&EPA
             United States
             Environmental Protection
             Agency
             Office of Water Regulations
             and Standards
             Washington, O.C. 20460
EPA-440/2-82-004
August 1982
             Water
Economic Analysis of Proposed
Effluent Standards and
Limitations for the Metal
Finishing Industry
                        QUANTITY

-------
      ECONOMIC ANALYSIS OF PROPOSED
    EFFLUENT STANDARDS AND LIMITATIONS
     FOR THE METAL FINISHING INDUSTRY
              Prepared for:

Office of Water Regulations and Standards
  U.S.  Environmental Protection Agency
         Washington, D.C.  20460
               Prepared by:

       Booz • Allen & Hamilton Inc.
          4330  East West Highway
        Bethesda, Maryland  20814
            Under Contract No.
                68-01-6214
                               •  ••  ^ ".  'i   . .. 1 , ins Street
               August 1982     Giii^^vi,  iiiiirJJ;>  60604.

-------

-------
                         PREFACE
    The attached document is a contractor's study prepared
for the Office of Water Regulations and Standards of the
Environmental Protection Agency ("EPA").  The purpose of
the study is to analyze the economic impact which could
result from the application of alternative BPT/BAT,  PSES/
PSNS limitations and standards established under the
Federal Water Pollution Control Act (the Act), as amended.

    The study supplements the technical study ("EPA Devel-
opment Document") supporting the proposal of regulations
under the Act.  The Development Document surveys existing
and potential waste treatment control methods and technol-
ogy within particular industrial source categories and
supports proposed limitations based upon an analysis of
the feasibility of these limitations in accordance with
the requirements of the Act.  Presented in the Development
Document are the investment and operating costs associated
with various alternative control and treatment technolo-
gies.  The attached document supplements this analysis by
estimating the broader economic effects which might  result
from the required application of various control methods
and technologies.  This study investigates the effect of
alternative approaches in terms of price increases,  ef-
fects upon employment and the continued viability of af-
fected plants, effects on production, and other competi-
tive effects.

    The study has been prepared with the supervision and
review of the Office of Analysis and Evaluation of the
EPA.  This report was submitted in fulfillment of Contract
No. 68-01-6214 by Booz, Allen & Hamilton.  This report
reflects work completed as of August 1982.

    This report is being released and circulated at  ap-
proximately the same time as publication in the Federal
Register of a notice of proposed rule making.  It will be
considered along with the information contained in the
Development Document and any comments received by EPA on
either document before or during proposed rule making pro-
ceedings necessary to establish final regulations.

-------
                TABLE   OF   CONTENTS
                                                      Page
                                                     Number
   I.   EXECUTIVE SUMMARY                               1-1

  II.   INTRODUCTION                                   II-l

 III.   CHARACTERISTICS OF THE METAL                  III-l
        FINISHING UNIVERSE

  IV.   ECONOMIC IMPACT METHODOLOGY                    IV-1

   V.   DERIVATION OF COMPLIANCE COSTS                  V-l

  VI.   ECONOMIC IMPACTS                               VI-1

 VII.   REGULATORY FLEXIBILITY ANALYSIS               VII-1

VIII.   SOCIAL COSTS                                 VTII-1

  IX.   NEW SOURCE PERFORMANCE STANDARDS/              IX-1
        PRETREATMENT STANDARDS FOR NEW
        SOURCES

   X.   LIMITS OF THE ANALYSIS                          X-l


        APPENDIX
                             - 11 -

-------
                   I.   EXECUTIVE SUMMARY
1.  INTRODUCTION

    This report presents the economic impacts of BPT/BAT
and PSES/PSNS Water Pollution limitations and standards on
the Metal Finishing Industry.  This study was prepared
under the supervision of the Office of Analysis and Eval-
uation, U.S. Environmental Protection Agency.  As required
by the Clean Water Act, this study presents for considera-
tion the economic impacts of regulations proposed which
would control the industry's discharge of its effluents.
Specifically this report includes:

         The economic characteristics and size of the
         Metal Finishing Industry

         Derivation of compliance costs

         Economic impact methodology

         Economic impacts on the Metal Finishing Industry
         due to the additional costs of meeting the
         regulations

         Analysis of new sources

         Regulatory Flexibility Analysis

         Estimates of social costs.

    This Executive Summary presents a brief discussion of
the following:

         Industry structure and characteristics
         Derivation of compliance costs
         Economic impact methodology
         Economic impacts
         Limits of the analysis.

    The study is based on data from various sources,
including a 1976-77 Survey of the Metal Finishing Indus-
try, Dun and Bradstreet's Market Identifiers File,  the
Permit Compliance System (PCS) File and the 1977 Census of
Manufacturers.
                           1-1

-------
2.  INDUSTRY STRUCTURE AND CHARACTERISTICS

    The Metal Finishing Industry is divided into two seg-
ments: job shops and captives.   The job shops are mostly
small, privately-owned operations classified as SIC's 3471
and 3479 by the Standard Industrial Classification Man-
ual.  300 of the estimated 5500 job shops are forecasted
to be affected by the BATEA regulations.   The remainder of
the job shops are either indirect dischargers,  and subject
to earlier pretreatment regulations, or they do not per-
form regulated operations.  The 300 job shop plants
accounted for $300 million in sales in 1980 and employed
7200 people.  The job shop segment is competitive, with a
4-firm concentration ratio of 7.2 percent in SIC 3471 and
15.1 percent in SIC 3479.

    The captive segment of the Metal Finishing  Industry is
comprised of plants or production centers found within
manufacturing firms who provide metal finishing services
to the parent company.  There are an estimated  10,000 cap-
tive plants that will be covered by these effluent guide-
lines, 7500 indirect dischargers and 2500 direct dischar-
gers.  Captive metal finishing operations occur in an es-
timated 150 four digit SIC's and vary in size from $.5
million to over $100 million in sales.  Further, captive
plants vary with respect to the relative importance of
their metal finishing operations and the degree with which
they provide finishing services to outside customers.
Plants range from those which metal finish all  of their
own goods to those which occasionally finish in-house
goods, and take in finishing work from other producers.
The total value of shipments of the captives is estimated
at $140 billion, shipments of just the metal finished
goods is on the order of $76 billion, and total employment
in captive plants is 6.5 million.  Metal finishing process
employment is estimated at 0.6 million.  [Exhibit 1-1 on
the following page contains summary data on the industry.]

3.  REGULATORY OPTIONS

    The Agency evaluated two regulatory options for exist-
ing industrial sources.  A third option was also evaluated
for new sources.  Each option sets a compliance standard
and, where applicable, assigns the following pollution
abatement equipment:

         Option I equipment includes:

              Chemical oxidation unit
              Reduction converter
              Clarifying units
              Sludge drying beds.
                            1-2

-------
      o
      O
      o
                              r-
                              **
                              in
                                                  r~
                                                  O
*
>
w
CD
>
        u
JC    O
o    o
w    in
•H     «.
Q    CN

 •
D
                      oo
                       •
                      "T
                      ro
                                                  o
                                                  ID







rH
1
H
EH
M
CD
M
02
X
W








.c
w
•iH
C
•H
Cb
i— 1
03
4J
CD
s
PrT
AJ
CD
4J
m
rH
3
CT>
CD
PS
0
X!
4-1
4-1
O

w
0
•
x:
o
w
•H
Q
•
)_j



fH
fC
ij
4— '
O
PH
(T1


•
w x;
£X U
o ra
JC -H
cn Q
rt •
•*J •
0 Q
h-j


O O O
o • o
m vo 
r~ I-H ^



o
o m
o ro CN
^ * •
ro o r~





ro
o ro CM
o • •
m a r-





O
IT)
•^l






CN
•
t—
•
c
o
•iH
4J
03
CN rH
3
r- cn
CD
^]

CO
.,_)
x:
W
4J
O
03
U
03
x:
a
     JS   *
     o   o
     w   o
CN
cn
4-1
C
03
i— (
CU

•4-1
O

i_l
CD
£}
e
3
z


a
•iH
x:
w

14-1
o

CD
3
rH
03
>
W
C
0
•rH
rH
rH
•rH
CD
. 	 . , — *
•W-
w
4J O
C 00
0) CTl
SrH
4J
C
1)
E
>,
0
rH
a
6
Ed

rH
03
4->
O
EH










, — .
O
O
0
—


c
•H

4J
C
0)
e
>i
o
i-H

D>
C
•H
x;
to
•H
C
•H
14H
rH
03
Cb-U
g
W
CD
S


*-*
O
O
0
• — .

CO
(0
0!
CJ
o
u
CM
af fecte
CD
X)

4_>
O
c

rH
r-H
•rH
s


•K
                                               1-3

-------
         Option II equipment includes all of the equipment
         necessary to meet Option I with the addition of a
         Multimedia Filtration Unit.

         Option III equipment includes all of the
         equipment necessary to meet Option I plus
         in-plant controls on cadmium.

4.   DERIVATION OF COMPLIANCE COSTS

    The Agency generated costs for both the direct and
indirect dischargers.  Different methods for applying the
costs were developed for job shops and captives.  Each is
discussed below.

    (1}  JOB SHOP PLANT COSTING

    Plant process information from the 1976-77 Survey of
Metal Finishers on 28 direct discharging job shops was
submitted to the  technical contractor.  In this fashion
each plant could  be run through the technical contractor's
cost generating program for each Regulatory Option.
Information existed on such parameters as:

         plant flow
         flow constituents
         plant layout
         materials finished
         hours of operations
         finishing processes
         amperage, thickness of plate
         pollution control equipment in place
         tooling, piping
         construction, laboratory costs.

    (2)  CAPTIVE PLANT COSTING

    Information on captive plants' processes was not
available from the 1976-77 Survey of Metal Finishers.  As
a result the determination of the expected water pollution
control expenditures was performed by taking the following
steps:

         The technical contractor accumulated manufactur-
         ing process data on a sample of 100 direct dis-
         charging captives and 100 indirect discharging
         captives.

         Each plant in the sample was costed by the tech-
         nical contractor using its costing program.
                            1-4

-------
     The cost data was then grouped and classified
     according to water use and discharge category.

     In the calculation of the .baseline and Option II
     costs, the 1087 captive plants from the 1976-77
     Survey of Metal Finishers, for which water use
     figures were available, were grouped by metal
     finishing water usage and then individually
     matched with the technical contractor's cost
     estimates.

     For the computation of the cost of the metal
     finishing flow only, the technical contractor
     costed the metal finishing flow and the electro-
     plating flow of the integrated indirect discharg-
     ing captive plants.*  The electroplating flow
     costs were subtracted from the total metal
     finishing flow costs to obtain the cost of treat-
     ing the metal finishing flow only.  These plants
     were then matched with plants in the 1976-77
     Survey data base according to water flow
     characteristics.

     In this fashion all model plants could be costed
     and results extrapolated to the 7500 indirect
     discharging plants and 2500 direct discharging
     plants.
Amongst the indirect dischargers,  EPA drew a
distinction between "integrated" and "non-integrated1
plants.  Integrated plants are those whose metal
finishing processes include both electroplating and
non-electroplating processes.   Non-integrated plants
have only electroplating processes.   The 1979
Electroplating Pretreatment Standards regulate all
electroplating processes in metal  finishing plants.
These proposed metal finishing guidelines, which
regulate electroplating processes  at more stringent
levels than the 1979 Guidelines, will have an
incremental cost impact only on integrated plants.
The pollution control technology basis for the 1979
Standard is the same as that prescribed for this
proposed regulation.  It is therefore sufficient for
compliance with the proposed metal finishing
regulation without any additional  investment.
                       1-5

-------
    (3)   NEW SOURCE PERFORMANCE STANDARDS/PRETREATMENT
         STANDARDS  FOR NEW SOURCES

         New Source Performance Standards/Pretreatment
    Standards for New Sources were  estimated  for  new
    source metal finishers that will  need  in-plant  cadmium
    controls.  The  cadmium controls will  result  in  an
    additional annual compliance cost of  between  $17,800
    and  $24,000 (in 1980  dollars) per plant.   These costs
    are  not expected to have adverse  competitive  impacts.
    A more detailed analysis is presented  in  chapter IX.

5.   ECONOMIC IMPACT METHODOLOGY

    As in costing,  two different analyses  were developed
for each industry segment.  Each method  is presented below.

    (1)   JOB SHOPS

    The  economic impact analysis for  the  job  shops  con-
sisted of a financial assessment of the  28 model  plants
and their capacity  to handle the incremental  cost of the
capital  investment.

    Estimated BATEA costs and a linked price  increase that
passes through these costs to customers  are used  to
calculate new financial statements  for each model plant.
These statements forecast what the  firm's  financial per-
formance would be in the first year after  a BATEA invest-
ment.  A closure is a plant that would close  either
because  of inadequate cash flow to  support a  bank loan  to
purchase the BATEA  equipment or from  inadequate  profits  to
the owners.  Plants with a coverage ratio  (which  basically
is projected cash flow divided by scheduled loan  repay-
ments) of less than 1.5 are deemed  closures in this
analysis.

    (2)   CAPTIVE ANALYSIS

    The  analysis of the economic effects on captives of
the investment in pollution control equipment is  based
upon considerations of both changes in plant  costs  and
industry-wide changes in price.  Throughout the  analysis
two key assumptions are made:

         A captive operation owned  wholly or  in  part by a
         parent operation experiences no significant capi-
         tal availability problems.
                            1-6

-------
         The demand curve facing the industry is inelas-
         tic. This means that all cost increases will be
         passed on to consumers in the form of price
         increases.

    Firms which will experience much higher relative cost
increases than the industry average or price leaders in
the industry will not be able to raise prices sufficiently
to fully recover their added cost.  The closure routine
identifies those captive firms which are candidates to
close or divest their metal finishing operations because
they are unable to change prices sufficiently to fully
recover the costs of pollution control.  The flow chart on
the next page illustrates the plants'  closure logic.

6.  ECONOMIC IMPACTS

    The estimated economic impacts of the regulations are
significantly different for Option I compliance levels and
Option II compliance levels.  Compliance with Option I
standards will have very small effects on the metal fin-
ishing universe, while compliance with Option II will have
a measurable impact.  In both cases, the costs of
complying with the 1979 Electroplating Pretreatment
Standards were factored into the baseline conditions.

