Comment submitted by Alfonse Mannato, American Petroleum Institute (API)

This is a Comment on the Environmental Protection Agency (EPA) Proposed Rule: Regulation of Fuels and Fuel Additives: Renewable Fuel Standard Program

For related information, Open Docket Folder  Docket folder icon


Comment




November 12, 2006

U.S. Environmental Protection Agency
Air Docket
1200 Pennsylvania Ave., NW
Mail Code 6102T
Washington, DC 20460
Attention: Docket ID Number: EPA-HQ-OAR-2005-0161

Re:The U.S. Environmental Protection Agency's Proposed Rule on
Regulation of Fuels and Fuel Additives: Renewable Fuel Standard Program
71 Fed. Reg. 55551

The American Petroleum Institute (API) is pleased to provide the attached
comments on the above referenced notice of proposed rulemaking. API is a
national trade association representing more than 400 member companies
involved in all aspects of the oil and natural gas industry. API members are
dedicated to meeting environmental requirements, while economically developing
and supplying energy resources for consumers. API members provide the fuels
that keep America running.

API believes that it is critical that fuel regulation retain sufficient flexibility to allow
industry to meet America?s energy needs. In order to meet the growing needs of
consumers, we understand that we must produce, refine and distribute more fuel
than ever before under continually restricting controls. Flexibility in the national
renewable fuel standard (RFS) plan is essential in order to integrate renewables
into the nation?s gasoline pool in the quickest and most effective way possible.
One of the greatest impediments to an efficient and workable national RFS is the
rising tide of state biofuels mandates. These threaten to remove most of the
flexibility that Congress and now EPA have carefully and intentionally put into the
national RFS program.

When API along with other stakeholders began talking to EPA about this
regulation, we developed a set of principles whose overall goal was to maximize
market efficiency. These principles can be described with just a few words:
workability; transparency; enforceability; low cost, and certainty.

There are some technical issues yet to be resolved, but overall the approach that
EPA has chosen conforms to all of these principles. Moreover, EPA has worked
cooperatively with all stakeholders ? demonstrating the benefits of dialogue and
idea exchange between business and the regulatory community.

U.S. EPA Air Docket
November 12, 2006
Page 2

API recommends that EPA make several changes to the proposed rule to help
ensure a smooth transition into the program during 2007. A smooth transition is
made more difficult because the program provision have yet to be established and
once set, will apply for a partial year API member companies are committed to
meeting the requirements of the RFS in a timely and effective manner and oppose
the suggestion made by several parties at the hearing that full implementation of
the program should be delayed until 2008. There is no legal justification for the
delayed implementation. EPA must finalize the regulation before the end of
March and must start program implementation at least by July 1, 2007. If EPA is
unable to meet this date, we will have to reevaluate this issue.

The changes that should be made include:
? The final regulations should be prospective and become effective no sooner than
60-120 days after they are issued. Moreover, they should apply to RIN generators
and obligated parties at the same time. That is, the date on which the RIN
obligation commences must be identical with the date on which RIN creation
begins.
? EPA needs to better coordinate the small refineries exemption. EPA has
assumed in calculating the 2007 RFS required percentage of 3.71, that all small
refineries (in addition to the small refiners) will be exempt. However, under the
proposal these parties are obligated parties until their application for exemption is
approved. These applications are not due to EPA until September 1, 2007. We
suggest that EPA finalize (and announce in a supplemental notice) an earlier,
accelerated schedule for small refinery exemption application so that small
refineries can qualify for the exemption prior to the effective date of the final rule.
This will avoid uncertainty regarding the RFS obligation status of small refineries
for 2007 (and beyond) while preserving the principle that small refineries must
qualify for the exemption. Also, the RFS required percentage for 2007 should take
into account only those small refineries that have been fully confirmed as exempt.

The proposed credit banking and trading program should allow refiners to use
renewables where they are most efficient and cost effective. This flexibility is of
critical importance and should facilitate the integration of ethanol and other
renewable fuels into the gasoline and diesel marketplace in the most efficient
manner.

Several of the important proposed flexibility measures that API believes are
critically important to the operation of this program are:
? The registration of renewable fuel producers and importers with the assignment
of Renewable Identification Numbers (RINs) by renewable fuel producers to every
batch of renewable fuel produced or imported.
? The limitation that only obligated parties, defined as a party that produces or
imports gasoline or parties that actually blend the renewable fuel into a finished
fuel, can separate RINs from renewable fuels for use in demonstrating compliance
or trading.
? EPA's definition of RIN life to include the current year and the year following.
U.S. EPA Air Docket
November 12, 2006
Page 3

? Allowing obligated parties to carry forward to the following year at least a portion
of their unused RINs.
? The use of fuel energy content as compared to corn ethanol to determine RIN
equivalence values for other renewables on a complete and equitable basis.
? The RIN rollover cap should be at least 30% to provide flexibility for response to
potential ethanol shortages arising from drought conditions.
? The inclusion of a deficit carryover provision for obligated parties to allow
optimized compliance approaches.
? Annual reporting by obligated parties is sufficient while quarterly reporting would
be extremely burdensome and add nothing to the enforceability and efficiency of
the program.

