Nonconformance Penalties for Heavy-
Duty Diesel Engines Subject to the 2010
NOx Emission Standard
Response to Comments
&EPA
United States
Environmental Protection
Agency
-------
Nonconformance Penalties for Heavy-
Duty Diesel Engines Subject to the 2010
NOx Emission Standard
Response to Comments
Assessment and Standards Division
Office of Transportation and Air Quality
U.S. Environmental Protection Agency
NOTICE
This technical report does not necessarily represent final EPA decisions or
positions. It is intended to present technical analysis of issues using data
that are currently available. The purpose in the release of such reports is to
facilitate the exchange of technical information and to inform the public of
technical developments.
&EPA
United States
Environmental Protection
Agency
EPA-420-R-12-015
August 2012
-------
List of Commenters
Commenter
Anonymous public comment
Clean Air Task Force, et al.
Cummins, Inc
Daimler Trucks North America
LLC and Detroit Diesel Corp.
Environmental Defense Fund
Ford Motor Company
International Council on Clean
Transportation
Jennifer V. Sinisi
Jennifer Whittaker
Mack Trucks, Inc. and Volvo
Group North America, LLC
Manufacturers Of Emission
Controls Association
Navistar, Inc.
Northeast States for Coordinated
Air Use Management
New Jersey Department of
Environmental Protection
PACCAR, Inc
Abbreviation/Acronym
Anonymous public comment
CATF
Cummins
Daimler
EOF
Ford
ICCT
J. V. Sinisi
J. Whittaker
Mack
MECA
Navistar
NESCAUM
NJDEP
PACCAR
Docket number of comment(s)
EPA-HQ-OAR-20 11-1000- . . .
0013
0021-0022
0015, 0025 & 0048
0028, 0043-0045 & 0049
0026
0029
0030
0020
0023
0024, 0046, 0047 & 0051
0017 & 0040
0027, 0031-0039, 0042 &
0050
0018-0019
0016
0041
-------
Table of Contents
1. Legal Issues 2
1.1. Statutory Basis for NCPs 2
1.1.1. Removing All Competitive Disadvantages for Complying Manufacturers... 3
1.1.2. Consideration of Environmental Impacts 5
1.2. Regulatory Criteria for NCPs 6
1.2.1. EPA Required to Follow Regulatory Criteria 7
1.2.2. New or Revised Standard More Stringent than Previous Standard and for
Which Substantial Work is Required 7
1.2.3. Technological Laggard 13
1.2.4. Other Criteria 20
1.3. Interim Final Rule 20
2. Penalty Level and Cost Analysis 21
2.1. Appropriateness of Penalty Level 21
2.2. Differences from Cost Methodologies Used in Prior NCP Rulemakings 28
2.3. Consideration of Manufacturer Costs 30
2.4. Upper Limit 33
2.5. Basing Compliance Costs on the Upper Limit Engine 36
2.6. Baseline Engine for Cost Analysis 39
2.7. Hardware Costs 44
2.8. Operating Costs 46
2.8.1. Cost Savings and Performance Improvements 46
2.8.2. DEF Usage and Fuel Consumption Rates 51
2.8.3. DEF and Fuel Prices 57
2.8.4. Discounting Operating Costs 59
-------
2.9. Other Costs 62
2.9.1. Warranty and Post-Warranty Repairs 62
2.9.2. Research and Development 63
2.9.3. Costs not included 65
2.10. Comparison of Commenters' and EPA's Estimates of Total Compliance Costs
and Penalties 68
3. Regulatory Issues 72
3.1. Base Year for Calculating NCPs 72
3.2. Retaining Emission Credit While Using NCPs 75
3.3. Prohibition against generating emission credits from NCP engines 75
3.4. NCP Formula 76
3.5. Significant figures of the upper limit 77
3.6. Dollar values 77
4. Other Issues 78
4.1. Sunsetting NCPs 78
4.2. Alternative Penalties 78
4.3. Regional Impacts of NCPs 80
4.4. Comments about SCR unrelated to costs 80
4.5. Timing of NCP Rulemaking 81
4.6. Late Comments 82
APPENDIX: Section 206(g) of the Clean Air Act 83
in
-------
Introduction
On January 31, 2012 EPA proposed to establish nonconformance penalties (NCPs) for
heavy-duty diesel engines. On that same day, we also promulgated an Interim Final Rule to
establish interim NCPs for heavy heavy-duty diesel engines. We held a public hearing on March
5, 2012, at which oral comments on the NPRM were received and recorded. Additionally, a
written comment period remained open until April 4, 2012. A complete list of organizations,
their abbreviations, and individuals that provided comments on the NPRM is contained in this
document.
This Response to Comments contains a summary of all comments we received on the
NPRM as well as our responses to these comments. Note that quotation marks are not used in all
cases in which comments are repeated verbatim; rather quotation marks are used only where we
believe it to be worth noting that the commenter's exact words are repeated. In some cases
where several commenters raised the same point, we did not separately summarize the
comments. Also, to the extent that commenters' written comments repeated comments made
during the public hearing, our response focuses on the written comments, which were generally
more detailed. The comments and responses are organized by topic (see Table of Contents) to
help the reader find comments and responses of interest.
-------
1. Legal Issues
1.1. Statutory Basis for NCPs
Section 206(g) of the Clean Air Act (the Act), 42 U.S.C. 7525(g), requires EPA to
promulgate regulations permitting manufacturers of heavy-duty engines (HDEs) or heavy-duty
vehicles (HDVs) to receive a certificate of conformity for HDEs or HDVs that exceed a federal
emissions standard, but do not exceed an upper limit associated with that standard, if the
manufacturer pays a nonconformance penalty (NCP) established by rulemaking. Congress
adopted section 206(g) in the Clean Air Act Amendments of 1977 as a response to a concern
with requiring technology-forcing emissions standards for heavy-duty engines. The concern was
if strict technology-forcing standards were promulgated, then some manufacturers might be
unable to comply initially and would be forced out of the marketplace. NCPs were intended to
remedy this concern. With this provision, the nonconforming manufacturers would have a
temporary alternative that would permit them to sell their engines or vehicles by payment of a
penalty. At the same time, conforming manufacturers would not suffer a competitive
disadvantage compared to nonconforming manufacturers, because the amount of the NCPs
would be determined, in part, based on the money saved by the nonconforming manufacturer.
Under section 206(g)(l), NCPs may be offered for HDVs or HDEs. The penalty may
vary by pollutant and by class or category of vehicle or engine. Section 206(g)(3) requires that
NCPs:
• Account for the degree of emission nonconformity;
• Increase periodically to provide incentive for nonconforming manufacturers to
achieve the emission standards; and
• Remove the competitive disadvantage to conforming manufacturers.
Section 206(g) authorizes EPA to require testing of production vehicles or engines in
order to determine the emission level upon which the penalty is based. If the emission level of a
vehicle or engine exceeds an upper limit of nonconformity established by EPA through
regulation, the vehicle or engine would not qualify for an NCP under section 206(g) and no
certificate of conformity could be issued to the manufacturer. If the emission level is below the
upper limit but above the standard, that emission level becomes the "compliance level," which is
also the benchmark for warranty and recall liability. The manufacturer who elects to pay the
NCP is liable for vehicles or engines that exceed the compliance level in use. The manufacturer
does not have in-use warranty or recall liability for emissions levels above the standard but
below the compliance level.
-------
1.1.1. Removing All Competitive Disadvantages for Complying
Manufacturers
This section addresses general comments related to competitive advantages. More
specific comments are addressed later in this document.
Summary of Comments
CATF EPA-HQ-OAR-2011-1000-0021-0022
Cummins EPA-HQ-OAR-2011-1000-0015, 0025 & 0048
Daimler EPA-HQ-OAR-2011-1000-0028, 0043-0045 & 0049
EOF EPA-HQ-OAR-2011-1000-0026
Ford EPA-HQ-OAR-2011-1000-0029
ICCT EPA-HQ-OAR-2011-1000-0030
Mack EPA-HQ-OAR-2011-1000-0024, 0046, 0047 & 0051
MECA EPA-HQ-OAR-2011-1000-0017 & 0040
NESCAUM EPA-HQ-OAR-2011-1000-0018 - 0019
PACCAR EP A-HQ-O AR-2011-1000-0041
Several manufacturers of SCR engines commented that the proposed NCPs do not meet
the statutory requirement to remove the competitive disadvantage to complying manufacturers.
These commenters believe that EPA proposed penalties that are too low to remove the
competitive disadvantage. Some of these comments focused on the competitive advantage of
engines that do not use SCR. Others addressed potential competitive advantages for SCR
engines that could be reoptimized to have NOx emissions near 0.50 g/hp-hr. These commenters
further stated that failing to fully remove all competitive advantages for noncomplying engines
would create an incentive for noncompliance.
Cummins emphasized that neither manufacturers of compliant engines, nor operators
who purchase them should be penalized relative to noncompliant engines.
Mack also commented that low NCP levels will encourage compliant manufacturers to
develop engines that could take advantage of a 0.50 g/hp-hr upper limit and gain a market
advantage. Mack cited the testimony from Cummins and its own market decisions to indicate
this is not hypothetical. Mack also noted that the availability of NCPs affects the analysis of
market share impacts.
Mack commented that EPA's analysis of market advantages is fundamentally flawed
because it "fails to consider what the market share would have been if all manufacturers had to
meet the same standard." It also stated that "EPA apparently assumes that if this happened each
manufacturer's share of the market would not have changed, which is not true." Mack
considered two scenarios. Under the first, Navistar would have been forced to comply with the
0.20 g/bhp-hr standard, which Mack argued would have reduced Navistar's share of the heavy-
HDDE market to zero. Under the second scenario, all other manufacturers would have been
given the opportunity to comply with a 0.50 g/bhp-hr standard and reaped a "dramatic shift in
market share."
-------
Mack also commented that EPA "must both demonstrate a true technological need for
NCPs (which it has not) and, if able to do that, must ensure the NCP is high enough to
discourage production of engines beyond those absolutely necessary (which it again has not)."
EDF commented that NCPs must be large enough to discourage complying
manufacturers from backsliding to the upper limit. Ford made similar comments.
NESC AUM stated that quantifying the competitive advantage of an engine manufacturer
is a potentially significant problem with respect to engines are not equipped with selective
catalytic reduction (SCR) systems for controlling NOx emissions. It commented that consumers
may disproportionately choose to purchase vehicles powered by this manufacturer's engines (the
nonconformance penalty notwithstanding) on the basis of a perceived convenience of not having
to maintain appropriate levels of diesel exhaust fluid for proper functioning of the SCR system.
There were also comments involving competitive impacts that are more directly related to
the actual level of the penalty, but these are addressed in Chapter 2.
Response
The purpose of adopting NCPs is to allow a noncompliant manufacturer to continue
selling its engines. However, the Clean Air Act directs EPA to set the NCPs at a level that will
"remove any competitive disadvantage" to complying manufacturers. Thus, the statute
effectively requires us to set the penalties at a level that we reasonably expect to protect the
complying manufacturers, but not so high that it cripples any noncomplying manufacturers. The
purpose of the NCP is not to punish noncomplying manufacturers, but to remove competitive
advantages and to incentivize compliance. To the extent there is uncertainty about compliance
costs and/or competitive impacts, we believe that the explicit statutory requirement to remove the
competitive disadvantage for complying manufacturers means that complying manufacturers
should generally be given the benefit of these uncertainties. Thus, we are somewhat
conservative (i.e. high) in our estimate of costs, including looking at our estimate of the highest
cost for complying manufacturers in developing the COC90 calculation.
For the proposal and interim rule, EPA estimated compliance costs relative to an SCR
baseline engine similar to today's compliant engines, but also evaluated the NCPs in the context
of competition between SCR engines with NOx emissions at or below 0.20 g/hp-hr and non-SCR
engines with NOx emissions at or just below 0.50 g/hp-hr. This was appropriate because these
two categories represented all heavy heavy-duty engines being sold currently. However, EPA
agrees with commenters that the final NCPs should be high enough to also protect complying
manufacturers from a competitive disadvantage relative to SCR engines that are more fully
optimized for 0.50 g/hp-hr than were considered for the proposal and interim rule. The final rule
should be, and is, more forward looking to also address potential competitive advantages that
could arise in the future. The Clean Air Act's requirements to "remove any competitive
disadvantage" to complying manufacturers effectively requires EPA to consider not only existing
engines with NOx emissions over the standard, but also engines that could reasonably be
developed during the period in which NCPs are available. In this case, this requires EPA to also
consider SCR engines that would be reoptimized for 0.50 g/hp-hr. With respect to Mack's,
Cummins's, and EDF's comments, we believe that the level of the final penalty, which was
-------
developed using a baseline engine emitting at 0.50 g/hp-hr that has been fully optimized, is high
enough to discourage optional use of NCPs by compliant manufacturers.
Mack is incorrect when it stated that we assumed that each manufacturer's share of the
market would not have changed if all manufacturers had to meet the same emission level. What
would have happened if all manufacturers had to meet the same emission level because NCPs or
credits were not available is not relevant. We believe that Congress intended us to consider
competitive impacts based on the market conditions that existed before the new standard took
effect, which is what we have done.
It is unclear to us what Mack meant when it commented that EPA must ensure the NCP is
high enough to "discourage production of engines beyond those absolutely necessary". If it
meant that we should calculate some number of engines that would be necessary for Navistar to
stay in business and set the NCP at a level that made it unlikely that Navistar would sell more
than that number, then we disagree with the comment. It is clear that Congress intended us to
allow Navistar (or any other noncompliant manufacturer) to continue to sell its engines but also
intended us to set the NCP at a level that ensures that Navistar (or any other noncompliant
manufacturer) does not gain any competitive advantage over compliant manufacturers. We
believe the final rule meets this intent.
With respect to the NESCAUM comment, the commenter provides no evidence, nor have
we seen any evidence, that there is a competitive disadvantage for SCR manufacturers beyond
the difference in compliance costs.
1.1.2. Consideration of Environmental Impacts
Summary of Comments
EOF EPA-HQ-OAR-2011-1000-0026
Ford EPA-HQ-OAR-2011-1000-0029
ICCT EPA-HQ-OAR-2011-1000-0030
J. V. Sinisi EPA-HQ-OAR-2011-1000-0020
Mack EPA-HQ-OAR-2011-1000-0024, 0046, 0047 & 0051
MECA EPA-HQ-OAR-2011-1000-0017 & 0040
NESCAUM EPA-HQ-OAR-2011-1000-0018 - 0019
NJDEP EPA-HQ-OAR-2011-1000-0016
Some commenters argued that EPA should consider environmental impacts when
deciding whether to establish NCPs and/or when setting the penalty level. In particular, Mack
commented that allowing Navistar to continue production of current engines using NCPs would
result in more than 55,440 excess tons of NOx emissions for model year 2012 and more than
110,000 tons for model year 2013. Its argues that "this is not in keeping with EPA's central
mission of protecting the environment and the air we breathe."
-------
ICCT commented that it believe that the Agency should also consider taking into account
the societal costs of increased air pollution resulting from noncompliance. Given the goals of the
original rule and the Agency's mission, ICCT believes that any accounting of costs is incomplete
without estimating the damages to human health and the environment that would result from
increased NOx emissions. ICCT suggested using the methodology developed in the regulatory
impact assessment as the basis for evaluating damages from noncompliance.
MECA stated that NCPs should not be based simply on costs of emission reduction
technologies because such an approach doesn't take into effect the societal costs of higher
emissions.
Response
The Clean Air Act does not direct EPA to consider environmental impacts in setting
NCPs. Congress was clear in its mandate in section 206(g) that EPA shall permit manufacturers
to certify using NCPs. While Congress did require EPA to take certain factors into account
when setting NCPs, evaluation of environmental impacts or "societal costs" is not set out as a
criterion. The language and history of this provision indicate that Congress effectively
determined that requiring EPA to issue standards for heavy-duty vehicles and engines that are
technology forcing in nature, with NCPs acting in part as a safety valve preventing a
manufacturer from being forced from the market, serves the public interest in the long term and
outweighs any short-term harm to the environment caused by allowing NCPs. We note that we
do escalate the amount of the NCP annually "in order to create incentives for the development of
production vehicles which achieve the required degree of emission reduction."
1.2. Regulatory Criteria for NCPs
Since the promulgation of the first NCP rule in 1985, subsequent NCP rules generally
have been described as continuing "phases" of the initial NCP rule. The first NCP rule (Phase I),
sometimes referred to as the "generic" NCP rule, established three basic criteria for determining
the eligibility of emission standards for nonconformance penalties in any given model year (50
FR 35374, August 30, 1985). (For regulatory language, see 40 CFR 86.1103-87.) The first
criterion is that the emission standard in question must become more difficult to meet. This can
occur in two ways, either by the emission standard itself becoming more stringent, or due to its
interaction with another emission standard that has become more stringent. Second, substantial
work must be required in order to meet the emission standard. EPA considers "substantial work"
to mean the application of technology not previously used in that vehicle or engine
class/subclass, or a significant modification of existing technology, in order to bring that
vehicle/engine into compliance. EPA does not consider minor modifications or calibration
changes to be classified as substantial work. Third, EPA must find that a manufacturer is likely
to be noncomplying for technological reasons (referred to in earlier rules as a "technological
laggard"). Prior NCP rules have considered such a technological laggard to be a manufacturer
who cannot meet a particular emission standard due to technological (not economic) difficulties
and who, in the absence of NCPs, might be forced from the marketplace.
-------
1.2.1. EPA Required to Follow Regulatory Criteria
Summary of Comments
Cummins EPA-HQ-OAR-2011-1000-0015, 0025 & 0048
Daimler EPA-HQ-OAR-2011-1000-0028, 0043-0045 & 0049
Mack EPA-HQ-OAR-2011-1000-0024, 0046, 0047 & 0051
MECA EPA-HQ-OAR-2011-1000-0017 & 0040
PACCAR EP A-HQ-O AR-2011-1000-0041
Several commenters argued (implicitly or explicitly) that EPA cannot establish NCPs
unless all of the regulatory criteria for NCPs (in 40 CFR 86.1103-87) are met. Mack also stated
that EPA "must demonstrate" that the criteria are met.
Response
When adopted in 1985, EPA intended to use the criteria of 40 CFR 86.1103-87 in
determining whether to establish NCPs. However, we note that actual regulatory text does not
state that EPA may establish NCPs only //all criteria are met, but rather that EPA shall establish
NCPs "provided that EPA finds" the criteria are met. These criteria were included in the
regulations to clarify that manufacturers should not expect EPA to initiate a rulemaking to
establish NCPs where these criteria were not met. Moreover, EPA may revise the criteria at any
time through notice and comment rulemaking as long as the statutory criteria of section 206(g)
are met. Nevertheless, as noted elsewhere, EPA has found that the criteria have been met.
Regarding Mack's comment that EPA must demonstrate that the criteria are met, the
regulations clearly defer to EPA's judgment for finding that the criteria are met. While we must
explain the basis of our finding, the regulatory language does not require us to prove or
demonstrate that the criteria are met.
1.2.2. New or Revised Standard More Stringent than Previous
Standard and for Which Substantial Work is Required
Summary of Comments
Cummins EPA-HQ-OAR-2011-1000-0015, 0025 & 0048
Daimler EPA-HQ-OAR-2011-1000-0028, 0043-0045 & 0049
EOF EPA-HQ-OAR-2011-1000-0026
J. Whittaker EPA-HQ-OAR-2011-1000-0023
Mack EPA-HQ-OAR-2011-1000-0024, 0046, 0047 & 0051
Navistar EPA-HQ-OAR-2011-1000-0027, 0031-0039, 0042 & 0050
PACCAR EP A-HQ-O AR-2011-1000-0041
Several commenters opposed EPA's determinations that the 2010 NOx standard qualifies
as a new or revised standard that is more stringent than the previous standard and that substantial
work is required to meet the standard. The commenters made arguments similar to one another.
-------
First, they argued that the 2010 NOx standard does not qualify as a new standard because it was
promulgated in 2001. Second, they argued that nothing happen in 2012 to make the standard
"more difficult to achieve". Third, with regard to the "substantial work" criterion, these
commenters stated that substantial work is no longer required to meet the standard.1 Most
repeated EPA's statement in the preamble to the proposal that:
EPA considers "substantial work" to mean the application of technology not previously
used in that vehicle or engine class/subclass, or a significant modification of existing
technology, in order to bring that vehicle/engine into compliance. EPA does not consider
minor modifications or calibration changes to be classified as substantial work.
These commenters argued that viewed from 2012, this "substantial work" criterion has not been
met because SCR is now an existing technology that has been used within the engine class.
Finally, more than one commenter noted that Navistar cannot claim that it would need to do
substantial work to use SCR because one of its subsidiaries uses SCR on trucks in Brazil. Other
comments that we received are summarized below.
Mack stated that "EPA's 'substantial work' finding is at odds with its own use of a
'hypothetical' 0.5 g/bhp-hr engine, and its assumption in setting the NCP level that all
manufacturers save Navistar would have adopted SCR to meet a 0.5 g/bhp-hr standard" and that
"EPA cannot, on one hand, declare that SCR did not previously exist for purposes of justifying
its substantial work finding in the NCP rule, and, on the other, decide that it always existed and
would only require minor modification for purposes of developing an NCP calculation." Mack
also commented that it is incorrect to presume that Navistar will need to innovate further and use
"technology not previously used in an engine or vehicle class or subclass" to meet the 0.20
g/bhp-hr standard. Cummins made a similar argument that it is inconsistent to find that
substantial work is required to meet the 2010 NOx standard while proposing as its "baseline"
engine an engine capable of emitting 0.20 g/hp-hr NOx that is de-tuned to meet the upper limit.
Mack also stated that "Navistar's public statements also confirm that it also does not
believe substantial work is required for its current EGR engines to meet the 0.20 g/bhp-hr NOx
standard." It quoted Navistar's Senior Vice President of North American Sales Operations as
saying:
"[d]rivers and owners 'won't see the difference' in the point-2 engine's performance ...
because there are no equipment changes... The lower NOx emissions can be achieved
with modified fuel pressures, altered introduction of inlet air, and recalibration of
electronic controls."2
1 In particular, Cummins noted that verb tense is significant and determinative of the regulation's meaning, stating
that "the Supreme Court has recognized as much in the analogous task of interpreting statutes. Ingalls Shipbuilding
v. Director, OWCP, 519 U.S. 248, 255 (1997) (tense of verb is an element of plain meaning); United States v.
Wilson, 503 U.S. 329, 333 (1992) ("Congress's use of a verb tense is significant"); see also Carrv. United States,
130 S. Ct. 2229, 2236-37 (2010) (collecting cases)."
2 Mack cited: Truckinglnfo.com, "Customers Wouldn't Pay Extra for Any Non-Compliance Penalties Imposed on
Navistar, Hebe Says,"(Feb. 1, 2012)
-------
Daimler stated that "the regulatory provisions cannot reasonably be read to allow for such
a belated NCP determination" and that EPA's conclusion in the standard-setting rule that the
standard is a technology forcing standard "has no bearing on whether there now exists a new or
revised and more stringent emission standard applicable to the current model year for which EPA
seeks to set NCPs." It also stated that for EPA to find that substantial work is still required
"would justify NCPs any time new technology is used to meet an emission standard - a result
Congress surely did not intend under the Clean Air Act and a result EPA does not permit under
its regulations."
Navistar agreed with EPA's determination that the first two criteria for establishing NCPs
were met, stating that "under the plain language set forth in EPA's regulations as well as the
agency's longstanding application and interpretation of both CAA § 206(g) and the agency's
own regulations, this is clearly a case where Navistar is applying new and/or significantly
modified technology—that it spent hundreds of millions of dollars developing—in order to
comply with the new stringent NOx standard set at near-zero levels." Finally, Navistar noted that
"the 0.20 g NOx Standard did not become fully effective across all engine families in model year
2007, or even in 2010" due to the availability of banked NOx emission credits. It argues that
SCR engine manufacturers' disregard of these regulatory flexibilities is "illogical" and "jettisons
EPA's well-established interpretation of the NCP statute and rules."
Response
When EPA initially proposed the NCP criteria, we noted that the first two criteria
addressed whether there was ^possibility for a technological laggard to develop.3 We stated that
when the first criterion is met, it "creates the possibility for a technological laggard to exist." We
also stated that "When manufacturers must perform substantial work, it is possible that at least
one will be unsuccessful and will become a laggard. Thus, when evaluating these criteria, the
purpose is to determine whether the standard created the possibility for a laggard to exist."
The first regulatory NCP criterion is met when "any new or revised emission standard is
more stringent than the previous standard for the pollutant, or when an existing standard for that
pollutant becomes more difficult to achieve because of a new or revised standard." This is the
case with the 2010 NOx standard. The previous emission standard for this category is a
combined NMHC + NOx standard of 2.4 g/hp-hr, or optionally a 2.5 g/hp-hr NMHC + NOx with
a limit of 0.5 g/hp-hr NMHC.4 (EPA has viewed this standard as being equivalent to a 2.3 g/hp-
hr NOx-only standard.5) The 2010 (i.e., current) standards are 0.20 g/hp-hr for NOx and 0.14
g/hp-hr for NMHC. When promulgated, the Agency concluded that the 0.20 g/hp-hr NOx
standard was a technology forcing standard.
3 50 FR 9206, March 6, 1985.
4 NMHC stands for non-methane hydrocarbons, which is a measure of total hydrocarbons with the methane
emissions subtracted out. For typical on-highway diesel fueled heavy-duty engines, methane emissions are on the
order of 10 percent of the total hydrocarbon emissions.