    Exhibit 1-3 shows that the total compliance burden of
Option I is $271.0 million*. tOnly indirect discharging
captives will have to make investment to comply with this
regulation.  Surveys** conducted by the Agency show that
*   This figure includes $26 million of investment in
    compliance with RCRA standards.   These RCRA costs are
    incremental to a plants' total RCRA compliance
    requirements;  i.e.,  they are attributable directly to
    compliance with the  metal finishing effluent
    guidelines.  The costs were derived through the
    technical contractor's solid waste disposal costing
    model.

**  The results of the first survey are outlined in a July
    30, 1981 memorandum  from Mr. Richard Kinch to Mr.  Art
    Berman.   The second  survey's sample design and
    selection procedure  is described in an August 20,  1981
    memorandum from Mr.  Richard Kotz to Mr.  David Pepson.
    The results of the survey are contained in a
    memorandum from Mr.  Henry D. Kahn to Mr.  David
    Pepson.   The results of the last survey are described
    in a memorandum dated December 3,  1981 from Mr.  Kinch
    to Mr.  Berman.
                           1-7

-------
       LU
I
M
CQ
M


X
<
O  O
    CC

    LU
    o:
                                UJ
                                  2
                                 Ul O
                                .0 =
       u.
       o«


       5"
          CO
          (O



          Ul £
                        (0
                  tc\o
                  a
       o
                                                        UJ Ul
                                ? "J'
                                Q. eci

                                02
                                O
                             o
                             0-
                                          t-

                                          <  .
                                          -Jo.
                                          a. oe
OJ
cr
(0
a,

tr
c
•H

o
o
<4-t

C
O

en
cu
-p
o
C
4J
o
o
                                                                CA
                                o
                                   1-8

-------
  **,
 ***.
**** ,
Price increase is computed as
Annual Pollution Control Costs
       Value of Sales.

Due to lack of demand elasticity information,
this computation does not reflect the projected
increase in prices of metal finished goods.   It
is rather, an estimate of the necessary price
increase to recover the costs of pollution
control equipment and operations.

Segmentation into multi-plant corporations and
single plant corporations is due but not limited
to the following reasons:

     Multi-plant corporations have greater access
     to capital due to their size

     The annual PCC for a plant which is part  of
     a multi-plant firm is very small relative to
     the firm's value of sales

     Multi-plant firms' structure indicates  an
     economic and/or financial advantage in  hav-
     ing a captive metal finishing operation.

All plants with a projected price increase of
more than 25 percent where less than 50 percent
of the products are metal finished.   For these
plants it would effectively mean that the cost of
the metal finishing process will increase at
least 50% in comparison with an industry wide
price increase of less than 1%.

This is a measure of the elasticity of demand
facing the firm.  The greater the percentage of
parts from outside customers the more elastic  the
demand will be.
                           1-9

-------
all direct dischargers and 84  percent  of  indirect
discharging captives are in full compliance with Option  I
control levels.   One reason for the high  rate of equipment
in place is that indirect dischargers  are subject  to
earlier pretreatment regulations.   Also,  direct
dischargers have to comply with state  and regional
standards.  Exhibit 1-4 shows  that  no  plant closures,
divestitures or  employment losses are  expected due to
compliance with  Option I.  The estimated  price increase  is
0.2 percent.

    Exhibits 1-3 and 1-4 summarize  the estimated impacts
of compliance with Option II.   The  total  investment costs
for the job shop sector is estimated at $13.2 million,
while investment for captives  will  be  slightly less than a
billion dollars.  Total job shop closures are estimated  at
42 (14 percent)  while captives are  expected to shut down
29 (0.2 percent) establishments and divest of 10 more.   A
total of 2069 jobs would be lost due to Option II
compliance requirements, with  1122  (15.6  percent)
accounted for by job shops.  The price increase in the  job
shop sector is calculated to be 4 percent, while in the
captive sector it will be 0.5  percent.
                           1-10

-------
        EXHIBIT 1-3

 Total  Capital  Investment
And Annual Costs By Options
    (In Million 1980 $)
Option I
Investment Costs
Annual Cost
Option II
Investment Cost
Annual Cost
Job Shops
0
0
13.2
4.0
Captives D.D.
0
0
380.0
114.0
Captives I.D.
271.0
81.0
601.0
180.0
Total
271.0
81.0
994.2
298.0
             1-11

-------
Plant Closures
MF Divestiture
Employment Loss
Price Increase
0
0
0
0
                 Job Shops
Plant Closures       42
Plant Closures(%)    14
MF Divestiture        0
Employment Loss    1122
Employment Loss(%)   15.6
Price Increase     4.0%
Captive I.D
    0
    0
    0
    0.2%
               EXHIBIT 1-4
Estimated Impacts of Regulatory Compliance
                 Option I
    Job Shops    Captive D.D
                     0
                     0
                     0
                     0
                 Option II
                 Captive D.D
                     21
                      0.8
                     10
                    760
                      0.04
                    0.5%
                         Captive I.D
                                8
                                0.1
                                0
                              185
                                0.05
                             0.5%
Total
  0
  0
  0
 0.08%

Total
  71
   0.7
  10
2067
   0.04
0.5%
                                 1-12

-------
                     II.   INTRODUCTION
    This chapter provides a brief overview of the content
and direction of the report.  It serves to highlight the
rationale for the rule-making effort as well as the ap-
proach chosen for the industry analysis.  To guide a re-
view of the work the chapter deals with the following
major issues:

         History of the metal finishing regulation
         Unique nature of the industry
         Organization of the economic impact analysis

1.  THE REGULATION

    The metal finishing industry has been affected by EPA
pollution control regulations since 1973.  The 1972 Clean
Water Act (as amended) states:

         "By July 1, 1977, existing industrial dischargers
         were required to achieve effluent limitations
         requiring the application of the best practicable
         control technology currently available1  ("BPT"),
         Section 301(b)(1)(A); and by July 1, 1983, these
         dischargers were required to achieve 'effluent
         limitations requiring the application of the best
         available technology economically achievable...
         which will result in reasonable further  progress
         toward the national goal of eliminating  the dis-
         charge of all pollutants' ("BAT"), Section
         301(b)(2)(A).

         New industrial direct dischargers were required
         to comply with Section 306 new source performance
         standards ("NSPS"), based on best available dem-
         onstrated technology, and new and existing dis-
         chargers to publicly owned treatment works
         ("POTWs") were subject to pretreatment standards
         under Sections 307(b) and (c) of the Act.  While
         the requirements for direct dischargers  were to
         be incorporated into National Pollutant  Discharge
         Elimination System (NPDES) permits issued under
         Section 402 of the Act, pretreatment standards
         were made enforceable directly against dischar-
         gers to POTWs (indirect dischargers)."

                           II-l

-------
    As with many regulatory initiatives there were court
challenges, delays,  revisions and negotiations in the
course of promulgated regulations in final form.   Pre-
treatment Guidelines for Electroplating were promulgated
in September of 1979.

    With revisions to the Clean Water Act (1977), Sections
301(b)(2)(a) and 301(b)(2)(C) of the Act now require the
achievement by July 1,  1984 of effluent limitations re-
quiring application of  BAT for "toxic" pollutants, includ-
ing the 65 "priority" pollutants and classes of pollutants
which Congress declared "toxic" under Section 307(a) of
the Act.

    This body of work is the culmination of Agency efforts
to define the costs and pollution control technologies
appropriate for effluent limitations for existing sources
after July 1, 1977,  ("Best Practicable Control Currently
Available") and after July 1, 1984 ("Best Available Tech-
nology Economically Achievable").  Levels of technology
appropriate for pretreatment of wastewater discharges to
POTW's from both new and existing sources were also iden-
tified.

2.  THE INDUSTRY

    For purposes of this regulation, the metal finishing
industry consists of any manufacturing establishment per-
forming one or more of the 44 unit operations — one of
which must be an Electroplating operation — shown in
Exhibit II-l on the following page.

    Two commercial sectors are most likely to perform
these operations.  One is the small, job shop covered
under SIC  (Standard Industrial Classification) 3471 and
3479.  The other sector is any establishment producing a
final good in which metal finishing is an intermediary
step in the production cycle.  Such establishments are
called "captives" and are primarily associated with SIC's
34 through 39.  The two sectors comprise the metal
finishing  industry.  They are, however, structurally
different, and different economic analyses are required
for each:

         The job shop sector constitutes a definable in-
         dustry (SIC's 3471/3479).  It produces a final
         product which is metal finishing services.  The
         captive sector is not an industry; rather it
         represents plants within which a metal finishing
         process is performed.

                           II-2

-------
              EXHIBIT II-l

Metal Finishing Category Unit Operations


             UNIT  OPERATIONS

     1.      Electroplating
     2.      Electoless Plating
     3.      Anodizing
     4.      Conversion Coating
     5.      Etching (Chemical Milling)
     6.      Cleaning
     7.      Machining
     8.      Grinding
     9.      Polishing
    10.      Tumbling
    11.      Burnishing
    12.      Impact Deformation
    13.      Pressure Deformation
    14.      Shearing
    15.      Heat Treating
    16.      Thermal Cutting
    17.      Welding
    18.      Brazing
    19.      Soldering
    20.      Flame Spraying
    21.      Sand Blasting
    22.      Other Abrasive Jet Machining
    23.      Electric Discharge Machining
    24.      Electrochemical Machining
    25.      Electron Beam Machining
    26.      Laser Beam Machining
    27.      Plasma Arc Machining
    28.      Ultrasonic Machining
    29.      Sintering
    30.      Laminating
    31.      Hot Dip Coating
    32.      Sputtering
    33.      Vapor Plating
    34.      Thermal Infusion
    35.      Salt Bath Descaling
    36.      Sovlent Degreasing
    37.      Paint Stripping
    38.      Painting
    39.      Electrostatic Painting
    40.      Electroplating
    41.      Vacuum Metalizing
    42.      Assembly
    43.      Calibration
    44.      Testing
                  II-3

-------
         A job shop faces a demand curve for  its  services
         while the demand curve facing  captive  metal  fini-
         shers is for a final product.   That  is,  the  de-
         mand for metal finishing in  captives is  a  derived
         demand for production inputs,  rather than  a  pri-
         mary demand for a specific good or service.

         The prices for job shop services are determined
         by the market.  However, there are no  market
         prices for captive metal finishers.   The "prices"
         of metal finishing services  performed  by captives
         are determined by accounting procedures  and  are
         used for internal transfers  alone.

    Given the fundamental differences between the two sec-
tors, two separate designs and methods  were developed for
the industry impact analysis.

3.  ORGANIZATION

    This report consists of ten chapters of which the
third through the sixth represent the major substantive
sections.  In the third chapter the characteristics of the
metal finishing universe are presented.  This basically
defines the object of the regulation.  Chapter  five pre-
sents the costs developed by the Agency's technical con-
tractor coupled with the guidance from the Agency on  such
matters as prior investments in controls, degree  of
equipment in place and stringency of  the regulatory sce-
narios.  Chapters four and six are tightly linked with de-
signing the methods for calculating impacts and displaying
the estimates of resulting impacts on the industry.  Sub-
sequent chapters reflect all the added considerations of
the regulatory review including:

         Social Costs

         Regulatory Flexibility

         New Source Performance Standards/Pretreatment
         Standards for New Sources

    The report closes with a discussion of the possible
limitations on the reliability or validity of the analysis
                           II-4

-------
   III.  CHARACTERISTICS OF THE METAL FINISHING UNIVERSE
1.  INTRODUCTION

    This description of the economic and financial charac-
teristics of the metal finishing universe is organized on
the basis of the two industry sectors,  job shops and
captives.  Both job shops and captives  are engaged in sim-
ilar production activities, but face different economic
variables.  The job shop sector constitutes an industry
and provides finishing services as a final product to out-
side customers.  Job shops fall into the standard indus-
trial classifications (SIC) of 3471 and 3479.  The captive
sector does not constitute an industry  in the normal
sense.  Captives perform metal finishing operations as
part of the production process and for  the most part do
not provide finishing services to outside customers as a
final good.  Further, captive operations occur in an esti-
mated 150 four-digit SIC's, with metal  finishing applied
to hundreds of products.  Products and  markets for final
goods are used as the basis for dividing the metal finish-
ing universe into sectors since they are the focus of the
economic activity affecting the universe.  It is these
economic considerations which will determine how the in-
dustry responds to increased effluent control costs.  This
chapter characterizes the two sectors of the metal finish-
ing universe in terms of economic significance, size,
financial strength and competitive structure.

2.  JOB SHOP SECTOR

    A metal finishing job shop is defined as a firm whose
primary operations are classified as SIC's 3471 or 3479.
There are an estimated 3000 firms performing regulated
operations in these SIC's but 2700 are  indirect dis-
chargers and fall under an earlier pretreatment regula-
tion.  In general, job shops are small, owner-operated,
single-plant firms that provide metal finishing services
to outside customers.  The job shop industry sector
appears to be competitive with a four firm concentration
ratio of 7.2 percent in SIC 3471 and 15.1 percent in
                          III-l

-------
SIC 3479.   The 1976-77 survey of the industry* indicated
that the demand curve facing the industry is inelastic.
The 424 owners surveyed reported that an average price
increase of 10 percent did not have a measurable affect  on
quantity sold.

    The average direct discharging job shop in 1980 had:

         Sales of $1,110,000
         Employment of 24, with 16 of these in production
         Water use of 1,630 gallons per hour.

    The total job shop direct discharging population ac-
counted for $330 million in sales, 7200 employees,  and
468,000 gallons per hour.   Exhibit III-l below shows a
classification of a representative sample of 28 direct
discharging job shops by employment and sales.

