Air Quality Standards: The Decisionmaking Process

CRS Report for Congress
Air Quality Standard s:
The Decisionmaking Process
Upda ted April 9, 2002
John E. Blodgett and Larry B. Pa rker
Specialists in Energy and En vironmental Po licy
Resources, Sc ience, and Industry Division

Congressional Research Service ˜ The Library of Congress

Air Quality S tandards: The Decisionmaking P rocess
Summary
The d ecisions by the Administrator of the Environm e n t a l P rotection Agency
(EPA) i n J uly 1997 to revise ambient air quality standards (NAAQS) for oz one and
particulat e m atter refocused attention on t he criteria and the proces s by which thes e
deci si ons a r e m ade. The n ew st andard s were t he subject of numerous oversight
hearings as well as litigation, which culminat ed in a S uprem e C ourt ruling February
27, 2001. The court upheld the NAAQS-setting procedures at question, in particular
definitively rejecting t he consideration of costs in setting NAAQS.
However, the court’s ruling also raised questions concerning the implementation
of the EPA’s new o z one standard. W ith co n t i n u i ng controversy over t he PM and
o z o n e s tandards, along with other concerns about the C lean Air Act (CAA), t he
ex piration i n 1998 of the authoriz ations fo r approp r i ations in the s tatute, and the
Bu sh Administration’s p roposal for amendmen ts, i t i s possible t hat t he Congress may
take up amendments to the CAA. If so, t he NAAQS decisionmaking process may
command attention, es peci ally with respect to how scientific evidence is used.
Because of the role NAAQS might play in bringi ng amendments onto t he legi slative
agenda, t his report p rovides b ackground on the p rocesses and procedures for s etting
and revising NAAQS . The basic s teps are as follows:
! EPA i dentifies a pollutant that is emitted from numerous or diverse m obile or
stationary sources and t hat enda ngers public health or welfare.
! EPA prepares a “criteria document” that summarizes t he scientific information
relevant to the pollutant; t his docum ent i s form al l y revi ewed by a C l ean Ai r
Scientific Advisory Committee (CASAC).
! EP A p repares a “staff paper” that summariz es t he criteria document and lays
out policy options for t he Administrat or; i t i s also reviewed by C ASAC.
! Based o n t he criteria document, the s t a f f paper, and C ASAC’s “closure
letters,” the Administrator proposes a NAAQS; t his p roposal is published i n
the Federal Regi ster, a “docket” creat ed, and an opportunity for public review
and comment provided. And,
! The Administrator’s final d ecision, “which, i n t he judgment of the A d m i n -
istrator, ... [ i s] requisite to protect the public health ... or public welfare.”
The C AA spells out requirements for the criteria document, the C ASAC review,
the b a s i s o n which t he Administrator c hooses the s tandard, and the p rocedural
process for promulgating t he standard. EPA administrativel y added t he preparation
of a “staff p aper”; in addition, Ex ecutive Order 12866 requires a Regu latory Im pact
Analys is (RIA), although t he economic a n a l ys is is essentially irrelevant to the
decision a NAAQS . Other laws also raise r egulatory assessment i ssues. T h e A ct
requires E P A to revisit each NAAQS every 5 years, following the s ame p rocess.
Several aspects i n t he NAAQS-setting p rocess h ave b een the foci o f attention
in the past and might be revisited: these i nclude the Act’s requirement that NAAQS
be set t o p rotect health with an adequate margin of safety, without consideration o f
costs; the p rocess for verifying t he scientific underpinnings o f a proposed standard;
th e boundaries on the Administrator’s judgment in accounting for risk and
uncertainty in setting NAAQS; EPA’s responsiveness t o public comments; and t he

EPA t o consider impact s of its regulations.

Contents
In troduction ......................................................1
Background ......................................................2
CriteriaAirPollutants ..............................................3
National Ambient Air Quality Standards ................................4
PrimaryStandards .............................................4
“MarginofSafety” .........................................5
Secondary Standards ...........................................6
Setting NAAQS ...................................................7
Setting and Reviewing NAAQS ..................................7
CriteriaDocument .........................................7
StaffPaper ...............................................9
Clean Air S ci entific Advisory Committee .......................9
TheAdministrator’s Decision ...................................11
TheEvidence ............................................13
TheCriteria .............................................13
Health ..............................................13
Sensitive populations ..............................14
Adversehealth effect ..............................15
Exposureevaluationand risk assessment ..............15
Marginofsafety ..................................16
Costs ...............................................17
P romulgating NAAQS .............................................19
TheAdministrativeRulemakingProcess ...........................19
Notice o f P roposed Rulemaking .............................20
PublicComment .........................................20
PromulgationoftheRule ...................................20
RegulatoryImpact Assessments .............................21
Costs and Benefits — Ex ecutive Order 12866 ..............21
Unfunded M andates — Unf unded M andates R eform Act ......22
Smal l Business — Regulatory Fl ex ibility Act ...............23
OtherRegulatoryImpact Assessments ....................24
Consultations — Office of Management and Budget; Other Departments
andAgencies ........................................25
Judicial Review ..............................................25
CongressionalReview .........................................26
Appendix I. The NAAQS-Setting P rocess ..............................28
Appendix II: Setting NAAQS — a Typical Chronology (Oz one) ............29

Air Q uality Standards:
The D ecisionmaking Process
Introduction
Decisions in J u l y 1997 by the Administrator of the Environmental P rotection
Agency (EPA) t o revise t he national ambient air quality standards (NAAQS) for
ozone and particulat e m atter (PM) refocused attention on t he criteria and the process
by whi ch t hese deci si ons are m ade. The n ew st andards h ave b een at t acked bot h as
ov e r l y stringent and as inadequately prot ective of health; as i gnoring costs and as
gi ving costs t oo much deference; as go ing b eyond what is scientifically conclusive;¹
and as failing t o b e s ufficiently precautionary. These i ssu e s were t h e s ubj ect of
numerous oversight hearings as well as litigation, which culminat ed in a S uprem e
Court ruling February 27, 2001. The court upheld those NAAQS setting p rocedures
at question, in particular definitivel y rej ecting t he consid eration of costs in setting
NAAQS.
However, the court’s ruling also raised questions concerning the implementation
of the EPA’s new o z one standard. W ith the court’s d ecision, continuing controversy
over t he PM and o z one standards, along with other concerns about the C lean Air Act
(CAA), t he ex piration i n 1998 of the authoriz ations for appropriations in the s tatute,²
and t he Bush Administration’s air quality initiative t hat would amend the Act , it is
possible t hat t he Congress may t ake u p amendments to the C AA. If so, t he NAAQS
deci si onm aki n g p rocess m ay com m and at t ent i on, especi al l y wi t h respect t o how
scient i f i c evidence is used – and possibly reopening the question o f whether costs
should b e considered in setting NAAQS.
Because of discrepan cies i n v iews on what transpired — o r s hould h ave
transpired — i n EPA’s development o f t he 1997 oz one and P M s tandards, this report
provides b ack grou n d on the p rocesses and procedures for s etting and revising
NAAQS . It l ays out the s t e p s o f the d ecision process and identifies t he statutory
criteria established b y t he Act for NAAQS, to aid the reader’s understanding of the
Act’s policy for NAAQS . It does not evaluate the d ecisionmaking p rocess nor assess
EP A’s d ecisions on the o z one and P M NAAQS ; nor does i t address t he several o ther
cat egories o f pollutants (e.g. , h azardous ai r pollutants) or other s tandards-setting
activities under t he Act.

¹ See CRS Report 97-8, Air Quality: EPA’s New Oz one and Particulate M atter Standards.
² Ge o r ge W . Bu s h, T he White House, “T he Cl ear Skies Initiative,” Feb. 14, 2002. For
current information on clean air l egislation, see J ames McCarthy, Cl ean Air Act I ssues in
the 107th Congress , CRS Issue Brief IB10065.

Background
The p rocess b y which EP A s ets and revises NAAQS evolved over m any years.
In itially, federal air quality law focused on supporting s tate programs, t hrough t he
conduct o f research (a central, federal res earch program w as more efficient t han 5 0
state research efforts) and t hrough t ec h n i cal and financial support (regu lation was
seen as a s tate and l ocal matter). By the l ate 1960s, t he federal re s e a rch effort,
locat ed in the Nat i onal C enter for Air P ollution C ontrol, Public Health Service,
Department of Health, E ducation, and W el fare (HEW ), included t he preparation o f
“criteria” — a summary of scientific knowledge concer n i n g s e lected air pollutants
— and “guidelines” — a s ummary of control t echnologi e s . These criteria and
gu idelines were to support s tate programs to control air pollution.
As environmental awar e n e s s heighten ed in the l ate 1960s, t he federal role
enlarged. W ith the C lean Air Act Ame ndments of 1970, the federal clean air³
p r o gram was moved t o t he newly created EPA and reconstituted, with the federal
government becoming responsible for estab lishing “national ambient air quali t y
standards” for air pollutants endangering public health or welfare and resulting from
numerous or diverse m obile or stationary sources . The st at es remained primarily
responsible for d eveloping implementation plans to attain and m aintain compliance
with those national s tandards . Major amendments in 1977 and 1990 refined t he
NAAQS setting and implementation process. The 1977 amendments formally
es tablished t he Clean Air S ci entific Advisory Committee (CASAC) and t he 5-year
review process. The 1990 amendment s i n creased federal guidance t o s tates i n
implementing ex i sting NAAQS .
Thus t h e p resent process for setting and revising NAAQS consists of the
s t atutory s teps incorporated in the C AA over a series of amendments. In additi o n ,
several o ther steps o f v aryi ng degrees of i m p o r t h a v e b een added b y t he EPA, by
ex ecut i v e o rders, and b y s ubsequent regu l atory reform enactments b y t he C ongress.
The s teps are d epicted i n appendix I; for an ex ample chronology o f t h e steps, see
appendix II. T h e d o c umentary language spelling out those s teps, along with
ex planatory notes, follows.

N^OTE: C AA language is in bold ; l egislative
history l angu a g e , i.e., from reports on
am endments to the C AA, is in b old italics;
language from s tatutes other than the C AA is
initalics.
Citations to ozone or PM standards setting
refer t o t he documentation for the 1997 oz one
andPMNAAQS.
³ EPA was created by Reorga niza tion Plan No . 3 i n 1970, reprinted a s 42 U.S.C. §4321 note;
the Agency went i nto operation December 2 of that year.

