Proposed RFS Small Refinery
Exemption Decision
£% United States
Environmental Protect
Agency
-------
Proposed RFS Small Refinery
Exemption Decision
United States Environmental Protection Agency
United States
Environmental Protection
^1 Agency
EPA-420-D-21-001
December 2021
-------
Table of Contents
EXECUTIVE SUMMARY 1
I. Proposed Adjudication and Request for Comment 6
II. Background and Program Interpretation 8
A. RFS Program 8
B. Renewable Identification Number Credits 9
C. RFS Compliance and RIN Market Dynamics 10
D. History of SREs 12
III. EPA's Proposed Approach to Determining DEH When Evaluating SRE Petitions 16
IV. EPA Evaluation 17
A. Eligibility to Petition 17
1. Definition of Small Refinery 18
2. Initial Blanket Statutory Exemption 18
3. Changed Approach to Eligibility 19
4. Proposed Eligibility Determination for Two Petitioning Small Refineries 20
B. Compliance with Petition Requirements 20
C. DOE Consultation and EPA Consideration of the DOE Study and Other Economic
Factors 21
D. Hardship Must Be Caused by RFS Compliance 23
1. The CAA Requires That DEH Must Be Caused by RFS Compliance 23
2. DEH and RIN Cost Passthrough 26
3. EPA Responses to Small Refinery Arguments for Exemption 52
V. Conclusion 62
l
-------
EXECUTIVE SUMMARY
With this notice, EPA is seeking comment on a proposal to establish a change in our
statutory interpretation of the Clean Air Act (CAA or "the Act") small refinery provisions and,
applying this interpretation, to deny 65 pending petitions from small refineries seeking
extensions of their small refinery exemptions (SREs) from the Renewable Fuel Standard (RFS)
program. The petitions concern one or more compliance years between 2016 and 2021. This
proposed adjudication and EPA's accompanying change in its interpretation of the relevant CAA
provision are compelled by the 2020 ruling in Renewable Fuels Association, et al. v. EPA, 948
F.3d 1206 (10th Cir. 2020) (RFA), the Supreme Court's subsequent decision inHollyFrontier
Cheyenne Refining, LLC, et al. v. Renewable Fuels Ass 'n, et al., 114 S.Ct. 2172 (2021), EPA's
experience implementing the RFS program for more than a decade, and our exhaustive analysis
of how the RFS credit market functions. EPA here proposes to establish that the small refineries
with pending petitions before the Agency have failed to demonstrate the disproportionate
economic hardship (DEH) that the CAA requires for EPA to grant such exemptions, and
therefore the 65 pending SRE petitions should be denied.
The conclusions we rely on in presenting this proposal are: (1) Regardless of the
mechanism by which small refineries and other obligated parties comply with their RFS
obligations, the RFS compliance costs are the same for all obligated parties and thus no party
bears RFS compliance costs that are disproportionate relative to others' costs; (2) Obligated
parties, including small refineries, recover their compliance costs through the market price they
receive when they sell their fuel products and thus do not bear a hardship created by compliance
with the RFS program; and (3) With no disproportionality and no economic hardship, there can
be no disproportionate economic hardship pursuant to the statute.
In addition to seeking comment on these conclusions, EPA requests comment on general
as well as specific aspects of this proposed adjudication, including but not limited to the
following1:
The proposed conclusion that DEH must be caused by compliance with the RFS
program to warrant an exemption.
The proposed finding that the structure of the RFS program places a proportional
burden on all obligated parties based on their gasoline and diesel fuel production
volume.
The proposed finding that the structure of the RFS compliance system EPA put in
place, which enables the use of renewable fuel credits known as Renewable
Identification Numbers (RINs), provides equal access to all obligated parties to the
same means of compliance.
The proposed finding that the fuel and RIN markets are competitive and that RIN
costs per gallon are the same to all obligated parties regardless of their role in the
market, their size, or whether they acquire RINs through blending renewable fuel or
by purchasing RINs (credits) representing that volume of fuel.
The proposed finding that RIN costs are passed through to consumers.
1 EPA has published a Federal Register notice that announces this proposed action and established a public docket
to receive these comments (Docket ID No. EPA-HQ-OAR-2021-0566).
1
-------
The proposed conclusion that, since passed through to consumers, the compliance
costs of the RFS program do not impose economic harm to obligated parties and,
therefore, the RFS program does not cause DEH for small refineries or any obligated
party.
Additional data that may illustrate the relationship between RFS compliance costs
and the price of transportation fuel and blendstocks.
The proposed change in approach to SRE eligibility described in Section IV.
The proposed decision to deny all pending SRE petitions for the reasons described
herein.
CAA section 21 l(o)(9)(A)(i) exempted small refineries from the RFS program through
compliance year 2010. CAA section 21 l(o)(9)(B)(i) authorizes the Administrator to temporarily
extend those exemptions "for the reason of disproportionate economic hardship." Under CAA
section 21 l(o)(9)(B)(ii), in evaluating SRE petitions, EPA shall "consider the findings" of a
study conducted by the Secretary of Energy pursuant to CAA section 21 l(o)(9)(A)(2) and "other
economic factors" to decide "whether compliance with the [RFS] requirements . . . would
impose a disproportionate economic hardship on small refineries." The statute does not define
DEH and identifies no particular "economic factors" to be considered, and case law establishes
that EPA has substantial discretion in how it implements the SRE provision. Relying on this
discretion, EPA's approach to SREs has evolved since 2011, shaped at various times by
prevailing administration policies, litigation, and the increasing knowledge and real-world
market data acquired through implementation of the RFS program.
The impetus for the statutory interpretation we are proposing for comment (while also
seeking comment on the proposed SRE adjudications) is the RFA opinion issued by the U.S.
Court of Appeals for the Tenth Circuit. The court acknowledged EPA's shifting approaches to
SREs since 2011 and ruled that, for the SRE grants at issue before that court, EPA had exceeded
its statutory authority by extending temporary exemptions to those small refineries. In remanding
those actions, the court held that: (1) In granting exemptions based on economic factors
unrelated to compliance with the RFS program, EPA had exceeded its statutory authority to
exempt small refineries from their RFS obligations "for purpose of [DEH]" because the statute
authorizes EPA to extend exemptions only where RFS compliance costs are the cause of the
small refinery's hardship; and (2) EPA had acted arbitrarily and capriciously in granting
exemptions without explaining whether and how the subject SRE grants were consistent with
EPA's firmly established position that all parties subject to RFS obligations recover their
compliance costs through a feature of the market EPA identified as "RIN cost passthrough."
Following the Tenth Circuit Court ruling, small refinery petitioners began submitting to
EPA supplemental materials to support their requests for exemptions. Many petitioners argue
that they bear comparatively high RFS compliance costs because they must buy more expensive
RINs (the renewable fuel credits that are the currency of the RFS compliance market), compared
to large integrated refiners that can acquire cheaper RINs by blending renewable fuel. Petitioners
also argue that they are unable to increase the price they charge for the petroleum fuel2 they
2 Generally, when we refer to "fuel" or "petroleum fuel" in this proposal, we are including in that term petroleum
blendstocks. A "blendstock" is defined as "any liquid compound or mixture of compounds (not including fuel or
fuel additive) that is used or intended for use as a component of a fuel." 40 CFR 1090.80.
2
-------
produce to recover their higher compliance costs because they are "price takers" and do not
control market prices. As we present in this proposal, EPA has carefully evaluated these claims;
we find them not to be valid, and, in some cases, the supplemental submissions even provide
support to our prior analyses, which the Tenth Circuit cited in RFA and which we propose to
extend here.
Defeating the small refineries' claims at the threshold and at the same time fundamental
to EPA's analyses are two economic phenomena in the RFS fuels market. Contrary to the first
claim of the small refineries, we find that all RFS obligated parties, including small refineries,
bear the same cost to acquire RINs whether the RINs are purchased in the market or acquired by
blending renewable fuel. This occurs because parties that acquire RINs by blending renewable
fuel must discount the price of the renewable fuel feedstocks (e.g., the ethanol in E103) by the
current market value of the RIN when they sell renewable fuel. Thus, a refinery that blends
renewable fuel pays the market price for the RINs it acquires from blending by offering the
discount it must give purchasers when it sells the renewable fuel. We refer to this first market
phenomenon, where the market price of renewable fuel is discounted by the RIN price, as the
"RIN discount."
The second phenomenon is refineries' recovery of the costs they bear in acquiring RINs
in the prices they charge for the petroleum fuel they sell. Throughout this document we will refer
to this economic outcome as "RIN cost passthrough" because refineries' RFS compliance costs
are passed through in higher prices for the petroleum fuel (i.e., gasoline and diesel fuel) they
produce. While it is true that small refineries have no control over the market price for the
petroleum fuel they sell, the price that they receive in the market does reflect the full RFS
compliance cost that is the same for all parties per gallon of fuel, allowing these costs to be
passed through. Hence even as price takers, small refineries recover their RFS compliance costs
when they receive the market price for the fuel they sell. Together, the two distinct market
phenomena of the "RIN discount" and "RIN cost passthrough" reflect the cross-subsidy nature
inherent to the RFS program. The "RIN discount" lowers the market price of renewable fuel to
facilitate their sale while "RIN cost passthrough" raises the market price of petroleum fuel to
recover those costs.
These economic outcomes are distinct, but are often conflated, especially in connection
with blended fuels like E10 that include both elements (a discount in the price of the ethanol that
is blended to make the fuel while the cost of the petroleum portion of the fuel increases). To help
illustrate how these two distinct aspects of the fuels market operate for different market
participants, the figure below shows the inputs and outputs of the fuels market for merchant
refiners, blenders, and integrated refiners. In the following sections, we highlight which of these
two phenomena—the "RIN discount" or "RIN cost passthrough"—we are referring to as we
detail our evaluation of the information submitted by the small refinery petitioners.4
3 E10 is a finished fuel blend composed of 90% gasoline blendstock and 10% ethanol. The gasoline blendstock is
often referred to as BOB (blendstock for oxygenate blending).
4 A comprehensive discussion of the RIN market impacts on different obligated parties can be found in Section
IV.D.2.C.
3
-------
Figure ES-1: Simplified Illustration of Fuels Market Participants
Before issuing this proposal, however, EPA exempted small refineries from their RFS
obligations solely based on metrics that did not account for RIN cost passthrough. Thus, EPA did
not require any demonstration that the DEH experienced by the refinery was due to the RFS
program. Neither did EPA reconcile this reasoning with EPA's own finding that the costs of
RINs used for compliance with the RFS program are the same for all obligated parties and
passed through by all obligated parties to consumers (RIN cost passthrough).
Now, as briefly described above and based on the detailed analyses herein, EPA has
reached the following proposed conclusions: (1) Regardless of the mechanism by which small
refineries and other obligated parties comply with their RFS obligations, the RFS compliance
costs are the same for all obligated parties and thus no party bears RFS compliance costs that
disproportionate relative to others' costs; (2) Obligated parties, including small refineries,
recover their compliance costs through the market price they receive when they sell their fuel
products and thus do not bear a hardship created by compliance with the RFS program; and (3)
With no disproportionality and no economic hardship, there can be no DEH pursuant to the
statute. Further, because EPA is here proposing to alter its statutory interpretation of its CAA
authority to extend SREs only to small refineries whose claimed DEH is caused by the cost of
complying with the RFS program, and not by other factors, EPA is here proposing to deny all of
the pending SRE petitions currently before the Agency. This proposed action is supported by
extensive market data and analyses, including information provided to EPA through recent active
engagement with small refineries.
4
-------
We are departing from past practice regarding SRE decisions by inviting comment on
aproposed adjudication of the pending petitions, along with the analysis we provide to support
our proposal. The goal of this public notice and comment process is to achieve the broadest
possible dissemination of EPA's preliminary analysis, to gain public input on the analysis, and
to solicit any further data or evidence interested parties may be able to provide to inform the
final decision EPA must issue with respect to the pending SRE petitions. Commenters are
encouraged to provide input on any aspect of this new statutory interpretation and our proposed
adjudication of the pending SRE petitions, including but not limited to the RIN market analysis
and conclusions detailed herein. We particularly request information or data to bolster or refute
our analysis of the RIN discount and RIN cost passthrough principles and our proposed decision
to deny the pending SRE petitions based on that analysis. This notice is not a final agency
action. EPA is requesting comment on general as well as specific aspects of this proposal.
5
-------
I. Proposed Adjudication and Request for Comment
EPA is proposing to require that small refineries: (1) Demonstrate that any DEH they
experience is caused by compliance with the RFS program; and (2) Reconcile any such showing
with RIN cost passthrough. This approach is described in more detail in Section III. The RIN
cost passthrough phenomenon is explained in Section IV.D.2. EPA is also proposing to change
its criteria for assessing a refinery's eligibility to receive an exemption from its RFS obligations.
We are proposing to now require a small refinery to have received the original statutory
exemption in order to be eligible to petition for an extension of that exemption, though a small
refinery need not have had continuous exemptions since the original statutory exemption,
consistent with the Supreme Court's holding in HollyFrontier5 and as explained in Section IV.A.
EPA currently has 65 pending petitions from small refineries seeking SREs pursuant to
CAA section 21 l(o)(9)(B) for one or more compliance years between 2016 and 2021.6 EPA has
completed its initial analysis of these petitions. After applying the new approach to DEH, and for
the reasons described in this document, EPA is proposing to deny all of the pending SRE
petitions as follows:
Deny 1 SRE petition for the 2016 compliance year.
Deny 1 SRE petition for the 2017 compliance year.
Deny 3 SRE petitions for the 2018 compliance year.
Deny 29 SRE petitions for the 2019 compliance year.
Deny 28 SRE petitions for the 2020 compliance year.
Deny 3 SRE petition for the 2021 compliance year.
In addition to proposing to deny all pending SRE petitions, EPA is proposing to make the
following eligibility determinations for two refineries with pending petitions:7
Determine that one refinery is ineligible to petition for an exemption for the 2019 and
2020 compliance years because it exceeded the crude oil throughput limit of 75,000
barrels per day in 2019, thereby making the refinery ineligible for an exemption in both
2019 and 2020 pursuant to applicable EPA regulations.8
- Determine that two refineries (including the one identified in the previous bullet) are
ineligible to petition for an exemption for the 2019 and 2020 compliance years (and
future compliance years) because they did not receive the original statutory exemption.
5 See HollyFrontier Cheyenne Refining, LLC, et al. v. Renewable FuelsAss'n, etal., 114 S.Ct. 2172, 2181 (2021)
(HollyFrontier).
6 Should we receive additional petitions for SREs subsequent to the release of this proposed action, we may decide
to include those petitions in our final action, if appropriate.
7 Though we are proposing to determine that these refineries are ineligible to petition for an exemption, we have also
included them in the 65 pending petitions that would be denied by this proposed action because we are, in the
alternative, also proposing to deny those petitions for the reasons described herein whether or not the small
refineries are found to be eligible to petition for an exemption.
8 40 CFR 80.1401 & 80.1441(e)(2)(iii).
6
-------
EPA requests comment on general as well as specific aspects of this proposed action in a
notice published in the Federal Register, including but not limited to the following:
The proposed conclusion that DEH must be caused by compliance with the RFS
program to warrant an exemption.
The proposed finding that the structure of the RFS program places a proportional
burden on all obligated parties based on their gasoline and diesel fuel production
volume.
The proposed finding that the structure of the RIN system put in place by EPA
provides equal access to all obligated parties to the same means of compliance.
The proposed finding that the fuel and RIN markets are competitive and that RIN
costs per gallon are the same to all obligated parties regardless of their role in the
market, their size, or whether they acquire RINs through blending or by purchasing
RINs.
The proposed finding that RIN costs are passed through to consumers.
The proposed conclusion that, since passed through to consumers, the compliance
costs of the RFS program do not impose economic harm to obligated parties and,
therefore, the RFS program does not cause DEH for small refineries or any obligated
party.
Additional data that may illustrate the relationship between RFS compliance costs
and the price of transportation fuel and blendstocks.
The proposed change in approach to SRE eligibility described in Section IV.
The proposed decision to deny all pending SRE petitions for the reasons described
herein.
In sum, EPA requests comment on all of these proposals and especially solicits any
additional information that either supports or refutes the Agency's findings on RIN discount
and/or on RIN cost passthrough. EPA in particular seeks information on the effects of these
phenomena on diverse small refineries and on the extent to which small refineries suffer DEH as
a result of compliance with their obligations. To be clear, this proposal seeks comment from the
general public as well as comment and additional information and analysis from the individual
petitioning refineries that supports or refutes these propositions as they apply to their respective
petitions.
This notice is not a final agency action for purposes of CAA section 307(b)(1); it is not a
rulemaking and is not subject to the various statutory and other provisions applicable to a
rulemaking. Instead, this is a proposed adjudication of the 65 pending SRE petitions by: (1)
Clearly articulating EPA's current interpretation of its statutory authority to grant SREs; (2)
Presenting our analysis of all available data on RFS costs and market dynamics for public review
and comment; and (3) Proposing to deny all of the pending petitions based on the current
statutory interpretation and analysis described herein in a single action. EPA intends to take final
action on the pending petitions based on the legal and factual analysis proposed herein, after
considering the public comments submitted in response to this proposed adjudication.
This document is structured to provide a sequential explanation of EPA's new approach
to SRE petition evaluation and of the data we analyzed to support this approach. It begins, in
7
-------
Section II, by providing background on the RFS program, compliance with the RFS program,
and the SRE provision of that program. Section II also provides a brief history of EPA's
approach to evaluating SRE petitions and judicial review of EPA's SRE decisions. Section III
presents the statutory requirements for EPA's evaluation of SRE petitions and EPA's new
approach to SRE evaluation. Section IV provides EPA's analysis of the SRE eligibility and
petition requirements and statutory construction of the CAA's SRE provision. It also presents a
detailed explanation of RFS market economics including the costs of RFS compliance on
obligated parties, and the implications of those costs on DEH. Section IV also includes a
description of how EPA satisfied the statutory requirements for this proposed action,9 then
summarizes and responds to the arguments advanced by the petitioning small refineries to date as
to how and why RFS compliance has caused their DEH. Section V provides EPA's proposed
conclusions based on all the information presented herein.
II. Background and Program Interpretation
A. RFS Program
In 2005 and 2007, Congress amended the CAA to establish the RFS program.10 Congress
enacted this program to "move the United States toward greater energy independence and
security" and to "increase the production of clean renewable fuels," among other purposes.11 The
statute specifies increasing annual "applicable volumes" for four categories of renewable fuel for
the transportation sector: total renewable fuel, advanced biofuel, cellulosic biofuel, and biomass-
based diesel.12 The specified applicable volumes for renewable fuel, advanced biofuel, and
cellulosic biofuel are prescribed for each year through 2022, and for biomass-based diesel
through 2012; EPA must determine the applicable volumes for subsequent years.13
Congress directed EPA to establish a compliance program and annual percentage
standards to ensure that the applicable volumes are used each year.14 To calculate these
percentage standards, EPA divides the applicable volume for each type of renewable fuel
established in the CAA or determined by EPA15 by the Energy Information Administration's
estimate of the national volume of transportation fuel that will be introduced into commerce in
that year.16 For example, if EPA set the percentage standard for total renewable fuel at 10 percent,
an obligated party that produced 1,000,000 gallons of gasoline one year would need to ensure that
100,000 gallons of renewable fuel was introduced into the market that year.
9 CAA section 21 l(o)(9)(B)(ii) requires EPA to: (1) Consult with the Secretary of Energy; and (2) Consider the
findings of the DOE study performed under CAA section 21 l(o)(9)(A)(ii)(I) and "other economic factors."
10 See Energy Policy Act of 2005 (EPAct), Pub. L. No. 109-58, 119 Stat. 594; Energy Independence and Security
Act of 2007 (EISA), Pub. L. No. 110-140, 121 Stat. 1492
11 121 Stat. 1492.
12 CAA section 21 l(o)(2)(B)(i)(I)-(IV).
13 Id.
14Id.; CAA section 211(o)(2)(A)(i), (iii) & (3)(B)(i).
15 CAA section 21 l(o)(2)(B), (7)(A), & (7)(D)-(F).
16 CAA section 21 l(o)(3)(A).
8
-------
Congress authorized EPA to place the obligation to satisfy the applicable percentage
standards on "refineries, blenders, and importers, as appropriate."17 By regulation, EPA
determined that refineries and importers of gasoline and diesel fuel must fulfill the requirements
of the RFS program.18 These "obligated parties" apply the percentage standards to their own
annual production or importation of gasoline and diesel fuel to calculate their individual
renewable volume obligation (RVO or "RFS obligation") for each type of renewable fuel. Thus,
the RFS standards absent SREs place the same obligation on all producers and importers of
gasoline and diesel fuel in proportion to their production volume.
