[[pp. 26053-26102]] Control of Air Pollution From New Motor Vehicles: Proposed Tier 2

Related Material

Other Related Documents

Note: EPA no longer updates this information, but it may be useful as a reference or resource.

[Federal Register: May 13, 1999 (Volume 64, Number 92)]
[Proposed Rules]
[Page 26053-26102]
From the Federal Register Online via GPO Access [wais.access.gpo.gov]
[DOCID:fr13my99-29]

[[pp. 26053-26102]] Control of Air Pollution From New Motor Vehicles: Proposed Tier 2
Motor Vehicle Emissions Standards and Gasoline Sulfur Control
Requirements

[[Continued from page 26052]]

[[Page 26053]]

our ultimate goal of the 30 ppm standard in an orderly fashion, while
limiting the negative environmental consequences. The temporary nature
of the ABT program would ensure that any negative consequences for Tier
2 vehicles of these higher sulfur levels (120 ppm average in 2004, 90
ppm in 2005) would be minimal. By the time that the majority of new
vehicles sales would be required to meet the Tier 2 standards (2006 and
beyond), average sulfur levels in gasoline would meet the 30 ppm annual
average standard.
We are interested in comment on the corporate pool average values,
and their associated caps. A higher pool average would obviously ease
implementation (e.g., 150 ppm average with an appropriate cap in 2004,
for example), but we have not proposed a higher average because of our
concerns that higher in-use sulfur levels after 2004 are undesirable
for emissions from Tier 2 vehicles. We request that commenters
supporting higher corporate pool average values discuss how such higher
values would affect in-use emission levels of Tier 2 vehicles, as well
as NLEV and Tier 1 vehicles.
We also ask for comment on an alternative approach that would
implement the corporate average requirement for 2004 (120 ppm) but not
require compliance with the 30 ppm standard (with or without credit
use) until 2005. The 120 ppm corporate pool average would continue in
2005 and the 90 ppm corporate pool average would be implemented in
2006, with the requirement to meet the 30 ppm standard (with or without
credits) beginning in 2005 and extending indefinitely, consistent with
the proposed program.
Finally, we request comment on whether refiners should be allowed
to comply with the corporate average standards through the use of
sulfur credits generated under the ABT program (within the limits of
the proposed caps). This would likely render the refinery-specific
standards in 2004 and 2005 unnecessary, and thus refiners would only
have to comply with the per-gallon caps and corporate averages in 2004
and 2005. However, in 2006 and beyond refiners would have to meet the
30 ppm average at every refinery (with limited use of sulfur credits,
to the extent that the 80 ppm cap permits).
We have proposed per-gallon caps of 300 ppm in 2004 and 180 ppm in
2005 at the refinery gate, with slightly higher caps imposed downstream
(as explained in Section VI.B below). We believe that downstream caps
would be necessary to ensure compliance and protect Tier 2 vehicles. At
the same time, we believe caps at the refinery gate would be necessary
to guarantee that the environmental goals of this program were met; the
corporate and refinery averages alone wouldn't provide the full
emissions reductions and environmental benefits we have estimated
because, by themselves, they could allow gasoline with high sulfur
levels in the system as long as the refiner offset any such high sulfur
batches with very low sulfur gasoline. However, there are some
arguments for eliminating the per-gallon standard at the refinery gate
and simply enforcing a per-gallon cap at the retail level (or some
intermediate point downstream). This approach would give refiners and
blenders greater flexibility in blending occasional batches of gasoline
that exceed the proposed cap standards. These refiners/blenders could
sell and transport these high sulfur batches to another party who would
blend down the sulfur level to make gasoline meeting the downstream
caps. One shortcoming of such an approach (removing the per-gallon cap
at the refinery) is that not all gasoline passes through multiple
parties before ending up at the retail level; some refiners ship part
or all of their production directly from refinery to retail outlet. We
welcome comment on whether caps at both the refinery gate and
downstream are appropriate. We also encourage your input on whether the
caps we have proposed to coincide with the corporate average standards
are appropriate. Keep in mind that we need some limitation on sulfur
levels to protect the first Tier 2 vehicles that would begin entering
the marketplace as early as the fall of 2003.
b. Proposed Standards for Small Refiners. As explained in the
regulatory flexibility analysis discussion in Section VIII.B. of this
document, we have considered the impacts of these proposed regulations
on small businesses. As part of this process, we convened a Small
Business Advocacy Review Panel for this proposed rulemaking, as
required under the Small Business Regulatory Enforcement Fairness Act
of 1996 (SBREFA). The Panel was charged with reporting on the comments
of small business representatives regarding the likely implications of
possible control programs, and to make findings on a number of issues,
including:
<bullet> A description and estimate of the number of small entities
to which the proposed rule would apply;
<bullet> A description of the projected reporting, recordkeeping,
and other compliance requirements of the proposed rule;
<bullet> An identification of other relevant federal rules that may
duplicate, overlap, or conflict with the proposed rule; and
<bullet> A description of any significant alternatives to the
proposed rule that accomplish the objectives of the proposal and that
may minimize any significant economic impact of the proposed rule on
small entities.
The final report of the Panel is available in the docket. The Panel
concluded that small refiners were the group most likely to be
negatively impacted by the proposed program. (The Panel noted that
small gasoline marketers would also have to comply with some portions
of a gasoline sulfur program, but did not recommend any regulatory
relief for this group of small businesses.) Many of the small refiners
the Panel met with indicated their belief that their businesses may
close if relief were not considered due to the substantial capital and
other costs required to reduce sulfur levels to the 30/80 standard. The
Panel recommended that EPA solicit comments on a number of options to
provide relief to small refiners, which include some or all of these
provisions:
<bullet> Providing small refiners a four-to six-year period during
which less stringent gasoline sulfur requirements would apply; comment
was also recommended on extending this period for up to a total of 10
years.
<bullet> Basing each small refinery's gasoline sulfur limit on its
individual average sulfur level based on the most recent report(s) to
EPA; and
<bullet> Granting temporary hardship relief on a case-by-case
basis, following the four-to six-year period of relief common to all
small refiners, based on a showing of economic need.
The Panel stated its belief that additional time would allow
sulfur-reduction technologies to be proven out by larger refiners,
thereby reducing the risks to be incurred by small refiners who choose
to incorporate these technologies. The added time would likely allow
for costs of these desulfurization units to drop, thereby limiting the
economic consequences for small refiners. Nationally, giving small
refiners more time to comply would help ensure that cross-industry
engineering and construction resources would be available. Finally,
extending the compliance deadlines would provide small refiners with
additional time to raise capital for infrastructure changes.
i. What Standards Would Small Refiners Have to Meet Under Today's
Proposal?

[[Page 26054]]

Upon evaluating the impacts of our proposed gasoline sulfur
requirements on small refiners and careful review of the Panel's
recommendations, we have determined that regulatory relief in the form
of delayed compliance dates is appropriate to allow small refiners to
comply without disproportionate burdens. We propose that, for a period
of four years after other refiners must start meeting the standards
proposed in Table IV.C-2, refiners meeting clearly defined company size
criteria be allowed to comply with somewhat less stringent requirements
than those just described for refiners and gasoline importers. We
propose to define a small refiner as any company employing no more than
1,500 employees throughout the corporation, including any subsidiaries,
regardless of the number of individual gasoline-producing refineries
owned by the company or the number of employees at any one refinery.
This number is based on the Small Business Administration definition of
a small refiner for the purposes of regulation.⁴⁹ The
proposed annual average small refiner standards beginning with 2004 are
shown in Table IV.C-3 below, although the cap standards begin October
1, 2003.
---------------------------------------------------------------------------

\49\ SBA uses a different definition of small refiner for the
purposes of federal procurements of petroleum products, and EPA in
the past has used criteria based on the processing capacity of the
individual refinery and of all refineries owned by one company.

Table IV.C-3.--Proposed Temporary Gasoline Sulfur Requirements for Small
Refiners in 2004-2007
------------------------------------------------------------------------
Temporary sulfur standards
Refinery baseline sulfur level (ppm) (ppm)
------------------------------------------------------------------------
0 to 30................................ Average: 30.
Cap: 80.^a
31 to 80............................... Average: no requirement.
Cap: 80.^a
81 to 200.............................. Average: baseline level. Cap:
Factor of 2 above the
baseline.^a
201 and above.......................... Average: 200 ppm minimum, or
50% of baseline, whichever is
higher, but in no event
greater than 300 ppm.
Cap: Factor of 1.5 above
baseline level.^a
------------------------------------------------------------------------
^a The cap standard takes effect at the refinery gate October 1, 2003.

We also propose to apply these provisions to any foreign refiner
that can establish that they meet this same definition of small. Since
few if any foreign refiners send all of their gasoline production to
the U.S., allowing eligible small foreign refiners to meet these less
restrictive standards, even on a temporary basis, would be a less
restrictive requirement than it will be for small domestic gasoline
producers since they may be able to send lower sulfur gasoline to the
U.S. without having to incur capital expenses. Furthermore, in many
cases foreign refiners are not subject to the same stringent permitting
and other regulatory requirements that domestic refiners face. At the
same time, we believe many foreign refiners will be installing gasoline
desulfurization equipment because of the various international
requirements that have been proposed and/or finalized (for example, in
Europe, Canada, Japan) that require gasoline sulfur levels to be
reduced to levels similar to our proposed standards and thus these
companies will not avoid all of these costs. In addition, in most cases
we expect importers to be the party responsible for the sulfur level of
imported gasoline, and importers are not eligible for the less
stringent standards applied to small refiners. Hence, the number of
foreign refiners who could benefit (financially and otherwise) from
gaining small refiner status is likely to be very small. However, we
welcome comments on the competitive and other marketplace implications
of this proposal.
We believe that these proposed small refiner standards are
reasonable and that they would not conflict with our overall goals of
reducing gasoline sulfur levels nationwide as soon as possible and of
reducing gasoline sulfur levels sufficiently to enable and protect the
emissions performance of Tier 2 vehicles. Our conclusions are based in
part on the fact that only a very small volume of gasoline will be
eligible for these lesser standards. We have estimated that small
refiners produce approximately 2.5 percent of all gasoline in the U.S.
Furthermore, of the 17 refineries that we have identified as meeting
SBA's definition of small business, nine already have gasoline sulfur
levels less than 90 ppm. Hence, only a very small fraction of the
gasoline sold in the U.S. would take advantage of the higher small
refiner standards through 2007. By the time that a large number of Tier
2 vehicles could have been impacted by residing in or traveling to
areas where higher sulfur fuel is sold, the temporary exemptions for
small refiners would have expired. Furthermore, in most cases, gasoline
produced by small refiners is mixed with substantial amounts of other
gasoline prior to retail distribution (due to the functioning of the
gasoline distribution system), likely resulting in only marginal
increases in overall sulfur levels. Thus, the sulfur level of gasoline
actually used by Tier 2 vehicles should generally be much lower than
that produced by individual small refineries who receive unique
compliance standards through 2007.
As explained above, we are proposing that compliance under the
proposed standards be based on a refiner's being able to show that it
meets specific criteria. If a refiner were able to qualify as a small
refiner under our definition, it would need to then establish a sulfur
baseline for each participating refinery. For small refiners,
compliance with the proposed sulfur regulations would be determined on
the basis of the sulfur baseline for each refinery owned by that
company. The following sections explain these proposed requirements in
more detail, to supplement the information be presented above. We also
explain how small refiners could obtain an additional two-year
exemption upon establishing a hardship case, as well as how small
foreign refiners could establish eligibility for compliance under the
small refiner provisions.
ii. Application for Small Refiner Status.
We are proposing that refiners seeking small refiner status under
our gasoline sulfur program would have to apply to us in writing no
later than June 1, 2002, requesting this status. In this application,
the refiner must demonstrate that as of January 1, 1999, the business
and any subsidiaries, including all refining, distribution, and
marketing activities, as well as any other activities worldwide,
employed 1,500 or fewer employees. We are proposing that in the case of
refineries owned by joint ventures, the total employment of both (all)
companies would be considered in determining whether the 1,500 employee
limit is reached. If a refiner that is not small as of January 1, 1999
subsequently sells part of its business and as a result has fewer than
1500 employees, it would not be eligible for a small refiner status.
These provisions would provide stability to the regulated and
regulatory parties and ensure that no ``gaming'' of the program occurs.
However, we are also proposing that any new refinery built between
January 1, 1999 and January 1, 2001, or a refinery that was not
operational as of January 1, 1999, owned by a refiner that meets our
proposed definition, could apply for small refiner status no later than
June 1, 2002. In this case, we would consider carefully the history of
the refinery and

[[Page 26055]]

the company in determining whether it is appropriate to grant this
refiner small refiner status.
We are also proposing that if a refiner with approved small refiner
status later exceeds the 1,500 employee threshold without merger or
acquisition, its refineries could keep their individual refinery
standards. This is to avoid stifling normal company growth and is
subject to our finding that the refiner did not apply for and receive
the small refiner status in bad faith. An example of an inappropriate
application for small refiner status would be a refiner that
temporarily reduced its workforce from 1,600 employees to 1,495
employees prior to January 1, 1999, and then rehired employees after
the cutoff date. This would be a bad faith attempt to avoid the intent
of the rule. We are requesting comment on this provision.
At any time after June 1, 2002, a refiner with approved small
refiner status could elect to cease complying with the small refiner
standards and, in the next calendar year, begin complying with the
standards specified in Table IV.C-2 and related provisions. However,
this decision would apply to all refineries owned by that refiner and
once a refiner dropped its small refiner status, it would not be
eligible to be reinstated as a small refiner at some later date.
iii. Application for a Small Refiner Sulfur Baseline.
A qualifying small refiner could apply for an individual sulfur
baseline by June 1, 2002 for any refinery owned by the company by
providing a calculation of its sulfur baseline using its average
gasoline sulfur level based on 1997 and 1998 production data, and the
average volume of gasoline produced in these two years. The proposed
regulations specify the information to be submitted to support the
baseline application. The baseline calculations should include any
oxygen added to the gasoline at the refinery. This application would be
submitted at the same time that the refiner applied for small business
status; confirmation of small business status would not be required to
apply to EPA for an individual sulfur baseline. If the baseline were
approved, we would assign standards to each of the company's refineries
in accordance with Table IV.C.-2.
Blenders would not be eligible for the small refiner individual
baselines and standards because they would not have the burden of
capital costs to install desulfurization equipment, which is the
primary reason for allowing small refiners to have a relaxed compliance
schedule.
iv. Volume Limitation on Use of a Small Refinery Standard.
We are proposing that the volume of gasoline subject to the small
refinery's individual standards would be limited to the volume of
gasoline the refinery produced from crude oil, excluding the volume of
gasoline produced using blendstocks produced at another
refinery.⁵⁰
---------------------------------------------------------------------------

\50\ In addition to gasoline produced from crude oil, a small
refinery's baseline volume would include gasoline produced from
purchased blendstocks where the blendstocks are substantially
transformed using a refinery processing unit.
---------------------------------------------------------------------------

Under this approach, the baseline volume for a small refinery would
reflect only the volume of gasoline produced from crude oil during the
baseline years. In addition, use of the refinery's individual baseline
sulfur level during each calendar year averaging period (beginning with
2004) would be limited to the volume of gasoline that is the lesser of:
(1) 105% of the baseline volume, or (2) the volume of gasoline produced
during the year from crude oil. Any volume of gasoline produced during
an averaging period in excess of this limitation would be subject to
the standards applicable to refiners not subject to a small refiner
standard. In this case, the small refiner's annual average standard
would be adjusted based on the excess volume in a manner similar to the
compliance baseline equation for conventional gasoline under Section
80.101(f) of Part 40 of the Code of Federal Regulations. However, the
small refiner's per-gallon cap standard would not be adjusted.
This limitation would assure that small refiners receive relief
only for gasoline produced from crude oil, the portion of the refinery
operation requiring capital investment to meet lower sulfur standards.
We are requesting comment on this provision and whether an alternative
approach may be more appropriate for the stated purpose.
v. Hardship Extensions Beyond 2007 for Small Refiners.
Beginning January 1, 2008, all small companies' refineries would
have to meet the permanent national sulfur standard of 30 ppm on
average and the 80 ppm cap, except small refineries that apply for and
receive a hardship extension. A hardship extension would provide the
small refiner an additional two years to comply with these national
standards. A hardship extension would need to be requested in writing
and would specify the factors that qualify the refiner for such an
extension. Factors considered for a hardship extension could include,
but would not be limited to, the refiner's financial position; its
efforts to procure necessary equipment and to obtain design and
engineering services and construction contractors; the availability of
desulfurization equipment, and any other relevant factors.
By January 1, 2010 all refiners would be required to meet the
permanent national average standard and cap. We are requesting comment
on the proposed hardship extension, including the factors to be
considered in petitions for extension, and the proposed time periods.
vi. What Alternative Provisions for Small Refiners Are Possible?
We have proposed one type of program to address the needs of small
refiners. We solicit comment on other options so that we can consider
these options as we finalize this rule. We encourage comments. We
request comment on a range of alternatives, including those listed
below, which could be considered when developing unique regulatory
requirements for small refiners. We specifically request that the
comments address not only the economic but also the environmental
implications of the alternative, relative to the program we've
proposed.
<bullet> Are there alternative or additional criteria that could/
should be used to define a small refiner, such as the volume of crude
oil processed or the volume of gasoline produced (since the gasoline
sulfur standard applies specifically to gasoline)? Other criteria may
also be acceptable, such as a different employee number for
qualification as a small entity, or basing the count on employees
employed in gasoline production only. We welcome your recommendations.
Our desire is to limit the number of companies meeting the small
refiner definition in order to provide regulatory relief only to those
companies that have the economic concerns unique to small businesses.
If you recommend criteria other than number of employees, please
comment on how those criteria can be shown to limit the number of
refineries that will be eligible for the proposed relief.
<bullet> Are the caps and averages of the proposed interim
standards for small refiners (see Table IV.C.-3) appropriate for the
corresponding individual sulfur baseline levels?
<bullet> What is an appropriate and sufficient time period for the
proposed small refiner interim standards? Would most qualifying small
refiners be able to meet the 30/80 standards within four years (six if
a hardship extension is granted, which is dependent on the case made by
the individual refiner), as proposed? The Panel report suggested that a
period of six to ten years could

[[Page 26056]]

be desirable to provide sufficient time for small refiners to comply
with the proposed standards. What are the arguments for granting more
than four years of additional time and what are the environmental
implications (and implications for Tier 2 vehicles) of such an
extension?
<bullet> Should small refineries of multi-refinery companies
(companies too large to meet the proposed small refiner criteria) be
eligible for small refiner interim standards? Should refineries not
producing gasoline as a major product (for example, refineries engaged
primarily in the production of lubricants where gasoline is a small
volume by-product) be eligible for small refiner interim standards
regardless of corporate size/employment?
<bullet> If a small refiner operates more than one refinery (while
still meeting our proposed small refiner criteria), should that refiner
be permitted to aggregate the sulfur baselines and comply with the
small refiner standards applicable to that aggregate baseline? Under
the sulfur ABT program described below, we are proposing to require
refiners to aggregate data from all of their refineries when
determining compliance with the 2004 and 2005 corporate average
standards (Table IV.C.-2) (but not the refinery gate standards,
although we seek comment on that alternative).
<bullet> Rather than providing unique standards for qualifying
small refiners, would the need for separate small refiner provisions be
addressed if we were to adopt a regional sulfur program? In Section
IV.C.1. above, we explained our concerns that a regional sulfur program
would not achieve the same emission reductions we project for our Tier
2/gasoline sulfur program. However, some have suggested to us that a
regional program would address the need for small refiner provisions
since the majority of small refiners are thought to sell gasoline in
the West. We know of several refiners that appear to meet our proposed
criteria for being small that sell at least some of their gasoline
production in the eastern U.S. (as defined by the oil industry's
proposed program) and thus a regional program would not cover all small
refiners. We encourage comments on this alternative, particularly from
refiners who could be impacted by such a decision.
<bullet> Would a more general hardship provision that would be
based on a showing of substantial economic hardship, such a discussed
in Section IV.C.4.c., provide sufficient compliance flexibility to
address the needs of small refiners?
4. Compliance Flexibilities
In addition to the basic standards applicable to refiners that were
explained above, we are proposing two additional programs that will
provide flexibility for refiners when complying with the proposed
standards. The first is the sulfur ABT program mentioned previously.
The second is a program to streamline the construction permitting
process so that refiners can make the required process modifications by
2004.
a. Sulfur Averaging, Banking, and Trading (ABT) Program. We are
proposing that any refiner or importer be allowed to generate, bank,
and trade sulfur credits. A sulfur ABT program would accelerate the
reduction of sulfur in gasoline and provide refiners with additional
flexibility in achieving compliance with the 30 ppm standard in 2004
and beyond. The following paragraphs provide additional information
about our proposed sulfur ABT program, to supplement that presented in
Section IV.C.-3.a above. We encourage comments on the design elements
we have proposed for the sulfur ABT program. If you believe alternative
approaches would make the program more useful to the refining industry,
please share your specific recommendations with us.
i. Why Are We Proposing a Sulfur Averaging, Banking, and Trading
Program?
A sulfur ABT program, if properly implemented, would provide the
opportunity for a win for both the refining industry and the
environment. The flexibility provided by an ABT program could provide
refiners more lead time to bring all of their refineries into
compliance with the 30 ppm standard, by allowing them to use credits
generated at one refinery to delay having to desulfurize gasoline from
another refinery. ABT would provide the opportunity for reduced costs
by allowing the industry the flexibility to average sulfur levels among
different refineries, between companies, and across time. Since, under
banking, early reductions have a value during program implementation,
ABT provides an incentive for technological innovation and the early
implementation of refining technology.
The ABT program could provide meaningful early benefits for the
environment because it would allow the Tier 2 standards to be
implemented earlier than might otherwise have been possible, and
because it would provide direct environmental benefits. The first
direct benefit relates to atmospheric sulfur loads. This benefit is
largely independent of when credits are generated and used. However,
atmospheric deposition and transformation rates of sulfur compounds
tend to vary geographically and seasonally and thus we must consider
whether a broad averaging program would have different pollutant
effects when compared to a more constrained averaging program or a
program without averaging. Any potential negative effects of a broad
ABT program should be mitigated by the geographic distribution of
refineries, the widespread distribution pipelines, and the fungible
nature of gasoline. All of these factors, taken together, lead us to
believe that any negative effect on atmospheric sulfur levels from ABT
(relative to a single 30 ppm average/80 ppm cap in 2004) would be
negligible. It should be noted that this situation is further moderated
by the pool averages and caps proposed for 2004 and 2005, since these
averages and caps would reduce actual gasoline sulfur levels as the ABT
program phases in.
Another environmental benefit is related to the effect of gasoline
sulfur on catalyst performance, as discussed in the draft RIA. Since
catalyst performance depends in part on gasoline sulfur levels, we must
consider whether the emissions benefits (measured in g/mi-per-ppm) of
early sulfur reductions when credits are generated are essentially the
same as the g/mi-per-ppm benefits when the credits are used. The effect
of sulfur on emissions from Tier 0 and Tier 1 vehicles, which will
dominate the fleet in 2000-2005, is approximately the same when sulfur
levels increase from 30 to 150 ppm as it is when sulfur levels increase
from 150 ppm to 330 ppm. In other words, for each ppm increase in
sulfur levels, approximately the same effect on emissions results
regardless of whether the increase is from low levels (e.g., from 30
ppm up to 150 ppm) or from higher levels (e.g., from 150 ppm up to
current average levels). Therefore, the emissions benefits from credits
generated before 2004 would essentially offset the emissions effects of
those credits being used in 2004 and beyond, especially since corporate
pool average sulfur levels could not exceed 120 ppm in 2004 and 90 ppm
in 2005, and sulfur levels will be capped at 80 ppm in 2006 and beyond.
Nonetheless, there remains concern about the sensitivity of later
models (NLEV and Tier 2) to sulfur and about the reversibility of the
effect of higher sulfur levels on catalyst efficiency. More explicitly,
the relatively few Tier 2 vehicles that would see somewhat higher
sulfur levels than 30 ppm in 2004 and 2005 (about three-quarters of

[[Page 26057]]

a model year of production) would not be able to fully recover the loss
in emissions performance due to the higher sulfur levels. Hence, the
corporate averages and caps would be necessary in these interim years.
In 2006 and beyond, the 80 ppm cap and the 30 ppm average refinery
standard, even with the ongoing use of credits to comply with the 30
ppm standard, would keep in-use sulfur levels very close to 30 ppm.
Thus, Tier 2 vehicles sold in 2006 and beyond would receive appropriate
protection from gasoline sulfur.
ABT programs must be designed and implemented carefully to be
certain that they are sensitive to equity and competitive issues in the
industry and do not create the potential for inadvertent emission
increases. In the context of gasoline sulfur control, concerns about
different baseline sulfur levels and different technological
capabilities among refiners must be considered. Even with the proposed
lead time, some refiners would find it easier to achieve reductions
than would others. This is due to a number of factors, including
refinery configuration, product mix (gasoline versus distillates),
crude oil sulfur levels, and the ability to generate capital to fund
the investment. At the same time the program must be designed to
eliminate the possibility of windfall credits and to be sure that the
environmental benefits associated with early sulfur reductions offset
the potential forgone benefits when the credits are used.
The program we are proposing today attempts to strike a balance
among all of these factors. Some of the elements and design features
(such as the eligibility trigger and the baseline requirement) were
included to address concerns such as timing, disparate capabilities
among refineries, and the potential for excessive (``windfall'')
credits. We are seeking comment on options for dealing with all of the
issues we have identified.
The ABT program is voluntary. No refiner or importer qualifying for
credits is required to generate them, use them, or make them available
to others (except as discussed in Section IV.C.4.a.vi. below). The
process for establishing a sulfur baseline and generating and using
credits is outlined below.
ii. How Would Refiners Establish a Sulfur Baseline?
To establish a sulfur baseline against which credits would be
calculated, we propose that by July 1, 2000, each refiner or importer
that wants to generate credits submit two pieces of information to the
Agency. One would be the volume-weighted average sulfur content for
conventional gasoline (CG) for each refinery (or imported by that
importer) for 1997 and 1998. The second would be the annual average
volume of CG produced by that refinery (or imported by the importer) in
those years. ⁵¹ ⁵²
---------------------------------------------------------------------------

\51\ Since participation in the sulfur ABT program is voluntary,
refines opting not to generate or use sulfur credits do not have to
establish a sulfur baseline for this program.
\52\ We believe that variations in specific gravity, which could
affect the sulfur content of gasoline as determined on a mass basis,
will average out over the year and need not be included in the
calculations. However, we request comment on whether specific
gravity should be considered in the calculation of sulfur baselines
(including whether such data exists for 1997-98) and subsequently,
in calculating credits generated relative to this baseline.
---------------------------------------------------------------------------

Since we expect summer RFG sulfur levels to decrease in 2000 to
approximately 150 ppm (due to the actions refiners will take to meet
the Phase II NO_X standards for RFG), we are proposing to set
the individual refinery sulfur baseline for summer RFG at 150 ppm,
regardless of volume produced in 1997 and 1998. Winter RFG production
would be assigned the same sulfur baseline as the refinery's
conventional gasoline, without regard to the volume of winter RFG
produced in 1997-98. Hence, no reporting of RFG sulfur levels or
volumes would be required in setting a sulfur baseline. We encourage
comments on the use of different sulfur baselines for summer and winter
RFG, particularly regarding whether this could create a disincentive to
produce RFG in the summer months. We do not want to jeopardize our RFG
program, but at the same time, we want sulfur credits to reflect
actions taken by refiners above and beyond their current operations
and/or regulatory obligations.
Conventional gasoline produced in 2000 and beyond that exceeded
105% of the CG baseline volume produced at that refinery would be
assigned a sulfur baseline (from which credits would be generated) of
150 ppm. This provision is intended to prevent increases in average
sulfur levels resulting from increases in CG production. A refiner/
importer of conventional gasoline to which oxygenate is added
downstream during 1997-1998 could include the downstream oxygenate
volume in that refinery's CG baseline, if the refiner can substantiate
that oxygenate was added to that gasoline.
A refinery/importer that did not produce/import gasoline during
1997-1998 would be assigned a baseline of 150 ppm each for CG and RFG
for the purposes of sulfur credit generation in 2000 and beyond. This
provision would also apply to blenders of natural gasoline, butane, or
similar non-oxygenated blending components. Such parties would be
considered refiners and would need to meet all requirements, such as
analyzing each batch of the blending component for sulfur prior to its
addition to gasoline. Credits would be based only on the volume of the
blending components. We encourage comments on alternative provisions
for establishing baselines for refiners/importers that could not
establish a 1997-98 sulfur baseline as described above. In particular
would 150 ppm be appropriate, or would a greater or lesser sulfur
content be most equitable and most environmentally neutral? Should this
baseline be tied in some way to the trigger for credit generation in
(as discussed below) 2000-2003?
We request comment on several aspects of this baseline provision.
The 1997-1998 years for the baseline represent the latest available
data and thus best reflects the present state of each refinery's
gasoline sulfur levels. However, we already have established baseline
sulfur levels for 1990 for most refineries. Except for changes related
to RFG, average gasoline sulfur levels have changed little since 1990.
Hence, we request comment on whether that 1990 baseline would be a
suitable substitute. Alternately, we request comment on whether 1997
and 1998 are the appropriate years to average when establishing a
sulfur baseline, given that mandatory use of the Complex Model starting
in 1998 could have led to changes in sulfur levels between 1997 and
1998. Since our purpose in proposing to establish sulfur baselines is
to try to capture current sulfur levels (within a reasonable date of
the 2000 start date for credits to be generated), the sulfur baseline
could be based on a single year's data (for example, 1998) rather than
a two-year average. We proposed a two-year average to try to capture
and accommodate operational fluctuations and changes. However, a single
year's data may adequately capture current sulfur levels.
We are not proposing a formal baseline review and/or approval
process since the proposal envisions a self-certifying process.
Refiners would submit their 1997 and 1998 sulfur baseline data for each
refinery to us, and then would generate credits from that baseline in
2000-2003. If we determined, through a refinery audit or other action,
that the sulfur baseline was calculated with incorrect data, we would
establish a new sulfur baseline and the refinery would subject to that
baseline, even if it meant recalculating

[[Page 26058]]

the number of credits generated in subsequent years. We have used this
baseline review process in other mobile source programs and believe it
works well, but we request comment this approach.
We considered the possibility that, since refiners report annual
production information to EPA, we could issue baselines for each
refinery rather than refiners having to submit them to us. However, we
do not think this is a possible solution because many refiners comply
with our RFG and CG requirements by aggregating the data from all of
their refineries. Thus, the data we currently receive from refiners
would not allow us to establish an individual baseline for every
refinery in the U.S. (unless we went back to 1990 data). However, we
would like comment on whether a more formal sulfur baseline approval
process (say, a letter from the Agency or a date by which approval can
be assumed unless the refiner hears otherwise) would be desirable. Keep
in mind that even with a more formal baseline approval process, the
baseline could be changed at a later date if we found, during an audit
of refinery records, errors in compliance with the proposed baseline
requirements. Hence, any up-front approval would only provide certainty
that, based on the data reported to us, we believe the refiner had
correctly applied the mathematical equations proposed today for
establishing a sulfur baseline.
Some have raised the concern that if imported gasoline were allowed
to be used for credit generation, as we propose today, foreign refiners
might be able to gain an unfair advantage. For example, it is possible
that foreign refiners could simply re-blend their gasoline (without
installing new capital equipment) and send their lowest-sulfur refinery
streams to the U.S. at a lower cost than gasoline produced by domestic
refiners that had to reduce overall sulfur levels through
desulfurization. Since importers, not foreign refiners, would be the
parties assigned a sulfur baseline and eligible for generating credits,
we do not believe foreign refiners would have a strong incentive to
send lower sulfur gasolines to the U.S. We believe that the benefits of
allowing importers to participate in the sulfur ABT program (more
players in the credit trading field, more chance for early reductions
in gasoline sulfur levels) outweigh the potential detriments. However,
we encourage comment on the implications of the decision to allow
imported gasoline to be used for credit generation.
Oxygenate blenders would not be able to participate in this
proposed credit program because they would not be subject to the sulfur
standard. Special provisions would exempt them from having to measure
the sulfur content of the oxygenate they blend and from the
recordkeeping and reporting requirements of the sulfur program, other
than the requirements that apply to all parties that handle gasoline
and gasoline blendstocks downstream of the refinery.
iii. How Would Refiners Generate Credits?
During the period 2000-2003, credits could be generated annually by
any refinery that produced conventional gasoline averaging 150 ppm
sulfur or less on an annual, volume-weighted basis. Credits would be
calculated based on the amount of reduction from the refinery's CG
sulfur baseline.⁵³ Credits could also be generated from
winter RFG based on reductions from the sulfur baseline, if the winter
RFG sulfur level averaged 150 ppm or less (on a seasonal volume-
weighted basis). Similarly, summer RFG would need to have a seasonal
volume-weighted average sulfur level below 150 ppm to be eligible for
credit generation, although credits would only be created based on the
difference between 150 ppm and the summer RFG sulfur average. Thus,
credits would need to be generated separately for conventional gasoline
and RFG. Conventional gasoline produced in excess of 105% of the
baseline volume could only generate credits for sulfur reductions below
150 ppm, not for the cumulative reduction from the baseline sulfur
level. Winter RFG would not be subject to any volume limitations, and
thus refineries could generate credits for any volume of winter RFG
that contains 150 ppm sulfur or less.
---------------------------------------------------------------------------

\53\ If a refinery's baseline average were 150 ppm or less,
credits could only be generated for annual average reduction's below
the baseline level.
---------------------------------------------------------------------------

For example, if in 2002 a refinery reduced its annual average
sulfur level for conventional gasoline from a baseline of 450 ppm to
150 ppm, its sulfur credits would be determined based on the difference
in annual sulfur level (450-150=300 ppm) multiplied by the volume of
conventional gasoline produced (up to 105% of the baseline CG volume).
If this refinery produced more CG than 105% of the baseline volume, it
would only generate credits from that incremental volume if the
incremental gasoline were below 150 ppm. (For example, if the
refinery's 2002 average CG sulfur level were 100 ppm, it would get 150-
100=50 ppm sulfur credits on any volume in excess of 105% of its
baseline CG volume, as well as 450-100=350 ppm for the baseline volume
up to 105%.)
If this same refinery also produced RFG with an annual average
sulfur content of 90 ppm in 2002, it could also receive sulfur credits
calculated based on the difference between 150 ppm and 90 ppm (60 ppm)
times the volume of summer RFG produced plus 360 ppm (450-90) times the
volume of winter RFG produced. A refinery with a sulfur baseline lower
than 150 ppm sulfur would only generate credits relative to reductions
from its baseline, for either CG or winter RFG. Credits from summer RFG
would be based on reductions from 150 ppm.
Several states have implemented or are considering gasoline sulfur
control programs. To avoid double-counting of emission benefits, lower
sulfur gasoline produced to comply with these state programs would not
be eligible for early banking credits under this program.
In 2004 and beyond we propose that credits could only be generated
for actual annual sulfur averages below the 30 ppm standard (combining
conventional and reformulated gasolines), and only for the difference
between the standard and the actual annual sulfur average. (For
example, a refinery producing gasoline in 2004 that averaged 25 ppm
could generate 30-25=5 ppm, while a refinery producing gasoline that
averaged 40 ppm would not be eligible for any credits.)
We encourage comments on this credit generation concept. In
particular, would these formulas permit sufficient credits to be
generated industry-wide to provide adequate credits for use in
compliance in 2004 and beyond? If not, what are the limitations on
credits and what changes could be made to improve the likelihood that
sufficient credits would be generated?
Our proposal to cap volumes on which credits could be generated at
105 percent of baseline levels is intended to preclude the possibility
of closely-located refineries generating credits by moving blendstocks.
This could occur if a refinery with a relatively low baseline level
moved blendstocks to a refinery with relatively higher levels, thus
allowing the somewhat artificial generation of credits. We request
comment on whether such a provision is necessary and whether the 5
percent cap should be increased to as high as 10 percent to reasonably
accommodate normal growth in volume. We raise some potential
alternatives to these provisions in Section IC.C.4.a.vi. below, and
encourage your consideration of all of these issues in your comments.

