[Federal Register Volume 85, Number 247 (Wednesday, December 23, 2020)]
[Rules and Regulations]
[Pages 84130-84157]
From the Federal Register Online via the Government Publishing Office [www.gpo.gov]
[FR Doc No: 2020-27368]



[[Page 84129]]

Vol. 85

Wednesday,

No. 247

December 23, 2020

Part II





Environmental Protection Agency





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





40 CFR Part 83





Increasing Consistency and Transparency in Considering Benefits and 
Costs in the Clean Air Act Rulemaking Process; Final Rule

  Federal Register / Vol. 85 , No. 247 / Wednesday, December 23, 2020 / 
Rules and Regulations  

[[Page 84130]]


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

ENVIRONMENTAL PROTECTION AGENCY

40 CFR Part 83

[EPA-HQ-OAR-2020-0044; FRL 10018-56-OAR]
RIN 2060-AU51


Increasing Consistency and Transparency in Considering Benefits 
and Costs in the Clean Air Act Rulemaking Process

AGENCY: Environmental Protection Agency (EPA).

ACTION: Final rule.

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

SUMMARY: This rule establishes processes that the Environmental 
Protection Agency (EPA) will be required to undertake in promulgating 
regulations under the Clean Air Act (CAA) to ensure that information 
regarding the benefits and costs of regulatory decisions is provided 
and considered in a consistent and transparent manner. The EPA is 
establishing procedural requirements governing the preparation, 
development, presentation, and consideration of benefit-cost analyses 
(BCA), including risk assessments used in the BCA, for significant 
rulemakings conducted under the CAA. Together, these requirements will 
help ensure that the EPA implements its statutory obligations under the 
CAA, and describes its work in implementing those obligations, in a way 
that is consistent and transparent.

DATES: This final rule is effective December 23, 2020, but does not 
apply to final rules for which a proposal was published prior to the 
effective date.

ADDRESSES: The EPA has established a docket for this action under 
Docket ID No. EPA-HQ-OAR-2020-0044. All documents in the docket are 
listed on the http://www.regulations.gov website. Although listed in 
the index, some information is not publicly available, e.g., CBI or 
other information whose disclosure is restricted by statute. Certain 
other material, such as copyrighted material, is not placed on the 
internet and will be publicly available only in hard copy form. 
Publicly available docket materials are available electronically 
through http://www.regulations.gov.

FOR FURTHER INFORMATION CONTACT: Leif Hockstad, Office of Air Policy 
and Program Support, Office of Air and Radiation, Environmental 
Protection Agency, Mail Code 6103A,1200 Pennsylvania Avenue NW, 
Washington, DC 20460; (202) 343-9432; email address: 
hockstad.leif@epa.gov.

SUPPLEMENTARY INFORMATION: Preamble acronyms and abbreviations. The EPA 
uses multiple acronyms and terms in this preamble. While this list may 
not be exhaustive, to ease the reading of this preamble and for 
reference purposes, the EPA defines the following terms and acronyms:

ANPRM Advanced Notice of Proposed Rulemaking
BCA Benefit-cost analysis
BenMAP Benefits Mapping and Analysis Program (BenMAP)
CAA Clean Air Act
CBI Confidential business information
CFR Code of Federal Regulation
CRA Congressional Review Act
EPA Environmental Protection Agency
IOM Institute of Medicine
NAAQS National Ambient Air Quality Standards
NHTSA National Highway Traffic Safety Administration
NPRM Notice of Proposed Rulemaking
IRIS Integrated Risk Information System
ISA Integrated Science Assessments
PII Personally identifiable information
SAB Science Advisory Board
WTA Willingness-to-accept
WTP Willingness to pay

    Organization of this document. The following outline is provided to 
aid in locating information in this preamble.

I. Executive Summary
    A. Purpose of the Regulatory Action
    B. Summary of the Major Provisions of the Regulatory Action
II. General Information
    A. Does this action apply to me?
    B. What is the Agency's authority for taking this action?
III. Background
    A. Summary of Executive Orders, Guidances, and Court Rulings 
Related to Regulatory BCA
    B. Summary of Proposed Rule
IV. Description of the Final Rule
V. Responses to Significant Comments
    A. Purpose of the Action
    B. Authority To Promulgate a Procedural Rule
    C. Definitions
    D. Preparation and Consideration of BCA in Rulemaking
    E. Best Practices for the Development of BCA
    1. Key Elements of a BCA
    2. Statement of Need
    3. Regulatory Options
    4. Baseline
    5. Measuring Benefits and Costs
    6. Methods for Estimating Benefits and Costs
    7. Selecting and Quantifying Health Endpoints in a BCA
    8. Uncertainty Analysis
    9. Principle of Transparency
    F. Requirements for the Presentation of BCA Results
    G. Additional Comment Responses
    1. Planning for Retrospective Analysis
    2. Comments Pertaining to Executive Order 12898
VI. References
VII. Statutory and Executive Order Reviews
    A. Executive Order 12866: Regulatory Planning and Review and 
Executive Order 13563: Improving Regulation and Regulatory Review
    B. Executive Order 13771: Reducing Regulations and Controlling 
Regulatory Costs
    C. Paperwork Reduction Act
    D. Regulatory Flexibility Act
    E. Unfunded Mandates Reform Act
    F. Executive Order 13132: Federalism
    G. Executive Order 13175: Consultation and Coordination With 
Indian Tribal Governments
    H. Executive Order 13045: Protection of Children From 
Environmental Health Risks and Safety Risks
    I. Executive Order 13211: Actions Concerning Regulations That 
Significantly Affect Energy Supply, Distribution or Use
    J. National Technology Transfer and Advancement Act
    K. Executive Order 12898: Federal Actions To Address 
Environmental Justice in Minority Populations and Low-Income 
Populations
    L. Congressional Review Act

I. Executive Summary

A. Purpose of the Regulatory Action

    Thorough and careful economic analysis is informative for 
developing sound environmental policies. High quality economic analyses 
enhance the effectiveness of environmental policy decisions by 
providing policy makers and the public with information needed to 
assess the likely consequences of various actions or options. 
Transparency about how these economic analyses are developed and how 
they are used in decision-making is essential to allowing interested 
parties to hold decision makers accountable for their decisions. BCA, a 
type of economic analysis, can serve an integral informative role in 
the regulatory development process. It provides detailed information 
about the value of benefits and costs of a policy to affected parties 
and whether a policy change has the potential to improve the aggregate 
well-being of society.
    The purpose of this action is to codify procedural best practices 
for the preparation, development, presentation, and consideration of 
BCA in regulatory decision-making under the CAA. This codification will 
help ensure that the EPA implements its statutory obligations under the 
CAA, and describes its work in implementing those obligations, in a way 
that is consistent and transparent. This transparency is important to 
allow interested parties to understand and evaluate the adequacy and 
accuracy of the BCA and the role the analysis

[[Page 84131]]

played in significant regulatory decision-making.
    The Agency is taking this action pursuant to CAA section 301(a). 42 
U.S.C. 7601(a)(1). Section 301(a)(1) provides authority to the 
Administrator ``to prescribe such regulations as are necessary to carry 
out his functions'' under the CAA. Such authority extends to internal 
agency procedures that increase the Agency's ability to provide 
consistency and transparency to the public in regard to the rulemaking 
process under the CAA. See NRDC v. EPA, 22 F.3d 1125, 1148 (D.C. Cir. 
1994) (``[Section 301] is sufficiently broad to allow the promulgation 
of rules that are necessary and reasonable to effect the purposes of 
the Act.'').

B. Summary of the Major Provisions of the Regulatory Action

    This final rule consists of three elements. First, it requires the 
EPA to prepare a BCA for all future significant proposed and final 
regulations under the CAA. The rule also requires that the Agency 
consider the BCA in promulgating the regulation except where the 
statutory provision or provisions under which a significant regulation 
is promulgated prohibit it.
    Second, the rule requires EPA to develop the BCA using the best 
available scientific information and in accordance with best practices 
from the economic, engineering, physical, and biological sciences. The 
final rule codifies best practices consistent with the EPA's Guidelines 
for Preparing Economic Analyses (hereafter ``Guidelines'') and the 
Office of Management and Budget's (OMB) Circular A-4, and also requires 
that risk assessments used to support BCAs should follow best 
methodological practices for risk characterization and risk assessment.
    Third, the rule imposes additional procedural requirements to 
increase transparency in the presentation and consideration of the BCA 
results. Specifically, the rule provides that the preambles of 
significant proposed and final CAA regulations must include a section 
that contains:
    a. A summary presentation of the overall BCA results for the rule, 
including total costs, benefits, and net benefits;
    b. An additional reporting of the public health and welfare 
benefits that pertain to the specific objective(s) of the CAA 
provision(s) under which the rule is promulgated;
    c. A transparent presentation of how specific costs contemplated in 
the CAA provision(s) under which the rule is promulgated (to the extent 
specified), relate to total costs, to the extent possible; and
    d. When the CAA statutory provision or provisions under which the 
rule is promulgated permit consideration of the BCA, a description of 
how the Agency considered the BCA.
    Together, these requirements will help ensure that the EPA 
implements its statutory obligations under the CAA in a way that is 
consistent and transparent. The provisions of the final rule codify 
best practices for the preparation, development, presentation, and 
consideration of BCA as articulated in the principles and requirements 
of Executive Order 12866. This final rule does not change any other 
requirements pertaining to CAA rules specified in executive orders and 
existing guidance documents. For example, this final rule does not 
change the requirements for what types of analysis should be included 
in regulatory impact analyses prepared under E.O. 12866.

II. General Information

A. Does this action apply to me?

    This rule does not regulate the conduct or determine the rights of 
any entity or individual outside the Agency, as this action pertains 
only to internal EPA practices. However, the Agency recognizes that any 
entity or individual interested in EPA's regulations may be interested 
in this rule. For example, this rule may be of particular interest to 
entities and individuals concerned with how the EPA conducts BCA.

B. What is the Agency's authority for taking this action?

    The Agency is taking this action pursuant to CAA section 301(a). 42 
U.S.C. 7601(a)(1). Section 301(a)(1) provides authority to the 
Administrator ``to prescribe such regulations as are necessary to carry 
out his functions'' under the CAA. Such authority extends to internal 
agency procedures that increase the Agency's ability to provide 
consistency and transparency to the public in regard to the rulemaking 
process under the CAA. See NRDC v. EPA, 22 F.3d 1125, 1148 (D.C. Cir. 
1994) (``[Section 301] is sufficiently broad to allow the promulgation 
of rules that are necessary and reasonable to effect the purposes of 
the Act.'').
    This is a rulemaking of agency organization, procedure, or 
practice. This procedural rule would not regulate any person or entity 
outside the EPA and would not affect the rights or obligations of 
outside parties. As a rule of Agency procedure, this rule is exempt 
from the notice-and-comment and delayed effective-date requirements set 
forth in the Administrative Procedure Act. See 5 U.S.C. 
553(a)(2),(b)(A),(d). Nonetheless, the Agency voluntarily sought public 
comment on the proposed rule because it believed that the information 
and opinions supplied by the public would inform the Agency's views. 
Vt. Yankee Nuclear Power Corp. v. Nat. Res. Def. Council, Inc., 435 
U.S. 519, 524 (1978) (``Agencies are free to grant additional 
procedural rights in the exercise of their discretion.'') In addition, 
even assuming arguendo that the notice-and-comment requirements of the 
Act applied to this action, EPA has determined that there would be good 
cause, consistent with 5 U.S.C. 553(d)(3), for making this final rule 
effective immediately because the goals of the rule, ensuring 
transparency and consistency in BCAs for significant CAA rulemakings, 
are crucial for ensuring confidence in EPA decision-making. Because 
this is a procedural rule that only applies internally to ensure that 
EPA follows existing best practices with respect to BCA and to ensure 
that EPA explains how EPA considered the results, the rationale for 
delayed effectiveness to allow time to adjust to the new requirements 
does not apply.
    In addition, the EPA received comments and recommendations on the 
proposed rule from the EPA Science Advisory Board (SAB), pursuant to 
its statutory duties to offer advice and comments on the scientific and 
technical basis of certain planned EPA actions pursuant to the 
Environmental Research, Development, and Demonstration Authorization 
Act of 1978 (ERDDAA).\1\ Finally, the EPA also reviewed comments 
received from the SAB during the course of its review of the 
forthcoming update of the EPA's Guidelines.\2\
---------------------------------------------------------------------------

    \1\ The ERDDAA requires the EPA to make available to the SAB 
proposed criteria documents, standards, limitations, or regulations, 
together with relevant scientific and technical information on which 
the proposed action is based. On the basis of this information, the 
SAB may provide advice and comments. The SAB final report on the 
proposed rule is available at: https://yosemite.epa.gov/sab/sabproduct.nsf/0/82e89c7a596e9efa852585a50064d32e!OpenDocument&TableRow=2.3#2.
    \2\ Information about the SAB review of the forthcoming update 
of the EPA's Guidelines is available at: https://yosemite.epa.gov/sab/sabproduct.nsf//LookupWebProjectsCurrentBOARD/30D5E59E8DC91C2285258403006EEE00?OpenDocument.
---------------------------------------------------------------------------

III. Background

A. Summary of Executive Orders, Guidances, and Court Rulings Related to 
Regulatory BCA

    As the EPA works to advance its mission of protecting public health 
and

[[Page 84132]]

the environment, it seeks to ensure that its analyses of regulatory 
decisions provided to the public continue to be rooted in sound, 
transparent, and consistent approaches to evaluating benefits and 
costs.
    The Supreme Court noted in Michigan v. EPA that ``[c]onsideration 
of cost reflects the understanding that reasonable regulation 
ordinarily requires paying attention to the advantages and the 
disadvantages of agency decisions.'' Michigan v. EPA, 135 U.S. 2699, 
2707 (2015). Many environmental statutes, including the CAA, 
contemplate the consideration of costs as part of regulatory decision-
making in many instances. Several of these statutes, including the CAA, 
contain provisions that explicitly require some form of cost 
consideration when establishing a standard. Additionally, several other 
statutory provisions use terminology that in context implicitly direct 
or allow the EPA to consider costs, alone or in conjunction with 
benefits and other factors. For example, section 112(n)(1)(A) of the 
CAA directs the Administrator to ``regulate electric utility steam 
generating units under [section 112], if the Administrator finds such 
regulation is appropriate and necessary.'' ``Read naturally in the 
present context, the phrase `appropriate and necessary' requires at 
least some attention to cost.'' Michigan, 135 S. Ct. at 2707 (2015). 
Therefore, in light of the varying statutory provisions in the CAA that 
apply to or otherwise address cost consideration, the Agency is 
finalizing procedural requirements to provide analysis to the public 
that will present all of the benefits and costs in a consistent manner 
for all significant CAA rulemakings.
    Thorough and careful economic analysis is informative for 
developing sound environmental policies. High quality economic analyses 
enhance the effectiveness of environmental policy decisions by 
providing policy makers and the public with information needed to 
systematically assess the likely consequences of various actions or 
options. BCA, a type of economic analysis, can serve an integral 
informative role in the regulatory development process. In general 
terms, a BCA is an evaluation of both the benefits and costs to society 
as a result of a policy and the difference between the two (i.e., the 
calculation of net benefits (benefits minus costs)). It provides 
information about whether a policy change has the potential to improve 
the aggregate well-being of society.
    The usefulness of BCA in informing the development of environmental 
regulations has been recognized both within and outside government for 
decades. As discussed below, Presidential Executive Orders and statutes 
have been in place for decades formally requiring the preparation of 
BCA in the development of major Federal regulations, and the courts 
have examined the use of BCA in several regulatory contexts. In 
addition, the usefulness of formal BCA in informing regulatory policy 
debates on protecting and improving public health, safety, and the 
natural environment has been emphasized in the academic literature. For 
example, as explained in seminal work by prominent economists Arrow et 
al. (1996a, 1996b), BCA ``can provide an exceptionally useful framework 
for consistently organizing disparate information, and in this way, it 
can greatly improve the process and, hence, the outcome of policy 
analysis. If properly done, BCA can be of great help to agencies 
participating in the development of environmental regulations . . .'' 
(1996b). Arrow et al. recommend that ``Benefit-cost analysis should be 
required for all major regulatory decisions,'' and that ``the precise 
definition of `major' requires judgment.''
    Benefit-cost analyses have been an integral part of executive 
branch rulemaking for decades. Presidents since the 1970s have issued 
executive orders requiring agencies to conduct analysis of the economic 
consequences of regulations as part of the rulemaking development 
process. President Ford's 1974 Executive Order (E.O.) 11821 required 
government agencies to prepare inflation impact statements before 
issuing major regulations.\3\ These inflation impact statements 
essentially turned into benefit-cost analyses based on the 
understanding that a regulation would not be truly inflationary unless 
its costs to society exceeded the benefits it produced,\4\ and the E.O. 
was renamed as Economic Impact Statements with E.O. 11949 in 1976.\5\ 
President Carter's 1978 E.O. 12044, Improving Government Regulations, 
included formal requirements for conducting regulatory analysis at a 
minimum ``for all regulations which will result in (a) an annual effect 
on the economy of $100 million or more; or (b) a major increase in 
costs or prices for individual industries, levels of government or 
geographic regions.'' \6\ Regulatory analyses under E.O. 12044 were 
required to contain ``a succinct statement of the problem; a 
description of the major alternative ways of dealing with the problem 
that were considered by the agency; an analysis of the economic 
consequences of each of these alternatives and a detailed explanation 
of the reasons for choosing one alternative over the others.''
---------------------------------------------------------------------------

    \3\ Executive Order 11821--Inflation Impact Statements, Federal 
Register, VOL. 39, NO. 231--Friday, November 29, 1974 (pages 41501-
41502) .
    \4\ https://obamawhitehouse.archives.gov/omb/inforeg_chap1#tnfrp.
    \5\ Executive Order 11949--Economic Impact Statements, Federal 
Register, VOL. 42, NO. 3--Wednesday, January 5, 1977 (page 1017). 
https://www.govinfo.gov/content/pkg/FR-1977-01-05/pdf/FR-1977-01-05.pdf.
    \6\ Executive Order 12044--Improving Government Regulations, 
Federal Register, VOL 43, NO. 58--Friday, March 24, 1978 (Pages 
12659-12670).
---------------------------------------------------------------------------

    In 1981, President Reagan issued E.O. 12291, Federal Regulation, 
which imposed the first requirements for conducting formal benefit-cost 
analysis in the development of new major Federal regulations. Among its 
provisions, E.O. 12291 explicitly required that: ``(a) Administrative 
decisions shall be based on adequate information concerning the need 
for and consequences of proposed government action; (b) Regulatory 
action shall not be undertaken unless the potential benefits to society 
for the regulation outweigh the potential costs to society; (c) 
Regulatory objectives shall be chosen to maximize the net benefits to 
society; (d) Among alternative approaches to any given regulatory 
objective, the alternative involving the least net cost to society 
shall be chosen; and (e) Agencies shall set regulatory priorities with 
the aim of maximizing the aggregate net benefits to society, taking 
into account the condition of the particular industries affected by 
regulations, the condition of the national economy, and other 
regulatory actions contemplated for the future.'' \7\ Under E.O. 12291, 
major regulations included ``any regulation that is likely to result 
in: (1) An annual effect on the economy of $100 million or more; (2) A 
major increase in costs or prices for consumers, individual industries, 
Federal, State, or local government agencies, or geographic regions; or 
(3) Significant adverse effects on competition, employment, investment, 
productivity, innovation, or on the ability of United States-based 
enterprises to compete with foreign-based enterprises in domestic or 
export markets.''
---------------------------------------------------------------------------

    \7\ Executive Order 12291--Federal Regulation, Federal Register, 
Vol 46--February 19, 1981 (Page 13193).
---------------------------------------------------------------------------

    In 1993, E.O. 12291 was revoked and replaced by President Clinton's 
E.O. 12866, Regulatory Planning and Review, which is still in effect 
today. E.O. 12866

[[Page 84133]]

requires that for all significant regulatory actions pursuant to 
Section 3(f), an agency provide ``an assessment of the potential costs 
and benefits of the regulatory action, including an explanation of the 
manner in which the regulatory action is consistent with a statutory 
mandate . . .'' For regulatory actions meeting criteria listed under 
Section 3(f)(1)--that is, any regulatory action that is ``likely to 
result in a rule that may . . . have an annual effect on the economy of 
$100 million or more or adversely affect in a material way the economy, 
a sector of the economy, productivity, competition, jobs, the 
environment, public health or safety, or State, local, or tribal 
governments or communities''--E.O. 12866 further requires that this 
assessment include a quantification of benefits and costs to the extent 
feasible. In addition, E.O. 12866 states that, to the extent permitted 
by law, agencies ``should assess both the costs and the benefits of the 
intended regulation and, recognizing that some costs and benefits are 
difficult to quantify, propose or adopt a regulation only upon a 
reasoned determination that the benefits of the intended regulation 
justify its costs''; ``in choosing among alternative regulatory 
approaches . . . should select those approaches that maximize net 
benefits (including potential economic, environmental, public health 
and safety, and other advantages; distributive impacts; and equity), 
unless a statute requires another regulatory approach''; and that 
``[e]ach agency shall base its decisions on the best reasonably 
obtainable scientific, technical, economic, and other information 
concerning the need for, and consequences of, the intended 
regulation.''
    In 1995, the Unfunded Mandates Reform Act of 1995 (UMRA) included 
analytical requirements for all regulatory actions that include federal 
mandates ``that may result in the expenditure by State, local, and 
tribal governments, in the aggregate, or by the private sector, of $100 
million or more (adjusted annually for inflation) in any one year.'' An 
action contains a federal mandate if it imposes an enforceable duty on 
state, local or tribal governments, or the private sector. The 
analytical requirements under UMRA are similar to the analytical 
requirements under E.O. 12866, and thus the same analysis may permit 
compliance with both analytical requirements.\8\
---------------------------------------------------------------------------

