September 14, 2000

Michael Trutna

Information Transfer and Program

  Integration Division  (MD-12)

U.S. Environmental Protection Agency

Research Triangle Park, NC  27711

Re:	Comments on White Paper 3

Dear Mr. Trutna:

This letter provides the comments of the Pharmaceutical Research and
Manufacturers of America (PhRMA) on the August 7, 2000 draft document
entitled “Design of Flexible Air Permits – Operating Permits Program
White Paper Number 3” (White Paper 3).  PhRMA represents the
country’s leading research-based pharmaceutical and biotechnology
companies, which are devoted to inventing medicines that allow patients
to lead longer, healthier and more productive lives.  Investing over $26
billion in 2000 in discovering and developing new medicines, PhRMA
companies are leading the way in the search for cures.

The pharmaceutical industry is characterized by the need for constant
change.  Not only must pharmaceutical manufacturing plants produce a
broad and diverse range of existing medicines, but they must also have
the flexibility to accommodate the new and improved products that are
constantly being developed.  EPA recently acknowledged the critical
importance of rapid change to the pharmaceutical industry by investing
substantial time and resources in developing an innovative change
management strategy for the NESHAP for pharmaceuticals production. 
Similar innovative approaches for minimizing regulatory delay are needed
for all Clean Air Act programs, especially the Title V operating permit
program.

Accordingly, PhRMA supports EPA’s efforts to promote operational
flexibility through the development and implementation of White Paper 3.
 Alternative operating scenarios, replicable operating procedures, and
advance approvals are keys elements of the pharmaceutical MACT change
management process and should similarly form a strong foundation for
establishing more broadly applicable flexible permitting opportunities. 
In addition, we believe that other ideas presented in the draft – such
as Plantwide Applicability Limits and the Clean Building approach –
provide innovative ways of handling major New Source Review
applicability and other specific flexible permitting challenges.

While PhRMA strongly support EPA’s efforts to develop White Paper 3
and encourage the Agency to issue it as soon as practicable, we do have
a number of comments and suggestions on specific aspects of the draft as
discussed below.

EPA should support PALs as a valuable flexible permitting tool, but
should delete the detailed PAL criteria from the final White Paper.  

            As noted above, Plant-wide Applicability Limits (PALs) stand
out as a real opportunity to simplify and facilitate major NSR
applicability determinations.  Moreover, as evidenced by the PALs
recently issued in the pharmaceutical, auto assembly, and specialty
chemical manufacturing industries, there is ample legal authority to
implement PALs under the existing Part 70 and NSR regulations.  Thus, we
believe it is important for EPA to send the clear message that, where
sources and permitting authorities determine that they are appropriate,
PALs can and should be issued.

The draft White Paper acknowledges that PALs are permissible under
existing regulations; however, the draft outlines detailed criteria that
“shape the scope and expectations of PALs that [EPA] will support
under the current NSR regulations.”  WP p.31.  We have several
concerns with these criteria.

First, a number of the criteria appear to place unnecessary restrictions
on the use of PALs.  For example, the draft suggests that PAL emission
limits must be based on past actual emissions plus an increase not to
exceed the applicable NSR significance threshold.  Such an approach
would not be practical, however, for a new unit or facility that
undergoes major NSR at the time of construction.  Similarly, the draft
states that PALs have a duration of up to five years.  In our view,
there is no legal impediment to issuing PALs with a longer duration and,
given the time and effort required to establish a PAL, sources and
permitting authorities will often want to establish robust PALs that are
designed to last much longer.  Also, the draft states that a PAL would
limit emissions from all emissions units at the source; however, there
are clearly circumstances where this approach would not be necessary or
practical (e.g., two entirely distinct manufacturing operations owned by
the same company at the same site). Finally, the draft prohibits
so-called “52.21(r)(4) relaxations” unless the changes are subject
to major NSR.  It would be preferable for the guidance to simply state
that deletions or modifications of synthetic minor limits must assure
that major NSR would not be triggered.  PhRMA believes 52.21(r)(4) can
be interpreted to allow changes to synthetic minor limits to occur
without triggering major NSR, because 52.21(r)(4) prohibits a relaxation
of a synthetic minor limit.  It does not, however, prohibit changes to a
limit for other reasons (e.g., when a source has been modified in a way
unrelated to the construction or modification that established the
limit, and needs the emission limit to be raised).

