%
0%
0%
0%
0%
0%
22%
7%
0%
0%
4%
0%
0%
10%
1%
LRRT2
100%
100%
100%
100%
100%
99%
100%
100%
98%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
100%
0%
89%
100%
97%

-------
Chapter 3
       3.8.5   Additional Detail on Mass Reduction Technology

       For MY 2021 and MY 2025, additional details are presented on the distribution of
mass reduction in the fleet by vehicle class. For presentation in this analysis, we aggregated
the 19 vehicle types into five vehicle classes.

       Table 3.8-50 Aggregation of Vehicle types for Mass Reduction Presentation
Vehicle Type
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
Aggregated Type
Subcompact/Compact
Subcompact/Compact
Midsize Car
Subcompact/Compact
Midsize Car
Large Car
Midsize MPV/Small Track
Midsize MPV/Small Track
Large Track/MPV/SUV
Large Track/MPV/SUV
Large Track/MPV/SUV
Large Track/MPV/SUV
Large Track/MPV/SUV
Large Track/MPV/SUV
Large Car
Large Track/MPV/SUV
Large Track/MPV/SUV
Large Track/MPV/SUV
Large Track/MPV/SUV
       After aggregations here are the weight reductions by vehicle class.
Net Mass Reduction

Category
Subcompact/Compact
Midsize Car
Midsize MPV/Small Truck
Large Truck/MPV/SUV
Large Car
Fleet
Reference
MY 2021
-2%
-7%
-5%
-7%
-8%
-5%
MY 2025
-2%
-7%
-5%
-7%
-8%
-5%
Control
MY 2021
-2%
-8%
-7%
-8%
-9%
-6%
MY 2025
-3%
-12%
-11%
-13%
-9%
-9%
       3.8.6   Air Conditioning Cost

       As previously referenced, once the OMEGA costs were determined, the estimated air
conditioning costs, as discussed in Chapter 5 of the draft Joint TSD were added onto the total
cost.  These costs are shown below.
                                         3-74

-------
                                               2017 Draft Regulatory Impact Analysis
        Table 3.8-51 Total Costs for A/C Control Used in This Proposal (2009$)
Car/
Truck
Car
Truck
Fleet
Case
Reference
Control
Total
Reference
Control
Total
Total
2021
$67
$78
$145
$51
$94
$145
$145
2025
$63
$69
$132
$48
$84
$132
$132
       3.8.7   Stranded Capital

Because the production of automotive components is capital-intensive, it is possible for
substantial capital investments in manufacturing equipment and facilities to become
"stranded" (where their value is lost, or diminished). This would occur when the capital is
rendered useless (or less useful) by some factor that forces  a major change in vehicle design,
plant operations, or manufacturer's product mix, such as a shift in consumer  demand for
certain vehicle types. It can also be caused by new standards that phase-in at a rate too rapid
to accommodate planned replacement or redisposition of existing capital to other activities.
The lost value of capital equipment is then amortized in some way  over production of the new
technology components. A discussion of this issue is presented in Chapter 3  of the TSD. To
help ensure a conservative cost analysis for the rule (i.e., an analysis that might err on the side
of over-costing), EPA asked FEV to calculate potential stranded capital on six  specific
technologies, using a set of conservative assumptions described in the TSD.  EPA then
included these potential additional technology costs as a post-process to the OMEGA model
(Table 3.8-53 ). These "stranded capital" costs were not directly incorporated into the
technology inputs because they are a function of how rapidly technologies are phased in.
Costs for potential stranded capital (as  shown in) depend both on the stranded technology and
the replacing technology.
                    Table 3.8-52 Potential Stranded Capital Costs
Replaced
technology
6-speed AT
6-speed AT
6-speed DCT
Conventional V6
Conventional V8
Conventional V6
New
technology
6-speed DCT
8-speed AT
8-speed DCT
DSTGDI 14
DSTGDI V6
Power-split HEV
Stranded capital cost per vehicle
when replaced technology's production is ended
after:
3 years
$55
$48
$28
$56
$60
$111
5 years
$39
$34
$20
$40
$43
$79
8 years
$16
$14
$8
$16
$17
$32
                                         3-75

-------
Chapter 3

    DSTGDI=Downsized, turbocharged engine with stoichiometric gasoline direct injection.

       For 2008-2016, the eight year stranded capital costs were used. For 2016-2021 and
2021-2021, the five year stranded capital costs were used.  This properly reflects EPA's
analytic assumption that redesign schedules are evenly spread through time.

       For transmissions, EPA determined the change in quantity of 6 and 8 speed automatic
and dual clutch transmissions. For each of these transmissions, manufacturers that increased
their production quantity had no stranded capital, otherwise, we applied a per piece cost
corresponding to the table above. For engines, the stranded capital work done by FEV does
not precisely correspond to the technologies considered in OMEGA; significantly,  the pieces
of "stranded" technology were often not those that were similarly "stranded" by the OMEGA
projections. As an example, OMEGA might forecast a 24 bar BMEP turbo-charged
downsized engine in 2021, and then 27 bar BMEP engine technology in 2025. The stranded
24 bar engine, while based on a FEV cost analysis, does not directly correspond to any
technology listed above. As a result, EPA created a projection that for each manufacturer
listed the number of engines with 8, 6, 4 or 3, as well as the number of EVs and Atkinson
cycle HEVs.  A decrease in any of these quantities resulted in a $50 per engine increase in
cost, which is a rough average of the five year stranded capital cost for the three engine
technologies.

       Total potential stranded capital determined by this analysis is shown below, and
includes all manufacturers including SVMs. These costs are not differentiated between car
and truck. As the values are small, we applied these same potential stranded capital costs to
all alternatives. The highest costs are in MY 2021, reflecting the rapid technology change
during the time leading up to that MY.
                                         3-76

-------
                                                       2017 Draft Regulatory Impact Analysis
                    Table 3.8-53 Estimated Potential Stranded Capital
                                                                             ss

Manufacturer
Aston Martin
BMW
Chrysler/Fiat
Daimler
Ferrari
Ford
Geely
General
Motors
Honda
Hyundai
Kia
Lotus
Mazda
Mitsubishi
Nissan
Porsche
Spyker
Subaru
Suzuki
Tata
Tesla
Toyota
Volkswagen
Fleet
MY2016
Engine
$60
$20
$54
$18
$5
$8
$14
$12
$2
$-
$-
$40
$5
$-
$6
$23
$45
$3
$20
$14
$-
$0
$13
$9
Trans-
mission
$15
$3
$-
$6
$1
$-
$0
$-
$-
(D
J>-
$-
$-
$-
$-
$-
$1
(D
J>-
$-
$-
$4
$-
$-
$2
$0
Total
$75
$23
$54
$24
$6
$8
$14
$12
$2
$-
$-
$40
$5
$-
$6
$25
$45
$3
$20
$18
$-
$0
$16
$10
MY 2021
Engine
$21
$29
$21
$17
$22
$17
$22
$8
$11
$11
$25
$15
$32
$25
$10
$15
$14
$18
$19
$20
$-
$17
$21
$15
Trans-
mission
$8
$15
$14
$9
$12
$12
$16
$10
$10
$6
$16
$0
$15
$15
$9
$10
$8
$7
$8
$12
$-
$6
$10
$10
Total
$29
$45
$35
$27
$34
$29
$38
$18
$21
$17
$41
$16
$47
$40
$19
$24
$22
$25
$27
$32
$-
$23
$31
$25
MY 2025
Engine
$14
$3
$9
$6
$16
$6
$8
$16
$11
$14
$16
$1
$14
$12
$9
$4
$3
$7
$2
$14
$-
$14
$2
$11
Trans-
mission
$3
$4
$5
$4
$3
$5
$5
$5
$4
$4
$4
$2
$4
$4
$4
$3
$3
$3
$4
$4
$-
$9
$4
$5
Total
$17
$7
$15
$10
$19
$10
$13
$21
$15
$18
$21
$2
$18
$16
$14
$7
$6
$10
$6
$19
$-
$23
$6
$16
ss Note that the total potential stranded capital for Aston Martin engines is greater than $50, the cost of the
potential stranded capital.  This is because the market forecast includes a decrease in sales for Aston Martin, and
a projected change in number of cylinders for every one of their engines.  Also note, as described in section
III.B.5 of the preamble, small volume manufacturers with U.S. sales of less than 5,000 vehicles would be able to
petition EPA for an alternative standard for MY 2017 and later. Manufacturers currently meeting the 5,000
vehicle cut point include Lotus, Aston Martin, and McLaren. Thus, these potential stranded capital costs may be
overstated for these small volume manufacturers.
                                               3-77

-------
Chapter 3
3.9  Per Vehicle Costs 2021 and 2025

       As described above, the per-vehicle technology costs for this program alone must
account for any cost that are incurred due to compliance with existing vehicle programs. EPA
first used OMEGA to calculate costs reflected in the existing 2012-2016 program, which is
the reference case for this analysis. The OMEGA estimates indicate that, on average,
manufacturers will  need to spend $830 to meet the 2016MY standards in the 2021MY, and
$776 to meet the 2016MY standards in the 2025MY per vehicle.  Reference case costs,
inclusive of AC costs, are provided in Table 3.9-1 .

                          Table 3.9-1 Reference Case Costs
Company
Aston Martin
BMW
Chrysler/Fiat
Daimler
Ferrari
Ford
Geely-Volvo
GM
Honda
Hyundai
Kia
Lotus
Mazda
Mitsubishi
Nissan
Porsche
Spyker-Saab
Subaru
Suzuki
Tata-JLR
Tesla
Toyota
VW
Fleet
2021
Cars
$2,589
$1,988
$921
$2,227
$2,470
$893
$2,159
$859
$320
$441
$384
$1,691
$611
$1,046
$409
$1,934
$2,000
$714
$1,068
$2,529
$0
$290
$1,870
$776
Trucks
$0
$2,220
$945
$2,290
$0
$1,220
$2,199
$910
$465
$785
$641
$0
$1,091
$1,235
$1,021
$1,935
$2,460
$1,080
$1,328
$2,529
$0
$462
$1,909
$930
Fleet
$2,589
$2,049
$931
$2,243
$2,470
$1,004
$2,172
$884
$365
$511
$442
$1,691
$696
$1,112
$598
$1,934
$2,066
$801
$1,115
$2,529
$0
$357
$1,878
$830
2025
Cars
$2,376
$1,827
$803
$2,058
$2,270
$856
$1,987
$815
$314
$424
$374
$1,563
$578
$981
$391
$1,783
$1,840
$721
$1,004
$2,324
$0
$273
$1,717
$728
Trucks
$0
$2,029
$896
$2,118
$0
$1,134
$2,031
$869
$440
$761
$576
$0
$1,032
$1,162
$931
$1,777
$2,272
$1,016
$1,251
$2,338
$0
$436
$1,747
$873
Fleet
$2,376
$1,880
$843
$2,072
$2,270
$942
$2,000
$840
$351
$491
$417
$1,563
$654
$1,041
$551
$1,781
$1,896
$787
$1,046
$2,331
$0
$332
$1,723
$776
       EPA then used OMEGA to calculate the costs of meeting the proposed standards in
the years 2021 and 2025, which are shown in Table 3.9-2 . EPA has accounted for the cost to
meet the standards in the reference case. In other words, Table 3.9-2 contains per-vehicle
costs that are incremental to the reference case costs shown in Table 3.9-1 .
                                        3-78

-------
                                       2017 Draft Regulatory Impact Analysis
Table 3.9-2 Control Case Costs for the Proposed Standards MY 2021 (2009$)
Company
Aston
Martin
BMW
Chrysler/Fiat
Daimler
Ferrari
Ford
Geely- Volvo
GM
Honda
Hyundai
Kia
Lotus
Mazda
Mitsubishi
Nissan
Porsche
Spyker-Saab
Subaru
Suzuki
Tata-JLR
Tesla
Toyota
VW
Fleet
2021 Costs
Cars
$6,424
$945
$569
$1,949
$6,351
$655
$2,035
$502
$467
$614
$483
$3,324
$924
$813
$759
$5,455
$3,335
$1,017
$1,160
$2,220
$0
$332
$1,624
$718
Trucks
$0
$915
$853
$956
$0
$776
$1,086
$680
$756
$884
$927
$0
$897
$998
$662
$1,328
$898
$922
$1,000
$1,648
$0
$713
$797
$764
Fleet
$6,424
$937
$698
$1,702
$6,351
$696
$1,741
$590
$556
$669
$582
$3,324
$919
$877
$729
$4,482
$2,986
$994
$1,132
$1,935
$0
$481
$1,457
$734
2021 Sales
Cars
1,058
359,098
421,013
300,378
7,059
1,401,617
92,726
1,564,277
1,198,880
613,355
331,319
278
274,740
65,851
912,629
36,475
21,294
230,780
95,725
58,677
28,623
1,903,706
585,607
10,505,165
Truck
-
128,724
348,613
99,449
-
714,181
41,768
1,530,020
535,916
156,466
95,432
-
59,227
35,309
408,029
11,242
3,560
72,773
20,767
58,153
-
1,215,539
148,734
5,683,902
Fleet
1,058
487,822
769,626
399,827
7,059
2,115,798
134,494
3,094,297
1,734,796
769,821
426,751
278
333,967
101,160
1,320,658
47,716
24,854
303,553
116,492
116,830
28,623
3,119,245
734,341
16,189,066
                                3-79

-------
Chapter 3
      Table 3.9-3 Control Case Costs for the Proposed Standards MY 2025 (2009$)
Company
Aston
Martin
BMW
Chrysler/Fiat
Daimler
Ferrari
Ford
Geely- Volvo
GM
Honda
Hyundai
Kia
Lotus
Mazda
Mitsubishi
Nissan
Porsche
Spyker-Saab
Subaru
Suzuki
Tata-JLR
Tesla
Toyota
VW
Fleet
2025
Cars
$6,862
$2,251
$1,914
$2,931
$7,109
$2,051
$3,228
$2,209
$1,452
$1,677
$1,442
$3,705
$2,196
$2,114
$1,997
$5,827
$4,001
$2,236
$2,307
$3,255
$0
$1,399
$2,618
$1,942
Trucks
$0
$1,959
$2,212
$1,952
$0
$2,463
$2,040
$1,834
$1,937
$1,988
$1,675
$0
$1,806
$2,171
$2,212
$2,054
$1,468
$2,087
$1,832
$2,653
$0
$1,631
$2,048
$1,954
Fleet
$6,862
$2,174
$2,043
$2,707
$7,109
$2,178
$2,876
$2,030
$1,595
$1,739
$1,491
$3,705
$2,131
$2,133
$2,060
$5,012
$3,670
$2,202
$2,225
$2,976
$0
$1,483
$2,506
$1,946
2025 Sales
Cars
1,182
405,256
436,479
340,719
7,658
1,540,109
101,107
1,673,936
1,340,321
677,250
362,783
316
306,804
73,305
1,014,775
40,696
23,130
256,970
103,154
65,418
31,974
2,108,053
630,163
11,541,558
Truck
-
145,409
331,762
101,067
-
684,476
42,588
1,524,008
557,697
168,136
97,653
-
61,368
36,387
426,454
11,219
3,475
74,722
21,374
56,805
-
1,210,016
154,284
5,708,900
Fleet
1,182
550,665
768,241
441,786
7,658
2,224,586
143,696
3,197,943
1,898,018
845,386
460,436
316
368,172
109,692
1,441,229
51,915
26,605
331,692
124,528
122,223
31,974
3,318,069
784,447
17,250,459
       EPA estimates that the additional technology required for manufacturers to meet the
GHG standards for this proposed rule will cost on average $734/vehicle and $l,946/vehicle in
the 2021 and 2025 MYs, respectively. These costs include our estimates of stranded capital
and costs associated with the A/C program from above.
                                        3-80

-------
                                             2017 Draft Regulatory Impact Analysis
      The OMEGA results project that under the primary proposal approximately 1% of the
vehicles sold in MYs 2017-2025 will be EVs or PHEVs.

                          Table 3.9-4 Sales by Technology
MY
2017
2018
2019
2020
2021
2022
2023
2024
2025
Total
Fraction
ICE Sales
14,940,135
14,648,056
14,575,393
14,795,940
14,991,075
14,804,015
14,573,553
14,385,507
14,214,379
131,928,053
90%
HEV Sales
840,896
878,510
928,488
998,265
1,068,478
1,417,930
1,773,810
2,146,396
2,535,818
12,588,590
9%
EV+PHEV
Sales
25,290
49,845
74,778
101,734
129,513
217,827
308,127
402,185
500,263
1,809,560
1%
Total Sales
15,806,322
15,576,410
15,578,658
15,895,939
16,189,066
16,439,772
16,655,489
16,934,087
17,250,459
146,326,204
100%
3.10Alternative Program Stringencies

 Table 3.10-1 Control Case Costs for the Alternative 1 (Trucks +20) Standards (2009$)
Company
Aston Martin
BMW
Chrysler/Fiat
Daimler
Ferrari
Ford
Geely-Volvo
GM
Honda
Hyundai
Kia
Lotus
Mazda
Mitsubishi
Nissan
Porsche
Spyker-Saab
Subaru
Suzuki
Tata-JLR
Tesla
Toyota
VW
Fleet
2021
Cars
$6,424
$553
$121
$1,344
$6,351
$371
$1,290
$99
$305
$528
$395
$3,324
$809
$491
$403
$4,929
$2,981
$790
$867
$688
$0
$233
$1,092
$436
Trucks
$0
$76
$490
$954
$0
$361
$953
$316
$665
$680
$791
$0
$536
$540
$602
$953
$805
$682
$1,000
$1,567
$0
$472
$797
$487
Fleet
$6,424
$427
$280
$1,255
$6,351
$368
$1,190
$202
$411
$559
$479
$3,324
$763
$507
$462
$4,070
$2,696
$766
$890
$1,097
$0
$320
$1,034
$453
2025
Cars
$6,862
$1,945
$1,247
$2,461
$7,109
$1,445
$2,515
$1,296
$1,249
$1,491
$1,234
$3,705
$1,918
$1,706
$1,674
$5,244
$3,630
$2,052
$2,147
$2,506
$0
$1,000
$2,197
$1,484
Trucks
$0
$1,320
$1,728
$1,952
$0
$2,180
$2,040
$1,418
$1,515
$1,580
$1,464
$0
$1,777
$1,865
$1,478
$2,054
$1,397
$1,486
$1,588
$2,143
$0
$1,366
$2,048
$1,580
Fleet
$6,862
$1,780
$1,455
$2,345
$7,109
$1,671
$2,374
$1,355
$1,327
$1,509
$1,282
$3,705
$1,895
$1,758
$1,616
$4,555
$3,338
$1,925
$2,051
$2,337
$0
$1,133
$2,168
$1,516
                                      3-81

-------
Chapter 3
  Table 3.10-2 Control Case Costs for the Alternative 2 (Trucks -20) Standards (2009$)
Company
Aston Martin
BMW
Chrysler/Fiat
Daimler
Ferrari
Ford
Geely-Volvo
GM
Honda
Hyundai
Kia
Lotus
Mazda
Mitsubishi
Nissan
Porsche
Spyker-Saab
Subaru
Suzuki
Tata-JLR
Tesla
Toyota
VW
Fleet
2021
Cars
$6,424
$1,511
$841
$2,579
$6,351
$1,010
$3,006
$1,121
$640
$815
$628
$3,324
$1,135
$1,358
$1,066
$6,182
$3,709
$1,434
$1,407
$2,800
$0
$494
$2,032
$1,055
Trucks
$0
$915
$1,498
$956
$0
$1,404
$1,086
$1,126
$1,043
$884
$988
$0
$1,000
$1,438
$793
$1,400
$898
$922
$1,194
$2,845
$0
$999
$914
$1,121
Fleet
$6,424
$1,354
$1,125
$2,208
$6,351
$1,131
$2,437
$1,123
$758
$829
$704
$3,324
$1,113
$1,384
$985
$5,148
$3,342
$1,319
$1,370
$2,821
$0
$678
$1,812
$1,077
2025
Cars
$6,862
$2,840
$2,570
$3,475
$7,109
$2,558
$4,181
$2,753
$1,854
$2,008
$1,635
$3,705
$2,440
$2,775
$2,561
$6,421
$4,250
$2,558
$2,561
$3,981
$0
$2,009
$3,072
$2,443
Trucks
$0
$1,959
$2,808
$1,952
$0
$2,923
$2,040
$3,013
$2,307
$1,988
$2,011
$0
$1,882
$2,171
$2,311
$2,054
$1,468
$2,087
$1,832
$3,563
$0
$2,023
$2,048
$2,501
Fleet
$6,862
$2,607
$2,673
$3,127
$7,109
$2,670
$3,546
$2,877
$1,987
$2,004
$1,715
$3,705
$2,347
$2,574
$2,487
$5,477
$3,887
$2,452
$2,436
$3,787
$0
$2,014
$2,871
$2,462
                                      3-82

-------
                                          2017 Draft Regulatory Impact Analysis
Table 3.10-3 Control Case Costs for the Alternative 3 (Cars +20) Standards (2009$)
Company
Aston Martin
BMW
Chrysler/Fiat
Daimler
Ferrari
Ford
Geely-Volvo
GM
Honda
Hyundai
Kia
Lotus
Mazda
Mitsubishi
Nissan
Porsche
Spyker-Saab
Subaru
Suzuki
Tata-JLR
Tesla
Toyota
VW
Fleet
2021
Cars
$6,424
-$240
$121
$592
$6,351
$278
$923
$109
$215
$304
$253
$2,114
$356
$284
$323
$3,732
$1,733
$574
$321
$688
$0
$201
$36
$244
Trucks
$0
-$154
$471
$481
$0
$156
$44
$394
$476
$374
$597
$0
$350
$409
$202
$953
$625
$151
$381
$1,567
$0
$466
$797
$390
Fleet
$6,424
-$218
$272
$567
$6,351
$240
$662
$245
$292
$318
$326
$2,114
$355
$326
$287
$3,131
$1,588
$478
$331
$1,097
$0
$298
$186
$292
2025
Cars
$5,348
$1,108
$1,242
$1,623
$6,292
$1,322
$1,946
$1,350
$913
$1,016
$776
$2,628
$1,251
$1,371
$1,323
$4,135
$2,431
$1,375
$1,341
$2,336
$0
$780
$1,230
$1,161
Trucks
$0
$1,320
$1,663
$1,591
$0
$1,246
$1,420
$1,456
$1,428
$1,156
$1,397
$0
$1,118
$1,501
$1,063
$2,054
$1,397
$1,486
$1,588
$2,143
$0
$1,334
$2,048
$1,394
Fleet
$5,348
$1,164
$1,424
$1,616
$6,292
$1,299
$1,790
$1,400
$1,064
$1,044
$908
$2,628
$1,229
$1,414
$1,246
$3,685
$2,296
$1,400
$1,383
$2,246
$0
$982
$1,391
$1,238
                                    3-83

-------
Chapter 3
   Table 3.10-4 Control Case Costs for the Alternative 4 (Cars -20) Standards (2009$)
Company
Aston Martin
BMW
Chrysler/Fiat
Daimler
Ferrari
Ford
Geely-Volvo
GM
Honda
Hyundai
Kia
Lotus
Mazda
Mitsubishi
Nissan
Porsche
Spyker-Saab
Subaru
Suzuki
Tata-JLR
Tesla
Toyota
VW
Fleet
2021
Cars
$6,424
$2,583
$866
$3,734
$6,351
$1,333
$4,111
$1,064
$909
$1,335
$1,049
$4,861
$2,064
$1,926
$1,485
$6,519
$5,406
$1,959
$2,276
$2,877
$0
$563
$3,267
$1,415
Trucks
$0
$915
$1,505
$1,025
$0
$1,878
$1,086
$1,017
$1,194
$1,440
$1,126
$0
$1,420
$1,555
$1,429
$1,678
$898
$1,724
$1,410
$2,845
$0
$1,099
$1,166
$1,275
Fleet
$6,424
$2,143
$1,142
$3,114
$6,351
$1,501
$3,215
$1,042
$993
$1,356
$1,066
$4,861
$1,957
$1,803
$1,469
$5,473
$4,817
$1,906
$2,128
$2,862
$0
$758
$2,854
$1,369
2025
Cars
$7,231
$3,956
$2,718
$4,693
$7,109
$3,235
$5,262
$2,689
$2,224
$2,901
$2,500
$4,812
$3,312
$3,184
$2,965
$7,428
$5,814
$3,091
$3,324
$4,291
$0
$2,140
$4,438
$2,923
Trucks
$0
$1,959
$2,808
$2,044
$0
$3,169
$2,040
$3,013
$2,777
$2,249
$2,064
$0
$3,117
$2,780
$2,938
$2,299
$1,575
$2,866
$3,032
$3,563
$0
$2,449
$2,219
$2,760
Fleet
$7,231
$3,428
$2,757
$4,087
$7,109
$3,214
$4,307
$2,843
$2,387
$2,771
$2,408
$4,812
$3,279
$3,050
$2,957
$6,320
$5,261
$3,040
$3,274
$3,953
$0
$2,252
$4,001
$2,869
3.11 Comparative cost of advanced technologies under credit scenarios

       As part of the analysis of the flexibility programs, EPA calculated an illustrative
example of the relative cost-effectiveness of certain advanced technologies.

       Table 3.11-1 shows the cost per gram per mile of going from the 2016 type
technologies to MY 2021 technologies.  Note that in all cases, the advanced technologies are
significantly more expensive than the average costs per vehicle from the OMEGA, even when
considering the impacts of the incentives.

                Table 3.11-1 Gram/mile cost of advanced technologies



Reference
Case CO2
MY 2021
CO2
(Proposed)

Delta
g/mile

Delta
CostA

$per
g/mile
                                        3-84

-------
                                               2017 Draft Regulatory Impact Analysis
OMEGA projection of
average 2021 Car in
Proposal Case
EV100 (45 sqft, VT 3,
no multiplier)
EV100 (45 sqft, VT 3,
1.5 multiplier)

OMEGA projection of
average 2021 Truck in
Proposal Case
HEV (65 sqft, VT 13,
no credit)
HEV (65 sqft, VT 13,
20 g credit)

225

263
263


297

344

344

178

0
0


239

243

223

47

263
395


58

101

121

$625

$16,066
$16,066


$654

$6,264

$6,264

$13

$61
$41


$11

$62

$52
ANote that we use average reference case cost of $704 for cars and $858 for trucks, not the vehicle specific cost.
If these vehicles reference case costs were higher than average, then their costs under the proposal would be less,
and conversely if their costs were lower than averages, then their compliance costs would be greater.

       The reference case CC>2 values are determined in the case of the OMEGA projections,
from the actual OMEGA runs, and in the case of the 45 and 65 square foot vehicles from the
applicable GHG curve. In this table, the EV is assumed to have a compliance value of zero
grams per mile without the multiplier incentive. For the incentive, we simply multiplied the
delta gram per mile by 1.5.  This overstates the impact of the credit, because the multiplier
would also increase the number of vehicles in a manufacturer's fleet by 1.5.  The cost per
gram/mile is actually greater than shown in this illustrative table because the size of the fleet
impacts the benefit of the multiplier. The HEV in this example has an effectiveness of 51.4%
relative to a baseline (no technology) vehicle with a CO2 of 500 g/mile.

       HEVs and EVs, regardless of their cost-effectiveness, are more effective than the
conventional technologies, and retain that advantage. Further in MY 2025, when the average
cost per gram/mile is higher, these technologies are more cost effective.
3.12 How Many of Today's Vehicles Can Meet or Surpass the Proposed MY 2017-2025
       CO2 Footprint-based Targets with Current Powertrain Designs?

       As part of its evaluation of the feasibility of the proposed standards , EPA evaluated
all MY2011 and MY2012 vehicles sold in the U.S. today against the proposed CO2 footprint-
based standard curves to determine which of these vehicles would meet or be lower than the
proposed MY 2017 -MY 2025 footprint-based CO2 targets assuming air conditioning credit
generation consistent with today's proposal. Under the proposed 2017-2025 greenhouse gas
emissions standards, each vehicle will have a unique CO2 target based  on the vehicle's
footprint (with each manufacturer having its own unique fleetwide standard )) .   In this
analysis, EPA assumed air conditioner credits because air conditioner improvements are
considered to be among the cheapest and easiest technologies to reduce greenhouse gas
emissions, manufacturers are already investing in air conditioner improvements, and air
conditioner changes do not impact engine, transmission,  or aerodynamic designs so assuming
such credits does not affect consideration of cost and leadtime for use of these other
                                         3-85

-------
Chapter 3

technologies. EPA applied increasing air conditioner credits over time with a phase-in of
alternative refrigerant for the generation of HFC leakage reduction credits consistent with the
assumed phase-in schedule discussed in Preamble Section III.C.  No adjustments were made
to vehicle CO2 performance other then this assumption of air conditioning credit generation.
Under this analysis, a wide range of these existing vehicles would meet the MY2017 proposed
CO2 targets, and a few meet even the proposed MY2025  CO2 targets.

       Using publicly available dataTT, EPA compiled a list of all available vehicles and their
2-cycle CC>2 g/mile performance (that is, the performance over the city and highway
compliance tests).  Data is currently available for all MY2011 vehicles and some MY2012
vehicles. EPA gathered vehicle footprint data from EPA  reports,23 manufacturer submitted
CAFE reports, and manufacturer websites. .

       Table 3.12-1  shows that a significant number of vehicles sold today would meet or be
lower than the proposed footprint-based CO2 targets with current powertrain designs,
assuming air conditioning credit generation consistent with our  proposal. The table
highlights the vehicles with CO2 emissions that meet or are lower than the applicable
proposed footprint targets from MY 2017 to 2025 in green,  and shows the percentage below
the proposed target for each year.   The list of vehicles includes midsize cars, minivans, sport
utility vehicles, compact cars, and small pickup trucks - all  of which meet the proposed MY
2017 target values with no technology improvements other then air conditioning system
upgrades. These vehicles utilize a wide variety of powertrain technologies, including internal
combustion, hybrid-electric, plug-in hybrid-electric, and full electric, and operate on a variety
of different fuels including gasoline,  diesel, electricity, and compressed natural gas. Nearly
every major manufacturer produces some vehicles that would meet or be lower than the
proposed MY2017 footprint CO2 target with only simple improvements in air conditioning
systems.

       Vehicles that are above,  but within 5%,  of the proposed targets are highlighted  in
yellow. This list also includes vehicles from multiple classes, including large cars  and
standard  pickup trucks. Four versions of the F-150 pickup truck are within 5% of the
proposed targets through  at least 2021. This includes two engine options (the 3.7L V6 and the
3.5L V6), and three wheelbase options1111.

       EPA also receives projected sales data prior to each  model year from each
manufacturer.  Based on this data, approximately 7% of MY2011 sales will be vehicles that
would meet or be better than the proposed MY 2017 targets for those vehicles, requiring only
improvements in air conditioning systems. In addition, nearly 30% of projected MY2011
sales would be within 10% of the proposed MY2017 footprint CO2 target with only simple
improvements to air conditioning systems, a full six model years before the proposed standard
would take effect.
TT www.fueleconomy.gov
 J The F-150 engine and v
available. Not all possible engine and wheelbase combinations are produced.
1111 The F-150 engine and wheelbase combinations listed in Table 3.12-1 correspond to models that are currently
                                         3-86

-------
                                               2017 Draft Regulatory Impact Analysis
       With improvements to air conditioning systems, the most efficient gasoline internal
combustion engines would meet the MY 2020 proposed footprint targets.  After MY2020, the
only current vehicles that continue to meet the proposed footprint-based CO2 targets
(assuming improvements in air conditioning) are hybrid-electric, plug-in hybrid-electric, and
fully electric vehicles.  However, the proposed MY 2020 standards will not be in effect for
another nine years. EPA expects that gasoline vehicles will continue to improve in that
timeframe and will be able to meet the standard (using the technologies discussed in Chapter
3 of the draft Joint TSD and as discussed in Preamble Section HID) assuming air conditioner
improvements . Today's Toyota Prius, Ford Fusion Hybrid, Chevrolet Volt, Nissan Leaf,
Honda Civic Hybrid, and Hyundai Sonata Hybrid all meet or surpass the proposed footprint-
based CO2 targets through MY2025.  In fact, the current Prius, Volt, and Leaf meet the
proposed 2025 CO2 targets without air conditioning credits.

       This assessment of MY2011 and MY2012 vehicles also makes clear that substantial
additional technology penetration across the fleet, and lead time in which to do so, is needed
for manufacturers to meet the proposed standards.  Notably, based on the OMEGA modeling,
we project that the MY2017-2025 standards can primarily be achieved by advanced gasoline
vehicles - for example, in MY2025, we project more than 80 percent of the new vehicles
could be advanced gasoline powertrains. The assessment of MY2011 and MY2012 vehicles
available in the market today indicates advanced gasoline vehicles (as well as diesels) can
achieve the targets for the early model years of the proposed standards (i.e., model years
2017-2020) with only improvements in air conditioning systems. However, significant
improvements in technologies are needed and penetrations of those technologies must
increase substantially in order for individual manufacturers (and the  fleet overall) to achieve
the proposed standards for the early years of the program,  and certainly for the later years
(i.e., model years 2021-2025). These technology improvements include: gasoline direct
injection fuel systems;  downsized and turbocharged gasoline engines (including in some cases
with the application of cooled  exhaust gas recirculation); continued improvements in engine
friction reduction and low friction lubricants; transmissions with an increased number of
forward gears (e.g., 8 speeds); improvements in transmission shifting logic; improvements in
transmission gear box efficiency; vehicle mass reduction; lower rolling resistance tires, and
improved vehicle aerodynamics. In many (though not all) cases these technologies are
beginning to penetrate the U.S. light-duty vehicle market.
       In general, these technologies must go through the automotive product development
cycle in order to be introduced into the U.S. fleet, and in some cases additional research is
needed before the technologies CC>2 benefits can be fully realized and large-scale
manufacturing can be achieved.  This topic is  discussed in more detail in Chapter 3.5 of the
draft Joint Technical Support Document.  In that Chapter, we explain that many CO2
reducing technologies should be able to penetrate the new vehicle market at high levels
between now and MY2016, there are also many of the key technologies we project as being
needed to achieve the proposed 2017-2025 standards which will only be able to penetrate the
market at relatively low levels (e.g., a maximum level of 30%) or less by MY2016, and which
even by MY2021 will still be constrained.  These include important powertrain technologies
                                        3-87

-------
Chapter 3

such as 8-speed transmissions and second or third generation downsized engines with
turbocharging,

             The majority of these technologies must be integrated into vehicles during the
product redesign schedule, which is typically on a  5-year cycle. EPA discussed in the
MY2012-2016 rule the significant costs and potential risks associated with requiring major
technologies to be added in-between the typical 5-year vehicle redesign schedule, (see 75 FR
at 25467-68,).  In addition, engines and transmissions generally have longer lifetimes then 5
years, typically on the order of 10 years or more. Thus major powertrain technologies
generally take longer to penetrate the new vehicle fleet then can be done in a 5-year redesign
cycle. As detailed in Chapter 3.5 of the draft Joint TSD, EPA projects that 8-speed
transmissions could increase their maximum penetration in the fleet from 30% in MY2016 to
80% in 2021 and to 100% in MY2025. Similarly,  we project that second generation
downsized and turbocharged engines (represented  in our assessment as engines with a brake-
mean effective pressure of 24 bars) could penetrate the new vehicle fleet at a maximum  level
of 15% in MY2016, 30% in MY2021, and 75% in  MY2025.  When coupled with the typical 5
year vehicle redesign schedule, EPA project that is not possible for all of the advanced
gasoline vehicle technologies we have assessed to  penetrate the fleet in a single 5 year vehicle
redesign schedule.

       Given the status of the technologies we project to be used to achieve the proposed
MY2017-2025 standards and the product development and introduction process which is
fairly standard in the automotive industry today, our assessment of the MY2011 and MY2012
vehicles in comparison to the proposed standards supports our overall feasibility assessment,
and reinforces our assessment of the lead time needed for the industry to achieve the proposed
standards.
23 EPA's "Light Duty Automotive Technology, Carbon Dioxide Emissions, and Fuel
Economy Trends Report, 1975 through 2010"

-------
                                                                2017 Draft Regulatory Impact Analysis
Table 3.12-1 Vehicles that Meet or Exceed Proposed Targets With Current Powertrain Designs
Model
Year
2011
2011
2011
2011
2011
2012
2011
2012
2012
2011
2011
2011
2011
2011
2011
2011
2011
2011
2011
2011
2011
2012
2011
2011
2011
2012
2012
2011
2011
2012
2011
2012
2012
2011
2011
2012
2011
2012
2012
2012
2012
2011
2011
2012
2011
2011
2011
2011
2011
2012
2012
2012
2012
Manufacturer
Mercedes-Benz
Mercedes-Benz
Nissan
Chevrolet
Toyota
Honda
Hyundai
Ford
Lincoln
Lexus
Honda
Toyota
Lexus
Honda
Chevrolet
CMC
Chevrolet
CMC
Nissan
Toyota
Lexus
Ford
Honda
Mercedes-Benz
Mercedes-Benz
Hyundai
Hyundai
Toyota
Toyota
Chevrolet
Lexus
Ford
Honda
Honda
Ford
Ford
Toyota
Hyundai
Infiniti
Honda
Ford
Honda
Ford
Fiat
Cadillac
Chevrolet
CMC
Chevrolet
CMC
Chevrolet
Volkswagen
Honda
Ford
Vehicle
Smart fortwo (cabriolet)
Smart fortwo (coupe)
LEAF
VOLT
PRIUS
CIVIC HYBRID
SONATA HYBRID
FUSION HYBRID FWD
MKZ HYBRID FWD
CT200h
INSIGHT
HIGHLANDER HYBRID4WD
RX 450h AWD
CIVIC CNG
SILVERA DO 2WD HYBRID
SIERRA 2WD HYBRID
SILVERA DO 4WD HYBRID
SIERRA 4WD HYBRID
ALTIMA HYBRID
CAMRY HYBRID
HS250h
ESCAPE HYBRID AWD
CR-Z
Smart fortwo (cabriolet)
Smart fortwo (coupe)
ELANTRA
ELANTRA
TACOMA 2WD
SIENNA
CRUZE ECO
RX450h
Focus SFE FWD
CIVIC HF
ODYSSEY 2WD
RANGER 2WD
ESCAPE HYBRID FWD
TACOMA 2WD
ACCENT
M35h
CIVIC
FOCUS FWD
CR-Z
Fiesta SFE FWD
500
ESCALADE 2WD HYBRID
TAHOE2WD HYBRID
YUKON 2WD HYBRID
TAHOE4WD HYBRID
YUKON 4WD HYBRID
CRUZE ECO
Passat
CIVIC
FOCUS FWD
Unadjusted
Fuel Economy
(mpg|
123.9
123.9
141.7
48.4
70.8
63.1
52.2
54.2
54.2
57.5
57.3
38.7
38.6
37.5
28.5
28.5
28.4
28.4
46.7
45.9
47.3
39.0
50.1
49.5
49.5
44.7
44.4
30.2
29.4
44.4
40.4
43.6
44.3
29.0
31.2
44.1
28.3
45.5
38.8
43.0
42.1
44.9
44.9
44.5
28.5
28.5
28.5
28.4
28.4
40.9
46.4
41.8
41.1
Tailpipe CO2
(ft2)
0.0
0.0
0.0
56.0
125.6
140.9
170.3
164.0
164.0
154.6
155.1
229.7
230.4
175.7
311.4
311.4
313.2
313.2
190.3
193.4
188.0
227.6
177.3
179.5
179.5
198.7
200.2
294.5
302.0
200.3
220.2
203.7
200.6
306.7
284.5
201.4
313.6
195.1
229.1
206.7
211.0
197.9
198.0
199.6
311.4
311.4
311.4
313.2
313.2
217.1
219.5
212.4
216.1
Footprint
(ft2)
26.8
26.8
44.7
45.3
44.2
43.5
48.0
45.6
45.6
42.6
40.8
48.8
48.6
43.4
67.3
67.3
67.3
67.3
46.3
46.9
44.5
43.3
39.5
26.8
26.8
45.2
45.2
55.9
56.1
44.8
48.6
44.2
43.4
55.9
50.6
43.3
55.9
41.7
49.1
43.4
44.2
39.5
39.3
34.7
54.8
54.8
54.8
54.8
54.8
44.8
45.3
43.4
44.2
Powertrain
Type
EV
EV
EV
PHEV
HEV
HEV
HEV
HEV
HEV
HEV
HEV
HEV
HEV
CNG
HEV
HEV
HEV
HEV
HEV
HEV
HEV
HEV
HEV
Gasoline
Gasoline
Gasoline
Gasoline
Gasoline
Gasoline
Gasoline
HEV
Gasoline
Gasoline
Gasoline
Gasoline
HEV
Gasoline
Gasoline
HEV
Gasoline
Gasoline
HEV
Gasoline
Gasoline
HEV
HEV
HEV
HEV
HEV
Gasoline
Diesel
Gasoline
Gasoline
Transmission
Al
Al
Al
CVT
CVT
A5
A6
CVT
CVT
CVT
CVT
CVT
CVT
A5
CVT
CVT
CVT
CVT
CVT
CVT
CVT
CVT
CVT
A5
A5
M6
A6
M5
A6
M6
CVT
A6
A5
A6
M5
CVT
A4
A6
A7
A5
A6
M6
A6
M5
CVT
CVT
CVT
CVT
CVT
A6
M6
M5
A6
Engine
Displacement
(L)
n/a
n/a
n/a
1.4
1.8
1.5
2.4
2.5
2.5
1.8
1.3
3.5
3.5
1.8
6.0
6.0
6.0
6.0
2.5
2.4
2.4
2.5
1.5
1.0
1.0
1.8
1.8
2.7
2.7
1.4
3.5
2.0
1.8
3.5
2.3
2.5
2.7
1.6
3.5
1.8
2.0
1.5
1.6
1.4
6.0
6.0
6.0
6.0
6.0
1.4
2.0
1.8
2.0
Vehicle Class
Two Seaters
Two Seaters
Midsize Cars
Compact Cars
Midsize Cars
Compact Cars
Midsize Cars
Midsize Cars
Midsize Cars
Compact Cars
Compact Cars
Sport UtilityVehide
Sport Utility Vehicle
SubcompactCars
Standard Pick-up Trucks
Standard Pick-up Trucks
Standard Pick-up Trucks
Standard Pick-up Trucks
Midsize Cars
Midsize Cars
Compact Cars
Sport Utility Vehicle
Two Seaters
Two Seaters
Two Seaters
Midsize Cars
Midsize Cars
Small Pick-up Trucks
Minivan
Midsize Cars
Sport Utility Vehicle
Compact Cars
Compact Cars
Minivan
Small Pick-up Trucks
Sport Utility Vehicle
Small Pick-up Trucks
Compact Cars
Midsize Cars
Compact Cars
Compact Cars
Two Seaters
SubcompactCars
MinicompactCars
Sport UtilityVehide
Sport Utility Vehicle
Sport Utility Vehicle
Sport Utility Vehicle
Sport Utility Vehicle
Midsize Cars
Midsize Cars
Compact Cars
Compact Cars
Carl
Truck
C
C
C
C
C
C
C
C
C
C
C
T
T
C
T
T
T
T
C
C
C
T
C
C
C
C
C
T
T
C
C
C
C
T
T
C
T
C
C
C
C
C
C
C
T
T
T
T
T
C
C
C
C
Compliance
2017
100%
100%
100%
76%
46%
38%
30%
30%
30%
30%
27%
21%
21%
21%
14%
14%
13%
13%
19%
18%
17%
13%
15%
14%
14%
13%
12%
9%
7%
11%
9%
9%
9%
5%
4%
8%
3%
8%
6%
6%
5%
5%
5%
4%
2%
2%
2%
1%
1%
4%
3%
3%
3%
2018
100%
100%
100%
76%
44%
35%
28%
27%
27%
27%
24%
20%
19%
17%
14%
14%
13%
13%
15%
15%
13%
12%
12%
11%
11%
9%
9%
7%
5%
8%
6%
5%
5%
3%
2%
4%
1%
4%
2%
2%
1%
1%
1%
0%
-1%
-1%
-1%
-1%
-1%
-1%
-1%
-1%
-2%
2019
100%
100%
100%
75%
42%
33%
24%
24%
24%
24%
21%
18%
18%
14%
14%
14%
14%
14%
12%
11%
9%
10%
8%
7%
7%
5%
4%
5%
3%
3%
1%
0%
0%
1%
0%
0%
-1%
-1%
-2%
-3%
-4%
-4%
-4%
-5%
-2%
-2%
-2%
-3%
-3%
-5%



2020
100%
100%
100%
73%
40%
30%
21%
21%
21%
21%
17%
16%
16%
10%
13%
13%
12%
12%
8%
7%
5%
8%
4%
2%
2%
1%
0%
3%
1%
-1%
-3%
-4%
-4%
-1%
-2%
-5%
-4%
-5%






-5%
-5%
-5%






2021
100%
100%
100%
72%
37%
27%
18%
17%
17%
17%
13%
10%
10%
6%
7%
7%
6%
6%
3%
3%
1%
1%
-1%
-2%
-2%
-4%
-5%
-4%

























2022
100%
100%
100%
71%
34%
23%
14%
13%
13%
13%
9%
6%
5%
1%
2%
2%
1%
1%
-1%
-2%
-4%
-4%































2023
100%
100%
100%
69%
31%
20%
10%
9%
9%
9%
5%
1%
0%
-3%
-3%
-3%
-3%
-3%



































2024
100%
100%
100%
68%
28%
16%
6%
5%
5%
5%
0%
-4%
-5%








































2025
100%
100%
100%
66%
24%
12%
1%
0%
0%
0%
-4%










































                                         5-89

-------
Chapter 3
Model
Year
2011
2012
2011
2012
2011
2012
2012
2011
2011
2011
2011
2011
2011
2011
2012
2011
2012
2011
2012
2011
2011
2011
2011
2012
2011
2012
2012
2011
2011
2012
2011
2012
2011
2012
2012
2011
2012
2011
2011
2011
2012
2011
2012
2012
2011
2012
2011
2012
2011
2011
2012
2011
2011
2012
Manufacturer
Ford
Buick
Kia
Chevrolet
Mini
Chevrolet
Buick
Ford
Ford
Ford
Ford
Cadillac
CMC
Mercedes-Benz
Ford
Mini
Volkswagen
Ford
Volkswagen
Toyota
Nissan
Nissan
Mazda
Ford
Toyota
Kia
Kia
Suzuki
Toyota
Volkswagen
Honda
Nissan
Toyota
Ford
Hyundai
Kia
Hyundai
Kia
Toyota
Kia
Chevrolet
Mini
Volkswagen
Volkswagen
Kia
Volkswagen
Kia
Audi
Nissan
Hyundai
Kia
Mini
Mini
Audi
Vehicle
Fiesta FWD
LACROSSE
FORTE ECO
CRUZE
Mini Cooper
CRUZE
REGAL
F150 PICKUP 2WD
F150 PICKUP 2WD
F150 PICKUP 2WD
F150 PICKUP 2WD
ESCALADE 4WD HYBRID
YUKON DENALI HYBRID 4WD
ML450 HYBRID 4MATIC
TRANSITCONNECTFWD
Mini CooperCountryman
Jetta
Fiesta FWD
Jetta
VENZAAWD
QUEST
FRONTIER 2WD
MAZDA2
Transit Connect Van
SIENNA
SORENTO 4WD
SPORTAGE 4WD
EQUATOR 2WD
YARIS
Passat
FIT
SENTRA
COROLLA
FOCUS FWD
SONATA
OPTIMA
SONATA
FORTE
YARIS
OPTIMA
CRUZE
Mini Cooper
GOLF
GOLF
RIO
JETTA SPORTWAGEN
FORTE KOUP
A6
VERSA
ENTOURAGE
SEDONA
Mini Clubman
Mini Convertible
A3
Unadjusted
Fuel Economy
(mpg|
44.0
38.1
40.7
40.4
43.6
40.1
38.1
23.9
23.9
23.9
24.4
28.0
28.0
29.6
31.1
41.0
46.1
42.7
46.1
30.2
27.2
27.4
42.6
30.5
26.7
30.6
31.0
27.3
42.6
44.6
42.5
39.5
41.0
39.4
36.5
36.5
36.5
38.9
41.9
36.3
38.5
41.7
46.1
46.1
41.1
46.1
38.3
35.4
40.8
26.8
26.8
41.0
41.0
46.1
Tailpipe CO2
(ft2)
202.2
233.3
218.3
219.8
203.9
221.7
233.3
372.3
372.3
372.3
363.8
317.4
317.4
300.4
286.0
216.8
220.7
208.1
220.9
294.3
326.7
324.8
208.6
291.8
333.0
290.8
286.9
325.3
208.7
228.2
208.9
224.9
217.0
225.4
243.3
243.8
243.6
228.4
212.2
245.1
230.5
213.3
220.7
220.9
216.0
220.9
232.1
251.0
217.9
331.6
331.6
216.8
216.8
220.7
Footprint
(ft2)
39.3
48.0
44.5
44.8
38.8
44.8
46.8
75.9
72.8
67.2
67.2
54.8
54.8
51.0
47.9
43.0
43.8
39.3
43.8
48.8
55.9
54.8
39.4
47.9
56.1
47.1
46.0
54.0
39.9
45.3
39.9
44.3
42.5
44.2
48.0
48.1
48.0
44.5
39.9
48.1
44.8
38.8
42.4
42.4
41.3
42.3
44.6
48.6
41.4
54.8
54.7
40.1
38.8
41.8
Powertrain
Type
Gasoline
Gasoline
Gasoline
Gasoline
Gasoline
Gasoline
Gasoline
Gasoline
Gasoline
Gasoline
Gasoline
HEV
HEV
HEV
Gasoline
Gasoline
Gasoline
Gasoline
Diesel
Gasoline
Gasoline
Gasoline
Gasoline
Gasoline
Gasoline
Gasoline
Gasoline
Gasoline
Gasoline
Diesel
Gasoline
Gasoline
Gasoline
Gasoline
Gasoline
Gasoline
Gasoline
Gasoline
Gasoline
Gasoline
Gasoline
Gasoline
Diesel
Diesel
Gasoline
Diesel
Gasoline
Gasoline
Gasoline
Gasoline
Gasoline
Gasoline
Gasoline
Diesel
Transmission
A6
A6
A6
M6
M6
A6
A6
A6
A6
A6
A6
CVT
CVT
CVT
M5
M6
A6
A6
CVT
A4
M6
M5
M5
A4
A6
A6
A6
M5
M5
A6
A5
CVT
M5
M5
A6
M6
M6
A6
A4
A6
M6
A6
A6
M6
M5
M6
A6
CVT
CVT
A6
A6
M6
M6
A6
Engine
Displacement
(L)
1.6
2.4
2.0
1.4
1.6
1.4
2.4
3.5
3.5
3.5
3.7
6.0
6.0
3.5
1.6
1.6
2.0
2.7
3.5
2.0
2.0
2.5
1.5
2.0
3.5
2.4
2.4
2.5
1.5
2.0
1.5
2.0
1.8
2.0
2.4
2.4
2.4
2.0
1.5
2.4
1.8
1.6
2.0
2.0
1.6
2.0
2.0
2.0
1.8
3.5
3.5
1.6
1.6
2.0
Vehicle Class
SubcompactCars
vlidsize Cars
vlidsize Cars
vlidsize Cars
vlinicompactCars
vlidsize Cars
vlidsize Cars
Standard Pick-up Trucks
Standard Pick-up Trucks
Standard Pick-up Trucks
Standard Pick-up Trucks
Sport UtilityVehide
Sport Utility Vehicle
Sport Utility Vehicle
Special Purpose Vehicle
Compact Cars
Compact Cars
SubcompactCars
Compact Cars
Sport UtilityVehide
vlinivan
Small Pick-up Trucks
Compact Cars
vans, Cargo Types
vlinivan
Sport UtilityVehide
Sport Utility Vehicle
Small Pick-up Trucks
SubcompactCars
vlidsize Cars
Small Station Wagons
vlidsize Cars
Compact Cars
Compact Cars
arge Cars
vlidsize Cars
arge Cars
vlidsize Cars
SubcompactCars
vlidsize Cars
vlidsize Cars
vlinicompactCars
Compact Cars
Compact Cars
Compact Cars
Small Station Wagons
Compact Cars
vlidsize Cars
vlidsize Cars
vlinivan
vlinivan
SubcompactCars
vlinicompactCars
Small Station Wagons
Car/
Truck
C
C
C
C
C
C
C
T
T
T
T
T
T
T
T
C
C
C
C
T
T
T
C
T
T
T
T
T
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
T
T
C
C
C
Compliance
2017
3%
3%
2%
2%
2%
1%
0%
-4%
-4%
-4%
-2%
0%
0%
-1%
-1%
0%
0%
0%
0%
-2%
-1%
-3%
-1%
-3%
-3%
-4%
-5%
-4%
-1%
-1%
-1%
-1%
-1%
-2%
-2%
-2%
-2%
-2%
-2%
-2%
-3%
-3%
-3%
-4%
-4%
-4%
-4%
-4%
-4%
-5%
-5%
-5%
-5%
-5%
2018
-2%
-2%
-2%
-2%
-3%
-3%
-4%
-4%
-4%
-4%
-1%
-3%
-3%
-3%
-3%
-5%
-5%
-5%
-5%
-5%
-4%
-5%
-5%































2019







-4%
-4%
-4%
-1%
-4%
-4%
-5%
-5%







































2020







-4%
-4%
-4%
-1%











































2021







-4%
-4%
-5%
-3%











































2022







-4%
-5%













































2023






















































2024






















































2025






















































                                                      5-90

-------
                                               2017 Draft Regulatory Impact Analysis
3.13 Analysis of Ferrari & Chrysler/Fiat

       Note that in the primary analyses, Ferrari is shown as a separate entity, but in this side-
analysis, is combined with other Fiat-owned companies for purposes of GHG compliance.
Ferrari could be combined with other Fiat-owned companies for purposes of GHG
compliance at the manufacturer's discretion. We conducted an OMEGA run to evaluate a
scenario where Ferrari's compliance would be included with other Fiat-owned companies,
including Chrysler.  Unlike Ferrari under the scenario in which Ferrari was modeled as a
stand-alone company, Chrysler/Fiat would comply, even with the Ferrari vehicles included.
Also note that in Section III.B., EPA is requesting comment on the concept of allowing
companies  that are able to demonstrate "operational independence" to be eligible for SVM
alternative  standards.  If EPA were to adopt such provisions, and Ferrari were to qualify, they
would likely petition for an alternative standard under the proposed SVM provisions, rather
than comply as part  of Chrysler/Fiat.

       Under the MY 2025 OMEGA projections,  Ferrari falls short of its 2025 target (150
grams/mile CO2) by nine grams. vv Under this scenario, Ferrari would produce a fleet
consisting of almost entirely HEVs (50%), EVs (23%) and PHEVs (22%) with a MY 2025
compliance cost of approximately $7,100 relative to the MY 2016 standards.

       If Ferrari is included in the Chrysler/Fiat GHG compliance fleet, Chrylser/Fiat's
starting 2008 CO2 is 2 grams higher (347.6 vs. 345.6).  As a result, the cost of complying with
the reference case standards would increase by approximately $65, and the cost of complying
with the proposed standards would increase by $91 for a net average increase in MY 2025
compliance costs of $36 per vehicle for Chrysler/Fiat.  Net program costs would not change
significantly.
3.14Cost Sensitivities

       3.14.1  Overview

       We have conducted several sensitivity analyses on a variety of input parameters. For
the analyses presented in and have run the OMEGA model to generate 2025MY results for
each of these sensitivities. We have looked at different levels of mass reduction costs, battery
pack costs, indirect cost multipliers, and learning rates.  These sensitivities are summarized in
vv Assuming that Ferrari complied with the primary proposed standards.


                                         3-91

-------
Chapter 3

       Table 3.14-1 , with the summarized results in

       Table 3.14-10  .  Additional sensitivities with regard to benefits are shown in DRIA
Chapter 4.
                                        3-92

-------
                                               2017 Draft Regulatory Impact Analysis
                      Table 3.14-1  Summary of Cost Sensitivities
Sensitivity parameter
Mass reduction direct
manufacturing costs
Battery pack direct
manufacturing costs
Indirect cost multipliers
Learning ratesa
Low side sensitivity
40% lower
10% lower for P2 HEVs
20% lower for PHEV/EV
Low side of 95% confidence
interval of modified Delphi
survey results
P-value of 30% on steep
portion of the curve; cost
reductions of 4%/3%/2%
per year for each 5 year
increment on the flat portion
of the learning curve
High side sensitivity
40% higher
10% higher for P2 HEVs
20% higher for PHEV/EV
High side of 95%
confidence interval of
modified Delphi survey
results
P-value of 10% on steep
portion of the curve; cost
reductions of 2%/l%/0%
per year for each 5 year
increment on the flat portion
of the learning curve
a Higher learning rates results in lower costs, hence the low side sensitivity uses the higher learning rates
while the high side sensitivity uses the lower learning rates.
       3.14.2  Mass Sensitivity

       For the mass reduction cost sensitivity, we adjusted the mass reduction DMC cost
equation by +/-40%.  That cost equation is shown in Table 3.14-2 along with the cost equation
used for each side of the mass reduction cost sensitivity.
                    Table 3.14-2 Mass Reduction Cost Sensitivities
Sensitivity parameter
Low side
Primary case
High side
Mass reduction DMC equation used
DMC=$2.60x, where x=% mass reduction
DMC=$4.33x, where x=% mass reduction
DMC=$6.06x, where x=% mass reduction
       We did not re-rank OMEGA packages for the mass reduction cost sensitivities but
rather used the same input files used for our primary analysis with new mass costs.  This
should have no impact on the results other than making them conservative since re-ranking
packages would serve to move the modeling to more cost effective technologies, thus
reducing the cost impact of this sensitivity. Because mass reduction is a cost effective
technology, even with higher costs, OMEGA would still choose a similar degree of mass
reduction given the stringency of the MY 2025 standards. By contrast, even with lower costs,
mass reduction would still be limited by the fatality analysis. As a result, the mass reduction
sensitivity does not have any impact on the relative ranking of packages or the engine and
hybridization technologies that would be added in response to our proposed standards.

       The high mass cost inputs increased the average compliance costs of the reference
case by $31 and the control case by $133, for a net average per-vehicle cost increase of $102
                                         3-93

-------
 Chapter 3

 in MY 2025. The low mass cost inputs decreased  the  average compliance costs of the
 reference case by $31 and the control case by $133 for a net average cost decrease of $102
 These impacts would be greater on manufacturers that use more mass reduction technology,
 and less on those that use less.

        3.14.3 Battery Sensitivity

        For the battery pack cost sensitivities, we decreased/increased the battery pack DMCs
 by the amounts shown in Table 3.14-3. As presented in Chapter 3 of the draft joint TSD, we
 have developed linear regressions for our battery pack costs.  These linear regressions provide
 battery pack DCM as a function of net weight reduction of the vehicle.  Table 3.14-3  and
 Table 3.14-5  show the linear regressions used for our low side  and high side sensitivity
 analyses, respectively, while Table 3.14-4 presents the linear regressions used for our primary
 analysis (as presented in Chapter 3 of the draft joint TSD).

   Table 3.14-3 Linear Regressions of Battery Pack Direct Manufacturing Costs  vs Net
                  Weight Reduction used for Low Side Sensitivity (2009$)
Vehicle Class
Subcompact
Small car
Large car
Minivan
Small truck
Minivan+towing
Large truck
P2HEV
-$159x+$645
-$196x+$682
-$270x+$777
-$265x+$763
-$249x+$740
-$265x+$836
-$285x+$868
PHEV20
-$690x+$2,082
-$798x+$2,197
-$l,255x+$2,664
-$l,152x+$2,637
-$l,071x+$2,514


PHEV40
-$694x+$2,919
-$l,675x+$3,229
-$2,521x+$4,153
-$l,672x+$4,028
-$l,955x+$3,829


EV75
-$l,080x+$4,339
-$l,626x+$4,710
-$2,768x+$5,738
-$2,784x+$5,761
-$2,518x+$5,463


EV100
-$l,651x+$5,088
-$2,279x+$5,603
-$3,215x+$6,481
-$2,472x+$6,731
-$2,377x+$6,436


EV150
-$l,615x+$6,633
-$2,480x+$7,350
-$3,016x+$8,791
-$3,653x+$9,397
$9,003


Notes:
"x" in the equations represents the net weight reduction as a percentage, so a subcompact P2 HEV battery pack
with a 20% applied weight reduction and, therefore, a 15% net weight reduction would cost (-
$159)x(15%)+$645=$621.
The small truck EV150 regression has no slope since the net weight reduction is always 0 due to the 20% weight
reduction hit.
The agencies did not regress PHEV or EV costs for the minivan+towing and large truck vehicle classes since we
do not believe these vehicle classes would use the technologies.

   Table  3.14-4 Linear Regressions of Battery Pack Direct Manufacturing Costs vs Net
                 Weight Reduction used for the Primary Analysis (2009$)
Vehicle Class
Subcompact
Small car
Large car
Minivan
Small truck
Minivan+towing
Large truck
P2HEV
-$177x+$716
-$218x+$758
-$300x+$864
-$294x+$848
-$277x+$822
-$294x+$929
-$317x+$964
PHEV20
-$862x+$2,602
-$998x+$2,746
-$l,568x+$3,331
-$l,439x+$3,296
-$l,338x+$3,143


PHEV40
-$867x+$3,649
-$2,093x+$4,037
-$3,152x+$5,192
-$2,090x+$5,035
-$2,444x+$4,787


EV75
-$l,350x+$5,424
-$2,033x+$5,888
-$3,460x+$7,173
-$3,480x+$7,201
-$3,148x+$6,828


EV100
-$2,064x+$6,360
-$2,849x+$7,004
-$4,019x+$8,101
-$3,090x+$8,414
-$2,971x+$8,045


EV150
-$2,019x+$8,292
-$3,100x+$9,187
-$3,770x+$10,989
-$4,566x+$l 1,746
$11,253


Notes:
"x" in the equations represents the net weight reduction as a percentage, so a subcompact P2 HEV battery pack
with a 20% applied weight reduction and, therefore, a 15% net weight reduction would cost (-
$177)x(15%)+$716=$689.
The small truck EV150 regression has no slope since the net weight reduction is always 0 due to the 20% weight
reduction hit.
The agencies did not regress PHEV or EV costs for the minivan+towing and large truck vehicle classes since we
do not believe these vehicle classes would use the technologies.
                                            3-94

-------
                                                 2017 Draft Regulatory Impact Analysis
   Table 3.14-5 Linear Regressions of Battery Pack Direct Manufacturing Costs vs Net
               Weight Reduction used for the High Side Sensitivity (2009$)
Vehicle Class
Subcompact
Small car
Large car
Minivan
Small truck
Minivan+towing
Large truck
P2HEV
-$194x+$788
-$240x+$834
-$330x+$950
-$324x+$933
-$304x+$904
-$324x+$l,022
-$348x+$l,061
PHEV20
-$l,034x+$3,123
-$l,197x+$3,295
-$l,882x+$3,997
-$l,727x+$3,955
-$l,606x+$3,771


PHEV40
-$l,041x+$4,378
-$2,512x+$4,844
-$3,782x+$6,230
-$2,508x+$6,042
-$2,933x+$5,744


EV75
-$l,620x+$6,509
-$2,439x+$7,065
-$4,152x+$8,607
-$4,176x+$8,641
-$3,777x+$8,194


EV100
-$2,477x+$7,632
-$3,419x+$8,404
-$4,823x+$9,722
-$3,708x+$10,097
-$3,565x+$9,654


EV150
-$2,423x+$9,950
-$3,720x+$l 1,024
-$4,524x+$13,187
-$5,479x+$14,096
$13,504


Notes:
"x" in the equations represents the net weight reduction as a percentage, so a subcompact P2 HEV battery pack with
a 20% applied weight reduction and, therefore, a 15% net weight reduction would cost (-$194)x(15%)+$788=$759.
The small truck EV150 regression has no slope since the net weight reduction is always 0 due to the 20% weight
reduction hit.
The agencies did not regress PHEV or EV costs for the minivan+towing and large truck vehicle classes since we do
not believe these vehicle classes would use the technologies.

        For the battery pack sensitivities, unlike the mass reduction sensitivities, we did re-
 rank OMEGA packages. However, we started with the master-sets of packages as described
 in Chapter 1 of this draft RIA for both the 2016MY and 2025MY and did not start with
 preliminary-sets of packages and conduct a full package building/ranking process.  Using the
 master-sets of packages, we inserted new battery pack costs, re-ranked the master-sets of
 packages to get the proper ordering of packages for each sensitivity case, then ran OMEGA.
 As noted above, this should have no impact on the results other than making them
 conservative since starting with a preliminary-set of packages and re-ranking would serve to
 move the modeling to more cost effective technologies, thus reducing the cost impact.

        The high battery cost inputs increased the average compliance costs of the reference
 case by $2 and the control case by $63, for a net cost increase of $61. In the high case, the
 penetration of EVs decreased from 2.8% to 2.2%, as companies chose more cost effective
 options. MY 2025 HEV penetration increased from 15% to  17%.  The low battery cost inputs
 decreased the average compliance costs of the reference case by $2 and the control case by
 $103  for a net decrease of $101.  In the low cost case, the MY 2025 penetration of EVs
 increased to 3.9%, while the HEV penetration decreased to  10%.  In general, changing the
 battery costs shifted the choice between HEVs and EVs. As both EVs and HEVs are less cost
 effective (in this set of inputs) than conventional technologies, the penetrations of non-battery
 dependent technologies was little changed.

        3.14.4 ICM Sensitivity

        For the ICM sensitivity, we looked  at the 95%  confidence intervals of the survey
 responses gathered as part of the modified Delphi process used to generate our low, medium
 and high2 complexity ICMs. We discuss this modified Delphi process in Chapter 3 of the
 draft joint TSD  and provide details in a memorandum  to the docket (EPA-HQ-OAR-2010-
                                          3-95

-------
Chapter 3
      ww
0799).    In that memorandum, the survey responses from each respondent are presented for
each element of the ICM along with average responses, standard deviations and other
statistical measures. Using these, we calculate the ICM elements at the low side of the 95%
confidence interval and at the high side. Table 3.14-6 and Table 3.14-8 show the ICMs used
for the low side and high side sensitivity analyses, respectively, while Table 3.14-7  shows
the ICMs used for our primary analysis. For the Highl ICM, since it was generated using a
consensus approach rather than blind surveys, we have scaled the ICM elements using the
same ratios as resulted from the 95% confidence intervals for the High2 ICM.

                  Table 3.14-6 ICMs used for the Low Side Sensitivity

Complexity
Low
Medium
Highl
High2
Near term
Warranty
0.004
0.037
0.043
0.048
Non-warranty
0.113
0.225
0.361
0.479
Long term
Warranty
0.001
0.025
0.027
0.041
Non-warranty
0.090
0.148
0.217
0.272
Summed
Near term
1.118
1.262
1.404
1.528
Long term
1.091
1.174
1.243
1.313
                   Table 3.14-7  ICMs used for the Primary Analysis

Complexity
Low
Medium
Highl
High2
Near term
Warranty
0.012
0.045
0.065
0.074
Non-warranty
0.230
0.343
0.499
0.696
Long term
Warranty
0.005
0.031
0.032
0.049
Non-warranty
0.187
0.259
0.314
0.448
Summed
Near term
1.242
1.387
1.564
1.770
Long term
1.193
1.290
1.345
1.497
                 Table 3.14-8 ICMs used for the High Side Sensitivity

Complexity
Low
Medium
Highl
High2
Near term
Warranty
0.019
0.052
0.087
0.099
Non-warranty
0.347
0.461
0.637
0.914
Long term
Warranty
0.010
0.037
0.037
0.057
Non-warranty
0.284
0.369
0.411
0.623
Summed
Near term
1.366
1.513
1.723
2.012
Long term
1.294
1.406
1.447
1.680
       Like done for the battery pack sensitivities, we re-ranked OMEGA packages using the
master-sets of packages as described in Chapter 1  of this draft RIA for the 2016MY and
2025MY. We did not start with preliminary-sets of packages and conduct a full package
building/ranking process. Using the master-sets of packages, we inserted new costs
calculated using the low/high IMCs, re-ranked the master-sets of packages to get the proper
ordering of packages for each sensitivity case, then ran OMEGA.  As noted above, this should
have no impact on the results other than making them conservative since starting with a
preliminary-set of packages and re-ranking would serve to move the modeling to more cost
effective technologies, thus reducing the cost impact.
   "Documentation of the Development of Indirect Cost Multipliers for Three Automotive Technologies,'
Helfand, G., and Sherwood, T., Memorandum dated August 2009.
                                         3-96

-------
                                                2017 Draft Regulatory Impact Analysis
       The high ICM inputs increased the average compliance costs of the reference case by
$63 and the control case by $312, for a net average per-vehicle cost increase of $249 in MY
2025. The low ICM inputs decreased the average compliance costs of the reference case by
$105 and the control case by $349 for a net average cost decrease of $244.

       3.14.5  Learning Rate Sensitivity

       For the learning rate sensitivity, we increased the learning effects for the low side case
and decreased the learning effects for the high side case. This sounds counterintuitive, but we
have done this because the increased learning rates result in lower technology costs so,
therefore, are more appropriate for the low side sensitivity.  The reverse is true when
decreasing the learning rates. For our primary analysis, as described in Chapter 3 of the draft
joint TSD, we have used a 20% p-value for technologies on the steep portion of the learning
curve and then have used learning rates of 3% per year for five years, 2% per year for 5 years,
then 1% per year for 5 years for technologies on the flat portion of the learning curve.  Table
3.14-9 shows how we have adjusted these learning rates for both the low and high side
sensitivities.

          Table 3.14-9  Learning Rates used for our Learning Rate Sensitivity
Sensitivity
Low side
Primary case
High side
Steep learning rate
30%
20%
10%
Flat learning rate
4%, 3%, 2%
3%, 2%, 1%
2%, 1%, 0%
        For the learning sensitivity, we re-ranked OMEGA packages using the master-sets of
packages as described in Chapter 1 of this draft RIA for the 2016MY and 2025MY. We did
not start with preliminary-sets of packages and conduct a full package building/ranking
process. Using the master-sets of packages, we inserted new costs calculated using the
low/high learning rates, re-ranked the master-sets of packages to get the proper ordering of
packages for each sensitivity case, then ran OMEGA.  As noted above, this should have no
impact on the results other than making them conservative since starting with a preliminary-
set of packages and re-ranking would serve to move the modeling to more cost effective
technologies, thus reducing the cost impact.

       The high learning inputs increased the average compliance costs of the reference case
by $48 and the control case by $154, for a net average per-vehicle cost increase of $106 in
MY 2025.  The low ICM inputs decreased the average compliance costs of the reference
case by $63 and the control case by $158  for a net average cost decrease of $97.