B. Renewable Identification Number Credits
The CAA requires EPA to establish a credit trading program allowing obligated parties
that acquire excess credits in one year to apply credits toward compliance in a subsequent year or
to sell the credits to another obligated party for use in its own compliance.19 In conjunction with
EPA's authority under CAA section 21 l(o)(2)(B) to put in place implementing regulations for
the RFS, and in compliance with CAA section 21 l(o)(5), we designed a much more flexible and
comprehensive system of tradable credits (RINs). Section 21 1 (a)(5) required only that EPA
allow for the generation and trading of credits for obligated parties who refine, blend, or import
excess renewable fuel. The RIN system fulfills that statutory provision, and also creates a
fungible system of credit trading by not just obligated parties but also renewable fuel producers
and others, creating an open, liquid market for RINs to allow obligated parties to comply with
their RFS obligations. Under the RIN system, producers and importers of renewable fuel
generate RINs for each gallon of renewable fuel they import or produce for use in the United
States.20 RINs are "assigned" to batches of renewable fuel by the producers and importers of
renewable fuel.21 RINs may be "separated" from those batches by a party that blends the
renewable fuel into conventional fuel.22 Once separated, RINs may be kept for compliance or
sold.23 A RIN may be used to demonstrate compliance during the calendar year in which it is
generated, or the following calendar year (for up to 20% of an obligated party's RVO).
Thereafter, the RIN is considered to be "expired" and cannot be used for compliance purposes.24
Obligated parties meet their RFS obligations by accumulating RINs and "retiring" them in an
annual compliance demonstration.25 The statute and regulations also provide that, in lieu of
retiring the requisite number of RINs to show compliance for a particular year, an obligated party
may choose to carry a RIN deficit into the following year under certain conditions.26 An
obligated party may carry forward a RIN deficit equal to its full or partial RFS obligation.
17 CAA section 21 l(o)(3)(B)(ii)(I).
18 40 CFR 80.1406. For simplicity this document focuses on refiners; however, the same concepts of RIN costs, RIN
cost passthrough, and RIN discount for blended fuel also apply to importers.
19 CAA section 21 l(o)(5)(A)-(C).
20 40 CFR 80.1426(a).
21 40 CFR 80.1426(e).
22 40 CFR 80.1429(b).
23 40 CFR 80.1425-29.
24 40 CFR 80.1427(a)(6), 80.1428(c), & 80.1431(a).
25 40 CFR 80.1427(b).
26 CAA section 21 l(o)(5)(D), 40 CFR 80.1427(b).
9
-------
The price of the RIN is expected to reflect the marginal difference between the supply
price for the renewable fuel and the demand price for the renewable fuel, which is the price the
market is willing to pay for the renewable fuel as a transportation fuel.27 In other words, if it
costs more to produce the renewable fuel than consumers are willing to pay for the renewable
fuel, the RIN price would be expected to match that cost difference so that, in the end, the fuel
price for consumers is the same. The price of the RIN, therefore, allows the "discount" on the
renewable fuel necessary to have the market consume the renewable fuel. This dynamic
functions to incentivize blending and use of renewable fuel up to the mandated volume even if
the market demand price for the renewable fuel would not cover the cost of its production. In this
way, the RIN price facilitates greater use of renewable fuel as the RFS program was designed to
do.
The design of the RIN trading system enabled parties that were already producing and
blending renewable fuel to continue to do so. They could then sell excess RINs to obligated
parties that lacked blending capability. This open trading market for RINs provides three main
benefits. First, it allows all obligated parties, regardless of size or situation, equal ability to
comply with their RFS obligations immediately without having to invest capital or resources.
They can contract with others already providing the services and/or go into the open market to
acquire RINs. Second, this system averts the need for each individual obligated party to purchase
and blend renewable fuel into its own gasoline and diesel fuel.28 Thus, the program was designed
to "preserve^ existing business practices for the production, distribution, and use of both
[petroleum] and renewable fuel."29 Third, it levels the playing field for the cost of compliance,
with all obligated parties having access to the RINs needed for compliance at the same cost,
regardless of whether they acquire the needed RINs by purchasing them on the open market or
by blending renewable fuel themselves. The RFS program, through the RIN system, was
designed to avoid creating DEH based on whether compliance is obtained through blending of
renewable fuel or through purchasing of RINs.
C. RFS Compliance and RIN Market Dynamics
Congress structured the RFS program to impose proportional requirements on all
obligated parties, including small refineries. The RFS obligations are established as a percentage
of an obligated party's production;30 therefore, by definition, the obligation is proportional to the
quantity of gasoline and diesel fuel that party produces or imports each year.31 Obligated parties
must acquire RINs to meet their RFS obligations,32 either through their own blending of
27 "A Preliminary Assessment of RIN Market Dynamics, RIN Prices, and Their Effect," Dallas Burkholder, Office
of Transportation and Air Quality, US EPA, May 14, 2015 [hereinafter, "the Burkholder memo"].
28 Complying with such a requirement would have been difficult, if not impractical for obligated parties, as different
renewable fuels are blended into gasoline and diesel fuel and pipeline operators normally do not allow gasoline or
diesel fuel containing renewable fuel to be transported through their pipelines.
29 "RFS1 Summary and Analysis of Comments," EPA-420-R-07-006 at 1-6, April 2007.
30 See supra, Sections II. A and B.
31 See CAA section 211(oX3)(B); 40 CFR 80.1407.
32 For purposes of the RFS program, transportation fuel is defined as "fuel for use in motor vehicles, motor vehicle
engines, nonroad vehicles, or nonroad engines (except fuel for use in ocean-going vessels)." 40 CFR 80.1401. The
regulations at 40 CFR 80.1406 establish that "[a]n obligated party is any refiner that produces gasoline or diesel fuel
10
-------
renewable fuel or through the purchase of RINs from other parties that produce or blend
renewable fuel. Obligated parties must demonstrate compliance annually by retiring RINs
requisite with their RFS obligations.
The cost of acquiring RINs is the same for all parties regardless of whether the RINs
needed to comply are acquired by blending renewable fuel or by procuring RINs from others.33
This occurs through the phenomena of RIN discount and RIN cost passthrough, introduced in the
Executive Summary and explained in detail throughout this document. Parties that blend more
renewable fuel than they need to satisfy their RFS obligations may show an apparent revenue
source from the sale of those RINs. However, in the competitive fuels market, price competition
prevents these parties from retaining the revenue from their RIN sales.34 If parties that blend
renewable fuel do not discount the price of their blended fuel by the market price of the RIN,
then their blended fuel would be priced higher than the same fuel where the producer has
discounted the fuel by the price of the RIN, and the non-discounted fuel would never sell.
Therefore, in order to price their products competitively in the fuels market, parties that blend
renewable fuel must reduce the price of their blended fuel by the price of the RIN (RIN
discount). Thus, the revenue from the RIN sale is used to offset the discounted sales price of the
blended fuel and is passed through to consumers through reduced market prices for the blended
fuels. Moreover, the RFS program imposes the same cost on all parties that produce or import
gasoline or diesel fuel nationwide,35 so the market price for gasoline and diesel fuel increases to
reflect this cost, much as it would increase in response to a new tax (RIN cost passthrough). This
relationship between RIN prices and the market prices for blended fuels was first analyzed by
EPA in 2015.36
In this document we refer to an obligated party's ability to recover the cost of the RINs it
acquires for compliance as "RIN cost passthrough," since obligated parties are passing these
costs through to consumers. We refer to the lower prices received for blended fuel (e.g., gasoline
and diesel fuel blended with renewable fuel) enabled by the sale of RINs as "RIN discount,"
since the sale of the RIN allows blenders to discount the price of the blended fuel. We find that
all types of obligated parties have the same cost to acquire RINs, and that all types of obligated
parties recover these costs when they sell the gasoline and diesel fuel they produce or import at
the market price (RIN cost passthrough). Further, we find that blenders use revenue from RIN
within the 48 contiguous states or Hawaii, or any importer that imports gasoline or diesel fuel into the 48 contiguous
states or Hawaii during a compliance period." The regulations at 40 CFR 80.1407 establish that, in practice, an RFS
obligation is imposed only on gasoline and ultra-low sulfur diesel (ULSD) used in motor vehicles, nonroad engines,
locomotives, and marine engines (historically called MVNRLM diesel fuel). Such gasoline and diesel fuel only
incur an obligation if used in the RFS covered location as defined in 40 CFR 80.1401. Throughout this document we
refer to fuel that incurs an RFS obligation (e.g., gasoline, diesel fuel) as "obligated fuel" and fuel that does not incur
an RFS obligation (e.g., heating oil, jet fuel) as "non-obligated fuel."
33 See infra, Section IV.D.2.
34 "A Preliminary Assessment of RIN Market Dynamics, RIN Prices, and Their Effect," Dallas Burkholder, Office
of Transportation and Air Quality, US EPA, May 14, 2015.
35 In this document, the term "nationwide" refers to the RFS "covered location," which the regulations at 40 CFR
80.1401 define as "the contiguous 48 states of the United States, Hawaii, and any state or territory that has received
an approval from the Administrator to opt-in to the RFS program under §80.1443."
36 "A Preliminary Assessment of RIN Market Dynamics, RIN Prices, and Their Effect," Dallas Burkholder, Office
of Transportation and Air Quality, US EPA, May 14, 2015.
11
-------
sales to discount the price of blended fuel (RIN discount). We therefore conclude that
compliance with the RFS program cannot cause a DEH for small refineries.37
I). History of SREs
A small refinery is defined by the statute as "a refinery for which the average aggregate
daily crude oil throughput for a calendar year . . . does not exceed 75,000 barrels."38 Both the
original RFS statutory provisions enacted pursuant to Energy Policy Act (EPAct), and the current
text of the statute as amended by the Energy Independence and Security Act (EISA), provided all
small refineries an initial blanket exemption from their obligations under the RFS program until
calendar year 2011.39 Under EPA's regulations, exempt small refineries were required to notify
EPA that they qualified for the temporary exemption by submitting verification letters stating
their average crude oil throughput rate during the applicable qualification period.40 Furthermore,
EPA's regulations allowed for small refineries that had submitted verification letters to qualify
for the original statutory exemption under EPAct to also qualify under the SRE provision in
EISA. The small refineries were not required to re-certify their throughput to maintain eligibility
under the amended RFS program. Further discussion of EPA's past and current interpretation of
small refinery eligibility criteria is provided in Section IV. A.41
The CAA includes two additional provisions regarding extensions of the small refinery
exemption for the period after the initial blanket exemption expired:
1) Under the first statutory mechanism, applicable to 2011 and 2012, if DOE determined,
through a study mandated under the CAA, that compliance with the RFS requirements
would impose DEH on a small refinery, EPA was required to extend the small refinery's
exemption by at least two years.42 In 2009, DOE completed its study and found that, in a
liquid and competitive RIN market, compliance with the RFS requirements would not
impose DEH on any small refinery. Subsequently, some members of Congress directed
DOE to revisit the 2009 DOE Small Refinery Study (hereinafter "the 2009 DOE Study")
and in so doing to solicit input from the small refineries themselves.43 In 2011, DOE
completed a second study that used the small refinery input to develop a set of financial
and operational metrics intended to inform DOE whether a small refinery was likely to
experience DEH. Contrary to the 2009 DOE Study, the 2011 DOE Small Refinery Study
(hereinafter "the 2011 DOE Study")44 did not assume that RFS compliance costs would
be the same for all refineries in a competitive market, and instead, assumed that small
37 The economic theory supporting EPA's findings on RIN cost passthrough and the RIN discount, the market data
we have evaluated in reaching these findings, and more detailed explanations on how various parties in the fuels
market are affected by the RFS program are discussed in greater detail in Section IV.D.2.
38 CAA section 21 l(o)(l)(K).
39 CAA section 21 l(o)(9)(A)(i).
40 40 CFR 80.1441(b).
41 40 CFR 80.1441(e)(2)(i); EPA, Financial and Other Information to be Submitted with 2016 RFS Small Refinery
Hardship Exemption Requests, December 6, 2016.
42 CAA section 21 l(o)(9)(A)(ii)(II).
43 Senate Report 111-45, at 109 (2009).
44 "Small Refinery Exemption Study, An Investigation into Disproportionate Economic Hardship," Office of Policy
and International Affairs, U.S. Department of Energy, March 2011.
12
-------
refineries could face higher compliance costs by purchasing RINs when compared to
large integrated refiners that would acquire RINs through blending. Furthermore, neither
study considered the possibility that refineries would recover the cost of RINs through
higher prices for their products.45 DOE organized the metrics into a two-part matrix with
sections addressing "disproportionate impacts" and "viability impairment."46 DOE also
developed a scoring protocol for the matrix that required the score in both sections of the
matrix to exceed an established threshold for DOE to find that DEH existed at a given
refinery. Using this regime, the 2011 DOE Study found that DEH existed at 14 small
refineries, but again, assumed the refineries did not recover RIN costs and did not address
the cause of the hardship. As required by the statute, EPA granted those small refineries a
two-year extension of the original exemption (through 2012).
2) The second statutory mechanism provided that small refineries "may at any time petition
the Administrator for an extension of the exemption in [section 21 l(o)(9)(A)] for the
reason of [DEH]."47 The Supreme Court recently opined on the meaning of "extension"
in the context of CAA section 21 l(o)(9)(B), overturning a holding by the U.S. Court of
Appeals for the Tenth Circuit that required a small refinery to have continuous
exemptions to be eligible for further exemption extensions.48 The Act directs EPA to
evaluate such petitions "in consultation with the Secretary of Energy[.]"49 EPA is also to
"consider the findings of the study under [CAA section 21 l(o)(9)(A)(ii)(I)] and other
economic factors."50 After DOE conducted its 2011 Study and EPA granted two-year
extensions to the 14 refineries the study identified, additional refineries came forward to
EPA to seek exemptions for 2011 and 2012. EPA shared these new petitions with DOE,
which applied the matrix scoring methodology developed in the 2011 DOE Study and
shared the scoring results with EPA. EPA chose to satisfy the statutory requirements for
consultation and consideration of the DOE Study by using DOE's scoring results in its
evaluation of each SRE petition. Consistent with the extensions of exemptions it granted
to the 14 small refineries through the 2011 DOE Study, EPA then decided to grant an
extension of the exemption to an additional ten small refineries for 2011, and to nine for
2012. Since 2013, EPA has shared all incoming SRE petitions with DOE, and DOE has
continued to make recommendations to EPA based on its scoring matrix, which does not
assess the degree to which small refineries recover the RFS compliance cost in higher
prices for their refined products (i.e., it does not consider RIN cost passthrough).
In the years since 2013, DOE and EPA have changed their treatment of the scoring
matrix several times as informed by direction from members of Congress, judicial review, and
changing administration policies. For DOE, the most significant change in approach did not
involve the matrix evaluation or the scoring methodology, but rather modified the
recommendation DOE provided to EPA for a given score on the matrix (i.e., DOE implemented
the direction from Congressional report language to recommend 50% exemptions, as opposed to
45 See infra, Section IV.D.
46 2011 DOE Study at 32-36.
47 CAA section 21 l(o)(9)(B)(i).
48 See HollyFrontier, 114 S.Ct. at 2181. Consistent with that decision, small refineries that have not received
continuous exemptions remain eligible to petition for future exemptions.
49 CAA section 21 l(o)(9)(B)(ii).
50 Id.
13
-------
exclusively 0% or 100%, as described further below). For EPA, the changes involved the weight
EPA afforded DOE's recommendation relative to the "other factors" EPA considered when
evaluating SRE petitions. However, in none of these years did EPA require small refineries to
demonstrate that they faced RFS compliance costs that were higher than for other obligated
parties (i.e., disproportionate), nor did EPA require a demonstration that the hardship was caused
by compliance with the RFS program, including an explanation for how compliance costs
harmed them in a market characterized by RIN cost passthrough.
In some prior decisions, DOE and EPA concluded that DEH existed only when a small
refinery experienced both disproportionate impacts and viability impairment, as measured by the
matrix. In response to concerns that the two agencies' threshold for establishing DEH was too
stringent, Consolidated Appropriations Act report language directed DOE to recommend 50%
relief when a small refinery's score on either section of the matrix exceeded the applicable
threshold.51 Subsequent Senate Report language directed EPA to follow DOE's
recommendation, and to report to Congress if it did not.52
The Congressional direction, along with changing administration policies, prompted EPA
to change its approach to finding DEH at a small refinery. Whereas EPA had previously
exercised discretion in evaluating "other economic factors" in its analysis of a small refinery's
petition, EPA changed its approach to rely on DOE's recommendation and began granting a full
exemption whenever DOE findings indicated that the small refinery could receive at least 50%
relief, based on its matrix score.53 Under this approach, EPA exempted small refineries from
their RFS obligations solely based on this DOE finding, which was based on metrics that did not
account for RIN cost passthrough. Thus, neither EPA nor DOE required any demonstration that
the DEH experienced by the small refinery was due to the RFS program. Nor did EPA reconcile
this reasoning with EPA's own finding that the costs of RINs used for compliance with the RFS
program are the same for all obligated parties and passed through by all obligated parties to
consumers (RIN cost passthrough).
EPA's approach to evaluating SRE petitions has been challenged several times by small
refineries and other parties in different U.S. Courts of Appeals, as well as in the Supreme Court
51 Consolidated Appropriations Act, 2016, Pub. L. No. 114-113 (2015). The Explanatory Statement is available at
161 Cong. Rec. H9693, H10105 (daily ed. Dec. 17, 2015): "If the Secretary finds that either of these two
components exists, the Secretary is directed to recommend to the EPA Administrator a 50 percent waiver of RFS
requirements for the petitioner."
52 Senate Report 114-281, 71 ("When making decisions about small refinery exemptions under the RFS program,
the Agency is directed to follow DOE's recommendations which are to be based on the original 2011 Small
Refinery Exemption Study prepared for Congress and the conference report to division D of the Consolidated
Appropriations Act of 2016. Should the Administrator disagree with a waiver recommendation from the Secretary of
Energy, either to approve or deny, the Agency shall provide a report to the Committee on Appropriations and to the
Secretary of Energy that explains the Agency position. Such report shall be provided 10 days prior to issuing a
decision on a waiver petition.").
53 We note that under this approach, EPA granted full SREs to some very profitable refineries. A substantial number
of small refineries that showed no viability impairment on the matrix received a 50% waiver finding from DOE,
based only on the small refinery's disproportionate impacts score.
14
-------
of the United States.54 The approach to evaluating DEH we propose in this action is informed by
the outcome of litigation filed in the U.S. Court of Appeals for the Tenth Circuit Court in 2018.
Biofuels groups led by the Renewable Fuels Association (RFA) challenged EPA's actions in
granting three individual SREs, and the affected small refineries intervened on EPA's behalf.55
The court vacated and remanded EPA's actions for three reasons. First, under the Tenth Circuit's
reading of the CAA, a small refinery would be eligible for SRE relief only if it has received
extensions of the initial exemption in every year since 2010.56 Second, the court found that EPA
may grant relief only when it finds that the small refinery would suffer DEH due to compliance
with the RFS program and not due, even in part, to other factors.57 Third, the court held that EPA
had acted arbitrarily and capriciously by failing to explain how granting the exemptions was
consistent with the Agency's longstanding findings on RIN cost passthrough.58
After the Tenth Circuit's RFA opinion, the small refinery intervenors petitioned the
Supreme Court for a writ of certiorari, appealing only the Tenth Circuit's holding that, in order to
be eligible for exemption, a small refinery needed a continuous, uninterrupted exemption
history.59 The Supreme Court granted the writ of certiorari and reviewed the Tenth Circuit's
holding. EPA, which changed its prior litigation position, and RFA, filed briefs in opposition,
arguing that the Court should uphold the Tenth Circuit's ruling. On June 25, 2021, the Supreme
Court held that the term "extension" as used in CAA section 21 l(o)(9)(B) does not include a
continuity requirement and reversed the Tenth Circuit opinion only on that issue.60 The Supreme
Court did not review the other two holdings in RFA as those were not appealed by the small
refineries.
Since the Supreme Court issued its opinion in the HollyFrontier case, EPA has met with
several of the petitioning small refineries in individual meetings, accepted additional
supplemental information from the small refineries, made all petitioning small refineries aware
that they have the opportunity to submit additional information to EPA for consideration, and
conducted an open meeting among the small refineries, inviting them to participate and provide
feedback. In this proposed action, we are again soliciting information from all interested parties
to inform our final analysis and decision, especially seeking additional information that would
support or refute the proposed finding that small refineries do not face disproportionate cost or
economic hardship caused by compliance with the RFS program. We are also soliciting
information demonstrating that the cost of compliance with the RFS program is the same for all
obligated parties and passed on to consumers and that small refineries are not experiencing DEH
caused by compliance with the RFS program.
54 Hermes Consol., LLC v. EPA, 787 F.3d 568 (D.C. Cir. 2015) (reversing EPA's action due to errors EPA admitted
in calculating the small refinery's net income and net refining margins); Lion Oil Co. v. EPA, 792 F.3d 978 (8th Cir.
2015); Sinclair Wyoming Refining Co. v. EPA, 887 F.3d 986 (10th Cir. 2017); Ergon-West Virginia, Lnc. v. EPA,
896 F.3d 600 (4th Cir. 2019) (EWV-I); Ergon-West Virginia, Lnc. v. EPA, 980 F.3d 403 (4th Cir. 2020) (EWV-LL);
Renewable Fuels Ass 'n, et al. v. EPA, 948 F.3d 1206 (10th Cir. 2020) (RFA).
