Docket
Number
F­
97­
CS4A­
FFFFF
EVALUATION
AND
USE
OF
DATA
SUBMITTED
BY
THE
CEMENT
KILN
RECYCLING
COALITION
(
CKRC)

Economics,
Methods,
and
Risk
Analysis
Division
Office
of
Solid
Waste
U.
S.
Environmental
Protection
Agency
401
M
Street,
SW
Washington,
D.
C.
20460
June
30,
1999
ACKNOWLEDGMENTS
The
Agency
recognizes
Industrial
Economics,
Incorporated
(
IEc),
2067
Massachusetts
Avenue,
Cambridge,
MA
02140,
for
the
overall
organization
and
development
of
this
report.
IEc
developed
the
database,
methodology,
and
analytical
model
that
allowed
for
the
comparative
review
and
comprehensive
analyses
of
the
CKRC
data.
Personnel
from
EPA's
Office
of
Solid
Waste
and
Office
of
General
Council
provided
guidance
and
review.
1
INTRODUCTION
This
report
compares
the
information
found
in
the
Cement
Kiln
Recycling
Coalition's
(
CKRC)
September
27,
1997
memo,
prepared
by
Environomics,
with
the
data
and
results
from
EPA's
recent
modeling
(
December
1998)
of
economic
impacts
for
the
hazardous
waste
combustion
MACT
standards.
The
key
findings
from
our
review
and
analysis
are
outlined
below:


EPA's
baseline
costing
approach
no
longer
relies
on
the
concept
of
model
plants,
and
therefore
is
more
likely
to
accurately
predict
outcomes
at
specific
facilities,
addressing
a
major
CKRC
concern
about
the
structure
of
EPA's
analytic
approach.


In
shifting
to
a
plant
specific
approach,
EPA
was
unable
to
use
CKRC's
information
directly
in
the
economic
impacts
model
because,
for
reasons
of
confidentiality,
CKRC
presented
the
data
in
ways
that
prevent
us
from
linking
specific
costs
and
revenues
to
individual
plants.
CKRC's
information
is
useful,
however,
for
evaluating
the
means
and
ranges
of
key
model
inputs
and
outputs.


The
data
on
the
median
total
costs
per
ton
of
burning
HWDF,
reported
in
the
Environomics
1997
memorandum,
are
within
7
percent
of
the
median
values
used
in
the
1998
economic
analysis.
The
ranges
of
costs
per
ton
across
plants
are
also
relatively
consistent
(
CKRC
$
11
to
$
103
versus
EPA
$
35
to
$
120).


The
major
difference
between
the
EPA
analysis
and
CKRC's
is
the
estimated
revenue
per
ton
of
hazardous
waste
received
by
the
cement
kilns.
EPA's
projected
average
revenue
for
liquids
burned
at
cement
kilns
is
approximately
$
136/
ton
versus
$
67/
ton
reported
by
CKRC.


As
a
result,
EPA's
December
1998
modeling
of
economic
impacts
projects
higher
revenues
and
consequently
higher
profits
for
the
cement
kilns,
relative
to
CKRC's
estimates.


The
results
of
our
sensitivity
analysis
using
the
EPA
economic
impacts
model
and
CKRC's
mean
values
for
waste­
burning
revenues
and
fuel
savings
indicates
that
the
same
number
of
incremental
facility
exits
occur
under
both
scenarios
examined
(
SVM
level
of
650
ug/
dscm
vs.
240
ug/
dscm).


The
sensitivity
analysis
also
shows
that,
over
the
longer
term,
one
to
two
facilities
move
from
positive
to
negative
profits
as
a
result
of
the
MACT
rule.
This
is
consistent
with
our
findings
from
the
December
1998
analysis.
1
We
define
the
"
long­
term"
as
the
capital
replacement
cycle,
and
the"
short­
term"
as
the
period
when
no
new
capital
investments
are
required.
Thus,
"
long
term"
viability
means
that
kilns
cover
their
operation,
maintenance,
and
capital
replacement
costs,
while
"
short­
term"
viability
means
that
kilns
only
need
to
cover
their
operation
and
maintenance
costs.

2
This
is
due
to
the
fewer
number
of
facility
exits
under
the
short­
term
vs.
the
long­
term
baseline.
That
is,
over
the
long
term
capital
replacement
cycle,
we
estimate
a
greater
number
of
facilities
will
exit
the
waste
burning
market
(
5
facilities)
even
in
the
absence
of
the
HWC
MACT.
In
the
short
term,
when
cement
plants
only
need
to
cover
their
short
term
operating
and
maintenance
costs,
only
two
facilities
currently
appear
non­
viable.
Thus,
at
the
50%
price
pass­
through,
the
total
number
of
market
exits
is
4
in
the
short
term
(
6
total
minus
2
in
the
baseline)
and
only
1
in
the
long
term
(
6
total
minus
5
in
the
baseline).
Our
view
remains
that
closures
of
plants
with
negative
profits
today
should
not
be
attributed
to
this
rule.
Plants
that
are
currently
operating
at
a
loss
are
expected
to
exit
the
market
under
baseline
conditions.