Our second concern with the draft PAL criteria is that they will
unavoidably impact the PAL regulations that EPA is planning to
promulgate later this year as part of the “NSR reform” rule.  While
the draft states that the “intent of this guidance is not to present
[EPA’s] final positions on these issues, nor is it to prejudge the
outcome of that rulemaking” (WP p. 31), we believe that the detailed
criteria in the White Paper could, as a practical matter, become the
point of departure for future decisions about the precise contours of
the PAL rule.  Perhaps more importantly, the PAL policy outlined in the
White Paper could actually limit EPA’s latitude in the future
rulemaking because the Agency has a legal duty to justify changes in
established policy (i.e., rather than starting with a clean slate, EPA
may conclude that it must reconcile the PAL rule with its preexisting
policy).

Our last concern is that, because of the detailed, mandatory, and
prescriptive nature of the PAL policy outlined in the White Paper,
opponents of PALs and flexible permitting may attempt to challenge the
White Paper on the grounds that EPA issued binding “regulations”
without using required notice and comment rulemaking procedures.  A
successful challenge could cast unnecessary uncertainty on the ability
of EPA and permitting authorities to issue PALs during the period before
the NSR reform rule is promulgated.

PhRMA’s recommended solution to all of these potential problems is for
EPA to delete the detailed PAL criteria from the final White Paper.  The
Agency should simply explain PALs and why they are a valuable flexible
permitting tool, emphasize that PALs are permissible under current law,
and indicate that a PAL regulation will be issued in the near future.

Facilities with proportionally high fugitive emissions should be
eligible for flexible permitting. 

            On page 38, the draft explains that a source is not a good
candidate for a plant-wide emissions cap if an “appreciable portion
(e.g., 20 per cent)” of emissions from the site are fugitives.  This
comment is included in the discussion on compliance assurance and
reflects a concern that fugitive emissions may not be reliably
quantifiable.

As explained at the beginning of these comments, we believe that the
pharmaceutical industry has a particular need for flexible permitting
tools because of the rapid pace at which existing drugs are improved and
new drugs are developed.  It is not uncommon, however, for fugitive
emissions to make up a substantial portion of the overall emissions from
a pharmaceutical manufacturing plant (particularly at plants that have
installed state-of-the-art emissions controls, such as those that will
be imposed under the recently promulgated MACT for pharmaceuticals
production).  Thus, the draft White Paper would prohibit the use of
important flexible permitting tools at many facilities where these tools
are needed the most.

While quantifying fugitive emissions can be a difficult technical
problem to solve, it is certainly not insurmountable.  In our
experience, sources, permitting authorities, and other interested
stakeholders can, in fact, develop methods that are mutually acceptable
and sufficiently accurate that they can be relied upon for making
applicability and compliance determinations.  Moreover, for certain
facilities and certain applicable requirements, fugitive emissions do
not even have regulatory significance and, therefore, quantification is
not even relevant for applicability or compliance purposes (e.g.,
facilities that do not belong to one of the 28 listed source categories
do not need to count fugitives in determining the applicability of major
NSR).

Thus, PhRMA believes that it is appropriate for EPA to point out the
unique challenges that may be presented by fugitive emissions; however,
the final White Paper should not categorically prohibit facilities with
a high proportion of fugitive emissions from using the flexible permit
tools described in the White Paper.

Clean Buildings:  

             Under this option, a source undergoes major NSR for a given
building and then is allowed to make subsequent changes to operations in
the building without triggering NSR again as long as NSR applicable
requirements are met.  BACT/LAER and the air quality analysis would have
to be reviewed and, as necessary, revised every 18 months.  We support
the concept behind this approach – i.e., that major NSR should not
apply to operations covered by state-of-the-art controls and that have
been shown to have no significant adverse impact on air quality. 
However, we think that this approach could be made more workable without
compromising its environmental benefits.