55 RFA, 948 F.3d 1206 (10th Cir. 2020).
56 RFA, 948 F.3d at 1244-49.
51 Id. at 1253-54.
58 Id.
59 Pet. for Writ of Certiorari at (i), HollyFrontier.
60HollyFrontier, 141 S.Ct. at 2183.
15
-------
III. EPA's Proposed Approach to Determining DEH When Evaluating SRE Petitions
Section 21 l(o)(9)(B)(i) of the CAA authorizes the EPA Administrator to temporarily
exempt small refineries from their RFS obligations for the reason of DEH. The statute directs
EPA, in consultation with DOE, to consider the DOE Study and "other economic factors" in
evaluating SRE petitions. The statute does not define "disproportionate economic hardship" and
identifies no particular "economic factors" to be considered, giving EPA "substantial discretion"
for purposes of implementing these exemption provisions.61 EPA, however, must interpret these
provisions in a reasonable manner, consistent with the purpose of the statutory provisions at
issue.
EPA's previous approach to interpreting these statutory provisions and evaluating SRE
petitions was that a small refinery could receive an exemption from its RFS obligations by
demonstrating it was experiencing DEH for any reason, including reasons unrelated to RFS
compliance.62 In this action, EPA is proposing to change its approach to require the small
refinery to demonstrate that compliance with the RFS program is the cause of the DEH
experienced by the small refinery. EPA has previously performed analyses, and reviewed
academic studies, on the RIN market that verify the passthrough of RFS compliance costs to
consumers. However, our prior approach to evaluating SRE petitions did not rely on a showing
that DEH was caused by RFS compliance because we concluded that our consideration of "other
economic factors" extended beyond economic factors addressing DEH caused by RFS
compliance. The Tenth Circuit in RFA clarified that EPA's prior approach was contrary to the
language of the CAA authorizing exemptions only due to DEH caused by compliance with the
requirements of the RFS program.63 Under the approach we propose here, the small refinery
must demonstrate a direct causal relationship between its RFS compliance costs and the DEH it
alleges; assertions regarding other real but unrelated financial difficulties the refinery may be
experiencing would not satisfy this requirement. Additionally, the small refinery must
demonstrate how its specific RFS compliance costs are disproportionate compared to other
refineries' RFS compliance costs and are of sufficient magnitude to warrant the exemption. EPA
has weighed several considerations in proposing this new approach and this interpretation is
consistent with the language of the Act, the purpose of the SRE provision, and is the most
reasonable approach for implementing the RFS program.64
Our change in approach is primarily informed by the RFA opinion, which laid out a
rationale for its conclusion that the statutory provisions require DEH to be caused by RFS
compliance.65 Additionally, the court in RFA held that EPA had acted arbitrarily and
capriciously when the Agency ignored the relevant evidence and granted three SREs without
addressing its long-standing position that RIN costs are passed through by refineries and
61 Hermes, 787 F.3d at 575 ("The statute gives no further instruction and identifies no particular economic factors or
metrics to be considered. That sort of statutory silence about the particular factors that an agency must consider
conveys 'nothing more than a refusal to tie the agency's hands' (internal citation omitted). As long as EPA consults
with DOE and considers the 2011 Study and 'other economic factors,' EPA retains substantial discretion to decide
how to evaluate hardship petitions.").
62 See supra, Section II.D.
63 RFA, 948 F.3d at 1253-54.
64 See infra, Section IV.D. 1.
65 Id.
16
-------
ultimately borne by consumers. After review of the court's decision, EPA agrees that these
holdings both reflect a reasonable interpretation of the Act and comport with EPA's longstanding
conclusions regarding RIN cost passthrough.66
Using this new approach, we evaluated the information and data available to us to assess
whether any of the petitioning small refineries demonstrated DEH. The very market-based
design of the RFS program with the RIN system for compliance has equalized the cost of
compliance among all market participants, making it highly unlikely any one refinery would face
a disproportionate cost of compliance.67 We have evaluated an extensive amount of data and
available literature, and our analysis shows the cost of RINs is the same whether refineries
acquire the RINs by blending renewable fuel or by buying RINs on the open market.68 The data
and available literature also informed our finding that the RFS compliance cost is passed through
in the price of refined products. Therefore, considering all of this information and analysis as
more fully explained in later sections of this document, we are proposing to find that no small
refinery experiences DEH due to their compliance with the RFS program.
When an agency changes its position, it must "provide a reasoned explanation for its
action" and "display awareness that it is changing position."69 Furthermore, EPA does not need
to show "that the reasons for the new policy are better than the reasons for the old one; it suffices
that the new policy is permissible under the statute, that there are good reasons for it, and that the
agency believes it to be better, which the conscious change of course adequately indicates."
(emphasis in the original).70 This proposed approach is reasonable as it is supported by the
language and construction of the CAA and data analyses performed by EPA and independent
parties.71 For the reasons described herein, EPA believes that this approach is the most logical
interpretation of the statutory provisions and the most reasonable way to implement the SRE
program.
IV. EPA Evaluation
A. Eligibility to Petition
In this action, we are proposing to find two refineries ineligible to petition for an SRE.
One refinery is ineligible because its throughput exceeded 75,000 bpd in the petitioning year and
the prior year and because it did not receive the initial exemption. The second refinery is
ineligible because the refinery did not receive the initial exemption.
In this action, we are proposing to revert to our prior approach to eligibility to petition for
an SRE on the basis of having received the initial exemption. We are proposing to read the
66 See infra, Section IV.D.2.
67 See supra, Section II.B.
68 See infra, Section IV.D.2.
69 FCC v. Fox Television Stations, Inc., 556 U.S. 502, 515 (2009).
70 Id. (emphasis in the original).
71 See infra, Section IV.D.
17
-------
statute such that only small refineries that received the initial blanket exemption are eligible to
petition for an extension of that initial exemption.72
L Definition of Small Refinery
As part of EPAct, Congress defined a small refinery as "a refinery for which the average
aggregate daily crude oil throughput for a calendar year (as determined by dividing the aggregate
throughput for the calendar year by the number of days in the calendar year) does not exceed
75,000 barrels."73 This definition was maintained in EISA.74 These definitions informed EPA's
implementing regulations in 2007 and 2010, which similarly defined a small refinery as
processing less than 75,000 bpd in 2004 and 2006, respectively, for purposes of determining
eligibility for the initial blanket statutory exemption.75 In 2014, EPA promulgated regulations
related to eligibility and requirements for SRE petitions.76 In these regulations, EPA modified
the eligibility requirements such that small refineries qualified for exemptions based on their
crude oil throughput for the petition year and the prior year.77 This means that, to qualify as a
small refinery, a refinery must have processed no more than 75,000 bpd of crude oil in both the
year for which the refinery requests an exemption and the prior year.78
2. Initial Blanket Statutory Exemption
In 2016, EPA deemed a refinery ineligible to petition for an exemption because the
refinery did not exist in 2006, and, thus, could not have received the initial blanket exemption.79
In that decision, EPA relied on the RFS regulations that state "a refiner may petition the
Administrator for an extension of its small refinery exemption...." (emphasis added).80
Additionally, EPA reasoned that "newer small refineries have the ability to consider whether
they believe the establishment of the RFS program and its requirements will cause economic
hardship before beginning operations." Beginning in 2017, EPA took a different approach to
small refinery eligibility and granted exemptions for refineries that did not receive the initial
exemption. We now propose to revert to the approach taken in 2016 - to be eligible to petition
for an SRE, a refinery must have received the initial blanket exemption from the RFS program,
72 We are maintaining our approach to size-based eligibility, such that only small refineries with an average
aggregate daily crude oil throughput that does not exceed 75,000 barrels per day (bpd) for the calendar year they
petition, and the prior year, are eligible to petition for an SRE. See CAA section 21 l(o)(l)(K), 40 CFR 80.1401, 40
CFR 80.1441(e)(2)(iii).
73 CAA section 21 l(o)(l)(K); EPAct of 2005, Pub. L. No. 109-58, 119 Stat. 594 (2005).
74 EISA of 2007, Pub. L. No. 110-140, 121 Stat. 1492 (2007).
75 40 CFR 80.1101, 80.1141(a)(1), 72 FR 23900 (May 1, 2007); 40 CFR 80.1401, 80.1441(a)(1), 75 FR 14670
(March 26, 2010).
76 79 FR 42128 (July 18, 2014).
77 40 CFR 80.1441(e)(2)(iii), "In order to qualify to an extension of its small refinery exemption, a refinery must
meet the definition of 'small refinery' in §80.1401 for the most recent full calendar year prior to seeking an
extension and must be projected to meet the definition of 'small refinery' in §80.1401 for the year oryears for which
an exemption is sought. Failure to meet the definition of small refinery for any calendar year for which an
exemption was granted would invalidate the exemption for that calendar year" (emphasis added). See also 79 FR
42128 (July 18, 2014).
78 40 CFR 80.1401. We are not proposing to modify this regulation in this action.
79 See Petition for Review, Dakota Prairie Refining, LLC v. EPA, No. 16-2692, at 8 of 17 (8th Cir. June 13, 2016).
80 40 CFR 80.1441(e)(2).
18
-------
though a small refinery need not have had a continuous exemption since the original statutory
exemption, consistent with the Supreme Court's holding in HollyFrontier.
3. Changed Approach to Eligibility
In this action, EPA is proposing to change its approach to SRE eligibility to require that a
petitioning small refinery receive the initial statutory exemption through 2010 in order to qualify
for an extension of the initial exemption under CAA section 21 l(o)(9)(B). We believe this policy
aligns with the text of the CAA, which describes a small refinery's right "to at any time petition
the Administrator for an extension of the exemption in subparagraph (A) for the reason of
[DEH]."81 Furthermore, we believe this policy aligns with policy interests of implementing the
RFS program. This is particularly true since the exemptions would provide a significant windfall
profit to the exempted small refinery.82 Additionally, refineries that exceeded the 75,000 bpd
throughput threshold in 2006 were not the intended recipients of the initial exemption for small
refineries, and, thus, we are acting consistent with congressional intent by continuing to exclude
these parties from receiving an SRE.
While the Supreme Court recently held that a small refinery need not have had a
continuous exemption since the initial exemption, the Court's decision implies that an exemption
must have existed at some point for it to be extended.83 Regarding the definition of "extension,"
the Court agreed with the Tenth Circuit that, as used in CAA section 21 l(o)(9), the word has a
temporal meaning (i.e., an extension of time), and not the alternative meaning of "extension" to
grant or offer.84 The Court, however, clarified that an extension may still be given after a lapse.85
The Court applied several analogies to illustrate this, including that of a student requesting an
extension of a deadline to submit a paper after the deadline has already passed.86 Applying that
analogy to a small refinery that did not receive the original exemption, but requests an extension
of that exemption, would be like a student that was never in the class asking the professor for an
extension of a deadline for a paper it was never assigned to begin with (i.e., there is no due date
for the professor to extend just as there is no exemption period for EPA to extend). Thus, the
language of the statute indicates that, without having received "the exemption under
81 CAA section 21 l(o)(9)(B)(i).
82 See infra, Section IV.D.2
83 See HollyFrontier, 141 S. Ct. at 2177 ("It is entirely natural—and consistent with ordinary usage—to seek an
"extension" of time even after some lapse."); id. at 2181 ("And fairly read, the key phrase at issue before us—'A
small refinery may at any time petition the Administrator for an extension of the exemption under subparagraph (A)
for the reason of disproportionate economic hardship'—simply does not contain the continuity requirement the court
of appeals supposed."); id. at 2184 (Barrett, J. dissenting) ("Yet, HollyFrontier insists, the term "extension" is not
always used that way. Instead, it might sometimes refer to a "non-continuous extension"—in other words, an
extension of something that used to exist but no longer does. . . . [T]he Court concludes that Holly-Frontier's reading
must be right—which means that EPA can provide an "extension" of an exemption that is no longer in effect.").
84 See supra, Section II.D.
85 HollyFrontier, 141 S.Ct. at 2177 ("Ultimately, however, we agree with the renewable fuel producers and the court
of appeals that subparagraph (B)(i) uses "extension" in its temporal sense—referring to the lengthening of a period
of time."). The HollyFrontier decision is further discussed in Section II.D.
86 Id. at 2177-78.
19
-------
subparagraph (A)," there is nothing for a small refinery to petition EPA to extend temporally.87
Ergo, if a small refinery did not receive the original statutory exemption, it is ineligible to have
EPA extend the duration of that exemption.88
4. Proposed Eligibility Determination for Two Petitioning Small Refineries.
EPA has received 2019 and 2020 SRE petitions from a refinery that exceeded the 75,000
bpd throughput limit in 2019, thereby making the refinery ineligible to petition for an SRE in
both 2019 and 2020.89 In this action, we propose to find this refinery ineligible to petition for an
SRE in 2019 and 2020. In the alternative, even if the refinery were eligible to petition, we have
reviewed the petitions and supplemental information supporting them and propose to find that
the refinery did not demonstrate that it experienced DEH caused by its RFS compliance as
described generally for all small refineries in Section IV.D.2.
EPA has received two 2019 and two 2020 SRE petitions from two refineries (including
the one identified in the previous paragraph) that would be ineligible under the proposed changed
approach described in Section IV.A.2 In this action, we propose to determine that these refineries
are ineligible to petition for an exemption because these two refineries did not receive the initial
exemption. In the alternative, even if the refineries were eligible to petition, EPA has reviewed
the petitions and supplemental information supporting them and is proposing to find the
refineries did not demonstrate that they experienced DEH caused by their RFS compliance
described generally for all small refineries in Section IV.D.2.
B. Compliance with Petition Requirements
When submitting an SRE petition to EPA, the small refinery bears the burden of
demonstrating that compliance with the requirements of the RFS program causes DEH for that
small refinery. EPA regulations require that an SRE petition specify the factors that demonstrate
a "disproportionate economic hardship," provide a detailed discussion regarding the hardship the
refinery would face in complying with the RFS requirements, and identify the date the refinery
anticipates that compliance with the RFS requirements can reasonably be achieved at the small
refinery.90 Since the Tenth Circuit issued its opinion in RFA, many small refineries have
contacted EPA to supplement their original SRE petitions and to provide additional information
about their financial situations. EPA greatly appreciates this information. EPA has completed a
preliminary evaluation of the data and information provided in the SRE petitions to determine if
any of the petitioners have demonstrated that the cost of compliance with the RFS is the cause of
the DEH, and that such costs are not passed through by that small refinery to the consumer under
the RIN cost passthrough principle.
87 Id. at 2181-82 ("Indeed, the dissent finds it "odd" that our reading would permit hardship relief only to small
refineries in existence in 2008 and not to new ones, post, at 2189-2190... Nor is there anything odd about the fact
that Congress chose only to protect existing small refineries rather than new entrants. Often Congress chooses to
protect existing market participants from shifts in the law while applying new restrictions fully to future entrants.")
88 We note that this issue was neither before the courts in RFA or in HollyFrontier because the three small refineries
at issue in those cases all received the initial blanket exemption.
89 40 CFR 40.144l(e)(2)(iii).
90 40 CFR 80.1441(e)(2).
20
-------
C. DOE Consultation and EPA Consideration of the DOE Study and Other Economic
Factors
CAA section 21 l(o)(9)(A)(ii) required that EPA grant exemptions upon DOE's
determination that a small refinery "would be subject to a disproportionate economic
hardship."91 Section 21 l(o)(9)(B), in contrast, provides that EPA must "consult[] with the
Secretary of Energy," but does not dictate any particular action that EPA must take following
consultation. It also does not provide any further direction on EPA's consultation with DOE, nor
does it inform DOE on the form its consultation must take. In fact, "Congress placed no limits on
how DOE should provide its consultation to EPA under [the RFS]."92 This absence of direction
provides "substantial discretion" to the agencies to determine how DOE will provide
consultation for the pending SRE petitions.93 Both agencies previously relied on DOE's findings
through its application of the DOE scoring matrix to effectuate DOE's consultation on each SRE
petition.94 For this proposed action, EPA has consulted with DOE through discussions in
meetings and phone conversations regarding the pending SRE petitions, the supplemental
supporting information the small refineries provided, and the analysis and proposed
determinations in this document.
In evaluating petitions for SREs under CAA section 21 l(o)(9)(B), EPA is directed to
"consider the findings of the [DOE] study." DOE, in fact, conducted two studies, one in 2009
and an update to the study in 2011.95 The original 2009 DOE Study concluded that small
refineries would not face DEH from compliance with the RFS program given the proportional
obligations of the program as a function of their gasoline and diesel fuel production and the
opportunity for refineries to comply by blending or by purchasing RINs, provided that the RIN
market proved to be liquid and competitive. The RIN market has developed to be open,
competitive, liquid, and functioning as intended;96 hence, the 2009 DOE Study accurately
forecasted what was likely to occur given the highly competitive fuels market with which DOE
was familiar.
When DOE expanded its study in 2011, it posited that small refineries could face DEH "if
blending renewable fuel into their transportation fuel or purchasing RINs increase[d] their cost of
products relative to competitors." (emphasis added).97 DOE expressed a similar possibility
another way noting, "If certain small refineries must purchase RINs that are far more expensive
than those that may be generated through blending, this will lead to disproportionate economic
hardship for those effected entities." (emphasis added).98 Looking to a potential future where
RIN prices rose significantly (as they have since done) DOE projected, "there are numerous
circumstances when RIN prices could rise, increasing the cost of compliance and perhaps
increasing the cost of compliance more for refineries that rely on RINs for compliance compared
91 See supra, Section II.D.
92 Hermes, 787 F.3dat577.
93 Id. at 575.
94 See supra, Section II.D.
95 See supra, Section II.D.
96 See infra, Section IV.D.2
97 2011 DOE Study at vii.
98 Id. at 2.
21
-------
to those that do not." (emphasis added).99 To make clearer the circumstances it was envisioning
where such disproportionate costs could arise, DOE provided a detailed appendix, Appendix B,
which laid out scenarios for three refiners in different circumstances relative to the RFS
program.100 The first case was a refiner that blends all its production with ethanol and does not
have to purchase ethanol RINs. The second case was for a refiner that does not do any blending
and must purchase RINs to meet all of its RVO. Finally, the third case was for a refiner with
excess RINs to sell into the market. What Appendix B assumed was that the refiner that got its
RINs through blending ethanol would get the RINs at nearly no cost, while the refiners that had
to buy RINs would be forced to pay the higher market cost for compliance. In this way, DOE
projected that some refineries could face a disproportionate cost of compliance. Through the
matrices in its report, DOE evaluated whether those disproportionate costs rose to a level such
that a refinery faced DEH due to those higher costs. DOE articulated bringing those two
elements together thusly, "[disproportionate economic hardship must encompass two broad
components: a high cost of compliance relative to the industry average, and an effect sufficient to
cause a significant impairment of the refinery operations."101 However, DOE did not assess in
this study whether their assumptions that refiners bear different costs for RINs and that they may
not be able to pass these costs on to consumers in the marketplace actually occurred.
A number of small refineries have stated to EPA that DOE's projection in the 2011 DOE
Study is exactly what has come to pass. Ethanol (D6) RIN prices have risen significantly, and
small refineries argue that they bear these higher RIN costs while integrated refiners (blenders)
receive RINs at almost no cost. Further, they argue that these disproportionate costs are
significant enough that they constitute DEH for the refineries just as DOE articulated. EPA has
carefully reviewed data, contracts and other information from small refineries to evaluate if, as
DOE posited in 2011, refineries that acquire RINs through blending get them at a lower cost than
do refineries that purchase RINs on the open market.102 What we have found is that the RIN
discount phenomenon applies—blenders, in fact, discount their sales price for E10 by the market
price of the RIN (i.e., the sales price of E10 reflects the cost to buy ethanol minus the market
price for selling the RIN). Hence, while the blender gets the RIN for "free" when it purchases a
gallon of ethanol, it has to discount the price of that ethanol when sold as E10 by the full current
market price of the RIN. This means the blending refinery pays the full market cost of the RIN
through the discount it gives in the price of the E10 it sells. The 2011 DOE Study did not
consider that blending refineries would have to discount blended fuel by the price of the RIN,
and hence, projected a potential outcome in finding DEH that has not occurred in practice.
Rather, as the 2009 DOE Study anticipated, the competitive market forces have resulted in the
same cost of compliance whether that cost comes in the purchase of RINs on the open market or
through the discounting of the price blended fuel sold by blenders. Neither the 2009 DOE Study
nor the 2011 DOE Study anticipated the even more significant finding that, without regard to
how refineries experience the RFS compliance cost, the RIN cost passthrough phenomenon
applies—refineries pass those higher costs through to their customers in higher prices for the
refined products they sell.
99 Id. at 3.
100 Id. at B-4.
101 Id. at 3.
102 See infra, Section IV.D.2.
22
-------
For the reasons described above and after considering the "other economic factors"
described in Sections IV.D.2 and 3, we propose to find small refineries do not face
disproportionate costs to comply with RFS. Further, we propose to find there is no economic
harm, much less a hardship significant enough to impair refinery operations, that qualifies as
DEH caused by RFS compliance. For these reasons, we propose to find, consistent with the
broad criteria for relief described in the 2009 and 2011 DOE Studies, that DEH is not
demonstrated in the petitions EPA has evaluated.