       3.14.6  Summary of Sensitivity Impacts

           The average per-vehicle impacts of the sensitivity runs are shown in

       Table 3.14-10. Note that the majority of these impacts are less than $100 relative to
the primary analysis costs. The ICM impacts are larger.
                                         3-97

-------
Chapter 3

Table 3.14-10 Summary of Per-vehicle Cost Impacts of Sensitivity Analyses in MY 2025
                            relative to Primary Analysis
Sensitivity Title
Primary Case
Mass Cost
Battery Cost
ICM
Learning Rates
Low ($)
High ($)
$1946
-$102
$101
$244
$97
$102
$61
$249
$106
       3.14.7 NAS report

       We note that EPA has decided not to base a sensitivity case on the 2010 National
Academy of Science Report "Assessment of Technologies for Improving Light-Duty Vehicle
Fuel Economy, Assessment for Fuel Economy Technologies for Light-Duty Vehicles" (The
National Academies Press, June 2010).

       As discussed in detail in Chapter 3 of the draft Joint Technical Support Document for
this proposal, EPA and NHTSA have utilized the best available information in order to
estimate the cost and effectiveness for a large number of technologies which can be used to
reduce GHG emissions and improve fuel  efficiency.

       In 2007, NHTSA commissioned the National Academy of Science to perform an
assessment of, among other things, the cost and effectiveness of technologies for improving
the fuel economy of light-duty vehicles. The 2010 NAS Committee published their results of
their assessment in June of 2010. EPA has reviewed this report in detail and for the reasons
discussed below, we have not relied upon this report as a primary assessment for our cost and
effectiveness estimates for this proposal, and we have also not used the report to perform a
sensitivity assessment based on the 2010 NAS report for the  same reasons.

       Our principal reasons are twofold. First, the 2010 NAS Committee focused their
report on the near-term, specifically the 2010-2015  time frame, and not on the time frame of
this proposal, which is 2017 to 2025.  Second, on a range of topics EPA and NHTSA have
relied upon newer information for cost and effectiveness estimates.

       With respect to the time frame of interest, in the Summary of the NAS 2010 report
(pages S-l and S-2), the NAS Committee discusses that their costs estimates are for the 2010-
2015 time frame.  The 2010 NAS Report  also discusses that there are longer-term
technologies which are in the 5 to 15 year time horizon which are not the focus of the NAS
2010 report.  There are a number of specific examples where this difference in time frame is
relevant to any potential comparison between the 2010 NAS  report and the EPA & NHTSA
assessment for this proposal. For example, there are a number of technologies that EPA and
NHTSA discuss in Chapter 3 of the draft  Joint TSD which are not a single, discrete piece of
hardware,  but rather a continuum of improvements where the level of improvement can
change given the potential time horizon.  The 2010  NAS Committee considered at least six  of
these technologies: low friction lubricants, engine friction reduction, improved accessories,
lower rolling resistance tires, aerodynamic drag improvement, and improved internals for
automatic  transmissions. The 2010 NAS  report provides cost and effectiveness estimates for
                                       3-98

-------
                                              2017 Draft Regulatory Impact Analysis
one increment of improvement for each of these technologies applicable to the 2010-2015
time frame. This is similar to the approach utilized by NHTSA and EPA for the 2012-2016
rulemaking.  However, for the 2017-2025 proposals, where EPA and NHTSA are using a
model year 2008 baseline set of vehicles, the agencies estimate that for each these
technologies two increments of improvement can be implemented across the fleet between
2008 and 2025.  Using the NAS Report estimates for these technologies thus, without basis,
would not consider the further projected incremental improvements in these technologies.

       A second example of the importance of the time frame is evaluation of the
effectiveness of gasoline direct injection with turbocharging and downsizing.  The 2010 NAS
Committee considered one level of downsizing in the 2010-2015 time frame,  and EPA and
NHTSA took a similar approach for the 2012-2016 rule. But, for the 2017-2025 proposal,
based on data in the literature, our discussions with the auto companies and automotive
suppliers, and a 2011 Ricardo study commissioned by EPA, in the longer term additional
levels of downsizing are achievable, including in some cases with the use of cooled exhaust
gas recirculation, that provide additional CO2/fuel consumption reductions. Those additioal
levels of downsizing were not considered by the 2010 NAS Committee in their assessment of
near-term costs and effectiveness.

       In addition to the difference in time frames being considered by the 2010 NAS report
and this proposal, a second significant difference between the two assessments were the
additional studies and information available to EPA and NHTSA which were not reviewed by
2010 NAS Committee.  In many cases this was due to the additional two years EPA and
NHTSA had available (while the NAS  Committee's report was published in 2010, the bulk of
their assessments occurred between 2007 and 2009), and in other areas this new information
was the result of the many confidential meetings EPA and NHTSA had with auto companies
and auto suppliers over the past two years.

       The additional publically available studies which EPA and NHTSA utilized included
new studies on the costs for  mass reduction, lithium-ion battery packs, 8 speed automatic
transmissions, 8 speed dual-clutch transmissions, hybrid electric vehicle, plug-in hybrid
electric vehicle, and all electric vehicles. EPA and NHTSA also utilized new reports dealing
with the use of indirect cost multipliers for estimating indirect manufacturing costs.  EPA and
NHTSA also are using a number of new studies which were not available to the 2010 NAS
Committee for the estimation of the effectiveness of a large number of the 2017-2025
technologies; these include peer reviewed papers in the  literature as well as the 2011 Ricardo
study (discussed in detail in Chapter 3 of the draft Joint TSD). A partial list of the studies and
data sources  regarding technology feasibility, costs, lead time, and effectiveness considered
by EPA which were not reviewed by the 2010 NAS Committee or were published after they
completed their work, or was obtained  confidentially from automotive suppliers includes:

             2011 Ricardo Report "Computer Simulation of Light-duty Vehicle
       Technologies for Greenhouse Gas Emission Reductions in the 2020-2025
       Timeframe"24, this report has been peer reviewed and the peer review report  and the
       response to peer review comments are available in the EPA docket EPA-HQ-OAR-
       2010-0799.
                                        3-99

-------
Chapter 3
             Argonne National Laboratories 2011 Report "Modeling the Performance and
       Cost of Lithium-Ion Batteries for Electric-Drive Vehicles"25 and the accompanying
       Battery Performance and Cost Model, which estimates lithium-ion battery pack cost
       for the 2020 time frame.  This report was peer reviewed and revised in 2011, and the
       model, report, and peer review report are available in the EPA docket EPA-HQ-OAR-
       2010-0799.

             2011 FEV Report "Light-Duty Technology Cost Analysis Power-split and P2
       HEV Case Studies."26  This report was peer reviewed, and a copy of the report, the
       peer review report, and the response to peer review comments report are available in
       the EPA docket EPA-HQ-OAR-2010-0799.

             2011 FEV Report "Light-Duty Technology Cost Analysis: Advanced 8-speed
       Transmissions"27. A copy of this report is available in the EPA docket EPA-HQ-
       OAR-2010-0799.

             2010 Lotus Engineer Study "An Assessment of Mass Reduction Opportunities
       for a 2017 - 2020 Model Year Vehicle Program"28, this report has been peer reviewed,
       and a copy of the report and the peer review report are available in the EPA docket
       EPA-HQ-OAR-2010-0799.

             EPA vehicle fuel economy certification data from MY2011  and MY2012
       vehicles, including for example the MY2011 Ford F-150 with the 3.5L Ecoboost
       engine, MY2011 Sonata Hybrid, MY2012 Infmiti M35h hybrid, and several other
       advanced technology production vehicles.

              "EBDI - Application of a Fully Flexible High BMEP Downsized Spark
       Ignited Engine." Society of Autmotive Engineers (SAE) Technical Paper No. 2010-
       01-0587, Gruff, L., Kaiser, M., Krause, S., Harris, R., Krueger, U., Williams, M.,
       2010.29

              "Water Cooled Exhaust Manifold and Full Load EGR Technology Applied to
       a Downsized Direct Injection Spark Ignition Engine." SAE Technical Paper Series No.
       2010-01-0356. Taylor, J., Fraser, N., Wieske, P., 2010.30

              "Requirements of External EGR Systems for Dual Cam Phaser Turbo GDI
       Engines."  SAE Technical Paper Series No. 2010-01-0588. Roth, D.B., Keller, P,
       Becker, M., 2010.31

              "Doing More with Less - The Fuel Economy Benefits of Cooled EGR on a
       Direct  Injected Spark Ignited Boosted Engine," SAE Technical Paper Series, No.
       2010-01-0589. Kaiser, M., Krueger, U., Harris, R., Gruff, L., 2010.32 EPA-HQ-OAR-
       2010-0799

              "Using indirect cost multipliers to estimate the total cost of adding new
       technology in the automobile industry," International Journal of Production
       Economics Rogozhin, A.,et al., 2009.33
                                      3-100

-------
                                               2017 Draft Regulatory Impact Analysis
          "Documentation of the Development of Indirect Cost Multipliers for Three Automotive
          Technologies/' EPA Technical Memorandum, Helfand, G., Sherwood, T., August 2009.34

          Confidential business information regarding the development status, effectiveness and
          costs for a large number of technologies obtained by EPA in meetings during 2010 and
          2011 with more than a dozen worldwide automotive suppliers involved in the
          development and production of a wide range of technologies, including but not limited
          to fuel injection systems, transmissions, turbochargers, lower mass automotive
          components, tires, and automotive lithium-ion  batteries.
       With the exception of the confidential business information and copyrighted
information, copies of the reports and studies listed above are available in the EPA docket for
this proposal, EPA-HQ-OAR-2010-0799. Information on how to obtain copies of the SAE
papers is also available in the EPA docket, or they can be order from SAE on-line at
http://papers.sae.org/.

       For the reasons described above, EPA has elected not to perform a sensitivity
assessment based on the 2010 NAS Report, nor have we used the 2010 NAS Report as our
primary basis for assessing the costs and effectiveness of technologies for this proposal.

       EPA requests comment on our overall approach for this proposal of basing our
assessment on technology feasibility, lead time, costs and effectiveness on the full range of
information described in the draft Joint Technical Support Document (which includes
consideration of the 2010 NAS Study),  as opposed to an alternative approach in which EPA
would base our technology feasibility, lead time, costs and effectiveness primarily on the
2010 NAS Study and place lower weighting or no weighting on the additional information
which has become available since the 2010 NAS Study (including those data sources, studies
and reports listed above).

       EPA also requests comment specifically on EPA's use of the  2011 Ricardo study
(listed above), and we seek comment on any ways to improve our estimates of technology
effectiveness, including the use of full vehicle simulation modeling as was used in the 2011
Ricardo study or alternative approaches. We also request comment on the 2011 Ricardo
Study and the Ricardo response to comments report with respect to the peer review conducted
on the draft Ricardo report. These  documents are all available in the EPA docket for this
rulemaking (EPA-HQ-OAR-2010-0799).  Significant additional detail regarding the 2011
Ricardo study and how it was used to inform EPA's estimates of technology effectiveness is
contained in Chapter 3 of the draft  Joint Technical Support Document.
                                        3-101

-------
Chapter 3

References

24 2011 Ricardo Report "Computer Simulation of Light-duty Vehicle Technologies for
Greenhouse Gas Emission Reductions in the 2020-2025 Timeframe"

25 Argonne National Laboratories 2011 Report "Modeling the Performance and Cost of
Lithium-Ion Batteries for Electric-Drive Vehicles"

26 2011 FEV Report "Light-Duty Technology Cost Analysis Power-split and P2 HEV Case
Studies"

27 2011 FEV Report "Light-Duty Technology Cost Analysis: Advanced 8-speed
Transmissions"

28 2010 Lotus Engineer Study "An Assessment of Mass Reduction Opportunities for a 2017 -
2020 Model Year Vehicle Program"

29  "EBDI - Application of a Fully Flexible High BMEP Downsized Spark Ignited Engine."
Society of Autmotive Engineers (SAE) Technical Paper No. 2010-01-0587, Gruff, L., Kaiser,
M., Krause, S., Harris, R., Krueger, U., Williams, M., 2010

30 "Water Cooled Exhaust Manifold and Full Load EGR Technology Applied to a Downsized
Direct Injection Spark Ignition Engine." SAE Technical Paper Series No. 2010-01-0356.
Taylor, I, Fraser, N., Wieske, P., 2010

31  "Requirements of External EGR Systems for Dual Cam Phaser Turbo GDI Engines." SAE
Technical Paper Series No. 2010-01-0588. Roth, D.B., Keller, P, Becker, M., 2010

32 "Doing More with Less - The Fuel Economy Benefits of Cooled EGR on a Direct Injected
Spark Ignited Boosted Engine," SAE Technical Paper Series, No. 2010-01-0589.  Kaiser, M.,
Krueger, U., Harris, R., Cruff, L., 2010

33 "Using indirect cost multipliers to estimate the total cost of adding new technology in the
automobile industry," International Journal of Production Economics Rogozhin, A.,et al.,
2009

34 "Documentation of the Development of Indirect Cost Multipliers for Three Automotive
Technologies," EPA Technical Memorandum, Helfand, G., Sherwood, T., August 2009
                                      3-102

-------
                                               2017 Draft Regulatory Impact Analysis
4      Projected Impacts on Emissions, Fuel Consumption, and
       Safety

   4.1 Introduction

    This chapter documents EPA's analysis of the emission, fuel consumption and safety
impacts of the proposed emission standards for light duty vehicles.  Light duty vehicles
include passenger vehicles such as cars, sport utility vehicles, vans, and pickup trucks. Such
vehicles are used for both commercial and personal uses and are significant contributors to the
total United States (U.S.) GHG emission inventory.

       Mobile sources represent a large and growing share of U.S. GHG emissions and
include light-duty vehicles, light-duty trucks, medium duty passenger vehicles, heavy duty
trucks, airplanes, railroads, marine vessels and a variety of other sources. In 2007, all mobile
sources emitted 30% of all U.S. GHGs, and have been the source of the largest absolute
increase in U.S. GHGs since 1990. Transportation sources, which do not include certain off
highway sources such as farm and construction equipment, account for 27% of U.S. GHG
emissions, and motor vehicles (CAA section 202(a)), which include light-duty vehicles, light-
duty trucks, medium-duty passenger vehicles, heavy-duty trucks, buses, and motorcycles
account for 23% of total U.S. GHGs.

    This proposal, if finalized, will significantly decrease the magnitude of these emissions.
Because of anticipated changes to driving behavior, fuel production, and electricity
generation, a number of co-pollutants would also be affected by this proposed rule. This
analysis quantifies the proposed program's impacts on the greenhouse gases (GHGs) carbon
dioxide (CC^), methane (CH4), nitrous oxide (TS^O) and hydrofluorocarbons (HFC-134a);
program impacts on  "criteria" air pollutants, including carbon monoxide (CO), fine particulate
matter (PM2.5) and sulfur dioxide (SOx) and the ozone precursors hydrocarbons (VOC) and
oxides of nitrogen (NOx); and impacts on several air toxics including benzene, 1,3-butadiene,
formaldehyde, acetaldehyde, and acrolein.

       CC>2 emissions from automobiles are largely the product of fuel combustion, and
consequently, reducing CO2 emissions will also produce a significant reduction in projected
fuel consumption. EPA's projections of these impacts are also shown in this chapter.

       In addition to the intended effects of reducing CC>2 emission, the agencies also
consider the potential of the standards to affect vehicle safety.  This topic is discussed in
Preamble Section II. G.  EPA's analysis of the change in fatalities due to projected usage of
mass reduction technology is shown in this chapter.

       This chapter  primarily describes the methods used by EPA in its analysis.  Detailed
discussion of the inputs, such as VMT, emission factors, and safety coefficients are found in
Chapter 4 of the Draft Joint TSD.
                                         4-1

-------
Chapter 4

4.2 Analytic Tools Used

       As in the MYs 2012-2016 rule, EPA used its Optimization Model for reducing
Emissions of Greenhouse gases from Automobiles (OMEGA) post-processor to project the
impacts of this proposal. Broadly speaking, the OMEGA core model is used to predict the
most likely paths by which manufacturers would meet tailpipe CO2 emission standards.
OMEGA applies technologies with varying degrees of cost and  effectiveness to a defined
vehicle fleet in order to meet a specified GHG emission target and calculates the  costs and
benefits of doing so.  The projections of impacts in OMEGA are conducted in a Microsoft
Excel Workbook (the benefits post-processor).  The OMEGA benefits post-processor
produces a national scale analysis of the impacts (emission reductions, monetized co-benefits,
safety impacts) of the analyzed program.

    The benefits post-processor incorporates the inputs discussed (many extensively) in the
Draft Joint Technical  Support Document.  Specifically, Draft Joint TSD Chapter 1 discusses
the development of the vehicle fleet, Draft Joint TSD Chapter 2 discusses the attribute based
curves which define the CO2 targets, Draft Joint TSD Chapter 3 discusses the technologies
                                      "V"V                	
which may be used to meet those targets,   and Draft Joint TSD Chapter 4 discusses other
relevant inputs (such as vehicle sales, vehicle miles traveled (VMT), and survival schedules).

    The remainder of this chapter provides a summary of the discussion of the TSD inputs,
additional data on methodology and inputs, and the results of the analysis.

4.3 Inputs to the emissions analysis

       4.3.1   Methods
       EPA estimated greenhouse impacts from several sources including: (a) the impact of
the standards on tailpipe CO2 emissions, (b) projected improvements in the efficiency of
vehicle air conditioning systems, YY (c) reductions in direct emissions of the potent
greenhouse gas refrigerant JTFC-134a from air conditioning systems, (d) "upstream" emission
reductions from gasoline extraction, production and distribution processes as a result of
reduced gasoline demand associated with this rule, and (e) "upstream" emission increases
from power plants as electric powertrain vehicles increase in prevalence as a result of this rule
(Table 4.3-16).zz EPA additionally accounted for the greenhouse gas  impacts of additional
vehicle miles travelled (VMT) due to the "rebound" effect discussed in Section III.H.
** Specifically, the power consumption of plug-in hybrid and battery electric vehicles are discussed in Draft
Joint TSD Chapter 3 and used in this analysis. Mass reduction, an input to the mass-safety analysis, is also
discussed therein.
YY While EPA anticipates that the efficiency of the majority of mobile air conditioning systems will be improved
in response to the MY 2012-2016 rulemaking, the agency expects that the remainder will be improved as a result
of this proposed action.
zz As discussed in TSD Chapter 4, the increased emissions from power plants includes feedstock gathering.
This includes GHG emissions from the extraction of fuel for power plants, including coal and natural gas.


                                          4-2

-------
                                                2017 Draft Regulatory Impact Analysis
       Our estimates of non-GHG emission impacts from the GHG program are broken down
by the three drivers of these changes: a) "downstream" emission changes, reflecting the
estimated effects of VMT rebound (discussed in Sections III.F and III.H) and decreased
consumption of fuel; b) "upstream" emission reductions due to decreased extraction,
production and distribution of motor vehicle gasoline; c)  "upstream" emission increases from
power plants as electric powertrain vehicles increase in prevalence as a result of this rule. For
all criteria and air toxic pollutants the overall impact of the proposed program would be small
compared to total U.S. inventories across all sectors.

       As discussed in the preamble, while electric vehicles have zero tailpipe
emissions, EPA assumes that manufacturers will plan for these vehicles in their
regulatory compliance strategy for criteria pollutant and air toxics emissions, and will
not over-comply with those standards for non-GHG air pollutants.   Since the Tier 2
emissions standards are fleet-average standards, we assume that if a manufacturer
introduces EVs into its fleet, that it would correspondingly compensate through
changes to vehicles elsewhere in its fleet, rather than produce  an overall lower fleet-
average emissions level.35  Consequently, EPA assumes neither tailpipe pollutant
benefit (other than CO2) nor an evaporative emission benefit from the introduction of
electric vehicles into the fleet.Two basic elements feed into OMEGA's calculation of
vehicle tailpipe emissions. These elements are vehicle miles traveled (VMT) and
emission rates.

                       Total Emissions = VMT miles * Emission rate grams/miie

                                   Equation 2 - Emissions

       This  equation is adjusted in calculations for various emissions, but provides the basic
form used throughout this analysis.  As an example, in an analysis of a single calendar year,
the emission equation is repeatedly applied to determine the contribution of each model year
in the calendar year's particular fleet. Appropriate VMT and emission factors by age are
applied to each model year within the calendar year, and the products are then summed.
Similarly, to determine the emissions of a single model year,  appropriate VMT and emission
factors by age are applied to each calendar year between when the model year fleet is
produced and projected to be scrapped

       SO2; which is largely controlled by the  sulfur content of the fuel, is an exception to this
basic equation.  As discussed inTSD 4, decreasing  the quantity of fuel consumed decreases
tailpipe SO2 emissions roughly proportionally  to the decrease  in fuel consumed.

           4.3.1.1   Global Warming Potentials

    Throughout this document, in order to refer to the four inventoried greenhouse gases on
an equivalent basis, Global Warming Potentials (GWPs) are used. In simple terms, GWPs
provide a common basis with which to combine several gases with different heat trapping
abilities into a single inventory (Table 4.3-1).  When expressed in CO2 equivalent (CO2 EQ)
terms, each gas is weighted by its heat trapping ability relative to that of carbon dioxide.  The
global warming potentials used in this rule are consistent with 100-year time frame values in
the 2007 Intergovernmental Panel on Climate Change (IPCC) Fourth Assessment Report
                                         4-3

-------
Chapter 4
       36
(FAR).   At this time, the 100-year global warming potential values from the 1996 IPCC
Second Assessment Report are used in the official U.S. greenhouse gas inventory submission
to the United Nations Framework Convention on Climate Change (per the reporting
requirements under that international convention, which were last updated in 2006) .  The
FAR values were used in the MYs 2012-2016 light duty rule and the MY 2014-2018 Medium
and Heavy duty rule.
            Table 4.3-1 Global Warming Potentials for the Inventory GHGs
Gas
CO2
CH4
N2O
HFC (R134a)
Global Warming potential
(CO2 Equivalent)
1
25
298
1430
           4.3.1.2   Years consi dered

       This analysis presents the projected impacts of this proposal in calendar years 2020,
2030, 2040 and 2050. We also present the emission impacts over the estimated full lifetime
of MYs 2017-2025.AAA The program was quantified as the difference in mass emissions
between a control case under proposed standards and a reference case as described in Section
4.3.3.
       4.3.2  Activity

           4.3.2.1   Vehicle Sales

  Vehicle sales projections from MY 2012 through MY 2025 were developed jointly by
   NHTSA and EPA and are discussed in Chapter 1 of the Draft Joint TSD. For MYs
 between 2025  and 2035, EPA used the Volpe Center run of the NEMs model (discussed
 in Draft Joint TSD Chapter 1) in order to project the sales of cars and trucks by "pre-
     MY 2011" definitions .  23 percent of "pre-MY 2011" defined trucks were then
                                 converted to cars (

       Table 4.3-2), consistent with the percent that changed in MY 2025 within the
reference fleet forecast. This action reflects the assumption that the vehicle mix within the car
and truck classes stops changing after MY 2025.
AAA The "full lifetime" is the timespan between sales and scrappage for a given MY, and includes estimates of
sales, scrappage, and VMT accumulation by year. For a given vehicle, it is the mileage between when it is
driven for its first and last miles.
                                         4-4

-------
                                                 2017 Draft Regulatory Impact Analysis
              Table 4.3-2 MY 2011 and later Car and Truck Definitions
                                                                       BBB
CAR DEFINITION
Passenger Car -Vehicles defined pre-MY 20 lias Cars + 2
wheel drive SUVs below 6,000 GVW
TRUCK DEFINITION
Light!
fleet


       As the NEMS analysis only goes through 2035, and this analysis goes through 2050,
sales from 2035-2050, the sales of cars and trucks were each projected to grow at the average
annual rates of sales growth from 2017-2035 (1.16%).
           4.3.2.2   Survival schedules000

       TSD 4 also describes the derivation of the survival schedule, which is shown below.
The proportions of passenger cars and light trucks expected to remain in service at each age
are drawn from a 2006 NHTSA study, and are shown in (Table 4.3-16)DDD'37  Note that these
survival rates were calculated against the pre-MY 2011 definitions of cars and light trucks,
because the NHTSA study has not been updated since 2006.  Because the agencies are
unaware of a better data source, these values were used unchanged, and are the same values
used in the MYs 2012-2016 rule and the interim Joint TAR. No changes in survival rates were
explicitly projected into the  future.

       The survival and annual mileage estimates reported in this section's tables reflect the
convention that vehicles are defined to be of age 1 during the calendar year that coincides
with their model year.  Thus for example, model year 2017 vehicles will be considered to be
of age 1 during calendar year 2017. This convention is used in order to account for the fact
that vehicles produced during a model year typically are first offered for sale in June through
September of the preceding  calendar year (for example, sales of a model year typically begin
in June through September of the previous calendar year, depending on manufacturer). Thus,
virtually all of the vehicles produced during a model year will be  in use for some or all of the
calendar year coinciding with their model year, and they are considered to be of age 1 during
that year.EEE
BBB While the formal definitions are lengthy, brief summaries of the classifications are shown here.
ccc A lengthier discussion of both survival and mileage schedules are provides in Draft Joint TSD Chapter 4
DDD The maximum age of cars and light trucks was defined as the age when the number remaining in service has
declined to approximately two percent of those originally produced. Based on an examination of recent
registration data for previous model years, typical maximum ages appear to be 26 years for passenger cars and
36 years for light trucks.
EEE Historic values are derived from the Fuel Economy Trends report (http://www.epa.gov/otaq/fetrends.htm).
future values are discussed in Table 4.3-7.
                                           4-5

-------
Chapter 4
                            Table 4.3-3  Survival Rates
VEHICLE AGE
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
ESTIMATED
SURVIVAL
FRACTION
CARS
0.9950
0.9900
0.9831
0.9731
0.9593
0.9413
0.9188
0.8918
0.8604
0.8252
0.7866
0.7170
0.6125
0.5094
0.4142
0.3308
0.2604
0.2028
0.1565
0.1200
0.0916
0.0696
0.0527
0.0399
0.0301
0.0227
0
0
0
0
0
0
0
0
0
0
ESTIMATED
SURVIVAL
FRACTION
LIGHT TRUCKS
0.9950
0.9741
0.9603
0.9420
0.9190
0.8913
0.8590
0.8226
0.7827
0.7401
0.6956
0.6501
0.6042
0.5517
0.5009
0.4522
0.4062
0.3633
0.3236
0.2873
0.2542
0.2244
0.1975
0.1735
0.1522
0.1332
0.1165
0.1017
0.0887
0.0773
0.0673
0.0586
0.0509
0.0443
0.0385
0.0334
                                       4-6

-------
                                               2017 Draft Regulatory Impact Analysis
           4.3.2.3   VMT schedules

       To estimate total miles driven, the number of cars and light trucks projected to remain
in use during each future calendar year is multiplied by the average number of miles a
surviving car or light truck is expected to be driven at the age it will have reached in that year.
Estimates of average annual miles driven by calendar year 2001 cars and light trucks of
various ages were developed by NHTSA from the Federal Highway Administration's 2001
National Household Transportation Survey (NHTS), and are reported in(Table 4.3-4).  These
estimates represent the typical number of miles driven by a surviving light duty vehicle at
each age over its estimated full lifetime.  To determine the number of miles a typical vehicle
produced during a given model year is expected to be driven at a specific age, the average
annual mileage for a vehicle of that model year and age is multiplied by the corresponding
survival rate for vehicles of that age.
                                         4-7

-------
Chapter 4
                      Table 4.3-4 CY 2001 Mileage Schedules
VEHICLE AGE
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
ESTIMATED
VEHICLE MILES
TRAVELED
CARS
14,231
13,961
13,669
13,357
13,028
12,683
12,325
11,956
11,578
11,193
10,804
10,413
10,022
9,633
9,249
8,871
8,502
8,144
7,799
7,469
7,157
6,866
6,596
6,350
6,131
5,940
0
0
0
0
0
0
0
0
0
0
ESTIMATED
VEHICLE MILES
TRAVELED
LIGHT TRUCKS
16,085
15,782
15,442
15,069
14,667
14,239
13,790
13,323
12,844
12,356
11,863
11,369
10,879
10,396
9,924
9,468
9,032
8,619
8,234
7,881
7,565
7,288
7,055
6,871
6,739
6,663
6,648
6,648
6,648
6,648
6,648
6,648
6,648
6,648
6,648
6,648
                                     4-8

-------
                                                  2017 Draft Regulatory Impact Analysis
       4.3.2.4 Adjusting vehicle use for years after 2001

       For this rulemaking, the agencies updated the estimates of average vehicle use
reported in Table 4-4 using forecasts of future fuel prices reported in the AEO 2011 Reference
Case and projected fuel economy levels in the reference and control cases.FFF This adjustment
accounts for the difference between the average retail price per gallon of fuel forecast during
each calendar year over the expected lifetimes of model year 2017-25 passenger cars and light
trucks, and the average price that prevailed when the NHTS was conducted in 2001.000 This
adjustment also accounts for the potential rebound effect from vehicles of a specific age in
future years having higher fuel economy than vehicles of the same age in 2001 (discussed
further in section 4.5.1). Like the survival schedule, this VMT schedule was not adjusted for
differences in car and truck definitions post MY 2011.

       Specifically, the elasticity of annual vehicle use with respect to fuel  cost per mile
corresponding to the 10 percent fuel economy rebound effect used in this analysis (i.e.., an
elasticity of annual vehicle use with respect to fuel cost per mile driven of-0.10; see section
4.5.1) was  applied to the difference between the combination of each future year's fuel prices
and vehicle fuel economy and those prevailing in 2001.HHH'm

       The estimates of annual miles driven by passenger cars and light trucks at each  age
were also adjusted to reflect projected future growth trends in average use for vehicles  of all
ages. In order to develop reasonable estimates of future growth in the average number of
miles driven by cars and light trucks of all ages, the agencies calculated the rate of growth in
the reference mileage schedules necessary for total car and light truck travel to increase at the
rate forecast in the AEO 2011 Reference Case. The growth rate in average  annual car and
light truck  use produced by this calculation is approximately 1.1 percent per year through
2030, and 0.5% per year from 2031-2050.JJJ   As shown in TSD 4,  this roughly calibrates the
total calculated VMT to an extrapolation from AEO 2011. This growth was applied to the
mileage figures  reported in Table 4.3-4 (after adjusting them as described previously for
future fuel  prices, fuel economy, and expected vehicle survival rates) to estimate average
annual mileage during each calendar year analyzed and during the expected lifetimes of
FFF Historic values are derived from the Fuel Economy Trends report (http://www.epa.gov/otaq/fetrends.htm),
future values are discussed in Table 4.3-7
GGG Under the assumption that people tend to drive more as the cost of driving decreases, the higher fuel prices
that are forecast for future years would be expected to reduce average vehicle use. We assume that fuel prices
will be the same in both the reference and control cases;, however, in section III.H.7 of the preamble to the
proposed rule, we discuss the potential for this proposal to decrease world oil prices.
111111 See Draft Joint TSD Chapter 4
m For our VMT analysis, we assume consumers respond the same way to changes in fuel efficiency and fuel
prices. Consistent with this assumption, we use the same elasticity to measure consumer responses to changes in
fuel prices as we do to measure the rebound effect of consumers driving more in response to increased fuel
efficiency.  See section Draft Joint TSD Chapter 4 and section 4.5.1 for additional discussion.
111 It was not possible to estimate separate growth rates in average annual use for cars and light trucks, because of
the significant reclassification of light truck models as passenger cars discussed previously.


                                           4-9

-------
Chapter 4

model year 2017-25 cars and light trucksKKK The net impact resulting from these two separate
adjustments is continued growth over time in the average number of miles that vehicles of
each age are driven, although at slower rates than those observed from 1985 - 2005. Observed
aggregate VMT in recent years has actually declined, but it is unclear if the underlying cause
is general shift in behavior or a response to a  set of temporary economic conditions. The
agencies intend to consider new data on the VMT growth estimates as it becomes available,
and are requesting  comment on this topic.LLL
       Because the effects of fuel prices, fuel economy, and growth in average vehicle use
differ for each year, these adjustments result in different VMT schedules for each future year.
While the adjustment for future fuel prices generally reduces average mileage at each age
from the values tabulated from the 2001 NHTS, the adjustment for expected future growth in
average vehicle use and improvements in fuel economy increases it.  The net impact resulting
from these separate adjustments is continued growth over time in the average number of miles
that vehicles of each age are driven, although at slower rates.
       4.3.2.5 Final VMT equation

       The following equation summarizes in mathematical form the adjustments that are
made to the values of average miles driven by vehicle age derived from the 2001 NHTS to
derive the estimates of average miles driven by vehicles of each model year (denoted MY)
during future calendar years (denoted CY) that are used in this analysis.  The equation has
three multiplicative components; the CY2001  VMT by age, the adjustment for a growth rate,
and the adjustment for changes in fuel prices and fuel economy.
                               Equation 3 - VMT growth

  VMTcalendaryear Xiagey = (Vy) * (1 + G/?l)ra * (1 + G2YS2 *(!-/?* (FCPM2001,y
       Where:
       Vy = Average miles driven in CY 200 1 (from NHTS A analysis of 200 1 NHTS data) by a vehicle of age
       y during 2001
       GR1 = Growth Rate for average miles driven by vehicles of each age from 200 1 to 2030
       YS1 = Lesser of (Years since 2001) and (29).
       GR2 = Growth Rate for average miles driven by vehicles of each age from 2030 to 2050
       YS2 = Greater of (Years since 2030) and (0).
KKK As indicated previously, a vehicle's age during any future calendar year is uniquely determined by the
difference between that calendar year and the model year when it was produced.
LLL The agencies note that VMT growth has slowed, and because the impact of VMT is an important element in
our benefit estimates, we will continue to monitor this trend to see whether this is a reversal in trend or
temporary slow down. See the 2009 National Household Travel Survey (http://nhts.ornl.gov/2009/pub/stt.pdf)
and National transportation Statistics
                                          4-10

-------
                                                 2017 Draft Regulatory Impact Analysis
       R= Magnitude of consumer responsiveness to changes in fuel cost per mile, equivalent to the rebound
effect rate assumption and expressed as an elasticity (-0.10)
       FCPM^ = Fuel cost per mile of a vehicle of age y in calendar year x

       In turn, fuel  cost per mile of an age y vehicle in calendar year x is determined by the
following equation, which can be extended for any number of fuels:
FCPMCalendaryearx = ECy * EPX
GCy * GPX
       Where:
       ECy= Electricity consumption of age y vehicle (in KWh) per mile
       EPX = Electricity Price (in $ per KWh) during calendar year x
       GCy = Gasoline Consumption of age y vehicle (in gallons) per mile
       GPX = Gasoline Price (in $ per gallon) during calendar year x
       DCy = Diesel Consumption of age y vehicle (in gallons) per mile

       DP* = Diesel Price (in $ per gallon) during calendar year x
                                                              DCy * DPX
       Table 4.3-5 presents the EPA's estimates of the average number of miles driven by
model year 2021 and 2025 cars and light trucks at over their estimated average lifetimes.
While these values may appear large relative to current vehicles, the full useful life of MY
2025 vehicles (36 years) ends in CY 2061.  A more extensive discussion of the VMT schedule
development relative to AEO and current data is presented in TSD 4. The control case VMT
schedules, due to the lower cost per mile, have somewhat higher VMT.
                  Table 4.3-5-Reference VMT used in EPA's analyses
MY 2021
Cars
204,668
Light
Trucks
242,576
MY 2025
Cars
210,898
Light
Trucks
249,713
           4.3.2.6   Non CO2 Emission Factors

    As documented in Draft Joint TSD Chapter 4, emission factors for this analysis were
derived from several sources. A more complete documentation of these sources is provided in
that chapter. Tailpipe emission factors other than CO2 were derived from MOVES 2010a.38
Upstream emission factors for petroleum refining, transport and distribution were derived
from EPA's "Impact spreadsheet" based on Argon National Labs Greet 1.8.39'40 Electricity
related emission factors for were derived from EPA's Integrated Planning Model (TPM).
These emission factors were used as inputs to the OMEGA post-processor.41
                                          4-11

-------
Chapter 4

       4.3.3   Scenarios

           4.3.3.1   Air conditioning

    HFC-134a (refrigerant) emission factors were applied on a gram per mile basis, and are
consistent with the Interim Joint TAR analysis of the on-road HFC impact per mile of 11.5
gram/mile for cars and 13.0 gram/mile for trucks.  For this analysis, the per-mile impact of
HFC reduction was determined by multiplying the fractional phase in of the credit by the
Interim Joint TAR assessment of the g/mile impact. Relative to the NPRM estimates, the
TAR estimates of HFC-134a leakage are smaller.  See TSD 5 for a detailed  discussion of the
TAR estimates of HFC-134a emissions, and why the total reductions estimated here may be
conservative in this regard.  As VMT is increasing, and the impact for HFC-134a control
programs are calculated on a gram/mile basis, this analysis implicitly assumes that a vehicle
driven more miles will have its HFC-134a reservoir refilled more times.
                              Table 4.3-6 - A/C Credits



MY 2017
MY 2018
MY2019
MY2020
MY2021
MY2022
MY2023
MY2024
MY2025
Reference
Car
Indirect
4.8
4.8
4.8
4.8
4.8
4.8
4.8
4.8
4.8
ts
OJ
Q
5.4
5.4
5.4
5.4
5.4
5.4
5.4
5.4
5.4
tc
4-»
£
10.2
10.2
10.2
10.2
10.2
10.2
10.2
10.2
10.2
Truck
Indirect
4.8
4.8
4.8
4.8
4.8
4.8
4.8
4.8
4.8
ts
OJ
Q
6.6
6.6
6.6
6.6
6.6
6.6
6.6
6.6
6.6
tc
4-»
£
11.5
11.5
11.5
11.5
11.5
11.5
11.5
11.5
11.5
Control
Car
Indirect
5.0
5.0
5.0
5.0
5.0
5.0
5.0
5.0
5.0
t;
OJ
Q
7.8
9.3
10.8
12.3
13.8
13.8
13.8
13.8
13.8
tc
4-»
£
12.8
14.3
15.8
17.3
18.8
18.8
18.8
18.8
18.8
Truck
Indirect
5
6.5
7.2
7.2
7.2
7.2
7.2
7.2
7.2
t;
OJ
Q
7
11
13.4
15.3
17.2
17.2
17.2
17.2
17.2
tc
4-»
£
12
17.5
20.6
22.5
24.4
24.4
24.4
24.4
24.4
    Indirect air conditioning emissions, or the additional load put on the engine by the
operation of the air conditioning unit, were  modeled similarly to the modeling in the MY
2012-2016 rulemaking, although with slightly different values.  The credits for air
conditioning efficiency improvements from the tables above (i.e. "indirect") were applied
directly to the two cycle emissions projected by OMEGA.

    Air conditioning credits, are modeled similarly to the MYs  2012-2016 rule, and their
derivation is more fully described in TSD 5. In the impacts modeling, both credits are
modeled as environmentally neutral, or that the impacts of the credits are larger than their 2
cycle credit values by the on-road gap.  See TSD 5 for more details.

           4.3.3.2   Reference Case

       As described in DRIA chapter 3 and Preamble HID, we assume a flat reference case
of MY 2016 standards. No additional compliance flexibilities were explicitly modeled for the
                                        4-12

-------
                                               2017 Draft Regulatory Impact Analysis
MY 2016 standards. The EPA flexible fueled vehicle (FFV) credit expires before MY
2016.MMM The Temporary Leadtime Allowance Alternative Standards (TLAAS), as analyzed
in RIA chapter 5 of the MY 2012-2016 rule, is projected have an impact of approximately 0.1
g/mile in MY 2016, and (by rule) will expire afterwards.  Therefore, no incentive credits are
projected to be available to the reference case. Off-cycle credits, which are designed to be
environmentally neutral, would only lower costs.  These credits are not modeled here due to
the difficult in predicting manufacturers use of these credits under the MY 2016 program.

       Consistent with the MYs 2012-2016 rule analysis, EPA did not allow EVs and PHEVs
(maximum penetration caps of zero) in the reference case. While the penetration of EVs and
PHEVs in MY 2016 will like be non-zero, as they are being sold in MY 2011, EPA chose not
to include these technologies  in the reference case assessment due to their cost-distorting
effects on the smallest companies. For further discussion see DRIA Chapter 3.

       CO2 emission rates for MY 2016, 2021 and 2025 were taken from OMEGA model
outputs. Intermediate years were interpolated, and CO2 g/mile rates past MY 2025  were kept
the same.  Two cycle CO2 emission rates for the reference case are shown below, and
continue changing on a fleet basis due to mix shifts (Table 4.3-7).  As no EVs were modeled,
there is no increase in electricity consumption in the reference Case. The air conditioning
impacts as discussed in Section 4.3.3.1 were also  incorporated.

                     Table 4.3-7 - Reference Case Two Cycle COi
MY
2017
2018
2019
2020
2021
2022
2023
2024
2025
Car
233
233
233
233
233
233
232
232
232
Truck
314
315
315
315
315
315
315
314
314
Fleet
263
263
262
262
262
261
261
260
259
           4.3.3.3   Control Case

MY 2017-2025 CO2 emission estimates were derived from the curves that determine the
targets and from projected credit usage on an industry wide basis. These values slightly
differ from those produced by the OMEGA modeling, which includes car-truck trading,
  but the results should be environmentally equivalent. A/C refrigerant and efficiency
  credit estimates are discussed in Section4.3.3.1, while the EV/PHEV/FCV multiplier
   credit and pickup related credits are discussed in the following sections. Off-cycle
    The credit available for producing FFVs will have expired, although the real world usage credits will be
available.
                                        4-13

-------
Chapter 4

  credits were not modeled explicitly, as they have been designed to be environmentally
           neutral. These estimates are summarized in Table 4.3-8 through
                                      4-14

-------
                                              2017 Draft Regulatory Impact Analysis
       Table 4.3-10. In the impacts modeling, both credits are modeled as environmentally
neutral, or that the impacts of the credits are larger than their 2 cycle credit values by the on-
road gap. See TSD 5 for more details.
                    Table 4.3-8 Passenger Cars (Grams per mile)
Model
Year
2016 (base)
2017
2018
2019
2020
2021
2022
2023
2024
2025
Projected
C02
Compliance
Target
225
213
202
192
182
173
165
158
151
144
EV/PHEV/FCV
Multiplier
—
2.2
2.1
2.0
1.5
1.0
—
—
—
—
A/C
Refrigerant
5.4
7.8
9.3
10.8
12.3
13.8
13.8
13.8
13.8
13.8
A/C
Efficiency
4.8
5.0
5.0
5.0
5.0
5.0
5.0
5.0
5.0
5.0
Projected
2-cycle
C02
235
228
219
210
201
193
184
177
169
163
                     Table 4.3-9 Light Trucks (Grams per mile)
Model
Year
2017
2018
2019
2020
2021
2022
2023
2024
2025
Projected
C02
Compliance
Target
295
285
277
270
250
237
225
214
203
EV/PHEV/FCV
Multiplier
0.0
0.0
0.1
0.1
0.0
—
—
—
—
Pickup
Mild
HEV +
Perf
0.3
0.4
0.6
0.7
0.8
—
—
—
—
Pickup
Strong
HEV +
Perf
0.0
0.1
0.2
0.2
0.4
0.5
0.6
0.6
0.7
A/C
Refrigerant
7.0
11.0
13.4
15.3
17.2
17.2
17.2
17.2
17.2
A/C
Efficiency
5.0
6.5
7.2
7.2
7.2
7.2
7.2
7.2
7.2
2-
cycle
C02
(3)
307
303
299
293
275
262
250
239
228
                                       4-15

-------
Chapter 4
               Table 4.3-10 Combined Cars and Trucks (Grams per mile)
Model
Year
2017
2018
2019
2020
2021
2022
2023
2024
2025
Projected
C02
Compliance
Target
243
232
223
213
200
190
181
172
163
EV/PHEV/FCV
Multiplier
1.4
1.3
1.3
1.0
0.6
—
—
—
—
Pickup
Mild
HEV
+ Perf
0.1
0.2
0.2
0.3
0.3
—
—
—
—
Pickup
Strong
HEV
+ Perf
0.0
0.0
0.1
0.1
0.1
0.2
0.2
0.2
0.2
A/C
Refrigerant
7.5
9.9
11.7
13.4
15.0
15.0
15.0
14.9
14.9
A/C
Efficiency
5.0
5.5
5.8
5.8
5.8
5.8
5.8
5.7
5.7
2-cycle
C02
257
249
242
234
222
211
202
193
184
4.3.3.3.1  EV/PHEV/FCVs

       As discussed in Section III.B of the preamble, the compliance cap for EVs and PHEVs
at zero g/mile is related to the standard level proposed.  For purposes of this modeling, we
assume that this cap is never reached.  This does not imply that EPA has proposed a cap based
on this criteria. A discussion of the potential impacts of these credits can be found in
preamble section III.C.2  and Section 4.5.2of the DRIA Costs beyond MY 2025 assume no
technology changes on the vehicles, and implicitly assume EVs used for compliance remain at
zero gram/mile.NNN Upstream emissions from electric vehicles, regardless of the zero-gram
mile credit, are always modeled in this analysis.
       For the benefits analysis, we assumed the following penetration of electric vehicles,
where the MY 2021 and MY 2025 values come from OMEGA, with the earlier and later
values interpolated. 2017 EV penetrations were setat 1% of the fleet. PHEV sales, as
                                       ^~~
projected by OMEGA, are not significant.
^^ The costs for PHEVs and EVs in this rule reflect those costs discussed in Draft Joint TSD Chapter 3, and do
not reflect any tax incentives, as the availability of those tax incentives in this time frame is uncertain.
000 Please note that the OMEGA technology projection for EVs and PHEVs does not include the multiplier
provision. Including that provision would presumably increase EV penetration in the 2017-2021 timeframe.
                                          4-16

-------
                                                  2017 Draft Regulatory Impact Analysis
                      Table 4.3-11 - EV Fraction of the MY Fleets
Model Year
2017
2018
2019
2020
2021
2022
2023
2024
2025
Cars
1.0%
1.0%
1.1%
1.1%
1.1%
1.7%
2.4%
3.0%
3.6%
Truck
0.0%
0.0%
0.0%
0.0%
0.0%
0.3%
0.5%
0.8%
1.0%
EV
multiplier
2
2
2
1.75
1.5
0
0
0
0
       The EV multiplier credit was calculated by following formula

                          Equation 4 - Impact of EV multiplier

       GHG Target with multiplier = (GHG Target without multiplier * (Total MY Sales + Multiplier *
Number of EV sales))/Total sales

       So for MY 2021, which had car sales of 10.5 million and a car GHG target of 172.8,
the formula would yield

                     Equation 5 - Impact of EV multiplier: example

       GHG Target with multiplier =

       (172.8 * (10.5 million+ 1.5 * 1.1% EV sales * 10.5 million sales))/10.5 million sales

       = 173.8 or a delta of 1.0 grams.



4.3.3.3.2  Mild and Strong HEV Pickup Credits

       Between MY 2017 and MY 2025, full-size pickup sales vary as a fraction of the fleet
sales as well as a fraction of light truck sales.  As we  did not consider these credits directly in
the OMEGA cost modeling, we did two post-process exercises to project likely benefits.
                                          4-17

-------
Chapter 4
                  Table 4.3-12 Pickup Trucks as a Fraction of the Fleet
Model Year
2017
2018
2019
2020
2021
2022
2023
2024
2025
Projected Sales of
Full Size Pickup
TrucksPPP
1,240,844
1,186,474
1,133,605
1,157,114
1,122,173
1,103,058
1,045,507
1,011,897
1,002,806
Pickup Trucks
(of Trucks)
21%
21%
20%
21%
20%
19%
18%
18%
18%
Pickup Trucks
(of Fleet)
8%
8%
7%
7%
7%
7%
6%
6%
6%
Trucks
(of fleet)
37%
36%
36%
35%
35%
35%
34%
34%
33%
       Based on these fleet fractions, and the credit available the maximum potential credit
can be calculated.

       Table 4.3-13 Maximum Potential Impact of Pickup Credits on Truck Fleet
Model
Year
2017
2018
2019
2020
2021
2022
2023
2024
2025
Mild
HEV
Credit
10.0
10.0
10.0
10.0
10.0
0.0
0.0
0.0
0.0
Mild
HEV
Max
impact
(Trucks)
2.1
2.1
2.0
2.1
2.0
0.0
0.0
0.0
0.0
Strong
HEV
Credit
20.0
20.0
20.0
20.0
20.0
20.0
20.0
20.0
20.0
Strong HEV Max
Credit (Trucks)
4.3
4.2
4.1
4.1
3.9
3.9
3.7
3.6
3.5
       Not every pickup truck will get these credits.  For the each credit, there is a minimum
fleet fraction required for a manufacturer to receive the credit. For the mild credit, we
assumed that one-half of this minimum percentage received the credit. For the strong HEV
credit we assumed that 0% received the credit in MY 2017, 10% received the credit in MY
2021, and 20% received the credit in MY 2025. Because these penetrations are in all cases
   These totals include 1 model with 30,000 sales which would not be classified as a full size pickup under the
proposal. Therefore, the credit impact is be overstated by about 3%.
                                          4-18

-------
                                                2017 Draft Regulatory Impact Analysis
higher than the penetrations projected by the OMEGA model, we consider these estimates to
adequately bound the potential utilization of the performance based credit.

                             Table 4.3-14 - Pickup Credits
MY




2017
2018
2019
2020
2021
2022
2023
2024
2025
Mild
Credit(%
of
Pickups)

30%
40%
55%
70%
80%




Truck
Credit
from
Mild
(g/mile)
0.3
0.4
0.6
0.7
0.8




Strong
Credit(%
of
Pickups)

0%
2%
4%
6%
10%
13%
15%
18%
20%
Truck
Credit
from
Strong
(g/mile)
0.0
0.1
0.2
0.2
0.4
0.5
0.6
0.6
0.7
           4.3.3.4    Consumption of Electricity

       Based on the OMEGA model outputs, we estimated electricity consumption and
emission impacts from the consumption of electricity due to the electric vehicles and plug-in
electric hybrids.  EPA accounts for all electricity consumed by the vehicle.  For calculations
of GHG emissions from electricity generation, the total energy consumed from the battery is
divided by 0.9 to account for charging losses, and by 0.93 to account for losses during
transmission. Both values were discussed in the MYs 2012-2016 rule as well as the Interim
Joint TAR, and are unchanged from those analyses. The estimate of charging losses is based
upon engineering judgment and manufacturer CBI. The estimate of transmission losses is
consistent, although not identical to the 8% estimate used in GREET, as well as the 6%
estimate in eGrid 2010.42'43  The upstream emission factor is applied to total electricity
production, rather than simply power consumed at the wheel. QQQ  It is assumed that
electrically power vehicles drive the same drive schedule as the rest of the fleet.
QQQ By contrast, consumer electricity costs would not include the power lost during transmission. While
consumers indirectly pay for this lost power through higher rates, this power does not appear on their electric
meter.
                                         4-19

-------
Chapter 4
                   Table 4.3-15 Average Electricity Consumption

Model
Year
2017
2018
2019
2020
2021
2022
2023
2024
2025
Average 2 cycle
Electricity
Consumption for the
fleet (kwh/mile)
Cars
0.000
0.001
0.001
0.002
0.002
0.004
0.005
0.007
0.009
Trucks
0.000
0.000
0.000
0.000
0.000
0.001
0.002
0.002
0.003
                                       4-20

-------
                                       2017 Draft Regulatory Impact Analysis
4.3.4  Emission Results




    4.3.4.1   Calendar Year Analyses




          Table 4.3-16 Impacts of Program on GHG Emissions
Calendar Year:
Net Delta*
Net CO 2
Net other GHG
Downstream
CO 2 (excluding A/C)
A/C - indirect CO2
A/C - direct HFCs
CH4 (rebound effect)
N2O (rebound effect)
Gasoline Upstream
C02
CH4
N2O
Electricity Upstream
C02
CH4
N2O
2020
-29
-24
-4
-24
-19
-1
-4
0
0
-6
-5
-1
0
1
1
0
0
2030
-297
-268
-29
-249
-224
-3
-21
0
0
-63
-55
-8
0
15
15
0
0
2040
-462
-420
-42
-389
-355
-4
-30
0
0
-100
-87
-12
0
27
26
0
0
2050
-547
-497
-50
-461
-421
-4
-36
0
0
-119
-103
-15
0
32
32
0
0
                                4-21

-------
Chapter 4
        Table 4.3-17 Annual Criteria Emission Impacts of Program (short tons)


Total
Downstream
Fuel Production
and Distribution
Electricity

Pollutant
VOC
CO
NOX
PM2.5
SOX
VOC
CO
NOX
PM2.5
SOX
VOC
CO
NOX
PM2.5
SOX
VOC
CO
NOX
PM2.5
SOX
CY 2020
Impacts
(Short Tons)
-12,467
21,242
-2,449
-351
-1,650
379
22,212
779
63
-449
-12,860
-1,229
-3,846
-524
-2,353
14
259
617
110
1,153
% of Total
US Inventory
-0.1%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
-0.1%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
0.0%
CY 2030
Impacts
(Short Tons)
-135,566
397,861
-16,008
-3,123
-9,443
8,623
405,260
14,872
1,023
-5,051
-144,503
-13,810
-43,215
-5,890
-26,443
6,411
6,411
12,335
1,743
22,051
% of Total
US Inventory
-1.1%
0.7%
-0.2%
-0.1%
-0.1%
0.1%
0.7%
0.1%
0.0%
-0.1%
-1.1%
0.0%
-0.4%
-0.1%
-0.3%
0.1%
0.0%
0.1%
0.0%
0.3%
                                      4-22

-------
                                     2017 Draft Regulatory Impact Analysis
Table 4.3-18 Annual Air Toxic Emission Impacts of Program (short tons)


Total
Downstream
Fuel Production
and Distribution
Electricity

Pollutant
1,3 -Butadiene
Acetaldehyde
Acrolein
Benzene
Formaldehyde
1,3- Butadiene
Acetaldehyde
Acrolein
Benzene
Formaldehyde
1,3 -Butadiene
Acetaldehyde
Acrolein
Benzene
Formaldehyde
1,3 -Butadiene
Acetaldehyde
Acrolein
Benzene
Formaldehyde
CY 2020
Impacts
(Short Tons)
2
4
0
-15
-5
2
6
0
13
5
0
-1
0
-28
-10
0
0
0
0
0
% of Total
US Inventory
0.02%
0.00%
0.01%
-0.01%
0.00%
0.02%
0.01%
0.01%
0.01%
0.00%
0.00%
0.00%
-0.01%
-0.01%
0.00%
0.00%
0.00%
0.01%
0.00%
0.00%
CY 2030
Impacts
(Short Tons)
47
112
-6
-26
3
49
124
5
285
118
-3
-15
-15
-313
-115
2
3
4
2
1
% of Total
US Inventory
0.4%
0.2%
0.0%
0.0%
0.0%
0.4%
0.2%
0.0%
0.1%
0.1%
0.0%
0.0%
0.0%
-0.1%
-0.1%
0.0%
0.0%
0.0%
0.0%
0.0%
                               4-23

-------
Chapter 4
            4.3.4.2   Model Year Analyses

          Table 4.3-19 Projected Net GHG Deltas (MMTCO2eq per model year)
MY
2017
2018
2019
2020
2021
2022
2023
2024
2025
Total
Downstream
-24
-58
-90
-125
-181
-226
-268
-311
-354
-1,637
Upstream
(Gasoline)
-6
-14
-21
-30
-44
-56
-68
-79
-91
-408
Electricity
1
2
O
4
5
9
13
18
23
77
Total
CO2e
-29
-70
-108
-151
-220
-273
-322
-372
-422
-1,967
              Table 4.3-20 Projected Net Non-GHG Deltas (MMT per model year)
Criteria Emission Impacts of Program (short tons)
MY
2017
2018
2019
2020
2021
2022
2023
2024
2025
Sum
voc
-11,990
-27,915
-43,158
-60,523
-89,344
-114,882
-138,334
-162,591
-187,085
-835,821
CO
49,315
115,855
179,978
253,119
373,036
477,853
574,020
673,516
774,071
3,470,763
NOx
-1,324
-3,324
-5,169
-7,281
-11,199
-12,202
-12,700
-13,127
-13,404
-79,729
PM2.5
-290
-723
-1,132
-1,603
-2,450
-2,821
-3,112
-3,413
-3,705
-19,247
S02RRR
-1,569
-4,108
-6,425
-9,107
-14,516
-14,590
-13,918
-13,086
-12,006
-89,322
Model Year Lifetime Air Toxic Emissions (short tons)
MY
2017
2018
Benzene
15
35
1,3 Butadiene
7
17
Formaldehyde
7
17
Acetaldehyde
17
40
Acrolein
1
2
111111 Note that one source of SO2 emission reductions are a result of the reduction in gasoline fuel use. Existing
EPA regulations require that highway gasoline fuel must not contain more than SOppm sulfur, and the average
content must be SOppm sulfur.
                                           4-24

-------
                                               2017 Draft Regulatory Impact Analysis
2019
2020
2021
2022
2023
2024
2025
Sum
54
76
112
141
167
194
222
1,014
26
37
54
69
83
97
112
502
27
38
56
71
84
98
113
512
63
88
130
166
200
234
269
1,207
3
4
6
8
11
13
16
63
       4.3.5  Fuel Consumption Impacts

       The fuel consumption analyses relied on the same set of fleet and activity inputs as the
emission analysis. Because the OMEGA penetrations of diesel technology are small (<1% in
MY 2025), EPA modeled the entire fleet as gasoline, and used a conversion factor of 8887
grams of CC>2 per gallon petroleum gasoline in order to determine the quantity of fuel savings.
The term petroleum gasoline is used here to mean fuel with 115,000 btu/gallon. This is
different than retail fuel, which is typically blended with ethanol and has a a lower energy
content. This topic is further discussed in TSD 4.

               Table 4.3-21 Calendar Year Fuel Consumption Impacts
CY
2020
2030
2040
2050
Sum 20 17-
2050
Fuel Delta
(Billion
Gallons
petroleum
gasoline)
-2
-26
-40
-48
-942
Fuel Delta
(Billion
Barrels
petroleum
gasoline)
-0.1
-0.6
-1.0
-1.1
-22.4
Electricity
Delta
(Billion
kwh)
1
26
46
55
1,023
                                        4-25

-------
Chapter 4
                     Table 4.3-22 Model Year Fuel Consumption Impacts
MY
2017
2018
2019
2020
2021
2022
2023
2024
2025
Sum
Fuel Delta
( Billion
Gallons
petroleum
gasoline)
-2
-6
-9
-12
-18
-23
-27
-32
-37
-165
Fuel Delta
(Billion
Barrels
petroleum
gasoline)
-0.1
-0.1
-0.2
-0.3
-0.4
-0.5
-0.6
-0.8
-0.9
-3.9
Electricity Delta
(Billion kwh)
2
3
5
6
8
16
23
32
40
135
      4.3.6  GHG and Fuel Consumption Impacts from Alternatives




             Table 4.3-23 Calendar Year Impacts of Alternative Scenarios

Scenario
Primary
A - Cars +20
g/mile
B - Cars -20
g/mile
C - Trucks +20
g/mile
D - Trucks -20
g/mile
GHG Delta
(MMT2 CO2eq)
2020
-29
-20
-35
-28
-39
2030
-297
-248
-335
-275
-322
2040
-462
-396
-511
-431
-492
2050
-547
-471
-604
-510
-582
Fuel Savings
(B. Gallons petroleum
gasoline)
2020
-2.3
-1.4
-2.9
-2.2
-3.2
2030
-25.6
-20.3
-30.8
-23.0
-28.6
2040
-40.4
-33.0
-48.1
-36.5
-44.4
2050
-47.9
-39.2
-56.9
-43.3
-52.7
                                       4-26

-------
                                              2017 Draft Regulatory Impact Analysis
          Table 4.3-24 Model Year Lifetime Impacts of Alternative Scenarios
                          (Summary of MY 2017-MY2025)

Primary
A - Cars +20
g/mile
B - Cars -20
g/mile
C - Trucks +20
g/mile
D - Trucks -20
g/mile
Total
CO2e
-1,967
-1,567
-2,283
-1,788
-2,254
Fuel Delta
(bgal
petroleum
gasoline)
-165
-125
-202
-146
-194
Fuel Delta
(b. barrels
petroleum
gasoline)
-3.9
-3.0
-4.8
-3.5
-4.6
4.4 Safety Analysis

       As described in Preamble Section II.G and DRIA Chapter 3, EPA used the OMEGA
model to conduct a similar analysis of the impacts of mass reduction on vehicle safety.  After
applying these percentage increases to the estimated weight reductions per vehicle size by
model year assumed in the OMEGA model, Table 6-6 shows the results of EPA's safety
analysis separately for each model year. These are estimated increases or decreases in
fatalities over the lifetime of the model year fleet. A positive number means that fatalities are
projected to increase;  a negative number means that fatalities are projected to decrease.  For
details, see the EPA DRIA Chapter 3.

                        4.4-1 - Summary of Fatality Analysis

Reference
Case
Control
Case
Delta

Passenger
cars
Light
trucks
Total
Passenger
cars
Light
trucks
Total
Passenger
cars
Light
trucks
Total
MY
2017
135
-160
-24
133
-160
-28
-3
-1
-3
MY
2018
137
-156
-18
132
-157
-25
-5
-1
-7
MY
2019
142
-153
-12
133
-155
-22
-8
-2
-10
MY
2020
149
-154
-5
138
-156
-19
-11
-2
-13
MY
2021
155
-156
-1
141
-160
-19
-14
-4
-18
MY
2022
160
-157
3
169
-192
-24
8
-35
-27
MY
2023
166
-158
8
198
-224
-26
32
-67
-34
MY
2024
172
-158
14
230
-257
-27
58
-99
-41
MY
2025
178
-159
19
264
-292
-28
86
-133
-47
Total
1,395
-1,411
-16
1,538
-1,754
-217
143
-343
-201
                                        4-27

-------
Chapter 4



4.5  Sensitivity Cases

       4.5.1   Rebound

       EPA conducted a sensitivity analysis regarding the GHG and fuel savings benefits of
the program under different rebound rates.

       As discussed in TSD 4, the rebound effect refers to the increase in vehicle use that
results if an increase in fuel efficiency lowers the cost per mile of driving, which can
encourage people to drive slightly more.  The rebound effect is measured directly by
estimating the change in vehicle use, often expressed in terms  of vehicle miles traveled
(VMT), with respect to changes in vehicle fuel efficiency.888  However, it is a common
practice in the literature to measure the rebound effect by estimating the change in vehicle use
with respect to the fuel cost per mile driven, which depends on both vehicle fuel efficiency
and fuel  prices.TTT  When expressed as a positive percentage, these two parameters give the
ratio of the percentage increase in vehicle use that results from a percentage increase in fuel
efficiency or reduction in fuel cost per mile, respectively. For example, the 10 percent
rebound effect we assume in this proposal means that a 10 percent decrease in fuel  cost per
mile is expected to result in a 1 percent increase in VMT.UUU

       As described in TSD 4 and section 4.3.2 of this DRIA, we estimate the VMT impact
from consumer responses to changes in fuel prices and fuel efficiency in both the control and
reference cases against CY 2001 NHTS data.vvv Below, we use the same  1.1% per-vehicle
VMT growth rates as in the primary case. As shown in Equation 3 - VMT growth, varying
the rebound  rate changes both the control and reference VMT  schedules. Therefore, this
sensitivity varies the total amount of both reference and control VMT.WWW
sss Vehicle fuel efficiency is more often measured in terms of fuel consumption (gallons per mile) rather than
fuel economy (miles per gallon) in rebound estimates.
TTT Fuel cost per mile is equal to the price of fuel in dollars per gallon divided by fuel economy in miles per
gallon (or multiplied by fuel consumption in gallons per mile), so this figure declines when a vehicle's fuel
efficiency increases.
    Please note that increasing VMT by 1% in response to a 10% decrease in fuel cost per mile is not equivalent
to decreasing the benefits from the rule by 1% due to the decreased fuel consumption and GHG emissions in the
control case. To a lesser extent, the issue is also complicated due to compliance strategies that do not directly
impact fuel cost per mile, such as HFC emission reduction strategies, and the use of electric vehicles for
compliance, which do not reduce cost per mile to the same extent that gasoline technologies do.
vvv As discussed in above in 4.3.2.4, we assume consumers respond the same way to changes in fuel efficiency
and fuel prices. Consistent with this assumption, we use the same elasticity to measure consumer responses to
changes in fuel prices as we do to measure the rebound effect of consumers driving more in response to
increased fuel efficiency.
www One important validation of the VMT equations used in this analysis is a strong resemblance to total
historic VMT data, and future AEO projections. Changing the rebound rate without changing the growth rates
may weaken that relationship, which is why they must be evaluated together when parameterizing the equation.
Consequently, while this sensitivity analysis varies the rebound rate in isolation, were any of these rebound rates
to be used for the primary analysis,EPA would also revisit the per-vehicle VMT growth rates.


                                            4-28

-------
                                                2017 Draft Regulatory Impact Analysis
                          4.5-1 - Rebound Sensitivity Results

Rebound
Rate
0%
5%
10%
15%
20%
MY Lifetime
2017-2025
GHG Benefits
(MMT CO2e)
2,342
2,155
1,967
1,780
1,593
Fuel Savings
(B. Gallons)
197
181
165
149
133
CY 2030
GHG Benefits
(MMT CO2e)
352
325
297
270
242
Fuel Savings
(B. Gallons)
30
28
26
23
21
       In the analysis, EPA applies the rebound rate to the change in the fuel cost of driving
in future years relative to CY 2001 NHTS values for all MYs, all ages, and in both the
reference and control cases.  This allows the agency to directly tie the future VMT schedules
back to known source data. A major benefit of this approach is the consistency, in that future
values of the fuel cost per mile in both the reference and control cases are always compared
back against the same reference point. However, it also means that the theoretical consumer
is comparing back against 2001 driving costs, which may not be an accurate representation of
the real-world process (i.e., in practice, consumers are more likely to be internalizing a change
in the fuel cost of driving that they experienced recently, rather than 15-40 years ago).

       An alternative approach considered by the agency is to calculate the rebound effect
relative to the reference case. Thus, the reference case would be calculated relative to the
2001 data, and the control case would be calculated relative to the derived reference VMT
schedules.  Tying the rebound effect to a shifting reference point (the current year of the
reference case) would make VMT increase proportional to the difference in the fuel cost per
mile between the two cases, rather than the difference being proportional to the CY 2001 fuel
cost per mile. However, this change implies that consumers  are responding to the impact of
changes in fuel cost per mile differently between the reference case and the control case than
between the reference case and CY 2001. Hence the absolute change in VMT from a given
rebound rate is different for the control case in this alternative approach.

       The current application of the rebound effect in both the reference and control cases is
demonstrated in Equation 2.  The alternative approach for evaluating the rebound effect in the
control case would be to use  Equation 5 (below) after calculating the reference case with
Equation 2. Following this alternative approach would also allow EPA to vary the rebound
effect in the control case while holding the reference case VMT constant for sensitivity cases.
                                         4-29

-------
Chapter 4
            Equation 5 - Rebound Equation Relative to the Reference Case
       VM1 ^caiendar year x,ageY = (VR X,y ) *  (1 - R
                      PCPM                    _ PCPM
                      r L,r I'ireference Case year x,y   r L'rl ^control Calendar year x,y^
                                     FTPM                                 '
                                     i  ^.i ^'reference case year x,y


       Where:
       VR ^ = Average miles driven in a vehicle  of age y in calendar year x in the reference case
       R= Magnitude of the rebound effect, expressed as an elasticity (e.g., -0. 10)
            jj, = Fuel cost per mile of a vehicle of age y in calendar year x
       EPA may consider alternative methods of implementing the rebound effect in the final
rulemaking.

       4.5.2                EV impacts

       In section III.C.2 of the preamble, EPA presented an analysis of the GHG impacts of
the EV zero gram/mile and EV/PHEV multiplier impacts on the cumulative GHG savings
from the fleet.  In this projection, EPA varied the number of electric and plug-in hybrid
electric assumed in the future fleet.

       This projection of the impact of the EV/PHEV/FCV incentives on the overall program
GHG emissions reductions assumes that EPA would have proposed exactly the same standard
if the 0 gram per mile compliance value were not allowed for any EV/PHEV/FCVs.  While
EPA has not analyzed such a scenario, it is clear that not allowing a 0 gram per mile
compliance value would change the technology mix and cost projected for the proposed
standard.

  To conduct this analysis, EPA first ran the OMEGA model post-processor assuming
 that no vehicles operated  on wall electricity. Thus, the 2 cycle standard was simply the
  CO2 targets adjusted for air conditioning and the pickup related credits (Table 4.3-8,

       Table 4.3-9). The OMEGA scenario results were drawn from the primary analysis,
but were adjusted as for a different ratio of EVs and PHEVs, as discussed below. The final
scenario, involving 2 million EVs+PHEVs sold from 2022-2025, was modeled through the
same method as the proposal. The EV phase in schedule from the primary scenario was
multiplied by -1.495 in order to produce the phase-in corresponding  to 2.0 million EVs sold
in 2022-2025.  2 cycle performance was then adjusted accordingly for the multiplier credits
and electricity usage was included in the accounting.
                                        4-30

-------
                                               2017 Draft Regulatory Impact Analysis
       For this analysis, we assumed that 50% of the plug-in vehicles would be PHEVs, and
subtracted 25% from the total impacts of the EVs+PHEVs in order to approximate the lesser
reliance of EVs on electric power. xxx

       In table 4.5-2, the number of metric tons represents the number of additional tons that
would be reduced if the standards stayed the same and there was no 0 gram per mile
compliance value.  The percentage change represents the ratio of the cumulative decrease in
GHG emissions reductions from the prior column to the total cumulative GHG emissions
reductions associated with the proposed standards and the proposed 0 gram per mile
compliance value.

       If EPA proposed the exact same tailpipe standards, and provided no additional
flexibilities, the program impacts would be estimated at 2,180 MMT if there were no electric
vehicles or plug-in electric vehicles used for compliance.