                           EXHIBIT  III-l

                 DIRECT  DISCHARGERS  CLASSIFICATION

                    Sales (Thousands of 1980 Dollars)
Employment Under $250
1 to 4
5 to 9
10 to
20 to
50 to
100 to


19
49
99
249
1
1
0
0
0
0
$250 to 499
2
3
3
0
0
0
$500 to 999 $1,000 and up Totals
0
1
0
3
0
0
0
0
2
7
4
1
3
c
^
5
10
4
1
Totals         28            4           14        28
    From a financial standpoint job shops appear to be
moderately profitable.  Of the sample of 28 firms, 21
showed an average profit before taxes of $65,000 while 7
showed losses averaging $209,000 in 1980 dollars.  (The
*U.S. Environmental Protection Agency, economic analysis
of pretreatment standards for existing sources of the
electroplating point source category.  Information on the
characteristics of the metal finishing universe was
derived from the 1976-77 survey of the industry.
                           III-2

-------
net average is a loss of $5,000.)  A more important mea-
sure for small businesses, though, is provided by the re-
turn to a working owner.  The average return to an owner
is $19,000.  Of the 28 firms, 24 show returns now to each
owner averaging $47,000 while 4 show losses averaging
$143,000 in 1980 dollars—although 2 of these 4 lost less
than $20,000.

    The average balance sheet for a direct discharger is
shown in Exhibit II-2 below.

                       EXHIBIT  III-2

         Typical Direct Discharger Balance  Sheet
                (Thousands of 1980 Dollars)
   Current Assets    $283      Current Liabilities    $194

   Fixed and Other             Long-Term Debt          137
    Long-Term Assets  320
                     	      Owners Equity
                                (Net Worth)
   Total Assets      $603

                               Total Liabilities

3.  CAPTIVE SECTOR

    Captive metal finishing operations occur in approxi-
mately 150 four digit SIC's.*  The Environmental Protec-
tion Agency estimates that 10,000** captive  plants perform
regulated processes.  There are virtually hundreds of dif-
ferent products that are wholly or partially metal fin-
ished.  Each one of these products has a unique price,
demand, and demand elasticity.  In the 1976-77 survey of
captives it was found that 24 percent of the plants judge
their in-house metal finishing contributes at least 10
percent of the final cost of the finished good.  For 40
percent of the sample, metal finishing's contribution to
the value of the finished good was less than 3 percent.
In general, captives use metal finishing as  part of the
production process for the purpose of protecting the final
product and/or enhancing it's aesthetic value.

    Metal finishing application may occur in the same
plant where the final good is produced, or the firm may
use a specialized plant to provide finishing services to
all its other plants.  Captive plants that engage in metal
 *Source - Products Finishing magazine.
**More detailed explanation follows in chapter IV,
                           III-3

-------
finishing operations are rather large.   One-sixth of all
establishments (16.7 percent)  have at least 1000 total
employees, with 57 percent having between 100 and 999
men.  An average of 60 people  are employed in metal fin-
ishing activities, while only  an average of 20 people are
employed in actual finishing operations.  The average
sales of a captive plant is $14 million a year, while the
average sales of the whole firm is more than $50 million a
year.  Exhibit II-3 below summarizes the major character-
istics of the captive universe.

                       EXHIBIT  III-3

          Characteristics  of the  Captive Universe


                              I. Disch.   D. Disch.  Total

Number of Plants               7,500       2,500     10,000

Sales (Billions $)             106.0        34.8      140.8

Total Employment  (000)         4,900       1,640      6,540

Employment in Metal
 Finishing Process (000)         450         150        600

Total Process Water Use          3.6         1.5        5.1
   (BCD)

Total Metal Finishing
 Water Use (BCD)                 1.5         0.6       2.1

Source:   1976-77  Survey of Metal Finishers.
                           III-4

-------
             IV.   ECONOMIC  IMPACT METHODOLOGY
1.  INTRODUCTION

    Two distinct impact methodologies were designed.   One
examines impacts on the job shop sector while the other
analyzes the captive sector of the metal finishing uni-
verse.  The rationale for this segmentation is summarized
below:

         The job shop sector constitutes an industry
         (SIC's 3471/3479).  It produces a final product
         which is metal finishing services.  The captive
         sector is not an industry, but rather a metal-
         finishing process performed across 150 four-digit
         SIC's.

         A job shop faces a demand curve for its products
         while the demand curve facing a captive is for a
         final product.  That is, the demand for metal
         finishing in captives is derived demand for
         production inputs.

         The prices for job shop services are determined
         by the market.  However, there are no market
         prices for captive metal finishers.  The prices
         of metal finishing services employed by captives
         are determined by accounting procedures and  are
         used for internal transfers alone.

         The 1976-77 Survey of Metal Finishers provides an
         elaborate financial and economic data base for
         individual job shops and the entire job shop sec-
         tor.  The availability of data for the investiga-
         tion of the captive sector was limited and the
         use of secondary data sources was extensive.

    Due to the above reasons, the job shop impact method-
ology relied on the use of a capital budgeting model  while
the captive sector methodology is driven by price consid-
erations .
                           IV-1

-------
2.  JOB SHOP SECTOR METHODOLOGY

    The job shop impact methodology centers around a plant
closure model developed specifically for job shops in the
metal finishing industry.   The reason for pursuing a plant
specific closure model is  four-fold:

         Primary data on a large,  representative sample of
         job shops had been gathered previously* and was
         judged still valid.

         Critical data on  a plant's fiscal condition and
         financial performance had been obtained for a
         sizable number of respondents.

         Plants are known  to be small and undercapital-
         ized, making cash flow and liquidity key factors
         in their continued survival.

         Job shops can be  defined  as a discrete market and
         economic segment.  Given  that price rises can be
         computed directly and demand elasticity estimated
         reliably, a sound basis exists for forecasting
         the plants' direct response to incremental capi-
         tal burdens.

    Since the job shop sector is characterized by a large
number of small, heterogeneous producers, the relevance of
the impact methodology rests on meeting the following
criteria:

         Primary surveys of plants and companies that es-
         tablish the affected population and characterize
         accurately the industry's economic, fiscal and
         wastewater control position.

         Technical costing geared  toward the unique con-
         ditions and needs of individual plants using cost
         and sizing rules that mirror the "real world".

         Estimation of candidates  for closure based on an
         analysis of the price/cost impacts of the pollu-
         tion control investment decision from the stand-
         point of owners,  managers, competitors, custo-
         mers, and lenders.
information gathered from the 1976-77 Survey of Metal
Finishers.
                           IV-2

-------
         Extrapolation of sample plant impacts to the in-
         dustry as a whole based on a defensible decision
         rule that reflects accurately the dynamics of the
         marketplace; e.g./ baseline closures, attrition
         rates, induced closures and new entrants.

    The closure model developed for the job shop metal
finishers is a financially driven set of operational
analyses designed to identify economically disadvantaged
firms by virtue of:

         Limited capital access or
         Insufficient profits.

Determinations of which firms would close are based on
projections of what the firm's financial standing would be
one year after the pollution control investment was made.
The basic premise is that those future conditions would be
evaluated by:

         A banker to determine if he would lend the firm
         sufficient funds for the investment, or by

         The owner(s) to determine if sufficient profits
         are projected to make it worthwhile for him
         (them) to remain in business, or whether the
         state of the business warrants an investment of
         further funds--called an equity infusion--into
         the firm in order to secure a bank loan.

    These closure determinations are predicted by the clo-
sure model based on pro forma balance and income statement
forecasts and quantitative decision rules.  The method-
ology is developed in six steps:

         Determination of allowable price increase, in-
         terest rates and lengths of loans

         Forecast of financial statements

         Coverage ratio test

         Equity infusion test

         Profitability test

         Classification of firms.
                           IV-3

-------
(1)   Determination of Variable Values

     Sxhibit IV-1, on the following page,  displays the
data and variables used by the model and the abbrevia-
tions used in the equations.   The model accepts as
inputs from the user the five variables shown.   The
first four—the interest rate and length of loans—are
input as numbers; the fifth variable—allowable price
increase (PlC)--is a set of options including:

          forecast price increase of respondent

          Average forecast price increase of respon-
          dents

          Cost pass through (CPT)

          Weighted cost pass through (CPTW).

Cost pass through is calculated as follows:

CPT = PCC (0.2 + Ip - .02) + PCO

Where the 0.2 reflects a 5 year depreciation schedule
and the .02 a 5 year flow through (accounting term;
the alternate method is capitalization) of investment
tax credit.

     Weighted cost pass through is the average of cost
pass through of the respondents weighted by sales
values as follows:

   28
    I,   CPT(i) X SA(i)
        28
         2   SA ( i )
     The cost pass through case is used to estimate
price level changes.  This assumes that each firm has
sufficient market protection for that firm to pass its
unique pollution control cost increase on to its cus-
tomers; the aggregate industry-wide price increase,
therefore, would be that of average cost producers.

( 2 )  Forecast of financial Statements

     Two sets of equations are used to produce finan-
cial statements:
                     IV-4

-------
 I
>
EH
M
CO
M
ac
x
     CO
     u
     tJ
     CO
     Q
     Z
     EH
     3
w
a
o
2
M
U
z
     a
     u
     tsi
     W
     EH
     3
     CU

































































<
Ij
Q

Q
W
a
M
0

CU
EH
Z
a
Q
z
2
CO
a
X
M Ul
t CU O
3 S< S
O -'
rH O
rH  4-1
CUrH C O
cy fp 3 M
a co O cu








_
^J
fci
-~ •—•
en cj
4J ^
CU
.-» en en
(0
4J
a

4-1
CU
cu
Si
CO

cu
o
c
03
rH
03
CO
< en cu
CJ 03 *H
— jj
1-1 -H
CO CU rH
4J .C -H
0) 4J ^Q
CO O 03
CO -H
^4 rO J
C
4J 03 4J
C C
CU 'O CU
>-i CU l-i
U X IH
3 -H 3
CJ Et, rj



























^K.
EH
<;
CU
^--

en
cu
X
03
EH

l_i
cu
4J

O

^-»
CO
CO
Q
,J
• — •
4J
•H
<4-4
O
1-1
CU





rtj
EH
^*
Q

W
-J
§
M
-. «
Q ^
EH >
J \
" EH
4J CU
j2 13
a M
j
£ «3I
i-i Z
CU O
EH M
EH
CJ1 M
C? C«i
0 Q
J  M M CC t4 <





*~*
-§
u*
CJ
cu en
-" 4J
en
4J O
en U
o
CJ 01
c
rH -H
03 4J
•H 1-1
CU CU
03 CU
o o

rH rH
O 0
IH 1-1
4J 4J
C C
o o
U CJ

c c
o o
•H -H
en 4J 4J
4J 33
3 rH rH
Q. rH rH
C 00
M CU CU
en
O
3
CO

to
0
•H
4J
03
U

•H
03
•H
O
C
03
C
•H
b



4J
C
CU
o
u
cu
a

4J
J3
n\
Q




•


O
•H
4J
03
U

4J
C
cu
u
u
3
CJ




•
U
cu
a
cu
•H
3
O1
CU
U

c
o
•H
CO
3
4-1
C
•H

>,
4J
•H
3
cr
w







^
cu
c
3
O

01
c
•H
**s
I-l
o
3
C
o
•H
4J
03
O
•H
U-4
•H
CO
to
03
rH
CJ

CU

3
CO
o
rH
CJ
CO
C
0
•H
4J
OJ
01
•H
iH
J2
O
T3
CU
X
•H
U-l
>.
jO
*•
T3 4-4
CU O























CO
EH
D
CU
EH
D
O
T3
•H
•H
rrj

3
0
rH
U-4

.C
en
03
O


0
•H
4J
OJ
a

cu
01

u
cu
>
o
u

cu
03
4J
C
cu
CJ

cu
a

to
03

X
03


1-1
CU
4J
U-l
03

4J
-H
U-4
O
1-1
CU













CO
4J
cu
en js
(0 4-1
03 SH
O
CO rH 3
CU 03
rH 4-1 4J
03 O CU
CO EH Z




• * •
mpensation as:
0
u
CO
1-1
cu
c
3
o

en
3
rH
Cu

X
03
4->

l_l
CU
4J
4-1
03

4J
•H
4-1
0
J.J
CU


1 |
+J
0
3

4-1
CU
c

U-4
O

CU
01
03
4J
c
CU
U
l-i
CU
CU
E
•H
4J
rH
rH
3
U-l
CO
-*
o
3

O
.C
3

S-l
cu
c
3
O

&.J
cu
a

u
03
r-l
CUrH

«c




*






•
                                                       IV-5

-------
          Current statements are prepared using the
          respondent's balance sheet—taken directly
          from the questionnaire—and supplemented by
          tne calculations shown on Exhibit IV-2.

          Projected statements are prepared using  the
          current balance sheet information and the
          input variables in the formulas shown in
          Exhibit IV-3.

(3)  Coverage Ratio Test

     The coverage ratio test is straightforward.  If
PCR is greater than or equal to 1.5, a firm is con-
sidered to be able to obtain a loan.  This level is
typical for small businesses with the owner's guaran-
tee, but still represents a moderate level of risk to
the banker.  The 1.5 provides for some coverage of
seasonal trends and temporary business down terms.  If
PCR is less than 1.5, then an equity infusion test is
made.