For example, the requirement that the controls be re-evaluated for
BACT/LAER every 18 months is a significant disincentive for companies to
pursue this option. The periodic re-evaluation should occur over a much
longer period of time, such as 10 years, so that a source would have
some long-term assurance that it’s investment in highly efficient
controls will be worthwhile.  Without such assurances, sources are not
likely to sign up for what is otherwise a very promising option.  For
most pollutants, especially the VOC sources that appear to be the best
candidate for flexible permits, state of the art technology does not
advance as quickly as every 18 months.  A 10-year time frame is a much
more appropriate period for re-evaluating whether technology is
appropriate.  The 18 month review period is also inconsistent with the
10-year “clean unit” exemption proposed by EPA in its July 1996 NSR
reform proposal.  

Advance approvals should be adequately described in the permit, but
should not be unnecessarily limited. 

           On page 15, the draft explains that flexible permits with
advance approvals must include “both a description of the types and
capacity ranges of advance approved equipment and some limitation on the
amount of new unit additions should be placed in the permit.”  Similar
comments are made on page 18.  EPA suggests that appropriate limits may
include an emissions cap or a limit on the number of types or
combinations of new units.  The draft indicates that such limits are
needed for several reasons – to make sure that the advance approvals
qualify as reasonably anticipated alternative operating scenarios under
Part 70, to allow for meaningful review, to assure that there will be
adequate control equipment capacity, and to assure that no unacceptable
ambient loading will occur pursuant to applicable minor NSR program
requirements.

 PhRMA agrees that advance approvals must satisfy all substantive and
procedural requirements under applicable emissions standards and
permitting programs; however, we believe that there are other ways to
accomplish this important objective besides limiting the number of
advance approvals or imposing an emissions cap.  The approach used for
any given permit necessarily will depend on unique, site and situation
specific circumstances.  It will also depend on the nature of the
applicable requirements that apply to the permitted operations.  If
applied rigorously, the “one size fits all” approach described in
the draft could substantially and unnecessarily limit potential permit
flexibility and, ultimately, the value of the flexible permit tools
provided by the White Paper.  Thus, the final White Paper should make it
clear that sources and permitting authorities have an obligation to meet
all applicable substantive and procedural requirements, but have
substantial latitude and discretion in crafting appropriate permit
terms.

For example, clearly explaining the nature and scope of advance
approvals when a draft flexible permit is developed and issued for
public comment is a key to satisfying procedural permit issuance
requirements.  Such clear and up-front communication with all interested
stakeholders in the permit also makes sense as a practical matter. 
However, an emissions cap or limit on the number of advance approvals is
not necessary to accomplish full, complete, and accurate public
disclosure or for interested parties to achieve a thorough understanding
of the proposed permit.

Similarly, as noted above, Part 70 allows alternative operating
scenarios to be included in Title V operating permits; however, they
must be “reasonably anticipated.”  Again, the nature and scope of
advance approvals must be clearly spelled out in a permit to satisfy
this legal requirement.  A cap or numeric limit is not the only way to
provide the necessary clarity.  For instance, the scope of advance
approvals could be reasonably bounded in some cases by specifying the
throughput of a given operating area, limiting the amounts of particular
raw materials that may be used, imposing hours of operation
requirements, or imposing reasonable bounds on other process or plant
specific characteristics.  Again, appropriate bounds for advance
approvals should be developed according to the unique circumstances of
each permitted facility.

Notably, the draft also suggests that flexible permits may need to
include emissions caps “to meet the ambient protection provisions of
minor NSR.”  See pp. 18 and 42.  A requirement to establish NAAQS and
increment-protective caps could substantially complicate flexible
permitting (e.g., a PSD-type air quality analysis could conceivably be
required), increase the time and effort required to establish a flexible
permit, and serve as a significant practical barrier to flexible
permitting.  Moreover, in our experience, state and local permitting
authorities use a variety of methods – including but not limited to
emissions caps – to assure protection of ambient air quality through
their minor NSR programs.  As explained above, the final White Paper
should acknowledge the importance of satisfying all applicable
requirements when a flexible permit is established, but indicate that
permitting authorities have the latitude to accommodate site-specific
characteristics in crafting appropriate permit terms and conditions.