    \8\ While the analytical requirements are the same, the dollar 
thresholds do not exactly coincide because the $100 million 
threshold is not adjusted for inflation under E.O. 12866.
---------------------------------------------------------------------------

    More recent Executive Orders also reaffirm the requirements and 
principles in E.O. 12866. E.O. 13563, issued in 2011 and still in 
effect today, reaffirms the requirements and other principles and 
definitions in E.O. 12866 and embraces benefit-cost analysis: ``In 
applying these principles, each agency is directed to use the best 
available techniques to quantify anticipated present and future 
benefits and costs as accurately as possible.'' \9\ More recently, E.O. 
13777, issued in 2017, directs agencies to identify regulations that 
``impose costs that exceed benefits.'' \10\ E.O. 13783, also issued in 
2017, similarly reaffirms the importance of benefit-cost analysis: ``In 
order to ensure sound regulatory decision-making, it is essential that 
agencies use estimates of costs and benefits in their regulatory 
analyses that are based on the best available science and economics.'' 
\11\
---------------------------------------------------------------------------

    \9\ https://obamawhitehouse.archives.gov/the-press-office/2011/01/18/executive-order-13563-improving-regulation-and-regulatory-review.
    \10\ Enforcing the Regulatory Reform Agenda (82 FR 12285, March 
1, 2017).
    \11\ https://www.govinfo.gov/content/pkg/FR-2017-03-31/pdf/2017-06576.pdf.
---------------------------------------------------------------------------

    The Office of Management and Budget's (OMB's) Circular A-4 (OMB 
2003), which remains in effect today, provides guidance to Federal 
agencies on the development of regulatory analysis as required under 
E.O. 12866 and a variety of related authorities.\12\ In developing 
Circular A-4, OMB first developed a draft that was subject to public 
comment, interagency review, and external peer review. As summarized in 
E.O. 13783, ``. . . OMB Circular A-4 . . . was issued after peer review 
and public comment and has been widely accepted for more than a decade 
as embodying the best practices for conducting regulatory cost-benefit 
analysis.'' \13\ The document encourages transparency in practices, 
including the expression of costs and benefits in monetary units that 
allow for the evaluation of ``incremental benefits and costs of 
successively more stringent regulatory alternatives'' such that an 
agency can ``identify the alternative that maximizes net benefits.'' 
\14\
---------------------------------------------------------------------------

    \12\ https://obamawhitehouse.archives.gov/omb/circulars_a004_a-4/. Circular A-4 refines and replaces OMB's ``best practices'' 
document of 1996, which was issued as a guidance in 2000 and 
reaffirmed in 2001. All these versions of the 1996 document were 
superseded by Circular A-4.
    \13\ https://www.govinfo.gov/content/pkg/FR-2017-03-31/pdf/2017-06576.pdf.
    \14\ https://obamawhitehouse.archives.gov/omb/circulars_a004_a-4/.
---------------------------------------------------------------------------

    EPA's Guidelines for Preparing Economic Analyses (hereafter, the 
Guidelines) \15\ complements Circular A-4 by providing the Agency with 
more detailed peer-reviewed guidance on how to conduct BCA and other 
types of economic analyses for both environmental regulatory actions 
and non-regulatory management strategies, with the intent of improving 
compliance with E.O. 12866 and other executive orders and statutory 
requirements (e.g., Small Business Regulatory Enforcement Fairness Act 
of 1996 provisions). The Guidelines are updated periodically--building 
on work issued in 1983 (then titled Guidelines for Performing 
Regulatory Impact Analysis), 2000, and most recently in 2010--to 
account for growth and development of economic tools and practices. The 
Guidelines establish a scientific framework for analyzing the benefits, 
costs, and other economic impacts of regulations and policies, 
including assessing the distribution of costs and benefits among 
various segments of the population. In addition to presenting the well-
established scientific foundations for economic analysis, the 
Guidelines incorporate recent advances in theoretical and applied work 
in the field of environmental economics. Updates of the Guidelines are 
led by the EPA's National Center for Environmental Economics in 
consultation with economists from across the Agency and OMB. All 
chapters undergo an external peer review, either through EPA's Science 
Advisory Board or through independent reviews by external experts, 
prior to be being finalized.\16\
---------------------------------------------------------------------------

    \15\ https://www.epa.gov/environmental-economics/guidelines-preparing-economic-analyses.
    \16\ The EPA is in the process of a periodic update of the 
Guidelines. The EPA anticipates that among the changes within this 
update, the current Section 9.2.3.3, ``Impacts on employment'', will 
be replaced with a discussion based on more recent literature and 
feedback from the Economy Wide Modeling Science Advisory Board 
Panel. For more details regarding Chapter 9, see: https://www.epa.gov/sites/production/files/2017-09/documents/ee-0568-09.pdf. 
For more details regarding the update of the Guidelines in general, 
see: https://yosemite.epa.gov/sab/sabproduct.nsf//LookupWebProjectsCurrentBOARD/30D5E59E8DC91C2285258403006EEE00?OpenDocument.
---------------------------------------------------------------------------

    Given the history described above pertaining to the use of BCA by 
executive agencies, and given that several statutes, including the CAA, 
include provisions that require some form of cost consideration, the 
federal courts have also developed significant case law regarding 
regulatory cost consideration and the usefulness of BCA. This case law 
addresses when, and if, such use is required or permissible and how it 
may be employed in reasoned decision-making.

[[Page 84134]]

As a general matter, while certain statutory provisions may prohibit 
reliance on BCA or other methods of cost consideration in decision-
making,\17\ such provisions do not preclude the Agency from providing 
additional information regarding the impacts of a proposed or final 
rule to the public. For example, while the CAA prohibits the EPA from 
considering cost when establishing or revising requisite National 
Ambient Air Quality Standards (NAAQS) for criteria pollutants,\18\ the 
EPA nonetheless provides Regulatory Impact Analyses (RIAs) \19\ to the 
public for these rulemakings.\20\
---------------------------------------------------------------------------

    \17\ See, e.g., Whitman v. Am. Trucking Ass'ns, 531 U.S. 457 
(2001) (holding that Section 109(b) of the CAA unambiguously barred 
cost considerations when setting the National Ambient Air Quality 
Standards.
    \18\ Id.
    \19\ A regulatory impact analysis, or ``regulatory analysis'' 
for brevity, as prepared under E.O. 12866, consists of a benefit-
cost analysis and any related cost-effectiveness analyses and 
assessments of economic and distributional impacts (OMB 2003).
    \20\ See, e.g., U.S. EPA, Regulatory Impact Analysis of the 
Proposed Revisions to the National Ambient Air Quality Standards for 
Ground-Level Ozone (2014), https://www3.epa.gov/ttn/ecas/regdata/RIAs/20141125ria.pdf.
---------------------------------------------------------------------------

    The Supreme Court has held that agencies may conduct and consider a 
BCA even when a statute does not explicitly require one. In Entergy 
Corp. v. Riverkeeper, Inc., 556 U.S. 208, 222-224 (2009), the Supreme 
Court clarified that neither American Textile Mfrs. Inst. V. Donovan, 
452 U.S. 490 (1981) (American Textile Mfrs.) nor Whitman v. Am. 
Trucking Ass'ns, 531 U.S. 457 (2001) (American Trucking), stands for 
the broad proposition that statutory silence in regard to a potential 
factor always implies prohibition of consideration of that factor. 
Therefore, the Supreme Court concluded that the EPA was permitted to 
use BCA in determining the content of regulations promulgated under 
Clean Water Act section 1326(b). The Court reasoned ``that [CWA] Sec.  
1326(b)'s silence is meant to convey nothing more than a refusal to tie 
the agency's hands as to whether cost-benefit analysis should be used, 
and if so to what degree.'' Id. at 222; see also id. at 212, 219-20, 
226.
    The Supreme Court noted that its decisions in American Trucking and 
American Textile Mfrs. ``do not undermine this conclusion.'' 556 U.S. 
at 223. The Court highlighted that in American Trucking, it had held 
that the text of section 109 of the Clean Air Act, ``interpreted in its 
statutory and historical context . . . unambiguously bars cost 
considerations'' when air quality standards are set pursuant to that 
provision. American Trucking, 531 U.S. at 471, quoted in Entergy Corp., 
556 U.S. at 223. The Entergy Corp. Court further elaborated that 
``[t]he relevant 'statutory context' [in American Trucking] included 
other provisions in the [CAA] that expressly authorized consideration 
of costs, whereas Sec.  109 did not.'' 556 U.S. at 233. The Court 
concluded, not that American Trucking stands for the proposition that 
statutory silence always unambiguously bars cost consideration, but, 
rather that American Trucking ``stands for the rather unremarkable 
proposition that sometimes statutory silence, when viewed in context, 
is best interpreted as limiting agency discretion.'' 556 U.S. at 223. 
The Court further noted that in American Textile, the Court had relied, 
in part, on the absence of mention of BCA in the statute to hold that 
the agency was not required to conduct a BCA when setting certain 
health and safety standards. 556 U.S. at 223. ``[U]nder Chevron, that 
an agency is not required to [engage in cost-benefit analysis] does not 
mean that an agency is not permitted to do so.'' Id. Thus, the Supreme 
Court has confirmed that a statute need not have explicitly required 
that the agency conduct a BCA in its decision-making process for the 
agency to do so.
    The Supreme Court additionally acknowledged in Entergy Corp. that 
``whether it is `reasonable' to bear a particular cost may well depend 
on the resulting benefits.'' 556 U.S. at 225-226. This concept was 
further elaborated upon by the Court in Michigan v. EPA, which held, in 
the context of the term ``appropriate and necessary'' contained in 
Section 112(n)(1)(A) of the CAA, that the term required consideration 
of cost. 135 S. Ct. 2699, 2706 (2015). In doing so, the Supreme Court 
stated that ``[o]ne would not say that it is even rational, never mind 
`appropriate,' to impose billions of dollars in economic costs in 
return for a few dollars in health or environmental benefits'', 
concluding that ``[n]o regulation is `appropriate' if it does 
significantly more harm than good.'' Id. at 2707. The D.C. Circuit 
recently echoed this concept in Mingo Logan Coal Co. v. EPA. While the 
D.C. Circuit panel ultimately concluded that the cost issue had been 
forfeited by petitioners, in response to then Judge Kavanaugh's dissent 
which argued that cost consideration should be required, the panel 
stated, ``[i]ndeed, we do not quibble with his general premise--and 
that of the many legal luminaries he cites--that an agency should 
generally weigh the costs of its action against its benefits.'' 829 
F.3d 710, 723 (D.C. Cir. 2016). In general, when cost consideration is 
either required or permitted by the CAA, the courts have not mandated a 
specific approach for cost consideration but have granted the Agency 
broad discretion in determining its methodology. See Michigan, 135 S. 
Ct. at 2711 (``We need not and do not hold that the law unambiguously 
required the Agency, when making this preliminary estimate, to conduct 
a formal cost-benefit analysis in which each advantage and disadvantage 
is assigned a monetary value. It will be up to the Agency to decide (as 
always, within the limits of reasonable interpretation) how to account 
for cost.''); see also Sierra Club v. Costle, 657 F.2d 298, 345 (D.C. 
Cir. 1981) (``[S]ection 111(a) explicitly instructs the EPA to balance 
multiple concerns when promulgating a NSPS.''); id. at 321 (``The text 
gives the EPA broad discretion to weigh different factors in setting 
the standard.''); Lignite Energy Council v. EPA, 198 F.3d 930, 933 
(D.C. Cir. 1999) (``Because section 111 [of the CAA] does not set forth 
the weight that [should be] assigned to each of these factors, we have 
granted the agency a great degree of discretion in balancing them''); 
Husqvarna AB v. EPA, 254 F.3d 195, 200 (D.C. Cir. 2001) (``Section 213 
[of the CAA] . . . simply directs the EPA to consider cost. . . . 
Because section 213 does not mandate a specific method of cost 
analysis, we find reasonable the EPA's choice to consider costs on the 
per ton of emissions removed basis.'').
    Additionally, courts have noted the usefulness of BCA and have 
utilized the information provided therein to inform their analysis when 
reviewing agency regulations. Several of these cases utilize 
information from agency-created BCAs and/or RIAs as evidence that an 
agency ignored alternatives or acted in an arbitrary and capricious 
manner when taking action.
    For example, in Advocates for Highway and Auto Safety v. FMCSA, 429 
F.3d. 1136 (D.C. Cir. 2005), the D.C. Circuit relied in part on a BCA 
in invalidating, as arbitrary and capricious, a final rule promulgated 
by Federal Motor Carrier Safety Administration (FMCSA) intended to 
ensure that drivers of commercial motor vehicles received adequate 
training. In its analysis, the D.C. Circuit highlighted an incongruity 
between methods of training shown to be effective and the final rule, 
noting that ``[f]rom a purely economic perspective, the agency's 
disregard of the Adequacy Report [containing a BCA] is baffling in 
light of the evidence in the record.'' Id. at 1146. The D.C. Circuit 
pointed to a training regimen that ``according to the agency's

[[Page 84135]]

own calculations, [would] produce benefits far in excess of costs.'' 
Id. Noting the agency's findings that ``the program's estimated 10-year 
cost of between $4.19 billion to $4.51 billion would yield a benefit 
ranging from $5.4 billion to $15.27 billion, depending on analytic 
assumptions,'' the court concluded that the BCA for the rule ``lends no 
support to FMCSA's position. In the final rule, FMCSA says practically 
nothing about the projected benefits.'' Id.
    In Public Citizen, Inc. v. Mineta, 340 F.3d 39 (2nd Cir. 2003), the 
Second Circuit determined that a National Highway Traffic Safety 
Administration (NHTSA) rule regarding tire pressure monitoring system 
(TPMS) requirements was arbitrary and capricious, as the NHTSA BCA 
showed that alternatives would be safer and more cost-effective. The 
court stated that it may ``be difficult to weigh economic costs against 
safety benefits. But the difficulty of the task does not relieve the 
agency of its obligation to perform it under [certain vehicle safety 
laws] and State Farm.'' Id. at 58 (citing Motor Vehicles Mfrs. Ass'n v. 
State Farm Mut. Auto. Ins. Co., 463 U.S. 29 (1983)). The Second Circuit 
observed that NHTSA ``instead, presents us with a rulemaking record 
that does not explain why the costs saved were worth the benefits 
sacrificed.'' Id. The court noted that the BCA ``discloses that the 
added cost for a system that worked all of the time, rather than half 
of the time, was less than $10 per car, and that the adoption of the 
four-tire, 25 percent standard alone was the most cost effective means 
of preventing crashes caused by significantly under-inflated tires.'' 
Id.
    Finally, in NRDC v. EPA, 824 F.2d 1258 (1st Cir. 1987), the First 
Circuit vacated, in part, and remanded rules for long-term disposal of 
high-level radioactive waste under Nuclear Waste Policy Act of 1982 
based in part on the Agency's selection of a 1,000-year design 
criterion rather than a longer-term one. The court determined that it 
was unreasonable agency action to not adopt cheap methods of increasing 
protections. In doing so, the court observed that ``[l]ikewise, EPA's 
Final [RIA] of 40 CFR part 191 demonstrates that more rigorous site 
selection could produce sites with such impermeable geologic media that 
compliance with the individual protections for a much longer duration 
would not even require the extra cost of `very good' engineered 
canisters.'' Id. at 1289.

B. Summary of the Proposed Rule

    With the history discussed above in mind as a backdrop and 
following E.O. 13777 noted above, the EPA opened a public docket \21\ 
in April 2017 to solicit feedback and identify regulations that 
``impose costs that exceed benefits.'' Among the public comments 
received, a large cross-section of industry stakeholders stated that 
the agency either underestimated costs, overestimated benefits, or 
evaluated benefits and costs inconsistently in its rulemakings. Per 
E.O. 13777 and based on these public comments, the EPA decided to take 
further action to evaluate opportunities for reform.
---------------------------------------------------------------------------

    \21\ See EPA, Evaluation of Existing Regulations (82 FR 17793). 
All public comments are accessible online in our docket on the 
Regulations.gov website identified by Docket ID No. EPA-HQ-OA-2017-
0190.
---------------------------------------------------------------------------

    In June 2018, the EPA issued an Advance Notice of Proposed 
Rulemaking (ANPRM), ``Increasing Consistency and Transparency in 
Considering Costs and Benefits in the Rulemaking Process'' (83 FR 
27524, June 13, 2018), to solicit public input on potential approaches 
for increasing consistency and transparency in how the EPA considers 
benefits and costs in the rulemaking process. Informed by the public 
comments received on that ANPRM, on May 13, 2019, the Administrator 
issued a memorandum \22\ to EPA's Assistant Administrators announcing 
the intention to propose statute-specific rules that outline how 
consistency and transparency concepts will be implemented in future 
rulemakings. The memorandum outlined the following principles for 
developing these regulatory proposals, consistent with applicable laws 
and regulations: Ensuring that the Agency balances benefits and costs 
in regulatory decision-making; increasing consistency in the 
interpretation of statutory terminology; providing transparency in the 
weight assigned to various factors in regulatory decisions; and 
promoting adherence to best practices in conducting the technical 
analysis used to inform decisions.
---------------------------------------------------------------------------

    \22\ Available at: https://www.epa.gov/environmental-economics/administrator-wheeler-memorandum-increasing-consistency-and-transparency.
---------------------------------------------------------------------------

    In June 2020, the EPA issued a Notice of Proposed Rulemaking 
(NPRM), ``Increasing Consistency and Transparency in Considering Costs 
and Benefits in the Rulemaking Process'' (85 FR 35612, June 11, 2020). 
The proposed rule was the first statute-specific rulemaking in this 
effort. The EPA proposed to codify the procedural requirements 
governing the development of BCA, including risk assessments used as 
inputs to the BCA, for significant rulemakings conducted under the CAA, 
and proposed additional procedural requirements to increase 
transparency in the presentation of the benefits and costs resulting 
from significant CAA regulations. Together, these requirements were 
proposed to ensure a consistent approach to the EPA's BCAs under the 
CAA and to provide transparency by requiring the provision of relevant 
information in all significant rulemakings. In the proposed rule, the 
EPA also solicited comment on how the Agency should take into 
consideration the results of a BCA in future rulemakings under specific 
provisions of the CAA, among other topics. Discussion of topics where 
the EPA solicited comment, and comments and responses where EPA has 
made modifications in the final rule, is included in Section V of this 
preamble. Responses to the rest of the comments are provided in the 
Response to Comments Document.

