Half a Century of Supreme Court Clean Air Act Interpretation: Purposivism, Textualism, Dynamism, and Activism
Half a Centur y of Supreme Court Clean Air Act Interpretation: Purposivism, Textualism, Dynamism, and Activism
Brandon T. Metroka 0 1 2 3 4 5
0 David M. Driesen, Thomas M. Keck, and Brandon T. Metroka, Half a Century of Supreme Court Clean Air Act Interpretation: Purposivism , Textualism, Dynamism, and Activism, 75 Wash. & Lee L. Rev. 1781 (2018) , https://scholarlycommons.law.wlu.edu/wlulr/vol75/iss4/3
1 Part of the Environmental Law Commons, Judges Commons, Jurisprudence Commons, Legislation Commons, and the Supreme Court of the United States Commons
2 This Article is brought to you for free and open access by the Washington and Lee Law Review at Washington & Lee University School of Law Scholarly Commons. It has been accepted for inclusion in Washington and Lee Law Review by an authorized editor of Washington & Lee University School of Law Scholarly Commons. For more information , please contact , USA
3 David M. Driesen Syracuse University College of Law
4 University of the Incarnate Word
5 Thom as M. Keck Syracuse University Maxwell School of Citizenship and Public Affairs
Half a Century of Supreme Court Clean
Air Act Interpretation: Purposivism,
Textualism, Dynamism, and Activism
David M. Driesen*
Thomas M. Keck**
Brandon T. Metroka***
I. Introduction.................................................................... 1782?
* University Professor, Syracuse University
** Michael O. Sawyer Chair of Constitutional Law and Politics, Maxwell
School of Citizenship, Syracuse University.
*** Assistant Professor, Political Science, University of the Incarnate Word.
Justice Antonin Scalia?s death raised growing anxiety about
ideological judging to a fever pitch. Congressional Republicans
refused to schedule hearings for President Obama?s nominee to fill
the vacated seat, Merrick Garland, lest a liberal judicial majority
emerge.1 The Democrats retaliated with an attempted filibuster
designed to stop the confirmation of President Trump?s nominee,
Neil Gorsuch, which was thwarted by a change in Senate rules.2
Public concern about Supreme Court nominees reflects a
perception that a Justice?s political leaning affects constitutional
rulings and usually focuses on the abortion rights issue. Most
scholarship addressing the effect of Supreme Court Justices?
ideology on their decision-making likewise focuses on rulings
interpreting the Constitution.3
1. See, e.g., Nina Totenberg, 170-Plus Days and Counting: GOP Unlikely to
End Supreme Court Blockade Soon, NPR (Sept. 6, 2016)
https://www.npr.org/2016/09/06/492857860/173-days-and-counting-gop-unlikelyto-end-blockade-on-garland-nomination-soon (last visited Dec. 4, 2018) (on file
with Washington and Lee Law Review).
2. See, e.g., Matt Flegenheimer, Senate Republicans Deploy ?Nuclear
Option? to Clear Path for Gorsuch, N.Y. TIMES (Apr. 6, 2017)
https://www.nytimes.com/2017/04/06/us/politics/neil-gorsuch-supreme-courtsenate.html (last visited Dec. 4, 2018) (on file with the Washington and Lee Law
3. See, e.g., THOMAS M. KECK, THE MOST ACTIVIST SUPREME COURT IN
HISTORY (2004) (focusing on constitutional law cases); MARK J. RICHARDS, THE
POLITICS OF FREEDOM OF EXPRESSION: THE DECISIONS OF THE SUPREME COURT OF
THE UNITED STATES (2013) (focusing on free speech cases); JEFFREY A. SEGAL &
HAROLD J. SPAETH, THE SUPREME COURT AND THE ATTITUDINAL MODEL REVISITED,
316?18, 324?25 (2002) (finding strong support for the attitudinal model of judging
Many opponents of Neil Gorsuch?s nomination, however,
suggested that his political leanings influence his statutory
interpretation and that interpretation of statutes can have a huge
effect on the law.4 In particular, campaigners against the Gorsuch
nomination focused on the so-called ?freezing trucker? case, in
which Judge Gorsuch dissented from a 10th Circuit ruling
affirming an award of damages to a trucker who was fired for
disobeying an order to drive his unsafe rig or remain in an
unheated cab in subzero temperatures.5 His opponents argued, in
effect, that Gorsuch?s conclusion that the whistleblower provision
of the Surface Transportation Assistance Act did not forbid firing
the trucker reflected a bias against workers.6 Senators at his
in Fourth Amendment search-and-seizure cases); Robert A. Dahl,
Decision-Making in a Democracy: The Supreme Court as a National Policymaker,
6 J. PUB. L. 279 (1957) (outlining a theoretical and empirical argument based
solely on cases where the Court considers the constitutionality of federal
legislation for viewing the Court as a member of national partisan regimes that
generally track the preferences of those regimes); cf. Mark Latham, The
Rehnquist Court and the Pollution Control Cases: Anti-Environmental and
Pro-Business?, 10 U. PA. J. CONST. L. 133, 135 (2007) (concluding that the
Rehnquist Court was not anti-environmental in pollution control cases); Richard
L. Revesz, Environmental Regulation, Ideology, and the D.C. Circuit, 83 VA. L.
REV. 1717, 1718 (1997) (systematically studying the role of ideology in D.C. Circuit
environmental law cases, rather than in Supreme Court cases).
4. See, e.g., Nomination of the Honorable Neil M. Gorsuch to be an Associate
Justice of the Supreme Court of the United States, 115th Cong. 1?2 (2017)
[hereinafter Nomination] (statement of Sen. Sheldon Whitehouse, Member, S.
Comm. on the Judiciary) (citing numerous statutory cases in which the Supreme
Court was divided 5?4, with a Republican majority favoring business interests
over ordinary citizens? interests).
5. See TransAm Trucking, Inc. v. Admin. Review Bd., 833 F.3d 1207,
1215?17 (10th Cir. 2016) (Gorsuch, J. dissenting); Elie Mystal, The ?Frozen Truck
Driver? Case Democratic Senators are Hanging on Neil Gorsuch, ABOVE THE LAW
(Mar. 20, 2017, 4:04 PM),
http://abovethelaw.com/2017/03/the-frozen-truckdriver-case-democratic-senators-are-hanging-on-neil-gorsuch/ (last updated Mar.
21, 2017, 10:55 AM) (last visited Dec. 4, 2018) (noting that several Democratic
senators emphasized Gorsuch?s callous dissent in this ?Frozen Trucker? case in
the nomination hearing) (on file with the Washington and Lee Law Review).
6. See TransAm Trucking, Inc., 833 F.3d at 1208 (noting that this case arose
under the whistleblower provision); cf. Camille E. Peeples, Essay: Rights, Facts,
and Relevant Inquiries: Surveying Judge Neil M. Gorsuch?s Employment Law
Jurisprudence, 69 STAN. L. REV. ONLINE 193, 197 n.16 (2017) (citing TransAm as
showing that Gorsuch does not defer to agencies, especially when their actions
contradict statutory language).
confirmation hearing also relied heavily on his statutory rulings in
arguing that Judge Gorsuch favors corporations over workers
across a broad range of cases.7
The effect of judicial ideology on the Supreme Court?s
statutory interpretation, however, has received less systematic
study than questions about ideology?s role in constitutional law
cases.8 When the Justices interpret statutes, do they simply enact
their political views into law under the guise of a judicial opinion,
so that party affiliation rather than legal considerations explain
what the Court does?9 Or should we view the Supreme Court as
7. See, e.g., Nomination, supra note 4, at 1?4 (questions for the record of
Sen. Al Franken, S. Comm. on the Judiciary) (questioning Gorsuch about pro-big
business holdings in an antitrust case); Nomination, supra note 4 (questions for
the record of Sen. Patrick Leahy, S. Comm. on the Judiciary) (questioning
Gorsuch on a narrow interpretation of statutory protection against a hostile work
8. Cf. Mario Bergara, Barak Richman & Pablo T. Spiller, Modeling
Supreme Court Strategic Decision Making: The Congressional Constraint, 28
LEGIS. STUD. Q. 247, 262, 267?68 (2003) (finding that from 1947?1992, as the
liberalism of congressional ?players? increased, so too did the percentage of liberal
statutory decisions); but cf. Jeffrey A. Segal, Separation-of-Powers Games in the
Positive Theory of Congress and Courts, 91 AM. POL. SCI. REV. 28, 42?43 (1997)
(finding that from 1947?1992, Supreme Court Justices did not generally temper
sincere ideological preferences to rationally anticipate overrides by Congress). For
a mixed assessment of whether credible threats from Congress affect votes of
ideological ?outlier? Justices in statutory cases, see Thomas G. Hansford & David
F. Damore, Congressional Preferences, Perceptions of Threat, and Supreme Court
Decision Making, 28 AM. POL. Q. 490, 504?05 (2000). Earlier work also found
variance in the relationship between judicial and congressional ideological
composition. See William N. Eskridge Jr., Overriding Supreme Court Statutory
Interpretation Decisions, 101 YALE L.J. 331, 340, 346?47, 349?52 (1991) (finding
from 1967?1990, Congress was more likely to override decisions featuring an
ideologically divided Court, but from 1986?1990 Congress was more likely to
override ?identifiably conservative? decisions than liberal ones). On the role of
statutory decisions in assessing the Court?s influence on policymaking, see
Jonathan D. Casper, The Supreme Court and National Policy Making, 70 AM.
POL. SCI. REV. 50, 56?57 (1976) (arguing that accounts of the Court?s influence on
national policymaking are incomplete without attention to the Court?s power of
statutory interpretation); Dahl, supra note 3, at 293?94 (finding a limited role for
the Court as part of a ?dominant alliance,? but focusing on constitutional
9. See JEFFREY A. SEGAL & HAROLD J. SPAETH, THE SUPREME COURT AND THE
ATTITUDINAL MODEL 65 (
) (providing what is generally understood to be the
canonical statement of the relationship between judicial ideological preferences
and voting patterns); see also SEGAL & SPAETH, supra note 3. Some political
scientists have modeled empirical support for the influence of both legal and
obeying statutory language, which often proves more specific and
more recent than constitutional language? And how does the
Court?s approach square with democracy and the rule of law? Does
the Court reflect the will of the Congress that enacted the statute?
Or is the Court perhaps performing a different democratic function
by updating the statute to reflect contemporary political
preferences and/or address new problems?10 And, perhaps most
importantly, should the Court be doing things differently? If so,
how? These questions prove difficult to study, because they require
a combination of political science, an understanding of the statutes
being interpreted, and familiarity with approaches to statutory
This Article addresses these questions by studying the history
of the Supreme Court?s interpretation of the Clean Air Act (CAA),11
which now goes back almost half a century. Many scholars have
argued that the Court has shifted from an approach to statutory
interpretation that relied heavily on purposivism?the custom of
giving statutory goals weight in interpreting statutes?toward one
that relies more heavily on textualism during this period.12 At the
same time, proponents of dynamic statutory interpretation have
argued that courts, in many cases, do not so much excavate a
statute?s meaning as adapt a statute to contemporary
The CAA provides a useful prism for evaluating these accounts
descriptively and normatively. The Court has taken a keen interest
ideological factors in U.S. Supreme Court decision-making. See MICHAEL A.
