Statement of
George C. Freeman, Jr.
for
The Utility Air Regulatory Group
before
The Environmental Protection Agency
December 12, 1978 Hearing on
Revised New Source Performance
Standards for Fossil-fuel Fired Electric
____Utility Steam Generating Units____
I.

Introduction

I am George C. Freeman, Jr.
behalf of the

I am appearing here today on

Air Regulatory Group HTARG^ 3 which is an

ad hoc group established to participate in EPA's rulemakings
implementing the 1977 Amendments to the Clean Air Act.

JARG^s

membershi

covers t]___________________

industry^

It consists of the Edison Electric InstituteffEEIl

___

thp^National Rural Electric Cooperative Association^/and 87 in-_
dividual utility systems that collectively own a majority of the _
generating capacity in the United States.I
I am accompanied today by Mr. A. Joseph Dowd, Senior Vice
President and General Counsel of the American Electric Power Service Corporation who is also Chairman of the UARG Policy Committee;
Mr. Randall E. Rush of Southern Company Services who is Chairman
of the UARG Scrubber Subcommittee; Mr. L. Kenneth Newsome of Virginia Electric and Power Company who is Chairman of the UARG
Continuous Monitoring Subcommittee; and Mr. Paul C. Bailey of
Southern Company Services who is Chairman of the UARG Particulate
Subcommittee.

A number of UARG's consultants are also present

 -2today to help respond to any questions members of the EPA panel
may have concerning the factual bases for UARG positions.
Over the last year UARG has cooperated with EPA in its
development of a macro-economic model for the NSPS rulemaking.
UARG has also mounted its own substantial modeling effort.
At the request of EPA technical staff in the fall of 1977,
UARG attempted to define potential major issues in the rule
making, and stated its preliminary position with respect to
those issues prior to the Staff's revelation of its initial
recommendations. These recommendations and a large body of
supporting materials were the subject of a two day NAPCTAC
Since that time, EPA has been
meeting in December 1977.
developing its record on technical issues independently.
UARG has undertaken a major fact-finding program of its own
s
and has modified its preliminary positions on several occasion
as a result of new data and information.
In light of the 1977 Amendments to the Clean Air Act and
the facts as they now stand, UARG has recommended new major
,
limitations on its own new coal-fired generating plants. Thus
scrubbers
requirements that would impose_____
UARG proposes new S0«~__
_____^^MM*I
_____ ___
___
,___
g needed
on virtually all new plants, while at the same time providin
on
flexibility for further development of scrubber technology
and
higher sulfur coals and greater incentives for the socially
environmentally preferable front-end technologies.1 The UARG
val
sliding scale proposals would also reconcile percentage remo
by
requirements with national energy and economic objectives,

 minimizing adverse inflationary and oil consumption effects of
the new regulations. Finally, the UARG proposal of a 30-day
averaging time is more consistent with current technology than
EPA's 24-hour proposal, would minimize future enforcement problems
and would not impose additional det-pTi-<ar>t-^ r>-p
sulfur coals.
UARG's particulate recommendations are based on the current
state of technology and are in accord with the technical assessments of DOE.

UARG's position on NO JL is based on practical con-

siderations of technological possibility and the balancing of the
competing environmental, energy and economic (including anticompetitive effects) aspects of NO Jv regulation.
The differences between the UARG proposals and those EPA
staff proposals published in regulatory form on September 19 (the
September 19 EPA proposals) can be simply stated: UARG's proposals
take into account all of the statutory factors spelled out in
section 111(a) including technological feasibility, costs, nonair
quality health and environmental impacts and energy requirements.
The September 19 EPA proposal does not.
II.

Procedural Aspects of the Rulemaking

Before turning to the substantive questions which must be
addressed during the next three days by the public and the
Agency, I would like to discuss our views on the manner in which
the Agency has run this rulemaking to date and the continuing
failure of the Agency to meet the procedural requirements of
the Clean Air Act and the Administrative Procedure Act (APA).

 -4Under both section 307(d) of the Clean Air Act and section
553 of the APA, the Agency has an obligation to disclose fully
at the time it publishes proposed rules the entire record upon
which it bases its proposal.

Section 307 makes the minimum

bounds of this obligation explicit:

at the time of initial

publication, EPA must provide (1) the factual data on which
the proposed rule is based, (2) the methodology used in obtaining and analyzing that data and (3) the major policy considerations and legal interpretations underlying the proposed rule.
EPA has to date failed to comply with these requirements.
As the preamble to the September 19 proposed regulations points
out, data and other information to support certain aspects of the
Agency's proposals is not yet available.

For example, EPA's

macro-economic analysis is still being carried out, and the
assumptions used in that analysis have changed significantly
since September.

The Agency has only recently undertaken a

program designed to obtain information to support its proposed
24-hour averaging time for SC^-

And data on the impact of

flue gas desulfurization systems on the ability of units burning higher sulfur coals to meet the proposed particulate
standard has yet to be provided by the Agency.

Other informa-

tion vital to a determination of the feasibility and appropriateness of EPA's proposal has been coming out piecemeal
since the September 19 publication.