IV. Description of the Final Rule

    This final rule consists of three elements. In the first element, 
it requires the EPA to prepare a BCA for all future significant 
proposed and final regulations promulgated under the CAA and to 
consider the BCA in the decision-making process when permitted for 
consideration under the specific provision of the CAA under which the 
future regulation is promulgated. The EPA believes that in keeping with 
OMB's Circular A-4 and Executive Order 12866 that the requirement to 
prepare a BCA would create consistency with well-understood and 
established processes and determinations for what constitutes a 
``significant'' rulemaking. Therefore, in this final rule, a 
significant regulation will include any proposed or final regulation 
that is determined to be a ``significant regulatory action'' pursuant 
to Section 3(f) E.O. 12866 or is otherwise designated as significant by 
the Administrator. Consideration of the results of BCA in regulatory 
decision-making is also consistent with the requirements of E.O. 12866. 
If the provision or provisions under which the rule is promulgated 
prohibit the consideration of the BCA, the final rule requires the 
Agency to identify the specific provision that bars such consideration.
    The second element of the final rule requires EPA to develop the 
BCA using the best available scientific information and in accordance 
with best practices from the economic, engineering, physical, and 
biological sciences. The final rule codifies general best practices 
consistent with the existing guidances that EPA relies upon to develop 
high

[[Page 84136]]

quality regulations (e.g., EPA's Guidelines for Preparing Economic 
Analyses (hereafter ``Guidelines'') and the Office of Management and 
Budget's (OMB) Circular A-4), and also requires that risk assessments 
used to support BCAs should follow best methodological practices for 
risk characterization/assessment. The final rule does not replace any 
detailed guidance for Agency analysis, including Executive Orders 
(e.g., E.O. 12866), OMB Circulars (e.g., Circular A-4), and EPA 
documents (e.g., Guidelines for Preparing Economic Analyses).
    The specific best practices that are required in this final rule 
are as follows. The BCA must include a statement of need, an 
examination of regulatory options which would contribute to the stated 
objectives of the CAA, and to the extent feasible, an assessment of all 
benefits and costs of these regulatory options relative to the baseline 
scenario. The baseline used in the BCA must appropriately consider 
relevant factors and rely on transparent and reasonable assumptions. In 
preparing the BCA, the Agency must rely on the use of a framework for 
estimating costs and benefits that is appropriate for the 
characteristics of the regulation being evaluated and must provide an 
explanation for the approach adopted. In estimating costs and benefits, 
the Agency must consider how costs and benefits may be affected by 
consumer and producer behavior both in the baseline and in the policy 
scenarios. The BCA must include, to the extent supported by scientific 
literature as well as practicable in a given rulemaking: A 
quantification of all benefits; a monetization of benefits that follows 
well-defined economic principles using well-established economic 
methods, appropriate data and/or studies; and a qualitative 
characterization of benefits that cannot be quantified or monetized.
    Regarding the process of selecting health benefit endpoints for 
quantification, the final rule requires that this process will be based 
upon scientific evidence that indicates there is a clear causal or 
likely causal relationship between pollutant exposure and effect, and 
that sufficient data and understanding allows the agency to reasonably 
model the anticipated change in that effect in response to changes in 
environmental quality or exposures expected as a result of the 
regulation under analysis. The evaluation of the scientific evidence 
necessary to select and quantify health benefit endpoints should follow 
the systematic review process, must emphasize transparency and 
replicability, and give more weight to higher quality data, models, 
and/or analyses that have been peer reviewed. The models used to 
quantify the concentration-response relationships should take into 
account the breadth and quality of the available evidence regarding the 
nature and magnitude of the risk to the populations affected by the 
regulation. The presentation of results should characterize the 
sensitivity of the choice of the concentration-response function on the 
magnitude and the uncertainty associated with estimated benefits.
    The BCA must include an identification of uncertainties underlying 
the estimation of both benefits and costs and, to the extent feasible 
and appropriate, quantitatively analyze those that are most 
influential; and must present benefits and cost estimates in ways that 
convey their uncertainty, including acknowledging unquantified benefits 
and costs, where appropriate. The BCA must include a reasoned 
explanation for the scope and specific quantitative or qualitative 
methods chosen to analyze uncertainties.
    The final rule also requires that the overall results of the BCA 
(benefits, costs, and net benefits of each regulatory option evaluated 
in the BCA) be presented and described in a manner designed to be 
objective, comprehensive, reproducible to the extent reasonably 
possible, and easily understood by the public. To the extent permitted 
by law, the Agency must ensure that all information (including data and 
models) used in the development of the BCA is publicly available. If 
data and models are proprietary, the Agency must make available, to the 
extent practicable, the underlying inputs and assumptions used, 
equations, and methodologies used by EPA. The BCA shall provide a 
reasoned explanation for any departures from best practices in the BCA, 
including a discussion of the likely effect of the departures on the 
results of the BCA.
    The third element of the final rule imposes additional procedural 
requirements to increase transparency in the presentation and 
consideration of the BCA results. Specifically, the rule requires the 
preamble of significant proposed and final CAA regulations to include a 
section that contains a summary presentation of the overall BCA results 
for the rule, including total benefits, costs, and net benefits. Within 
this summary presentation, if any benefits and costs accrue to non-U.S. 
populations they must be reported separately to the extent possible. 
This section of the preamble should also provide an additional 
reporting of the public health and welfare benefits that pertain to the 
specific objective(s) of the CAA provision(s) under which the rule is 
promulgated and a transparent presentation of how specific costs 
contemplated in the CAA provision(s) under which the rule is 
promulgated (to the extent specified), relate to total costs, to the 
extent possible. Finally, when the CAA statutory provision or 
provisions under which the rule is promulgated permit consideration of 
the BCA, this section of the preamble should contain a description of 
how the Agency considered the BCA.
    Together, these requirements will help ensure that the EPA 
implements its statutory obligations under the CAA with high quality 
regulations in a way that is consistent and transparent and that these 
procedures are made enforceable upon the Agency. The provisions of the 
final rule codify into regulation best practices for the preparation, 
development, presentation, and consideration of BCA as articulated in 
the principles and requirements of Executive Order 12866.

V. Responses to Significant Comments

    The EPA had a 45-day public comment period on the proposed rule, 
and also hosted a virtual public hearing on July 1, 2020, which 
included 50 speakers registered to provide testimony. In total, the EPA 
received 24,740 public comments, including several mass mail campaigns 
and 513 unique comment letters (including transcripts from the July 1 
virtual public hearing). Of these, a total of 143 letters provided 
detailed, substantive comments. Commenters included environmental and 
health advocacy organizations, industry trade groups, academics, and 
State, Local, and Tribal governments.

A. Purpose of the Action

    Commenters supporting the EPA's proposed rulemaking argued that the 
proposed requirements, if finalized, would provide more clarity and 
transparency, make common sense, enhance public accountability and 
understanding of the scientific inputs that drive the EPA's decisions, 
improve the integrity of the rulemaking process, and lead to better 
public policy. Commenters also stated that codification of best 
practices for conducting and presenting BCA would standardize 
procedures and would achieve consistency over time and provide for 
better transparency. Some commenters further argued the rule would 
deliver continued environmental improvement as well as a more

[[Page 84137]]

predictable and achievable set of outcomes for the regulated community. 
In addition, a commenter stated that EPA's proposed rule, if finalized, 
would supersede, rather than duplicate, existing non-justiciable, non-
statutory sources of guidance for Agency analysis, including EOs (e.g., 
E.O. 12866), OMB Circulars (e.g., Circular A-4), and EPA documents 
(e.g., EPA's Guidelines).
    Commenters opposed to the proposed rule argued that the EPA does 
not explain how any of the Agency's previous BCAs have fallen short of 
any applicable legal requirements or failed to deliver on their 
purported policy benefits. Commenters stated that EPA has also not 
specifically detailed how the Agency's use of its own economic guidance 
(e.g., EPA's Guidelines) and OMB's Circular A-4 guidance has resulted 
in inadequate, inconsistent, or nontransparent practices or has 
compromised the Agency's abilities and disagreed with the need for a 
rulemaking. These commenters said that the EPA's proposal does not make 
the case that such shortcomings are so widespread among the EPA's 
existing BCA practices that the proposal was necessary. These 
commenters further stated the EPA does not identify any deficiencies in 
existing laws, orders, and guidelines, and, therefore, did not fully 
demonstrate how the proposed changes will address the alleged problem. 
Some commenters further stated that the EPA's proposed rule creates an 
excessively burdensome set of procedures for completing a BCA that 
would be difficult for the agency to satisfy and would be prohibitively 
costly to complete. One commenter stated that increasing transparency 
and consistency in the analysis upon which regulatory decisions are 
based should not come at the cost of undermining the flexibility and 
accuracy needed for regulatory decision-making on the wide variety of 
air pollutants and sources regulated under the CAA. The commenter added 
that many of the consistency and transparency goals in the proposal are 
already being met through existing EPA practices, particularly 
requirements in E.O. 12866, and contended that setting a prescriptive 
process for conducting BCAs will lead to inflexibility that could prove 
detrimental to public health and the environment. One commenter argued 
that, given the clear credibility and reliability of the peer-reviewed 
and longstanding methodologies for developing BCAs (as acknowledged by 
the EPA itself throughout the proposal), it was arbitrary and 
capricious for the EPA to constrain its methodologies. A few commenters 
objected to the proposal's approach, as they believed that a regulation 
establishes rigid practices that then make it difficult for the EPA to 
readily adopt future improvements to best practices. On this issue, a 
few commenters further suggested that because analytical requirements 
evolve, the EPA should create a requirement to periodically update the 
best practices through a public notice and comment rulemaking process.
    The EPA disagrees with commenters that this rule is unnecessary. 
The EPA continues to believe that codifying best practices into 
regulation provides additional certainty and increases the consistency 
and transparency of its analysis of the benefits and costs of 
significant regulations under the CAA. The requirements promulgated in 
this action address the comments, by many, that the Agency has not 
consistently estimated, presented, and considered benefits and costs in 
line with best practices and principles set forth in longstanding 
executive orders governing regulatory analysis. Some commenters 
asserted that these inconsistencies were not identified by EPA and were 
not so widespread among the EPA's existing BCA practices that the 
proposal was necessary. However, EPA has not had procedural enforceable 
regulations in place to ensure consistency in its past BCA practices. 
To the extent that commenters assert that EPA's past practice has been 
consistent and transparent, it is not due to an enforceable 
standardized approach that would ensure such a result. Other commenters 
have noted the contrary belief, that EPA's practices in regard to BCA 
have indeed been inconsistent and have lacked transparency. Without 
enforceable procedural regulations for BCA, future regulations may be 
promulgated without consideration of, and public accountability 
concerning, their costs and benefits. Thus, the EPA has determined that 
the Final Rule is necessary to ensure that BCA practices are 
implemented in a consistent fashion prospectively. The requirements 
provide a practical framework to ensure that the BCA of significant CAA 
regulations follow best practices and complement more detailed existing 
guidances the EPA relies upon (e.g., OMB's Circular A-4 and EPA's 
Guidelines) to develop quality regulations consistent with the CAA, and 
that these procedures are made enforceable upon the Agency. The final 
rule does not replace detailed guidance for Agency analysis, including 
Executive Orders (e.g., E.O. 12866), OMB Circulars (e.g., Circular A-
4), and EPA documents (e.g., EPA's Guidelines).

B. Authority To Promulgate a Procedural Rule

    The EPA received comments on its legal authority to promulgate the 
proposed rule. We respond to some of the major comments below and to 
the rest in Chapter 4 of the Response to Comments Document. In 
particular, the EPA received comments that Section 301(a)(1) of the CAA 
both does and does not provide adequate authority to promulgate the 
proposed rule. Commenters asserted that Section 301(a)(1) explicitly 
authorizes the EPA Administrator ``to prescribe such regulations as are 
necessary to carry out his functions'' under the statute, noting the 
D.C. Circuit holding that Section 301(a)(1) ``is sufficiently broad to 
allow the promulgation of rules that are necessary and reasonable to 
effect the purposes of the Act.'' NRDC v. EPA, 22 F.3d 1125, 1148 (D.C. 
Cir. 1994). Commenters further noted how consistency and transparency 
advance the goals of the CAA. Other commenters argued that Section 
301(a)(1) was not an adequate authority as the rule was not necessary, 
noting that Section 301(a)(1) does not provide the Administrator 
``carte blanche authority to promulgate any rules, on any matters 
relating to the Clean Air Act, in any manner that the Administrator 
wishes,''' and only permits ``the promulgation of rules that are 
necessary and reasonable to effect the purposes of the Act.'' Id.
    The EPA agrees with the commenters stating that Section 301(a)(1) 
of the CAA provides adequate authority for this final rulemaking. The 
EPA has determined that the authority in Section 301(a)(1) extends to 
internal agency procedures that increase the Agency's ability to 
provide consistency and transparency to the public in regard to the 
rulemaking process under the CAA. In NRDC, the court stated that 
``[a]lthough section 301 does not provide the Administrator `carte 
blanche authority to promulgate any rules, on any matter relating to 
the Clean Air Act, in any manner that the Administrator wishes,' 
Spencer County, 600 F.2d at 873, it is sufficiently broad to allow the 
promulgation of rules that are necessary and reasonable to effect the 
purposes of the Act.'' Id. Further finding that ``[w]here, as here, 
Congress has erected no clear impediment to the issuance of binding 
rules, section 301 takes the agency as far as the second step of 
Chevron. Once there, the EPA provided a reasoned explanation for 
resorting to rulemaking.'' Id. Likewise, the Agency is not aware of any 
clear

[[Page 84138]]

impediment to this rulemaking and this preamble provides a reasoned 
explanation of the purpose and need for this rulemaking.
    The Agency believes that the information provided as a result of 
the procedural requirements of this rule will increase transparency and 
consistency across CAA rulemakings; provide the public with additional 
information in the CAA rulemaking process; and provide the Agency with 
supplemental information for use by the Agency when it is appropriate 
to be considered. These outcomes will better allow the Agency to 
fulfill the purpose described in Section 101(b)(1) of the CAA ``to 
protect and enhance the quality of the Nation's air resources so as to 
promote the public health and welfare and the productive capacity of 
its population''. Further, Section 101(c) of the CAA states that ``a 
primary goal of [the Act] is to encourage or otherwise promote 
reasonable Federal, State, and local governmental actions, consistent 
with the provisions of [the] Act, for pollution prevention.'' As noted 
above, the Supreme Court has stated that ``reasonable regulation 
ordinarily requires paying attention to the advantages and the 
disadvantages of agency decisions.'' Michigan v. EPA, 135 U.S. 2699, 
2707 (2015). The information provided as a result of the procedural 
requirements of this rule will be in addition to the information 
provided by other methodologies and analyses as directed by specific 
CAA statutes and regulations. Such an approach is consistent with 
reasonable rulemaking standards.
    The EPA also received public comments asking for clarification as 
to whether the procedures in this final rule are enforceable against 
the Agency. The EPA received comments arguing that the procedures in 
this final rule are enforceable against the agency and comments that 
such procedures would not be and asking for clarification. The EPA 
agrees with commenters asserting that the procedures in this final rule 
are enforceable against the Agency. Generally, a court reviews an 
agency's compliance with its regulations, even where the regulatory 
requirements go beyond what is required by statute. See, e.g., Service 
v. Dulles, 354 U.S. 363, 388 (1957) (``While . . . the Secretary was 
not obligated to impose upon himself these more rigorous substantive 
and procedural standards, neither was he prohibited from doing so, as 
we have already held, and having done so he could not, so long as the 
Regulations remained unchanged, proceed without regard to them.''). See 
generally Wright & Miller, 32 FED. PRAC. & PROC. JUDICIAL REVIEW Sec.  
8165 (1st ed. Oct. 2020 Update) (``One of the most firmly established 
principles in administrative law is that an agency must obey its own 
rules.''). See also, e.g., United States v. Nixon, 418 U.S. 683, 696 
(1974) (``So long as this regulation remains in force the Executive 
Branch is bound by it, and indeed the United States as sovereign 
composed of the three branches is bound to respect and to enforce 
it.''); Vitarelli v. Seaton, 359 U.S. 535, 540 (1959); United States ex 
rel. Accardi v. Shaughnessy, 347 U.S. 260, 266-67 (1954). Indeed, many 
courts have enforced non-legislative procedural rules against the 
agency. See, e.g., Morton v. Ruiz, 415 U.S. 199, 235 (1974) (enforcing 
an agency manual even though the manual was not a ``legislative rule'' 
but ``solely an internal-operations brochure intended to cover policies 
that do not relate to the public,'' because ``[b]efore the BIA may 
extinguish the entitlement of these otherwise eligible beneficiaries, 
it must comply, at a minimum, with its own internal procedures.''); 
NRDC v. Perry, 940 F.3d 1072, 1077 (9th Cir. 2019). Thus, the Agency 
believes that this Final Rule is binding upon the Agency for 
significant CAA regulations, and that EPA's compliance with these 
procedural requirements is subject to judicial review in challenges to 
such rulemakings.
    Finally, the EPA received comments that the proposed rule was a 
procedural rule and comments, to the contrary, that the proposed rule 
was non-procedural because it altered the rights and interests of 
parties beyond EPA. The EPA disagrees with commenters asserting that 
the proposed rule was non-procedural because it altered the rights and 
interests of parties beyond EPA. The D.C. Circuit has explained that 
``the critical feature of a rule that satisfies the so-called 
procedural exception [to the APA's notice and comment requirements] is 
that it covers agency actions that do not themselves alter the rights 
or interests of parties, although it may alter the manner in which the 
parties present themselves or their viewpoints to the agency.'' James 
A. Hurson Assocs. v. Glickman, 229 F.3d 277, 280 (D.C. Cir. 2000); 
National Mining Association v. McCarthy, 758 F.3d 243 (D.C. Cir. 2014) 
(holding that EPA's interagency plan for enhanced consultation and 
coordination is a procedural rule because it does not alter the rights 
or interests of parties); Batterton v. Marshall, 648 F.2d 708 (D.C. 
Cir. 1980) (``The critical question is whether the agency action 
jeopardizes the rights and interests of parties.''). In addition, the 
Supreme Court explained in Chrysler Corp. v. Brown, that rules of 
internal agency management are considered procedural rules as opposed 
to substantive rules under the APA. 441 U.S. 281, 301-02 (1979). As the 
Supreme Court explained in Chrysler Corp., ``the central distinction 
among agency regulations found in the APA is that between `substantive 
rules' on the one hand and `interpretive rules, general statements of 
policy, or rules of agency organization, procedure, or practice on the 
other.' '' 441 U.S. at 301. The Supreme Court further clarified that 
unlike procedural rules, substantive rules have legal force and effect 
on individual rights and obligations, and noted that whether a rule 
affects individual rights and obligations is an ``important 
touchstone'' for distinguishing substantive rules from other types of 
rules. Chrysler Corp., 441 U.S. 281 at 302.
    Because this rule covers requirements that apply to the agency's 
rulemaking procedure and does not impose any obligations or grant any 
rights to third parties, it is procedural.
    In this Final Rule, the EPA does not interpret or apply other 
provisions of the CAA. Subsequent substantive CAA rulemakings applying 
this rule will be subject to judicial review. By contrast, in this 
action, the EPA finalizes a rule governing internal agency procedures. 
This rule does not require any outside entity to take any action. 
Further, this rule would not regulate the conduct or determine the 
rights of any entity outside the federal government in the manner 
described above. Several comments noted that the rule would potentially 
create an enforcement mechanism were the Agency to fail to follow its 
own internal procedures. The Agency, as discussed above, believes that 
this Final Rule is binding upon the Agency for significant CAA 
regulations, and EPA's compliance with these procedural requirements is 
subject to judicial review in challenges to such rulemakings. However, 
this does not render a rule non-procedural. As discussed above, courts 
have generally enforced non-legislative procedural rules against 
agencies. Commenters assert that such enforcement in turn renders the 
rule non-procedural. If enforcement of a procedural rule rendered the 
rule substantive, there could be no history of enforcement of 
procedural rules; all such rules would simply be substantive. Clearly 
this cannot be the standard. The rule itself must alter the rights and 
interests of parties beyond EPA, rather than simply be binding upon the 
Agency, and this final rule does not regulate any party

[[Page 84139]]

outside of the EPA, but, rather, exclusively governs the EPA's internal 
procedure.

C. Definitions

    Several commenters and the SAB provided specific recommendations 
for changes to some of the definitions in the proposed rule. Examples 
of terms that commenters or the SAB provided specific definitions for 
include, but are not limited to, ``Benefit-cost analysis (BCA)'', 
``Opportunity cost,'' ``Social benefits,'' ``Compliance cost,'' 
``Regulatory Options'', and ``Significant'' regulation. These 
commenters provided references for their suggested definitions, which 
included guidance published by OMB, the EPA's Guidelines, and published 
economic journal articles, and they recommended that the EPA finalize 
the rule with these definitions. Discussed below are the definitions 
that we are revising or finalizing as proposed based on the comments 
received. Complete responses to other specific suggestions for 
additional terms to be defined are provided in Chapter 10 of the 
Response to Comments document, and in some of the remaining sections in 
this preamble where relevant.
    Baseline. The EPA did not receive specific suggestions in the 
public comments on the definition of baseline. However, based on 
feedback from the EPA SAB on the EPA Guidelines update, the EPA has 
decided to adopt a minor revision to the definition to clarify that it 
provides the counterfactual situation against which a policy should be 
assessed. The revision does not change the substantive meaning of the 
term. In the final rule, the definition of baseline is as follows: 
``Baseline means the best assessment of the way the world would evolve 
absent the regulation. It is the primary point of comparison for 
assessing the effects of the regulatory options under consideration.''
    Benefit-cost analysis (BCA). Some commenters recommended that EPA 
provide a more detailed definition of benefit-cost analysis. For 
example, one commenter claimed that as written, ``benefit-cost 
analysis'' lacks clarity, because a key term ``favorable effects of a 
policy action'' is undefined. The commenter further argued that 
evaluation of a benefits-cost analysis is incomplete without concise, 
clear directive to the EPA on what favorable effects may balance 
opportunity costs.
    In their review of the proposed rule, the SAB recommended that the 
definition for BCA be revised to more closely align with the definition 
provided in OMB's Circular A-4. Specifically, the SAB recommended 
revising the definition to clearly state that BCA provides decision 
makers with a clear indication of the most efficient alternative, that 
is, the alternative that generates the largest net benefits (benefits 
minus costs) to society (ignoring distributional effects) (OMB, 2003). 
The SAB also recommended that the definition should indicate that costs 
should be opportunity costs and benefits represent the willingness-to-
pay for a policy outcome valued by United States individuals.
    The EPA agrees with the SAB and public comments that it would be 
helpful to provide a more comprehensive definition of BCA, drawing 
language more explicitly from OMB's Circular A-4 and avoiding undefined 
phrases such as ``favorable effects''. Thus, in this final rule the 
definition of BCA is revised to eliminate the phrase ``favorable 
effects.'' The definition is also expanded to clarify that the social 
benefits of a policy are measured by society's willingness-to-pay for 
the policy outcome, and the social costs are measured by the 
opportunity costs of adopting the policy. Finally, the definition 
explains that where all benefits and costs can be quantified and 
expressed in monetary units, BCA provides decision makers with a clear 
indication of the most economically efficient alternative, that is, the 
alternative that generates the largest net benefits to society 
(ignoring distributional effects).
    The EPA does not agree with the SAB's recommendation to add 
``valued by United States individuals'' because limiting the geographic 
scope of a BCA does not belong in a general definition of BCA. OMB 
Circular A-4 allows impacts accruing to non-U.S. populations to be 
estimated and reported separately: ``Where you choose to evaluate a 
regulation that is likely to have effects beyond the borders of the 
United States, these effects should be reported separately'' (OMB 
2003). The EPA is including in this final rule a presentational 
requirement consistent with this guidance. See Section V.F of this 
Preamble.
    Compliance cost. One commenter stated that the definition provided 
in the proposed rule fails to include all necessary costs of 
compliance, because costs of professional service and interrelated 
effects appear to be excluded. While the EPA believes that the 
definition provided in the proposed rule was broad enough to cover all 
private costs associated with compliance, the final rule revises the 
definition to explain that this could include, for instance, costs 
incurred through planning, design, installation and operation of 
pollution abatement equipment.
    Data. The EPA received limited specific suggestions in the public 
comments on the definition of data. Some commenters expressed concern 
that this language could be interpreted to exclude anonymized medical 
data from the definition of ``data'' and therefore preclude use of 
studies relying on such medical data in the EPA's BCAs. The EPA notes 
that the proposed definition for ``data'' is consistent with the EPA's 
``Strengthening Transparency in Pivotal Science Underlying Final 
Significant Regulatory Actions and Influential Scientific Information'' 
rulemaking.\23\ Therefore, the EPA is finalizing this definition as 
proposed to maintain consistency with related EPA actions.
---------------------------------------------------------------------------

