MP&
M
EEBA:
Appendices
Appendix
A:
Detailed
Economic
Impact
Analysis
Information
Appendix
A:
Detailed
Economic
Impact
Analysis
Information
INTRODUCTION
This
appendix
provides
information
to
support
the
economic
analyses
of
MP&
M
industries
evaluated
for
the
final
rule
and
presented
in
Chapter
3
through
Chapter
11
of
the
EEBA.

The
first
section
below
provides
the
SIC
and
NAICS
codes
that
define
the
MP&
M
industrial
sectors.
The
second
section
presents
information
on
the
annual
turnover
of
establishments
( 
births 
and
 
deaths )
in
the
industrial
sectors.
The
third
section
provides
a
description
of
the
MP
&
M
surveys
that
supported
the
economic
impact
and
benefits
analyses
presented
in
the
EEBA
(
see
Section
3
of
the
TDD
).

A.
1
MP&
M
SIC
AND
NAICS
CODES
Standard
Industrial
Classification
(
SIC)
codes
and
North
American
Industrial
Classification
System
(
NAICS)
codes
are
hierarchical
systems
that
allow
for
detailed
classification
of
industries
using
numerical
codes.
This
section
lists
and
describes
the
SIC
codes
that
make
up
the
MP&
M
industry
sectors.
It
also
describes
the
process
by
which
data
organized
by
NAICS
code
was
converted
to
SIC
code
format.
CHAPTER
CONTENTS
A.
1
................
A­
1
A.
1.1
SIC
Codes
by
Sector
.................
...
A­
1
A.
1.2
.....
A­
7
A.
2
l
Establishment
 
Births 
and
 
Deaths 
in
MP&
M
Industries
.................
.....
A­
26
A.
3
escription
of
MP&
M
Surveys
...............
A­
28
A.
3.1
reener
Surveys
.................
....
A­
28
A.
3.2
io
Screener
Surveys
................
A­
28
A.
3.3
ailed
MP&
M
Industry
Surveys
.......
A­
28
A.
3.4
.................
A­
29
A.
3.5
.................
.
A­
29
A.
3.6
................
A­
29
A.
3.7
.................
......
A­
29
References
.................
.................
.
A­
31
MP&
M
SIC
and
NAICS
Codes
Bridge
Between
NAICS
and
SIC
codes
Annua
D
Sc
Oh
Det
Iron
and
Steel
Survey
Municipality
Survey
Federal
Facility
Survey
POTW
Survey
A.
1.1
SIC
Codes
by
Sector
Table
A.
1
lists
and
describes
the
4­
digit
SIC
codes
that
make
up
the
MP&
M
industry
sectors.
These
codes
were
used
until
recently
to
define
industries
for
reporting
of
Federal
Census
data,
and
are
the
framework
for
the
part
of
the
industry
profile
(
Chapter
3)
based
on
publicly
available
material.

Table
A.
1:
MP&
M
Sectors
and
SIC
Codes
Evaluated
for
the
Final
Rulea
SIC
Code
Standard
Industrial
Classification
Groups
Aerospace
3761
Guided
Missiles
and
Space
Vehicles
3764
Guided
Missile
and
Space
Vehicle
Propulsion
3769
Other
Space
Vehicle
and
Missile
Parts
Aircraft
3721
Aircraft
3724
Aircraft
Engines
and
Engine
Parts
3728
Aircraft
Parts
and
Auxiliary
Equipment
4581
Airports,
Flying
Fields,
Airport
Terminal
Services
Bus
And
Truck
3713
Truck
and
Bus
Bodies
3715
Truck
Trailers
A­
1
MP&
M
EEBA:
Appendices
Appendix
A:
Detailed
Economic
Impact
Analysis
Information
Table
A.
1:
MP&
M
Sectors
and
SIC
Codes
Evaluated
for
the
Final
Rulea
SIC
Code
Standard
Industrial
Classification
Groups
4111
Local
And
Suburban
Transit
4119
Local
Passenger
Transit,
N.
E.
C.

4131
Intercity
And
Rural
Bus
Transportation
4141
Local
Bus
Charter
Service
4142
Bus
Charter
Service,
Except
Local
4173
Bus
Terminal
And
Service
Facilities
4212
Local
Trucking
without
Storage
4213
Trucking,
Except
Local
4214
Local
Trucking
with
Storage
4215
Courier
Services,
Except
by
Air
4231
Trucking
Terminal
Facilities
Electronic
Equipment
3661
Telephone
and
Telegraph
Apparatus
3663
Radio
and
Television
Broadcast
and
Communications
Equipment
3669
Communications
Equipment,
N.
E.
C.

3671
Electron
Tubes
3675
Electronic
Capacitors
3677
Electronic
Coils
and
Transformers
3678
Connectors
for
Electronic
Applications
3679
Electronic
Components,
N.
E.
C.

3699
Electrical
Machinery,
Equipment,
And
Supplies,
N.
E.
C.

Hardware
2796
Platemaking
and
Related
Services
3398
Metal
Heat
Treating
3412
Metal
Shipping
Barrels,
Drums,
Kegs,
Pails
3421
Cutlery
3423
Hand
And
Edge
Tools,
Except
Machine
Tools
and
Handsaws
3425
Hand
Saws
and
Saw
Blades
3429
Hardware,
N.
E.
C.

3433
Heating
Equipment,
Except
Electric
and
Warm
Air
Furnace
3441
Fabricated
Structural
Metal
3443
Fabricated
Plate
Work
(
Boiler
Shops)

3444
Sheet
Metal
Work
3446
Architectural
and
Ornamental
Metal
Work
3448
Prefabricated
Metal
Buildings
And
Components
3449
Miscellaneous
Metal
Work
3451
Screw
Machine
Products
3452
Bolts,
Nuts,
Screws,
Rivets,
and
Washers
3462
Iron
and
Steel
Forgings
3466
Crowns
and
Closures
3469
Metal
Stamping,
N.
E.
C.

3492
Fluid
Power
Valves
and
Hose
Fittings
3493
Steel
Springs
3494
Valves
And
Pipe
Fittings,
Except
Brass
A­
2
MP&
M
EEBA:
Appendices
Appendix
A:
Detailed
Economic
Impact
Analysis
Information
Table
A.
1:
MP&
M
Sectors
and
SIC
Codes
Evaluated
for
the
Final
Rulea
SIC
Code
Standard
Industrial
Classification
Groups
3495
Wire
Springs
3496
Miscellaneous
Fabricated
Wire
Products
3498
Fabricated
Pipe
and
Fabricated
Pipe
Fitting
3499
Fabricated
Metal
Products,
N.
E.
C.

3541
Machine
Tools,
Metal
Cutting
Types
3542
Machine
Tools,
Metal
Forming
Types
3544
Special
Dies
and
Tools,
Die
Sets,
Jigs
and
Fixtures,
and
Industrial
Molds
3545
Machine
Tool
Access
and
Measuring
Devices
3546
Power
Driven
Hand
Tools
3965
Fasteners,
Buttons,
Needles,
Pins
Household
Equipment
2514
Metal
Household
Furniture
2522
Office
Furniture,
Except
Wood
2531
Public
Building
and
Related
Furniture
2542
Partitions
and
Fixtures,
Except
Wood
2591
Drapery
Hardware
and
Window
Blinds/
shades
2599
Furniture
and
Fixtures,
N.
E.
C.

3431
Metal
Sanitary
Ware
3432
Plumbing
Fittings
and
Brass
Goods
3442
Metal
Doors,
Sash,
and
Trim
3631
Household
Cooking
Equipment
3632
Household
Refrigerators
and
Home
and
Farm
and
Freezers
3633
Household
Laundry
Equipment
3634
Electric
Housewares
and
Fans
3635
Household
Vacuum
Cleaners
3639
Household
Appliances,
N.
E.
C.

3641
Electric
Lamps
3643
Current­
carrying
Wiring
Devices
3644
Noncurrent­
carrying
Wiring
Devices
3645
Residential
Electrical
Lighting
Fixtures
3646
Commercial,
Industrial,
and
Institutional
3648
Lighting
Equipment,
N.
E.
C.

3651
Radio/
television
Sets
Except
Communication
Types
7623
Refrigeration
and
Air­
conditioning
Service
and
Repair
Shops
Instruments
3812
Search,
Detection,
Navigation,
Guidance,
Aeronautical,
Nautical
Systems
and
Instruments
3821
Laboratory
Apparatus
and
Furniture
3822
Automatic
Environmental
Controls
3823
Process
Control
Instruments
3824
Fluid
Meters
and
Counting
Devices
3825
Instruments
to
Measure
Electricity
3826
Laboratory
Analytical
Instruments
3827
Optical
Instruments
and
Lenses
3829
Measuring
and
Controlling
Devices,
N.
E.
C.

A­
3
MP&
M
EEBA:
Appendices
Appendix
A:
Detailed
Economic
Impact
Analysis
Information
Table
A.
1:
MP&
M
Sectors
and
SIC
Codes
Evaluated
for
the
Final
Rulea
SIC
Code
Standard
Industrial
Classification
Groups
3841
Surgical
and
Medical
Instruments
and
Apparatus
3842
Orthopedic,
Prosthetic
and
Surgical
Suppl.

3843
Dental
Equipment
and
Supplies
3844
X­
ray
Apparatus
and
Tubes
3845
Electromedical
Equipment
3851
Ophthalmic
Goods
7629
Electric
Repair
Shop
Iron
and
Steel
3315
Steel
Wiredrawing
and
Steel
Nails
and
Spikes
3316
Cold­
Rolled
Steel
Sheet,
Strip,
and
Bars
3317
Steel
Pipe
and
Tubes
Job
Shop
3471
Plating
and
Polishing
3479
Metal
Coating
and
Allied
Services
Mobile
Industrial
Equipment
3523
Farm
Machinery
and
Equipment
3524
Garden
Tractors
and
Lawn
and
Garden
Equipment
3531
Construction
Machinery
and
Equipment
3532
Mining
Machinery
and
Equipment,
Except
Oil
Field
3536
Hoists,
Industrial
Cranes
and
Monorails
3537
Industrial
Trucks,
Tractors,
Trailers
3795
Tanks
and
Tank
Components
Motor
Vehicle
3465
Automotive
Stampings
3592
Carburetors,
Piston
Rings,
Valves
3647
Vehicular
Lighting
Equipment
3694
Electrical
Equipment
for
Motor
Vehicles
3711
Motor
Vehicle
and
Automobile
Bodies
3714
Motor
Vehicle
Parts
and
Accessories
3716
Mobile
Homes
3751
Motorcycles
3792
Travel
Trailers
and
Campers
3799
Miscellaneous
Transportation
Equipment
4121
Taxicabs
5013
Motor
Vehicle
Supplies
and
New
Parts
5511
Motor
Vehicle
Dealers
(
New
and
Used)

5521
Motor
Vehicle
Dealers
(
Used
Only)

5561
Recreational
Vehicle
Dealers
5571
Motorcycle
Dealers
5599
Automotive
Dealers,
N.
E.
C.

7514
Passenger
Car
Rental
7515
Passenger
Car
Lease
7519
Utility
Trailer
and
Recreational
Vehicle
Rental
7532
Top,
Body,
and
Upholstery
Repair
and
Paint
Shops
A­
4
MP&
M
EEBA:
Appendices
Appendix
A:
Detailed
Economic
Impact
Analysis
Information
Table
A.
1:
MP&
M
Sectors
and
SIC
Codes
Evaluated
for
the
Final
Rulea
SIC
Code
Standard
Industrial
Classification
Groups
7533
Auto
Exhaust
Systems
7537
Auto
Transmission
Repair
7538
General
Automotive
Repair
7539
Auto
Repair
Shop,
N.
E.
C.

7549
Auto
Services,
Except
Repair
and
Carwashes
Office
Machine
3571
Electronic
Computers
3572
Typewriters
3575
Computer
Terminals
3577
Computer
Peripheral
Equipment,
N.
E.
C.

3578
Calculating,
Accounting
Machines
Except
Computers
3579
Office
Machines,
N.
E.
C.

7378
Computer
Maintenance
and
Repairs
7379
Computer
Related
Services,
N.
E.
C.

Ordnance
3482
Small
Arms
Ammunition
3483
Ammunition,
Except
for
Small
Arms
3484
Small
Arms
3489
Ordnance
and
Accessories,
N.
E.
C.

Miscellaneous
Metal
Products
3497
Metal
Foil
and
Leaf
3861
Photographic
Equipment
and
Supplies
3931
Musical
Instruments
3944
Games,
Toys,
Children's
Vehicles
3949
Sporting
and
Athletic
Goods,
N.
E.
C.

3951
Pens
and
Mechanical
Pencils
3953
Marking
Devices
3993
Signs
and
Advertising
Displays
3995
Burial
Caskets
3999
Manufacturing
Industries,
N.
E.
C.

7692
Welding
Repair
7699
Repair
Shop,
Related
Service
Precious
Metals
and
Jewelry
3873
Watches,
Clocks,
and
Watchcases
3911
Jewelry,
Precious
Metal
3914
Silverware,
Plated
Ware
and
Stainless
3915
Jewelers'
Materials
and
Lapidary
Work
3961
Costume
Jewelry
7631
Watch,
Clock,
Jewelry
Repair
Printed
Circuit
Boards
3672
Printed
Circuit
Boards
Railroad
3743
Railcars,
Railway
Systems
4011
Railroad
Transportation
A­
5
MP&
M
EEBA:
Appendices
Appendix
A:
Detailed
Economic
Impact
Analysis
Information
Table
A.
1:
MP&
M
Sectors
and
SIC
Codes
Evaluated
for
the
Final
Rulea
SIC
Code
Standard
Industrial
Classification
Groups
4013
Railroad
Transportation
Ships
and
Boats
3731
Ship
Building
and
Repairing
3732
Boat
Building
and
Repairing
4412
Deep
Sea
Foreign
Transportation
4424
Deep
Sea
Domestic
Transportation
4432
Freight
Transportation
Great
Lakes
4449
Water
Transportation
of
Freight,
N.
E.
C.

4481
Deep
Sea
Passenger
Transportation
4482
Ferries
4489
Water
Passenger
Transportation,
N.
E.
C.

4491
Marine
Cargo
Handling
4492
Towing
and
Tugboat
Service
4493
Marinas
4499
Water
Transportation
Services,
N.
E.
C.

Stationary
Industrial
Equipment
3511
Steam,
Gas,
Hydraulic
Turbines,
Generating
Units
3519
Internal
Combustion
Engines,
N.
E.
C.

3533
Oil
Field
Machinery
and
Equipment
3534
Elevators
and
Moving
Stairways
3535
Conveyors
and
Conveying
Equipment
3543
Industrial
Patterns
3547
Rolling
Mill
Machinery
and
Equipment
3548
Electric
and
Gas
Welding
and
Soldering
3549
Metal
Working
Machinery,
N.
E.
C.

3552
Textile
Machinery
3553
Woodworking
Machinery
3554
Paper
Industries
Machinery
3555
Printing
Trades
Machinery
and
Equipment
3556
Food
Products
Machinery
3559
Special
Industry
Machinery,
N.
E.
C.

3561
Pumps
and
Pumping
Equipment
3562
Ball
and
Roller
Bearings
3563
Air
and
Gas
Compressors
3564
Blowers
and
Exhaust
and
Ventilation
Fans
3565
Industrial
Patterns
3566
Speed
Changers,
High
Speed
Drivers
and
Gears
3567
Industrial
Process
Furnaces
and
Ovens
3568
Mechanical
Power
Transmission
Equipment,
N.
E.
C.

3569
General
Industrial
Machinery,
N.
E.
C.

3581
Automatic
Merchandising
Machines
3582
Commercial
Laundry
Equipment
3585
Refrigeration
and
Air
and
Heating
Equipment
3586
Measuring
and
Dispensing
Pumps
A­
6
MP&
M
EEBA:
Appendices
Appendix
A:
Detailed
Economic
Impact
Analysis
Information
Table
A.
1:
MP&
M
Sectors
and
SIC
Codes
Evaluated
for
the
Final
Rulea
SIC
Code
Standard
Industrial
Classification
Groups
3589
Service
Industry
Machines,
N.
E.
C.

3593
Fluid
Power
Cylinders
and
Actuators
3594
Fluid
Power
Pumps
and
Motors
3596
Scales
and
Balances,
Except
Laboratory
3599
Machinery,
Except
Electrical,
N.
E.
C.

3612
Transformers
3613
Switchgear
and
Switchboard
Apparatus
3621
Motors
and
Generators
3629
Electric
Industrial
Apparatus,
N.
E.
C.

7353
Heavy
Construction
Equip
Rental,
Leasing
7359
Equipment
Rental,
Leasing,
N.
E.
C.

a
EPA
evaluated
options
for
these
industrial
sectors
but
did
not
regulate
them
all
under
the
final
rule.

N.
E.
C.
=
Not
Elsewhere
Classified
Source:
Executive
Office
of
the
President,
Office
of
Management
and
Budget,
Standard
Industrial
Classification
Manual
1987.

A.
1.2
Bridge
Between
NAICS
and
SIC
codes
In
1997,
the
Census
Bureau
switched
from
using
SIC
codes
to
using
NAICS
codes.
NAICS
codes
allow
for
greater
comparability
with
the
International
Standard
Industrial
Classification
System
(
ISIC),
which
is
developed
and
maintained
by
the
United
Nations.
NAICS
codes
also
better
reflect
the
structure
of
today s
economy,
including
the
growth
of
the
service
sectors
and
new
technologies,
than
do
the
decades­
old
SIC
codes.
Because
EPA
chose
to
create
regulatory
subgroups
for
the
MP&
M
industries
based
on
aggregated
four­
digit
SIC
codes,
it
was
necessary
for
EPA
to
convert
some
data
based
on
NAICS
codes
into
SIC
code
format.

The
SIC­
NAICS
conversion
is
not
always
straightforward
because
NAICS
and
SIC
codes
often
don t
map
on
a
one­
to­
one
basis.
Specific
industries
that
were
grouped
together
in
one
SIC
code
sometimes
map
to
several
NAICS
codes,
and
sometimes
several
SIC
codes
were
aggregated
together
in
one
NAICS
code.

To
address
this
conversion
problem,
EPA
created
a
 
bridge 
that
converts
the
NAICS
classification
structure
to
the
SIC
structure
using
share
values
computed
from
Economic
Census
data.
This
bridge
is
based
on
data
from
the
1997
Census,

which
reported
the
share
of
number
of
establishments
and
value
of
output
that
each
SIC
code
that
contributed
to
each
NAICS
code,
and
vice
versa.

The
first
step
in
creating
the
bridge
was
to
obtain
a
table
that
listed
the
value
of
shipments
(
VOS)
that
each
NAICS
code
contributed
to
each
SIC
code.
Since
the
total
VOS
for
each
NAICS
code
was
known,
EPA
computed
share
values
for
each
NAICS,
which
were
equal
to
the
percent
of
total
VOS
in
that
NAICS
code
that
was
classified
in
a
certain
SIC
code.
The
equation
is:

Share
of
NAICSx
going
to
SICy
=
(
VOS
that
NAICSx
contributed
to
SICy)
/
(
total
VOS
for
NAICSx)
(
A­
1)

Using
these
share
values,
EPA
converted
data
classified
by
NAICS
to
SIC
format,
simply
by
multiplying
VOS
for
each
NAICS
by
its
share
value,
for
each
SIC,
and
then
summing
up
the
totals
for
each
SIC.
For
example,
if
NAICS
codes
333121,

332456,
and
332457
all
contributed
a
portion
of
their
output
to
SIC
3322,
then:

VOS
for
SIC3322
=
(
share
of
NAICS333121
going
to
SIC3322
)
*
(
VOS
for
NAICS333121
)

+
(
share
of
NAICS332456
going
to
SIC3322)
*
(
VOS
for
NAICS332456)
(
A­
2)

+
(
share
of
NAICS332457
going
to
SIC3322
)
*
(
VOS
for
NAICS332457
)

A­
7
MP&
M
EEBA:
Appendices
Appendix
A:
Detailed
Economic
Impact
Analysis
Information
Occasionally
it
was
not
possible
to
compute
share
values
because
the
Census
Bureau
withheld
some
1997
VOS
data
because
1
of
disclosure
issues
.
In
those
cases,
EPA
estimated
1997
VOS
based
on
1992
Census
data
and
then
used
those
estimates
to
compute
share
values.
First,
EPA
calculated
the
average
VOS
per
establishment
in
1992
for
each
relevant
SIC
code:

VOS
per
establishment
for
SICy
=
(
A­
3)
(
VOS
for
SICy
in
1992)
/
(
number
of
establishments
for
SICy
in
1992)

EPA
then
multiplied
this
average
VOS
per
establishment
for
a
certain
SIC
by
the
number
of
establishments
that
each
NAICS
contributed
to
that
SIC
in
1997:

Estimated
VOS
that
NAICSx
contributed
to
SICy
in
1997
=
(
A­
4)
(
VOS
per
establishment
for
SICy)
*
(
number
of
establishments
NAICSx
contributed
to
SICy
in
1997)

EPA
used
this
estimated
VOS
to
compute
an
estimated
share
value.

To
gain
a
rough
measure
of
how
accurately
the
NAICS
codes
could
be
broken
into
sectors,
EPA
calculated,
by
sector:
(
1)
the
percentage
of
NAICS
codes
that
matched
 
one­
to­
one 
with
an
SIC
code,
(
2)
the
percentage
that
did
not
match
one­
to­
one
but
were
contained
in
a
single
sector,
and
(
3)
the
percentage
that
didn t
match
one
to
one
and
were
contained
in
multiple
sectors
(
Figure
A.
1,
Table
A.
2).

Figure
A.
1:
Percentage
of
VOS
1997
to
1999
Attributable
to
One­
to­
One
NAICS­
SIC
Match,
Not
One­
to­
One
but
in
the
Same
Sector,
and
Not
One­
to­
One
but
in
Different
Sectors
Sectors:
1
Hardware;
2
Aircraft;
3
Electronic
Equipment;
4
Stationary
Industrial
Equipment;
5
Ordnance;
6
Aerospace;
7
Mobile
Industrial
Equipment;
8
Instruments;
9
Precious
Metals
and
Jewelry;
10
Ships
and
Boats;
11
Household
Equipment;
12
Railroad;
13
Motor
Vehicle;
14
Bus
and
Truck;
15
Office
Machine;
16
Printed
Circuit
Boards;
17
Job
Shop;
18
Miscellaneous
Metal
Products;
19
Iron
and
Steel
Source:
Department
of
Commerce,
Bureau
of
the
Census,
Manufacturing
Industry
Series;
U.
S.
EPA
analysis.

1
The
Bureau
of
the
Census
does
not
release
any
data
that
could
reveal
data
about
a
specific
firm.
In
cases
when
a
NAICS
or
SIC
code
is
so
specific
that
it
includes
only
a
few
firms,
information
about
VOS
is
not
released.
However,
the
number
of
establishments
in
a
specific
industry
is
not
considered
private
information.

A­
8
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
MP&
M
EEBA:
Appendices
Appendix
A:
Detailed
Economic
Impact
Analysis
Information
Table
A.
2:
Percentage
of
Input
One­
to­
One,
Not
One­
to­
One
but
in
the
Same
Sector,
and
Not
One­
to­
One
and
in
Different
Sectors
Sector
VOS
One­
to­
One
Employment
One­
to­
One
VOS
Same
Sector
Employment
Same
Sector
VOS
Different
Sectors
Employment
Different
Sectors
YEAR:
1997
62.5%
64.3%
18.2%
16.5%
19.3%
19.2%

100.0%
100.0%
0.0%
0.0%
0.0%
0.0%

46.7%
47.2%
47.2%
43.2%
6.2%
9.7%

63.3%
68.1%
3.9%
4.4%
32.8%
27.6%

100.0%
100.0%
0.0%
0.0%
0.0%
0.0%

100.0%
100.0%
0.0%
0.0%
0.0%
0.0%

91.8%
88.1%
5.5%
7.8%
2.7%
4.1%

30.4%
30.2%
14.4%
14.4%
55.2%
55.4%

10.2%
8.3%
0.0%
0.0%
89.8%
91.7%

100.0%
100.0%
0.0%
0.0%
0.0%
0.0%

67.5%
60.6%
6.3%
4.5%
26.3%
34.9%

0.0%
0.0%
0.0%
0.0%
100.0%
100.0%

85.3%
69.5%
1.1%
3.1%
13.6%
27.4%

39.1%
42.8%
0.0%
0.0%
60.9%
57.2%

73.1%
59.9%
26.4%
38.6%
0.5%
1.5%

100.0%
100.0%
0.0%
0.0%
0.0%
0.0%

99.9%
99.9%
0.0%
0.0%
0.1%
0.1%

83.1%
76.5%
12.2%
17.8%
4.6%
5.7%

98.1%
95.3%
0.0%
0.0%
1.9%
4.7%

YEAR:
1998
62.8%
64.9%
17.9%
16.3%
19.3%
18.8%

100.0%
100.0%
0.0%
0.0%
0.0%
0.0%

47.6%
47.3%
46.0%
42.7%
6.4%
10.0%

62.0%
68.3%
3.8%
4.4%
34.2%
27.3%

100.0%
100.0%
0.0%
0.0%
0.0%
0.0%

100.0%
100.0%
0.0%
0.0%
0.0%
0.0%

91.8%
88.0%
5.5%
8.0%
2.7%
4.1%

29.4%
29.3%
15.1%
14.7%
55.5%
55.9%

8.4%
8.7%
0.0%
0.0%
91.6%
91.3%

100.0%
100.0%
0.0%
0.0%
0.0%
0.0%

66.2%
60.0%
6.9%
4.8%
26.9%
35.2%

0.0%
0.0%
0.0%
0.0%
100.0%
100.0%

84.2%
68.0%
1.3%
3.4%
14.6%
28.6%

40.7%
43.4%
0.0%
0.0%
59.3%
56.6%

73.5%
58.9%
26.0%
39.7%
0.5%
1.4%

100.0%
100.0%
0.0%
0.0%
0.0%
0.0%

99.9%
99.9%
0.0%
0.0%
0.1%
0.1%

82.1%
76.1%
12.9%
18.2%
4.9%
5.8%

97.9%
95.3%
0.0%
0.0%
2.1%
4.7%

A­
9
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
MP&
M
EEBA:
Appendices
Appendix
A:
Detailed
Economic
Impact
Analysis
Information
Table
A.
2:
Percentage
of
Input
One­
to­
One,
Not
One­
to­
One
but
in
the
Same
Sector,
and
Not
One­
to­
One
and
in
Different
Sectors
Sector
VOS
One­
to­
One
Employment
One­
to­
One
VOS
Same
Sector
Employment
Same
Sector
VOS
Different
Sectors
Employment
Different
Sectors
YEAR:
1999
62.3%
64.3%
18.3%
16.4%
19.4%
19.3%

100.0%
100.0%
0.0%
0.0%
0.0%
0.0%

48.7%
47.9%
45.4%
42.3%
5.9%
9.8%

61.6%
67.8%
3.6%
4.3%
34.7%
27.9%

100.0%
100.0%
0.0%
0.0%
0.0%
0.0%

100.0%
100.0%
0.0%
0.0%
0.0%
0.0%

89.6%
87.0%
7.4%
8.8%
3.0%
4.2%

29.9%
29.8%
15.2%
15.2%
54.9%
55.1%

7.1%
7.5%
0.0%
0.0%
92.9%
92.5%

100.0%
100.0%
0.0%
0.0%
0.0%
0.0%

65.5%
57.8%
7.7%
5.4%
26.9%
36.8%

0.0%
0.0%
0.0%
0.0%
100.0%
100.0%

84.6%
68.3%
1.3%
3.9%
14.1%
27.9%

40.5%
45.8%
0.0%
0.0%
59.5%
54.2%

75.6%
56.6%
23.8%
41.9%
0.6%
1.6%

100.0%
100.0%
0.0%
0.0%
0.0%
0.0%

99.9%
99.9%
0.0%
0.0%
0.1%
0.1%

82.0%
76.8%
13.0%
17.1%
5.0%
6.1%

97.7%
95.0%
0.0%
0.0%
2.3%
5.0%

Sectors:
1
Hardware;
2
Aircraft;
3
Electronic
Equipment;
4
Stationary
Industrial
Equipment;
5
Ordnance;
6
Aerospace;
7
Mobile
Industrial
Equipment;
8
Instruments;
9
Precious
Metals
and
Jewelry;
10
Ships
and
Boats;
11
Household
Equipment;
12
Railroad;
13
Motor
Vehicle;
14
Bus
and
Truck;
15
Office
Machine;
16
Printed
Circuit
Boards;
17
Job
Shop;
18
Miscellaneous
Metal
Products;
19
Iron
and
Steel
Source:
Department
of
Commerce,
Bureau
of
the
Census,
Manufacturing
Industry
Series;
U.
S.
EPA
analysis.

Table
A.
3
presents
the
data
that
was
used
to
calculate
the
relationship
between
NAICS
and
SIC
codes.
The
table
lists
the
MP&
M
sector
to
which
each
SIC
code
belongs,
gives
a
short
description
of
each
SIC,
and
lists
NAICS
codes
that
encompass
similar
industries.
The
table
also
lists
the
number
of
establishments,
the
value
of
shipments,
and
the
number
of
employees
that
are
contributed
to
each
SIC
by
each
NAICS,
as
well
as
the
share
values,
i.
e.
the
portion
of
its
total
value
of
shipments
that
a
given
NAICS
code
contributes
to
a
given
SIC
code.

A­
10
MP&
M
EEBA:
Appendices
Appendix
A:
Detailed
Economic
Impact
Analysis
Information
Table
A.
3:
Relationships
between
SIC
and
NAICS
Codes
Based
on
1997
Economic
Census
for
MP&
M
Industries
Evaluated
for
the
Final
Rulea
(
thousands,
1997$)

SIC
SIC
Industry
NAICS
Code
1997
NAICS
Industry
Number
of
Establishments
Sales,

Shipments
or
Receipts
Share
Value
Aerospace
3761
Guided
Missiles
and
Space
Vehicles
336414
Guided
Missile
and
Space
Vehicle
Manufacturing
22
14,791,466
100.0%

3764
Guided
Missile
and
Space
Vehicle
Propulsion
336415
Guided
Missile
and
Space
Vehicle
Propulsion
Unit
and
Propulsion
Unit
Parts
Manufacturing
28
3,239,033
100.0%

3769
Other
Space
Vehicle
and
Missile
Parts
336419
Other
Guided
Missile
and
Space
Vehicle
Parts
and
Auxiliary
Equipment
Manufacturing
49
898,758
100.0%

Aircraft
3721
Aircraft
336411
Aircraft
Manufacturing
204
56,273,651
100.0%

3724
Aircraft
Engines
and
Engine
Parts
336412
Aircraft
Engine
and
Engine
Parts
Manufacturing
369
22,617,284
100.0%

3728
Aircraft
Parts
and
Auxiliary
Equipment
336413
Other
Aircraft
Parts
and
Auxiliary
Equipment
Manufacturing
1,138
20,073,061
100.0%

4581
Airports,
Flying
Fields,

Airport
Terminal
Services
488111
Air
Traffic
Control
114
43,450
100.0%

488119
Other
Airport
Operations
1,699
3,243,149
99.8%

488190
Other
Support
Activities
for
Air
Transportation
2,400
5,859,631
100.0%

561720
Janitorial
Services
127
203,918
1.0%

Bus
&
Truck
3713
Truck
and
Bus
Bodies
336211
Motor
Vehicle
Body
Manufacturing
715
8,719,326
96.2%

3715
Truck
Trailers
336212
Truck
Trailer
Manufacturing
390
5,507,768
100.0%

4111
Local
And
Suburban
Transit
485111
Mixed
Mode
Transit
Systems
28
51,567
100.0%

485113
Bus
and
Other
Motor
Vehicle
Transit
Systems
542
1,152,525
100.0%

485999
All
Other
Transit
and
Ground
Passenger
Transportation
534
601,988
89.9%

4119
Local
Passenger
Transit,

N.
E.
C.
485320
Limousine
Service
3,234
1,873,924
100.0%

485410
School
and
Employee
Bus
Transportation
158
158,947
3.6%

485991
Special
Needs
Transportation
1,789
1,141,413
100.0%

485999
All
Other
Transit
and
Ground
Passenger
Transportation
232
67,395
10.1%

487110
Scenic
and
Sightseeing
Transportation,
Land
307
462,186
82.9%

621910
Ambulance
Services
3,275
4,443,174
88.4%

4131
Intercity
And
Rural
Bus
Transportation
485210
Interurban
and
Rural
Bus
Transportation
407
1,147,432
100.0%

4141
Local
Bus
Charter
Service
485510
Charter
Bus
Industry
482
459,953
26.0%

4142
Bus
Charter
Service,
Except
Local
485510
Charter
Bus
Industry
1,049
1,308,246
74.0%

4173
Bus
Terminal
And
Service
Facilities
488490
Other
Support
Activities
for
Road
Transportation
26
15,253
3.9%

A­
11
MP&
M
EEBA:
Appendices
Appendix
A:
Detailed
Economic
Impact
Analysis
Information
Table
A.
3:
Relationships
between
SIC
and
NAICS
Codes
Based
on
1997
Economic
Census
for
MP&
M
Industries
Evaluated
for
the
Final
Rulea
(
thousands,
1997$)

SIC
SIC
Industry
NAICS
Code
1997
NAICS
Industry
Number
of
Establishments
Sales,

Shipments
or
Receipts
Share
Value
4212
Local
Trucking
without
Storage
484110
General
Freight
Trucking,
Local
14,545
11,108,345
90.5%

484210
Used
Household
and
Office
Goods
Moving
3,259
1,198,983
9.5%

484220
Specialized
Freight
(
except
Used
Goods)

Trucking,
Local
34,935
18,932,851
96.0%

562111
Solid
Waste
Collection
7,083
18,211,495
100.0%

562112
Hazardous
Waste
Collection
414
1,095,553
100.0%

562119
Other
Waste
Collection
827
837,625
100.0%

4213
Trucking,
Except
Local
484121
General
Freight
Trucking,
Long­
Distance,

Truckload
23,111
51,142,148
100.0%

484122
General
Freight
Trucking,
Long­
Distance,

Less
Than
Truckload
6,210
25,010,091
100.0%

484210
Used
Household
and
Office
Goods
Moving
3,555
9,111,477
72.4%

484230
Specialized
Freight
(
except
Used
Goods)

Trucking,
Long­
Distance
14,439
20,500,392
100.0%

4214
Local
Trucking
with
Storage
484110
General
Freight
Trucking,
Local
915
1,164,931
9.5%

484210
Used
Household
and
Office
Goods
Moving
2,286
2,273,241
18.1%

484220
Specialized
Freight
(
except
Used
Goods)

Trucking,
Local
543
782,939
4.0%

4215
Courier
Services,
Except
by
Air
492110
Couriers
2,362
19,289,602
53.1%

492210
Local
Messengers
and
Local
Delivery
5,384
3,519,100
100.0%

4231
Trucking
Terminal
Facilities
488490
Other
Support
Activities
for
Road
Transportation
14
12,989
3.3%

Electronic
Equipment
3661
Telephone
and
Telegraph
Apparatus
334210
Telephone
Apparatus
Manufacturing
598
38,300,044
100.0%

334416
Electronic
Coil,
Transformer,
and
Other
Inductor
Manufacturing
7
8,904
0.6%

334418
Printed
Circuit
Assembly
(
Electronic
Assembly)
Manufacturing
20
1,364,671
5.2%

3663
Radio
and
Television
Broadcast
and
Comm
Eq
334220
Radio
and
Television
Broadcasting
and
Wireless
Communications
Equipment
Manufacturing
1,091
37,042,241
94.2%

3669
Communications
Eq,
N.
E.
C.
334290
Other
Communications
Equipment
Manufacturing
497
4,233,288
100.0%

3671
Electron
Tubes
334411
Electron
Tube
Manufacturing
159
3,858,499
100.0%

3675
Electronic
Capacitors
334414
Electronic
Capacitor
Manufacturing
129
2,482,163
100.0%

3677
Electronic
Coils
and
Transformers
334416
Electronic
Coil,
Transformer,
and
Other
Inductor
Manufacturing
426
1,512,232
97.9%

3678
Connectors
for
Electronic
Applications
334417
Electronic
Connector
Manufacturing
347
5,598,906
100.0%

A­
12
MP&
M
EEBA:
Appendices
Appendix
A:
Detailed
Economic
Impact
Analysis
Information
Table
A.
3:
Relationships
between
SIC
and
NAICS
Codes
Based
on
1997
Economic
Census
for
MP&
M
Industries
Evaluated
for
the
Final
Rulea
(
thousands,
1997$)

SIC
SIC
Industry
NAICS
Code
1997
NAICS
Industry
Number
of
Establishments
Sales,

Shipments
or
Receipts
Share
Value
3679
Electronic
Components
N.
E.
C.
334220
Radio
and
Television
Broadcasting
and
Wireless
Communications
Equipment
Manufacturing
126
2,265,873
5.8%

334418
Printed
Circuit
Assembly
(
Electronic
Assembly)
Manufacturing
695
24,704,154
94.8%

334419
Other
Electronic
Component
Manufacturing
1,851
10,547,090
100.0%

336322
Other
Motor
Vehicle
Electrical
and
Electronic
Equipment
Manufacturing
253
1,420,996
8.4%

3699
Electronic
Mach.,

Equipment,
&
Suppl.
N.
E.
C.
332212
Hand
and
Edge
Tool
Manufacturing
4
140,811
2.1%

333293
Printing
Machinery
and
Equipment
Manufacturing
5
0
0.9%
b
333314
Optical
Instrument
and
Lens
Manufacturing
5
7,320
0.2%

333319
Other
Commercial
and
Service
Industry
Machinery
Manufacturing
57
934,728
10.0%

333512
Machine
Tool
(
Metal
Cutting
Types)

Manufacturing
8
151,363
2.8%

333618
Other
Engine
Equipment
Manufacturing
2
0
0.7%
b
333992
Welding
and
Soldering
Equipment
Manufacturing
6
11,101
0.2%

334510
Electromedical
and
Electrotherapeutic
Apparatus
Manufacturing
11
52,855
0.5%

334511
Search,
Detection,
Navigation,
Guidance,

Aeronautical,
and
Nautical
System
and
Instrument
Manufacturing
7
77,832
0.2%

334516
Analytical
Laboratory
Instrument
Manufacturing
10
36,473
0.5%

334519
Other
Measuring
and
Controlling
Device
Manufacturing
5
6,174
0.1%

335129
Other
Lighting
Equipment
Manufacturing
4
859
0.0%

335999
All
Other
Miscellaneous
Electrical
Equipment
and
Component
Manufacturing
567
4,051,267
58.8%

Hardware
2796
Platemaking
and
Related
Services
323122
Prepress
Services
1,276
2,663,020
53.2%

3398
Metal
Heat
Treating
332811
Metal
Heat
Treating
808
3,485,459
100.0%

3412
Metal
Shipping
Barrels,

Drums,
Kegs,
Pails
332439
Other
Metal
Container
Manufacturing
151
1,310,595
57.8%

3421
Cutlery
332211
Cutlery
and
Flatware
(
except
Precious)

Manufacturing
164
2,198,365
99.6%

3423
Hand
&
Edge
Tools,
Except
Mach.
Tools,
Saws
332212
Hand
and
Edge
Tool
Manufacturing
1,069
5,677,903
86.0%

3425
Hand
Saws
and
Saw
Blades
332213
Saw
Blade
and
Handsaw
Manufacturing
176
1,452,540
100.0%

A­
13
MP&
M
EEBA:
Appendices
Appendix
A:
Detailed
Economic
Impact
Analysis
Information
Table
A.
3:
Relationships
between
SIC
and
NAICS
Codes
Based
on
1997
Economic
Census
for
MP&
M
Industries
Evaluated
for
the
Final
Rulea
(
thousands,
1997$)

SIC
SIC
Industry
NAICS
Code
1997
NAICS
Industry
Number
of
Establishments
Sales,

Shipments
or
Receipts
Share
Value
3429
Hardware
N.
E.
C.
332439
Other
Metal
Container
Manufacturing
117
402,378
17.7%

332510
Hardware
Manufacturing
952
10,359,952
96.0%

332919
Other
Metal
Valve
and
Pipe
Fitting
Manufacturing
16
0
3.9%
b
3433
Heatg.
Equip.
Except
Elec.
&

Warm
Air
Frnc.
333414
Heating
Equipment
(
except
Warm
Air
Furnaces)
Manufacturing
453
3,387,391
91.1%

3441
Fabricated
Structural
Metal
332312
Fabricated
Structural
Metal
Manufacturing
2,900
14,200,270
86.8%

3443
Fabricated
Plate
Work
(
Boiler
Shops)
332313
Plate
Work
Manufacturing
1,035
2,806,913
100.0%

332410
Power
Boiler
and
Heat
Exchanger
Manufacturing
472
3,849,100
100.0%

332420
Metal
Tank
(
Heavy
Gauge)
Manufacturing
614
4,764,118
100.0%

333415
Air­
Conditioning
and
Warm
Air
Heating
Equipment
and
Commercial
and
Industrial
Refrigeration
Equipment
Manufacturing
9
43,264
0.2%

3444
Sheet
Metal
Work
332322
Sheet
Metal
Work
Manufacturing
4,479
15,957,992
100.0%

332439
Other
Metal
Container
Manufacturing
126
275,440
12.1%

3446
Architectural
and
Ornamental
Metal
Work
332323
Ornamental
and
Architectural
Metal
Work
Manufacturing
1,744
3,536,413
88.2%

3448
Prefabricated
Metal
Buildings
&
Components
332311
Prefabricated
Metal
Building
and
Component
Manufacturing
604
4,199,550
100.0%

3449
Miscellaneous
Metal
Work
332114
Custom
Roll
Forming
401
3,074,662
100.0%

332312
Fabricated
Structural
Metal
Manufacturing
152
2,166,021
13.2%

332321
Metal
Window
and
Door
Manufacturing
33
364,564
3.6%

332323
Ornamental
and
Architectural
Metal
Work
Manufacturing
6
91,939
2.3%

3451
Screw
Machine
Products
332721
Precision
Turned
Product
Manufacturing
2,745
8,326,077
100.0%

3452
Bolts,
Nuts,
Screws,
Rivets,

and
Washers
332722
Bolt,
Nut,
Screw,
Rivet,
and
Washer
Manufacturing
1,040
8,134,661
100.0%

3462
Iron
and
Steel
Forgings
332111
Iron
and
Steel
Forging
421
4,924,426
100.0%

3466
Crowns
and
Closures
332115
Crown
and
Closure
Manufacturing
67
969,982
100.0%

3469
Metal
Stamping
N.
E.
C.
332116
Metal
Stamping
2,166
12,041,638
100.0%

332214
Kitchen
Utensil,
Pot,
and
Pan
Manufacturing
77
1,369,914
100.0%

3492
Fluid
Power
Valves
and
Hose
Fittings
332912
Fluid
Power
Valve
and
Hose
Fitting
Manufacturing
424
6,602,909
100.0%

3493
Steel
Springs
332611
Spring
(
Heavy
Gauge)
Manufacturing
129
761,711
100.0%

3494
Valves
&
Pipe
Fittings,

Except
Brass
332919
Other
Metal
Valve
and
Pipe
Fitting
Manufacturing
222
2,753,397
94.4%