This failure to provide

data to support critical portions of the Agency's proposal
at the time the proposal was published has led to EPA's

 -5failure to adhere to the second major requirement of both
section 307(d) of the Act and section 553 of the APA -- provision of adequate opportunity for the interested public to
review and comment on the factual data, methodologies and
rationale underlying the Agency's proposed rules.
For over two and one-half years, EPA has been developing
information upon which to base its proposal.

On September 19,

EPA gave the public just 60 days to comment on the fruits of
that two and one-half year effort much of which was made available
for the first time on that date.

We requested an extension

until January 15, 1979, of the schedule for submittal of initial
comments on the September 19 proposal.

We considered this

extension the minimum necessary to enable UARG and its consultants to review adequately the voluminous, albeit incomplete,
record.

Yet, EPA refused our request.

Instead, the Agency

provided only 25 extra days for submittal of initial comments.
This refusal to provide an adequate period for meaningful
public review of and comment on the material disclosed to the
public on September 19, if left unremedied, will jeopardize
the substantive adequacy of the record ultimately developed in
this rulemaking.
But the Agency's failure to provide adequate time for
comment on the September 19 record is only one part of the
problem.

As I have just mentioned, much of the information

upon which the September 19 proposal was based was not released

 -6until after that date.

Some has not yet been released.

Indeed,

some of the data which would be necessary to support EPA's
proposals still do not exist and are being feverishly generated
Certainly, the public cannot be said to have
had an adequate opportunity to comment on EPA's proposal if it
is being denied the opportunity to see the material on which EPA

by the EPA Staff.

bases that proposal.

Moreover, this failure to provide the public

with the data and rationale underlying the Agency's proposal
continues here today.
EPA could have made this hearing a forum at which the
public could question the Agency's staff in order to probe the
basis for the Staff's conclusions.
this opportunity on its head.

Instead, EPA has turned

Rather than providing the

public an opportunity to explore the factual predicates for
its proposal, EPA has indicated its intent to channel the flow
of information at this hearing in one direction --to the
Agency. This leaves UARG and others here today in a curious
These hearings will disclose the basis underlying
the positions taken by all the participants but EPA -- the only
participant which has a statutory obligation to disclose that

position.

information.
This leads to my last point.

The September 19 record was

defective in a number of critical areas and clearly inadequate
to support the Agency's proposal. UARG has submitted detailed
questions to the Agency pointing out many of the major areas

 -7where the record is presently either devoid of information or
contains inadequate information upon which to base a rational
conclusion.

The Agency must fix a date by which its

data gathering will be complete, and then republish its proposal
as revised to reflect the data gathered by the Agency since
September 19.

It must then provide the public an opportunity

to comment on its revised proposal and all of the information
contained in the record at that time, including the information
provided in response to the questions UARG and other members of
the public have submitted concerning the September 19 record.
If the Agency determines not to follow this approach, it
must provide another route for interested menbers of the public
to ascertain the factual basis for and the rationale underlying
the conclusions EPA has drawn in developing its proposal.

Thus,

if EPA determines not to repropose its regulations as I have
suggested, it should reconvene this hearing 30 days after the
close of the period for filing responses to comments on the
September 19 proposal.

At the reopened hearing, the Agency

should make available a panel of EPA Staff experts to answer
questions submitted by members of the public.

We will submit

to the Agency shortly detailed proposals for the procedural
format of such a hearing.
If the Agency follows either of these suggestions it will
of course be difficult, if not impossible, to promulgate revised NSPS by the end of March.

But we do not believe that

 -8this should deter the Agency from developing an adequate record or providing the public a meaningful opportunity to comment on that record.

The courts in this Circuit have recog-

nized that statutory deadlines for promulgation of regulations
must give way when the Agency cannot perform the statutorily
mandated task within the time provided.

In view of the com-

plexity of this rulemaking and the substantial impact it will
have on the American public over the next 20 years, it is
reasonable to assume that those same courts would not judge
EPA harshly for failing to meet a March promulgation date
here.

Moreover, the settlement order in the recent District

Court suit expressly reserved EPA's right to seek an extension
of the March 19 promulgation deadline for just such a contingency as now exists.
III.

The Applicable Law

I'll turn next to some of the substantive issues raised by
this rulemaking.

There was substantial debate in Congress during

consideration of the Clean Air Act Amendments and the National
Energy Act regarding the need to reconcile our national energy
and environmental objectives.

More recently, we have been told

by the White House that our national energy and environmental
policies have to be reconciled with our pressing national economic
priorities.