5 "Regulatory Impact Analysis: Control of Emissions of Air Pollution from Highway Heavy-Duty Engines," page
109, EPA420-R-00-010, July 2000.
-------
The second criterion requires that substantial work be need to meet the new standard.
Since all heavy heavy-duty diesel engines currently certified to the 0.20 g/hp-hr standard
(without using credits) are using new aftertreatment systems to meet this standard, it is
appropriate to conclude that the standard qualifies as more difficult to meet and that substantial
work was required to meet the emission standard. Thus, when we adopted the 2010 NOx
standard, it created the possibility for a technological laggard to develop.
We note that the commenters generally do not dispute that the first two criteria were met
before 2010, and that we could have set NCPs at that time had we determined that a
technological laggard was likely to develop. In fact, none of the manufacturers questioned
EPA's determination when we notified them in 2010 that we had found that the new NOx
standard required substantial work.6 When making that determination, EPA noted that the
reason that we were not establishing NCPs at that time was because we had not determined that a
technological laggard was likely to develop. Had we known before 2010 that Navistar would run
out of credits in 2012 and that it would have not yet brought its NOx levels down to 0.20 g/hp-hr,
we would have established NCPs. We do not believe that these commenters would have
opposed establishing NCPs at that time.
Regarding the first criterion, some commenters now claim that EPA no longer has the
authority to establish NCPs for the 2010 NOx standard because they claim it is no longer a new
or revised standard. One commenter noted that the standard was promulgated in 2001. Yet it is
often not apparent, and was not apparent in this case, that there would be a need for NCPs until
close to, or in this case after, the implementation of a standard. In this case, EPA did not
promulgate NCPs earlier than 2012 because no manufacturer needed them, as they all met the
standard either directly or through credits. The 0.20 g/hp-hr NOx standard clearly is a revised
standard, and a more stringent one, since the previous standard was approximately ten times
higher than the 0.20 g/hp-hr standard. The fact that EPA did not promulgate NCPs prior to the
standard being implemented does not forever preclude EPA from promulgating them when they
are needed.
Although we reject the commenters' argument that we may not establish NCPs once a
standard has been implemented, it is worthwhile to consider the logical result of this argument.
Many of these commenters seem to believe that the NOx standard was new in 2010. However,
strictly speaking this standard was not new in 2010. It was actually first applicable to model year
2007 engines. The commenters provide no basis for treating the phase-in provisions that applied
in model year 2007 through 2009 differently than the flexibility of the emission credit program
that Navistar has used since then to certify engines near 0.50 g/hp-hr NOx. Thus, the logic of the
commenters opposing our finding that the first criteria was met would lead to a result in which
we could not have set NCPs after 2007. If we accepted this logic, it would severely hinder our
ability to set NCPs for standards for which there is extensive phase-in flexibility. We do not
believe that such a result would be consistent with the intent of the statute.
6 "Nonconformance Penalties for Heavy-Duty Diesel Engines in 2010 Model Year", Letter from Karl J. Simon,
Director, EPA Compliance and Innovative Strategies Division, February 22, 2010.
10
-------
Regarding the substantial work criterion, there can be no argument that substantial work
was necessary to meet the 2010 NOx standard. Every manufacturer other than Navistar has for
the first time included NOx aftertreatment (selective catalytic reduction), on their engines to
meet the standard, and Navistar has also greatly modified in exhaust gas recirculation (EGR)
system to reduce its NOx emissions and has now stated that it will also use NOx aftertreatment
to meet the standard. These are substantial changes to the emission control systems of these
engines. While manufacturers may currently be using SCR systems, they were not doing so until
they were required to meet the 2010 NOx standard. Therefore, it is clear that substantial work
was needed to meet the standard.
Commenters do not dispute that substantial work was required to meet the 2010 standard,
and in fact, complying engine manufacturers generally emphasized in their comments how much
work they had done to meet the standards. Nevertheless, some commenters claim that this
criterion has not been met because they claim it is no longer true that substantial work is required
because some manufacturers have met the standard. We disagree for two reasons. First, this is
not how EPA interprets this criterion. This criterion is to be evaluated based on actual work
needed to go from meeting the previous standard to meeting the current standard, regardless of
the timing of such changes. EPA looks at whether "substantial work" is or was required to meet
the revised standard at any time after the standard was issued - the important question is whether
manufacturers who were using technology that met the previous standard would need to build
upon that technology to meet the revised standard. Indeed, the commenters' argument would
seem to be directly contrary to the purpose of the statute, which is designed to allow
technological laggards to be able to certify engines even if other manufacturers have met the
standard. Commenters' approach would prevent this by apparently preventing lagging
manufacturers from certifying in exactly those circumstances. To avoid this confusion for future
NCPs, we are clarifying in the regulatory text that this criterion is to be evaluated based on the
need for new or modified technology or design to meet the new or revised standard regardless of
the timing for such changes.
Second, even under the current circumstances, we find that Navistar will need to do
substantial work to meet the standard, whether it meets the standards through advanced EGR or
through SCR. Mack cited public statements by a vice president of Navistar responsible for sales,
but we do not believe these statements reflect the full extent of the challenge facing Navistar in
its attempt to meet the standard through advanced EGR. Navistar has been working to bring all
its engines into compliance with the 0.20 g/hp-hr NOx standard without credits, and is devoting a
substantial fraction of its engineering resources to the task, but it has not yet been able to do so.
Moreover, Navistar has subsequently withdrawn its application for a 0.20 g/hp-hr
certificate for its non-SCR engines and announced that it has begun work to redesign its engines
and vehicles to use SCR. While our analysis of this technological challenge is discussed in more
11
-------
detail in Section 1.2.3, we note here that Navistar must complete the work that other
manufacturers have already done, including:
Selecting an SCR system design
• Making arrangements with component suppliers
• Validating components
• Recalibrating its engine to work with the SCR system
Redesigning it trucks to fit the SCR hardware
• Completing its emission testing and durability testing for certification
• Obtaining EPA approval for the new engine-SCR system
Thus, we also disagree with the views of the commenters who claim that Navistar could now add
SCR to its engines without substantial work, simply because its competitors have already done
so. Even if it Navistar could take full advantage of the research of its competitors, this would not
eliminate the substantial work needed for such things as redesigning its vehicles and retooling its
factories. Regarding the use of SCR by one of Navistar's subsidiaries in Brazil, this does not
mean that Navistar can apply SCR to its trucks for the U.S. market in the near future. Neither the
trucks being built in Brazil nor the market served by those trucks, are the same as those being
sold in the United States. Even though Navistar has announced that it will incorporate SCR
technology in its U.S. engines and trucks, it will take substantial time and resources to
accomplish this. More importantly, while these comments have some relevance in the context of
determining whether Navistar is a technological laggard, they are not relevant to whether
substantial work was required to meet the 2010 standard. Whether manufacturers have used
SCR or advanced EGR to reduce emissions, it is clear that the changes to their engines have
required considerable work.
With respect to Mack's assertion that "EPA cannot" declare that SCR required substantial
work and "decide that it always existed", we note that we never declared it always existed.
Cummins made a similar comment, but both Cummins and Mack are mixing two separate
provisions. First, our evaluation of the regulatory criteria for NCPs must necessarily be based on
the factual differences between the prior standard and the new standard (as well as the
technology required to meet them). This is clearly separate from issues related to the baseline
engine that EPA used to estimate compliance costs relative to the upper limit because the upper
limit is well below the prior standard. The engines being considered when evaluating whether a
new standard requires substantial work are engines meeting the previous standard (i.e. model
year 2006 or earlier engines), while the baseline engine represents engines that are currently
practicable (i.e. current model year engines).
We also disagree with Daimler's comment that EPA's logic "would justify NCPs any
time new technology is used to meet an emission standard- a result Congress surely did not
intend." It is important to emphasize that this comment was made in the context of the first two
criteria, which merely address the possibility for a laggard to develop. It ignores the third
criterion that requires a likelihood of a technological laggard. We do not believe that using a
new technology would justify NCPs if the manufacturer had already met the standard and was
simply choosing to change its emission controls because we would not consider such a
manufacturer to be a technological laggard. Additionally, it is clear that Congress indeed
intended that manufacturers be permitted to use NCPs when new technology is used to meet an
12
-------
emission standard. Nothing in section 206(g) requires that a manufacturer whose technology is
lagging be denied the ability to use NCPs because its technology is different than the technology
of compliant manufacturers.
Finally, the regulatory text states that we will establish NCPs when "EPA finds" that
substantial work was needed. Thus, the regulations clearly leave the determination up to the
Agency's discretion, provided we have a reasonable basis for making such a finding. See
Section 1.2.1 for additional discussion of the extent to which EPA the regulatory criteria
constrain EPA.
1.2.3. Technological Laggard
Summary of Comments
Cummins EPA-HQ-OAR-2011-1000-0015, 0025 & 0048
Daimler EPA-HQ-OAR-2011-1000-0028, 0043-0045 & 0049
EOF EPA-HQ-OAR-2011-1000-0026
J. V. Sinisi EPA-HQ-OAR-2011-1000-0020
J. Whittaker EPA-HQ-OAR-2011-1000-0023
Mack EPA-HQ-OAR-2011-1000-0024, 0046, 0047 & 0051
Navistar EPA-HQ-OAR-2011-1000-0027, 0031-0039, 0042 & 0050
PACCAR EP A-HQ-O AR-2011-1000-0041
Several commenters opposed our finding that there is a likely to be a technological
laggard, arguing that the technology exists for all manufacturers to comply with the 2010 NOx
standard.
Mack
Mack commented that EPA must demonstrate that a "true" technological laggard exists,
arguing that EPA has identified nothing to date to suggest that Navistar is a technological
laggard. It noted further Navistar has "never publicly expressed any inability to meet the
standard" and has actually submitted two applications for certificates of conformity for engine
families that purport to comply with the 0.20 g/bhp-hr 2010 NOX standard. Mack argued that
EPA has failed to examine the relevant data and articulate a satisfactory explanation for its
action, including a rational connection between the facts found and the choice made and that
EPA must provide technical analysis to demonstrate that Navistar:
1. Selected technology that cannot meet the 0.20 g/bhp-hr standard for purely
technological reasons.
2. Was incapable of using a proven, demonstrated, and available technology, such as
SCR, to meet the standard.
3. Actually submitted a formal request for NCPs or claimed that it was unable to
meet the standard.
13
-------
On the other hand, Mack did acknowledge that "it was readily apparent as far back as (and
indeed, even earlier than) 2009, when Navistar sued EPA seeking to overturn the 2010 standard
based on allegations that it was not feasible, that Navistar would need NCPs."
Cummins
Cummins stated that Navistar has chosen not to utilize SCR technology to comply with
the 0.20 g/hp-hr NOx standard and instead pursued a different path, not based on a lack of
technological ability to employ SCR, but because of a conscious business calculation. It quoted
a Navistar official as saying: "There's no question that SCR works. It's just not the choice our
management wanted to take. It's just too easy."7 It also noted that Navistar is currently using
SCR technology in some of its engines in South America, and has submitted an application for a
certificate of conformity that does not rely on the use of emissions credits to comply. Finally,
Cummins noted that because "the factual predicate for the agency's proposed action has
completely changed" (i.e. after the proposal, Navistar submitted a request for certification of a
0.20 g/hp-hr NOx engine family) EPA must reconsider its finding that NCPs are appropriate.
Nevertheless, in its hearing testimony, Cummins did not seem to object to EPA finalizing the
rule stating, "We urge the Agency to follow due process expeditiously and finalize a sound rule
that ensures all manufacturers are treated fairly . . ."
Daimler
Daimler emphasized the distinction between "true technological laggards and economic
laggards-particularly intentional economic laggards", stating that "one federal court has held,
"NCPs were intended to give a manufacturer that has made every effort to comply, but has been
unable to achieve compliance, a chance to continue to participate in the market." (United States
v. Caterpillar, Inc., 227 F. Supp. 2d 73,88 (D.D.C. 2002)). Daimler argued that "Navistar has
failed to demonstrate that it cannot use SCR" and noted that Navistar's wholly owned subsidiary
is already using SCR in Brazil. It argues that this means "that Navistar purposefully chose not to
use SCR in the United States, thinking it could gain some economic or competitive advantage."
Daimler further noted that Navistar stated in its annual reports that it "chose [ ] EGR, combined
with other technologies, as our solution to meet the 2010 emissions standards [because] [w]e
believe that our customer-friendly solution provides our products with a significant competitive
advantage in North America" over all of its competitors, which "have chosen liquid-based urea
SCR as the solution to meet 2010 emission standards." Daimler also stated that "a critical
factual predicate of the Agency's rulemakings has been proven incorrect" because a Navistar
official stated that the company could meet the standard but did not want to produce the engine
until it further improves fuel economy and noted that Navistar has submitted an application for a
0.20 g/hp-hr engine.
7 E. Ballam, 2010 Emission Standards Limit Apparatus Engine Choices, Fire Apparatus & Emergency Equipment,
Mar. 2009, available at http://www.fireapparatusmagazine.com/ index/display/article-
display/4258450219/articles/fire-apparatus/Volume-14/issue-03/features/2010-emissions-standards-limit-apparatus-
engine-choices.html.
14
-------
Ms. Whittaker
Ms.Whittaker identified several emission control technologies that she believes could be
used by Navistar to meet the emission standard and stated that a technological laggard does not
exist because Navistar could have used other technologies to meet the standard.
Ms. Sinisi
Ms. Sinisi commented that EPA should not establish NCPs "every time a single engine
manufacturer is in danger of becoming noncompliant."
EDF
EDF stated that NCPs provide a narrow exception to other requirements for truly exigent
circumstances where compliance is impossible for technological reasons
Navistar
Navistar acknowledged during the public hearing that it "needs" NCPs and agrees with
EPA's determination that there is a reasonable possibility that Navistar will be unable to comply
for technological reasons because it will run out of credits in model year 2012. Navistar stated
that without the interim NCPs, it would have already run out of HHDDE credits.
It acknowledged that it submitted an application for a certificate at 0.20 g NOx, but stated
that even were EPA to approve that application tomorrow, it would still take Navistar until June
or July 2012 to produce road-ready engines covered by the application. It also noted that the
submitted 0.20 g NOx application only covers one family of Navistar's 13 liter MaxxForce
heavy HDDEs. Navistar also has two other 13 liter families, one 11 liter family, and two 15 liter
families in its heavy HDDE line. Navistar indicated that it expected there would be more than a
year of lag time to prepare and have approved its certificates, including all other requirements
needed for introduction into commerce, for its other HHDDEs. Navistar also provided evidence
that it needs NCPs to ensure continued operations.
Navistar agrees with EPA's finding that its issues arise from technological as opposed to
economic reasons (noting that it had invested $700 million to develop its emission controls).
Navistar noted that long term economic interest influences all decisions of for-profit companies,
but that the delay in compliance came because of the technological decision to use a
technologically innovative path, rather than using "the more primitive SCR option." Navistar
stated that it needs NCPs as a temporary bridge to keep certain engine families in the
marketplace.
Navistar stated that NCPs are precisely intended to provide relief for manufacturers who
have chosen a different technology path for compliance. Navistar noted that if it were required
that a noncomplying manufacturer must adopt the technology of those who do comply, there
could never be a finding of a likelihood of a technological laggard. Navistar stated that it did not
"choose" not to comply, but chose a different technology.
15
-------
Navistar also stated that the fact that one of its international facilities manufacturers
engines using SCRs is not relevant as those engines are designed to meet Europeans standards
and requirements that are substantially different from EPA requirements and would require an
entirely new development program, taking approximately 22 months.
Finally, although Navistar does not expect a need for NCPs for its medium HDDE
families during the 2012 model year, it stated that it may need NCPs for at least some of these
families in the time period covered by the proposed rule.
Navistar also submitted a late comment in which it responded to the claims of SCR
engine manufacturers that Navistar does not need NCPs because it has submitted an application
for certification for one of its engines. Navistar responded by stating:
SCR engine makers arc blatantly wrong both legally and factually. As discussed in detail
below, should EPA accept Navistar's current certificate application for one rating of one
of its 13 liter engines - and it should - that in no way changes Navistar's need for, or the
appropriateness of issuing, NCPs for either the accepted 13 liter application (as a
temporary bridge between certification and actual production and sale for that one
configuration), or for Navistar's many other HDDE families and ratings none of which
have 0.20 g applications submitted. Moreover, the claims of SCR engine makers reflect
either a complete disregard, or a complete misunderstanding, of not only the law and
intent behind NCPs, but also EPA's application and interpretation of those rules. To
contend, as they do, that Navistar does not qualify for NCPs because it consciously
"chose" what unknowingly resulted in a late-complying technology path (when they say
they made it on time with a separate inferior path and that Navistar could have too) is
simply nonsensical. Their claims have no legal basis and, thus, should be summarily
rejected.
Navistar noted that its decision to embark alone on development of advanced EGR, rather
than using "off-the-shelf technology" has slowed development and increased the costs of the
necessary advancements. It argued this reinforces the appropriateness of NCPs in this instance.
Response
The Clean Air Act places no explicit restrictions on when EPA can set NCPs. In fact, it
seems to create a presumption that NCPs will be available. The Act requires EPA to allow
certification of engines that do not meet the standard (unless EPA determines the practicable
upper limit to be equal to the new emission standard). The commenters arguing that either EPA
or Navistar must prove the existence of a technological laggard are misreading both the statute
and the regulations. In 40 CFR 86.1103-87(a)(2), EPA specifies that we will issue NCPs,
provided we find "that there is likely to be a technological laggard." The regulations do not
require any explicit proof that a technological laggard currently exists, only a finding that one is
likely.
Nor do the regulations require any sort of formal request by a manufacturer that it needs
NCPs. EPA's decision is based on its own evaluation of the facts. In any case, commenters
cannot now dispute the fact that Navistar has effectively requested in its comments on this rule
16
-------
that NCPs be available. Moreover, commenters can also not dispute the fact that Navistar
certified heavy heavy-duty engines to use the interim NCPs, and therefore, can be considered to
have formally requested NCPs.
While the regulations do not define "technological laggard", the commenters correctly
noted that EPA has previously interpreted this as meaning a manufacturer who cannot meet the
emission standard due to technological difficulties, not merely economic difficulties (67 Fed.
Reg. 51,464-5, Aug. 8, 2002). The regulations do not require EPA to be certain that one or more
manufacturers will actually be unable to meet the standard for technological reasons. Rather, the
regulations specify that it is sufficient for EPA to find this will likely be the case. Moreover, the
regulatory text states that we will establish NCPs when "EPA finds" that there is likely to be a
technological laggard. Thus, just as with the finding that substantial work is needed, the
determination of the likelihood of a technological laggard developing is one that is made at
EPA's discretion, provided we have a reasonable basis for making such a finding.
EPA proposed NCPs for heavy-duty diesel engines because we found that there was a
significant likelihood that they would be needed by Navistar and that Navistar had not met the
requirements for technological reasons. Navistar was then using NOx credits to certify all of its
heavy heavy-duty diesel engines at nearly the FEL cap level of 0.50 g/hp-hr and did not have
sufficient credits to cover its entire model year 2012 production. The comments opposing this
determination can be summarized as the following three reasons:
1. Navistar should have used SCR to meet the standard.
2. Navistar could still use SCR (or other advanced emission controls) to meet the
standard.
3. Navistar could meet the standard without SCR, but is choosing to not meet the
standard for economic reasons.
With respect to the first reason, we agree that Navistar could have decided over two years
ago to apply SCR to its engines (as the rest of the industry did) and that it would likely be able to
meet the standard if it had. However, it made a decision to attempt to meet the emission standard
without SCR. The emission standard adopted by EPA is a performance standard, and does not
require that all manufacturers use the same technology to meet the standards. Congress,
understanding that manufacturers may not all be in the same place regarding compliance with
technology-forcing standards, specifically permitted manufacturers to emit higher levels of
pollutants using NCPs. EPA believes commenters are incorrect in interpreting the NCP
provisions to effectively discourage technological innovation by requiring all manufacturers to
use the same technology once one manufacturer has met the standard using that technology. As
Navistar noted, such an interpretation would seem to negate the entire purpose of NCPs, which
presume that at least one manufacturer has met the standard. Having made its decision to use a
different technology to meet the standards, Navistar has not yet developed and produced engines
that have been certified to meet the 0.020 standard. This is similar to the circumstances in 2002
17
-------
when Caterpillar developed its "ACERT" technology rather than use cooled exhaust gas
recirculation (EGR) technology and needed to use NCPs while developing ACERT.8'9
As Navistar noted, all for-profit companies look at economics in making long-term
decisions. However, this does not refute the clear evidence that Navistar made a decision to use
a different technological solution to meet the 0.20 g/hp-hr standard than other manufacturers, and
that this technological solution has delayed Navistar's ability to meet the standard. It is for this
technological reason that Navistar cannot meet the standard, not because it has made an
economic decision not to try.
With respect to Daimler's comment that a federal court has held that "NCPs were intended to
give a manufacturer that has made every effort to comply, but has been unable to achieve
compliance, a chance to continue to participate in the market", we believe that this was quoted
out of context. The broader context for the court's statement is:10
EPA also refutes Caterpillar's suggestion that the purpose of the NCPs was to allow a
manufacturer to weigh the costs of compliance against the costs of paying NCPs. If that
were the case, EPA points out, engine manufacturers would be able to calibrate the
intensity of their compliance efforts to the NCP for each new standard, allowing them to
opt for noncompliance when compliance becomes more expensive than the NCP. This
kind of second-guessing, however, was clearly not Congress' intent in providing for
NCPs:
The principal purpose of the Clean Air Act is to protect public health. The mere
payment of an economic penalty required by the delayed compliance penalty
provision should not be insulation against achieving requirements related to
protection of public health.... [T]he delayed compliance penalty ... is not intended
to provide an opportunity for continued non-compliance.
Instead, NCPs were intended to give a manufacturer that has made every effort to
comply, but has been unable to achieve compliance, a chance to continue to participate
in the market (emphasis added). Thus, NCPs serve their purpose even if promulgated
after a company has made its engine design decisions, since those decisions should be
based on whether compliance can be achieved, not on whether compliance is less
expensive than paying NCPs.
Q
See "Caterpillar Announces Plans to Phase Out Bridge Engines," Transport Topics, Sept. 9, 2003.
9 Final Technical Support Document: Nonconformance Penalties for 2004 Highway Heavy Duty Diesel Engines,
EPA420-R-02-021 August 2002, at 11-12 ("Engine manufacturers generally agree with us that cooled EGR is one of
the principal technologies capable of achieving the 2004 emission standards. In the past several months, a number
of engine manufacturers have announced they are pursuing cooled EGR technology as their principle means of
complying with the 2004 standards. In addition, at least one engine manufacturer [identified as Caterpillar] has
announced they are pursuing an alternative technology for complying with the 2004 HDDE standards which does
not include the use of cooled EGR.")
10 United States v. Caterpillar, Inc., 227 F. Supp. 2d 73, 88 (D.D.C. 2002).
18
-------
When considered in its broader context, it becomes clear that this statement had nothing to do
with evaluating in retrospect whether a manufacturer had made every effort to comply, but rather
addresses the possibility of a manufacturer choosing which technology to pursue based on the
availability of NCPs. This is not the case for Navistar. Clearly, Navistar chose to pursue a non-
SCR solution long before NCPs were available. Navistar has made considerable efforts to meet
the standard, and has not simply weighed the costs of compliance with the cost of NCPs.
Navistar's choice of innovative in-cylinder control has led to a delay in its full compliance
should not be confused with a manufacturer that has failed to meet standards based on a
weighing of costs.
With respect to the second reason, we agree it is reasonable to conclude that Navistar can
meet the 0.20 g/hp-hr NOx standard by adding SCR to its engines. In fact, Navistar has recently
announced that it will do just that. However, this does not mean that it is not currently a
technological laggard. Even before Navistar's announcement, our technological laggard
determination considered Navistar's ability to add SCR to its engines, but found this to be an
unrealistic near-term strategy (i.e., for 2012) due to the amount of time it would take to redesign
engines and vehicles for an alternate compliance path that would use SCR. The same would be
true for other advanced technologies. These limitations are technological rather than economic
in nature, and no amount of money could be spent by Navistar to bring its engines into
compliance during the current model year. We continue to believe that there is no viable
technological path available to Navistar that would allow it to produce engines that fully comply
with 0.20 g/hp-hr NOx emission standard for model year 2012, or to use compliant engines made
by another engine manufacturer in its vehicles.
With respect to the third reason, commenters are correct that prior to the end of the
comment period, Navistar submitted an application for certification of engines to the 0.20 g/hp-
hr standard without using credits. However, after EPA's review of the initial application, there
were substantial issues regarding the extent to which Navistar's engines truly meet the standards
and other regulations, and that request was withdrawn. Subsequently, Navistar submitted a new
application for certification for one of its 13 liter engines. This application would have covered a
single engine configuration and would not cover Navistar's other 13 liter engines or any of its 11
liter or 15 liter engines. Navistar noted that even if its application were approved, it would take
more than a year to incorporate appropriate revisions for all of its engines. Our initial review of
this revised application did not convince us that Navistar's newer engine design met the
standards and other regulations. Navistar has since withdrawn this application.
When considered in totality, we continue to believe that there is a significant likelihood
that Navistar is a technological laggard with respect to a significant number of engines. We also
note that if we are incorrect and Navistar is able to comply with the standard for all its engines in
the very near future, the final NCP is high enough that it would discourage Navistar from using
NCPs in those circumstances.
With respect to the comment from Ms. Sinisi, we note that the Clean Air Act does not
require more than one manufacturer to be at risk of noncompliance in order for EPA to set NCPs.
