Comments of the Coalition for Clean Air Implementation 

on EPA’s August 7, 2000 Review Draft of White Paper 3 

“Design of Flexible Air Permits” 

Introduction And Summary 

On August 7, 2000 the Environmental Protection Agency released for
comment a draft of the latest in its series of Title V permit guidance
“White Papers”, White Paper 3.  That draft White Paper focuses
exclusively on ways in which Title V permits can be written to avoid the
need for later permit revisions as industrial operations change.  The
White Paper identifies three such approaches and discusses them in
detail.  They are:

Writing permits that would eliminate unnecessary detail not needed to
comply with “applicable requirements”.

Describing, and approving in advance, alternative operating scenarios
that would become “pre-permitted”.  A source could switch between
these pre-approved scenarios simply by giving notice. 

Establishing “caps” that would prevent changes at a facility from
triggering “major new source review” (“MNSR”) under the Clean
Air Act. 

The Coalition for Clean Air Implementation ("CCAI) is an industry group
focused on implementation of the Clean Air Act.  Our members are the
American Forest and Paper Association (AF&PA), the American Iron and
Steel Institute (AISI), the American Petroleum Institute (API), and the
National Mining Association (NMA).

The members of CCAI’s component organizations own and operate numerous
sources subject to Title V requirements.  Accordingly, CCAI has
participated in the development and revision of Title V regulations from
the beginning.  CCAI has commented on every Title V regulatory proposal
and every draft White Paper, and has participated in numerous
“stakeholder” meetings on Title V.

CCAI supports the purpose of White Paper 3 and many of its specific
positions.  In general, we believe the White Paper is constructive and
thoughtful and would improve the operations of the Title V program.  The
fact that comments are, by nature, critical should not obscure our
overall position.  However, we also believe the White Paper could
benefit from a broader focus and eliminating some excessively detailed
recommendations. In particular: 

The whole tone of the White Paper discussion would encourage –
probably inadvertently – the multiplication of the paperwork
requirements and overly complex procedures that hamper the Title V
program at present. That discussion  tends to presents a
paperwork-intensive solution for every real or hypothetical problem.
CCAI does not object to paperwork solutions where they are the only way
to address real problems.  However, there is little recognition in the
White Paper draft that the procedures and accountability mechanisms of
Part 70 would, in actual operation, already provide protection against
many of the problems identified. 

In many cases the operating flexibility at which the White Paper aims
could be achieved more efficiently by streamlined permit amendment
procedures than by writing permits to avoid amendment.  The White Paper
should expressly acknowledge that fact. 

The draft White Paper would require every detail of State “minor New
Source review” (“mNSR”) permits to be reflected in a
“streamlined” Title V permit, regardless of the environmental need
for that detail.  This position, if adopted in final form, would reverse
the position of White Paper 1.  This level of detail is in no way
required by the Clean Air Act.  It would make streamlining permits far
more difficult.

The draft White Paper (perhaps inadvertently) suggests that a “cap”
must be included in all permits with flexible operating scenarios.  CCAI
believes this is clearly not the case, and that any such requirement
would make flexible permitting unworkable.

The discussion of the relationship between Title V permits and MNSR is
confusing.  We have a number of suggestions for clarifying it. 

Finally, in some respect the White Paper addresses topics beyond EPA’s
authority, or is too prescriptive for a guidance document.  This is
particularly true of the provisions that address compliance monitoring.

Our full discussion follows. 

Discussion 

Flexibility and Permit Amendments

A glance at the White Paper shows how complicated the drafting of
“alternate operating scenarios” could become.  At the meeting EPA
held last June on a leaked draft of White Paper 3, John Paul of Dayton
said that his agency simply did not have the resources to review
intricate sets of alternate scenarios for preapproval.  He urged EPA
instead to adopt flexible and workable permit amendment procedures. 
That would make the permit amendment process more attractive and remove
any need for elaborate efforts to avoid it. 

CCAI fully agrees.  We know that addressing these amendment procedures
is not the purpose of White Paper 3.  However, the complexity of some of
the scenarios outlined for avoiding permit amendments itself shows the
need for workable permit amendment procedures that would reduce the need
for such complicated alternative approaches.

