United State*
Environmental Protection
Agency
Office of Water Regulation!
and Standards
Washington, DC 20460
EPA 440/2-86O28
September 1086
Water
ECONOMIC IMPACT ANALYSIS OF
EFFLUENT LIMITATIONS GUIDELINES
AND STANDARDS FOR THE METAL
MOLDING AND CASTING (FOUNDRY)
INDUSTRY
QUANT
-------
ECONOMIC IMPACT ANALYSIS OF EFFLUENT
LIMITATIONS GUIDELINES AND STANDARDS
FOR THE METAL MOLDING
AND CASTING (FOUNDRY) INDUSTRY
September 1985
Submitted to:
Economic Analysis Branch
Analysis and Evaluation Division
Office of Water Regulation and Standards
Environmental Protection Agency
401 M Street, South West
Washington, D.C. 20460
Submitted by:
Policy Planning & Evaluation, Inc.
8301 Greensboro Drive, Suite 460
McLean, VA 22102
Contract No. 68-01-6731
-------
TABLE OF CONTENTS
Page No.
EXECUTIVE SUMMARY 1
I. INTRODUCTION 1-1
II. STRUCTURE OF THE INDUSTRY II-1
A. TECHNOLOGY II-1
B. TRENDS IN INDUSTRY SHIPMENTS II-1
C. NUMBER OF FOUNDRIES AFFECTED BY THE REGULATION .... II-6
D. FIRM FINANCIAL STRUCTURE II-7
E. DISTINCTION BETWEEN JOBBERS AND CAPTIVES II-7
F. ANALYSIS OF IMPORTS AND EXPORTS 11-21
III. METHODOLOGY III-1
A. OVERVIEW III-1
B. ESTIMATION OF THE NUMBER OF AFFECTED PLANTS III-1
1. Baseline Year for Compliance III-2
2. Use of Publicly Available Censuses III-2
3. Incorporation of EPA Survey Data III-3
4. Comparison to Analyses Previously Developed
for This Industry III-3
C. ESTIMATION OF COMPLIANCE COSTS III-4
1. Adjustment of Costs for Revised Production
Estimates III-5
2. Estimate of Cotreatment Savings III-6
3. Development of Annual Costs III-6
D. DEVELOPMENT OF MODEL PLANTS III-7
1. Use of Subcategories III-9
a. Metal Type III-9
b. Size Category III-9
c. Jobber/Captive Category 111-10
d. Economic Quartile Category 111-10
i
-------
TABLE OF CONTENTS (Continued)
Page No.
2. Estimation of Precompliance Financial
Statements III-11
a. Estimation of Average Sales Per Foundry ... Ill11
b. Estimation of Ratios Ill-13
(1) Description of the FINSTAT Database .. 111-16
(2) Use of FINSTAT by EPA Ill-16
c. Construction of Financial Statements 111-17
d. Comparison to Previous Analyses 111-19
3. Incorporation of Cost Estimates 111-21
E. ESTIMATION OF IMPACTS 111-21
1. Choice of Tests 111-21
2. Firm Failure Criterion ........................ 111-22
3. Description of the Threshold Values and
Application of Tests 111-22
a. Return on Assets 111-22
b. Total Debt to Total Assets 111-23
c. Beaver's Ratio 111-24
4. Sample Calculations 111-25
IV. EFFLUENT CONTROL AND GUIDELINE COSTS IV-1
V. ANALYSIS OF ECONOMIC IMPACTS V-1
A. BASIS FOR COMPLIANCE COSTS V-1
1. Option 1: Recycle and Simple Settling V-1
2. Option 2: Recycle, Lime Addition, and
Settling V-2
3. Option 3: Recycle, Lime Addition,
Settling, and Filtration V-2
4. Option 4: Recycle, Lime Addition,
Settling, Filtration, and Carbon
Adsorption V-2
B. ECONOMIC IMPACTS OVERVIEW V-2
1. Plant Closure and Employment Loss Impacts V-2
2. Other Economic Impacts V-7
C. POTENTIAL CLOSURES AND EMPLOYMENT LOSSES FOR
INDIVIDUAL METALS V-7
1. Potential Impacts on Gray Iron Foundries V-8
2. Potential Impacts on Ductile Iron Foundries ... V-8
3. Potential Impacts on Malleable Iron
Foundries V-17
ii
-------
TABLE OF CONTENTS (Continued)
Page No.
4. Potential Impacts on Steel Foundries V-26
5. Potential Impacts on Aluminum Foundries V-26
6. Potential Impacts on Copper-Base Foundries .... V-26
7. Potential Impacts on Zinc Foundries V-35
8. Potential Impacts on Magnesium Foundries V-35
D. SELECTION OF OPTIONS V-44
E. OTHER IMPACTS V-44
1. Potential Price Increases V-46
2. Potential Production Loss Due to the
Regulation V-46
3. Potential Balance of Trade Impacts V-46
4. Community Effects V-50
VI. NEW SOURCE IMPACTS VI-1
VII. SMALL BUSINESS ANALYSIS VII-1
A. SMALL FOUNDRY SIZE CRITERIA VII-1
B. IMPACT ANALYSIS FRAMEWORK VII-1
C. CLOSURES FOR SMALL AND LARGE FOUNDRIES VII-2
D. OTHER POTENTIAL IMPACTS VII-3
E. REDUCTION OF IMPACT ON SMALL BUSINESSES VII-3
F. REGULATORY FLEXIBILITY ACT VII-7
VIII. LIMITATIONS OF THE ANALYSIS VIII-1
A. FORECASTS OF SHIPMENTS VIII-1
B. SELECTION OF RATIOS VIII-1
C. USE OF SAME TESTS FOR CAPTIVES AND JOBBERS VIII-2
D. DERIVATION OF COMPOSITE FINANCIAL STATEMENTS
FROM QUARTILE RATIOS VIII-2
APPENDIX A REVIEW OF FINANCIAL RATIOS AS PREDICTORS
OF BANKRUPTCY A-1
iii
-------
LIST OF TABLES
Table No. Page No.
1 Selected Options for Effluent Guidelines 8
2 Compliance Costs and Economic Impacts Foundry Industry
(Option 1 Recycle/Simple Settle) 10
3 Compliance Costs and Economic Impacts Foundry Industry
(Option 2 Recycle/Lime Addition/Settle) 11
4 Compliance Costs and Economic Impacts Foundry Industry
(Option 3 Recycle/Lime Addition/Settle/
Filtration) 12
5 Compliance Costs and Economic Impacts Foundry Industry
(Option 4 Recycle/Lime Addition/Settle/
Filtration/Carbon Adsorption) 13
6 Potential Production Impacts for Selected Options 16
II-1 Specialization and Coverage Ratios for the Metal
Molding amd Casting Industry II-2
II-2 Quantity of Shipments II-3
II-3 Value of Shipments II-5
II-4 Foundry Population Operating in 1984 II-8
II-5 Projected Number of Active Wet Plants in Industry
(1984) II-9
II-6 Calculation of Average Sales Per Foundry 11-11
II-7 Financial Ratios for Gray Iron Foundries 11-13
II-8 Financial Ratios for Malleable Iron Foundries 11-14
II9 Financial Ratios for Steel Foundries 11-15
11-10 Financial Ratios for Aluminum Foundries II16
11-11 Financial Ratios for Copper Foundries 11-17
11-12 Financial Batios for Nonferrous, NEC Foundries 11-18
iv
-------
LIST OF TABLES (Continued)
Table No. Page No.
11-13 Separation of Ferrous Employment-Size Segments
Between Jobber and Captive Foundries 11-19
II-14 Separation of Nonferrous Employment-Size Segments
between Jobber and Captive Foundries 11-20
11-15 List of End Markets Used for Export Analysis 11-23
11-16 Results of Import/Export Analysis 11-24
III-1 Sample Compliance Costs Per Plant - Aluminum III-8
III-2 Trends in Shipments Ill-12
III-3 Calculation of Average Sales Per Foundry 111-14
III-4 Sample Derivation of Precompliance Financial
Statements 111-20
III-5 Sample Derivation of Postcompliance Financial
Statements 111-26
IV-1 Compliance Costs and Economic Impacts Foundry Industry
(Option 1 Recycle/Simple Settle) IV-2
IV-2 Compliance Costs and Economic Impacts Foundry Industry
(Option 2 Recycle/Lime Addition/Settle) IV-3
IV-3 Compliance Costs and Economic Impacts Foundry Industry
(Option 3 Recycle/Lime Addition/Settle/
Filtration) IV-4
IV-4 Compliance Costs and Economic Impacts Foundry Industry
(Option 4 Recycle/Lime Addition/Settle/
Filtration/Carbon Adsorption) IV-5
IV-5 Contribution of the Most Important Discharger Process or
Process Combination to the Total Cost for Each Metal IV-7
V-1 Compliance Costs and Economic Impacts Foundry Industry
(Option 1 Recycle/Simple Settle) V-3
V-2 Compliance Costs and Economic Impacts Foundry Industry
(Option 2 Recycle/Lime Addition/Settle) V-4
V-3 Compliance Costs and Economic Impacts Foundry Industry
(Option 3 Recycle/Lime Addition/Settle/Filtration) V-5
v
-------
LIST OF TABLES (Continued)
Table No. Page No.
V-4 Compliance Costa and Economic Impacts Foundry Industry
(Option 1 Recycle/Lime Addition/Settle/Filtration/
Carbon Adsorption) V-6
V-5 Compliance Costs and Economic Impacts Gray Iron
(Option 1 Recycle/Simple Settle) V-9
V-6 Compliance Costs and Economic Impacts Gray Iron
(Option 2 Recycle/Lime Addition/Settle) V-10
V-7 Compliance Costs and Economic Impacts Gray Iron
(Option 3 Recycle/Lime Addition/Settle/
Filtration) V-11
V-8 Compliance Costs and Economic Impacts Gray Iron
(Option 4 Recycle/Lime Addition/Settle/
Filtration/Carbon Adsorption) V-12
V-9 Compliance Costs and Economic Impacts -- Ductile Iron
(Option 1 Recycle/Simple Settle) V13
V-10 Compliance Costs and Economic Impacts Ductile Iron
(Option 2 Recycle/Lime Addition/Settle) V-14
V11 Compliance Costs and Economic Impacts Ductile Iron
(Option 3 Recycle Lime Addition/Settle/
Filtration) V-15
V-12 Compliance Costs and Economic Impacts Ductile Iron
(Option 1 Recycle/Lime Addition/Settle/
Filtration) V 16
V13 Compliance Costs and Economic Impacts Malleable Iron
(Option 1 Recycle/Simple Settle) V-18
V-14 Compliance Costs and Economic Impacts Malleable Iron
(Option 2 Recycle/Lime Addition/Settle) V-19
V-15 Compliance Costs and Economic Impacts -- Malleable Iron
(Option 3 Recycle/Lime Addition/Settle/Filtration) V-20
V16 Compliance Costs and Economic Impacts Malleable Iron
(Option 1 Recycle/Lime Addition/Settle/
Filtration/Carbon Adsorption) V-21
V-17 Compliance Costs and Economic Impacts Steel
(Option 1 Recycle/Simple Settle) V-23
vi
-------
LIST OF TABLES (Continued)
Table No. Page No.
V-18 Compliance Costs and Economic Impacts Steel
(Option 2 Recycle/Lime Addition/Settle) V-24
V-19 Compliance Costs and Economic Impacts Steel
Option 3 Recycle/Lime Addition/Settle/
Filtration) V-25
V-20 Compliance Costs and Economic Impacts Steel
(Option 11 Recycle/Lime Addition/Settle/
Filtration/Carbon Adsorption) V-26
V-21 Compliance Costs and Economic Impacts Aluminum
(Option 1 Recycle/Simple Settle) V-27
V-22 Compliance Costs and Economic Impacts Aluminum
(Option 2 Recycle/Lime Addition/Settle) V-28
V-23 Compliance Costs and Economic Impacts Aluminum
(Option 3 Recycle/Lime Addition/Settle/
Filtration) V-29
V-24 Compliance Costs and Economic Impacts Aluminum
(Option 4 Recycle/Lime Addition/Settle/
Filtration/Carbon Adsorption) V-30
V-25 Compliance Costs and Economic Impacts Copper
(Option 1 Recycle/Simple Settle) V-31
V-26 Compliance Costs and Economic Impacts Copper
(Option 2 Recycle/Lime Addition/Settle) V-32
V-27 Compliance Costs and Economic Impacts Copper
(Option 3 Recycle/Lime Addition/Settle/
Filtration) V-33
V-28 Compliance Costs and Economic Impacts Copper
(Option 1 Recycle/Lime Addition/Settle/
Filtration/Carbon Adsorption) V-31
V-29 Compliance Costs and Economic Impacts Zinc
(Option 1 Recycle/Simple Settle) V-36
V-30 Compliance Costs and Economic Impacts Zinc
(Option 2 Recycle/Lime Addition/Settle) V-37
vii
-------
LIST OF TABLES (Continued)
Table No. Page No.
V-31 Compliance Costa and Economic Impacts Zinc
(Option 3 Recycle/Lime Addition/Settle/
Filtration) V-38
V-32 Compliance Costs and Economic Impacts Zinc
(Option 4 Recycle/Lime Addition/Settle/
Filtration/Carbon Adsorption) V-39
V-33 Compliance Costs and Economic Impacts Magnesium
(Option 1 Recycle/Simple Settle) V-40
V-3^ Compliance Costs and Economic Impacts Magnesium
(Option 2 Recycle/Lime Addition/Settle) V-41
V-35 Compliance Costs and Economic Impacts Magnesium
(Option 3 Recycle/Lime Addition/Settle/
Filtration) V-42
V-36 Compliance Costs and Economic Impacts Magnesium
(Option 4 Recycle/Lime Addition/Settle/
Filtration/Carbon Adsorption) V-43
V-37 Selected Options for Effluent Guidelines V-45
V-38 Price Pass Through Requirements for All Regulated
Segments V-47
V-39 Potential Production Impacts for Selected Options V-49
V-40 List of Regions and States Within Regions V-51
V-41 Projected Regional Distribution of Closures in
Employment-Size Segments V-52
VII1 Annual Compliance Costs as a Percentage of Sales for
Affected Small and Large Foundries VII-4
VII-2 Annual Compliance Costs as a Percentage of Cost of
Production for Affected Small and Large Foundries VII-5
VII-3 Change in Return on Assets for Affected Small and
Large Foundries VII-6
viii
-------
EXECUTIVE SUMMARY
-------
EXECUTIVE SUMMARY
A. PURPOSE
This report presents the results of the economic impact study
developed for the effluent guidelines, limitations, and standards
applicable to the metal molding and casting (foundry) industry. These
regulations are based on Best Practicable Technology Currently Available
(BPT), Best Available Technology Economically Achievable (BAT), New
Source Performance Standards (NSPS), and Pretreatment Standards for New
and Existing Sources (PSNS and PSES) which are being issued under
authority of Sections 301, 304, 306, and 307 of the Federal Water
Pollution Control Act, as amended by the Clean Water Act of 1977. The
primary economic impact variables assessed in this report include the
costs of the effluent regulations and the potential for these
regulations to cause plant closures, the increase in foundry costs as a
percentage of sales, and impacts on small businesses.
B. INDUSTRY COVERAGE
For purposes of this study, the foundry industry consists of plants
that cast one of the following metals:
The analysis in this study differentiates between Jobber operations and
captive operations. A plant is considered to be a captive operation if
more than 50 percent of its output is consumed by its parent company.
Conversely, a plant selling 50 percent or more of its output to outside
customers is considered to be a Jobber. Publicly available data show
that most foundries produce castings of a single metal.
C. METHODOLOGY
EPA anticipates that all direct dischargers will have complied with
this regulation by 1986, and that all indirect dischargers will have
complied by 1988. In estimating impacts, however, EPA has assumed a
common basis of compliance in 1986. (Because of Increasing levels of
castings shipments, this assumption may tend to overstate the impacts on
indirect dischargers.) EPA then reviewed the characteristics of each
metal subcategory over the 1978 to 1981 time frame. From this
historical base, estimates of the population of plants and the number of
employees per plant were made. In addition, a variety of sources were
used to estimate plant financial characteristics, including shipments
and financial ratios. Finally, the likely plant-speciflo responses to
the Imposition of compliance costs were assessed. Supplemental analyses
Ferrous Metals
Nonferrous Metals
Gray Iron
Malleable Iron
Ductile Iron
Steel
Aluminum
Copper-base
Zinc
Magnesium
-1-
-------
were used to link the conditions in the foundry industry to other
effects such as community and balance of trade Impacts.
1. Description of Industry Characteristics
The first step in the analysis is to develop a description of
the basic industry operations and financial characteristics. For the
purposes of this report, the operating characteristics included:
the number of foundries casting each metal type;
the distribution of foundries by the number of employees in each
plant; and
the value of shipments.
Data were collected for the period 1960 through 1984. From this
information, EPA Judged that the population of foundries in 1984
represents a reasonable estimate of the population that will exist in
1986. Principal support for this judgment is the projected rise in
castings shipments countered by long-term trends of consolidation.
Assessing the basic financial characteristics of the industry
required data from the Small Business Administration's (SBA's) FINSTAT
database, the Census Bureau's Annual Survey of Manufactures, and
corporate annual reports. The financial information is condensed into
the following financial ratios:
return on sales;
sales to net worth;
debt to net worth;
net fixed assets to net worth;
gross fixed assets to net worth; and
depreciation to gross fixed assets.
Estimates of average shipments in 1978 were taken from EPA's
survey of 438 wet foundries over the 1978-1983 time period. The average
1978 shipments were revised downward by the ratio of the industry
shipments decline between 1978 and 1982. In making this adjustment, EPA
is compromising between the shipments data supplied by firms in its
database and average plant shipments reported in other sources, and is
accounting for the economic downturn that occurred in the industry
during the early 1980s.
These estimates of sales, combined with estimates of financial
ratios, are used to establish the overall financial statement of model
firms in the industry. These statements form the basis for determining
the impacts of compliance costs on model foundries, which are then
projected to the industry as a whole.
2. Establishing the Affected Population
EPA estimated the affected population in three steps. First,
EPA used the 1978 directory of foundries published by Penton
-2-
-------
Publications to prepare a list of potentially affected foundries.
Second, EPA surveyed a sample of the industry to determine the types of
discharging processes, the nature and amounts of effluent discharged,
and plant operating characteristics typical of the industry. Two
results of this survey were a revised estimate of the number of
foundries in each of the ferrous subcategories and an estimate of the
proportion of wet plants in each metal and size category. The third
step was to develop a count of the number of foundries in 1984 by
tracing openings and closings from the 1983 directory of foundries
issued by Penton Publications.
Because of the large discrepancies between the categorization of
ferrous foundries in the Penton directory and the results of EPA's
survey, EPA has not tried to extrapolate the foundry population to
1986. Because the foundry population has shown steady declines over the
past 20 years, EPA does not anticipate any noticeable growth in the
number of potentially affected foundries. At the same time, however,
EPA expects the recent and anticipated increases in foundry shipments to
prevent significant numbers of closures between 1984 and 1986.
3. Costs of Compliance
The water treatment control systems, costs, and effluent
limitations and pretreatment standards recommended for the foundry
industry are presented in the Development Document for Effluent
Limitation Guidelines and Standards for the Metal Molding and
Casting (Foundry) Point Source Category. The Development Document
contains detailed descriptions of the technologies recommended and the
development of their costs, as developed by EPA's technical staff. For
this economic analysis, EPA has made four modifications to the treatment
costs presented in the Development Document. First, EPA reduced the
treatment costs to coincide with the production figures used in the
economic analysis. Second, EPA has considered savings attributable to
the cotreatment of several discharging processes at a single plant.
Third, EPA has combined the operating and maintenance costs with
annualized capital costs in estimating the increases in plant production
costs. Fourth, EPA has considered possible Increases in compliance
costs by the 30 percent of plants found to commingle large volumes of
noncontact cooling water with small volumes of regulated waste streams.
4. Plant Closure Analysis
The pro forma financial statements developed for foundries in
each size category are formed from metal type-employment size level
financial ratios and plant-level value of shipments. Financial ratios
in 1986 are taken to be a composite of financial ratios for the 1975 to
1984 time period. Plant level shipments data are estimated based on
average yield and production data from EPA's collection of 438 data
collection portfolios (DCP's) from discharging foundries. The values
estimated from the DCP's have been adjusted downwards to capture
reductions in shipments during the recession in 1978-1982 and to make
the values more consistent with average sales data from other sources.
-3-
-------
The values of financial ratios for 1978 are used with projected
plant-level shipments to construct model 1985 financial statements.
Compliance costs are then imposed on the model plant financial
statements. The following three closure tests and the plant closure
criterion are used to assess the impacts of these costs. Where
estimated postcompliance ratios exceed the boundaries chosen (threshold
values), the affected foundries are considered potential closures. The
selection of ratios and the threshold values were based on an extensive
literature search and on data from actual foundries. The tests are
applicable to both jobbers and captives.
a. Return on Assets Test
This is a profitability test; it measures a foundry's
efficiency at generating income from its asset base. It is defined as
net income after taxes divided by total book assets. A threshold of 2.5
percent is used based on Beaver's work (1966) and a recent examination
of data for failed foundries as well as for foundries still solvent.
b. Total Debt to Total Assets Test
For this solvency test, total debt is considered to be any
liability that is not owner's equity. Beaver's cut-off for debt to
assets ranged from 50 percent to 57 percent. Due to recent structural
changes in the economy, it is not uncommon for foundries to carry
upwards of 60 percent to 65 percent debt. Based on a review of
foundries that have actually filed for bankruptcy, a threshold of 70
percent is used.
c. Beaver's Ratio
Beaver's ratio, defined as cash flow to total debt, is
another solvency test. Cash flow is measured as net income after taxes
plus depreciation. Total debt is assumed the same as in the debt to
assets test. A threshold of 8 percent has been chosen.
d. Plant Closure Criterion
Foundry closure estimates are based primarily on the
quantitative estimates of after-compliance profitability and the ability
to raise capital developed in the above tests. A foundry falling any
two of the three tests Is considered a potential closure for the
purposes of this analysis.
The identification of potential closures In this analysis
should be interpreted as an indication of the extent of plant impact
rather than as a prediction of certain closure. The decision to close a
foundry also involves consideration of other, highly uncertain and
unquantiftable factors. However, the Agency's review of recently
bankrupt metals companies, financial literature, Dun & Bradstreet
composite ratios, and case-study foundries all tend to support the
method chosen.
-4-
-------
5. Other Economic Impacts
Besides potential foundry closures, other economic impacts are
also analyzed and presented. These include effects on employment,
communities, production, foreign trade, and small businesses.
a. Employment Impacts
In developing the model plants, changes in average plant
employment for each employment size category of each metal were
considered. Estimated employment impacts are based on the number of
forecast closures and the average number of employees per closed plant.
b. Potential Price Impacts
To estimate potential price impacts, it was assumed that
foundries could pass along the entire cost of the regulation to
customers. Thus, the potential price impact is the ratio of the annual
cost of the regulation to annual revenues of foundries incurring
costs. (It should be noted that the assumption used for estimating
closures was that no price pass-through would be possible, and foundries
would bear the entire cost of the regulations.)
c. Production Impacts
To estimate potential production impacts, it was assumed
that production capacity is proportional to sales. Thus, the proportion
of sales accounted for by the model foundries forecast to fall Is
assumed to equal the proportion of foundry production capacity lost. It
is likely, however, that production lost by closed foundries will be
made up by expanded production at remaining foundries.
d. Balance of Trade Impacts
To estimate potential impacts on the balance of trade,
imports and exports of castings both as castings and as components of
products in end markets were considered. Of principal concern were both
the trends in the balance of trade and the importance of imports and
exports to the overall domestic castings market.
e. Community Effects
To estimate potential community effects, it was assumed that
the plants that close are distributed nationally in the same manner as
the total foundry population. This is necessary because the analysis
uses model foundries. There is no way to determine either which actual
foundries may correspond to the model foundries, or where,
geographically, the specific impacts may occur.
f. Small Business Impacts
To measure the impacts on small businesses, both model
compliance costs and financial data were gathered for plants of
-5-
-------
different sizes. After classifying the foundries by metal type and size
(number of employees), foundries with 50 employees or less were
identified as small businesses for this analysis. In establishing small
business cut-offs for the final rule, EPA will relate production per
plant to the employee size groups in this analysis.
D. INDUSTRY CHARACTERISTICS
EPA has determined that there were approximately 3,850 foundries in
the U.S. in 1984. Of these, 1,059 generated process waste waters,
including 499 indirect dischargers, 301 direct dischargers, and 259 zero
dischargers. Only indirect and direct dischargers will incur capital
and annual costs as a result of this regulation. Therefore, only these
dischargers are analyzed in this report. As a basic industry, foundries
are found throughout the country, with some concentration in the
Industrial areas of the East (New York, New Jersey, Pennsylvania,
Massachusetts), the Midwest (Illinois, Indiana, Missouri, Ohio), the
South (Texas), and the West (California).
Castings are critical components for various durable goods
industries. In fact, 90 percent of all durable goods contain some
castings. The castings industry was hit hard by the 1982 recession with
capacity utilization and employment suffering their worst declines in
many years.
Historical data indicate that the foundry industry picked up sharply
after the recession, showing increases in shipments ranging from 14 to
43 percent (depending on metal cast) between 1982 and 1984. Shipments
growth for ferrous castings is expected to be 2 percent through the rest
of the decade, while growth of shipments of nonferrous castings is
expected to be 5 percent per year through the end of the decade
(Department of Commerce, Bureau of Industrial Economics).
Competition among metal types is strong in certain markets. The
automotive industry is moving towards the use of lightweight castings as
it strives to increase mileage ratings. Ferrous foundries must attempt
to satisfy the need for these special castings in the face of
competition from aluminum and zinc foundries. New markets must be found
for all foundries that feel the effects of product substitution.
E. BASIS FOR COMPLIANCE COSTS
Listed below are brief descriptions of the various treatment levels
being considered as possible bases for the regulation. A complete
description of these technologies can be found in the Development
Document for Effluent Limitations Guidelines and Standards for the Metal
Molding and Casting (Foundry) Industry.
1. Option 1: Recycle and Simple Settling
Option 1 comprises high rate recycle achieved by settling
(including surface skimming for free oil removed in certain segments)
and recycle to the process (including pH adjustment as required to
-6-
-------
maintain water chemistry balance between scaling and corrosion), and
including cooling towers for some segments, followed by settling of the
blowdown stream prior to discharge. Option 1 costs Include the costs
for the grinding scrubber operations of aluminum, copper, ductile iron,
gray iron, malleable iron, magnesium, and steel plants. Treatment for
grinding scrubber operations is similar to Option 1, but requires
complete recycle with no blowdown, and thus no blowdown treatment.
Option 1 costs were not developed for the aluminum and zinc die casting
segments or for the ferrous dust collection and wet sand reclamation
segments because the treatment systems would be inadequate for the
treatment of wastes from these segments.
2. Option 2: Recycle, Lime Addition, and Settling
Option 2 is designed as an "add-on" to the Option 1 facility and
consists of the addition of flocculation with lime and polymer to
facilitate metals precipitation and solids settling for blowdown
treatment. This option also includes emulsion breaking for the aluminum
and zinc die casting segments and chemical oxidation of organic matter
for these two segments and for ferrous dust collection and wet sand
reclamation.
3. Option 3: Recycle, Lime Addition. Settling, and Filtration
Option 3 adds filtration of the effluent from the Option 2
facility through cartridge filters, multimedia filters, and pressure
filters, depending on the size of the systems.
4. Option 4: Recycle, Lime Addition, Settling, Filtration, and
Carbon Adsorption
Option 4 adds carbon adsorption treatment of the effluent from
the Option 3 facility. Option 4 costs were determined only for those
segments where the Option 3 effluent contained toxic organic chemicals
at a level that could be reduced by this method of treatment.
F. FINDINGS
This section provides a brief summary of the potential economic
impacts.
1. Selection of Options
Table 1 shows the selected options for the foundry effluent
guidelines. The options chosen are based on EPA's estimates of economic
impacts and other factors. Effluent guidelines have been chosen for all
metals except the magnesium subcategory, which is not being regulated.
Effluent guidelines for BPT (Best Practicable Control Technology
Currently Achievable) are set based on removal using Option 2 technology
(partial recycle of process water followed by lime addition and
settling). For steel and aluminum, removals under BAT (Best Available
Technology Economically Achievable), PSES (Pretreatment Standards for
-7-
-------
TABLE 1
SELECTED OPTIONS FOR EFFLUENT GUIDELINES
<1
BPT
BAT
PSES
NSPS
PSNS
Gray Iron
2
3
3a
3
3a
Ductile Iron
2
3
3
3
3
Malleable Iron
2
3b
3b
3b
3b
Steel
2
2
2
2
2
Aluminum
2
2
2
2
2
Copper-Base
2
3
3
3
3
Zinc
2
3
3
3
3
Magnesium
n.r.c
n.r.
n.r.
n.r.
n.r.
Option
1:
Option
2:
Option
3:
Option
4:
Recycle and simple settle
Recycle, lime addition, and settling
Recycle, lime addition, settling and filtration
Recycle, lime addition, settling, filtration, and
carbon adsorption
BPT: Best practicable control technolgy currently available
BAT: Best available technology economically achievable
NSPS: New source performance standards
PSES: Pretreatment standards for existing sources
PSNS: Pretreatment standards for new sources
aFor plants with fewer than 50 employees, PSES and PSNS are set at
Option 2.
^or plants with fewer than 100 employees, BAT, PSES, NSPS, and
PSNS are set at Option 2.
C t
n.r. means not regulated.
-------
Existing Sources), NSPS (New Source Performance Standards) and PSNS
(Pretreatment Standards for New Sources) have been set equal to BPT.
In general, standards for gray iron, ductile iron, malleable iron,
copper-based metals, and zinc have been set at the more stringent Option
3 treatment (partial recycle of process water followed by lime addition,
settling and filtration). However, EPA has established lower levels of
stringency for small gray and malleable iron foundries. For malleable
iron foundries with fewer than 100 employees, BAT, PSES, NSPS, and PSNS
are set equal to BPT. For gray iron foundries with fewer than 50
employees, PSES and PSNS have also been set equal to BPT.
2. Potential Plant Closure and Employment Loss Impacts
EPA has used the potential loss of employment and closure of
plants as the primary measure of economic impacts. Precompliance
financial statements were established using the model financial ratios
presented in Chapter II. Estimated compliance costs, in 1983 dollars,
were imposed on the model financial statements to estimate
postcompliance financial conditions. Where the model postcompliance
financial statements failed two of three tests, the number of firms
estimated to have those financial statements has been forecast to fail.
As shown in Tables 2 through 5, overall impacts under each of
the four options are expected to be low. Under Option 1, only four
foundries (two casting gray iron and two casting magnesium) are
projected to close. The associated Job loss of 100 employees represents
0.07 percent of the 149,287 employees of discharging foundries. Under
Option 4, a total of 24 foundries are Judged potential closures. The
associated job loss of 724 employees represents 0.50 percent of the
employment of discharging foundries.
In complying with the regulations, the estimated 800 discharging
foundries would incur capital costs ranging from $43.2 million under
Option 1 to $102.4 million under Option 4.^ Total annual costs,
including operating costs, interest, and depreciation, would range from
$16.0 million under Option 1 to $47 million under Option 4. Aggregate
Impacts at all four levels are:
As a conservative measure, Impacts are estimated assuming all foundries
incur costs to segregate noncontact cooling water. This was done
because EPA was not able to determine the specific foundries that would
incur these costs. Actually, only 30 percent are expected to incur the
incremental cost.
o
All costs in this chapter are in 1983 dollars.
-9-
-------
TABLE 2
COMPLIANCE COSTS AND ECONOMIC IMPACTS FOUNDRY INDUSTRY
(Option 1 Recycle/Simple Settle)
Number of
Discharging
Foundries
(in
Compliance Costs
thousands of 1983 dollars)
Closures
Capital
Investment
Annual Costs
Number of Foundries
Number of
Employees
Direct
Indirect
Direct
Indirect
Direct
Indirect
Direct
Indirect
Total
Total
Gray Iron
91
115
4,662
8,758
1,640
3,113
2
0
2
51
Ductile Iron
27
25
3,036
1,167
1,058
113
0
0
0
0
Malleable Iron
21
29
573
1,064
205
373
0
0
0
0
Steel
*3
64
1,706
2,350
614
861
0
0
0
0
Total Ferrous
182
263
9,978
13,339
3,517
1,759
2
0
2
51
Aluminum
15
131
2,524
3,627
919
1,100
0
0
0
0
Copper-base
63
51
7,916
1,355
3,369
1,650
0
0
0
0
Zinc
9
19
151
1,175
73
460
0
0
0
0
Magnesium
2
2
17
57
22
20
1
1
2
16
Total Nonferrous
119
236
10,669
9,211
1,113
3,530
1
1
2
46
Orand Total
301
199
20,647
22,553
7,930
8,290
3
1
4
100
Jobber
Gray Iron
65
10
3,274
6,145
1,151
2,167
1
0
1
27
Ductile Iron
23
21
2,197
1,087
861
386
0
0
0
0
Malleable Iron
15
23
131
878
153
308
0
0
0
0
Steel
35
53
1,123
1,871
513
685
0
0
0
0
Total Ferrous
138
207
7,624
9,982
2,684
3,517
1
0
1
27
Aluminum
37
107
2,213
2,876
824
1,114
0
0
0
0
Copper-base
12
38
1,003
3,091
2,002
1,177
0
0
0
0
Zlne
7
38
105
831
52
331
0
0
0
0
Magnesium
2
2
17
57
22
20
1
1
2
16
Total Nonferrous
88
185
7,168
6,861
2,900
2,615
1
1
2
16
Grand Total
227
392
14,792
16,843
5,584
6,192
2
1
3
73
Cap tire
Gray Iron
26
35
1,389
2,613
486
915
1
0
1
27
Ductile Iron
4
1
539
80
193
27
0
0
0
0
Malleable Iron
6
6
113
186
51
64
0
0
0
0
Steel
6
11
283
478
102
176
0
0
0
0
Total Ferrous
41
56
2,351
3,358
832
1,213
1
0
1
27
Aluminum
8
21
311
751
125
286
0
0
0
0
Copper-base
21
16
3,113
1,261
1,367
173
0
0
0
0
Zinc
2
11
47
311
21
126
0
0
0
0
Magnesium
0
0
0
0
0
0
0
0
0
0
Total Nonferrous
31
51
3,501
2,353
1,513
886
0
0
0
0
Grand Total
76
107
5,855
5,711
2,315
2,099
1
0
1
27
-------
TABLE 3
COMPLIANCE COSTS AND ECONOMIC IMPACTS FPUTORY INDUSTRY
(Option 2 Recycle/Lime Addition/Settle)
Number of
Discharging
Foundries
(in
Compliance Coats
thousands of 1983 dollars)
Closures
Capital
Investment
Annual Costs
Number of Foundries
Number of
Employees
Direct
Indirect
Direct
Indirect
Direct
Indirect
Direct
Indirect
Total
Total
Cray Iron
91
115
13,899
19,772
6,091
8,522
3
2
5
135
Ductile Iron
27
25
6,546
2,475
2,836
1,061
0
0
0
0
Malleable Iron
21
29
2,387
2,432
1,109
1,072
0
0
0
0
Steel
43
61
4,377
5,409
1,649
2,370
0
0
0
0
Total Ferrous
182
263
27,209
30,088
11,885
13,025
3
2
5
135
Aluminum
45
131
3,040
6,005
1,337
3,230
0
0
0
0
Copper-base
63
54
8,208
4,607
3,607
1,871
0
0
0
0
Zinc
9
19
197
1,700
123
880
0
0
0
0
Magnesium
2
2
59
65
26
23
1
1
2
46
Total Nonferrous
119
236
11,504
12,377
5,093
6,004
1
1
2
46
Grand Total
301
199
38,713
42,466
16,979
19,029
4
3
7
181
Jobber
Gray Iron
65
110
9,421
13,980
4,125
5,987
2
2
4
108
Ductile Iron
23
21
5,484
2,171
2,367
935
0
0
0
0
Malleable Iron
15
23
1,774
1,986
822
876
0
0
0
0
Steel
35
53
3,712
4,393
1,592
1,920
0
0
0
0
Total Ferroua
136
207
20,391
22,531
8,907
9,718
2
2
4
108
Aluminum
37
107
2,628
4,833
1,135
2,624
0
0
0
0
Copper-base
42
38
4,988
3,281
2,165
1,340
0
0
0
0
Zinc
7
38
138
1,243
89
665
0
0
0
0
Magnesium
2
2
59
65
26
23
1
1
2
46
Total Nonferrous
68
165
7,812
9,422
3,414
4,651
1
1
2
46
Grand Total
227
392
28,203
31,953
12,321
14,369
3
3
6
154
Captlvs
Gray Iron
26
35
4,478
5,792
1,966
2,535
1
0
1
27
Ductile Iron
1)
4
1,062
304
468
127
0
0
0
0
Malleable Iron
6
6
613
446
288
196
0
0
0
0
Steel
8
11
665
1,015
257
450
0
0
0
0
Total Ferrous
56
6r816
7,558
2,979
3,307
1
0
1
27
Aluminum
8
24
412
1, 172
202
606
0
0
0
0
Copper-base
21
16
3,220
1,326
1,443
531
0
0
0
0
Zinc
2
11
59
457
34
215
0
0
0
0
Hagnesiue
0
a
0
0
0
0
0
0
0
0
Total Nonferrous
31
51
3,692
2,956
1,679
1,353
0
0
0
0
Qrand Total
76
107
10,510
10,514
4,658
4,660
1
0 1
1 ,
27
-------
TABLE 1
COMPLIANCE COSTS AND ECONOMIC IMPACTS FOUNDRY INDUSTRY
(Option 3 Recycle/Lime Addition/Settle/Filtration)
Number of
Discharging
Foundries
(in
Compliance Costa
thousands of 1983 dollars)
Closures
Capital
Investment
Annual
Costs
Number of Foundries
Number of
Direct
Indirect
Direct
Indirect
Direct
Indirect
Direct
Indirect
Total
Employees
Total
Gray Iron
91
115
15,702
22,152
7,099
9,892
3
6
9
213
Ductile Iron
27
25
7,150
2,799
3,311
1,215
0
1
1
27
Malleable Iron
21
29
2,611
2,736
1,259
1,215
1
0
1
71
Steel
13
61
1,851
5,900
2,111
2,691
0
0
0
0
Total Ferrous
182
263
30,617
33,587
13,813
15.073
1
7
11
341
Aluminum
15
131
3,353
6,110
1,599
3,652
0
0
0
0
Copper-base
63
51
9,012
1,911
1,173
2,112
0
0
0
0
Zinc
9
19
212
1,828
163
1,012
0
0
0
0
Magnesium
2
2
63
68
30
26
1
1
2
46
Total Nonferrous
119
236
12,669
13,280
5,961
6,801
1
1
2
46
Grand Total
301
199
13,316
16,867
19,806
21,875
5
8
13
387
Jobber
Gray Iron
65
110
10,671
15,686
1,823
6,961
2
5
7
189
Ductile Iron
23
21
6,237
2,460
2,789
1, 100
0
1
1
27
Malleable Iron
15
23
1,962
2,228
932
1,015
1
0
1
71
Steel
35
53
1,096
1,798
1,829
2,185
0
0
0
0
Total Ferrous
138
207
22,969
25,172
10,372
11,265
3
6
9
287
Aluminum
37
107
2,893
5,183
1,357
2,965
0
0
0
0
Copper-base
12
38
5,193
3,526
2,515
1,511
0
0
0
0
Zinc
7
38
170
1,338
117
765
0
0
0
0
Magnesium
2
2
63
68
30
26
1
1
2
46
Total Nonferrous
88
185
8,619
10, 111
1,020
5,270
1
1
2
46
Grand Total
227
392
31,588
35,295
11,390
16,535
4
7
11
333
Captive
Gray Iron
26
35
5,028
6,466
2,277
2,927
1
1
2
54
Ductile Iron
4
1
1,213
339
552
115
0
0
0
0
Malleable Iron
6
6
680
508
327
230
0
0
0
0
Steel
8
11
758
1,102
315
506
0
0
0
0
Total Ferrous
44
56
7,678
6,115
3,171
3,809
1
1
2
51
Aluminum
8
21
160
1,257
212
687
0
0
0
0
Copper-base
21
16
3,519
1,118
1,658
598
0
0
0
0
Zinc
2
11
72
190
15
217
0
0
0
0
Magnesium
0
0
0
0
0
0
0
0
0
0
Total Nonrerrous
31
51
1,051
3,165
1,915 ,
1,532
0
0
0
0
Grand Total
76
107
U ,729
11,580
5,116
5,311
1
1
2
54
-------
TABLE 5
COMPLIANCE COSTS AMD ECONOMIC IMPACTS FOUNDRY INDUSTRY
(Option 1 Recycle/Lime Addition/Settle/Flltration/Carbon Adsorption)
Number of
Discharging
Foundries
(in
Compliance Costs
thousands of 1983 dollars)
Closures
Capital
Investment
Annual Costs
Number of Foundries
Naaber of
Direct
Indirect
Direct
Indirect
Direct
Indirect
Direct
] Indirect
Total
Employees
Total
Cray Iron
Ductile Iron
Malleable Iron
Steel
91
27
21
13
115
25
29
64
17,8l6
8,171
2,931
5,155
25,011
3,111
2,991
6,381
7,915
3,621
1,376
2,352
1 1,1(50
1,131
1,398
2,891
5
0
1
0
13
1
1
0
18
1
2
0
K86
27
1K2
0
Total Ferrous
182
263
31,373
37,526
15,295
17,172
6
15
21
655
Aluminum
Copper-base
Zinc
Magnesium
15
63
9
2
131
51
19
2
1,086
9,563
<416
01
8,052
5,700
2,1(68
66
1,929
1,H60
213
40
1,152
2,187
1,335
26
0
0
0
2
0
0
0
1
0
0
0
3
0
0
0
69
Total Nonferrous
Grand Total
119
301
239
199
11,166
18,538
16,289
53,815
6,673
21,968
8,299
25,171
2
8
1
16
3
21
69
721
Jobber
Gray Iron
Ductile Iron
Malleable Iron
Steel
65
23
15
35
110
21
23
53
12,129
6,816
2.176
1,572
17,766
2,737
2,135
5,209
5,121
3,028
1,018
1,993
8,135
1,272
1,133
2,355
K
0
1
0
11
1
1
0
15
1
2
0
405
27
142
0
Total Ferrous
130
207
25,723
28,1MB
11,1461
12,891
5
13
18
571
Aluminum
Copper-ba3e
Zinc
Magnesium
37
12
7
2
107
38
38
2
3,522
5,695
299
81
6,519
1,061
1,826
68
1,61(0
2,716
176
1(0
3,629
1,792
1,012
26
0
0
0
2
0
0
0
1
0
0
0
3
0
0
0
69
Total Nonferrous
Grand Total
88
227
185
392
9,797
35,520
12,197
10,615
1,573
16,036
6,159
19,353
2
7
1
114
3
21
69
643
Captive
Cray Iron
Ductile Iron
Malleable Iron
Steel
26
14
6
8
35
6
11
5,686
1,325
755
883
7,274
371
558
1,171
2,522
592
358
359
3,315
162
265
536
1
0
0
0
2
0
0
0
3
0
0
0
81
0
0
0
Total Ferrous
44
56
8,619
9,378
3,831
1,278
1
2
3
81
Aluninua
Copper-base
Zinc
Magnesium
8
21
2
0
21
16
11
0
563
3,688
117
0
1,531
1,616
613
0
289
1,711
67
0
822
695
323
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
Total Nonferrous
Grand Total
31
76
51
107
1,368
13,017
3,792
13,170
2, 100
5,931
1,81(0
6, 118
0
1
0
2
0
3
0
81
-------
Potential
Total
Total
Potential
Number of
Capital Cost
Annual Cost
Job
Option
Closures
($ Thousands)
($ Thousands)
Loss
1
4
43,200
16,220
100
2
7
81,179
36,008
181
3
13
90,183
41,681
387
4
24
102,353
47,439
724
The potential industry impacts are concentrated in four
segments: gray iron, ductile iron, malleable iron, and magnesium.