The detailed API comments are attached.

Please contact me at 202-682-8180 if you have any questions about these
comments.

Thank you,

Alfonse Mannato


Docket ID No. EPA-HQ-OAR-2005-0161
Re: 40 CFR Parts 80; Regulation of Fuels and Fuel Additives: Renewable Fuel
Standard Program; Proposed Rule, 71 FR 55551, September 22, 2006

API Comments ? November 12, 2006

API is a national trade association representing more than 400 member
companies involved in all aspects of the oil and natural gas industry. API
members are dedicated to meeting environmental requirements, while
economically developing and supplying energy resources for consumers. API
members provide the fuels that keep America running.

Before EPA issued this proposed rule, API and NPRA developed a set of
principles whose overall goal was to maximize program efficiency. These publicly
stated principles can be described with just a few words: workability;
transparency; enforceability; low cost, and certainty.

There are some technical issues yet to be resolved, but overall the approach that
EPA has chosen conforms to all of these principles. Moreover, EPA has worked
cooperatively with all stakeholders ? demonstrating the benefits of dialogue and
idea exchange between business and the regulatory community.

As a result API supports most aspects of the proposal as follows:

API supports EPA's definition of the parties subject to the renewable volume
obligation that is based on the renewable fuel standard. Parties producing
gasoline, including refiners, importers, and blenders (other than oxygenate
blenders) should be the obligated parties.

API supports EPA's definition of the specific gasoline volumes that would serve as
the basis of the obligation, namely all finished gasoline produced or imported for
use in the contiguous 48 states, and any reformulated or conventional
blendstocks for oxygenate blending, or in the case of blenders, blendstocks
added to finished or unfinished gasoline. We concur with EPA that renewable fuel
volumes should not be counted as gasoline for this purpose.

API supports EPA?s use of fuel energy content, as compared to the energy
content of corn ethanol, as the basis of establishing RIN equivalence values (with
one exception as noted below). Equivalence values should be uniformly applied to
all renewables, regardless of production means or use location. A well-to-wheels
analysis would be preferred if an accepted, consistent calculation method were
available to determine respective well-to-wheel values. However, such calculations
are too variable at this juncture for reliable use and energy content is a reasonable
surrogate.

API supports the registration of renewable fuel producers and importers and the
assignment of Renewable Identification Numbers (RINs) by renewable fuel
producers. We support the use of RIN as the basic mechanism for compliance
demonstration by obligated parties, and as the mechanism for fulfilling the credit
trading provision in the Energy Policy Act of 2005 (EPACT05).

API supports the limitation that only obligated parties or parties that actually blend
the renewable fuel into a finished fuel can separate RINs for use in demonstrating
compliance or trading. We suggest that distributors of neat renewable fuels for
use as motor vehicle fuel be treated in a manner similar to oxygenate blenders.

API supports EPA's definition of RIN life to include the current year and the year
following.

API supports the provision for obligated parties to carry forward to the following
year a portion of their renewable obligation as long as the obligation carried
forward and the following year obligation are fully met the following year.

API supports the requirement that any party holding RINs be subject to the
reporting and recordkeeping requirements of the program. Annual, not quarterly,
reporting should be sufficient. Quarterly reporting would be unduly burdensome
and add little in the way of useful information since the compliance basis is annual
and multiple "interim" reports may not be indicative of ultimate compliance status.

API supports the requirement that RINs associated with volumes of renewable fuel
exported must be retired, as well as the other aspects of the program for
renewable fuel exporters. We suggest that EPA clarify in the final rule preamble
that volumes of renewable fuel produced in the contiguous 48 states and delivered
to Alaska or Hawaii are exports and their RINs must also be retired unless Alaska
or Hawaii decides to opt-in to the RFS program. Similarly, EPA should also
clarify that any RINs attached to or associated with renewable fuel blending into
distillate fuel intended for use in space heating or as furnace fuel must also be
retired.

API recommends that EPA make several changes to the proposed rule to help
ensure a smooth transition into the program during 2007. A smooth transition is
made more difficult because the program provision have yet to be established and
once set, will apply for a partial year API member companies are committed to
meeting the requirements of the RFS in a timely and effective manner and oppose
the suggestion made by several parties at the hearing that full implementation of
the program should be delayed until 2008. There is no legal justification for the
delayed implementation. EPA must finalize the regulation before the end of
March and must start program implementation at least by July 1, 2007. If EPA is
unable to meet this date, we will have to reevaluate this issue.