Indeed, at the same time the President called on

labor and business to exercise self-restraint and to cooperate
with his program of voluntary wage and price controls, he promised

 -9that his Administration would "keep to a minimum those actions
that directly increase private costs and prices -- actions that
include regulations that add unnecessarily to costs and prices
. . . ." This rulemaking will furnish the first solid test of
Ee^COYWX£-^
that Presidential pledge.
Fortunately, section 111 of the Clean Air Act demands that
EPA take potential inflationary impacts into account in promulgating new source performance standards. That section of the Act
_______________

_____

______-__^___^^^^^^^^»»^^^^^^^^^^^^^^a»«

•••••^•••••i^^MMBBIMMBlMBBMi^BB^MBBMMBMW

expressly requires EPA to base tts revised new source performance
standards on the "best technological system of continuous emission
reduction which (taking into consideration the cost of achievi
such emission reduction, any nonair quality health and environmental impact and energy requirements) the Administrator deter-^
mines has been adequately demonstrated." The Administrator must
cons ider these factors in setting both emission limitations
and percentage reduction requirements for new powerplants.. Q/

/£>

It is also important to note that the Conference Report
indicates a (Congressional intentjbhat the percentage removal.
requirement need not be uniform by specifically stating that
"in establishing a national percent reduction foj
fuel-fired sources, the Conferees agreed that the Administrator
may in his discretion, set a range of pollutant reductioj
reflects varying fuel, characteristics."

The structure of the

Act provides additional support for the proposition that a
flexible percent reduction requirement can and should be set by
EPA.

 -10Part C of the amended Act ,
of significant deterioratio

ntains the prevention

PSD) requirements ^requires the

states to set the best: available r-nnt-rnl

new plants on a case^by-case basi

ich governs non-

attainment areas, requires the states to set lowest achievable _
emission rates (LAER) for new plants located in those areas.
The fact that these BACT and LAER requirements cannot be less
stringent than the revised new source performance standards
Indicates Congressional intent that the NSPS not be set by EPA
at the outer limits of available technology.

Some technological

leeway must be left to permit the states to balance at each new
plant local environmental conditions, local economic con'siderations
and local employment needs.
IV.
A,

Proposed NSPS for S02

What is Not at Issue Here - Public Health

The incremental differences in total national S(>2 emissions
I ^resulting from adoption of any of the three SC^ standards under
ffjf

nronosal.
fiill RoniKhincr
f>Vi*a F.PA
"1* ^a-rai-T rvn
co1"10
scrubbing proposj
EPA full
EPA (the
by F.PA
n "hv
consideratio

0.8 floor or the UARG sliding scale) are small.

I-"hp DOE_

Regardless of

which approach is ultimately chosen, all new fossil-fuel fired
powerplants will use SC^ scrubbers or emerging technologies that
assure significant further emission reductions of SO^.

And re-

gardless of which alternative is chosen, other air quality requirements designed to ensure the public health and welfare independent
of revised new source performance standards will have to be met.

 -11Even if the marginal decreases in SC^ emissions at issue
cy
here could be quantified in terms of lengthened life expectan
or increased relief from discomfort, these benefits might well
be more than offset b)' decreases in life expectancies and in*/
s.
creased discomfort resulting from the higher energy cost
of
Electric air conditioning and space heating are major means
controlling temperature and humidity levels in the indoor enble
vironment where most of the persons in our society suscepti
.
to marginal levels of air pollution spend most of their time
Cost is, of course, a major factor in the availability of elec
ftricity,for those purposes.

And under both the EPA-DOE and

NERA analyses, EPA 1 s "'full scrubbing" option imposes greater
on
aggregate national costs and also greater costs in every regi
than do the UARG and DOE proposals.

_jr

B.

What is at Issue

1.

Full Scrubbing versus the Sliding Scale
UARG believes that both its sliding scale proposal and

to a lesser degree the DOE proposal are preferable to the EPA
uuniform percentage reduction proposal under all of the stat
tory factors which the Administrator is required to consider
*/

ing homes
A recent study of deaths in New York City nurs
number of deaths
in heat waves in 1972 and 1973 concludes that the
significantly
in nursing homes without air conditioning "was
on the basis of
greater than the number of deaths expected The
y recommends
mortality during cooler control periods." for stud
the elderly lo"that nursing homes and other institutions
ired to provide
cated in climates like that of New York bect requ
studies also sugair conditioning." Early EPA health effe
dity may be more
gested that high temperature and high humiy
in so-called "air
closely associated with increased mortalit ls.
pollution episodes" than air pollutant leve

 -12under section 111 of the Act.

At the time he announced the

September 19 proposal, the Administrator announced that he had
an open mind on which of the various SO^ proposals would ultimately
be determined to meet the factors set forth in seciton 111 of
the Act.

It appears, however, that certain EPA regional officials

may have already prejudged the outcome of this rulemaking since
they are requiring that applicants for PSD permits demonstrate
compliance with the EPA Staff's "full scrubbing11 proposal.

We

hope that this is not national policy and that the Administrator
continues to have an open mind on the question.

We would also

request that the Administrator direct his regions to do what the
statute commands -- make BACT decisions on a case-by-case basis
after fully considering all of the statutory factors.

If at the

conclusion of this rulemaking, revised NSPS for a source is more
stringent than its BACT limits, the PSD permit can be modified.
Turning to the statutory factors the Administrator must
consider in this rulemaking, let us look first at the question
of what constitutes a "demonstrated technology."
emonstrated
There is apparent agreement among EPA, DOE and UARG that all
of their SQ2 percentage removal proposals will "push" the development of better scrubber technology.

The key issue is how far

and how fast that technology may and should be "pushed."