Projected closures of gray iron foundries ranged from two under Option 1
to 18 under Option 1. One ductile iron foundry is Judged a potential
closure under Options 3 and 1. One malleable iron foundry may close
under Option 3 while two may close under Option 4. The magnesium
subcategory has the highest potential impacts, with two potential
closures at Options 1 through 3 and three potential closures under
Option *4. These closures represent 50 and 75 percent, respectively, of
the magnesium foundries incurring costs.
3. Impacts of the Selected Options
Under the selected options (Table 1), six plants are projected
to close. Three of these are small directly discharging gray iron
foundries, two are small indirectly discharging gray iron foundries, and
one is a small indirectly discharging ductile iron foundry. These six
plants represent about 1 percent of the 796 foundries directly affected
by this regulation. Approximately 162 jobs will be lost as a result of
the six plant closures.
Total capital costs for discharging plants as a result of this
regulation are estimated to be $90.4 million. Total annualized costs
(including depreciation and interest) are estimated at $41.2 million
(1985 dollars). Of these total amounts, BPT regulations (which are
being promulgated for direct dischargers in all subcategories except
magnesium) account for $39.7 million in investment and $17.4 million in
annual costs. The BAT requirements that exceed BPT requirements (which
affect the gray iron, ductile iron, malleable iron, zinc, and copper
subcategories) amount to an additional $3.9 million in capital costs and
$2.3 million in annualized costs. BAT limitations for steel, zinc, and
aluminum foundries are based on BPT technology; therefore, no
incremental BAT compliance costs are Incurred in these subcategories.
Pretreatment standards are being promulgated for indirect
dischargers in all subcategories except magnesium. Capital costs to
comply with PSES are estimated at $46.7 million and annualized costs are
$21.5 million (1985 dollars).
New source standards (NSPS and PSNS) are based on the same
technology levels as BAT and PSES, respectively. There are no
-14-
-------
Incremental costs and therefore no barriers to entry attributable to the
new source standards.
The costs associated with the selected options are summarized
below (in 1985 dollars):
Limitations
BPT
BAT
PSES
4. Price Impacts
The closure analysis was predicated on the inability of
foundries incurring costs as a result of the regulation to pass along
the costs to customers. Specifically, discharging foundries compete
with both nondischarging domestic foundries (dry or zero dischargers)
and with foreign foundries, neither of which incur costs as a result of
these regulations. Nevertheless, an estimation is made of the potential
price impacts that would result if the entire cost of compliance could
be passed through by regulated foundries. The potential price impacts,
measured as annual costs as a percent of total foundry sales, were less
than 1 percent for 25 of the 32 metal/size categories. For complete
recovery of investment costs, the affected foundries would require an
average price pas3-through of less than 0.5 percent for most metal
segments at the selected options:
Metal
Cost/Sales (%)
Gray Iron
0.49
Ductile Iron
0.51
Malleable Iron
0.30
Steel
0.12
Aluminum
0.12
Copper-base
0.28
Zinc
0.13
5. Potential Production Logs Due to the Regulation
EPA expects that production losses caused by this regulation
will be minor. Under the selected options only six foundries (five gray
iron and one ductile iron) are expected to close. Those six closures
could lead to a loss of about 14,000 tons per year of production, or
about 0.2 percent of combined gray and ductile iron production (Table
6). Production losses of this size can be easily made up by the
remaining foundries in the industry.
Capital Costs
(millions)
$39.7
3.9
46.7
$90.4
Annualized Costs
(millions)
$17.4
2.3
21.5
$41.2
-15-
-------
TABLE 6
POTENTIAL PRODUCTION IMPACTS FOR SELECTED OPTIONS
Gray Iron
Ductile Iron
Foundries Closed
5
1
Annual Sales per Foundry
($ thousands)
947
1,053
Sales Lost ($ thousands)
4,735
1,053
Sales by Dischargers in
Size Category ($ thousands)
76,708
11,712
% of Category Sales Lost
6.17
7.11
Sales by Dischargers
($ millions)
1,182
1,231
% of Sales Lost
0.11
0.09
Tons Shipped per Foundry
2,102
2,038
Tons Lost
12,010
2,038
1982 Shipments of Metals
(thousand tons)
6,393
1,822
% of Metal Shipments Lost
0.19
0.11
-16-
-------
6. Potential Balance of Trade Impacts
This regulation Is expected to have no significant impact on the
U.S. balance of trade. This conclusion is based on three factors:
Imports have a minor share in the U.S. market.
Potential price increases of affected foundries are minor.
Most U.S. foundries incur no cost increase at all.
As shown in the EIA, foreign imports have a very small share of
the U.S. market. Although some specific casting types have had strong
competition from imports, foreign castings overall account for only 2.6
percent of the total castings market. International Trade Commission
figures also show that exports of U.S. castings have grown at the same
time that imports have grown. Based on the data, it appears that
factors such as transportation costs, service and responsiveness are
strong enough to outweigh the price advantage of some foreign castings.
The second factor precluding large balance of trade effects is
the small potential effect on prices. For almost all affected segments,
price increases are less than 0.5 percent of costs. For comparison, it
should be noted that the value of the dollar fell 11 percent between
February and August, 1985, leading to an equivalent increase in the cost
of imported castings. Relative to such fluctuations in the cost of
imports, the cost increase to affected foundries Is minimal.
The third factor reflects the small number of affected foundries
relative to the U.S. foundry population as a whole. Although 800
foundries discharge process waters and thus incur costs, more than 3,000
foundries do not. The competitiveness of the 3,000 foundries not
incurring costs will not be affected by this regulation.
To summarize, only a fraction of foundries incur cost increases,
which sire minor relative to recent changes in the value of the U.S.
dollar. As a result, EPA concludes that potential balance of trade
impacts are minor.
7. Community Effects
Because of the use of model plant analysis to determine Impacts,
no way exists to determine which specific foundries will close rather
than comply with the regulations. In the absence of precise community
location of the affected foundries, the analysis assumes that the
distribution of closures will be the same as for foundries in general.
Foundries are located in four regions composed of various states, as
defined in the Census of Manufactures.
The analysis of community effects has been confined to an
Illustrative distribution of the closures among the four regions.
Closed foundries are assumed to have the same distribution as all
foundries casting the metal. Half of the seven plant closures at the
selected options might occur in the North Central region, with the
remainder distributed evenly in each of the other regions.
-17-
-------
Because of the small number of closures spread over the four
regions and the low total employment loss, significant adverse Impacts
in any one community are not expected.
8. Small Business Impacts
For this analysis, EPA has generally taken a small foundry to be
one employing fewer than 50 employees. Based on their generally larger
size, EPA has used a size cut-off of 100 employees for malleable iron
foundries.
At the selected options, six foundries may close rather than
comply. Of these, five are small (10-49 employee) gray iron foundries,
while one la a small (10-49 employee) ductile iron foundry. These
closures represent 3 percent of the 250 directly and indirectly
discharging foundries. In setting standards, EPA has mitigated small
business impacts by exempting magnesium foundries from the regulation
(all of these are small plants) and by setting less stringent standards
for small gray and malleable iron foundries. These less stringent
requirements result in approximately seven fewer plant closures.
-18-
-------
CHAPTER I
INTRODUCTION
-------
I. INTRODUCTION
This study assesses the economic impacts likely to result from the
effluent guidelines, limitations, and standards applicable to the
foundry industry. These regulations are based on Best Practicable
Control Technology Currently Available (BPT), Best Available Technology
Economically Achievable (BAT), New Source Performance Standards (NSPS),
and Pretreatment Standards for New and Existing Sources (PSNS and PSES)
which are being issued under authority of Sections 301, 304, 306, and
307 of the Federal Water Pollution Control Act, as amended by the Clean
Water Act of 1977. The primary economic impact variables assessed in
this study include the costs of the effluent regulations and the
potential for these regulations to cause plant closures, price changes,
Job losses, changes in industry profitability, structure and
competition, shifts in the balance of foreign trade, new source Impacts,
and impacts on small businesses. However, this study Includes new cost
information developed since proposal. The basis for the costs estimates
is presented elsewhere in the Development Document for Effluent
Limitations Guidelines and Standards for the Metal Molding and Casting
(Foundries) Point Source Category.
This report is organized as follows:
Chapter II presents the structure of the industry. No
extrapolation is made to baseline conditions in Chapter II.
Instead, the basic, historical production and financial
Information used in the analysis is presented.
Chapter III presents the methodology. The emphasis in this
chapter has been on an overview of the analytical techniques
used. More detailed discussions of the justification for and
Implications of certain methodological parameters, and actual
calculations are given in the Appendices and in Chapter VIII,
Limitations of the Analysis.
Chapter IV summarizes the effluent control and guideline costs
used as the basis for the analysis. These costs reflect EPA's
most recent cost estimates, based on extensive studies of the
EPA's survey data base. Costs have been developed in 1983
dollars.
Chapter V presents the analysis of economic Impacts. This
chapter contains the estimated closures by metal type and
employment size, and ancillary analyses of community effects,
production impacts, potential price Impacts and balance of trade
Impacts.
Chapter VI presents the regulatory flexibility analysis. In
accordance with the Regulatory Flexibility Act, this study has
analyzed the Impacts on small business. For this study, a
delineation of size based on number of employees has been
1-1
-------
used. This chapter also describes the alternatives chosen by
EPA to mitigate impacts on small plants.
o Chapter VII presents estimates of the effects of these standards
on new sources (plants opening subsequent to promulgation of the
rules).
o Chapter VIII presents the limitations of the analysis.
Particular attention has been given to reconciling data from
different sources.
In addition to the body of the report, two appendices have been
provided. Appendix A contains a detailed discussion of the literature
on financial statement analysis, the basis for the financial tests
chosen for this study, and the basis for the threshold values chosen.
1-2
-------
CHAPTER II
STRUCTURE OF THE INDUSTRY
-------
II. STRUCTURE OF THE INDUSTRY
This chapter presents the historical basis for the economic analysis
presented in this report, including information on foundry technology
and markets, historical trends in foundry shipments, and the financial
performance of foundries. In most cases, the information is presented
in terms of how it is used in the analysis, which is explained in
Chapter III.
A. TECHNOLOGY
The unique feature of the foundry industry is the pouring or
injection of molten metal into a mold. The cavity of the mold
represents, within close tolerances, the final dimensions of the
product. One of the major advantages of this process is that intricate
metal shapes, which are not easily made by other methods of fabrication,
can be produced. A second advantage is the potential to rapidly develop
a finished product from a new design.
The Department of Commerce categorizes industries into Standard
Industrial Classifications, with major groupings at the 2-, 3-ป and
5-digit levels. For this analysis, EPA has categorized foundries by the
major metal cast. In most cases this corresponds to Standard Industrial
Classifications (SIC) codes at the 1-digit level:
gray iron (SIC 3321, except 33211 and 33212)
ductile iron (SIC 33211 and 33212)
malleable iron (SIC 3322)
steel (SIC 3324 and 3325)
aluminum (SIC 3361)
copper and copper-based alloys (SIC 3362)
zinc and zinc-based alloys (SIC 33691)
magnesium and magnesium-based alloys (SIC 33692)
These categorizations recognize that metals have sufficiently
different characteristics that most foundries choose to cast only one
metal. Data from the Census of Manufactures show that foundries
typically receive 80 to 95 percent of their total revenues from castings
of their primary metal, and that plants whose principal business is
casting account for 80 to 90 percent of all castings (See Table II-1).
B. TRENDS IN INDUSTRY SHIPMENTS
Individual metals have physical properties that make them
particularly well-suited to different purposes. Thus, while castings
shipments generally reflect overall trends in the economy, shipments of
individual metals show varying trends.
As seen in Table II-2, foundry shipments dropped sharply during the
recession of 1982. Relative to 1978, the year of EPA^ survey of the
industry, the quantity of shipments declined between 30 percent (for
II-1
-------
TABLE II-1
SPECIALIZATION AND COVERAGE RATIOS
FOR THE METAL MOLDING AND CASTING INDUSTRY
Metal
Ratio
1972
1977
1982
Gray and Ductile Iron
Specialization
(*)a
91
96
94
Coverage ($)b
87
88
91
Malleable Iron
Specialization
(*)
87
86
83
Coverage ($)
91
93
76
Steel0
Specialization
(*)
89
87
91
Coverage ($)
88
91
90
Aluminum
Specialization
(*)
84
87
87
Coverage (%)
89
92
92
Copper, Brass,
Specialization
(*)
84
85
88
and Bronze
Coverage ($)
81
74
85
Nonferrous Metals,
Specialization
(*)
83
85
85
NEC
Coverage (%)
79
77
77
SOURCE: 1982 Census of Manufactures, Preliminary Report Industry Series.
aThe specialization ratio Is the ratio of primary product shipments
(products In the primary 4-digit industry) to total product shipments
for establishments classified in the industry.
bThe coverage ratio is the ratio of primary products shipped by
establishments classified in the industry to total shipments of such
products by all manufacturing establishments, wherever classified.
cBased on steel, not elsewhere classified (SIC 3325).
II-2
-------
TABLE II-2
QUANTITY OF SHIPMENTS
Gray
Ductile
Malleable
Steel
Copper-
Iron
Iron
Iron
NEC
Aluminum
Base
Zinc
Magnesium
(millions
(millions
(millions
(millions
(billions
(billions
(billions
(billions
Year
of tons)
of tons)
of tons)
of tons)
of pounds)
of pounds)
of pounds)
of pounds)
1972
13.167
1.835
0.961
1.584
1.916
0.920
0.938
0.050
1973
14.801
2.246
1.031
1.894
2.113
0.966
1.080
0.054
1971
13.458
2.203
0.912
2.091
1.857
0.857
0.843
0.058
1975
10.547
1.823
0.729
1.974
1.455
0.700
0.712
0.038
1976
11.923
2.245
0.848
1.767
1.971
0.682
0.869
0.053
1977
12.371
2.736
0.829
1.677
2.153
0.702
0.789
0.058
1978
13.140
2.984
0.790
1.857
2.287
0.743
0.760
0.051
1979
12.512
2.890
0.715
2.039
2.303
0.793
0.665
0.028
1980
9.399
2.400
0.450
1.879
1.690
0.592
0.486
0.025
1981
9.610
2.191
0.422
1.743
1.820
0.581
0.471
0.023
1982
6.393
1.822
0.284
1.017
1.605
0.456
0.405
0.018
1983
7.180
2.067
0.291
0.729
1.822
0.552
0.516
0.024
1984
8.014
2.607
0.360
0.956
1.830
0.625
0.565
0.024a
SOURCE: U.S. Bureau of the Census, Current Industrial Reports.
aEstlmated. Bureau of the Census estimates approximately 2-1/2 to 5 percent increase in
magnesium castings from 1983 to 1984.
-------
aluminum) and 65 percent (for magnesium). As shown in Table II-3,
however, increases in castings prices have meant that the decline in
castings value has not been as sharp.
Gray iron castings are used in applications requiring high strength
without necessarily having maximum workability and resistance to
impact. The largest market for gray iron castings is in automotive
markets, but they are also used for piping, molds for steel ingots,
construction, and other uses. Shipments of gray iron castings reached
their peak in 1973, as trends to smaller and lighter cars have reduced
automotive consumption since then. Gray iron castings suffered fairly
sharp drops in shipments between 1978 and 1982, losing 51 percent of the
tonnage shipped. Between 1982 and 1984, gray iron castings shipments
increased 25 percent, to about 60 percent of 1978 values.
Ductile iron is a variant of gray iron with improved workability and
resistance to fracture. It is used in similar markets, particularly
pressure pipe and fittings and automotive applications. Ductile iron is
also used in many of the applications formerly served by malleable iron,
and has taken over some of those markets because of lower cost.
Shipments of ductile iron castings reached their peak in 1978. In 1982
ductile iron shipments slipped to 61 percent of 1978 shipments. By
1984, shipments of ductile iron castings had reached 87 percent of 1978
values.
Malleable iron castings are produced by annealing a brittle "white
iron" to transform the carbon content to agglomerations of graphite.
The resulting material has relatively high strength and workability,
properties that led to its wide use in automotive markets. The market
share of malleable iron has declined considerably because of
displacement by ductile iron, while automotive use of all iron castings
has declined because of increased automobile Imports and trends to
smaller, lighter cars. Since 1972, malleable iron shipments reached a
peak volume in 1973. Malleable iron shipments suffered severely in
1982, reaching only 37 percent of the 1978 value. By 1984, shipments
had risen 26 percent, to 46 percent of 1978 values.
Steel castings are preferred because of their high strength,
weldability, strength and resistance to impacts. On average, steel
castings have the highest price per ton of all ferrous castings. The
major markets for steel castings are for heavy capital goods and
railroads. Because of the difference in markets, steel castings
shipments increased between 1978 and 1980, while shipments of other
castings declined. Steel casting shipments reached their lowest point
in 1983, at only 39 percent of shipments in 1978. In 1984, steel
castings shipments recovered to Just over half the value of 1978.
Aluminum is easily cast and machinable, with good thermal and
electrical conductivity properties. Because of* aluminum's light weight,
aluminum castings are widely used in transportation markets, such as
motor vehicles and aerospace. Shipments of aluminum castings rose
fairly steadily through the 1970*s, reaching their peak in 1979.
Aluminum castings suffered the smallest decline of any metal,
II-4
-------
TABLE II-3
VALUE OF SHIPMENTS
(billions of dollars)
Gray
Ductile
Malleable
Copper-
Iron
Iron
Iron
Steel
Aluminum
Base
Zinc
Magnesium
1972
3.^28
0.605
0.485
1.288
1.172
0.463
0.402
0.041
1973
4.281
0.809
0.534
1.505
1.398
0.535
0.505
0.050
1974
5.085
0.981
0.627
2.066
1.745
0.620
0.417
0.056
1975
4.849
1.230
0.561
2.486
1.536
0.494
0.308
0.045
1976
5.498
1.454
0.640
2.480
1.963
0.566
0.405
0.053
1977
6.212
1.623
0.670
2.640
2.294
0.616
0.610
0.083
1978
6.971
1.873
0.680
3.047
2.614
0.690
0.606
a
1979
7.184
1.967
0.708
3.754
3.160
0.862
0.658
0.125
1980
6.142
1.697
0.494
4.047
3.135
0.879
0.556
0.139
1981
6.757
1.816
0.508
3.953
3.326
0.888
0.572
0.118
1982
6.288
1.655
0.371
2.973
2.811
0.715
0.563
0.092
SOURCE: U.S. Bureau of the Census, Annual Survey of Manufactures,
aData not reported by Census because of statistical problems.
II-5
-------
maintaining more than 70 percent of 1978 values. By 1984, aluminum
castings reached 80 percent of 1978 values.
Copper-base castings, including brass and bronze, are commonly used
in the water-handling and plumbing markets. As a result, copper
castings production generally fluctuates with the domestic housing
market. After reaching a peak in 1973* copper casting shipments first
slumped through the mid-1970*s, then rose until 1979. At the low point
of the 1981 to 1983 recession, shipments of copper-based castings were
61 percent of 1978 values. By 1984, shipments recovered 37 percent, to
84 percent of 1984 values.
Historically, most demand for zinc castings has been in the
automotive industry. Through the 1970's zinc lost some of that market
to aluminum castings and molded plastics because of price and weight
considerations. Zinc casting shipments reached a peak in 1973, dropped
through 1975, recovered in 1976, and dropped steadily through 1982.
Shipment values in 1982 were 53 percent of those in 1978. By 1984, zinc
shipments had recovered to 74 percent of 1978 values.
Magnesium is the smallest of the eight major cast metals both in
dollars and tonnage. Magnesium castings command a premium price because
of their production costs and very light weight. In 1980, only 120
foundries cast magnesium, a factor that would contribute to its scarcity
and price. Magnesium tonnage declined sharply from peaks in 1977.
Shipments in 1978 were only 35 percent of 1978 values. By 1984,
shipments rose to approximately 47 percent of 1978 values.
C. NUMBER OF FOUNDRIES AFFECTED BY THE REGULATION
Because of the large size of the foundry Industry, EPA did not
conduct an industry-wide survey to determine the specific foundries that
would be affected by this regulation. Instead, EPA has combined
publicly available data with the results of its own section 308
survey (including follow-up efforts) to estimate the number of affected
foundries.
In summary, EPA used the 1978 directory of the foundry industry
developed by Penton Publications as the starting point for a detailed
survey. EPA then sent out questionnaires to 1,147 plants, receiving a
total of 919 responses. Since 1978, the year of the survey, EPA has
received information about an additional 3^7 plants.
On reviewing the information from these 1,266 plants, EPA found that
many foundries were misclassified as to principal metal or size in the
Penton directory. Misclassifications of metal type were particularly
severe in the ferrous metal subcategories, where foundries reported that
their castings Included higher fractions of ductile and malleable iron,
and lower amounts of gray iron and steel. Therefore, EPA relied heavily
on its survey data to allocate the total foundry count into sub-
categories .
II-6
-------
TABLE II-4
FOUNDRY POPULATION OPERATING IH 1984
Fewer than
250
10
10 to 19
50 to 99
100 to 249
or more
Metal
Employees
Employees
Employees
Employees
Employees
Total
Gray Iron
81
360
167
169
101
878
Ductile Iron
7
66
22
43
25
163
Malleable Iron
5
23
13
54
22
147
Steel
36
108
91
67
36
338
Total Ferrous
129
557
323
333
184
1,526
Aluminum
476
575
142
114
44
1,351
Copper-base
232
324
77
27
6
666
Zinc
86
116
41
34
10
287
Magnesium
6
10
4
0
3
23
Total Nonferrous
800
1,025
264
175
63
2,327
Grand Total
929
1,582
587
508
247
3,853
SOURCE: Penton Publications and EPA surveys. Data from Penton Publications were
used for the count of foundries In each nonferrous metal and for the
aggregate cost of ferrous foundries. Proportions of ferrous foundries In
each specific metal Here determined from EPA survey data.
II-8
-------
TABLE II-5
PROJECTED NUMBER OF ACTIVE WET PLANTS IN INDUSTRY (1984)
Metal
Employee
Group
Discharge Modes
Direct
Indirect
Zero
Discharge
Total
Gray Iron
Fewer than 10
0
2
2
4
10 to 19
14
38
29
81
50 to 99
14
27
24
65
100 to 249
32
48
38
118
250 or more
31
-JO
14
75
Overall
91
145
107
343
Ductile Iron
Fewer than 10
0
0
0
0
10 to 49
0
9
5
14
50 to 99
0
3
5
8
100 to 249
16
11
3
30
250 or more
11
2
6
19
Overall
27
25
19
71
Malleable Iron
Fewer than 10
0
0
0
0
10 to 49
0
0
0
0
50 to 99
3
5
8
16
100 to 249
11
22
5
38
250 or more
7
2
6
15
Overall
21
29
19
69
Steel
Fewer than 10
2
0
0
2
10 to 49
0
10
9
19
50 to 99
11
21
5
37
100 to 249
19
19
10
48
250 or more
11
14
28
Overall
43
64
27
13*
Aluminum
Fewer than 10
7
0
6
13
10 to 49
9
61
33
103
50 to 99
6
20
7
33
100 to 249
14
41
7
62
250 or more
9
9
21
Overall
45
131
56
232
(continued)
II-9
-------
TABLE II-5 (Continued)
Metal
Employee
Group
Discharge Modes
Total
Direct
Indirect
Zero
Discharge
Copper
Fewer than 10
16
11
0
27
10 to 49
20
28
4
52
50 to 99
16
5
8
29
100 to 249
6
6
3
15
250 or more
5
4
1
10
Overall
63
54
16
133
Zinc
Fewer than 10
0
2
1
3
10 to 49
0
17
6
23
50 to 99
0
13
2
15
100 to 249
7
13
2
22
250 or more
2
4
I
7
Overall
9
49
12
70
Magnesium
Fewer than 10
0
0
1
1
10 to 49
2
2
0
4
50 to 99
0
0
2
2
100 to 249
0
0
0
0
250 or more
0
0
0
0
Overall
2
2
3
7
TOTALS OF
ALL METALS
301
499
259
1,059
SOURCE: EPA
11-10
-------
TABLE II-6
CALCULATION OF AVERAGE SALES PEH FOUNDRY
Metal/Size
Production8
Yield
<ปa
Tons Shipped
per Year
1983
Average
Price
(thousand
dollars per
short ton)
1983
Average
Sales per
Foundry
(thousand
dollars per
short ton)
Ratio of
Shipments
1982-78
(Steel
1983-78)
Revised
Sales
per
Foundry
Number
of
Dischargers0
Sales by
Dischargers
(thousand
dollars)
Short Tons
per Day
Short Tons
per Year
Grey Iron
Fewer than 10
20
5,200
0.66
3,132
0.810
2,781
0.186530
1,353
1
5,111
10 to U9
11
3,610
0.66
2,102
0.810
1,916
0.186530
917
81
76,708
50 to 99
38
9,880
0.66
6,521
0.810
5,283
0.186530
2,570
65
167,080
100 to 219
161
11,860
0.66
27,628
0.810
22,381
0.186530
10,891
118
1,285,095
250 or more
581
151,060
0.66
99,700
0.810
80,778
0.186530
39,301
75
2,917,575
Total
3^3
1,181,869
Duetlle Iron
Fewer than 10
20
5,200
0.56
2,912
0.816
2,161
0.610590
1,501
0
0
10 to 19
11
3,610
0.56
2,038
0.816
1,725
0.610590
1,053
11
11,712
50 to 99
38
9,880
0.56
5,533
0.816
1,681
0.610590
2,858
8
22,865
100 to 219
161
11,860
0.56
23,112
0.816
19,832
0.610590
12,109
30
363,278
250 or more
581
151,060
0.56
81,591
0.816
71,568
0.610590
13,699
19
830,275
Total
71
1,231,159
Malleable Iron
Fewer than 10
20
5,200
0.18
2,196
1.666
1,160
0.359191
1,195
0
0
10 to 19
11
3,610
0.18
1,717
1.666
2,912
0.359194
1,017
0
0
50 to 99
38
9,880
0.18
1,712
1.666
7,903
0.359191
2,811
16
15,158
100 to 219
161
11,860
0.18
20,093
1.666
33,181
0.359191
12,037
38
157,117
250 or more
581
151,060
0.18
72,509
1.666
120,831
0.359191
13,139
15
651,583
Total
69
1,151,158
Steel
Fewer than 10
20
5,200
0.59
3,068
2.719
8,312
0.392569
3,275
2
6,550
10 to 19
11
3,610
0.59
2,118
2.719
5,810
0.392569
2,292
19
13,557
50 to 99
38
9,880
0.59
5,829
2.719
15,850
0.392569
6,222
37
230,227
100 to 219
161
11,860
0.59
21,697
2.719
67,156
0.392569
26,363
18
1,265,132
250 or more
581
151,060
0.59
89,125
2.719
212,311
0.392569
95,137
28
2,663,825
Total
131
1,209,591
(Continued)
-------
TABLE II-6 (Continued)
CALCULATION OF AVERAGE SALES PER FOUNDRY
Metal/Size
Productionฎ
Yield
(*)a
Tons Shipped
per Year
1983
Average
Price
(thousand
dollars per
short ton)
1963
Average
Sales per
Foundry
(thousand
dollars per
short ton)
Ratio of
Shipments
1982-78
(Steel
1983-78)
Revised
Sales
per
Foundry
Number
or
Dischargers0
Sales by
Dischargers
(thousand
dollars)
Short Tons
per Day
Short Tons
per Year"
UuBinuo
Fewer than 10
12
3,120
0.71
2,215
4.074
9,024
0.701990
6,335
13
82,356
10 to 49
6
1,560
0.71
1,108
4.074
4,512
0.701990
3,168
103
326,257
50 to 99
32
8,320
0.71
5,907
4.074
24,065
0.701990
16,894
33
557,487
100 to 249
65
16,900
0.71
11,999
4.074
48,882
0.701990
31,315
62
2,127,532
250 or more
131
34,840
0.71
24,736
4.074
100,773
0.701990
70,742
21
1,485,577
Total
23 2
4,579,208
Copper-base
Fewer than 10
e
2,080
0.69
1,435
1.786
2,563
0.612874
1,571
27
42,416
10 to 49
110
28,600
0.69
19,734
1.786
35,245
0.612874
21,601
52
1,123,250
50 to 99
72
18,720
0.69
12,917
1.786
23,070
0.612B74
11,139
29
410,025
100 to 249
258
67,080
0.69
46,285
1.786
82,666
0.612874
50,664
15
759,961
250 or more
153
39 , 780
0.69
27,448
1.786
49,023
0.612874
30,045
10
300,450
Total
133
2.636,102
Zioo
Fewer than 10
0.9
234
0.83
194
3-570
693
0.533279
370
5
1,849
10 to 49
7
1,820
0.83
1,511
3-570
5,393
0.533279
2,876
23
66,145
50 to 99
15
3,900
0.83
3,237
3-570
11,556
0.533279
6,163
15
92,439
100 to 249
62
21,320
0.83
17,696
3-570
63,173
0.533279
33,689
22
741,158
250 or more
25
6,500
0.83
5,395
3-570
19,260
0.533279
10,271
7
71,897
Total
72
973,489
Magnesium
Fewer than 10
0.2
52
0.62
32
6.342
204
0.357499
73
1
73
10 to 49
0.8
208
0.62
129
6-342
818
0.357499
292
M
1,170
50 to 99
10
2,600
0.62
1,612
6.312
10,224
0.357499
3,655
2
7,310
100 to 249
0
0
0.62
0
6.342
0
0.357499
0
0
0
250 or more
0
0
0.62
0
6.342
0
0.357499
0
0
0
Total
7
8,553
Grand Total
19,274,428
aBased or data collected by EPA between 1978-1984.
^Baaed on 260 operating days per year.
cIncludes direct, indirect and zero dischargers.
-------
TABLE II-7
FINANCIAL RATIOS FOR GRAY IRON FOUNDRIES6
(Includes Ductile Iron)
Upper
Quartlle
Median
Lower
Quartlle
Employee Size: Fewer than 10
Return on Sales (%)
13.58
8.53
4.70
Sales to Net Worth (times)
1.61
2.60
5.15
Debt to Net Worth ($)
9.59
51.73
188.78
Net Fixed Assets to Net Worth
(*)
25.93
37.51
89.19
Employee Size: 10 to 49
Return on Sales (%)
7.18
4.32
2.61
Sales to Net Worth (times)
1.96
3.01
4.39
Debt to Net Worth (%)
27.63
67.92
136.42
Net Fixed Assets to Net Worth
($)
6.21
36.50
71.62
Employee Size: 50 to 99
Return on Sales (%)
6.41
4.10
2.64
Sales to Net Worth (times)
2.27
3.32
4.99
Debt to Net Worth (%)
32.04
82.47
184.13
Net Fixed Assets to Net Worth
(%)
23.15
56.62
93.59
Employee Size: 100 to 249
Return on Sales (?)
5.23
3.82
1.42
Sales to Net Worth (times)
2.30
3.26
4.37
Debt to Net Worth ($)
41.11
68.98
122.24
Net Fixed Assets to Net Worth
(%)
42.31
59.27
74.34
Employee Size: 250 or more
Return on Sales ($)
5.57
3.86
2.40
Sales to Net Worth (times)
2.17
2.66
3.13
Debt to Net Worth (%)
64.92
82.47
121.84
Net Fixed Assets to Net Worth
(%)
52.25
86.30
109.52
All
Depreciation to Gross Fixed Assets
6.9941
6.9941
6.9941
Gross Fixed Assets to Net Fixed Assets
(times)
2
2
2
development of the quartiles and quartlle financial ratios is explained in
Chapter III.
SOURCES:
1. For ratios that vary by size: Review of FINSTAT.
2. Depreciation to Gross Fixed Assets: Annual Survey of Manufactures,
1978.
3. Gross Fixed Assets to Net Fixed Assets: EPA estimates.
11-13
-------
TABLE II-8
FINANCIAL RATIOS FOR MALLEABLE IRON FOUNDRIES6
Upper
Quartile
Median
Lower
Quartile
Employee Size: Fewer than 10
Return on Sales ($)
Sales to Net Worth (times)
No discharging foundries.