The changes that should be made include:
? The final regulations should be prospective and become effective no sooner than
60-120 days after they are issued. Moreover, they should apply to RIN generators
and obligated parties at the same time. That is, the date on which the RIN
obligation commences must be identical with the date on which RIN creation
begins.
? EPA needs to better coordinate the small refineries exemption. EPA has
assumed in calculating the 2007 RFS required percentage of 3.71, that all small
refineries (in addition to the small refiners) will be exempt. However, under the
proposal these parties are obligated parties until their application for exemption is
approved. These applications are not due to EPA until September 1, 2007. We
suggest that EPA finalize (and announce in a supplemental notice) an earlier,
accelerated schedule for small refinery exemption application so that small
refineries can qualify for the exemption prior to the effective date of the final rule.
This will avoid uncertainty regarding the RFS obligation status of small refineries
for 2007 (and beyond) while preserving the principle that small refineries must
qualify for the exemption. Also, the RFS required percentage for 2007 should take
into account only those small refineries that have been fully confirmed as exempt.

Recognizing the nearly instantaneous compliance imposition of the proposed
provisions, the Agency must recognize that any major rule changes from that
proposed can significantly alter our obligated member?s ability to meet the
administrative aspects of the rule, e.g., renewable fuel producers certifying RIN on
time. We, therefore, continue to urge EPA to promulgate this rule expeditiously,
simplify the process for exempting small refiners, line up the effective date for
obligated parties and registration of renewable producers, and reach out to the
renewable producers and importers to educate them concerning the rule
requirements, and encourage them to register early to ease implementation of the
rule.

We offer the following additional comments and suggestions regarding some
detailed aspects of the program that should be changed:

EPACT05 specifies that small refineries are exempt until 2011. EPA has
proposed that both small refineries and small refiners be exempt through 2010.
API opposes expansion of this exemption to include small refiners. Since
Congress was explicit in exempting small refineries, it is clear that they did not
intend that the exemption be broadened to also include small refiners. EPA has
exceeded its discretionary authority by extending the statutory small refinery
exemption to small refiners. When this exemption is eliminated, EPA will need to
recalculate the proposed 2007 RFS percentage. Proposed section 80.1142
should be deleted in its entirety.

EPA should require that all RINs remain attached to the renewable fuel until the
renewable fuel is either obtained by an obligated party or blended into a finished
fuel. There is no logical basis for allowing extra-value RINs to be retained by the
renewable fuel producer any more than there is for them to retain standard-value
RINs. This proposed approach would allow these producers to withhold these
RINs from the market and greatly increases the complexity and administrative
burden of the program. The extra-value RINs are more likely to serve as a
production incentive (i.e. demand) if the parties obligated under the RFS receive
the full RIN value (standard and extra value combined).

Rollover RINs should not be capped. EPA has proposed an RIN ?rollover cap? of
20%. That is, an obligated party?s RIN obligation for a current year can be met
with a maximum of 20% RINs not used during the previous year. API believes this
restriction needlessly restricts RIN utilization and requests that this cap be
removed or increased at least to 30%. While a 20% cap may compensate for the
most recent drought (during which ethanol production was reduced approximately
20%), it provides little insurance against bigger droughts or against other
conditions that may constrain supply. Prudence dictates a larger cap. If EPA
establishes a cap that is too generous, there are few, if any, negative
consequences besides a large RIN bank. However, if the cap is too small to
compensate for a severe drought or other unforeseen circumstance, there could
be severe economic consequences because the RIN market will be unable to
match supply and demand. If EPA does finalize a cap on the use of rollover RINs,
the Agency should make the cap as large as possible to provide flexibility for
response to potential ethanol shortages arising from drought conditions.

Obligated parties will be reporting obligations and RIN reports for annual
compliance under aggregated approaches. On this basis, a corporation will need
to be recognized as a "facility" to facilitate trades between obligated entities. The
regulation should be clear and providing this approach.

EPA should not impose a compliance system that would require the tracking of
RINs to the point where they are blended.

To address permeation emissions, EPA should require that new vehicles be
certified on E10. API recommends that EPA begin to make the necessary vehicle
emission regulatory revisions to correct the impact of increased fuel permeation
on new vehicles due to increased ethanol blending. EPA could insure that new
vehicles continue to meet current emission standards by changing the certification
fuel to E10. E10 is now the predominant fuel in urban areas most challenged for
ozone attainment and new vehicles should be designed to comprehend this and
the growing use of ethanol as a gasoline blend stock.

