United States
Environmental Protection
Agency
Office of Air Quality
Planning and Standards
Research Triangle Park, NC 27711
November 1998
Air
EPA Economic Impact Analysis for the
Proposed Secondary Aluminum
Industry NESHAP
-------
CONTENTS
Page
TABLES
vi
FIGURES . . vii
1.0 INTRODUCTION . . 1
1.1 BACKGROUND 2
2.0 AN INDUSTRY PROFILE
OF THE SECONDARY ALUMINUM INDUSTRY 3
2.1 INTRODUCTION .... ... 3
2.1.1 BACKGROUND . ... . .... 3
2.1.1.1 Types of Scrap .... 3
2.1.1.2 Relationship Between the Primary Aluminum and
Secondary Aluminum Industries 4
2.1.1.3 Problems in Characterizing the Secondary Aluminum
Industry .... ... 5
2.1.2 ORGANIZATION OF CHAPTER 2 . 7
2.2 MARKET STRUCTURE 8
2.2.1 AFFECTED FIRMS AND FACILITIES . ... ..14
2.2.1.1 Scrap Dealers '. .15
2.2.1.2 Secondary Smelters 15
2.2.2 LOCATION AND EMPLOYMENT 16
2.2.3 MARKET CONCENTRATION 16
2.2.4 INTEGRATION 17
2.2.4.1 Vertical Integration 22
2.2.4.2 Horizontal Integration 24
2.2.5 TOTAL CAPACITY 24
2.3 PRODUCTION, SHIPMENTS, AND CAPACITY UTILIZATION 25
2.3.1 PRODUCTION AND CAPACITY TREND 26
11
-------
2.3.2 COMPETITION AND SUBSTITUTES 28
2.3.3 IMPORTS AND EXPORTS 31
2.3.4 SUPPLY DETERMINANTS 34
2.3.5 CAPACITY EXPANSION 36
2.4 DEMAND AND END USE MARKETS 37
2.4.1 CONSUMPTION TRENDS 37
2.4.2 DEMAND DETERMINANTS 40
2.4.2.1 Primary and Secondary Aluminum 40
2.4.2.2 Secondary Aluminum 42
2.4.3 PRICES 44
2.4.3.1 Scrap Prices 46
2.4.3.2 Secondary Aluminum Ingot Prices 46
2.5 FINANCIAL DATA 48
2.5.1 REVENUES 48
2.5.2 CAPITAL FORMATION 50
2.5.3 PROFITABILITY 50
2.5.4 BALANCE SHEETS AND INCOME STATEMENTS 52
2.6 MARKET OUTLOOK 52
2.6.1 PRODUCT DEMAND 52
2.6.2 CAPACITY DEMAND 55
2.6.3 RAW MATERIAL OR PROCESS CHANGES 55
2.6.4 NEW SOURCES-GROWTH AND REPLACEMENT 56
2.6.5 PRICE PROJECTIONS 57
2.6.6 IMPORTS AND EXPORTS 57
3.0 ECONOMIC METHODOLOGY 59
3.1 INTRODUCTION 59
3.2 MARKET MODEL 59
3.2.1 Partial-Equilibrium Analysis 59
3.2.2 Market Demand and Supply 60
3.2.3 Market Supply Shift 61
3.2.4 Impact of the Supply Shift on Market Price and Quantity 64
3.2.5 Trade Impacts 66
3.2.6 Plant Closures 67
3.2.7 Changes in Economic Welfare : 67
3.2.7.1 Changes in Consumer Surplus 67
3.2.7.2 Change in Producer Surplus 68
3.2.7.4 Total Economic Costs 69
3.2.8 Labor Input Impact 69
111
-------
3.2 BASELINE INPUTS 71
3.3 INDUSTRY SUPPLY AND DEMAND ELASTICITIES 72
3.3.1 Introduction 72
3.3.2 Price Elasticity of Demand 74
3.3.2.1 Approach 75
3.3.2.2 Data 77
3.3.2.3 Statistical 77
3.3.3 Price Elasticity of Supply 79
3.3.3.1 Model Approach 79
3.3.3.2 Estimated Model 83
3.3.3.3 Data 83
3.3.3.5 Statistical Results 85
3.3.3.6 Limitations of the Supply Elasticity Estimates 86
3.4 FINANCIAL IMPACT ANALYSIS 86
4.0 CONTROL COSTS, ENVIRONMENTAL IMPACTS, COST
EFFECTIVENESS, AND ESTIMATES OF ECONOMIC COSTS
88
4.1 INTRODUCTION 88
4.2 CONTROL COST ESTIMATES 88
4.3 ENVIRONMENTAL IMPACTS AND COST EFFECTIVENESS 89
4.4 ESTIMATE OF ECONOMIC COSTS 92
5.0 PRIMARY ECONOMIC IMPACTS AND FINANCIAL IMPACT
ANALYSIS 95
5.1 INTRODUCTION 95
5.2 ESTIMATES OF PRIMARY IMPACTS 95
5.3 FINANCIAL IMPACT ANALYSIS 98
5.4 LIMITATIONS 98
5.5 SUMMARY 102
6.0 SECONDARY ECONOMIC IMPACTS 103
* *
6.1 INTRODUCTION 103
iv
-------
6.2 LABOR MARKET IMPACTS 103
6.3 FOREIGN TRADE 104
6.4 REGIONAL IMPACTS 105
6.5 SUBSTITUTE PRODUCT MARKETS 105
6.6 LIMITATIONS 106
6.7 SUMMARY 106
7.0 POTENTIAL SMALL BUSINESS IMPACTS 108
7.1 INTRODUCTION 108
7.2 SMALL BUSINESS CLASSIFICATION 108
7.3 METHODOLOGY 109
7.4 SMALL BUSINESS IMPACTS 110
APPENDIX A 112
SENSITIVITY ANALYSIS 112
A-1 INTRODUCTION 112
A-2 SENSITIVITY OF THE MODEL ESTIMATES OF THE PRICE
ELASTICITY OF DEMAND AND THE PRICE ELASTICITY OF
SUPPLY 112
A-3 SENSITIVITY ANALYSIS OF THE PER UNIT COSTS OF
EMISSION CONTROLS 114
REFERENCES 116
-------
TABLES
Page
2-1. EMPLOYMENT IN THE SECONDARY ALUMINUM 19
INDUSTRY
2-2. SECONDARY ALUMINUM SUPPLY IN RELATION TO 27
TOTAL ALUMINUM SUPPLY: 1984-1994
2-3. TRENDS IN SECONDARY ALUMINUM RECOVERY: 29
1982-1997
2-4. TRENDS IN U.S. ALUMINUM CAN RECLAMATION: 30
1983-1994
2-5. TRENDS IN ALUMINUM SCRAP FOREIGN TRADE: 33
1983-1995
2-6. TRENDS IN SECONDARY ALUMINUM AVERAGE PRICES, 45
1985-1997
2-7. YEAR-END PRICES FOR A SAMPLING OF 47
SECONDARY ALUMINUM INGOT, 1993-1994
2-8. CAPITAL INVESTMENT IN SIC'S 3341 AND 3353 51
2-9. KEY BUSINESS RATIOS FOR SIC'S 3341 AND 3353 53
FOR 1992
3-1. BASELINE INPUTS FOR THE SECONDARY 73
ALUMINUM INDUSTRY
3-2. DATA INPUTS FOR THE ESTIMATION OF DEMAND 78
EQUATIONS FOR THE SECONDARY ALUMINUM
INDUSTRY
3-3. DERIVED DEMAND COEFFICIENTS 78
3-4. DATA INPUTS FOR THE ESTIMATION OF THE 84
PRODUCTION FUNCTION FOR THE SECONDARY
ALUMINUM INDUSTRY
3-5. ESTIMATED SUPPLY MODEL COEFFICIENTS 85
4-1. CONTROL COST ESTIMATES FOR MODEL PLANTS 90
AND NATIONWIDE COST ESTIMATES
-VI-
-------
4-2. ENVIRONMENTAL IMPACTS AND COST 91
EFFECTIVENESS
4-3. BASELINE EMISSIONS AND EMISSION 92
REDUCTIONS
4-4. ANNUAL ECONOMIC COST ESTIMATES 94
5-1. SUMMARY OF PRIMARY ECONOMIC IMPACTS 97
5-2. MODEL PLANT COST-TO-SALES RATIOS 99
6-1. SUMMARY OF SECONDARY IMPACTS 104
7-1. MODEL PLANT COST/SALES RATIOS 111
7-2. COMPANY SPECIFIC COST/SALES RATIOS 111
A-1. PRICE ELASTICITY OF DEMAND AND SUPPLY 113
ESTIMATES
A-2. SENSITIVITY ANALYSIS FOR ESTIMATED PRIMARY 113
IMPACT: LOW-END AND HIGH-END ELASTICITY
ESTIMATES
A-3. SENSITIVITY ANALYSIS FOR CONTROL COSTS 115
-Vll-
-------
FIGURES
Page
2-1. METHODS FOR PROCESSING SCRAP ALUMINUM 6
2-2. FLOW OF ALUMINUM SCRAP AND SECONDARY 11
ALUMINUM IN THE US
2-3. LOCATIONS OF PRIMARY AND SECONDARY 18
SMELTERS IN THE US
2-4. MARKET CONCENTRATION FOR SIC 3341 20
2-5. MARKET CONCENTRATION FOR SIC 3353 21
2-6. 1994-1995 GROSS SHIPMENTS OF ALUMINUM 23
INGOT
2-7. TRENDS IN DEMAND FOR ALUMINUM SCRAP: 38
1983-1993
2-8. CONSUMPTION OF SCRAP ALUMINUM BY 39
CONSUMER CLASS, 1994
2-9. ALUMINUM CONSUMPTION BY SECTOR, 1994 41
2-10. PERCENTAGE OF TOTAL FIRMS IN GIVEN SALES 49
RANGES
3-1. ILLUSTRATION OF POST-NESHAP MODEL 65
3-2. CHANGE IN CONSUMER SURPLUS, PRODUCER 70
SURPLUS, AND NET CHANGE IN SURPLUS
-Vlll-
-------
1.0 INTRODUCTION
The purpose of this economic impact analysis (El A) is to evaluate the economic effects of
the emission control costs estimated to occur as a result of the National Emission Standard for
Hazardous Air Pollutants (NESHAP) for the secondary aluminum industry. The El A was
conducted for one regulatory option chosen by the United States Environmental Protection
Agency (EPA ) for the regulation of affected secondary aluminum facilities in the secondary
aluminum industry. This analysis compares the economic impacts of the regulation to baseline
industry conditions that would occur in absence of the regulation.
Section 112(d) of the Clean Air Act (CAA) lists source categories of major sources and
specific area sources of hazardous air pollutants (HAP's) for which regulations must be
developed. The U.S. Environmental Protection Agency (EPA) is currently preparing maximum
achievable control technology (MACT) standards for emission sources in the secondary
aluminum industry. For the purposes of developing the MACT standards, the industry is defined
to include any secondary aluminum facility that performs any of the following:
• crushing and shredding;
• drying;
• delaquering;
• melting;
• holding;
• sweating; and
• dross cooling.
These secondary aluminum processes are carried out at facilities classified under several
standard industrial classification (SIC) codes, including SIC 3334 (primary aluminum); SIC 3341
(secondary smelting and refining of nonferrous metals); and SIC 3353 (aluminum sheet, plate,
and foil). For a more detailed discussion of the primary aluminum industry, see the EPA
document entitled, .Industry Profile for the Primary Aluminum MACT, dated March 1996.
1
-------
The purpose of this economic impact analysis is to assess the market impacts of this
regulation for the secondary aluminum industry and other affected markets and entities. Chapter
2 of this report is an industry profile which contains a compilation of economic and financial data
on the affected industry. Included in this profile are an identification of affected secondary
aluminum facilities, a characterization of market structure, discussions of the factors that affect
supply and demand, a discussion of foreign trade, a financial profile, and the quantitative data
inputs for the economic impact analysis model. Chapter 3 outlines the economic methodology
used in this analysis, the structure of the market model, and the process used to estimate industry
supply and demand elasticities. Chapter 4 presents the control costs used in the model, the
estimated emission reductions expected as a result of regulation, the cost-effectiveness of the
regulatory alternative, and an estimate of the economic costs of the regulation. Chapter 5
presents the estimates of the primary impacts determined by the model, which include estimates
of price, output, and industry revenue impacts. A financial impact analysis is included, as well as
a discussion of the limitations of the model. Chapter 6 presents the secondary economic impacts,
which are estimated quantitative impacts on the industry's labor market, foreign trade, substitute
products, and regional markets. Chapter 7 specifically addresses the potential impacts of
regulation on small affected firms. Lastly, Appendix A presents the results of sensitivity
analyses conducted to quantify the extent to which the model results are sensitive to specific
input data.
1.1 BACKGROUND
The Clean Air Act stipulates that HAP emission standards for existing sources must at
least match the percentage reduction of HAPs achieved by either: (1) the best performing 12
percent of existing sources, or (2) the best 5 sources in a category or subcategory consisting of
fewer than 30 sources. This minimum level of control is referred to as the MACT floor. The
level of control proposed in this regulation is the MACT floor level of control. The analysis
evaluates the economic impacts of the MACT floor regulatory alternative for the secondary
aluminum industry.
-------
2.0 AN INDUSTRY PROFILE
OF THE SECONDARY ALUMINUM INDUSTRY
2.1 INTRODUCTION
The purpose of the industry profile is to present financial and economic data on the
secondary aluminum industry to facilitate an accurate assessment of the economic impacts of the
alternative MACT standards on price, output, and consumption in the secondary aluminum
market, and firm-level impacts such as unemployment and plant closures. This chapter provides
a broad overview of the market structure of the industry.
2.1.1 BACKGROUND
2.1.1.1 Types of Scrap
Aluminum scrap can be divided into two major categories: "new" and "old" scrap. New-
scrap is generated in the manufacturing of primary aluminum, semifabricated aluminum mill
products, or finished industrial and consumer products. New scrap includes classifications such
as borings and turnings (from machinery operations); clippings, cuttings, and other solids (from
the aircraft industry, fabricators, and manufacturing plants); residues formed during the melting
process (dross, skimmings, spillings, and sweepings); and obsolete or surplus products (e.g.,
castings and mill products). New scrap is the easiest to process because its components and
alloys are known by the company performing the recovery.1
New scrap is further categorized as either "runaround" (also known as "home") scrap or
"purchased" scrap. Runaround scrap is new scrap that is recycled by the same company that
generates it; because it is used by the company that generates it, such scrap never enters the
-------
marketplace. Purchased scrap is new scrap that is imported, purchased, or processed under a toll
agreement (i.e., an arrangement whereby one facility contracts with another to convert scrap,
according to specifications, for a pre-set fee) by secondary smelters, aluminum product
manufacturers, or others. An important consideration is that aluminum scrap recovery and
consumption data do not include measures of runaround scrap because this scrap is not traded in
the aluminum scrap market.