[[Page 26059]]

iv. How Would Refiners Use Credits?
Credits generated prior to 2004 would have to be used or
transferred by 2007. Credits generated in 2004 and beyond would have to
be used or transferred within five years of the year in which they were
generated. If these credits were traded to another party, they would
have to be used by the new owner within five years of the year of
transfer. Since the transfer could occur any time within five years of
generation, some credits could have a life of up to ten years.
Our proposed ABT program is designed to ease implementation of the
new standards and credits would be of their greatest value during
phase-in periods. ABT is not necessarily intended to permit a refinery
to operate above the standard for a protracted time period. While
limiting credit life might reduce the incentive to generate credits and
could create a ``use or lose'' mentality, the credit program would seem
to be of relatively small value to any refiner/importer that held
credits for five years and did not need to use them. We believe that
limiting credit life is appropriate since we must also consider the
basic reason for ABT and address concerns about our ability and the
ability of the refiners to maintain the integrity of the credit system
over many years. EPA requests comment on credit life including options
such as limiting life by depreciating their value over a period of
years as well as longer or shorter periods of fixed credit value.
We propose that credits could be withdrawn from a refinery's/
importer's credit bank or purchased from another refinery/importer to
bring the annual sulfur average for each refinery down to the 30 ppm
standard beginning in 2004. There would be no geographic constraints on
credit trades. However, as explained in Section IV.C.3.a above, in 2004
no batch of domestically produced or imported gasoline could exceed 300
ppm, and a refinery's/importer's actual annual corporate pool average
sulfur level could not exceed 120 ppm. (A refiner owning more than one
refinery would have to aggregate the respective sulfur levels of
gasoline produced at those refineries for determining compliance with
the 120 ppm standard.) In 2005, gasoline sulfur would be capped at 180
ppm and the corporate pool average could not exceed 90 ppm. The
aggregation requirement would also apply in 2005. As described above,
credits would apply only to compliance with the 30 ppm refinery
standard, not to the corporate pool average or the cap.
A refiner or importer choosing to participate in the ABT program
would be required to file annual reports with the Agency indicating the
applicable baselines or standard(s) in ppm sulfur, the annual
average(s) in ppm sulfur, and the annual volume(s) in gallons (for each
refinery). These calculations would be reported, along with an
accounting of credits banked, transferred (sold), or acquired (bought).
(For 2000-2003, the reports would only cover credits banked and
traded.) The credits would be in units of ppm-gallons.
Thus, for each purchase of credits, as reported on the buyer's
annual report, there should be a corresponding entry on the seller's
annual report. Through the report, refiners would have to demonstrate
that their average sulfur levels (with the use of credits, if
necessary) comply with the 30 ppm standard at each refinery. Refiners
would also have to demonstrate that the combined production from all
refineries meets the corporate average standard. As mentioned above,
the actual corporate averages could not exceed 120 ppm in 2004 and 90
ppm in 2005. The identity of refiners/refineries and importers involved
in these transactions would be reported, along with the registration
numbers assigned to them by the Agency under the RFG/CG program (40 CFR
part 80, Subparts D, E, and F).
In addition, we are concerned that the potential exists for credits
to be generated by one party and subsequently purchased or used in good
faith by another, and later found to have been calculated or created
improperly or otherwise determined to be invalid. In this case, both
the seller and purchaser would have to adjust their sulfur calculations
to reflect the proper credits and either party (or both) could be
deemed in violation of the standards and other requirements if the
adjusted calculations demonstrate noncompliance with an applicable
standard. We have taken this approach in our other fuels enforcement
programs. We welcome comments on this provision. In particular, we
request comment on whether our program should be designed such that
only the seller should be deemed in violation if that party sold
invalid credits and, upon correction for this error, was found to have
violated one or more standards. In general, mobile source ABT programs
hold both parties liable.
For the duration of the credit program, each participating refinery
and importer could make deposits to and withdrawals from its ``bank
account''. All transactions would have to be concluded by the last day
of February after the close of the annual compliance period (2004,
2005, etc.). It would be up to the industry to establish any mechanisms
for linking buyers and sellers. The Agency does not intend to become
involved in this marketplace activity.
We are also proposing to allow refiners to miss the 30 ppm standard
for an individual refinery and to carry forward the credit debt that
would have brought that refinery into compliance in the year the
deficit occurred. This is very similar to provisions proposed today for
auto manufacturers in complying with the averaging provisions Tier 2
standards. Under this provision, the refiner would have to make up the
credit deficit and bring that refinery into compliance with the 30 ppm
standard the next calendar year, or face penalties. This program would
in no way absolve the refiner from having to meet the applicable per-
gallon cap standard. This provision would provide some relief for
refiners faced with an unexpected shutdown or that otherwise were
unable to obtain sufficient credits to meet the 30 ppm standard. We
welcome comment on this provision.
The following Table IV.C.-4 summarizes the compliance dates and
program requirements of this proposed sulfur ABT program. See Section
VI for more specific information, particularly about the dates that the
sulfur caps would apply and the standards that would apply downstream
of the refinery.

BILLING CODE 6560-50-P

[[Page 26060]]

[GRAPHIC] [TIFF OMITTED] TP13MY99.003

BILLING CODE 6560-50-C
v. Could Small Refiners Participate in the ABT Program?
We believe that refiners complying under the small refiner
provisions outlined in the previous section should not be permitted to
use sulfur credits to meet the average standard applicable to their
refineries. We are proposing to exclude small refiners from using
credits to meet the small refiner standards because the small refiner
standards are generally more lenient than the 30 ppm standard and thus
these refiners should have less need for a credit trading program than
the rest of the industry. Furthermore, small refiners, even those
currently producing gasoline near the 30 ppm average, are given an
additional two years (until 2008) to meet the 30 ppm standard compared
to refiners complying under the sulfur ABT program. We want to ensure
that the sulfur levels of the majority of gasoline are reduced on
average, and overall, in 2004 and 2005; permitting small refiners to
meet the more lenient standards through the purchase of credits could
jeopardize that goal by resulting in in-use sulfur levels that are even
greater than the maximum small refiner standard (300 ppm average). If a
small refiner believed it could generate sufficient sulfur credits in
2000-2003, or obtain such credits through purchases from other
refiners, to be able to meet the 30 ppm average and the corporate
averages of 120 ppm in 2004 and 90 ppm in 2005, it should choose not to
participate in the small refiner program and take full advantage of the
sulfur ABT program.
However, small refiners would be permitted to generate and trade
sulfur credits if they reduced sulfur levels early in 2000-2003, per
the requirements outlined above. Furthermore, a small refiner could
sell credits that were generated in 2000-2003 in 2004 and 2005 while at
the same time meeting the small refinery standards. A small refiner
wishing to generate and sell credits would have to establish the
individual refinery sulfur baseline by the deadline specified above for
the ABT program (July 1, 2000) but could wait until June 1, 2002 to
apply for small refiner status. However, the standards assigned to that
refinery (as presented in Table IV.C-3) would be based on the sulfur
level from which credits were generated, not the 1997-98 baseline
sulfur level, since the refiner would have already demonstrated the
ability to meet the lower sulfur level (in this case, 150 ppm or lower
on an annual average basis).
At any time, a small refiner could ``opt out'' of the small refiner
program and, beginning the next calendar year, comply with the
standards in Table IV.C-2. The refiner would have to notify us of this
change in compliance program. Once a small refiner left the small
refiner program, however, we propose that it would not be eligible to
re-enter the small refiner program. We encourage comments on this
provision.
The sulfur ABT program could provide an alternative to offering any
small refiner standards, if small refiners were capable of complying
with the proposed pool average standards and caps in 2004 and 2005 just
as larger refiners could. In this case, all refiners, large or small,
could obtain credits necessary to meet the 30 ppm average standard for
the two intervening years. However, EPA recognizes that this may not be
the best response to the needs of small refiners, and has proposed, as
a result of the SBREFA Panel process, alternate standards in section
IV.C.3.b of this document. Indeed many small refiners expressed concern
during the Panel process that an ABT program would not address their
needs. However, we welcome comments on the pros and cons of using the
sulfur ABT program to provide regulatory relief for small refiners in
lieu of additional regulatory standards unique to small refiners.
vi. What Alternative Implementation Approaches Are Possible?
As we were developing this proposal, members of the oil industry
and others expressed concern that the ABT program as described above
may not be of great value in providing flexibility in complying with
the 30 ppm standard in 2004. Several different concerns have been
expressed.
Industry representatives have asserted that the opportunity to
generate early credits is limited because the proposed lead time would
be too short to implement enough of the refinery operational changes
and capital investments needed to achieve sulfur reductions before
2004. Additionally, the industry is concerned that relying on early
credits generated with what is perhaps the best long-term
technology(ies) is problematic because the preferred technology(ies) is
new and

[[Page 26061]]

does not yet have a proven performance record. Their concern is further
exacerbated by the uncertainty in the diesel fuel sulfur
picture, the MTBE /oxygenates situation developing in California, and
the DI petition discussed below, as well as ongoing state initiatives
to reduce sulfur in gasoline before this action is decided upon.
When credits are generated, there is a fear that those that
generate them will hoard them, particularly refiners that operate
several refineries. And when credits are made available for trade, they
may not become publicly available in enough time for them to be
considered by others in their capital investment planning, so
essentially all refineries would have to take steps to implement 30 ppm
technology by 2004. These issues may be of special concern to those
moderate sized refiners that are too large to qualify as small entities
but do not have enough refineries or refineries of the right gasoline
production volume to internally optimize their operations under the ABT
program.
Given these uncertainties about credit availability, the refiners
may need additional flexibility as a means to provide relief to those
that make a good faith effort to comply but are precluded by
circumstances beyond their control. These may include unanticipated
technological and commercial concerns, credit availability problems, or
force majeure type events.
We have examined this issue of credit availability and our
analysis, which is presented in the Draft RIA, indicates that credits
should be available by 2004 for the 2004/5 phase-in. This is based on
the fact that the 300 ppm cap in 2004 would require that all refineries
with a baseline above 300 ppm reduce sulfur by 2004. And, while they
could choose to just achieve 300 ppm, some would need greater
reductions to comply with the 120 ppm corporate pool average standard
and all would be facing increasingly more stringent requirements in
2005 and beyond. Quite simply, we believe that good business sense
would dictate that once a hardware investment is made the refinery
would shoot for 30 ppm or less. As the analysis shows, this approach
implemented over just three years would yield compliance with the 120
ppm corporate pool average and would generate ample credits. We
requested comment on our analysis in the Draft RIA and the underlying
analytical approach.
EPA is proposing the ABT program described above in order to
increase the refiners'/importers' confidence that they could comply in
2004. And, while our analysis indicates that credits would be available
for 2004/2005 compliance, we realize that the ABT program might not
meet its objective if the industry did not have confidence that credits
would be available in enough time and in sufficient quantities to
enable them to make economically efficient investment decisions. It is
our desire to provide the industry as much flexibility as possible to
ease implementation and phase-in while still meeting the objectives of
the program as described above. Toward that end we are asking for
comment on several variations on the above proposal that might increase
its overall value as a means to provide flexibility in meeting the
proposed standards. These can be divided into four categories: (1)
Modifications to the design elements of the proposed ABT program, (2) a
compliance supplement pool, (3) an allowance-based system, and (4)
reserved credits. As constructed below, the compliance supplement pool,
an allowance-based system, and reserved credits could be implemented in
varying ways to complement the early ABT program. EPA asks comments on
the cost and air quality impact implications of these concepts, which
are described in more detail below.

Potential Modifications to Proposed ABT Program

Modifications to the base program to increase the potential
availability of credits and the time over which these credits could be
used might increase the effectiveness of the proposed ABT program.
These changes could potentially affect both the near-term when the
program was phasing-in and the long term when the 30 ppm standard was
fully implemented.
The 150 ppm trigger value is designed to ``level the playing
field'' between companies with relatively low baselines and those with
relatively high baselines. Those with high baselines could potentially
generate more credits than those with lower baselines, but at a
somewhat greater cost since achieving 150 ppm or less becomes
increasing more difficult with higher sulfur gasoline. Those with
baselines closer to 150 ppm may be able to generate fewer credits, but
generate them more easily.
However, requiring that gasoline be below 150 ppm before credits
could be generated might preclude credit generation from higher sulfur
gasolines that could achieve large, real reductions in sulfur. The size
of the potential credit pool could be increased, perhaps dramatically,
if the trigger were relaxed or eliminated. We would like comment on
trigger values higher than 150 ppm for CG and winter RFG. We would also
request comment on expressing the trigger as a percent reduction from
baseline levels (e.g., 10-25%) rather than as an absolute value. In
addition, we request comment on a hybrid concept under which credits
would be generated for CG and winter RFG depending on initial 1997/1998
baseline sulfur levels (gasoline less than 150 ppm sulfur would
qualify, gasoline between 150 ppm and 350 ppm sulfur would need a 10-15
percent reduction, and gasoline greater than 350 ppm sulfur would need
a 15-20 percent reduction to qualify.) It would be helpful for those
suggesting the ``no-trigger'' approach to also address the issue of
equity among refiners with different baselines.
In combination with comments on the trigger, we also ask for
comment on the proposed phase-in approach. The 300 ppm cap effective
October 1, 2003 and the timing for the 30 ppm average standard would
both be important factors affecting the transition to low-sulfur
gasoline. Our analysis of the potential availability of credits
(discussed above and presented in the Draft RIA) indicates that most of
the credits needed to smooth out the transition would be generated by
low-sulfur winter RFG. Our analysis also assumes that a substantial
number of credits would be generated by refiners investing in
technology capable of producing 30 ppm gasoline prior to 2004 to ensure
compliance with the 300 ppm cap. If refiners take another approach to
meeting the 300 ppm cap (i.e., one that does not result in significant
credit generation), fewer excess credits would be available. However,
as long as some refiners invest in 30 ppm technology before 2004, we
believe sufficient credits would be available. We encourage comment on
our proposed phase-in approach.
Specifically, should the interim phase-in program be extended by an
additional year to provide an even smoother transition to the 30 ppm
standard (e.g., 120/300, 105/210, 90/180 for 2004, 2005, and 2006)?
Should the time frame for the 30 ppm average standard be shifted to
2005, for example, while retaining the 120/300 ppm caps for 2004, to
provide more time for transition to the 30 ppm standard? Should credits
expire after 2007 (as proposed) or would a shorter (or longer) credit
life be appropriate?
We are also seeking comment on a concept that would provide an
incentive to introduce clean technology early. Under this concept, any
sulfur credits generated before 2004 would be banked at a rate of 1.5
to 2.0 times the amount generated, if the annual average for that

[[Page 26062]]

refinery were equal to or less than 30 ppm and if the credits resulted
from the implementation of gasoline sulfur reduction technology
(hardware) not previously used at that refinery. This multiplier would
not be available for credits generated from modest operational changes
or product separation at the refinery or downstream. Calculation of the
un-multiplied credits would be at the refinery level. Neither domestic
refiners nor importers could qualify by segregating product or product
streams either from their refinery(ies) or in the case of importers
from one or more offshore refineries. Also, while refiners/importers
could get sulfur credits under ABT through the use of allowable
oxygenates, these could not be used as part of the basis for achieving
the 30 ppm average. EPA seeks comment on the need for and utility of
such an approach and on whether it is appropriate to encourage
implementation of sulfur control technology in this manner.

Compliance Supplement Pool

To address concerns about credit supply and the timeliness of the
availability of credits, and as a way of providing additional
flexibility, particularly to refiners that encounter unexpected
problems in complying, we are considering the concept of a government-
created and -operated compliance supplement pool for the sulfur ABT
program. Under this concept, the government would create a pool of
additional credits that could be provided to refiners/importers. This
pool would build refiner confidence that a supply of credits would be
available in the market and that credits could in fact be considered as
part of the business plan for 2004-2005 compliance. Credits from this
pool could first be made available in the 2000-2001 time frame and
perhaps in subsequent years and could only be used in 2004-2005. This
program would supplement the 2000-2003 early credit approach under ABT.
There are a number of issues related to implementing such a
program. The size of the pool potentially available for use in 2004 and
2005 would be a critical issue. A larger pool would lower the chance
that a refiner/importer could not get credits, but would reduce the
environmental benefits of the overall program. Clear rules on the
availability of credits would need to be established at the outset so
that refiners/importers could make correct investment decisions. In
addition, EPA would not want a compliance supplement pool to supplant
the need for each refiner to make aggressive efforts to comply in the
appropriate time or for a pool to create a disincentive for refiners to
generate early credits. If credits from early reductions were available
at a reasonable price, EPA would prefer that refiners/importers
purchase such credits rather than looking to a compliance supplement
pool. EPA seeks comment on the appropriate size of a compliance
supplement pool in light of these factors.
The conditions under which a refiner/importer would be eligible for
credits are important. For example, the pool could be made available
only to refiners that had demonstrated that they had made a good faith
effort to comply with the 2004 requirements, but, due to circumstances
beyond their control could not do so. Providing credits to a refiner
that failed to make good faith efforts to procure and install the
technology would create the wrong incentives and could be unfair to
competitors that had invested resources to comply.
Options for distributing credits in the pool might include granting
credits as rewards to those that generated some early reductions,
distribution based primarily or solely on need, equal distribution to
all, pro-rata distribution based on volume, making credits available at
a fixed price, or a credit auction. These approaches could be
considered singly or in combination. For example, the majority of the
compliance supplement pool could be distributed based on need, with due
consideration of the effect of lack of credits on gasoline supply in a
given area. In this case, the remaining portion might be set aside and
auctioned off to provide a price signal and a certain source of
credits.
It would seem that any such compliance pool should be administered
by the government or its agent, but decisions on credit applications
would include a public process. As part of our deliberations on this
concept we need to decide whether credits could be used to meet the
interim corporate pool averages (120/90 ppm) or just the 30 ppm
standard or both. Unlike credits generated by refiners/importers
reducing actual sulfur levels, any credits under this program would
expire after 2005.
Credits from the compliance supplement pool would be government-
created and not derived from actual reductions in gasoline sulfur. If
credits from the compliance supplement pool were distributed at little
or no cost to the receiver, such an approach might create an inequity
between those using credits and those who invested in technology to
reduce sulfur. As a means to address the potential environmental
effects of these government credits and to correct financial inequities
among refiners/importers, we seek comment on a provision that would
require those awarded these credits from the compliance supplement pool
to repay them. The credits to be used for repayment could be generated
internally in 2004-2006, purchased surplus credits from other refiners/
importers, or simply unused credits originally distributed from the
compliance supplement pool. These credits would have to be repaid by
the expiration of the period to close credit balances under the interim
program (2006, taking into account the one-year credit debt carry-
forward provision).
If, as mentioned above, credits were sold at a fixed price or
auction, several issues would arise. Should payment be through monetary
means? If so, what is EPA's authority to engage in such monetary
transactions, and what would be done with any proceeds? There is also
an issue with regard to a requirement to both buy credits for cash and
then also repay with credits. Alternatively, credits could be allocated
based on a determination that a refiner/importer needs the credits, in
conjunction with a determination regarding the refiner's/importer's
ability and willingness to repay the credits to the pool in the future
at a rate greater than 1:1. A credit auction could be held in a similar
way, that being the willingness of the bidder to repay the credits in
the future at a rate greater than 1:1. In these approaches, a refiner/
importer seeking credits might be willing to repay them at a rate of
say 1.2:1, thus essentially offering or bidding a 20 percent premium.
This could be done as a one-time premium or perhaps as a discount at
the time the credits are issued from the pools. Under this system no
money exchange would be required. This would simplify set-up of the
compliance supplement pool, allow refiners to conserve capital for
purposes of capital investment, and create an environmental return for
the compliance supplement pool. In addition, it would result in credits
being provided to refiners/importers that need them, and that are
expected to achieve additional environmental benefits in the future by
generating or purchasing excess credits.
The ``reasonableness'' of the price of credits is critical to any
approach requiring repayment from those entities using these credits.
We request comment and suggestions on ways to establish reasonable
credit prices. For example, as an upper bound, EPA might

[[Page 26063]]

set a credit price based on information received during the rulemaking
on the cost of sulfur removal for different technologies.
EPA also seeks comment on whether refiners/importers that used
credits from the compliance supplement pool should be excused from the
repayment of some or all of the credits if they could demonstrate that
it was not feasible for them to generate credits themselves and
insufficient credits were available at a reasonable price. Finally, EPA
seeks comment on how to ensure that refiners/importers that used
credits from the compliance supplement pool would in fact repay those
credits. One option would be to hold such refiners/importers liable for
failure to meet the sulfur standards over the averaging period during
which they relied on credits from the compliance supplement pool, if
such credits were not repaid in time. EPA seeks comment on this option,
as well as other alternatives that would ensure that compliance
supplement pool credits were repaid.
EPA has some experience with the compliance supplement pool
approach as part of the NO_X SIP Call (ROTR) discussed in
Section III above. In this process, a compliance supplement pool was
created to address concerns raised by industry about how the
requirements might affect the reliability of the supply of electric
power. The size of the NO_X compliance supplement pool was
created based on an EPA projection of what compliance shortfalls might
result if problems developed in implementing the control technology.
The NO_X SIP Call pool may be allocated through direct
distribution based on need or as a reward for early reductions.

Allowance-Based System

In the context of gasoline sulfur, a traditional allowance program
would provide more confidence in the availability of ``credits''
(surplus allowances) by creating sulfur budgets that the industry
(refiners and importers) would be required to meet during the 2004-5
phase-in and perhaps beyond. This budget would be created on a mass
basis using gasoline volume and the applicable regulatory standard.
This budget would then have to be allocated to individual refiners and
importers. If an individual refinery or importer had sulfur levels
below its allocation this would create surplus allowances that could be
traded. Allowances for 2004 and later would be made available in 2001.
This would facilitate the development of a market in allowances, since
those planning to beat the requirements for 2004/5 could market their
allowances early. This could significantly contribute to the certainty
that surplus allowances would be available in time for consideration by
others in their 2004 business planning.
While there are other possibilities, it would seem reasonable to
allocate the budgets to individual refiners/importers in the 2004 and
later time period based upon their individual percentages of the
gasoline market. To be consistent with other aspects of this proposal
this could be done at the corporate level in 2004/5 and at the
individual refinery/importer level in 2006 and later.
One major benefit of such an approach is that refiners/importers
could trade part or all of their 2004 and later allowances for future
use without EPA involvement and those purchasing these allowances could
do so early enough to allow a more orderly and reasoned set of capital
investment decisions. Also, since it would be allowances, not credits,
that would be traded, the seller could be held solely responsible for
failure to meet its budget without involving the buyer. The trading of
allowances would be relatively unencumbered. Allowances could be used
to meet the budgets allocated under the regulatory standard.
This approach would provide increased flexibility and certainty, it
is not clear that a large number of surplus allowances would be
created, since surplus allowances would only exist relative to a budget
based on the 30 ppm standard. Obviously the number of allowances
created in 2004 and 2005 could be increased if the budget were based on
a value higher than the 30 ppm regulatory standard, but this would
require a fundamental change in overall program design. Alternatively,
the number of surplus allowances might be increased if the allowances
program were started earlier. For example, refiners/importers could be
allocated budgets beginning in 2001 based on the product of their 1997/
1998 sulfur baselines in ppm (with appropriate adjustments for RFG
Phase II) and their gasoline volume. Any reductions in the average
sulfur levels or volume from the baseline level during that 2001-2003
time period would result in surplus allowances.
While the idea of pre-2004 allowances has merit, it requires the de
facto implementation of a standard before 2004 (since each refiner's/
importer's budget would in effect be a standard), in order to establish
allowances. And, in contrast to the ABT program where participation is
voluntary and no requirements exist before 2004, an allowance system
would require refiners subject to the allowance program to hold
sufficient allowances to cover their calculated mass emissions starting
in 2001.
In principle, an allowance system could be designed to incorporate
all of the features of an ABT credit system as described above. We are
interested in comment on the viability of such an allowance program as
an alternative to the traditional ABT program and whether such a
program would have to be mandatory for all refiners/importers in order
to be effective. For example, could we structure an allowance program
such that the refiner opts into if it intends to generate or use
allowances or opts out of if it does not? We are also interested in
comment on the parameters of such a program, including the appropriate
budget levels, methods for distributing the budgets to refiners/
importers, and whether allowances could be used to meet the corporate
pool averages, the regulatory standard, or both. As with the ABT
program, we would like to hear your views on the years over which such
a program should apply (e.g., should it start in 2001?, should it
extend beyond 2005?), as well as the other regulatory requirements that
should apply in each year.
We also request comment on whether the allowance program could be
established as a supplement to the credit program. If an allowance
program is implemented along with a compliance supplement pool and/or
early ABT we are interested in comments on how to make credits fully
exchangeable among the programs. We are also interested in comments on
how the programs could/should be integrated. For example, could we let
a refiner/importer generate early ABT credits and at the same time sell
2004-2005 allowances?

Reserved Credits

EPA is also aware of concerns regarding whether refiners that
earned or received credits would make them available in a timely manner
to those that needed them, particularly to small- to mid-sized
refiners/importers. If an adequate number of credits were not available
in a timely manner and for a reasonable price, small- to mid-size
refiners would have no choice but to pursue near term capital
investment to comply in 2004. This might be the appropriate course for
many of these refineries, but we do not think it is appropriate for
them to be precluded from the same flexibility as larger refineries.
We are seeking comment on whether we should require that a set
percentage (e.g., 1015%) of all credits generated in early ABT (2000-
2003), awarded

[[Page 26064]]

through the compliance supplement pool, or earned through the
allowance-based approach either must be retired or offered for trade
outside of the refining company that originally generated or was
granted them. Under such a provision, refiners/importers would be
required to set aside a percentage of credits/allowances they generate,
but could choose whether to retire them or offer them for sale at a
fair market price to another refiner/importer. Regardless of which
option the refiner/importer chose, the results would be beneficial--the
environment would benefit if credits are retired, and credit
availability would improve if the refiner chose to sell credits. We are
also interested in your views as to how this objective might be
accomplished.
EPA also asks comment on the disposition of credits that were put
up for trade one or more times during the period 2004-2006 but did not
sell during that period. This could be the case if a credit owner
offered credits for sale at a price in excess of fair market value and
thus they were not purchased by another party or if credit supply
significantly exceed demand. In this kind of situation, should the
credits be retired or revert to the generator at a full or reduced rate
(e.g., 50%) for future use in compliance determinations? We request
comment on whether such a provision for reserved credits would be
needed by small- to mid-sized refiners and whether the reservation of
10-15 percent of credits would be sufficient to address the concerns.
We also seek comment on whether such a pool should be supplemented by
the government through an auction to ensure that the pool size is
adequate and whether such a pool could be useful in helping to
establish a market price for company owned credits.
b. Refinery Air Pollution Permitting Requirements. As discussed
previously in this document, this proposed program would result in
significant emission reductions from reducing sulfur in gasoline
nationally, through the emission reductions from the current fleet of
vehicles and ensuring the efficacy of new technologies in future
vehicles. In order to achieve this environmental benefit as soon as
possible, we want to be sure the public is aware of the full range of
available methods for expediting permits required for refinery process
changes to reduce gasoline sulfur. Expedited permitting also will
facilitate refiners' ability to generate sulfur credits, under today's
proposed sulfur Averaging, Banking and Trading program, described in
the previous section.
There are two key Clean Air Act permitting programs that refiners
must comply with when making changes at their existing facilities to
implement gasoline sulfur control--the New Source Review (NSR) program
and the Title V operating permit program. Typically, both of these
programs are administered by state/local permitting agencies, with EPA
oversight. While the basic requirements of these programs are dictated
by the Clean Air Act and EPA regulations, the specific requirements of
each state/local permitting program may vary.
We recognize that compliance with these air permitting requirements
is an integral component in any plan to implement the gasoline sulfur
control program under the schedule proposed today. To help refiners
meet the permit requirements, below we discuss the possible mechanisms
to address the substantive requirements of the major NSR and Title V
programs, including possible opportunities to streamline and expedite
the processing of permit applications. Finally, we conclude this
section by discussing possible tools that we are currently testing in
the experimental Pollution Prevention in Permitting Program (P4), which
promotes permit streamlining and flexibility for Title V operating
permits, along with increased pollution prevention activities. We
encourage commenters to provide suggestions for additional
opportunities to streamline the permitting process to accommodate the
implementation of the proposed gasoline desulfurization requirements
for the refining industry sector.
The American Petroleum Institute (API) has sent a letter to EPA
outlining its concerns about the potential impact of various permitting
requirements on the industry's ability to meet future gasoline sulfur
standards, as well as their suggested options for permit
streamlining.⁵⁴ This letter is included in the docket for
this rulemaking. We are aware that individual refineries are in
different situations regarding the modification to current operation
that would be needed to meet the proposed sulfur standard and the
regulatory requirements applicable to those modifications. Based on the
limited information available at present, some refineries may not
increase emissions significantly, and others may find it most
economical to make on-site emission reductions at the plant to avoid
emission increases. Accordingly, we request comment on the extent to
which the various mechanisms to streamline the permitting process
discussed in this section are in fact needed or useful. We request that
commenters supporting such streamlining describe the specific refiner
situations in which they believe streamlining is needed, and encourage
them to provide any suggestions for additional opportunities to
streamline the permit process to expedite refineries' preparation to
meet the proposed sulfur standards.
---------------------------------------------------------------------------

\54\ Letter from William F. O'Keefe, Executive Vice President,
American Petroleum Institute, to Bruce Jordan, U.S. EPA, Office of
Air Quality Planning and Standards, dated February 12, 1999 (Docket
item IIG-304).
---------------------------------------------------------------------------

i. New Source Review Program.
The New Source Review (NSR) program,⁵⁵ as it applies to
existing major sources of air pollution, requires that a
preconstruction permit be issued before a source begins construction of
any project that would result in a significant net emissions increase.
With respect to NSR, we anticipate that refineries will fall into one
of two categories if the proposed sulfur standards are implemented. The
first category consists of those refineries that would be able to avoid
major NSR by demonstrating that the physical and operational changes
needed to reduce gasoline sulfur do not result in a net emission
increase of the quantity that would require a major NSR permit. Major
NSR would not apply where: (1) The proposed changes would not result in
an emissions increase at the refinery; (2) the increase is, in and of
itself, less than ``significant'' ⁵⁶; or (3) the refinery
``nets'' the project out of review. In most cases, even where a
refinery change to accommodate the production of lower sulfur gasoline
does not trigger the major source NSR program, the project still will
be subject to a state's general, or ``minor,'' NSR
program.⁵⁷ The second category consists of those refineries
that would experience a significant net emissions increase as a result
of process changes necessary to accommodate gasoline sulfur control
and, therefore, will trigger major NSR applicability and the attendant
permit process (e.g., nonattainment NSR or Prevention of Significant
Deterioration). Accordingly, such facilities must obtain a major source
preconstruction permit prior to making these process changes.
---------------------------------------------------------------------------

\55\ See 40 CFR 51.165, 40 CFR 51.166, 40 CFR 52.21, 42 U.S.C.
7475, and 42 U.S.C. 7503.
\56\ EPA's and state/local regulations for major NSR define
``significance'' levels for various pollutants.
\57\ This permitting program applies to the construction or
modification of any stationary source. See 40 CFR 51.160 and 42
U.S.C. 7410(a)(2)(C).
---------------------------------------------------------------------------

As described previously in today's document, there are several
types of process changes refineries could make to meet the proposed
gasoline sulfur

[[Page 26065]]

levels. Traditional sulfur removal technologies include installing a
hydrocracker upstream, or a hydrotreater upstream or downstream, of the
fluidized catalytic cracker (FCC) unit, the unit that produces the
largest fraction of gasoline. There also are improved desulfurization
technologies, CDHydro and CDHDS (licensed by the company CDTECH) and
OCTGAIN 220 (licensed by Mobil Oil). These technologies use
conventional refining processes combined in new ways, with either
improved catalysts or other design changes to maximize gasoline
desulfurization effectiveness with minimal negative effects, such as
octane loss. To different degrees, all these technologies involve the
use of a furnace and, thus, have the potential to increase pollutants
associated with combustion, such as NO_X, VOCs, PM, CO, and
SO<INF>2</INF>. The addition of these technologies also could result in
equipment leaks of petroleum compounds, which could increase emissions
of VOCs and other pollutants. It also is possible that the increased
removal of sulfur from the gasoline stream might require increased
capacity of a number of refinery processes, such as the sulfur recovery
unit (SRU), which converts hydrogen sulfide into elemental sulfur and
is associated with SO<INF>2</INF> emissions. The emission increase
associated with a desulfurization project will vary from refinery to
refinery, depending on a number of source-specific factors, such as the
specific refinery configuration, choice of desulfurization technology,
amount of gasoline production, and type of fuel used to fire the
furnace.
While we do not have sufficient information at this time to
estimate the number of refineries nationwide that will trigger major
NSR, we believe it could be substantial, given that over 100 refineries
in the country would be required to make desulfurization process
changes under today's proposal. Estimates from one vendor indicate that
its desulfurization process could result in emission increases that are
considered ``significant'' in severe ozone nonattainment areas (i.e.,
greater than 25 tons/year of NO_X and VOC), which would
trigger major source nonattainment NSR review. Since the significance
threshold generally is lower in certain nonattainment areas (i.e.,
those nonattainment areas classified as serious and above for ozone),
refineries located in those nonattainment areas may be the most likely
to trigger major NSR review. There are many refineries located in ozone
nonattainment areas (e.g., parts of the Gulf Coast).