I). Hardship Must Be Caused by RFS Compliance
1. The CAA Requires That DEH Must Be Caused by RFS Compliance
As discussed briefly above, the best reading of the statutory provisions at CAA section
21 l(o)(9) is that EPA's authority to grant an SRE "for the purpose of (DEH)" requires that the
hardship be caused by RFS compliance. This interpretation aligns with both the statutory text
and the purpose of the RFS program and the exemption.
a. The Text of the Statute Provides That DEH Must Be Caused by Compliance with
the RFS Program
EPA did not require that DEH be caused by compliance with the RFS program in prior
decisions to grant SREs.103 On January 24, 2020, the Tenth Circuit in RFA held that the EPA
only has the authority to grant SREs when the refinery experiences DEH caused by the RFS
program.104 The court pointed to statements in the three decision documents at issue indicating
that relief from the RFS obligations could relieve the refinery's hardship "in whole or in part,"
and concluded that granting relief on the basis of something other than DEH caused by RFS
compliance was impermissible.105 We have evaluated the court's opinion and the text of the
statute, and, going forward, we propose to require that the petitioning small refineries
demonstrate that DEH is caused by RFS compliance as discussed further in this section.
The CAA's SRE provision is structured in two sections. Section "(A) Temporary
exemption" provides the blanket exemption to all small refineries through 2010 and then lays out
the conditions in which a small refinery may receive an extension of the initial exemption based
on the DOE study. Section "(B) Petitions based on [DEH]" addresses ongoing case-by-case SRE
petitions and how EPA must evaluate those petitions.
Section A refers to the "requirements of paragraph [21 l(o)(2)]," which provides, among
other things, the applicable annual volume targets for the required categories of renewable fuel.
The "requirements of paragraph [21 l(o)(2)]" are utilized in describing what an exemption
means: "The requirements of paragraph [21 l(o)(2)] shall not apply to small refineries until
calendar year 2011,"106 as well as identifying the subject of the DOE's study: "[T]he Secretary of
Energy shall conduct for the Administrator a study to determine whether compliance with the
103 See supra, Section III.
104 RFA, 948 F.3d at 1254.
105 Id.
106 CAA section 21 l(o)(9)(A)(i).
23
-------
requirements of paragraph [21 l(o)(2)] would impose a [DEH] on small refineries."107 It also
describes the basis under which an exemption can be extended: "[i]n the case of a small refinery
that the Secretary of Energy determines under subclause (I) would be subject to a [DEH] if
required to comply with paragraph [21 l(o)(2)], the Administrator shall extend the exemption
under clause (i) for the small refinery for a period of not less than 2 additional years." (emphasis
added).108 These references to paragraph 21 l(o)(2) indicate a direct link between the RFS
requirements and DEH. Thus, the best reading of the statutory language is that compliance with
the RFS program must be the impetus for DEH warranting an SRE under section A, meaning
that a small refinery may not simply experience a year of poor economic performance or struggle
with disadvantageous operational or market constraints to merit an SRE. Nor can a refinery rely
on unplanned events like a fire, natural disaster, or planned events such as paying out stock
dividends or other capital purchases/loans to qualify for relief from its RFS obligations.109
Rather, section A of the SRE provision provides that DEH must be caused by the small
refinery's compliance with the requirements of the RFS program.110
Section B of the SRE provision states that a small refinery may "at any time petition the
Administrator for an extension of the exemption under subparagraph (A) for the reason of
[DEH] "m By making any ongoing SREs mere "extensions] of the exemption under
subparagraph (A)," Congress carried over the causal requirement in section A to section B.112
While section B uses the language "for the reason of [DEH]" without a modifying clause tying it
to compliance with the RFS program, section B cannot be read outside of the context of section
A; section B is merely providing an opportunity for small refineries to request continuation of
the exemption in section A. Therefore, the causal requirement in section A tying DEH to RFS
compliance applies to section B as well. Additionally, it is section A that provides the basis on
which DEH must be founded: compliance with the RFS program. Thus, even if the exemption
under section B could be interpreted as a distinct exemption from the exemption under section A,
it must be "for the reason of [DEH]" as defined in section A as being "impose[d]" by, or existing
"if [a small refinery was] required to comply with" its RFS obligations. In this way, the use and
meaning of "disproportionate economic hardship" is the same in both sections A and B.
Therefore, we agree with the Tenth Circuit that the "language of these provisions indicates that
renewable fuels compliance must be the cause of any disproportionate hardship."113 As described
above, EPA believes this is the best interpretation of the inter-related provisions of CAA sections
21 l(o)(9)(A) and (B) and is therefore proposing to adopt this interpretation going forward.
107 CAA section 21 l(o)(9)(A)(ii)(I).
108 "In the case of a small refinery that the Secretary of Energy determines under subclause (I) would be subject to a
[DEH] if required to comply with paragraph [21 l(o)(2)], the Administrator shall extend the exemption under clause
(i) for the small refinery for a period of not less than 2 additional years." CAA section 21 l(o)(9)(A)(ii)(II).
109 RFA, 948 F.3d at 1254. ("Granting extensions of exemptions based at least in part on hardships not caused by
RFS compliance was outside the scope of the EPA's statutory authority.")
110 Id.
111 CAA section 21 l(o)(9)(B)(i) (emphasis added).
112RFA, 948 F.3d at 1253.
113 Id.
24
-------
b. The Purpose of the RFS Program Supports a Requirement That DEH Must Be
Caused by Compliance with the RFS Program
Requiring that DEH be caused by RFS compliance also furthers the purpose of the RFS
program, which exists to encourage the use of renewable fuel. Historically, SREs have resulted
in reductions in the volume of renewable fuel used in the United States.114 In general, hardship
provisions exist to provide particular parties additional time to come into compliance with new
regulations if they are expected to or are facing difficulties adjusting to a new regulatory
requirement. This is historically how EPA has designed and implemented hardship provisions for
other fuels programs, often with phase-out provisions (i.e., only available for a set number of
years after implementation).115 Therefore, it is logical to interpret the SRE provision to also
provide initial time for small refineries to come into compliance, with the expectation that they
would do so, and would only be eligible for an extension of the exemption if they suffered
hardship specifically due to the RFS program itself.116
Additionally, allowing relief from RFS obligations for hardship completely unrelated to
the RFS program would be an inappropriate use of an exemption provision, particularly one
where the text of the statute requires demonstration of a causal relationship between the hardship
and the program. Had Congress intended that EPA provide relief for hardship due to something
other than the RFS program, the statutory language likely would have been more explicit in
providing such broad authority. It would be illogical for the "temporary hardship" provision to
have been established as an opportunity to prop up businesses and provide relief for reasons
wholly unrelated to the RFS program, the program from which it is providing relief. It would
only make sense that, in implementing the RFS program, we provide relief from impacts of the
RFS program that result from the RFS program itself. It is hard to imagine that Congress
intended the SRE provision be used to cure the financial distress some small refineries may
otherwise face, especially when other legal and policy options exist to provide compliance
flexibility, and, significantly, when that distress may be caused by a broad array of circumstances
unrelated to the RFS program, ranging from difficult geography to adverse business decisions.
Finally, the granting of SREs in light of RIN cost passthrough means that exempted small
refineries are not only relieved of their RFS obligations, but also get a financial benefit through
the sale of RINs obtained through blending, or through the sale of their petroleum fuel that
includes the value of the RIN.117 This windfall to small refineries does not further the goals of
the RFS program, and only provides a disproportionate net benefit to small refineries granted
exemptions in comparison to other refineries that are either ineligible to petition for exemption or
are denied an exemption on the lack of merit of their petition.118 Furthermore, when small
refineries gain this benefit through exemption, RFS compliance is incrementally shifted to other
114 We acknowledge that beginning in 2020, we have projected the amount of SREs such that when the projections
accurately reflect the volume of fuel exempted, the volume of renewable fuel required under the RFS program is not
reduced by the granting of SREs.
115 See, e.g., the diesel fuel hardship provisions at 40 CFR 80.560(a), which effectively allowed a delay of up to four
years in complying with EPA's diesel fuel sulfur standards.
116 We note that such an approach is consistent with Congressional intent for the SRE provision, whether that intent
is considered to be a "funnel to compliance" or a "safety valve," as considered by the Supreme Court in deciding
HollyFrontier. See HollyFrontier, 141 S. Ct. at 2181-82.
117 See infra, Section IV.D.2.
118 See, e.g., Comments from API on 2020 RFS Annual Rule, Docket Item No. EPA-HQ-OAR-2019-0136-0721.
25
-------
parties, which, in turn, pass on that increment in their compliance costs to consumers. In essence,
the significant financial benefit of exemptions granted to small refineries is paid for by
consumers in still higher transportation fuel costs.119
2. DEH and RIN Cost Passthrough
An additional part of the Tenth Circuit's holding was that EPA failed to explain how a
finding of DEH comports with EPA's findings on RIN cost passthrough.120 In this action, we are
proposing to adopt an interpretation of the statute that DEH must be caused by compliance with
the RFS program. It follows, then, that in making a finding of DEH we would have to explain
how the RFS program could cause DEH for a small refinery in light of EPA's longstanding and
consistent findings on RIN cost passthrough.
After reviewing the available data and analysis, including analyses conducted by EPA
and outside parties, as well as data and analyses submitted by petitioning small refineries, we
propose to find that all obligated parties recover the cost of acquiring RINs by selling the
gasoline and diesel fuel they produce at the market price, which reflects these RIN costs (RIN
cost passthrough). Further, we propose to find that blenders use the revenue from RIN sales to
discount the price of the blended fuel they sell (RIN discount). We further propose to find that,
since refining and fuel blending markets are highly competitive, the RFS obligation is the same
for every gallon of gasoline and diesel fuel, RINs are widely available in an open and liquid
market, and the cost of acquiring RINs is the same for all parties, all types of obligated parties
bear the same cost from compliance with the RFS program as these aspects of the RFS program
and the RIN market facilitate the RIN cost passthrough and the RIN discount principles
discussed above. While some parties dispute EPA's findings on RIN cost passthrough and the
RIN discount, those same parties have not made business decisions over the last decade that
would be logical if RIN cost passthrough and RIN discount were not occurring. For example, if
RIN cost passthrough did not exist, we would expect to see refiners avoiding RFS obligations by
shifting production to non-obligated fuel (e.g., heating oil, jet fuel) and/or export fuel. We would
also expect to see actions to expand or modify their business models to include additional
blending of renewable fuel to reap the alleged rewards that they claim independent blenders and
marketers enjoy.121 Taken together, we propose to conclude that the RFS program does not
impose a DEH on small refineries.
Assessing the impact of the RFS program on refiners and blenders is complicated for
several reasons. First, many parties may operate in several different roles, such as merchant
refiners, integrated refiners, and blenders, in any given year.122 Second, the impact of RIN costs
119 In the 2020 RFS Annual Rule, EPA finalized regulations that shift the projected exempted volumes for small
refineries to the remaining obligated parties instead of reducing the renewable fuel volumes as had been common
practice in prior years. 85 FR 7016 (February 6, 2020).
120 RFA, 948 F.3d at 1256-57.
121 There has been an expansion of independent gasoline blenders and marketers in recent years. We anticipate this
trend will continue. This is driven by the gasoline market being oversupplied, making it economical for companies
to invest in blending and purchase gasoline blendstocks on the spot market. The typical fraction of a cent return is
sufficient to offset capital investments in blending infrastructure. The market dynamic has swung back and forth
over the decades independent of the RFS program and RIN costs.
122 See infra, Section IV.D.2.C.
26
-------
on the price of fuels is not often apparent in the market pricing data.123 Third, while market
prices for renewable fuel with RINs attached are readily available in posted prices, renewable
fuel is less commonly traded without RINs and hence prices of renewable fuel without the RIN
are also rarely available outside of contracts between parties that are claimed as confidential.124
Finally, terminology and accounting practices vary between different parties, often making
apples to apples comparisons difficult.125
In this section, we present the data and analysis to support our proposed findings that
small refineries do not suffer DEH from their RFS obligations because RIN costs are fully
passed through to consumers. Further, we show that any such RFS compliance costs are not
disproportionate because the cost to acquire RINs, whether via blending, or through the RIN
market, are the same, making the costs of RIN acquisition the same for all parties. First, we
provide a brief description of prior publications on RIN cost passthrough and the RIN discount.
Second, we reiterate the general economic theory that supports the premises of RIN cost
passthrough and the RIN discount. Third, we briefly discuss the different market participants,
and how we expect their operations to be affected based on economic theory. Fourth, we analyze
the most current data available to the Agency to determine whether the finished fuel and RIN
markets move in the way the economic theory predicts.
Small refineries alleging DEH generally claim that: (1) They are unable to recover the
cost of the RINs they purchase in the sales prices of the gasoline and diesel fuel they produce
because of their geography or market position; and/or that (2) They face higher costs for
acquiring RINs than their competitors (usually integrated refiners or non-obligated blenders) that
acquire RINs by blending qualifying renewable fuel. In the first case, petitioners argue that they
are unable to recover the added cost of RIN purchases needed for RFS compliance and/or that
the market price for gasoline and diesel fuel does not fully reflect these costs. In the second case,
petitioners argue that their competitors (non-obligated blenders and/or integrated refiners) do not
have to discount the blended fuel they sell to consumers by the price of the RIN and, therefore,
are able to acquire these RINs at a lower net cost than parties that purchase RINs. EPA has not
found evidence to support either of these arguments, as shown by the data and analysis presented
below. We note that the data we looked at in doing this analysis and the market behavior they
describe are consistent across all the markets we observed.
a. Previous Assessments of RIN Market Dynamics
The degree to which the cost is "passed through" to consumers (RIN cost passthrough)
and revenue from RIN sales is used to discount the price of blended fuel (RIN discount) has been
a longstanding area of interest, especially since D6 RIN prices increased dramatically in 2013.
EPA first published results of an assessment of obligated parties' ability to "pass through" RIN
costs and the impact of RIN prices on the price of blended fuel in a technical memorandum in
2015.126 EPA explained the economic principles at work that enabled obligated parties to recover
123 See infra, Section IV.D.2.b.
124 See infra, Section IV.D.2.d.
125 See infra, Section IV.D.2.d.ii.
126 -A Preliminary Assessment of RIN Market Dynamics, RIN Prices, and Their Effect," Dallas Burkholder, Office
of Transportation and Air Quality, US EPA, May 14, 2015.
27
-------
their RIN costs through RIN cost passthrough and the discount of renewable fuel blends by the
price of the RIN. EPA then examined several sources of market data to test those principles. We
concluded that both the costs in refined products and discounts in blended fuel prices due to
RINs were being fully passed through to consumers.
EPA next considered this issue in the context of petitions to reconsider the point of
obligation in the RFS program.127 While RIN cost passthrough was not the only topic at issue in
our consideration of changing the point of obligation in the RFS program, the degree to which
RIN costs and the RIN discount were passed through to consumers was a central argument in the
various petitions. In considering these requests, EPA again examined available market data, as
well as studies by outside parties and numerous public comments.128 Once again, EPA concluded
that the RIN costs and RIN discount were fully passed through to consumers and reflected in the
market prices of petroleum fuel and blended fuel, and that blenders used revenue from RIN sales
to discount the price of blended fuel. This decision was reviewed and upheld by the U.S. Court
of Appeals for the D.C. Circuit.129
In evaluating the SRE petitions currently before the Agency, EPA has again evaluated the
available market data. EPA has examined data through 2020 to determine whether more recent
data continues to support EPA's views on the economic principles at play in the RIN market and
whether these new data reconfirm our prior conclusions about both RIN cost passthrough and the
RIN discount. EPA's prior analyses were generally based on publicly available data reported by
the Energy Information Administration (EIA), which reports spot fuel prices for large fuels
markets such as the New York Harbor or the Gulf Coast. Several small refineries claimed that,
while RIN cost passthrough and the RIN discount may occur in these larger and more
competitive fuels markets, RIN cost passthrough and the RIN discount were not occurring in the
local markets into which these small refineries sold gasoline and diesel fuel. To assess these
claims, EPA analyzed the data we received, including data sets provided by some of the small
refinery petitioners located in smaller markets. The petitioners submitted the datasets to disprove
EPA's conclusions on RIN cost passthrough. However, EPA found that the data, including the
more recent data through 2020, either could not be used to draw conclusions regarding RIN
market dynamics, or supported the conclusions that RIN costs are passed through in higher
127 "Denial of Petitions for Rulemaking to Change the RFS Point of Obligation," EPA-420-R-17-008 at 21-31,
November 2017.
128 C.R. Knittel, B.S. Meiselman, & J.H. Stock, "The Pass-Through of RIN Prices to Wholesale and Retail Fuels
under the Renewable Fuel Standard," Journal of the Association of Environmental and Resource Economists, 2017.
C.R. Knittel, B.S. Meiselman, & J.H. Stock, "The Pass-Through of RIN Prices to Wholesale and Retail Fuels under
the Renewable Fuel Standard: Analysis of Post-March 2015 Data," Working Paper. See also Letter from RaceTrac
to Administrator McCarthy, August 17, 2016, Docket Item No. EPA-HQ-OAR-2016-0544-0014; Letter from
QuikTrip to Administrator McCarthy, August 17, 2016, Docket Item No. EPA-HQ-OAR-2016-0544-0013;
Presentation from Murphy USA to EPA, August 16, 2016, Docket Item No. EPA-HQ-OAR-2016-0544-0028.
129 A Ion Refining Krotz Springs, Inc v. EPA, 936 F.3d 628 (D.C. Cir. 2019). In its decision, the D.C. Circuit found
that in determining whether refiners recover the cost of the RINs they purchase for RFS compliance, EPA
"grounded that conclusion in studies and data in the record." Id. at 649. The D.C. Circuit also supported EPA's
findings that there is a cost for integrated refiners and non-obligated blenders to acquire RINs, even if they do not
purchase separated RINs, through lower prices for blended fuels. "In a competitive market there's no such thing as a
free lunch, and blenders and integrated refiners pay their tab just as other do; they just do so indirectly. To offer
finished fuel without attached RINs at a competitive price, these entities must discount their blended fuel by roughly
the value of the RINs that they detach and kept for themselves." Id. at 650.
28
-------
refined product prices and that blended fuel prices are discounted by the price of the RIN and,
hence, passed through to consumers.130 This means that no obligated party has a structural
advantage or disadvantage from the RFS program. EPA found these conclusions held not only in
the large fuels market previously assessed, but also in the smaller markets EPA examined using
non-public market data, as well as the data submitted by the small refineries. Each of these
assessments is discussed in further detail in the following sections.
b. Economic Principles of RIN Cost Passthrough
The market for gasoline and diesel fuel in the United States is extremely competitive at
all levels, including the retail and wholesale levels. At the retail level, there are currently about
145,000 retail stations across the United States.131 The majority of these stations are owned by
parties that own fewer than ten retail stations, and, in many cases, only a single retail station.132
All of these parties are selling fungible products (gasoline and diesel fuel) to a consumer base
that is very sensitive to fuel prices, with prices posted on large signs making prices transparent.
At the wholesale level, there are 129 petroleum refineries in the United States.133 The market for
renewable fuel and RINs is similarly very competitive. In 2020, more than 300 companies
generated RINs for qualifying renewable fuel.134 On average, approximately 5 billion RINs are
traded between registered parties each month.135 Prices for petroleum fuel, renewable fuel, and
RINs are regularly reported by a variety of price reporting services.136
Refineries within the United States compete with each other, as well as with many other
refineries overseas, and importers capable of sourcing gasoline and diesel fuel from a global
fuels market. Low transportation costs for gasoline and diesel fuel, enabled by an extensive
pipeline network, and the low cost of shipping these fuels via pipeline, barge, and petroleum
tankers, mean that fuels markets across the United States are linked and that refiners are not only
competing with other local refineries, but with parties across the country and in many cases the
world. This can be seen clearly in the structure of many fuel supply contracts across the country
that establish pricing based on the price of fuel at a major market (such as Houston or New York
Harbor) plus or minus transportation costs between the local market and the major market,
130 See infra, Section IV.D.2.d.
131 National Association of Convenience Stores, Convenience Stores Sell the Most Fuel, March 24, 2021,
https://www.conven.ience.org/Topics/Fneis/Who-Seils-Americas-Fiiei.
132 Id. According to this data, 57.1% of retail fuel stations are owned by parties that own only one station, and an
additional 3.8% of all retail fuel stations are owned by parties that own 2-10 retail stations.
133 According to data from EIA, there were 129 operable refineries in the United States as of January 1, 2021 (EIA,
When was the last refinery built in the United States?, Frequently Asked Questions (FAQs), June 25, 2021,
https://www.eia. gov/tools/faas/faa.php?id=29&t=6). Some of these refineries are located outside of the RFS covered
location or do not produce gasoline or diesel fuel, and thus are not subject to the RFS program.
134 The number of companies that generated RINs is from data accessed from EPA's Moderated Transaction System
(EMTS).
135 RIN trade and price information reported to EMTS is available at: https://www.epa.gov/fuels-registration-
reporting-and-compliance-help/rin-trades-and-p rice-information.
136 See, e.g., fuel price data from EIA (https://www.eia.gov/dnav/pet/pet pri spt s i d.htm) and RIN price data from
EPA (https://www.epa.gov/fuels-registration-reporting-and-compliance-help/rin-trades-and-price-information).