19
-------
1.2.4. Other Criteria
Summary of Comments
Daimler EPA-HQ-OAR-2011-1000-0028, 0043-0045 & 0049
Daimler commented that NCPs are only appropriate where EPA finds that the regulatory
lead time it has provided is insufficient to enable manufacturers to develop the technology
necessary to meet the standards.
Response
We disagree with this comment. EPA would not set standards such that the regulatory
lead time is not sufficient to meet the standard - see section 202(a)(3). In any case, while it is
true that allowing certification through NCPs might be appropriate where the regulatory lead
time has proved insufficient for manufacturers to meet the standards, NCPs are generally
intended for circumstances like the current circumstances, where only a portion of the industry
has met the standards. Navistar needs NCPs not because we did not provide sufficient lead time,
but rather because it chose to pursue an emission control technology path that has not yet
achieved 0.20 g/hp-hrNOx.
1.3. Interim Final Rule
On January 31, 2012 EPA simultaneously published an Interim Final Rule establishing
interim NCPs for heavy heavy-duty engines and a parallel Notice of Proposed Rulemaking
(NPRM). Several engine manufacturers petitioned EPA to rescind that Interim Final Rule.
These petitions and EPA's responses to them have been placed into the Docket for this rule.
Summary of Comments
Cummins EPA-HQ-OAR-2011-1000-0015, 0025 & 0048
Daimler EPA-HQ-OAR-2011-1000-0028, 0043-0045 & 0049
Mack EPA-HQ-OAR-2011-1000-0024, 0046, 0047 & 0051
Mack, Daimler and Cummins reiterated in their comments objections that they made to
the Interim Final Rule in their petitions. Navistar commented in support of the Interim Final
Rule.
Response
EPA has placed into the docket for this rule responses to the issues raised by the
commenters in their petitions. The procedural issues raised in the petitions with regard to the
interim final rule are not relevant to this final rule that was issued after notice and comments.
The substantive comments regarding the provisions of the Interim Final Rule are moot as the
final rule has superseded that rule, and are generally answered in responses to comments
regarding the notice of proposed rulemaking.
20
-------
2. Penalty Level and Cost Analysis
This chapter responds to comments addressing our cost analysis. Note that additional
discussion of the basis of our cost analysis can be found in the Final Technical Support
Document for this rulemaking.
2.1. Appropriateness of Penalty Level
Section 206(g)(3)(B) of the CAA instructs EPA to set NCPs at a level that "remove[s]
any competitive disadvantage to manufacturers whose engines or vehicles achieve the required
degree of emission reduction." This section discusses general comments about the level of the
penalty. Later sections in this chapter discuss more specific comments related to the penalty
level, such as those related to the methodology or data used to support our analysis.
Summary of Comments
CATF EPA-HQ-OAR-2011-1000-0021-0022
Cummins EPA-HQ-OAR-2011-1000-0015, 0025 & 0048
Daimler EPA-HQ-OAR-2011-1000-0028, 0043-0045 & 0049
Ford EPA-HQ-OAR-2011-1000-0029
ICCT EPA-HQ-OAR-2011-1000-0030
J. V. Sinisi EPA-HQ-OAR-2011-1000-0020
Mack EPA-HQ-OAR-2011-1000-0024, 0046, 0047 & 0051
MECA EPA-HQ-OAR-2011-1000-0017 & 0040
Navistar EPA-HQ-OAR-2011-1000-0027, 0031-0039, 0042 & 0050
NESC AUM EP A-HQ-O AR-2011-1000-0018-0019
NJDEP EPA-HQ-OAR-2011-1000-0016
PACCAR EP A-HQ-O AR-2011-1000-0041
Navistar
Navistar commented that EPA set the NCP penalties too high in the proposal and Interim
Final Rule. Navistar believes that EPA overestimated DEF costs and underestimated investment.
It commissioned Compass Lexecon ("Lexecon") to analyze the economics of EPA's Proposed
NCP Rule and Interim Final NCP Rule. Navistar reports that Lexecon concludes that:
• EPA's method is an "economically appropriate approach" in balancing the CAA's twin
goals of not forcing a manufacturer out of the market and of not placing a complying
manufacturer at a competitive disadvantage.
• If Navistar were eliminated from or significantly restricted in the market, there would
likely be a substantial competitive effect and consumer harm.
Navistar noted that the purpose of a NCP "is not to penalize nonconforming manufacturers" but
to ensure that there is no competitive advantage received from the failure to comply.
21
-------
Ford
Ford commented that if it is cheaper and easier simply to pay the NCPs rather than install
the aftertreatment equipment necessary to meet the standards, many if not all manufacturers will
elect to pay the NCPs. Such an outcome would be contrary to the CAA goals of improving air
quality. Ford also commented that the proposed NCP levels for medium heavy duty diesel
engines (MHDDEs) are substantially lower than necessary to achieve the goal of removing the
competitive advantages of certifying to a less stringent Oxides of Nitrogen (NOx) standard. Ford
stated that the COC 50 value of $462 and COC 90 value of $682 are as much as an order of
magnitude less than would be suggested by comparisons of retail or wholesale prices of fully
compliant engines to engines that will require NCPs.
NJDEP and CATF
NJDEP and CATF commented that the proposed NCP values were too low to meet the
applicable NCP statutory and regulatory provisions. NJDEP stated that "the NCP does not
appear to be large enough to incentivize the one non-conforming manufacturer to spend the
required money and resources to develop a compliant emission control system quickly." It also
stated that "the amount of the NCP should act as a deterrent to prevent manufacturers from
failing to invest the necessary resources to develop a compliant system" and should be greater
than the costs of a compliant system. CATF submitted a memorandum that evaluates potential
NCP values that would potentially derive from a "high engine-out" nonconformance path, using
best available recent projections of the future relationship between DEF and diesel fuel prices.
That analysis indicates that a year 1 NCP value in the range of $4,000-$5,000 might be
warranted to preclude backsliding by current conforming engine manufacturers.
J. Sinisi
Ms. Sinisi commented that noncompliance penalties really only offer engine
manufacturers an incentive to remain noncompliant with applicable emissions standards. She
stated that though noncompliance penalties are designed to increase over time in order to
encourage more efficient compliance; in reality it may actually be cheaper for some engine
manufacturers to produce noncompliant engines for a very long time. She further stated that
certain manufacturers may only bring their engines up to code on the very day that it becomes
too expensive to pay the increased noncompliance penalty. She stated that engine manufacturers
will continue to pump dangerous nitrogen oxides into the environment because they have an
incentive to do so in the form of a noncompliance penalty. Ms. Sinisi commented that the
penalty may only buy them time to increase their profits until they are either finally forced to
comply or until the next noncompliance penalty is instituted.
MECA
MECA stated that a powertrain manufacturer may decide to take short term profits rather
than investing millions of dollars and years of resources into R&D, manufacturing upgrades and
other new product commercialization costs and in the end fall back on NCPs to buy their way
into compliance. Incentivizing such an approach is a significant concern to MECA members and
threatens the establishment of a fair market, based on a level playing field. MECA urges the
22
-------
agency to manage the NCP compliance option so as to not threaten the future development of
advanced technologies that actually deliver the emission reductions used to justify the original
rule.
Cummins
Cummins commented that EPA's proposed NCPs would create a situation where paying
NCPs to produce noncompliant engines is a lower cost option compared to compliance, stating:
When Congress authorized NCPs, it foresaw this possibility and spoke directly to it:
"The Committee does not intend to encourage noncompliance with the revised standards.
For example, if a manufacturer opts to pay the penalty and to design or tune the vehicle
or engine to higher emission levels, the nonconformance penalty would probably be
inadequate and should be revised." House Rep. No. 95-294 at 276 (95th Cong. 2d Sess.
1977), 2 U.S.C.C.A.N. 1355 (1977). But that is just what EPA has done here.
Cummins also commented that the too-low NCPs create a much greater issue with regard to
compliance among SCR manufacturers than between EGR and SCR manufacturers. Cummins
continued to say that since NCPs are available to all manufacturers once enacted, and since the
cost of the proposed NCPs are much lower than the economic value that can be created by
paying them and optimizing engines at higher, noncompliant NOx levels, SCR manufacturers
that are otherwise capable of producing compliant engines will be driven by competitive market
forces to pay NCPs as a license to produce higher emitting engines. Cummins also commented
that a correct analysis would recognize the first cost and operating cost benefits of operating at
higher NOx levels up to 0.50 g/hp-hr and would incorporate a more appropriate DEF/diesel price
ratio which would result in a much higher NCP level of $8,100 for a heavy heavy-duty diesel
engine and $2,600 for a medium heavy-duty diesel engine.
Daimler
Daimler stated that the penalty levels set forth in the Interim Final Rule and NPRM do
not meet the paramount statutory obligation to remove any competitive disadvantage to
manufacturers whose engines or vehicles achieve the required degree of emissions reduction."
Daimler also commented that EPA focuses in on an emissions reduction that is 1/8 the amount
that has been achieved by manufacturers to meet the current standard, thereby ensuring that any
calculation under this new method will always be grossly inadequate. Daimler stated that EPA
explained in response to comments on its most recent prior NCP rulemaking: "The statute
requires we remove any disadvantage which the complying manufacturer may encounter. . . .
[T]he statute does not require that we establish an NCP which removes any disadvantage to the
non-complying manufacturer."
PACCAR
PACCAR stated that the proposed NCP level is so low that it creates an additional
competitive disadvantage for manufacturers that have invested considerable time and resources
in developing emission control technology that complies with the 2010 NOx standard.
PACCAR commented that in order to eliminate the competitive advantage Navistar gains under
23
-------
the current NCP, EPA should add the value of Navistar's fuel economy advantage to the SCR
hardware, R&D and warranty component costs, and increase the NCP for 2012 to a minimum of
$14,378.
Mack
Mack stated that EPA's NCP for the 2010 standard is a fraction of its previous NCPs.
Mack commented that the maximum penalty, or penalty associated with meeting the upper
limit of 6.0 g/bhp-hr established by the 1991 NCP, was $2,250. Adjusted for inflation to 2012
dollars, the penalty is $3,744. In addition, it commented that this penalty was established for
an upper limit that was only 17 percent higher than the applicable 5.0 g/bhp-hr standard.
Mack stated that the maximum penalty associated with meeting the upper limit of 5.0 g/bhp-
hr established by the 1998 NCP was $2,540. Adjusted for inflation to 2012 dollars, the
penalty is $3,532. Mack further stated that the upper limit for this rulemaking, meanwhile,
was set at about 20 percent higher than the applicable 4.0 g/bhp-hr standard. It said for 2004,
the maximum penalty associated with meeting the upper limit of 6.0 g/bhp-hr was $12,210, or
$14,651 in 2012 dollars. The upper limit for that rulemaking was 2.4 times the standard.
Further, by comparison, the maximum penalty established for meeting the upper limit of 0.50
g/bhp-hr in the NCP under the Proposed NCP Rule is $1,919, which is lower than all
previously established NCPs forNOX emissions from heavy heavy-duty diesel engines, even
if not adjusted for inflation. Mack concluded that this is despite the fact that the rule was the
most expensive emissions mandate the industry had ever faced.
Mack concluded that by setting an extraordinarily low NCP for MY 2010, EPA not
only would remove any incentive by Navistar to develop a compliant engine, it also would
create new incentives for conforming manufacturers to develop engines that can take
advantage of a 0.50 g/bhp-hr upper limit through payment of a nominal penalty.
Response
NCPs must conform to the statute and the regulations. With respect to the statutory
requirements, we note that while the purpose of adopting NCPs is to allow a noncompliant
manufacturer to continue selling its engines, the Clean Air Act directs EPA to set the NCPs at a
level that will "remove any competitive disadvantage" to complying manufacturers. We believe
that the statute effectively requires us to set the penalties at a level that we reasonably expect to
protect the complying manufacturers, but not so high that it effectively forces any noncomplying
manufacturers from the market, (unless that is the only way to protect the complying
manufacturers). In this context we agree with commenters that state EPA must consider
competitive disadvantages for complying manufacturers relative to both noncomplying SCR and
noncomplying non-SCR engines. We agree with commenters that setting NCPs too low could
cause manufacturers who can now comply with the standard to instead make a decision to
comply using NCPs, which would lead to more NOx emissions. We believe that the NCPs
promulgated in this final rule, which are substantially larger than those proposed, are sufficient to
protect complying manufacturers from competitive disadvantage, but not so high as to create a
substantial competitive advantage for complying manufacturers.
24
-------
While commenters provided arguments why Navistar could have a competitive
advantage over compliant engines, none provided any evidence of an actual competitive
advantage. Commenters did not dispute EPA's statement in the Interim and Proposed Technical
Support Document that Navistar has not gained any market share since the 2010 standard has
gone into effect, during a period when Navistar was not paying any penalties at all. Some
commenters claim that Navistar has saved money by not developing SCR, but Navistar claims
that it has actually spent more on R&D than its competitors. Thus, while we agree that
producing engines with NOx emissions at 0.50 g/hp-hr provides Navistar some competitive
advantage, we do not believe it is as high as some commenters claim. Moreover, we agree with
Cummins that an optimized SCR engine with NOx emissions at 0.50 g/hp-hr would have an even
greater competitive advantage than Navistar's engine.
It is also worth noting that Navistar has revealed in its quarterly reports that it has had
higher than expected warranty claims for its 2010 and 2011 engines, which led to the company
posting a net loss so far for 2012.n We believe those commenters focused on costs of SCR
hardware overestimate Navistar's competitive advantage because they did not consider the high
warranty costs for Navistar's high-EGR engines.
EPA has also set regulatory requirements for penalty levels. Most significantly, the
regulations state that we must set penalties based on total incremental costs of compliance
relative to engines at the upper limit. These requirements apply in addition to the statutory
requirements. Unlike the statutory requirements, the regulatory requirements may be revised at
any time through rulemaking. Thus, while EPA follows the regulatory requirements, this does
not allow us to ignore the statutory requirements. In fact, should we find that conforming to the
regulatory requirements would not conform to the statutory requirements, we would revise the
regulatory requirements. Having said this, as noted elsewhere, we believe that the final NCPs
being established conform to the regulatory requirements.
As is described in the Final Technical Support Document, the final NCP levels meet both
the statutory and regulatory criteria. They are also consistent with the legislative history cited by
Cummins, which stated "if a manufacturer opts to pay the penalty and to design or tune the
vehicle or engine to higher emission levels, the nonconformance penalty would probably be
inadequate." In summary, we have found that the final NCP parameters are set at a level that
will:
1. Remove any competitive advantage Navistar may have otherwise had by selling
non-SCR engines.
2. Allow Navistar to remain in the market.
3. Eliminate the incentive for SCR engine manufacturers to reoptimize their engines
for 0.50 g/hp-hr NOx.
While many commenters addressed this first conclusion, none provided any evidence that the
actual competitive advantage for Navistar is so large that it is not eliminated by our penalty, and
11 "Navistar Announces Management Reorganization, Reports Second Quarter Results", Navistar Press Release,
June?, 2012.
25
-------
none of the comments even suggest that the penalty level would not allow Navistar to remain in
the market. However, several commenters provided information to support their belief that the
proposed penalty was not large enough to eliminate the incentive for SCR engine manufacturers
to reoptimize their engines for 0.50 g/hp-hr NOx. As is described in later sections, we have
made changes to our analysis in response to these comments so that our final penalty now does
eliminate this incentive.
Neither the statute nor regulations state (or even imply) that NCPs should be evaluated
based on prior NCPs. Thus, comments comparing the proposed NCPs to prior NCPs are not
relevant without specific reference to differences in methodology and costs. Nevertheless, we
have compared them to prior NCPs to show that, while the commenters are correct that the
proposed maximum penalty was lower than those for recent NCPs, the reason for this is that the
difference between the standard and upper limit is so much lower than for these recent NCP
rules. As shown in the following table, when considered on a dollar per g/hp-hr basis, the
proposed NCPs are actually higher than those of these recent NCP rules.
2-1 Comparison of Proposed NCPs to Prior NCPs for Heavy Heavy-Duty Engines.
Comparison of Proposed NCPs to Most Recent NCPs for Heavy Heavy-Duty Engines
Standard
4.0 g/hp-hr 1998 NOx
Standard
2.4 g/hp-hr 2004
NOx+NMHC
Standard
0.20 g/hp-hr 20 10
NOx Standard
Upper Limit
5.0 g/hp-hr
6.0 g/hp-hr
0.50 g/hp-hr
COC90
(in 20 11 Dollars)
$3,855
$15,508
$1,919
Dollars per g/hp-hr
$3,855
$4,308
$6,397
Daimler's comment about our focus on the costs of going from 0.50 g/hp-hr to 0.20
g/hp-hr, rather than the previous standard (about 2.0-2.3 g/hp-hr) to 0.20 g/hp-hr, disregards the
fact that, pursuant to section 206(g)(2), the upper limit should be set to the level we find to be
practicable, which in this case is 0.50 g/hp-hr, which every manufacturer has met. Moreover, it
is not true that looking at the smaller emissions difference will "ensure" lower NCPs. Whether
or not we based our analysis on the smaller or larger difference, NCPs would still only be
permitted to the upper limit (0.50 g/hp-hr). There is no reason to believe that an NCP for an
engine emitted at 0.50 g/hp-hr will necessarily be higher under the existing regulatory structure
because it is based on the calculation of going from the previous standard to 0.20 g/hp-hr, rather
than going from 0.50 g/hp-hr to 0.20 g/hp. Indeed, in our analysis of alternatives in the Final
26
-------
Technical Support Document, we reviewed a possible NCP based on going from a baseline
engine emitting NOx at 1.2 g/hp-hr (the effective interim standard from model year 2007 to
2009) to 0.20 g/hp-hr, and we found that the NCPs at 0.50 g/hp-hr were actually lower in that
scenario than in the scenario we used in the final rule. See Sections 2.4 and 2.5 for additional
discussion of this comment.
With respect to Ford's comments regarding the appropriateness of the medium heavy-
duty NCP values, we have decided to seek additional comment before finalizing medium heavy-
duty engine NCPs. The Final Technical Support Document includes an analysis of three
alternative methods that could be used to calculate medium heavy-duty NCP values.
We did receive other relevant and more specific comments on the level of the penalty.
These comments are addressed in separate sections of this document, as specified in the
following table:
2-2 Sections Addressing Details of NCP Level Derivation
Comments related to ...
Consideration of cost information from engine manufacturers
Basing compliance costs on engines at the upper limit
Baseline engines
Hardware costs
Operating costs
Warranty and repair costs
Research and development costs
Other costs
Are addressed in Section . . .
2.3
2.5
2.6
2.7
2.8
2.9.1
2.9.2
2.9.3
While we are responding separately to comments on specific aspects of our cost analysis,
it is important to note that the costs must ultimately be evaluated as a whole. This is true in the
context of the regulatory definition of COC90 as well as in the context of whether the resulting
penalties meet the statutory requirements. The COCgo cost is defined as the 90th percentile cost
of compliance, although it is often thought of as a worst case cost of compliance. This leads
some commenters to argue that each individual cost should be set at the highest possible value.
However, there is uncertainty associated with each cost component and we have attempted to
address this uncertainty so that the resulting total COCgo cost represents the 90th percentile cost
of compliance and removes the competitive disadvantage for complying manufacturers.
Finally, we note that several commenters recommended specific values for COCgo (in
some cases stated as the recommended penalty). These recommendations, which each differ
from our final COCgo value, are summarized in Section 2.10. In those cases where it is possible
to do so, we identify the primary reasons why the recommended value differs from our estimate.
27
-------
2.2. Differences from Cost Methodologies Used in Prior NCP
Rulemakings
This section addresses comments on the general topic of the extent to which EPA must
follow the same cost methodologies used in prior NCP rules. Similar comments addressing
specific differences between this rule and prior NCP rules are addressed elsewhere in this
document.
Summary of Comments
Daimler EPA-HQ-OAR-2011-1000-0028, 0043-0045 & 0049
Mack EPA-HQ-OAR-2011-1000-0024, 0046, 0047 & 0051
Navistar EPA-HQ-OAR-2011-1000-0027, 0031-0039, 0042 & 0050
Daimler and Mack argued against EPA's cost methodology because it differs from
methodologies used in prior NCP rules.
Daimler stated that EPA acknowledged that its methodology "differs from that used in all
past NCP rules", but added that "EPA never offers a rational explanation for the change and has
failed to recognize the fact that this methodology inevitably leads to a gross underestimation of
the true cost of compliance." (Daimler's other comments related to specific aspects of prior
methodologies, as well as Mack's, are discussed in later sections.)
Navistar stated that EPA's overall analysis and methodology is a rational and reasonable
path for calculating the cost parameters and NCP penalty levels in the circumstances presented.
It said that EPA applies the exact same methodology and NCP formula that the agency has
applied for decades. Navistar continued to state that although the exact cost inputs for each NCP
are dependent upon the specific circumstances that apply at the time, as with every other
previous NCP, the Agency has based the NCP on incremental compliance costs, including
engine and vehicle manufacturing costs (such as variable costs including hardware and assembly
and fixed costs including tooling, research and development, and warranty costs) in addition to
operating costs (such as fuel and DEF costs). Navistar also stated that the baseline engine
selected by EPA is reasonable for calculating NCPs.
Response
This section addresses the general issue of whether EPA is obligated to follow past
practices used in other NCP rules. See Sections 2.4, 2.6, and 2.8.1, for additional discussion of
ways in which this current rule differs from prior NCP rules.
As noted earlier, the Clean Air Act directs EPA to set the NCPs at the level necessary to
"remove any competitive disadvantage" to complying manufacturers. However, since it is
generally not possible to precisely quantify the competitive disadvantage to complying
manufacturers, EPA's original "generic rule" provides for the NCPs to be calculated from total
28
-------
incremental costs of compliance. In that rule, EPA noted that to evaluate the total incremental
10
costs of compliance, it would "assess both manufacturers' and users' cost impacts." The logic
of this approach was that penalties based on total incremental costs of compliance would
generally be the best surrogate for penalties based directly on competitive disadvantages.
Nevertheless, this has never meant that we would knowingly ignore other factors related to
competitive advantages.
The commenters' suggestion that EPA not deviate from prior precedents for calculating
costs implies that EPA has always used the same methodology. However this is not true. EPA
explained this in the Technical Support Document for the NPRM when we said:
In each of our six previous NCP rulemakings, we estimated costs using a methodology
appropriate for the specific circumstances that applied at the time. None were approached in
exactly the same way. In each case we considered key factors such as differences in
calibration, hardware, and operating costs, but there have been some NCP calculations where
other potential individual cost or cost saving elements have been included or excluded for
various reasons.
Even if it were true that EPA had used the same methodology for all prior NCP rules, this would
not obligate us to continue using the same methodology in perpetuity. By arguing that EPA may
never change from approaches taken in prior NCP rules, the commenters are essentially saying
that we should consider only certain compliance costs, and may never consider new costs or
other competitive impacts that we have not considered previously. However, this would clearly
not be consistent with the plain meaning of the statutory language.
With respect to Daimler's comment, we believe it was referring to following statement of
page 7740 of the proposal:
This approach differs slightly from that used in previous NCP rules, where EPA based
the NCPs directly on an average of actual compliance costs for all manufacturers.
If so, then Daimler's statement that this represents an acknowledgement "that the methodology it
adopted in the NPRM differs from that used in all past NCP rules" suggests that it is missing the
intent of this sentence. This statement was intended to address the narrow issue of how to
calculate costs when the upper limit is set at a level different than the previous standard. As we
discussed in the proposal, the upper limit in this rule is lower than the previous standard, and
therefore we had to determine the best way to approximate the cost of going from this upper
limit to the new standard, even though no manufacturer had actually done this in reality. See
Section 2.5 for additional discussion.
12 50 FR 35382, August 30, 1985.
29
-------
2.3. Consideration of Manufacturer Costs
Summary of Comments
Daimler EPA-HQ-OAR-2011-1000-0028, 0043-0045 & 0049
Mack EPA-HQ-OAR-2011-1000-0024, 0046, 0047 & 0051
Mack stated that "the statutory mandate to remove competitive harm effectively requires
EPA to consult with and obtain actual cost data from complying manufacturers." It also stated
that "every manufacturer, save one, has information that can demonstrate, accurately and
unequivocally, what it cost them to comply with a 0.20 g/bhp-hr NOX standard." It stated that
EPA solicited no input on actual compliance costs and "asked manufacturers to speculate what
technology and approaches they would undertake to achieve compliance with a 0.2 g/bhp-hr
engine when starting with a 0.5 g/bhp-hr engine." It stated further that EPA dismissed this input
citing our concern that manufacturers were aware that their estimated costs could be used to
determine the amount of the NCP paid by a competitor, but that "there would have been no
opportunity for such alleged strategic estimates" if EPA had it used actual compliance cost
information. Mack cited the following EPA statements from the 2002 Response to Comments
for the Phase VI NCP rulemaking:
Our request for data from manufacturers was not for them to speculate on costs for a
development program which they did not undertake. Rather, our request from
manufacturers was for them to provide us with data that reflected the actual development
programs they were performing (i.e., making a model year 2001 engine comply with the
2004 emission standards).
Again, EPA's request prior to the proposal was for the cost estimate for the development
programs that the manufacturer was actually undertaking keyed to costs for the 2004
model year, not for a theoretical development program, or a prior model year.