4
To
evaluate
the
effect
of
these
revenue
differences
on
the
projected
economic
impacts
of
the
hazardous
waste
combustion
MACT
standards,
we:

(
1)
Conducted
a
sensitivity
analysis
of
EPA's
economic
impact
model
results
using
CKRC's
average
data
on
revenues;
and
(
2)
Compared
the
estimated
compliance
costs
of
the
rule
(
after
price
pass
through)
with
CKRC's
plant­
specific
profitability
estimates
to
determine
whether
using
CKRC's
information
would
yield
significantly
different
estimates
of
economic
impacts.

These
analyses
suggest
that,
over
the
long
term1
capital
replacement
cycle,
the
rule
would
cause
only
one
to
two
facilities
to
move
from
positive
to
negative
hazardous
waste­
burning
profits,
the
same
result
indicated
by
EPA's
December
1998
modeling.
In
the
short­
term,
however,
when
firms
only
need
to
burn
enough
waste
to
cover
the
operation
and
maintenance
costs
associated
with
wasteburning
the
sensitivity
analysis
predicts
that
between
three
and
four
cement
plants
will
find
it
no
longer
profitable
to
burn
waste.
2
Also,
if
demand
for
HWDF
is
more
elastic,
use
of
CKRC's
plantspecific
data
suggests
impacts
on
cement
kilns
would
be
greater
(
four
to
five
facility
exits)
than
estimated
by
EPA's
December
1998
modeling
results
(
one
to
two
facility
exits).

For
both
analyses,
we
evaluated
economic
impacts
using
compliance
cost
estimates
for
a
SVM
emissions
standard
of
650
µ
g/
dscm
and
for
a
more
stringent
standard
of
240
µ
g/
dscm.
Economic
impact
results
under
these
two
scenarios
are
comparable.

Throughout
this
report,
it
is
important
to
recognize
that
the
revenues
reported
in
CKRC's
survey
vary
widely,
presenting
considerable
uncertainty
about
the
"
correct"
level
to
assume
in
EPA's
model.
EPA
is
concerned
about
two
facets
of
the
revenues
reported
by
CKRC
members.
3
CKRC
figures
converted
from
metric
tons.

4
In
fact,
Environomics
notes
that
the
Group
1
plants
"
apportion
a
percentage
of
HWDF
revenue
to
the
cement
companies.
Within
Group
1
there
is
a
great
variety
in
the
specifics
of
the
contractual
arrangements
regarding
shared
costs,
revenues
and
savings
on
conventional
fuels."
(
Environomics
1997,
11­
12).
This
statement
contributes
to
EPA's
uncertainty
about
whether
the
revenues
reported
by
these
plants
reflect
the
full
waste­
burning
revenues
ultimately
received
by
the
plants.

5
Because
the
specific
CKRC
Group
plants
cannot
be
identified,
we
cannot
adjust
the
waste­
burning
costs
in
the
EPA
model.

6
Environomics
received
"
reasonably
complete"
responses
to
the
CKRC
survey
from
17
plants.
In
CKRC's
original
August
1996
comments,
they
indicated
that
twenty
cement
plants
used
HWDF.
In
the
September
1997
memo,
CKRC
noted
that
"[
s]
ince
August
1996,
two
of
the
twenty
plants
that
were
using
HWDF
have
ceased
doing
so."
This
corresponds
with
EPA's
universe,
which
includes
18
waste­
burning
cement
plants.

5

First,
EPA
is
concerned
that
profit­
sharing
arrangements
between
cement
plants
and
fuel
blenders/
managers
may
result
in
under
reported
revenues
by
some
of
the
kilns.
The
average
revenues
reported
by
kilns
with
integrated
operations
(
CKRC's
"
Group
3"
plants)
is
$
90.96
per
ton
whereas
the
average
revenues
reported
by
kilns
that
contract
with
a
separate
company
to
handle
HWDF
operations
(
CKRC's
"
Group
1"
plants)
is
only
$
33.09
per
ton
(
Environomics
1997,
11).
3

Second,
because
EPA's
baseline
cost
estimates
include
costs
for
waste
handling,
storage,
and
testing,
using
the
revenues
reported
by
CKRC's
Group
1
plants
in
the
EPA
economic
impact
model
may
lead
to
underestimated
waste­
burning
profits
in
the
EPA
model.
4
That
is,
while
the
lower
revenues
reported
by
the
Group
1
CKRC
plants
may
represent
actual
waste­
burning
revenues
for
these
plants,
using
the
EPA
costs
with
the
CKRC
revenues
will
lead
to
underestimated
profits.
5
This
occurs
because
EPA's
analysis
includes
baseline
costs
that
will
not
be
incurred
by
plants
that
contract
for
hazardous
waste
fuel
handling
such
as
storage,
testing,
and
other
handling
costs.