(4)  Equity Infusion Test

     The amount of equity that the owner(s) would  have
to invest to qualify for a smaller loan, thus raising
PCR to 1.5, is defined as the equity infusion (El).
An owner would invest El and borrow PCC-EI.  The test
is taat the owners would make the investment if they
could maintain an income of 23,610* during the year of
the investment and the loss in income for that one
year would be less than one third the total, i.e., an
equity infusion would be made:

      (PPAT + QC - El)
             WO         2 $23,610, and     El    /1/3
                                        PAT + OC

     In cases where equity infusion was not possible,
the above test was aade using PCC instead of El.
Again, if the value were greater than $23,610, it  was
assumed tiiat the investment would be made by the
owners.
$15,000 was used as the cut-off in 1975.  $23,610
adjusts $15,000 to 1980 dollars.
                     IV-6

-------
(N
 I
       EH
       2
EH
cn
EH     M
M     CJ
CO     Z
M     «C
X     Z
X     M
W     fri

       EH
       Z
       U

••
•^
c
<0


-P 0
en
U EH
•H CD
M-l CU

i| 1 i| j
0 O

dP aft
CM m
fN tl*

II
s
Cd o
EH J
rf^ fcj
CC
X
x en

EH CJ

II II
(^ C^
CJ






















O
M
EH
<
(X

w
u

OS
frl
>
O
u

II
a:
U
calculated

cu
u
(0

CO
cu
u
3
en
(C
cu
£

,-t

-------
 I
>
H

&H
H
CQ
X
a
               co

               z
               a
               2
               a
               £—i
               <

               CQ
CJ
z
fa

Q
a
£"*
CJ
a
»-3
O
(X
Cu
1
<0
a;
3J
5
CJ
r"
• H
T3
CU
4J
O

4J
3
a
c
•H
T3
C
ra
tn
OJ
S-l
3
tn
ro
OJ


r— I
fl
•H
CJ
c
(0
c
•H
IM

•JJ
C

U
3
tn
to
OJ
T3
OJ
4J
CJ
0)
1-1
o
S-i
a
T3
0)
S-l
(15
a
0}
S-l
a
a
S-4
(0
4J
CU

\
CJ
CJ
Cu

+

Cu
a
Q
II

Cu
a
Q
Cu










*—.
CJ
H
Cu

4-

r-l
— '
<
cn
II

<
cn
Cu


0
CJ
Cu
1
Cu
M
X
CJ
CJ
Cu
1
Cu
JH
CJ
U
Cu

1

^-*
CJ
M
Cu
— -
<
cn

+

s-«
ca
Cu
u

j-t
CO
Cu
04



X.
Q
=1
t-3
-^
X
fa
CJ
Cu
II

a
CJ
Cu


                            01
                                                                                          H
                                                                                          a
                                                                                          CJ
                                                                                          z
                                                                                          <
                                                                                          Et
                                                                                          CJ
                                                                                          a

                                                                                          3
                                                                                          ex
                                                                                          Cw
                                                                                          CN
a
CJ
z
<
J
<
CO
                                                                                                   CJ

                                                                                                    II
                                                                                                   CJ
                                                                                                   Cu
                                                                                    CJ

                                                                                     II
                                                                                    CJ
                                                                                    Cu
                                                                                         CJ
                                                                                         u
                                                                                         Cu
CJ
CJ
Cu
      Cu
      a
      Q
      Cu

       I

      CJ
      CJ
      Cu
                                                                                                         fa
      fa
      Cu
                                                                                                              Cu
                                                                                                              Cu
      Q

      %
      II
      Q
      &*   3
      J   Z
      Cu   Cu
      Z

      II
                   tn
                   CU
                   tn
                   3
                   (0
                   O

                   tn
              tn
              tn
              CU
             r-1
              c
              3
             Cu
             Cu
Cu
a
Q
Cu
                                                                                                                      CJ
                                                                                                                      CJ
                                                                                                                      Cu
                                                                                                                      a
                                                                                                                      CJ
                                                                                                                      z
                    CU
                    s-t
                    0)
      0)
      tn
      fl
      0
                                                                                                                            o
                                                                                                                            G
                                                                                                                           •H
O

 C
                                                                                                       ^-   4J
                                                                                                       Cu
                                                                                                       >
                                                                                                             tn
                                                                                                             en
                                                                                                             cu
                                                                                                                            0)
                                                                                                                            .a
                                                                                                                            J
                                                                                                                            CJ
                                                                                                                            04
                                                                                                                                   CJ  rH

                                                                                                                                   O4  O
                                                                                                                    < T3
                                                                                                                    CJ r-4

                                                                                                                    CM  O
                                                                                                                    <   s
                                                                                                                    cj  
-------
(5)  Profitability Test

     No profitability test was incorporated into the
model, but the following rule was used when examining
matrices depicting either (PPAT + OC) or (PAT + OC)
versus PCC, and (PPAT + OC)  versus CR.  A baseline
closure is a firm that would close regardless of the
pollution control investment because of poor pre-
investment profitability, defined to be PAT + OC of
less than $10,000 per working owner.

(6)  Classification of Firms

     As a result of the foregoing tests—all of which
are incorporated into the model except for the profit-
ability test—the firms are classified into categories,

          Non-closure, no equity infusion needed

          Non-closure due to equity infusion

          Candidate for closure due to lack of
          profitability

          Candidate for closure due to lack of capital
          access

          Vulnerable firm on pre-investment basis,
          i.e., baseline closure.

     The relationship between the model plants and  the
universe of direct discharging job shops was estab-
lished using variables and values provided from
several sources.  This analysis revealed that model
plants show sufficient similarity to the universe to
allow closures for the model plants to represent im-
pacts in the universe on a directly proportional basis.

     Development of the extrapolation rules entailed
several sequential steps summarized as follows:

          The number of baseline closures (i.e.,
          plants likely to close prior to BATBA
          investment) were estimated.

          The model plant data base was corrected for
          the baseline closures.
                     IV-9

-------
              Extrapolations from the model  plant  impacts
              to the corrected universe  were conducted  on
              a straight  line basis.

         The impact closure rate estimated by the  closure
    routine was applied across the balance of job  shops
    remaining after purging the universe of  baseline clo-
    sures.   Tne rationale for this approach  is that there
    is not  sufficient sample data to  support the develop-
    ment of a probability distribution from  which  unique
    closure probability estimates could  be developed for
    selected plant characteristics,  such as  size,  sales,
    water use, etc.  Employing this  technique does not
    alter the aggregate result, but  may  over-or under-
    estimate the number of closures  and  primary economic
    impacts associated with a specific category of job
    shops.

3.  CAPTIVE SECTOR METHODOLOGY

    Economic analysis of  the captive  metal finishing uni-
verse is extremely complex.  Captive  metal finishing
operations occur across 150 industries (four-digit SIC
level).  Each industry markets products  which are  differ-
ent tnan products manufactured by other  industries.  Con-
sequently,  any one industry faces a  unique demand  curve
for its products and a particular pricing structure.
Moreover, the economic structure of  different industries
varies, especially degrees of competition and concentra-
tion ratios.  In addition, individual firms  within each
industry face different demand elasticities  than other
firms in the same industry.  There are well  established
empirical techniques for  the investigation of markets and
for estimating demand and demand elasticities, but there
are several difficulties involved when the objective func-
tion is the market conditions for metal  finished products:

         Output for metal finishing cannot be measured
         readily as it is not a final product; rather,  it
         is an input into the production function  with
         varying degrees of use in many final products.

         The prices of metal finished products vary with
         their final use.  The price charged by any one
         firm will depend on:

              The extent of price or non-price competition
              in the industry
                         IV-10

-------
              Types of manufacturing processes used

              The availability of substitutes

              The geographic distribution of the firm's
              customers and suppliers.

         In captive firms there is no market price for
         metal finishing since all output is for interme-
         diate consumption.

         The elasticity of demand facing each firm will be
         different from the industry as a whole.

    The analysis of the metal finishing industry is fur-
ther complicated because individual firms process more
than one product and use several types of metal finishing
processes.  Many firms can substitute between inputs,
products and processes to meet external requirements.  Any
substitution between products can change total revenues as
well as operating and capital costs.  Moreover, a firm
will not necessarily pass on cost to consumers in the
products uiost affected oy the cost increases; rather, it
will increase costs according to the elasticities of
demand.  The more inelastic the demand the higher the
probable price increase.

    In general, there are three methods availaole for es-
timating the impacts of pollution control expenditures.
They are the following:

         A plant specific, financially driven closure
         model that assesses the cash flow and profitabil-
         ity of the plant under two conditions:  pre- and
         post-expenditures for pollution controls.   Plants
         whose cash flows and equity positions cannot sup-
         port the purchase are judged candidates for clo-
         sure.  This is essentially the method that was
         used for the analysis of the job shop metal
         finishers.

         A general industry model that estimates the abil-
         ity of industries and individual firms to pass on
         the incremental direct costs associated with pol-
         lution control expenditures.  These models rely
         on the knowledge of the following:

              Concentration in the industry
              General structure of the industry
              Industry growth trends
                         IV-ll

-------
              Prices of final goods
              Prices of inputs
              Production technology
              Substitute products.

         In addition to the above,  the analysis depends
         upon formulations of scenarios of demand elas-
         ticities and cross elasticities.

         A macro-economic approach  which attempts to esti-
         mate the impact of the capital expenditures on
         the economy as a whole and derive values for the
         following variables:

              Employment level in the economy
              Price level [inflationary pressures]
              Relative prices
              Interest rates
              Aggregate level of demand and supply
              Capital availability.

    None of the approaches cited above were used in their
pure form in this analysis due to the following reasons:

         Plant specific financial or operating data do not
         exist for cases at the four-digit SIC level.

         With more than 150 four-digit SIC industries a
         sector analysis is not cost-effective.

         Adopting a macro approach  similarly poses limita-
         tions for this analysis.  Not only are the avail-
         able macro models static,  they often require in-
         put specifications for demand coefficients or
         price changes, the very factors one wishes to
         predict.  Moreover, macro  models are insensitive
         to small price changes (as is the case here) and
         thus their application is  unwarranted.

    The chosen approach is effectively a blending of
methods; a modified microeconomic analysis which best sat-
isfies the prevailing data constraints.  This analysis
relies on a large sample of plants  (1087)  for which key
data requirements are available with the use of sim-
plifying yet realistic assumptions.  A comparative pricing
model is applied to generate estimations of price changes,
plant closure candidates, divestiture candidates, and
employment effects.

    The analysis of the effects of the pollution control
expenditures on an individual plant is based upon consi-
derations of, first, the projected change in the price
level of the entire metal finishing universe required to
                           IV-12

-------
offset the investment for pollution control equipment, and
second, the ability of individual firms to change their
own prices vis-a-vis universe-wide price changes.  Due to
the complexity of estimating demand elasticities in the
metal finishing universe (described in detail above),  it
is assumed that markets for metal finished goods are com-
petitive.  Firms able to raise prices sufficiently to
fully recover the costs of pollution control will exper-
ience negligible economic impact.  The impacted firms are
those unable to raise prices which are vastly different
than the industry's average.  That is, firms with a ratio
of pollution control expenditures to total revenues which
significantly exceed the industry average will not be able
to raise prices by an amount (percent) equal to their
added costs.  These plants are singled out as impacted
firms.

    The plant impact model which is illustrated by the
flow chart in Exhibit IV-4 on the following page identi-
fies those captive firms which are candidates to close or
divest their metal finishing operations because they are
found to require price increases that are untenable in the
marketplace.  The following steps explain in some detail
the logic of the model.

         The model is based on information available from
         the 1976-77 Survey of Metal Finishers from which
         key data is present for 1087 plants.  It is
         important to note that the survey results do  not
         classify the individual plants by SIC's.

         The sample of 1087 plants is divided into two
         groups by mode of discharge:

              Direct dischargers (270 plants)
              Indirect dischargers (817 plants).

         The distinction is used to reflect the Agency's
         costing rules which assign different costs to
         direct and indirect discharging plants.

         Each plant's estimated capital pollution control
         expenditure obtained from the technical contrac-
         tor is annualized* and divided by the plant's
    Annualization computations were based on information
    obtained from the 1976-77 survey of metal finishers.
    The calculations assumed a five-year depreciation
    schedule,  a ten year asset life period,  operating and
    maintenance costs of 12 percent of total investment
    costs,  and an average cost of capital of 10  percent.
                          IV-13

-------
    LU
EH
M
<
O  O
Uj  CC
X  UJ
I-  CC
u.  =>
og
                                         iu
(A

ii
Q
CO


0- 10

a
    o
                                       •*• ~
                                       o£
                                            o
                                       =  2=
                                       Q-OC UJCL
                                       So  o
                                       02  a.
                                       o
                                    £  «
                                    
-------
  *1.     Price increase is computed as

         Annual Pollution Control Costs
              Value of Sales.

         Due to lack of demand elasticity information,
         this computation does not reflect the projected
         increase in prices of metal finished goods.   It
         is rather, an estimate of the necessary price
         increase to recover the costs of pollution
         control equipment and operations.

  **2.    Segmentation into multi-plant corporations and
         single plant corporations is due but not limited
         to the following reasons:

              Multi-plant corporations have greater access
              to capital due to their size

              The annual pollution control cost for a
              plant which is part of a multi-plant firm is
              very small relative to the  firm's value of
              sales

              Multi-plant firms'  structure indicates an
              economic and/or financial advantage in
              having a captive metal finishing operation.

 ***3.    All plants with a projected price increase of
         more than 25 percent where less  than 50 percent
         of the products are metal finished.   For these
         plants it would effectively mean that the cost of
         the metal finishing process will increase at
         least 50% in comparison with an  industry wide
         price increase of less than 1%.

****4.    This is a measure of the elasticity  of demand
         facing the firm.  The greater the percentage of
         parts from outside customers the more elastic the
         demand will be.
                          IV-15

-------
value of sales to obtain the price increase
necessary to cover the added costs of pollution
control for each individual plant.  To clarify,
the price increase per plant is computed as,

Price Increase = Annual Pollution Control Costs
                        Value of Sales

The price increase for the total population of
plants is then computed using a weighting scheme
based on the sales volume of each plant.

The entire population is divided into those
plants with projected price increases lower than
the sample average and those higher than the sam-
ple average.  Those plants with projected neces-
sary price increases of less than the average
were considered to experience minor economic
impact and flagged out.

All remaining plants are further distinguished
according to those with a projected price
increase of greater than either the sample aver-
age or five percent, whichever is larger.  The
five percent figure was chosen because five per-
cent price increase is considered a major change,
as it is significant enough to alter demand-
supply relationships.  It is especially
appropriate in lieu of the assumption of no
capital availability constraints.

All the remaining plants are divided into those
which are multi-plant corporations and those
which are single plant corporations.  A multi-
plant corporation is one where the production
process is performed in more than one plant.  In
these cases the firm uses one, or any one of its
plants for peforming metal finishing services for
all its products.  A single plant corporation is
one that operates only a single plant and metal
finishing is a major part of the production
process.  The segmentation of the plants into
multiplant companies and single plant firms was
done for the following reasons:

     Multi-plant firms' structure indicates an
     economic and/or financial advantage in hav-
     ing a captive metal finishing operation.