Daimler commented that EPA must gather cost data from manufacturers in order to
determine the costs of compliance with the emission standard and use these data to derive
various factors and inputs necessary to calculate the appropriate penalty amount, so that any
competitive disadvantage to complying manufacturers is removed. Daimler continued to note
that in the past, EPA has requested information in writing from engine manufacturers, proposed
the values for a new NCP for a given emission standard with proper public notice, accepted
comments from the manufacturers and other members of the public, revised its proposal
accordingly, and then issued its final rule. Daimler also said that "EPA contacted various engine
manufacturers by telephone in late 2011 requesting preliminary information for a proposed NCP
rulemaking but did not take written submissions and refused requests for additional meetings
with individual manufacturers."
Navistar commented that it previously provided cost data to EPA for meeting 0.20 g NOx
using only Advanced EGR, but noted that the data is protected as confidential business
information. Navistar's estimates that the projected costs provided by Navistar (using an
alternative "baseline") would result in a significantly lower NCP level than EPA proposed.
30
-------
Response
Neither the Clean Air Act nor our regulations require us to give any special consideration
to manufacturers' estimates of compliance costs. In fact, for some prior NCP rules EPA did not
explicitly solicit any cost information from manufacturers other than the general solicitation of
public comments on a proposal. In the end, the compliance costs must be those that EPA
determines to most appropriately conform to the intent of the Clean Air Act. As EPA noted in
the generic Phase I NCP rule, NCP costs are to be based on "the best cost and emission
performance data available to EPA during the specific NCP rulemaking." The basis for our final
cost estimates is described in detail in the Final Technical Support Document.
Moreover, we did seek input from heavy heavy-duty engine manufacturers before the
proposal and did consider that input. However, as we noted in the Interim Technical Support
Document, each manufacturer was aware that its estimated costs could be used to determine the
amount of the NCP paid by a competitor, or by itself. We stated that we were "concerned about
this because we cannot independently verify the validity of the manufacturers' costs." We relied
on our own experience in determining compliance costs from all information available to us,
including information provided by manufacturers. We believe that, especially in this context,
this is appropriate because we do not have the same incentive to skew our results as competing
manufacturers do. For example, as we note in the Final Technical Support Document,
information about operating costs provided by both SCR and non-SCR manufacturers is not
necessarily consistent with information provided to customers in marketing materials and press
statements.
Our cost estimates are neither arbitrary nor speculative, but are based on our unbiased
view of the collection of data provided by manufacturers (both publicly and confidentially) as
well as other data collected by EPA. In particular, we note that our methodology for
aftertreatment costs is an approach that we have used in several previous notice-and-comment
rulemakings. This and other examples of independent sources of cost informations are
summarized in Table 2-3. (See the Final Technical Support Document for additional discussion
of the basis of our cost estimates.)
31
-------
2-3 Examples of Independent Data Sources.
Cost Components
Aftertreatment Hardware Cost
Vehicle Miles Travelled
Fuel Price
DBF Price
DBF Consumption Rates
Data Sources
Independent EPA methodology used in prior
rulemakings, which has been made available
for public comment several times.
EPA Motor Vehicle Emissions Simulator
DOE Energy Information Administration's
Annual Energy Outlook 2012
Integer Research
EPA test data
With respect to Mack's concern that EPA had requested speculative data, we note that we
did specifically request cost data relative to engine emitting at the upper limit of 0.50 g/hp-hr,
consistent with the regulatory provision specifying that compliance costs are to be estimated
relative to the upper limit (see Section 2.5 for additional discussion). Since Mack has not
manufactured engines at this upper limit, this necessarily required some speculation on its part.
While it is true that Mack would not have needed to speculate about costs relative to its 2009
engines, these costs do not supply the critical information needed to determine COC90 because
the 2009 engines had NOx emissions well above the upper limit (Mack did not object to setting
the upper limit at 0.50 g/hp-hr.) The prior EPA statements cited by Mack are taken out of
context. Mack suggests that these statements are evidence that EPA has previously argued that
manufacturers should not speculate on costs for NCPs. However, these statements were made in
response to comments about what EPA actually did when requesting preproposal information for
that Phase VI rule, rather than what we should do. Thus, these statements are not relevant to this
rule.
With respect to Daimler's comments related to the process of soliciting cost information,
we note that this seems to be a misunderstanding caused by the very short timeline of the
proposal and Interim Final Rule. We did not solicit written comments because we wanted to
gather the information quickly. While we did inform manufacturers that we needed to end the
discussions so that we could get the rulemaking out as soon as possible, we did not actually
refuse requests for meetings.
It is import to note that we also received cost information from manufacturers during the
public comment period and met with those manufacturers who wanted to discuss their cost
information further. We considered this new information along with the preproposal information
32
-------
in our final analysis of compliance costs. In particular, we found the cost information provided
by Cummins to be especially helpful because it provided detailed information about how an SCR
engine could be optimized for 0.50 g/hp-hr NOx. While we did not agree with Cummins' fuel
and DEF prices (see Section 2.8.3), the rest of our cost analysis is similar to the analysis it
provided. We should also note that information provided by manufacturers other than Cummins
(those manufacturers who addressed costs relative to other baseline engines) was considered in
our analysis of alternative scenarios as described in Appendix B of the Final Technical Support
Document.
Finally, there are two primary reasons why we are not basing our NCPs on the costs
estimates provided by Navistar. First, just as with its competitors' costs, we are concerned
because we cannot verify Navistar's costs. Second, even if we could verify these costs, they
would clearly not represent the COCgo costs, which are supposed to represent the costs for the
90th percentile manufacturer. While we cannot divulge Navistar's exact estimate, we can say
that it much lower than our final estimate of COC90. Because they are lower than our estimated
COCgo, we believe they would not be large enough to remove the competitive advantage of an
optimized 0.50 g/hp-hr SCR engine.
2.4. Upper Limit
The upper limit used in the NCP derivation is the emission level established by regulation
above which NCPs are not available, and a heavy-duty engine cannot be certified or introduced
into commerce if its emissions are above the upper limit. CAA section 206(g)(2) refers to the
upper limit as a level above the emission standard, set by regulation, that corresponds to an
emission level EPA determines to be "practicable." The upper limit is an important aspect of the
NCP regulations not only because it establishes an emission level above which no engine can be
certified, but it is also a critical component of the cost analysis used to develop the NCP factors.
The regulations specify that the relevant NCP costs for determining the COCso and the COCgo
factors are the cost difference between an engine emitting at the upper limit and one that meets
the new standards (see 40 CFR 86.1113-87). In the proposal, we set the upper limit at 0.50 g/hp-
hr based on the existing family emissions limit (FEL) cap and the fact that all manufacturers are
producing products today at NOx levels at or below 0.50 g/hp-hr.
Summary of Comments
CATF EPA-HQ-OAR-2011-1000-0021-0022
Daimler EPA-HQ-OAR-2011-1000-0028, 0043-0045 & 0049
EOF EPA-HQ-OAR-2011-1000-0026
ICCT EPA-HQ-OAR-2011-1000-0030
Several commenters noted their support for setting the upper limit at 0.50 g/hp-hr. No
commenters opposed setting the upper limit at 0.50 g/hp-hr.
The ICCT believes that the upper certification limit of 0.50 g/hp-hr outlined in the NPRM
is justified and necessary to ensure a minimum level of emissions performance from engines.
33
-------
EDF supports EPA's proposal to "revise the regulations in 40 C.F.R. § 86.1104-91 to clarify
that EPA may set (during rulemaking) the upper limit at a level below the previous standard if we
determine that the lower level is achievable by all engines." EDF also agrees that the upper limit
should be set at the tightest standard achievable by all manufacturers. EDF supports EPA's proposal
to set a limit of 0.50 g/hp-hr for this NCP rule, as all manufacturers are already meeting this standard.
CATF supports EPA's proposal to set the limit at 0.50 g/hp-hr NOx.
Daimler supports setting the upper limit at the maximum technologically feasible by the
laggard rather than the prior emissions standard, but stated that doing so "is no reason to
understate the compliant manufacturers' full cost of compliance." Daimler also stated that "if
EPA changes the definition of upper limit, the Agency needs to implement a corresponding
change to 40 C.F.R. § 86.1113-87(a)(4) to define COC50 and COC90 in relevant part as the total
incremental cost to comply with the standard relative to complying with the previous emission
standard." Daimler commented that failure to do this would render the NCP regulation and
formula incapable of measuring compliant manufacturers' full cost of moving from one
generation of emissions standards to the next.
Response
The Clean Air Act directs EPA to establish an upper limit for emissions that is equal to
the level EPA determines "to be practicable." The regulatory approach adopted under the prior
NCP rules sets the default upper limit at the prior emission standard when a prior emission
standard exists and that standard is changed and becomes more stringent. EPA concluded that
the upper limit should be set at a level reasonably achievable by all manufacturers with vehicles
in the relevant class. It should be within reach of all manufacturers of HDEs or HDVs that are
currently allowed so that they can continue to sell their engines and vehicles while finishing their
development of complying engines. A manufacturer of a previously certified engine or vehicle
should not be forced to immediately remove an HDE or HDV from the market when an emission
standard becomes more stringent. The prior emissions standard generally meets these goals,
because manufactures have already certified their vehicles to that standard. However, in the
prior NCP rule, we did set the upper limit for heavy heavy-duty engines at a level different than
the prior standard. We set the upper limit above the prior standard because of a Consent Decree
that allowed manufacturers to exceed the otherwise applicable standard.
In the past, EPA has rejected suggestions that the upper limit should be more stringent
than the prior emission standard because it would be very difficult to identify a limit that could
be met by all manufacturers. For this rule, however, all manufacturers are currently certifying all
of their engines at or below the 0.50 g/hp-hr FEL cap. Thus, since NCPs were not intended to
allow manufacturers to increase emissions, we are setting the upper limit at 0.50 g/hp-hr. This
will conform to the purpose of NCPs, which is to allow manufacturers to continue selling
engines they are producing, but not to allow backsliding. No commenter opposed using 0.50
g/hp-hr as the upper limit.
34
-------
Daimler supports setting the upper limit at 0.50 g/hp-hr, but stated that EPA must also
redefine COC50 and COC90 as the compliance costs relative to the previous emission standard.
We do not agree that this is an appropriate condition for setting the upper limit below the prior
standard. Daimler's comment is discussed in more detail in Section 2.5.
Although no commenters support setting an upper limit above 0.50 g/hp-hr, it is worth
considering a scenario in which the upper limit was set at 1.2 g/hp-hr (the level widely
considered to be the effective standard during the 2007-2009 phase-in period). As shown in
Figure 2-1, if we had set the upper limit at 1.2 g/hp-hr, the COCgo would have had to be set at
$12,583 to result in the same penalty at 0.50 g/hp-hr as the penalty we are finalizing - $3,775. In
fact, however, as described in Appendix B of the Final Technical Support Document, we
estimated the actual net 90th percentile costs of reducing emissions from 1.2 g/hp-hr to 0.20
g/hp-hr to be only $6,990. Thus, if we had set the upper limit at 1.2 g/hp-hr, we would have
likely set COCgo equal to $6,990 which would have resulted in a $2,097 penalty for engines at
0.50 g/hp-hr.
«ii A nnn
^ „-''
^
?U i i i i i i i i i
0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.90 1.00 1.10 1.20
NOx Emissions Level (g/hp-hr)
Figure 2-1 - Example of NCPs with Upper Limit at 1.2 g/hp-hr
35
-------
2.5. Basing Compliance Costs on the Upper Limit Engine
We estimated compliance costs relative to a baseline engine with NOx emissions at
the upper limit, as specified in §86. 1 1 13-87(a)(4) of the existing regulations, which states
that:
COCso = Estimate of the average total incremental cost to comply with the
standard relative to complying with the upper limit (emphasis added).
= Estimate of the 90th percentile total incremental cost to comply with the
standard relative to complying with the upper limit (emphasis added).
Summary of Comments
Daimler EPA-HQ-OAR-201 1-1000-0028, 0043-0045 & 0049
Mack EPA-HQ-OAR-201 1-1000-0024, 0046, 0047 & 0051
Mack stated that the fact that the "starting point of compliance - the phase-in standard
that manufacturers met between 2007 and 2009 - was different from EPA's upper limit" is not a
sufficient reason to not include the full cost of SCR hardware. It argued that the "NCP formula
is designed to reflect actual compliance costs, which are then adjusted based on the degree a
given compliance level exceeds the standard" and that "there is no reason to arbitrarily link the
engine used to develop an NCP to the upper limit set by EPA." Mack also commented that EPA
needs to consider the full engineering and development costs related to meeting the 0.20 g/hp-hr
engine.
Daimler supports setting the upper limit at 0.50 g/hp-hr and acknowledged that §86.1113-
87(a)(4) defines COC50 and COCgo in relevant part as the "total incremental cost to comply with
the standard relative to complying with the upper limit." However, it nevertheless argues that
compliance costs should be measured based on the emissions delta between the prior and current
standards because EPA has historically set the upper limit at the previous pollutant emission
standard.
Daimler also commented that the "cost for a manufacturer that has already developed
emissions control technology to meet the standard will always be significantly less than the cost
for a manufacturer who has not yet developed the necessary enabling technology." It further said
that "establishing a penalty amount based on a small fraction of the compliant manufacturer's
marginal cost of compliance does nothing to eliminate the competitive advantage that accrues to
a noncompliant manufacturer when that manufacturer is allowed to sell engines that do not meet
the emissions standards" and that "anything less would incentivize laggards to simply wait and
let others make the investment necessary to develop required emissions control technology, and
then piggy-back on that development by paying only the incremental cost of using the other
manufacturers' technology."
36
-------
Response
The regulations specify that costs of compliance are to be determined relative to the
upper limit. Thus, our linkage of the costs of compliance to the upper limit in our cost analysis is
not arbitrary, but rather it is consistent with our long-standing regulations and all prior NCP
rulemakings. For example, in the Phase VI rule, when we adopted an upper limit above the prior
standard, we estimated compliance costs relative to the upper limit rather than the prior standard.
Given the extent to which Mack argues in other areas that EPA must follow the regulations and
be consistent with prior NCP rulemakings, it is unclear how it finds this to be an insufficient
reason.
We fundamentally disagree with Daimler that compliance costs used for the penalty can
be disconnected from the upper limit. The regulations are based on analysis of costs related to
emission differences between the current standard and the upper limit (i.e., the only emission
levels allowed for NCP engines). We do not see how costs outside this range can determine the
appropriate NCP within that range. The determining factors are the costs that reflect the
competitive advantage for engines within this range. The flaw in Daimler's argument can be
readily seen by considering the counterfactual scenario in which the 2004 standard never existed
and the prior NOx standard would have been 4.0 g/hp-hr. Clearly the compliance costs for going
from 4.0 g/hp-hr to the current standard would have been greater than the cost to reach the
standard from 2.3 g/hp-hr (the effective NOx level of the 2004 NOx standard) or 1.2 g/hp-hr (the
effective intermediate standard during the 2007-2009 phase-in period). However, this would
have had no impact on the actual competitive advantage of current engines with NOx emissions
between 0.20 and 0.50 g/hp-hr. Since the competitive advantage of engines at the upper limit is
independent of the level of the prior standard, basing the penalty on the prior standard would be
arbitrary.
The NCP formula is intended to result in a penalty at any given compliance level equal to
the cost for the worst case manufacturer (i.e., the manufacturer with the highest compliance
costs) to reduce its emissions from that compliance level and meet the standard. For example,
the penalty for compliance levels near the standard is set based on the worst case marginal
compliance costs (MCgo) so that the NCP associated with being 0.10 g/hp-hr above the standard
is equal to the highest cost paid by any compliant manufacturer (more precisely, the 90th
percentile costs) to achieve that last 0.10 g/hp-hr. Since the regulations set the penalty equal to
COCgo for engines at the upper limit, and the Clean Air Act requires us to set the penalty at a
level that removes the competitive disadvantage for complying manufacturers, Daimler's
argument that COC90 should be based on engines at the prior standard could only be valid if it
were true that engines with emissions at the prior standard and engines with emissions at the
upper limit have equivalent competitive advantages over compliant engines. This is clearly not
true. No manufacturer has claimed that the cost of reducing emissions from 1.2 g/hp-hr to 0.50
g/hp-hr is zero, and the comments all indicate, either directly or through implication, that there
are considerable costs to do so. Daimler provides no basis to believe that the additional costs it
wants to be included have any relation to competitive advantage for manufacturers producing
engines at the upper limit.
Daimler's comment that we need to take into account the costs of going from the old
standard to the new standard in developing NCPs, even though the NCPs would not allow any
37
-------
manufacturer to actually emit at the old standard, is not inherently more rational than EPA
determining NCPs based on our view of what a 0.20 g/hp-hr engine would cost compared to an
optimized 0.50 g/hp-hr engine. We did not ignore the costs that manufacturers had to pay to
move from the previous standard to 0.50 g/hp-hr, whether they did so using SCR or advanced
EGR. (Mack and other manufacturers who use SCR seem never to acknowledge that Navistar
also experienced costs in moving from the old standard to 0.50 g/hp-hr, and that such costs
would also be relevant if we were to use such an approach.) Our analysis simply begins with the
understanding that whatever costs manufacturers incurred to reduce emissions, those emissions
will never be permitted to be above 0.50 g/hp-hr and it is not necessary to determine NCPs for
values above 0.50 g/hp-hr. Moreover, as noted above, we actually did analyze the costs for
manufacturers of reducing emissions from the previous effective standard (1.2 g/hp-hr), and
determined the NCPs for a 0.50 g/hp-hr engine based on that analysis, and the resulting NCPs
were actually lower than when EPA used the optimized 0.50 g/hp-hr engine as the baseline.
It is unclear how Daimler reconciles its comment that the "cost for a manufacturer that
has already developed emissions control technology to meet the standard will always be
significantly less than the cost for a manufacturer who has not yet developed the necessary
enabling technology" with its comment that a competitor could save money if it would "let
others make the investment necessary to develop required emissions control technology, and
then piggy-back on that development by paying only the incremental cost of using the other
manufacturers' technology." It seems that they cannot both be correct.
Mack's comment that EPA must consider full compliance costs relative to 2009 engines
appear to ignore the fact that, like all manufacturers, Navistar has expended capital to reduce
emissions from their engines from the previous standard to 0.50 g/hp-hr, or lower. It also
appears to ignore the fact that all new engine designs have some effect on operating costs.
Whatever those capital and operating costs may have been, all manufacturers have had to face
such costs to reduce their emissions. What is more relevant is the cost associated with reducing
engine emissions from the upper limit to the standard. With the upper limit set at 0.50 g/hp-hr,
the only purpose for evaluating the costs to get to 0.50 g/hp-hr would be to allow an indirect
calculation of the costs to get from 0.50 to 0.20.13 Moreover, as noted above, EPA did review an
alternative approach that included a baseline engine of 1.2 g/hp-hr, not using SCR, and found
that the NCP calculated for an engine emitting at 0.50 g/hp-hr using that baseline engine was
actually lower than the final NCP.
Finally, we note that while both Mack and Daimler provided information to support their
beliefs that the total costs associated with SCR are higher than the proposed COC90, as well as
arguments why the penalty should be based on these costs, neither one provided any evidence
that Navistar has a competitive advantage over their engines that is equivalent to these costs.
One of the reasons Navistar's advantage is less than the total SCR costs is that it invested
considerable money to achieve its 0.50 emissions level, which effectively offsets much of the
13 For example, if we knew the cost associated with reducing NOx emissions from 1.2 g/hp-hr to 0.50 g/hp-hr was
$7,000, and the cost associated with reducing NOx emissions from 1.2 g/hp-hr to 0.20 g/hp-hr was $10,000, we
could indirectly calculate the cost associated with reducing NOx emissions from 0.50 g/hp-hr to 0.20 g/hp-hr as the
difference between the two ($3,000).
38
-------
savings associated with not using SCR. Navistar has also recently acknowledged that its
warranty costs for its EGR engines are higher than expected. Neither Mack nor Daimler
considered these costs. Basing the NCP on the total costs of SCR without accounting for
Navistar's costs would actually give a substantial competitive advantage to complying
manufacturers - a result clearly not consistent with the intent of the statute.
2.6. Baseline Engine for Cost Analysis
To calculate the 90th percentile costs of compliance relative to the upper limit, we have
to define a baseline engine with NOx emissions at the upper limit. For the proposal, we assumed
a baseline engine with engine-out NOx emissions equivalent to the compliant engines (and thus
having no impact on fuel consumption), but that included less expensive hardware and a lower
DEF consumption rate. We calculated compliance costs relative to such an engine.
Note that comments related to the numerical value of the upper limit are discussed in
Section 2.4 and comments related to whether compliance costs should be calculated based on the
upper limit are discussed in Section 2.5. Also, many comments about the baseline engine that
relate to specific cost categories (such as specific hardware elements or fuel consumption rates)
are addressed in other sections of this document.
Summary of Comments
CATF EPA-HQ-OAR-2011-1000-0021-0022
Cummins EPA-HQ-OAR-2011-1000-0015, 0025 & 0048
Daimler EPA-HQ-OAR-2011-1000-0028, 0043-0045 & 0049
ICCT EPA-HQ-OAR-2011-1000-0030
Mack EPA-HQ-OAR-2011-1000-0024, 0046, 0047 & 0051
Navistar EPA-HQ-OAR-2011-1000-0027, 0031-0039, 0042 & 0050
Mack commented that the only way to avoid what it calls the arbitrary, speculative and
unreliable results associated with EPA's 0.50 g/bhp-hr hypothetical engine would be to engage in
extensive engineering work to determine what engine configuration would be appropriate to
meet a 0.50 g/bhp-hr NOx level and then determine the full set of hardware, operating,
maintenance, and research and development costs associated with such an engine. It stated that it
is unreasonable for EPA to expect manufacturers to engage in such a burdensome, resource-
intensive and time-consuming exercise for purposes of developing an NCP. Mack commented
that absent such an undertaking, however, EPA's proposed approach does not yield reliable
results. It continued to say that given that EPA is not an engine manufacturer itself, its
hypothetical engine cost analysis cannot substitute for actual cost data from manufacturers.
Mack concluded that in light of the ample amount of actual cost information available to EPA,
there simply is no need to invent entirely new, hypothetical engines.
Mack also commented that though every other manufacturer opted to use SCR
technology, it did so for purposes of complying with a 0.20 g/bhp-hr standard, not a 0.50 g/bhp-
hr standard. It commented that it is inappropriate for EPA to "conclude that SCR manufacturers
39
-------
would have relied on SCR technology to comply with a 0.50 g/bhp-hr limit without considering
and analyzing a host of additional costs potentially associated with the use of this technology."
Mack stated that these "include additional warranty costs associated with SCR systems; costs
associated with designing and adapting vehicles to accommodate SCR technology; and operating
and maintenance costs associated with SCR systems" and that "these costs must be considered in
developing an NCP, and cannot be conveniently ignored based on unsubstantiated speculation
that all manufacturers would have incurred them equally under a 0.5 g/bhp-hr standard as under a
0.2 g/bhp-hr standard." Mack stated that EPA needs to consider the full engineering and
development costs related to meeting the 0.20 g/hp-hr engine.
Mack also argues that EPA speculated that "manufacturers would have pursued SCR
technology to meet the fuel-economy demands of their customers and the Agency's own
greenhouse gas standards for heavy-duty diesel engines" but that "there is no reason why
manufacturers could not have achieved substantially improved fuel economy using technologies
other than SCR, if given the opportunity to design to a less stringent standard." It also stated
"EPA's conclusion that new fuel economy mandates under the GHG rule would have driven
manufacturers to use SCR makes no sense, given that the GHG rule was not even proposed until
November 30, 2010, after the MY 2010 standards took effect, and years after manufacturers had
made their decisions regarding the technologies they would use to meet the 2010 standard."
CATF agreed with the proposed baseline engine approach, but recommended changes in
other areas, particularly regarding its view that manufacturers would attempt to optimize fuel
economy. Responses to these comments are included in Section 2.8.2.
ICCT stated that the definition of the baseline is a significant flaw that led to undervalued
non-compliance penalties because it does not consider reoptimization of an SCR engine for fuel
efficiency and because it "does not account fully" for different levels of investment in
technology. It supports using two different baseline engines to produce two different penalty
curves. (See Section 3.4 for our response to this last comment.)
Daimler commented that SCR engine manufacturers have invested hundreds of millions
of dollars to develop SCR and that "any NCP cost methodology which assumes, contrary to
reality, that the noncompliant manufacturer has already spent money to develop and install SCR
on each truck, radically understates actual costs." It also stated during the hearing that "EPA
should not predicate its rule on a nonexistent hypothetical baseline engine with SCR simply to
address an operating cost disparity that does not exist, according to the company seeking NCPs."
Navistar commented that EPA's baseline selection is reasonable for calculating NCPs
and it supports the agency's determination, but that costs estimated using Navistar's alternative
baseline would result in a significantly lower NCP level than EPA selected.
Navistar also responded to comments made about fuel consumption by SCR engine
manufacturers at the public hearing. Those manufacturers objected to EPA's assumption that
there would be no difference in fuel consumption between the baseline and compliant engines.
Navistar stated that their objections miss the point. It said that while there are other paths to
reduce NOx emissions from 0.50 g to 0.20 g, EPA should not (and legally cannot) select a
particular path because it is the most expensive. Rather, in Navistar's view, EPA reasonably
40
-------
selected the path that was available and would have been selected by the manufacturers (who are
the "competitors" that should not be disadvantaged).
Cummins supports our proposal to base compliance costs on an optimized SCR engine.