The
remainder
of
this
memo
is
organized
into
three
sections.
We
first
present
a
detailed
comparison
of
data
input
assumptions
­­
hazardous
waste­
burning
costs,
revenues,
and
conventional
energy
savings.
Second,
we
present
the
analysis
comparing
compliance
costs
with
CKRC's
profitability
estimates
for
kilns.
The
final
section
provides
a
sensitivity
analysis
of
EPA's
economic
impacts
modeling
using
CKRC's
mean
values
for
revenues
and
fuel
savings
at
17
hazardous
wasteburning
cement
plants.
6
7
The
baseline
cost
model
randomly
imputes
11
clinker
production
penalty
values
reported
by
CKRC,
stratified
into
two
categories
(
HWDF
content
as
a
percentage
of
total
energy
usage).

8
Converted
from
metric
tons
(
Environomics
1997,
15).
We
assume
that
CKRC
figures
are
in
1995
dollars
because
they
were
reported
in
a
survey
conducted
in
1995.

9
Converted
from
metric
tons
(
Environomics
1997,
32).
We
assume
that
CKRC
figures
are
in
1995
dollars
because
they
were
reported
in
a
survey
conducted
in
1995.

6
COMPARISON
OF
DATA
ASSUMPTIONS
Costs
of
Waste
Burning
at
Cement
Kilns
In
response
to
CKRC's
comments
on
our
proposed
rule,
and
our
own
concerns
about
the
baseline
cost
modeling,
EPA
has
significantly
revised
the
baseline
cost
methodology.
Instead
of
using
a
model
plant
approach
which
bases
waste­
burning
costs
on
kiln
characteristics
such
as
"
small
vs.
large"
and
"
wet
vs.
dry,"
the
1998
economic
analysis
provides
cost
estimates
based
on
specific
kiln
parameters.
This
change
addresses
one
of
CKRC's
three
primary
concerns:

The
small/
large
and
wet/
dry
distinctions
that
form
the
basis
for
the
model
plant
analysis
in
the
draft
RIA
do
not
appropriately
reflect
the
variation
across
actual
plants
in
the
economics
of
using
HWDF.
(
Environomics
1997,
2)

In
addition,
the
operating
and
maintenance
costs
have
been
divided
into
fixed
and
variable
cost
components;
variable
costs
include
costs
such
as
clinker
production
penalty
costs
and
acid
gasabsorbing
sorbents
that
vary
with
the
amount
of
waste
being
burned.
Clinker
production
penalties
are
imputed
from
CKRC's
1996
comments
to
the
proposed
rule.
7
The
clinker
penalties
reported
in
CKRC's
survey
range
from
zero
to
$
46.85
per
ton;
for
those
cement
plants
that
reported
experiencing
adverse
effects
on
clinker
production,
the
median
clinker
penalty
is
$
20.64
per
ton.
8
Overall,
the
total
cost
of
waste
burning
reported
in
the
Environomics
September
1997
memo
ranges
from
$
11
to
$
103
per
ton
with
a
median
value
of
$
72
per
ton,
9
which
closely
mirrors
the
baseline
cost
range
used
in
EPA's
December
1998
economic
analysis:
$
35
to
$
121
per
ton,
with
a
median
value
of
$
67
per
ton,
a
difference
of
only
7
percent
at
the
median.

Hazardous
Waste­
Burning
Revenues
The
price
estimates
we
use
in
the
1998
Assessment
to
calculate
hazardous
waste
burning
revenues
at
cement
kilns
represent
the
prices
received
by
cement
kilns,
and
not
by
fuel
managers.
This
approach
agrees
with
Environomics'
recommendation:

We
believe
that
the
revenues
that
accrue
to
the
cement
kilns
are
far
more
relevant
to
assessing
the
impact
of
the
proposed
MACT
rule
than
are
revenues
received
by
the
fuels
managers.
(
Environomics
1997,
10)
10
When
speaking
with
fuel
blenders/
managers,
we
asked
the
price
charged
by
cement
plants,
and
not
the
price
they
charge
to
generators.

11
We
categorized
waste
burned
at
specific
cement
kilns
by
using
the
three
form
categories
reported
in
the
EPA's
Biennial
Reporting
System
(
BRS)
database.
We
then
further
segmented
the
waste
as
more­
or
less­
highly
contaminated
by
using
average
figures.
That
is,
for
all
energy
recovery
systems
included
in
the
NHWCS,
we
determined
the
average
tonnage
of
highly
contaminated,
liquids
with
suspended
solids,
and
comparable
fuels.
We
define
"
highly
contaminated
liquids"
as
those
liquids
with
more
than
5
percent
halogens,
17.5
percent
water,
5
ppm
mercury,
3,000
ppm
lead,
or
200
ppm
cadmium.
Comparable
fuels
are
defined
using
EPA's
comparable
fuel
specifications.