                              4.5-2 - EV/PHEV Impacts
Scenario





No EV/PHEVs
EPA OMEGA
model
projection
EPA alternative
projection
Cumulative
EV/PHEV/FCV
Sales 2017-2025



0
1.9 million


2.8 million

Cumulative
EV/PHEV/FCV
Sales
2022-2025


0
1.3 million


2.0 million

Cumulative
Decrease in
GHG Emissions
Reductions
2017-2025

0 millon metric tons
80 million metric
tons

110 million metric
tons
Percentage
Decrease in
GHG Emissions
Reductions
2017-
2025
0
3.6%


5.4%

4.6 Calculation of Impacts from An Electric Vehicle

       As one illustrative example, using the most recent national average electricity GHG
emissions factors to calculate upstream fuel production and distribution GHG emissions, the
Nissan Leaf would have an upstream GHG emissions value of 161 grams per mile. This is
calculated as follows.
   While have PHEVs rather than EVs would also change the multiplier, this 2.0 million vehicle scenario is
meant to approximate the impacts of a larger fleet of electric vehicles.
                                         4-31

-------
Chapter 4
       The Leaf consumes 238 watt-hours of electricity per mile over the EPA city and highway
       tests.  Note that the EPA electricity consumption values reported here and on the fuel
       economy label include both the electricity needed to propel the vehicle as well as vehicle
       charging losses, which are typically on the order of 10 percent. This EPA test value is divided
       by 0.7 to get the official label value for the 2011 Leaf of 340 watt-hours per mile, or 0.34
       kilowatt-hours per mile.
       To reflect average electricity grid/transmission losses of about 7 percent, we divide the 238
       watt-hours per mile by 0.93 to get 256 watt-hours per mile, which is the amount of electricity
       that would to be generated at the powerplant to power the Leaf for one mile.
       Multiplying the 256 watt-hours/mile value by a nationwide average electricity upstream GHG
       emissions  rate (powerplant plus feedstock) of 0.628 grams GHG per watt-hour at the
       powerplantYYY to get 161 grams GHG per mile.
YYY The most recent nationwide average electricity upstream GHG rate of 0.628 grams GHG per watt-hour at the
powerplant was calculated from 2007 nationwide powerplant data for CO2, CH4, and N2O emissions from
eGRID2010 at http://www.epa.gov/cleanenergv/energy-resources/egrid/index.html, converting to CO2-e using
Global Warming Potentials of 25 for CH4 and 298 for N2O, yielding a value of 0.592 grams GHG per watt-hour
generated at the powerplant, and multiplying by a factor of 1.06 to account for GHG emissions associated with
feedstock extraction, transportation, and processing (based on Argonne National Laboratory's The Greenhouse
Gases, Regulated Emissions, and Energy Use in Transportation (GREET) Model, Version l.Sc.O, available at
http://www.transportation.anl.gov/modelingsimulation/GREET/'). EPA Docket EPA-HQ-OAR-2009-0472. Of
course, EVs sold in various areas of the country would have different upstream GHG gram per mile values. For
example, using an average California electricity upstream GHG rate of 0.349 grams GHG per watt-hour at the
powerplant, from the same EPA eGRID 2010 database, would yield a Leaf upstream GHG emissions value of
89 grams per mile.
                                             4-32

-------
                                                       2017 Draft Regulatory Impact Analysis
     References

35Historically, manufacturers have reduced precious metal loading in catalysts in order to reduce costs. See
http://www.platinum.matthev.eom/media-room/our-view-on-.-.-./tnrifting-of-precious-metals-in-autocatalvsts/
Accessed 11/08/2011. Alternatively, manufacturers could also modify vehicle calibration.

36 Intergovernmental Panel on Climate Change. Chapter 2. Changes in Atmospheric Constituents and in
Radiative Forcing. September 2007. http://www.ipcc.ch/pdf/assessment-report/ar4/wgl/ar4-wgl-chapter2.pdf.
Docket ID: EPA-HQ-OAR-2009-0472-0117

37 Lu, S., NHTSA, Regulatory Analysis and Evaluation Division, "Vehicle Survivability and Travel Mileage
Schedules," DOT HS 809 952, 8-11 (January 2006). Available at http://www-nrd.nhtsa.dot.gov/pdf/nrd-
30/NCSA/Rpts/2006/809952.pdf (last accessed Sept. 9, 2011).

38EPA MOVES 2010a. August 2010. http://www.epa.gov/otaq/models/moves/index.htm

39 Craig Harvey, EPA, "Calculation of Upstream Emissions for the GHG Vehicle Rule." 2009. Docket ID:
EPA-HQ-OAR-2009-0472-0216

40 Argonne National Laboratory.  The Greenhouse Gases, Regulated Emissions, and Energy Use in
Transportation (GREET) Model versions 1.7 and 1.8.
http://www.transportation.anl.gov/modeling  simulation/GREET/.  Docket ID: EPA-HQ-OAR-2009-0472-0215

41 OMEGA Benefits post-processor.

42 Argonne National Laboratory's The Greenhouse Gases, Regulated Emissions, and Energy Use in
Transportation (GREET) Model, Version l.Sc.O, available at
http://www.transportation.anl.gov/modeling_simulation/GREET/).  EPA Docket EPA-HQ-OAR-2009-0472.

43 EPA.  eGrid2010, http://www.epa.gov/cleanenergv/energy-resources/egrid/index.html
                                               4-33

-------

-------
                                               2017 Draft Regulatory Impact Analysis
5      Vehicle Program Costs and Fuel Savings

       In this chapter, EPA presents our estimate of the costs associated with the proposed
vehicle program.  The presentation here summarizes the vehicle level costs associated with
the new technologies expected to be added to meet the proposed GHG standards, including
hardware costs to comply with the proposed A/C credit program. The analysis summarized
here provides our estimate of incremental costs on a per vehicle basis and on an annual total
basis.

       The presentation here summarizes the outputs of the OMEGA model that were
discussed in some detail in Chapter 3 of this draft RIA.  For details behind the analysis, such
as the OMEGA model inputs and the estimates of costs associated with individual
technologies, the reader is directed to Chapter 1 of this draft RIA, and Chapter 3  of the draft
Joint TSD.
5.1 Costs per Vehicle

       To develop costs per vehicle, EPA has used the same methodology as that used in the
recent 2012-2016 final rule and the 2010 TAR. Individual technology direct manufacturing
costs have been estimated in a variety of ways—vehicle and technology tear down, models
developed by outside organizations, and literature review—and indirect costs have been
estimated using the updated and revised indirect cost multiplier (ICM) approach that was first
developed for the 2012-2016 final rule. All of these individual technology costs are described
in detail in Chapter 3 of the draft joint TSD.  Also described there are the ICMs used in this
proposal and the ways the ICMs have been updated and revised since the 2012-2016 final rule
which results in considerably higher indirect costs in this proposal than estimated  in the 2012-
2016 final rule. Further, we describe in detail the adjustments to technology costs to account
for manufacturing learning and the cost reductions that result from that learning. We note
here that learning impacts are applied only to direct manufacturing costs.  This approach
differs from the 2012-2016 final rule which applied learning to both direct and indirect costs.
Lastly, we have included costs associated with stranded capital (i.e., capital investments that
are not fully recaptured by auto makers because they would be forced to update vehicles on a
more rapid schedule than they may have intended absent this proposal). Again, this is
detailed in Chapter 3 of the draft joint TSD.

       EPA then used the technology costs to build GHG and fuel consumption reducing
packages of technologies for each of 19 different vehicle types meant to fully represent the
range of baseline vehicle technologies in the marketplace (i.e., number of cylinders, valve
train configuration, vehicle class).  This package building process as well  as the process we
use to determine the most cost effective packages for each of the 19 vehicle  types  is detailed
in Chapter 1 of this draft RIA.  These packages are then used as inputs to the OMEGA model
to estimate the most cost effective means of compliance with the proposed standards giving
due consideration to the timing required for manufacturers to implement the needed
technologies. That is, we assume that manufacturers cannot add the full suite of needed
technologies in the first year of implementation. Instead, we expect them to add technologies
                                         5-1

-------
Chapter 5
to vehicles during the typical 4 to 5 year redesign cycle. As such, we expect that every
vehicle can be redesigned to add significant levels of new technology every 4 to 5 years.
Further, we do not expect manufacturers to redesign or refresh vehicles at a pace more rapid
than the standard industry four to five year cycle.

       We then ran the OMEGA model for the 2021 and 2025 MYs as described in detail in
Chapter 3  of this draft RIA.  The control case OMEGA cost outputs for the 2021 and 2025
MYs were presented there and are repeated here in Table 5.1-1.

  Table 5.1-1 2021MY & 2025MY Control Case OMEGA Costs, including AC-Related
                        Costs but no Stranded Capital (2009$)
Company
Aston Martin
BMW
Chrysler/Fiat
Daimler
Ferrari
Ford
Geely-Volvo
GM
Honda
Hyundai
Kia
Lotus
Mazda
Mitsubishi
Nissan
Porsche
Spyker-Saab
Subaru
Suzuki
Tata-JLR
Tesla
Toyota
Volkswagen
Fleet
202 1MY
Car
$6,400
$905
$540
$1,928
$6,323
$631
$2,003
$489
$451
$602
$448
$3,314
$882
$778
$745
$5,436
$3,318
$998
$1,139
$2,194
$0
$314
$1,599
$697
Truck
$73
$860
$808
$945
$73
$736
$1,038
$651
$724
$857
$875
$73
$887
$947
$632
$1,293
$865
$887
$963
$1,606
$0
$680
$755
$728
Combined
$6,400
$893
$661
$1,683
$6,323
$667
$1,703
$569
$536
$654
$543
$3,314
$883
$837
$710
$4,460
$2,967
$971
$1,108
$1,901
$0
$457
$1,428
$708
2025MY
Car
$6,850
$2,249
$1,905
$2,926
$7,095
$2,046
$3,220
$2,193
$1,442
$1,664
$1,426
$3,708
$2,184
$2,103
$1,988
$5,825
$4,000
$2,230
$2,306
$3,242
$0
$1,381
$2,618
$1,931
Truck
$59
$1,942
$2,187
$1,942
$59
$2,443
$2,017
$1,803
$1,912
$1,960
$1,645
$59
$1,797
$2,145
$2,189
$2,038
$1,452
$2,067
$1,816
$2,625
$0
$1,598
$2,032
$1,928
Combined
$6,850
$2,168
$2,027
$2,701
$7,095
$2,168
$2,864
$2,007
$1,580
$1,723
$1,473
$3,708
$2,119
$2,117
$2,048
$5,007
$3,667
$2,194
$2,222
$2,955
$0
$1,460
$2,503
$1,930
                                         5-2

-------
                                               2017 Draft Regulatory Impact Analysis
       To get the costs per vehicle for the intervening years 2017-2020 and 2022-2024, we
have interpolated costs based on target CO2 levels for each individual company.  For this
proposal, those target CC>2 levels were presented in Chapter 3 of this draft RIA and are
repeated here for cars in Table 5.1-2 and for trucks in Table 5.1-3.

                 Table 5.1-2 Target CO2 Levels by MY for Cars (g/mi)
Company
Aston Martin
BMW
Chrysler/Fiat
Daimler
Ferrari
Ford
Geely-Volvo
GM
Honda
Hyundai
Kia
Lotus
Mazda
Mitsubishi
Nissan
Prosche
Spyker-Saab
Subaru
Suzuki
Tata-JLR
Tesla
Toyota
Volkswagen
Fleet
2016
233
239
243
248
245
239
243
241
232
233
235
216
232
230
236
216
232
226
218
260
216
231
229
235
2017
223
229
230
239
235
231
232
230
223
224
226
208
223
220
111
208
223
217
209
250
208
222
220
226
2018
214
220
221
230
226
221
223
220
214
215
216
199
214
211
218
199
214
208
201
240
199
213
211
217
2019
206
211
212
220
217
212
214
211
205
206
208
191
206
203
209
191
205
200
193
230
191
204
202
208
2020
197
202
204
211
208
204
205
203
197
198
199
183
197
194
201
183
197
192
185
221
183
196
194
200
2021
189
194
195
203
200
196
197
195
189
190
191
176
190
187
193
176
189
184
177
212
176
188
186
192
2022
182
186
187
194
191
188
189
187
181
182
183
169
182
179
185
169
181
176
170
203
169
181
179
184
2023
174
179
179
187
184
180
182
179
174
175
176
162
175
172
178
162
174
169
163
195
162
173
171
176
2024
167
172
172
179
176
173
174
172
167
168
169
156
168
165
171
156
167
163
157
187
156
166
165
169
2025
161
165
165
172
169
166
167
165
160
161
162
149
161
158
164
149
160
156
150
180
149
160
158
162
                                         5-3

-------
Chapter 5
                Table 5.1-3 Target CO2 Levels by MY for Trucks (g/mi)
Company
Aston Martin
BMW
Chrysler/Fiat
Daimler
Ferrari
Ford
Geely-Volvo
GM
Honda
Hyundai
Kia
Lotus
Mazda
Mitsubishi
Nissan
Prosche
Spyker-Saab
Subaru
Suzuki
Tata-JLR
Tesla
Toyota
Volkswagen
Fleet
2016
0
294
307
306
0
318
292
324
292
290
301
0
282
281
306
298
291
278
283
284
0
306
304
309
2017
0
295
305
311
0
317
290
321
292
289
301
0
284
278
305
298
290
275
281
282
0
304
307
307
2018
0
289
301
306
0
313
284
316
286
283
297
0
277
271
300
292
283
268
274
275
0
298
301
302
2019
0
284
296
301
0
309
279
311
282
278
292
0
272
267
295
287
278
264
269
270
0
293
296
298
2020
0
278
289
295
0
305
272
306
275
272
285
0
266
261
288
280
272
257
263
264
0
288
290
292
2021
0
261
271
277
0
287
255
286
258
255
267
0
251
244
272
262
255
241
247
247
0
270
272
274
2022
0
249
259
265
0
274
244
273
246
244
255
0
240
233
260
251
243
231
236
236
0
258
260
262
2023
0
238
247
254
0
261
233
261
236
233
244
0
230
223
248
240
233
220
225
226
0
247
249
250
2024
0
228
236
243
0
249
223
249
225
223
233
0
220
213
237
229
222
211
215
216
0
235
238
238
2025
0
218
226
232
0
237
213
238
215
213
223
0
210
204
226
219
213
201
206
206
0
225
227
228
       Interpolating the costs shown in Table 5.1-1 by CC>2 targets shown in

       Table 5.1-2 and Table 5.1-3 is straight forward enough, but the costs shown in Table
5.1-1 include our estimated AC-related costs (see Chapter 5 of the draft joint TSD). Because
2-cycle CC>2 targets do not include AC-related GHG controls, we first backed out the AC-
related costs prior to conducting the interpolations. The non-AC Costs were interpolated first
between 2016MY costs (set to $0 for the Control case) and 2021MY costs, and were
interpolated again between 2021MY and 2025MY costs. Also included in this step was a
scalar that was applied to costs in an effort to estimate the effects of learning on costs for the
intervening years. This scalar was generated by simply averaging package costs year-over-
year using the ranked-set of packages used for our 2021MY OMEGA runs and the ranked-set
of OMEGA packages for our 2025MY OMEGA runs. We note that ranked-sets of packages
and how they were developed is described in detail in Chapter 1 of this draft RIA. These
averaged package costs were then expressed as a percentage of the 2021MY costs and then
2025MY costs, respectively.  The former scalar was used for the interpolations between 2016
and 2021 while the latter scalar was used for the interpolations between 2021 and 2025.
These scalars are shown in Table 5.1-4.
                                         5-4

-------
                                                2017 Draft Regulatory Impact Analysis
      Table 5.1-4 Scalars Applied to Interpolated Costs to Reflect Learning Effects
Sealer
Costs as % of 2021
Costs as % of 2025
2017
118%
133%
2018
114%
129%
2019
105%
120%
2020
102%
116%
2021
100%
114%
2022

113%
2023

111%
2024

110%
2025

100%
      Note that scalars exclude AC-related costs.

       AC-related costs as presented in Chapter 5 of the draft joint TSD were then added
back in to the interpolated costs by year. Note that the same cost for AC was used for each
manufacturer as we do not have unique AC-related costs by manufacturer.

The final step was to include our estimates of stranded capital.  The stranded capital costs
used were based on those presented in Chapter 3 of this draft RIA where we presented
estimates of stranded capital for the 2016, 2021 and 2025 MYs.  To estimate stranded capital
for the intervening years, we have done straight line interpolations to arrive at the stranded
capital costs shown in
Table 5.1-5. Note that the same stranded capital costs were used for both cars and trucks
except that no truck stranded capital costs were included for those manufacturers with no
truck sales (Aston Martin, Ferrari, Lotus and Tesla).
         Table 5.1-5 Interpolated Estimates of Stranded Capital Costs (2009$)
Company
Aston Martin
BMW
Chrysler/Fiat
Daimler
Ferrari
Ford
Geely-Volvo
GM
Honda
Hyundai
Kia
Lotus
Mazda
Mitsubishi
Nissan
Prosche
Spyker-Saab
Subaru
Suzuki
Tata-JLR
Tesla
Toyota
Volkswagen
Fleet
2017
$66
$28
$50
$25
$12
$12
$19
$13
$6
$3
$8
$35
$14
$8
$9
$25
$40
$7
$21
$21
$0
$5
$19
$12
2018
$57
$32
$46
$25
$17
$17
$24
$14
$10
$7
$16
$30
$22
$16
$12
$25
$36
$12
$22
$24
$0
$10
$22
$16
2019
$48
$36
$42
$26
$23
$21
$29
$16
$14
$10
$25
$26
$30
$24
$14
$25
$31
$16
$24
$26
$0
$14
$25
$19
2020
$39
$40
$38
$26
$28
$25
$33
$17
$17
$14
$33
$21
$39
$32
$17
$25
$27
$20
$25
$29
$0
$19
$28
$22
2021
$29
$45
$35
$27
$34
$29
$38
$18
$21
$17
$41
$16
$47
$40
$19
$24
$22
$25
$27
$32
$0
$23
$31
$26
2022
$26
$35
$30
$23
$30
$25
$32
$19
$20
$17
$36
$12
$40
$34
$18
$20
$18
$21
$22
$29
$0
$23
$25
$23
2023
$23
$26
$25
$18
$27
$20
$25
$20
$18
$18
$31
$9
$33
$28
$16
$16
$14
$18
$16
$25
$0
$23
$18
$21
2024
$20
$16
$20
$14
$23
$15
$19
$20
$17
$18
$26
$6
$25
$22
$15
$11
$10
$14
$11
$22
$0
$23
$12
$18
2025
$17
$7
$15
$10
$19
$10
$13
$21
$15
$18
$21
$2
$18
$16
$14
$7
$6
$10
$6
$19
$0
$23
$6
$16
The end results are presented in Table 5.1-6 for cars, Table 5.1-7 for trucks and Table 5.1-8
for the combined fleet.
                                         5-5

-------
Chapter 5
   Table 5.1-6 Control Case Costs by Manufacturer by MY including AC & Stranded
                           Capital Costs - Cars (2009$)
Company
Aston Martin
BMW
Chrysler/Fiat
Daimler
Ferrari
Ford
Geely-Volvo
GM
Honda
Hyundai
Kia
Lotus
Mazda
Mitsubishi
Nissan
Prosche
Spyker-Saab
Subaru
Suzuki
Tata-JLR
Tesla
Toyota
Volkswagen
Fleet
2017
$1,708
$263
$215
$476
$1,634
$168
$553
$152
$126
$155
$123
$887
$232
$217
$193
$1,420
$893
$269
$314
$586
$0
$89
$427
$194
2018
$3,157
$470
$326
$914
$3,080
$318
$1,007
$259
$228
$293
$230
$1,635
$441
$397
$366
$2,657
$1,643
$496
$572
$1,086
$0
$161
$790
$353
2019
$4,245
$631
$411
$1,254
$4,170
$436
$1,351
$343
$312
$404
$318
$2,200
$609
$539
$502
$3,590
$2,209
$674
$772
$1,466
$0
$222
$1,070
$479
2020
$5,341
$787
$482
$1,604
$5,267
$542
$1,691
$419
$387
$507
$399
$2,764
$765
$674
$629
$4,527
$2,774
$844
$965
$1,844
$0
$275
$1,348
$595
2021
$6,424
$945
$569
$1,949
$6,351
$655
$2,035
$502
$467
$614
$483
$3,324
$924
$813
$759
$5,455
$3,335
$1,017
$1,160
$2,220
$0
$332
$1,624
$718
2022
$7,353
$1,442
$1,047
$2,482
$7,367
$1,147
$2,633
$1,066
$808
$999
$818
$3,847
$1,399
$1,292
$1,212
$6,243
$3,944
$1,497
$1,639
$2,798
$0
$682
$2,116
$1,165
2023
$7,383
$1,794
$1,401
$2,730
$7,487
$1,525
$2,938
$1,527
$1,072
$1,283
$1,076
$3,905
$1,738
$1,642
$1,543
$6,268
$4,082
$1,820
$1,943
$3,055
$0
$972
$2,371
$1,492
2024
$7,409
$2,125
$1,758
$2,950
$7,598
$1,885
$3,226
$1,975
$1,329
$1,555
$1,324
$3,960
$2,068
$1,975
$1,861
$6,291
$4,211
$2,129
$2,228
$3,297
$0
$1,250
$2,608
$1,806
2025
$6,862
$2,251
$1,914
$2,931
$7,109
$2,051
$3,228
$2,209
$1,452
$1,677
$1,442
$3,705
$2,196
$2,114
$1,997
$5,827
$4,001
$2,236
$2,307
$3,255
$0
$1,399
$2,618
$1,942
                                      5-6

-------
                                         2017 Draft Regulatory Impact Analysis
Table 5.1-7 Control Case Costs by Manufacturer by MY including AC & Stranded
                       Capital Costs - Trucks (2009$)
Company
Aston Martin
BMW
Chrysler/Fiat
Daimler
Ferrari
Ford
Geely-Volvo
GM
Honda
Hyundai
Kia
Lotus
Mazda
Mitsubishi
Nissan
Prosche
Spyker-Saab
Subaru
Suzuki
Tata-JLR
Tesla
Toyota
Volkswagen
Fleet
2017
$0
-$148
$43
-$284
$0
$106
-$117
$123
-$104
-$132
-$93
$0
-$317
-$144
$58
-$142
-$71
-$112
-$104
-$233
$0
$42
-$121
$55
2018
$0
$91
$194
-$62
$0
$243
$168
$235
$78
$93
$86
$0
-$4
$130
$194
$196
$163
$137
$159
$197
$0
$194
$70
$198
2019
$0
$253
$313
$150
$0
$339
$347
$321
$213
$248
$249
$0
$185
$298
$294
$409
$305
$292
$323
$472
$0
$301
$224
$305
2020
$0
$443
$465
$370
$0
$422
$557
$401
$371
$433
$453
$0
$378
$501
$400
$667
$471
$472
$515
$817
$0
$400
$384
$417
2021
$0
$915
$853
$956
$0
$776
$1,086
$680
$756
$884
$927
$0
$897
$998
$662
$1,328
$898
$922
$1,000
$1,648
$0
$713
$797
$764
2022
$0
$1,331
$1,352
$1,364
$0
$1,345
$1,497
$1,096
$1,194
$1,316
$1,254
$0
$1,262
$1,464
$1,184
$1,702
$1,171
$1,377
$1,364
$2,144
$0
$1,062
$1,254
$1,200
2023
$0
$1,610
$1,724
$1,629
$0
$1,814
$1,749
$1,414
$1,509
$1,608
$1,449
$0
$1,496
$1,776
$1,616
$1,886
$1,318
$1,689
$1,583
$2,403
$0
$1,312
$1,589
$1,525
2024
$0
$1,873
$2,070
$1,877
$0
$2,269
$1,984
$1,712
$1,816
$1,890
$1,636
$0
$1,734
$2,072
$2,026
$2,059
$1,456
$1,982
$1,788
$2,646
$0
$1,547
$1,914
$1,834
2025
$0
$1,959
$2,212
$1,952
$0
$2,463
$2,040
$1,834
$1,937
$1,988
$1,675
$0
$1,806
$2,171
$2,212
$2,054
$1,468
$2,087
$1,832
$2,653
$0
$1,631
$2,048
$1,954
                                    5-7

-------
Chapter 5
   Table 5.1-8 Control Case Costs by Manufacturer by MY including AC & Stranded
                       Capital Costs - Combined Fleet (2009$)
Company
Aston Martin
BMW
Chrysler/Fiat
Daimler
Ferrari
Ford
Geely-Volvo
GM
Honda
Hyundai
Kia
Lotus
Mazda
Mitsubishi
Nissan
Prosche
Spyker-Saab
Subaru
Suzuki
Tata-JLR
Tesla
Toyota
Volkswagen
Fleet
2017
$1,708
$154
$137
$287
$1,634
$147
$345
$138
$55
$97
$75
$887
$134
$91
$151
$1,052
$755
$178
$239
$178
$0
$71
$316
$146
2018
$3,157
$370
$266
$671
$3,080
$293
$746
$247
$182
$253
$198
$1,635
$362
$304
$312
$2,077
$1,431
$410
$498
$644
$0
$174
$644
$299
2019
$4,245
$531
$367
$980
$4,170
$403
$1,039
$332
$281
$372
$303
$2,200
$534
$455
$437
$2,840
$1,936
$582
$692
$972
$0
$253
$898
$418
2020
$5,341
$696
$475
$1,297
$5,267
$501
$1,339
$410
$382
$492
$411
$2,764
$696
$614
$558
$3,618
$2,444
$755
$885
$1,333
$0
$324
$1,153
$533
2021
$6,424
$937
$698
$1,702
$6,351
$696
$1,741
$590
$556
$669
$582
$3,324
$919
$877
$729
$4,482
$2,986
$994
$1,132
$1,935
$0
$481
$1,457
$734
2022
$7,353
$1,413
$1,179
$2,226
$7,367
$1,208
$2,297
$1,080
$922
$1,062
$910
$3,847
$1,377
$1,349
$1,204
$5,262
$3,582
$1,470
$1,592
$2,494
$0
$820
$1,947
$1,176
2023
$7,383
$1,746
$1,541
$2,478
$7,487
$1,614
$2,585
$1,473
$1,201
$1,347
$1,155
$3,905
$1,697
$1,687
$1,565
$5,321
$3,721
$1,790
$1,881
$2,752
$0
$1,096
$2,218
$1,503
2024
$7,409
$2,058
$1,893
$2,704
$7,598
$2,003
$2,858
$1,850
$1,472
$1,622
$1,391
$3,960
$2,012
$2,007
$1,910
$5,377
$3,851
$2,096
$2,153
$2,994
$0
$1,358
$2,472
$1,815
2025
$6,862
$2,174
$2,043
$2,707
$7,109
$2,178
$2,876
$2,030
$1,595
$1,739
$1,491
$3,705
$2,131
$2,133
$2,060
$5,012
$3,670
$2,202
$2,225
$2,976
$0
$1,483
$2,506
$1,946
       These costs per vehicle are then carried forward for future MYs to arrive at the costs
presented in Table 5.1-9, including costs associated with the air conditioning program and
estimates of stranded capital.

  Table 5.1-9 Industry Average Vehicle Costs Associated with the Proposed Standards
                                      (2009$)
Model Year
$/car
$/truck
Combined
2017
$194
$55
$146
2018
$353
$198
$299
2019
$479
$305
$418
2020
$595
$417
$533
2021
$718
$764
$734
2022
$1,165
$1,200
$1,176
2023
$1,492
$1,525
$1,503
2024
$1,806
$1,834
$1,815
2025
$1,942
$1,954
$1,946
2030
$1,926
$1,938
$1,930
2040
$1,926
$1,938
$1,929
2050
$1,926
$1,938
$1,929
5.2 Annual Costs of the Proposed National Program

       The costs presented here represent the costs for newly added technology to comply
with the proposed program incremental to the costs of the 2012-2016 standards. Together
with the projected increases in car and truck sales, the increases in per-car and per-truck
average costs shown in Table 5.1-9 above result in the total annual costs presented in Table

-------
                                               2017 Draft Regulatory Impact Analysis
5.2-1 below. Note that the costs presented in Table 5.2-1 do not include the fuel savings that
consumers would realize as a result of driving a vehicle with improved fuel economy. Those
impacts are presented in Chapter 5.4 below.  Note also that the costs presented here represent
costs estimated to occur presuming that the proposed MY 2025 standard would continue in
perpetuity.  Any changes to the proposed standards would be considered as part of a future
rulemaking. In other words, the proposed standards  do not apply only to 2017-2025 model
year vehicles - they do, in fact, apply to all 2025 and later model year vehicles.
  Table 5.2-1 Undiscounted Annual Costs & Costs of the Program Discounted back to
                   2012 at 3% and 7% Discount Rates (2009 dollars)
Calendar
Year
2017
2018
2019
2020
2021
2022
2023
2024
2025
2030
2040
2050
NPV, 3%
NPV, 7%
Sales
Cars
9,987,667
9,905,364
9,995,696
10,291,562
10,505,165
10,735,777
10,968,003
11,258,138
11,541,560
12,535,870
14,097,092
15,822,370


Trucks
5,818,655
5,671,046
5,582,962
5,604,377
5,683,902
5,703,996
5,687,486
5,675,949
5,708,899
5,986,092
6,505,226
7,301,371


$/unit
$/car
$194
$353
$479
$595
$718
$1,165
$1,492
$1,806
$1,942
$1,926
$1,926
$1,926


$/truck
$55
$198
$305
$417
$764
$1,200
$1,525
$1,834
$1,954
$1,938
$1,938
$1,938


$Million/year
Cars
$1,940
$3,500
$4,780
$6,120
$7,540
$12,500
$16,400
$20,300
$22,400
$24,100
$27,100
$30,500
$373,000
$165,000
Trucks
$322
$1,130
$1,700
$2,340
$4,340
$6,840
$8,680
$10,400
$11,200
$11,600
$12,600
$14,100
$177,000
$78,300
Combined
$2,300
$4,660
$6,510
$8,470
$11,900
$19,300
$25,000
$30,700
$33,600
$35,700
$39,800
$44,600
$551,000
$243,000
       Note that costs are estimated to decrease slightly in years beyond 2025. This
represents the elimination of stranded capital that is included in the costs for 2017 through
2025. These costs are described in detail in Chapter 3 of the draft Joint TSD.

5.3 Cost per Ton of Emissions Reduced

       EPA has calculated the cost per ton of GHG reductions associated with the proposed
GHG standards on a CC^eq basis using the costs and the emissions reductions described in
Section III.F. These values are presented in Table 5.3-1 for cars, trucks and the combined
fleet.  The cost per metric ton of GHG emissions reductions has been calculated in the years
2020, 2030,  2040, and 2050 using the annual vehicle compliance costs and emission
reductions for each of those years.  The value in 2050 represents the long-term cost per ton of
the emissions reduced. EPA has also calculated the cost per metric ton of GHG emission
reductions including the savings associated with reduced fuel consumption (presented below
in Section 5.4).  This latter calculation does not include the other benefits associated with this
program such as those associated with energy security benefits as discussed later in  Chapter 7.
By including the fuel savings, the cost per ton is generally less than $0 since the estimated
value of fuel savings considerably outweighs the program costs.
                                         5-9

-------
Chapter 5
       Table 5.3-1 Annual Cost per Metric Ton of CO2eq Reduced (2009 dollars)

Cars
Trucks
Combined
Calendar
Year
2020
2030
2040
2050
2020
2030
2040
2050
2020
2030
2040
2050
Undiscounted
Annual Costs
($millions)
$6,120
$24,100
$27,100
$30,500
$2,340
$11,600
$12,600
$14,100
$8,470
$35,700
$39,800
$44,600
Undiscounted Annual
Pre-tax Fuel Savings
($millions)
$4,840
$54,300
$91,200
$117,000
$2,340
$34,000
$57,500
$76,000
$7,180
$88,300
$149,000
$193,000
Annual CO2eq
Reduction
(mmt)
19
183
284
332
10
114
178
215
29
297
462
547
$/ton
(w/o fuel
savings)
$318
$132
$95
$92
$245
$102
$71
$66
$294
$120
$86
$81
$/ton
(w/ fuel
savings
$67
-$165
-$226
-$260
$0
-$196
-$252
-$288
$45
-$177
-$236
-$271
5.4 Reduction in Fuel Consumption and its Impacts

       5.4.1  What Are the Proj ected Changes in Fuel Consumption?

       The proposed CC>2 standards will result in significant improvements in the fuel
efficiency of affected vehicles. Drivers of those vehicles will see corresponding savings
associated with reduced fuel expenditures. EPA has estimated the impacts on fuel
consumption for both the tailpipe CO2 standards and the A/C credit program.  While gasoline
consumption would decrease under the proposed GHG standards, electricity consumption
would increase slightly due to the small penetration of EVs and PHEVs (1-3% for the 2021
and 2025 MYs).  The fuel savings includes both the gasoline consumption reductions and the
electricity consumption increases. Note that the total number of miles that vehicles are driven
each year is different under the control case than in the reference case due to the "rebound
effect," which is described in Chapter 4 of the draft joint TSD.  EPA also notes that
consumers who drive more than our average estimates for vehicle miles traveled (VMT) will
experience more fuel savings; consumers who drive less than our average VMT estimates will
experience less fuel savings.

       The expected impacts on fuel consumption are shown in Table 5.4-1.  The gallons
reduced and kilowatt hours increased (kWh) as shown in the tables reflect impacts from the
proposed CO2 standards, including the A/C credit program, and include increased
consumption resulting  from the rebound effect.
                                        5-10

-------
                                                  2017 Draft Regulatory Impact Analysis
   Table 5.4-1 Fuel Consumption Impacts of the Proposed Standards and A/C Credit
                                        Programs
Calendar Year
2017
2018
2019
2020
2021
2022
2023
2024
2025
2030
2040
2050
Total
Petroleum-based
Gasoline Reference
(million gallons)
130,544
129,503
128,680
128,229
128,387
128,599
129,312
130,087
131,289
140,602
159,582
184,136
5,079,096
Petroleum-based
Gasoline Reduced
(million gallons)
194
641
1,326
2,277
3,673
5,424
7,520
9,919
12,658
25,581
40,391
47,883
941,839
Electricity Increased
(million kWh)a
115
345
695
1,177
1,796
2,952
4,673
6,980
9,911
24,298
42,369
51,123
951,392
             1 Electricity increase by vehicles not by power plants.
       5.4.2  What are the Fuel Savings to the Consumer?

       Using the fuel consumption estimates presented in Section 5.4.1, EPA can calculate
the monetized fuel savings associated with the proposed standards. To do this, we multiply
reduced fuel consumption in each year by the corresponding estimated average fuel price in
that year, using the reference case taken from the AEO 2011 Final Release.zzz AEO is a
standard reference used by NHTSA and EPA and many other government agencies to
estimate the projected price of fuel. The agencies also used the AEO's fuel price estimate for
the 2012-2016 rulemaking.

       However, these estimates do not account for the significant uncertainty in future fuel
prices. AEO also provides a "low" fuel price case and a "high" fuel price case. The
monetized fuel savings would be understated if actual fuel prices are higher, or overstated if
fuel prices are lower, than estimated.AAAA In addition, since future fuel prices are not known
with certainty, there could be a distribution of possible fuel price outcomes, as opposed to a
zzz In the Preface to AEO 2011, the Energy Information Administration describes the reference case used in
AEO 2011. They state that, "Projections by El A are not statements of what will happen but of what might
happen, given the assumptions and methodologies used for any particular scenario. The Reference case
projection is a business-as-usual trend estimate, given known technology and technological and demographic
trends.
AAAA While EPA did not conduct an uncertainty analysis on the future price of fuel, NHTSA has conducted both
a sensitivity analysis on fuel prices and a probabilistic uncertainty analysis where fuel price is one of the
uncertain parameters (See Chapters X and XII of NHTSA's DRIA). Because the agencies' analyses are
generally consistent and feature similar parameters, the results of NHTSA's sensitivity and uncertainty analyses
are indicative of the uncertainty present in EPA's results.
                                           5-11

-------
Chapter 5
set of known higher price- and a set of known lower price-pathways. For the final rule, EPA
may to do a sensitivity analysis on future fuel prices.

       EPA's assessment uses both the pre-tax and post-tax gasoline prices.  Since the post-
tax gasoline  prices are the prices paid at fuel pumps, the fuel savings calculated using these
prices represent the savings consumers would see. The pre-tax fuel savings are those savings
that society would see.  Assuming no change in gasoline tax rates, the difference between
these two columns represents the reduction in fuel tax revenues that will be received by state
and federal governments - about $82 million in 2017 and $17 billion by 2050. These results
are shown in Table 5.4-1. Note that in Chapter 7 of this DRIA, the overall benefits and costs
of the proposal are presented and, for that reason, only the pre-tax fuel savings are presented
there.

Table 5.4-1 Undiscounted Annual Fuel Savings & Fuel Savings Discounted back to 2012
                at 3% and 7% Discount Rates (millions of 2009 dollars)
Calendar
Year
2017
2018
2019
2020
2021
2022
2023
2024
2025
2030
2040
2050
NPV, 3%
NPV, 7%
Gasoline
Savings
(pre-tax)
$581
$1,950
$4,120
$7,180
$11,600
$17,400
$24,400
$32,700
$42,400
$88,300
$149,000
$193,000
$1,550,000
$596,000
Gasoline
Savings
(taxed)
$663
$2,230
$4,670
$8,110
$13,100
$19,700
$27,500
$36,800
$47,200
$98,100
$164,000
$210,000
$1,720,000
$660,000
Electricity
Costs
$11.1
$32.8
$66.0
$113
$172
$286
$458
$691
$1,000
$2,550
$4,850
$6,350
$47,800
$17,800
Total Fuel
Savings
(pre-tax)
$570
$1,920
$4,060
$7,060
$11,400
$17,100
$24,000
$32,000
$41,400
$85,800
$144,000
$187,000
$1,510,000
$579,000
Total Fuel
Savings
(taxed)
$652
$2,200
$4,600
$7,990
$12,900
$19,400
$27,000
$36,100
$46,200
$95,600
$159,000
$204,000
$1,670,000
$642,000
Annual values represent undiscounted values; net present values represent annual costs discounted to 2012.
       As shown in Table 5.4-1, the agencies are projecting that consumers would realize
very large fuel savings as a result of the proposed standards. These calculations are based on
the assumption, discussed in Preamble Section HID., that the fuel economy of vehicles would
be constant at MY 2016 levels in the absence of the rule. As discussed further in Chapter
8.1.2.6 of this DRIA, it is a conundrum from an economic perspective that these large fuel
savings have not been provided by automakers and purchased by consumers. A number of
behavioral and market phenomena may lead to this disparity between the fuel economy that
makes financial sense to consumers. Regardless how consumers make their decisions on how
much fuel economy to purchase, EPA expects that, in the aggregate, they will gain these fuel
savings, which will provide actual money in consumers' pockets.
                                        5-12

-------
                                                2017 Draft Regulatory Impact Analysis
5.5 Consumer Payback Period and Lifetime Savings on New Vehicle Purchases

       Another factor of interest is the payback period that consumers would experience on
the purchase of a new vehicle that meets the proposed standards. In other words, how long
would it take for the expected fuel savings to outweigh the increased cost of a new vehicle?
For example, a new 2025 MY vehicle is estimated to cost $1,946 more (on average, and
relative to the reference case vehicle) due to the addition of new GHG reducing/fuel economy
improving technology.  This new technology will result in lower fuel consumption and,
therefore, savings in fuel expenditures.  But how many months or years would pass before the
fuel savings exceed the upfront costs?

       Table 5.5-1 provides the answer to this question for a vehicle purchaser who pays for
the new vehicle upfront in cash (we discuss later in this section the payback period for
consumers who finance the new vehicle purchase with a loan).  The table uses annual miles
driven (vehicle miles traveled, or VMT) and survival rates consistent with the emission and
benefits analyses presented in Chapter 4 of the draft Joint TSD.  The control case includes
fuel savings associated with A/C controls.  Not included here are the possible A/C-related
maintenance savings as discussed in Chapter 5 of the draft joint TSD.  Further,  this analysis
does not include other private impacts, such as reduced refueling events, or other societal
impacts, such as the potential rebound miles driven or the value of driving those rebound
miles, or noise, congestion and accidents, since the focus is meant to be on those factors
consumers think about most while in the showroom considering a new car purchase. To
estimate the upfront vehicle cost (i.e., the lifetime increased cost discounted back to
purchase), we have included not only the sales tax on the new car purchase but also the
increased insurance premiums that would result from the more valuable vehicle.44 Car/truck
fleet weighting is handled as described in Chapter 1 of the draft Joint TSD. The present value
of the increased vehicle costs shown in the table are $2,189 at a 3% discount rate and $2,180
at a 7% discount rate. As can be seen in the table, it will take just over 3.5 years at a 3%
discount rate, and just under 4 years at a 7% discount rate, for the cumulative discounted fuel
savings to exceed the present value of increased vehicle costs.

    Table 5.5-1 Payback Period on a 2025 MY New Vehicle  Purchase via Cash (2009
                                        dollars)
Year of
Ownership
1
2
3
4
5
6
7
8
Increased Vehicle
Cost3
(undiscounted)
-$2,087
-$31
-$26
-$21
-$16
-$11
-$6
-$1
Annual Fuel
Savings'3
(undiscounted)
$643
$634
$630
$614
$601
$572
$543
$512
Cumulative
Discounted Fuel
Savings at 3%
$634
$1,240
$1,826
$2,379
$2,906
$3,392
$3,840
$4,250
Cumulative
Discounted Fuel
Savings at 7%
$622
$1,195
$1,728
$2,213
$2,656
$3,051
$3,401
$3,709
  a Increased vehicle cost due to the rule is $1,946; the value here includes nationwide average sales tax of
  5.32% and increased insurance premiums of 1.85% in year one decreasing to 0% by year 9. Both of these
  percentages are discussed in Section 8.1.1 of this DRIA. This results in a present value of increased costs
  of $2,189 at 3% discounting and $2,180 at 7% discounting. These present value costs are used in
                                         5-13

-------
Chapter 5

  determining the payback period.
  b Calculated using AEO 2011 reference case fuel prices including taxes.

       However, most people purchase a new vehicle using credit rather than paying cash up
front. A common car loan today is a five year, 60 month loan.  As of July, 2011, the national
average interest rate for a 5 year new car loan was 5.52 percent.45  If the increased vehicle
cost is spread out over 5 years at 5.52 percent, the analysis would look like that shown in
Table 5.5-2. As can be seen in this table, the fuel savings immediately outweigh the increased
payments on the car loan, amounting to $145 in discounted net savings (3% discount rate) in
the first year and similar  savings for the next four years although savings decline somewhat
due to reduced VMT as the average vehicle ages.  Results are similar using a 7% discount
rate. This means that for every month that the average owner is making a payment for the
financing of the average new vehicle their monthly fuel savings would be greater than the
increase in the loan payments. This amounts to a savings on the order  of $12 per month
throughout the duration of the 5 year loan. Note that in year six when the car loan is paid off,
the net savings  equal the  fuel savings less the increased insurance premiums (as would be the
case for the remaining years of ownership).

    Table 5.5-2 Payback Period on a 2025 MY New Vehicle  Purchase via Credit (2009
                                        dollars)
Year of
Ownership
1
2
3
4
5
6
7
8
Increased Vehicle
Cost3
(undiscounted)
-$489
-$488
-$487
-$485
-$484
-$11
-$6
-$1
Annual Fuel
Savings'3
(undiscounted)
$643
$634
$630
$614
$601
$572
$543
$512
Annual Discounted
Net Savings at 3%c
$145
$133
$127
$109
$96
$477
$443
$409
Annual Discounted
Net Savings at 7% c
$133
$117
$107
$88
$74
$387
$346
$308
  a This uses the same increased cost as Table 5.5-1 but spreads it out over 5 years assuming a 5 year car loan
  at 5.52 percent.
  b Calculated using AEO 2011 reference case fuel prices including taxes.
  0 Note that the cumulative discounted fuel savings are identical to those shown in Table 5.5-1. Here we
  show discounted net savings.

       The lifetime fuel savings and net savings can also be calculated for those who
purchase the vehicle using cash and for those who purchase the vehicle with credit. This
calculation applies to the vehicle owner who retains the vehicle for its entire life and drives
the vehicle each year at the rate equal to the national projected average.  The results are
shown in Table 5.5-3.  In either case, the present value of the lifetime net savings is greater
than $4,200 at a 3% discount rate, or $2,900 at a 7% discount rate.
                                          5-14

-------
                                               2017 Draft Regulatory Impact Analysis
Table 5.5-3 Lifetime Discounted Net Savings on a 2025 MY New Vehicle Purchase (2009
                                       dollars)
Purchase
Option
Increased Discounted
Vehicle Cost
Lifetime Discounted
Fuel Savings b
Lifetime Discounted
Net Savings
3% discount rate
Cash
Credit a
$2,189
$2,310
$6,568
$6,568
$4,378
$4,258
7% discount rate
Cash
Credit a
$2,180
$2,147
$5,154
$5,154
$2,972
$3,004
          a Assumes a 5 year loan at 5.52 percent.
          bFuel savings here were calculated using AEO 2011 reference case fuel prices
          including taxes.

       Note that throughout this consumer payback discussion, the analysis reflects the
average number of vehicle miles traveled per year.  Drivers who drive more miles than the
average would incur fuel-related savings more quickly and, therefore, the payback would
come sooner. Drivers who drive fewer miles than the average would incur fuel related
savings more slowly and, therefore, the payback would come later.
                                         5-15

-------
Chapter 5
                                         References

44 U.S. Department of Energy, 2011. "Transportation and the Economy," Chapter 10 in
"Transportation Energy Data Book," http://cta.ornl.gov/data/tedb30/Edition30_Chapter!0.pdf,
accessed 8/22/11, Table 14..

45 "National Auto Loan Rates for July 21, 2011,"
http://www.bankrate.com/fmance/auto/national-auto-loan-rates-for-july-21 -2011. aspx,
accessed 7/26/11.
                                         5-16

-------
                                                2017 Draft Regulatory Impact Analysis
6      Health and Environmental Impacts

6.1 Health and Environmental Impacts of Non-GHG Pollutants

       6.1.1   Health Effects Associated with Exposure to Non-GHG Pollutants

       In this section we will discuss the health effects associated with non-GHG pollutants,
specifically: paniculate matter, ozone, nitrogen oxides (NOx), sulfur oxides (SOx), carbon
monoxide and air toxics. These pollutants will not be directly regulated by the standards, but
the standards will affect emissions of these pollutants and precursors.

       6.1.1.1 Background on Particulate Matter

       Particulate matter (PM) is a generic term for a broad class of chemically and
physically diverse substances. It can be principally characterized as discrete particles that
exist in the condensed (liquid or solid) phase spanning several orders of magnitude in size.
Since 1987, EPA has delineated that subset of inhalable particles small enough to penetrate to
the thoracic region (including the tracheobronchial and alveolar regions) of the respiratory
tract (referred to as thoracic particles).BBBB Current National  Ambient Air Quality Standards
(NAAQS) use PM2.5 as the indicator for fine particles (with PM2.5 generally referring to
particles with a nominal mean aerodynamic diameter less than or equal to 2.5 micrometers
(jim)), and use PMio as the indicator for purposes of regulating the coarse fraction of PMio
(referred to as thoracic coarse particles or coarse-fraction particles; generally including
particles with a nominal mean aerodynamic diameter greater than 2.5 jim and less than or
equal to 10 jim, or PMio-2.5). Ultrafine particles (UFPs) are a subset of fine particles,
generally less than  100 nanometers (0.1 um) in diameter.

       Particles span many sizes and shapes and consist of numerous different components.
Particles originate from sources and are also formed through atmospheric chemical reactions;
the former are often referred to as "primary" particles, and the latter as "secondary" particles.
In addition, there are also physical, non-chemical  reaction mechanisms that contribute to
secondary particles. Particle pollution also varies by time of year and location and is affected
by several weather-related  factors, such as temperature, clouds, humidity,  and wind.  A
further layer of complexity comes from a particle's ability to shift between solid/liquid and
gaseous phases, which is influenced by concentration, meteorology, and temperature.

       Fine particles are produced primarily by combustion processes and by transformations
of gaseous emissions (e.g., SOx, NOx and volatile organic compounds (VOCs)) in the
atmosphere. The chemical  and physical properties of PM2.5 may vary greatly with time,
region, meteorology and source category. Thus, PM2.5 may include a complex  mixture of
different components including sulfates, nitrates, organic compounds, elemental carbon and
metal compounds.  These particles can remain in the atmosphere for days to weeks and travel
through the atmosphere hundreds to thousands of kilometers.46
BBBB Regulatory definitions of PM size fractions, and information on reference and equivalent methods for
measuring PM in ambient air, are provided in 40 CFR Parts 50, 53, and 58.


                                         6-1

-------
Chapter 6

       6.1.1.2 Particulate Matter Health Effects

       This section provides a summary of the health effects associated with exposure to
ambient concentrations of PM.CCCC  The information in this section is based on the
information and conclusions in the Integrated Science Assessment (ISA) for Particulate
Matter (December 2009) prepared by EPA's Office of Research and Development
(ORD)DDDD

       The ISA concludes that ambient concentrations of PM are associated with a number of
adverse health effects.EEEE The ISA characterizes the weight of evidence for different health
effects associated with three PM size ranges:  PM2.5, PMio-2.s, and UFPs. The discussion
below highlights the ISA's conclusions pertaining to these three size fractions of PM,
considering variations in health effects associated with both short-term and long-term
exposure periods.

       6.1.1.2.1      Effects Associated with Short-term Exposure to PM2.5

       The ISA concludes that cardiovascular effects and mortality are causally associated
with short-term exposure to PM2.5.47  It also concludes that  respiratory effects are likely to be
causally associated with short-term exposure to PM2 5, including respiratory emergency
department (ED) visits and hospital admissions for chronic  obstructive pulmonary disease
(COPD), respiratory infections, and asthma; and exacerbation of respiratory symptoms in
asthmatic children.

       6.1.1.2.2      Effects Associated with Long-term Exposure to PM2 5

       The ISA concludes that there are causal associations between long-term exposure to
PM2.s and cardiovascular effects, such as the development/progression of cardiovascular
disease (CVD), and premature mortality,  particularly from cardiovascular causes.48  It also
concludes that long-term exposure to PM2 5 is likely to be causally associated with respiratory
effects, such as reduced lung function growth, increased respiratory symptoms, and asthma
development.  The  ISA characterizes the  evidence as suggestive of a causal relationship for
associations between long-term PM2.5 exposure and reproductive and  developmental
outcomes,  such as low birth weight and infant mortality.  It also characterizes the evidence as
suggestive of a causal relationship between PM2 5 and cancer incidence, mutagenicity, and
genotoxicity.
cccc personaj exposure includes contributions from many different types of particles, from many sources, and in
many different environments. Total personal exposure to PM includes both ambient and nonambient
components and collectively these components may contribute to adverse health effects.
DDDD  -pj^ j<^ -g avaiiabie at http://cfpub.epa.gov/ncea/cfm/recordisplay.cfm?deid=216546
EEEE -pj^ j<^ evaluates the health evidence associated with different health effects, assigning one of five "weight
of evidence" determinations: causal relationship, likely to be a causal relationship, suggestive of a causal
relationship, inadequate to infer a causal relationship, and not likely to be a causal relationship. For definitions
of these levels of evidence, please refer to Section 1.5 of the ISA.
                                           6-2

-------
                                                2017 Draft Regulatory Impact Analysis
       6.1.1.2.3      Effects Associated with PMi0-2.5

       The ISA summarizes evidence related to short-term exposure to PMi0.2.5. PMi0-2.5 is
the fraction of PMio particles that is larger than PM2.5.49 The ISA concludes that available
evidence is suggestive of a causal relationship between short-term exposures to PMi0-2.5 and
cardiovascular effects.  It also concludes that the available evidence is suggestive of a causal
relationship between short-term exposures to PMi0-2.5 and respiratory effects, including
respiratory-related ED visits and hospitalizations. The ISA also concludes that the available
literature suggests a causal relationship between short-term exposures to PMi0-2.5 and
mortality. Data are inadequate to draw conclusions regarding health effects  associated with
long-term exposure to PMi0-2.5.5°

       6.1.1.2.4      Effects Associated with Ultrafine Particles

       The ISA concludes that the evidence is suggestive of a causal relationship between
short-term exposures to UFPs and cardiovascular effects, including changes  in heart rhythm
and vasomotor function (the ability of blood vessels to expand and contract).51

       The ISA also concludes that there is suggestive evidence of a causal relationship
between short-term UFP exposure and respiratory effects.  The types of respiratory effects
examined in epidemiologic studies include respiratory symptoms and asthma hospital
admissions, the results of which are not entirely consistent. There is evidence from
toxicological and controlled human exposure studies that exposure to UFPs may increase lung
inflammation and produce small asymptomatic changes in lung function. Data are inadequate
to draw conclusions regarding health effects associated with long-term exposure to UFPs.52

       6.1.1.3 Background on Ozone

       Ground-level ozone pollution is typically formed by the reaction of VOCs and NOx in
the lower atmosphere in the presence of sunlight.  These pollutants, often referred to as ozone
precursors, are emitted  by many types of pollution sources such as highway  and nonroad
motor vehicles and engines, power plants, chemical plants, refineries, makers of consumer
and commercial products, industrial facilities, and smaller area sources.

       The science of ozone formation, transport, and accumulation is complex.  Ground-
level ozone is produced and destroyed in a cyclical set of chemical reactions, many of which
are sensitive to temperature and sunlight. When ambient temperatures and sunlight levels
remain high for several days and the air is relatively stagnant,  ozone and its precursors can
build up and result in more ozone than typically occurs on a single high-temperature day.
Ozone can be transported hundreds of miles downwind of precursor emissions, resulting in
elevated ozone levels even in areas with low VOC or NOx emissions.

       The highest levels of ozone are produced when both VOC and NOx emissions are
present in significant quantities on clear summer days.  Relatively small amounts of NOx
enable ozone to form rapidly when VOC levels are relatively high, but ozone production is
quickly limited by removal  of the NOx. Under these conditions NOx reductions are highly
effective in reducing ozone while VOC reductions have little effect.  Such conditions are
called "NOx-limited."  Because the contribution of VOC emissions from biogenic (natural)
                                         6-3

-------
Chapter 6

sources to local ambient ozone concentrations can be significant, even some areas where man-
made VOC emissions are relatively low can be NOx-limited.

       Ozone concentrations in an area also can be lowered by the reaction of nitric oxide
(NO) with ozone, forming nitrogen dioxide (NO2); as the air moves downwind and the cycle
continues, the NO2 forms additional ozone. The importance of this reaction depends, in part,
on the relative concentrations of NOX, VOC, and ozone, all of which change with time and
location. When NOx levels are relatively high and VOC levels relatively low, NOx forms
inorganic nitrates (i.e., particles) but relatively little ozone. Such conditions are called "VOC-
limited." Under these conditions, VOC reductions are effective in reducing ozone, but NOx
reductions can actually increase local ozone under certain circumstances. Even in VOC-
limited urban areas, NOx reductions are not expected to increase ozone levels if the NOx
reductions are sufficiently large. Rural areas are usually NOx-limited, due to the relatively
large amounts of biogenic VOC emissions in such areas. Urban areas can be either VOC- or
NOx-limited, or a mixture of both, in which ozone levels exhibit moderate sensitivity to
changes in either pollutant.

       6.1.1.4 Ozone Health Effects

       Exposure to ambient ozone contributes to a wide range of adverse health effects.FFFF
These health effects are well documented and are critically assessed in the EPA ozone air
quality criteria document (ozone AQCD) and EPA staff paper.53'54 We are relying on the data
and conclusions in the ozone AQCD and staff paper, regarding the health effects associated
with ozone exposure.

       Ozone-related health effects include lung function decrements, respiratory symptoms,
aggravation of asthma, increased hospital and emergency room visits, increased asthma
medication usage, and a variety of other respiratory effects.  Cellular-level effects,  such as
inflammation of lungs, have been documented as well.  In addition, there is suggestive
evidence of a contribution of ozone to cardiovascular-related morbidity and highly suggestive
evidence that short-term ozone exposure directly or indirectly contributes to non-accidental
and cardiopulmonary-related mortality, but additional research is needed to clarify the
underlying mechanisms causing these effects. In a report on the estimation of ozone-related
premature mortality published by  the National Research Council (NRC), a panel of experts
and reviewers concluded that short-term exposure to ambient ozone is likely to contribute to
premature deaths and that ozone-related mortality should be included in estimates of the
health benefits of reducing ozone  exposure.55 People who appear to be more susceptible to
effects associated with exposure to ozone include children, asthmatics and the elderly. Those
with greater exposures to ozone, for instance due to time spent outdoors (e.g., children and
outdoor workers), are also of concern.
FFFF Human exposure to ozone varies over time due to changes in ambient ozone concentration and because
people move between locations which have notable different ozone concentrations. Also, the amount of ozone
delivered to the lung is not only influenced by the ambient concentrations but also by the individuals breathing
route and rate.
                                          6-4

-------
                                                2017 Draft Regulatory Impact Analysis
       Based on a large number of scientific studies, EPA has identified several key health
effects associated with exposure to levels of ozone found today in many areas of the country.
Short-term (1 to 3 hours) and prolonged exposures (6 to 8 hours) to ambient ozone
concentrations have been linked to lung function decrements, respiratory symptoms, increased
hospital admissions and emergency room visits for respiratory problems.56'57'58'59'60'61
Repeated exposure to ozone can increase susceptibility to respiratory infection and lung
inflammation and can aggravate preexisting respiratory diseases, such as asthma.62'63'64'65'66
Repeated exposure to sufficient concentrations of ozone can also cause inflammation of the
lung, impairment of lung defense mechanisms, and possibly irreversible changes in lung
structure, which over time could affect premature aging of the lungs and/or the development
of chronic respiratory illnesses, such as emphysema and chronic bronchitis.67'68'69'70

       Children and outdoor workers tend to have higher ozone exposure because they
typically are active outside, working, playing and exercising, during times of day and seasons
(e.g., the summer) when ozone levels are highest.71  For example, summer camp studies have
reported statistically significant reductions in lung function in children who are active
outdoors.72'73'74'75'76'77'78'79 Further, children are more at risk of experiencing health effects
from ozone exposure than adults because their respiratory systems are still developing. These
individuals (as well as people with respiratory illnesses, such as asthma, especially asthmatic
children) can experience reduced lung function and increased respiratory symptoms, such as
chest pain and cough, when exposed to relatively low ozone levels during prolonged periods
of moderate exertion.80'81'82'83

       6.1.1.5 Background on Nitrogen Oxides and Sulfur Oxides

       Sulfur dioxide (802), a member of the sulfur oxide (SOx) family of gases, is formed
from burning fuels containing sulfur (e.g., coal or oil), extracting gasoline from oil,  or
extracting metals from ore.  Nitrogen dioxide (NO2) is a member of the nitrogen oxide (NOx)
family of gases.  Most NO2 is formed in the air through the oxidation of nitric oxide (NO)
emitted when fuel is burned at a high temperature. SO2 andNO2 can dissolve in water
droplets and further oxidize to form sulfuric and nitric acid which react with ammonia to form
sulfates and nitrates, both of which are important components of ambient PM. The  health
effects of ambient PM are discussed in Section 6.1.1.2. NOx along with non-methane
hydrocarbons (NMHC) are the two major precursors of ozone. The health effects of ozone
are covered in Section 6.1.1.4.

       6.1.1.6 Health Effects of SO2

       This section provides an overview of the health effects associated with SO2.
Additional information on the health effects of SO2 can be found in the EPA Integrated
                                    Q A
Science Assessment for Sulfur Oxides.   Following an extensive evaluation of health
evidence from epidemiologic and laboratory studies, the U.S. EPA has concluded that there is
a causal relationship between respiratory health effects and short-term (from 5 minutes to 24
hours) exposure to SO2. The immediate effect of SO2 on the respiratory system in humans is
bronchoconstriction. Asthmatics are more sensitive to the effects of SO2 likely resulting from
preexisting inflammation associated with this disease. In laboratory studies involving
controlled human exposures to SO2, respiratory effects have consistently been observed
                                          6-5

-------
Chapter 6

following 5-10 min exposures at SC>2 concentrations > 0.4 ppm in asthmatics engaged in
moderate to heavy levels of exercise, with more limited evidence of respiratory effects among
exercising asthmatics exposed to concentrations as low as 0.2-0.3 ppm. A clear
concentration-response relationship has been demonstrated in these studies following
exposures to SC>2 at concentrations between 0.2 and 1.0 ppm, both in terms of increasing
severity of respiratory symptoms and decrements in lung function, as well as the percentage
of asthmatics adversely affected.

       In epidemiologic studies, respiratory effects have been observed in areas where the
mean 24-hour SC>2 levels range from 1 to 30 ppb, with maximum 1 to 24-hour average SC>2
values ranging from 12 to 75 ppb.  Important new multicity studies and several other studies
have found an association between 24-hour average ambient 862 concentrations and
respiratory symptoms in children, particularly those with asthma. Generally consistent
associations also have been observed between ambient SC>2 concentrations and emergency
department visits and hospitalizations for all respiratory causes, particularly among children
and older adults (> 65  years), and for asthma. A limited subset of epidemiologic studies have
examined potential confounding by copollutants using multipollutant regression models.
These analyses indicate that although copollutant adjustment has varying degrees of influence
on the SO2 effect estimates, the effect of SO2 on respiratory health outcomes  appears to be
generally robust and independent of the effects of gaseous and particulate copollutants,
suggesting that the observed effects of SC>2 on respiratory endpoints occur independent of the
effects of other ambient air pollutants.  In addition, this epidemiologic evidence is plausible
and coherent given the consistency of the effects observed in the epidemiologic and controlled
human exposure studies along with toxicological evidence related to the mode of action of
SC>2 on the human respiratory system.

       Consistent associations between short-term exposure to SC>2 and mortality have been
observed in epidemiologic studies, with larger effect estimates reported for respiratory
mortality than for cardiovascular mortality.  While this finding is consistent with the
demonstrated effects of SC>2 on respiratory morbidity, uncertainty remains with respect to the
interpretation of these associations due to potential confounding by various copollutants.  The
U.S. EPA has therefore concluded that the overall evidence is suggestive of a causal
relationship between short-term exposure to 862 and mortality.  Significant associations
between short-term exposure to  SC>2 and emergency department visits and hospital admissions
for cardiovascular diseases have also been reported. However, these findings have been
inconsistent across studies and do not provide adequate evidence to infer a causal relationship
between 862 exposure and cardiovascular morbidity.

       6.1.1.7 Health Effects of NO2

       Information on the health effects of NC>2 can be found in the EPA Integrated Science
Assessment (ISA) for Nitrogen Oxides.85  The EPA has concluded that the findings of
epidemiologic, controlled human exposure, and animal toxicological studies provide evidence
that is sufficient to infer a likely causal relationship between respiratory effects and short-term
NC>2 exposure. The ISA concludes that the strongest evidence for such a relationship comes
from epidemiologic studies of respiratory effects including symptoms, emergency department
visits, and hospital admissions.  The ISA also draws two broad conclusions regarding  airway
                                          6-6

-------
                                                 2017 Draft Regulatory Impact Analysis
responsiveness following NC>2 exposure. First, the ISA concludes that NC>2 exposure may
enhance the sensitivity to allergen-induced decrements in lung function and increase the
allergen-induced airway inflammatory response following 30-minute exposures of asthmatics
to NC>2 concentrations as low as 0.26 ppm. Second, exposure to NC>2 has been found to
enhance the inherent responsiveness of the airway to subsequent nonspecific challenges in
controlled human exposure studies of asthmatic subjects. Small but significant increases in
non-specific airway hyperresponsiveness were reported following 1-hour exposures of
asthmatics to 0.1 ppm NC>2.  Enhanced airway responsiveness could have important clinical
implications for asthmatics since transient increases in airway responsiveness following NC>2
exposure have the potential to increase symptoms and worsen asthma control.  Together, the
epidemiologic and experimental data sets form a plausible, consistent, and coherent
description of a relationship between NC>2 exposures and an array of adverse health effects
that range from the onset of respiratory symptoms to hospital admission.

       Although the weight of evidence supporting a causal  relationship is somewhat less
certain than that associated with respiratory morbidity, NC>2 has also been linked to other
health endpoints. These include all-cause (non-accidental) mortality, hospital admissions or
emergency department visits for cardiovascular disease,  and  decrements in lung function
growth associated with chronic exposure.

       6.1.1.8 Health Effects of Carbon Monoxide

       Information on the health effects of carbon monoxide (CO) can be found in the EPA
Integrated Science Assessment (ISA) for Carbon Monoxide.86  The ISA concludes that
                                                                               f~~*
ambient concentrations of CO are associated with a number of adverse health effects.
This section provides a summary of the health effects associated with exposure to ambient
concentrations  of COHHHH

       Human clinical studies of subjects with coronary artery disease show a decrease in the
time to onset of exercise-induced angina (chest pain) and electrocardiogram changes
following CO exposure. In addition, epidemiologic studies show associations between short-
term CO exposure and cardiovascular morbidity, particularly increased emergency room visits
and hospital admissions for coronary heart disease (including ischemic heart disease,
myocardial infarction, and angina).  Some epidemiologic evidence is also available for
increased hospital admissions and emergency room visits for congestive heart failure and
cardiovascular  disease as a whole.  The ISA concludes that a causal relationship is likely to
exist between short-term exposures to CO and cardiovascular morbidity.  It also concludes
      The ISA evaluates the health evidence associated with different health effects, assigning one of five
"weight of evidence" determinations: causal relationship, likely to be a causal relationship, suggestive of a
causal relationship, inadequate to infer a causal relationship, and not likely to be a causal relationship. For
definitions of these levels of evidence, please refer to Section 1.6 of the ISA.
101101 Personal exposure includes contributions from many sources, and in many different environments. Total
personal exposure to CO includes both ambient and nonambient components; and both components may
contribute to adverse health effects.
                                          6-7

-------
Chapter 6

that available data are inadequate to conclude that a causal relationship exists between long-
term exposures to CO and cardiovascular morbidity.

       Animal studies show various neurological effects with in-utero CO exposure.
Controlled human exposure studies report inconsistent neural and behavioral effects following
low-level CO exposures.  The ISA concludes the evidence is suggestive of a causal
relationship with both short- and long-term exposure to CO and central nervous system
effects.

       A number of epidemiologic and animal toxicological studies cited in the ISA have
evaluated associations between CO exposure and birth outcomes such as preterm birth or
cardiac birth defects. The epidemiologic studies provide limited evidence of a CO-induced
effect on preterm births and birth defects, with weak evidence for a decrease in birth weight.
Animal toxicological studies have found associations between perinatal  CO exposure and
decrements in birth weight,  as well as other developmental outcomes. The ISA concludes
these studies are suggestive of a causal relationship between long-term exposures to CO and
developmental effects and birth outcomes.

       Epidemiologic studies provide evidence of effects on respiratory morbidity such as
changes in pulmonary function, respiratory symptoms, and hospital admissions associated
with ambient CO concentrations.  A limited number of epidemiologic studies considered co-
pollutants such as ozone,  SO2, and PM in two-pollutant models and found that CO risk
estimates were generally robust, although this limited evidence makes it difficult to
disentangle effects attributed to CO itself from those of the larger complex air pollution
mixture.  Controlled human exposure studies have not extensively evaluated the effect of CO
on respiratory morbidity.  Animal studies at levels of 50-100 ppm CO show preliminary
evidence of altered pulmonary vascular remodeling and oxidative injury. The ISA concludes
that the evidence is suggestive of a causal relationship between short-term CO exposure and
respiratory morbidity, and inadequate to conclude that a causal relationship exists  between
long-term exposure  and respiratory morbidity.

       Finally, the ISA concludes that the epidemiologic evidence is suggestive of a causal
relationship between short-term exposures to CO and mortality. Epidemiologic studies
provide evidence of an association between short-term exposure to CO and mortality, but
limited evidence is available to evaluate cause-specific mortality outcomes associated with
CO exposure.  In addition, the attenuation of CO risk estimates which was often observed in
co-pollutant models contributes to the uncertainty as to whether CO is acting alone or as an
indicator for other combustion-related pollutants. The ISA also concludes that there is not
likely to be a causal relationship between relevant long-term exposures to CO and mortality.

       6.1.1.9 Health Effects of Air Toxics

       Motor vehicle emissions contribute to ambient levels of air toxics known or suspected
as human or animal  carcinogens,  or that have noncancer health effects. The population
experiences an elevated risk of cancer and other noncancer health effects from exposure to air
toxics.87  These compounds include, but are not limited to, benzene, 1,3-butadiene,
formaldehyde, acetaldehyde, acrolein, polycyclic organic matter (POM), and naphthalene.
                                         6-8

-------
                                                 2017 Draft Regulatory Impact Analysis
These compounds were identified as national or regional risk drivers or contributors in the
2005 National-scale Air Toxics Assessment (NATA) and have significant inventory
contributions from mobile sources.  Although the 2005 NATA did not quantify cancer risks
associated with exposure to diesel exhaust, EPA has concluded that diesel exhaust ranks with
the other emissions that the 2005 NATA suggests pose the greatest relative risk. According to
NATA for 2005, mobile sources were responsible for 43 percent of outdoor toxic emissions
and over 50 percent of the cancer risk and noncancer hazard attributable to direct emissions
from mobile and stationary sources.1111

       Noncancer health effects can result from chronic,JJJJ subchronic,KKKK or acuteLLLL
inhalation exposures to  air toxics, and include neurological, cardiovascular, liver, kidney, and
respiratory effects as well as effects on the immune and reproductive systems. According to
the 2005 NATA, about three-fourths of the U.S. population was exposed to an average
chronic concentration of air toxics that has the potential for adverse noncancer respiratory
health effects. This will continue to be the case in 2030, even though toxics concentrations
will be lower.88

       The NATA modeling framework has a number of limitations which prevent its use as
the sole basis for setting regulatory  standards. These limitations and uncertainties are
discussed on the 2005 NATA website.89  Even  so, this modeling framework is very useful in
identifying air toxic pollutants and sources of greatest concern, setting regulatory priorities,
and informing the decision making process.

       6.1.1.9.1      Benzene

       The EPA's IRIS database lists benzene  as a known human carcinogen (causing
leukemia) by all routes of exposure, and concludes that exposure is associated with additional
health effects, including genetic changes in both humans and animals and increased
proliferation of bone marrow cells in mice.90'91'92 EPA states in its IRIS database that data
indicate a causal relationship between benzene  exposure and acute lymphocytic leukemia and
suggest a relationship between benzene exposure and chronic non-lymphocytic leukemia and
chronic lymphocytic leukemia. The International Agency for Research on  Carcinogens
(IARC) has determined that benzene is a human carcinogen and the U.S. Department of
Health and Human Services (DHHS) has characterized benzene as a known human
           93 94
carcinogen. '
mi NATA also includes estimates of risk attributable to background concentrations, which includes contributions
from long-range transport, persistent air toxics, and natural sources; as well as secondary concentrations, where
toxics are formed via secondary formation.  Mobile sources substantially contribute to long-range transport and
secondarily formed air toxics.
1111 Chronic exposure is defined in the glossary of the Integrated Risk Information (IRIS) database
(http://www.epa.gov/iris) as repeated exposure by the oral, dermal, or inhalation route for more than
approximately 10% of the life span in humans (more than approximately 90 days to 2 years in typically used
laboratory animal species).
KKKK Defined in the IRIS database as exposure to a substance spanning approximately 10% of the lifetime of an
organism.
LLLL Defined in the IRIS database as exposure by the oral, dermal, or inhalation route for 24 hours or less.
                                           6-9

-------
Chapter 6

       A number of adverse noncancer health effects including blood disorders, such as
preleukemia and aplastic anemia, have also been associated with long-term exposure to
benzene.95'96  The most sensitive noncancer effect observed in humans, based on current data,
is the depression of the absolute lymphocyte count in blood.97'98 In addition, published work,
including studies sponsored by the Health Effects Institute (HEI), provides evidence that
biochemical responses are occurring at lower levels of benzene exposure than previously
known.99'100'101'102 EPA's IRIS program has not yet evaluated these new data.