BAILEY & FORREST MALTZMAN, THE CONSTRAINED COURT 47?63 (2011); LEE
EPSTEIN, WILLIAM A. LANDES & RICHARD A. POSNER, THE BEHAVIOR OF FEDERAL
JUDGES 25?64 (2013) (finding a substantial effect of ideological preferences on
decision-making?particularly at the U.S. Supreme Court?while also noting that
a variety of legal and professional factors may also affect judicial behavior); Mark
J. Richards & Herbert M. Kritzer, Jurisprudential Regimes in Supreme Court
Decision Making, 96 AM. POL. SCI. REV. 305 (2002).
10. See WILLIAM N. ESKRIDGE, JR., DYNAMIC STATUTORY INTERPRETATION
11. 42 U.S.C. ? 7401 (
12. See infra note 20 and accompanying text.
13. See ESKRIDGE, supra note 10, at 9?10 (arguing that as a statute ages the
legislature?s original intent loses relevance and courts adapt the statute to
in the CAA.14 Indeed, its interest in recent years has led it to review
some cases that generated neither circuit splits nor, arguably,
important national issues.15 Over the years, the Court has decided
20 cases interpreting the CAA, a body of case law sufficiently large
to ground a focused study of approaches to statutory
interpretation, but not so huge that it defies coherent qualitative
study. In recent years, the Court has issued a series of rulings
addressing EPA?s attempt to grapple with climate disruption, the
most important new environmental issue of the last fifteen years.16
These recent decisions provide a case study in dynamic statutory
interpretation, as the Court has struggled to adapt the CAA to an
important problem not fully anticipated when Congress amended
the CAA in 1990, let alone when Congress first enacted it in its
modern form in 1970.17 At the same time, we have some older cases
that establish a baseline which facilitates inquiry into whether the
shifts that have occurred over time reflect new politics, new
problems, or new judicial philosophies.18
Focusing on the CAA provides an opportunity to discuss the
role of political shifts in attitudes toward the statute itself in
Supreme Court adjudication. Specifically, elite opinion and
political opinion have shifted in ways that matter to CAA
14. See infra notes 15, 16, 25, 26 and accompanying text.
15. See Michigan v. EPA, 135 S. Ct. 2699, 2706?07 (2015) (reviewing a single
ruling about whether EPA must consider cost in deciding whether regulation of
an electric utility?s hazardous air pollution is appropriate and necessary, when it
has already considered cost in crafting the regulation); Util. Air Regulatory Grp.
v. EPA, 573 U.S. 302, 308, 313 (2014) (reviewing a single ruling about reducing
the applicability of general CAA program to greenhouse gases even when more
specific standards will likely prove more important).
16. See Util. Air Regulatory Grp., 573 U.S. at 333?34 (reversing EPA rule
tailoring source coverage of greenhouse gas emitters); Am. Elec. Power Co. v.
Connecticut, 564 U.S. 410, 415 (2011) (holding that the CAA displaces a federal
common law nuisance claims against carbon dioxide emitters); Massachusetts v.
EPA, 549 U.S. 497, 528 (2007) (holding that the CAA authorizes EPA regulation
of greenhouse gases).
17. See, e.g., Robert Barnes, Supreme Court: EPA Can Regulate Greenhouse
Gas Emissions, with Some Limits, WASH. POST (June 23, 2014),
https://www.washingtonpost.com/politics/supreme-court-limits-epas-ability-toregulate-greenhouse-gas-emissions/2014/06/23/c56fc194-f1b1-11e3-914c1fbd0614e2d4_story.html?utm_term=.ef8b3cb5c8fb (last visited Dec. 4, 2018) (on
file with the Washington and Lee Law Review).
18. See KECK, supra note 3, at 11 (noting that ideas affect judicial behavior).
interpretation since Congress last amended the statute in 1990.19
One can compare shifts of attitude in judicial opinions with
changes in elite and political opinion to develop a specific account
of dynamic statutory interpretation?s response to political
Part II provides a basic account of the CAA and its evolution
to provide a grounding for understanding the particular cases to
come. But the CAA is so vast and the problems that arise under it
so varied that the presentation of many statutory details must wait
until the cases implicating them become the focus of analysis. This
section also discusses the evolution of political and elite attitudes
toward the CAA in order to provide a basis for the evaluation of
dynamic statutory interpretation to follow.
Part III develops a baseline in the interpretation of the statute
from 1970 to 2004. This Part develops the concepts of purposivism
and textualism in this context and examines how they play out in
the case law of this period. The standard accounts of statutory
interpretation suggest that the Supreme Court?s approach shifted
from a heavy reliance on purpose to a much heavier reliance on
text.20 This section tests this account in the CAA context and
uncovers a surprise?judicial pursuit of goals unmoored from text
and purpose in the 1980s, which delayed the shift to textualism in
the CAA context.
Part IV argues that more recent Supreme Court CAA cases
evince a shift to dynamic statutory interpretation, but in two
different senses. The Court?s dynamic interpretation in some cases
adapts the CAA to a new problem, that of global climate disruption.
In other cases, the Court?s dynamic interpretation adapts the
statute to elite political opinion favoring what Cass Sunstein calls
the ?cost-benefit state??a state devoted to cost-benefit
balancing?or to political opinion opposing regulation.21
Part V evaluates the shift to dynamic statutory interpretation,
which one sees in the previous decade. It argues that the problem
many judges associate with purposivism?its supposed tendency
to foster judicial opinions reflecting judges? political views?has
proven more pervasive in the recent era of dynamic statutory
interpretation than in the 1970s, when the Court regularly gave
substantial weight to statutory purpose. It examines the question
of how best to carry out statutory adaptation in a way that does
not undermine democracy and the rule of law; in particular,
considering what role elite views should play in that adaptation.
II. The Clean Air Act
A. Key Statutory Features
Congress enacted the modern CAA in 1970 in response to an
environmental crisis that led to mass demonstrations and a broad
consensus favoring strict environmental protection.22 The CAA?s
stated goal is to protect public health and the environment, rather
than to achieve a balance between environmental protection and
The CAA pursues this goal by establishing a comprehensive
program of pollution control. It requires EPA to list pollutants that
endanger public health and the environment.24 For ubiquitous
pollutants, called ?criteria pollutants,? the CAA operationalizes its
goals by requiring EPA to promulgate and periodically revise
national ambient air quality standards (NAAQS), set at levels that
protect public health and the environment with an adequate
margin of safety.25 As the term ?ambient? suggests, these
22. Luke W. Cole, Foreword: A Jeremiad on Environmental Justice and the
Law, 14 STAN. ENVTL. L.J. ix, xii (1995) (discussing mass demonstrations seeking
environmental protection, such as Earth Day in 1970, which involved millions of
23. See 42 U.S.C. ? 7401(b)(1) (
). This paragraph defines the purpose as
?to protect and enhance the quality of the Nation?s air resources so as to promote
the public health and the productive capacity of its population.? Id. It expresses
the view that clean air will not only protect public health but also make people
more productive, because pollution-induced illness can increase absenteeism and
harm productivity in the workplace. Id.
24. See 42 U.S.C. ? 7408(a)(1) (
25. See Whitman v. Am. Trucking Ass?ns, 531 U.S. 457, 465 (2001)
(explaining that the CAA requires EPA to establish the NAAQS to protect public
standards apply to levels of pollution in the air surrounding us. To
reduce levels of ambient pollution, many polluters must reduce
their emissions. Accordingly, the CAA requires states to develop
state implementation plans (SIPs), containing enforceable
emissions limits for stationary sources (e.g., factories and power
plants) emitting criteria pollutants and their precursors.26 These
plans must contain standards adequate to secure attainment of the
NAAQS, and EPA reviews the plans to make sure that they do.27
Although the CAA primarily relies upon state regulation to achieve
the NAAQS, it carves out a substantial federal role as well.28 EPA
must establish New Source Performance Standards for major new
and modified sources under ? 111 of the CAA.29 The CAA also
establishes a ?technology-forcing? program of federal regulation of
?mobile sources? such as cars and trucks, which includes
regulation of fuel.30
The CAA also operationalizes its health and environmental
protection goal by requiring federal regulation of hazardous air
pollutants?non-criteria pollutants associated with very serious
health effects such as cancer and birth defects?designed to protect
public health with an ?ample margin of safety.?31 This program did
not work very well under the 1970 and 1977 CAA Amendments,
primarily because EPA listed only eight pollutants for regulation.32
So Congress itself listed 191 hazardous air pollutants for
mandatory federal regulation in the 1990 Amendments.33 The
health with an adequate margin of safety based on health effects information in
the ?criteria? documents required by the CAA).
26. See Train v. Nat. Res. Def. Council, Inc., 421 U.S. 60, 65 (1975)
(discussing the CAA requirement that states develop and submit plans to
implement and enforce the NAAQS).
27. See id. at 65?67, n.2 (characterizing the requirement that the plans
secure attainment as ?the heart of the 1970 Amendments? and citing the language
requiring EPA review).
28. 42 U.S.C. ? 7411 (
30. See 42 U.S.C. ?? 7521?7589.
31. See Nat. Res. Def. Council v. EPA, 529 F.3d 1077, 1079 (D.C. Cir. 2008)
(explaining that the CAA required regulation to protect public health from
pollutants increasing mortality or serious illness).
1990 Amendments require a round of technology-based regulation
of major sources of hazardous air pollution based on maximizing
feasible emission reductions.34 Congress did not, however, abandon
the goal of fully protecting public health, requiring a second round
of regulation designed to protect public health with an ample
margin of safety from any residual risk, thereby largely mirroring
the standard setting approach of the 1970 Amendments in a second
phase of contemplated regulation.35
The CAA introduced an enforcement innovation consistent
with its effort to vigorously pursue the goal of fully protecting
public health and the environment in spite of likely resistance by
regulated industries and government officials influenced by
them?the citizen suit. It empowered ?any person,? including
individual citizens, to sue violators of the CAA and to seek judicial
review of EPA?s implementing decisions.36
The CAA became increasingly lengthy and complex over time.
The 1970 Amendments established the statute as perhaps the
most complex and lengthy statute other than the Internal Revenue
Code.37 In 1977, Congress added complex provisions requiring a
program for Prevention of Significant Deterioration (PSD) of air
quality in areas that had attained the NAAQS, in order to prevent
them from becoming nonattainment areas and to preserve
visibility in the national parks.38 The PSD program requires new
and modified ?major sources? to use Best Available Control
Technology (BACT) to limit their emissions.39 The Congress
enacting the 1990 Amendments sought to correct EPA?s and the
states? failure to achieve the CAA?s health protection goals through
extremely detailed instructions with respect to both hazardous and
criteria air pollutants.40 It also added new titles addressing acid
rain and stratospheric ozone depletion.41 Furthermore, it adapted
the statute to reflect elite opinion favoring emissions trading, by
creating a trading program regulating sulfur dioxide in the acid
rain title and by authorizing market-based mechanisms in SIPs.42
Finally, it required that owners and operators of major sources
obtain an operating permit detailing how they would comply with
all of the CAA requirements applicable to each source in order to
improve compliance and enforcement.43 All of these amendments
created a broad federal role in securing clean air and passed with
overwhelming bipartisan support.44
B. The Evolution of Philosophy Toward Clean Air
The modern CAA reflects a particular environmental and
health protection philosophy. Senator Edmund Muskie, a major
architect of the CAA, believed that the government should make
sure that the public has clean and safe air to breathe.45 Although
the CAA contains numerous provisions that require consideration
of cost (for example, in the technology-based standard setting
provisions that one finds throughout much of the CAA), ultimately
the CAA reflects a philosophy of giving primacy to protection of the
Conservation v. EPA, 540 U.S. 461, 470?73 (2004) (describing the PSD program).