7
In
addition,
we
have
moved
from
the
"
pricing­
by­
waste­
form"
approach
that
used
average
prices
for
liquids
and
solids
(
from
the
EI
Digest
survey),
to
a
more
complex
pricing
approach
in
which
prices
vary
by
waste
form
as
well
as
by
heat
content,
water
content,
and
contamination
level
(
e.
g.,
halogen,
metals).

Pricing
data
come
from
conversations
with
industry
representatives
such
as
fuel
blenders/
managers,
as
well
as
industry
trade
publications.
10
Waste
management
prices
vary
widely
(
prices
reported
range
from
$
20
to
$
1,281
per
ton)
and
depend
on
several
factors.
These
include
the
waste
form
(
solid/
liquid/
sludge),
heat
content,
method
of
delivery
(
e.
g.,
bulk
versus
drum),
and
contamination
level
(
e.
g.,
metals
or
chlorine
content).
In
addition,
regulatory
constraints
(
e.
g.,
prohibitions
against
burning
certain
types
of
wastes)
and
technical
constraints
(
e.
g.,
adverse
effects
of
certain
waste
streams
on
cement
product
quality)
also
influence
combustion
prices.

Although
data
limitations
prevent
us
from
accounting
for
all
factors,
the
information
on
heat
content
and
constituent
concentrations
from
EPA's
National
Hazardous
Waste
Constituent
Survey
(
NHWCS)
allows
us
to
improve
the
characterization
of
combusted
waste.
11
The
result
from
our
data
analysis
of
the
NHWCS,
along
with
discussions
with
industry
representatives,
is
seven
categories
of
waste
types
to
which
we
assign
prices.
8
Exhibit
1
HAZARDOUS
WASTE
COMBUSTION
PRICE
COMPARISON
Hazardous
Waste
Revenues
per
Ton
Received
by
Cement
Kilns
Type
EPA
Prices
CKRC
Prices
Liquids
$
20
"
clean
liquids"
$
70
"
with
suspended
solids"
$
301
"
highly
contaminated"
$
58.40
Solids
$
683
"
less
contaminated"
$
309.20
Sludges
$
320
"
less
contaminated"
­­

Weighted
Average
$
136
(
liquids)
$
172
(
liquids,
solids,
and
sludges)
(
see
Note
1)
$
67
($
15.93
to
$
138.22)
(
see
Note
2)

NOTES:
1.
EPA's
weighted
average
price
for
cement
kilns
assumes
liquids
are
comprised
of
6%
clean
liquids,
64%
liquids
with
suspended
solids,
and
30%
more
highly
contaminated
liquids;
solids
are
comprised
solely
of
"
less
contaminated"
solids.
The
average
mix
of
hazardous
waste
burned
in
cement
kilns
in
the
EPA
economic
impact
model
is
91%
liquids,
6%
solids,
and
3%
sludges.
2.
CKRC
HWDF
revenues
range
from
$
15.93
per
ton
to
$
138
per
ton
(
converted
to
short
tons
from
Table
2,
Environomics
1997,
11).
3.
CKRC
weighted
average
prices
for
liquids
and
solids
is
from
Table
1,
Environomics
1997,
9.
4.
We
assume
that
CKRC
figures
are
in
1995
dollars
because
they
were
reported
in
a
survey
conducted
in
1995.
5.
EPA's
prices
are
in
1996
dollars.
(
Note:
The
one­
year
difference
between
the
1995
CKRC
prices
and
the
1996
EPA
numbers
may
contribute
to
a
small
portion
of
the
difference
between
the
EPA
and
CKRC
revenue
estimates.
6.
CKRC
does
not
report
revenues
received
for
burning
sludges.

There
are
a
number
of
possible
explanations
for
the
differences
between
CKRC's
and
EPA's
revenue
estimates.
We
may
not
be
accurately
categorizing
wastes
based
on
their
levels
of
contamination.
NHWCS
only
represents
about
50
percent
of
the
waste
burned
at
kilns,
and
this
may
be
biased
toward
more
contaminated
materials.
Alternatively,
our
decision
rule
for
identifying
highly
contaminated
waste
may
not
be
fully
reflective
of
industry
practices.
If
either
of
these
possibilities
is
true,
we
will
tend
to
overstate
cement
kiln
revenues
and
subsequently
profitability
(
since
profits
are
revenues
minus
costs).
12
EPA
is
also
concerned
that
these
profit­
sharing
arrangements
may
already
be
captured
in
the
wasteburning
revenues.
That
is,
if
the
fuels
managers
bring
in
higher­
Btu
wastes,
the
cement
plants
will
charge
a
lower
price
to
burn
these
wastes
(
representing
lower
costs
to
the
fuels
manager).
If
this
is
the
case,
using
CKRC's
values
may
understate
the
total
benefits
realized
by
the
cement
plants
of
burning
hazardous
waste.