    \23\ https://www.epa.gov/osa/strengthening-transparency-regulatory-science.
---------------------------------------------------------------------------

    Expected value. The EPA did not receive specific suggestions in the 
public comments on the definition of expected value. However, based on 
feedback from the EPA SAB on the EPA Guidelines update, the EPA has 
decided to expand the definition for clarity. The revision does not 
change the substantive meaning of the term. In the final rule, the 
definition of expected value is as follows: ``Expected value means the 
probabilistically weighted outcome that defines a statistical mean and 
a measure of the central tendency of a set of data. For a variable with 
a discrete number of outcomes, the expected value is calculated by 
multiplying each of the possible outcomes by the likelihood that each 
outcome will occur and then summing all of those values.''
    Model. The EPA did not receive specific suggestions in the public 
comments on the definition of model. Therefore, the EPA is finalizing 
the definition as proposed.
    Opportunity cost. One commenter recommended that the EPA expand the 
definition of opportunity cost to explain how other concepts like 
willingness to pay capture the notion of opportunity cost. Further 
discussion of opportunity cost and how to measure it is provided in 
section V.E.5 of this Preamble. The EPA disagrees that an expanded 
definition of this term is needed in the regulatory text. Therefore, 
the EPA is finalizing this definition as proposed.
    Publicly available. The EPA did not receive specific suggestions in 
the public comments on the definition of publicly available. Therefore, 
the EPA is finalizing this definition as proposed.
    Regulatory options. One commenter criticized the proposed 
definition of

[[Page 84140]]

``regulatory options'' for bracketing the selected proposed or final 
option with one more stringent alternative and one less stringent 
alternative. In the commenter's view, this bracketing results in 
biasing the EPA in favor of ultimately choosing central options rather 
than a more environmentally protective one that is more consistent with 
statutory guidance or requirements. In their review of the proposed 
rule, the SAB recommended that the definitions for regulatory options 
be revised to make clearer that for BCA, as opposed to cost-
effectiveness analysis, the regulatory options should only help to 
solve a problem, not accomplish a goal or objective. For example, a 
less stringent option might accomplish less, but at lower cost.
    The EPA disagrees with the comment that analyzing one more 
stringent and one less stringent alternative than the selected option 
biases the Agency's decision. The analysis of these alternative options 
provides the public and decision makers information about the 
consequences of options that are more or less stringent than the 
selected option. The EPA agrees with the SAB's comment and is adopting 
the SAB recommended revisions to the definition to improve clarity. 
Specifically, the EPA is revising parts of the definition of regulatory 
options to clarify that the options should only help to solve a 
problem, not accomplish a goal or objective. For example, the 
definition describes a more stringent option as one that ``contributes 
to'' the stated objectives of the Clean Art Act and achieves additional 
benefits (and presumably costs more) beyond those realized by the 
proposed or finalized option.
    Sensitivity Analysis. The EPA did not receive specific suggestions 
in the public comments on the definition of sensitivity analysis. 
Therefore, the EPA is finalizing this definition as proposed.
    Significant regulation. Several commenters were broadly supportive 
of the proposed definition of ``significant regulation''. Additionally, 
several commenters supported the concept that the definition of a 
``significant regulation'' should include ``those that would 
disproportionately affect an industry, group or area'' or ``those that 
are novel or relevant for other policy reasons,'' with one commenter 
arguing that such inclusion is important to avoid adverse impacts on 
small businesses. One commenter stated that the E.O. 12866 language 
should be inserted into the BCA rather than referencing E.O. 12866, 
because executive orders can be changed or withdrawn in the future.
    Some commenters advocated using the definition of ``significant'' 
from the Congressional Review Act (CRA). The commenters argued that 
adopting a definition from U.S. law is preferable to one from an 
executive order. Furthermore, the commenters also argued that the CRA 
is not limited to a narrow economic impact analysis that ignores the 
indirect impacts of a regulation on the broader economy. The commenters 
further stated that the EPA's economic impact statements for any 
significant proposal should be consistent with the CRA and give 
approximate quantitative estimates of the potential economic impacts, 
the expected timing of these impacts, and the sectors of the economy 
that will experience the impact.
    Several commenters objected to giving the Administrator the 
discretion to decide what constitutes a significant regulation, because 
with no specific decision criteria specified in the rule, the decisions 
would be arbitrary and contrary to the stated goals of the BCA rule for 
consistency and transparency. And some commenters expressed opposition 
to expanding rules requiring a BCA because it would deplete the EPA's 
analytic, financial, and expertise resources without providing any 
benefit to public health or the environment.
    As discussed in more detail below, after reviewing the comments on 
applicability, in this final rule, EPA maintains the same definition of 
significant regulation as in the proposal and concludes it represents 
an appropriate scope for the rule. Specifically, EPA requires that all 
future significant proposed and final regulations promulgated under the 
CAA be accompanied by a BCA using the definition that a significant 
regulation is a proposed or final regulation that is determined to be a 
``significant regulatory action'' pursuant to E.O. 12866 Section 3(f) 
\24\ or is otherwise designated as significant by the Administrator. 
Regulations meeting either of these factors are generally those that 
the EPA anticipates would have the largest annual impact on the economy 
(i.e., greater than $100 million) or are important to analyze for other 
policy reasons. For example, a rule projected to have less than a $100 
million annual effect on the economy could disproportionately affect a 
single industry, population subgroup, or geographic area. Such rules, 
or ones that are notably novel or significant for other policy reasons, 
will benefit from rigorous analysis to inform the public and decision 
makers about the magnitude and disposition of both their benefits and 
costs on affected entities.
---------------------------------------------------------------------------

    \24\ Separate from and independent of the requirements in this 
rulemaking, E.O. 12866 establishes broadly applicable conditions for 
regulatory analysis. More specifically, section 6 of E.O. 12866 
establishes the analytic requirements for those actions OIRA 
determines to be a ``significant regulatory action'' and 
``significant regulatory actions within the scope of section 
3(f)(1).'' Sec. 6(a)(3)(B)-(C).
---------------------------------------------------------------------------

    Social benefits, or benefits. One commenter argued that the 
definition of ``social benefit or benefits'' is overly broad and vague. 
Another recommended an expanded definition that included discussion of 
how to measure benefits. Another said the EPA's definition is arbitrary 
and capricious and potentially unlawful because the proposed definition 
of ``social costs'' included the ``sum'' of all costs, but the proposed 
definition of social benefits, did not. The commenter contended that 
this apparent direction to include all costs but not necessarily all 
benefits would be inconsistent with the general principles of BCA and 
would bias any such analyses. The EPA did not intend to create a 
disparity between the calculations of costs and benefits, so the Agency 
is adjusting the definition of social benefits to be consistent with 
the phrasing of the definition of social costs to avoid any confusion. 
In this final rule, social benefits, or benefits, means ``the sum of 
all positive changes in societal well-being experienced as a result of 
the regulation or policy action.'' Additional discussion of how 
benefits can be measured is provided in section V.E.5 of this Preamble.
    Social costs, or costs. One commenter recommended an expanded 
definition of social cost to elaborate on how costs are measured. In 
this final rule, the EPA is adding a second sentence to the definition 
of social costs to further clarify what is included in opportunity 
costs. Additional discussion of how these costs can be measured is 
provided in section V.E.5 of this Preamble.

D. Preparation and Consideration of BCA in Rulemaking

    In the proposed rule, the EPA proposed to require that all future 
significant proposed and final regulations promulgated under the CAA be 
accompanied by a BCA. Commenters supportive of the proposal were 
generally supportive of conducting BCA for all significant regulatory 
actions, though some commenters argued for a less expansive approach 
and others argued for broader application than the proposal. For 
example, as discussed above, some commenters argued that the

[[Page 84141]]

EPA should use the definition of significant from the CRA. Other 
commenters recommended expanding the scope, for example, to (1) apply 
not only to BCA, but also to any related risk assessment to estimate 
both baseline risk and the risk-reduction benefits estimated in the 
BCA, and (2) clarify that its information quality standards apply to 
BCA, risk assessments, and related risk analyses (e.g., IRIS 
assessments). Commenters opposed to the proposal found the scope too 
expansive and questioned the resource burden of the requirements.
    After considering these comments, the EPA is finalizing the 
requirement that all future significant proposed and final regulations 
promulgated under the CAA be accompanied by a BCA. The EPA believes 
that in keeping with OMB's Circular A-4 and Executive Order 12866 that 
this requirement would create consistency with well-understood and 
established processes and determinations for what constitutes a 
``significant'' rulemaking. Therefore, in this final rule, a 
significant regulation will include any proposed or final regulation 
that is determined to be a ``significant regulatory action'' pursuant 
to Section 3(f) E.O. 12866 or is otherwise designated as significant by 
the Administrator.
    At proposal, in addition to proposing the preparation of a BCA for 
all significant regulation, the EPA also solicited comment on how or 
whether the results of the BCA should inform significant CAA regulatory 
decisions. The EPA requested comment on how the Agency ``could take 
into consideration the results of a BCA in future rulemakings under 
specific provisions of the CAA.'' 85 FR 35624. The EPA received 
numerous comments including recommendations that the Agency formulate a 
mandatory test that the benefits justify the costs of future 
significant rulemakings subject to this final rule, recommendations 
that the Agency not address how BCAs would be taken into consideration 
in future rules, and recommendations that no final rule be promulgated. 
Several commenters noted the importance of BCA and how it can inform 
decision makers. Commenters emphasized that consideration of benefits 
and costs is part of long held requirements imposed by executive order. 
As one commenter summarized, ``the clear direction of every president 
over the last four decades [is] that, to the extent permitted by law, 
executive agencies `shall . . . propose or adopt a regulation only upon 
a reasoned determination that the benefits of the intended regulation 
justify its costs.' '' In addition, the proposal highlighted the 
historical use of BCA by courts to inform their view of the 
appropriateness of agency actions and that ``[c]onsideration of cost 
reflects the understanding that reasonable regulation ordinarily 
requires paying attention to the advantages and the disadvantages of 
agency decisions.'' Michigan v. EPA, 135 U.S. 2699, 2707 (2015), see 85 
FR 35615-617.
    Based on the comments received, executive orders, and judicial 
decisions, the EPA has determined that, when permitted for 
consideration under the specific provision of the CAA under which a 
future regulation is promulgated, the Agency should consider in the 
decision-making process the BCA developed pursuant to this Final Rule, 
which would be part of the record of such a future rulemaking. See 42 
U.S.C. 7607(d)(9); 5 U.S.C. 706(2); see also Motor Vehicles Mfrs. Ass'n 
v. State Farm Mut. Auto. Ins. Co., 463 U.S. 29, 43 (1983) (``Normally, 
an agency rule would be arbitrary and capricious if the agency has 
relied on factors which Congress has not intended it to consider, 
entirely failed to consider an important aspect of the problem, offered 
an explanation for its decision that runs counter to the evidence 
before the agency, or is so implausible that it could not be ascribed 
to a difference in view or the product of agency expertise.''). The 
benefits and costs of a potential regulation, when permitted to be 
considered under the specific provision of the CAA under which a future 
regulation is promulgated, are of clear importance to decision-making 
and can provide justification for whether and how the Agency decides to 
regulate. Consideration of the results of BCA in regulatory decision-
making is also consistent with the requirements of E.O. 12866. However, 
the EPA declines to formulate a specific test or mandate of how to 
consider the BCA or what weight it should be given in such a future 
rulemaking. The precise details of what test would be appropriate could 
differ from one CAA provision to another, and the EPA has not proposed 
or requested comment on how such tests would be formulated under those 
specific provisions. Some commenters also expressed concern that the 
rule as proposed would limit or prohibit the Agency from considering 
other metrics or analyses, either generated by the Agency or submitted 
by commenters into the record of a future rulemaking proceeding. There 
is nothing in this final rule that would create such an outcome, as 
consideration of one metric does not bar consideration of another; 
commenters will retain the ability to provide the Agency with 
information, and the Agency will be required to consider such 
information and respond to comment as is dictated by the process 
governing the future CAA rulemaking. To provide the public with as much 
information and transparency as possible, the EPA is finalizing a 
requirement to identify when the CAA provision or provisions under 
which the future rule is promulgated permit consideration of the BCA, 
and if so, the Agency is required to provide a description in the 
preamble of how the Agency considered the results of the BCA. If the 
provision or provisions under which the rule is promulgated prohibit 
the consideration of the BCA, the final rule requires the Agency to 
identify the specific provision that bars such consideration.

E. Best Practices for the Development of BCA

    The EPA received a wide range of comments on the proposed 
requirements to codify best practices for the development of the BCA 
into a procedural regulation. In its review of the proposed rule, the 
SAB sought to limit its review to requirements in the proposed rule 
that would not be addressed by the SAB's review of the forthcoming 
update to the EPA's Guidelines. Therefore, the SAB did not advise on 
the details of each BCA best practice that the EPA proposed to codify. 
However, the SAB did emphasize that the EPA should consider carefully 
which aspects of BCA should be included in the final rule versus which 
aspects should be addressed in guidance, given the case-by-case nature 
of BCA. The EPA appreciates all the comments received and agrees with 
the SAB that it is important to think carefully about which best 
practices should be made enforceable and which best practices (or 
details thereof) should be addressed in guidance. The best practices 
codified in this final rule include the high-level best practices in 
conducting regulatory BCA. The EPA's Guidelines will continue to 
provide detailed guidance on how to implement these best practices. The 
EPA does not expect the forthcoming update of the EPA's Guidelines to 
include any changes to these high-level elements. We respond to some of 
the major comments in the discussions in the subsections below and to 
the rest in Chapter 7 of the Response to Comments Document.
    After reviewing the comments, the EPA has included in this final 
rule the requirements outlined in the following subsections, which are 
the high-level best practices outlined in existing peer-

[[Page 84142]]

reviewed OMB and EPA guidance documents developed in response to 
longstanding presidential orders discussed above, OMB's Circular A-4 
(2003) and its associated guidance (2010, 2011a, 2011b),\25\ EPA's 
Guidelines (2010). These guidance documents are grounded in the 
economics literature pertaining to the conduct of BCA. Benefit-cost 
analysis as a discipline is a branch of applied microeconomic welfare 
economics and is summarized in numerous textbooks such as Boardman et 
al. (2018), Farrow (2018), Brent (2006), Mishan and Quah (2007), and 
Hanley and Spash (1996).\26\ This discipline is applied routinely to 
environmental economics issues and the theory of BCA and its 
application can be found in standard environmental economic textbooks 
such as Phaneuf and Requate (2016) and Perman et al. (2012).\27\ 
Specific lists of best practices and guidance for practitioners can 
also be found in articles by Robinson and Hammit (2016), Sunstein 
(2014), Farrow (2013), Farrow and Viscusi (2011), Krutilla (2005), and 
notably in an article on the principles and standards by Nobel laureate 
Kenneth Arrow and a number of prominent economists (Arrow et al., 
1996).\28\
---------------------------------------------------------------------------

    \25\ Office of Management and Budget, U.S., 2003. Circular 
A[hyphen]4: Regulatory Analysis. Office of Management and Budget, 
U.S., 2010. Agency Checklist: Regulatory Impact Analysis. Office of 
Management and Budget, U.S., 2011a. Circular A-4, ``Regulatory 
Analysis'' Frequently Asked Questions (FAQs). Office of Management 
and Budget, U.S., 2011b. Circular A-4, ``Regulatory Impact Analysis: 
A Primer''.
    \26\ Farrow, S. ed., 2018. Teaching Benefit-Cost Analysis: Tools 
of the Trade. Edward Elgar Publishing. Brent, R.J. ed., 2004. 
Applied Cost-Benefit Analysis. Edward Elgar Publishing. Mishan, E.J. 
and Quah, E., 2007. Cost-benefit analysis. Routledge. Hanley, N. and 
Spash, C., 1996. Cost benefit analysis and the environment.
    \27\ Phaneuf, D.J. and Requate, T., 2016. A course in 
environmental economics: Theory, policy, and practice. Cambridge 
University Press. Perman, R., Ma, Y., McGilvray, J. and Common, M., 
2003. Natural resource and environmental economics. Pearson 
Education. Krutilla, K., 2005. Using the Kaldor[hyphen]Hicks tableau 
format for cost[hyphen]benefit analysis and policy evaluation. 
Journal of Policy Analysis and Management: The Journal of the 
Association for Public Policy Analysis and Management, 24(4), 
pp.864-875.
    \28\ Robinson, L.A. and Hammitt, J.K., 2013. Skills of the 
trade: Valuing health risk reductions in benefit-cost analysis. 
Journal of Benefit-Cost Analysis, 4(1), pp.107-130. Sunstein, C.R., 
2014. The real world of cost-benefit analysis: Thirty-six questions 
(and almost as many answers). Columbia Law Review, pp.167-211. 
Farrow, S., 2013. How (not) to lie with benefit-cost analysis. The 
Economists' Voice, 10(1), pp.45-50. Farrow, S. and Viscusi, W.K., 
2011. Towards principles and standards for the benefit-cost analysis 
of safety. Journal of Benefit-Cost Analysis, 2(3), pp.1-25.
---------------------------------------------------------------------------

    Since best practices for the conduct of BCA inherently require that 
the inputs to the analysis reflect the best available information,\29\ 
the EPA is also finalizing the requirement that the EPA follow certain 
best practices regarding the incorporation of information as an input 
to BCA for significant CAA regulations. In particular, risk assessments 
often provide key inputs to the development of the EPA's health benefit 
estimates in a BCA, and several commenters recommended that additional 
consistency and transparency be applied in the assessment of risks 
leading to the estimation of benefits. Through this rulemaking, the EPA 
requires a consistent and transparent use of risk assessments in BCA of 
CAA regulations. These requirements include elements that are 
responsive to recommendations from the National Academies of Science, 
Engineering and Medicine (hereafter, ``National Academies'') and the 
EPA's SAB to improve the utility of risk assessment for use in BCAs for 
CAA regulations, as well as recommendations offered by the SAB in their 
review of the proposed rule. As an example, the National Academies has 
previously provided advice to the Agency regarding best practices for 
selecting concentration-response parameters, when it is appropriate to 
pool (or, combine) risk estimates and how to characterize uncertainty 
in those estimates. This rule is also consistent with the 2007 OMB and 
Office of Science and Technology Policy's Updated Principles for Risk 
Analysis,\30\ which also builds off the National Academies and SAB 
recommendations as well as the EPA's Risk Characterization 
Handbook.\31\
---------------------------------------------------------------------------

    \29\ See EPA, Guidelines for Ensuring and Maximizing the 
Quality, Objectivity, Utility and Integrity of Information 
Disseminated by the Environmental Protection Agency (https://www.epa.gov/sites/production/files/2019-08/documents/epa-info-quality-guidelines_1.pdf).
    \30\ https://www.whitehouse.gov/sites/whitehouse.gov/files/omb/memoranda/2007/m07-24.pdf.
    \31\ https://www.epa.gov/risk/risk-characterization-handbook 
(EPA 100-B-00-002, December 2000).
---------------------------------------------------------------------------

    1. Key elements of a BCA. The EPA did not receive comments on the 
proposed requirement that a BCA should include three key elements. The 
specific comments received on each element are provided in the 
corresponding subsections below. Therefore, EPA is finalizing the key 
elements of a BCA as proposed. The key elements of a rigorous 
regulatory BCA include: (1) A statement of need; (2) an examination of 
regulatory options; and (3) to the extent feasible, an assessment of 
all benefits and costs of these regulatory options relative to the 
baseline (no action) scenario.
    2. Statement of Need. Some commenters supported the EPA requiring a 
statement of need in the BCA stating that the requirement is consistent 
with agency guidance detailed in OMB's Circular A-4 and Executive Order 
12866. These commenters argued that a concise and coherent statement of 
need helps to set the foundation for developing the subsequent analysis 
of benefits and costs, particularly as it relates to assessing 
environmental or public health improvements targeted by the relevant 
statutory provision from which the rule derives its authority.
    Some commenters opposed the EPA requiring a statement of need in 
the BCA. These commenters argued a statement of need would be in 
conflict with many, if not most, of the EPA's rulemaking 
responsibilities under the CAA. Commenters further asserted that a 
citation to the provision of the CAA that requires the rulemaking 
should be sufficient for any statement of need. Furthermore, one 
commenter also argued that the EPA cannot apply the ``statement of 
need'' requirement to rulemakings subject to CAA section 307(d) 
requirements, because CAA section 307(d)(2) already includes a 
requirement that the notice of rulemaking shall be accompanied by ``a 
statement of its basis and purpose.''
    None of the comments received have led the EPA to materially change 
its views from the proposal regarding the requirement for a statement 
of need. The EPA disagrees with the comment that a statement of need 
would conflict with the EPA's rulemaking responsibilities under the 
CAA. There is nothing in this final rule that would create such an 
outcome, since an articulation of the statement of need does not bar 
the Agency from complying with any requirements of the CAA, including 
those of CAA section 307(d)(2). The EPA is codifying into regulation a 
procedure that is already prescribed as a best practice in OMB's 
Circular A-4 (OMB, 1993) and EPA's Guidelines (EPA, 2010), which are 
the existing peer reviewed guidance documents implementing E.O. 12866. 
Therefore, the EPA is finalizing the requirement that each regulatory 
BCA should include a statement of need that provides (1) a clear 
description of the problem being addressed, (2) the reasons for and 
significance of any failure of private markets or public institutions 
causing this problem, and (3) the compelling need for federal 
government intervention in the market to correct the problem. This 
statement sets the stage for the subsequent analysis of benefits and 
costs and allows one to judge whether the problem is being adequately 
addressed by the