       6.1.1.9.2      1,3-Butadiene

       EPA has characterized 1,3-butadiene as carcinogenic to humans by inhalation.103'104
The IARC has determined that 1,3-butadiene is a human carcinogen and the U.S. DHHS has
characterized 1,3-butadiene as a known human carcinogen.105'106'107 There are numerous
studies consistently demonstrating that 1,3-butadiene is metabolized into genotoxic
metabolites by experimental animals and humans.  The specific mechanisms of 1,3-butadiene-
induced carcinogenesis are unknown; however, the scientific evidence strongly suggests that
the carcinogenic effects are mediated by genotoxic metabolites. Animal data suggest that
females may be more sensitive than males  for cancer effects associated with 1,3-butadiene
exposure; there are insufficient data in humans from  which to draw conclusions about
sensitive subpopulations.  1,3-butadiene also causes a variety of reproductive and
developmental effects in mice; no human data on these effects are available. The most
sensitive effect was ovarian atrophy observed in a lifetime bioassay of female mice.108

       6.1.1.9.3     Formaldehyde

       Since 1987, EPA has classified formaldehyde as a probable human carcinogen based
on evidence in humans and in rats, mice, hamsters, and monkeys.109 Substantial additional
research since that time informs current scientific understanding of the health effects
associated with exposure to formaldehyde. These include recently published research
conducted by the National Cancer Institute (NCI) which found an increased risk of
nasopharyngeal cancer and lymphohematopoietic malignancies such as leukemia among
workers exposed to formaldehyde.110'111 In an analysis of the lymphohematopoietic cancer
mortality from an extended follow-up of these workers, NCI confirmed an association
between lymphohematopoietic cancer risk  and peak formaldehyde exposures.112 A recent
NIOSH study of garment workers also found increased risk of death due to leukemia among
workers exposed to formaldehyde.113 Extended follow-up of a cohort of British chemical
workers did not find evidence of an  increase in nasopharyngeal or lymphohematopoietic
cancers, but a continuing statistically significant excess in lung cancers was reported.114

       In the past 15 years there has been substantial research on the inhalation dosimetry for
formaldehyde in rodents and primates by the Chemical Industry Institute of Toxicology (CUT,
now renamed the Hamner Institutes  for Health Sciences), with a focus on use of rodent data
for refinement of the quantitative cancer dose-response assessment.115'116'117 CIIT's risk
assessment of formaldehyde incorporated mechanistic and dosimetric information on
formaldehyde. These data were modeled using a biologically-motivated two-stage clonal
growth model for cancer and  also a point of departure based on a Benchmark Dose approach.
However, it should be noted that recent research published by EPA indicates that when two-
                                        6-10

-------
                                               2017 Draft Regulatory Impact Analysis
stage modeling assumptions are varied, resulting dose-response estimates can vary by several
orders of magnitude.118'119'120'121 These findings are not supportive of interpreting the CUT
model results as providing a conservative (health protective) estimate of human risk.122 EPA
research also examined the contribution of the two-stage modeling for formaldehyde towards
characterizing the relative weights of key events in the mode-of-action of a carcinogen. For
example, the model-based inference in the published CUT study that formaldehyde's direct
mutagenic action is not relevant to the compound's tumorigenicity was found not to hold
under variations of modeling assumptions.123

       Based on the developments of the last decade, in 2004, the working group of the IARC
concluded that formaldehyde is carcinogenic to humans (Group 1), on the basis of sufficient
evidence in humans and sufficient evidence in experimental animals - a higher classification
than previous IARC evaluations.  After reviewing  the currently available epidemiological
evidence, the IARC (2006) characterized the human evidence for formaldehyde
carcinogenicity as "sufficient," based upon the data on nasopharyngeal cancers; the
epidemiologic evidence on leukemia was characterized as "strong."124

       Formaldehyde exposure also causes a range of noncancer health effects, including
irritation of the eyes (burning and watering of the eyes), nose and throat. Effects from
repeated exposure in humans include respiratory tract irritation, chronic bronchitis and nasal
epithelial lesions such as metaplasia and loss of cilia.  Animal studies suggest that
formaldehyde may also cause airway inflammation - including eosinophil infiltration into the
airways. There are several studies that suggest that formaldehyde may increase the risk of
asthma - particularly in the young.125'126

       The above-mentioned rodent and human studies, as well as mechanistic information
and their analyses, were evaluated in EPA's recent Draft Toxicological Review of
Formaldehyde - Inhalation Assessment through the Integrated Risk Information System
(IRIS) program. This draft IRIS assessment was released in June 2010 for public review and
comment and external peer review by the National Research Council (NRC). The NRC
released their review report in April 2011
(http://www.nap.edu/catalog.php?record_id=l3142).  The EPA is currently revising the draft
assessment in response to this review.

       6.1.1.9.4      Acetaldehyde

       Acetaldehyde is classified in EPA's IRIS database as a probable human carcinogen,
based on nasal tumors in rats, and is considered toxic by the inhalation, oral, and intravenous
      1 T7
routes.    Acetaldehyde is reasonably anticipated to be a human carcinogen by the U.S.
DHHS in the 11th Report on Carcinogens and is classified as possibly carcinogenic to humans
(Group 2B) by the IARC.128'129 EPA is currently conducting a reassessment of cancer risk
from inhalation exposure to acetaldehyde.

       The primary noncancer effects of exposure to acetaldehyde vapors include irritation of
the eyes,  skin, and respiratory tract.130 In short-term (4 week) rat studies, degeneration of
olfactory epithelium was observed at various concentration levels of acetaldehyde
exposure.131'132 Data from these studies were used by EPA to develop an inhalation reference
                                        6-11

-------
Chapter 6

concentration. Some asthmatics have been shown to be a sensitive subpopulation to
decrements in functional expiratory volume (FEV1 test) and bronchoconstriction upon
acetaldehyde inhalation.133  The agency is currently conducting a reassessment of the health
hazards from inhalation exposure to acetaldehyde.

       6.1.1.9.5      Acrolein

       Acrolein is extremely acrid and irritating to humans when inhaled, with acute
exposure resulting in upper respiratory tract irritation, mucus hypersecretion and congestion.
The intense irritancy of this carbonyl has been demonstrated during controlled tests in human
subjects, who suffer intolerable eye and nasal mucosal sensory reactions within minutes of
exposure.134  These data and additional studies regarding acute effects of human exposure to
acrolein are summarized in EPA's 2003 IRIS Human Health Assessment for acrolein.135
Evidence available from studies in humans indicate that levels as low as 0.09 ppm (0.21
mg/m3) for five minutes may elicit subjective complaints of eye irritation with increasing
concentrations leading to more extensive eye, nose and respiratory symptoms.136  Lesions to
the lungs and upper respiratory tract of rats, rabbits, and hamsters have been observed after
subchronic exposure to acrolein.137 Acute exposure effects in animal studies report bronchial
hyperresponsiveness.138 In one study, the acute respiratory irritant effects of exposure to 1.1
ppm acrolein were more pronounced in mice with allergic  airway disease by comparison to
non-diseased mice which also showed decreases in respiratory rate.139 Based on these animal
data and demonstration of similar effects in humans (e.g., reduction in respiratory rate),
individuals with compromised respiratory function (e.g., emphysema, asthma) are expected to
be at increased risk of developing adverse responses to strong respiratory irritants such as
acrolein.

       EPA determined in 2003 that the human carcinogenic potential of acrolein could not
be determined because the available data were inadequate.  No information was available on
the carcinogenic effects of acrolein in humans and the animal  data provided inadequate
evidence of carcinogenicity.140 The IARC determined in 1995 that acrolein was not
classifiable as to its carcinogenicity in humans.141

       6.1.1.9.6      Polycyclic Organic Matter (POM)

       The term polycyclic organic matter (POM) defines a broad class of compounds that
includes the polycyclic aromatic hydrocarbon compounds  (PAHs). One of these compounds,
naphthalene, is discussed separately below. POM compounds are formed primarily from
combustion and are present in the atmosphere in gas and particulate form. Cancer is the
major concern from exposure to POM. Epidemiologic studies have reported an increase in
lung cancer in humans exposed to  diesel exhaust, coke oven emissions, roofing tar emissions,
and cigarette  smoke; all of these mixtures contain POM compounds.MMMM142 Animal studies
have reported respiratory tract tumors from inhalation exposure to benzo[a]pyrene and
MMMM Agency for Toxic Substances and Disease Registry (ATSDR). 1995. Toxicological profile for Polycyclic
Aromatic Hydrocarbons (PAHs) Atlanta, GA: U.S. Department of Health and Human Services, Public Health
Service. Available electronically athttp://www.atsdr.cdc.gov/ToxProfiles/TP.asp?id=122&tid=25.


                                         6-12

-------
                                                2017 Draft Regulatory Impact Analysis
alimentary tract and liver tumors from oral exposure to benzo[a]pyrene. In 1997 EPA
classified seven PAHs (benzo[a]pyrene, benz[a]anthracene, chrysene, benzo[b]fluoranthene,
benzo[k]fluoranthene, dibenz[a,h]anthracene, and indeno[l,2,3-cd]pyrene) as Group B2,
probable human carcinogens.143 Since that time, studies have found that maternal exposures
to PAHs in a population of pregnant women were associated with several adverse birth
outcomes, including low birth weight and reduced length at birth, as well as impaired
cognitive development in preschool children (3 years of age).144'145 EPA has not yet evaluated
these studies.

       6.1.1.9.7     Naphthalene

       Naphthalene is found in small quantities in gasoline and diesel fuels. Naphthalene
emissions have been measured in larger quantities in both gasoline and diesel exhaust
compared with evaporative emissions from mobile sources, indicating it is primarily a product
of combustion. EPA released an external review draft of a reassessment of the inhalation
carcinogenicity of naphthalene based on a number of recent animal  carcinogenicity studies.146
The draft reassessment completed external peer review.147  Based on external peer review
comments received, additional analyses are being undertaken.  This external review draft does
not represent official agency opinion and was released solely for the purposes of external peer
review and public comment. The National Toxicology Program listed naphthalene as
"reasonably anticipated to be a human carcinogen" in 2004 on the basis of bioassays reporting
clear evidence of carcinogenicity in rats and some evidence of carcinogenicity in mice.148
California EPA has released a new risk assessment for naphthalene, and the IARC has
reevaluated naphthalene and re-classified it as Group 2B: possibly carcinogenic to humans.149
Naphthalene also causes a number of chronic non-cancer effects in animals, including
abnormal cell changes and growth in respiratory and nasal tissues.150

       6.1.1.9.8     Other Air Toxics

       In addition to the compounds described above, other compounds in gaseous
hydrocarbon and PM emissions from vehicles will be affected by this proposal.  Mobile
source air toxic compounds that would potentially be impacted include ethylbenzene,
propionaldehyde, toluene, and xylene.  Information regarding the health effects of these
compounds can be found in EPA's IRIS database.NNNN

       6.1.1.10      Exposure and Health Effects Associated with Traffic-Related Air
       Pollution

       Populations who live, work, or attend school near major roads experience elevated
exposure to a wide range of air pollutants, as well as higher risks for a number of adverse
health effects.  While the previous sections of this RIA have focused on the health effects
associated with individual criteria pollutants or air toxics, this section discusses the mixture of
different exposures near major roadways, rather than the effects of any single pollutant.  As
    U.S. EPA Integrated Risk Information System (IRIS) database is available at: www.epa.gov/iris


                                         6-13

-------
Chapter 6

such, this section emphasizes traffic-related air pollution, in general, as the relevant indicator
of exposure rather than any particular pollutant.

       Concentrations of many traffic-generated air pollutants are elevated for up to 300-500
meters downwind of roads with high traffic volumes.151 Numerous sources on roads
contribute to elevated roadside concentrations, including exhaust and evaporative emissions,
and resuspension of road dust and tire and brake wear. Concentrations of several criteria and
hazardous air pollutants are elevated near major roads. Furthermore, different semi-volatile
organic compounds and chemical components of particulate matter, including elemental
carbon, organic material, and trace metals,  have been  reported at higher concentrations near
major roads.

       Populations near major roads experience greater risk of certain adverse health effects.
The Health Effects Institute published a report on the  health effects of traffic-related air
pollution.152 It concluded that evidence is "sufficient  to infer the presence of a causal
association" between traffic exposure and exacerbation of childhood asthma symptoms. The
HEI report also concludes that the evidence is either "sufficient" or "suggestive but not
sufficient" for a causal association between traffic exposure and new childhood asthma cases.
A review of asthma studies by Salam et al.  (2008) reaches similar conclusions.153 The HEI
report also concludes that there is "suggestive" evidence for pulmonary function deficits
associated with traffic exposure, but concluded that there is "inadequate and insufficient"
evidence for causal associations with respiratory health care utilization, adult-onset asthma,
COPD symptoms, and allergy. A review by Holguin  (2008) notes that the effects of traffic on
asthma may be modified by nutrition status, medication use, and genetic factors.154

       The HEI report also concludes that  evidence is "suggestive" of a causal association
between traffic exposure and all-cause and  cardiovascular mortality.  There is also evidence
of an association between traffic-related air pollutants and cardiovascular effects such as
changes in heart rhythm, heart attack, and cardiovascular disease.  The HEI report
characterizes this evidence as "suggestive" of a causal association, and an independent
epidemiological literature review by Adar and Kaufman (2007) concludes that there is
"consistent evidence" linking traffic-related pollution  and adverse cardiovascular health
outcomes.155

       Some studies have reported associations between traffic exposure and other  health
effects, such as birth outcomes (e.g., low birth weight) and childhood cancer.  The HEI report
concludes that there is currently "inadequate and insufficient" evidence for a causal
association between these effects and traffic exposure. A review by Raaschou-Nielsen and
Reynolds (2006) concluded that evidence of an association between childhood cancer and
traffic-related air pollutants is weak, but noted the inability to draw firm conclusions based on
limited evidence.156

       There is a large population in the U.S. living in close proximity of major roads.
According to the Census Bureau's American Housing Survey for 2007, approximately 20
million residences in the U.S., 15.6% of all homes, are located within 300 feet (91 m) of a
highway with 4+ lanes, a railroad, or an airport.157 Therefore, at current population of
approximately 309 million, assuming that population  and housing are similarly distributed,
                                         6-14

-------
                                                2017 Draft Regulatory Impact Analysis
there are over 48 million people in the U.S. living near such sources.  The HEI report also
notes that in two North American cities, Los Angeles and Toronto, over 40% of each city's
population live within 500 meters of a highway or 100 meters of a major road.  It also notes
that about 33% of each city's population resides within 50 meters of major roads. Together,
the evidence suggests that a large U.S. population lives in areas with elevated traffic-related
air pollution.

       People living near roads are often socioeconomically disadvantaged.  According to the
2007 American Housing Survey, a renter-occupied property is over twice as likely as an
owner-occupied property to be located near a highway with 4+ lanes, railroad or airport.  In
the same survey, the median household income of rental housing occupants was less than half
that of owner-occupants ($28,921/$59,886). Numerous studies in  individual urban areas
report higher levels of traffic-related air pollutants in areas with high minority or poor
populations.158'159'160

       Students may also be exposed in situations where schools are located  near major
roads.  In a study of nine metropolitan areas across the U.S., Appatova et al. (2008) found that
on average greater than 33% of schools were  located within 400 m of an Interstate, US, or
state highway, while 12% were located within 100 m.161 The study also found that among the
metropolitan areas studied, schools in the Eastern U.S. were more  often sited near major
roadways than schools in the Western U.S.

       Demographic studies of students in schools near major roadways suggest that this
population is more likely than the general  student population to be of non-white race or
Hispanic ethnicity, and more often live in  low socioeconomic status locations.162'163'164 There
is  some inconsistency in the evidence, which  may be due to different local development
patterns and measures of traffic and geographic scale used in the studies.161

       6.1.2   Environmental Effects Associated with Exposure to Non-GHG Pollutants

       In this section we will  discuss the environmental effects associated with non-GHG
pollutants, specifically: particulate matter, ozone, NOx, SOx and air toxics.

       6.1.2.1 Visibility Degradation

       Airborne particles degrade visibility by scattering and absorbing light. Good visibility
increases the quality of life where individuals live and work, and where they engage in
recreational activities.

       EPA is pursuing a two-part strategy to address visibility impairment.  First, EPA
developed the regional haze program (64 FR  35714) which was put in place in July 1999  to
protect the visibility in Mandatory Class I Federal areas. There are 156 national parks, forests
and wilderness areas categorized as Mandatory Class I Federal areas (62 FR 38680-38681,
July 18, 1997).  These areas are defined in CAA section 162 as those national parks exceeding
6,000 acres, wilderness areas and memorial parks exceeding 5,000 acres, and all international
parks which were in existence on August 7, 1977. Second, EPA has concluded that PM2.s
causes adverse effects on visibility in other areas that are not protected by the Regional Haze
Rule, depending on PM2 5 concentrations and other factors that control their visibility impact


                                         6-15

-------
Chapter 6

effectiveness such as dry chemical composition and relative humidity (i.e., an indicator of the
water composition of the particles), and has set secondary PM2.5 standards to address these
areas. The existing annual primary and secondary PM2.5 standards have been remanded and
are being addressed in the currently ongoing PM NAAQS review.  Figure 6.1-1 shows the
location of the 156 Mandatory Class I Federal areas.
                              -- Rainbow Lake, W and BradweH Bay, FL are Class 1 Ar
                              where visibility is not an important air quality related
       Figure 6.1-1 Mandatory Class I Federal Areas in the U.S.

       6.1.2.1.1      Visibility Monitoring

       In conjunction with the U.S. National Park Service, the U.S. Forest Service, other
Federal land managers, and State organizations in the U.S., the U.S. EPA has supported
visibility monitoring in national parks and wilderness areas since 1988. The monitoring
network was originally established at 20 sites, but it has now been expanded to 1 10 sites that
represent all but one of the 156 Mandatory Federal Class I areas across the country (see
Figure 6.1-1).  This long-term visibility monitoring network is known as IMPROVE
(Interagency Monitoring of Protected Visual Environments).

       IMPROVE provides direct measurement of fine particles that contribute to visibility
impairment. The IMPROVE network employs aerosol measurements at all sites, and optical
and scene measurements at some of the sites.  Aerosol measurements are taken for PMio and
PM2.5mass, and for key constituents of PM2 5, such as sulfate, nitrate, organic and elemental
carbon, soil dust, and several other elements. Measurements for specific aerosol constituents
are used to calculate "reconstructed" aerosol light extinction by multiplying the mass for each
constituent by its empirically-derived scattering and/or absorption efficiency, with adjustment
for the relative humidity. Knowledge of the main constituents of a site's light extinction
"budget" is critical for source apportionment and control strategy development.  In addition to
this indirect method of assessing light extinction, there are optical measurements which
                                         6-16

-------
                                                2017 Draft Regulatory Impact Analysis
directly measure light extinction or its components.  Such measurements are made principally
with either a nephelometer to measure light scattering, some sites also include an
aethalometer for light absorption, or at a few sites using a transmissometer, which measures
total light extinction.  Scene characteristics are typically recorded using digital or video
photography and are used to determine the quality of visibility conditions (such as effects on
color and contrast) associated with specific levels of light extinction as measured under both
direct and aerosol-related methods. Directly measured light extinction is used under the
IMPROVE protocol to cross check that the aerosol-derived light extinction levels are
reasonable in establishing current visibility conditions. Aerosol-derived light extinction is
used to document spatial and temporal trends and to determine how changes in atmospheric
constituents would affect future visibility conditions.

       Annual average visibility conditions (reflecting light extinction due to both
anthropogenic and non-anthropogenic sources) vary regionally across the U.S. Visibility is
typically worse in the summer months and the rural East generally has higher levels of
impairment than remote sites in the West. Figures 9-9 through 9-11 in the PM ISA detail the
percent contributions to particulate light extinction for ammonium nitrate and sulfate, EC and
OC, and coarse mass and fine soil, by season.165

       6.1.2.2 Plant and Ecosystem Effects of Ozone

       There are a number of environmental or public welfare effects associated with the
presence of ozone in the ambient air.166  In this section we discuss the impact of ozone on
plants, including trees, agronomic crops and urban ornamentals.

       The Air Quality  Criteria Document for Ozone and related Photochemical Oxidants
notes that, "ozone affects vegetation throughout the United States, impairing crops, native
vegetation, and ecosystems more than any other air pollutant."167 Like carbon dioxide (CO2)
and other gaseous substances, ozone enters plant  tissues primarily through apertures (stomata)
in leaves in a process called "uptake."168 Once sufficient levels of ozone (a highly reactive
substance), or its reaction products, reaches the interior of plant cells, it can inhibit or damage
essential cellular components and functions, including enzyme activities, lipids, and cellular
membranes, disrupting the plant's osmotic (i.e., water) balance and energy utilization
patterns.169'170  If enough tissue becomes damaged from these effects,  a plant's capacity to fix
carbon to form carbohydrates, which are the primary form of energy used by plants, is
reduced,171 while plant respiration increases. With fewer resources available, the plant
reallocates existing resources away from root growth and storage, above ground growth or
yield, and reproductive processes, toward leaf repair and maintenance, leading to reduced
growth and/or reproduction.  Studies have shown that plants stressed in these ways may
exhibit a general loss of vigor, which can lead to  secondary impacts that modify plants'
responses to other environmental factors. Specifically, plants may become more sensitive to
other air pollutants, more susceptible to disease, insect attack, harsh weather (e.g., drought,
frost) and other environmental stresses. Furthermore, there is evidence that ozone can
interfere with the formation of mycorrhiza, essential symbiotic fungi associated with the roots
of most terrestrial plants, by reducing the amount of carbon available for transfer from the
host to the symbiont.172'173
                                         6-17

-------
Chapter 6

       This ozone damage may or may not be accompanied by visible injury on leaves, and
likewise, visible foliar injury may or may not be a symptom of the other types of plant
damage described above. When visible injury is present, it is commonly manifested as
chlorotic or necrotic spots, and/or increased leaf senescence (accelerated leaf aging).  Because
ozone damage can consist of visible injury to leaves, it can also reduce the aesthetic value of
ornamental vegetation and trees in urban landscapes, and negatively affects scenic vistas in
protected natural areas.

       Ozone can produce both acute and chronic injury in sensitive species depending on the
concentration level and the duration of the exposure. Ozone effects also tend to accumulate
over the growing season of the plant, so that even lower concentrations experienced for a
longer duration have the potential to create chronic stress on sensitive vegetation.  Not all
plants, however, are equally sensitive to ozone. Much of the variation in sensitivity between
individual plants or whole species is related to the plant's ability to regulate the extent of gas
exchange via leaf stomata (e.g., avoidance of ozone uptake through closure of
stomata)174'175'176 Other resistance mechanisms may involve the intercellular production of
detoxifying substances.  Several biochemical substances capable of detoxifying ozone have
been reported to occur in plants, including the antioxidants ascorbate and glutathione.  After
injuries have occurred, plants may be capable of repairing the damage to a limited extent.177

       Because of the differing sensitivities  among plants to ozone, ozone pollution can also
exert a selective pressure that leads to changes in plant community composition.  Given the
range of plant sensitivities and the fact that numerous other environmental factors modify
plant uptake and response to ozone,  it is not possible to identify threshold values above which
ozone is consistently toxic for all plants.  The next few paragraphs present additional
information on ozone  damage to trees, ecosystems, agronomic crops and urban ornamentals.

       Assessing the impact of ground-level ozone on forests in the United States involves
understanding the risks to sensitive tree species from ambient ozone concentrations and
accounting for the prevalence of those species within the forest.  As a way to quantify the
risks to particular plants from ground-level ozone, scientists have developed ozone-
exposure/tree-response functions by exposing tree seedlings to different ozone levels and
measuring reductions  in growth as "biomass loss."  Typically, seedlings are used because they
are easy to manipulate and measure  their growth loss from ozone pollution.  The mechanisms
of susceptibility to ozone within the leaves of seedlings and mature trees are identical, though
the magnitude of the effect may be higher or lower depending on the tree species.178

       Some of the common tree  species in the United States that are sensitive to ozone are
black cherry (Prunus serotina), tulip-poplar (Liriodendron tulipifera), and eastern white pine
(Pinus strobus). Ozone-exposure/tree-response functions have been developed for each of
these tree species, as well as for aspen (Populus tremuliodes), and ponderosa pine (Pinus
ponderosd). Other common tree species,  such as oak (Quercus sppj and hickory (Carya
spp.), are not nearly as sensitive to ozone.  Consequently, with knowledge of the distribution
of sensitive species and the level of ozone at particular locations, it is possible to estimate a
"biomass loss" for each species across their range.
                                         6-18

-------
                                                2017 Draft Regulatory Impact Analysis
       Ozone also has been conclusively shown to cause discernible injury to forest
trees.179'180 In terms of forest productivity and ecosystem diversity, ozone may be the
pollutant with the greatest potential for regional-scale forest impacts.  Studies have
demonstrated repeatedly that ozone concentrations commonly observed in polluted areas can
have substantial impacts on plant function.181'182

       Because plants are at the base of the food web in many ecosystems, changes to the
plant community can affect associated organisms and ecosystems (including the suitability of
habitats that support threatened or endangered species and below ground organisms living in
the root zone). Ozone impacts at the community and ecosystem level vary widely depending
upon numerous factors, including concentration and temporal variation of tropospheric ozone,
species composition, soil properties and climatic factors.183  In most instances, responses to
chronic or recurrent exposure in forested ecosystems are subtle and not observable for many
years.  These injuries can cause stand-level forest decline in sensitive ecosystems.184'185'186 It
is not yet possible to predict ecosystem responses to ozone with much certainty; however,
considerable knowledge of potential ecosystem responses has been acquired through  long-
term observations in highly damaged forests in the United States.

       Air pollution can have noteworthy cumulative impacts on forested ecosystems by
affecting regeneration, productivity, and species composition.187 In the U.S., ozone in the
lower atmosphere is one of the pollutants of primary concern. Ozone injury to forest plants
can be diagnosed by examination of plant leaves.  Foliar injury is usually the first visible sign
of injury to plants from ozone exposure and indicates impaired physiological processes in the
leaves.188 However, not all impaired plants will exhibit visible symptoms.

       Laboratory and field experiments have also shown reductions in yields for agronomic
crops exposed to ozone, including vegetables (e.g., lettuce) and field crops (e.g., cotton and
wheat). The most extensive field experiments, conducted under the National Crop Loss
Assessment Network (NCLAN) examined  15 species and numerous cultivars. The NCLAN
results show that "several economically important crop species are sensitive to ozone levels
typical of those found in the United States."189 In addition, economic studies have shown
reduced economic benefits as a result of predicted reductions in crop yields associated with
observed ozone levels.190'191'192

       Urban ornamentals represent an additional vegetation category likely to experience
some degree of negative effects associated  with exposure to ambient ozone levels.  It is
estimated that more than $20 billion (1990  dollars) are spent annually on landscaping using
ornamentals, both by private property owners/tenants and by governmental units responsible
for public areas.193 This is therefore a potentially costly  environmental effect. However, in
the absence of adequate exposure-response functions and economic damage functions for the
potential range of effects relevant to these types of vegetation, no direct quantitative analysis
has been conducted.

       6.1.2.2.1      Recent Ozone Visible Foliar Injury Data for the U.S.

       In the U.S. the national-level visible foliar injury indicator  is based on data from the
U.S. Department of Agriculture (USD A) Forest Service Forest Inventory and Analysis (FIA)
                                         6-19

-------
Chapter 6

program.  As part of its Phase 3 program, formerly known as Forest Health Monitoring, FIA
examines ozone injury to ozone-sensitive plant species at ground monitoring sites in forest
land across the country.  For this indicator, forest land does not include woodlots and urban
trees. Sites are selected using a systematic sampling grid, based on a global sampling
design.194'195 At each site that has at least  30 individual plants of at least three ozone-sensitive
species and enough open space to ensure that sensitive plants are not protected from ozone
exposure by the forest canopy, FIA looks for damage on the foliage of ozone-sensitive forest
plant species. Because ozone injury is cumulative over the course of the growing season,
examinations are conducted in July and August, when ozone injury is typically highest.
Monitoring of ozone injury to plants by the USDA Forest Service has expanded over time
from monitoring sites in 10 states in 1994  to nearly 1,000 monitoring sites in 41 states in
2002.

       There is considerable regional variation in ozone-related visible foliar injury to
sensitive plants in the U.S. The U.S. EPA has  developed an environmental indicator based on
data from the USDA FIA program which examines ozone injury to ozone-sensitive plant
species at ground monitoring sites in forest land across the country. The data underlying the
indicator in Figure 6.1-2 is based on averages of all observations collected in 2002, the latest
year for which data are  publicly available  at the time the study was conducted, and is broken
down by U.S. EPA Region. Ozone damage to  forest plants is classified using a subjective
five-category biosite index based on expert opinion, but designed to be equivalent from site to
site. Ranges of biosite values translate to no injury, low or moderate foliar injury (visible
foliar injury to highly sensitive or moderately sensitive plants, respectively), and high or
severe foliar injury, which would be expected to result in tree-level or ecosystem-level
responses, respectively.196'197

        The highest percentages of observed high and severe foliar injury, those which are
most likely to be associated with tree or ecosystem-level responses, are primarily found in the
Mid-Atlantic and Southeast regions. In EPA Region 3 (which comprises the States of
Pennsylvania, West Virginia, Virginia, Delaware, Maryland and Washington D.C.), 12% of
ozone-sensitive plants showed signs of high or severe  foliar damage, and in Regions 2 (States
of New York, New Jersey), and 4 (States of North Carolina, South Carolina, Kentucky,
Tennessee, Georgia, Florida, Alabama, and Mississippi) the values were 10% and 7%,
respectively. The sum of high and severe  ozone injury ranged from 2% to 4% in EPA Region
1 (the six New England States), Region 7 (States of Missouri, Iowa, Nebraska and Kansas),
and Region 9 (States of California, Nevada, Hawaii and Arizona).  The percentage of sites
showing some ozone damage was about 45% in each of these EPA Regions.
                                         6-20

-------
                                                    2017 Draft Regulatory Impact Analysis
                              Degree of inpry:
                                Nope


                              Pf Bsri of rnofiitori^io Ettas in inch ;ategnr,:
                      Rectors "
                      ;54

                      Reqitm 2
                      •;42 sites.
                      Region 3
                      {1l"l srtesj

                      Ragisn 4
                                    11?
              -3.7
                   111
   21.4
7.1
                  55.9
111
                                           75.3
                                      10.1
                      Region 5
                      •iiebstfls
                            6
                      75.6
             B.T
                                               94.9
                         icri 7
                      {63 sites)

                      Region 8
                      [72 sites/

                           9
                      •|80 sites)
                       35.7
          B.I
 TO.2
  .-1.5
                          100,0
                                      12.5
           as
 F-1.3
   •1.3
                      ;57 stesj
                                               100.D
          45 mcnitn ftn§ sites,
 'ocatad in 4^
:Totate may not add te 1 CC*% due to
 rounding

 Sjfa so-ttPw; USDA Forest Serviss.
 2006
                                                        EP*
                                                    -t< 5
        Figure 6.1-2 Ozone Injury to Forest Plants in U.S. by EPA Regions, 2002ab

        6.1.2.2.2      Indicator Limitations

        The categories for the biosite index are subjective and may not necessarily be directly
related to biomass loss or physiological  damage to plants in a particular area.  Ozone may
have other adverse impacts on plants (e.g., reduced productivity) that do not show signs of
visible foliar injury.198 The presence of diagnostic visible ozone injury on indicator plants
                                            6-21

-------
Chapter 6

does provide evidence that ozone is having an impact in an area. However, absence of ozone
injury in an area does not necessarily mean that there is no impact from ozone exposure.

       Field and laboratory studies were reviewed to identify the forest plant species in each
region that are sensitive to ozone air pollution and exhibit diagnostic injury.  Other forest
plant species, or even genetic variants of the same species, may not show symptoms at ozone
levels that cause effects on the selected ozone-sensitive species.

       Because species distributions vary regionally, different ozone-sensitive plant species
were examined in different parts of the country.  These target species could vary with respect
to ozone sensitivity, which might account for some of the apparent differences in ozone injury
among regions of the U.S. Ozone damage to foliage may be reduced under conditions of low
soil moisture, but most of the variability in the index (70%) was explained by ozone
             199
concentration.

       Though FIA has extensive spatial coverage based on a robust sample design, not all
forested areas in the U.S. are monitored for ozone injury.  Even though the biosite data have
been collected over multiple years, most biosites were not monitored over the entire period, so
these data cannot provide more than a baseline for future trends.

       6.1.2.3 Particulate Matter Deposition

       Particulate matter contributes to adverse  effects on vegetation and ecosystems,  and to
soiling and materials damage.  These welfare effects result predominately from exposure to
excess amounts of specific chemical species, regardless of their source or predominant form
(particle, gas or liquid). The following characterizations  of the nature of these  environmental
effects are based on information contained in the 2009 PM ISA and the 2005 PM Staff Paper
as well as the Integrated Science Assessment for Oxides of Nitrogen and Sulfur- Ecological
Criteria.200'201'202

       6.1.2.3.1      Deposition of Nitrogen and Sulfur

       Nitrogen and sulfur interactions in the environment are highly complex. Both nitrogen
and sulfur are essential, and sometimes limiting, nutrients needed for growth and productivity.
Excesses of nitrogen or sulfur can lead to acidification, nutrient enrichment, and
eutrophication of aquatic ecosystems.203

       The process of acidification affects both  freshwater aquatic and terrestrial ecosystems.
Acid deposition causes acidification of sensitive surface waters.  The effects of acid
deposition on aquatic systems depend largely upon the ability of the ecosystem to neutralize
the additional acid.  As acidity increases, aluminum leached from soils and sediments,  flows
into lakes and streams and can be toxic to both terrestrial and aquatic biota. The lower pH
concentrations and higher aluminum levels resulting from acidification make it difficult for
some fish and other aquatic organisms to survive, grow, and reproduce.  Research on effects
of acid deposition on forest ecosystems has come to focus increasingly on the biogeochemical
processes that affect uptake, retention, and cycling of nutrients within these ecosystems.
Decreases in available base cations from soils are at least partly attributable to acid
deposition. Base cation depletion is a cause for concern because of the role these ions play in


                                         6-22

-------
                                                2017 Draft Regulatory Impact Analysis
acid neutralization, and because calcium, magnesium and potassium are essential nutrients for
plant growth and physiology. Changes in the relative proportions of these nutrients,
especially in comparison with aluminum concentrations, have been associated with declining
forest health.

       At current ambient levels, risks to vegetation from short-term exposures to dry
deposited particulate nitrate or sulfate are low.  However, when found in acid or acidifying
deposition, such particles do have the potential to cause direct leaf injury.  Specifically, the
responses of forest trees to acid precipitation (rain, snow) include accelerated weathering of
leaf cuticular surfaces, increased permeability of leaf surfaces to toxic materials, water, and
disease agents; increased leaching of nutrients from foliage; and altered reproductive
processes—all which serve to weaken trees so that they are more susceptible to other stresses
(e.g., extreme weather, pests, pathogens). Acid deposition with levels of acidity associated
with the leaf effects described above  are currently found in some locations in the eastern
U.S.204 Even higher concentrations of acidity can be present in occult depositions (e.g., fog,
mist or clouds) which more frequently impacts higher elevations. Thus, the risk of leaf injury
occurring from acid deposition in some areas of the eastern U.S. is high. Nitrogen deposition
has also been shown to impact ecosystems in the western U.S. A study conducted in the
Columbia River Gorge National Scenic Area (CRGNSA), located along a portion of the
Oregon/Washington border, indicates that lichen communities in the CRGNSA have shifted
to a higher proportion of nitrophilous species and the nitrogen content of lichen tissue is
elevated.205  Lichens are sensitive indicators of nitrogen deposition effects to terrestrial
ecosystems and the lichen studies in the Columbia River Gorge clearly show that ecological
effects from air pollution are occurring.

       Some of the most significant detrimental effects associated with excess nitrogen
deposition are those associated with a condition known as nitrogen saturation.  Nitrogen
saturation is the condition in which nitrogen inputs from atmospheric deposition and other
sources exceed the biological requirements of the ecosystem. The effects associated with
nitrogen saturation include: (1) decreased productivity, increased mortality, and/or shifts in
plant community composition, often leading to decreased biodiversity in many natural
habitats wherever atmospheric reactive nitrogen deposition increases significantly above
background and critical thresholds are exceeded; (2) leaching of excess nitrate  and associated
base cations from soils into streams, lakes, and rivers, and mobilization of soil  aluminum; and
(3)  fluctuation of ecosystem processes such as nutrient and energy cycles through changes in
the  functioning and species composition of beneficial soil organisms.206

       In the U.S. numerous forests now show severe symptoms of nitrogen saturation.
These forests include: the northern hardwoods and mixed conifer forests in the Adirondack
and Catskill Mountains of New York; the red spruce forests at Whitetop Mountain,  Virginia,
and Great Smoky Mountains National Park, North Carolina; mixed hardwood watersheds at
Fernow Experimental Forest in West Virginia; American beech forests in Great Smoky
Mountains National Park, Tennessee;  mixed conifer forests and chaparral watersheds in
southern California and the southwestern Sierra Nevada in Central California; the alpine
tundra/subalpine conifer forests of the Colorado Front Range; and red alder forests in the
Cascade Mountains in Washington.
                                         6-23

-------
Chapter 6

       Excess nutrient inputs into aquatic ecosystems (i.e. streams, rivers, lakes, estuaries or
oceans) either from direct atmospheric deposition, surface runoff, or leaching from nitrogen
saturated soils into ground or surface waters can contribute to conditions of severe water
oxygen depletion; eutrophication and algae blooms; altered fish distributions, catches, and
physiological states; loss of biodiversity; habitat degradation; and increases in the incidence
of disease.

       Atmospheric deposition of nitrogen is a significant source of total nitrogen to many
estuaries in the United States. The amount of nitrogen entering estuaries that is ultimately
attributable to atmospheric deposition is not well-defined. On an annual basis, atmospheric
nitrogen deposition may contribute significantly to the total nitrogen load, depending on the
size and location of the watershed. In addition, episodic nitrogen inputs, which may be
ecologically important, may play a more important role than indicated by the annual average
concentrations. Estuaries in the U.S. that suffer from nitrogen enrichment often experience a
condition known as eutrophi cation.  Symptoms of eutrophi cation include changes in the
dominant species of phytoplankton, low levels of oxygen  in the water column,  fish and
shellfish kills, outbreaks of toxic algae, and other population changes which can cascade
throughout the food web.  In  addition, increased phytoplankton growth in the water column
and on surfaces can attenuate light causing declines in submerged aquatic vegetation, which
serves as an important habitat for many estuarine fish and shellfish species.

       Severe and persistent eutrophi cation often directly impacts human activities.  For
example, losses in the nation's fishery resources may be directly caused by fish kills
associated with low dissolved oxygen and toxic blooms. Declines in tourism occur when low
dissolved oxygen  causes noxious smells and floating mats of algal blooms create unfavorable
aesthetic conditions.  Risks to human health increase when the toxins from algal blooms
accumulate in edible fish and shellfish, and when toxins become airborne, causing respiratory
problems due to inhalation. According to a NOAA report, more than half of the nation's
estuaries have moderate to high expressions of at least one of these symptoms - an indication
that eutrophi cation is well developed in more  than half of U.S. estuaries.207

       6.1.2.3.2     Deposition of Heavy Metals

       Heavy metals, including cadmium, copper, lead, chromium, mercury, nickel and zinc,
have the greatest potential for impacting forest growth.208 Investigation of trace metals  near
roadways and industrial facilities indicate that a substantial load of heavy metals can
accumulate on vegetative  surfaces.  Copper, zinc, and nickel have been documented  to cause
direct toxicity to vegetation under field conditions. Little research has been conducted on the
effects associated  with mixtures of contaminants found in ambient PM.  While metals
typically exhibit low  solubility, limiting their  bioavailability and direct toxicity, chemical
transformations of metal compounds occur in the environment, particularly in the presence of
acidic or other oxidizing species.  These chemical changes influence the mobility and toxicity
of metals in  the environment. Once taken up  into plant tissue, a metal compound can undergo
chemical changes, exert toxic effects on the plant itself, accumulate and be passed along to
herbivores or can re-enter the soil and further cycle in the environment.  Although there has
been no direct evidence of a physiological association between tree injury and heavy metal
exposures, heavy metals have been implicated because of similarities between metal
                                         6-24

-------
                                                2017 Draft Regulatory Impact Analysis
deposition patterns and forest decline.  This hypothesized relationship/correlation was further
explored in high elevation forests in the northeastern U.S.  These studies measured levels of a
group of intracellular compounds found in plants that bind with metals and are produced by
plants as a response to sublethal concentrations of heavy metals.  These studies indicated  a
systematic and significant increase in concentrations of these compounds associated with the
extent of tree injury.  These data strongly imply that metal stress causes tree injury and
contributes to forest decline in the northeastern United States.209  Contamination of plant
leaves by heavy metals can lead to elevated soil levels.  Trace metals absorbed into the plant
frequently bind to the leaf tissue, and then are lost when the leaf drops.  As the fallen leaves
decompose, the heavy metals are transferred into the soil.210'211 Upon entering the soil
environment, PM pollutants can alter ecological processes of energy flow and nutrient
cycling, inhibit nutrient uptake, change ecosystem structure, and affect ecosystem
biodiversity.  Many of the most important effects occur in the soil.  The soil environment is
one of the most dynamic sites of biological interaction in nature. It is inhabited by microbial
communities of bacteria, fungi, and actinomycetes.  These organisms are essential participants
in the nutrient cycles that make elements available for plant uptake. Changes in the soil
environment that influence the role of the bacteria and fungi in nutrient cycling determine
plant and ultimately ecosystem response.212

       The environmental sources and cycling of mercury are currently of particular concern
due to the bioaccumulation and biomagnification of this metal in aquatic ecosystems and  the
potent toxic nature of mercury in the forms in which is it ingested by people and other
animals.  Mercury is unusual compared with other metals in that it largely partitions into the
gas phase (in elemental form), and therefore has a longer residence time in the atmosphere
than a metal found predominantly in the particle phase. This property enables mercury to
travel far from the primary source before being deposited and accumulating in the aquatic
ecosystem. The major source of mercury in the Great Lakes is from atmospheric deposition,
accounting for approximately eighty percent of the mercury in Lake Michigan.213'214 Over
fifty percent of the mercury in the Chesapeake Bay has been attributed to atmospheric
deposition.215 Overall, the National Science and Technology Council identifies atmospheric
deposition as the primary source of mercury to aquatic systems.216  Forty-four states have
issued health advisories for the consumption offish contaminated by mercury; however, most
of these advisories are issued in areas without a mercury point source.

       Elevated levels of zinc and lead have been identified in streambed sediments, and
these elevated levels have been correlated with population density and motor vehicle
use.217'218 Zinc and nickel have  also been identified in urban water and soils. In addition,
platinum, palladium, and rhodium, metals found in the catalysts of modern motor vehicles,
have been measured at elevated  levels along roadsides.219  Plant uptake of platinum has been
observed at these locations.

       6.1.2.3.3      Deposition of Polycyclic Organic Matter

       Polycyclic organic matter (POM) is a byproduct of incomplete combustion and
consists of organic compounds with more than one benzene ring and a boiling point greater
than or equal to 100 degrees centigrade.220  Polycyclic aromatic hydrocarbons (PAHs) are a
class of POM that contains compounds which are known or suspected carcinogens.
                                         6-25

-------
Chapter 6

       Major sources of PAHs include mobile sources. PAHs in the environment may be
present as a gas or adsorbed onto airborne paniculate matter. Since the majority of PAHs are
adsorbed onto particles less than 1.0 jim in diameter, long range transport is possible.
However, studies have shown that PAH compounds adsorbed onto diesel exhaust particulate
and exposed to ozone have half lives of 0.5 to 1.0 hours.221

       Since PAHs are insoluble, the compounds generally are particle reactive and
accumulate in sediments.  Atmospheric deposition of particles is believed to be the major
source of PAHs to the sediments of Lake Michigan.222'223 Analyses of PAH deposition in
Chesapeake and Galveston Bay indicate that dry deposition and gas exchange from the
atmosphere to the surface water predominate.224'225 Sediment concentrations of PAHs are
high enough in some segments of Tampa Bay to pose an environmental health threat. EPA
funded a study to better characterize the sources and loading rates for PAHs into Tampa
Bay.226 PAHs that enter a water body through gas exchange likely partition into organic  rich
particles and can be biologically recycled, while dry deposition of aerosols containing PAHs
tend to be more resistant to biological recycling.227 Thus, dry deposition is likely the main
pathway for PAH concentrations in sediments while gas/water exchange at the surface may
lead to PAH distribution into the food web, leading to increased health risk concerns.

       Trends in PAH deposition levels are difficult to discern because of highly variable
ambient air concentrations, lack of consistency in monitoring methods, and the significant
influence of local sources on deposition levels.228 Van Metre et al. noted PAH concentrations
in urban reservoir sediments have increased by 200-300% over the last forty years and
correlate with increases in automobile use.229

       Cousins et al. estimate that more than ninety percent of semi-volatile organic
compound (SVOC) emissions in the United Kingdom deposit on soil.230 An analysis of PAH
concentrations near a Czechoslovakian roadway indicated that concentrations were thirty
times greater than background.231

       6.1.2.3.4     Materials Damage and Soiling

       The effects of the deposition of atmospheric pollution, including ambient PM, on
materials are related to both physical damage and impaired aesthetic qualities. The deposition
of PM (especially sulfates and nitrates) can physically  affect materials, adding to the effects of
natural weathering processes,  by potentially promoting or accelerating the corrosion of
metals, by degrading paints, and by deteriorating building materials such as concrete and
limestone. Only chemically active fine particles or hygroscopic coarse particles contribute to
these physical effects.  In  addition, the deposition of ambient PM can reduce the aesthetic
appeal of buildings and culturally important articles through soiling. Particles consisting
primarily of carbonaceous compounds cause soiling of commonly used building materials and
culturally important items such as statues and works of art.

       6.1.2.4 Environmental Effects of Air Toxics

       Emissions from producing, transporting and combusting fuel contribute to ambient
levels of pollutants that contribute to adverse effects on vegetation.  Volatile organic
                                         6-26

-------
                                               2017 Draft Regulatory Impact Analysis
compounds (VOCs), some of which are considered air toxics, have long been suspected to
play a role in vegetation damage.232 In laboratory experiments, a wide range of tolerance to
VOCs has been observed.233  Decreases in harvested seed pod weight have been reported for
the more sensitive plants,  and some studies have reported effects on seed germination,
flowering and fruit ripening.  Effects of individual VOCs or their role in conjunction with
other stressors (e.g., acidification, drought, temperature extremes) have not been well studied.
In a recent study of a mixture of VOCs including ethanol and toluene on herbaceous plants,
significant effects on seed production, leaf water content and photosynthetic efficiency were
reported for some plant species.234

       Research suggests an adverse impact of vehicle exhaust on plants, which has in some
cases been attributed to aromatic compounds and in other cases to nitrogen oxides.235'236'237
The impacts of VOCs on plant reproduction  may have long-term implications for biodiversity
and survival of native species near major roadways. Most of the studies of the impacts of
VOCs on vegetation have focused on short-term exposure and few studies have focused on
long-term effects of VOCs on vegetation and the potential for metabolites of these compounds
to affect herbivores or insects.

6.2 Air Quality Impacts of Non-GHG Pollutants

       Chapter 4 of this DRIA presents the projected emissions changes due to the proposed
rule. Once the emissions  changes are projected the next step is  to look at how the ambient air
quality would be impacted by those emissions changes.  Although the purpose of this proposal
is to address greenhouse gas emissions, this proposed rule would also impact emissions of
criteria and air toxics.  Section 6.2.1 describes current ambient levels of PM,  ozone and some
air toxics without the standards being proposed in this rule. No air quality modeling was  done
for this DRIA to project the impacts of the proposed rule.  EPA plans to conduct such
modeling, however, and those plans are discussed in Section 6.2.2.

       6.2.1   Current Levels of Non-GHG Pollutants

       6.2.1.1  Particulate Matter

       As described in  Section 6.1.1.2, exposure to PM25 causes adverse health effects, and
the U.S. government has set national standards to provide requisite protection against those
health effects.  There are two U.S. national ambient air quality standards (NAAQS) for PM2 5:
an annual standard (15 ug/m3) and a 24-hour standard (35 ug/m3). The most recent revisions
to these standards were in 1997 and 2006.  In 2005 the U.S. EPA designated nonattainment
areas for the 1997 PM2.5 NAAQS (70 FR 19844, April 14, 2005)0000 As of April 21, 2011,
approximately 88 million people live in the 39  areas that are designated as nonattainment for
the 1997  PM2s NAAQS.  These PM2 5 nonattainment areas are comprised of 208 full or
partial counties.  On October 8, 2009, the EPA issued final nonattainment area designations
for the 2006 24-hour PM2.5 NAAQS (74 FR  58688, November 13, 2009).  These designations
0000 A nonattainment area is defined in the Clean Air Act (CAA) as an area that is violating an ambient standard
or is contributing to a nearby area that is violating the standard.


                                        6-27

-------
Chapter 6

include 32 areas composed of 121 full or partial counties with a population of over 70 million.
In total, there are 54 PM2 5 nonattainment areas composed of 245 counties with a population
of 101 million people.

       States with PM2 5 nonattainment areas will be required to take action to bring those
areas into compliance in the future.  Most 1997 PM2.5 nonattainment areas are required to
attain the  1997 PM2.5 NAAQS in the 2010 to 2015 time frame and then be required to
maintain the 1997 PM2.5 NAAQS thereafter.238 The 2006 24-hour PM2.5 nonattainment areas
will be required to attain the 2006 24-hour PM2.5 NAAQS in the 2014 to 2019 time frame and
then be required to maintain the 2006 24-hour PM2.5 NAAQS thereafter.239 The vehicle
standards  proposed here first apply to model year 2017 vehicles.

       6.2.1.2 Ozone

       As described in Section 6.1.1.4, ozone causes adverse health effects, and the U.S.
government has set national standards to protect against those health effects.  The primary
NAAQS for 8-hour ozone was set at 0.075 ppm in 2008. The previous 8-hour ozone standard,
set in 1997, had been 0.08 ppm. In 2004 the U.S. EPA designated nonattainment areas for the
1997 8-hour ozone NAAQS (69 FR 23858,  April 30, 2004).pppp As of August 30, 2011 there
are 44 1997 8-hour ozone nonattainment areas comprised of 242 full or partial counties with a
total population of over 118 million.240 Nonattainment designations for the 2008 8-hour
ozone standard are currently under development.

       States with ozone nonattainment areas are required to take action to bring those areas
into compliance in the future. The attainment date assigned to an ozone nonattainment area is
based on the area's classification.  Most ozone nonattainment areas are required to attain the
1997 8-hour ozone NAAQS in the 2007 to 2013 time frame and then to maintain it
thereafter.QQQQ The attainment dates associated with the potential nonattainment areas based
on the 2008 8-hour ozone NAAQS will likely be in the 2015 to 2035 timeframe, depending on
the severity of the problem in each area. In addition, EPA is working to complete the current
review of the ozone NAAQS by mid-2014.  The attainment dates associated with the potential
nonattainment areas based on the 2014 8-hour ozone NAAQS will likely be in the 2019 to
2036 timeframe, depending on the severity of the problem in each area.  As mentioned above,
the vehicle standards proposed here first apply to model year 2017 vehicles.
pppp A nonattainment area is defined in the Clean Air Act (CAA) as an area that is violating an ambient standard
or is contributing to a nearby area that is violating the standard.
QQQQ -pjjg LOS Angeles South Coast Air Basin 8-hour ozone nonattainment area is designated as severe and will
have to attain before June 15, 2021. The South Coast Air Basin has requested to be reclassified as an extreme
nonattainment area which will make its attainment date June 15, 2024. The San Joaquin Valley Air Basin 8-hour
ozone nonattainment area is designated as serious and will have to attain before June 15, 2013. The San Joaquin
Valley Air Basin has requested to be reclassified as an extreme nonattainment area which will make its
attainment date June 15, 2024.
                                          6-28

-------
                                                 2017 Draft Regulatory Impact Analysis
       6.2.1.3 Air Toxics

       According to the National Air Toxics Assessment (NATA) for 2005, mobile sources
were responsible for 43 percent of outdoor toxic emissions and over 50 percent of the cancer
risk and noncancer hazard attributable to direct emissions from mobile and stationary
sources. RRRR'SSSS'241  According to the 2005 NATA, about three-fourths of the U.S. population
was exposed to an average chronic concentration of air toxics that has the potential for
adverse noncancer respiratory health effects. In 2007 EPA finalized vehicle and fuel controls
to reduce mobile source air toxics.242  In addition, over the years, EPA has implemented a
number of mobile source and fuel controls resulting in VOC reductions, which also reduce air
toxic emissions. Modeling from the Mobile Source Air Toxics (MSAT) rule suggests that the
mobile source contribution to ambient benzene concentrations is projected to decrease over
40% by 2015, with a decrease in ambient benzene concentration from all sources of about
25%.  Although benzene is used as an example, the downward trend is projected for other air
toxics as well.  See the RIA for the final MSAT rule for more information  on ambient air
                 243
toxics projections.

       6.2.2  Impacts on Future Air Quality

       Air quality models use mathematical and numerical techniques to simulate the
physical and chemical processes that affect air pollutants as they disperse and react in the
atmosphere. Based on inputs of meteorological data and source information, these models are
designed to characterize primary pollutants that are emitted directly into the atmosphere and
secondary pollutants that are formed as a result of complex chemical reactions within the
atmosphere. Photochemical air quality models have become widely recognized and routinely
utilized tools for regulatory analysis by assessing the effectiveness of control strategies.
These models are applied at multiple spatial scales from local, regional, national, and global.
Section 6.2.2.1 provides more detail on the photochemical model, the Community Multi-scale
Air Quality (CMAQ) model, which will be utilized for the final rule analysis.

       6.2.2.1 Community Multi-Scale Air Quality (CMAQ) Modeling Plans

       Full-scale photochemical air quality modeling is necessary to accurately project levels
of PM2.5, ozone and air toxics. For the final rule, a national-scale air quality modeling
analysis will be performed to analyze  the impacts of the vehicle standards on PM2.s, ozone,
and selected air toxics (i.e., benzene, formaldehyde, acetaldehyde, acrolein and 1,3-
butadiene).  The length of time needed to prepare the necessary emissions inventories, in
addition to the processing time associated with the modeling itself, has precluded us from
performing air quality modeling for this proposal.
11111111 NATA also includes estimates of risk attributable to background concentrations, which includes
contributions from long-range transport, persistent air toxics, and natural sources; as well as secondary
concentrations, where toxics are formed via secondary formation.  Mobile sources substantially contribute to
long-range transport and secondarily formed air toxics.
ssss NATA relies on a Guassian plume model, Assessment System for Population Exposure Nationwide
(ASPEN), to estimate toxic air pollutant concentrations. Projected air toxics concentrations presented in this final
action were modeled with CMAQ 4.7.1.


                                          6-29

-------
Chapter 6

       Section III.G of the preamble presents projections of the changes in criteria pollutant
and air toxics emissions due to the proposed vehicle standards; the basis for those estimates is
set out in Chapter 4 of the DRIA.  The atmospheric chemistry related to ambient
concentrations of PM2 5, ozone and air toxics is very complex, and making predictions based
solely on emissions changes is extremely difficult. However, based on the magnitude of the
emissions changes predicted to result from the proposed vehicle standards, we expect that
there will be an improvement in ambient air quality, pending a more comprehensive analysis
for the final rule.

       For the final rule, EPA intends to use a Community Multi-scale Air Quality (CMAQ)
modeling platform as the tool for the air quality modeling. The CMAQ modeling system is a
comprehensive three-dimensional grid-based Eulerian air quality model designed to estimate
the formation and fate of oxidant precursors, primary and secondary PM concentrations and
deposition, and air toxics, over regional and urban spatial scales (e.g., over the contiguous
U.S.).244'245'246 The CMAQ model is a well-known and well-established tool and is commonly
used by EPA for regulatory analyses, for instance the 2012-2016 final rule, and by States in
developing attainment demonstrations for their State Implementation Plans.247 The CMAQ
model (version 4.7) was most recently peer-reviewed in February of 2009 for the U.S.
EPA.248 The CMAQ model also has been used in numerous national and international
applications.249'250'251

       CMAQ includes many science modules that simulate the emission, production, decay,
deposition and transport of organic and inorganic gas-phase and particle-phase pollutants in
the atmosphere. EPA intends to use the most recent version of CMAQ, which reflects updates
                                            0 SO
to version 4.7 to improve the underlying science.   These include aqueous chemistry mass
conservation improvements, improved vertical convective mixing and lowered CB05
mechanism unit yields for acrolein from 1,3-butadiene tracer reactions which were updated to
be consistent with laboratory measurements.

       The CMAQ modeling domain encompasses all of the lower 48 States and portions of
Canada and Mexico.  The modeling domain is made up of a large continental U.S. boundary
within which air quality is modeled at the 36 kilometer (km) grid cell level and two 12 km
boundaries (an Eastern US and a Western US domain) within which air quality is modeled at
the 12 km grid cell level. These are shown in Figure 6.2-1. The modeling domain contains 14
vertical layers with the top of the modeling domain at about 16,200 meters, or 100 millibars
(mb).
                                        6-30

-------
                                               2017 Draft Regulatory Impact Analysis
        Figure 6.2-1 CMAQ 12-km Eastern and Western US modeling domains

       The key inputs to the CMAQ model include emissions from anthropogenic and
biogenic sources, meteorological data, and initial and boundary conditions.  The CMAQ
meteorological input files are derived from annual simulations of the Pennsylvania State
University / National Center for Atmospheric Research Mesoscale Model.253 This model,
commonly referred to as MM5, is a limited-area, nonhydrostatic, terrain-following system
that solves for the full set of physical and thermodynamic equations which govern
atmospheric motions.254 The meteorology for the national 36 km grid and the 12 km Eastern
and Western U.S.  grids are developed by EPA and will be described in more detail within the
final RIA and the technical support document for the final rule air quality modeling.1111  The
meteorological outputs from MM5 are processed to create model-ready inputs for CMAQ
using the Meteorology-Chemistry Interface Processor (MCIP) version 3.4, for example:
horizontal wind components (i.e., speed and direction), temperature, moisture, vertical
diffusion rates, and rainfall rates for each grid cell in each vertical layer.255

       The lateral boundary and initial species concentrations are provided by a three-
dimensional global atmospheric chemistry model, the GEOS-CHEM model.256 The global
GEOS-CHEM model simulates atmospheric chemical and physical processes driven by
assimilated meteorological observations from the NASA's Goddard Earth Observing System
(GEOS). This model will be run with a grid resolution of 2 degree x 2.5 degree (latitude-
longitude) and 20  vertical layers. The predictions will be used to provide one-way dynamic
   In addition background information can be found in the final RIA and TSD for the 2012-2016 final rule,
http://www.epa.gov/otaq/climate/regulations.htnrfl-l.
                                         6-31

-------
Chapter 6

boundary conditions at three-hour intervals and an initial concentration field for the 36 km
CMAQ simulations. The future base conditions from the 36 km coarse grid modeling will be
used as the initial/boundary state for all subsequent 12 km finer grid modeling.

6.3 Quantified and Monetized Non-GHG Health and Environmental Impacts

       This section presents EPA's analysis of the non-GHG, or co-pollutant, health and
environmental impacts that can be expected to occur as a result of the proposed light-duty
vehicle GHG rule.  GHG emissions are predominantly the byproduct of fossil fuel combustion
processes that also produce criteria and hazardous air pollutants. The vehicles that are subject
to the proposed standards are also significant sources of mobile source air pollution such as
direct PM, NOx, VOCs and air toxics.  The proposed standards would affect exhaust
emissions of these pollutants from vehicles.  They would also affect emissions from upstream
sources related to changes in fuel consumption and electricity generation. Changes in
ambient ozone, PM2.5, and air toxics that would result from the proposed standards are
expected to affect human health in the form of premature deaths and other serious human
health effects, as well as other important public health and welfare effects.

       It is important to quantify the health and environmental impacts associated with the
proposed standard because a failure to adequately consider these ancillary co-pollutant
impacts could lead to an incorrect assessment of their net costs and benefits. Moreover, co-
pollutant impacts tend to accrue in the near term, while effects from reduced climate change
mostly accrue over a time frame of several decades or longer.

       EPA typically quantifies and monetizes the health and environmental impacts related
to both PM and ozone exposure in its regulatory impact analyses (RIAs), when possible.  To
estimate these impacts, EPA conducts full-scale photochemical modeling to provide the
needed spatial and temporal detail to estimate the changes in ambient levels of these
pollutants and their associated health and welfare impacts. However, we were unable to do so
in time for this proposal, as explained above. EPA  attempts to make emissions and air quality
modeling decisions early in the analytical process so that we can complete the photochemical
air quality modeling and use that data to inform the health and environmental impacts
analysis. Resource and time constraints precluded the Agency from completing this work in
time for the proposal.  Instead, EPA is using PM2.5-related benefits-per-ton values as an
interim approach to estimating the PM2.5-related benefits of the proposal.  We also provide a
complete characterization of the health and environmental impacts that will be quantified and
monetized for the final rulemaking.

       This section is split into two sub-sections: the first presents the PM2.5-related benefits-
per-ton values used to monetize the PM2.5-related co-benefits associated with the proposal; the
second explains what PM2 5- and ozone-related health and environmental impacts EPA will
quantify and monetize in the analysis for the final rule. EPA bases its analyses on peer-
reviewed studies of air quality and health and welfare effects  and peer-reviewed studies of the
monetary values  of public health and welfare improvements, and is generally consistent with
benefits analyses performed for the analysis of the final Transport Rule,257 the final 2012-
2016 MY Light-Duty Vehicle Rule,258 and the final Portland Cement National Emissions
Standards for Hazardous Air Pollutants (NESHAP) RIA.259
                                         6-32

-------
                                                2017 Draft Regulatory Impact Analysis
       Though EPA is characterizing the changes in emissions associated with toxic
pollutants, we will not be able to quantify or monetize the human health effects associated
with air toxic pollutants for either the proposal or the final rule analyses.  Please refer to
Chapter 4.5 of this DRIA for more information about the air toxics emissions impacts
associated with the proposed standards.

       6.3.1   Economic Value of Reductions in Criteria Pollutants

        As described in Chapter 4.5, the proposed standards would reduce emissions of
several criteria and toxic pollutants and precursors. In this analysis, EPA estimates the
economic value of the human health benefits associated with reducing PM2.5 exposure.  Due
to analytical limitations, this analysis does not estimate benefits related to other criteria
pollutants (such as ozone, NC>2 or 862) or toxic pollutants, nor does it monetize all of the
potential health and welfare effects associated with PM2.5.

       This analysis uses a "benefit-per-ton" method to estimate a selected suite of PM2.5-
related health benefits described below.  These PM2.5 benefit-per-ton estimates provide the
total monetized human health benefits (the sum of premature mortality and premature
morbidity) of reducing one ton of directly emitted PM2.s, or its precursors (such as NOx, SOx,
and VOCs),  from a specified source. Ideally, the human health benefits would be estimated
based on changes in ambient PM2.5 as determined by full-scale air quality modeling.
However, this modeling was not possible in the timeframe for this proposal, but EPA plans to
perform such modeling for the final rulemaking.

       The dollar-per-ton estimates used in this analysis are provided in Table 6.3-1. In the
summary of costs and benefits, Chapter 7.4 of this RIA, we present the monetized value of
PM-related improvements associated with the proposal.

    Table 6.3-1 Benefits-per-ton Values (2009$) Derived Using the American Cancer
Society Cohort Study for PM-related Premature Mortality (Pope et al., 2002)a and a 3%
                                    Discount Rateb
Yearc
All Sources'1
SOX
voc
Stationary (Non-EGU)
Sources"
NOX
Direct PM2. 5
Mobile Sources
NOX
Direct PM2. 5
Estimated Using a 3 Percent Discount Rate"
2015
2020
2030
2040
$29,000
$32,000
$38,000
$44,000
$1,200
$1,300
$1,600
$1,900
$4,800
$5,300
$6,300
$7,500
$230,000
$250,000
$290,000
$340,000
$5,000
$5,500
$6,600
$7,900
$280,000
$300,000
$360,000
$430,000
Estimated Using a 7 Percent Discount Rate"
2015
2020
2030
2040
$27,000
$29,000
$34,000
$40,000
$1,100
$1,200
$1,400
$1,700
$4,400
$4,800
$5,700
$6,800
$210,000
$220,000
$260,000
$310,000
$4,600
$5,000
$6,000
$7,200
$250,000
$280,000
$330,000
$390,000
a The benefit-per-ton estimates presented in this table are based on an estimate of premature mortality derived
from the ACS study (Pope et al., 2002). If the benefit-per-ton estimates were based on the Six-Cities study
(Laden et al., 2006), the values would be approximately 245% (nearly two-and-a-half times larger. See below
for a description of these studies.
                                          6-33

-------
Chapter 6

b The benefit-per-ton estimates presented in this table assume either a 3 percent or 7 percent discount rate in the
valuation of premature mortality to account for a twenty-year segmented cessation lag.
0 Benefit-per-ton values were estimated for the years 2015, 2020, and 2030. For intermediate years, such as
2017 (the year the standards begin), we interpolated exponentially. For years beyond 2030 (including 2040),
EPA and NHTSA extrapolated exponentially based on the growth between 2020 and 2030.
d Note that the benefit-per-ton value for SOx is based on the value for Stationary (Non-EGU) sources; no SOx
value was estimated for mobile sources.  The benefit-per-ton value for VOCs was estimated across all sources.
e Non-EGU denotes stationary sources of emissions other than electric generating units (EGUs).

        The benefit per-ton technique has been used in previous analyses, including EPA's
2012-2016 Light-Duty Vehicle Greenhouse Gas Rule,260 and the Portland Cement National
Emissions Standards for Hazardous Air Pollutants (NESHAP) RIA.261  Table 6.3-2  shows the
quantified and unquantified PM2.5-related co-benefits captured in those benefit-per-ton
estimates.

                 Table 6.3-2 Human Health and Welfare Effects of PM2.5
Pollutant /
Effect
Quantified and Monetized
in Primary Estimates
Unquantified Effects
Changes in:	
PM9
Adult premature mortality
Bronchitis: chronic and acute
Hospital admissions: respiratory and
cardiovascular
Emergency room visits for asthma
Nonfatal heart attacks (myocardial
infarction)
Lower and upper respiratory illness
Minor restricted-activity days
Work loss days
Asthma exacerbations (asthmatic
population)
Infant mortality
Subchronic bronchitis cases
Low birth weight
Pulmonary function
Chronic respiratory diseases other than chronic
bronchitis
Non-asthma respiratory emergency room visits
Visibility
Household soiling
        Consistent with the cost-benefit analysis that accompanied the NC>2 NAAQS,11111111'262
the benefits estimates utilize the concentration-response functions as reported in the
epidemiology literature. To calculate the total monetized impacts associated with quantified
health impacts, EPA applies values derived from a number of sources.  For premature
mortality, EPA applies a value of a statistical life (VSL) derived from the mortality valuation
literature. For certain health impacts, such as chronic bronchitis and a number of respiratory-
related ailments, EPA applies willingness-to-pay estimates derived from the valuation
literature. For the remaining health impacts, EPA applies values derived from current cost-of-
illness and/or wage estimates.
uuuu Although we summarize the main issues in this chapter, we encourage interested readers to see benefits
chapter of the NO2 NAAQS for a more detailed description of recent changes to the PM benefits presentation
and preference for the no-threshold model. Note that the cost-benefit analysis was prepared solely for purposes
of fulfilling analysis requirements under Executive Order 12866 and was not considered, or otherwise played
any part, in the decision to revise the NO2 NAAQS.
                                             6-34

-------
                                                   2017 Draft Regulatory Impact Analysis
       A more detailed description of the benefit-per-ton estimates is provided in Chapter 4
of the Draft Joint TSD that accompanies this rulemaking. Readers interested in reviewing the
complete methodology for creating the benefit-per-ton estimates used in this analysis can
consult the Technical Support Document (xSD)VVVV'WWWW accompanying the recent final
ozone NAAQS RIA (U.S. EPA, 2008).  Readers can also refer to Fann et al. (2009)xxxx for a
detailed description of the benefit-per-ton methodology.YYYY

       As described in the documentation for the benefit per-ton estimates cited above,
national per-ton estimates were developed for selected pollutant/source category
combinations. The per-ton values calculated therefore apply only to tons reduced from those
specific pollutant/source combinations (e.g., NC>2 emitted from mobile sources; direct PM
emitted from stationary sources).  Our estimate of PM2 5 benefits is therefore based on the
total direct PM2.5 and PM2.s-related precursor emissions controlled by sector and multiplied by
each per-ton value.