9
CKRC's
reported
revenues
may
also
not
be
truly
reflective
of
the
full
revenues
received
by
kilns
for
waste
burning.
Due
to
the
different
arrangements
between
cement
plants
and
fuel
blenders/
managers,
a
disproportionate
amount
of
revenues
(
and
profits)
may
be
captured
by
the
fuel
blenders/
managers,
when
in
fact
they
are
spread
more
evenly
across
the
blenders/
managers
and
the
cement
plants.
The
wide
variability
in
revenues
and
the
confusing
aspects
of
the
cement
plant/
fuels
operations
adds
further
uncertainty
to
the
selection
of
the
"
correct"
revenue
levels
for
cement
plants.

Conventional
Fuel
Savings
A
second
area
where
our
estimates
of
the
economic
benefits
of
burning
HWDF
differ
from
CKRC's
is
in
our
estimates
of
savings
on
purchases
of
conventional
fuels.
The
Environomics'
1997
memo
reports
a
wide
range
of
variability
in
conventional
energy
savings
from
the
use
of
hazardous
waste­
derived
fuels.
As
shown
in
Exhibit
2,
CKRC
energy
savings
range
from
$
7.86
to
$
42.81
per
ton,
while
heating
values
range
only
from
10,260
to
12,814
Btu/
lb.

EPA
has
some
concerns
about
using
the
CKRC
energy
savings
numbers
directly
in
the
model.
First,
as
is
the
case
for
other
model
inputs,
we
have
no
way
to
link
savings
to
specific
plants.
Due
to
the
wide
variability,
use
of
an
average
or
median
value
may
not
provide
a
reliable
estimate
of
actual
savings.
Second,
we
have
a
concern
that
the
energy
savings
estimated
by
CKRC
have
the
potential
to
understate
the
full
energy
savings
because
the
reported
savings
"
represent
only
the
portion
of
savings
that
accrues
to
the
cement
plants.
There
are
several
plants
which
share
savings
with
their
fuels
managers
based
on
contractual
arrangements."
(
CKRC
1997,
13)
As
such,
we
would
expect
the
CKRC
estimate
to
be
lower
than
the
EPA
estimate.
12
13
EPA's
analysis
of
waste
management
alternatives
is
found
in
Appendix
F
of
the
Final
Assessment
document,
"
Economic
Analysis
of
Waste
Minimization
Alternatives
to
Hazardous
Waste
Combustion."

14
Recent
history
also
suggests
inelastic
demand.
As
prices
fell
through
the
1990s,
demand
remained
roughly
constant.
We
have
no
basis
for
believing
that
the
factors
influencing
demand
have
changed
dramatically
during
this
time
period.
As
a
result,
we
would
expect
the
same
inelastic
response
as
the
supply
curves
shift
upward
to
reflect
the
new
regulatory
costs.

10
Exhibit
2
Comparison
of
Data
Assumptions
for
Estimating
Conventional
Energy
Savings
HWDF
Energy
Content
(
Btu/
lb)
Price
per
Million
Btu
of
Convention
Fuel
Average
Energy
Savings
per
Ton
of
HWDF
Burned
Reported
by
CKRC
11,698
(
10,260
­
12,814)
$
1.27
($
0.52
­
$
1.72)
$
26.39
($
7.86
­
$
42.81)

Assumption
used
by
EPA
12,811
$
1.31
$
33.62
NOTES:
1.
CKRC
reports
that
coal,
or
less
often
coke,
would
serve
as
the
replacement
fuel.
EPA
assumes
a
conventional
fuel
mix
of
91.1%
coal
and
8.9%
natural
gas
(
Portland
Cement
Association
1996).

