EPA; OVERSIGHT OF
AIR NEW SOURCE REVIEWS
A Report, .by the
Evaluation Division
C he ry 1 Wa s s e'rma n
Stan Meiburfcj
July, 1981-
-------
TABLE OF CONTENTS
Executive Summary
Introduction
Part I: States Want But Are Not Getting Meaningful
Oversight
Part II: EPA Is Not Delivering Where We Have A Comparative
Advantage
Part III: Options for Future Oversight
The Goals of New Source Review
An Alternative Look at EPA Goals
Oversight Philosophies: How Much Is Enough?
EPA the Franchiser: Focusing on Quality
Tailoring Oversight: Adapting to States' Needs
Mechanisms of Oversight
Routine State Submissions
Direct Involvement in Permit Actions
Program Audits
Plant Audits
Objectives and Measures
Consequences of Oversight
Summary of Recommendations
Part IV: Appendices
1) Summary of Existing Regional NSR Oversight Approaches
2) Assessment of Routine Submissions to EPA
3) State Comments on Direct EPA Involvement in
State Permit Actions
4) Summary of State Views on EPA Program Audits
5) Present Implementation of New Source Review and
Oversight
-------
EXECUTIVE SUMMARY
The review, prior to construction, of proposed new and
modified stationary sources of air pollution has been among
the most complex and controversial areas of EPA activity in
the past ten years. Recently, however, EPA has been
attempting to reduce its own direct involvement in individual
new source decisions and return the full responsibility
for new source review (NSR) to State and local agencies.
In doing so, questions have arisen concerning EPA's role and
methods in overseeing the conduct of new source review by
State and local agencies.
The present study was commissioned by EPA's Office of
Air Quality Planning and Standards to evaluate current over-
sight and look for ways it might be done better. It involved
site visits to four SPA Regions (III, IV, V, and VI) and seven
State air pollution control agencies (Pennsylvania, Virginia,
Georgia, South Carolina, Texas, Ohio, and Michigan), along
with surveys of EPA's Regional offices on their current over-
sight" practices and discussions of the study with EPA Head-
quarters and Regional managers. The study looked both at
existing State practices in reviewing new sources and at
ways which EPA has used to oversee them. "Oversight" in this
study refers both to mechanisms such as permit reviews and program
audits, and to the delivery of EPA assistance to State programs.
FINDINGS ON CURRENT OVERSIGHT
I. States Want But Have Not Been Getting Meaningful Oversight
All of the State agencies we visited agreed that some N
form of EPA oversight of new source reviews is necessary,
important, and an appropriate part of EPA's national responsi-
bility for the control of air pollution. The chief reason why
is to ensure equity among the States in the treatment of new
sources of pollution. State agencies are concerned that without
EPA oversight, they will lose new industry to other States which
relax pollution control requirements to attract industry.
EPA's most important oversight role is to ensure that new sources
meet equivalent minimum national standards no matter where they
locate.
However, existing oversight of new source review varies
tremendously from Region to Region, ranging from a minimal
program based on informal contacts to a detailed system of
program audits, and from extensive review of draft permits
to no routine reviews. The reasons for this include continuously
changing program regulations, the inability of some Regions to
delegate parts of the program, lack of effective oversight guidance
from Headquarters, and uncertain organizational roles in the
Regions.
Many States have been subject to only minimal EPA over-
sight of new source review. Where more extensive oversight
has occurred, States had several concerns.
ES-l.
-------
HS-2
o Approach Taken by EPA
Reviews have sometimes been conducted by inexperienced
staff, or by different staff each year, with the
result that a lot of time is spent going over old
ground and State's gain little of value from oversight.
Some EPA staff were criticized for "unnecessarily adver-
sarial attitudes" towards State agencies.
o Standards of Judgment Used by EPA
States feel they are subject to "evolving criteria"
and that they do not know what EPA's expectations
are. In the absence of clear criteria, "zero errors"
becomes the effective standard of judgment and
. EPA fails to distinguish between trivial, isolated
errors in specific permits and serious flaws in
State programs.
o Consequences Which Follow Oversight
When EPA identified problems in specific permits, States
felt EPA has sometimes acted without first giving the
State a chance to correct the problem on its own, which
undermines the State's credibility with sources. Inappro-
priate EPA comments on draft permits after the close of
State public comment periods were also cited as a problem.
II. EPA Is Not Delivering Where It Has A Comparative Advantage
State officials identified four roles where EPA has a
"comparative advantage" which it is not fully using to support
State operating programs. These are:
1) Improving Levels of Pollution Control
2) Advancing Air Quality Modeling
3) Making Decisions on Policy Questions
4) Training of State Staff
1) Improving Levels of Pollution Control
EPA now fosters improvements in the level of pollution
control by promulgating new source performance standards
- (NSPS) and by providing information about new control
techniques for use in State permits.
State air pollution control officials unanimously
endorse the development of NSPS, and would like EPA to move
faster in issuing NSPS for more categories of sources
particularly industrial boilers. EPA also needs to improve
its ability to transfer technical information about control
approaches to States by improving the usefulness of the
present "BACT Clearinghouse."
-------
ES-3
2) Advancing Air Quality Modeling
EPA has a potentially fruitful comparative advantage
in advancing the accuracy and utility of models which predict
the ambient impact of source emissions. However, States
criticized EPA's modeling program for a) making it difficult
to get alternative models approved, and b) discouraging
the development by States and industry of more "user-oriented,"
lower cost models while failing to reduce the cost of applying
EPA's own models.
3 ) Decisions^ on Policy Guidance
Several States expressed frustration about the length
of time it took to get guidance on policy questions from
EPA. They commented that EPA Regional offices tend to
refer all such questions to Headquarters, which treats.each
question as a new occasion for national policy guidance.
States also noted that organizational conflicts within EPA
Regional offices .can delay authoritative rulings by forcing
even minor issues up to the Regional Administrator for
resolution.
4) Training of State Staff
Several States requested that EPA continue and expand
its program of training State staff in areas related to new
source review and were critical of recent cutbacks which
the Agency has made in this area. Recent EPA workshops to
explain the revisions of the PSD regulations and training
sessions for State enforcement personnel were praised by
State officials. Budget restrictions and constraints on
travel by State employees were cited as the two biggest
drawbacks to further State participation in EPA training.
FINDINGS ON CURRENT IMPLEMENTATION
o States generally believe that they understand and can
conduct sound reviews of permit applications by new
sources. Most applications are from relatively small
sources and are routine State actions. However, for
large or unusual sources, States depend upon NSPS or
control guidance from EPA or neighboring States. The
absence of this guidance creates substantial difficulties
for States in trying to independently evaluate the claims
made by applicants.
o Most of the air quality analyses which States conduct are
for minor new sources where local downwash conditions threaten
to violate ambient standards in the vicinity of the plant.
This analysis may affect stack heights at some minor
sources. However, air quality analysis has little effect
on the siting or the emissions limits of most major new
sources.
o State modeling capability has improved since the imposition
of the new PSD requirements and their emphasis on the use
of guideline models.
-------
SS-4
o States do little design review of prospective new/modified
sources from the standpoint of determining the applicant's
ability to comply continuously with air pollution standards.
o States pride themselves on being able to process new source
permits quickly (e.g., between 40 and 120 days). States
appear to take substantially less tine to review major
source applications than EPA does. This is an incentive
to States to take on programs such as PSD.
o States have increasingly tried to formalize their procedures
for reviewing permit applications and recording the results
of these reviews. However, several States were not satisfied
with their performance on such items as keeping accurate
files, and would like additional assistance from EPA in
this area.
o States would like to make better use of EPA plant audits
for inspector training. EPA inspectors could function
like district sales managers, observing State performance
and giving subsequent suggestions on how performance could
be improved.
Appendix 5 contains a more specific discussion of imple-
mentation in each of the following areas:
o Applicability Determinations
o . Control Technology Review
o Air Quality Analysis
o Permit Review for Enforceability
o Public Participation
o Operating Procedures
Each discussion notes our findings, describes current EPA
oversight mechanisms and possible problems, summarizes State
comments and concerns, and suggests options for EPA oversight
in each area. These options are consistent with the overall
oversight recommendations summarized below and laid out in cnore
detail in Part III of the report.
RECOMMENDATIONS (Also see summary, pages 111-30 and 31)
I. EPA Should Adopt a "Franchiser" Approach to Oversight
1) Focus Attention on Goals
EPA should use oversight to determine whether State programs
are carrying out their overall mandate, rather than whether
every action on every individual permit has been taken exactly
as EPA would have done.
2) Success Is Improving State Programs
EPA should direct its efforts towards improving the quality
of State programs. This requires that oversight be tailored to
the conditions of each State program. However, oversight may
also include a special emphasis on a few problem areas each
year which have been identified as national concerns.
-------
ES-5
3) Use Clear Measures of Performance
SPA should establish clear, objective measures and standards
of performance for oversight of a State's overall program
and for specific program elements. To be useful, these
measures and standards must be developed and applied through
close consultation vith each individual State agency. (Note:
possible measures are discussed on pages 111-23 through 111-27).
4) Change Present Oversight Mechanisms
EPA should decrease its direct involvement in specific permit
actions except in unusual cases (specified .on pp. 111-13 and
111-14 of the report). However, EPA should use program
audits conducted by experienced EPA staff. Audits should
include site visits to State offices, discussions with State
staff and managers, and a review of a sample of final permits
selected according to targeting criteria which have been
discussed in advance with the State. Results of the audit should
be discussed with State managers in an exit interview and written
up in a formal report of findings. EPA should also change its
present enforcement oversight ("plant audits") to make better use
of the opportunities they offer for training State personnel.
5) Consequences of Oversight: State Actions
EPA should give States a full opportunity to correct problems
with either a specific permit or the State program which oversight
may uncover. When program problems are at issue, EPA should
focus on mechanisms it can use (training, technical assistance,
grant assistance) to help correct them. Grant conditions
can be used, but they should focus on producing State program
improvements rather than "bean counting." Independent EPA
action to change individual permits should be limited only
to extreme cases and taken only with the approval of top Regional
management after the State has had time to correct the problem
on 'their own.
II. EPA Should Improve Delivery in Areas of Comparative Advantage
1) Improve Promotion of Better Control Strategies
EPA should improve its dissemination of technical infor-
mation on control technologies, in a form which States have
determined will be useful to them. EPA should also investigate
alternative means to promote the development of good control
strategies (e.g., an award to new facilities which use exemplary
control systems). As suggested in EPA's Continuous Compliance
strategy, EPA should develop design review guidance which helps
State and local design reviewers address potential future
operation and maintenance problems.
2) Encourage More Useful Models
EPA should expedite the approval and availability of model
modifications which decrease their cost and complexity to users.
-------
ES-6
3) Improve Headquarters/Regional Communications
Existing mechanisms (e.g., quarterly Air Branch Chief
meetings; Air Programs Policy and Guidance Notebook, annual
conferences of Headquarters and Regional staff) should be
continued and developed further to reduce delays in Head-
quarters/Regional communication on policy questions.
4) Improve Delivery of Training
EPA should review its air pollution training activities
and approaches to see how changes can be made to increase
delivery to States at reduced cost.
-------
INTRODUCTION
The review, prior 'to construction, of proposed new and
modified stationary sources of air pollution has been among the
most complex and controversial areas of EPA activity in
the past ten years. EPA's involvement in new source review
had its beginning in the 1970 Clean Air Act, through the
mechanism of 'the State Implementation Plans. Two requirements
which the Act laid out for implementation plans were of
particular importance for new sources: the requirement that
State plans provide for the maintenance of ambient air quality
standards once they were attained or where they were already
being met; and the requirement that State plans include
procedure for review of the location of new sources to which
a new source performance standard (NSPS) would apply.
The requirement to "maintain" standards in the 1970 Act
involved the review of new and modified sources because it
attempted to anticipate the air quality problems created by
future industrial growth. While State Implementation Plans
were to control existing sources of pollution so that all
areas of the country attained the ambient standards by 1975,
it was recognized that industrial growth could cause renewed
violations of the standards unless emissions resulting from
that growth were carefully controlled. EPA therefore required
State plans to contain provisions for review of all new and
modified sources (both "majors" and "minors") to ensure
maintenance of the national standards.
As it was originally envisioned, NSPS were to be the
chief means by which EPA and State and local agencies were
to control these "growth" emissions. NSPS required that a
source achieve the degree of emissions limitation achievable
by the best demonstrated system of emissions reduction,
taking cost and other factors into account. Standards covering
most types of new sources were to be promulgated by EPA, and
all new sources locating anywhere in the country would have
to meet them. The presumption was that: 1) NSPS would cover
a wide variety of industries, and 2) that sources which
met NSPS emissions limits would generally also not threaten
the attainment and maintenance of the ambient air quality
standards. The review of the location of NSPS sources,
however, was made a special requirement out of concern that
even with NSPS control, too many new sources locating in the
same place could cause violations of an ambient standard.
The Growing Complexity of New Source Review
New source review has become considerably more complex
since the passage of the 1970 Act. In issuing regulations in
1971 for the approval of State implementation plans, EPA
-------
11
included provisions which were interpreted as allowing the
deterioration of air quality in areas which had relatively
clean air. These provisions were challenged in court, a
challenge which eventually resulted in a requirement that EPA
develop regulations to "prevent significant deterioration."
These "PSD" regulations were first issued in December, 1974.
They included a case by case expansion of the NSPS concept
to major source categories not yet covered by NSPS (which
were suffering from delays in development). They also included
highly controversial provisions which limited the allowable
deterioration of air quality in areas which were cleaner than
national standards (the "increment" system).
Similarly, although all areas of the country were supposed
to attain the national ambient air quality standards by
1975, it became clear very quickly that this goal could not
be reached. In theory, EPA could have been required under the
terms of the Clean Air Act to forbid all new source growth in
areas of the country which had failed to attain the ambient
air quality standards by the statutory deadline. Instead,
however, EPA proposed a policy which would allow major new
sources to construct in non-attainment areas as long as they
obtained "offsetting" emissions reductions and employed the
"lowest achievable emissions rate" the "emissions offset
interpretive ruling".
These two provisions became the basis for substantial
legislative revisions to the Clean Air Act in 1977. The
amendments affecting new source review were contained primarily
in a new Part C and Part D of the Act, which specified require-
ments for "major" new and modified sources in PSD and in
non-attainment areas respectively. Both sets of requirements
were still to be imposed through revisions to State Implementation
Plans (required by the 1977 amendments). The Part C provisions
governing PSD, however, required sources to get a direct
federal permit in order to be able to construct if State
plans were not approved by EPA. Areas which had not attained
the standards, and which did not submit approvable Part D
SIP revisions (including provisions for new source review)
to attain them, were to be subject to a.construction moratorium.
on all major new sources or modifications. Both Part C
and Part D continued the EPA approach of requiring stringent
new source emissions controls on a case by case basis.
Meanwhile, EPA continued to set both NSPS and national
emissions standards for hazardous air pollutants (NESHAP), and
was delegating responsibility for implementing these to States
on a category by category basis as new standards were established.
EPA Concern About "Oversight" of New Source Review
EPA had originally provided for an "oversight" function by
requiring States to submit all draft permits and public notices
to EPA, though in practice this procedure was seldom followed.
-------
Ill
However, EPA became more concerned about the oversight of
State programs at the time of its offset ruling, due to the
belief that air quality in existing non-attainment areas could
worsen significantly due to growth unless new enissions sources
in these areas were wel'l controlled and offset by further
reductions. Moreover, the controversy and complexity surrounding
PSD resulted in delays by States in adopting their own PSD
rules and a greater direct EPA role in issuing new source
permits than had been originally envisioned.
In recent years EPA has been attempting to reduce
its own direct involvement in individual new source decisions
and return the full responsibility for new source review
to State and local agencies. However, this effort has not
been fully successful, nor has it been carried out with equal
force in all of the EPA regions. As long as EPA retains the
national responsibilities assigned to it by the 1977 amendments
to the Act, and retains its responsibility for ensuring the
attainment and maintenance of national ambient air quality
standards, EPA will continue to have some role in ensuring
.that new sources do not result in violations of these national
goals. This role, however, is becoming less and less one of a
direct actor and more and more one of overseeing the actions
of others.