40. See, e.g., 42 U.S.C. ?? 7412, 7511 (
41. Id. ?? 7651?7651o, 7671?7671q.
42. Id. ?? 7410(a)(2)(A), 7651?7651e.
43. Util. Air Regulatory Grp., 573 U.S. at 309 (noting that Title V requires a
?comprehensive operating permit?).
44. See CLEAN AIR ACT AMENDMENTS OF 1990: S. ROLL CALL VOTE NO. 324,
136 CONG. REC. 36,138 (1990) (recording that the Amendments passed the Senate
89?10); CLEAN AIR ACT AMENDMENTS OF 1990: H.R. ROLL CALL VOTE NO. 525, 136
CONG. REC. 35,084 (1990) (recording that the Amendments passed the House
45. See Union Elec. Co. v. EPA, 427 U.S. 246, 258 (1976) (describing the
government?s responsibility as determining what is necessary to ?protect the
health of persons? (quoting 116 CONG. REC. 32901?02 (1970) (statement of Sen.
public health and the environment.46 Indeed, the Senate Report
accompanying the 1970 Amendments make it clear that owners of
polluting facilities should figure out how to operate without
harming public health and the environment or shut down.47 Thus
the CAA reflects a technology-forcing philosophy?a view that
given sufficiently strict standards, polluters would figure out how
to operate without damaging public health and accomplish feats
that appeared infeasible.48 The primacy afforded public health
protection is congruent with a rights-based view of environmental
protection that one often finds reflected in the common law,
especially in earlier cases.49
This rights-based view, however, attracted criticism, which
gained strength during the 1980s. During the 1970s, Richard
Posner helped establish law and economics as an overarching
framework to guide legal decision-making.50 He claimed that law
often aims to achieve economic efficiency and argued for the
46. Compare 42 U.S.C. ?? 7411(a)(1), 7412(d)(2), with id. ?? 7401(b)(1),
47. See Union Elec. Co., 427 U.S. at 259 (noting that the Senate committee
determined that air pollution sources must either meet the health-based
standards or ?be closed down? (quoting S. REP. NO. 91-1196, at 2?3 (1970))).
48. See id. at 258?59 (concluding that the entire Congress demanded
attainment of the NAAQS within three years ?even if attainment does not appear
49. See John P. Dwyer, The Pathology of Symbolic Legislation, 17 ECOLOGY
L.Q. 233, 233?35 (1990) (characterizing CAA ? 112?s language as creating a right
to a risk-free environment); Amy Sinden, In Defense of Absolutes: Combating the
Politics of Power in Environmental Law, 90 IOWA L. REV. 1405, 1410?12
(suggesting that National Ambient Air Quality Standards under the CAA reflect
a rights-based approach to addressing power imbalances); see also Mary Jane
Angelo, Embracing Uncertainty, Complexity and Change: An Eco-Pragmatic
Reinvention of a First-Generation Environmental Law, 33 ECOLOGY L. Q. 105, 108
(2006) (describing risk-based approaches to environmental harm as ?absolutist?);
David M. Driesen, The Ends and Means of Pollution Control: Toward a Positive
Theory of Environmental Law, 2017 UTAH L. REV. 57, 66?68 (discussing the
risk-based protective goals in the CAA and other statutes); see, e.g., Georgia v.
Tenn. Copper Co., 206 U.S. 230 (1907) (granting an injunction of air pollution
constituting a nuisance as of right at the behest of a sovereign state); Boomer v.
Atl. Cement Co., 26 N.Y.2d 219, 223 (1970) (noting that traditionally New York
did not balance equities in deciding whether to enjoin a nuisance, before deciding
to balance equities in the case before it).
50. See generally RICHARD POSNER, THE ECONOMIC ANALYSIS OF LAW 16 (2d
normative desirability of economic efficiency.51 Cost-benefit
analysis (CBA) of regulation can, in principle, weed out
economically inefficient rules.52 The law-and-economics movement
strongly influenced lawyers and other policy-making elites and
came to exercise a great deal of influence over law, including
environmental law.53 Regulated corporations championed CBA
from early on, recognizing its potential to delay and weaken
In 1982, Ronald Reagan promulgated an executive order
demanding CBA of major regulations.55 This order also ended the
independence of EPA, which Richard Nixon had helped establish
in order to implement the rights-based view of environmental law
reflected in the CAA and other major environmental statutes of the
1970s.56 The executive order did this by authorizing the Office of
Information and Regulatory Affairs (OIRA) in the Office of
Management Budget to oversee EPA implementation of the CAA
and other environmental statutes.57
51. See id. at 18 (suggesting that judges employ economic reasoning, albeit
without explicitly invoking economic concepts); Richard A. Posner, The Ethical
and Political Basis of the Efficiency Norm in Common Law Adjudication, 8
HOFSTRA L. REV. 487, 488 (1980) (arguing that economic efficiency has value
because it maximizes wealth); cf. David M. Driesen & Robin Malloy, Critiques of
Law and Economics, in OXFORD HANDBOOK OF LAW AND ECONOMICS 300, 300?03
(2017) (Francisco Parisi ed., 2017) (briefly reviewing major critiques of this claim).
52. See David M. Driesen, Distributing the Costs of Environmental, Health,
and Safety Protection: The Feasibility Principle, Cost-Benefit Analysis, and
Regulatory Reform, 32 B.C. ENVTL. AFF. L. REV. 1, 55 (
) [hereinafter Driesen,
Distributing Costs] (noting that ?allocatively efficient regulation requires that the
cost of environmental regulation equal the benefits derived from it?).
54. See THOMAS O. MCGARITY, REINVENTING RATIONALITY: THE ROLE OF
REGULATORY ANALYSIS IN THE FEDERAL BUREAUCRACY 149?50 (1991) (noting that
?regulatees? support CBA).
Exec. Order No. 12,291, 3 C.F.R. 127 (1981).
56. See Robert V. Percival, Checks Without Balance: Executive Office
Oversight of the Environmental Protection Agency, 54 L. & CONTEMPORARY
PROBLEMS 127, 131?32 (1991) (discussing Nixon?s support for an independent
EPA in preference to a proposal calculated to balance development and
57. See Exec. Order No. 12,291, 3 C.F.R. 127, ? 3 (1981) (detailing the steps
needed to obtain OMB approval of agency action).
Over the years, elite policy experts came to embrace CBA and
many of the attitudes that often came with it. Legal scholars at
elite law schools, for example, associated CBA not only with
economic efficiency, but also with enhanced rationality, better
priority setting, and improvement of ?overall well-being.?58 These
views enjoy little support from environmental law professors, who
have grave concerns about the application of CBA to
difficult-to-quantify environmental effects and who tend to support
the values embedded in the original statutes.59
Law and economics usually treats law as being about
balancing costs and benefits. Its proponents see rational
environmental regulation as a product of some sort of balancing
and identify CBA with balancing. They often treat a law aimed at
a very specific goal as likely to trigger unintended consequences
and as creating unhealthy ?tunnel vision.? Justice Stephen
Breyer?s academic work fits within this elite tradition and offers
perhaps the best articulation of concerns about tunnel vision and
priority setting derived from a view of environmental law as just
another form of economic resource allocation.60 Thus, elite opinion,
defined as the opinion of policy experts and business leaders,
shifted toward a balancing approach during the 1980s.
Political opinion, defined as the opinion of elected politicians,
shifted more slowly. Although Congress appeared initially hostile
to CBA, by the mid-1990s it enacted a law basically ratifying the
58. See Driesen, Distributing Costs, supra note 52, at 60?66 (reviewing these
59. See, e.g., FRANK ACKERMAN & LISA HEINZERLING, PRICELESS: ON KNOWING
THE PRICE OF EVERYTHING AND THE VALUE OF NOTHING (2004) (critiquing pricing
environmental benefits and claiming that some things are priceless); DOUGLAS A.
KYSAR, REGULATING FROM NOWHERE: ENVIRONMENTAL LAW AND THE SEARCH FOR
OBJECTIVITY (2010) (arguing that CBA ducks essential normative question that
environmental law has forthrightly addressed); MCGARITY, supra note 54; SIDNEY
A. SHAPIRO & ROBERT L. GLICKSMAN, RISK REGULATION AT RISK: RESTORING A
PRAGMATIC APPROACH 51?55 (2003) (suggesting that feasibility regulation avoids
serious economic disruption while declining to treat injury and death as fungible
like a dollar cost).
60. See STEPHEN BREYER, BREAKING THE VICIOUS CIRCLE: TOWARD EFFECTIVE
RISK REGULATION 10?19 (
) (discussing a problem of ?tunnel vision? and
recommending better priority setting as the cure); Lisa Heinzerling, Political
Science, 62 U. CHI. L. REV. 449, 461 (1995) (reading Breyer?s book as ?most of all,
a call for better prioritization?).
reform contained in the executive orders.61 Furthermore, while
initially a Republican reform, Democratic presidents promulgated
executive orders retaining OIRA review and CBA as part of the
regulatory structure.62 Thus, by the mid-1990s both elite and
political opinion had shifted away from supporting rights-based
environmental protection toward support of a ?balanced? approach
of some kind.63
Although politicians and other elites have moved toward
cost-benefit balancing and skepticism toward regulation, public
opinion has remained much more stable in support of the CAA?s
original philosophy. From 1994 to 2016, between 71% and 80% of
the public indicated that we should do whatever it takes to protect
64. On eleven occasions from 1994?2016, the Pew Research Center asked
respondents whether ?[t]he country should do whatever it takes to protect the
environment? or ?[t]he country has gone too far in its efforts to protect the
environment.? PEW RESEARCH CENTER, MARCH 2016 POLITICAL SURVEY (2016),
http://www.people-press.org/files/2016/03/03-31-2016-Political-topline-forrelease.pdf. Support for the former option ranged from 71?80%, support for the
latter from 15?25%. Id. When asked to weigh a tradeoff between environmental
protection and economic growth, popular majorities usually still favor
environmental protection, though by a narrower margin. Environment, GALLUP,
http://www.gallup.com/poll/1615/environment.aspx (last visited Dec. 4, 2018) (on
file with the Washington and Lee Law Review). In twenty-nine polls conducted
from 1984?2016, a majority of respondents supported environmental protection
over economic growth in eighteen and a plurality did so in another six. Id. Only
five times, all from 2009?2013, did a plurality favor economic growth. Id.
C. Global Climate Disruption
By the 1990s, a new environmental problem had become
prominent?global climate disruption. Greenhouse gas emissions,
mostly from burning fossil fuels, have warmed the earth?s average
mean surface temperature and will increase that temperature
further unless emissions are eliminated.65 Moreover, this warming
triggers very serious consequences: increasingly severe extreme
weather events, rising seas, inundation of coastal areas, killer heat
waves, drought, the spread of infectious diseases, destruction of
ecosystems, and the elimination of many species figure among its
myriad effects.66 Because this problem stems from greenhouse gas
emissions around the world, avoiding dangerous climate
disruption requires global effort.67
In 1990, Congress required study of global climate disruption68
but did not include specific provisions explicitly establishing
standards for greenhouse gas emissions.69 The 1990 Amendments
do contain specific provisions to tackle a related global problem,
stratospheric ozone depletion, requiring a phase-out of the major
ozone depleting chemicals to implement, and in some ways go
beyond, the Montreal Protocol on Ozone Depleting Substances,
which the United States ratified in 1988.70 But no international
treaty addressing global climate disruption existed in 1990, and
the science was just becoming established.