2.
CKRC
convention
fuel
savings
data
from
Table
3,
Environomics
1997,
14.

3.
EPA's
price
and
energy
content
for
conventional
fuel
is
from
the
Energy
Information
Administration.

ALTERNATIVE
ANALYSIS
OF
ECONOMIC
IMPACTS
USING
CKRC
PLANT­
SPECIFIC
DATA
The
fundamental
issue
of
concern
to
both
EPA
and
CKRC
is
the
degree
to
which
the
baseline
data
assumptions
affect
the
estimated
economic
impacts
of
the
MACT
Standards.
While
we
cannot
directly
input
the
plant­
specific
CKRC
data
into
our
model,
it
is
possible
to
use
their
data
to
develop
an
alternative
assessment
of
the
MACT
rule's
impacts
on
cement
plants.
This
can
be
done
in
a
rough
fashion
by
comparing
compliance
costs
per
ton
(
after
price
pass­
through)
with
CKRC's
plant­
by­
plant
profitability
estimates.
We
compare
compliance
costs
for
MACT
standards
that
include
an
SVM
emissions
limit
of
650
ug/
dscm,
and
at
a
more
stringent
240
ug/
dscm
SVM
emissions
limit.
To
do
this,
we
must
first
evaluate
the
ability
of
cement
kilns
to
increase
prices
for
hazardous
waste
burning,
thus
passing
through
a
portion
of
the
compliance
costs
to
generators
or
waste
management
intermediaries
(
e.
g.,
fuel
blenders
or
fuel
managers).
EPA's
analysis
of
waste
management
alternatives
suggests
that
substitutes
for
combustion
are
not
readily
available,
resulting
in
a
relatively
inelastic
demand
for
waste
combustion
services.
13,
14
Moreover,
in
the
current
market,
further
cost
absorption
for
the
combustion
industry
seems
unlikely,
given
the
price
competition
in
the
15
At
a
starting
price
of
$
70/
ton,
the
waste
minimization
analysis
suggests
a
price
elasticity
of
demand
at
­
0.21,
an
inelastic
demand.
This
is
roughly
consistent
with
our
75%
price
pass­
through
assumption.
(
Please
refer
to
Table
5
of
Appendix
F
of
the
Final
Assessment
document
for
the
complete
analysis.)

11
1990s
that
has
reduced
profit
margins.
In
light
of
these
two
considerations
­­
the
inelasticity
of
demand,
and
the
inability
for
the
industry
to
absorb
much
of
the
additional
regulatory
costs
­­
we
expect
combustion
firms
will
be
able
to
pass
through
a
significant
portion
of
the
cost
increase
due
to
the
new
regulations.

As
shown
in
Exhibit
3,
if
we
assume
that
combustion
demand
is
relatively
inelastic15,
such
that
cement
kilns
can
pass
through
75
percent
of
the
compliance
costs
to
their
"
customers"
(
which
in
the
case
of
cement
kilns
are
fuels
managers
and
fuel
blenders),
we
find
that
only
one
cement
kiln
moves
from
positive
to
negative
hazardous
waste­
burning
profits.
Only
one
cement
kiln,
therefore,
is
likely
to
exit
the
waste­
burning
market
as
a
result
of
the
MACT
rule
at
the
SVM
emission
standard
of
650
ug/
dscm
(
Our
view
remains
that
closures
of
plants
with
negative
profits
today
should
not
be
attributed
to
this
rule.
Plants
that
are
currently
operating
at
a
loss
are
expected
to
exit
the
market
under
baseline
conditions).
This
result
does
not
change
with
the
more
stringent
SVM
standard
of
240
ug/
dscm,
as
shown
in
Exhibit
4.
This
is
the
same
result
we
get
from
our
economic
model,
as
reported
in
EPA's
December
1998
Assessment
document.

A
key
awareness
derived
from
directly
applying
the
CKRC
profitability
data
is
that
waste
burning
activities
become
more
sensitive
to
the
price
elasticity.
At
more
elastic
waste­
burning
demand,
the
number
of
facilities
that
stop
burning
waste
could
increase
more
than
is
predicted
by
the
EPA
model.
While
this
is
possible,
our
best
estimate
of
elasticity
suggests
that
this
will
not
be
the
case.
As
a
result,
while
we
recognize
the
differences
between
CKRC's
and
EPA's
estimates
of
facility
revenues
and
profits,
we
do
not
believe
these
differences
have
a
substantial
impact
on
estimates
of
the
number
of
facilities
that
will
cease
burning
hazardous
waste
as
a
result
of
the
rule.
12
Exhibit
3
HAZARDOUS
WASTE­
BURNING
PROFITABILITY
OF
CEMENT
KILNS
USING
CKRC
BASELINE
PROFIT
ESTIMATES
(
SVM
=
650ug/
dscm)

CKRC
HWDF
Profits
($/
ton)
Profitable
in
the
Baseline?
Profitable
with
MACT
cost
increases
of
$
17/
ton?
(
50%
price
pass­
through)
Profitable
with
MACT
cost
increases
of
$
8.50/
ton?
(
75%
price
pass­
through)

­
34.18
no
no
no
­
28.56
no
no
no
­
12.12
no
no
no
­
10.93
no
no
no
5.74
yes
NO
NO
10.44
yes
NO
yes
13.93
yes
NO
yes
14.29
yes
NO
yes
17.4
yes
yes
yes
21.38
yes
yes
yes
24.26
yes
yes
yes
24.93
yes
yes
yes
61.11
yes
yes
yes
61.93
yes
yes
yes
64.17
yes
yes
yes
64.84
yes
yes
yes
74.80
yes
yes
yes
Number
of
plants
that
will
exit
the
waste
burning
market,
incremental
to
the
baseline:
4
1
NOTES:
1.
Compliance
cost
estimates
are
based
on
EPA
engineering
costs
that
incorporate
a
design
safety
factor
of
70
percent
(
i.
e.,
pollution
control
measures
are
assigned
such
that
kilns
can
meet
emission
levels
that
are
70%
of
the
MACT
standards,
on
average).
2.
Compliance
costs
do
not
include
PM
CEM
costs.
3.
CKRC
HWDF
waste­
burning
profits
from
the
1997
Environomics
memo,
Table
7,
page
19
(
converted
from
metric
tons).
13
Exhibit
4
HAZARDOUS
WASTE­
BURNING
PROFITABILITY
OF
CEMENT
KILNS
USING
CKRC
BASELINE
PROFIT
ESTIMATES
(
SVM
=
240ug/
dscm)