EPA

concludes in the preamble to the proposed NSPS that Wellman-Lord,
magnesium oxide, dual alkali, lime and limestone FGD systems are

 -13capable of achieving the levels of removal efficiency (85
average) and system reliability (90 percent with
one spare module) reflected in the proposed NSPS.

Available

data, however, simply do not support the conelus ion that long term, reliable operation of non-lime/limestone FGD technologies
have been demons trate<f:

Reliability jandsludgeASisposal problems

also continue to plague lime/limestone FGD technology.
an

And

85 percent removal efficiency has not been shown to be

consistently achievable on a short-term (24-hour) averaging
basis. Indeed, the Agency's data base has not yet related
removal efficiency to any averaging time.

In fact, there are

grave inadequacies in virtually every aspect of EPA's evaluation
of the state of lime/limestone technology. For example,
While EPA states that "certain practical changes
1.
in design, operation, and maintenance practices" will
be necessary to enable lime/limestone FGD systems to
meet the staff-recommended NSPS, nowhere are these
"practical changes" identified.
EPA has not addressed certain critical operating
2.
parameters such as the impact of high sulfur coal and
closed-loop operation on scrubber reliability.
Most of the data on which EPA bases its reliability
3.
and removal efficiency conclusions are the product of
short-±arm te&£4ng. These data do not establish that
scrubbers can continuously operate at the levels of
reliability and removal efficiency reflected in the staffrecommended standard. The continuous monitoring data
available are of questionable validity, and UARG has been
informed that EPA has recently returned to the field to
gather additional data on continuous scrubber performance.
EPA cites Japanese scru^frpr ^p^-*-Q*-i'^ as a basis for
4.
conclusions concerning scrubber capabilities in the United
States, but does not evaluate similarities and differences
between Japanese and American scrubber applications and
technology. Available data indicate that there are significant differences which prevent superficial comparisons.

 -14Significant uncertainties remain in designing &
5.
scrubber to perform at high levels of reliability and removal efficiency. EPA has given little attention to the
difficulty of predicting the scrubber design level required at a particular facility.
EPA has not examined sludge disposal and water quality
6.
problems associated with scrubber operation in light of
recent water quality and solid waste legislation.
Finally, available data indicate that scrubber technology^
is further advanced on lower sulfur coal than higher sulfur coal.
Thus, a percentage reduction standard that provides for a
sliding scale on "middling" coals would afford greater flexibility
for new plants that attempt to burn higher sulfur coal but then
find that their scrubber does not meet design
\
Under a sliding scale approach, plants thai
in this manner could comply with the NSPS by shifting to a
slightly lower sulfur coal and thus having to meet a lesser
percentage reduction requirement.

In contrast, the conse-

quences of "missing the mark" under a uniform percentage reujlreinent may well deter use of higher sulfur coals.
Cost
The second factor that must be considered under section
111 is the "cost" of the proposed standard.

The EPA-DOE (IGF)

and UARG (NERA) "Round One" and "Round Two" and "Round Three"
macro-economic analyses show that the UARG and DOE alternatives
are less costly overall and more cost effective than the EPA
alternative.

In particular, the most recent analyses performed

by NERA for UARG indicate that:
1.

The proposed NSPS revisions will substantially increase

 -15the cost to the American public of the Clean Air Act including
the cost of electric generation but the increase would be
markedly less under the UARG proposal than under that made by
EPA.
(a) The UARG proposal would increase the cost impacts
Q EPA's proof the Clean Air Act by 3 Q r^r^f
posal would increase those costs by ]7^L percfip*"^
(b) Clean Air Act costs under the currently appl i
NSPS would be 4.38 billion per year bv 19ftQ: thff
be 6. 1 billion under the UARG proposal and
would
*
billion^ under trie E^A proposal In fhAf
"

1nmmt-ry proposal wmilH

by 1.7 billion per vear in 1990 and the EPA pro*posal would raise costs by 3.22 billion.
(c) The present value of the increase iq
thg
ith th
associa
1984-2020
ana about PZ billion under the EP
sal-- a differen
2.

Total utility emissions of sulfur dioxide would be

little affected by any of the alternatives.

At a $15 oil

price, total SC^ emissions would be 25.04 million tons in
1990 under the current NSPS, 23.20 million tons, or 7.2 percent
less, under the UARG proposal and 22.20 million tons, or 12
percent less, under the EPA proposal.
*/The results described above assume: (1) $15 oil prices
and eTectricity growth of 5.3 percent per year; (2) scrubber
reliability of 70 percent per module for full scrubbing of high
sulfur coal and 90 percent per module for partial scrubbing of
low sulfur coal; (3) scrubbers containing sufficient redundant
modules to assure 97 percent reliability; and (4) bypass of
scrubbers if they are unavailable under emergency conditions.
At higher oil prices these costs would be substantially increased.
At $18.20 per barrel the incremental cost of the UARG proposal
rises from 1.74 to $2.08 billion per year while the cost of the
EPA proposal rises from 3.22 to 4.51 billion.

 -163.