However, it emphasized that an SCR engine truly optimized for 0.50 g/hp-hr NOx would have
lower hardware and operating costs than EPA's baseline engine, and recommended specific
changes to our estimates of such an engine's hardware and performance characteristics (see
Sections 2.7 and 2.8). Cummins noted that it is currently producing heavy-duty SCR-equipped
engines with NOx emissions up to 0.50 g/hp-hr.
Response
We have determined that the appropriate upper limit is 0.50 g/hp-hr, and no manufacturer
supported setting the upper limit above that level. As noted in Section 2.5, the regulations have
always required compliance costs to be calculated relative to engines at the upper limit, and we
are not revising this requirement. Since manufacturers cannot introduce engines that emit above
the upper limit, it is appropriate to determine the competitive disadvantage to the complying
manufacturer from this upper limit.
Prior to the proposal, we considered other technology paths suggested by manufacturers
in additional to our proposed baseline engine. Some manufacturers suggested baseline engines
with EGR but not SCR, or baseline engines with SCR but not EGR. However, we believed (and
continue to believe) it is likely that these baseline engines would have had high operating costs.
In general, relying on EGR to reduce NOx emissions significantly increases fuel consumption.
Although Daimler is correct that Navistar has stated in marketing material that SCR engines do
not have an operating cost advantage over its engines, this does not change our assessment that
increasing EGR increases fuel consumption. Similarly, relying on SCR to reduce NOx
emissions without EGR requires a larger amount of DEF. In either case, over the life of a truck,
the increased operating costs could be greater than the original hardware cost. We stated that we
did not have the information required to calculate these operating costs with the accuracy needed
to use these scenarios as the basis of our NCPs. However, the only comments we received on the
technology of the baseline engine during the comment period supported using an optimized SCR
baseline engine that included EGR, or a non-SCR engine.
We agree with commenters that our proposed baseline SCR engine does not represent
how an engine manufacturer would optimize an SCR engine to meet 0.50 g/hp-hr based on the
latest price projections for fuel and DEF. We have revised our baseline engine significantly so
that it now is very similar to the baseline engine recommended by Cummins. Mack objected to
our use of any hypothetical baseline engine for compliance costs. However, since the baseline
engine must reflect the upper limit (see Section 2.5) and Navistar is the only manufacturer
41
-------
producing heavy heavy-duty engines at the 0.50 g/hp-hr upper limit14, there are essentially only
three options:
1. Setting the upper limit at 1.2 g/hp-hr (or higher) and using a pre 2010 engine as the
baseline engine.
2. Using the Navistar engine (or a similar engine) as the baseline engine.
3. Using a hypothetical optimized SCR engine as the baseline engine.
Setting the upper limit at 0.50 g/hp-hr means that, at least for heavy heavy-duty engines,
the only alternative to using a hypothetical engine would have been to use Navistar's engine. To
address this comment, we evaluated using a baseline engine equivalent to Navistar's engine.
However, contrary to the view of many commenters, we found that basing our NCPs on costs of
compliance relative to a reoptimized SCR engine was the best approach to removing the
competitive advantage for both SCR and non-SCR manufacturers that could potentially use
NCPs. This is because we calculated the full lifetime cost savings of using an optimized SCR
engine that emitted 0.50 g/hp-hr NOx, compared to an engine meeting the standard, to be greater
than the cost savings of using an advanced EGR engine to that emitted 0.50 g/hp-hr NOx. See
Appendix B of the Final Technical Support Document for additional details about this
comparison.
Contrary to Mack's comment, our cost estimates are neither arbitrary nor speculative. They are
based on our unbiased view of the collection of data provided by manufacturers (both publicly
and confidentially) as well as other data collected by EPA. In particular, we note that our costs
are based on many unbiased and publicly available sources of data, such as those discussed in
Section 2.3. (See the Final Technical Support Document for additional discussion of the basis of
our cost estimates.) While there is still some uncertainty associated our estimates, they are more
reliable than relying solely on the word of specific manufacturers who have an interest in either
maximizing (for complying manufacturers) or minimizing (for manufacturers using NCPs) the
NCP level. Moreover, because much of the information manufacturers have provided was
provided under protection as confidential business information, EPA must in part provide
information other than information from specific manufacturers to justify its cost estimates in
this final rule.
With respect to Mack's comment that it is inappropriate for EPA to "conclude that SCR
manufacturers would have relied on SCR technology to comply with a 0.50 g/bhp-hr limit
without considering and analyzing a host of additional costs potentially associated with the use
of this technology", we note that we are not concluding what manufacturers would have done.
We also did not speculate or conclude that "manufacturers would have pursued SCR technology
to meet the fuel-economy demands of their customers and the Agency's own greenhouse gas
standards for heavy-duty diesel engines." This confusion seems to be caused by Mack's
misreading of the following text from the Interim Technical Support Document:
14 Cummins is producing medium heavy-duty engines with NOx emission near 0.50 g/hp-hr and heavy heavy-duty
engines with NOx emission near 0.35 g/hp-hr, but not heavy heavy-duty engines with NOx emission near 0.50
g/hp-hr
42
-------
The greenhouse gas requirements are potentially relevant to this NCP rulemaking
because they affect the types of emission controls manufactures would pursue. As noted
later, we believe that it is appropriate to assume that even if the NOx standard was higher,
manufacturers would not have chosen emission controls that would have increased fuel
consumption rates because they must also meet the greenhouse gas emission standards.
Note that this text does not mention SCR. We drew no conclusions about what manufacturers
would have specifically done. Rather, we are basing our compliance costs on what we believe to
be the greatest cost savings that a manufacturer could achieve by meeting a 0.50 g/hp-hr upper
limit compared to the 0.20 g/hp-hr standard, considering what could be achieved currently or in
the near future. Our calculation is that the greatest cost savings a manufacturer could obtain
would be by reoptimizing SCR engines for 0.50 g/hp-hr in the near future.
Similarly, we disagree with Daimler's comment that implies that our baseline engine
selection results in a "cost methodology which assumes, contrary to reality, that the
noncompliant manufacturer has already spent money to develop and install SCR on each truck."
We made no such assumption. We estimated compliance costs for different noncompliant
baseline engines (most notably, a hypothetical SCR engine and an engine equivalent to
Navistar's non-SCR engine), and selected the COCgo value that will result in penalties that
protect compliant manufacturers from being at a competitive disadvantage with respect to any of
them. The extent to which Daimler comments address which technologies should be included on
the baseline engine representing engines at 0.50 g/hp-hr is unclear because Daimler conflates the
issue of whether compliance costs should be estimated relative to the upper limit with the issue
of which technologies should be included with the baseline engine. We included SCR-equipped
engines in some of our baseline alternatives, and indeed used an SCR-equipped engine as our
final baseline engine, because it is a completely reasonable choice for an engine manufacturer to
use SCR technology to meet a 0.50 g/hp-hr emission level. Cummins already has done for some
of its medium heavy duty engines. See Section 2.5 for our response to Daimler's objections to
calculating compliance costs relative to the upper limit and Appendix B of the Final Technical
Support Document for our consideration of alternate baseline engine scenarios.
While we are not basing our costs on the baseline engine recommended by Navistar, we
did evaluate such a scenario (as described in Appendix B of the Final Technical Support
Document). Consistent with Navistar's comment, we found that such a scenario would probably
have resulted in lower penalties. However, we also found this scenario to be inappropriate. First
and foremost, the COCgo value we estimated was about one-quarter of the COCgo for optimized
SCR engines. Thus, NCPs based on this scenario would not large enough to ensure that SCR
engine manufacturers could not gain a competitive advantage by reoptimizing their engines for
0.50 g/hp-hr and paying the lower NCP. In addition, the reason the COCgo for the non-SCR
baseline scenario is so much lower than the SCR baseline scenario is that we estimate that
compliant SCR engines actually have significantly lower operating costs than engines like
Navistar's 0.50 g/hp-hr engines, which largely offsets the higher hardware costs. However,
Navistar does not agree that its engines have higher operating costs. Thus, we cannot be sure
that NCPs based on the lower COCgo value would completely remove the competitive advantage
for Navistar, let alone optimized SCR engines.
43
-------
Finally, while we appreciate Navistar's support for our proposed baseline engine
analysis, it is important to address Navistar's statement that "EPA should not (and legally
cannot) select a particular path because it is the most expensive." The Clean Air Act directs
EPA to set the NCPs at a level that will "remove any competitive disadvantage" to complying
manufacturers. To do this, EPA determines the baseline engine using the most cost-effective
way for a non-complying manufacture to, in this case, manufacture an engine meeting a 0.50
g/hp-hr level, compared to a 0.20 g/hp-hr. Were we to determine the baseline engine using an
engine that did not maximize cost-effectiveness, complying manufacturers would be potentially
subject to a competitive disadvantage if a noncomplying manufacturer chooses a more cost-
effective 0.50 g/hp-hr than the one EPA used for its analysis. EPA would not characterize this as
the most expensive path, but it is the path that maximizes the potential benefits that a
manufacturer may achieve through noncompliance, and thus is appropriate for evaluating the
proper NCP.
2.7. Hardware Costs
For the NPRM, we projected that the hardware costs to reduce NOx emissions from 0.50
to 0.20 g/hp-hr would fall into two categories -an additional sensor to improve the dosing
control, and improved SCR catalyst to increase the NOx conversion efficiency. We assumed the
lower tailpipe NOx emissions levels would require the addition of a sensor for better control over
the urea injection. This cost would apply for both the typical and worst case engines. In
addition, for the worst case manufacturer, we projected a larger SCR catalyst would be required
to increase the NOx conversion efficiency. We estimate that the size and loading of the catalyst
would need to increase by about 20 percent.
We did not estimate total costs for SCR for the NPRM. See Sections 2.3, 2.5, and 2.10
for comments related to total SCR costs.
Summary of Comments
Cummins EPA-HQ-OAR-2011-1000-0015, 0025 & 0048
Daimler EPA-HQ-OAR-2011-1000-0028, 0043-0045 & 0049
Mack Undocketed Confidential Business Information
Cummins was the only manufacturer that provided hardware cost information related to
our proposed cost methodology. (As noted below, other manufacturers provided hardware cost
information that was not directly related to our proposed cost methodology.) Cummins stated
the following in comments:
EPA did allow for precious metal reduction in the SCR catalyst;
however, there are other hardware changes between compliant and
noncompliant engines that were not taken into account. Precious metals used
for the diesel oxidation catalyst (DOC) and ammonia oxidation catalyst could
be reduced and reformulated with a greater substitution of cheaper precious
metals for the optimized NCP baseline. The cost reduced DOC is possible
44
-------
because the optimized NCP baseline engine would experience more passive
DPF regeneration due to the higher engine out NOx and would not need as
much nitrogen dioxide (NO2) production from the DOC. This results in lower
precious metal content required to produce the needed levels of NO2 used for
the DPF regeneration process. The SCR catalyst performance over the
lifetime of the vehicle is not diminished using the cost reduced DOC because
the SCR catalyst will experience less high temperature exhaust due to lower
temperature DPF regeneration (i.e., less thermal aging of the SCR catalyst).
The NCP baseline engine will have a reduced need for aftertreatment and
vehicle insulation to prevent excessive heat loss. Thermal management of the
aftertreatment system and providing acceptable driving performance to the
end user will be easier to achieve on the NCP baseline engine so cost reduced
turbomachinery can be utilized. Additionally, improved emissions
measurement devices for NOx-related, on-board diagnostics would be
removed, because the NCP baseline engine will not have to measure the
smaller amounts of NOx and a sensor with a higher level of accuracy will not
be needed. These hardware differences are important to account for when
going between a nonconforming and compliant engine. A customer
purchasing a 0.50 g/hp-hr NOx engine would not be willing to pay the
additional upfront costs of extra precious metal, insulation, turbomachinery,
and OBD equipment that the engine does not require for regulatory or
performance reasons.
Daimler commented that even if there were some rational basis for using a hypothetical
0.50 g/bhp-hr engine as the baseline engine for NCP cost calculations, the engine hypothesized
by EPA bears no resemblance to the engine that DTNA would have developed for 0.50 g/bhp-hr
NOx performance. Specifically, DTNA's 0.50 g/bhp-hr SCR engine would have had much better
fuel economy and lower material costs than the 0.50 g/bhp-hr engine that EPA hypothesizes.
Daimler also provided confidential information on total SCR costs, including hardware
costs. The costs reflected the difference between a baseline 2008 engine and a compliant 2010
engine. The baseline engine included turbocharging, aftercooling, electronically controlled
common rail fuel injection, cooled EGR, oxidations catalyst, a catalyzed diesel particulate filter,
and an open crankcase breather with separator. Daimler's compliant 2010 engine had similar
technology as their baseline engine, with the addition of a SCR system with urea injection, post-
SCR oxidation catalyst, and OBD diagnostics associated with the aftertreatment system.
Mack also provided confidential information on total SCR costs, including hardware
costs.
45
-------
Response
For the final rulemaking, we determined that an SCR aftertreatment system originally
designed for 0.50 g/hp-hr NOx would require hardware modifications to achieve tailpipe
emissions levels of 0.20 g/hp-hr. Based on conversations with manufacturers, we believe that the
lower NOx emissions levels would require:
• The addition of SCR catalyst volume to gain NOx efficiency,
• Improvements to the diesel oxidation catalyst (DOC) due to the lower engine-out
NOx emissions and higher soot conditions,
• An optimized turbocharger to cover a broader range of EGR flow, and
• A sensor for better control over the urea injection.
We note that these hardware elements are largely consistent with the information provided by
Cummins in its comments.
As noted in the previous response and included in detail in the Appendix B of the Final
Technical Support Document, we considered and analyzed alternative baseline engines,
including baseline engines with only EGR, in the final rulemaking. The comments from Daimler
and Mack were helpful to EPA in developing our costs for those alternative baseline engines and
in evaluating these alternative baseline engines. However, as discussed above, those alternative
baseline engines were determined not to be appropriate in developing our final NCPs and they
were not relevant to our estimates of compliance costs in our final methodology.
2.8. Operating Costs
We received comments on several aspects of our proposed analysis of operating costs.
While the comments on these costs are addressed separately in the following subsections, we
also considered the costs collectively to ensure that our final estimate of the COCgo operating
costs accurately reflected the 90th percentile value for the potential market impact of lower
operating costs for the baseline engine.
2.8.1. Cost Savings and Performance Improvements
Summary of Comments
Cummins EPA-HQ-OAR-2011-1000-0015, 0025 & 0048
Daimler EPA-HQ-OAR-2011-1000-0028, 0043-0045 & 0049
Mack EPA-HQ-OAR-2011-1000-0024, 0046, 0047 & 0051
PACCAR EP A-HQ-O AR-2011-1000-0041
Cummins supports EPA's use of total life cycle cost and the inclusion of operating costs
for all future years in the NCP calculation because engines that use NCPs will have a cost
advantage for the entire life of the engine. This cost advantage will include lower fuel
consumption and better resale value based on this lower fuel consumption. It further stated
46
-------
engines that use NCPs will emit a higher level of NOx for the entire life of that engine, not just
the first five years. Cummins also noted that if the entire life cycle cost is not included, this
could result in a lower NCP which would be more enticing for manufacturers to use, resulting in
a greater NOx impact on the environment.
PACCAR stated that "EPA should include in the NCP calculation the fuel consumption
advantages for engines that do not meet the 0.20 gram standard" because the "fuel economy
advantage is extraordinarily important in today's engine market."
Daimler commented that customers' operating costs are arguably relevant to
manufacturers' compliance costs only to the extent they impact the purchase price that customers
are willing to pay for compliant manufacturers' trucks. Based on industry sales and marketing
experience, DTNA's Product Planning Department generally estimates that truck customers are
willing to pay for of net fluid superiority when making new truck purchase decisions. Daimler
noted that SCR engines have a net fluid superiority but that it is less than the fixed and variable
costs of SCR. Therefore it concluded the net fluid superiority of SCR is no reason not to use
manufacturers' actual costs of compliance, even if EPA erroneously includes the fuel benefit in
the calculation.
Daimler also commented that EPA has never counted the fuel efficiency benefit of
increased emissions controls against compliant manufacturers, and there is no reason to start
now. Daimler stated that in the past, EPA included the fuel economy penalty associated with
emissions compliance in the NCP because failing to do so would have effectively created a
financial incentive to use the NCP rather than meet emissions standards. Daimler said that the
"entire focus on fuel costs has always related to the need to disgorge any potential benefit of
noncompliance from the laggard, thus ensuring no harm to compliant manufacturers." The
purpose was not to fully equalize the operating cost and net present value of compliant and
noncompliant products. Daimler also stated that attempting to deduct all commercial benefits of
environmental compliance would be contrary to the congressional intent behind NCPs and EPA's
mission more generally. It stated that Congress provided NCPs as actual "penalties," and that
attempting to extract all of the indirect benefits of environmental compliance would also
eliminate the deterrence element that is critically important to the NCP program.
Daimler and Mack's comments stated that in the 1990 NCP rule, EPA explained that it
did not account for any side benefits of the emissions reduction technologies that compliant
manufacturers adopted and that it would be inappropriate to do so. Mack stated that "by factoring
in the value of the side benefits of SCR emissions technology, the Agency is creating incentive
for companies such as Navistar to claim an artificially high fuel economy penalty associated with
the use of its technology (relative to the complying technologies) for purposes of setting an
NCP." Daimler and Mack both noted that EPA excluded side benefits from NCP calculations in
1990 for the following three reasons:
1. That it was difficult to place a dollar value on the side benefits, especially
intangible benefits like increased power or improved driveability.
2. That not all potential buyers will place the same value on these side benefits such
as less intensive users placing a lower value on improved fuel consumption.
47
-------
3. That if a conforming manufacturer is forced for emissions control reasons to add a
technology to an engine line whose users do not value the side benefits as much
as EPA's estimate assumes they do, that manufacturer will be at a disadvantage in
comparison to a non-conforming manufacturer.
Daimler cited our prior statement that "EPA has an overriding obligation to protect the
conforming manufacturer." Daimler and Mack commented that if customers do not value the
side benefit as much as EPA has assumed they will, the compliant manufacturer who uses that
emissions reduction technology will be at a disadvantage compared to noncompliant
manufacturers.
Mack also commented that it believes that "EPA has stated in its generic NCP
rulemaking that user costs, such as fuel economy penalties, should be taken into account only to
the extent they result in placing compliant engines at a disadvantage." Mack's comments stated
that EPA noted:
Users' costs may include the present value of any fuel economy penalty,
requirements for unleaded fuel, and changes in maintenance costs
associated with operating a conforming engine or vehicle over its lifetime.
To the extent that these cost impacts are known by potential buyers, a
purchaser should be willing to pay more for an engine or vehicle with lower
total user costs. Therefore, in order to remove a manufacturer's economic
incentive for nonconformance and to prevent manufacturers of conforming
engines or vehicles from being placed at a competitive advantage, the NCP
must also reflect user cost increase.15
Mack also stated:
This is in keeping with Congress' intent that the calculation "would include
such items as the actual cost of compliance for complying vehicles, the
capital costs foregone as a result of non-compliance, the market value of
any fuel economy gains made by noncomplying vehicles compared to
complying vehicles and the competitive advantage that may arise because
of the lesser warranty and recall obligations for noncomplying vehicles
compared to complying vehicles." H. Conf. Report No. 95-564, p. 163,
reprinted at 77 U.S.C.A.A.N. 1544. It is clear from both the legislative
history of the Clean Air Act and EPA's generic NCP rule that fuel economy
penalties are only to be considered to the extent they give a non complying
manufacturer a competitive advantage over a complying manufacturer.
15 50 FR 35383, August 30, 1985.
48
-------
Response
We agree with Cummins that our cost analysis should consider full life cycle costs, which
our analysis does. As discussed earlier, suggestions that EPA not deviate from prior precedents
for calculating costs incorrectly imply that EPA has always used the same methodology. Even if
it were true that EPA had used the same methodology for all prior NCP rules, this would not
obligate us to continue using the same methodology in perpetuity. The methodology EPA uses
to account for cost savings is dependent on the factual circumstances for each NCP rule. The
methodology we used to develop operating costs is fully in compliance with the regulations and
the statute, and is a completely reasonable method for accounting for operating costs in this rule.
We do not agree with Mack's assertion that EPA stated in the generic rule that "user
costs, such as fuel economy penalties, should be taken into account only (emphasis added) to the
extent they result in placing compliant engines at a disadvantage." Mack inappropriately inserts
the word "only" where it was not used. The text Mack cites to support its claim includes the
recognition by EPA that "to the extent that these cost impacts are known by potential buyers, a
purchaser should be willing to pay more for an engine or vehicle with lower total user costs."
Mack does not dispute that its fuel savings are known by potential buyers, or that buyers would
be willing to pay more for an engine with lower total user costs.
We do not believe the reasons previously given in the 1990 NCP rule for not considering
side benefits apply for fuel consumption savings in this NCP rule - at least not to point of
justifying ignoring the fuel savings associated with SCR. First and foremost, as PACCAR and
Cummins noted in their comments, current purchasers of heavy heavy-duty diesel engines care
very much about changes in fuel consumption. Thus, it is not possible to accurately reflect the
competitive market if we ignore fuel savings. As noted in Section 1.1.1, while the statute
requires us protect the competitive interests of complying manufacturers, we cannot knowingly
set the first year NCP at a level that is much higher than the actual dollar value of the
competitive disadvantage of the complying manufacturers. We believe that simply ignoring fuel
savings could easily result in penalties that are several thousand dollars higher than needed or
appropriate to remove the competitive disadvantage for complying manufacturers. Moreover, it
is contrary to the clear evidence that fuel consumption of these vehicles is a key element in the
purchasing decisions of customers, and thus the value of the engines. As noted in the excerpt
from the legislative history included in Mack's comment, Congress intended EPA to address the
"market value" of differences in the engines. Nothing in the legislative history indicates that our
evaluation of competitive disadvantage requires us to ignore any real tangible benefits that exist
for complying engines and that would affect the competitive balance of manufacturers.
It is also relevant that EPA noted in 1990 that its concerns about assigning dollar values
for side benefits were more significant for "intangible benefits like increased power or improved
driveability." There was somewhat less concern in that rule about fuel economy benefits. As
Daimler noted, historically compliance with NOx standards hurt fuel economy. This is in
substantial contrast to the very tangible (if somewhat difficult to project) benefits of higher fuel
efficiency, particularly for these vehicles. In fact, for this current rulemaking, we received
comments from manufacturers insisting that we can and must include costs associated with
changing fuel consumption. Indeed, Mack, in criticizing our decision in the proposal not to
include potential fuel economy benefits associated with complying with a 0.50 g/hp-hr upper
49
-------
limit, rather than a 0.20 g/hp-hr standard, noted the importance of accounting for increased fuel
efficiency. While we acknowledge that precisely quantifying the value that purchasers will place
on fuel savings can be difficult, we believe that this issue is more appropriately addressed by
choosing a cost methodology that takes this difficulty into account in estimating fuel savings.
See the following sections for our discussions of how to best estimate the values of fuel savings.
Even if we were to exclude operating cost savings from our COC90 calculations, we
believe that it would be appropriate to determine whether such operating cost savings exist by
analyzing the combined consumption of'fuel andDEF, which is commonly described as "fluid
consumption. " In other words, if we were to agree with Daimler and Mack that our cost analysis
should ignore operating cost benefits, we should only ignore them to extent that a noncompliant
engine's cost savings with regard to reduced DBF use exceeds the increased fuel cost of such an
engine. Our understanding of the market is that operators evaluating SCR equipped engines
consider fluid consumption costs rather than the traditional consideration of fuel consumption.
Our conversations with manufacturers, as well as Daimler's comments, further support this
concept. When considered this way, our methodology is consistent with the prior approaches
because it counts the additional fluid consumption costs for compliant engines, relative to the
upper limit engine. Neither Daimler nor Mack provides any basis for separating DEF
consumption from fuel consumption when both manufacturers and operators consider them
together as fluid consumption. Moreover, Daimler acknowledges that purchasers are willing to
pay for net fluid savings.
As described in the Final Technical Support Document, with the revised baseline engine
used in the final analysis, we estimate that reducing emissions to 0.20 g/hp-hr would require
recalibrating the engine to decrease engine-out NOx which would increase fuel consumption.
These costs would be partially offset by lower DEF costs. Thus, while we are including the
benefit of reduced DEF usage, it is only in the context of a net increase in fluid costs. We are
doing this because it reflects how the market actually values these changes.
Mack's comment that "the Agency is creating incentive for companies such as Navistar
to claim an artificially high fuel economy penalty associated with the use of its technology",
would be true only to the extent that EPA relies on Navistar's claims. However, this issue is not
relevant because EPA is independently estimating costs. Moreover, applying this logic would
preclude EPA from considering any costs or savings. For example, according to this logic, EPA
could not consider DEF costs because doing so would create an incentive for SCR manufacturers
to claim artificially high DEF costs. Clearly this would be unworkable.
It is also important to note that manufacturers could possibly have used compliance
strategies that would have had lower upfront hardware costs. When first adopting the 0.20 g/hp-
hr NOx standard, EPA projected that manufacturers would have used lean NOx traps (LNTs -
also known as NOx adsorbers) to meet the standard. We still believe that manufacturers could
have used LNTs and that they may have had less expensive hardware. Instead, manufacturers
chose to add SCR hardware in part because it would result in lower fuel consumption for
operators. Clearly, this implies that fuel savings would make engines more desirable for
purchasers and have an impact on the competitive balance in the marketplace.