332999
All
Other
Miscellaneous
Fabricated
Metal
Product
Manufacturing
23
73,983
0.7%

3495
Wire
Springs
332612
Spring
(
Light
Gauge)
Manufacturing
394
2,481,151
100.0%

334518
Watch,
Clock,
and
Part
Manufacturing
2
0
2.5%
b
A­
14
MP&
M
EEBA:
Appendices
Appendix
A:
Detailed
Economic
Impact
Analysis
Information
Table
A.
3:
Relationships
between
SIC
and
NAICS
Codes
Based
on
1997
Economic
Census
for
MP&
M
Industries
Evaluated
for
the
Final
Rulea
(
thousands,
1997$)

SIC
SIC
Industry
NAICS
Code
1997
NAICS
Industry
Number
of
Establishments
Sales,

Shipments
or
Receipts
Share
Value
3496
Miscellaneous
Fabricated
Wire
Products
332618
Other
Fabricated
Wire
Product
Manufacturing
1,253
4,587,656
87.3%

3498
Fabricated
Pipe
and
Fabricated
Pipe
Fitting
332996
Fabricated
Pipe
and
Pipe
Fitting
Manufacturing
856
4,024,999
100.0%

3499
Fabricated
Metal
Products
N.
E.
C.
332117
Powder
Metallurgy
Part
Manufacturing
128
1,317,301
100.0%

332439
Other
Metal
Container
Manufacturing
98
273,541
12.1%

332510
Hardware
Manufacturing
58
435,815
4.0%

332919
Other
Metal
Valve
and
Pipe
Fitting
Manufacturing
7
0
1.7%
b
332999
All
Other
Miscellaneous
Fabricated
Metal
Product
Manufacturing
2,592
7,558,137
71.9%

337215
Showcase,
Partition,
Shelving,
and
Locker
Manufacturing
78
123,057
1.5%

339914
Costume
Jewelry
and
Novelty
Manufacturing
82
49,953
3.9%

3541
Machine
Tools,
Metal
Cutting
Types
333512
Machine
Tool
(
Metal
Cutting
Types)

Manufacturing
393
5,183,521
97.2%

3542
Machine
Tools,
Metal
Forming
Types
333513
Machine
Tool
(
Metal
Forming
Types)

Manufacturing
225
2,255,011
100.0%

3544
Special
Dies
&
Tools,
Die
Sets,
Jigs,
Etc.
333511
Industrial
Mold
Manufacturing
2,529
5,116,635
100.0%

333514
Special
Die
and
Tool,
Die
Set,
Jig,
and
Fixture
Manufacturing
4,746
8,244,855
100.0%

3545
Machine
Tool
Access
&

Measuring
Devices
332212
Hand
and
Edge
Tool
Manufacturing
185
714,277
10.8%

333515
Cutting
Tool
and
Machine
Tool
Accessory
Manufacturing
1,920
5,347,173
100.0%

3546
Power
Driven
Hand
Tools
333991
Power­
Driven
Handtool
Manufacturing
217
3,609,779
100.0%

3965
Fasteners,
Buttons,
Needles,

Pins
339993
Fastener,
Button,
Needle,
and
Pin
Manufacturing
249
0
99.2%
b
Household
Equipment
2514
Metal
Household
Furniture
337124
Metal
Household
Furniture
Manufacturing
420
2,422,853
100.0%

2522
Office
Furniture,
Except
Wood
337214
Office
Furniture
(
except
Wood)

Manufacturing
359
8,230,935
100.0%

2531
Public
Buildng
&
Relatd
Furniture
336360
Motor
Vehicle
Seating
and
Interior
Trim
Manufacturing
184
6,060,320
57.1%

337127
Institutional
Furniture
Manufacturing
267
1,697,870
41.9%

339942
Lead
Pencil
and
Art
Good
Manufacturing
17
110,985
9.0%

2542
Partitions
&
Fixtures,
Exc
Wood
337215
Showcase,
Partition,
Shelving,
and
Locker
Manufacturing
926
5,249,474
65.6%

2591
Drapery
Hardware
and
Window
Blinds/
Shades
337920
Blind
and
Shade
Manufacturing
488
2,393,564
100.0%

2599
Furniture
and
Fixtures,

N.
E.
C.
337127
Institutional
Furniture
Manufacturing
727
2,305,770
57.0%

339113
Surgical
Appliance
and
Supplies
Manufacturing
16
645,688
4.2%

A­
15
MP&
M
EEBA:
Appendices
Appendix
A:
Detailed
Economic
Impact
Analysis
Information
Table
A.
3:
Relationships
between
SIC
and
NAICS
Codes
Based
on
1997
Economic
Census
for
MP&
M
Industries
Evaluated
for
the
Final
Rulea
(
thousands,
1997$)

SIC
SIC
Industry
NAICS
Code
1997
NAICS
Industry
Number
of
Establishments
Sales,

Shipments
or
Receipts
Share
Value
3431
Metal
Sanitary
Ware
332998
Enameled
Iron
and
Metal
Sanitary
Ware
Manufacturing
88
1,575,505
100.0%

3432
Plumbing
Fittings
and
Brass
Goods
332913
Plumbing
Fixture
Fitting
and
Trim
Manufacturing
116
3,590,128
100.0%

332999
All
Other
Miscellaneous
Fabricated
Metal
Product
Manufacturing
5
118,059
1.1%

3442
Metal
Doors,
Sash,
and
Trim
332321
Metal
Window
and
Door
Manufacturing
1,384
9,876,049
96.4%

3631
Household
Cooking
Equipment
335221
Household
Cooking
Appliance
Manufacturing
84
3,543,231
100.0%

3632
Household
Refrig.
&
Home
&
Farm
&
Freezers
335222
Household
Refrigerator
and
Home
Freezer
Manufacturing
27
4,887,364
100.0%

3633
Household
Laundry
Equipment
335224
Household
Laundry
Equipment
Manufacturing
17
3,723,375
100.0%

3634
Electric
Housewares
and
Fans
333414
Heating
Equipment
(
except
Warm
Air
Furnaces)
Manufacturing
16
329,270
8.9%

335211
Electric
Housewares
and
Household
Fan
Manufacturing
138
3,488,251
100.0%

3635
Household
Vacuum
Cleaners
335212
Household
Vacuum
Cleaner
Manufacturing
34
2,399,206
100.0%

3639
Household
Appliances
N.
E.
C.
333298
All
Other
Industrial
Machinery
Manufacturing
4
0
0.2%
b
335228
Other
Major
Household
Appliance
Manufacturing
36
3,300,662
100.0%

3641
Electric
Lamps
335110
Electric
Lamp
Bulb
and
Part
Manufacturing
82
3,306,009
100.0%

3643
Current­
Carrying
Wiring
Devices
335931
Current­
Carrying
Wiring
Device
Manufacturing
519
5,877,522
100.0%

3644
Noncurrent­
Carrying
Wiring
Devices
335932
Noncurrent­
Carrying
Wiring
Device
Manufacturing
219
4,451,186
100.0%

3645
Residential
Electrical
Lighting
Fixtures
335121
Residential
Electric
Lighting
Fixture
Manufacturing
497
2,177,355
96.6%

3646
Commercial,
Industrial,
and
Institutional
335122
Commercial,
Industrial,
and
Institutional
Electric
Lighting
Fixture
Manufacturing
356
4,047,437
100.0%

3648
Lighting
Equipment
N.
E.
C.
335129
Other
Lighting
Equipment
Manufacturing
327
3,054,806
100.0%

3651
Radio/
Television
Sets
Except
Commun.
Types
334310
Audio
and
Video
Equipment
Manufacturing
554
8,454,194
100.0%

7623
Refrig,
air
condition
811310
Commercial
and
Industrial
Machinery
and
Equipment
(
except
Automotive
and
Electronic)
Repair
and
Maintenance
2,343
1,890,237
10.8%

811412
Appliance
Repair
and
Maintenance
1,671
789,622
19.9%

Instruments
3812
Search,
Det,
Nav,
Ggnc,

Aero,
Naut
Sys/
Inst
334511
Search,
Detection,
Navigation,
Guidance,

Aeronautical,
and
Nautical
System
and
Instrument
Manufacturing
680
32,497,776
99.8%

3821
Laboratory
Apparatus
and
Furniture
339111
Laboratory
Apparatus
and
Furniture
Manufacturing
385
2,471,153
100.0%

A­
16
MP&
M
EEBA:
Appendices
Appendix
A:
Detailed
Economic
Impact
Analysis
Information
Table
A.
3:
Relationships
between
SIC
and
NAICS
Codes
Based
on
1997
Economic
Census
for
MP&
M
Industries
Evaluated
for
the
Final
Rulea
(
thousands,
1997$)

SIC
SIC
Industry
NAICS
Code
1997
NAICS
Industry
Number
of
Establishments
Sales,

Shipments
or
Receipts
Share
Value
3822
Automatic
Environmental
Controls
334512
Automatic
Environmental
Control
Manufacturing
for
Residential,
Commercial,

and
Appliance
Use
317
2,935,692
100.0%

3823
Process
Control
Instruments
334513
Instruments
and
Related
Products
Manufacturing
for
Measuring,
Displaying,

and
Controlling
Industrial
Process
Variables
1,002
7,890,923
100.0%

3824
Fluid
Meters
and
Counting
Devices
334514
Totalizing
Fluid
Meter
and
Counting
Device
Manufacturing
222
3,765,769
100.0%

3825
Instruments
to
Measure
Electricity
334416
Electronic
Coil,
Transformer,
and
Other
Inductor
Manufacturing
17
24,303
1.6%

334515
Instrument
Manufacturing
for
Measuring
and
Testing
Electricity
and
Electrical
Signals
826
13,852,897
100.0%

3826
Laboratory
Analytical
Instruments
334516
Analytical
Laboratory
Instrument
Manufacturing
664
7,157,038
99.5%

3827
Optical
Instruments
and
Lenses
333314
Optical
Instrument
and
Lens
Manufacturing
495
3,174,652
99.8%

3829
Measuring
and
Controlling
Devices
N.
E.
C.
334519
Other
Measuring
and
Controlling
Device
Manufacturing
853
5,114,547
99.9%

339112
Surgical
and
Medical
Instrument
Manufacturing
6
62,148
0.3%

3841
Surgical
&
Medical
Instruments
&
Apparatus
339112
Surgical
and
Medical
Instrument
Manufacturing
1,598
18,450,024
99.7%

3842
Orthopedic,
Prosthetic
&

Surgical
Suppl.
322121
Paper
(
except
Newsprint)
Mills
2
0
1.4%
b
322291
Sanitary
Paper
Product
Manufacturing
16
651,398
6.7%

334510
Electromedical
and
Electrotherapeutic
Apparatus
Manufacturing
74
807,427
7.1%

339113
Surgical
Appliance
and
Supplies
Manufacturing
1,636
14,743,779
95.8%

3843
Dental
Equipment
and
Supplies
339114
Dental
Equipment
and
Supplies
Manufacturing
877
2,699,867
100.0%

3844
X­
Ray
Apparatus
and
Tubes
334517
Irradiation
Apparatus
Manufacturing
155
3,942,256
100.0%

3845
Electromedical
Equipment
334510
Electromedical
and
Electrotherapeutic
Apparatus
Manufacturing
460
10,567,566
92.5%

3851
Ophthalmic
Goods
339115
Ophthalmic
Goods
Manufacturing
575
3,607,813
100.0%

7629
Electric
repair
shop
811212
Computer
and
Office
Machine
Repair
and
Maintenance
1,538
913,258
10.7%

811213
Communication
Equipment
Repair
and
Maintenance
201
231,458
14.4%

811219
Other
Electronic
and
Precision
Equipment
Repair
and
Maintenance
2,033
2,509,452
86.1%

811411
Home
and
Garden
Equipment
Repair
and
Maintenance
579
185,507
18.5%

811412
Appliance
Repair
and
Maintenance
4,327
3,125,853
78.6%

A­
17
MP&
M
EEBA:
Appendices
Appendix
A:
Detailed
Economic
Impact
Analysis
Information
Table
A.
3:
Relationships
between
SIC
and
NAICS
Codes
Based
on
1997
Economic
Census
for
MP&
M
Industries
Evaluated
for
the
Final
Rulea
(
thousands,
1997$)

SIC
SIC
Industry
NAICS
Code
1997
NAICS
Industry
Number
of
Establishments
Sales,

Shipments
or
Receipts
Share
Value
Iron
and
Steel
3315
Steel
Wiredrawing
and
Steel
Nails
and
Spikes
331222
Steel
Wire
Drawing
273
4,920,798
100.0%

332618
Other
Fabricated
Wire
Product
Manufacturing
31
370,492
7.0%

3316
Cold­
Rolled
Steel
Sheet,

Strip,
and
Bars
331221
Rolled
Steel
Shape
Manufacturing
186
6,343,466
100.0%

3317
Steel
Pipe
and
Tubes
331210
Iron
and
Steel
Pipe
and
Tube
Manufacturing
from
Purchased
Steel
235
7,565,377
100.0%

Job
Shop
3471
Plating
and
Polishing
332813
Electroplating,
Plating,
Polishing,
Anodizing,

and
Coloring
3,404
5,979,405
100.0%

3479
Metal
Coating
&
Allied
Services
332812
Metal
Coating,
Engraving
(
except
Jewelry
and
Silverware),
and
Allied
Services
to
Manufacturers
2,156
8,460,896
100.0%

339911
Jewelry
(
except
Costume)
Manufacturing
22
5,798
0.1%

339914
Costume
Jewelry
and
Novelty
Manufacturing
16
2,257
0.2%

339912
Silverware
and
Hollowware
Manufacturing
12
6,296
0.7%

Mobile
Industrial
Equipment
3523
Farm
Machinery
and
Equipment
332212
Hand
and
Edge
Tool
Manufacturing
1
0
0.1%
b
332323
Ornamental
and
Architectural
Metal
Work
Manufacturing
140
380,152
9.5%

333111
Farm
Machinery
and
Equipment
Manufacturing
1,339
15,921,455
100.0%

333922
Conveyor
and
Conveying
Equipment
Manufacturing
28
33,377
0.5%

3524
Garden
Tractors
&
Lawn
&

Garden
Equipment
332212
Hand
and
Edge
Tool
Manufacturing
3
0
0.3%
b
333112
Lawn
and
Garden
Tractor
and
Home
Lawn
and
Garden
Equipment
Manufacturing
145
7,454,511
100.0%

3531
Constr
Mach
and
Eq
333120
Construction
Machinery
Manufacturing
785
21,965,455
100.0%

333923
Overhead
Traveling
Crane,
Hoist,
and
Monorail
System
Manufacturing
87
1,805,198
57.4%

336510
Railroad
Rolling
Stock
Manufacturing
25
346,760
4.2%

3532
Mining
Mach.
&
Equip.,

Except
Oil
Field
333131
Mining
Machinery
and
Equipment
Manufacturing
292
2,710,923
100.0%

3536
Hoists,
Industrial
Cranes
&

Monorails
333923
Overhead
Traveling
Crane,
Hoist,
and
Monorail
System
Manufacturing
220
1,340,561
42.6%

3537
Industrial
Trucks,
Tractors,

Trailers
332439
Other
Metal
Container
Manufacturing
4
6,775
0.3%

332999
All
Other
Miscellaneous
Fabricated
Metal
Product
Manufacturing
19
27,488
0.3%

333924
Industrial
Truck,
Tractor,
Trailer,
and
Stacker
Machinery
Manufacturing
461
5,538,326
100.0%

3795
Tanks
and
Tank
Components
336992
Military
Armored
Vehicle,
Tank,
and
Tank
Component
Manufacturing
37
0
86.0%
b
A­
18
MP&
M
EEBA:
Appendices
Appendix
A:
Detailed
Economic
Impact
Analysis
Information
Table
A.
3:
Relationships
between
SIC
and
NAICS
Codes
Based
on
1997
Economic
Census
for
MP&
M
Industries
Evaluated
for
the
Final
Rulea
(
thousands,
1997$)

SIC
SIC
Industry
NAICS
Code
1997
NAICS
Industry
Number
of
Establishments
Sales,

Shipments
or
Receipts
Share
Value
Motor
Vehicle
3465
Automotive
Stampings
336370
Motor
Vehicle
Metal
Stamping
810
23,668,110
100.0%

3592
Carburetors,
Piston
Rings,

Valves
336311
Carburetor,
Piston,
Piston
Ring,
and
Valve
Manufacturing
141
2,755,311
100.0%

3647
Vehicular
Lighting
Equipment
336321
Vehicular
Lighting
Equipment
Manufacturing
106
3,282,824
100.0%

3694
Electrical
Equipment
for
Motor
Vehicles
336322
Other
Motor
Vehicle
Electrical
and
Electronic
Equipment
Manufacturing
569
9,074,335
53.6%

3711
Motor
Vehicle
and
Automobile
Bodies
336111
Automobile
Manufacturing
194
95,385,563
100.0%

336112
Light
Truck
and
Utility
Vehicle
Manufacturing
112
110,400,169
100.0%

336120
Heavy
Duty
Truck
Manufacturing
84
14,490,344
100.0%

336211
Motor
Vehicle
Body
Manufacturing
76
82,633
0.9%

336992
Military
Armored
Vehicle,
Tank,
and
Tank
Component
Manufacturing
6
0
14.0%
b
3714
Motor
Vehicle
Parts
and
Accessories
336211
Motor
Vehicle
Body
Manufacturing
23
265,552
2.9%

336312
Gasoline
Engine
and
Engine
Parts
Manufacturing
881
25,974,369
100.0%

336322
Other
Motor
Vehicle
Electrical
and
Electronic
Equipment
Manufacturing
193
6,446,681
38.1%

336330
Motor
Vehicle
Steering
and
Suspension
Components
(
except
Spring)
Manufacturing
212
10,750,312
100.0%

336340
Motor
Vehicle
Brake
System
Manufacturing
269
10,033,288
100.0%

336350
Motor
Vehicle
Transmission
and
Power
Train
Parts
Manufacturing
523
33,288,093
100.0%

336399
All
Other
Motor
Vehicle
Parts
Manufacturing
1,508
34,193,298
99.6%

3716
Mobile
Homes
336213
Motor
Home
Manufacturing
88
3,943,709
100.0%

3751
Motorcycles
336991
Motorcycle,
Bicycle,
and
Parts
Manufacturing
385
0
99.0%
b
3792
Travel
Trailers
and
Campers
336214
Travel
Trailer
and
Camper
Manufacturing
315
3,076,049
67.4%

3799
Miscellaneous
Transportation
Equipment
332212
Hand
and
Edge
Tool
Manufacturing
1
0
0.1%
b
336214
Travel
Trailer
and
Camper
Manufacturing
498
1,485,367
32.6%

336999
All
Other
Transportation
Equipment
Manufacturing
378
4,557,989
100.0%

4121
Taxicabs
485310
Taxi
Service
3,184
1,280,597
100.0%

5013
Motor
Vehicle
Supplies
and
New
Parts
421120
Motor
Vehicle
Supplies
and
New
Parts
Wholesalers
12,620
83,214,728
100.0%

441310
Automotive
Parts
and
Accessories
Stores
16,253
22,093,428
51.2%

5511
Motor
Vehicle
Dealers
(
New
and
Used)
441110
New
Car
Dealers
25,897
518,971,824
100.0%

5521
Motor
Vehicle
Dealers
(
Used
Only)
441120
Used
Car
Dealers
23,340
34,680,468
100.0%

5561
Recreational
Vehicle
Dealers
441210
Recreational
Vehicle
Dealers
3,014
10,069,749
100.0%

A­
19
MP&
M
EEBA:
Appendices
Appendix
A:
Detailed
Economic
Impact
Analysis
Information
Table
A.
3:
Relationships
between
SIC
and
NAICS
Codes
Based
on
1997
Economic
Census
for
MP&
M
Industries
Evaluated
for
the
Final
Rulea
(
thousands,
1997$)

SIC
SIC
Industry
NAICS
Code
1997
NAICS
Industry
Number
of
Establishments
Sales,

Shipments
or
Receipts
Share
Value
5571
Motorcycle
Dealers
441221
Motorcycle
Dealers
3,635
7,369,260
100.0%

5599
Automotive
Dealers,
N.
E.
C.
441229
All
Other
Motor
Vehicle
Dealers
1,678
2,517,267
100.0%

7514
Passenger
Car
Rental
532111
Passenger
Car
Rental
4,367
14,783,704
100.0%

7515
Passenger
Car
Lease
532112
Passenger
Car
Leasing
879
3,800,424
100.0%

7519
Utility
Trailer
and
Recreational
Vehicle
Rental
532120
Truck,
Utility
Trailer,
and
RV
(
Recreational
Vehicle)
Rental
and
Leasing
360
256,119
2.5%

7532
Top,
Body,
and
Upholstery
Repair
and
Paint
Shops
811121
Automotive
Body,
Paint,
and
Interior
Repair
and
Maintenance
35,569
17,755,296
100.0%

7533
Auto
Exhaust
Systems
811112
Automotive
Exhaust
System
Repair
5,251
1,985,377
100.0%

7537
Auto
Transmission
Repair
811113
Automotive
Transmission
Repair
6,768
2,431,584
100.0%

7538
Gen
Automotive
Repair
811111
General
Automotive
Repair
77,751
25,598,455
100.0%

7539
Auto
Repair
Shop,
N.
E.
C.
811118
Other
Automotive
Mechanical
and
Electrical
Repair
and
Maintenance
9,674
3,494,643
100.0%

7549
Auto
Services,
Except
Repair
and
Carwashes
488410
Motor
Vehicle
Towing
5,893
2,295,188
100.0%

811191
Automotive
Oil
Change
and
Lubrication
Shops
7,413
2,787,318
100.0%

811198
All
Other
Automotive
Repair
and
Maintenance
1,646
798,626
73.5%

Office
Machine
3571
Electronic
Computers
334111
Electronic
Computer
Manufacturing
563
66,331,909
100.0%

3572
Typewriters
334112
Computer
Storage
Device
Manufacturing
211
13,907,367
100.0%

3575
Computer
Terminals
334113
Computer
Terminal
Manufacturing
142
1,483,460
100.0%

3577
Computer
Peripheral
Eq
N.
E.
C.
334119
Other
Computer
Peripheral
Equipment
Manufacturing
1,006
25,130,308
93.1%

3578
Calculating,
Accounting
Machines
Except
Computers
333313
Office
Machinery
Manufacturing
35
144,380
4.5%

334119
Other
Computer
Peripheral
Equipment
Manufacturing
61
1,870,426
6.9%

3579
Office
Machines,
N.
E.
C.
333313
Office
Machinery
Manufacturing
134
3,047,549
95.5%

334518
Watch,
Clock,
and
Part
Manufacturing
16
0
19.6%
b
339942
Lead
Pencil
and
Art
Good
Manufacturing
13
257,020
20.8%

7378
Computer
Maintenance
and
Repairs
811212
Computer
and
Office
Machine
Repair
and
Maintenance
6,087
7,565,169
89.0%

7379
Computer
Related
Services,

N.
E.
C.
334611
Software
Reproducing
124
1,258,435
100.0%

541512
Computer
Systems
Design
Services
20,233
15,942,861
31.1%

541519
Other
Computer
Related
Services
8,405
4,339,989
100.0%

Ordnance
3482
Small
Arms
Ammunition
332992
Small
Arms
Ammunition
Manufacturing
113
938,818
100.0%

3483
Ammunition,
Except
for
Small
Arms
332993
Ammunition
(
except
Small
Arms)

Manufacturing
53
1,497,045
100.0%

3484
Small
Arms
332994
Small
Arms
Manufacturing
198
1,251,792
100.0%

A­
20
MP&
M
EEBA:
Appendices
Appendix
A:
Detailed
Economic
Impact
Analysis
Information
Table
A.
3:
Relationships
between
SIC
and
NAICS
Codes
Based
on
1997
Economic
Census
for
MP&
M
Industries
Evaluated
for
the
Final
Rulea
(
thousands,
1997$)

SIC
SIC
Industry
NAICS
Code
1997
NAICS
Industry
Number
of
Establishments
Sales,

Shipments
or
Receipts
Share
Value
3489
Ordnance
and
Accessories,

N.
E.
C.
332995
Other
Ordnance
and
Accessories
Manufacturing
70
1,750,485
100.0%

Miscellaneous
Metal
Products
3497
Metal
Foil
and
Leaf
322225
Laminated
Aluminum
Foil
Manufacturing
for
Flexible
Packaging
Uses
43
1,546,143
100.0%

332999
All
Other
Miscellaneous
Fabricated
Metal
Product
Manufacturing
64
1,711,600
16.3%

3861
Photographic
Equipment
&

Supplies
325992
Photographic
Film,
Paper,
Plate,
and
Chemical
Manufacturing
311
12,895,637
100.0%

333315
Photographic
and
Photocopying
Equipment
Manufacturing
428
8,410,124
100.0%

3931
Musical
Instruments
339992
Musical
Instrument
Manufacturing
576
1,356,651
100.0%

3944
Games,
Toys,
Children's
Vehicles
336991
Motorcycle,
Bicycle,
and
Parts
Manufacturing
4
0
1.0%
b
339932
Game,
Toy,
and
Children's
Vehicle
Manufacturing
785
4,534,497
100.0%

3949
Sporting
and
Athletic
Goods,

N.
E.
C.
339920
Sporting
and
Athletic
Goods
Manufacturing
2,571
10,591,160
100.0%

3951
Pens
and
Mechanical
Pencils
339941
Pen
and
Mechanical
Pencil
Manufacturing
112
1,590,770
100.0%

3953
Marking
Devices
339943
Marking
Device
Manufacturing
634
643,007
100.0%

3993
Signs
and
Advertising
Displays
339950
Sign
Manufacturing
5,709
7,910,809
100.0%

3995
Burial
Caskets
339995
Burial
Casket
Manufacturing
177
1,271,184
100.0%

3999
Manufacturing
Industries,

N.
E.
C.
314999
All
Other
Miscellaneous
Textile
Product
Mills
52
173,353
2.8%

316110
Leather
and
Hide
Tanning
and
Finishing
26
24,625
0.7%

325998
All
Other
Miscellaneous
Chemical
Product
and
Preparation
Manufacturing
9
80,624
0.6%

326199
All
Other
Plastics
Product
Manufacturing
140
319,241
0.5%

332212
Hand
and
Edge
Tool
Manufacturing
7
0
0.6%
b
332999
All
Other
Miscellaneous
Fabricated
Metal
Product
Manufacturing
185
285,362
2.7%

335121
Residential
Electric
Lighting
Fixture
Manufacturing
53
69,864
3.1%

337127
Institutional
Furniture
Manufacturing
5
28,296
0.7%

339999
All
Other
Miscellaneous
Manufacturing
2,284
7,183,815
85.4%

7692
Welding
Repair
811490
Other
Personal
and
Household
Goods
Repair
and
Maintenance
4,840
1,640,808
36.8%

A­
21
MP&
M
EEBA:
Appendices
Appendix
A:
Detailed
Economic
Impact
Analysis
Information
Table
A.
3:
Relationships
between
SIC
and
NAICS
Codes
Based
on
1997
Economic
Census
for
MP&
M
Industries
Evaluated
for
the
Final
Rulea
(
thousands,
1997$)

SIC
SIC
Industry
NAICS
Code
1997
NAICS
Industry
Number
of
Establishments
Sales,

Shipments
or
Receipts
Share
Value
7699
Repair
Shop,
Related
Service
488390
Other
Support
Activities
for
Water
Transportation
12
4,737
0.7%

561622
Locksmiths
3,799
1,081,317
100.0%

561790
Other
Services
to
Buildings
and
Dwellings
1,254
0
22.4%
b
562991
Septic
Tank
and
Related
Services
2,538
0
81.8%
b
811212
Computer
and
Office
Machine
Repair
and
Maintenance
104
23,844
0.3%

811219
Other
Electronic
and
Precision
Equipment
Repair
and
Maintenance
838
404,627
13.9%

811310
Commercial
and
Industrial
Machinery
and
Equipment
(
except
Automotive
and
Electronic)
Repair
and
Maintenance
16,404
13,600,413
77.7%

811411
Home
and
Garden
Equipment
Repair
and
Maintenance
3,032
816,008
81.5%

811412
Appliance
Repair
and
Maintenance
181
59,338
1.5%

811430
Footwear
and
Leather
Goods
Repair
82
18,294
7.0%

811490
Other
Personal
and
Household
Goods
Repair
and
Maintenance
3,946
1,362,271
30.6%

3873
Watches,
Clocks,
and
Watchcases
334518
Watch,
Clock,
and
Part
Manufacturing
128
718,191
77.9%

Precious
Metals
and
Jewelry
3911
Jewelry,
Precious
Metal
339911
Jewelry
(
except
Costume)
Manufacturing
2,272
5,416,836
99.9%

3914
Silverware,
Plated
Ware
&

Stainless
332211
Cutlery
and
Flatware
(
except
Precious)

Manufacturing
11
8,032
0.4%

339912
Silverware
and
Hollowware
Manufacturing
151
899,684
99.3%

3915
Jewelers'
Materials
&

Lapidary
Work
339913
Jewelers'
Material
and
Lapidary
Work
Manufacturing
394
919,066
100.0%

3961
Costume
Jewelry
339914
Costume
Jewelry
and
Novelty
Manufacturing
826
1,223,475
95.9%

7631
Watch,
Clock,
Jewelry
Repair
811490
Other
Personal
and
Household
Goods
Repair
and
Maintenance
1,716
345,774
7.8%

Printed
Circuit
Boards
3672
Printed
Circuit
Boards
334412
Bare
Printed
Circuit
Board
Manufacturing
1,401
9,787,576
100.0%

Railroad
3743
Railcars,
Railway
Systems
336510
Railroad
Rolling
Stock
Manufacturing
207
7,916,635
95.8%

Ships
and
Boats
3731
Ship
Building
and
Repairing
336611
Ship
Building
and
Repairing
700
10,571,810
100.0%

3732
Boat
Building
and
Repairing
336612
Boat
Building
1,043
5,622,040
100.0%

811490
Other
Personal
and
Household
Goods
Repair
and
Maintenance
1,739
821,273
18.4%

4412
Deep
Sea
Foreign
Transportation
483111
Deep
Sea
Freight
Transportation
487
11,570,718
100.0%

4424
Deep
Sea
Domestic
Transportation
483113
Coastal
and
Great
Lakes
Freight
Transportation
292
3,114,639
66.6%

A­
22
MP&
M
EEBA:
Appendices
Appendix
A:
Detailed
Economic
Impact
Analysis
Information
Table
A.
3:
Relationships
between
SIC
and
NAICS
Codes
Based
on
1997
Economic
Census
for
MP&
M
Industries
Evaluated
for
the
Final
Rulea
(
thousands,
1997$)

SIC
SIC
Industry
NAICS
Code
1997
NAICS
Industry
Number
of
Establishments
Sales,

Shipments
or
Receipts
Share
Value
4432
Freight
Transportation
Great
Lakes
483113
Coastal
and
Great
Lakes
Freight
Transportation
32
519,863
11.1%

4449
Water
Transportation
of
Freight,
N.
E.
C.
483211
Inland
Water
Freight
Transportation
222
2,821,121
83.3%

4481
Deep
Sea
Passenger
Transportation
483112
Deep
Sea
Passenger
Transportation
80
3,908,143
100.0%

483114
Coastal
and
Great
Lakes
Passenger
Transportation
64
89,597
49.2%

4482
Ferries
483114
Coastal
and
Great
Lakes
Passenger
Transportation
61
92,493
50.8%

483212
Inland
Water
Passenger
Transportation
76
121,992
41.6%

4489
Water
Passenger
Transportation,
N.
E.
C.
483212
Inland
Water
Passenger
Transportation
154
171,135
58.4%

487210
Scenic
and
Sightseeing
Transportation,
Water
654
861,001
76.3%

4491
Marine
Cargo
Handling
488310
Port
and
Harbor
Operations
168
889,125
100.0%

488320
Marine
Cargo
Handling
623
4,456,243
100.0%

4492
Towing
&
Tugboat
Service
483113
Coastal
and
Great
Lakes
Freight
Transportation
292
1,043,440
22.3%

483211
Inland
Water
Freight
Transportation
161
566,027
16.7%

488330
Navigational
Services
to
Shipping
361
1,014,026
67.0%

4493
Marinas
713930
Marinas
4,217
2,541,481
100.0%

4499
Water
Transporation
Services,
N.
E.
C.
488330
Navigational
Services
to
Shipping
504
499,176
33.0%

488390
Other
Support
Activities
for
Water
Transportation
640
444,499
67.7%

532411
Commercial
Air,
Rail,
and
Water
Transportation
Equipment
Rental
and
Leasing
126
454,392
7.1%

Stationary
Industrial
Equipment
3511
Steam,
Gas,
Hydraulic
Turbines,
Generator
Units
333611
Turbine
and
Turbine
Generator
Set
Units
Manufacturing
86
5,783,057
100.0%

3519
Internal
Combustion
Engines,

N.
E.
C.
333618
Other
Engine
Equipment
Manufacturing
297
0
99.3%
b
336399
All
Other
Motor
Vehicle
Parts
Manufacturing
7
123,954
0.4%

3533
Oil
Field
Machinery
and
Equipment
333132
Oil
and
Gas
Field
Machinery
and
Equipment
Manufacturing
563
6,240,079
100.0%

3534
Elevators
and
Moving
Stairways
333921
Elevator
and
Moving
Stairway
Manufacturing
196
1,607,066
100.0%

3535
Conveyors
and
Conveying
Equipment
333922
Conveyor
and
Conveying
Equipment
Manufacturing
871
6,346,525
99.5%

3543
Industrial
Patterns
332997
Industrial
Pattern
Manufacturing
673
623,927
100.0%

3547
Rolling
Mill
Machinery
and
Equipment
333516
Rolling
Mill
Machinery
and
Equipment
Manufacturing
100
700,084
100.0%

3548
Electric
and
Gas
Welding
and
Soldering
333992
Welding
and
Soldering
Equipment
Manufacturing
244
4,433,877
99.8%

3549
Metal
Working
Machinery,

N.
E.
C.
333518
Other
Metalworking
Machinery
Manufacturing
474
3,463,811
100.0%

A­
23
MP&
M
EEBA:
Appendices
Appendix
A:
Detailed
Economic
Impact
Analysis
Information
Table
A.
3:
Relationships
between
SIC
and
NAICS
Codes
Based
on
1997
Economic
Census
for
MP&
M
Industries
Evaluated
for
the
Final
Rulea
(
thousands,
1997$)

SIC
SIC
Industry
NAICS
Code
1997
NAICS
Industry
Number
of
Establishments
Sales,

Shipments
or
Receipts
Share
Value
3552
Textile
Machinery
333292
Textile
Machinery
Manufacturing
478
1,779,034
100.0%

3553
Woodworking
Machinery
333210
Sawmill
and
Woodworking
Machinery
Manufacturing
327
1,321,752
100.0%

3554
Paper
Industries
Machinery
333291
Paper
Industry
Machinery
Manufacturing
366
3,438,235
100.0%

3555
Printing
Trades
Machinery
and
Equipment
333293
Printing
Machinery
and
Equipment
Manufacturing
546
0
99.1%
b
3556
Food
Products
Mach
333294
Food
Product
Machinery
Manufacturing
597
2,877,841
100.0%

3559
Special
Industry
Machinery,

N.
E.
C.
333220
Plastics
and
Rubber
Industry
Machinery
Manufacturing
455
3,584,992
100.0%

333295
Semiconductor
Machinery
Manufacturing
257
11,158,627
100.0%

333298
All
Other
Industrial
Machinery
Manufacturing
1,677
0
99.8%
b
333319
Other
Commercial
and
Service
Industry
Machinery
Manufacturing
78
644,019
6.9%

3561
Pumps
and
Pumping
Equipment
333911
Pump
and
Pumping
Equipment
Manufacturing
489
6,826,043
100.0%

3562
Ball
and
Roller
Bearings
332991
Ball
and
Roller
Bearing
Manufacturing
185
6,120,940
100.0%

3563
Air
and
Gas
Compressors
333912
Air
and
Gas
Compressor
Manufacturing
314
5,633,008
100.0%

3564
Blowers
and
Exhaust
and
Ventilation
Fans
333411
Air
Purification
Equipment
Manufacturing
370
2,174,729
100.0%

333412
Industrial
and
Commercial
Fan
and
Blower
Manufacturing
204
1,901,196
100.0%

3565
Industrial
Patterns
333993
Packaging
Machinery
Manufacturing
689
4,858,270
100.0%

3566
Speed
Changers,
High
Speed
Drivers
&
Gears
333612
Speed
Changer,
Industrial
High­
Speed
Drive,

and
Gear
Manufacturing
268
2,402,392
100.0%

3567
Industrial
Process
Furnaces
and
Ovens
333994
Industrial
Process
Furnace
and
Oven
Manufacturing
404
2,871,475
100.0%

3568
Mechanical
Power
Transmission
Equipment,

N.
E.
C.
333613
Mechanical
Power
Transmission
Equipment
Manufacturing
299
3,301,091
100.0%

3569
General
Industrial
Machinery,
N.
E.
C.
333999
All
Other
Miscellaneous
General
Purpose
Machinery
Manufacturing
1,257
7,991,746
87.5%

3581
Automatic
Merchandising
Machines
333311
Automatic
Vending
Machine
Manufacturing
121
1,325,960
100.0%

3582
Commercial
Laundry
Equipment
333312
Commercial
Laundry,
Drycleaning,
and
Pressing
Machine
Manufacturing
68
604,966
100.0%

3585
Refrigeration
&
Air
and
Heating
Equipment
333415
Air­
Conditioning
and
Warm
Air
Heating
Equipment
and
Commercial
and
Industrial
Refrigeration
Equipment
Manufacturing
792
22,846,865
99.8%

336391
Motor
Vehicle
Air­
Conditioning
Manufacturing
60
5,626,596
100.0%

3586
Measuring
and
Dispensing
Pumps
333913
Measuring
and
Dispensing
Pump
Manufacturing
71
1,316,899
100.0%

3589
Service
Industry
Machines,

N.
E.
C.
333319
Other
Commercial
and
Service
Industry
Machinery
Manufacturing
1,165
7,596,253
81.3%

A­
24
MP&
M
EEBA:
Appendices
Appendix
A:
Detailed
Economic
Impact
Analysis
Information
Table
A.
3:
Relationships
between
SIC
and
NAICS
Codes
Based
on
1997
Economic
Census
for
MP&
M
Industries
Evaluated
for
the
Final
Rulea
(
thousands,
1997$)

SIC
SIC
Industry
NAICS
Code
1997
NAICS
Industry
Number
of
Establishments
Sales,

Shipments
or
Receipts
Share
Value
3593
Fluid
Power
Cylinders
and
Actuators
333995
Fluid
Power
Cylinder
and
Actuator
Manufacturing
320
3,528,906
100.0%

3594
Fluid
Power
Pumps
and
Motors
333996
Fluid
Power
Pump
and
Motor
Manufacturing
170
2,712,058
100.0%

3596
Scales
and
Balances,
except
Laboratory
333997
Scale
and
Balance
(
except
Laboratory)

Manufacturing
122
682,940
100.0%

3599
Machinery,
Except
Electrical,

N.
E.
C.
332710
Machine
Shops
23,619
27,143,131
100.0%

332999
All
Other
Miscellaneous
Fabricated
Metal
Product
Manufacturing
132
506,611
4.8%

333319
Other
Commercial
and
Service
Industry
Machinery
Manufacturing
50
172,536
1.8%

333999
All
Other
Miscellaneous
General
Purpose
Machinery
Manufacturing
836
1,146,348
12.5%

3612
Transformers
335311
Power,
Distribution,
and
Specialty
Transformer
Manufacturing
318
4,716,162
100.0%

3613
Switchgear
and
Switchboard
Apparatus
335313
Switchgear
and
Switchboard
Apparatus
Manufacturing
583
7,609,164
100.0%

3621
Motors
and
Generators
335312
Motor
and
Generator
Manufacturing
528
11,788,281
96.3%

3629
Electric
Industrial
Apparatus,

N.
E.
C.
335999
All
Other
Miscellaneous
Electrical
Equipment
and
Component
Manufacturing
413
2,838,366
41.2%

7353
Heavy
Construction
Equip
Rental,
Leasing
234990
All
Other
Heavy
Construction
2,295
2,734,732
8.7%

532412
Construction,
Mining,
and
Forestry
Machinery
and
Equipment
Rental
and
Leasing
3,286
5,339,163
77.4%

7359
Equip
Rental,
Leasing,

N.
E.
C.
532210
Consumer
Electronics
and
Appliances
Rental
3,011
1,790,890
100.0%

532299
All
Other
Consumer
Goods
Rental
3,133
2,133,450
99.1%

532310
General
Rental
Centers
6,509
3,910,618
100.0%

532411
Commercial
Air,
Rail,
and
Water
Transportation
Equipment
Rental
and
Leasing
498
0
74.3%
b
532412
Construction,
Mining,
and
Forestry
Machinery
and
Equipment
Rental
and
Leasing
671
1,555,089
22.6%

532420
Office
Machinery
and
Equipment
Rental
and
Leasing
400
436,178
7.1%

532490
Other
Commercial
and
Industrial
Machinery
and
Equipment
Rental
and
Leasing
3,408
6,775,140
69.7%

562991
Septic
Tank
and
Related
Services
563
0
18.2%
b
a
EPA
evaluated
options
for
these
industrial
sectors
but
did
not
regulate
them
all
under
the
final
rule.

Share
values
were
calculated
using
estimated
value
of
shipments
data.

N.
E.
C.
=
Not
Elsewhere
Classified
b
Source:
Department
of
Commerce,
Bureau
of
the
Census,
1997
Economic
Census,
Bridge
Between
NAICS
and
SIC;
and
EPA
analysis.

A­
25
MP&
M
EEBA:
Appendices
Appendix
A:
Detailed
Economic
Impact
Analysis
Information
A.
2
ANNUAL
ESTABLISHMENT
 
BIRTHS 
AND
 
DEATHS 
IN
MP&
M
INDUSTRIES
EVALUATED
FOR
THE
FINAL
RULE
EPA
used
the
Statistics
of
U.
S.
Businesses
(
SUSB)
dynamic
data
to
estimate
the
rate
at
which
MP&
M
facilities
evaluated
for
the
final
rule
enter
and
leave
the
industry
each
year.
The
SUSB
dynamic
data
report
numbers
of
facilities
starting
up,
closing,

expanding
employment
and
contracting
employment
each
year
from
1989
through
1997
(
the
latest
currently
available.)