“Smart Permits” and mNSR

The White Paper strongly recommends the use of “smart permits”.  A
“smart permit” is a permit that “does not prevent the source from
operating as flexibly as the applicable requirement(s) allow by
including restrictive terms in the permit that are not included in the
applicable requirement(s).” p. 7. It adds that “[t]he first two
White Papers on the operating permit program describe many smart
permitting techniques”.

However, in one key respect White Paper 3 would withdraw much of the
flexibility to write “smart permits” created by White Paper 1. 
State mNSR permits are by far the largest single source of “applicable
requirements” for many Title V permits.  Yet many of those permits are
obsolete or duplicative or have little relation to air quality. 

White Paper 1 recognized this and said (p. 12)

The Agency has concluded ... that only environmentally significant terms
need to be included in part 70 permits.  The EPA recognizes that NSR
permits contain terms that are obsolete, extraneous, environmentally
insignificant, or otherwise not required as part of the SIP or a
federally enforceable NSR program.  Such terms ... need not be
incorporated into the part 70 permit.

Minor NSR, in particular, is a program which the State has discretion to
mold as necessary to be consistent with the goals of the SIP. 
Therefore, the permitting authority has very broad discretion in
determining the terms of minor NSR. 

	However, White Paper 3 as currently drafted would specifically require
all mNSR permit terms to be included in the Title V permit, subject only
to a limited authority to amend them in a way that did not affect their
substantive stringency. pp. 6-7.  Such an approach would only perpetuate
a workload that greatly impedes Title V implementation with no
environmental benefit.  (Indeed, the workload creates an environmental
detriment in that it drains resources from more useful tasks.)  We urge
EPA to reaffirm the message of White Paper 1 that obsolete and
environmentally marginal mNSR permits may be omitted from the Title V
permit. 

We also urge EPA to make clear in White Paper 3 that if a State finds
that its mNSR permit program is generating too many permits of marginal
environmental usefulness, it has “broad discretion”, to quote White
Paper 1, to amend that program to reduce its burdens as long as the
basic mNSR function of protecting air quality is preserved.  EPA should
encourage States to interpret or amend unreasonably stringent or
obsolete mNSR programs to reduce their negative impacts.  That would
greatly increase the ability of States to write “smart permits”, and
would be of unquestionable legality under the Clean Air Act. 

Alternate Scenarios and Emission Caps 

Where “smart permits” do not provide enough flexibility, the White
Paper suggests (pp. 10-21) the use of “advance approval” of
different operating scenarios that the source could then switch between
simply by giving notice.  The discussion repeatedly suggests that such
permit terms must include limits on the changes that can be approved in
advance, or even an emissions cap. See, e.g. pp. 2 (‘”[a]dvance
approvals ... typically include one or more emissions limits(known as
emissions ‘caps’”), 4 (“advance approvals routinely require the
establishment of one or more emissions caps which act to freeze
emissions and overall capacity use at current levels.”) 10 (“a ...
permit containing an advance approval must include a description of the
advance approved changes and a limitation on their magnitude”), 13
(“some limitation [is required and] can be accomplished by either an
emissions cap or a numerical limit on the expected amount of certain
types or combinations of new emissions units”), p. 18 (“[e]ither an
emissions cap or numerical limit on the amount of expected changes that
can occur under the advance approval [is needed for the advance approval
to be acceptable.]”). 

No blanket requirement for such a limit can be justified, except to the
extent that a limit might be needed to protect local air quality.  Where
the new units to be added are subject only to a technology requirement
like an NSPS or MACT standard, it should be enough to describe those
requirements and provide for compliance with them.  The applicability of
such a standard would not justify any limit on the number of units. As
EPA said in White Paper 1 (p. 1)

[O]perating permits required by Title V are meant to accomplish the
largely procedural task of identifying and recording existing
substantive requirements applicable to regulated sources and to assure
compliance with these existing requirements.  Accordingly, operating
permits and their accompanying applications should be vehicles for
defining existing compliance obligations rather than for imposing new
requirements or accomplishing other objectives. 

If enough units were added under a flexible permit, the new emissions
might trigger MNSR.  But that does not justify requiring a cap to avoid
that result.  MNSR in this case would not be a currently applicable
requirement, and so it would be the source’s decision whether to seek
such a cap.  Indeed White Paper 3 (p. 17) expressly endorses the
conclusion of White Paper 1 that ton per year emissions estimates need
not be included in a permit at all “where they would serve no useful
purpose, where a quantifiable emissions rate is not applicable, or where
emissions units are subject to a generic requirement.”  Clearly, if a
permit need not even contain estimates of overall annual emissions, it
need not contain an annual cap on them.