NSR Applicability Principles

A refiner's ability to avoid triggering major NSR by keeping
emission increases below the major NSR applicability cutoffs will
depend primarily on the case-by-case circumstances of each refinery.
Nevertheless, numerous means by which a source can otherwise legally
avoid major NSR permitting are available to all refineries for
consideration and possible use. In addition, as discussed below, the
Agency is prepared to work with refineries to explore the use of
certain NSR applicability mechanisms (i.e., plant wide applicability
limits or ``PALs''), where appropriate.
To the extent needed, we intend to work with state/local permitting
authorities to provide assistance with the proper application of the
NSR rules on an expedited basis for permits involving refinery
desulfurization projects. We want to ensure that applicability
decisions are made at the earliest possible opportunity and consider
the full spectrum of options available so that a refiner can adjust, or
possibly reconfigure, planned desulfurization projects so as to prevent
significant emission increases and thereby avoid major NSR within the
framework of the current regulations. In addition, timely applicability
decisions will provide added certainty as to the applicable NSR
requirements and, where a major NSR permit is needed, how to best to
expedite the issuance of a permit.
Depending on the nature of the physical or operational changes
necessary to accommodate desulfurization projects, the NSR
applicability process for major modifications can be a complex and time
consuming exercise. The NSR regulatory provisions require that a
proposed physical change result in a significant net emissions increase
in order for the change to be considered a modification and therefore
subject to NSR. We expect that there likely will be questions regarding
which, and how, existing emission units are affected by the change,
including how to calculate the magnitude of the emissions change for
major NSR applicability purposes. We are committed to working with
refiners and state/local air pollution control agencies to clarify and
ensure that, in applicability analyses for gasoline desulfurization
projects, only those emissions increases resulting from the physical or
operational changes necessary to comply with gasoline desulfurization
requirements are included in the applicability analysis.
In doing an applicability analysis for major NSR, refineries should
analyze their past, current, and future operations and emissions to
determine whether it is possible to avoid major NSR based upon their
facility-specific circumstances, including the use of previous emission
reductions at the facility to ``net'' out of NSR. Similarly, sources
might avoid NSR by using Plantwide Applicability Limits (PALs) to cap
emissions. Emissions netting is a term that refers to the process of
considering certain previous and prospective emission changes at an
existing major source to determine if a net emissions increase will
result from the proposed new project. Where the sum total of creditable
increases and decreases across the refinery is less than significant,
major NSR would not apply. In addition, if the proposed emissions
increase from a proposed project (in this case, a project undertaken to
reduce gasoline sulfur levels) is by itself, without considering any
decreases, less than significant, major NSR would also not apply.
PALs may provide another opportunity for refineries to avoid
triggering major NSR applicability. The voluntary, source-specific PAL
is a straightforward, flexible approach to determine whether changes at
an existing major source of air pollution result in a significant net
emissions increase. By restricting (or ``capping'') a facility's
emissions to a level representative of current actual emissions, a PAL
allows a source to change operations and equipment without having to
undergo major NSR permitting. For example, as long as refinery
activities do not result in emissions above the PAL cap level, the
refinery would not be subject to major NSR, regardless of the nature of
the activity. Under a PAL, instead of a case-by-case assessment of
whether a proposed change is subject to or excluded from major NSR, the
refinery manager knows that as long as the refinery stays within its
emissions cap, major NSR will not be triggered. Production units may be
started and stopped, production lines reconfigured, and products
changed and revamped without delay from major NSR permitting.
Because of these advantages, the Agency previously has proposed to
incorporate PALs in all of its NSR regulations (see 61 FR 38250, 38264,
July 23, 1996), and has worked with state permitting authorities to
develop PALs for individual sources. Likewise, the Agency is committed
to exploring the propriety of authorizing PALs for refineries subject
to the final gasoline

[[Page 26066]]

sulfur control rules. We are examining our authorities to assure they
support these approaches. Should it be necessary, EPA stands prepared
to issue final regulations to make PALs available to sources making
changes to comply with these gasoline sulfur control requirements.
We are further committed to investigating with affected refineries
whether a PAL might be a valuable tool for managing a number of other
Clean Air Act requirements. For instance, depending on the relevant
state rules, a PAL also could include terms that allow facility changes
to be made without triggering minor NSR. It is our experience that, in
the cases where PALs have been applied, both industry and air pollution
regulators have benefitted from the regulatory certainty and simplicity
a PAL provides. The use of a PAL can enhance a refinery's ability to
make appropriately designated changes quickly, without having to
evaluate a baseline for each modification, determine the
contemporaneous increases and decreases, and engage in other time-
consuming netting procedures required under the major NSR program on a
case-by-case basis. A PAL also can encourage a source to reduce
emissions voluntarily (e.g., from pollution prevention or other
emission reduction efforts), so that it has sufficient room for growth
(under the PAL) to accommodate increased emissions from future process
changes.

Approaches to Expedite the Processing of NSR Permit Applications

Notwithstanding the availability of the major NSR applicability
principles and mechanisms discussed above, we anticipate that it will
not be possible for all refineries subject to the gasoline
desulfurization requirements to prevent significant emission increases
and avoid major NSR. Additionally, even those facilities that are able
to avoid major NSR likely will be required to obtain a state minor NSR
permit. For facilities subject to major NSR, the timing of permit
issuance could vary depending on many factors, including the complexity
of process changes, the type of permit required, air quality impact,
control technology reviews, and the state's overall permit workload. It
is not uncommon for issuance of a major source preconstruction permit
to take six to 12 months from the receipt of a source's complete permit
application. In addition, determining the applicable permitting
requirements for refineries is often complex, due to the wide array of
emission points and processes.
To help expedite the NSR permitting process, we suggest the
following streamlining approaches. Since state/local governments
typically are the lead permitting agencies, we will work closely with
them on any of these efforts. We solicit comments on the efficacy of
these approaches and opportunities for additional streamlining. We are
particularly interested in understanding whether these permit
streamlining approaches could enable refineries to begin voluntarily
producing lower-sulfur gasoline earlier than the compliance dates
proposed today, so that the environmental benefits may be realized
sooner than 2004 and ABT credits (see previous Section) could be
generated.
<bullet> Federal guidance on streamlining certain major NSR
permitting requirements, such as control technology and compliance
parameters. Although the major NSR permit is a case- and source-
specific evaluation, we could provide guidance on certain aspects of
refinery projects designed to reduce fuel sulfur that share a common
requirement or circumstance. For example, for refinery projects
permitted in the same time frame, the Lowest Achievable Emission Rate
(LAER) requirement should be the same for identical emissions units
regardless of the location of the individual refinery. In this case, we
could define for the industry what emissions levels would be expected
to meet LAER and provide model permit conditions, including appropriate
monitoring, record keeping, and reporting. Although Best Available
Control Technology (BACT) determinations require case-by-case
considerations, we also could issue guidance setting out a level of
emissions that, in our view, satisfies BACT for the class or category
of emission units associated with refinery desulfurization. We expect
that providing BACT and LAER guidance would help to expedite major
source permitting and add more certainty to the permit process.
Consequently, for any applications processed within a discrete time
frame, a presumptive federal LAER and/or BACT could be established.
<bullet> Availability of offsets. The major NSR permitting
provisions require that a significant emissions increase of
nonattainment pollutants must be offset by emission reductions from
other sources. We solicit comment on the need for offsets by refineries
making modifications to meet the proposed sulfur standards, and the
expected size or volume of any offsets that may be necessary. In
addition, to the extent offsets may be useful or necessary, EPA
requests comment on whether on-site emissions reductions at the
refinery could be used to avoid the expected emissions increases that
would otherwise occur. We will work with refiners and state/local air
pollution control agencies to explore options and possible new
approaches that would help ensure the availability of offsets. For
example, it may be possible to establish pre-funded offset pools,
designed specifically for offsetting emissions increases resulting from
gasoline desulfurization projects. We believe that the establishment of
preapproved offset banks or pools could greatly expedite permitting in
nonattainment areas.
To help give certainty that offsets will be available, we seek
comment on how and whether emission reductions resulting from vehicles
operated on low sulfur gasoline could be used as offsets by refineries
implementing gasoline sulfur controls. For example, it may be possible
for a state, within a given nonattainment area, to set aside a portion
of the emission reductions expected from vehicles operating on low
sulfur gasoline and dedicate those reductions for use as offsets by
refineries. These offsets would have to meet all the criteria currently
established for being creditable, and could not be ``double-counted''
by the state for other SIP planning purposes. We request comment on the
ability of emission reductions from the use of low sulfur gasoline to
meet the Clean Air Act's criteria for creditable offsets for NSR
purposes. Since securing offsets can be a significant challenge to
sources undergoing major NSR permitting in nonattainment areas, we
believe this approach could substantially speed up, and add certainty
to, the permitting process. We believe this approach is worth
evaluating, given the enormous emission reductions resulting from the
use of low sulfur gasoline, and given that some refineries will trigger
major NSR solely as a result of the process changes needed to produce
this new gasoline. Finally, EPA seeks comment on whether providing the
ability to use the emissions reductions resulting from the use of low
sulfur gasoline in vehicles as offsets for refineries producing low
sulfur gasoline can be limited to this specific situation.
Specifically, EPA requests comment on the concern that providing this
option to refineries would allow the use of such emissions reductions
as offsets for other stationary sources.
As discussed above, we believe that refineries in ozone
nonattainment areas could be the most likely to trigger major NSR
review, based on net emission increases of NO_X and/or VOCs.
The proposed Tier 2/gasoline sulfur control program is expected to
result in over

[[Page 26067]]

500,000 tons of NO_X reductions and over 100,000 tons of VOC
reductions nationwide in 2004 (the first year of implementation), as
well as substantial reductions in particulate matter and sulfur
dioxide, as described elsewhere in this document and the draft
Regulatory Impact Analysis.⁵⁸ In a given nonattainment area,
the program could result in hundreds to thousands of tons of
NO_X and VOC reductions, depending on the inventory of cars
and light-trucks in the area. For example, for the New York
metropolitan area, EPA projects NO_X emission reductions of
7,344 tons and VOC emission reductions of 1,285 tons in 2004 resulting
from the proposed Tier 2/gasoline sulfur control program.⁵⁹
We anticipate that only a small fraction of these total emission
reductions in a given area would be needed for use as offsets for
refineries implementing gasoline sulfur control projects.
---------------------------------------------------------------------------

\58\ Although these emission reduction estimates are for the
combined Tier 2 emission standards/gasoline sulfur control program,
in 2004, nearly all these emission reductions would be attributed
solely to vehicles fueled by low sulfur gasoline, since vehicles
meeting the Tier 2 emission standards would comprise only a small
fraction of the vehicle fleet.
\59\ See draft Regulatory Impact Analysis, Chapter III.
---------------------------------------------------------------------------

<bullet> Model permits and permit applications. It may be possible
to develop an individual, or series of, model permits or permit
applications for gasoline desulfurization projects. Rather than each
individual refinery having to develop its own permit application from
scratch, a generic permit application form could be developed to
address common issues. To file a major source application, a refinery
would only need to fill in the blanks as they may relate to case-
specific assessments, such as air quality impacts. Similarly, a model
permit could contain all necessary compliance measures avoiding the
time spent in developing individual permit conditions. Model permits or
permit applications would serve as templates, thereby eliminating much
of the time and uncertainty associated with processing each
application.
<bullet> EPA refinery permitting teams. We could establish a team
of experts to be available as a resource, as needed, to refineries and
state/local agencies to troubleshoot permitting issues that may develop
with individual applications. The team could be made up of EPA
permitting experts empowered to make decisions and resolve issues
quickly.
In addition to the above opportunities to streamline the permitting
process, we encourage states to process a refinery's request to
implement changes at a facility to meet gasoline desulfurization
requirements as a priority and on an expedited basis. Priority
treatment, in combination with the above opportunities to streamline
the process, would ensure that permit applications associated with
gasoline desulfurization changes are processed as expeditiously as
possible. Given the enormous environmental benefits that we estimate
would be achieved as a result of the proposed gasoline sulfur control
requirements, we believe such expedited and special processing is
appropriate.
ii. Title V Operating Permit Program.
We recognize that the changes to be made by refiners to implement
gasoline sulfur controls typically would involve not only NSR
preconstruction permitting requirements but also those of the title V
operating permit program. Title V requires owners or operators of
``major'' and certain other sources to obtain an operating permit--a
document that identifies all emissions units, their applicable
requirements as developed in accordance with the Clean Air Act, and
monitoring and other permit conditions to provide a reasonable
assurance of compliance with each of the applicable requirements on an
ongoing basis. Most of the refiners likely are ``major'' sources
subject to title V, due to their plant-wide level of emissions. As with
other process changes, prior to implementing gasoline sulfur controls,
refiners would need to work with their state, local, or tribal
permitting agency to determine what requirements apply and what changes
might be required to the source's title V permit application or permit
(if one has been issued).
A critical element of any successful title V permitting strategy to
accomplish the necessary desulfurization is how best to integrate the
procedural and substantive requirements of the title V and NSR permit
programs. We believe the title V permitting process provides an
excellent opportunity to accomplish this integration and to impart
greater certainty into the ultimate approvability of a gasoline
desulfurization project under both permit programs. Depending on a
specific permitting authority's program and when the desulfurization
activity would occur relative to the issuance of the refinery's initial
title V permit, the NSR preconstruction permit and the title V permit
processes might be done in parallel or in sequence.
Where the title V permit is issued before the desulfurization
activity commences, this permit must be updated before operation of the
changes that would also be subject to NSR. In this case, we suggest
that the preconstruction permit review process, managed by the
permitting authority, be merged with the title V permit revision
process so as to satisfy the procedural safeguards and the same
substantive requirements of the NSR and title V programs at the same
time.⁶⁰ If this is done, the title V permit may be
administratively amended to incorporate the contents of the NSR permit
prior to operation of the desulfurization process changes. Where the
appropriate NSR action (major or minor) approving the desulfurization
changes precedes the issuance of a source's initial title V permit, the
applicable NSR process can still be ``enhanced'' to address title V
obligations. Here, in order to determine approvability under both title
V and NSR, the permitting authority can issue a separate title V permit
specifically for the desulfurization project in advance of the title V
permit that will be issued subsequently for the rest of the site.
Finally, if issuance of the title V permit issuance for the entire
source would precede the NSR construction, depending on several
factors, the permitting authority could conduct simultaneous permit
processes to accomplish preconstruction approval of the desulfurization
project and title V approval for the operation of the project in
conjunction with the entire refinery source.
---------------------------------------------------------------------------

\60\ The concept of a merged NSR/title V process refers to the
combination of the title V review process with any otherwise
applicable state preconstruction review process, where such process
satisfies the procedural requirements of the title V's permit
revision, permit review, and public participation provisions.
Example state review processes that may be eligible for merger
include, but are not limited to, preconstruction review of major or
minor NSR, source-specialized State Implementation Plan revisions,
and procedures implementing section 112(g) of the Clean Air Act.
Under a merged process, activities are only presented in a public
forum once, rather than in sequence, to avoid duplication of
process. Upon completion of the merged process, a successful project
would have met all federal permitting requirements, including review
by the public, EPA and affected States, and opportunities for EPA
objection and public petition, and can implement both processes
without delay. Qualifying activities that have received
preconstruction review permits meeting the requirements of 40 CFR
70.7(d)(1)(v) may be incorporated into title V permits as
administrative permit amendments.
---------------------------------------------------------------------------

Beyond synchronizing when the two permit programs would be
implemented, we recommend that permitting authorities take approaches
in the substantive permitting of the desulfurization projects that will
both assure compliance with all applicable air requirements and result
in a more flexible and efficient permit design. We encourage that the
approaches in the

[[Page 26068]]

title V ``White Papers'' ⁶¹ be considered to focus both the
content of title V applications and permits. In particular, we
recommend that permitting authorities and owners or operators of
refineries consider the ``streamlining'' of multiple applicable
requirements applying to the same project. Under the streamlining
concept, where multiple applicable requirements apply to the same
emission unit(s), the permitting authority may develop one emission
limit (with associated monitoring, recordkeeping, and reporting) that
assures compliance with all applicable requirements. For example,
several aspects of the control requirements necessary to implement our
maximum available control technology (MACT) and new source performance
standards (NSPS) requirements, State Implementation Plan (SIP), and NSR
programs (including both major and minor NSR, as applicable) could be
considered for streamlining per White Paper Number 2. Where successful,
this streamlining will result in a single control requirement (or
emission limit), coupled with appropriate monitoring, recordkeeping,
reporting, and testing requirements that yield a reasonable assurance
of compliance for all subsumed requirements.⁶²
---------------------------------------------------------------------------

\61\ White Paper for Streamlined Development of Part 70 Permit
Applications, Lydia N. Wegman, Deputy Director, Office of Air
Quality Planning and Standards, U.S. EPA, July 10, 1995 and White
Paper Number 2 for Improved Implementation of the Part 70 Operating
Permits Program, Lydia N. Wegman, Deputy Director, Office of Air
Quality Planning and Standards, U.S. EPA, March 5, 1996.
\62\ See Section II.A. of White Paper Number 2.
---------------------------------------------------------------------------

We also are willing to explore applying to the varying situations
of sulfur removal at refineries certain permit design approaches that
have previously been limited to some permitting pilot projects. In
particular, in partnership with permitting authorities, we have been
working with selected industries at specific sites to conduct Pollution
Prevention in Permitting Project (P4) pilots. These projects respond to
the Administration's goals for reinvention in order to implement
environmental permit programs in a more streamlined fashion, while
assuring required levels of environmental protection. Based on our
prior experience with these regulatory reinvention projects, permit
design options for refiners implementing gasoline desulfurization
projects might include, but are not limited to, any of the following
approaches:
<bullet> Advance approvals of certain types of changes in title V,
including those subject to minor NSR.# ⁶³
---------------------------------------------------------------------------

\63\ Advance approval means that a particular project (or class
of projects) like one to accomplish gasoline desulfurization and its
support activities would be preapproved for title V purposes before
its actual construction, provided that the terms of the title V
permit governing the advance approval are met. The Agency has a
possible non-binding interpretation of the Title V regulations that
would provide for the advance approval of certain new emission units
and control devices. See 63 FR 50279, 50315-20 (Sept. 21, 1998)
(Section IV.L., Permitting and Compliance Options/Change Management
Strategy, in National Emission Standards for Hazardous Air
Pollutants for Source Categories: Pharmaceuticals Production).
---------------------------------------------------------------------------

<bullet> Provisions that where met would prevent another
requirement from applying (e.g., plant wide applicability limits (as
noted above) to address potential major NSR applicability).
<bullet> Model permit conditions, such as a presumptive,
streamlined approach to meet all applicable control technology
requirements to expedite permitting decisions, where applicable.
<bullet> Adding terms to a title V permit so as to preauthorize a
faster permit revision process where one is necessary to add further
details within an approved approach (e.g., the minor instead of
significant permit modification process).
<bullet> Permitting the worst-case emissions scenario to address
all applicable requirements applying in a range of possible operating
scenarios or to prevent certain requirements from applying.
<bullet> Permitting alternative compliance options where an owner
or operator of a source needs the flexibility to vary the compliance
approach with changing refinery conditions.
<bullet> Using pollution prevention approaches to facilitate
compliance with applicable requirements and/or required permit terms.
We recognize that the situations for refineries affected by the
proposed gasoline sulfur control program can vary widely (e.g., sulfur
level in the gasoline, size of the stream, air quality status of the
area, etc.), and that the actual permit approach for an individual
refinery may be a combination of certain options outlined above and
previously for streamlining NSR. Any title V approach must, however,
assure compliance with all applicable requirements linked to the
necessary construction and provide a meaningful opportunity for all
affected parties to review the appropriateness of a proposed approach
as it would apply to a particular site. For example, where new
desulfurization units would be required and would be well controlled so
as to result in emissions below the threshold for triggering major NSR,
then an advance approval of minor NSR requirements in combination with
certain operationally limiting conditions might be an appropriate
strategy. Where the addition of such a unit would trigger major NSR,
then the strategies that combine the reviews and streamline the
requirements of both title V and major NSR offer promise. In a few
cases, reblending of high sulfur gasoline blend stocks, blending in low
sulfur oxygenates, or using sweeter crude oil might be sufficient to
achieve the necessary sulfur reductions and require few, if any,
additional title V permit terms to implement.
iii. EPA Assistance to Explore Permit Streamlining Options and
Solicitation of Comment.
We are committed to exploring the possible approaches described
above. Accordingly, if there is sufficient interest and need, as
expressed in comments on this proposed rule, within the refining
industry and among state permitting authorities, we will hold a P4/
flexible permit workshop focused on the permitting of the refining
industry arising from the gasoline desulfurization program.
Additionally, should a permitting authority and owners or operators of
affected facilities within a common jurisdiction express a desire for a
specific flexible permit project aimed at the development of permit
language to facilitate refinery activities to reduce gasoline sulfur,
then in accordance with already established principles for initiating
similar permit projects, we would be willing to work with a designated
refinery. We intend that the approaches derived from such efforts could
then serve as a template as needed for use by other refineries and
state permitting authorities, provided the approaches are modified to
conform with all applicable state title V and NSR requirements.
We believe that application of one or more of the approaches
described in today's document would reduce any burden of meeting NSR
permit requirements and revisions to title V permit applications or
permits to incorporate the gasoline desulfurization requirements
adopted in the final rule. However, the use of one or more of these
approaches would have accompanying resource requirements. For example,
it is possible that the initial resources required to establish a PAL,
and the attendant monitoring, recordkeeping and reporting requirements,
could involve as much time and resources as associated with a typical
NSR permit. However, once established, a PAL could provide more
flexibility and minimize future resource demands than more traditional
permit approaches. Accordingly, we request that permitting authorities,
owners or operators of affected facilities, and the public comment on
whether use of the

[[Page 26069]]

approaches described in today's document will achieve appropriate
streamlining of controls and requirements arising out of this rule and
meet the objectives of the NSR and title V permitting programs.
c. Should Hardship Relief Be Available? Elsewhere in this document
(Section IV.C.3.b.), we propose a hardship provision that would apply
to small refiners. EPA seeks additional comment on whether it should
adopt a hardship provision allowing for compliance with standards less
stringent than those proposed today during the early years of the
program. While EPA believes that it is feasible for most refiners to
meet the proposed standard by 2004, the Agency is seeking comment on
whether it may be appropriate to allow refiners with substantial
economic hardship circumstances to apply for relief from compliance
with the sulfur standard for a limited time period.
Such a hardship provision would need to contain appropriate
criteria to limit the provision to a narrowly drawn set of
circumstances. This might include criteria such as ability to raise
capital to make necessary refinery investments in time for 2004, given
the current size and ownership of the refinery, the physical
characteristics of the refinery, the volume of gasoline at issue,
ability to purchase credits to comply, and any efforts by the refiner
to limit sulfur that are already underway or have been attempted. The
provision would also need to contain criteria to ensure that it would
not undermine the emissions reduction goals of the Tier 2/sulfur
program and would not allow large amounts of gasoline with sulfur
levels significantly above 30 ppm into the market. For example, this
might include a volume limit on the use of less stringent standards in
hardship circumstances. It would also need to include an endpoint, so
that the relief is short-term and the refinery would then have to meet
the same standard as all other refineries. For example, EPA would not
expect that hardship relief will be needed beyond 2009.
Under such a provision, we expect that refiners would be subject to
a reasonable level of control, albeit less stringent than the proposed
standards. At a minimum, sulfur levels at a particular refinery should
not be permitted to be higher than 1997-1998 baseline levels and in no
event should the average sulfur level be greater than 300 ppm. EPA also
seeks comment on the appropriate time frame for allowing relief in
hardship circumstances. EPA solicits comments on whether any refiners
would encounter significant hardship in meeting the proposed standard.
EPA solicits comment on the implications of any such hardship provision
on small refiners and its relationship to the small refiner provisions
proposed in this document. Finally, EPA seeks comment on the
implications of a hardship provision on the proposed ABT program.
5. Consideration of Diesel Fuel Control
As explained in Section IV.B. above, the proposed Tier 2 standards
would apply to both gasoline- and diesel fuel-fueled vehicles.
Currently very few light-duty vehicles operate on diesel fuel. Given
what we know about gasoline vehicles, we believe it is reasonable to
anticipate that the use of exhaust aftertreatment devices may be
required, and that these technologies may have similar sensitivities to
sulfur that the catalysts used on gasoline engines have. However, we do
not yet have enough information to be able to conclude that diesel
sulfur levels need to be reduced in the same time frame that Tier 2
vehicles are introduced. A decision to require reductions in diesel
sulfur levels could have significant implications for the refining
industry, both because it would likely require capital expenditures
over and above the significant costs that would be incurred in
controlling gasoline sulfur, and because for some refiners concurrent
control of gasoline and diesel sulfur may be the most economical
solution. Hence, due to the implications for automotive manufacturers
and for diesel fuel producers, a decision on whether to require diesel
fuel sulfur reductions needs to be made as soon as possible.
Automobile and diesel engine manufacturers and state air quality
agencies have recently asked us to set new fuel quality requirements
for diesel fuel used in highway vehicles.⁶⁴ The
manufacturers believe that such requirements, especially controlling
diesel fuel sulfur content to very low levels, could produce large
environmental benefits by enabling dramatically lower-emitting diesel
engines equipped with exhaust aftertreatment devices. The viability of
such technologies would, of course, affect the feasibility of the
proposed Tier 2 emission standards for diesel vehicles. Currently,
highway diesel fuel is regulated under standards we set in 1990. These
standards, which became effective in 1993, limit the concentration of
sulfur in diesel fuel to a maximum of 500 ppm; they also control the
amount of aromatic compounds in the fuel (55 FR 34120, August 21,
1990).
---------------------------------------------------------------------------

\64\ See the following contained in the docket for this
rulemaking: Letter from Robert J. Eaton, Chrysler Corporation, Alex
Trotman, Ford Motor Company and John F. Smith, Jr., General Motors
Corporation, to Vice President Al Gore, July 16, 1998; ``STAPPA/
ALAPCO Resolution on Sulfur in Diesel Fuel,'' October 13, 1998;
Letter from S. William Becker, Executive Director of STAPPA/ALAPCO,
to Carol Browner, Administrator of U.S. EPA, October 16, 1998;
Letter from Jed R. Mandel, Engine Manufacturers Association, to
Margo T. Oge, Director, Office of Mobile Sources, EPA, November 6,
1998.
---------------------------------------------------------------------------

Diesel engine manufacturers have argued that implementing Tier 2
standards without concurrent diesel fuel changes would be unfair to
diesels because diesel fuel quality is worse than gasoline fuel
quality, especially considering that the Tier 2 rulemaking includes
proposed improvements in gasoline quality to enable advanced three-way
catalytic converters. Some argue that, beyond fuel-neutrality
considerations, diesel fuel quality improvement is needed to combat
global warming because it will facilitate the marketing of more diesel
vehicles and, in their opinion, thereby reduce emissions of global
warming gases. Others counter that such benefits are illusory and that
diesel vehicles should be discouraged because diesel exhaust is a
serious health hazard, a hazard that improvements in fuel quality would
do little to mitigate.
To address the issue of diesel fuel changes, we will issue an
Advance Notice of Proposed Rulemaking (ANPRM) in the near future. We
encourage interested parties to review and comment on the issues raised
in the ANPRM. On the basis of this information, if appropriate, we plan
to publish a proposal on standards for diesel fuel in the next several
months. This would provide some degree of clarity regarding our plans
in this area in time to help affected industries to then make their own
plans without undue disruption. This is especially important for the
petroleum refining industry in planning capital outlays to accomplish
sulfur reduction in gasoline, and potentially diesel fuel, at the most
economical point in the refining process.
Several diesel vehicle manufacturers have raised the concern that
unless or until lower sulfur diesel fuel is available, the sulfate
component of diesel PM may be particularly difficult to control to very
low emission levels. They have encouraged us to express the proposed PM
standards in terms of non-sulfate PM to provide manufacturers
flexibility in how they balance the control of sulfate and non-sulfate
PM components.

[[Page 26070]]

We request comment on such an approach, including specific comments
on the following:
<bullet> Whether or not such an approach could be justified on an
air quality basis, given the potential for very high sulfate PM
emissions due to unrestrained sulfate production in diesel catalytic
converters;
<bullet> Whether such an approach should be limited to the interim
PM standards and be discontinued when the Tier 2 standards are fully
phased in;
<bullet> How this approach should be phased out if low-sulfur
diesel fuel were to be phased in; and
<bullet> Whether a cap on sulfate PM should accompany such an
approach and what value (in grams per mile) would be appropriate for a
cap.

D. What Are the Economic Impacts, Cost Effectiveness and Monetized
Benefits of the Proposal?

Consideration of the economic impacts of new standards for vehicles
and fuels has been an important part of our decision making process for
this proposal. The following sections describe first the costs
associated with meeting the new vehicle standards and the new fuel
standards. This will be followed with a discussion of the cost
effectiveness of the proposal. Lastly, we will discuss the results of a
preliminary benefit-cost assessment that we have prepared.
Full details of our cost analyses, including information not
presented here, can be found in the Draft RIA associated with this
rule. We invite comments on all aspects of these analyses.
1. What Are the Estimated Costs of the Proposed Vehicle Standards?
To perform a cost analysis for the proposed standards, we first
determined a package of likely technologies that manufacturers could
use to meet the proposed standards and then determined the costs of
those technologies. In making our estimates we have relied both on
publicly available information, such as that developed by California,
and confidential information supplied by individual manufacturers.
In general, we expect that the Tier 2 standards will be met through
refinements of current emissions control components and systems rather
than through the widespread use of new technology. Furthermore, lighter
vehicles will generally require less extensive improvements than larger
vehicles and trucks. More specifically, we anticipate a combination of
technology upgrades such as the following:
<bullet> Improvements to the catalyst system design, structure, and
formulation plus some increase in average catalyst size and loading.
<bullet> Air and fuel system modifications including changes such
as improved microprocessors, improved oxygen sensors, leak free exhaust
systems, air assisted fuel injection, and calibration changes including
improved precision fuel control and individual cylinder fuel control.
<bullet> Engine modifications, possibly including an additional
spark plug per cylinder, an additional swirl control valve, or other
hardware changes needed to achieve cold combustion stability.
<bullet> Increased use of fully electronic exhaust gas
recirculation (EGR).
<bullet> Increased use of secondary air injection for 6 cylinder
and larger engines.
<bullet> Heat optimized exhaust pipes and low thermal capacity
manifolds.
Using a typical mix of changes for each group, we projected costs
separately for LDVs, the different LDT classes, and for different
engine sizes (4, 6, 8-cylinder) within each class. For each group we
developed estimates of both variable costs (for hardware and assembly
time) and fixed costs (for R&D, retooling, and certification).
Cost estimates based on the current projected costs for our
estimated technology packages represent an expected incremental cost of
vehicles in the near-term. For the longer term, we have identified
factors that would cause cost impacts to decrease over time. First,
since fixed costs are assumed to be recovered over a five-year period,
these costs disappear from the analysis after the fifth model year of
production. Second, the analysis incorporates the expectation that
manufacturers and suppliers will apply ongoing research and
manufacturing innovation to making emission controls more effective and
less costly over time. Research in the costs of manufacturing has
consistently shown that as manufacturers gain experience in production,
they are able to apply innovations to simplify machining and assembly
operations, use lower cost materials, and reduce the number or
complexity of component parts.⁶⁵ These reductions in
production costs are typically associated with every doubling of
production volume. Our analysis incorporates the effects of this
``learning curve'' by projecting that the variable costs of producing
the Tier 2 vehicles decreases by 20 percent starting with the third
year of production. We applied the learning curve reduction only once
since, with existing technologies, there would be less opportunity for
lowering production costs than would be the case with the adoption of
new technology.
---------------------------------------------------------------------------

\65\ ``Learning Curves in Manufacturing,'' Linda Argote and
Dennis Epple, Science, February 23, 1990, Vol. 247, pp. 920-924.
---------------------------------------------------------------------------

We have prepared our cost estimates for meeting the Tier 2
standards using a baseline of NLEV technologies for LDVs, LDT1s, and
LDT2s, and Tier 1 technologies for LDT3s and LDT4s. These are the
standards that vehicles would be meeting in 2003. ⁶⁶ We have
not specifically analyzed smaller incremental changes to technologies
that might occur due to the interim standards between the baseline and
Tier 2. In many cases, we believe these changes will not be significant
based on current certification levels. For others, manufacturers can
use averaging and other program flexibilities to avoid redesigning
vehicles twice within a relatively short period of time. We believe
this is likely to be an attractive approach for manufacturers due to
the savings in R&D and other resources.
---------------------------------------------------------------------------

\66\ Even though the NLEV program ends in the Tier 2 time frame,
we have not included the NLEV program costs or benefits in our
analysis, since EPA analyzed and adopted NLEV previously.
---------------------------------------------------------------------------

For the total annual cost estimates, we projected that
manufacturers will start the phase-in of Tier 2 vehicles with LDVs in
2004 and progress to heavier vehicles until all LDT2s meet Tier 2
standards in 2007. For LDT3s and LDT4s, we projected some sales of Tier
2 LDT3s prior to 2008 for purposes of averaging in the interim program
and that the phase-in of Tier 2 vehicles would end with LDT4s in 2009.
Finally, we have incorporated what we believe to be a high level of
R&D spending at $5,000,000 per vehicle line (with annual sales of
100,000 units per line). We have included this large R&D effort because
calibration and system optimization is likely to be a critical part of
the effort to meet Tier 2 standards. However, we believe that the R&D
costs may be overstated because the projection ignores the carryover of
knowledge from the first vehicle lines designed to meet the standard to
others phased-in later.
The evaporative emissions standards we are proposing today for LDVs
and LDTs are feasible with relatively small cost impacts. We estimate
the cost of system improvements to be about $4 per vehicle, for all
vehicle classes. This incremental cost reflects the cost of moving to
low permeability materials, improved designs or low-loss

[[Page 26071]]

connectors. R&D for the evaporative emissions standard is included in
the R&D estimates given above for the tailpipe standards. We have made
no projections of learning curve reductions for the evaporative
standard.
Table IV.D.-1 provides our estimates of the per vehicle increase in
purchase price for LDVs and LDTs. The near-term cost estimates in Table
IV.D.-1 are for the first years that vehicles meeting the standards are
sold, prior to cost reductions due to lower productions costs and the
retirement of fixed costs. The long-term projections take these cost
reductions into account. We have sales weighted the cost differences
for the various engine sizes (4-, 6-, 8-cylinder) within each category.

Table IV.D.-1.--Estimated Purchase Price Increases Due to Proposed Tier 2 Standards
----------------------------------------------------------------------------------------------------------------
LDV LDT1 LDT2 LDT3 LDT4
----------------------------------------------------------------------------------------------------------------
Tailpipe standards:
Near-term (year 1)......................... $76 $69 $132 $270 $266
Long-term (year 6 and beyond).............. 46 43 99 214 209
Evaporative Standard........................... 4 4 4 4 4
----------------------------------------------------------------------------------------------------------------

2. What Are the Estimated Costs of the Proposed Gasoline Sulfur
Standards?
As explained in Section IV.C., most refiners will have to install
capital equipment to meet the proposed gasoline sulfur standard.
Presuming that refiners will want to minimize the cost involved,
refiners are expected to desulfurize the gasoline blendstock produced
by the fluidized catalytic cracker (FCC) unit. Recent advances have led
to significant improvements in hydrotreating technology by CDTECH and
Mobil Oil (OCTGAIN) that lower the cost of desulfurizing FCC gasoline;
we understand that similar technologies are being developed by other
parties. Since these improved desulfurization technologies represent
the lowest cost options and are expected to be used by most refiners
needing to install desulfurization equipment, we estimated the cost of
desulfurization based on their use.
For our analysis, we estimated the cost of lowering gasoline sulfur
levels in five different regions of the country (Petroleum
Administration Districts for Defense, or PADD), starting from the
current regional average in each PADD down to 30 ppm. We then converted
the regional cost to a national average per-refinery cost, and
calculated a national aggregate cost and cents-per-gallon cost.
Based on this analysis we estimate that, on average, refiners in
the year 2004 would be expected to invest about $45 million for capital
equipment and spend about $16 million per year for each refinery to
cover the operating costs associated with these desulfurization units.
Since this average represents many refineries diverse in size and
gasoline sulfur level, some refineries would pay more and others less
than the average costs. When the average per-refinery cost is
aggregated for all the gasoline expected to be produced in this country
in 2004, the total investment for desulfurization processing units is
estimated to be about $4.7 billion dollars, and operating costs for
these units is expected to be about $1.5 billion per year. We believe
that the $4.7 billion in capital costs would be spread over several
years by the refiners' participation in the proposed averaging,
banking, and trading program.
These capital and operating costs represent our estimates for
domestic costs. While we think that many foreign refiners might incur
capital costs to meet the requirements of our gasoline sulfur program,
particularly in light of similar programs being enacted
internationally, others will argue that most foreign refiners would not
incur new costs as a result of our program because they can simply send
the lowest-sulfur fraction of their current production to the U.S.
Furthermore, some will argue that most foreign refiners do not face the
same permitting limitation and environmental and other regulatory costs
that domestic refiners face, and thus that their costs of producing low
sulfur gasoline will be minimal even if some investment is required.
While we have developed cost estimates with and without consideration
of possible costs attributed to imported gasoline, our estimates of
national and average costs do not include any costs attributed to
foreign refiners.
Using our estimated capital and operating costs we calculated the
average per-gallon cost of reducing gasoline sulfur down to 30 ppm.
Using a capital cost amortization factor based on a seven percent rate
of return on investment, and including no taxes, we estimated the
average national cost for desulfurizing gasoline to initially be about
1.7 cents per gallon. This cost is the cost to society of reducing
gasoline sulfur down to 30 ppm that we used for estimating cost
effectiveness. If we amortize the costs based on a rate of return on
investment of six to ten percent and a tax rate of 39 percent, which
may more closely represent the actual economic situation facing
refiners today, the average national cost for desulfurizing gasoline
down to 30 ppm would be 1.7-1.9 cents per gallon.
We anticipate that these costs will decrease in future years due to
improvements in technology, similar to the learning curve improvements
discussed above for vehicle cost. This improvement is estimated to
result in a 20 percent reduction in operating costs after the second
complete year of use. This estimated rate of improvement is similar to
previous cost reductions observed with desulfurization technologies as
they were being developed.
Additional cost reduction is expected as refiners increase the
throughput (debottleneck) of their refineries to lower their per-gallon
fixed costs. This increase in throughput for the industry as a whole is
termed capacity creep and it is has allowed a shrinking number of U.S.
refineries to handle the increasing demand for refined products. Our
analysis presumes that as an industry, refiners will debottleneck their
refineries at a rate consistent with the forecasted increase in
gasoline demand, which is about 2 percent per year. Thus, the fixed
operating cost, and a portion of the capital costs for these
desulfurization technologies, would decrease over time on a per gallon
basis as the volume of gasoline processed at each refinery increased.
Table IV.D.-2 below summarizes our estimates of per-gallon gasoline
cost increases for the years 2004, 2010 and 2015.