Table
A.
4
shows
the
average
number
of
facilities
(
establishments)
operating
at
the
beginning
of
each
year
for
the
period
1989
through
1997,
the
number
of
facility
 
births 
and
 
deaths ,
and
the
average
 
birth
rate 
and
 
death
rate 
for
each
of
the
major
3­
digit
manufacturing
SIC
codes
that
include
MP&
M
4­
digit
SIC
codes
evaluated
for
the
final
rule.
2
This
table
shows
that,

over
the
period
1989­
1997,
annual
closure
rates
ranged
from
6
to
over
12
percent
in
the
different
industries,
with
an
overall
average
of
almost
8
percent.

Table
A.
4:
Annual
Births
and
Deaths
for
MP&
M
Establishments
Evaluated
for
the
Final
Rule
by
3
Digit
SIC
Codes
(
1989­
1997)

SIC
SIC
Description
Average
#

Establishments
at
the
Beginning
of
the
Year
Average
Establishment
Births
Average
Establishment
Deaths
%
Births
%
Deaths
3410
Metal
Cans
And
Shipping
Containers
464
22
35
4.7%
7.5%

3420
Cutlery,
Handtools,
And
Hardware
2,294
143
139
6.2%
6.1%

3430
Plumbing
And
Heating,
Except
Electric
687
45
53
6.6%
7.8%

3440
Fabricated
Structural
Metal
Products
12,268
853
908
7.0%
7.4%

3450
Screw
Machine
Products,
Bolts,
Etc.
2,436
84
111
3.4%
4.6%

3460
Metal
Forgings
And
Stamping
3,812
199
226
5.2%
5.9%

3470
Metal
Services,
N.
E.
C.
5,028
341
340
6.8%
6.8%

3480
Ordnance
&
Accessories,
N.
E.
C.
390
39
40
10.0%
10.2%

3490
Misc.
Fabricated
Metal
Products
7,084
606
531
8.6%
7.5%

3510
Engines
And
Turbines
346
26
24
7.5%
6.8%

3520
Farm
And
Garden
Machinery
1,711
133
129
7.8%
7.5%

3530
Construction
And
Related
Machinery
3,165
217
230
6.9%
7.3%

3540
Metalworking
Machinery
11,072
672
660
6.1%
6.0%

3550
Special
Industry
Machinery
4,427
307
317
6.9%
7.1%

3560
General
Industrial
Machinery
3,961
243
225
6.1%
5.7%

3570
Computer
And
Office
Equipment
2,025
262
246
12.9%
12.1%

3580
Refrigeration
And
Service
Machinery
2,104
154
165
7.3%
7.9%

3590
Industrial
Machinery,
N.
E.
C.
21,972
1,996
1,659
9.1%
7.5%

3610
Electric
Distribution
Equipment
764
53
51
6.9%
6.6%

3620
Electrical
Industrial
Apparatus
2,024
117
130
5.8%
6.4%

3630
Household
Appliances
461
44
41
9.5%
8.9%

2
The
data
are
disaggregated
only
to
the
3­
digit
SIC
level,
and
EPA
therefore
was
unable
to
calculate
closure
rates
for
the
specific
4­
digit
SICs
that
comprise
the
MP&
M
industries
evaluated
for
the
final
rule.
The
analysis
does
not
include
3­
digit
SICs
that
may
include
large
numbers
of
non­
metal
products
producers,
for
example
SIC
241
(
furniture,
both
wood
and
metal.)

A­
26
MP&
M
EEBA:
Appendices
Appendix
A:
Detailed
Economic
Impact
Analysis
Information
Table
A.
4:
Annual
Births
and
Deaths
for
MP&
M
Establishments
Evaluated
for
the
Final
Rule
by
3
Digit
SIC
Codes
(
1989­
1997)

SIC
SIC
Description
Average
#

Establishments
at
the
Beginning
of
the
Year
Average
Establishment
Births
Average
Establishment
Deaths
%
Births
%
Deaths
3640
Electric
Lighting
And
Wiring
Equipment
1,905
123
143
6.5%
7.5%

3650
Household
Audio
&
Video
Equip
766
96
87
12.5%
11.4%

3660
Communications
Equipment
1,794
169
159
9.4%
8.9%

3670
Electronic
Components
And
Accessories
6,068
614
522
10.1%
8.6%

3690
Misc.
Electrical
Equipment
&

Supplies
1,890
136
157
7.2%
8.3%

3710
Motor
Vehicles
And
Equipment
4,477
387
372
8.6%
8.3%

3720
Aircraft
And
Parts
1,633
122
127
7.5%
7.8%

3730
Ship
And
Boat
Building
And
Repairing
2,669
343
339
12.9%
12.7%

3740
Railroad
Equipment
189
15
15
7.9%
7.7%

3750
Motorcycles,
Bicycles,
&
Parts
256
38
25
14.8%
9.7%

3760
Guided
Missiles,
Space
Vehicles,

Parts
127
7
11
5.5%
8.4%

3790
Miscellaneous
Transportation
Equipment
962
106
109
11.0%
11.3%

3810
Search
&
Navigation
Equipment
758
34
60
4.5%
7.9%

3820
Measuring
And
Controlling
Devices
4,209
275
295
6.5%
7.0%

3840
Medical
Instruments
And
Supplies
3,770
334
289
8.9%
7.7%

3850
Ophthalmic
Goods
536
40
48
7.5%
8.9%

3860
Photographic
Equip
&
Supplies
784
71
72
9.1%
9.1%

3870
Watches,
Clocks,
Watchcases
&

Parts
159
12
20
7.5%
12.7%

3910
Jewelry,
Silverware,
And
Plated
Ware
2,606
246
275
9.4%
10.6%

3930
Musical
Instruments
434
46
35
10.6%
8.0%

3940
Toys
And
Sporting
Goods
2,843
384
345
13.5%
12.1%

3950
Pens,
Pencils,
Office,
&
Art
Supplies
975
62
70
6.4%
7.2%

3960
Costume
Jewelry
And
Notions
1,010
105
128
10.4%
12.7%

3990
Miscellaneous
Manufactures
7,338
784
740
10.7%
10.1%

TOTAL
136,653
11,103
10,698
8.1%
7.8%

N.
E.
C.
=
Not
Elsewhere
Classified
Source:
Small
Business
Administration,
Statistics
of
U.
S.
Businesses.

A­
27
MP&
M
EEBA:
Appendices
Appendix
A:
Detailed
Economic
Impact
Analysis
Information
A.
3
DESCRIPTION
OF
MP&
M
SURVEYS
EPA
used
two
screener
and
seven
detailed
questionnaires
(
surveys)
issued
between
1989
and
1996
to
collect
financial
and
technical
data
from
a
sample
of
facilities
that
were
evaluated
for
regulation
under
the
final
MP&
M
rule
(
see
Section
3
of
the
TDD).
The
responses
to
these
surveys
provided
the
basic
financial
and
economic
information
used
in
the
facility
and
firm
impact
analyses.
In
addition,
the
POT
W
Survey
provided
information
on
facility
permitting
costs
associated
with
regulatory
options
considered
by
EPA.
The
various
surveys
are
described
below
as
they
relate
to
the
financial
and
economic
analyses.

The
M
P&
M
rulemaking
docket
provides
copies
of
the
survey
instruments
and
detailed
information
on
the
conduct
of
the
surveys.

A.
3.1
Screener
Surveys
In
1990,
EPA
distributed
8,342
screener
surveys
to
sites
believed
to
be
engaged
in
the
original
seven
Phase
I
MP&
M
sectors.

In
1996,
EPA
distributed
5,325
screener
surveys
to
sites
believed
to
be
engaged
in
the
eleven
Phase
II
MP&
M
sectors.
The
screener
surveys
helped
EPA
to
identify
sites
to
receive
the
more
detailed
follow­
up
surveys
and
to
make
a
preliminary
assessment
of
the
MP
&
M
industry
evaluated
for
the
final
rule.
EPA
identified
the
SIC
codes
applicable
to
the
respective
MP&
M
sectors
evaluated
for
the
final
rule
and
randomly
selected
names
and
addresses
in
those
SICs
to
receive
the
screener
surveys
based
on
Dun
&
Bradstreet
databases.

A.
3.2
Ohio
Screener
Surveys
EPA
also
sent
the
1996
screener
survey
to
1,600
randomly
selected
sites
in
Ohio
to
support
the
Ohio
case
study.

A.
3.3
Detailed
MP&
M
Industry
Surveys
Based
on
responses
to
the
screener
surveys,
EPA
sent
a
more
detailed
survey
to
a
selected
group
of
water­
using
MP&
M
facilities
evaluated
for
the
final
rule.
EPA
collected
financial
and
technical
data
from
sample
facilities
in
two
phases.

Based
on
responses
to
the
1990
screener,
EPA
sent
the
Phase
I
detailed
survey
to
a
select
group
of
water­
using
facilities.
The
Agency
designed
this
survey
to
collect
detailed
technical
and
financial
information.
EPA
selected
1,020
detailed
survey
recipients
from
water­
discharging
screener
respondents,
water­
using
screener
respondents
that
did
not
discharge
process
water,
and
a
non­
randomly
selected
group
of
known
water­
discharging
facilities
that
did
not
receive
the
screener.

EPA
used
information
from
the
first
two
groups
of
survey
recipients
to
develop
pollutant
loadings
and
reductions
and
to
develop
compliance
cost
estimates.
Because
EPA
did
not
randomly
select
the
third
group
of
recipients,
EPA
did
not
use
the
data
to
develop
national
estimates.

To
reduce
burden
on
survey
recipients
for
Phase
II
of
the
data
collection
effort,
EPA
developed
two
similar
detailed
surveys.

Based
on
the
development
of
the
1995
MP&
M
proposal,
EPA
chose
to
collect
more
detailed
information
from
sites
with
annual
process
wastewater
discharges
greater
than
one
million
gallons
per
year
(
1
MG
Y).
EPA
sent
the
 
long 
detailed
survey
to
all
353
1996
screener
respondents
evaluated
for
the
final
rule
who
indicated
they
discharged
one
million
or
more
gallons
of
process
wastewater
annually
and
performed
MP&
M
operations.
The
Agency
sent
the
 
short 
detailed
survey
to
101
randomly
selected
1996
screener
respondents
evaluated
for
the
final
rule
who
indicated
they
discharged
less
than
one
million
gallons
of
process
wastewater
annually
and
performed
MP&
M
operations.

The
detailed
survey
responses
provide
financial,
economic,
and
employment
information
about
the
site
or
the
company
owning
the
facility.
In
addition,
the
1996
long
detailed
questionnaire
included
a
section
that
requested
supplemental
information
on
other
facilities
owned
by
the
company.
EPA
included
this
voluntary
section
to
measure
the
impact
of
the
final
MP&
M
effluent
guidelines
on
companies
with
multiple
facilities
that
discharge
process
wastewater.
This
section
requested
the
same
information
collected
in
the
1996
MP&
M
screener
survey.
Responses
to
questions
in
this
section
provided
information
on
the
size,
industrial
sector,
revenue,
unit
operations,
and
water
usage
of
the
company s
other
facilities.

The
1996
short
survey
included
the
identical
general
facility
and
economic
information
collected
in
the
long
detailed
survey,

with
one
exception.
Short
survey
recipients
were
not
asked
to
provide
information
on
the
liquidation
value
of
their
plant.

A­
28
MP&
M
EEBA:
Appendices
Appendix
A:
Detailed
Economic
Impact
Analysis
Information
A.
3.4
Iron
and
Steel
Survey
EPA
also
developed
a
detailed
survey,
under
a
separate
rulemaking
effort,
to
collect
detailed
information
from
facilities
covered
by
the
Iron
and
Steel
Manufacturing
effluent
guidelines
(
40
CFR
Part
420).
Following
field
sampling
of
iron
and
steel
sites
and
review
of
the
completed
industry
surveys,
EPA
decided
at
proposal
that
some
iron
and
steel
operations
would
be
more
appropriately
covered
by
the
MP&
M
rule
because
they
were
more
like
MP&
M
operations.
EPA
relied
on
the
Iron
&

Steel
survey
for
financial
and
economic
information
on
47
iron
and
steel
facilities.
Commenters
on
the
proposed
rule
stated
that
these
operations
and
resulting
wastewaters
are
comparable
to
those
at
facilities
subject
to
the
Iron
and
Steel
Manufacturing
effluent
guidelines
and
that
these
discharges
should
remain
subject
to
Part
420
rather
than
the
final
MP&
M
rule.
Also
at
NODA,
EPA
considered
including
in
the
Steel
Forming
and
Finishing
subcategory
wastewater
discharges
resulting
from
continuous
electroplating
of
flat
steel
products
(
e.
g.,
strip,
sheet,
and
plate).
EPA
also
relied
on
the
Iron
&

Steel
survey
for
financial
and
economic
information
on
these
24
iron
and
steel
facilities.
EPA
re­
examined
its
database
for
facilities
that
perform
continuous
steel
electroplating,
and
found
that,
contrary
to
its
initial
finding,
continuous
electroplaters
do
not
perform
operations
similar
to
other
facilities
in
this
subcategory
(
i.
e.,
steel
forming
and
finishing
facilities
performing
cold
forming
on
steel
wire,
rod,
bar,
pipe,
and
tube).
Thus,
EPA
included
continuous
electroplaters
performing
electroplating
and
coating
operations
in
the
General
Metals
subcategory
for
analyses
supporting
the
final
rule.
As
described
in
Section
VI
of
the
preamble
to
the
final
rule,
EPA
is
not
revising
limitations
or
standards
for
any
of
these
facilities.
Such
facilities
will
continue
to
be
regulated
by
the
General
Pretreatment
Standards
(
Part
403),
local
limits,
permit
limits,
and
Iron
&
Steel
effluent
limitations
guidelines
(
Part
420),
as
applicable.

A.
3.5
Municipality
Survey
EPA
distributed
surveys
in
1996
to
city
and
county
facilities
that
might
operate
facilities
engaged
in
MP&
M
operations
evaluated
for
the
final
rule.
The
Agency
designed
this
survey
to
measure
the
rule s
impact
on
municipalities
and
other
government
entities
that
perform
maintenance
and
rebuilding
operations
on
MP&
M
products
(
e.
g.,
bus
and
truck,

automobiles).
The
Agency
sent
the
municipality
survey
to
150
city
and
county
facilities
randomly
selected
from
the
Municipality
Year
Book­
1995
based
on
population
and
geographic
location.
EPA
allocated
sixty
percent
of
the
sample
to
municipalities
and
40
percent
to
counties.
The
60/
40
distribution
was
approximately
proportional
to
their
aggregate
populations
in
the
frame.
EPA
divided
the
municipality
sample
and
the
county
sample
into
three
size
groupings
as
measured
by
population.
The
surveys
collected
information
on
costs
of
service
and
on
the
financial
and
economic
characteristics
of
the
governments
operating
these
facilities.

A.
3.6
Federal
Facility
Survey
EPA
designed
this
survey
to
assess
the
impact
of
the
MP&
M
effluent
limitations
guidelines
and
standards
on
federal
agencies
that
operate
MP&
M
facilities.
EPA
distributed
the
survey
to
federal
agencies
likely
to
perform
industrial
operations
on
metal
products
or
machines.
The
Agency
requested
that
the
representatives
of
the
seven
chosen
federal
agencies
voluntarily
distribute
copies
of
the
survey
to
sites
they
believed
performed
MP&
M
operations.
The
information
collected
in
the
1996
federal
survey
was
identical
to
the
long
survey.
After
engineering
review
and
coding,
EPA
entered
data
from
44
federal
surveys
into
the
database.
Because
EPA
did
not
randomly
select
the
survey
recipients,
data
from
these
questionnaires
were
not
used
to
develop
national
estimates.

A.
3.7
POTW
Survey
EPA
distributed
the
Publicly­
Owned
Treatment
Works
(
POTW
)
survey
in
November
1997.
The
Agency
designed
this
survey
to
estimate
possible
costs
and
burden
that
POTWs
might
incur
in
administering
MP&
M
permits
or
other
control
instruments
and
to
estimate
benefits
from
implementation
of
the
options
considered
for
the
final
rule.
The
Agency
sent
the
POTW
Survey
to
150
POTWs
with
flow
rates
greater
than
0.50
million
gallons
per
day.
EPA
randomly
selected
the
recipients
from
the
1992
Needs
Survey
Review,
Update,
and
Query
System
Database
(
RUQus),
and
divided
the
POTW
sample
into
two
strata
by
daily
flow
rates:
0.50
to
2.50
million
gallons,
and
2.50
million
gallons
or
more.

In
addition
to
the
total
volume
of
wastewater
treated
at
the
site,
the
POTW
Survey
requested
the
number
of
industrial
permits
written,
the
cost
to
write
the
permits,
the
permitting
fee
structure,
the
percentage
of
industrial
dischargers
covered
by
National
Categorical
Standards
(
i.
e.,
effluent
guidelines),
and
the
percentage
of
permits
requiring
specific
administrative
activities.

EPA
used
this
information
to
estimate
administrative
burden
and
costs.
In
addition,
EPA
requested
information
on
the
use
or
disposal
of
sewage
sludge
generated
by
the
POTW.
The
Agency
only
required
POTWs
that
received
discharges
from
an
MP&
M
facility
to
complete
those
questions.
The
POTW
Survey
requested
the
following
sewage
sludge
information:
amount
A­
29
MP&
M
EEBA:
Appendices
Appendix
A:
Detailed
Economic
Impact
Analysis
Information
generated,
use
or
disposal
method,
metal
levels,
use
or
disposal
costs,
and
the
percentage
of
metal
loadings
from
MP&
M
facilities.
The
Agency
used
this
information
to
assess
the
potential
changes
in
sludge
handling
resulting
from
the
MP&
M
rule
and
to
estimate
economic
benefits
of
these
options
to
the
POTW.

A­
30
MP&
M
EEBA:
Appendices
Appendix
A:
Detailed
Economic
Impact
Analysis
Information
REFERENCES
U.
S.
Department
of
Commerce.
2000.
U.
S.
Bureau
of
the
Census.
The
Bridge
Between
NAICS
and
SIC
Report.
March.

http://
www.
census.
gov/
epcd/
www/
naicensu.
html
U.
S.
Department
of
Commerce.
2001.
U.
S.
Bureau
of
the
Census.
Manufacturing
Industry
Series:
Industry
Stats
on
NAICS
Basis
with
Distribution
Among
1987
SIC­
Based
Industries.
ECON97S
Report
Series
CD­
Rom.

Small
Business
Administration.
Statistics
of
U.
S.
Businesses.
http://
www.
sba.
gov/
advo/
stats/
int_
data.
html
A­
31
MP&
M
EEBA:
Appendices
Appendix
A:
Detailed
Economic
Impact
Analysis
Information
THIS
PAGE
INTENTIONALLY
LEFT
BLANK
A­
32
MP&
M
EEBA:
Appendices
Appendix
B:
Cost
Pass­
Through
Analysis
INTRODUCTION
This
appendix
presents
the
methodology
and
results
from
the
analysis
of
cost
pass­
through
(
CPT)
potential
for
19
MP&
M
sectors.
1
This
analysis
consists
of
two
parts:

1.
an
econometric
analysis
of
the
historical
relationship
of
output
prices
to
changes
in
input
costs,
and
2.
an
analysis
of
market
structure
characteristics.

These
two
analyses
together
provide
a
numerical
estimate
of
how
much
of
compliance­
related
cost
increases
a
sector
can
be
expected
to
pass
on
to
its
consumers.

The
rest
of
this
appendix
is
organized
into
the
following
six
sections:

 
B.
1:
Rationale
for
developing
sector­
specific
CPT
coefficients
as
opposed
to
firm­
specific
CPT
coefficients;

 
B.
2:
Econometric
analysis
of
CPT
potential,

based
on
the
historical
changes
in
output
prices
relative
to
changes
in
input
costs;

 
B.
3:
Analysis
of
the
market
structure
factors
expected
to
affect
cost
recovery;
Appendix
B:
Cost
Pass­
Through
Analysis
APPENDIX
CONTENTS
B.
1
hoice
of
Sector­
Specific
CPT
Coefficients
........
B­
1
B.
2
etric
Analysis
.................
............
B­
2
B.
2.1
ework
.................
...............
B­
3
B.
2.2
Used
to
Estimate
the
Regression
Equation
.................
.................
.
B­
4
B.
2.3
ession
Results
.................
.........
B­
6
B.
3
t
Structure
Analysis
.................
........
B­
9
B.
3.1
es
Descriptions
.................
......
B­
9
B.
3.2
s
.................
.................
B­
13
B.
4
dation
of
Econometrically­
Estimated
CPT
Coefficients
.................
.................
.
B­
16
B.
4.1
etal
Products
.................
......
B­
17
B.
4.2
.................
...............
B­
17
B.
4.3
le
.................
...........
B­
18
B.
4.4
raft
.................
................
B­
18
B.
4.5
trial
Equipment
................
B­
18
B.
4.6
.................
..............
B­
18
B.
5
.....
B­
18
Attachment
B.
A:
Selected
Review
of
CPT
Literature
.......
B­
20
B.
A.
1
Firm­
Specific
Cost
Pass­
Through
Rate. 
.........
B­
20
B.
A.
2
ge
Rate
Pass­
Through
................
B­
20
B.
A.
3
ax
Pass­
Through
.................
.......
B­
20
B.
A.
4
.................
............
B­
20
Acronyms
.................
.................
.......
B­
22
The
C
Econom
Fram
Data
Regr
Marke
Measur
Result
Vali
Other
M
Job
Shops
Motor
Vehic
Airc
Mobile
Indus
Aerospace
Adjusting
Estimates
of
Compliance
CPT
Potential
Ashenfelter
et
al.
(
1998),
 
Identifying
the
Exchan
T
Studies
Cited
 
B.
4:
Validation
of
econometric
estimates
of
the
CPT
coefficients;

 
B.
5:
Adjustment
of
estimated
CPT
coefficients
to
reflect
the
portion
of
an
MP&
M
sector
that
will
incur
compliance
costs;
and
 
B.
6:
Attachment:
Findings
from
a
review
of
the
CPT
literature.

B.
1
THE
CHOICE
OF
SECTOR­
SPECIFIC
CPT
COEFFICIENTS
EPA
believes
the
use
of
sector­
specific
CPT
coefficients
instead
of
firm­
specific
CPT
coefficients
in
the
impact
analysis
is
an
appropriate
and
practical
way
of
analyzing
compliance
CPT.
The
sector­
wide
rate
provides
an
estimate
of
the
change
in
each
facility s
output
prices
as
a
function
of
the
regulation­
induced
increase
in
its
production
costs,
assuming
that
the
same
cost
increase
is
experienced
by
all
establishments
competing
with
the
facilities
in
question.
For
MP&
M
sectors
in
which
a
large
fraction
of
establishments
will
be
affected
by
the
regulation,
it
is
reasonable
to
assume
that
the
MP&
M
compliance
cost
acts
1
The
analysis
of
cost
pass­
through
potential
presented
here
refines
in
several
places
the
methodology
developed
for
the
Phase
I
MP&
M
analysis.
These
refinements
are
highlighted
at
the
appropriate
stages
of
the
discussion
that
follows.

B­
1
MP&
M
EEBA:
Appendices
Appendix
B:
Cost
Pass­
Through
Analysis
like
an
industry­
wide
cost
shock.
As
noted
below
in
section
five,
EPA
applies
an
additional
adjustment
to
the
estimated
CPT
rate
to
reflect
the
fraction
of
total
sector
output
that
is
estimated
to
incur
regulation­
induced
production
cost
increases.

In
contrast
to
the
concept
of
a
sector­
specific
CPT
adjustment,
a
firm­
specific
CPT
rate
relates
a
change
in
the
prices
charged
by
a
specific
firm
to
a
change
in
its
production
costs,
assuming
no
change
in
the
production
cost
for
rival
producers
of
that
product.
Not
surprisingly,
previous
studies
have
found
that
the
CPT
rate
for
changes
on
an
individual
firm s
costs
differs
from
the
rate
at
which
a
firm
would
pass
through
cost
changes
that
are
common
to
all,
or
a
substantial
fraction
of,
firms
in
an
industry
(
e.
g.,
Ashenfelter
et
al.,
1998).
It
is
true,
however,
that
firms
in
an
industry
will
have
differing
CPT
among
each
other
to
some
extent
for
reasons
such
as,
differentiated
products
(
e.
g.,
products
of
different
firms
are
not
commodities
and
are
not
perfectly
substitutable);
imperfectly
competitive
markets
(
e.
g.,
markets
in
which
individual
firms
possess
different
degrees
of
market
power);
and
segmented
markets
(
e.
g.,
geographically
segmented
markets).
In
the
presence
of
such
imperfections,

individual
firms
will
very
likely
respond
differently
in
their
ability
to
pass
on
cost
increases
in
higher
output
prices
even
when
the
production
cost
increase
applies
to
all,
or
a
substantial
fraction,
of
an
industry s
production.
Nonetheless,
estimating
the
CPT
ability
of
individual
firms
or
sub­
sector
groups
of
firms
within
the
MP&
M
sectors
would
require
a
detailed
analysis
of
market
segments
and
substitutability
of
MP&
M
products.
While
this
effort
may
be
theoretically
possible,
it
would
be
highly
expensive
and
an
overall
daunting
challenge
given
the
breadth
of
the
MP&
M
industry
sectors.

Therefore,
this
analysis
of
CPT
potential
in
the
MP&
M
industry
is
undertaken
at
the
sector­
specific
level
under
the
assumption
of
perfect
competition
in
these
sectors
­­
including
product
homogeneity
(
i.
e.,
products
produced
by
one
firm
are
perfect
substitutes
for
products
produced
by
other
firms),
and
homogeneity
of
production
technology
and
cost
across
firms
(
i.
e.,
pricing
is
at
marginal
cost).
2
Under
these
conditions,
the
price
response
to
a
general
industry­
wide
change
in
production
costs
is
likely
to
be
industry­
wide
and
similar
across
all
firms.

B.
2
ECONOMETRIC
ANALYSIS
EPA
performed
an
econometric
analysis
of
input
costs
and
output
prices
to
estimate
historical
CPT
elasticities
for
18
of
the
19
Phase
I
and
Phase
II
MP&
M
Sectors.
EPA
could
not
estimate
historical
CPT
coefficients
for
Aerospace
due
to
data
limitations.
These
elasticities
indicate
the
changes
in
output
prices
by
sector
that
have
occurred
historically
in
relation
to
changes
in
the
cost
of
production
inputs.
Two
factors
determine
the
share
of
a
cost
increase
that
a
facility
can
pass
through
to
its
customers:
the
elasticity
of
demand
and
the
elasticity
of
supply
in
the
facility's
market.
Both
factors
are
difficult
to
measure
accurately;
among
other
reasons,
observed
changes
in
price
are
due
to
simultaneous
changes
in
demand
and
supply.

In
view
of
this
difficulty,
this
pass­
through
analysis
does
not
decompose
cost
pass­
through
into
the
separate
effects
stemming
from
elasticity
of
demand
and
elasticity
of
supply.

An
additional
analytic
challenge
involves
joint
consideration
of
quantity
and
price
effects.
Specifically,
the
amount
of
cost
increase
that
a
firm
may
recover
through
a
revenue
increase
may
generally
be
decomposed
into
a
change
in
price
and
a
change
in
quantity
sold.
In
most
markets,
increased
prices
(
in
response
to
increased
costs)
translate
into
reduced
quantity
of
sales.

The
interaction
of
supply
and
demand
elasticities
determines
whether
or
not
total
revenue
increases.

For
practical
reasons,
this
analysis
focused
on
the
change
in
equilibrium
price
due
to
a
change
in
input
costs
and
furthere
assumes
that
the
sale
quantities
of
businesses
complying
with
the
regulation
do
not
change.
The
analysis
determined
changes
in
market
quantities
from
closures
rather
than
by
estimating
output
changes
in
non­
closing
facilities.
The
analysis
assumed
that
the
quantity
of
shipments
or
sales
does
not
vary
with
the
increase
in
fixed
and
average
costs
unless
the
facility
closes.

The
following
grounds
support
this
restriction:

 
The
cost
model
for
the
individual
facility
reflects
a
constant
marginal
cost
relationship.
The
change
in
quantity
of
output
at
a
facility
is
a
function
of
the
change
in
equilibrium
price
and
the
marginal
cost
relationship
at
the
facility.
For
instance,
in
the
case
in
which
marginal
cost
increases
with
output,
an
upward
shift
in
the
marginal
cost
relationship
due
to
compliance
costs
will
generally
cause
a
facility
to
reduce
its
production
quantity.
The
extent
of
changes
in
production
quantity
will
vary
across
facilities
based
on
the
shift
in
marginal
cost
and
the
rate
at
which
marginal
cost
changes
with
production.
Engineering
analysis
of
facilities
provides
no
information,
however,
about
any
change
in
the
marginal
cost
relationship
for
a
given
facility,
providing
only
lump­
sum
costs.
In
lieu
of
this
information,
the
analysis
uses
constant
marginal
costs,
which
in
turn
means
that
2
These
assumptions
likely
approximate
the
real
world
for
those
MP&
M
sectors
that
consist
of
a
large
number
of
small,
highly
competitive
firms
such
as
Job
Shops
or
Printed
Wiring
Boards.

B­
2
MP&
M
EEBA:
Appendices
Appendix
B:
Cost
Pass­
Through
Analysis
profit­
seeking
facilities
will
tend
not
to
change
their
output
quantities
in
response
to
added
costs
resulting
from
regulation.
As
a
result,
the
only
quantity­
related
decision
that
can
be
meaningfully
analyzed
at
the
facility
level
is
whethe
r
to
term
inate
production
com
pletely.

 
An
estimate
of
quantity
response
would
be
based
on
the
aggregate
industry
response
and
would
not
be
logically
ap
plicable
to
the
facility­
level
analysis.
An
an
alysis
can
estimate
quan
tity
elasticity
resp
onse
to
changes
in
input
costs,
but
this
value
would
represent
the
aggregate
quantity
response
in
the
particular
MP&
M
sector.
The
aggregate
response
encompasses
a
diversity
of
responses
across
facilities:
a
few
facilities
may
eliminate
production
entirely
while
others
may
reduce,
keep
the
same,
or
even
increase
output.
Applying
the
aggregate
quantity
response
to
individual
facilities
while
simultaneously
allowing
for
terminated
production
would
exaggerate
the
likely
facility­
level
quantity
response
and
the
likelihood
of
facility
closures.
The
current
analysis
simulates
the
aggregate
response
from
a
micro­
analytic
perspective:
exiting
facilities
that
found
compliance
to
be
an
uneconomic
proposition
affect
the
industry­
wide
quantity
response.

B.
2.1
The
analysis
measured
the
sensitivity
of
equilibrium
prices
to
changes
in
input
costs.
t
elasticity
of
price, 
denoted
Ep,
measured
the
percentag
e
change
in
o
utput
price
pe
r
percent
cha
nge
in
unit
input
costs.
3
EPA
estim
ated
the
cost
elasticity
of
price
by
regressing
annual
output
price
indices
on
annual
input
price
indices.
The
methodology s
direct
estimation
measured
actual
changes
in
output
price
with
respect
to
changes
in
input
costs.
ook
into
account
the
full
range
of
possible
mechanisms
by
which
input
costs
affect
output
prices,
including
technical
changes,
substitution,
non­
competitive
pricing
mechanisms,
imperfect
information
phenome
na,
and
any
other
shifts
or
irregularities
in
the
supply
and
demand
functions.

Th
e
19
M
P&
M
industry
se
ctors
e
ncompa
ss
224
industrial
4­
d
igit
SIC
cod
es.
A,
ho
weve
r,
could
estim
ate
the
c
ost
elastic
ity
of
price
based
on
historical
data
for
only
170
manufacturing
SIC
codes.
EPA
could
not
estimate
the
cost
elasticity
of
price
for
Aerospace
and
non­
manufacturing
industries
due
to
data
limitations,
but
assigned
a
CPT
coefficient
to
the
aerospace
sector
based
on
the
market
structure
analysis
(
see
Section
2
for
details,
below).
4
EPA
assumed
zero
CP
T
for
non­
manufacturing
industries
because
these
industries
tend
to
be
very
competitive.

For
each
MP&
M
sector,
EPA
estimated
a
relationship
for
the
k
=
1
to10
yearly
observations
(
from
1987
to
1996)
by
least­
squares
linear
regression,
as
follow
s:

(
B­
1)

where
:

Pout,
k
=
price
index
for
the
bundle
of
goods
produced
by
the
MP
&
M
sector,
year
k;

Ep
=
elasticity
of
output
price
with
respect
to
input
costs
for
a
given
MP&
M
sector;

Pin,
k­
1
=
price
index
of
inputs
(
labor
and
non­
labor)
to
a
given
sector,
year
k­
1;

b
=
elasticity
of
output
price
with
respect
to
em
ploymen
t
costs;

 
=
error
term;
and
ln(
x)
=
natural
log
of
x
Specifying
the
key
regression
variables
as
logarithms
permitted
EPA
to
estimate
the
elasticities
of
output
prices
with
respect
to
the
ind
ependent
variab
les
directly.
at
is:
Framework
The
 
cos
This
practice
t
EP
Th
3
The
elasticity
measure
also
applies
to
revenue
because
quantity
of
production
is
assumed
constant.

4
Output
Price
Index
data
for
the
Aerospace
sector
were
unavailable.
EPA
attempted
to
use
proxy
data
for
missile
manufacturing,,
a
component
of
the
defense
sector,
to
estimate
a
CPT
coefficient
for
the
Aerospace
sector.
This
analysis
did
not
produce
meaningful
results.
The
missile
manufacturing
industry
witnessed
a
sharp
decline
in
producer
prices
during
the
1987­
1996
time
period,
therefore
yielding
a
negative
CPT
coefficient
for
the
Aerospace
sector.
Since
the
Aerospace
sector
and
the
missile
manufacturing
industry
are
sufficiently
different
from
each
other,
EPA
decided
not
to
use
the
estimated
CPT
coefficient
and
instead
derive
a
coefficient
for
the
Aerospace
sector
based
solely
on
the
market
structure
analysis.

B­
3
MP&
M
EEBA:
Appendices
Appendix
B:
Cost
Pass­
Through
Analysis
(
B­
2)

which
is
the
elasticity
of
output
price
with
respect
to
input
cost
changes
in
the
previous
year.

EP
A s
use
of
the
logarithm
ic
transfo
rmatio
ns
also
eliminated
an
y
linear
tren
d
ov
er
time;
in
effect,
the
ind
ividua
l
yearly
observatio
ns
becom
e
cross­
sectional
variables.
he
mo
del
therefore
required
no
specific
time­
serie
s
structure
.

EPA
considered
additional
independent
variables
that
might
aid
in
explaining
output
price
changes.
or
example,
EPA
included
some
measures
of
aggregate
income,
but
these
measures
did
not
contribute
significantly
to
the
estimated
relationships.

The
coefficients
Ep
from
this
regression
are
the
estimated
cost­
elasticities
of
price
for
each
MP&
M
sector.
imated
coefficients
address
the
question:
over
the
period
of
analysis,
by
how
much
did
output
prices
change
as
input
costs
increased?

Th
e
value
of
Ep
for
each
secto
r,
linked
with
other
information
o
n
market
structure,
yielded
a
comp
osite
measure
of
cost
pass­
through
potential
by
M
P&
M
sector.
ussed
below,
EP
A
used
the
re
sults
of
the
m
arket
stru
cture
analysis
to
valida
te
the
estim
ated
v
alues
o
f
Ep,
which
represent
the
expected
CPT
potential
for
the
different
MP&
M
sectors.

validated
Ep
values
are
the
CP
T
co
efficients
ultimately
assigned
to
sec
tors
for
the
econo
mic/
financial
impact
ana
lysis.

B.
2.2
ata
Used
to
Estimate
the
Regression
Equation
Estimating
Ep
required
a
measure
of
the
change
over
time
in
input
costs
and
a
measure
of
the
change
in
output
price
for
each
MP
&
M
sector.
akes
time
to
respond
to
price
changes
(
i.
e.,
input
prices
from
1988
would
pred
ict
output
prices
in
1989).
udies
found
the
lags
associate
with
price
pass­
through
can
extend
from
5
to
8
quarters
(
J.
Menon,
1995).
a
on
changes
of
annual
output
price
indices
from
1987
to
1996
and
input
price
indices
from
1986
to
1995.
The
final
data
set
contains
ten
years
of
data
for
each
of
the
18
industrial
sectors
of
concern.
ysis
estimated
the
relationship
between
change
in
output
price
index
(
dependent
variable)
and
change
in
input
cost
index.
nput
cost
index
(
independent
variables)
combines
a
wide
range
of
no
n­
labor
co
st
values,
including
energy,
with
emp
loyment
cost
value
s.

a.

The
dependent
variable
is
the
output
price
index.
The
Producer
Price
Index
(
PPI),
an
appropriate
measure
of
output
price,

me
as
ur
es
ch
an
ge
s
in
the
p
rice
that
the
p
ro
du
ce
r
rec
eive
s
at
th
e
p
lant
ga
te
an
d
is
th
er
efor
e
the
releva
nt
pr
ice
for
the
p
ro
du
ce
r's
production
decisions.
ducts
are
often
intermediate
goods
whose
market
prices
are
producer
prices.
A
estimated
the
dep
endent
variable
a
s
the
weighted
avera
ge
of
P
PIs
fo
r
the
go
ods
produced
b
y
the
indu
stries
in
each
sector.

EP
A
calculated
the
output
price
ind
ex
for
the
sectors
as
follow
s:

(
B­
3)

where
:

Pout,
k
=
average
output
price
index
value
for
a
given
MP
&
M
sector
in
year
k;

qi,
k
=
value
o
f
shipments
for
S
IC
ind
ustry
i,
year
k;
and
PPIi,
k
=
Pro
ducer
Price
Ind
ex
for
the
outp
ut
of
SIC
industry
i,
year
k.

EPA
used
the
following
information
to
fill
in
data
gaps
for
all
output
prices
when
the
PPI
series
had
missing
data:

 
Inform
ation
at
the
3­
digit
SIC
cod
e
level
if
data
were
unavailable
at
the
4­
digit
SIC
cod
e
level;

 
Th
e
per
centage
cha
nge
in
p
rice
at
the
3­
digit
lev
el,
app
lied
to
the
4­
digit
level
to
calculate
missing
values,
if
da
ta
at
the
4­
digit
level
were
available
for
several
years;
and
T
F
The
est
As
disc
The
D
EPA
lagged
output
prices
by
a
year
because
the
market
t
For
example,
exchange
rate
pass­
through
st
EPA
used
dat
The
anal
The
i
Dependent
variable
MP&
M
proEP
B­
4
MP&
M
EEBA:
Appendices
Appendix
B:
Cost
Pass­
Through
Analysis
 
A
best­
fit
line
to
extrapolate
data
for
years
with
missing
data
when
at
least
five
years
worth
of
data
were
available.

b.

The
independ
ent
variable
is
the
input
cost
index.
The
input
cost
index
averages
the
producer
price
index
values
for
commo
dity
inputs
to
the
sector
in
question,
weighted
by
the
share
of
each
input
to
sector
output.
The
weighted
average
calculation
involves
two
steps:
(
1)
estimating
input
cost
indice
s
at
the
4­
digit
SIC
level
and
(
2)
deve
loping
the
input
co
st
index
at
the
MP&
M
sector
level.
eps
are
discussed
in
detail
below.

 
Estimating
Input
Cost
Indices
at
the
4­
digit
SIC
level
EPA
first
identified
the
composition
of
production
inputs
required
to
produce
output
from
a
given
industry
by
obtaining
direct
requirement
coefficients
from
the
1992
Benchm
ark
Input­
Output
Tables
of
the
United
States.
5
The
direct
requirement
coefficients
are
defined
as
fo
llows:
for
each
dollar
of
ou
tput
from
industry
i,
the
dire
ct
requirem
ents
co
efficient
rj
indicates
the
value
of
inp
ut
j
required
to
achiev
e
one
dollar
of
ou
tput
from
industry
i.
The
sum
of
all
requiremen
ts
coefficients
rj
for
industry
i
equals
one.
Note
that
the
direct
requirements
coefficients
from
the
input­
output
table
include
information
on
the
purchase
of
capital
goods.
Changes
in
the
cost
of
capital
goods
are
therefore
reflected
in
the
PPI
series
for
the
associated
industrie
s.
cause
only
on
e
set
of
d
irect
req
uirements
co
efficients
we
re
ava
ilable
for
and
are
use
d
in
the
a
nalysis,
this
analysis
assumes
that
the
input
mix
rem
ains
constant
ove
r
the
ten­
year
period
considere
d
in
the
analysis.

EPA
then
used
yearly
PPI
values
and
the
Employm
ent
Cost
Index
(
EC
I)
from
the
Bureau
of
Lab
or
Sta
tistics
to
estim
ate
changes
in
the
labor
and
non­
labor
com
ponents
of
production
cost
over
time.

to
estimate
changes
in
labor
cost
for
all
sectors
except
for
aircraft
manufacturing,
for
which
a
sector­
specific
ECI
is
available.

EPA
calculated
the
input
cost
index
for
a
4­
digit
SIC
group
as
a
weighted
average
of
prices
for
(
a)
all
non­
labor
inputs
for
which
the
PPI
series
data
were
available
and
(
b)
labor
input.
The
percentage
of
inputs
accounted
for
in
our
regression
model
range
s
from
39
p
ercen
t
to
10
0
pe
rcent,
w
ith
an
average
of
66
perc
ent.

To
summ
arize,
E
PA
calcula
ted
the
input
co
st
index
as
follow
s.
ach
4
­
digit
SIC
industry,
i,
that
uses
non­
lab
or
inp
uts,
j,

the
average
input
price
for
the
year
k
is:

(
B­
4)

where
:

Pi,
k
=
avera
ge
inp
ut
price
index
for
SIC
industry
i,
year
k;

rj
=
direc
t
requireme
nts
coe
fficient
for
inp
ut
com
mod
ity
j
by
indu
stry
i;
and
PPIj,
k
=
Pro
ducer
Price
Ind
ex,
co
mmodity
j,
year
k.

rl
=
direct
requirements
coe
fficient
for
wages
and
salaries
by
industry
i;
and
ECIk
=
Employement
Co
st
Index
in
year
k.

 
Developing
the
input
cost
index
at
the
MP&
M
sector
level
EPA
developed
the
input
cost
index
at
the
MP&
M
sector
level
by
weighting
the
individual
4­
digit
SIC
group
cost
index
values
by
4­
digit
SIC
value
of
shipments
from
the
Census
of
Manufactures
and
various
Annual
Surveys
of
Manufactures
for
the
co
rresp
ond
ing
years.
is
analysis
assume
s
that
weig
hts
by
p
rod
uction
value
a
re
co
nstants
o
ver
time
.