[[Page 84143]]

policy. Additional discussion of the regulatory statement of need can 
be found in OMB's Circular A-4 (1993, B. Introduction, The Need for 
Federal Regulatory Action) and the EPA's Guidelines (2010, Chapter 3).
    3. Regulatory Options. Commenters supporting the requirement to 
analyze the benefits and costs of at least three regulatory options 
argued that the proposed requirement provides decision makers and the 
public with important perspective on not only the various options' 
relative impact on net social benefits, but also the sensitivity of 
stringency options on other individual factors that comprise the 
overall forecasts. One commenter further suggested that the Agency also 
consider including a fourth option, the implementation of voluntary 
programs if appropriate to the circumstances.
    Some commenters opposed the requirement to analyze the benefits and 
costs of at least three regulatory options. These comments provided 
various reasons including, but not limited to: The EPA incorrectly 
assumes that a continuum of options is possible; requiring three 
regulatory options may lead to patently inappropriate or otherwise 
unacceptable options; requiring three regulatory options may lead the 
agency to put forward intentionally poor choices; and requiring three 
regulatory options may lead to unintended consequences such as leading 
the agency to evaluate options that are infeasible and impractical.
    None of the comments received have led the EPA to materially change 
its views from the proposal. The EPA is codifying into regulation a 
procedure that is already prescribed as a best practice in OMB's 
Circular A-4 (OMB, 1993) and EPA's Guidelines (EPA, 2010), which are 
the existing peer reviewed guidance documents implementing E.O. 12866. 
These guidance documents provide additional details for how to select 
appropriate regulatory options for evaluation. OMB's Circular A-4 also 
allows for the possibility of evaluating an option whose selection 
would be prohibited under the specific statutory provision under which 
the rule is being promulgated because the identification of these 
statutory constraints and an estimate of their opportunity costs may 
provide useful information to Congress under the Regulatory Right-to-
Know Act. The requirement to analyze at least three regulatory options 
also provides for cases where a continuum of options is not possible, 
which is further clarified below. Finally, there is nothing in this 
final rule that would prevent an additional evaluation of a voluntary 
program to address the problem articulated in the statement of need if 
appropriate to the circumstances. Therefore, the EPA is finalizing the 
requirement that the BCA analyze the benefits and costs of regulatory 
options. The final rule requires the BCA to analyze at least three 
options that contribute to the stated objectives of the CAA (unless the 
BCA explains the rationale for analyzing fewer than three options, as 
further described below) and to explain why they were selected. Where 
there is a continuum of options (such as options that vary in 
stringency), the three options are required to include at a minimum: 
The proposed or finalized option; a more stringent option that achieves 
additional benefits (and presumably costs more) beyond those realized 
by the proposed or finalized option; and a less stringent option that 
costs less (and presumably generates fewer benefits) than the proposed 
or finalized option. When a continuum of options is not applicable, an 
analysis of three regulatory options provides an opportunity to analyze 
a variety of parameters including different compliance dates, 
enforcement methods, standards by size or location of facilities, and 
regulatory designs (e.g., performance vs. technology standards). If 
fewer than three options are analyzed relative to the baseline, or if 
there is a continuum of options and the options analyzed do not include 
at least one more stringent (or otherwise more costly) and one less 
stringent (or otherwise less costly) option than the proposed or 
finalized option, then the final rule requires the BCA to explain why 
it is not appropriate to consider more alternatives. For further 
discussion, see OMB's Circular A-4 (specifically, see section E. 
Identifying and Measuring Benefits and Costs, General Issues, 3. 
Evaluation of Alternatives).
    4. Baseline. Many commenters supported the proposed requirement 
regarding the development of a baseline as consistent with best 
practices for BCA. Several commenters noted that defining the baseline 
scenario is one of the most important elements of a regulatory impact 
analysis, and multiple commenters supported the proposed requirements 
to develop a baseline that appropriately considers relevant factors 
based on transparent and reasonable assumptions. Additionally, some 
commenters supported the explicit use of more than one baseline: ``one 
baseline based solely on current standards and another based on the 
agency's reasoned assumptions regarding the effect of all related 
pending regulations''; and stated that this is consistent with OMB's 
Circular A-4.
    Several commenters stated that the proposed requirements for 
developing a baseline will prevent ``double-counting.'' The commenters 
added that the issue of double counting of benefits has been a 
particular concern with past EPA BCAs under the CAA. Commenters 
referenced a report that found that the simultaneous advancement of 
multiple CAA-related rulemakings resulted in changes between proposed 
and final BCAs' baseline assumptions about implementation of other 
regulations that created inconsistencies in BCA estimates between the 
proposed and final stages and revealed examples of double-counting. One 
commenter suggested that where ancillary benefits exist and have not 
been counted before by the EPA, the EPA must determine the most cost-
effective regulatory means of achieving them. The commenter argued that 
this should ensure that the EPA properly and efficiently utilizes its 
regulatory authorities to achieve optimal results to enhance societal 
well-being.
    Some commenters opposed the requirements for developing a baseline 
in a BCA in the proposed rule as they argued OMB and EPA policies 
already establish the process for establishing a baseline, for assuring 
that benefits will not be double-counted, and for being transparent in 
those explanations. Creating a new rule for the purpose of preventing 
an oversight in a pre-existing mechanism for assessing BCA is 
unnecessarily ``reinventing the wheel.'' The commenters further argued 
the proposed requirements for developing a baseline bias the analyses 
against regulations that otherwise meet statutory requirements and 
provide important environmental benefits, in contravention of the CAA's 
public-health protective mandate.
    Other commenters opposing the proposed requirements contended that 
the EPA provides no specific cases to support its assertion that there 
is a risk of ``double-counting.'' Some of the commenters contended that 
recent research indicates some claimed mechanisms of ``double-
counting'' are either inaccurate or can be addressed by the EPA 
following its own guidelines on BCA baselines assuming full compliance 
with existing rules. The commenters added that the proposed rule 
provides no evidence that there is a gap that needs to be filled in 
this regard beyond its existing guidance, and, in fact, adds no 
additional insight into these issues.
    None of the comments received have led the EPA to materially change 
its

[[Page 84144]]

views from the proposal. The EPA is codifying into regulation a 
procedure that is already prescribed as a best practice in OMB's 
Circular A-4 (OMB 1993) and EPA's Guidelines (EPA 2010), which are the 
existing peer reviewed guidance documents implementing E.O. 12866. 
Nothing in the public comments have suggested specific additional 
factors that should be codified into the final rule as factors to be 
considered when developing the baseline in a BCA. Therefore, the EPA is 
finalizing the requirement to develop a suitable baseline as proposed, 
as described below.
    The baseline in a BCA serves as a basis of comparison with the 
regulatory options considered. It is the best assessment of the way the 
world would look absent the regulatory action. The choice of a baseline 
requires consideration of a wide range of potential factors, including 
exogenous changes in the economy that may affect relevant benefits and 
costs (e.g., changes over time in demographics, economic activity, 
consumer preferences, and technology); impacts of regulations that have 
been promulgated by the agency or other government entities; and the 
degree of compliance by regulated entities with other regulations. 
Accounting for other existing regulations in the baseline is especially 
important in order to avoid double counting of the incremental benefits 
and costs from other existing regulatory actions affecting the same 
environmental condition (e.g., ambient air quality). When the EPA 
determines that it is appropriate to consider more than one baseline 
(e.g., one that accounts for another EPA regulation being developed at 
the same time that would affect the same environmental condition), the 
final rule requires the BCA to provide a reasoned explanation for the 
baselines used and to identify the key uncertainties in the 
forecast(s). These requirements for developing a baseline are 
consistent with best practices as outlined in OMB's Circular A-4 (1993) 
and EPA's Guidelines (2010).
    5. Measuring Benefits and Costs. Some commenters contended that the 
proposal identifies the willingness to pay (WTP) metric as the 
``correct measure'' of changes from the baseline, but the proposal 
fails to acknowledge the existence of other metrics and does not 
justify their exclusion in favor of WTP. One commenter further argued 
the proposal also fails to acknowledge or consider the greater 
difficulty in estimating willingness-to-pay for non-market goods, such 
as air quality and associated health risk. Another commenter further 
added that WTP studies are helpful, but not the only source of 
information for monetizing benefit and WTP studies are particularly 
helpful in estimating the value of mortality risk reduction, which 
typically comprise the bulk of monetized benefits in CAA rules.
    Several commenters opposed including the WTP concept in the 
proposed rule. The commenters expressed concern that the proposed rule 
will continue practices to propagate the understatement of CAA 
benefits, to the detriment of all, but particularly to low-income and 
minority communities. Several commenters stated that WTP is strongly 
affected by factors such as ability to pay and by the awareness of the 
respondent of the harms being inflicted or avoided. A commenter then 
asserted that a WTP analysis will lead to higher measured monetary 
benefits for wealthier communities than for poorer communities for the 
same level of health and wellbeing benefit. At least two commenters 
focused on particular methods used for estimating WTP. These commenters 
advised EPA against using survey approaches to estimate WTP because 
they contend that such studies often overstate WTP that does not align 
with reality.
    None of the comments received have led the EPA to materially change 
its views from the proposal on the appropriate measure of benefits and 
costs in a BCA. The EPA is codifying into regulation a procedure that 
is already prescribed as a best practice in OMB's Circular A-4 (OMB, 
1993) and EPA's Guidelines (EPA, 2010), which are the existing peer 
reviewed guidance documents implementing E.O. 12866. As discussed in 
Section V.B of this Preamble, the EPA agrees with the SAB's 
recommendation, per their review of the proposed rule, to provide more 
clarity in the definition of Benefit-Cost analysis and the measurement 
of benefits and costs. Therefore, in this final rule EPA has provided a 
more fulsome definition of BCA to clarify that it is consistent with 
OMB Circular A-4. The EPA disagrees with commenters who stated that the 
proposed rule did not acknowledge the existence of metrics other than 
willingness-to-pay, as discussed below. In addition, the EPA disagrees 
with commenters who advised to include more discussion in the rule 
about particular methods for estimating WTP. The EPA's Guidelines and 
OMB's Circular A-4 include discussion of particular methods for 
estimating WTP, which can generally be broadly categorized as either 
revealed preference or stated preference methods. As described in these 
guidance documents and standard textbooks on BCA, some methods will be 
more suitable than others in a given scenario for a variety of reasons, 
and some will be better able to capture certain types of benefits than 
others. Since research on all of these methods is ongoing, the 
limitations and qualifications of each method is best described in 
guidance and the EPA has decided not to include any requirements 
related to particular valuation methods in this final rule.
    A BCA evaluates the social benefits and social costs of a policy 
action. The social benefits of a policy are measured by society's 
willingness-to-pay for the policy outcome. The social costs are 
measured by the opportunity costs of adopting the policy. Opportunity 
cost is the value of the next best alternative to a particular activity 
or resource.\32\ A BCA addresses the question of whether the benefits 
from the policy action are sufficient for those who gain to 
theoretically compensate those burdened such that everyone would be at 
least as well off as before the policy. In other words, many 
regulations can be thought of as a requirement to divert resources from 
activities with a higher net return in private markets alone to those 
with a higher net return when all impacts are counted, thus the 
calculation of net benefits (benefits minus costs) helps ascertain the 
economic efficiency of a regulation. Where all benefits and costs can 
be quantified and expressed in monetary units, BCA provides decision 
makers with a clear indication of the most economically efficient 
alternative, that is, the alternative that generates the largest net 
benefits to society (ignoring distributional effects).
---------------------------------------------------------------------------

    \32\ Opportunity cost need not be assessed in monetary terms. It 
can be assessed in terms of anything that is of value to the person 
or persons doing the assessing. For example, a grove of trees used 
to produce paper may have a next-best-alternative use as habitat for 
spotted owls. Assessing opportunity costs is fundamental to 
assessing the true cost of any course of action. In the case where 
there is no explicit accounting or monetary cost (price) attached to 
a course of action, ignoring opportunity costs could produce the 
illusion that the action's benefits cost nothing at all. The unseen 
opportunity costs then become the implicit hidden costs of that 
course of action.
---------------------------------------------------------------------------

    In keeping with best practices, the appropriate measures of 
benefits and costs to use in a regulatory BCA are social benefits and 
social costs. When assessing a regulation, the social benefits are the 
society-wide positive changes in well-being, and social costs are the 
society-wide opportunity costs, or reductions in well-being. WTP is the 
correct measure of these changes in BCA.
    Willingness to pay means the largest amount of money that an 
individual or

[[Page 84145]]

group would pay to receive the benefits (or avoid the damages) 
resulting from a policy change, without being made worse off. The 
principle of WTP captures the notion of opportunity cost by measuring 
what individuals are willing to forgo to enjoy a particular benefit. In 
general, economists tend to view WTP as the most appropriate measure of 
opportunity cost, but an individual's ``willingness-to-accept'' (WTA) 
compensation for not receiving the improvement can also provide a valid 
measure of opportunity cost. WTP is generally considered to be more 
readily measurable. Market prices provide rich data for estimating 
benefits and costs based on WTP if the goods and services affected by 
the regulation are traded in well-functioning competitive markets. See 
Hanley and Spash (1993), Freeman (2003), Just et al. (2005), and 
Appendix A of the EPA's Guidelines (2010).
    WTP provides a full accounting of an individual's preference for an 
outcome by identifying what the individual would give up to attain that 
outcome. WTP is measured in monetary terms to allow a comparison of 
benefits to costs in the net benefit calculation. If the BCA departs 
from these best practices (e.g., where WTP is hard to measure), this 
final rule requires a robust explanation for doing so. For further 
discussion, see OMB's Circular A-4 (specifically, see section E. 
Identifying and Measuring Benefits and Costs, General Issues, 2. 
Developing a Baseline and Guidelines (2010), Chapter 5. Baseline).
    While based on the same underlying conceptual framework, social 
benefits and social costs are often evaluated separately due to 
practical considerations. The social benefits of reduced pollution are 
often attributable to changes in outcomes not exchanged in markets, 
such as improvements in public health or ecosystems. In contrast, the 
social costs generally are measured through changes in outcomes that 
are exchanged in markets. As a result, different techniques are used to 
estimate social benefits and social costs however, in both cases the 
goal is to estimate measures of WTP to provide consistency.
    6. Methods for Estimating Benefits and Costs. The EPA received a 
range of comments on the proposed requirements regarding the methods 
for estimating benefits and costs. Comments were divided on the idea of 
codifying best practices, with many commenters supporting codification 
in a procedural regulation, but others noting possible inconsistency 
when practices are updated in the future.
    Many comments pertained to whether more specific or additional best 
practices should be codified as requirements in the final rule. For 
example, when estimating costs, some recommended that the final rule be 
expanded to include procedural requirements for determining whether an 
engineering base cost estimation, partial-equilibrium model, general 
equilibrium model, or a combination of these models should be used. One 
commenter argued that when a regulation will affect a sector that 
supplies a wide swath of the economy, then the final rule should 
specify that the presumptive cost evaluation method be a general 
equilibrium model, and if a general equilibrium model is not used, then 
the BCA should be accompanied by a detailed explanation of why small 
price effects in the affected sector's outputs would not be expected to 
have economy-wide effects. Others pointed out that systems are so large 
and complex that evaluative tools are not adequate for these types of 
analyses to be accurate and useful for decision-making. Another of 
these commenters said that although the EPA is correct to highlight the 
potential value added to be gained by using general equilibrium models, 
there still are a number of reasons why general equilibrium models may 
not yet be ready to be used as a principal analytic framework for 
undertaking cost-benefit analysis of environmental regulations. The 
commenter argued that general equilibrium models provide insights 
rather than answers about the economic effects of policies; for 
example, general equilibrium models are calibrated using parameter 
estimates to ``fit'' predetermined values providing a certain degree of 
``realism'' but only up to a point.
    Finally, some commenters argued that the proposed rule provided an 
unbalanced treatment of benefits and costs by setting more stringent 
standards for benefit estimation than cost estimation, and therefore, 
aside from being unnecessary and unjustified, they stated the proposed 
requirements were also biased and arbitrary. These commenters' 
recommended solution to the proposed rule's problem of treating costs 
and benefits differently is simply to withdraw the proposed rule and 
revert to relying on existing guidance, like OMB's Circular A-4 and the 
EPA's Guidelines, which already offer a more balanced treatment to both 
costs and benefits. Other commenters stated the proposed rule 
arbitrarily fails to address the likelihood that compliance costs will 
be overestimated and benefits will be underestimated.
    None of the public comments received have led the EPA to materially 
change its views from the proposal. The EPA disagrees with the comments 
that more specific procedures should be codified into regulation 
pertaining to the use of particular estimation methods or models. The 
EPA also disagrees with commenters stating that the rule imposes uneven 
requirements. The EPA is codifying into regulation procedures that are 
consistent with best practices for estimating both benefits and costs 
as discussed at length in OMB's Circular A-4 (OMB 1993) and the EPA's 
Guidelines (EPA 2010), which are the existing peer reviewed guidance 
documents implementing E.O. 12866. In this final rule, the EPA is 
codifying these best practices as proposed, as described below.
    Although the most appropriate methods for estimating social costs 
and social benefits can often be regulation-specific, there are best 
practices for selecting these methods. With this final rule, the EPA 
requires that all BCAs will rely on such best practices and will 
provide reasoned explanations for methods selected. These best 
practices include the use of a framework that is appropriate for the 
characteristics of the regulation being evaluated. As discussed in OMB 
Circular A-4, a good regulatory analysis cannot be developed according 
to a formula. Conducting high-quality analysis requires competent 
professional judgment. Different regulations may call for different 
emphases in the analysis, depending on the nature and complexity of the 
regulatory issues and the sensitivity of the benefit and cost estimates 
to the key assumptions. For example, the extent to which compliance 
cost is a sufficient measure of social costs will depend on whether a 
regulation is expected to result in changes in prices and quantities 
within and across markets. Other considerations when selecting an 
estimation method include the ability of an estimation approach to 
capture certain types of costs, to adequately reflect the geographic 
and sectoral detail and scope of the rule, and to reflect how costs may 
change over time, among other considerations.
    During the estimation process, the final rule requires analysts to 
consider how social cost and benefit endpoints may be affected by 
behaviors in the baseline and potential behavioral changes from the 
policy. For example, three broad frameworks for estimating social 
cost--compliance cost, partial equilibrium, and general equilibrium--