       The benefit-per-ton estimates are  subject to a number of assumptions and
uncertainties:

           •   They do not reflect local variability in population density, meteorology, exposure,
               baseline health incidence rates, or other local factors that might lead  to an
               overestimate or underestimate  of the actual benefits of controlling fine particulates.
               EPA will conduct full-scale air quality modeling for the final rulemaking in an effort
               to capture this variability.
           •   This analysis assumes that all fine particles, regardless of their chemical composition,
               are equally potent in causing premature mortality. This is an important assumption,
               because PM2 5 produced via transported precursors emitted from stationary sources
               may differ significantly from direct PM2 5 released from diesel engines and other
               industrial sources, but no clear scientific grounds exist for supporting differential
               effects estimates by particle type.
           •   This analysis assumes that the  health impact function for fine particles is linear within
               the range of ambient concentrations under consideration. Thus, the estimates include
vvvv jj g Environmental Protection Agency (U.S. EPA).  2008. Technical Support Document: Calculating
Benefit Per-Ton estimates, Ozone NAAQS Docket EPA-HQ-OAR-2007-0225-0284. Office of Air Quality
Planning and Standards, Research Triangle Park, NC. March.  Available on the Internet at
.
wwww ^ote jj^ ^e cost-benefit analysis was prepared solely for purposes of fulfilling analysis requirements
under Executive Order 12866 and was not considered, or otherwise played any part, in the decision to revise the
Ozone NAAQS.
xxxx Fann, N. et al. (2009). The influence of location, source, and emission type in estimates of the human
health benefits of reducing a ton of air pollution. Air Qual Atmos Health. Published online: 09 June, 2009.
YYYY -pj^ vajues inciudeci in this report are different from those presented in the article cited above. Benefits
methods change to reflect new information and evaluation of the science.  Since publication of the June 2009
article, EPA has made two significant changes to its benefits methods: (1) We no longer assume that a threshold
exists in PM-related models of health impacts; and (2) We have revised the Value of a Statistical Life to equal
$6.3 million (year 2000$), up from an estimate of $5.5 million (year 2000$) used in the June 2009 report. Please
refer to the following website for updates to the dollar-per-ton estimates:
http://www.epa.gov/air/benmap/bpt.html
                                            6-35

-------
Chapter 6

               health benefits from reducing fine particles in areas with varied concentrations of
               PM2 5  including both regions that are in attainment with fine particle standard and
               those that do not meet the standard down to the lowest modeled concentrations.
           •   There are several health benefit categories that EPA was unable to quantify due to
               limitations associated with using benefits-per-ton estimates, several of which could be
               substantial.  Because the NOXand VOC emission reductions associated with this
               proposal are also precursors to ozone, reductions in NOX and VOC would also reduce
               ozone formation and the health effects associated with ozone  exposure.
               Unfortunately, ozone-related benefits-per-ton estimates do not exist due to issues
               associated with the complexity of the atmospheric air chemistry and nonlinearities
               associated with ozone formation. The PM-related benefits-per-ton estimates also do
               not include any human welfare or ecological benefits.  Please refer to Chapter 6.3.2
               for a description of the agency's plan to quantify and monetize the PM- and ozone-
               related health impacts planned for the FRM and a description of the unqualified co-
               pollutant benefits associated with this rulemaking.
           •   There are many uncertainties associated with the health impact functions used in this
               modeling effort. These include:  within-study variability (the precision with which a
               given study estimates the relationship between air quality changes and health effects);
               across-study variation (different published studies of the  same pollutant/health effect
               relationship typically do not report identical findings and in some instances the
               differences are substantial); the application of concentration-response functions
               nationwide (does not account for any relationship between region and health effect, to
               the extent that such a relationship exists); extrapolation of impact functions across
               population (we assumed that certain health impact functions applied to age ranges
               broader than that considered in the original epidemiological study);  and various
               uncertainties in the concentration-response function, including causality and
               thresholds. These uncertainties may under- or over-estimate benefits.
           •   EPA has investigated methods to characterize uncertainty in the relationship between
               PM2 5 exposure and premature mortality. EPA's final PM2 5 NAAQS analysis provides
               a more complete picture about the overall uncertainty in PM2 5 benefits estimates.  For
               more information, please consult the PM25 NAAQS RIA (Table 5.5).263
           •   The benefit-per-ton estimates used in this analysis incorporate projections of key
               variables, including atmospheric conditions, source level emissions, population, health
               baselines and incomes, technology. These projections introduce some uncertainties to
               the benefit per ton estimates.
           •   As described above, using the benefit-per-ton value derived from the ACS study
               (Pope et al., 2002) alone provides an incomplete characterization  of PM25 benefits.
               When placed in the context of the Expert Elicitation results, this estimate falls toward
               the lower end of the distribution. By contrast, the estimated PM2 5 benefits using the
               coefficient reported by Laden in that author's reanalysis of the Harvard Six Cities
               cohort fall toward the upper end of the Expert Elicitation distribution results (Laden et
               al., 2006).

       As mentioned above, emissions changes and benefits-per-ton estimates alone are not a
good indication of local or  regional air quality and health impacts, as there  may be localized
impacts associated with the proposed rulemaking. Additionally, the atmospheric chemistry
related to ambient concentrations of PM2 5, ozone and air toxics is very complex.  Full-scale
photochemical modeling is therefore necessary to provide the needed spatial  and temporal
detail to more completely and accurately estimate the changes in ambient levels of these
pollutants and their associated health and welfare impacts. As discussed above, timing and
                                            6-36

-------
                                               2017 Draft Regulatory Impact Analysis
resource constraints precluded EPA from conducting a full-scale photochemical air quality
modeling analysis in time for the NPRM. For the final rule, however, a national-scale air
quality modeling analysis will be performed to analyze the impacts of the standards on PM2.5,
ozone, and selected air toxics. The benefits analysis plan for the final rulemaking is discussed
in the next section.

       6.3.2  Human Health and Environmental Benefits for the Final Rule

       6.3.2.1 Human Health and Environmental Impacts

       As noted above, to model the ozone and PM air quality benefits for the final rule, EPA
plans to use the Community Multiscale Air Quality (CMAQ) model (see Chapter 6.2.2.1 for a
description of the CMAQ model). The modeled ambient air quality data will serve as an input
to the Environmental Benefits Mapping and Analysis Program (BenMAP).264 BenMAP is a
computer program  developed by EPA that integrates a number of the modeling elements used
in previous RIAs (e.g., interpolation functions, population projections,  health impact
functions,  valuation functions, analysis and pooling methods) to translate modeled air
concentration estimates into health effects incidence estimates and monetized benefits
estimates.

       Table 6.3-3 lists the PM- and ozone-related health effect exposure-response functions
we will use to quantify the non-GHG incidence impacts associated with the final light-duty
vehicles standard.
                                        6-37

-------
Chapter 6
  Table 6.3-3: Health Impact Functions Used in BenMAP to Estimate Impacts of PM2.s
                             and Ozone Reductions
ENDPOINT
POLLUTANT
STUDY
STUDY
POPULATION
Premature Mortality
Premature mortality -
daily time series
Premature mortality —
cohort study, all-cause
Premature mortality, total
exposures
Premature mortality —
all-cause
03
PM25
PM25
PM25
Multi-city
Bell et al (2004) (NMMAPS study)265 -
Non-accidental
Huang et al (2005)266 - Cardiopulmonary
Schwartz (2005)267 - Non-accidental
Meta-analyses:
Bell et al (2005)268 - All cause
Ito et al (2005)269 - Non-accidental
Levy et al (2005)270 - All cause
Popeetal. (2002)2'1
Laden et al. (2006)272
Expert Elicitation (lEc, 2006)273
Woodruff etal. (1997)274
All ages
>29 years
>25 years
>24 years
Infant (<1 year)
Chronic Illness
Chronic bronchitis
Nonfatal heart attacks
PM25
PM25
Abbey etal. (1995)275
Peters etal. (200 1)276
>26 years
Adults (>18
years)
Hospital Admissions
Respiratory
Cardiovascular
03
PM25
PM25
PM25
PM25
PM25
Pooled estimate:
Schwartz (1995) - ICD 460-519 (all resp)277
Schwartz (1994a; 1994b) - ICD 480-486
(pneumonia)278'279
Moolgavkar et al. (1997) - ICD 480-487
(pneumonia)280
Schwartz (1994b) - ICD 491-492, 494-496
(COPD)
Moolgavkar et al. (1997) - ICD 490-496
(COPD)
Burnett etal. (200 1)281
Pooled estimate:
Moolgavkar (2003)— ICD 490-496
(COPD)282
Ito (2003)— ICD 490-496 (COPD)283
Moolgavkar (2000)— ICD 490-496
(COPD)284
Ito (2003)— ICD 480-486 (pneumonia)
Sheppard (2003)— ICD 493 (asthma)285
Pooled estimate:
Moolgavkar (2003)— ICD 390-429 (all
cardiovascular)
>64 years
<2 years
>64 years
20-64 years
>64 years
<65 years
>64 years
                                     6-38

-------
                                                     2017 Draft Regulatory Impact Analysis

Asthma-related ER visits
Asthma-related ER visits
(con't)

PM25
03
PM25
Ito (2003)— ICD 410-414, 427-428
(ischemic heart disease, dysrhythmia, heart
failure)
Moolgavkar (2000)— ICD 390-429 (all
cardiovascular)
Pooled estimate:
Peel et al (2005)286
Wilson et al (2005)287
Norrisetal. (1999)288

20-64 years
All ages
All ages
0-18 years
Other Health Endpoints
Acute bronchitis
Upper respiratory
symptoms
Lower respiratory
symptoms
Asthma exacerbations
Work loss days
School absence days
Minor Restricted Activity
Days (MRADs)
PM25
PM25
PM25
PM25
PM25
03
03
PM25
Dockeryetal. (1996)289
Popeetal. (1991)290
Schwartz and Neas (2000)291
Pooled estimate:
Ostro et al. (200 1)292 (cough, wheeze and
shortness of breath)
OQQ
Vedal et al. (1998) (cough)
Ostro (1987)294
Pooled estimate:
Gilliland et al. (2001)295
Chenetal. (2000)296
Ostro and Rothschild (1989)297
Ostro and Rothschild (1989)
8-12 years
Asthmatics, 9-11
years
7-14 years
6-18 years3
18-65 years
5-17 years'3
18-65 years
18-65 years
Notes:
a The original study populations were 8 to 13 for the Ostro et al. (2001) study and 6 to 13 for the Vedal et al.
(1998) study. Based on advice from the Science Advisory Board Health Effects Subcommittee (SAB-HES), we
extended the applied population to 6 to 18, reflecting the common biological basis for the effect in children in
the broader age group. See: U.S. Science Advisory Board. 2004. Advisory Plans for Health Effects Analysis in
the Analytical Plan for EPA's Second Prospective Analysis -Benefits and Costs of the Clean Air Act, 1990—
2020. EPA-SAB-COUNCIL-ADV-04-004. See also National Research Council (NRC). 2002. Estimating the
Public Health Benefits of Proposed Air Pollution Regulations.  Washington, DC: The National Academies
Press.
b Gilliland et al. (2001) studied children aged 9 and 10. Chen et al. (2000) studied children 6 to 11. Based on
recent advice from the National Research Council and the EPA SAB-HES, we have calculated reductions in
school absences for all school-aged children based on the biological similarity between children aged 5 to 17.

        6.3.2.2  Monetized Estimates of Impacts of Reductions in  Co-Pollutants

        Table 6.3-4 presents the monetary values we will apply to changes in the incidence of
health and welfare effects associated with reductions in non-GHG pollutants that will occur
when these GHG control strategies are finalized.
                                              6-39

-------
 Chapter 6
   Table 6.3-4: Valuation Metrics Used in BenMAP to Estimate Monetary Co-Benefits
Endpoint
Valuation Method
Valuation (2009$)
Premature mortality
Assumed Mean VSL
    $7,850,000
Chronic Illness
 Chronic Bronchitis
 Myocardial Infarctions,
Nonfatal
WTP: Average Severity
Medical Costs Over 5 Years. Varies by age and
discount rate. Russell (1998)298
Medical Costs Over 5 Years. Varies by age and
discount rate. Wittels (1990)299
    $424,193
Hospital Admissions
Respiratory, Age 65+
Respiratory, Ages 0-2
COI: Medical Costs -
COI: Medical Costs
i- Wage Lost
$26,433
$11,149
 Chronic Lung Disease (less
 Asthma)
 Pneumonia
 Asthma
COI: Medical Costs + Wage Lost

COI: Medical Costs + Wage Lost
COI: Medical Costs + Wage Lost
     $17,827

     $21,161
     $9,555
 Cardiovascular
COI: Medical Costs + Wage Lost (20-64)
COI: Medical Costs + Wage Lost (65-99)
     $32,806
     $30,520
ER Visits, Asthma
COI: Smith etal. (1997)
COI: Standford et al. (1999)
                                                    301
      $449
      $376
Other Health Endpoints
 Acute Bronchitis
 Upper Respiratory Symptoms
 Lower Respiratory Symptoms
 Asthma Exacerbation

 Work Loss Days
 Minor Restricted Activity
Days
 School Absence Days
 Worker Productivity
WTP: 6 Day Illness, CV Studies
WTP: 1 Day, CV Studies
WTP: 1 Day, CV Studies
WTP: Bad Asthma Day, Rowe and Chestnut (1986)
302

Median Daily Wage, County-Specific
WTP: 1 Day, CV Studies

Median Daily Wage, Women 25+
Median Daily Wage, Outdoor Workers, County-
Specific	
      $444
       $31
       $20
       $54
       $64

       $93
Environmental Endpoints
 Recreational Visibility
WTP: 86 Class I Areas
Source: Dollar amounts for each valuation method were extracted from BenMAP and adjusted to year 2009
dollars (from year 2000 dollars) using the Consumer Price Urban Index (CPI-U).  For endpoints valued using
measures of VSL, WTP, or are wage-based, we use the CPI-U for "all items": 214.537 (2009) and 172.2 (2000).
For endpoints valued using a Cost-of-Illness measure, we use the CPI-U for "medical care": 375.613 (2009) and
260.8 (2000)..

        6.3.2.3 Other Unquantified Health and Environmental Impacts

         In addition to the co-pollutant health and environmental impacts we plan to quantify
for the analysis of the Light-Duty Vehicle GHG standard, there are  a number of other health
and human welfare endpoints that we will not be able to quantify because of current
limitations in the methods or available data.  These impacts are associated with emissions of
air toxics (including benzene, 1,3-butadiene, formaldehyde, acetaldehyde, acrolein, and
ethanol), ambient ozone, and ambient PM2 5 exposures. For example, we have not quantified
a number of known or suspected health effects  linked with ozone and PM for which
appropriate health impact functions are not available or which do not provide easily
                                            6-40

-------
                                                 2017 Draft Regulatory Impact Analysis
interpretable outcomes (i.e., changes in heart rate variability).  In addition, we are currently
unable to quantify a number of known welfare effects, including reduced acid and particulate
deposition damage to cultural monuments and other materials, and environmental benefits due
to reductions of impacts of eutrophication in coastal areas.  Table 6.3-5 lists these
unquantified health and environmental impacts.

       Although there will be impacts associated with air toxic pollutant emission changes
that result from this action, we do not attempt to monetize those impacts. This is primarily
because currently available tools and methods to assess air toxics risk from mobile sources at
the national scale are not adequate for extrapolation to incidence estimations or benefits
assessment. The best suite of tools and methods currently available for assessment at the
national scale are those used in the National-Scale Air Toxics Assessment (NATA). The EPA
Science Advisory Board  specifically commented in their review of the 1996 NATA that these
tools were not yet ready for use in a national-scale benefits  analysis, because they did not
consider the full distribution of exposure and risk, or address sub-chronic health effects.303
While EPA has since improved these tools, there remain critical limitations for estimating
incidence and assessing benefits of reducing mobile source air toxics.

       As part of the second prospective analysis of the benefits and costs of the Clean Air
Act,304 EPA conducted a case study analysis of the health effects associated with reducing
exposure to benzene in Houston from implementation of the Clean Air Act. While reviewing
the draft report, EPA's Advisory Council on Clean Air Compliance Analysis concluded  that
"the challenges for assessing progress in health improvement as a result of reductions in
emissions of hazardous air pollutants (HAPs) are daunting.due to a lack of exposure-response
functions, uncertainties in emissions inventories and background levels, the difficulty of
extrapolating risk estimates to low doses and the challenges of tracking health progress for
diseases, such as cancer,  that have long latency periods."305 EPA continues to work to
address these limitations; however, we did not have the methods and tools available for
national-scale application in time for the analysis of this action.zzzz We seek public comment
to inform how the Agency might do this in the future.
             Table 6.3-5: Unquantified and Non-Monetized Potential Effects
POLLUTANT/EFFECTS
Ozone Healtha
EFFECTS NOT INCLUDED IN ANALYSIS - CHANGES
IN:
Chronic respiratory damage
Premature aging of the lungs
Non-asthma respiratory emergency room visits
zzzz In April, 2009, EPA hosted a workshop on estimating the benefits or reducing hazardous air pollutants.
This workshop built upon the work accomplished in the June 2000 Science Advisory Board/EPA Workshop on
the Benefits of Reductions in Exposure to Hazardous Air Pollutants, which generated thoughtful discussion on
approaches to estimating human health benefits from reductions in air toxics exposure, but no consensus was
reached on methods that could be implemented in the near term for a broad selection of air toxics. Please visit
http://epa.gov/air/toxicair/2009workshop.html for more information about the workshop and its associated
materials.
                                          6-41

-------
Chapter 6

Ozone Welfare
PM Healthb
PM Welfare
Nitrogen and Sulfate
Deposition Welfare
CO Health
Hydrocarbon (HC)/Toxics
Health6
HC/Toxics Welfare'
Exposure to UVb (+/-)d
Yields for
-commercial forests
-some fruits and vegetables
-non-commercial crops
Damage to urban ornamental plants
Impacts on recreational demand from damaged forest aesthetics
Ecosystem functions
Exposure to UVb (+/-)
Premature mortality - short term exposures0
Low birth weight
Pulmonary function
Chronic respiratory diseases other than chronic bronchitis
Non-asthma respiratory emergency room visits
Exposure to UVb (+/-)
Residential and recreational visibility in non-Class I areas
Soiling and materials damage
Damage to ecosystem functions
Exposure to UVb (+/-)
Commercial forests due to acidic sulfate and nitrate deposition
Commercial freshwater fishing due to acidic deposition
Recreation in terrestrial ecosystems due to acidic deposition
Existence values for currently healthy ecosystems
Commercial fishing, agriculture, and forests due to nitrogen deposition
Recreation in estuarine ecosystems due to nitrogen deposition
Ecosystem functions
Passive fertilization
Behavioral effects
Cancer (benzene, 1,3 -butadiene, formaldehyde, acetaldehyde, ethanol)
Anemia (benzene)
Disruption of production of blood components (benzene)
Reduction in the number of blood platelets (benzene)
Excessive bone marrow formation (benzene)
Depression of lymphocyte counts (benzene)
Reproductive and developmental effects (1,3 -butadiene, ethanol)
Irritation of eyes and mucus membranes (formaldehyde)
Respiratory irritation (formaldehyde)
Asthma attacks in asthmatics (formaldehyde)
Asthma-like symptoms in non-asthmatics (formaldehyde)
Irritation of the eyes, skin, and respiratory tract (acetaldehyde)
Upper respiratory tract irritation and congestion (acrolein)
Direct toxic effects to animals
Bioaccumulation in the food chain
Damage to ecosystem function
Odor
   In addition to primary economic endpoints, there are a number of biological responses that have been
associated with ozone health effects including increased airway responsiveness to stimuli, inflammation in the
lung, acute inflammation and respiratory cell damage, and increased susceptibility to respiratory infection. The
public health impact of these biological responses may be partly represented by our quantified endpoints.
                                                6-42

-------
                                                 2017 Draft Regulatory Impact Analysis
 In addition to primary economic endpoints, there are a number of biological responses that have been
associated with PM health effects including morphological changes and altered host defense mechanisms. The
public health impact of these biological responses may be partly represented by our quantified endpoints.
c While some of the effects of short-term exposures are likely to be captured in the estimates, there may be
premature mortality due to short-term exposure to PM not captured in the cohort studies used in this analysis.
However, the PM mortality results derived from the expert elicitation do take into account premature mortality
effects of short term exposures.
 May result in benefits or disbenefits.
e Many of the key hydrocarbons related to this rule are also hazardous air pollutants listed in the Clean Air Act.
Please refer to Chapter 6.1 for additional information on the health effects of air toxics.
f Please refer to Chapter 6. Ifor additional information on the welfare effects of air toxics.
6.4    Changes in Atmospheric COi Concentrations, Global Mean Temperature, Sea
       Level Rise, and Ocean pH Associated with the Proposed Rule's GHG Emissions
       Reductions

       6.4.1   Introduction

       The impact of GHG emissions on the climate has been reviewed in the 2012-2016
light-duty rulemaking and recent heavy-duty GHG rulemaking. See 75 FR at 25491; 76 FR at
57294.  This section briefly discusses again some of the climate impact context for
transportation emissions. These previous discussions noted that once emitted, GHGs that are
the subject of this regulation can remain in the atmosphere for decades to millennia, meaning
that 1) their concentrations become well-mixed throughout the global atmosphere regardless
of emission origin, and 2) their effects on climate are long lasting. GHG emissions come
mainly from the combustion of fossil fuels (coal, oil, and gas), with additional contributions
from the clearing of forests, agricultural activities, cement production, and some industrial
activities. Transportation activities, in aggregate, were the second largest contributor to total
U.S. GHG emissions in 2009 (27 percent of total emissions).306

       The Administrator relied on thorough and peer-reviewed assessments of climate
change science prepared by the Intergovernmental Panel on Climate Change ("IPCC"), the
United States Global Change Research Program ("USGCRP"), and the National Research
Council of the National Academies ("NRC") AAAAA as the primary scientific and technical
basis for the Endangerment and Cause or Contribute Findings for Greenhouse Gases Under
Section 202(a) of the Clean Air Act (74 FR 66496, December 15, 2009).  These assessments
comprehensively address the scientific issues the Administrator had to examine, providing her
both data and information on a wide range  of issues pertinent to the Endangerment Finding.
These assessments have been rigorously reviewed by the expert community, and  also by
United States government agencies and scientists,  including by EPA itself.
     For a complete list of core references from IPCC, USGCRP/CCSP, NRC and others relied upon for
development of the TSD for EPA's Endangerment and Cause or Contribute Findings see section l(b),
specifically, Table 1.1 of the TSD. Docket: EPA-HQ-OAR-2009-0171-11645.
                                          6-43

-------
Chapter 6

       Based on these assessments, the Administrator determined, in essence, that greenhouse
gases cause warming; that levels of greenhouse gases are increasing in the atmosphere due to
human activity; the climate is warming; recent warming has been attributed to the increase in
greenhouse gases; and that warming of the climate threatens human health and welfare.  The
Administrator further found that emissions of well-mixed greenhouse gases from new motor
vehicles and engines contribute to the air pollution for which the endangerment finding was
made.  Specifically, the Administrator found under section 202 (a) of the Act that six
greenhouse gases (carbon dioxide, methane, nitrous oxide, hydrofluorocarbons,
perfluorocarbons, and  sulfur hexafluoride) taken in combination endanger both the public
health and the public welfare of current and future generations, and further found that the
combined emissions of these greenhouse gases from new motor vehicles and engines
contribute to the greenhouse gas air pollution that endangers public health and welfare.

       More recent assessments have produced similar conclusions to those of the
assessments upon which the Administrator relied. In May 2010, the NRC published its
comprehensive assessment, "Advancing the Science of Climate Change."307  It concluded that
"climate change is occurring, is caused largely by human activities,  and poses significant risks
for—and in many cases is already affecting—a broad range of human and natural systems."
Furthermore, the NRC stated that this conclusion is based on findings that are "consistent with
the conclusions of recent assessments by the U.S. Global Change Research Program, the
Intergovernmental Panel on Climate Change's Fourth Assessment Report, and other
assessments of the state of scientific knowledge on climate change." These are the same
assessments that served as the primary scientific references underlying the Administrator's
Endangerment Finding.  Another NRC assessment, "Climate Stabilization Targets: Emissions,
Concentrations, and Impacts over Decades to Millenia", was published in 2011. This report
found that climate change due to carbon dioxide emissions will persist for many centuries.
The report also estimates a number of specific climate change impacts, finding that every
degree Celsius (C) of warming could lead to increases in the heaviest 15% of daily rainfalls of
3 to 10%, decreases of 5 to 15% in yields for a number of crops (absent adaptation measures
that do not presently exist), decreases of Arctic sea ice extent of 25% in September and 15%
annually averaged,  along with changes in precipitation and streamflow of 5 to 10% in many
regions and river basins (increases in some regions, decreases in others). The assessment also
found that for an increase of 4 degrees C nearly all land areas would experience summers
warmer than all but 5% of summers in the 20th century, that for an increase of 1 to 2 degrees
C the area burnt by wildfires in western North America will likely more than double, that
coral bleaching  and erosion will increase due both to warming and ocean acidification, and
that sea level will rise  1.6 to 3.3 feet by 2100 in a 3 degree C scenario. The assessment notes
that many important aspects of climate change are difficult to quantify but that the risk of
adverse impacts is likely to increase with increasing temperature, and that the risk of abrupt
climate changes can be expected to increase with the duration and magnitude of the warming.

       In the 2010 report cited above, the NRC stated that some of the largest potential risks
associated with future  climate change may come not from relatively smooth changes that are
reasonably well understood, but from extreme events, abrupt changes,  and surprises that
might occur when climate or environmental system thresholds are crossed. Examples cited as
warranting more research include the release of large quantities of GHGs  stored in permafrost
(frozen soils) across the Arctic, rapid disintegration of the major ice sheets, irreversible drying
                                        6-44

-------
                                                 2017 Draft Regulatory Impact Analysis
and desertification in the subtropics, changes in ocean circulation, and the rapid release of
destabilized methane hydrates in the oceans.

       On ocean acidification, the same report noted the potential for broad, "catastrophic"
impacts on marine ecosystems.  Ocean acidity has increased 25 percent since pre-industrial
times, and is projected to continue increasing. By the time atmospheric CO2 content doubles
over its preindustrial value, there would be virtually no place left in the ocean that can sustain
coral reef growth. Ocean acidification could have dramatic consequences for polar food webs
including salmon, the report said.

       Importantly, these recent NRC assessments represent another independent and critical
inquiry of the state of climate change science, separate and apart from the previous  IPCC and
USGCRP assessments.

       Based on modeling analysis performed by the EPA, reductions in CO2 and other GHG
emissions associated with this proposed rule will affect future climate change. Since GHGs
are well-mixed in the atmosphere and have long atmospheric lifetimes, changes in GHG
emissions will affect atmospheric concentrations of greenhouse gases and future climate for
decades to millennia, depending on the gas. This section provides estimates of the projected
change in atmospheric CO2 concentrations based on the emission reductions estimated for this
proposed rule, compared to the reference case. In addition, this section analyzes the response
to the changes in GHG concentrations of the following climate-related variables:  global mean
temperature, sea level rise, and ocean pH. See Chapter 4 in this DRIA for the estimated net
reductions in global emissions over time by GHG.BBBBB

       6.4.2  Projected Change in Atmospheric CO2 Concentrations, Global Mean Surface
      Temperature and Sea Level Rise

         To assess the impact of the emissions reductions from the proposed rule, EPA
estimated changes in projected atmospheric CO2 concentrations, global mean surface
temperature and sea-level rise to 2100 using the  GCAM (Global Change Assessment Model,
formerly MiniCAM), integrated assessment modelccccc'308 coupled with the MAGICC
(Model for the Assessment of Greenhouse-gas Induced Climate  Change) simple climate
model.DDDDD'309'310 GCAM was used to create the globally and temporally consistent set of
          tmijng constraints, the modeling analysis in this section was conducted with preliminary estimates
of the emissions reductions projected from this proposal, which were similar to the final estimates presented in
Chapter 4 of this DRIA.
ccccc QCAM is a long-term, global integrated assessment model of energy, economy, agriculture and land use
that considers the sources of emissions of a suite of greenhouse gases (GHG's), emitted in 14 globally
disaggregated regions, the fate of emissions to the atmosphere, and the consequences of changing concentrations
of greenhouse related gases for climate change. GCAM begins with a representation of demographic and
economic developments in each region and combines these with assumptions about technology development to
describe an internally consistent representation of energy, agriculture, land-use, and economic developments that
in turn shape global emissions.
             consists of a suite of coupled gas-cycle, climate and ice-melt models integrated into a single
framework. The framework allows the user to determine changes in greenhouse-gas concentrations, global-mean
                                          6-45

-------
Chapter 6

climate relevant emissions required for running MAGICC. MAGICC was then used to
estimate the projected change in relevant climate variables over time. Given the magnitude of
the estimated emissions reductions associated with the proposal, a simple climate model such
as MAGICC is appropriate for estimating the atmospheric and climate response.
       6.4.2.1 Methodology

       Emissions reductions associated with this proposal were evaluated with respect to a
baseline reference case. An emissions scenario was developed by applying the estimated
emissions reductions from the proposed rule relative to the baseline to the GCAM reference
(no climate policy) scenario (used as the basis for the Representative Concentration Pathway
RCP4.5).311 Specifically, the annual CO2, N2O, CH4, HFC-134a, NOx, CO, and SO2
emissions reductions estimated from this proposal were applied as net reductions to the
GCAM global baseline net emissions for each sub stance.EEEEE The emissions reductions past
2050 for all emissions were scaled with total U.S. road transportation fuel consumption from
the GCAM reference scenario. This was chosen as a simple scale factor given that both direct
and upstream emissions changes are included in the emissions reduction scenario provided.
Road transport fuel consumption past 2050 does not change significantly and thus emissions
reductions remain relatively constant from 2050 through 2100.

       The GCAM reference scenario312 depicts a world in which global population reaches a
maximum of more than 9 billion in 2065 and then declines to 8.7 billion in 2100 while global
GDP grows by an order of magnitude and global energy consumption triples.  The reference
scenario includes no explicit policies to limit carbon emissions, and therefore fossil fuels
continue to dominate global energy consumption, despite substantial  growth in nuclear and
renewable energy.  Atmospheric CO2 concentrations rise throughout the century and reach
760 to 820 ppmv by 2100, depending on climatic parameters, with total radiative forcing
increasing more than 5 Watts per square meter (W/m2) above 1990 levels by 2100. Forest
land declines in the reference scenario to accommodate increases in land use for food and
bioenergy crops. Even with the assumed agricultural productivity increases, the amount of
land devoted to crops increases in the first half of the century due to increases in population
and income (higher income drives increases in land-intensive meat consumption). After 2050
the rate of growth in food demand slows, in part due to declining population. As a result the
amount of cropland and also land use change (LUC) emissions decline as agricultural crop
productivity continues to increase.
surface air temperature and sea-level resulting from anthropogenic emissions of carbon dioxide (CO2), methane
(CH4), nitrous oxide (N2O), reactive gases (CO, NOx, VOCs), the halocarbons (e.g. HCFCs, HFCs, PFCs) and
sulfur dioxide (SO2). MAGICC emulates the global-mean temperature responses of more sophisticated coupled
Atmosphere/Ocean General Circulation Models (AOGCMs) with high accuracy.


EEEEE Que to jjjjjjjjg constraints, the modeling analysis in this section was conducted with preliminary estimates
of the emissions reductions projected from this proposal, which were similar to the final estimates presented in
Chapter 4 of this DRI A.


                                         6-46

-------
                                                2017 Draft Regulatory Impact Analysis
       The GCAM reference scenario uses non-CO2 and pollutant emissions implemented as
described in Smith and Wigley (2006); land-use change emissions as described in Wise et al.
(2009); and updated base-year estimates of global GHG emissions.  This scenario was created
as part of the Climate Change Science Program (CCSP) effort to develop a set of long-term
global emissions scenarios that incorporate an update of economic and technology data and
utilize improved scenario development tools compared to the IPCC Special Report on
Emissions Scenarios (SRES) (IPCC 2000).

       Using MAGICC 5.3 v2,313 the change in atmospheric CC>2 concentrations, global
mean temperature, and sea level were projected at five-year time steps to 2100 for both the
reference (no climate policy) scenario and the emissions reduction scenario specific to the
proposed rule. To capture some of the uncertainty in the climate system, the changes in
projected atmospheric CC>2 concentrations, global mean temperature and sea level were
estimated across the most current Intergovernmental Panel on Climate Change (IPCC) range
of climate sensitivities, 1.5°C to 6.0°C.FFFFF The range as illustrated in Chapter 10, Box 10.2,
Figure 2  of the IPCC's Working Group I is approximately consistent with the 10-90%
probability distribution of the individual cumulative distributions of climate sensitivity.314
Other uncertainties, such as uncertainties regarding the carbon cycle, ocean heat uptake,
different baseline emissions  scenarios, or aerosol forcing, were not addressed.

       MAGICC calculates  the forcing response at the global scale from changes in
atmospheric concentrations of CC>2, CH4, N2O, HFCs, and tropospheric ozone. It also includes
the effects of temperature changes on stratospheric ozone and the effects of CFLt  emissions on
stratospheric water vapor. Changes in CH4, NOx, VOC, and CO emissions affect both O3
concentrations and CH4 concentrations. MAGICC includes the relative climate forcing effects
of changes in sulfate concentrations due to changing SO2 emissions, including both  the direct
effect of sulfate particles and the indirect effects related to cloud interactions. However,
MAGICC does not calculate the effect of changes in concentrations of other aerosols such as
nitrates, black carbon, or organic carbon, making the assumption that the sulfate cooling
effect is a proxy for the sum of all the aerosol effects. Therefore, the climate effects  of
changes in PM2.5 emissions and precursors (besides 802) which were presented in Chapter 4
were not included in the calculations in this chapter. MAGICC also calculates all climate
effects at the global scale. This global scale captures the climate effects of the long-lived,
well-mixed greenhouse gases, but does not address the fact that short-lived climate forcers
such as aerosols and ozone can have effects that vary with location and timing of emissions.
Black carbon in particular is known to cause a positive forcing or warming effect by
absorbing incoming solar radiation, but there are uncertainties about the magnitude of that
FFFFF
     In IPCC reports, equilibrium climate sensitivity refers to the equilibrium change in the annual mean global
surface temperature following a doubling of the atmospheric equivalent carbon dioxide concentration. The IPCC
states that climate sensitivity is "likely" to be in the range of 2°C to 4.5°C, "very unlikely" to be less than 1.5°C,
and "values substantially higher than 4.5°C cannot be excluded." IPCC WGI, 2007, Climate Change 2007 - The
Physical Science Basis, Contribution of Working Group I to the Fourth Assessment Report of the IPCC,
http://www.ipcc.ch/.
                                         6-47

-------
Chapter 6

warming effect and the interaction of black carbon (and other co-emitted aerosol species) with
clouds.  While black carbon is likely to be an important contributor to climate change, it
would be premature to include quantification of black carbon climate impacts in an analysis
of the proposed standards.  See generally, EPA, Response to Comments to the Endangerment
Finding Vol. 9 section 9.1.6.1 and the discussion of black carbon in the endangerment finding
at 74 FR at 66520. Additionally, the magnitude of PM2.5 emissions changes (and therefore,
black carbon emission changes) related to these proposed standards are small in comparison
to the changes in the pollutants which have been included in the MAGICC model simulations.

       To compute the changes in atmospheric CC>2 concentration, global mean temperature,
and sea level rise specifically attributable to the impacts of the proposal, the difference in
emissions between the proposal and the baseline scenario was subtracted from the GCAM
reference emissions scenario. As a result of the emissions reductions from the proposed rule
relative to the baseline case, the concentration of atmospheric CC>2 is projected to be reduced
by approximately 3.3 to 3.7 parts per million by volume (ppmv), the global mean temperature
is projected to be reduced by approximately 0.008-0.018°C, and global mean sea level rise is
projected to be reduced by  approximately 0.07-0.17 cm by 2100. For sea level rise, the
calculations in MAGICC do not include the possible effects of accelerated ice flow in
Greenland and/or Antarctica.

       Figure 6.4-1 provides the results over time for the estimated reductions in
atmospheric CC>2 concentration associated with the proposal compared to the baseline case.
Figure 6.4-2 provides the estimated change in projected global mean temperatures associated
with the proposal. Figure 6.4-3 provides the estimated reductions in global mean sea level
rise associated with the proposal.  The range of reductions in global mean temperature and sea
level rise due to uncertainty in climate sensitivity is larger than that for CC>2 concentrations
because CC>2 concentrations are only weakly coupled to climate sensitivity through the
dependence on temperature of the rate of ocean absorption of CO2, whereas the magnitude of
temperature change response to CC>2 changes (and therefore sea level rise) is more tightly
coupled to climate sensitivity in the MAGICC  model.
                                        6-48

-------
                                             2017 Draft Regulatory Impact Analysis
                        Change in CO2 Concentration
                          (Proposed Rule - Baseline)
             o.o
             -4.0
               2000
2020
2040
2060
2080
2100
Figure 6.4-1 Projected Reductions in Atmospheric COi Concentrations (parts per
million by volume) from the Proposed Rule (climate sensitivity (CS) cases ranging from
1.5-6.0°C)
                     Change in Global Mean Temperature
                          (Proposed Rule - Baseline)
            0.000
         u
         01
           -0.016  -
           -0.020
                2000
2020
2040
2060
2080
2100
Figure 6.4-2 Projected Reductions in Global Mean Surface Temperatures from the
Proposed Rule (climate sensitivity (CS) cases ranging from 1.5-6.0°C)
                                       6-49

-------
Chapter 6
                           Change in Global Mean Sea Level Rise
                                (Proposed Rule - Baseline)
                0.000
                -0.200
                    2000
2020
2040
2060
2080
2100
  Figure 6.4-3 Projected Reductions in Global Mean Sea Level Rise from the Proposed
              Rule (climate sensitivity (CS) cases ranging from 1.5-6.0°C)
       The results in Figure 6.4-2 and Figure 6.4-3 show reductions in the projected global
mean temperature and sea level respectively, across all climate sensitivities. The projected
reductions are small relative to the change in temperature (1.8 - 4.8 °C) and sea level rise (23
- 55 cm) from 1990 to 2100 from the MAGICC simulations for the GCAM reference case.
However, this is to be expected given the magnitude of emissions reductions expected from
the proposal in the context of global emissions. Again, it should be noted that the calculations
in MAGICC do not include the possible effects of accelerated ice flow in Greenland and/or
Antarctica: the recent NRC report estimated a likely sea level increase for the business-as-
usual A1B SRES scenario of 0.5 to 1.0 meters, almost double the estimate from MAGICC, so
projected reductions in sea level rise may be similarly underestimated.315 If other uncertainties
besides climate sensitivity were included in the analysis, the resulting ranges of projected
changes would likely be slightly larger.
       6.4.3  Projected Change in Ocean pH

       For this proposal, EPA analyzes another key climate-related variable and calculates
projected change in ocean pH for tropical waters. For this analysis, changes in ocean pH are
related to the change in the atmospheric concentration of carbon dioxide (CO2) resulting from
                                         6-50

-------
                                               2017 Draft Regulatory Impact Analysis
the emissions reductions associated with the proposed rule.GGGGG EPA used the proposal
developed for CO2 System Calculations CO2SYS,316 version 1.05, a proposal which performs
calculations relating parameters of the carbon dioxide (€62) system in seawater.  The
proposal was developed by Ernie Lewis at Brookhaven National Laboratory and Doug
Wallace at the Institut fur Meereskunde in Germany,  supported by the U.S. Department of
Energy, Office of Biological and Environmental Research, under Contract No. DE-ACO2-
76CH00016.

       The proposal uses two of the four measurable parameters of the CC>2 system [total
alkalinity (TA), total inorganic CC>2 (TC), pH, and either fugacity (fCO2) or partial pressure of
CO2 (pCO2)] to calculate the other two parameters given a specific set of input conditions
(temperature and pressure) and output conditions chosen by the user. EPA utilized the DOS
version (Lewis and Wallace, 1998)317 of the program to compute pH for three scenarios: the
reference scenario at a climate sensitivity of 3 degrees for which the CO2 concentrations was
calculated to be 784.868 in 2100, the proposed rule relative to the baseline with a CO2
concentration of 781.419, and a calculation for 1990 with a CO2 concentration of 353.633. .

       Using the set of seawater parameters detailed below, the EPA calculated pH levels for
the three scenarios. The pH of the proposed emissions standards relative to the baseline
scenario pH was +0.0018 pH units (more basic). For comparison, the difference between the
reference scenario in 2100 and the pH in  1990 was -0.30 pH units (more acidic).

       The CO2SYS program required the input of a number of variables and constants for
each scenario for calculating the result for both the reference case and the proposed rule's
emissions reduction case. EPA used the following inputs, with justification and references for
these inputs provided in brackets:

       1)     Input mode: Single-input [This simply means that the program calculates pH
for one set of input variables at a time,  instead of a batch of variables. The choice has no
affect on results].

       2)     Choice of constants: Mehrbach et al. (1973)318, refit by Dickson and Millero
(1987)319

       3)     Choice of fCO2 or pCO2:  pCO2 [pCO2 is the partial pressure of CO2 and can
be converted to fugacity (fCO2) if desired]

       4)     Choice of KSO4: Dickson (1990)320 [Lewis and Wallace (1998)321 recommend
using the equation of Dickson (1990) for this dissociation constant. The model also allows the
use of the  equation of Khoo et al. (1977).322 Switching this parameter to Khoo et al. (1977)
instead of Dickson (1990) had no effect on the calculated result].
GGGGG Que to jjjjjjjjg constraints, the modeling analysis in this section was conducted with preliminary estimates
of the emissions reductions projected from this proposal, which were highly similar to the final estimates
presented in Chapter 4 of this DRIA.


                                         6-51

-------
Chapter 6

       5)     Choice of pH scale: Total scale [The model allows pH outputs to be provided
on the total scale, the seawater scale, the free scale, and the National Bureau of Standards
(NBS) scale. The various pH scales can be interrelated using equations provided by Lewis and
Wallace (1998)].

       The program provides several choices of constants for saltwater that are needed for the
calculations. EPA calculated pH values using all choices and found that in all cases the choice
had an indistinguishable effect on the results. In addition, EPA ran the model using a variety
of other required input values to test whether the model was sensitive to these inputs. EPA
found the model was not sensitive to these inputs in terms of the incremental change in pH
calculated for each climate sensitivity case.  The input values are derived from certified
                                                                            101
reference materials of sterilized natural sea water (Dickson, 2003, 2005, and 2009).  Based
on the projected atmospheric CO2 concentration reductions that would result from this
proposed rule (784.868 ppmv for a climate sensitivity of 3.0), the modeling program
calculates an increase in ocean pH of approximately 0.0018 pH units in 2100. Thus, this
analysis indicates the projected decrease in atmospheric CO2 concentrations from the
proposed standards yields an increase in ocean pH.  Table 6.4-1 contains the projected
changes in ocean pH based the change in atmospheric CO2 concentrations which were
derived from the MAGICC modeling.

    Table 6.4-1 Impact of the Proposal's GHG Emissions Reductions On Ocean pHa
CLIMATE
SENSITIVITY
3.0
DIFFERENCE
INCO2a
-3.45 ppm
YEAR
2100
PROJECTED
CHANGE
+0.0018
a represents the change in atmospheric CO2 concentrations in 2100 based on the difference from the proposed
rule relative to the base case from the GCAM reference scenario used in the MAGICC modeling.

       6.4.4   Summary of Climate Analyses

       EPA's analysis of the impact of the emissions reductions from this proposal on global
climate conditions is intended to quantify these potential reductions using the best available
science.  While EPA's modeling results of the impact of this proposal alone show small
differences in climate effects (CO2 concentration, global mean temperature,  sea level rise, and
ocean pH), in comparison to the total projected changes, they yield results that are repeatable
and directionally consistent within the modeling frameworks used.  The results are
summarized in Table 6.4-2, Impact of GHG Emissions Reductions On Projected Changes in
Global Climate Associated with the Proposal.

       These projected reductions are proportionally representative of changes to U.S. GHG
emissions in the transportation sector. While not formally estimated for this  proposal, a
reduction in projected global mean temperature change, sea level rise, and ocean acidification
implies a reduction in the risks associated with climate change. The figures for these variables
illustrate that across a range of climate sensitivities projected global mean temperature and sea
level increase less in the proposed rule scenario than in the reference (no climate policy) case,
and the ocean does not become as acidic as it does in the reference case. The benefits of GHG
emissions reductions can be characterized both qualitatively and quantitatively, some of
which can be monetized (see Chapter 7). There are substantial uncertainties  in modeling the
                                         6-52

-------
                                                2017 Draft Regulatory Impact Analysis
global risks of climate change, which complicates quantification and cost-benefits
assessments. Changes in climate variables are a meaningful proxy for changes in the risk of
most potential impacts—including those that can be monetized, and those that have not been
monetized but can be quantified in physical terms (e.g., water availability), as well as those
that have not yet been quantified or are extremely difficult to quantify (e.g., forest disturbance
and catastrophic events such as collapse of large ice sheets and subsequent sea level rise).

   Table 6.4-2 Impact of GHG Emissions Reductions On Projected Changes in Global
Climate Associated with the Proposal (based on a range of climate sensitivities from 1.5-
                                        6°C)
VARIABLE
Atmospheric CC>2 Concentration
Global Mean Surface Temperature
Sea Level Rise
Ocean pH
UNITS
ppmv
°C
cm
pH units
YEAR
2100
2100
2100
2100
PROJECTED CHANGE
-3. 29 to -3. 68
-0.0076 to -0.0184
-0.074 to -0.166
+0.0018a
' The value for projected change in ocean pH is based on a climate sensitivity of 3.0.
                                         6-53

-------
Chapter 6
                                               References

46U.S. EPA (2009). Integrated Science Assessment for Paniculate Matter (Final Report). U.S. Environmental
Protection Agency, Washington, DC, EPA/600/R-08/139F, 2009.  Docket EPA-HQ-OAR-2010-0799.

47 U.S. EPA (2009). Integrated Science Assessment for Paniculate Matter (Final Report). U.S. Environmental
Protection Agency, Washington, DC, EPA/600/R-08/139F, 2009. Section 2.3.1.1. Docket EPA-HQ-OAR-2010-
0799.

48 U.S. EPA (2009). Integrated Science Assessment for Paniculate Matter (Final Report). U.S. Environmental
Protection Agency, Washington, DC, EPA/600/R-08/139F, 2009. Section 2.3.1.2. Docket EPA-HQ-OAR-2010-
0799.

49 U.S. EPA (2009). Integrated Science Assessment for Paniculate Matter (Final Report). U.S. Environmental
Protection Agency, Washington, DC, EPA/600/R-08/139F, 2009. Section 2.3.4. Docket EPA-HQ-OAR-2010-
0799.

50 U.S. EPA (2009). Integrated Science Assessment for Paniculate Matter (Final Report). U.S. Environmental
Protection Agency, Washington, DC, EPA/600/R-08/139F, 2009. Table 2-6. Docket EPA-HQ-OAR-2010-0799.

51 U.S. EPA (2009). Integrated Science Assessment for Paniculate Matter (Final Report). U.S. Environmental
Protection Agency, Washington, DC, EPA/600/R-08/139F, 2009. Section 2.3.5.1. Docket EPA-HQ-OAR-2010-
0799.

52 U.S. EPA (2009). Integrated Science Assessment for Paniculate Matter (Final Report). U.S. Environmental
Protection Agency, Washington, DC, EPA/600/R-08/139F, 2009. Table 2-6. Docket EPA-HQ-OAR-2010-0799.

53 U.S. EPA. (2006;. Air Quality Criteria for Ozone and Related Photochemical Oxidants (Final). EPA/600/R-
05/004aF-cF. Washington, DC: U.S. EPA. Docket EPA-HQ-OAR-2010-0799.

54U.S. EPA. (2007/ Review of the National Ambient Air Quality Standards for Ozone: Policy Assessment of
Scientific and Technical Information, OAQPS Staff Paper. EPA-452/R-07-003. Washington, DC, U.S. EPA.
Docket EPA-HQ-OAR-2010-0799

55 National Research Council (NRC), 2008. Estimating Mortality Risk Reduction and Economic Benefits from
Controlling Ozone Air Pollution.  The National Academies Press: Washington, D.C. Docket EPA-HQ-OAR-
2010-0799.

56Bates, D.V., Baker-Anderson, M., Sizto, R. (1990). Asthma attack periodicity: a study of hospital emergency
visits in Vancouver. Environ. Res., 57,51-70. Docket EPA-HQ-OAR-2010-0799.

57 Thurston, G.D., Ito, K., Kinney, P.L., Lippmann, M. (1992). A multi-year study of air pollution and
respiratory hospital admissions in three New York State metropolitan areas:  results for 1988 and 1989 summers.
J. Exposure Anal. Environ. Epidemiol, 2,429-450. Docket EPA-HQ-OAR-2010-0799.

58 Thurston, G.D., Ito, K., Hayes, C.G., Bates, D.V., Lippmann,  M. (1994) Respiratory hospital admissions and
summertime haze air pollution in Toronto, Ontario: consideration of the role of acid aerosols. Environ. Res., 65,
271-290. Docket EPA-HQ-OAR-2010-0799.

59Lipfert, F.W., Hammerstrom, T. (1992). Temporal patterns in air pollution and hospital admissions. Environ.
Res., 59,374-399. Docket EPA-HQ-OAR-2010-0799.

60Burnett, R.T., Dales, R.E., Raizenne, M.E., Krewski, D., Summers, P.W., Roberts, G.R., Raad-Young, M.,
Dann,T., Brook, J. (1994). Effects of low ambient levels of ozone and sulfates on the frequency of respiratory
admissions to Ontario hospitals. Environ. Res., 65, 172-194. Docket EPA-HQ-OAR-2010-0799.

61 U.S. EPA. (2006/ Air Quality Criteria for Ozone and Related Photochemical Oxidants (Final). EPA/600/R-
05/004aF-cF. Washington, DC: U.S. EPA. Docket EPA-HQ-OAR-2010-0799

62 U.S. EPA. (2006). Air Quality Criteria for Ozone and Related Photochemical Oxidants (Final). EPA/600/R-
05/004aF-cF. Washington, DC: U.S. EPA. Docket EPA-HQ-OAR-2010-0799.
                                               6-54

-------
                                                       2017 Draft Regulatory Impact Analysis
63Devlin, R. B., McDonnell, W. F., Mann, R., Becker, S., House, D. E., Schreinemachers, D., Koren, H. S.
(1991). Exposure of humans to ambient levels of ozone for 6.6 hours causes cellular and biochemical changes in
the lung. Am. J. Respir. Cell Mo 1. Biol, 4,  72-81.

64Koren, H. S., Devlin, R. B., Becker, S., Perez, R., McDonnell, W. F. (1991). Time-dependent changes of
markers associated with inflammation in the lungs of humans exposed to ambient levels of ozone. Toxicol.
Pathol., 19, 406-411. Docket EPA-HQ-OAR-2010-0799.

65Koren, H. S., Devlin, R. B., Graham, D. E., Mann, R., McGee, M. P., Horstman, D. H., Kozumbo, W. J.,
Becker, S., House, D. E., McDonnell, W. F., Bromberg, P. A. (1989). Ozone-induced inflammation in the lower
airways of human subjects. Am. Rev. Respir. Dis., 39, 407-415. Docket EPA-HQ-OAR-2010-0799.

66 Schelegle, E.S., Siefkin, A.D., McDonald, R.J. (1991).  Time course of ozone-induced neutrophilia in normal
humans. Am. Rev. Respir. Dis., 143,1353-135$. Docket EPA-HQ-OAR-2010-0799.

67 U.S. EPA. (1996). Air Quality Criteria for Ozone and Related Photochemical Oxidants. EPA600-P-93-004aF.
Washington. D.C.: U.S. EPA. Docket EPA-HQ-OAR-2010-0799.

68Hodgkin, J.E., Abbey, D.E., Euler, G.L., Magie, A.R. (1984). COPD prevalence in nonsmokers in high and
low photochemical air pollution areas. Chest, 86, 830-838. Docket EPA-HQ-OAR-2010-0799.

69Euler, G.L., Abbey, D.E., Hodgkin, J.E., Magie, A.R. (1988).  Chronic obstructive pulmonary disease
symptom effects of long-term cumulative exposure to ambient levels of total oxidants and nitrogen dioxide in
California Seventh-day Adventist residents. Arch. Environ. Health, 43, 279-285. Docket EPA-HQ-OAR-2010-
0799.

70 Abbey, D.E., Petersen, F., Mills, P.K., Beeson, W.L. (1993).  Long-term ambient concentrations of total
suspended particulates, ozone, and sulfur dioxide and respiratory symptoms in a nonsmoking population. Arch.
Environ. Health, 48, 33-46. Docket EPA-HQ-OAR-2010-0799.

71 U.S. EPA. (2006). Air Quality Criteria for Ozone and Related Photochemical Oxidants (Final). EPA/600/R-
05/004aF-cF. Washington, DC: U.S. EPA. Docket EPA-HQ-OAR-2010-0799.

72 U.S. EPA. (2006). Air Quality Criteria for Ozone and Related Photochemical Oxidants (Final). EPA/600/R-
05/004aF-cF. Washington, DC: U.S. EPA. Docket EPA-HQ-OAR-2010-0799.

73 Avol, E.L., Trim, S. C., Little, D.E., Spier, C.E., Smith, M. N., Peng, R.-C., Linn, W.S., Hackney,  J.D., Gross,
K.B., D'Arcy, J.B., Gibbons, D., Higgins, I.T.T. (1990 June). Ozone exposure and lung function in children
attending a southern California summer camp. Paper no. 90-150.3. Paper presented at the 83rd annual meeting
and exhibition of the Air &  Waste Management Association, Pittsburgh, PA. Docket EPA-HQ-OAR-2010-0799.

74Higgins, I. T.T., D'Arcy, J. B., Gibbons, D. I., Avol, E. L., Gross, K.B. (1990). Effect of exposures to ambient
ozone on ventilatory lung function in children. Am. Rev. Respir. Dis., 141, 1136-1146. Docket EPA-HQ-OAR-
2010-0799.

75Raizenne, M.E., Burnett, R.T., Stern, B., Franklin, C.A., Spengler, J.D. (1989) Acute lung function responses
to ambient acid aerosol exposures in children. Environ. Health Perspect, 79,179-185. Docket EPA-HQ-OAR-
2010-0799.

76Raizenne, M.; Stern, B.; Burnett, R.; Spengler, J. (1987 June) Acute respiratory function and transported air
pollutants: observational studies. Paper no. 87-32.6. Paper presented at the 80th annual meeting of the Air
Pollution Control Association, New York, NY. Docket EPA-HQ-OAR-2010-0799.

77 Spektor, D. M., Lippmann, M. (1991). Health effects of ambient ozone on healthy children at a summer camp.
In: Berglund, R. L.; Lawson, D. R.; McKee, D. J., eds. Tropospheric ozone and the environment: papers from an
international conference', March 1990; Los Angeles, CA.  Pittsburgh, PA: Air & Waste Management
Association; pp. 83-89.  (A&WMA transaction series no.  TR-19). Docket EPA-HQ-OAR-2010-0799.

78 Spektor, D. M., Thurston, G.D., Mao, J., He, D., Hayes, C., Lippmann, M. (1991). Effects of single- and
multiday ozone exposures on respiratory function in active normal children. Environ. Res, 55,107-122. Docket
EPA-HQ-OAR-2010-0799.
                                               6-55

-------
Chapter 6


79 Spektor, D. M., Lippman, M., Lioy, P. J., Thurston, G. D., Citak, K., James, D. J., Bock, N., Speizer, F. E.,
Hayes, C. (1988). Effects of ambient ozone on respiratory function in active, normal children. Am. Rev. Respir.
Dis., 137, 313-320. Docket EPA-HQ-OAR-2010-0799.

80 U.S. EPA. (2006). Air Quality Criteria for Ozone and Related Photochemical Oxidants (Final). EPA/600/R-
05/004aF-cF. Washington, DC: U.S. EPA. Docket EPA-HQ-OAR-2010-0799.

81 Hazucha,  M. J., Folinsbee, L. J., Seal, E., Jr. (1992). Effects of steady-state and variable ozone concentration
profiles on pulmonary function. Am. Rev. Respir. Dis., 146, 1487-1493. Docket EPA-HQ-OAR-2010-0799.

82Horstman, D.H., Ball, B.A., Folinsbee, L.J., Brown, J., Gerrity, T. (1995) Comparison of pulmonary responses
of asthmatic and nonasthmatic subjects performing light exercise while exposed to a low level of ozone.
Toxicol. Ind. Health., 11(4), 369-85.

83Horstman, D.H.,; Folinsbee, L.J., Ives, P.J., Abdul-Salaam, S., McDonnell, W.F. (1990). Ozone concentration
and pulmonary response relationships for 6.6-hour exposures with five hours of moderate exercise to 0.08, 0.10,
and 0.12 ppm.Am. Rev. Respir. Dis., 142, 1158-1163. Docket EPA-HQ-OAR-2010-0799.

84 U. S. EPA (2008). Integrated Science Assessment (ISA) for Sulfur Oxides - Health Criteria (Final Report).
EPA/600/R-08/047F. Washington, DC,: U.S.EPA. Retrieved on March 19, 2009 from
http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm?deid= 198843. Docket EPA-HQ-OAR-2010-0799.

85 U.S. EPA (2008). Integrated Science Assessment for Oxides of Nitrogen -Health Criteria (Final Report).
EPA/600/R-08/071. Washington, DC,: U.S.EPA. Retrieved on March 19, 2009 from
http://cfpub.epa.gov/ncea/cfm/recordisplay.cfm?deid= 194645. Docket EPA-HQ-OAR-2010-0799.

86 U.S. EPA, 2010. Integrated Science Assessment for Carbon Monoxide (Final Report). U.S. Environmental
Protection Agency, Washington, DC, EPA/600/R-09/019F, 2010.
http://cfpub.epa.gov/ncea/cfrn/recordisplay.cfm?deid=218686. Docket EPA-HQ-OAR-2010-0799.

87 U.S. EPA. (2011) Summary of Results  for the 2005 National-Scale Assessment.
www.epa.gov/ttn/atw/nata2005/05pdf/sum_results.pdf. Docket EPA-HQ-OAR-2010-0799.

88 U.S. EPA (2010). Final Rulemaking to Establish Light-Duty Vehicle Greenhouse Gas Emission Standards and
Corporate Average Fuel Economy Standards: Regulatory Impact Analysis. Chapter 7, section 7.2.2.3.2. This
material is available electronically at http://www.epa.gov/oms/climate/regulations/420rl0009.pdf. Docket EPA-
HQ-OAR-2010-0799.

89 U.S. EPA (2011) 2005 National-Scale Air Toxics Assessment, http://www.epa.gov/ttn/atw/nata2005. Docket
EPA-HQ-OAR-2010-0799.

90U.S. EPA. 2000. Integrated Risk Information System File for Benzene. This material is available
electronically at: http://www.epa.gov/iris/subst/0276.htm. Docket EPA-HQ-OAR-2010-0799.

91 International Agency for Research on Cancer, IARC monographs on the evaluation of carcinogenic risk of
chemicals to humans, Volume 29, Some industrial chemicals and dyestuffs, International Agency for Research
on Cancer, World Health Organization, Lyon, France 1982. Docket EPA-HQ-OAR-2010-0799.

92 Irons, R.D.; Stillman, W.S.; Colagiovanni, D.B.; Henry, V.A. (1992) Synergistic action of the benzene
metabolite hydroquinone on myelopoietic stimulating activity of granulocyte/macrophage colony-stimulating
factor in vitro, Proc. Natl. Acad. Sci. 89:3691-3695. Docket EPA-HQ-OAR-2010-0799.

93 International Agency for Research on Cancer (IARC). 1987. Monographs on the evaluation of carcinogenic
risk of chemicals to humans, Volume 29, Supplement 7, Some industrial chemicals and dyestuffs, World Health
Organization, Lyon, France. Docket EPA-HQ-OAR-2010-0799.

94 U.S. Department of Health and Human Services National Toxicology Program 11th Report on Carcinogens
available at: http://ntp.niehs.nih.gov/go/16183.

95Aksoy, M. (1989). Hematotoxicity and carcinogenicity of benzene. Environ. Health Perspect. 82:193-197.
EPA-HQ-OAR-2010-0799
                                               6-56

-------
                                                      2017 Draft Regulatory Impact Analysis
96 Goldstein, B.D. (1988).  Benzene toxicity.  Occupational medicine. State of the Art Reviews.  3:541-554.
Docket EPA-HQ-OAR-2010-0799.

97Rothman, N., G.L. Li, M. Dosemeci, W.E. Bechtold, G.E. Marti, Y.Z. Wang, M. Linet, L.Q. Xi, W. Lu, M.T.
Smith, N. Titenko-Holland, L.P. Zhang, W. Blot, S.N. Yin, andR.B. Hayes (1996) Hematotoxicity among
Chinese workers heavily exposed to benzene. Am. J. Ind. Med. 29: 236-246.  Docket EPA-HQ-OAR-2010-0799.

98 U. S. EPA 2002 Toxicological Review of Benzene (Noncancer Effects). Environmental Protection Agency,
Integrated Risk Information System (IRIS), Research and Development, National Center for Environmental
Assessment, Washington DC. This material is available electronically at http://www.epa.gov/iris/subst/0276.htm.
Docket EPA-HQ-OAR-2010-0799.

99Qu, O.; Shore, R.; Li, G.; Jin, X.; Chen, C.L.; Cohen, B.; Melikian, A.; Eastmond, D.; Rappaport, S.; Li, H.;
Rupa, D.; Suramaya, R.; Songnian, W.;  Huifant,  Y.; Meng, M.; Winnik, M.; Kwok, E.; Li, Y.; Mu, R.; Xu,
B.; Zhang, X.; Li, K. (2003).  HEI Report 115, Validation & Evaluation of Biomarkers in Workers Exposed to
Benzene in China. Docket EPA-HQ-OAR-2010-0799.

100 Qu, Q., R. Shore, G. Li, X. Jin, L.C. Chen, B. Cohen, et al. (2002).  Hematological changes among Chinese
workers with a broad range of benzene exposures. Am. J. Industr. Med. 42: 275-285. Docket EPA-HQ-OAR-
2010-0799.

101 Lan, Qing, Zhang, L., Li, G., Vermeulen, R., et al. (2004). Hematotoxically in Workers Exposed to Low
Levels of Benzene.  Science 306: 1774-1776. Docket EPA-HQ-OAR-2010-0799.

102Turtletaub, K.W. and Mani, C. (2003). Benzene metabolism in rodents at doses relevant to human exposure
from Urban Air. Research Reports Health Effect Inst. Report No. 113. Docket EPA-HQ-OAR-2010-0799.

103 U.S. EPA. 2002.  Health Assessment of 1,3-Butadiene. Office of Research and Development, National Center
for Environmental Assessment, Washington Office, Washington, DC.  Report No. EPA600-P-98-00IF. This
document is available electronically at http://www.epa.gov/iris/supdocs/buta-sup.pdf. Docket EPA-HQ-OAR-
2010-0799.

104U.S. EPA. 2002 "Full IRIS Summary for 1,3-butadiene (CASRN 106-99-0)" Environmental Protection
Agency, Integrated Risk Information System (IRIS), Research and Development, National Center for
Environmental Assessment, Washington, DC http://www.epa.gov/iris/subst/0139.htm. Docket EPA-HQ-OAR-
2010-0799.

105 International Agency for Research on Cancer (IARC) (1999) Monographs on the evaluation of carcinogenic
risk of chemicals to humans, Volume 71, Re-evaluation of some organic chemicals,  hydrazine and hydrogen
peroxide and Volume 97 (in preparation), World Health Organization, Lyon, France. Docket EPA-HQ-OAR-
2010-0799.

   International Agency for Research on Cancer (IARC) (2008) Monographs on the evaluation of carcinogenic
risk of chemicals to humans, 1,3-Butadiene, Ethylene Oxide and Vinyl Halides (Vinyl Fluoride, Vinyl Chloride
and Vinyl Bromide) Volume 97, World Health Organization, Lyon, France. Docket EPA-HQ-OAR-2010-0799.

107 U.S. Department of Health and Human Services National Toxicology Program 11th Report on Carcinogens
available at: http://ntp.niehs.nih.gov/go/16183.

108Bevan, C.; Stadler, J.C.; Elliot, G.S.; etal. (1996) Subchronic toxicity of 4-vinylcyclohexene in rats and mice
by inhalation. Fundam. Appl. Toxicol. 32:1-10. Docket EPA-HQ-OAR-2010-0799.

109U.S. EPA. 1987.  Assessment of Health Risks to Garment Workers and CertainHome Residents from
Exposure to Formaldehyde, Office of Pesticides and Toxic Substances, April 1987. Docket EPA-HQ-OAR-
2010-0799.

110Hauptmann, M..; Lubin, J. H.; Stewart, P. A.; Hayes, R. B.; Blair, A. 2003. Mortality from
lymphohematopoetic malignancies among workers in formaldehyde industries. Journal of the National Cancer
Institute 95: 1615-1623. Docket EPA-HQ-OAR-2010-0799.
111
   Hauptmann, M..; Lubin, J. H.; Stewart, P. A.; Hayes, R. B.; Blair, A.  2004. Mortality from solid cancers
among workers in formaldehyde industries. American Journal of Epidemiology 159: 1117-1130. Docket EPA-
HQ-OAR-2010-0799.
                                              6-57

-------
Chapter 6


112 Beane Freeman, L. E.; Blair, A.; Lubin, J. H.; Stewart, P. A.; Hayes, R. B.; Hoover, R. N.; Hauptmann, M.
2009. Mortality from lymphohematopoietic malignancies among workers in formaldehyde industries: The
National Cancer Institute cohort. J. National Cancer Inst. 101: 751-761. Docket EPA-HQ-OAR-2010-0799.

113 Pinkerton, L. E. 2004. Mortality among a cohort of garment workers exposed to formaldehyde: an update.
Occup. Environ. Med. 61: 193-200. Docket EPA-HQ-OAR-2010-0799.

114 Coggon, D, EC Harris, J Poole, KT Palmer. 2003. Extended follow-up of a cohort of British chemical
workers exposed to formaldehyde. J National Cancer Inst. 95:1608-1615. Docket EPA-HQ-OAR-2010-0799.

115 Conolly, RB, JS Kimbell, D Janszen, PM Schlosser, D Kalisak, J Preston, and FJ Miller. 2003. Biologically
motivated computational modeling of formaldehyde carcinogenicity in the F344 rat. Tox Sci 75: 432-447.
Docket EPA-HQ-OAR-2010-0799.

116 Conolly, RB, JS Kimbell, D Janszen, PM Schlosser, D Kalisak, J Preston, and FJ Miller. 2004. Human
respiratory tract cancer risks of inhaled formaldehyde: Dose-response predictions derived from biologically-
motivated computational modeling of a combined rodent and human dataset. Tox Sci 82: 279-296. Docket EPA-
HQ-OAR-2010-0799.

117 Chemical Industry Institute of Toxicology (CUT). 1999. Formaldehyde: Hazard characterization and dose-
response assessment for carcinogenicity by the route of inhalation.  CUT, September 28, 1999. Research
Triangle Park, NC. Docket EPA-HQ-OAR-2010-0799.

118 U.S. EPA.  Analysis of the Sensitivity and Uncertainty in 2-Stage Clonal Growth Models for Formaldehyde
with Relevance to Other Biologically-Based Dose Response (BBDR) Models. U.S. Environmental Protection
Agency, Washington, D.C., EPA/600/R-08/103, 2008. Docket EPA-HQ-OAR-2010-0799.

119 Subramaniam, R; Chen, C; Crump, K; .et  .al. (2008) Uncertainties in biologically-based modeling of
formaldehyde-induced cancer risk: identification of key issues. Risk Anal 28(4):907-923. Docket EPA-HQ-
OAR-2010-0799.

120 Subramaniam RP; Crump KS; Van Landingham C; et. al. (2007) Uncertainties  in the CUT model for
formaldehyde-induced carcinogenicity in the rat: A limited sensitivity analysis-I. Risk Anal, 27: 1237-1254.
Docket EPA-HQ-OAR-2010-0799.

121 Crump, K; Chen,  C; Fox, J; .et .al. (2008) Sensitivity analysis of biologically motivated model for
formaldehyde-induced respiratory cancer in humans. Ann Occup Hyg 52:481-495. Docket EPA-HQ-OAR-2010-
0799.

122 Crump, K; Chen,  C; Fox, J; .et .al. (2008) Sensitivity analysis of biologically motivated model for
formaldehyde-induced respiratory cancer in humans. Ann Occup Hyg 52:481-495. Docket EPA-HQ-OAR-2010-
0799.

123 Subramaniam RP; Crump KS; Van Landingham C; et. al. (2007) Uncertainties  in the CUT model for
formaldehyde-induced carcinogenicity in the rat: A limited sensitivity analysis-I. Risk Anal, 27: 1237-1254.
Docket EPA-HQ-OAR-2010-0799.

124 International Agency for Research on Cancer (2006) Formaldehyde, 2-Butoxyethanol and 1-tert-
Butoxypropan-2-ol.  Monographs Volume 88. World Health Organization, Lyon, France. Docket EPA-HQ-
OAR-2010-0799.

125 Agency for Toxic Substances and Disease Registry (ATSDR). 1999. Toxicological profile for Formaldehyde.
Atlanta, GA: U.S. Department of Health and Human Services, Public Health Service.
http://www.atsdr.cdc.gov/toxprofiles/tpl 1 l.html. Docket EPA-HQ-OAR-2010-0799.

126 WHO (2002) Concise International Chemical Assessment Document 40: Formaldehyde.  Published under the
joint sponsorship of the United Nations Environment Programme, the International Labour Organization, and the
World Health Organization, and produced within the framework of the Inter-Organization Programme for the
Sound Management of Chemicals. Geneva. Docket EPA-HQ-OAR-2010-0799.

127U.S. EPA (1988). Integrated Risk Information System File of Acetaldehyde. Research and Development,
National Center for Environmental Assessment, Washington, DC. This material is available electronically at
http://www.epa.gov/iris/subst/0290.htm. Docket EPA-HQ-OAR-2010-0799.
                                               6-58

-------
                                                  2017 Draft Regulatory Impact Analysis
128 U.S. Department of Health and Human Services National Toxicology Program 11th Report on Carcinogens
available at: http://ntp.niehs.nih.gov/go/16183.
129 International Agency for Research on Cancer (IARC). 1999. Re-evaluation of some organic chemicals,
hydrazine, and hydrogen peroxide. IARC Monographs on the Evaluation of Carcinogenic Risk of Chemical to
Humans, Vol 71. Lyon, France.
130 U.S. EPA (1988). Integrated Risk Information System File of Acetaldehyde. This material is available
electronically at http://www.epa.gov/iris/subst/0290.htm. Docket EPA-HQ-OAR-2010-0799.
131 U.S. EPA. (2003). Integrated Risk Information System File of Acrolein. Research and Development,
National Center for Environmental Assessment, Washington, DC. This material is available electronically at
http://www.epa.gov/iris/subst/0364.htm. Docket EPA-HQ-OAR-2010-0799.
132 Appleman, L.M., R. A. Woutersen, and V. J. Feron. (1982). Inhalation toxicity of acetaldehyde in rats. I. Acute
and subacute studies. Toxicology. 23: 293-297. Docket EPA-HQ-OAR-2010-0799.
133 Myou, S.; Fujimura, M; Nishi K.; Ohka, T.; and Matsuda, T.  (1993) Aerosolized acetaldehyde induces
histamine-mediated bronchoconstriction in asthmatics. Am. Rev. Respir.Dis. 148(4 Pt 1): 940-943. Docket EPA-
HQ-OAR-2010-0799.
134 U.S. EPA. (2003) Toxicological review of acrolein in support of summary information on
Integrated Risk Information System (IRIS) National Center for Environmental Assessment,
Washington, DC. EPA/635/R-03/003.  p. 10.  Available online at:
http://www.epa.gov/ncea/iris/toxreviews/0364tr.pdf. Docket EPA-HQ-OAR-2010-0799.
135 U.S. EPA. (2003) Toxicological review of acrolein in support of summary information on
Integrated Risk Information System (IRIS) National Center for Environmental Assessment,
Washington, DC. EPA/635/R-03/003.  Available online at:
http://www.epa.gov/ncea/iris/toxreviews/0364tr.pdf. Docket EPA-HQ-OAR-2010-0799.
136 U.S. EPA. (2003) Toxicological review of acrolein in support of summary information on
Integrated Risk Information System (IRIS) National Center for Environmental Assessment,
Washington, DC. EPA/635/R-03/003.  p. 11.  Available online at:
http://www.epa.gov/ncea/iris/toxreviews/0364tr.pdf. Docket EPA-HQ-OAR-2010-0799.
137 U.S. EPA. (2003). Integrated Risk Information  System File of Acrolein.  Office of Research
and Development, National Center for Environmental Assessment, Washington, DC. This
material is available at http://www.epa.gov/iris/subst/0364.htm. Docket EPA-HQ-OAR-2010-
0799.
138 U.S. EPA. (2003) Toxicological review of acrolein in support of summary information on
Integrated Risk Information System (IRIS) National Center for Environmental Assessment,
Washington, DC. EPA/635/R-03/003.  p. 15.  Available online at:
http://www.epa.gov/ncea/iris/toxreviews/0364tr.pdf. Docket EPA-HQ-OAR-2010-0799.
139 Morris JB, Symanowicz PT, Olsen JE, et al. 2003. Immediate sensory nerve-mediated
respiratory responses to irritants in healthy and allergic airway-diseased mice. J Appl Physiol
94(4):1563-1571. Docket EPA-HQ-OAR-2010-0799.
140 U.S. EPA. (2003). Integrated Risk Information System File of Acrolein. Research and
Development, National Center for Environmental Assessment, Washington, DC. This
material is available at http://www.epa.gov/iris/subst/0364.htm. Docket EPA-HQ-OAR-2010-0799.
141 International Agency for Research on Cancer (IARC). 1995. Monographs  on the evaluation
of carcinogenic risk of chemicals to humans, Volume 63. Dry cleaning, some chlorinated
solvents and other industrial chemicals, World Health Organization, Lyon, France. Docket
EPA-HQ-OAR-2010-0799.
                                          6-59

-------
Chapter 6



142 U.S. EPA (2002). Health Assessment Document for Diesel Engine Exhaust. EPA/600/8-
90/057F Office of Research and Development, Washington DC.
http://cfpub.epa.gov/ncea/cfm/recordisplay.cfm?deid=29060.