     The annual pollution control cost for a
     plant which is part of a multi-plant firm is
     very small relative to the firm's value of
     sales.
                  IV-16

-------
              Multi-plant  corporations  have a  greater  ac-
              cess  to capital  due  to  their  size.

         All single plant  firms were  singled out  as  candi-
         dates for  closure if  they met  one  of  two condi-
         tions:   they faced price  increases greater  than
         ten percent; or they  faced price  increases
         greater than five percent and  the  percentage  of
         metal finishing performed for  outside customers
         was greater than  50 percent.

              Any firm that obtains more  than  50  percent
              of its total revenue by performing  metal
              finishing services for  outside customers is
              a  quasi-job  shop and faces  a  relatively
              elastic demand curve; i.e.,  the  firm's
              services are relatively price sensitive.
              The reason that  the  demand  curve is more
              elastic in these cases  is that in a compe-
              titive industry  outside customers can
              readily observe  prices  and  by careful
              shopping, always choose the  lowest  price.

              A  firm with  a ten percent increase  will  not
              be able to compete in an  industry where  the
              average price increase  is less than one
              percent (0.2 percent is the  estimated  price
              increase for the chosen option).

         All multi-plant corporations are  singled out  as
         candidates to divest  their captive operation  if
         the projected price increase in  the captive plant
         is 25 percent or  more and the  percentage of
         products receiving metal  finishing is less  than
         50 percent.  As mentioned above,  multi-plant
         corporations are  not  as vulnerable to the rising
         costs of production and thus would only  divest
         their metal finishing operation  under extreme
         conditions.  A captive plant that  metal  finishes
         less than  50 percent  of its  products  and has  an
         overall price increase of more than 25 percent
         due to  metal finishing pollution  control costs
         will effectively  experience  at least  a 50 percent
         increase in the cost  of the  metal  finishing
         process.  Thus,  it will have a great  incentive to
         contract out its  metal finishing  operations or do
         away with  the process altogether  in favor of  a
         less costly alternative.

    In addition, an analysis of secondary  impacts on final
consumption goods is carried out with the  use  of  an  input-
output analysis.  (I/O model is discussed  in detail  in the
appendix.)
                          IV-17

-------
         V.   REGULATORY OPTIONS AND DERIVATION OF
                     COMPLIANCE COSTS
    Effluent treatment options and their associated costs
are obviously instrumental in the assessment of the eco-
nomic impacts of water pollution controls.   This chapter
addresses the following:

         Regulatory options
         Derivation of compliance costs
         Derivation of the size of the affected universe.

1.  REGULATORY OPTIONS

    The Environmental Protection Agency evaluated two
regulatory options for existing industrial  sources.  A
third option was also evaluated for new sources.  Each
option sets a compliance  standard and assigns specific
pollution abatement equipment.

         Option I equipment includes:

              Chemical Oxidation Unit
              Chemical Reduction Converter
              Clarifying  Units
              Sludge Drying Beds.

         Option II equipment includes all the equipment
         necessary to meet Option I/ with the addition of
         a multimedia-filtration unit.

         Option III equipment includes all  the equipment
         necessary to meet Option I plus in-plant controls
         on cadmium.

    The pollution control equipment requirements are the
same for both job shops and captives.

2.  DERIVATION OF COMPLIANCE COSTS

    The derivation of water pollution control costs dif-
fers somewhat between job shops and captives.

    (1)  Job Shop Plant Costing

         Information from the 1976-77 Survey of Metal
    Finishers on 28 direct discharging job  shops was sub-
    mitted to the technical contractor for  costing.  In
                           V-l

-------
this fashion each plant was run individually through
the technical contractor cost generating program.
Information exists on such key parameters as:

          Flow constituents
          Plant layout
          Materials finished
          Hours of operations
          Finishing processes
          Amperage, thickness of plate
          Equipment in place
          Tooling, piping
          Construction, laboratory costs.

Thus, each plant used as a model for the economic
analysis has a unique set of costs corresponding to
the regulatory options defined by the agency.

(2)  Captive Plant Costing

     Information on processes of captive plants was
not available from the 1976-77 Survey of Metal Finish-
ers.  As a result the determination of the expected
water pollution control expenditures was performed by
taking the following steps:

          The technical contractor accumulated manu-
          facturing process data on a sample of 100
          direct discharging captives and 100  indirect
          discharging captives.

          Each plant in the sample was costed  by the
          technical contractor using its costing
          program.

          The cost data was then grouped and classi-
          fied according to a plant's water use and
          discharge category.

          In the calculation of the baseline and
          Option II costs, the 1087 captive plants
          from the 1976-77 Survey of Metal Finishers
          were grouped by metal finishing water usage
          and then individually matched with the
          technical contractor's cost estimates.

          For the computation of the cost of the metal
          finishing flow only, the technical contrac-
          tor costed the metal finishing flow and  the
          electroplating flow of the integrated in-
          direct discharging captive plants.  The
          electroplating flow costs were subtracted
          from the total metal finishing flow costs to
          obtain the cost of treating the metal

                        V-2

-------
              finishing flow only.   These plants were then
              matched with plants in the 1976-77 Survey
              data base according to water flow
              characteristics.

              In this fashion all 1087 plants could be
              costed and results extrapolated to the 7,500
              indirect discharging  plants and 2,500 direct
              discharging plants.

3.  ESTIMATION OF THE SIZE OF THE UNIVERSE

    The agency relied on the Permit Compliance System
(PCS) File for the estimation of the captive metal finish-
ing universe.  That file was also used to estimate a total
of 300 direct discharging job shops in SIC's 3471 and 3479.

    The PCS File showed that there  are approximately 2,500
direct discharging captives in SIC's 25-39 that will be
covered by the metal finishing effluent guidelines.  Most
of these plants are in SIC's 34-39.  The 1976-77 Survey of
Metal Finishers indicated that there were three times as
many indirect dischargers as direct dischargers, and thus,
it was determined that there are 7,500 indirect discharg-
ing captives in the metal finishing universe.

    (1)  Segment of the Universe Affected By This
         Regulation

         In order to determine the  affected universe the
    Agency recently conducted three surveys to update its
    information on pollution control equipment in place.*
    The findings of these surveys are as follows:

              None of the 300 direct discharging job shops
              will have to invest in order to comply with
              Option I standards.

              None of the 2,500 direct discharging cap-
              tives will have to invest in order to meet
              Option I standards.
    The results of the first survey are outlined in a July
    30, 1981 memorandum from Mr.  Richard Kinch to Mr. Art
    Herman. The second survey's sample design and selec-
    tion procedure is described in an August 20, 1981
    memorandum from Mr. Richard Kotz to Mr.  David Pepson.
    The results of the survey are contained  in a memoran-
    dum from Mr.  Henry D.  Kahn to Mr. David  Pepson.  The
    results of the last survey are described in a mem-
    orandum dated December 3, 1981 from Mr.  Kinch to
    Mr. Berman.
                           V-3

-------
          Of the 7500  indirect discharging captives,
          the Electroplating Pretreatment regulation
          covers all process flows at the non-
          integrated plants (3750 plants).  The elec-
          troplating standards also regulate the elec-
          troplating processes of integrated plants
          which also number 3750 plants.   Of the inte-
          grated plants:

               17 percent will not need to install
               treatment  because they meet EPA
               limitations through the use of
               in-process controls

               51 percent have already installed
               Option 1 level pollution treatment
               equipment  for all relevant processes.

This leaves 1200 integrated plants that EPA estimates
will bear additional costs to meet Option I standards.

     To summarize, the surveys conducted by the
Environmental Protection  Agency found that all direct
dischargers, job shops and captives have already met
Option I compliance standards, and 1200 indirect
discharging captives require an addition to their
existing equipment to meet Option I levels.  Nearly
all plants in the metal finishing universe will have
to make additional investments to meet Option II com-
pliance standards.
                        V-4

-------
                  VI.  ECONOMIC  IMPACTS
    The estimated economic impacts of the regulations are
significantly different for Option I compliance levels and
Option II compliance levels.   Compliance with Option I
standards will have very small effects on the metal
finishing universe, while compliance with Option II will
have measurable impacts.

1.  BASELINE CONDITIONS

    The indirect discharging  segment of the metal finish-
ing industry is also covered  by the Electroplating Pre-
treatment regulation promulgated by EPA in 1979.  Com-
pliance with the Electroplating standards requires a con-
siderable investment cost.  Because the process coverage
of the Metal Finishing regulation is more comprehensive,
it will require an additional capital investment for
approximately 16 percent of all indirect discharging
plants.  A vast majority of the metal finishing plants
will not need an additional investment in treatment
technology because they either:

         Have only electroplating process flows (and are
         therefore completely covered by the
         Electroplating Pretreatment regulation)

         Have already installed pollution control
         equipment that treats all their metal finishing
         processes

         Have in-plant "process" controls that treat
         process flows sufficiently to meet proposed EPA
         Limitations.*

    For those plants requiring an additional investment in
pollution control equipment,  electroplating compliance
costs are factored into the baseline conditions.  The
costs attributable strictly to this Metal Finishing regu-
lation are then added on so that incremental economic
impacts can be measured.  The baseline costs for this seg-
ment of the metal finishing population -- derived by
employing assumptions identical to those for deriving the
metal finishing costs and using 1980 as the baseline
year — are $453 million in capital costs and $136 million
*This calculation is partially based on EPA surveys  of  the
metal finishing industry.

                           VI-1

-------
in annual costs.   As a result of these compliance expendi-
tures EPA estimates that there will be 24 baseline plant
closures and 6 electroplating process line divestitures.
The incremental costs and impacts for the Metal  Finishing
regulation are described below.

2.  CAPITAL AVAILABILITY CONSIDERATIONS

    The job shop methodology explicitly takes into consi-
deration the ability of plants to finance new invest-
ments.  In fact,  the closure model is based on financial
variables, the magnitude of some are determined  by the
market.

    No systematic analysis of the ability of captive
plants to finance investment was conducted because the
majority of the captive plants that require investment in
pollution control equipment are large (more than $50 mil-
lion in annual sales).  Moreover, the magnitude  of the
average required annual investment is small relative to
the size of the plants.  In addition, no finance informa-
tion was available for captive plants.

3.  ESTIMATED IMPACTS OF OPTION I

    As Exhibit VI-1 on the following page shows  the total
compliance burden of Option I is $271 million in invest-
ment costs and $81 million in annual costs.  Annualized
costs include an annualized portion of investment costs
plus yearly operation and maintenance costs.

    These investment and annual costs apply to an esti-
mated 1200 of the 3750 integrated indirect discharging
captive shops.  They arise from treating the
non-electroplating portion of the effluent from integrated
plants.  The electroplating effluent portion was regulated
already by the 1979 Electroplating Pretreatment  Standard.
Exhibit VI-2 shows that no plant closures, divestiture or
employment losses are expected due to compliance with
Option I.  The estimated price increase is approximately
0.2 percent.

    Other sectors covered by the metal finishing
regulation — the balance of the integrated indirect
discharging captives, the direct discharging job and
captive shops, and the non-integrated indirect discharging
captives are not expected to incur additional costs from
the regulation.  A series of 1981 agency surveys indicated
that approximately 2550 of the integrated indirect dis-
charging captives and all direct discharging job and
captive shops already have treatment  in place sufficient
                           VI-2

-------
            EXHIBIT  VI-1

TOTAL CAPITAL INVESTMENT AND ANNUAL
          COSTS  BY OPTIONS
       (In Millons of  1980  $)
OPTION I
Investment Costs
Annual Cost
OPTION II
Investment Costs
Annual Cost
JOB SHOPS
0
0

13.2
4.0
CAPTIVES D.D
0
0

380.0
114.0
CAPTIVES I.D
271.0
81.0

601.0
180.0
TOTAL
271.0
81.0

994.2
298.0
             VI-3

-------
to meet option I standards or do not need treatment.   The
3750 non-integrated indirect discharging captives are also
not expected to incur costs from the metal finishing  regu-
lation.  These shops are already covered by the Electro-
plating Treatment Standards.  Although the limitations in
the metal finishing regulation are more stringent,  they
can be met with the same technology basis as that employed
to meet the electroplating standards, and, thus,  are  not
expected to give rise to additional costs.

4.  ESTIMATED IMPACTS OF OPTION II

    Exhibits VI-1 and VI-2 summarize the estimated impacts
of compliance with Option II.  The total investment cost
for the job shop sector is estimated at $13.2 million,
while investment for captives will be slightly less than a
billion dollars.  Total job shop closures are estimated at
42 (14 percent of job shop universe) while captives are
expected to shut down 29 establishments (less than one
percent of captive universe) and divest of 10 more.  A
total of 2,067 jobs will be lost due to Option II com-
pliance requirements, with 1,122 accounted for by job
shops.  The price increase in the job shop sector is  cal-
culated to be 4 percent, while in the captive sector  it
will be 0.5 percent.

5.  ESTIMATED IMPACTS OF OPTION III

    Option III compliance costs were estimated for New
Source Performance Standards/Pretreatment Standards for
New Sources.  These standards will apply only to those
metal finishing plants that will need in-plant cadmium
controls.  The incremental annual cost of the cadmium
controls will be between $17,800 and $24,000 (in 1980
dollars) per plant.  These costs are not expected to  have
adverse competitive impacts.  Chapter IX contains a more
detailed discussion of Option III costs and impacts.

6.  ESTIMATED SECONDARY PRICE IMPACTS

    The secondary price impacts were derived with the use
of an input-output analysis.  The analysis estimates  the
inflationary impact of a change in the price of metal
finishing services.  The input-output analysis provides a
method of examining in a simple but quantifiable way  the
relationship between prices in a particular economic
system.  It makes possible an estimate of the consequences
of a change in any one price upon the others in the
system.  (Appendix A provides a detailed exposition of the
input-output model).  The input-output analysis assumes
that all increases in costs, direct or indirect, are
passed on: i.e., that each sector raises the price of its
primary inputs, plus the rise in the price of the inputs
                           VI-4

-------
absorbed from other industries.   However,  it is important
to note that the input-output model employed assumes that
the whole economy could be adequately represented by
static input-output technical coefficients.   Therefore/
estimation results based on the  I/O model  should be read
with this inherent weakness in mind.