During the 1990s, however, the global community established
a treaty regime to address global climate disruption.71 At the
United Nations Conference on Environment and Development in
1992, many countries signed on to the United Nation?s Framework
Convention on Climate Change (Framework Convention), which
the United States subsequently ratified.72 This agreement provides
a set of principles and goals for addressing climate disruption, but
does not contain binding obligations to reduce greenhouse gas
emissions.73 In 1997, the Conference of the Parties to the
Framework Convention adopted the Kyoto Protocol to the
Framework Convention, which requires developed countries to
reduce greenhouse gas emissions.74 The United States, however,
never ratified the Kyoto Protocol, and President George W. Bush
expressly repudiated it in 2001.75 This failure to ratify the Kyoto
Protocol suggests a significant change in political climate, as the
United States throughout the 1970s and 1980s regularly led
international efforts to combat serious international
In 2008, however, it appeared that the United States would at
least follow other nations? lead and address global climate
disruption. In that year?s presidential election, both Republican
candidate John McCain and his rival Barack Obama favored action
on global climate disruption.77 And, consistent with this bipartisan
consensus, the House passed comprehensive legislation mandating
reductions in greenhouse gas emissions primarily through an
emissions trading program shortly after President Obama?s
inauguration.78 This legislation, however, failed to pass the
Since then, the political landscape has shifted in ways that
make the political climate today radically different from the
political climate that existed when Congress created and amended
the CAA and even from the climate prevailing in 2008. Opposition
to action on global climate disruption has become an article of faith
among Republican politicians and presidential candidates.80 Thus,
political opinion has recently departed markedly from educated
elite opinion, which tends to favor some action on global climate
disruption in light of the strong evidence of significant harm, even
though it favors basing that action on CBA.
This shift in political opinion made it impossible to pass new
legislation to address climate disruption even after 2008.
Accordingly, President Obama addressed it primarily under the
CAA.81 President Trump, however, declared climate disruption a
77. See Michael B. Gerrard, McCain vs. Obama on Environment, Energy,
and Resources, 23 NAT. RES. & ENV?T, Fall 2008, at 3, 4. (noting that both John
McCain and Barack Obama supported cap-and-trade legislation to reduce
greenhouse gas emissions).
78. See MARK HOLT & GENE WHITNEY, CONG. RESEARCH SERV., R40643,
GREENHOUSE GAS LEGISLATION: SUMMARY AND ANALYSIS OF H.R. 2454 AS REPORTED
BY THE HOUSE COMMITTEE ON ENERGY AND COMMERCE (2009) (analyzing the bill
that passed in the House).
79. See Kassie Siegel et al., Strong Law, Timid Implementation: How the
EPA Can Apply the Full Force of the Clean Air Act to Address the Climate Crisis,
30 UCLA J. ENVTL. L. & POL?Y 185, 186 (
) (mentioning the ?defeat of
economy-wide climate legislation in the 111th Congress?).
80. See id. at 186 (describing the Congress elected in 2010 as ?openly hostile
to any form of greenhouse regulation?); see also Art Swift, Americans Again Pick
Environment Over Economic Growth, GALLUP (Mar. 20, 2014),
http://www.gallup.com/poll/168017/americans-again-pick-environmenteconomic-growth.aspx (last visited Dec. 4, 2018) (indicating that about two-thirds
of Democrats but only one-third of Republicans would give environmental
protection priority over economic growth in 2014 and noting that this is the
largest partisan gulf since 1997) (on file with the Washington and Lee Law
81. See Uma Outka, The Obama Administration?s Clean Air Act Legacy and
hoax and seeks to unwind Obama Administration initiatives to
address the issue.82
D. Summary of Background
In sum, the CAA reflects a rights-based view of environmental
law that enjoyed broad bipartisan and public support for at least
twenty years. Since its enactment, however, this view has become
less popular among both elites and politicians.83 In addition, EPA
and thus the courts have recently applied the CAA to a new
problem not fully anticipated in 1970 or even 1990?the problem
of global climate disruption.84 These background realities invite
consideration of whether the Court has dynamically interpreted
the statute either to adapt to a new problem or to adapt to new
attitudes among elites or politicians. That question requires some
consideration of broader judicial trends in statutory
interpretation, which also may influence cases? outcomes.
III. Purpose, Text, and Judge-Made Law: The Court?s
Interpretation of the Clean Air Act from 1970 to 2004
A. Purpose and Text in Statutory Interpretation
A venerable canon of statutory construction urges courts to
construe statutes in a way that effectuates their purposes.85
the UNFCC, 48 CASE W. RES. J. INT?L L. 109, 117?18 (2016) (explaining that
several of President Obama?s environmental goals ?have proven integrally
connected by the Administration?s regulatory agenda under the CAA?).
82. See, e.g., Coral Davenport, Donald Trump Could Put Climate Change on
Court for Danger Zone, N.Y. TIMES (Nov. 10, 2016),
https://www.nytimes.com/2016/11/11/us/politics/donald-trump-climatechange.html (last visited Dec. 4, 2018) (on file with the Washington and Lee Law
83. Siegel et al., supra note 79.
84. Supra note 68 and accompanying text.
85. See 3 NORMAN J. SINGER & SHAMBIE SINGER, SUTHERLAND STATUTES AND
STATUTORY CONSTRUCTION ? 58:6 (7th ed. 2011) (referring to construction to
realize a statute?s purpose as ?ancient wisdom?).
will . . . contribute significantly? to nonattainment of the NAAQS
in neighboring states.332 This provision strongly suggests that EPA
should require abatement from pollution sources sufficient to avoid
significantly interfering with NAAQS achievement, a task
requiring air quality modeling to make sure that the amount of
reductions suffice to attain this goal.333 This focus on
environmental effects might imply no consideration of cost or
technology in setting targets, a conclusion in keeping with Union
Electric?s understanding that the CAA subordinates cost
considerations to the goal of achieving the NAAQS.334
Such a view of the statute, while textually supported, does not
comport with elite intellectual trends. Accordingly, EPA
considered the marginal cost effectiveness of pollution reductions
and based its plan for abating these emissions on conclusions about
sensible cost per ton of pollution reduction numbers.335
Furthermore, it used a mechanism favored by law and
economics?emissions trading?to realize these reductions.336 But
emissions trading works for pollutants where location does not
matter, and location does matter to the question of whether a set
of reductions will avoid interfering with neighboring states?
When the issue of whether EPA acted properly in considering
cost in setting the targets reached the Supreme Court in EPA v.
EME Homer City Generation (Homer City)338 the Court affirmed
EPA?s Clean Air Transport Rule in a 6?2 decision.339 The Homer
City majority clearly treated EPA?s core methodology of
considering costs in establishing state abatement obligations as an
appropriate adaptation of the statute to the problems of addressing
interstate air pollution.340 It also approved of employing cost in
part based on contemporary elite thinking, finding its use in
allocating emission reduction obligations among the states
?efficient.?341 The majority also linked this adaptation to technical
necessity, as it found the alternative suggested by the D.C. Circuit,
reducing emissions in proportion to each source?s contribution to
each state?s attainment problem, unworkable.342 Justice Scalia?s
dissent, however, provides a textually strong argument that the
statute requires an allocation based on air quality alone, not
cost.343 And the dissent did not agree that workability
considerations required upholding EPA?s decision about the proper
methodology for allocating reduction obligations to states.344
On a second issue?whether EPA could promulgate a federal
plan implementing its allocations?the dissent and majority
flipped their approaches to statutory interpretation. The majority
closely followed text that clearly showed that EPA may promulgate
a federal implementation plan after disapproving state plans that
fail to satisfy the Good Neighbor obligation.345 The dissent,
however, adapted the statute creatively to solve a perceived
338. 572 U.S. 489 (2014).
339. See id. at 524.
340. See id. (finding that the D.C. Circuit requirement to allocate emission
reductions in proportion to each state?s contribution to neighboring states? air
quality problems unworkable).
341. See id. at 519?20.
342. See id. at 515?16 (finding that the ?proportionality approach could
scarcely be satisfied in practice?).
343. See id. at 527?29 (Scalia, J., dissenting) (pointing out that the statutory
reference to significant contribution to air pollution has no logical relationship to
344. See id. at 530?35 (Scalia, J., dissenting) (rejecting the majority
arguments that proportional reductions are impossible and result in extensive
345. See id. at 507?08 (noting that the statute authorizes a FIP anytime
within two years of EPA disapproval of a SIP).
problem. Justice Scalia, the apostle of textualism, abandoned text
on the ground that states cannot be expected to comply with the
Good Neighbor Provision without effective EPA guidance and
therefore needed another opportunity to do so now that EPA had
issued guidance.346 The Scalia dissent buttresses this argument
with a claim that the CAA?s cooperative federalism structure
requires giving states a second chance now that EPA has provided
Thus, the Court ratified EPA?s textually problematic
adaptation of goal-setting to elite views, but simply followed text
to ratify its implementation procedures. The dissent favored
honoring textual constraints on EPA?s goal-setting while creatively
adapting the statute to reign in EPA?s textually based approach to
346. See id. at 538?39 (Scalia, J., dissenting) (accusing the majority of forcing
states to guess at what their responsibilities might be with respect to the Good
Neighbor Provision). The relevant text requires SIPs to include
?provisions . . . prohibiting . . . any source . . . within the State from emitting any
air pollutant in amounts which will . . . contribute significantly to nonattainment
in . . . any other State? within three years of EPA?s promulgation of a NAAQS. 42
U.S.C. ? 7410(a)(1) (
); id. ? 7410(2)(D)(i). If a state fails to submit an
adequate SIP, the relevant text demands that EPA ?promulgate a [f]ederal
implementation plan at any time within [two] years after? disapproving a SIP. Id.
? 7410(c)(1). Because EPA had disapproved the relevant SIPs for noncompliance
with the Good Neighbor Provision, it had clear statutory authority to issue a
federal implementation plan (FIP). EPA v. EME Homer City Generation, L.P.,
572 U.S. 489, 538?39 (2014) (Scalia, J., dissenting). Justice Scalia argued that the
statutory principle that states have ?the primary responsibility? for air pollution
control requires that EPA issue guidance before disapproving a SIP or
promulgating a FIP but admits that the statute does not require any EPA
guidance at all. See id. at 537 (Scalia, J., dissenting). Scalia argues that EPA has
discretion to postpone the SIP submission deadline until after it issues guidance
on how to comply with the Good Neighbor Provision, but points to no statutory
language authorizing such a postponement. See id. at 542 (Scalia, J., dissenting).
He also argues that EPA need not promulgate a FIP before two years elapse, and
that doing so is an abuse of discretion in this case. See id. at 542?43 (Scalia, J.,
dissenting). In short, Scalia?s argument was based on his policy preference for
guidance and runs counter to very clear text respecting SIP content and the
deadline for SIP submission.
347. See EPA v. EME Homer City Generation, 572 U.S. 489, 537 (2014)
(Scalia, J., dissenting) (finding that the majority?s reading abandons cooperative
federalism in favor of ?centralized federal control?).