EPA should modify the RIN character numerical code to accommodate larger
batch volumes. The proposed 34 character RIN code contains two six digit codes
representing the batch starting gallons (SSSSSS) and the batch ending gallons
(EEEEEE). Having only six digits limits batch size to six significant figures, or
999, 999 gallons, which equals slightly less than 24,000 barrels. Many batches of
renewable fuel could be larger than this, particularly water borne cargos.
Expanding the starting and ending gallon codes to seven digits would correct this
deficiency.

API supports the proposed regulation?s definition and treatment of renewables ?
with one major exception. The proposal covers renewable fuels that can be
blended in or are mixtures in either gasoline or diesel and those that can be used
in their neat (unblended) form as motor vehicle fuel, with the exception of
biodiesel. EPA proposes that biodiesel must be blended into conventional diesel
at a concentration of 80% or less. API does not support this limitation and
believes that biodiesel blends up to 99.99% should be allowed. Also, as with
other renewables, EPA should allow the stripping of biodiesel RINs by obligated
parties after purchase rather than after blending.

EPA?s proposal to assign RINs valued at 1.0 to ?a portion of a batch? for a
renewable fuel that has an equivalence value of <1.0 will cause accounting
confusion. For accuracy, understanding and accountability, an appropriate
fractional value should be assigned to every gallon of a renewable with an
equivalence value of <1.0.

As a matter of environmental policy, even though ETBE is made partially from a
renewable fuel, ETBE production should not be encouraged by establishing an
equivalence value for ETBE at this time. Since ethanol can be blended directly
into gasoline, there is no need to use it to produce its ether counterpart that has
unfavorable qualities in terms of its potential impact on groundwater in event of a
spill or leak of blended gasoline. Like MTBE, ETBE in ground water has very low
odor and taste thresholds. EPA's own Blue Ribbon Panel on Oxygenates in
Gasoline recommended in 1999 that the use of MTBE and other ethers in gasoline
be reduced. The EPACT05 requires that the Agency examine the health and
environmental impacts of potential substitutes for MTBE, including ETBE. ETBE
is expected to show similar hydrogeologic characteristics to MTBE and its use
should not be encouraged by EPA by establishing an equivalence value for ETBE
until EPA's evaluation is complete and then only if the Agency determines that
ETBE does not pose a threat to the environment in a manner similar to MTBE.

The EPA proposal includes a requirement that the RIN be included on a product
transfer document (PTD). EPA states that the PTD can be included in any form of
standard documentation that is already associated with or used to identify title to
a batch. API has a concern with EPA?s use of a bill-of-lading as an example of a
document that would require a RIN number. A bill-of-lading documents a transfer
of custody and not a transfer of title. A common carrier terminal operator that
blends oxygenates into gasoline should not be required to include the RIN on the
bill-of-lading issued to the driver at the time of loading. API believes that, in this
situation, the appropriate document to reflect the RIN is the document which
transfers title to the purchaser which is often an invoice (not a bill-of-lading)
between the renewable producer and the blender.

Renewable diesel that is produced through co-processing in a hydrotreater should
have an equivalence value greater than 1.0 assigned to the biocrude. The
equivalence value of biocrude based renewable fuel generated at refineries should
also be based on the energy value of the bio-produced materials. The equivalence
value of 1 proposed by EPA is inappropriate given biodiesel and non-ester
renewable diesels are 1 to 1.5 and 1.7. This unequal approach handicaps refinery
operations and the effective utilization of biocrude. Therefore, if biocrude to
renewable motor fuel yield is one-to-one in volume, this biocrude should receive an
equivalency value based on the energy content of the bio-produced materials.

In a parenthetical statement at page 55575, the Preamble states that ?... the
refiner operates as an oxygenate blender.? EPA should clarify that what was
meant there was ?refinery? not ?refiner.? This would be consistent with the
regulations as proposed.

Obligated party using RINs that are later found to be invalid should be given
opportunity to ?cure? a shortfall caused by ?invalid? RINs without penalty.

Independent audit or attestation provisions should not be required of obligated
parties. EPA can easily check producer versus obligated party use of RINs.
Also, the data is not such that an attest is needed. There is no verification of raw
data as with other programs (such as lab results for RFG, batch volumes to
pipeline tickets, overall volume balances, etc.). Reports allow EPA to crosscheck
data with other reporting entities. Renewable producers need an attest to verify
reported production and RIN volumes. If there are attest requirements for
obligated parties, then the requirement needs to change significantly. The current
proposed language requires the auditors to check documentation for every RIN
transaction (there will be thousands). Instead, at most, a limited sample should
be required.