50
-------
2.8.2. DEF Usage and Fuel Consumption Rates
Summary of Comments
CATF EPA-HQ-OAR-2011-1000-0021-0022
Cummins EPA-HQ-OAR-2011-1000-0015, 0025 & 0048
Daimler EPA-HQ-OAR-2011-1000-0028, 0043-0045 & 0049
Ford EPA-HQ-OAR-2011-1000-0029
ICCT EPA-HQ-OAR-2011-1000-0030
Mack EPA-HQ-OAR-2011-1000-0024, 0046, 0047 & 0051
Navistar EPA-HQ-OAR-2011-1000-0027, 0031-0039, 0042 & 0050
PACCAR EP A-HQ-O AR-2011-1000-0041
Ford estimates that, for a medium heavy-duty engine manufacturer who would otherwise
fully comply with the NOx standard via a SCR aftertreatment system, relaxing the NOx standard
from 0.20 g/hp-hr to 0.50 g/hp-hr would allow for a 2.7% improvement in fuel economy at
constant DEF dosing. It stated that from a marketing perspective, the ability to provide such a
fuel efficiency benefit to the customer offers a significant competitive advantage. Ford noted that
from the perspective of compliance with the 2014+ MY EPA and NHTSA Heavy Duty
Greenhouse Gas and Fuel Consumption Standards, it provides an additional competitive
advantage by displacing the need to deploy other, more costly technologies that would
potentially be required to achieve a 2.7% improvement in fuel economy.
ICCT stated that the assumptions about fuel efficiency are significant flaws that lead to
undervalued non-compliance penalties. ICCT commented that the assumption that an engine
fully optimized at 0.50 g/hp-hr would have the same engine-out NOx as an engine design to meet
0.20 g/hp-hr is likely unrealistic. It further stated that it would be reasonable to assume that
manufacturers would likely look to maximize fuel efficiency in order to deliver savings to their
customers and enhance their competitiveness (and that this would affect NOx emissions given
the relationship between engine-out NOx and fuel consumption). The ICCT recommends that if
non-compliance provides a fuel efficiency benefits and a competitive advantage, these benefits
should be included in the assessment of NCPs.
Cummins commented that its SCR system is delivering up to 6% better fuel economy
than the engines it produced in 2009 and that fuel consumption is a major consideration when
customers are purchasing a new heavy-duty vehicle. Cummins stated that an NCP engine at the
upper limit would have a 2% fuel consumption advantage over a compliant engine when fuel and
DEF consumption are optimized. It stated that rather than holding engine-out NOx as constant, a
manufacturer could hold constant the percent reduction of NOx across the SCR catalyst, which is
a more realistic optimized baseline. This leads to a fuel consumption advantage due to higher
engine-out NOx, reduced exhaust backpressure, and reduced particulate filter regenerations for
the NCP engine. Cummins provided fuel economy data based on a proprietary drive cycle
which demonstrated a 2% fuel economy benefit.
CATF commented that given the obvious advantages of reduced fuel consumption, it
would be entirely reasonable for an engine manufacturer to meet a 0.50 g standard by allowing
engine-out emissions to increase substantially above 2.0 g/hp-hr in order to take advantage of the
51
-------
high removal efficiencies of SCR. Thus, assuming a 90% SCR efficiency, engine-out emissions
could approach or slightly exceed 4.0 g/hp-hr and still meet the 0.50 g upper limit at the tailpipe.
CATF also submitted a memo from M.J. Bradley and Associates with their comments. M.J.
Bradley conducted an analysis of the NCP values in response to comments made by engine
manufacturers in the public hearing. It noted that Volvo indicated a 2 percent reduction in fuel
use was possible for an engine meeting 0.50 g/hp-hr NOx tailpipe levels relative to an engine
with 0.20 g/hp-hr NOx. MJ. Bradley agreed with this assessment and estimated that it could be
achieved with a baseline engine emitting 4.0 g/hp-hr NOx engine-out, while the engine meeting
the 2010 NOx standards would emit 2.0 g/hp-hr. MJ. Bradley also noted that Cummins has
publicly stated that the brake thermal efficiency of their 1998 engines with 4.0 g/hp-hr engine-
out emissions was 3 percent higher than their 2002 engines with 2.5 g/hp-hr engine-out NOx
emissions.
Daimler commented that even if there were some rational basis for using a hypothetical
0.50 g/bhp-hr engine as the baseline engine for NCP cost calculations, the engine hypothesized
by EPA bears no resemblance to the engine that Daimler would have developed for 0.50 g/bhp-
hr NOx performance. Specifically, Daimler's 0.50 g/bhp-hr SCR engine would have had much
better fuel economy and lower material costs than the 0.50 g/bhp-hr engine that EPA
hypothesizes.
PACCAR commented that it "estimates that the fuel consumption advantage for the 0.50
gram design is approximately 4% over a 0.20 gram design, which, allowing for a 2% increase in
DEF usage, would result in a net cost reduction of $13,833." (PACCAR cited its experience
designing engines meeting emissions standards of EPA 2007, EPA 2010 with both EGR and
EGR/SCR technologies, Euro V, and Euro IV). PACCAR commented that "this fuel economy
advantage is extraordinarily important in today's engine market." PACCAR noted that Cummins
produces only one rating of its ISX15 engine that is certified at 0.20 g/bhp-hr NOx which is sold
primarily to customers who must meet municipal bid requirements. PACCAR stated that the
overwhelming majority of Cummins' large block engines are sold at higher certification levels
using credits because of the advantage for customers in engine performance and for Cummins in
the market compared to other engine manufacturers. PACCAR concluded that this example
highlights PACCAR's position that EPA should include in the NCP calculation the fuel
consumption advantages for engines that do not meet the 0.20 gram standard.
Navistar commented that EPA inappropriately overestimates the lifetime costs of DEF,
which must be corrected in the final rule. EPA also assumes that DEF will always be used in
the SCR-equipped truck during those 30 years. Navistar commented that later assumption has
no basis in fact, as the SCR system is authorized to operate many thousands of miles (and,
indeed, operates indefinitely) without using any DEF. It also stated that there is thus no record
basis for such significant DEF costs, and EPA must recalculate its NCP using appropriate
assumptions regarding DEF use. It based its concern in part on the Federal Motor Carrier
Safety Administration roadside inspection statistics between 2008 and 2012, which it claimed
show that "trucks" are in some type of violation over 60% of the time, including violations that
result in an "out-of-service" order nearly 20% of the time. It stated that if 20% of the vehicles
inspected contain violations serious enough to require them to be taken out of service, it seems
completely reasonable to assume that at least 20% of the time DEF will not be used. Navistar
concluded that for purposes of calculating the current NOx NCP, EPA cannot estimate the
52
-------
lifetime cost of DEF on the basis that DEF is always used—when it is not— because it grossly
miscalculates the true SCR operating costs for end users and, as a result, goes well past
removing any purported competitive disadvantage to complying manufacturers. Thus, Navistar
is effectively arguing that we should assume a smaller change in the DEF consumption.
Mack also stated that EPA's position in the NCP NPRM is inconsistent with the
Agency's position in its Final Rule establishing GHG emissions standards for heavy-duty
engines and vehicles. Mack commented that in the GHG regulation, EPA provides an exception
for increased fuel consumption where NOx emissions are decreased from 0.50 g/bhp-hr to 0.20
g/bhp-hr, thus verifying that the Agency recognizes a fuel efficiency benefit associated with
complying with the 0.5 g/bhp-hr standard over the more stringent standard. 76 Fed. Reg. 57,205
(Sept. 15, 2011). Mack also noted in the public hearing held on March 5, 2012, that it is possible
to achieve a 2 percent fuel savings with a 0.50 g/hp-hr engine versus a 0.20 g/hp-hr engine.
Response
Based on new information and comments we received, we are revising our baseline
engine for the heavy heavy-duty service class. Specifically, we are revising the COCgo baseline
engine to be more optimized for lower fuel consumption at 0.50 g/hp-hr NOx than was assumed
for the proposal. This change is consistent with the majority of the comments we received on the
fuel consumption rate of the baseline engine. For the proposal, we estimated that reducing NOx
emissions from 0.50 g/hp-hr to 0.20 g/hp-hr would require an increase in DEF consumption but
would not change fuel consumption because we projected that there would be little price
difference between DEF and fuel. For most engines, without a difference in fuel and DEF
prices, there is little to be gained by reoptimizing for lower fuel consumption because much or
all of the savings would be offset by higher DEF costs. However, as is described in Chapter 3 of
the Final Technical Support Document, we now have new information (some of which came
from comments) indicating that fuel prices will likely be significantly higher than DEF prices for
the foreseeable future. We agree with commenters that heavy heavy-duty engine manufacturers
designing engines for 0.50 g/hp-hr NOx would have responded (and could still respond) to this
price difference by optimizing the engines to have higher engine-out NOx, which would reduce
fuel consumption, and reduce the excess NOx by increasing DEF consumption. We now
estimate that the difference in fuel consumption between the baseline engine and compliant
engines would be 1.9 percent.
53
-------
Estimating the relationship between NOx emissions and fuel consumption is difficult for
several reasons. In particular, we note the following:
• The relationship between tailpipe NOx emissions and fuel consumption is a
function of both aftertreatment efficiency and the relationship between engine-out
NOx emissions and fuel consumption.
• Manufacturers generally consider both aftertreatment efficiency and the
relationship between engine-out NOx emissions and fuel consumption to be
confidential business information.
• Manufacturers providing test data for this rulemaking have an incentive to
provide data selectively to support either a higher or lower penalty value.
• Some manufacturers providing fuel consumption data did not provide the
corresponding engine-out NOx emission data.
• The relationship between engine-out NOx emissions and fuel consumption varies
by engine model and is highly dependent on how the vehicle is operated and
factors such as injection pressure, aftertreatment heating control strategies, and
EGR rates.
Facing these challenges, we chose to estimate fuel consumption impacts by developing a curve
to represent the relationship between engine-out NOx emissions and fuel consumption and using
the curve in conjunction with our estimates of catalyst efficiencies. The fuel curve, which is
shown in Figure 2-2, largely reflects the majority of the fuel consumption data provided by
engine manufacturers. This relationship is based on the CBI data provided by manufacturers.16
While most of the data on fuel consumption versus engine-out NOx that we received are CBI,
CATF did include in its comments public data for increasing engine-out NOx emissions from a
compliant engine that was provided by M. J. Bradley. These data are also shown in Figure 2-2.
To the extent possible, manufacturer data was weighted based on the amount of detail provided
by the manufacturer and each manufacturer's approximate share of the heavy heavy-duty engine
market, and data provided in the 3.0 to 4.8 g/hp-hr range was given a greater weight (because
this is the range that represents the expected change in engine-out emissions and resulting fuel
consumption from the baseline engine to a compliant engine). The goal was to derive a curve
that is somewhat conservative to protect complying manufacturers, but not unreasonably
conservative.
16 "Consideration of Confidential Business Information on the Relationship between Engine-Out NOx Emissions
and Fuel Consumption", EPA Memorandum from Charles Moulis to Matt Spears, August 3, 2012.
54
-------
1.04
1.03
1.02
0
tL
E
2 1.01
o
u
HI
3
ai •*•
>
~5
ce
0.99
0.98
0.97
C
Engine-Out NOx Emissions Impact on Relative Fuel Consumption
\
EPAFRM
• MJ Bradley
^V
N.
_\^
^S.
^v. 1.9% Fuel
^s. Impact
3123456
Engine-out NOx Emissions (g/hp-hr)
Figure 2-2: Engine-Out NOx and Fuel Consumption Relationship
With respect to whether 1.9 percent is the correct value for COCgo, it is important to
emphasize that we do not believe that a 1.9 percent increase in fuel consumption is necessarily
the worst case possible. There is a significant amount of uncertainty related to fuel consumption
impacts and some commenters claimed that the impact could be higher than 1.9 percent - either
in public comments or in confidential business information. However, we cannot verify these
claims. For example, PACCAR stated in its comments that an SCR engine reoptimized for 0.50
g/hp-hr could achieve four percent lower fuel consumption than its current compliant engines.
While we cannot verify this number and do not know if it broadly represents in-use operation, it
does suggest that the actual fuel consumption impact for some vehicles could be somewhat
higher than our 1.9 percent. It would be more conservative (in terms of protecting complying
manufacturers) to assume the fuel impact is equal to the highest value within possible range of
impacts (perhaps as high as the value suggested by PACCAR). Nevertheless, we note that our
value is equivalent (when rounded to one significant figure) to the values recommended by
CATF, Cummins, and Mack (as Summarized in Table 2-4). Thus, we believe that using
PACCAR's four percent would be unreasonably conservative, especially since this value is not
supported by any other information available in the record for this rulemaking.
55
-------
2-4 Summary of Commenters Recommendations for the Fuel Consumption Difference between the baseline and
compliant engines
Commenter
Cummins
Mack
CATF
PACCAR
Ford
Percent Difference Between
Baseline SCR engine and
Compliant Engine
2 percent
2 percent
2 percent
4 percent
2.7 percent
Notes
Also provided CBI
Public Hearing
From MJ Bradley
Also provided CBI
Medium Heavy-Duty
We also believe that a higher value may very well overstate the market value of reduced
fuel consumption. As is described in the Final Technical Support Document, and as several
commenters have noted, there is reason to believe that many truck purchasers may undervalue
future fuel savings (relative to our estimate of the net present value). Therefore, by using the full
lifetime impacts of our estimated increase in fuel economy, we are already using a conservative
value for the market value of fuel savings benefit. We did not believe it was appropriate to make
an even further conservative assumption by using a fuel consumption rate that was higher than
the value we are using. For example, if customers typically discount fuel consumption impacts
so much that they value them at one-quarter of the lifetime cost, then the actual competitive
impact of a four percent increase in fuel consumption (the highest value recommended by
commenters) would be equivalent to the full lifetime cost of a one percent increase. Compared
to this value, 1.9 percent calculated for the full lifetime would be conservative (i.e. it would lead
to higher presumed fuel impacts for complying engines, which would lead to higher NCPs). See
Section 2.8.4 for additional discussion of comments on how to address discounting of future
costs by operators. See the Final Technical Support Document for additional discussion of how
we addressed uncertainty related to fuel consumption impacts.
We disagree with Navistar's assertion that we should not assume that DBF is always used
in the NCP calculations. EPA has provided detailed guidance to manufacturers of engines using
SCR technology regarding possible methods for safeguarding their engines against operation
without proper DBF. EPA noted that a series of warnings that alert operators to potential
operation without proper DBF, followed by engine operational controls that reduce the
effectiveness of engine performance (e.g. derates of engine power, reductions in possible speed),
would be particularly beneficial in reducing the likelihood that the vehicle will be used outside of
manufacturer's settings in use. EPA has certified many engine families that have incorporated
such warnings and engines controls into their systems. EPA has received evidence that these
_ 1 "7 1 Q
engines, when in actual use, have operated almost exclusively with proper DEF. ' Navistar has
produced no evidence that actual operators of these engines have been operating for any
17 Greuel, Justin. Cummins Memo to Docket. Docket EPA-HQ-OAR-2010-0162.
18 California Air Resources Board. Heavy-Duty Vehicle Selective Catalytic Reduction Technology Field
Evaluation. May 2011. Docket EPA-HQ-OAR-2010-0162. Last viewed on July 29, 2011 at
http://www.arb.ca.gov/msprog/cihd/resources/reports/scrreport.pdf
56
-------
significant amount of time with no SCR controls. Thus, the evidence leads us to believe that the
warnings and engine controls put in place by manufacturers are having the intended effect of
reducing the likelihood that engines equipped with SCR are operated without proper DBF.
Finally, we also note that using the new baseline engine means that lowering DEF costs would
actually increase penalties, which Navistar does not support.
283 DEF and Fuel Prices
For the NPRM, we used a DEF price of $2.99 per gallon for calendar years 2012 and
beyond. That DEF price represented the national average retail pump price of on-highway
DEF.19
EPA did not include additional fuel operating costs as part of the NCP calculation for the
proposal. However, we noted that if we did project an impact on fuel consumption, we would
have used the Annual Energy Outlook (AEO) 2011 fuel price projections adjusted to 2011
dollars.
Summary of Comments
CATF EPA-HQ-OAR-2011-1000-0021-0022
Cummins EPA-HQ-OAR-2011-1000-0015, 0025 & 0048
PACCAR EP A-HQ-O AR-2011-1000-0041
CATF recommended that EPA conduct an assessment of projected fuel and DEF prices.
It recommends using Integer Research as a source for DEF prices and that the fuel prices be
updated to reflect the latest Annual Energy Outlook prices.
Cummins stated the following in its comments:
Since those initial conservative estimates, the view of the DEF market has
changed significantly in that there has been a shift towards more bulk
distribution thereby driving down the average price of DEF. This trend is
expected to continue for several more years. Cummins has combined data
from large fleets, truck stops, and Integer Research to show the relative
use of each distribution method and price information. An extensive set of
data has been gathered by Cummins Filtration which sells packaged and
bulk DEF. These data have been compiled and are shown in the Cummins
comments. The fact that more customers are continuing to convert their
DEF purchases to bulk distribution is supported by the front page article in
Transport Topics titled Truck Stops, Fleets Respond to DEF Demand By
Installing More Bulk Filling Dispensers published April 2, 2012. As
stated in the article, "All the major truck-stop chains have bulk DEF
available, and many small and midsize operators are following suit." An
19 DieselExhuastFluid.com. Last accessed on November 14, 2011 at http://www.dieselexhaustfluid.com/.
57
-------
increasing number of customers are installing bulk DEF sites even if the
customers do not distribute their own fuel. This trend toward bulk
distribution of DEF is further supported by Integer Research's DEF
Tracker monthly price reporting service. According to Figure 2, over
forty percent of the total volume of DEF distributed in North America will
be by bulk distribution channels by 2017. Customers site the ability to
transition to on-site bulk dispensing due to the safe and economical
solutions provided by an increasing number of companies that offer
equipment assistance programs. With the growing number of vehicles
utilizing SCR engines, customers are increasingly moving to bulk
dispensing. This market dynamic is driving the cost of bulk and retail
pump prices lower as shown in Attachment A. The retail pump operators
(truck stops) are challenging other DEF supply formats on cost (bulk and
totes) as a necessity to preserving diesel fuel sales.
Combining the information on DEF distribution method with diesel price
information from http://www.eia.gov/petroleum/gasdiesel and with the
DEF price data provided by Integer Research, which can be found in the
Cummins comments. This shows that a 0.4 DEF/diesel price ratio is a
good forward looking representative value. It incorporates the DEF price
variation associated with the distribution methods along with the
downward price pressure due to market competition. This type of price
projection is typical of a new market where the volume of DEF consumed
is increasing rapidly while the distribution channels mature.
Additionally, we expect that going forward the price of DEF will more
closely track the price of natural gas. Urea, the active component in DEF,
is manufactured from natural gas, and the price of natural gas has been and
is projected to remain lower relative to the price of diesel. This likely
results in the cost ratio of DEF to diesel decreasing even further.
[Cummins comments at 11-13, figures and attachments not reprinted.]
PACCAR commented that based on the European experience with AdBlue, the DEF
equivalent for Europe, it is expected that the DEF price will continue to drop until an equilibrium
price is reached. PACCAR further stated that the Integer presentation dated July 12, 2010,
previously provided to EPA, outlines the downward pricing trend of AdBlue.
Response
For the final rule, we updated the fuel price projections and used the Energy Information
Administration's (EIA) Annual Energy Outlook 2012 (AEO2012) to project fuel prices through
2035 and applied the annual projected price increase found in AEO2012 to project the fuel prices
through 2042.
We also revised the source for current and future DEF prices in the Final Technical
Support Document based on the comments we received. Instead of using a constant DEF price
58
-------
through 2042, we are projecting that the long term DEF prices will trend with the prices of
industrial natural gas, which is used in the production of the ammonia used in urea, at an annual
increase of 1.3 percent based on AEO 2012. For the near term, we used the Integer Research
retail DEF pricing projections for 2012 through 2014, which range between $2.60 and $2.50 per
gallon. Note that our analysis uses an annual average DEF price, which does not consider
seasonal variations in DEF prices, or other short term fluctuations.
We are not using bulk DEF prices, as recommended by some commenters, because this
would be inconsistent with our fuel price, which reflects the retail pump price. Bulk prices of
both fuel and DEF are typically substantially lower than retail prices, although sometimes
reported price differences can be overstated because the specified bulk price excludes any
delivery charges.20 We believe that the majority of operators paying retail price for fuel are
paying retail price for DEF, and the majority of operators paying bulk price for DEF are paying
bulk price for fuel. Thus, using a DEF price based even in part on bulk DEF prices while using a
retail price for fuel would significantly overestimate actual costs.
2.8.4. Discounting Operating Costs
In the NPRM, we evaluated the full lifetime operating costs based on a typical vehicle's
annual mileage over a 30 year period. The annual mileage took into consideration the impact of
vehicle scrappage. We also discounted the operating costs by 7 percent, consistent with past
NCP rulemakings.
Summary of Comments
Cummins EPA-HQ-OAR-2011-1000-0015, 0025 & 0048
Daimler EPA-HQ-OAR-2011-1000-0028, 0043-0045 & 0049
Ford EPA-HQ-OAR-2011-1000-0029
Navistar EPA-HQ-OAR-2011-1000-0027, 0031-0039, 0042 & 0050
Daimler stated that fuel efficiency is relevant only to the extent it impacts the purchase
price that the manufacturer receives and that operating cost savings for the customer after the
sale is irrelevant if there is no related impact on purchase price. It generally estimates that truck
customers are willing to pay for [confidential business information redacted] months of reduced
operating costs when making new truck purchase decisions. Ford stated that purchasers do not
fully value the lifetime operating costs, and suggested that the direct market value of reduce
operating costs is less than one-half of the total.
Cummins supports using full lifetime costs to reflect both reduce operating costs for the
first purchaser as well as better resale value. Cummins stated that engines that use NCPs will
20 For example, if the delivery charge for 200 gallons of DEF was $200, then a bulk price without delivery charge
would understate the actual price of DEF by $1.00 per gallon.
59
-------
have a cost advantage for the entire life of the engine and that not including the entire life cycle
cost could be incentive to manufacturers to use NCPs.
Navistar believes that EPA's proposed 7% discount rate is too low and, thus, improperly
overestimates the NCP cost parameters. Navistar uses a 15% standard rate for discounting future
cash flows for any new product program. The reason is that making investments in differentiated
engineered products is far from a risk-free proposition. Navistar continued to state that there are
huge variables in product programs, including the investment required, the timing of the program
launch, the length of life of the product, market acceptance (volumes and pricing), and product cost.
Each of these factors has a high degree of variability and thus risk.
Navistar also commented that EPA's use of a 30-year truck life and the lifetime mileage
over that 30 years for estimating owner operating costs is inappropriate because EPA's emission
standards are only applicable for the prescribed "useful life" of the engine, and that EPA has no
basis for including DEF-maintenance costs that fall outside the prescribed regulatory life of the
vehicle when the NOx emissions standards no longer apply.
Navistar commented that truck purchasers only consider operating costs that occur in the
first 5 years (or less) of the truck's life. Navistar stated that the purpose of including owner costs
in the NCP formula is to eliminate the impact on purchase decisions "not to penalize
nonconforming manufacturers." It stated that to the extent that operating costs do not impact a
buyer's purchasing decision—i.e., the costs are not considered by the buyer to be relevant and,
thus, do not factor into creating a market disadvantage—those costs then do not serve any NCP
purpose and cannot be included without unlawfully penalizing the nonconforming manufacturer.
Navistar stated that operating costs beyond those five years should be excluded because they
have no impact on a buyer's purchasing decision and cannot create a competitive disadvantage
for a complying manufacturer.
Response
We disagree with commenters arguing that the NCP analysis should not include the
operating costs over the full lifetime of the vehicle. The difference in operating costs due to fuel
consumption and DEF consumption exist throughout the life of a vehicle and are not limited to
the vehicle's regulatory useful life or any other arbitrary time period. Thus, we are considering
the full lifetime operating costs in the final rule because it is essential to take into account the
operating cost difference over the full life of the engine to properly incorporate this market factor
into the NCP calculation. It is also worth noting in response to Navistar's comment that, while
manufacturers are liable for the emission performance throughout the engine's regulatory useful
life, we generally expect the emission controls to continue functioning beyond this period. We
also continue to prohibit operators from tampering with emission controls including urea
inducements.
We agree with commenters that the full amount of future operating costs is discounted by
consumers in purchasing new and used vehicles. Commenters provided a range of suggestions
for discounting operating costs. We believe that a seven percent discount rate appropriately
60
-------
reflects the uncertainty of the lifetime vehicle mileage, fuel price, and DEF price. This value is
consistent with OMB guidance on discounting future costs.21 Of note, the use of a seven percent
discount rate is essentially the same as considering 57 percent of the total fuel cost, similar in
magnitude to the discount rate comments provided by Ford.22 This is shown in 2-5 below, which
compares different methods of discounting future fuel costs.
2-5 Comparison of Discounting Methods
Method
Include the first two years of fuel costs without
calculating net present value
Net present value (7%) of costs occurring within
the first 5 years
Net present value (15%) of lifetime fuel cost
Include the one-half of the undiscounted lifetime
fuel cost
Include the first five years of fuel costs without
calculating net present value
Net present value (7%) of lifetime fuel cost
(EPA Method)
Include total undiscounted lifetime fuel cost
Percent of Total Undiscounted
Lifetime Fuel Costs
25%
30%
36%
50%
53%
57%
100%
We also recognize that not all purchasers will value lifetime operating costs to the same
extent that we estimate. We considered this uncertainty in evaluating how to include operating
costs in our analysis. For example, as noted in Section 2.8.2, we did not use the worst case fuel
consumption rates that some commenters suggested should be incorporated into our cost
analysis. One reason we did not use those suggested rates is because it would have substantially
overestimated the competitive impact for customers who discount lifetime operating costs more
than we do in our analysis. While we do not have sufficient information to precisely analyze the
distribution of how much customers will value changes in fuel consumption, we believe that the
inclusion of full lifetime costs for a 1.9 percent increase discounted by seven percent
appropriately reflects the 90th percentile compliance costs. We note that this is about the same
net result as would have been calculated if we had assumed a fuel consumption impact of four
percent (as PACCAR recommended), but only considered the first two years of operating costs.