Even if EPA were to respond that a limitation on future activities was
required to make sure that the changes approved in advance were
“reasonably anticipated”, that would at most justify a generic
qualitative description, not an emissions cap.  Any such cap requirement
would virtually assure that the advance approval mechanism would be
little used.  Moreover, it would serve little to no practical purpose.

That is clear from the White Paper’s own list (p. 6) of permitting
issues that are likely candidates for being addressed by “advance
approval”.  It runs as follows:

(1)	adding a new emissions unit or new component equipment;

(2)	reconstructing an emissions unit; 

(3)	modifying emissions units (e.g., redesigning for efficiency,
redesigning for new material or products, modifying materials use or
storage and/or production rate, and replacements); 

(4)	relocating/reconfiguring equipment; 

(5)	rerouting emissions to another control device; 

(6)	adding a new control device; and conducting specific activities
(e.g., factory experiments, remediation, test burning, emergency
generators, and pilot operations).  

It is clear from the list itself that in none of the cases listed would
there be any justification for a cap requirement to guard against abuse
of the advance approval mechanism. 

White Paper 3 and MNSR

1.	NSR Permitting in Advance

White Paper 3 states that in general, MNSR permit terms may not be
approved in advance because of the highly discretionary nature of the
decisions they require.  However, it suggests that exceptions could be
made for “clean buildings” and “clean replacements”.  A “clean
building” is a building containing a varying set of pieces of
equipment ducted to a common control device; a “clean replacement”
program would authorize the addition of specific types of equipment such
as “turbines, boilers, compressors, degreasers, tanks, emergency
generators, etc.”  (p. 24).  In both cases the overall emissions
increase and BACT would have to be defined in advance, and BACT would
have to be revisited every eighteen months. 

CCAI supports the concept of advance NSR permitting as outlined in the
draft White Paper.  However, we think the discussion should be focussed
more clearly on defining what current NSR regulations allow.  Those
regulations, § 52.21(j), (r)(2) already provide expressly for granting
“phased construction” permits subject to revisiting the control
technology determination every eighteen months.  They also provide
expressly for accepting the limits in an NSR permit as the
“baseline” for determining whether an emissions increase exists that
might trigger NSR in the future.  § 52.21(b) (21) (iii).  Although the
use of “permit allowables” as the baseline has been controversial in
many contexts, it has been accepted by all participants including David
Hawkins when applied to a new unit or facility that is “ramping up”
to full production over several years.  In substance, the White Paper
draft appears to be describing either a variation on “phased
construction permits”, or simply a way of writing “smart” NSR
permits similar to the “smart” Title V permits that the draft White
Paper describes.

CCAI believes, therefore, that this portion of the White Paper draft
would work much better as guidance explaining what the current MNSR
regulations allow than as an interpretation of Title V.  In that
context, the guidance would state: 

As long as control technology and ambient impacts can be adequately
defined, MNSR permits can address a variety of scenarios and permit a
variety of pieces of equipment.  CCAI does not believe there is any
valid reason why this authority should be limited to the specific
scenarios described in the White Paper draft.  Instead, any scenario for
which air quality can be adequately modeled, and technology adequately
defined, should be eligible for “smart permitting”.

Control technology determinations would need to be revisited every
eighteen months as the current regulations require.

As long as the source was moving forward with the program of
construction envisioned in its permit, the “permit allowables”
defined by the MNSR permit would be the “baseline” for determining
whether any future change at the facility had caused an emissions
increase under the NSR rules.

In fact, CCAI believes that sources should have the right to use
“permit allowables” as their “baseline” in many contexts beyond
the ones just described.  However, we have purposely limited our
comments in order to show that EPA would not need to broaden the policy
coverage of its White Paper 3 discussion in order to transpose these
provisions into NSR guidance and make them broader and more useful. 

2.	Limits and PALs

CCAI fully supports the use of limits on “potential to emit” and
“plantwide applicability limits” to make MNSR inapplicable to
plants, or activities within plants, that stay within defined emissions
limits. 