Debt to Net Worth ($)
Net Fixed Assets to Net Worth
($)
Employee Size: 10 to 49
Return on Sales ($)
8.87
6.62
.64
Sales to Net Worth (times)
1.26
1.94
4.08
Debt to Net Worth (%)
25.11
44.69
124.99
Net Fixed Assets to Net Worth
($)
0.0
7.99
49.49
Employee Size: 50 to 99^
Return on Sales ($)
7.38
3.10
2.31
Sales to Net Worth (times)
1.04
2.66
17.74
Debt to Net Worth (%)
68.98
108.54
1,185.96
Net Fixed Assets to Net Worth
(%)
26.81
68.50
504.26
Employee Size: 100 to 249^
Return on Sales ($)
7.38
3.10
2.31
Sales to Net Worth (times)
1.04
2.66
17.74
Debt to Net Worth (%)
68.98
108.54
1,185.96
Net Fixed Assets to Net Worth
(*)
26.81
68.50
504.26
Employee Size: 250 or more
Return on Sales ($)
7.01
4.42
1.04
Sales to Net Worth (times)
2.18
2.67
4.04
Debt to Net Worth (%)
32.60
55.69
103.18
Net Fixed Assets to Net Worth
(%)
38.95
57.26
70.01
All
Depreciation to Gross Fixed Assets
6.3805
6.3805
6.3805
Gross Fixed Assets to Net Fixed Assets
(times)
2
2
2
development of the quartlles and quartile financial ratios is explained in
Chapter III.
^Financial data for the 50 to 99 and 100 to 249 employment size groups were
merged because of insufficient sample size.
SOURCES:
1. For ratios that vary by size: Review of FINSTAT.
2. Depreciation to Gross Fixed Assets: Annual Survey of Manufactures,
1978.
3. Gross Fixed Assets to Net Fixed Assets: EPA estimates.
11-14
-------
TABLE II-9
FINANCIAL RATIOS FOR STEEL FOUNDRIESฎ
Upper
Quartile
Median
Lower
Quartile
Employee Size: Fewer than 10
Return on Sales (%)
11.00
7.07
1.67
Sales to Net Worth (times)
1.01
2.61
3.11
Debt to Net Worth (J)
11.53
36.17
121.20
Net Fixed Assets to Net Worth
($)
1.72
38.13
121.76
Employee Size: 10 to 19
Return on Sales ($)
11.55
8.00
6.15
Sales to Net Worth (times)
2.10
3.11
1.21
Debt to Net Worth ($)
55.60
113.71
201.99
Net Fixed Assets to Net Worth
(St)
0.0
7.39
66.01
Employee Size: 50 to 99
Return on Sales (%)
3.83
2.55
1.90
Sales to Net Worth (times)
2.17
3.55
5.19
Debt to Net Worth ($)
33.21
76.65
139.27
Net Fixed Assets to Net Worth
(%)
26.82
29.56
63.98
Employee Size: 100 to 219
Return on Sales (%)
8.71
1.67
1.01
Sales to Net Worth (times)
2.52
2.97
1.65
Debt to Net Worth (%)
72.52
115.52
186.83
Net Fixed Assets to Net Worth
(%)
0.0
13.98
97.75
Employee Size: 250 or more
Return on Sales ($)
6.98
1.61
2.68
Sales to Net Worth (times)
2.68
3.33
1.36
Debt to Net Worth ($)
77.66
111.32
213.17
Net Fixed Assets to Net Worth
(%)
59.36
73.19
101.07
All
Depreciation to Gross Fixed Assets
7.1135
7.1135
7.1135
Gross Fixed Assets to Net Fixed Assets
(times)
2
2
2
development of the quartlles and quartile financial ratios Is explained In
Chapter III.
SOURCES:
1. For ratios that vary by size: Review of FINSTAT.
2. Depreciation to Gross Fixed Assets: Annual Survey of Manufactures,
1978.
3. Gross Fixed Assets to Net Fixed Assets: EPA estimates.
11-15
-------
TABLE 11-11
FINANCIAL RATIOS FOR COPPER FOUNDRIESฎ
Upper
Quartile
Median
Lower
Quartile
Employee Size: Fewer than 10
Return on Sales (%)
23.73
7.08
2.81
Sales to Net Worth (times)
1.62
3.14
5.54
Debt to Net Worth ($)
10.34
34.26
84.45
Net Fixed Assets to Net Worth
(*)
5.68
34.01
73.39
Employee Size: 10 to 19
Return on Sales ($)
8.24
5.44
3.51
Sales to Net Worth (times)
2.19
3.59
5.51
Debt to Net Worth ($)
21.91
67.60
143.31
Net Fixed Assets to Net Worth
(*)
12.84
32.20
62.15
Employee Size: 50 to 99
Return on Sales (%)
11.58
8.24
3.61
Sales to Net Worth (times)
2.07
2.33
4.09
Debt to Net Worth (%)
33.38
71.51
148.05
Net Fixed Assets to Net Worth
(.%)
18.63
24.96
56.97
Employee Size: 100 to 249
Return on Sales ($)
3.36
2.75
2.08
Sales to Net Worth (times)
3.76
3.84
3.98
Debt to Net Worth (%)
52.12
72.66
158.25
Net Fixed Assets to Net Worth
(%)
33.54
37.51
40.79
Employee Size: 250 or more
Return on Sales ($)
6.11
5.24
4.36
Sales to Net Worth (times)
2.41
3.03
3.48
Debt to Net Worth ($)
118.97
133.43
156.40
Net Fixed Assets to Net Worth
(%)
52.25
73.14
93.40
All
Depreciation to Gross Fixed Assets
6.2122
6.2122
6.2122
Gross Fixed Assets to Net Fixed Assets
(times)
2
2
2
development of the quartiles and quartile financial ratios is explained in
Chapter III.
SOURCES:
1. For ratios that vary by size: Review of FINSTAT.
2. Depreciation to Gross Fixed Assets: Annual Survey of Manufactures,
1978.
3. Gross Fixed Assets to Net Fixed Assets: EPA estimates.
11-17
-------
TABLE 11-12
FINANCIAL RATIOS FOR NONFERROUS. NEC FOUNDRIES8
(includes Magnesium and Zinc)
Upper
Quartile
Median
Lower
Quartile
Employee Size: Fewer than 10
Return on Sales ($)
17.51
11.86
4.43
Sales to Net Worth (times)
1.59
2.52
6.14
Debt to Net Worth (%)
9.09
40.13
104.28
Net Fixed Assets to Net Worth
(%)
3.54
30.9M
60.24
Employee Size: 10 to 49
Return on Sales ($)
7.87
4.72
2.58
Sales to Net Worth (times)
2.50
3.30
5.43
Debt to Net Worth ($)
47.98
83.26
141.92
Net Fixed Assets to Net Worth
(t)
16.46
35.84
59.52
Employee Size: 50 to 99
Return on Sales ($)
4.84
3.59
2.45
Sales to Net Worth (times)
3.17
4.36
6.05
Debt to Net Worth (j)
60.74
154.32
192.55
Net Fixed Assets to Net Worth
(?)
28.60
47.42
70.62
Employee Size: 100 to 249
Return on Sales (%)
5.23
2.80
2.29
Sales to Net Worth (times)
3.13
4.03
4.83
Debt to Net Worth ($)
57.62
122.08
146.83
Net Fixed Assets to Net Worth
(*)
22.20
51.12
76.09
Employee Size: 250 or more
Return on Sales ($)
5.37
4.59
3.25
Sales to Net Worth (times)
2.80
3.46
4.26
Debt to Net Worth (?)
88.51
137.94
207.18
Net Fixed Assets to Net Worth
(*)
75.12
98.92
107.71
All
Depreciation to Gross Fixed Assets
6.8369
6.8369
6.8369
Gross Fixed Assets to Net Fixed Assets
(times)
2
2
2
development of the quartiles and quartile financial ratios is explained in
Chapter III.
SOURCES:
1. For ratios that vary by size: Review of FINSTAT.
2. Depreciation to Gross Fixed Assets: Annual Survey of Manufactures,
1978.
3. Gross Fixed Assets to Net Fixed Assets: EPA estimates.
11-18
-------
TABLE II-13
SEPARATION OF FERROUS EMPLOYMENT-SIZE SEGMENTS
BETWEEN JOBBER AND CAPTIVE FOUNDRIES
Employment Size Segment
Proportions
Jobber
Captive
Gray Iron
Fewer than 10
78$
22$
10 to i49
83
17
50 to 99
78
22
100 to 249
78
22
250 or more
61
_22
Total
78
22
Ductile Iron
Fewer than 10
50
50
10 to 49
63
37
50 to 99
89
11
100 to 249
89
11
250 or more
80
20
Total
77
23
Malleable Iron
Fewer than 10
67
33
10 to 49
100
0
50 to 99
77
23
100 to 249
77
23
250 or more
77
_2!
Total
77
23
Steel
Fewer than 10
63
37
10 to 49
86
14
50 to 99
84
16
100 to 249
84
16
250 or more
80
20
Total
82$
18$
11-19
-------
TABLE II-14
SEPARATION OF NONFERROUS EMPLOYMENT-SIZE
SEGMENTS BETWEEN JOBBER AND CAPTIVE FOUNDRIES
Employment Size Segment
Proportions
Jobber
Captive
Aluminum
Fewer than 10
80|
20$
10 to 49
83
17
50 to 99
84
16
100 to 249
84
16
250 or more
74
26
Total
82
18
Copper Base
Fewer than 10
80
20
10 to 49
79
21
50 to 99
56
44
100 to 249
56
44
250 or more
29
71
Total
76
24
Zinc
Fewer than 10
52
48
10 to 49
77
23
50 to 99
83
17
100 to 249
83
17
250 or more
60
40
Total
70
30
Magnesium
Fewer than 10
0
100
10 to 49
71
29
50 to 99
100
0
100 to 249
100
0
250 or more
Total
78*
22%
SOURCE: EPA Surveys.
11-20
-------
regarded as captive foundries. The Jobber/captive separation was based
on data from the 1978 308 survey and Its 1981 telephone update. To
determine the number of Jobber and captive foundries for each employment
segment of each metal, the proportion of jobber and captive foundries
was applied to the numbers of foundries in each segment. Table 11-13
shows the proportion of Jobber and captive foundries for the ferrous
metals, while Table 11-14 presents comparable information for the
nonferrous metals.
In computing the economic impacts, captive foundries were treated as
though the financial decisions were made on a plant-by-plant basis and
subject to the same financial tests as jobbers. The Agency has
concluded, in accordance with current financial theory and in line with
comments received by many parties on previous foundry analyses, that
parent corporations treat their operations in different industries as
separate companies, subject to the normal financial structure and
restrictions of those industries. The Agency recognizes that this may
undervalue any benefits of conglomeration, such as centralized
accounting and scheduling, lower corporate borrowing costs, etc.
F. ANALYSIS OF IMPORTS AND EXPORTS
Domestic industries operate increasingly in a competitive world
market. Foreign competition is important to this analysis in two major
ways. If foreign Imports are a significant fraction of domestic
consumption, then the ability of domestic foundries to pass along any
cost increases may be greatly constrained. Also, if domestic producers
perceive significant pressure from importers of castings, then there may
be impacts on profits as domestic foundries seek to keep prices low.
Recently, the U.S. International Trade Commission (ITC) reviewed foreign
trade in the castings market. The investigation, made at the request of
the U.S. Trade Representative, was intended to determine whether
imported castings were restraining the performance of the foundry
industry.
Overall, imports and exports represent a very small fraction of the
domestic casting market. According to ITC data, exports rose from 2.4
percent of domestic shipments in 1979 to 4.2 percent of domestic
shipments in 1982, declining to 3ซ9 percent of shipments in 1983*
Imported castings rose from 1.0 percent of domestic shipments in 1979 to
2.6 percent in 1983. These numbers show Imports and exports to be a
small fraction of the total domestic market, but also show that Imports
have been making progressive inroads. They also show that net exports
as a percentage of U.S. markets are still quite close to 1979 levels.
11-21
-------
Total
Shipments
Exports
Imports
Percent
Percent
Year
(million $)
(million $)
(million $)
Exports
Imports
1979
21551
508
, 210
2.4
1.0
1980
20560
749
253
3.6
1.2
1981
22197
805
358
3.6
1.6
1982
16349
684
387
4.2
2.3
1983
15873
614
424
3.9
2.6
Source: U.S. International Trade Commission, 1984, p. 39
The foundry industry experiences competition not only directly from
raw castings, but indirectly from imported end products that contain
castings. To evaluate the potential impact, EPA examined the castings
content and the1 import/export patterns of 39 end markets. The end
markets represented 47 percent of total castings demand in 1977, and an
estimated 54 percent in 1982. Table 11-15 shows the end markets used
for the analysis, while Table 1116 shows the results. Assuming that
average castings content in each of the end markets remained constant
between 1977 and 1982, it is apparent that net exports (value of exports
minus value of imports) grew In both nominal and constant dollars
between 1972 and 1982. Also, net exports grew in nominal dollars
between 1977 and 1982, although they remained nearly constant in
constant 1972 dollars.
Although overall levels of imports are low, the ITC found that
import penetration has been severe in some individual product lines.
Imports as a fraction of consumption were found to range from 10 percent
to 37 percent for some categories of castings. Import penetration has
been and is expected to continue to be most significant in the area of
standardized, simple-to-manufacture, price-sensitive castings, such as
iron construction castings, fittings, and valves, where foreign
competitors can take advantage of the large U.S. market, lower labor
cost, and other price-related advantages. The International Trade
Commission found that average prices on imported products range from 15
percent to 28 percent lower than comparable prices on domestically
produced products. Respondents to the ITC investigation claim various
responses to the foreign competition: lowered prices, suppressed price
increases, and cost reduction efforts. Some producers reported reduced
production and curtailed expansion plans.
In the long run, the ability of U.S. foundries to compete with
foreign sources hinges on three factors: (1) the maintenance of
existing servicing and other marketing advantages, (2) the restriction
of price increases through productivity gains, and (3) the value of the
U.S. dollar relative to other currencies. The first two of these
factors are largely within the control of U.S. producers; the third is
not. Of these three factors, the value of the U.S. dollar is probably
preeminent. Between 1980 and February 1985 the value of the dollar rose
11-22
-------
SIC
3441
3444
3448
3465
3494
3511
3523
3524
3531
3532
3533
3534
3535
3536
3537
3541
3542
3544
3546
3551
3552
TABLE 11-15
LIST OF END MARKETS USED FOR EXPORT ANALYSIS
Industry
Fabricated Structural Metal
Sheet Metal Work
Prefabricated Metal Buildings
Automotive Stampings
Valves and Pipe Fittings
Turbines and Generator Sets
Farm Machinery and Equipment
Lawn and Garden Equipment
Construction Machinery
Mining Machinery
Oil Field Machinery
Elevators and Moving Stairs
Conveyors and Conveying
Equipment
Hoists, Overhead Cranes,
Monorails
Industrial Trucks and Tractors
Machine Tools-Metal Cutting
Machine Tools-Metal Forming
Special Dies, Tools, Etc.
Power Hand Tools
Food Products Machinery
Textile Machinery
SIC Industry
3554 Paper Industries Machinery
3555 Printing Trades Machinery
3561 Pumps and Pumping Equipment
3562 Ball and Roller Bearings
3563 Air and Gas Compressors
3564 Blowers and Fans
3567 Industrial Furnaces and Ovens
3573 Electronic Computing Equipment
3579)
3572( 0ffice Machines & Typewriters
3585 Air Conditioning
3711 Motor Vehicles and Car Bodies
3714 Motor Vehicle Parts and
Accessories
3715 Truck Trailers
3721 Aircraft
3724 Aircraft Engines and Parts
3728 Aircraft Equipment, NEC
3769 Space Vehicles and Equipment
3811 Engineering and Scientific
Instruments
3825 Electricity-Measuring
Instruments
11-23
-------
TABLE 11-16
RESULTS OF IMPORT/EXPORT ANALYSIS
(values in million dollars)
1972
1977
1982
Iron
Value of materials consumed in end markets
2,680.9
4,488.3
4,750.6
Value as included in net exports, nominal $
75.1
228.5
300.7
Value as included in net exports, 1972 $
75.1
129.1
116.1
Steel
Value of materials consumed in end markets
591.0
1,338.9
1,771.6
Value as included in net exports, nominal $
65.3
202.0
307.3
Value as included in net exports, 1972 $
65.3
114.1
118.6
Aluminum
Value of materials consumed in end markets
629.3
1,153.6
1,355.9
Value as included in net exports, nominal $
21.4
65.1
91.2
Value as Included in net exports, 1972 $
21.4
37.2
37.2
Copper
Value of materials consumed in end markets
88.1
145.1
193.9
Value as included in net exports, nominal $
5.5
12.7
12.5
Value as included in net exports, 1972 $
5.5
7.3
5.0
SOURCE: Bureau of Census, Census of Manufactures; U.S. Industrial Outlook.
Note: Net exports = (value of exports) - (value of imports).
11-24
-------
53 percent against foreign currencies. Between February 1985 and August
1985, the dollar fell about 11 percent. Swings of this magnitude
outweigh probable gains in productivity. Nevertheless, continued gains
in productivity and maintenance of marketing-related advantages will be
essential for U.S. foundries to retain their current market share.
In summary, imports have had a real effect in some specific casting
markets, while the threat of imports may be restricting price increases
for a wide range of castings. As yet, however, the overall importation
of castings is very small, less than three percent of the total U.S.
castings shipments. Overall, the principal source of competition to the
800 discharging foundries that are projected to incur costs is still the
3,053 dry or non-discharging, wet foundries that will not incur costs
because of this rulemaking.
11-25
-------
CHAPTER III
METHODOLOGY
-------
III. METHODOLOGY
A. OVERVIEW
This chapter shows how EPA dealt with three crucial issues in the
development of the economic impact analysis:
how to estimate the number of affected plants
how to estimate compliance costs
how to estimate impacts.
Although these issues are vital in the development of any effluent
guideline, they took on a special prominance for this industry. The
foundry industry is broad and diverse. In 1981, there were almost 4,000
foundries employing more than 300,000 people. Many of the plants are
independently owned (Jobbers), for which there are few reliable data
sources. These data have been supplemented by publicly-available
composites such as U.S. Bureau of the Census reports and with privately-
generated databases such as Dun & Bradstreet financial profiles. By
combining these data sources, EPA has obtained a comprehensive financial
database for the industry using the best available data sources. EPA
has further categorized the industry by conducting surveys of the
industry, with data collected between 1978 and 1981.
Nine foundry processes have the potential to generate process
waters, with no foundry in the survey database having more than seven
in-plant processes. Because of the large number of plants, and the
small number of individual discharging processes, EPA has determined
economic impacts by developing "model" plants. Each model plant has
average sales and compliance costs, but one of several sets of financial
ratios.
The remainder of this chapter is arranged in the following steps:
estimation of the number of affected plants,
estimation of compliance costs,
development of model plants,
estimation of Impacts.
This chapter does not provide detailed results, but rather shows the
methodology used.
B. ESTIMATION OF THE NUMBER OF AFFECTED PLANTS
A critical element of this analysis is estimating the number of
plants subject to the regulation. That estimate is used to determine
the aggregate national cost of the regulation, the number of plants
potentially suffering economic distress, and the significance of the
impacts in terms of the overall industry.
III-1
-------
1. Baseline Year for Compliance
The first step is to establish the baseline year for
compliance. Over the past 15 years, the foundry industry has been
marked by a gradual decline in the foundry population. The deep
recession of 1982 induced additional closures, while economic growth
3ince then has led to some plant openings.
Although indirect dischargers need not comply until 1988, EPA
has assumed that all foundries, both direct and indirect, will have
complied with the guidelines by 1986. Use of 1986 as the baseline year
assumes that foundries may require up to a year to arrange financing for
design, construction, and installation of pollution control equipment.
EPA has not forecast conditions to 1988 because of the difficulty in
projecting beyond a year or two. If conditions continue to improve as
they have since 1982, impacts forecast for Indirect dischargers may be
mitigated.
2. Use of Publicly Available Censuses
Although there are several sources of data on the number of
foundries, the two prominent ones are the U.S. Bureau of the Census
Census of Manufactures and Penton Publications Metal Casting Industry
Directory. The Census of Manufactures is more complete than the Penton
directory, in that it includes data on sales, production, and
employment. However, the Census of Manufactures has two deficiencies:
The most recent complete Census of Manufactures available for
this analysis represents data from 1977;
It excludes data for many foundries that are part of larger
operations in other SIC codes.
The most recent Penton directory represents data from 1983. EPA
was able to update the number of plant openings and closings through
1984 through telephone conversations with Penton Publications. EPA
believes these data represent the most complete and current count of the
foundry industry.
EPA is using the 1981 count of foundries obtained from Penton as
the population potentially affected by the regulation. EPA believes
that the 1986 foundry population will approximate the 1984 population.
If plants in EPA's database have closed since 1984 because of
general economic conditions, the impacts caused by the regulation may be
overstated. If new plants have opened since 1984, EPA feels it likely
that they will either be in a stronger position to absorb regulatory
costs, or will have incorporated effluent controls into the plant design
in anticipation of the regulation. EPA believes It extremely unlikely
that a foundry opening after 1984 would close because of costs
associated with this regulation.
III-2
-------
3. Incorporation of EPA Survey Data
Neither the Bureau of the Census nor Penton Publications
distinquish between foundries that generate process waters and those
that do not. As documented in the technical record, EPA conducted a
survey of the foundry industry in 1978 . The data collected were
specifically oriented, among other things, towards obtaining an estimate
of the number of discharging foundries, the types of discharging
processes, and the type and amount of pollutant discharged'. In response
to comments, EPA has expanded and verified the survey.
Results of the survey and additional data gathering have been
used in the estimation of the number of affected plants in two ways.
First, the 308-based distribution of plants casting ferrous metals
(which did not agree with data from Penton) has been used to estimate
the number of plants in each ferrous subcategory. The disagreement was
particularly large in the malleable iron and ductile iron
subcategories. In 1978, Penton estimated that there were 56 malleable
iron foundries. However, between 1978 and 1984, EPA obtained data from
63 foundries principally casting malleable iron. Based on the 308-based
ratio of malleable iron plants to the total count of ferrous plants, EPA
estimates that a total of 147 foundries currently cast malleable iron.
Penton also reported a plant count of 83 ductile iron foundries in
1978. However, EPA's survey results showed that many foundries not
listed as ductile iron foundries by Penton cast ductile iron as their
principal metal. Based on EPA's data, ductile iron foundries represent
11 percent of all ferrous foundries, leading to an overall estimate of
163 ductile iron foundries.
Second, EPA used its survey data to estimate the number of
foundries generating and discharging process waste waters. EPA noted
substantial differences between metals in the casting processes used,
the extent of wastewater generation, and the fraction of plants
discharging process waste waters. Details of this analysis may be found
in the technical record supporting this regulation.
4. Comparison to Analyses Previously Developed For This Industry
In previous analyses supporting this regulation, EPA projected
the foundry population at promulgation from a basis in the 1978 or 1981
Penton directories. At proposal, (EPA, Economic Analysis of Proposed
Effluent Limitations and Standards for the Foundry Industry, 1982) EPA
projected the industry population by using historical rates for the
creation and closing of foundries. In subsequent analyses (EPA,
Economic Analysis of Proposed Effluent Limitations and Standards for the
Metal Molding and Casting (Foundry) Industry, Supplemental Analysis.
The survey was conducted under the authority of Section 308 of the
Clean Water Act, and is thus referred to as the "308 survey."' Its
product was a group of more than 1,200 Data Collection Portfolios
(dcp's).
III-3
-------
downward to reflect decreased industry shipments between 1978 and
1982. Second, EPA has developed costs for the various combinations of
discharging processes that occur commonly in the foundry industry.
Third, EPA has combined the one-time capital costs and the operating
costs into a single equivalent annual cost. Fourth, the capital costs
have been increased to account for those foundries that commingle
process water and noncontact cooling water. A detailed study of 20
plants showed that 6 (or 30 percent) would benefit from installing
separate piping to carry noncontact cooling water. These 30 percent
incur increases in capital costs ranging from 1H.8 percent at Option 1
to 7.4 percent at Option 1. Estimates of potential closures due to
compliance are based on an assumption that all foundries incur costs to
segregate noncontact cooling water. This is a conservative assumption
that reflects EPA's inability to specify those foundries in the affected
population that will have to Install a segregation system. However, the
compliance cost estimates presented in this report accurately reflect
the proportion actually incurring the additional cost (30 percent).
1. Adjustment of Costs for Revised Production Estimates
Since EPA's survey in 1978, the foundry industry suffered a
severe recession in 1982. EPA recognizes that the decline has reduced
average foundry shipments, and has adjusted its per plaint revenue
estimates accordingly (see discussion below). The adjustments are based
on estimated production declines in the industry from 1978 through the
recession in 1982. (For one segment, steel, production continued to
decline through 1983.) Production declines for individual metals range
from 30 to 65 percent. Use of these production decline factors also
serves to bring estimated revenues more in line with other published
sources, notably Census. EPA has chosen to use its own estimates, as
revised, rather than other reported values, for three reasons:
the production estimates are based on data submitted by 438 wet
foundries in all size groups;
not all foundries will have recovered equally from the
recession; and
conflicting estimates from other sources suggest the use of a
lower estimate is appropriate to ensure that impacts are not
underestimated.
To maintain consistency, EPA has also estimated corresponding
reductions in capital and operating costs. These cost reductions
incorporated the concept of "economies of scale." Engineers have found
that increases in capacity generally do not require proportionate
increases in cost. As an approximation, engineers use cost curves of
the form
III-5
-------
Where: P is production
C is cost
x is the cost adjustment exponent
Thus, the cost at production level 2 is equal to the cost at
production level 1 times the ratio of production at level 2 to
production at level 1, raised to the cost adjustment exponent.
These cost adjustment exponents may be different for each metal,
employment size, and discharging process. In reviewing its data, EPA
developed a total of 22 cost adjustment factors (11 for capital costs,
11 for annual costs), with values ranging from 0.05 to 0.93. As an
example of the impact of the cost adjustment factors, we can analyze the
affect of a 50 percent reduction in production:
Impacts of 50 Percent Reduction in Production
Cost Exponent
Cost Reduction (percent)
0.05
3
0.93
18
1.00
50
Using a factor of 0.05, a 50 percent reduction in revenue will lead to
only a 3 percent reduction in treatment costs.
2. Estimate of Cotreatment Savings
Although cost estimates were separately developed for each
individual discharging process, many foundries have several processes
that each create a flow of process wastewater. For the same reasons
used to develop cost factors, it is generally less expensive to treat a
single larger stream than to install facilities to treat several smaller
streams separately.
EPA analyzed several combinations of processes that it believes
are typical of the foundry industry, and found that foundries cotreating
the discharges from several processes gain significant benefits 36
percent of operating costs and 28.9 percent of capital costs. EPA has
incorporated these savings into the estimated costs for any plant with
two or more discharging processes.
3. Development of Annual Costs
EPA estimated compliance impacts by combining capital costs and
operating and maintenance costs into a single equivalent annual cost.
^These cost adjustment factors may be found in the economic record
supporting these effluent guidelines.
III-6
-------
Annual costs are composed of two segments: operating costs Imposed by
requirements for power, labor, maintenance, chemicals, monitoring, and
sludge and oil disposal, and capital recovery costs incurred in
financing the equipment. As indicated, the operating costs were based
on technical considerations as described in the Development Document in
the public record. Capital recovery costs consist of the charges for
depreciation and interest. As presented at proposal, depreciation
charges were based upon a 10-year straight line depreciation. Interest
charges were calculated as follows:
average interest charges = capital recovery - average principal payment
s P (1(1+1)" ) _ ฃ
Where: P = capital cost of control technology
i = rate of interest
n = number of years over which the capital equipment is
depreciated and financed.
The analysis uses the DRI 1986 prime rate projection of 10.89
percent as the basis for computing interest expense with all plants
paying a premium over the prime rate. Interest charges are baaed on a
sliding scale with larger plants paying a lower interest premium:
Size
Premium over Prime Rate
(percent)
Fewer than 10 employees
6
10 to 49 employees
6
50 to 99 employees
5
100 to 249 employees
5
250 or more employees
3
Thus, the interest rate used for foundries with 250 or more
employees is 13.89 percent.
An example of capital and annual costs for the aluminum category
for 10 to 49 employees is shown in Table III-1.
D. DEVELOPMENT OF MODEL PLANTS
This analysis relies on the use of model plants to represent the
industry. Both technical and economic models were established. To
provide sufficient detail, the industry was subcategorized at several
levels: by metal, by employment size category, by discharge mode, by
type of foundry (Jobber or captive) and by financial status.
To estimate Impacts, EPA developed financial models of typical
plants. Data for the analysis were obtained from four major sources,
III-7
-------
TABLE III-1
SAMPLE COMPLIANCE COSTS PEB PLANT - ALUMINUM
(size = 10-49 employees)
(thousands of 1983 dollars)
I
oo
Combination
of Discharging
Technologies
Number
of
Jobbers
Number
of
Captives
Total
Dis-
chargers
Option 1
Option 2
Option 3
Option 4
Mode
Capital
Annual
Capital
Annual
Capital
Annual
Capital
Annual
Casting Cleaning/
Casting Quench
Direct
3
1
4
34.6
14.0
43.1
16.8
45.8
19.3
52.6
22.9
Casting Cleaning/
Casting Quench/
Die Casting
Direct
1
0
1
34.6
14.0
61.5
30.5
65.6
34.3
79.7
41.6
Die Casting
Direct
3
1
4
0.0
0.0
25.9
20.8
27.8
22.7
38.1
28.2
Casting Quenoh
Indirect
17
3
20
17.4
9.4
23.4
11.4
25.3
13.1
34.9
18.5
Casting Quenoh/
Die Casting
Indirect
3
1
4
12.4
6.3
35.0
21.4
37.6
23.5
51.7
30.8
Casting Quench/
Hold Cooling
Indirect
3
1
4
44.7
19.2
53.5
23.2
56.9
26.1
73.3
33.9
Die Casting
Indirect
10
2
12
0.0
0.0
25.9
20.8
27.5
22.3
37-8
27.9
Die Casting/
Mold Cooling
Indirect
4
1
5
32.3
12.9
55.3
29.2
58.5
32.0
75.4
40.0
Die Casting/
Melting Furnace
Scrubber
Indirect
3
1
4
15.6
7.1
40.5
26.4
44.1
29.3
51.4
33.0
Investment Casting
Indirect
4
1
5
37.1
14.4
42.6
17.1
51.9
21.8
67.4
29.4
Mold Cooling
Indirect
1
0
1
45.5
19.1
51.8
23.1
54.8
25.7
68.3
32.1
Melting Furnace
Scrubber
Indirect
0
1
1
21.9
10.6
31.1
19.2
34.4
22.0
34.4
22.0
Dust Collection
Indirect
4
1
5
37.0
13.0
42.5
17.0
44.9
19-2
44.9
19.2
-------
discussed below. Briefly, EPA combined estimates of average sales per
foundry with values of financial ratios to construct pro forma,
precompliance financial statements (balance sheets and income
statements). EPA then incorporated the capital and annual costs of
compliance into the financial statements to estimate the pro forma,
postcompliance financial status of each plant.
1. Use of Subcategories
As stated, EPA developed model plants to estimate impacts. To
obtain a sufficient degree of differentiation, EPA established models
for many subcategories of the industry.
a. Metal Type
The first subcategorization was by metal type. Available
data show that foundries generally specialize in a single metal. Census
data show that in general the foundries casting a specific type of metal
cast more than 90 percent of all production of that metal, and typically
derive more than 90 percent of their revenue from casting that metal
(see Chapter II). Also, different metals have different
characteristics, and thus different potential end markets. Because of
these differences, EPA has used eight metal types to represent the
industry. In this analysis, the metals are ordered by Standard
Industrial Classification (SIC) code:
SIC
METAL
3321
Gray iron (except ductile iron)
33211
Ductile iron (includes 33212)
3322
Malleable iron
3325
Steel (includes 3321)
3361
Aluminum
3362
Copper, Brass and Bronze (copper-
base)
33691
Zinc
33692
Magnesium
b. Size Category
The foundry industry encompasses a wide range of sizes. The
Census of Manufactures reports plant sizes ranging from fewer than five
to more than 2,500 employees. EPA recognizes the potential for
different impacts for foundries of different sizes. Based on its
analysis of the industry and public comments, EPA has used five
employment size segments to represent the industry:
fewer than 10 employees
10 to 19 employees
50 to 99 employees
III-9
-------
100 to 249 employees
250 or more employees.
Use of these size categories, combined with each of the
eight metal categories, leads to a potential of 40 separate
metal/employment size subcategories. In fact, EPA data show several
metal/employment size subcategories with no discharging foundries.
The subcategories with no discharging foundries are ductile
iron with fewer than 10 employees, malleable iron with fewer than 49
employees, steel with fewer than 10 employees, copper with fewer than 10
employees, and magnesium with fewer than 10 or more than 49 employees.
c. Jobber/Captive Category
EPA further recognizes that plant ownership may affect a
plant's response to regulation. Previous EPA regulations have shown two
categories of ownership. Jobber foundries are independently owned and
operated plants selling castings on the open market. Captive foundries
are plants that sell or transfer their products to other operations of
the same company. The percentages of jobber and captive foundries in
each subcategory were reported in Chapter II.
In computing the economic impacts, captive foundries were
treated as though the financial decisions were made on a plant-by-plant
basis and subject to the same financial tests as jobbers. The Agency
has concluded, in accordance with current financial theory, that parent
corporations most often treat their operations in different industries
as separate companies, subject to the normal financial structure and
restrictions of those industries. The Agency recognizes that this may
undervalue any benefits of conglomeration, such as centralized
accounting and scheduling, lower corporate borrowing costs, etc. For
reasons explained more fully in Chapter VIII, Limitations of the
Analysis, this treatment is believed to overstate impacts on the captive
segment of the industry.
d. Economic Quartlle Category
EPA recognizes that different foundries in the same size
category casting the same metal may have different financial health.
EPA has addressed this issue by using economic quartiles.
The concept of quartiles originates in statistics. If
several items are measured, the individual measurements can be arranged
in order of size. For example, if a group of similar castings were
weighed, one would find that they did not have identical weights. The
values of the weights can then be sorted, smallest to largest, and
broken into four segments. The lowest one-fourth (one-quarter) of all
values is the lowest quartlle range. The upper one-fourth of values is
the upper quartlle range. The lower quartile value is the value
separating the lower quartile from the second quartile, and is the value
that exceeds one-fourth of all values. The upper quartile value Is the
value separating the upper quartile from the third quartile, and is the
111-10
-------
value that exceeds three-fourths of all values. The value separating
the second and third quartlles Is the median, and is the value that
exceeds one-half of all values.
For this analysis, EPA has grouped financial ratios into
quartiles. EPA has used the lower quartile value to represent all
values in the lower quartile range. The upper quartile value is used to
represent all values in the upper quartile range. The median value is
used to represent all values in the second and third quartiles.
As discussed below, EPA has developed quartile and median
values for four separate financial ratios: return on sales, sales to
net worth, debt to net worth, and net fixed assets to net worth. These
ratios have been combined to form three sets of financial ratios for
each metal/employment size category, and are used to create three
separate financial statements. EPA assumes that the lower and upper
quartile statements each represent 25 percent of a metal/employment size
category, while the median represents 50 percent.
2. Estimation of Precompliance Financial Statements
In this analysis, EPA has used model plants to represent the
affected foundries. Precompliance financial statements were developed
in three steps: first, estimated sales were developed for each
metal/employment size subcategory; second, the ratios of various
financial statement items were developed from various data sources for
three quartiles of financial health; third, average sales were combined
with the financial ratios to create three separate financial statements
in each metal/employment size subcategory.
a. Estimation of Average Sales Per Foundry
Assumptions about sales per foundry play a critical role in
the economic analysis, because foundry sales are used to establish firm
size and income. The data from EPA's 308 survey form the basis for
establishing cost estimates. To maintain consistency, EPA has based its
estimate of average sales on the production figures reported by the
survey respondents. Forecasted sales were adjusted to reflect the
overall industry decrease in production from 1978 to the lowest point
since then. For most metals, 1982 represented the lowest industry
production level. For steel, however, 1983 was the year with the lowest
tonnage shipped. These adjustment serve (1) to provide a conservative
estimate of impacts, and (2) to make the revenue estimates more
consistent with other sources, such as Census.
Recent data show the industry is recovering to some extent
from the 1982 levels (see Table III-2). EPA has used the lower levels
as a prudent measure in capturing impacts. To the extent the economic
recovery in the industry continues, the impacts may be overstated. If
the recovery slackens to 1982 levels, however, the impacts will not be
understated. Given the improvements in most subcategories since 1982,
the impacts presented in this analysis may be overstated.
111-11
-------
TABLE III-2
TRENDS IN SHIPMENTS
QUANTITY
Gray
Ductile
Malleable
Steel
Copper-
Iron
Iron
Iron
NEC
Aluminum
Base
Zinc
Magnesium
Year
(millions of tons)
(billions of pounds)
1972
13.467
1.835
0.961
1.584
1.916
0.920
0.938
0.050
1973
14.801
2.246
1.031
1.894
2.113
0.966
1.080
0.054
1974
13.458
2.203
0.912
2.091
1.857
0.857
0.843
0.058
1975
10.547
1.823
0.729
1.974
1.455
0.700
0.712
0.038
1976
11.923
2.245
0.848
1.767
1.971
0.682
0.869
0.053
1977
12.371
2.736
0.829
1.677
2.153
0.702
0.789
0.058
1978
13.140
2.984
0.790
1.857
2.287
0.743
0.760
0.051
1979
12.512
2.890
0.715
2.039
2.303
0.793
0.665
0.028
1980
9.399
2.400
0.450
1.879
1.690
0.592
0.486
0.025
1981
9.610
2.191
0.422
1.743
1.820
0.581
0.471
0.023
1982
6.393
1.822
0.284
1.017
1.605
0.456
0.405
0.018
1983
7.180
2.067
0.291
0.729
1.822
0.552
0.516
0.024
1984
8.014
2.607
0.360
0.956
1.830
0.625
0.565
0.0243
PERCENTAGE CHANGE
1978-82
-51
-39
-64
-45
-30
-39
-47
-65
1982-84
+25
+43
+27
-6
+ 14
+37
+40
+33
1978-84
-39
-13
-54
-49
-20
-16
-26
-53
SOURCE: U.S. Bureau of the Census, Current Industrial Reports.
aEstlmated. Bureau of the Census estimates approximately 2-1/2 to 5 percent increase in magnesium
castings from 1983 to 1981.
-------
The calculation of average sales was conducted in four
steps. The first step calculated the average sales price per casting,
in dollars per unit weight. While different types of castings have
different values on a price per pound basis, these differences seem
unrelated to the size of the company. An average castings price was
estimated using preliminary data from the 1982 Census of Manufactures,
extrapolated to 1983 dollars using wholesale price indices. The Census
of Manufactures reports both the quantity of shipments and the value of
castings at the 7-digit SIC code level. For all categories listed, the
quantity of shipments and the value were obtained. The value divided by
the quantity is equal to the base price of castings in 1982 dollars.