Finally, while we disagree with Navistar's comment that we should not include operating
costs that occur after the first five years of an engine's life, we note that using a seven percent
discount rate means that about sixty percent of the NPV of the lifetime operating costs are due to
the costs that occur in the first five years. In other words, excluding operating costs that occur
21 OMB Circular A-4, September 17, 2003.
22 The 57 percent value reflects the increases in fuel prices and decreases in annual mileage accumulation for future
years, as well as the discounting for future costs. For example, this value would mean that an actual 30-year fuel
cost of $ 10,000 would have a net present value of $5,700.
61
-------
after this time would reduce the calculated NPV of operating costs by about 40 percent. As
noted above, when estimating operating costs, we did consider the possibility that some operators
may consider costs this way.
2.9. Other Costs
2.9.1. Warranty and Post-Warranty Repairs
Summary of Comments
Cummins EPA-HQ-OAR-2011-1000-0015, 0025 & 0048
Daimler EPA-HQ-OAR-2011-1000-0028, 0043-0045 & 0049
Ford EPA-HQ-OAR-2011-1000-0029
Mack EPA-HQ-OAR-2011-1000-0024, 0046, 0047 & 0051
Several commenters stated that EPA needs to consider costs for warranty in calculating
the NCP value, though the only details on warranty costs were included in the confidential
information submitted to the Agency. Much of this was related to total warranty costs for SCR
systems rather than incremental warranty costs relative to the upper limit. Several commenters
also commented that we needed to add costs for repairs that happen after the warranty period and
demurrage costs for time that the trucks will be out of service.
Cummins commented that an optimized NCP baseline engine will have higher warranty
and down time cost advantages compared to a compliant engine. First, the addition of the more
expensive hardware discussed by Cummins will be associated with warranty costs. Second, as
the level of NOx is reduced, the NOx OBD requirements become more stringent, which will
likely lead to more OBD-triggered warranty claims. Cummins further stated that it is important
to consider the cost to the customer for the additional down time associated with this additional
warranty, as well as the cost to the consumer for the additional down time associated with this
additional warranty.
Ford comments included the following:
The final NCP amount should also account for onboard diagnostic implications of
certifying to the higher standards. 2013 MY+ California Heavy Duty OBD requirements
(to which most manufacturers certify nationwide) included additive NOx thresholds of
0.2 g/bhphr for both NOx catalyst monitoring and NOx sensor monitoring. Manufacturers
certifying to a 0.2 g/bhphr standard are required to detect malfunctions before emissions
reach 0.4 g/bhphr. Manufacturers certifying to a 0.5 g/bhphr NOx FEL would only be
required to detect malfunctions before emissions reach 0.7 g/bhphr. This is effectively the
difference between a 2 x standard threshold and a 3.5 x standard threshold. The
implications of this would vary depending upon the emissions control and monitoring
strategies for the engine, with the most extreme case being the difference between a
62
-------
compliant engine with SCR aftertreatment and one or more NOx sensors, and an engine
certifying via NCPs with no NOx aftertreatment and no NOx sensors.
Response
We proposed to include additional warranty costs related to incremental hardware costs
and are not changing this analysis for the Final Rule. Commenters focused on the total warranty
costs associated with SCR ignore the substantial warranty costs Navistar is incurring for its non-
SCR engines. To be consistent with our baseline engine, we need to include only those warranty
costs associated with the changes to the baseline engine necessary to reach 0.20 g/hp-hr NOx.
However, based on the comments, we are adding costs for similar repairs that operators will incur
after the warranty period and demurrage costs for time that the trucks will be out of service.
These costs are discussed in more detail in the Final Technical Support Document.
We also agreed with commenters who stated that because OBD systems must detect
emission problems relative to thresholds specified as an addition to the family emission limit (or
compliance level for NCP engines) and notify the operator of concerns, using NCPs to certify
above the standard could result in fewer emission-related repairs. We estimated warranty and
post warranty repair costs of the OBD-related costs for the final rulemaking. It is important to
emphasize that such costs are appropriate for NCP rulemakings that address competitive effects;
however, such costs were not included in our OBD rule because the OBD system is only catching
failure modes for which manufacturers were already responsible.
2.9.2. Research and Development
Summary of Comments
Cummins EPA-HQ-OAR-2011-1000-0015, 0025 & 0048
ICCT EPA-HQ-OAR-2011-1000-0030
Mack EPA-HQ-OAR-2011-1000-0024, 0046, 0047 & 0051
Navistar EPA-HQ-OAR-2011-1000-0027, 0031-0039, 0042 & 0050
ICCT commented that compliant manufacturers have made significant investments over
the last decade, and as a result, are in a position to comply with the emission standards. ICCT
further stated that the current definition of the baseline technology package does not account
fully for these differences in investment in technology, which is critical if NCPs are to serve
deterrents to noncompliance for both this regulation and future rulemakings.
Cummins stated that it invested over $200 million dollars to comply, and that heavy-duty
truck prices increased by $10,000 and medium-duty truck prices increased by $6,000 as the
emissions surcharge for 2010-compliant equipment.
Mack commented that through EPA's invention of a hypothetical 0.5 g/bhp-hr baseline
engine, EPA conveniently kept out of the NCP calculation the hundreds of millions of dollars
spent by SCR manufacturers in researching, engineering, developing and implementing their
SCR technology.
63
-------
Navistar provided a lengthy discussion of its technology path in developing advanced
EGR. Navistar commented that their choice of Advanced EGR was not a gamble but a
significant, capital intensive commitment in response to EPA's 2001 Rule. Navistar stated that it
has devoted tens of thousands of employee hours and has invested approximately $700 million in
the development of its clean-burning, in-cylinder solution. Navistar stated that these costs
include:
$543 million - Engineering/Research & Development. These are costs to design,
develop, test and validate the product and include both engine and vehicle costs.
$32 million - Capital Expenditures. These are costs primarily for property, plant and
equipment used in Navistar facilities required to test and manufacturer the new
product.
$81 million - Supplier Tooling. These costs include unique fixtures in suppliers'
facilities that physically touch Navistar parts in the supplier production process.
$36 million - Other Costs. These include remaining costs associated with the project,
including such costs as manufacturing start-up costs in Navistar's plants as well as field
testing of the product.
Ford commented that investment costs for vehicle manufacturers associated with
packaging SCR aftertreatment systems and the necessary sensors that go with them have been
underestimated in the proposed analysis.
Response
Several commenters provided information about the total research expenditures
associated with meeting the 0.20 g/hp-hr, but did not provide such information relative to
engines at the upper limit. As explained in the Final Technical Support Document, we estimated
R&D costs as a fraction of the total incremental hardware costs (relative to the upper limit)
using the near term, low complexity indirect cost multiplier (ICM) value of 1.15 which includes
a portion (0.02) for research and development costs.
It is important to emphasize that we are not developing costs to reflect a specific
manufacturer, but rather the general difference between compliant engines and those at the upper
limit. We see no evidence that research and development costs for the noncomplying
manufacturer are substantially lower than for complying manufacturers. For example, Navistar
commented that it spent $700 million in the development of their current engines, while
Cummins stated that it had spent $200 million to comply with the 2010 standards. Thus, while
commenters may feel that we have underestimated compliance costs associated with R&D, there
is no evidence that we have underestimated the competitive advantage or disadvantage
associated with R&D costs.
Finally, we note that §1113-87(h) of the NCP regulations contain provisions that allow a
manufacturer that pays NCPs to recover up to 90 percent of the portion of the penalty which
EPA determines to be related to research and development costs. Therefore, R&D costs only
64
-------
have a short term impact on a manufacturer using NCPs because the majority of these costs
would be refunded. While this would not justify knowingly using the wrong R&D costs, it does
provide context for how important this cost component is compared to other costs.
2.9.3. Costs not included
Summary of Comments
Daimler EPA-HQ-OAR-2011-1000-0028, 0043-0045 & 0049
Ford EPA-HQ-OAR-2011-1000-0029
Mack EPA-HQ-OAR-2011-1000-0024, 0046, 0047 & 0051
Ford commented that the final NCP amount should also account for the additional
marketing advantages that a manufacturer of non-SCR engines or vehicles has over
manufacturers of comparable SCR equipped engines or vehicles. Ford further stated that these
include the ability to market vehicles that do not entail the cost and inconvenience associated
with having to regularly replace DEF over the life of the vehicle (likely offset in part by an
increase in fuel consumption). Daimler had similar comments and noted that "so-called
"convenience factors" have been the focus of Navistar marketing and publicity campaigns for
years. It argued that EPA ignored this factor because we had not identified any substantial shift
in market share in favor Navistar. It stated that this lack of market shift could be because
customers may "choose to purchase [SCR engines] because of their superior environmental
performance, and because they prefer the certainty afforded by manufacturers who do have the
technology to meet emission standards over a company that has not developed technology to
meet the standards. Finally, it argued that "if Navistar's non-SCR engines were no longer on the
market, then all trucks would have essentially equal NOx maintenance requirements, and this
would no longer be negative cost factor for DTNA." Neither Ford nor Daimler provided a dollar
value estimate of this advantage.
Daimler also stated that "Navistar's future costs are irrelevant", and that EPA should not
have taken these costs "into account". It argued that "the only costs that are relevant to
protecting the competitive position of compliant manufacturers are the compliant manufacturers'
total costs of compliance."
Mack argued that the compliance costs need to reflect differences in deterioration rates.
It said, a non-SCR engine certified at 0.5 g/bhp-hr, such as those Navistar would produce under
the NCP as proposed, does not require as significant a compliance margin as evidenced by
Navistar's certification levels, which range from 0.423 - 0.473 g/bhp-hr to achieve a 0.5 g/bhp-
hr FEL (i.e., a compliance margin of 5 - 15 percent). In addition Mack stated, Navistar applies a
deterioration factor multiplier of 1.0 for its 0.5-gram engines (i.e., no deterioration at all), while
Mack must use a multiplier of 1.25. Based on Mack's calculations, when the additional
compliance margin necessary to meet a 0.20 g/bhp-hr standard (versus a 0.50 g/bhp-hr standard)
is accounted for, the cost differential between achieving a 0.50 g/bhp-hr standard and a 0.20
g/bhp-hr standard is increased by approximately 53 percent.
65
-------
Response
Our final COC90 costs do not include any amount to address "convenience factors" like
those identified by Ford and Daimler. The regulations require that the NCP be based on
estimated compliance costs rather than perceived value. This is intended to capture marketing
advantage by setting the maximum penalty based on the worst case cost that a complying
manufacturer has to pay to reduce emissions from the upper limit to the standard and the
additional costs its customers pay to operate the compliant engine. Nevertheless, we
acknowledged the commenters' concern in our proposal, stating:
In establishing prior NCP rules, we have frequently made it clear that satisfying the
statutory objective of protecting the complying manufacturer was paramount. The
generic NCP rule established an approach which attempts to remove any competitive
disadvantage to complying manufacturers by assessing a cost to the manufacturer of a
non-complying engine in the form of an NCP, with the expectation that this cost is at
least equivalent to or exceeds the value of the competitive benefit gained by building a
noncomplying engine. Imposing such a cost is a way to level the playing field without
interfering in the actual marketing or pricing of the engines. However, since the issue of
competitive advantage involves many subjective factors, the regulatory structure cannot
by itself ensure that no competitive advantage remains.
In both the Interim and Final Technical Support Documents, we evaluated the market impacts of
the NCP to determine if our estimated costs, which do not include additional factors for
subjective market impacts, were sufficiently large to cover the competitive impacts including the
marketing differences. We concluded that they are. None of the commenters provided any
evidence that they are not. More specifically, we determined that Navistar has not gained market
share as a result of selling engines without SCR, even though it is selling its engines for a lower
price. Thus, if these "convenience factors" are impacting operators purchase decisions at all, the
impact must be small.
With respect to Daimler's assertion that we "ignored" this issue, we note that we
addressed this issue in Section 4.2 of the Interim Technical Support Document, and even
addresses one of Daimler's specific points when we stated that "a trucking company that
promotes itself as being an environmentally responsible company may be willing to pay a
premium to ensure that its trucks have low-emitting engines." On the broader issue, we stated:
EPA recognizes that insufficient information is available to conclusively prove that the
NCP removes all competitive disadvantages to all complying manufacturers. However, it
is still helpful to consider market data to put the amount of the NCPs into context. To do
this, EPA considered the available information about market prices and market share. As
described below, both market prices and market shares support (emphasis added) our
conclusion that the NCPs are large enough to remove the competitive disadvantage to
complying manufacturers. The analyses are presented here for the comparison of engines
equipped with SCR to those that are not equipped with SCR.
Also, while we do not fully agree with other Daimler comment that Navistar's future costs are
irrelevant, we did not use Navistar's future costs to calculate either the proposed or final NCPs.
66
-------
We disagree with Mack that costs need to be adjusted to reflect differences in
deterioration factors. Neither the regulations nor past precedent support any additional
adjustment for deterioration factors , beyond what is already taken into account in our cost
analysis and we do not believe that Mack's argument support such adjustments now. Mack
stated that it recommends increasing costs to reflect differences in deterioration factors (DFs)
because Navistar uses a lower deterioration factor near 0.50 g/hp-hr than Mack does at 0.20
g/hp-hr. However, it is unclear why Mack believes this is relevant. Therefore, we are addressing
this comment in the following three contexts:
1. Impact on compliance costs relative to our baseline SCR engine.
2. Impact on our analysis of alternative methodologies.
3. Impact on Navistar's compliance level and the penalty it would pay.
In the first context, we note that since our final COCgo is not based on Navistar's engine,
the Navistar DF is not directly relevant. Mack could be arguing that Navistar's lower
deterioration factor is intrinsic to its higher emission level. However, the difference in DFs
between Navistar and Mack is more likely to result from the fact that Navistar has not been using
catalytic aftertreatment to control NOx emissions than anything intrinsic to the emission level.
Even if it is not, our analysis bases costs on the end-of useful-life catalyst efficiencies of the
baseline and compliant engines, and thus fully addresses deterioration. In other words, EPA's
cost analysis presumes that the expected technology will meet emission levels at the end of the
useful life of the engine. Thus, cost analysis already includes assumptions regarding catalyst
deterioration that would affect the cost of the compliant technology. Inclusion of an additional
cost for deterioration would effectively double-count these costs. It is also worth noting that
Cummins was the only manufacturer to comment specifically on the hardware costs associated
with our cost scenario (i.e., optimizing engines for 0.50 versus 0.20 g/hp-hr NOx) and it did not
support including any additional costs for deterioration.
A second context to consider is how deterioration is addressed in our analysis of
alternatives. For the first alternative, we looked at the costs associated with an upper limit of 1.2
g/hp-hr NOx and our cost estimates were consistent with compliance costs provided by engine
manufacturers of going from a 1.2 g/hp-hr engine that does not use SCR to a compliant engine
that uses SCR. These manufacturers provided SCR hardware costs for the systems they are
actually producing, which necessarily include any costs associated with deterioration.
The only alternative scenario in which Mack's comment could apply would be the second
alternative, in which we estimated costs relative to an EGR engine similar to Navistar's current
engines. While, as discussed above, we believe we fully accounted for deterioration in this
alternative, even if we did underestimate the costs associated with deterioration of engines like
Navistar's engines, it would not have changed our conclusion. We estimated that the
costs for this alternative would have been $994, which is $2,781 less than our final
($3,775). Thus, even if omitted a deterioration-cost of $2,000 per engine (which would be much
more than any such error could reasonably be expected to be), we still would have concluded
that the optimized SCR engine is the more appropriate baseline for our cost analysis.
With respect to the third context, we note that the actual amount of the NCP paid by
noncomplying manufacturers is calculated from the compliance level, which includes a
67
-------
deterioration factor (see §86.1102-87). NCP cost parameters are based on the levels of emission
standards (or effective standards such as the 1.2 g/hp-hr phase-in NOx level or the 0.50 g/hp-hr
PEL cap), which is consistent with the regulatory provisions that treat the compliance level as the
in-use standard for NCP engines. Thus, Navistar would have to include an appropriate
deterioration factor in its application for certification to insure that its engines meet its NCP
compliance level at the end of their useful lives. Any question regarding the appropriateness of
Navistar's DF would be a certification issue that is not within the scope of this NCP rulemaking.
2.10. Comparison of Commenters' and EPA's Estimates of Total
Compliance Costs and Penalties
Several commenters recommended specific values for COCgo (in some cases stated as the
recommended penalty). These recommendations, which each differ from our final COCgo value,
are summarized in this section. Note that in many cases we cannot respond to these comments
in full detail because the underlying cost analyses shared with EPA are considered by the
commenters to be confidential business information.
Summary of Comments
CATF EPA-HQ-OAR-2011-1000-0021-0022
Cummins EPA-HQ-OAR-2011-1000-0015, 0025 & 0048
Daimler EPA-HQ-OAR-2011-1000-0028, 0043-0045 & 0049
Mack EPA-HQ-OAR-2011-1000-0024, 0046, 0047 & 0051
Navistar EPA-HQ-OAR-2011-1000-0027, 0031-0039, 0042 & 0050 PACCAR
EP A-HQ-O AR-2011-1000-0041
CATF submitted a memorandum that evaluates potential NCP values that would
potentially derive from a "high engine-out" nonconformance path, using best available recent
projections of the future relationship between DEF and diesel fuel prices. CATF believes this
analysis indicates that a year 1 NCP value in the range of $4,000-$5,000 might be warranted to
preclude backsliding by current conforming engine manufacturers.
Cummins also commented that NCPs that are too low create a much greater issue with
regard to compliance among SCR manufacturers than between EGR and SCR manufacturers.
Cummins continued to say that since NCPs are available to all manufacturers once enacted, and
since the cost of the proposed NCPs are much lower than the economic value that can be created
by paying them and optimizing engines at higher, noncompliant NOx levels, SCR manufacturers
that are otherwise capable of producing compliant engines will be driven by competitive market
forces to pay NCPs as a license to produce higher emitting engines. Cummins also commented
that a correct analysis would recognize the first cost and operating cost benefits of operating at
higher NOx levels up to 0.50 g/hp-hr and would incorporate a more appropriate DEF/diesel price
ratio which would result in a much higher NCP level of $8,100 for a heavy heavy-duty diesel
engine.
68
-------
PACCAR also stated that the proposed NCP level is so low that it creates an additional
competitive disadvantage for manufacturers that have invested considerable time and resources
in developing emission control technology that complies with the 2010 NOx standard.
PACCAR commented that in order to eliminate the competitive advantage Navistar gains under
the current NCP, EPA should add the value of "Navistar1 s fuel economy advantage" to the SCR
hardware, R&D and warranty component costs, and increase the NCP for 2012 to a minimum of
$14,378.
Mack provided confidential business information summarizing its costs associated
with developing, producing, operating and maintaining an engine that complies with a 0.20
g/bhp-hr standard compared with an engine that complies with a 1.2 g/bhp-hr NOx limit (the
effective standard during the 2007-2009 phase-in period). It stated the costs ranges from
$14,520 to $21,390, depending on the DEF dosing level adopted. Mack's cost estimates did
not include any fuel savings associated with going from 1.2 to 0.20 g/hp-hr, because it
believes EPA's past practice was to neglect any side benefits associated with compliance.
Mack stated that if the total cost represented 1.2 g/hp-hr NOx and was adjusted to represent
the cost difference between meeting a 0.5 g/bhp-hr standard and a 0.20 g/bhp-hr standard, it
would result in COC90 of $4,356 to $6,417 per engine (which is 30 percent of the total SCR
cost). However, it argued that a higher value is needed to reflect differences in deterioration
rates. It said, a non-SCR engine certified at 0.5 g/bhp-hr, such as those Navistar would
produce under the NCP as proposed, does not require as significant a compliance margin as
evidenced by Navistar's certification levels, which range from 0.423 - 0.473 g/bhp-hr to
achieve a 0.5 g/bhp-hr PEL (i.e., a compliance margin of 5 - 15 percent). In addition Mack
stated, Navistar applies a deterioration factor multiplier of 1.0 for its 0.5-gram engines (i.e.,
no deterioration at all), while Mack must use a multiplier of 1.25. Based on Mack's
calculations, when the additional compliance margin necessary to meet a 0.20 g/bhp-hr
standard (versus a 0.5 g/bhp-hr standard) is accounted for, the cost differential between
achieving a 0.5 g/bhp-hr standard and a 0.20 g/bhp-hr standard is increased by approximately
53 percent. Mack commented that the base range of costs of $4,356 to $6,417 per engine,
therefore, must be adjusted to account for these factors. Once so adjusted, Mack's estimated
costs of compliance range from $6,655 to $9,804 per engine. Note: Mack also submitted
comments more than three months after the close of the comment period that argued for a
penalty of approximately $19,000 per engine based on fundamentally redesigned SCR
engines.
Daimler provided a similar analysis of the costs it incurred to meet the 2010 NOx
standard. However, it entire analysis was identified as confidential business information.
69
-------
Response
We considered each of the specific values for COCgo (or recommended penalty)
presented by commenters. Where it is possible, we identify the primary reasons why the
recommended value differs from our estimate. Note that this summary of reasons is not intended
to be exhaustive, but rather is intended to identify the most significant difference between the
commenters' analyses and ours. We also identify other sections in this chapter where these
differences are discussed in more detail.
CATF recommended a COC90 of $4,000-$5,000 based on using an optimized 0.50 g/hp-
hr SCR engine as the baseline (similar to EPA's final baseline engine). This value is slightly
higher than our final COCgo value. While it did not provide full details of its calculations, the
difference appears to be due primarily to its use of a slightly lower DEF price (about 10 percent
less than our DEF price) and a slightly larger fuel consumption impact. See Section 2.8.3 for a
discussion of why our projected DEF price is more appropriate than the value assumed by
CATF.
The baseline engine for Cummins' recommended COCgo value of $8,100 is very similar
to our final baseline engine. While the details of Cummins calculations are considered to be
confidential business information, we can say that the difference between our COCgo and
Cummins' COC90 is due primarily to Cummins' assumption that the price of DEF ($/gal) will be
only 40 percent of the cost of fuel, while we project that the DEF price will be 65 to 70 percent
of the price of fuel. Its confidential estimates for other costs (such as hardware, warranty, and
fuel consumption) also differ from our estimates, but these differences are less significant than
the different DEF price. See Section 2.8.3 for a discussion of why our projected DEF price is
more appropriate than the value assumed by Cummins.
PACCAR's baseline engine is also very similar to our baseline engine. However,
PACCAR believes the fuel consumption cost should be $20,108. It calculated essentially the
same DEF consumption as our final estimate, and assumed all other costs are the same as our
proposed COCgo. See Section 2.8.2 for a discussion of why we believe PACCAR overestimates
the COCgo fuel consumption impact. PACCAR argued that the penalty should also include the
value of "Navistar's fuel economy advantage", but we see no evidence that Navistar has any fuel
economy advantage over compliant engines.
Mack did not use the same type of baseline engine for its analysis as we did. Its baseline
engine is a 2009 non-SCR engine that has NOx emissions at 1.2 g/hp-hr, which is above the
upper limit. It attempted to address this difference by scaling its costs to be proportional to the
difference in emissions. As described in Appendix B of the Final Technical Support Document,
we evaluated this methodology using our own independent cost inputs. Our analysis using this
methodology results in a COC90 that is slightly less than our final COC90 based on our SCR-
baseline methodology, which is much less than Mack's estimate. The primary reasons for the
difference are that: (1) we account for the fuel benefit associated with adding SCR, while Mack
does not; and (2) we do not agree with Mack's adjustment for deterioration factors. We estimate
the NPV of the lifetime fuel benefit of adding SCR to about $10,000. When scaled by 30 percent
to represent the upper limit, this benefit reduces Mack's COCgo values by about $3,000. So our
final COCgo value is consistent with the range recommended by Mack, as long as Mack's
70
-------
estimates are corrected for the fuel benefit. Moreover, our final COCgo value is actually higher
than the range recommended by Mack without its deterioration factor adjustment when those
estimates are corrected for the fuel benefit. See Section 2.8.1 for our discussion of why it is not
appropriate to neglect fuel savings in this NCP rule, and Section 2.9.3 for a discussion of why
Mack's deterioration adjustment is not appropriate. Regarding, Mack's very late comments
pertaining to fundamentally redesigned SCR engines, these comments came in too late for EPA
to include in our cost analysis or for EPA to provide detailed analysis. However, see Section B.3
of the Final Technical Support Document for a discussion of why we are not basing our costs on
fundamentally redesigned SCR engines, such as those discussed in Mack's late comments.
While we cannot respond in detail to Daimler's analysis because Daimler considers it to
be confidential business information, we can note that Daimler's costs were relative to a model
year 2009 baseline engine with NOx emissions near 1.2 g/hp-hr rather than the a baseline engine
at the upper limit. Daimler does not scale its estimate of costs or resulting COC90 to take into
account the upper limit of 0.50 g/hp-hr. Its costs also excluded fuel savings associated with
adding SCR to model year 2009 engine. See Sections 2.5 and 2.8.1 for our discussions of why
these assumptions are not appropriate.