Old scrap includes any aluminum product recovered after its useful life. Major sources of
old scrap are used aluminum cans and utensils, old wire and cable, obsolete aircraft, and
aluminum engine and body parts from junked automobiles.
The method for processing scrap depends on the type of scrap handled. Figure 2-1
presents an overview of the methods used to process scrap aluminum.2
2.1.1.2 Relationship Between the Primary Aluminum and Secondary Aluminum Industries
For the purposes of this El A, the term "primary aluminum producer" refers to any of the
13 U.S. firms that produce aluminum from alumina/bauxite. It is important to note that these
firms may also include secondary aluminum processes, as defined by EPA, in their plant
operations. Unless otherwise noted for this EIA, the term "secondary aluminum producer" refers
to those firms whose sole purpose is to process aluminum scrap into molten aluminum or
aluminum ingot. Alternatively, the term "secondary aluminum industry" refers to all firms that
conduct secondary aluminum processes.
The secondary and primary aluminum industries are somewhat interrelated. There is an
overlap in marketing between the two segments of the industry—both in the purchasing of scrap
and in the selling of processed aluminum. In processed aluminum, primaries and secondaries
compete for molten metal sales in the deoxidizing market and in the alloy market—especially in
the low-copper alloys. Because secondary aluminum is usually cheaper than primary aluminum,
-------
secondary aluminum sells first. The price of secondary aluminum is usually lower than primary
aluminum due to lower production costs. Secondary production saves more than 95 percent of
the energy required to produce aluminum from bauxite because secondary recovery eliminates
the most electricity-intensive phases of production.3 The two industries can also directly
compete in the scrap purchasing market. However, the two industries also have a number of
common differences, including:
Raw materials. The primary producer uses bauxite, secondary uses scrap.
Prices. Primary and secondary prices can fluctuate somewhat independently of one
another because each has different factors of production. (However, the two often move
in the same direction—for instance, as primary prices go up, scrap usually becomes more
expensive as well.)
Plant sites. Primaries locate near plentiful, low-cost power sources since electricity is the
most expensive factor in producing primary aluminum. Secondaries cluster in areas
which provide them close proximity to supplies of scrap and their customers, such as the
heavily industrialized Midwestern States.4
2.1.1.3 Problems in Characterizing the Secondary Aluminum Industry
As defined for regulatory purposes, the secondary aluminum industry is a complex mix of
firms and entities that includes both large international conglomerates, as well as small
independent scrap dealers that shred aluminum cans before shipping them for further processing.
Unfortunately, the wide variety and scale of entities and the high level of integration complicate
efforts to compile data to characterize the industry.
-------
FIGURE
2-1. Methods for Processing Scrap Aluminum
Group I
cSheet and castings
Group II
Clippings
Group
Borings and turnings
Group IV
Residues
Crusher
Crusher
Dryer
_
Crusher
Highjron scrap
Magnetic
Separator
Screening W
Magnetic L
Separator i
Sweating
furnace
Reverberatory
furnace
Smelter
products
-------
A second problem is that data are often not available at the level of detail desired. The
economic and financial data that are often available are for the primary aluminum industry. Even
though all primary aluminum firms use new scrap in the production of aluminum, for the most
part, data are not available for determining what proportion, if any, of their operations are
devoted to secondary aluminum processes. For example, the available data characterizing
aluminum demand do not differentiate between primary and secondary aluminum sources of
metal and, as previously mentioned, data that are available do not include runaround scrap
consumption. (Note that hereafter, all data presented in this EIA do not include estimates of
runaround scrap.) Another example of the lack of the detailed data necessary to profile the
industry involves data that are provided by SIC. In SIC 3341. data do not break out aluminum
operations from other secondary nonferrous metal production operations, such as copper
recycling.
2.1.2 ORGANIZATION OF CHAPTER 2
Section 2.2 provides a characterization of the market structure of the secondary aluminum
industry. This section describes the facilities that may be affected by the MACT standards, the
distribution of production by firm, an assessment of market concentration and integration, and
the most current distribution of employment by firm and facility, as permitted by the available
data.
Section 2.3 focuses on the supply side of the industry, including historical data on
secondary aluminum recovery. The extent to which product substitutes are available for
secondary aluminum is included in this section. Foreign trade levels for the last few years are
presented, as are the determinants of the supply of secondary aluminum.
Section 2.4 presents historical consumption trends and the factors that determine the level
of demand for secondary aluminum. Historical price levels are also discussed in this section.
However, price elasticity of demand estimates available in the literature are reported in Chapter 3
-------
of this report.
Section 2.5 describes the financial conditions of firms in the industry. Revenue estimates
are presented for two of the major sectors (SIC's 3341 and 3353) in the industry. In addition,
financial data are presented for average firms in these sectors of the industry in three different
states of financial health—below average, average, and above average.
Section 2.6 discusses the outlook for secondary aluminum in terms of supply, demand,
and price levels. This discussion is based on a literature search for industry projections as well as
an assessment of the historical industry trends presented in previous chapters. Any changes in
economic conditions that may affect domestic supply and demand are identified.
2.2 MARKET STRUCTURE
An EIA requires that the facilities affected by a regulation be identified and classified by
some production factor or other quantifiable characteristic. Unfortunately, consistent and
complete data are unavailable to accurately characterize the firms and facilities that comprise the
secondary aluminum industry. Throughout this EIA, the word "firm" refers to the company or
producer, while facility or plant refers to the actual secondary aluminum processing site. Given
the lack of quantifiable data characterizing the entities in the industry, this chapter provides a
qualitative discussion of the industry. A brief review of the data that were collected in the
Secondary Aluminum/Aluminum Processing Follow-up Information Collection Request (ICR)
immediately follows the qualitative description of the industry.
The secondary aluminum industry has developed into a major market force in the
domestic aluminum industry. The recycling of scrap provides a source of aluminum that not
only helps the aluminum industry to maintain its growth, but also helps to conserve energy and
slow the depletion of bauxite resources.5
-------
Secondary aluminum is sold in a highly competitive market with numerous sellers, no
one of which is large enough to influence market price.6 For many applications, secondary
aluminum is comparable to primary aluminum, however, for some applications only primary
aluminum is employed. There is competition between the sectors for those grades of metal which
secondary smelters can produce.
The major aluminum producers have been very astute at recognizing the trend in the
secondary aluminum market and have adapted extremely well by incorporating the recycling
process into their own operations. All of the major primary producers now include secondary in
their total production, in addition to their primary operations. In fact, scrap recovery by primary
producers has been the fastest growing component of total secondary consumption in the U.S.
over the past 2 decades. Primary producers consumed nearly 50 percent of all secondary
aluminum in 1992, up from 32 percent in 1980, and only 16 percent in 1970.7
Figure 2-2 presents the overall flow of scrap aluminum by sector. The following are the
major sectors involved in the processing of scrap aluminum:8
• The primary aluminum producers are also part of the secondary industry. Unlike the
secondary smelters and non-integrated producers who have to go into the market to buy
scrap, primaries have a captive scrap supply, which they generate from their operations.
They sometimes also purchase scrap from manufacturers that they supply.
• • Secondary smelters are the traditional core of the secondary industry. In 1993, they
consumed approximately 36 percent of all the aluminum scrap generated in the United
States.
• Non-integrated fabricators, foundries, and chemical producers represent the second
major industrial group within the secondary industry. Since they are not affiliated with
the primaries, they are dependent on scrap for a good portion of their raw materials.
-------
These entities utilized about 22 percent of the total scrap consumed in 1993.9
• Dealers and brokers are also included in the secondary aluminum industry. The dealers
and brokers gather scrap from numerous sources, segregate the scrap by type, shred and
bale the scrap, and ship it in truckload or carload quantities to secondaries, fabricators, or
primaries.
Waste and scrap are generated at various stages during the production of primary
aluminum, the manufacture of aluminum mill and foundry products, and finishing of
industrial and consumer end products containing aluminum. The primary aluminum producers,
aluminum foundries, and independent aluminum fabricators recycle a portion of the waste and
scrap they generate. They also recycle scrap they purchase or receive on toll from others. The
manufacturers of industrial and consumer products usually sell or toll their aluminum scrap to
scrap dealers, to secondary smelters, or to their aluminum suppliers.
Most manufacturers do not have the space nor the time to store, segregate, and bale scrap.
As a result, most scrap is gathered by dealers, who then either sell to larger dealers or segregate
and bale the scrap themselves and then ship trailer loads or carloads to smelters or independent
fabricators.10 A dealer operates a collection plant in which he processes scrap received from
many sources, segregates by type, cleans contaminated scrap to a commercially-usable condition,
and packages it for shipment in carload or truckload quantities to an industrial consumer.1' The
dealer is the most important factor in the collection of scrap, and while there is no accurate data
on what percentage of scrap moves through dealers, it is thought to be quite high.12
10
-------
Manufacturing
Plants
Primary Aluminum
Producers
Scrap
Dealers
Secondary
Smelters
Total
Scrap
Supply
Imports
Primary
Aluminum
Producers
Foundries,
Fabricators
Sheet Mills
Chemical
Producers
g
For the purpose of this profile, this grouping refers to the 13 firms
that produce primary aluminum from alumina/bauxite (and may conduct
secondary aluminum processes at their plants as well).
b
Does not include runaround scrap.
FIGURE 2-2. FLOW OF ALUMINUM SCRAP
AND SECONDARY ALUMINUM IN THE U.S., 1991
11
-------
Dealer processing may involve sweating, cleaning, shearing, cutting, and/or grading.
Scrap is commonly identified according to its source. Sometimes the material is sorted as it is
delivered or picked up by the dealer. For example, a scrap source may use only one alloy; or it
may keep all scrap segregated, often earning a premium price or minimum quantity waiver for
this service.13
After the scrap is processed by the dealer, it follows one of many possible routes. The
aluminum may be sold to a broker who is an intermediary in the distribution process, dealing
usually in large quantity transactions between any type of scrap producer and scrap consumer
whether there may be a dealer, private industry, or government entities involved; or the scrap
may be directly exported by either a dealer or a broker; or it may be sold by any of the foregoing
directly to a scrap consumer.14
There are 13 primary aluminum producers in the United States. All participate in either
the collection or utilization of new aluminum scrap. They frequently purchase scrap from their
industrial customers or on a contract conversion basis.15 Over the last decade or so, major
primary aluminum producers have also operated used beverage can (UBC) recycling programs to
recover the increasing amounts of old aluminum scrap that are being generated by use of
aluminum containers. The tremendous increase in the amount and percentage of aluminum cans
recovered in the United States over the last decade has resulted in an upsurge of secondary
aluminum recovery by the primary producers.
Up until they were surpassed by the primary aluminum producers in 1987, the largest
consumers of aluminum scrap reaching the marketplace were the secondary aluminum producers.
Unlike scrap dealers who buy and sell many different metals, and diecasters who cast other
metals, the secondary aluminum smelter is totally dependent on one metal—aluminum. The
secondary producers are entirely dependent on scrap for raw material and cannot operate without
it. Because of this, scrap buyers for secondary aluminum producers occupy a position of vital
importance to their companies. The secondary smelters use aluminum scrap for at least 90
12
-------
percent of their raw manufacturing material.
16
Secondaries generally purchase scrap in trailerload--30,000 pounds (Ib)-and carload-
40,000 lb--quantities, but some buying may even be done in truckloads of 10,000 Ib. These
shipments are usually mixed lots-not necessarily one type of scrap. Buying is handled through
the smelter's own staffs, which keep in constant contact with the market.17
13
-------
The secondary producer does not, in general, use the most selective, higher purity
aluminum scrap except as an alloy "sweetener." A dealer, who may offer these more expensive
grades to primary producers, fabricators, or billet makers, risks rejection if the scrap does not
conform to specifications. If rejected, it is often downgraded to a smelter's specification and a
resulting lower price.18
Non-integrated fabricators, including foundries, extruders, chemical producers, and some
sheet mills producing building products comprise the balance of the aluminum scrap
consumers.19 These sectors of the industry are the most difficult to characterize due to the lack of
secondary aluminum data pertaining to their use of scrap.
2.2.1 AFFECTED FIRMS AND FACILITIES
The U.S. Department of Commerce's Bureau of the Census publishes the number of
companies reporting aluminum inventory data. For December 1994 they reported that there were
12 integrated companies with inventories of scrap (i.e., primary aluminum producers, which may
also produce mill products); 141 nonintegrated companies with scrap inventories (nonintegrated
is defined as companies that produce mill products in the U.S. and that are not affiliated with a
domestic primary ingot producer); and 25 smelters with aluminum scrap inventories (smelters are
defined by the Census as companies whose aluminum facilities are exclusively devoted to
producing specification ingot from scrap).20 Information obtained from the ICR shows that there
are 135 facilities that may be affected by this regulation.
The remainder of this section discusses the data that have been used in other publications
to describe two of the individual sectors-scrap dealers and secondary smelters—of the secondary
aluminum industry.
14
-------
2.2.1.1 Scrap Dealers
In 1985, the U.S. Bureau of Mines estimated that several hundred scrap dealers purchased
and collected scrap from industrial plants as well as scrap contained in discarded industrial or
commercial products, and that such scrap was then sold to the independent secondary smelters,
the integrated producers, and others, including about a dozen firms that used it to produce
aluminum chemicals and for other dissipative applications.21 Also in 1985, the Aluminum
Association estimated that aluminum scrap was commercially recovered from about 200
operational automobile shredders and from shredded white goods (major appliances) that were
also processed in these shredders in about 20 specialized residue processing facilities in the
U.S.22
2.2.1.2 Secondary Smelters
The five key products included in SIC 3341 are secondary aluminum, secondary precious
metals, secondary lead, secondary copper and secondary zinc. Of all the metal produced by the
secondary smelting industry, secondary aluminum represented the largest product category.
accounting for 38.5 percent of the industry's aggregate shipments in 1990.23
In the early 1970's, it was estimated that there were 72 full-time secondary aluminum
smelters operating 86 plants, and thousands of small foundries that occasionally processed small
quantities of scrap aluminum. At that time, the industry was dominated by small producers with
53 of the 86 plants producing less than 500 tons per month. Seventeen plants produced between
500 and 1,000 tons per month, while 13 plants produced between 1,000 and 3,000 tons per
month. Only three plants produced in the range of 3,000 to 7,000 tons per month.24 In 1982,
there were 76 secondary ingot plants in operation. By 1985, this number had decreased to 69.25
In 1992, the Aluminum Association estimated that there were 49 companies operating 68
secondary aluminum ingot plants.26 The following section presents the geographic break-down
of these plants in'1992.