143 U.S. EPA (1997). Integrated Risk Information System File of indeno(l,2,3-cd)pyrene. Research and
Development, National Center for Environmental Assessment, Washington, DC. This material is available
electronically at http://www.epa.gov/ncea/iris/subst/0457.htm.
144 Perera, P.P.; Rauh, V.; Tsai, W-Y.; et al. (2002) Effect of transplacental exposure to environmental pollutants
on birth outcomes in a multiethnic population. Environ Health Perspect. Ill: 201 -205.
145 Perera, P.P.; Rauh, V.; Whyatt, R.M.; Tsai, W.Y.; Tang, D.; Diaz, D.; Hoepner, L.; Barr, D.; Tu, Y.H.;
Camann, D.; Kinney, P. (2006) Effect of prenatal exposure to airborne polycyclic aromatic hydrocarbons on
neurodevelopment in the first 3 years of life among inner-city children. Environ Health Perspect 114: 1287-1292.

146U. S. EPA.  1998. Toxicological Review of Naphthalene (Reassessment of the Inhalation Cancer Risk),
Environmental Protection Agency, Integrated Risk Information System, Research and Development, National
Center for Environmental Assessment, Washington, DC.  This material is available electronically at
http://www.epa.gov/iris/subst/0436.htm. Docket EPA-HQ-OAR-2010-0799.

147 Oak Ridge Institute for Science and Education.  (2004). External Peer Review for the IRIS Reassessment of
the Inhalation Carcinogenicity of Naphthalene. August 2004.
http://cfpub.epa.gov/ncea/cfm/recordisplay.cfm?deid=84403. Docket EPA-HQ-OAR-2010-0799.

148 National Toxicology Program (NTP). (2004). 11th Report on Carcinogens. Public Health Service, U.S.
Department of Health and Human Services, Research Triangle Park, NC. Available from: http://ntp-
server.niehs.nih.gov.

149 International Agency for Research on Cancer (IARC). (2002). Monographs on the Evaluation of the
Carcinogenic Risk of Chemicals for Humans. Vol. 82. Lyon, France. Docket EPA-HQ-OAR-2010-0799.

150U. S. EPA. 1998. Toxicological Review of Naphthalene, Environmental Protection Agency, Integrated Risk
Information System, Research and Development, National Center for Environmental Assessment, Washington,
DC. This material is available electronically at http://www.epa.gov/iris/subst/0436.htm

151 Zhou, Y.; Levy, J.I.  (2007) Factors influencing the spatial extent of mobile source air pollution impacts: a
meta-analysis.  BMC Public Health 7: 89. doi:10.1186/1471-2458-7-89. Docket EPA-HQ-OAR-2010-0799.

152 HEI Panel on the Health Effects of Air Pollution. (2010) Traffic-related air pollution: a critical review of the
literature on emissions, exposure, and health effects.  [Online at www.healtheffects.org1. Docket EPA-HQ-OAR-
2010-0799.

153 Salam, M.T.; Islam, T.; Gilliland, F.D. (2008) Recent evidence for adverse effects of residential proximity to
traffic sources on asthma. Current Opin Pulm Med 14: 3-8. Docket EPA-HQ-OAR-2010-0799.

154 Holguin, F.  (2008) Traffic, outdoor air pollution, and asthma. Immunol Allergy Clinics North Am 28: 577-
588.

155 Adar, S.D.; Kaufman, J.D. (2007) Cardiovascular disease and air pollutants: evaluating and improving
epidemiological data implicating traffic exposure.  Inhal Toxicol 19:  135-149. Docket EPA-HQ-OAR-2010-
0799.

156 Raaschou-Nielsen, O.; Reynolds, P. (2006) Air pollution and childhood cancer: a review of the
epidemiological literature. Int J Cancer 118: 2920-2929. Docket EPA-HQ-OAR-2010-0799.

157 U.S. Census Bureau (2008) American Housing Survey for the United States in 2007. Series H-150 (National
Data), Table 1A-6.  [Accessed at http://www.census.gov/hhes/www/housing/ahs/ahs07/ahs07.html on January
22, 2009]

158 Lena, T.S.; Ochieng, V.; Carter, M.; Holguin-Veras, J.; Kinney, P.L. (2002) Elemental carbon and PM2.5
levels in an urban community heavily impacted by truck traffic. Environ Health Perspect 110:  1009-1015.
Docket EPA-HQ-OAR-2010-0799.
                                               6-60

-------
                                                       2017 Draft Regulatory Impact Analysis
159 Wier, M; Sciammas, C.; Seto, E.; Bhatia, R.; Rivard, T. (2009) Health, traffic, and environmental justice:
collaborative research and community action in San Francisco, California. Am J Public Health 99: S499-S504.
Docket EPA-HQ-OAR-2010-0799.

160Forkenbrock, D.J. and L.A. Schweitzer. Environmental Justice and Transportation Investment Policy. Iowa
City: University of Iowa, 1997.

161 Appatova, A.S.; Ryan, P.H.; LeMasters, O.K.; Grinshpun, S.A. (2008) Proximal exposure of public schools
and students to major roadways: a nationwide US survey.  J Environ Plan Mgmt. Docket EPA-HQ-OAR-2010-
0799.

162 Green, R.S.; Smorodinsky, S.; Kim, J.J.; McLaughlin, R.; Ostro, B. (2004) Proximity of California public
schools to busy roads. Environ Health Perspect 112: 61-66. Docket EPA-HQ-OAR-2010-0799.

163 Houston, D.; Ong, P.; Wu, J.; Winer, A. (2006) Proximity of licensed child care facilities to near-roadway
vehicle pollution.  Am J Public Health 96:  1611-1617. Docket EPA-HQ-OAR-2010-0799.

164 Wu, Y.; Batterman, S. (2006) Proximity of schools in Detroit, Michigan to automobile and truck traffic.  J
Exposure Sci Environ Epidemiol 16: 457-470. Docket EPA-HQ-OAR-2010-0799.

165 U.S. EPA (2009). Integrated Science Assessment for Paniculate Matter (Final Report). U.S. Environmental
Protection Agency, Washington, DC, EPA/600/R-08/139F, 2009. pg 9-19 through 9-23. Docket EPA-HQ-OAR-
2010-0799.

166U.S. EPA. 1999. The Benefits and Costs of the Clean Air Act, 1990-2010. Prepared for U.S. Congress by
U.S. EPA, Office of Air and Radiation, Office of Policy Analysis and Review, Washington, DC, November;
EPA report no. EPA410-R-99-001.  Docket EPA-HQ-OAR-2010-0799.

167 U.S. EPA (2006). Air Quality Criteria for Ozone and Related Photochemical Oxidants (Final). U.S. EPA,
Washington, DC, EPA/600/R-05/004aF-cF, 2006. Docket EPA-HQ-OAR-2010-0799.

168 Winner, W.E., and C.J. Atkinson. 1986. "Absorption of air pollution by plants, and consequences for growth."
Trends in Ecology and Evolution 7:15-18.  Docket EPA-HQ-OAR-2010-0799.

169U.S. EPA (2006). Air Quality Criteria for Ozone and Related Photochemical Oxidants (Final). U.S. EPA,
Washington, DC, EPA/600/R-05/004aF-cF, 2006. Docket EPA-HQ-OAR-2010-0799.

170 Tingey, D.T., and Taylor,  G.E. (1982) Variation in plant response to ozone: a conceptual model of
physiological events. In M.H. Unsworth & D.P. Omrod (Eds.), Effects of Gaseous Air Pollution in Agriculture
and Horticulture, (pp.113-138). London, UK: Butterworth Scientific. Docket EPA-HQ-OAR-2010-0799.

171 U.S. EPA (2006). Air Quality Criteria for Ozone and Related Photochemical Oxidants (Final). U.S. EPA,
Washington, DC, EPA/600/R-05/004aF-cF, 2006. Docket EPA-HQ-OAR-2010-0799.

172 U.S. EPA (2006). Air Quality Criteria for Ozone and Related Photochemical Oxidants (Final). U.S. EPA,
Washington, DC, EPA/600/R-05/004aF-cF, 2006. Docket EPA-HQ-OAR-2010-0799.

173 U.S. EPA (2006). Air Quality Criteria for Ozone and Related Photochemical Oxidants (Final). U.S. EPA,
Washington, DC, EPA/600/R-05/004aF-cF, 2006. Docket EPA-HQ-OAR-2010-0799.

174 U.S. EPA (2006). Air Quality Criteria for Ozone and Related Photochemical Oxidants (Final). U.S. EPA,
Washington, DC, EPA/600/R-05/004aF-cF, 2006. Docket EPA-HQ-OAR-2010-0799.

175Ollinger, S.V., Aber, J.D., Reich, P.B. (1997). Simulating ozone effects on forest productivity: interactions
between leaf canopy  and stand level processes. Ecological Applications,  7, 1237-1251.  Docket EPA-HQ-OAR-
2010-0799.

176 Winner, W.E. (1994). Mechanistic analysis of plant responses to air pollution. Ecological Applications, 4(4),
651-661.  Docket EPA-HQ-OAR-2010-0799.

177 U.S. EPA (2006). Air Quality Criteria for Ozone and Related Photochemical Oxidants (Final). U.S. EPA,
Washington, DC, EPA/600/R-05/004aF-cF, 2006. Docket EPA-HQ-OAR-2010-0799.

178 Chappelka, A.H., Samuelson, L.J. (1998).  Ambient ozone effects on forest trees of the eastern United States:
a review. New Phytologist, 139, 91-108.  Docket EPA-HQ-OAR-2010-0799
                                               6-61

-------
Chapter 6
179U.S. EPA (2006). Air Quality Criteria for Ozone and Related Photochemical Oxidants (Final). U.S. EPA,
Washington, DC, EPA/600/R-05/004aF-cF, 2006. Docket EPA-HQ-OAR-2010-0799.

180 Fox, S., Mickler, R. A. (Eds.). (1996). Impact of Air Pollutants on Southern Pine Forests, Ecological Studies.
(Vol. 118, 513 pp.) New York: Springer-Verlag.

181 De Steiguer, I, Pye, I, Love, C. (1990). Air Pollution Damage to U.S. Forests. Journal of Forestry, 88(8),
17-22. Docket EPA-HQ-OAR-2010-0799.

182 Pye, J.M. (1988).  Impact of ozone on the growth and yield of trees: A review. Journal of Environmental
Quality, 77,347-360. Docket EPA-HQ-OAR-2010-0799.

183 U.S. EPA (2006). Air Quality Criteria for Ozone and Related Photochemical Oxidants (Final). U.S. EPA,
Washington, DC, EPA/600/R-05/004aF-cF, 2006. Docket EPA-HQ-OAR-2010-0799.

184 U.S. EPA (2006). Air Quality Criteria for Ozone and Related Photochemical Oxidants (Final). U.S. EPA,
Washington, DC, EPA/600/R-05/004aF-cF, 2006. Docket EPA-HQ-OAR-2010-0799.

185 McBride, J.R., Miller, P.R., Laven, R.D. (1985). Effects of oxidant air pollutants on forest succession in the
mixed conifer forest type of southern California.  In: Air Pollutants Effects On Forest Ecosystems, Symposium
Proceedings,  St. P, 1985, p. 157-167.  Docket EPA-HQ-OAR-2010-0799.

186 Miller, P.R., O.C. Taylor, R.G. Wilhour. 1982. Oxidant air pollution effects on a western coniferous forest
ecosystem. Corvallis, OR: U.S. Environmental Protection Agency, Environmental Research Laboratory
(EPA600-D-82-276). Docket EPA-HQ-OAR-2010-0799.

187 U.S. EPA (2006). Air Quality Criteria for Ozone and Related Photochemical Oxidants (Final). U.S. EPA,
Washington, DC, EPA/600/R-05/004aF-cF, 2006. Docket EPA-HQ-OAR-2010-0799.

188 Grulke, N.E. (2003). The physiological basis of ozone injury assessment attributes  in Sierran conifers. In A.
Bytnerowicz, M.J. Arbaugh, & R. Alonso  (Eds.), Ozone air pollution in the Sierra Nevada: Distribution and
effects on forests, (pp. 55-81). New York, NY: Elsevier Science, Ltd. Docket EPA-HQ-OAR-2010-0799.

189U.S. EPA (2006). Air Quality Criteria for Ozone and Related Photochemical Oxidants (Final). U.S. EPA,
Washington, DC, EPA/600/R-05/004aF-cF, 2006. Docket EPA-HQ-OAR-2010-0799.

190Kopp, R. I, Vaughn, W. J., Hazilla, M., Carson, R. (1985).  Implications of environmental policy for U.S.
agriculture: the case  of ambient ozone standards.  Journal of Environmental Management, 20, 321-331. Docket
EPA-HQ-OAR-2010-0799.

191 Adams, R. M., Hamilton,  S. A., McCarl, B. A. (1986). The benefits of pollution control: the case of ozone
and U.S.  agriculture. American Journal of Agricultural Economics, 34, 3-19. Docket EPA-HQ-OAR-2010-
0799.

192 Adams, R. M., Glyer, J. D., Johnson, S. L., McCarl, B. A. (1989). A reassessment of the economic effects of
ozone on U.S. agriculture. Journal of the Air Pollution Control Association, 39, 960-968. Docket EPA-HQ-
OAR-2010-0799.

193 Abt Associates, Inc.  1995.  Urban ornamental plants: sensitivity to ozone and potential economic losses.
U.S. EPA, Office of Air Quality Planning and Standards, Research Triangle Park. Under contract to RADIAN
Corporation, contract no. 68-D3-0033,  WAno. 6. pp. 9-10.  Docket EPA-HQ-OAR-2010-0799.

194 White, D., Kimerling, A.J., Overton, W.S. (1992). Cartographic and geometric component of a global
sampling design for environmental monitoring. Cartography and Geographic Information Systems, 19, 5-22.
Docket EPA-HQ-OAR-2010-0799.

195 Smith, G., Coulston, J., Jepsen, E., Prichard, T. (2003). A national ozone biomonitoring program—results
from field surveys of ozone sensitive plants in Northeastern forests (1994-2000).  Environmental Monitoring and
Assessment, 87,211-291. Docket EPA-HQ-OAR-2010-0799.

196 Coulston, J.W., Riitters, K.H., Smith, G.C. (2004). A preliminary assessment of the Montreal process indica-
tors of air pollution for the United States. Environmental Monitoring and Assessment,  95, 57-74. Docket EPA-
HQ-OAR-2010-0799.
                                               6-62

-------
                                                      2017 Draft Regulatory Impact Analysis
197 U.S. EPA. (2006). Air Quality Criteria for Ozone and Related Photochemical Oxidants. EPA/600/R-
05/004aF-cF. Docket EPA-HQ-OAR-2010-0799.

198 U.S. EPA (2006). Air Quality Criteria for Ozone and Related Photochemical Oxidants. EPA/600/R-
05/004aF-cF. Docket EPA-HQ-OAR-2010-0799.

199 Smith, G., Coulston, J., Jepsen, E.,  Prichard, T. (2003). A national ozone biomonitoring program—results
from field surveys of ozone sensitive plants in Northeastern forests (1994-2000). Environmental Monitoring and
Assessment, 87,211-291. Docket EPA-HQ-OAR-2010-0799.

200 U.S. EPA (2009). Integrated Science Assessment for Paniculate Matter (Final Report). U.S. Environmental
Protection Agency, Washington, DC, EPA/600/R-08/139F, 2009.  Docket EPA-HQ-OAR-2010-0799.

201 U.S. EPA (2005) Review of the National Ambient Air Quality Standard for Paniculate Matter: Policy
Assessment of Scientific and Technical Information, OAQPS Staff Paper. EPA-452/R-05-005. Docket EPA-
HQ-OAR-2010-0799.

202 U.S. EPA, 2008. Integrated Science Assessment for Oxides of Nitrogen and Sulfur- Ecological Criteria
(Final). U.S. EPA, Washington D.C., EPA/600/R-08/082F. Docket EPA-HQ-OAR-2010-0799.

203 U.S. EPA, 2008. Integrated Science Assessment for Oxides of Nitrogen and Sulfur- Ecological Criteria
(Final). U.S. EPA, Washington D.C., EPA/600/R-08/082F. Docket EPA-HQ-OAR-2010-0799.

204 Environmental Protection Agency (2003). Response Of Surface Water Chemistry to the Clean Air Act
Amendments of 1990. National Health and Environmental Effects Research Laboratory, Office of Research and
Development, U.S. Environmental Protection Agency. Research Triangle Park, NC. EPA 620/R-03/001. Docket
EPA-HQ-OAR-2010-0799.

205 Fenn, M.E. and Blubaugh, TJ. (2005) Winter Deposition of Nitrogen and Sulfur in the Eastern Columbia
River Gorge National Scenic Area, USD A Forest Service. Docket EPA-HQ-OAR-2010-0799.

206 Galloway, J. N.; Cowling, E. B. (2002). Reactive nitrogen and the world: 200 years of change. Ambio 31:  64-
71. Docket EPA-HQ-OAR-2010-0799.

207 Bricker, Suzanne B., et al., National Estuarine Eutrophication Assessment, Effects of Nutrient Enrichment in
the Nation's Estuaries, National Ocean Service, National Oceanic and Atmospheric Administration, September,
1999. Docket EPA-HQ-OAR-2010-0799.

208 Smith, W.H. 1991. "Air pollution and Forest Damage." Chemical Engineering News, 69(45): 30-43. Docket
EPA-HQ-OAR-2010-0799.

209 Gawel, J.E.; Ahner, B.A.; Friedland, A.J.; and Morel, F.M.M. 1996. "Role for heavy metals in forest decline
indicated by phytochelatin measurements." Nature, 381: 64-65.  Docket EPA-HQ-OAR-2010-0799.

210 Cotrufo, M.F.; DeSanto, A.V.; Alfani, A.; et al. 1995. "Effects of urban heavy metal pollution on organic
matter decomposition in Quercus ilix L. woods." Environmental Pollution, 89: 81-87. Docket EPA-HQ-OAR-
2010-0799.

211 Niklinska, M.; Laskowski, R.; Maryanski, M. 1998. "Effect of heavy metals and storage time on two types of
forest litter: basal respiration rate and exchangeable metals." Ecotoxicological Environmental Safety, 41: 8-18.
Docket EPA-HQ-OAR-2010-0799.

212 U.S. EPA (2009). Integrated Science Assessment for Paniculate Matter (Final Report). U.S. Environmental
Protection Agency, Washington, DC, EPA/600/R-08/139F, 2009. Section 9.4.5.2.  Docket EPA-HQ-OAR-2010-
0799

213 Mason, R.P. and Sullivan, K.A.  1997. "Mercury in Lake Michigan." Environmental Science & Technology,
31: 942-947. (from Delta Report "Atmospheric deposition of toxics to the Great Lakes"). Docket EPA-HQ-
OAR-2010-0799.

214 Landis, M.S. and Keeler, G. J. 2002. "Atmospheric mercury deposition to Lake Michigan during the Lake
Michigan Mass Balance Study." Environmental Science & Technology, 21: 4518-24.  Docket EPA-HQ-OAR-
2010-0799.
                                               6-63

-------
Chapter 6
215 U.S. EPA. 2000. EPA453/R-00-005, "Deposition of Air Pollutants to the Great Waters: Third Report to
Congress," Office of Air Quality Planning and Standards, Research Triangle Park, North Carolina. Docket EPA-
HQ-OAR-2010-0799.

216 National Science and Technology Council (NSTC) 1999. "The Role of Monitoring Networks in the
Management of the Nation's Air Quality." Docket EPA-HQ-OAR-2010-0799.

217 Callender, E. and Rice, K.C. 2000. "The Urban Environmental Gradient: Anthropogenic Influences on the
Spatial and Temporal Distributions of Lead and Zinc in Sediments." Environmental Science & Technology, 34:
232-238. Docket EPA-HQ-OAR-2010-0799.

218 Rice, K.C. 1999. "Trace Element Concentrations in Streambed Sediment Across the Conterminous United
States." Environmental Science & Technology, 33: 2499-2504. Docket EPA-HQ-OAR-2010-0799.

219 Ely, JC; Neal, CR; Kulpa, CF; et al. 2001. "Implications  of Platinum-Group Element Accumulation along
U.S. Roads from Catalytic-Converter Attrition." Environ. Sci. Technol. 35: 3816-3822. Docket EPA-HQ-OAR-
2010-0799.

220 U.S. EPA. 1998. EPA454/R-98-014, "Locating and Estimating Air Emissions from Sources of Fob/cyclic
Organic Matter," Office of Air Quality Planning and Standards, Research Triangle Park, North Carolina. Docket
EPA-HQ-OAR-2010-0799.

221 U.S. EPA. 1998. EPA454/R-98-014, "Locating and Estimating Air Emissions from Sources of Fob/cyclic
Organic Matter," Office of Air Quality Planning and Standards, Research Triangle Park, North Carolina. Docket
EPA-HQ-OAR-2010-0799.

222 Simcik, M.F.; Eisenreich, S.J.; Golden, K.A.; et al. 1996. "Atmospheric Loading of Fob/cyclic Aromatic
Hydrocarbons to Lake Michigan as Recorded in the Sediments." Environmental Science and Technology, 30:
3039-3046.  Docket EPA-HQ-OAR-2010-0799.

223 Simcik, M.F.; Eisenreich, S.J.; and Lioy, P.J. 1999. "Source apportionment and source/sink relationship of
PAHs in the coastal atmosphere of Chicago and Lake Michigan." Atmospheric Environment, 33: 5071-5079.
Docket EPA-HQ-OAR-2010-0799.

224 Arzayus, K.M.; Dickhut, R.M.; and Canuel, E.A. 2001. "Fate of Atmospherically Deposited Fob/cyclic
Aromatic Hydrocarbons (PAHs) in Chesapeake Bay." Environmental Science & Technology, 35, 2178-2183.
Docket EPA-HQ-OAR-2010-0799.

225 Park, J.S.; Wade, T.L.; and  Sweet, S. 2001. "Atmospheric distribution of polycyclic aromatic hydrocarbons
and deposition to Galveston Bay, Texas, USA." Atmospheric Environment, 35: 3241-3249.  Docket EPA-HQ-
OAR-2010-0799.

226 Poor, N.; Tremblay, R.; Kay, H.;  et al. 2002. "Atmospheric concentrations and dry deposition rates of
polycyclic aromatic hydrocarbons (PAHs) for Tampa Bay, Florida, USA." Atmospheric Environment 38: 6005-
6015. Docket EPA-HQ-OAR-2010-0799.

227 Arzayus, K.M.; Dickhut, R.M.; and Canuel, E.A. 2001. "Fate of Atmospherically Deposited Polycyclic
Aromatic Hydrocarbons (PAHs) in Chesapeake Bay." Environmental Science & Technology, 35, 2178-2183.
Docket EPA-HQ-OAR-2010-0799.

228 U.S. EPA. 2000. EPA453/R-00-005, "Deposition of Air Pollutants to the Great Waters: Third Report to
Congress," Office of Air Quality Planning and Standards, Research Triangle Park, North Carolina.  Docket EPA-
HQ-OAR-2010-0799.

229 Van Metre, P.C.; Mahler, B.J.; and Furlong, E.T. 2000. "Urban Sprawl Leaves its PAH Signature."
Environmental Science & Technology, 34: 4064-4070. Docket EPA-HQ-OAR-2010-0799.

230 Cousins, I.T.; Beck, A.J.; and Jones, K.C. 1999. "A review of the processes involved in the exchange of semi-
volatile organic compounds across the air-soil interface." The Science of the Total Environment, 228: 5-24.
Docket EPA-HQ-OAR-2010-0799.

231 Tuhackova, J. et al. (2001) Hydrocarbon deposition and soil microflora as affected by highway traffic.
Environmental Pollution,  113:255-262. Docket EPA-HQ-OAR-2010-0799.
                                              6-64

-------
                                                      2017 Draft Regulatory Impact Analysis
232 U.S. EPA. 1991. Effects of organic chemicals in the atmosphere on terrestrial plants. EPA/600/3-91/001.
Docket EPA-HQ-OAR-2010-0799.

233 cape JN, ID Leith, J Binnie, J Content, M Donkin, M Skewes, DN Price AR Brown, AD Sharpe. 2003.
Effects of VOCs on herbaceous plants in an open-top chamber experiment. Environ. Pollut. 124:341-343.
Docket EPA-HQ-OAR-2010-0799.

234 Cape JN, ID Leith, J Binnie, J Content, M Donkin, M Skewes, DN Price AR Brown, AD Sharpe.  2003.
Effects of VOCs on herbaceous plants in an open-top chamber experiment. Environ. Pollut. 124:341-343.
Docket EPA-HQ-OAR-2010-0799.

235 Viskari E-L. 2000. Epicuticular wax of Norway spruce needles as indicator of traffic pollutant deposition.
Water, Air, and Soil Pollut. 121:327-337.  Docket EPA-HQ-OAR-2010-0799.

236 Ugrekhelidze D, F Korte, G Kvesitadze. 1997. Uptake and transformation of benzene and toluene by plant
leaves. Ecotox. Environ. Safety 37:24-29.  Docket EPA-HQ-OAR-2010-0799.

237 Kammerbauer H, H Selinger, R Rommelt, A Ziegler-Jons, D Knoppik, B Hock. 1987. Toxic components of
motor vehicle emissions for the spruce Picea abies. Environ. Pollut. 48:235-243. Docket EPA-HQ-OAR-2010-
0799.

238 U.S. EPA. (2007). PM2.5 National Ambient Air Quality Standard Implementation Rule (Final). Washington,
DC: U.S.  EPA. Retrieved on May 14, 2009 from Docket EPA-HQ-OAR-2003-0062 at
http://www.regulations.gOv/.72 FR 20586.
U.S. EPA. (2006;. Air Quality Criteria for Ozone and Related Photochemical Oxidants (Final). EPA/600/R-
05/004aF-cF. Washington, DC: U.S. EPA. Retrieved on March 19, 2009 from Docket EPA-HQ-OAR-2003-
0190 at http://www.regulations.gov/.

239 PM Standards Revision - 2006: Timeline. Retrieved on March 19, 2009 from
http://www.epa.gov/oar/particlepollution/naaqsrev2006.htmlttimeline
240 US EPA: 8-hour Ozone Nonattainment Areas, http://www.epa.gov/oar/oaqps/greenbk/gnsum.html

241 U.S. EPA. (2011)  Summary of Results  for the 2005  National-Scale Assessment.
http://www.epa.gov/ttn/atw/nata2005/05pdf/sum_results.pdf.

242 Control of Hazardous Air Pollutants From Mobile Sources (72 FR 8428; February 26, 2007)

243 US EPA (2007) Control of Hazardous Air Pollutants from Mobile Sources Regulatory Impact Analysis. EPA
document number 420-R-07-002, February 2007.

244U.S. Environmental Protection Agency, Byun, D.W., and Ching, J.K.S., Eds, 1999. Science algorithms of
EPA Models-3 Community Multiscale Air Quality (CMAQ modeling system, EPA/600/R-99/030, Office of
Research and Development).

245 Byun, D.W., and Schere, K.L., 2006. Review of the  Governing Equations, Computational Algorithms, and
Other Components of the Models-3 Community Multiscale Air Quality (CMAQ) Modeling System, J. Applied
Mechanics Reviews, 59 (2), 51-77.

246Dennis, R.L., Byun, D.W., Novak, J.H., Galluppi, K.J.,  Coats, C.J., and Vouk, M.A., 1996. The next
generation of integrated air quality modeling: EPA's Models-3, Atmospheric Environment, 30, 1925-1938.

247 US EPA (2010). Regulatory Impact Analysis of the Final Rulemaking to Establish Light-Duty Vehicle
Greenhouse Gas Emission Standards and Corporate Average Fuel Economy Standards. EPA document number
420-R-10-009, April 2010.
248 Allen, D., Burns, D., Chock, D., Kumar, N., Lamb, B., Moran, M. (February 2009). Report on the Peer
Review of the Atmospheric Modeling and Analysis Division, NERL/ORD/EPA. U.S. EPA, Research Triangle
Park, NC.,  http://www.epa.gov/amad/peer/2009_AMAD_PeerReviewReport.pdf.
                                              6-65

-------
Chapter 6
249 Hogrefe, C., Biswas, I, Lynn, B., Civerolo, K., Ku, J.Y., Rosenthal, J., et al. (2004). Simulating regional-
scale ozone climatology over the eastern United States: model evaluation results. Atmospheric Environment,
38(17), 2627-2638. Docket EPA-HQ-OAR-2009-0472-11350

250 Lin, M, Oki, T., Holloway, T., Streets, D.G., Bengtsson, M., Kanae, S. (2008). Long-range transport of
acidifying substances in East Asia-Part I: Model evaluation and sensitivity studies. Atmospheric Environment,
42(24), 5939-5955. Docket EPA-HQ-OAR-2009-0472-11341

251 U.S. Environmental Protection Agency. (2008). Technical support document for the final locomotive/marine
rule: Air quality modeling analyses. Research Triangle Park, N.C.: U.S. Environmental Protection Agency,
Office of Air Quality Planning and Standards, Air Quality Assessment Division. Docket EPA-HQ-OAR-2009-
0472-11329

252 CMAQ v4.7.1 release notes are available at: http://www.cmaq-model.org/

253 Grell, G., Dudhia,  J., Stauffer, D. (1994). A Description of the Fifth-Generation Penn State/NCAR Mesoscale
Model (MM5), NCAR/TN-398+STR., 138 pp, National Center for Atmospheric Research, Boulder CO

254 Grell, G., Dudhia, J., Stauffer, D. (1994). A Description of the Fifth-Generation Penn State/NCAR Mesoscale
Model (MM5), NCAR/TN-398+STR., 138 pp, National Center for Atmospheric Research, Boulder CO.

255 Byun,  D.W., Ching, J. K.S. (1999). Science algorithms of EPA Models-3 Community Multiscale Air Quality
(CMAQ) modeling system, EPA/600/R-99/030, Office of Research and Development).  Please also see:
http://www.cmascenter.org/. Docket EPA-HQ-OAR-2009-0472-1915

256Yantosca, B. (2004). GEOS-CHEMv7-01-02 User's Guide, Atmospheric Chemistry Modeling Group,
Harvard University, Cambridge, MA, October 15, 2004.

257 U.S. Environmental Protection Agency. (2011). Regulatory Impact Analysis (RIA) for the final Transport
Rule Docket ID No. EPA-HQ-OAR-2009-0491. June 2011.

258 U.S. Environmental Protection Agency. (2010). FinalRulemaking to Establish Light-Duty Vehicle
Greenhouse Gas Emission Standards and Corporate Average Fuel Economy Standards: Regulatory Impact
Analysis,  Assessment and Standards Division, Office of Transportation and Air Quality, EPA-420-R-10-009,
April 2010. Available on the internet: http://www.epa.gov/otaq/climate/regulations/420rl0009.pdf

259 U.S. Environmental Protection Agency (U.S. EPA). 2010. Regulatory Impact Analysis: National Emission
Standards for Hazardous Air Pollutants from the Portland Cement Manufacturing Industry. Office of Air
Quality Planning and Standards, Research Triangle Park, NC. Augues.  Available on the Internet at <
http://www.epa.gov/ttn/ecas/regdata/RIAs/portlandcementfinalria.pdf >. EPA-HQ-OAR-2009-0472-0241

260 U.S.  Environmental Protection Agency (U.S. EPA), 2010. Regulatory Impact Analysis,
Final Rulemaking to Establish Light-Duty Vehicle Greenhouse Gas  Emission Standards and
Corporate Average Fuel Economy Standards.  Office of Transportation and Air Quality.
April. Available at http://www.epa.gov/otaq/climate/regulations/420rl0009.pdf. EPA-420-R-
10-009

261 U.S.  Environmental Protection Agency (U.S. EPA). 2010. Regulatory Impact Analysis:
National Emission Standards for Hazardous Air Pollutants from the  Portland Cement
Manufacturing Industry. Office of Air Quality Planning and Standards, Research  Triangle
Park, NC. Augues.  Available on the Internet at <
http://www.epa.gov/ttn/ecas/regdata/RIAs/portlandcementfmalria.pdf>. EPA-HQ-OAR-
2009-0472-0241
                                             6-66

-------
                                                       2017 Draft Regulatory Impact Analysis
262 U.S. Environmental Protection Agency (U.S. EPA). 2010. Final NO2 NAAQS Regulatory Impact Analysis
(RIA). Office of Air Quality Planning and Standards, Research Triangle Park, NC. April. Available on the
Internet at http://www.epa.gov/ttn/ecas/regdata/RIAs/FinalNO2RIAfulldocument.pdf. Accessed March 15,
2010. EPA-HQ-OAR-2009-0472-0237U.S. Environmental Protection Agency (U.S. EPA).  2009.

263 U.S. Environmental Protection Agency.  October 2006.  Final Regulatory Impact Analysis
(RIA) for the Final National Ambient Air Quality Standards for Particulate Matter.  Prepared
by: Office of Air and Radiation.

264 Information on BenMAP, including downloads of the software, can be found at http://www.epa.gov/ttn/ecas/
benmodels.html.

265 Bell, M.L., et al. (2004). Ozone and short-term mortality in 95 US urban communities, 1987-2000. JAMA,
2004. 292(19): p. 2372-8.

266 Huang, Y.; Dominici, F.; Bell, M. L.  (2005) Bayesian hierarchical distributed lag models for summer ozone
exposure and cardio-respiratory mortality. Environmetrics. 16: 547-562.

267 Schwartz, J. (2005) How sensitive is the association between ozone and daily deaths to control for
temperature? Am. J. Respir. Crit. CareMed. 171: 627-631.

268 Bell, M.L., F. Dominici, and J.M. Samet. (2005). A meta-analysis of time-series studies of ozone  and
mortality with comparison to the national morbidity, mortality, and air pollution study. Epidemiology. 16(4): p.
436-45.

269 Ito, K., S.F. De Leon, and M. Lippmann (2005).  Associations between ozone and daily mortality:  analysis
and meta-analysis. Epidemiology. 16(4): p. 446-57.

270 Levy,  J.I., S.M. Chemerynski, and J.A. Sarnat. (2005).  Ozone exposure and mortality: an empiric bayes
metaregression analysis. Epidemiology.  16(4): p. 458-68.

271 Pope,  C.A., III, R.T.  Burnett, M.J. Thun, E.E. Calle, D. Krewski, K. Ito, and G.D. Thurston.  (2002). "Lung
Cancer, Cardiopulmonary Mortality, and Long-term Exposure to Fine Particulate Air Pollution." Journal of the
American Medical Association 287:1132-1141.

272 Laden, F., J.  Schwartz, F.E. Speizer, and D.W. Dockery. (2006).  Reduction in Fine Particulate Air Pollution
and Mortality. American Journal of Respiratory and Critical Care Medicine.  173: 667-672.

273 Industrial Economics, Incorporated (IEc). (2006). Expanded Expert Judgment Assessment oj'the
Concentration-Response Relationship Between PM2.5 Exposure and Mortality.  Peer Review Draft.  Prepared
for: Office of Air Quality Planning and Standards, U.S. Environmental Protection Agency, Research Triangle
Park, NC. August.

        274 Woodruff, T.J., J. Grille, and K.C. Schoendorf. (1997).  The Relationship Between Selected Causes
of Postneonatal Infant Mortality and Particulate Air Pollution in the United States. Environmental Health
Perspectives. 105(6):608-612.

        275 Abbey, D.E., B.L. Hwang, R.J. Burchette, T. Vancuren, and P.K. Mills. (1995).  Estimated Long-
Term Ambient Concentrations of PM(10) and Development of Respiratory Symptoms in a Nonsmoking
Population. Archives of Environmental Health.  50(2): 139-152.

        276 Peters, A., D.W. Dockery, J.E. Muller, and M.A. Mittleman.  (2001). Increased Particulate Air
Pollution and the Triggering of Myocardial Infarction.  Circulation. 103:2810-2815.

277 Schwartz J. (1995).  Short term fluctuations in air pollution and hospital admissions of the elderly for
respiratory disease.  Thorax. 50(5):531-538.

278 Schwartz J. (1994a). PM(10) Ozone, and Hospital Admissions For the Elderly in Minneapolis St Paul,
Minnesota. Arch Environ Health. 49(5):366-374.

279 Schwartz J. (1994b). Air Pollution and Hospital Admissions For the Elderly in Detroit, Michigan. AmJ
Respir Crit Care Med. 150(3):648-655.
                                               6-67

-------
Chapter 6
280 Moolgavkar SH, Luebeck EG, Anderson EL. (1997). Air pollution and hospital admissions for respiratory
causes in Minneapolis St. Paul and Birmingham. Epidemiology. 8(4):364-370.

281 Burnett RT, Smith-Doiron M, Stieb D, Raizenne ME, Brook JR, Dales RE, et al. (2001).  Association
between ozone and hospitalization for acute respiratory diseases in children less than 2 years of age. Am J
Epidemiol.  153(5):444-452.

        282 Moolgavkar, S.H.  (2003). "Air Pollution and Daily Deaths and Hospital Admissions in Los
Angeles and Cook Counties."  In Revised Analyses of Time-Series Studies of Air Pollution and Health.  Special
Report.  Boston, MA: Health Effects Institute.

        283 Ito, K. (2003). "Associations of Paniculate Matter Components with Daily Mortality and Morbidity
in Detroit, Michigan." In Revised Analyses of Time-Series Studies of Air Pollution  and Health. Special  Report.
Health Effects Institute, Boston, MA.

284 Moolgavkar, S.H.  (2000).  Air Pollution and Hospital Admissions for Diseases of the Circulatory System in
Three U.S. Metropolitan Areas. Journal of the Air and Waste Management Association 50:1199-1206.

285 Sheppard, L. (2003). Ambient Air Pollution and Nonelderly Asthma Hospital Admissions in Seattle,
Washington, 1987-1994. In Revised Analyses of Time-Series Studies ofAif Pollution and Health. Special
Report.  Boston, MA: Health Effects Institute.

286 Peel, J. L., P. E. Tolbert, M. Klein, et al.  (2005).  Ambient air pollution and respiratory emergency department
visits. Epidemiology. Vol. 16 (2): 164-74.

287 Wilson, A. M., C. P. Wake, T. Kelly, et al. (2005). Air pollution, weather, and respiratory emergency room
visits in two northern New England cities: an ecological time-series study. EnvironRes. Vol. 97 (3): 312-21.

        288 Norris, G., S.N. YoungPong, J.Q. Koenig, T.V. Larson, L. Sheppard, and J.W. Stout.  (1999). An
Association between Fine Particles and Asthma Emergency Department Visits for Children in Seattle.
Environmental Health Perspectives 107(6):489-493.

        289 Dockery, D.W., J. Cunningham, A.I. Damokosh, L.M. Neas, J.D. Spengler, P. Koutrakis, J.H. Ware,
M. Raizenne, and F.E. Speizer. (1996). Health Effects of Acid Aerosols On North American Children-
Respiratory Symptoms. Environmental Health  Perspectives 104(5):500-505.

290 Pope, C.A., III, D.W. Dockery, J.D. Spengler, and M.E. Raizenne. (1991). Respiratory Health and PM10
Pollution: A Daily Time Series Analysis. American Review of Respiratory Diseases 144:668-674.

291 Schwartz, J., and L.M. Neas.  (2000). Fine Particles are More Strongly Associated than Coarse Particles with
Acute Respiratory Health Effects in Schoolchildren. Epidemiology 11:6-10.

        292 Ostro, B., M. Lipsett, J. Mann, H. Braxton-Owens, and M. White. (2001).  Air Pollution and
Exacerbation of Asthma in African-American Children in Los Angeles.  Epidemiology 12(2):200-208.

        293 Vedal, S., J. Petkau, R. White, and J. Blair. (1998). Acute Effects of Ambient Inhalable Particles in
Asthmatic and Nonasthmatic Children. American Journal of Respiratory and Critical Care Medicine
157(4): 1034-1043.

        294 Ostro, B.D. (1987).  Air Pollution and Morbidity Revisited: A Specification Test.  Journal  of
Environmental Economics Management 14:87-98.

295 Gilliland FD, Berhane K, Rappaport EB, Thomas DC, Avol E, Gauderman WJ, et al. (2001).  The effects of
ambient air pollution on school absenteeism due to respiratory illnesses. Epidemiology 12(l):43-54.

296 Chen L,  Jennison BL, Yang W, Omaye ST.  (2000). Elementary school absenteeism and air pollution. Inhal
7b;dco/12(ll):997-1016.

        297 Ostro, B.D. and S. Rothschild. (1989).  Air Pollution and Acute Respiratory Morbidity: An
Observational Study of Multiple Pollutants. Environmental Research 50:238-247.

        298 Russell, M.W., D.M. Huse, S. Drowns,  B.C. Hamel, and S.C. Hartz. (1998). Direct Medical Costs
of Coronary Artery Disease in the United States. American Journal of Cardiology 81(9):1110-1115.
                                                6-68

-------
                                                      2017 Draft Regulatory Impact Analysis
        299 Wittels, E.H., J.W. Hay, and A.M. Gotto, Jr. (1990). Medical Costs of Coronary Artery Disease in
the United States. American Journal of Cardiology 65(7):432-440.

        300 Smith, D.H., D.C. Malone, K.A. Lawson, L.J. Okamoto, C. Battista, and W.B. Saunders. (1997). A
National Estimate of the Economic Costs of Asthma.  American Journal of Respiratory and Critical Care
Medicine 156(3 Pt l):787-793.

        301 Stanford, R., T. McLaughlin, and L.J. Okamoto. (1999).  The Cost of Asthma in the Emergency
Department and Hospital. American Journal of Respiratory and Critical Care Medicine  160(1):211-215.

302 Rowe, R.D., and L.G. Chestnut. (1986). Oxidants and Asthmatics in Los Angeles: A Benefits Analysis—
Executive Summary. Prepared by Energy and Resource Consultants, Inc. Report to the U.S. Environmental
Protection Agency, Office of Policy Analysis. EPA-230-09-86-018. Washington, DC.

303 Science Advisory Board.  2001. NATA - Evaluating the National-Scale Air Toxics Assessment for 1996 -
an SAB Advisory, http://www.epa.gov/ttn/atw/sab/sabrev.html.

304 U.S. Environmental Protection Agency (U.S. EPA). 2011. The Benefits and Costs of the Clean Air Act from
1990 to 2020. Office of Air and Radiation, Washington, DC. March.  Available on the Internet at
.

305 U.S. Environmental Protection Agency—Science Advisory Board (U.S. EPA-SAB). 2008. Benefits of
Reducing Benzene Emissions in Houston, 1990-2020. EPA-COUNCIL-08-001. July. Available at
.

306 U.S. EPA (2011) Inventory of U.S. Greenhouse Gas Emissions and Sinks:  1990-2009. EPA 430-R-l 1-005.

307 National Research Council (NRC) (2010). Advancing the Science of Climate  Change. National Academy
Press.  Washington, DC. Docket EPA-HQ-OAR-2010-0799.
308
   Brenkert A, S. Smith, S. Kim, andH. Pitcher, 2003: Model Documentation for the MiniCAM. PNNL-14337,
Pacific Northwest National Laboratory, Richland, Washington. Docket EPA-HQ-OAR-2010-0799.

309 Wigley, T.M.L. and Raper, S.C.B. 1992. Implications for Climate And Sea-Level of Revised IPCC Emissions
Scenarios Nature 357, 293-300. Raper, S.C.B., Wigley T.M.L. and Warrick R.A. 1996. in Sea-Level Rise and
Coastal Subsidence: Causes, Consequences and Strategies J.D. Milliman, B.U. Haq, Eds., Kluwer Academic
Publishers, Dordrecht, The Netherlands, pp. 11-45. Docket EPA-HQ-OAR-2010-0799.

310 Wigley, T.M.L. and Raper, S.C.B. 2002. Reasons for larger warming projections in the IPCC Third
Assessment Report J. Climate 15, 2945-2952. Docket EPA-HQ-OAR-2010-0799.

311 Thompson AM, KV Calvin, SJ Smith, GP Kyle, A Volke, P Patel, S Delgado-Arias, B Bond-Lamberty, MA
Wise, LE Clarke and JA Edmonds. 2010. "RCP4.5: A Pathway for Stabilization of Radiative Forcing by 2100."
Climatic Change (in review)

312 Clarke, L., J. Edmonds, H. Jacoby, H. Pitcher, J. Reilly, R. Richels, (2007) Scenarios of Greenhouse Gas
Emissions and Atmospheric Concentrations. Sub-report 2.1A of Synthesis and Assessment Product 2.1 by the
U.S. Climate Change Science Program and the Subcommittee on Global Change Research (Department of
Energy, Office of Biological & Environmental Research, Washington, DC., USA, 154 pp.). Docket EPA-HQ-
OAR-2010-0799.

313 Wigley, T.M.L. 2008. MAGICC 5.3.v2 User Manual. UCAR - Climate and Global Dynamics Division,
Boulder, Colorado, http://www.cgd.ucar.edu/cas/wiglev/magicc/. Docket EPA-HQ-OAR-2010-0799.

314 Meehl, G. A. et al. (2007) Global Climate Projections. In:  Climate Change 2007: The Physical Science Basis.
Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on  Climate
                                               6-69

-------
Chapter 6


Change [Solomon, S., D. Qin, M. Manning, Z. Chen, M. Marquis, K.B. Averyt, M. Tignor and H.L. Miller
(eds.)]. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA. Docket EPA-HQ-
OAR-2010-0799.

315 National Research Council, 2011. Climate Stabilization Targets: Emissions, Concentrations, and Impacts over
Decades to Millenia. Washington, DC: National Academies Press. Docket EPA-HQ-OAR-2010-0799.

316 Lewis, E., and D. W. R. Wallace. 1998. Program Developed for CO2 System Calculations. ORNL/CDIAC-
105. Carbon Dioxide Information Analysis Center, Oak Ridge National Laboratory, U.S. Department of Energy,
Oak Ridge,  Tennessee. Docket EPA-HQ-OAR-2010-0799.

317 Lewis, E., and D. W. R. Wallace. 1998. Program Developed for CO2 System Calculations. ORNL/CDIAC-
105. Carbon Dioxide Information Analysis Center, Oak Ridge National Laboratory, U.S. Department of Energy,
Oak Ridge,  Tennessee. Docket EPA-HQ-OAR-2010-0799.

318 Mehrbach, C., C. H. Culberson, J. E. Hawley, and R. N. Pytkowicz.  1973. Measurement of the apparent
dissociation constants of carbonic acid in seawater at atmospheric pressure. Limnology and Oceanography
18:897-907. Docket EPA-HQ-OAR-2010-0799.

319 Dickson, A.  G. and F. J. Millero. 1987. A comparison of the equilibrium constants for the dissociation of
carbonic acid in seawater media. Deep-Sea Res.  34, 1733-1743. (Corrigenda. Deep-Sea Res. 36, 983). Docket
EPA-HQ-OAR-2010-0799.

320 A. G. Dickson. 1990. Thermodynamics of the dissociation of boric acid in synthetic sea water from 273.15 to
318.15 K. Deep-Sea Res. 37,  755-766. Docket EPA-HQ-OAR-2010-0799.

321 Lewis, E., and D. W. R. Wallace. 1998. Program Developed for CO2 System Calculations. ORNL/CDIAC-
105. Carbon Dioxide Information Analysis Center, Oak Ridge National Laboratory, U.S. Department of Energy,
Oak Ridge,  Tennessee. Docket EPA-HQ-OAR-2010-0799.

322Khoo, K.H., R.W. Ramette, C.H. Culberson,  andR. G. Bates. 1977. Determination of hydrogen ion
concentrations in seawater  from 5 to 40°C: Standard potentials at salinities from 20 to 45%o. Analytical
Chemistry 49(1): 29-34. Docket EPA-HQ-OAR-2010-0799.

323 Dickson, A.  G. 2003. Certificate of Analysis  - Reference material for oceanic CO2 measurements (Batch
#62, bottled on August 21,  2003). Certified by Andrew Dickson, Scripps Institution of Oceanography. November
21, 2003. Docket EPA-HQ-OAR-2010-0799.

Dickson, A. G.  2005. Certificate of Analysis - Reference material for oceanic CO2 measurements (Batch #69,
bottled on January 4, 2005). Certified by Andrew Dickson, Scripps Institution of Oceanography. July 12, 2005.
Docket EPA-HQ-OAR-2010-0799.

Dickson, A. G.  2009. Certificate of Analysis - Reference material for oceanic CO2 measurements (Batch #100,
bottled on November 13, 2009). Certified by Andrew Dickson, Scripps Institution of Oceanography. February
10, 2010. Docket EPA-HQ-OAR-2010-0799.
                                              6-70

-------
                                                  2017 Draft Regulatory Impact Analysis
7      Other Economic and Social Impacts

       This Chapter presents a summary of the total costs and benefits of EPA's proposed GHG
standards. We note that this summary of costs and benefits of EPA's GHG standards does not
change the fact that both the CAFE and GHG standards, jointly, will be the source of the benefits
and costs of the National Program. These costs and benefits are appropriately analyzed
separately by each agency and should not be  added together.

       For several reasons, the estimates for  costs and benefits presented by NHTSA and EPA,
while consistent, are not directly comparable, and thus should not be expected to be identical.
Most important, NHTSA and EPA's standards would require different fuel efficiency
improvements. EPA's proposed GHG standard is more  stringent in part reflecting our
projections regarding manufacturers' use of air conditioning leakage credits, which result from
reductions in air conditioning-related emissions of HFCs. NHTSA is proposing standards at
levels of stringency that assume improvements in the efficiency of air conditioning systems, but
that do not account for reductions in HFCs, which are not related to fuel economy or energy
conservation. In addition, the CAFE and GHG standards offer somewhat different program
flexibilities and provisions, and the agencies' analyses differ in their accounting for these
flexibilities (examples include the treatment of EVs, dual-fueled vehicles, and restrictions on
transfer of credits between car and truck fleets), primarily because NHTSA is statutorily
prohibited from considering some flexibilities when establishing CAFE standards,HHHHH while
EPA is not. Also, manufacturers may opt to pay a civil penalty in lieu of actually meeting CAFE
standards, but they  cannot pay a fine to avoid complying with EPA's proposed GHG standards.
Some manufacturers have traditionally paid CAFE penalties instead of complying with the
CAFE standards. These differences contribute to differences in the agencies' respective
estimates of costs and benefits resulting from the new standards.  Nevertheless, it is important to
note that NHTSA and EPA have harmonized the programs as  much as possible, and this proposal
to continue the National Program would result in significant cost and other advantages for the
automobile industry by allowing them to manufacture one fleet of vehicles across the U.S., rather
than comply with potentially multiple state standards that may occur in the absence of the
National Program.

       For the reader's reference, Table 7.1-1 below summarizes the values of a number of joint
economic and other values that the agencies used to estimate the overall costs and benefits
associated with each agency's proposed standard. Note, however, that the values presented in
this table are summaries of the inputs used for the agencies' respective models.  See draft Joint
TSD Chapter 4 for expanded discussion and details on each of these joint economic and other
values.

       This Chapter includes an expanded description of the agencies' approach to the
monetization of CC>2 emission reductions.  Though the underlying unit values are consistent with
those used in NHTSA's analysis of the proposed CAFE  standards, the specific stream of CC>2-
1111111111 See 49 U.S.C. 32902(h).
                                          7-1

-------
Chapter 7
related benefits are unique to each program and EPA's benefits are therefore presented in section
7.1.

    Table 7.1-6.4-1 Joint Economic and other Values for Benefits Computations (2009$)
Fuel Economy Rebound Effect
"Gap" between test and on-road MPGfor liquid-fueled vehicles
"Gap" between test and on-road wall electricity consumption for
electric and plug-in hybrid electric vehicles
Value of refueling time per ($ per vehicle-hour)
Average tank volume refilled during refueling stop
Annual growth in average vehicle use
Fuel Prices (2017-50 average, $/gallon)
Retail gasoline price
Pre-tax gasoline price
Economic Benefits from Reducing Oil Imports ($/gallon)
"Monopsony" Component
Price Shock Component
Military Security Component
Total Economic Costs ($/gallon)
Emission Damage Costs (2020, $/short ton)
Carbon monoxide
Volatile organic compounds (VOC)
Nitrogen oxides (NOX) - vehicle use
Nitrogen oxides (NOX) - fuel production and distribution
Paniculate matter (PM2 5) - vehicle use
Paniculate matter (PM2 5) - fuel production and distribution
Sulfur dioxide (SO2)

Annual CO2 Damage Cost (per metric ton)
External Costs from Additional Automobile Use ($/vehicle-mile)
Congestion
Accidents
Noise
Total External Costs
External Costs from Additional Light Truck Use ($/vehicle-mile)
Congestion
Accidents
Noise
Total External Costs
Discount Rates Applied to Future Benefits
10%
20%
30%
$ 22.02
57%
1% through 2030,
0.5% thereafter

$3.71
$3.35

$0.00
$0.185 in 2025
$0.00
$0.185 in 2025

$0
$ 1,300
$ 5,500
$ 5,300
$ 300,000
$ 250,000
$ 32,000

variable,
depending on
discount rate and
year (see PJA
Chapter 7.1
below)

$ 0.056
$ 0.024
$0.001
$ 0.080

$0.049
$0.027
$0.001
$0.077
3%, 7%
                                         7-2

-------
                                                   2017 Draft Regulatory Impact Analysis
7.1 Monetized CO2 Estimates

       We assigned a dollar value to reductions in carbon dioxide (CO2) emissions using recent
estimates of the "social cost of carbon" (SCC). The SCC is an estimate of the monetized
damages associated with an incremental increase in carbon emissions in a given year.  It is
intended to include (but is not limited to) changes in net agricultural productivity, human health,
property damages from increased flood risk, and the value of ecosystem services due to climate
change. The  SCC  estimates used in this analysis were developed through an interagency process
that included  EPA, DOT/NHTSA, and other executive branch entities,  and concluded in
February 2010. The SCC  Technical Support Document (SCC TSD) provides a complete
discussion of the methods  used to develop these SCC estimates.324

        The interagency group selected four SCC values for use in regulatory  analyses, which
we have applied in this analysis:  $5, $22, $36, and $67 per metric ton of CC>2 emissionsmn'JJJJJ in
the year 2010, and in 2009 dollars. The first three values are based on the average SCC from
three integrated assessment models,  at discount rates of 5, 3, and 2.5 percent, respectively. SCCs
at several discount rates are included because the literature shows that the SCC is quite sensitive
to assumptions about the discount rate, and because no consensus exists on the appropriate rate
to use in an intergenerational context. The fourth value is the 95th percentile of the SCC from all
three models  at a 3 percent discount rate. It is included to represent higher-than-expected impacts
from temperature change further out in the tails of the SCC distribution. Low probability, high
impact events are incorporated into all of the SCC values through explicit consideration of their
effects in two of the three models as well as the use of a probability density function for
equilibrium climate sensitivity. Treating climate sensitivity probabilistically results in more high
temperature outcomes, which in turn lead to higher projections of damages.

       The SCC increases over time because future emissions are expected to produce larger
incremental damages as physical and economic systems become more stressed in response to
greater climatic change. Note that the interagency group estimated the growth rate of the SCC
directly using the three integrated assessment models rather than assuming a constant annual
growth rate. This helps to ensure that the estimates are internally consistent with other modeling
assumptions.  Table 7.1-2  presents the SCC estimates used in this analysis.

       When attempting to assess the incremental economic impacts of carbon dioxide
emissions, the analyst faces a number of serious challenges. A recent report from the National
Academies of Science (NRC 2009) points out that any assessment will suffer from uncertainty,
speculation, and lack of information about (1) future emissions of greenhouse gases, (2) the
effects of past and  future emissions on the climate system, (3) the impact of changes in climate
mn The interagency group decided that these estimates apply only to CO2 emissions. Given that warming profiles
and impacts other than temperature change (e.g. ocean acidification) vary across GHGs, the group concluded
"transforming gases into CO2-equivalents using GWP, and then multiplying the carbon-equivalents by the SCC,
would not result inaccurate estimates of the social costs of non-CO2 gases" (SCC TSD, pg 13).
11111 The SCC estimates were converted from 2007 dollars to 2008 dollars using a GDP price deflator (1.021) and
again to 2009 dollars using a GDP price deflator (1.009) obtained from the Bureau of Economic Analysis, National
Income and Product Accounts Table 1.1.4, Price Indexes for Gross Domestic Product.


                                           7-3

-------
Chapter 7	

on the physical and biological environment, and (4) the translation of these environmental
impacts into economic damages.325 As a result, any effort to quantify and monetize the harms
associated with climate change will raise serious questions of science, economics, and ethics and
should be viewed as provisional.

       The interagency group noted a number of limitations to the SCC analysis, including the
incomplete  way in which the integrated assessment models capture catastrophic and non-
catastrophic impacts, their incomplete treatment of adaptation and technological change,
uncertainty in the extrapolation of damages to high temperatures, and assumptions regarding risk
aversion. The limited amount of research linking climate impacts to economic damages makes
the interagency modeling exercise even more difficult.  The interagency group hopes that over
time researchers and modelers will work to fill these gaps and that the SCC estimates  used for
regulatory analysis by the Federal government will continue to evolve with improvements in
modeling. Additional details on these limitations are discussed in the SCC TSD.

       In light of these limitations, the interagency group has committed to updating the current
estimates as the science and economic understanding of climate change and its impacts on
society improves over time. Specifically,  the interagency group has set a preliminary goal of
revisiting the SCC values in the next few years or at such time as substantially updated models
become available, and to continue to support research in this  area.

Applying the global SCC estimates, shown in Table 7.1-2
                                          7-4

-------
                                                   2017 Draft Regulatory Impact Analysis
Table 7.1-, to the estimated reductions in CO2 emissions under the proposed standards, we
estimate the dollar value of the GHG related benefits for each analysis year. For internal
consistency, the annual benefits are discounted back to net present value terms using the same
discount rate as each SCC estimate (i.e. 5%, 3%, and 2.5%) rather than 3% and 7%.KKKKK The
SCC estimates are presented in and the associated CC>2 benefit estimates for each calendar year
are shown in Tables 7.1-3.
KKKKK jt -g pOSSibie tjjgj other benefits or costs of proposed regulations unrelated to CO2 emissions will be
discounted at rates that differ from those used to develop the SCC estimates.
                                            7-5

-------
Chapter 7
                 Table 7.1-2 Social Cost of CO2 2017-2050a (2009 dollars)
YEAR
2017
2020
2025
2030
2035
2040
2045
2050
DISCOUNT RATE AND STATISTIC
5% AVERAGE
$6.36
$7.01
$8.53
$10.05
$11.57
$13.09
$14.63
$16.18
3% AVERAGE
$25.59
$27.10
$30.43
$33.75
$37.08
$40.40
$43.34
$46.27
2. 5% AVERAGE
$40.94
$42.98
$47.28
$51.58
$55.88
$60.19
$63.59
$66.99
3%95m
PERCENTILE
$78.28
$83.17
$93.11
$103.06
$113.00
$122.95
$131.66
$140.37
       "The SCC values are dollar-year and emissions-year specific.
 Table 7.1-3 Undiscounted Annual Upstream and Downstream COi Benefits for the Given
 SCC Value, and CO2 Benefits Discounted back to 2012, Calendar Year Analysis3 (Millions
                                     of 2009 dollars)
YEAR
2017
2018
2019
2020
2021
2022
2023
2024
2025
2030
2040
2050
NPVb
5%
(AVERAGE SCC =
$6 IN 20 17)
$13
$45
$97
$171
$289
$443
$635
$866
$1,140
$2,690
$5,490
$8,050
$32,800
3%
(AVERAGE SCC =
$26 IN 20 17)
$53
$179
$378
$662
$1,100
$1,650
$2,330
$3,130
$4,070
$9,040
$17,000
$23,000
$172,000
2.5%
(AVERAGE SCC =
$4 UN 20 17)
$85
$286
$602
$1,050
$1,730
$2,600
$3,650
$4,890
$6,320
$13,800
$25,300
$33,300
$292,000
3%
(95™ PERCENTILE =
$78 IN 20 17)
$162
$549
$1,160
$2,030
$3,360
$5,060
$7,150
$9,600
$12,500
$27,600
$51,600
$69,800
$522,000
"The SCC values are dollar-year and emissions-year specific.
b Note that net present value of reduced GHG emissions is calculated differently than other benefits.
discount rate used to discount the value of damages from future emissions (SCC at 5, 3, 2.5 percent)
calculate net present value of SCC for internal consistency.  Refer to SCC TSD for more detail.
The same
is used to
       We also conducted a separate analysis of the CO2 benefits over the model year lifetimes
of the 2017 through 2025 model year vehicles. In contrast to the calendar year analysis, the
model year lifetime analysis shows the impacts of the proposed standards on each of these MY
fleets over the course of its lifetime. Full details of the inputs to this analysis can be found in
Chapter 4 of this DRIA. The CC>2 benefits of the full life of each of the nine model years from
2017 through 2025 are shown in Table 7.1-4 through Table 7.1-7 for each of the four different
social cost of carbon values.  The CC>2 benefits are shown for each year in the model year life
and in net present value. The same discount rate used to discount the value of damages from
                                           7-6

-------
                                                  2017 Draft Regulatory Impact Analysis
future emissions (SCC at 5, 3, 2.5 percent) is used to calculate net present value of SCC for
internal consistency.

  Table 7.1-4 Undiscounted Annual Upstream and Downstream CO2 Benefits for the 5%
 (Average SCC) Value, CO2 Benefits Discounted back to the 1st Year of each MY, and Sum
          of Values Across  MYs, Model Year Analysis3 (Millions of 2009 dollars)
YEAR
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
2031
2032
2033
2034
2035
2036
2037
2038
2039
2040
2041
2042
2043
2044
2045
2046
2047
2048
2049
2050
NPV,
5%
MY
2017
$13
$13
$13
$13
$13
$13
$13
$13
$12
$12
$11
$10
$9
$8
$7
$6
$5
$4
$3
$3
$2
$2
$2
$1
$1
$1
$1
$1
$1
$1
$1
$0
$0
$0
$142
MY
2018
$0
$32
$32
$32
$33
$33
$33
$32
$32
$31
$29
$28
$26
$23
$19
$16
$13
$11
$9
$7
$6
$5
$4
$4
$3
$3
$2
$2
$1
$1
$1
$1
$1
$1
$344
MY
2019
$0
$0
$51
$52
$53
$53
$53
$53
$52
$51
$49
$47
$45
$42
$36
$30
$25
$21
$17
$14
$12
$9
$8
$6
$6
$5
$4
$3
$2
$2
$2
$2
$1
$1
$552
MY
2020
$0
$0
$0
$74
$76
$77
$77
$77
$77
$76
$74
$72
$69
$66
$60
$52
$44
$36
$30
$25
$20
$16
$13
$11
$9
$8
$6
$5
$5
$3
$3
$2
$2
$2
$802
MY
2021
$0
$0
$0
$0
$114
$116
$118
$118
$118
$117
$115
$112
$108
$105
$98
$90
$78
$66
$56
$46
$39
$32
$26
$21
$18
$15
$13
$11
$9
$8
$6
$5
$5
$4
$1,230
MY
2022
$0
$0
$0
$0
$0
$151
$153
$155
$156
$156
$154
$152
$147
$142
$136
$128
$117
$102
$86
$72
$60
$50
$41
$34
$28
$23
$20
$17
$14
$12
$11
$8
$7
$6
$1,610
MY
2023
$0
$0
$0
$0
$0
$0
$187
$190
$193
$193
$192
$190
$186
$182
$175
$167
$156
$143
$124
$105
$88
$74
$61
$51
$41
$34
$28
$24
$21
$18
$15
$13
$9
$8
$1,980
MY
2024
$0
$0
$0
$0
$0
$0
$0
$227
$230
$234
$233
$233
$229
$226
$219
$209
$200
$187
$171
$148
$126
$105
$88
$73
$60
$50
$41
$34
$29
$25
$21
$18
$16
$11
$2,390
MY
2025
$0
$0
$0
$0
$0
$0
$0
$0
$270
$274
$276
$276
$275
$273
$266
$257
$245
$234
$219
$200
$173
$147
$123
$102
$85
$71
$58
$48
$40
$33
$29
$25
$21
$18
$2,810
SUM
$13
$45
$97
$171
$289
$443
$635
$866
$1,141
$1,143
$1,134
$1,120
$1,094
$1,066
$1,016
$955
$885
$804
$716
$622
$527
$441
$367
$304
$252
$209
$173
$145
$122
$103
$88
$74
$62
$52
$11,900
a The SCC values are dollar-year and emissions-year specific. Note that annual data extend to 2052 for the 2017MY
and to 2060 for the 2025MY. These data are not shown but are included in the NPV values.
                                          7-7

-------
Chapter 7
  Table 7.1-5 Undiscounted Annual Upstream and Downstream CO2 Benefits for the 3%
 (Average SCC) SCC Value, CO2 Benefits Discounted back tothe 1st Year of each MY, and
        Sum of Values Across MYs, Model Year Analysis3 (Millions of 2009 dollars)
YEAR
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
2031
2032
2033
2034
2035
2036
2037
2038
2039
2040
2041
2042
2043
2044
2045
2046
2047
2048
2049
2050
NPV,
5%
MY
2017
$53
$53
$52
$51
$51
$50
$48
$46
$44
$42
$39
$36
$31
$27
$22
$19
$16
$13
$11
$9
$7
$6
$5
$5
$4
$3
$2
$2
$2
$2
$2
$1
$1
$1
$598
MY
2018
$0
$127
$126
$126
$124
$123
$120
$117
$113
$108
$102
$96
$87
$76
$64
$53
$44
$36
$29
$24
$19
$16
$13
$12
$10
$8
$7
$5
$4
$4
$3
$3
$3
$2
$1,430
MY
2019
$0
$0
$200
$199
$200
$197
$195
$191
$186
$180
$171
$163
$152
$140
$120
$100
$83
$68
$56
$45
$36
$29
$24
$20
$17
$14
$12
$10
$7
$6
$5
$5
$4
$4
$2,260
MY
2020
$0
$0
$0
$286
$287
$287
$284
$280
$275
$267
$257
$246
$234
$220
$200
$171
$143
$118
$96
$79
$64
$51
$41
$34
$28
$24
$20
$16
$14
$9
$8
$7
$6
$5
$3,240
MY
2021
$0
$0
$0
$0
$434
$433
$433
$427
$422
$413
$400
$386
$367
$351
$327
$297
$256
$214
$178
$147
$122
$99
$81
$66
$54
$45
$39
$33
$27
$23
$17
$15
$13
$11
$4,900
MY
2022
$0
$0
$0
$0
$0
$563
$564
$563
$556
$549
$535
$520
$500
$478
$453
$421
$383
$329
$277
$230
$190
$157
$129
$105
$85
$70
$59
$51
$43
$36
$31
$22
$19
$17
$6,350
MY
2023
$0
$0
$0
$0
$0
$0
$689
$687
$688
$679
$668
$654
$633
$613
$580
$549
$510
$462
$398
$335
$279
$230
$191
$156
$127
$104
$85
$72
$63
$52
$44
$38
$27
$24
$7,730
MY
2024
$0
$0
$0
$0
$0
$0
$0
$822
$822
$822
$809
$799
$778
$760
$728
$689
$651
$603
$547
$472
$396
$330
$273
$226
$185
$150
$123
$101
$85
$74
$62
$52
$45
$32
$9,200
MY
2025
$0
$0
$0
$0
$0
$0
$0
$0
$964
$964
$960
$948
$933
$916
$885
$846
$801
$755
$701
$635
$547
$460
$383
$316
$262
$214
$174
$142
$117
$98
$86
$71
$60
$52
$10,700
SUM
$53
$179
$378
$662
$1,096
$1,652
$2,334
$3,134
$4,070
$4,024
$3,942
$3,848
$3,716
$3,580
$3,379
$3,145
$2,885
$2,599
$2,293
$1,976
$1,661
$1,379
$1,139
$938
$771
$633
$521
$432
$361
$304
$256
$213
$178
$149
$46,400
aThe SCC values are dollar-year and emissions-year specific. Note that annual data extend to 2052 for the 2017MY
and to 2060 for the 2025MY. These data are not shown but are included in the NPV values.