A clearer example of judicial adaptation of a statute to elite
views comes from Michigan v. EPA,348 which reverses an EPA
decision finding regulation of hazardous air pollutants from power
plants ?appropriate and necessary.?349 EPA found such regulation
appropriate and necessary because of evidence that the emissions
remaining after application of the acid rain program pose a
significant hazard to public health.350 In making this
determination, EPA did not consider cost, deferring such
consideration to the first round of regulation, when it considered
cost in promulgating technology-based standards to regulate
hazardous air pollutants from power plants.351
The statutory phrase ?appropriate and necessary? on its face
seems extraordinarily open-ended.352 One might take the view that
environmental regulation is appropriate and necessary when a
pollution problem seriously impacts public health or the
environment.353 Or one might take the view that one should also
consider the costs of regulation in deciding about the
appropriateness of regulation, a view more in keeping with elite
thinking about the importance of CBA. On its face, this sort of
open-ended language offers a compelling context for the
application of deference to EPA?s decision, as it seems to require a
political decision of the sort not appropriate to the judiciary.354
The Court, however, in a 5?4 opinion split along the Court?s
liberal?conservative fault line, based its decision on the majority?s
preference for elite views and refused to defer to EPA?s preferred
reading of a statutory provision that seems devoid of specific
content, finding EPA?s cost-blind interpretation unreasonable
under Chevron.355 The Court notes early in its opinion that EPA
?refused to consider whether the costs of its decision outweighed
the benefits.?356 In keeping with the ?comprehensive rationality?
traditionally associated with CBA, the Court read the term
?appropriate? as an ?all-encompassing term.?357 The
appropriateness inquiry requires consideration of cost, because it
would not be appropriate to impose ?billions of dollars? of cost ?for
a few dollars in health or environmental benefits.?358 Thus, the
Court assumes that environmental and health benefits can be
reasonably described in dollar terms and adopts the views of Cass
Sunstein, who has frequently argued that a major benefit of CBA
involves avoidance of costs grossly disproportionate to benefits.359
Echoing Justice Breyer?s academic contributions to the literature
advocating a ?cost-benefit state,? the Court notes that spending too
much on one problem may leave fewer resources to devote to more
serious problems, a frequently asserted (and cogently contested)
assumption of those viewing regulation as just another form of
resource allocation.360 Although the Court stopped short of
deference should have applied here); see generally KECK, supra note 3, at 186
(noting that ?if judicial restraint means anything . . . it must mean that the
unelected judiciary? should play a ?smaller role in settling divisive conflicts?).
355. See id. at 2711 (finding it unworkable to read the statute ?to mean that
cost is irrelevant to the initial decision to regulate power plants?).
356. Id. at 2706.
357. Id. at 2707; cf. MCGARITY, supra note 54, at 5, 10?11 (explaining the
concept of comprehensive rationality and its relationship to CBA).
358. Michigan v. EPA, 135 S. Ct. 2699, 2707 (2015).
359. See Timur Kuran & Cass R. Sunstein, Availability Cascades and Risk
Regulation, 51 STAN. L. REV. 683, 758 (1999) (arguing for a judicial presumption
that regulation is unreasonable if it generates costs ?grossly disproportionate to
360. See Michigan, 135 S. Ct. at 2707?08 (?Consideration of cost . . . reflects
the reality that ?too much wasteful expenditure devoted to one problem may well
mean considerably fewer resources available to deal effectively with other
(perhaps more serious) problems.?? (quoting Entergy Corp. v. Riverkeeper, Inc.,
556 U.S. 208, 233 (2009) (Breyer, J., concurring in part and dissenting in part)));
mandating consideration of formal CBA, it justified its decision to
require consideration of cost in evaluating a statutory trigger with
many of the same arguments that lie behind the law and economics
movement?s support for CBA, including the views of elite
moderates, such as Breyer and Sunstein.361
The dissenting Justices (including Justice Breyer) would defer
to EPA?s decision, largely because EPA considered cost and indeed
conducted a CBA when it developed the regulation that the
appropriate and necessary finding led to.362 The dissenters,
however, echoed the elite views found in the majority opinion as to
general regulatory philosophy and suggested that those views
would have proven dispositive if EPA had not conformed its
subsequent actions to those views. It opined, contrary to the
general philosophy of the CAA, that power plant regulation would
be ?unreasonable? if EPA did not consider cost at all in regulating
power plants.363 Echoing the majority?s articulation of elite
orthodoxy, the dissent supported a presumption favoring the new
elite consensus?requiring EPA to consider costs ?[u]nless
Congress provides otherwise.?364 Thus, the dissent articulates a
position strikingly at odds with the Court?s cases following the
CAA?s original philosophy, as both Union Electric and American
Trucking hold that Congress does not intend to require cost
consideration unless it specifically mentions cost.365 The dissent
also echoes the majority?s articulation of concerns about costs
cf. David M. Driesen, Getting Our Priorities Straight: One Strand of the
Regulatory Reform Debate, 31 ENVTL. L. REP. 10003, 10015?16 (2001) (noting that
Breyer and other regulatory reformers do not explain how relaxing ?even
ridiculously stringent government regulation would increase the funding of
childhood vaccination programs, or . . . other . . . preferred public health
programs?); Thomas O. McGarity, A Cost-Benefit State, 50 ADMIN. L. REV. 7, 34
(1998) (explaining that no vehicle exists for channeling cost savings from relaxed
regulation to ?deserving social programs?).
361. See Michigan, 135 S. Ct. at 2711 (declining to mandate formal CBA).
362. See id. at 2717 (Kagan, J., dissenting) (finding EPA?s action in not
considering cost at the first stage reasonable because it considered cost
subsequently, including a CBA).
363. See id. at 2717.
365. Whitman v. Am. Trucking Ass?n, 531 U.S. 457, 467 (2001); Union Elec.
Co. v. EPA, 427 U.S. 246, 271 n.5 (1976).
grossly disproportionate to benefits and mandates to clean up one
form of pollution wasting resources that might be better spent on
more important problems.366
The elite view disfavoring the CAA?s philosophy of fully
protecting public health regardless of apparent cost or feasibility
so thoroughly pervades the thinking of the Supreme Court, and
perhaps even EPA in this case, that none of the Justices even
considered the incongruity of declining to regulate based on cost
when the very program before the Court ultimately aimed to
provide a basic level of safety regardless of cost.367 Although the
Court extensively discussed the relationship between the
?appropriate and necessary? trigger and the immediately
subsequent cost-sensitive, technology-based rulemaking, it did not
discuss the CAA?s requirement to eliminate residual risk through
a standard protecting public health with an ample margin of safety
if the technology-based standard proves insufficiently
protective.368 The Court therefore failed to note that relying on cost
considerations in declining to regulate hazardous air pollutants
found to harm public health and the environment contradicts the
CAA?s overall philosophy, which is plainly reflected in the program
Thus, recent CAA cases evince a marked turn away from
literal statutory interpretation and toward dynamic adaptation of
the CAA, either adapting it to elite views or to the problem of
366. See Michigan, 135 S. Ct. at 2717 (Kagan, J., dissenting) (expressing
concern about imposing ?massive costs far in excess? of benefits and ?wasteful
expenditure[s]? meaning ?considerably fewer resources available to deal? with
more serious problems).
367. See 42 U.S.C. ? 7412(n)(1)(A) (
); Michigan, 135 S. Ct. at 2709
(responding to an EPA argument that it need not consider cost in making an
appropriate and necessary finding because it ?can consider cost when deciding
how much to regulate,? thereby suggesting that EPA did not make an argument
based on the incompatibility of cost consideration and Section 112?s health
368. See Michigan, 135 S. Ct. at 2705 (discussing EPA?s technology-based
rulemaking under 42 U.S.C. ? 7412(d)).
369. See 42 U.S.C. ? 7412(n)(1)(A) (stating that the purpose of the studies
under this section of the statute is to find ?hazards to public health reasonably
anticipated to occur?); see also 116 CONG. REC. 19,204 (1970) (discussing and
debating the importance of developing strategies and technology to achieve clean
air because of pollution?s threat to people?s health and wellbeing).
climate disruption. The Court during this period did not consider
statutory purpose, even in cases replete with statutory
ambiguity.370 In its place, we find the Court advancing the
regulatory philosophy of the majority of Justices. And as Tables 1
and 2 indicate, in every case except Homer City (a 6?2 ruling) and
AEP (a unanimous decision), the Roberts Court divided 5?4 along
ideological lines on the question of whether to uphold EPA?s
C. Summary Description of the Evolution of the Supreme Court?s
Approach to Interpreting the Clean Air Act
Thus, the Court?s approach to interpreting the CAA has
evolved over time. The first decade?s jurisprudence usually focused
on text and purpose, but the Court misinterpreted the CAA, in the
eyes of Congress, when it invoked quasi-constitutional clear
statement rules to resolve cases or chose a fairly natural textual
reading completely at odds with the statutory purpose.372 The
Rehnquist Court abandoned purpose in the 1980s and began
treating statutory interpretation under the CAA as an occasion to
elaborate judge-made common law.373 But this emphasis on
judge-made law took place in the context of issues regarding
attorney fees, which judges may feel competent to resolve
according to their own policy views.374 When abandoning purpose
in the context of a more technical regulatory issue, the Court
deferred to EPA in Chevron.375 Beginning in the 1990s, the Court
mostly based its decisions on statutory text and turned to purpose
only in one especially difficult statutory case (ADEC).376
Throughout this latter period (
), the Court almost
always decided cases unanimously or with a large majority, and
the one split decision (again ADEC) involved a genuinely difficult
The last decade, however, has featured a turn toward dynamic
statutory interpretation. Some dynamic statutory interpretation
cases adapted the CAA to the climate disruption problem, but
others adapted the CAA to trends in elite thinking.378
This turn toward dynamic interpretation coincided with a
trend toward ideological decision-making. As Table 2 indicates,
four of the five decisions issued in the last decade were divided on
ideological lines.379 Moreover, the Justices? tendency to adapt the
statute to their own preferences through split decisions dominates
cases where the text is clear (e.g. Massachusetts v. EPA, where the
majority followed it, and UARG, where nobody followed it) and
where the text says very little (e.g. Michigan v. EPA).380
The recent division and emphasis on judicial policymaking
reminds one of the Rehnquist Court?s common law experiment.381
But the modern cases differ from the 1980s common law cases in
at least one respect; the Rehnquist Court made policy decisions in
an area where one might expect judicial competence, namely the
attorney fee policies for federal litigation.382 The more recent cases
feature judicial decisions about core regulatory questions, where
376. See Alaska Dep?t of Envtl. Conservation v. EPA, 540 U.S. 461, 470 (2004)
(noting that the purpose of the relevant statutory provisions is to protect public
health and welfare).
377. Supra Table 1.
378. See, e.g., Michigan, 135 S. Ct. at 2707 (interpreting ?appropriate and
necessary? to require some consideration of costs); Massachusetts v. EPA, 549
U.S. 497, 534 (2007) (discussing global warming in relation to the regulation of
379. Supra Table 2.
380. See generally Michigan v. EPA, 135 S. Ct. 2699 (2015); Util. Air
Regulatory Grp. v. EPA, 573 U.S. 302 (2014); Massachusetts v. EPA, 549 U.S. 497
381. See supra note 373 and accompanying text.
382. See supra note 374 and accompanying text.
one can expect relatively little judicial expertise and substantial
congressional and EPA engagement.383
General common law principles do play some role in the recent
dynamic cases, but these general principles do not constitute
traditional background principles of substantive law, but rather
recently minted interpretive canons, as Lisa Heinzerling has
shown.384 In UARG, the Court applied a presumption against
?unheralded? economically and politically important exertions of
regulatory power under ?long extant statutes.?385 This constitutes
an important canon from the standpoint of dynamic statutory
interpretation. It suggests a judicial presumption in favor of
adapting a statute to important new problems by not allowing it to
reach the problem even if its text seems to apply.386 It comports
with the recent anti-regulatory turn in the Republican Party.387
But as Massachusetts v. EPA demonstrates, the Court has not
consistently applied this anti-regulation presumption.388 In
Michigan v. EPA, the Court created a presumption in favor of
considering cost, thus putting itself in the camp of elite opinion at
383. Compare, e.g., Michigan, 135 S. Ct. at 2707, with Pennsylvania v. Del.
Valley Citizens? Council, 478 U.S. 546, 562?68 (1986).