Criteria Ai r P ollutants
The pollutants for which NAAQS are s et ar e o ften called “criteria pollutants.”
This term reflects t he evolution o f t he process. Before there were NAAQS , t he Air
Quality Act o f 1967 (a predecessor o f t he CAA) required t he Federal government to
list air pollutants and to prepare “criteria” — a report s ummarizing scientific
evi d ence concerni ng t h ei r h eal t h effect s. The s t at es t hen s et st andards. Thus t h e ai r
pollutants for which NAAQS were later s et we re originally air pollutants for which
the Office of Air Quality in HEW p repared “criteria. ” The cu rrent language
specifying t he listing o f “criteria” air po llutants, as modified by the 1970 and 1977
am endments to the C AA, is as follows:
The Administrator shall . . . p u b lish, and shall f rom time to time
th ereaf ter revi s e, a l i s t w h i ch i n c l u d e s each a i r p o l l u tan t — (A)
emissions of w hich , in h is judgment, cause or contribute to air
pollution w h ich may reasonably b e anticipated to en d a n g er public
health or welfare; [and] (B) the presen ce of which in the ambien t air
resu l ts f rom n u merou s or d i verse mob i l e o r station a ry sou rces ....
[ C AA §108(a)(1)]
W ith the C lean Air Act Amendments of 1970, the EP A became responsible for
determining national ambient ai r quality standards for the “cri t eria pollutants” —
those o f s uch n ational s c o p e in releas e o r effect that national s tandards are
appropri a t e . Also, EPA was to assist stat es in attaining t hese standards t hrough
grant s and t echni cal assi st ance,⁴ and could impose s anctions on states for failures t o
fulfill thei r obligations. At present, s ix pollutants are designat ed “criteria pollutants”
for which NAAQS have been set: particulate m atter⁵ (P M), o z one (O³, a key m easure⁶
of smog), nitrogen diox ide (NO^2,, or, inclusively, nitrogen oxides, NOx ), sulfur
ox i d es (S Ox , o r, speci fi cal l y, S O²), carbon monox ide (CO), and l ead (Pb).

⁴ For example, i n conj unction with setting NAAQS, t he CAA r equires EPA t o p r e p are
information on air pollution control t echnologies:
Si multaneously w i th the i ssuance of crite r i a . . ., t he Admini strator shall,
af ter c onsultation w ith appropriat e advisory c ommittees and F ederal
depart m e nt s a nd agenci e s, i ssue t o t he St at es and a ppropri at e a i r pol l ut i on
control agencies i nf ormation on air p ollut i o n control t echnologies, w hich
inf ormation shall include data relating t o t he cost of installation and
operation, e nergy requirements, emission reduction b e nef i ts, and
environmental i mpact of t he emission control t echnology. [CAA, §108(b)(1)]
T he preparation of t hese “guidance documents,” as they are known, is not revi ewed in this
report.
⁵ T h e o r i gi na l P M NAAQS wa s f or “ T ot a l Sus p e nde d P a r t i c ul a t e s ” ; t h e s t a nda r d wa s l a t e r
focused on particles smaller t h a n 1 0 mi c rons (PM¹⁰) a nd in 1997 EPA proposed that a
standard be added f or particles s maller t han 2.5 mi crons (PM^2.5 ).
⁶ T h e NAAQS i s f o r NO² ; nitrogen gases t hat are ozone precursors are referred t o as NOx.

National Ambient Ai r Quality S tandards
The Administrator ... sh a l l p ublish ... regulations prescri bing a
national primary amb ient air quality standard and a n ational
secondary ambien t air quality standard for each air polluta n t f o r
w h ich a ir quality criteria h ave been issued .... [ C AA §109(a)(1)(A)]
A NAAQS is a uniform, n ational s tanda r d establishing the m ax imum
permissible concentration of an air pollutant in the ambient ai r — the “portion of t he
at m o sphere, ex t ernal t o bui l d i n gs , t o w hi ch t h e general p u b l i c has access.”⁷ The
CAA direct s t he Administrat or to set a N A AQS at a l evel that provides protection
from adverse effects o n t he public health and welfare. T he Act p rovides for “primary
standards” to protect health with a m argi n of s afety and for “secondary standards” to
protectwelfare.
Pr i m ar y S tandar ds
Nati o n a l primary amb ient air quality standard s ... shall b e amb ient
air quality standard s the attainment and mainten ance of which in the
judgment of the Admi n istrator, b ased on such criteria and allow i ng
a n ad eq u a te ma rgi n of saf ety, a re req u i si te to p rotect th e p u b l i c
health. [ C AA §109(b)(1)]
An am bien t air quality standard ... should be the m axim u m perm i ssible
am bi en t a i r l evel of an ai r pol l u t i o n agen t or cl ass o f s u ch agen t s
(rel ated to a period of tim e) which w ill protect t h e h ealth of any group
of the population. [ S enate R eport No. 91-1196, p. 10]
As stated in §109, a NAAQS defines t he acceptabl e concentration o f an air
pollutant in the ambient ai r necessary to protect health. As discussed l at er, costs are
not considered in setting primary standards. Areas — defined as ai r quality control
regi ons — complyi ng with the s tandard are des ignated “attainment areas ” and areas
in which t he pollutant ex ceeds t he standard are d esignated “nonatta i n m e n t areas.”
An area can be i n at t ai n m ent for one ai r pollutant and out of attainment for another.
Although a primary NAAQS has a federally enforceab l e d e adline, a NAAQS
does not i t s el f est abl i s h what t o d o when concent rat i ons ex ceed t h e s t andard (nor
how t o prot ect cl ean ai r i n at t ai n m ent areas ). Other p rovisions of the C AA lay out
the p rocess o f implemen t ation (specifi cally, §110). For nonattainment areas, t his
incl udes m onitoring am bient air quality to determine compliance; requiring stat es to
prepare s tate implementation plans (SIP s) to bring areas not in compliance i nto
com p l i ance; and requi ri ng st at es t o i m p l em ent t h ei r p l ans t o achi eve and m ai nt ai n t he
NAAQS by a s pecified deadline (Title I, Part D o f t he A c t). Federal measures to
protect air q u a l i t y i nclude new s ource performance standards (§111), which are
nat i onal s t andards o n s peci fi ed cat egori es o f n ew sources of sel ect ed ai r pol l u t ant s,
and m obile source emission standard s (§202). A l s o , the federal government
continues t o prepare “guidance documents” s pelling out available control m eas ures

⁷ 40 CFR §50.1(e). Air within workplaces is subj ect to regulation by t he Occupational
Health and Safety Administration.

for t he NAAQS pollutants. Finally, i n attainment areas where t he air i s cleaner than
NAAQS, t he CAA establishes a p rogram for new s ource permitting t o “maintain”
that cl ean ai r quality (Title I, Part C).
“Margin of S afety”.
... [A]llow i ng an adequate margin of saf ety .... [ C AA §109(b)(1)]
In setting such [ national am bient air quality] standards the
[Adm inistrator] should consider and incorporate n ot only the results of
research sum m arized i n air quality criteria docum en ts, but also the n eed
for m a r g i n s o f safety. Margins of safety are essential t o any health-
related environm ental standards if a r easonable d egree o f p rotection i s
to be provided against h azards which r esearch h as not yet iden t i fied.
[ S enate C ommittee on P ublic Works, Report No. 91-1196 (1970), pp. 9-

10]

The phrase, “m argi n o f s afet y, ” h as been cent ral t o several d ebat es.
The phrase s eems t o imply that NAAQS are b ased on thresholds — t hat t here
is a concentration of a pollutant below which adverse health effect s do not occur. In
this view, t he Administrat or determines the “no effect ” t hres hold for the pollutant
from t he sci ent i fi c evi d ence i n t h e “cri t eri a docum ent ” and t hen adds a s afet y fact or.
From this perspective, t h e i ssue i s how large t he margin of safety should b e. In a
report o n automobile emissions in 1974, a p anel of the Nati o n al Academy o f
S ci ences observed t hat “ t h e s a f et y fact or s p rovided b y t he air s tandards are much
smaller t han i s usual in regulating other environmental pollut a nts s uch as⁸
radi o a c t i v i t y....” Thi s coul d b e i nt erpret ed as suggest i n g t he need for t i ght er
standards,⁹ thus providing a l arger m argi n o f s afety.
In fact , s ci ent i s t s fi nd t h at t h ere s eem not t o be cl ear t h reshol ds for ai r
pollutants. The 1974 National A c a d emy o f S ciences panel o n auto emissions
observed t hat—
... i n n o case i s t here evi d ence t h at t h e t hreshol d l evel s h ave a cl ear
physiological meaning, in the s ense that there are genuine adverse h ealth
effects at and above some level o f pollution, but no effects at all below t hat
level. On the contrary, evidence i ndicates that the amount of health
damage varies with the upward and downward v ariations in the¹⁰

concentration of t he pollutant, with no sharp l ower limit.
⁸ National Academy of Sciences, Ai r Quality and Automobile Emi s s i on Control, Vol. 1
(September 1974), p. 6.
⁹ U.S. Congress, House, Committee on Interstate and Foreign Commerce, Cl e a n A i r A c t
Amendments of 1977, House Rept. No. 95-294, to accompany H.R. 6161 (95^th Congress, 1^st
session) (Washington, D.C.: U.S . Govt. Print. Off., 1977), p. 182.
¹⁰National Academy of Sciences, Ai r Quality and Automobile Emission Control, Vol. 1
(September 1974), p. 17.

The 1996 reviews o f t he oz one and P M s tandards likewi s e concluded t hat n o
threshold o f adverse effects could b e found for either pollutant.
The i nability of scientists to find a t hres hold has led t o contention i n t he setting
of NAAQS. S ome argue that if there i s n o t hreshold, then there cannot be a m argi n
of safety and as a result the whole NAAQS process o f n ecessity becomes a risk
management deci sion — t hat i s, one in which t he Administrat or balances risks with
cost s t o d eci de where t o s et t h e s t andard. Ot h ers argue t h at t h e l ack of a t hreshol d
justifies t he tightes t possible s tandards.
Another, related d ebate comes from t he view that only i n adding a “margin of
safety” does t he administrat or layer a policy j ud gm e n t onto an objective,
scientifically determined NAAQS . S ome argue that this j u dgmental aspect means
that the “margin of safety” phrase implicitly endorses t he consideration of costs in
setting NAAQS; as d iscussed l ater, t he lead industry s ued EPA over l ead standards
on the b asis that the “margin of safety” requi r e d EPA to take costs i nto account in
s e t t i ng NAAQS , but the court ruled that the s tatute and its legi slative h ist o r y a r e
agai nst t hat i nt erpret at i o n ( Lead Industries Association v. Envi ronment a l P rot ect i o n
Agency , 647 F.2d 1130 (D.C. C ir. 1980)).
The l egislative history of the C AA only b riefly touches on t he “m argi n of
safety” phrase. The S enate R eport 90-1196 (accompanyi ng legi slation t hat b ecame
the C lean Air Amendments of 1970), quoted at the b eginning of this section, clearly
indicat es that the “margin of safety” i s des igned t o protect agai nst t he potential for
adverse effects t o o ccur at pollutant concentrations below t hose known t o cause
harm. T hus, regardless o f t he ex istence o f a t h reshol d, t h e m argi n o f s afet y i s a fact or
t h e Administrator would consider in making choices involving uncertainti e s
embedded i n t he definitions of which vulnerable population groups to protect and o f
what effect s are ad v e rs e h ealth effect s: these i ssues are discussed l at er. S imilarly,
House R eport 95-294 (accompanyi ng legi slation t hat b ecam e t he Clean Air Act
Amendments of 1977), after quoting t h e N a t i o n a l Academy o f S ciences about the
“smaller t h a n u s u al” s afety factor in NAAQS and about the l ack of evidence for
thresholds, s uggested “greater n ot lesser control of em is s i o n s a r e likely t o b e
n eeded .”¹¹ The p recaut i onary prem i s e o f t he act seem s m ani fest i n t h e p h rase
“m argi n o f s afet y” regardl ess of t h e ex i st ence of t h reshol ds; and t h e i nference t h at
the phras e calls for consideration of costs has been consistently reject ed.
Secondar y Standar ds
An y n ational secondary ambien t air quality standard ... shall specify
a l evel of air quality the attainment and maintenance of w h ich i n the
judgment of the Administrator, b ased on su ch cri teri a, i s req u i si te to
p r o t ect th e public welfare f rom any known or anticipated adver s e
ef f ects a ssoci ated w i t h p resen ce of su ch ai r p ol l u tan t i n th e a mb i en t
air. [ C AA §109(b)(2)]

¹¹ U. S. Congress, House, Committee on Interstate and Foreign Commerce, Cl e a n A i r Ac t
Amendments of 1977, House Rept. No. 95-294, to accompany H.R. 6161 (95^th Congress, 1^st
session) (Washington, D.C.: U.S . Govt. Print. Off., 1977), p. 182.