-------
                                                 2017 Draft Regulatory Impact Analysis
 Table 7.1-6 Undiscounted Annual Upstream and Downstream CO2 Benefits for the from
2.5% (Average SCC) SCC Value, CO2 Benefits Discounted back tothe 1st Year of each MY,
     and Sum of Values Across MYs, Model Year Analysis3 (Millions of 2009 dollars)
YEAR
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
2031
2032
2033
2034
2035
2036
2037
2038
2039
2040
2041
2042
2043
2044
2045
2046
2047
2048
2049
2050
NPV,
2.5%
MY
2017
$85
$84
$83
$81
$80
$78
$76
$72
$69
$65
$61
$55
$48
$41
$34
$28
$24
$19
$16
$13
$11
$9
$8
$7
$6
$5
$4
$3
$3
$2
$2
$2
$2
$2
$968
MY
2018
$0
$202
$201
$199
$196
$193
$188
$182
$175
$167
$158
$147
$134
$116
$97
$80
$66
$54
$44
$36
$29
$24
$20
$17
$14
$12
$10
$7
$6
$5
$5
$4
$4
$3
$2,310
MY
2019
$0
$0
$318
$316
$316
$310
$306
$298
$289
$278
$263
$251
$233
$213
$183
$152
$126
$103
$84
$68
$55
$44
$36
$30
$26
$21
$17
$15
$10
$9
$8
$7
$6
$5
$3,640
MY
2020
$0
$0
$0
$453
$454
$451
$445
$437
$427
$414
$397
$378
$358
$337
$305
$260
$217
$178
$145
$119
$96
$77
$62
$50
$41
$35
$29
$24
$20
$13
$12
$10
$9
$8
$5,190
MY
2021
$0
$0
$0
$0
$684
$680
$679
$666
$656
$640
$618
$594
$562
$537
$498
$451
$388
$324
$269
$221
$182
$149
$120
$98
$80
$67
$58
$48
$40
$34
$24
$21
$19
$17
$7,820
MY
2022
$0
$0
$0
$0
$0
$885
$882
$877
$864
$850
$825
$800
$766
$731
$691
$640
$580
$498
$417
$346
$285
$235
$192
$156
$127
$104
$87
$76
$62
$52
$45
$32
$28
$25
$10,100
MY
2023
$0
$0
$0
$0
$0
$0
$1,078
$1,072
$1,068
$1,051
$1,031
$1,005
$971
$936
$884
$834
$773
$699
$601
$504
$418
$345
$285
$233
$189
$154
$126
$106
$92
$76
$64
$55
$39
$35
$12,300
MY
2024
$0
$0
$0
$0
$0
$0
$0
$1,282
$1,277
$1,273
$1,248
$1,229
$1,193
$1,161
$1,109
$1,046
$986
$912
$825
$709
$595
$494
$407
$337
$275
$222
$181
$149
$125
$109
$90
$75
$65
$47
$14,600
MY
2025
$0
$0
$0
$0
$0
$0
$0
$0
$1,498
$1,492
$1,481
$1,458
$1,430
$1,399
$1,348
$1,286
$1,213
$1,141
$1,056
$955
$821
$688
$571
$471
$388
$317
$257
$209
$172
$144
$125
$104
$87
$75
$16,900
SUM
$85
$286
$602
$1,050
$1,730
$2,597
$3,654
$4,888
$6,324
$6,231
$6,082
$5,918
$5,696
$5,472
$5,149
$4,779
$4,372
$3,928
$3,457
$2,971
$2,491
$2,064
$1,701
$1,398
$1,145
$936
$769
$636
$530
$445
$374
$310
$258
$216
$73,800
"The SCC values are dollar-year and
and to 2060 for the 2025MY. These
                            emissions-year specific. Note that annual data extend to 2052 for the 2017MY
                           data are not shown but are included in the NPV values.
                                          7-9

-------
Chapter 7
  Table 7.1-7 Undiscounted Annual Upstream and Downstream CO2 Benefits for the 3%
(95th Percentile) SCC Value, CO2 Benefits Discounted back tothe 1st Year of each MY, and
       Sum of Values Across MYs, Model Year Analysis3 (Millions of 2009 dollars)
YEAR
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
2031
2032
2033
2034
2035
2036
2037
2038
2039
2040
2041
2042
2043
2044
2045
2046
2047
2048
2049
2050
NPV,
3%
MY
2017
$162
$161
$160
$158
$156
$152
$148
$142
$135
$129
$120
$110
$95
$81
$68
$57
$48
$39
$32
$26
$22
$19
$16
$14
$12
$10
$7
$7
$6
$5
$5
$4
$4
$3
$1,830
MY
2018
$0
$388
$387
$385
$382
$376
$368
$358
$345
$329
$312
$293
$266
$232
$194
$162
$133
$110
$89
$72
$59
$48
$40
$35
$29
$24
$21
$14
$13
$11
$10
$9
$8
$7
$4,380
MY
2019
$0
$0
$613
$611
$613
$604
$598
$584
$569
$549
$522
$498
$465
$426
$366
$306
$253
$207
$170
$138
$111
$90
$73
$61
$52
$43
$36
$30
$20
$18
$16
$14
$12
$11
$6,920
MY
2020
$0
$0
$0
$877
$881
$879
$870
$858
$841
$818
$787
$751
$714
$673
$610
$523
$436
$359
$294
$240
$194
$156
$126
$102
$84
$73
$59
$49
$42
$27
$24
$21
$19
$17
$9,910
MY
2021
$0
$0
$0
$0
$1,330
$1,326
$1,328
$1,308
$1,293
$1,264
$1,224
$1,180
$1,121
$1,072
$998
$906
$780
$654
$544
$448
$370
$303
$245
$200
$164
$137
$120
$99
$83
$71
$50
$44
$39
$35
$15,000
MY
2022
$0
$0
$0
$0
$0
$1,725
$1,726
$1,723
$1,701
$1,680
$1,636
$1,590
$1,527
$1,461
$1,383
$1,285
$1,167
$1,004
$843
$702
$579
$479
$392
$318
$260
$213
$179
$156
$129
$108
$94
$67
$59
$52
$19,400
MY
2023
$0
$0
$0
$0
$0
$0
$2,110
$2,104
$2,104
$2,077
$2,043
$1,997
$1,934
$1,871
$1,772
$1,675
$1,556
$1,409
$1,215
$1,021
$849
$701
$581
$476
$386
$316
$260
$218
$190
$158
$132
$114
$82
$73
$23,600
MY
2024
$0
$0
$0
$0
$0
$0
$0
$2,517
$2,516
$2,515
$2,473
$2,442
$2,377
$2,320
$2,221
$2,101
$1,985
$1,840
$1,668
$1,438
$1,208
$1,005
$830
$687
$562
$457
$373
$307
$258
$225
$187
$157
$135
$98
$28,100
MY
2025
$0
$0
$0
$0
$0
$0
$0
$0
$2,951
$2,948
$2,935
$2,898
$2,850
$2,796
$2,701
$2,583
$2,442
$2,302
$2,135
$1,936
$1,667
$1,400
$1,165
$962
$796
$651
$529
$432
$356
$299
$261
$216
$182
$157
$32,700
SUM
$162
$549
$1,160
$2,031
$3,361
$5,064
$7,149
$9,595
$12,455
$12,309
$12,052
$11,759
$11,350
$10,932
$10,314
$9,597
$8,800
$7,925
$6,990
$6,021
$5,059
$4,200
$3,468
$2,855
$2,345
$1,924
$1,584
$1,313
$1,097
$923
$778
$647
$540
$452
$142,000
a The SCC values are dollar-year and
and to 2060 for the 2025MY. These
emissions-year specific. Note that annual data extend to 2052 for the 2017MY
data are not shown but are included in the NPV values.
7.2 Summary of Costs and Benefits

       In this section, EPA presents a summary of costs, benefits, and net benefits of the
proposed program.  Table 7.2-1  shows the estimated annual monetized costs of the proposed
program for the indicated calendar years.  The table also shows the net present values of those
                                          7-10

-------
                                                   2017 Draft Regulatory Impact Analysis
                                                                                  LLLLL
costs for the calendar years 2012-2050 using both 3 percent and 7 percent discount rates.
Table 7.2-2 shows the estimated annual monetized fuel savings of the proposed program. The
table also shows the net present values of those fuel savings for the same calendar years using
both 3 percent and 7  percent discount rates. In this table, the aggregate value of fuel savings is
calculated using pre-tax fuel prices since savings in fuel taxes do not represent a reduction in the
value of economic resources utilized in producing and consuming fuel. Note that fuel savings
shown here result from reductions in fleet-wide fuel use. Thus, they grow over time as an
increasing fraction of the fleet meets the 2025 standards. Table 7.2-3 shows the annual
reductions in petroleum-based imports and the monetized energy security benefits of the
proposed program for the indicated calendar years.  The table also shows the net present values
of monetized energy  security benefits for the calendar years 2012-2050 using both 3 percent and
7 percent discount rates.

  Table 7.2-1  Undiscounted Annual Costs & Costs of the Proposed Program Discounted
               Back to 2012 at 3% and 7% Discount Rates (Millions, 2009$)a



Technology
Costs
2017


$2,300
2020


$8,470
2030


$35,700
2040


$39,800
2050


$44,600
NPV, Years
2012-2050, 3%
Discount Rate
$551,000
NPV, Years
2012-2050,7%
Discount Rate
$243,000
Note:
" Technology costs for separate light-duty vehicle segments can be found in Chapter 5 of this DRIA. Annual costs
shown are undiscounted values.
    Table 7.2-2 Undiscounted Annual Fuel Savings & Proposed Program Fuel Savings
         Discounted Back to 2012 at 3% and 7% Discount Rates (Millions, 2009$) a

Fuel Savings
(pre-tax)
2017
$570
2020
$7,060
2030
$85,800
2040
$144,000
2050
$187,000
NPV, Years
2012-2050, 3%
Discount Rate
$1,510,000
NPV, Years
2012-2050, 7%
Discount Rate
$579,000
Note:
a Fuel savings for separate light-duty vehicle segments can be found in Chapter 5 of this DRIA. Annual costs shown
are undiscounted values.
    For the estimation of the stream of costs and benefits, we assume that after implementation of the proposed
MY 2017-2025 standards, the 2025 standards apply to each year out to 2050.
                                           7-11

-------
Chapter 7	

 Table 7.2-3 Undiscounted Annual Energy Security Benefits, & Proposed Program Benefits
         Discounted back to 2012 at 3% and 7% Discount Rates (Millions, 2009$)a

Petroleum-based
imports reduced
(mmb)
Monetized
benefits
2017
4.4
$30
2020
51.5
$366
2030
579
$4,810
2040
914
$7,860
2050
1,083
$9,310
NPV, Years
2012-2050, 3%
Discount Rate

$81,500
NPV, Years
2012-2050,7%
Discount Rate

$31,500
Note:
a EPA developed estimates of energy security premiums (i.e., $^arrel of imported crude oil and finished petroleum
products) as result of the proposed rule for 2020, 2030, 2040 and 2050 using a method developed by the Oak Ridge
National Laboratory. The method and estimated premiums are discussed in detail in Chapter 4 of the Joint TSD
along with our approach for estimating the reductions in petroleum-based imports from the propose rule. EPA
linearly interpolated the premium values for the years 2017 through 2035, using the 2015 and 2035 values as
endpoints and the 2020, 2025, and 2030 values as midpoints. Since ORNL uses AEO 2011 forecasts that end in
2035, EPA assumed that the post-2035 energy security premium did not change through 2050. Annual costs shown
are undiscounted values.

       Table 7.2-4 presents estimated annual monetized benefits for the indicated calendar
years.  The table also shows the net present values of those benefits  for the calendar years 2012-
2050 using both 3 percent and 7 percent discount rates. The table shows the benefits of reduced
CC>2 emissions—and consequently the annual quantified benefits (i.e., total benefits)—for each
of four SCC values estimated by the interagency working group. As discussed above in  section
7.1  of this DRIA, there are some limitations to the SCC analysis, including the incomplete way
in which the integrated assessment models capture catastrophic and  non-catastrophic impacts,
their incomplete treatment of adaptation and technological change, uncertainty in the
extrapolation of damages to high temperatures, and assumptions regarding risk aversion.

       In addition, these monetized GHG benefits exclude the value of net reductions in non-
CC>2 GHG emissions (CH4, N2O, HFC)  expected under this action.  Although EPA has not
monetized the benefits of reductions in non-CC>2 GHGs, the value of these reductions should not
be interpreted as zero. Rather, the net reductions in non-CO2 GHGs will contribute to this
program's climate benefits, as explained in Chapter 6.4 of this DRIA.
                                           7-12

-------
                                                          2017 Draft Regulatory Impact Analysis
 Table 7.2-4 Monetized Undiscounted Annual Benefits & Benefits of the Proposed Program
           Discounted Back to 2012 at 3% and 7% Discount Rates (Millions, 2009$)

2017
2020
2030
2040
2050
NPV, Years
2012-2050,
3% Discount
Rate3
NPV, Years
2012-2050,
7% Discount
Rate3
Benefits of Reduced CO2 Emissions at each assumed SCC value b
5% (avg SCC)
3% (avg SCC)
2.5% (avg SCC)
3%(95th%ile)
Energy Security
Benefits
Accidents,
Congestion, Noise
Costs8
Increased Travel
Benefits
Refueling Time
Savings
PM2 5 Related Impacts
c,d,e
Non-CO2 GHG
Impactsf
$13
$53
$85
$162
$30
$66
$89
$25
$11
n/a
$171
$662
$1,050
$2,030
$366
$844
$1,090
$301
$150
n/a
$2,690
$9,040
$13,800
$27,600
$4,810
$9,960
$12,900
$3,780
$1,360
n/a
$5,490
$17,000
$25,300
$51,600
$7,860
$16,900
$23,600
$6,650
$2,190
n/a
$8,050
$23,000
$33,300
$69,800
$9,310
$22,000
$33,600
$8,800
$2,970
n/a
$32,800
$172,000
$292,000
$522,000
$81,500
$176,000
$244,000
$68,700
$23,800
n/a
$32,800
$172,000
$292,000
$522,000
$31,500
$67,700
$92,100
$26,200
$9,280
n/a
Total Annual Benefits at each assumed SCC value "
5% (avg SCC)
3% (avg SCC)
2.5% (avg SCC)
3% (95th %ile)
$101
$141
$173
$250
$1,240
$1,730
$2,120
$3,100
$15,600
$22,000
$26,700
$40,500
$29,000
$40,400
$48,700
$75,100
$40,700
$55,600
$65,900
$102,00
0
$275,000
$413,000
$534,000
$764,000
$124,000
$263,000
$384,000
$614,000
Notes:
" Net present value of reduced CO2 emissions is calculated differently than other benefits.  The same discount rate
used to discount the value of damages from future emissions (SCC at 5, 3, 2.5 percent) is used to calculate net
present value of SCC for internal consistency. Refer to the SCC TSD for more detail. Annual costs shown are
undiscounted values.
* DRIA Chapter 7.1 notes that SCC increases over time.  For the years 2012-2050, the SCC estimates range as
follows: for Average SCC at 5%:  $5-$16; for Average SCC at 3%: $23-$46; for Average SCC at 2.5%:  $38-$67;
and for 95th percentile SCC at 3%: $70-$ 140.
0 Note that the co-pollutant impacts associated with the proposed standards presented here do not include the full
complement of endpoints that, if quantified and monetized, would change the total monetized estimate of rule-
related impacts. Instead, the co-pollutant benefits are based on benefit-per-ton values that reflect only human health
impacts associated with reductions in PM2 5 exposure.  Ideally, human health and environmental benefits would be
based on changes in ambient PM2 5 and ozone as determined by full-scale air quality modeling. However,  EPA was
unable to conduct a full-scale air quality modeling analysis in time for the proposal.  We intend to more  fully capture
the co-pollutant benefits for the analysis of the final standards.
d The PM2 s-related benefits (derived from benefit-per-ton values) presented in this table are based on an estimate of
premature mortality derived from the ACS study (Pope etal., 2002). See Chapter6.3.1 ofthisDRIA. If the benefit-
per-ton estimates were based on the Six Cities study (Laden et al., 2006), the values would be nearly two-and-a-half
times larger. Id..
e The PM25-related benefits (derived from benefit-per-ton values) presented in this table assume a 3% discount rate
in the valuation of premature mortality to account for a twenty-year segmented cessation lag.  If a 7% discount rate
had been used, the values would be approximately 9% lower.
f The monetized GHG benefits presented in this analysis exclude the value of changes in non-CO2 GHG emissions
expected under this program as discussed above in section 7.1.  Although EPA has not monetized changes in non-
                                                 7-13

-------
Chapter 7	

CO2 GHGs, the value of any increases or reductions should not be interpreted as zero. We seek comment on a
method of quantifying non-CO2 GHG benefits in Section III.H.5 of the preamble.
g The values shown for Accidents, Congestion, and Noise are costs and are treated as negative values in the total
benefits.
       Table 7.2-5 presents estimated annual net benefits for the indicated calendar years.  The
table also shows the net present values of those net benefits for the calendar years 2012-2050
using both 3 percent and 7 percent discount rates.  The table includes the benefits of reduced CC>2
emissions (and consequently the annual net benefits) for each of four SCC values considered by
EPA.

 Table 7.2-5 Undiscounted Annual Monetized Net Benefits & Net Benefits of the Proposed
     Program Discounted Back to 2012 at 3% and 7% Discount Rates (Millions, 2009$)

Technology Costs
Fuel Savings
2017
$2,300
$570
2020
$8,470
$7,060
2030
$35,700
$85,800
2040
$39,800
$144,000
2050
$44,600
$187,000
NPV, 3%a
$551,000
$1,510,000
NPV, 7%a
$243,000
$579,000
Total Annual Benefits at each assumed SCC value b
5% (avg SCC)
3% (avg SCC)
2.5% (avg SCC)
3% (95th %ile)
$101
$141
$173
$250
$1,240
$1,730
$2,120
$3,100
$15,600
$22,000
$26,700
$40,500
$29,000
$40,400
$48,700
$75,100
$40,700
$55,600
$65,900
$102,000
$275,000
$413,000
$534,000
$764,000
$124,000
$263,000
$384,000
$614,000
Monetized Net Benefits at each assumed SCC value °
5% (avg SCC)
3% (avg SCC)
2.5% (avg SCC)
3% (95th %ile)
-$1,630
-$1,590
-$1,560
-$1,480
-$166
$325
$712
$1,690
$65,600
$72,000
$76,800
$90,500
$133,000
$144,000
$153,000
$179,000
$183,000
$198,000
$208,000
$244,000
$1,230,000
$1,370,000
$1,490,000
$1,720,000
$460,000
$599,000
$719,000
$950,000
Notes:
" Net present value of reduced CO2 emissions is calculated differently than other benefits. The same discount rate
used to discount the value of damages from future emissions (SCC at 5, 3, 2.5 percent) is used to calculate net
present value of SCC for internal consistency. Refer to the SCC TSD for more detail. Annual costs shown are
undiscounted values.
* DRIA Chapter 7.1 notes that SCC increases over time. For the years 2012-2050, the SCC estimates range as
follows: for Average SCC at 5%: $5-$16; for Average SCC at 3%:  $23-$46; for Average SCC at 2.5%: $38-$67;
and for 95th percentile SCC at 3%: $70-$ 140. DRIA Chapter 7.1 also presents these SCC estimates.
0 Net Benefits equal Fuel Savings minus Technology Costs plus Benefits.
        EPA also conducted a separate analysis of the total benefits over the model year
lifetimes of the 2017 through 2025 model year vehicles. In contrast to the calendar year analysis
presented above in Table 7.2-1 through Table 7.2-5, the model year lifetime analysis below
shows the impacts of the proposed program on vehicles produced during each of the model years
2017 through 2025 over the course of their expected lifetimes. The net societal benefits over the
full lifetimes of vehicles produced during each of the nine model years from 2017 through 2025
are shown in Table 7.2-6 and Tables 7.2-7 at both 3 percent and 7 percent discount rates,
respectively.
                                            7-14

-------
                                                         2017 Draft Regulatory Impact Analysis
     Table 7.2-6  Monetized Technology Costs, Fuel Savings, Benefits, and Net Benefits
   Associated with the Lifetimes of 2017-2025 Model Year Light-Duty Vehicles (Millions,
                                   2009$; 3% Discount Rate)11

Technology Costs
Fuel Savings (pre-
tax)
Energy Security
Benefits
Accidents,
Congestion, Noise
Costs f
Increased Travel
Benefits
Refueling Time
Savings
PM2 5 Related
Impacts"'4"
Non-CO2 GHG
Impacts8
2017
MY
$2,270
$6,060
$322
$721
$1,040
$262
$117
n/a
2018
MY
$4,590
$14,30
0
$763
$1,740
$2,480
$618
$302
n/a
2019
MY
$6,410
$22,400
$1,200
$2,740
$3,850
$967
$481
n/a
2020
MY
$8,340
$31,80
0
$1,710
$3,880
$5,380
$1,370
$692
n/a
2021
MY
$11,700
$47,300
$2,550
$5,600
$7,720
$2,040
$1,090
n/a
2022
MY
$19,100
$61,000
$3,310
$7,150
$9,770
$2,650
$1,210
n/a
2023
MY
$24,700
$73,700
$4,030
$8,560
$11,600
$3,230
$1,300
n/a
2024
MY
$30,300
$87,000
$4,790
$10,000
$13,600
$3,840
$1,380
n/a
2025
MY
$33,100
$100,00
0
$5,560
$11,500
$15,500
$4,470
$1,450
n/a
Sum
$140,000
$444,000
$24,200
$52,000
$70,900
$19,500
$8,020
n/a
Benefits of Reduced CO2 Emissions at each assumed SCC value "• b
5% (avg SCC)
3% (avg SCC)
2.5% (avg SCC)
3% (95th %ile)
$142
$598
$968
$1,830
$344
$1,430
$2,310
$4,380
$552
$2,260
$3,640
$6,920
$802
$3,240
$5,190
$9,910
$1,230
$4,900
$7,820
$15,000
$1,610
$6,350
$10,100
$19,400
$1,980
$7,730
$12,300
$23,600
$2,390
$9,200
$14,600
$28,100
$2,810
$10,700
$16,900
$32,700
$11,900
$46,400
$73,800
$142,000
Monetized Net Benefits at each assumed SCC value "• b
5% (avg SCC)
3% (avg SCC)
2.5% (avg SCC)
3% (95th %ile)
$4,960
$5,420
$5,790
$6,650
$12,50
0
$13,60
0
$14,50
0
$16,60
0
$20,300
$22,100
$23,400
$26,700
$29,50
0
$32,00
0
$33,90
0
$38,60
0
$44,600
$48,300
$51,200
$58,400
$53,300
$58,100
$61,800
$71,100
$62,600
$68,400
$72,900
$84,300
$72,600
$79,400
$84,800
$98,300
$85,700
$93,600
$99,800
$116,00
0
$386,000
$421,000
$448,000
$516,000
Notes:
a Net present value of reduced CO2 emissions is calculated differently than other benefits. The same discount rate
used to discount the value of damages from future emissions (SCC at 5, 3, 2.5 percent) is used to calculate net
present value of SCC for internal consistency.  Refer to the SCC TSD for more detail.
* DRIA Chapter 7.1 notes that SCC increases over time. For the years 2012-2050, the SCC estimates range as
follows: for Average SCC at 5%: $5-$16; for Average SCC at 3%: $23-$46; for Average SCC at 2.5%: $38-$67;
and for 95th percentile SCC at 3%:  $70-$ 140.  DRIA Chapter 7.1 also presents these SCC estimates.
c Note that the co-pollutant impacts associated with the proposed standards presented here do not include the full
complement of endpoints that, if quantified and monetized, would change the total monetized estimate of rule-
related impacts. Instead, the co-pollutant benefits are based on benefit-per-ton values that reflect only human health
impacts associated with reductions in PM2 5 exposure. Ideally, human health and environmental benefits would be
based on changes in ambient PM2 5 and ozone as determined by full-scale air quality modeling. However, EPA was
unable to conduct a full-scale air quality modeling analysis in time for the proposal.  We intend to more fully capture
the co-pollutant benefits for the analysis of the final standards.
d The PM2 s-related benefits (derived from benefit-per-ton values) presented in this table are based on an estimate of
premature mortality derived from the ACS study (Pope etal., 2002).  See Chapter6.3.1 ofthisDRIA.  If the benefit-
per-ton estimates were based on the Six Cities  study (Laden et al., 2006), the values would be nearly two-and-a-half
times larger. Id.
                                                7-15

-------
Chapter 7	

e The PM25-related benefits (derived from benefit-per-ton values) presented in this table assume a 3% discount rate
in the valuation of premature mortality to account for a twenty-year segmented cessation lag. If a 7% discount rate
had been used, the values would be approximately 9% lower
/The values shown for Accidents, Congestion, and Noise are costs and are treated as negative values in the net
benefits.
g The monetized GHG benefits presented in this analysis exclude the value of changes in non-CO2 GHG emissions
expected under this action as discussed above in section 7.1. Although EPA has not monetized changes in non-CO2
GHGs, the value of any increases or reductions should not be interpreted as zero. We seek comment on a method of
quantifying non-CO2 GHG benefits in Section III.H.5 of the preamble.
* Model year values are discounted to the first year of each model year; the "Sum" represents those discounted
values summed across model years.
     Table 7.2-7 Monetized Technology Costs, Fuel Savings, Benefits, and Net Benefits
   Associated with the Lifetimes of 2017-2025 Model Year Light-Duty Vehicles (Millions,
                                  2009$; 7% Discount Rate)11

Technology Costs
Fuel Savings (pre-
tax)
Energy Security
Benefits
Accidents,
Congestion, Noise
Costs f
Increased Travel
Benefits
Refueling Time
Savings
PM2 5 Related
Impacts''46
Non-CO2 GHG
Impacts8
2017
MY
$2,220
$4,720
$250
$562
$808
$203
$93
n/a
2018
MY
$4,500
$11,20
0
$593
$1,360
$1,930
$481
$240
n/a
2019
MY
$6,290
$17,500
$934
$2,140
$3,000
$754
$382
n/a
2020
MY
$8,190
$24,90
0
$1,330
$3,040
$4,190
$1,070
$551
n/a
2021
MY
$11,500
$37,000
$1,980
$4,390
$6,010
$1,590
$864
n/a
2022
MY
$18,700
$47,700
$2,580
$5,600
$7,620
$2,070
$964
n/a
2023
MY
$24,200
$57,700
$3,150
$6,720
$9,080
$2,520
$1,030
n/a
2024
MY
$29,700
$68,100
$3,750
$7,880
$10,600
$2,990
$1,100
n/a
2025
MY
$32,500
$78,700
$4,360
$9,060
$12,100
$3,480
$1,160
n/a
Sum
$138,000
$347,000
$18,900
$40,800
$55,300
$15,200
$6,390
n/a
Benefits of Reduced CO2 Emissions at each assumed SCC value "• b
5% (avg SCC)
3% (avg SCC)
2.5% (avg SCC)
3% (95th %ile)
$142
$598
$968
$1,830
$344
$1,430
$2,310
$4,380
$552
$2,260
$3,640
$6,920
$802
$3,240
$5,190
$9,910
$1,230
$4,900
$7,820
$15,000
$1,610
$6,350
$10,100
$19,400
$1,980
$7,730
$12,300
$23,600
$2,390
$9,200
$14,600
$28,100
$2,810
$10,700
$16,900
$32,700
$11,900
$46,400
$73,800
$142,000
Monetized Net Benefits at each assumed SCC value "• b
5% (avg SCC)
3% (avg SCC)
2.5% (avg SCC)
3%(95th%ile)
$3,420
$3,880
$4,250
$5,110
$8,920
$10,00
0
$10,90
0
$13,00
0
$14,700
$16,400
$17,800
$21,100
$21,60
0
$24,00
0
$26,00
0
$30,70
0
$32,800
$36,400
$39,400
$46,500
$38,200
$43,000
$46,700
$56,000
$44,500
$50,200
$54,800
$66,100
$51,300
$58,100
$63,500
$77,000
$61,100
$69,000
$75,200
$91,000
$277,000
$311,000
$338,000
$406,000
                                               7-16

-------
                                                       2017 Draft Regulatory Impact Analysis
Notes:
a Net present value of reduced CO2 emissions is calculated differently than other benefits. The same discount rate
used to discount the value of damages from future emissions (SCC at 5, 3, 2.5 percent) is used to calculate net
present value of SCC for internal consistency. Refer to the SCC TSD for more detail.
* DRIA Chapter 7.1 notes that SCC increases over time. For the years 2012-2050, the SCC estimates range as
follows: for Average SCC at 5%: $5-$16; for Average SCC at 3%: $23-$46; for Average SCC at 2.5%: $38-$67;
and for 95th percentile SCC at 3%:  $70-$ 140.  DRIA Chapter 7.1 also presents these SCC estimates.
0 Note that the co-pollutant impacts associated with the proposed standards presented here do not include the full
complement of endpoints that, if quantified and monetized, would change the total monetized estimate of rule-
related impacts. Instead, the co-pollutant benefits are based on benefit-per-ton values that reflect only human health
impacts associated with reductions in PM2 5 exposure.  Ideally, human health and environmental benefits would be
based on changes in ambient PM2 5 and ozone as determined by full-scale air quality modeling. However, EPA was
unable to conduct a full-scale air quality modeling analysis in time for the proposal. We intend to more fully capture
the co-pollutant benefits for the analysis of the final standards.
d The PM2 5-related benefits (derived from benefit-per-ton values) presented in this table are based on an estimate of
premature mortality derived from the ACS study (Pope etal., 2002). See Chapter6.3.1 ofthisDRIA.  If the benefit-
per-ton estimates were based on the Six Cities study (Laden et al., 2006), the values would be nearly two-and-a-half
times larger. Id.
e The PM2 ^-related benefits (derived from benefit-per-ton values) presented in this table assume a 3% discount rate
in the valuation of premature mortality to account for a twenty-year segmented cessation lag. If a 7% discount rate
had been used, the values would be approximately 9% lower.
/The values shown for Accidents, Congestion, and Noise are costs and are treated as negative values in the net
benefits.
g The monetized GHG benefits presented in this analysis exclude the value of changes in non-CO2 GHG emissions
expected under this action as discussed above in section 7.1.  Although EPA has not monetized changes in non-CO2
GHGs, the value of any increases or reductions should not be interpreted as zero. We seek comment on a method of
quantifying non-CO2 GHG benefits in Section III.H.5 of the preamble.
* Model year values are discounted to the first year of each model year; the "Sum" represents those discounted
values summed across model years.
                                           References

324 Docket ID EPA-HQ-OAR-2010-0799, Technical Support Document: Social Cost of Carbon
for Regulatory Impact Analysis Under Executive Order 12866, Interagency Working Group on
Social Cost of Carbon, with participation by Council of Economic Advisers, Council on
Environmental Quality, Department of Agriculture, Department of Commerce, Department of
Energy, Department of Transportation, Environmental Protection Agency, National Economic
Council, Office of Energy and Climate Change, Office of Management and Budget, Office of
Science and Technology Policy, and Department of Treasury (February 2010). Also available at
http://epa.gov/otaq/climate/regulations.htm

325 National Research Council (2009).  Hidden Costs of Energy: Unpriced Consequences of
Energy Production and Use. National Academies Press. See docket ID EPA-HQ-OAR-2010-
0799.
                                              7-17

-------

-------
                                                 2017 Draft Regulatory Impact Analysis
8      Vehicle Sales and Employment Impacts

8.1 Vehicle Sales Impacts

       8.1.1  Vehicle Sales Impacts and Payback Period

       Predicting the effects of this rule on vehicles entails comparing two competing effects.
On the one hand, as a result of this rule, the vehicles will become more expensive, which would,
by itself, be expected to discourage sales. On the other hand, the vehicles will have improved
fuel economy and thus lower operating costs, producing lower total costs over the life of
vehicles, which makes them more attractive to consumers. Which of these effects dominates for
potential vehicle buyers when they are considering a purchase will determine the effect on sales.
However, assessing the net effect of these two competing effects is complex and uncertain, as it
rests on how consumers value fuel  savings at the time of purchase and the extent to which
manufacturers and dealers reflect them in the purchase price.  The empirical literature does not
provide clear evidence on whether  consumers fully consider the value of fuel savings at the time
of purchase. It also generally does not speak to the efficiency of manufacturing and dealer
pricing decisions. Thus, for the proposal we do not provide quantified estimates of potential sales
impacts. Rather, we solicit comment on the issues raised here and on methods for estimating the
effect of this rule on vehicle sales.

       For years, consumers have been gaining experience with the benefits that accrue to them
from owning and operating vehicles with greater fuel efficiency. Many households already own
vehicles with a fairly wide range of fuel economy, and thus already have an opportunity to learn
about the value of fuel economy on their own. Among two-vehicle households, for example, the
least fuel-efficient vehicle averages just over 22 mpg (EPA test rating), and the range between
this and the fuel economy of their other vehicle averages nearly 7 mpg.  Among households that
own 3 or more vehicles, the typical range of the fuel economy they offer is much wider.
 Consumer demand may have shifted towards such vehicles, not only because of higher fuel
prices but also if many consumers are learning about the value of purchases based not only on
initial costs but also on the total cost of owning and operating a vehicle over its lifetime.  This
type of learning should continue before and during the model years affected by this rule,
particularly given the new fuel economy labels that clarify potential economic effects and should
therefore reinforce that learning..

       Today's proposed rule, combined with the new and easier-to-understand fuel economy
label required to be on all new vehicles beginning in 2012, may increase sales above baseline
levels by hastening this very type of consumer learning. As more consumers experience, as a
result of the rule, the savings in time and expense from owning more fuel efficient vehicles,
demand may shift yet further in the direction  of the vehicles mandated under the rule. This
social learning can take place both  within and across households, as consumers learn from one
another.

       First and most directly, the  time and fuel savings associated with operating more fuel
efficient vehicles may be more salient to individuals who own them, which might cause their
subsequent purchase decisions to shift closer to minimizing the total cost of ownership over the
lifetime of the vehicle.
                                          3-1

-------
Chapter 8	

       Second, this appreciation may spread across households through word of mouth and other
forms of communications.

       Third, as more motorists experience the time and fuel savings associated with greater fuel
efficiency, the price of used cars will better reflect such efficiency, further reducing the cost of
owning more efficient vehicles for the buyers of new vehicles (since the resale price will
increase).

       If these induced learning effects are strong,  the rule could potentially increase total
vehicle sales over time.  It is not possible to quantify these learning effects years in advance and
that effect may be speeded or slowed by other factors that enter into a consumer's valuation of
fuel efficiency in selecting vehicles.  The possibility that the rule will (after a lag for consumer
learning) increase sales need not rest on the assumption that automobile manufacturers are
failing to pursue profitable opportunities to supply the vehicles that consumers demand. In the
absence of the rule, no individual automobile manufacturer would find it profitable to move
toward the more efficient vehicles mandated under the rule.  In particular, no individual company
can fully internalize the future boost to demand resulting from the rule. If one company were to
make more efficient vehicles, counting on consumer learning to  enhance demand in the future,
that company would capture only a fraction of the extra sales so  generated, because the learning
at issue is not specific to any one company's fleet. Many of the extra sales would accrue to that
company's competitors.

       In other words, consumer learning about the benefits of fuel efficient vehicles involves
positive externalities (spillovers) from one company to the othersMMMMM.  These positive
externalities may lead to benefits for manufacturers as a whole.  We emphasize that this
discussion has been tentative and qualified. To be sure,  social  learning of related kinds has been
identified in a number of contexts.326 Comments are invited on the discussion offered here, with
particular reference to any relevant empirical findings.

       In previous rulemakings, EPA and NHTSA conducted vehicle sales analyses by
comparing the up-front costs of the vehicles with the present value of five years' worth of fuel
savings.  We assumed that the costs for the fuel-saving technologies would be passed along fully
to vehicle buyers in the vehicle prices.  The up-front vehicle costs were adjusted to take into
account several factors that would affect consumer costs:  the  increased sales tax that consumers
would  pay, the increase in insurance premiums, the increase in loan payments that buyers would
face, and a higher resale value, with all of these factors due to the higher up-front cost of the
vehicle.  Those calculations resulted in an adjusted  increase in costs to consumers.  We then
assumed that consumers considered the present value of five years of fuel savings in their vehicle
purchase, which is consistent with the length of a typical new  light-duty vehicle loan, and is
MMMMM Industrywide positive spillovers of this type are hardly unique to this situation. In many industries,
companies form trade associations to promote industry-wide public goods. For example, merchants in a given locale
may band together to promote tourism in that locale. Antitrust law recognizes that this type of coordination can
increase output.


                                            8-2

-------
                                                    2017 Draft Regulatory Impact Analysis
similar to the average time that a new vehicle purchaser holds onto the vehicle.NNNNN  The
present value of fuel savings was subtracted from technology costs to get a net effect on vehicle
cost of ownership. We then used a short-run demand elasticity of-1 to convert a change in price
into a change in quantity demanded of vehicles.00000 An elasticity of-1 means that a 1%
increase in price leads to a 1% reduction in quantity sold.

       We do not here present a vehicle sales analysis using this approach. This rule takes effect
for MY 2017-2025.  In the intervening years, it is possible that the assumptions underlying this
analysis, as well as market conditions, might change. Instead, we present a payback period
analysis to estimate the number of years of fuel savings  needed to recover the up-front costs of
the new technologies.  In other words, the payback period identifies the break-even point for new
vehicle buyers. The calculation of the payback period is discussed in DRIA Chapter 5.3. Table
8.1-1 shows the estimated payback period for MY 2021  and 2025.  We present MY 2021
because it is the last year before the mid-term review impacts, if any, will take place, and MY
2025 because it is the last year of the program.  The payback period in 2021 is shorter than that
in 2025, because the technologies required to meet the proposed MY 2021 standards are more
cost-effective than those for MY 2025.  In all cases, the  payback periods are less than 4 years.

       Table 8.1-1 Estimated Payback Period for Model Years 2021 and 2025 (Years)
Model
Year

2021
2025
Estimated
Payback Period
for Cash
Purchase, 3%
Discount Rate
2.7
3.7
Estimated
Payback Period
for Cash
Purchase, 7%
Discount Rate
2.9
3.9
Estimated
Payback Period
for Purchase on
Credit, 3%
Discount Rate
2.9
3.9
Estimated
Payback Period
for Purchase on
Credit, 7%
Discount Rate
2.8
3.9
       We welcome comments on all aspects of this discussion, including the full range of
considerations and assumptions which influence market behavior and outcomes; and associated
uncertainties. We welcome comments on the methodology described here for quantitative
estimates of the effects of this proposal on sales and its appropriateness for this rulemaking; we
also welcome proposals for other methods.
NNTOIN In this proposal, the 5-year payback assumption corresponds to an assumption that vehicle buyers take into
account between 30 and 50 percent of the present value of lifetime vehicle fuel savings (with the variation
depending on discount rate, model year, and car vs. truck).
ooooo
     por a Durable good such as an auto, the elasticity may be smaller in the long run:  though people may be able
to change the timing of their purchase when price changes in the short run, they must eventually make the
investment. We request comment on whether or when a long-run elasticity should be used for a rule that phases in
over time, as well as how to find good estimates for the long -run elasticity.
                                            8-3

-------
Chapter 8	

       We here provide further detail on some of the assumptions included in the payback
analysis. We seek comment on all these factors as well.  The analysis starts with the increase in
costs estimated by the OMEGA model. We assume that these costs are fully passed along to
consumers.  This assumption is appropriate for cost increases in perfectly competitive markets.
In less than perfectly competitive markets, though, it is possible that the cost increase is split
between consumers and automakers, and possibly suppliers, and the price may not increase as
much as costs.327  Thus, the assumption of full cost pass-through is possibly an overestimate.

       The next step in the analysis is to adjust this cost increase for other effects on the
consumer.  The higher vehicle price is likely to lead to an increase in sales tax, insurance, and
vehicle financing costs.

       The increase in insurance costs is estimated from the average value of collision plus
comprehensive insurance as  a proportion of average new vehicle price. Collision plus
comprehensive insurance is the portion of insurance costs that depend on vehicle value. The
Insurance Information Institute328 provides the average value of collision plus comprehensive
insurance in 2008 as $432. The average value of a new vehicle in 2008, according to the U.S.
Department of Energy, was $23,334.329 Dividing the insurance cost by the average price of a
new vehicle gives the proportion of comprehensive plus collision insurance as  1.85 percent of
the  price of a vehicle.  If this same proportion holds for the increase in price of a vehicle, then
insurance costs should go up by 1.85 percent of the increase in vehicle cost.  We use information
on depreciation of vehicle value from the  same U.S. Department of Energy report to estimate a
reduction in insurance costs, due to reduction in the estimated value of the vehicle, over 9 years.

       Calculating the average increase in sales tax starts with the vehicle sales tax for each state
in 2006, the most recent source identified.330 The sales tax per state was then multiplied by the
2010 population of the state;331 those values were summed and divided by total U.S. population,
to give a population-weighted sales tax. That estimate of the state sales taxes for vehicles in the
U.S. is 5.3 percent. This value is assumed to be a one-time cost incurred when the vehicle is
purchased.

       As of July, 2011, the national average interest rate for a 5 year new car loan was 5.51
percent.332 We use this loan rate to calculate the increase in vehicle costs due to financing a loan.

       8.1.2  Consumer Vehicle Choice Modelingppppp

       An alternative to the  vehicle sales  analysis discussed above is the use of consumer
vehicle choice models. In this section we describe some of the consumer vehicle choice models
EPA has reviewed in the literature, and we describe the  models' results and limitations that we
have identified. The evidence from consumer vehicle choice models indicates  a huge range of
estimates for consumers' willingness to pay for additional fuel economy.  Because consumer
surplus estimates from consumer vehicle choice models depend critically on this value, we
ppppp This section is drawn heavily from Helfand, Gloria, and Ann Wolverton, "Evaluating the Consumer Response
to Fuel Economy:  A Review of the Literature." International Review of Environmental and Resource Economics 5
(2011):  103-146.


                                           8-4

-------
                                                  2017 Draft Regulatory Impact Analysis
would consider any consumer surplus estimates of the effect of our rule from such models to be
unreliable.  In addition, the predictive ability of consumer vehicle choice models may be limited.
While vehicle choice models are based on sales of existing vehicles, vehicle models are likely to
change, both independently and in response to this rule. The models may not predict well in
response to these changes.  Instead, we compare the value of the fuel savings associated with this
rule with the increase in technology costs. EPA will continue its efforts to review the literature,
but, given the known limitations and uncertainties of vehicle choice models, EPA has not
conducted an analysis using these models for this proposal.

       This rule will lead automakers to change characteristics - in particular, the fuel  economy
- of the vehicles they produce. These changes will affect the cost of manufacturing the vehicle;
as a result, the prices of the vehicles will also change.

       In response to these changes, the number and types of vehicles sold is likely to change.
When consumers buy vehicles, they consider both their personal characteristics (such as age,
family composition, income,  and their vehicle needs)  and the characteristics of vehicles (e.g.,
vehicle size, fuel economy, and price). In response to the changes in vehicle characteristics,
consumers will reconsider their purchases. Increases  in fuel economy are likely to be attractive
to consumers, but increases in price,  as well as any detrimental changes in other vehicle
characteristics, may be deterrents to purchase.  As a result, consumers may choose a different
vehicle than they would have purchased in the absence of the rule. The changes in prices and
vehicle characteristics are likely to influence consumers on multiple market scales: the total
number of new vehicles sold; the mix of new vehicles sold; and the effects of the  sales  on the
used vehicle market.

       Consumer vehicle choice modeling (CCM) is  a method used to predict what vehicles
consumers will purchase based on vehicle characteristics and prices. In principle, it should
produce more accurate estimates of compliance costs  compared to models that hold fleet mix
constant, since it predicts changes in the fleet mix that can affect compliance costs. It can also be
used to measure changes in consumer surplus, the benefit that consumers perceive from a good
over and above the purchase  price. (Consumer surplus is the difference between what consumers
would be willing to pay for a good, represented by the demand curve, and the amount they
actually pay. For instance, if a consumer were willing to pay $30,000 for a new vehicle, but
ended up paying $25,000, the $5000 difference is consumer surplus.)

       A number of consumer vehicle choice models have been developed. They vary in the
methods used, the data sources, the factors included in the models, the research questions they
are designed to answer, and the results of the models related to the effects of fuel economy on
consumer decisions.  This section will give some background on these differences among the
models.

              8.1.2.1       Methods

       Consumer choice models (CCMs) of vehicle purchases typically use a form of discrete
choice modeling. Discrete choice models seek to explain discrete rather than continuous
decisions.  An example of a continuous decision is how many pounds of food a farm might grow:
the pounds of food can take any numerical value.  Discrete decisions can take only a limited set
                                           3-5

-------
Chapter 8	

of values. The decision to purchase a vehicle, for instance, can only take two values, yes or no.
Vehicle purchases are typically modeled as discrete choices, where the choice is whether to
purchase a specified vehicle.  The result of these models is a prediction of the probability that a
consumer will purchase a specified vehicle. A minor variant on discrete choice models estimates
the market share (a continuous variable between 0 and 1) for each vehicle. Because the market
share is, essentially, the probability that consumers will purchase a specific vehicle, these
approaches are similar in process; they differ mostly in the kinds of data that they use.

       The primary methods used to model vehicle choices are nested logit and mixed
logit.QQQQQ In a nested logit, the model is structured in layers.  For instance, the first layer may
be the choice of whether to buy a new or used vehicle. Given that the person chooses a new
vehicle, the second layer may be whether to buy a car or a truck. Given that the person chooses  a
car, the third layer may be the choice among an economy, midsize, or luxury car. Examples of
nested logit models  include Goldberg,333 Greene et al.,334 and McManus.335

       In a mixed logit, personal characteristics of consumers play a larger role than in nested
logit. While nested logit can look at the effects of a change in average consumer characteristics,
mixed logit allows consideration of the effects of the distribution of consumer characteristics.
As a result, mixed logit can be used to examine the distributional effects on various
socioeconomic groups, which nested logit is not designed to do. Examples of mixed logit
models include Berry, Levinsohn, and Pakes,336 Bento et al.,337 and Train and Winston.338

       While  discrete choice modeling appears to be the primary method for consumer choice
modeling, others (such as Kleit339 and Austin and Dinan340) have used a matrix of demand
elasticities to estimate the effects of changes in cost. The discrete choice models can produce
such elasticities. Kleit as well as Austin and Dinan used the elasticities from an internal GM
vehicle choice model.

               8.1.2.2       Data Sources

       The predictions of vehicle purchases from CCMs are based on consumer and vehicle
characteristics. The CCMs identify the effects of changing the  characteristics on the purchase
decisions. These effects are typically called the parameters or coefficients of the models. For
instance, the model  parameters might predict that an increase in a person's income  of 10% would
increase the probability of her purchasing vehicle A by 5%, and decrease the probability of her
purchasing vehicle B by 10%.

       The parameters in CCMs can be developed either from original data sources (estimated
models), or using values taken from other studies (calibrated models).

       Estimated models use  datasets on consumer purchase patterns, consumer characteristics,
and vehicle characteristics to develop their original sets of parameters.  The datasets used in
these studies sometimes come from surveys of individuals' behaviors.341  Because they draw on
QQQQQ Logit refers to a statistical analysis method used for analyzing the factors that affect discrete choices (i.e.,
yes/no decisions or the choice among a countable number of options).

-------
                                                  2017 Draft Regulatory Impact Analysis
the behavior of individuals, they provide what is sometimes called micro-level data.  Other
studies, that estimate market shares instead of discrete purchase decisions, use aggregated data
that can cover long time periods.342

       Calibrated models rely on existing studies for their parameters. Researchers may draw on
results from a number of estimated models, or even from research other than CCM, to choose the
parameters of the models.  The Fuel Economy Regulatory Analysis Model developed for the
Energy Information Administration343 and the New Vehicle Market Model developed by NERA
Economic Consulting344 are examples of calibrated models.

              8.1.2.3       Factors Included  in the Models

       Consumer choice models vary in their complexity and levels of analysis. Some focus
only on the new vehicle market;345 others consider the choice between new vehicles and an
outside good (possibly including a used vehicle);346 others explicitly consider the relationship
between the new and used vehicle markets.347  Some models include consideration of vehicle
miles traveled,348 though most do not.

       The models vary in their inclusion of both consumer and vehicle information.  One model
includes only vehicle price and the distribution of income in the population influencing
choice;349 others include varying numbers and kinds of vehicle and consumer attributes.

       Some models include only the consumer side of the vehicle market;350 others seek to
represent both consumer and producer decisions.351  Models that include only the consumer side
are suitable for reflecting consumer choices, but they do not allow for revisions of vehicle
characteristics in response to consumer preferences. Including producer behavior allows for
vehicle characteristics such as price and fuel economy to be the result of market forces rather
than characteristics of the existing fleet. For instance, in the context of "feebates" (subsidizing
fuel-efficient cars with revenue collected by taxing fuel-inefficient vehicles), Greene et al.
estimated that 95% of the increase in fuel economy was due to addition of technology rather than
changes in vehicles sold.352 Including auto maker response is a complex exercise.  Auto makers
are commonly considered to have market power; they can influence the prices that consumers
pay to increase their profits. As a result, the price increases that consumers face may reflect
strategic factors that could make them higher or lower than the technology costs. In addition,
auto makers may seek to influence consumer preferences through marketing and advertising.353
Even those vehicle choice models that include a producer model may not include much detail,
due to computational limits: it is unusual for models to allow both buyers and producers to
choose one vehicle characteristic, much less multiple characteristics.354

              8.1.2.4       Research Questions for the Models

       Consumer choice models have been developed to analyze many different research and
policy questions. In part, these models have been developed to advance the state of economic
modeling. The work of Berry, Levinsohn, and  Pakes,355 for instance, is often cited outside the
motor vehicle context for its incorporation of multiple new modeling issues into its framework.
In addition, because the vehicle sector is a major part of the U.S. economy and involved in many
public policy discussions, research questions cover a wide gamut. These topics have included
                                           3-7

-------
Chapter 8	

the effects of voluntary export restraints on Japanese vehicles compared to tariffs and quotas,356
the market acceptability of alternative-fuel vehicles,357 the effects of introduction and exit of
vehicles from markets,358 causes of the decline in market shares of U.S. automakers,359 and the
effects of gasoline taxes360 and "feebates"361 (subsidizing fuel-efficient cars with revenue
collected by taxing fuel-inefficient vehicles).

               8.1.2.5       The Effect of Fuel Economy on Consumer Decisions

       Consumer vehicle choice models typically consider the  effect of fuel economy on vehicle
purchase decisions.  It can appear in various forms.

       Some models362 incorporate fuel economy through its effects on the cost of owning a
vehicle. With assumptions on the number of miles traveled per year and the cost of fuel, it is
possible to estimate the fuel savings (and perhaps other operating costs) associated with a more
fuel-efficient vehicle.  Those savings are considered to reduce the cost of owning a vehicle:
effectively, they reduce the purchase price. This approach relies on the assumption that, when
purchasing vehicles, consumers can estimate the fuel savings that they expect to receive from a
more fuel-efficient vehicle and consider the savings equivalent  to a reduction in purchase price.
Turrentine and Kurani363 question this assumption; they find, in fact, that consumers do not make
this calculation when they purchase a vehicle.  The question remains, then, how or whether
consumers take fuel economy into account when they purchase their vehicles.

       Most estimated consumer choice models, instead of making assumptions about how
consumers incorporate fuel economy into their decisions, use data on consumer behavior to
identify that effect.  In some models, miles per gallon is one of the vehicle characteristics
included to explain purchase decisions. Other models use fuel consumption per mile, the inverse
of miles per gallon,  as a measure:364  since consumers pay for gallons  of fuel, then this measure
can assess fuel savings relatively directly.365  Yet other models  multiply fuel consumption per
mile by the cost of fuel to get the cost of driving a mile,366 or they divide fuel economy by fuel
cost  to get miles per dollar.367 It is worth  noting that these last two measures assume that
consumers respond the same way to an increase in fuel economy as they do to a decrease in the
price of fuel when each has the same effect on cost per mile driven.RRRRR On the one hand,
while this assumption does not rely on as complex a calculation as the present value of fuel
savings that Turrentine and Kurani examined, it suggests a calculating consumer.  On the other
hand, using a form of cost per mile is a way to recognize the role of fuel prices in consumers'
purchase of fuel economy: recent research368 presents results that higher fuel prices play a major
role in that decision.

       Greene and Liu,369 in a paper published in  1988, reviewed 10 papers using consumer
vehicle choice models and estimated for each one how much consumers would be willing to pay
1111111111 Likewise, these measures assume consumers respond the same way to increases and decreases in cost per mile
of driving, as well as if those increases and decreases are large shocks rather than small, gradual changes. The issue
of potential asymmetric consumer response to increased fuel efficiency compared to other types of changes to the
cost of driving also arises and is discussed in the context of the VMT rebound effect (see Section III.H.4 of the
Preamble and Chapter 4.2.5.2 of the TSD).

-------
                                                  2017 Draft Regulatory Impact Analysis
at time of purchase to reduce vehicle operating costs by $1 per year. They found that people
were willing to pay between $0.74 and $25.97 for a $1 decrease in annual operating costs for a
vehicle. This is clearly a very wide range: while the lowest estimate suggests that people are not
willing to pay $1 once to get $1 per year reduced costs of operating their vehicles, the maximum
suggests a willingness to pay 35 times as high.  For comparison, the present value of saving $1
per year for 15 years at a 3% discount rate is $11.94, while a 7% discount rate produces a present
value of $8.78.  While this study is quite old, it suggests that, at least as of that time, consumer
vehicle choice models produced widely varying estimates of the value of reduced vehicle
operating costs.

       A newer literature review from David Greene370 suggests continued lack of convergence
on the value of increased fuel economy to consumers.  Of 27 studies, willingness to pay for fuel
economy as a percent of the expected value of fuel savings varied from  highly positive to highly
negative. Significant numbers of studies found that consumers overvalued fuel economy,
undervalued fuel economy, or roughly valued fuel economy correctly relative to fuel savings.
Part of the difficulty may be, as these papers note, that fuel economy may be correlated (either
positively or negatively) with other vehicle attributes, such as size, power, or quality, not all of
which may be included in the analyses; as a result, "fuel economy" may in fact represent several
characteristics at the same time.  Indeed, Gramlich371 includes both fuel cost (dollars per mile)
and miles per gallon in his analysis, with the argument that miles per gallon measures other
undesirable quality attributes, while fuel cost picks up the consumer's demand for improved fuel
economy. Greene finds that, while some of the variation  may be explainable due to issues in
some of the studies, the variation shows up in studies that appear to be well conducted.  As a
result, further work needs to be conducted before it is possible to identify the role of fuel
economy in consumer purchase decisions.

       Some studies372 argue that automakers could increase profits by  increasing fuel economy
because the amount that consumers are willing to pay for increased fuel economy outweighs the
costs of that improvement. Other studies373 have found that increasing fuel economy standards
imposes welfare losses on consumers and producers, because consumers should already be
buying as much fuel economy as they want.  In the course of reaching this result, though, at least
one of these studies374 notes that its baseline model implies  that consumers are willing to buy
more fuel economy than producers have provided; they have to adjust their model to eliminate
these "negative-cost" fuel economy improvements.

       The models do not appear to yield very consistent results on the  role of fuel economy in
consumer and producer decisions.

               8.1.2.6      Why Market Outcomes May Not Reflect Full Appreciation for
               Fuel Economy which Pays for Itself

       A detailed and wide ranging literature attempts to explain why market outcomes for
energy-using products appear to reflect under-investment in energy saving technologies that - at
least using a present value calculation based on engineering estimates - appear to pay for
themselves.  Existing research does not provide a definitive answer to this question. Potential
explanations are bounded by two scenarios. On the one hand, purely private benefits of fuel
economy (fuel savings, time savings, increases in driving time) must be accompanied by private
                                          8-9

-------
Chapter 8	

losses of the same magnitude. However, if there is no such private loss, or if it is small or
insignificant, then there is a market or behavioral failure.

       This disconnect between net present value estimates of energy-conserving cost savings
and what consumers actually spend on energy conservation is often referred to as the Energy
Paradox,375 since consumers appear to undervalue a wide range of investments in energy
conservation. There are many possible explanations for the paradox discussed in the
literature.376 Some explanations point to costs or aspects of consumer decision-making
unaccounted for in a simple present value calculation, while others point to potential behavioral
or market failures.  There is little empirical literature to help the analyst determine which
combination of hypothesis offers the most credible explanation.  Some possibilities include:

       •  Consumers put little weight on benefits from fuel economy in the future and show
          high discount rates;

       •  Consumers do not find the benefits from fuel economy to be sufficiently salient at the
          time of purchase, even if it would be in consumers' economic interest to take account
          of those benefits;

       •  Consumers consider other attributes more important than fuel economy at the time of
          vehicle purchase, especially if fuel economy is a relatively "shrouded" attribute;

       •  Consumers have difficulty in calculating expected fuel savings;

       •  Consumers may use imprecise rules of thumb when deciding how much fuel
          economy to purchase;

       •  Consumers might associate higher fuel economy with inexpensive, less well designed
          vehicles;

       •  Fuel savings in the future are uncertain, while at the time of purchase the increased
          costs of fuel-saving technologies are certain and immediate;

       •  Consumers may not be able to find the vehicles they want with improved fuel
          economy;

       •  The level of cost savings may be affected by the underlying reasons for the gap:
          factors such as transactions costs and differences in quality may not be adequately
          measured;377

       •  There is likely to be variation among consumers in the benefits they get from
          improved fuel economy, due to different miles driven and driving styles;

       •  Consumers may give particular weight to the losses associated with upfront costs, and
          less so to the costs over time (a version of the phenomenon of "myopic loss
          aversion").
                                          8-10

-------
                                                   2017 Draft Regulatory Impact Analysis
       The extent to which fuel economy is optimized relative to other, potentially more salient
vehicle attributes (such as engine horsepower and seating capacity) in market outcomes for new
vehicles remains an important area of uncertainty.  There are significant challenges involved in
effectively interpreting and anticipating consumer preferences for various vehicle attributes and
amenities. There are significant lead times to market, potential return to scale limits on the range
of options provided for a given attribute or amenity, market transaction frictional factors, and
other factors inherent to the nature of these costly durable goods which may contribute to
imperfect satisfaction of market demand for fuel economy among a highly heterogeneous
customer base.  . Both sides of the market would be expected to attempt to maximize the utility
they gain from these transactions, they presumably rely heavily in their calculations on the
uncertain benefits of savings from fuel economy improvements, and yet market outcomes may
still appear to reflect potential foregone opportunities to increase utility. We remain interested in
these market dynamics, their underlying causes, and their potential significance for assessing the
potential incremental effects of pollution control standards. We welcome comments on any
aspect of this discussion.

               8.1.2.7      Modeling Electric Vehicles and Other New  Vehicles

       Modeling the introduction of new vehicles can be a greater challenge than modeling the
existing vehicle market, because the modeler does not have data on how many of the new
vehicles consumers buy. Nevertheless, it can be possible to estimate the effects of new vehicle
introduction by identifying characteristics for the new vehicles and using those in a vehicle
choice model.  For instance, as discussed above, the models can estimate effects on the vehicle
market when vehicles change their fuel economy or price. If the model incorporates other
vehicle attributes important to the new vehicles, such as size, performance, or range, then the
effect of the introduction can be modeled by applying the parameters for those features to the
new vehicle characteristics.

       As discussed above, some models rely on vehicle price as the primary or only
explanatory  variable. Even in these  models, it is possible, with some additional information, to
consider the effects of new vehicle introduction. The first step is to find a vehicle similar on as
many dimensions as possible to the new vehicle. For instance, if the change is to create an
electric vehicle (EV) version of an existing model, then the existing model serves as the base
vehicle. Next, it is necessary to measure the changes in vehicle attributes of interest to potential
vehicle buyers.  For an EV, changes in vehicle driving range and cost of fueling may be two such
attributes. The next requirement is information on the value to consumers of the attributes that
change between the new and the base vehicle.  Multiplying the value for that attribute by the
change in the attribute provides an estimate of the benefit or cost associated with changing that
characteristic.  That amount can then be added to or subtracted from the vehicle purchase price to
give an adjusted purchase price reflecting the changed characteristic. This procedure is just an
extension of the approach, discussed above, used to incorporate fuel economy improvements into
vehicle choice models, by calculating future fuel savings and subtracting them (either in whole
or a fraction) from  vehicle purchase  price.

       Incorporating new vehicles into a vehicle choice model, then, requires estimates of the
changes in key attributes from conventional vehicles, and estimates of the value, also called the
willingness to pay (WTP), that consumers put on those attributes.
                                          8-11

-------
Chapter 8	

       Electric vehicles (EVs) will have a number of changes in vehicle characteristics from any
baseline model.  EVs are likely to have a smaller driving range between refuelings than
conventional vehicles, due to the large battery capacity needed to increase range.  The ability to
recharge at home may be a convenient, desirable feature for people who have garages with
electric hookups, but not for people who park on the street.  If an infrastructure develops for
recharging vehicles with the convenience approaching that of buying gasoline, range or home
recharging may become less of a barrier to purchase.  The reduced tailpipe emissions and
reduced noise may be attractive features to some consumers.sssss They may have different
performance or storage capacity. If sufficient data were available, the changes in these
attributes, combined with WTP for each of the attributes, could be used to adjust the purchase
price of the baseline vehicle to estimate consumers' WTP for the electric version of a vehicle.
Greene (2001), for instance, used this approach for a model that simulates choice, not only for
EVs, but also for other alternative-fuel vehicles.378 In that model, he considers only one base
vehicle, a passenger car, but considers the effect on WTP of fuel cost per mile, range,
acceleration, and several other vehicle attributes.

       Vehicle driving range has received attention because of the current paucity of recharging
infrastructure: if the driver of an EV gets low on fuel, it may be difficult to find a place to
recharge.  Because range appears to be a major factor in EV acceptability, it is starting to draw
attention in the research community.

       In several studies, researchers have used stated preference conjoint analysis to estimate
the effect of vehicle range on consumer vehicle choice. In a conjoint analysis, consumers are
given a choice between several vehicles with different attributes.  One choice might be, for
instance, between a baseline car and another car with higher range and a higher purchase price.
The choices that consumers make (e.g., how much higher does the purchase price have to be for
the consumer not to choose more range?) provide data that can be used to estimate the role of
vehicle attributes in the consumer's choice. Stated preference analysis is sometimes considered
less reliable than actual market behavior, because what people  say they will do in hypothetical
situations may not match what they would do in actual situations.  On the other hand, stated
preference methods can be used to study goods where market data do not exist, such as future
market products undergoing development (marketing studies often use stated preference
methods), or environmental goods. Because electric vehicles are not in widespread enough use
for market studies, stated preference studies are, at this point, one of the few options to examine
consumer behavior  relating to these vehicles.

       Table 8.1-2  summarizes results from several conjoint studies that include the effects of
extending range (in the table, from 150 to 300 miles, to present standardized results). Variation
of results in the table is from income or other demographic factors,  not from confidence
intervals.  The results suggest that the value of additional range varies among consumers, and the
amount of that variation is changing (perhaps shrinking) in more recent studies.
sssss For instance, Hidrue et al. (Hidrue, Michael K., George R. Parsons, Willett Kempton, and Meryl P. Gardner.
"Willingness to Pay for Electric Vehicles and their Attributes." Resource and Energy Economics 33(3) (2011):
686-705) find that some consumers are willing to pay $5100 more for vehicles with 95% lower emissions than the
vehicles they otherwise aim to purchase.


                                           8-12

-------
                                                       2017 Draft Regulatory Impact Analysis
   Table 8.1-2 Willingness to Pay for Increasing Range Calculated from Various Studies
             Study (Date)
  Value of extending range
from 150 to 300 miles (dollar
            year)
Value of additional
  range in 2009$a
Bunch etal. (1993)b
                $7,600(1991$)
             $11,100
Kavelek (1996) for California Energy
Commission0
       $2600 - $41,900 (1993$)
     $3700 - $58,700
Resource Systems Group (2009) for
California Energy Commissiond
        $2900 - $7500 (2009$)
       $2900 - $7500
Hess et al. (2009), using the same data
as Resource Systems Group (2009)e
        $2400 - $8500 (2009$)
       $2400 - $8500
Hidrueetal. (2011)f
       $3776 - $10,399 (2009$)
     $3776 - $10,399
"Values adjusted to 2009$ using the Bureau of Economic Analysis GDP deflator.
bBunch, David S., Mark Bradley, Thomas F. Golob, and Ryuichi Kitamura. "Demand for Clean-Fuel Vehicles in
California: A Discrete-Choice Stated Preference Pilot Project."  Transportation Research Part A 27A(3) (1993):
237-253.  The value of range was, in their model, assumed to be the same for all people.
°Kavelek, Chris. "CALCARS: The California Conventional and Alternative Fuel Response Simulator." Demand
Analysis Office, California Energy Commission, April 1996. The variation in values is due to willingness to pay
(WTP) varying by income levels and for one-car and two-car households.  The coefficient on range for one-car
households was not statistically significantly different from zero (t-statistic = 1.5), but it was for 2-car households (t-
statistic = 3.02). The minima and maxima presented here represent the values across both ownership and income
categories.
dResource Systems Groups, Inc. "Transportation Fuel Demand Forecast Household and Commercial Fleet Survey
Task 8 Report: Logistic Regression Analysis and Results." Prepared for California Energy Commission, June 2009.
eHess, S., T. Adler, M. Fowler and A. Bahreinian "The Use of Cross-nested Logit Models for Multi-Dimensional
Choice Processes:  The Case of the Demand for Alternative Fuel Vehicles," Proceedings of the 2009 European
Transport Conference, Leiden, Netherlands, 2009.  This study uses the same data as the Resource Systems Group
study. The coefficient on range was not statistically significantly different from zero in these regressions: t-
statistics varied from 1.29 to 1.52. The variation in values is due to willingness to pay (WTP) varying by income
levels and statistical specification. The minima and maxima presented here represent the values across both income
categories and specifications.
fHidrue, Michael K., George R. Parsons, Willett Kempton, and Meryl P. Gardner. "Willingness to Pay for Electric
Vehicles and their Attributes." Resource and Energy Economics 33(3) (2011): 686-705. The range of values is due
to the model separating consumers into "gasoline vehicle-oriented" and "electric vehicle-oriented" groups.  The EV-
oriented group has higher WTP for additional range.

       Driving range may be a major factor in consumers' decisions on EVs, but it is not the
only attribute that may be important to potential buyers (e.g., as noted, Hidrue et al. find that
some consumers appear willing to pay substantially for reduced tailpipe emissions).  A model
                                              8-13

-------
Chapter 8	

that does not incorporate the other factors important to consumers' decisions may not perform
well in predicting vehicle purchases. In addition, as mentioned above, and as seen in Table
8.1-2, it is likely that the WTP values for attributes of EVs will change over time, particularly if
EVs are used more widely, the infrastructure to fuel the vehicles becomes more accessible, and
consumers develop more familiarity and understanding of the vehicles.  Thus, challenges
associated with predicting market shares for EVs are even more serious than those already
serious challenges associated with predicting market shares for conventional vehicles.

               8.1.2.8       EPA Exploration of Vehicle Choice Modeling

       In order to develop greater understanding of these models, EPA is in the process of
developing a vehicle choice model.  In its current form, the model assumes that the vehicle fleet
and all characteristics of each vehicle, except vehicle prices and fuel economy, stay the same.
The model will predict changes in the vehicle fleet, at the individual-configuration level and at
more aggregated levels, in response to changes in vehicle fuel economy and price.

       The draft EPA model uses a nested logit structure common in the vehicle choice
modeling literature, as discussed above in Chapter 8.1.2.1.  "Nesting" refers to the decision-tree
structure of the model, and "logit" refers to the fact that the choices are discrete (i.e., yes/no
decisions about which vehicles to purchase, instead  of continuous values).

       The nesting involves a hierarchy of choices.  In its current form, at the initial decision
node, consumers choose between buying a new vehicle or not. Conditional on choosing a new
vehicle, consumers then choose between passenger vehicles, cargo vehicles, and ultra-luxury
vehicles. The next set of choices subdivides each of these categories into vehicle type (e.g.,
standard car, minivan, SUV, etc.).  Next, the vehicle types are divided into classes (small,
medium, and large SUVs, for instance),  and then, at the bottom, are the individual vehicle
configurations.

       At this bottom level, vehicles that are similar to each other (such as standard
subcompacts, or prestige large vehicles) end up in the same "nest." Substitution within a nest is
considered much more likely than substitution across nests, because  the vehicles within a  nest
are more similar to each other than vehicles in different nests. For instance, a person is more
likely to substitute between a Chevrolet  Aveo and a Toyota Yaris than between an Aveo and a
pickup truck.  In addition, substitution is greater  at low decision nodes (such as individual
configurations) than at higher decision nodes (such as the buy/no buy decision), because there
are more choices at lower levels than at higher levels.

       Parameters for the model (including demand elasticities and the value of fuel economy in
purchase decisions) are  being selected based on a review of values found in the literature on
vehicle choice modeling. As discussed above, a  number of studies have estimated these
parameters. Those estimates, combined with some theoretical requirements,11111 assist in
TTTTT rpj^ ^eory of nested logit requires that the price slopes (the change in utility as vehicle full price changes, a
measure of consumer responsiveness to price changes) must be higher in absolute value for lower nests. This
condition reflects the point, discussed above, that substitution is greater at lower decision notes than at higher ones.


                                           8-14

-------
                                                   2017 Draft Regulatory Impact Analysis
assigning values for the parameters. The model will allow individual users to change those
parameters.

       The fuel economy of a vehicle is used to adjust the price of the vehicle, using a version of
the procedure discussed in Chapter 8.1.2.7:  the value that the consumer places on fuel  economy
is multiplied by the change in fuel economy and incorporated into the "effective price" of the
vehicle.  In practice, implementing this calculation involves calculating the change in
expenditures on fuel based on schedules of VMT, vehicle survival, and fuel prices in the future
consistent with those in OMEGA. As discussed in Chapter 8.1.2.5, there is no consensus value
for consumers' willingness to pay for improved fuel economy: estimates vary tremendously.
The model assumes that consumers will use some years of discounted fuel savings, with the
modeler able to input both the number of years and the discount rate to be used in the analysis.

       The vehicle choice model will take as inputs an initial fleet of vehicles (including the
initial sales and fuel economy) in the absence of standards, the cost of technologies added to
each vehicle to comply with standards, and the change in fuel economy. With the initial sales
mix, for each vehicle, the model calculates a vehicle-specific constant that summarizes the value
of all attributes of the vehicle other than price and fuel economy. This constant ensures that the
model will predict changes in consumer response that would result only from changes in price
and fuel economy. This constant substitutes for estimating the effects of changes in all other
vehicle characteristics;  the underlying assumption is that  these other vehicle characteristics do
not change.uuuuu For instance, it assumes that a Ford Escape will not change in size, power, or
accessories; the only changes will be to its cost and its fuel economy.

       The model  assumes that the increase in vehicle cost associated with increased technology
is fully passed through  as an increase in vehicle price, and some years of fuel savings offset this
price increase. It then calculates changes in total fleet size and in sales mix, at the individual-
configuration level and at the level of vehicle class, due to the changes in fuel economy and
vehicle prices. It also calculates changes in consumer surplus associated with the changes in fuel
economy and vehicle prices.