The
resulting
values
provided
an
aggregate
measure
of
input
costs
over
the
ten­
year
period
1986­
1995
for
each
MP&
M
sector.
&
M
industry
sector,
containing
N
4­
digit
SIC
industries,
the
average
input
price
in
each
year
k
is:
Independent
variables
These
st
Be
The
Agency
used
ECI
for
private
manufacturers
For
e
Th
For
each
MP
5
The
Bureau
of
Economic
Analysis'
Input­
Output
Table
uses
its
own
industry
classification
system,
which
is
similar
to
the
Standard
Industrial
Classification
(
SIC)
used
in
the
Economic
Censuses.
This
discussion
refers
to
that
classification
system
as
the
BEA
classification.
Although
the
BEA
classification
has
more
categories
than
the
SIC
system,
EPA
grouped
and
mapped
the
BEA
classification
codes
to
the
more
aggregate
SIC
codes
that
form
the
MP&
M
sectors.
EPA
calculated
an
average
price
when
one
BEA
input
classification
code
corresponded
to
more
than
one
SIC
code.

B­
5
MP&
M
EEBA:
Appendices
Appendix
B:
Cost
Pass­
Through
Analysis
(
B­
5)

where:

Pin,
k
=
average
input
price
index
value
for
a
given
MP&
M
sector
in
year
k;

Pi,
k
=
input
price
index
value
for
SIC
industry
i,
year
k
;
and
qi,
k
=
value
of
shipments
for
SIC
industry
i,
year
k.

B.
2.3
Regression
Results
Table
B.
1
below
gives
the
estimated
parameter
values
(
corrected
for
autocorrelation)
and
t­
statistics
for
each
of
the
sectors.

Most
of
the
estimated
parameters
have
the
expected
sign
and
are
statistically
significant
at
95th
percentile.
The
estimated
parameters
show
that
16
of
the
18
MP&
M
sectors
have
been
able
to
increase
prices,
at
the
margin,
between
42
percent
and
121
percent
for
every
one
percent
increase
in
non­
labor
input
costs.
The
estimated
input
cost
coefficients
are
negative
for
two
industrial
sectors:
Printed
Circuit
Boards
and
Office
Machines.
This
finding
suggest
that
additional
market
factors
such
strong
domestic
and
global
competition
drive
output
prices
down.

Figure
B.
1
below
depicts
output
price
and
input
cost
trends
from
1987
to
1996
for
these
two
industries.
It
shows
that
in
both
sectors,
output
prices
decreased
faster
than
input
costs.
This
difference
indicates
that
significant
competition
in
these
sectors
drives
output
prices
down,
undoubtedly
through
rapid
technology
innovation.
An
inverse
relationship
between
labor
cost
and
output
prices
also
indicates
presence
of
strong
competition
in
these
two
sectors.
Based
on
these
findings,
it
is
reasonable
to
assume
that
the
printed
circuit
board
and
office
machine
sectors
have
zero
CPT
ability.

B­
6
MP&
M
EEBA:
Appendices
Appendix
B:
Cost
Pass­
Through
Analysis
Table
B.
1:
CPT
Regression
Results
By
Sector
MP&
M
Sector
Regression
Coefficients
(
t­
statistics
in
parenthesis)

Phase
1
Proposed
Rule
(
1982
to
1991)
Phase
2
Model
(
1987
to
1996)

Non­
Labor
Input
Costs
Labor
Input
Costs
Intercept
Total
Input
Costs
(
Labor+
Non­
Labor)

Aerospace
.774
(
12.73)
.001
(
4.21)
N/
A
N/
A
Aircraft
.924
(
37.22)
.003
(
3.32)
­
0.9280
(­
1.45)
1.20
(
8.90)

Bus
&
Truck
.930
(
30.91)
.003
(
2.46)
0.629
(
1.00)
0.864
(
6.52)

Electronic
Equipment
.899
(
25.28)
.005
(
3.46)
2.79
(
4.06)
0.395
(
2.72)

Hardware
.889
(
27.02)
.005
(
3.68)
1.06
(
1.80)
0.772
(
6.22)

Household
Equipment
.921
(
43.03)
.003
(
4.16)
1.69
(
2.91)
0.636
(
5.22)

Instruments
.923
(
46.44)
.003
(
4.34)
1.06
(
1.79)
0.771
(
6.18)

Iron
and
Steel
N/
A
N/
A
1.12
(
1.57)
0.767
(
5.14)

Job
Shop
N/
A
N/
A
1.97
(
3.33)
0.575
(
4.61)

Mobile
Industrial
Equipment
.901
(
23.94)
.004
(
2.68)
0.546
(
0.92)
0.884
(
7.05)

Motor
Vehicle
.898
(
27.85)
.004
(
3.36)
0.833
(
1.03)
0.820
(
4.76)

Office
Machines
.920
(
35.05)
.004
(
3.52)
47.5
(
17.2)
­
9.33
(­
15.6)

Ordnance
.907
(
29.05)
.004
(
3.18)
1.89
(
3.63)
0.591
(
5.41)

Other
Metal
Products
N/
A
N/
A
1.71
(
3.04)
0.631
(
5.34)

Precious
Metals
&
Jewelry
.938
(
24.82)
.002
(
1.68)
1.69
(
2.47)
0.640
(
4.42)

Printed
Circuit
Boards
n/
a
n/
a
6.23
(
9.07)
­
0.337
(­
2.31)

Railroad
.911
(
30.52)
.004
(
3.23)
0.548
(
0.914)
0.881
(
6.98)

Ships
and
Boats
.970
(
34.68)
.001
(
0.93)
0.817
(
1.53)
0.823
(
7.32)

Stationary
Industrial
Equipment
.909
(
28.09)
.004
(
3.06)
0.973
(
1.78)
0.791
(
6.88)

N/
A
=
Not
available
from
the
Phase
I
analysis.

Source:
U.
S.
EPA
analysis
B­
7
MP&
M
EEBA:
Appendices
Appendix
B:
Cost
Pass­
Through
Analysis
Table
B.
2:
Output
Prices
and
Unit
Input
Cost
Trends
in
the
Printed
Circuit
Board
and
Office
Machine
Sectors
Source:
EPA
Analysis.

Table
B.
1
also
presents
Phase
1
results
for
comparison.
Note
the
following
differences
in
the
Phase
1
and
Phase
2
analyses:

1.
Time
period:

 
Phase
1
analysis
covers
1982
to
1991;

 
Phase
2
analysis
covers
1987
to
1996.

2.
Explanatory
variables:

 
Phase
1
analysis
included
non­
labor
and
labor
cost
variables
separately.
The
model
has
no
intercept
term.
Note
that
EPA
then
used
only
the
non­
labor
input
cost
coefficient
to
estimate
a
CPT
potential
for
a
given
sector;

 
Phase
2
analysis
combines
labor
and
non­
labor
input
costs
because
compliance
costs
are
associated
with
both.

The
intercept
term
captures
additional
market
trends
(
e.
g.,
increased
import
penetration)
not
reflected
in
the
input
cost
indices.

3.
Industrial
sectors:

 
Phase
1
analysis
included
15
industrial
sectors.
It
excluded
iron
and
steel,
job
shops,
other
metal
products,
and
printed
circuit
boards
industries;

 
Phase
2
analysis
includes
18
of
the
19
industrial
sectors
and
excludes
the
aerospace
industry.
The
Phase
1
analysis
included
aerospace,
but
EPA
used
proxies
from
the
aircraft
industries
to
estimate
output
price
indices
for
the
aerospace­
related
4­
digit
SICs.
EPA
now
estimates
the
CPT
potential
for
this
sector
based
on
the
market
structure
analysis
alone.

EPA
assigned
MP&
M
sectors
to
low,
average,
and
high
CPT
categories
based
on
the
natural
breaks
in
the
estimated
parameter
values.
The
estimated
parameter
values
exhibit
two
distinct
breaks
in
their
distribution,
between
Precious
Metals
and
Jewelry
(
65.89
percent)
and
Hardware
(
78.17
percent)
and
between
Motor
Vehicle
(
82.45
percent)
and
Railroad
(
88.49
percent).
EPA
added
the
Aerospace
sector
to
the
high
CPT
category
based
on
results
from
the
market
structure
analysis.

Table
B.
3
summarizes
results
from
this
analysis.

B­
8
MP&
M
EEBA:
Appendices
Appendix
B:
Cost
Pass­
Through
Analysis
Table
B.
3:

Low
CPT
Average
CPT
High
CPT
Office
Machine
Hardware
Railroad
Printed
Circuit
Boards
Instruments
Mobile
Industrial
Equipment
Electronic
Equipment
Iron
&
Steel
Bus
&
Truck
Job
Shop
Stationary
Industrial
Equipment
Aircraft
Ordnance
Ships
&
Boats
Aerospacea
Other
Metal
Products
Motor
Vehicle
Household
Equipment
Precious
Metals
&
Jewelry
Classification
of
MP&
M
Sectors
by
CPT
Ability
a
Aerospace
assigned
to
High
category
based
on
results
from
the
market
structure
analysis
(
discussed
in
the
next
section).

Source:
U.
S.
EPA
analysis
B.
3
MARKET
STRUCTURE
ANALYSIS
The
second
part
of
the
analysis
of
cost
pass­
through
potential
is
based
on
an
analysis
of
the
current
market
structure
of
the
MP
&
M
industry
sectors.
Information
on
the
competitive
structure
and
market
characteristics
of
an
industry
provide
insight
into
the
likely
ranges
of
supply
and
demand
elasticities
and
the
sensitivity
of
output
prices
to
input
costs.
For
example,
when
input
costs
increase,
the
profit­
maximizing
firm
attempts
to
maintain
its
profits
by
increasing
output
prices
accordingly.
The
amount
of
the
cost
increase
that
the
firm
can
pass
on
as
higher
prices
depends
on
the
relative
market
power
of
the
firm
and
its
customers.
The
market
structure
analysis
described
in
this
section
attempts
to
measure
the
relative
market
power
enjoyed
by
firms
in
each
MP
&
M
sector
and
provides
ordinal
rankings
used
to
validate
the
CPT
coefficients
estimated
by
the
econometric
analysis.
The
analysis
represents
the
current
market
structure
and
CPT
ability
of
firms
in
the
MP&
M
sectors
and
in
no
way
attempts
to
forecast
the
future
market
structure
of
these
sectors.

B.
3.1
Measures
Descriptions
The
following
discussion
describes
five
indicators
of
market
power
used
to
assess
cost
pass­
through
potential
for
the
19
MP&
M
sectors.
Only
manufacturing
firms
have
been
considered;
non­
manufacturing
firms
have
been
excluded
due
to
data
limitations.
As
noted
above,
EPA
assigned
zero
CPT
ability
to
non­
manufacturing
firms.
The
five
indicators
of
market
power
analyzed
are
:
the
eight­
firm
concentration
ratio,
import
competition,
export
competition,
long
term
growth,
and
competition
barriers.
Each
of
these
factors
are
discussed
in
detail
below.

a.
Concentration
The
extent
of
concentration
among
a
group
of
market
participants
is
an
important
determinant
of
that
group's
market
power.

A
group
of
many
small
firms
typically
has
less
market
power
than
a
group
of
a
few
large
firms,
because
the
latter
are
in
a
more
advantageous
position
to
collude
with
each
other.
All
else
being
equal,
highly­
concentrated
industries
are
therefore
6
expected
to
pass­
through
a
higher
proportion
of
the
compliance
costs
that
will
result
from
this
regulation.

This
analysis
uses
the
eight­
firm
concentration
ratio,
which
measures
the
percentage
of
the
value
of
shipments
concentrated
in
the
top
eight
firms
in
each
four­
digit
SIC
category,
as
an
indicator
of
market
concentration.
The
analysis
estimates
sector
concentration
ratios
as
the
weighted
averages
of
component
industry
concentration
ratios,
weighted
by
SIC
value
of
shipments.
7
An
increase
in
the
sector
concentration
ratio
makes
firms
in
an
industry
better
able
to
pass
on
larger
portions
of
their
input
cost
increases
without
adversely
affecting
quantities
sold
to
a
significant
extent.

6
A
substantial
body
of
empirical
research
exists
that
has
addressed
the
relationship
between
industry
concentration
and
market
power.
Eg.,
see
Waldman
&
Jensen,
1997.

7
The
eight­
firm
concentration
ratio
and
value
of
shipments
data
used
are
for
the
year
1992.

B­
9
MP&
M
EEBA:
Appendices
Appendix
B:
Cost
Pass­
Through
Analysis
This
analysis
is
potentially
limited
by
the
necessity
to
aggregate
component
industries
into
sectors.
The
accuracy
of
any
analysis
to
characterize
market
power
originating
from
industry
concentration
depends
to
a
great
extent
on
defining
the
relevant
market.
A
well­
defined
market
requires
including
all
competitors
and
excluding
all
non­
competitors.
Defining
the
relevant
market
too
narrowly
overstates
market
power,
while
defining
the
market
too
broadly
would
underestimate
it.

Aggregating
concentration
ratios
for
the
four­
digit
SIC
categories
into
a
sector
concentration
ratio
results
in
a
sector
average
that
may
overstate
market
power
for
some
portions
of
the
sector
and
understate
market
power
for
other
portions.
This
analysis
would
likely
estimate
concentration
ratios
for
markets
that
in
general
are
too
broadly­
defined.
8
Even
so,
the
sectoral
concentration
ratios
estimated
should
provide
meaningful
information
that
will
assist
in
determining
relative
market
power
for
each
sector,
because
firms
producing
similar
or
related
products
are
still
classified
within
the
same
sector
and
each
sector
produces
a
distinctly
different
family
of
products
(
e.
g.,
motor
vehicles,
aircrafts,
ships
and
boats).

Another
important
determinant
of
the
relevant
market
is
its
geographical
extent.
Given
the
nature
of
the
MP&
M
industry,

however,
this
factor
is
not
important
because
it
pertains
more
to
industries
dealing
with
perishable
commodities
and
those
with
high
transportation
costs.

b.
Import
competition
Theory
suggests
that
imports
as
a
percent
of
domestic
sales
are
negatively
associated
with
market
power
because
competition
from
foreign
firms
limits
domestic
firms 
ability
to
exercise
such
power.
Firms
belonging
to
sectors
in
which
imports
make
up
a
relatively
large
proportion
of
domestic
sales
will
therefore
be
at
a
relative
disadvantage
in
their
ability
to
pass­
through
costs
compared
to
firms
belonging
to
sectors
with
lower
levels
of
import
penetration,
a
measure
of
import
competition.

Import
penetration,
the
ratio
of
imports
in
a
sector
to
the
total
value
of
domestic
consumption
in
that
sector,
is
particularly
significant
because
foreign
producers
will
not
incur
costs
as
a
result
of
this
regulation.

In
the
market
structure
analysis,
higher
import
penetration
generally
means
that
firms
are
exposed
to
greater
competition
from
foreign
producers
and
will
thus
possess
less
market
power
to
increase
prices
in
response
to
regulation­
induced
increases
in
production
costs.
The
Census
Bureau
provides
import
data
at
the
four­
digit
SIC
level.
EPA
estimated
sector
import
penetration
ratios
as
the
ratio
of
the
sum
of
component
industry
imports
divided
by
the
sum
of
component
industry
value
of
domestic
consumption9.

c.
Export
competition
The
MP&
M
regulation
will
not
increase
the
production
costs
of
foreign
producers
with
whom
domestic
firms
must
compete
in
export
markets.
As
a
result,
sectors
that
rely
to
a
greater
extent
on
export
sales
will
have
less
latitude
in
increasing
prices
to
recover
cost
increases
resulting
from
regulation­
induced
increases
in
production
costs.
They
will
therefore
have
a
lower
CPT
potential,
all
else
being
equal.

This
analysis
uses
export
dependence,
defined
as
the
percentage
of
shipments
from
a
sector
that
is
exported,
to
measure
the
degree
to
which
a
sector
is
exposed
to
competitive
pressures
abroad
in
export
sales.
EPA
used
export
data
at
the
four­
digit
SIC
level
and
derived
sector
export
dependence
ratios:
the
sum
of
component
industry
exports
divided
by
the
sum
of
component
industry
value
of
shipments.

That
domestic
producers
export
a
substantial
share
of
their
product
does
not
necessarily
imply
that
they
are
subject
to
greater
competitive
pressures
abroad
compared
to
what
they
face
in
domestic
markets.
Such
would
be
the
case
in
sectors
where
U.
S.

producers
are
the
dominant
suppliers
worldwide.
To
account
for
this
possibility,
EPA
analyzed
in
more
detail
those
sectors
showing
high
export
dependence
to
see
if
domestic
firms
in
those
sectors
appear
to
dominate
the
world
market.
10
Based
on
information
presented
in
the
profile
of
MP&
M
industry
profile,
EPA
determined
that
firms
in
all
four
of
these
sectors
(
i.
e.,

precious
metals
and
jewelry,
ordnance,
office
machine,
and
aircraft)
operate
in
highly
competitive
international
markets.
The
conventional
theory
that
higher
export
dependence
results
in
relatively
lower
market
power
is
therefore
assumed
to
hold
true
for
all
MP&
M
sectors.

8
The
four­
digit
SIC
category,
while
not
a
perfect
delineation,
is
most
often
used
by
industrial
organization
economists
in
their
studies
because,
among
publicly
available
data
sources,
these
industries
appear
to
correspond
most
closely
to
economic
markets
(
Waldman
&
Jensen,
1997).

9
Census
data
on
imports,
exports,
and
value
of
shipments
for
the
year
1996
were
used
for
estimating
this
and
the
next
market
structure
indicator.

10
EPA
considered
sectors
with
export
dependence
exceeding
30
percent
for
this
part
of
the
analysis.

B­
10
MP&
M
EEBA:
Appendices
Appendix
B:
Cost
Pass­
Through
Analysis
A
substantial
body
of
literature
studies
the
link
between
environmental
regulation
and
competitiveness
in
international
trade.

Overall,
little
empirical
evidence
seems
to
support
the
hypothesis
that
environmental
regulations
have
had
a
significant
adverse
effects
on
the
international
competitiveness
of
domestic
firms
(
Jaffe
et
al.,
1995).
Nonetheless,
export
dependence
as
an
important
independent
factor
in
assessing
the
validity
of
the
estimated
CPT
coefficients.
If
historical
changes
in
input
costs
have
affected
both
domestic
and
foreign
firms
more
or
less
uniformly,
then
the
econometrically
estimated
Ep
would
not
address
situations
in
which
only
domestic
firms
face
higher
costs.
Determining
the
exact
extent
to
which
changes
in
input
costs
have
affected
both
domestic
and
foreign
producers
uniformly
is
beyond
the
scope
of
this
analysis.
Such
changes,

however,
can
affect
a
significant
proportion
of
cost
changes
related
to
the
non­
environmental
aspect
of
inputs,
such
as
those
for
energy,
imported
raw
materials,
and
imported
manufactured
inputs.

Given
the
above,
European
and
other
developed
countries
have
also
implemented
strict
environmental
regulations
comparable
to
U.
S.
regulations;
even
changes
in
environmental
costs
have
therefore
often
been
relatively
uniform
across
domestic
and
foreign
firms.
This
uniformity
may
account
for
the
fact
that
past
studies
do
not
show
substantial
impacts
of
U.
S.

environmental
regulation
on
the
balance
of
trade.

Because
this
regulation
will
affect
only
domestic
firms,
and
the
analysis
assumes
that
no
similar
regulatory
response
is
expected
in
foreign
countries
at
least
in
the
short
term,
domestic
firms
will
face
relatively
higher
production
costs
compared
to
their
international
competitors
as
a
result
of
regulation.
To
study
the
impact
of
this
regulation
on
the
change
in
MP&
M
industry
competitiveness
in
international
markets,
the
market
structure
analysis
must
therefore
include
measures
that
assess
the
effect
of
each
sector s
dependence
on
export
markets
on
its
ability
to
pass
through
costs.

d.
Long­
term
industry
growth
An
industry s
competitiveness
and
the
ability
of
firms
to
engage
in
price
competition
are
likely
to
differ
between
declining
and
growing
industries.
Most
studies
have
found
that
recent
growth
in
revenue
is
positively
related
to
profitability
(
Waldman
&
Jensen,
1997),
which
suggests
a
greater
ability
to
recover
costs
fully.

Based
on
Census
Bureau
data,
EPA
estimated
the
average
growth
rate
in
the
value
of
shipments
between
1988
and
1996
for
each
sector,
with
the
value
of
shipments
for
each
component
industry
also
serving
as
the
weights
for
deriving
average
sector
growth
rates.
EPA
expects
firms
in
sectors
with
higher
growth
rates
to
be
better
positioned
to
pass
through
compliance
costs
rather
than
being
forced
to
absorb
such
cost
increases
in
order
to
retain
market
share
and
revenues.

e.
Competition
barriers
Barriers
to
entry
and
exit
help
a
concentrated
industry
exert
market
power
by
deterring
potential
competitors
from
entering
the
market.
Without
these
barriers,
a
firm
that
tries
to
pass
through
compliance
costs
by
raising
its
prices
risks
losing
its
market
share
to
new
firms
that
see
an
opportunity
to
compete
at
higher
prices.

 
Entry
barriers
are
the
fixed
costs
of
beginning
business
in
an
industry.
Entry
barriers
include
high
capital
costs,
brand
name
reputations
that
require
a
large
advertising
expense
to
overcome,
a
long
learning
curve,
and
any
other
factors
that
make
the
costs
for
new
entrants
higher
than
the
costs
of
existing
firms.

 
Exit
barriers
are
the
fixed
costs
that
cannot
be
salvaged
upon
leaving
the
industry.
They
are
sometimes
called
sunk
costs
and
are
measured
as
the
difference
between
the
replacement
value
of
a
facility's
capital
and
its
liquidation
value.
Exit
barriers
include
factors
that
make
it
difficult
for
a
firm
to
liquidate
its
assets,
such
as
specialized
machinery
that
cannot
be
sold
or
converted
to
alternative
uses,
brand
names
that
cannot
transfer
well
to
other
products,
or
substantial
shutdown
liabilities
that
would
offset
the
value
of
assets
in
liquidation.
The
capital
valuations
are
typically
needed
to
measure
exit
barriers.

An
analysis
measuring
entry
and
exit
barriers
can
avoid
problems
of
data
availability
by
identifying
directly
the
presence
of
above­
normal
profits
that
such
barriers
would
permit.
This
analysis
uses
a
sector s
risk­
normalized
return
on
assets
(
ROA)
as
an
indicator
of
profit
rates
and
the
likely
presence
of
entry
and
exit
barriers.
A
popular
measure
used
by
managers
for
measuring
firm
performance,
the
ROA
is
used
an
indicator
of
firm
profitability.
This
analysis
estimates
an
ROA
before
interest
payments
and
taxes
to
compare
firms
with
different
capital
structures.
Using
the
pre­
tax
ROA
results
in
the
adding
back
of
the
interest
tax
shield
and
permits
comparing
ROAs
among
firms
assumed
to
be
entirely
equity­
financed.
The
analysis
measures
firm
riskiness
by
the
Asset
Beta,
which
is
the
firms 
Equity
Beta
(
i.
e.,
measure
of
the
firm s
riskiness
as
an
investment
relative
to
the
market
for
equity
investments
as
a
whole),
adjusted
to
remove
their
financing
decision
from
the
beta
calculation.
With
this
adjustment,
the
analysis
can
compare
firms
with
different
capital
structures
because
the
Asset
Beta
represents
the
beta
of
common
stock
had
the
firm
been
entirely
equity­
financed.

B­
11
MP&
M
EEBA:
Appendices
Appendix
B:
Cost
Pass­
Through
Analysis
The
Capital
Asset
Pricing
Model
(
CAPM)
states
that
the
expected
risk
premium
on
an
investment
(
return
earned
over
and
above
the
risk­
free
rate)
reflect
investment s
riskiness
relative
to
the
market
(
beta).
The
Treynor
Ratio,
a
commonly
used
performance
measure
that
uses
betas
as
a
measure
of
risk,
embodies
this
principle
of
the
CAPM:

Treynor
Ratio
=
(
Return
from
Investment
­
Risk
Free
Interest
Rate)
/
(
Beta
of
Investment)

For
this
analysis,
however,
the
Treynor
Ratio,
or
any
other
performance
measure
requiring
estimation
of
the
risk
premium
on
an
investment,
could
not
be
used.
More
than
60
percent
of
the
firms
in
the
analysis
had
five
year,
pre­
tax
ROAs
that
were
lower
than
the
risk­
free
interest
rate
of
5.21
percent
(
return
on
the
three­
month
U.
S.
Treasury
Bill
for
the
five­
year
period
1996­
2000).
The
analysis
using
the
Treynor
Ratio
yielded
results
that
did
not
permit
a
meaningful
comparison
of
risk­

normalized
ROAs
among
sectors.
This
analysis
therefore
used
a
modified
form
of
the
Treynor
Ratio
that
adjusts
the
total
return
and
not
just
the
risk
premium
by
the
riskiness
of
an
investment.
Applying
this
modification,
the
analysis
estimated
the
risk­
normalized
ROAs
as
follows:

Risk­
Normalized
ROA
=
ROA
/
Asset
Beta
The
analysis
estimated
risk­
normalized
ROAs
for
sectors
using
firm
level
data
as
opposed
to
data
at
the
4­
digit
SIC
level,
and
identified
firms
belonging
to
each
MP&
M
sector
using
a
two
step
process:

 
First,
EPA
assigned
facilities
(
and
their
parent
firms)
responding
to
the
MP&
M
facilities
survey
to
the
sector
from
which
they
received
the
largest
portion
of
their
revenues.

 
Second,
EPA
identified
additional
facilities
belonging
to
each
sector
using
a
financial
information
Web
site
(
marketguide.
com),
which
provides
a
classification
of
publicly­
traded
firms
by
the
4­
digit
SIC
code
of
their
largest
business
segment
based
on
revenues.

EPA
estimated
ROA
and
Beta
values
for
a
five­
year
time
period,
and
estimated
sector
risk­
normalized
ROAs
by
weighting
11
each
firm s
risk­
normalized
ROA
by
its
market
capitalization.

The
use
of
the
risk­
normalized
ROA
measure
only
assigns
MP&
M
sectors
relative
rankings
and
does
not
imply
that
they
face
high
or
low
barriers
to
competition
in
absolute
terms.
The
analysis
assumes
that
higher
risk­
adjusted
profits
in
general
indicate
potential
entry
and
exit
barriers
and
above
average
market
power.

11
EPA
further
studied
the
business
activities
of
firms
belonging
in
the
MP&
M
facilities
survey
that
were
identified
as
conglomerates
or
found
to
own
multiple
facilities
belonging
to
more
than
one
MP&
M
sector,
and
of
firms
in
the
broader
sample
having
a
market
capitalization
exceeding
$
25
billion.
This
additional
step
ensured
that
the
market
capitalization
weight
used
in
the
analysis
represented
only
the
fraction
of
revenues
that
the
firm
receives
from
its
business
activities
in
the
MP&
M
sector(
s)
of
interest.

B­
12
MP&
M
EEBA:
Appendices
Appendix
B:
Cost
Pass­
Through
Analysis
B.
3.2
Results
EPA
used
these
five
indicators
to
assign
each
sector
a
cost
pass­
through
score.
Higher
numerical
values
indicate
greater
CPT
potential
for
some
indicators
(
e.
g.,
industry
concentration)
and
lesser
CPT
potential
for
others
(
e.
g.,
import
competition).

Table
B.
4
summarizes
the
specific
ranking
definitions
for
each
indicator.

Table
B.
4:
Summary
of
Ranking
Rules
for
Assessing
Relative
Pass­
Through
Potential
Based
on
Market
Structure
Considerationsa
Variable
Indicates
Greater
Pass­

Through
Potential
(
High
Rank)
Variable
Indicates
Lesser
Pass­

Through
Potential
(
Low
Rank)

8­
Firm
Concentration
Ratio
Greater
than
median
Lesser
than
median
Ratio
of
Imports
to
Shipments
Lesser
than
median
Greater
than
median
Ratio
of
Exports
to
Shipments
Lesser
than
median
Greater
than
median
Average
Growth
Rate
of
Shipments
Greater
than
median
Lesser
than
median
Risk­
Normalized
Pre­
Tax
Return
on
Assets
Greater
than
median
Lesser
than
median
a
All
assessments
of
pass­
through
potential
are
relative
among
the
19
MP&
M
Sectors.

Source:
U.
S.
EPA
analysis.

For
each
of
the
five
indicators,
EPA
ranked
sectors
from
1
to
19,
with
1
assigned
to
the
sector
assessed
to
have
the
lowest
CPT
potential
and
19
assigned
to
the
sector
assessed
to
have
the
highest
CPT
potential.
12
Based
on
this
scoring
system,
the
possible
score
for
a
sector
when
all
five
of
its
ranks
are
summed
ranges
from
5
to
95.
Table
B.
5
presents
a
summary
of
the
results
for
the
market
structure
analysis.

12
This
ranking
scale
differs
from
the
scale
used
to
assign
scores
in
the
market
structure
analysis
undertaken
for
the
Phase
I
MP&
M
analysis.
In
the
Phase
I
analysis,
depending
on
the
variable
under
consideration,
a
sector
received
a
value
of
+
1
if
it
indicated
a
greater
CPT
potential
relative
to
the
median
and
a
value
of
­
1
if
it
indicated
a
lesser
CPT
potential
relative
to
the
median
.
The
sector
at
the
median
received
a
value
of
0.
The
use
of
the
median
value
as
the
threshold
for
determining
relatively
higher
or
lower
(+
1
or
­
1)
market
power
was
somewhat
arbitrary,
especially
for
values
closely
centered
around
the
median.
The
new
scale,
since
it
considers
individual
sector
ranks,
is
superior
because
it
explicitly
recognizes
that
extreme
values
are
more
likely
to
be
indicative
of
high
or
low
market
power,
and
accordingly
assigns
them
a
higher
or
lower
score.
For
example,
the
old
scale
would
assign
a
sector
with
industry
concentration
just
above
the
median
(
e.
g.,
other
metal
products)
the
same
score
of
+
1
as
a
very
highly­
concentrated
industry,
such
as
aerospace.
The
new
scale,
however,
recognizes
the
difference
in
industry
concentration
between
the
two
sectors
and
therefore
assigns
the
first
sector
a
rank
close
to
10
and
aerospace
a
rank
close
to
19.

B­
13
MP&
M
EEBA:
Appendices
Appendix
B:
Cost
Pass­
Through
Analysis
Table
B.
5:
Results
of
the
Market
Structure
Analysisa
Overall
Rank
Sector
8­
firm
Concentration
Ratio
Import
Penetration
(%)
Export
Dependence
(%)
Avg.
Annual
Growth
Rate
(%)
Risk­

Normalized
ROA
(%)
Aggregate
Score
Value
Rank
Value
Rank
Value
Rank
Value
Rank
Value
Rank
1
Precious
Metals
and
Jewelry
35.0
4
77.36
1
49.85
2
­
1.9
3
14.43
10
20
2
Printed
Circuit
Boards
35.0
3
21.99
8
17.07
10
1.5
8
7.50
2
31
3
Ordnance
76.90
16
18.92
10
50.17
1
­
7.3
2
12.30
6
35
4
Household
Equipment
54.22
10
33.18
3
17.02
11
1.5
9
12.02
5
38
4
Office
Machine
61.38
14
51.85
2
43.41
4
3.1
15
9.58
3
38
6
Electronic
Equipment
47.27
9
24.55
6
24.04
6
5.1
18
7.21
1
40
7
Aircraft
85.3
18
22.74
7
46.43
3
­
1.7
4
16.15
13
45
8
Iron
and
Steel
41.87
6
4.54
16
1.32
17
0.4
6
11.38
4
49
9
Other
Metal
Products
54.27
11
32.40
4
17.57
9
1.1
7
26.60
19
50
10
Stationary
Industrial
Equipment
41.16
5
17.71
11
23.64
7
3.7
16
16.78
14
53
11
Hardware
24.52
2
14.31
14
11.37
13
2.1
11
17.18
15
55
12
Instruments
44.2
8
15.33
12
23.07
8
1.8
10
19.64
18
56
13
Mobile
Industrial
Equipment
58.56
13
21.42
9
29.62
5
2.8
13
18.13
17
57
14
Ships
and
Boats
58.20
12
6.49
15
6.48
15
­
1.5
5
16.11
12
59
15
Job
Shop
19.26
1
0.00
19
0.00
19
3.1
14
13.44
9
62
15
Motor
Vehicle
77.30
17
27.56
5
15.74
12
2.6
12
18.10
16
62
17
Aerospace
92.29
19
0.75
18
0.75
18
­
7.6
1
13.19
8
64
17
Bus
&
Truck
42.51
7
2.86
17
3.04
16
4.8
17
12.31
7
64
19
Railroad
71.00
15
15.16
13
10.26
14
7.6
19
14.62
11
72
a
Shaded
values
are
the
medians
for
each
market
structure
indicator.

Source:
U.
S.
EPA
analysis
This
rank
scoring
system
has
some
important
limitations:

1.
This
grading
scale
implicitly
assigns
equal
weights
to
each
of
the
five
market
structure
indicators.
Clearly,
the
impact
of
each
of
these
five
indicators
on
market
power
will
vary
from
sector
to
sector,
and
some
indicators
are
likely
to
dominate
others
within
each
sector.

2.
Although
the
ranking
scale
distinguishes
between
sectors
with
extreme
values
and
those
that
are
close
to
the
median,

it
does
not
permit
an
accurate
judgement
about
how
significant
a
particular
value
may
be
in
determining
market
power.
For
each
indicator,
sectors
are
simply
ranked
from
1
to
19
based
on
the
lowest
to
highest
market
power
potential.
The
change
in
market
power
expected
as
one
moves
from
sector
1
to
sector
5
is
not
likely
to
be
equal,

however,
to
the
change
in
market
power
expected
as
one
moves
from
sector
6
to
sector
10.

In
general,
the
market
structure
analysis
revealed
that
a
discernable
gap
exists
in
the
estimated
parameters
around
rank
4/
5
and
around
rank
14/
15
for
most
indicators
(
see
Table
B.
6).
For
each
indicator,
two
small
groups,
each
containing
about
four
to
B­
14
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
MP&
M
EEBA:
Appendices
five
sectors,
therefore
seem
to
have
relatively
low
and
high
market
power.
A
much
larger
group
of
about
nine
to
ten
sectors
exhibit
average
market
power.
Appendix
B:
Cost
Pass­
Through
Analysis
Table
B.
6:
istribution
of
Estimated
Parameters
for
Market
Structure
Variables
Rank
8­
firm
Concentration
Ratio
Import
Penetration
Export
Dependence
Average
Annual
Growth
Rate
Risk­
Normalized
ROA
19.26
77.36%
50.17%
­
7.6%
7.21
24.52
51.85%
49.85%
­
7.3%
7.50
35.00
33.18%
46.43%
­
1.9%
9.58
35.07
32.40%
43.41%
­
1.7%
11.38
a
41.16
27.56%
29.62%
­
1.5%
12.02
41.87
24.55%
24.04%
0.4%
12.30
42.51
22.74%
23.64%
1.1%
12.31
44.22
21.99%
23.07%
1.5%
13.19
47.27
21.42%
17.57%
1.5%
13.44
a
54.22
18.92%
17.07%
1.8%
14.43
54.27
17.71%
17.02%
2.1%
14.62
58.20
15.33%
15.74%
2.6%
16.11
58.56
15.16%
11.37%
2.8%
16.15
61.38
14.31%
10.26%
3.1%
16.78
a
71.00
6.49%
6.48%
3.1%
17.18
76.90
4.54%
3.04%
3.7%
18.10
77.30
2.86%
1.32%
4.8%
18.13
85.32
0.75%
0.75%
5.1%
19.64
92.29
0.00%
0.00%
7.6%
26.60
D
a
Highlighted
rows
mark
the
natural
gaps
in
the
various
indicators.

Source:
U.
S.
EPA
analysis
The
aggregate
market
structure
scores
for
all
sectors
range
from
a
low
of
19
to
a
high
of
71.
Apart
from
the
lowest
score
(
precious
metals
and
jewelry)
and
the
highest
score
(
railroad),
all
the
other
scores
are
uniformly
distributed
with
no
clear
breaks
in
their
distribution
that
can
be
used
for
classifying
sectors
by
their
CPT
potential
(
see
Table
B.
5).
EPA
therefore
used
an
alternative
classification
system
for
the
market
structure
analysis.
Based
on
the
average
aggregate
score
of
50
(
average
rank
of
10),
EPA
assigned
sectors
with
an
aggregate
score
of
40
or
below
(
average
rank
of
8
or
less)
to
the
low
CPT
category,

and
assigned
sectors
with
an
aggregate
score
of
60
or
above
(
average
rank
of
12
or
more)
to
the
high
CPT
category.
EPA
assigned
sectors
with
aggregate
scores
between
these
cutoffs
to
the
average
CPT
category.
Table
B.
7
shows
the
categorization
of
all
19
sectors
by
their
CPT
potential
based
on
this
classification
system.
In
total,
EPA
classified
six,
eight,

and
five
sectors
in
the
low,
average,
and
high
CPT
categories,
respectively.
The
classification
cutoffs,
though
somewhat
arbitrary,
result
in
a
sector
classification
similar
to
the
trends
witnessed
for
most
individual
indicators,
such
that
about
five
sectors
are
classified
in
the
low
and
high
CPT
categories
and
the
remaining
sectors
are
classified
as
having
average
CPT
potential.

B­
15
MP&
M
EEBA:
Appendices
Appendix
B:
Cost
Pass­
Through
Analysis
Table
B.
7:

Low
CPT
Average
CPT
High
CPT
Precious
Metals
&
Jewelry
Aircraft
Job
Shop
Printed
Circuit
Boards
Iron
&
Steel
Motor
Vehicle
Ordnance
Other
Metal
Products
Aerospace
Household
Equipment
Stationary
Industrial
Equipment
Bus
&
Truck
Office
Machine
Hardware
Railroad
Electronic
Equipment
Instruments
Mobile
Industrial
Equipment
Ships
&
Boats
Classification
of
MP&
M
Sectors
by
CPT
Ability
Source:
U.
S.
EPA
analysis
Although
recognizing
the
limitations
of
the
ranking
scale,
EPA
believes
that
it
is
useful
for
presenting
the
results
succinctly
and
provides
a
basis
for
validating
the
estimated
CPT
coefficients.
Analyzing
the
relative
importance
of
each
indicator
for
each
of
the
sectors
is
beyond
the
scope
of
this
analysis.

B.
4
VALIDATION
OF
ECONOMETRICALLY­
ESTIMATED
CPT
COEFFICIENTS
The
econometric
analysis
provides
a
quantitative
assessment
of
what
the
cost
pass­
through
ability
of
each
sector
appears
to
be.
The
market
structure
analysis
yields
a
judgment
of
what
the
pass­
through
ability
of
each
sector
ought
to
be.
In
this
section
the
two
analyses
are
brought
together,
with
the
results
of
the
market
structure
analysis
used
to
validate
the
CPT
coefficients
estimated
by
the
econometric
analysis.

Table
B.
8
shows
a
comparison
of
each
sector s
CPT
classification
based
on
the
econometric
analysis
and
the
market
structure
analysis.
The
two
analyses
classify
13
of
the
19
sectors
in
the
same
CPT
category.
For
these
sectors,
the
market
structure
analysis
appears
to
validate
the
CPT
coefficient
derived
using
the
econometric
analysis.
No
econometric
estimate
is
available
for
one
sector
(
aerospace);
for
this
sector,
EPA
used
only
the
market
structure
analysis.
For
the
remaining
five
sectors,

however,
the
two
analyses
assign
sectors
to
different
CPT
categories.
EPA
undertook
a
more
detailed
analysis
of
these
sectors 
market
structure
to
validate
their
CPT
coefficient.
Specifically,
EPA
examined
the
following
two
factors
affecting
firm s
market
power
in
a
given
industrial
sector:

 
Whether
any
(
i.
e.,
one
or
more)
of
the
five
structural
indicators
may
be
extremely
important
or
irrelevant
for
a
particular
sector,
and
therefore
whether
its
effect
on
market
power
is
being
under­
weighted
or
over­
weighted,

respectively.

 
Whether
other
factors
affecting
market
power
for
these
sectors
have
not
been
included
in
the
market
structure
analysis,
but
which
possibly
have
substantial
effects
on
market
power/
CPT
ability
in
particular
sectors.

The
discussion
below
summarizes
EPA s
review
and
conclusions
for
each
of
these
six
sectors.

B­
16
MP&
M
EEBA:
Appendices
Appendix
B:
Cost
Pass­
Through
Analysis
Table
B.
8:
Comparison
of
Sectoral
Classification
Based
on
Econometric
and
Market
Structure
Analysis
Sector
Econometric
Analysis
Market
Structure
Analysis
CPT
Categorization
Matches
Electronic
Equipment
Low
Low
Household
Equipment
Low
Low
Office
Machine
Low
Low
Ordnance
Low
Low
Precious
Metals
and
Jewelry
Low
Low
Printed
Circuit
Boards
Low
Low
Hardware
Average
Average
Instruments
Average
Average
Iron
and
Steel
Average
Average
Ships
and
Boats
Average
Average
Stationary
Industrial
Equipment
Average
Average
Bus
&
Truck
High
High
Railroad
High
High
CPT
Categorization
Does
Not
Match
Other
Metal
Products
Low
Average
Job
Shop
Low
High
Motor
Vehicle
Average
High
Aircraft
High
Average
Mobile
Industrial
Equipment
High
Average
CPT
Comparison
Not
Possible
Aerospace
N/
A
High
Source:
U.
S.
EPA
analysis.

B.
4.1
Other
Metal
Products
This
sector
is
assigned
to
the
low
category
by
the
econometric
analysis
and
the
average
category
by
the
market
structure
analysis.
EPA
believes
that
the
estimated
CPT
coefficient
for
this
sector
is
accurate
and
that
the
market
structure
score
for
this
sector
is
somewhat
misleading
because
of
the
exceptionally
high
risk­
normalized
ROA
derived
for
it.
A
priori,
there
appears
to
be
no
reason
why
firms
in
this
sector
should
be
able
to
earn
significantly
higher
returns
than
in
other
sectors,
and
the
high
risk­
normalized
ROA
estimated
is
likely
an
artifact
of
the
small
sample
of
firms
for
which
financial
data
were
available
to
estimate
risk­
normalized
returns
for
this
sector.
The
other
four
indicators
of
market
power
suggest
below­
average
CPT
for
this
sector,
which
agrees
with
the
CPT
coefficient
estimated
from
the
econometric
analysis.

B.
4.2
Job
Shops
EPA
assigned
this
sector
to
the
low
category
by
the
econometric
analysis
and
the
high
category
by
the
market
structure
analysis.
EPA
believes
that
the
market
structure
analysis
may
be
misleading
due
to
the
high
CPT
ranks
assigned
to
the
Import
Penetration
and
Export
Dependence
indicators
of
market
power
for
this
sector.
These
two
indicators
of
market
power
are
not
relevant
for
this
sector,
however,
because
the
sector
is
not
trade­
oriented.
EPA
expects
the
level
of
domestic
competition
among
job
shops
to
be
the
single
most
important
factor
that
determines
market
power
and
the
ability
of
firms
to
pass
through
costs
in
the
sector.
The
Job
Shop
sector
has
the
lowest
concentration
ratio
among
all
the
sectors,
suggesting
that
the
sector
is
characterized
by
a
substantial
number
small
firms
(
see
Table
3.8
in
the
MP&
M
Industry
Profile)
that
are
most
likely
engaged
B­
17
MP&
M
EEBA:
Appendices
Appendix
B:
Cost
Pass­
Through
Analysis
in
intense
competition
among
each
other.
The
estimated,
low,
CPT
coefficient
for
this
sector
therefore
appears
to
be
appropriate.