Project Goals and Conduct
Our charge in this project, as specified by the Director
of EPA's Office of Air Quality Planning and Standards in the
fall of 1980, was to look for ways in which this oversight
might be done better. To do this, we conducted a series of
site visits to four EPA Regions (III, IV, V, and VI) and seven
State air pollution control agencies (Pennsylvania, Virginia,
Georgia, South Carolina, Texas, Ohio, and Michigan). These
visits were conducted in conjunction with staff from the New
Source Review Office in EPA's Office of Air Quality Planning
and Standards, Following each State visit, we prepared a
draft summary of our discussions with State managers and staff
sent this draft to the State for review-and comment, and
incorporated their comments into a final report on each site
visit. We have also surveyed EPA's Regional Offices on their
current oversight practices, and have discussed the study with
numerous EPA Headquarters and Regional managers and with repre-
sentatives of the State and Territorial Air Pollution Program
Administrators (STAPPA).
A Note on Definitions
The term, "new source review" is often used by different
people to mean different things. For the purposes of this
discussion, the term "new source review" will include all
Federal air pollution control requirements which any major new
source or modification must meet. This includes the emissions
-------
IV
offset, PSD, NSPS, and MESHAP program requirements. This is-.
because all these different parts are ultimately supposed to
be elements of a single review program administered by the
States.
Each unified State "new source review" program would also
include, of course, any State-specific requirements for new or
modified sources. However, for convenience's sake, provisions
which new or modified sources must meet solely because of
State regulations will be referred to as "State new source
review requirements."
The term "oversight" is also subject to different definitions
We use the terra here to encompass all of EPA's interaction with
and involvement in State new source review programs following
SIP approval or delegation. This includes not only program
"audits" or direct EPA involvement with individual State
permits, but also the delivery of EPA technical assistance
to States in targeted areas where experience has shown that
such assistance can help State and local agencies exercise
their responsibilities for new source review.
-------
PART I: STATES WANT BUT HAVE NOT BEEN GETTING MEANINGFUL
OVERSIGHT
All of the State agencies we visited agreed that some form
of EPA oversight of new source reviews is necessary, important,
and an appropriate part of EPA's national responsibility for
the control of air pollution. However, some States commented
that, they had seen very little evidence of any meaningful EPA
oversight to date. Others which had experienced oversight had
definite ideas about what they did and did not want to see
continued in any future oversight program.
Why Is EPA Oversight Needed?
According to the State_ agency officials, the chief reason
why some form of EPA oversight of new source review is needed,
even after all NSR programs become the full responsibility of
a State, is to ensure equity among the States in the treatment
of new sources of pollution. State agencies are concerned that
without EPA oversight, both they and other States would come
under irresistable pressure to relax pollution control require-
ments for new sources as part of their general competition to
attract new industry. EPA's most important oversight role, in
the view of these States, is to referee this competition by
ensuring that new sources have to meet equivalent minimum
national standards set forth by the Clean Air Act no matter
where they locate, so that discounting pollution control does
not become a legitimate incentive for drawing industry into
some States and out of others.
Although equity considerations are the greatest concerns
of State agencies, some States also see other opportunities
which an effective oversight program could promote. In some
cases, State officials feel that oversight could be a chance
for the State to assess its own program in conjunction with
EPA, .identifying areas which need improvement and ways in
which this could be done. Some States also feel that oversight
will help them because they can demonstrate the quality of their
program to EPA at first hand, and a "good" EPA rating would help
the State agency to respond to their own public officials and
constituencies. In some cases, States feel that the chance to
communicate directly and openly with EPA Regional staff can help
to avoid some of the problems which have characterized State-EPA
relations in the past.
Current EPA Oversight of New Source Review
The most striking characteristic of current EPA oversight
of State new source reviews is its variability from Region to
Region. Existing approaches to NSR oversight range from a
minimal program based almost exclusively on informal contacts
1-1
-------
1-2
to a detailed system of audits, and from extensive review of
draft permits to review of nothing on a routine basis. (See
Appendix 1 for a fuller discussion of this point).
These differences are not surprising. EPA Headquarters
has never spelled out an oversight philosophy which went
beyond the review of 10?s of PSD permits to "ensure" that
States followed the regulations. In the absence of guidance from
Headquarters, only a few Regions have independently articulated an
oversight philosophy of their own. Much "oversight" has therefore
consisted of ad hoc reactions to the pressures of events. An example
of this is the emphasis which EPA has placed on issuing PSD permits,
not because PSD sources posed inherently greater pollution problems or
threatened more severe health effects, but rather because in most
instances PSD permits were federally-issued permits. In contrast, new
source permits for non-attainment areas, because they were "State"
permits, generally have received less EPA attention.
There are several underlying reasons why new source
review oversight has taken its present fragmented course.
o The Clean Air Act and its associated regulations
have been in constant turmoil ever since 1970.
To cite only one example, the rules governing PSD
were:
Required by court suit in 1972;
- Initially promulgated in 1974;
- The subject of major CAA Amendments in 1977;
Repromulgated in 1978;
- The subject of a massive court remand in 1979;
Repromulgated again in 1980 with major changes; and
A primary focus for current efforts to amend the CAA again.
The uncertainty surrounding the future of PSD has made
States reluctant to assume responsibility for it until
they could see "what they were buying."
o For this and other reasons, many Regions have not
succeeded in delegating major portions of new source
review, or in getting States to submit approvable SIP
revisions. As a result, the practical and operational
demands of getting permits out prevented some Regions
from having the resources or the opportunity to develop
coherent oversight.
o In some Regions, organizational roles and responsi-
bilities for oversight have not always been clear,
and have hindered the development of a coordinated
approach.
-------
1-3
o EPA has not established clear, objective measures
of performance for State new source review programs.
Consequently, Regional offices have had no assistance
in focusing their oversight efforts either on selected types
of sources (e.g., power plants) or selected key activities
(e.g., applicability determinations). In the absence
of such measures, EPA has criticized State programs for any
deviation from the regulations governing new source review,
no matter how small or insignificant. With a few exceptions,
oversight "targeting" has been done only on an ad hoc basis.
Moreover, the performance measures against which State
programs are evaluated now are highly subjective and appear
to depend excessively on personal relationships.
State Views On Present Oversight Practices
As noted earlier, some States have few concerns about present
EPA oversight of new source reviews because they have seen little
evidence that any has taken place. In such cases, the State
generally assumes that their program is satisfactory in the absence
of any indication to the contrary. Moreover, in cases where PSD
has not been delegated, "oversight" at Regional offices has been
secondary to the pressing need to get PSD source permits prepared
and approved.
As delegation has progressed and more and more States
have taken on full new source review responsibilities, however,
oversight questions have come more and more to the forefront
of EPA and State attention. Some Regions which moved quickly
to delegate NSR programs have followed up on these delegations
with their own oversight approaches. Our conversations with
States which have experienced this kind of oversight revealed
several State concerns with the quality of the oversight they
now receive from EPA. However, it is important to note that
all of these concerns do not apply in each State or to each
individual case. State officials did cite some perceived benefits
from past program audits. Moreover, the States recognize that
some of their present concerns result from good faith efforts by the
Regions to carry out national guidance and regulations from EPA
Headquarters.
Concerns About Approach
o Review By Inexperienced Staff State officials emphasized
the importance of having experienced staff conduct over-
sight reviews. If the State has to spend a lot of time
explaining their program to inexperienced reviewers, or
if State staff know more than EPA staff about what EPA
requirements are, then the State feels that they are
wasting their time.
-------
1-4
o Lack of Continuity States prefer to be able to see
the same reviewer for repeated oversight visits, rather
than different ones each time. When this does not
happen, States feel that they have to spend a lot of
time repeating .past debates.
o Attitudes Although this is by no means universal,
States reported that in a few cases they have felt
that EPA "auditors" had an unnecessarily adversarial
attitude toward the State program and were more
interested in "finding mistakes" than in building the
capability of the State to do better work in the
future. In making their observations and judgments
about State programs, EPA reviewers, in the States'
view, should be sensitive to the "real world"
political and resource constraints under which State
agencies must operate. Virginia in particular felt
that it was most important that the proper tone be
established for an EPA ."liaison visit" constructive,
helpful, and supportive.
Concerns About Standards of Judgment
o Evolving Criteria States are not sure about what
criteria their program is being judged against, short
of total conformance in every detail with all regulations
and guidance governing the program. To the extent that
they do have a sense of these criteria, it is rarely
because EPA has spelled them out, but rather because
of personal relationships with particular EPA reviewers
which allow them to informally set ground rules together.
All too often, States feel that they are not aware of what
EPA's expectations are, but face instead an attitude of
"I'll know you have a problem when I see it." South
Carolina, for example, noted that EPA had criticized
them for failing to indicate the distance of a source
from the> nearest Class I area in their permit files, but
had never specified that this was an item against which
the State's program was to be judged. This tendency was
believed to result from EPA's "need" to find something to
comment about to satisfy their professional pride.
o EPA Expects "Zero Errors" State staff felt that EPA
holds up, at least in principle, a standard of "zero
errors" as the measure of a successful State program.
"Zero errors" means not only that each State-issued permit
is substantively identical to what it would have looked like
if EPA had issued it, but also that the State followed the
exact procedural steps which EPA would supposedly follow in
issuing such a permit (e.g., detailed, explicit, case by
case BACT analysis; use of EPA guideline models; case by
case exemption from pre-application monitoring requirements,
-------
1-5
accumulation of de miniinis emission changes). States
noted that "short cuts" on sone permits result from
resource limitations and attempts by the State to follow
the "letter of the law" for complex sources. Moreover
State officials are well aware that EPA does not have the
resources itself to be able to do the kind of review and
documentation on individual new source permits that EPA
appears to expect the States to perform.
States feel that some differences between what they do
and what EPA would do and in fact some outright errors by
States are inevitable in a program of the magnitude and
complexity of new source review. At present, however, States
feel SPA assumes that a problem with a single permit is endemic
to the entire State program, and that EPA should do additional
checks to see whether or not this is in fact the case. They
believe that EPA must be prepared to judge among State "errors"
that are inadvertant, minor, and not worth further attention
(e.g., failure to fully document routine decisions on common
sources); errors that are so major that they threaten the
program's integrity (e.g., will cause NAAQS violations); and
errors which reflect fundamental flaws in the State's conduct
of new source review. Such a perspective would make EPA "permit
audits" a part of an overall review of the State program. a
tool for identifying needs for program improvements rather
than a means of attacking minute discrepancies in individual
permits. :
Wrong Emphasis By EPA States commented on the apparent
preoccupation of EPA with oversight of PSD permits and
the relative neglect of permits in non-attainment areas.
They questioned whether this focus was appropriate in
light of the relative environmental effects of sources
in non-attainment areas.
Texas raised a further concern about the shifts in
emphasis to which EPA was subject. They noted that EPA
seemed at times to have a "project of the year" approach,
and felt that this resulted from attempts by Region VI
to respond to "constantly changing priorities and guidance
from EPA Headquarters."
Finally, several States felt that EPA oversight of
new source review should focus primarily on the appropriate-
ness and adequacy of the control technology required of
new and modified sources in both attainment and non-attain-
ment areas. At present, however, they feel that EPA places
too much emphasis on the review of modeling methods and
results. According to the States, modeling requires the
use of considerable judgment, and should not be a subject
for EPA "second-guessing."
-------
1-6
Concerns About Consequences
o No Sense of Priority While in practice EPA may regard
certain aspects of new source review'as "more critical"
than others, th-is is sometimes not reflected in either
EPA guidance or comments in EPA audits. These comments
may not distinguish between isolated errors in particular
cases, and systemmatic flaws in a State's new source review
program. Consequently, comments ranging from improper
LAER determinations to lack of a specific procedure to
notify PSD sources that their applications were complete
(which, the audit noted, was done by the State as., "a matter
of course") appear to receive equal weight in audit reports.
States feel that EPA needs to be explicit in specifying
aspects of new source review it considers to be most
critical.
o Inappropriate EPA Follow-Up States expressed concern
that when EPA audits uncover problems, EPA's reaction
is to refer the cases for enforcement action rather than
first raise the issue with the State. This sometimes has
resulted in EPA enforcement on the heels of a similar
action by the State. States feel that EPA enforcement
against a source whose actions have been approved by the
State, or where the State has already taken corrective action,!
undercuts the State's credibility and encourages sources to
deal directly with EPA (rather than the State) to get "real"
answers to their questions.
States also objected to the timing of some of EPA's
actions. The most common complaint was that EPA on
occasion will comment on a proposed new source permit
action after the close of the State's public comment
period. This places the State in the awkward position
of having to respond to EPA's comments separately from
everyone else's, and has the effect of 1) causing delays
in the issuing of permits, and 2) preventing EPA's
comments from being aired in a public forum. The first
of these effects is particularly onerous to States because
of the importance they and industry attach to speed in.
processing new source permits.
-------
PART II: SPA IS NOT DELIVERING WHERE WE HAVE A COMPARATIVE
ADVANTAGE
In our site visits, State officials identified four roles
where EPA has a "comparative advantage" which it should use
to support State operating programs. States felt, however,
that EPA is not now making the best use of its unique role
in the areas of:
1) Improving Levels Of Pollution Control
2) Advancing Air Ouality Modeling
3) Making Decisions on Policy Questions
4) Training Of State Staff
1) Improving Levels of Pollution Control
There are two ways in which EPA now fosters improvements
in the level of pollution control. The first is the promulgation
of new source performance standards (NSPS). The second is the
provision of technical support for and the dissemination of
information about new control techniques for use in State permit
actions. States supported both types of EPA efforts but were
critical of some aspects of how EPA is doing them now.
New Source Performance Standards State air pollution
control officials unanimously and almost unequivocally endorse
the development of NSPS for several reasons.
o NSPS address interstate equity concerns by providing
nationally uniform minimum standards which sources must
meet everywhere.
o NSPS result in real emissions reductions, as opposed to other
parts of new source review which some States regard as "paper
exercises."
o The development documents which accompany proposed NSPS are
good sources of technical information which States can use
in doing reviews of similar sources.
o NSPS give clear signals about what levels of control will
meet EPA's expectations.
o NSPS represent a clear baseline which EPA can use to measure
State performance.
o Though NSPS create pressure on States not to go beyond
them, States agree that total emissions loadings on
balance are less with NSPS than they would be without them.
II-l
-------
II-2
State officials v/ould like EPA to move faster in issuing
NSPS for more categories of sources particularly industrial
boilers. Texas also criticized the scope of some existing NSPS
specifically the failure of the NSPS for fluid catalytic cracking
units at refineries to contain an emission limit for
Technical Support for New Control Technology States
h-ave attempted to use EPA as a source of technical expertise
in evaluating applications for sources unfamiliar to State
staff. The results of these attempts have been mixed; in
some cases EPA has provided useful information, while other
times the States have not felt that their needs were met.
There has been general agreement, however, that EPA ought
to perform this technical support function.
Just what this function should include is less clear.
States have different needs, expertise, and capabilities;
a source which is unusual and unfamiliar in one State can
be relatively routine in another. How EPA can best encourage
the transfer .of expertise among States when it is available,
provide expertise for sources (e.g., synthetic fuel plants)
which are new to all States, or otherwise support States in
this area needs further study.
At present, States seeking technical support from EPA
turn to either the staff of the Emissions Standards and
Engineering Division (ESED) or to the BACT Clearinghouse.
We found that States were familiar with the technical
expertise available in ESED on a given industry or control
technique, and heard reports of where this expertise had
been helpful. However, limits to its usefulness were also
noted.
o ESED expertise is tied to areas of NSPS development.
It is more difficult to find an "expert" on a source
which is not subject to or a pending candidate for
NSPS.
o EPA staff are unwilling to ;offer technical advice or
judgments on types of control strategies or equipment
if decisions about their value are "in process" within
EPA. This is particularly frustrating when a State is
trying to get an independent evaluation of the efficiency
claims made by manufacturers of control equipment.
BACT Clearinghouse The other EPA source of technical
information available to States is that contained in the
"BACT Clearinghouse." However, States did not feel that the
present Clearinghouse is meeting their needs.
-------
II-3
o Most determinations in the Clearinghouse are for
NSPS sources, while the greatest State need for
information is for non-NSPS sources including
minors.
o States and EPA Regions regard the reporting which
the Clearinghouse requires as burdensome and they
generally do not do it. Moreover, a disincentive to
report strong determinations exists because States do
not want to be labeled as "too tough on industry."
o The present Clearinghouse document cannot be kept
current, and getting a response to requests for
more information is time-consuming.
o The Clearinghouse contains only "bare-bones" infor-
mation on source types, emissions limits, and control
equipment. More detailed information is needed to be
useful for comparison purposes in detailed engineering
review. Follow up phone calls to State or industry
staff involved in the decision may prove to be so
burdensome that it would be a decided disincentive
next time around.