Table IV.D.-2.--Estimated Per-Gallon Cost for Desulfurizing Gasoline in
Future Years
------------------------------------------------------------------------
Cost (cents/
Year gallon)
------------------------------------------------------------------------
2004....................................................... 1.7
2010....................................................... 1.5
2015....................................................... 1.4
------------------------------------------------------------------------

[[Page 26072]]

3. What Are the Aggregate Costs of the Tier 2/Gasoline Sulfur Proposal?
Using current data for the size and characteristics of the vehicle
fleet and making projections for the future, the per-vehicle and per-
gallon fuel costs described above can be used to estimate the total
cost to the nation for the proposed emission standards in any year.
Figure IV.D.-1 portrays the results of these projections.⁶⁷

BILLING CODE 6560-50-P
[GRAPHIC] [TIFF OMITTED] TP13MY99.004

BILLING CODE 6560-50-C
As can be seen from the figure, the annual cost starts out at just
over $2.5 billion per year and increases over the phase-in period to a
maximum of $3.7 billion in 2008. Thereafter, the annual cost declines
to a level of about $3.5 billion. The effect of projected growth in
vehicle sales and fuel consumption causes a slow, gradual rise in
annual cost to set in after about 2012.
4. How Does the Cost Effectiveness of This Program Compare to Other
Programs?
This section summarizes the cost effectiveness analysis done by EPA
and its results. The purpose of this assessment is to determine whether
reductions from the vehicle and fuel controls are cost effective,
taking into consideration alternative means of attaining or maintaining
the national primary ambient air quality standards. This involves a
comparison of our proposed program not only with past measures, but
with other new measures that might be employed to attain and maintain
the NAAQS. Both EPA and states have already adopted numerous control
measures, and remaining measures tend to be more expensive than those
previously employed. Therefore, there is no single cost effectiveness
level that defines what is acceptable. Rather, as we employ the most
cost effective available measures first, more expensive ones tend to
become necessary over time.
---------------------------------------------------------------------------

\67\ Figure IV.D.-1 is based on the amortized costs from Tables
IV.D.-1 and IV.D.-2. Actual capital investments, particularly
important for fuels, would occur prior to and during the initial
years of the program, as described above in section IV.D.2.
---------------------------------------------------------------------------

a. What Is the Cost Effectiveness of This Program? We have
calculated the per-vehicle cost effectiveness of the exhaust/gasoline
sulfur standards and the evaporative emission standards, based on the
net present value of all costs and emission reductions over the life of
an average Tier 2 vehicle subject to today's proposal. As described
earlier in the discussion of the cost of this proposal, the cost of
complying with the new standards will decline over time as
manufacturing costs are reduced and amortized capital investments are
recovered. To show the effect of declining cost on the cost
effectiveness, we have developed both near term and long term cost
effectiveness values. More specifically, these correspond to

[[Page 26073]]

vehicles sold in years one and six of the vehicle and fuel programs.
Vehicle cost is constant from year six onward. Fuel costs per gallon
continue to decline slowly in the years past year six; however, the
overall impact of this decline is small and we have decided to use year
six results for our long term cost effectiveness. Chapter V of the
draft RIA contains a full description of this analysis, and you should
look in that document for more details on the results summarized here.
Table IV.D.-3 summarizes the net present value lifetime cost, NMHC
+ NO_X emission reduction and cost effectiveness results for
the Tier 2/gasoline sulfur proposal using sales weighted averages of
the costs (both near term and long term) and emission reductions of the
various vehicle classes affected.
Table IV.D.-3 also displays cost effectiveness values based on two
approaches to account for the small reductions in SO<INF>2</INF> and
tailpipe emitted sulfate particulate matter (PM) associated with the
reduction in gasoline sulfur. While these reductions are not central to
the proposal and are therefore not displayed with their own cost
effectiveness, they do represent real emission reductions due to the
proposed rule. The first set of cost effectiveness numbers in Table
IV.D.-3 simply ignores these reductions and bases the cost
effectiveness on only the NMHC + NO_X reductions from Tier 2/
gasoline sulfur. The second set accounts for these reductions by
crediting some of the cost of the program to SO<INF>2</INF> and PM
reduction. The amount of cost allocated to SO<INF>2</INF> and PM is
based on the cost effectiveness of SO<INF>2</INF> and PM emission
reductions from other EPA actions. You may refer to the RIA for details
about these actions and how the specific allocations were developed.

Table IV.D.-3.--Cost Effectiveness of the Proposed Standards (1997 dollars)
----------------------------------------------------------------------------------------------------------------
Discounted
Discounted Discounted Discounted lifetime cost
lifetime lifetime NMHC lifetime cost effectiveness
Cost basis vehicle and + NO_X effectiveness per ton with
fuel costs reduction per ton SO<INF>2</INF> and direct
(tons) PM credit^a
----------------------------------------------------------------------------------------------------------------
Near term cost (production year 1).............. $230 0.108 $2,134 $1,599
Long term cost (production year 6).............. 188 0.109 1,748 1,213
----------------------------------------------------------------------------------------------------------------
^a $54 credited to SO<INF>2</INF> ($4800/ton), $4 to direct PM ($10,000/ton).

b. How Does the Cost Effectiveness of this Program Compare with
Other Means of Obtaining Mobile Source NO_X + NMHC
Reductions? In comparison with other mobile source control programs, we
believe that today's proposal represents the most cost effective new
mobile source control strategy currently available that is capable of
generating substantial NO_X + NMHC reductions. This can be
seen by comparing the cost effectiveness of today's program with a
number of new mobile source standards that EPA has adopted in recent
years. Table IV.D.-4 summarizes the cost effectiveness of several
recent EPA actions.

Table IV.D.-4.--C/E of Previously Implemented Mobile Source Programs
------------------------------------------------------------------------
$/ton
Program NO_X+NMHC
------------------------------------------------------------------------
2004 Highway HD Diesel stds................................... 300
Nonroad Diesel engine stds.................................... 410-650
Tier 1 vehicle controls....................................... 1,980-2,
690
NLEV.......................................................... 1,859
Marine SI engines............................................. 1,128-1,
778
On-board diagnostics.......................................... 2,228
------------------------------------------------------------------------
(Costs adjusted to 1997 dollars.)

We can see from the table that the cost effectiveness of the Tier
2/gasoline sulfur standards falls within the range of these other
programs. Engine-based standards (the 2004 highway heavy-duty diesel
standards, the nonroad diesel engine standards and the marine spark-
ignited engine standards) have generally been less costly than Tier 2/
gasoline sulfur. Vehicle standards, most similar to today's proposal,
have values comparable to or higher than Tier 2/gasoline sulfur.
It is tempting to look at the engine standards and conclude that
more reductions at a similar low cost effectiveness should still be
available. This is especially true for the two largest categories
(highway and nonroad diesel engines) where new standards have been
adopted that were highly cost effective. However, cost effectiveness
was not a limiting consideration in either case. Rather, the level of
the standards selected was based primarily on technical feasibility in
the time available. That is, the maximum level of control that we found
to be feasible in these actions was driven more by what technology we
believed would be available than by cost. It will be important to
consider the potential for further control in these categories as we
move forward.
We do not believe that significant further control is available
from highway or nonroad diesel engines through more stringent standards
at the same cost effectiveness that these standards realized, in the
time frame proposed. Based on current knowledge, the next generation of
controls for these diesel engines would require advanced after-
treatment devices, still in the research and development phase. Such
controls have not yet been employed and when they become available will
be more costly and will have difficulty functioning without changes to
diesel fuel. We fully expect that, as the development of new technology
progresses and cost declines, future new standards for both of these
source categories will be developed. But we also expect that the cost
effectiveness of future standards will be higher and is not likely to
be significantly less than the cost effectiveness of today's proposal.
On the light duty vehicle side, the last two sets of standards were
Tier 1 and NLEV, which had cost effectiveness comparable to or higher
than Tier 2/gasoline sulfur. Compared to engines, these levels reflect
the advanced (and more expensive) state of vehicle control technology,
where standards have been in effect for a much longer period than for
engines. In fact, considering the increased stringency of the Tier 2
standards,⁶⁸ it is remarkable that the cost effectiveness of
Tier 2/gasoline sulfur is in the same range as these actions. Based on
these results, Tier 2/gasoline sulfur appears to be a logical and
consistent next step in vehicle control.
---------------------------------------------------------------------------

\68\ Tier 2/gasoline sulfur will yield about a 75% reduction in
NO_X emissions compared to NLEV vehicles.
---------------------------------------------------------------------------

In conclusion, we believe that the Tier 2/gasoline sulfur proposal
is a cost effective program for mobile source NO_X + NMHC
control. We are unable to

[[Page 26074]]

identify another mobile source control program that would be more cost
effective than Tier 2/gasoline sulfur for making substantial further
progress in reducing NO_X + NMHC emissions.
c. How Does the Cost Effectiveness of this Proposed Program Compare
with Other Known Non-Mobile Source Technologies for Reducing
NO_X + NMHC? In evaluating the cost effectiveness of the Tier
2/gasoline sulfur proposal, we also considered whether our proposal is
cost effective in comparison with alternative means of attaining or
maintaining the NAAQS other than mobile source programs. As described
below, we have concluded that Tier 2/gasoline sulfur is cost effective
considering the anticipated cost of other technologies that will be
needed to help attain and maintain the NAAQS.
For purposes of estimating the cost of implementing the new ozone
and PM NAAQS, the Agency assumed certain baseline controls and compiled
a list of additional known technologies that could be considered in
devising emission reductions strategies.⁶⁹ Through this
broad review, over 50 technologies were identified as reducing
NO_X or VOC. The average cost effectiveness of these
technologies varied from hundreds of dollars a ton to tens of thousands
of dollars a ton. The Agency selected from this list all those
technologies that could be applied with an average cost effectiveness
of $10,000/ton or less, and showed that substantial progress toward
attainment could be made when operating within that limit.
---------------------------------------------------------------------------

\69\ ``Regulatory Impact Analyses for the Particulate Matter and
Ozone National Ambient Air Quality Standards and Proposed Regional
Haze Rule,'' Appendix B, ``Summary of control measures in the PM,
regional haze, and ozone partial attainment analyses,'' Innovative
Strategies and Economics Group, Office of Air Quality Planning and
Standards, U.S. Environmental Protection Agency, Research Triangle
Park, NC, July 17, 1997.
---------------------------------------------------------------------------

While many areas still remained in nonattainment under the NAAQS
analysis, we assumed that other methods would be identified in the
future that on average could help achieve the NAAQS at $10,000 per ton
or less. We believe that Tier 2/gasoline sulfur is one of those
methods. In fact, it will deliver critical further reductions that are
not readily obtainable by any other means known to the Agency. By way
of comparison, if all of the technologies identified for the NAAQS
analysis costing less than $10,000/ton were implemented nationwide,
they would produce NO_X emission reductions of about 2.9
million tons per year. The Tier 2/gasoline sulfur proposal by itself
will generate about 2.8 million tons per year once fully implemented.
To obtain significant further reductions using the other technologies
identified in the NAAQS analysis rather than Tier 2/gasoline sulfur
could mean adopting measures costing well beyond $10,000/ton. Given the
continuing need for further emission reductions, we believe that Tier
2/gasoline sulfur control is clearly a cost effective approach, in
addition to those technologies assumed for the NAAQS analysis, for
attaining and maintaining the NAAQS.
We recognize that the cost effectiveness calculated for Tier 2/
gasoline sulfur is not strictly comparable to a figure for measures
targeted at nonattainment areas, since Tier 2/gasoline sulfur is a
nationwide program. However, there are several additional
considerations that have led us to conclude that Tier2/gasoline sulfur
is cost effective considering alternative means of attaining and
maintaining the NAAQS.
First, given the fact that Tier 2/gasoline sulfur is at most only
20 percent as costly per ton as the NAAQS figure for additional control
measures, we believe that there can be little doubt that the cost
effectiveness of Tier 2/gasoline sulfur is well within the cost
effectiveness range that the NAAQS cost analysis anticipated for
unspecified additional technologies that will be needed to attain the
NAAQS--technologies that the analysis noted might be applied in limited
areas or nationwide. Furthermore, as a national program, Tier 2/
gasoline sulfur can be implemented as a single unified rule without the
need for individual action by each of the states. Moreover, as noted
above, for states to obtain further substantial emission reductions
beyond those identified in the NAAQS could mean adopting measures
costing well beyond $10,000/ton, something that few areas of the
country to date have done.
In dealing with the question of comparing local and national
programs, it is also relevant to point out that, because of air
transport, the need for NO_X control is a broad regional
issue not confined to non-attainment areas only. To reach attainment,
future controls will need to be applied over widespread areas of the
country. In the analyses supporting the recent NO_X standards
for highway diesel engines,⁷⁰ we looked at this question in
some detail and concluded that the regions expected to impact ozone
levels in ozone nonattainment areas accounted for over 85% of total
NO_X emissions from a national heavy-duty engine control
program. Similarly, NO_X emissions in attainment areas also
contribute to particulate matter nonattainment problems in downwind
areas. Thus, the distinction between local and national control
programs for NO_X is less important than it might appear.
---------------------------------------------------------------------------

\70\ Final Regulatory Impact Analysis: Control of Emissions of
Air Pollution from Highway Heavy-Duty Engines, September 16, 1997.
---------------------------------------------------------------------------

Finally, the statute indicates that in considering the cost
effectiveness of Tier 2/gasoline sulfur EPA should consider not only
attainment, but also maintenance of the standards. Tier 2/gasoline
sulfur--unlike nonattainment area measures--will achieve attainment
area reductions that, among other effects, will help to maintain air
quality that meets the NAAQS. These reductions relate not only to the
ozone and PM NAAQS, but also to SO<INF>2</INF> and NO<INF>2</INF>, and
to CO.
In summary, given the array of controls that will have to be
implemented to make progress toward attaining and maintaining the
NAAQS, we believe that the weight of the evidence from alternative
means of providing substantial NO_X + NMHC emission
reductions indicates that the Tier 2/gasoline sulfur proposal is cost
effective. This is true from the perspective of other mobile source
control programs or from the perspective of other stationary source
technologies that might be considered.
5. Does the Value of the Benefits Outweigh the Cost of the Proposed
Standards?
While relative cost effectiveness is the principal economic policy
criterion established for these standards in the Clean Air Act (see CAA
202(i)), further insight regarding the merits of the proposed standards
can be provided by benefit-cost analysis. The purpose of this section
is to summarize the methods we used and results we obtained in
conducting a preliminary analysis of the economic benefits of the
proposed standards, and to compare these economic benefits with the
estimated costs of the proposal. In summary, the results of our
analysis indicate that the economic benefits of the proposed standards
will likely exceed the costs of meeting the standards by a substantial
margin, and the significant uncertainties underlying the analysis are
unlikely to alter this outcome of positive net benefits.
a. What Is the Purpose of this Benefit-Cost Comparison? Benefit-
cost analysis (BCA) is a useful tool for evaluating the economic merits
of proposed changes in environmental programs and policies. In its
traditional application, BCA

[[Page 26075]]

estimates the economic ``efficiency'' of proposed changes in public
policy by organizing the various expected consequences and representing
those changes in terms of dollars. Expressing the effects of these
policy changes in dollar terms provides a common basis for measuring
and comparing these various effects. Because improvement in economic
efficiency is typically defined to mean maximization of total wealth
spread among all members of society, traditional BCA must be
supplemented with other analyses in order to gain a full appreciation
of the potential merits of new policies and programs. These other
analyses may include such things as examinations of legal and
institutional constraints and effects; engineering analyses of
technology feasibility, performance and cost; or assessment of the air
quality need.
In addition to the narrow, economic efficiency focus of most BCAs,
the technique is also limited in its ability to project future economic
consequences of alternative policies in a definitive way. Critical
limitations on the availability, validity, or reliability of data;
limitations in the scope and capabilities of environmental and economic
effect models; and controversies and uncertainties surrounding key
underlying scientific and economic literature all contribute to an
inability to estimate the economic effects of environmental policy
changes in exact and unambiguous terms. Under these circumstances, we
consider it most appropriate to view BCA as a tool to inform, but not
dictate, regulatory decisions such as the ones reflected in today's
proposal.
Despite the limitations inherent in BCA of environmental programs,
we considered it useful to estimate the potential benefits of today's
proposed standards both in terms of physical changes in human health
and welfare and environmental change, and in terms of the estimated
economic value of those physical changes. The BCA presented herein
should be considered preliminary, however, due to limitations in the
data and models available for analysis in advance of today's proposal.
Additional, more refined analysis will be conducted prior to issuance
of final standards. This post-proposal analysis will take account of
public comments on the proposed standards and this BCA and will also
make use of more extensive and refined data and models currently being
developed. Our expectation is that the more extended and refined
economic analysis conducted prior to final rulemaking will further help
inform and guide decisions on the appropriateness of the final rules.
Toward this end, we are presenting this preliminary BCA and requesting
public comments on the assumptions, data, and modeling efforts
supporting the analysis and its results, and the appropriate
interpretations and uses of those results.
b. What Was Our Overall Approach to the Benefit-Cost Analysis? The
basic question we sought to answer in the preliminary BCA was: ``What
are the net yearly economic benefits to society of the reduction in
mobile source emissions likely to be achieved by today's proposed
standards?'' In designing an analysis to answer this question, we
adopted an analytical structure and sequence similar to that used in
the so-called ``section 812 studies'' ⁷¹ to estimate the
total benefits and costs of the entire Clean Air Act. Moreover, we used
many of the same data sets, models, and assumptions actually used in
the Section 812 studies and/or the recent Regulatory Impact Analyses
(RIAs) for the Particulate Matter and Ozone National Ambient Air
Quality Standards and for the NO_X SIP Call (also known as
the Regional Ozone Transport Rule, as discussed in Section III
above).⁷² By adopting the major design elements, data sets,
models, and assumptions developed for the recent RIAs, we have largely
relied on methods that have already received extensive review by the
public and by other federal agencies. Furthermore, the data sets
adopted from the Section 812 studies have received extensive review by
the independent Science Advisory Board and by the public.
---------------------------------------------------------------------------

\71\ The ``section 812 studies'' refers to (1) USEPA, Report to
Congress: The Benefits and Costs of the Clean Air Act, 1970 to 1990,
October 1997 (also known as the ``section 812 Retrospective); and
(2) the first in the ongoing series of prospective studies
estimating the total costs and benefits of the Clean Air Act,
expected to be published later in 1999.
\72\ Regulatory Impact Analysis for the NO_X SIP Call,
FIP, and Section 126 Petitions'' September 1998, EPA-452/R-98-003.
---------------------------------------------------------------------------

As described in more detail in the Draft RIA for today's proposal,
this overall analytical design involves the following sequential steps:
1. Identify the technologies likely to be used to comply with the
proposed standards
2. Estimate the costs society would incur to employ the
technologies
3. Estimate the emissions reductions achieved by application of the
technologies
4. Estimate the change in air quality conditions resulting from the
estimated emissions reductions
5. Estimate the changes in human health and well-being and
environmental quality associated with the estimated changes in air
quality
6. Estimate the economic value of the estimated changes in human
health, human welfare, and environmental outcomes
7. Compare the resulting estimate of economic benefits with the
estimated costs, and calculate the net monetized benefits of the
proposed standards
8. Evaluate the uncertainty surrounding the estimate of net
monetized benefit by developing ranges of results that reflect the key
underlying scientific, economic, data, and modeling uncertainties
c. What Are the Significant Limitations of the Benefit-Cost
Analysis? Every BCA examining the potential effects of a change in
environmental protection requirements is limited to some extent by data
gaps, limitations in model capabilities (such as geographic coverage),
and uncertainties in the underlying scientific and economic studies
used to configure the benefit and cost models. Deficiencies in the
scientific literature often result in the inability to estimate changes
in health and environmental effects, such as potential increases in
premature mortality associated with increased exposure to carbon
models. Deficiencies in the economics literature often result in the
inability to assign economic values even to those health and
environmental outcomes that can be quantified, such as changes in lung
function caused by increased exposure to ozone. While these general
uncertainties in the underlying scientific and economics literatures
are discussed in detail in the RIA and its supporting documents and
references, the key uncertainties that have a bearing on the results of
the preliminary BCA of today's proposed standards are:
1. The exclusion of potentially significant benefit categories
(e.g., health and ecological benefits of incidentally controlled
hazardous air pollutants)
2. Scientific uncertainties regarding whether the observed
statistical relationship between exposure to elevated particulate
matter and incidences of adverse health effects reflects a causal
relationship (especially premature mortality and chronic bronchitis)
3. Scientific uncertainty regarding the potential existence of a
concentration threshold below which adverse health effects of exposure
to particulate matter might not occur
4. Scientific uncertainty regarding whether tropospheric ozone
exposure contributes to premature mortality
In addition to these uncertainties and shortcomings that pervade
all analyses of criteria air pollutant control

[[Page 26076]]

programs, a number of limitations apply specifically to the preliminary
BCA of today's proposed rules. Though we used the best data and models
currently available, we were required to adopt a number of simplifying
assumptions and to use data sets that, while reasonably close, did not
match precisely the conditions and effects expected to result from
implementation of the standards proposed today. For example, the year
2010 emissions data sets available for use in this analysis do not
fully reflect the emissions reductions expected to be achieved by other
recently-enacted standards and by expected near-future control
programs, such as additional measures aimed at full attainment of the
new fine particulate matter National Ambient Air Quality Standards. In
addition, we have used the year 2010 as a proxy for the time (actually
circa 2040) when all non-complying vehicles would be fully retired from
the fleet and full implementation of today's proposed standards would
be finally achieved, requiring adjustments described more fully in the
next section. The key limitations and uncertainties unique to the
preliminary BCA of today's proposed rules, therefore, include:
1. A mismatch between the 2010 air quality base year adopted for
the BCA and the eventual timing of fleet turnover
2. Potential mis-estimation of future year emissions inventories,
such as those associated with nonroad vehicle emissions and with
measures aimed at attaining and maintaining compliance with newly
revised ambient air quality standards
3. Uncertainties associated with the extrapolation of air quality
monitoring data to distant sites required to capture the effects of the
proposed standards on all affected populations
Despite these additional important uncertainties, which are
discussed in more detail or referenced in the Draft RIA, we believe the
preliminary BCA does provide a reasonable indication of the potential
range of net economic benefits of the standards proposed today. This is
because the analysis focuses on estimating the economic effects of the
changes in air quality conditions expected to result from today's
proposed rules, rather than focusing on developing a precise prediction
of the absolute levels of air quality likely to prevail at some
particular time in the future. An analysis focusing on the changes in
air quality can give useful insights into the likely economic effects
of emission reductions of the magnitude expected to result from today's
proposed rule.
d. How Did We Perform the Benefit-Cost Analysis? As summarized
above, the analytical sequence begins with a projection of the mix of
technologies likely to be deployed to comply with the new standards,
and the costs incurred and emissions reductions achieved by these
changes in technology. The program proposed today has various cost and
emission related components, as described earlier in this section.
These components would begin at various times and in some cases would
phase in over time. This means that during the early years of the
program there would not be a consistent match between cost and
benefits. This is especially true for the vehicle control portions of
the proposal, where the full vehicle cost would be incurred at the time
of vehicle purchase, while the fuel cost along with the emission
reductions and benefits would occur throughout the lifetime of the
vehicle. To deal with this question, we might have wished to perform a
per-vehicle analysis corresponding to the cost effectiveness analysis
described above. However, the modeling used for benefits estimates
cannot be done on a per-vehicle basis, so we have instead used an
annual cost and annual benefit approach.
To develop a representative benefit-cost number, we need to have a
stable set of cost and emission reductions to use. This means using a
future year where the fleet is fully turned over and there is a
consistent annual cost and annual emission reduction. For today's
proposal this stability wouldn't occur until well into the future.
However, for the purpose of the benefit calculations, we have no
available baseline data set beyond the year 2010. We have therefore
made adjustments to allow use of 2010 as a surrogate for a future year
in which the fleet consists entirely of Tier 2 vehicles.
For emissions, we calculated reductions by treating 2010 as if the
fleet had already turned over. We did this by applying the control case
emission factor from a fully turned over fleet year (from the year
2040) to the fleet mileages for this year. Clearly, this approach does
not, nor is it intended to, predict actual expected emission reductions
for 2010. This is not its purpose. It is intended to portray the
characteristics of the vehicle fleet after it is fully turned over,
within the constraint that 2010 was the latest year for which we could
perform an analysis.
The resulting analysis represents a snapshot of benefits and costs
in a future year in which the light-duty fleet consists entirely of
Tier 2 vehicles. As such, it depicts the maximum emission reductions
(and resultant benefits) and among the lowest costs that would be
achieved in any one year by the program on a ``per mile'' basis. (Note,
however, that net benefits would continue to grow over time beyond
those resulting from this analysis, but only because of growth in
vehicle miles traveled.) Thus, based on the long-term costs for a fully
turned over fleet, the resulting benefit-cost ratio will be close to
its maximum point (for those benefits that we have been able to value).
Costs to be compared to the monetized value of the benefits were
also developed for a fleet the size of the year 2010 fleet. For this
purpose we used the long term cost once the capital costs have been
recovered and the manufacturing learning curve reductions have been
realized, since this most closely represents the makeup of a fully
turned over fleet.
We also made adjustments in the costs to account for the fact that
there is a time difference between when some of the costs are expended
and when the benefits are realized. The vehicle costs are expended when
the vehicle is sold, while the fuel related costs and the benefits are
distributed over the life of the vehicle. We resolved this difference
by using costs distributed over time such that there is a constant cost
per ton of emissions reduction and such that the net present value of
these distributed costs corresponds to the net present value of the
actual costs.
The resulting adjusted costs are somewhat greater than the expected
actual annual cost of the program, reflecting the time value
adjustment. Thus, both because of the assumption of a fully turned over
fleet and because of the time value adjustment, the costs presented in
this section do not represent expected actual annual costs for 2010.
Rather, they represent an approximation of the steady-state cost per
ton that would likely prevail in 2015 and beyond. The benefit cost
ratio for the earlier years of the program would be expected to be
lower than that based on these costs, since the fleet-adjusted costs
are larger in the early years of the program while the benefits are
smaller.
Finally, at the time that we undertook the development of the
benefit estimates for this rule, we did not have quantitative estimates
of the VOC emission reductions that would result from the evaporative
emission standards in the proposal. Therefore, the benefit estimates do
not include the value of the evaporative emission standard. Consistent
with this, the program cost estimates also exclude the evaporative
emission control cost. Since the evaporative emission reductions and
costs are both relatively small compared to the rest of the program,
they are not

[[Page 26077]]

expected to significantly affect the overall cost-benefit ratio.
In order to estimate the changes in air quality conditions that
would result from these emissions reductions, we developed two
separate, year 2010 emissions inventories to be used as inputs to the
air quality models. The first, baseline inventory reflects the best
available approximation of the county-by-county emissions for
NO_X, NMHC, and SO<INF>2</INF> expected to prevail in the
year 2010 in the absence of the standards proposed today. To generate
the second, control case inventory, we first estimated the change in
vehicle emissions, by pollutant and by county, expected to be achieved
by the 2010 control scenario described above. We then took the baseline
emissions inventory and subtracted the estimated reduction for each
county-pollutant combination to generate the second, control case
emissions inventory. Taken together, the two resulting emissions
inventories reflect two alternative states of the world and the
differences between them represent our best estimate of the reductions
in emissions that would result from our control scenario.
With these two emissions inventories in hand, the next step was to
``map'' the county-by-county and pollutant-by-pollutant emission
estimates to the input grid cells of two air quality models and one
deposition model. The first model, called the Urban Airshed Model
(UAM), is designed to estimate the tropospheric ozone concentrations
resulting from a specific inventory of emissions of ozone precursor
pollutants, particularly NO_X and NMHC. The second model,
called the Climatological Regional Dispersion Model Source-Receptor
Matrix model (S-R Matrix), is designed to estimate the changes in
ambient particulate matter and visibility that would result from a
specific set of changes in emissions of primary particulate matter and
secondary particulate matter precursors, such as SO<INF>2</INF>,
NO_X, and NMHC. Also, separate factors relating nitrogen
emissions to watershed deposition were developed using the Regional
Acid Deposition Model (RADM). By running both the baseline and control
case emissions inventories through these models, we were able to
estimate the expected 2010 air quality conditions and the changes in
air quality conditions that would result from the emissions reductions
expected to be achieved by the standards proposed today.
After developing these two sets of year 2010 air quality profiles,
we used the same health and environmental effect models used in the 812
studies to calculate the differences in human health and environmental
outcomes projected to occur with and without the proposed standards.
Specifically, we used the Criteria Air Pollutant Modeling System
(CAPMS) to estimate changes in human health outcomes, the Agricultural
Simulation Model (AGSIM) to estimate changes in yields of a selected
few agricultural crops, and a Household Soiling Damage function to
estimate the value of reduced household soiling due to particulate
matter. In addition, the benefits of reduced visibility impairment were
estimated using the same overall methodology used in the 812 studies,
updated to reflect recent advancements in the literature. Finally, we
developed estimates of the effect of changes in nitrogen deposition to
sensitive estuaries using methodologies applied in the PM/Ozone NAAQS
RIA (1997) and in the recent NO_X SIP Call rulemaking. (These
benefits models and methodologies are described in detail in the RIAs
associated with these actions.) Several air quality-related health and
environmental benefits, however, could not be calculated for the
preliminary BCA of today's proposed standards. Changes in human health
and environmental effects due to changes in ambient concentrations of
carbon monoxide (CO), gaseous sulfur dioxide (SO<INF>2</INF>), gaseous
nitrogen dioxide (NO<INF>2</INF>), and hazardous air pollutants could
not be included, though some of these may be included in the extended
analysis to be conducted for the final rule.
To characterize the total economic value of the reductions in
adverse effects achieved across the lower 48 states,⁷³ we
used the same set of economic valuation coefficients and models used in
the section 812 studies and the recent NO_X SIP Call RIA to
convert each type of adverse effect into a dollar value equivalent. The
net monetary benefits of today's proposed standards were then
calculated by subtracting the estimated costs of compliance from the
estimated monetary benefits of the reductions in adverse health and
environmental effects.
---------------------------------------------------------------------------

\73\ Though California is included based on the expectation that
reductions in surrounding states will achieve some benefits in
California, this analysis does not assume additional reductions in
California emissions beyond those already achieved by prevailing
standards.
---------------------------------------------------------------------------

In the final step of the analysis, we estimated the range of net
benefit estimates that might occur if important but uncertain
underlying factors were allowed to vary. By conducting this
``uncertainty analysis,'' we sought to demonstrate how much the overall
net benefit estimate might vary based on the particular uncertainties
underlying the estimates for human health and environmental effect
incidence and the economic valuation of those effects. To accomplish
this, we calculated a range of possible monetized benefit estimates
using two sets of assumptions surrounding the modeling techniques.
The method for presenting uncertainty, referred to here as the
sensitivity approach, identifies the uncertain variables that appear to
most strongly influence the overall uncertainty in the monetized
benefit estimate. These included, among others, (1) The potential that
a concentration threshold exists below that adverse PM-related health
effects may not occur, (2) alternative methods for valuing mortality,
(3) the potential contribution of tropospheric ozone to premature
mortality, (4) alternative methods for valuing reduced cases of chronic
bronchitis, (5) the extent to which agricultural crops included in our
benefits model are resistant to damage from tropospheric ozone, (6)
alternative approaches for valuing visibility. After identifying these
key variables, we defined lower bound and upper bound values for each
variable and combined these into a Low Case and a High Case. This
approach allowed us to demonstrate the sensitivity of the total
benefits to uncertainties in important variables. For example, there is
no compelling scientific evidence that a PM concentration threshold
exists below that adverse health effects do not occur. However, there
is also no scientific evidence ruling out the potential existence of a
threshold. As a result, there are no data available that would support
estimating the probability that a threshold exists at any particular PM
concentration. Under these circumstances, using the sensitivity
approach allows us to demonstrate the effect of assuming different
levels for a PM threshold.
This uncertainty calculation method does not provide a definitive
or complete picture of the true range of monetized benefits estimates.
This approach, as implemented in this preliminary BCA, does not reflect
important uncertainties in earlier steps of the analysis, including
estimation of compliance technologies and strategies, emissions
reductions and costs associated with those technologies and strategies,
and air quality and deposition changes achieved by those emissions
reductions. Nor does this approach provide a full accounting of all
potential benefits (or disbenefits) associated with the Tier 2
standards, due to data or methodological

[[Page 26078]]

limitations. Therefore, the uncertainty range is only representative of
those benefits that we were able to quantify and monetize.
e. What Were the Results of the Benefit-Cost Analysis? The
preliminary BCA for the proposed standards reflects a single year
``snapshot'' indicative of the relative yearly benefits and costs
expected to be realized once the proposed standards have been fully
implemented and non-compliant vehicles have all been retired. By
necessity, we chose to model the year 2010 because essential data on
emissions and air quality were available for this year, but not for
later years, even though the complete turnover of the fleet to Tier 2
compliant vehicles will not occur until well after 2010. Consequently,
these results are best viewed as a representation of yearly benefits
and costs over the long-term and should not be interpreted as
reflecting actual benefits and costs likely to be realized for the year
2010 itself. Benefits of the amounts shown here are likely to be
realized in the 2015-2020 time frame. In reality, near-term costs will
be higher than long-run costs as vehicle manufacturers and oil
companies invest in new capital equipment and develop and implement new
technologies. In addition, near-term benefits will be lower than long-
run benefits because it will take a number of years for Tier 2-
compliant vehicles to fully displace older, more polluting vehicles.
However, as described earlier, we have adjusted the cost estimates
upward to compensate for this discrepancy in the timing of benefits and
costs and to ensure that the benefits and costs are calculated on a
consistent basis. Because of this adjustment, the cost estimates also
should not be interpreted as reflecting the actual costs expected to be
incurred in the year 2010. Actual program costs can be found in Section
IV.D.3.
Earlier in this section, we described in more detail our approach
to estimating and adjusting our cost estimates, based upon the long-run
costs expected to be incurred in future years after the initial capital
and technology investments have been made. The resulting adjusted cost
values are given in Table IV.D.-5. Since the long term costs are not
representative of the per vehicle costs in the early phases of the
program, we also estimated an adjusted cost based on the near term cost
effectiveness value. Using the near term cost effectiveness value of
$2134/per ton, the adjusted cost would be $4.3 billion. While no actual
in-use fleet could consist entirely of vehicles experiencing this near
term cost, this value does present an upper bound on the cost figure.