The incremental cost per ton of sulfur removal woul^v

be higher under the EPA proposal than under the UARG proposal I
I
but in either case these costs vastly exceed both the cost
per ton of SO^ removed under the applicable NSPS and state

^J

implementation plan requirements in effect prior to the 1977
Amendments to the Act and the estimates of the social benefit
of sulfur dioxide removal that have been made to date.
(a) Incremental cost per ton of SOn rgmov
$912 per ton""uncier the UARG proposal but $
peT LUll'lIlllltiT cne trA pr6!5osai.

-is

(b) Costs per ton of SOo removed under the Clean
Air Act prior to the 1977 Amendments were $158 per
ton. Thus, the incremental costs per ton of S02
removed under the EPA proposal are six times the
historic average. Moreover, the National Academy
of Sciences' liberal estimates put the benefits of
SOo removal to the public health and welfare at
$200 per ton.
4.

Each of the proposals will substantially decrease

I

utility consumption of coal and increase consumption of oil
and gas, but the EPA proposal is markedly worse in this regard
than either the UARG or DOE proposals.
(a) At a $15 oil price utility coal consumption
which is projected to be 1.23 billion tons in 1990
under current NSPS, falls by 77.8 million tons (to
1.15 billion) under the UARG proposal, and by
144.9 million tons (to 1.08 billion) under the EPA
proposal. Oil and gas consumption would be 1.14
million barrels per day in 1990 under current NSPS
and rises to 1.92 million barrels per day under the
UARG proposal (a rise of 780,000 barrels of oil per
day) and 2.63 million under the EPA proposal (a
rise of 1.4 million barrels per day).
(b) These shifts are less but still quite significant under an $18.20 per barrel oil price. At this
price oil consumption is 0.33 million barrels per day
in 1990 under current NSPS, 0.45 million under the
UARG proposal, and 0.59 million under the EPA proposal.

 -175.

The UARG proposal has little impact on the regional

distribution of coal production in the United States.

The EPA

proposal markedly diminishes Western coal production while
leaving
Eastern
production
virtually unchanged.
C
.-—...MI. .. •-- •••---•••• • - ••• ••••• ———————————

In the

Appalachian coal producing region, which is the most vulnerable
to employment losses during this period, the EPA proposal would
diminish projected production in 1990 under the current NSPS by
19 million tons.

The EPA proposal would result in increased pro­

duction in 1990 in only the Central region -- Ohio, Indiana,
Illinois -- and the increase is only 6.0 million tons.
6.

While reducing total SO^ emissions modestly, the EPA

proposal increases reliance on high sulfur coal.
^-mg^

,————^^.^^^E^^^^MMaaMBMMMMa^^^-M-^.^^-

.

__

_________

High sulfur

_____^^^mm,^^m^^,^^^^^

coal production in 1990 would increase from 27 percent of total
production under current NSPS to 30 percent under the EPA pro­
posal.

Under the UARG proposal, reliance on high sulfur coal

is essentially unchanged.
7.

Each of the S0 0 proposals substantially increases the

production of scrubber sludge whfrch may be designated as a
^ vXX^ y
^^•••^^^•••^•^^^^^
hazardous waste under the Resource Conservation and Recovery
Act.

However, the increase in sludge production under EPA's full

scrubbing proposal is nearly seven times that which would re­
sult from UARG's sliding scale.
(a) Scrubber sludge production in 1990 would be 44.4
million tons under the EPA proposal but only 33.3
million tons under the UARG proposal. Sludge produc­
tion would be 31.5 million tons in 1990 under the
current NSPS.

 -18(b) If scrubber sludge is classified as a hazardous
waste under the Resource Conservation and Recovery
Act, and if disposal costs were an additional $20
per ton as one EPA source has suggested, these addi­
tional disposal costs would increase the costs of
the EPA proposal by $222 million relative to the UARG
proposal.
All of the proposed NSPS revisions have a marked impact

8.

on the marginal cost of electricity but the impact of the EPA
proposal on marginal cost is much greater than UARG's.

The

EPA proposal increases marginal electric costs by 10 percent
over current NSPS.

The impact of the UARG proposal is about

3.0 percent.
9.

The impacts of the EPA proposal are extremely unequal

on a regional basis. In Texas, Kansas, Oklahoma, the Northern
Rockies and the Pacific Northwest the EPA proposal increases ^\f^
£+
^
is
impact
cost
The
percent.
16-19
by
marginal electric costs
5-6 percent in the Northeast, 9-10 percent in the Sout

&•

the remainder of
the Midwest anda»»*"IM^gll'IIH"BI"MM^^
10-15 percent ^^^in ^^^^^^_^»^^^^^^^^
^^^^^
^^^^MBM»a*Bi»»»MM
_

_____

The UARG proposal reduces these marginal cost increases to 7 per­
cent in the most heavily impacted areas, 2-3 percent in most of
the East and 3.5 percent in the remainder of the West.
11.

These sharply differential cost impacts of the EPA

proposal are likely to lead to regional disruption of employ­
ment in industries using substantial amounts of electricity.
Assuming this marked increase in marginal cost will be passed
on to industrial customers, the competitive positions in
international trade of the stone, clay, glass, primary metal,
chemical, petroleum, textile, and paper industries, which are
substantial users of electricity, are likely to be adversely
impacted by the EPA proposal.