We particularly support the White Paper’s discussion of “minor
ongoing modification” (“MOM”) programs.  A MOM program would allow
the grouping of “certain interrelated changes ...independent of other
non-support activities at the site” into what would be, in essence a
PAL applicable to the MOM units only.  The difference between the MOM
PAL limits and the “past actual” emissions of the MOM units would
count as an “emissions increase” for purposes of NSR accounting for
the rest of the source.  In return, the MOM units would then be allowed
to make any change within their emissions caps without triggering NSR.

CCAI and its members have long believed there was no justification in
law or logic for limiting PALs to entire facilities.  We are encouraged
to see White Paper 3 put forward the same position.

However, the White Paper specifies detailed requirements for
incorporation of PALs into permits, including continuous monitoring
requirements.  A PAL could not be used to relax minor NSR limits, would
be limited to a 5-year duration, and could not be increased while it is
effective.  No modification that would cause exceedance of a PAL could
be preapproved absent major NSR review.  Ongoing minor modifications
could be approved under a PAL in limited circumstances.

Here again, the procedure is much more cumbersome than required.  With
or without a PAL, minor modifications should not require permit
revisions.  A PAL should be a tool to make it easier to determine when a
modification is minor, but EPA has made it needlessly complicated with
the detailed requirements for incorporating PALs into permits, and
needlessly narrow by restricting the changes that can be made under PALs
and to PALs.  Continuous monitoring to document the effects of an
operational change should not be required where it is not reasonably
necessary or where the applicable requirements already contain
monitoring requirements.  

In addition, we do not see the point of the White Paper’s discussion
of “cap and track”. This would allow sources to provide in a mNSR
permit for the approval and tracking of changes over a five-year period
to quantify the emissions increases (without regard to any decreases)
that those changes might cause. 

The White Paper discussion (p. 34) describes this as an approach “to
address project aggregation”.  But to really address aggregation, the
White Paper would need to define when different projects at a facility
will be aggregated and when they will not be aggregated, a subject it
does not mention.  This issue has been discussed repeatedly in the NSR
“stakeholder” meetings.  CCAI urges EPA to follow the
recommendations of those discussions and clarify when projects will, and
will not, be aggregated.

Enhanced Monitoring

White Paper 3 provides that flexible permits must not only meet Title V
monitoring requirements but also additional requirements to ensure
certainty of compliance with emissions caps and predictions for
alternative operating scenarios.  A hierarchy of acceptable monitoring
methods is specified, including continuous monitoring where available,
mass-balance calculations and emission factors developed from
site-specific testing.  The draft White Paper says forthrightly that
“we require use of CEMS or their equivalent, where either is
feasible”. id. , and provides detailed, mandatory procedures for
establishing parameter monitoring standards. See, e.g. p. 39 (calling,
among many other requirements, for a “correlation coefficient of 0.80
or greater”, together with use of the “Student’s t-value at the
eighty percent (two-sided) confidence interval”).  The data must
account for all increments of operating time, including process or
monitoring malfunctions and startup/shutdown situations.  Monitoring
must include fugitive emissions, and the White Paper states that Aa
source is not a good candidate for a plant-wide emissions cap if it
emits fugitive emissions that are an appreciable portion (e.g. 20
percent) of the total emissions emitted from the source or that cannot
be well quantified.” (p. 38). 

Such provisions are misguided in concept. Where applicable requirements
include specified monitoring approaches, no additional monitoring should
be required and no monitoring changes should be necessary when process
equipment is added or modified. Beyond that basic issue, these
provisions, which clearly would have binding regulatory effect, cannot
legally be included in a guidance document under the standards laid down
in the Appalachian Power case and must be removed from the final
version.  Further, the statute does not provide that additional
monitoring requirements must be specified in flexible permits, and the
degree of monitoring that the White Paper would require is excessive. 
If the permit already includes monitoring that bears on compliance with
PALs or alternative operating scenarios, no additional monitoring should
be required.  If additional monitoring is necessary, it should not be
required to cover unusual situations such as malfunctions or
startup/shutdown scenarios.  

The White Paper should specify that monitoring of fugitive emissions can
be required only where the emissions are reasonably quantifiable.  In
such cases, sources of fugitive emissions should be no less eligible for
permitting flexibility than other sources.  If the emissions are not
reasonably quantifiable, they should not be counted for Title V
purposes. 