Al l l an gu age ref erri n g to ef f ects o n w el f a re i n cl u d es, b u t i s n o t l i mi ted
to, ef f ects o n s oi l s , w ater, crop s , v egetati o n , ma n -ma d e ma teri al s,
animals, w ildlife, w eather, visibility, and climate, damage to and
deterioratio n o f p roperty, and h azards to transportation, a s w ell a s
ef f ects o n econ o mi c val u es a n d on personal comf ort an d w e ll-being,
w h eth er cau sed b y tran s f o rmati o n , con v ersi o n , or comb i n ati o n w i th
other air pollutants. [ C AA §302(h)]
Secondary NAAQS define the concentratio n of an air pollutant in the ambient
air n ecessary to protect the “public welfar e.” S econdary standards are implemented
in the s ame m anner as p r i m a r y NAAQS , with the k ey difference t hat t here is no
federally enforceable s p ecified deadline for attainment. M ost s econdary NAAQS
have been set at t he same level as t he primary NAAQS , but they can be set at l evels
more or less stringent i f j ustified b y available evidence.
SettingNAAQS
Setting and Review i ng NAAQ S
The p rocess for setting a NAAQS is a m ultistage one, and repeats regularly as
the C AA requires each NAAQS to be reviewed every 5 years u sing the s ame p rocess,
to ensure that each NAAQS is based o n t he most recent s cientific information. The
C AA i s qui t e speci fi c o n cert ai n st eps o f t he process: i n part i cul ar, o n t he preparat i o n
of a “criteria document” summarizing the s ci entific information, on the review of t hat
document by an i ndependent scientific committee, on the criteria t o be used by t he
Administrator in deciding on the final standard, and on the p rocedural p r o c e s s for
promulgating t he standard. In addition, EP A has administrativel y added a key s tep,
the p reparation o f a “s taff paper” that su mmariz es t he criteria document and lays out
policy options; a n d E x e cutive Order 12866 requires a Regu latory Im pact Analys is
(RIA), although t he e c o nomic analys is it contains is legally irrelevant to the actual
deci s i o n o n the s tandard. Finally, t here are a number o f regulatory assessment
requirements i n l aw that i m pinge on t he proces s — but have limited s ubstantive
i m p act on t h e d eci si on i t s el f. These s t ages are di scussed b el ow (see al so Appendi x
I).
Criteria Document.
T h e Administrator shall i ssue air quality criteria f or [each] air
pollutant ... included ... [on the] list .... [ C AA §108(a)(2)]
The “criteria document” precedes the NAAQS both i n its appearance in air
pollution control l aw and i n t he process o f s etting NAAQS. In t he early stages of the
evol ut i o n o f Federal ai r pol l u t i o n cont rol l aw, t he Federal role focused primarily on
research and o n p roviding financial and t echni ca l a d v i c e t o st at es. T hi s rol e was
exemplified i n t he requirement that the P ub lic Health Service (which was responsible
for Federal ai r pollution activities before EPA) prepare a “criteria document.”
“Air quality criteria [docum en ts] are an expression of the sci en tific
knowledge of the r elationship bet w een var i o u s con cen t r at i o n s of

pollutants i n t he air and thei r adverse effect s on m an, anim als,
veget ation, m aterials, vi sibility, and so on.
“ A i r quality criteria can and s hould be u sed i n developing a i r
quality standards. C riteria and standards are n ot synonym ous. A ir
quality criteria are descriptive; that is, t hey describe t he effect s t hat can
be expect ed t o occu r w h en ever an d w h erever t h e a m b i en t ai r l evel of a
pollu tan t reach es o r exceeds a specific figu r e for a s pecific tim e p eriod.”
[Dr. Middleton, Director, National C enter for Air P ollution C ontrol, Public
Health Service, quoted in Senate Committee on P ublic Works R eport 403
on the Air Quality Act of 1967, p. 26.]
Ai r q u a l i t y cri teri a f or an ai r p ol l u tan t sh al l a ccu ratel y ref l ect th e
latest scientific knowled ge usef ul in indicating th e k ind and extent of
all identifiable effects on public health o r w e lfare w hich may b e
exp ected f rom th e p resen ce of su ch p o l l u tan t i n th e a mb i en t a i r , i n
varying quantities. [ C AA, §108(a)(2)]
In 1970, when the C lean Air Amendments of 1970 established t he federal role
in setting NAAQS, t he “criteria document” became t he basic t echnical underpinning
of the s tandards-setting process. Ex plicitly, t he preparation, review, and use of t he
“criteria document” was t o b e o b j ective and scientifically validat ed. It i s
scientifically peer reviewed by an advisory committee — es tablished by s tatute in the
1977 amendments — and by other federal departments and agencies. M uch o f t he
review process i s open t o t he public, and th e final criteria document i s m ade public:
The i ssuance of air quali t y c r i t eri a ... sh al l b e a n n o u n ced i n th e
Federal R eg i s ter and copies shall b e made available to the general
public. [ C AA, §108(d)]
The cri t eri a docum ent i s p repared i n t he Offi ce of R esearch and D evel opm ent
by EPA s ci entists (with the advice of and review by a scientific advisory committee,
des c ri b e d bel ow). Reviewing t he scientific literature for all studies relevant to the
air pollutant, t he preparers consolidate i nformation p ertinent t o i ndicating t he kinds
and magnitudes of effects resulting from the pollutant’s presence in ambient air,
assess the robustness o f t he studies, endeavor to resolve i nconsistencies, and evaluate
findings. Key components o f t he scientific evidence include epidemiologi cal studies
that ex amine t he relationships between ambient pollutant levels and public health and
welfare; clinical studies that examine human responses to controlled levels of
pollutants, for ex ample in ai r chambers; and animal studies. Typically, t he studies
included h ave undergone peer review and b een published i n t he open literature, but
on occasion s ome o ther studies, s uch as preliminary reports on ongoing research, m ay
be included i f t hey m eet other s tandards o f s ci en tific reporting.
Criteria documents are m ajor undertakings : t he 1996 oz one criteria document
contains over 1,500 pages and evaluates nearly 190 scientific studies; t he 1996 PM
criteria document contains 2,400 pages a n d e v aluates s ome 8 0 s tudies. Each
document t ook about one year to draft; r e v i e w a n d revisions took another year to
com e t o cl osure.

StaffPaper.
This [ s taf f paper] assessment i s i nt endedtohelpbridgethegapbetween
t h e s ci ent i fi c revi ew cont ai ned i n t he [ criteria document] and t he
judgments required of t he Administrat or in setting ambient standards for
P M . Thus, em phasi s i s p l aced o n i d ent i fyi ng t hose concl u si ons and
uncertainties i n t he available s ci entific literature that the s taff believes
should b e considered in selecting p articulate pollutant in d i c ators, forms,
averaging times, and l evels for the p ri m a r y (health) and secondary
(welfare) s tandards. [ E P A, Particulate Matter Staff Paper (1996), p . I-1]
Based on t he criteria document, EPA s ci entists and policy ex perts prepare a staff
paper. It is developed i n t he Office of Air Quality Planning and S tandards o f t he
Office of Air and Radiation. This document i s not re q u i red b y t he CAA; it is an
administrative s tep des igned t o facilitate the EPA Administrat or’s deci sion. It lays
out options for a NAAQS standard — e.g., whether t o s et a s tandard, at what l evel(s)
it might be set, and m ethods for m easuring compliance — along with justifications
from t he criteria document. Li ke the criteri a document, the s taff paper i s reviewed
by the s ci entific advisory committee.
Clean Air S cientific Advisory Committee.
The Administrator shall appoint an independent [cl ean air]
sci en ti f i c revi ew commi ttee [CAS AC] comp osed of seven memb ers
including at least one memb er of th e National Academy of Sci ences,
on e p h y si ci an , a n d on e p erson rep resen ti n g S tate a i r p o l l u ti on con trol
agencies.
... [T]he [cl ean air scientific advi sory] commi ttee ... shall ... review
... th e criteria published under section 108 and the national primary
and secondary ambien t air quality standard s p romu lgated under this
secti o n a n d sh al l recommen d to th e Ad mi n i s trator an y n ew n a ti on al
ambien t air quality standard s and rev i sions of existing criteria and
standards as may be appropriate ....
Such committee shall also (i) advise the Ad ministrator of areas in
which additional knowled ge is required to appraise th e adequacy and
basis of existing, new , or revised national a mb i e n t air quality
stan d a rd s, (i i ) d escri b e th e research ef f o rts n ecessary to p rovi d e th e
required information, (iii) a d vise the Administrator on the rel ative
con tri b u ti o n to a i r p o l l u ti on con cen trati o n s of n a tu ral a s w el l a s
anth ropogenic activity , a n d (iv) advise the Ad ministrator of any
ad verse p u b l i c h eal th , w el f a re, soci al , econ o mi c, o r en ergy ef f ects
w h ich may resu l t f rom various strategies f o r a ttainment and
mainten a nce of such n ational amb ient air quality standard s. [ C AA,
§109(d)(2)]
T h e 1 967 Air Quality Act required t he Secret ary of HEW to consult w i t h
“appropriate advisory committees ” (al ong with Federal d epartments and agenci es )
when preparing criteria documents. The Secret ary established a National Air Quality
Criteria Advisory Committee, havi n g a m em bership broadly repres entative of
industry, universities, conservation i nterests, and all l ev el s of government. This
committee actively participated in the rew riting of the Sulfur Ox ides Criteria

Document published i n 1969, and i n t he preparation o f s u b s equent criteria
documents. T his general requirement fo r consultation was replaced in 1977 by the
specific requirements creating C ASAC , w i t h its responsibilities for reviewing t he
scientific basis o f t he Administrator’s decisions on NAAQS . The legi slative h istory
concerning the C ASAC provision emphasiz es its independence:
This com m ittee i s i nten ded t o assist the A dm inistrator, but it is also
intended to have com plete independen ce. This independen ce will help
provide an outside m echanism for eva luating w hether any pollu tant m ay
reasonably be anticipated to endanger public health or en vi ronm en t, for
evaluating t he scientific and m edical data w hich bear on this question,
and for reviewing gaps in the ava ilable data and recom m ending
additional n eeds for research . [Comm ittee on Interstat e and Foreign
Commerce, House R ept. No. 95-294 (1977), p . 182]
The s even-member C AS AC creates a p anel to review each NAAQS . T his p anel
consists of the m embers of CASAC plus cons ultant members t o assure full coverage
of the expertises needed to assess fully the issues involved. For the ozone review,
a p anel of 15 was convened; for t he PM review, t he panel consisted o f 21.
The p anel m em b ers m eet t o revi ew each cri t eri a docum ent and st aff p aper as i t
i s prepared, recom m end i m p rovem ent s, and aft er furt her m eet i n gs and revi ews si gn
off only when t hey are convinced that each accurately reflects t he status of the¹²
sci ence. C A S A C p anel m eet i n gs are open t o t he publ i c. T h ese “cl osure
documents” b ecome part of the record for rulemaking. Thus —
Th e i n d epen den t , s ci en t i fi c r evi ew com m i t t ee’s r ecom m en dat i o n s
on these i ssues will not only aid the A dm inistrator and t he Congress, but
also the courts i n j u d i c ial revi ew of any n ational am bient air quality
standard or of t he Adm i nistrator’s failure or refusal t o s et or revise such
a standard w ith respect to a n y pollutant. [Committee on Interstat e and
Fo reign C ommerce, House R ept. No. 95-294 (1977), p . 182]
What the closure letter m eans i s t hat t he CASAC panel members agree that the
criteria docum e n t a n d the s taff paper p r ovide an adequate scientific basis for
regulatory deci sionmaking. Using formulaic sentences , a cl osure l etter for a criteria
docum ent t yp i cal l y reads:
At the S eptember 1995 meeting t he Panel came t o closure on the
C riteria Document. It was t he consen sus o f t he Panel m ember s t h a t t h e
Criteria Document p rovides an adequate review of the available s ci entific
data a n d relevant s tudies of oz one an d related photochemical ox idants.
The document i s q u i t e comprehens ive and will provide an adequate
scientific basis for r e gulatory d ecisions on oz one and related
photochemical ox idants based o n available i nformation. [ C ASAC Closure
on the Air Quality Criteria for Ozone and R el at ed Photochemical

¹²CASAC i s s ubj ect to the Federal Ad vi s o r y Committee Act (5 U.S.C. App. 2) which
governs public accessibility to committee meetings and products.