 -19Nonair Quality Health and Environmental Impact
The third factor that must be considered by the Administrator
in establishing revised NSPS is "nonair quality health and
environmental impacts."

As discussed above, the EPA full.

scrubbing proposal would result in the production of much more
scrubber sludge than the UARG and DOE alternatives^

Less sludge

production means greater flexibility in power plant siting, which
can mean more environmentally compatible sites.

In addition,

the EPA proposal would require much more lime scrubbing^ as
distinguished from limestone scrubbing, than the UARG proposal.
The resulting increase in lime production would result in increased air and wateiT pollution and solid waste storage problems
at the lime production facilities.

Adverse nonair quality health

and environmental impacts are not, however, unique to the SC^
proposal.
Energy Requirements
The fourth factor that must be analyzed under section 111
is the "energy
NSPS.

requirements" associated with various alternative

As indicated above, the EPA full scrubbing proposals

results in greater oil consumption in the United States than
either the UARG or DOE alternatives.

This difference could

range between 300,000 to 800,000 barrels a day in 1990 under
various assumptions.

Most of this increased oil consumption

results from the economic penalty full scrubbing imposes and
which in turn discourages "phase out" or "phase down" of existing
oil and gas fired plants.

The EPA proposal would also preclude

use of bypass for reheat at plants burning lower sulfur coal and

 -20thus would impose a greater "energy penalty" on electric
generation than either the UARG or DOE alternatives. Moreover,
increased lime production required by the EPA proposal would
increase national energy consumption. And the EPA option would
also discourage strengthening of transportation facilities to
carry low sulfur coals from the West to the Midwest and East
(such as by rail to the Great Lakes, then to the Midwest and
East over the Lakes), thus, lessening the availability of low
sulfur coal for non-utility use.
2. Encouragement of "Innovative Technologies" and Use of
"Local Coals"_________________________________
The UARG proposal also better satisfies two other
legislative concerns reflected in the 1977 Amendments and their
legislative history than does EPA's proposal. Section 111 by its
nature and subsection lll(j) by its express language reflect
Congressional intent to encourage innovation. But section lll(j)
expressly provides that, in order to qualify for a waiver,
the EPA Administrator must find that the new technology will
achieve reductions in emissions at least equivalent to that
required by the then current NSPS. The EPA fixed percentage
reduction proposal would discourage, and in certain casespreclude, the granting of section lll(j) waivers for promising new_
technologies that may have fewer side effects than those upon
which EPA's proposal is based. Solvent refined coal, dry scrubbers,
^^_^^^__^f*~
^^_
and other promising technologies which may achieve 50-80 percent
removal might be abandoned by the utility industry if an 85 percent

 -21across-the-board standard were adopted.

And fluidized bed

combustion could be discouraged in view of the risk a company
would have to assume if the system did not fulfill design
expectations.
The-UARG and^DOE proposals would also better foster the use
pan would EPA's proposal.

While we do not

lective of the House Committee rises to the
point of transcending the express statutory requirements of
section lll(a)(l) to consider and balance the four statutory
factors, it (like the anticompetitive and other economic effects of
proposed standards) is appropriately taken into account in considering
"costs."

Employment effects in the coal industry, like all other

major secondary economic effects, are obviously relevant, just
as are effects on employment in the steel, aluminum, cement,
chemical, and other electric-energy intensive industries resulting
from higher electricity costs.
The 1CF and NERA analyses both show that none of the three
different proposals for S0 9 percentage reduction
is likely by itself to increase substantially regional employment
in the coal mining industry over that which would n
the other alternatives.

Coal production in 1990 under all of

the alternatives should be substantially greater than current
coal production.

The greatest increases are projected for the

West where coal production rises five to six-fold between 1977
and 1990 and in the Midwest where production triples.

The

 -22Appalachian region is projected to sustain the slowest growth
between 1976 and 1990.

These patterns emerge under all of the

alternative proposals; and consequently, for the West and Midwest
sections of the country employment in coal production is expected
to grow substantially under all of the proposed alternatives.
The only region of the country in which reduced employment is
possible is in the Appalachian region.

However, Appalachian —-^ l\j^tO

coal production is substantially higher under the DOE and UARG^-^ .^^
sliding scale alternatives than under the EPA full scrubbing
alternative.

>Ku*£>

( OCQ&

As a result, the risk of reduced employment in

s

the Appalachian region is less under either the UARG or DOE
proposal than under the EPA proposal.
3.

"Emergency" Bypass
Scrubber bypass provisions should permit operation of a

generating unit whenever required to preserve system integrity.
Criteria governing bypass should consider (1) necessary spinning
reserves; (2) area supply problems; (3) undesirability of con­
centrating spinning reserves in one unit; and (4) necessity to
look ahead to the load projections and reserves for the next
day before shutting down a malfunctioning unit.

We believe

that the EPA Staff and UARG are in substantial agreement on
this issue.
4. X^TJTlb/MBtu versus 1.5 Ib/MBtu "Ceiling1
While this issue has been obscured by the debate over the
full scrubbing option versus the sliding scale, it should not

 -23be forgotten.