384. See Lisa Heinzerling, The Power Canons, 58 WM. & MARY L. REV. 1933,
1948?54 (2017) (discussing the newness of the ?UARG canon?); cf. Manning, supra
note 20, at 113 (suggesting that only ?firmly established? background principles
legitimately allow textualists to fill in textual gaps).
385. Util. Air Regulatory Grp. v. EPA, 573 U.S. 302, 323?24 (2014). This
canon has an antecedent. See FDA v. Brown & Williamson Tobacco Corp., 529
U.S. 120, 159?60 (2000) (applying the canon to reject FDA regulation of cigarettes
as nicotine delivery devices in light of evidence that Congress has rejected
cigarette regulation other than labeling); cf. Jody Freeman, Why I Worry About
UARG, 39 HARV. ENVTL. L. REV. 9, 16 (2015) (noting that after Massachusetts v.
EPA, 549 U.S. 497 (2007), the major question canon appeared to be dead or in
386. See Heinzerling, supra note 384, at 1946 (defining the canon as
promising skepticism toward agency interpretations of ?long extant statutes? as
authorizing regulation in an area of ?vast economic and political significance?)
(internal quotations omitted).
387. See Buzbee, supra note 311, at 75 (characterizing this canon as
388. See Heinzerling, supra note 384, at 1953?54 (discussing the
inconsistency between UARG, 573 U.S. 302 (2014), and Massachusetts v. EPA,
549 U.S. 487 (2007)).
the expense of the philosophy behind the CAA.389 This canon
constitutes a general principle favoring adaptation of the CAA to
the elite opinion emanating from the law and economics
In spite of the Court?s unity on the need to conform the statute
to the prevailing principles of the law and economics movement,
the Court remains divided on how to resolve cases. The only time
in the last decade where the Court managed a unanimous CAA
ruling?AEP?involves following precedent slighting text in favor
of quasi-constitutional judicial lawmaking.391 Thus, the cases
suggest a willingness to adhere to judicial policy, but no consensus
on whether to follow the overall policy behind the CAA.
V. Lessons for Dynamic Statutory Interpretation
In this Part, we normatively critique the Court?s dynamic
interpretation with a goal of advancing the art of dynamic
statutory interpretation. We focus on how and when judges should
adapt statutes to new problems with consideration of the propriety
of judges updating statutes to reflect changes in elite or political
A. Purpose as Dynamic Archaeology
William Eskridge, the leading proponent of dynamic statutory
interpretation, does not claim to have answered the key normative
question that the existence of dynamic statutory interpretation
389. See Michigan v. EPA, 135 S. Ct. 2699, 2707 (2015) (finding that cost
should be a factor in deciding whether to regulate power plans and failing to
discuss health and safety considerations? role in the regulatory process).
390. See generally ESKRIDGE, supra note 10, at 297, 301 (discussing the idea
of statutory canons as an expression of judicial ideology and some canons as
responding to the scarcity concerns at the heart of law and economics).
391. See Am. Elec. Power Co. v. Connecticut, 564 U.S. 410, 423 (2011)
(following Milwaukee II, 451 U.S. 304 (1981), in holding that the CAA supplants
the need for federal common law claims related to regulating emissions).
raises: how and when should judges dynamically construe
Eskridge presents dynamic interpretation as an alternative to
both intentionalism and purposivism, which he characterizes as
archaeological approaches.393 He concludes that purposivism, like
intentionalism, ?can establish no connection with majority-based
preferences in the hard cases.?394 At the same time, he describes
implementing the legislature?s ??general intent? about the goals of
a statute? as a form of dynamic interpretation.395 And he favors
?bending? the statute?s literal terms in order to achieve this
general intent in adapting the statute to changing
circumstances.396 Eskridge?s position suggests the possibility that
judges can adapt a statute to new circumstances by using the
values embedded in the statute?s goals to guide the adaptation and
that doing so conforms to both the precepts of a ?faithful agent?
theory of statutory interpretation and the exigencies of dynamic
interpretation.397 This means that archaeological and dynamic
statutory approaches overlap when a court uses a very general
congressional intent to decide how to adapt a statute to a new
problem.398 We might call this ?dynamic purposivism.?
392. See ESKRIDGE, supra note 10, at 107?08 (identifying the normative issues
raised by an understanding of dynamic statutory interpretation as whether it is
justifiable and if so when and how it should be used).
393. See id. at 13, 25?34 (characterizing discovery of congressional intent
through text and purpose as ?archaeological? and explaining why purposivism
fails as a ?foundational theory of statutory interpretation?).
394. Id. at 26.
395. See id. at 121 (pointing out that serving a statute?s goals requires
dynamic interpretation as circumstances change in order to conform to ?general
intent?); see also Jonathan R. Macey, Promoting Public-Regarding Legislation
Through Statutory Interpretation: An Interest Group Model, 86 COLUM. L. REV.
223, 249?52 (1986) (advocating adherence to statutory purpose as an antidote to
ESKRIDGE, supra note 10, at 121.
397. See supra note 396 and accompanying text; Manning, supra note 20, at
9 (explaining that purposivism allows ?federal judges to fulfill their presumed
duty as Congress?s faithful agents?).
398. See ESKRIDGE, supra note 10, at 26 (characterizing purposivism as
?attractive? because it both ?allows a statute to evolve? and ties ?interpretation to
original legislative expectations?).
One of us has argued previously that courts should construe
statutes to effectuate their stated purposes, at least where the
statute is ambiguous, primarily on the grounds that stated
purposes are likely to reflect goals enjoying broad public
support.399 This argument suggests that dynamic statutory
interpretation should follow the general intent of the enacting
Congress when possible.
Eskridge?s suggestion that judges might appropriately bend
text to dynamically interpret statutes in keeping with their
purposes would support Justice Stevens? dissent in Adamo.400 The
term ?emission standard? might most naturally indicate a
numerical limit on the amount of air pollution, but perhaps the
Court should have creatively read the term to include a work
practice requirement that has the effect of limiting air pollution in
order to allow the CAA to meet its goals.401 It is linguistically
plausible to read the term ?emission standard? as including any
requirement that has the effect of reducing emissions, and such a
reading would avoid the problem of hindering the enforcement of
a properly enacted work practice standard.402
Eskridge?s claim that neither purposivism nor intentionalism
connects results to majority preferences in hard cases suggests an
acknowledgment that they can do so in easy cases.403 From the
standpoint of purposivism and intentionalism, Massachusetts v.
EPA should have been an easy case.
399. See Driesen, supra note 86, at 98; see also Macey, supra note 395, at 250
(claiming that a statute?s stated purpose will almost always be
400. Adamo Wrecking Co. v. United States, 434 U.S. 275, 293?307 (1978)
(Stevens, J., dissenting); cf. ESKRIDGE, supra note 10, at 50?55 (discussing
dynamic interpretation of a prohibition on immigration of people with a ?mental
defect? to permit immigration of homosexuals in spite of original expectation that
the term would exclude them from the U.S.).
401. See Adamo, 434 U.S. at 294?95 (Steven, J., dissenting) (arguing that the
term ?emission standard? should be interpreted to allow the requirement to water
down asbestos prior to demolition because this requirement has the ?effect? of
?curtail[ing] the quantity of asbestos . . . emitted?).
402. See id. at 306 (pointing out that the majority?s reading makes work
practice standards unenforceable).
403. See supra note 396 and accompanying text.
This case may have divided the Court not because it is hard to
discern what the enacting Congresses intended based on
purposivism and intentionalism (as required by the archaeological
approach), but because applying the statute faithfully to this
particular new problem puts judges in the uncomfortable position
of deciding an important question that is politically controversial
now. This adaptive aspect of Massachusetts v. EPA, not textual
ambiguity or any legitimate doubt about the CAA?s purpose, may
explain why it generated a 5?4 decision.404 Justice Stevens wrote
about the congruity of adaptation to a new problem with following
Congressional intent in his opinion for the Court:
While the Congresses that drafted ? 202(a)(1) might not have
appreciated the possibility that burning fossil fuels could lead
to global warming, they did understand that without regulatory
flexibility, changing circumstances and scientific developments
would soon render the Clean Air Act obsolete. The broad
language of ? 202(a)(1) reflects an intentional effort to confer
the flexibility necessary to forestall such obsolescence.405
Justice Stevens did not explain why failing to apply the CAA to
greenhouse gases renders the statute obsolete, but the answer is
plain enough. The CAA is intended to protect public health and the
environment, and it cannot do so absent coverage of greenhouse
gases. Thus, this passage can be read as supporting construing
general capacious language naturally to allow adaptations serving
the statute?s purposes.
Although the dissent?s statutory argument on the merits is
cast in terms of deference to EPA, the dissent on standing (for the
same four Justices) suggests a desire to follow the policy judgment
of current elected officials.406 Justice Scalia has elsewhere
expressed concern that adaptive interpretation might prevent
statutes from dying a natural death and therefore interfere with
the prerogatives of new Congresses.407 This conservative desire to
404. See Massachusetts v. EPA, 549 U.S. 497, 528?30 (2007).
405. Id. at 532.
406. See id. at 535 (Roberts, J., dissenting) (noting ongoing consideration of
climate disruption in Congress and the Executive Branch and describing the
litigation as an expression of dissatisfaction with the elected branches? progress
on the issue).
407. See K Mart Corp. v. Cartier, Inc., 486 U.S. 281, 324?25 (1988).
avoid expansive interpretation of regulatory statutes might help
explain why textually strained arguments appeared sensible to
dissenters who often find much more constraint in more
All of this suggests that dynamic statutory interpretation
based on the judges? views becomes inevitable when the statute
and its purpose bear an uncertain relationship to the problem at
hand, but that in cases where the text or purpose bear pretty
plainly on a new issue before the Court, dynamic purposivism
remains an available option. Indeed, the faithful agent theory (that
the Courts should be a ?faithful agent? of the enacting Congress)
would suggest that the dissenters erred because they should have
simply applied the text and purpose to the new pollutant.409 In
other words, recognizing that dynamic interpretation based on the
judges? views becomes inevitable in hard cases does not rule out
the possibility that dynamic purposivism can establish a
connection to majority preferences in easy cases.
Legislation usually attempts to govern the future by
establishing broad general principles that apply to future
circumstances, not through exclusive examples.410 Thus, the CAA
is about air pollutants, substances that harm public health and the
environment when emitted into the air, not just about the
particular pollutants causing the problems that Congress knew
about in 1970.411 The rule of law suggests that the Court should
408. Compare Massachusetts, 549 U.S. at 556?60 (Scalia, J., dissenting)
(advancing a textually strained interpretation of ?air pollutant?), with Michigan
v. EPA, 135 S. Ct. 2699, 2706?07 (2015) (majority opinion) (Scalia, J.) (reading
the phrase ?appropriate and necessary? as requiring consideration of cost and
declining to defer to EPA?s contrary interpretation under Chevron, U.S.A., Inc. v.