Table IV.D.-5.--Adjusted Cost for Comparison to Benefits
------------------------------------------------------------------------
Adjusted
cost
Cost basis (billions
of dollars)
------------------------------------------------------------------------
Long term.................................................. 3.5
------------------------------------------------------------------------

With respect to the benefits, several different measures of
benefits can be useful to compare and contrast to the estimated
compliance costs. These benefit measures include: (a) The tons of
emissions reductions achieved, (b) the reductions in incidences of
adverse health and environmental effects, and (c) the estimated
economic value of those reduced adverse effects. Calculating the cost
per ton of pollutant reduced is particularly useful for comparing the
cost effectiveness of proposed new standards or programs against
existing programs or alternative new programs achieving reductions in
the same pollutant or combination of pollutants. The cost-effectiveness
analysis presented earlier in this preamble provides such calculations
on a per-vehicle basis. Considering the absolute numbers of avoided
adverse health and environmental effects can also provide valuable
insights into the nature of the health and environmental problem being
addressed by the rule as well as the magnitude of the total public
health and environmental gains potentially achieved by the proposed
rule. Finally, when considered along with other important economic
dimensions--including environmental justice, small business financial
effects, and other outcomes related to the distribution of benefits and
costs among particular groups--the direct comparison of quantified
economic benefits and economic costs can provide useful insights into
the overall estimated net economic effect of the proposed standards.
Table IV.D.-6 presents our range of estimates of both the estimated
reductions in adverse effect incidences and the estimated economic
value of those incidence reductions. Specifically, the table lists the
avoided incidences of individual health and environmental effects, the
pollutant associated with each of these endpoints, and the range of
estimated economic value of those avoided incidences. For several
effects, particularly environmental effects, direct calculation of
economic value in response to air quality conditions is performed,
eliminating the intermediate step of calculating incidences. Table
IV.D.-7 supplements Table IV.D.-6 by listing those additional health
and environmental benefits that could not be expressed in quantitative
incidence and/or economic value terms. A full appreciation of the
overall economic consequences of today's proposed standards requires
consideration of all benefits and costs expected to result from the new
standards, not just those benefits and costs that could be expressed
here in dollar terms.

Table IV.D.-6.--Avoided Incidence and Monetized Benefits Associated With the Tier 2 Rule for a Range of
Assumption Sets
----------------------------------------------------------------------------------------------------------------
Avoided incidence (cases/ Monetary benefits (millions
year) 1997$)
Endpoint -----------------------------------------------------------------
Low ^a High ^b Low High
----------------------------------------------------------------------------------------------------------------
PM:
Mortality (long-term exp.--ages 30+)...... 832 2,416 2,275 14,256
Mortality (long-term exp.--infants)....... .............. 10 ............... 56
Chronic bronchitis........................ 3,885 3,914 281 1,354
Hosp. Admissions--all respiratory (all 504 836 4.6 7.6
ages)....................................
Hosp. Admissions--congestive heart failure 127 138 1.5 1.7
Hosp. Admissions--ischemic heart disease.. 146 159 2.2 2.4
Acute bronchitis.......................... 984 4,072 0.1 0.2
Lower respiratory symptoms (LRS).......... 19,782 37,437 0.3 0.5
Upper respiratory symptoms (URS).......... 3,093 3,387 0.1 0.1
Work loss days (WLD)...................... 233,000 415,000 23.8 42.3
Minor restricted activity days (MRAD)..... 1,856,000 3,370,000 87.7 159.3

[[Page 26079]]

Household soiling damage.................. .............. .............. 60.1 60.1
Ozone:
Mortality (short-term; four U.S. studies). .............. 388 ............... 2,312
Hospital admissions--all respiratory (all 549 736 5.3 7.1
ages)....................................
Any of 19 acute symptoms.................. 54,101 71,545 1.3 1.7
Decreased worker productivity............. .............. .............. 43.0 60.4
Agricultural crop damage.................. .............. .............. -1 301
Visibility.................................... .............. .............. 165 701
Nitrogen Deposition........................... .............. .............. 200 200
-----------------------------------------------------------------
Total (PM + ozone + visibility + N .............. .............. 3,150 19,525
deposition)..............................
----------------------------------------------------------------------------------------------------------------
^a The low assumption set assumes effects from PM do not occur below concentrations of 15 <greek-m>g/m³, that all
mortality and chornic bronchitis effects occur within the same year of the PM reduction (see Section 7.a. of
the Draft RIA for a discussion of this uncertainty), utilizes the value of statistical life year lost
approach, ozone-related mortality and PM-related infant mortality are not included in the benefits estimate,
chronic bronchitis valued with the cost of illness approach, plantings of commodity crop cultivars are assumed
to be insensitive to ozone, does not value residential visibility benefits, and uses the lower-bound estimate
of ``willingness to pay'' for recreational visibility to reflect variation.
^b The high assumption set assumes a PM threshold of background, utilizes the value of a statistical life
approach, both ozone-related mortality and PM-related mortality are included in the estimation of benefits,
chronic bronchitis valued with a willingness-to-pay approach, plantings of commodity crop cultivars are
assumed to be sensitive to ozone, and full accounting for recreational and residential visibility benefits.

Table IV.D.-7.--Additional, Non-monetized Benefits of Proposed Tier 2 Standards
----------------------------------------------------------------------------------------------------------------
Pollutant Nonmonetized adverse effects
----------------------------------------------------------------------------------------------------------------
Particulate Matter................ Large Changes in Pulmonary Function.
Other Chronic Respiratory Diseases.
Inflammation of the Lung.
Chronic Asthma and Bronchitis.
Ozone............................. Changes in Pulmonary Function.
Increased Airway Responsiveness to Stimuli.
Centroacinar Fibrosis.
Immunological Changes.
Chronic Respiratory Diseases.
Extrapulmonary Effects (i.e., other organ systems).
Forest and other Ecological Effects.
Materials Damage.
Carbon Monoxide................... Premature Mortality.
Decreased Time to Onset of Angina.
Behavioral Effects.
Other Cardiovascular Effects.
Developmental Effects.
Sulfur Dioxide.................... Respiratory Symptoms in Non-Asthmatics.
Hospital Admissions.
Agricultural Effects.
Materials Damage.
Nitrogen Oxides................... Increased Airway Responsiveness to Stimuli.
Decreased Pulmonary Function.
Inflammation of the Lung.
Immunological Changes.
Eye Irritation.
Materials Damage.
Acid Deposition.
Hazardous Air Pollutants.......... All Human Health Effects.
Ecological Effects.
----------------------------------------------------------------------------------------------------------------

These results indicate that, based on the particular assumptions,
models, and data used in this preliminary BCA, the range of monetary
benefits realized after full turnover of the fleet to Tier 2 vehicles
would be approximately 3.2 billion to 19.5 billion dollars per year.
Comparing this estimate of the economic benefits with the adjusted cost
estimate indicates that the net economic benefit of the proposed
standards to society could be from a net cost of 0.4 billion to a net
benefit of 16.0 billion dollars per year.
The breadth of the ranges of net economic benefit estimates
presented in this preliminary BCA reinforces our conclusion that these
BCA results may be indicative of potential overall economic effects,
but they should by no means dictate whether or not the standards
proposed today should be promulgated.
f. What Additional Efforts Will Be Made Following Proposal? While
we believe that the preliminary BCA provides a strong indication that
the standards proposed today will yield positive overall economic
benefits, we

[[Page 26080]]

believe it is important to do additional analysis prior to the final
decision regarding these standards. In particular, we plan to develop
an updated and extended set of emissions inventories, and to expand the
range of pollutant-specific effects to include the benefits of
reductions in carbon monoxide (CO), sulfur dioxide (SO<INF>2</INF>),
nitrogen dioxide (NO<INF>2</INF>), and perhaps hazardous air
pollutants. We will also carefully review the public comments submitted
on the preliminary BCA and review each of the assumptions and methods
used in light these public comments and the advice of the Science
Advisory Board charged with reviewing these and other methods being
used in the pending section 812 Prospective Study Report to Congress.

E. Other Program Design Options We Have Considered

In addition to the proposed program combining Tier 2 vehicle
standards and gasoline sulfur controls, we have considered two other
major alternatives to a comprehensive vehicle/fuel program. This
section identifies these two alternatives and seeks comment on specific
aspects of each.
1. Corporate Average Standards Based on NMOG or NMOG+NO_X
We have described in great detail in previous sections of this
preamble why NO_X is our main pollutant of concern for this
rulemaking. Based on this conclusion, we are proposing a Tier 2 program
that is centered around a full useful life corporate average
NO_X standard (0.07 g/mi). Our proposed interim program for
non-Tier 2 vehicles is also centered around a corporate average
NO_X standard (0.30 or 0.20
g/mi, depending on vehicle type).
California's program, by contrast, is centered on corporate average
NMOG standards. We recognize that for Tier 2 vehicles we could also set
up the bins of emission standards and impose an average NMOG standard
in a similar fashion. A program centered on corporate average NMOG
standards could even be defined in such a way that NO_X
emissions would be indirectly driven down to the levels we have defined
with our proposed Tier 2 standards. Such an approach would provide more
consistency with California's program, and would be consistent with our
own NLEV program. However, we believe it is best, for the federal
program, to use a NO_X average standard.
With a NO_X average standard we can better tailor the
various aspects of the program to reduce the pollutant with which we
are most concerned. Thus, our averaging, banking and trading program
has been set up to provide NO_X credits for early compliance
with the Tier 2 NO_X average standard and to provide
additional NO_X credits for manufacturers certifying to
extended useful lives. Also, the NO_X average standard allows
us to set up bins in such a way as to provide manufacturers with
incentives to strive for additional NO_X reductions.
Although the use of an average NO_X requirement conflicts
with California's requirements, we do not believe any additional burden
is imposed on manufacturers. Under an NMOG averaging requirement,
manufacturers would still have to compute separate NMOG averages for
their California and Federal vehicles. This would be no smaller burden
than computing an NMOG average for California vehicles and a
NO_X average for Federal vehicles. We request comment on the
appropriateness and burden of our NO_X averaging standards
and on what benefits, if any, might be afforded by an NMOG standard for
the federal program in lieu of the proposed NO_X average.
2. More Stringent Tier 2 NO_X and Gasoline Sulfur Standards
We considered whether average NO_X levels even lower than
0.07 g/mi (which would likely result in lower NO_X standards
for all of the Tier 2 certification bins and substantially limit the
number of vehicles certified at NO_X emissions levels
significantly higher than 0.07 g/mi) might be possible and cost
effective in a scenario where sulfur levels in gasoline would be
reduced to an average level on the order of 10 ppm (with perhaps a 20
ppm cap). Manufacturers have requested that California consider such a
``near zero'' sulfur limit to help them to meet the mandatory bins in
the CAL LEV II program, which are more stringent than what would be
required in the proposed Tier 2 program. We believe our proposed Tier 2
standards can be met with the proposed gasoline sulfur standards.
However, tighter Tier 2 standards could require even lower gasoline
sulfur limits.
We selected our proposed Tier 2 standards and gasoline sulfur
levels based on air quality need, technical feasibility, and cost
effectiveness. Hence, we believe the proposed requirements are
reasonable and are as stringent as is warranted. However, in
consideration of the alternative discussed here, we request comment on
the ability of manufacturers to produce vehicles meeting a corporate
average NO_X emission level substantially lower than 0.07 g/
mi. How would the cost of producing such a vehicle differ from the
costs estimated for the proposed Tier 2 vehicles? How sensitive would
such a vehicle be to the sulfur level of gasoline, and what sulfur
level would be required? How soon could manufacturers be expected to be
able to comply with a lower NO_X standard, given that they
will be producing LEVII vehicles for California beginning in 2004?
We also request comment on the magnitude of additional sulfur
reduction that would be necessary to reduce average full useful life
NO_X to levels significantly below 0.07 g/mi, and whether
such low levels of sulfur can be met with the technology EPA expects
refiners to use to meet the requirements we are proposing today. We
request comment on the costs of such sulfur reductions and the timing
needed to acquire and implement any additional refinery controls. If
refiners invest today to achieve 30 ppm average sulfur levels, will
those investments be rendered obsolete by a future sulfur requirement
of a near-zero average, or would the technologies complement one
another? How much time would refiners need to comply with a near-zero
sulfur standard following compliance with a 30 ppm standard?

V. Additional Elements of the Proposed Vehicle Program and Areas
for Comment

The section describes several additional provisions of the vehicle
proposal and issues on which we are requesting comment that were not
previously discussed in this preamble.

A. Other Vehicle-Related Elements of the Proposal

1. Proposed Tier 2 CO, HCHO and PM Standards
Table IV.B.-1 in Section IV.B.4.a. above presented the proposed
Tier 2 standards for carbon monoxide (CO), formaldehyde (HCHO), and
particulate matter (PM). The following paragraphs discuss our selection
of these specific standards for proposal.
a. Carbon Monoxide (CO) Standards. Beyond aligning carbon monoxide
(CO) standards for all LDVs and LDTs, and allowing harmonizing with
California vehicle technology, reduction in CO emissions is not a
primary goal of the Tier 2 program. Thus the CO standards we are
proposing for all Tier 2 LDVs and LDTs are essentially the same as
those from the NLEV program for LDVs and LDT1s. These standards would
harmonize with CalLEV II CO standards except at California's SULEV
level (EPA Bin 2). This lone divergence would not pose additional
burden to

[[Page 26081]]

manufacturers because the proposed federal Tier 2 CO standards for
these vehicles would be less stringent than California's. Our proposed
interim standards during the phase-in of Tier 2 standards would apply
these same CO standards.
As we indicated in the Tier 2 Report to Congress, the number and
severity of CO NAAQS violations have decreased greatly in recent years.
Presently, CO exceedances occur primarily during cold weather. The need
for more stringent cold CO standards is a subject of a separate EPA
study that is now underway. Consequently, in this rulemaking we propose
to simply align CO standards for all categories with those applicable
to LDVs and LDT1s under NLEV. This alignment is consistent with our
goal of bringing all LDVs and all categories of LDTs under common
standards that allow for technology to be harmonized to the extent
possible with California.
We believe that technological changes to bring LDT2s and HLDTs
⁷⁴ under tighter NMOG standards should easily ensure
compliance with the CO standards at no additional cost. In fact,
certification data on current model year LDTs indicate that there are
LDTs in all categories that can already meet the LDV/LDT1 NLEV CO
standard.
---------------------------------------------------------------------------

\74\ As defined earlier, the category called HLDT, or heavy
light-duty truck, includes all LDTs greater than 6000 pounds GVWR.
This term includes the categories LDT3 and LDT4.
---------------------------------------------------------------------------

We recognize that the vast majority of CO emissions are from motor
vehicles and that increases in population in some areas combined with
increases in vehicle miles traveled could lead to additional incidences
of CO nonattainment. Consequently, we request comment on the need for
and implications of tighter CO standards for any category of vehicles
affected by today's document.
b. Formaldehyde (HCHO) Standards. Similar to our approach to the
proposed CO standards, we are proposing to align all Tier 2 LDVs and
LDTs under the formaldehyde standards for LDVs and LDT1s from the NLEV
program. For new bins below Bin No. 4, we propose to adopt the CalLEV
II standards for formaldehyde. HLDTs, which are not subject to the NLEV
program, would become subject to HCHO standards for the first time
under the provisions of this rulemaking. The Tier 2 formaldehyde
standards would be essentially replicated in the interim standards we
are proposing for LDVs and LDTs.
Formaldehyde is a component of NMOG but is primarily of concern for
methanol-fueled vehicles, because it is chemically similar to methanol
and is likely to occur when methanol is not completely burned in the
engine. HLDTs are not included under the NLEV program and will
therefore not face formaldehyde standards as LDVs and LLDTs will in
2001 (1999 in the northeast states). We believe it is appropriate to
bring HLDTs under HCHO standards in this rulemaking. Applying
formaldehyde standards to HLDTs would be consistent with our goals of
aligning standards for all LDVs and LDTs regardless of fuel type and
harmonizing technologically with California standards wherever possible
and reasonable and the burden would be minimal.
Consequently, we are proposing to include formaldehyde standards
for HLDTs under the Tier 2 program as well as under the interim
programs. We note that HCHO is actually a component of NMOG, and as
with CO, we expect that all vehicles able to meet the Tier 2 or interim
NMOG standards (including methanol-fueled vehicles) would readily
comply with the HCHO standards.
c. Particulate Matter (PM) Standards. We are proposing to adopt
tighter PM standards, although in this case only full useful-life
standards. For Tier 2 vehicles, we are proposing a 0.01 g/mi standard
for all categories at the Tier 2 (Bin 5) level or below (except ZEV
which, of course, is 0.0). To provide manufacturers with additional
flexibility, we are proposing a 0.02 g/mi PM standard for vehicles that
certify to Bins 6 or 7 standards.
For non-Tier 2 LDV/LLDTs during the phase-in period, we are
proposing a PM standard of 0.06 g/mi for Bins 4 and 5. The other
standards would be 0.04 for Bin 3 and 0.01 for Bin 2. For non-Tier 2
HLDTs, similar standards would apply except that the highest bin would
have a PM standard of 0.06 g/mi, gradually decreasing in the other bins
to 0.01
g/mi (Bin 2).
PM standards are primarily a concern for diesel-cycle vehicles, but
they also apply to gasoline and other otto-cycle vehicles. We propose
to continue to permit otto-cycle vehicles to certify to PM standards
based on representative test data from similar technology vehicles. We
request comment on the degree to which these standards would affect the
certification of diesel-fueled vehicles.
2. Useful Life
The ``useful life'' of a vehicle is the period of time, in terms of
years and miles, during which a manufacturer is formally responsible
for the vehicle's emissions performance. For LDVs and LDTs, there have
historically been both ``full useful life'' values, approximating the
average life of the vehicle on the road, and ``intermediate useful
life'' values, representing about half of the vehicle's life. We are
proposing several changes to the current useful life provisions for
LDVs and LDTs.
a. Mandatory 120,000 Mile Useful Life. We are today proposing to
equalize full useful life values for all 2004 and later model year LDVs
and LDTs at 120,000 miles. This value would apply to Tier 2 and interim
non-Tier 2 vehicles. California, in its LEV II program, has adopted
full useful life standards for all LDVs and LDTs of 10 years or 120,000
miles, whichever occurs first. We are proposing that the time period
for federal LDV/LLDTs would be 10 years, but it would remain at 11
years for HLDTs consistent with the Clean Air Act.⁷⁵
Intermediate useful life values, where applicable, would remain at 5
years or 50,000 miles, whichever occurs first. Where manufacturers
elect to certify Tier 2 vehicles for 150,000 miles to gain additional
NO_X credits, as discussed below, the useful life of those
vehicles would be 15 years and 150,000 miles. We are not proposing to
harmonize with California on the mandatory useful life for evaporative
emissions of 15 years and 150,000 miles, but rather we are proposing
that this useful life be mandatory for evaporative emissions only when
a manufacturer elects optional 150,000 mile exhaust emission
certification.
---------------------------------------------------------------------------

\75\ Section 202(h) of the Clean Air Act specifies a useful life
of 11 years/120,000 miles for HLDTs. California is able to use a 10
year figure because it has a waiver under section 209 of the Act to
implement its own emission control program when such program is
found to be at least as protective of public health and welfare ``in
the aggregate'' as the federal program.
---------------------------------------------------------------------------

b. 150,000 Mile Useful Life Certification Option. We are proposing
to adopt a provision to provide additional NO_X credit in the
fleet average calculation for vehicles certified to a useful life of
150,000 miles. In our proposal, a manufacturer certifying an engine
family to a 150,000 mile useful life would incorporate those vehicles
into its corporate NO_X average as if they were certified to
a full useful life standard 0.85 times the applicable 120,000 mile
NO_X standard. To use this option, the manufacturer would
have to agree to (1) certify the engine family to the applicable
120,000 mile exhaust and evaporative standards at 150,000 miles for all
pollutants; and (2) increase the mileage on the single extra-high
mileage in-use test vehicle from a minimum of

[[Page 26082]]

90,000 miles to a minimum of 105,000 miles.
Congress, in directing EPA to perform the Tier 2 study, also
directed EPA to consider changing the useful lives of LDVs and LDTs.
Manufacturers have made numerous advances in quality, materials and
engineering that have led to longer actual vehicle lives and data show
that each year of a vehicle's life, people are driving more miles.
Current data indicate that passenger cars are driven approximately
120,000 miles in their first ten years of life. Trucks are driven
approximately 150,000 miles. Current regulatory useful lives are 10
years/100,000 miles for LDV/LLDTs and 11 years/120,000 miles for HLDTs.
We project based on our Tier 2 model that approximately 13 percent of
light-duty NO_X and 11 percent of light-duty VOCs is produced
between 100,000 and 120,000 miles. Given the trend toward longer actual
vehicle lives and increases in annual mileage, we believe that it is
reasonable to propose extension to the regulatory useful life
requirements.
Additionally, 41 percent of light-duty NO_X and 59
percent of light-duty VOC is produced beyond 120,000 miles. Based on
this data, we believe it is also appropriate to propose incentives to
manufacturers to certify their vehicles to extended useful lives beyond
120,000 miles. This is why we are proposing, as discussed above, to
provide additional NO_X credits for Tier 2 vehicles certified
to a useful life of 150,000 miles.
3. Light Duty Supplemental Federal Test Procedure (SFTP) Standards
Supplemental Federal Test Procedure (SFTP) standards require
manufacturers to control emissions from vehicles when operated at high
rates of speed and acceleration (the US06 test cycle) and when operated
under high ambient temperatures with air conditioning loads (the SC03
test cycle). The existing light duty SFTP requirements begin a three
year phase-in in model year 2000 for Tier 1 LDV/LLDTs . For HLDTs, SFTP
requirements begin a similar phase-in in 2002. Intermediate and full
useful life standards exist for all categories. SFTP standards do not
apply to diesel fueled Tier 1 LDT2s and HLDTs. Table V.A.-1 shows the
full useful life federal SFTP requirements applicable to Tier 1
vehicles.

Table V.A.-1.--Full Useful Life Federal SFTP Standards Applicable to Tier 1 Vehicles
----------------------------------------------------------------------------------------------------------------
NMHC + NO_X CO (g/mi) ^b
Vehicle category (weighted g/ -----------------------------------------------
mi) ^a US06 SC03 Weighted
----------------------------------------------------------------------------------------------------------------
LDV/LDT1 (gasoline)............................. 0.91 11.1 3.7 4.2
LDV/LDT1 (diesel)............................... 2.07 11.1 .............. 4.2
LDT2............................................ 1.37 14.6 5.6 5.5
LDT3............................................ 1.44 16.9 6.4 6.4
LDT4............................................ 2.09 19.3 7.3 7.3
----------------------------------------------------------------------------------------------------------------
^a Weighting for NMHC+NO_X and optional weighting for CO is 0.35 x (FTP)+0.28 x (US06)+0.37 x (SC03).
^b CO standards are stand alone for US06 and SC03 with option for a weighted standard.

The NLEV program includes SFTP requirements for LDVs, LDT1s and
LDT2s. These requirements impose the Tier 1 intermediate and full
useful life SFTP standards on Tier 1 and TLEV vehicles, but impose only
4000 mile standards on LEVs and ULEVs.⁷⁶ NLEV SFTP standards
for LEVs and ULEVs are shown in Table V.A.-2. These standards do not
provide for a weighted standard for NMHC+NO_X or for CO, but
rather employ separate sets of standards for the US06 and SC03 tests.
Also, while the NLEV SFTP standards apply to gasoline and diesel
vehicles, they do not include a standard for diesel particulates (PM).

\76\ This disparity in useful lives arose because neither EPA
nor CARB had full useful life SFTP standards for LEVs or ULEVs when
the NLEV program was adopted. Since a major requirement of the NLEV
program was harmony with California standards, EPA adopted the
California SFTP standards in place for the NLEV time frame (2001 and
later).

Table V.A.-2.--SFTP Standards for LEVs and ULEVs in the NLEV Program
----------------------------------------------------------------------------------------------------------------
US06 SC03
---------------------------------------------------------------
NMHC+NO_X (g/ NMHC+NO_X (g/
mi) CO (g/mi) mi) CO (g/mi)
----------------------------------------------------------------------------------------------------------------
LDV/LDT1........................................ 0.14 8.0 0.20 2.7
LDT2............................................ 0.25 10.5 0.27 3.5
----------------------------------------------------------------------------------------------------------------

Since no significant numbers of vehicles certified to SFTP
standards below TLEV levels will enter the fleet until 2001,
manufacturers have raised concerns regarding significant changes to the
SFTP program before its implementation. At this point, it seems
reasonable not to increase SFTP stringency for the Tier 2 program, but
we are proposing to substitute SFTP standards adjusted for intermediate
and full useful life deterioration where there are currently only 4000
mile standards.
Full useful life standards for Tier 2 vehicles are consistent with
our mandate under the Clean Air Act. The 4000 mile standards exist in
the federal program only because they were adopted in the NLEV
program--a voluntary program under which California requirements were
adopted nationwide. We derived the full and intermediate useful life
standards by applying deterioration allowances proposed for our MOBILE
6 model to the existing 4000 mile standards for LDVs and LLDTs. For
HLDTs we applied similarly derived deterioration allowances to
California's LEV I SFTP standards for MDV2s and MDV3s, which are the
corresponding categories to LDT3s and LDT4s in the California program.
The full and intermediate useful life SFTP standards we are proposing
are shown in Tables V.A.-3

[[Page 26083]]

and V.A.-4. These standards would apply to all Tier 2 vehicles
including Tier 2 LDT3s and LDT4s.

Table V.A.-3.--Proposed Full Useful Life Supplemental Emission Standards
[(SFTP Standards (grams/mile)]
----------------------------------------------------------------------------------------------------------------
USO6 NMHC+NO_X USO6 CO SCO3 NMHC+NO_X SCO3 CO
----------------------------------------------------------------------------------------------------------------
LDV/LDT1........................................ 0.2 11.1 0.26 4.2
LDT2............................................ 0.37 14.6 0.39 5.5
LDT3............................................ 0.53 16.9 0.44 6.4
LDT4............................................ 0.78 19.3 0.62 7.3
----------------------------------------------------------------------------------------------------------------

Table V.A.-4.--Proposed Intermediate Useful Life Supplemental Emission Standards
[(SFTP Standards)(grams/mile)]
----------------------------------------------------------------------------------------------------------------
USO6 NMHC+NO_X USO6 CO SCO3 NMHC+NO_X SCO3 CO
----------------------------------------------------------------------------------------------------------------
LDV/LDT1........................................ 0.16 9.0 0.22 3.0
LDT2............................................ 0.30 11.6 0.32 3.9
LDT3............................................ 0.45 11.6 0.36 3.9
LDT4............................................ 0.67 13.2 0.51 4.4
----------------------------------------------------------------------------------------------------------------

Because our proposed interim standards for LDV/LLDTs (see section
VI.A.3.d. above) are derived from NLEV standards, we believe that the
SFTP standards we are proposing for Tier 2 vehicles should also apply
to the interim non-Tier 2 LDV/LLDTs. However, we propose that TLEV
vehicles (EPA interim Bin 5 in Table IV.B.-6), which are not subject to
new SFTP standards under NLEV, could continue to meet Tier 1 SFTP
standards, and HLDTs under the interim programs could continue to meet
Tier 1 SFTP standards that do not fully phase in until the 2004 model
year.
LDT3 and LDT4 SFTP standards do not currently apply to diesels.
Further, the standards applicable to Tier 1 diesel LDVs and LDT1s are
less stringent than gasoline standards and do not apply to the SC03
cycle. We are proposing to apply the approach we are using with other
standards in this document to the Tier 2 and interim SFTP standards.
Consequently, we are proposing that Tier 2 and interim LDVs and LDTs
with diesel or gasoline engines comply with the same
NMHC+NO_X and CO SFTP limits. We are also requesting comment
on the appropriate SFTP PM standards for diesel vehicles. We believe it
would be appropriate to establish a margin between 10% and 50% above
the applicable FTP PM standard to serve as the SFTP standard. As an
example of how EPA has recently used such a margin, in recent consent
decrees, heavy-duty engine manufacturers have agreed not to exceed
emission levels 1.25 times the applicable exhaust standards (including
PM standards) when engines are operated over a wide range of operating
conditions. We request comment on the appropriate standard for PM in
the SFTP.
4. LDT Test Weight
Historically, HLDTs (LDT3s and LDT4s) have been emission tested at
their adjusted loaded vehicle weight (ALVW), while LDVs, LDT1s, and
LDT2s have been tested at their loaded vehicle weight (LVW). ALVW is
equivalent to the curb weight of the truck plus half its maximum
payload, while LVW is equivalent to the curb weight of the truck plus a
driver and one adult passenger (300 pounds). As we are proposing in
this document to equalize standards and useful lives across LDVs and
all categories of LDTs, we believe it is appropriate to test all the
vehicles under the same conditions. Therefore, consistent with the
CalLEV II program, we are proposing to test HLDTs at their loaded
vehicle weight. We recognize that removing all but 300 pounds of load
from these trucks during the test provides them with a somewhat
``easier'' test cycle than they currently have. However, the standards
we are proposing for HLDTs under Tier 2, are considerably more
stringent than the Tier 1 standards. Further, one of our reasons for
bringing HLDTs under the same standards as passenger cars is that these
trucks include many vans and sport utility vehicles that are often used
as passenger cars with just one or two passengers. Consequently, we
believe it is appropriate to test them at LVW.
5. Test Fuels
As discussed elsewhere in this preamble, the NLEV program was
adopted virtually in its entirety from California's program. Because
California's standards were developed around the use of California
Phase II reformulated gasoline (RFG) as the exhaust emission test fuel,
we adopted California Phase II test fuel as the exhaust emission test
fuel for gasoline-fueled vehicles in the federal NLEV program, although
we recognized at the time that vehicles outside of California would be
unlikely to operate on that fuel in use.
We believe that it is best to establish compliance with standards
based on the fuel that the vehicles will operate upon. However, we also
believe that the major exhaust emission related issues between
California Phase II fuel and federal test fuel are related to sulfur
and we do not believe the other differences between the two fuels will
significantly impact NMOG, CO or NO_X exhaust emissions in
Tier 2 (or interim) gasoline fueled vehicles.
In this document, we are proposing to reduce the sulfur in federal
test fuel to reflect the reductions in sulfur we are proposing for
commercial gasoline. Currently, federal test gasoline is subject to a
limit of 0.10 percent by weight. We are proposing to amend that to an
allowable range of 30 to 80 ppm (0.003 to 0.008 percent by weight). We
also propose that vehicles be certified and in-use tested using federal
test fuel. However, where vehicles are certified for 50 state sale, and
where other testing issues do not arise, we are proposing to accept the
results of testing done for California certification on California
Phase II fuel. We would reserve the right to perform or require in-use
testing on

[[Page 26084]]

federal fuel. Where vehicles are only certified for non-California
sale, we propose to require certification and in-use testing on federal
fuel. We request comments with supporting emission data on all aspects
of these two possible test fuels.
Because differences exist between the California and federal
evaporative emission testing procedures, we propose to continue to
require the use of federal certification fuel as the test fuel in
evaporative emission testing. Under current programs, where California
and federal evaporative emission standards are nearly identical,
California accepts evaporative results generated on the federal
procedure (using federal test fuel), because available data indicates
the federal procedure to be a ``worst case'' procedure. The evaporative
standards California has adopted for their LEV II program are more
stringent than those we are proposing in this document. We request
comment and supporting emission test data on whether vehicles certified
to CalLEV II evaporative standards using California fuels will
necessarily comply with the federal Tier 2 evaporative standards,
including ORVR standards, when tested with federal test fuel.
6. Changes to Evaporative Certification Procedures to Address Impacts
of Alcohol Fuels
Current certification procedures, including regulations under the
CAP2000 program,⁷⁷ allow manufacturers to develop their own
durability process for calculating deterioration factors for
evaporative emissions. The regulations (Sec. 86.1824-01) permit
manufacturers to develop service accumulation (aging) methods based on
``good engineering judgement'', subject to review and approval by EPA.
The manufacturer's durability process must be designed to predict the
expected evaporative emission deterioration of in-use vehicles over
their full useful lives. We are proposing to require that these aging
methods include the use of alcohol fuels to address concerns that
alcohol fuels increase the permeability and thus the evaporative losses
from hoses and other evaporative components.
---------------------------------------------------------------------------

\77\ The Compliance Assurance Program, CAP2000, was proposed in
an NPRM (63 FR 39654, July 23, 1998). The final rule was signed on
March 15, 1998. As today's NPRM went forward for signature, the
CAP2000 final rule had not been published, so no citation for the
final rule is available. You should check our web site (http://
www.epa.gov/omswww/) for the most current information on publication
of the CAP2000 rule takes effect in the 2000 model year.
---------------------------------------------------------------------------

We have reviewed data indicating that the permeability, and
therefore the evaporative losses, of hoses and other evaporative
components can be greatly increased by exposure to fuels containing
alcohols.⁷⁸ Alcohols have been shown to promote the passage
of hydrocarbons through a variety of different materials commonly used
in evaporative emission systems. Data from component and fuel line
suppliers indicate that alcohols cause many elastomeric materials to
swell, which opens up pathways for hydrocarbon permeation and also can
lead to distortion and tearing of components like ``O'' ring seals.
Ethers such as MTBE and ETBE have a much smaller effect. Alcohol-
resistant materials such as fluoroelastomers are available and are
currently used by manufacturers to varying extents.
---------------------------------------------------------------------------

\78\ Numerous SAE papers examine the permeability of fuel and
evaporative system materials as well as the influence of alcohols on
permeability. See, for example SAE Paper #s 910104, 920163, 930992,
970307, 970309, 930992, and 981360, copies of which are in the
docket for this rulemaking.
---------------------------------------------------------------------------

Alcohols do not impact evaporative components and hoses
immediately, but rather it may take as long as one year of exposure to
alcohol fuels for permeation rates to stabilize. The end result in
higher permeation and increased in-use evaporative emissions.\79\
---------------------------------------------------------------------------

\79\ Ibid.
---------------------------------------------------------------------------

Today, roughly 10% of fuel sold in the U.S. contains alcohol,
mainly in the form of ethanol, and such fuels are often offered in
ozone nonattainment areas. We believe it is appropriate to ensure that
evaporative certification processes expose evaporative components to
alcohols and do so long enough to stabilize their permeability.
Therefore, we are proposing to amend evaporative certification
requirements to require manufacturers to develop their deterioration
factors using a fuel that contains the highest legal quantity of
ethanol available in the U.S.
To implement this change, we are proposing to modify the Durability
Demonstration Procedures for Evaporative Emissions found at
Sec. 86.1824-01. Our proposal would require manufacturers to age their
systems using a fuel containing the maximum concentration of alcohols
allowed by EPA in the fuel on which the vehicle is intended to operate,
i.e., a ``worst case'' test fuel. (Under current requirements, this
fuel would be about 10% ethanol, by volume.) We are also proposing to
modify the Durability Demonstration Procedures to require manufacturers
to ensure that their aging procedures are of sufficient duration to
stabilize the permeability of the fuel and evaporative system
materials.
It is our desire to find an alternative way by which a manufacturer
could document or demonstrate that its tanks, hoses, connectors and
other evaporative components are made of materials whose permeability
is not significantly affected by alcohols. Successful manufacturers
would not have to use alcohol fuel in certification. There are a
variety of test methods to evaluate permeation losses from materials,
components or subassemblies described in the literature.⁸⁰
However, from our discussions with component and materials suppliers,
we conclude that there is currently no consensus test procedure or
standard available that we could rely on to establish whether a fuel/
evaporative system is likely to be sufficiently impermeable to alcohol
fuels. We request comment on the availability and appropriateness of
such procedures and standards and we request comment on the need for
and benefits of certification enhancements to account for the effects
of alcohols in fuels. We also seek comment on whether certification
test fuel for evaporative emissions should include 10% ethanol.
---------------------------------------------------------------------------