29
-------
depending on the direction of product flow.137 If a small refinery is facing competition in its
local market from a larger remote market, the local price will typically be higher than the price in
the major market, reflecting the cost of shipping the fuel to the local market from the larger
remote market.138 Conversely, if the small refinery is shipping its fuel to the larger market to sell,
it will need to price its fuel below the remote market price to cover the cost of shipping the fuel
to the remote market. Through thousands of decisions made by all the market participants each
day, the prices between the markets equilibrate to the same level, offset by the transportation
costs between the markets.
Economic theory suggests that in competitive markets like the fuels market, competitive
market forces would drive market participants to pass through the costs and revenue from RINs
to consumers in the prices of the products they sell. This means that higher RIN prices should not
advantage any one group of refineries over another, and that RIN prices should not impact
refining margins. As an initial assessment of the impact of RIN prices on refineries, EPA
examined the refining margins for three groups of refineries—small refineries, large refineries,
and all refineries—based on available public data (e.g., financial data from publicly traded
companies) and confidential data, including data provided by petitioners. We compared these
refining margins (operating profit per gallon of fuel produced) to the average RIN cost per gallon
(the per gallon cost to acquire the RINs necessary to meet a refinery's RVO).139 These data are
presented in Figure IV.D.2.b-l. Consistent with the economic theory, we see no correlation
between refining margins and RIN prices, nor do we see any indication that higher RIN prices
put small refineries at an advantage or disadvantage relative to large refineries. Figure IV.D.2.b-
1 also includes an estimate of the refining margin for small refineries if they received an
exemption from their RFS obligations. The estimate was calculated by adding the RFS RIN
compliance cost per gallon to the refining margins for small refineries each year, since
exempting small refineries from their RFS obligations means they do not have to acquire RINs.
This estimate demonstrates that exempting small refineries from their RFS obligations results in
small refineries, as a class, having consistently higher refining margins than large refineries or
the average of all refineries. This advantage is significant and increases as RIN prices increase.
137 Several small refinery petitioners included examples of contracts, some of which were based on the fuel price at a
larger fuel market plus (or minus) transportation costs. This information has been claimed as confidential by the
petitioners.
138 This is because the price in the local market will be set by the marginal supplier of fuel. In a market with both a
local and remote supplier, the marginal supply price will be no lower than the fuel sourced from the remote market,
which will include transportation costs.
139 We calculated the RIN cost per gallon based on the RFS obligation and the average RIN prices for each year.
30
-------
Figure IV.D.2.b-l: Refining Margins and RIN Costs (2009-2019)a
$0.35
tS
<-> $0.30
z
% $0.25
~ $0.20
c
_o
re $0.15
-2?
— $0.10
V)
£? $0.05
CO
S <_
60 ¦?
c
c
4-
u
^ All Renneries Large Rerinenes
Small Refineries ¦ RIN Cost
— — — Small Refinery with Exemption
Data from SRE petitions and financial statements from publicly traded companies.
a The "Small Refinery with Exemption" line was calculated by adding the "RIN cost" line to the "Small Refineries"
line. If a small refinery had already accounted for the financial benefit of an SRE in their reported margin for a given
year, the effect would be to make the "Small Refinery with Exemption" line slightly less than shown for that year.
Understanding the impacts of the RFS program on the various parties that participate in
the fuels market is complicated by the fact that different parties may participate in different
activities within the fuels market. When analyzing the impact of the RFS program on the fuels
market, we generally consider three different types of market participants: (1) Parties that
produce and sell petroleum fuel, including blendstocks140 (generally referred to as merchant
refiners); (2) Parties that purchase petroleum fuel and renewable fuel, and sell blended fuel
(blenders); and (3) Parties that produce petroleum fuel, purchase renewable fuel, and sell blended
fuel (integrated refiners). A simplified version of the business activities each of these parties
engage in, as well as the impact of the RFS program on their costs and revenue, is illustrated in
Figure IV.D.2.b-2 below.
Merchant refiners produce, market, and sell petroleum fuel and buy the RINs they need
for compliance with their RFS obligations; they do not purchase or blend renewable fuel.
Integrated refiners also produce petroleum fuel, but unlike merchant refiners, they also purchase
and blend renewable fuel to produce, and ultimately sell, blended fuel that contains some volume
of renewable fuel. Integrated refiners generally do not purchase RINs, but instead purchase
renewable fuel with attached RINs and acquire most of the RINs they need for compliance when
they blend the renewable fuel.141 Non-obligated blenders do not produce petroleum fuel
1411A "blendstock" is defined as "any liquid compound or mixture of compounds (not including fuel or fuel additive)
that is used or intended for use as a component of a fuel." 40 CFR 1090.80.
141 Very few, if any, integrated refiners acquire all the RINs they need by blending renewable fuel. Petroleum fuel is
subject to an RFS obligation for all four categories of renewable fuel, but it is generally only blended with one type
of renewable fuel (i.e., ethanol in the case of gasoline and biodiesel or renewable diesel in the case of diesel fuel).
31
-------
components, but instead purchase these products from merchant refiners. They then purchase
renewable fuel with attached RINs that they use to produce, and ultimately sell, blended fuel
(e.g., E10 and B5).142 Because these parties do not have RFS obligations, they can also sell the
RINs associated with the renewable fuel they blend.
In practice there are few refineries that fall entirely into a single category, with most
refiners having business interests that fall into at least two categories. Nevertheless, these
distinctions help to clarify the context for RIN cost passthrough and the RIN discount in the price
of blended fuel. While the RFS program results in new costs and revenue streams for various
parties, we expect that, for each of these types of market participants (merchant refiners,
integrated refiners, and blenders), these new costs are passed through in higher prices for refined
products, while these new revenues from the sale of RINs are reflected in lower market prices for
blended fuels, such that no refiner or blender realizes an advantage or disadvantage from the RFS
143
program.
Based on the 2020 RFS percentage standards, integrated refiners would generate a small amount of excess
conventional biofuel (D6) RINs when blending ethanol as E10, but would need to purchase a small number of
advanced biofuel (D5), biomass-based diesel (D4), and cellulosic biofuel (D3) RINs to meet the RFS obligation
associated with the petroleum-based portion of the E10 blend. Similarly, integrated refiners that blend biodiesel as
B5 would generate excess D4 RINs but would need to purchase D6 and D3 RINs to meet the RFS obligation
associated with the petroleum-based portion of the B5 blend. In practice, nearly every gallon of blended fuel
produced by an integrated refiner generates some quantity of excess RINs of one type and simultaneously incurs an
obligation for other types of RINs.
142 B5 refers to diesel fuel blended with 5% biodiesel.
143 See infra, Section IV.D.2.C.
32
-------
Figure IV.D.2.b-2: Simplified Illustration of Fuels Market Participants
The place in the fuel supply chain where we can see the cost of the RIN being passed
through to consumers is in the price of the petroleum products. Since all parties have the same
cost to acquire RINs (on a per gallon basis),144 whether they blend renewable fuel or purchase
separated RINs, one would expect the price for petroleum fuel subject to an RFS obligation (i.e.,
gasoline and diesel fuel) to increase when RIN prices increase and to decrease when RIN prices
decrease. Just as the prices of gasoline and diesel fuel increase if fuel taxes increase,145 they also
increase when RIN prices increase. Merchant refiners fully recover the cost of their RFS
obligations when the difference between the market price of gasoline and diesel fuel and the
market price for these fuel in the absence of the RFS obligation is equal to the cost of purchasing
the RINs to satisfy the RFS obligation. Equations showing the expected RIN price impacts on the
prices of gasoline and diesel fuel, assuming RIN costs are fully passed through, are shown
below.
144 See infra, Section IV.D.2.C.
145 EIA, Gasoline explained: Factors affecting gasoline prices, March 2, 2021,
https://www.eia.gov/energvexplained/gasoline/factors-affecting-gasoline-prices.php.
33
-------
Equation 1: Expected Impact on Gasoline (EO) Prices Assuming Full RIN Cost Passthrough
Gasoline Price = Gasoline Price with no RFS Obligation + RIN Costs
Equation 2: Expected Impact on Diesel Fuel (BO) Prices Assuming Full RIN Cost Passthrough
Diesel Fuel Price = Diesel Fuel Price with no RFS Obligation + RIN Costs
EPA once again examined these economic principles by looking at available market data,
including more recent market data.146 The data EPA examined show that the market prices for
gasoline and diesel fuel operate as shown in Equations 1 and 2, supporting EPA's findings that
all obligated parties recover the cost of their RFS obligations in the sale prices for the gasoline
and diesel fuel they produce.147 The ability for an obligated party to recover its RIN costs is not
dependent on the obligated party's ability to set the price for these fuels in the markets where
they are sold. Rather, because all obligated parties face the same RIN costs per gallon of gasoline
and diesel fuel produced nationwide,148 the market prices for these fuels rise and fall with
changes in RIN prices in all markets by the same amount on any given day (after accounting for
other factors that impact the prices of these fuels), such that all parties that sell gasoline and
diesel fuel recover their RIN costs.149
The place in the fuel supply chain where we see the RIN discount is the point at which
renewable fuel is blended with gasoline or diesel fuel and sold to fuel retailers. Parties that blend
renewable fuel with gasoline or diesel fuel to produce transportation fuel must discount the price
of the blended fuel by the price of the associated RIN.150 These parties can then separate any
RINs that are attached to the renewable fuel and either use these RINs to demonstrate
compliance with their RFS obligations (if they are an obligated party) or sell these RINs to other
parties. In either case, the point at which they acquired the RIN at the market price, or, rather,
incurred a market rate cost for the RIN, is the point when they had to discount the blended fuel
by the price of the RIN.
The sale of a RIN by a party that blends renewable fuel and separates the RIN creates a
separate revenue stream in addition to the revenue from the sale of the blended fuel itself.
Competitive forces require that blenders price their blended fuel based on the net price of
renewable fuel, or the price of the renewable fuel less the price of the RIN associated with the
fuel (e.g., net ethanol price = ethanol price - D6 RIN price; net biodiesel price = biodiesel -
1.5*D4 RIN price151). Any party that attempts to retain the revenue from the RIN sales, rather
than passing it on to consumers via the RIN discount, is unable to offer blended fuel at a
146 EPA's analysis of the market data to determine the degree to which RIN costs are passed through to consumers
through higher prices for gasoline and diesel fuel is provided in Section IV.D.2.d.i.
147 See infra, Figures IV.D.2.d.i. 1 through 4, where EPA compared the price difference between a fuel subject to an
RFS obligation to a very similar fuel not subject to an RFS obligation and the RIN cost per gallon of diesel fuel.
148 See infra Section IV.D.2.d.ii.
149 See infra Section IV.D.2.d.i.
150 Another way to think about the RIN discount is that, to remain competitive, parties that blend renewable fuel
must base the final price for the blended fuel on the net price of the renewable fuel (after accounting for the sale of
the RIN) rather than on the price they paid for the renewable fuel with an attached RIN.
151 Each gallon of biodiesel generates 1.5 RINs.
34
-------
competitive price. If the market price for blended fuel is equal to the prices of the fuels used to
create the blended fuel (e.g., 0.9 gallons of gasoline blendstock and 0.1 gallons of ethanol in the
case of E10) without discounting the price for the renewable fuel by the price of the RIN, the
RIN sales would result in profits for the blender. In the competitive fuels market, however,
blenders are forced to reduce the price of the blended fuel to be competitive, consistent with the
RIN discount phenomenon. If they do not, their competitors will give up the revenue from the
sale of RINs to gain a greater market share. These competitive forces require that blenders use
the revenue from the RIN sales to effectively subsidize the price of the blended fuel they sell.
This market phenomenon has been relatively obvious to program participants looking at
the market for biodiesel blends where it was understood from the start of the RFS2 program that
a higher D4 RIN price was necessary to reduce the effective market price of biodiesel to make it
equivalent to petroleum diesel fuel. Integrated refiners and non-obligated blenders pay the higher
cost for renewable fuel through their purchase and blending. Merchant refiners pay the non-
obligated blenders the incremental cost of the renewable fuel for doing the blending of renewable
fuel on their behalf when they purchase the separated RINs. As an illustrative example, if
petroleum diesel fuel is selling at $3.00 per gallon, and it costs $4.50 per gallon to produce
biodiesel and generate 1.5 D4 RINs, the price of a D4 RIN would need to be $1.00 for biodiesel
to compete with petroleum diesel fuel so that the revenue from the sale of the 1.5 D4 RINs for
$1.50 would lower the effective cost of the biodiesel to match the cost of the petroleum diesel
fuel.152 Any blender attempting to retain the revenue from the sale of the D4 RINs (rather than
using it to discount the price of the blended fuel) could not offer a competitively priced blended
fuel, since any biodiesel the blender used in its product would increase the cost of the fuel blend.
As described in greater detail below both in terms of economic principles and the recent
data EPA received from small refineries, this market dynamic was previously not well
understood when applied to the blending of ethanol to make E10. From the start of the RFS
program until recently, there was no need to discount ethanol to create parity with gasoline
blendstocks because ethanol has been relatively inexpensive and highly valued as an octane
improver when blended to produce E10. As a result, both in the period prior to the RFS program
and for the early parts of the RFS program, the market price for E10 was simply the weighted
price for gasoline blendstock and ethanol. When D6 RIN prices increased, it was not obvious to
many program participants how these high RIN prices impacted E10 prices, which many
program participants simply assumed should continue to reflect the weighted costs of gasoline
blendstock and ethanol. In fact, what has happened is that the high RIN prices have increased the
production cost of gasoline blendstock (i.e., the RIN cost passthrough described in the preceding
section) while simultaneously lowering the net cost of ethanol in almost equal proportion (the
RIN discount), resulting in little change in the actual cost of E10 to consumers. While this
competitive market response has meant little change in E10 prices due to the RFS program, it has
created confusion among market participants who perceive that D6 RINs are "free" to parties
that blend E10 while obligated parties that must buy the D6 RINs at market prices bear a very
high cost. Instead, as we will show here based both on economic theory and the new small
refinery data submissions, all sellers of E10 discount the price of E10 by the price of the D6
152 In this example we are ignoring the impact of the federal biodiesel tax credit for simplicity. If we included the
impact of the tax credit, it would reduce the RIN price needed to bring the net or effective price of biodiesel to parity
with diesel fuel.
35
-------
RIN, meaning fuel blenders pay for the RIN through this discounted E10 price at the same cost
as if they purchased the RIN on the open market. As a result, parties that acquire RINs through
fuel blending and parties that acquire RINs from the open market incur the same cost to acquire
RINs.
Equations showing a generalized fuel blending example, and an example specific to E10,
are provided below. These equations and the discussion that follows describe what one would
expect if RIN prices are fully passed through to consumers. The subsequent sections examine
market data to test these equations and determine the degree to which RIN prices are passed
through to consumers.
Equation 3: Generalized Fuel Blending Example Assuming Full RIN Discount
Blended Fuel Price = PFP * PF% + (RFP - RIN Value) * RF%
Where: PFP = Petroleum Fuel Price
PF% = Petroleum Fuel Percentage in the fuel blend
RFP = Renewable Fuel Price
RIN Value = RIN Price * Equivalence Value153
RF% = Renewable Fuel Percentage in the fuel blend
Equation 4: Fuel Blending Example for E10 Assuming Full RIN Discount
E10 Price = Gasoline Blendstock Price * 90% + (Ethanol Price - D6 RIN Price) * 10%
EPA's analysis of the market data confirms these economic principles that the RIN value
is passed through to consumers in the price of blended fuel.154 The analysis—comparing the
market prices for petroleum fuel, ethanol, RINs, and E10—shows that the market prices for
blended fuel operate as shown in Equations 3 and 4, supporting EPA's findings that blenders are
passing on the value of the RIN to consumers.155 Importantly, this means that, although blenders
do not purchase RINs directly, there is still a cost for blenders to acquire RINs. This cost is
realized when blenders discount the price for the finished blended fuel, pricing it based on the
net price of the renewable fuel, after accounting for the sale of any RINs attached to the
renewable fuel. The data EPA analyzed support our finding that the RIN value is fully passed
through from blenders to consumers, as described in Equations 3 and 4. Because the market is
competitive, a blender cannot attempt to sell RINs at higher prices, consumers would merely go
153 The equivalence value is an RFS regulatory term that relates the number of RINs generated per gallon of
renewable fuel produced. Ethanol has an equivalence value of 1.0. Other renewable fuels have equivalence values
that are determined by their energy content relative to ethanol. For example, biodiesel has an equivalence value of
1.5 RINs per gallon of biodiesel reflecting that biodiesel has approximately 150% the energy content of ethanol.
154 See infra, Section IV.D.2.d.
155 See infra, Section IV.D.2.d.ii.
36
-------
to a competitor selling at the market price. Thus, the cost of acquiring a RIN by blending
renewable fuel and the cost of purchasing a separated RIN are equal as would be expected from
the design of the RFS program and RIN system.
c. Impacts on Different Market Participants
Before turning to the data that support EPA's proposed conclusion that RIN cost
passthrough and the RIN discount are reflected in the prices of refined products and blended fuel,
respectively, we first provide an illustrative example to examine the implications of RIN cost
passthrough and the RIN discount on the three types of market participants described above: a
merchant refiner, an integrated refiner, and a non-obligated blender. We present examples for
producing both E10 and B5, two common fuel blends present in many fuels markets. Each of
these parties produces, purchases, and sells different products within the E10 and B5 markets,
but, as this example demonstrates, no party has a structural advantage or disadvantage since both
the RIN cost and the RIN discount are passed through to consumers.
As previously discussed briefly, in reality very few parties fit entirely within only one of
these three categories. Most refiners, both small and large, sell some volume of petroleum fuel
(acting as merchant refiners) and blend some of their petroleum fuel with renewable fuel (acting
as integrated refiners). Some also purchase gasoline or diesel fuel from other parties and blend it
with ethanol to sell as E10 (acting as non-obligated blenders). Further, some refiners are also
renewable fuel producers that produce the renewable fuel they blend rather than purchasing it
from other parties and sell excess renewable fuel to others. Therefore, to better understand how
various parties are affected by the RFS program and RIN prices, it is better to consider the role
the party is playing in the fuels market (producing gasoline or diesel fuel, blending renewable
fuel, etc.) than the predominant role of the company.
To illustrate the impact of the RFS program and RIN prices on parties acting in each of
these roles, EPA evaluated scenarios with fuel prices, RIN prices, and RVOs as they existed on
December 30, 2020. EPA also evaluated an alternative scenario where there was no RFS
obligation. The fuel and RIN prices used in these scenarios, as well as the sources of these prices,
are shown in Table IV.D.2.C-1 for the E10 example and Table IV.D.2.C-3 for the B5 example.
The costs, revenue, and profit/loss for each party, both with and without the RFS program, are
shown in Table IV.D.2.C-2 for E10 and Table IV.D.2.C-4 for B5.
The 2011 DOE Study included a very similar hypothetical value breakdown for various
types of refiners in Appendix B of that study.156 At the time DOE projected that //integrated
refiners did not have to discount the E10 that they sell, then they could acquire RINs through
blending at little or no cost. In this hypothetical scenario, integrated refiners that acquired RINs
at little or no cost through blending renewable fuel would have a significant advantage relative to
merchant refiners that purchased RINs at a higher market price. However, as the examples below
illustrate, integrated refiners must compete with non-obligated blenders in the blended fuels
market. To offer competitively priced blended fuel, integrated refiners (like blenders) must
discount the price of the blended fuel by the price of the RIN attached to the renewable fuel
contained in the blended fuel. Market data reviewed by EPA confirms that the price of blended
156 See supra, Section II.D.
37
-------
fuel reflects the RIN discount.157 Thus, contrary to the hypothetical example in the 2011 DOE
Study,158 we find that all obligated parties have the same cost to acquire RINs, whether they
acquire RINs through blending renewable fuel or purchasing separated RINs.
Table IV.D.2.C-1: BOB159, Ethanol, E10, and RIN Prices on December 30, 2020
Product
Price
Data Source
BOB Cost of Production
$1.34
Assumed to be equal to the BOB Market Price
without RIN Cost
BOB Market Price without RIN
Cost
$1.34
Calculated (BOB Market Price with RIN Cost
less RIN Cost)
BOB Market Price with RIN Cost
$1.44
EIA
Ethanol Market Price
$1.50
OPIS
E10 Market Price with the RFS
Program
$1.37
Calculated using BOB Market Price with RIN
Cost, Ethanol Market Price, and D6 RIN Price
E10 Market Price without the
RFS Program
$1.36
Calculated using BOB Market Price without
RIN Cost and Ethanol Market Price
D6 RIN Price
$0.77
OPIS
RIN Cost per Gallon of BOB
$0.10
Calculated from 2020 RVO and OPIS RIN
Prices
D6 RIN Cost per Gallon of E10
$0.06
Calculated from 2020 RVO and OPIS RIN
Prices
D3, D4, and D5 RIN cost per
gallon of E10
$0.03
Calculated from 2020 RVO and OPIS RIN
Prices
157 See infra, Section IV.D.2.d.ii.
158 DOE's example in Appendix B of the 2011 DOE Study included a comparison of Company A that blends all its
production with ethanol and does not need to purchase ethanol RINs, with Company B that does not do any blending
and must purchase RINs to meet its entire RFS obligation, and Company C that blends in excess of its obligation
and has RINs to sell into the market. In DOE's hypothetical case, Company A acquired RINs at no cost (n/a in the
estimate) while Company B faced a 15 cent per RIN cost to purchase RINs. 2011 DOE Study at B-4.