[[Page 84146]]

offer different scopes in terms of the degree to which behavioral 
response and other market imperfections are included. In general, 
analysts can improve the accuracy of cost estimates by reducing known 
biases due to the omission of potentially important behavioral 
responses or missing opportunity costs. However, adopting more complex 
approaches can reduce the precision of estimates due to data and 
modeling limitations. A compliance cost approach typically identifies 
the private expenditures associated with compliance in the regulated 
sector(s). Compliance cost estimates typically exclude behavioral 
responses outside of the choice of compliance activity and may, 
therefore, not capture some opportunity costs associated with 
regulations. However, with adequate data, this approach can generate 
highly detailed and relatively precise information on compliance 
options and costs, reflecting the heterogeneity of regulated entities. 
This can provide a reasonable estimate of the social cost of a 
regulation when changes in the regulated sector's outputs and input mix 
are expected to be minimal and no large market effects are anticipated. 
A partial equilibrium analysis captures supply and demand responses in 
the regulated sector due to compliance activities and may, therefore, 
provide a more complete estimate of compliance costs in addition to any 
lost profits and consumer welfare due to reductions in output. In other 
words, behavioral responses can have important impacts on both the size 
and distribution of benefits and costs, and therefore can provide a 
fuller picture of the social impact of a particular regulation. Partial 
equilibrium analyses may be extended to consider a small number of 
related sectors in addition to those directly regulated (e.g., upstream 
markets that supply intermediate goods to the regulated sector, or 
markets for substitute or complementary products). A partial 
equilibrium approach is preferred for estimating social cost when the 
regulation will result in appreciable behavioral change, but the 
effects will be confined primarily to a single market or a small number 
of markets. When broader economy-wide impacts are expected as a result 
of the regulation, a partial equilibrium approach will miss these 
effects. In this case, a general equilibrium approach may be more 
appropriate to more adequately estimate social cost.
    A general equilibrium approach, which captures linkages between 
markets across the entire economy, is most likely to add value when 
both relevant relationships among sectors and pre-existing market 
distortions are expected to be significant. Market distortions are 
factors such as pre-existing taxes, externalities, regulations, or 
imperfectly competitive markets that move consumers or firms away from 
what would occur in the absence of such distortions. For example, when 
an environmental regulation affects the real wage such that individuals 
opt to work fewer hours, it can exacerbate pre-existing inefficiencies 
in the labor market due to taxes, regulatory barriers, or other market 
imperfections. This represents a welfare cost not captured by 
compliance cost estimates. The impacts of a regulation also may 
interact with pre-existing distortions in other markets, which may 
cause additional impacts on welfare either positively or negatively. In 
cases such as these, a general equilibrium approach may be capable of 
identifying how the costs of complying with a regulation flow through 
the economy, such as through changes in substitution among factors of 
production, trade patterns, and demand for goods and services. These 
effects are partially or wholly missed by compliance cost and partial 
equilibrium approaches. For further discussion, see EPA's Guidelines 
(2010), Chapter 8, Analyzing Costs, 8.1. The Economics of Social Cost.
    The estimated social benefits reported in a BCA should link 
regulatory requirements to the value that individuals place on the 
beneficial outcomes,\33\ or benefit endpoints, that can be meaningfully 
expected as a result of those requirements. Benefits assessment is, 
therefore, typically a multi-step process. The starting point is 
identifying the changes in environmental contaminants or stressors that 
are likely to result from policy options relative to the baseline. 
These changes are often characterized through air quality modeling. The 
next step is to identify the benefit endpoints that may be affected by 
changes in environmental quality, such as human health improvements, 
ecological improvements, aesthetic improvements, and reduced materials 
damages. The EPA recognizes that the strength of scientific evidence 
for different health or environmental endpoints varies, and that 
strength of scientific evidence should be strongest when the benefits 
are estimated. As further discussed in OMB's M 19-15, this concept is 
referred to as ``fitness for purpose,'' whereby information anticipated 
to have a higher impact must be held to higher standards of 
quality.\34\
---------------------------------------------------------------------------

    \33\ As a practical matter, the value of any adverse public 
health or welfare outcomes (sometimes referred to as 
``disbenefits'') resulting from the regulatory requirements are 
usually also included on the benefits side of the ledger in 
regulatory BCAs, although it is theoretically appropriate to include 
them on the cost side. Such adverse outcomes could include adverse 
economic, health, safety, or environmental consequences that occur 
due to a rule (e.g., adverse safety impacts from vehicle emission 
standards) and are not already accounted for in the direct cost of 
the rule.
    \34\ OMB's M-19-15 refers back to OMB's 2002 Guidelines, which 
characterize a subset of agency information as ``influential 
scientific, financial, or statistical information'' that is held to 
higher quality standards. This is scientific, financial, or 
statistical information that ``the agency can reasonably determine . 
. . will have or does have a clear and substantial impact on 
important public policies or important private sector decisions.''
---------------------------------------------------------------------------

    Once benefit endpoints are identified, analysts need to decide 
whether and how to quantify changes in each endpoint. From among the 
endpoints identified above, the EPA will quantify effects for endpoints 
which scientific evidence is robust enough to support such 
quantification. If the Agency determines that some benefits should be 
discussed only qualitatively, for example, due to limited scientific 
evidence or limited resources for developing concentration response 
functions, the final rule requires the Agency to provide a reasoned 
explanation for that decision. Additional requirements for choosing and 
quantifying health endpoints are described further below.
    Quantification is then followed by valuation of these endpoints 
when data and methods allow. There are well-defined economic principles 
and well-established economic methods for valuation as detailed in OMB 
and Agency guidance, including OMB's Circular A-4 and the EPA's 
Guidelines. It will not always be possible to express in monetary units 
all of the important benefits and costs. When it is not, the most 
efficient alternative will not necessarily be the one with the largest 
quantified and monetized net-benefit estimate. In such cases, the EPA 
will exercise its subject matter expertise in determining how important 
the non-quantified benefits or costs may be in the context of the 
overall analysis. Even when a benefit or cost cannot be expressed in 
monetary units, the EPA will try to measure it in terms of its physical 
units. If it is not possible to measure the physical units, the EPA 
will describe material benefits or costs qualitatively.
    Finally, the valued endpoints should be aggregated to the extent 
possible and supported by scientific and economic practice to provide 
the basis for characterizing the benefits of each policy option.

[[Page 84147]]

    In some instances, it may be possible to value bundles of 
attributes or endpoints using reduced-form techniques, such as the 
hedonic property method. Care and professional judgment are necessary 
in determining the appropriateness of bundling of several endpoints 
versus modeling separate endpoints. Even if bundling is thought to be 
appropriate, it can be useful to think through the multi-step process 
above conceptually to: (a) Assess whether there are benefit endpoints 
not reflected in the reduced form valuation estimate that should be 
included through additional analysis, or (b) compare the magnitudes of 
multi-step and reduced-form, revealed-preference benefits estimates so 
that each can provide a check on the reliability of the other.
    In summary, this final rule requires that, to the extent supported 
by the scientific criteria, as discussed above, as well as practicable 
in a given rulemaking, (1) BCAs will quantify all benefits; (2) BCAs 
will monetize all the benefits by following well-defined economic 
principles using well-established economic methods, appropriate data 
and/or studies; and (3) BCAs will qualitatively characterize benefits 
that cannot be quantified or monetized. In addition, the final rule 
requires the Agency to explain any departure from the best practices 
for the BCA described in Circular A-4; this includes discussing the 
likely effect of the departures on the size of the benefits estimate. 
More discussion of these best practices and estimation methods is 
provided in OMB's Circular A-4 and the EPA's Guidelines, and the 
literature cited therein.
    7. Selecting and Quantifying Health Endpoints in a BCA. The EPA 
received numerous comments on the proposed requirements for selecting 
and quantifying health endpoints in a BCA. Many public commenters were 
critical of the lack of definitions for key terms in this section, 
especially ``causal'' and ``likely causal'' though some of these 
commenters supported the proposed requirements while providing more 
specific definitions that could improve the terms. Other commenters 
were generally critical of the proposed requirements that any linkage 
between regulatory requirements and benefits be based on ``a clear 
causal or likely causal relationship'' and argued such requirements 
will restrict the assessment of the health benefits of proposed CAA 
regulations. With respect to determining what concentration-response 
functions to use to quantify changes in the selected endpoints, some 
commenters argued that the proposed criteria for selecting studies from 
the literature are too restrictive. Others recommended that the EPA 
consider different criteria entirely or require a more systematic 
review approach for evaluating the scientific literature to quantify 
health impacts. For example, one commenter noted that while the list of 
proposed criteria referred to study features that should be evaluated 
under a systematic review framework, it was not exhaustive or complete 
and does not provide a systematic approach for the integration of this 
evidence to prioritize studies that provide the accurate 
characterization of health impacts. Some commenters stated that the 
rule would contradict advice the EPA has received from the National 
Academies and SAB and/or questioned why, in their view, the EPA is re-
inventing the wheel. Some commenters emphasized that best practices for 
characterizing uncertainty should reflect more probabilistic techniques 
and that EPA should also use a risk of bias approach when selecting 
among studies.
    In their review of the proposed rule, the SAB also provided 
recommendations related to the selection and quantification of health 
endpoints. First, the SAB recommended that the EPA clarify the 
requirements for estimation of benefits to incorporate systematic 
review approaches, better define causality, and include effects for 
which causal or likely causal relationships may be less certain. In 
particular, the SAB advised that no ``one size fits all'' approach to 
causality should be mandated because a variety of approaches may need 
to be taken (some data driven, some based on systematic review of the 
biology, toxicology and epidemiology). Instead, the SAB recommended 
that the EPA should include reference to and support for relevant 
guidance from current best Agency practices for evaluating causality. 
The SAB also advised that the EPA modify the proposed requirement to 
include in the benefits analyses the effects for which causal or likely 
causal relationships may be less certain, but the impact would be 
substantial.
    Second, the SAB provided recommendations for how the EPA could 
adjust the proposed requirements for selection of health endpoints to 
provide greater clarity and transparency, especially with regard to the 
selection of concentration response functions. The SAB recommended that 
the final rule should clarify the specific scientific rationale for 
endpoint selection and promote transparency by defining specific terms 
used in the requirements, or the Agency should replace all of the 
specific criteria on the selection of health endpoints with ``an 
overall framework outline of the systematic review principles it would 
follow for the evaluation of human health hazard data for the purposes 
of concentration-response selection and quantification of benefits.'' 
The SAB also advised the Agency to discuss how relevant advice from the 
National Academies and the SAB on systematic review as well as the 
approaches under development by the EPA in the Consolidated Human 
Toxicity Assessment Guidelines \35\ will be evaluated and incorporated. 
The EPA agrees with the recommendations from the SAB and commenters on 
the importance of using a systematic review process to evaluate the 
scientific literature for the purposes of determining which health 
endpoints to include in a BCA and what concentration-response functions 
to use to quantify changes in these endpoints. Therefore, the EPA is 
revising the requirements in this section of the rule as described 
below.
---------------------------------------------------------------------------

    \35\ For more information about the development of the 
Consolidated Human Toxicity Assessment Guidelines, see: https://yosemite.epa.gov/sab/sabproduct.nsf//LookupWebProjectsCurrentBOARD/DF0F42C34645448685258570005ADFFF?OpenDocument.
---------------------------------------------------------------------------

    It is essential for analyses to characterize health effects for 
which the science indicates the likelihood that changes in exposure 
would provide positive benefits. The EPA requires that BCAs performed 
under this final rule will include benefit endpoints for which the 
scientific evidence indicates there is (a) a causal or likely causal 
relationship between pollutant exposure and effect, and subsequently, 
(b) sufficient data and understanding to allow the agency to reasonably 
model the anticipated change in that effect in response to changes in 
environmental quality or exposures expected as a result of the 
regulation under analysis.
    As stated in the proposal, decisions about whether and which 
changes in the health endpoints should be quantified should be informed 
by an evaluation of the relevant scientific literature studying the 
strength of the association between exposure to a pollutant and the 
health endpoint and the nature of the concentration-response function 
(i.e., the amount of change in the frequency or severity of the health 
endpoint expected as the distribution of air quality changes). Benefits 
may be quantified for associations that meet the criteria for 
causality, considering, for example, the biologic plausibility, 
consistency, temporality, strength, and specificity of the effect.

[[Page 84148]]

    In this final rule, the EPA is clarifying that for human health 
endpoints, a systematic review process must be used to evaluate the 
hazard data for the purposes of determining which endpoints to include 
in a BCA and what concentration-response functions to use to quantify 
changes in these endpoints. As described by Institute of Medicine 
(IOM), ``systematic review is a scientific investigation that focuses 
on a specific question and uses explicit, pre-specified scientific 
methods to identify, select, assess, and summarize the findings of 
similar but separate studies. The goal of systematic review methods is 
to ensure that the review is complete, unbiased, reproducible, and 
transparent'' (IOM, 2011).
    The systematic review process, at a minimum, consists of: Problem 
formulation and protocol development, evidence identification, evidence 
evaluation, and evidence integration (National Research Council, 2014). 
Problem formulation should identify the specific question to be 
addressed in the review and the protocol should specify the methods 
used to address the question, making these methods and the review 
process transparent. Evidence identification should follow a search 
strategy written into the protocol that explicitly states the inclusion 
and exclusion criteria for studies. Importantly, a study's inclusion in 
the review should not depend upon that study's findings. When feasible, 
the evidence evaluation should include a risk of bias assessment to 
determine how confidently conclusions can be drawn from the data. For 
example, the EPA began incorporating a risk of bias assessment into its 
Integrated Science Assessments (ISAs), starting with the recently 
published ozone ISA (EPA, 2020).\36\ Finally, evidence integration 
should provide a structured approach to drawing conclusions considering 
all appropriate and available lines of scientific evidence, including 
epidemiologic, toxicologic, and mechanistic lines of evidence.
---------------------------------------------------------------------------

    \36\ The EPA prepares ISAs to provide the scientific foundation 
for setting standards for the 6 criteria air pollutants under the 
National Ambient Air Quality Standards program. This assessment is a 
comprehensive review, synthesis, and evaluation of the most policy-
relevant science, including key science judgments that are important 
to inform the development of the risk and exposure assessments, as 
well as other aspects of the NAAQS review. The preamble to the ISAs 
describes the five-level causal framework for evaluating weight of 
evidence and drawing scientific conclusions and causal judgments. 
See https://www.epa.gov/isa.
---------------------------------------------------------------------------

    Applying the systematic review process described above, the final 
rule requires the EPA to identify concentration-response relationships 
from the scientific literature that take into account the breadth and 
quality of the available evidence regarding the nature and magnitude of 
the risk to the populations affected by the regulation. More weight 
should be given to higher quality studies or analyses that have been 
peer reviewed. To the extent possible, the studies or analyses should: 
(1) Be based upon human data when available; (2) specify the exposure 
route, duration, and levels, with preference given to those studies 
assessing exposure similar to those experienced by the general 
population; (3) employ a design or analysis that adequately addresses 
relevant sources of potential critical confounding; (4) consider how 
exposure is measured, particularly those that provide measurements at 
the level of the individual and that provide actual measurements of 
exposure; and (5) be able to reliably distinguish the presence or 
absence (or degree of severity) of health outcomes. Studies 
demonstrating more of the attributes listed above, and those which 
demonstrate the considerations to a greater extent, are expected to 
provide more accurate concentration-response relationships and 
associated risk estimates. Consistent with the general process of 
systematic review, the evaluation should emphasize transparency and 
replicability in the evaluation process.
    When utilizing multiple concentration-response functions to 
estimate impacts on a single health outcome, the BCA must quantify 
risks in such a way that the heterogeneity in the estimated health 
impacts is clearly characterized. The EPA will present results in a 
manner that promotes transparency in the assessment process by 
selecting and clearly identifying concentration-response functions best 
characterizing risk for affected populations, as well as evidence 
necessary to demonstrate the sensitivity of the choice of the 
concentration-response function on the magnitude and the uncertainty 
associated with air pollution-attributable effects. Evidence from 
epidemiologic, experimental, and controlled human exposure studies may 
suggest that certain demographic subgroups are subject to risks that 
differ from the general population; in these instances, it may be 
appropriate to select concentration-response relationships that 
quantify risks among these specific subgroups, abiding by the overall 
framework of the systematic review process.
    In cases where existing Agency documents (e.g., ISA for criteria 
pollutants) provide the review and synthesis consistent with the 
process described above, the final rule allows a BCA to reference this 
synthesis.
    Conceptually, BCA requires a comparison of expected costs and 
expected benefits, so BCA for CAA regulations should include the 
determination of expected benefits. When sufficient data exist, a 
probability distribution of risk is appropriate to use when determining 
the expected benefits for CAA regulations. When it is infeasible to 
estimate a probability distribution, measures of the central tendency 
of risk may be used. Upper-bound risk estimates must not be used 
without also presenting lower bound and central tendency estimates.
    8. Uncertainty Analysis. Many public commenters supported the 
proposed rule's codification of best practices for uncertainty analysis 
and further contended that the EPA's past uncertainty analyses in CAA 
BCA vary in their quality, scope, and rigor. Some of these commenters 
provided additional recommendations for uncertainty analyses in the BCA 
including using probability distributions of risk when calculating 
benefits. For example, one commenter recommended that the EPA analyze 
assumptions embedded in the EPA's environmental Benefits Mapping and 
Analysis Program (BenMAP) tool \37\ in its uncertainty assessment as 
well as further aligning with numerous EPA recommendations from the SAB 
and the National Academies. Some commenters recommended that the EPA 
should also quantify the effect of the major sources of uncertainty and 
variability on the risk estimates, benefit estimates, and cost 
estimates as well as transparently documenting key assumptions that 
drive uncertainty analyses.
---------------------------------------------------------------------------

    \37\ https://www.epa.gov/benmap.
---------------------------------------------------------------------------

    Some commenters opposed the EPA's proposed requirements for an 
uncertainty analysis in the BCA, stating that these proposed provisions 
are arbitrary, capricious and not appropriate. One of these commenters 
said that the EPA unjustifiably weights the burden of uncertainty 
assessment on benefits rather than costs by placing more prescriptive 
requirements on the analysis of the uncertainty of benefits, thus 
skewing the assessment of uncertainty towards benefits more than costs, 
and by depicting benefits as more uncertain than costs. Additional 
commenters opposed to the EPA's proposal argued that the proposed 
requirements add seemingly endless layers of analyses and potentially 
import substantive constraints and judgments under the guise of 
characterizing uncertainty.

[[Page 84149]]

    The SAB also made several recommendations related to the proposed 
requirements for uncertainty analysis. First, the SAB recommended that 
the preamble of the final rule discuss the broader purposes of 
uncertainty analysis beyond simple transparency. Second, the SAB 
explained that because best practices require that the analysis be 
appropriate for the policy context, uncertainty analysis should only be 
required to the extent feasible ``and appropriate.'' Third, the SAB 
advised that the discussion in the final rule be broadened to reflect 
the fact that outcomes other than the expected value may be very 
important for policies involving low-probability, high consequence 
hazards. Also, when presenting quantitative results, the SAB 
recommended that the final rule require the EPA to clearly note when 
there are unquantified benefits or costs that could be significant. 
Finally, the SAB recommended that the EPA acknowledge in the final rule 
that uncertainty analysis will not correct errors resulting from the 
inclusion of ``poor science'', which arguably has a greater impact on 
policy choices than the lack of uncertainty analysis.
    None of the public comments received have led the EPA to materially 
change its views from the proposal. The EPA disagrees with the comment 
that the requirement to conduct uncertainty analysis is arbitrary, 
capricious and not appropriate. The EPA is codifying into regulation 
procedures that are consistent with the principle of transparency 
discussed at length in OMB's Circular A-4 (OMB, 1993) and the EPA's 
Guidelines (EPA, 2010), which are the existing peer reviewed guidance 
documents implementing E.O. 12866. The EPA agrees with the principles 
emphasized in the SAB's comments on the proposed rule. The Agency has 
reviewed the discussion of uncertainty analysis below to ensure it is 
consistent with these principles and has made clarifying revisions in 
this preamble and final regulatory text where helpful. The final rule 
includes requirements pertaining to uncertainty analysis as provided 
below.
    For various reasons, including the reason that the future is 
unpredictable, the benefits and costs of future regulatory options are 
not known with certainty. The EPA is finalizing requirements for BCAs 
to identify uncertainties underlying the estimation of both benefits 
and costs and, to the extent feasible and appropriate, quantitatively 
analyze those that are most influential. Specifically, the final rule 
requires the EPA to characterize, preferably quantitatively, sources of 
uncertainty in the assessment of costs, changes in air quality, 
assessment of likely changes in health and welfare endpoints, and the 
valuation of those changes. The EPA will be required to also present 
benefit and cost estimates in ways that convey their uncertainty, 
including acknowledging unquantified benefits and costs, where 
appropriate. Because information on the range of outcomes from policy 
may be an important consideration in decision-making, the final rule 
requires EPA to also characterize the range of likely outcomes. BCAs 
will be required to include a reasoned explanation for the scope of the 
uncertainty analysis and to specify specific quantitative or 
qualitative methods chosen to analyze uncertainties. Quantitative 
uncertainty analyses may consider both statistical and model 
uncertainty where the data are sufficient to do so. Furthermore, where 
data are sufficient to do so, the rule requires BCAs to consider 
sources of uncertainty both independently and jointly. The BCA should 
also discuss the extent to which qualitatively assessed costs or 
benefits are characterized by uncertainty.
    Probabilistic uncertainty analysis involves greater effort than 
other quantitative characterizations of uncertainty but can add 
insights into the role of uncertainty in a BCA. When simpler 
quantitative analysis may not sufficiently describe uncertainty, and 
where probability distributions for relevant input assumptions are 
available and can be feasibly and credibly combined, BCAs should 
characterize how the probability distributions of the relevant input 
assumption uncertainty would impact the resulting distribution of 
benefit and cost estimates. The EPA should report probability 
distributions for each health benefit whenever feasible. In addition to 
characterizing these distributions of outcomes, it is useful to 
emphasize summary statistics or figures that can be readily understood 
and compared to achieve the broadest public understanding of the 
findings. In instances when calculating expected values is not feasible 
or appropriate due to data or other limitations, the EPA should strive 
to present a range of benefits and costs. Additional discussion of 
these best practices related to uncertainty analysis is provided in 
OMB's Circular A-4, Treatment of Uncertainty, and throughout the EPA's 
Guidelines.
    9. Principle of Transparency. Several commenters supported the 
general concept of transparency in conducting BCA, because transparency 
improves the quality of regulatory decision-making. Some commenters 
further stated that providing information on the data, models, 
assumptions, and uncertainties will increase public participation by 
improving the dialog between the EPA and stakeholders and creating a 
better-informed public.
    Several commenters objected to the transparency provisions of the 
rule with one commenter stating that it is unclear what is meant by the 
statement that the EPA's presentation of BCA results should be 
``reproducible to the extent reasonably possible.'' Commenters argued 
that the preamble offers no basis for concluding that the EPA in the 
past has not been transparent in presenting the results of their 
analysis of regulatory options. Other commenters further contended that 
the proposed requirements would obscure the basis for the EPA's 
decisions and the proposal is inappropriate to require ``consistency 
across the Clean Air Act'' given the differences in statutory 
obligations for different pollutants. Several of these commenters 
claimed that the EPA's regulatory assessments already are transparent, 
and the proposed rule would lead to confusion on the regulatory 
analysis and not increase transparency. One of these commenters further 
claimed that BCA does not increase transparency because it can distract 
from the statutory basis of regulations, since most CAA standards are 
health-based or technology-based standards, which involve a unique set 
of factors to consider.
    None of the comments received have led the EPA to materially change 
its views from the proposal. The EPA disagrees with the comment that it 
is inappropriate to impose consistent requirements related to 
transparency across the CAA given the differences in statutory 
obligation for different pollutants in various provisions of the Act. 
The requirements in this final rule aimed at providing transparency do 
not bar the Agency from complying with any requirements of the Act. The 
EPA is codifying into regulation procedures that are consistent with 
the principle of transparency discussed at length in OMB's Circular A-4 
(OMB, 1993) and the EPA's Guidelines (EPA, 2010), which are the 
existing peer reviewed guidance documents implementing E.O. 12866. For 
example, the practice of ensuring that results are reproducible is 
taken directly from OMB's Circular A-4. Therefore, after reviewing 
public comments, the EPA is finalizing the transparency requirements as 
proposed.
    This final rule provides that BCA of significant CAA regulations 
will include, at a minimum, a detailed and clear explanation of:

[[Page 84150]]

     The overall results of the BCA. The benefits, costs, and 
net benefits of each regulatory option evaluated in the BCA will be 
presented in a manner designed to be objective, comprehensive, and 
easily understood by the public.
     How the benefits and costs were estimated, including the 
assumptions made for the analysis. BCAs must include a clear 
explanation of the models, data, and assumptions used to estimate 
benefits and costs, and the evaluation and selection process for these 
analytical decisions. This explanation must also include an explanation 
of procedures used to select among input parameters for the benefit and 
cost models. Such an explanation could include methods used to quantify 
risk and to model the fate and transport of pollutants.
     A description, consistent with the best available 
scientific information, of the non-monetized and non-quantified 
benefits and costs of the action. The description must include 
available evidence on all non-monetized and non-quantified benefits and 
costs, including explanations as to why they are not being monetized or 
quantified and what the potential impact of those benefits and costs 
might be on the overall results of the BCA.
     The primary sources and potential effects of uncertainty. 
The BCA must present the results of the assessment of the sources of 
uncertainty that are likely to have a substantial effect on the 
results. Any data and models used to analyze uncertainty must be fully 
identified, and the quality of the available data must be discussed.
    Finally, to the extent permitted by law, the Agency must ensure 
that all information (including data and models) used in the 
development of the BCA is publicly available while consistent with 
protections for privacy, confidentiality, confidential business 
information (CBI), and national and homeland security. If data and 
models are proprietary, the Agency must make available, to the extent 
practicable, the underlying inputs and assumptions, equations, and 
methodologies used by EPA.
    Additional discussion of these best practices related to 
transparency is provided in OMB's Circular A-4, Transparency and 
Reproducibility of Results, and throughout the EPA's Guidelines (2010).