15
-------
2.2.2 LOCATION AND EMPLOYMENT
Because transportation costs limit the effective distance over which firms buy and sell
locally, or at most regionally, there is a heavy concentration of secondary aluminum plants in
areas that are near scrap supplies or consumers of secondary aluminum. A cursory look at the
distribution of secondary smelters in the U.S. reveals a heavy concentration of smelters in the
automotive and appliance manufacturing areas of the country.27 Figure 2-3 presents the locations
of primary and secondary smelters in the U.S.28
Table 2-1 presents employment data for the secondary smelting of nonferrous metals
(SIC 3341) and the aluminum plate, sheet, and foil industries (SIC 3353). For SIC 3341, which
includes non-aluminum smelting firms, a significant decline in employment of approximately
35 percent occurred from 1982 to 1987. Between 1987 and 1990, employment in this industry
increased approximately 15 percent; the year-to-year change between 1990 and 1991 was a
decrease of about 10 percent. Over the 1982-1991 period, SIC 3353 has seen employment
decline by 12 percent.
Of the 14,700 people employed in the secondary smelting and refining industry (SIC
3341) in 1990, almost 40 percent were salaried employees, or those performing managerial,
administrative or technical duties. The balance of the industry's work force were production
workers. A typical secondary smelter and refining facility in 1989 employed 27 production
workers and 11 salaried employees. In terms of the number of people employed per
establishment, the secondary smelting and refining industry has been, historically, predominantly
populated by relatively small manufacturing facilities.29
2.2.3 MARKET CONCENTRATION
U.S. Bureau of Census publishes data that characterize market concentration for SIC
16
-------
3341 and SIC 3353. There does not appear to be a great amount of market concentration in SIC
3341. In 1992, the 4 largest companies in SIC 3341 accounted for 28 percent of total shipments,
the 8 largest for 41 percent, the 20 largest for 60 percent, and the 50 largest for 79 percent.
Figure 2-4 displays the market concentration in SIC 3341 for 1982, 1987. and 1992. Overall,
there has been a slight increase in market concentration between 1982 and 1992.30
Figure 2-5 presents the 1982, 1987, and 1992 market concentration data for SIC 3353.31
There is a substantial level of market concentration in this industry. Four major producers-
Alcoa, Alcan, Reynolds Metals, and Alumax-dominate the North American aluminum industry
and thus dominate the market for sheet, plate, and foil. Alcoa, Alcan, and Reynolds combined
have 78 percent of the key can market that accounts for about half of all sheet, plate, and foil
shipments.32 In 1992, the 4 largest companies in this SIC accounted for 68 percent of total
shipments, the 8 largest combined for 86 percent, the 20 largest for more than 99 percent, and the
50 largest for 100 percent, respectively. The concentration for the 4 largest companies decreased
from 74 percent in 1987 to 68 percent in 1992; the ratio for the 8 largest companies decreased
from 91 percent to 86 percent over the same period of time.
2.2.4 INTEGRATION
Integration is a measure of the control firms have over the product and factor markets for
its output. Figure 2-6 presents the 1994 and 1995 gross shipments of aluminum ingot (including
primary ingot) for integrated, nonintegrated, and smelting firms using the Bureau of the Census'
definitions for these three terms33. Again, the Census defines integrated firms as companies that
produce primary aluminum ingot (from alumina) and may also produce mill products;
nonintegrated producers are defined to be firms that produce mill products that are not affiliated
with a domestic primary ingot producer; smelters are defined as companies whose aluminum
facilities are exclusively devoted to producing specification ingot from scrap. These data show
17
-------
FIGURE 2-3. LOCATIONS OF PRIMARY AND SECONDARY SMELTERS IN THE U.S.
SOURCE: The Aluminum Association. Inc.
if Primary
m Secondary
18
-------
TABLE 2-1. EMPLOYMENT IN THE SECONDARY ALUMINUM INDUSTRY 34
(Thousands)
Year
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
SIC 3341
19.2
17.2
17.7
16.0
14.6
12.5
13.2
14.6
14.7
13.2
12.0P
11. 4P
10.9P
10.3P
9.7P
SIC 3353
21.0
21.9
21.0
20.8
20.4
19.8
19.5
19.3
18.7
18.4
18.1P
17.8P
17.4P
17.1P
16.7P
P = projections by editors
19
-------
FIGURE 2-
4. MARKET SHARES OF LARGEST FIRMS IN SIC 3341
100
{2 90
z
HI
80
70
60
50
40
u_
O
30
20
4 LARGEST
8 LARGEST 20 LARGEST
COMPANIES
1982 1987 1992
50 LARGEST
20
-------
FIGURE 2-5.
MARKET SHARES OF LARGEST FIRMS IN SIC 3353
100
c
CD
.9- 90
CO
0
80
70
60
4 LARGEST
8 LARGEST 20 LARGEST
Companies
50 LARGEST
1982 1987 1992
21
-------
that of the total ingot shipped in 1995, over 50 percent were from primary producers and 26
percent were from secondary producers. When compared to the 1994 data, this represents an
increase in the percentage of total gross shipments of ingot by integrated producers, at the
expense of nonintegrated firms. The integrated producers had the largest percentage of gross
shipments prior to 1992, at which time the shipments by nonintegrated producers increased to
take the lead. It is important to distinguish between vertical and horizontal integration and
assess the characteristics of each with respect to the secondary aluminum industry.
2.2.4.1 Vertical Integration
For the purposes of this El A, a vertically integrated producer (or supplier) is a firm which
either collects the scrap necessary to produce secondary aluminum (i.e., vertical integration
backward to suppliers), or manufactures products using secondary aluminum as an input (i.e.,
vertical integration forward to product markets).
Vertical integration in the aluminum industry is extensive. It includes backward
integration in the mining, refining and smelting of primary aluminum (including sheet ingot.
casting ingot, and extrusion billet), and forward integration to the production of semi fabricated
and fabricated products downstream, including sheet, plate, and foil offerings.
It is known that primary aluminum producers run thousands of UBC collection centers
throughout the U.S. It is also known that some aluminum fabricators (e.g., of can sheet metal)
are integrated backward to primary aluminum producers, some of which, as mentioned above,
collect UBC.
22
-------
FIGURE 2-6. GROSS SHIPMENTS OF ALUMINUM INGOT, 1994-1995
1994 1995
PERCENTAGE OF TOTAL SHIPMENTS
Integrated U Smelters H Nonintegrated
23
-------
The vast majority of secondary aluminum firms rely upon the demand for their metal
from other industry's firms. This derived-demand relationship is typified by the automobile
industry which is completely separate from the scrap dealer and secondary aluminum ingot
producer. In fact, for aluminum automobile products, the cast aluminum producer is often times
a separate entity located between the secondary aluminum industry and the final product market.
2.2.4.2 Horizontal Integration
Horizontal integration exists when a firm owns a number of different facilities, each of
which functions at the same stage of the production process. Horizontal integration is common
in the secondary aluminum industry. A number of companies own more than one facility.
According to data from the preliminary ICR, 33 of the firms own more than one secondary
aluminum facility, with Reynolds Metals Company owning 18 plants, Alcan Aluminum owning
16 facilities, Alumax owning 12 plants, Kaiser Aluminum owning 8 plants, and Southwire
Company and Aluminum Company of America owning 6 plants apiece.
2.2.5 TOTAL CAPACITY
Capacity for scrap production is a matter of managing collection and installing processing
equipment. Neither has very long lead times and, with the heightened importance of recycled
containers, an increased proportion of aluminum scrap will be returned to production channels
semiautomatically. Another portion of the scrap may require more elaborate separation
equipment than heretofore—particularly that which is potentially recoverable from municipal
waste streams and junked motor vehicles.35
The following are excerpts from a 1994 Bureau of Mines publication that detail recent
plant additions:
24
-------
U.S. Reduction Company reported that construction had begun on a new 27,220-
ton-per-year aluminum recycling plant in Tipton, Indiana. The company expected
the plant to reach full capacity by the second quarter of 1993. The plant
reportedly would provide specification liquid metal and ingot to aluminum
automotive diecasting and foundries.
Barmet Aluminum Corp. and Imco Recycling Inc., announced plans to construct a
$16 million 120,200-ton-per-year aluminum recycling facility near Barmet's
Urichsville, OH rolling mill. Under the reported agreement, Imco would have the
exclusive rights to recycle Barmet-purchased scrap at this plant for a period of 10
years. As the new plant comes on stream, Barmet's recycling plant at Urichsville
will be phased out. The plant, which reportedly recycles UBC's and dross and
would increase Imco's aluminum recycling capacity by 50 percent, was expected
to be fully operational by mid-1993.36
2.3 PRODUCTION, SHIPMENTS, AND CAPACITY UTILIZATION
Increased costs for energy, growing concerns over the siting of waste disposal sites, and
improvements in recycling technologies have provided the impetus for increased recycling rates.
Additionally, some of the increase in aluminum scrap recovery can be attributed to a changing
and growing end-use consumption pattern. Aluminum products developed for the construction,
transportation, and electrical industries tend to have a fairly long life and are slow to enter the
scrap supply stream. The emergence of the aluminum beverage can in the mid-1970's with a life
cycle of 3 months has added dramatically to the scrap aluminum supply.37
It is estimated that over 25 percent of the total "old" aluminum scrap potentially available
to the industry is recycled; on the other hand, almost 100 percent of the prompt industrial or
"new" scrap is currently being recycled.38
25
-------
2.3.1 PRODUCTION AND CAPACITY TREND
Historical data of the total supply of secondary aluminum are provided by the Aluminum
Association. It is important to note that the production and capacity data reported in this section
do not accurately capture captive production in the industry. Table 2-2 shows that secondary
aluminum supply has maintained approximately 25 to 30 percent of total supply over the 1984-
1994 period (the drop in the percentage that occurred from the early 1980's to the mid-1980's can
largely be attributed to a decline in energy prices over that period, which made primary
aluminum production relatively more competitive with secondary aluminum). In the 1983-1993
period, secondary recovery grew at a 6.4 percent annual rate, while primary production increased
at a 1.5 percent annual rate.39 While primary production in the U.S. has increased by about 430
thousand metric tons since 1970, secondary production has increased by more than triple that
amount, or 1.4 million metric tons.40
Table 2-3 shows that old scrap recovery, at 1,632 thousand metric tons in 1993, was up
approximately 2 percent from 1992, then decreased by about 8 percent in 1994. The percentage
of the total secondary supply from old scrap has increased from 46 percent in 1983 to 55 percent
in 1993. Between 1993 and 1997, old scrap decreased as a percentage of total secondary supply
from 55 percent to 41 percent.
From 1982 to 1993, aluminum can reclamation programs have returned over 18 billion
pounds of aluminum beverage cans to the supply stream. Most of this scrap was returned to
domestic primary producers for production of can sheet. To a lesser degree, UBC's have also
been used by secondary aluminum producers in the manufacturing of casting alloys, and some
have been exported to foreign countries for conversion there into can sheet. The large-scale
aluminum can recovery programs of the major primary aluminum producers have substantially
added to the rate of aluminum recycling from old scrap—UBC's share of total old scrap has
doubled since 1975.41 Can recycling has grown into a very real component of the U.S. supply
picture, accounting for more than 10 percent of total aluminum supply consumed in 1992.42
26
-------
TABLE 2-2. SECONDARY ALUMINUM SUPPLY IN RELATION TO TOTAL ALUMINUM
SUPPLY: 1984-1994 (in thousands of metric tons)4344
YEAR
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
TOTAL SUPPLY*
7,235
6,594
6,655
7,035
7,534
7,437
7,863
7,805
8,371
8,966
9,514
SECONDARY
ALUMINUM SUPPLY
1,760
1,762
1,773
1,986
2,122
2,054
2,393
2,286
2,756
2,944
3,080
RATIO OF
SECONDARY
ALUMINUM
SUPPLY TO TOTAL
0.24
0.27
0.27
0.28
0.28
0.28
0.30
0.29
0.33
0.33
0.32
* Total supply = domestic primary production + imports of ingot and mill products + metal
recovered from scrap (i.e., secondary aluminum).
27
-------
Table 2-4 presents the trends in aluminum can recovery between 1983 and 1994. Over
this time period, the number of aluminum cans collected each year has increased, except for a
slight decline in the recovery rate of UBC's in 1991 from the previous high of almost 64 percent
in 1990. Industry analysts believe that low UBC prices in 1991 caused can stockpiling by UBC
collectors. The percentage of aluminum cans collected increased by more than 8 percent year-to-
year in 1992 for a record high of almost 68 percent. The recovery rate was somewhat lower in
1993 and 1994, but was still above the 1992 rate. Discussion of price trends for UBC and other
types of aluminum scrap are provided in section 2.4.3.
2.3.2 COMPETITION AND SUBSTITUTES
Substitutes for aluminum include: plastics, magnesium, steel, copper, wood, titanium,
graphite, paper, and fiber epoxies. Copper can replace aluminum in electrical applications;
magnesium, titanium, and steel can substitute for aluminum in construction; glass, plastics,
paper, and steel can substitute for aluminum in packaging. However, substitution of one of the
above materials for aluminum in an ongoing manufacturing process is seldom possible. The
substitute must offer advantages in long-term costs and performance to justify the cost of change.