State Suggestions For Improvement State officials
suggested several steps which EPA could take to improve its
activities in stimulating advances in pollution control and
transfering this information among States.
o Texas suggested the establishment of an EPA award to
facilities which adopted and used exemplary control
systems. This would represent "positive feedback"
from EPA and differ from the adversarial relationship
which so often characterizes EPA - industry relations.
States could nominate candidates for such an award.
o Ohio and Virginia would like direct computer access to
a "BACT Clearinghouse" data bank. This would enable
the State to get quick, up-to-date information on
determinations.for particular source categories.
o Virginia suggested that Clearinghouse determinations
include a phone number of a State reviewer who
could be contacted for further information. Texas
expressed concern, however, that such requests would
create an excessive burden on reviewers. Michigan
noted that if an industry contact number would be used,
it would be a constant source of irritation to the
facility, especially if they came to regret having
accepted a relatively low emissions limit.
-------
II-4
o Texas was most interested in information on control
approaches which could transfer among industries
especially within the chemical industry.
o Texas and Michigan suggested that EPA identify
advances in pollution control and incorporate them
in the Clearinghouse by writing up case studies of
the determination which would contain sufficient
detail that they could stand on their own.
o Michigan suggested that EPA identify candidates for
the Clearinghouse by reviewing the relatively few
major source determinations it makes each year when
these are submitted to the Regional office.
2) Advancing Air Quality Modeling
A second area in which EPA has a potentially fruitful
comparative advantage is in advancing the accuracy and utility
of models which project the ambient impact of source emissions.
EPA's technical expertise in this area is acknowledged, and
most of the models which are described in the Guideline on Air
Quality Models were developed either by EPA staff or with EPA
support.
. However, States particularly South Carolina, Texas,
and Michigan were critical of EPA's modeling efforts.
These criticisms were of two types.
EPA Makes It Difficult to Get Alternative Models Approved
The Guideline states that the EPA Regional Administrator may
approve the use of alternatives to the models specified in the
Guideline. However, South Carolina and Michigan indicated that
the requirements to gain acceptance by EPA of such alternatives
were so "ominous" in practice as to effectively inhibit the
development of "better" models in the interest of maintaining
"consistent" ones. Texas officials commented along the same
lines that while they could gain approval of their alternative
models from the Regional office, the State felt that their
models would not be "fully" sanctioned until they had also been
approved by OAQPS. Texas had delayed accepting PSD delegation
until they received assurances from OAQPS that their models were
approvable.
EPA Discourages User-Oriented Models A chief reason cited
by South Carolina, Texas, and Michigan for wanting to use alter-
native models was that each State felt their models were more
"user-oriented," particularly in that they are considerably
cheaper to run without sacrificing accuracy. These States
generally felt that EPA's models had been developed from a
research perspective and had not been reviewed from a stand-
point of trying to improve the cost-effectiveness of their
-------
II-5
repetitive use by States. They would like EPA assistance, which'
they do not feel they are getting now, in making better use of
model improvements which can reduce costs to States and industry.
State Suggestions for Improvements
o South Carolina has modified the PTMAX model to accept
data in CRSTER format, which reduces the time required
to prepare data for a model run from 45 minutes to
20 minutes. This also substantially reduces the
difficulty of shifting from one model to another.
Michigan also commented on the incompatibility of
different models and suggested additional work to .
make inputs and formats more consistent.
o All States were critical of the requirement that
five years of meteorological data be used in
modeling, and felt that this requirement "tripled
costs" (South Carolina), yet made any difference
at all in "less than 20% of the cases" (Ohio).
Michigan has developed a "persistence program which
screens five years of meteorological data and draws
out only those conditions which are limiting for a
particular type of source with savings which the
State characterized as "a factor of 100."
o EPA should pursue a "centers of excellence" idea in
developing, approving and disseminating better
alternative models.
o EPA Regions should convene regular meetings of modelers
in their States to transfer information about advances
and problems.
3) Decisions on Policy Guidance
Georgia, Texas, and Michigan all expressed frustration
about the length of time it took to get guidance on policy
questions from EPA. These States indicated that they found it
difficult to get someone in EPA who would take the responsibility
for making a decision. Rather than do so, they found that
Regional offices tended to refer all such decisions to Head-
quarters, which then treated each case as a new occasion for
needing national policy guidance. Moreover, States noted
that organizational conflicts within EPA Regional offices (e.g.,
between Air Programs Branches, Air Enforcement Branches, and
Regional Counsels) could also delay authoritative rulings by
forcing even minor issues up to the Regional Administrator for
resolution. Georgia officials in particular felt that these
-------
II-6
problems caused the "decentralization" of SPA to work much
less effectively than it otherwise might, and created
an obstacle to good State-EPA relations.
4) Training of State Staff
Several States requested that EPA continue and expand
its program of training State staff in areas related to new
source review and were critical of recent cutback's which
the Agency has made in this area. Recent EPA. efforts to
1) provide workshops explaining the most recent revision
of the PSD regulations, and 2) conduct training for State
enforcement staff were praised by State officials. Budget
restrictions and constraints on travel by State employees
were cited as the two biggest drawbacks to further State
participation in EPA training.
-------
PART III: OPTIONS FOR FUTURE OVERSIGHT
THE GOALS OF NEW SOURCE REVIEW
Both the overall {purposes of the Clean Air Act as specified
in Section 101 and the purposes specified in Part C for
prevention of significant deterioration establish general
goals regarding the review of new or modified sources of air
pollution. As stated in the Act, these general goals include:
o Protecting public health and welfare from the
adverse effects of air pollution;
o Ensuring that economic growth occurs in a manner
consistent with the clean up of dirty air areas and
the preservation of existing clean air resources;
o Protecting air quality in special areas, such as
national parks and wilderness areas; and
o Ensuring that any decision to permit increased air
pollution is made only after careful evaluation by
all concerned parties of the consequences of such a
decision.
The actual implementation of programs to meet these goals
is not left solely to EPA's discretion. The Clean Air Act
imposes a number of more specific requirements which individual
major new or modified sources must meet, and charges EPA or
an implementing State or local agency with the responsibility
of ensuring that such sources in fact comply with the Act's
provisions. The most basic requirement is that. any proposed
"major" new or modified source of air pollution must obtain
a permit to construct from EPA or a State/local agency. The
Clean Air Act spells out certain terms which the source must
meet to get this permit, terms which may require a source
to:
o Apply the best available control technology (BACT)
or achieve the lowest achievable' emissions rate
(LAER), depending on location;
o Analyze its projected impact on the air quality of
the surrounding area to demonstrate that it will
1) in non-attainment areas, provide a net air
quality benefit, or 2) in attainment areas, not
violate an applicable ambient air quality standard
or increment of deterioration an analysis which
may require monitoring and modeling;
III-l
-------
III-2
o Obtain "offsetting" emissions reductions to compensate
for any increases;
o Meet explicit standards which have been specified by
EPA for selected categories of new sources (NSPS) or
sources of certain hazardous pollutants (NESHAP); and
o Meet other procedural requirements relating to public
participation, compliance by other sources with the
same owner as the prospective new source or modification,
or implementation of the SIP in the area where the source
proposes to locate.
The applicability-of each of these requirements varies
in the statute, depending upon the type of source involved,
its size, its location, and whether it is a new source or a
modification. EPA or the implementing State or local agency
is responsible for determining which of these requirements
apply in individual cases.
The Clean Air Act is silent on any goal for the review
of proposed new or modified "non-major" sources of air pollution,
except to the extent that such sources pose a potential threat
to the attainment and maintenance of any National Ambient Air j
Quality Standard or increment of deterioration. EPA has,
however, specified (40 CFR 51.18) that State Implementation
Plans shall contain procedures which enable the State or a
local agency to determine whether or not any new/modified
source will interfere with the attainment and maintenance of
national standards.
EPA Goals and Implementing Regulations
While the Clean Air Act specifies many requirements for
new source review, the program has had to spell out considerably
more detail in the form of implementing regulations. These:
1) specify the precise definitions of terms used in the Act
(e.g., a "major stationary source"); and 2) describe the pro-
cedures which EPA or State/local agencies must follow to approve
a permit which allows a major new or modified source to construct,
In practice, EPA has used compliance with these regulations
as its operational "goal" for new source review. Yet defining
its "goal" in terms of simple compliance with the details of
the implementing regulations has created real problems for EPA
and the States because the regulations have been in almost
constant turmoil due to court suits. From the States' perspec-
tive, every time a new judgment is rendered, EPA goes through
wild swings from one extreme to another on such fundamental
questions as which sources and what pollutants are subject
to Federal new source review requirements. States feel that
EPA has exhibited very little consistency in its goals for
NSR except that it has tried in a legalistic way to react to
the perceived instructions of the latest court suit.
-------
III-3
AN ALTERNATIVE LOOK AT OVERSIGHT GOALS
The troubled history of the Air new source review program
in the statute and in the courts has produced an overwhelming
tendency within EPA to look at the program's regulations as
equal to its goals. 'By this standard, only a State program
which did exactly what the regulations required in exactly the
way EPA would have done it could satisfy the Agency's own
goals for new source review. Even where EPA has wanted to
give State and local agencies more flexibility in implementing
the provisions of the Act, the Agency has felt constrained
by legal interpretations which raised the spectre of even
more court suits if EPA or States deviated in their requirements
from the precise specifications of the regulations. The result
has been mutual frustration: EPA feels that its hands are tied
in imposing, requirements, and States feel that EPA is unnecessarily
rigid in specifying requirements for their programs.
Legislative changes to increase the flexibility of State
governments in such areas as modeling and monitoring requirements,
applicability for modifications, allowable increments of deteriora-
tion, and public participation requirements offer the best way
out of this dilemma. Even if legislative change is forthcoming,
however, issues regarding what parts of new source review should
be nationally uniform, and which should be within the sole purview
of State/local agency judgment, will continue to be resolved as
part of EPA's future oversight efforts.
A later section of this paper discusses, in detail,
measures of performance which we believe are appropriate for EPA
oversight of new source review, based on our discussions with
State and EPA officials. It is important first, however, to
put these measures in context by providing a set of general
national goals which we believe should guide future oversight
of new source review by EPA under either the present or proposed
revisions to the statute.
The following discussion touches on three such goals as
candidates. They are not the only possible goals, but each has
been stated at one time or another as-a "purpose" of new source
review and can serve as examples of the type of focus which EPA
might want to have in overseeing new source review. Each
goal is related, but they differ in concept and emphasis. EPA
programs which support these goals are mentioned.
Goal #1; Protecting Environmental Equity Among the States
The first goal for EPA in overseeing new source reviews would
be to promote equity among the States and protect the basic
integrity of State air pollution control programs. EPA actions
in pursuit of this goal involve the setting of minimum national
standards of new source pollution control (e.g., NSPS/NESHAPS),
to minimize the extent to which siting decisions for new sources
are influenced by competition among potential locations in
loosening pollution control requirements, and the review of
State programs to make sure that these minimum standards are
being applied.
-------
III-4
Goal #2; Reducing the Total Tonnage of New Pollution Being
Emitted Into the Atmosphere
Emissions from new (since 1970) sources of air pollution are
still a relatively small percentage of emissions from existing
sources in most parts of the country, with the exception of
areas which have experienced rapid growth only in the last
20 years. However, one of the key assumptions of air pollution
control has been that air quality will improve, over time, as
a part of routine industrial turnover in which older "polluting"
equipment wears out and is replaced by new equipment which is
either inherently less polluting or for which better emissions
controls are designed in prior to construction. Moreover, an
emphasis on putting the "best" new pollution control on both
"replacement" sources and "growth" sources yields two additional
benefits a reduction in emissions into the atmosphere which
are transported over long distances with uncertain ultimate
effects; and the protection of areas which are currently cleaner
than the national ambient standards. EPA actions in pursuit of
this goal focus on efforts to stimulate the transfer of
knowledge about and use of. new methods of pollution control,
including methods which promise greater reliability and long-
term compliance with emissions standards.
Goal $3; Limiting Air Quality Deterioration
A third goal for EPA limiting the deterioration of air
quality is strongly linked to the second goal. Existing
evidence appears to indicate that in most cases, air quality
deterioration can be limited by the installation and use of
good pollution control technology on new and modified sources.
However, in selected cases (e.g., extremely hazardous pollu-
tants, areas of special national significance such as
national parks, or areas where a high concentration of new
and existing sources threatens the national ambient standards),
there may be an additional concern that the source will need
to take special measures to protect existing air quality.
EPA's actions in pursuit of this goal include the provision
and/or dissemination of improved air quality models, and
the review of permit applications or determinations which
appear to affect an area of special significance or threaten
the attainment and maintenance of an ambient standard.
Opportunities and Limits of Goal-Oriented Oversight
An oversight approach which focuses on goals rather than
on the specifics of implementing regulations seems most appro-
priate for a program such as new source review where issues
frequently turn on the judgments made by agency reviews and
officials. Moreover, it calls EPA attention to the funda-
mental question of whether the program under review was succeeding
in carrying out its overall mandate, rather than whether every
action on an individual permit had been taken exactly as EPA
would have done. Finally, the goals listed above, or alter-
natives selected by top management, provide a context for
EPA's assessment of a State/local program which goes beyond
-------
III-5
the procedural requirements of the regulations and
focuses on the real environmental benefits which were being
achieved by new source review.
However, EPA cannot move directly to a goal oriented
oversight program withdut specifying several other items
upon which such a program would be based. Different
"generic philosophies" exist about oversight and EPA's role,
and the choice of a philosophy will affect the mechanisms
which EPA establishes to implement oversight. Moreover, any
oversight should be tailored to some extent to differences
in individual State abilities to meet the goals of new source
review. Finally, even if philosophies are agreed upon and
some consensus can be achieve about the "need" of particular
State programs for oversight, difficult questions remain to
be resolved about precisely how EPA would carry out its
oversight role, and what measures EPA could use in assessing
how well States are meeting these goals.
These topics are the subject of the following sections.
OVERSIGHT PHILOSOPHIES: HOW MUCH IS ENOUGH?
There is disagreement about the overall level of oversight
which State air pollution control agencies "need." There is a
general consensus on the importance of some oversight for the
purpose of ensuring environmental equity among the States, and
the need for EPA to continue some of its basic standard-setting,
technology transfer, and training functions. However, EPA actions
which go beyond these involve different values (e.g., willingness
to live with uncertainty or less than'"zero errors") and require
policy direction and choices from top management.
In brief, the prevailing perspectives on oversight may
be summarized as follows.
"States Require Considerable Oversight" Proponents of
the view that State new source review programs need considerable
oversight usually cite three support arguments. The first points
to the Clean Air Act as establishing a "Federal mandate" for the
protection of air quality as a resource. While this is held to
be especially true for areas of special Federal concern national
parks, or areas where air quality is worse than Federally-set
standards proponents of the "considerable oversight" view feel
that the -very nature of air pollution requires active coordination
and control by the Federal government, and feel that the Act
supports this interpretation.
A second argument in support of the "considerable oversight"
view points to limited State staff capabilities as another reason
for EPA involvement. Low State salary scales and high turnover are
cited as evidence which detracts, in this view, from the ability
of States to follow the requirements which Clean Air Act contains.
Perhaps more importantly, States are felt to have limited "political
resources" as well, and to be more likely to be subject to undue
political pressure than the Federal government is, particularly in
bringing enforcement actions against recalcitrant sources.
-------
III-6
A third contention on behalf of the "considerable oversight"
philosophy is that the experience with air pollution control
prior to 1970 shows that only a strong Federal involvement can
bring about real progress in achieving air quality goals. The
importance of State and local implementation is acknowledged,
but active involvement by the Federal government is seen as the
"backbone" which supports State and local efforts and has brought
about real improvement ...in their quality over the past ten years.
"States Require Minimal Oversight" Proponents of a
"minimal oversight" philosophy point to the Clean Air Act's
injunction that air pollution control is "the primary responsi-
bility of States and local governments" as a touchstone for an
EPA role which is much less involved with day to day decisions.