Ox idants, l etter, from Dr. George T. W o lff, Chair, to Honorable Carol M .
Browner, Administrator, EPA (November 28, 1995)]
A closure letter for a s taff paper t ypically says :
It was t he consensus o f t he Panel t hat although our underst a n d i ng of the
health effects o f o z one is far from complete, the document p rovides an
adequate scienti f ic basis for making regu latory decisions regarding a
primary oz one standard. [CASAC Closure on the Primary Standard Portion
of the S taff Paper for Oz one, l etter, from D. George T . W olff , C h a ir, t o
Honorable Carol M . Browner, Administrator, EPA (November 30, 1995)]
The Admi ni str a tor’s Decision
... [I]n the judgment of the Administrator .... [ C AA, §109(b)(1)]
The C AA speci fies that the Administrat or shal l use her “judgment ..., base d
on [th e] cri teri a [docu ment] and allowing an adequate margin of safety” to
determine t he NAAQS “requisite to protect th e public health ” [ CAA, §109(b)(1)] .
Thus the d ecision i nvolves weigh ing o f t he scientific evidence, collected and
anal yz ed in the criteria document, of the policy options laid out in the s taff paper, and
of the comments of C ASAC — following the criteria dictated i n t he CAA.
EP A’s i nt erpret at i o n o f t he cri t eri a h as been t h e s ubj ect of vari ous chal l enges.
One key issue arose from litigation t hat act ually concerned another part of t he CAA;
the d ispute l ed to amendments in 1977 that clarified p rovisions relating t o NAAQS.
This landmark cas e concerning the appropri a te criteria for ai r quality regulatory
judgments i s Ethyl C orp. v. EPA. A 3-judge p anel, voting 2 -1, i nvalidated the
Administrat or’s regulation of l ead in fuel s bas ed on § 211 (regulation of fuels) of t he
CAA [ No. 73-2205 ( D .C. C ir. J an. 28, 1975)] . The d ecision hinged on the burden
of proof the Administrator had t o m eet in order t o j ustify t he proposed standard, with
the m aj ority holding that the Administrat or had t o s how actual h arm rather t han t he
threat or risk of harm. S u b s e q u e n tly, t he initial d ecision was v acated and t he U.S.
Court o f Appeals granted a rehearing en b a n c ; t h e full court upheld the
Admini s t r a t or’s regu lations on a 5 -4 vote, holding that §211 states a m erely
precaut i onary st andard — i .e., does not requi re act ual h arm .
Even though v acated, t he initial Ethyl deci si on rai s ed i ssues affect i n g al l cl ean
air s tan d a r d s s etting t hat were d irectly addressed i n t he 1977 Amendments to the¹³
CAA. In its report on t he bill am ending the C AA, the House C ommittee on
Interstate and Forei gn Commerce i dentified s ix issues raised by the cas e t hat
“required further congre ssional clarification”:

¹³ T his discussion, including the quotations, are from U.S. Congress, House, Committee on
Interstate and Foreign Commerce, Clean Air Act Amendments of 1977, House Rept. No. 95-^th^st

294, to accompany H.R. 6161 (95 Congress, 1 session) (Washington, D.C.: U.S. Govt.

Print. Off., 1977), pp. 43-51.

(1) w het h er t h e A dm i n i s t rat or coul d “act t o prevent h arm b efore i t o c c u rs or
should h e b e authoriz ed to regu late an air pollutant only i f h e finds actual h arm h as
al ready o ccurred”;
(2) whether the Administrator could assess risks o r only “make findings o f p ast
fact”;
(3) whether the Administrat or could consider the cumulative risk of a pollutant
from m ultiple sources, or only the risk from the single class of sources being
regulated;
(4) whether different standards o f p roof apply t o d ifferent sources;
(5) whether “protection of public health” referred only t o healthy normal adults
or included “susceptible individuals within the ex posed population”; and
(6) whether a “Court’s s tandard of r eview of inform a l r u l e m aking by the
Adm i nistrator t o protect public health be whet her h e h as relied on concl usive or
indisputable facts or whether h e h as reached reasonable conclusions which are
rationally justified.”
To resolve t hese issues , t he Committee added l anguage amending the act , using
“a standardized basis for future rulem aking to protect the public h e a l t h : t he
Adm i nistrator m ay regulate a pollutant, em i ssions of ‘which i n h is judgm en t cause
or contribute to air pollution w hich m ay r easonably b e a n t i c i p a t ed to endanger
public health or welfare.’ Thi s s am e basic form ula” w as used i n t h e s everal
standards-setting p rovisions of the Act, i ncluding sections 108 (criteria for NAAQS),
111 (new source performance standards), 112 (haz ardous air pollutants), 202 (motor
vehicl e emission standards), 211 (regulation of fuels and fuel additives ), an d 231
(aircraft emission s ) , a s w el l as s ubtitle B of Title I (ozone and s trat ospheric
protection).
In using t his l anguage i n amending the act , as discussed i n t he report on t he bill,
the C ommittee i ntended —
(1) T o em p h a si ze t h e p reven t a t i ve o r p recau t i o n a ry n a t u re of t h e a ct , i .e., t o
assu re th at regu latory action can effectively p reven t h arm b efore i t o ccu rs ...;
( 2 ) To authorize the A dm inistrator to weigh risks and m ake reaso n a b l e
projections of future trends ...;
(3 ) To assure consideration of t he cu m u lative i m pact of all source s o f a
pollutant in setting am bient and em i ssion standards, n ot just the exten t of t he risk
from the em i ssions from a single source or class of s ources of the pollutant; ...
(4) To provide the s am e s tandard of proof for r egulation of any air pollutant
...;
(5) To assure that the h ealth of susceptible indivi duals, as well as healthy
adults, w ill be en com passed i n t he term ‘public health,’ regardless of t he section
of t h e a ct u n d er w h i ch t h e A d m i n i st rat o r p roceeds; an d
(6) To reflect awaren ess of the uncertainties and lim itations in the data w hich
w i l l b e available t o t he Adm i nistrator i n t he foreseeable future t o enable h im t o
execu te his rulem aking duties under this act, becaus e o f t he lim itations on
research resources and t he fact that d ecisionm aking about t he risks t o public
health from air pollution falls on ‘the frontier s o f scientific and m edical
knowledge.’
In short, “the com m ittee l a n g u age is intended to em phasize t he necessarily
judgm en tal elem ent in the t ask of predicting future h ealth risks of present action

and t o confer u pon the A dm inistrator the requisite a u thority to exerci se such
judgm en t.”
Thi s i s not , however, a b l a n k c h eck: “... the com m ittee does n ot intend this
language as a licen se for ‘crystal ball’ specu lation. The A dm inistrator’s judgm en t
m u st, o f course, r em ain s ubject to rest raints of reasoned decisionm aking.”
TheEvidence.
...[B]ased on such criteria [document] ... [ C AA, §109(b)(1)]
Ai r q u a l i ty cri t e r i a f or an ai r p ol l u tan t sh al l a ccu ratel y ref l ect th e
latest scientific knowled ge usef ul in indicating th e k ind and extent of
all identifiable effects on public health or welfare w h ich may b e
e x p e c t e d f rom th e p resen ce of su ch p o l l u tan t i n th e a mb i en t ai r, i n
varying quantiti e s . T h e criteria f or an air p ollutant, to the extent
practicable, shall include information on —
(A) t h ose variable f actors (including atmospheric conditions)
w h ich o f thems elves o r in combination w i th other f actors ma y a lter the
effects on public health or welfare of such air pollutant;
(B) the types of air pollutants w hich , w hen p re sent in th e
atmo sp h ere, ma y i n teract w i th su ch p o l l u tan t to p rod u ce a n a d v erse
ef f ect on public health or w elf are; and
(C) a n y k n o w n or an ti ci p a ted a d v erse ef f ects o n w el f a re.
[ C AA, §108(a)(2)]
Each agency shal l b ase i t s [ regul at ory] d e ci si ons on t h e b est reasonabl y
o b t a i nabl e s ci ent i fi c, t echni cal , economic, and other i nformatio n
concerning the n eed for, an d c o n sequences of, t he intended regulation.
[ Ex ecutive Order 12866]
The evidence for a NAAQS decision ultimately resides i n t he criteria document,
the s taff paper, and t he CASAC l etters of cl osure, plus material s s ubmitted during t he
public comment period. Ancillary information i s i ncluded i n t he Regulatory Im pact
Analys is (discussed b elow). Summaries ar e i ncluded i n t he proposed and final rules
as published i n t he Federal R egister. In se tting NAAQS, t he generic d irection o f E.O.
12866 to r e gu l a t o ry agencies to base their “decisions on the b est reasonably
obtainable s cientific, t echnical, economic, and other i nformation....” is constrained
by the C AA, which s peci fies that health protection i s t o be t he criterion in
determining t he standard.
TheCriteria.
Health.
Health is the s ole criterion for s etting t he primary NAAQS. (Fo r a discussion of
the i ssue of considering costs, s ee below.)