F. Requirements for the Presentation of BCA Results

    In the proposed rule, the EPA proposed to codify a standardized 
presentation of the results of the BCA in the preamble of significant 
regulations. Regarding these presentational requirements, many 
commenters supported providing additional details and disaggregated 
data with a focus on the specific objective of the CAA provision or 
provisions under which the rule is promulgated. These commenters 
supported the increased transparency that this presentation of BCA 
results in the preamble will provide to the public on an EPA rulemaking 
action. Some commenters were supportive of adding even more 
requirements to enhance transparency (e.g., to include a disaggregation 
of impacts on small entities).
    Other commenters opposed the proposal's presentational 
requirements, especially the requirement to provide an additional 
reporting in the preamble of the public health and welfare benefits 
that pertain to the specific objective of the CAA provision under which 
the rule is promulgated. Commenters interpreted this proposed 
requirement as barring consideration of all benefits that do not stem 
directly from the statutory objective and they argued that such 
ancillary benefits developed for a BCA are important for the EPA to 
take into consideration. Some commenters stated that distinguishing 
between benefits ``targeted by the statutory provision'' versus ``other 
welfare effects'' can be a complex, controversial, and ultimately 
fruitless endeavor, and that analysts should not assume, absent 
explicit statutory language, that any statute has the objective of 
barring consideration of important indirect effects. For example, any 
broad statutory language, like ``reasonable'' or ``appropriate,'' 
should be read broadly to authorize consideration of all important 
effects, whether direct or indirect. The SAB did not comment on this 
element of the proposed rule.
    The proposed rule also solicited comment as to whether non-domestic 
benefits and costs of regulations, when examined, should be reported 
separately from domestic benefits and costs of such regulations, 
analogous to the proposed requirement for a separate presentation of 
benefits limited to those targeted by the relevant statutory provision 
or provisions. The EPA received wide ranging comments on this issue. 
Many commenters voiced support for separately reporting, or only 
reporting, domestic benefits and costs. These commenters stated that 
separate reporting of domestic and non-domestic benefits and costs 
would allow stakeholders to better understand who would experience the 
costs and benefits before regulatory action is taken. Several 
commenters also stated that a disaggregated reporting would be 
consistent with guidance in OMB Circular A-4 that states that the ``. . 
. . analysis should focus on benefits and costs that accrue to citizens 
and residents of the United States;'' and in the case where a 
regulation is evaluated that ``is likely to have effects beyond the 
borders of the United States, these effects should be reported 
separately.'' One commenter stated that separate reporting of domestic 
impacts would assist EPA in transparently fulfilling the CAA's primary 
purpose ``to protect and enhance the quality of the Nation's air 
resources.'' Many other commenters were opposed to disaggregated 
reporting of domestic and non-domestic benefits and costs. Some stated 
that separate reporting is unnecessary and counterproductive. For 
example, one commenter stated that identification and communication of 
subcategories of benefits (such as benefits accruing outside the United 
States), where practical, is already accommodated and frequently done 
under existing procedures. Others stated that a policy of breaking out 
non-domestic benefits only ``when examined'' de-values non-domestic 
benefits and ignores the impacts that occur outside of the United 
States but that harm individuals in and outside of the United States 
directly and indirectly. Others emphasized that certain classes of 
effects cannot be meaningfully disaggregated. Some argued that a BCA 
which does not allow for benefits and costs to be calculated outside of 
the United States fails to include the ``best available science''. 
These commenters stated that EPA's request for comment on separate 
presentation of domestic benefits and costs vs. non-domestic benefits 
presumes, wrongly, that ``non-domestic'' benefits and costs can be 
accounted separately while meeting the agency's obligations to use the 
``best available science'' and reasoned decision-making. One commenter 
pointed to recent National Academies findings that the calculation of a 
domestic benefit in the case of greenhouse gas emissions reductions 
cannot be credibly done using current models, as they ignore important 
spillover effects given the global nature of climate change (National 
Academies 2017).
    None of the comments received pertaining to the proposed additional 
presentation of benefits limited to those targeted by the relevant 
statutory provision have led the EPA to materially change its views 
from the proposal. The EPA disagrees with the comment that 
distinguishing the benefits pertaining to the CAA statutory objective 
means that other benefits (or disbenefits) are not to

[[Page 84151]]

be considered. The proposed presentational requirements do not bar 
consideration of any part of the BCA. As described in Section V.D of 
this preamble, the final rule requires that the Agency consider the BCA 
in the decision-making process when permitted to do so. However, the 
EPA declines to formulate a specific test or mandate of how to consider 
the BCA or what weight the BCA, or particular elements of it, should be 
given in such a future rulemaking. The precise details of what test 
would be appropriate could differ from one CAA provision to another, 
and the EPA has not proposed or requested comment on how such tests 
would be formulated under those specific provisions.
    On the issue of separate reporting of domestic and non-domestic 
benefits and costs, the EPA agrees with commenters who stated that this 
disaggregation would enhance transparency. Separate reporting is 
consistent with both guidance in OMB's Circular A-4 and with the CAA 
which is concerned with ``enhanc[ing] the quality of the Nation's air 
resources so as to promote the public health and welfare and the 
productive capacity of its population'' (CAA 101(b)). The EPA disagrees 
with commenters who stated that a disaggregation would de-value non-
domestic benefits and ignore the impacts that occur outside of the 
United States but that harm individuals in and outside of the United 
States directly and indirectly. A separate reporting does not prohibit 
calculating or considering non-domestic benefits, but rather helps to 
allow costs and benefits to be compared in an apples-to-apples manner, 
whether domestic or not.
    Aside from separate reporting of domestic impacts, the EPA 
disagrees with commenters who stated that additional disaggregation of 
benefit and cost results in the preamble presentation are needed to 
enhance transparency. For example, CAA rules will continue to comply 
with the requirements of the Regulatory Flexibility Act so it is 
unclear why an additional requirement to discuss or present impacts to 
small entities is needed in this final rule. Therefore, the EPA is 
finalizing the presentational requirements as proposed, as described in 
detail below, along with two additional requirements. First, the final 
rule requires that any benefits and costs accruing to non-U.S. 
populations be reported separately to the extent possible in the 
summary of BCA results in the preamble. Second, the final rule requires 
that the BCA include a description in the preamble of how the Agency 
considered the results of the BCA.
    Following the principle of transparency, the EPA agrees with 
commenters that when presenting the results of a BCA, it is important 
to clearly distinguish between the social benefits attributable to the 
specific pollution reductions or other environmental quality goals that 
are targeted by the statutory provisions that give rise to the 
regulation, and other welfare effects. The disaggregation of welfare 
effects will be important to ensure that the BCA may provide, to the 
maximum extent feasible, transparency in decision-making. These other 
welfare effects could include both favorable and adverse impacts on 
societal welfare. Analogous to how a regulation's interactions with 
existing imperfections or distortions in other markets (e.g., due to 
pre-existing taxes) could lead to additional social costs, a regulation 
could ameliorate or exacerbate other pre-existing externalities. For 
example, more stringent vehicle emissions standards could affect 
upstream refinery emissions or reduce the marginal cost of driving due 
to greater fuel efficiency and could lead to an increase in vehicle 
miles traveled that affects road safety, congestion, and other 
transport-related externalities.
    Other welfare effects could also occur as a direct or indirect 
result of the compliance approaches used by regulated entities. For 
example, changes in other environmental contaminants may arise from the 
regulated sources. Likewise, the use of an abatement technology that 
reduces the emissions of hazardous air pollutants into one medium 
(e.g., air) may change the emissions of another pollutant into the same 
medium (e.g., coming out of the same smokestack) or cause changes in 
emissions of pollutants into another medium (e.g., water) by the 
regulated sources. Changes in other environmental contaminants may also 
occur as a result of market interactions induced by the regulation. For 
example, a regulation may cause consumers or firms to substitute away 
from one commodity towards another, whose increased production may be 
associated with changes in various environmental contaminants or other 
externalities.
    The welfare effects associated with these changes should be 
accounted for in a BCA to the extent feasible, as it is the total 
willingness to pay for all changes induced by a regulation that 
determines their relative importance in evaluating economic efficiency.
    Disaggregating benefits into those targeted and ancillary to the 
statutory objective of the regulation may cause the EPA to explore 
whether there may be more efficient, lawful and defensible, or 
otherwise appropriate ways of obtaining ancillary benefits, as they may 
be the primary target of an alternative regulation that may more 
efficiently address such pollutants, through a more flexible regulatory 
mechanism, better geographic focus, or other factors. This may be 
relevant when certain benefits are the result of changes in pollutants 
that the EPA regulates under a different section of the CAA or under 
another statute.
    In this final rule, the EPA is codifying into regulation several 
presentational requirements for the preamble of all future significant 
CAA regulations.
    First, in order to ensure standardized presentation of the summary 
of the BCA results consistent with E.O. 12866 in CAA rulemakings, the 
EPA is codifying into regulation the requirement to present a summary 
in the preamble of the overall BCA results, including total benefits, 
costs, and net benefits. Within this summary presentation, if any 
benefits and costs accrue to non-U.S. populations they must be reported 
separately to the extent possible.
    Second, to enhance transparency about the extent to which a rule is 
achieving its statutory objectives, the EPA is required to provide, in 
addition to a clear reporting of the overall results of the BCA, an 
additional presentation in the preamble of the public health and 
welfare benefits that pertain to the specific objective (or objectives, 
as the case may be) of the CAA provision or provisions under which the 
rule is promulgated. This second presentation would include a listing 
of the benefit categories arising from the environmental improvement 
that is targeted by the relevant statutory provision, or provisions and 
would report the monetized value to society of these benefits. If these 
benefit categories cannot be monetized, the final rule requires the EPA 
to report the quantified estimates of these benefits to the extent 
practicable and to provide a qualitative characterization if they 
cannot be quantified. Similarly, if the statute directs or allows the 
Agency to consider costs, the EPA should also provide a disaggregation 
of all relevant cost categories to the extent feasible in this section. 
This requirement would serve as a supplement to the BCA that is 
developed and presented according to best practices as outlined in 
Section V.E of this preamble. It does not replace or change any part of 
the RIA or the section of the preamble that summarizes the BCA results 
consistent with E.O. 12866.
    Finally, as described in Section V.D of this Preamble, to provide 
the public with as much information and

[[Page 84152]]

transparency as possible, the EPA will be required per the final rule 
to identify when the CAA provision or provisions under which the future 
rule is promulgated permit consideration of the BCA, and if so, the 
Agency is required to provide a description in the preamble of how the 
results of the BCA were considered. If the provision or provisions 
under which the rule is promulgated prohibit the consideration of the 
BCA, the final rule requires the Agency to identify the specific 
provision which bars such consideration. The presentational 
requirements described above should be provided in the same section of 
the preamble of future CAA significant rulemakings.

G. Additional Comment Responses

    1. Planning for Retrospective Analysis. As discussed in the ANPRM, 
a lack of data, and a lack of a regularized process for ongoing or 
retrospective review after rules have been implemented, inhibits the 
EPA's ability to gain insights about the realized costs and benefits of 
actions that may help inform how the Agency designs future regulations 
and conducts prospective BCA of future rules. Many previous 
administrations have periodically undertaken programs of retrospective 
review or issued executive orders urging or requiring agencies to 
reassess existing regulations and to eliminate, modify, or strengthen 
those regulations that have become outmoded in light of changed 
circumstances. But for the most part, retrospective review has not 
become institutionalized practice within the EPA. When they occur, 
these reviews rarely involve ex post BCA of the original EPA 
regulations. The EPA received many comment letters on the ANPRM voicing 
support for increased retrospective analysis of Agency rules or 
programs to evaluate the effectiveness of regulations, to design future 
improvements to increase efficiency, and to improve methods of ex ante 
analysis. In the proposed rule, the EPA requested comments on this 
issue, including whether EPA should include a requirement for 
conducting retrospective analysis of significant CAA rulemakings and 
how the Agency can overcome the challenges for conducting retrospective 
analysis in cases where the EPA's ability to collect information about 
the costs of compliance is limited or otherwise influenced by other 
statutes.
    The EPA received comments from a variety of stakeholders supporting 
the idea of conducting more retrospective analysis. Many commenters 
emphasized that retrospective analyses could provide useful data to 
help the EPA improve environmental outcomes while minimizing regulatory 
burdens, promulgate better regulations, and improve the analytical 
framework the Agency uses to make regulatory decisions. However, some 
questioned the need and appropriateness of a rule-based approach to 
institutionalizing the practice of retrospective analysis of existing 
regulations. Some commenters stated that the Agency should not compel 
companies to provide information necessary to conduct high quality 
retrospective analysis unless the impacted industry is interested and 
willing to participate in a retrospective review prior to beginning the 
information collection process. Others recommended that the EPA adopt 
specific guidance establishing a retrospective analytic process within 
its rulemaking procedures. One commenter specified that this guidance 
should include criteria for selecting the set of rules to be studied 
and establishing at the outset a rule design that facilitates such 
analyses; that the plan for ex post review should identify at the time 
of rulemaking the measurable outcomes to be chosen for retrospective 
analysis, the data needs, the time period for evaluation, and set out 
and justify a specific plan for data collection. Others stated that any 
potential requirements regarding retrospective analysis should be 
concretely proposed in a separate notice that fully explains the need 
for a rule-based solution to this issue and that allows a new and 
adequate opportunity for public comment. Finally, some commenters 
voiced concern that retrospective economic analyses have always been 
problematic and have many practical challenges. These commenters noted 
the difficulty in obtaining updated, accurate data for use in 
retrospective analyses and believe the EPA should focus its efforts to 
invest in high-quality, robust economic analyses using best-available 
science and following best economic practices in BCAs prepared for 
current rulemakings. Additionally, some commenters argued that 
retrospective analyses could lead to unacceptable regulatory and legal 
uncertainty especially should previously implemented regulations be 
undone and past investments based on those regulatory decisions be 
undermined or reversed.
    The EPA agrees with commenters that conducting retrospective 
analyses of an implemented regulation can provide valuable information 
that, if considered, can more fully inform public decision-making. In 
many cases, retrospective analysis provides an opportunity to 
understand whether a regulation achieved its objectives--for example, 
whether the regulation, once implemented, promoted economic efficiency 
as expected compared to a baseline without the regulation. 
Retrospective analyses may also lead to improved methods for 
prospective analysis and ultimately improvements in regulatory design. 
The Agency also agrees with those commenters that said guidance was a 
more appropriate way to better institutionalize best practices when 
planning for and conducting retrospective analysis. This approach is 
also consistent with recent recommendations the EPA received from the 
SAB during the course of their review of the forthcoming update of the 
EPA's Guidelines. In that review, the SAB recommends that the EPA 
should consider expanding discussion in the Guidelines of how 
regulatory approaches can be designed to promote effective 
retrospective analysis and, in the future, possibly devote a chapter to 
best practices for conducting such analysis.
    Given this advice, the EPA is not including a requirement in this 
final rule that retrospective analysis be undertaken for all 
significant regulations. Instead, EPA is committing to taking 
additional steps to better institutionalize the practice of conducting 
high quality retrospective review and analysis, which could be 
accomplished through the development of guidance on best practices for 
conducting retrospective analysis and how to plan for different types 
of retrospective analysis within its rulemaking procedures including 
how to address data needs. This guidance could, for example, include 
criteria for identifying rules that might be most amenable to 
retrospective analysis and direction on how to identify analytic 
requirements for such analysis at the outset when a regulation is 
promulgated. Data needs could be identified and avenues for ex post 
data collection integrated into the regulation (while also accounting 
for the cost and time needed for firms to collect such information). In 
this way, the EPA could learn from past experience and improve both 
policy designs and analytic approaches to prospective benefit and cost 
estimation. Regardless of the specific administrative procedure pursued 
for institutionalizing retrospective analysis at the EPA, it is the 
intention of the Agency to engage experts, including academics and 
practitioners, and to ultimately peer review any guidance that is 
developed.
    2. Comments pertaining to Executive Order 12898. Numerous 
commenters

[[Page 84153]]

contended that the EPA's proposed rule did not consider E.O. 12898 
(Federal Actions To Address Environmental Justice in Minority 
Populations and Low-Income Populations) and commenters stated that the 
proposal language incorrectly asserts that ``this proposed action is 
not subject to Executive Order 12898 . . . because it does not 
establish an environmental health or safety standard.'' Commenters 
further stated that air pollution disproportionately impacts minority 
communities and the proposed rule would obstruct efforts to address 
this disparity. Commenters further argued the proposed rule was unclear 
on how the proposal's BCA analysis requirements would ascribe benefits 
to communities of color that frequently bear the brunt of environmental 
risks. One of these commenters contended that, although the list of 
elements to consider in the BCA includes vulnerable and highly impacted 
communities, the proposal failed to describe how these communities are 
to be ``considered.''
    The EPA considered these comments but reiterates that this rule, as 
a procedural rule, is focused on best practices for conducting BCA 
analysis for CAA rulemaking with an aim to increase consistency and 
transparency for these BCA analyses. As such, it does not establish an 
environmental health or safety standard and is not subject to E.O. 
12898. However, the EPA asserts that with the focus on increased 
transparency and providing access to the underlying data as provided in 
this final rule's provisions, the requirements will increase the 
consistency and transparency of E.O. 12898 analyses. The additional 
information available as a result of compliance with this final rule's 
requirements will provide a better foundation for upcoming E.O. 12898 
analyses of future CAA rulemakings and will improve the understanding 
of the underlying issues highlighted by the commenters.

VI. References

    The following is a listing of the documents that are specifically 
referenced in this document. The docket includes these documents and 
other information considered by the EPA, including documents referenced 
within the documents that are included in the docket, even if a 
referenced document is not physically located in the docket. For 
assistance in locating these other documents, please consult the person 
listed under the ``For Further Information Contact'' section above.