They generally agree that the magnitude of the pre-1970 problem.
required a strong Federal presence at that time. However, since
1970 significant improvements have occurred in most State programs,
and many would argue that the more advanced States have staff who
are as good or better than staff at EPA.
Two other arguments support the "minimal oversight" view.
The proper Federal role is seen as being that of setting basic,
minimum standards of performance, both for important parts of
new source review (e.g., NSPS) and for State performance as a
whole (though, as noted earlier, EPA has not done this).
"Standard setting" is a function which the Federal government
can do well, and which is efficient as compared to having 250
State and local agencies each doing original research on the
same problem. However, the Federal government does not do a
good job of implementation, which requires the adaptation of
programs to local conditions in order to achieve specific
objectives. It would also require the Federal government to
employ many more staff than it now does. Federal involvement
in implementation, it is argued, is an inefficient use of
Federal, State, and local resources, and can produce counter-
productive uncertainty, delay, and confusion about who is
responsible for what.
Secondly, oversight beyond a "minimal" level inevitably
gets into questions of judgment. Much of new source review is
predicated on "judgment" about (for example), control technologies
and air quality impacts. Experts can and often do disagree on
these questions. Given the importance which .judgments play in
new source review, proponents of "minimal .-oversight" argue that
it is appropriate for them to be made at the- level of government
which are closest to those who will actually be affected
that is, the State and local level.
EPA THE "FRANCHISER"; FOCUSING ON QUALITY
Past work on delegation in other EPA programs has suggested
an alternative model for delegation and oversight which goes
beyond the discussion of "how much or how little" oversight is
needed. This has been referred to as the "Franchiser" model,
because it suggests a relationship between EPA and State and
local governments which is similar, though not identical, to
the relationship between a franchiser and its franchises.
-------
III-7
The key to the Franchiser model is a division of labor among
levels of government, with each providing those services for which
it is best suited. where it differs most from present practices
is that it requires SPA to consciously acknowledge that "success"
for the Agency comes not from the direct conduct of activities
(e.g., issuing permits), but rather from what State and local
governments accomplish'using tools and methods supplied by
EPA. This means that EPA must spend more of its time and
resources in identifying common problems which affect many
State.and local programs and which EPA can solve as a common
problem rather than as one unique to each State and local
agency. Some such problems, such as information transfer
about new control technology and stimulating development of
more "user-oriented" models, have already been identified.
A "Franchiser" EPA might not totally give up its role in the
quality control of individual items if they were of sufficient
importance to the national interest. PSD Class I areas or inter-
state issues might be one example of where such a residual
Federal role might be appropriate. EPA's focus, however,
would shift away from the review of individual actions and
toward building the capability of State and local programs
to handle these actions without EPA intervention. "Oversight"
would then become one of a series of opportunities (others
being targeted technical support or training of State staff)
to do this, rather than an occasion for looking for ways to
criticize the State program.
The shift to a Franchiser model of oversight will, in some
cases, require a greater degree of trust among EPA and State and
local agencies than has existed in the recent past. It will
also require a willingness by EPA to allow some "mistakes" to
occur as an inevitable part of learning. The benefits of such
a shift, however, could be a national pollution control effort
which is more effective because it better uses the particular
skills and talents of each of the participants.
TAILORING OVERSIGHT; ADAPTING TO STATE NEEDS
The need for consistency in the results of new source
review need not mean that "oversight" would mean the same
thing at all times and in all places. In fact, there are good
reasons why it should mean different things. State needs and
capabilities differ, and whatever approach EPA takes to oversight
should be flexible enough to respond appropriately to these
differences.
In general, oversight should vary according to the size
and technical competence of the State's new source review program,
and according to the length of time during which the State has
had full responsibility for the program's implementation. Some
State programs now have capabilities in certain areas which stem
from experience and are equal to those available to EPA's (e.g.,
in the area of modeling or control technology reviews for
selected industries). It does not make much sense for EPA to
-------
III-8
invest in overseeing State decisions'in areas where State expertise
matches that of EPA. Where EPA should apply its effort is in
transferring the results of this expertise to other State and
local agencies.
Similarly, EPA may find it necessary in the initial stages
of (for example) PSD delegation to review a higher number of
actions to ensure that the State agency fully understands the
application of the rules of the program. (Note: most States
will have less than 20 PSD actions each year under current rules,
and even fewer if certain applicability provisions are changed).
However, this intensive review would not ordinarily last long.
After a specified period of time (e.g., a year or two), EPA's
level of oversight would decline to a "maintenance" level in
the absence of evidence that the State was failing to properly
conduct the program.
MECHANISMS OF OVERSIGHT
Four oversight mechanisms are now in use by EPA:
o Review of Routine State Submissions
o Direct Involvement in State Permit Actions
o Program Audits
o Plant Audits
The following sections discuss the present use of each of these
and suggest the conditions under which they would be most useful
in the future.
Routine State Submissions; What Should EPA See?
EPA Regions vary in what they now see of state permit
actions. Among the items EPA now sees are:
- lists of new source actions
applications received or
final permit actions
selected permit applications
draft public notices and draft permits
final permits
Sometimes this information is submitted for major sources only,
and sometimes for all new or modified sources. EPA has used
methods ranging from grant agreements to informal letter requests
to institute these requirements.
The regulatory or legislative bases for routine submissions
include:
o a requirement in 40 CFR 51.18 for States to submit
all draft public notices and draft permits to EPA.
-------
III-9
o a requirement in §165(d) of the Clean Air Act for
states to submit copies of permit applications for
all major PSD sources and notice of all major actions
related to processing the PSD perm.it application.
Because these requirements have been viewed as unrealistic or
onerous by States and Regions alike, EPA Regions have negotiated
different submissions on a State-by-State basis. However, there
is no overall program logic or clearly established purpose for
establishing these requirements. There is also no mechanism to
re-evaluate the need for the submissions at regular intervals.
Often a prior request is never rescinded even as new ones are
substituted. Region V enforcement officials noted that routine
submissions can drive the work of the Regional Office staff
"if we get it we should be reviewing it."
Table III-l contains our recommendations on when routine
submissions can best satisfy the various information needs we
identified. It is important to note that major sources constitute
a very small percentage (10% or less) of State permit actions.
We asked States for their comments on routine submissions during
our site visits; these are summarized in Appendix 2.
As Table III-l shows, the principal reasons why routine
submissions might be useful to EPA are to provide a basis for
active involvement in selective permit reviews, or to provide
duplicate files in EPA's Regional office for enforcement purposes.
Public.notices for major source permits may be useful to alert
EPA to permits of special concern and allow EPA to request more
information of the State and comment within the State's public
comment period. We believe it is questionable whether routine
final permit submissions are of value to EPA. We also believe
that the issue of whether EPA should have a complete record
of SIP requirements goes beyond new source review and should
be addressed in a wider context than new source permit sub-
missions alone.
-------
I11-10
TABLE III-l
ROUTINE SUBMISSIONS
TO EPA
PURPOSE
- Identify potential sources
requiring EPA permits/review
Familiarizing EPA staff with
State permit actions in
preparation for the program
audit
Provide a basis for active
involvement in selective
permit reviews (i.e. public
comment)
RECOMMENDATIONS
Request list of candidates
for EPA review only in non-
delegated programs
List of applications received
is preferrable to list of
final permit actions, but
EPA should require what is
easiest for the State to compile
This can be achieved through
receipt of public notice/draft
permits or final permits if
there are other reasons to
receive them. If not, we
recommend reliance on informal
contacts.
Notification of EPA for early
opportunity to comment on '
state permit action requires
draft permit/public notice.
We suggest that public notice/
draft permits be reviewed
selectively:
- for all majors or perhaps
a small portion of minors
where the State has evidenced
problems with its NSR program
and more frequent audits are
not feasible; or
- only for selected types of
situations where we decide
EPA involvement may be
required. (See section
on when EPA should get
directly involved).
Permit applications or further
technical support analysis
should only be requested in
highly selective cases where
EPA has decided it must get
actively involved. No routine
submission should be required.
-------
III-ll
Maintain a complete record
of applicable SI? requirements
Support enforcement
inspections with duplicate
final permit files in
Regional office
Identify exemplary permit
limits for the BACT Clearing-
house
Identify and correct
erroneously issued permits
The need for EPA to have a
complete record of SIP require-
ments and how this can best
be achieved should be reviewed
more comprehensively than just
trying to obtain copies of
final MSR permits. Such a
review should include what
to do about operating permits
and SIPs in general.
Because final permits are no
substitute for review of
complete files and a visit to
State offices prior to inspec-
tion we do not recommend sub-
missions for this purpose.
This can be accomplished during
a program audit.
This can be accomplished during
a program audit.
-------
111-12
Direct Involvement In State Permit Actions
EPA now makes very few if any comments on State permitting
actions. The bulk of EPA's interactions with States on new
source reviews for specific sources have been on cases where
a source had to get both a State permit to construct and a
PSD permit from EPA. The States we visited which had a PSD
SIP or full PSD delegation reported a substantial decline
in the level of EPA involvement with the State on specific
permit actions.
We asked States when they believed EPA should get directly
involved in State permit actions, both in general and in cases
involving:
a PSD Class I area
- a hazardous waste facility (e.g., incinerator)
- potential interstate impacts
- environmental impact statements
Their responses are summarized in Appendix 3.
Table III-2 lists our recommendations concerning direct
EPA involvement in permit actions. In general, we recognize
that in certain highly controversial cases, EPA involvement
is almost inevitable. However, we recommend that EPA issue
a formal comment on a State permit action only in unusual
instances when that comment will make a significant difference
in the outcome of the permit. These comments should be made
within the period allowed by the State for public comment, and
only after prior efforts to resolve the matter with the State
agency. If comments will not make a difference in the permit
at issue but could affect future actions, EPA should withhold
formal comment but take the matter up with the State at the
next State program audit. EPA involvement in all other permit
actions should be at the request of the State agency.
-------
111-13
TABLE III-2
DIRECT INVOLVEMENT IN STATE
PERMIT ACTIONS
Type of Case
General Cases
Controversial Cases
Cases of Potential
Interstate Impacts
Permitting of Hazardous
Waste Facilities
Recommended Comment
Procedure
Record comments (if any) for
discussion during annual
program audits
All EPA comments should be made
within the State's public comment
period. EPA should issue comments
only when the comment will:
o Make a difference in the
outcome (e.g./ propose a
change in the source's
emissions limit to prevent
NAAQS violation)
o Prevent the source from
.violating a regulation or
statute
o Rectify a significant technical
error (e.g., failure to meet
an applicable NSPS).
EPA should give advance notice
to the State of pending comments
as a professional courtesy.
EPA should get involved only at
the request of either party to
a dispute, and only after good
faith efforts by both parties
to resolve it.
EPA should ensure that if the
State has permitting authority
under both air new source review
and RCRA programs, there is a
coordinated procedure to issue
any necessary permits and resolve
disputes. If EPA retains authority
under RCRA, the Agency should treat
the action as a controversial case.
-------
111-14
Cases with Potential
Class I Area Impacts
EPA should not routinely get
involved in these cases.
When significant controversy
or a request by a Federal Land
Manager triggers involvement,
treat as a controversial case.
EPA should act as an arbiter
between a State and a Federal
Land Manager only if requested
to do so by either party.
-------
111-15
Program Audits
Three Regions now conduct program audits of one sort or
another, and others are planning to as a part of their oversight
of PSD delegation. Region II staff review a percentage of State
new source pemits each'year, write up a summary of each permit
reviewed, and take up any problems with the State reviewing
engineer. Issues of general concern which these permit audits
reveal are discussed by the Region in quarterly meetings with
State staff.
Region IV has sponsored contractor audits of two of its
State programs. In both cases the contractor, TRW, went to the
State office, spent time going through the State files, and
prepared a draft written report outlining problems with the State
program. Both reports were critical of the State programs involved.
The Region felt the contractor audits had been useful in identifying
problems and prompting management changes in the agencies concerned.
Region V has the most extensive and formal audit program of
any Region. They use an "audit team" composed of a senior team
leader and staff with various specialties, including modeling.
The team visits the State office (or district/local offices in
decentralized States) for a two day visit, during 'which time they
review selected permit files and talk with State staff. At the
conclusion of the visit, they hold an exit interview with the
State program manager, then prepare a written report in the form
of a letter to the State which contains findings and recommendations,
The recommendations are divided between those which the Region feels
the State must implement to satisfy a regulatory or statutory
requirement, and those which the Region believes would be good
program practice but are not required.
Region V is now expanding its audits to cover air enforce-
ment and monitoring activities as well as those related to new
source review.
State Reactions _to Program Audits All of the States we
visited endorsed the program audit in some form as an oversight
mechanism. As the States see it, this audit:
o Would occur annually or semi-ahnually;
o Would replace day to day EPA involvement in State permit
actions;
o Would include:
Discussions with State officials and staff on the
program's progress; and
-------
111-16
The review by EPA staff of a selected number of State
permit files at the State office. As noted below,
This might include the review of some ninor source
permits for applicability purposes. States would
like to see some "randomness" in the selection of
these permits to ensure that EPA does not just look
at the "best" permits.
State views on program audits are summarized in Appendix 4.
All States stressed the importance of the approach which
EPA uses in such an audit. States feel that EPA1s- approach:
o Should be cooperative and not adversarial, in nature;
o Should test whether the State is, in practice, following
its own rules;
o Should use, as auditors, staff who are familiar with
the State program;
o Should vary from State to State, depending on the different
needs of each program; and
o Should forego excessive interest in "process" and
focus instead on "results".
States would like EPA to spell out just what an "auditable"
permit file should contain, though one State specifically commented
that such specifications should not take the form of regulations.
They suggested that an EPA request that certain items be
included in permit files, without rigidly specifying an
exact management system, would be saleable and was. probably
necessary if an audit system was to work. Some of the States
we visited already have permit filing systems in place which
appear workable, and EPA's role would be to stimulate information
transfer between States with good systems and States which
need them.
Finally, State officials felt that the State must have
the first opportunity to correct any deficiencies which program
audits may reveal. Unilateral EPA enforcement action based on
audit findings would destroy the cooperative tone which States
feel should be the backdrop for program audits.
Permit vs. Program Audits The chief purpose of permit
audits is not for EPA to "hunt" for errors in specific permits,
but to identify problems which the State is having in the conduct
of its reviews. To have useful outcomes result from this exercise
requires communication between EPA and the State and preferably
directly between the EPA staff person reviewing the permit and
the State staff person who prepared it. For this reason, we
recommend that any EPA review of State permits occur at the
State office, where both State reviewers and complete state permit
-------
111-17
files are readily available. This has the added advantage of
encouraging both EPA and the State to clarify, explictly and
face to face, what their mutual expectations are for an
"adequate" permit review. It also placed the "permit audit"
in its proper context as part of the continuing discussions
between EPA and the State on how the State's program can be
improved.
There is not a simple answer to the question of "how many"
permits EPA should review in such an audit. Most States do
not review enough "major" sources per year for a truly "random"
sample to be taken from these sources. While some high growth
States may process up to fifty such permits each year, this
level of activity is highly unusual. Twenty to thirty major
source permits per year is a more common level, even in major
industrial States. Moreover, differences in the types of.
sources involved (e.g., power plants versus refineries) mean
that the skills needed for review may be vastly different and
that the sample "universe" is far from uniform.
Generally, therefore, a percentage "random" sample of
major source permits will not yield particularly useful
information about a State program. In the absence of a simple
"sample" approach, the most important factor which determines
which and how many final permits get reviewed should be
discussions which EPA has in advance with the State agency.
EPA, in these discussions, may use several factors as a basis
for estimating how many and which permits it would like to
see, such as:
- how long the State, has been responsible for major
source review under offset or PSD provisions
how well the State performed in its last audit
industries which are unfamiliar to the State or
about which the State has expressed concern to EPA
types of industry within the State which are experiencing
rapid growth
selected source categories in which EPA may be trying
to improve State capabilities nationwide.
Review of Minor Source Permits We examined the question
of whether EPA should review any minor source permits for
applicability concerns
- local ambient impacts
toxic or other hazardous pollutants
general SIP adequacy
-------
111-18
States generally endorsed some review of minor source permits
to help ensure the uniform application of Federal requirements
about the kind of review to which new sources should be subject.
A look at selected "borderline" minor permits might be an
appropriate way to determine whether, in these instances,
the State is applying its rules properly. States also expressed
the desire for additional technical support in making control
technology determinations for selected categories of new or
unfamiliar minor sources (e.g., an alcohol fuels plant).