This price was then escalated to 1983 dollars using the price indices
for ferrous and nonferrous metals in the 198M U.S. Industrial Outlook,
published by the Department of Commerce Bureau of Industrial Economics.
The second step determined the average shipments of
castings. Average production and yield data were derived from responses
to EPA's data collection efforts. These responses, collected in Data
Collection Portfolios, are in the technical record. Table III-3
presents the average production and yield data, as well as the average
annual sales per foundry, for each metal and size group.
The third step consisted of a simple multiplication of the
average shipments times the average price of castings. The fourth step
adjusted the average sales figures downward to reflect reductions in
industry shipments from 1978 to 1982. In making this adjustment, EPA
recognizes that sales vary from year to year. Industry sales in 1982
(or 1983 for steel) were the lowest in decades and represent the deepest
recession since World War II. Use of 1982 shipment data thus present
the lowest shipment data consistent with EPA's survey of 438 wet
plants. As previously stated, the shipment data are now more consistent
with other sources.
b. Estimation of Ratios
In the second phase of establishing precompliance financial
statements, EPA estimated values for six financial statement ratios:
sales to net worth (S/NW)
return on sales (ROS)
debt to net worth (D/NW)
net fixed assets to net worth (NFA/NW)
gross fixed assets to net fixed assets, and
depreciation to gross fixed assets.
For the first four ratios, EPA used the FINSTAT database to
estimate quartile values. This database (described below) has been
developed and maintained by the Small Business Administration, and
incorporates data from Dun 4 Bradstreet and Standard and Poor. In using
this database, EPA eliminated the records of firms with ratios failing
the closure criteria, as described below. To estimate the ratio of
gross fixed assets to net fixed assets, EPA reviewed annual reports and
Form 10-Ks submitted to the Securities and Exchange Commission by
111-13
-------
TABLE III-3
CALCULATION OF AVERAGE SALES PER FOUNDRY
Metal/Size
Productionฎ
Yield
(*)a
Tons Shipped
per Year
1983
Average
Price
(thousand
dollars per
short ton)
1983
Average
Sales per
Foundry
(thousand
dollars per
short ton)
Ratio of
Shipments
1982-78
(Steel
1983-78)
Revised
Sales
per
Foundry
Number
of
Dischargersc
Sales by
Dischargers
(thousand
dollars)
Short Tons
per Day
Short Tons
per Year
Gray Iron
Fewer than 10
20
5,200
0.66
3,432
0.810
2,781
0.486530
1,353
4
5,411
10 to 49
14
3,640
0.66
2,402
0.810
1,946
0.486530
947
81
76,708
50 to 99
38
9,880
0.66
6,521
0.810
5,283
0.486530
2,570
65
167,080
100 to 249
161
41,860
0.66
27,628
0.810
22,384
0.486530
10,891
118
1,285,095
250 or more
581
151,060
0.66
99,700
0.810
80,778
0.486530
39,301
75
2.947,575
Total
343
4,481,869
Duo tile Iron
Fewer than 10
20
5,200
0.56
2,912
0.846
2,464
0.610590
1,504
0
0
10 to 19
14
3.640
0.56
2,038
0.846
1,725
0.610590
1,053
14
14,742
50 to 99
38
9,880
0.56
5,533
0.846
4,681
0.610590
2,858
8
22,865
100 to 249
161
41,860
0.56
23,442
0.846
19,832
0.610590
12,109
30
363,278
250 or more
581
151,060
0.56
84,594
0.846
71,568
0.610590
43,699
19
830,275
Total
71
1.231,159
Malleable Iron
Fewer than 10
20
5,200
0.48
2,496
1.666
4,160
0.359494
1,495
0
0
10 to 49
14
3,640
0.48
1,747
1.666
2,912
0.359494
1,047
0
0
50 to 99
38
9,880
0.48
4,742
1.666
7,903
0.359494
2,841
16
45,458
100 to 249
161
41,860
0.48
20,093
1.666
33,484
0.359494
12,037
38
457,417
250 or more
581
151,060
0.48
72,509
1.666
120,834
0.359494
43,439
15
651,583
Total
69
1.154,458
Steel
Fewer than 10
20
5,200
0.59
3,068
2.719
8,342
0.392569
3,275
2
6,550
10 to 49
14
3,640
0.59
2,148
2.719
5,840
0.392569
2,292
19
43,557
50 to 99
38
9,880
0.59
5,829
2.719
15,850
0.392569
6,222
37
230,227
100 to 249
161
41,860
0.59
24,697
2.719
67,156
0.392569
26,363
48
1,265,432
250 or more
581
151,060
0.59
89,125
2.719
242,344
0.392569
95,137
28
2,663,825
Total
134
4,209,591
(Continued)
-------
TABLE III-3 (Continued)
CALCULATION OF AVERAGE SALES PER FOUNDRY
Metal/Size
Productionฎ
Yield
<*)a
Tons Shipped
per Year
1983
Average
Price
(thousand
dollars per
short ton)
1983
Average
Sales per
Foundry
(thousand
dollars per
short ton)
Ratio of
Shipments
1982-78
(Steel
1983-78)
Revised
Sales
per
Foundry
Number
of
Dischargers0
Sales by
Dischargers
(thousand
dollars)
Short Tons
per Day
Short Tons
per Year
Uialnw
Fewer than 10
12
3,120
0.71
2,215
4.074
9,024
0.701990
6,335
13
82,356
10 to 19
6
1,560
0.71
1,108
4.074
4,512
0.701990
3,168
103
326,257
50 to 99
32
8,320
0.71
5,907
4.074
24,065
0.701990
16,894
33
557,487
100 to 249
65
16,900
0.71
11,999
4.074
48,882
0.701990
34,315
62
2,127,532
250 or more
134
34,840
0.71
24,736
4.074
100,773
0.701990
70,742
21
1,485,577
Total
232
4,579,208
Copper-teae
1.786
2,563
0.612874
Fewer than 10
8
2,080
0.69
1,435
1,571
27
42,416
10 to 19
110
28,600
0.69
19,734
1.786
35,245
0.612874
21,601
52
1,123,250
50 to 99
72
18,720
0.69
12,917
1.786
23,070
0.612874
14,139
29
410,025
100 to 219
258
67,080
0.69
46,285
1.786
82,666
0.612874
50,664
15
759,961
250 or more
153
39,780
0.69
27,448
1.786
49,023
0.612874
30,045
10
300,450
Total
133
2,636,102
Zlno
1,b49
Fewer than 10
0.9
234
0.83
194
3.570
693
0.533279
370
5
10 to 49
7
1,820
0.83
1,511
3.570
5,393
0.533279
2,876
23
66,145
50 to 99
15
3,900
0.83
3,237
3.570
11,556
0.533279
6,163
15
92,439
100 to 249
82
21,320
0.83
17,696
3.570
63,173
0.533279
33,689
22
741,158
250 or more
25
6,500
0.83
5,395
3.570
19,260
0.533279
10,271
7
71,897
Total
72
973,489
MagnealM
Fewer than 10
0.2
52
0.62
32
6.342
204
0.357499
73
1
73
10 to 49
0.8
208
0.62
129
6.342
818
0.357499
292
1
1,170
50 to 99
10
2,600
0.62
1,612
6.342
10,224
0.357499
3,655
2
7,310
100 to 249
0
0
0.62
0
6.342
0
0.357499
0
0
0
250 or more
0
0
0.62
0
6.342
0
0.357499
0
0
0
Total
7
8,553
Grand Total
.
19,274,428
aBased on data collected by EPA between 1978-1981.
^Based on 260 operating days per year.
ฐlnoludea direct, indirect and zero dischargers.
-------
publicly held companies. These data showed that net (depreciated) fixed
assets range from *10 to 60 percent of gross (historical value) fixed
assets. EPA assumed that net fixed assets are 50 percent of gross fixed
assets for all metals and employment size categories. EPA used data
from the Bureau of Census Annual Survey of Manufactures to obtain the
ratio of depreciation to gross fixed assets. Census reports the total
gross fixed assets and depreciation for each industry at the 4-digit SIC
level.
In summary, the ratios used and their sources are
Ratio
Source
Return on Sales
Sales to Net Worth
Debt to Net Worth
Net Fixed Assets to Net Worth
Gross Fixed Assets to
Net Fixed Assets
Depreciation to Gross Fixed Assets
FINSTAT
FINSTAT
FINSTAT
FINSTAT
Review of Financial Statements
Annual Survey of Manufactures
(1) Description of the FINSTAT Database
The Small Business Administration (SBA) maintains the
FINSTAT database so that it may assess the impacts of policies or
regulations on firms of different sizes. The data in FINSTAT are
originally collected by Dun 4 Bradstreet (D&B) as part of its credit
reporting activities, and includes more than 3 million records spanning
the 1975-1984 period.
SBA has leased the database from D&B and made several
improvements. First, SBA merged data for large publicly-held
corporations with the D&B database by incorporating data from Compustat
(a database established by Standard and Poor). Second, where firms have
supplied both interim and final statements to D&B, SBA removed the
interim statements as being less reliable. Third, SBA reviewed each
financial statement against five financial criteria that test the
validity of balance sheet identities. Fourth, SBA developed "outlier"
tests that removed a small fraction of firms having financial ratios
substantially removed from the norm. By modifying the database through
these four steps, SBA reduced the number of financial statements from
about 3.4 million to about 1.4 million valid statements. Although
ratios available from Dun & Bradstreet come from the same original data
as FINSTAT, the 'data in FINSTAT have undergone more rigorous
verification.
(2) Use of FINSTAT by EPA
SBA supplied approximately 2,000 financial records for
firms whose SIC codes corresponded to the foundry industry. These
111-16
-------
records were of financial statements from the 1975 to 1984 time
period. Upon review of the data, EPA determined that many of the
records were not suitable for the analysis because they (1) failed to
meet SBA's criteria for financial reasonableness, or (2) had financial
ratios that failed the criteria established by EPA for its closure tests
(see below). EPA made a further decision that quartile values would be
developed only where the subcategory contained five or more financial
records. Because there were only three valid statements for malleable
iron foundries with 50 to 99 employees, the statements for this group
were merged with statements for the 100 to 249 size category. In all
1,302 financial statements were used in establishing financial ratios.
EPA chose not to include financial statements that
failed the financial criteria used for estimating impacts. EPA
determined that when such financial statements were included, many
metal/size groups showed lower quartile values indicating closure before
the imposition of compliance costs. EPA considers that the inclusion of
these ratios interferes with the analysis in two ways.
First, EPA believes that firms whose financial
statements indicate failure represent baseline closures. In other
words, EPA*s studies suggest that those firms will be already closed by
the date final regulations are promulgated, so that use of their ratios
distorts the estimate of the ratios that will prevail at promulgation.
Second, use of those ratios would almost completely
negate the use of an incremental analysis to estimate closures due to
the regulation. If EPA were to include the financial statements that
show closure, almost one-fourth of all foundries would be designated as
"baseline" closures, and almost no foundries would be shown as
incremental closures caused by the regulations. EPA believes that
excluding those financial statements that portray precompliance closure
will lead to more accurate estimates of the potential incremental
closures caused by the regulations.
In summary, EPA rejected the records of firms with the
weakest financial condition. If the records had not been rejected, the
1984 foundry population estimate would have shown a large number of
baseline closures. By dropping the records, EPA is basing its analysis
on those "better than marginal" firms that will survive to be subject to
this rulemaking.
c. Construction of Financial Statements
To construct the financial statements, EPA had to address
two Issues:
any one of the first four ratios can be used to allocate
data in the upper, median, and lower quartiles; hence one of
the ratios should be selected for construction purposes.
once one of the ratios has been selected, internally
consistent financial statements must be constructed.
111-17
-------
In this study, the return on sales ratio is used for the
initial allocation. Selection of return on sales as the ratio to define
the quartile has no impact on the outcome of the tests; it only
determines whether the model financial statements are considered "upper
quartile" or "lower quartile." The problem of constructing internally
consistent financial statements is illustrated below. The solution
chosen by EPA is also given.
Consider three firms, Able, Baker, and Charley, with the
following financial characteristics:
Able
Baker
Charley
Sales
1,000
2,000
1,000
Income
100
80
240
Assets
1,000
1,000
1,000
Return on Sales (ROS)
10}
4$
6 $
Return on Assets (ROA)
10?
8*
21*
Sales to Assets (S/A)
1
2
4
Using our three companies as the sample population, the
ratios in the quartiles are as follows:
Upper
Quartile
Median
Lower
Quartile
ROS
10$ (A)
6$ (C)
4$ (B)
ROA
24$ (C)
10$ (A)
8$ (B)
S/A
4 (C)
2 (B)
1 (A)
It is not true for any quartile that return on sales times sales to
assets equals return on assets. Although this example is hypothetical,
the same results are observed when examining the quartiles from the
FINSTAT database. In deriving balance sheets from the quartile data, we
have maintained the general relationship that increasing debt imposes
interest costs that decrease net income and that the fraction of debt is
smaller for larger companies. For deriving the model financial
statements, we used the following characteristics:
highest ROS with lowest D/NW, S/NW, and NFA/NW;
median ROS with median D/NW, S/NW, and NFA/NW; and
lowest ROS with highest D/NW, S/NW, and NFA/NW.
This procedure increases the likelihood of at least one quartile falling
more than one of the closure tests and thus may overestimate potential
Impacts.
111-18
-------
The following shows the specific construction of the financial
statements using the above ratios:
NI = Sales x ROS (by quartile)
Net Worth = Sales x Net Worth to Sales (by quartile)
Debt = Debt to Net Worth x Net Worth (by quartile)
Assets = Debt + Net Worth
NFA = Net Worth x (Net Fixed Assets/Net Worth) (by quartile)
GFA = NFA x (Gross Fixed Assets/Net Fixed Assets)
Depreciation = GFA x (Depreciation/GFA)
Gross Income = NI adusted for taxes.
Where:
NI = Net Income;
ROS = Return on Sales;
NFA = Net (depreciated) Value of Fixed Assets; and
GFA = Gross (historical) Value of Fixed Assets.
Gross income (net before taxes) is estimated by "backing
out" taxes. Taxes are assumed to be based on the following schedule:
20 percent of the first $25,000 of gross income
22 percent of the second $25,000 of gross income
U6 percent of gross income greater than $50,000
Table III-4 provides an example of the development of precompliance
financial statements in the aluminum, 10 to 19 employment size segment.
d. Comparison to Previous Analyses
This analysis has two main differences from previous
analyses. First, EPA no longer assumes that 1985 will resemble any
particular year for the purposes of estimating precompliance financial
statements. Instead, EPA has used financial data from the entire period
of 1975 to 1964 to estimate the financial statements. EPA believes the
quartiles developed represent the widest feasible range, using the
largest amount of data.
Second, EPA is basing its sales estimates on the results of
its surveys, adjusted for sales declines between 1978 and 1982 (or 1983
for steel). EPA recognizes that the adjusted sales estimates are still
higher than those shown in other data sources. However, EPA believes
the data it gathered are the most reliable because they were supplied
directly by 1,266 foundries. As shown in the technical record, EPA has
made extensive efforts to verify its data sources, and has recontacted
many foundries to confirm the values for shipments and employment.
111-19
-------
TABLE III-4
SAMPLE DERIVATION OF PRECOMPLIANCE FINANCIAL STATEMENTS
Basis: Aluminum, 10 to 49 employees. Dollars In thousands.
Upper
Quartlle
Medism
Lower
Quartlle
1.
Sales3
3,167.5
3,167.5
3,167.5
2.
Return on Sales (%)
6.44
4.67
3.17
3.
Sales to Net Worth
2.76
3.85
5.93
4.
Debt to Net Worth (%)
43.6
84.34
165.92
5.
Fixed Assets to Net Worth (%)
6.25
28.43
59.17
6.
Gross Assets to Fixed Assets0
2
2
2
7.
Depreciation to Assets (%r
6.7464
6.7464
6.7464
8.
Net Income (1x2)
203.98
147.92
100.41
9.
Net Worth (1 ~ 3)
1,147.6
822.7
534.1
10.
Debt (4 x 9)
500.3
693.9
886.2
11.
Total Assets (9 + 10)
1,648.0
1,516.6
1,420.3
12.
Net Fixed Assets (5 x 9)
71.7
233.9
316.0
13.
Gross Fixed Assets (6 x 12)
143.5
467.8
632.1
14.
Depreciation (7 x 13)
9.7
31.6
42.6
15.
Cash Flow (8 + 14)
213.66
179.52
143.0
16.
R0A (8 4 11) (*)
12.4
9.8
7.1
17.
Debts/Assets (10 ~ 11)(%)
30.4
45.8
62.4
18.
Cash Flow to Debt (15 ~ 10)(%)
42.7
25.9
16.1
19.
Gross Income (8 + adj. for taxes)
354.5
250.8
162.8
aBased on 308 survey data.
bFINSTAT.
cStudy of financial statements.
dAnnual Survey of Manufactures.
Ill-20
-------
It will be noted that the declines In sales to 1982 and 1983
levels that are incorporated into this analysis range between 30 and 65
percent, providing a substantial downward adjustment to the revenue
figures. These adjustments provide the lowest estimate of shipments for
foundries similar to those in EPA's database. Because of the losses of
"economies of scale," they give the highest relative impact of the
imposition of compliances costs, such as cost as a percentage of
sales. They thus provide the most conservative estimate (highest
potential impact) consistent with EPA's data sources.
3. Incorporation of Cost Estimates
Closures are based on the financial ratio values after
compliance. These ratios are obtained by adjusting the precompliance
financial statements to reflect compliance costs. For this analysis,
EPA has assumed that compliance capital costs are financed entirely by
debt. Net income after compliance is estimated by first subtracting the
annual cost (including interest) and depreciation from the estimated
precompliance gross income and then subtracting estimated tax
liability. The increased depreciation is estimated assuming a schedule
of 10-year straight line depreciation.
E. ESTIMATION OF IMPACTS
The fourth major step in the analysis is the actual calculation of
impacts. The tests and threshold values used to estimate plant closures
have been chosen on the basis of a literature search, the observed data
for three firms in the metals processing industry that have gone
bankrupt since 1978, and data for solvent firms in the industry. The
background for the selection of tests and thresholds is explained in
detail in Appendix A. The impact analysis is discussed below.
1. Choice of Tests
An intensive search of the financial literature was made in an
effort to identify suitable tests and threshold values for the closure
analysis. (See Appendix A for a detailed discussion.) On the basis of
this search, the following three tests have been chosen to measure the
economic impacts:
return on assets;
total debt to total assets; and
"Beaver's ratio" (cash flow to total debt).
^These tests were found to occur most frequently in the literature as
significant tests of firm failure. Specific threshold values are
derived from the seminal article by William Beaver, "Financial Ratios as
Predictors of Business Failure, "Empirical Research In Accounting:
Selected Studies, 1966.
111-21
-------
2. Firm Failure Criterion
Compliance costs have been established for each known
combination of discharging technologies. For aluminum foundries in the
10 to 49 employee segment, for instance, five combinations of
discharging technologies were found in EPA's 1978 survey. The
compliance costs for each combination were used to establish three
separate, postcompliance balance sheets, one for each quartile.
Each test is applied separately to the derived financial
statements of each quartile. In each case that a model plants fails two
tests, the entire quartile represented by the financial statements is
considered to fail. Twenty-five percent of the employment segment is
represented by each of the upper and lower quartiles, and 50 percent by
the median. Thus, if there are eight plants in an employment size
category with a specific combination of discharging technologies, and
the lower quartile fails at least two tests, then two firms (one-fourth
of eight total) are estimated to close. By definition, the lower
quartile is intended to represent one-fourth of the plants using that
technology. The requirement that two of three tests fail recognizes
that some firms continue in business even though one measure is bad.
The requirement is also a recognition that the individual tests are not
100 percent effective, and may overstate closures. A close inspection
of the individual plant data obtained from the agency's independent case
study analysis showed that many firms do not close even when financial
conditions exceed these closure criteria.
The failure criterion is derived from an examination of the data
for bankrupt firms, which showed that companies did not file for
bankruptcy unless they failed at least two tests. (Details of the
examination are given in Appendix A, Section E.) This seems rational in
principle. If a company has a very high fraction of debt, but also has
sufficient income and cash flow to satisify investors and creditors, it
would likely stay in business. If a company has low return on assets,
but also fairly low debt and sufficient cash flow, again it would likely
remain in business. In the third case, if a company had a low Beaver's
ratio (cash flow to total debt) but low debt to assets and reasonable
return on assets, it would again probably stay in business. However, if
two ratios are below the threshold, there is much less chance of
recovery.
3. Description of the Threshold Values and Application of Tests
a. Return on Assets
Return on assets is defined as net income after taxes
divided by total book assets. In principle, this ratio measures the
efficiency of the firm at generating income from its asset base. To
apply the test, both Income and assets must be adjusted for the
compliance effects.
Postcompliance income was derived through the following
steps:
111-22
-------
Precompliance gross Income (GI) was estimated by adjusting
the net income by the Federal corporate income tax rate;
GI = NI + (estimated taxes)
This analysis assumes a progressive tax scale: 20 percent
tax on the first $25,000 of income, 22 percent on the next
$25,000, and 46 percent on all Income over $50,000.
Compliance costs have been assumed to be tax deductible as
an ordinary business expense, but the extent of the tax
savings depends on the Income of the foundry. Furthermore,
the marginal tax bracket may change as taxable Income
changes, so that the postcompliance income is adjusted for
taxes based on the postcompliance level.
Postcompliance gross income (PCGI) was calculated from pre-
compliance gross Income minus the estimated annual cost of
compliance (including debt service, depreciation, and
operating costs);
PCGI = GI - CAC (compliance annual cost).
Postcompliance net Income (PCNI) was taken from post-
compliance gross Income minus Federal corporate income
taxes.
PCNI = PCGI - (estlmate<:l taxes on
postcompliance income).
Postcompliance assets (PCA) equals precompliance assets plus
the capital cost of compliance, or
PCA = Assets + CCC (compliance capital cost).
The cut-off value for net income to total assets, as
determined by Beaver, ranges from an average of 1 percent one year
before failure to 3.5 percent five years before failure. On the basis
of the observed values of the ratio for failed firms and firms currently
in business, the Agency selected a reasonable cut-off value of 2.5
percent, or 0.025. (More detailed justification is given in Appendix A,
Section E.) The test can be written as follows. A firm will fail if
PCROA < 0.025,
where: PCROA = PCNI/PCA, and
PCROA = Postcompliance return on assets.
b. Total Debt to Total Assets
Total debt to total assets is the ratio of all debt to total
assets. Total debt is defined as anything that cannot be considered to
be owner's equity, and it thus includes accounts payable and accrued
expenses.
This ratio is computed as follows:
111-23
-------
Postcompliance debt equals precompliance debt plus the
capital cost of compliance (assumed to be financed at 100
percent debt); and
PC Debt = Debt + CCC.
Postcompliance total assets equals precompliance total
assets plus the capital cost of compliance.
PC Assets = Assets + CCC.
For this study a value of 0.7 is selected as a reasonable
cut-off for the debt to assets ratio. (Detailed justification is
presented in Appendix A, Section E.) The test is written as follows.
The firm will fail if
PC Debt/PC Assets > 0.7.
c. Beaver's Ratio
Beaver's ratio is defined as cash flow divided by total
debt. For purposes of computing the ratio, cash flow is defined as net
income after taxes plus depreciation. Methods for computing pre- and
postcompliance net Income and total debt have been explained above.
Postcompliance cash flow will be computed in the following steps:
Postcompliance depreciation is precompliance depreciation
plus the depreciation of the compliance equipment, assumed
to take place on a 10-year, straight line basis;' and
PCDepreciation = Depreciation + CCC/10.
Postcompliance cash flow (PCCF) is postcompliance income
(computed as explained above) plus postcompliance
depreciation.
PCCF = PCNI + PCDepreciation.
This measure may be interpreted as an indication of a firm's
ability to repay the interest and principal of borrowings. In Beaver's
study, the cut-off value ranged from 0.05 one year before failure to
0.11 five years before failure. After analyzing the data for a few
failed firms, examining the data for solvent firms, and evaluating the
conditions in the economy, the study has selected 0.08 as an appropriate
value. (More detailed justification is found in Appendix A, Section
E.) The test is written as follows. The firm will fail if
^Recent changes in the tax laws allow more rapid depreciation for tax
purposes. For companies using a more rapid depreciation, reported ROA
will be less, but cash flow will be commensurately better. Although EPA
believes the tax law changes benefited the foundry Industry, the average
effects on ROA and cash flow, and thus the required adjustments to
threshold values, are unknown. As a result, this study assumes
accounting practices in line with historical practices.
111-24
-------
PCCF < 0.08
Total Debt
4. Sample Calculations
For every subcategory (at the level of metal/employment
size/jobber-captive) all relevant costs were applied to all three
quartiles. For example, for aluminum, 10 to 49 employees, there are
three processes or combinations used by direct dischargers and ten
processes or combinations used by indirect dischargers. Financial
models were created for each process, for both jobbers and captives, for
all three quartiles, for four levels of treatment, leading to a total of
276 financial statements.
EPA made no assumptions as to which processes or combinations of
processes are used by foundries in any quartile of financial health.
Instead, EPA adopted an "expected value" approach. The method assumes
that plants with a given process are equally well represented in all
three financial quartiles. If a given process is estimated to be used
by four plants, EPA assumes that one plant (25$) is in the lower
quartile, two plants (50$) are at the median, and one plant (25$) is in
the upper quartile. The financial tests are then performed for each
quartile separately.
The concept of expected value extends even where there is only
one occurrence of a process or combination of processes. In such an
instance, 0.25 plants are allocated to the lower quartile, 0.5 plants to
the median, and 0.25 plants to the upper quartile. To calculate
aggregate impacts, EPA added up any fractional closures, and rounded to
the nearest whole number.
Table III-5 presents an example of the calculations used to
develop post-compliance balance sheets, using data for aluminum plants
in the 10 to 19 employee subcategory that directly discharge process
water from the casting cleaning process.
^Three process combinations are used by only one discharger. As shown
in Table III-1, there are a total of 23 process combinations used by
Jobbers and captives.
Ill-25
-------
TABLE III-5
SAMPLE DERIVATION OF POSTCOMPLIANCE FINANCIAL STATEMENTS
Basis: Aluminum, 10 to 49 Employee Size
Treatment of Wastewater from Directly Discharging
Jobber Foundries with the Casting Cleaning/Casting Quench
Process Combination at Option 1
Upper
Lower
Quartile
Median
Quartile
1.
Number of Dischargers (3 total)
0.75
1.5
0.75
2.
Capital Cost
34.6
34.6
34.6
3.
Annual Cost
14.0
14.0
14.0
4.
Precompliance Net Worth
1,147.6
822.7
534.2
5.
Precompliance Debt
500.3
693.2
886.3
6.
Precompliance Total Assets
1,648.0
1,515.6
1,420.3
7.
Precompliance Gross Income
354.5
250.8
162.8
8.
Precompliance Depreciation
9.7
31.6
42.6
9.
Postcompliance Debt (5 + 2)
534.3
727.8
920.9
10.
Postcompliance Total Assets (4 + 9)
1,681.9
1,550.5
236.8
1,455.1
11.
Postcompliance Gross Income (7-3)
340.5
148.8
12.
Postcompliance Net Income (11 - taxes)
196.4
140.2
92.9
13.
Postcompliance Depreciation (8 + (2 x (.1)))
13.2
35.1
46.1
14.
Return on Assets (12 ~ 10) {%)
11.7
9.0
6.4
15.
Debts to Assets (9 ~ 10) (%)
31.8
47.0
63.3
16.
Cash Flow to Debt ((12 + 13)/9) (*)
39.2
24.1
15.1
III-26
-------
CHAPTER IV
EFFLUENT CONTROL AND GUIDELINE COSTS
-------
IV. EFFLUENT CONTROL AND GUIDELINE COSTS
EPA has developed compliance cost estimates for foundries within the
framework of the metal type and employment size segmentation format
described in Chapter III, Methodology. Two types of compliance cost
estimates were developed: (1) those pertaining to equipment that
discharging foundries have already collectively installed; and (2) those
pertaining to required equipment that .discharging foundries must install
to comply with various levels of pollutant removal. For the analysis,
only those costs still required to meet the standards for direct and
indirect dischargers were considered. Expenditures already made for
equipment in-place were regarded as having been spent for operational
reasons.
Foundries incurring costs as a result of this regulation may incur
two kinds of costs: one-time capital costs, and recurring operating and
maintenance costs. The capital costs are those costs incurred when the
water treatment equipment is first installed, and include costs for
equipment design and installation. Operating and maintenance costs are
costs incurred on a periodic basis throughout the operation of the
facility, and include operation and maintenance, labor and materials,
sludge and oil disposal, energy and chemicals, and monitoring costs.
Details of the cost estimation procedure are given In the Development
Document in the technical record.
The costs were considered by EPA as the total costs to treat all
occurrences of each individual process that were in existence as of the
1984 survey of plants. Costs have been provided for treatment at four
levels, which involve increasingly higher levels of pollutant removal.
From the costs of treating individual processes, costs to treat the
process combinations commonly found in the industry were developed.
First, average costs per plant to treat discharges from each individual
process were calculated. These average costs were then added together
to provide an estimate of the costs to treat the processes and
combinations of processes forecast to exist in 1986. An EPA analysis
showed significant economies in cotreating the discharges from several
processes. These economies averaged 28.9$ of capital costs and 36$ of
operating costs, and were applied to each process combination.
The total treatment costs are shown in Tables IV-1 through IV-4.
Additional tables, presenting the costs of treatment within each size
category of each metal, are shown in Chapter VI.
1As shown below, this analysis Is based on an assumption that capital
costs are paid for by loans. In estimating total annual expenditures,
the capital cost is converted to a series of payments of principal and
interest, with charges to income for interest expense and depreciation.
(Continued)
IV-1
-------
TABLE IV-1
COMPLIANCE COSTS AND ECONOMIC IMPACTS FOUNDRY INDUSTRY
(Option 1 Recycle/Simple Settle)
Number of
Discharging
Foundries in 1986
1986 Compliance Costs
(in thousands of 1983 dollars)
Capital Investment
Annual Costs
Direct
Indirect
Direct
Indirect
Direct
Indirect
Total
Gray Iron
91
145
4,662
8,758
1,640
3,113
Ductile Iron
27
25
3,036
1,167
1,058
413
Malleable Iron
21
29
573
1,064
205
373
Steel
43
64
1,706
2,350
614
861
Total Ferrous
182
263
9,978
13,339
3,517
4,759
Aluminum
45
131
2,524
3,627
949
1,400
Copper-base
63
54
7,946
4,355
3,369
1,650
Zinc
9
49
151
1,175
73
460
Magnesium
2
2
47
57
22
20
Total Nonferrous
119
236
10,669
9,214
4,413
3,530
Grand Total
301
499
20,647
22,553
7,930
8,290
Jobber
Gray Iron
65
10
3,274
6,145
1,154
2,167
Ductile Iron
23
21
2,497
1,087
864
386
Malleable Iron
15
23
431
878
153
308
Steel
35
53
1,423
1,871
513
685
Total Ferrous
138
207
7,624
9,982
2,684
3,547
Aluminum
37
107
2,213
2,876
824
1,114
Copper-base
42
38
4,803
3,094
2,002
1,177
Zinc
7
38
105
834
52
334
Magnesium
2
2
47
57
22
20
Total Nonferrous
88
185
7,168
6,861
2,900
2,645
Grand Total
227
392
14,792
16,843
5,584
6,192
Captive
Gray Iron
26
35
1,389
2,613
486
945
Ductile Iron
4
4
539
80
193
27
Malleable Iron
6
6
143
186
51
64
Steel
8
11
283
478
102
176
Total Ferrous
44
56
2,354
3,358
832
1,213
Aluminum
8
24
311
751
125
286
Copper-base
21
16
3,143
1,261
1,367
473
Zinc
2
11
47
341
21
126
Magnesium
0
0
0
0
0
0
Total Nonferrous
31
51
3,501
2,353
1,513
886
Grand Total
76
107
5,855
5,711
2,345
2,099
IV-2
-------
TABLE IV-2
COMPLIANCE COSTS AND ECONOMIC IMPACTS FOUNDRY INDUSTRY
(Option 2 Recycle/Lime Addition/Settle)
Number of
Discharging
Foundries in 1986
1986 Compliance Costs
(in thousands of 1983 dollars)
Capital Investment
Annual Costs
Direct
Indirect
Direct
Indirect
Direct
Indirect
Total
Gray Iron
91
145
13,899
19,772
6,091
8,522
Ductile Iron
27
25
6,546
2,475
2,836
1,061
Malleable Iron
21
29
2,387
2,432
1,109
1,072
Steel
13
64
4,377
5,409
1,849
2,370
Total Ferrous
182
263
27,209
30,088
11,885
13,025
Aluminum
45
131
3,040
6,005
1,337
3,230
Copper-base
63
54
8,208
4,607
3,607
1,871
Zinc
9
49
197
1,700
123
880
Magnesium
2
2
59
65
26
23
Total Nonferrous
119
236
11,504
12,377
5,093
6,004
Grand Total
301
499
38,713
42,466
16,979
19,029
Jobber
Gray Iron
65
110
9,421
13,980
4,125
5,987
Ductile Iron
23
21
5,484
2,171
2,367
935
Malleable Iron
15
23
1,774
1,986
822
876
Steel
35
53
3,712
4,393
1,592
1,920
Total Ferrous
138
207
20,391
22,531
8,907
9,718
Aluminum
37
107
2,628
4,833
1,135
2,624
Copper-base
42
38
4,988
3,281
2,165
1,340
Zinc
7
38
138
1,243
89
665
Magnesium
2
2
59
65
26
23
Total Nonferrous
88
185
7,812
9,422
3,414
4,651
Grand Total
227
392
28,203
31,953
12,321
14,369
Captive
Gray Iron
26
35
4,478
5,792
1,966
2,535
Ductile Iron
4
4
1,062
304
468
127
Malleable Iron
6
6
613
446
288
196
Steel
8
11
665
1,015
257
450
Total Ferrous
44
56
6,818
7,558
2,979
3,307
Aluminum
8
24
412
1,172
202
606
Copper-base
21
16
3,220
1,326
1,443
531
Zinc
2
11
59
457
34
215
Magnesium
0
0
0
0
0
0
Total Nonferrous
31
51
3,692
2,956
1,679
1,353
Grand Total
76
107
10,510
10,514
4,658
4,660
IV-3
-------
TABLE IV-3
COMPLIANCE COSTS AND ECONOMIC IMPACTS FOUNDRY INDUSTRY
(Option 3 Recycle/Lime Addition/Settle/Filtration)
Number of
Discharging
Foundries in 1986
1986 Compliance Costs
(in thousands of 1983 dollars)
Capital Investment
Annual Costs
Direct
Indirect
Direct
Indirect
Direct
Indirect
Total
Gray Iron
91
145
15,702
22,152
7,099
9,892
Ductile Iron
27
25
7,450
2,799
3,341
1,245
Malleable Iron
21
29
2,641
2,736
1,259
1,245
Steel
43
64
4,854
5,900
2,144
2,691
Total Ferrous
182
263
30,647
33,587
13,843
15,073
Aluminum
15
131
3,353
6,440
1,599
3,652
Copper-base
63
54
9,012
4,944
4,173
2,112
Zinc
9
49
242
1,828
163
1,012
Magnesium
2
2
63
68
30
26
Total Nonferrous
119
236
12,669
13,280
5,964
6,801
Grand Total
301
499
43,316
46,867
19,806
21,875
Jobber
Gray Iron
65
110
10,674
15,686
4,823
6,964
Ductile Iron
23
21
6,237
2,460
2,789
1,100
Malleable Iron
15
23
1,962
2,228
932
1,015
Steel
35
53
4,096
4,798
1,829
2,185
Total Ferrous
138
207
22,969
25,172
10,372
11,265
Aluminum
37
107
2,893
5,183
1,357
2,965
Copper-base
42
38
5,493
3,526
2,515
1,514
Zinc
7
38
170
1,338
117
765
Magnesium
2
2
63
68
30
26
Total Nonferrous
88
185
8,619
10,114
4,020
5,270
Grand Total
227
392
31,588
35,295
14,390
16,535
Captive
Gray Iron
26
35
5,028
6,466
2,277
2,927
Ductile Iron
4
4
1,213
339
552
145
Malleable Iron
6
6
680
508
327
230
Steel
8
11
758
1,102
315
506
Total Ferrous
44
56
7,678
8,415
3,471
3,809
Aluminum
8
24
460
1,257
242
687
Copper-base
21
16
3,519
1,418
1,658
598
Zinc
2
11
72
490
45
247
Magnesium
0
0
0
0
0
0
Total Nonferrous
31
51
4,051
3,165
1,945
1,532
Grand Total
76
107
11,729
11,580
5,416
5,341
IV-4
-------
TABLE IV-4
COMPLIANCE COSTS AND ECONOMIC IMPACTS FOUNDRY INDUSTRY
(Option 1 Recycle/Lime Addition/Settle/Filtration/Carbon Adsorption)
Number of
Discharging
Foundries in 1986
1986 Compliance Costs
(in thousands of 1983 dollars)
Capital Investment
Annual Costs
Direct
Indirect
Direct
Indirect
Direct
Indirect
Total
Gray Iron
91
145
17,816
25,041
7,945
11,450
Ductile Iron
27
25
8,171
3,111
3,621
1,434
Malleable Iron
21
29
2,931
2,994
1,376
1,398
Steel
43
64
5,455
6,381
2,352
2,891
Total Ferrous
182
263
34,373
37,526
15,295
17,172
Aluminum
45
131
4,086
8,052
1,929
4,452
Copper-base
63
54
9,583
5,700
4,460
2,487
Zinc
9
49
416
2,468
243
1,335
Magnesium
2
2
81
68
40
26
Total Nonferrous
119
239
14,166
16,289
6,673
8,299
Grand Total
301
499
48,538
53,815
21,968
25,471
Jobber
Gray Iron
65
110
12,129
17,766
5,424
8,135
Ductile Iron
23
21
6,846
2,737
3,028
1,272
Malleable Iron
15
23
2,176
2,435
1,018
1,133
Steel
35
53
4,572
5,209
1,993
2,355
Total Ferrous
138
207
25,723
28,148
11,464
12,894
Aluminum
37
107
3,522
6,519
1,640
3,629
Copper-base
42
38
5,895
4,084
2,716
1,792
Zinc
7
38
299
1,826
176
1,012
Magnesium
2
2
81
68
40
26
Total Nonferrous
86
185
9,797
12,497
4,573
6,459
Grand Total
227
392
35,520
40,645
16,036
19,353
Captive
Gray Iron
26
35
5,686
7,274
2,522
3,315
Ductile Iron
4
4
1,325
374
592
162
Malleable Iron
6
6
755
558
358
265
Steel
8
11
883
1,171
359
536
Total Ferrous
44
56
8,649
9,378
3,831
4,278
Aluminum
8
24
563
1,534
289
822
Copper-base
21
16
3,688
1,616
1,744
695
Zinc
2
11
117
643
67
323
Magnesium
0
0
0
0
0
0
Total Nonferrous
31
51
4,368
3,792
2,100
1,840
Grand Total
76
107
13,017
13,170
5,931
6,118
IV-5
-------
Although there are only nine foundry processes with the potential to
produce process wastewaters, foundries vary greatly in the types and
combinations of discharging processes. EPA's study showed that ferrous
foundries in general have a wider diversity of discharging process
combinations:
Metal-Type
Direct
Indirect
Industry
Dischargers
Dischargers
Gray iron
20
18
Ductile iron
13
7
Malleable iron
7
8
Steel
10
8
Aluminum
16
13
Copper-base
8
7
Zinc
6
7
Magnesium
1
1
Additionally, the total costs indicated in the tables for the metal-
type foundry industries differed significantly in their content. Table
IV-5 shows the dominant discharging process combination as measured by
the required treatment costs. For some metals, the treatment costs for
one combination of discharging processes are so high that relatively few
foundries contribute most of the costs. For other metals, the treatment
costs are similar for all discharging processes.