CKRC
HWDF
Profits
($/
ton)
Profitable
in
the
Baseline?
Profitable
with
MACT
cost
increases
of
$
18.50/
ton?
(
50%
price
pass­
through)
Profitable
with
MACT
cost
increases
of
$
9/
ton?
(
75%
price
pass­
through)

­
34.18
no
no
no
­
28.56
no
no
no
­
12.12
no
no
no
­
10.93
no
no
no
5.74
yes
NO
NO
10.44
yes
NO
yes
13.93
yes
NO
yes
14.29
yes
NO
yes
17.4
yes
NO
yes
21.38
yes
yes
yes
24.26
yes
yes
yes
24.93
yes
yes
yes
61.11
yes
yes
yes
61.93
yes
yes
yes
64.17
yes
yes
yes
64.84
yes
yes
yes
74.80
yes
yes
yes
Number
of
plants
that
will
exit
the
waste
burning
market,
incremental
to
the
baseline:
5
1
NOTES:
1.
Compliance
cost
estimates
are
based
on
EPA
engineering
costs
that
incorporate
a
design
safety
factor
of
70
percent
(
i.
e.,
pollution
control
measures
are
assigned
such
that
kilns
can
meet
emission
levels
that
are
70%
of
the
MACT
standards,
on
average).
2.
Compliance
costs
do
not
include
PM
CEM
costs.
3.
CKRC
HWDF
waste­
burning
profits
from
the
1997
Environomics
memo,
Table
7,
page
19
(
converted
from
metric
tons).
14
SENSITIVITY
ANALYSIS
OF
EPA'S
ECONOMIC
IMPACT
MODELING
USING
CKRC
DATA
This
last
section
presents
a
sensitivity
analysis
of
EPA's
economic
impact
modeling
using
the
mean
values
of
CKRC's
waste­
burning
revenues
and
fuel
savings
data
reported
for
the
17
wasteburning
cement
plants.
Because
CKRC's
hazardous
waste­
burning
costs
line
up
well
with
the
cost
assumed
in
EPA's
analysis,
our
sensitivity
analysis
focuses
on
revenue
and
energy
savings
data.
As
explained
earlier
in
this
report,
we
are
unable
to
use
CKRC's
plant­
specific
information
directly
in
the
economic
impacts
model
because,
for
reasons
of
confidentiality,
CKRC
presented
the
data
in
ways
preventing
us
from
linking
specific
costs
and
revenues
to
individual
plants.
For
this
reason,
we
rely
on
mean
values
of
CKRC's
data.
Our
sensitivity
analysis
replaces
the
current
EPA
baseline
data
with
CKRC's
data,
as
shown
in
Exhibit
5.

Exhibit
5
CKRC
MEAN
DATA
TO
USE
IN
EPA
ECONOMIC
IMPACTS
MODEL
Data
Parameter
Current
EPA
Data
Assumption
($/
ton)
CKRC
Data
to
Use
(
average
$/
ton)

Revenues
received
by
kilns
for:

"
Clean"
liquids
$
20
$
58.40
Liquids
with
suspended
solids
$
70
$
58.40
Highly
contaminated
liquids
$
301
$
58.40
Sludges
$
320
$
309.20
Solids
$
683
$
309.20
Energy
savings
$
33.62
$
26.39
Note:
CKRC
data
converted
from
metric
tons.

It
is
important
to
recognize
that
using
these
mean
values
from
the
CKRC
data
tends
to
mask
important
differences
across
the
waste
streams
handled
by
kilns,
and
the
associated
waste­
burning
revenues.
As
stated
in
the
Environomics
1997
memo,
the
data
parameters
driving
the
financial
performance
of
waste­
burning
at
cement
kilns
vary
widely
across
plants:
16
This
sensitivity
analysis
finds
that,
in
the
baseline,
a
total
of
five
facilities
are
expected
to
stop
burning
hazardous
waste
over
the
long
term
capital
replacement
cycle.
However,
two
facilities
are
expected
to
stop
burning
in
the
short­
term
baseline.
Thus,
we
find
that,
based
on
this
analysis,
three
incremental
facilities
(
with
respect
to
the
short
term
baseline)
are
expected
to
exit
over
the
longer
term.
This
will
occur
even
in
the
absence
of
the
MACT
standards.