More Detailed Comments

The White Paper contains a number of statements that are beyond the
authority of a guidance document, or beyond EPA’s authority
altogether.  They should be removed in the final version. Specifically: 

There is no justification for the statements in the draft White Paper at
pp. 13 and 19 that notice is required for switches within an approved
operating scenario.  As the White Paper itself explains, approved
operating scenarios are part of the permit.  There is no justification
for requiring a source to give notice when it moves from operating in
compliance with its permit to operating in compliance with its permit,
nor do the regulations require notice. See 40 C.F.R. 70.6 (a)(9)(i). 

On pages 8 and 9 the draft White Paper sets out a “sample” permit
term to establish a rolling twelve-month emission limit for a printing
line.  It consists of an equation that specifies averaging times,
inputs, and test methods.  After describing every logical component of a
standard, the discussion on page 10 adds that these components must
“be supplemented, in many locations, by additional limitations on
control efficiency, fountain solution VOC content, and cleaning solvent
VOC content or vapor pressure”. This statement is indefensible.  Since
the equation EPA has offered already contains every logically necessary
component, there is no justification for “additional limitations”. 
The fact that no reasons are given makes the statement even more
gratuitous.

On page 21, EPA states that the “permit shield” cannot apply to
permit terms generated by use of “replicable operating procedures”
(“ROPs”), because “[t]he part 70 permit shield may extend only to
permit terms and conditions that have undergone prior public and EPA
review.”  CCAI agrees with the principle; but it does not justify
denying a permit shield to ROP permit terms.  The ROPs   will have
undergone public and EPA review.  Accordingly, any permit term generated
by applying them properly should qualify for the permit shield.  Permit
terms resulting from their improper application would not qualify for
the permit shield.  The enforcement authorities would be able to readily
distinguish between the two, since any proper use of the ROP will yield
the same inputs no matter who applies it, as the White Paper explains
(p. 20).  Indeed, EPA’s current regulations expressly provide that
“alternate operating scenarios” can qualify for the permit shield. 
See 40 CFR 70.6(a)(9)(ii).

At page 26, the White Paper says that States may be allowed to grant
“advance approval” for changes under their own air toxics program

 “provided that the emissions associated with such changes are routed
to a state-of-the-art control device which is accommodative of these
emissions and the resultant emissions cause no ambient impacts beyond
those that your have determined to be acceptable using a specified air
toxics screening model under conservative assumptions defined in the
permit. 

Every element of this “guidance” exceeds EPA’s authority, and the
italicized language is particularly egregious.  Since the programs being
described exist only under State law, EPA has no mandate to tell States
how to run them. 

  Page 32 contains a particularly striking example of this insistence on
the letter of all current mNSR permits. It says that when a PAL is
written “[c]ertain minor NSR limits can be re-formatted, but only to
the extent that they would still meet their intended purpose and be
practical in their enforceability.” Since PALs will be written to
protect air quality and will require extensive public participation,
there is no reason why they should not be allowed to rationalize, amend,
or even repeal existing mNSR permits to the full extent that protection
of air quality and the underlying State law would allow. 

 In addition, CCAI wants to point out one clear misstatement in this
part of the draft. At page 23, the draft says that a “clean
building” must require LAER on “any applicable debottlenecked
activities”. However, it has always been EPA’s position, aside from
very special circumstances, that NSR technology is not required on
debottlenecked units because those units have not been “modified”.
CCAI believes that this long-standing position is clearly required by
the Clean Air Act. 

 There is no justification, however, for the suggestion (p. 34) that MOM
PALs should be limited to “specific advance approved, related
changes”. As with regular PALs, all changes to a MOM PAL should be
eligible for coverage as long as compliance with the cap can be
quantified, 

 See, e.g., Essex Chemical Corp. v. Ruckelshaus, 486 F.2d 427, 432-33
(1972)(NSPS must include variance for excess emissions from startup,
shutdown or malfunction situations). 

 See, e.g., 59 Fed. Reg. 44514 (August 29, 1994)(AEPA believes the Act
requires that fugitive emissions, to the extent quantifiable, be
counted)(emphasis added); accord., National Mining Assn. v. EPA, 59 F.3d
1351, 1360 n.19 (1995).  

Footnote continued from previous page

Footnote continued on next page

  PAGE  6 