       It is possible that the predicted changes in fleet mix will lead to predictions of vehicle
sales for  auto makers that do not meet the proposed standards, because the mix and volume of
vehicles sold changed from the initial levels. To correct this problem, it would be necessary to
feed the new fleet mix into OMEGA (which calculates costs and compliance) and get a new set
of output, which could then be fed back into the vehicle choice model.  OMEGA would increase
technologies, and thus costs, to improve compliance; those adjustments would then again affect
vehicle demand. We expect that this iterative process would converge to a fleet mix that would
meet standards. Performing this iteration requires development of an interface between the
vehicle choice model and OMEGA to ensure accurate transmission of data between the models.
At this time, the vehicle choice model takes output from OMEGA, but the results of the
modeling do not feed easily back into OMEGA. Building this interface is an expected  part of
our future modeling work.
1111111111 As explained in Section III.D of the preamble, as part of the technology cost analysis for the proposed rule,
the agencies have estimated the cost of maintaining all vehicle utility, with minor exceptions.


                                           8-15

-------
Chapter 8	

       The model is still undergoing development; EPA will seek peer review on it before it is
utilized.  In addition, concerns remain over the ability of any vehicle choice model to make
reasonable predictions of the response of the vehicle fleet to changes in prices and vehicle
characteristics.  EPA seeks comments on the use of vehicle choice modeling for predicting
changes in fleet mix due to policies, and on methods to test the ability of a vehicle choice model
to produce reasonable estimates of changes in fleet mix.

               8.1.2.9       Summary and Additional Considerations

       Consumer vehicle choice modeling in principle can provide a great deal of useful
information for regulatory analysis, helping to answer  some of the central questions about
relevant effects on consumer welfare. In practice, the advantages depend on the success of
models in predicting changes in fleet size and mix.

       First, consumer vehicle choice modeling has the potential to describe more accurately the
impact of a policy, by identifying market shifts. More accurate description of the market
resulting from a policy can improve other estimates of policy impacts, such as the change in total
vehicle emissions or vehicle miles traveled.  The predictive ability of models, though, is not
proven.

       Vehicle choice models can incorporate the effects on consumer decisions of changes in
vehicle characteristics, if there are estimates of the value that consumers put on changes in those
characteristics.  These willingness-to-pay values may, however, be sensitive to the ways they are
estimated, as indicated in the discussion of the value that consumers place on fuel economy in
their purchase decisions. Especially for characteristics associated with advanced technology
vehicles, such as EVs, the willingness-to-pay values may change over time as consumers
develop more experience with the vehicles and these characteristics. Models based on current
estimates may not predict well for the future.

       The modeling may improve estimates of the compliance costs of a rule. Consumers can
either accept the new costs and vehicle characteristics, or they can change which vehicles they
buy. Using a vehicle choice model is likely to reduce compliance costs: because the model
allows consumers to choose among accepting the new  vehicle, buying a different vehicle, or not
buying a vehicle, consumers have additional options, which improves their welfare relative to the
assumption that consumers will not change their buying behavior. .

       An additional complication associated with consumer choice modeling is accurate
prediction of producers' responses to the rule. While it is possible to include auto makers'
decisions (for instance, on setting prices) into vehicle choices, computational limits affect the
richness of these models.  Technology costs, while an accurate measure of the opportunity cost
of resources to society, may overestimate or underestimate the effect on the prices that
consumers face.

       Consumer choice models can be used to calculate consumer surplus impacts on vehicle
purchase decisions. Because these values  are based on the estimates of changes in vehicle sales
and fleet mix, consumer surplus measures  may not be accurate if the changes in vehicle sales and
fleet mix are not well estimated.
                                          8-16

-------
                                                    2017 Draft Regulatory Impact Analysis
       Principles of welfare analysis can be useful for understanding the role of consumer
vehicle choice models in benefit-cost analysis. In particular, except for EVs, the technology cost
estimates developed in this proposal take into account the costs to hold other vehicle attributes,
such as size and performance, constant.  In addition, the analysis assumes that the full technology
costs are passed along to vehicle buyers. With these assumptions, because welfare losses are
monetary estimates of how much buyers would have to be compensated to be made as well off as
   11        r- 1    1     WVVV 1    •   •                   11       11      WWWWW
in the absence of the change,       the price increase measures the loss to the buyer.
Assuming that the full technology cost gets passed along to the buyer as an increase in price, the
technology cost thus measures the welfare loss to the buyer. Increasing  fuel economy would
have to lead to other changes in the vehicles that buyers find undesirable for there to be
additional losses  not included in the technology costs.

       Given the current limitations in modeling the role of fuel economy in vehicle purchase
decisions, and limitations in modeling market responses to the new regulations, in this proposal
EPA holds constant the vehicle fleet size and mix in its calculations of the impacts of this rule,
and compares the fuel and other savings that consumers will receive with the technology costs of
the vehicles.  EPA continues to explore options for including consumer and producer choice in
modeling the impacts of fuel economy-related regulations.  This effort includes further review of
existing consumer vehicle choice models, the  estimates of consumers' willingness to pay for
increased fuel economy, and overall effects on consumer welfare, as well as EPA's exploration
of a vehicle choice model for use in the future.

8.2 Employment Impacts

       8.2.1   Introduction

       Although analysis of employment impacts is not part of a cost-benefit analysis (except to
the extent that labor costs contribute to costs), employment impacts of federal rules are of
particular concern in the current economic climate of sizeable unemployment. When President
Obama requested that the agencies develop  this program, he sought a program that would
"strengthen the [auto] industry and enhance job creation in the United States."379'xxxxx  The
recently issued Executive Order 13563, "Improving Regulation and Regulatory Review"
vvvw -pj^s approacn describes the economic concept of compensating variation, a payment of money after a change
that would make a consumer as well off after the change as before it. A related concept, equivalent variation,
estimates the income change that would be an alternative to the change taking place. The difference between them
is whether the consumer's point of reference is her welfare before the change (compensating variation) or after the
change (equivalent variation).  In practice, these two measures are typically very close together.
wwwww jncjeecj5 it is likely to be an overestimate of the loss to the buyer, because the buyer has choices other than
buying the same vehicle with a higher price; she could choose a different vehicle, or decide not to buy a new
vehicle. The buyer would choose one of those options only if the alternative involves less loss than paying the
higher price. Thus, the increase in price that the buyer faces would be the upper bound of loss of consumer welfare,
unless there are other changes to the vehicle due to the fuel economy improvements that make the vehicle less
desirable to buyers.
xxxxx rpj^ jyjay 21 2010 Presidential Memorandum also requested that EPA and NHTSA, in developing the
technical assessment to inform the rulemaking process (which was issued by the agencies and CARB on September
30, 2010), include, among other things, the "impacts on jobs and the automotive manufacturing base in the United
States."
                                            8-17

-------
Chapter 8	

(January 18, 2011), states, "Our regulatory system must protect public health, welfare, safety,
and our environment while promoting economic growth, innovation, competitiveness, and job
creation" (emphasis added). EPA is accordingly providing partial estimates of the effects of this
proposal on domestic employment in the auto manufacturing and parts sectors, while
qualitatively discussing how it may affect employment in other sectors more generally.

       This proposal is expected to affect employment in the United States through the regulated
sector - the auto manufacturing industry - and through several related sectors, specifically,
industries that supply the auto manufacturing industry (e.g., vehicle parts), auto dealers, the fuel
refining and supply sectors, and the general retail sector. According to the U.S. Bureau of Labor
Statistics, in 2010, about 677,000 people in the U.S. were employed in Motor Vehicle and Parts
Manufacturing Sector (NAICS  3361, 3362, and 3363). About 129,000 people in the U.S. were
employed in the Automobile and Light Truck Manufacturing Sector (NAICS 33611) in
December 2010; this is the directly regulated sector, since it encompasses the auto manufacturers
that are responsible for complying with the proposed  standards.380  Changes in light duty vehicle
sales, discussed in Chapter 8.1.1, could affect employment for auto dealers. The employment
effects of this rule are expected to expand beyond the regulated sector.  Though some of the parts
used to achieve the proposed standards are likely to be built by auto manufacturers themselves,
the auto parts manufacturing sector also plays a significant role in providing those parts, and will
also be affected by changes in vehicle sales. As discussed in Chapter 5.4 of the RIA, this
proposal is expected to reduce the amount of fuel these vehicles use, and thus affect the
petroleum refinery and supply industries. Finally, since the net reduction in cost associated with
this proposal is  expected to lead to lower household expenditures on fuel net of vehicle costs,
consumers then will have additional discretionary income that can be spent on other goods and
services.

       When the economy is at full employment, an environmental regulation is unlikely to have
much impact on net overall U.S. employment; instead, labor would primarily be shifted from one
sector to another.  These shifts in employment impose an opportunity cost on society,
approximated by the wages of the employees, as regulation diverts workers from other activities
in the economy. In this situation, any effects on net employment are likely to be transitory as
workers change jobs (e.g., some workers may need to be retrained or require time to  search for
new jobs, while shortages in some sectors or regions could bid up wages to attract workers).

       On the other hand, if a regulation comes into effect during a period of high
unemployment, a  change in labor demand due to regulation may affect net overall U.S.
employment because the labor market is not in equilibrium. In such a period, both positive and
negative employment effects are possible.YYYYY Schmalansee and Stavins point out that  net
positive employment effects are possible in the near term when the economy is at less than full
employment due to the potential hiring of idle labor resources by the regulated sector to meet
new requirements (e.g., to install new equipment) and new economic activity in sectors related to
the regulated sector.381  In the longer run, the net effect on employment is more difficult to
predict and will depend on the way in which the related industries respond to the regulatory
YYYYY Masur and Posner, available at http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1920441
                                          8-18

-------
                                                  2017 Draft Regulatory Impact Analysis
requirements. As Schmalansee and Stavins note, it is possible that the magnitude of the effect on
employment could vary over time, region, and sector, and positive effects on employment in
some regions or sectors could be offset by negative effects in other regions or sectors. For this
reason, they urge caution in reporting partial employment effects since it can "paint an inaccurate
picture of net employment impacts if not placed in the broader economic context."

       It is assumed that the official unemployment rate will have declined to 5.3 percent by the
time by the time this rule takes effect and so the effect of the regulation on labor will be to shift
workers from one sector to another.22222 Those shifts in employment impose an opportunity cost
on society, approximated by the wages of the employees, as regulation diverts workers from
other activities in the economy. In this situation, any effects on net employment are likely to be
transitory as workers change jobs (e.g., some workers may need to be retrained or require time to
search for new jobs, while shortages in some sectors or regions could bid up wages to attract
workers). It is also possible that the state of the economy will be such that positive or negative
employment effects will occur.

       A number of different approaches have been used in published literature to conduct
employment analysis.  This section describes  some of the common methods,  as well as some of
their limitations.

       8.2.2  Approaches to Quantitative Employment Analysis

       Measuring the employment impacts of a policy depend on a number of inputs and
assumptions.  For instance, as discussed, assumptions about the overall  state of unemployment in
the economy play a major role in measured job impacts.  The inputs to the models  commonly are
the changes in quantities or expenditures in the affected sectors; model results may vary in
different studies depending on the assumptions about the levels of those inputs, and which
sectors receive those changes.  Which sectors are included in the study can also affect the results.
For instance, a study of this program that looks only at employment impacts in the refinery
sector may find negative effects, because consumers will purchase less gasoline; a study that
looks only at the auto parts sector, on the other hand, may find positive  impacts, because the
program will require redesigned or additional parts for vehicles. In both instances, these would
only be partial perspectives on the overall change in national employment due to Federal
regulation.

              8.2.2.1       Conceptual Framework for Employment Impacts in the Regulated
              Sector

       One study by Morgenstern, Pizer, and Shih382  provides a retrospective look at the impacts
of regulation in employment in the regulated sectors by estimating the effects on employment of
spending on pollution abatement for four highly polluting/regulated U.S. industries (pulp and
paper, plastics, steel, and petroleum refining) using data for six years between 1979 and 1991.
The paper provides a theoretical framework that can be useful for examining the impacts of a
zzzzz Qfflce Of Management and Budget, "Fiscal Year 2012 Mid-Session Review: Budget of the U.S. Government.'
http://www.whitehouse.gov/sites/default/files/omb^udget/fy2012/assets/12msr.pdf, p. 10.


                                          8-19

-------
Chapter 8	

regulatory change on the regulated sector in the medium to longer term.  In particular, it
identifies three separate ways that employment levels may change in the regulated industry in
response to a new (or more stringent) regulation.

       •  Demand effect:  higher production costs due to the regulation will lead to higher
          market prices; higher prices in turn reduce demand for the good, reducing the demand
          for labor to make that good.  In the authors' words, the "extent of this effect depends
          on the cost increase passed on to consumers as well as the demand elasticity of
          industry output."

       •  Cost effect: as costs go up, plants add more capital and labor (holding other factors
          constant), with potentially positive  effects on employment. In the authors' words, as
          "production costs rise, more inputs, including labor, are used to produce the same
          amount of output."

       •  Factor shift effect:  post-regulation production technologies may be more or less
          labor-intensive (i.e., more/less labor is required per dollar of output).  In the authors'
          words, "environmental activities may be more labor intensive than conventional
          production," meaning that "the amount of labor per dollar of output will rise," though
          it is also possible that "cleaner operations could involve automation and less
          employment, for example."

According to the authors, the "demand effect" is expected to have a negative effect on
employment,AAAAAA  the "cost effect" to have a positive effect on employment, and the "factor
shift effect" to have an ambiguous effect on employment. Without more information with
respect to the magnitudes of these competing effects, it is not possible to predict the total effect
environmental regulation will have on employment levels in a regulated sector.

       The authors conclude that increased abatement expenditures  generally have not caused a
significant change in employment in those sectors. More specifically, their results show that, on
average across the industries studied, each additional $1  million spent on pollution abatement
results in a (statistically insignificant) net increase of 1.5 jobs.

       This approach to employment analysis has the advantage of carefully controlling for
many possibly confounding effects in order to separate the effect of changes in regulatory costs
on employment. It was, however, conducted for only four sectors.  It could also be very difficult
to update the study for other sectors, because one of the databases on which it relies, the
Pollution Abatement Cost and Expenditure survey, has been conducted infrequently since 1994,
with the last survey conducted in 2005.  The empirical estimates provided by Morgenstern et al.
are not relevant to the case of fuel economy standards, which are very different from the
      As will be discussed below, the demand effect in this proposal is potentially an exception to this rule. While
the vehicles become more expensive, they also produce reduced fuel expenditures; the reduced fuel costs provide a
countervailing impact on vehicle sales. As discussed in Preamble Section III.H. 1, this possibility that vehicles may
become more attractive to consumers after the program poses a conundrum: why have interactions between vehicle
buyers and producers not provided these benefits without government intervention?


                                           8-20

-------
                                                  2017 Draft Regulatory Impact Analysis
pollution control standards on industrial facilities that were considered in that study. In addition,
it does not examine the effects of regulation on employment in sectors related to but outside of
the regulated sector. Nevertheless, the theory that Morgenstern et al. developed continues to be
useful in this context.

       The following discussion of additional methodologies draws from Berck and Hoffmann's
review of employment models.383

              8.2.2.2       Computable General Equilibrium (CGE) Models

       Computable general equilibrium (CGE) models are often used to assess the impacts of
policy. These models include a stylized representation of supply and demand curves for all
major markets in the economy.  The labor market is commonly included. CGE models are very
useful for looking at interaction effects  of markets:  "they allow for substitution among inputs in
production and goods in consumption." Thus, if one market experiences a change, such as a new
regulation, then the effects can be observed in all other markets. As a result, they can measure
the employment changes in the economy due to a regulation. Because they usually assume
equilibrium in all markets, though, they typically lack involuntary unemployment.  If the total
amount of labor changes, it is due to people voluntarily entering or leaving the workforce. As a
result, these models may not be appropriate for measuring effects of a policy on unemployment,
because of the assumption that there is no involuntary unemployment.  In addition, because of
the assumptions of equilibrium in all markets and forward-looking consumers and firms, they are
designed for examining the long-run effects of a policy but may offer little insight into its short-
run effects.

              8.2.2.3       Input-Output (IO) Models

       Input-output models represent the economy through a matrix of coefficients that describe
the connections between supplying and consuming sectors. In that sense, like CGE models, they
describe the interconnections of the economy. These interconnections look at how changes in
one sector ripple through the  rest of the economy. For instance, a requirement for additional
technology  for vehicles requires additional steel, which requires more workers in both the auto
and steel sectors; the additional workers in those sectors then have more money to spend, which
leads to more employment in retail sectors.  These are known as "multiplier" effects, because an
initial impact in one sector gets multiplied through the economy.  Unlike CGE models, input-
output models have fixed, linear relationships among the sectors (e.g., substitution among inputs
or goods is not allowed), and quantity supplied need not equal quantity demanded. In particular,
these models do not allow for price changes - an increase in the demand for labor or capital  does
not result in a change in its price to help reallocate it to its best use. As a result, these models
cannot capture opportunity costs from using resources in one area of the economy over another.
The multipliers take an initial impact and can increase it substantially.

       IO models are commonly used for regional analysis of projects.  In a regional analysis,
the markets are commonly considered small enough that wages and prices are determined
outside the region, and any excess supply or demand is due to exports and imports (or, in the
case of labor, emigration or immigration).  For national-level employment analysis, the use of
input-output models requires  the assumption that workers flow into or out of the labor market
                                          8-21

-------
Chapter 8	

perfectly freely. Wages do not adjust; instead, people join into or depart from the labor pool as
production requires them. For other markets as well, there is no substitution of less expensive
inputs for more expensive ones.  As a result, IO models provide an upper bound on employment
impacts.  As Berck and Hoffmann note, "For the same reason, they can be thought of as
simulating very short-run adjustment," in contrast to the CGE's implicit assumption of long-run
adjustment.  Changes in production processes, introductions of new technologies, or learning
over time due to new regulatory requirements are also generally not captured by IO models, as
they are calibrated to already established relationships between inputs and outputs.

              8.2.2.4       Hybrid Models

       As Berck and Hoffmann note, input-output models and CGE models "represent a
continuum of closely related models." Though not separately discussed by Berck and Hoffmann,
some hybrid models combine some of the features of CGE models (e.g., prices that can change)
with input-output relationships. For instance, a hybrid model may include the ability to examine
disequilibrium phenomena, such as labor being at less than full employment. Hybrid models
depend on assumptions about how adjustments in the economy occur. CGE models characterize
equilibria but say little about the pathway between them, while IO models assume that
adjustments  are largely constrained by previously defined relationships; the effectiveness of
hybrid models depends on their success in overcoming the limitations of each of these
approaches.  Hybrid models could potentially be used to model labor market impacts of various
vehicle policy options although a number of judgments need to be made about the appropriate
assumptions underlying the model  as well the empirical basis for the modeling results.

              8.2.2.5       Single Sectors

       It is possible to conduct a bottom-up analysis of the partial effect of regulation on
employment in a single sector by estimating the change in output or expenditures in a sector and
multiplying it by an  estimate of the number of workers per unit of output or expenditures, under
the assumption that labor demand is proportional to output or expenditures.  As Berck and
Hoffmann note, though, "Compliance with regulations may create additional jobs that are not
accounted for." While such an analysis can approximate the effects in that one sector in a simple
way, it also may miss important connections to related sectors.

              8.2.2.6       Ex-Post Econometric Studies

       A number of ex-post econometric analyses examine the net effect of regulation on
employment in regulated sectors. Morgenstern, Pizer, and Shih (2002), discussed above, and
Berman and  Bui (2001) are two notable examples that rely on highly disaggregated
establishment-level time series data to estimate longer-run employment effects.384 While often a
sophisticated treatment of the issues analyzed, these studies commonly analyze specific scenarios
or sectors in  the past; care needs to be taken in extrapolating their results to other scenarios and
to the future.  For instance, neither of these two studies examines the auto industry and are
therefore of limited applicability in this context.
                                          8-22

-------
                                                  2017 Draft Regulatory Impact Analysis
              8.2.2.7       Summary

       All methods of estimating employment impacts of a regulation have advantages and
limitations. CGE models may be most appropriate for long-term impacts, but the usual
assumption of equilibrium in the employment market means that it is not useful for looking at
changes in overall employment: overall levels are likely to be premised on full employment.  IO
models, on the other hand, may be most appropriate for small-scale, short-term effects, because
they assume fixed relationships across sectors and do not require market equilibria. Hybrid
models, which combine some features of CGEs with IO models, depend upon key assumptions
and economic relationships that are built into them. Single-sector models are simple and
straightforward, but they are often based on the assumptions that labor demand is proportional to
output, and that other sectors are not affected. Finally, econometric models have been developed
to evaluate the longer-run net effects of regulation on sector employment, though these are ex-
post analyses commonly of specific sectors or situations, and the results may not have direct
bearing for the regulation being reviewed.

       8.2.3  Employment analysis of this proposal

       As mentioned above,  this program is expected to affect employment in the regulated
sector (auto manufacturing) and other sectors directly affected by the proposal: auto parts
suppliers, auto dealers, the fuel supply market (which will face reduced petroleum production
due to reduced fuel demand but which may see additional demand for electricity or other fuels),
and consumers (who will face higher vehicle costs and lower fuel expenditures).  In addition, as
the discussion above suggests, each of these sectors could potentially have ripple effects in the
rest of the economy. These ripple effects depend much more heavily on the state of the
macroeconomy than do the direct effects. At the national level, employment may increase in one
industry or region and decrease in another, with the net effect being smaller than either
individual-sector effect.  EPA does not attempt to quantify the net effects of the regulation on
overall national employment.

       The discussion that follows provides a partial, bottom-up quantitative estimate of the
effects of this proposal on the regulated sector (the auto industry; for reasons discussed below,
we include some quantitative assessment of effects on suppliers to the industry although they are
not regulated  directly). It also includes qualitative discussion of the effects of the proposal on
other sectors.  Focusing quantification of employment impacts on the regulated sector has some
advantages over quantifying all impacts. First, the analysis relies on data generated as part of the
rulemaking process, which focuses on the regulated sector;  as a result, what is presented here is
based on internally consistent assumptions and estimates made in this proposal.  Secondly, as
discussed above, net effects on employment in the economy as a whole depend heavily on the
overall state of the economy when this rule has its effects. Focusing on the regulated sector
provides insight into employment effects in that sector without having to make assumptions
about the state of the economy when this rule has its impacts.  We include  a qualitative
discussion of employment effects on other sectors to provide a broader perspective on the
impacts of this rule.

       As noted  above, in a full-employment economy, any changes in employment will result
from people changing jobs or voluntarily entering or exiting the workforce. In a full-

-------
Chapter 8	

employment economy, employment impacts of this proposal will change employment in specific
sectors, but it will have small, if any, effect on aggregate employment. This rule would take
effect in model years 2017 through 2025; by then, the current high unemployment may be
moderated or ended.  For that reason, this analysis does not include multiplier effects, but instead
focuses on employment impacts in the most directly affected industries. Those sectors are likely
to face the most concentrated employment impacts.  The agencies seek comment on other sectors
that are likely to be significantly affected and thus warrant further analysis in the final
rulemaking analysis.

               8.2.3.1      Employment Impacts in the Auto Industry

       Following the Morgenstern et al. conceptual framework for the impacts of regulation on
employment in the regulated sector, we consider three effects for the auto sector: the demand
effect, the cost effect, and the factor shift effect. However, we are only able to offer quantitative
estimates for the cost effect.  We note that these estimates, based on extrapolations from current
data, become more uncertain as time goes on.

               8.2.3.1.1     The  Demand Effect

       The demand effect depends on the effects of this proposal on vehicle sales.  If vehicle
sales increase, then more people will be required to assemble vehicles and their components.  If
vehicle sales decrease, employment associated with these activities will unambiguously decrease.
Unlike in Morgenstern et al.'s study, where the demand effect unambiguously decreased
employment, there are countervailing effects in the vehicle market due to the fuel savings
resulting from this program.  On one hand, this proposal will increase vehicle costs; by itself, this
effect would reduce vehicle sales.  On the other hand, this proposal will reduce the fuel costs of
operating the vehicle; by itself, this effect would increase vehicle sales, especially if potential
buyers have an expectation of higher fuel prices. The sign of demand effect will depend on
which of these effects dominates. Because, as described in Chapter 8.1, we have not quantified
the impact on sales for this proposal, we do not quantify the demand effect.

               8.2.3.1.2    The  Cost Effect

       The demand effect, discussed above, measures employment changes due to new vehicle
sales only. The cost effect measures employment impacts due to the new or additional
technologies needed for vehicles to comply with the proposed standards.

       One way to estimate the cost effect, given the cost estimates for complying with the rule,
is to use the ratio of workers to each $1 million of expenditures in that sector.  The use of these
ratios has both advantages and limitations.  It is often possible to estimate these ratios for quite
specific sectors of the economy: for instance, it is possible to estimate the average number of
workers in the light-duty vehicle manufacturing sector per $1 million spent in the sector, rather
than use the ratio from another, more aggregated sector, such as motor vehicle manufacturing.
As a result, it is not necessary to extrapolate employment ratios from possibly unrelated sectors.
On the other hand, these estimates are averages for the sectors, covering all the activities in those
sectors; they may  not be representative of the labor required when expenditures are required on
specific activities, as the factor shift effect (discussed below) indicates. For instance, the ratio
                                          8-24

-------
                                                 2017 Draft Regulatory Impact Analysis
for the motor vehicle manufacturing sector represents the ratio for all vehicle manufacturing, not
just for fuel efficiency improvements.  In addition, these estimates do not include changes in
sectors that supply these sectors, such as steel or electronics producers. They thus may best be
viewed as the effects on employment in the auto sector due to the changes in expenditures in that
sector, rather than as an assessment of all employment changes due to these changes in
expenditures.

       Some of the costs of this proposal will be spent directly in the auto manufacturing sector,
but some of the costs will be spent in the auto parts manufacturing sector. Because we do not
have information on the proportion of expenditures in each sector, we separately present the
ratios for both the auto manufacturing sector and the auto parts manufacturing sector.  These are
not additive, but should instead be considered as a range of estimates for the cost effect,
depending on which sector adds technologies to the vehicles to comply with the regulation.

       We use several public sources for estimates of employment per $1 million expenditures.
The U.S. Bureau of Labor Statistics (BLS) provides its Employment Requirements Matrix
(ERM),385 which provides direct estimates of the employment per $1 million in sales of goods in
202 sectors.  The most recent estimates, used here, are from 2008 (adjusted to 2009$).  The
tables used here are adjusted to remove the employment effects of imports through use of the
ratio of domestic production to domestic sales, described above, of 0.667. The values reported
are for Motor Vehicle Manufacturing (NAICS 3361) and Motor Vehicle Parts Manufacturing
(NAICS 3363).

       The Annual Survey of Manufactures386 (ASM) provides another source of estimates
based on a sample of 50,000 establishments out of a universe of 346,000 manufacturing
establishments.  It includes more sectoral detail than the BLS ERM:  for instance, while the ERM
includes the Motor Vehicle Manufacturing sector, the ASM has detail at the 6-digit NAICS code
level (e.g., automobile manufacturing vs. light truck and utility vehicle manufacturing). While
the ERM provides direct estimates of employees/Si million in expenditures, the ASM separately
provides number of employees and value shipments; the direct employment estimates here are
the ratio of those values. The data in the ASM are updated annually, except for years when the
full Economic Census occurs.  The tables presented here use data from 2009.  As with the ERM,
we adjust for the ratio of domestic production to domestic sales. The values reported are  for
Motor Vehicle Manufacturing (NAICS 3361), Automobile and Light Duty Motor Vehicle
Manufacturing (NAICS 33611), and Motor Vehicle Parts Manufacturing (NAICS 3363).

       The Economic Census includes all large companies and a sample of smaller ones. The
ASM is a subset of the Economic Census; though the Census itself is more complete, it is
conducted only every 5 years,  while the ASM is annual.  The values presented here use data
from 2007 (adjusted to 2009$), with the domestic production-to-sales adjustment. The values
reported are for Motor Vehicle Manufacturing (NAICS 3361), Automobile and Light Duty
Motor Vehicle Manufacturing (NAICS 33611), and Motor Vehicle Parts Manufacturing (NAICS
3363).

       Table 8.2-1 provides the values, either given (BLS) or calculated (ASM, Economic
Census) for employment per $1 million of expenditures, all based on 2009 dollars, though the
underlying data come from different years (which may account for some of the differences).  The
                                         8-25

-------
    Chapter 8	

    different data sources provide similar magnitudes for the estimates for the sectors. Parts
    manufacturing appears to be more labor-intensive than vehicle manufacturing; light-duty vehicle
    manufacturing appears to be slightly less labor-intensive than motor vehicle manufacturing as a
    whole.

Table 8.2-1 Employment per $1 Million Expenditures (2009$) in the Motor Vehicle Manufacturing
                                            Sector*
Source
BLSERM
ASM
ASM
Economic
Census
Economic
Census
BLSERM
ASM
Economic
Census
Sector
Motor Vehicle Mfg
Motor Vehicle Mfg
Light Duty Vehicle
Mfg
Motor Vehicle Mfg
Light Duty Vehicle
Mfg
Motor Vehicle Parts
Mfg
Motor Vehicle Parts
Mfg
Motor Vehicle Parts
Mfg
Ratio of
workers per
$1 million
expenditures
0.834
0.824
0.757
0.674
0.610
3.073
3.093
2.749
Ratio of workers per $1
million expenditures,
adjusted for domestic vs.
foreign production
0.556
0.549
0.505
0.449
0.407
2.049
2.063
1.833
    BLS ERM refers to the U.S. Bureau of Labor Statistics' Employment Requirement Matrix. ASM refers to the U.S.
    Census Bureau's Annual Survey of Manufactures. Economic Census refers to the U.S. Census Bureau's Economic
    Census.

           Over time, the amount of labor needed in the auto industry has changed: automation and
    improved production methods have led to significant productivity increases.  The BLS ERM, for
    instance, provided estimates that, in 1993, 1.52 workers were needed per $1 million of 2000$,
    but only 0.83 workers by 2008 (in 2000$)/S7 Because the ERM is available annually for 1993-
    2008, we used these data to estimate productivity improvements over time. We regressed logged
    ERM values on year (to estimate percent change per year) for both the Motor Vehicle
    Manufacturing and Motor Vehicle Parts Manufacturing sectors.  The results suggest a 4.4
    percent per year productivity improvement in the Motor Vehicle Manufacturing Sector, and a 3.6
    percent per year improvement in the Motor Vehicle Parts Manufacturing Sector. We then used
    the regression relationship to project the ERM through 2025. In the results presented below,
    these projected values (adjusted to 2009$) were used directly for the BLS ERM estimates.  For
    the ASM, we used the ratio of the projected value in the future to the projected value in 2009 (the
    base year for the ASM); for the Economic Census estimates, we used the ratio of the projected
                                              8-26

-------
                                                  2017 Draft Regulatory Impact Analysis
value in the future to the projected value in 2007 (the base year for that estimate).  As noted
above, we adjusted the estimate of workers per vehicle for the demand effect in Chapter
8.2.3.1.1, above, using the productivity improvement for the Motor Vehicle Manufacturing
sector; because the estimate of workers per vehicle for the demand effect was based on data from
2001 to 2010, we used the projected value of the ERM in 2005 as the denominator of the ratio.
This is a simple way to examine the relationship between labor required and expenditure and we
seek comment on refining this method.

       Table 8.2-2 shows the cost estimates developed for this rule, discussed in Chapter 5. The
maximum value in Table 8.2-2 for employment impacts per $1 million expenditures (after
accounting for the share  of domestic production) is 2.049 in 2009 if all the additional costs are in
the parts sector; the minimum value is 0.407 in 2009, if all the additional costs are in the light-
duty vehicle manufacturing sector: that is, the range of employment impacts is between 0.4 and
2 additional jobs per $1 million expenditures in the sector in 2009. The results in Table 8.2-2
include the productivity adjustment described above.

       While we estimate employment impacts beginning with the first year of the standard
(2017), some of these job gains may occur earlier as auto manufacturers and parts  suppliers hire
staff in anticipation of compliance with the standard.

Table 8.2-2 Employment due to Cost Effect in the Motor Vehicle Manufacturing Sector
Year
2017
2018
2019
2020
2021
2022
2023
2024
2025
Costs (before
adjustment for
domestic proportion of
production) (SMillions)
$ 2,300
$ 4,656
$ 6,507
$ 8,467
$ 11,878
$ 19,340
$25,036
$30,738
$33,561
Minimum
employment effect (if
all expenditures are in
the parts sector)
600
1,200
1,600
1,900
2,600
4,100
5,000
5,900
6,200
Maximum employment
effect (if all expenditures
are in the light duty
vehicle mfg sector)
3,600
7,000
9,400
11,800
15,900
25,000
31,200
37,000
39,000
              8.2.3.1.3
The Factor Shift Effect
       The factor shift effect looks at the effects on employment due to changes in labor
intensity associated with a regulation. As noted above, the estimates of the cost effect assume
constant labor per $1 million in expenditures, though the new technologies may be either more or
less labor-intensive than the existing ones. An estimate of the factor shift effect would either
increase or decrease the estimate used for the cost effect.

       We are not quantifying the factor shift effect here, for lack of data on the labor intensity
of all the possible technologies that manufacturers could use to comply with the proposed
standards. For a subset of the technologies, though, EPA-sponsored research (discussed in
                                          8-27

-------
Chapter 8	

Chapter 3.2.1.1 of the Joint TSD) which compared new technologies to existing ones at the level
of individual components provides some insights into the factor shift effect.

       The comparison involved tearing down the selected technologies to their individual
components and looking at the differences in materials and labor needs in moving from the
conventional to the new technologies/*5 For instance, the analysis compared all the parts and
labor associated with an 8-speed automatic transmission to those needed for a 6-speed automatic
transmission.

       Because labor cost was one of the sources of differences between the technologies, it is
possible, for those technologies, to see whether labor needs increase or decrease with the switch
to technologies that might contribute to compliance with the proposed standards. An increase in
labor cost for the new technology indicates an increase in the labor needed for the new
technology compared to the baseline technology.  For instance, an 8-speed transmission requires
$15.11 more in labor costs than a 6-speed transmission (as accounted for in EPA's cost estimates
for the proposed rule). Dividing the labor cost by a wage per hour estimate provides an estimate
of the additional hours (and thus the additional labor) needed for the new technology compared
to the baseline technology. As with labor cost, an increase in labor hours per technology
indicates greater employment needs for the new technologies. For this conversion, a weighted
average wage rate (90 percent of the average wage in the Motor Vehicle Parts Manufacturing
sector, and 10 percent of the average wage in the Motor Vehicle Manufacturing Sector) of
$46.36/hour in 2015, using 2008 dollars (the unit of analysis for the FEV study).  For the change
from a 6-speed to an 8-speed transmission, we thus estimate an additional 0.33 hours of labor per
transmission.

       Table 8.2-3 shows the changes in labor hours in moving from baseline to new fuel-saving
technologies for technologies in the FEV study. It indicates that, in switching from the baseline
to the new technologies, labor use per technology increased:  the fuel-saving technologies use
more labor than the baseline technologies. For a subset of the technologies likely to be used to
meet the standards in this proposal, then, the factor shift effect increases labor demand, at least in
the short run; in the long run, as with all technologies, the cost structure is likely to change due to
learning, economies of scale, etc. The technologies examined in this research are, however, only
a subset of the technologies that auto makers may use to comply with the standards proposed in
this program.  As  a result, these results cannot be considered definitive evidence that the factor
shift effect increases employment for this rule. We therefore do not quantify the factor shift
effect for this proposal.
                                          8-28

-------
                                                  2017 Draft Regulatory Impact Analysis
Table 8.2-3 Estimated Change in Labor for Selected Compliance Technologies
Technology
Downsized
Turbo GDI 4
Downsized
Turbo GDI V6
Downsized
Turbo GDI V6
Electric A/C
compressor
Power split
hybrid
6- to 8-speed
transmission
FEV
Case
Study
0101
0102
0104
0602
0502
0803
Vehicle Class
Compact C
Mid/Large C
SUV/Trucks

Mid/Large C
Mid/Large C
Labor
Costs
$72.58
$25.76
$84.19
$4.68
$395.85
$15.11
Total
Costs
$537.70
$87.38
$789.53
$167.54
$3,435.01
$61.84
Hours/
Technology
1.57
0.56
1.82
0.10
8.54
0.33
              8.2.3.1.4     Summary of Employment Effects in the Auto Sector

       While we are not able to quantify the demand or factor shift effects, the cost effect results
show that the employment effects of the increased spending in the regulated sector (and,
possibly, the parts sector) are expected to be positive and on the order of a few thousand in the
initial years of the program. As noted above, motor vehicle and parts manufacturing sectors
employed about 677,000 people in 2010, with automobile and light truck manufacturing
accounting for about 129,000 of that total.

       8.2.4  Effects on Employment for Auto Dealers

       The effects of the proposed standards on employment for auto dealers depend principally
on the effects of the standards on light duty vehicle sales.  In addition, auto dealers may be
affected by changes in maintenance and service costs.  Increases in those costs are likely to
increase labor demand in dealerships.

       Although this proposal predicts very small penetration of advanced technology vehicles,
the uncertainty on consumer acceptance of such technology vehicles  is even greater.  As
discussed in Chapter 8.1.2.7, consumers may find some characteristics of electric vehicles and
plug-in hybrid electric vehicles, such as the ability to fuel with electricity rather than gasoline,
attractive; they may find other characteristics, such as the limited  range for electric vehicles,
undesirable.  As a result, some consumers will find that EVs will meet their needs, but other
buyers will choose more conventional vehicles. Auto dealers may play a major role in
explaining the merits and disadvantages of these new technologies to vehicle buyers.  There may
be a temporary need for increased employment to train sales staff in the new technologies as the
new technologies become available.
                                          8-29

-------
Chapter 8	

       8.2.5   Effects on Employment in the Auto Parts Sector

       As discussed in the context of employment in the auto industry, some vehicle parts are
made in-house by auto manufacturers; others are made by independent suppliers who are not
directly regulated, but who will be affected by the proposed standards as well. The additional
expenditures on technologies are expected to have a positive effect on employment in the parts
sector as well as the manufacturing sector; the breakdown in employment between the two
sectors is difficult to predict. The effects on the parts sector also depend on the effects of the
proposed standards on vehicle sales and on the labor intensity of the new technologies,
qualitatively in the same ways as for the  auto manufacturing sector.

       8.2.6   Effects on Employment for Fuel Suppliers

       In addition to the effects on the auto manufacturing and parts sectors, these rules will
result in changes in fuel use that lower GHG emissions.  Fuel saving, principally reductions in
liquid fuels such as gasoline and diesel, will affect employment in the fuel suppliers industry
sectors throughout the supply chain, from refineries to gasoline stations. To the extent that the
proposed standards result in increased use of electricity or other new fuels, employment effects
will result from providing these fuels and developing the infrastructure to supply them to
consumers.

       Expected petroleum fuel consumption reductions can be found in Chapter 5.3. While this
reduced consumption represents fuel savings for purchasers of fuel, it represents a loss in value
of output for the petroleum refinery industry, fuel distributors, and gasoline stations.  The loss of
expenditures to petroleum fuel  suppliers  throughout the petroleum fuel supply chain, from the
petroleum refiners to the gasoline stations, is likely to result in reduced employment in these
sectors.

       This rule is also expected to lead  to increases in electricity consumption by vehicles, as
discussed in Chapter 5.3.  This new fuel may require additional infrastructure, such as electricity
charging locations. Providing this infrastructure will require some increased employment. In
addition, the generation of electricity is likely to require some additional labor. We have
insufficient information at this time to predict whether the increases in labor associated with
increased infrastructure provision and generation for electricity will be greater or less than the
employment reductions associated with reduced demand for petroleum fuels.

       8.2.7   Effects on Employment due to Impacts on Consumer Expenditures

       As a result of these proposed standards, consumers will pay a higher up-front cost for the
vehicles, but they will recover those costs in a fairly short payback period (see Preamble Section
III.H.lO.b); indeed, people who finance their vehicles are expected to find that their fuel savings
per month exceed the increase in the loan cost (though this depends on the particular loan rate a
consumer receives).  As a result, consumers will have additional  money to spend on other goods
and services, though, for those who do not finance their vehicles, it will occur after the initial
payback period.  These increased expenditures will support employment in those sectors where
consumers spend their savings.
                                          8-30

-------
                                                  2017 Draft Regulatory Impact Analysis
       These increased expenditures will occur in 2017 and beyond. If the economy returns to
full employment by that time, any change in consumer expenditures would primarily represent a
shift in employment among sectors. If, on the other hand, the economy still has substantial
unemployment, these expenditures would contribute to employment through increased consumer
demand.

       8.2.8   Summary

       The primary employment effects of this proposal are expected to be found throughout
several key sectors: auto manufacturers, auto dealers, auto parts manufacturing, fuel production
and supply, and consumers.

       These proposed standards initially take effect in model year 2017, a time period
sufficiently far in the future that the current sustained high unemployment at the national level
may be moderated or ended. In an economy with full employment, the primary employment
effect of a rulemaking is likely to be to move employment from one sector to another, rather than
to increase or decrease employment. For that reason, we  focus our partial quantitative analysis
on employment in the regulated sector, to examine the impacts on that sector directly.  We
discuss the likely direction of other impacts in the regulated sector as well as in other directly
related sectors, but we do not quantify those impacts, because they are more difficult to quantify
with reasonable accuracy, particularly so far into the future.

       For the regulated sector, the cost effect is expected to increase employment by 600 -
3,600 workers in 2017, depending on the share of that employment that is in the auto
manufacturing sector compared to the auto parts manufacturing sector.  As mentioned above,
some of these job gains may occur earlier as auto manufacturers and parts suppliers hire staff to
prepare to comply with the standard. The demand effect is ambiguous and depends on changes in
vehicle sales, which are not quantified for this proposal. Though we do not have estimates of the
factor shift effect for all potential compliance technologies, the evidence which we do have for
some technologies suggests that many of the technologies will have increased labor needs.

       Effects in other sectors that are predicated on vehicle sales are also ambiguous.  Changes
in vehicle sales are expected to affect labor needs in auto  dealerships and in parts manufacturing.
Increased expenditures for auto parts are expected to require increased labor to build parts,
though this effect also depends on any changes in the labor intensity of production; as noted, the
subset of potential compliance technologies for which data are available show increased labor
requirements. Reduced fuel production implies less employment in the petroleum sectors.
Finally, consumer spending is expected to affect employment through changes in expenditures in
general retail sectors; net fuel savings by consumers are expected to increase demand (and
therefore employment) in other sectors.
                                          References
                                          8-31

-------
Chapter 8	

326 See Hunt Alcott, Social Norms and Energy Conservation, Journal of Public Economics
(forthcoming 2011), available at http://web.mit.edu/allcott/www/Allcott%20201
!%20JPubEc%20-%20 Social% 20Norms% 20and%20Energy%20Conservation.pdf; Christophe
Chamley, Rational Herds: Economic Models of Social Learning (Cambridge, 2003).

327 See, for instance, Gron, Ann, and Deborah Swenson, 2000. "Cost Pass-Through in the U.S.
Automobile Market," Review of Economics and Statistics 82: 316-324 (Docket EPA-HQ-OAR-
2010-0799).

328 Insurance Information Institute, 2008, "Average Expenditures for Auto Insurance By State,
2007-2008," http://www.iii.org/media/facts/statsbyissue/auto/, accessed 8/22/11 (Docket EPA-
HQ-OAR-2010-0799).

329 U.S. Department of Energy, 2011. "Transportation and the Economy," Chapter 10 in
"Transportation Energy Data Book," http://cta.ornl.gov/data/tedb30/Edition30_Chapterl0.pdf,
accessed 8/22/11, Table 11 (Docket EPA-HQ-OAR-2010-0799).

330 Solheim, Mark, 2006 "State Car Tax Rankings,"
http://www.kiplinger.com/features/archives/2006/04/cartax.html, accessed April 23, 2009
(Docket EPA-HQ-OAR-2009-0472-0010) (Docket EPA-HQ-OAR-2010-0799).

331 U.S. Census Bureau,  "Table 2. Resident Population of the 50 States, the District of Columbia,
and Puerto Rico: 2010 Census" http://2010.census.gov/news/press-
kits/apportionment/apport.html .

332"National Auto Loan Rates for July 21, 2011,"
http://www.bankrate.com/fmance/auto/national-auto-loan-rates-for-july-21-2011 .aspx, accessed
7/26/11 (Docket EPA-HQ-OAR-2010-0799).

333 Goldberg, Pinelopi Koujianou, "Product Differentiation and Oligopoly in International
Markets: The Case of the U.S. Automobile Industry," Econometrica 63(4) (July 1995): 891-951
(Docket EPA-HQ-OAR-2010-0799); Goldberg, Pinelopi Koujianou, "The Effects of the
Corporate Average Fuel Efficiency Standards in the US," Journal of Industrial Economics 46(1)
(March 1998):  1-33  (Docket EPA-HQ-OAR-2010-0799).

334 Greene, David L., K.G. Duleep, Doug Elliott, and Sanjana Ahmad, "A Fuel Economy
Regulatory Analysis Model (FERAM) For the Energy Information Administration," prepared by
the Oak Ridge National Laboratory for the U.S. Department of Energy under contract No. DE-
AC0500OR22725, 2005 (Docket EPA-HQ-OAR-2010-0799); Greene, David L., Philip D.
Patterson, Margaret Singh, and Jia Li, "Feebates, Rebates, and Gas-Guzzler Taxes: A Study of
Incentives for Increased Fuel Economy," Energy Policy 33 (2005):  757-775 (Docket EPA-HQ-
OAR-2010-0799).

335 McManus, Walter M., "Can Proactive Fuel Economy Strategies Help Automakers Mitigate
Fuel-Price Risks?" University of Michigan Transportation Research Institute, September  14,
2006 (Docket EPA-HQ-OAR-2010-0799).
                                         8-32

-------
                                               2017 Draft Regulatory Impact Analysis
336 Berry, Steven, James Levinsohn, and Ariel Pakes, "Automobile Prices in Market
Equilibrium," Econometrica 63(4) (July 1995): 841-940 (Docket EPA-HQ-OAR-2010-0799);
Berry, Steven, James Levinsohn, and Ariel Pakes, "Differentiated Products Demand Systems
from a Combination of Micro and Macro Data: The New Car Market," Journal of Political
Economy 112(1) (2004): 68-105 (Docket EPA-HQ-OAR-2010-0799).

337 Bento, Antonio M., Lawrence H. Goulder, Emeric Henry, Mark R. Jacobsen, and Roger H.
von Haefen, "Distributional and Efficiency Impacts of Gasoline Taxes" American Economic
Review 99(3) 2009):  667-699 (Docket EPA-HQ-OAR-2010-0799).

338 Train, Kenneth E., and Clifford Winston, "Vehicle Choice Behavior and the Declining
Market Share of U.S. Automakers," International Economic Review 48(4) (November 2007):
1469-1496 (Docket EPA-HQ-OAR-2010-0799).

339 Kleit, Andrew N., "Impacts of Long-Range Increases in the Fuel Economy (CAFE)
Standard," Economic Inquiry 42(2) (April 2004): 279-294 (Docket EPA-HQ-OAR-2010-0799).

340 Austin, David, and Terry Dinan, "Clearing the Air: The Costs and Consequences of Higher
CAFE Standards and Increased Gasoline Taxes," Journal of Environmental Economics and
Management 50 (2005):  562-582 (Docket EPA-HQ-OAR-2010-0799).

341 E.g., Bento, Antonio M., Lawrence H. Goulder, Emeric Henry, Mark R. Jacobsen, and Roger
H. von Haefen, "Distributional and Efficiency Impacts of Gasoline Taxes," American Economic
Review 99(3) 2009):  667-699 (Docket EPA-HQ-OAR-2010-0799); Train, Kenneth E., and
Clifford Winston, "Vehicle Choice Behavior and the Declining Market Share of U.S.
Automakers," International Economic Review 48(4) (November 2007): 1469-1496 (Docket
EPA-HQ-OAR-2010-0799).

342 E.g., Berry,  Steven, James Levinsohn, and Ariel Pakes, "Automobile Prices in Market
Equilibrium," Econometrica 63(4) (July 1995): 841-940 (Docket EPA-HQ-OAR-2010-0799).

343 Greene, David L., K.G. Duleep, Doug Elliott, and Sanjana Ahmad,  "A Fuel Economy
Regulatory Analysis Model (FERAM) For the Energy Information Administration," prepared by
the Oak Ridge National Laboratory for the U.S. Department of Energy under contract No. DE-
AC0500OR22725, 2005 (Docket EPA-HQ-OAR-2010-0799).

344 NERA Economic Consulting, "Evaluation of NHTSA's Benefit-Cost Analysis of 2011-2015
CAFE Standards," 2008, available at
http://www.heartland.org/policybot/results/23495/Evaluation_of_NHTSAs_BenefitCost_Analysi
s_Of_20112015_CAFE_Standards.html (Docket EPA-HQ-OAR-2010-0799).

345 E.g., Train, Kenneth E., and Clifford Winston, "Vehicle Choice Behavior and the Declining
Market Share of U.S. Automakers," International Economic Review 48(4) (November 2007):
1469-1496 (Docket EPA-HQ-OAR-2010-0799).

346 E.g., Berry,  Steven, James Levinsohn, and Ariel Pakes, "Automobile Prices in Market
Equilibrium," Econometrica 63(4) (July 1995): 841-940 (Docket EPA-HQ-OAR-2010-0799).
                                        8-33

-------
Chapter 8	

347 E.g., Bento, Antonio M., Lawrence H. Goulder, Emeric Henry, Mark R. Jacobsen, and Roger
H. von Haefen, "Distributional and Efficiency Impacts of Gasoline Taxes," American Economic
Review 99(3) 2009): 667-699  (Docket EPA-HQ-OAR-2010-0799).

348 Bento, Antonio M., Lawrence H. Goulder, Emeric Henry, Mark R. Jacobsen, and Roger H.
von Haefen, "Distributional and Efficiency Impacts of Gasoline Taxes," American Economic
Review 99(3) 2009): 667-699 (Docket EPA-HQ-OAR-2010-0799); Feng, Yi, Don Fullerton,
and Li Gan, "Vehicle Choices, Miles Driven and Pollution Policies," National Bureau of
Economic Analysis Working Paper 11553, available at http://econweb.tamu.edu/gan/wl 1553.pdf
, accessed 5/12/09 (Docket EPA-HQ-OAR-2010-0799).

349 NERA Economic Consulting, "Appendix B:  New Vehicle Market Model," "Impacts of the
California Greenhouse Gas Emission Standards on Motor Vehicle Sales," comments submitted
to the U.S. Environmental Protection Agency at Regulations.gov, document number EPA-HQ-
O AR-2006-0173-9053.1 (Docket EPA-HQ-OAR-2010-0799).

350 E.g., Train, Kenneth E., and Clifford Winston, "Vehicle Choice Behavior and the Declining
Market Share of U.S. Automakers," International Economic Review 48(4) (November 2007):
1469-1496 (Docket EPA-HQ-OAR-2010-0799)

351 E.g., Berry,  Steven, James Levinsohn, and Ariel Pakes, "Automobile Prices in Market
Equilibrium." Econometrica 63(4) (July  1995):  841-940 (Docket EPA-HQ-OAR-2010-0799).

352 Greene, David L., Philip D. Patterson, Margaret Singh, and Jia Li, "Feebates, Rebates, and
Gas-Guzzler Taxes: A Study of Incentives for Increased Fuel Economy," Energy Policy 33
(2005):  757-775  (Docket EPA-HQ-OAR-2010-0799).

353 Rex, Emma, and Henrikke Baumann (2007).  "Beyond ecolabels: what green marketing can
learn from conventional marketing." Journal of Cleaner Production 15: 567-576  (Docket EPA-
HQ-OAR-2010-0799).

354 Whitefoot, Kate, Meredith Fowlie, and Steven Skerlos, "Product Design Responses to
Industrial Policy:  Evaluating Fuel Economy Standards Using an Engineering Model of
Endogenous Product Design," Energy Institute at Haas Working Paper 214, University of
California Energy Institute, February 2011  (Docket EPA-HQ-OAR-2010-0799).

355 Berry, Steven, James Levinsohn, and Ariel Pakes, "Automobile Prices in Market
Equilibrium," Econometrica 63(4) (July  1995):  841-940 (Docket EPA-HQ-OAR-2010-0799).

356 Goldberg, Pinelopi Koujianou, "Product Differentiation and Oligopoly in International
Markets:  The Case of the U.S. Automobile Industry," Econometrica 63(4) (July 1995):  891-951
(Docket EPA-HQ-OAR-2010-0799).

357 Brownstone, David, and Kenneth Train, "Forecasting New Product Penetration with Flexible
Substitution Patterns," Journal of Econometrics 89 (1999):  109-129 (Docket EPA-HQ-OAR-
2010-0799); Brownstone, David, David S. Bunch, and Kenneth Train,  "Joint Mixed Logit
Models of Stated and Revealed Preferences for Alternative-Fuel Vehicles," Transportation
Research Part B 34 (2000):  315-338 (Docket EPA-HQ-OAR-2010-0799); Greene David L.,
                                        8-34

-------
                                                2017 Draft Regulatory Impact Analysis
"TAFV Alternative Fuels and Vehicles Choice Model Documentation," prepared by the Oak
Ridge National Laboratory for the U.S. Department of Energy, July 2001 (Docket EPA-HQ-
OAR-2010-0799); Greene, David L., K. G. Duleep, and Walter McManus, "Future Potential of
Hybrid and Diesel Powertrains in the U.S. Light-Duty Vehicle Market," prepared by the Oak
Ridge National Laboratory for the U.S. Department of Energy, August 2004 (Docket EPA-HQ-
OAR-2010-0799).

358 Petrin, Amil, "Quantifying the Benefits of New Products:  The Case of the Minivan," Journal
of Political Economy 110 (2002): 705-729 (Docket EPA-HQ-OAR-2010-0799); Berry, Steven,
James Levinsohn, and Ariel Pakes, "Differentiated Products Demand Systems from a
Combination of Micro and Macro Data: The New Car Market," Journal of Political Economy
112(2004): 68-105 (Docket EPA-HQ-OAR-2010-0799).

359 Train, Kenneth E., and Clifford Winston, "Vehicle Choice Behavior and the Declining
Market Share of U.S. Automakers," International Economic Review 48 (November 2007):
1469-1496 (Docket EPA-HQ-OAR-2010-0799).

360 Bento, Antonio M., Lawrence H. Goulder, Emeric Henry, Mark R. Jacobsen, and Roger H.
von Haefen, "Distributional and Efficiency Impacts of Gasoline Taxes," American Economic
Review 99(3) 2009): 667-699 (Docket EPA-HQ-OAR-2010-0799); Feng, Yi, Don Fullerton,
and Li Gan, "Vehicle Choices, Miles Driven and Pollution Policies," National Bureau of
Economic  Analysis Working Paper 11553, available at http://econweb.tamu.edu/gan/wl 1553.pdf
, accessed  5/12/09 (Docket EPA-HQ-OAR-2010-0799).

361 Greene, David L., Philip D. Patterson, Margaret Singh, and Jia Li, "Feebates, Rebates,  and
Gas-Guzzler Taxes: A Study of Incentives for Increased Fuel Economy," Energy Policy 33
(2005): 757-775  (Docket EPA-HQ-OAR-2010-0799); Feng, Yi, Don Fullerton, and Li Gan,
"Vehicle Choices, Miles Driven and Pollution Policies," National Bureau of Economic  Analysis
Working Paper 11553, available at http://econweb.tamu.edu/gan/wl 1553.pdf, accessed 5/12/09
(Docket EPA-HQ-OAR-2010-0799); Greene, David L., "Feebates, Footprints and Highway
Safety," Transportation Research Part D 14 (2009):  375-384.

362 E.g., Austin, David, and Terry Dinan, "Clearing the Air: The Costs and Consequences of
Higher CAFE Standards and Increased Gasoline Taxes," Journal of Environmental Economics
and Management 50 (2005):  562-582  (Docket EPA-HQ-OAR-2010-0799).

363 Turrentine, Thomas S., and Kenneth S. Kurani, "Car Buyers and Fuel Economy?" Energy
Policy 35 (2007):  1213-1223  (Docket EPA-HQ-OAR-2010-0799).

364 E.g., Espey, Molly, and Santosh Nair, 2005. "Automobile Fuel Economy: What Is It Worth?"
Contemporary Economic Policy 23: 317-323  (Docket EPA-HQ-OAR-2010-0799).

365 Larrick, Richard P., and Jack B. Soil, 2008. "The MPG Illusion,"  Science 320(5883): 1593-
1594 (Docket EPA-HQ-OAR-2010-0799).

366 E.g., Goldberg, Pinelopi Koujianou, "Product Differentiation and Oligopoly in International
Markets: The Case of the U.S. Automobile Industry," Econometrica 63(4) (July 1995): 891-951
(Docket EPA-HQ-OAR-2010-0799).


                                        8-35

-------
Chapter 8	

367 E.g., Berry, Steven, James Levinsohn, and Ariel Pakes, "Automobile Prices in Market
Equilibrium," Econometrica 63(4) (July 1995):  841-940  (Docket EPA-HQ-OAR-2010-0799).

368 Busse, Meghan, Christopher R. Knittel, and Florian Zettelmeyer, 2009.  "Pain at the Pump:
How Gasoline Prices Affect Automobile Purchasing in New and Used Markets," working paper,
http://web.mit.edu/knittel/www/papers/gaspaper_latest.pdf (accessed 11/1/11) (Docket EPA-
HQ-OAR-2010-0799); Li, Shanjun, Christopher Timmins, and Roger H. von Haefen, 2009.
"How Do Gasoline Prices Affect Fleet Fuel Economy?" American Economic Journal:
Economic Policy 1(2):  113-137; Congressional Budget Office (2008) (Docket EPA-HQ-OAR-
2010-0799);  Congressional Budget Office (2008). Effects of Gasoline Prices on Driving
Behavior and Vehicle Markets. The Congress of the United States, Pub. No.  2883 (Docket EPA-
HQ-OAR-2010-0799); West. Sarah (2007). "The Effect of Gasoline Prices on the Demand for
Sport Utility Vehicles." Working Paper. Macalester College,
http://www.macalester.edu/~wests/SarahWestMEA2007.pdf, accessed 1/25/10 (Docket EPA-
HQ-OAR-2010-0799).

369 Greene, David L., and Jin-Tan Liu, 1988.  "Automotive Fuel Economy Improvements and
Consumers' Surplus," Transportation Research Part A 22A (3):  203-218 (Docket EPA-HQ-
OAR-2010-0799).

370 Greene, David L. (2010). "How Consumers Value Fuel Economy: A Literature Review."
EPA Report EPA-420-R-10-008 (Docket EPA-HQ-OAR-2010-0799).

371 Gramlich, Jacob, "Gas Prices, Fuel Efficiency, and Endogenous Product Choice in the U.S.
Automobile Industry," http://faculty.msb.edu/jpg72/Autos_GramlichJacob.pdf, accessed 11/1/11
(Docket EPA-HQ-OAR-2010-0799).

372 Gramlich, Jacob, "Gas Prices and Endogenous Product Selection in the U.S. Automobile
Industry," http://faculty.msb.edu/jpg72/Autos_GramlichJacob.pdf, accessed  11/1/11 (Docket
EPA-HQ-OAR-2010-0799); McManus, Walter, 2007. "The Impact of Attribute-Based
Corporate Average Fuel Economy (CAFE) Standards: Preliminary Results." Ann Arbor, MI:
University of Michigan Transportation Research Institute, Report No. UMTRI-2007-31 (Docket
EPA-HQ-OAR-2010-0799).

373 For instance, Kleit, Andrew N., "Impacts of Long-Range Increases in the Fuel Economy
(CAFE) Standard," Economic Inquiry 42(2) (April 2004): 279-294 (Docket EPA-HQ-OAR-
2010-0799); Austin, David, and Terry Dinan,  "Clearing the Air:  The Costs and Consequences of
Higher CAFE Standards and Increased Gasoline Taxes," Journal of Environmental Economics
and Management 50 (2005): 562- (Docket EPA-HQ-OAR-2010-0799); Klier, Thomas, and
Joshua Linn, 2008. "New Vehicle Characteristics and the Cost of the Corporate Average Fuel
Economy Standard," Resources for the Future Paper RFF DP  10-50 (December 2010),
http://www.rff.org/RFF/Documents/RFF-DP-10-50.pdf) (Docket EPA-HQ-OAR-2010-
0799) Jacob sen, Mark, 2010. "Evaluating U.S. Fuel Economy Standards in a Model with
Producer and Household Heterogeneity," working paper,  at
http://econ.ucsd.edu/~m3jacobs/Jacobsen_CAFE.pdf (accessed 11/1/11) (Docket EPA-HQ-
OAR-2010-0799) .
                                        8-36

-------
                                                2017 Draft Regulatory Impact Analysis
374 Austin, David, and Terry Dinan, "Clearing the Air:  The Costs and Consequences of Higher
CAFE Standards and Increased Gasoline Taxes," Journal of Environmental Economics and
Management 50 (2005): 562-582 (Docket EPA-HQ-OAR-2010-0799).

375 LEA. 2007. "Mind the Gap: Quantifying Principal-Agent Problems in Energy Efficiency."
Paris, France: International Energy Agency (Docket EPA-HQ-OAR-2010-0799); Jaffe, Adam
B., Richard G. Newell, and Robert N. Stavins (2001). "Energy Efficient Technologies and
Climate Change Policies: Issues and Evidence." In Climate Change Economics and Policy,
Toman, Michael A., ed., Washington, D.C.: Resources for the Future, p.171-181 (Docket EPA-
HQ-OAR-2010-0799); Metcalf, Gilbert E., and Kevin A. Hassett (1999).  "Measuring the Energy
Savings From Home Improvement Investments: Evidence From Monthly Billing Data."  The
Review of Economics and Statistics 81(3):  516-528 (Docket EPA-HQ-OAR-2010-0799);
Tietenberg, T. (2009). "Reflections - Energy Efficiency Policy: Pipe Dream or Pipeline to the
Future?"  Review of Environmental Economics and Policy 3(2): 304-320 (Docket EPA-HQ-
OAR-2010-0799).

376 Helfand, Gloria, and Ann Wolverton (2011). "Evaluating the Consumer Response to Fuel
Economy: A Review of the Literature." International Review of Environmental and Resource
Economics 5 (2011): 103-146 (Docket EPA-HQ-OAR-2010-0799).

377 Jaccard, Mark.  "Paradigms  of Energy Efficiency's Cost and their Policy Implications: Deja
Vu All Over Again." Modeling the Economics of Greenhouse Gas Mitigation: Summary of a
Workshop, K. John Holmes, Rapporteur. National Academies Press, 2010.
http://www.nap.edu/openbook.php?record_id=13023&page=42 (Docket EPA-HQ-OAR-2010-
0799)

378 Greene David L., "TAFV Alternative Fuels and Vehicles Choice Model Documentation,"
prepared by the Oak Ridge National Laboratory for the U.S. Department of Energy, July 2001
(Docket EPA-HQ-OAR-2010-0799).

379 President Barack Obama.  "Presidential  Memorandum Regarding Fuel Efficiency Standards.
The White House, Office of the Press Secretary, May 21, 2010.  http://www.whitehouse.gov/the-
press-office/presidential-memorandum-regarding-fuel-efficiency-standards

380 U.S. Bureau of Labor Statistics, Quarterly Census of Employment and Wages, as accessed on
August 9, 2011.

381 Schmalensee, Richard, and Robert N. Stavins. "A Guide to Economic and Policy Analysis of
EPA's Transport Rule." White paper commissioned by Excel on Corporation, March 2011
(Docket EPA-HQ-OAR-2010-0799).

382 Morgenstern, Richard D., William A. Pizer, and Jhih-Shyang Shih. "Jobs Versus  the
Environment: An Industry-Level Perspective." Journal of Environmental Economics and
Management 43 (2002): 412-436 (Docket EPA-HQ-OAR-2010-0799).

383 Berck, Peter, and Sandra Hoffman.  "Assessing the Employment Impacts of Environmental
and Natural Resource Policy."  Environmental and Resource Economics 22 (2002): 133-156
(Docket EPA-HQ-OAR-2010-0799).


                                        8-37

-------
Chapter 8	

384 Berman, Eli, and Linda T. Bui, (2001) "Environmental Regulation and Labor Demand:
Evidence from the South Coast Air Basin," Journal of Public Economics, 79, 265 - 295 (Docket
EP A-HQ-O AR-2010-0799).

385 http://www.bls.gov/emp/ep_data_emp_requirements.htm

386 http://www.census.gov/manufacturing/asm/index.html

387 http://www.bls.gov/emp/ep_data_emp_requirements.htm; this analysis used data for sectors
88 (Motor Vehicle Manufacturing) and 90 (Motor Vehicle Parts Manufacturing) from "Chain-
weighted (2000 dollars) real domestic employment requirements table. . . adjusted to remove
imports."

388FEV, Inc. "Light-Duty Technology Cost Analysis: Power-split and P2 HEV Case Studies,"
Report FEV07-069-3 03, October 10, 2011 (Docket EP A-HQ-O AR-2010-0799).
                                        8-38

-------
                                                2017 Draft Regulatory Impact Analysis
9     Small Business Flexibility Analysis

       The Regulatory Flexibility Act, as amended by the Small Business Regulatory
Enforcement Fairness Act of 1996 (SBREFA), generally requires an agency to prepare a
regulatory flexibility analysis of any rule subject to notice-and-comment rulemaking
requirements under the Administrative Procedure Act or any other statute. As a part of this
analysis, an agency is directed to convene a Small Business Advocacy Review Panel (SBAR
Panel or 'the Panel'), unless the agency certifies that the rule will not have a significant
economic impact on a substantial number of small entities.  During such a Panel process, the
agency would gather information and recommendations from Small Entity Representatives
(SERs) on how to reduce the impact of the rule on small entities.  As discussed below, EPA is
proposing to certify that this proposed rule would not have a significant economic impact on a
substantial number of small entities, and thus we have not conducted an SBAR Panel for this
rulemaking

       The following discussion  provides an overview  of small entities in the vehicle market.
Small entities include small businesses, small organizations, and small  governmental
jurisdictions. For the purposes of assessing the impacts of the rule on small entities, a small
entity is defined as: (1) a small business that meets the definition for business based on the
Small Business Administration's (SBA) size standards (see Table 9.1-1);  (2) a small
governmental jurisdiction that is a government of a city, county, town,  school district or
special district with a population  of less than 50,000; and (3) a small organization that is any
not-for-profit enterprise which is  independently owned  and operated and is not dominant in its
field. Table 9.1-1 provides an overview of the primary SBA small business categories
potentially affected by this regulation.

              Table 9.1-1 Primary Vehicle SBA Small Business Categories
Industry a
Vehicle manufacturers (including
small volume manufacturers)
Independent commercial
importers
Alternative Fuel Vehicle
Converters
Defined as Small Entity
by SBA if Less Than or
Equal to:
1,000 employees
$7 million annual sales
$23 million annual sales
100 employees
750 employees
1,000 employees
$7 million annual sales
NAICS Codes b
336111,336112
811111,811112,811198
441120
423110
336312,336322,336399
335312
811198
 ' Light-duty vehicle entities that qualify as small businesses would not be subject to this proposed rule. We are
proposing to exempt small business entities from the proposed standards.
b North American Industrial Classification System

       We compiled a list of vehicle manufacturers, independent commercial importers
(ICIs), and alternative fuel converters that would be potentially affected by the rule from our
2011 model year certification database.  These companies are already certifying their vehicles
for compliance with applicable EPA emissions standards (e.g., Tier 2). We then identified
                                         9-1

-------
Chapter 9	

companies that appear to meet the definition of small business provided in the table above.
We were able to identify companies based on certification information and previous
rulemakings where we conducted Regulatory Flexibility Analyses.

       Based on this assessment, EPA identified a total of about 21 entities that appear to fit
the Small Business Administration (SBA) criterion of a small business. EPA estimates there
are about 4 small vehicle manufacturers, including three electric vehicle manufacturers, 8
independent commercial importers (ICIs), and 9 alternative fuel vehicle converters in the
light-duty vehicle market which may qualify as small businesses.389  Independent commercial
importers (ICIs) are companies that hold a Certificate (or Certificates) of Conformity
permitting them to import nonconforming vehicles and to modify these vehicles to meet U.S.
emission standards.  ICIs are not required to meet the emission standards in effect when the
vehicle is modified, but instead they must meet the emission standards in effect when the
vehicle was originally produced (with an annual production cap of a total of 50 light-duty
vehicles and trucks). Alternative fuel vehicle converters are businesses that convert gasoline
or diesel  vehicles to operate on alternative fuel (e.g., compressed natural gas), and converters
must seek a certificate for all of their vehicle models. Model year 1993 and newer vehicles
that are converted are required to meet the standards applicable  at the time the vehicle was
originally certified.  Converters serve a niche market, and these  businesses primarily convert
vehicles to operate on compressed natural gas (CNG) and liquefied petroleum gas (LPG), on a
dedicated or dual fuel basis.

       EPA is proposing to exempt from the proposed GHG standards any manufacturer,
domestic or foreign, meeting SBA's size definitions  of small business as described in 13 CFR
121.201.  EPA adopted the same type of exemption for small businesses in the MY 2012-
2016 rulemaking.390 Together, we estimate that small entities comprise less than 0.1 percent
of total annual vehicle sales and exempting them will have a negligible impact on the GHG
emissions reductions from the standards.  Because we are proposing to exempt small
businesses from the GHG standards, we are proposing to certify that the rule would not have a
significant economic impact on a substantial number of small entities. Therefore, EPA has
not conducted a Regulatory Flexibility Analysis or a SBREFA SBAR Panel for the rule.

       Based on input we have heard from at least one small business vehicle manufacturer,
EPA is proposing to allow small businesses to voluntarily waive their small entity exemption
and optionally certify to the GHG standards. This would allow  small entity manufacturers to
earn CC>2 credits under the GHG program, if their actual fleetwide CC>2 performance was
better than their fleetwide CO2 target standard. EPA is proposing to make the GHG program
opt-in available starting in MY 2014, as the MY 2012,  and potentially the MY 2013,
certification process will have already occurred by the time this rulemaking is finalized. EPA
is also proposing that manufacturers certifying to the GHG standards for MY 2014 would be
eligible to generate early credits for vehicles sold in MY 2012 and MY 2013.  EPA is
proposing that manufacturers waiving their small entity exemption would be required to meet
all aspects of the GHG standards and program requirements  across their entire product line.
However, the exemption waiver would be optional for small entities and thus we believe that
manufacturers would only opt into the GHG program if it is  economically advantageous for
them to do so, for example in order to generate and sell  CC>2 credits.  Therefore, EPA believes
                                         9-2

-------
                                              2017 Draft Regulatory Impact Analysis
adding this voluntary option does not affect EPA's determination that the proposed standards
would impose no significant adverse impact on small entities.
                                       References



389 "List of Potential Small Businesses in the Light-duty Vehicle Market," Memorandum from
Chris Lieske to Docket EPA-HQ-OAR-2010-0799.

390 75 FR 25424, May 7, 2010.
                                        9-3

-------