Other options should be available for addressing major NSR in flexible
permits.  

          If a flexible permit requires the source to connect new
equipment added under an advance approval to a control device that
achieves a designated control efficiency, that requirement could serve
as a means of restricting potential to emit (PTE), and thus ensure the
PTE does not exceed NSR significance thresholds.  A similar approach
could be used for changes in existing equipment if the flexible permit
contained appropriate control requirements, replicable operating
procedures (ROP), and emission limits.  PhRMA has not fleshed out all
the elements of this approach, but we believe EPA should include PTE
limiting provisions as another means of addressing major NSR.

Discussion of the permit shield should be clarified.  

          On p. 21, the draft states that the permit shield “may not
extend to the outcomes of ROP equations, applicability or
nonapplicability determinations, or other compliance determinations
recorded exclusively in the on-site log.”  The justification for this
conclusion is that, “The part 70 permit shield may extend only to
permit terms and conditions that have undergone prior public and EPA
review.”  ROPs, the obligation to appropriately implement ROPs, and
the obligation to comply with requirements that flow from the ROPs are,
however, all provisions that have undergone prior public and EPA review.
 Therefore, the justification for limiting the permit shield when ROPs
are used is inconsistent with actual practice.  If ROPs are used
correctly, then the permit shield should, in fact, apply because the
source will be operating in a manner that is wholly consistent with the
scope of the publicly reviewed permit.  On the other hand, if a source
misapplies a ROP, then that source may operate in a manner that was not
anticipated at the time of permitting (i.e., may be operating outside
the range of the approved alternative operating scenario).  In PhRMA’s
 view, the final White Paper should provide that the permit shield can
apply when ROPs are properly implemented, but does not apply to
applicability or compliance determinations resulting from the incorrect
implementation of ROPs.

The White Paper should not suggest limitations on the use of flexible
permits.

           Page 4 of the draft includes a list of several circumstances
where flexible permitting may not be appropriate.  The list is preceded
by the suggestion that flexible permits should be developed only where
the benefits justify the extra burden.  While PhRMA agrees that flexible
permitting is not appropriate in all circumstances, we think that this
discussion should be omitted from the final White Paper.  Sources,
permitting authorities, and interested stakeholders will quickly be able
to assess the relative value of creating a flexible permit for a
particular site.  The guidance provided in the draft would not be
particularly helpful in making informed decisions in these site-specific
circumstances and, more generally, could discourage the development of
flexible permits in the first instance.

Greater flexibility should be provided in demonstrating compliance with
cap permits.

           We have two specific suggestions.  First, the draft states
that emissions factors should generally not be used unless validated by
site-specific testing.  WP p. 38.  Moreover, extremely conservative
safety factors must be applied to emissions estimates derived from
emissions factors not validated by site-specific testing.  WP p. 39. 
Also, the draft states that the source assumes all the risk if factors
are erroneous.  WP p. 39.  While the use of emissions factors is
certainly not ideal in many applications, we believe that they are
useful and should be usable in certain specific circumstances.  For
example, if a facility operates a small package gas-fired heater that
represents only a tiny fraction of NOx emissions from the overall site,
then the use of emissions factors provided by the manufacturer should be
recognized as reasonable and appropriate.  In such a circumstance,
site-specific testing should not be required to validate the factors
because the possibility of significant variations would be small and the
factors would have a high degree of reliability since they were
developed for the particular heater by its manufacturer.  Also,
conservative safety factors should not be needed because, even if the
factors are not perfectly accurate, compliance assurance will not be
compromised because the heater represents a small fraction of overall
site emissions.  Finally, as long as the source, the permitting
authority, and interested stakeholders have a chance to review the
factors prior to permit issuance, all of the risk of error should not
automatically flow to the source.  Thus, as this example shows, the
final White Paper should acknowledge that emissions factors can be used
– without site-specific testing, extremely conservative safety
factors, and assignment of all risk of error to the permittee –
provided informed decisions are made through the permitting process.