28
-------
TABLE 2-3. TRENDS IN SECONDARY ALUMINUM RECOVERY:
1982-19974546
(in thousands of metric tons)
YEAR
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
TOTAL
SECONDARY
RECOVERY
1,773
1,760
1,762
1,773
1,986
2,122
2,054
2,393
2,286
2,756
2,944
3,080
3,190
3,310
3,690
FROM NEW
SCRAP
953
935
912
989
1,134
1,077
1,043
1,034
969
1,144
1,312
1,580
1,595
1.772
2,177
FROM OLD
SCRAP
820
825
850
784
852
1,045
1,011
1,359
1,317
1,612
1,632
1,500
1.595
1.538
1,513
RATIO OF
OLD SCRAP
TO TOTAL
0.46
0.47
0.48
0.44
0.43
0.49
0.49
0.57
0.58
0.58
0.55
0.49
0.50
0.55
0.49
29
-------
TABLE 2-4. TRENDS IN U.S. ALUMINUM CAN RECLAMATION: 1983-199447
YEAR
1984
1985
1986
1987b
1988
1989
1990
1991
1992
1993
1994
POUNDS OF
ALUMINUM CANS
COLLECTED (IN
MILLIONS)
1,226
1,245
1,233
1,335
1,505
1,688
1,934
1,969
2,142
2,015
2,149
NO. OF ALUMINUM
CANS COLLECTED
(IN BILLIONS)
31.9
33.1
33.3
36.6
42.5
49.4
55.0
56.8
62.7
59.5
64.7
PERCENT OF
ALUMINUM CANS
COLLECTED3
52.8
51.0
48.7
50.5
54.6
60.8
63.6
62.4
67.9
63.1
65.4
"Based on aluminum beverage can shipment data—lagged on quarter.
bBeginning in 1987, UBC data includes estimates of exported can scrap.
30
-------
The strong growth of aluminum markets over the past 30 years has occurred largely at the
expense of other materials. For example, aluminum has replaced a significant part of the steel
and glass beverage container market; has penetrated the wood and steel markets in residential,
industrial, and commercial construction; and is increasing its share in the castings markets in
competition with zinc and iron. It has also replaced copper in the high-voltage transmission of
electrical power. With the exception of copper, also a highly strategic commodity, aluminum
could be replaced by many of the original materials should it become in short supply.48
The enormous strides that aluminum has made in displacing steel and other materials in
the container/packaging, automotive, aerospace, and construction markets demonstrates the
substantial investment the industry has made in research and development.49
As alluded to earlier, primary aluminum can often substitute for secondary aluminum,
though usually at a higher price. However, there are applications where substitution of
secondary aluminum for higher purity metal is not acceptable, such as in automobile and
appliance trim applications.50
2.3.3 IMPORTS AND EXPORTS
Aluminum scrap is also traded in the international marketplace. Price and shipping costs
usually determine whether scrap is sold in domestic or international markets. U.S. trade in
aluminum scrap has grown dramatically over the past 30 years. Most of the scrap shipped into
the U.S. comes from Canada. Japan has been the major recipient of U.S. scrap exports since the
mid-1970's.
Table 2-5 shows that in 1990, exports of aluminum scrap began to decline after reaching
a 5-year high of 586,000 metric tons in 1989. This downward trend in scrap exports continued
until 1994, when there was a 30 percent increase from 214,000 metric tons to 307,000 metric
31
-------
tons. 1993 was the first year since 1982 that the U.S. had a trade deficit in scrap aluminum.
Japan continues to be the principal destination of aluminum scrap exported, accounting for
almost one-half of total 1993 exports.
As can be seen in Table 2-5, the amount of scrap that was imported rose from about 7
percent of total supply in 1984 to 12 percent in 1995. Scrap imports in 1991 decreased
approximately 18 percent to 218,000 metric tons from a 1990 level of 265,000 metric tons.
Imports had steadily increased over the preceding 5 years and have increased each of the 3
subsequent years. Canada remains the major scrap importer to the United States, supplying
almost 58 percent of U.S. total aluminum scrap imports in 1993.51
According to the Aluminum Association, exports of sheet and plate rose about 7 percent
in 1993 to 1.178 billion pounds; foil shipments increased approximately 17 percent to 83 million
pounds. Sheet and plate imports to the U.S. in 1993 rose 8 percent to 813 million pounds. Foil
imports increased 20 percent in 1993 to 61 million pounds after a decrease of 8 percent between
1991 and 1992.52
32
-------
TABLE 2-5. TRENDS IN ALUMINUM SCRAP FOREIGN TRADE:
1983-1995 (thousand metric tons)5354
YEAR
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
TOTAL U.S.
SCRAP
CONSUMPTION3
2,010
1,978
1,986
2,204
2,348
2,280
2,561
2,456
2,960
3,186
3,336
3,480
IMPORTS
OF SCRAP"
145
131
165
191
207
223
265
218
272
312
390
419
EXPORTS
OF SCRAP"
258
375
347
368
487
586
538
464
298
214
307
430
NET
EXPORTS
113
244
182
177
280
363
273
246
26
-98
-83
-11
"As reported (rather than as estimated).
"Aluminum scrap includes both scrap and dross.
33
-------
2.3.4 SUPPLY DETERMINANTS
The distinction between new and old scrap was first described in section 2.1.1.1. The
amount of new scrap supplied tends to change with the changes in fabrication loss attendant from
changing applications of aluminum; old scrap supply tends to vary according to the changing
lifetimes of different applications, but is also responsive to prices and governmental regulation.55
Some in the aluminum industry believe that the recovery of aluminum cans is coming
close to reaching its maximum levels. However, the companies involved in the secondary
aluminum industry say that they are not about to abandon any plans for the future growth of their
field, and increasingly are turning to other sources of used aluminum to sustain their successes.56
Some believe that the increase in aluminum can recycling, which is viewed as part of a general
recycling movement in this country, can spread to other aluminum products such as lawn chairs
and building materials.
Many in the industry expect to see an increase in plants built to improve the scrap
recovery and recycling process because the major obstacle to recovering scrap is that, except for
UBC's, it usually does not come in a condition that is conducive to aluminum recovery. Often
the presence of other alloys and post-production elements such as paint can make it impossible or
uneconomical to recover useful aluminum. Many see that these plants will be needed in order
for growth to continue in the secondary aluminum market.
In addition to the obstacles sometimes found when recovering aluminum from scrap,
scrap is also subject to the effects of the free market as shown by the drop in aluminum can
recycling rate between 1990 and 1991. It all boils down to competition, according to Michel
Primeau, a recycling official with Alcan, one of the big three aluminum can sheet producers
along with Alcoa and Reynolds. "At Alcan we grab as many used cans as possible. The
company that has access to the most cans has a competitive advantage."57
34
-------
Another problem is its seasonal nature—beverage can use is slack during the winter and
high during the summer. However, it is relatively easy for suppliers to stockpile scrap to offset
the seasonal highs and lows.
Because most processors of aluminum scrap do not recover it themselves directly from
the consumer market, it has become a commodity collected and sold by dealers who attempt to
set the most favorable prices by affecting the supply of scrap on the open market. John
Dickinson, of the Aluminum Association notes:
I tend to think supply and demand may be the more driving force . . as demand
for scrap has increased in recent years and more companies are reliant on its use,
the price has started to increase even though recovery methods are more efficient
than they used to be and it's easier to obtain certain types of scrap.58
One factor that may be helping the aluminum recycling rate is that aluminum can and
sheet makers have invested heavily in recycling over the years, which gives them a strong
incentive to use secondary materials. Industry has established a recycling infrastructure that
includes 10,000 buyback centers nationally.
In the nine bottle bill States, up to 90 percent of aluminum cans are recycled, according to
Ralph Cheek, president of Imco Recycling Inc., a major processor of UBC's. He estimates the
other States average roughly about 50 percent. The rapid expansion of curbside programs across
the country (in 1991, nearly 4,000 were reported, an increase of 40 percent over 1990) is bound
to have a positive impact on supply, and should lower the cost of obtaining UBC scrap (assuming
that dealers/companies do not hold onto their UBC's until the price of UBC's increases).59
Recent developments in the industry related to supply include a promotional effort by
Aluminum Company of America (Alcoa) and Reynolds Metals Company to increase the
recycling of aluminum foil wrap and formed aluminum containers (such as those used for frozen
35
-------
entrees, baking items, and deli trays).60
2.3.5 CAPACITY EXPANSION
Little information was located that projected secondary aluminum plant capacities. The
following excerpts provide some general background on capacity that may affect capacity levels
in the industry.
Can reclamation and remelt facilities are being expanded in lieu of building new
primary aluminum smelters. This added remelt capacity can be built in half the
time and at one-tenth the cost of primary facilities.61
... the highest reported recycling rate in the world is Ontario's 88 percent and in
Europe, Sweden's 85 percent. Thus, aluminum can reclamation in the U.S. has
the potential to sustain its steady growth.62
Imco Recycling Inc. plans to increase capacity at its UBC recycling plant in
Urichsville, Ohio. The $2.5 million expansion was expected to increase annual
capacity at the plant by 25 percent.
Alcan Aluminum Corp. announced plans to more than triple the UBC processing
capacity at its Oswego, New York rolling complex. Upon completion of the $23
million expansion project, the recycling unit reportedly would have the capacity
to recycle 5 billion UBC's (75,000 tons) per year.63
36
-------
2.4 DEMAND AND END USE MARKETS
This chapter presents a discussion of secondary aluminum demand. The demand for
secondary aluminum is derived from the demand for the manufactured products which use it as a
raw material. Therefore, the level of secondary aluminum demand is determined by the market
conditions present in the industries that use secondary aluminum as an input into their
manufacturing process. Unfortunately, data characterizing the end uses of secondary aluminum
are lacking. In particular, descriptions of the use of aluminum in product markets do not
differentiate primary metal demand from secondary metal demand.
2.4.1 CONSUMPTION TRENDS
Figure 2-7 presents the historical trend in demand for aluminum scrap over the last
decade.64 This figure shows that the percentage of total consumption of aluminum scrap by
secondary smelters declined from approximately 44 percent in 1983 to slightly greater than 33
percent in 1993. It is likely that much of this decline is attributable to the increase in UBC
recycling activity from the primary producers.
Figure 2-8 displays the consumption in 1994 of secondary aluminum scrap by consumer
class and by type of scrap.65 This figure shows that in 1994 secondary smelters consumed a
higher proportion of their total scrap in the form of new scrap. Prior to this, the highest
percentage of scrap consumption by secondary smelters had been old non-UBC scrap. A higher
percentage of old scrap than new scrap is consumed by primary producers, foundries, fabricators.
and chemical plants. Of this old scrap, a significantly higher percentage comes from aluminum
cans than for the secondary smelters.
37
-------
FIGURE 2-7. TRENDS IN DEMAND FOR ALUMINUM SCRAP
1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994
Year
Secondary
Smelters
Integrated
Producers
'Others
*M
Others" includes fabricators, foundries, and chemical
producers
38
-------
FIGURE 2-8. CONSUMPTION OF SCRAP ALUMINUM BY CONSUMER CLASS, 1994
100
Secondary
Other*
New ED Old Non-UBC D Old UBC
Primary, Foundry, Fabricator, Chemical Plant
39
-------
2.4.2 DEMAND DETERMINANTS
There is a lack of specific information on the end-use product markets for secondary
aluminum. It is known that a significant proportion of secondary aluminum comes in the form of
aluminum UBC's, and that currently aluminum holds 96 percent of the beverage can market, and
76 percent of the total beer and soft drink market-a complete reversal from 20 years ago, when
steel cans and glass bottles dominated the market.66 It is also known that many of these
aluminum UBC's are used in the production of additional can stock, although figures were not
available to characterize the percentage use of UBC's for this purpose. Given the paucity of data
on the specific products where secondary aluminum is a significant input, the following section
describes the demand determinants for aluminum in general (both primary and secondary). This
discussion is followed by a brief description of the products that are known to use secondary
aluminum.
2.4.2.1 Primary and Secondary Aluminum
Figure 2-9 displays the breakdown of aluminum consuming sectors in 1994 (the
consumption of aluminum is not broken down between primary and secondary aluminum).67
Important end-use sectors for aluminum include packaging, construction, and transportation. In
1994, these three end-use sectors accounted for 64 percent of domestic industry shipments as
compared to 72 percent in 1992. The transportation industry, with almost 25 percent of the
aluminum consumption, was the largest consumer. The automotive sector has been increasing
their use of aluminum in recent years and is expected to continue to do so in the future. This was
the first year since the mid-70's that the container and packaging industry did not dominate the
industry. They consumed approximately 24 percent of the total. The building and construction
industry accounted for approximately 15 percent of consumption. Approximately 7 percent of
the total aluminum consumed in end products was used in the electrical and communications
40
-------
FIGURE 2-9. ALUMINUM CONSUMPTION BY SECTOR, 1994
Transportation
Bldg./Const.
Containers/Packaging
Other
Machinery
Consumer Durables
Electrical
41
-------
industries, about 7 percent went into consumer durable goods, including refrigerators, air
conditioners, washing machines, and other appliances. Aluminum use for industrial equipment
and machinery was about 6 percent of consumption.
Over time, technical developments, structural changes within sectors, and competition
from other materials will account for changes in the intensity of aluminum use in these end use
sectors. In the building and construction sectors, for example, the main demand determinant is
new construction activity, as well as the price of substitute materials, such as steel. The
accelerated use of aluminum for doors and windows in residential housing is attributed to higher
levels of new housing construction in recent years. In the transportation sector, demand is
derived from the production of motor vehicles and parts. Aluminum's substitution for heavier
materials could result in higher demand levels in future years.68 In the electrical sector, demand
is derived from the use of aluminum in household appliances. Because of consumer resistance
and local electrical code restrictions based on poor performance in the past, the future use of
aluminum is not projected to be high in this sector. In the packaging sector, demand is based on
the demand for food and food products. The increased use of convenience foods will be
beneficial in this sector.
2.4.2.2 Secondary Aluminum
Aluminum scrap is not used to any appreciable extent for aluminum extrusions, but
largely goes to aluminum casting. The product mix in 1994 shows that about 61 percent of
secondary aluminum alloys produced are used in the production of die cast alloys, with most of
the remainder going for sand and permanent mold casting.69 Some secondary metal is also used
for deoxidizing in processing steel, and in aluminum wrought products and extrusion billets.
42
-------
Secondaries sell primarily 15-lb and 30-lb ingots to the casting industry, with the 30-lb
size being the most widely used. Some casters, however, buy much larger ingots. The ingots
have several notches in them, which permits the caster to divide each ingot into smaller
segments. Sales policies may vary from company to company, but contracts between secondary
smelters and foundries normally run no longer than 3 months at a time.70 Because casters
generally keep only a one- or two-week supply of ingot on hand, deliveries from the smelter to
the foundry are frequent.
There is competition between secondaries and primaries in hot metal and in the secondary
ingot market. Secondaries are used for most casting alloy ingot because their prices are normally
lower than primaries' casting alloy ingot. They are also preferred in many cases because they
will often do quite a bit of custom tailoring of alloys for specific customers. However, primary
producers generally are called on when a customer desires a low-iron, low-zinc, or low-
manganese alloy. These three metallics are more difficult to dilute. Primaries also occasionally
get the business when a customer wants a slightly greater degree of purity. In most instances,
though, secondary and primary alloy ingot have virtually equal purities.