Beside the t echni c a l information underlyi ng a h ealth standard,¹⁴ the
Administrat or must consider several cruci al policy i ssues : These incl ude (1) defining
whose health the s tandard is to protect , given that some people wi l l b e more
susceptible to pollution t han others; (2) defining which health effect s of pollution are
adverse h eal t h i m p act s t o b e p rot ect ed agai nst ; (3) eval u at i n g ex posure and
characteriz ing risk; and (4) considering what factors t o t ake i nto account in providing
“an ad equate margin of safety.” The statute and l egislative history provide some
gu idance o n t h ese questions. As t hey are key elements o f t he judgment that the
A d ministrat or makes i n deci ding the s tandard, t hese questions are m ore o r l e s s
ex plicitly posed in the s taff paper.
Sensitive populations. Fo r answering the first question, about whose h ealth
a primary standard should protect, the Senate Report on the Clean Air Act
Amendments of 1970 provided ex p licit gu idance:
In requiring t h at n ational am bient air quality standards be
est abl i s h ed a t a l evel n ecessary t o prot ect t h e h eal t h of person s , t h e
Com m ittee r ecognizes that such stan dards will not n ecessarily provide
for t he quality of air required t o protect those i ndividuals who are
ot h e r w ise dependent on a controlled i nternal environm ent such as
patients in inten sive care units or newborn infants i n nurseries.
However, t he Com m ittee em phasizes that i ncluded am ong those persons
whose h ealth should be protect ed b y the am bient standard are
particularly sensitive citizen s such as bronchial asthm atics and
em physem atics w ho in the n orm al course of daily activity are exposed to
the am bient environm ent. In establishing an am bient s t a ndard
n ecessary t o prot ect t h e h eal t h of t h ese p erson s , r eferen ce w o u l d b e
m ade to a representative sam ple of p er s o n s com prising the sen sitive
group r ather t han t o a s ingle person i n s uch a group.
Am bien t air quality is suffici en t t o protect the h ealth of such
person s w h en ever t h ere i s a n absen ce of adverse effect on t h e h eal t h of
a statistically related sam ple of persons in sensitive groups from
exposure t o am bient air. [Senate C ommittee on P ublic W o rk s , Report
No. 91-1196 (1970), p . 10]
Also, as discussed earlier, the C ommittee on Interstat e and Foreign C ommerce,
in its report l anguage discussing am endments relating t o t he basis for administrative
standards, ex pressly noted its i n t ent to “assure that the health of susceptible

¹⁴ Cont r ove r s y a bout t he s ci ence under l yi ng t he 1997 NAAQS st andar d s c ont r i but ed t o l a t e r
legi slation. A key concern f ocused on the availability of raw data t o assess the scientific
robustness of a study of health effects of air pollution i n s everal cities and of the conclusions
drawn. T his led t o a provision added t o P.L. 105-277 regarding public availability of data;
see Eric A. Fischer and Genevieve J . Kn e zo , P u b l i c access t o data f rom f ederally funded
research: OMB Circular A-110 and issues for Congress (November 18, 1999), CRS Report
RL30376, pp. 3-4. Relatedly, a provi sion was i nc l u d e d i n P.L. 106-554 to ensure and
maximi ze the i ntegrity of information used by agencies, and t o r equire agencies to provide
“mechanisms allowing affected persons to seek and obtain correction of i nformation ...”; s ee^th
J ohn E. Blodgett, Environmental Reauthorizations and Regulatory Reform: From the 104
Congress t hrough the 106^th (December 27, 200), CRS Report 96-949 ENR, p. 6.

individuals, a s w el l as healthy adults, will be encompassed i n t he term ‘public
health,’ regardless of the s ection o f t he act under which the Administrator proceeds.”
Adverse health effect. The C lean Air Act does not define the adverse h ealth
effects against whi c h NAAQS must protect, nor does t he legi slative h istory of the
CAA contai n m uch discussion of what constitutes a n a d verse health effect . The
National C ommission on Air Quality¹⁵ noted proposals t hat t he Act contain greater
specificity, but observed:
any attempt to identify specific types of health effects could i nhibit
identification o f unanticipated effect s t h a t s hould b e considered. In
addition, most ai r pollutants affect public health in more than one way. To
carry out its responsibilities under t h e A c t, EPA, before making a final
deci sion to set or revise an air quality standard, m ust have all relevant and
reliable s ci entific information on t he full range of possible health effect s
of a pollutant.¹⁶
Indications of what constitutes an adverse h ealth effect can be found in the
distinction b etween primary and s econda ry s t andards and in the definition of
haz ardous air pollutants. W h il e a p r i m ary NAAQS protects h ealth, a secondary
NAAQS protects welfare, which includes, by definition, “personal comfort and well-
being. ” Thus, pollution effects t hat i nvolve only “personal comfort and well-being”
are not health effect s t hat p rimary standards are direct ed toward. The definition o f
a h az ardous air pollutant in the 1970 Act i s “an air pollutant ... which ... may cause,
or contribute t o, an increas e i n m ortality or an i n creas e i n s erious irreversible, or
incapacitating reversible, illness.” Unque stionably, “increase i n m ortality,”
“irreversible” and “incapacitating” signal health effects. Are bouts o f c o u ghing or
decreased lung functioning adverse h ealth effect s ? If they are i rreversible or
i n capaci t at i n g, t h e answer w oul d s eem cl earl y “yes.” As a p ract i cal m a t t e r , i f an
effect necessitates m edical intervention or l eads t o curtailed activity, such as missing
school or work, t he answer would s eem “yes.” The question o f whether an effect is
an adverse h eal t h effect can be di ffi cul t t o answer, p art i cul arl y i f t h e effect appears
minor but may cumulat e over a lifetime.
Exposure e valuatio n and risk assessment.
In order t o em p h a si ze t h e p recau t i o n a ry or preven t i ve p u r p o s e o f
the act (and, therefore, the Adm inistra t o r ’ s duty t o assess risks r ather
than wait for proof of actual harm ), the com m ittee n ot only retained the
concept of endangerm ent t o h ealth; t he com m ittee also added t he words
‘m ay reasonably b e a nticipated’. ...
By its use of t h e w ords ‘cause or contribute t o air pollution’, t he
com m ittee i nten ds to require t he Adm i nistrator t o cons i d e r a l l sources
of the contam inant which contribute to air pollution and to consider all

¹⁵The National Commi ssion on Ai r Quality was created by the Clean Ai r Act Amendments
of 1977, §323, to revi ew t h e Act and t o r eport t o t he Congress the e ffects of t he Act,
alternative ways of controlling air pollution, and any recommended midcourse corrections.
¹⁶National Commi ssion on Ai r Quality, To Br eathe Clean Ai r (Washington, D.C.: 1981), p.

3.1-2.

sources of ex p osure to the contam i nant — food, water, air, etc. — i n
determ ining h ealth risk.
Finally, t he term ‘in t he judgm en t of t he Adm i nistrator’ is intended
to m odify both t he ‘cause and contribute t o’ phrase and t he ‘reasonably
m ay be anticipated’ phrase. [Committee on In t ers t at e and Foreign
Congress, Clean Air Act A m e n d m e nts of 1977, R eport 95-294 (1970), p .

51.]

As indicated in the p revious discussions of the Ethyl and Lead Industries v. EPA
cases, and as emphasized in House R eport 95-294, the Administrator may p romulgate
a NAAQS in “reasonable” anticipation o f pub lic health endangerment. The practical
effect of this criterion is to allow a “w ei ght of evidence” approach to be used in
setting a NAAQS¹⁷; E P A need not prove t h at adve rse h eal t h effect s h ave al ready
occurred, nor that the air-borne form of the pollutant solely caused t he harm.
Ma rg in o f sa fe ty. On the question of what c o nstitutes a “m argi n of s afet y,”
EPA’s S taff Paper on P articulat e Matter summarizes t he situation as follows:
The U.S. C ourt of Appeals for th e District of C olumbia C ircuit has
held that the requirement for an adequate margin of safety for p ri m ary
standards was intended t o addres s uncertainties a s s ociated with
inconclusive scientific and t echnical information available at t he time o f
standard setting. It was also i ntended t o provide a reas onable degree of
protection aga inst haz ards t hat res earch has not yet i dentified (Lead
Industries Association v. EPA, 647 F.2d 1130, 1154 (D.C. C ir. 1980), cert.
denied , 101 S. Ct. 621 (1980); American Petroleum Institute v. Costle, 665
F.2d 1176, 1177 (D.C. C i r . 1981), cert. denied, 102 S. Ct. 1737 (1982)).
Bo th kinds of uncertainties are components o f t he ri s k a s s o ciated with
pol l u t i o n at l evel s b el ow t hose at w hi ch hum an heal t h effect s can be sai d
to occur with reas onable s ci entific certainty. Thus, by s el ecting primary
standards t hat p rovide an adequate margin of safety, t he Administrat or is
seeking not only t o p revent pollution l evel s t hat h ave b een dem onst rat ed
to be harmful but also to prevent l ower pollutant levels that she finds may
pose an unaccept abl e ri s k o f h arm , even i f t h e ri s k i s not preci sel y
i d ent i f i ed as t o n at ure o r d egree.
In selecting a margin of safety, t he EPA considers s uch factors as t he
nature and s everity of the h ealth effects i nvolved, the s ize of t he sensitive
population(s) at risk, and the k ind and degree of the uncertainties t hat m ust
be addressed. Given t hat t he “m argi n of s afet y” requirement by definition
only comes into play where n o conclusive showing o f adverse ef f e c t s
ex ists, s uch factors which i n v o lv e unknown o r only p artially quantified

¹⁷In r e j e c t i n g obj e c t i ons t o t h e o zone NAAQS, t he D.C. Ci r c ui t Cour t obs e r ve d: “ One f i na l
a s pe c t of EPA’ s d i s c u s s i on of t h e p r i ma r y NAAQS l e ve l i s r e l e va n t h e r e : T h e A ge nc y’ s
response t o certain comments questioning its reliance on s pecific f ield, epidemi ological, and
clinical studies. According t o EPA, t he comments ‘ did not reflect an integr ative assessment
of the e vi dence – the a pproach CASAC has hist orically urged [ the Agency] to follow – but
r a t her a pi ecemeal l ook at each i ndi vi dual s t udy’ [ Ozone NAAQS, 6 2 Fed. R e g. 38,868] .
EPA t herefore dismissed t he comments, arguing t hat s uch an i ncremental critique ‘tends to
mi ss the s trengt h of t he entire body of evidence taken t ogether’ [Ibid]. American Trucking
Association, Inc. v. EPA, 2002 Westlaw 452092 (D.C. Cir. M ar. 26, 2002).

risks have t heir inherent limits as guides to action. The s el ect i o n o f any
particular approach to providing an ad equate margin of safety is a policy
choice left speci fically to the Administrat or's judgment (Lead Industries
Association v. EPA, s upra, 647 F.2d at 1161-62). [EPA, Particulate
Matter Staff Paper (1996), pp. II-1 - II-2]
EPA’s view, then, i s t hat t he “m argi n of s afet y” allows the Administrat or to take
into account the dimensions of the decision that scientific data cannot resolve. This
view has also b e e n v o i ced by the C ourts. For ex ample, concerning the o z one
NAAQS , t he D.C. Circuit C ourt o f Appeals observed, “... EP A m ust err on the s ide
of caution, just as it did h ere – setting t he NAAQS at whatever l e v e l i t d eems
necessary and s uffi ci ent t o p r o t ect t h e publ i c heal t h wi t h an adequat e m argi n of
safety, t aking i nto account both t he available evidence and the i nevitable s cientific
uncertainties.” American Trucking Association, Inc. v. EPA, 2002 Westlaw 452092 (D.C.
Cir. Mar. 26, 2002).
Costs.
The question o f whether costs o r t echnical feasibility should b e t aken into
account in setting NAAQS has b een an enduring d ebate. The National C ommission
on Air Quality ex plai ned t he princi ple as follows:
The s tatutory basis for setting national ambient ai r quality standards
does not take economic factors i nto account. In t he Act, Congres s
recognized that while the l evel s of air pollution at which public health is
affected generally do not vary amo n g d i fferent locations, t he costs o f
meet i n g a specific s tandard can vary substantially from area t o area,
depending upon the s everity of the pollution. Thus, i f a national air quality
standard were based i n p a r t o n t he co sts of complyi ng with it, the high
cost s o f m eet i n g t he st andard i n a few heavi l y pol l u t ed areas coul d resul t
in the s tandard’s being s et at a l es s protective l evel than is achievable i n a
reasonabl e, econom i c fashi o n i n o t h er areas. The health benefits of good
air quality and t he economic, s ocial, energy, a n d o t her costs of meeting
heal t h -based st andards can be bal anced m o re effect i v el y and appropri at el y
when cont rol p rogram s are est abl i s hed for part i cul ar areas t h an when¹⁸
national p rimary standards are set.
The concept of deferring cost considerations unt il the implementation of
standards, rather than incorporating t hem i n s tandards-setting, arose early in the
evolution o f t he CAA. The S enate R eport o n t he Air Quality Act o f 1967 affirmed
the p rimacy of health protection i n s etting NAAQS:
Considerations of technology and econom i c feasibility, w hile
i m port a n t i n h el p i n g t o d evel op al t ern a t i v e p l a n s an d s ch edu l es for
achieving goals of air quality, s hould n ot be used to m itigate against
protection of t he public health and w elfare. [Senate C ommittee on P ublic
W o rks, Report No. 403 (1967), pp. 28-29]

¹⁸National Commi ssion on Ai r Quality, To Br eathe Clean Ai r (Washington, D.C.: 1981), p.