State permit reviews for new minor sources are generally
routine and repetitive in nature. Most States have established
standard operating procedures to handle such reviews, by
establishing "standard" permits for small sources (e.g.,
Texas) and by developing control efficiency curves to establish
allowable emissions levels for small sources (e.g., Virginia
for small boilers). Ideally, any necessary EPA oversight of
minor source permitting activities would occur in the context
of a general review of the adequacy of a State's Implementation
Plan to achieve the ambient standards. In practice, however,
it appears that EPA has conducted little implementation review
of SIP's (as opposed to "paper reviews" of State submissions).
Whether there is a need for such a review, and if so how it
should be conducted, is an issue which goes beyond new source
review and would require further study.
Table III-3 summarizes our recommendations regarding
program audits.
-------
111-19
TABLE III-3
AUDITS OF STATE PROGRAMS
ISSUE
o Who conducts?
How should audit be
conducted?
How should permits be
selected for review?
Recommendations
o Prefer Regional office staff
to contractors. Staff should
be familiar with the State program
in general, but have a variety
of skills (e.g., a modeler, an
enforcement staff, and design
engineer). A "team of experts"
approach is desireable; another
possibility is the inclusion
of staff from other States on
the audit team.
o Prior contact with State staff
to arrange time.
o Prefer permit review during
site visit to State office to
desk review of permits at
Regional office.
o Include discussions with State
staff during visit and an exit
interview with program manager
o Prepare formal written report
of findings; send in draft to
State and get comments. Formal
comments should include an out-
line of the criteria used to
judge "adequacy".
o The criteria for permit
selection should be discussed
with the State in advance.
Possible criteria include:
- Selected sample of a minimum
number of majors
Controversial cases, (e.g.,
Class I, interstate, hazardous
air pollutants)
Special focus on selected
source categories of national
concern
-------
111-20
o How often should audits
be conducted?
What should the consequences
of a permit audit be?
What should the audit
focus on?
Permits of special Regional
concern (e.g., energy
facilities; high-growth
industries)
- Minor source permits with
applicability questions
o Negotiate with State; may
be as often as quarterly
or as seldom as biannually,
depending upon State pre-
ference and program quality
o The State should be given
the final formal audit
report and a chance to act
on its recommendations before
any EPA action is taken
o Problems with specific permits
should be distinguished from
overall concerns with the program.
o If problems with a specific
permit are not satisfactorily
resolved, the State should be
formally notified of any action
EPA intends to take.
o Eve'ry effort should be made to
work out a joint course of
action with the Sta-te and
offer EPA assistance to State .
action in preference to direct
EPA action.
o The audit's focus should ideally
be on the overall conduct of the
State's program -- the general
level and quality of permit
reviews, the emphasis given by
State program managers to
environmentally sound new source
reviews, and the integration of
new source review with other
parts (e.g. compliance) of a
State's program.
o EPA may target selected industrial
categories from year to year
a national effort to expand
capabilities.
-------
111-21
o In a given year, by agreement
with the State, specific areas
may be targeted for special
attention:
Modeling
Enforceability of Conditions
Public Participation.
o A focus on minor source permitting
as part of a general review of
SIP adequacy is a question for
further study.
-------
111-22
Plant Audits
A few States commented on the possibility of uses EPA plant
inspections as an "audit" tool for new source review. In such
a "plant audit," EPA would check to see whether the State's permit
review had accurately characterized the source's emissions and
control strategy, and whether the facility was operating in con-
formance with permit specifications.
Judging by our interviews, present EPA plant inspection
practices are a source of controversy. The biggest items of
contention were:
o Use of contractors for inspections (Michigan,
Virginia, Ohio).
o Poor communications with EPA about which sources EPA
would inspect (Virginia, South Carolina, Texas, Michigan).
o Overlapping State-EPA inspections (Virginia, Texas).
o Failure to receive copies of EPA inspection reports
(Virginia, South Carolina, Michigan).
o Delay in getting reports up the line at EPA for
action (Texas).
o Arbitrary selection of 10% of plants for audits (South
Carolina)
o Lack of inspector expertise (Virginia, South Carolina).
o Objection to the philosophy of "independent" EPA
inspections (Michigan, Ohio, Virginia).
South Carolina, Ohio, Virginia, and Michigan all expressed
an interest in using EPA inspections as a vehicle for training
their staff to do better inspections on their own, with Michigan
comparing a proper relationship to that of a salesman and a sales
supervisor. Only Texas indicated that they preferred not to
accompany EPA inspectors on site visits.
-------
111-23
OBJECTIVES AND MEASURES
The key to the use of any of the oversight approaches
discussed above is the establishment .of agreed-upon measures
of program performance.' Without such measures, States will
have no idea of what standards they are supposed to meet in
implementing new source review other than the impossible one
of doing things just .as EPA would theoretically do then. The
absence of such measures has contributed to EPA's treatment
of the program's regulations as equal to its goals. It has
fostered reluctance by the States to accept delegation because
they do not know what standards they must meet, and it has
fostered reluctance by EPA to delegate because we have no
criteria for comparing State performance or judging what a
successful State program is.
Suggested objectives and performance measures are specified
below. The standards against which these measures of performance
apply will, to some extent have to be negotiated on a State by
State basis. While EPA is in the position of "national manager"
for new source review, the very fact of federalism means that
the manager-supervisor relationship can never be a true characteri-
zation of EPA-State relations. We have earlier suggested a
"Franchiser" as a model for EPA's role; this model applies here
with particular force. EPA cannot and should not try to dictate
national standards of performance except in selected areas covered
by the Clean Air Act (e.g., .NAAQS or NSPS). The Agency can,
however, work with States in partnership to improve their joint
performance against measures of effectiveness to which both can
agree.
Possible Measures
1) Can the basis for State decisions on major source permits
be determined?
o Example #1: What are BACT determinations based on?
NSPS
CTG
AP-42
Stack tests at similar sources
State SIP
Vendor-supplied data or applicant data
Judgment of State engineer
-------
111-24
o Example #2: Is air quality analysis being done properly?
-- Model selection justified (Guideline or Alternative)
Assumptions Justified (e.g., urban vs. rural model;
single vs. multi-source podel).
o Example £3: Have the proper rules been applied to the
source?
Is a checklist of applicability considerations
routinely completed for each source?
Types of decisions for which no justification is consistently
provided may lead to further evaluation of that element of a
State's program and recommendations for changes (e.g., if a State
relies exclusively on applicant data and fails to perform any
independent engineering evaluation of a source applications).
However, EPA needs to support information transfer among States,
to identify good systems now in use for recording such decisions
and make these available to other States.
2) Does the State have adequate hardware and software to
perform permit reviews?
o Example: Air Quality Analysis
Does the State have the most up-to-date EPA Guideline
models?
Does the State have programmable calculators or
computer capability needed to do modeling
Does the State have screening model packages which
would allow them to check source-submitted modeling
3) Does the State have adequate and qualified staff to perform:
o Engineering review (for sources likely to locate in State)
o Air Quality Modeling
Is there any provision for training of rvew staff or existing
staff in new areas?
4) Does the State have clear ground rules to determine:
o Which sources will be modeled or must model;
o When an application is considered "complete";
o What regulations apply to which sources;
-------
111-25
o How exemption requests (if any) will be handled;
o How "special cases" requiring coordination with EPA
or other State agencies (e.g., RCRA hazardous waste
incinerators) wij.1 be handled.
5) Are permit limits enforceable?
a) Legally*
o Identifies all points of emission
o Specifies allowable emissions in mass emissions rate or
concentration terms wherever feasible
o Specifies limits on operating hours/rates where these
are part of an applicability determination or the
result of an impact analysis
o Specifies applicable offsets in the permit and makes
them a condition of source operation
o Provides that permit expires if the source does not
commence construction within a certain time frame
o Specifies whether source or State responsibility for
s amp1i ng
o Provides for reporting of milestone data (e.g., dates
when construction begins, construction complete,
compliance test conducted)
b) Practically
o Specify method of determining compliance for each
emission point and each requirement (e.g., reference
method, GEM, record-keeping if necessary, etc.)
o Preventive requirements if necessary
o Opacity limits if appropriate specify averaging time
o Parameter monitoring specify measures and desireable ranges
6) Results
Ideally, all oversight would use "results" as the measure of
performance. In air pollution control, however, significant
levels of professional judgment must be used in determining
what the proper outcome of any new source permit action is.
Since one of the fundamental principles of delegation is that
'Source: Region VI "Minimum Enforceability Criteria"
-------
111-26
such judgments should be made by State and local programs,
the issue of oversight of outcomes is almost inextricably
tied up with the question of oversight of judgment. There-
fore, the following measures are offered as suggestions,
in the expectation that there will be further discussion
on this point.
o Do State permits require NSPS/NESHAP sources to meet
applicable emissions limits, performance test, and
continuous monitoring requirements?
.o Did the State follow its own rules in making permit
decisions (e.g., regarding public comment)
o Are State BACT/LAER determinations "reasonable"?
o Did the State apply the Guideline models or approve
alternatives properly?
o Are new sources complying at start-up? On a continuous
basis?
Measures and the Exercise of Judgment
These measures, while more "objective" than those currently
in use, still require judgment. This is a necessary and proper
component of evaluation. However, the measures attempt to set
some guidelines for the exercise of this judgment by EPA in order
to avoid the problem of "second guessing." For instance, they
focus more on determining whether a State 1) has the staff and
tools it needs to make certain decisions and 2) has procedures
which make it likely that certain factors will be taken into
account in making these decisions, rather than on attempting to
evaluate the outcome of any particular decision.
To some extent, as was noted in the section on "Results,"
these are "procedural" measures which substitute for direct
measures of the outcomes of specific decisions (as would be
the case if EPA commented on the propriety of an individual BACT
or LAER determination). The problem with "outcome" measures,
again, is that information on (for example) what "BACT" or "LAER"
is for a particular source is not always available. Even when
it is, judgment is an essential part of establishing what these
are for a particular source. EPA can help its own case by
increasing the availability of exemplary determinations,
through the "Clearinghouse" or some other mechanism. Yet if EPA
is to avoid "second guessing," it must allow the judgments made
by States in such cases to stand unless there are flagrant cases
where it is immediately apparent that a State has not applied
BACT or LAER. The one exception to this is for NSPS/NESHAP
sources. Here States agree that the importance of interstate
environmental equity combines with the existence of clear national
standards to make an outcome measure necessary, appropriate, and
practical.
-------
111-27
The point must be reinforced that performance against these
measures is an item which EPA and the State must negotiate and
agree upon in advance. EPA cannot develop performance standards
on its own and apply them without warning to the State. iMoreover,
"zero errors" from EPA's point of view is not an acceptable
standard; EPA cannot expect States to agree with its views 100%
of the time. What EPA can expect is States will make good faith,
professional efforts to comply with the letter and spirit of the
law, and be willing to work with an EPA which sees good performance
by States as a measure of its own success.
Consequences of Oversight
"Consequences" of oversight are generally thought of as
bad things. This should not necessarily be the case. EPA
should note areas where State programs are performing well and
command these in any formal report. Individuals who are doing
good jobs should be cited specifically for special recognition.
Praise is an important oversight consequence and EPA should
remember in giving in that substantive achievements in new source
review, if they are to occur at all, must come primarily from
the efforts of State and local staff.
Problems with State new source review programs may have been
identified through any of the mechanisms suggested above. They
may be of two types a problem with an individual permit or a
flaw with the overall conduct of the State program which is affecting
many or all permits.
Specific Permits; State Views on Consequences States agreed
on some "consequences" of oversight and disagreed on others. For
example, where EPA has a comment on a draft State permit action,
States felt that comment should be expressed during the public
comment period and not afterwards. However, on whether EPA should
retroactively challenge permits, a variety of views existed.
o Pennsylvania believed that EPA could require retroactive
permit changes if it found serious flaws in a State's
determination. This was thought to be unlikely, however.
o Virginia said that permit problems should be brought to
the attention of the State for future reference but should
not be used to make retroactive permit changes.
o Georgia officials felt that the key to EPA's actions in
such case should be the magnitude of effect of the error.
If a NAAQS or increment were violated as a result of an
error, the case should be reopened; if not, it should be
noted for future State reference but no further action
should be taken.
-------
111-28
o Texas, Ohio and Michigan stressed that the State should
have the opportunity to correct any identified problems,
and that any independent EPA action should come only as
a last resort.
o Michigan also sdid that EPA should sue the permittee if
a permit was discovered which was "illegal" and would
adversely affect air quality only after working with
the State. However, EPA should continue to work with the
State even while threatening the court suit, and withdraw
suits which become moot due to State correction of the
problem.
Overall Program; State Views on Consequences States
recommend that more frequent or more extensive audits be used
if a problem which affects the entire program is found. Targeted
EPA technical assistance (e.g., supplying additional hardware
needed for modeling) or training to help correct the problem was
also suggested.
Audit reports have only been used in a few instances. The
States which received them had some complaints about specific
items which they contained. However, they also noted their belief
that their program and the programs of other States had improved
in response to EPA's comments. Ohio has adopt.ed a similar system
of audits and formal reports for use in evaluating its district
offices and local agencies.
Michigan and Ohio felt that the audit reports should be
based on items discussed in an "exit interview" with State
program managers, and subsequent discussions with the State.
Grant conditions have been used primarily to get States to
accept PSD delegation and have not been widely used to follow up
on oversight findings. They have had mixed success. Michigan
reported that it was objecting to EPA's past use of grant
conditions to impose procedural requirements on the State (e.g.,
for permit submissions) and would not agree to such conditions in
the future. Texas, however, noted that some of the grant conditions
imposed by Region VI had been beneficial because they forced the
State to become more explicit and systemmatic about its management
choices (in this case, the subject was inspection priorities).
However, Texas also objected to the vast amount of reporting which
grant conditions required.
Only Ohio reported any use of State-EPA Agreement.s_ for "over-
sight and related" purposes. In this case, Ohio inserted as an
SEA objective its desire that communications improve between the
State and Region V Air Enforcement. Ohio officials were quite
positive about this use of the SEA as a means of communicating
its major concerns to the Regional Office.
Recommendations on Consequences As has been noted earlier,
the most important element of consequences is that, along with
-------
111-29
measures of performance and standards of judgment, they he agreed
upon in advance through discussions with State officials. This
should produce realistic assessments of what can be done and
reasonable agreement on what should occur (e.g., training, technical
assistance) if standards are not met. The emphasis for EPA should
be on how it can help States meet such standards in the future.
While the correction of program deficiencies should take
precedence over individual source problems, it is nevertheless
likely that individual State-issued permits will come up as
disputes from time to time. These are inherently controversial
because they may require major investments by the source. In
these cases we recommend the following:
o EPA should assess the severity of the problem. If it
does not threaten NAAQS or an increment, or is not a
flagrant violation of BACT, it should not force a
retroactive permit change.
o EPA should discuss with .the State whether the problem
is an isolated event or a symptom of a broader concern.
A more extensive audit, focused on other potential examples
of the problem, should be performed if EPA and the State
disagree on this point.
o If a permit change is felt necessary, then the State
should be given every opportunity to make the change, and
discretion about precisely how to implement it (e.g., terms
of a compliance schedule). The State, not EPA, should
negotiate the change; EPA should provide technical backup
to the State if requested and participate in source
negotiations only if requested both parties.
o Direct action by EPA should be taken only as a last resort,
only in extreme cases, and only with the approval of top
Regional management.
-------
111-30
Summary of Recommendations
A long list of recommendations have been made in the course
of Part III. These are summarized here for the convenience of
the reader.
Item
1. Goals of New Source
Review and Oversight
2. Extent of EPA Oversight
3. Tailoring EPA Oversight
4. Mechanisms of Oversight
Recommendations
1. Focus oversight more on achieve-
ment of broad goals than on
exact compliance with every detail
of implementing regulations.
2. Adopt an "EPA the Franchiser"
model; focus EPA actions on areas
where EPA has the greatest compara-
tive advantage, such as improving
pollution control technology,
disseminating information about
new control techniques, and stimu-
lating advances in modeling. Shift
toward building the capability of
State and local programs to handle
problems without EPA intervention.
3. Adapt the extent of oversight and
the methods used to the conditions
of each State program; assume
declining oversight over time
unless problems develop.