The presence of several unwanted metals in aluminum scrap tends to limit its use for
producing wrought alloys. The amount of dilution with primary aluminum or high purity scrap
(such as electrical conductor wire) required to bring the scrap to a specific wrought alloy
composition is in direct proportion to the amount by which a given contaminant exceeds the
specification.
Because die-cast alloys contain silicon, zinc, and copper, and auto-shredder scrap
contains high levels of other-metal "impurities," (generally zinc and copper), aluminum now
recovered from autos is more generally suited to the production of specification casting ingot
43
-------
than for use in the production of wrought alloys. Comparison of the non-automotive alloys show
that even when mixed they are essentially compatible for many common end-use applications. '
2.4.3 PRICES
Given the fact that the secondary aluminum industry as defined by EPA consists of all the
entities that collect, process, and make semifabricated products from scrap aluminum, there are
several product prices of interest to the industry. First of all, the price of scrap is important to
scrap dealers who must be able to make a profit from the revenue that scrap sales bring;
similarly, the price of secondary ingot must provide a margin over the costs to produce the ingot.
Finally, the price of semifabricated aluminum products must also allow foundries and fabricators
to make a profit or else they will eventually withdraw from the industry. The following
discusses the data that were readily available on the price of aluminum scrap and secondary
ingot.
In October 1992, the London Metal Exchange began 3-month trading on its new
secondary aluminum alloy ingot contract. Cash trading began in January 1993.72 As the scrap
market becomes more developed, the pricing of scrap has become more efficient. As scrap
dealers become more consolidated, for example, their ability to find inventories and thus make
more market savvy decisions has been enhanced. In past periods of rising prices, secondary
operations have normally benefitted from a lag between scrap and ingot prices. Because of the
scrap markets increased efficiencies, much of that lag has been largely eliminated.73
Table 2-6 presents the primary ingot, number 380 secondary ingot, cast aluminum scrap,
and scrap aluminum clippings prices for the last several years. These data, which are unadjusted
for inflation, show that the price of scrap and secondary aluminum ingot has declined over the
last few years after peaking in 1988. Further discussion of these prices, including the reasons for
their recent declines, is provided in the following sections.
44
-------
TABLE 2-6. TRENDS IN SECONDARY ALUMINUM AVERAGE PRICES,
1985-1997 (in cents per pound)74 7576-7778-798081
Year
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
No. 380
Secondary
Aluminum
Ingot-3% Zn
57.0
59.8
71.4
98.1
89.9
76.2
65.2
64.6
64.8*
74.7**
80.5
67.3
75.5
Cast
Aluminum
Scrap
(Midwest)
24.5
25.5
35.8
52.1
45.9
35.9
24.9
22.7
20.0
N.A.
N.A.
N.A.
N.A.
Used
Beverage
Can Scrap
33.8
38.1
49.7
69.0
63.0
49.7
40.3
40.2
36.7
53.1
N.A.
N.A.
N.A.
Scrap
Aluminum
Clipping
(Midwest)
32.4
37.1
49.5
74.3
63.1
49.8
36.0
34.0
32.0
N.A.
N.A.
N.A.
N.A.
Note: Primary aluminum ingot and number 380 secondary aluminum ingot (a popular type of secondary ingot)
should not be compared as pure substitutes since number 380 secondary aluminum ingot includes alloying agents
such as zinc and copper, which are not included in the primary ingot.
*First seven months of 1993. In mid-July, American Metal Market discontinued the publication of secondary
aluminum ingot prices and substituted an indicator price series.
**The prices of No. 380 ingot (3% zinc) for 1994 through 1997 are from Platt's Metals Week and cannot be directly
compared to the prices shown for 1985 through 1993, which are from American Metals Market.
45
-------
2.4.3.1 Scrap Prices
The price of aluminum scrap is affected by the fact that such scrap is largely a substitute
for alumina consumption (in the production of primary aluminum). Scrap prices should,
therefore, follow approximately the same relative trend as alumina prices-modified slightly
downward by virtue of the fact that, for direct users of scrap, the alternative is the more slowly
changing price of refined aluminum.82 Scrap prices are also affected somewhat by the degree of
recycling activity in the U.S. For example, if more States adopt mandatory beverage deposit
legislation, then the price of scrap aluminum should be affected downward somewhat due to the
increase in supply of UBC's.
Over the past few years, the cost differential between primary aluminum and scrap has
narrowed considerably. In 1988, the price for virgin (primary) aluminum hit an all-time high,
averaging about Sl.lO/pound. In that year, the price difference between primary and UBC was
about 40 cents/pound, a margin which made it much cheaper for can sheet makers to recycle than
use virgin material. After that, the price of primary aluminum declined, aided by the former
Soviet Union dumping aluminum onto the world markets. By late 1992, the price for primary
aluminum had dropped to 57 cents/pound-almost half of what it was in 1988. At that time, the
difference in price between primary aluminum and UBC's stood at about 17 cents/pound. In
1994, the price of both primary aluminum and UBC's rose and the price margin decreased to
approximately 14 cents/pound.83 At that margin, recycling is less profitable due to the necessary
decontamination and remelting process, which costs money and results in UBC weight loss.84
2.4.3.2 Secondary Aluminum Ingot Prices
Secondary aluminum ingot prices, as quoted by American Metal Market, have followed
the trend of scrap aluminum prices over the last several years. The year end 1993 and 1994
prices for selected secondary aluminum ingots are presented in Table 2-7.
46
-------
TABLE 2-7. YEAR-END PRICES FOR A SAMPLING OF SECONDARY
ALUMINUM INGOT, 1993-199485-86
Type of Ingot
Alloy 3 19
Alloy 360 (0.6% copper content)
Alloy 380 (1% zinc content)
Alloy 413 (0.6% copper content)
Price (cents per pound)
1993 1994
64.52
67.00
61.96
67.09
102.30
102.79
98.98
102.51
47
-------
2.5 FINANCIAL DATA
Although the economic impact analysis will be conducted on a market level, the
examination of firm-level revenue and profitability trends are useful as a preliminary indicator of
the baseline conditions of affected firms in the industry. Firm-level data also provide an
indication of the financial resources available to affected firms and the ability of their resources
to cover the increased compliance costs associated with the MACT standards.
2.5.1 REVENUES
Dun and Bradstreet (D&B) collects the financial statements of both public and privately
owned U.S. firms in all size ranges and most lines of business. These data are then compiled by
SIC code numbers into "typical" balance sheets and financial statements for firms in that SIC.
For example, in 1992. the median firm in D&B's database for SIC 3341 (secondary smelting of
nonferrous metals) had revenues of almost $3.2 million; similarly, for SIC 3353 (aluminum
sheet, plate, and foil), the typical firm had sales of nearly $1.4 million. In 1994. the revenues for
the median firm in SIC 3341 were almost $5.8 million, and in SIC 3353. a little over
$600.000.87-88 D&B also publishes a break-down of sales by SIC into 7 sales range categories.
Figure 2-10 presents the percentage of firms with either primary or secondary classifications in a
particular SIC (firms were classified into SIC's on a secondary capacity if at least 10 percent of
the total revenue from that business was in a particular SIC).89 These data show that there is a
higher percentage of firms with sales over $5 million in SIC 3353 than in SIC 3341, probably
due to a greater degree of integration in the aluminum, sheet, and plate sector of the industry.
A second source of firm-level revenue data is published by Gale Research Incorporated. This
source provides a list of the largest companies categorized by SIC code. Unfortunately, for SIC
3341, it is not possible to differentiate the aluminum companies from the other nonferrous metals
companies, and for SIC 3353, there is no way to distinguish which companies use secondary
aluminum in their production process.
48
-------
FIGURE 2-10. PERCENTAGE OF TOTAL FIRMS IN GIVEN SALES RANGES
CD
HI
Q.
100
90
80
70
60
50
40
30
20
10
0
1994
SIC 3341
;.-:.*. -, r?m-e*M<(Smt$£m<.
SIC 3353
$1-49k D$50-99k
$100-249kB$250-499k
$500-999k E3 $1,000-4,999k D $5,000k+
49
-------
2.5.2 CAPITAL FORMATION
Gale Research Incorporated publishes capital expenditure data for SIC's 3341 and 3353.
although there is no way of knowing how much of this activity is due to outlays related to
secondary aluminum operations. Table 2-8 presents the trends in capital expenditures for these
SIC's over the 1982-1990 period. These data indicate that levels of capital investment are much
higher in the aluminum plate, sheet, and foil industry than in the secondary nonferrous smelting
industry, and that while this latter industry has seen capital investment fluctuate over the time
period, the aluminum plate, sheet, and foil industry has seen capital investment increase fairly
steadily over the period. This increase by SIC 3353 may be a result of increased investments to
reduce the amount of aluminum needed to manufacture a can, and perhaps increased secondary
aluminum recovery investments.
Because of the relatively high energy costs that U.S. primary aluminum companies must
pay. few expect to see any additional capital investment in primary smelters in the near future.
Instead of investing capital in smelters, some companies are turning instead to perfecting the
scrap recovery processes. One such firm, Nichols-Homeshield Casting, recently built a new $65
million plant in Davenport, Iowa, to handle scrap recovery and recycling.90 Investments in
recycling infrastructure are a bargain when compared to the cost of a new smelter required for
virgin aluminum.91
2.5.3 PROFITABILITY
A recent article noted that because of the tight margin that exists between the price of
UBC's and primary aluminum, recycling is only profitable now at efficient plants with state of
the art technology.92 According to conventional wisdom in the industry, under standard
50
-------
TABLE 2-8. CAPITAL INVESTMENT IN SIC'S 3341 AND 335393
(in millions of dollars)3
YEAR
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
SIC 3341
146.4
54.4
116.7
103.0
58.2
62.6
67.1
108.3
103.9
72.2
73.7P
70.9P
68.1P
65.2P
62.4P
SIC 3353
260.4
296.9
359.3
348.0
439.4
439.2
524.0
551.9
681.3
567.2
678.2P
720.3P
762.3P
804.4P
846.5P
"Nominal dollars
P = Projections by editors
51
-------
operating conditions-a recycling company will make no money for ten months, and then have a
month or so of high financial return.94 Companies tend to streamline their operations by closing
smaller plants and enlarging others.
2.5.4 BALANCE SHEETS AND INCOME STATEMENTS
The only data that have been compiled to date characterizing the industry's balance sheets
and income statements is from D&B's Industry Norms and Kev Business Ratios. This source has
the limitation of providing data for all companies in a particular SIC, therefore, not
disaggregating companies that are involved in secondary aluminum operations from those that
are not. Table 2-9 displays some of the key ratios that can be calculated from the typical balance
sheets and income statements of all firms classified in SIC's 3341 and 3353. It is important to
note that none of the lower quartile firms show a negative ratio for any of these key ratios.
2.6 MARKET OUTLOOK
The previous chapters indicate that the secondary aluminum industry is currently
characterized as having both significant problems, such as the tight profit margins due to the
former Soviet Union's dumping of aluminum on the world market, and notable cause for
optimism, such as the increased environmental consciousness on the part of U.S. businesses and
consumers. This chapter presents the supply, demand, and price projections that are available from the
literature as well as analysis of recent trends in the industry. These projections are important because the
economic impact analysis of the MACT standards is conducted for the fifth year following promulgation
of the regulation.
2.6.1 PRODUCT DEMAND
The U.S. Bureau of Mines predicts that both domestic and world secondary aluminum
52
-------
TABLE 2-9. KEY BUSINESS RATIOS FOR SIC'S 3341 AND 3353 FOR 199295
BUSINESS RATIO
Current
(total current assets divided by
total current liabilities)
Assets to sales
(total assets divided by annual
net sales)
Return on sales
(net profit after taxes divided
by annual net sales)
Return on assets
(net profit after taxes divided
by total assets)
MEASURES
Solvency
Efficiency of
use of assets
Profitability
Profitability
SIC 3341
UQ
3.1
19.1
4.0
10.7
M
1.5
33.6
1.8
5.5
LQ
1.2
51.5
0.4
1.2
SIC 3353
UQ
2.9
33.6
7.9
7.2
M
1.7
52.9
2.1
2.9
LQ
1.2
11
5.7
0.6
0.6
Note: UQ - represents the ratio for the firm whose ratio falls at the mid-point between the ratios for the
median firm in the industry and the firm with the best ratio in the industry; .
— represents the ratio for the firm whose ratio falls at the mid-point of all ratios for this industry;
and
LQ- represents the ratio for the firm whose ratio falls at the mid-point between the ratios for the
median firm in the industry and the firm with the worst ratio in the industry.
53
-------
industries should continue to expand. For the short term, they view the increasing acceptance of
aluminum beverage cans in foreign countries leading to a more rapid growth rate in secondary
aluminum recovery in overseas markets than the United States where the aluminum beverage can
already dominates the market. However, as more and more aluminum is used in products with
longer life cycles, such as automobiles, they expect that more scrap will enter the market for
recovery, increasing secondary production levels in both the domestic and world markets of the
future.96 In general, it is expected that the average annual increase in U.S. secondary aluminum
demand through the year 2000 will be greater than that in the primary aluminum industry.
Industry analysts believe that a call for additional weight reduction in automobiles will
translate into a substantial increase in aluminum use for vehicles, particularly for castings and
perhaps for extrusions. Although plastics will be a direct competitor of aluminum for this
business, one industry analyst notes that a significant advantage that aluminum currently has is
that it is recyclable, whereas many types of plastics are not. Projections for the amount of
aluminum per car for the year 2000 center around 200 pounds (up from an estimated 160 pounds
per car). One industry analyst projects that the initial increase in the amount of aluminum going
into cars will come from the primary sector. This he believes because once aluminum goes into a
car, it will not be recycled for 10 to 15 years. One industry official estimates that approximately
two-thirds of all the aluminum going into the automobile industry currently is secondary
aluminum, but expects to see that figure drop to less than one-third by the year 2000, and rebound
back up to better than three-fourths by 2010.97
The three largest domestic automakers, Chrysler Corporation, Ford Motor Co., and GM.
announced the formation of the Vehicle Recycling Partnership (VRP) to research and promote the
recovery and reuse of materials from junked cars. According to the companies, the partnership
was expected to propose methods for increasing the amount of recycled materials used in autos
and trucks as well as recovering more from the vehicles. The automakers also stated that the
partnership was expected to look into developing industry guidelines in such areas as materials
selection and compatibility, bonding methods, and materials, painting, and design for
54
-------
disassembly. Research in these areas reportedly would be carried out in the laboratories of the
three automakers, and by other industries, universities, and research institutes.98
General Motors Corporation has seen its annual aluminum consumption rise by around 40
percent over the last five years. About half the total consists of outside purchased castings made
from secondary aluminum, plus another 300 million Ibs of secondary or scrap purchased for GM's
in-house casting operations, and close to 400 million Ibs in the form of aluminum wheels,
wrought raw material and fabricated products."