    1. U.S. EPA (U.S. Environmental Protection Agency). Increasing 
Consistency and Transparency in Considering Costs and Benefits in 
the Rulemaking Process; Advance notice of proposed rulemaking. (83 
FR 27524, June 13, 2018).
    2. OMB (Office of Management and Budget). (1996). Economic 
Analysis of Federal Regulations Under Executive Order 12866.
    3. OMB (Office of Management and Budget). (2003). Circular A-4, 
``Regulatory Analysis.''
    4. U.S. EPA (U.S. Environmental Protection Agency). (2010). 
Guidelines for Preparing Economic Analyses.
    5. Arrow, K., M. Cropper, G. Eads, R. Hahn, L. Lave, R. Noll, P. 
Portney, M. Russell, R. Schmalensee, V. Smith, and R. Stavins. 
1996a. Benefit-Cost Analysis in Environmental, Health, and Safety 
Regulation: A Statement of Principles. Washington, DC: American 
Enterprise Institute, The Annapolis Center, and Resources for the 
Future.
    6. Arrow et al. 1996b. Is There a Role for Benefit-Cost Analysis 
in Environmental, Health, and Safety Regulation? Science 272: 221-
222.
    7. Institute of Medicine (IOM). 2011. Finding What Works in 
Health Care: Standards for Systematic Reviews. Washington, DC: The 
National Academies Press. https://www.nap.edu/catalog/13059/finding-what-works-in-health-care-standards-for-systematic-reviews.
    8. National Research Council. 2014. Review of EPA's Integrated 
Risk Information System (IRIS) Process.

VII. Statutory and Executive Order Reviews

    Additional information about these statutes and Executive Orders 
can be found at https://www.epa.gov/laws-regulations/laws-and-executive-orders.

A. Executive Order 12866: Regulatory Planning and Review and Executive 
Order 13563: Improving Regulation and Regulatory Review

    This action is a significant regulatory action that was submitted 
to the OMB for review. Any changes made in response to OMB 
recommendations have been documented in the docket. The EPA does not 
anticipate that this rulemaking will have an economic impact on 
regulated entities. This is a rule of agency procedure and practice.

B. Executive Order 13771: Reducing Regulations and Controlling 
Regulatory Costs

    This action is not subject to Executive Order 13771 because this 
final rule is a rulemaking of agency organization, procedure, or 
practice.

C. Paperwork Reduction Act (PRA)

    This action does not contain any information collection activities 
and therefore does not impose an information collection burden under 
the PRA.

D. Regulatory Flexibility Act (RFA)

    I certify that this action would not have a significant economic 
impact on a substantial number of small entities under the RFA. This 
action would not impose any requirements on small entities. This action 
would not regulate any entity outside the federal government and is a 
rule of agency procedure and practice.

E. Unfunded Mandates Reform Act (UMRA)

    This action does not contain any unfunded mandate as described in 
UMRA, 2 U.S.C. 1531-1538, and does not significantly or uniquely affect 
small governments. The action imposes no enforceable duty on any state, 
local or tribal governments or the private sector.

F. Executive Order 13132: Federalism

    This action does not have federalism implications. It would not 
have substantial direct effects on the states, on the relationship 
between the national government and the states, or on the distribution 
of power and responsibilities among the various levels of government.

G. Executive Order 13175: Consultation and Coordination With Indian 
Tribal Governments

    This action does not have tribal implications as specified in 
Executive Order 13175. Thus, Executive Order 13175 does not apply to 
this action.

H. Executive Order 13045: Protection of Children From Environmental 
Health Risks and Safety Risks

    The EPA interprets Executive Order 13045 as applying only to those 
regulatory actions that concern environmental health or safety risks 
that the EPA has reason to believe may disproportionately affect 
children, per the definition of ``covered regulatory action'' in 
section 2-202 of the Executive Order. This action is not subject to 
Executive Order 13045 because it does not concern an environmental 
health risk or safety risk.

I. Executive Order 13211: Actions Concerning Regulations That 
Significantly Affect Energy Supply, Distribution or Use

    This action is not a ``significant energy action'' because it is 
not likely to have a significant adverse effect on the supply, 
distribution or use of energy and has not otherwise been designated as 
a significant energy action by the Administrator of the Office of 
Information and Regulatory Affairs.

[[Page 84154]]

J. National Technology Transfer and Advancement Act (NTTAA)

    This rulemaking does not involve technical standards.

K. Executive Order 12898: Federal Actions To Address Environmental 
Justice in Minority Populations and Low-Income Populations

    The EPA believes that this action is not subject to Executive Order 
12898 (59 FR 7629, February 16, 1994) because it does not establish an 
environmental health or safety standard.

L. Congressional Review Act (CRA)

    This rule is exempt from the CRA because it is a rule of agency 
organization, procedure, or practice that does not substantially affect 
the rights or obligations of non-agency parties.

List of Subjects in 40 CFR Part 83

    Environmental protection, Administrative practice and procedure, 
Reporting and recordkeeping requirements.

Andrew Wheeler,
Administrator.

    For the reasons set forth in the preamble, the EPA amends title 40, 
chapter I of the Code of Federal Regulations by adding part 83 to read 
as follows:

PART 83--INCREASING CONSISTENCY AND TRANSPARENCY IN CONSIDERING 
BENEFITS AND COSTS IN CLEAN AIR ACT RULEMAKING PROCESS

Sec.
Subpart A--Analysis of Air Regulations
83.1 What definitions apply to this subpart?
83.2 How do the provisions of this subpart apply?
83.3 What requirements apply to EPA's preparations of Benefit-Cost 
Analyses (BCAs) under the Clean Air Act?
83.4 What additional requirements apply to EPA's presentation of BCA 
results for all significant rules promulgated under the Clean Air 
Act?

    Authority: 42 U.S.C. 7601(a)(1).

Subpart A--Analysis of Air Regulations


Sec.  83.1   What definitions apply to this subpart?

    Baseline means the best assessment of the way the world would 
evolve absent the regulation. It is the primary point of comparison for 
assessing the effects of the regulatory options under consideration.
    Benefit-cost analysis (BCA) means an evaluation of the social 
benefits and social costs of a policy action and other policy 
alternatives. The social benefits of a policy are measured by society's 
willingness-to-pay for the policy outcome. The social costs are 
measured by the opportunity costs of adopting the policy. BCA addresses 
the question of whether the benefits for those who gain from the action 
are sufficient to, in principle, compensate those burdened by costs 
such that everyone would be at least as well off as before the policy. 
The calculation of net benefits (benefits minus costs) answers this 
question and helps ascertain the economic efficiency of the policy. 
Where all regulation attributable benefits and costs can be quantified 
and expressed in monetary units, BCA provides decision makers with a 
clear indication of the most economically efficient alternative, that 
is, the alternative that generates the largest net benefits to society 
(ignoring distributional effects).
    Compliance cost means the private cost that a regulated entity 
incurs to comply with a regulation, such as through planning, design, 
installation, and operation of pollution abatement equipment.
    Data means the set of recorded factual material commonly accepted 
in the scientific community as necessary to validate research findings 
in which obvious errors, such as keystroke or coding errors, have been 
removed and that is capable of being analyzed by both the original 
researcher and an independent party.
    Endpoint is the specific manifestation of the documented effect 
that is to be quantified for the benefits analysis. It is a metric 
(e.g., number of hospital admissions) that acts as a surrogate for some 
aspect of a health or public welfare effect (e.g., respiratory system 
effects).
    Expected value is the probabilistically weighted outcome that 
defines a statistical mean and a measure of the central tendency of a 
set of data. For a variable with a discrete number of outcomes, the 
expected value is calculated by multiplying each of the possible 
outcomes by the likelihood that each outcome will occur and then 
summing all of those values.
    Model means a simplification of reality that is constructed to gain 
insights into select attributes of a physical, biological, economic, or 
social system. A formal representation of the behavior of system 
processes, often in mathematical or statistical terms. The basis can 
also be physical or conceptual.
    Opportunity cost means the value of the next best alternative to a 
particular activity or resource.
    Publicly available means lawfully available to the general public 
from federal, state, or local government records; the internet; widely 
distributed media; or disclosures to the general public that are 
required to be made by federal, state, or local law.
    Regulatory options means:
    (1) The proposed or finalized option, and at a minimum the 
following;
    (2) A more stringent option which contributes to the stated 
objectives of the Clean Air Act and that achieves additional benefits 
(and presumably costs more) beyond those realized by the proposed or 
finalized option; and
    (3) A less stringent option which contributes to the stated 
objectives of the Clean Air Act and that costs less (and presumably 
generates fewer benefits) than the proposed or finalized option.
    Sensitivity Analysis means an analysis that is used to assess how 
the final results or other aspects of an analysis change as input 
parameters change, particularly when only point estimates of parameters 
are available. Typically, a sensitivity analysis measures how a model's 
output changes as one of the input parameters change. Joint sensitivity 
analysis (varying more than one parameter at a time) is sometimes 
useful as well.
    Significant regulation means a proposed or final regulation issued 
pursuant to authority provided by the Clean Air Act that is determined 
to be a ``significant regulatory action'' pursuant to Section 3(f) of 
E.O. 12866 or is otherwise designated as significant by the 
Administrator.
    Social benefits, or benefits, means the sum of all positive changes 
in societal well-being experienced as a result of the regulation or 
policy action.
    Social costs, or costs, means the sum of all opportunity costs, or 
reductions in societal well-being, incurred as a result of the 
regulation or policy action. These opportunity costs consist of the 
value lost to society of all the goods and services that will not be 
produced and consumed as regulated entities reallocate resources to 
comply with the regulation.
    Systematic Review Process is the process for evaluating the 
scientific literature that includes:
    (1) Identification of the specific question to be addressed in the 
review;
    (2) Pre-specified methods used to address the question, making 
these methods and the review process transparent);
    (3) A search strategy written into the protocol that explicitly 
states the inclusion and exclusion criteria for studies; and
    (4) A description of the structured approach used to draw 
conclusions

[[Page 84155]]

considering all appropriate and available lines of evidence, including 
epidemiologic, toxicologic, and mechanistic lines of evidence.


Sec.  83.2   How do the provisions of this subpart apply?

    (a) After December 23, 2020, the Agency must prepare a benefit-cost 
analysis (BCA) for all significant proposed and final regulations, 
except that the requirement to prepare a BCA for significant final 
regulations does not apply to final regulations proposed on or before 
December 23, 2020. Except where explicitly stated otherwise, the 
provisions of this subpart do not apply to any other type of agency 
action, including individual party adjudications, enforcement 
activities, or actions taken in permit proceedings.
    (b) Except where the provision or provisions under which a 
significant regulation is promulgated prohibit the consideration of the 
BCA, the Agency must consider the BCA in promulgating the regulation.


Sec.  83.3   What requirements apply to EPA's preparations of Benefit-
Cost Analyses (BCAs) under the Clean Air Act?

    (a) A BCA prepared pursuant to this subpart must be developed by 
the Agency in accordance with best available scientific information and 
best practices from the economic, engineering, physical, and biological 
sciences according to paragraphs (a)(1) through (12) of this section.
    (1) The BCA must include the following information:
    (i) A statement of need as defined in paragraph (a)(2) of this 
section;
    (ii) An examination of regulatory options as defined in paragraph 
(a)(3) of this section; and
    (iii) To the extent feasible, an assessment of all benefits and 
costs of these regulatory options relative to the baseline scenario.
    (2) The BCA must include a statement of need that provides a clear 
description of the problem being addressed, the reasons for and 
significance of any failure of private markets or public institutions 
causing this problem, and the compelling need for federal government 
intervention in the market to correct the problem.
    (3) The BCA must include an analysis of the benefits and costs of 
regulatory options, which would contribute to the stated objectives of 
the Clean Air Act and an explanation as to why these regulatory options 
were selected. Where there is a continuum of options (such as options 
that vary in stringency), the regulatory options must include at a 
minimum (as provided in Sec.  83.1): The proposed or finalized option; 
a more stringent option that achieves additional benefits (and 
presumably costs more) beyond those realized by the proposed or 
finalized option; and a less stringent option that costs less (and 
presumably generates fewer benefits) than the proposed or finalized 
option. When a continuum of options is not applicable, the regulatory 
options can include variation of key parameters, such as different 
compliance dates, enforcement methods, standards by size or location of 
facilities, and regulatory designs. If fewer than three options are 
analyzed relative to the baseline, or if there is a continuum of 
options and the options analyzed do not include at least one more 
stringent (or otherwise more costly) and one less stringent (or 
otherwise less costly) option than the proposed or finalized option, 
then the Agency must provide an explanation of why it is not 
appropriate to analyze more options.
    (4) The BCA must include a baseline that appropriately considers 
relevant factors and relies on transparent and reasonable assumptions. 
The baseline must account for, but is not limited to, the following 
factors:
    (i) Exogenous changes in the economy that may affect benefits and 
costs (e.g., changes in demographics, economic activity, consumer 
preferences, or technology);
    (ii) Regulations promulgated by the Agency or other government 
entities; and
    (iii) The degree of compliance by regulated entities with other 
regulations.
    In rulemaking actions where the Agency determines it is appropriate 
to consider more than one baseline (e.g., one that accounts for another 
EPA regulation being developed at the same time that affects the same 
environmental condition), the BCA must include a reasoned explanation 
for the selection of the baselines used and must identify the key 
uncertainties in the forecast(s).
    (5) In preparing the BCA, the Agency must rely on the use of a 
framework that is appropriate for the characteristics of the regulation 
being evaluated and must provide an explanation for the approach 
adopted.
    (6) The Agency must consider how costs and benefits may be affected 
by consumer and producer behavior in the baseline and potential 
behavioral changes from the policy scenarios.
    (7) The BCA must include an estimation of benefits that links 
regulatory requirements to the value that individuals place on the 
change in benefit endpoints that can be meaningfully attributed to 
those requirements.
    (8) The BCA must include, to the extent supported by scientific 
literature as well as practicable in a given rulemaking:
    (i) A quantification of all benefits;
    (ii) A monetization of all the benefits that follows well-defined 
economic principles using well-established economic methods, 
appropriate data and/or studies; and
    (iii) A qualitative characterization of benefits that cannot be 
quantified or monetized.
    (9) The process of selecting and quantifying human health benefit 
endpoints in the BCA must be conducted according to paragraphs 
(a)(9)(i) through (vii) of this section:
    (i) The process of selecting human health benefit endpoints will be 
based upon scientific evidence that indicates there is:
    (A) A clear causal or likely causal relationship between pollutant 
exposure and effect, and
    (B) Sufficient data and understanding to allow the agency to 
reasonably model the anticipated change in that effect in response to 
changes in environmental quality or exposures expected as a result of 
the regulation under analysis.
    (ii) For human health endpoints, a systematic review process must 
be used to evaluate the hazard data for the purposes of determining 
which endpoints to include in a BCA and what concentration-response 
functions to use to quantify changes in these endpoints. A study's 
inclusion in the review must not depend upon that study's findings. 
More weight should be given to higher quality studies or analyses that 
have been peer reviewed.
    (iii) The studies or analyses used to quantify the concentration-
response relationships should take into account the breadth and quality 
of the available evidence regarding the nature and magnitude of the 
risk to the populations affected by the regulation. To the extent 
possible, the studies or analyses should be:
    (A) Based upon human data when available;
    (B) Specific to the exposure route, duration, and levels, with 
preference given to those studies assessing exposure similar to those 
experienced by the general population;
    (C) Employ a design or analysis that adequately addresses relevant 
sources of potential critical confounding;
    (D) Consider how exposure is measured, particularly those that 
provide measurements at the level of the individual and that provide 
actual measurements of exposure; and

[[Page 84156]]

    (E) Reliably distinguish the presence or absence (or degree of 
severity) of health outcomes.
    (iv) When utilizing multiple concentration-response functions to 
estimate impacts on a single health endpoint, the BCA must quantify 
risks in such a way that the heterogeneity in the estimated health 
impacts is clearly characterized.
    (v) The presentation of results should characterize the sensitivity 
of the choice of the concentration-response function on the magnitude 
and the uncertainty associated with estimated benefits.
    (vi) When sufficient data exist, a probability distribution of risk 
is appropriate to use when determining the expected benefits for CAA 
regulations. When it is infeasible to estimate a probability 
distribution, measures of the central tendency of risk may be used. 
Upper-bound risk estimates must not be used without also presenting 
lower bound and central tendency estimates.
    (vii) Consistent with the general systematic review process, the 
evaluation and model specification processes conducted under all 
subsections of (9) must emphasize transparency and replicability. This 
includes:
    (A) An explanation of the basis for significant judgments, 
assumptions, data, models, and inferences used or relied upon in the 
assessment and decisions regarding the selection and quantification of 
health endpoints; and
    (B) A description of the sources, extent and magnitude of 
significant uncertainties associated with the assessment.
    (10) The BCA must include an identification of uncertainties 
underlying the estimation of both benefits and costs and, to the extent 
feasible and appropriate, quantitatively analyze those that are most 
influential; and must present benefits and cost estimates in ways that 
convey their uncertainty, including acknowledging unquantified benefits 
and costs, where appropriate. The BCA must include a reasoned 
explanation for the scope and specific quantitative or qualitative 
methods chosen to analyze uncertainties. Specifically, the explanation 
must include the following:
    (i) To the extent feasible and appropriate, the BCA must apply 
quantitative methods to analyze uncertainties that have the largest 
potential effect on benefits or cost estimates and include a 
description of such methods.
    (ii) The BCA must characterize, preferably quantitatively, sources 
of uncertainty in the assessment of costs, changes in air quality, 
assessment of likely changes in health and welfare endpoints, and the 
valuation of those changes. For example, the BCA could characterize 
statistical, model or parameter uncertainty.
    (iii) Where data are sufficient to do so, the BCA must include a 
consideration of sources of uncertainty both independently and jointly.
    (iv) To the extent feasible and appropriate, the BCA must also 
include a consideration, and transparent acknowledgement of, the extent 
to which qualitatively-assessed costs or benefits are characterized by 
uncertainty.
    (v) When simpler quantitative analysis may not sufficiently 
describe uncertainty, and where probability distributions for relevant 
input assumptions are available and can be feasibly and credibly 
combined, the BCA must include a characterization of how the 
probability distributions of the relevant input assumption uncertainty 
would impact the resulting distribution of benefit and cost estimates.
    (vi) Except as provided in this paragraph, the BCA must include a 
characterization of the range of likely outcomes, including expected 
value estimates of benefits and costs as well as distributions about 
each of the estimates. In cases where estimates based on expected 
values are not feasible or appropriate, the BCA must present a range of 
benefits and costs.
    (11) The BCA must include a presentation that includes the 
following elements:
    (i) A presentation of the overall results of the BCA (benefits, 
costs, and net benefits of each regulatory option evaluated in the BCA) 
in a manner designed to be objective, comprehensive, reproducible to 
the extent reasonably possible, and easily understood by the public.
    (ii) A description of how the benefits and costs were estimated in 
the BCA, including the assumptions made for the analysis. The 
description must include the models, data, and assumptions used to 
estimate benefits and costs, and the evaluation and selection process 
for these analytical decisions. The description must also include an 
explanation of procedures used to select among input parameters to the 
benefit and cost models, and any methods used to quantify risk and to 
model fate and transport of pollutants.
    (iii) A description, consistent with the best available scientific 
information, of the non-monetized and non-quantified benefits and costs 
of the action. The description must include available evidence on non-
monetized and non-quantified benefits and costs, including explanations 
as to why they are not being monetized or quantified and discussions of 
what the potential impact of those benefits and costs might be on the 
overall results of the BCA.
    (iv) A presentation of the results of an assessment of the sources 
of uncertainty that are likely to have a substantial effect on the 
results of the BCA and present the results of this assessment. The 
presentation must identify any data and models used to analyze 
uncertainty in the BCA, and the quality of the available data shall be 
discussed.
    (v) A reasoned explanation for any departures from best practices 
in the BCA, including a discussion of the likely effect of the 
departures on the results of the BCA.
    (12) To the extent permitted by law, the Agency must ensure that 
all information (including data and models) used in the development of 
the BCA is publicly available while consistent with protections for 
privacy, confidentiality, confidential business information (CBI), and 
national and homeland security. If data and models are proprietary, the 
Agency must make available, to the extent practicable, the underlying 
inputs and assumptions used, equations, and methodologies used by EPA.
    (b) [Reserved]


Sec.  83.4   What additional requirements apply to EPA's presentation 
of BCA results for all significant regulations promulgated under the 
Clean Air Act?

    (a) The Agency must provide a summary in the preamble of each 
significant regulation of the overall BCA results, including total 
benefits, costs, and net benefits. Within this summary, if any benefits 
and costs accrue to non-U.S. populations they must be reported 
separately to the extent possible.
    (b) The Agency must provide an additional presentation in the 
preamble of each significant regulation of the public health and 
welfare benefits that pertain to the specific objective (or objectives, 
as the case may be) of the CAA provision or provisions under which the 
significant regulation is promulgated.
    (1) This presentation must list the benefit categories arising from 
the environmental improvement that is targeted by the relevant 
statutory provision and report the monetized value to society of these 
benefits.
    (2) If these benefit categories cannot be monetized, the Agency 
must report the quantified estimates of these benefits to the extent 
possible and provide a qualitative characterization if they cannot be 
quantified.

[[Page 84157]]

    (c) When the CAA provision or provisions under which the 
significant regulation is promulgated require the consideration of 
specific costs, the Agency must provide a transparent presentation of 
how those specific costs relate to total costs, to the extent possible.
    (d) When the CAA statutory provision or provisions under which the 
significant regulation is promulgated does not prohibit the 
consideration of the BCA, the Agency must provide a description in the 
preamble of how the Agency considered the BCA. If the provision or 
provisions under which the significant regulation is promulgated 
prohibit the consideration of the BCA, the Agency must identify the 
specific provision which bars such consideration.
    (e) The summary, description and presentations specified in 
paragraphs (a), (b), (c), and (d) of this section must be placed in the 
same section in the preamble of the regulation.

[FR Doc. 2020-27368 Filed 12-22-20; 8:45 am]
BILLING CODE 6560-50-P