2-6 Summary of Why EPA's Final COC90 Differs from Commenter Recommendations
Commenter
CATF
Cummins
Daimler
Mack
PACCAR
Recommended
COC90/Penalty
$4,000 to $5,000
$8,100
At least $19,000
$6,655 to $9,804
At least $14,378
Primary Difference from EPA
Analysis
Uses lower DBF price and larger
increase in fuel consumption.
Uses lower DBF price.
Costs not calculated relative to the
upper limit and fuel savings are
neglected.
Fuel savings are neglected, and
costs are calculated using an
alternative methodology.
Assumes larger increase in fuel
consumption.
Related Sections
2.8.2
2.8.3
2.8.3
2.5
2.8.1
2.8.1
2.9.3
2.8.2
71
-------
3. Regulatory Issues
3.1. Base Year for Calculating NCPs
As required by the Clean Air Act, the existing regulations include a formula that increases
the penalty rates with each new model year. We proposed to apply this formula to the NCPs
beginning with the 2013 model year by setting the 2012 model year as year number one. This
would be consistent with the pre-existing regulatory text in §1113-87(a)(4) which states that
"n=l for the first year that the NCP is available" (emphasis added).
Summary of Comments
CATF EPA-HQ-OAR-2011-1000-0021-0022
Cummins EPA-HQ-OAR-2011-1000-0015, 0025 & 0048
Daimler EPA-HQ-OAR-2011-1000-0028, 0043-0045 & 0049
EOF EPA-HQ-OAR-2011-1000-0026
Mack EPA-HQ-OAR-2011-1000-0024, 0046, 0047 & 0051
Navistar EPA-HQ-OAR-2011-1000-0027, 0031-0039, 0042 & 0050
PACCAR EP A-HQ-O AR-2011-1000-0041
Cummins, Daimler, Mack, and PACCAR each commented that EPA incorrectly selected
2012 as the proposed "first year" for purposes of the NCP annual escalator adjustment factor.
Cummins commented that EPA provided "scant justification for this proposal." It argued that
"setting the 'first year' for purposes of penalty escalation based on the time that one particular
manufacturer runs out of credits is wholly at odds with the purpose of the statutory provision
requiring NCPs to escalate over time: to increase the incentive to come into compliance quickly.
See 42 U.S.C. § 7525(g)(3)(D)." It also stated that "the availability of credits should not exempt
any manufacturer from the escalation of NCPs required by Congress."
Mack commented that EPA's "perception of when NCPs might be needed is wholly
irrelevant to the separate question of when an escalation factor should be initiated." It also stated
that "since NCPs are designed to protect both the environment and the competitive interests of
compliant manufacturers, the relevant point at which escalation factors must apply is the point at
which the industry must comply (i.e., the year the standard takes effect). "
Daimler commented that the only reason that EPA finds itself setting NCPs in 2012 is
because Navistar has used credits to sell its engines for the past two years, and that compliant
manufacturers have been meeting the NOx standard since 2010, and it is the competitive
disadvantage to compliant manufacturers that EPA must remove. It stated that where NCPs are
made available beginning one or more model years after the effective date of the pertinent
emission standard, either the "n" value or the initial cost level must be adjusted upward to reflect
that fact and that failure to make such an adjustment would breach EPA's statutory obligation to
ensure that penalties incentivize compliance and remove "any competitive disadvantage" to
compliant manufacturers.
72
-------
EDF believes EPA was "misguided in choosing model year 2012 as the first year for
NCPs," noting that "every year the NCP does not escalate, there is less incentive for
nonconforming manufacturers to conform." EDF opposes rewarding the nonconforming
manufacturer by allowing them two model years without an increase in penalties. It stated that
"EPA should have established the NCP by 2010 so that the manufacturer would have had to
choose then between using credits or paying the NCP," but that using model year 2010 as the
first year for the escalator would "prevent conforming manufacturers from suffering a
competitive advantage (sic), to give the nonconforming manufacturer greater incentive to
conform, and to disincentivize conforming manufacturers from backsliding."
CATF stated that EPA's selection of 2012 as the first year of the NCP escalation clause is
unreasonable and unlawful, rewards noncompliance, and is inconsistent with the Clean Air Act
and EPA precedent. CATF noted that CAA section 206(g)(3)(D) provides that penalties "shall
be increased periodically" and that EPA has traditionally structured NCPs as escalating
beginning with the effective date of the new emission standard. CATF noted that the purpose of
the escalator clause is to minimize the production of noncompliant engines.
Navistar commented that choosing model year 2012 as the base year was correct. It
argued that "selecting an earlier model year would run counter to EPA's longstanding
interpretation of when NCPs (and the corresponding annual adjustment) are appropriate." To
support this, it cited EPA's statement in the proposal for the Phase I rule (50 Fed. Reg. at 9212)
that "annual adjustment factors would be used to increase the amount of the nonconformance
penalty from year to year, beginning with the second year that the NCP is available for a
particular standard." Navistar also stated that the SCR engine manufacturers' argument is flawed
because the 0.20 g NOx standard actually was in effect in model year 2007 and there was no
standard change in 2010 but a lowering of a cap under the phase-in process. It also argued that
selecting an earlier model year would run counter to Congressional intent and there could be no
other purpose for moving "year one" earlier other than to unlawfully and unfairly penalize
Navistar. Finally, Navistar commented that we "must make NCPs available retroactively" if we
set 2010 as year one.
73
-------
Response
Many of the commenters addressing the appropriate base year for NCP inferred what
Congress' intent was with respect to this issue. However, we believe that this is not appropriate.
Congress did not specifically address circumstances in which the NCPs are established after a
standard has already gone into effect. Nevertheless, Congress did establish two very clear
requirements that are relevant here. First, Congress requires the NCP to always be high enough
to protect the complying manufacturers. As is described elsewhere, the final NCPs (that are
based on 2012 compliance costs) are high enough to meet this requirement without setting 2010
as the base year. Second, Congress requires the NCP to increase periodically. The existing
regulations fulfill this requirement without regard to which base year is used. However, since
the purpose of this escalator is to provide an incentive for manufacturers who use NCPs for more
than one model year to achieve compliance quickly rather than continuing to use NCPs for
multiple model years, using 2012 as the base year is consistent with the requirements in the
Clean Air Act.
We also note that the existing regulatory text states that year one is the first year that NCPs
are available, which in this case is 2012. Thus, using 2012 as the base year is also consistent
with the existing regulations.
Setting 2012 as the base year is consistent with both the Clean Air Act and the existing
regulations, and we believe that there would need to be a significant policy reason to finalizing
2010 as the base year. The commenters have not provided such a reason. On the contrary, using
the first year of NCP availability as the first year for the escalator calculation, the NCPs for the
first year of availability are sufficient to protect the complying manufacturers while allowing
Navistar to continue making engines, as Congress intended. Adding an extra penalty equivalent
to two years of escalation seems contrary to the intent for this escalation, which is to encourage
manufacturers who are using NCPs to come into compliance. As no manufacturer has had
access to NCPs prior to 2012, requiring an escalator for the two previous years is not consistent
with the purpose of the escalator. Moreover, EPA disagrees with commenters who apparently
believe that Navistar's use of emission credits in model years 2010 and 2011 amounted to
noncompliance. This is false. Emission credits are a completely valid and environmentally
neutral (if not beneficial) method for manufacturers to reduce emissions early and use credits to
have higher emissions later. Commenters who wish to penalize Navistar for using a credit
program that has been part of the regulations since the NOx standard was promulgated, and that
can be credited with achieving NOx reductions earlier than otherwise required, are mistaken.
74
-------
3.2. Retaining Emission Credit While Using NCPs
Summary of Comments
Mack EPA-HQ-OAR-2011-1000-0024, 0046, 0047 & 0051
Mack commented that since we justified our adoption of NCPs based on concerns that
Navistar will exhaust emissions credits, we should require that Navistar exhaust all available
emissions credits before it is allowed to utilize NCPs, noting that we have previously stated that
"NCPs have always been intended for manufacturers that cannot meet an emission standard for
technological reasons rather than manufacturers choosing not to comply." Mack argues that this
logic means that "EPA should not make NCPs available to any manufacturer until such time as
the Agency has confirmed the manufacturer has exhausted all credits available for use in the
United States."
Response
We neither proposed nor requested comment on any requirement to exhaust emission
credits before using NCPs. We are not adding such a restriction to the regulations in this Final
Rule because we do not believe it is necessary as long as the penalty is high enough to remove
all competitive disadvantages for complying manufacturers and as long as there is a
technological laggard. We note that Navistar would have already exhausted its emission credits
had interim NCPs not been available, and that Navistar has been using its remaining credits to
certify its engines for use in California. There does not appear to be any environmental benefit
to including the requirement to exhaust credits, since in either case the engines would be emitting
above the level of the standard.
3.3. Prohibition against generating emission credits from NCP
engines
Summary of Comments
Cummins EPA-HQ-OAR-2011-1000-0015, 0025 & 0048
Navistar EPA-HQ-OAR-2011-1000-0027, 0031-0039, 0042 & 0050
Cummins commented that disallowing the ability to generate greenhouse gas credits for
engines that use NCPs is not a sufficient deterrent to prevent otherwise compliant engine
manufacturers from using NCPs to economic advantage.
Navistar stated that it "EPA overstates the issue and thus proposes a limit that is too
broad." However Navistar "agrees that SCR engine makers should not be able to "game" the
system to generate CO2 credits in the manner described by EPA." It suggested that we "should
clarify that engine manufacturers that implement approved methods for generating CO2
75
-------
credits—i.e., actions other than simply dialing up NOx and paying an NCP—are not precluded
from generating CO2 credits for future use."
Response
We are not relying on this prohibition to prevent compliant engine manufacturers from
using NCPs to gain an economic advantage. We believe our final regulations set the penalties
high enough to prevent this.
With respect to Navistar's comment, we note that the new 86.1105-87(j)(2) states only
that:
(j)(2) Manufacturers may not generate emission credits for any pollutant from engines for
which the manufacturer pays an NCP for the NOx standard identified in paragraph (j)(l) of
this section.
Thus a manufacturer is precluded from generating credits only from those engines for which it
pays the NCP. This rule does not preclude manufacturers from generating credits from any
engines for which they do not pay the NCP. Nor does this rule preclude manufacturers from
generating vehicle credits (i.e., non-engine credits) from any vehicles, even if the vehicles
include engines for which NCPs are paid.
3.4. NCP Formula
Summary of Comments
ICCT EPA-HQ-OAR-2011-1000-0030
The ICCT believes that the proposed methodology rewards noncompliant manufacturers
who have not made sufficient investments in technology. ICCT proposes an alternative
methodology wherein two sets of NCPs are developed based on the particular technology
pathway the manufacturer has chosen. Using this approach, the Agency would develop one set of
NCPs for noncompliant manufacturers using only EGR and another set of NCPs for potential
noncompliant manufacturers using SCR+EGR. Then, the appropriate penalties can be applied
based on the manufacturer's specific technology platform. This system avoids having to set
penalty levels that do not reflect the technology choices of a given manufacturer. Under the
current cost estimation methodology, choosing a baseline SCR+EGR engine to determine the
NCPs for a manufacturer using EGR discounts the years of substantial research and development
costs that the compliant manufacturers have devoted to achieving the current standard.
76
-------
Response
The generic Phase INCP rule established that a single NCP curve should apply for all
engines, and that the curve should be based on the full distribution of compliance costs for all
engines. We did not propose to revise this approach. While we understand the ICCT's concern,
we do not believe it would be appropriate to adopt such a significant change without full notice
and opportunity for public comment. Moreover, we believe that this recommendation was made
to address the ICCT's broader goal of ensuring that there is no incentive for SCR engine
manufacturers to reoptimize their engines for higher NOx emissions. We are confident that our
final NCPs will achieve this broader goal. We believe the best approach is to have a single NCP
level that reasonably represents the 90th percentile costs for manufacturers.
3.5. Significant figures of the upper limit.
Summary of Comments
Cummins EPA-HQ-OAR-2011-1000-0015, 0025 & 0048
Cummins noted that the preamble and proposed regulatory language for the upper limit
vary by one significant digit, and that the regulatory upper limit should contain the extra
significant digit.
Response
We agree and the final regulatory text has been corrected.
3.6. Dollar values
Summary of Comments
Anonymous public commentEPA-HQ-OAR-2011-1000-0013
We received an anonymous comment asking the reason some of the regulatory values are
expressed in monetary amounts from another decade (e.g., "December 1991 dollars" or
"December 1984 dollars").
Response
These references in the regulations identify the basis of costs listed in the regulations for
purposes of adjusting for inflation in later years. For example, the statement in the regulations
that the new NCP parameters for the 2010 NOx standard are "expressed in December 2011
dollars" means that these values are consistent with the Consumer Price Index (CPI) for the
month of December in 2011.
77
-------
4. Other Issues
4.1. Sunsetting NCPs
Summary of Comments
NJDEP EPA-HQ-OAR-2011-1000-0016
The State of New Jersey commented that there should be a definitive deadline to the
length of time nonconformance penalties will be allowed as a substitute for a complying engine.
Response
Section 206(g)(l) of the Clean Air Act states "a certificate of conformity shall be issued
... if such manufacturer pays a nonconformance penalty as provided under regulations ..." The
statute does not refer to EPA setting end dates for NCPs and EPA's has not included end dates
in its previous generic or specific NCP regulations. However, we believe that the yearly increase
in the penalty rates, which is authorized by the Act, will effectively accomplish the same goal by
making the penalty much higher than the cost of compliance. Nevertheless, it is possible that we
could determine in the future that NCPs are no longer needed and are being used inappropriately,
in which case we could effectively end the NCPs by revising the upper limit to be equal to 0.20
g/hp-hr through a new rulemaking.
4.2. Alternative Penalties
EPA requested comment on whether we could or should also include a non-monetary
value as an option in the definition of the noncompliance penalty. Specifically, we asked if there
should be an option where the penalty could be defined as the amount of NOx emission
reductions that would not be achieved by the engine compared to the applicable standard.
Summary of Comments
Cummins EPA-HQ-OAR-2011-1000-0015, 0025 & 0048
NESC AUM EP A-HQ-O AR-2011-1000-0018-0019
NJDEP EPA-HQ-OAR-2011-1000-0016
NESCAUM and NJDEP support the alternative penalty approach discussed in the
NPRM. NESCAUM stated that this "would provide a means to recover the environmental loss,
calculated as the total excess NOx emissions expected from the non-conformance engines over
their lifetimes, through a specific offset plan developed by the engine manufacturer and at a cost
equivalent to or greater than what the nonconformance penalty would be under the conventional
approach outlined in the proposed rule." NJDEP stated that due to "the uncertainty in
determining actual tons of NOx that should be offset, the level of NOx to be mitigated should be
78
-------
set above the excess tons of expected NOx emissions that will occur from the engines certified
using NCPs."
Navistar stated that EPA's past practice shows that a non-monetary NCP is a desirable,
equitable option and that an alternative penalty would ensure that excess NOx emitted by
nonconforming engines are off-set. It also noted that because EPA would require that the cost of
achieving these reductions would be at least as great as the dollar amount of the NCP, any
purported competitive disadvantage to complying manufacturers would be eliminated to the
same extent as simply paying a dollar amount. Navistar stated that "under EPA's proposal, it
makes most sense to continue to use the existing penalty formula, but then to offer an option that
the penalty be paid in terms of money spent on a "green project" rather than money spent by
writing a check" and that this approach "is well suited for ensuring that the CAA § 206(g)(3)
requirement is met and that the penalty will remove any competitive disadvantage. " Navistar
recommends that NCPs continue to be defined in monetary terms, but that manufacturers
choosing to pay penalties have the added option of instead spending the "penalty money" on
green projects. Finally, Navistar stated that "EPA does not need to fully establish the alternative
penalty/green project regulatory structure in the current rulemaking" but could promulgate
regulations now that allow for/create the alternative penalty payment option later.
Cummins stated that it would be inappropriate to allow for nonmonetary NCPs because
"Section 206(g) calls for manufactures to "pay" a nonconformance "penalty" in "amounts" set
by EPA." It cited the legislative history of the NCP provision to support its belief that Congress
intended the penalties to be monetary:
"The provision requires that the nonconforming technology penalty to be set at a level
which will eliminate the competitive advantage, if any, for the manufacturer of a
nonconforming vehicle or engine. Thus, its calculation would include such items as the
actual cost of compliance for complying vehicles, the capital costs foregone as a result of
noncompliance, the market value of any fuel economy gains made by non-complying
vehicles compared to complying vehicles...." House Conf. Rept. No. 95-564 at 163
(95th Cong., 2d Sess. 1977), 2 U.S.C.C.A.N. 1544 (1977) (emphasis added).
Response
While we recognize the potential environmental benefits of allowing manufacturers to
offset the emission impacts of NCP engines instead of paying a cash penalty, we are not
finalizing such an approach at this time. We have concerns regarding the complexity of such a
program and the enforceability (in terms of monitoring compliance) of this approach. Such an
approach would be a substantial shift from the current NCP program and would need
considerable review regarding its parameters and compliance mechanisms. EPA also notes the
comments of Cummins, regarding the intent of Congress regarding NCPs and possible
inconsistencies of such a program with Congressional intent, though we do not conclude that
such a program would be necessarily inappropriate.
79
-------
4.3. Regional Impacts of NCPs
Summary of Comments
NESC AUM EP A-HQ-O AR-2011-1000-0018-0019
NJDEP EP A-HQ-O AR-2011 -1000-0016
NESC AUM estimates as a result of the sale and use of these higher emitting engines that
NOx emissions within our eight-state region will increase by as much as 2590 tons in 2012, and
by 45,500 tons over the useful life of these model year 2012 engines as compared with the use of
fully compliant engines. The NESCAUM states are concerned about the effect that this
unanticipated increase in NOx emissions will have on attainment and maintenance of the ozone
standard.
NJDEP stated that operation of nonconforming engines could increase in NOx emissions
throughout the northeast region because states upwind of New Jersey have a significant impact
on New Jersey's ozone levels. It argued that the non-compliant engine manufacturer should also
offset NOx emission increases in those upwind states.
Response
There is little question that the promulgation of NCPs could have a negative impact in
terms of the emissions of new heavy heavy-duty diesel engines in the near future. However,
Congress clearly intended that NCPs be made available. The language and history of this
provision indicate that Congress effectively determined that requiring EPA to issue standards for
heavy-duty vehicles and engines that are technology forcing in nature, with NCPs acting in part
as a safety valve preventing a manufacturer from being forced from the market, serves the public
interest in the long term and outweighs any short-term harm to the environment caused by
allowing NCPs.
4.4. Comments about SCR unrelated to costs
Summary of Comments
Navistar EP A-HQ-O AR-2011-1000-0027, 0031-0039, 0042 & 0050
Navistar also included comments arguing against the use of SCR by its competitors. In
particular, it stated that "EPA unlawfully eliminated the inconvenience and cost of DEF-
maintenance, by authorizing SCR-equipped vehicles to run for thousands of miles (and, in
practice, indefinitely) with the SCR system turned off."
Response
These comments are not relevant to the current proceeding. We based our NCP rule on
the statutory and regulatory provisions for developing NCPs and the reasonable cost estimates
80
-------
we developed for the rule. Other tangential issues regarding the heavy duty diesel engine
manufacturing industry were not relevant to our actions during the proceeding.
4.5. Timing of NCP Rulemaking
Summary of Comments
Daimler EPA-HQ-OAR-2011-1000-0043 & 0045
EOF EPA-HQ-OAR-2011-1000-0026
Mack EPA-HQ-OAR-2011-1000-0046 & 0047
EDF commented that EPA's failure to timely set NCPs before 2010 contradicts the
agency's regulatory mandate to set such penalties "when any new or revised emission standard is
more stringent than the previous standard" and frustrates the Act's directive mandating that
NCPs "shall remove any competitive disadvantage to manufacturers whose engines or vehicles
achieve the required degree of emission reduction." 42 U.S.C. § 7525(g)(3)(E).
Mack stated in it hearing testimony that if Navistar was in danger of running out of
emission credits before achieving the standard, it should have requested NCPs soon enough to
allow for a proper process. It added that the fact that Navistar did not "should not work to its
advantage.
Daimler commented that all previous NCP rules were "conducted before or at the same
time that the emission standards were taking effect."
Response
We notified manufacturers in 2010 that we had found that the new NOx standard required
substantial work.23 When making that determination, EPA noted that the reason that we were
not establishing NCPs at that time was because we had not determined that a technological
laggard was likely to develop. Had we known before 2010 that Navistar would run out of credits
in 2012 and that it would have not yet brought its NOx levels down to 0.20 g/hp-hr, we would
have established NCPs. It was not apparent that there would be a need for NCPs until after, the
implementation of a standard. Daimler is correct that the timing of this rule is different than for
other NCP rules. However, the appropriateness of the NCP rule is determined by its compliance
with the statute, not its timing. The fact that Navistar did not need NCPs until after the initial
implementation of the standard does not change our obligation under the Clean Air Act.
23 "Nonconformance Penalties for Heavy-Duty Diesel Engines in 2010 Model Year", Letter from Karl J. Simon,
Director, EPA Compliance and Innovative Strategies Division, February 22, 2010.
81
-------
It is also important to note that although many commenters seem to believe that the NOx
standard was new in 2010, it was actually first applicable to model year 2007 engines. There is
no basis for treating the phase-in provisions that applied in model year 2007 through 2009
differently than the flexibility of the emission credit program that Navistar has used since then to
certify engines near 0.50 g/hp-hr NOx.
With respect to Mack's comment, while it may have been better had Navistar notified us
earlier of its need for NCPs, we do not agree that its failure to notify us sooner has worked to its
benefit, particularly not in with regard to this final rule completed after notice and comment.
Mack's comment may have been addressing the Interim Final Rule, rather than the proposal.
Indeed, given the events that have occurred since the publication of the proposal and interim
final rule, it is likely that an earlier proposal and final rule promulgating NCPs would have been
more advantageous to Navistar. It is also likely that, had we finalized an NCP rule last year, the
final rule may have been based on the fuel and DEF prices similar to those used for the proposal,
which were appropriate at that time, and would likely have resulted in lower penalties.
4.6. Late Comments
Daimler EPA-HQ-OAR-2011-1000-0043 & 0045
Mack EPA-HQ-OAR-2011-1000-0046 & 0047
Navistar EPA-HQ-OAR-2011-1000-0042
PACCAR EPA-HQ-OAR-2011-1000-0041
Daimler, Mack, Navistar, and PACCAR submitted these comments after the close of the
comment period. In most cases, these were supplemental comments. We considered these late
comments to the extent practicable and many details of these late comments are addressed in the
respective other sections of this document. For example, PACCAR's comments were received
shortly after the close of the comment period and have been addressed in this document.
Navistar submitted comments in May to rebut arguments raised by other commenters, and these
comments are also addressed to some extent. On the other hand, comments from Daimler and
Mack that were received in July (more than three months after the close of the comment period
and after we had completed our analysis of costs) are generally not explicitly addressed in this
document. Nevertheless, we did summarily review these very late comments and determined
that they did not contain any new information that would have changed our conclusions or our
final NCP level.
82
-------
APPENDIX: Section 206(g) of the Clean Air Act
(g) Nonconformance penalty
(1) In the case of any class or category of heavy-duty vehicles or engines to which a standard
promulgated under section 7521 (a) of this title applies, except as provided in paragraph (2), a
certificate of conformity shall be issued under subsection (a) of this section and shall not be
suspended or revoked under subsection (b) of this section for such vehicles or engines
manufactured by a manufacturer notwithstanding the failure of such vehicles or engines to meet
such standard if such manufacturer pays a nonconformance penalty as provided under
regulations promulgated by the Administrator after notice and opportunity for public hearing. In
the case of motorcycles to which such a standard applies, such a certificate may be issued
notwithstanding such failure if the manufacturer pays such a penalty.
(2) No certificate of conformity may be issued under paragraph (1) with respect to any class or
category of vehicle or engine if the degree by which the manufacturer fails to meet any standard
promulgated under section 7521 (a) of this title with respect to such class or category exceeds the
percentage determined under regulations promulgated by the Administrator to be practicable.
Such regulations shall require such testing of vehicles or engines being produced as may be
necessary to determine the percentage of the classes or categories of vehicles or engines which
are not in compliance with the regulations with respect to which a certificate of conformity was
issued and shall be promulgated not later than one year after August 7, 1977.
(3) The regulations promulgated under paragraph (1) shall, not later than one year after August 7,
1977, provide for nonconformance penalties in amounts determined under a formula established
by the Administrator. Such penalties under such formula—
(A) may vary from pollutant-to-pollutant;
(B) may vary by class or category or vehicle or engine;
(C) shall take into account the extent to which actual emissions of any air pollutant
exceed allowable emissions under the standards promulgated under section 7521 of this
title;
(D) shall be increased periodically in order to create incentives for the development of
production vehicles or engines which achieve the required degree of emission reduction;
and
(E) shall remove any competitive disadvantage to manufacturers whose engines or
vehicles achieve the required degree of emission reduction (including any such
disadvantage arising from the application of paragraph (4)).
(4) In any case in which a certificate of conformity has been issued under this subsection, any
warranty required under section 7541(b)(2) of this title and any action under section 754 l(c) of
this title shall be required to be effective only for the emission levels which the Administrator
determines that such certificate was issued and not for the emission levels required under the
applicable standard.
(5) The authorities of section 7542(a) of this title shall apply, subject to the conditions of section
7542(b) 2 of this title, for purposes of this subsection.
83
------- |