EPA should clarify how the rule applies to transmix processors. API believes that
EPA should expressly state that, only to the extent that previously unaccounted
blendstocks are added, transmix processors should be treated similarly
to "blenders" so that, as EPA treats blenders in general, they ?would count as
their gasoline production only the volumes of blendstocks added to finished or
unfinished gasoline? Also, transmix blending operations (as opposed to transmix
processors) should be exempt from RIN obligations when blending at levels not
requiring blendstock reporting.

EPA should generally reference ASTM standards for biodiesel and other
renewables for which standards exist. However, these references should be
generic and not tied to a particular version of a standard. In this way, EPA would
not be required to conduct additional rulemaking merely to update an ASTM
reference. For example, EPA stated that biodiesel (ester-based) is defined as
meeting ASTM D-6751-02a. In fact, the most current specification is D-6751-06a.
Other ASTM specifications for fuel ethanol (D 4806) and E85 (D 5798) should also
be included to help promote product quality and eliminate ambiguity in what
constitutes a specific renewable. For example, on page 11 of the pre-publication
preamble EPA says that E85 is "a blend of 85 percent ethanol and 15 percent
gasoline." In fact, the ethanol can range from 70% to 85% according to the ASTM
specification and varies accordingly in actual practice, as it must to meet vapor
pressure requirements. EPA also asks for comment on whether non-ester
biodiesel blends should meet D-975. API supports this approach, again provided
the reference to the standard is to ?the most current version? of the standard and
is not tied to a specific version.

With regard to the RIN number, API supports the voluntary labeling concept that
EPA requests comments on. EPA should build the voluntary labeling concept
into the RIN at this time, and work out the details of the voluntary program later.

The following are comments on, or questions about the specific regulatory
language in the proposal. They are not intended to be all inclusive of the
comments made.

The requirement to evaluate the term RBx in 80.1107(b) seems to require tracking
all renewables and their volumes to the blend point. The RBx term simply should
be dropped since 80.1107(d) prevents counting renewables volume as gasoline
volume.

Table 1 of 80.1115: ?ethanol from corn, starches, or sugar? should be ?Denatured
ethanol from?? in accordance with p.51 of the pre-publication version preamble.

80.1128(b)(1)(i): Extra value RINs should be separable only by obligated parties
and not by the producer, as discussed above.

80.1128(b)(4): There is no clear rationale for limiting splitting of renewables
batches into only two pieces. In fact, 80.1128(a)(3) implies a batch can be split
into more than two pieces. API suggests replacing ?two? with ?any number of? in
80.1128(b)(4).

80.1129(a)(2)(v): As discussed above, this provision would require tracking of RINs
all the way to fuel blending. EPA should make the same valid assumption for
biodiesel as for ethanol, i.e., that once produced, biodiesel will be used for motor
fuel.

80.1130(a)(2): Confining applicability of 80.1130 to exporters from an applicable
region will not achieve the desired result since an exporter could be anywhere in
the world. This provision should apply to the physical product exported from an
applicable region.

80.1141(e), 80.1142(f), 80.1152(a)(1): These are the only mentions of refinery
aggregation for compliance purposes in the regulatory language. Aggregation
should be made an explicit option.

80.1141(f)(3): "80.1105" doesn?t refer to the production of gasoline. This should be
a demonstration of the inability to meet the provisions of 80.1105.

As discussed above, 80.1142 should be eliminated. Congress was clear about
exemptions that would be allowed and small refiners were not mentioned.
Exacerbating this problem is that per 80.1142(d) two small refiners could merge to
become large and still be exempted.

80.1143: There needs to be a provision for small refineries and non-contiguous
states and territories to opt-in effective the first compliance period (2007). As the
proposed rule now reads, such opt-ins could not become effective until the 2008
compliance period.

80.1125(h): Renewable batches may need to be 7 characters (SSSSSSS and
EEEEEEE) to accommodate a large batch of renewables, i.e., over 999,999 gals,
as discussed above.

The EPA?s assessment of environmental impacts does not consider all media
and is therefore incomplete
API does not disagree with the conclusions reached by the EPA regarding the
general magnitude and direction of the environmental impacts of the proposed
renewable fuels program. However, as the EPA itself notes in the Preamble to the
Proposal, the assessment is incomplete for it ignores impacts on water quality
that might result from the increased use of renewable fuels. This is a serious
omission. The recent rapid expansion in the construction of ethanol production
facilities has elevated public concern in some areas about the adequacy of
existing groundwater supplies. The EPA acknowledges that ?? changes in
agriculture as a result of increased use of renewable fuels can have significant
adverse effects on water quality, either locally or on a more broad basis, ? but it
fails to provide any evaluation of the potential environmental damage from
increased acreages of agricultural feedstocks for renewable fuels, including the
impacts of pesticides, nitrate runoff into water supplies, or the increased demand
on water, as ?energy crops? like corn and soy begin to displace more drought
tolerant crops such as wheat in several Midwestern states.