Nat. Res. Def. Council, Inc., 467 U.S. 837 (1984)), and EPA v. EME Homer City
Generation, L.P., 572 U.S. 489, 542 (2014) (Scalia, J., dissenting) (interpreting
the requirement that EPA promulgate a federal implementation plan ?at any time
within 2 years? after disapproving a SIP as not permitting exercise of that
authority prior to giving states additional guidance not explicitly required by the
409. See Manning, supra note 20, at 9.
410. See Pub. Citizen v. U.S. Dep?t of Justice, 491 U.S. 440, 475 (1989)
(?[Congress] usually does not legislate by specifying examples, but by identifying
broad and general principles that must be applied to particular factual
411. See Massachusetts, 549 U.S. at 506 (quoting the relevant statutory
follow the statute?s text and purpose when they provide an answer
to the question before them, as it did in Massachusetts v. EPA.
B. How and When to Adapt Statutes to Changes in Elite and
It may appear troubling to Justices to let legislation apply to
new problems within the ambit of applicable statutory language or
purpose, when the current legislature would not endorse the
result. That problem, we have suggested, caused a 5?4 split in
Massachusetts v. EPA.
Scalia?s concern about preserving the prerogatives of the
current Congress cannot supply an adequate answer to the
question of how to address a split between current political opinion
and the opinion of the drafters of an old statute. No matter what
the Court does, congressional prerogatives remain intact. Congress
remained free, after Massachusetts v. EPA, to pass legislation
forbidding federal regulation of greenhouse gases.412 And had the
Court held that the CAA currently does not authorize regulation of
greenhouse gases, Congress would remain free to override that
decision and authorize standards limiting greenhouse gas
emissions. The concept of ?congressional prerogative? must refer to
the right to pass legislation, which is a right of the institution as a
whole, not a right belonging to an individual member or a faction
within the Congress.
When the Court issues holdings contrary to statutory purpose
and when the enacting coalition remains in power in Congress and
the White House, judicial decisions may not matter very much. As
Hancock and Adamo illustrate, the congressional coalitions that
enacted the original statute can override judicial decisions
contrary to the enacting coalition?s intentions to correct the Court?s
misreading.413 Hence, the Court?s deviations from congressional
provisions, which require regulation of air pollutants defined as substances
emitted into the air that endanger public health and the environment rather than
list particular substances).
412. See id. at 533?35.
413. See United States v. Pa. Envtl. Hearing Bd., 584 F.2d 1273, 1287 n.22
(3d Cir. 1978) (recognizing that the 1977 Amendments superseded Hancock v.
intent in Hancock and Adamo had a very limited impact, because
Congresses that consistently favored environmental protection
promptly superseded these decisions.414
But when the enacting coalition does not remain in power,
judicial decisions, even incorrect ones, may matter a lot. The
Court?s CAA decisions have become more important lately,
precisely because political opinion has become divided and the
prospects for legislative correction of judicial decisions
correspondingly diminished. In this context, the policy impacts of
the Court?s decisions about how to adapt an old statute to changing
opinions and circumstances are sharpened.
The analysis above suggests that the rule of law requires that
judges follow the general intention of the enacting Congress even
when adapting the statute to a new problem that Congress did not
specifically consider. The Massachusetts v. EPA dissent did not
follow the law, nor would its opinion preserve some threatened
prerogative of Congress as a whole. Instead, the dissenting
Justices voted, in effect, to conform the statute to the views of much
of the Republican Party.415 Judges should not conform a statute to
current political opinion when it matters?when political opinion
is too divided to allow an override?because in that circumstance
they can only conform the statute to one party?s opinion not to a
widely shared political preference. Instead, judges should follow
statutory purpose and text revealing an intention about the matter
before them, even when the consensus undergirding the original
statute has shattered.
But in cases where such a general intention does not provide
clear guidance about how to resolve an issue before the Court, the
Court must make a policy decision about how the statute should
apply to the new problem. In other words, sometimes dynamic
statutory interpretation that reflects changes in views is inevitable
because, as Eskridge argues, archaeological approaches sometimes
Train, 426 U.S. 167 (1976)); United States v. Ethyl Corp., 576 F. Supp. 80, 82
(M.D. La. 1983), rev?d on other grounds, 761 F.2d 1153 (5th Cir. 1985), cert.
denied, 474 U.S. 1070 (1986).
414. See supra note 413 and accompanying text.
415. See Massachusetts v. EPA, 549 U.S. 497, 560 (2007) (concluding that
EPA should be granted discretion in deciding not to regulate greenhouse gases
and noting that the ?alarm over global warming may or may not be justified?).
provide no convincing connection to the intentions of the enacting
As a practical matter, judges will tend to resolve such open
questions based on their own views. Such a practice raises
constitutional concerns, which we analyze, but do not completely
The Court has frequently rejected judicial policymaking and
suggested it is constitutionally suspect.418 Doctrinally, this view
emerges perhaps most clearly in cases disfavoring the creation of
federal common law.419 But if the Constitution disfavors judicial
policymaking and congressional intent does not resolve a case,
what should a judge do?
In practice, judges tend to enact the views of the elites of which
they are a part into law.420 The CAA case study provides examples
of this tendency in what we might call the ?law and economics
cases??the cases construing ?appropriate and necessary? and the
Good Neighbor Provision to embrace consideration of cost.421 These
cases echo the Lochner-era cases, discussed by Eskridge, in which
the Supreme Court conformed the antitrust statutes to elite views
by using them to authorize injunctions against labor actions.422 In
all of these cases, the dynamic statutory constructions conformed
statutes to fairly well developed views amongst intellectual and
416. See ESKRIDGE, supra note 10, at 14.
417. Cf. id. at 108 (admitting that he cannot fully answer the question of how
judges should conduct dynamic statutory interpretation).
418. See, e.g., Tenn. Valley Auth. v. Hill, 437 U.S. 153, 194 (1977) (stating that
judges must ?put aside? their views of the wisdom of a congressionally chosen
course of action in interpreting a statute).
419. See, e.g., Milwaukee II, 451 U.S. 304, 312?13 (1981) (stating that federal
courts, ?purposefully insulated from democratic pressures,? normally lack
authority to develop rules of decision).
420. See ESKRIDGE, supra note 10, at 82 (discussing how judges created the
federal labor injunction in conjunction with other legal elites).
421. See supra notes 352, 367 and accompanying text.
422. See ESKRIDGE, supra note 10, at 82 (identifying the roots of the labor
injunction in the ?economic ideology held by legal elites?); cf. David M. Driesen,
Regulatory Reform: The New Lochnerism?, 36 ENVTL. L. 603, 614?21, 626?29
(2006) (discussing parallels between Lochner-era statutory interpretation and
judicial support for CBA under environmental statutes).
This approach to dynamic statutory interpretation raises
constitutional concerns, because it seems to conflict with popular
sovereignty. In the case of the CAA, opinion polls indicate that the
public does not generally buy into the balancing approach that has
captured the imagination of the elites. The labor injunction cases
harmed the reputation of the Lochner-era Court, because
ideological decision-making appears illegitimate, and Congress
repudiated many of these cases on several occasions, finally
leading the Court to renounce its elitist approach to antitrust
The idea that popular opinion should govern the resolution of
issues requiring dynamic statutory interpretation draws support
from the Constitution?s preamble, which states that ?we, the
people . . . establish? the Constitution and by its provisions
establishing policymaking authority in an elected Congress and, to
some extent, in an elected President.424 The idea that the entire
government derives its authority from the people suggests that the
Court should follow popular rather than elite views.
On the other hand, the Constitution tempers popular
democracy, reflecting some anxiety about the possibility of
democratic excess. These anxieties led to creation of the Senate,
the electoral college, the separation of powers, and an unelected
judiciary with lifetime tenure.425 For these reasons, it is hard to
423. See Milk Wagon Drivers? Union v. Lake Valley Dairy Prod., Inc., 311 U.S.
91, 101?03 (1940) (discussing legislative findings that the Court had largely
nullified Clayton Act restrictions on the labor injunction and holding that
Congress had drastically curtailed this practice in the Norris LaGuardia Act); cf.
ESKRIDGE, supra note 10, at 95?104 (providing a more nuanced account of the
labor injunction cases and congressional efforts to temper them).
424. See U.S. CONST. pmbl., arts. I?II; see generally BRUCE ACKERMAN, WE THE
PEOPLE (1991); LARRY D. KRAMER, THE PEOPLE THEMSELVES: POPULAR
CONSTITUTIONALISM AND JUDICIAL REVIEW (2004); Jack M. Balkin & Sanford
Levinson, Understanding the Constitutional Revolution, 87 VA. L. REV. 1045
425. U.S. CONST. art. I, ? 3 (requiring legislators to choose Senators); id. art.
II, ? 1 (providing that an electoral college select the President); id. art. II, ? 2
(providing for Presidential appointment of Supreme Court Justices); id. art. III,
? 1 (providing lifetime tenure for Supreme Court Justices); id. amend. XVII
(providing for popular election of U.S. Senators, but leaving the number at two
per state regardless of population). See generally JESSE H. CHOPER, JUDICIAL
REVIEW AND THE NATIONAL POLITICAL PROCESS: A FUNCTIONAL RECONSIDERATION
OF THE ROLE OF THE SUPREME COURT (1980) (trying to reconcile the
argue that the Constitution wholly precludes independent elitist
policymaking by the judiciary, however troubling that might be.426
We have some doubts about whether the Court could
accurately follow current popular views in resolving questions that
past congressional intent cannot resolve even if it chose to do so.
Opinion polls can be unreliable and may not track the issues before
the Court with sufficient precision. Yet, some awareness by judges
that their most natural inclinations may prove at odds with
popular views should temper their approach to dynamic statutory
The tendency of judges to follow elite views appears most
troubling when it influences the outcomes in cases that do not pose
new problems and therefore do not seem to require dynamic
statutory interpretation. The law and economics cases appear
troubling in part because the problems they dealt with, while
perhaps not resolved clearly by statutory text, were anticipated by
Congress.427 One would expect the CAA?s philosophy, which
reflects popular opinion, rather than the judiciary?s elitist
approach to control such cases. Dynamic statutory interpretation
based on elite views seems more appropriate when confronting a
new problem not anticipated by Congress, such as the problem of
the scope of greenhouse gas regulation that the UARG Court
Chevron, whatever its weaknesses, does provide a resource for
ameliorating some of the dilemmas arising in cases demanding or
tending to trigger dynamic statutory interpretation.428 The Court
?countermajoritarian? nature of judicial review with democracy); JOHN HART ELY,
DEMOCRACY AND DISTRUST: A THEORY OF JUDICIAL REVIEW (1980) (same); Deni
Gewirtzman, Glory Days: Popular Constitutionalism, Nostalgia, and the True
Nature of Constitutional Culture, 93 GEO. L.J. 897 (
) (critiquing popular
constitutionalism as insufficiently attentive to problems of lack of informed
popular participation and the ability of elites to manipulate public opinion).
426. See Fareed Zakaria, America?s Democracy has Become Illiberal, WASH.
POST (December 29, 2016),
https://www.washingtonpost.com/opinions/america-isbecoming-a-land-of-less-liberty/2016/12/29/2a91744c-ce09-11e6-a747d03044780a02_story.html?noredirect=on&utm_term=.4208ce175a10 (last visited
Dec. 4, 2018) (pointing out that Alexander Hamilton expected lawyers and other
professionals to make sure that democracy served broad national interests rather
than special interests) (on file with the Washington and Lee Law Review).