IV-6
-------
TABLE IV-5
CONTRIBUTION OF THE MOST IMPORTANT DISCHARGER PROCESS
OR PROCESS COMBINATION TO THE TOTAL COST FOR EACH METAL
(in thousands of dollars)
Option 3 Recycle/Lime Addition/ Settle/Filtration
Most Important
Most
Total
Total Capital
Total Annual
Process as % of
Important
Dischargers with
Cost for
Cost for
Dischargers
Annual
Industry
Process
Process
Process
Process
Costs
Gray Iron
Melting Furnace Scrubber/Slag
Quench/Dust Collection
45
11,077
5,013
.19
.30
Ductile Iron
Slag Quench/Dust Collection
8
1,458
716
.15
.16
Malleable Iron
Dust Collection
19
1,834
961
.38
.38
Steel
Cast Quench/Dust Collection
22
3,043
1,404
.21
29
Aluminum
Dust Collection
52
1,474
1,323
.30
.25
Copper-base
Direct Chill Casting
19
3,806
1,890
.16
.30
Zinc
Casting Quench
22
623
316
37
.27
Magnesium
Casting Quench
2
64
30
.50
.54
-------
CHAPTER V
ANALYSIS OF ECONOMIC IMPACTS
-------
V. ANALYSIS OF ECONOMIC IMPACTS
EPA has estimated the potential economic impacts at four levels of
stringency. The levels were based on EPA's judgement of potential
levels of technology, and require increasing amounts of pollution
control equipment. Based on a review of its database and the potential
economic impacts, EPA has established effluent guidelines corresponding
to the Best Practicable Control Technology Currently Available (BPT) and
Best Available Control Technology Economically Achievable (BAT). In
some instances BPT was set equal to BAT. New Source Performance
Standards (NSPS) were considered identical to BAT. EPA has also
reviewed the costs and impacts attributable to effluent guidelines for
indirect dischargers (foundries whose effluent is treated by a publicly-
owned treatment works (POTWs) before discharge to surface waters).
Pretreatment Standards for New Sources (PSNS) were considered identical
to Pretreatment Standards for Existing Sources (PSES), and generally
need the same technologies as BAT.
This chapter presents a brief description of the four treatment
options reviewed, and then discusses the potential economic impacts of
each option.
A. BASIS FOR COMPLIANCE COSTS
Compliance-cost estimates for pollution control systems by foundries
pertain to the "Best Practicable Technology Currently Available" (BPT)
regulations and to "Best Available Control Technology Economically
Achievable" (BAT) regulations. Pretreatment technologies (PSES) for
foundries discharging indirectly to POTW's were considered identical to
the BPT and BAT treatment alternatives for directly discharging
foundries. Collectively, the proposed regulations involve four
treatment options, and two alternative options considered where small
business Impacts initially appeared high. Listed below are brief
descriptions of the various treatment technologies used for the four
principal options.
1. Option 1: Recycle and Simple Settling
Option 1 is comprised of high rate recycle achieved by settling
(including surface skimming for free oil removed in certain segments),
and recycle to the process (including pH adjustment as required to
maintain water chemistry balance between scaling and corrosion) and
including cooling towers for some segments, followed by settling of the
blowdown stream prior to discharge. Option 1 costs include the costs
for the grinding scrubber operations of aluminum, copper, ductile iron,
gray iron, malleable iron, magnesium, and steel plants. The grinding
scrubber treatment is similar to Option 1, but requires complete recycle
with no blowdown, and thus no blowdown treatment. Option 1 costs were
not developed for the aluminum and zinc die casting segments or the
ferrous dust collection and wet sand reclamation segments because the
treatment systems would be inadequate for the. treatment of wastes from
these processes.
V-1
-------
2. Option 2: Recycle, Lime Addition, and Settling
Option 2 consists of the addition of flocculation with lime and
polymer to the Option 1 technology to facilitate metals precipitation
and solids settling for blowdown treatment. This option also Includes
emulsion breaking for the aluminum and zinc die casting segments and
chemical oxidation of organic matter for these two segments and also for
ferrous dust collection and wet sand reclamation.
3. Option 3: Recycle, Lime Addition, Settling, and Filtration
Option 3 adds filtration of the effluent from the Option 2
facility through cartridge filters, multimedia filters, and pressure
filters, depending on the size of the systems.
4. Option 4: Recycle, Lime Addition, Settling, Filtration, and
Carbon Adsorption
Option 4 adds carbon adsorption treatment of the effluent from
the Option 3 facility. Option 4 costs were determined only for those
segments where the Option 3 effluent contained toxic organics at a level
that could be reduced by this method of treatment.
B. ECONOMIC IMPACTS OVERVIEW
This section provides a brief summary of the potential economic
impacts. More detailed discussions, with supporting information, are
presented later in the chapter.
1. Plant Closure and Employment Loss Impacts
EPA has used the potential loss of employment and closure of
plants as the primary measure of economic impacts. Precompliance
financial statements were established, using the model financial ratios
presented in Chapter II. Estimated compliance costs, in 1983 dollars,
were imposed on the model financial statements to estimate
postcompliance financial conditions. Where the model postcomplince
financial statements failed two of three tests, the number of firms
estimated to have those financial statements has been forecast to fall.
As shown in Tables V-1 through V-4, overall impacts under each
of the four options are expected to be low. Under Option 1, only four
foundries (two casting gray iron and two casting magnesium) are expected
to close. The associated Job loss of 100 employees represents 0.07
percent of the 149,287 employees of discharging foundries. Under Option
4, a total of 24 foundries are judged potential closures. The
^Because EPA could not determine the specific identities of foundries
that would Incur costs to segregate noncontact cooling water, impacts
are estimated assuming all foundries would incur these losses.
Actually, only 30 percent are expected to incur the incremental cost.
V-2
-------
TABLE V-1
COMPLIANCE COSTS AND ECONOMIC IMPACTS FOUNDRY INDUSTRY
(Option 1 Recycle/Simple Settle)
Number of
Discharging
Foundries
(in
Compliance Costs
thousands of 1983 dollars)
Closures
Capital
Investment
Annual Costs
Number of Foundries
Number of
Direct
Indirect
Direct
Indirect
Direct
Indirect
Direct
Indirect
Total
Employees
Total
Gray Iron
91
115
1,662
8,758
1,610
3,113
2
0
2
51
Ductile Iron
27
25
3,036
1,167
1,058
113
0
0
0
0
Malleable Iron
21
29
573
1,061
205
373
0
0
0
0
Steel
13
61
1,706
2,350
611
861
0
0
0
0
Total Ferrous
162
263
9,978
13,339
3,517
1,759
2
0
2
51
Aluminum
15
131
2,521
3,627
919
1,100
0
0
0
0
Copper-base
63
51
7,916
1,355
3,369
1,650
0
0
0
0
Zinc
9
19
151
1,175
73
160
0
0
0
0
Magnesium
2
2
17
57
22
20
1
1
2
16
Total Nonferrous
119
236
10,669
9,211
1,113
3,530
1
1
2
16
Grand Total
301
199
20,617
22,553
7,930
8,290
3
1
1
100
Jobber
Cray Iron
65
10
3,271
6,115
1,151
2,167
1
0
1
27
Ductile Iron
23
21
2,197
1,087
861
386
0
0
0
0
Malleable Iron
15
23
131
676
153
308
0
0
0
0
Steel
35
53
1,123
1,671
513
685
0
0
0
0
Total Ferrous
138
207
7,621
9,982
2,681
3,517
1
0
1
27
Aluminum
37
107
2,213
2,876
821
1,111
0
0
0
0
Copper-base
42
38
1,803
3,091
2,002
1,177
0
0
0
0
Zlno
7
38
105
831
52
331
0
0
0
0
Magnesium
2
2
17
57
22
20
1
1
2
46
Total Nonferroua
68
185
7,168
6,861
2,900
2,615
1
1
2
16
Grand Total
227
392
11,792
16,813
5,581
6,192
2
1
3
73
Captive
Cray Iron
26
35
1,389
2,613
186
915
1
0
1
27
Ductile Iron
1
1
539
80
193
27
0
0
0
0
Malleable Iron
6
6
113
186
51
61
0
0
0
0
Steel
8
11
283
178
102
176
0
0
0
0
Total Ferrous
11
56
2,351
3,358
832
1,213
1
0
1
27
Aluminum
6
21
311
751
125
286
0
0
0
0
Copper-base
21
16
3,113
1,261
1,367
173
0
0
0
0
Zinc
2
11
17
311
21
126
0
0
0
0
Magnesium
0
0
0
0
0
0
0
0
0
0
Total Nonferrous
31
51
3,501
2,353
1,513
886
0
0
0
0
Grand Total
76
107
5,855
5,711
2,315
2,099
1
0
1 .
27
-------
TABLE V-2
COMPLIANCE COSTS AND ECONOMIC IHPACTS FOUNDRY INDUSTRY
(Option 2 Recycle/Lime Addition/Settle)
Number of
Discharging
Foundries
(in
Compliance Costs
thousands of 1983 dollars)
Closures
Capital
Investment
Annual Costs
Number of Foundries
Number of
Employees
Dlreot
Indirect
Dlreot
Indirect
Direct
Indirect
Direct
Indirect
Total
Total
Cray Iron
91
115
13,899
19,772
6,091
8,522
3
2
5
135
Ductile Iron
27
25
6,516
2,175
2,836
1,061
0
0
0
0
Malleable Iron
21
29
2,387
2,132
1,109
1,072
0
0
0
0
Steel
13
61
1,377
5,109
1.8U9
2,370
0
0
0
0
Total Ferroua
182
263
27,209
30,088
11,885
13,025
3
2
5
135
Aluminum
15
131
3,010
6,005
1,337
3,230
0
0
0
0
Copper-base
63
51
8,208
1,607
3,607
1,871
0
0
0
0
Zinc
9
19
197
1,700
123
880
0
0
0
0
Magnesium
2
2
59
65
26
23
1
1
2
16
Total Nonferrous
119
236
11,501
12,377
5,093
6,001
1
1
2
16
Grand Total
301
199
38,713
12,166
16,979
19,029
1
3
7
181
Jobber
108
Gray Iron
65
110
9,121
13,980
1,125
5,987
2
2
1
Ductile Iron
23
21
5,181
2,171
2,367
935
0
0
0
0
Malleable Iron
15
23
1,771
1,986
8 22
876
0
0
0
0
Steel
35
53
3,712
1,393
1,592
1,920
0
0
0
0
Total Ferrous
138
207
20,391
22,531
8,907
9,718
2
2
1
108
Aluminum
37
107
2,628
1,833
1,135
2,621
0
0
0
0
Copper-base
12
38
1,988
3,281
2,165
1,310
0
0
0
0
Zlno
7
38
138
1,213
89
665
0
0
0
0
Magnesium
2
2
59
65
26
23
1
1
2
16
Total Nonferrous
88
185
7,812
9,122
3,111
1,651
1
1
2
16
Grand Total
227
392
28,203
31,953
12,321
11,369
3
3
6
151
Captive
Gray Iron
26
35
1,178
5,792
1,966
2,535
1
0
1
27
Ductile Iron
4
1
1,062
301
168
127
0
0
0
0
Malleable Iron
6
6
613
116
2 88
196
0
0
0
0
Steel
8
11
665
1,015
257
150
0
0
0
0
Total Ferrous
11
56
6,818
7,558
2,979
3,307
1
0
1
27
Aluminum
8
21
112
1,172
202
606
0
0
0
0
Copper-base
21
16
3,220
1,326
1 ,113
531
0
0
0
0
Zinc
2
11
59
157
31
215
0
0
0
0
Magnesium
0
0
0
0
0
0
0
0
0
0
Total Nonferrous
31
51
3,692
2,956
1,679
1,353
0
0
0
0
Grand Total
76
107
10,510
10,511
1,658
1,660
1
0
1
27
-------
TABLE V-3
COMPLIANCE COSTS AND ECONOMIC IMPACTS FOUNDRY INDUSTRY
(Option 3 Fecycle/Lime Addition/Settle/Filtration)
Number of
Discharging
Foundries
(in
Compliance Costs
thousands of 1983 dollars)
Closures
Capital
Investment
Annual Costs
Number of Foundries
Numhpr Af
Direct
Indirect
Direct
Indirect
Direct
Indirect
Direct
Indirect
Total
liUUUCI VI
Employees
Total
Gray Iron
91
145
15,702
22,152
7,099
9,892
3
6
9
243
Duotile Iron
27
25
7,450
2,799
3,311
1,245
0
1
1
27
Malleable Iron
21
29
2,641
2,736
1,259
1,245
1
0
1
71
Steel
13
64
4,854
5,900
2,144
2,691
0
0
0
0
Total Ferrous
182
263
30,647
33,587
13,843
15,073
4
7
11
341
Aluminum
45
131
3,353
6,440
1,599
3,652
0
0
0
0
Copper-base
63
54
9,012
4,944
1,173
2,112
0
0
0
0
Zlno
9
49
24 2
1,828
163
1,012
0
0
0
0
Magnesium
2
2
63
68
30
26
1
1
2
46
Total Nonferrous
119
236
12,669
13,280
5,964
6,801
1
1
2
46
Grand Total
301
499
13,316
46,867
19,806
21,875
5
8
13
387
Jobber
Gray Iron
65
110
10,674
15,686
1,823
6,964
2
5
7
189
Duotile Iron
23
21
6,237
2,460
2,789
1,100
0
1
1
27 ^
Malleable Iron
15
23
1,962
2,228
932
1,015
1
0
1
71
Steel
35
53
1,096
1,798
1,829
2,185
0
0
0
0
Total Ferrous
138
207
22,969
25,172
10,372
11,265
3
6
9
287
Aluminum
37
107
2,893
5,183
1,357
2,965
0
0
0
0
Copper-base
42
38
5,493
3,526
2,515
1,514
0
0
0
0
Zinc
7
38
170
1,338
117
765
0
0
0
0
Magnesium
2
2
63
68
30
26
1
1
2
46
Total Nonferrous
88
185
8,619
10,114
4,020
5,270
1
1
2
46
Grand Total
227
392
31,588
35,295
14,390
16,535
4
7
11
333
Captive
Gray Iron
26
35
5,028
6,466
2,277
2,927
1
1
2
54
Duotile Iron
<1
4
1,213
339
552
145
0
0
0
0
Malleable Iron
6
6
680
508
327
230
0
0
0
0
Steel
8
11
758
1,102
315
506
0
0
0
0
Total Ferrous
44
56
7,678
6,415
3,171
3,809
1
1
2
54
Aluminum
8
24
460
1,257
242
687
0
0
0
0
Copper-base
21
16
3,519
1,418
1,658
598
0
0
0
0
Zinc
2
11
72
490
45
247
0
0
0
0
Magnesium
0
0
0
0
0
0
0
0
0
0
Total Nonferrous
31
51
4,051
3,165
1,945
1,532
0
0
0
0
Grand Total
76
107
11,729
11,580
5,416
5,341
1
1
2
54
-------
TABLE V-4
COMPLIANCE COSTS AMD ECONOMIC IMPACTS FOUNDRY INDUSTRY
(Option 4 Recycle/Lime Additlon/Settle/Flltratlon/Carbon Adsorption)
Numl
Disci
Foul
ier of
lAPfflnir
(in
Compliance Costs
thousands of 1983 dollars)
Closures
Jul qmII^
idries
Capital
Investment
Annual Costs
Number of Foundries
MumKaf ar
Direct
Indirect
Direct
Indirect
Direct
Indirect
Direct
Indirect
Total
numuer 01
Employees
Total
Grey Iron
91
145
17,816
25,041
7,945
11,450
5
13
18
486
Ductile Iron
27
25
8,171
3,111
3,621
1,434
0
1
1
27
Malleable Iron
21
29
2,931
2,994
1,376
1,398
1
1
2
142
Steel
43
64
5,455
6,381
2,352
2,891
0
0
0
0
Total Ferrous
182
263
34,373
37,526
15,295
17,172
6
15
21
655
Aluminum
15
131
4,086
8,052
1,929
4,452
0
0
0
0
Copper-base
63
54
9,583
5,700
4,460
2,487
0
0
0
0
Zlno
9
49
416
2,468
243
1,335
0
0
0
0
Magnesium
2
2
81
68
40
26
2
1
3
69
Total Nonferrous
119
239
14,166
16,289
6,673
8,299
2
1
3
69
Grand Total
301
499
48,538
53,815
21,968
25,471
8
16
24
724
Jobber
Gray Iron
65
110
12,129
17,766
5,424
8,135
4
11
15
405
Ductile Iron
23
21
6,846
2,737
3,028
1,272
0
1
1
27
Malleable Iron
15
23
2.176
2,435
1,018
1,133
1
1
2
142
Steel
35
53
4,572
5,209
1,993
2,355
0
0
0
0
Total Ferrous
138
207
25,723
28,148
11,464
12,894
5
13
18
574
Aluminum
37
107
3,522
6,519
1,640
3,629
0
0
0
0
Copper-base
42
38
5,895
4,084
2,716
1,792
0
0
0
0
Zinc
7
38
299
1,826
176
1,012
0
0
0
0
Magnesium
2
2
81
68
40
26
2
1
3
69
Total Nonferrou9
88
185
9,797
12,497
4,573
6,459
2
1
3
69
Grand Total
227
392
35,520
40,645
16,036
19,353
7
14
21
643
Captive
Gray Iron
26
35
5,686
7,274
2,522
3,315
1
2
3
81
Ductile Iron
I)
4
1,325
374
592
162
0
0
0
0
Malleable Iron
6
6
755
558
358
265
0
0
0
0
Steel
8
11
883
1,171
359
536
0
0
0
0
Total Ferrous
44
56
8,649
9,378
3,831
4,278
1
2
3
81
Aluminum
8
24
563
1,534
289
822
0
0
0
0
Copper-base
21
16
3,688
1,616
1,744
695
0
0
0
0
Zinc
2
11
117
643
67
323
0
0
0
0
Magnesium
0
0
0
0
0
0
0
0
0
0
Total Nonferrous
31
51
4,368
3,792
2,100
1,840
0
0
0
0
Grand Total
76
107
13,017
13,170
5,931
6,118
1
2
3
81
-------
associated Job loss of 724 employees represents 0.50 percent of the
employment of discharging foundries.
In complying with the regulations, the estimated 800 discharging
foundries would incur capital costs ranging from $43.2 million under
Option 1 to $102. 4 million under Option 4. Total annual costs
including operating costs, interest, and depreciation, would range from
$16.0 million under Option 1 to $47 million under Option 4. Aggregate
Impacts at all four levels are:
Potential
Total
Total
Potential
Number of
Capital Cost
Annual Cost
Job
Option
Closures
($ Thousands)
($ Thousands)
Loss
1
4
43,200
16,220
100
2
7
81,179
36,008
181
3
13
90,183
41,681
387
4
24
102,353
47,439
724
2. Other Economic Impacts
Although EPA has used plant closures and employment loss as the
primary measure of economic impacts, other potential adverse impacts
have also been examined. These include:
potential price impacts
potential production Impacts
potential balance of trade impacts
potential community effects
None of these four potential impacts is expected to be major. In
general, potential price Increases are less than 0.5 percent, and EPA
expects that competition will preclude any price increase. The small
number of closures, limited to small foundries, is expected to have a
minor effect on the production capacity of the foundry industry. Under
the selected options, five gray iron and one ductile iron foundries are
listed as potential closures. These six closures account for only about
0.2 percent of total gray and ductile iron production. This regulation
will have negligible impact on the U.S. balance of trade. Because of
the low number of potential closures, EPA believes community effects
will be small and widely scattered.
C. POTENTIAL CLOSURES AND EMPLOYMENT LOSSES FOR INDIVIDUAL METALS
Although overall impacts are very low, they are not uniformly spread
over all metals and size categories. This section discusses the impacts
on each metal in more detail. Closure estimates are made assuming that
2A11 costs in this chapter are in 1983 dollars, unless otherwise noted.
V-7
-------
all plants incur the incremental cost of segregating noncontact cooling
water. The estimates of costs are based on the assumption that only 30
percent of plants must segregate noncontact cooling water.
1. Potential Impacts on Gray Iron Foundries
As shown in Tables V-5 through V-8, estimated annual costs of
treatment for gray iron range from ,752 thousand at Option 1 to
$19,396 thousand at Option 1. These costs could lead to closures of
foundries with fewer than 50 employees:
Potential
Total
Total
Number of
Capital Cost
Annual Cost
Option
Closures
($ Thousands)
($ Thousands)
1
2
13,^20
1,752
2
5
33,671
11,613
3
9
37,851
16,991
1
18
12,857
19,396
The potential closures, which result from failure of the return
on assets and Beaver's ratio tests, are all expected to occur in
foundries with 10 to 19 employees, distributed proportionately between
direct and indirect dischargers. Except for four potential closures of
indirect dischargers treating only dust collection effluent at Option 1,
all potential closures are for foundries treating effluent from melting
furnace scrubbers. As expected, almost all of the potential closures
are for firms in the lower quartile of financial health. In all cases
of closures, annual costs were greater than 3.2 percent of sales.
2. Potential Impacts on Ductile Iron Foundries
As shown in Tables V-9 through V-12, the annual cost to treat
the effluent from ductile iron foundries ranges from $1.5 million at
Option 1 to $5.1 million at Option 4. Assuming that all foundries incur
the cost to treat noncontact cooling water, one foundry with 10 to 19
employees could potentially close at Options 3 and 1. Overall costs and
impacts are:
Potential
Total
Total
Number of
Capital Cost
Annual Cost
Option
Closures
($ Thousands)
($ Thousands)
1
0
1,203
1,171
2
0
9,021
3,897
3
1
10,219
1,586
1
1
11,282
5,055
V-8
-------
TABLE V-5
COMPLIANCE COSTS AND ECONOMIC IMPACTS GRAY IRON
(Option 1 Recycle/Simple Settle)
Number of
Discharging
Foundries
Compliance Costs
(in thousands of 1983 dollars)
Closures
Capital Investment
Annual Costs
Number of Foundries
Number of
Employees
Direct
Indirect
Direct
Indirect
Direct
Indirect
Direct
Indirect
Total
All Foundries
Fewer than 10
0
2
0
80
0
29
0
0
0
0
10-19
11
38
791
1,128
273
511
2
0
2
51
50-99
11
27
396
901
135
309
0
0
0
0
100-2149
32
18
1,111
3,039
191
1,029
0
0
0
0
250 or more
11
_20
2.028
3.310
739
1,232
0
0
ฃ
_0
Total
91
115
1,662
8,758
1,610
3,112
2
0
2
51
Jobber Foundries
Fewer than 10
0
2
0
80
0
29
0
0
0
0
10-19
10
32
510
1,198
186
131
1
0
1
27
50-99
11
21
316
667
118
228
0
0
0
0
100-219
25
37
1,112
2,311
380
791
0
0
0
0
250 or more
21
18
1,275
1,857
170
685
0
0.
0
_0
Total
65
110
3,271
6,115
1,151
2,167
1
0
1
27
Captive Foundries
Fewer than 10
0
0
0
0
0
0
0
0
0
0
10-19
1
6
251
231
88
83
1
0
1
27
50-99
3
6
50
231
17
81
0
0
0
0
100-219
7
11
332
695
113
235
0
0
0
0
250 or more
iS
11
75?
1,153
269
517
0
0
0
_0
Total
26
35
1,389
2,613
186
915
1
0
1
27
Note: Numbers may not add up to totals due to rounding.
-------
TABLE V-6
COMPLIANCE COSTS AMD ECONOMIC IMPACTS GRAY IRON
(Option 2 Reoyole/Llme Addition/Settle)
<
I
o
Number of
Discharging
Foundries
Compliance Costs
(in thousands of 1983 dollars)
Closures
Capital
Investment
Annual Costs
Number of Foundries
Number of
Employees
Direct
Indirect
Direct
Indirect
Direct
Indirect
Direct
Indirect
Total
All Foundries
Fewer than 10
0
2
0
91
0
36
0
0
0
0
10-19
11
38
1,182
2,861
112
1,086
3
2
5
135
50-99
11
27
990
2,029
366
761
0
0
0
0
100-219
32
48
3,125
5,538
1,336
2,320
0
0
0
0
250 or more
11
_3ฐ
8.602
9.217
3.9*5
",31?
a
0
0
0
Total
91
115
13,899
19,772
6,091
8,522
3
2
5
135
Jobber Foundries
Fewer than 10
0
2
0
91
0
36
0
0
0
0
10-19
10
32
837
2,111
315
915
2
2
1
108
50-99
11
21
787
1,588
291
581
0
0
0
0
100-219
25
37
2,380
1,273
1,017
1,791
0
0
0
0
250 or more
12
18
5,118
5,611
2,502
2.661
0
0
0
0
Total
65
110
9,121
13,980
1,125
5,987
2
2
1
108
Captive Foundries
Fewer than 10
0
0
0
0
0
0
0
0
0
0
10-19
1
6
315
150
127
170
1
0
1
27
50-99
3
6
203
171
77
177
0
0
0
0
100-219
7
11
715
1,265
319
529
0
0
0
0
250 or more
12
12
3.186
3.606
1.113
1.659
2
0
0
_0
Total
26
35
1,178
5,792
1,966
2,535
1
0
1
27
Note: Numbers may not add up to totals due to rounding.
-------
TABLE V-7
COMPLIANCE COSTS AND ECONOMIC IMPACTS GRAY IRON
(Option 3 Recycle/Lime Addition/Settle/Filtration)
Compliance Costs
Number of
(In
thousands of 1983 dollars)
Closures
T)1 flnhnr
-------
TABLE V-8
COMPLIAHCE POSTS AMD ECONOMIC IMPACTS GRAY IRON
(Option 4 Recycle/Lime Addltlon/Settle/Flltratlon/Carbon Adsorption)
Number of
Discharging
Foundries
Coapllanoe Costs
(In thousands of 1983 dollars)
Closures
Capital Investment
Annual Costs
Number of Foundries
Number of
Employees
Direct
Indirect
Direct
Indirect
Direct
Indirect
Direct
Indirect
Total
All Foundries
Fewer than 10
0
2
0
138
0
77
0
0
0
0
10-19
11
38
1,582
3,735
613
1,691
5
13
18
186
50-99
11
27
1,298
2,615
518
1,136
0
0
0
0
100-219
32
18
1,121
7,053
1,900
3,198
0
0
0
0
250 or more
11
_20
10.811
11.500
1.856
5,3"5
0
0
_0
0
Total
91
115
17,816
25,011
7,916
11,150
5
13
18
186
Jobber Foundries
Fewer than 10
0
2
0
138
0
77
0
0
0
0
10-19
10
32
1,128
3,118
159
1,127
1
11
15
105
50-99
11
21
1,035
2,011
136
873
0
0
0
0
100-219
25
37
3,151
5,118
1,152
2,172
0
0
0
0
250 or more
21
18
6.815
7.02 2
3.077
3.286
J)
0
0
0
Total
65
110
12,129
17,766
5,121
8,135
1
11
15
105
Captive Foundries
Fewer than 10
0
0
0
0
0
0
0
0
0
0
10-19
1
6
151
587
181
268
1
2
3
81
50-99
3
6
263
601
112
726
0
0
0
0
100-219
7
11
973
1,605
117
727
0
0
0
0
250 or more
12
12
3,??6
1.178
1.779
2,058
0
0
j)
0
Total
26
35
5,685
7,271
2,522
3,315
1
2
3
81
Note: Numbers may not add up to totals due to rounding.
-------
TABLE V-9
COMPLIANCE COSTS AND ECONOHIC IMPACTS DUCTILE IRON
(Option 1 Reoyole/Stmple Settle)
Number of
Discharging
Foundries
(In
Compliance Costs
thousands of 1983 dollars)
Closures
Capital
Investment
Annuel Costs
Number of Foundries
Number of
Employees
Direct
Indirect
Direct
Indirect
Direct
Indirect
Direct
Indirect
Total
All Foundries
Fewer than 10
0
0
0
0
0
0
0
0
0
0
10-19
0
9
0
181
0
56
0
0
0
0
50-99
0
3
0
168
0
57
0
0
0
0
100-249
16
11
1,163
561
512
198
0
0
0
0
250 or more
11
2
1,553
258
516
102
0
0
0
0
Total
27
25
3,036
1,167
1,058
113
0
0
0
0
Jobber Foundries
Fewer than 10
0
0
0
0
0
0
0
0
0
0
10-19
0
6
0
112
0
11
0
0
0
0
50-99
0
3
0
166
0
57
0
0
0
0
100-219
11
10
1,150
520
397
183
0
0
0
0
250 or more
_!
_2
1.316
258
168
102
0
0
0
0
Total
23
21
2,197
1,087
861
386
0
0
0
0
Captive Foundries
Fewer than 10
0
0
0
0
0
0
0
0
c
0
10-19
0
3
0
38
0
12
0
0
0
0
50-99
0
0
0
0
0
0
0
0
0
0
100-219
2
1
333
11
115
15
0
0
0
0
250 or more
2
ฃ
206
_0
_0
0
0
0
0
Total
1
1
539
80
193
27
0
0
0
0
Note: Numbers may not add up to totals due to rounding.
-------
TABLE V-10
COMPLIANCE COSTS AMD ECOHOMIC IMPACTS DUCTILE IRON
(Option 2 Reoyole/Lioe Addition/Settle)
<
I
*
-O
Number of
Discharging
Foundries
(in
Compliance Costs
thousands of 1983 dollars)
Closures
Capital Investment
Annual Coats
Number of Foundries
Number of
Employees
Direct
Indirect
Direct
Indlreat
Direct
Indirect
Direct
Indirect
Total
All Foundries
Fewer than 10
0
0
0
0
0
0
0
0
0
0
10-19
0
9
0
529
0
196
0
0
0
0
50-99
0
3
0
189
0
68
0
0
0
0
100-219
16
11
2,111
1,259
993
566
0
0
0
0
250 or more
_2
1.131
196
1.813
ill
0
0
0
0
Total
27
25
6,516
2,175
2,836
1,061
0
0
0
0
Jobber Foundries
Fewer than 10
0
0
0
0
0
0
0
0
0
0
10-19
0
6
0
312
0
121
0
0
0
0
50-99
0
3
0
189
0
68
0
0
0
0
100-219
11
10
2,073
1,112
869
511
0
0
0
0
250 or mors
_!
2
3.110
198
1."??
221
0
0
0
0
Total
23
21
5,181
2,171
2,367
935
0
0
0
0
Captln Foundries
Fewer than 10
0
0
0
0
0
0
0
0
0
0
10-19
0
3
0
187
0
72
0
0
0
0
50-99
0
0
0
0
0
0
0
0
0
0
100-219
2
1
311
118
121
55
0
0
0
0
250 or more
2
0
721
0
311
0
0
0
0
0
Total
1
1
1,062
301
168
127
0
0
0
0
Note: Numbers may not add up to totals due to rounding.
-------
TABLE V-11
COMPLIANCE COSTS AMD ECONOMIC IMPACTS DUCTILE IRON
(Option 3 Recycle/Lime Addition/Settle/Filtration)
<3
I
w
(Ji
Compliance Costs
Number of
(In thousands of 1983 dollars)
Closures
n1A^hAPffinff
VlDvllQl KAII^
Foundries
Capital
Investment
Annual Costs
Number of Foundries
Direct
Indirect
Direct
Indirect
Direct
Indirect
Direct
Indirect
Total
nuouci ui
Employees
All Foundries
Fewer than 10
0
0
0
0
0
0
0
0
0
0
10-49
0
9
0
559
0
212
0
1
1
27
50-99
0
3
0
225
0
87
0
0
0
0
100-249
16
11
2.893
1,467
1,249
684
0
0
0
0
250 or more
_n
2
4,557
549
2.092
262
0
0
0
_0
Total
27
25
7,450
2,799
3,341
1,245
0
1
1
27
Jobber Foundries
Fewer than 10
0
0
0
0
0
0
0
0
0
0
10-149
0
6
0
359
0
133
0
1
1
27
50-99
0
3
0
225
0
87
0
0
0
0
100-249
14
10
2,477
1,328
1,086
617
0
0
0
0
250 or more
_2
_2
3.760
549
1,703
262
0
ฃ
0
_0
Total
23
21
6,237
2,460
2.789
1,099
0
1
1
27
Captive Foundries
Pewer than 10
0
0
0
0
0
0
0
0
0
0
10-49
0
3
0
200
0
79
0
0
0
0
50-99
0
0
0
0
0
0
0
0
0
0
100-249
2
1
416
140
163
67
0
0
0
0
250 or Dore
2
2
7?7
0
389
0
0
0
0
0
Total
4
4
1,213
339
552
145
0
0
0
0
Note: Numbers may not add up to totals due to rounding.
-------
TABLE V-12
COMPLIANCE COSTS AND ECOHOHIC IMPACTS DUCTILE IROH
(Option 'I Recycle/Line Addition/Settle/Filtration/Carbon Adsorption)
<
I
tฆ
On
Compliance Costs
Number of
(in
thousands of 1983 dollars)
Closures
Diirhnrfflnff
VlsViilU glllg
Foundries
Capital
Investment
Annual Costs
Number of Foundries
NurnhAr* nf
Direct
Indirect
Direct
Indirect
Direct
Indirect
Direct
Indirect
Total
nuuiwUi vl
Employees
All Foundries
Fewer than 10
0
0
0
0
0
0
0
0
0
0
10-19
0
9
0
615
0
236
0
1
1
27
50-99
0
3
0
266
0
130
0
0
0
0
100-219
16
11
3,125
1,624
1,389
785
0
0
0
0
250 or more
_n
_2
5.017
606
2,231
285
0
0
0
_0
Total
27
25
8,171
3,111
3,621
1,434
0
1
1
27
Jobber Foundries
Fewer than 10
0
0
0
0
0
0
0
0
0
0
10-49
0
6
0
392
0
147
0
1
1
27
50-99
0
3
0
266
0
130
0
0
0
0
100-249
11
10
2,660
1,174
1,210
710
0
0
0
0
250 or store
_i
_2
1.167
606
1,819
285
0
0
0
_0
Total
23
21
6,816
2,737
3,028
1,272
0
1
1
27
Captive Foundries
Fewer than 10
0
0
0
0
0
0
0
0
0
0
10-1)9
0
3
0
224
0
89
0
0
0
0
50-99
0
0
0
0
0
0
0
0
0
0
100-249
2
1
416
150
180
73
0
0
0
0
250 or more
2
0
880
0
m
0
0
0
JD
0
Total
1
1
1,325
371
592
162
0
0
0
0
Note: Numbers may not add up to totals due to rounding.
-------
A closer review shows that the two closures occur for foundries
treating effluent from dust collection. Closures are not predicted for
the 70 percent of foundries that will not incur the cost to segregate
noncontact cooling water. Further the two potential closures are based
on an estimated one jobber and one captive closing. In fact, the
potential closures occur for lower quartile firms only. Because there
are only four plants incurring the cost to treat dust collection wastes,
only one plant is a potential closure; the second "closure" is a result
of disaggregation into Jobber or captive status and subsequent rounding
of partial closures. Taking all factors into account, EPA estimates
only one ductile iron closure as a result of the regulation:
Closures shown
Actual estimated closures if all
plants segregate noncontact
cooling water
Fraction of plants incurring cost
to segregate noncontact cooling
water
Actual estimated closures
2.0
1.0 (25$ of 4 plants)
0.3
0.3
The plant expected to fail has insufficient return on assets and cash
flow to total debt.
3. Potential Impacts on Malleable Iron Foundries
As shown in Tables V-13 through V16, the annual cost to treat
discharges from the 50 discharging malleable iron foundries range from
$0.6 million under Option 1 to $2.8 million under Option 4. Assuming
that all foundries incur the incremental cost to segregate noncontact
cooling water, EPA estimates that there would be one potential closure
under Option 3 and two potential closures under Option 4. Overall costs
and impacts are as follows:
Potential
Potential
Total
Total
Number of
Employment
Capital Cost
Annual Cost
Option
Closures
Lost
($ Thousands)
($ Thousands)
1
0
0
1,637
578
2
0
0
4,819
2,181
3
1
71
5,377
2,505
4
2
142
5,925
2,774
All potential closures occur in the 50 to 99 employment size
category and are due to failure of the return on assets and debt to
assets tests. A close review of the data shows that the failures occur
even though the total annual cost represents less that 1 percent of
sales for the foundries considered potential closures.