The EPA should better characterize the uncertainty associated with its estimates
of the impacts of the renewable fuels program on emissions and air quality
The EPA evaluated the emissions inventory and air quality impacts associated
with its proposed renewable fuels program. Their assessment concluded that the
expanded use of renewable fuels will provide reductions in carbon monoxide and
some air toxics such as benzene, while other emissions, notably VOC and NOx,
will increase. Overall, the EPA found that the ozone air quality impacts
associated with the program would be minimal ? an average ozone increase of
0.06 ppb of <0.1% of the standard for the eastern US.

API does not disagree with the EPA assessment. However, we note that the
EPA?s conclusions are supported by underlying models, data and assumptions
which contain numerous elements of uncertainty that warrant further investigation,
testing and research. Some of these areas of uncertainty include, for instance:
? Use of either old or limited data which may not be reliable for estimating the
effects of fuel property changes particularly for vehicles equipped with advanced
emissions controls in the 2012 fleet. We agree with the EPA?s observation that
that existing models (such as the Complex Model, the Predictive Model and
MOBILE6.2) for evaluating fuel factor effects are based on technology that is more
representative of the 1990s than of the present or oncoming decade. There is
clearly a need to update these models with information from fuel factor effect test
programs on newer technology vehicles such as those contained in recent reports
of the Coordinating Research Council (CRC) relating to the effects of ethanol and
gasoline volatility on exhaust emissions and the effects of ethanol on permeation
emissions.
? The impact of ethanol on emissions from non-road equipment. The EPA
projections of emissions inventories for mobile sources that were provided in the
nonroad diesel rulemaking suggest that the non-road sector will account for an
increasingly larger proportion of the total in the future. However, while a crude
sensitivity analysis was performed for the on-highway mobile source sector (e.g.,
assuming fuel factor sensitivity for all vehicles, not just Tier 0), the EPA failed to
conduct any sensitivity analysis for criteria pollutant emissions from non-road
vehicles even though they have acknowledged that the underlying data were
extremely limited ? both with respect to emission factors, activity levels and
equipment population estimates. We strongly encourage the EPA to perform
such an analysis. It would provide, at a minimum, a preliminary measure of the
uncertainty associated with the contribution of the non-road sector to future mobile
source emissions inventories.

? Need for more test data to evaluate the emissions effects of E85 in FFVs. As
EPA notes in the draft RIA, current projections of E85 show its usage to be very
small compared to E10. While this indicates that the overall emissions impacts
of FFVs fueled with E85 will be small, this expectation is based on extremely
limited published information available on the emissions characteristics of modern
technology FFVs. The available data relate to tests performed on FFVs produced
in the early and mid-1990s and standard EPA emissions certification tests and
suggest that E85 will increase NMOG and acetaldehyde emissions while having
mixed effects on other criteria pollutants and air toxics. To the extent that FFV
penetration and usage increases in the future, whether by market incentives or
other means, it will be important to for EPA to collect more data to better
characterize the emissions implications associated with fueling these vehicles on
E85 blends.

? Uncertainty in the level of future ethanol usage will influence ozone impacts.
The EPA noted that the ozone analysis did not include consideration of the
impacts of CO reduction from ethanol usage nor did it include consideration of the
impact of ethanol on changes in the types of compounds comprising VOC
emissions ? factors which might ameliorate the projected ozone increases
resulting from the proposed program. However, the EPA only focused on a 7.5
billion gallon renewable fuels scenario in a 37-state eastern area of the US. The
projected ozone impacts will likely be larger and more widespread if the EPA
includes the 9.9 billion gallon scenario (which was covered in the emissions
inventory assessment) as well as those western regions of the US which are likely
to see expanded ethanol usage.

Impacts on Fossil Fuel Consumption and Related Implications
The EPA makes extensive use of lifecycle assessment (LCA) modeling to
estimate the impacts of increased renewable fuel use. Specifically, the Preamble
to the proposed rule notes usage of the GREET model developed by Argonne
National Laboratory as well as an EPA model entitled ?FUEL-CO2? for the
estimation of lifecycle greenhouse gas emissions and fossil energy usage. With
respect to the latter, no information is provided on the ?FUEL-CO2? model either in
the draft RIA or on the EPA website, so it is impossible to comment on the
appropriateness of this tool for fuel providers. If - as the text of the Preamble
implies - the ?FUEL-CO2? model is to be used for regulatory purposes, then the
EPA should provide appropriate notice and opportunity for public input and
comment on the specific details of this tool.