159 BOB is an intermediate petroleum product that is used in making finished gasoline and is generally blended with
ethanol to make E10. BOB represents the petroleum-based portion of blended gasoline that has a RIN obligation
attached to it. Therefore, BOB can be used to show the price impacts of the RIN market on the petroleum
component of blended fuel.
38
-------
Table IV.D.2.C-2: Illustrative Costs, Revenue, and Profit for E10 Production
Line
Merchant
Refiner
Integrated
Refiner
Non-Obligated
Blender
With
RFS
No
RFS
With
RFS
No
RFS
With
RFS
No
RFS
2-1
0.9*BOB Cost of
Production
$(1.21)
$(1.21)
$(1.21)
$(1.21)
-
-
2-2
0.9*RIN Cost
$(0.09)
-
$(0.09)
-
-
-
2-3
0.9*BOB Market Price
$1.30
$1.21
-
-
$(1.30)
$(1.21)
2-4
0.1 *Ethanol Market Price
(with RIN)
-
-
$(0.15)
$(0.15)
$(0.15)
$(0.15)
2-5
0.1 *Net Ethanol Market
Price (no RIN)
-
-
$(0.07)
$(0.15)
$(0.07)
$(0.15)
2-6
E10 Market Price (per
Gallon)
-
-
$1.37
$1.36
$1.37
$1.36
2-7
D6 RIN Purchases
$(0.06)
-
-
-
-
-
2-8
D3, D4, and D5 RIN
Purchases
$(0.03)
-
$(0.03)
-
-
-
2-9
D6 RIN Sales
-
-
$0.02
-
$0.08
-
2-10
Profit/Loss per Gallon
E10
$0.00
$0.00
$0.00
$0.00
$0.00
$0.00
39
-------
Table IV.D.2.C-3: Diesel Fuel, Biodiesel, B5 and RIN Prices on December 30, 2020
Product
Price
Data Source
ULSD Cost of Production
$1.38
Assumed to be equal to the ULSD
Market Price without RIN Cost
ULSD Market Price without RIN Cost
$1.38
Calculated (ULSD Market Price with
RIN Cost less RIN Cost)
ULSD Market Price with RIN Cost
$1.48
EIA
Biodiesel Market Price
$3.66
OPIS
Biodiesel Tax Credit
$1.00
N/A
B5 Market Price with the RFS Program
$1.46
Calculated using ULSD Market Price
with RIN Cost, Biodiesel Market Price,
and D4 RIN Price, and Tax Credit Price
B5 Market Price without the RFS
Program
$1.44
Calculated using ULSD Market Price
without RIN Cost, Biodiesel Market
Price, and Tax Credit Price
D4 RIN Price
$1.00
OPIS
RIN Cost per Gallon of ULSD
$0.10
Calculated from 2020 RVO and OPIS
RIN Prices
D4 RIN Cost per Gallon of B5
$0.02
Calculated from 2020 RVO and OPIS
RIN Prices
D3, D5, and D6 RIN cost per gallon of B5
$0.07
Calculated from 2020 RVO and OPIS
RIN Prices
Table IV.D.2.C-4: Illustrative Costs, Revenue, and Profit for B5 Production
Merchant
Refiner
Integrated
Refiner
Non-Obligated
Blender
With
No
With
No
With
No
Line
RFS
RFS
RFS
RFS
RFS
RFS
4-1
0.95HJLSD Cost of
Production
$(1.31)
$(1.31)
$(1.31)
$(1.31)
-
-
4-2
0.95*RIN Cost
$(0.09)
-
$(0.09)
-
-
-
4-3
0.95*ULSD Market Price
$1.41
$1.31
-
-
$(1.41)
$(1.31)
4-4
0.05*Biodiesel Market Price
(with RIN)
-
-
$(0.18)
$(0.18)
$(0.18)
$(0.18)
4-5
0.05*Tax Credit
-
-
$0.05
$0.05
$0.05
$0.05
4-6
0.95*Net Biodiesel Price
$(0.06)
$(0.13)
$(0.06)
$(0.13)
4-7
B5 Market Price (per Gallon)
-
-
$1.46
$1.44
$1.46
$1.44
4-8
D4 RIN Purchases
$(0.02)
-
-
-
-
-
4-9
D3, D5, and D6 RIN
Purchases
$(0.07)
-
$(0.07)
-
-
-
4-10
D4 RIN Sales
-
-
$0.05
-
$0.07
-
4-11
Profit/Loss per Gallon E10
$0.00
$0.00
$0.00
$0.00
$0.00
$0.00
40
-------
The illustrative examples presented in Tables IV.D.2.C-2 and 4 demonstrate several
important points about the impact of the RFS program and RIN prices on merchant refiners,
integrated refiners, and non-obligated blenders. First, since the RIN cost (lines 2-2 and 4-2) and
the RIN discount (blended fuel prices based on net renewable fuel prices; lines 2-6 and 4-7) are
fully passed through to consumers, no party benefits or is harmed by the RFS program, either in
absolute terms or relative to their competitors.160 This can be seen in lines 2-10 and 4-11. In each
of the examples, the revenues and costs of various products change as a result of the RFS, but the
profit/loss and, thus, the potential harm for each of these three parties is identical with and
without the RFS.
Second, a merchant refiner's ability to recover its RIN costs in the price of the fuel it
produces does not depend on its ability to be a "price setter" or to receive a price for its fuel that
is above the market price. Instead, the market price for fuel increases to account for the RIN cost
associated with producing the fuel (RIN cost passthrough). Whether and the degree to which a
refiner is a "price setter" or "price taker" is not influenced by the RFS program. Rather, the RFS
program merely shifts upward the price at which this competitive dynamic is at play. This price
impact can be seen by comparing the market prices for gasoline and diesel fuel (lines 2-2 and 4-
2, respectively) with and without the RFS program. Merchant refiners automatically receive a
price for their fuel that reflects the cost increase due to the RFS program (i.e., the cost of the
RIN) when they sell the fuel at the market price.
Third, if a refiner (merchant or integrated) has a higher cost of production than the market
price without the RFS program, it will lose money for each gallon of fuel it produces. This is true
both in the absence of the RFS program and with the RFS program. Any party that has a higher
cost of production than the market price for the goods it produces will lose money when selling
those goods. However, the higher market prices for fuels can obscure these underlying
fundamentals. In the example presented in Table IV.D.2.C-1, if a merchant refiner's cost to
produce 0.9 gallons of gasoline is $1.30, it may appear that the refiner would break even by
selling gasoline at the market price (line 2-3) but for the RIN purchases (lines 2-7 and 2-8).
Several petitioners have made this very claim, that their refineries would be profitable if they did
not have to purchase RINs but are not profitable after accounting for their RIN costs. However,
such claims ignore the fact that in the absence of the RFS program, the market price for 0.9
gallons of gasoline (line 2-3) would fall to $1.21, resulting in a $0.09 loss. If a refiner's cost of
production exceeds the marginal supply price for its market, the refiner will lose money for every
gallon of fuel it produces due to its high cost of production, regardless of the presence or absence
of the RFS program. As demonstrated by the identical results for all parties in Tables IV.D.2.C-2
and 4, the RIN compliance costs associated with the RFS program do not have a differential
impact on the refiner's situation.
Fourth, while integrated refiners that do their own blending have the same cost to acquire
RINs as merchant refiners, they spend less on separated RIN purchases when they produce E10
or B5 (lines 2-7 and 4-8, respectively). Integrated refiners are acting both as merchant refiners
(producing fuel that carries an RFS obligation) and as blenders (blending renewable fuel and
separating the attached RINs) at the same time. However, rather than purchasing all the RINs
160 Throughout Section IV.D.2.C, references to "lines" are to Table IV.D.2.C-2 (lines beginning with 2-) and Table
IV.D.2.C-4 (lines beginning with 4-).
41
-------
they need from other parties or selling all the RINs they acquire through blending renewable
fuel, integrated refiners acquire many of the RINs they need for compliance from blending
renewable fuel rather than purchasing these RINs. The transfer of RINs from the blending
operation of an integrated refiner to the refining operation is an internal transfer, rather than an
external purchase or sale that is easier to see in financial reports. While it may appear that
integrated refiners are at an advantage relative to merchant refiners under the RFS program
because they purchase fewer RINs per gallon of fuel produced (see lines 2-7 and 4-8) than
merchant refiners, they also sell fewer RINs than non-obligated blenders (lines 2-9 and 4-10).
These two impacts—the higher RIN purchases relative to merchant refiners and the lower RIN
sales relative to non-obligated blenders—offset each other such that integrated refiners neither
benefit from the RFS program, nor are at a disadvantage relative to merchant refiners or non-
obligated blenders under the RFS program.
Another way to understand the impact of the RFS program on integrated refiners is to
consider the opportunity cost to these parties of selling blended fuel rather than petroleum fuel.
Integrated refiners are competing with non-obligated blenders when they sell blended fuel (lines
2-6 and 4-7). These blenders must discount the price of the blended fuel they sell because of the
revenue they realize when they sell the RINs associated with the renewable fuel (lines 2-9 and 4-
10). Integrated refiners generally keep the RINs they acquire when they blend renewable fuel, so
they do not have this revenue source to reduce the price of their blended fuel to compete with
blenders. Instead of revenue from RIN sales, integrated refiners can use their own production of
petroleum fuel, which has a lower cost of production than the market price for the fuel (lines 2-1
and 2-3 and lines 4-1 and 4-3), to produce blended fuel. Access to these lower-cost fuels allows
integrated refiners the ability to offer blended fuel at the same price as non-obligated blenders—
which use the revenue from RIN sales to discount the price of their blended fuel—despite the
fact that they use the RINs they acquire through blending for RFS compliance, rather than selling
them to other parties. In doing so they give up the opportunity to sell their petroleum fuel at the
higher market rate, which reflects the RIN cost (lines 2-2 and 4-2).
Fifth, the fact that refiners are able to recover the cost of the RINs they need for
compliance and that blenders pass through the RIN discount to consumers does not mean that the
RFS program has no impact on fuel prices. The RFS program functions as a cross-subsidy,
where RINs increase the market price of petroleum fuel (lines 2-3 and 4-3) and decrease the net
price of renewable fuel (lines 2-5 and 4-6). This means that the RFS program reduces the market
price for fuel with higher renewable fuel content (e.g., E85 or B20) and increases the market
price for fuel with little or no renewable content (e.g., E0 or BO). Notably, the RIN cost and the
RIN discount are not the same for all blended fuels. RIN costs (lines 2-2 and 4-2) are
proportional to the quantity of petroleum fuel in the blended fuel while the RIN value used to
discount the price of the renewable fuel is proportional to the quantity and type (D6 ethanol, D4
biodiesel, etc.) of renewable fuel in the blended fuel. In the two examples in Tables IV.D.2.C-2
and 4, the RIN cost and the RIN discount for E10 and B5 are very similar and as a result the
prices for E10 and B5 with and without the RFS program (lines 2-6 and 4-7, respectively) are
very similar. This is not the case for fuels with significantly higher or lower proportions of
renewable fuel.
42
-------
Finally, while non-obligated blenders realize revenue from RIN sales (lines 2-9 and 4-
10), this revenue is not a windfall profit. Instead, RIN revenues result in lower net prices for
renewable fuels (lines 2-5 and 4-6). The prices of the blended fuel (lines 2-6 and 4-7) then reflect
the lower net cost for the renewable fuel under the RFS program. For fuels such as E10 and B5,
when the RIN value of the renewable fuel in the blend is approximately equal to the RIN cost
associated with the petroleum fuel in the blend, it can be difficult to see the impact of the RFS
program in the blended fuel price. For fuels with significantly higher or lower renewable fuel
content, the impact is more pronounced. RINs decrease the price for fuel with a high renewable
content (e.g., B20 or E85), while RINs increase the price for fuel with little or no renewable
content (e.g., E0 or BO). This is the mechanism by which the RFS program was intended to
increase the production and use of renewable fuel in the United States.
In the calculations in Tables IV.D.2.C-2 and 4, we have made several simplifying
assumptions. First, we have assumed that the fuel cost of production for both the merchant
refiner and the integrated refiner (lines 2-1 and 4-1) is equal to the market price for the fuel
without the RFS program. In practice, the marginal cost to supply fuel to any given market sets
the market price. Each refiner's refining margin would, therefore, be determined by its actual
fuel cost of production relative to the market price for the fuel. RIN costs increase the market
price for the fuel by an amount equal to the RIN cost, since all parties have the same RIN costs.
However, since the market price for fuel reflects the RIN cost, the merchant refiner's profit/loss
is determined by its cost of production relative to the marginal cost of production for its market,
with or without the RFS program. Said another way, different refineries in a market will have
differing profit margins for the fuel they produce. But since the RFS compliance costs (i.e.,
RINs) apply equally to every gallon of fuel produced, these costs directly impact all gasoline and
diesel fuel volumes equally, raising the marginal supply price for these products. Thus, RIN
prices increase a refinery's costs and the market price for their production, but the difference
between the refining margins for the different refineries will remain the same with and without
the RFS program.
Similarly, in this example we have assumed no blending margin or cost for blending
beyond the purchase of petroleum fuel and renewable fuel. This is a simplification that does not
reflect the fact that, in addition to the cost of purchasing fuel, blenders also have operating costs
and fixed costs. These costs include, among others, labor costs, maintenance costs, and capital
recovery costs. Blenders must earn a margin when they sell blended fuel to cover these fixed and
operating costs, and the market price for blended fuel reflects the fixed and operating costs of the
marginal fuel blender. However, not all blenders will have the same fixed and operating costs.
Much like the previous example, we would expect a blender's (or integrated refiner's) profit/loss
for blending renewable fuel to be equal to its fixed and operating costs relative to the fixed and
operating costs of the marginal blender. Blenders and integrated refiners with relatively low
blending costs are expected to earn greater profits through blending, while blenders and
integrated refiners with relatively high blending costs are expected to earn relatively lower
profits (or losses) through blending. This is true independent of the RFS program, as RIN
costs/revenues are neutral. Notably, the design of the RFS program enables the market to
function efficiently by allowing those refiners that have relatively high fixed and operating costs
of blending renewable fuel to purchase RINs from blenders that have lower fixed and operating
costs of blending renewable fuel. We acknowledge this simplification and note that our decision
43
-------
to exclude a blending margin from the examples presented in Tables IV.D.2.C-2 and 4 does not
affect the conclusions highlighted above.
d. EPA Evaluation of Available Market Data
EPA analyzed the available market data to verify the economic principles at work and to
verify that the RIN cost and the RIN discount are being reflected in the retail price of blended
fuel.161 These analyses, including analyses conducted for previous assessments of the
passthrough of both the RIN cost and the RIN discount, as well as new analyses using more
recent data, are presented in this section. These analyses confirm that both the cost of the RINs—
which is reflected in the prices for fuel and blendstocks—and the discount of the RINs are passed
through to consumers in the marketplace in the price they pay for blended fuel.
i. Assessment of Data on RIN Cost Passthrough
EPA first assessed available data to determine whether refiners are able to recover the
cost of the RINs they need to demonstrate compliance with their RFS obligations through higher
prices for the petroleum fuel they produce, as described in Equations 1 and 2. This analysis is
complicated by the fact that the terms in Equations 1 and 2 for the gasoline price with no RFS
obligation and the diesel fuel price with no RFS obligation cannot be found in market data from
the United States, as the reported data will always reflect the cost of the RFS obligation. As
described below, however, there are market data on the prices of fuels that are very similar (and
in some cases identical) where one fuel has an RFS obligation and the other does not.
In 2015, EPA identified prices for near-identical fuels (in terms of technical fuel
specifications, and, therefore, presumably cost of production) except for the fact that one fuel
was subject to an RFS obligation while the other was not.162 We then used the price of the non-
obligated fuel to approximate what the cost of the obligated fuel would be in the absence of the
RFS obligation. We then compared the price difference between these two fuels, which
represents the increase in the market price of the obligated fuel as a result of its RFS obligation,
to the RIN cost for producing or importing a gallon of fuel subject to an RFS obligation. The
strong correlations between the price differences for similar fuels with and without an RFS
obligation and the RIN cost per gallon of obligated fuel led to the conclusion that the market
prices for gasoline and diesel fuel are higher than they would otherwise be in the absence of the
RFS program. Further, the observed price difference was equal to the cost of purchasing the
RINs needed to meet the compliance obligations for a gallon of gasoline or diesel fuel. We
therefore concluded that all refiners recovered the full cost of the RINs they purchase through the
prices of the fuel they sell.
EPA has since repeated the analytical techniques first developed in 2015 using more
recent data from 2017-2020. Figure IV.D.2.d.i-l shows the price difference in New York Harbor
between ULSD, which is subject to an RFS obligation, and heating oil, which is essentially an
identical product except that it is not subject to an RFS obligation. As expected, there is a very
161 See supra, Section IV.D.2.b.
162 -A Preliminary Assessment of RIN Market Dynamics, RIN Prices, and Their Effect," Dallas Burkholder, Office
of Transportation and Air Quality, US EPA, May 14, 2015.
44
-------
strong correlation between these data sets, as shown in Figure IV.D.2.d.i-2. The market price
premium for ULSD over that for heating oil consistently matches the RIN cost (i.e., the cost of
purchasing the RINs needed to meet the RFS obligation).
Similarly, Figure IV.D.2.d.i-3 shows the price difference in the Gulf Coast between
ULSD, which is subject to an RFS obligation, and jet fuel, which is not. However, as shown in
Figure IV.D.2.d.i-4, the correlation between the price difference of ULSD and jet fuel and the
RIN cost is not as strong as the correlation between the price difference of ULSD and heating oil
and the RIN cost. This is to be expected, as there are more significant product quality differences
between ULSD and jet fuel such that they are not one-for-one replacements of each other, and
they are used primarily in different markets with distinct supply/demand issues that would also
contribute to differences in their market prices.163 Thus, there is more noise in these data, but a
general relationship between the price difference among these fuels and the RIN cost can be
seen. Also apparent in Figure IV.D.2.d.i-3 is the impact of the COVID-19 pandemic. In late
March 2020 air travel and demand for jet fuel decreased dramatically, resulting in an over-supply
of jet fuel and a spike in the price premium for ULSD over jet fuel.164 Over time, as demand for
jet fuel gradually increased and refiners adjusted their production to better match fuel demand,
the price difference between jet fuel and ULSD returned to match the RIN cost. Taken together,
these more recent data confirm EPA's original conclusion that the market prices for gasoline and
diesel fuel reflect the RIN cost, and, therefore, all refiners are able to recover their RIN costs
through the sales prices of these fuels.
163 Jet fuel generally contains more sulfur than ULSD. While the properties of jet fuel are closer to #1 diesel than to
#2 diesel, EPA's public data does not contain prices for #1 diesel.
164 EIA, COVID-19's impact on commercial jet fuel demand has been significant and uneven, Today in Energy
(August 7, 2020), littps ://www,eia. gov/todav inenergv/detai t. php?id=44676.
45
-------
Figure I V.D.2.d.i-l: Price Difference Between ULSD and Heating Oil in New York Harbor
and RIN Cost (2017-2020)165
UL
$0.20
$0.15
SD - Heating Oil and RIN Value (NYH)
p
1/3/2017
$(0.05)
1/3/2018 1/3/2019 1/3/2020
ULSD - Heating Oil RIN Cost
\
Figure IV.D.2.d.i-2: Correlation Between Price Difference of ULSD and Heating Oil and
RIN Cost (2017-2020)
$0.20
$0.15
$0.10
c
ro
CD .
X $0.05
Q
55
${0.05)
4
t
i.
R
d
0
S&0E
• • •
$0
02 $0
04 $0
•
#6 $0
OS $0
10 $0.
RIN Cost
165 Prices for ULSD and heating oil are reported by EIA and are available at:
https://www.eia.gov/dnav/pel/pet pri sot si d.htm.
46
-------
Figure IYf.D.2.d.i-3: Price Difference Between ULSD and Jet Fuel in the Gulf Coast and
RIN Cost (2017-2020)166
1/3/2017
$(0.10)
$(0.20)
$(0.30)
1/3/2020
¦ ULSD-Jet Fuel
»RIN Cost
Figure IV.D.2.d.i-4: Correlation Between Price Difference of ULSD and Jet Fuel and RIN
Cost (2017-2020)
u
Q
in
$0.40
$0.30
$0.20
$0.10
$-
$(0.10)
$(0.20)
$(0.30)
March - May 2020
( -.-v*. 1
V «•<
J
'* *jfA
•
1 $0,
!o2^P40.
.04 $0,
.06 $0,
.08 $Jl0 $0.
i
4
t
~
•
RIN Cost
In their SRE petitions, several small refineries submitted examples of fuel pricing
contracts in their local markets. Notably, many of these contracts indexed the sales price for fuel
in the typically smaller markets into which the small refineries sell fuel to larger fuels markets,
usually with the addition of transportation costs. The structure of these contracts supports EPA's
166 prjccs for ULSD and jet fuel are reported by EIA and are available at:
https://www.eia.gov/dnav/pei/pet pri spl si d.htm.