15
Our
data
show
that
revenues,
costs
and
net
profits
per
ton
of
HWDF
used
each
vary
across
plants
by
more
than
an
order
of
magnitude,
in
a
manner
unrelated
to
kiln
size
and
type.
...
[
T]
he
economics
of
using
HWDF
at
a
cement
plant
depends
on
plant­
specific
combinations
of
market,
physical,
regulatory
and
institutional
circumstances.
The
several
key
elements
that
determine
HWDF
revenues,
savings
and
costs
can
be
identified
but
they
are
not
uniformly
important
across
plants.
[
T]
he
relative
magnitude
and
importance
of
each
of
the
different
elements
vary
in
different
combinations
across
plants.
(
Environomics
1997,
3)

Thus,
the
results
from
the
model
runs
which
use
CKRC's
mean
values
for
revenues
and
energy
savings
will
not
result
in
accurate
estimates
of
economic
impacts.
The
model
results
described
below
should
be
strictly
viewed
as
sensitivity
analyses
of
model
input
parameters.

Using
the
average
CKRC
values
for
waste­
burning
revenues
and
fuel
savings
in
the
EPA
economic
impact
model
results
in
a
greater
number
of
cement
plants
that
are
expected
to
stop
burning
hazardous
waste.
As
with
the
previous
analysis,
economic
impacts
from
the
sensitivity
analysis
are
remarkably
comparable
across
the
different
SVM
emission
standards.
As
shown
in
Exhibit
6,
for
both
MACT
scenarios,
the
sensitivity
analysis
shows
that
between
three
to
five
cement
plants
will
exit
the
waste­
burning
market
(
incremental
to
the
baseline)
in
the
face
of
increased
costs
associated
with
the
MACT
standards.
However,
as
indicated
in
Exhibit
7,
three
of
these
facilities
are
expected
to
stop
burning
over
the
long
term
capital
replacement
cycle
even
if
the
MACT
standards
are
not
implemented.
16
Thus,
the
sensitivity
analysis
shows
that
over
the
longer
term,
one
to
two
facilities
move
from
positive
to
negative
profits
as
a
result
of
the
MACT
rule.
Again,
this
is
consistent
with
the
conclusions
of
the
EPA
model.
16
Exhibit
6
SENSITIVITY
ANALYSIS
RESULTS
USING
CKRC
DATA:
FACILITY
MARKET
EXITS
IN
THE
SHORT
TERM
Price
"
Pass­
Through"
(
see
Note
1)

0%
25%
50%
75%

Baseline
2
2
2
2
MACT
Scenario
A
(
SVM=
650ug/
dscm)
4­
5
4
4
3­
4
MACT
Scenario
B
(
SVM=
240ug/
dscm)
4­
5
4
4
3­
4
NOTES:
1.
The
price
pass­
through
is
the
percentage
of
median
compliance
costs
that
cement
kilns
can
pass
through
to
their
"
customers"
(
which
in
the
case
of
cement
kilns
are
fuels
managers,
fuel
blenders,
and
generators).

2.
Ranges
reflect
differences
across
engineering
design
levels
that
incorporate
different
emissions
"
safety
factors."

3.
Results
from
the
model
do
not
include
PM
CEM
costs.
17
Exhibit
7
SENSITIVITY
ANALYSIS
RESULTS
USING
CKRC
DATA:
FACILITY
MARKET
EXITS
IN
THE
LONG
TERM
Price
"
Pass­
Through"
(
see
Note
1)

0%
25%
50%
75%

Baseline
5
5
5
5
MACT
Scenario
A
(
SVM=
650ug/
dscm)
1­
2
1­
2
1
1
MACT
Scenario
B
(
SVM=
240ug/
dscm)
1­
2
1­
2
1
1
NOTES:
1.
The
price
pass­
through
is
the
percentage
of
median
compliance
costs
that
cement
kilns
can
pass
through
to
their
"
customers"
(
which
in
the
case
of
cement
kilns
are
fuels
managers,
fuel
blenders,
and
generators).

2.
Ranges
reflect
differences
across
engineering
design
levels
that
incorporate
different
emissions
"
safety
factors."

3.
Results
from
the
model
do
not
include
PM
CEM
costs.
18
REFERENCES
Portland
Cement
Association
1996.
"
U.
S.
Cement
Industry
Fact
Sheet:
Fourteenth
Edition,"
Table
24:
Fossil
Fuel
Mix.

Susel,
Irv
and
Stu
Sessions,
Environomics.
Memorandum:
"
Response
to
IEc
Follow­
up
Questions
on
CKRC
Comments
on
EPA's
Proposed
Hazardous
Waste
Combustion
MACT
Standards
(
March
13,
1997),"
September
22,
1997.