    Exhibits Vl-3 and VI-4 on the next page  show the se-
condary price impact on the following:

         SIC's 34-39
         Personal consumption expenditures
         Gross private fixed capital formative
         Net exports
         Total federal government purchases.
                          VI-5

-------
                         EXHIBIT VI-2
          ESTIMATED  IMPACTS OF REGULATORY  COMPLIANCE

                           OPTION  I
                JOB SHOPS  CAPTIVE D.D  CAPTIVE I.D.
Plant Closures      000
MF Divestiture      000
Employment Loss     000
Price Increase      0           0           0.2%

                           OPTION II
Plant Closures
Plant Closures (%)
MF Divestiture
Employment Loss
Employment Loss (%;
Price Increase
TOTAL
  0
  0
  0
 0.08%
OB SHOPS
42
14
0
1122
15.6
4.0%
CAPTIVE D.D
21
0.8
10
760
0.04
0.5%
CAPTIVE I.D.
8
0.1
0
185
0.05
0.5%
TOTAL
21
0.7
10
2067
0.04
0.5%
                           VI-6

-------
                       EXHIBIT VI-3

               ESTIMATED  PRICE INCREASES  FOR
                2-DIGIT SIC'S  (IN PERCENT)

SIC             OPTION I             OPTION II

34                0.3                   0.9
35                0.3                   0.9
36                0.1                   0.3
37                0.1                   0.2
38                0.1                   0.2
39                0.1                   0.3
                       EXHIBIT VI-4

         ESTIMATED  TOTAL  PRICE  IMPACT  OF  EPA BATEA
         METAL FINISHING REGULATIONS ON PRINCIPAL
                 FINAL DEMAND  COMPONENTS
FINAL DEMAND SECTORS	PERCENTAGE PRICE INCREASE

	Option I	Option II

Personal Consumption Expenditures       0.01         0.02

Gross Private Fixed Capital Formation   0.005        0.01

Net Exports                             0.01         0.02

Total Federal Government Purchases      0.007        0.01
                           VI-7

-------
        VII.  REGULATORY FLEXIBILITY ANALYSIS
     The Regulatory Flexibility Act  (Public Law 96-354) is
a regulatory reform initiative designed to ensure that,
while achieving statutory goals, regulations do not impose
unnecessary costs on "small entities."  Small entities are
defined in Section 2(a)(3) as "small businesses, small or-
ganizations, and small governmental jurisdictions with
limited resources."  The analytical requirements for regu-
latory flexibility analysis are enumerated in Sections 603
and 604 of this statute.  Section 605 (b) qualifies these
requirements and states that:

          "Sections 603 and 604 of this title shall
          not apply to any proposed or final rule if
          the head of the agency certifies that the
          rule will not, if promulgated, have a sig-
          nificant economic impact on a substantial
          number of small entities."

     Option I, the selected option, will affect only inte-
grated captive shops.   These tend to be extremely large
operations.  The economic impact analysis set forth in this
document indicates that for Option I there will be no eco-
nomic impacts in terms of plant closures for any metal
finishing establishments, including those affected that
are considered to be small entities.  Thus, there will not
be "a significant economic impact on a substantial number of
small entities."  Exhibits VII-1 and VII-2 display the
industry-wide costs and impacts attributable to compliance
with Option I standards.

     Despite the fact that there is, thus, no statutory re-
quirement to perform this analysis, a preliminary investiga-
tion into regulatory flexibility issues has been undertaken.
These issues include:

          Definition of a small firm

          Establishment of an "optimal" small firm cri-
          terion

          Regulatory impacts on small firms.
                           VII-1

-------
               EXHIBIT VII-1
 Total Captive Investment and Annual Costs
               (In Million $)
Option I
Investment Costs
Annual Costs
Job Shops
0
0
Captives D.D.
0
0
Captives I.D.
271.0
81.0
Total
271.0
81.0
               EXHIBIT VII-2
Estimated Impacts of Regulatory Compliance
                  OPTION I

Plant Closures
MF Divestiture
Employment Loss
Price Increase
Job Shops
0
0
0
0
Captive D.D.
0
0
0
0
Captive I.D.
0
0
0
0.2%
Total
0
0
0
0.2%
                   VII-2

-------
1.   DEFINITION OF A SMALL FIRM

     In considering reasonable definitions of small firms
in the metal finishing industry, four separate approaches
appeared promising.  The four methods considered the follow-
ing criteria:

          Number of employees, using either total employ-
          ment or metal finishing production employees

          Sales volume or value added by manufacturing

          Pollution generation volume either in terms of
          metric tons of contaminant or types of contami-
          nants in the wastestream

          Production/process water volume in gallons per
          day.

     Several problems are apparent with each index, although
a basis for selecting one,  the last criterion,  is readily
defensible.  Briefly, the operant difficulties  in relying on
these measures include:

          Number of employees in the firm does  not show a
          large positive correlation with sales of the firm.
          Businesses with relatively few employees can gen-
          erate sales volumes greater than that of a plant
          with ten times the employment.  The problem here
          is plant technology and automation levels.  In
          metal finishing a highly automated plant could be
          a production giant, yet appear, on the basis of
          employment, to be a small business.

          Dollar volume in  sales or value added is equally
          unsatisfactory due to the fact that by itself it
          fails to reflect  market vulnerability.  As SBA has
          long maintained,  a small business is  small only in
          relation to its competitors.   Small in this sense
          means lacking the stature to  influence price or
          production trends in the industry.   There is no
          obvious basis for establishing any sales level as
          a distinguishing  cut-off for  a small  firm in this
          industry because  of the heterogeneity of the pro-
          ducers.
                           VII-3

-------
          For regulatory purposes establishing pollution
          generating volume is a highly relevant factor for
          defining plant size.  Regardless of employment or
          sales, plants generating large hazardous waste
          volumes are of more importance than those generating
          trivial quantities.  The problem here is that waste
          volume is a function of many plant-specific con-
          siderations (processes, chemicals, and operations),
          and data on a plant's pollution volume are not
          readily available.

          Plant water volume correlates moderately well with
          both employment and sales.   While not a predictor
          of economic size, water volume is at least asso-
          ciated with plant economics.  Additionally, plant
          water volume correlates with but does not predict
          pollution generation volume.  It serves, therefore,
          as a reasonable measure of plant size both on eco-
          nomic and technological grounds.

     Extensive sensitivity analysis that was performed in
the economic impact analysis for the earlier electroplating
pretreatment regulation showed that the most vulnerable
plants had a water flow level of less than 10,000 G.P.D.

     EPA, therefore, chose plant water volume as the pri-
mary criterion to use in confirming its earlier conclusions
that there would be no significant economic impact on a
substantial number of small entities.  It selected the
10,000 G.P.D. flow level used in the earlier Electroplating
Pretreatment Regulation as the cut-off level for identifying
potential small business economic impacts.  Exhibit VII-3
on the following page shows a profile of the metal finish-
ing revenues for a 10,000 G.P.D flow level.

2.   REGULATORY IMPACTS ON SMALL FIRMS

     As stated earlier, only integrated captive shops will
incur incremental investment costs to comply with the metal
finishing BAT/PSES effluent guidelines.  The EPA data base
indicates that plants in this subcategory are much larger
than other metal finishing plants.  Integrated captives
typically have multi-million dollar sales levels and a
metal finishing process water flow of several hundred
thousand gallons per day.  It is likely, therefore, that
very few small establishments, as defined in this chapter,
will be affected by these effluent guidelines.
                          VII-4

-------
                      EXHIBIT VII-3
            Profile of Captive Metal Finishers
               by 10,000 G.P.D. Flow Volume
                            < 10,000 G.P.D.     >10,000 G.P.D.

Population Size                 3400                 6600

Average Plant Employment         420                1,205

Average Plant Sales (Millions)    7.5                 48.1

Average MF Value (Millions)      .53                 1.83

Average MF G.P.D.              5,190              355,000
*  53.6% of all the firms in the 1976-77 survey reported
   sales of greater than $50 million in 1976 dollars.
                          VII-5

-------
     In any case, this report clearly documents that small
firms are expected to experience no economic impacts due to
this regulation.  In effect, this means that in no instance
will a plant's annual pollution control costs exceed five
percent of its sales.  This conclusion is based on an anal-
ysis of the 1,087 plants surveyed.  The plant population
covers an extensive range of plant sizes, processes, sales,
metal finishing water flow, levels of pollution treatment
operating, and other pertinent parameters.
                          VII-6

-------
           VIII.  A DISCUSSION OF SOCIAL  COSTS
    The purpose of this chapter is to explore some practi-
cal means by which the social costs of the metal finishing
regulation can be analyzed, and highlight some of the
difficulties in performing such an analysis on the metal
finishing industry.  It should be noted, however, that it
is not the intent here to formally conduct a study of so-
cial costs.

    Given the complexity of the metal finishing industry
and the lack of good data it would be most appropriate to
use a static, partial equilibrium/method to estimate the
social costs.  Under this approach costs are measured by
the total amount individuals would be willing to pay for
goods and services that could be rendered if the resources
to be employed in compliance with the regulation were
instead used in their next highest valued use.  This
framework relies on an analysis of supply and demand
relationships in the directly affected markets.  The main
difference between this method and the Hicks-Kaldor/
general equilibrium criterion is that the partial equilib-
rium approach focuses attention solely on the direct
effects of the regulation.  The general equilibrium
approach considers all effects, whether directly or indi-
rectly related to the regulation.

    For illustrative purposes/ the basic social costs of
the metal finishing regulation can be segmented into the
following components.

         The most obvious social cost is the cost of
         compliance; that is, the present and future real
         resource expenditures necessitated by law.

         The metal finishing regulation is expected to
         lead to higher input and final product prices.
         The resulting dead-weight loss of consumers' and
         producers' surplus constitutes a real social cost
         and if possible should be quantified.

         The regulation could give rise to some resource
         unemployment, which translates into lost output.
         The resulting costs are important and should be
         quantified.
                          VIII-1

-------
         Additional resources will be expended on the es-
         tablishment and enforcement of the metal finish-
         ing regulation.  This includes government
         expenditures and private sector expenditures.
         These costs should also be taken into account.

In reality, however, an accurate assessment of the social
costs of the metal finishing regulation is extremely com-
plex because the estimation of demand and supply elastici-
ties, output changes, and price changes for all 150 indus-
tries performing metal finishing is very costly, if not
impossible.  Further, due to the size of the Metal Finish-
ing universe, the estimated small price increase, and the
magnitude of the overall compliance investment, the real
resource cost will account for more than 95 percent of the
total social costs.  As a result of the difficulty in the
assessment of the dead-weight loss in consumer surplus and
the relatively small size of monitoring, implementing, and
estimated litigation costs, calculation of the net pre-
sent value of the resources to be used directly in compli-
ance with the regulation constitutes an appropriate esti-
mation of the total social costs.

    To calculate the net present value of the resources  to
be used directly in complying with the metal finishing
regulation, the discounted (using a real discount rate of
10% as recommended by the Office of Management and Budget)
stream of operating and maintenance costs was added to the
initial investment costs.  This present value was then
annualized in perpetuity by multiplying by 10 percent.*
The annual social cost for Option 1 is estimated as
$62 million.  Option II annual social costs are estimated
to be $138 million.

    A regulation may affect innovation, market structure,
or productivity in a manner which may impose additional
cost but are untraceable in a static, partial equilibrium
analysis.  In the case or this metal finishing regulation
the relative importance of these effects is expected to be
minor.
    Formula:  Annual Social Cost = .1 (Investment Cost) +
                                   (Annual Operating and
                                   Maintenance Cost)
                          VIII-2

-------
        IX.  NEW SOURCE PERFORMANCE STANDARDS/
          PRETREATMENT STANDARDS FOR NEW SOURCES
     The Federal Water Pollution Control Act of 1972  (the  "Act")
requires that New Source Performance Standards  (NSPS) represent
the best available demonstrated control technology, processes,
and operating methods.  Where practicable, no pollutant dis-
charge at all is supposed to be allowed.  Where pollutant  dis-
charge is unavoidable, these standards are to represent the
greatest degree of effluent reduction achievable.  NSPS applies
only to direct dischargers.

     Pretreatment Standards for New Sources  (PSNS) will regulate
indirect dischargers.  The Act states that pretreatment standards
shall prevent the discharge to a publicly-owned treatment  works
(POTW)  of any pollutant that may interfere with, pass through,
or otherwise be incompatible with the POTW.  The amendments to
the Act further stipulate that industrial discharges must  not
interfere with use and disposal of municipal sludges.

     Both NSPS and PSNS apply specifically to new sources.  New
sources are defined as any building, structure, facility,  or in-
stallation that discharges pollutants and for which construction
is started after proposal of the relevant standards.  Captive
shops and job shops may be regulated under either NSPS or  PSNS.

     One option was selected for all new source standards.  The
NSPS/PSNS treatment system consists of the Option 1 end-of-pipe
treatment system (described earlier in this report)  plus in-
plant controls for cadmium.  In-plant controls could include
evaporative recovery, ion exchange, and recovery rinses.   The
purpose of these in-plant controls is to reduce cadmium concen-
tration levels in the raw waste stream.

     Due to the nature of this option, only new sources that
perform cadmium plating operations will incur additional com-
pliance requirements beyond the proposed BPT/BAT/PSES stan-
dards.   EPA estimates that between 13 and 16 percent of exist-
ing sources plate with cadmium.  This information is presented
in Exhibit IX-1.  It is likely, therefore, that the NSPS/PSNS
requirements will concern only a small segment of the metal
finishing population.
                           IX-1

-------
                        EXHIBIT IX-1
               Prevalence of Cadmium Platers
                  Amongst Metal Finishers*

                                 Metal Finishing Subcategory
                                 Captives           Job Shops

Percent Cadmium Platers           13.0               16.7

*Source: March 5,  1981 memo from Mr. Jack Nash to Mr.  Richard Kinch.
     Information  collected on the existing  cadmium users indi-
cates that they are generally larger and more diverse than non-
cadmium users**,  specifically:

          Job  shop cadmium platers are generally much larger
          than job shops that do not plate  with cadmium.