3.1-2.

S ubsequent l y, i n b ri ngi n g t o t he fl oor t h e S enat e b i l l t h at becam e t he C l ean Ai r Act
Amendments of 1970, S e n a t o r M uskie repeated that language and called i t a
“warning” t h a t h a d been ignored by those who sought to compromise h ealth
st andards b ecause of cost s:
T h a t w arning ... has been on the books of this com m ittee f o r 3
years, for a ll to read.
Contrary to this intent, t hese [cost and feasibility] c onsiderations
have been used as argum ents t o com prom ise t he public health.
Therefore, the com m ittee h as m ade explicit in this bi l l w h a t is im plicit
to standards d e s i g n e d to protect our h ealth. That concept and that
ph ilosoph y a re beh i n d every page of th e p roposed legislation .
The first responsibility of Congress is not t he m aking of
t ech n o l ogi cal o r econ o m i c j u dgm en t s — o r even t o be l i m i t ed
by what is or appears t o be t echnologically f e a s i ble. Our
responsibility is to establish w hat t he publ i c interest requires
to protect the h ealth of persons.
[ S enat or Muski e, d ebat e o n t he Nat i onal Ai r Qual i t y S t andards Act
of 1970 (Congressional Record, September 21, 1970, pp. 32901-02)]
The question o f t aking costs into account in setting NAAQS was litigated in the
case o f t he lead NAAQS. The D.C. Circuit C ourt’s d ecision was unambiguous:
... [T]he statute and its legi s l at i v e history make cl ear that economic
considerations play no part in the promulgation of ambient ai r q u a lity
standards under S ection 109.
... Section 109(b) speaks only of protecting t he public health and
welfare. Nothing i n its language suggests t hat t h e A d m i nistrat or is to
consider economic or tech nological feasibility in setting ambient ai r
quality standards.
The l egislative history of the Act al so shows t he Administrat or may
not consider economic and t echnological feasibility in setting air quality
standards; the absence of any p rovisi on r e quiring consideration o f t hese
fact ors w as no acci dent ; i t w as t h e resul t o f a del i b erat e d eci si on by
Congress to subordinate such concerns to the achievement of health go als.
[ Lead Industries Association v. Environmental Protection Agency, 647
F.2d 1130 (D.C. C ir. 1980)]
The i ssue o f costs recurred following the 1997 oz one a n d P M^2.5 NAAQS .
Among the m any comments s ubmitted o n t he proposed rules were objections that the
EPA h ad not considered costs. EPA res ponded t o t hese comments at s ome l engt h,¹⁹
argu ing t hat costs should not be considered in setting t he NAAQS. EPA’s position
was challenged in court, and again the D.C. C ircuit Court was ex plicit:

¹⁹ See, for e xample, EPA, “ Cost Considerations,” National Ambient Air Quality Standards
for Ozone; Final Rule,62Federal Register 38878-38883 (J uly 18, 1997).

As this court l ong ago m ade clear, i n s etting NAAQS under §109(b)
of th e C l ean A i r Act , t he EPA i s not permitted t o consider the cost of
implem e n t i n g t hose s tandards. [ American Trucking Associations v. U.S.
E.P.A., 175 F.3d 1027 (D.C. C ir. 1999)]
This holding was among those appealed to the S upreme C ourt, w h i ch
unan i m o u s ly upheld the D.C. C ircuit Court’s ruling on t his point. The Supreme
Courtstated:
Secti o n 1 09(b) does not permit the Administrator to consider
implementation costs in setting NAAQS. [ Whitman v. Ameri can T r ucki ng
Associations, Inc. 531 U.S. 457 (2001).
Promulgating NAAQS
The p rocedural steps for promulgatin g o r revising NAAQS are s et forth i n t he²⁰
CAA itsel f — not, as i s usual el sewhere, in the Administrative P roced u r e A ct .
Affect i n g t hi s p rocess t o v aryi ng degrees are s ever a l o t h er s t at u t o ry requi rem ent s
affecting regulations, notably the R egul atory Flex i bility Act, the Unfunded Mandates
R eform Act , and t he S m al l Bus i n e s s R egul at ory E nforcem ent Fai rness A ct . A l s o
affecting t he process are ex ecutive m a n dates, notably Ex ecutive Order 12866,
Regu latory Planning and R eview, and E x ecutive Order 12848, Federal Actions to
Address Environmental J ustice i n M inority Populations and Low-Income
Populations.
The Administrative Rulemaking Process
Fo r n u m e r o u s CAA rulemakings , i ncluding the p rocess for promulgating o r
revising CAA rules, §307 of the C AA modifies the p rocedures of the Administrative
Procedure Act governing rulemaking. Th e p rocess i s codified at 42 U.S.C. 7607(d).
In general, the p rocedures require EPA t o establ i s h a “docket” that contains all t he
crucial elements of t he rulemaking, t hat i s open t o public inspection, a n d t hat
represents all t he inform a t i o n a vailabl e for the Administrat or’s deci sion. (An
additional p rovision allows evidence not in the docket t o be considered in certain
c ases when t he omission was “reasonable.”) The k ey steps o f t he rulemak i n g
include: (1) notice o f t he proposed rulemaking, (2) a p eriod ava i l a b l e for public
comment, (3) promulgation o f t he rule, which shall i nclude responses to sign ificant
comments o n t he proposal, and (4) an opportunity for j udicial review and challenges
to the procedural determinations.

²⁰T he Clean Air Act Amendments of 1977, adding § 307 to the CAA: the r easons for t he
changes from t he Admi nistrative Procedures Act and their i ntent are discussed at l engt h i n
U.S. Con gr e s s, House, Committee on Interstate and Foreign Commerce, Cl e a n A i r Ac t^th^st
Amendments of 1977, House Rept. No. 95-294, to accompany H.R. 6161 (95 Congress, 1
session) (Washington, D.C.: U.S. Govt. Print. Off., 1977), pp. 318-325.

Noti ce of Proposed Rulemaking.
... [T]he Administrator shall publish, simultaneously w i th the i ssuance
of su c h cri teri a [d o cu men t] a n d i n f o rmati o n , p rop osed n a ti on al
primary and secondary ambient air quality s tandards .... [ C AA,
§109(a)(2)]
... [N]otice of proposed rulemaking shall b e p u b l i s h e d i n the
Fed eral Regi s ter, ... sh al l b e a ccomp an i ed b y a s tatemen t of i ts b asi s
and purpose and shall specify th e p eriod available f or public comment
.... The n otice o f p roposed r ulemak ing s hall also state the docket
number, the l ocation ... of the d ocket, and the times it w ill be open to
public inspection. [ C AA §307(d)(3)]
As an illustration o f t he process, on J une 12, 1996 EPA published an “Advance
Not i ce o f P roposed R u l em aki ng (ANP R ) for N ational Ambient Air Quality Standards
for Oz one and P art i cul at e M at t er” (61 Federal Register 29719-2925). This ANPR
outlined the bas is for EP A h a v i ng to make the deci sion, identified t he key
d o c u m ents, and indicated the m ajor op tions under consideration. The P roposed
Rules o n o z one and P M were released November 27, 1996, and published December

13 (61 Federal Register 6 5638-65872). Each laid out the p roposed decision,

requested public comment generally and o n specific options, t old where and how to
access t he docket, and p rovided for a 60-day public comment period (later ex tended

21 days , until March 12, 1997).

PublicComment.
... [A ] f t e r a reason a b l e ti me f or i n terested p erson s to s u b mi t w ri tten
commen ts th ereon .... [ C AA, §109(a)(1)(B)]
In p romu l gatin g a [NAAQS ], ... (i) th e Ad mi n i strator s h a ll allow
an y p erson to su b mi t w ritten commen ts, d a ta, o r d ocu men tary
i n f o rmati o n ; (i i ) th e Ad mi n i s trator sh al l g i v e i n terested p erson s a n
opportunity f or the oral presentation of data, view s , or arguments ....
[ C AA, §307(d)(5)]
In the case o f t he oz one a n d p a r t i c ulate m atter NAAQS proposed by EPA i n
December 1996, the Agency received over 25,000 comments duri n g t h e public
comment period. Also, EPA held 4 public hearings .
Promulgation of the Rule.
...[A]fter a reasonable time f or interes ted p ersons to submit written
commen ts th ereon (b u t n o l a ter th a n 9 0 d ays af ter th e i n i ti a l
publication of such p roposed standard s) [th e Administrator] shall by
regulation promulgate such p roposed n ational amb ient air quality
standard s w ith such modifications as he deems ap p r o p riate. [ C AA,
§109(a)(1)(B)]
(A) T he promulgated rule shall b e accomp anied b y (i) a s tatement of
basi s a n d purpose ... and (ii) an explanation of the reasons for any

ma j o r c h a n g es i n th e p romu l gated ru l e f rom th e p rop o sed ru l e. (B )
T h e p romu l gated ru l e sh al l a l s o b e a ccomp an i ed b y a resp o n s e to each
of th e significant comments , criticisms, and n ew data submitted in
w ri tte n o r o ral p resen tati on s d u ri n g th e commen t p eri od . [ C AA,
§307(d)(6)]
The o z one and p art i cul at e m at t er fi n al deci si o n s w e r e si gn ed by t h e
Administrator on J uly 16, 1997 and published on J uly 18, 1997 (62 Federal Register
38652-38896). Each final rule contained l engt hy discussions of issues raised by
commentat ors and the EPA’s final disposition of t hem.
Re gulatory Impact Assessments .
Costs a nd Be nefits — E xecutive Order 12866.
Each agency shal l assess bot h t he c o s t s a n d t h e b enefi t s of t h e i nt ended
regulation....
Fo r ... a s ignifica n t r e gulatory action ... the agency s hall ... provide ... (i)
An assessment, including the underlyi ng analys is, o f b enefits anticipated
from t he regulatory action (such as , but not limited t o, the promotion of t he
efficient functioning of the economy and private m arkets, t he enhancement
of health and s afety, the p rotection o f t he natural environment, and t he
elimination or reduction of discrimination or bias) together with, t o t h e
ex tent feas ible, a quantification of t hose benefits; (ii) An assessment,
including the underlyi ng analys is, o f costs anticipated from t he regu latory
action (such as , but not limited t o, the direct cost both t o t he government
in administering the regulation and to businesses and others in complying
with the regulation, and any adverse effects o n t he efficient functioning of
the econ o m y, private m arkets (including productivity, employm ent, and
competitiveness), health, s afet y, and t he natural environment, together
with, t o t he ex tent feas ib l e , a quantification of t hose costs; and (iii) An
assessment, including the underlyi ng analys is, o f costs and b e n e f i t s of
potentially effective an d reas onably feas ible alternatives to the planned
regulation....
[ Ex ecutive Order 12866, 58 FR 51735 (4 October 1993)]
EP A h as concluded t hat NAAQS reviews are “significant” regu latory actions²¹
requiring preparation o f a R egu latory Im pact Analys is (R IA). However, EP A also
ex plicitly states that “Because judicial decisions make clear that co s t c a n not be
considered in setting NAAQS, t he results of the d raft RIA h ave not been considered
in developing this proposal” [ National Amb i e n t A i r Quality Standard for Ozone:
Proposed Decision, pp. 157-158] .
The 1996 oz one and P M NAAQS proposals are the first NAAQS rulemakings
undertaken since enactment of the Unfunded M andates R eform Act (UMRA) and t he
Regu latory Fl ex ibility Act (RFA). W hile EPA concluded t hat t h e proposals were