\80\ Ibid.
---------------------------------------------------------------------------

7. Other Test Procedure Issues
California's LEV II program implements a number of minor changes to
exhaust emissions test procedures. We have evaluated these changes and
found that, for tailpipe emissions, the California test procedures fall
within ranges and specifications permitted under the Federal Test
Procedure.
With regard to HEVs and ZEVs, we believe that these vehicles will
be predominantly available in California, or that they will typically
be first offered for sale in California, because of California's ZEV
requirement, which promotes the sale of HEVs and ZEVs. Where
manufacturers market HEVs or ZEVs outside of California, it is likely
that they will market the same vehicles in California. Consequently, we
intend to incorporate by reference California's exhaust emission test
procedures for HEVs and ZEVs.⁸¹ We request comment on the
appropriateness of this proposed incorporation and an emission
allowance for HEVs.
---------------------------------------------------------------------------

\81\ California Zero-Emission and Hybrid Electric Vehicle
Exhaust Emission Standards and Test Procedures for 2003 and
Subsequent Model Year Passenger Cars, Light-Duty Trucks and Medium-
Duty Vehicles. September 18, 1998 for the Board Hearing of November
5, 1998.
---------------------------------------------------------------------------

In the NLEV program, we provided a specific formula used by
California that could be used to compute an HEV contribution factor to
NMOG emissions. This formula took into consideration the

[[Page 26085]]

range without engine operation of various types of HEVs and had the
effect of reducing the NMOG emission standard for a given emission bin
(for HEV vehicles only). This would have obvious beneficial effects on
a manufacturer's calculation of its corporate NMOG average.
The technology of HEVs is under rapid change and we do not believe
that we can design a formula now that will accurately predict the
impact of HEVs on corporate average NO_X emissions in the
Tier 2 time frame. Consequently, we are including a provision by which
manufacturers could propose HEV contribution factors for NO_X
to EPA. If approved, these factors could be used in the calculation of
a manufacturer's fleet average NO_X emissions and would
provide a mechanism to credit an HEV for operating with no emissions
over some portion of its life.
These factors would be based on good engineering judgement and
would consider such vehicle parameters as vehicle weight, the portion
of the time during the test procedure that the vehicle operates with
zero emissions, the zero emission range of the vehicle, NO_X
emissions from fuel-fired heaters and any measurable NO_X
emissions from on-board electricity production and storage.
The final NLEV rule (See 62 FR pg 31219, June 6, 1997) incorporates
by reference California's NMOG measurement procedure and adopts
California's approach of using Reactivity Adjustment Factors (RAFs) to
adjust vehicle emission test results to reflect differences in the
impact on ozone formation between an alternative-fueled vehicle and a
vehicle fueled with conventional gasoline. While we intend to bring all
LDVs and LDTs under NMOG standards beginning in 2004 and while we
desire to harmonize with California when practical and reasonable, we
are not proposing to allow the use of RAFs for Tier 2 vehicles and
interim non-Tier 2 vehicles. As has been discussed elsewhere in this
preamble, the NLEV program is a special case in which California
standards and provisions were adopted virtually in their entirety. In
the preamble to the final NLEV rule (See 62 FR 31203), we expressed our
reservations about the use of RAFs. We also addressed our reservations
about the use of reactivity factors developed in California in a
program that spans a range of climate and geographic locations across
the United States in the final rule on reformulated gasoline (RFG) (see
59 FR 7220). We are concerned about the validity of RAFs to predict
ozone formation nationwide and have asked the National Academy of
Sciences to look at the scientific evidence in support of the use of
these factors nationwide. We expect to receive their report prior to
making our final decisions about the Tier 2 standards.
Recognizing that we are not proposing a corporate average NMOG
standard, and that RAFs impact only the calculation of NMOG emissions,
we request comment on all aspects of RAFs including the impact of not
using them on the severity of our proposed standards, their validity to
predict ozone formation nationwide, and any impact the lack of RAFs may
have on alternative fueled vehicles.
In its LEV II program, California is also implementing a number of
changes to evaporative emission test procedures.⁸² Many of
these changes address the evaporative emission testing of hybrid
electric vehicles. We are generally not proposing to adopt California's
changes, because California uses different test temperatures and
different test fuel in its evaporative emission testing of gasoline
vehicles than we use in the federal program. The preamble to the final
NLEV rule (See 62 FR 31227) explains that California and EPA are
reviewing an industry proposal to streamline and reconcile the
California and federal procedures. That work has not been completed.
However, where California proposes procedures specific to HEVs and
ZEVs, we do intend to adopt those procedures, except that our testing
would occur at lower temperatures, and use a fuel determined by EPA to
be representative of federal usage (for HEVs only). Given the small
number of HEVs and ZEVs likely to be sold in states other than
California early in the Tier 2 program, and given the small quantities
of fuel likely to be used by HEVs in any event, we request comment on
the appropriateness of simply accepting California evaporative results
for HEVs and ZEVs to show compliance with the less stringent federal
evaporative standards. We also request comment on whether any or all of
the changes California has adopted for evaporative emission testing
should be adopted into federal testing requirements.
---------------------------------------------------------------------------

\82\ California Evaporative Emission Standards and Test
Procedures for 2001 and Subsequent Model Motor Vehicles; September
18, 1998. Prepared for the November 5, 1998 Hearing of the
California Air Resources Board.
---------------------------------------------------------------------------

8. Small Volume Manufacturers
Our proposal includes the following flexibilities intended to
assist all manufacturers in complying with the stringent proposed
standards without harm to the program's environmental goals: (1) A four
year phase-in of the standards for LDV/LLDTs; (2) a delayed phase-in
for HLDTs; (3) the freedom to select from specific bins of standards;
(4) a standard that can be met through averaging, banking and trading
of NO_X credits; (5) provisions for NO_X credit
deficit carryover; and (6) provisions by which a manufacturer may
generate additional NO_X credits.
These flexibilities would apply to all manufacturers, regardless of
size, and in general we believe they eliminate the need for more
specific provisions for small volume manufacturers. However, we are
proposing one additional flexibility for small volume
manufacturers.⁸³ Our proposal would exempt small volume
manufacturers from the 25%, 50% and 75% Tier 2 phase-in requirements
applicable to the 2004, 2005 and 2006 LDV/LLDTs and the 50% phase-in
requirement applicable to 2008 HLDTs. Instead, small volume
manufacturers would simply comply with the appropriate 100% requirement
in the 2007 or 2009 model year. Our proposal would also exempt small
volume manufacturers from the 25%, 50% and 75% phase-in requirements
applicable to interim HLDTs in 2004-2006. Instead, small volume HLDT
manufacturers would simply comply with the interim standards, including
the corporate average NO_X standard, in 2007 for 100% of
their vehicles. During model years 2004-2006, these same small volume
manufacturers would comply with any of the interim bins of HLDT
standards for 100% of their HLDTs.⁸⁴
---------------------------------------------------------------------------

\83\ We define small volume manufacturers to be those with total
U.S. sales of less than 15,000 highway units per year. Independent
commercial importers (ICIs) with sales under 15,000 per year would
be included under this term.
\84\ For a graphical illustration of the phase-ins through time,
see Figure IV.B.-1.
---------------------------------------------------------------------------

Also, we will continue to apply the federal small volume
manufacturer provisions, which provide relief from emission data and
durability showing and reduce the amount of information required to be
submitted to obtain a certificate of conformity. In addition, the
CAP2000 program contains reduced in-use testing requirements for small
volume manufacturers. Under section V.B.1. below, we describe and
request comment on possible additional special provisions for
certifiers that qualify as small businesses.
Our proposal to exempt small volume manufacturers from the Tier 2
phase-in requirements eliminates a dilemma that the phase-in
percentages might pose to a manufacturer that has a limited product
line, i.e., how to address percentage phase-in requirements if the

[[Page 26086]]

manufacturer makes vehicles in only one or two test groups. We have
proposed similar provisions for small entities in other rulemakings.
Approximately 15-20 manufacturers that currently certify vehicles, many
of which are independent commercial importers (ICIs), would qualify.
These manufacturers represent just a fraction of one percent of LDVs
and LDTs produced. We do not believe that this provision would have any
measurable impact on air quality.
9. Compliance Monitoring and Enforcement
a. Application of EPA's Compliance Assurance Program, CAP2000. The
CAP2000 program (final rule signed March 15, 1998; Federal Register
cite not yet available) streamlines and simplifies the procedures for
certification of new vehicles and would also require manufacturers to
test in-use vehicles to monitor compliance with emission standards. The
CAP2000 program was developed jointly with the State of California and
involved considerable input and support from manufacturers. As the name
implies, it can be implemented as early as the 2000 model year.
In today's document, we are proposing that the Tier 2 and the
interim requirements would be implemented subject to the requirements
of the CAP2000 program. Certain CAP2000 requirements would be slightly
modified to reflect changes to useful lives, standard structure and
other aspects of the Tier 2 program, but we are proposing no major
changes to fundamental principles of the CAP2000 program.
Although we are proposing changes to useful lives in this document,
we are not proposing to amend the 50,000 mile minimum mileage used in
manufacturer in-use verification testing or in-use confirmatory testing
under the CAP2000 program at this time. The CAP2000 in-use program is
not yet implemented and we believe it is appropriate to allow
manufacturers to gain experience with procuring and testing vehicles at
the 50,000 mile level before making significant changes. However, where
one vehicle from each in-use test group would have a minimum mileage of
75,000 miles under the CAP2000 program, we are proposing, consistent
with California, to change that figure to 90,000 miles for Tier 2
vehicles.
We may, in our own in-use program, procure and test vehicles at
mileages higher than 50,000 and pursue remedial actions (e.g. recalls)
based on that data. We may also use that data as the basis to initiate
a rulemaking to make changes in theCAP2000 in-use requirements, if the
data indicate significant non-conformity at higher mileages.
b. Compliance Monitoring. We plan no new compliance monitoring
activities or programs for Tier 2 vehicles. These vehicles would be
subject to the certification and manufacturer in-use testing provisions
of the CAP2000 rule. Also, we expect to continue our own in-use testing
program for exhaust and evaporative emissions. We will pursue remedial
actions when substantial numbers of properly maintained and used
vehicles fail any standard in either in-use testing program.
We retain the right to conduct Selective Enforcement Auditing of
new vehicles at manufacturer's facilities. In recent years, we have
discontinued SEA testing of new light-duty vehicles and trucks, because
compliance rates were routinely at 100%. We recognize that the need for
SEA testing may be reduced by the low mileage in-use testing
requirements of the CAP2000 program. However, we expect to re-examine
the need for SEA testing as standards tighten under the NLEV and Tier 2
programs.
We have established a data base to record and track manufacturers'
compliance with NLEV requirements including the corporate average NMOG
standards. We expect to monitor manufacturers' compliance with the Tier
2 and interim corporate average NO_X standards in a similar
fashion and also to monitor manufacturers' phase-in percentages for
Tier 2 vehicles.
c. Relaxed In-Use Standards for Tier 2 Vehicles Produced During the
Phase-in Period. As we have indicated numerous times in this preamble,
the Tier 2 standards we are proposing would be challenging for
manufacturers to achieve, and some vehicles would pose more of a
challenge than others. Not only would manufacturers be responsible for
assuring that vehicles can meet the standards at the time of
certification, they would also have to ensure that the vehicles could
comply when tested in-use by themselves under the provisions of the
CAP2000 program, and by EPA under its in-use (``Recall'') test program.
With any new technology, or even with new calibrations of existing
technology, there are risks of in-use compliance problems that may not
appear in the certification process. In-use compliance concerns may
discourage manufacturers from applying new technologies or new
calibrations. Thus, it may be appropriate for the first few years, for
those bins most likely to require the greatest applications of effort,
to provide assurance to the manufacturers that they will not face
recall if they exceed standards by a specified amount.
We are proposing, for Tier 2 vehicles only, that for the first two
years after a test group meeting a new standard is introduced, that
test group be subject to more lenient in-use standards. These ``in-use
standards'' would apply only to Tier 2 Bins 5 and below, only for the
pollutants indicated, and only for the first two model years that a
test group was certified under that bin. The in-use standards would not
be applicable to any test group first certified to a new standard after
2007 for LDV/LLDTs or after 2009 for HLDTs.
The in-use standards we are proposing are shown in Table V.A.-5
below.

Table V.A.-5.--In-use Compliance Standards for Tier 2 Vehicles (g/mi)
[Certification standards shown for reference purposes]
--------------------------------------------------------------------------------------------------------------------------------------------------------
Durability NO_X
Bin No. period (miles) NO_X In-use certification NMOG in-use NMOG certification
--------------------------------------------------------------------------------------------------------------------------------------------------------
5, 4........................................... 50,000 0.07 0.05 N/a 0.075, 0.04.
5, 4........................................... 120,000 0.10 0.07 N/a 0.090, 0.055.
3.............................................. 120,000 0.06 0.04 N/a 0.070.
2.............................................. 120,000 0.03 0.02 0.02 0.010.
--------------------------------------------------------------------------------------------------------------------------------------------------------

We believe manufacturers should and will strive to meet the Tier 2
certification standards for the full useful lives of the vehicles, but
we recognize that the existence of such in-use standards poses some
risk that a

[[Page 26087]]

manufacturer might aim for the in-use standard in its design efforts
rather than the certification standard, and thus market less durable
designs. We do not believe that risk to be significant. We believe that
such risks are more than balanced by the gains that could result from
earlier application of new technology or new calibration techniques
that might occur in a scenario where in-use liability is slightly
reduced. Further, we believe that the in-use standards will be of short
enough duration that any risks are minimal.
We note that the in-use provisions proposed above are similar to
those included in California's LEV II program. We request comment on
all aspects of the proposed in-use standards including the
appropriateness of and need for separate in-use compliance standards
for the early years of the Tier 2 program.
d. Enforcement of the Tier 2 and Interim Corporate Average
NO_X Standards. Under the proposed programs, manufacturers
could either report that they met the relevant corporate average
NO_X standard in their annual reports to the Agency or they
could show via the use of NO_X credits that they have offset
any exceedence of the corporate average NO_X standard.
Manufacturers would also report their NO_X credit balances or
deficits.
The averaging, banking and trading program would be enforced
through the certificate of conformity that the manufacturer would need
to obtain in order to introduce any regulated vehicles into commerce.
The certificate for each test group would require all vehicles to meet
the applicable Tier 2 emission standards from the applicable bin of the
Tier 2 program, and would be conditioned upon the manufacturer meeting
the corporate average NO_X standard within the required time
frame. If a manufacturer failed to meet this condition, the vehicles
causing the corporate average NO_X exceedence will be
considered to be not covered by the certificate of conformity for that
engine family. A manufacturer would be subject to penalties on an
individual vehicle basis for sale of vehicles not covered by a
certificate. These provisions would also apply to the interim corporate
average standards.
As outlined in detail in the preamble to the final NLEV rule, EPA
would review the manufacturer's sales to designate the vehicles that
caused the exceedence of the corporate average NO_X standard.
We would designate as nonconforming those vehicles in those test groups
with the highest certification emission values first, continuing until
a number of vehicles equal to the calculated number of noncomplying
vehicles as determined above is reached. In a test group where only a
portion of vehicles would be deemed nonconforming, we would determine
the actual nonconforming vehicles by counting backwards from the last
vehicle produced in that test group. Manufacturers would be liable for
penalties for each vehicle sold that is not covered by a certificate.
We are proposing in today's action to condition certificates to
enforce the requirements that manufacturers not sell NO_X
credits that they have not generated. A manufacturer that transferred
NO_X credits it did not have would create an equivalent
number of debits that it would be required to offset by the reporting
deadline for the same model year. Failure to cover these debits with
NO_X credits by the reporting deadline would be a violation
of the conditions under which EPA issued the certificate of conformity,
and nonconforming vehicles would not be covered by the certificate. EPA
would identify the nonconforming vehicles in the same manner described
above.
In the case of a trade that resulted in a negative credit balance
that a manufacturer could not cover by the reporting deadline for the
model year in which the trade occurred, we propose to hold both the
buyer and the seller liable. This is consistent with other mobile
source rules, except for the NLEV rule as discussed below. We believe
that holding both parties liable will induce the buyer to exercise
diligence in assuring that the seller has or will be able to generate
appropriate credits and will help to ensure that inappropriate trades
do not occur.
In the NLEV program we implemented a system in which only the
seller of credits would be liable. In the preamble to the final NLEV
rule (See 62 FR 31216), we explained that a multiple liability approach
would be unnecessary in the context of the NLEV program given that the
main benefit to a multi-party liability approach would be to ``protect
against a situation where one party sells invalid credits and then goes
bankrupt, leaving no one liable for either penalties or compensation
for the environmental harm.'' Our preamble stated further that EPA
would not necessarily take the same approach for ``other differently
situated trading programs.''
The NLEV program was implemented to be a relatively short duration
program, during which time we could expect relative stability in the
industry. Also, given that NLEV is a voluntary program of lower than
mandated standards, we did not expect that the smallest manufacturers
would opt in. These are the companies whose stability is most in
jeopardy in a dynamic and very competitive worldwide business.
We currently believe that the Tier 2 program and its framework will
remain for many years. We note that the program is not scheduled for
complete phase-in for almost nine years after the publication of this
proposal. All manufacturers, large and small, will ultimately have to
meet the Tier 2 standards. We cannot predict that in the Tier 2 time
frame there will not be companies that leave the market or are divided
between other companies in mergers and acquisitions. Thus we believe it
is prudent to implement a program to provide inducements to the seller
to assure the validity of any credits that it purchases or contracts
for. However, we request comment on whether we should implement a
program that would only deem the seller to be in violation if it sold
credits it could not supply.
10. Miscellaneous Provisions
We are proposing to continue existing emission standards from Tier
1 and NLEV that apply to cold CO, certification short testing,
refueling, running loss, idle CO for LDTs, and highway NO_X.
We are not proposing to continue the 50 degree (F) standards and
testing included in the NLEV program. The 50 degree standards are a
part of the NLEV program because that national program adopted
California requirements virtually in their entirety. These standards
had not previously been part of any federal program. We request comment
on the need and the associated burden for any of the standards
mentioned in this paragraph.

B. Other Areas on Which We are Seeking Comment

1. LDV/LDT Program Options
The alternatives for which we seek comment would have impacts on
the level of emission reductions achieved by the program as well as on
the cost and technological impacts of the program. Any decision to
adopt an alternative would have to consider those factors. We welcome
comments on all of the options described below. Commenters should
address cost, technological feasibility and emission impact whenever
possible.
a. Alternatives to Address Stringency of the Standards.
i. Alternative Standards and Implementation Schedules.
We believe that the Tier 2 standards and phase-in schedule
contained in this proposal provide appropriate lead time and
flexibility for manufacturers to

[[Page 26088]]

achieve cost-effective emission reductions in a reasonable time period.
Further, our standards and phase-in schedules are reasonably harmonized
with California's LEV II program to facilitate the sale of 50-state
vehicles and to minimize the administrative burdens involved with
having to meet the requirements of both California and EPA
simultaneously. We believe our proposed fuels provisions will ensure
that appropriate fuels are available to enable Tier 2 vehicles to
provide substantive in-use emission reductions. Some have suggested
delays in the program to 2007 and later. However, many states need
reductions as soon as possible for 2007 NAAQS compliance, so there is a
need for an aggressive but achievable implementation schedule.
Nevertheless, we are interested in reviewing alternative standards,
implementation schedules and averaging schemes. Therefore we request
comment on all aspects of the standards and schedules we are proposing
today, including the interim standards and schedules, and we request
comment on what alternative standards and implementation approaches
might provide comparable emission reductions that are cost-effective in
the same time frame as our proposal.
We recognize that the Tier 2 program as proposed today does not
provide for further reductions in average certification levels after
2008 as California's LEV II program does. We request comment on the
technological feasibility, necessity, cost and likely benefits of
further reductions in corporate average standards after 2009, including
comments on the reduction of the corporate average NO_X
standard to a level of approximately 0.05 g/mi in the 2011-2012 time
frame. We also request comment on a traditional, non-averaging standard
of 0.07 g/mi NO_X with related standards for NMOG, CO, HCHO,
and PM in the 2011-2012 time frame, applicable to all LDVs and LDTs.
ii. Use of Family Emission Limits (FELs) Rather than Bins.
A bins-based program with an overarching corporate average standard
has worked well in California for many years and is being implemented
nationwide beginning in 1999 under the NLEV program. We believe that a
phased in, bins-based program is the best way to implement the Tier 2
exhaust emission standards and, at the same time, encourage the
development of advanced emission control technology. We believe that
manufacturers of light duty vehicles and trucks are accustomed to such
programs and will appreciate the flexibility and opportunities for 50-
state certification that a bins-based program affords.
We are aware, of course, that in other EPA mobile source emission
programs, we have implemented averaging standards that were not based
upon bins. In these programs, manufacturers declare a family emission
limit (FEL) either above or below the averaging standard set by EPA.
The FEL becomes the standard for that family. Similar to the bins
approach, manufacturers compute a sales weighted average for the
subject pollutant at the end of the model year and then determine
credits generated or needed based on the distance of that average above
or below the standard.
In an FEL based program, every test group can have a different
FEL--essentially there is an unlimited continuum of bins to choose from
(although there is usually an upper limit or cap on the FELs). The FEL
approach adds flexibility and could increase the incentive for cost-
effective improvements in vehicle emissions performance. Under a bins
approach, a manufacturer is limited to step-wise improvements. An FEL
approach could provide incentive for manufacturers to realize smaller,
low cost emissions improvements that could be achieved, for example,
through engine re-calibration.
However, FEL-based programs create other concerns. One concern with
an FEL approach is that it may be viewed as providing too much
flexibility since a manufacturer could request a change in an FEL based
on a change in desired compliance margin above the certification level
or based on concern about its credit balance rather than a change in
technology. In EPA's FEL-based programs, it is not uncommon for a
manufacturer to declare an FEL that is identical to its certification
level. It is also not uncommon for a manufacturer to change its FEL
several times during a model year, based, among other reasons, on the
availability of or need for credits. In a bins approach, such changes
are unlikely, since a change in bins involves more of an increment in
emissions and involves compliance with all pollutants in that bin.
Consequently, a bins approach eases EPA's compliance monitoring burden.
It provides additional assurance that expected emission reductions will
occur in use because some vehicles may ``over-qualify'' for their bin
resulting in greater than expected reductions than if they exactly met
the standard for that bin. Of course, an FEL approach could be modified
to restrict or prohibit changes in certification levels during a model
year.
Also, in an FEL-based program, it may be necessary to establish
corporate average standards for other pollutants besides
NO_X. These standards would then require manufacturers to
establish FELs for additional pollutants. In a bins-based program, the
standards for the other pollutants are simply set by the different
bins.
An FEL approach could also lead to additional complexity in
manufacturer in-use testing under the CAP2000 program and in EPA in-use
testing because if FEL changes are made, the issue of which standard to
measure compliance against arises as does the issue of how many
vehicles to test for each different FEL. If we were to adopt an FEL
approach, we would have to consider significant changes to the in-use
provisions of the CAP2000 program to assure that all variations of a
test group were adequately covered by manufacturer in-use testing.
We request comment on the appropriateness and need for an FEL-based
program for the Tier 2 and/or interim standards. Commenters supporting
the use of an FEL-based program should also provide comment as to how
EPA can best manage the issues related to in-use testing and how EPA
can best assure that FEL changes are closely linked to real changes in
vehicle emissions.
iii. Use of Different Averaging Sets.
We chose for our proposal the broadest possible--and therefore most
flexible--averaging set for the Tier 2 vehicles. We are proposing that,
beginning in 2009 when phase-in of all vehicles is complete, all LDVs
and LDTs could be averaged together to meet the corporate average
NO_X standard. We believe this approach is appropriate
because it treats LDTs like LDVs, considering that LDTs are used as
passenger cars much of the time. Also, by permitting this broad
averaging, a manufacturer of larger LDTs that might have difficulty
meeting a 0.07 g/mi NO_X level can certify the LDTs to Bin 6
or 7 and offset the emissions of these trucks with cars or smaller
trucks that it certifies to levels below 0.07 g/mi.
While we believe our proposed averaging program is appropriate, we
recognize that most manufacturers do not produce larger LDTs and may be
able to meet the corporate average NO_X standard of 0.07 g/mi
with less overall effort. Therefore, we request comment as to whether
another approach to averaging might be more appropriate such as a
segregated approach where LDTs are averaged separately from LDVs or
where HLDTs (LDT3s and 4s) are averaged separately from LDV/LLDTs.

[[Page 26089]]

iv. Different Standards for Different Categories of Vehicles.
We have explained several times in this preamble that we believe
the same standards should apply to all LDVs and LDTs because LDTs are
so often used as passenger vehicles, and because the standards are
feasible for all LDVs and LDTs. The technological challenge may be
greater for larger trucks, so our proposal provides additional leadtime
and a later start date for HLDTs to provide more opportunity to resolve
potential problems. However, we recognize that other approaches exist
that could yield comparable environmental benefit. Therefore, we
request comment on other approaches such as one that would employ a
lower corporate average NO_X standard for LDV/LLDTs, with a
higher corporate average standard for HLDTs.
v. Consideration of Special Provisions for the Largest LDTs and
Advanced Technology.
California has adopted a provision in its LEV II program, under
which a manufacturer could certify up to 4 percent of its larger LDTs
to a higher NO_X standard. These trucks could meet a 0.10 g/
mi NO_X standard rather than a 0.07 g/mi NO_X
standard, provided they have a payload of at least 2500 pounds.
California chose the figure of 4% because it approximates the fraction
of such trucks in the largest volume manufacturer's fleet.
We have not proposed such an option in the federal program because
we are providing additional lead time and compliance on average for all
cars and trucks beginning in 2009. Nevertheless, we do recognize that
the largest trucks will likely require the greatest application of
emission control technology to comply with Tier 2 standards and we
expect that larger trucks will likely be the last, and the most
difficult, vehicles to phase into the Tier 2 program.
In the context of the flexibilities already proposed for the
federal program, we request comment on the need for and environmental
impact of additional program flexibility for the largest trucks. One
option we have considered would allow manufacturers to exclude a small
fraction (perhaps 4 percent) of their largest Tier 2 trucks (HLDTs)
from the corporate average NO_X calculation beginning in 2009
and lasting through approximately model year 2011. These trucks would
still be subject to a NO_X standard of 0.20 g/mi and all
other standards and provisions of the Tier 2 program, including the
requirement to fit within a Tier 2 bin for other emission standards.
This provision would provide a less stringent standard for the
heaviest LDTs. We believe these LDTs are the most likely to be used
primarily for work and commercial purposes, while at the same time
having the most difficulty complying with Tier 2 requirements. We
request comment on all aspects of this provision, including whether the
allowable sales fraction (4%) and payload minimum (2500 pounds) set by
California would be appropriate for the federal provision, and whether
such a concept should also be applied to only LDT4s or both LDT3s and
4s. Supporters of such an approach should comment on the appropriate
allowable sales fraction for the interim vehicles.
Some have suggested that a potential way of providing flexibility
for advanced technology vehicles would be to provide bins with less
stringent standards while retaining the stringency of the 0.07
NO_X average. These additional bins would augment the current
flexibilities offered to manufacturers. We request comment on this
idea, specifically on including additional bins with NO_X
standards up to 0.60 g/mi, with any other modifications that are
appropriate. We also ask comment on whether such bins should be a
temporary part of the Tier 2 program.
vi. Measures to Prevent LDT Migration to Heavy-Duty Vehicle
Category.
Existing regulations define a light-duty truck to be any motor
vehicle rated at 8500 pounds gross vehicle weight rating (GVWR) or less
that has a curb weight of 6000 pounds or less and that has a basic
frontal area of 45 square feet or less, which is:
<bullet> Designed primarily for purposes of transportation of
property or is a derivation of such a vehicle, or
<bullet> Designed primarily for transportation of persons and has a
capacity of more than 12 persons, or
<bullet> Available with special features enabling off-street or
off-highway operation and use.
For the heaviest LDTs, we are concerned that manufacturers may, in
some cases, find it attractive to add GVWR capacity, curb weight or
frontal area to their vehicles such that they would no longer meet one
or more of the criteria to be considered an LDT. The vehicles would
then fall into the heavy-duty category and would be subject to less
technologically challenging standards.
We would like to develop reasonable restrictions to prevent this
``gaming'' of the LDT definition. The ideal restrictions would prevent
migration of LDTs above the limiting criteria, but would not impact
vehicles with legitimate needs to be outside, but close to, the LDT
definition. Our objective is complicated by the fact that many LDTs
currently have derivatives or corresponding models that are over 8500
pounds GVWR.
We have considered various approaches to restrictions on LDTs. Some
of the ideas we have considered are as follows:
<bullet> Require all complete trucks in the 8500-10,000 pound GVWR
range to meet light-duty standards.
<bullet> Raise the GVWR cutoff from 8500 pounds to some other
number such as 8750, 9000 or 9500 pounds.
<bullet> Require manufacturers of vehicles that are above but close
to any of the three size criteria to provide justification that they
cannot accomplish their intended function if built to a lower size
criterion.
<bullet> Require manufacturers to provide supporting data, surveys,
etc., that vehicles above, but close to, any of the LDT cutoffs are
primarily used for commercial purposes.
We request comment on all aspects of this vehicle migration issue,
including specific comment on the ideas presented above and on other
approaches that might be appropriate. This discussion serves as notice
that we are very likely to finalize a provision to address this vehicle
migration issue. You are encouraged to consider the approaches we have
outlined above and provide specific suggestions on other approaches as
well as comments as to the need for such controls, their feasibility
and their cost.
In the longer term, the best way to address the vehicle migration
issue is to implement standards for complete heavy-duty vehicles that
have a stringency comparable to their HLDT counterparts. In the near
future, we expect to publish an NPRM addressing emissions from
gasoline-fueled heavy-duty engines and vehicles for 2004 and later
model years. As part of that effort we are considering chassis-based
standards for gasoline-fueled complete vehicles between 8,500 and
14,000 lbs GVWR. The degree to which such standards discourage
migration depends upon the relative stringency of the standards. EPA
requests comment on the potential effectiveness of such a strategy in
addressing migration concerns and the timing and level of emission
standards necessary to do so.
vii. Use of Non-conformance Penalties (NCPs).
NCPs are monetary payments that manufacturers can pay to meet an
adjusted standard in lieu of complying with a prescribed emission
standard or set of emission standards. See CAA

[[Page 26090]]

section 206(g). Current regulations at 40 CFR part 86 Subpart L provide
for NCPs for HLDTs, and for heavy-duty engines. However, in order to
establish NCPs for a specific standard or set of standards for these
vehicles and engines, EPA must first determine that (1) substantial
work will be required to meet the standard for which the NCP is
offered; and (2) that there will be a manufacturer that is a
technological laggard in complying with that standard. EPA must also,
through rulemaking, determine compliance costs so that the penalty
rates can be established appropriately.
NCPs were used extensively by manufacturers of on-highway heavy-
duty engines in the late 1980s, prior to the implementation of our
heavy-duty averaging, banking and trading program. Since that time,
their use has been rare. We believe manufacturers have used the
flexibility of an averaging, banking and trading scheme as a preferred
alternative to incurring the monetary losses associated with NCPs.
We are not proposing NCPs for HLDTs in the primary Tier 2 program
or in the interim programs. This is because we believe that the
NO_X averaging program we are proposing makes it unlikely
that the criteria for NCPs mentioned above will be met, as
NO_X credits from other vehicles may be used to enable HLDTs
to meet the 0.07 g/mi average NO_X standard.
We have considered whether NCPs might be appropriate for the Tier 2
diesel particulate standards, for which our proposal contains no
averaging provisions. We are not proposing PM NCPs for those diesel
powered trucks, but we request comment on whether such NCPs would be
appropriate. We believe that appropriate technologies will be available
from component vendors and diesel engine suppliers. We request comment
on the need for and appropriateness of NCPs for any Tier 2 standard for
HLDTs.
viii. Additional NO_X Credits for Vehicles Certifying to
Low NO_X Levels.
There is currently substantial work underway to develop vehicles
with extremely low emissions. We believe that it is appropriate to
encourage such technology by providing incentives for its use.
Consequently, we are requesting comment as to whether we should
implement a provision by which manufacturers can earn additional
NO_X credits for certifying to levels below 0.07 g/mi. As we
envision such a provision, manufacturers would be allowed, in the
calculation of their year end corporate average NO_X level,
to multiply the number of vehicles sold which are certified to bins
below 0.07 g/mi NO_X by some preset multiplier, or set of
multipliers. For example, the number of vehicles certified to the 0.04
bin might be multiplied by 1.5, those in the 0.02 bin might be
multiplied by 2.0 and those in the 0.0 bin (ZEVs) might be multiplied
by 3.0.
We recognize that such a program would enable manufacturers to use
more credits than actually generated in use, and that the use of these
credits would likely result in some additional NO_X
emissions. However, we believe that it may be appropriate to provide
inducements to manufacturers to strive for ever lower NO_X
emissions and that these inducements may help pave the way for greater
and/or more cost effective emission reductions from future vehicles. We
request comment on all aspects of such incentive credits. Issues
related to these credits include the value of a multiplier or
multipliers, whether early credits should be subject to the
multipliers, and whether there should be a ``sunset'' provision to
limit the time period in which manufacturers could obtain and/or use
these extra credits. We request comment on a sunset year of 2009, since
it is the end of the proposed Tier 2 program phase-in.
ix. Incentives for Manufacturers to Bank Additional Early NO_X
credits.
We are interested in exploring any reasonable approaches that would
provide incentives to manufacturers to produce vehicles meeting the
0.07 g/mi NO_X standard earlier than required. We believe
that early certification to this level will help manufacturers gain
experience with new or enhanced technologies on a limited scale before
they must be applied to the entire fleet, and that such experience
would have a positive, although hard to quantify, environmental
benefit.
We have proposed an approach elsewhere in this preamble that
permits manufacturers to utilize alternative phase-in schedules.
Manufacturers that introduce Tier 2 vehicles before the first required
year in the primary phase-in schedule could follow a more flexible
phase-in path to 100% compliance than required under the primary
option. Manufacturers would also be able to generate NO_X
credits if these ``early'' vehicles met a corporate average
NO_X level of less than 0.07 g/mi.
We have considered whether a mechanism that provided additional
NO_X credits could induce manufacturers to introduce more
Tier 2 vehicles sooner than required. Such a mechanism might substitute
a number higher than the 0.07 g/mi NO_X standard in the
credit calculation so that the manufacturer would subtract its
corporate average NO_X level from, say, 0.10 and then
multiply the difference by the number of Tier 2 vehicles to determine
credits earned. While we believe such a scheme might induce
manufacturers to accelerate the introduction of Tier 2 vehicles, we
have concerns about whether this approach would lead to windfall
credits and whether we would need to employ a discount to compensate
for them. Should the resulting credits have finite or infinite life?
Should we apply such a scheme to LDV/LLDTs only; or should we also
apply it to HLDTs; and should we apply such a scheme to the interim
standards for HLDTs? We request comment on these and all other aspects
of permitting additional NO_X credits for Tier 2 and interim
vehicles.
x. Flexibilities for Small Volume Manufacturers and Small
Businesses.
In section V.A.8. above, we propose to waive the Tier 2 phase-in
requirements for small volume manufacturers.⁸⁵ These
manufacturers, which each produce 15,000 or fewer vehicles per year,
would simply comply with the 100 % requirement in 2007 (2009 for
HLDTs).
---------------------------------------------------------------------------

\85\ A ``small volume manufacturer'' is not necessarily a
``small business''. Rather, ``small volume manufacturer'' is an EPA
term that refers to entities whose annual on-highway sales are
15,000 or fewer vehicles per year. However, most if not all small
businesses covered under this discussion are also ``small volume
manufacturers,'' though most small volume manufacturers are not
small businesses.
---------------------------------------------------------------------------

Some very small volume manufacturers of LDVs and LDT1s and LDT2s
elected not to opt into NLEV and thus will produce Tier 1 vehicles
during the NLEV program. We are seeking comment about the burden that
our interim standards might impose on very small manufacturers in 2004
given that they will have to meet the Tier 2 standards no later than
2007 under today's proposal. Similarly we are concerned about the
burden that the interim standards might impose on any small volume HLDT
manufacturers. We request comment on the need for and appropriateness
of a provision that would waive the interim standards for very small
volume manufacturers who produce, say, less than 1,000 vehicles per
year, or who qualify as small businesses (see below).
The panel convened under the Small Business Regulatory Enforcement
Fairness Act (SBREFA),⁸⁶ recommended that we seek comment on
five provisions outlined below to ease our

[[Page 26091]]

proposal's impact on small businesses. These provisions, if adopted,
would apply to ``small businesses'' as defined by Small Business
Administration. The size of a ``small business'' varies by industry
type as represented by SIC codes. Tables V.B.-2 and V.B.-3 contain the
SIC codes that could potentially be impacted by the Tier 2 rule and the
maximum number of employees or maximum revenue a business can have to
be considered a small business.
---------------------------------------------------------------------------

\86\ This panel was convened, consistent with SBREFA, by EPA,
the Small Business Administration, and the Office of Management and
Budget to review of the likely impact of Tier 2 requirements on
small businesses.