Lifecycle Analysis of Alternative Fuels
The Preamble and draft RIA for the proposed program appropriately recognize that
there are numerous inputs, parameters and assumptions in lifecycle analysis
models that have both direct and indirect influence on model outputs. These
models are extensive and extremely complex. The EPA makes use of the
GREET model developed by Argonne National Laboratory as a means to estimate
the impacts of increased use of renewable fuels. However, as the EPA correctly
notes ?? a consensus on all the assumptions, including point estimates, that are
used as inputs into that model does not exist.? (This is evidenced by the ongoing
public debate over the magnitude of the energy balance associated with the
production and usage of ethanol versus gasoline.) Moreover, there are numerous
other lifecycle models that the EPA could have employed in its analysis. Two
such models include, for instance, those developed by Mark Delucchi at the
University of California, Davis, and Jason Hill et al at the University of Minnesota.

API conducted a review of a prior version of the GREET model that was developed
in conjunction with the EPA MOVES model. As a result of that review, we
concluded that there were a number of parameters and inputs in the model which
appeared to bias the model output in favor of the ethanol pathway. We have not
reviewed the current version of GEEET (version 1.7) used in the proposed
rulemaking. However, we urge the EPA to initiate the public dialogue on lifecycle
modeling that it references in the Preamble to the proposed rule (perhaps under
the auspices of the National Academy of Sciences). Such a dialogue should
include, for instance, discussion of the ?boundaries? of lifecycle models, i.e., how
the overall problem is defined. For example, the GREET model covers air
emissions, but does not address environmental issues like water use, discharges
to water bodies, or the related ecological impacts. Also, there is significant
uncertainty regarding the ability of such models to accurately characterize all of
the relevant and important factors that collectively contribute to net impacts /
emissions / energy use, and whether or not there is sufficient good data that can
be used as input to most or all of the variables.
Comments on Chapters VI, VII and IX

Chapter VI. Current and Projected Renewable Fuel Production and Use

Page 117, 3rd paragraph: ?Over the last 25 years, domestic fuel ethanol
production has steadily increased due to technological advances, environmental
regulation, and the rising cost of crude oil.?

Comment: Statement is misleading, the cost of crude oil has increased and
decreased over the last 25 years. It would be more accurate to state that
domestic fuel ethanol production has steadily increased over the last 25 years due
to environmental regulations, federal and state subsidies and mandates, and
market demand.

Page 119, 1st paragraph: ?EPA forecasts ethanol production to continue to grow
into the future. In addition to the past impacts of federal and state tax incentives,
as well as the more recent impacts of state ethanol mandates and the removal of
MTBE from all U.S. gasoline, record-high crude oil prices are expected to
continue to drive up demand for ethanol.?

Comment: The statement ?record-high crude oil prices are expected to continue to
drive up demand for ethanol? is pure speculation and should be removed.
Expected by whom? Current crude prices are below $60 per barrel.

Page 121, 1st paragraph: ?However ethanol production is not expected to stop
here ? If all these plants come to fruition, the combined domestic ethanol
production could exceed 20 billion gallons as shown in Table VI.A.2-2?.

Comment: Pure speculation. Many of the proposed ethanol plants will never be
built which begs the question of the need for Table VI.A.2-2.

Chapter VII. Impacts on Cost of Renewable Fuels and Gasoline

Page 134, 3rd paragraph: ?We have estimated an average corn ethanol production
cost of $1.20 per gallon in 2012 (2004 dollars) in the case of 7.5 billion gallons per
year and $1.26 per gallon in the case of 9.9 billion gallons per year. For cellulosic
ethanol, we estimate it will cost approximately $1.65 in 2012 (2004 dollars) to
produce a gallon of ethanol using corn stover as a cellulosic feedstock.?

Comment: These corn ethanol production costs ($1.20 - $1.26 per gallon) seem
low, typical estimates seem to be roughly $1.35 to $1.50 per gallon with corn at
$2.25 per bushel (this is roughly the corn price used in your analysis), even
accounting for DDG sale credits. You also assume only a 3.6% increase in corn
prices between the 7.2 billion gallon use case and the 9.6 billion gallon use
cases. Given the implied increase in demand for corn between these 2 cases, it
is likely that the corn price increase will be significantly above 3.6%.

Also, corn futures for December 2006 delivery are $3.44 per bushel. Suggest
rechecking your production cost estimates and some sensitivity analysis using
various corn price assumptions. Also, personnel at DOE indicate $2.26 per gallon
as an estimate of cellulosic ethanol production costs, significantly above your
cost estimate of $1.65 per gallon.