47
-------
finding that the inclusion of the RIN cost in the price of obligated fuel is not unique to larger,
coastal fuels markets, but is true across the United States. If the RIN cost is reflected in the sales
price of fuel in New York Harbor and the Gulf Coast, it is certainly reflected in markets,
including smaller markets, that index their pricing to these larger markets. Furthermore, because
of the highly connected and competitive nature of fuels markets across the United States, one
would expect every fuels market to reflect these same pricing dynamics. To date, no petitioning
small refinery has provided EPA with data nor have we found other data that is in conflict with
this expectation. In fact, small refineries that participate in both larger markets and smaller
markets have consistently highlighted to EPA that they are in direct competition with larger and
better resourced refineries regardless of their location. Even in cases where the small refineries
themselves may not distribute fuel beyond a relatively small geographic area, the large integrated
refiners with which they compete in those local markets do sell fuels into the larger distributed
markets. It would not make economic sense for these large integrated refiners, which have access
to larger fuels markets where market prices reflect the cost of RINs, to choose to sell into the
smaller markets occupied by small refineries unless the market prices in those smaller markets
also reflected the RIN cost.
Another important observation from these data is that neither the RIN cost nor the
additional revenue a refiner receives for an obligated fuel compared to a non-obligated fuel (the
premium for obligated fuel versus a similar non-obligated fuel) are static. There has been
significant variation in these prices from 2017-2021, from approximately $0.10 per gallon in late
2017 and late 2020, to a low of approximately $0.03-0.04 per gallon throughout 2019. RIN
prices have continued to increase in 2021, with prices for most RIN categories currently 50-
100% greater than RIN prices at the end of 2020 (see Figure IV.D.2.d.i-5).
Figure IV.D.2.d.i-5: RIN Cost Per Gallon by RFS Category (2011-2020)
$0.16
<3J
£ $0.14
"O
I 50.12
i * 44 J
MMM
g $0-02 , | ^ M
E tf' qV1
a? a? >.v
J? ^ ^ ^ ^ J? J? ^ ^
¦ D6 RIN Cost ¦ D5 RIN Cost ¦ D4 RIN Cost ¦ D3 RIN Cost
Obligated parties that choose to purchase the RINs they need for compliance on a ratable
basis (i.e., purchase on a systematic, regular basis the number of RFNs needed to satisfy their
obligation for all the fuel sold each day) will recover the cost of the RINs they purchase in the
48
-------
sales price of the petroleum fuel they sell. Conversely, obligated parties that choose to delay RIN
purchases, or to purchase excess RINs in advance of producing or importing petroleum fuel, may
recover more or less than the price they paid for RINs in the sales price of the petroleum fuel
they sell, depending on whether the RIN price on the purchase date is higher or lower than the
RIN price on the date the petroleum fuel is sold. For example, based on the data presented in
Figures IV.D.2.d.i-l and 3, an obligated party that sold fuel in July 2020 received approximately
$0.06 per gallon more than it would have in the absence of the RFS program. If that obligated
party delayed purchasing RINs until the end of 2020, the RIN cost would have been
approximately $0.10 per gallon. Conversely, if the obligated party had purchased excess RINs in
January 2020, the RIN cost would have been approximately $0.03 per gallon. Thus, the decision
to delay RIN purchases until December 2020 would have cost an obligated party an additional
$0.04 per gallon of fuel produced in July 2020; whereas purchasing excess RINs in January 2020
would have resulted in an additional $0.03 per gallon profit for every gallon of fuel produced in
July 2020. By purchasing RINs ratably, all obligated parties have the ability to match their RIN
costs with the price they receive when they sell their fuel (i.e., to pass through their RIN costs).
Alternatively, refineries can try to time their purchases in the RIN market, which may result in
greater or lesser RIN costs. EPA strongly disputes any notion that costs resulting from individual
refinery's business decisions, including the choice to delay RIN procurement in hopes of
receiving an SRE, or an attempt to time the transaction to profit from the fluctuation in the RIN
market prices over time, represent DEH caused by the RFS program. Rather, individual business
decisions made by an obligated party not to ratably accrue RINs as the obligation accrues, but
instead to either purchase RINs in advance or delay RIN purchases until a later date, are
speculation in the RIN market, a business activity not required to comply with the RFS program.
Thus, these costs cannot be considered to represent DEH caused by compliance with the RFS
program.
ii. Assessment of Data on the RIN Discount
To verify that fuel blenders are passing through the RIN discount to consumers through
the price of blended fuel as described by Equations 3 and 4, EPA considered information from a
variety of sources. We evaluated the issue by analyzing market pricing data for petroleum fuel,
renewable fuel, RINs, and blended fuel (including data submitted by petitioners), statements
from blenders in publicly available earnings reports, and fuel pricing contracts submitted by
petitioners. Each of these data sources support EPA's finding that revenue from RIN sales does
not represent a windfall profit for fuel blenders. Rather, they demonstrate that blenders pass
through the full value of the RIN to consumers in discounts on the price of the blended fuel they
sell and, therefore, do not retain any revenue from the sale of RINs.
There are a limited number of markets where prices for each of these fuels are reported,
but all of those we have seen confirm our conclusions that fuel blenders are passing through the
RIN discount to consumers through the price of blended fuel.167 In 2015, EPA analyzed market
data from Des Moines, Iowa and demonstrated that there was a very strong correlation between
the difference in the posted price for E10 in Des Moines and the calculated E10 price based on
167 This same point was raised in one small refinery's petition, along with data to illustrate it. The small refinery
claimed its petition and all supporting information as confidential business information.
49
-------
the component fuels (gasoline blendstock and ethanol), and the RIN price per gallon of E10.168
These data indicated that fuel blenders are selling blended fuel based on the net price of the
renewable fuel (after accounting for the sale of any associated RINs). This means that the price
of the blended fuel was lower than the cost to purchase the components of the fuel blend
(gasoline blendstock and ethanol with a RIN) and that revenue from RIN sales offset these costs.
The result of this pricing behavior is that 100% of the revenue from RIN sales was passed on to
consumers.
Two petitioning small refineries submitted data to EPA on fuel prices in their markets
that enabled EPA to analyze current data in additional markets using a methodology similar to
the analysis we conducted for Des Moines in 2015.169 One small refinery used monthly gasoline
and ethanol pricing data from a local terminal, along with RIN pricing data, to determine a
monthly calculated E10 price from 2010 to the present using an equation nearly identical to
Equation 2.170 That petitioner then plotted these calculated E10 prices, which assume that 100%
of the RIN value is passed through to consumers through lower prices for blended fuel, against
the posted prices for E10 at that same terminal. The petitioner found an extremely strong
correlation (R2 = 0.9976) between the calculated E10 price (assuming 100% RIN passthrough)
and the posted E10 price, demonstrating for this terminal that the RIN value has been fully
passed through to consumers since 2010.171
Another petitioning small refinery's fuel pricing data allowed EPA to conduct a similar
analysis for yet another market.172 This small refinery provided daily pricing information for E10
from a local terminal, as well as daily pricing information for gasoline blendstock and ethanol
from a nearby market along with the cost to transport these fuels to the refinery's local market.
Daily prices were provided from January 1, 2019 through June 21, 2021. EPA used the data to
calculate an E10 price using Equation 2 and compared these calculated E10 prices (assuming the
E10 price was based on the net price of the ethanol, passing through 100% of the RIN in the
discounted price of E10) to the posted E10 prices at the local terminal. As with the data provided
by the other petitioning small refinery, we again find an extremely strong correlation (R2 =
0.9991) between these two prices, further confirming our previous findings that the RIN price is
fully passed through to consumers as a discount on the price of the renewable fuel when
petroleum fuel and renewable fuel are blended and then sold.
Support for EPA's finding that the RIN discount is fully reflected in the price of blended
fuels and is accordingly passed through to consumers by fuel blenders can also be found in
168 -A Preliminary Assessment of RIN Market Dynamics, RIN Prices, and Their Effect," Dallas Burkholder, Office
of Transportation and Air Quality, US EPA, May 14, 2015.
169 We do not present the data here because the petitioners have claimed it contains CBI.
170 The only difference between Equation 2 and the equation used by the petitioner to determine the calculated E10
price was that the petitioner included an additional terminaling and throughput charge that applies regardless of the
RFS program and is not relevant to this discussion.
171 This petitioner acknowledged that the RIN was used to discount the price of blended fuel at their terminal.
However, the petitioner further argued that the RIN cost could not be recovered in the cost of the gasoline and used
to discount the price of the blended fuel. As discussed further in Section IV.D.2.C, both the economic principles and
the market data demonstrate that this is incorrect. Refiners recover the cost of the RIN through the sales of their
petroleum fuel and the RIN is used to discount the price of blended fuel.
172 We do not present the data here because the petitioners have claimed it contains CBI.
50
-------
public statements by the blenders themselves. Several parties directly involved in fuel blending
supported EPA's findings in comments173 on EPA's Point of Obligation denial.174 More recently,
R. Andrew Clyde, President, CEO & Director of Murphy USA, a large fuel blender and retailer,
was asked if the recent high RIN prices positively affected Murphy USA's margins in a Q1 2021
earnings report. He responded:
The reality is RINs and RIN prices are immaterial to our business. Historically, and you
can look back over the last 3 years annual results, we've made $0.02 to $0.03 per gallon
on product supply and wholesale net of RINs. And so during the quarter on the average,
we generated about the equivalent of $0.07 a gallon per RIN, but net of the negative spot
to rack margins of $0.04, we netted a little bit over $0.03.. If RINs are high, the refiner
gate price is high and like it was in this quarter, our refinery gate spot to rack margin is
negative... So RIN prices don't matter. The product supply margin plus the RINs is going
to be about $0.02 to $0.03.175
Mr. Clyde describes a market dynamic wherein blenders experience negative blending
margins (due to competitive market forces requiring that the RIN price be reflected in the market
price of blended fuel) that are offset by revenue from selling RINs, with total margins (including
fuel blending and RIN sales) relatively stable and independent of RIN prices.176 These dynamics
are exactly what one would expect to see if blenders are passing through 100% of the RIN price
as a discount to consumers in the price of blended fuel.177
Several petitioning small refineries also provided EPA with examples of contracts for
fuel sales.178 While there were some differences among these contracts, they generally showed
that the sales price for blended E10 was discounted by the value of the RIN associated with the
ethanol blended into the fuel blend. Many of the pricing formulas shown in these contracts
looked very similar to Equation 4, with some referencing petroleum fuel and/or ethanol prices in
nearby markets and including transportation costs. In some cases, the contracts stipulated that the
purchase price would be the lower of the calculated price based on the prices of the petroleum
fuel and the net price of ethanol (thus passing through 100% of the RIN price to consumers) or
the posted price of El 0 at the local terminal, whichever was lower. These contracts provide yet
more evidence that the price of the RIN is reflected in the sales price for blended fuel, and
further that the passthrough of the RIN price to consumers is not limited to any particular U.S.
market.
173 See Letter from RaceTrac to Administrator McCarthy, August 17, 2016, Docket Item No. EPA-HQ-OAR-2016-
0544-0014; Letter from QuikTrip to Administrator McCarthy, August 17, 2016, Docket Item No. EPA-HQ-OAR-
2016-0544-0013; Presentation from Murphy USA to EPA, August 16, 2016, Docket Item No. EPA-HQ-OAR-2016-
0544-0028.
174 81 FR 83776 (November 22, 2016) and 82 FR 56779 (November 30, 2017).
175 Murphy USA Inc. FQ1 2021 Earnings Call Transcripts (April 29, 2021).
176 Petitioners' claims of "RIN theft" and windfall profits from RIN sales by Murphy USA and other blenders are
further addressed in Section IV.D.2.a.
177 See supra, Section IV.D.2.b.
178 We do not present the contract data here because the petitioners have claimed it contains CBI.
51
-------
3.
EPA Responses to Small Refinery Arguments for Exemption
The petitioning small refineries raise many similar arguments in their petitions and in
supplemental information they submitted to support receiving an exemption from their RFS
obligations. Because these arguments are repeated by most, if not all, SRE petitioners, EPA is
addressing them in this section at a level of generality needed to maintain the claims of CBI
asserted by the small refineries in their respective petitions. The refineries generally argue eight
overarching themes in their petitions and supplemental information. However, EPA recognizes
that this list is not comprehensive. Some refineries provided information and perspective that we
do not address in this proposal, primarily because we are still seeking the underlying data or
additional explanation to understand and respond to the individual or market-specific
circumstances the petitioners describe.
The general themes small refineries have articulated are: (a) They face unique challenges
that prevent them from achieving RIN cost passthrough and that EPA must consider their
specific circumstances; (b) EPA's Point of Obligation denial did not address their situations and
does not apply to them; (c) The Point of Obligation denial is out of date and inapplicable; (d) The
revenue from RIN sales allows large retailers to undercut small refineries; (e) Large integrated
refiners set prices in fuels markets, undercutting small refineries on price because of their market
position and because large integrated refiners have lower or no RIN costs; (f) EPA is incorrect
about parity between the cost of obtaining a RIN through blending and the cost of buying a RIN
on the market; (g) Single site refineries are disadvantaged relative to large integrated refiners
because they only have access to a limited market; and (h) Small refineries that produce
primarily diesel fuel are at a disadvantage since they cannot blend as much renewable fuel into
their product as can refineries that produce gasoline.
EPA evaluates and responds to each of these general themes below.
a. Small refineries face unique challenges that prevent them from passing through
their RIN costs. EPA must consider each small refinery's specific situation.
Small refineries assert that "EPA must do more than cite to the Burkholder Report's
conclusion 'that the refining industry as a whole is not burdened by rising RIN prices because
refineries may pass that cost to purchasers of the blended fuel.' Ergon-W. Va., Inc. v. EPA, 896
F.3d 600, 613 (4th Cir. 2018) (emphasis added)."179 The small refineries further assert that EPA
has, in the past, ignored information specific to individual refineries that demonstrates that they
cannot pass through the prices they pay for RINs due to unique operational or local market
circumstances.
The small refineries misstate the holding from EWV-I and completely ignore the follow-
up to that case, EWV-II. The court in EWV-I held that EPA had acted arbitrarily and capriciously
when it "failed to squarely address Ergon's petition with regards to RIN costs"180 and instead
relied on the Burkholder memo "as the sole basis for its conclusion."181 (emphasis added). The
179 Confidential submissions by several small refineries made this assertion.
180 EWV-I, 896 F3d at 613.
181 EWV-II, 980 F.3d at 417, rev'd on other grounds.
52
-------
court found that EPA was not arbitrary and capricious in relying on the Burkholder memo as one
of many factors considered in the decision, but rather, failed to adequately illustrate how the
analysis in that study applied to the circumstances at a particular small refinery (Ergon-West
Virginia). On remand, EPA reached the same conclusion as in its first decision and again faced
challenge from Ergon before the Fourth Circuit. The court, in EWV-II, reviewed EPA's post-
remand denial, which again relied heavily on the Burkholder memo, and found that "EPA's post-
remand discussion of Ergon's evidence connected the dots left unaddressed in its original
decision[,]" because "EPA thoroughly discussed Ergon's purported evidence of hardship,
explained why it rejected Ergon's arguments, and set out other factors that led it to reach an
opposite conclusion."182 Accordingly, in this proposed action, EPA has again evaluated the
question in depth, starting with an evaluation of the underlying structure of the RFS program and
RIN system to ascertain whether and how it might be possible for compliance with the RFS
program to cause DEH. EPA then conducted a careful analysis of how the cost and value of
RINs would be expected to flow through to consumers, and analyzed a substantial amount of
data, including available local market-specific data, that show how the findings in the
Burkholder memo regarding the refining industry as a whole are true for all obligated parties,
including small refineries in general and individual small refineries in particular.183 However,
due to the confidential nature of much of the information included in SRE petitions, we cannot
present any refinery-specific data in this proposal. Nevertheless, we have reviewed the
information in the SRE petitions, and nothing presented in them leads us to conclude that the
small refineries are affected by RFS compliance differently than other obligated parties or that
they are not able to pass along the RFS compliance cost to consumers. However, to ensure we
base our decision on the best data available, we are in this proposed action requesting additional
market data, including additional data specific to any petitioning small refinery.
The small refineries also state that there are many diverse factors that affect each
refinery's profitability and ability to recover the full cost of fuel production, including their RFS
compliance costs. The small refineries cite to the 2011 DOE Study to support their assertion,
quoting the following language:
The degree to which the costs burdening small refineries will be passed through to the
market depends on many factors, including the market power and the relative cost level of a
small refiner relative to other market participants.... The cost for small refiners to comply with
the RFS2 requirements can be substantial.... Their limited product slates coupled with an
inability to blend renewable fuels means that many of the small refiners must enter the market to
buy RINs. The cost to meet their individual RVO makes this aspect the most significant cost of
compliance.184
EPA understands that there are, indeed, many diverse factors that affect refinery
profitability; however, RFS compliance costs are not among them.185 The metrics DOE included
in the scoring matrix were only intended to identify small refineries that may face adverse
182M
183 See supra, Section IV.D.2.
184 2011 DOE Study at 22-23.
185 See supra, Section IV.D.2.
53
-------
business conditions,186 and therefore potentially face DEH if they also experienced "a high cost
of compliance relative to the industry average."187
The small refineries fail to acknowledge the fact that they may not be profitable or able to
pass through the full cost of their fuel production despite their RIN costs being passed through. It
is important to reiterate that independent market analyses, as well as EPA's own, support the
premise that RIN costs are incorporated into the price of finished fuels.188 This is to say that even
without RFS compliance costs, these small refineries may not be profitable. This kind of
economic hardship is not caused by the RFS program, but rather, by the refinery's business
model, geographic location, or other factors independent of the RFS program. The CAA only
speaks in terms of DEH caused by compliance with the RFS program. Congress tied SREs to
compliance with the RFS program by using the language "compliance with the requirements of
paragraph (2) would impose a [DEH]"189 and "would be subject to a [DEH] if required to
comply with paragraph (2)."190 The CAA does not authorize or require EPA to subsidize through
compliance exemptions any refinery whose economic hardship is not caused by compliance with
the RFS program no matter the seriousness of the economic conditions the refinery may face,
particularly since the magnitude of the RIN cost per gallon in comparison to typical refinery
margins could turn the least profitable refineries into the most profitable ones.191
Additionally, the DOE language the small refineries quote comes from the "[o]ther
observations from the interview process,"192 which DOE "compiled through interviews with
several industry participants, including two refineries, three importers, a fuel marketer, and a
corn ethanol marketer."193 This section does not state DOE's own conclusions, but rather
summarizes what DOE heard from the stakeholders it reached out to. This language cannot be
treated as DOE's findings, but rather, DOE's statement of the input it solicited and considered.
EPA believes the conclusions in the Burkholder memo are applicable to all gasoline and
diesel fuel markets nationwide, and, therefore, also applicable to all refineries, including small
refineries.194 Nevertheless, some petitioning small refineries have provided refinery-specific
supplemental information attempting to explain why the conclusions in the Burkholder memo
don't apply to them. EPA has analyzed the supplemental information and found no evidence
supporting the assertions from the petitioning small refineries that their RFS compliance costs
are disproportionately greater than for other refineries or that they are not able to pass along the
RFS compliance cost to consumers.195 In fact, the data petitioners provided to EPA reflected the
price behavior for both RINs and finished fuels that EPA would have expected based on
186 The adverse business conditions considered are quite varied and not directly related to the RFS compliance cost.
187 2011 DOE Study at 3.
188 See supra, Section IV.D.
189 CAA section 21 l(o)(9)(A)(ii)(I), paragraph (2) refers to the section where Congress provided the annual
applicable renewable volume mandates.
190 CAA section 21 l(o)(9)(A)(ii)(II).
191 See supra, Section IV.D.2.b. See also infra, Section IV.D.3.e.
192 2011 DOE Study at 22.
193 Id. at 21.
194 See supra, Section IV.D.2.
195 See supra, Section IV.D.2.
54
-------
economic principles.196 Additionally, other stakeholders with interest and expertise in RIN
market behavior and RFS compliance have provided support for and approved of EPA's analysis
and conclusions regarding RIN cost passthrough.197
b. The small refineries' situations are distinguishable from the findings provided in the
Point of Obligation denial, and the Point of Obligation denial did not address small
refineries.
Petitioners claim that EPA's assessment of RIN cost passthrough in the Point of
Obligation denial covered three categories of parties: integrated refiners, non-obligated fuel
blenders, and merchant refiners. The petitioners note that small refineries as a group do not fit
neatly within any of these categories. They further claim that EPA's conclusions about merchant
refiners' ability to recover their RIN costs were based on representations from Valero, which
they note is a large, international refiner with efficiency, geographic range, and pricing power.
The petitioners state that while these types of merchant refiners may be able to recover the cost
of purchased RINs, small refineries without these characteristics cannot.