V-17
-------
TABLE V-13
COMPLIANCE COSTS AND ECONOMIC IMPACTS MALLEABLE IRON
(Option 1 Reoycle/Slmple Settle)
<
I
iฆ
oo
Number of
Discharging
Foundries
(In
Compliance Costa
thousands of 1963 dollara)
Closures
Capital
tnvestoent
Annual Costa
Number of Foundries
Number or
Direct
Indirect
Direct
Indireot
Direct
Indireot
Direct
Indirect
Total
Employees
All Foundries
Fewer than 10
0
0
0
0
0
0
0
0
0
0
10-49
0
0
0
0
0
0
0
0
0
0
50-99
3
5
33
161
13
56
0
0
0
0
100-249
11
22
46
600
15
209
0
0
0
0
250 or more
_7
_2
494
30?
m
107
0
0
ฃ
0
Total
21
29
573
1,064
205
373
0
0
0
0
Jobber Foundries
Fewer than 10
0
0
0
0
0
0
0
0
0
0
10-49
0
0
0
0
0
0
0
0
0
0
50-99
2
4
22
125
9
44
0
0
0
0
100-249
B
17
31
449
10
157
0
0
0
0
250 or more
_2
Hi
m.
131
m.
0
0
ฃ
0
Total
15
23
431
878
153
308
0
0
0
0
Captive Foundries
Fewer than 10
0
0
0
0
0
0
0
0
0
0
10-49
0
0
0
0
0
0
0
0
0
0
50-99
1
1
11
36
4
12
0
0
0
0
100-249
3
5
15
150
5
52
0
0
0
0
250 or more
2
0
116
0
42
_0
0
ฃ
0
0
Total
6
6
143
186
51
64
0
0
0
0
Note: Numbers may not add up to totals due to rounding.
-------
TABLE V-11
COMPLIANCE COSTS AND ECONOMIC IMPACTS MALLEABLE IRON
(Option 2 Recyole/Llme Addition/Settle)
<
I
VO
Number of
Discharging
Foundries
Compliance Coats
(In thousands of 1983 dollars)
Closures
Capital Investment
Annual Costs
Number of Foundries
NumhAr ftf
Direct
Indirect
Direct
Indirect
Direct
Indirect
Direct
Indirect
Total
UlUvvl vl
Employees
All Foundries
Fewer than 10
0
0
0
0
0
0
0
0
0
0
10-19
0
0
0
0
0
0
0
0
0
0
50-99
3
5
169
163
65
69
0
0
0
0
100-219
11
22
715
1,638
361
735
0
0
0
0
250 or more
_2
611
681
269
0
0
0
0
Total
21
29
2,387
2,132
1,109
1,072
0
0
0
0
Jobber Foundries
Fewer than 10
0
0
0
0
0
0
0
0
0
0
10-19
0
0
0
0
0
0
0
0
0
0
50-99
2
1
112
113
13
51
0
0
0
0
100-219
e
17
513
1,232
261
553
0
0
0
0
250 or more
_ฃ
_2
1,11?
611
515
269
0
0
J)
0
Total
15
23
1,771
1,986
822
876
0
0
0
0
Captive Foundries
Fewer than 10
0
0
0
0
0
0
0
0
0
0
10-19
0
0
0
0
O
0
0
0
0
0
50-99
i
1
56
11
22
11
0
0
0
0
100-219
3
5
203
106
97
181
0
0
0
0
250 or more
2
0
0
169
0
0
0
0
0
Total
6
6
613 j
416
288
196
0
0
0
0
Note: Numbers may not add up to totals due to rounding.
-------
TABLE V-15
COMPLIANCE COSTS AND ECONOMIC IMPACTS MALLEABLE IRON
(Option 3 Recycle/Lime Addltlon/Settle/Flltratlon)
<
I
to
O
Compliance Costs
Number of
(in thousands of
1983 dollars)
Closures
Q^harff 1 ncr
Foundries
Capital Investment
Annual
Costs
Number of Foundries
Number of
Direct
Indirect
Direct
Indlreot
Direct
Indirect
Direct
Indlreot
Total
Employees
All Foundries
Fewer than 10
0
0
0
0
0
0
0
0
0
0
10-19
0
0
0
0
0
0
0
0
0
0
50-99
3
5
189
217
77
87
1
0
1
71
100-219
11
22
836
1,855
115
858
0
0
0
0
250 or more
_2
1,616
665
768
301
0
0
0
0
Total
21
29
2,611
2,736
1,259
1,216
1
0
1
71
Jobber Foundries
-
Fewer than 10
0
0
0
0
0
0
0
0
0
0
10-H9
0
0
0
0
0
0
0
0
0
0
50-99
2
1
126
168
51
68
1
0
1
71
100-219
8
17
609
1,395
303
617
0
0
0
0
250 or more
_5
_2
1.227
665
m
301
0
0
0
_0
Total
15
23
1,962
2,228
93 2
1,015
1
0
1
71
Captive Foundries
Fewer than 10
0
0
0
0
0
0
0
0
0
0
10-49
0
0
0
0
0
0
0
0
0
0
50-99
1
1
63
18
26
18
0
0
0
0
100-219
3
5
227
159
112
212
0
0
0
0
250 or more
2
0
389
0
190
0
0
0
0
0
Total
6
6
680
508
327
230
0
0
0
0
Note: Numbers may not add up to totals due to rounding.
-------
TABLE V-16
COMPLIANCE COSTS AMD ECONOMIC IMPACTS MALLEABLE IRON
(Option 1 Recycle/Lime Addltion/Settle/Flltratlon/Carbon Adsorption)
Number of
Discharging
Foundries
Compliance Costs
(in thousands of 1983 dollars)
Closures
Capital Investment
Annual Costs
Number of Foundries
Number of
Employees
Direct
Indirect
Direct
Indlreot
Direct
Indirect
Direct
Indirect
Total
All Foundries
Fewer than 10
0
0
0
0
0
0
0
0
0
0
10-19
0
0
0
0
0
0
0
0
0
0
50-99
3
5
213
210
87
110
1
1
2
112
100-219
11
22
925
2,015
172
959
0
0
0
0
250 or more
_2
1.792
739
817
32?
0
0
0
0
Total
21
29
2,931
2,991
1,376
1,398
1
1
2
112
Jobber Foundries
-
Fewer than 10
0
0
0
0
0
0
0
0
0
0
10-19
0
0
0
0
0
0
0
0
0
0
50-99
2
1
112
183
58
83
1
1
2
112
100-219
8
17
671
1,511
315
721
0
0
0
0
250 or more
_5
_2
1.36ฐ
73?
615
32?
0
0
2
0
Total
15
23
2,176
2,135
1,018
1.133
1
1
2
112
Captive Foundries
Fewer than 10
0
0
0
0
0
0
0
0
0
0
10-19
0
0
0
0
0
0
0
0
0
0
50-99
1
1
71
57
29
27
0
0
0
0
100-219
3
5
251
501
127
238
0
0
0
0
250 or more
2
0
122
0
202
0
0
0
0
0
Total
6
6
755
558
358
265
0
0
0
0
Note: Numbers may not add up to totals due to rounding.
-------
4. Potential Impacts On Steel Foundries
As shown in Tables V-17 through V-20, potential total annual
costs for the 107 directly and indirectly discharging steel foundries
range from about $1.5 million under Option 1 to $5.2 million under
Option 4. In all cases, the annual costs represent less than 1 percent
of sales for all individual combinations of discharging processes.
Aggregate costs at each Option are:
Potential
Potential
Total
Total
Number of
Employment
Capital Cost
Annual Cost
Option
Closures
Lost
($ Thousands)
($ Thousands)
1
0
0
4,056
1,475
2
0
0
9,786
4,219
3
0
0
10,754
4,835
4
0
0
11,836
I 5,244
5. Potential Impacts on Aluminum Foundries
As shown in Tables V-21 through V-24, potential aggregate annual
costs for the 176 directly and indirectly discharging aluminum foundries
range from $2.35 million under Option 1 to $6.4 million under Option
4. The maximum annual cost for any model plant is only about 1.3
percent of sales. EPA does not anticipate any aluminum foundry closures
as a result of the regulation. Total potential costs under each Option
are:
Option
Potential
Number of
Closures
Potential
Employment
Lost
Total
Capital Cost
($ Thousands)
Total
Annual Cost
($ Thousands)
1
2
3
4
0
0
0
0
0
0
0
0
6,151
9,045
9,793
12,138
2,349
4,567
5,251
6,381
6.
Potential Impacts on Copper-Base Foundries
As shown in Table V-25 through V-28, potential annual costs for
treating the discharge from the 117 directly and indirectly discharging
foundries range from $5.0 million under Option 1 to $6.9 million under
Option 4. Under Option 4, the incremental cost for treating the
discharge from direct chill casting reaches a peak of 2.92 percent for
direct dischargers with fewer than 10 employees. EPA expects no
potential closures as a result of this regulation. Potential total
costs under each Option are:
V-22
-------
TABLE V-17
COMPLIANCE COSTS AND ECONOMIC IMPACTS STEEL
(Option 1 Recycle/Simple Settle)
<
I
N>
OJ
Number of
Discharging
Foundries
Compliance Costs
(in thousands of 1983 dollars)
Closures
Capital
Investment
Annual Costs
Number of Foundries
Number of
Employees
Direct
Indirect
Direct
Indirect
Direct
Indirect
Direct
Indirect
Total
All Poundrles
Fewer than 10
2
0
57
0
20
0
0
0
0
0
10-19
0
10
0
97
0
12
0
0
0
0
50-99
11
21
279
667
107
239
0
0
0
0
100-219
19
19
756
525
271
206
0
0
0
0
250 or n>ore
11
_11
611
1.061
HI
371
ฃ
0
2
0
Total
13
61
1,706
2,350
6l1
861
0
0
0
0
Jobber Foundries
Fewer than 10
1
0
29
0
10
0
0
0
0
0
10-19
0
8
0
77
0
33
0
0
0
0
50-99
9
18
235
566
90
203
0
0
0
0
100-219
16
16
633
136
230
172
0
0
0
0
250 or more
_2
21
526
791
182
277
0
0
0
0
Total
35
53
1,123
1,871
513
685
0
0
0
0
Captive Foundries
Fewer than 10
1
0
29
0
10
0
0
0
0
0
10-19
0
2
0
19
0
8
0
0
0
0
50-99
2
3
11
101
17
36
0
0
0
0
100-219
3
3
123
89
11
31
0
0
0
0
250 or more
2
_2
87
269
_2I
0
0
0
0
Total
8
ii
283
178
102
176
0
0
0
0
Note: Numbers may not add up to totals due to rounding.
-------
TABLE V-18
COMPLIANCE COSTS AND ECONOMIC IMPACTS STEEL
(Option 2 Recycle/Lime Addition/Settle}
<
I
ho
-P-
Number of
Discharging
Foundries
Compliance Costs
(in thousands of 1983 dollars)
Closures
Capital Investment
Annual Costs
Number of Foundries
Number of
Direct
Indirect
Direct
Indirect
Dlreot
Indireot
Dlreot
Indireot
Total
Employees
All Foundries
Fewer than 10
2
0
70
0
21
0
0
0
0
0
10-19
0
10
0
283
0
111
0
0
0
0
50-99
11
21
325
1,102
111
129
0
0
0
0
100-219
19
19
1,589
1,363
668
639
0
0
0
0
250 or more
_n
J1
2,393
2.660
1,01?
1.188
0
0
0
0
Total
13
61
1.377
5,109
1,819
2,370
0
0
0
0
Jobber Foundrios
Fewer than 10
1
0
35
0
12
0
0
0
0
0
10-19
0
8
0
227
0
91
0
0
0
0
50-99
9
18
272
918
121
370
0
0
0
0
100-219
16
16
1,321
1,133
557
533
0
0
0
0
250 or more
_2
11
2.085
2.066
902
?2 7
0_
2
0
0
Total
35
53
3,712
1,393
1,592
1,920
0
0
0
0
Captive Foundries
Fewer than 10
1
0
35
0
12
0
0
0
0
0
10-19
0
2
0
57
0
23
0
0
0
0
50-99
2
3
51
151
23
59
0
0
0
0
100-219
3
3
268
230
112
107
0
0
0
0
250 or more
2
_i
308
575
111
261
0
0
0
0
Total
8
11
665
1,015
257
150
0
0
0
0
Note: Numbers may not add up to totals due to rounding.
-------
TABLE V-19
COMPLIANCE COSTS AND ECONOMIC IMPACTS STEEL
(Option 3 Recycle/Lime Additlon/Settle/Filtratlon)
<
I
to
Cn
Number of
Discharging
Foundries
(in
Compliance Costs
thousands of 1983 dollars)
Closures
Capital
Investment
Annual Costs
Number of Foundries
Nunhflr nf
Direct
Indirect
Direct
Indirect
Direct
Indirect
Direct
Indirect
Total
H UlUI/vl VI
Employees
All Foundries
Fewer than 10
2
0
71
0
31
0
0
0
0
0
10-19
0
10
0
326
0
110
0
0
0
0
50-99
11
21
316
1,166
168
188
0
0
0
0
100-2149
19
19
1 ,817
1,520
798
731
0
0
0
0
250 or more
11
Jl
2.617
2,890
1.116
1132?
0
0
0
0
Total
13
61
1,851
5,900
2,111
2,691
0
0
0
0
Jobber Foundries
Fewer than 10
1
0
37
0
16
0
0
0
0
0
10-19
0
8
0
260
0
112
0
0
0
0
50-99
9
18
289
1,001
111
121
0
0
0
0
100-219
16
16
1,508
1,263
663
611
0
0
0
0
250 or more
_2
11
2.261
2.271
1.009
1.011
0
0
0
0
Total
35
53
1,096
1,798
1,829
2,185
0
0
0
0
Captive Foundries
Fewer than 10
1
0
37
0
16
0
0
0
0
0
10-19
0
2
0
65
0
28
0
0
0
0
50-99
2
3
57
162
27
67
0
0
0
0
100-219
3
3
310
257
135
122
0
0
0
0
250 or more
2
_1
i51
618
JL3I
289
0
ฃ
J)
0
Total
e
11
758
1,102
315
506
0
0
0
0
Note: Numbers may not add up to totals due to rounding.
-------
TABLE V-20
COMPLIANCE COSTS AND ECONOMIC IMPACTS STEEL
(Option <4 Recycle/Lime Addltlon/Settle/Filtratlon/Carbon Adsorption)
Compliance Costs
Number of
(in
thousands of 1983 dollars)
Closures
D1srhflpclnff
VXwVfllQI pAI
Foundries
Capital Investment
Annual Costs
Number of Foundries
Number of
Direct
Indirect
Direct
Indirect
Direct
Indirect
Direct
Indirect
Total
Employees
All Foundries
Fewer than 10
2
0
73
0
32
0
0
0
0
0
10-19
0
10
0
375
0
166
0
0
0
0
50-99
11
21
311
1,230
168
518
0
0
0
0
100-219
19
19
2,076
1,620
909
800
0
0
0
0
250 or more
_n
ii
2.961
3,155
1.211
1.106
2
0
0
0
Total
13
61
5,155
6,381
2,353
2,891
0
0
0
0
Jobber Foundries
Fewer than 10
1
0
37
0
16
0
0
0
0
0
10-19
0
8
0
300
0
133
0
0
0
0
50-99
9
18
287
1,061
111
118
0
0
0
0
100-219
16
16
1,711
1,316
751
667
0
0
0
0
250 or more
_2
11
2,535
2.502
1.086
1.107
ฃ
0
0
0
Total
35
53
1,572
5,209
1,993
2,355
0
0
0
0
Captive Foundries
Fewer than 10
1
0
37
0
16
0
0
0
0
0
10-19
0
2
0
75
0
33
0
0
0
0
50-99
2
3
57
170
27
71
0
0
0
0
100-219
3
3
363
271
158
131
0
0
0
0
250 or more
2
_2
126
65?
158
299
0
0
ฃ
0
Total
B
"
883
1,172
359
536
0
0
0
0
Note: Numbers may not add up to totals due to rounding.
-------
TABLE V-21
COMPLIANCE COSTS AND ECONOMIC IMPACTS ALUMINUM
(Option 1 Recycle/Simple Settle)
<
I
to
-j
Compliance Costs
Number of
(in
thousands of 1983 dollars)
Closures
DiflrhAro 1 riff
vxoviiui giiyj
Foundries
Capital
Investment
Annual Costs
Number of Foundries
Number of
Direct
Indirect
Direct
Indirect
Direct
Indirect
Direct
Indirect
Total
Employees
All Foundries
Fewer than 10
7
0
595
0
213
0
0
0
0
0
10-19
9
61
181
1,291
72
561
0
0
0
0
50-99
6
20
113
527
60
231
0
0
0
0
100-219
11
11
1,026
1,292
357
112
0
0
0
0
250 or more
_ฃ
9
609
175
218
161
0
0
0
0
Total
15
131
2,521
3,627
919
1,100
0
0
0
0
Jobber Foundries
Fewer than 10
6
0
500
0
205
0
0
0
0
0
10-19
7
19
115
1,029
57
117
0
0
0
0
50-99
5
17
103
172
53
195
0
0
0
0
100-219
12
31
932
991
323
310
0
0
0
0
250 or more
533
381
186
133
0
0
ฃ
ฃ
Total
37
107
2,213
2,876
821
1,111
0
0
0
0
Captive Foundries
Fewer than 10
1
0
95
0
38
0
0
0
0
0
10-19
2
12
36
265
11
111
0
0
0
0
50-99
1
3
10
95
7
39
0
0
0
0
100-219
2
7
91
301
33
102
0
0
0
0
250 or more
2
2
-Iฃ
_2ฐ
_3ฃ
_2ฐ
ฃ
ฃ
0
0
Total
8
21
311
751
125
286
0
0
0
0
Note: Numbers may not add up to totals due to rounding.
-------
TABLE V-22
COMPLIANCE COSTS AND ECONOMIC IMPACTS ALUMINUM
(Option 2 Recycle/Lime Addition/Settle)
<
I
ro
oo
Number of
Discharging
Foundries
(in
Compliance Costa
thousands of 1983 dollars)
Closures
Capital
Investment
Annual Costs
Number of Foundries
Number of
Employees
Direct
Indirect
Direct
Indirect
Direct
Indirect
Direct
Indireot
Total
All Foundries
Fewer than 10
7
0
681
0
311
0
0
0
0
0
10-19
9
61
318
2,115
183
1,135
0
0
0
0
50-99
6
20
172
1,102
105
603
0
0
0
0
100-219
11
11
1,100
2,068
120
1,171
0
0
0
0
250 or more
_i
_i
736
689
318
321
0
ฃ
0
0
Total
15
131
3,010
6,005
1,337
3,230
0
0
0
0
Jobber Foundries
Fewer than 10
6
0
575
0
262
0
0
0
0
0
10-19
7
19
277
1,710
115
903
0
0
0
0
50-99
5
17
151
9 22
86
506
0
0
0
0
100-219
12
31
1,000
1,638
383
951
0
0
0
0
250 or more
JL
1
621
562
259
263
0
()
0
0
Total
37
107
2,628
1,833
1,135
2,621
0
0
0
0
Captive Foundries
Fewer than 10
1
0
109
0
50
0
0
0
0
0
10-19
2
12
71
136
38
232
0
0
0
0
50-99
1
3
21
180
18
97
0
0
0
0
100-219
2
7
100
130
37
220
0
0
0
0
250 or more
2
_2
112
127
60
58
0
ฃ
0
0
Total
8
21
112
1,172
202
606
0
0
0
0
Notซ: Numbers may not add up to totals due to rounding.
-------
TABLE V-23
COMPLIANCE COSTS AHD ECONOMIC IMPACTS ALUMINUM
(Option 3 Recycle/Line Addltion/Settle/Flltratlon)
<
1
to
VO
Number of
Discharging
Foundries
(In
Compliance Coats
thousands of 1983 dollars)
Closures
Capital
Investment
Annual Costs
Number of Foundries
NitinKat* n f
Direct
Indirect
Direct
Indirect
Direct
Indirect
Direc t
Indirect
Total
nuioucr oi
Employees
All Foundries
Fewer than 10
7
0
734
0
386
0
0
0
0
0
10-49
9
61
370
2,322
205
1,273
0
0
0
0
50-99
6
20
205
1,177
133
688
0
0
0
0
100-249
14
41
1,197
2,207
491
1,330
0
0
0
0
250 or more
_2
846
734
384
361
0
0
0
0
Total
45
131
3,353
6,440
1,599
3,652
0
0
0
0
Jobber Pouodrlea
Fewer than 10
6
0
617
0
325
0
0
0
0
0
10-49
7
49
295
1,849
162
1,012
0
0
0
0
50-99
5
17
177
985
110
577
0
0
0
0
100-249
12
34
1,094
1,751
450
1,081
0
0
0
0
250 or more
_I
709
597
310
295
0
ฃ
0
ฃ
Total
37
107
2,893
5,183
1,357
2,965
0
0
0
0
Captive Foundries
Fewer than 10
1
0
117
0
61
0
0
0
0
0
10-49
2
12
76
473
42
261
0
0
0
0
50-99
1
3
27
192
23
110
0
0
0
0
100-249
2
7
103
456
41
250
0
0
0
0
250 or more
2
_2
211
'37
66
0
0
0
0
Total
8
24
460
1 ,257
24 2
687
0
0
0
0
Note: Numbers may not add up to totals due to rounding.
-------
TABLE V-24
COMPLIANCE POSTS AND ECONOMIC IMPACTS ALUMINUM
(Option *4 Recycle/Lime Addition/Settle/Flltration/Carbon Adsorption}
<5
I
LO
o
Number of
Discharging
Foundries
Compliance Costs
(In thousands of 1983 dollars)
Closures
Capital Investment
Annual Costs
Number of Foundries
Number of
Employees
Dlreot
indirect
Dlreot
Indirect
Direct
Indirect
Direct
Indirect
Total
All Foundries
Fewer than 10
7
0
860
0
452
0
0
0
0
0
10-19
9
61
453
2,963
248
1 ,606
0
0
0
0
50-99
6
20
306
1,494
179
847
0
0
0
0
100-249
14
41
1,384
2,712
575
1,574
0
0
0
0
250 or more
_2
1.082
883
474
424
0
ฃ
ฃ
0
Total
45
131
1,086
8,052
1,929
4,452
0
0
0
0
Jobber Foundries
Fewer than 10
6
0
724
0
381
0
0
0
0
0
10-49
7
49
360
2,371
197
1,284
0
0
0
0
50-99
5
17
266
1,250
151
711
0
0
0
0
100-219
12
34
1,262
2,171
526
1,285
0
0
0
0
250 or more
_L
_1
910
727
385
350
ฃ
ฃ
0
0
Total
37
107
3,522
6,519
1,640
3,630
0
0
0
0
Captive Foundries
Fewer than 10
1
0
136
0
71
0
0
0
0
0
10-49
2
12
93
592
52
322
0
0
0
0
50-99
1
3
40
244
29
136
0
0
0
0
100-249
2
7
122
542
49
290
0
0
0
0
250 or more
2
2
172
157
89
_Ei
0
0
ฃ
0
Total
8
24
563
1,534
289
822
0
0
0
0
Note: Numbers nay not add up to totals due to rounding.
-------
TABLE V-25
COMPLIANCE COSTS AMD ECONOMIC IMPACTS COPPER
(Option 1 Recycle/Simple Settle)
Number of
Discharging
Foundries
Compliance Costs
(in thousands of 1983 dollars)
Closures
Capital Investment
Annual Costs
Number of Foundries
Number of
Employees
Direct
Indirect
Direct
Indirect
Direot
Indirect
Direct
Indirect
Total
All Poundrloo
Fewer than 10
16
11
187
136
208
181
0
0
0
0
10-19
20
28
1,912
2,613
723
1,008
0
0
0
0
50-99
16
5
1,017
730
130
271
0
0
0
0
100-249
6
6
3,323
200
1,118
69
0
0
0
0
250 or more
_5
_1
1.207
315
560
122
0
0
0
0
Total
63
51
7,916
1,355
3,369
1,650
0
0
0
0
Jobber Foundries
Fewer than 10
13
9
392
351
167
117
0
0
0
0
10-19
16
22
1,577
2,037
591
777
0
0
0
0
50-99
9
3
589
161
219
170
0
0
0
0
100-219
3
3
1,781
H5
761
10
0
0
0
0
250 or store
_1.
1
166
126
227
11
0
0
2
0
Total
12
38
1,803
3,091
2,002
1,177
0
0
0
0
Captlw Foundries
Fewer than 10
3
2
95
82
10
31
0
0
0
0
10-19
1
6
336
605
129
231
0
0
0
0
50-99
7
2
128
269
181
101
0
0
0
0
100-219
3
3
1,513
85
681
29
0
0
0
0
250 or more
_1
JL
711
219
333
Iฎ.
0
ฃ
0
ฃ
Total
21
16
3,112
1,261
1,367
173
0
0
0
0
Note: Numbers may not add up to totals due to rounding.
-------
TABLE V-26
COMPLIANCE COSTS AND ECONOMIC IMPACTS COPPER
(Option 2 Recycle/Lime Addition/Settle)
<
I
u>
N>
Number of
Discharging
Foundries
Compliance Costs
(in thousands of 1983 dollars)
Closures
Capital Investment
Annual Costs
Number of Foundries
Number of
Employees
Dlreot
Indirect
Dlreot
Indirect
Direct
Indirect
Direct
Indirect
Total
All Foundries
Fewer than 10
16
11
511
195
212
209
0
0
0
0
10-19
20
28
2,038
2,785
832
1,153
0
0
0
0
50-99
16
5
1,019
755
161
301
0
0
0
0
100-219
6
6
210
1,187
71
0
0
0
0
250 or more
_ฃ
_1
1.232
362
582
135
0
0
0
0
Total
63
51
8,208
1,607
3,607
1,871
0
0
0
0
Jobber Foundries
Fewer than 10
13
9
136
102
195
170
0
0
0
0
10-19
16
22
1,678
2,118
682
891
0
0
0
0
50-99
9
3
608
177
268
188
0
0
0
0
100-219
3
3
1,799
122
789
13
0
0
0
0
250 or more
1
1
167
132
230
1?
ฃ
0
0
0
Total
12
38
1,988
3,281
2,165
1,310
0
0
0
0
Captive Foundries
Power than 10
3
2
105
93
17
39
0
0
0
0
10-19
1
6
360
637
150
263
0
0
0
0
50-99
7
2
111
2 79
195
112
0
0
0
0
100-219
3
3
1,519
88
698
31
0
0
0
0
250 or more
_1
765
230
352
86
0
0
ฃ
0
Total
21
16
3,220
1,326
1,113
531
0
0
0
0
Note: Numbers may not add up to totals due to rounding.
-------
TABLE V-27
COMPLIANCE POSTS AMD EOOHOMIC IMPACTS COPPER
(Option 3 Feoyole/Llme Addltlon/Settle/Flltratlon)
<
I
u>
OJ
Compliance Costs
Number of
(In
thousands of 1983 dollars)
Closures
T>1 M/^hnrffl no
Foundries
Capital
Investment
Annual Costs
Number of Foundries
Niinhpr nf
Direct
Indirect
Direct
Indirect
Direct
Indirect
Direct
Indirect
Total
llylilwvi i
Employees
All Foundries
Fewer than 10
16
11
622
526
338
267
0
0
0
0
10-149
20
28
2,231
3,005
931
1,269
0
0
0
0
50-99
16
5
1,235
798
598
337
0
0
0
0
100-219
6
6
3,606
231
1 ,662
85
0
0
0
0
250 or more
_5
_1
i .31?
381
611
151
0
ฃ
ฃ
ฃ
Total
63
51
9,012
1,911
1,173
2,112
0
0
0
0
Jobber Foundries
Fewer than 10
13
9
501
128
272
217
0
0
0
0
10-19
16
22
1,838
2,318
761
979
0
0
0
0
50-99
9
3
715
501
315
211
0
0
0
0
100-219
3
3
1,910
136
882
50
0
0
0
0
250 or more
1
1
500
ill
251
56
ฃ
0
0
ฃ
Total
12
j 38
5,193
3,526
2,515
1,511
0
0
0
0
Captive Foundries
Fewer than 10
3
2
121
98
66
50
0
0
0
0
10-19
1
6
391
688
167
290
0
0
0
0
50-99
7
2
520
295
253
126
0
0
0
0
100-219
3
3
1,666
95
780
35
0
0
0
0
250 or more
_1
_a
819
213
m
_9ฃ
ฃ
0
ฃ
0
Total
21
16
3,519
1,120
1,658
598
0
0
0
0
Note: Numbers may not add up to totals due to rounding.
-------
TABLE V-28
COMPLIANCE COSTS AMD ECONOMIC IMPACTS COPPER
(Option 1 Recycle/Lime Addition/Settle/Plltration/Carbon Adsorption)
<
u>
.e-
Number of
Discharging
Foundries
Complianoe Costs
(in thousands or 1983 dollars)
Closures
Capital
Investment
Annual Costs
Number of Foundries
Number of
Employees
Direct
Indirect
Dlreot
Indirect
Dlreot
Indirect
Direct
Indirect
Total
All Foundries
Fewer than 10
16
11
719
615
102
311
0
0
0
0
10-19
20
28
2,117
3,182
1,028
1,185
0
0
0
0
50-99
16
5
1,336
903
651
395
0
0
0
0
100-219
6
6
3,721
230
1,713
81
0
0
0
0
250 or nore
_5
_1
1,35?
111
663
116
0
0
0
0
Total
63
51
9,583
5,700
1,160
2,187
0
0
0
0
Jobber Foundries
Fewer than 10
13
9
581
525
325
280
0
0
0
0
10-19
16
22
2,031
2,689
850
1,118
J 0
0
0
0
50-99
9
3
766
570
375
217
0
0
0
0
100-219
3
3
2,021
135
916
50
0
0
0
0
250 or more
1
1
196
166
250
6J.
0
0
2
0
Total
12
38
5,896
1,081
2,716
1,792
0
0
0
0
Captive Foundries
Fewer than 10
3
2
139
121
77
65
0
0
0
0
10-19
1
6
115
793
178
337
0
0
0
0
50-99
7
2
570
333
280
118
0
0
0
0
100-219
3
3
1,701
91
797
31
0
0
0
0
250 or more
_1
863
275
112
111
0
0
0
0
Total
21
16
3,688
1,616
1,711
695
0
0
0
0
Note: Numbers may not add up to totals due to rounding.
-------
Potential
Potential
Total
Total
Number of
Employment
Capital Cost
Annual Cost
Option
Closures
Lost
($ Thousands)
($ Thousands)
1
0
0
12,301
5,019
2
0
0
12,815
5,178
3
0
0
13,956
6,285
4
0
0
15,283
6,917
7. Potential Impacts on Zinc Foundries
As shown in Tables V-29 through V-32, potential costs for
treating the discharges from the 58 directly and indirectly discharging
zinc foundries range from $0.5 million under Option 1 to $1.6 million
under Option 4. The annual cost to treat the discharge from casting
quench operations for the two foundries with fewer than 10 employees
range from 2.5 percent of sales under Option 1 to 5.5 percent of sales
under Option 4. Because of the high profitablity and low debt of the
foundries in this subcategory, however, EPA does not anticipate any
potential closures a3 a result of the regulation.
Potential
Potential
Total
Total
Number of
Employment
Capital Cost
Annual Cost
Option
Closures
Lost
($ Thousands)
{$ Thousands)
1
0
0
1,326
533
2
0
0
1,897
1,003
3
0
0
2,070
1,175
1
0
0
2,884
1,575
8. Potential Impacts on Magnesium Foundries
As shown in Tables V-33 through V-36, annual costs for treating
the discharges from the four directly and indirectly discharging
magnesium foundries range from $12 thousand under Option 1 to $66
thousand under Option 4. Based on these costs, EPA estimates potential
closures ranging from two under Option 1 to three under Option 4. These
potential closures are based on average annual costs of 3>7 percent of
sales or higher, and result from failure of the return on assets and
Beaver's ratio tests. Potential costs and impacts under each of the
four Options are:
V-35
-------
TABLE V-29
COMPLIANCE COSTS AND ECONOHIC IMPACTS ZINC
(Option 1 Recyole/Slmple Settle)
<
u>
Number of
Discharging
Foundries
Compliance Costs
(in thousands of 1983 dollars)
Closures
Capital
Investment
Annual Costs
Number of Foundries
Nufflbpr nf
Direct
Indirect
Direct
Indirect
Direct
Indirect
Direct
Indirect
Total
IVUiilUvl wl
Employees
All Foundries
Fewer than 10
0
2
0
15
0
18
0
0
0
0
10-1)9
0
17
0
176
0
173
0
0
0
0
50-99
0
13
0
285
0
111
0
0
0
0
100-219
7
13
97
253
52
115
0
0
0
0
250 or more
2
_1
51
116
11
11
0
0
0
ฃ
Total
9
19
151
1,175
73
160
0
0
0
0
Jobber Foundries
Fewer than 10
0
1
0
22
0
9
0
0
0
0
10-19
0
13
0
353
0
130
0
0
0
0
50-99
0
11
0
207
0
81
0
0
0
0
100-219
6
11
67
200
38
91
0
0
0
0
250 or more
2
2
_2ฃ
51
J1
_I2.
0
0
0
0
Total
7
38
105
831
52
331
0
0
0
0
Captive Foundries
Fewer than 10
0
1
0
22
0
9
0
0
0
0
10-19
0
1
0
122
0
13
0
0
0
0
50-99
0
2
0
79
0
27
0
0
0
0
100-219
1
2
30
52
11
23
0
0
0
0
250 or more
2
2
J6
65
JL
21
0
0
0
0
Total
2
11
17
311
21
126
0
0
0
0
Note: Numbers may not add up to totals due to rounding.
-------
TABLE V-30
COMPLIANCE COSTS AMD ECONOMIC IMPACTS ZINC
(Option 2 Recycle/Lime Addition/Settle)
<
I
Number of
Discharging
Foundries
Compliance Costs
(in thousands of 1983 dollars)
Closures
Capital Investment
Annual Costs
Number of Foundries
Number of
Employees
Direct
Indirect
Direct
Indirect
Direct
Indirect
Direct
Indirect
Total
All Foundries
Fewer than 10
0
2
0
55
0
22
0
0
0
0
10-19
0
17
0
670
0
301
0
0
0
0
50-99
0
13
0
102
0
200
0
0
0
0
100-219
7
13
135
122
97
283
0
0
0
0
250 or more
2
_1
62
152
26
_Ii
0
()
0
0
Total
9
19
197
1,700
123
880
0
0
0
0
Jobber Foundries
Fewer than 10
0
1
0
28
0
11
0
0
0
0
10-19
0
13
0
500
0
229
0
0
0
0
50-99
0
11
0
316
0
168
0
0
0
0
100-219
6
11
95
310
71
231
0
0
0
0
250 or more
2
2
Jtl
5?
II
25
0
0
ฃ
0
Total
7
38
138
1,213
89
665
0
0
0
0
Captive Foundries
Fewer than 10
0
1
0
28
0
11
0
0
0
0
10-19
0
1
0
170
0
75
0
0
0
0
50-99
0
2
0
86
0
32
0
0
0
0
100-219
1
2
10
81
25
52
0
0
0
0
250 or more
2
_2
J!
_22
_2
16
0
0
0
0
Total
2
11
59
157
31
215
0
0
0
0
Note: Numbers may not add up to totals due to rounding.
-------
TABLE V-31
COMPLIANCE COSTS AMD ECONOMIC IMPACTS ZINC
(Option 3 Recycle/Line Addition/Settle/Filtration)
<
I
U)
oo
Number of
Discharging
Foundries
Compliance Costs
(in thousands of 1963 dollars)
Closures
Capital Investment
Annual Costs
Number of Foundries
Number of
Employees
Direct
Indirect
Direct
Indirect
Direct
Indirect
Direct
Indirect
Total
All Foundries
Fewer than 10
0
2
0
58
0
28
0
0
0
0
10-19
0
17
0
709
0
311
0
0
0
0
50-99
0
13
0
137
0
236
0
0
0
0
100-219
7
13
172
159
129
323
0
0
0
0
250 or more
_2
_1
_6ฃ
165
-21
81
0.
0
0
2
Total
9
19
212
1,828
163
1,012
0
0
0
0
Jobber Foundries
Fewer than 10
0
1
0
29
0
11
0
0
0
0
10-t9
0
13
0
530
0
260
0
0
0
0
50-99
0
11
0
316
0
199
0
0
0
0
100-249
6
11
122
371
96
261
0
0
0
0
250 or more
2
2
18
62
22
_2i
0
0
0.
0
Total
7
38
170
1,338
118
765
0
0
0
0
Captive Foundries
Fewer than 10
0
1
0
29
0
11
0
0
0
0
10-19
0
1
0
179
0
81
0
0
0
0
50-99
0
2
0
91
0
37
0
0
0
0
100-219
1
2
51
88
31
59
0
0
0
0
250 or more
ฑ
_2
21
102
11
52
0
0
0
0
Total
2
11
72
190
15
217
0
0
0
0
Note: Numbers may not add up to totals due to rounding.
-------
TABLE V-^2
COMPLIANCE COSTS AMD ECONOMIC IMPACTS ZINC
(Option 1 Recycle/Lime Addltlon/Settle/Filtratlon/Carbon Adsorption)
<
I
u>
VO
Number of
Discharging
Foundries
Compliance Costs
(in thousands of 1983 dollars)
Closures
Capital Investment
Annual Costs
Number of Foundries
Number of
Employees
Direct
Indirect
Direct
Indirect
Direct
Indirect
Direct
Indirect
Total
All Foundries
Fewer than 10
0
2
0
75
0
10
0
0
0
0
10-19
0
17
0
923
0
500
0
0
0
0
50-99
0
13
0
591
0
321
0
0
0
0
100-2^9
7
13
317
657
196
4412
0
0
0
0
250 or more
2
_1
_9ฃ
219
17
109
0
0
0
0
Total
9
49
4416
2,4468
212
1,333
0
0
0
0
Jobber Foundries
Fewer than 10
0
1
0
38
0
20
0
0
0
0
10-19
0
13
0
692
0
310
0
0
0
0
50-99
0
11
0
1477
0
271
0
0
0
0
100-2149
6
11
232
53*4
116
338
0
0
0
0
250 or more
_2
67
844
_2ฃ
3?.
0
0
0
0
Total
7
38
299
1,826
176
1,012
0
0
0
0
Captive Foundries
Fewer than 10
0
1
0
38
0
20
0
0
0
0
10-H9
0
0
231
0
109
0
0
0
0
50-99
0
2
0
117
0
50
0
0
0
0
100-2449
1
2
85
122
50
744
0
0
0
0
250 or more
2
_2
_22
m
27
69
0
ฃ
0
ฃ
Total
2
11
117
613
67
323
0
0
0
0
Note: Numbers may not add up to totals due to rounding.