Second, the draft states that, when parametric monitoring is used,
on-site emissions testing must be performed to correlate monitoring with
actual emissions.  WP p. 39.  PhRMA believes that the requirement for
emissions testing is not appropriate in all circumstances and will
unnecessarily drive up the cost and burden of obtaining a flexible
permit.  For example, for safety and product purity reasons, certain
pharmaceutical manufacturing plants have adopted a strategy of using
point of generation air pollution control devices (rather than
installing manifolded systems that combine uncontrolled emissions from
multiple operations and route them to a limited number of large,
end-of-line control devices).  Such facilities may have dozens of
relatively small condensers, scrubbers, thermal oxidizers, and other
types of control devices for which parametric monitoring would be the
most appropriate method of assuring ongoing compliance with applicable
requirements.  Each of these air pollution control devices would handle
a relatively small portion of the total uncontrolled emissions from the
site.  If the requirements of the draft White Paper were applied, dozens
of performance tests would have to be conducted to establish the
necessary correlations to actual emissions.  In addition to being
prohibitively costly and time consuming, such testing is simply not
needed to provide a reasonably accurate correlation to actual emissions.
 For such devices, engineering analyses or other scientific methods of
assessment could be used to establish reliable correlations.  In fact,
in the recently-promulgated MACT standard for pharmaceutical production,
such engineering assessments may be used to make initial compliance
demonstrations for practically all control devices handling less than
ten tons per year of uncontrolled HAP emissions.  See, e.g., 40 C.F.R.
§§ 63.1257(c) and (d)(3)(i) (1999).  A similar approach must be
adopted in the final White Paper in order for cap permits to be workable
as a practical matter.

EPA should refrain from addressing the applicability of NSR control
requirements to units that experience collateral emissions increases.  

          At several points, the draft suggests that BACT/LAER applies
to emissions units that experience collateral emissions increases (i.e.,
units that are not physically changed, but have an emissions increase as
a result of a change to another emissions unit at the facility).  See,
e.g., WP p. 23 (BACT/LAER under the Clean Building approach must cover
“activities advance approved to occur within the clean building and
for any applicable de-bottlenecked emissions.”).  In PhRMA’s view,
this suggestion is plainly inconsistent with EPA’s NSR regulations. 
See, e.g., 40 C.F.R. § 52.21(b)(2)(iii)(f) (a physical change or change
in the method of operation does not include increases in production or
hours of operation that are not otherwise prohibited by existing
federally enforceable permit conditions or regulations).  Moreover, this
suggestion is inconsistent with EPA’s longstanding interpretation of
its regulations.  Accordingly, we recommend that EPA correct or delete
potentially confusing or inconsistent statements regarding the
applicability of NSR to units that experience collateral emissions
increases.

Thank you for the opportunity to provide comments on the draft White
Paper 3.  In PhRMA’s view, environmental protection and regulatory
flexibility can and should go hand in hand.  We support EPA’s efforts
to develop this guidance on flexible permitting and encourage you to
move ahead expeditiously with the final version.

If you have any questions relating to these comments or need additional
information, please call Bernie Paul of Eli Lilly at (317) 276-0331,
Catherine Ehlhardt of Eli Lilly at (317) 276-3095, or Bill Wehrum at
(202) 637-2284.

						Sincerely,

						Thomas X. White

|||

 	The notice asking for comments on White Paper 3 was published at 65
Fed. Reg. 49803 (Aug. 15, 2000).

 	See 63 Fed. Reg. 50280, 50309 (Sept. 21, 1998).

 	See, e.g., Small Re. Lead Phase-Down Task Force v. EPA, 705 F.2d 506,
519 (D.C. Cir. 1983) (holding that the Clean Air Act imposes a duty on
the Agency to explain and justify its actions and that this duty is more
stringent when the Agency alters established policy).

 	See, e.g., Appalachian Power Co. v. EPA, 208 F.3d 1015 (D.C. Cir.
2000).

Michael Trutna

Re: Comments on EPA White Paper 3

September 14, 2000

Page  page  6 