There are substantial reserves of very high

sulfur Midwestern and Appalachian coal which will be barred
from use by the 1.2 Ib/MBtu emission limitation even if it
proves possible to achieve sustained, reliable 857e removal with
scrubbers on high sulfur coals.
VI. Proposed NSPS for Particulate
The notice accompanying the proposed NSPS and related EPA
background information discloses that several important issues
that are critical to developing a complete record have not been
assessed adequately by EPA in the formulation of its proposed
particulate standard.

For example,

(1) Critical information (control equipment design and
operating parameters, fuel characteristics, and operating
characteristics of the generating unit) necessary to eval­
uate and interpret the performance of particulate control
systems has not been included in much of the EPA data base.
(2) EPA 1 s background information contains virtually no
information on the overall cost-effectiveness on the pro­
posed particulate NSPS or the cost-effectiveness of various
control technologies in conjunction with FGD systems at
levels of control.
(3) EPA has not addressed adequately the potential effects
of proposed NSPS for NOX and SC>2 on the proposed particu­
late NSPS.
(4) EPA's data base is too small and incomplete to provide
sufficient statistical support for the proposed particulate
NSPS.
(5) EPA 1 s data base on baghouses is not representative of
the size and type of generating units that will be subject
to the particulate NSPS. That data base is insufficient
to support a claim that at present baghouses are adequately
"demonstrated" technology for control of particulate emis­
sions from large pulverized coal-fired generating units,
particularly those firing higher sulfur coals.

 -24-

(6) EPA's data base, comprised of results from a small
number of isolated tests (i.e., performance/acceptance
and compliance), does not represent the continuous per­
formance of particulate control systems during day-to­
day operation, including normal variations caused by
such occurrences as soot blowing and load fluctuations.
(7) EPA has not evaluated the ability of test methods to
measure accurately and precisely the very small mass of
particulate matter necessary to determine compliance with
the proposed NSPS.
UARG believes that the Agency's record does not include suffi­
cient data to support a finding that baghouses "have been demonstra­
ted" for large pulverized coal-fired generating units.

Moreover,

while UARG has no argument with the notion that, given suffi­
cient siting space and monetary resources, an ESP can be operated
to achieve the proposed standard under controlled performance
test conditions, EPA's data does not support the conclusion
that such a standard could be achieved on a continuous basis.
And we believe that, taking into consideration the cost, nonair quality health and environmental impacts and energy require­
ments associated with achieving the proposed standard with an
ESP, the data presented by EPA do not support the conclusion
that ESPs have been "adequately demonstrated" at the 0.03
Ib/MBtu level.
EPA's proposed particulate NSPS would also entirely pre­
clude the use of wet particulate matter scrubbers for both
low and high sulfur coals.

We believe a more prudent approach

would be to establish a standard that would encourage the use
(and thus further development) of baghouses, while at the same

 -25-

time allowing the use of other particulate control technologies,
i.e., ESP, and, where conditions are favorable, wet particulate
matter scrubbing in conjunction with SO^ scrubbing.
Finally, because of a lack of data regarding scrubber
slurry solids entrainment and sulfuric acid mist and other sulfate information, EPA's rulemaking record is at present inade­
quate to support any conclusion concerning an achievable
particulate standard as measured after the FGD device. Data
on these effects must be obtained and be reflected in the par­
ticulate NSPS ultimately established by EPA.
VII. Proposed NSPS for NOx
Neither EPA's proposed emission limitations nor the
continuous monitoring compliance requirement for nitrogen oxides
are supported by the September 19 record.
Most of the data available to the Agency is from perfor­
mance tests which do not support a continuous compliance require­
ment. The Agency does present continuous monitoring data in the
record but information just made available shows that it is no
longer valid. As a result, EPA has no basis for determining
an appropriate emission limitation or averaging time when a
continuous compliance requirement is imposed.
EPA should also be cautious in using the limited data that
is available to revise the numerical emission limitations, even

 -26when compliance is determined solely on a performance test basis,
for the following reasons:
1.

Most of the data is from existing boilers
retrofitted with individual NOx control
techniques.

2.

These tests do not show how much the combi­
nation of recommended control techniques can
reduce NOx emissions from modern boilers.

3.

Most of EPA's data are derived from the
inherently low NOx emitting tangential boiler.

4.

EPA recognizes that the other pulverized coalfired boiler types have not been shown to be
as effective in reducing NOx emissions as the
tangential boiler.

5.

A sufficient: analysis of the adverse side
effects of low NOx operation has not been
performed. The potential side effects
include boiler corrosion, slagging, reduced
unit efficiency, decreased operating margins,
enhanced safety hazards, and increased emission
of other air pollutants.

6.

EPA has failed to undertake an adequate assess­
ment of the cost and energy requirements of
complying with the proposed NOX NSPS.

7.

Data from the boilers subject to current NSPS
contradict EPA's conclusion that each boilertype/fuel combination can attain the proposed
emission limitations even on a performance test
basis.