4a. Minimize requests for routine State
submissions. In most cases public
" notices should suffice. Further
study is needed on whether EPA
needs a duplicate file of final
State permits.
b. Direct involvement in State permit
actions should be kept to a minimum,
and limited to controversial cases
or cases where a State requests
EPA involvement.
c. Program audits are potentially the
most useful oversight mechanism.
They should be conducted by EPA
staff at the State office, and
include both discussions with
State staff and a review of selected
State permits, with selection criteria
to be discussed in advance with tnj
State. Audit frequency may vary
depending upon the preference of the
State and the results of past reviews.
-------
111-31
5. Objectives and Measures
of Performance
6. Consequences of Oversight
The focus of audits should
generally be on the overall
capability of State programs,
but areas (e.g., selected indus-
trial categories; program compo-
nents) may be targeted for special
attention from year to year.
EPA should prepare formal reports
which discuss audit findings and
specify the criteria used in
evaluating the program.
d. Plant audits offer some oversight
possibilities but difficulties
remain with this option.
5a. Measures and standards of perfor-
mance should be established in
advance with the State. "Zero
errors" should not be explicitly
or implicitly applied as a standard.
b. Suggested measures of results
should be used where available (e.g.,
NSPS sources). Procedural measures
may be a substitute where clear
measures of results cannot be used.
c. EPA should avoid "second-guessing"
judgment decisions made by States.
6a. Positive oversight findings should
be stressed as well as negative
ones.
b. EPA should emphasize positive steps
it can take (e.g., technical support,
training) to build the capability
of State programs in problem areas.
c. EPA'action to correct problems with
specific permits should vary,
depending on the environmental
effect of the problem and whether
it reflects a single incident or
a general problem with the State
program.
d. EPA should give States every oppor-
tunity to correct identified
problems, and intervene only as
a last resort and with the approval
of top Regional management.
-------
PART IV: APPENDICES
-------
APPENDIX 1: Summary of Regional NSR Oversight Approaches
We have surveyed EPA's Regional offices regarding their
present oversight practices for State New Source reviews.
We have also supplemented this survey with site visits to
Regions III, IV, V, and VI. These visits were conducted prior
to our interviews with State officials in each Region.
Existing approaches to Regional NSR oversight range
from a minimal program based almost exclusively on informal
contacts (Region I) to a system of program audits (Region
V), with most Regions relying primarily on informal contacts
and parallel reviews where the Region still has PSD responsi-
bility (See Table I). The attention paid to State reviews
of PSD sources is almost always more intensive than that
given to other components of new source review. Regions
with the most PSD-delegated States tend to have the most
formal oversight approaches, and other Regions are moving to
develop oversight programs under the pressure of approaching
PSD delegation.
WHERE ARE OVERSIGHT APPROACHES SPELLED OUT?
Where formal oversight approaches exist, they may be
spelled out either in a delegation agreement (for PSD/ NSPS/
NESHAPS) or in the State § 105 grant agreement with the
Region. These formal arrangements, however, only detail what the
State will routinely send to EPA. They have not been used
to describe a broader oversight approach and philosophy, or to
establish "standards of performance" for State programs.
While some Regions which do not now have formal oversight
approaches are in the process of developing them in conjunction
with negotiations over the delegation of PSD, only Region IV
has developed a formal statement of NSR oversight policy
which is linked to a larger policy statement governing all
delegated programs (e.g., construction grants, NPDES permits).
-------
-2-
TABLE I: BASIC REGIONAL OVERSIGHT APPROACHES AND EXTENT OF PSD DELEGATION
Reg_ion
Basic Over-
sight
Approach
States/
Region
Extent of
PSD
Defeg_ation
(As of
6/30/81;
does not
count
pending
actions)
I
Little
Over-
sight
6 States
2 PSD
SIP
0 Dele-
gation
II
Post-hoc
review
of about
10% of
all
State
NSR per-
mits
( includ-
ing mi-
nors)
4 States1
0 PSD
SIP
0 Dele-
gation
III
Program
and Per-
mit Au-
dit (PSD
only -
prospec-
tive)
6 States2
1 Full
Deleg .
1 Par-
tial
Deleg.
IV
1) Post-
hoc 10%
permit
review
2) App-
lication
review
for
selected
"criti-
cal"
facili-
ties
3) In-
depth
program
review
(1
state/
year)
8 States
2 PSD
SIP
3 Full
Deleg.
3 Par-
tial
Deleg.
V
Annual
or Bi-
annual
Progam
Audits,
includ-
ing
post-hoc
review
of
selected
permits
6 States
4 Full
Deleg.
2 Par-
tial
Deleg.
VI
Post-hoc
Enforce-
ment
review
of
selected
% of
permits;
percent-
age
around
10% but
less for
Texas
5 States
1 Par-
tial
Deleg.
VII
Review
of con-
trover-
sial
State
NSR per-
mits
4 States
1 Par-
tial
Deleg.
VIII
Review
of all
major
new
source
permits
proposed
by
States
6 States
2 PSD
SIP
IX
Review
of a
small
number
of large
sources
at the
request
of local
agencies
6 States3
0 PSD
SIP
0 Deleg.
X
Permit
reviews
gen-
erally
post-hoc
except
for off-
sets
4 States
0 PSD
SIP
0 Deleg.
-------
-3-
SPECIFIC METHODS NOW USED TO CONDUCT OVERSIGHT
The Regions reported the use of a variety of specific
oversight mechanisms, ranging from informal contacts to
formal program audits.
I n f o rm_al_ _C_ont a c ts
Most Regions rely heavily on informal contacts to both conduct
oversight and encourage States to correct problems uncovered
by oversight. Region I, for example, relies almost exclusively
on informal methods of both oversight and correction, reflecting
relatively low levels of new source activity in that Region.
Region IX has relied heavily on informal contacts because the
resource demands of direct PSD permitting have prevented the
development of any more formal oversight approaches.
Permit Review
Regions with higher levels of new source activity or
more extensive delegation of PSD to States also use informal
methods of oversight and correction, but have adopted other
methods as well. The routine review of permits issued by
States is the most common mechanism, as all Regions other
than Region I noted that at least some such review took
place.
Methods used for permit review, however, varied substan-
tially among the Regions. They differed in four respects:
o Who reviews permits;
o Whether review occurs before or after the State issues
the permit;
o What percentage of permits are selected for review;
and
o How permits are selected for review.
No two Regions do these four things in the same way. Each
Region's review system is a custom-built composite which
reflects the Region's general approach to dealing with State
programs. Table II summarizes the Regional practices in this
area, and they are discussed further below.
On the question of who reviews permits, Regions III, IX and
X reported that Regional Air Enforcement Branches do most of
these reviews at present, though Region III expects this to
-------
-4-
TABLE II: NEW SOURCE PERMIT REVIEW APPROACHES USED BY REGIONAL OFFICES
Region
Method
:> reviews?
a-or post
suance
view?
rcentage
viewed
w selected
I
Little
review
II
Air
Facilities
Branch
Post
10% State
NSR. Expect
to review
all PSD
first year
then de-
clining^ %
Applica-
bility
questions ;
some
selected
at random
III
Air
Enforce-
ment, ex-
cept APB
review w/
PSD dele-
gation
Post
Contro-
versial
IV
Air
Facili-
ties
Branch
Pre:
"critical
facili-
ties"
Post-1 0%
audit
Use 10%
as
guidance
Critical
facili-
ties,
outlined
in policy
state-
ment
plus
random
selection
of 10%
V
Air
Programs
and Air
Enforce-
ment
Pre: APB
if State
asks ;
and PSD
in dele-
gated
States
Post-AEB
routine
All
majors
by AEB;
variable
degree
of re-
view
Audit
team
selects
permits
at ran-
dom or
at State
sugges-
tion
VI
Air
Programs
and Air
Enforce-
ment
Pre:
Contro-
versial
Post:
APB/AEB
10%
guide-
lines;
varies
by State
Selected
at
random
VII
Air
Programs
Pre: Iowa
Post:
Nebraska
Kansas
Missouri
Less
than 5%
Contro-
versial
VIII
Air
Programs
Pre and
Pos_t
100%
majors
N/A
IX
Air
Enforce-
ment
(Permits
Branch)
Post
Infrequent
Controver-
sial or
suspected
applica-
bility
problems
t
X
Air Enforc<
ment (NSPS
Post
10% NSPS, )
all NSPS ti
ed for con)
mance w/de.
tion agreei
All offset!
Controvers:
or select*
at randoi
ote: The u^p^erse of permits reviewed by each Region J^r "oversight" purposes wiil vary,
pending on CTe extent of PSD delegation in the Region
-------
-5-
change as PSD is delegated to the States. Regions V and VI
have more of a joint review process (although with greater
involvement by the Region V Air Programs Branch), while
review responsibilities in Regions VII and VIII appear to
rest primarily with the Air Programs Branch. Reviews in
Regions II and IV are conducted by the Air Facilities Branches.
Regional involvement before or after the State issues a
new/modified source varies. Region IV is the most explicit
in distinguishing permit facilities in which the Region will
get involved before final State action ("critical facilities")
from those which will receive at most a posj^ hoc review (all
others.) Region IV's involvement is particularly distinctive
because they are requesting copies of the facility's application
for joint review by EPA staff. Most other Regions who do
preissuance reviews (e.g., Region V for PSD) get a copy of
a draft permit and supporting analysis from the State.
Involvement prior to permit issuance occurs in other Regions,
but as a matter of practice appears to be limited to permits
for sources which are subject to PSD, offset requirements,
or are otherwise controversial (or where the source is subject
to direct EPA permitting authority under PSD). The Region VI
Air Programs Branch used to do a more extensive review of draft
State NSR permits but has had to abandon this under the pressure
of time and staff resources.
The percentage of State permits reviewed also differs
across Regions. Region VIII reported that it reviews all
major new source permits issued by all States in the Region.
In contrast, Region VI enforcement reviews about 10% of all
(major and minor) NSR permits issued by its States. Region
X also used the 10% figure as a reference point for reviews
of State permits by its Air Enforcement Branch for States
with delegation under NSPS. However, Region X also noted
that ajll State actions under NSPS authority were "tracked to
ensure their conformance with the terms of the delegation
agreement." Region IX reviews only a small number of permits
usually those on which applicability questions exist.
The percent of NSR permits reviewed in Region
VII varies from State to State, depending on the number
of major .or controversial permits in a particular State.
The Region reviews all such major or controversial actions
in each State; these reviews occur before final action in
one State and after final action in the other three. These
reviews result in Regional intervention in State permit
actions in less than 5% of all permits.
Region III expects to use the review of selected
permits in overseeing State programs for PSD when delegation
occurs. These reviews will be post hoc with the percentage
of reviewed permits an item for negotiation between the
-------
-6-
State and Regional office in the course of agreeing upon the
terms of PSD delegation. Region IV is also applying a 10%
post hoc review approach to State NSR permits, with the
provison noted above for joint review of permit applications
for selected "critical facilities."
i
Region II makes the most extensive and formal use of
permit reviews as an oversight method. In 1980, for example,
Region II Air Facilities Branch reviewed about 100 of the
1000 new source permits issued by New York State. (Note:
New York State issues .a separate permit for each emissions
point; the 100 permits consist of emissions points at about
30 facilities). At the same time, the Regional office
conducted 6 PSD permit reviews, since New York has no PSD SIP
and Region II's General Counsel does not believe EPA has the
authority to delegate PSD. About 50-60 of the State new
source permits reviewed by EPA were looked at because of
applicability questions; the remainder of permits for review
are chosen at random from final permit submissions by the
State. AFB staff will review the State determination and
write up and independent assessment, which concludes with the
Regional engineer's judgment of whether or not the permit was
properly issued. Feedback on the results of these audits
is usually given at quarterly meetings of Regional and State
personnel or through informal referrals between Regional and
State staff; the Regional is trying to formalize the referral
process somewhat more this year. Region II has never had a
case where they had to ask the State to revoke an issued
permit, although this has come close to happening on at least
one occasion. Generally, some revision to a "faulty" permit
can be worked out among the Region, the State, and the source.
Region II expects New York State to submit a PSD SIP
soon. When this SIP is approved, the Region expects to audit
all New York PSD permits during the first year, then gradually
reduce the audit percentage over time unless problems are
uncovered in the State program during the first year.
Conversations with several Regional Offices have indicated
that the percentage figures for pjasjt hoc permit audits (e.g.,
10%) are estimates only, and few Regional staff expect to
rigidly adhere to them. Moreover, criteria for selecting
permits for audit (random, targeted, etc.) were not clear, nor
was it clear whether certain parts of the permit action
are or would be reviewed more closely than others. The one
exception to this was the review of emissions limits and permit
conditions for enforceability by Regional Air Enforcement
Branches.
Grant Agreements and SEA1s
Four Regions indicated that they use grant reviews as
an opportunity to either conduct oversight or institute
corrections. Grant agreements, of course cover the entire
range of a State or local air pollution control program, and
-------
-7-
responses do not indicate how effective grant reviews are in
identifying and correcting program deficiencies. They do,
however, offer an opportunity to integrate oversight of New
Source Review with more general concerns about State or
local agency activities. Regions V and VI appear to make
the most extensive formal use of grant agreements to specify,
in delegated States, what documents the States had to submit
to the Regional Office (e.g., public notices, draft permits,
etc.) Table III shows how Region V used grant agreements for
this purpose.
Only three Regions (III, IV, and VI) indicate any use of
State-EPA Agreements for "oversight". Region IV notes that few
air oversight issues "fall into a category that is appropriate
for inclusion in the SEA as that concept is evolving." The
other two Regions indicate the use of SEA's for oversight and
correction, but provided no discussion of just what this means.
A quick review of Region VI's 1981 SEA's, for instance, shows
that no Air issues are included in the Texas SEA, while commit-
ments in other SEA's which relate to NSR focus almost exclusively
on getting States to take on PSD. Commitments in Region III '
SEA's are of a similar nature. Given this, it is premature
to say that SEA's are now being used for NSR oversight.
Program Audits
Six Regions indicated that they either make or are
planning to make some use of program audits in overseeing
various parts of the NSR program. Air Enforcement Branches
in Regions I and X have used these audits to oversee State
enforcement programs for NSPS/NESHAPS sources. Regions III,
IV and VII have proposed such audits for future oversight,
particularly of State PSD programs. Region IV has sponsored
contractor audits of two of its State programs which resulted
in written reports and changes in State management practices.
The Air Programs and Air Enforcement Branches in Region VI
conduct annual audits of State NSR programs which include
site visits and reviews of State permit files.
Region V has made the most extensive use of program
audits in overseeing State NSR performance. Region V has
conducted site visits to all of its State agencies and some
local agencies as well, visits which lasted for two or three
days and involved both reviews of State permit files and
discussions with State staff. The Region conducted exit
interviews with program managers, and prepared formal reports
with findings and recommendations that were sent to each
agency.
-------
-8-
TABLE III: REGION V's USE OF GRANT AGREEMENTS TO
SPECIFY STATE PERMIT SUBMISSION REQUIREMENTS
State Submits
Copies of;
STATE
Michigan
Ohio*
Wisconsin
Indiana
Minnesota
Illinois
Permit
Applications
PSD
permits
No. of NSR/
cations
reviewed
quarterly
(FY 80)
PSD
Staff
Analyses
For PSD permits
and others
ijeijuiLxny a
public comment
period
NSR permits
PSD permits
Public
Notice
_ X
PSD permits
Preliminary
Approvals
NSR
permits
Final
Permits
-
PSD
permits
(quarterly)
NSR permits
/ c- v Q '1 \
(t 1 oj. ;
PSD permits
NSR permits
-external offsets
as SIP revisions
-internal offsets
for info.
PSD permits
(quarterly)
PSD permits
(quarterly)
PSD is not covered in the Ohio grant agreement
The table above is based on an analysis of the conditions contained in the §105 grant agreements
each State in Region V. It shows what each agi^pent requires the State to submit to Region V
either NSR or PSD requirements. "NSR" in this tabl^nrefers only to sources with emissions greater
-------
-9-
The Region V audit reports for 1980 identified a number of
problems with individual State/local programs. Still, with
one exception, the reports indicated that the Region was
generally satisfied with the progress of the State or local
program, and problems and recommendations were presented as
exceptions to be corrected rather than as symptoms of broader
program deficiencies. The one exception was a State report
which contained an unusually long list of problems which the
Region considered serious.