-------
TABLE V-33
COMPLIANCE COSTS AND ECONOMIC IMPACTS MAGNESIUM
(Option 1 Recyole/Slmple Settle)
Number of
Discharging
Foundries
Compliance Costs
(in thousands of 1903 dollars)
Closures
Capital Investment
Annual Costs
Number of Foundries
Number of
Employees
Direct
Indirect
Direct
Indlreot
Direct
Indirect
Direct
Indirect
Total
All Foundries
Fewer than 10
0
0
0
0
0
0
0
0
0
0
10-19
2
2
17
57
22
20
1
1
2
16
50-99
0
0
0
0
0
0
0
0
0
0
100-219
0
0
0
0
0
0
0
0
0
0
250 or more
0
0
_0
0
_ฃ
_0
ฃ
0
ฃ
_ฃ
Total
2
2
17
57
22
20
1
1
2
16
Jobber Foundries
Fewer than 10
0
0
0
0
0
0
0
0
0
0
10-19
2
2
17
57
22
20
1
1
2
16
50-99
0
0
0
0
0
0
0
0
0
0
100-219
0
0
0
0
0
0
0
0
0
0
250 or more
ฃ
ฃ
0
0
_0
_0
ฃ
ฃ
0
0
Total
2
2
17
57
22
20
1
1
2
16
Captive Foundries
Fewer than 10
0
0
0
0
0
0
0
0
0
0
10-19
0
0
0
0
0
0
0
0
0
0
50-99
0
0
0
0
0
0
0
0
0
0
100-219
0
0
0
0
0
0
0
0
0
0
250 or more
ฃ
ฃ
ฃ
ฃ
0
ฃ
ฃ
ฃ
0
ฃ
Total
0
0
0
0
0
0
0
0
0
0
Note: Numbers nay not add up to totals due to rounding.
-------
TABLE V-3t
COMPLIANCE COSTS AND ECONOMIC IMPACTS MAGNESIUM
(Option 2 Recycle/Lime Addition/Settle)
Number of
Discharging
Foundries
Compliance Costs
(in thousands of 1983 dollars)
Closures
Capital Investment
Annual Costs
Number of Foundries
Number of
Employees
Direct
Indirect
Direct
Indirect
Dlreot
Indirect
Direct
Indirect
Total
All Foundries
Fewer than 10
0
0
0
0
0
0
0
0
0
0
10*19
2
2
59
65
26
23
1
1
2
16
50-99
0
0
0
0
0
0
0
0
0
0
100-2149
0
0
0
0
0
0
0
0
0
0
250 or more
0
0
0
_0
_0
_0
0
0
0
_0
Total
2
2
59
65
26
23
1
1
2
16
Jobber Foundries
Fewer than 10
0
0
0
0
0
0
0
0
0
0
10-IJ9
2
2
59
65
26
23
1
1
2
16
50-99
0
0
0
0
0
0
0
0
0
0
100-219
0
0
0
0
0
0
0
0
0
0
250 or more
0
0
0
0
0
0
ฃ
0
ฃ
_0
Total
2
2
59
65
26
23
1
1
2
16
Captive Foundries
Fewer than 10
0
0
0
0
0
0
0
0
0
0
10-19
0
0
0
0
0
0
0
0
0
0
50-99
0
0
0
0
0
0
0
0
0
0
100-219
0
0
0
0
0
0
0
0
0
0
250 or more
0
0
0
0
0
0
0
0
0
0
Total
0
0
0
0
0
0
0
0
0
0
Note: Numbers may not add up to totals due to rounding.
-------
TABLE V-35
COMPLIANCE COSTS AND ECONOMIC IMPACTS MAGNESIUM
(Option 3 Recycle/Lime Addltion/Settle/Filtratlon)
<
Number of
Discharging
Foundries
(in
Compliance Coats
thousands of 1983 dollars)
Closures
Capital
Investment
Annual Costs
Number of Foundries
Number of
Direct
Indirect
Dlreot
Indirect
Direct
Indirect
Direct
Indirect
Total
Employees
All Foundries
Fewer than 10
0
0
0
0
0
0
0
0
0
0
10-19
2
2
63
68
30
26
1
1
2
16
50-99
0
0
0
0
0
0
0
0
0
0
100-219
0
0
0
0
0
0
0
0
0
0
250 or more
0
ฃ
0
_0
0
_0
0
ฃ
ฃ
_ฃ
Total
2
2
63
68
30
26
1
1
2
16
Jobber Foundries
Fewer than 10
0
0
0
0
0
0
0
0
0
0
10-19
2
2
63
68
30
26
1
1
2
16
50-99
0
0
0
0
0
0
0
0
0
0
100-219
0
0
0
0
0
0
0
0
0
0
250 or more
ฃ
ฃ
0
_0
_0
_0
ฃ
ฃ
ฃ
0
Total
2
2
63
68
30
26
1
1
2
16
Captive Foundries
Fewer than 10
0
0
0
0
0
0
0
0
0
0
10-19
0
0
0
0
0
0
0
0
0
0
50-99
0
0
0
0
0
0
0
0
0
0
100-219
0
0
0
0
0
0
0
0
0
0
250 or more
0
ฃ
ฃ
ฃ
0
0
0
ฃ
ฃ
ฃ
Total
0
0
0
0
0
0
0
0
0
0
Note: Numbers may not add up to totals due to rounding.
-------
TABLE V-36
COMPLIANCE COSTS AND ECONOMIC IMPACTS MAGNESIUM
(Option i) Recycle/Line Additlon/Settle/Flltratlon/Carbon Adsorption)
<
I
U)
Number of
Discharging
Foundries
Compliance Costs
(in thousands of 1983 dollars)
Closures
Capital Investment
Annual Costs
Number of Foundries
Number of
Employees
Direct
Indirect
Direct
Indirect
Dlreot
Indirect
Direct
Indirect
Total
*11 Foundries
Fewer than 10
0
0
0
0
0
0
0
0
0
0
1019
2
2
81
68
10
26
2
1
3
69
50-99
0
0
0
0
0
0
0
0
0
0
100-219
0
0
0
0
0
0
0
0
0
0
250 or more
0
0
0
0
0
_0
0
0
0
_0
Total
2
2
81
68
10
26
2
1
3
69
Jobber Foundries
Fewer than 10
0
0
0
0
0
0
0
0
0
0
10-19
2
2
81
68
10
26
2
1
3
69
50-99
0
0
0
0
0
0
0
0
0
0
100-219
0
0
0
0
0
0
0
0
0
0
250 or more
0
0
_0
_0
0
_0
0
0
ฃ
_0
Total
2
2
81
68
10
26
2
1
3
69
Captive Foundries
Fewer than 10
0
0
0
0
0
0
0
0
0
0
10-19
0
0
0
0
0
0
0
0
0
0
50-99
0
0
0
0
0
0
0
0
0
0
100-219
0
0
0
0
0
0
0
0
0
0
250 or more
0
0
0
0
0
0
0
0
0
0
Total
0
0
0
0
0
0
0
0
0
0
Note: Numbers may not add up to totals due to rounding.
-------
Potential
Potential
Total
Total
Number of
Employment
Capital Cost
Annual Cost
Option
Closures
Lost
($ Thousands)
($ Thousands)
1
2
46
104
42
2
2
46
124
49
3
2
46
131
56
4
3
69
149
66
D. SELECTION OF OPTIONS
Table V-37 shows the selected options for the Metal Molding and
Casting effluent guidelines. The options chosen are based on EPA's
estimates of economic impacts and other factors. National regulations
have been chosen for all metals except magnesium.
Effluent guidelines for BPT (best practicable control technology
current achievable) are set based on removal using Option 2 technology
(partial recycle of process water followed by lime addition and
settling). For steel and aluminum, removals under BAT (best available
technology economically achievable), PSES (pretreatment standards for
existing sources), NSPS (new source performance standards) and PSNS
(pretreatment standards for new sources) have been set equal to BPT.
In general, standards for gray iron, ductile iron, malleable iron,
copper-based metals and zinc have been set at the more stringent Option
3 treatment (partial recycle of process water followed by lime addition,
settling and filtration). However, EPA has established lower levels of
stringency for small gray and malleable iron foundries. For malleable
iron foundries with fewer than 100 employees, BAT, PSES, NSPS, and PSNS
are set equal to BPT. For gray iron foundries with fewer than 50
employees, PSES and PSNS have also been set equal to BPT.
E. OTHER IMPACTS
In estimating potential impacts, EPA places primary emphasis on
potential closures and employment loss. However, as previously
mentioned there are several other measures of economic impact, including
potential price increases due to the regulation
potential production loss due to the regulation
potential balance of trade impacts
potential community effects
The remainder of this section provides an analysis of each of these
potential impacts.
V-44
-------
TABLE V-37
SELECTED OPTIONS FOR EFFLUENT GUIDELINES
BPT
BAT
PSES
NSPS
PSNS
Gray Iron
2
3
3a
3
3a
Ductile Iron
2
3
3
3
3
Malleable Iron
2
3b
3b
3b
3b
Steel
2
2
2
2
2
Aluminum
2
2
2
2
2
Copper-Base
2
3
3
3
3
Zinc
2
3
3
3
3
Magnesium
n.r.c
n.r.
n.r.
n.r.
n.r.
Option 1: Recycle and simple settle
Option 2: Recycle, lime addition, and settling
Option 3: Recycle, lime addition, settling and filtration
Option if: Recycle, lime addition, settling, filtration, and
carbon adsorption
BPT: Best practicable control technolgy currently available
BAT: Best available technology economically achievable
NSPS: New source performance standards
PSES: Pretreatment standards for existing sources
PSNS: Pretreatment standards for new sources
aFor plants with fewer than 50 employees, PSES and PSNS are set at
Option 2.
^or plants with fewer than 100 employees, BAT, PSES, NSPS, and
PSNS are set at Option 2.
cn.r. means not regulated.
V-45
-------
1. Potential Price Increases
In estimating potential impacts, EPA has assumed that foundries
would be unable to pass the compliance costs on to customers. The
assumption was based on estimates of competition both from domestic
foundries not incurring costs and foreign foundries. Less than one-
fourth of domestic foundries discharge process waters, and thus may
incur costs. Further, several respondents to the International Trade
Commission study claimed that they are already holding prices down in
response to foreign competition. Although EPA is basing its estimates
of impacts on the inability of foundries to raise prices, EPA has
assessed the potential price increase required for foundries to fully
pass along the cost increases to customers.
Table V-38 shows price pass-through requirements for all segments
incurring costs. The values represent the highest increase in cost
needed to fully recover compliance costs under the selected options. It
can be seen that potential price increases are generally very low (less
than one percent). In only seven cases do potential increases exceed
one percent. Potential closures are shown only where the required price
increase exceeds about three percent.
2. Potential Production Loss Due to the Regulation
EPA expects that production losses caused by this regulation
will be minor. Under the selected options only six foundries (five gray
iron and one ductile iron) are expected to close. Those six closures
could lead to a loss of about 14,000 tons per year of production, or
about 0.2 percent of combined gray and ductile iron production (see
Table V-39). Production losses of this magnitude can be easily made up
by the remaining foundries in the industry.
3. Potential Balance of Trade Impacts
This regulation is expected to have no significant impact on the
U.S. balance of trade. This conclusion is based on three factors:
Impacts have a minor share in the U.S. market.
Potential price increases on affected foundries are minor.
Most U.S. foundries incur no cost increase at all.
As shown in Chapter II, foreign imports have a very small share
of the U.S. market. Although some specific casting types have had
strong competition from Imports, foreign castings overall account for
only 2.6 percent of the total castings market. International Trade
Commission figures also show that exports of U.S. castings have grown at
the same time that imports have grown. Based on the data, it appears
that factors such as transportation costs, service and responsiveness
are strong enough to outweigh the price advantage of some foreign
castings.
The second factor precluding large balance of trade effects is
the small potential effect on prices. For almost all affected segments,
V-46
-------
TABLE V-38
PRICE PASS THROUGH REQUIREMENTS FOR
ALL REGULATED SEGMENTS
(selected options)
Metal
Size
Number
% Change
In Price
Closures
Gray Iron
Fewer than 10
2
1.31
no
10 to 49
52
3.10
yes
50 to 99
41
1.29
no
100 to 249
80
0.51
no
250 or More
61
0.39
no
Overall
236
0.49
Ductile Iron
10 to 19
9
2.24
yes
50 to 99
3
1.02
no
100 to 249
27
0.60
no
250 or More
13
0.42
no
Overall
52
0.51
Malleable Iron
50 to 99
8
0.59
no
100 to 249
33
0.32
no
250 or More
9
0.27
no
Overall
50
0.31
Steel
Fewer than 10
2
0.37
no
10 to 49
10
0.50
no
50 to 99
32
0.29
no
100 to 249
38
0.13
no
250 or More
25
0.09
no
Overall
107
0.12
Aluminum
Fewer than 10
7
0.70
no
10 to 49
70
0.60
no
50 to 99
26
0.16
no
100 to 249
55
0.08
no
250 or More
28
0.05
no
Overall
176
0.12
(Continued)
V-47
-------
TABLE V-3S (Continued)
Metal
Size
Number
% Change
In Price
Closures
Copper
Fewer than 10
27
1.43
no
10 to 49
48
0.21
no
50 to 99
21
0.31
no
100 to 249
12
0.29
no
250 or More
9
0.29
no
Overall
117
0.28
Zinc
Fewer than 10
2
3.79
no
10 to 49
17
0.70
no
50 to 99
13
0.29
no
100 to 249
20
0.07
no
250 or More
6
0.18
no
Overall
60
0.13
V-48
-------
TABLE V-39
POTENTIAL PRODUCTION IMPACTS FOB SELECTED OPTIONS
Gray Iron
Ductile Iron
Foundries Closed
5
1
Annual Sales per Foundry
($ thousands)
917
1,053
Sales Lost ($ thousands)
4,735
1,053
Sales by Dischargers in
Size Category ($ thousands)
76,708
14,712
% of Category Sales Lost
6.17
7.1
Sales by Dischargers
($ millions)
M82
1,231
% of Sales Lost
0.11
0.09
Tons Shipped per Foundry
2,102
2,038
Tons Lost
12,010
2,038
1982 Shipments of Metals
(thousand tons)
6,393
1,822
$ of Metal Shipments Lost
0.19
0.11
V-lป9
-------
price increases are less than 0.5 percent of costs. For comparison, it
should be noted that the value of the dollar fell 11 percent between
February and August, 1985, leading to an equivalent Increase in the cost
of imported castings. Relative to such fluctuations in the cost of
imports, the cost increase to affected foundries is minimal.
The third factor is the small population of affected
foundries. Although 800 foundries discharge process waters and thus
incur costs, more than 3,000 foundries do not. The competitiveness of
the 3.000 foundries not incurring costs will not be affected by this
regulation.
To summarize, only a fraction of foundries incur cost increases,
which are minor relative to recent changes in the value of the U.S.
dollar. As a result, EPA concludes that potential balance of trade
impacts are minor.
4. Community Effects
Because of the use of model plant analysis to determine impacts,
there is no way to determine which specific foundries will close rather
than comply with the regulations. In the absence of precise community
location of the affected foundries, the analysis assumes that the
distribution of closures will be the same as for foundries in general.
Foundries are located in four regions composed of various states, as
defined in the Census of Manufactures. These regions have been used as
the basis for an analysis of community effects. Table V-10 lists the
four regions and the states included in those regions.
The analysis of community effects has been confined to an
illustrative distribution of the closures among the four regions.
Closed foundries are assumed to have the same distribution as all
foundries casting the metal. Table V-41 shows that nearly half of the
six plant closures at the selected options might occur in the North
Central region, with another one-quarter of them in the Northeast
region.
Because of the small number of closures spread over the four
regions and the 1cm total employment loss, significant adverse impacts
in any one community are not expected.
V-50
-------
TABLE V-40
LIST OF REGIONS AND STATES WITHIN REGIONS
Northeast
North Central
South
West
Maine
Ohio
Delaware
Washington
Vermont
Indiana
Maryland
Oregon
Massachusetts
Illinois
Virginia
California
Rhode Island
Michigan
West Virginia
Montana
Connecticut
Wisconsin
North Carolina
Idaho
New York
Minnesota
South Carolina
Nevada
New Jersey
Iowa
Georgia
Utah
Pennsylvania
Missouri
Florida
Arizona
New Hampshire
North Dakota
Kentucky
New Mexico
South Dakota
Tennessee
Colorado
Nebraska
Alabama
Wyoming
Kansas
Mississippi
Hawaii
Arkansas
Alaska
Louisiana
Oklahoma
Texas
V-51
-------
TABLE V-ln
PROJECTED REGIONAL DISTRIBUTION OF CLOSURES
IN EMPLOYMENT-SIZE SEGMENTS
Employment-Size
Total
North
Segment
Closures
Northeast
Central
South
West
Gray Iron
10 to 49
5
1
2
1
1
Ductile Iron
10 to 19
ฑ
ฑ
ฃ
Total
6
1
3
1
1
Distribution
100$
17%
50%
17$
17$
V-52
-------
CHAPTER VI
NEW SOURCE IMPACTS
-------
VI. NEW SOURCE IMPACTS
The basis for new source performance standards (NSPS) and
pretreatment standards for new sources (PSNS), as established under
Section 306 of the Clean Water Act, is the best available demonstrated
control technology. Builders of new facilities have the opportunity to
install the best available production processes and waste-water
treatment technologies, without incurring the added costs and
restrictions encountered in retrofitting an existing facility.
Therefore, Congress directed EPA to require that the best demonstrated
process changes, in-plant controls, and end-of-pipe treatment
technologies be installed in new facilities. For regulatory purposes
new sources include greenfield plants and major modifications to
existing plants.
The potential economic impact of concern to EPA in evaluating new
source regulations is the extent to which these regulations represent a
barrier to the construction of new facilities or exert pressures on
existing plants to modernize, and thereby reduce the growth potential of
the industry.
In evaluating the potential economic impact of the NSPS/PSNS
regulations on new sources, it is necessary to consider the costs of the
regulations relative to the costs incurred by existing sources under the
BAT/PSES regulations. Under this regulation, new source requirements
are the same as those for existing sources. Therefore, no incremental
costs will be incurred by new source plants. Consequently, new sources
will not be operating at a cost disadvantage relative to existing
sources due to this regulation. The economic effects resulting from the
regulations are not significant (production cost increases range from
0.11 to 0.43 percent across all subcategories except magnesium) and,
therefore, will not in themselves pose a barrier to entry.
The magnesium subcategory is exempt from coverage under this
regulation. As previously reported, the costs of the treatment options
were projected to result in closure of from two (under Options 1, 2 or
3) to three (under Option 1) out of four existing discharging magnesium
plants. These closures reflected annual compliance costs amounting to
4.2 percent of production costs. Given the significance of the
compliance costs, plants in this subcategory are not included under the
regulation. This extends to new magnesium foundries, where the Agency
believes that the compliance costs would create a significant barrier to
entry.
This includes the provisions for small gray and malleable iron
plants. The PSNS requirements for small gray iron foundries are less
stringent, as are the NSPS and PSNS requirements for small malleable
iron foundries.
VI-1
-------
CHAPTER VII
SMALL BUSINESS ANALYSIS
-------
VII. SMALL BUSINESS ANALYSIS
This chapter analyzes the possible economic consequences resulting
from small foundry compliance with the proposed regulations. The
purpose is to determine if the regulations will impose a significant
economic Impact on a substantial number of small entities (i.e., small
businesses).
A. SMALL FOUNDRY SIZE CRITERIA
Under Section 3 of the Small Business Act (13 CFR Part 121), "small
business" is defined by the number of a firm's employees and by the
dollar volume of a firm's net income. For the foundry industry
specifically, the Small Business Act also specifies that the maximum
employee size for "small" foundries ranges from 500 for ferrous
foundries to 1,000 for nonferrous foundries, and that the maximum net
income size for all "small" foundries is $2 million. On the basis of
the SBA size criteria, most foundries qualify as small businesses. Of
all 3,661 foundries that were operating in 1978, 96 percent were small
according to the SBA employee size criteria, and 98 percent were small
according to the SBA net income criteria.
However, the Small Business Act recognizes that basic, narrow
definitions may not be applicable to an entire industry, particularly
when it has an extreme diversity of plant sizes. In such instances, the
Act permits the use of alternate criteria that more realistically
delineate the maximum size of "small business."
In the foundry industry, there is an extreme diversity of plant
sizes. In 1978, 61 percent of the 3,664 foundries had fewer than 50
employees, and those plants shipped only 6 percent of the industry's
tonnage. In sharp contrast, 29 percent of the foundries had between 50-
249 employees, and they collectively had a 31 percent shipments share.
Another 10 percent of the foundries having at least 250 employees
accounted for 63 percent of all tonnage shipped by the foundry industry.
Foundry managers and trade groups recognize operational differences
between foundries in three employment-size groups, and they frequently
describe those groups as small, medium, and large, respectively.
However, disagreement exists as to the precise cutoff for small
foundries. Based on these considerations, and the apparent threshold of
economic impacts shown in Chapter V, EPA is continuing to define small
foundries as those foundries having fewer than 50 employees.
B. IMPACT ANALYSIS FRAMEWORK
The analysis of economic impacts for the foundry industry was
confined to foundries that cast one of eight metal types as their major
metal. The analysis started by assigning foundries to categories based
on the major metal cast, the foundry's number of employees, and the
relative importance of castings shipments to Jobber and captive
VII-1
-------
markets. Estimates of the foundry population in 1986 were based on an
enumeration of foundries operating in 1983ป with modifications made to
reflect known changes from 1983 to 1984. The projected 1986 segment
populations were then distributed between dry and wet foundries, with
the wet plants being further distributed between (1) zero dischargers,
and (2) direct and indirect dischargers. Compliance investment costs
(capital costs) and annual costs (operating costs plus capital recovery)
were developed by EPA within the framework of the segmentation format.
C. CLOSURES FOR SMALL AND LARGE FOUNDRIES
The most visible and critical portion of the overall impact analysis
pertained to determination of the number of foundries that might close
rather than comply with the proposed regulations. Both jobber and
captive plants were expected to be subject to the same economic
criteria. It is assumed that economic factors on a plant-level basis
are the determining factors, and that small plants owned by larger
corporate entities are treated similarly to any other investment.
Closure thresholds were based on a review of the literature on financial
distress1 and a review of the historical operating behavior of
foundries.
Using the methodology described in Chapter III, EPA estimates that
800 of the 3,853 foundries projected for 1986 would be direct or
indirect dischargers. Compliance with Option 1 treatment by all of
those dischargers would require $43.2 million of capital costs and $16.2
million of annual costs, based on 1983 dollars. Application of the
financial tests indicated that four foundries might close rather than
install Option 1 technology. This number is 0.4 percent of all wet
foundries (direct, indirect, and zero dischargers).
For this analysis, the Agency has determined that small foundries
are those with fewer than 50 employees. Consequently, in 1986 there
would be 346 small wet foundries among a total of 2,511 small foundries
having 42,323 employees. Compliance with Option 1 treatment by the 250
small dischargers would, in 1983 dollars, involve $10.8 million of
capital costs and $4.2 million of annual costs. The financial tests
indicated that the four potential closures at Option 1 consist of two
small gray iron foundries and two small magnesium foundries, employing a
total of 100 persons.
The four small foundry closures attributable to Option 1 would be
equivalent to 1.6 percent of the small wet foundries. The 100 workers
that would face unemployment because of the closures would represent 1.3
percent of the employees for all small wet foundries.
^Most studies on financial distress use bankruptcy as the definition of
distress. A few use wider definitions, such as failure to pay preferred
stock dividends or lack of sufficient funds to cover checks.
VII-2
-------
Compliance by the 250 small foundry dischargers with Option 4, which
Is the most stringent of the alternate treatment levels, would Increase
the compliance capital and annual costs to $19.2 million and $9.0
million, respectively, in 1983 dollars. Applying those costs to the
financial tests indicated that 22 small foundries employing 582 workers
might close rather than comply with Option 4. Those 22 closures
represent 6.4 percent of small wet foundries, and their employment is
7.8 percent of the employees of small wet foundries.
To provide perspective, closure determinations for the larger
foundries (i.e., those with 50 or more employees) are also detailed. Of
the 1,342 larger foundries projected to be operating in 1986, 713 would
generate process wastewater, either as direct, indirect, or zero
dischargers. For them to comply with Option 1, capital and annual costs
of $32.4 million and $12.1 million, respectively, would be required,
based on 1983 dollars.
The 550 larger discharging plants would be employing more than
144,015 workers in 1986, which is 51 percent of the 284,140 total
employment by all 1,342 larger foundries. The Agency expects that one
malleable iron foundry employing 76 persons mights close rather than
comply with the regulations if Option 3 were the selected option, while
2 malleable iron foundries employing a total of 142 persons might close
rather than install Option 4 technology.
D. OTHER POTENTIAL IMPACTS
EPA also investigated the relative impact of the regulations as
measured by changes in financial performance. Three ratios in
particular were examined: the annual cost as a percentage of sales
(Table VII-1) and the annual cost as a percentage of the cost of
production (Table VII-2), and the change in return on assets (Table VII-
3). As is expected, costs are relatively greater for smaller
foundries. For four metals (steel, aluminum, copper-base and zinc),
costs remains less than one percent of sales for both size groups under
all options. Costs for gray and ductile iron are more substantial,
exceeding one percent of sales at Option 1 for small gray iron
foundries, and at Option 2 for small ductile iron foundries. Costs
exceed 3 percent of sales for magnesium foundries for all options.
Comparison of compliance costs as a percentage of the costs of
production yields similar trends.
E. REDUCTION OF IMPACT ON SMALL BUSINESSES
The Clean Water Act allows EPA to apply less stringent regulations
to small plants if the Agency determines that the regulations are not
economically achievable for small plants. Accordingly, the regulation
establishes less stringent limitations and standards for small foundries
in two subcategories where Impacts were significant.
For two metals, gray iron and malleable iron, EPA is establishing
less stringent standards for smaller foundries. Based on the estimates
of Impacts, EPA is establishing PSES and PSNS for gray iron foundries
VII-3
-------
TABLE VII-1
ANNUAL COMPLIANCE COSTS AS A PERCENTAGE OF SALES
FOR AFFECTED SMALL AND LARGE FOUNDRIES
(percent)
Metal
Option 1
Option 2
Option 3
Option 4
Small
Large
Small
Large
Small
Large
Small
Large
Gray Iron
1.57
0.11
3.01
0.39
3.56
0.45
4.69
0.49
Ductile Iron
0.52
0.16
2.08
0.41
2.24
0.49
2.49
0.54
Malleable Iron
NA
0.73
NA
0.27
NA
0.31
NA
0.34
Steel
0.21
0.04
0.47
0.11
0.58
0.13
0.67
0.14
Aluminum
0.33
0.04
0.61
0.08
0.70
0.09
0.87
0.11
Copper-base
0.20
0.36
0.22
0.26
0.26
0.29
0.30
0.31
Zinc
0.39
0.04
0.66
0.08
0.75
0.10
0.99
0.13
Magnesium
3.63
NA
4.20
NA
4.80
NA
5.69
NA
NA = Not applicable.
VII-4
-------
TABLE VII-2
ANNUAL COMPLIANCE COSTS AS A PERCENTAGE OF COST
OF PRODUCTION FOR AFFECTED SMALL AND LARGE FOUNDRIES
(percent)
Metal
Option 1
Option 2
Option 3
Option 4
Small
Large
Small
Large
Small
Large
Small
Large
Gray Iron
1.69
0.14
3.23
0.12
3.82
0.49
4.98
0.54
Ductile Iron
0.63
0.17
2.22
0.44
2.39
0.52
2.66
0.58
Malleable Iron
NA
0.08
NA
0.29
NA
0.33
NA
0.37
Steel
0.25
0.04
0.55
0.12
0.68
0.14
0.79
0.15
Aluminum
0.36
0.01
0.68
0.09
0.77
0.10
0.96
0.12
Copper-base
0.23
o.4o
0.25
0.28
0.29
0.32
0.34
0.34
Zinc
0.^2
0.04
0.72
0.09
0.82
0.10
1.08
0.14
Magnesium
3.87
NA
4.48
NA
5.12
NA
6.07
NA
NA = Not applicable.
VII-5
-------
TABLE VII-3
CHANGE IN RETURN ON ASSETS FOR
AFFECTED SMALL AND LARGE FOUNDRIES
(percent)
Option 1
Option 2
Option 3
Option 4
Metal
Small
Large
Small
Large
Small
Large
Small
Large
Gray Iron
Ductile Iron
Malleable Iron
Steel
-28.38
-10.75
NA
-2.07
-2.13
-2.89
-1.19
-0.63
-51.01
-35.95
NA
-4.61
-6.67
-7.17
-4.28
-1.72
-58.93
-38.57
NA
-5.54
-7.67
-8.40
-4.88
-1.95
-73.74
-42.77
NA
-6.32
-8.61
-9.24
-5.40
-2.11
Aluminum
Copper-base
Zinc
Magnesium
-4.90
-2.86
-5.99
-64.41
-0.69
-3.80
-0.84
NA
-8.42
-3.20
-9.54
-72.93
-1.26
-3.96
-1.58
NA
-9.55
-3.63
-10.68
-81.07
-1.43
-4.47
-1.85
NA
-11.78
-4.19
-13.95
-93.04
-1.73
-4.73
-2.53
NA
NA = Not applicable.
VII-6
-------
with fewer than 50 employees based on Option 2 technology, while larger
gray iron foundries must comply with standards based on Option 3. For
the malleable iron subcategory, EPA is establishing across the board
standards (BAT, NSPS, PSES, and PSNS) based on Option 2 for foundries
with fewer than 100 employees, while larger malleable iron foundries
must achieve removals based on Option 3*
EPA found that impacts on magnesium foundries are sufficiently
severe to warrant an exemption of magnesium foundries from the
regulation. It should be noted that all discharging foundries in the
magnesium foundry were found in the 10-49 employment size subcategory,
and that closures were projected beginning at 50 percent at Option 1.
F. REGULATORY FLEXIBILITY ACT
This regulation does not cause significant adverse economic impact
upon small foundries. The Agency has incorporated less stringent
requirements into the regulation for small foundries (gray iron and
malleable iron) where the compliance costs had significant effects on
plants in the small size categories. Additionally, the Agency has
excluded one subcategory (magnesium) from the regulation due largely to
the effects of compliance costs on plants in the subcategory, all of
which are small. Because the regulation does not create significant
economic impacts on a substantial number of small foundries, a separate
Regulatory Flexibility Analysis has not been prepared.
VII-7
-------
CHAPTER VIII
LIMITATIONS OF THE ANALYSIS
-------
VIII. LIMITATIONS OF THE ANALYSIS
A. FORECASTS OF SHIPMENTS
EPA has used the shipments data collected in its 308 survey as the
basis for estimating plant revenues. A review of other sources, notably
Census, shows that average revenues for some plants, particularly in the
smaller size categories, may be lower. EPA has made many efforts to
confirm the survey results, calling many foundries in 1982 to verify
production and employment values. The values used reflect the
submissions of 438 wet foundries responding directly to this
rulemaking. EPA has reduced the per plant shipment values by the
decline in industry production measured between 1978 (the base year for
the survey) and 1982 (or 1983 for steel). The use of the production
decline factors produces a lower bound on the production from plants
similar to those in EPA's data base, while also leading to values more
consistent with other data sources. Despite the Agency's efforts, it is
possible that 1986 shipments may be underestimated.
B. SELECTION OF RATIOS
EPA has selected three ratios to estimate impacts: return on
assets, debt to assets, and cash flow to total debt. The estimated
post-compliance value of each ratio for each quartile is used. If
values for two of the ratios fall below the threshold values, a model
plant is forecast to close.
Much of the literature on financial statement analysis has been
oriented to multivariate functions, which Include in one function
several important ratios. The benefit of a multivariate function is
that the effect of countervailing influences can be measured. That is,
a company with some poor financial ratios and some good financial ratios
will show up as less likely to fail than one with uniformly poor
financial ratios. Unfortunately, the existing multivariate functions
cannot be applied to the foundry industry, because they require data not
available, such as (1) the market value of stock (not relevant to
privately held companies), (2) the year-to-year changes in ratios (not
available if looking at how imposition of compliance cost would impact
forecast financial ratios), and (3) the relationship of a firm's ratios
to the industry median and quartiles (irrelevant if only examining
median and quartile ratios).
Although multivariate functions could be more accurate for closure
analysis, they require data that are only available at the level of
individual firms, and frequently require data only available for
publicly held companies. The three univariate measures we have chosen
were extensively investigated by Beaver. Although Beaver examined many
other1, similar ratios, these were the most effective. They, or very
similar ratios, have appeared as parts of the multivariate functions.
The interpretations of the ratios in the context of forecasting failure
are clear. We believe they represent the best of the techniques that
can be applied to the data available in the foundry Industry.
VIII-1
-------
C. USE OF SAME TESTS FOR CAPTIVES AND JOBBERS
This analysis assumes that captive plants and Jobbers operate under
the same financial conditions. Beaver's original threshold values were
established by examining larger, publicly held companies, so the values
should be consistent with those applied by parent corporations.
If the analysis errs, we believe that it may overestimate impacts on
captive plants. In general, the Dun & Bradstreet data show that larger
companies are financially in better shape. Small, captive plants,
supported by the overall financial structure of the parent organization,
should be healthier than small, independent firms. Furthermore, captive
plants satisfy the economic needs of assured supply and dedicated
scheduling to the parent firm, functions whose value cannot be
quantified by an outside observer. Captive plants owned by large
corporations may also have access to professional management techniques
generally available in large corporations, including electronic record
keeping and improved financial management.
Last of all, larger corporations have easier access to credit
markets. This is not to say that the controllers of large corporations
would automatically allocate funds to foundry operations. Instead, it
suggests that if the decision is made to install pollution control
equipment rather than close down foundry operations, a larger
corporation is more likely to be able to borrow the funds. Hence, small
foundries owned by larger corporations should have readier access to
debt markets than small, independent foundries.
D. DERIVATION OF COMPOSITE FINANCIAL STATEMENTS FROM QUARTILE RATIOS
Consider three firms, Able, Baker, and Charley, with the following
financial characteristics:
Able
Baker
Charley
Sales
1,000
2,000
1,000
Income
100
80
2U0
Assets
1,000
1,000
1,000
ROS
10$
4*
6%
ROA
10%
8*
2H%
Sales to Assets
1
2
H
By definition, the return on assets of a firm equals the return on
sales times the sales to assets ratio:
ROA = ROS x S/A
This is true for the three firms shown. When compiling quartiles,
however, the same does not hold true. Using our three companies as the
sample population, the ratios in the quartiles are as follows:
VIII-2
-------
Upper Quartile
Median
Lower Quartile
ROS
10$ (A)
6* (C)
4$ (B)
ROA
2k% (C)
10$ (A)
8$ (B)
S/A
4 (C)
2 (B)
1 (A)
It is not true for any quartile that return on sales times sales to
assets equals return on assets. Although this example is hypothetical,
the same results are observed when examining the quartiles published by
Dun & Bradstreet. In deriving balance sheets from the quartile data, we
have attempted to maintain the general relationship that increasing debt
imposes interest costs that decrease net income, and that the fraction
of debt is smaller for larger companies. In 1978, Dun & Bradstreet
published quartile data on return on sales, debt to net worth (D/NW),
and sales to net worth (S/NW). For deriving the model financial
statements, we used the following characteristics:
highest ROS with lowest D/NW and S/NW;
median ROS with median D/NW and S/NW; and
lowest ROS with highest D/NW and S/NW.
This procedure would increase the likelihood of at least one
quartile failing more than one of the closure tests, and may overstate
impacts.
VIII-3
-------
APPENDIX A
REVIEW OF FINANCIAL RATIOS AS PREDICTORS OF BANKRUPTCY
-------
APPENDIX A
REVIEW OF FINANCIAL RATIOS AS PREDICTORS OF BANKRUPTCY
A. INTRODUCTION
EPA 19 required to determine the estimated economic impact of the
regulations it promulgates. Frequently, the calculation has been
performed by estimating the impact of the additional compliance costs on
the financial statements of the impacted firms and inferring the number
of closures by the extent of the impact. Where the impact on a firm
resulted in a financial ratio, such as debt divided by total assets,
that exceeded a threshold value, the firm or class of firms was deemed
to be a potential closure.
Public comments on proposed regulations have questioned both the
ratios used and the threshold values selected. This paper addresses the
Issues of theoretical and empirical Justification Inherent in the use of
any financial test.
Part B, Summary of the Use of Financial Ratios, shows the broad use
of financial ratios to predict financial distress, both in the academic
literature and within EPA. As will be shown, several types of ratios
have recurred consistently as being reasonable predictors. While the
different studies used different statistical methodologies and different
ratios, all of them demonstrated that the financial ratios of failed and
non-failed firms are consistently different, and that the financial
ratios of failed and non-failed firms differ before failure.
Part C, Discussion of Specific Tests, presents the tests that (1)
have appeared most frequently in the literature, or (2) seem to have
rational explanations for their effectiveness. For each test, the
discussion highlights the theoretical considerations of the use of the
A-1
-------
test, the empirical history of its use, and any available threshold
values.
Part D, Financial Tests Proposed for Forecasting Foundry Closures,
gives three ratios considered to have both empirical Justification and
sufficient publicly available data to allow for their use.
Part E, Selection of Thresholds, presents the basis of the threshold
values chosen for each test. In particular, observations from the Dun &
Bradstreet financial data and from a review of recently bankrupt metal
companies are given, along with an interpretation of the data.
Part F, Interpretations of Results, explains why this analysis uses
the criteria that a model plant must fail two of three financial tests.
Part G, Summary, briefly presents the ratios chosen for this study,
and the reason for their selection, and the threshold values chosen.
B. SUMMARY OF THE USE OF FINANCIAL RATIOS
1. Academic Literature
Prediction of financial distress is important to many segments of
the business community such as bankers, investors, company managers,
regulatory bodies, and business competition. As early as 1908, bankers
and lenders were using the current ratio (the ratio of the current
assets of a company to its current liabilities) to predict loan
repayment (Beaver, 1966, p. 71). As financial accounting developed more
structure, more ratios could be examined. "(T)he development of
financial statement analysis in the 1920's and 1930's was characterized
by extensive data collection and the proliferation of new ratios (Lev,
1974, p. 3).n Threshold values for each ratio were determined on tin ad
hoc basis, with no theoretical or empirical Justification. Further,
analysts could receive conflicting estimates of the solvency of a firm
when looking at ratios individually.
A-2
-------
Starting in the mid-1960s, researchers started using statistical
techniques to determine the actual effectiveness of the various
financial ratios. The academicians had widely varying goals, and in
many cases used widely variant tests. Virtually all, however, found
that financial ratios of failed and non-failed firms differ before
failure.