2.6.2 CAPACITY DEMAND
The large increase in the use of scrap is due primarily to the recycling of aluminum
beverage containers. This increase in the use of scrap has placed greater demands on smelters to
improve quality. As the demand for aluminum scrap and the use of imported ingot increases, the
demand for aluminum melting capacity will increase.
2.6.3 RAW MATERIAL OR PROCESS CHANGES
Any secondary aluminum industry raw material or process changes that may take place in
the near future could affect the amount of secondary aluminum consumed in the U.S. In addition.
any technological changes in the use of secondary aluminum's substitutes, for example,
innovations in the use of plastics for automotive applications, will affect the level of secondary
aluminum demand. The following qualitatively describes a few recent innovations in the
secondary aluminum industry. A brief discussion of implications for the industry is provided.
A recent patented process developed by Alcan recycles dross without the need for a salt
flux, which is required in the conventional dross treatment process. This innovation may reduce
the cost of recycling dross, thereby increasing the supply of recycled dross-derived secondary
aluminum.
55
-------
Alcoa reports a new process for handling non-segregated scrap which results in a low-
lithium secondary ingot. This new process may broaden the market for secondary aluminum
ingot due to the resulting lithium reduction in the ingot.100
Additionally, a new aluminum can sheet plant has the ability to use 70 percent recycled
content - almost 20 percent higher than the industry average. This innovation may have a
substantial positive effect on the amount of secondary aluminum used in the can sheet production
sector of the industry.
A new technology called continuous casting is being used by Quanex and ACX
technologies. In traditional secondary operations, scrap is remelted and cast into ingot. The ingot
is then remelted and cast into some form of mill product, primarily sheet. This new technology
skips the middle step in which the aluminum is cast into ingot. The scrap can be melted and cast
directly into sheet saving a great deal in both labor and energy costs. This technology is well
positioned to displace much of the secondary recovery being performed by secondary producers
and within the operations of the integrated producers.101
2.6.4 NEW SOURCES-GROWTH AND REPLACEMENT
U.S. Companies Imco Recycling and Alchem Aluminum formed a joint venture to build a
$10 million aluminum recycling plant that will begin production by the fourth quarter of 1996.
This plant will have a rated capacity of 150 million Ibs/yr. but will be capable of expanding.102
Century Aluminum, that holds interests in Ravenswood Aluminum and the Mt. Holly
smelter announced plans to double its heat-treated plate capacity at the Ravenswood facility from
25 million Ibs/yr to 50 million Ibs/yr. This will be done by investing $28 million in new plate
equipment. Work on the plate expansion will be completed in phases with the first startup
beginning in the third quarter of 1998. The expansion is in response to rapid growth in the
commercial aircraft market, which Century estimates will grow by nearly 30 percent by the year
56
-------
2000.l03
2.6.5 PRICE PROJECTIONS
Trends in the price of secondary aluminum have historically tended to follow the
movements in the price of primary aluminum which is cyclical in nature. No specific price
projections were found for either secondary aluminum or scrap. The following forecasts were
found in the literature which pertain to overall aluminum and scrap prices.
Gerald A. Meyers, president and CEO of Century Aluminum forecasts some
improvement in the supply and demand fundamentals, resulting in aluminum prices averaging in
the low- to mid-70 cents/lb range in 1997. For 1998, further price increases are expected, but will
rely on many issues, including the European economy.104
According to Yale Brandt, vice chairman of Reynolds Metals Company, the amount of
scrap available will increase as the first- and second-generations of aluminum-intensive vehicles
enter the recycling stream. This will keep the price of scrap from rising steeply and pushing up
aluminum auto parts costs. In addition, prices will continue to depend on the overall total world
supply-demand equation for all aluminum.105
2.6.6 IMPORTS AND EXPORTS
According to Jorn DeLinde, director of ferroalloys at Resource Strategies, "Lack of
growth in export markets has not only dampened the U.S. recovery rate to a level far less robust
than those of the 1980's, but it has also negatively impacted U.S. metals sellers who are exporting
less and competing with more importers. The inability of US metal sellers to insulate themselves
from the rest of the world is apparent to anyone who has to compete with China's exports. While
many external conditions have undermined U.S. metals producers, some uncontrollable factors are
starting to benefit them. For example, the value of the U.S. dollar and the economic problems in
57
-------
Japan and Europe have made the U.S. into a preferred location for new business investment.
More businesses are recognizing that the economic ills of Japan and Europe are not just cyclical.
but, in fact, structural."106
58
-------
3.0 ECONOMIC METHODOLOGY
3.1 INTRODUCTION
The purpose of this chapter is to outline the economic methodology used in this analysis.
Baseline values used in the partial equilibrium analysis are presented, and the analytical methods
used to conduct the following analyses are described individually in this chapter:
• Partial equilibrium model used to compute post-control price, output, and trade
impacts;
• Economic surplus changes:
• Labor and trade impacts; and
• Financial impacts.
3.2 MARKET MODEL
The framework for the analysis of economic impacts on the secondary aluminum industry
is a partial-equilibrium model. A partial-equilibrium analysis is an analytical tool often used by
economists to analyze the single market model. This method assumes that some variables are
exogenously fixed at predetermined levels. The goal of the partial-equilibrium model is to specify
market supply and demand, estimate the post-control shift in market supply, estimate the change
in market equilibrium (price and quantity), and predict plant closures. This section presents the
framework of the partial equilibrium model, baseline equilibrium conditions, the calculation of
the supply curve shift, and the methodology used to calculate impacts on trade, facility closures.
and labor inputs. The baseline inputs for the secondary aluminum industry are also presented.
3.2.1 Partial-Equilibrium Analysis
A partial-equilibrium analysis was used to estimate the economic impacts of the chosen
regulatory option (the MACT floor) for the secondary aluminum industry. For modeling
59
-------
purposes, it was assumed that this industry is operating in a perfectly competitive market.
Perfectly competitive industries are characterized by the following conditions: the presence of
many sellers: production of a homogeneous product; a small market share owned by each firm in
the industry; freely available information regarding prices, technology, and profit opportunities;
freedom of entry and exit by firms in the industry; and competing sellers which are not considered
as a threat to market share by other firms in the industry.107 The implication of an assumption of
perfect competition to this analysis is that perfect competition constrains firms in the industry to
be price takers due to the absence of the market power necessary to affect market price. Firms
which operate in a perfectly competitive industry are also assumed to minimize costs. Although
the secondary aluminum industry does not strictly meet all of the criteria for perfect competition,
the market does have many buyers and sellers. The assumption of a competitive market is
reasonable for analytical purposes.
3.2.2 Market Demand and Supply
The baseline, or pre-control levels for the secondary aluminum market, is defined with a
domestic market demand equation, a domestic market supply equation, a foreign supply equation
(imports), and a foreign demand equation (exports). It is assumed that each of these markets will
clear, or achieve an equilibrium. The following equations identify the market demand, supply.
and equilibrium conditions for each affected industry:
Q°d = aPe
QDf = 6Pe
Qs" =
Q s/ = pPY
60
-------
Q = QD* + QD> =QS* + QSf
where:
= the quantity of the secondary aluminum demanded by domestic consumers
annually,
Of* ~ the quantity of the secondary aluminum demanded by foreign consumers
and produced by domestic producers annually (or exports),
0s'1 = the quantity of secondary aluminum produced by domestic supplier(s)
annually,
<2V = the quantity of secondary aluminum produced by foreign suppliers and sold
in the United States annually (or imports).
P = the price of secondary aluminum,
e = the price elasticity of demand for secondary aluminum, and
y = the price elasticity of supply for secondary aluminum.
The constants, a, 6, p. and p, are parameters estimated by the model, which are computed
such that the baseline equilibrium price is normalized to one. The market specification assumes
that domestic and foreign supply elasticities are the same, and that domestic and foreign demand
elasticities are identical. These assumptions are necessary, since data were not readily available to
estimate the price elasticity of supply for foreign suppliers and the price elasticity of demand for
foreign consumers.
3.2.3 Market Supply Shift
The domestic supply equation shown above may be solved for the price, P, of secondary
aluminum, respectively, to derive an inverse supply function that serves as the baseline supply
function for the industry. The inverse domestic supply equation for the industry is as follows:
j_
P = (QS"/3)Y
61
-------
A rational profit maximizing business firm will seek to increase the price of the product it
sells by an amount that recovers the capital and operation costs of the regulatory control
requirements over the useful life of the emission control equipment. This relationship is
identified in the following equation:
[(C • Q) - (V + D)] + D _
S
where:
C = the increase in the supply price,
Q = output,
V = a measure of annual operating and maintenance control costs.
D = annual depreciation (straight line depreciation is assumed),
S = a capital recovery factor, and
k = the investment cost of emission controls.
Thus, the model assumes that secondary aluminum facilities will seek to increase the
product supply price by an amount, C, that equates the investment costs in control equipment, k,
to the present value of the net revenue stream (revenues less expenditures) related to the
equipment. Solving the equation for the supply price increase, C, yields the following equation:
r kS - D V + D
C = +
Q Q
Estimates of the annual operation and maintenance control costs and of the investment
cost of emission controls, Fand k, respectively, were obtained from engineering studies
conducted by an engineering contractor for EPA and are based on 1994 price levels. Production
levels reflect values at the time the EPA industry survey was conducted. The variables for annual
62
-------
depreciation and the capital recovery factor, D and 5, respectively, are computed as follows:
S =
[(1
where:
r = the discount rate faced by producers, which is assumed to be 7 percent, and
T = the life of the emission control equipment, which is 20 years for most of the
proposed emission control equipment.
Emission control costs will increase the supply price for secondary aluminum by an
amount equivalent to the per unit cost of the annual recovery of investment costs plus the annual
operating costs of emission control equipment, or C, (/ denotes the number of affected facilities in
the industry). The baseline product cost curve for secondary aluminum is unknown because
production costs for the individual facilities are unknown. Therefore, an assumption is made that
the affected facilities in each industry with the highest per unit control costs are marginal in the
post-control market. In other words, those firms with the highest per unit control costs also have
the highest per-unit pre-control production costs. This is a worst-case scenario model assumption
for price and quantity impacts that may not be the case in reality. The assumption, however.
results in the upper bound of possible price and quantity impacts occurring as the result of
regulation. Based upon this assumption, the post-control supply function can be expressed as
follows:
where:
63
-------
C (C, q,) = a function that shifts the supply function to reflect the incurrence of
control costs,
C, = the vertical shift that occurs in the supply curve for the ith facility to
reflect the increased cost of production in the post-control market,
and
q, = the quantity produced by the zth facility producing secondary
aluminum.
This shift in the supply curve is shown graphically in Figure 3-1.
3.2.4 Impact of the Supply Shift on Market Price and Quantity
The impact of the control standards on market equilibrium price and output is derived by solving
for the post-control market equilibrium and comparing the new equilibrium price and quantity to
the baseline equilibrium conditions. Since post-control domestic supply is assumed to be
segmented, or a step function, a special algorithm was developed to solve for the post control
market equilibrium. The algorithm first searches for the segment in the post-control supply
function at which equilibrium occurs, and then solves for the post-control market price that clears
the market.
Since the market-clearing price occurs where the sum of domestic demand and foreign
demand of domestic production equals post-control domestic supply plus foreign supply, the
64
-------
FIGURE 3-1. ILLUSTRATION OF POST-NESHAP MODEL
Price
S,
Annualized
Per tMiT
Compliance Cost
Quantity
-------
algorithm simultaneously solves for the following post-control variables:
• Equilibrium market price;
• Equilibrium market quantity;
• Change in the value of domestic production or revenues to producers;
• Quantity supplied by domestic producers;
• Quantity supplied by foreign producers (imports);
• Quantity demanded (domestic production) by foreign consumers (exports); and
• Quantity demanded by domestic consumers.
The changes in these equilibrium variables are estimated by comparing baseline equilibrium
values to post-control equilibrium values.
3.2.5 Trade Impacts
Trade impacts are reported as the change in the volume of exports, imports, and net
exports (exports minus imports). The price elasticity of demand for each of the products has been
assumed to be identical for foreign and domestic consumers, and the price elasticity of supply is
presumed the same for foreign and domestic producers. As the volume of imports rises and the
volume of exports falls, the volume of net exports will decline.
The following algorithms are used to compute the trade impacts of the proposed regulatory
alternative:
' =
= Q' - QQf
= .QD'
66
-------
where:
A£>V = the change in the volume of imports,
A£XV = the change in the volume of exports, and
*NX = the volume change in net exports.
The subscripts 1 and 0 refer to the post- and pre-control equilibrium values, respectively, and all
other variables have been previously identified.
3.2.6 Plant Closures
It is assumed that a facility will close if its post-control supply price exceeds the post-
control market equilibrium price. Closures in this analysis relate to facilities. In the event the
firm owns multiple facilities or has diversified interests, the firm itself may not shut down.
however, an individual facility owned by the firm may close.
3.2.7 Changes in Economic Welfare
Regulator}' control requirements will result in changes in the market equilibrium price and
quantity of secondary aluminum produced and sold. These changes in the market equilibrium
price and quantity will affect the welfare of consumers of products manufactured by secondary
aluminum producers, producers of secondary aluminum and secondary aluminum products, and
society as a whole. The methods used to measure these changes in welfare are described below.
3.2.7.1 Changes in Consumer Surplus
Consumers will bear a loss in consumer surplus, or a dead-weight loss, associated with the
reduction in the amount of secondary aluminum sold due to higher prices charged for this product.
67
-------
This loss in consumer surplus represents the amount consumers would have been willing to pay
over the pre-control price for production eliminated. Additionally, consumers will have to pay a
higher price for post-control output. This consumer surplus change for domestic consumers is
defined as the AGSd, as follows:
°° 1 DO
Drf/_.\ e _i/^v Orf . r> /•*> d O f~\ a
(Qu'/a)e dQu" + P,Q, d- P0Q
00
The change in consumer surplus is an estimate of the losses of surplus incurred by domestic
consumers only. Although both domestic and foreign consumers may suffer a loss in surplus as a
result of emission controls, this study focuses on the change in domestic consumer surplus only.