Region V is moving, at the request of its States, to
coordinate the NSR oversight audits conducted by the Air
Programs Branch with other State Program reviews conducted
by Air Enforcement and S&A Branches. While each Branch now
still sends out separate audit team, they will pool their
findings in a combined report on each State agency. These
"comprehensive air agency audits" are now getting underway,
and the Region is now planning to use a combined audit team
in the future. The Region expects these "comprehensive audits"
to replace the §105 grant "mid-year review" which has occurred
in the past.
-------
APPENDIX 2: Assessment of Routine Submissions to EPA
o List of New Source Actions:
Pennsylvania, Virginia and South Carolina submit lists
of permit actions. Pennsylvania submits applications
received; Virginia and South Carolina submit final
actions. All indicated it is not a bother but they
questioned infrequent inquiries on whether PSD would
apply to specific sources.
Possible Purposes;
identification of potential sources requiring EPA
action where program (PSD, NSPS, NESHAP) is not
delegated.
familiarization of EPA staff with state permit
actions in preparation for the program audit.
o Selective Permit Applications;
None of the States interviewed are routinely submitting
copies of permit applications for major sources.
Region IV is instituting its seletive/active oversight
program and has recently requested copies of permit
applications in cases involving interstate impacts, hazardous
waste facility, EIS, energy facility and highly controversial
cases. .
States uniformly thought that permit applications are
most difficult items to send to EPA. They would be of
value, in the State's view, only if EPA.were going to
perform a parallel review which would lead to second
guessing. While States had a problem with permit
applications as a routine submission, States did
indicate that they had no problem providing applications
to EPA on a selective basis in isolated cases where EPA
must get involved (see next section on direct involvement).
Possible Purposes;
basis for active involvement in selective application
reviews
duplicate final permit file in Regional Office to support
enforcement and inspections.
-------
o Draft Public Notice/Permit
Virginia, South Carolina, Georgia, Texas, Ohio and Michigan
have had to submit draft notices and permits. All but
Virginia send for major sources only £/ and also now submit
final permits. '
Two months ago, Region IV changed South Carolina's require-
ment from draft permits to final permits. Texas indicated
that they felt they should discontinue the practice of
sending drafts to EPA when they are operating under full
delegation or their own PSD SIP. Region VI' s Air Programs
Branch indicated that they do not have the staff resources
to be reviewing all of the draft permits for sources other
than PSD and have asked the States to stop sending them.
Region V's Air Enforcement Branch used to review draft
State major source permits but is moving away from this
practice to focus more on the enforceability of final
permits. The Region V Air Programs Branch does still
review draft PSD permits submitted by the States and
comment on some of them.
Possible Purposes;
- basis for active involvement (e.g. formal public
comment on proposed state action).
familiarization of EPA staff with state permit actions
in preparation for the program audit.
o Final Permits
All of the states interviewed except Virginia and Pennsylvania
submit final permits for all majors. Ohio submits all
final new source permits including minors. In addition,
Georgia was recently asked to send to Region IV 10% of
all of its final permits in support of EPA's audit.
Region III enforcement is contemplating requests of its
states to submit final permits in order to have a complete
record of the state's SIP requirements. In Region V final
permits are sent to Enforcement whereas the drafts are
sent to the air program.
All states felt that a final permit is the briefest
document they could send to EPA. They did indicate it
is discouraging because they get no feedback and do not
understand its value to EPA. State staff indicated that
final permits often reference applications, and unless
you were very experienced you would need the entire source
file to understand what is required.
_V Michigan has its own criteria for what is sent to EPA.
-------
Michigan has been asked on several occasions to
provide final permits for all new sources, including
the backlog. They have refused to send the estimated
10,000 permits.
EPA's Regional enforcement offices felt the final
permits were necessary as a record of SIP limits and as
an alternative to having to visit State offices to review
permit files prior to making an inspection. It is
DSSE policy, however, that the Regions and/or their contractors
visit States prior to such an inspection. Region V enforce-
ment staff intend to review final permits for enforceability
problems which they will try to get the States to correct
(e.g. are sources of offsets specified in the permit).
Region VI enforcement staff review about 10% of State new
source permits here for this purpose, using written
"Minimum Criteria for Enforceability" as their standard.
These ."Criteria" are imposed on States as conditions of
the State's S105 grant.
Possible Purposes;
formal record of state final permit limits for:
enforcement files (inspections)
compliance data system
- identification of exemplary permit limits for
the BACT Clearinghouse.
familiarization of EPA staff with state permit
actions in preparation for the program audit.
identification of erroneously issued permits.
-------
APPENDIX 3: State Comments on Direct EPA Involvement
in State Permit Actions
General Comments
All of the States we interviewed believed that EPA should
not get involved in State permit actions as a general rule.
However, most States recognized that some EPA involvement was
probably inevitable in certain cases, and did not have philo-
sophical objections to it as long a_s EPA followed certain rules
in making its comments.
Rule #1; Comment _Within The Public Comment Period Virginia,
Texas, Ohio, and Michigan all stressed that if EPA chose to get
involved in a specific permit action, it must do so within the
State's public comment period. Failure by EPA to do so is a
major cause of friction between EPA and the States. Post-closure
comments can delay (at best).the issuance of a State permit by
forcing the re-opening of the comment period a major headache.
Acceptance of ejc parte comments by EPA can cause administrative
and legal difficulties for the State.
Rule #2; Alert the^ Stat^ in Advance Michigan was critical
of comments on a pending permit action which were transmitted to
the State one day before their governing Commission was to hold a
public hearing on a permit. Ohio expressed concern that EPA reviews
of a permit file might prompt an EPA enforcement action without
prior consultation with the State. Virginia stressed the importance
of prior EPA-State contact so that when comments came there were
"no surprises."
Rule^ _|_31_;;__ Comment Only. Qn Important Issues Georgia suggested
that the key to any EPA action on a State determination with which
EPA disagreed should be the magnitude of the effect of the States
action. If EPA felt a permit action was in error, but this action
would not violate a NAAQS or applicable increment, EPA should note
the error for future discussions in a State program review. If,
on the other hand, a NAAQS or increment were violated, EPA should
intervene in the action.
Sources in Class I Areas
(Note: Most of the States we interviewed had few or no Class I areas)
o Pennsylvania - No extra EPA involvement at the "front end"
of the permit process
o Virginia - No EPA involvement unless the State requests
assistance (e.g., in modeling, in arbitrat-
ing a dispute with a Federal Land Manager or
in dealing with other Federal agencies).
o Georgia - No special EPA attention; target if necessary
for a higher percentage audits
-------
o South Carolina-No special EPA attention
Texas -
Any EPA comments it should be made
within the State public comment period.
Sources With Potential Interstate Impacts
o Pennsylvania- New legislation needed to allow States to
challenge permit actions taken by other
States; §126 has not worked well; greater
direct EPA involvement is not the answer
if EPA cannot resolve the conflict and gets
involved unnecessarily
o Virginia, - The area is important but.EPA should get
Georgia, involved only if asked by States; EPA
Texas, "shouldn't go looking for problems"
South Carolina,
Ohio, Michigan
Permits_ Involving Hazardous Air Pollutants
o Pennsylvania,'
Georgia
o Virginia
o Texas
o Ohio
o Michigan
No special reason for direct EPA
involvement audit both RCRA and air programs
Would like more EPA technical assistance
EPA should assess State capabilities to
ensure needed coordination occurs
Would like more EPA technical assistance;
EPA should ensure that States specify
clear lines of authority for hazardous
waste permitting
EPA involvement in "controversial" actions;
possible "probation" for problem States
Environmental Impact Statement Cases
o Pennsylvania -
o Michigan
EPA should find a way to delegate EIS
review to State agencies
EIS should not produce EPA involvement
in specific permit actions, especially
since air permits often follow an EIS due
to lack of detailed information on plant design
EIS review should be delegable since
Michigan has an equivalent EIS requirement.
-------
APPENDIX 4: Sjjm_m_ary_ o f _S_ta_ t e V i e_ws_ o n E PA Prog ram A u d its
Pennsylvania
o Program reviews are generally desireable on an annual
or semi-annual basis.
o _Audits would include a review of a percentage of State
permits/ both majors and minors.
o Audits should replace day to day involvement.
o Audits should be conducted by staff who are familiar
with the State program and its strengths and weaknesses.
o EPA could request items which a permit file should contain
without requiring it.
o Criteria for evaluation should include
- Whether the State has been following its own rules
- Whether NSPS/NESHAPS are met
They should not focus on management issues (e.g., permit
processing times).
o EPA could challenge "serious" permit flaws but need for
this is unlikely.
o Audits could be tailored to the "adequacy" of each State
program.
Virginia
o Suggest focus on specific source categories where EPA
has special concern and expertise; source categories
could vary from year to year.
o The tone and conduct of "audits" is critical; suggest
calling them "liaison visits."
o Saw audit as an opportunity to learn what other States
are doing.
o They are trying now to better organize their own files
and documentation.
o Audit problems should be brought to the State's attention
but not be used to make retroactive permit changes.
-------
3
Georgia
Audits could occur as often as "three or four times
per year."
Regional staff,' in doing permit reviews, should both
look at the files and talk to responsible State staff.
Criteria for evaluation should be 1) whether individual
source permits conformed with applicable State and
Federal standards, and 2) whether State reviews were
adequately protecting NAAQS.
EPA oversight activity should increase as its item by
item responsibilities decline.
Oversight must and should vary from State to State,
depending on the State's capabilities and performance.
Delegation should not mean that EPA and the State must
agree on every decision, and oversight must reflect this.
Oversight must focus on majpjr issues, not on every single
detail.
South Carolina
Texas
Existing semi-annual program reviews are very useful,
but the mid-year review of the grant agreement is a
"bean count."
The percentage of major source permits reviewed should
be higher at first and decline as the State demonstrates
its ability to perform the reviews.
It is fair for EPA to ask for auditable files; doing this
will require additional effort by the State.
Once basic State capability is established, EPA audits
should focus on 1) items which affect economic competition
among States and 2) aspects of permits for which national
uniformity is essential.
Texas agrees with the concept of oversight as long as
it is "visible" but not "persuasive".
V
State staff suggested two audits per year, one to be sent
to Washington, and one solely for Regional office use.
-------
Audits should include the examination of State files for
a few selected permits and discussions with State staff
to suggest how reviews might be done better in the future.
EPA should not expect to review every PSD or other permit
action.
EPA must not take an action against a source as a result
of an audit without first giving the State a chance to act.
If EPA concerns about a specific judgment or interpretation
are so strong that EPA feels it must exercise control over
it, the interpretation should be published as a formal
requirement governing the program.
The general system of annual site visits and permit
reviews now used by Region V is satisfactory.
Ohio would like all air. program reviews consolidated
into a single audit.
States must be given a chance to act on audit findings
before EPA enforcement action is taken.
Oversight should decrease under a PSD SIP (as opposed
to delegation).
Mi_chic[_an
o Present EPA audits have tended to result in "nitpicking"
comments. However, Michigan believes that other States
have improved their programs in response to EPA audits.
o Audit staff must be knowledgeable about the State
program or much time will be wasted.
o EPA should treat State oversight like a supervisor
and an employee.
Surface problems when controversy exists
Do not reverse a State decision except in extreme cases
Do not expect "zero errors"
o If serious problems in State administration are uncovered,
a more extensive audit should be performed of actions in
the problem area.
o EPA should bring court suits against a source for an illegal
permit only if problems cannot otherwise be corrected by the
State. Even if a court suit is threatened, EPA should
combine to work with the State.
-------
o EPA grant conditions were not especially successful in
imposing changes on the State program.
o Minor source permits are a legitimate focus for EPA oversight,
o Final audit reports on State programs should be public;
this is "good advertising" for a good State program.
-------
Appendix 5: Present Implementation of New Source Review and
Oversight
As part of our site visits to States, we asked questions
about the present status of State implementation of new source
review. The following appendix discusses implementation in five
areas:
o Applicability Determinations
o Control Technology Determninations
o Air Quality Analysis
o Permit Review for Enforceability
o Public Participation.
o Operating Procedures and Management
Both actual implementation by States and present oversight
by EPA in each area are discussed, along with future EPA oversight'
options for each activity. These options specifically apply the
oversight mechanisms discussed in Part III of this report.
-------
-2-
APPLICABILITY DETERMINATIONS
State/EPA Implementation
o Applicability questions come up primarily only for
sources subject to PSD, offset, or NSPS requirements.
The overwhelming majority of a State's new sources
are not subject to specific Federal requirements.
o There is little focus on "marginal cases" e.g., only
one State we visited systematically tracks minor
emissions increases/decreases
o Modifications cause the greatest difficulty in determining
applicability
EPA Oversight
o Regions II and V have used audits of State permits to
check applicability determinations. Region II in
particular has reviewed both major and minor source
permits for these purposes.
o If a State is in doubt on an applicability question,
they usually call EPA and request a ruling.
o EPA has reviewed applications in "controversial"
cases such as:
Where there appeared to be a need for an EPA-issued
PSD permit
- Where questions arose about whether a source was
subject to the construction ban.
Cases of significant public interest.
EPA Oversight Potential Problems
o Sources which are in fact "majors" may erroneously
qualify as "minors" and be subject to less stringent
requirements.
o Sources which should require offsets because of their
geographic location in a non-attainment area may escape
this requirement.
o Sources may avoid new source review altogether. Michigan
estimated that 70% of source apply for a new source permit
prior to construction, 20% apply after the installation of
equipment, and 10% are caught in violation, having failed
to apply. They noted that after one push to enforce appli-
cability, in which two major sources were fined $8000 each
-------
-3-
for construction without a permit, the State subsequently
received "hundreds of applications" from other sources
which had built without permits. They felt that other
non-permitted sources might exist in the State, but
felt that most such sources were minors.
State Comments/Concerns
o States generally endorsed EPA involvement in applicability
determination to the extent it was needed to ensure
national consistency.
o States suggested or endorsed the inclusion of some
"minor source" permits in an audit review for
applicability
o States urged that present applicability requirements
be simplified
o States asked that if EPA uncovers applicability
problems, that
a) EPA assess the magnitude of impact of the problem
before pursuing further action, as the cost of
remedial action can be quite high in specific instances.
b) EPA consult with State officials before taking
independent enforcement action
Future Oversight Options
o EPA permit audits could include a sample of "minor"
or "borderline major" permits to check State
applicability procedures
o EPA could focus on the State's process for making
applicability determinations (e.g., tracking of de minimis
changes; use of standard operating procedures);
o EPA could use "plant audits" inspections of selected
facilities to check whether the emissions points and
operational limits identified on a State's permit (e.g.,
capacity or operating hour restrictions) correspond to
those present at the plant site.
o EPA could focus its oversight exclusively on sources
determined by the State to be "major" and accept State
applicability determinations on minors as final
To successfully implement any of these will require
o Simplified applicability rules
o A clear checklist of applicability consideration which
States could use in reviewing applications from, major sources
-------
-4-
CONTROL TECHNOLOGY DETERMINATION
State/EPA Implementation
o No case by case'determination
BACT/LAER usually equals NSPS where available; if
no NSPS, States generally have standard "rules of
thumb" for defining BACT, particularly for small
boilers. Most State determinations are routine.
o "BACT" is Chief State goal
The States we visited had their own "BACT" requirements
for new sources. One case (Texas) explicitly requires
"BACT" for each emissions point, regardless of "netting."
o States want better control guidance from EPA
States now look for guidance on what control technology
to use in unusual cases from wherever they can find it.
NSPS, CTG, staff in other States, and OAQPS staff are
all helpful in certain cases, but State staff felt EPA
could do a better job of developing and disseminating
information.
o States do not want to be tagged as "requiring more than
the national minimum"
Many State agencies felt under pressure from industry
and elected officials to not go beyond "the national
minimum" in making BACT/LAER determinations. While
States do sometimes succeed in getting sources to use
more stringent controls than Federal requirements, four
of the States we visited indicated explicitly that they
were reluctant to have this publicized.
EPA Oversight; Current Mechanisms
o Regions II and V have reviewed State BACT/LAER determina-
tions as part of their program audits. Region V has
criticized LAER determinations in 3 of its State audits.
o In non delegated States, EPA has used its own PSD review
to impose its interpretation of."BACT" on sources.
o EPA has established a "BACT Clearinghouse" to catalogue .
information on "good" source controls and spread this
among States. Its value and use to States is limited,
however.