          Job  shop cadmium platers use twice as many types
          of metals as non-cadmium platers

          Indirect discharging captive cadmium platers con-
          sume twice as much water as either their direct
          discharging counterparts or the non-cadmium users.

          Direct  discharging cadmium platers use less water
          than all other types of captives, but they generally
          employ  more people and work with  more metals.

     The incremental cost to new sources of controlling cad-
mium was used  as  the basis for measuring these standards'
competitive effects.  Annual control costs  were calculated
for five different water flow categories.   The average cost
results are presented in Exhibit IX-2.  The data indicates
that a plant's annual costs of cadmium control will vary be-
tween $17,800  and $24,100.***  These costs  are insensitive to
a plant's water flow volume; i.e., unlike the existing source
standards, there  is not a strong functional relationship be-
tween a plant's water flow and compliance costs.
     This information is based on two memos.  The first is dated
     March 5, 1981 and is from Mr. Jack Nash to Mr. Richard Kinch.
     The second is dated May 1, 1981 and is from Mr. Lior Samuelson
     to Mr. Art Herman.

     Annual costs are derived from investment cost data in the
     Technical Development Document. They are annualized according
     to information contained in the 1976-77 Survey of Metal Finishers.
                           IX-2

-------
                       EXHIBIT IX-2
        Incremental NSPS/PSNS Annual Costs by Metal
             Finishing Plant Water Flow Category
                         (in 1980 $)

       Plants With Metal            Incremental Annual
      Finishing Flow (GPP)           Costs of Option 3

             <1,000                    $  21,924
           1,000-10,000                   18,802
          10,000-50,000                   21,791
          50,000-500,000                  17,811
            >500,000                     24,090

       Source:  Technical Development Document
     The cost effects for five plant sales categories  due  to
NSPS/PSNS requirements were calculated by the  following  three
step process:

          For each plant sales category, model plants  were
          clustered according to water flow.*

          Within each sales category, cost effects were  cal-
          culated for each water flow cluster by dividing  the
          incremental Option 3 compliance cost by plant  sales.
          This approach is similar to the one used to  measure
          economic impacts of Options 1 and 2 on the captive
          metal finishing population.

          A weighted average cost effect was calculated  for
          each plant sales category based on the relative
          importance of the water flow clusters.

     The results of these calculations are presented in
Exhibits IX-3 to IX-8.  The cost effects range from  .04  per-
cent for the largest plants to 2 percent for the smallest.
In general, as plants' sales volumes grow larger, the  cost
effects become very small.  In addition, the cost effects  for
nearly 90 percent of the plants  (assuming that new sources
have a size distribution equivalent to existing sources) are
.7 percent or less.
     Source:  1976-77 Survey of Metal Finishers.
                           IX-3

-------
   C
   O
   •H
   cn
   tu
   rH
   (0
   Ul

   s:
   -P
   •H
rri ^
 i    ;
X tn<
H -P

EH (0 <
H rH <
03 Cli '
H
33 SH
x o-
W 4H

   cn
   -P
   u
   0)
   4H
   4H
   W
    cn
    o
   u

   C/3
   eu

   2
-p
u
-P 0)
C 4-1
0) 4H
U W
J_l
(U 4J
Cn 10
0
u
.p
rH cn
to o
-P\U
C rH
a) to n
Er-l
-J
CD C C
!H C O
U r< -rH
C 4J
H a
o
C
o
•rH
rH CO
rH -rH
•-H .c
S EH
3
rH £! 0
V> -P rH
•H fa
4H S
o
cn fa
M .]_) v
CCS
CU 03
O rH




ctf>
CN O> CN 00 ^
CN rH CN rH CN






•^ CN rH rH O
CN O CTi rH 0s!
01 co r^ oo o
rH 00 rH I-» ^T
CN rH CN rH CN


-------
       4->
       U
    4-1 0)
    C 4-1
    0) 4-1
    U H


    
0
4-1 fo
o tn •
4-1 a
4-> C
C (0
O CU

0)



                                     m
                                     CN
4-1
4-1
cn
0
U

cn
2
CO
a
\
CO
Cn
CO
2
    Cn
    C
   •H  -
    C
   •H
    (0  Pn
    -p
             O   O
             O   (H
             O    I
              -   O
             i—I   O
             w   o
          o
o   o   o
000
o   o    •>
 ••    ••   o
o   o   o   o
<—i   in   in   o
           I    O
          o
     o   o   o
     000
      •    -   m
     °   °   A
     rH   m
 i
o
                    en
                    Cn
i
Perc
Exi
                          OO

                          CN
                                                IX-5

-------
   c
   o
   •H
   •H
   2

   in
    •
   r-
   (0
   CO
   .p
   •H
   s
H
03
    u
    S
*
4-1 01 fa
O -P •
C 2
-P «S
C rH
ni o
UJ M-(
O
c.
rH
0)
ft


CW
ro ro ro CN ro
• • • • •
0 0 0 0 O






•=»< CN rH rH O
(N O Cn rH CTv
c^ co r^ co o
rH CO rH r» i*
CN rH CN rH CN

V*-












CO VO O CN •*
CN ro ro









C7<
c
•r^ *^~*
J2 Q
01 04
•H e>

•H
fa 3
o
rH rH
03 fa
-P
CU
2





o
0
o
^
rH
V




0
o
o
•.
o
rH
1
O
0
O
v
r-H



O
0
o
*
0
in
I
o
o
0
•»
o
H

O
O
0
^
o
0
ITl
1
0
o
0
*
o
LO





o
o
0
^
o
0
in



                      01
                      •p
                   4-1

                   O
                            co
                    03

                   04
                                                IX-6

-------
   c
   o
   S _

X CO <*

Msg

HHS
CQ 04 ^
M    e
X rH .5
x o C.
w -H
S •
iti rri
**"• W
O eo •
-U S
-P C
C (C
CD rH
U 04
SH
CD
04


dft
Oi 00 W [-• O
O O O O r-H
O O O O O






•<* CN rH rH 0
CN O  ^r
CN H CN rH CN

•00-











VO *^
•H
fr4 S
O
__j __i
^^ r~^
(0 PH
JJ
CD
2
, CO
rH ^
<~H — •
< £
rH
HH 71
0 fc
Percent
Existing
000
o o o
00-
- - o
0 O O O O
o i— i m m o
0 1 1 1 O
- o o o -
rH O O O O
.. o o o o
— — — in
rH 0 O *
rH IP ^






00
•
in
ro
                                IX-7

-------
                   -P
                   o
                4J 0)
                C 4H
                CD 4H
                U W

                CD -P
                HI cn
                   O
                   U
           o  o  o  o  o

           o  o  o  o  o
   C
   o
  •H
  •H
  2

  o
  in

rH
nj
i ^
C
(D
s
CD
rH
U
C



^v^
rH
(0
3
C
C




CN
0
CO
>.
CO
rH


                                     co
                                CN
                                         O
                                         O-l
                                         o
                                         CN
   a)
   cn
X
EH cn co
1  I I I o^

CQ C5 i—(
H ffl
a rH  C
X ft -H
H   -^

   O
   4-4
 C
 o
•H

rH
•H
2

O
in
•o>
Jr>

 C
 0)

 U


 0)
                   o
                   rH
                   CM
                         CN
                                 O   rH
                                 CN   m
   cn
   -P
   o
   Q)
   MH
   4H
   W
   03
   O
   U

cn
C
•H *-•
.C Q
cn ft
•H O
c —
•H
CM 3
o
rH rH
ra EM
-P
o
O
0
-
0 O
0 rH
0 1
- O
rH 0
N/ ^
rH


O
O
O
*h
0
in
1
0
o
o
o
rH
o
0
o
^
0
0
m
I
o
o
o
V
o
m



o
o
o
i»
0
o
m
A


   03
   ft
   cn
   2
                 
-------
                   EXHIBIT IX-3
         NSPS/PSNS Average Cost Effects
                   (in 1980 $)
                Percent of          Average Cost
  Sales        Existing Plants      Effects Due to
Category     in Sales Category     Option 3 Costs
  1mm               11.0%                2.0%
  3mm               23.2                0.7
  7.5 mm            14.8                0.3
 25mm               35.8                0.08
>50mm               15.2                0.04
                             IX-9

-------
     Based on the available data, it does not seem that the
additional costs to comply with the NSPS/PSNS standards will
erect significant entry barriers or create competitive dis-
advantages.  The main reasons are the following:

          The incremental cost of compliance with Option 3
          is small.   (Between 0.04 percent and 2.0 percent
          of the value of sales).

          Only a small percentage of the metal finishing
          universe uses cadmium.  Assuming that this trend
          holds for new source metal finishers, the vast
          majority of plants will have no compliance re-
          quirements at all due to NSPS/PSNS cadmium controls.

          In some processes, cadmium plating may be sub-
          stituted for by other metals.  This will relieve
          these new sources of NSPS/PSNS cadmium controls.
          In cases where substitution is not possible due
          to the characteristics of cadmium, cost can be
          passed through as a result of inelastic demand.

     In general, the decision of a firm to enter into the
market for metal finishing will be insensitive to the incre-
mental burden of NSPS/PSNS.  For captives, metal finishing
is an input into the production process.  As such, they per-
form metal finishing in-house to ensure continuous supply,
to minimize work flow disruptions, and to lower transporta-
tion and packaging costs.  If the relative costs of NSPS/
PSNS were high, captives could decide to obtain finishing
services from existing job shops.  New job shop sources will
have the same effluent guideline requirements as existing
sources unless the new plants use cadmium.  The decision to
use cadmium will be dictated by the demand for cadmium plat-
ing.  Elastic demand will mean a high probability of no cad-
mium use while inelastic demand means little or no competitive
impacts.
                            IX-10

-------
               X.   LIMITS OF THE ANALYSIS
    The purpose of this chapter is to summarize the issues
that bear upon the "power" of the findings presented here-
in.  The data and analytic constraints must be understood
in order for the estimates of industry impacts to be held
in perspective.  Accordingly, the applicability of the
results rests with how well the data, logic and assump-
tions of the models reflect reality.

    The focus of this chapter are the major limitations
involving study issues relating to the:

         Quality and quantity of the data
         Agency survey updates on equipment in place
         Type of models used.

1.  QUALITY AND QUANTITY OF THE DATA

    A major strategic consideration in the planning of
this study was the appropriate source of information.
Age'ncy budgetary and timing constraints regarding new
financial and economic survey work necessitated the use of
data from prior analyses.   This decision may have a
bearing on the findings of this study.  The Economic
Impact Analysis relies on data obtained in the 1976-77
Survey of Metal Finishers.  The forecasted impacts may be
sensitive to any change in the economic and/or financial
characteristics of the Metal Finishing universe between
1976-77 and 1982.

2.  AGENCY SURVEYS ON EQUIPMENT IN PLACE

    The Agency conducted two surveys of the Metal
Finishing universe.  The purpose of the surveys was to
update information on the proportion of metal finishers
that would require additional pollution control equipment
in order to meet proposed BAT regulations.  The survey of
direct dischargers* was conducted on a sample derived from
the NPDES permit records,  while the surveys of indirect
    The survey's sample design and selection procedure is
    described in an August 20, 1981 memorandum from
    Mr. Richard Kotz to Mr.  David Pepson.   The results of
    the survey are contained in a memorandum from
    Mr. Henry D. Kahn to Mr. David Pepson

                            X-l

-------
dischargers* relied on the technical contractor's data
base.  The projected magnitude of compliance costs and
plant closures is sensitive to any changes in the results
of the surveys due to sampling or data base biases.

3.  TYPE OF MODELS USED

    Two distinct models were used to analyze the economic
impacts.  In general, the design of the models was
constrained by data availability.  The job shop model is
essentially a capital budgeting closure analysis that
assumes full cost pass through.  Different assumptions on
the elasticity of demand, interest rates, and the model
plants' financial condition may yield different results.

    The model used to analyze the impacts on the captive
sector assumes various critical price increase limits.
Any plant projected to require a greater price increase is
considered a candidate for closure.  A change in the
critical price limits would alter the results of the
analysis.  In addition, the captive analysis assumes a
full cost pass through.  A precise prediction of price and
output changes due to the regulation is not possible due
to the complexity of the Metal Finishing universe.  The
approach taken here therefore, limits somewhat the
accuracy of the forecasts.
    Results of these surveys are described in a July 30,
    1981 memorandum and a December 3, 1981 memorandum from
    Mr. Richard Kinch to Mr. Art Herman.
                            X-2

-------
                        APPENDIX A
                   INPUT-OUTPUT  ANALYSIS
    This appendix first sets out a framework within which
the interrelationships between consumption, production and
metal finishing requirements can be measured.  Then a
method of quantifying the price relationships of the sec-
tors in an economy/ the input-output price model/ is dis-
cussed.

    All economic activities can De divided into two compo-
nents, final demand and intermediate production.  Final
demand includes the personal consumption expenditures of
consumers, the expenditures of business for capital goods
replacement and/or augmentation, the expenditures of
government at all levels, and the net exports.  Interme-
diate production represents the inter-industry transac-
tions necessary to produce the final goods and services.
The inter-industry model, termed "input-output analysis",
developed by Wassily Leontief [1] and recently estimated
for the year 1972 ay the Department of Commerce, provides
the required framework [2] for estimating and analyzing
tne above mentioned interrelationships.