²¹ E.O. 12866 defines “ significan t r egulatory action” to include a r ule t hat may “have an
annual e ffect on the economy of $100 million or more ....”

“significant regulatory actions” as defined by E.O. 12866, EPA concluded that the
proposals d o not trigger t he regu latory analys is provisions of UMRA or RFA.
Unfunded Mandates — Unfunde d Mandates Reform Act.
U nless otherwise prohibited by l aw, before promulgating any gene r a l
notice o f p roposed rulemaking that is likely t o r es u l t i n p romulgation o f
any rule that includes any Fe deral mandate t hat may r esult i n t he
expenditure by State, local, and tribal governments, i n t he aggregate, or
by the private sector, of $100,00 0,000 or more (adjusted annually for
inflation) in any 1 year, and before promulgating any final rule for w hich
a g eneral notice o f p roposed rulemaking was published, the agency s hall
prepare a written s tatement containing--
... a qualitative and quantitative assessment of t he anticipated costs
and benefits of the Federal mandate, including t h e c osts and benefits to
State, local, and tribal governments or the p rivate sector, a s w ell a s t he
effect of the Federal mandate o n h ealth, s afety, and the n atural
envi ronment and ... a d escri p t i o n o f t h e ext ent o f t he agency' s p ri or
consul t a t i o n w i t h el ect ed represent a t i ves ... of t h e a f f ect ed St at e, l o cal , and
tribal governments .... [ Unfunded M andates R eform Act of 1995, §202(a)]
On the i ssue o f unfunded m a ndates, EPA concluded:
As indicated previously, EPA cannot consider in setting a NAAQS the
economic or technological feasibility of attaining ambient air quality
standards, although s uch factors m ay be c o n s idered to a d egree i n t he
development of S tate plans t o implement the s tandards. Accordingl y, EPA
has d etermined that the p rovision of sections 202, 203, and 2 0 5 of the
UMRA do not apply t o t his p roposed [ NAAQS] d e c i s i o n . The EPA
acknow ledges, however, t hat any corresponding revisions to associated
Stat e implementation plan requirements and ai r quality survei l l ance
requirements, 40 CFR p art 5 1 an d 4 0 C FR p art 58, respect i v el y, m i ght
result in such effects. Accordingly, EPA will address unfunded m andates
as appropriate when it proposes any revisions to 40 CFR p arts 51 and 58.
[ Proposed De c i s i on: Particulate Matter, 61 FR 65670 (December 13,

1996)]

However, in its PM RIA, EPA does prepare a governm e n t al entities analysis.
According t o EPA, “This ... is not an unfunded m andates analysis, but provides
estimates o f t he potential budgetary impact of the co n t r o l m easures used in the
cont rol s t rat egy-cost anal ys i s affect i n g S t at e and l ocal governm ent agenci es.” EP A
notes that it “will be useful in gu iding future implementation activities....” [ P M R IA,
p.8-20]
EPA’s d enial t hat i t h ad to prepare an unfunded m andates analysis (regu latory
i m p act st at em ent ) was chal l enged i n court as p art o f American Trucking Associations
v. U.S. E.P.A. The D.C. C ircuit Court noted, however, t hat t he Unfunded M andates
Reform Act ex plicitly stat es that a failure to prepare an impact anal ys is “s hall not be
used as a b asis for s tayi ng, enjoining, i nvalid ating or otherwise affecting [ an] agency
rule” [UMRA, § 1571] . Further, the court held that “the failure to prepare a
regu latory impact statement does not re nder t he NAAQS arbitrary and capricious.”

[ American Tr u c k i n g A s sociations v. U.S. E.P.A., 175 F.3d 1027 (D.C. C ir. 1999),
modifi ed on other grounds, 195 F.3 d 4 (D.C. Cir. 1999). T hi s issue was not
addressed i n t he appeal to the S upreme C ourt, see 531 U.S. 457 (2001).]
Small Business — Regulatory Flexibility Act.
Wh enever an agency i s r equired ... to publish g eneral notice o f p roposed
rulemaking for any proposed rule, t he agency s hall p repare and make
available f or public comment an initial r egulatory flexibility analysis.
Such analysis shall d escribe t he i m p a ct of the p roposed rule on small
entities.... [ 5 U.S.C. §603(a)]
Wh en an agency p romulgates a final rule ..., after b ei n g r equired ... to
publish a g eneral notice o f p roposed rulemaking, the agency s hall p repare
a final regulatory flexibility analysis.... [ 5 U.S.C. §604(a)]
Sections 603 and 604 of this title shall not apply t o any proposed or final
rule i f t h e h ead of t he agency certifies t hat the rule w ill not, i f
promulgated, have a significant economic impact on a s ubstantial number
of small entities. [ 5 U.S.C. §605(b)]
O n the i ssue o f assessing the impact of regu lations on small businesses, E P A
said the o z one and P M p roposed NAAQS —
will not have a significant economic i m p act o n smal l entities within the
meaning of t he RFA. Instead, i t will es tablish a standard of ai r quality that
other Act provisions will call o n s tates (or in c ase of state d efault, the
federal govern ment), to achieve by adopting implementation plans
containing specific contro l m easures for t hat purpose. In other words, s tate
(or federal) regulations implementing t he NAAQS might establish
requirements applicable to small entities, but the NAAQS itself would not.
Fo r t hese reasons, t he Administrator certifies t hat t his p roposed rule will
not have a s ignificant economic impact on a s ubstantial number o f s mall
entities.” [ National Ambient Air Quality S t andard for Ozone: P roposed
Decision, pp. 163-164] .
EP A n evertheless c o n c edes interest in the “potential impact” o f t he NAAQS and
note s t h a t d i s cussion of those impacts are included i n t he RIA. In the R IAs, EPA
performs a “S creening Analysis” to “eval uate smal l entity impact s.” This i dentifies
impacts o n i ndustries classified by SIC codes.
Prior t o publication of an i nitial r egulatory flexibilit y analysis which a
covered agency i s r equired t o conduct b y t his chapter —
(1 ) a covered agency s hall notify the C hief Counsel for Advocacy o f
t h e Small Business Administration and provide t he Chief C ounsel wi t h
information on t he potential i mpacts of the proposed rule on small entities
and the t ype of s mall entities t hat might be affect ed;
(2 ) not l a t er t han 15 days a f t er t he dat e o f r ecei pt of t h e mat eri a l s
described i n paragraph (1 ), the C hief Counsel shall i den tify individuals
repres entative of affect ed small entities f or the purpose of obtaini ng advice

and recommendations from those i ndividuals about the potential i mpacts
of the p roposed rule;
(3 ) t he agency s hall convene a review panel f or such rule consisting
wholly of full time Federal employees of the o ffice w ithin the agency
responsible for carrying out the proposed rule, t he Office of Information
and Regulatory Affairs w ithin the Office o f Management and Budget, and
the C hief Counsel;
(4 ) t he panel s hall r eview any material the agency has p r epared in
connection w ith this chapter, i ncluding any draft p ropose d rule, collect
advice and r ecommendations of each individual small entity
representative i dentified by t he agency after consultation w ith t h e C hief
Counsel....
(5 ) not later t han 60 days a fter t he date a covered agency convenes
a r eview panel pursuant to paragraph (3 ), the r eview panel s hall r eport o n
the comments of t he small entity representatives and its findings ...; and
(6 ) w here appropriate, t he agency s hall modify the p roposed rule, t he
initial r egulatory flexibility analysis or the decision on whether an i nitial
regu l a t o r y flexibility analysis is required. [Small Business Regulatory
Enforcement Fairness Act of 1996, §244]
On the s ame b asis that it decided t he Unfunded M andates and the R e g u l a t ory
Fl e x i bility Act requirements did not apply when s etting NAAQS, EPA conclude s
“t hat t he small-entity provisions in Section 244 of the S mall Business Regulatory
Enforcem ent Fai rnes s Act (SBR EFA) do not apply.” [ National Ambient Air Quality
Standard for Ozone: P roposed Decision, p. 164] And s imilarly, EPA s ays i t “intends
to fulfill the s pirit of S BR EFA on a voluntary bas is” by wo r k i n g w i t h t he S mall
Business Administration t o hold panel ex erci ses t o s olicit comments and advice from
repres entatives of smal l entities.
As part of American Trucking Associations v. U.S. E.P.A., this certification t hat
SBREFA did n o t apply was challenged . However, t he D.C. Circuit C ourt found
“incontestable” EPA’s argument t hat s etting a NAAQS has n o d irect impact on small
busi n esses b ecause t h e s t at es, t h rough t he S IP p l an p rocess, det erm i n e what s ources
will be affected. Thus the court conclud e d t hat “EPA properly certified t hat its
NAAQS woul d n o t h ave a s ignificant impact on a s ubstantial number o f s mall
entities.” [ American Trucking Associations v. U.S. E.P.A., 175 F.3d 1027 (D.C. C ir.
1999), m odified on other grounds, 195 F.3d 4 (D.C. Cir. 1999). This i ssue was not
addressed i n t he appeal to the S upreme C ourt, see 531 U.S. 457 (2001).]
Other Regulatory Impact Assessments .
Besides t he regu latory impact assessmen ts re q u i red b y t he statutes discussed
a bove, p rovisions of the P aperwork Reduction Act may b e t riggered b y report i n g
requirements. In the o z one and P M p roposed rules, EPA s aid t hat t his i ssue would
arise only i n implementation.
Also, Ex ecutive Order 12848, Federal Actions To Address Environ m ental
J u stice i n M inority P opulations and Low-Income P opulations, requires each federal
agency to identify and address, as appropriate, d isproportionate adverse h ealth and
environmental impact s of program s, policies, and activities on minorities and low-
income populations. Again, EPA indicates that analys is of effects o n minorities and