EPA recognizes that few, if any, small refineries (or any refineries) fit neatly into a single
category of integrated refiner, non-obligated blender, and merchant refiner.198 Rather, we explain
that refiners, whether large or small, may operate as an integrated refiner, non-obligated blender,
and/or a merchant refiner in various fuels markets and in different aspects of their business
operations. EPA demonstrates that because both the RIN cost and the RIN discount are
ultimately passed through to consumers for all three categories, the RFS program does not
advantage or disadvantage any of these parties over the others, regardless of how much of their
operations fall into one or more of these categories. Importantly, a small refinery's ability to
recover its RIN costs in the price of the fuel it produces does not depend on factors such as
geographic range or pricing power.199 Instead, the data and analysis EPA presents demonstrate
that the market prices for both refined products and blended fuel reflect the cost of acquiring the
RINs necessary to satisfy the RFS obligation associated with the fuel. Merchant refiners do not
need to exercise market power and demand a price that is higher than the market price to recover
their RIN costs; all parties selling into these competitive markets are recovering the cost of
acquiring RINs when they sell their fuel at the market price. Thus, although size and market
power can be an advantage for reasons other than RFS compliance, they provide no advantage to
non-small refineries in recovering their RFS compliance costs.
196 See supra, Section IV.D.2.
197 See supra, Section IV.D.2. See also Letter from RaceTrac to Administrator McCarthy, August 17, 2016, Docket
Item No. EPA-HQ-OAR-2016-0544-0014; Letter from QuikTrip to Administrator McCarthy, August 17, 2016,
Docket Item No. EPA-HQ-OAR-2016-0544-0013; Presentation from Murphy USA to EPA, August 16, 2016,
Docket Item No. EPA-HQ-OAR-2016-0544-0028. See also Comments from API on 2020 RFS Annual Rule, Docket
Item No. EPA-HQ-OAR-2019-0136-0721.
198 See supra, Section IV.D.2.C.
199 See infra, Section IV.D.3.e.
55
-------
c. EPA's assessment in the 2017 Point of Obligation Denial is out of date and not
applicable in 2019 or 2020.
Many petitioners state that EPA could not rely on the conclusions of the assessment
conducted in 2017 in the context of the Point of Obligation denial to evaluate their recent
petitions. The petitioners state that the information considered in 2017 is now out of date and
does not reflect the present realities of the fuels market.
We believe that the analyses conducted in 2017 continue to inform our understanding of
the ways in which the RFS program affects small refineries and other fuels market participants,
and we are not reconsidering or taking additional comments on EPA's decision not to initiate a
rulemaking to change the point of obligation in the RFS program. The fact that the data reviewed
in 2017 was consistent with what would be expected based on the design of the RFS program
with its RIN system and economic principles is strong evidence that it is highly unlikely that the
RFS program will cause DEH, and is strong evidence that the conclusions in that action remain
true today. Our finding in that decision that the fuels market operates as we would expect in a
competitive market remains relevant. As long as the fuels and RIN markets remain competitive,
we do not anticipate that the RFS program will cause DEH to small refineries.
Nevertheless, in this proposed decision, we have considered more recent data since
2017—including data the small refinery petitioners themselves submitted—and we find that the
more recent data are consistent with the data EPA reviewed in 2017.200 These data continue to
support our finding that both the RIN cost and the RIN discount are passed through to consumers
and continue to show that the RIN market works in the same way for all market participants,
including individual small refineries.
d. Revenue from RIN sales allows large retailers to undercut small refineries.
Petitioners claim that EPA had not considered clear evidence that revenue from RIN sales
enabled large retailers such as Murphy USA to undercut the small refineries they compete with
that are unable to sell RINs for a profit. The petitioners argue that large retailers (which are
generally not obligated parties) can sell blended fuel at a lower cost than the cost of the
petroleum fuel and renewable fuel they are composed of because of the revenue they receive by
selling RINs. Small refineries must price their blended fuel at the same price as large retailers to
be competitive, but they do not receive the benefit of revenue from RIN sales.
Contrary to the petitioners' claims, EPA has considered the ability for non-obligated
blenders to sell RINs and to use the RIN sales revenue to discount the price of blended fuel while
remaining profitable.201 We present an illustrative example of how RIN prices affect integrated
refiners (which is the role small refineries are taking in the fuels market when they are blending
the petroleum fuel they produce with renewable fuel) and non-obligated blenders in Section
IV.D.2.C. As shown in Tables IV.D.2.C-2 and 4, neither integrated refiners nor non-obligated
blenders benefit from, or are harmed by, higher RIN prices.
200 The data, and the conclusions we have drawn from the more recent data, are presented in Section IV.D.2.d.
201 See supra, Section IV.D.2.
56
-------
The petitioners' description of blenders using revenue from RIN sales to enable them to
offer lower prices for the blended fuel they sell is consistent with EPA's findings (i.e., the RIN
discount).202 We also recognize that competitive forces require small refineries selling blended
fuel to sell at the market price (which reflects the passthrough of the RIN price as a discount to
consumers). In their claims about the advantages that the RFS program provides to non-obligated
blenders, however, the petitioners have not considered the impact of RIN prices on the market
price for fuels.
When small refineries produce and sell blended fuel from the petroleum fuel that they
produce, they are acting as integrated refiners for that volume of fuel. Generally speaking,
integrated refiners are not able to sell the RINs associated with the renewable fuel they blend, as
they need these RINs to meet their RFS obligations. But unlike non-obligated blenders,
integrated refiners do not typically purchase petroleum fuel to produce blended fuel; instead,
they are producing the petroleum fuel themselves. This means that for an integrated refiner, the
cost of the petroleum fuel is not the market price for these products (which reflects the marginal
cost of production of the fuels plus the cost of purchasing the RINs needed to satisfy the RFS
obligation associated with the fuel), but rather simply the cost of production for the petroleum
diesel fuel. The lower cost of the petroleum fuel relative to the market price for these products
allows the integrated refiner to price its blended fuel competitively with non-obligated blenders
and still maintain a positive margin for producing blended fuel even though they do not realize
revenue from RIN sales.203
Both the economic principles and the data EPA reviewed support our finding that the
RFS program does not advantage non-obligated blenders over integrated refiners. While RIN
sales provide an additional source of revenue for non-obligated blenders, this is offset by the
higher price (which reflects the RIN cost) for the petroleum fuel that the blenders pay to
merchant refiners to produce blended fuel. Integrated refiners, who are producing petroleum fuel
rather than purchasing them at the market price, have access to lower cost petroleum fuel but do
not realize revenue from RIN sales. Thus, while the RFS program impacts these parties in
different ways, neither enjoys an advantage or disadvantage over the other.
e. Large integrated refiners set the prices in fuels markets, undercutting small
refineries on price because of their market position and because the large,
integrated operations have no or lower RIN costs.
Petitioners claim that they compete in markets with large integrated refiners, and that
they have no market pricing power relative to these parties. Petitioners also state that, because
these large integrated refiners have no or lower RIN costs, they are able to undercut small
refineries when they price their product. They further note several other advantages that large
integrated refiners have relative to small refineries, such as a broader range of assets, economies
of scale, and access to more fuels markets (including exports). We address each of these points in
turn.
202 See supra, Section IV.D.2.
203 A further description of the impact of the RFS program on merchant refiners, integrated refiners, and non-
obligated blenders is provided in Section IV.D.2.c.
57
-------
The market for gasoline and diesel fuel in the United States is extremely competitive.204
EPA's finding that merchant refiners are able to pass through their RIN costs through higher
market prices for the fuel they produce does not depend on merchant refiners having market
pricing power in the markets where they sell fuel. Rather, we find that the market price for fuel
reflects the RIN value, and therefore all parties in all markets that sell fuel recover their RIN
costs when they sell their fuel (RIN cost passthrough).
In Section IV.D.2.C, EPA presented an example of the impact of higher RIN prices on
merchant refiners, integrated refiners, and non-obligated blenders, and discussed the impact on
each of these parties. In short, integrated refiners spend less money to purchase RINs than
merchant refiners; unlike the non-obligated blenders they are competing with in the blended
fuels market (i.e., large fuel retailers without refining or import businesses), they do not benefit
from revenue from RIN sales. Merchant refiners do benefit from the higher market prices for
gasoline and diesel fuel that are the result of higher RIN prices, but they must use this additional
revenue to purchase RINs. Said another way, there is an opportunity cost when these integrated
refiners blend renewable fuel with the petroleum fuel they produce instead of selling it
unblended, because these parties sell blended fuel for a lower price than they could sell the
petroleum fuel. This opportunity cost is equal to the savings these parties experience from
acquiring RINs by blending renewable fuel rather than purchasing separated RINs.
The many factors mentioned by the petitioners, such as a broader range of assets
(upstream, downstream, etc.), economy of scale, and access to more fuels markets, may in fact
provide a competitive advantage to large integrated refiners. However, the fact that small
refineries have continued to remain in the marketplace and compete with large integrated refiners
is evidence of the fact that small refineries typically have other market advantages, such as
access to local crude supplies and local markets lowering their distribution costs, specialty
products, and niche markets with fewer competitors. None of these market advantages and
disadvantages are the result of the RFS program. Each of these factors offered potential
advantages (and potential liabilities) before the RFS program existed and continue to do so
today. The petitioners have not presented any evidence, nor is EPA aware of any evidence, that
would suggest that the RFS program has exacerbated any of the advantages large integrated
refiners may have over small refineries.205 In other words, the competitiveness of small refineries
in the fuels market, be it favorable or unfavorable, does not change as a result of RFS
compliance obligations.
On the other hand, granting SREs has provided small refineries a unique and significant
competitive advantage. When small refineries are exempted from their RFS obligations, they
continue to sell their petroleum fuel at the market price, which reflects the RIN cost via RIN cost
passthrough. Thus, exempted small refineries recover the cost of the RINs (receive RIN revenue)
through their product sales, but do not have any RIN costs when they are granted an exemption.
The number of small refineries receiving exemptions, the total volume of gasoline and diesel fuel
204 See supra, Section IV.D.2.
205 EPA acknowledges that the Tenth Circuit has found that Congress may have understood large integrated refiners
to have certain advantages, and EPA has cited that decision itself in support of its prior approach to SRE decisions.
However, as noted, EPA does not believe that the available evidence supports the conclusion that small refineries
are structurally disadvantaged by the RFS program itself.
58
-------
exempted, the total value of the exemptions, and the value of the exemptions on a per gallon
basis are shown in Table IV.D.3-1. This table also shows the average net refining margins (an
indicator of profitability) for the exempted small refineries, for comparison with the value of the
exemptions. The value of the exemptions is typically significant relative to the average net
refining margin. For all exemptions granted from 2013 through 2018, the average value of the
exemptions (6.76 cents per gallon) was approximately 64% of the average net refining margin of
the exempted refineries (10.61 cents per gallon). Any exemptions granted in 2021 would likely
be of even greater value since current RIN prices, and therefore the current RIN cost per gallon
of fuel produced, are higher than RIN prices when the exemptions for 2013-2018 were granted.
Table IV.D.3-1: Value of SREs (2013-2018)
Volume of
Average Net
Number
Gasoline and
Diesel Fuel
Total Value of
Value of
Refining Margin
for Exempted
Year
of Grants
Issued
Exempted
(billion gallons)
the Exemptions
($ Million)206
Exemptions
(0 per gallon)
Refineries
(0 per gallon)207
2013
8
1.98
118
5.98
-0.65
2014
8
2.30
105
4.57
4.98
2015
7
3.07
171
5.57
12.05
2016
19
7.84
676
8.63
2.11
2017
35
17.05
1,459
8.56
11.76
2018
31
13.42
558
4.16
17.00
Total
108
45.66
3,088
6.76
10.61
f. EPA's conclusion that there is parity between the cost of obtaining a RIN through
blending and the cost of buying a RIN on the market is incorrect. It costs much
more to buy RINs, which many small refineries must do.
Several petitioners claim that EPA's analyses are based on the assumption that the cost of
obtaining a RIN through blending and the cost of purchasing a RIN is the same, and that this
assumption is unfounded. To support this claim, the petitioners note that the cost to purchase
RINs increased significantly in 2019, 2020, and 2021 and that the cost to purchase RINs in these
years was much greater than the cost to blend renewable fuel. The petitioners further state that if
there was no cost advantage to blending then there would be no reason for non-obligated parties
to continue blending. Rather, these parties would stop blending if they could not recoup the loss
by selling the RINs on the market.
We are aware that RIN prices increased significantly recently and we extended our
analysis of the impact of RIN prices on the fuels market through the end of 2020 to determine
206 Based on annual average RIN prices calculated by EPA from OPIS data for D3, D4, D5 and D6 RINs.
207 EPA often grants exemptions in the year(s) following the year for which an exemption is requested. Because of
this time lag, refineries sometimes financially account for the value of their exemption in the following year(s).
Thus, the value of the exemptions for some refineries may be included in the net refining margin for the following
year(s). For example, EPA granted some 2013 exemption in 2014 or later years, so the value of some 2013
exemptions may be included in financial statements for 2014 or later.
59
-------
whether our previous findings on RIN cost passthrough were supported by more recent data.208
We concluded that all the data available to EPA, including data submitted by the petitioners,
continues to support EPA's findings on RIN cost passthrough.
EPA's finding that there is parity between the cost to obtain a RIN through blending and
the price to purchase a RIN is not an arbitrary assumption. Rather, it is strongly supported by
both economic principles and the fuels market data. As stated previously, the market for blended
fuel is highly competitive. If the cost of obtaining a RIN by blending renewable fuel was lower
than the market price for a RIN, we would expect to see new blenders enter the market and/or
existing blenders increasing their blending to capitalize on this profit opportunity. This activity
would result in an increase in the supply of RINs for sale until the demand price for a RIN was
equal to the cost of obtaining a RIN through blending. Competitive market situations where the
sales price of a good is appreciably higher than the cost to produce a good are short-lived, as
market participants will increase production to take advantage of this opportunity until the
supply price and demand price are equal.
The market data EPA reviewed support this finding as well.209 The cost to obtain a RIN
by blending renewable fuel is not simply the fixed and operating costs for fuel blending (which
are relatively minor), nor is it simply the price difference between renewable fuel and the
petroleum fuel into which they are blended (e.g., the price difference between ethanol and
gasoline or between biodiesel and diesel fuel). Instead, the cost to a blender to obtain a RIN is
the price difference between the cost of the petroleum fuel (e.g., gasoline or diesel fuel) and the
renewable fuel used to produce blended fuel and the sales price of the blended fuel (e.g., E10 or
B5). The data presented in Section IV.D.2.d demonstrate that the difference between the cost of
the petroleum fuel and the renewable fuel used to produce blended fuel and the sales price of the
blended fuel is equal to the market price for the RINs associated with the blended fuel.210
The finding that there is parity between the cost of obtaining RINs by blending renewable
fuel and purchasing RINs does not mean that RINs do not provide an incentive for the blending
of renewable fuel. While blending renewable fuel does not result in windfall profits for blenders
(since the revenue from RIN sales is passed through to consumers in a discount on the price for
blended fuel), RIN revenue lowers the effective cost of renewable fuel, allowing blenders to
offer blended fuel containing renewable fuel at lower prices. The examples presented in Section
IV.D.2.C illustrate this point. In the E10 blending example, the price of the gasoline is $1.44 per
gallon and the price of ethanol is $1.50 per gallon, which is higher than the price of the gasoline
(prices from Table IV.D.2.C-2). However, the RIN discount allows E10 to sell for $1.37 per
gallon, which is lower than the price of the gasoline (line E4 from Table IV.D.2.C-2). Similarly,
in the B5 blending example, the price for ULSD is $1.48 and the price for biodiesel is $3.66
(prices from Table IV.D.2.C-3). Here again the RIN revenue, when combined with the federal tax
credit, allows B5 to sell for a lower price ($1.46 from line B5 in Table IV.D.2.C-4) than the price
of diesel fuel. Fuel buyers are extremely sensitive to prices. The incentive for blenders to
continue to blend renewable fuel when there is parity between the cost of obtaining a RIN
through blending and the cost to purchase a RIN is not that the revenue from the sale of the RIN
208 See supra, Section IV.D.2.
209 See supra, Section IV.D.2.d.
210 See supra, Figures IV.D.2.C-2 and 4.
60
-------
represents a windfall profit, but rather that the RIN discount allows blended fuel to sell at a lower
price relative to unblended fuel after passing through the revenue of the RIN sales to the
consumer.
g. Single site refineries only have access to a limited market and are therefore at a
disadvantage relative to large integrated refiners.
Several petitioners claim that because they own a single refinery and have access to
limited markets for their fuels, they are at a disadvantage compared to large integrated refiners.
The petitioners claim that because of their size, they cannot set the market price in such a way as
to recover their RIN costs, nor can they sell their fuel into other markets if their local market
prices are unfavorable.
As previously discussed, a refiner's ability to recover its RIN costs does not depend on
the refiner's ability to set the market price for the fuel it produces.211 Rather, because all parties
have the same cost to acquire RINs, whether they acquire RINs through blending renewable fuel
or by purchasing RINs, the market price for fuel reflects the cost of the RINs.
We are aware that the economics of refining crude oil to produce transportation fuel
changes over time, and that fuels markets vary in their profitability relative to other markets. At
times it can be an advantage to be in limited markets, and at other times not. Refiners with better
access to pipelines and other low-cost ways to transport the fuel they produce are better
positioned to react to changes in market dynamics, whether these changes are positive, negative,
short-term, or long-term in nature. These varying circumstances, and any hardship they might
cause to small refineries, are independent of and not caused by compliance with the RFS
program.
We received claims of disadvantage from small refineries in isolated markets where they
were the main supplier of fuel, from small refineries in markets readily accessible to many other
refineries, and from small refineries in every situation in-between. The identical claims from
such a broad diversity of refinery situations demonstrates that a small refinery's market has
nothing to do with potential impacts from the RFS program. As a result of the nationwide RIN
trading program, all refineries have equal access to the RINs they need for compliance with the
RFS program and at the same nationwide price.
h. Refineries that produce primarily diesel fuel are at a disadvantage since they
generally cannot blend as much renewable fuel into their product as can refineries
that produce gasoline.
The claim that small refineries producing a disproportionately high amount of diesel fuel,
relative to the amount of gasoline produced, suffer DEH from the RFS program presumes that
parties that acquire RINs by blending renewable fuel do so at a lower cost than parties that
purchase RINs. These small refineries generally assert that their ability to acquire RINs by
211 See supra, Sections IV.D.2 and IV.D.3.e.
61
-------
blending biodiesel or renewable diesel is limited relative to their competitors that have the ability
to blend greater quantities of ethanol into the gasoline they produce.
As previously discussed, all parties have the same cost to acquire RINs, whether they do
so by blending renewable fuel or by purchasing RINs.212 A party's cost of acquiring RINs,
therefore, is unrelated to its ability to blend renewable fuel. Further, it is not necessarily the case
that greater quantities of renewable fuel can be blended into gasoline relative to diesel fuel. With
the exception of very small quantities of higher-level ethanol blends such as El 5 and E85,
blending of ethanol into gasoline is limited to 10% by volume. Conversely, many parties
regularly sell diesel fuel blended with up to 20% biodiesel or renewable diesel.213 Parties
blending 20% biodiesel or renewable diesel into diesel fuel would acquire more RINs than
parties blending 10% ethanol into gasoline, especially after accounting for the higher
equivalence values of biodiesel and renewable diesel.
V. Conclusion
Section 21 l(o)(9)(B) of the CAA and 40 CFR 80.1441(e)(2) give EPA the authority to
grant an SRE petition only when a small refinery demonstrates it is experiencing DEH caused by
compliance with the RFS program. Based on our detailed evaluation, careful consideration of all
the available information, consultation with DOE, and consideration of the DOE study and other
economic factors, EPA proposes to find that none of the 65 pending small refinery petitions have
demonstrated DEH caused by the costs of compliance with the requirements of the RFS
program.
The market-based design of the RFS program and the RIN-based compliance system
have equalized the cost of compliance among all market participants, such that no refinery would
face DEH from its RFS obligations.214 We have evaluated an extensive amount of data and
available literature and have concluded that the cost of RINs is the same for all obligated parties,
whether the RINs are acquired by blending renewable fuel or by buying them on the market.215
Hence, small refineries do not face a disproportionate cost of compliance when compared to
other refineries, or to each other. Our analysis further shows that the costs of RFS compliance
(i.e., RINs) are passed through in the prices of refined products. Hence, in recovering their RIN
costs, refineries do not face economic hardship due to compliance with the RFS program.
Finding no disproportionate cost of compliance and no economic hardship due to the RFS
program, we conclude that small refineries do not face DEH. As such, EPA proposes to find that
compliance with the RFS program does not cause DEH to small refineries and, accordingly, to
deny all pending SRE petitions.
As stated above, this notice is not a final agency action for purposes of CAA section
307(b)(1). This action is not a rulemaking and is not subject to the various statutory and other
provisions applicable to a rulemaking. Rather, the purpose of this proposal is to clearly articulate
212 See supra, Sections IV.D.2 and IV.D.3.f.
213 See, e.g., diesel fuel offerings by Pilot Flying J—the largest diesel fuel retailer in the United States—available at:
https://pilotflvingi.com/fiiel-prices.
214 See supra, Section II.B.
215 See supra, Section IV.D.2.
62
-------
the Agency's current interpretation of its statutory authority to grant SREs and to present our
analysis of all available data on RFS costs and market dynamics for public review and comment.
63
------- |