B.
4.3
Motor
Vehicle
This
sector
is
assigned
to
the
average
category
by
the
regression
analysis
and
the
high
category
by
the
market
structure
analysis.
EPA
believes
that
this
sector
is
characterized
by
average
cost
pass­
through
potential
due
to
the
extremely
competitive
nature
of
the
motor
vehicle
industry
both
domestically
and
in
international
markets.
In
recent
years,
in
a
bid
to
remain
or
become
more
competitive,
the
trend
in
this
industry
has
been
towards
the
continual
consolidation
of
firms
into
globalized
manufacturers.
In
fact,
motor
vehicle
manufacturers
are
no
longer
constrained
within
national
boundaries,
as
mergers
and
joint
ventures
include
some
of
the
largest
firms
from
different
countries.
In
addition,
manufacturers
have
increasingly
standardized
the
design
of
motor
vehicles
and
their
parts,
changes
that
have
resulted
in
much
less
product
differentiation
(
but
greater
product
quality)
among
manufacturers.
The
increasing
intensity
of
global
competition
and
the
move
towards
decreasing
product
differentiation
are
likely
to
limit
the
ability
of
domestic
producers
to
pass­
through
significant
portions
of
their
cost
increases
associated
with
this
regulation.
Therefore,
the
finding
of
an
average
cost
pass­

through
coefficient
appears
to
be
justified.

B.
4.4
Aircraft
This
sector
is
assigned
to
the
high
category
by
the
econometric
analysis
and
the
average
category
by
the
market
structure
analysis.
Based
on
the
unique
nature
of
the
global
aircraft
industry,
EPA
believes
that
the
estimated
CPT
coefficient
for
this
sector
is
appropriate.
Not
only
is
the
industry
concentrated
domestically
(
concentration
ratio
of
85.3),
but
this
is
also
true
of
the
global
aircraft
manufacturing
industry.
In
recent
years,
the
industry
has
witnessed
substantial
restructuring
through
mergers
and
consolidation,
both
nationally
and
internationally
(
see
section
3.2.2
in
the
MP&
M
Industry
Profile).
The
highly
concentrated
nature
of
the
industry,
combined
with
the
sizeable
share
of
the
domestic
market
that
is
controlled
by
domestic
aircraft
manufacturers,
suggests
that
firms
in
this
sector
have
the
ability
to
pass
through
a
significant
portion
of
their
cost
increases.

B.
4.5
Mobile
Industrial
Equipment
EPA
assigned
this
sector
to
the
high
category
by
the
econometric
analysis
and
the
average
category
by
the
market
structure
analysis.
EPA
believes
that
this
sector
is
more
appropriately
characterized
by
average
CPT
because
the
sector
has
witnessed
certain
trends
in
recent
years
that
suggest
that
firms
in
this
sector
do
not
have
a
high
ability
to
pass
through
cost
increases.

Specifically,
growth
rates
in
the
construction
and
the
farm
and
machinery
equipment
industries
started
to
level
off
or
even
declined
in
recent
years
after
a
sustained
period
of
growth
(
see
section
3.2.10
in
the
MP&
M
Industry
Profile).
These
declining
trends
are
not
fully
represented
in
the
regression
analysis
because
the
last
year
of
analysis
is
1996.
EPA
therefore
revised
the
CPT
coefficient
for
this
sector
to
equal
the
average
CPT
value
for
all
sectors
classified
in
the
average
category
based
on
the
regression
analysis.

B.
4.6
Aerospace
Since
the
market
structure
analysis
categorizes
the
Aerospace
sector
in
the
high
CPT
category,
EPA
estimated
the
CPT
coefficient
for
this
sector
as
the
average
CPT
value
for
all
sectors
classified
in
the
high
category
based
on
the
regression
analysis
(
excluding
Mobile
Industrial
Equipment
whose
CPT
coefficient
was
revised
based
on
the
market
structure
analysis).

B.
5
ADJUSTING
ESTIMATES
OF
COMPLIANCE
CPT
POTENTIAL
The
CPT
values
estimated
above
reflect
sector
level
CPT
potential.
The
methodology
must
consider
that
ability
to
pass
on
cost
increases
through
price
increases
will
differ
at
the
industry
level
versus
the
facility
level.
Cost
increases
that
affect
all
facilities
in
an
industry
are
more
likely
to
be
recovered
through
industry­
wide
price
increases,
whereas
cases
where
only
a
few
facilities
in
an
industry
incur
cost
increases
are
less
likely
to
result
in
price
increases.
This
analysis
must
therefore
take
into
account
the
proportion
of
an
industry
that
will
experience
cost
increases
when
applying
industry­
level
cost
pass­
through
coefficients.

B­
18
MP&
M
EEBA:
Appendices
Appendix
B:
Cost
Pass­
Through
Analysis
For
the
final
MP&
M
rule,
EPA
will
use
the
method
used
in
the
Phase
I
analysis
where
EPA
adjusted
the
industry­
level
cost
pass­
through
coefficient
downward
in
proportion
to
the
percentage
of
sector
output
bearing
compliance
cost.
The
ratio
of
the
revenues
in
water­
discharging
facilities
affected
by
the
rule
divided
by
total
revenues
in
the
MP&
M
sector
provided
a
measure
of
the
fraction
of
production
in
the
MP&
M
sector
likely
to
be
affected
by
cost
increase.
That
is,
a
cost
pass­
through
percentage
of
90
percent
would
be
reduced
to
72
percent
if
80
percent
of
the
sector
output
was
subject
to
the
regulation
(.
80
×
.90
=
.72).
EPA
applied
this
adjusted
pass­
through
percentage
to
the
percentage
cost
increase
experienced
by
the
regulated
facilities
only
(
i.
e.,
sum
of
compliance
costs
divided
by
the
sum
of
baseline
costs
for
the
facilities
subject
to
the
rule).
Table
B.
9
presents
the
adjusted
CPT
coefficients
estimated
for
each
sector.

Table
B.
9:
Adjusted
Estimates
of
Compliance
Cost
Pass­
Through
Potential
by
MP&
M
Sector
Sector
Unadjusted
Cost
Pass­

Through
Potential
Estimated
Fraction
of
Sector s
Revenue
Subject
to
Regulation
(%)
Adjusted
Cost
Pass­
Through
Potential
Aerospacea
0.98
100.00
1.00
Aircraftb
1.20
100.00
1.00
Bus
&
Truck
0.86
100.00
0.96
Electronic
Equipment
0.39
100.00
0.42
Hardware
0.77
33.50
0.26
Household
Equipment
0.64
100.00
0.64
Instruments
0.77
100.00
0.77
Iron
and
Steel
0.77
100.00
0.77
Job
Shop
0.57
43.70
0.25
Mobile
Industrial
Equipmentc
0.79
100.00
0.79
Motor
Vehicle
0.82
44.10
0.36
Office
Machinesd
(
9.33)
34.50
0.00
Ordnance
0.59
100.00
0.59
Other
Metal
Products
0.63
100.00
0.63
Precious
Metals
&
Jewelry
0.64
42.90
0.27
Printed
Circuit
Boards
(
0.34)
53.60
0.00
Railroad
0.88
100.00
0.88
Ships
and
Boats
0.82
100.00
0.82
Stationary
Industrial
Equipment
0.79
32.20
0.25
a
CPT
coefficient
for
the
Aerospace
sector
estimated
based
on
the
market
structure
analysis.

b
For
the
Aircraft
sector,
the
cost­
pass
through
potential
is
capped
at
100%.
CPT
coefficient
for
the
Mobile
Industrial
Equipment
sector
revised
based
on
the
market
structure
analysis.

d
For
the
Office
Machine
and
Printed
Circuit
Boards
sectors,
the
cost­
pass
through
coefficients
are
set
to
zero
based
on
both
the
estimated
negative
regression
coefficient
and
the
results
of
the
market
structure
analysis.

Source:
U.
S.
EPA
analysis
B­
19
c
MP&
M
EEBA:
Appendices
Appendix
B:
Cost
Pass­
Through
Analysis
ATTACHMENT
B.
A:
SELECTED
REVIEW
OF
CPT
LITERATURE
To
support
the
CPT
analysis,
EPA
undertook
a
selected
review
of
previous
CPT
analyses.
The
two
most
studied
areas
in
the
literature
deal
with
exchange
rate
pass­
through
and
tax
pass­
through.
Unfortunately,
neither
of
these
study
types
is
useful
in
assessing
the
reliability
of
the
MP&
M
CPT
results.
Sections
B.
A.
2
and
B.
A.
3
provide
a
brief
summary
of
this
studies.
One
study
(
Ashenfelter
et
al,
1998)
estimates
the
pass­
through
rate
for
cost
changes
faced
by
an
individual
firm
and
compares
it
with
passes­
through
of
cost
changes
common
to
all
firms
in
an
industry.
This
appears
to
be
the
most
relevant
to
the
analysis
of
compliance
costs
pass
through.
Section
B.
A.
1
provides
a
brief
summary
of
findings
from
this
study.

B.
A.
1
Ashenfelter
et
al.
(
1998),
 
Identifying
the
Firm­
Specific
Cost
Pass­
Through
Rate. 

As
noted
above,
Ashenfelter
et
al.
(
1998)
examines
the
pass­
through
rate
for
cost
changes
faced
by
only
an
individual
firm
(
Staples,
an
office
superstore
chain),
and
distinguishes
that
rate
from
the
rate
at
which
a
firm
passes
through
cost
changes
common
to
all
firms
in
an
industry.
Based
on
their
analysis,
they
find
the
combined
firm­
specific
and
industry­
wide
pass­

through
rate
(
i.
e.,
with
no
distinction
between
cost
changes
specific
to
the
individual
firm
and
those
applicable
to
the
entire
industry)
to
be
57
percent.
Conversely,
the
pass­
through
rate
estimated
for
only
firm­
specific
cost
changes
is
about
15
percent
and
the
pass­
through
rate
for
only
industry­
wide
cost
changes
is
close
to
85
percent.
The
finding
of
a
high
CPT
rate
for
industry­
wide
cost
changes
lends
support
to
EPA s
finding
of
similarly
high
historical
CPT
rates
for
many
of
the
MP&
M
sectors.

B.
A.
2
Exchange
Rate
Pass­
Through
The
exchange
rate
pass­
through
literature
examines
the
response
of
local
currency
import
prices
to
variation
in
the
exchange
rate
between
exporting
and
importing
countries.
Based
on
seven
studies
covering
the
period
1970
to
the
mid­
1980s,
Menon
(
1995)
finds
that
the
estimated
aggregate
pass­
through
of
exchange
rate
changes
to
import
prices
ranges
from
a
low
of
48.7
percent
to
a
high
of
91
percent.
The
mean
value
for
pass­
through
for
the
sample
of
studies
he
considered
is
69.9
percent.
In
contrast,
Feinberg
(
1989)
considers
the
impacts
of
exchange
rate
movements
on
U.
S.
domestic
prices
and
finds
an
average
pass­
through
of
16
percent
in
real
terms.
The
pass­
through
is
close
to
complete
for
industries
that
are
heavily
reliant
on
imported
inputs
and
producing
goods
highly
substitutable
for
imports.
Pass­
through
rates
are
much
lower
for
capital­
intensive
and
concentrated
industries
and
those
protected
by
barriers
to
entry.
The
exchange
rate
pass­
through
scenario,
however,
is
not
comparable
to
the
nature
of
compliance
cost
changes
expected
under
the
MP&
M
regulation
and
the
resultant
pass­
through
responses
from
domestic
producers
because
the
studies
focus
primarily
on
the
impact
of
exchange
rate
changes
on
prices
of
imported
goods
and
not
on
prices
of
domestically
produced
goods.
Feinberg s
study
appears
to
be
more
relevant,
but
he
does
not
present
pass­
through
rates
for
individual
industries,
and
does
not
explain
why
pass­
through
rates
are
much
lower
for
capital­
intensive
and
concentrated
industries
and
those
protected
by
barriers
to
entry.

B.
A.
3
Tax
Pass­
Through
The
literature
on
tax
pass­
through
examines
the
impact
of
excise
tax
changes
on
prices.
Of
the
several
studies
that
addressed
the
issue
of
tax
pass­
through,
the
majority
report
pass­
through
rates
slightly
in
excess
of
a
100
percent
(
Ashenfelter
et
al.,

1998).
This
literature
is
not
entirely
relevant
to
the
CPT
scenario
being
analyzed
for
this
rule
because
most
of
these
studies
analyze
changes
in
excise
tax
rates
in
the
cigarette
industry.
In
addition,
excise
tax
changes
on
final
goods
do
not
affect
manufacturing
costs,
and
they
have
a
uniform
impact
on
the
entire
industry.
Excise
taxes
do
affect
domestic
producers,

however,
by
altering
final
demand
and
therefore
revenues
received.

B.
A.
4
Studies
Cited
Ashenfelter,
Orley,
et
al.
(
1998),
 
Identifying
the
Firm­
Specific
Cost
Pass­
Through
Rate, 
FTC
Working
Paper
No.
217,

January.

Feinberg,
Robert
M
(
1989),
 
The
Effects
of
Foreign
Exchange
Movements
on
U.
S.
Domestic
Prices, 
The
Review
of
Economics
and
Statistics.

Jaffe,
A.
et
al.,
(
1995),
 
Environmental
Regulation
and
the
Competitiveness
of
U.
S.
Manufacturing:
What
does
the
Evidence
Tell
Us? 
Journal
of
Economic
Literature,
XXXIII
(
March):
132­
63.

B­
20
MP&
M
EEBA:
Appendices
Appendix
B:
Cost
Pass­
Through
Analysis
Menon,
Jayant
(
1995),
 
Exchange
Rate
Pass­
Through, 
Journal
of
Economic
Surveys,
9(
2).

Waldman,
Don
E.
and
Elizabeth
J.
Jensen
(
1997),
Industrial
Organization:
Theory
and
Practice.
Addison­
Wesley.

B­
21
MP&
M
EEBA:
Appendices
Appendix
B:
Cost
Pass­
Through
Analysis
ACRONYMS
CAPM:
Capital
Asset
Pricing
Model
CPT:
cost
pass­
through
ECI:
Employment
Cost
Index
PPI:
Producer
Price
Index
ROA:
risk­
normalized
return
on
assets
B­
22
MP&
M
EEBA:
Appendices
Appendix
C:
Summary
of
Moderate
Impact
Threshold
Values
by
Sector
INTRODUCTION
Facilities
subject
to
moderate
impacts
from
the
rule
are
expected
to
experience
financial
stress
short
of
closure.

This
analysis
uses
two
financial
indicators:
(
1)
Pre­
Tax
Return
on
Assets
(
PTRA)
and
(
2)
Interest
Coverage
Ratio
(
ICR).
These
threshold
values
were
compared
to
pre­
and
post­
compliance
PTRA
and
ICR
values
for
sample
facilities
to
determine
if
facilities
choosing
to
remain
in
business
after
promulgation
of
effluent
guidelines
would
Appendix
C:
Summary
of
Moderate
Impact
Threshold
Values
by
Sector
APPENDIX
CONTENTS
C.
1
Developing
Threshold
Values
for
Pre­
Tax
Return
on
Assets
(
PTRA)
.................
.................
..
C­
1
C.
2
Developing
Threshold
Values
for
t
Coverage
Ratio
(
ICR)
.................
.................
....
C­
2
C.
3
ary
of
Results
.................
.............
C­
4
References
.................
.................
.......
C­
5
Interes
Summ
experience
moderate
impacts
on
their
ability
to
attract
and
finance
new
capital.
The
remainder
of
this
appendix
describes
the
sources
and
methodology
used
to
derive
sector­
specific
moderate
impact
threshold
values.

EPA
calculated
the
thresholds
using
income
and
financial
structure
information
by
4­
digit
SIC
code
from
the
Risk
Management
Association
(
RMA)
Annual
Statement
Studies
for
eight
years
1994­
2001
(
RMA,
2001;
RMA
1998).
This
source
provides
quartile
values
derived
from
statements
of
commercial
bank
borrowers
and
loan
applicants
for
firms
having
less
than
$
250
million
in
total
assets.
These
criteria
may
introduce
bias,
since
firms
with
particularly
poor
financial
statements
might
be
less
likely
to
apply
to
banks
for
loans,
and
some
types
of
firms
may
be
more
likely
to
use
bank
financing
than
others.
However,
the
RMA
data
offers
the
advantage
of
being
available
by
4­
digit
SIC
codes
and
for
quartile
ranges.

RMA
did
not
provide
data
for
all
4­
digit
SIC
codes
associated
with
an
MP&
M
sector.
Out
of
174
manufacturing
SIC
codes
and
50
non­
manufacturing
SIC
codes,
52
manufacturing
SIC
codes
(
30
percent)
and
13
non­
manufacturing
SIC
codes
(
26
percent),
had
no
years
of
data
available.
RMA
did
not
compile
data
for
any
SIC
codes
in
two
manufacturing
sectors,

Ordnance
and
Aerospace
and
one
non­
manufacturing
sector,
Precious
Metals
and
Jewelry.
When
data
were
not
available
for
any
SIC
codes
within
the
sector,
EPA
calculated
an
average
manufacturing
or
non­
manufacturing
threshold
to
use
as
a
proxy.

The
4­
digit
SIC
code
data
were
consolidated
into
weighted
sector
averages,
weighted
by
1997
value
of
shipments
from
the
Economic
Censuses
(
U.
S.
DOC,
1997).
For
each
sector
and
impact
measure,
a
separate
threshold
was
calculated
for
manufacturing
and
non­
manufacturing
SIC
codes.
The
use
of
the
RMA
data
for
calculating
the
threshold
values
for
pre­
tax
return
on
assets
and
interest
coverage
ratio
is
outlined
below.

C.
1
DEVELOPING
THRESHOLD
VALUES
FOR
PRE­
TAX
RETURN
ON
ASSETS
(
PTRA)

Pre­
tax
return
on
total
assets
measures
the
effectiveness
of
management
in
employing
the
resources
available
to
it.
A
low
ratio
may
indicate
that
a
borrower
would
have
difficulty
financing
treatment
investments
and
continuing
to
attract
investment.

The
following
data
from
Risk
Management
Association
Annual
Statement
Studies
were
used
to
calculate
PTRA:

 
%
Profit
Before
Taxes
/
Total
Assets25th
Ratio
of
profit
before
taxes
divided
by
total
assets
and
multiplied
by
100
for
the
lowest
quartile
of
values
in
each
4­
digit
SIC
code.

 
Operating
Profit
Gross
profit
minus
operating
expenses.

 
Profit
Before
Taxes
Operating
profit
minus
all
other
expenses
(
net).

RMA
provides
a
measure
of
pre­
tax
return
on
assets
that
approximates
the
measure
that
EPA
defined
for
the
moderate
impact
analysis.
As
defined
by
RMA,
this
measure
is
the
ratio
of
pre­
tax
income
to
assets,
designated
ROARMA:

ROARMA
=
Pre­
Tax
Income
(
EBT)
/
ASSETS25th
C­
1
MP&
M
EEBA:
Appendices
Appendix
C:
Summary
of
Moderate
Impact
Threshold
Values
by
Sector
However,
as
defined
by
EPA
for
its
analysis,
the
numerator
of
the
PTRA
measure
requires
the
use
of
earnings
before
interest
and
taxes
(
EBIT)
instead
of
pre­
tax
income
(
EBT).
Defined
as
EBIT,
the
PTRA
numerator
will
capture
all
return
from
assets,

whether
going
to
debt
or
equity.
To
derive
a
pre­
tax,
total
return
value,
EPA
adjusted
RMA s
measure
of
PTRA
using
the
median
percentage
values
of
EBIT
and
EBT
available
from
RMA.
This
adjustment
yields
the
PTRA
measure
that
EPA
used
in
the
moderate
impact
analysis,
designated
ROAMP&
M
:

ROAMP&
M
=
ROARMA
*
EBIT
/
EBT
Negative
values
are
included
in
the
weighted­
sector
PTRA
averages
but
a
different
method
is
used
to
adjust
the
ROA
values
reported
in
RMA
to
the
value
used
in
the
moderate
impact
analysis.
Specifically,
using
only
those
observations
(
i.
e.,
4­
digit
SIC
code
and
year
combinations)
with
positive
values
for
%
Profit
Before
Taxes
/
Total
Assets,
Operating
Profit,
and
Profit
Before
Taxes,
EPA
calculated
an
adjustment
factor
by
subtracting
the
difference
between
ROAMP&
M
and
ROARMA
as
follows:

ROAMP&
M­
ROARMA
=
adjustment
factor.

Those
values
were
consolidated
into
sector­
specific
adjustment
factors,
weighted
by
1997
value
of
shipments
from
the
Economic
Censuses
(
U.
S.
DOC,
1997).
Each
negative
PTRA
observation
from
RMA
was
adjusted
by
its
sector
specific
adjustment
factor
to
approximate
the
measure
used
in
the
moderate
impact
analysis:

ROARMA
+
sector­
specific
adjustment
factor
=
ROAMP&
M
The
sector­
specific
adjustment
factors
average
0.47
for
manufacturing
sectors
and
range
from
0.13
for
the
Office
Machines
sector
to
0.60
for
the
Aircraft
and
Motor
Vehicle
sectors.
The
sector­
specific
adjustment
factors
average
0.22
for
non­

manufacturing
sectors
and
range
from
0.15
for
the
Motor
V
ehicle
sector
to
0.74
for
the
Railroad
sector.

C.
2
DEVELOPING
THRESHOLD
VALUES
FOR
INTEREST
COVERAGE
RATIO
(
ICR)

Interest
coverage
ratio
is
a
measure
of
a
firm's
ability
to
meet
current
interest
payments
and,
on
a
pro­
forma
basis,
to
meet
the
additional
interest
payments
under
a
new
loan.
A
high
ratio
may
indicate
that
a
borrower
would
have
little
difficulty
in
meeting
the
interest
obligations
of
a
loan.
This
ratio
also
serves
as
an
indicator
of
a
firm's
capacity
to
take
on
additional
debt.

The
following
data
from
Risk
Management
Association
Annual
Statement
Studies
were
used
to
calculate
ICR:

 
EB
IT/
Interest25th
Ratio
of
earnings
(
profit)
before
annual
interest
expense
and
taxes
(
EBIT)
divided
by
annual
interest
expense
for
the
lowest
quartile
of
values
in
each
4­
digit
SIC
code.

 
%
Dep
r.,
Dep.,
Amort./
Salesmed
Median
ratio
of
annual
depreciation,
amortization
and
depletion
expenses
divided
by
net
sales
and
multiplied
by
100.

 
Op
erating
Profit
Gro
ss
profit
minus
ope
rating
expense
s.

RMA
provides
a
measure
of
interest
coverage
that
approximates
the
measure
that
EPA
defined
for
the
moderate
impact
analysis.
As
defined
by
RMA,
this
measure
is
the
ratio
of
earnings
before
interest
and
taxes
to
interest,
designated
ICRRMA
:

ICRRMA
=
EBIT
/
INTEREST25th
However,
as
defined
by
EPA
for
its
analysis,
the
numerator
of
the
ICR
measure
requires
the
use
of
earnings
before
interest,

taxes,
depreciation,
and
amortization
(
EBITDA)
instead
of
earnings
before
interest
and
taxes
(
EBIT).
Defined
this
way,
the
ICR
numerator
will
include
all
operating
cash
flow
that
could
be
used
for
interest
payments.
To
derive
the
desired
ICR
value
(
designated
ICRMP&
M
),
EPA
adjusted
the
RMA
value
as
outlined
below:

ICRMP&
M
=
EBITDA
/
INTEREST
Therefore,
ICRMP&
M
=
ICRRMA
*
(
EBIT
+
DA)
/
EBIT
*
{
1+
[(
DA
/
SALES)
/
(
EBIT
/
SALES)]}
or
ICRMP&
M
=
ICRRMA
C­
2
MP&
M
EEBA:
Appendices
Appendix
C:
Summary
of
Moderate
Impact
Threshold
Values
by
Sector
For
consistency
of
calculation,
EPA
used
the
median
values
available
from
RMA
for
the
adjusting
both
the
numerator
(
DA
/
SALES)
and
denominator
(
EBIT
/
SALES)
terms.
1
EPA
used
the
same
method
as
described
above
to
adjust
the
negative
ICR
values
reported
in
RMA
to
the
value
used
in
the
moderate
impact
analysis.
Including
only
those
observations
with
positive
values
for
EBIT/
Interest,
%
Depr.,
Dep.,

Amort./
Sales,
and
Operating
Profit,
an
adjustment
factor
was
calculated
by
subtracting
the
difference
between
ICRMP&
M
and
ICRRMA
as
follows:

ICRMP&
M­
ICRRMA
=
adjustment
factor.

A
sector­
specific
adjustment
factor
was
calculated
for
ICR
values
similar
to
the
PTRA.
Each
negative
ICR
observation
from
RMA
was
adjusted
by
its
sector
specific
adjustment
factor
to
approximate
the
measure
used
in
the
moderate
impact
analysis:

ICRRMA
+
sector­
specific
adjustment
factor
=
ICRMP&
M
The
sector­
specific
adjustment
factors
average
0.59
for
manufacturing
sectors
and
range
from
0.28
for
the
Precious
Metals
and
Jewelry
sector
to
0.79
for
the
Printed
Circuit
Board
sector.
The
sector­
specific
adjustment
factors
average
0.50
for
non­

manufacturing
sectors
and
range
from
0.24
for
the
Office
Machines
sector
to
1.85
for
the
Aircraft
sector.

1
Numerator
(%
Depr.,
Dep.,
Amort./
Sales)
is
available
for
quartile
values;
denominator
(
Operating
Profit)
only
for
median
values.

C­
3
MP&
M
EEBA:
Appendices
Appendix
C:
Summary
of
Moderate
Impact
Threshold
Values
by
Sector
C.
3
SUMMARY
OF
RESULTS
Table
C.
1
shows
the
resulting
threshold
values
for
PTRA
and
ICR
by
sector.
The
PTRA
values
for
manufacturers
range
from
zero
percent
for
the
Office
Machine
sector
to
2.8
percent
for
the
Aircraft
and
Household
Equipment
sectors
and
for
the
non­

manufacturers
the
values
range
from
0.3
percent
for
the
Office
Machine
sector
to
3.1
percent
for
the
Railroad
sector.
The
ICR
values
for
manufacturers
range
from
1.4
for
the
Office
Machine
and
Railroad
sectors
to
2.3
for
the
Hardware,
Household
Equipment,
and
Printed
Circuit
Board
sectors
and
for
the
non­
manufacturers
the
values
range
from
1.2
for
the
Office
Machine
sector
to
2.9
for
the
Aircraft
sector.

In
assessing
moderate
impacts,
EPA
used
the
non­
manufacturing
threshold
for
facilities
that
reported
100
percent
of
their
revenues
came
from
rebuilding
and
maintenance;
otherwise,
EPA
used
the
manufacturing
threshold.

Table
C.
1:
Summary
of
Moderate
Impact
Thresholds
by
Sector
Sector
Pre­
Tax
Return
on
Assets
(
PTRA)
Interest
Coverage
Ratio
(
ICR)

Manufacturing
Non­

Manufacturing
Manufacturing
Non­

Manufacturing
Hardware
b
2.6%
1.6%
2.3
1.9
Aircraft
2.8%
0.4%
2.2
2.9
Electronic
Equipmentb
2.1%
1.6%
2.2
1.9
Stationary
Industrial
Equipment
2.1%
2.5%
2.1
2.8
Ordnancea
2.2%
1.6%
2.1
1.9
Aerospacea
2.2%
1.6%
2.1
1.9
Mobile
Industrialb
2.6%
1.6%
2.1
1.9
Instrument
2.2%
2.0%
2.1
2.0
Precious
and
Non­
Preciousa
1.8%
1.6%
1.7
1.9
Ships
and
Boats
1.7%
1.0%
1.6
2.0
Household
Equipment
2.8%
2.6%
2.3
2.0
Railroadb
1.1%
3.1%
1.4
2.7
Motor
Vehicle
2.4%
1.5%
2.0
1.7
Bus
and
Truck
2.3%
1.7%
2.0
2.8
Office
Machine
0.0%
0.3%
1.4
1.2
Printed
Circuit
Boardb
2.5%
1.6%
2.3
1.9
Job
Shopb
2.3%
1.6%
2.2
1.9
Other
Metal
Products
1.0%
1.7%
1.6
1.8
Iron
and
Steel
2.4%
N/
A
2.2
N/
A
Unknown
Sectora
2.2%
1.6%
2.1
1.9
a
When
data
were
not
available
for
any
SIC
codes
within
the
sector,
EPA
calculated
an
average
manufacturing
or
non­
manufacturing
threshold
to
use
as
a
proxy.
b
There
are
no
non­
manufacturing
SIC
codes
in
several
sectors,
but
in
these
sectors
there
are
some
facilities
who
reported
that
all
of
their
revenue
came
from
rebuilding
and
maintenance.
In
these
cases,
EPA
used
the
average
non­
manufacturing
thresholds
in
that
sector
as
a
proxy
for
the
non­
manufacturing
threshold.

Source:
RMA,
2001;
RMA,
1998;
U.
S.
Economics
Census,
1997;
U.
S.
EPA
Analysis,
2002.

C­
4
MP&
M
EEBA:
Appendices
Appendix
C:
Summary
of
Moderate
Impact
Threshold
Values
by
Sector
REFERENCES
U.
S.
Department
of
Commerce.
1997.
Bureau
of
the
Census.
Census
of
Manufacturers,
Census
of
Transportation,
Census
of
Wholesale
Trade,
Census
of
Retail
Trade,
Census
of
Service
Industries.

Risk
Management
Association
(
RMA).
1997­
1998.
Annual
Statement
Studies.

Risk
Management
Association
(
RMA).
2000­
2001.
Annual
Statement
Studies.

C­
5
MP&
M
EEBA:
Appendices
Appendix
C:
Summary
of
Moderate
Impact
Threshold
Values
by
Sector
THIS
PAGE
INTENTIONALLY
LEFT
BLANK
C­
6
MP&
M
EEBA:
Appendices
Appendix
D:
Estimating
Capital
Outlays
for
MP&
M
Discounted
Cash
Flow
Analyses
INTRODUCTION
The
economic
impact
analysis
for
the
Metal
Products
&

Machinery
Industry
(
MP&
M)
final
regulation
involved
calculation
of
the
business
value
of
sample
facilities
on
the
basis
of
a
discounted
cash
flow
(
DCF)
analysis
of
operating
cash
flow
as
reported
in
facility
questionnaires.

Business
value
is
calculated
on
a
pre­
and
post­
compliance
basis
and
the
change
in
this
value
serves
as
an
important
factor
in
estimating
regulatory
impacts
in
terms
of
potential
facility
closures.
For
proposal,
the
business
value
calculation
was
based
only
on
cash
flow
from
operations
and
did
not
recognize
cash
outlays
for
capital
acquisition
as
a
component
of
cash
flow.
EPA
Office
of
Water
(
OW)
previously
identified
that
the
omission
of
capital
acquisition
cash
outlays
from
the
DCF
analysis
may
lead
to
overstatement
of
the
business
value
of
sample
facilities
and,
as
a
consequence,
understatement
of
regulatory
impacts
in
terms
of
estimated
facility
closures.
Appendix
D:
Estimating
Capital
Outlays
for
MP&
M
Discounted
Cash
Flow
Analyses
APPENDIX
CONTENTS
D.
1
ncepts
Underlying
Analysis
of
Capital
Outlays
.................
.............
D­
2
D.
2
ying
Variables
for
the
Analysis
................
D­
4
D.
3
et
.............
D­
7
D.
4
.................
D­
8
D.
4.1
r
Model
Specification
.................
..
D­
9
D.
4.2
Linear
Model
Specification
..............
D­
10
D.
4.3
vity
Analysis
.................
.......
D­
12
D.
5
tion
.................
..............
D­
12
Attachment
D.
A:
Bibliography
of
Literature
Reviewed
for
this
Analysis
.................
...........
D­
17
Attachment
D.
B:
istorical
Variables
Contained
in
the
Value
Line
Investment
Survey
Dataset
................
D­
18
Analytic
Co
Specif
Selecting
the
Regression
Analysis
Datas
Specification
of
Models
to
be
Tested
Linea
Log­

Sensiti
Model
Valida
H
In
response
to
this
omission,
the
Office
of
Management
and
Budget
suggested
the
adoption
of
depreciation
as
a
surrogate
for
cash
outlays
for
capital
replacement
and
additions.
However,
for
several
reasons
EPA
believes
depreciation
is
a
poor
surrogate.
First,
depreciation
is
meant
to
capture
the
consumption/
use
of
previously
acquired
assets,
not
the
cost
of
replacing,

or
adding
to,
the
existing
capital
base.
Therefore,
depreciation
is
fundamentally
the
wrong
concept
to
use
as
a
surrogate
for
capital
outlays
for
capital
replacement
and
additions.
Second,
depreciation
is
estimated
based
on
the
historical
asset
cost,

which
may
understate
or
overstate
the
real
replacement
cost
of
assets.
Third,
both
book
and
tax
depreciation
schedules
generally
understate
the
assets 
useful
life.
Thus,
reported
depreciation
will
overstate
real
depreciation
value
for
recently
acquired
assets
that
are
still
in
the
depreciable
asset
base,
and
conversely,
understate
the
real
depreciation
value
of
assets
that
have
expired
from
the
depreciable
asset
base
but
still
remain
in
valuable
use.
Finally,
depreciation
does
not
capture
the
important
variations
in
capital
outlays
that
result
from
differences
in
revenue
growth
and
financial
performance
among
firms.

Businesses
with
real
growth
in
revenues
will
need
to
expand
both
their
fixed
and
working
capital
assets
to
support
business
growth,
and
all
else
being
equal,
growing
businesses
will
have
higher
ongoing
outlays
for
fixed
and
working
capital
assets.

Similarly,
the
ability
of
businesses
to
renew
and
expand
their
asset
base
depends
on
the
financial
productivity
of
the
deployed
capital
as
indicated
by
measures
such
as
return
on
assets
or
return
on
invested
capital.
As
a
result,
businesses
with
 
strong 

asset
productivity
will
attract
capital
for
renewal
and
expansion
of
their
asset
base,
while
businesses
with
 
weak 
asset
productivity
will
have
difficulty
attracting
the
capital
for
renewal
and
expansion
of
the
business 
asset
base.
All
else
being
equal,
businesses
with
strong
asset
productivity
will
have
higher
ongoing
outlays
for
capital
assets;
businesses
with
weak
asset
productivity
will
have
lower
ongoing
outlays
for
capital
assets.

As
an
approach
to
addressing
the
omission
of
capital
acquisition
cash
outlays
from
the
DCF
analysis,
EPA
undertook
to
estimate
a
regression
model
of
capital
outlays
using
capital
expenditure
and
relevant
explanatory
financial
and
business
environment
information
for
public­
reporting
firms
in
the
MP&
M
industry
sectors.
The
estimated
model
was
then
used
to
estimate
capital
outlays
for
facilities
in
the
MP&
M
sample
dataset.
The
estimated
capital
outlay
values
were
used
in
the
DCF
analyses
to
calculate
business
value
of
sample
facilities
and
estimate
regulatory
impacts
in
terms
of
facility
closures.

D­
1
MP&
M
EEBA:
Appendices
Appendix
D:
Estimating
Capital
Outlays
for
MP&
M
Discounted
Cash
Flow
Analyses
This
appendix
reports
the
results
of
this
effort,
including:
an
overview
of
the
analytic
concepts
underlying
the
analysis
of
capital
outlays;
specific
variables
included
in
the
regression
analysis;
summary
of
data
selection
and
preparation;
general
specification
of
regression
models
to
be
tested;
and
the
findings
from
the
regression
analyses.

D.
1
ANALYTIC
CONCEPTS
UNDERLYING
ANALYSIS
OF
CAPITAL
OUTLAYS
On
the
basis
of
general
economic
and
financial
concepts
of
investment
behavior,
EPA
began
its
analysis
by
outlining
a
framework
relating
the
level
of
a
firm s
capital
outlays
to
explanatory
factors
that:

 
can
be
observed
for
public­
reporting
firms
either
as
firm­
specific
information
or
general
business
environment
information
and
thus
be
included
in
a
regression
analysis;
and
 
for
firm­
specific
information,
are
also
available
from
the
MP&
M
sample
facility
dataset.

To
aid
in
identifying
the
explanatory
concepts
and
variables
that
might
be
used
in
the
analysis
and
as
well
in
specifying
the
models
for
analysis,
EPA
reviewed
recent
studies
of
the
determinants
of
capital
outlays.
EPA s
review
of
this
literature
generally
confirmed
the
overall
approach
in
seeking
to
estimate
capital
outlays
and
helped
to
identify
additional
specific
variables
that
other
analysts
found
to
contribute
important
information
in
the
analysis
of
capital
outlays
(
e.
g.,
the
decision
to
test
capacity
utilization
as
an
explanatory
variable,
see
below,
resulted
from
the
literature
review).
Articles
reviewed
are
listed
in
Attachment
D.
A
to
this
appendix.

Table
D.
1
beginning
below
and
continuing
the
following
two
pages
summarizes
the
conceptual
relationships
between
a
firm s
capital
outlays
and
explanatory
factors
that
EPA
sought
to
capture
in
this
analysis.
In
the
table,
EPA
outlines
the
concept
of
influence
on
capital
outlays,
the
general
explanatory
variable(
s)
that
EPA
identified
to
capture
the
concept
in
a
regression
analysis,
and
the
hypothesized
mathematical
relationship
(
sign
of
estimated
coefficients)
between
the
concept
and
capital
outlays.
Table
D.
2
identifies
the
specific
variables
included
in
the
analysis,
including
any
needed
manipulations
and
the
correspondence
of
the
variables
to
MP&
M
survey
information.

Table
D.
1:
Summary
of
Factors
Influencing
Capital
Outlays
Explanatory
Factor/
Concept
To
Be
Captured
in
Analysis
Translation
of
Concept
to
Explanatory
Variable(
s)
Expected
Relationship
Availability
of
attractive
opportunities
for
additional
capital
investment.
rm s
owners,
or
management
acting
on
behalf
of
owners,
should
expend
cash
for
capital
outlays
only
to
the
extent
that
the
expected
return
on
the
capital
outlays
whether
for
replacement
of,
or
additions
to,
existing
capital
stock
are
sufficient
to
compensate
providers
of
capital
for
the
expected
return
on
alternative,
competing
investment
opportunities,
taking
into
account
the
risk
of
investment
opportunities.
Historical
Return
On
Assets
of
establishment
as
a
indicator
of
investment
opportunities
and
management
effectiveness,
and,
hence,

of
desirability
to
expand
capital
stock
and
ability
to
attract
capital
investment.
plicitly
assumes
past
performance
is
indicative
of
future
expectations.
Positive
Business
growth
and
outlook
as
a
determinant
of
need
for
capital
expansion
and
attractiveness
of
investment
opportunities.
se
equal,
a
firm
is
more
likely
to
have
attractive
investment
opportunities
and
need
to
expand
its
capital
base
if
the
business
is
growing
and
the
outlook
for
business
performance
is
favorable.
Revenue
Growth,
from
the
prior
time
period(
s)
to
the
present,

provides
a
historical
measure
of
business
growth
and
is
a
potential
indicator
of
need
for
capital
expansion.
se
of
a
historical
variable
implicitly
assumes
past
performance
is
indicative
of
future
expectations.
Positive
Clearly,
the
theoretical
preference
is
for
a
forward­
looking
indicator
of
business
growth
and
need
for
capital
expansion.
Options
EPA
identified
include
Index
of
Leading
Indicators
and
current
Capacity
Utilization,
by
industry.
her
current
Capacity
Utilization
may
presage
need
for
capital
expansion.
Positive
A
fi
Use
of
a
historical
variable
im
All
el
U
Hig
D­
2
MP&
M
EEBA:
Appendices
Appendix
D:
Estimating
Capital
Outlays
for
MP&
M
Discounted
Cash
Flow
Analyses
Table
D.
1:
Summary
of
Factors
Influencing
Capital
Outlays
Explanatory
Factor/
Concept
To
Be
Captured
in
Analysis
Translation
of
Concept
to
Explanatory
Variable(
s)
Expected
Relationship
Importance
in
capital
in
business
production.
intensive
the
production
activities
of
a
business,
the
greater
will
be
the
need
for
capital
outlay
to
replenish,
and
add
to,
the
existing
capital
stock.
More
capital
intensive
businesses
will
spend
more
in
capital
outlays
to
sustain
a
given
level
of
revenue
over
time.
The
Capital
Intensity
of
production
as
measured
by
the
production
capital
required
to
produce
a
dollar
of
revenue
provides
an
indicator
of
the
level
of
capital
outlay
needed
to
sustain
and
grow
production.

As
an
alternative
to
a
firm­
specific
concept
such
as
Capital
Intensity
of
production,
differences
in
business
characteristics
might
be
captured
by
an
Industry
Classification
variable.
Positive
Life
of
capital
equipment
in
the
business.

All
else
equal,
the
shorter
the
useful
life
of
the
capital
equipment
in
a
business,
the
greater
will
be
the
need
for
capital
outlay
to
replenish,
and
add
to,
the
existing
capital
stock.
No
information
is
available
on
the
actual
useful
life
of
capital
equipment
by
business
or
industry
classification.
ever,
the
Capital
Turnover
Rate,
as
calculated
by
the
ratio
of
book
depreciation
to
net
capital
assets,
provides
an
indicator
of
the
rate
at
which
capital
is
depleted,
according
to
book
accounting
principles:

the
higher
the
turnover
rate,
the
shorter
the
life
of
the
capital
equipment.
However,
the
measure
is
imperfect
for
reasons
of
both
the
inaccuracies
of
book
reporting
as
a
measure
of
useful
life,
and
as
well
the
confounding
effects
of
growth
in
the
asset
base
due
to
business
expansion
which
will
tend
to
lower
the
indicated
turnover
rate,
all
else
equal,
without
a
real
reduction
in
life
of
capital
equipment.

As
above,
an
alternative
to
a
firm­
specific
concept,
differences
in
business
characteristics
might
be
captured
by
an
Industry
Classification
variable.
Positive,

generally,
but
with
recognition
of
the
potential
for
counter­
trend
effects
The
cost
of
financial
capital.
which
capital
both
debt
and
equity
is
made
available
to
a
firm
will
determine
which
investment
opportunities
can
be
expected
to
generate
sufficient
return
to
warrant
use
of
the
financial
capital
for
equipment
purchases.
se
equal,
the
higher
the
cost
of
financial
capital,
the
fewer
the
investment/
capital
outlay
opportunities
that
would
be
expected
to
be
profitable
and
the
lower
the
level
of
outlays
for
replacement
of,
or
additions
to,
capital
stock.
Preferably,
measures
of
cost­
of­
capital
would
be
developed
separately
for
debt
and
equity.