Table V.B.-2.--SBA Small Business Categories for Small Independent
Commercial Importers
------------------------------------------------------------------------
Size standard
(annual
SIC code Description revenues in
millions)
------------------------------------------------------------------------
7533........................... Auto Exhaust System $5
Repair Shops.
7549........................... Automotive Services.... 5
8742........................... Management Consulting 5
Services.
------------------------------------------------------------------------

Table V.B.-3.--SBA Small Business Categories for Alternative Fuel
Vehicle Converters
------------------------------------------------------------------------
Size standard ($
SIC code Description =annual revenues)
------------------------------------------------------------------------
3592........................ Carburetors, 500 employees.
Pistons, Rings and
Valves.
3714........................ Motor Vehicle Parts 750 employees.
and Accessories.
5172........................ Petroleum Products.. 100 employees.
5984........................ Liquefied Petroleum $5 million.
Gas Dealers.
7549........................ Automotive Services. $5 million.
8742........................ Management $5 million.
Consulting Services.
8931........................ Commercial Physical 500 employees.
Research.
------------------------------------------------------------------------

The vast majority of businesses in these categories are not subject
to these EPA requirements. However, some businesses in these categories
may in fact manufacture LDVs and LDTs or may modify vehicles produced
by others in a manner that will subject them to the requirements
applicable to manufacturers under EPA regulations. For example,
Independent Commercial Importers (ICIs) modify imported motor vehicles
into configurations that they certify to meet federal emission
requirements. Approximately 15-20 small businesses qualified as
manufacturers and received certificates of conformity each year over
the last five years.
For simplicity, and consistency with the report of the SBREFA
panel, we refer to these small businesses as small certifiers in the
following discussion. The requirements to certify continue to apply
only to parties that meet the definition of ``manufacturer.''
Consistent with the recommendations of the SBREFA panel, we request
comment on the following ideas:

For small certifiers that convert imported vehicles to U.S.
standards (independent commercial importers or ICIs) and for small
certifiers that convert vehicles to operate on alternative fuels,
provide a delay in required compliance of two years after the
particular model vehicle is certified to Tier 2 standards by the
original equipment manufacturer.

This provision would provide time for development of appropriate
emission control systems and test data for small businesses who may
need to first obtain a regular production vehicle certified by the OEM
before they can begin work.
Although it was not a specific recommendation of the SBREFA panel,
we are also requesting comment on whether ICIs should be exempted from
the Tier 2 and interim fleet average NO_X standards. ICIs may
not be able to predict their sales of vehicles and control their fleet
average emissions because they may be dependant upon vehicles brought
to them by individuals attempting to import uncertified vehicles.
Presently, the NLEV requirements are optional for ICIs and ICIs are
specifically exempted from complying with the fleet average NMOG
standard under the NLEV program. (See 40 CFR 85.1515(c)). Further, a
prohibition in the current ICI regulations specifically bars ICIs from
participating in any emission related averaging, banking or trading
program. (See 40 CFR 85.1515(d)). If we do not amend this prohibition,
the likely outcome would be that ICIs could choose any bin to certify
their vehicles and would pick the least stringent standards.
Given the historically very low sales of ICIs and the probable
challenges that even the least stringent Tier 2 and interim non-Tier 2
bins will impose upon ICIs, we do not expect ICIs to grow significantly
in number or size. Therefore, we do not expect that provisions
exempting or prohibiting ICIs from the fleet average NO_X
standard would have any air quality impact. However, we request comment
on all aspects of the applicability of the fleet average NO_X
standards to ICIs.
Establish a credit program and provide incentives for large
manufacturers so that they would make credits available to small
certifiers.
This provision would address the problem inherent with any emission
credit trading program that manufacturers holding credits don't have to
trade them. While the panel proposed this option, it did not provide
any thoughts on what type of incentives might be appropriate and
necessary to induce larger manufacturers to supply credits at
reasonable prices to small businesses.
Develop a program to provide credits to small certifiers for taking
older vehicles off of the road (i.e., a scrappage program).
Because older vehicles often have very high emissions, removing one
from use could more than offset the emissions of a new vehicle produced
by a small certifier that was unable to fully comply with the Tier 2
standards. Scrappage programs must be designed so that they remove
vehicles from the fleet that see significant annual mileage. They must
be adequately funded and managed. They must have controls and oversight
to ensure that they don't remove vehicles that would have been scrapped
anyway.
Design a case-by-case hardship relief provision that would delay
required

[[Page 26092]]

compliance for small certifiers that demonstrate that they would face a
severe economic impact from meeting the Tier 2 standards.
We have implemented case-by-case hardship provisions in some rules
subject to specific limiting constraints. Typically, these would
provide that small businesses that have tried all other regulatory
options and apply in writing before they experience nonconformity,
could obtain a 1 year delay in the implementation of the standards. The
small business would have to show that failure to comply was the fault
of external and extenuating circumstances and that inability to sell
the subject vehicles would have a major impact on the company's
solvency.
If the Tier 2 program involves a phase-in of standards, allow small
certifiers to comply at the end of such a phase-in.
As indicated at the beginning of this section, we are proposing
this option for all phase-ins associated with the Tier 2 program
including the phase-in of the Interim standards for HLDTs (see Section
V.A.8. above).
We request comment on the need for, appropriateness and
environmental impact of all of the items proposed by the SBREFA panel.
Also, we request comment on whether any such provisions would be
necessary and appropriate for the interim standards for non-Tier 2
vehicles.
xi. Adverse Effects of System Leaks.
For the emission control system to operate as designed, the air-
fuel (A/F) ratio must stay within strictly prescribed limits that vary
with vehicle/engine operating conditions and engine controls must
respond quickly to the slightest changes in this ratio. Even the
smallest air leak in either the exhaust manifold or exhaust pipe or any
related connection can provide the oxygen sensor incorrect information
on the oxygen content of the exhaust gas it uses to calibrate the
engine A/F ratio.
Some manufacturers have taken steps to address this concern as part
of their overall design process by incorporating features such as
corrosion-free flexible couplings, corrosion-free steel, and improved
welding of catalyst assemblies. EPA is concerned that either as a
result of manufacturing or installation errors or errors in a repair
action, there will be an unintentional and unobserved increase in
emissions and perhaps a failure to meet FTP and a SFTP emission
standards in-use.
EPA seeks comment on design or onboard monitoring requirements that
might be useful to address this concern. EPA would also seek comment on
a provision that would require a manufacturer to demonstrate through
engineering analysis or design that such possibilities have been taken
into account.
xii. Consideration of Other Corporate Averaging Approaches.
We welcome comments on the pros and cons, including regulatory
burden, of establishing a combined NMOG plus NO_X corporate
average standard in lieu of either the proposed NO_X average
or a California-like NMOG average. We also request comments, if not
provided in response to Section IV.B. above, on the concept of
requiring a declining corporate average NO_X standard or a
declining corporate average NMOG standard at the federal level. For
example, we would consider a declining average approach that reduces
NMOG/NO_X corporate average emissions by 20-25% over the
period 2008-2012, or nominally to 0.07 NMOG/0.05 NO_X. Such a
reduction might involve a reduction in gasoline sulfur levels as
discussed in Section IV.E.2. above. We also seek comment on the idea of
eliminating the averaging concept in 2011 or 2012 and setting the LDV/
LDT standards at the levels of Bin No. 5 in Table IV.B.-2 (0.07 g/mi
NO_X plus the other standards). Commenters should address the
cost and feasibility of these approaches.
2. Tighter Evaporative Emission Standards
We considered proposing tighter evaporative emission standards,
including California's LEV II standards for evaporative emissions,
shown in Table V.B.-4 below.

Table V.B.-4.--California's LEV II Evaporative Hydrocarbon Standards
[Grams per test]
------------------------------------------------------------------------
Supplemental
Three day two day
Vehicle class diurnal + diurnal +
hot soak hot soak
standard standard
------------------------------------------------------------------------
LDV............................................ 0.50 0.65
LDT1 AND LDT2.................................. 0.65 0.85
LDT3 AND LDT4.................................. 0.90 1.15
------------------------------------------------------------------------

These standards are based on an evaporative emission test procedure
that is conducted at different temperatures using fuel with lower vapor
pressure than the corresponding federal evaporative test procedure.
Under current evaporative standards, California accepts the results of
federal evaporative testing, because it represents a worst case test.
We do not know whether California's standards are feasible under the
federal test conditions.
We are concerned about evaporative hydrocarbons and we recognize
that they constitute a portion of the mobile source VOC inventory that
will be similar in size to the light duty exhaust contribution when
NLEV exhaust standards are in place. Our proposed standards, which are
found in section IV.B.4.a. above, are roughly in line with current
average certification levels but will nonetheless yield real in-use
evaporative reductions as manufacturers reduce certification levels to
gain safety margins under the new standards. These standards will also
prevent manufacturers from ``backsliding'' from their current low
certification levels upward toward the existing standards as they seek
cost reductions. Our proposed standards will require manufacturers to
capture the abilities of available fuel system materials to minimize
evaporative emissions. Further, we are proposing certification
enhancements to address the impact of alcohol fuels on evaporative
emissions, and we expect that these measures will lead to more uniform
use of lower permeability materials that will result in in-use
reductions in non-attainment areas where alcohol fuels are the most
prevalent.
We request comment on the appropriateness and cost effectiveness of
applying tighter evaporative standards in the federal program.
3. Credits for Innovative VOC, NO_X and Ozone Reduction
Technologies Not Appropriately Credited by EPA's Emission Test
Procedures
Compliance with the current and proposed EPA motor vehicle emission
standards is based on the emission performance of a vehicle over EPA's
prescribed test procedure. While this test procedure addresses many of
the aspects of a vehicle's impact on air quality, it does not address
all such impacts. Two developing technologies have been brought to
EPA's attention that have shown significant potential to improve ozone-
related air quality, but that would not do so over the current EPA test
procedure.
The first example is a device that removes ozone from the air as
the vehicle is driven. A major producer of automotive catalysts,
Englehard, has approached both California and EPA with a proposal for a
technology (called Premair) in which vehicle radiators would be coated
with a catalyst that converts ambient ozone to oxygen. In its CalLEVII
program, California has adopted some basic ground rules concerning the
types of information that

[[Page 26093]]

would have to be submitted in order to certify such ozone reduction
technologies and determine the amount of allowable NMOG
credits.⁸⁷ This determination would be made on a case-by-
case basis. The manufacturer would have to provide an evaluation of the
system's performance and durability, as well as a description of the
on-board diagnostic strategy to monitor the performance of the device
in use. The NMOG credit would be based upon the running of an approved
airshed model, which would determine the amount of NMOG emission
reductions that would produce the same change in one-hour peak ozone as
the use of the ozone reduction device being evaluated.
---------------------------------------------------------------------------

\87\ See page II-28 of the following California document for a
full discussion: Proposed Amendments to California Exhaust and
Evaporative Emission Standards and Test Procedures for passenger
Cars, Light-Duty Trucks and Medium Duty Vehicles (``LEV II'') and
Proposed Amendments to California Motor Vehicle Certification,
Assembly-Line and In-Use Test Requirements (``CAP2000''). Released
September 18, 1998 for the Air Resources Board Hearing of November
5, 1998.
---------------------------------------------------------------------------

Englehard has asked EPA to develop a similar procedure to that
adopted by ARB and to consider granting their technology a
NO_X credit, as well as an NMOG credit. The manufacturer of
the vehicle employing Premair would then have the option of which
credit to use.
There are a number of issues that would have to be resolved before
such credits could be granted, including:
<bullet> The methods to be used to certify in-use performance over
the useful life of the vehicle,
<bullet> The requirement for, and the design and certification of,
an onboard diagnostic system to monitor in-use performance, and
<bullet> Which airshed model to use, including what cities and
episodes to use in modeling the 8-hour peak ozone reduction, and
<bullet> The methods for determining either the NMOG or
NO_X credit, or both.
EPA has placed information provided to date by Englehard in the
docket to this rule, and requests comments on the appropriateness of
such credits, and on the procedures that should be used to determine
those credits, should we proceed.
The second example is an insulated catalyst. The insulation retains
heat for extended periods of time, increasing the catalyst temperature
when the engine is started and reducing the time required for the
catalyst to reach an operational temperature. This technology can
reduce cold start emissions for engine off times (called soaks) of 24
hours or less. The vast majority of engine soaks in-use are less than
24 hours. However, EPA's test procedure only tests emissions at two
fairly extreme soak times: 10 minutes and 12-36 hours. The 10 minute
soak is so short that even an uninsulated catalyst is warm enough to
quickly begin working upon restart. The 36 hour soak is beyond the
practical limit of cost-effective insulating techniques.
In 1994, as part of its proposed SFTP standards, EPA proposed
adding an intermediate soak of 1 hour to the test procedure, due both
to the large number of in-use soaks falling between the current 10
minute and 12-36 hour soaks and to the desire to encourage catalyst
technology that reduced cold start emissions for such intermediate
soaks. EPA did not promulgate this aspect of its SFTP standards, due in
part to concerns about the cost effectiveness of mandating such
controls. However, the efficacy of such technology was not questioned.
Thus, there appears to be little reason to prohibit a manufacturer from
using such technology to reduce in-use emissions in lieu of other
technology needed to meet the proposed Tier 2 standards.
As mentioned above concerning Premair, a methodology would need to
be developed to estimate the impact of an insulated catalyst, or other
any other similar technology, on in-use emissions so that equivalent
NMOG and NO_X emission credits could be determined. Also,
procedures for certifying in-use performance and durability and onboard
diagnostics would also have to be addressed. EPA requests comments on
the appropriateness of allowing emission credits for insulated
catalysts and other technologies not appropriately assessed under
current test procedures. EPA also requests comments on the procedures
to be used to develop such credits.
EPA also requests comments on whether the credits granted for
either ozone or emission reduction technologies should be restricted to
the proposed Tier 2 standards, or whether they should also be granted
under the current NLEV standards and the proposed interim standards for
non-Tier 2 vehicles, as well.
4. Need for Intermediate Useful Life Tier 2 Standards
For our Tier 2 and interim standards we have generally proposed
both full useful life and intermediate useful life FTP exhaust emission
standards. (See Tables IV.B.-2, -3, -6,-7,-10 and -11.) We have also
proposed full and intermediate life SFTP standards. (See Tables V.A.-3
and -4.) Intermediate useful life standards are more stringent than
full useful life standards and reflect our experience that better
emission performance can be expected at lower mileages.
We are not proposing intermediate useful life standards for the
three lowest Tier 2 FTP bins, and we are not proposing intermediate
standards for the lowest FTP bin (the Zero Emission Vehicle or ZEV bin)
in any case. This is because the full life standards in those bins are
already so low as to allow little deterioration between a new vehicle
and a vehicle at full useful life.
We request comment on the appropriateness of and need for
intermediate useful life and what the environmental consequences might
be from deleting intermediate useful life standards for all Tier 2
vehicles and from the interim standards bins that match those of the
Tier 2 program.

VI. Additional Proposed Elements and Areas for Comment: Gasoline
Program

Section VI.A. presents two additional issues that have some impact
on our proposed program: whetherstates are preempted from requiring
gasoline sulfur reductions as a result of today's action, and whether
other gasoline properties may also need to be controlled in the future.
We encourage your comment on all of these issues. Section VI.B.
provides additional detailed information about our proposed
requirements for establishing compliance with the gasoline sulfur
standards, as well as how we will enforce these standards. The major
details of our proposed gasoline sulfur control program were explained
in Section IV.C.; the information presented here is supplementary.

A. Other Areas for Comment

The following sections raise additional issues that are relevant to
our decisions regarding gasoline sulfur control and the design of our
gasoline sulfur program. We encourage you to comment on these issues if
they are of interest to you.
1. Would States Be Preempted From Adopting Their Own Sulfur Control
Programs?
When we adopt federal fuel standards, states are preempted from
adopting similar state-level controls. Section 211(c)(4)(A) of the CAAA
prohibits states from prescribing or attempting to enforce controls or
prohibitions respecting any fuel characteristic or component if EPA has
prescribed a control or prohibition applicable to such fuel
characteristic or component under section 211(c)(1). This preemption
applies to all states except California, as explained in section

[[Page 26094]]

211(c)(4)(B). For these states other than California, the Act provides
two mechanisms for avoiding preemption. First, section 211(c)(4)(A)(ii)
creates an exception to preemption for state prohibitions or controls
that are identical to the prohibition or control adopted by EPA.
Second, states may seek EPA approval of SIP revisions containing fuel
control measures, as described in section 211(c)(4)(C). EPA may approve
such SIP revisions, and thereby ``waive'' preemption, only if it finds
the state control or prohibition ``is necessary to achieve the national
primary or secondary ambient air quality standard which the plan
implements.''
We are proposing to adopt the sulfur standards pursuant to our
authority under section 211(c)(1). Thus, we believe final promulgation
of the sulfur standards would result in the clear preemption of future
state actions to adopt fuel sulfur controls.⁸⁸ States would
therefore need to obtain a waiver from us under the provisions
described in section 211(c)(4)(C) for all state fuel sulfur control
measures adopted following promulgation, unless the state standard were
identical to our final sulfur standard. We welcome your comments on our
interpretation of the source and effect of federal preemption.
---------------------------------------------------------------------------

\88\ Even in the absence of final promulgation of federal sulfur
standards, existing federal fuel controls for RFG and conventional
gasoline have raised issues of preemption of state fuel sulfur
measures. In any case, it is clear that state sulfur standards would
be preempted as of the date of promulgation of the proposed federal
sulfur standard.
---------------------------------------------------------------------------

Section 211(c)(4)(A) preempts state fuel controls if EPA has
``prescribed'' federal controls. We read this language to preempt non-
identical state standards on the effective date of the standards, as
opposed to the date the standards become enforceable. Thus, if the
proposed standards are finalized according to our expected schedule,
this rulemaking would preempt state actions upon promulgation at the
end of 1999, even though the standards would not require sulfur
reductions until 2004. This interpretation is consistent with EPA
actions applying other federal fuel measures. See 54 FR 19173 (May 4,
1989) (noting preemption of Massachusetts state RVP measure before
start of first control period for federal RVP). We also believe this
interpretation is consistent with the intent behind section
211(c)(4)(A). Though the standards are not immediately enforceable,
they will have an immediate impact on refiners' investment decisions.
We believe, by adopting 211(c)(4)(A), Congress intended to provide
security for these investment decisions by preventing unnecessary
conflict between state and federal fuel controls.
2. Potential Changes in Gasoline Distillation Properties
During the last several years, representatives of the automotive
industry have presented information to us suggesting that control of
certain gasoline distillation properties can provide reductions in both
exhaust hydrocarbon emissions as well as the frequency of performance
problems such as hesitation, cold startability, and impeded
acceleration. Automotive industry representatives contend that the
source of most performance problems--slower atomization and
vaporization due to fuels with higher boiling points--also leads to
less efficient combustion, and thus higher levels of hydrocarbons in
the exhaust.
With regard to Tier 2 vehicles, some automakers have claimed that
in-use fuels with high boiling points would impact their ability to
control the mixture of air and fuel entering the engine, and thus could
result in in-use emissions that are higher than expected based on
certification levels. Thus, automakers argue, controls on the
distillation properties of gasoline would not only produce emission
benefits for the in-use fleet, but would also ensure the viability and
benefits of Tier 2 vehicles.
On January 27, 1999, we received a petition ⁸⁹ from a
group of automakers in which they provided a more detailed analysis of
the costs and benefits of controlling gasoline distillation properties.
In this petition, they specifically requested that the Distillation
Index (DI) be capped at 1200 for all summer-grade gasolines nationwide.
They have defined the distillation index by the equation 1.5xT10 +
3xT50 + T90 +20xOxy, where T10 represents the temperature at which 10%
of the fuel has evaporated in a standard distillation test, and
likewise for T50 and T90, and Oxy is the oxygen content contributed by
ethanol. This petition includes a study conducted by MathPro
Inc.⁹⁰ to estimate the feasibility and cost to the refining
industry of capping all summer grade gasoline at a DI level of 1200.
MathPro concluded that the cost of such control would be approximately
0.4 cents/gal on average for all summer grade gasoline.
---------------------------------------------------------------------------

\89\ ``Petition to regulate gasoline distillation properties''.
Submitted by DaimlerChrysler Corporation, Ford Motor Company,
General Motors Corporation, and the Association of International
Automobile Manufacturers. Submitted to EPA Administrator Carol
Browner on January 27, 1999. EPA Air Docket A-97-10, Document No.
II-G-286.
\90\ ``Technical and economic implications of controlling the
distillation index of gasoline.'' MathPro Inc., October 21, 1998.
EPA docket A-97-10, document II-G-268.
---------------------------------------------------------------------------

We believe that the analyses presented by this petition have merit.
However, we do not believe that they are sufficient to justify capping
DI at 1200 at this time, since there are a number of issues that it
does not address. Before we could formally propose a DI cap, we would
need to have a justification for the cap based on air quality need,
peer-reviewed estimates of the cost to the refining industry and to
consumers, and comparisons of the cost effectiveness of this strategy
to that for other potential hydrocarbon control strategies. Therefore,
we are not today proposing controls on gasoline distillation
properties. However, we request comment on the automakers' DI petition
and the included MathPro report in terms of their sufficiency in
demonstrating that a DI cap of 1200 is appropriate.

B. Gasoline Sulfur Program Compliance and Enforcement Provisions

1. Overview
We are proposing enforcement mechanisms that track those of the
reformulated gasoline/conventional gasoline (RFG/CG) rule, because of
significant similarities between the two programs, including refinery
average standards, refinery level and downstream level caps, and the
generation and use of credits. These features raise similar compliance
issues for both programs. Because of the importance of assuring that
all gasoline meets the sulfur standards, measures are needed to assure
the accuracy of refiner and importer testing, and to assure that the
quality of gasoline is not adversely affected downstream of the
refinery. Downstream enforcement would be based primarily on EPA
sampling and testing, and examination of product transfer documents
(PTDs) and other evidence.
More specifically, we are proposing:
<bullet> That refiners and importers test each batch of RFG and CG
produced or imported for sulfur content and maintain testing records
and retain test samples.
<bullet> That refiners and importers of gasoline submit reports
regarding compliance with averaging and credits provisions.
<bullet> That the current attest procedures of the RFG/CG rule
⁹¹ be applied to sulfur rule compliance.
---------------------------------------------------------------------------

\91\ 40 CFR part 80 subpart F.

---------------------------------------------------------------------------

[[Page 26095]]

<bullet> Enforcement provisions regarding the credit program, to
prevent the use, sale or purchase of invalid credits, and to require
adjustments to compliance calculations based on use of invalid credits.
<bullet> Requirements to ensure compliance by small foreign
refiners subject to individual refinery sulfur standards and to ensure
the separation of such foreign gasoline from all other gasoline to the
U.S. port of entry.
<bullet> Downstream maximum sulfur caps, which would apply to all
persons in the chain of distribution of gasoline, including
distributors, resellers, carriers, retailers and wholesale purchaser-
consumers of gasoline.
<bullet> Voluntary downstream quality assurance testing by
distributors and refiners to help assure compliance.
The sulfur standards proposed today would apply, as in other fuels
programs, to all motor vehicle fuel that meets the definition of
gasoline. See 40 CFR 80.2. This definition typically includes all the
gasoline that is produced and distributed through the gasoline
distribution system, including gasoline, such as marina gas, that is
ultimately used in nonroad equipment. Such fuel meets the definition of
gasoline and is subject to the standards proposed today. For example,
where gasoline makes up only a small portion of what a refinery
produces, and is perhaps a byproduct of other processing, the refiner
could not avoid the sulfur standard by designating the product as
marina gasoline or nonroad gasoline. EPA would apply the sulfur
standard to the same broad group of products that meets the definition
of gasoline for its other gasoline fuel programs.
We are aware that there are certain fuels, such as aviation fuel
and racing fuel, that are generally segregated from gasoline throughout
the distribution system. Where such fuels are segregated from motor
vehicle gasoline and not made available for use in motor vehicles, the
fuel would not be subject to sulfur rule standards.⁹² We
propose that such fuel become subject to the sulfur standards and other
regulatory requirements and prohibitions if its segregation from
gasoline at any point in the distribution system is compromised.
Offering such fuel for motor vehicle use or dispensing such fuel for
motor vehicle use would be prohibited. We are also proposing specific
PTD requirements and labeling requirements to prevent introduction of
high sulfur fuels into motor vehicles. EPA invites comment on whether
such fuel should also be subject to refinery level sulfur standards, or
whether it should be subject to the standards from the point at which
it is made available for use in motor vehicles.
---------------------------------------------------------------------------

\92\ If a fuel is not segregated throughout the gasoline
distribution system, but is fungibly mixed with gasoline, then it
becomes a gasoline that is subject to the standard.
---------------------------------------------------------------------------

The proposal would clarify the definition of refinery at 40 CFR
80.2(h). Specifically, we are proposing to clarify that ``refinery''
means any facility, including a plant, tanker truck or vessel where
gasoline or diesel fuel is produced, including any facility at which
blendstocks are combined to produce gasoline or diesel fuel, or at
which blendstock is added to gasoline or diesel fuel.⁹³
---------------------------------------------------------------------------

\93\ This is consistent with all current EPA fuels rules,
interpretations, policies and question and answer documents, and is
only a clarification.
---------------------------------------------------------------------------

We propose that any oxygenate blender that only adds oxygenate to
gasoline or to ``reformulated blendstocks for oxygenate blending''
(RBOB), be exempt from sulfur standards and would not be required to
conduct any new testing, or perform any new recordkeeping or reporting,
because we believe the sulfur level of EPA-allowed oxygenates added
downstream from the refinery is very low. We believe it is an
appropriate assumption, barring special circumstances, that the sulfur
content of the gasoline will be diluted in proportion to the addition
of the oxygenate.
In the remainder of this section we address enforcement issues
regarding today's proposed rule that are not discussed in section
IV.C.3., above.
2. What Requirements is EPA Proposing for Foreign Refiners and
Importers?
As discussed in section IV.C, under today's proposal, standards for
gasoline produced by foreign refineries that are not subject to small
refiner individual refinery standards would be met by the importer.
Standards for gasoline produced by a foreign refinery subject to an
individual sulfur rule standard would be met by the foreign refinery,
with certain limited exceptions. The provisions would be very similar
to the foreign refinery provisions of the RFG/CG rule, under 40 CFR
80.94.

a. What Are the Proposed Requirements for Small Foreign Refiners with
Individual Refinery Sulfur Standards?

Under the RFG/CG rule, EPA has promulgated regulations
⁹⁴ addressing establishment and implementation of individual
baselines for CG produced by certain foreign refiners. The purpose of
these regulations is to assure the compliance of gasoline supplied from
foreign refineries with individual compliance baselines. It includes
comprehensive controls, requirements and enforcement mechanisms to
monitor the movement of gasoline from the foreign refinery to the U.S.,
to monitor gasoline quality and to provide for compliance and
enforcement as necessary.
---------------------------------------------------------------------------

\94\ 40 CFR 80.94.
---------------------------------------------------------------------------

Today we are proposing similar requirements that would apply to any
foreign refiner that can demonstrate that it meets the small refiner
criteria. Foreign refinery baselines would be based on average sulfur
levels and the volume of gasoline imported to the U.S. in 1997-98. Any
foreign refiners that obtain a foreign refinery sulfur rule baseline
would be subject to the same requirements as domestic small refiners
with individual refinery sulfur rule standards. Additionally,
provisions similar to the provisions at 40 CFR 89.94 would apply, that
include:
1. Segregating gasoline produced at the small refinery until it
reaches the U.S.;
2. Refinery registration;
3. Controls on product designation;
4. Load port and port of entry testing;
5. Attest requirements; and
6. Requirements regarding bonds and sovereign immunity.
The rationale for these enforcement provisions is discussed more
fully in the Agency's August 28, 1997 preamble to the final RFG/CG
foreign refineries rule. (See 62 FR 45533 (Aug. 28, 1997)).
By no later than January 1, 2010, ⁹⁵ all gasoline would
be subject to a single national averaged standard and one national
refinery level cap. Thus, EPA is proposing that, beginning on that
date, the use of foreign small refinery baselines would sunset and
standards for all imported gasoline would be met by U.S. importers.
With a single national standard and cap, gasoline sulfur content could
most readily be monitored at the U.S. importer level, since there would
no longer be a special class of gasoline with different standards that
would need to be monitored.
---------------------------------------------------------------------------

\95\ As stated in section IV.C. of the preamble, small refiner
individual refinery standards would sunset January 1, 2008, except
for any small refineries that receive a hardship extension not to
exceed two years.
---------------------------------------------------------------------------

b. What Are the Proposed Requirements for Truck Importers? The
proposed sampling and testing requirements for importers require
sampling and testing of each batch of gasoline. For parties that import
gasoline into the U.S. by truck, the every-batch testing requirement
would include testing the gasoline in each

[[Page 26096]]

truck compartment, or if the gasoline is homogeneous, testing the
gasoline in the truck. However, EPA is concerned that this testing
requirement may not be feasible for truckers hauling many small loads
of gasoline. Since some northern U.S. communities rely, in large part,
on gasoline transported into the U.S. by truck from Canadian terminals,
these communities could suffer gasoline shortages if this requirement
proves too burdensome for truck importers. We therefore propose to
allow alternative requirements for truck-imported gasoline only.
i. Truck Transports of Gasoline (Excluding Gasoline Subject to
Small Foreign Refiner Individual Refinery Standards).
EPA is proposing a limited alternative approach for truck importers
in lieu of every-batch testing. This proposal would be based on the
importer meeting the 30 ppm sulfur average standard on a per-gallon
basis. Under this proposal, the importer would be allowed to rely on
the sulfur results of sampling and testing conducted by the operator of
the truck loading terminal in Canada. The environmental consequences of
this proposal would be neutral, because by meeting the 30 ppm sulfur
standard on an every-gallon basis the standard also is being met on
average.
The importer would be required to demonstrate the gasoline meets
the 30 ppm sulfur standards on an every-gallon basis. The gasoline in
the storage tank from which the importer's trucks are loaded would have
to be sampled and tested subsequent to each receipt of gasoline into
the terminal tank, and these tests would have to show the gasoline
meets the 30 ppm sulfur standard. For each truck load of gasoline, the
importer would have to obtain documents that accurately state the
sulfur content of the gasoline. The importer then would treat each
truck load of imported gasoline as a separate batch for purposes of the
recordkeeping and reporting requirements.
The terminal operator in most cases would not be subject to United
States laws, so the proposal contains safeguards that are intended to
ensure the gasoline in fact meets the applicable standard. First, the
importer would be required to conduct an independent program of quality
assurance sampling and testing of the gasoline dispensed to the
importer. This sampling and testing would have to be at a rate
specified in the proposed regulations, and the sampling would have to
be unannounced to the terminal operator. In addition, EPA inspectors
would have to be given access to conduct inspections at the truck
loading terminal and at any laboratory where samples collected pursuant
to this proposed approach are analyzed. These inspections could be
unannounced, and would include gasoline sampling and testing, and
record reviews.
EPA requests comment on this proposal for parties that import
gasoline by truck. Specifically, EPA requests comment on the provisions
that apply to persons located outside the United States, and the need
for EPA inspectors to conduct inspections at terminals located outside
the United States. In addition, EPA recognizes that the proposed per-
gallon standard of 30 ppm is more restrictive than an annual average
standard with per-gallon caps, although it provides assurance that
gasoline imported by truck will meet the requirements of the sulfur
control program. However, establishing an averaged standard with per-
gallon caps for truck-imported gasoline would require more substantial
recordkeeping, reporting and auditing by the importers and more
compliance monitoring by the EPA. EPA requests comments on the
alternative of allowing an annual average standard with per-gallon caps
for truck importers and the appropriate sulfur standards that should
apply under such an approach.

ii. Truck-Imported Gasoline Subject to Small Foreign Refiner Individual
Refinery Standards

There are additional compliance concerns related to the gasoline
produced by small foreign refiners whose gasoline is imported into the
U.S. by truck. The proposed requirements for gasoline produced at a
small foreign refinery with an individual baseline, and certified as
subject to the individual standard (S-FRGAS), include the necessity of
segregating the gasoline from all other gasoline, from the refinery
gate to the U.S., so that compliance with standards can be tracked.
Under our proposed certified S-FRGAS provisions applicable to other
importers, each batch of gasoline must be tested at the load port and
port of entry. However, in the case of gasoline imported by truck, each
truckload of such gasoline would constitute a batch. Given the small
batch volumes for truck imports, the testing and other procedures
proposed for certified S-FRGAS may not be feasible. The issue is
further complicated because the load port, in effect, stretches from
the refinery, through a pipeline and to a terminal in Canada.
Therefore, EPA is proposing an alternative to the requirement for
testing every truckload of imported certified S-FRGAS.
EPA is proposing that small foreign refiners whose gasoline is
exported to the U.S. by truck would, as part of their petition for an
individual baseline, submit a plan designed to ensure that certified S-
FRGAS remains segregated from all other gasoline from the refinery to
the U.S. The proposed plan would be reviewed for approval in
conjunction with the baseline petition.
Rather than specifying the precise requirements of such a plan in
the regulations, EPA would allow the refiner to develop its own
procedures for ensuring that S-FRGAS remains segregated until it
reaches the U.S. However, EPA believes that any plan would have to
include certain elements. For example, PTDs would have to accompany
each transfer of certified S-FRGAS through the distribution system,
clearly identifying the origin of the gasoline and prohibiting its
commingling with any product other than certified S-FRGAS from that
refinery. The refiner may need to enter into contracts with pipelines
and terminals, if the gasoline is shipped in this manner, that ensure
segregation and prohibit commingling. This certified product could then
only be loaded into trucks if they were importing the gasoline into the
U.S.
The refiner of such gasoline would have to receive and maintain all
such product shipment documents, including U.S. import documents, for
five years and review these on an ongoing basis to ensure segregation
is maintained until reaching the U.S. To further ensure that this
review occurs, EPA is proposing that the refiner's plan would include
attest audit procedures to be conducted annually by an independent
third party that would review the refiner's procedures and records to
ensure that the certified S-FRGAS is segregated at all times. For
example, these procedures would likely include volume reconciliation to
confirm that product is transferred without commingling. However,
additional procedures may be needed to accomplish the goal of ensuring
that certified-S-FRGAS remains segregated from all other gasoline.
3. What Standards Would Apply Downstream?
EPA is proposing downstream per-gallon cap standards that would
apply to all parties in the distribution system downstream of the
refinery-level, including pipelines, terminals, distributors, carriers,
retailers and wholesale purchaser-consumers. Downstream standards would
help ensure the sulfur level of gasoline remains below the cap level
when dispensed for use in motor vehicles, thereby avoiding the adverse
emissions