Moreover, the Agency should not promulgate standards that
prejudice the viability of boiler types and cannot promulgate
such regulations without an adequate analysis of the potential
anticompetitive effects of such regulations.
While UARG recognizes EPA's duty to revise the NSPS for the

 -27the electric utility steam generating unit category, those new
standards can reflect only the best technological system of
continuous emission reduction that (taking into account cost,
nonair quality health and environmental impacts and energy
requirements) have been adequately demonstrated.

The limited

data available to the Agency do not justify either a change
in the emission limitation or the compliance requirement of
the current or proposed NOJv NSPS.

If the Agency feels com-

pelled to act on the data available, UARG recommends that EPA
continue its practice of subcategorizing electric utility steam
generators on the basis of boiler type and fuel type and that
the Agency act cautiously.
The better course would be the continued acquisition of
data from continuous monitors on continuous NOx emissions through
enforcement of its existing monitoring regulations.

As sufficient

data become available on each boiler-type/fuel combination, con­
clusions as to the appropriate emission limitation for either
performance tests or continuous compliance (and any averaging
time requirement) can be made at or before the next statutory
mandated review of these standards four years from now.
VIII.

Continuous .Monitors and Compliance Requirements

The Agency shoula not establish standards on the basis
of intermittent performance tests and require compliance with
those standards on a continuous basis, as it has done here.
The continuous compliance requirement can only be based on a
history of continuous emissions or an analysis of the variations

 -28expected from the performance test levels.

The record is

deficient in both respects for SC^, particulate, and NO .
In addition, continuous monitors are not an adequately
demonstrated technology for determining compliance with the
proposed NSPS.

The recognized errors inherent and allowed in

their output prevent an accurate measurement of emissions.

They

have not been developed to the point that even a primary and
redundant monitoring system can satisfy EPA's data acquisition
requirements.

As a result, EPA's recomirended backup, manual

sampling and analysis, may be the rule rather than the exception
even though the Agency has not shown that these manual procedures
can produce the data required in a reasonable and timely manner.
Finally, the Agency has not justified the need for continuous
data to show compliance with the proposed NSPS.
In light of these considerations, the Agency should opt for
a continuation of the present method of determining compliance,
intermittent performance testing.

Such testing cam be conducted

at the discretion of the regulatory agency and interim indications
of emissions levels can be provided by continuous monitors when
they operate or documentation of proper operation and main­
tenance when they do not.
IX.

DUE PROCESS

Finally, I'd like to put to rest an argument we have heard
with increasing frequency lately.

The thrust of that argument

is that any difficulties in complying with regulations due to

 -29impossibility can be remedied by the exercise of "prosecutorial
discretion," It must be remembered that the Agency, is formulating
regulations that are reenforced by severe criminal penalties.
It is important that these regulations be understandable to
those affected and command only what is possible.

Congress

clearly intended that when it designed Section 111(a).

By the

use of the word "demonstrated" Congress obviously intended that
the technology upon which an NSPS was based should be well known
and clearly capable of achieving the standard.
It is also important to note that the notion of "forcing"
technology in order to protect the public health and welfare is
not at issue here.

Accordingly, the Supreme Court's decision in

Union Electric cannot be invoked.

If EPA were to ignore

the statutory criteria of Section 111(a) and promulgate new
source performance standards that command the impossible or
nearly impossible, then the regulations would violate DUE PROCESS.
And this constitutional infirmity could not be cured by any
"law of prosecutorial discretion."
This "law of prosecutorial discretion" is a contradiction
in terms and cannot be squared with the concept of due process
and other constitutional safeguards.

One of the axioms under­

lying our constitutional system is that "We live under the rule
of law and not of men."

This presupposes, among other things,

that what is or is not a crime will be spelled out clearly in
advance so that men can voluntarily obey the law and so they
will not commit crimes in ignorance of the law.

See, e.g.,

 -30Republic Steel Co. v. Train, 559 F.2d 91 (6th Cir. 1977),
United States v. Pennsylvania Industrial Chemical Corp., 411 U.S.
655, 674 (1973).

Thus our Constitution expressly prohibits ex

post facto laws and the concept of due process clearly encompasses
a requirement of mens rea for most offenses.
Since those in government charged with the detection
and prosecution of crime were entrusted with the awesome
powers of the State, and could commit abuses from good
motives as well as bad ones, the Founding Fathers thought
it necessary to impose a number of specific checks on
government officials in the exercise of those powers.
All these limitations make the definition, detection,
and punishment of crime more difficult.

The Constitution

thus weighed the needs of those in charge of the government
for flexibility, convenience, and discretion against the
needs of a free society and deliberately chose to restrict the
former.
Applying those principles to this proceeding, it is
clear that the EPA staff proposal poses concerns of constitu­
tional dimensions.

In many instances, individuals could

not know, and could not fairly be expected to know, when they
might be committing a crime.

In some instances, they could also

have been ordered to do the physically impossible and their
failure to perform could be a crime.

The injustices of such

 -31a system are not cured by hoping that prosecutors will
exercise their discretion benignly.

Such an approach is

anathema to the centuries-old traditions of Anglo-American
law and clearly transgresses our Constitution.
Reid v. Covert, 354 U.S. 1, 41 (1957).

See, e.g.,