The
Cost
of
Debt
Capital,
as
measured
by
an
appropriate
benchmark
interest
rate,
provides
an
indication
of
the
terms
of
debt
availability
and
how
those
terms
are
changing
over
time.
Preferably,
the
debt
cost/
terms
would
reflect
the
credit
condition
of
the
firm,
which
could
be
based
on
a
credit
safety
rating
(
e.
g.,
S&
P
Debt
Rating).
While
such
information
would
be
available
for
public
firms,
EPA
judged
that
developing
a
comparable
concept
for
MP&
M
sample
facilities
would
not
be
possible
within
the
scope
of
this
analysis.
Negative
The
cost
of
equity
capital
is
more
problematic
than
the
cost
of
debt
capital
since
it
is
not
directly
observable
for
either
public­
reporting
firms
or,
in
particular,
private
firms
in
the
MP&
M
dataset.

However,
a
readily
available
surrogate
such
as
Market­
to­
Book
Ratio
provides
insight
into
the
terms
at
which
capital
markets
are
providing
equity
capital
to
public­
reporting
firms:
the
higher
the
Market­
to­
Book
Ratio,
the
more
favorable
the
terms
of
equity
availability.
lthough
such
information
would
not
be
available
for
private
firms
in
the
MP&
M
sample,
EPA
judged
that
it
would
be
possible
to
develop
a
industry­
level
value
for
use
with
the
MP&
M
facility
analysis.
Negative
All
else
equal,
the
more
capital
How
The
cost
at
All
el
A
D­
3
MP&
M
EEBA:
Appendices
Appendix
D:
Estimating
Capital
Outlays
for
MP&
M
Discounted
Cash
Flow
Analyses
Table
D.
1:
Summary
of
Factors
Influencing
Capital
Outlays
Explanatory
Factor/
Concept
To
Be
Captured
in
Analysis
Translation
of
Concept
to
Explanatory
Variable(
s)
Expected
Relationship
The
price
of
capital
equipment.

of
capital
equipment
in
particular,
how
capital
equipment
prices
are
changing
over
time
will
influence
the
expected
return
from
capital
outlays.
ll
else
equal,
when
capital
equipment
prices
are
increasing,
the
expected
return
from
incremental
capital
outlays
will
decline
and
vice
versa.

However,
although
the
generally
expected
effect
of
higher
capital
equipment
prices
is
to
remove
certain
investment
opportunities
from
consideration,
the
potential
effect
on
total
capital
outlay
may
be
mixed.

expected
returns
are
such
that
the
demand
to
invest
in
capital
projects
is
relatively
inelastic,
the
effect
of
higher
prices
for
capital
equipment
may
be
to
raise,
instead
of
lower,
the
total
capital
outlay
for
a
firm.
Index
provides
an
indicator
of
the
change
in
capital
equipment
prices.
Negative,

generally,
but
with
recognition
of
the
potential
for
counter­

trend
effects
The
price
A
If
Source:
U.
S.
EPA
analysis.

D.
2
SPECIFYING
VARIABLES
FOR
THE
ANALYSIS
Working
from
the
general
concepts
of
explanatory
variables
outlined
above,
EPA
defined
the
specific
explanatory
variables
to
be
included
in
the
analysis.
A
key
requirement
of
the
regression
analysis
is
that
the
firm­
specific
explanatory
variables
included
in
the
regression
analysis
later
be
able
to
be
used
as
the
basis
for
estimating
capital
expenditures
for
facilities
in
the
MP&
M
dataset.
As
a
result,
in
defining
the
firm­
specific
variables,
it
was
necessary
to
ensure
that
the
definition
of
variables
selected
for
the
regression
analysis
using
data
on
public­
reporting
firms
be
consistent
with
the
data
items
available
for
facilities
in
the
MP&
M
dataset.

Also,
EPA s
selection
of
firm­
specific
variables
was
further
constrained
by
an
earlier
decision
to
use
the
Value
Line
Investment
Survey
(
VL)
as
the
source
of
firm­
specific
information
for
the
regression
analysis.
The
decision
to
use
VL
as
the
source
of
firm­
specific
data
for
the
analysis
was
driven
by
several
considerations:

 
Considerably
lower
price
than
alternatives.
VL
data
were
available
at
a
price
of
$
95
for
a
one­
time
data
purchase;
the
price
for
other
data
sources
such
as
Bloomberg
and
Standard
&
Poor s
ranged
from
$
7,000
to
$
11,000.

 
Reasonable
breadth
of
public­
reporting
firm
coverage.
The
VL
dataset
includes
7,500
firms.

 
Reasonable
breadth
of
temporal
coverage.
VL
provides
data
for
the
most
recent
10
years
i.
e.,
1991­
2000.

Although
ideally
EPA
would
have
preferred
a
longer
time
series
to
include
more
years
not
in
the
 
boom 

investment
period
of
the
mid­
to
late­
1990s.

 
Timeliness
of
access.
The
VL
data
are
provided
as
a
standard
package
and
thus
could
be
available
within
a
week
of
ordering
while
other
data
sources
(
e.
g.,
Bloomberg)
would
have
required
more
time
because
data
would
have
provided
as
a
custom
purchase.

 
Reasonable
coverage
of
concepts/
data
needed
for
analysis.
The
VL
data
includes
a
wide
range
of
financial
data
that
are
applicable
to
the
analysis
(
VL
provides
37
data
items
over
the
10
reporting
years;
see
Attachment
DB).

However,
because
of
the
pre­
packaged
nature
of
the
VL
data,
it
was
not
possible
to
customize
any
data
items
to
support
more
precise
definition
of
variables
in
the
analysis.
In
particular,
EPA
found
that
certain
balance
sheet
items
were
not
reported
to
the
level
of
specificity
preferred
for
the
analysis.
Overall,
though,
EPA
expects
the
consequence
of
using
more
aggregate,
less­
refined
concepts
should
be
minor.

D­
4
MP&
M
EEBA:
Appendices
Appendix
D:
Estimating
Capital
Outlays
for
MP&
M
Discounted
Cash
Flow
Analyses
The
decision
to
use
VL
data
for
the
analysis
constrained,
in
some
instances,
EPA s
choice
of
variables
for
the
analysis.

Table
D.
2
reports
the
specific
definitions
of
variables
included
in
the
analysis
(
both
the
dependent
variable
and
explanatory
variables),
including
any
needed
manipulations,
the
data
source,
the
MP
&
M
estimation
analysis
equivalent
(
either
the
corresponding
variable(
s)
in
the
MP&
M
questionnaire
or
other
source
outside
the
questionnaire),
and
any
issues
in
variable
definition.

Table
D.
2:
Variables
For
Capital
Expenditure
Modeling
Analysis
Variables
for
Regression
Analysis
MP&
M
Analysis
Equivalent
Comment
/
IssueVariable
Source
Calculation
Dependent
Variable
Gross
expenditures
on
fixed
assets:

CAPEX,
includes
outlays
to
replace,
and
add
to,
existing
capital
stock
Value
Line
Obtained
from
VL
as
Capital
Spending
per
Share.

CAPEX
calculated
by
multiplying
by
Average
Shares
Outstanding.
None:
to
be
estimated
based
on
estimated
coefficients.
This
value
and
all
other
dollar
values
in
the
regression
analysis
were
deflated
to
1996
(
base
year
for
MP&
M
regulatory
analysis)
using
2­
digit
SIC
PPI
values.

Explanatory
Variables
Firm­
Specific
Variables
On
Assets:

ROA
Value
Line
ROA
=
Operating
Income
/

Total
Assets.

Operating
Income,
defined
as
Revenue
less
Operating
Expenses
(
CoGS+
SG&
A),

and
Total
Assets
were
obtained
directly
from
VL.
From
Survey:
Revenue
less
Total
Operating
Expenses
(
Material
&

Product
Costs
+

Production
Labor
+

Cost
of
Contract
Work
+
Fixed
Overhead
+
R&
D
+

Other
Costs
&

Expenses)
Would
have
preferred
a
post­
tax
concept
in
numerator
and
a
deployed
production
capital
concept
in
denominator.

no
tax
value
per
se
and
would
require
calculation
of
tax
using
an
estimated
tax
rate,
which
could
introduce
error.
neither
VL
nor
MP&
M
survey
data
provide
sufficient
information
to
get
at
deployed
production
capital.
Both
However,
VL
provides
Also
D­
5
MP&
M
EEBA:
Appendices
Appendix
D:
Estimating
Capital
Outlays
for
MP&
M
Discounted
Cash
Flow
Analyses
Table
D.
2:
Variables
For
Capital
Expenditure
Modeling
Analysis
Variables
for
Regression
Analysis
MP&
M
Analysis
Equivalent
Comment
/
IssueVariable
Source
Calculation
Revenue
Growth:

RVGR
Value
Line
Primary
formulation
tested
for
linear
models
was
percentage
change
in
revenue
over
two
years
prior
to
current
year:
RVGR
=
(
REVt
 
REVt­
2)
/
REVt­
2.
provides
10
years
of
financial
statement
values
1991­

2000,
including
Revenue
by
year.

For
log­
linear
models,
the
growth
concept
was
dropped
and
REV
was
used
as
the
explanatory
variable
(
see
below
and
also
see
later
discussion
under
model
specification).
No
equivalent
needed.

Analysis
proposed
to
set
this
value
to
zero
in
estimating
capital
outlay
values
for
MP&
M
facilities.
The
use
of
a
zero
growth
value
is
consistent
with
estimating
the
replacement
capital
expenditures
in
a
no­

growth
steady­
state.
Using
a
revenue
growth
term
in
the
analysis
defined
over
the
prior
two
years
requires
three
years
of
revenue
data
(
e.
g.,
current
year
plus
trailing
two
years)
and
effectively
eliminates
two
observation
years
from
the
analysis
(
1991
and
1992).
iven
that
these
data
years
occurred
at
the
end
of
a
recession
period
and
before
the
mid­
to
late­
90s
economic
boom
period,
EPA
was
very
concerned
about
the
potential
loss
of
these
years
from
the
analysis
dataset.

In
the
end,
the
use
of
a
log­
linear
model
eliminated
the
need
to
construct
the
lagged
difference
variables
and
thus
mooted
the
concern
over
loss
of
early
year
observations.
se
of
the
log­
linear
model,
however,
also
eliminated
the
potential
to
set
the
growth
term
to
zero
in
estimating
baseline
capital
outlays
for
MP&
M
facilities.

Revenue:

REV
Value
Line
In
the
linear
models,
REV
included
as
a
scale
variable
together
with
REVGR,
as
outlined
above.
or
log­
linear
models,
retained
only
REV
as
the
explanatory
variable.

REV,
captures
the
percent
change/
growth
concept
in
the
log­
linear
formulation.
From
Survey:
Revenue
Using
REV
only
and
not
REVGR
in
the
log­
linear
model
restored
the
two
data
years
at
the
beginning
of
the
analysis
period
(
1991
and
1992)
to
the
analysis
dataset.

including
data
for
the
first
two
observation
period
years
is
important
for
the
generality
of
the
analysis.

Also
tested
Total
Assets
as
a
scale
variable,
which
provided
good,
but
not
as
strong,
an
explanation,
as
REV.

Capital
Turnover
Rate:
CAPT
Value
Line
CAPT
=
Depreciation
/

Total
Assets.
Depreciation
and
Total
Assets
directly
available
from
VL.
From
Survey:

Depreciation
/
Total
Assets
Would
have
preferred
denominator
of
net
fixed
assets
instead
of
total
assets.

However,
VL
provides
detailed
balance
sheet
information
for
only
the
four
most
recent
years.
possible
to
separate
current
assets
and
intangibles
from
total
assets.

Capital
Intensity:

CAPI
Value
Line
CAPI
=
Total
Assets
/

Revenue.
l
Assets
and
Revenue
directly
available
from
VL
From
Survey:

Total
Assets
/
Revenue
As
above,
would
have
preferred
net
fixed
assets
instead
of
total
assets,
but
needed
data
are
not
available
from
VL
for
the
full
analysis
period.

Market­
to­

Book
Ratio:

MV/
B
Value
Line
MV/
B
=
average
market
price
of
common
equity
(
Price)
divided
by
book
value
of
common
equity
(
Book
Value
per
Share).
Price
and
Book
Value
per
Share
directly
available
from
VL.
Use
average
of
MV/
B
for
firms
by
MP&
M
industry
group
in
regression
analysis
dataset;
calculated
at
time
of
MP&
M
industry
survey.
Ultimately
found
MV/
B
highly
correlated
with
other,
more
important
explanatory
variables,
which
makes
sense,
given
that
equity
terms
would
be
derived
from
more
fundamental
factors,

such
as
ROA.
MV/
B
from
the
analysis
eliminated
the
need
to
define
an
approach
to
use
this
variable
with
MP&
M
survey
data.
VL
G
The
u
F
The
simple
variable,
EPA
believes
Not
Tota
Omitting
D­
6
MP&
M
EEBA:
Appendices
Appendix
D:
Estimating
Capital
Outlays
for
MP&
M
Discounted
Cash
Flow
Analyses
Table
D.
2:
Variables
For
Capital
Expenditure
Modeling
Analysis
Variables
for
Regression
Analysis
MP&
M
Analysis
Equivalent
Comment
/
IssueVariable
Source
Calculation
General
Business
Environment
Variables
Interest
on
10­
year,
A­

rated
industrial
debt:

DEBTCST
Bloomberg
Financial
Services
DEBTCST
=
annual
average
of
rates
for
each
data
year
Use
average
of
DEBTCST
rates
at
time
of
MP&
M
industry
survey.
10­
year
maturity,
industry
debt
selected
as
reasonable
benchmark
for
industry
debt
costs.
10
years
became
 
standard 
maturity
for
industrial
debt
during
1990s.

Index
of
Leading
Indicators:

ILI
Conference
Board
Monthly
index
series
available
from
Conference
Board.

For
linear
models,
ILI
=

percent
change
from
beginning
to
end
of
current
year.

For
log­
linear
models,
ILI
=

geometric
mean
of
current
year
values.
For
linear
formulation,
use
average
of
year­
to­

year
percent
change
in
ILI
at
time
of
MP&
M
industry
survey.

For
log­
linear
formulation,
use
average
of
ILI
values
at
time
of
MP&
M
industry
survey.

Capacity
Utilization
by
Industry:

CAPUTIL
Federal
Reserve
Board
(
Dallas
Federal
Reserve)
Monthly
index
series
available
from
Federal
Reserve.
For
linear
models,

CAPUTIL
=
percent
change
in
annual
average
values
from
prior
year
to
current
year.

For
log­
linear
models,

CAPUTIL
=
current
year
average
value.
For
linear
formulation,

use
average
of
year­
to­
year
percent
change
in
CAPUTIL
at
time
of
MP&
M
industry
survey.

For
log­
linear
formulation,
use
average
of
CAPUTIL
values
at
time
of
MP&
M
industry
survey.

Producer
Price
Index
series
for
capital
equipment:

CAPPRC
Bureau
of
Labor
Statistics
Annual
average
values
available
from
BLS.

For
linear
models,

CAPPRC
=
percent
change
from
prior
year
to
current
year.

For
log­
linear
models,

CAPPRC
=
current
year
average
value
as
reported
by
BLS.
For
linear
formulation,

use
average
of
year­
to­

year
percent
change
in
CAPPRC
at
time
of
MP&
M
industry
survey.

For
log­
linear
formulation,
use
average
of
CAPPRC
values
at
time
of
MP&
M
industry
survey.
BLS
reports
PPI
series
for
capital
equipment
based
on
 
consumption
bundles 
defined
for
manufacturing
and
non­
manufacturing
industries.
or
this
analysis,
EPA
used
the
PPI
series
based
on
the
manufacturing
industry
bundle.
F
Source:
U.
S.
EPA
analysis.

D.
3
SELECTING
THE
REGRESSION
ANALYSIS
DATASET
In
addition
to
specifying
the
variables
to
be
used
in
the
regression
analysis,
EPA
also
needed
to
select
the
public
firm
dataset
on
which
the
analysis
would
be
performed.

As
noted
above,
EPA
used
the
Value
Line
Investment
Survey
as
the
source
for
public
firm
data.
VL
includes
over
7,500
publicly
traded
firms
and
identifies
firms 
principal
business
both
by
a
broad
industry
classification
(
e.
g.,
Electrical
Equipment,
Machinery)
and
by
an
SIC
code
assignment.
In
most
instances,
the
SIC
codes
assignment
is
only
at
the
2­
digit
level.
To
build
the
public
firm
dataset
corresponding
to
the
MP&
M
industry
sectors,
EPA
initially
selected
all
firms
included
in
the
following
2­
digit
SIC
code
families:

D­
7
MP&
M
EEBA:
Appendices
Appendix
D:
Estimating
Capital
Outlays
for
MP&
M
Discounted
Cash
Flow
Analyses
 
2500:
Furniture
and
fixtures,

 
3300:
Primary
metal
industries,

 
3400:
Fabricated
metal
products,

 
3500:
Industrial
machinery
and
equipment,

 
3600:
Electrical
and
electronic
equipment,

 
3700:
Transportation
equipment,
and
 
3800:
Instruments
and
related
products.

From
manual
inspection,
EPA
deleted
firms
in
four­
digit
SIC
code
3579,
which,
in
the
VL
classification,
was
comprised
only
of
software
manufacturers.
In
addition,
in
SIC
code
group
3300,
EPA
included
firms
only
in
the
ferrous
metal
processing
sectors:
SIC
codes
3311,
3312,
3315,
3316,
3317,
and
3398.1
As
a
result
of
this
selection,
EPA
developed
an
initial
dataset
of
1,015
firms.
On
inspection,
EPA
found
that
a
substantial
number
of
firms
did
not
have
data
for
the
full
10
years
of
the
analysis
period.
The
general
reason
for
the
omission
of
some
years
of
data
is
that
the
firms
did
not
become
publicly
listed
in
their
current
operating
structure
whether
through
an
initial
public
offering,
spin­
off,
divestiture
of
business
assets,
or
other
significant
corporate
restructuring
that
renders
earlier
year
data
inconsistent
with
more
recent
data
until
after
the
beginning
of
the
10­
year
data
period.
2
As
a
result,
the
omission
of
observation
years
for
a
firm
always
starts
at
the
beginning
of
the
data
analysis
period.
This
systematic
front­
end
truncation
of
firm
observations
in
the
dataset
could
be
expected
to
bias
the
analysis
in
favor
of
the
capital
expenditure
behavior
nearer
the
end
of
the
1990s
decade.
To
avoid
this
problem,
EPA
removed
all
firm
observations
that
have
fewer
than
10
years
of
data.

As
a
result,
the
dataset
used
in
the
analysis
has
a
total
of
3,900
yearly
data
observations
and
represents
390
firms.

Table
D.
3
presents
the
number
of
firms
by
industry
classifications.

Table
D.
3:

SIC
Industry
Classification
Number
of
Firms
2500:
Furniture
and
fixtures
13
3300:
Primary
metal
industries
27
3400:
Fabricated
metal
products
24
3500:
Industrial
machinery
and
equipment
119
3600:
Electrical
and
electronic
equipment
101
3700:
Transportation
equipment
65
3800:
Instruments
and
related
products
41
Number
of
Firms
by
Industry
Classifications
D.
4
SPECIFICATION
OF
MODELS
TO
BE
TESTED
On
the
basis
of
the
variables
listed
above
and
their
hypothesized
relationship
to
capital
outlays,
EPA
specified
a
time­
series,

cross
sectional
model
to
be
tested
in
the
regression
analysis.
EPA s
dataset
consisted
of
390
cross
sections
observed
at
10
years
(
1991
through
2000).
The
general
structure
of
this
model
was
as
follows:

CAPEXi,
t
=
f(
ROAi,
t
,
REVi,
t
,
CAPTi,
t
,
CAPIi,
t
,
DEBTCST
i,
t
,
CAPPRCt
,
CAPU
TILj,
t
)

1
These
4­
digit
SIC
codes
include
all
MP&
M
sectors
in
SIC
2­
digit
code
33
plus
4­
digit
SIC
code
3311,
to
capture
information
for
the
steel
manufacturing
industry.

2
When
VL
adds
a
firm
to
its
dataset,
it
fills
in
the
public­
reported
data
history
for
the
firm
for
the
lesser
of
10
years
or
the
length
of
time
that
the
firm
has
been
publicly
listed
and
thus
subject
to
SEC
public
reporting
requirements.

D­
8
MP&
M
EEBA:
Appendices
Appendix
D:
Estimating
Capital
Outlays
for
MP&
M
Discounted
Cash
Flow
Analyses
Where:

CAPEXi,
t
=
capital
expenditures
of
firm
i,
in
time
period
t;
1
t
=
year
(
year
=
1991,
.
.
.
,
2000);

i
=
firm
i
(
i=
1,
.
.
.
,
390);

j=
industry
classification
j
ROAi,
t
=
return
on
total
assets
for
firm
i
in
year
t;

REVi,
t
=
revenue
($
millions)
for
firm
i
in
year
t;

CAPTi,
t
=
capital
turnover
rate
for
firm
i
in
year
t;

CAPIi,
t
=
capital
intensity
for
firm
i
in
year
t;

DEBTCSTt
=
financial
cost
of
capital
in
year
t;

CAPPRCt
=
price
of
capital
goods
in
year
t;

CAPUTILj,
t
=
the
Federal
Reserve
Board s
Index
of
Capacity
utilization
for
a
given
industry
j
in
year
t.

EPA
tested
both
linear
and
log­
linear
model
specifications.
Both
models
fit
quit
well,
achieving
overall
correlation
(
R2
)
in
the
upper
80
percent/
low
90
percent
range.
However,
the
pattern
of
coefficient
significance
was
better
in
the
log­
linear
model.
In
addition,
the
log­
linear
model
offered
advantages
in
terms
of
retention
of
early
time
period
observations
and
variable
specification,
as
discussed
below.
Therefore,
EPA
selected
a
log­
linear
specification
as
the
final
model.
The
following
paragraphs
briefly
discuss
testing
of
both
linear
and
log­
linear
forms
of
the
model.
Parameter
estimates
are
presented
for
the
final
log­
linear
model
only
because
this
specification
appeared
to
be
superior
to
a
linear
model.

D.
4.1
Linear
Model
Specification
EPA
first
tested
linear
models
of
CAPEX
as
a
function
of
the
proposed
explanatory
variables.
In
testing
linear
models
of
CAPEX,
EPA
tested
a
number
of
structural
modifications
within
the
overall
hypothesized
framework
of
explanatory
variables.
These
included:

 
Testing
the
influence
of
industry
classification
on
the
estimation
of
the
coefficients
for
certain
of
the
explanatory
variables:
e.
g.,
using
the
product
of
an
industry
classification
dummy
variable
and
CAPPRC
to
test
whether
certain
industries
in
particular,
 
high­
tech 
vs.
 
traditional 
industries
responded
differently
to
change
in
price
of
capital
equipment
over
time.

 
Testing
contemporary
vs.
lagged
specification
of
certain
explanatory
variables:
e.
g.,
using
prior,
instead
of
current,
period
revenue,
REV,
as
an
explanatory
variable.

 
Testing
scale­
normalized
specification
of
the
dependent
variable:
e.
g.,
using
CAPEX/
REV
as
the
dependent
variable
instead
of
simple
CAPEX.

 
Testing
flexible
functional
forms
that
included
quadratic
terms.

 
Testing
additional
explanatory
variables
including
the
index
of
10
leading
economic
indicators
(
ILI)
and
market­

to­
book
ratio
(
MV/
B).

EPA
also
tested
the
data
for
autocorrelation,
heteroscedasticity,
and
multicollinearity.

Cross­
sectional,
time­
series
datasets
typically
exhibit
both
autocorrelation
and
group­
wise
heteroscedasticity
characteristics.

Autocorrelation
is
frequently
present
in
economic
time
series
data
as
the
data
display
a
 
memory 
with
the
variation
not
being
independent
from
one
period
to
the
next.
Heteroscedasticity
usually
occurs
in
cross­
sectional
data
where
the
scale
of
the
dependent
variable
and
the
explanatory
power
of
the
model
vary
across
observations.
Not
surprisingly,
the
dataset
used
in
this
analysis
had
both
characteristics.

The
collinearity
diagnostic
showed
that
several
independent
variables
are
collinear.
In
particular,
EPA
found
that
the
index
of
leading
economic
indicators
(
ILI)
and
the
price
of
capital
equipment
(
CAPPRC)
variables
are
highly
correlated.
EPA
further
found
that
the
market­
to­
book
ratio
variable
(
MV/
B)
was
highly
correlated
with
both
capital
turnover
(
CAPT)
and
return­
on­

assets
(
ROA)
variables.
To
address
the
multicollinearity
issue,
EPA
substituted
capacity
utilization
(
CAPUT
IL)
for
the
index
of
leading
economic
indicators
(
ILI)
and
dropped
the
market­
to­
book
ratio
(
MV/
B)
variable
in
the
final
model.

1
All
dollar
values
were
deflated
to
1996
(
base
year
for
MP&
M
regulatory
analysis)
using
2­
digit
SIC
PPI
values.

D­
9
MP&
M
EEBA:
Appendices
Appendix
D:
Estimating
Capital
Outlays
for
MP&
M
Discounted
Cash
Flow
Analyses
D.
4.2
Log­
Linear
Model
Specification
The
main
advantage
of
the
log­
linear
model
is
that
it
incorporates
directly
the
concept
of
percent
change
in
the
explanatory
variables.
Specifying
the
key
regression
variables
as
logarithms
permitted
us
to
estimate
directly
as
the
coefficients
of
the
model,
the
elasticities
of
capital
expenditures
with
respect
to
firm
financial
characteristics
and
general
business
environment
factors.
In
addition,
by
eliminating
the
need
to
use
percent
change
variables,
EPA
was
able
to
avoid
losing
early
year
observations
in
the
analysis
dataset.
Finally,
the
logarithmic
transformations
helped
to
reduce
outlier
effects
in
the
model.

EPA
specified
a
log­
linear
model,
as
follows:

ln(
CAPEXi,
t)
=
 
+
 
[
 
x
ln(
Xi,
t)]
+
 
[
 
y
ln(
Yt)]
+
 
W
here:

CAPEXi,
t
=
capital
expenditures
of
firm
i,
year
t;

 
x
=
elasticity
of
capital
expenditures
with
respect
to
firm
characteristic
X;

Xi,
t,
=
a
vector
of
financial
characteristics
of
firm
i,
year
t;

 
y
=
elasticity
of
capital
expenditures
with
respect
to
economic
indicator
Y;

Yt
=
a
vector
of
economic
indicators,
year
t;
for
CAPU
TIL,
Y
is
also
differentiated
by
industry
classification
 
=
an
error
term;
and
ln(
x)
=
natural
log
of
x
Based
on
this
model,
the
elasticity
of
capital
expenditures
with
respect
to
an
explanatory
variable,
for
example,
return
on
assets
is
calculated
as
follows:

Because
the
log­
linear
specification
incorporates
directly
the
concept
of
percent
change
in
the
explanatory
variables,
EPA
dropped
the
 
change 
specification
variables
i.
e.,
revenue
growth
(
REVGR),
year­
to­
year
change
in
the
Index
of
Leading
Indicators
(
ILIGR),
and
year­
to­
year
change
in
the
Capital
Equipment
Price
Index
(
CAPPRC)
from
the
analysis.
For
these
variables,
EPA
used
the
logarithm
of
the
simple,
unadjusted
values
in
the
log­
linear
specification.

One
disadvantage
of
the
specified
log­
linear
model
is
that
the
logarithmic
transformation
is
not
feasible
for
negative
and
zero
values.
This
means
that
negative
and
zero
values
require
linear
transformation
to
be
included
in
the
analysis.
The
following
variables
in
the
sample
required
transformation:

 
CAPEX:
four
firms
in
the
sample
reported
zero
capital
expenditures
at
least
in
one
time
period.
EPA
set
these
expenditures
to
$
1.

 
REVENUE:
one
firm
reported
negative
revenue
(­$
1,018)
in
one
time
period.
Because
this
is
likely
due
to
accounting
adjustments
from
prior
period
reporting,
EPA
set
the
firm s
revenue
in
the
current
time
period
to
$
1.

 
ROA:
the
values
for
return
on
assets
in
the
public
firm
sample
range
from
­
1.1
to
0.6.
Approximately
25
percent
of
the
firms
in
the
dataset
reported
negative
ROAs
in
at
least
one
year.
To
address
this
issue
while
reducing
potential
effects
of
data
transformation
on
the
modeling
results,
EPA
used
the
following
data
transformation
approach:

R
EPA
excluded
12
firms
with
any
annual
ROA
values
below
the
99th
percentile
of
the
ROA
distribution
(
i.
e.,

ROA
 
­
0.31).

R
EPA
used
an
additive
data
transformation
to
ensure
that
remaining
negative
ROA
values
were
positive
in
the
logarithm
transformation.
The
additive
transformation
was
performed
by
adding
0.31
to
all
ROA
values.

D­
10
MP&
M
EEBA:
Appendices
Appendix
D:
Estimating
Capital
Outlays
for
MP&
M
Discounted
Cash
Flow
Analyses
The
analysis
tested
several
specifications
of
a
log­
linear
model,
including
models
with
slope
dummies
for
different
industrial
sectors
and
models
with
the
intercept
suppressed.
The
model
presented
below
was
most
successful
at
explaining
firms 

investment
behavior.

EPA
estimated
the
specified
model
using
the
generalized
least
squares
procedure.
This
procedure
involves
the
following
two
steps:

 
First,
EPA
estimated
the
model
using
simple
OLS,
ignoring
autocorrelation
for
the
purpose
of
obtaining
a
consistent
estimator
of
the
autocorrelation
coefficient
(
 
)
;

 
Second,
EPA
used
the
generalized
least
squares
procedure,
where
the
analysis
is
applied
to
transformed
data.

The
resulting
autocorrelation
adjustment
is
as
follows:

Zi,
t
=
Zi,
t
­
 
Zi,
t­
1
where
Zit
is
either
dependent
or
independent
variables.

EPA
was
unable
to
correct
the
estimated
model
for
group­
wise
heteroscedasticity
due
to
computational
difficulties.
The
statistical
software
used
in
the
analysis
(
LIMDEP)
failed
to
correct
the
covariance
matrix
due
to
the
very
large
number
of
groups
(
i.
e.,
390
firms)
included
in
the
dataset.
Application
of
other
techniques
to
correct
for
group­
wise
heteroscedasticity
was
not
feasible
due
to
time
constraints.
The
estimated
coefficients
remain
unbiased;
however,
they
are
not
minimum
variance
estimators.

Table
D.
4
presents
model
results.
The
model
has
a
fairly
good
fit,
with
adjusted
R2
of
0.89.
All
coefficients
have
the
expected
sign
and
all
but
two
(
constant
and
capital
price)
are
significantly
different
from
zero
at
the
95th
percentile.

Table
D.
4:
ime
Series,
Cross­
Sectional
Model
Results
Variable
Coefficient
t­
Statistics
Constant
­
2.077
­
0.97
Ln(
ROA)
0.618
9.353
Ln(
REV)
1.025
113.867
Ln(
CAPT)
0.6
20.285
Ln(
CAPI)
0.976
27.342
Ln(
DEBTCST)
­
0.205
­
2.653
Ln(
CAPPRC)
­
0.478
­
0.939
Ln(
CAPUTIL)
0.904
3.176
Autocorrelation
Coefficient
r
0.413
27.842
T
The
empirical
results
show
that
the
output
variable
(
REV)
is
a
dominant
determinant
of
firms 
investment
spending.
A
positive
coefficient
on
this
variable
means
that
larger
firms
invest
more,
all
else
equal,
which
is
clearly
a
simple
expected
result.
Very
important
for
the
MP&
M
analysis,
as
expected,
firms
with
higher
financial
performance
and
better
investment
opportunities
(
ROA)
invest
more,
all
else
equal:
for
each
one
percent
increase
in
ROA,
a
firm
is
expected
to
increase
its
capital
outlays
by
0.62
percent.
Other
firm­
specific
characteristics
were
also
found
important
and
will
aid
in
differentiating
the
expected
capital
outlay
for
MP&
M
facilities
according
to
firm­
specific
characteristics.
Firms
that
require
more
capital
to
produce
a
given
level
of
business
activity
(
i.
e.,
firms
that
have
high
capital
intensity,
CAPI)
tend
to
invest
more:
a
one
percent
increase
in
capital
intensity
leads
to
a
0.98
increase
in
capital
spending.
Higher
capital
turnover/
shorter
capital
life
(
CAPT)

also
has
a
positive
effect
on
investment
decisions:
a
one
percent
increase
in
capital
turnover
rate
translates
to
a
0.60
percent
in
capital
outlays.

The
model
also
shows
that
current
business
environment
conditions
play
an
important
role
in
firms 
decision
to
invest.
The
D­
11
MP&
M
EEBA:
Appendices
Appendix
D:
Estimating
Capital
Outlays
for
MP&
M
Discounted
Cash
Flow
Analyses
most
influential
factor
is
capacity
utilization
in
manufacturing
facilities.
A
one
percent
increase
in
the
Federal
Reserve
Index
of
Capacity
Utilization
for
the
relevant
industrial
sector
(
CAPUTIL)
leads
to
a
0.90
percent
increase
in
capital
investment.

Negative
signs
on
the
debt
cost
(
DEBTCST)
and
capital
price
(
CAPPRC)
variables
match
expectations,
indicating
that
less
costly
credit
and
falling
(
either
relatively
or
absolutely)
capital
equipment
prices
are
likely
to
have
a
positive
effect
on
firms 

capital
expenditures.
That
these
systematic
variables
are
significant
in
the
regression
analysis
means
that
EPA
will
be
able
to
control
for
economy­
and
industry­
wide
conditions
in
estimating
capital
outlays
for
MP&
M
facilities.

D.
4.3
Sensitivity
Analysis
To
examine
the
degree
to
which
the
estimated
model
was
affected
by
transformation
of
ROA
values
and
inclusion/
exclusion
of
firms
with
the
lowest
ROA
values,
EPA
ran
two
additional
models.
First,
EPA
estimated
a
model
based
on
a
subset
of
data
that
includes
only
firms
with
positive
ROA
values.
Second,
EPA
estimated
a
model
based
on
a
complete
dataset
that
includes
the
12
firms
with
the
lowest
ROA
values.
Although
all
three
models
produced
compatible
results,
the
first
model
shows
some
notable
differences
in
the
estimated
coefficients
compared
to
the
model
presented
in
the
preceding
section.
EPA
found
that
when
firms
with
the
lowest
negative
ROAs
are
excluded
from
the
analysis:

 
The
magnitude
of
the
ROA
effect
on
capital
expenditures
decreases;

 
The
magnitude
of
the
debt
cost
effect
on
capital
expenditures
decreases
slightly;

 
The
coefficient
on
the
capital
price
term
becomes
significant.

These
differences
can
be
expected
since
firms
with
negative
ROAs
are
weak
performers
and
therefore
are
less
likely
to
have
large
capital
outlays.
Not
surprisingly,
general
economic
indicators
that
affect
firms 
decisions
to
invest
can
be
less
or
more
important
if
a
firm s
financial
performance/
asset
productivity
is
weak.
For
financially
weaker
firms,
the
financial
cost
of
capital
is
a
more
important
factor
compared
to
firms
that
are
strong
financially.
This
finding
indicates
a
strong
 
threshold
of
adequate
financial
performance 
effect:
capital
outlays
fall
off
severely
at
the
lowest
financial
performance
levels
but
the
marginal
effect
of
financial
performance
becomes
more
moderate
as
asset
productivity
moves
into
a
more
acceptable
i.
e.,

positive
return
range.
Price
of
capital
goods
appears
to
be
an
insignificant
factor
in
firms 
decision
to
invest
when
weak
firms
are
included
in
the
analysis.
At
first,
this
finding
seems
to
be
counterintuitive:
previous
studies
of
investment
behavior
found
a
strong
capital
price
effect
on
firms 
decision
to
invest
in
high
tech
equipment.
However,
because
financially
weak
firms
are
less
likely
to
invest
in
general,
it
is
reasonable
to
assume
that
they
will
not
respond
as
strongly
to
changes
in
capital
equipment
prices.
Thus,
their
investment
decisions
were
relatively
less
affected
by
falling
high­
tech
equipment
prices
in
the
last
decade.

D.
5
MODEL
VALIDATION
To
validate
the
results
of
the
regression
analysis,
EPA
used
the
estimated
regression
equation
to
calculate
capital
expenditures
and
then
compared
the
resulting
estimate
of
capital
expenditures
with
actual
data.
EPA
used
two
methods
to
validate
its
results:

 
EPA
used
median
values
for
explanatory
variable
from
the
Value
Line
data
as
input
to
estimate
capital
expenditures
and
then
compared
the
estimated
value
to
the
median
reported
capital
expenditures,
and
 
EPA
used
MP&
M
survey
data
to
estimate
capital
expenditures
and
then
compared
the
estimated
values
to
depreciation
reported
in
the
survey.

First,
EPA
estimated
capital
expenditures
for
a
hypothetical
firm
based
on
the
median
values
of
the
four
dependent
variables
from
the
Value
Line
data
and
the
relevant
values
of
the
three
economic
indicators.
The
estimated
capital
expenditures
for
this
hypothetical
firm
are
$
10.9
million.
EPA
then
compared
this
estimate
to
the
median
value
of
capital
expenditures
from
the
Value
Line
data.
The
median
capital
expenditure
value
in
the
dataset
is
$
11.3
million,
which
provides
a
very
close
match
to
the
estimated
value.
This
is
not
surprising
since
the
same
dataset
was
used
to
estimate
the
regression
model
and
to
calculate
the
median
values
used
in
this
analysis.

EPA
also
used
MP&
M
survey
data
to
confirm
that
the
estimated
capital
expenditures
seem
reasonable.
Because
the
MP&
M
survey
does
not
provide
information
on
capital
expenditures,
EPA
compared
the
capital
expenditure
estimates
to
the
D­
12
MP&
M
EEBA:
Appendices
Appendix
D:
Estimating
Capital
Outlays
for
MP&
M
Discounted
Cash
Flow
Analyses
depreciation
values
reported
in
the
survey.
Depreciation
had
been
proposed
as
a
possible
surrogate
for
cash
outlays
for
capital
replacements
and
additions.
However,
depreciation
does
not
capture
important
variations
in
capital
outlays
that
result
from
differences
in
firms 
financial
performance.

For
this
analysis,
EPA
chose
a
representative
facility
from
each
of
the
nineteen
MP&
M
sectors
for
model
validation.
The
selected
facility
for
each
sector
corresponds
as
closely
as
possible
to
the
hypothetical
median
facility
in
the
sector
based
on
the
distribution
of
facility
revenues
and
facility
return
on
assets.
For
each
of
the
nineteen
facilities,
EPA
estimated
capital
expenditures
using
the
estimated
regression
equation
and
facility
financial
data.
Table
D.
5
shows
the
estimated
regression
coefficients,
financial
averages
for
the
nineteen
MP&
M
sectors,
estimated
facility
capital
expenditures,
reported
facility
depreciation,
and
the
comparison
of
capital
expenditures
and
depreciation.

As
shown
in
Table
D.
5,
the
estimated
model
provides
reasonable
estimates
of
capital
expenditures.
A
facility s
size,
as
indicated
by
revenue,
is
a
principal
determinant
of
the
general
range
of
value
for
capital
expenditures,
all
else
equal
(
i.
e.,

greater
revenues
correspond
to
greater
predicted
capital
expenditures).
However,
the
size
of
capital
expenditures
relative
to
the
depreciation
allowance
depends
substantially
on
a
facility s
return
on
assets.
Facilities
with
lower
return
on
assets
tend
to
invest
less
than
indicated
by
depreciation
while
facilities
with
higher
return
on
assets
tend
to
invest
more
than
depreciation.

This
finding
is
consistent
with
the
expectation
that
businesses
with
higher
financial
performance
will
have
relatively
more
attractive
investment
opportunities
and
are
more
likely
to
attract
the
capital
to
undertake
those
investments.
To
highlight
this
relationship
between
capital
expenditure,
depreciation
allowance,
and
a
facility s
return
on
assets,
EPA
presents
graphs
for
the
Hardware,
Iron
&
Steel,
Job
Shops,
and
Printed
Circuit
Board
sectors
that
plot
MP&
M
survey
facilities
in
these
sectors
along
with
linear
trend
lines
for
each
sector s
depreciation
and
capital
expenditures
with
respect
to
return
on
assets.
4
4
For
presentation
purposes,
some
outlier
facilities
were
excluded
from
the
graphs.

D­
13
MP&
M
EEBA:
Appendices
Appendix
D:
Estimating
Capital
Outlays
for
MP&
M
Discounted
Cash
Flow
Analyses
Table
D.
5:
ted
by
Revenue
and
ROA
Percentiles
Sectors
Pre­
Tax
Return
on
Assets
(
ROA)
Revenue
Capital
Turnover
Rate
Capital
Intensity
Cost
of
Debt
Price
of
Capital
Goods
Capacity
Utilization
Estimated
Capital
Expenditures
Depreciation
Difference
between
Depreciation
and
Capital
Expenditures
Coefficient
Intercept
(­
2.077)
0.62
1.03
0.60
0.98
(
0.21)
(
0.48)
0.90
Aerospace
0.02
90.66
0.02
1.29
7.11
135.4
73.67
2,113,741
1,821,434
­
0.14
Aircraft
0.05
18.39
0.06
0.54
9.8
115.87
80.01
440,385
558,478
0.27
Bus
&

Truck
0.06
58.09
0.03
0.25
7.11
135.4
73.69
471,199
503,124
0.07
Electronic
Equipment
0.05
36.85
0.12
0.4
7.11
135.4
86.37
1,100,627
1,730,023
0.57
Hardware
0.03
11.99
0.06
0.61
9.8
115.87
81.93
311,085
403,535
0.3
Household
Equipment
0.05
18
0.05
0.8
7.11
135.4
84.24
624,804
745,476
0.19
Instruments
0.15
62.47
0.04
0.47
7.11
135.4
77.21
1,195,144
1,139,873
­
0.05
Iron
&
Steel
0.12
23.17
0.06
0.47
6.4
136.9
90.82
617,740
613,834
­
0.01
Job
Shop
0.03
2
0.07
0.26
7.11
135.4
81.92
25,146
37,250
0.48
Mobile
Industrial
Equipment
0.07
37.6
0.03
0.63
9.8
115.87
79.45
670,447
586,609
­
0.13
Motor
Vehicle
0.1
104.44
0.06
0.46
7.11
135.4
81.24
2,473,215
2,810,386
0.14
Office
Machine
0.1
28.95
0.06
0.43
7.11
135.4
85.02
661,715
748,972
0.13
Ordnance
0.05
27.08
0.04
0.65
9.8
115.87
79.77
674,446
770,051
0.14
Other
Metal
Products
0.08
27.78
0.17
0.44
7.11
135.4
80.01
1,100,691
2,034,831
0.85
Precious
Metals
&

Jewelry
0.04
13.5
0.03
0.62
7.11
135.4
77.21
224,438
226,708
0.01
Estimation
of
Capital
Outlays
for
MP&
M
Sample
Facilities:
Median
Facilities
Selec
For
facilities
that
responded
to
the
Phase
1
survey,
EPA
calculated
a
3­
year
average
of
the
non­
facility
specific
information
over
the
years
in
which
survey
data
were
collected
(
1987­
1989).
Likewise,
for
facilities
that
responded
to
the
Phase
2
survey,
EPA
calculated
a
3­
year
average
of
the
non­
facility
specific
information
for
the
years
1994­
1996.
Since
the
Iron
and
Steel
sector
was
surveyed
in
1997,
EPA
calculated
a
3­
year
average
of
the
non­
facility
specific
information
for
the
years
1995­
1997.