The variable. ACS,,. represents the change in domestic consumer surplus that results from the
change in market equilibrium price and quantity occurring after the incurrence of regulatory
control costs. Figure 3-2 depicts the loss in consumer surplus.
3.2.7.2 Change in Producer Surplus
The change in producer surplus is composed of two elements. The first element relates to
output eliminated as the result of emission controls. The second element is associated with the
change in price and cost of production for the new market equilibrium quantity. The total change
in producer surplus is the sum of these two elements. Figure 3-2 depicts graphically the change in
producer surplus resulting from this regulation.
The lower output levels resulting from emission control costs cause producers to suffer a
welfare loss in producer surplus. Affected facilities which continue producing after the incurrence
of control costs realize a welfare gain on each unit of production produced attributable to the
incremental increase in the market price. Producers will also experience a decrease in welfare per
unit of production relating to the increased capital costs and operating cost of emission controls.
68
-------
The total change in producer surplus is specified by the following equation:
APS = [P, Q, * - P0Q0 - - f (Q/(3)v dQ - £ C,. g,.]
s M
Q,8'
Since domestic surplus changes are the object of interest, the welfare gain experienced by
foreign producers due to higher prices is not considered. This procedure treats higher prices paid
for imports as a dead-weight loss in consumer surplus. Higher prices paid to foreign producers
represent simply a transfer of surplus from the United States to other countries from a world
economy perspective, but a welfare loss from the perspective of the domestic economy.
3.2.7.4 Total Economic Costs
The total economic costs of the regulations are the sum of the changes in consumer surplus
and producer surplus. This relationship is defined in the following equation:
EC = ACS,, + APS
where:
EC = the economic cost of the controls.
All other variables have been previously defined.
3.2.8 Labor Input Impact
The estimate of the labor market impact associated with the standard is based on the
baseline input-output ratios and the estimated changes in domestic production. The labor market
impacts are measured as the number of jobs lost due to domestic output reductions. The
69
-------
FIGURE 3-2. PRODUCER AND CONSUMER SURPLUS CHANGES
$/Q
Q2 Q, Q/t
(a) Change In Consumer Surplus with Regulation
$/Q
Q2 Q1 Q/t
(b) Change in Producer Surplus with Regulation
$/Q
(c) Net Change in Economic Welfare with Regulation
-------
estimated number of job losses are a function of the change in level of production that is
anticipated to occur as a result of the emission controls. Employment information was provided
in the EPA information collection request for this regulation. Based on data in the information
collection request, secondary aluminum facilities have, on average, approximately 275 employees
per facility. Extrapolating this estimate to the industry, results in a total employment estimate of
23,650 employees. The change in employment level is estimated as follows:
AL = [L0 * %AQ Sd]
where:
A/, = the change in the employment level expressed in terms of number of
workers,
L,, = the total number of secondary aluminum employees,
0s'1 = Domestic supply.
3.2 BASELINE INPUTS
The partial equilibrium model used in this analysis requires, as data inputs, baseline values
for variables and parameters that have been previously described to characterize the secondary
aluminum industry. These data inputs include the number of domestic facilities currently in
operation, the annual capacity per facility, and the relevant control costs per facility. Table 3-1
lists the variable and parameter inputs to the model for the secondary aluminum industry.
In Table 3-1 the baseline parameters and variables used to characterize baseline market
conditions are listed. Baseline market price data are based upon information published by the
American Metals Market.108 Production data are based upon model plant production and the
number of model plants assumed for this industry. 109 Imports and exports of secondary aluminum
are estimated using scrap metal import and export ratios obtained from the Aluminum Association
71
-------
and applied to the secondary aluminum domestic production.110 The prices are staled in cents per
pound and industry output is stated in millions of tons produced annually. The price elasticities of
supply and demand were estimated econometrically and are discussed in Section 3.3 Industry
Supply and Demand Elasticities.
A discount rate of 7 percent is assumed for the analysis as a surrogate for the actual rates
in the economy. The 7 percent social discount rate is consistent with the most current United
States Office of Management and Budget (OMB) guidance.1" The equipment life of 20 years was
obtained from the engineering study of emission control costs conducted by an engineering
contractor for EPA. This equipment life is applicable for most of the pollution control equipment
considered in the analysis. The number of employees employed in the secondary aluminum
industry was calculated from data obtained from the Secondary Aluminum information collection
request."2
3.3 INDUSTRY SUPPLY AND DEMAND ELASTICITIES
3.3.1 Introduction
Demand and supply elasticities are crucial components of the partial equilibrium model
used to quantify the economic impact of regulatory control cost measures on the affected
secondary aluminum industry. The price elasticity of demand was unavailable in the literature for
secondary aluminum, specifically. However, an estimate of the price elasticity of demand for the
overall aluminum metal (primary and secondary) was available. Since the only published
elasticity estimate relates to demand for the overall metal and was published in 1979,
econometric estimates of the price elasticity of demand and the price elasticity of supply are
72
-------
TABLE 3-1. BASELINE INPUTS FOR THE SECONDARY ALUMINUM INDUSTRY
Variable/Parameter
Price (P0)'
Domestic Output, (Q0Sd)2
Imports, (Q0sf)3
Exports, (Q0Df)3
Demand Elasticity (e)
Supply Elasticity (y)
Equipment life (T)
Labor (Number of workers)4
Amount
74.71
13.016
1.526
1.23
-0.34
2.33
20
23,650
NOTES: ' Cents per pound (1994$).
; Millions of tons per year.
' The import ratio and export ratio are based upon scrap metal import and export ratios for 1994 applied to secondary aluminum
domestic production.
4 Number of workers estimated for the secondary aluminum industry' are based on the EPA Information Collection Request.
73
-------
estimated for this analysis. The following sections present the analytical approach, and the data
employed to estimate the price elasticities of demand and supply used in the partial equilibrium
analysis. The techniques utilized to estimate the price elasticity of demand and supply are
consistent with economic theory and, at the same time, utilize the data available.
3.3.2 Price Elasticity of Demand
The price elasticity of demand, or own-price elasticity of demand, is a measure of the
sensitivity of buyers of a product to a change in price of the product. The price elasticity of
demand represents the percentage change in the quantity demanded resulting from each 1 percent
change in the price of the product.
The only published estimates of the elasticity of demand for aluminum are for the overall
metal industry. No distinctions were found in the literature between the primary and secondary
aluminum price elasticity of demand estimates, nor between the elasticities of the various end-use
sectors for aluminum. Given the many substitutes for aluminum, an important concept to
consider is the cross-price elasticity of demand for aluminum with its substitute materials. One
estimate located of the cross-price elasticity of demand for aluminum with steel was quite high--
2.07.1I3 The existence of substitutes for aluminum such as steel are a relevant consideration for
secondary market impacts in the economic impact analysis.
Because of the likelihood that the substitution of aluminum with one of its alternatives
(e.g., steel) would require a process change, it is likely that the response of demand to price
changes in the overall aluminum industry would be relatively minor in the short-term; long-run
elasticities however would be higher. The results of a study published in 1979 confirms this
supposition. This study found that the aluminum price elasticities of demand were -0.13 in the
short-run, and -0.80 in the long-run.114 The market impacts associated with the published long run
estimate of the price elasticity of demand of-0.80 are addressed in a sensitivity analysis in
Appendix A.
74
-------
3.3.2.1 Approach
Secondary aluminum is used as intermediate products to produce final goods. The
demand for secondary aluminum is therefore derived from the demand for these final products.
Information concerning the end uses of aluminum is provided in Chapter 2. According to the
information contained in Chapter 2, secondary aluminum is used primarily as an input into
aluminum cans which are used for canned beverages such as beer and soft drinks.
The assumption was made that firms using secondary aluminum as input into their
productive processes seek to maximize profits. The profit function for these firms may be written
as follows:
Max n = PFP * f(Q, /) - (P * Q) - (POI - 1}
Q, /
where:
7t = profit,
PFP = the price of the final product or end-use product,
XQ. I) - the production function of the firm producing the final product,
P = the price of the secondary aluminum,
Q = the quantity input use of secondary aluminum,
POI = a vector of prices of other inputs used to produce the final product,
and
I = a vector of other inputs used to produce the final product.
All other variables have been previously defined.
The solution to the profit function maximization results in a system of derived demand
equations for secondary aluminum. The derived demand equations are of the following form:
75
-------
Q = 9(P, PFP' po/) ^
A multiplicative functional form of the derived demand equation is assumed because of the useful
properties associated with this functional form. The functional form of the derived demand
function is expressed in the following formula:
Q = AP*P*FPP
where:
P = the price elasticity of demand for the secondary aluminum, and
PFP = the final product price elasticity with respect to the use of the secondary
aluminum.
All other variables have been previously defined. P, PFP, and A are parameters to be estimated by
the model. P represents the own-price elasticity of demand. The price of other inputs
(represented by POI) has been omitted from the estimated model, because data relevant to these
inputs were unavailable. The implication of this omission is that the use of secondary aluminum
production is fixed by technology.
The market price and quantity sold of secondary aluminum are simultaneously determined
by the demand and supply equations. For this reason, it is advantageous to apply a systems
estimator to obtain unbiased and consistent estimates of the coefficients for the demand
equations.115 Two-stage least squares (2SLS) is the estimation procedure used in this analysis to
estimate the demand equations for secondary aluminum. Two-stage least squares uses the
information available from the specification of an equation system to obtain a unique estimate for
each structural parameter. The predetermined, or exogenous, variables in the demand and supply
equations are used as instruments. The supply-side variables used to estimate the demand
functions include: the real capital stock variable for SIC code 3341, a technology time trend (t),
real GDP, and the weighted-average price index for the cost of labor and materials for SIC code
3341(PLM).
76
-------
3.3.2.2 Data
Data relevant to the econometric modeling of the price elasticity of demand for secondary
aluminum including the variable symbol, units of measure, and variable descriptions are listed in
Table 3-3. Consistent time series data for the period 1975 through 1993 were obtained. Price
data for No 380 Alloy and secondary aluminum production data were obtained from the American
Metals Market116 and the Aluminum Association117, respectively. The Annual Survey of
Manufactures publishes price indices for the final products of interest (soft drinks, beer, and metal
cans and containers). Data relative to the supply-side variables were also obtained from the
Annual Survey of Manufactures.118 Finally, real GDP and the GDP deflator were obtained from
the Bureau of Economic Analysis.119
3.3.2.3 Statistical Results
SAS Release 6.12 for Windows was used to estimate econometric estimates of the price
elasticity of demand. Two-stage least square econometric models were estimated for the
secondary aluminum industry using the price of soft drinks, beer, and metal cans and containers
as the end-use products, respectively, and the methods and models discussed. The models using
the price of soft drinks and beer, respectively, as the final products were not successful. The
coefficient for the own-price variable in each model was not statistically different from zero. In
contrast, the model results for the demand model using the price of metal cans and containers is
reported in Table 3-4. Standard errors are shown in parenthesis. Each of the coefficients reported
has the anticipated sign and is statistically significant.
The price elasticity of demand estimate for secondary aluminum reflects that the demand
for secondary aluminum is inelastic. Regulatory control costs are more likely to be paid by
consumers of products with inelastic demand when compared to products with elastic demand,
77
-------
Variable
TABLE 3-2. DATA INPUTS FOR THE ESTIMATION OF DEMAND
- EQUATIONS FOR THE SECONDARY ALUMINUM INDUSTRY
Unit of Measure Description
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
Time Trend - 1
Price (Alloy No. 380)- P'
Sales Volume of Secondary Aluminum
-Q2
Price Final Goods: - PFP
a. Bottled and Canned Soft Drinks and
Carbonated Waters3
b. Malt Beverages3
c. Metal Cans and Containers3
Cost of Material Inputs3
Price index for Material Inputs3
Production Worker Wages3
Production Worker Hours3
Real Capital Stock3
Implicit Price Deflator4
Real Gross Domestic Product4
.
price per pound
thousands of metric
tons
index
index
index
index
index
millions of $1987
thousands of labor
man hours
millions of $1987
index
millions of $1987
-
Annual Average Price
Quantity sold of
secondary aluminum
SIC code 2086
SIC code 2082
SIC code 341 1
SIC code 3341
SIC code 3341
SIC code 3341
SIC code 3341
SIC code 3341
NOTES: 1. American Metals Market.
2. Aluminum Association.
3. Annual Survey of Manufactures.
4. United States Bureau of Economic Analysis.
TABLE 3-3. DERIVED DEMAND COEFFICIENTS
Secondary Aluminum
Own Price P'
Secondary Aluminum
End-Use pFP'
Metal Cans and Containers
Model F Value
Probability > F
-0.34
(0.185)
0.69
(0.207)
19.548
0.0001
Adjusted R-Sauare
0.6733
NOTES: Standard errors are shown in parenthesis.
78
-------
all other things held constant. Price increases for products with inelastic price elasticity of
demand lead to revenue increases for producers of the product. Thus, one can predict that price
increases resulting from implementation of regulatory control costs will lead to an increase in
revenues for the secondary aluminum industry.
A degree of uncertainty is associated with this method of demand estimation. The
estimation is not robust since the model results vary depending upon the instruments used in the
estimation process. For this reason, a sensitivity analysis of the price elasticity of demand
estimates is presented using a range of elasticities. A lower estimate that differs by minus one
standard deviation from that utilized in the analysis of-0.16. The estimate reported in the
literature for the overall aluminum of-0.80 is used for the upper bound. The results of the
sensitivity analysis are reported in Appendix A.
3.3.3 Price Elasticity of Supply
The price elasticity of supply, or own-price elasticity of supply, is a measure of the
responsiveness of producers to changes in the price of a product. The price elasticity of supply
indicates the percentage change in the quantity supplied of a product resulting from each 1
percent change in the price of the product.
3.3.3.1 Model Approach
Published sources of the price elasticity of supply using current data were not readily
available. For this reason, an econometric analysis of the price elasticity of supply for the
secondary aluminum was conducted. The approach used to estimate the price elasticity of supply
makes use of the production function. The theoretical methodology of deriving a supply
elasticity from an estimated production function will be briefly discussed with the industry
production function defined as follows:
79
-------
Q s = f (L,K,M,t)
where:
QF = output or production
L = the labor input, or number of labor hours,
K = real capital stock,
M = the material inputs, and
/ = a time variable to reflect technology changes.