-------
-5-
EPA Oversight: Performance Measures
o We have used NSPS as "baseline" for control where they
exist, and have not pushed sources to go beyond NSPS.
We have no performance measures for non-MSPS sources.
o .Audits in Regions II and V have criticized States for
. failure to document how control technology determina-
tions were made.
State Comments/Concerns
o Importance of NSPS
States use NSPS as the basis for setting standards for
.new sources to which .NSPS applies (and some related
sources). All States strongly endorsed NSPS and urged
their continued development. There was particular
interest in the rapid promulgation of the NSPS for
industrial boilers. Texas and Ohio objected to the
proposal for NSPS offsets.
o Interstate Equity
Pennsylvania, Georgia, and Ohio raised concerns about
equity among States in making BACT determinations.
o Improving Upon NSPS
Georgia and Ohio both indicated their belief that existing
NSPS for particulate matter from boilers could be improved
upon. Texas emphasized their desire to improve upon NSPS
in general, and Pennsylvania noted that NSPS sometimes
functioned as a "ceiling" which limited opportunities
for further control.
o BACT Clearinghouse
States generally supported upgrading and improving the
BACT Clearinghouse. Virginia and Ohio specifically want
a Clearinghouse with a data base which they can tie into
via their computer. However, several States raised caveats
Pennsylvania, Virginia, Texas, and Ohio are concerned
about the use of a Clearinghouse by industry to
criticize them as "overly aggressive" States.
No State was willing to invest many resources in
reporting to the Clearinghouse.
The greatest need for Clearinghouse data is for sources
not covered by NSPS, whereas most Clearinghouse reports
now are for NSPS sources.
-------
-6-
Sfcates disagreed on the level, extent, and type of
data which the Clearinghouse needs. Virginia, for
example, wanted a "range" of determinations for a
source category, while Texas wanted only "exemplary"
determinations.
Options for EPA Oversight
o Continue to promulgate NSPS but do no active review of
State BACT/LAER determinations.
o Target NSPS sources and review selected final permits
for such sources as part of program audit. Offer no
comments unless NSPS are not met.
o Target non-NSPS major sources and review selected final
permits for such sources to examine the State's compliance
with its own rules in making control determinations.
o Request draft State permits and technical support documents
for selected sources after receipt of a notice of a proposed
permit action from the State. Respond informally or formally
within the State's public comment period.
o Improve the dissemination of information through the BACT
Clearinghouse after consultation with State officials to
determine what information is most needed and how it should
be provided.
-------
-7-
AIR QUALITY ANALYSIS
State/EPA Implementation
o Different Effects on Major and Minor Sources
With the exception of Michigan, States felt that air
quality analysis has little effect on the siting, of
a source or the emissions limits which a "major" source
inust meet. They noted, however, that modeling does affect
plant stack heights for minor sources (e.g., asphalt
batch plants in Ohio). One State estimated that this
effect occurred for greater than 50% of minor sources.
Generally, these were cases where downwash conditions
produced local threats to NAAQS.
o Cause of State-EPA Contention
Modeling results are a source of friction between EPA
and States. Topics of discussion include:
- What models are approvable
What models are appropriate for a specific site
What meteorogical input is required
What emissions assumptions to use (e.g., hours of operation)
Cost of running EPA models
- Receptor spacing
o Few Sources Do Independent Monitoring
States do not generally require sources to conduct
extensive pre-construction monitoring, but try to work
with the source to use available data. In Texas, however,
sources are establishing consortia which will operate
monitoring networks for PSD application purposes.
o Many States Do Some Modeling
All the States we visited do some screening modeling
on minor sources to protect against local NAAQS violations.
EPA Oversight: Current Mechanisms
o Air quality analysis has only been an issue for PSD
permits in delegated States a relatively small number.
o EPA Region V has required some of their States to improve
their modeling capability (e.g., hire a modeler) as a
grant condition.
o Region V criticized State modeling practices in all
six of their 1980 audits though the type of practice
criticized varied from State to State. Examples included:
-------
Relying on EPA-derived background concentrations
Use of one year of meteorological data instead
of five years
- Use of rural dispersion coefficients in an urban area.
EPA Oversight: Performance Measures
o We criticized States who fail to use Guideline models
or apply the precise parameters specified in the
Guideline (e.g., 5 years of meteorological data;
spacing receptors at 100 meter intervals).
o We require States to get Regional approval before
allowing sources to use alternative models.
o We assess the capabilities of the modelers on State
staffs.
State Comments/Concerns
o Model Interpretation Where Experts Can Disagree
All the States we interviewed see modeling as an area
involving significant issues of judgment. Where these
issues arise, they feel that State modelers should
make these judgments and EPA should refrain from
"second-guessing" them (e.g., on elevated terrain impacts).
State modelers will accept some limited variations in a
source-provided model if they believe that these variations
will not affect the model's outcome.
o EPA Models Are Expensive
South Carolina, Texas, and Michigan all criticized EPA's
Guideline models as unnecessarily expensive and complex
to run. They argued that the models are "research
oriented, not user oriented."
o Approving Alternative Models Is Difficult
Michigan commented that an "ominous" amount of work would
be required to get an alternative model approved. Texas
felt that they could get an alternative model approved by
the Regional office, but they were not satisfied with this
and wanted Headquarters approval from OAQPS. This issue
in Texas had held up PSD delegation.
o Modeling for PSD
Michigan indicated that what made PSD modeling so difficult
was having to track and factor in the growth of surrounding
sources. The site-specific modeling itself would not
otherwise be so complex. A further complicating factor,
-------
-9-
however, was constant changes in EPA requirements for
pollutants, actual/allowable emissions, baseline/input
areas, and baseline dates.
Options for EPA Oversight
o Retain the Guideline on Air Quality Models, but do not
review specific air quality analyses. Establish only
that the State has qualified modelers.
o Formally increase State discretion in applying new
modeling techniques (e.g., eliminate the regulatory
status of the Guideline).
o Review State modeling as part of program audit, including
review of selected PSD permit files.
o Focus on State procedures for pre-application review
of source modeling/monitoring plans.
-------
-10-
PERMIT REVIEW FOR ENFORCEABILITY
State/EPA Implementation
o All States Use '.'Boilerplate" and Conditions
Every State program has a "boilerplate" list of general
conditions with which all new/modified sources must comply.
These usually covered such provisions as reporting of
malfunctions, compliance with orders of the Agency,
notification' of start-up, or revocation of the permit
if construction failed to begin within a certain time.
States add "special" conditions to this "boilerplate"
as appropriate. The frequency of special conditions
varies; Pennsylvania said about 50% of its permits
-have them, while Michigan reports that almost all of
their permits contain special conditions. Usually,
the larger the source, the more detailed the permit.
o Applications Are Enforceable By States
All States regard sources as subject to conditions or
claims made in their permit applications. Pennsylvania
noted a conflict with EPA on this, as they said EPA
"wants everything written in the permit document to
make it Federally enforceable."
o Permit Denial Is Rare
Data on new/modified source permit denials was available
in three States. Virginia has denied one permit in ten
years over a land use question. Georgia initially
denied less than 5% of its applications in 1980; upon
submission of further information by the applicant, all
of these were subsequently approved. Ohio denied four
applications for permits to install and approved 633 in
1980.
EPA Oversight; Current Mechanisms
o EPA Region VI has established "General Minimum Criteria
for Enforceability" (see attached copy) for all new source
permits in Region VI States, and has imposed a grant
condition on States that all their permits must meet
them. The Regional Air Enforcement .Branch reviews about
10% of final State permits against these criteria.
o Region V reviews permits for enforceability in the course
of"its program audits. The Region commented on enforce-
ability problems with State permits in five of its six
1980 audit reports. Enforceablity problems were particularly
noted in specifying emissions offsets as a condition of
-------
-li-
the applicant's permit. Other enforceability issues
involved failures to specify emissions limits in
permits, or failure to include operating hour restric-
tions as a permit conditions.
o Region II has commented on occasion when a permit appears
to contain a flaw with respect to legal enforceability.
EPA Oversight; Performance Measures
o Region VI's "General Minimum Criteria" is the most explicit
list of what an "enforceable" permit should contain.
State Comments/Concerns
p States generally believe that they issue legally
enforceable permits. States acknowledged difficulties,
however, with the practical enforceability of permits.
o Ohio and Pennsylvania follow their permit to install
with a renewable operating permit requirement. Ohio
felt strongly that the renewable operating permit gave
them additional leverage over so.urce behavior.
o Georgia and Texas require a source to get a non-renewable
permit to operate following the completion of construction.
Texas requires a source to submit stack test data before
it can be issued a permit to operate.
o Opacity requirements are key enforcement tools in most
States. In Ohio, where opacity enforcement is somewhat
restricted by State court decisions, their enforcement
program has been more cumbersome as a result.
o CEM and parameter monitoring are only in very limited use.
Options for EPA Oversight
o Review State "boilerplate" against minimum "enforceability
criteria."
o Review a sample of State final permits for consistency
with enforceability criteria.
-------
VI
. GENEP.U MINJHU?'. CRITERIA FCi-i DETERMINING ENFORCEA8IL1TY
OF NEW SOURCE PERMITS
An enforceable permit shall contain as a minimum:
1. Idcrifi fi cation of all points of emission (both stack and fugitive).
2. Specification of an allowable emission rate for each point of emission in
terms of mass rate or concentration limitations. If the existing SIP
regulations provide the applicable emission limitation, the appropriate
section of the regulation^ may be referenced in lieu of reiterating the
1 imitation. If emission testing based on a numerical emission linntation
is infeasible , .the permit may instead prescribe a design, operational,
cr equipment stsnnard. Any permits issued v/ithout numerical emission
limitations must contain conditions which assure that the design character-
istics or equipment will be properly maintained or that the operational
conditions v/ill be properly performed, so as to continuously achieve the
assumed degree of control. .
3. Limitations on factors which v/ere basis for air quality impact analysis
'must be specified (e.g., hours of operation, stack height, materials
processed v/hich affect emissions).
s
4. Methods of determining compliance for each point of emission must be
refersnccd (if part of the SIP) or explicitly identified if a reference
method is not .used.
5.- Where- continued source compliance is dependent on process variations or
events which can be reasonably anticipated, preventative requirements must
be specified (such as stockpiling of low sulfur coal and fuel specifications,
such as ash content and sulfur content).
6. Record-keeping requirements v/hich enable the agency to ascertain continued
compliance especially where factors such as hours of operation, through-put
of materials , type or quantity of materials processed are conditions of the
permit.
7. A condition that the permit will expire if the construction is not commenced
within- certain specified time frame.
r
8. The condition that the source is responsible for providing sampling and
testing facilities at its own expense. . .
9. Reporting requirements which enable the agency to monitor the progress of
source construction and compliance, including the date by which construction
is completed, and if different from the completion of construction date,
the date by which full compliance is to be achieved.
IV-28
-------
-13-
PUBLIC PARTICIPATION
State/EPA Implementation
o Public participation plays no significant role in
State permit decisions.
o The few cases where significant public participation
occurs are those which involve significant land use
questions.
o Public comment periods account for a significant portion
of total processing time for most permits.
o States take the procedural requirements seriously
but they have little substantive affect.
EPA Oversight; Current Mechanisms
o EPA has paid very little attention to State public
participation programs.
o Region V noted deficiencies in the public notice and
comment provisions in two of its six 1980 audits.
o EPA conducts public participation for non-delegated
PSD permits.
EPA Oversight; Performance Measures
o The procedural requirements governing PSD sources
(40 CFR 51.24 (g)) and other new/modified sources
(51.18(g)) differ. EPA has generally required States
to adhere in each case to the procedures governing
the separate programs, with the result that delegated
States have one set of rules for PSD permits, and a
different set for everything else. However, one State
noted that EPA has allowed de facto exceptions to these
requirements by tacit agreement to "look the other way"
at different State practices.
State Comments/Concerns
o No State we visited receives any public comment on
95% or more of its permits. When comments occur,
they concern road noise, traffic disruption, or
general "quality of life" or property value concerns.
o Scorecard on public participation in new source permit
decisions
Pennsylvania Two public hearings in five years
- Virginia No comments, no one shows up for
the hearing in 95% of cases
-------
-14-
Georgia- Of 21 PSD permits in 1980, only
1 required a public hearing
- Texas Receive public comment on less than
5% of all applications for large
sources to construct or modify
- Ohio Out of 600-700 permits to install
issued each year, they receive comments
on 12-18
Options for EPA Oversight
o No oversight of State public participation requirements.
o Continue to receive notices of public comment for
selected najor sources.
o Examination of State public participation program and
procedures in the course of program audit.
o Review quality and content of State summary analyses
or prelininary determinations which are the basis for
public review and comment.
-------
-15-
OPERATING PROCEDURES AND MANAGEMENT
State/EPA Implementation
While EPA generally has paid little attention to State
operating procedures, EPA has on occasion criticized States for:
o Failure to notify a source about whether its appli-
cation is complete or not;
o Failure to properly document and record the basis for
decisions; and
o Failure to record de minimis changes in sources which
could accumulate past "major modification" thresholds.
States have developed their own mechanisms for filing permits,
tracking their status, and documenting decisions. While some
States felt their present procedures were adequate, other States
indicated that they saw needs for improvement. Only Michigan
systematically tracks de minimis emissions changes.
States with PSD authority felt that their acceptance of PSD
dramatically cut permit processing times. Georgia cited a
90 day processing time, and Virginia expected to cut processing
times at least in half (to twelve weeks). States saw this as a
major advantage which they could offer sources subject to PSD.
EPA Oversight Current Mechanisms
o Region V has raised operating procedure issues in its
audit reports.
o The two contract audits performed for Region IV both
criticized State operating procedures and filing
practices.
EPA Oversight Potential Problems
o Under present regulations, failure to keep accurate records
of de minimis emissions changes could result in some
sources erroneously avoiding PSD review.
o Failure to maintain adequate files would render any EPA
audit program unworkable.
State Comments/Concerns
o EPA should not worry about State permit processing times.
-------
-16-
o States generally agree that EPA should specify, in some
form, what an "auditable" permit should contain.
Pennsylvania, Virginia and Texas specifically said,
however, that this should not take the form of a
regulation on permit files. Texas recommended that
SPA provide examples of what it would like to see in
an "auditable" file.
Future Oversight Options
o Use early audits to test the adequacy of a State's filing
system for audit purposes.
o EPA should develop a proposal which specifies what an
"auditable" file for a major source should contain,
and'identify samples (preferably State-conducted reviews)
and State systems which contain the elements necessary
for a auditable file system.
o EPA should stimulate information transfer among States on
recordkeeping and filing systems, and recommend that States
with problems in these areas visit other "successful"
States to study their systems.
-------
CRS-25
It is not ac all clear chat section 176(a) authorizes the withholding
sanction when a State refuses to submit revisions to an initially fully
aooroved Part D revision in response to an EPA finding that the initially
68/
revised SIP has failed to produce attainment by December 31, 1982.
There is colorable argument that the section 176(a) sanction was intended
to apply only to the first round of revisions required in nonextension
69/
areas by January 1, 1979. (The same argument applies in the event of
post-1987 nonattainment in extension areas.) Discussion.of the cut-off
sanction in the 1977 legislative history appears to revolve exclusively
around that initial revision, rather than later ones necessitated by
TO/ ,
post-1982, nonattainment. And ?gain there are sections 110(a)(2)(H) and
110(c)(l)(C), a revision mechanism more clearly applicable than 176(a) to
SIPS that prove to be. inadequate to produce attainment.
A "results'* interpretation, as has been plausibly made in connection
with the construction ban, is difficult to support for the funds sanction.
To reiterate, this view would hold that the simple, unadorned fact of
post-1982 nonattainment is enough to subject an AQCR to withholding of-
funds. A State with nonattainment areas, begins- this argument, oust sub-
mit a plan that considers each of the elements in section 172, one of
which is attainment by December 31, 1982 in nonextension areas. If non-
attainment persists beyond that date, then the State has not submitted a
68/ There is no indication at present how many States would fall into
this category. Presumably, States that were disposed to submit approvable
Part D revisions initially might be similarly inclined as to a second-round
revision.
69/ The January 1, 1979 deadline for submittal of the first Part D re-
vision is found in section 129(c) of the 1977 amendments, Pub. Law 95-95,
but was not codified in the Clean Air Act itself.
70/ Legislative History, supra note 31, at 536, 1061, and 1383.
------- |