Source:
U.
S.
EPA
analysis
D­
14
MP&
M
EEBA:
Appendices
Appendix
D:
Estimating
Capital
Outlays
for
MP&
M
Discounted
Cash
Flow
Analyses
Figure
D.
1:
Comparison
of
Estimated
Capital
Outlays
to
Reported
Depreciation
for
MP&
M
Survey
Facilities
in
the
Hardware
Sector
Source:
U.
S.
EPA
analysis.

Figure
D.
2:
Comparison
of
Estimated
Capital
Outlays
to
Reported
Depreciation
for
MP&
M
Survey
Facilities
in
the
Iron
&
Steel
Sector
Source:
U.
S.
EPA
analysis.

D­
15
MP&
M
EEBA:
Appendices
Appendix
D:
Estimating
Capital
Outlays
for
MP&
M
Discounted
Cash
Flow
Analyses
Figure
D.
3:
Comparison
of
Estimated
Capital
Outlays
to
Reported
Depreciation
for
MP&
M
Survey
Facilities
in
the
Job
Shop
Sector
Source:
U.
S.
EPA
analysis.

Figure
D.
4:
Comparison
of
Estimated
Capital
Outlays
to
Reported
Depreciation
for
MP&
M
Survey
Facilities
in
the
Printed
Circuit
Board
Sector
Source:
U.
S.
EPA
analysis.

D­
16
MP&
M
EEBA:
Appendices
Appendix
D:
Estimating
Capital
Outlays
for
MP&
M
Discounted
Cash
Flow
Analyses
ATTACHMENT
D.
A:
BIBLIOGRAPHY
OF
LITERATURE
REVIEWED
FOR
THIS
ANALYSIS
As
noted
above,
EPA
relied
on
previous
studies
of
investment
behavior
to
select
critical
determinants
of
firms 
capital
expenditures.
Empirical
results
from
these
studies
suggest
that
investment
is
most
sensitive
to
quantity
variables
(
output
or
sales),
return­
over­
cost,
and
capital
utilization
(
R.
Chirinko).
Empirical
results
from
more
recent
studies
further
found
that
increasing
depreciation
rates
and
capital
equipment
prices
were
of
first­
order
importance
in
the
equipment
investment
behavior
in
the
1990
(
T.
Tevlin,
K.
Whelan).
Specifically,
declining
prices
of
micro­
processor
based
equipment
played
a
crucial
role
in
the
investment
boom
in
the
1990.

Chirinko,
Robert
S.
1993.
 
Business
Fixed
Investment
Spending:
A
Critical
Survey
of
Modeling
Strategies,
Empirical
Results
and
Policy
Implications. 
Journal
of
Economic
Literature
31,
no.
4:
1875­
1911.

Goolsbee,
Austan.
1997.
 
The
Business
Cycle,
Financial
Performance,
and
the
Retirement
of
Capital
Goods. 
University
of
Chicago,
Graduate
School
of
Business
Working
Paper.

Greenspan,
Alan.
2001.
 
Economic
Developments. 
Remarks
before
the
Economic
Club
of
New
York,
New
York,
May
24.

Kiyotaki,
Nobuhiro
and
Kenneth
D.
West.
1996.
 
Business
Fixed
Investment
And
The
Recent
Business
Cycle
In
Japan. 

National
Bureau
of
Economic
Research
Working
Paper
5546.

McCarthy,
Jonathan.
2001.
 
Equipment
Expenditures
since
1995:
The
Boom
and
the
Bust. 
Current
Issues
In
Economics
And
Finance
7,
no.
9:
1­
6.

Opler,
Tim
and
Lee
Pinkowitz,
Rene
Stulz
and
Rohan
Williamson.
1997.
 
The
Determinants
and
Implications
of
Corporate
Cash
Holdings. 
Working
paper,
Ohio
State
University
College
of
Business.

Tevlin,
Stacey
and
Karl
Whelan.
2000.
 
Explaining
the
Investment
Boom
of
the
1990s. 
Board
of
Governors
of
the
Federal
Reserve
System
Finance
and
Economics
Discussion
Paper
no.
2000­
11
Uchitelle,
Louis.
2001.
 
Wary
Spending
by
Companies
Cools
Economy. 
New
York
Times,
May
14,
p.
A1.

D­
17
MP&
M
EEBA:
Appendices
Appendix
D:
Estimating
Capital
Outlays
for
MP&
M
Discounted
Cash
Flow
Analyses
ATTACHMENT
D.
B:
HISTORICAL
VARIABLES
CONTAINED
IN
THE
VALUE
LINE
INVESTMENT
SURVEY
DATASET
All
variables
are
provided
for
10
years
(
except
where
a
firm
has
been
publicly
reported
for
less
than
10
years):

 
Price
of
Common
Stock
 
Revenues
 
Operating
Income
 
Operating
Margin
 
Net
Profit
Margin
 
Depreciation
 
Working
Capital
 
Cash
Flow
per
share
 
Dividends
Declared
per
share
 
Capital
Spending
per
share
 
Revenues
per
share
 
Average
Annual
Price­
Earnings
Ratio
 
Relative
Price­
Earnings
Ratio
 
Average
Annual
Dividend
 
Return
Total
Capital
 
Return
Shareholders
Equity
 
Retained
To
Common
Equity
 
All
Dividends
To
Net
Worth
 
Employees
 
Net
Profit
 
Income
Tax
Rate
 
Earnings
Before
Extras
 
Earnings
per
share
 
Long
Term
Debt
 
Total
Loans
 
Total
Assets
 
Preferred
Dividends
 
Common
Dividends
 
Book
Value
 
Book
Value
per
share
 
Shareholder
Equity
 
Preferred
Equity
 
Common
Shares
Outstanding
 
Average
Shares
Outstanding
 
Beta
 
Alpha
 
Standard
Deviation
D­
18
MP&
M
EEBA:
Appendices
Appendix
E:
Calculation
of
Capital
Cost
Components
Appendix
E:
Calculation
of
Capital
Cost
Components
INTRODUCTION
APPENDIX
CONTENTS
E.
1
Calculation
of
One­
Time
Capital
Cost
Estimates
.........
E­
1
E.
1
CALCULATION
OF
ONE­
TIME
CAPITAL
COST
COMPONENTS
EPA
used
the
engineering
estimates
of
total
one­
time
capital
costs
to
calculate
the
purchase
cost
paid
to
manufacturers
of
compliance
equipment,
and
the
costs
of
shipping,
installation,
insurance,
engineering,
and
consultants.
Two
components
of
capital
costs
were
used
to
estimate
job
gains
due
to
compliance
requirements:
(
1)
the
estimated
direct
capital
equipment
cost
and
(
2)
the
labor
cost
of
installation.
Table
E.
1
shows
the
cost
components
that
comprise
the
total
capital
costs
attributed
to
the
regulation.

Table
E.
1:
Components
of
Total
Installed
Capital
Costs
(
millions,
2001$,
before
tax)
a
Cost
Component
Option
I:

Selected
Option
Option
II:

Proposed/

NODA
Option
Option
III:

413
to
433
Upgrade
Option
Option
IV:

All
to
433
Upgrade
Option
(
a)
Total
installed
direct
capital
costs
$
4,407,590
$
802,051,833
$
95,552,532
$
148,434,303
(
b)
Direct
capital
equipment
cost
$
3,070,680
$
558,773,471
$
66,569,538
$
103,411,210
(
c)
Shipping
(
20%
of
a)
$
881,518
$
160,410,367
$
19,110,506
$
29,686,861
(
d)
Labor
cost
of
installation
(
7%
of
f)
$
455,392
$
82,867,995
$
9,872,488
$
15,336,232
(
e)
Indirect
costs:
insurance,
engineering
&

consultants
(
47.6%
of
a)
$
2,098,013
$
381,776,672
$
45,483,005
$
70,654,728
(
f)
Total
installed
capital
costs
$
6,505,602
$
1,183,828,505
$
141,035,538
$
219,089,032
a
Excludes
costs
for
baseline
and
regulatory
closures.

Source:
U.
S.
EPA
analysis.

The
components
of
total
capital
costs
for
the
final
rule
in
Table
E.
1
are
discussed
below
in
reverse
order
of
the
table
presentation.

 
Total
installed
capital
costs:
EPA
estimated
the
total
one­
time
capital
cost
for
each
facility
expected
to
comply
with
the
regulation.
1
Compliance
costs
are
discussed
in
more
detail
in
Chapter
5:
Facility­
Level
Impact
Analysis
of
this
EEB
A.
The
national
estimate
of
capital
costs
for
the
regulation
is
$
6.5
million
($
2001).
2
1
See
the
Technical
Development
Document
for
a
description
of
the
methods
used
to
estimate
capital
costs.

2
The
$
6.5
million
is
the
sum
of
one­
time
outlays
for
purchasing
and
installing
the
capital
equipment
needed
to
comply
with
the
final
rule.
This
expense
is
not
the
annual
equivalent
of
that
capital
investment.
The
capital
outlay
is
annualized
in
the
economic
impact
analysis
over
a
15­
year
period.
The
resulting
value,
which
is
part
of
the
total
annual
cost
of
compliance,
is
$
0.7
million.

E­
1
MP&
M
EEBA:
Appendices
Appendix
E:
Calculation
of
Capital
Cost
Components
 
Indirect
Costs:
MP&
M
project
engineers
estimate
that
indirect
costs
such
as
insurance,
engineering,
and
consulting
are
47.6%
of
installed
direct
capital
cost.
EPA
calculated
the
total
direct
and
indirect
cost
using
the
total
capital
cost.

The
national
estimate
of
indirect
costs
for
the
regulation
is
$
2.1
million.

 
Total
Installed
Direct
Capital
Costs:
The
direct
capital
costs
include
the
cost
of
compliance
equipment,
shipping,

and
the
labor
cost
of
installation.
The
national
estimate
of
direct
costs
for
the
regulation
is
$
4.4
million.
MP&
M
project
engineers
estimate
that
shipping
costs
might
be
as
much
as
20
percent
of
the
total
installed
direct
capital
cost.

The
estimated
one­
time
shipping
cost
is
$
0.9
million
for
the
final
regulatory
option.
Installation
labor
costs
are
estimated
by
the
engineers
to
be
seven
percent
of
the
total
installed
capital
costs.
The
estimated
one­
time
cost
of
installation
labor
is
$
0.5
million
for
the
final
regulatory
option.
Therefore,
the
direct
capital
equipment
cost
is
$
3.1
million,
the
remainder
of
the
total
installed
direct
capital
cost
when
the
cost
of
shipping
and
installation
are
subtracted
out.

E­
2
MP&
M
EEBA:
Appendices
Appendix
F:
Administrative
Costs
Appendix
F:
Administrative
Costs
INTRODUCTION
Effluent
guidelines
and
limitations
are
implemented
by
Federal,
State,
and
local
government
entities
through
the
NPDES
permit
program
(
for
direct
dischargers)
and
the
General
Pretreatment
Regulations
(
for
indirect
dischargers).
A
new
effluent
guideline
rule
may
require
that
facilities:
(
1)
be
permitted
for
the
first
time;
(
2)
be
issued
a
different
form
of
permit,
if
they
already
have
a
permit
in
the
baseline;
and
(
3)
be
repermitted
sooner
than
would
otherwise
be
required.
In
these
cases,
the
permitting
authority
will
incur
additional
costs
to
implement
the
effluent
guideline
rule.

This
appendix
provides
information
on
the
unit
costs
of
these
permitting
activities
and
describes
the
calculation
of
government
permitting
costs
for
the
final
MP&
M
rule
and
regulatory
alternatives.
EPA
expects
no
additional
costs
for
permitting
direct
dischargers
under
the
final
rule.
Costs
for
issuing
permits
for
indirect
dischargers
are
based
on
information
reported
by
publicly­
owned
treatment
works
(
POTWs)
in
the
Metal
Products
and
Machinery
(
MP&
M)

POT
W
Survey.
EPA
also
used
the
data
provided
in
the
Association
of
Metropolitan
Sewerage
Agencies
(
AMSA)

survey
to
supplement
information
from
the
MP&
M
POTW
Survey.
EPA
evaluated
POTW
administrative
costs
for
pretreatment
options
for
the
final
rule.
As
discussed
in
Section
VI
of
the
preamble
to
the
final
rule,
EPA
is
not
establishing
any
pretreatment
standards
in
the
final
rule.
APPENDIX
CONTENTS
F.
1
luent
Guidelines
Permitting
Requirements
......
F­
1
F.
1.1
ermit
Program
....
F­
1
F.
1.2
.................
..
F­
2
F.
2
TW
Administrative
Cost
Methodology
........
F­
2
F.
2.1
.................
........
F­
2
F.
2.2
Overview
of
Methodology
...............
F­
3
F.
3
..............
F­
4
F.
3.1
..........
F­
4
F.
3.2
ion
.................
...........
F­
7
F.
3.3
.................
..........
F­
7
F.
3.4
ement
.................
.........
F­
9
F.
3.5
itting
.................
........
F­
10
F.
4
dministrative
Costs
by
Option
.........
F­
10
Appendix
F
Exhibits
.................
...........
F­
12
References
.................
.................
.
F­
25
Eff
NPDES
Basic
Industrial
P
Pretreatment
Program
PO
Data
Sources
Unit
Costs
of
Permitting
Activities
Permit
Application
and
Issuance
Inspect
Monitoring
Enforc
Reperm
POTW
A
The
remainder
of
this
appendix
is
organized
as
follows:
Section
F.
1
provides
an
overview
of
permitting
requirements
under
the
NPDES
Permit
Program
and
the
General
Pretreatment
Regulations.
Section
F.
2
describes
the
MP&
M
POTW
Survey
and
the
methods
used
to
develop
annualized
cost
estimates
for
permitting
indirect
dischargers.
Section
F.
3
presents
the
estimates
of
unit
costs
by
permitting
activity
for
indirect
dischargers.
The
final
Section
F.
4
lists
the
steps
involved
in
applying
these
unit
costs
to
calculate
administrative
costs
for
regulatory
options
evaluated
by
EPA
for
the
final
rule.

F.
1
EFFLUENT
GUIDELINES
PERMITTING
REQUIREMENTS
Any
facility
that
directly
discharges
wastewater
to
surface
water
is
required
to
have
a
permit
issued
under
the
National
Pollution
Discharge
Elimination
System
(
NPDES)
permit
program.
Facilities
that
discharge
indirectly
through
a
POTW
are
regulated
by
the
General
Pretreatment
Regulations
for
Existing
and
New
Sources
of
Pollution
(
40
CFR
Part
403).
The
major
portion
of
government
administrative
costs
associated
with
implementing
an
effluent
guidelines
rule
are
the
costs
of
managing
the
NPDES
and
Pretreatment
programs.

F.
1.1
NPDES
Basic
Industrial
Permit
Program
Best
Practical
Technology
(
BPT),
Best
Control
Technology
(
BCT),
and
New
Source
Performance
Standards
(
NSPS)
for
effluent
limitations
guidelines
are
implemented
through
the
NPDES
industrial
permit
program.
However,
EPA
does
not
expect
the
administrative
costs
associated
with
the
NPDES
industrial
permit
program
to
increase
as
a
result
of
the
final
rule.

The
Clean
Water
Act
prohibits
discharge
of
any
pollutant
to
a
water
of
the
U.
S.
except
as
permitted
by
a
NPDES
permit.

Therefore,
every
facility
that
discharges
wastewater
directly
to
surface
water
must
hold
a
permit
specifying
the
mass
of
pollutants
that
can
be
discharged
to
waterways.
The
final
rule
will
affect
the
terms
of
the
permits
but
is
unlikely
to
increase
the
administrative
costs
associated
with
permitting.

The
final
rule
may
decrease
the
administrative
burden
of
NPDES
permits.
The
TDD
and
rulemaking
record
for
the
final
rule
F­
1
MP&
M
EEBA:
Appendices
Appendix
F:
Administrative
Costs
provide
valuable
information
to
permitting
authorities
that
may
reduce
the
research
required
to
develop
Best
Professional
Judgment
(
BPJ)
permits.
1
Further,
establishing
discharge
standards
may
reduce
time
spent
by
permitting
authorities
establishing
limits
and
the
frequency
of
evidentiary
hearings.
The
promulgation
of
limitations
may
also
enable
EPA
and
the
authorized
States
to
cover
more
facilities
under
general
permits.
General
permits
are
single
permits
covering
a
common
class
of
dischargers
in
a
specified
geographic
area.

F.
1.2
Pretreatment
Program
The
General
Pretreatment
Regulations
(
40
CFR
Part
403)
establish
procedures,
responsibilities,
and
requirements
for
EPA,

States,
local
governments,
and
industry
to
control
pollutant
discharges
to
POTWs.
Under
the
Pretreatment
Regulations,

POTWs
or
approved
States
implement
categorical
pretreatment
standards
for
existing
sources
(
PSES)
and
new
sources
(
PSNS).

Discharges
from
an
MP&
M
facility2
to
a
POTW
may
already
be
permitted
in
the
baseline.
3
For
example,
industrial
users
subject
to
another
Categorical
Pretreatment
Standard
would
have
a
discharge
permit.
Other
significant
industrial
users
(
SIU)

that
are
typically
permitted
by
POTWs
include
industrial
users
that:

 
discharge
an
average
of
25,000
gallons
per
day
or
more
of
process
wastewater
to
a
POTW,

 
contribute
a
process
waste
stream
that
makes
up
five
percent
or
more
of
the
average
dry
weather
hydraulic
or
organic
capacity
of
the
POT
W
treatment
plant,
or
 
have
a
reasonable
potential
for
adversely
affecting
the
POTW
 
s
operation
or
for
violating
any
pretreatment
standard.

As
discussed
in
Section
VI
of
the
preamble
to
the
final
rule,
EPA
did
not
establish
or
revise
any
pretreatment
standards
in
the
final
rule.
Consequently,
there
are
no
POTW
administrative
costs
associated
with
the
final
rule.
Under
the
options
evaluated
for
the
final
rule,
which
include
options
for
setting
pretreatment
standards,
EPA
expects
no
increase
in
permitting
costs
for
indirect
dischargers
that
already
hold
a
permit
in
the
baseline.
However,
governments
will
incur
additional
permitting
costs
for
unpermitted
facilities
(
under
the
NODA/
Proposal
option
only)
and
to
accelerate
repermitting
for
some
indirect
dischargers
that
currently
hold
permits.
The
remainder
of
this
appendix
estimates
these
cost
increases.
As
with
direct
industrial
dischargers,
promulgation
of
the
MP&
M
rule
may
cause
some
administrative
costs
to
decrease.
For
example,
control
authorities
will
no
longer
have
to
repermit
facilities
that
are
estimated
to
close
as
a
result
of
some
of
the
options
EPA
evaluated
for
the
final
rule.
These
cost
savings
are
reflected
in
estimates
of
total
government
administrative
costs
associated
with
the
regulatory
options
considered
for
the
final
rule.

F.
2
POTW
ADMINISTRATIVE
COST
METHODOLOGY
F.
2.1
Data
Sources
EPA
collected
information
from
POTWs
to
support
development
of
the
MP&
M
effluent
guideline
(
see
Section
3
of
the
TDD).
Of
150
surveys
mailed,
EPA
received
responses
to
147,
for
a
98
percent
response
rate.
The
POTW
Survey
asked
respondents
to
provide
information
on
administrative
permitting
costs
for
indirect
dischargers,
sewage
sludge
use
and
disposal
costs
and
practices,
and
general
information
(
including
number
of
permitted
users
and
number
of
known
MP&
M
dischargers).

The
administrative
cost
information
included
the
number
of
hours
required
to
complete
specific
permitting
and
repermitting,

1
Permits
issued
to
facilities
not
covered
by
effluent
guidelines
or
water
quality­
based
standards
are
developed
based
on
BPJ
(
see
NPDES 
permit
writers
manual).

2
MP&
M
facilities
are
defined
on
the
basis
of
three
considerations:
(
1)
they
produce
metal
parts,
products,
or
machines
for
use
in
one
of
the
19
industry
sectors
evaluated
for
coverage
in
the
MP&
M
point
source
category;
(
2)
they
use
operations
in
one
of
the
eight
regulatory
subcategories
evaluated
for
coverage
in
the
MP&
M
point
source
category;
and
(
3)
they
discharge
process
wastewater,
either
directly
or
indirectly,
to
surface
waters.
In
this
document,
the
term
 
MP&
M
facilities 
refers
to
all
facilities
meeting
the
above
definition,
regardless
of
whether
a
facility s
industrial
sector,
subcategory,
or
discharger
category
is
covered
by
the
final
regulation.

3
Under
the
General
Pretreatment
Program,
a
facility's
discharges
may
be
controlled
through
a
"
permit,
order
or
similar
means".
For
simplicity,
this
document
refers
to
the
control
mechanism
as
a
permit.

F­
2
MP&
M
EEBA:
Appendices
Appendix
F:
Administrative
Costs
inspection,
monitoring,
and
enforcement
activities.
Respondents
were
also
asked
to
provide
an
average
labor
cost
for
all
staff
involved
in
permitting
activities.
EPA
used
the
survey
responses
on
administrative
costs
to
estimate
a
range
of
costs
incurred
by
POTWs
to
permit
a
single
MP&
M
facility.

The
Association
of
Metropolitan
Sewerage
Agencies
(
AMSA)
also
provided
data
on
administrative
costs
to
EPA
(
see
Section
3
of
the
TDD).
EPA
used
the
data
provided
in
the
AMSA
survey
to
verify
and,
in
some
cases,
supplement
its
own
analyses
of
POTW
administrative
costs
for
regulatory
options
evaluated
for
the
final
rule.
AMSA
provided
EPA
with
comments
on
the
proposed
MP&
M
rule
and
supplemented
these
comments
with
a
spreadsheet
database.
The
database
contains
data
from
an
AMSA
formulated
survey
and
covers
responses
from
176
POTW
s,
representing
66
pretreatment
programs.
The
AMSA
survey
was
conducted
to
verify
data
from
EPA's
survey
of
POTW
s
and
therefore
included
similar,
although
fewer,
variables
compared
to
EPA's
survey.
Elements
EPA
verified
using
the
AMSA
survey
include:
(
1)
the
estimated
number
of
indirect
dischargers;
and
(
2)
the
unit
costs
of
certain
permitting
activities,
including
permit
implementation,
sampling,
and
sample
analysis.
Elements
EPA
added
to
its
analysis
using
the
AMSA
data
include:
(
1)
screening
costs
for
POTW
s
that
do
not
currently
operate
under
a
pretreatment
program;
and
(
2)
oversight
costs
associated
with
implementing
the
MP&
M
regulation.

F.
2.2
Overview
of
Methodology
EPA
estimated
the
annualized
costs
of
permitting
indirect
dischargers
under
the
different
regulatory
options
evaluated
for
the
final
rule
using
the
following
steps:

 
Determine
the
number
and
characteristics
of
indirect
dischargers
that
will
be
permitted
under
each
regulatory
option
evaluated
for
the
final
rule.
Only
the
NODA
option
includes
costs
for
permitting
an
MP&
M
facility
for
the
first
time.
The
final
rule
does
not
cover
indirect
dischargers
while
the
other
regulatory
options
only
regulate
those
indirect
dischargers
that
already
hold
permits
in
the
baseline.
For
the
NODA
option,
EPA
determined
how
many
new
permits
would
be
issued.
The
NODA
option
requires
only
concentration­
based
permits,
no
mass­

based
permits.

 
Use
the
data
from
the
POTW
Survey
to
determine
a
high,
middle,
and
low
hourly
burden
for
permitting
a
single
facility.
EPA
defined
the
low
and
high
estimates
of
hours
such
that
90%
of
the
POTW
responses
fell
above
the
low
value
and
90%
of
responses
fell
below
the
high
value.
The
median
value
is
used
to
define
the
middle
hourly
burden.

 
Use
the
data
from
the
POTW
Survey
to
determine
the
average
frequency
of
performing
certain
administrative
functions.
For
administrative
functions
that
are
not
performed
at
all
facilities,
survey
data
were
used
to
calculate
the
portion
of
facilities
requiring
these
functions.
For
example,
the
survey
data
show
that
on
average
38.5%
of
facilities
submit
a
non­
compliance
report.

 
Multiply
the
per­
facility
burden
estimate
by
the
average
hourly
wage.
EPA
determined
a
high,
middle,
and
low
dollar
cost
of
administering
the
rule
for
a
single
facility
by
multiplying
the
per­
facility
hour
burden
by
the
average
hourly
wage.
The
POTW
Survey
reported
an
average
hourly
labor
rate
of
$
39.33
($
2001)
for
staff
involved
in
permitting.
This
is
a
fully­
loaded
cost,
including
salaries
and
fringe
benefits.

 
Calculate
the
annualized
cost
of
administering
the
rule.
The
number
of
facilities,
hourly
burden
estimate,

frequency
estimates,
and
hourly
wage
estimates
are
all
combined
to
determine
the
total
cost
of
administering
the
rule.

The
type
of
administrative
activities
required
varies
over
time
and
the
total
administrative
cost
is
calculated
over
a
15
year
time
period.
EPA
calculated
the
present
value
of
total
costs
using
a
seven
percent
discount
rate,
and
then
annualized
the
present
value
using
the
same
seven
percent
discount
rate.

F.
3
UNIT
COSTS
OF
PERMITTING
ACTIVITIES
This
section
presents
unit
costs
for
the
following
permitting
activities:

 
Permit
application
and
issuance:
developing
and
issuing
concentration­
based
permits
at
previously
unpermitted
facilities;
providing
technical
guidance;
and
conducting
public
and
evidentiary
hearings;

 
Inspection:
inspecting
facilities
both
for
the
initial
permit
development
and
to
assess
subsequent
compliance;

F­
3
MP&
M
EEBA:
Appendices
Appendix
F:
Administrative
Costs
 
Monitoring:
sampling
and
analyzing
permittee s
effluent;
reviewing
and
recording
permittee s
compliance
self­

monitoring
reports;
receiving,
processing,
and
acting
on
a
permittee s
non­
compliance
reports;
and
reviewing
a
permittee s
compliance
schedule
report
for
permittees
in
compliance
and
permittees
not
in
compliance;

 
Enforcement:
issuing
administrative
orders
and
administrative
fines;
and
 
Repermitting.

EPA
believes
that
theses
functions
constitute
the
bulk
of
the
required
administrative
activities.
To
these
costs,
EPA
added
a
provision
for
managerial
oversight
of
25
percent.
4
There
are
other
relatively
minor
or
infrequent
administrative
functions
(
e.
g.,
providing
technical
guidance
to
permittees
in
years
other
than
the
first
year
of
the
permit,
or
repermitting
a
facility
in
significant
non­
compliance),
but
their
costs
are
likely
to
be
insignificant
compared
to
the
estimated
costs
for
the
five
major
categories
outlined
above.
EPA
also
added
a
cost
for
identifying
facilities
to
be
permitted
for
POTWs
that
do
not
currently
operate
under
a
Pretreatment
Program.
EPA
estimates
this
cost
to
be
approximately
$
0.8
million.
This
cost
only
applies
to
the
NODA/
Proposal
Option
since
facilities
subject
to
the
upgrade
options
already
hold
permits.

For
each
major
administrative
function,
this
section
provides
below:
(
1)
a
description
of
the
activities
involved,
(
2)
the
estimated
percentage
of
facilities
that
require
the
administrative
function;
(
3)
the
frequency
with
which
the
function
is
performed,
and
(
4)
high,
middle,
and
low
estimates
of
per
facility
hours
and
costs.
All
costs
are
presented
in
year
2001
dollars.

F.
3.1
Permit
Application
and
Issuance
Before
issuing
a
wastewater
discharge
permit
to
a
facility,
the
permit
authority
typically
inspects
the
facility,
monitors
the
facility s
wastewater,
and
completes
pollutant
limits
calculations
and
permit
paperwork.
This
section
discusses
the
costs
of
completing
limits
calculations
and
paperwork;
subsequent
sections
address
inspection
and
monitoring
costs.
This
section
also
discusses
the
costs
of
technical
assistance
that
the
control
authority
may
provide
facilities
to
facilitate
compliance
with
new
limits.
Finally,
this
section
includes
the
costs
of
public
and
evidentiary
hearings
that
may
be
required
for
some
permits.

a.
Issue
a
concentration­
based
permit
at
a
previously
unpermitted
facility
To
issue
a
concentration­
based
permit,
permit
authorities
first
review
permit
applications
for
completeness.
If
an
application
is
incomplete,
the
authorities
notify
the
applicant
and
request
the
missing
information.
Completed
applications
are
assigned
to
permit
writers,
who
review
the
applications
in
more
detail
as
they
develop
permit
conditions.
The
effort
required
to
complete
these
activities
depends,
in
part,
on
the
extent
to
which
the
permit
authority
has
automated
the
permitting
process.

EPA
assumed
that
one­
third
of
facilities
will
be
permitted
in
each
of
the
three
years
following
the
rule s
effective
date
because
compliance
is
mandated
within
three
years
of
the
date
the
standard
is
effective
(
40
CFR
Section
403.6).
EPA
further
assumed
that
facilities
are
repermitted
in
five
year
cycles.
(
The
administrative
costs
of
repermitting
are
discussed
separately
below.)

The
actual
number
of
facilities
that
are
permitted
each
year
is
likely
to
differ
somewhat
from
EPA s
simplifying
assumption.

These
minor
differences
in
permit
timing
are
not
expected
to
significantly
change
the
estimated
administrative
costs.

4
The
25
percent
oversight
cost
provision
is
based
on
comments
and
data
received
from
the
Association
of
Metropolitan
Sewerage
Agencies
(
AMSA).

F­
4
MP&
M
EEBA:
Appendices
Appendix
F:
Administrative
Costs
Table
F.
1:
Administrative
Activity:
Develop
and
issue
a
concentration­
based
permit
at
a
previously
unpermitted
facility
Percent
of
facilities
for
which
activity
is
required
Frequency
of
activity
Typical
costs
(
2001$)

Low
Median
High
100%
of
unpermitted
MP&
M
facilities
(
applicable
to
NODA/
Proposal
option
only)
One
time
4.0
hours;

$
122
10.0
hours;

$
304
40.0
hours;

$
1,217
Source:
U.
S.
EPA
analysis
of
POTW
Survey
responses;
U.
S.
Department
of
Labor,
Bureau
of
Labor
Statistics.

b.
Issue
a
mass­
based
permit
for
a
previously
unpermitted
facility5
The
same
administrative
activities
required
to
issue
a
concentration­
based
permit
are
also
required
for
a
mass­
based
permit.

In
addition,
for
mass­
based
permits
issued
under
the
MP&
M
rule,
the
permit
writer
must
determine
whether
the
facility
practices
pollution
prevention
and
water
conservation
methods
equivalent
to
those
specified
as
the
basis
for
BPT.
If
so,
the
permitting
authority
must
determine
the
facility s
historical
flow
rate.
If
not,
the
authority
must
derive
a
mass­
based
limit
based
on
other
factors
such
as
production
rates.
When
a
facility
matches
BPT
water
conservation
practices
and
provides
historic
flow
data,
development
of
a
mass­
based
permit
is
a
relatively
straight­
forward
process.
However,
the
task
will
be
more
challenging
at
a
facility
practicing
only
limited
water
conservation,
particularly
if
the
facility
has
multiple
production
units
and
generates
integrated
process
and
sanitary
wastewaters.

Table
F.
2:
Administrative
Activity:
Develop
and
issue
a
mass­
based
permit
at
a
previously
unpermitted
facility
Percent
of
facilities
for
which
activity
is
required
Frequency
of
activity
Typical
costs
(
2001$)

Low
Median
High
100%
of
MP&
M
facilities
being
issued
a
new
mass­
based
permit
(
estimates
used
for
the
proposed
rule)
One
time
4.0
hours;
$
122
13.0
hours;
$
396
40.0
hours;
$
1,217
Source:
U.
S.
EPA
analysis
of
POTW
Survey
responses;
U.
S.
Department
of
Labor,
Bureau
of
Labor
Statistics.

c.
Issue
a
mass­
based
permit
for
a
facility
with
a
concentration­
based
permit6
Some
of
the
activities
described
above
for
issuing
a
mass­
based
permit
will
be
simplified
in
cases
where
the
facility
already
holds
a
concentration­
based
permit.
For
example,
much
of
the
basic
information
required
in
the
permitting
application
will
already
be
in
the
permitting
authorities 
records.
However,
the
potentially
labor­
intensive
task
of
determining
the
flow
basis
for
the
permit
remains.

5
None
of
the
regulatory
options
considered
for
the
final
rule
require
issuance
of
mass­
based
permits
for
previously
unpermitted
facilities.
However,
since
these
costs
were
developed
for
the
proposed
rule,
they
are
presented
in
this
appendix
even
though
they
are
not
used
in
the
administrative
costs
estimates.

6
None
of
the
regulatory
options
considered
for
the
final
rule
require
conversion
of
a
concentration­
based
to
a
mass­
based
permit.

However,
since
these
costs
were
developed
for
the
proposed
rule,
they
are
presented
in
this
appendix
even
though
they
are
not
used
in
the
administrative
costs
estimates.

F­
5
MP&
M
EEBA:
Appendices
Appendix
F:
Administrative
Costs
Table
F.
3:
Administrative
Activity:
Develop
and
issue
a
mass­
based
permit
at
a
facility
holding
a
concentration­
based
permit
Percent
of
facilities
for
which
activity
is
required
Frequency
of
activity
Typical
costs
(
2001$)

Low
Median
High
100%
of
MP&
M
facilities
with
permit
conversion
(
estimates
used
for
the
proposed
rule)
One
time
2.0
hours;

$
61
8.0
hours;

$
243
20.0
hours;

$
608
Source:
U.
S.
EPA
analysis
of
POTW
Survey
responses;
U.
S.
Department
of
Labor,
Bureau
of
Labor
Statistics.

d.
Provide
technical
guidance
to
a
permittee
Technical
guidance
is
frequently
provided
by
permit
authorities
to
permittees
concurrent
with
the
issuance
of
a
new
permit.

There
are
no
legal
requirements
that
a
permit
authority
provide
a
permittee
with
technical
guidance.
However,
such
guidance
is
generally
in
the
interest
of
all
parties
as
it
can
expedite
the
permitting
process,
accelerate
the
permittee s
compliance,
and
reduce
the
compliance
burden.
The
extent
of
technical
guidance
provided
varies
dramatically
among
permit
authorities.
In
some
cases,
a
permit
authority
may
hold
a
one­
day
workshop
to
provide
information
on
a
new
pretreatment
standard
to
facilities.
In
other
cases,
a
permit
authority
may
meet
extensively
with
individual
permittees
to
educate
them
regarding
their
responsibilities
under
pretreatment
standards.
The
range
of
technical
guidance
appears
to
depend
on
whether
the
permittee
already
has
a
wastewater
permit,
whether
the
permittee
is
part
of
a
multi­
facility
company,
the
resources
of
the
permit
authority,
and
the
extent
to
which
the
permit
authority
has
written
or
standardized
guidance
available
for
dissemination.

EPA
assumed
that
permit
authorities
provide
technical
guidance
to
all
facilities
being
issued
a
new
mass­
based
or
concentration­
based
permit
under
the
MP&
M
pretreatment
standards.
Costs
for
technical
guidance
were
estimated
separately
for
facilities
receiving
a
concentration­
based
permit
and
facilities
receiving
a
mass­
based
permit.
EPA
assumed
that
technical
guidance
is
provided
in
the
year
the
initial
permit
is
issued.

Table
F.
4:
Administrative
Activity:
Provide
technical
guidance
to
permittee
on
permit
compliance
Percent
of
facilities
for
which
activity
is
required
Frequency
of
activity
Typical
costs
(
2001$)

Low
Median
High
100%
of
MP&
M
facilities
being
issued
a
new
concentration­
based
permit
(
applicable
to
NODA/
Proposal
option
only)
One
time
1.5
hours;
$
46
4.0
hours;
$
122
12.0
hours;
$
365
100%
of
MP&
M
facilities
being
issued
a
new
mass­
based
permit
(
estimates
used
for
the
proposed
rule)
One
time
2.0
hours;

$
61
4.0
hours;

$
122
12.0
hours;

$
365
Source:
U.
S.
EPA
analysis
of
POTW
Survey
responses;
U.
S.
Department
of
Labor,
Bureau
of
Labor
Statistics.

e.
Conduct
a
public
or
evidentiary
hearing
on
a
proposed
permit
Federal
regulations
provide
for
a
period
during
which
the
public
may
submit
written
comments
on
a
proposed
permit
for
direct
dischargers
and/
or
request
that
a
public
hearing
be
held.
Permitting
authorities
for
indirect
dischargers
may
have
the
same
requirements.
Thus,
proposed
permits
for
indirect
dischargers
may
be
subject
to
public
comments
and
hearings.

Pretreatment
public
hearings
are
typically
conducted
at
a
scheduled
local
government
(
e.
g.,
City
Council)
meeting.
The
meetings
may
require
substantial
preparation.

Federal
regulations
also
provide
for
evidentiary
hearings
following
final
permit
determination
for
direct
dischargers.
Again,

permitting
authorities
for
indirect
dischargers
may
have
these
requirements
as
well.
Thus,
final
permit
determinations
for
indirect
dischargers
may
be
subject
to
evidentiary
hearings.

Data
from
the
POT
W
Survey
indicated
that
a
public
or
evidentiary
hearing
would
be
required
for
3.6%
of
indirect
dischargers
being
issued
a
new
mass­
based
or
concentration­
based
permit,
on
average.

F­
6
MP&
M
EEBA:
Appendices
Appendix
F:
Administrative
Costs
Table
F.
5:
Administrative
Activity:
Conduct
a
public
or
evidentiary
hearing
Percent
of
facilities
for
which
activity
is
required
Frequency
of
activity
Typical
costs
(
2001$)

Low
Median
High
3.2%
of
MP&
M
facilities
being
issued
a
new
mass­
based
or
concentration­
based
permit
(
applicable
to
NODA/
Proposal
option
only)
One
time
2.0
hours;
$
61
8.0
hours;
$
243
40.0
hours;
$
1,217
Source:
U.
S.
EPA
analysis
of
POTW
Survey
responses;
U.
S.
Department
of
Labor,
Bureau
of
Labor
Statistics.

F.
3.2
Inspection
Permit
authorities
may
choose
to
integrate
their
inspection
and
monitoring
work
force
or
to
administer
these
functions
separately.
This
discussion
covers
inspections
only;
monitoring
is
discussed
below.
Inspections
are
performed
both
to
assess
conditions
for
initial
permitting
and
to
evaluate
compliance
with
permit
requirements.
Inspections
involve
record
reviews,

visual
observations,
and
evaluations
of
the
treatment
facilities,
effluents,
receiving
waters,
etc.
EPA
assumed
that
the
initial
inspection
would
occur
in
the
same
year
a
new
permit
is
issued,
and
that
all
permitted
facilities
would
be
inspected
annually
to
assess
compliance.

Table
F.
6:
Administrative
Activity:
Inspect
facility
for
permit
development
Percent
of
facilities
for
which
activity
is
required
Frequency
of
activity
Typical
costs
(
2001$)

Low
Median
High
100%
of
MP&
M
facilities
being
issued
a
new
permit
(
applicable
to
NODA/
Proposal
option
only)
One
Time
2.2
hours;
$
66
5.0
hours;
$
152
12.0
hours;
$
365
Source:
U.
S.
EPA
analysis
of
POTW
Survey
responses;
U.
S.
Department
of
Labor,
Bureau
of
Labor
Statistics.

Table
F.
7:
Administrative
Activity:
Inspect
facility
for
compliance
assessment
Percent
of
facilities
for
which
activity
is
required
Frequency
of
activity
Typical
costs
(
2001$)

Low
Median
High
100%
of
MP&
M
facilities
being
issued
a
new
permit
(
applicable
to
NODA/
Proposal
option
only)
Annual
2.0
hours;
$
61
3.3
hours;
$
101
10.0
hours;
$
304
Source:
U.
S.
EPA
analysis
of
POTW
Survey
responses;
U.
S.
Department
of
Labor,
Bureau
of
Labor
Statistics.

F.
3.3
Monitoring
Permitting
authorities
monitor
facilities
both
to
gather
data
needed
for
permit
development
and
to
assess
compliance
with
permit
conditions.
Monitoring
includes
sampling
and
analysis
of
the
permittee s
effluent,
review
of
the
permittee s
compliance
self­
monitoring
reports,
receipt
of
non­
compliance
reports,
and
review
of
compliance
schedule
reports.
These
activities
are
discussed
below.

a.
Sample
and
analyze
permittee s
effluent
As
noted
above,
inspection
and
monitoring
staff
may
be
integrated
or
distinct.
The
costs
of
inspection
were
presented
above.

Federal
regulations
require
that
the
permit
authority
 
randomly
sample
and
analyze
the
effluent
from
industrial
users
...

independent
of
information
supplied
by
industrial
users 
(
40
CFR
Part
403.8).
The
permit
authority
obtains
samples
required
by
the
permit
and
performs
chemical
analyses.
The
results
are
used
to
verify
the
accuracy
of
the
permittee s
self­
monitoring
program
and
reports,
determine
the
quantity
and
quality
of
effluents,
develop
permits,
and
provide
evidence
for
enforcement
proceedings
where
appropriate.

F­
7
MP&
M
EEBA:
Appendices
Appendix
F:
Administrative
Costs
EPA
estimated
sampling
costs
for
all
facilities
issued
a
new
permit
under
the
MP&
M
rule,
and
assumed
annual
monitoring.

Although
EPA
requires
only
annual
effluent
sampling,
some
localities
sample
more
frequently.
EPA
encourages
this
practice.