[[Page 26097]]

consequences of using gasoline with a sulfur content above the cap
level.
EPA is proposing that downstream standards would be more lenient
than the refinery-level cap standards so that refiners and importers
can produce gasoline that equals the refinery-level cap standard. It
has been EPA's experience that if a refiner produces gasoline that
equals, or almost equals a standard, that gasoline may be shown to
violate the standard when subsequently tested at a location downstream
of the refinery due to testing variability. As a result, parties
downstream of the refinery (primarily pipelines) set commercial
specifications for the quality of the gasoline they will accept that
are more stringent than the standard that applies to the downstream
party. This, in effect, forces refiners to produce gasoline that is
``cleaner'' than the refinery-level standard.
In other fuels programs (for example, the benzene per-gallon
standard for RFG) EPA has resolved this concern by announcing
enforcement tolerances for fuels standards that apply downstream of the
refinery-level, thereby reducing the need for pipelines to set
specifications more stringent than the refinery level standards. EPA
believes the approach proposed for the gasoline sulfur cap standards--
more lenient downstream standards--would have the same effect as
announced enforcement tolerances.
EPA is proposing that the values of the downstream cap standards
would reflect the testing variability that could reasonably be expected
when different laboratories test gasoline for sulfur content, that is,
lab-to-lab variability, or reproducibility. For gasoline subject to the
80 ppm refinery-level sulfur cap the proposed downstream standard would
be 95 ppm. This difference reflects the lab-to-lab variability
established by the American Society for Testing and Materials
(ASTM).⁹⁶ For gasoline subject to refinery-level sulfur caps
higher than 80 ppm, which would be the case for gasoline produced
before 2006 and by certain small refiners, the proposed downstream cap
would be similarly established by using the most recent available ASTM
reproducibility data.
---------------------------------------------------------------------------

\96\ ASTM standard method D-2622-98, entitled ``Standard Test
Method for Sulfur in Petroleum Products by Wavelength Dispersive X-
ray Fluorescence Spectrometry.'' The California Air Resources Board
found nearly identical reproducibility under ASTM D-2622-94,
according to a round robin study conducted by ARB and received by
EPA Feb. 11, 1999.
---------------------------------------------------------------------------

As described in section IV.C.3, EPA is proposing that the cap
standards that apply to some small refiners would be higher than the
cap standards that apply to refiners generally. The downstream
standards that apply to this small refiner gasoline would be
correspondingly higher, based on ASTM reproducibility for each
refinery's assigned cap. If gasoline produced by a small refiner with a
higher cap standard is mixed in the distribution system with other
gasoline with a lower cap standard, the entire mixture then would be
subject to the higher cap standard. For this reason, EPA is concerned
that the small volume of small refinery gasoline could drive up the
downstream standard for all gasoline, most of which would have been
subject to the much lower national cap standard.
Therefore, EPA is proposing that during the period small refinery
individual standards are in effect, PTDs must identify whether gasoline
is comprised, in whole or in part, of gasoline produced at a small
refinery with a higher sulfur cap standard than the national cap
standard, and the level of the downstream cap applicable to the
gasoline. A downstream party could rely on the information contained in
the PTDs for gasoline received by that party as the basis for whether
gasoline contains any small refinery gasoline.
However, as gasoline is mixed, and re-mixed, in downstream
pipelines and tanks, the percentage of a particular gasoline that is
small refinery gasoline normally will progressively diminish. For this
reason EPA also is proposing that a downstream party must classify
gasoline as containing no small refinery gasoline if a test result for
the gasoline shows a sulfur content below the applicable national
downstream cap.
Under these proposed requirements, downstream parties and EPA would
know the downstream standard that applies to any particular gasoline.
If the gasoline contains no small refiner gasoline, the downstream
standard would be based on the national cap. If the gasoline is
comprised in whole or in part of small refiner gasoline subject to a
higher cap standard, the downstream standard would be based on this
higher cap standard. This approach would require regulated parties and
EPA to review and rely on the information contained in PTDs.
Following are two examples of how gasoline from small refineries
with individual standards (S-RGAS) would be identified downstream of
the refinery and how the downstream cap would apply:
(1) In 2005 the national refinery cap standard is 180 ppm. If a
small refinery with an individual sulfur cap standard produces a batch
of gasoline that contains 175 ppm sulfur, the transfer document that
accompanies that batch of gasoline into a pipeline may not indicate the
batch contains S-RGAS.
(2) In 2006, when the national downstream cap is 95 ppm, a terminal
receives three shipments of gasoline that are identified in the PTD's
as S-RGAS subject to downstream per-gallon cap standards of 205, 325
and 410 ppm. The terminal operator combines these shipments in a
storage tank. That gasoline mixture is subject to a downstream cap
standard of 410 ppm and any PTD subsequently provided to transferees
must identify the gasoline as containing S-RGAS and state the gasoline
is subject to a downstream cap standard of 410 ppm.
After several additional receipts of gasoline into the storage
tank, the terminal operator obtains a test result indicating the sulfur
level of the mixture is 90 ppm. Based on this test result, the gasoline
mixture becomes subject to the national cap standard of 95 ppm and any
PTD subsequently provided to transferees may not state the gasoline
contains S-RGAS.
EPA requests comment on these proposed downstream standards.
Specifically, we request comment on an alternative whereby gasoline
would be presumed to be subject to the national cap downstream
standard, unless the responsible regulated party were able to
demonstrate through PTDs the presence of small refinery gasoline. EPA
also requests comment on any alternatives that would allow enforcement
of the national downstream cap standards during the period small
refiner individual refinery standards were in effect.
4. What Are the Proposed Testing and Sampling Methods and Requirements?
a. What Is the Primary Test Method for Gasoline? We propose that
the ASTM standard method D 2622-98 be the primary test method for
testing for sulfur in gasoline by refiners and importers. This is the
regulatory method under the RFG/CG rule.⁹⁷ However, we are
requesting comment on whether ASTM method D 5453-93, entitled
``Standard Test Method for Determination of Total Sulfur in Light
Hydrocarbons, Motor Fuels and Oils by Ultraviolet Fluorescence,''
should be the primary method. We are specifically concerned about the
suitability of these test methods for sulfur levels between 0-10 ppm,
and invite comment on other appropriate test methods, including ASTM D
4045, which is used under the California fuels program for sulfur
levels below 10 ppm. We are also requesting

[[Page 26098]]

comment on relative costs of the methods. We believe that ASTM D 5453
would significantly reduce capital costs for test equipment and that
operational costs would be similar to ASTM D 2622. A description of
these ASTM test methods, as well as other methods discussed later in
this section, can be found in Table VI-1, below.
---------------------------------------------------------------------------

\97\ See 40 CFR 80.46(a). The proposed rule would update the
current method, ASTM D 2622-94.

Table VI.-1.--ASTM Standard Test Methods and Practices Described in This
Section
------------------------------------------------------------------------
ASTM No. Title
------------------------------------------------------------------------
D 2622................................. Standard Test Method for Sulfur
in Petroleum Products by
Wavelength Dispersive X-ray
Fluorescence Spectrometry.
D 4045................................. Standard Test Method for Sulfur
in Petroleum Products by
Hydrogenolysis and Rateometric
Colorimetry.
D 4057................................. Standard Practice for Manual
Sampling of Petroleum and
Petroleum Products.
D 4177................................. Standard Practice for Automatic
Sampling of Petroleum and
Petroleum Products.
D 5453................................. Standard Test Method for
Determination of Total Sulfur
in Light Hydrocarbons, Motor
Fuels and Oils by Ultraviolet
Fluorescence.
D 5842................................. Standard Practice for Sampling
and Handling of Fuels for
Volatility Measurement.
------------------------------------------------------------------------

b. What Is the Proposed Test Method for Sulfur in Butane? We are
proposing that ASTM D 5623 would be the regulatory method for testing
the sulfur content of butane. This is the sulfur test method for butane
that the Agency proposed under the RFG/CG rule (proposal published at
62 FR 37338 (July 11, 1997)). However, we received several negative
comments regarding this test method in response to our proposal. We are
requesting comments on other methods and correlation of those methods
to ASTM D 5623. We are also requesting comment on appropriate
correlation procedures and other issues such as bias, accuracy, and
precision.
c. Is EPA Proposing a Requirement To Test Every Batch of Gasoline
Produced or Imported? Under today's proposal, all refiners and
importers ⁹⁸ would be required to sample and test the sulfur
content of each batch of gasoline produced or imported. Test results
would be used to calculate a refiner's or importer's annual average
sulfur level. Any batch of gasoline that exceeded the applicable sulfur
cap could not be distributed or sold in the U.S., unless it was
exempted from this rule, as described later in this section. This
``every-batch'' testing requirement is not a new requirement for RFG
refiners and importers. However, it would be a new requirement for
refiners and importers of CG.
---------------------------------------------------------------------------

\98\ Except for certain truck importers, as noted above.
---------------------------------------------------------------------------

In the past, CG refiners and importers have been allowed to prepare
composite samples of gasoline from multiple gasoline batches and test
the composite sample. However, we believe that every-batch sulfur
testing by refiners and importers is necessary to ensure compliance
with upstream and downstream sulfur caps contained in the proposed
rule. We have proposed the use of alternative test methods to reduce
the cost of testing. We are requesting comment on this proposed
requirement.

i. Butane Blenders' Every-Batch Testing Requirement

Under the RFG rule, refiners that blend butane to previously
certified gasoline (PCG) must determine the volume and parameter values
of the butane, including sulfur content, by testing the gasoline,
before and after blending, and calculating the properties of the butane
by subtracting the volume and parameter values of the PCG. For CG only,
under certain conditions, we have allowed butane blenders to use the
parameter specifications of butane as tested by the butane producer.
This includes an assumed sulfur content of 140 ppm. We have allowed
this alternative to every-batch testing because of the costs of testing
each load of butane.⁹⁹
---------------------------------------------------------------------------

\99\ In addition, commercial grade butane easily meets
conventional gasoline standards, but that is not the case with
regard to the proposed gasoline sulfur standards.
---------------------------------------------------------------------------

We are proposing a similar alternative to every-batch testing for
butane blenders under today's sulfur program. We propose that butane
blenders could use the actual sulfur test result of their suppliers, if
the butane contained less than 30 ppm sulfur and if the butane blender
undertook a quality assurance program to ensure that the supplier's
sampling and testing was accurate. If the butane were tested and found
to violate the 30 ppm cap, the butane blender would be in violation for
the volume of product that exceeded the 30 ppm cap that was added to
gasoline and for any violations of the national downstream cap
resulting from the butane sulfur content. We believe this is a fair
alternative to every batch testing and the only alternative that gives
EPA reasonable ability to monitor compliance. We request comment on
this proposal.

ii. Refiners Blending Other Blendstocks into Previously Certified
Gasoline

Refiners that blend blendstock into PCG would be required to sample
and test each batch of gasoline produced. This would normally include
sampling and testing the PCG to determine its sulfur content and
volume; then sampling and testing the combined product subsequent to
blending; and calculating the sulfur content and volume of the
blendstock (which is the blender's batch for annual average compliance
and reporting purposes), by subtracting the volume and sulfur content
of the PCG from the volume and sulfur content of the combined product.
We are proposing to allow such refiners to meet an alternative testing
requirement in lieu of testing every batch of gasoline. Provided that
the refiner's test result for the sulfur content of each of the
blendstocks is less than the national refinery level per-gallon cap
standard, a refiner could sample and test each blendstock when received
at the refinery, and treat each blendstock receipt as a separate batch
for purposes of compliance calculations for the annual average sulfur
standard.
d. What Sampling Methods Are Proposed? Sampling methods apply to
all parties that conduct sampling and testing under the rule. We are
proposing requiring the use of sampling methods that were proposed in
the July 11, 1997 Federal Register notice (62 FR 37338, at 37341-37342,
37375-37376), which proposes modifications to the RFG/CG rule. These
sampling methods include ASTM D 4057-95 (manual sampling), D 4177-95
(automatic sampling from pipelines/in-line blending), and ASTM D 5842
(this sampling method is primarily concerned with sampling where
gasoline volatility is going to be tested, but it would also be an
appropriate sampling method to use when testing for sulfur). We are
proposing requiring use of these ASTM methods instead of the methods
provided in 40 CFR part 80, Appendix D. That is because the proposed
methods have been updated by ASTM, the updates have provided
clarification and they have eliminated certain requirements, such as
storage tank tap extensions, that are not necessary for sampling light
petroleum products such as gasoline.
e. What Are the Proposed Gasoline Sample Retention Requirements?
We are proposing a refiner and importer sampling and testing
program to establish the sulfur compliance of each batch of gasoline
produced or

[[Page 26099]]

imported. However, we are aware of the inherent drawbacks to a self-
testing scheme. There is the possibility that a party might sample or
test gasoline in a manner not consistent with the required procedures,
or that employees might inaccurately record the test results, by
mistake or otherwise. Under such a scheme, parties might also attempt
to conceal a discovered violation or to save money by not correcting a
violation.
In an attempt to address these concerns about self-testing, we
considered the option of requiring independent sampling and testing for
all gasoline, including conventional gasoline. Under current
regulations, only refiners or importers of reformulated gasoline are
obligated to do this. However, because of the costs of independent
sampling and testing ¹⁰⁰ EPA is instead proposing an
alternative strategy to help ensure refinery and importer sulfur
compliance. Refiners and importers would be required to retain for
thirty days a representative sample from each batch of gasoline
produced, and to provide such samples to the Agency upon request. By
means of this option, EPA could verify the refiner test results.
---------------------------------------------------------------------------

\100\ See the discussion on this subject in the preamble to the
reformulated gasoline program's final rule, 59 FR 7765 (Feb. 16,
1994).
---------------------------------------------------------------------------

This limited duration sample retention would be useful to address
many of the potential problems concerning a refiner self-testing
program. Through this requirement, parties would be faced with the
knowledge that EPA could easily and randomly confirm the accuracy of
the refiner's test results and could discover unrecorded violations. We
believe that this would create an incentive for refiners to sample,
test, and record their sulfur results in an accurate and truthful
manner.
The Agency also is proposing that refiners be required to certify
annually that the samples have been collected in the manner required
under the sulfur rule. This requirement is intended to assure that
refinery officials insist on accurate and honest sampling and retention
of samples at their refineries. We are also proposing that specific
procedures be followed by refiners to properly collect retain, and ship
the samples in a manner consistent with requirements already imposed or
proposed under the RFG program. Under today's proposal, a minimum
representative sample of 330 ml of each gasoline batch would need to be
retained.¹⁰¹
---------------------------------------------------------------------------

\101\ See 40 CFR 80.65(f)(3)(F)(ii), and the Proposed Rule for
Modifications to Standards and Requirements for Reformulated and
Conventional Gasoline, 62 FR 37337 et seq, proposed 40 CFR
80.101(i)(l)(i)(C)(iii).
---------------------------------------------------------------------------

The Agency does not believe that the proposed sulfur rule sample
retention requirements would impose an undue financial burden on
regulated parties. Many refineries already engage in some sample
retention for their own purposes, and the retention procedures proposed
in today's proposal would merely require that typical industry
retention standards be applied. Shipping samples to us would entail
some expense, but this shipping would only occur periodically, and
would certainly cost less than hiring an independent laboratory to
regularly sample and test gasoline.
The Agency requests comments on the costs and effectiveness of the
proposed sample retention requirements, and invites comments on any
alternative plan to promote accuracy of refiner self-testing of
gasoline for sulfur compliance. In particular, we are interested in
information on the cost and effectiveness of a nationwide, independent
sampling and testing program
5. What Federal Enforcement Provisions Would Exist for California
Gasoline and When Could California Test Methods Be Used to Determine
Compliance?
a. Requirement to Segregate Gasoline and To Use Product Transfer
Document Requirements. Today's proposal would generally exempt
California gasoline from regulation under the sulfur rule for the
reasons previously described in this preamble. However, today's NPRM
does propose two requirements that would apply to some California
gasoline. The first would require that gasoline produced outside of
California, that is intended for California use, be segregated from all
other gasoline at all points in the distribution system. Second, the
Agency is proposing that out-of-state producers of gasoline intended
for sale in California be required to create PTDs identifying the
product as California gasoline, and that such PTDs be provided to all
transferees of this gasoline in the distribution system. Such
documentation is intended to facilitate our enforcement of the proposed
sulfur control program through identifying the gasoline not covered by
the federal regulation, even though it is produced in areas otherwise
subject to this proposed regulation. This documentation would also
assist regulated parties in identifying the gasoline as non-federally
regulated to facilitate segregation of California gasoline from federal
gasoline.
The sulfur program PTD requirements for California gasoline
produced out-of-state should not create any new burdens on regulated
parties, since the same requirements currently apply under the RFG
program.¹⁰² Today's proposal would incorporate and restate
the RFG rule's PTD requirements for this California gasoline. The
Agency does not believe that it is necessary to impose additional PTD
requirements under the sulfur program, since the California gasoline
identification requirements under the RFG rule would also satisfy the
identification needs of this rule. Having the same requirements in both
rules means that regulated parties that fail to produce and transfer
the necessary PTD identification would be in violation of both
programs.
---------------------------------------------------------------------------

\102\ See CFR 80.81(g).
---------------------------------------------------------------------------

b. Use of California Test Methods for 49 State Gasoline. As stated
previously, we are proposing to exclude gasoline produced in California
for California use from federal sulfur standards. However, refineries
or importers located in California would have to meet the standards and
other requirements with regard to ``federal'' gasoline used outside of
California. Nevertheless, EPA is proposing that gasoline produced in
California for sale outside of California could be tested for
compliance under the federal sulfur rule using the methodologies
approved by the ARB, provided that the producer complies with the
procedures for such testing as already required under 40 CFR 80.81(h),
which permits California test methods not identical to federal test
methods to be used for conventional gasoline only.
6. What Are the Proposed Recordkeeping and Reporting Requirements?
a. What Are the Proposed Product Transfer Document Requirements? We
are proposing that the PTDs that accompany each transfer of custody or
title of gasoline that includes gasoline produced by any small refiner
subject to sulfur rule individual refinery standards be required to
identify the gasoline as such, including the applicable downstream cap,
as an aid to enforcing the national downstream cap. Other PTD
information is currently required under the RFG/conventional gasoline
regulations. We believe that the additional PTD information regarding
sulfur compliance required under today's proposal would impose little
additional burden on industry. We request comment on this proposed
requirement.

[[Page 26100]]

b. What Are the Proposed Recordkeeping Requirements? We are
proposing to require that refiners and importers keep and make
available to EPA certain records that demonstrate compliance with the
sulfur program standards and requirements. The RFG/CG regulations
currently require refiners and importers to retain records that include
much of the information proposed to be required under today's rule. As
a result, we believe that the proposed reporting requirements would
impose very little additional burden on these regulated parties.
We are proposing to require all parties in the gasoline
distribution system, including refiners, importers, retailers, and all
types of distributors to retain PTDs and records of quality assurance
programs that parties conduct to establish a defense to downstream
violations. All parties in the gasoline distribution system currently
are required to keep PTDs for RFG. However, since there are no
downstream CG standards, only refiners and importers are required to
retain PTDs for conventional gasoline. Because today's proposed sulfur
rule, like the RFG rule, includes downstream standards, we believe that
a requirement to retain PTDs for all parties in the gasoline
distribution system would be appropriate under the sulfur rule. The PTD
information would help us identify the source of any gasoline found to
be in violation of the sulfur standards. The PTDs would also provide
downstream parties with information regarding the applicable downstream
standard.
Today's proposal would require parties to keep records for a period
of five years, with additional requirements for records pertaining to
credits. Records pertaining to credits that were banked and never
transferred to another party would need to be retained for five years
after the credits are used for compliance purposes. Records pertaining
to credits that were transferred would need to be retained by both
parties (transferee and transferor) for ten years after the date the
credits were generated (which would ensure the records are retained at
least years after they are used, since use would have to occur within
five years of generation even if the credits were transferred).
Most of the records that would be required to be kept for five
years already are subject to that requirement by the RFG/CG rule. Five
years is the applicable statute of limitations for the RFG and other
fuels programs. See 28 U.S.C. 2462. We request comment on these
proposed recordkeeping requirements for refiners, importers and
downstream regulated parties. In particular, we request comment on the
record retention provisions specific to credits that were transferred.
While we recognize that retaining records for ten years could be
problematic for both parties, we believe that both parties would need
to retain records so that we could be reasonably sure that credits used
for compliance were appropriate. An alternative, raised earlier in this
proposal, would be to give a more finite life to credits or to require,
beginning in 2006, credits to be used in the same year they were
generated or transferred. We welcome comments on this solution or any
other way in which we can be assured that adequate records would be
available should a credit transaction come into question at some date
longer than five years after the transaction.
c. What Are the Proposed Reporting Requirements? Today's proposed
rule would require refiners and importers to submit to us, on an annual
basis, a report that demonstrated compliance with the applicable sulfur
standards and data on individual batches of gasoline, including batch
volume and sulfur content. The RFG/CG programs contain similar
reporting requirements. Based on our experience with these programs, we
believe that requiring an annual sulfur report and batch information
would provide an appropriate and effective means of monitoring
compliance with the average standards under the sulfur program. The
batch data also would serve to verify that each batch of gasoline met
the applicable sulfur cap standard when it left the refinery. In
addition, the annual report would provide a vehicle for accounting for
any sulfur credits created, sold or used to achieve compliance during
the averaging period.
d. What Are the Proposed Attest Requirements? We are also proposing
to require refiners and importers to arrange for a certified public
accountant or certified internal auditor to conduct an annual review of
the company's records that form the basis of the annual sulfur
compliance report (called an ``attest engagement''). The purpose of the
attest engagement is to determine whether representations by the
company are supported by the company's internal records. Attest
engagements are required under the RFG/CG regulations. We believe that
an attestation for sulfur could be included in a refiner's current
attest engagement with little additional burden.
We believe that the proposed reporting requirements under today's
rule would impose minimal additional reporting burdens on industry
while providing us with information necessary to monitor compliance
with the sulfur standards. We request comment on these proposed
reporting requirements.
7. What Are the Proposed Exemptions for Research, Development, and
Testing?
We are proposing to exempt from the sulfur requirements gasoline
used for research, development and testing purposes. We recognize that
there may be legitimate research programs that require the use of
gasoline with higher sulfur levels than those allowed under today's
proposed rule. As a result, today's rule contains proposed provisions
for obtaining an exemption from the prohibitions for persons
distributing, transporting, storing, selling or dispensing gasoline
that exceeded the standards, where such gasoline is necessary to
conduct a research, development or testing program.
Under the proposal, parties would be required to submit to EPA an
application for exemption that would describe the purpose and scope of
the program and the reasons why use of the higher sulfur gasoline is
necessary. In approving any application, EPA would impose reasonable
conditions such as recordkeeping, reporting and volume limitations. We
believe that the proposal includes the least onerous requirements for
industry that also would ensure that higher sulfur gasoline is used
only for legitimate research purposes. We request comment on these
proposed provisions. We also request comment on whether in lieu of an
approval process, parties should be required to submit the required
information to EPA at the start of the program, and annually
thereafter, with the condition that EPA could provide a party with
written notification in the event the Agency determines the exemption
is not justified. We also request comment on whether the regulations
should impose a volume limit on the amount of gasoline that could be
used in a research program, as a way of minimizing any adverse
environmental effects that could result from allowing such an exemption
from the sulfur requirements.
8. What Are the Proposed Liability and Penalty Provisions for
Noncompliance?
Today's proposed rule contains provisions for liability and
penalties that are similar to the liability and penalty provisions of
the RFG and other fuels regulations.¹⁰³ Under the proposed

[[Page 26101]]

rule, regulated parties would be liable for committing certain
prohibited acts, such as selling or distributing gasoline that does not
meet the sulfur standards, or causing others to commit prohibited acts.
In addition, parties would be liable for a failure to meet certain
affirmative requirements, or causing others to fail to meet affirmative
requirements. For example, persons who produce or import gasoline would
be liable for a failure to fulfill any of the requirements for refiners
and importers, including the sampling and testing requirements, the
reporting and attest audit requirements, the averaging requirements,
the small refinery requirements, and the credit creation and trading
requirements. In such cases the regulated party would also be liable
for any violation of the sulfur standard based on corrected
information. All parties in the gasoline distribution system, including
refiners, importers, distributors, carriers, retailers, and wholesale
purchaser-consumers, would be liable for a failure to fulfill the
recordkeeping requirements and the PTD requirements.
---------------------------------------------------------------------------

\103\ See section 80.5 (penalties for fuels violations); section
80.23 (liability for lead violations); section 80.28 (liability for
volatility violations); section 80.30 (liability for diesel
violations); section 80.79 (liability for violation of RFG
prohibited acts); section 80.80 (penalties for RFG/conventional
gasoline violations).
---------------------------------------------------------------------------

a. Presumptive Liability Scheme of Current EPA Fuels Programs.
Current EPA fuels programs include a presumptive liability scheme for
violations of prohibited acts. Under this approach, presumptive
liability is imposed on two types of parties: (1) That party in the
gasoline distribution system that controls the facility where the
violation was found or had occurred; and (2) those parties, typically
upstream in the gasoline distribution system from the initially listed
party, (such as the refiner, reseller, and any distributor of the
gasoline), whose prohibited activities could have caused the program
non-conformity to exist.¹⁰⁴ This presumptive liability
scheme has worked well in enabling us to enforce our fuels programs,
since it creates comprehensive liability for substantially all the
potentially responsible parties. The presumptions of liability may be
rebutted by establishing an affirmative defense.
---------------------------------------------------------------------------

\104\ Additional type of liability, vicarious liability, is also
imposed on branded refiners under these fuels programs.
---------------------------------------------------------------------------

To clarify the inclusive nature of these presumptive liability
schemes, today's proposed rule would explicitly include causing another
person to commit a prohibited act and causing the presence of non-
conforming gasoline to be in the distribution system as prohibitions.
This is consistent with the provisions and implementation of other
fuels programs.
Today's proposed rule, therefore, provides that most parties
involved in the chain of distribution would be subject to a presumption
of liability for actions prohibited, including causing non-conforming
gasoline to be in the distribution system and causing violations by
other parties. Like the other fuels regulations, a refiner also would
be subject to a presumption of vicarious liability for violations by
any downstream facility that displays the refiner's brand name, based
on the refiner's ability to exercise control at these facilities.
Carriers, however, would be presumed liable only for violations arising
from product under their control or custody, and not for causing non-
conforming gasoline to be in the distribution system, except where we
have specific evidence of causation.
b. Affirmative Defenses for Each Presumptively Liable Party. The
proposal includes affirmative defenses for each party that is deemed
presumptively liable for a violation, and all presumptions of liability
are refutable. The proposed defenses are similar to the defenses
available to parties for violations of the RFG regulations. We believe
that these defense elements set forth reasonably attainable criteria to
rebut a presumption of liability. The defenses include a demonstration
that: (1) the party did not cause the violation; and (2) except for
retailers and wholesale purchaser-consumers, the party conducted a
quality assurance program. For parties other than tank truck carriers,
the quality assurance program would be required to include periodic
sampling and testing of the gasoline. For tank truck carriers, the
quality assurance program would not need to include periodic sampling
and testing, but in lieu of sampling and testing, the carrier would be
required to demonstrate evidence of an oversight program for monitoring
compliance, such as appropriate guidance to drivers on compliance with
applicable requirements and the periodic review of records concerning
gasoline quality and delivery.
As in the other fuels regulations, branded refiners would be
subject to more stringent standards for establishing a defense because
of the control such refiners have over branded downstream parties.
Under today's rule, in addition to the other defense elements, branded
refiners would be required to show that the violation was caused by an
action by another person in violation of law, an action by another
person in violation of a contractual agreement with the refiner, or the
action of a distributor not subject to a contract with the refiner but
engaged by the refiner for the transportation of the gasoline.
Based on experience with other fuels programs, we believe that a
presumptive liability approach would increase the likelihood of
identifying persons who cause violations of the sulfur standards. We
normally do not have the information necessary to establish the cause
of a violation found at a facility downstream of the refiner or
importer. We believe that those persons who actually handle the
gasoline are in the best position to identify the cause of the
violation, and that a refutable presumption of liability would provide
an incentive for parties to be forthcoming with information regarding
the cause of the violation. In addition to identifying the party that
caused the violation, providing evidence to rebut a presumption of
liability would serve to establish a defense for the parties who are
not responsible. Presumptive liability is familiar to both industry and
to us, and we believe that this approach would make the most efficient
use of EPA's enforcement resources. For these reasons, we are proposing
a liability scheme for the sulfur program based on a presumption of
liability. We request comment on the proposed liability provisions.
c. Penalties for Violations. Section 211(d)(1) of the CAA provides
for penalties for violations of the fuels regulations.¹⁰⁵
Today's rule proposes penalty provisions that would apply this CAA
penalty provision to the sulfur rule. The proposed provisions would
subject any person who violates any requirement or prohibition of the
sulfur rule to a civil penalty of up to $27,500 for every day of each
such violation and the amount of economic benefit or savings resulting
from the violation. A violation of the applicable average sulfur
standard would constitute a separate day of violation for each day in
the averaging period. A violation of a sulfur cap standard would
constitute a

[[Page 26102]]

separate day of violation for each day the gasoline giving rise to the
violation remained in the gasoline distribution system. The length of
time the gasoline in question remained in the distribution system would
be deemed to be twenty-five days unless there is evidence that the
gasoline remained in the gasoline distribution system for fewer than or
more than twenty-five days. The penalty provisions proposed in today's
rule are similar to the penalty provisions for violations of the RFG
regulations. EPA requests comment on these provisions.
---------------------------------------------------------------------------

\105\ Section 211(d)(1) reads, in pertinent part:
(d)(1) Civil Penalties.--Any person who violates * * * the
regulations prescribed under subsection (c) * * * of this section *
* * shall be liable to the United States for a civil penalty of not
more than the sum of $25,000 for every day of such violation and the
amount of economic benefit or saving resulting from the violation. *
* * Any violation with respect to a regulation prescribed under
subsection (c) * * * of this section which establishes a regulatory
standard based upon a multi-day averaging period shall constitute a
separate day of violation for each and every day in the averaging
period. * * *
Pursuant to the Debt Collection Improvement Act of 1996 (31
U.S.C. 3701 note), the maximum penalty amount prescribed in section
211(d)(1) of the CAA was increased to $27,500. (See 40 CFR part 19.)
---------------------------------------------------------------------------

9. How Would Compliance With the Sulfur Standards Be Determined?
We have often used a variety of evidence to establish non-
compliance with requirements imposed under our current fuels
regulations. Test results of the content of gasoline have been used to
establish violations, both in situations where the sample has been
taken from the facility at which the violation is found, and where the
sample has been obtained from other parties' facilities when such test
results have had probative value of the gasoline's characteristics at
points upstream or downstream. The Agency has also commonly used
documentary evidence to establish non-compliance or a party's liability
for non-compliance. Typical documentary evidence has included transfer
documents identifying the gasoline as inappropriate for the facility it
is being delivered to, or identifying parties having connection with
the non-complying gasoline.
a. What Evidence Could Be Used to Establish Sulfur Rule Violations
and Liability for these Violations? A recent EPA Environmental Appeals
Board decision, (In re: Commercial Cartage Company, Docket No. CAA-93-
H-002, CAA Appeal No. 97-9) (the ``Cartage'' decision), interpreted the
regulatory language of one of EPA's fuels programs as restricting the
evidence that the Agency may use in establishing a violation of a
standard under that program. Under the Cartage decision, in order to
establish the existence of a violation of the gasoline volatility
standards ¹⁰⁶ at a particular carrier or retail outlet
facility, we would have to produce non-compliant test results obtained
only by using the regulatory method and only from a sample taken from
the facility itself. Other potentially persuasive evidence establishing
volatility standard violations would not be permitted under the Cartage
decision's interpretation of the volatility rule.¹⁰⁷
---------------------------------------------------------------------------

\106\ EPA's gasoline volatility regulations are found at 40 CFR
80.27 and 80.28.
\107\ See 40 CFR 80.27(b) and 80.28(b) and (e).
---------------------------------------------------------------------------

We believe that it would best serve the purposes of the proposed
sulfur rule to not limit the evidence that may be used to show whether
a violation occurred or liability for that violation. Our enforcement
experience in other programs has shown that the Cartage-permitted
evidence (test results from samples taken only from a particular
facility, and using only the regulatory test methods) often does not
exist, while other persuasive evidence of the existence of the
violations does exist. If we are not able to use other forms of
persuasive evidence to establish violations or other necessary facts
short of test results such as those permitted by the volatility
regulations under the Cartage interpretation, violators will continue
to avoid liability for their actions.
To ensure that evidence with probative value could be used under
the sulfur rule, the Agency is making explicit in today's proposal that
any probative evidence could be used to establish compliance or non-
compliance with the sulfur standards and requirements and liability for
non-compliance. This would not remove or change the obligation on
refiners and importers to perform testing on each batch of gasoline
using the procedures authorized under these regulations. Compliance or
non-compliance with sulfur standards would continue to be based on
regulatory test methods. However, other probative evidence could be
used to determine compliance with sulfur standards if the evidence is
relevant to whether the sulfur content would have been in compliance if
the appropriate sampling and testing methodologies had been performed.
Under today's proposal, the permitted probative evidence
specifically includes information obtained from any source or any
location, since Agency enforcement experience has proven the value of
such widely-obtained material. Respondents in EPA enforcement actions
would have the same right to present other evidence of compliance with
the sulfur rule as the Agency would have to establish non-compliance.

VII. Public Participation

We received many comments from a range of interested parties on our
Tier 2 Report to Congress. We have also received comments as part of
the our outreach to small entities (see section V.B.). These comments
have been very valuable in developing this proposal, and we look
forward to additional comment during the rulemaking process. You can
find comments on the issuance of Tier 2 standards and gasoline sulfur
control we received prior to this proposed action in the rulemaking
docket, and many of them are discussed in the context of various issues
in this preamble. We have considered comments received during the
development of the proposal and have addressed a number of them in
today's document.

A. Comments and the Public Docket

Publication of this document opens a formal comment period on this
proposal. You may submit comments during the period indicated under
DATES above. The Agency encourages all parties that have an interest in
the program described in this document to offer comment on all aspects
of the action. Throughout this proposal you will find requests for
specific comment on various topics.
The most useful comments are those supported by appropriate and
detailed rationales, data, and analyses. We also encourage commenters
who disagree with the proposed program to suggest and analyze alternate
approaches to meeting the air quality goals of this proposed program.
You should send all comments, except those containing proprietary
information, to the EPA's Air Docket (see ADDRESSES) before the date
specified above for the end of the comment period.
Commenters who wish to submit proprietary information for
consideration should clearly separate such information from other
comments. Such submissions should be labeled as ``Confidential Business
Information'' and be sent directly to the contact person listed (see
FOR FURTHER INFORMATION CONTACT), not to the public docket. This will
help ensure that proprietary information is not placed in the public
docket. If a commenter wants EPA to use a submission of confidential
information as part of the basis for the final rule, then a
nonconfidential version of the document that summarizes the key data or
information must be sent to the docket.
We will disclose information covered by a claim of confidentiality
only to the extent allowed by the procedures set forth in 40 CFR Part
2. If no claim of confidentiality accompanies a submission when we
receive it, we will make it available to the public without further
notice to the commenter.

B. Public Hearings

We will hold four public hearings as noted under ``DATES'' above.
If you would like to present testimony at the

[[Continued on page 26103]]

Notices For
2009
2008
2007
2006
2005
2004
2003
2002
2001
2000
1999
1998
1997
1996
1995
1994

[[pp. 26053-26102]] Control of Air Pollution From New Motor Vehicles: Proposed Tier 2

Related Material

Local Navigation