Page 135, 2nd paragraph. Comment: Your estimates of overall gasoline costs
given the fuel changes you assume (both with and without the subsidy) are low
given your underestimates of ethanol production costs, your underestimates of
biodiesel production costs and your underestimates of ethanol transportation
costs (see next comment below). You estimate production costs of soy-derived
biodiesel of $2.06 per gallon in 2004 and $1.89 per gallon in 2012. But with current
soy oil at roughly $2.00 per gallon, how can soy-derived biodiesel production cost
be $2.06? A better estimate would be $2.50 per gallon for soy-derived biodiesel.

Page 143, 1st paragraph: ??Therefore we believe that estimated freight costs for
ethanol of 9.2 cents per gallon adequately reflects the freight costs for biodiesel
for this analysis?.

Comment: Your estimated ethanol transportation costs of 9.2 cents per gallon are
low and should be adjusted. Check current rates, e.g., regular railcar movements
are roughly 17 ? 22 cents per gallon from Chicago to Philadelphia, and 15- 20
cents per gallon if shipped via unit train according to Jim Jordon and Associates.
Also, your assertion that ethanol transportation costs ?adequately reflects the
freight costs for biodiesel? is speculation with no basis in fact.

Chapter IX. Impacts on Fossil Fuel Consumption and Related Implications

Page 182, Displacement Indexes derived from Greet

Comment: The EPA uses Argonne?s GREET Model GHG emission reduction
estimates of 26% for corn-based ethanol compared with gasoline on an energy
equivalent basis, and a 53% GHG reduction for biodiesel compared with
conventional diesel. It should be noted that EPA seems to have chosen the upper
end of a range given by Argonne of 18 to 26% GHG emission reduction of corn-
based ethanol relative to conventional gasoline. These estimates are significantly
higher than those estimated by Hill et. al. (12% reduction for corn-based ethanol
and 41% reduction for biodiesel) and by Delucchi (virtually no reduction for corn-
based ethanol and a 50% increase for biodiesel) as seen in the table below. It is
not clear the extent to which the GREET model accounts for emissions from land
use changes associated with biofuels production but these factors are extensively
considered in the works by Delucchi and Hill et al.

A comparison of estimates across studies reveals significant differences in both
magnitude and sign of emissions reflecting both differing assumptions and
uncertainties in estimation. The Delucchi study finds virtually no improvement in
GHG emissions associated with corn-based ethanol relative to conventional
gasoline, and significantly worse GHG emissions of biodiesel compared with LS
diesel fuel. Hill et al. find modest GHG emission reductions associated with corn-
based ethanol though significant reductions in GHG emissions associated with
biodiesel are found. A primary factor behind the differences in these study
estimates is the assumption of ex-ante land use associated with the particular bio-
crop. The GREET model (developed at Argonne National Lab) shows greater
benefits associated with corn-based ethanol and biodiesel. This may reflect model
assumptions related to the energy output/input ratios of ethanol and fossil fuels
(ethanol assumed to have a greater ratio than gasoline which is clearly incorrect).
Finally, a report from the International Energy Agency (IEA) summarizes the
results from a dozen studies, many stemming from the 1990s.



EPA states on page 174 that ?we believe that GREET offers the most
comprehensive treatment of the transportation sector?. However, the Dulucchi
study also offers a comprehensive treatment of the transportation sector as well
as detailed emissions changes from changes in land use. Given the significant
differences in model results vis-?-vis GHG emissions, sole reliance on the GREET
model for estimates of displacement values of renewable fuels related to GHG and
petroleum should be avoided, and results from other relevant models taken into
account. In particular, Table IX.B.3-1 on page 182 should be amended to take
account of the results of other studies, and estimates of GHG and petroleum use
reductions due to increased renewables use should be re-estimated in this
chapter accounting for estimates from all relevant studies including Dulucchi..

Sources for Summary Table

Lifecycle Analyses of Biofuels, Draft Manuscript, May 2006, Mark A. Delucchi
Institute of Transportation Studies, University of California at Davis

Environmental, economic, and energetic costs and benefits of biodiesel and
ethanol biofuels, PNAS, vol. 103, no. 30, July 25, 2006 by Hill, Nelson, Tilman,
Polasky and Tiffanny.

Biofuels for Transport: An International Perspective
Released by the International Energy Agency

Attachments

 (1)

Comment attachment submitted by Alfonse Mannato, American Petroleum Institute (API)

View Attachment:

 
 
Comment Period Closed
Nov 12 2006, at 11:59 PM ET
ID: EPA-HQ-OAR-2005-0161-0185
Tracking Number: 801e0c3a

Document Information

Date Posted: Nov 13, 2006
RIN: Not Assigned
Show More Details