    The basic input-output structure is developed by
dividing the productive and final demand activities of an
economy into a number of sectors, which are arrayed in
matrix form.  The distribution of the sales and purchases
of each industry is then estimated for each sector during
a 1-year period.  As an example, Table A-l, panel (a),
shows all economic activities divided into three producing
sectors and one component of final demand.  Reading across
the rows, one finds the sales (in dollars) of the output
of the sector named at the beginning of the row to the
sector named at the head of each column.  Or, reading down
the column, one finds the purchases by each sector named
at the head of the column from the sector named at the
beginning of the row.  Final demand can be furcher disag-
gregated into the components used in the national income
accounts.  Thus, the total final demand for the output of
an industry is the sum of those components:

                    Y = C + I + G + I,      (1)
                        A-l

-------
                  Table A-l.  Example ot Input-Output  Tables



                   Ja)  Transactions table, X matrix  (Dollars)
Producing
sectors
Agriculture
Manufacturing
Services
Tot a L int-f»r-
Total inter Total
Consuming sectors mediate final
agriculture
40
40
0
manufacturers
80
40
60
services
0
20
20
output
120
100
80
demand
80
300
120
Gross
Output K
200
400
200
mediate inputs
80
180
40
Value added
Gross inputs
(b)
Producing
sectors
Agriculture
Manufacturing
Services
Value added
Total
120 220
200 400
Direct requirements

agriculture
0.20
0.20
0
0.60
1.00
(c) Total requirements (direct and
Producing
sectors

agriculture
160
200
500

table, A matrix (Dollars/Dollars)
Consuming sectors
manufacturers
0.20
0.10
0.15
0.55
1.00
indirect) table, S matrix
Consuming sectors
manufacturers

services
0
0.10
0.10
0.60
1.00
(Dollars/ Dollars)

services
Agriculture



Manufacturing



Services
 1.33




 0.30



 0.05
           0.30




           1.20



           0.20
                    0.03




                    0.13




                    1.13
                                        A-2

-------
where*

    Y =  [yjj      final demand for the output of industry
                   i/ where i = 1....n/

    C =  [cjj      personal consumption expenditure compo-
                   nent of final demand for industry  i
                   output,

    I =  [ijj      private investment expenditure compo-
                   nent of final demand for industry  i
                   output,

    G =  [gjj      government expenditure component of
                   final demand for industry i output, and

    T =  [tjj      net export component (exports minus
                   imports) of final demand for industry i
                   output.

    The gross output of an industry is the sum of its
sales to other industries and to final demand:

                        Z  = XL +  Y,               (2)

where

    Z =  [zj_]      gross output of  industry i,

    X =  l*ij]     sales of industry i to industry j,
                   where j = l....n, and

    L =  n         dimensional unit vector.

    Analogously, the gross input of industry is the sum of
its purchases from other industries and of value added:

                        Z  = X'L + V,               (3)

where

    V  = [vj_]      value added by industry i, and

    X*- = [Xji]     purchases of industry from other indus-
                   tries.
  * Square-oracketed, lower-or upper-case subscripted
    variables denote vectors or matrices.
                         A-3

-------
    Gross national product is measured as the sum of final
demand (expenditure approach) or the sum of value added
(income approach).

    Up to tnis point, the input-output table is essenti-
ally a system of accounting identities.  However, in situ-
ations where producers are regarded as having only a
limited choice regarding factor (i.e., input) intensities
and wnere adjustments to shifts in demand take the form of
quantity (i.e., output) rather than price adjustments, the
transactions table can be utilized to develop a general
set of production coefficients.  Specifically, a set of
technical coefficients can be derived from the transac-
tions table.  A technical coefficient is defined as the
dollar input purchases from industry i per dollar output
from industry j, or

                        A = Uij],               (4)

where
    Thus, continuing with our example in Table A-l, panel
(b), the values in each column represent the composition
of input to the industry named at the head of the column.
To produce $1.00 of output, the manufacturing sector re-
quires $0.20 of inputs from agriculture, $0.10 from manu-
facturing, $0.15 from services, and $0.5$ of valued added.

    Substituting the value of x-j_j from equation (4) into
equation (2) yields the result

                        Z= AZ + Y.               (5)

    This is equivalent to

                        (I-A)Z  =  Y,               (6)

where

    I = the identity matrix.

    From aquation  (6) one can find the "total requirements
matrix," S

                          Z = SY,                 (7)

where

    3 =   tsij] =  [I-A]-1.
                         A-4

-------
    Each s-j_j represents the dollar output of industry is
required both directly and indirectly per dollar of final
demand from industry j.

    In Taole A-l, panel (c),  the s-j_j elements of the
hypothetical economy are shown.  Reading down the column,
eacn entry represents the output of the industry named at
the beginning of the row per  dollar of final demand from
the industry named at the head of the column.  Thus, to
deliver $1.00 of manufactures to final demand requires
$0.30 of output by the agriculture sector,  $1.20 by
manufacturing (the $1.00 for  final demand plus the addi-
tional manufacturing output required to produce the
required output of all three  sectors), and $0.20 of the
output of the services sector.

    The three basic input-output tables thus provide the
framework for analyzing the interrelations in an economy.
To summarize:

    1.   The transactions table, X matrix,  shows the flows
         between sectors per  unit of time;

    2.   The direct coefficients table, A matrix, indi-
         cates the direct output requirements of each sec-
         tor to produce one dollar's worth of output by
         every other sector;  and

    3.   The total requirements coefficients table, S mat-
         rix, indicates the total (direct and indirect)
         output of each sector required to deliver one
         dollar's worth of output of every other sector to
         final demand.

    The Bureau of Economic Analysis (SEA),  U.S. Department
of Commerce has developed several input-output tables of
the U.S. economy.  The latest, a 478-producing-sectors
table based on the structure  of production for 1972, has
been employed in this study [2].
                         A-5

-------
(a)
            Table A-2.   Example of Diagrregated Personal
                  Consumption  Expenditures  Tables

Industrial composition of personal consumption expenditures table, U
                          matrix (Dollars)
Producing
sectors
Agriculture
Manufacturing
Services
TOTAL (E)
PCE item
1 2
0 80
200 0
80 20
280 Too
Total final
3 demand i
0 80
100 300
20 120
~L2Q ~500
(b) Distribution of the industrial
composition of personal consumption
expenditures table, B matrix
(Dollars/ Dollars)

Producing
sectors 1
Agriculture 0.00
Manufacturing 0.71
Services 0.29
1.00
PCE item
2 3
0.80 0.00
0.00 0.33
0.20 0.17
1.00 1.00
(c) Total requirements personal consumption
expenditures table, K matrix
(Dollars/Dollars)

Producing
sectors 1
Agriculture 0.22
Manufacturing 0.89
Services 0.47
PCE item
2 3
1.07 0.25
0.27 1.02
0.27 0.36
                                       A-6

-------
2.  pEASQilAL CONSUMPTION EXPENDITURES INTERRELATIONSHIPS

    One drawback of input-output tables in terms of the
objective of this study is their lack of resolution in the
final demand sectors.   Specifically, personal consumption
expenditures are usually represented by a single column
vector.  Because the purpose of the impact model is to
identify interrelationships between personal consumption
expenditure items and the requirements for materials pro-
duced by the affected sectors, such as metal finishing
requirements, this vector must be disaggregated into a set
of consumer expenditure items.  Unpublished BEA data used
in developing the 1972 input-output table were used to
develop a "bridge" between the 478 producing sectors in
the input-output table and 477 personal consumption ex-
penditure items,,
    Assuming there are m consumer products, one can define
aggregate expenditures on these products as

    E =  [em]      the total dollar expenditure
                   on each consumer product at
                   retail prices.                (8)

    The Personal Consumption Expenditure (PCE) oridge is
the allocation of the PCE vector among the m consumer
expenditure items:

                          C  = UL,                 (9)

where

    C =  [cj_]      the personal consumption expenditure
                   component of final demand for industry
                   i output,


    L =  [IjJ      m dimensional unit vector, and

    u =  LuinJ     the dollar amount of final demand sales
                   from sector  i required for production
                   or distribution of consumer product m.

    Using the hypothetical economy aoove, suppose all
final demand was for three consumption goods which could
be disaggregated as shown in Table A-2, panel (a).

    The bridge can be converted into a set of fixed coef-
ficients in a manner similar to that employed in develop-
ing the direct coefficients:
                         A-7

-------
                   B = bi,a  =  uim,              (10)
where

    bj_n =     dollar of final demand sales from sector i
              directly required per dollar expenditure on
              consumer product.

    The distribution of the industrial composition of each
expenditure item in the hypothetical economy is shown
Table A-2, panel (b).

    Last, from equations (7)  and (10)  we can create a
matrix of the total output of each sector required per
dollar of PCE on each item:

                          K = S3.                (11)

    The final matrix, Table A-2, panel (c), shows tne out-
put (in dollars) required by each industry named at the
beginning of the row to produce and deliver one dollar's
worth of the personal consumption expenditure item named
at the head of the column.

    In this hypothetical economy, private investment,
government expenditures, and net exports are zero.  There-
fore, from equation  (1) it is apparent that aggregate PCE
expenditures are assumed to equal aggregate income, i.e.,
C = Y.  Therefore, equations (7), (9), and (10) can be
combined to state the further requirement that total pro-
duction in the hypothetical economy equals the sum of the
output requirements of all PCE items;  i.e.,
                    Z  =  3Y  =  SUL  =  SBE,
or                                              (12)
                          Z = KE.
                         A-3

-------
So the effect of a 10 percent price increase in manufac-
turing has been a 2.5 percent increase in agriculture
prices and a 1 percent increase in the price of services.

    Using the share of final demand accounted for by each
sector, a cost of living index can be computed.  To illus-
trate this idea, recall the total final demand column from
Taole A-2, panel (1), in which the sector components of
personal consumption expenditures were as follows:
agriculture, 80; manufacturing, 300; services, 120.  Thus,
total final demand, or personal consumption expenditures,
is 500, so the relative share for each sector is,
respectively,

          80, 300 and 120.
         500  500     500

The respective prices from the preceding example are p]_
= 1.025, p2 = 1.10 and P3 = 1.01, so a cost-of-living
index, C  , can be obtained as follows:
       = 1.025
1.10
1.01
120
500
or
    c  = 1.0664
Thus, the increase in the cost of living as a result of a
10 percent price increase in manufacturing, was 6.6
percent.
                        A-ll

-------
4.  ESTIMATES OF PRICE IMPACTS OF EPA REGULATIONS

    The new level of prices for the 477 PCE items due to
the proposed EPA regulations are determined following the
increases of the price level of the metal finishing uni-
verse  (SIC's 34, 35, 36, 37, 38, 39).  Exhibit A-3 indi-
cates the estimated price increases for each 2-digit SIC.
The price impact on the major components of total final
demand in the economy is computed based on the price in-
crease in the metal finishing universe.
                          A-12

-------
3.  THS PRIGS SYSTEM

    Input-output analysis offers a method of quantifying
the price relationships of the sectors in an economy.
Since the cost of any sector's output is composed of the
costs of the materials inputs purchased from other sectors
and such items as wages, profits, taxes, and depreciation
which are a part of value added, there are implied price
relationships in an input-output table.  By defining unit
price equal to unit cost, the price relationships between
the price of different goods can be expressed as:
   PI » anpi + a2iP2 + ...  anipn +

   P2 = a!2Pl + a22P2 + •••  an2Pn + V2 2
   Pn = alnPl + a2nP2 + • • •  annPn + vn n
                                                      (13)

where

    Pi  = the price of good  i,

    aji = the technical coefficients,

    Vj_  = tne share of primary  inputs, and

     i  = the price of primary  inputs.

    By using the equation  system (7),  the effects of a
price change in one sector upon the relative prices  of
every other affected sector's output can be calculated
assuming each industry passes on its increased costs plus
the rise in costs of inputs  purchased from other indus-
tries to final demand.  For  the exposition of the input-
output price model, refer  to References 4 and 5.

    An example of the procedure for determining price im-
pacts on the value of final  demand is shown below for the
three sector economy shown in Table A-l.   It is assumed
here that prices of all sectors and all primary inputs in
the base periods are equal to unity for the sake of  sim-
plicity.  Further, suppose the  price level in manufactur-
ing increases 10 percent due to the increased costs  neces-
sary to comply with the proposed standard.  The problem  is
to determine the new level of prices for  agriculture and
services.
                         A-9

-------
    When we assuiae that the prices of primary inputs in
other sectors (i.e. /]_ and $3) are not permitted to
cnan-je, the relationship (13) for the three sector economy
can be easily solved as below by first transposing all the
know elements in each of the three equations to the right
hand side and the unknown variables (and their coeffi-
cients) to the left hand side.  Given our assumption, the
"unknowns" are PI, ?3, and $2, and all other elements
are known or assumed known.  Thus, equation (13) for a
three sector economy, with our stated assumption, may be
written as:
                    P2
    Pi = allPl + a3lP3 + <

    V24>2 = -ai2Pi - a32P3 + d -

    P3 = <*13P1 + a33P3 + U23P2

and solved as follows:
PI
P3
 a!2

-a13
° ~ a31

2 a32

0  (1 - 33)
                                -1
                              a2lP2 + vl*l

                            (1 - ai2) P2
«
0.8

0.2

0
                    .55
         .15

         .90
                                 -1
                               2(1.1) + .6(1.0)

                               8(1.2)

                               ,1(1.1) + .8(1.0)
Thus
        1.25

       -  .455

        0
   0      0

   1.313  -.303

   0      1.11

X

".82"
.83
.91

55

1.025
.951
1.0.0
    PI = 1.025

    P3 - 1.011, and

    ?2 = 1.10 by assumption,
                         A-10

-------
                    EXHIBIT A-3

           Estimated Price Increases For
            2-Digit SIC's  (In  Percent)

SIC            Option I             Option  II

 34               0.3                  0.9

 35               0.3                  0.9

 36               0.1                  0.3

 37               0.1                  0.2

 38               0.1                  0.2

 39               0.1                  0.3
                      A-13

-------
                             EXHIBIT A-4

              Estimated  Total Price Impact or EPA BATEA
              Metal Finishing Regulations or Principal
                       Final  Demand Components
Final Demand Sectors	Percentage  Price  Increase
                                            OPT  I	OPT II
Personal Consumption Expenditures             0.01       0.02

Gross Private Fixed Capital Formation         0.005      0.01

Net Exports                                   0.01       0.02

Total Federal Government Purchases            0.007      0.01
                                   A-14

-------
                REFERENCES
Leontief, Wassily, Input-Output Economics, Oxford
University Press, New York, 1966.

U.S. Department of Commerce, Office of Business
Economics, 1972 Input-Output Structure of the U.S.
Economy.

United Nations, Problems of Input-Output Tables and
Analysis, New York, 1966.

-------
                                                              1
,•-••• r rotation

-------