Table
F.
8:
Administrative
Activity:
Sample
and
analyze
permittee s
effluent
Percent
of
facilities
for
which
activity
is
required
Frequency
of
activity
Typical
costs
(
2001$)

Low
Median
High
100%
of
MP&
M
facilities
being
issued
a
new
permit
(
applicable
to
NODA/
Proposal
option
only)
Annual
1.0
hour;

$
30
3.0
hours;

$
91
17.7
hours;

$
537
Source:
U.
S.
EPA
analysis
of
POTW
Survey
responses;
U.
S.
Department
of
Labor,
Bureau
of
Labor
Statistics.

b.
Review
and
record
permittee s
compliance
self­
monitoring
reports
40
CFR
Part
403.12
specifies
that:
 
Any
Industrial
User
subject
to
a
categorical
pretreatment
standard
...
shall
submit
to
the
Control
authority
during
the
months
of
June
and
December
...
a
report
indicating
the
nature
and
concentration
of
pollutants
in
the
effluent
which
are
limited
by
such
categorical
pretreatment
standards. 
The
permit
authority
briefly
reviews
these
submissions
and
may
enter
the
information
into
a
computerized
system
and/
or
file
the
data.

EPA
estimated
the
costs
of
handling
annual
self­
monitoring
reports
for
all
facilities
being
issued
a
new
permit
under
the
MP&
M
rule.

Table
F.
9:
Administrative
Activity:
Review
and
enter
data
from
permittee s
compliance
self­

monitoring
reports
Percent
of
facilities
for
which
activity
is
required
Frequency
of
activity
Typical
costs
(
2001$)

Low
Median
High
100%
of
MP&
M
facilities
being
issued
a
new
permit
(
applicable
to
NODA/
Proposal
option
only)
2
reports
per
year
0.5
hours;
$
15
1.0
hour;
$
30
4.0
hours;
$
122
Source:
U.
S.
EPA
analysis
of
POTW
Survey
responses;
U.
S.
Department
of
Labor,
Bureau
of
Labor
Statistics.

c.
Receive,
process,
and
act
on
a
permittee s
non­
compliance
report
Generally,
when
a
permittee
violates
a
permit
condition,
it
must
submit
a
non­
compliance
report
to
the
permit
authority.

Permittees
report
both
unanticipated
bypasses
or
upsets
and
violations
of
maximum
daily
discharge
limits.
The
permit
authority
receives
and
processes
both
verbal
and
written
non­
compliance
reports.
In
some
cases,
immediate
action
by
the
permit
authority
is
required
to
mitigate
the
problem.

Data
from
the
POTW
Survey
indicate
that
38.5
percent
of
all
facilities
submit
at
least
one
non­
compliance
report
annually.

Of
facilities
that
submit
at
least
one
non­
compliance
report,
the
median
number
of
reports
filed
per
year
is
five
reports.

Table
F.
10:
Administrative
Activity:
Receive,
process
and
act
on
a
permittee s
non­
compliance
reports
Percent
of
facilities
for
which
activity
is
required
Frequency
of
activity
Typical
costs
(
2001$)

Low
Median
High
38.5%
of
all
indirect
dischargers
receiving
a
new
permit
(
applicable
to
NODA/
Proposal
option
only)
5
times
per
year
1.0
hour;
$
30
2.0
hours;
$
61
6.0
hours;
$
183
Source:
U.
S.
EPA
analysis
of
POTW
Survey
responses;
U.
S.
Department
of
Labor,
Bureau
of
Labor
Statistics.

F­
8
MP&
M
EEBA:
Appendices
Appendix
F:
Administrative
Costs
d.
Review
a
permittee s
compliance
schedule
report
Permittees
submit
reports
to
permit
authorities
that
state
whether
compliance
schedule
milestones
contained
in
their
permits
have
been
met.
If
the
facility
is
in
compliance,
the
permit
authority
reviews
and
files
the
report.

Data
from
the
POTW
Survey
indicate
that
approximately
17%
of
all
facilities
are
issued
compliance
milestones.
Of
these
facilities,
94%
meet
the
milestones.
Facilities
submit
an
average
of
two
compliance
milestone
reports
per
year.
The
cost
of
handling
the
report
depends
on
whether
the
facility
is
in
compliance
with
the
schedule.

Table
F.
11:
Administrative
Activity:
Review
a
compliance
schedule
report
Percent
of
facilities
for
which
activity
is
required
Frequency
of
activity
Typical
costs
(
2001$)

Low
Median
High
Meeting
milestones:
16.0%
of
all
facilities
issued
a
new
permit
 
94%
of
the
17%
who
have
compliance
milestones
(
applicable
to
NODA/
Proposal
option
only)
2
reports
per
year
0.5
hours;

$
15
1.0
hour;

$
30
2.7
hours;

$
81
Not
meeting
milestones:
1%
of
all
facilities
issued
a
new
permit
 
6%
of
the
17%
who
have
compliance
milestones
(
applicable
to
NODA/
Proposal
option
only
2
reports
per
year
1.0
hours;
$
30
2.0
hours;
$
61
6.0
hours;
$
183
Source:
U.
S.
EPA
analysis
of
POTW
Survey
responses;
U.
S.
Department
of
Labor,
Bureau
of
Labor
Statistics.

F.
3.4
Enforcement
When
a
permitting
authority
identifies
a
permit
violation,
the
authority
determines
and
implements
an
appropriate
enforcement
action.
Considerations
when
determining
enforcement
response
include
(
1)
the
severity
of
the
permit
violation,

(
2)
the
degree
of
economic
benefit
obtained
by
the
permittee
through
the
violation,
(
3)
previous
enforcement
actions
taken
against
the
violator,
(
4)
the
deterrent
effect
of
the
response
on
similarly
situated
permittees,
and
(
5)
considerations
of
fairness
and
equity.
EPA
estimated
administrative
costs
for
two
levels
of
enforcement
actions:
(
1)
less
severe
actions
such
as
issuing
an
administrative
order,
and
(
2)
more
severe
activities
such
as
levying
an
administrative
fine.

EPA
estimated
that,
annually,
seven
percent
of
facilities
issued
a
new
permit
under
the
MP&
M
rule
will
require
a
minor
enforcement
action,
such
as
issuing
an
administrative
compliance
order.
In
addition,
EPA
estimated
that
seven
percent
of
facilities
receiving
a
new
permit
will
require
more
severe
enforcement
actions
such
as
a
fine
or
penalty.

Table
F.
12:
Administrative
Activity:
Minor
enforcement
action
e.
g.,
issue
an
administrative
order
Percent
of
facilities
for
which
activity
is
required
Frequency
of
activity
Typical
costs
(
2001$)

Low
Median
High
7%
of
MP&
M
facilities
being
issued
a
new
permit
(
applicable
to
NODA/
Proposal
option
only)
Annual
1.0
hour;
$
30
3.7
hours;
$
112
12.0
hours;
$
365
Source:
U.
S.
EPA
analysis
of
POTW
Survey
responses;
U.
S.
Department
of
Labor,
Bureau
of
Labor
Statistics.

Table
F.
13:
Administrative
Activity:
Minor
enforcement
action,
e.
g.,
impose
an
administrative
fine
Percent
of
facilities
for
which
activity
is
required
Frequency
of
activity
Typical
costs
(
2001$)

Low
Median
High
7%
of
MP&
M
facilities
being
issued
a
new
permit
(
applicable
to
NODA/
Proposal
option
only)
Annual
1.0
hour;
$
30
5.0
hours;
$
152
24.0
hours;
$
730
Source:
U.
S.
EPA
analysis
of
POTW
Survey
responses;
U.
S.
Department
of
Labor,
Bureau
of
Labor
Statistics.

F­
9
MP&
M
EEBA:
Appendices
Appendix
F:
Administrative
Costs
F.
3.5
Repermitting
The
duration
of
permits
cannot
exceed
five
years.
Renewing
a
permit
for
a
facility
in
compliance
with
an
existing
permit
is
expected
to
be
a
relatively
straightforward
task.
The
data
submitted
in
the
permit
application
generally
require
few
changes,

although
pollutant
limits
may
need
to
be
recalculated
in
some
cases.
The
labor
required
for
repermitting
depends,
in
part,
on
the
extent
to
which
the
permit
authority
has
automated
the
paperwork.

Table
F.
14:
Administrative
Activity:
Repermit
Percent
of
facilities
for
which
activity
is
required
Frequency
of
activity
Typical
costs
(
2001$)

Low
Median
High
100%
of
MP&
M
facilities
being
issued
a
new
permit
(
applicable
to
NODA/
Proposal
option
only)
every
5
years
1.0
hour;
$
30
4.0
hours;
$
122
20.0
hours;
$
608
Source:
U.
S.
EPA
analysis
of
POTW
Survey
responses;
U.
S.
Department
of
Labor,
Bureau
of
Labor
Statistics.

In
addition
to
repermitting
MP&
M
facilities
being
issues
a
new
permit,
EPA
also
considered
two
other
types
of
cost:
(
1)
the
costs
associated
with
repermitting
facilities
that
already
hold
a
permit
in
the
baseline
sooner
than
would
otherwise
be
required;
and
(
2)
cost
savings
associated
with
no
longer
having
to
permit
facilities
that
already
hold
a
permit
in
the
baseline
but
that
are
estimated
to
close
as
a
result
of
the
rule.
Both
cost
components
are
reflected
in
the
POTW
administrative
costs
presented
in
the
next
section.

F.
4
POTW
ADMINISTRATIVE
COSTS
BY
OPTION
Exhibits
F.
1
through
F.
7
at
the
end
of
this
appendix
present
the
calculation
of
POT
W
permitting
costs
for
the
final
rule
and
the
three
regulatory
alternatives
considered
by
EPA.

Exhibit
F.
1
provides
an
overview
of
the
permitting
activities,
the
estimated
percentage
of
facilities
that
require
the
administrative
function,
the
frequency
with
the
function
is
performed,
and
per
facility
hours
and
costs
for
each
function.

Exhibit
F.
2
contains
the
per
facility
hour
burden
and
other
assumptions
described
above
for
each
of
the
three
types
of
permitting
(
new
concentration­
based
permit,
new
mass­
based
permit,
and
converting
a
concentration­
based
to
a
mass­
based
permit.)

Exhibits
F.
3
through
F.
5
show
hours
by
type
of
permit
for
the
low,
medium,
and
high
estimate
of
per­
facility
burden,

respectively.
These
exhibits
also
summarize
costs
and
dollars
by
year
and
permit
type.

Exhibit
F.
6
presents
the
number
of
facilities
requiring
different
types
of
permitting,
for
each
of
the
regulatory
options.
The
exhibit
shows
the
total
number
of
facilities
that
will
be
subject
to
requirements,
the
baseline
permit
status
of
those
facilities,

and
the
number
of
facilities
by
expected
post­
compliance
permit
status.
These
estimates
are
based
on
facility
survey
information
about
baseline
permit
status
and
the
results
of
the
facility
impact
analysis
described
in
Chapter
5
of
the
EEBA.

The
exhibit
also
shows
the
number
of
currently­
permitted
facilities
that
are
projected
to
close
as
a
result
of
the
rule,
and
which
will
therefore
no
longer
require
re­
permitting.

The
final
Exhibit
F.
7
shows
the
resulting
calculation
of
POTW
administrative
hours
and
costs
by
year
for
each
regulatory
option.
This
exhibit
also
shows
the
present
value
of
these
costs,
the
annualized
cost,
and
the
maximum
hours
and
costs
incurred
in
any
one
year,
for
each
option.
These
calculations
reflect
the
incremental
number
of
facilities
requiring
different
types
of
permitting,
inspection,
monitoring,
enforcement
and
repermitting
in
each
year
multiplied
by
the
unit
hours
and
cost
per
facility
for
those
activities.
All
facilities
are
assumed
to
receive
a
permit
under
the
final
rule
within
the
three­
year
compliance
period.
Some
facilities
with
existing
permits
are
repermitted
sooner
than
they
otherwise
would
be
on
the
normal
five­
year
permitting
cycle.
The
cost
analyses
calculates
incremental
costs
by
subtracting
the
costs
of
repermitting
these
facilities
on
a
five­
year
schedule
from
the
costs
of
repermitting
all
such
facilities
within
three
years.
EPA
assumes
that
the
required
initial
permitting
activities
will
be
equally
divided
over
the
three­
year
period.
The
analysis
also
calculates
the
net
F­
10
MP&
M
EEBA:
Appendices
Appendix
F:
Administrative
Costs
change
in
the
number
of
facilities
requiring
permitting
by
subtracting
the
number
of
facilities
that
close
due
to
the
rule
from
the
number
of
facilities
that
will
require
new
permits
under
each
regulatory
option.

More
detailed
information
on
these
cost
calculations
is
provided
in
the
docket
for
the
final
rule.

F­
11
MP&
M
EEBA:
Appendices
Appendix
F:
Administrative
Costs
APPENDIX
FEXHIBITS
Exhibit
F.
1:
Government
Administrative
Activities
for
Indirect
Dischargers:
Per
Facility
Hours
and
Costs
Exhibit
F.
2:
Per­
Facility
Hours
and
Assumptions
Exhibit
F.
3:
Low
Estimate
of
Hours
and
Costs
per
Facility
Exhibit
F.
4:
Medium
Estimate
of
Hours
and
Costs
per
Facility
Exhibit
F.
5:
High
Estimate
of
Hours
and
Costs
per
Facility
Exhibit
F.
6:
Number
of
Facilities
Requiring
Additional
Permitting
Exhibit
F.
7:
POTW
Administrative
Costs
by
Option
F­
12
MP&
M
EEBA:
Appendices
Appendix
F:
Administrative
Costs
Exhibit
F.
1:
Government
Administrative
Activities
for
Indirect
Dischargers:
Per
Facility
Hours
and
Costs
Administrative
Activity
Percent
of
facilities
for
which
activity
is
required
Frequency
of
activity
Typical
hours
and
costs
Low
Median
High
Develop
and
issue
a
concentration­

based
permit
at
a
previously
unpermitted
facility
100%
of
unpermitted
facilities
(
applicable
to
NODA/
Proposal
option
only)
One
time
4.0
hours;
$
122
10.0
hours;

$
304
40.0
hours;

$
1,217
Develop
and
issue
a
mass­
based
permit
at
a
previously
unpermitted
facility
100%
of
MP&
M
facilities
being
issued
a
new
mass­
based
permit
(
estimates
used
for
the
proposed
rule)
One
time
4.0
hours;
$
122
13.0
hours;

$
396
40.0
hours;

$
1,217
Develop
and
issue
a
mass­
based
permit
at
a
facility
holding
a
concentration­

based
permit
100%
of
MP&
M
facilities
with
permit
conversion
(
estimates
used
for
the
proposed
rule)
One
time
2.0
hours;
$
61
8.0
hours;
$
243
20.0
hours;

$
608
year
Provide
technical
guidance
to
a
permittee
on
permit
compliance
100%
of
MP&
M
facilities
being
issued
a
new
concentration­
based
permit
(
applicable
to
NODA/
Proposal
option
only)
One
time
1.5
hour;
$
46
4.0
hours;
$
122
12.0
hours;

$
365
100%
of
MP&
M
facilities
being
issued
a
new
mass­
based
permit
(
estimates
used
for
the
proposed
rule)
One
time
2.0
hours;
$
61
4.0
hours;
$
122
12.0
hours;

$
365
Conduct
a
public
or
evidentiary
hearing
3.2%
of
MP&
M
facilities
being
issued
a
new
mass­
based
or
concentration­
based
permit
(
applicable
to
NODA/
Proposal
option
only)
One
time
2.0
hours;

$
61
8.0
hours;

$
243
40.0
hours;

$
1,217
Inspect
facility
for
permit
development
100%
of
MP&
M
facilities
being
issued
a
new
permit
(
applicable
to
NODA/
Proposal
option
only)
One
Time
2.2
hours;

$
66
5.0
hours;

$
152
12.0
hours;

$
365
Inspect
facility
for
compliance
assessment
100%
of
MP&
M
facilities
being
issued
a
new
permit
(
applicable
to
NODA/
Proposal
option
only)
Annual
2.0
hours;

$
61
3.3
hours;

$
101
10.0
hours;

$
304
Sample
and
analyze
permittee s
effluent
100%
of
MP&
M
facilities
being
issued
a
new
permit
(
applicable
to
NODA/
Proposal
option
only)
Annual
1.0
hour;

$
30
3.0
hours;

$
91
17.7
hours;

$
537
Review
and
enter
data
from
permittee s
compliance
self­
monitoring
reports
100%
of
MP&
M
facilities
being
issued
a
new
permit
(
applicable
to
NODA/
Proposal
option
only)
2
reports
per
year
0.5
hours;

$
15
1.0
hour;

$
30
4.0
hours;

$
122
Receive,
process
and
act
on
a
permittee s
non­
compliance
reports
38.5%
of
all
indirect
dischargers
receiving
a
new
permit
(
applicable
to
NODA/
Proposal
option
only)
5
times
per
year
1.0
hour;

$
30
2.0
hours;

$
61
6.0
hours;

$
183
Review
a
compliance
schedule
report
Meeting
milestones:
16.0%
of
all
facilities
issued
a
new
permit
 
94%
of
the
17%
who
have
compliance
milestones
(
applicable
to
NODA/
Proposal
option
only)
2
reports
per
year
0.5
hours;

$
15
1.0
hour;

$
30
2.7
hours;

$
81
Not
meeting
milestones:
1%
of
all
facilities
issued
a
new
permit
 
6%
of
the
17%
who
have
compliance
milestones
(
applicable
to
NODA/
Proposal
option
only)
2
reports
per
year
1.0
hours;

$
30
2.0
hours;

$
61
6.0
hours;

$
183
Minor
enforcement
action
e.
g.,
issue
an
administrative
order
7%
of
MP&
M
facilities
being
issued
a
new
permit
(
applicable
to
NODA/
Proposal
option
only)
Annual
1.0
hour;
$
30
3.7
hours;
$
112
12.0
hours;
$
365
Minor
enforcement
action,
e.
g.,
impose
an
administrative
fine
7%
of
MP&
M
facilities
being
issued
a
new
permit
(
applicable
to
NODA/
Proposal
option
only)
Annual
1.0
hour;

$
30
5.0
hours;

$
152
24.0
hours;

$
730
Repermit
100%
of
MP&
M
facilities
being
issued
a
new
permit
(
applicable
to
NODA/
Proposal
option
only)
Every
5
years
1.0
hour;
$
30
4.0
hours;
$
122
20.0
hours;
$
608
Source:
Estimates
of
hours
by
activity
from
the
1996
MP&
M
POTW
Survey.
Average
hourly
rate
from
Bureau
of
Labor
Statistics
(
Sept.
2002
rate,
adjusted
to
$
2001).

F­
13
MP&
M
EEBA:
Appendices
Appendix
F:
Administrative
Costs
Exhibit
F.
2:
Per­
Facility
Hours
and
Assumptions
Activity
Low
Medium
High
%
Facil
x/
yr
Notes
New
concentration­
based
permit
develop
and
issue
permit
4.0
10.0
40.0
100.0%
1
one­
time
provide
technical
guidance
1.5
4.0
12.0
100.0%
1
one­
time
conduct
public
or
evidentiary
hearings
2.0
8.0
40.0
3.2%
1
one­
time,
3.2%
of
facilities
inspection
for
permit
development
2.2
5.0
12.0
100.0%
1
one­
time
inspection
for
compliance
assessment
2.0
3.3
10.0
100.0%
1
annual
sample
and
analyze
effluent
1.0
3.0
17.7
100.0%
1
annual
review
&
record
self­
monitoring
reports
0.5
1.0
4.0
100.0%
2
2x/
year
process
&
act
on
non­
compliance
reports
1.0
2.0
6.0
38.5%
5
5x/
year,
38.5%
of
facilities
review
compliance
schedule
report
­
in
compliance
with
schedule
0.5
1.0
2.7
16.0%
2
2x/
yr,
17%
of
facilities
with
compliance
milestones,
of
which
94%
in
compliance
review
compliance
schedule
report
­

not
in
compliance
with
schedule
1.0
2.0
6.0
1.0%
2
2x/
yr,
17%
of
facilities
with
compliance
milestones,
of
which
6%
not
in
compliance
minor
enforcement
action
(
e.
g.,
admin
order)
1.0
3.7
12.0
7.0%
1
annual,
7%
of
facilities
minor
enforcement
action
(
e.
g.,
admin
fine)
1.0
5.0
24.0
7.0%
1
annual,
7%
of
facilities
repermit
1.0
4.0
20.0
100.0%
1
every
three
years
New
mass­
based
permit
develop
and
issue
permit
4.0
13.0
40.0
100.0%
1
one­
time
provide
technical
guidance
2.0
4.0
12.0
100.0%
1
one­
time
conduct
public
or
evidentiary
hearings
2.0
8.0
40.0
3.2%
1
one­
time,
3.2%
of
facilities
inspection
for
permit
development
2.2
5.0
12.0
100.0%
1
one­
time
inspection
for
compliance
assessment
2.0
3.3
10.0
100.0%
1
annual
sample
and
analyze
effluent
1.0
3.0
17.7
100.0%
1
annual
review
&
record
self­
monitoring
reports
0.5
1.0
4.0
100.0%
2
2x/
year
process
&
act
on
non­
compliance
reports
1.0
2.0
6.0
38.5%
5
5x/
year,
38.5%
of
facilities
review
compliance
schedule
report
­

in
compliance
with
schedule
0.5
1.0
2.7
16.0%
2
2x/
yr,
17%
of
facilities
with
compliance
milestones,
of
which
94%
in
compliance
review
compliance
schedule
report
­

not
in
compliance
with
schedule
1.0
2.0
6.0
1.0%
2
2x/
yr,
17%
of
facilities
with
compliance
milestones,
of
which
6%
not
in
compliance
minor
enforcement
action
(
e.
g.,
admin
order)
1.0
3.7
12.0
7.0%
1
annual,
7%
of
facilities
minor
enforcement
action
(
e.
g.,
admin
fine)
1.0
5.0
24.0
7.0%
1
annual,
7%
of
facilities
repermit
1.0
4.0
20.0
100.0%
1
every
three
years
Converting
concentration­
based
to
mass­
based
develop
and
issue
permit
2.0
8.0
20.0
100.0%
1
one­
time
provide
technical
guidance
0.0
0.0
0.0
0.0%
0
N/
A
F­
14
MP&
M
EEBA:
Appendices
Appendix
F:
Administrative
Costs
Exhibit
F.
2:
Per­
Facility
Hours
and
Assumptions
Activity
Low
Medium
High
%
Facil
x/
yr
Notes
conduct
public
or
evidentiary
hearings
0.0
0.0
0.0
0.0%
0
N/
A
inspection
for
permit
development
0.0
0.0
0.0
0.0%
0
N/
A
inspection
for
compliance
assessment
2.0
3.3
10.0
100.0%
1
annual
sample
and
analyze
effluent
1.0
3.0
17.7
100.0%
1
annual
review
&
record
self­
monitoring
reports
0.5
1.0
4.0
100.0%
2
2x/
year
process
&
act
on
non­
compliance
reports
1.0
2.0
6.0
38.5%
5
5x/
year,
38.5%
of
facilities
review
compliance
schedule
report
­

in
compliance
with
schedule
0.5
1.0
2.7
16.0%
2
2x/
yr,
17%
of
facilities
with
compliance
milestones,
of
which
94%
in
compliance
review
compliance
schedule
report
­

not
in
compliance
with
schedule
1.0
2.0
6.0
1.0%
2
2x/
yr,
17%
of
facilities
with
compliance
milestones,
of
which
6%
not
in
compliance
minor
enforcement
action
(
e.
g.,
admin
order)
1.0
3.7
12.0
7.0%
1
annual,
7%
of
facilities
minor
enforcement
action
(
e.
g.,
admin
fine)
1.0
5.0
24.0
7.0%
1
annual,
7%
of
facilities
repermit
1.0
4.0
20.0
100.0%
1
every
three
years
Discount
rate:
7%

Average
hourly
rate:
$
30.42
($
2001)

Source:
Estimates
of
hours
by
activity
from
the
1996
MP&
M
POTW
Survey.
Average
hourly
rate
from
Bureau
of
Labor
Statistics
(
Sept.
2002
rate,
adjusted
to
$
2001).

F­
15
MP&
M
EEBA:
Appendices
Appendix
F:
Administrative
Costs
Exhibit
F.
3:
Low
Estimate
of
Hours
and
Costs
per
Facility
(
average
considering
frequency
of
activity
and
percent
of
facilities
requiring
activity)

Activity
Initial
Year
Annual
(
non­
permitting
year)
Repermit
Year
New
concentration­
based
permit
develop
and
issue
permit
4
provide
technical
guidance
2
conduct
public
or
evidentiary
hearings
0
inspection
for
permit
development
2
inspection
for
compliance
assessment
2
2
2
sample
and
analyze
effluent
1
1
1
review
&
record
self­
monitoring
reports
1
1
1
process
&
act
on
non­
compliance
reports
2
2
2
review
compliance
schedule
report
­
in
compliance
with
schedule
0
0
0
review
compliance
schedule
report
­
not
in
compliance
with
schedule
0
0
0
minor
enforcement
action
(
e.
g.,
admin
order)
0
0
0
minor
enforcement
action
(
e.
g.,
admin
fine)
0
0
0
repermit
1
Total
Hours
by
Year
14
6
7
Total
Dollars
by
Year
$
425
$
190
$
220
New
mass­
based
permit
develop
and
issue
permit
4
provide
technical
guidance
2
conduct
public
or
evidentiary
hearings
0
inspection
for
permit
development
2
inspection
for
compliance
assessment
2
2
2
sample
and
analyze
effluent
1
1
1
review
&
record
self­
monitoring
reports
1
1
1
process
&
act
on
non­
compliance
reports
2
2
2
review
compliance
schedule
report
­
in
compliance
with
schedule
0
0
0
review
compliance
schedule
report
­
not
in
compliance
with
schedule
0
0
0
minor
enforcement
action
(
e.
g.,
admin
order)
0
0
0
minor
enforcement
action
(
e.
g.,
admin
fine)
0
0
0
repermit
1
Total
Hours
by
Year
14
6
7
Total
Dollars
by
Year
$
440
$
190
$
220
Upgrading
from
concentration­
based
to
mass­
based
develop
and
issue
permit
2
provide
technical
guidance
0
conduct
public
or
evidentiary
hearings
0
inspection
for
permit
development
0
inspection
for
compliance
assessment
2
2
2
sample
and
analyze
effluent
1
1
1
F­
16
MP&
M
EEBA:
Appendices
Appendix
F:
Administrative
Costs
Exhibit
F.
3:
Low
Estimate
of
Hours
and
Costs
per
Facility
(
average
considering
frequency
of
activity
and
percent
of
facilities
requiring
activity)

Activity
Initial
Year
Annual
(
non­
permitting
year)
Repermit
Year
review
&
record
self­
monitoring
reports
1
1
1
process
&
act
on
non­
compliance
reports
2
2
2
review
compliance
schedule
report
­
in
compliance
with
schedule
0
0
0
review
compliance
schedule
report
­
not
in
compliance
with
schedule
0
0
0
minor
enforcement
action
(
e.
g.,
admin
order)
0
0
0
minor
enforcement
action
(
e.
g.,
admin
fine)
0
0
0
repermit
1
Total
Hours
by
Year
8
6
7
Total
Dollars
by
Year
$
251
$
190
$
220
Source:
Estimates
of
hours
by
activity
from
the
1996
MP&
M
POTW
Survey.
Average
hourly
rate
from
Bureau
of
Labor
Statistics
(
Sept.
2002
rate,
adjusted
to
$
2001).

F­
17
MP&
M
EEBA:
Appendices
Appendix
F:
Administrative
Costs
Exhibit
F.
4:
Medium
Estimate
of
Hours
and
Costs
per
Facility
(
average
considering
frequency
of
activity
and
percent
of
facilities
requiring
activity)

Activity
Initial
Year
Annual
(
non­
permitting
year)
Repermit
Year
New
concentration­
based
permit
develop
and
issue
permit
10
provide
technical
guidance
4
conduct
public
or
evidentiary
hearings
0
inspection
for
permit
development
5
inspection
for
compliance
assessment
3
3
3
sample
and
analyze
effluent
3
3
3
review
&
record
self­
monitoring
reports
2
2
2
process
&
act
on
non­
compliance
reports
4
4
4
review
compliance
schedule
report
­
in
compliance
with
schedule
0
0
0
review
compliance
schedule
report
­
not
in
compliance
with
schedule
0
0
0
minor
enforcement
action
(
e.
g.,
admin
order)
0
0
0
minor
enforcement
action
(
e.
g.,
admin
fine)
0
0
0
repermit
4
Total
Hours
by
Year
32
13
17
Total
Dollars
by
Year
$
986
$
400
$
522
New
mass­
based
permit
develop
and
issue
permit
13
provide
technical
guidance
4
conduct
public
or
evidentiary
hearings
0
inspection
for
permit
development
5
inspection
for
compliance
assessment
3
3
3
sample
and
analyze
effluent
3
3
3
review
&
record
self­
monitoring
reports
2
2
2
process
&
act
on
non­
compliance
reports
4
4
4
review
compliance
schedule
report
­
in
compliance
with
schedule
0
0
0
review
compliance
schedule
report
­
not
in
compliance
with
schedule
0
0
0
minor
enforcement
action
(
e.
g.,
admin
order)
0
0
0
minor
enforcement
action
(
e.
g.,
admin
fine)
0
0
0
repermit
4
Total
Hours
by
Year
35
13
17
Total
Dollars
by
Year
$
1,077
$
400
$
522
Upgrading
from
concentration­
based
to
mass­
based
develop
and
issue
permit
8
provide
technical
guidance
0
conduct
public
or
evidentiary
hearings
0
inspection
for
permit
development
0
inspection
for
compliance
assessment
3
3
3
F­
18
MP&
M
EEBA:
Appendices
Appendix
F:
Administrative
Costs
Exhibit
F.
4:
Medium
Estimate
of
Hours
and
Costs
per
Facility
(
average
considering
frequency
of
activity
and
percent
of
facilities
requiring
activity)

Activity
Initial
Year
Annual
(
non­
permitting
year)
Repermit
Year
sample
and
analyze
effluent
3
3
3
review
&
record
self­
monitoring
reports
2
2
2
process
&
act
on
non­
compliance
reports
4
4
4
review
compliance
schedule
report
­
in
compliance
with
schedule
0
0
0
review
compliance
schedule
report
­
not
in
compliance
with
schedule
0
0
0
minor
enforcement
action
(
e.
g.,
admin
order)
0
0
0
minor
enforcement
action
(
e.
g.,
admin
fine)
0
0
0
repermit
4
Total
Hours
by
Year
21
13
17
Total
Dollars
by
Year
$
643
$
400
$
522
Source:
Estimates
of
hours
by
activity
from
the
1996
MP&
M
POTW
Survey.
Average
hourly
rate
from
Bureau
of
Labor
Statistics
(
Sept.
2002
rate,
adjusted
to
$
2001).

F­
19
MP&
M
EEBA:
Appendices
Appendix
F:
Administrative
Costs
Exhibit
F.
5:
High
Estimate
of
Hours
and
Costs
per
Facility
(
average
considering
frequency
of
activity
and
percent
of
facilities
requiring
activity)

Activity
Initial
Year
Annual
(
non­
permitting
year)
Repermit
Year
New
concentration­
based
permit
develop
and
issue
permit
40
provide
technical
guidance
12
conduct
public
or
evidentiary
hearings
1
inspection
for
permit
development
12
inspection
for
compliance
assessment
10
10
10
sample
and
analyze
effluent
18
18
18
review
&
record
self­
monitoring
reports
8
8
8
process
&
act
on
non­
compliance
reports
12
12
12
review
compliance
schedule
report
­
in
compliance
with
schedule
1
1
1
review
compliance
schedule
report
­
not
in
compliance
with
schedule
0
0
0
minor
enforcement
action
(
e.
g.,
admin
order)
1
1
1
minor
enforcement
action
(
e.
g.,
admin
fine)
2
2
2
repermit
20
Total
Hours
by
Year
116
51
71
Total
Dollars
by
Year
$
3,529
$
1,543
$
2,151
New
mass­
based
permit
develop
and
issue
permit
40
provide
technical
guidance
12
conduct
public
or
evidentiary
hearings
1
inspection
for
permit
development
12
inspection
for
compliance
assessment
10
10
10
sample
and
analyze
effluent
18
18
18
review
&
record
self­
monitoring
reports
8
8
8
process
&
act
on
non­
compliance
reports
12
12
12
review
compliance
schedule
report
­
in
compliance
with
schedule
1
1
1
review
compliance
schedule
report
­
not
in
compliance
with
schedule
0
0
0
minor
enforcement
action
(
e.
g.,
admin
order)
1
1
1
minor
enforcement
action
(
e.
g.,
admin
fine)
2
2
2
repermit
20
Total
Hours
by
Year
116
51
71
Total
Dollars
by
Year
$
3,529
$
1,543
$
2,151
Upgrading
from
concentration­
based
to
mass­
based
develop
and
issue
permit
20
provide
technical
guidance
0
conduct
public
or
evidentiary
hearings
0
inspection
for
permit
development
0
inspection
for
compliance
assessment
10
10
10
F­
20
MP&
M
EEBA:
Appendices
Appendix
F:
Administrative
Costs
Exhibit
F.
5:
High
Estimate
of
Hours
and
Costs
per
Facility
(
average
considering
frequency
of
activity
and
percent
of
facilities
requiring
activity)

Activity
Initial
Year
Annual
(
non­
permitting
year)
Repermit
Year
sample
and
analyze
effluent
18
18
18
review
&
record
self­
monitoring
reports
8
8
8
process
&
act
on
non­
compliance
reports
12
12
12
review
compliance
schedule
report
­
in
compliance
with
schedule
1
1
1
review
compliance
schedule
report
­
not
in
compliance
with
schedule
0
0
0
minor
enforcement
action
(
e.
g.,
admin
order)
1
1
1
minor
enforcement
action
(
e.
g.,
admin
fine)
2
2
2
repermit
20
Total
Hours
by
Year
71
51
71
Total
Dollars
by
Year
$
2,151
$
1,543
$
2,151
Source:
Estimates
of
hours
by
activity
from
the
1996
MP&
M
POTW
Survey.
Average
hourly
rate
from
Bureau
of
Labor
Statistics
(
Sept.
2002
rate,
adjusted
to
$
2001).

F­
21
MP&
M
EEBA:
Appendices
Appendix
F:
Administrative
Costs
Exhibit
F.
6:
Number
of
Facilities
Requiring
Additional
Permitting
Option
II:
NODA/
Proposal
Option
Number
of
facilities
operating
post­
regulation
requiring
a
permit
3,687
Of
facilities
operating
post­
regulation:

existing
concentration­
based
692
existing
mass­
based
2,892
no
permit
in
baseline
103
concentration
based
to
be
converted
to
mass­
based
0
new
concentration­
based
103
new
mass­
based
0
Number
of
currently
permitted
facilities
closing
(
no
longer
requiring
a
permit)
722
Of
facilities
closing
due
to
the
rule:

existing
concentration­
based
209
existing
mass­
based
513
Option
III:
Directs
+
413
to
433
Upgrade
Number
of
facilities
operating
post­
regulation
requiring
a
permit
954
Of
facilities
operating
post­
regulation:

existing
concentration­
based
184
existing
mass­
based
770
no
permit
in
baseline
0
concentration
based
to
be
converted
to
mass­
based
0
new
concentration­
based
0
new
mass­
based
0
Number
of
currently
permitted
facilities
closing
(
no
longer
requiring
a
permit)
120
Of
facilities
closing
due
to
the
rule:

existing
concentration­
based
0
existing
mass­
based
120
Option
IV:
Directs
+
413+
50%
LL
Upgrade
Number
of
facilities
operating
post­
regulation
requiring
a
permit
1,414
Of
facilities
operating
post­
regulation:

existing
concentration­
based
515
existing
mass­
based
899
no
permit
in
baseline
0
concentration
based
to
be
converted
to
mass­
based
0
new
concentration­
based
0
new
mass­
based
0
Number
of
currently
permitted
facilities
closing
(
no
longer
requiring
a
permit)
120
Of
facilities
closing
due
to
the
rule:

existing
concentration­
based
0
existing
mass­
based
120
Source:
U.
S.
EPA
analysis.

F­
22
MP&
M
EEBA:
Appendices
Appendix
F:
Administrative
Costs
Exhibit
F.
7:
POTW
Administrative
Costs
by
Option
(@
7%
discount
rate)

Option
II:
NODA/
Proposal
Option
Year
Relative
to
Promulgation
of
Rule
1
10
11
12
13
14
15
Total
Hours
High
32,561
­
15,017
­
28,095
­
60,763
­
60,763
­
30,038
­
30,038
­
30,038
­
60,763
­
60,763
­
30,038
­
30,038
­
30,038
­
60,763
­
60,763
M
edium
33,603
­
4,289
­
7,680
­
14,480
­
14,480
­
8,335
­
8,335
­
8,335
­
14,480
­
14,480
­
8,335
­
8,335
­
8,335
­
14,480
­
14,480
Low
33,638
­
2,472
­
4,083
­
5,908
­
5,908
­
4,372
­
4,372
­
4,372
­
5,908
­
5,908
­
4,372
­
4,372
­
4,372
­
5,908
­
5,908
Tota
l
Cos
ts
High
$
990,604
$­
456,868
$­
854,738
$­
1,848,612
$­
1,848,612
$­
913,859
$­
913,859
$­
913,859
$­
1,848,612
$­
1,848,612
$­
913,859
$­
913,859
$­
913,859
$­
1,848,612
$­
1,848,612
M
edium
$
1,022,297
$­
130,480
$­
233,655
$­
440,526
$­
440,526
$­
253,575
$­
253,575
$­
253,575
$­
440,526
$­
440,526
$­
253,575
$­
253,575
$­
253,575
$­
440,526
$­
440,526
Low
$
1,023,378
$­
75,221
$­
124,220
$­
179,746
$­
179,746
$­
133,008
$­
133,008
$­
133,008
$­
179,746
$­
179,746
$­
133,008
$­
133,008
$­
133,008
$­
179,746
$­
179,746
High
M
edium
Low
N
P
V
$­
9,357,000
$­
1,802,000
$­
422,000
Annualized
Cost
$­
1,027,000
$­
198,000
$­
46,000
M
ax
O
ne
Y
ear
H
ours
32,561
33,603
33,638
M
ax
O
ne
Y
ear
C
osts
$
991,000
$
1,022,000
$
1,023,000
Option
III:
Directs
+
413
to
433
Upgrade
Year
Relative
to
Promulgation
of
Rule
1
10
11
12
13
14
15
Total
Hours
High
33
­
2,513
­
5,059
­
13,011
­
13,011
­
5,059
­
5,059
­
5,059
­
13,011
­
13,011
­
5,059
­
5,059
­
5,059
­
13,011
­
13,011
M
edium
­
144
­
805
­
1,465
­
3,055
­
3,055
­
1,465
­
1,465
­
1,465
­
3,055
­
3,055
­
1,465
­
1,465
­
1,465
­
3,055
­
3,055
Low
­
185
­
498
­
812
­
1,209
­
1,209
­
812
­
812
­
812
­
1,209
­
1,209
­
812
­
812
­
812
­
1,209
­
1,209
Tota
l
Cos
ts
High
$
1,000
$­
76,451
$­
153,901
$­
395,845
$­
395,845
$­
153,901
$­
153,901
$­
153,901
$­
395,845
$­
395,845
$­
153,901
$­
153,901
$­
153,901
$­
395,845
$­
395,845
M
edium
$­
4,394
$­
24,479
$­
44,563
$­
92,952
$­
92,952
$­
44,563
$­
44,563
$­
44,563
$­
92,952
$­
92,952
$­
44,563
$­
44,563
$­
44,563
$­
92,952
$­
92,952
Low
$­
5,616
$­
15,154
$­
24,692
$­
36,789
$­
36,789
$­
24,692
$­
24,692
$­
24,692
$­
36,789
$­
36,789
$­
24,692
$­
24,692
$­
24,692
$­
36,789
$­
36,789
High
M
edium
Low
NPV
$­
1,982,000
$­
509,000
$­
238,000
Annualized
Cost
$­
218,000
$­
56,000
$­
26,000
M
ax
O
ne
Y
ear
H
ours
33
­
144
­
185
M
ax
O
ne
Y
ear
C
osts
$
1,000
$­
4,000
$­
6,000
9
8
7
6
5
4
3
2
9
8
7
6
5
4
3
2
F­
23
MP&
M
EEBA:
Appendices
Appendix
F:
Administrative
Costs
Exhibit
F.
7:
POTW
Administrative
Costs
by
Option
(@
7%
discount
rate)

Option
IV:
Directs
+
413+
50%
LL
Upgrade
Year
Relative
to
Promulgation
of
Rule
1
10
11
12
13
14
15
Total
Hours
High
1,566
­
980
­
3,525
­
15,311
­
15,311
­
3,525
­
3,525
­
3,525
­
15,311
­
15,311
­
3,525
­
3,525
­
3,525
­
15,311
­
15,311
M
edium
162
­
498
­
1,158
­
3,515
­
3,515
­
1,158
­
1,158
­
1,158
­
3,515
­
3,515
­
1,158
­
1,158
­
1,158
­
3,515
­
3,515
Low
­
108
­
421
­
735
­
1,324
­
1,324
­
735
­
735
­
735
­
1,324
­
1,324
­
735
­
735
­
735
­
1,324
­
1,324
Tota
l
Cos
ts
High
$
47,645
$­
29,805
$­
107,256
$­
465,813
$­
465,813
$­
107,256
$­
107,256
$­
107,256
$­
465,813
$­
465,813
$­
107,256
$­
107,256
$­
107,256
$­
465,813
$­
465,813
M
edium
$
4,935
$­
15,150
$­
35,234
$­
106,945
$­
106,945
$­
35,234
$­
35,234
$­
35,234
$­
106,945
$­
106,945
$­
35,234
$­
35,234
$­
35,234
$­
106,945
$­
106,945
Low
$­
3,283
$­
12,822
$­
22,360
$­
40,288
$­
40,288
$­
22,360
$­
22,360
$­
22,360
$­
40,288
$­
40,288
$­
22,360
$­
22,360
$­
22,360
$­
40,288
$­
40,288
High
M
edium
Low
N
P
V
$­
1,940,000
$­
501,000
$­
236,000
Annualized
Cost
$­
213,000
$­
55,000
$­
26,000
M
ax
O
ne
Y
ear
H
ours
1,566
162
­
108
M
ax
O
ne
Y
ear
C
osts
$
48,000
$
5,000
$­
3,000
9
8
7
6
5
4
3
2
Source:
Estimates
of
hours
by
activity
from
the
1996
MP&
M
POTW
Survey.
Average
hourly
rate
from
Bureau
of
Labor
Statistics
(
Sept.
2002
rate,
adjusted
to
$
2001).

F­
24
MP&
M
EEBA:
Appendices
Appendix
F:
Administrative
Costs
REFERENCES
Association
of
Metropolitan
Sewage
Agencies
(
AMSA).
2000.
Survey
on
Proposed
MP&
M
Effluent
Guidelines.

U.
S.
Department
of
Labor,
Bureau
of
Labor
Statistics.
Average
Hourly
Rate.

U.
S.
Environmental
Protection
Agency.
1996.
MP&
M
POTW
Survey.

F­
25
MP&
M
EEBA:
Appendices
Appendix
F:
Administrative
Costs
THIS
PAGE
INTENTIONALLY
LEFT
BLANK
F­
26