427. See supra notes 333, 345, 346, 352, 367 and accompanying text.
428. See Chevron U.S.A., Inc. v. Nat. Res. Def. Council, 467 U.S. 837, 842?43
could defer to EPA, instead of acting on its own when the statute
is ambiguous.429 But the CAA cases involving Chevron deference
suggest that Chevron has not constrained ideological
decision-making, because the Justices apply it so selectively.430
They refuse to defer to the Obama EPA?s interpretation of the
completely open-ended clause ?appropriate and necessary? but
many of them find a provision defining a pollutant as including any
substance emitted into the ambient air sufficiently ambiguous to
justify deference to the Bush EPA.431
One of us has suggested that the Roberts Court reflects, to
some degree, the polarization of our times in its constitutional
decision-making.432 This case study suggests that the same thing
has happened in the context of an extraordinarily detailed and
prescriptive statute, the CAA. Increasingly, these divisions occur
in archaeologically easy cases (like Massachusetts v. EPA), not just
in hard ones. And this division often occurs as the Justices engage
in dynamic statutory interpretation in cases that might be resolved
fairly easily by intentionalism. While Eskridge is surely right that
dynamic statutory interpretation based on judicial views of
sensible adaptation is inevitable in hard cases, it is troubling to see
it have such a strong influence in easy ones.
While statutory text and purpose played a major role in the
CAA?s early years, increasingly the Court issues rulings divided
(1984) (articulating the two-step framework for reviewing agency statutory
429. See id. at 843.
430. Cf. Connor N. Raso & William N. Eskridge, Jr., Chevron as a Canon, Not
a Precedent: An Empirical Study of What Motivates Justices in Agency Deference
Cases, 110 COLUM. L. REV. 1727, 1741, 1784 (2010) (finding in a study of 667 cases
decided between 1984 and 2006 that ?ideology . . . correlates significantly with
how Justices vote on deference doctrine?).
431. Compare Michigan v. EPA, 135 S. Ct. 2699, 2706?07 (2015) (finding
EPA?s interpretation of appropriate and necessary as not requiring consideration
of cost unreasonable), with Massachusetts v. EPA, 549 U.S. 497, 558 (2007)
(Scalia, J., dissenting) (characterizing the CAA?s definition of air pollution as
ambiguous and arguing that the Court should therefore defer to EPA).
432. See KECK, supra note 300, at 13 (finding the Justices ?polarized along
partisan lines, but not as badly as members of Congress?).
II. The Clean Air Act........................................................... 1788? A . Key Statutory Features ........................................... 1788? B. The Evolution of Philosophy Toward Clean Air ..... 1791? C. Global Climate Disruption ...................................... 1796? D. Summary of Background......................................... 1799 ?
III. Purpose , Text, and Judge-Made Law : The Court's Interpretation of the Clean Air Act from 1970 to 2004 ........................................................... 1799? A. Purpose and Text in Statutory Interpretation ....... 1799? B. Text and Purpose: 1970 ? 1980 ................................. 1802 ? C. The Decline of Purposivism and the Rise of Judicial Policymaking in the 1980s . ................... 1810 ? D. The Rise of Textualism: 1990 - 2004 ........................ 1819 ? E. Summary of 1970 -2004 ........................................... 1825 ?
IV. Climate Disruption , Elite Thinking, and the Court: Dynamic Statutory Interpretation .................... 1826 ? A. The Supreme Court and Climate Disruption ......... 1827 ? B. Adaptation to Elite Views ....................................... 1835 ?
C. Summary Description of the Evolution of the Supreme Court's Approach to Interpreting the Clean Air Act ..................................................... 1843 ?
V. Lessons for Dynamic Statutory Interpretation ............ 1846 ? A. Purpose as Dynamic Archaeology ........................... 1846 ? B. How and When to Adapt Statutes to Changes in Elite and Political Opinion ...................................... 1851 ?
VI. Conclusion ...................................................................... 1856 ? 19 . See Clean Air Act Amendments of 1990 , Pub. L. No. 101 - 549 , 104 Stat.
2399 (codified as amended at 42 U .S.C. ? 7401 ( 2012 )). 20 . See , e.g., John F. Manning, Textualism and the Equity of the Statute, 101
COLUM. L. REV . 1 , 7 ( 2001 ) (stating that ?near the close of the twentieth
century . . . the 'new textualism' called . . . strong purposivism? into question) . 21. CASS R. SUNSTEIN , THE COST-BENEFIT STATE : THE FUTURE OF REGULATORY
PROTECTION 3 ? 18 ( 2003 ). 34 . See id. (explaining that these standards are to be based on the
?maximum achievable control technology?) . 35 . See id. at 1080 (noting that the second stage of regulation is
?health-based? as it was prior to the 1990 Amendments) . 36 . 42 U.S.C. ?? 7604 ( a ), 7607 (b). 37 . See , e.g., Errol Meidinger, On Explaining the Development of 'Emissions
Trading' in U.S. Air Pollution Regulation, 7 L . & POL 'Y 447 , 451 ( 1985 )
a modern industrial state?) . 38 . See U.S. Steel Corp. v. EPA , 444 U.S. 1035 , 1038 ( 1980 ) (Rehnquist , J. ,
Control-Compelling Versus Site-Shifting, 74 IOWA L. REV. 1 , 10 ( 1988 ) (discussing
the PSD program's structure and goals) . 39 . See Util . Air Regulatory Grp. v. EPA , 573 U.S. 302 , 309 ( 2014 )
(describing the PSD program's BACT provisions ); Alaska Dep't of Envtl. 61. See Unfunded Mandates Reform Act of 1995 , Pub. L. No. 104-4 , ? 202 ( a ),
109 Stat. 48 ( codified as amended at 2 U .S.C. ? 1532 ( 2018 )). 62 . See , e.g., Exec . Order No. 12 , 866 ? 1 (b)(6), 58 Fed . Reg. 51 , 735 (Oct. 4 ,
1993 ) (stating that a regulation's benefits should ?justify? the costs); see also Lisa
Heinzerling , Quality Control : A Reply to Professor Sunstein, 102 CALIF. L. REV.
1457 , 1462 ( 2014 ) (critiquing the views of President Obama's first head of OIRA,
Cass Sunstein) . 63 . See Andrew McFee Thompson , Comment, Free Market
Inconsistency , 45 EMORY L.J. 1329 , 1329 ? 30 ( 1996 ) (noting that by the time of the
environmental statutes) . 65 . See Massachusetts v. EPA , 549 U.S. 497 , 509 ( 2007 ) (noting that in 1990
activities are increasing the temperature) . 66 . See id. at 521-23 (referencing the scientific reports documenting these
consequences) . 67. See id. at 545 (Roberts, C.J. , dissenting) ( noting that 80% of greenhouse
gas emissions ?originate outside the United States?). 68. See Global Change Research Act of 1990 , Pub. L. No. 101 - 606 , 104 Stat.
3096 (establishing a program ?aimed at understanding and responding to global
L. 100 - 204 , 101 Stat. 1331, Title XI (codified in note following 15 U .S.C. ? 2901 )
to make diplomatic efforts to address it ). 69. Cf . 42 U.S.C. ? 7602 (h) ( 2012 ) (defining welfare effects to include effects
on climate) . 70. Montreal Protocol on Substances that Deplete the Ozone Layer, Sept . 16 ,
1987 , 1522 U.N.T.S. 29 (U.S. ratification Apr. 21 , 1988 ). 71 . See Massachusetts, 549 U.S. at 508-09 ( noting that the IPCC ?published
its first comprehensive report? in 1990 and that President Bush signed the United
Nations Framework Convention on Climate Change in 1992). 72. United Nations Framework Convention on Climate Change, May 9 ,
1992 , 1771 U.N.T.S. 107 (U.S. ratification Oct. 15 , 1992 ). 73 . See Massachusetts, 549 U.S. at 455 (mentioning the Framework
reduction commitments) . 74 . Kyoto Protocol to the United Nations Framework Convention on Climate
Change , adopted Dec. 11 , 1997 , 2303 U.N.T.S. 162 ( entered into force Feb . 16 ,
2005 ). 75 . Letter to Members of the Senate on the Kyoto Protocol on Climate
Change , 37 WEEKLY COMP. PRES. DOC. 444 (Mar. 13 , 2001 ). 76 . See , e.g., RICHARD ELLIOT BENEDICK , OZONE DIPLOMACY : NEW DIRECTIONS
IN SAFEGUARDING THE PLANET ( 1998 ) (describing the Reagan Administration's
international leadership in addressing stratospheric ozone depletion) . 332 . 42 U.S.C. ? 7410(a)(2)(D)(i)(I). 333 . See North Carolina, 531 F. 3d at 908 (stating that EPA's program to
significantly to nonattainment). 334. See supra notes 106-107 . 335 . See EPA v. EME Homer City Generation, L.P. , 572 U.S. 489 , 500 ( 2014 );
see also EME Homer City Generation , L.P. v. EPA, 696 F.3d 7 , 14 (D.C. Cir . 2012 )
(citing Michigan v . EPA , 213 F. 3d 663 (D.C. Cir . 2000 )) (discussing the D .C.
EPA , 531 F. 3d 896 (D.C. Cir . 2008 ), vacated, 572 U.S. 489 ( 2014 ). 336 . See Homer City, 572 U.S. at 544 n.10 (citing Interstate Transport of Fine
Particulate Matter and Ozone and Correction of SIP Approvals , 76 Fed. Reg.
48271- 72 ( Aug . 8, 2011 ) (codified at 40 C.F.R. pts. 51 , 52 , 72 , 78 , 97)). 337 . See North Carolina v. EPA , 531 F.3d 896 , 906 -08 (D.C. Cir . 2008 )
neighboring states' attainment) . 348 . 135 S. Ct . 2699 ( 2015 ). 349 . See id. at 2711-12 (citing 42 U.S.C. ? 7412(n)(1)(1) ( 2012 )). 350 . See id. at 2705 (explaining that EPA decided to regulate coal and oil-fired
environment? and ?controls were available to reduce these emissions?). 351. See id . at 2705-06 ( explaining that EPA ?concluded that 'costs should not
Performance for Fossil-Fuel-Fired Electric Utility, 77 Fed . Reg. 9326 ( Feb . 16,
2012 ) (codified at 40 C.F .R. pts. 60 , 63))). 352 . See id. at 2707 (characterizing the phrase ?appropriate and necessary?
as capacious (quoting 42 U .S.C. ? 7412(n)(1)(A) ( 2012 ))). 353 . See Indus . Union Dep't, AFL-CIO v . Am. Petroleum Inst. , 448 U.S. 607 ,
615 ( 1980 ) (plurality opinion) (interpreting a directive to regulate when
places of employment? to require a finding of significant risk ). 354. See Michigan , 135 S. Ct . at 2718 (Kagan, J., dissenting) ( arguing that 370 . See, e.g., Michigan v . EPA , 135 S. Ct . 2699 , 2709 ( 2015 ). 371. Supra Tables 1 and 2 . 372. See , e.g., Union Elec . Co., 427 U.S. at 258 ( using both the ?language and
technological or economic infeasibility?) . 373 . See , e.g., Pennsylvania v . Del. Valley Citizens' Council , 478 U.S. 546 ,
562- 68 ( 1986 ) (citing to other cases to derive a common-law interpretation of
reasonable attorney's fees) . 374 . See , e.g., id. at 548 . 375. See Chevron U.S .A., Inc. v. Nat. Res. Def. Council , 467 U.S. 837 , 865