low-income populations would b e appropriately ex amined in preparation o f R IAs i n
the implementation process.²²
Consultations — O ffice of Ma nagement and Budget; Other
De partments a nd Agencies.
T h e d rafts of ... ru les submitted b y the Ad ministrator t o t h e
O f f i ce of Man agemen t an d B u d g et f o r a n y i n teragen cy revi ew p rocess
..., all d ocuments accomp anying such draf ts, a nd all w ritten comments
th ereon b y o th er agen ci es a n d al l w ri tten resp o n s e s to su ch w ri tten
comments b y the Admi nistrator s hall be placed in the d ocket .... [ C AA,
§307(d)(4)(B)(ii)]
To the ex t ent permitted b y l aw, OMB ... shal l be t he entity that reviews
individual regulations .... [ Ex ecutive Order 12866]
EP A’s NAAQS decisions are s u b j ect to this OMB review: “In v iew o f its
important policy implicat i o n s , t his p ropos al has b een judged to be a ‘sign ificant
regulatory action’ within the m eaning of t he Ex ecutive Order, and EPA has submitted
it to OMB for review. C hanges made in response t o O MB suggestions or
recommendations will b e documented i n the public docket and made available for
public inspection....” [ National Ambient Ai r Quality Standard for Ozone: P roposed
Decision, p. 157] . Apparently no changes were m ade o n t his b asis, however.
Li kewise, d rafts o f t he proposed and final rule are circulated for review to other
depart m ent s and agenci es. In cases of s ubstantive controversy, a s w i t h the o z one
and P M^2.5 NAAQS , t he issue m ay go to the W hite House for final adjudication.
JudicialReview
A p eti ti o n f or revi e w o f a cti o n o f th e Ad mi n i strator i n
promulgating any national primary or secondary ambien t air quality
standard , ... may b e f iled only in the Un ited States Court of Appea l s
f o r the District of Colu mbia. ... Any p etition f or review ... shall b e
filed w ithin six ty days from the d ate n otice o f s uch p romu lgation ...
appears i n the Federal Register .... [ C AA, §307(b)(1)]
EP A’s NAAQS rulemaking is s u b j e c t to several s tatutory procedural
requirements, compliance with which i s s ubject to judicial review. The b a s i c
framework is spelled out in the C AA, §307(d), [ 42 U.S.C. §7607(d)] and details the
requirements for public notice and participation i n t he process. The final rule cannot
be based, i n whol e o r p ar t , o n a n y i nform at i o n o r d at a whi ch have not been pl aced
in the rulem aking docket as of t he date of its final promulgation. The final rule must
be accom p ani e d b y a st at em ent o f b asi s and purpose whi ch i n cl udes a sum m ary of
the factual d ata upon which t he rule is based, the m ethodology u sed i n obtaining and
anal yz i n g t he data, t he major l egal interpretations, and policy considerations

²²RIA f or Proposed Particulate M atter Ambient Air Quality Standard (December 1996), p.

8-26.

underlyi ng t h e remaking d ecision. The s tatement must also contain t he agency’s
response t o each of the s ignificant comments, criticisms, and new data s ubmitted i n
written and oral pres entations during the comment period. Courts have al so indicat ed
th a t t h ey will look at the alternatives th e agency considered (or b elieves i t s hould
have consi d ered) i n assessi ng necessary com p l i ance. On appeal t h e court m ay
reverse t he r u l e m a king action i f i t finds it to be “arbitrary, capricious, an abuse o f
di scret i o n o r o t h erwi se not i n accordance wi t h l aw,” a s t andard by whi ch t he court s
asses s the rule’s reas onablenes s and rationality based on review of t he rulemaking
record taken as a whole.
The first suits challenging t he oz one and P M NAAQS were filed i n t he D.C.
Circuit C ourt o f Appeals o n J uly 18, 1997, t h e d a y t h e f inal rules appeared in the
Federal R egister. During 1997 a t otal of 38 suits were filed for judicial review of
various asp e cts o f t he oz one and P M NAAQS . These suits were consolidated in
American Trucking Associations v. U.S. E.P.A., a r gu ed December 17, 1998, and
decided M ay 14, 1999 [ 175 F.3d 1027 (D.C. C i r . 1 9 99)] . Various parts o f t hat
decision were then appealed to the U.S . S upreme C ourt, argu ed November 7, 2000,
and d ecided February 27, 2001 [ Whitman, Administrator of Environm ental
Protection A gency, et al. v. American Trucking Assoc i ations, Inc., et al. , 531 U.S.
_______ (2001)] . W h ile the S upreme C ourt d ecided s everal aspects o f t he case –
incl uding that the C AA constitutional l y del egat ed authority to EPA and that costs
could not be taken i nto account in setting NAAQS, i t remanded t he question o f t he
validity of the s tandards t o t he D.C. Circuit C ourt. The C ircuit Court i ssued its ruling
on March 26, 2002, upholding EPA’s p ar ticulate and oz one standards.
For any rule subject to this chapter, a s mall entity th at is adversely
affected or aggrieved by final agency a ction i s entitled t o j udicial review
of agency compliance w ith the r e quirements o f s ections 601, 604, 605(b)
[... if the head of t he agency cer tifies t hat the r ule w ill not, i f promulgated,
have a significan t e conomic impact on a s ubstantial number of s mall
en t ities] , 608(b), and 610 in accordance w ith chapter 7. [5 U. S . C .
§611(a)(1)]
EPA’s d ecis i ons that requirements o f t he Small Business R egulatory
Enforcemen t F airness Act, t he Regu latory Fl ex ibility Act, and t he Unfunded
Mandates Act do not apply t o t he oz one and P M NAAQS were challenged. T hese
were am ong the s uits consolidat ed in American Trucking Associations v. U.S. E.P.A.
As previously noted, t he D.C. Circuit C ourt o f Appeals upheld EPA’s position.
Congr essi onal Revi e w
Before a r ule can take effect , t he Federal agency p romulgating s uch
rule shall s ubmit to each House of t he Congress ... a r eport ....
A r ule s hall not take effect ..., if the C ongress enacts a j oint resolution
of disapproval .... [ 5 U.S.C. §801(a)(1)(A), (b)(1)]
EP A’s publication o f its final d ecisions on the o z one and P M NAAQS on J u ly
18, 1997, triggered t he Congressional R eview o f Agency R ulemaking p rovision of
SBREFA. Under t his provision, Congress could conside r a j o i nt resolution of
disapproval, with special procedures i n the S enate t o ensure floor consideration

within 60 legi slative days. While the p rocedure was not invoked, bills to delay t he
new s tandards were i ntroduced, i ncludi ng H.R. 1984 and S . 1084, with a h earing h eld
on the l atter. No bill to rescind t he new NAAQS was reported from committee,
however.

Appendix I . The NAAQS-Setting Process

Cr iter ia D oc um en t I n itia tio n
R evi ew every 5 years
Assemble task fo rce to set L it eratu re search
C ond uct o f sci en t ifi c out purpose & scope of do cum en t revisi o n &
research cri teria do cu men t an aly si s
⁷²²^R^e^c^o^m^m^en^d^e^d^r^es^ea^rc^h
Closure
P r e p ar at i o n o f d r a f tCASAC reviewFin al cri teri a d ocume nt
ⁱ^ki/CRS-97- cri teria do cu men t
^g/w No closure
^s^.or
^le^ak^R^e^c^o^m^m^en^ded^r^e^s^e^a^r^c^h^/^r^e^vⁱ^sⁱ^oⁿ^s
^://wiki
^h^ttp Closure
Prepar at ion of draftCASAC revi ew
sta ff p aperFin al staf f pap er
Noclosure
^R^e^c^o^m^m^en^ded^r^e^s^e^a^r^c^h^/^r^e^vⁱ^sⁱ^oⁿ^s
Ad mi nistr at o r d ecisi on Proposed st andardpubli shed for publi c F inal stand ardpromulgat ed Im pl eme nt atio n
onNAAQSreview
C on g ressio n al review Jud icial r evi ew
^S^o^u^r^c^e^:^A^dap^t^e^d^f^ro^m^N^a^tⁱ^ona^l^C^om^mⁱ^s^sⁱ^oⁿ^oⁿ^Aⁱ^r^Q^ualⁱ^t^y^,^T^o^B^r^e^a^t^he^C^l^eaⁿ^Aⁱ^r⁽¹⁹⁸^1),^b^a^s^e^d^on^E^P^A^Iⁿ^f^o^r^m^a^tⁱ^oⁿ^.

Appendix II: Setting NAAQS — a Typical Chronology
(Ozone)
Augu st 1992 Oz one NAAQS review initiated
Summer/Fall 1993 W orkshops
Spring 1994 Draft C riteria Document available t o public and
CASAC
J uly 20 and 21, 1994 CASAC m eetings to review draft C riteria
Document
March 2 0 and 21, 1995 CASAC m eetings to review revised d raft of
Criteria Document; al so to review draft s ections
ofStaffPaper
September 19 and 20, 1995 CASAC closure on Criteria Document; also
cl osure o n p rimary standard section o f S taff
Paper
November 28, 1995 CASAC closure letter o n C riteria Document
sent to Administrat or
November 30, 1995 CASAC closure letter o n p rimary standard
section of S taff Paper s ent t o Administrat or
March 21, 1996 CASAC s ubpanel m eeting o n s econdary
standard section o f S taff Paper
April 4 , 1996 CASAC closure letter o n s econdary standard
section of S taff Paper s ent t o Administrat or
J une 12, 1996 EPA publishes Advance Notice o f P roposed
Rulemaking (ANP R) for Oz one & P M NAAQS
J uly 25 and August 8, 1996 Public meetings on ANPR
November 27, 1996/ Proposed Decision announced/published i n
December 13, 1996 Federal Register ; public comment period begi ns
J anuary 14 and 15, 1997 Public Meetings in four cities o n p roposal
March 12, 1997 End o f public comment period on proposal
W i nter/Spring 1997 Congressional h earings on the p roposed
NAAQS
J une 25, 1997 President C linton endorses t he proposed oz one
& P M NAAQS , with some modifications
late J une 1997 EPA s ubmits proposed final s tandards t o OMB

J u ly 16, 1997 Administrator sign s o ff on final Oz one & P M
NAAQS
J u ly 18, 1997 Final NAAQS published i n Federal Register
J uly 18, 1997 Fi rst s uit challenging t he final s tandards filed i n
the U.S. C ourt of Appeals for the D.C. District
September 16, 1997 Oz one & P M rules become effective
December 17, 1998 American Trucking Associations v. U.S. E.P.A.
argu ed before D.C. Circuit C ourt o f Appeals
May 14, 1999 D.C. Circuit C ourt o f Appeals i ssues decision
October 29, 1999 D.C. Circuit C ourt o f Appeals (en banc) d enies
petition for rehearing
May 2 2 and 29, 2000 S upreme C ourt accepts cert. on appeals b y EP A
and t he U.S. Chamber o f C ommerce
November 7, 2000 Oral argu ments o n Ameri can T r ucki ng
Associations v. U.S. E.P.A. before the U.S.
SupremeCourt
February 27, 2001 U.S. Supreme C ourt i ssues decision, which
among other findings unanimously concluded
that costs m ay not be considered in setting
NAAQS , remands issue o f adequacy of science
for s tandards t o C ircuit Court
March 26, 2002 D.C. Circuit C ourt o f Appeals i ssues decision,
basically upholding EPA’s p articulate and
oz one standards