For example, Tamari, in 1966, presented a formula for a weighted
average of several ratios. Although he was able to correctly classify
97$ of the non-failed and 52$ of the failed firms in his sample, the
weights used were arbitrary and the cut-off value was sample-specific.
In consequence, there was little reason to think the formula would be
applicable to a new selection of firms.
William Beaver, in 1966, was the first researcher to examine the
actual distribution of ratios for failed and non-failed firms to
determine appropriate thresholds. While Beaver did not solve the
problem of conflicting results from different interpretations, he did at
least demonstrate that some ratios were better than others, and that
threshold values could be determined from a review of actual firms.
Concurrently with Beaver, Horrigan (1966) investigated long-term
bond ratings as a function of financial ratios. Using multiple
regression analysis, he developed a function that could predict bond
ratings to within one classification. While not directly related to
financial distress, the bond rating of a company will affect its cost of
aqulring funds, and thus its cost of doing business. In addition, lower
bond ratings presumably reflect the analysts' opinions about potential
future financial distress.
Lev (reported in Moyer, 1977) used a univariate model based on the
balance sheet decomposition measure. The balance sheet decomposition
measure is a measure of the change in the relative proportions of
balance sheet measures from year to year. Failed firms show greater
changes, and thus have larger measures. The methods chosen by Lev and
Horrigan, although widely cited, were not used in many further studies.
A-3
-------
The next major analytical technique was promoted by Edward Altman In
1968. Using multiple discriminant analysis (MDA), Altman derived a
function of five ratios, four from the financial statement and one
incorporating market value of equity. Like Beaver, Altman used a paired
sample design, with bankrupt firms being matched with non-bankrupt firms
of equivalent asset size and comparable industry. In later years,
Altman continued to apply multiple discriminant analysis, determining
specific functions for railroads and brokerage houses. In 1977, he
estimated the parameters for a new, seven-variable discriminant model
(Altman, Haldeman, and Narayanan, 1977).
Following Altman's lead, other investigators applied multiple
discriminant analysis. Blum (1974) used discriminant analysis while
investigating the "failing company" doctrine, which is used as a defense
in antitrust cases. Deakin (1972) applied discriminant analysis to
firms that had gone bankrupt or were liquidated, and confined himself to
14 ratios obtained from balance sheet and income statement items.
Sinkey (1975) applied multiple discriminant analysis to identify problem
banks. Edmister (1972) compiled a quite complicated discriminant
function using a series of dummy variables. The study was supported by
the Small Business Administration and was specifically concerned with
small firms.
Moyer (1977) reviewed the performance of the original Altman model
against larger firms drawn from a later time period, finding the error
rate to be higher than for Altman*s original sample. When Moyer
reestimated the parameters of the model, he found that the coefficients
had changed. Further examination seemed to show that two of Altman's
variables, sales to total assets and market value of equity to book
value of debt, offer little additional Information to the model. In a
further test, he compared the Altman model to a two-variable
discriminant model composed of Beaver's cash flow to total debt ratio
and Lev's balance sheet decomposition measure, two univariate measures
expected to predict well. The Altman model had about the same error
rate, but the distribution of errors was different. Speolfically, the
A-iป
-------
Altman model predicted fewer non-falling firms as failing, while the
other model predicted fewer failing firms as non-failing.
Jarrod Wilcox (1971, 1973) sought to establish a theory of
bankruptcy. As a basis, he adopted the "Gamblers' Ruin" model. The
Gamblers' Ruin model presents the probability of ultimate bankruptcy
given the average gain or loss per period, the initial reserves, and the
probability of gains or losses. Reviews published with the 1973 paper
were sharply critical, and there has been little follow-up work.
As empirical tests using discriminant analysis proliferated, other
researchers examined the violations of statistical assumptions that came
from the use of ratios. Gupta and Huefner (1972) used IRS data to
cluster manufacturing companies by ratios, and presented what they
believed to be meaningful groupings for four financial ratios. To the
extent that industry characteristics affect financial ratios, tests
based on cross-industry patterns will be less valid for any specific
industry. Eisenbeis (1977) surveyed the literature on multiple
discriminant analysis and listed several common errors, with the most
severe error being the use of samples in which the classification is
either inexact (e.g., determined by subjective analysis), not inclusive
of the entire relevant population, or not necessarily discrete. Other
problems arise from Inconsistencies in _a priori probability estimates,
violation of the underlying statistical assumptions of the technique
used, and pooling of data across time.
Lev and Sunder (1979) examined the general issue of using ratios.
As they found, "a major reason for using financial variables In the form
of ratios is to control for the systematic effect of size on the
variables under examination .... (C)ontrol for size by the ratio
form is adequate only under very restrictive conditions (Lev and Sunder,
1979, pp. 187-188)." Deakin (1976) examined the distribution of
financial ratios. Use of multiple discriminant analysis assumes, among
other things, that the variables aire distributed normally. Deakin found
that most ratios are not distributed normally, that transformations
A-5
-------
(such as taking the logarithm or the square root) sometimes improve the
distribution, and that ratios of companies within a given industry may
be distributed more normally than ratios compiled from several
industries. Frecka and Hopwood (1983) also examined the distribution of
financial ratios. They found that the distributions are highly skewed,
but that a normal distribution could be obtained for most ratios by
deleting outliers and using a transformation of the data. Further, they
cite one study using discriminant analysis in which the results were
"strongly influenced by a small number of observations (Frecka and
Hopwood, 1983, p. 127)."
Ohlson (1980) reviewed the available literature and demonstrated
that, beyond statistical inconsistencies of the MDA model, previous
studies had potential timing problems (use of financial statements not
available before bankruptcy) and that the matched sample technique
potentially hid information. Instead, Ohlson used a conditional logit
analysis, which is claimed to avoid any prespecification of the
statistical distribution of predictors. In addition, his study used
recent (1970-1977) data, and had a relatively large sample (105
bankrupt, and 2,058 non-bankrupt firms). However, Ohlson specifically
excluded small, privately held companies (Ohlson, 1980, p. 111).
Further, Ohlson did not use any hold-out sample to provide a test of the
predictive ability outside of the original sample.
Not surprisingly, each author has attempted to extend the
literature. As seen, these attempts have resulted in the use of widely
divergent statistical techniques and forms of equations. They have also
resulted in the selection of quite different variables. Altaian (1968)
and Beaver (1966), the authors of the classical papers, used almost
entirely different sets of ratios. Following them, other authors
selected from either Beaver's list or Altaian13 list, but rarely looked
at both. In addition, the later authors frequently tried to develop new
ratios, incorporating either industry norms, the tenets of financial
theory, or financial variables peculiar to a specific industry. Table 1
shows the wide variety of ratios examined. From that list, we have
selected the financial ratios discussed in Section C.
A-6
-------
TABLE 1
FINANCIAL RATIOS EXAMINED IN STUDIES
OF THE PREDICTION OF FINANCIAL DISTRESS
AUTHORS
RATIOS
Beaver,
'66
Beaver,!
'68a
Altman,^
'68
Altnan2
et al '71
Altman?
73
Blum,^
74
DeaVin,^
*72
Edmls-
ter.*>^
Horrlgan^
66
Lev,1
69
Moyer, 4'*
*77
Ohlson,*
*80
Wilcox,
*71. *73
Cash flow/current liab.
A
Cash flow/sales
+
Cash flow/total assets
+
Cash flow/net worth
+
Cash flow/total debt
*
*
+
*
*
*
Net income/sales
~
*
Net lncone/total assets
*
*
*
*
Nat Income/net worth
+
Net Income/total debt
~
+
Current liab/total assets
+
Long-term liab/total assets
+
*
Current and long-term
llab/total assets
+
Current liab + long-term liab
+ oref. stock/total assets
*
*
+
*
*
+ - Variable tested. Not Included as "best" predictor.
* - Variable tested. Shoved discriminating ability.
^Variables tested In dlchotonous classification testa.
^Variables Incorporated Into multivariate functions.
^Edmister used dummy variables based on the relationship of a company's ratios to thresholds.
4
Moyer tested three oodels.
-------
TABLE 1
FINANCIAL RATIOS EXAMINED IN SIUDIIS
OF THE PREDICTION OF FINANCIAL DIMKISS
oo
AUTHORS
RATIOS
Bcjvcr,1
'66
Beaver,I
'68a
Altman,^
'68
Altman,z
et al '73
Altman,L
'73
'7 U
Deakin,^
'72
Edrais-
ter.2.3
llorrlgan^
'66
Lev,1
69
Moycr,1
77
Ohlson,I
'80
Wilcox,
^ 71, '73
Cash/total assets
+
~
*
Quick assets/total assets
+
+
*
Current assets/total assets
~
~
*
Working capital/total assets
*
+
*
+
*
* *
*
Cash/current Xlab
~
~
*
Quick assets/current llab
*
ฆf
*
Current assets/current llab
~
-f
~
*
*
Cash/sales
ฆf
~
*
Receivables/sales
~
Inventory/sales
+
Quick assets/sales
~
~
*
Current assets/sales
+
+
*
Worklns eanltal/sales
+
*
*
Net worth/sales
~
*
*
Total assets/sales
~
*
~
*
Cash Interval
~
nซfMfl1vp inrprvnl
~
1
-------
TABLE I
KINANC IAI KAT10S tXAMlNI.I) IN STUDIES
01- lllh I'KKIHCI ION OK KINANL1AL DISTRESS
=r
NO
AUTHORS
RATIOS
Beaver,1
*66
Bu.iver,1
'68a
Airman,"1
*68
Altaian,^
ec al '77
Altm.rn,*
*73
Hlum.l
*74
Deakln,'
72
Edmi s-
ter.2.3
Horrlgan'
'66
Lev,1
'69
Moyer,^
'77
Oh 1 son,'!
"80
Wilcox.
71. '73
No credit interval
*
Total assets
*
*
*
Retained earnings/total assets
*
*
EBIT/total assets
*
*
*
* *
Mkt value equity/book debt
*
*
*
Net available for total
cap/total cap
~
Sales/total capital
+
EBIT/sales
+
NATC/sales
+
Tangible assets
+
-
Interest coverage
*
Fixed charge coverage
+
~
Working capital/long-term debt
+
Retained earnings/total assets
*
*
Book equity/total capital
~
Other, off-balance sheet
~
*
*
-------
TABLE 1
HNANl IAI KAI LOS IWI1NII) IN MUIHIS
01 III) I'RI l)H I ION III MNANI.IAI IMMWSS
AUTIIOKS
RATIOS
Bv.ivur, '
*66
Heaver,1
' 68j
A1 tiiuii, ^
'68
Alrm.in,*
et al '7)
Altm.nl, ^
' 73
Ilium, ฃ
74
Dc.nkln,^
72
Kdral s-
ter,^.J
llorrigan^
'66
Lev, 1
"69
Hoyur,^
77
Oh1 son,i
*80
U1 lcox,
*71, * 7 J
"Oulck flow" ratio5
*
Industry specific
~ . *
Cash flow/fixed charges
Net aulck assets/lnventorv
*
Balance sheet decomposition
*
. .*
TL > TA dummy
*
Funds/total liabilities
*
"Canbler's Ruin"
ซ
3
Quick flow - cash + notes receivable ~ securities ~ (annual sales/12) i (COGS - Depreciation + G4A + Interest) ~ 12
-------
2. Summary
Financial ratios have a long history of use as predictors of
financial distress. Various ad hoc rules date back at least to 1908.
Since 1966, researchers have used a wide variety of tests to demonstrate
the validity of using financial ratios in this context. Financial
ratios of various sorts also have a wide precedent of use within several
of the EPA offices. Despite potential statistical problems, both single
variable threshold models and multiple variable functions have measured
differences between the balance sheets of firms that later failed and
those that did not.
C. DISCUSSION OF SPECIFIC TESTS
Over the years, financial analysts and accountants have grouped
individual financial ratios into four basic categories:
profitability;
solvency;
liquidity; and
efficiency.
More detailed statistical studies have found seven to eight
groupings of financial ratios (Pinches and others, 1973ป and Gombola and
Ketz, 1983). After excluding many of the ratios using short-term assets
and liabilities, however, the four categories listed provide a close
approximation to the statistically determined patterns.
The remainder of this section will discuss, in turn, the ratios
comprising each category. Ratios that are commonly used either in the
literature or in previous work by EPA have been considered. In
addition, the potential for using a multivariate test will be reviewed.
A 11
-------
1. Profitability Measures
a. Return on Assets
Return on assets (ROA) is computed as the ratio of net
income after taxes divided by total assets. Beaver tested this ratio in
1966, finding it to be the best of four "net income" ratios in
predictive ability. In 1980, Ohlson used the ratio as one of seven of
the variables making up his predictive function. ROA has a strong
advantage from the standpoint of availability, because it requires only
the net income and the total assets, both of which are reported in
financial statements.
From a more theoretical standpoint, the ratio has potential
disadvantages. Its principal drawback is that its use confuses the
separate issues of the productivity of the capital base and the
financing of the asset base. It is easy to show (see Table 2) that two
firms having identical assets, sales, and costs, but different financial
structures, will show different ROA. Depending on the specific numbers
chosen, the firm that provides the higher return on net worth (the
residual amount representing the owners' Interest In the firm) will have
the lower ROA. If an assumption is made that firms within an Industry
either seek, or should seek, a common financial structure, some of the
problems with the ratio seem less severe.
b. EBIT Divided by Total Assets
Earnings before interest and taxes (EBIT) divided by total
assets is another profitability measure that has been popular in the
literature. In particular, Altman used the measure in two of his
discriminant functions, while Moyer's review of Altman's work also
showed the measure to be a successful predictor. However, this measure
is much harder to determine if detailed financial statements are not
available. Dun & Bradstreet reports only values for return on net
worth. While D&B also reports ratios such as long-term debt to total
A-12
-------
TABLE 2
COMPARISON OF RETORN ON ASSETS FIGURES
Non-Leveraged
Leveraged
Debt
0
500
Net Worth
1000
500
Total Assets
1000
1000
Sales
2000
2000
EBIT
200
200
Interestฎ
0
50
Gross Income
200
150
Taxesb
100
li
Net Income
100
75
EBIT/TA ($)
20
20
ROA (%)
10
7.5
Return on Net Worth ($)
10
15
EBIAT/TA (%)c
10
10
Remarks: Firms have the same total assets, sales, and costs before
Interest. However, the firm with the higher return on net
worth, tiiich could be more valuable to the owners, shows a
lower return on total assets. From the standpoint of
economic value of the assets, however, the firms are
Identical, as shown by both the EBIT and the EBIAT.
aAssumes interest rate of 10$.
^Assumes tax rate of 50$.
CEBIAT (earnings before interest but after taxes) =
Net Income ~ (1-tax rate) x interest payments.
A-13
-------
assets, it is easily shown that the median values provided by D&B cannot
be combined to give a consistent balance sheet. Even using the ratio of
long-term debt to total assets to estimate the amount of interest-
bearing debt would not be sufficient, because both the amount of taxes
and the interest rate must be estimated as well.
Theoretically, EBIT divided by total assets is a little more
satisfactory than ROA because the earnings' value used is before the
considerations of financing. Because tax effects are not included, the
effects of special tax effects are not taken into account, which may
serve to penalize some industries. The literature has not provided
threshold values of EBIT divided by total assets on a univariate
basis. Altman provided the mean value for failed and non-failed firms
in both his 1968 find 1977 papers. From an investment standpoint, one
could say that any firm not providing an EBIT divided by total assets
greater than the before tax return on a comparable investment would be a
poor investment, and that the firm could be liquidated in favor of other
investments. However, doing so would confuse the historical cost basis
of the financial statements with the salvage value of the firm. The two
values are not necessarily related.
c. EBIAT Divided by Total Assets
The use of earnings before interest but after taxes (EBIAT)
has not been pursued in the literature on financial distress. As with
EBIT, calculation of the numerator requires manipulation of the
published data. Again, the D&B industry norms do not report a median
value by industry, so that use of EBIAT would require several
assumptions about interest rates, amount of interest-bearing debt, and
tax structures.
In principle, use of EBIAT gives the after-tax profitability of a
firm's assets without confusing the issue of financing. As a result, it
should provide the best estimate of the profitability of the firm, and
also give a value that would be directly comparable to other potential
A-14
-------
investments. It would be appropriate to compare EBIAT divided by total
assets to after-tax returns on treasury notes, possibly adjusted for a
risk premium. As with the use of EBIT, however, the use of "total
assets" as measured by the financial statements may not be a true
measure of the value of the assets employed.
d. Return on Sales
Return on sales (ROS) has been used in several of the EPA
economic impact studies. Its principal virtue is the relatively small
amount of data necessary only the value of production and the cost of
the compliance on an annualized basis. Unfortunately, the measure has
not been popular in studies on financial distress. Beaver (1966,
p. 106) showed ROS to be fairly successful, but not the best of the "net
income" measures (ROA, ROS, return on net worth, return on debt).
Altman's unsuccessful investigation of EBIT to sales is the only
reported attempt at using any measure of return on sales in multivariate
studies. A possible reason is that researchers assume that many firms
use "cost-plus" pricing, where the prices are developed to get a fairly
constant margin. Truth of the assumption would tend to violate
comparable assumptions in the field of economics.
Return on sales has been Included in factor analysis
studies. In those studies, which attempted to determine whether there
were common measures of the performance of firms, return on sales tended
to correlate highly with a dimension of "return on investment," along
with cash flow to total debt and income to assets.
2. Solvency Measures
a. Cash Flow to Total Debt
This ratio was Beaver's "best" predictor of financial
distress, and is commonly referred to as the "Beaver's ratio." As
defined by Beaver, cash flow consists of the net income plus
A15
-------
depreciation of a firm. Computationally, the ratio is easy to derive
from financial statements. Dun & Bradsteet does not provide an industry
norm for depreciation, but the Census does record total assets employed
and total depreciation by lJ-digit SIC. With appropriate manipulation
(ignoring the potential inconsistencies in the use of median data), an
estimate of depreciation can be calculated and added to a derived value
of the median net income to total debt. For use of Beaver's ratio, the
appropriate measure of total debt is all liabilities of the firm. This
includes such items as accounts payable, taxes payable, bank loans, and
capitalized leases.
b. Total Debt to Total Assets
The first issue to be resolved in the use of total debt to
total assets as a predictor is the definition to be used. Total debt to
total assets generally refers only to current plus long-term
liabilities, and does not include preferred stock. In Beaver's study,
however, the best of the "debt to total asset" ratios was the measure
using total debt plus preferred stock. For firms having no preferred
stock, the measures are identical. Another common measure, total debt
to net worth, is exactly equivalent to debt to total assets. Dun &
Bradstreet report an industry norm for total debt to net worth, but
apparently do not include preferred stock.
Use of some measure of debt to assets has been very popular
in the literature. Beaver tried several, and found total debt plus
preferred stock to be the best measure of debt. Horrigan used the same
measure in his study on bond ratings. Altman found that the measure was
not the best one from a multivariate sense when he studied the over-the-
counter brokerage industry. Instead, he found retained earnings to
total assets to be a better predictor, possibly because it captures
aspects of the age and past profitability of the firm. Ohlson used the
normal definition of total debt (i.e., not including preferred stock) in
his 1980 study.
A-16
-------
For a long time, analysts have used a threshold of total
debt to total assets of 0.5. The reasoning, as explained in the Dun &
Bradstreet publications, is that if the ratio is higher, the creditors
have more at stake than the owners. Beaver, however, found that the
actual threshold when predicting failure is somewhat higher. For
predictions one year, four years, and five years out, the threshold was
fairly stable at .57 to .58. Beaver's study used data from the period
1951-1961. Through the intervening years, average debt levels for most
firms have risen.
c. Interest Coverage
Interest coverage, computed as EBIT divided by total
Interest charges, is a measure sometimes mentioned as an indication of
the flexibility of the firm. Its only use in the academic literature on
financial distress is in Altaian's 1977 study, where interest coverage
entered as part of a multivariate function. Although Altman didn't
perform any dichotomous testing of the efficiency of the variable as a
predictor on its own, he did find that there were significant
differences between the interest coverages of failed and non-failed
firms. The threshold value frequently suggested by credit analysts and
stock analysts is two the earnings before interest and taxes should
be at least twice the fixed Interest charges. Using a threshold value
of two does not mean a company is extremely profitable, merely that it
is not in Imminent danger of falling.
3. Liquidity Measures
In the traditional literature, liquidity measures provide a
measure of a firm's ability to meet its short-term (less than one year
until due) obligations. Firms that may easily meet the obligations are
considered liquid; those which cannot are considered illiquid. Three
A-17
-------
common measures of liquidity are the current ratio (current assets1
divided by current liabilities), the quick ratio (quick assets divided
by current liabilities), and net working capital^ divided by total
assets. Net working capital divided by total assets has been a
frequently used variable in the empirical studies, and is one of the few
variables pursued by both Altman and Beaver. The potential effects of
the regulations on the short-term items in the balance sheet are
unknown, but are expected to be low. Thus, these tests will not be
useful in estimating the impacts of the regulations.
Efficiency Measures
Measures of efficiency are intended to reflect the extent to
which assets are used. Although many measures are possible, the one
that has received the most attention is sales to total assets. Beaver
found that this ratio is not a very good predictor of financial
distress. Altman included the ratio in his 1968 multivariate function,
but not in his 1977 version. Moyer's review in 1977 of Altman's work
also concluded that sales to total assets is not a particularly good
predictor. From the standpoint of assessing the economic impacts of
regulations, it is hard to see how the use of this variable would be
implemented.
5. Multivariate Measures
Concurrent use of several univariate tests suffers from the
problem of potentially inconsistent interpretations. A common trend in
^Current assets generally include cash, marketable securities, accounts
receivable, and inventories.
Quick assets are assets assumed to be readily convertible to cash:
cash, marketable securities, and accounts receivable (which can be sold
to external credit collection agencies).
%et working capital is the difference between current assets and
current liabilities.
A-18
-------
the academic literature has been the attempt to develop a multivariate
function that could be said to balance the impact of several aspects of
a firm's financial performance. Of all the tests proposed in the
literature, only the function given by Ohlson in 1980 uses variables
that can all be derived either from financial statements or by
manipulation of industry-wide data. Most of the variables are implicit
in the Dun & Bradstreet industry norms, but "funds divided by total
liabilitiesn corresponds to Beaver's ratio and must be computed by
adding the Census report of industry depreciation to the net income to
total debt figure. The problem still remains, however, of the potential
inconsistencies in the use of industry median values. In addition, two
of Ohlson1s variables require year-to-year changes in the ratios,
increasing the number of assumptions to be made and the potential error
in application.
D. FINANCIAL TESTS PROPOSED FOR FORECASTING FOUNDRY CLOSURES
A suitable test for assessing the economic impacts of regulations on
the foundry industry should meet three criteria:
strong empirical Justification;
threshold values derived from recent data; and
simple, consistent application to available data sources.
Table 3 provides a review of the most likely tests, the data
available for applying each test, and the issues Involved In using the
data to apply the tests. Section C has already reviewed the empirical
and theoretical Justifications of the tests.
Three ratios satisfy most of the criteria:
return on assets;
total debt to total assets; and
Beaver's ratio (cash flow to total debt).
A-19
-------
TABLE 3
COMPARISON OF UNIVARIATE RATIO TESTS
Test
Data Required
Data on Hand
Issues
ROA
1. Net Income
2. Assets
Can be computed from:
ROA plus assets;
ROA plus net income;
Net Income plus assets.
Dun & Bradstreet:
ROA (as ratio) and total assets:
median values by U digit SIC/
asset size class (see list).
1979-1982.
Iron Castings Society:
Mean "capital employed" by sales
class aggregate for iron foundries.
Steel Founders' Society
Profit before taxes and "capital
employed" by sales level-
aggregate for steel foundries.
Robert Morris Assoc.
Median profit before taxes/total
assets plus average assets, by
asset size class. Data only for
all ferrous and all non-ferrous
foundries.
FINSTAT:
Individual company data from 1977
1980 giving employment, asset size
from Dun & Bradstreet data base.
To calculate employment
effects, would like to
have relationship
between assets or net
income and employment.
Use of median values by
size category only allows
one test to be made for
each size class.
Median assets need not
correspond to median
ROA, leading to use of
incompatible data.
Use of before tax
earnings ratio would
require guess of tax
structure.
(Continued)
-------
COMPARISON OF UNIVARIATE RATIO TESTS (Continued)
Test
Data Required
Data on Hand
Issues
KB IT/Total Assets
1. Earnings before
Dun & Bradstreet
Multiplication of
interest and taxes
Median total assets
medians does not
(EBIT)
Median ROA
necessarily lead to
2. Total assets
Median total liab to NW
median value of another
Median Current liab to NW
ratio.
EBIT can be obtained:
Median Return on NW
- Directly from corporate
- derive long term liab to assets
Potential error in
financial statement;
- guess interest rate
guessing tax rates,
- From ratio plus total
- guess tax rate
interest payments.
assets;
- calculate EBIT/total assets.
- From net income plus
D&B provides no data on
taxes plus interest.
Robert Morris Associates
average or median tax
Median EBIT/interest
payments.
Median Profit before taxes/total
assets (can be combined to EBIT/
RMA asset sizes are not
total assets) average assets by
the same a9 Dun &
asset size category.
Bradstreet, and do not
correspond to SRI.
Steel Founders* Society
average profit before taxes
RMA data is only
average total assets
available at level of
- no data relevant to interest
ferrous or non-ferrous
payments.
foundries.
Iron Castings Society
average operating profit
average capital employed
- no data relevant to interest
payments.
KBIAT/Total Assets
1. Earnings Before Interest
Same as for EBIT/Total Assets.
Same restrictions and
and After Taxes (EBIAT)
issues as for EBIT,
2. Total Assets
plus the requirement
of additional
EBIAT requires EBIT
calculation.
plus taxes.
(Continued)
-------
COMPARISON OF UNIVARIATE RATIO TESTS (Continued)
T
K>
Is)
Test
Data Required
Data on Hand
Issues
ROS
1. Net Income
2. Sales
Can be derived from
ROS ratio plus either
net income or sales.
Dun and Bradstreet
Median ROS
Median net Income
Median sales.
Robert Morris Associates
Median sales/total assets
Median EBT/total assets
Median EBT/sales.
Steel Founders' Society
average profit before taxes
average sales.
Iron Castings Society
average operating profit
average sales.
FINSTAT
abbreviated balance sheets and
income statements 1977-1980
by 4-digit SIC.
Use of median net Income
divided by median sales
does not yield median
ROS.
RMA data available only
at level of ferrous or
non-ferrous foundries.
Would need to use
FINSTAT data to
correlate ROS to
predictor of failure.
Cash Plow/
Total Debt
1. Cash Flow
2. Total Debt
Cash flow requires net
income plus depreciation.
If net income to total debt
Is available, need either
total debt plus total
depreciation or depre-
ciation to total debt.
Dun and Bradstreet
Median ROS
Median total liab to NW
(combine to return on debt)
Median fixed assets to net worth
(combine to get fixed assets to
total debt).
May infer depreciation by guessing
average depreciation rate.
Combining median values
does not necessarily
yield median values.
Depreciation rate must
be obtained from separate
not necessarily
compatible source.
(Continued)
-------
COMPARISON OF UNIVARIATE RATIO TESTS (Continued)
Test
Data Required
Data on Hand
Issues
Cash Flow/
Total Debt
(Continued)
Total debt can be derived
from total assets plus
debt to assets.
Robert Morris Associates
Median sales to total assets
Median profit before taxes to
total assets
Median depreciation to sales
Median debt to net worth.
Census (Annual Survey of Manufactures)
Total gross depreciable assets
(not depreciated)
Total annual depreciation
(by M-digit SIC).
Iron Castings Society
average capital employed
average net worth
(difference is average total debt)
average operating profit
- Insufficient data.
RMA data available only
at level of ferrous or
non-ferrous foundries.
Census gives
depreciation vs gross
fixed assets; D&B
provides ratios
incorporating only net
fixed assets.
Total Debt/
Total Assets
Total Debt/Total Assets
plus total debt or
total assets
Total Debt/Net Worth
plus total debt or
net worth
Total Debt and Net Worth
Total Debt and Total
Assets.
Dun & Bradstreet
Median total liab to net worth
Median total assets
(also median total liab and
median net worth).
RMA
Median debt to net worth
Median total assets
Median net worth.
Medians not consistent.
RMA data not broken down
by metal type.
Projections of "typical"
asset size and relation
to employment unclear.
(Continued)
-------
COMPARISON OF UNIVARIATE RATIO TESTS (Continued)
Test
Data Required
Data on Hand
Issues
Interest Coverage
BBIT/lnterest
payments
EBIT
interest payments
Interest payment could
be obtained from debt
times interest rate.
See EBIT/Total Assets
Robert Morris Associates
EBIT/total interest directly profit
before taxes to total interest
from EBIT/total interest.
Many assumptions re:
interest rates, tax
rates, interest bearing
debt, etc.
Requires combinations of
median data.
-------
Applying these ratios as tests will require some assumptions about
the internal consistency of different ratios, but fewer than for the
other possible tests. Primarily, these three ratios have a strong
empirical base. No single ratio has been studied in recent, publicly
available financial distress studies, so threshold values for the
univariate tests must be Inferred from the values obtained in older
studies.
1. Return on Assets
Of the ratios examined, this best fulfills the requirements.
Although there are theoretical objections to its use (i.e., its
incorporation of the financing method), Beaver found it to be a
successful predictor. Use of the test requires relatively little
manipulation of the published data, because D&B publishes ROA
directly. All that is needed to apply the ratio is a measure of total
assets, which is needed for virtually all tests. The application of the
test is relatively insensitive to the financing method chosen, because
the cost of compliance is an asset whether financed through debt or
equity. The return variable, however, will be sensitive to financing
considerations.
Beaver's 1966 study provides an indication of an appropriate
threshold. Although the one year to failure cut-off was 0-2$, the
values for three to five years before failure consistently ranged
between 2$-4$.
2. Total Debt to Total Assets
This ratio has a long history in the ad hoc financial
literature, in the academic literature, and in EPA studies. Beaver
found it to be a fairly good predictor of failure. The principal
drawback to the ratio is its sensitivity to financing. Assumption of
all equity financing, for example, will show an improved financial
health for all firms by increasing the asset base and lowering the value
A-25
-------
of the ratio. Assumption of all debt financing will have a large Impact
on balance sheets, particularly of small firms. For purposes of
estimating financial Impact, however, all debt financing will be
assumed.
Another problem Is that the threshold is somewhat more difficult
to determine. Beaver found a value of .57-.58 in both the short-term
(one year) and long-term (four to five years) timeframes. In the mid-
term (two to three years), however, the threshold value dropped to the
range of .49-.51. Because the critical timeframe is likely to be the
short term, a base value of debt to assets of .57 seems appropriate.
3. Cash Flow to Total Debt
Cash flow to total debt was Beaver's most "successful" ratio.
It has a fairly simple economic interpretation operations must
generate enough cash to meet the debt service, or the company will fall
into a trap of borrowing to meet the Interest payments.
There are drawbacks to the use of Beaver's ratio. The principal
drawback is the inconsistency of the publicly available depreciation
data. The data available provides the industry total of gross
(undepreciated), fixed asset value and total depreciation claimed, in
dollars. Simple division gives a mean value of the ratio of
depreciation to historical cost, not depreciation as a fraction of net
fixed assets. The D&B industry composites only provide ratios using net
fixed assets. Assuming that depreciable lifetimes have historically
been based on average useful lifetimes, and that replacement of assets
will occur on a fairly steady level, assets will, on average, be 50$
depreciated. Thus, depreciation to net fixed assets will be about twice
depreciation to gross fixed assets. A review of publicly held metal
working companies verified that net fixed assets generally range from
JJ0ฃ to 60$ of gross fixed assets.
A-26
-------
E. SELECTION OF THRESHOLDS
Beaver's study used 195^196^ data. Since then, many changes have
occurred in the underlying financial and economic conditions. Inflation
has risen, debt levels have increased on balance sheets, and Interest
rates have risen.
We have selected as a basis Beaver's short-term (one year) threshold
values. Because of the variability of corporate performance from year
to year, it would be inappropriate to use long-term threshold values as
a comparison to quartile-base financial statements. Such a comparison
would imply that the same firms remain in the same relative financial
position year after year. In addition, the academic studies themselves
show declining performance for the models as the time span of the
forecast is increased. This is no doubt caused by the variability of
earnings for companies. From that standpoint, using a short-term test
is more relevant than using a long-term test.
The specific thresholds have been established after reviewing
Beaver's work, changes in fundamental economic conditions, the financial
ratios of the foundry industry, and the financial statements of three
metal working companies that have filed for Chapter XI bankruptcy in the
last three years: Revere Brass and Copper, Steelmet, and McLouth
Steel. None of the three filed for bankruptcy even when the return on
assets was as low as 2$-4$. Steelmet returned only 4$ on assets and had
a debt to assets ratio of 79$ in 1980, but did not file for bankruptcy
until 1982, after losing money and increasing its fraction of debt in
1981. Revere survived while making 5$ on assets, and having a Beaver's
ratio of 0.14. Only after return on assets dropped to 2$, and the
Beaver's ratio fell to 0.08, did Revere file for bankruptcy. McLouth
followed the same pattern. It made 2$ on assets in 1979, lost 12$ on
assets in 1980, and filed for bankruptcy even later. Between 1979 and
1980, its debt to asset ratio climbed from 62$ to 73$ป while Its
Beaver's ratio fell from 0.12 to -0.10.
A-27
-------
After reviewing these data and the ratio data for firms in the
foundry industry, the following threshold values have been selected.
1. Return on Assets
The threshold value for return on assets has been chosen as
2.5$. As seen in the review of bankrupt firms, bankruptcy was declared
only after return on assets fell to 2$ or less. Reviewing financial
ratios for the foundry industry, three segments with positive income
showed lower quartile ratios for return on assets of less than 3.5$.
These segments were the upper quartile of steel, NEC, with fewer than 10
employees (ROA = 2$), the lower quartile of aluminum, 10-49 employees
(ROA = 3-4$), and the lower quartile of magnesium, 50-249 employees
(ROA = 3.5$).
In the belief that the threshold should be somewhat above that
found for bankrupt firms, but below that normally found in the industry,
the analysis used an intermediate value of 2.5$.
2. Debt to Total Assets
The analysis uses a threshold value of 70$ debt. In Beaver's
study, the cut-off values for debt to assets ranged from .50 to .57.
There has been a substantial structural shift in the economy since the
research was done between 1961-1966, with companies at all levels taking
on more debt. Since 1978, the shift has been particularly noticeable.
For aluminum foundries, the lower quartile number for the debt to asset
ratio (exceeded by 25$ of the firms) has been 64$, 61$, 60$, and 57$ in
the years 1978-1981, consecutively. For gray iron foundries, the debt
to asset ratio was .57, .50, .57, and .61 in the years 1978-1981.
Revere's debt to asset ratio ranged from 58$ to 65$ in the three years
before bankruptcy, but climbed to 91$ in the year Revere declared
bankruptcy. Steelmet's debt to asset ratio was 79$-80$ in the two years
preceding bankruptcy, and dropped to 77$ the year of bankruptcy.
McLouth's debt to asset ratio rose from 62$ to 73$ in one year. We do
not know how soon thereafter McLouth declared bankruptcy.
A-28
-------
In reviewing computed debt to asset ratios for the foundry
industry, it was noted that a few quartiles of some metals have debt to
asset ratios above 60$, some as high as 68$. The highest values
computed were 74$ and 89$, for copper with fewer than 10 employees and
for steel, NEC, with fewer than 10 employees. This analysis uses a
threshold value of 70$ because it is above common values for the debt to
asset ratio, but below the value before failure for known bankrupt
firms.
3. Beaver's Ratio
The threshold value for Beaver's ratio was selected to be 8$.
The highest Beaver's ratio value for the three bankrupt firms studied
was 8$ (for Revere), with all others being lower.
Estimated Beaver's ratio values from the Dun & Bradstreet data
were generally high. For some quartiles, for very small (less than 10
employees) firms, the values were negative. The lowest positive value
for larger firms was 9.5$, for the lower quartile of aluminum, 10-49
employees. The lower quartile for malleable iron shows a negative
Beaver's ratio for firms with 50 or more employees.
4. Summary
Although the threshold values we have selected are stringent,
they seem to reflect the actual behavior of firms if faced with the
prospect and costs of bankruptcy.
F. INTERPRETATIONS OF RESULTS
The third issue in the use of financial ratio-based tests is the
interpretation of results. If one ratio falls below its threshold value
while the other two pass, how should the firm be rated?
A-29
-------
Examination of the data for the three bankrupt firms shows that
companies did not file for bankruptcy unless they failed at least two
tests. This seems rational in principle. If a company has a very high
fraction of debt, but also has sufficient income and cash flow to
satisfy investors and creditors, it would likely stay in business. If a
company has low income in one period, but still has both sufficient cash
flow and moderate debt, again it would likely stay in business. In the
third case, if a company has a low Beaver's ratio (cash flow to total
debt) but low debt to assets and reasonable return on assets, it would
again probably stay in business.
However, if two ratios are below the threshold, there is much less
chance of recovery. It is considerations such as these that have
promoted the development of multivariate functions.
G. SUMMARY
Financial ratios have a long history of use in the financial
distress literature. Statistical research since 1966 proves that
financial ratios are different between failed and non-failed firms.
Given the large number of foundries, and the need to predict the
economic impacts at the time of compliance, use of aggregate ratios
seems to be a reasonable means to predict industry-wide impacts.
Based on the data available for the foundry Industry and the
empirical findings, three ratios best meet the criteria used as a basis
for selection:
return on assets;
total debt to total assets; and
Beaver's ratio.
Financial variables to compute these ratios are available at the
4-digit SIC level for virtually all the foundry segments. With the
assumptions outlined for each ratio, the data may be applied In a
rational, consistent manner.
A-30
-------
However, several changes have occurred in average balance sheet
composition and overall economic conditions since Beaver's study. In
addition, several studies have shown that using industry-specific data
is desirable when drawing conclusions about financial ratios. Because
of these factors, normal financial ratios for foundries and ratios for
recently bankrupt metal companies, were reviewed to determine reasonable
threshold values for foundries in the 1980s. The review has resulted in
the use of the following threshold values and criteria for failure:
return on assets 2.5% minimum
total debt to total assets 10% maximum
Beaver's ratio 8$ minimum
Failure if two of three ratios surpass thresholds.
A-31
-------
|