In a competitive market, market forces constrain firms to produce at the cost minimizing
output level. Cost minimization allows for the duality mapping of a firm's technology
(summarized by the firm's production function) to the firm's economic behavior (summarized by
the firm's cost function). The total cost function for a secondary aluminum facility is as follows:
TC = h(C,K,t,Qs)
where:
TC = the total cost of production, and
C = the cost of production (including cost of materials and labor).
All other variables have been previously defined.
This methodology assumes that capital stock is fixed, or a sunk cost of production. The
assumption of a fixed capital stock may be viewed as a short-run modeling assumption. This
assumption is consistent with the objective of modeling the adjustment of supply to price
changes after implementation of controls. Firms will make economic decisions that consider
those costs of production that are discretionary or avoidable. These avoidable costs include
production costs, such as labor and materials, and emission control costs. In contrast, costs
associated with existing capital are not avoidable or discretionary. Differentiating the total cost
function with respect to Q5 derives the following marginal cost function:
80
-------
MC = h'(C,K,t,Qs)
where MC is the marginal cost of production and all other variables have been previously
defined.
Profit maximizing competitive firms will choose to produce the quantity of output that
equates market price, P, to the marginal cost of production. Setting the price equal to the
preceding marginal cost function and solving for Qs yields the following implied supply
function:
Qs = (P,PL,PM,K,t)
where:
P = the price of secondary aluminum.
P, = the price of labor, and
PA/ = the price of materials input.
All other variables have been previously defined.
An explicit functional form of the production function may be assumed to facilitate
estimation of the model. For this analysis, the Cobb-Douglas, or multiplicative form, of the
production function is postulated. The Cobb-Douglas production function has the convenient
property of yielding constant elasticity measures. The functional form of the production function
becomes:
Q, = A K?K tA L?L M?M
where:
Q, = output or production in year t,
K, = the real capital stock in year t,
L, = the quantity of labor hours used in year t,
81
-------
M, = the material inputs in year t, and
A, aK, a/, av/, A = parameters to be estimated by the model.
This equation can be written in linear form by taking the natural logarithms of both sides
of the equation. Linear regression techniques may then be applied. Using the approach
described, the implied supply function may be derived as:
In = p0 + Y In P + P2 In K +(33 In PL + 34 In PM + (35 In t
where:
PI = the factor price of the labor input,
PU = the factor price of the material input, and
K - fixed real capital.
The P, and ^coefficients are functions of the a,, the coefficients of the production function. The
supply elasticity, y. is equal to the following:
Y =
It is necessary to place some restrictions on the estimated coefficients of the production
function in order to have well-defined supply function coefficients. The sum of the coefficients
for labor and materials should be less than one. Coefficient values for a, and aM that equal to
one result in a price elasticity of supply that is undefined, and values greater than one result in
negative supply elasticity measures. For these reasons, the production function is estimated with
the restriction that the sum of the coefficients for the inputs equal one. This is analogous to
assuming that the secondary aluminum industry exhibits constant returns to scale, or is a long-
run constant cost industry. This assumption seems reasonable on an a priori basis and is not
inconsistent with the data.
82
-------
3.3.3.2 Estimated Model
The estimated model reflects the production function for secondary aluminum, using
annual time series data for the years from 1958 through 1994. The following model was
estimated econometrically:
In Qf = In A + aK In K + A In t + aL In L + aM In M
where each of the variables and coefficients have been previously defined.
3.3.3.3 Data
The data used to estimate the model are enumerated in Table 3-5. This table contains a
list of the variables included in the model, the units of measure, and a brief description of the
data. The data for the price elasticity of supply estimation model includes: the value of domestic
shipments in millions of dollars; the price index for value of domestic shipments (the value of
domestic shipments deflated by the price index represent the quantity variable, O, or the
dependent variable in the analysis); a technology time variable, /; real net capital stock, K, in
millions of dollars; the number of production labor man-hours, L,; the material inputs in millions
of dollars, M,; and the price index for value of materials. Data to estimate the production
function for the secondary aluminum industry exclusively were largely unavailable; therefore,
data for SIC code 3341 (Secondary Smelting and Refining of Nonferrous Metals) is utilized for
each of the variables previously enumerated with the exception of the time variable.
The capital stock variable was the most difficult variable to quantify for use in the
econometric model. Ideally, this variable should represent the economic value of the capital
stock actually used by each facility to produce secondary aluminum for each year of the study.
The most reasonable data for this variable would be the number of machine hours actually used
83
-------
TABLE 3-4. DATA INPUTS FOR THE ESTIMATION OF THE PRODUCTION FUNCTION
FOR THE SECONDARY ALUMINUM INDUSTRY
Variable
Unit of Measure
Description
fir
Millions of dollars
M,
Years
Millions of 1987
dollars
Thousands of labor man hours
Millions of dollars
The value of shipments for SIC code
3341 deflated by the price index for
value of shipments1
technology time trend
Real capital stock for SIC code 3341'
Production worker hours
for SIC code 3341'
Dollar value of material input for SIC
code 3341 deflated to real values using
the materials price index1
NOTES: 'Annual Survey of Manufactures.
to produce secondary aluminum each year. These data are unavailable. In lieu of machine hours
data, the dollar value of net capital stock in constant 1987 prices, or real net capital stock, is used
as a proxy for this variable. However, these data are imperfect in two ways. First, the data
represent accounting valuations of capital stock rather than economic valuations. This aberration
is not easily remedied, but is generally considered unavoidable in most studies of this kind. The
second flaw involves capital investment that is used in production of secondary nonferrous
metals other than secondary aluminum. Although aluminum represents the most significant
metal accounted for in SIC 3341, it is not possible to segregate data for aluminum versus other
metals. Thus the assumption is made that the production processes of secondary nonferrous
metals are highly correlated with the productive process of secondary aluminum. This
84
-------
assumption represents a limitation of the analysis.
3.3.3.5 Statistical Results
SAS Release 6.12 for Windows was used to estimate econometric estimates of the price
elasticity of supply for the secondary aluminum industry. A restricted least squares estimator
was used to estimate the coefficients of the production function model. A log-linear
specification was estimated with the sum of the a, restricted to unity. This procedure is
consistent with the assumption of constant returns to scale. The model was further adjusted to
correct for first-order serial correlation using the Yule-Walker estimation procedure. The results
of the estimated model are presented in Table 3-6. All of the coefficients have the expected sign.
but only the capital stock and materials coefficients are significantly different from zero with a
high degree of confidence.
TABLE 3-5. ESTIMATED SUPPLY MODEL COEFFICIENTS FOR THE SECONDARY
ALUMINUM INDUSTRY
Variable Estimated Coefficients'
t time -0.04559
(0.0301)
K, Capital stock 0.30073
(.0.0707)
L, Labor -0.00677
(0.0155)
M, Materials 0.70604
(0.0696)
NOTES: 'Standard errors are shown in parenthesis.
The coefficients for real capital and materials have the anticipated signs and are
significant at a high level of confidence. The labor coefficient does not have the anticipated sign
85
-------
and does not test significantly different from zero. Using the estimated coefficients in Table 3-6
and the formula for supply elasticity shown under Section 3.3.3.1, Model Approach, the price
elasticity of supply for secondary aluminum is derived to be 2.33. The calculation of statistical
significance for this elasticity measure is not a straightforward calculation since the estimated
function in non-linear. No attempt has been made to assess the statistical significance of the
estimated elasticity. The corrections for serial correlation and the restricted model results yield
inaccurate standard measures of goodness of fit (R2). However, the model that is unrestricted
and unadjusted for serial correlation has an R2 of 0.79.
3.3.3.6 Limitations of the Supply Elasticity Estimates
The estimated price elasticity of supply for the secondary aluminum industry reflects that
the industry in the United States will increase production of secondary aluminum products by
2.33 percent for every 1.0 percent increase in the price of these products. The preceding
methodology does not directly estimate the supply elasticities for secondary aluminum due to a
lack of necessary data. The assumption implicit in the use of this price elasticity of supply
estimate is that the supply elasticity of secondary aluminum will not differ significantly from the
price elasticity of supply for all products classified under SIC code 3341.
The uncertainty of the supply estimate is acknowledged. The results of a sensitivity
analysis of the price elasticity supply is included in Appendix A for a high and low estimate of
the price elasticity of supply of 1.33 and 3.33, respectively.
3.4 FINANCIAL IMPACT ANALYSIS
It is necessary to estimate the impact of the emission controls on the affected firms'
financial performance after investment in emission control equipment. The financial impact
analysis was conducted on a model plant basis due to lack of sufficient firm-specific financial
data. However, model plant data evaluated at the firm level have been analyzed to estimate the
86
-------
financial impact of the regulation to firms within the industry.
The measure of the financial impact of the regulation used in this analysis is the cost-to-
sales ratio. Cost-to-sales ratio refers to the change in annualized control costs divided by the
revenues of a particular good or goods being produced in the process for which additional
pollution control is required. It can be estimated for either individual firms or as an average for
some set of firms such as affected small firms. While it has different significance for different
market situations, it is a good rough gauge of potential impact. If costs for the individual (or
group) of firms are completely passed on to the purchasers of the good(s) being produced, it is an
estimate of the price change (in percentage form after multiplying the ratio by 100). If costs are
completely absorbed by the producer, it is an estimate of changes in pretax profits (in percentage
form after multiplying the ratio by 100). The distribution of costs-to-sales ratios across the
whole market, the competitiveness of the market, and profit-to-sales ratios are among the
obvious factors that may influence the significance of any particular cost-to-sales ratio for an
individual facility.
Due to the number of facilities and variety of processes used in this industry, model
plants were developed to categorize facilities based on possible combinations of processes that
are performed. Eight model plants were created and annual compliance costs were calculated for
each one. The individual facilities were then assigned to the model plant that most closely fit
their process structure, and the annual compliance cost for that model plant was used in
calculating the company's cost/sales ratio. Model plant revenues were estimated based upon the
estimated model plant annual production and the average 1994 price of secondary aluminum
alloy A-380120.121
87
-------
4.0 CONTROL COSTS, ENVIRONMENTAL IMPACTS, COST
EFFECTIVENESS, AND ESTIMATES OF ECONOMIC COSTS
4.1 INTRODUCTION
Inputs to the model outlined in the previous chapter include market data summarized in
Chapter 2.0 and control cost estimates provided by the EPA. This chapter summarizes the cost
inputs used in this EIA that were provided on a model plant basis for the affected facilities.
Environmental impacts associated with the proposed regulatory alternative are also presented in
this chapter. These impacts reflect estimates of emission reductions anticipated to result for this
regulation. Finally, estimates of the cost-effectiveness and economic costs of the proposed
regulation are shown.
4.2 CONTROL COST ESTIMATES
Control cost estimates and emission reductions were provided by EPA's engineering
contractor on a model plant basis for each of the affected facilities. The model plants are made
up of a series of model processes each of which is designed to represent processes found in the
industry as determined from an industry survey. For each model process at each model plant, the
baseline level of control is defined as uncontrolled, partially controlled, or fully controlled
relative the Maximum Achievable Control Technology (MACT) floor level of emission control.
In order to comply with a MACT standard, each model plant is assumed to need emission control
equipment equal to MACT floor technology. Depending on the baseline level of control, plants
may need to install an entire emission control system or may only need to update an existing
system (e.g., add a lime injection system to an existing baghouse). EPA estimates that some
furnaces can meet the proposed emission limits by using pollution prevention/work practices
rather than installing emission controls. For some of these furnaces, a monitoring option would
consist of a program for inspecting and testing the level of contaminants in the scrap. The cost
impacts include the estimated cost of such a scrap inspection and sampling procedure for these
furnaces.
-------
The emission processes for which costs were estimated include: crushing and shredding;
drying; delaquering; melting; holding; sweating; and dross cooling. The costs were estimated for
existing emission sources and potential new sources. New source costs represent the control of
new process units and equipment built (or reconstructed or replaced ) in the first 5 years after
promulgation. For the secondary aluminum industry, it was assumed that new facilities planned
and built in the near future will primary be built as replacement facilities for currently existing
sources. Thus on a net basis, the cost estimates assume no growth in industry production in the
near future beyond the current level of production. This assumption seems reasonable
considering the current economic environment worldwide.
Table 4-1 presents the model plant and national annualized cost estimates for the
secondary aluminum NESHAP.122 Emission control costs are the annualized capital and annual
operating and maintenance costs of controls based on the assumption that all affected facilities
install controls. Capital costs for this regulation are estimated to be approximately $147.9
million. Since capital costs relate to emission control equipment that will be utilized over a
period of years, this cost is annualized or apportioned to each year of the anticipated equipment
life. The annual capital costs include annual depreciation of equipment plus the cost of capital
associated with financing the capital equipment over its useful life. A seven percent discount
rate or cost of capital is assumed for this regulation. The total national annualized costs for
implementing the regulation are expected to be approximately $73.2 million including the costs
of monitoring, recordkeeping and reporting. All costs are shown in 1994 dollars.
4.3 ENVIRONMENTAL IMPACTS AND COST EFFECTIVENESS
Baseline emissions for each model plant are estimated on the basis of measured emissions
from existing emission controls. Emissions to the atmosphere after implementation of the
MACT standard are assumed to be equal to the proposed emission limits for each regulated
source within the industry. Emission reductions at each model plant are estimated as the
difference between emissions with existing controls in place and emissions to the atmosphere
89
-------
after adding the required emission controls. Where test data are incomplete, emission reductions
are based on assumed control efficiencies or projections of measured efficiency at plants with
similar processes.
TABLE 4-1. CONTROL COST ESTIMATES FOR MODEL PLANTS AND
NATIONWIDE COST ESTIMATES FOR THE SECONDARY ALUMINUM NESHAP
(thousands of 1994 dollars)
Model
Plants
1
2
3
4
5
6
7
8
Burden Costs
Nationwide
Costs
Capital
Model Plant
Costs
$1,390
1,660
1,833
2,944
2,159
3,731
198
0
Annualized
Model
Plant Cost
$541
574
702
1,203
1,400
2,142
134
0
Number
of
Plants
31
10
7
9
10
7
6
6
Total
Capital
Costs
$43,094
16,603
12,832
26,492
21,587
26,119
1,188
0
$147,915
Total
Annualized
Costs
$16,770
5,740
4.910
10.829
14.001
14.992
807
0
5,142
$73,191
The estimated emission reductions are based on the assumption that plants reduce
emissions to a level equal to the proposed emission limit. Table 4-2 contains a summary of
emission reductions for individual model plants and for the nation.123 Emission reductions are
shown in total and for hazardous air pollutants (HAPs). The estimated total HAP emission
reductions are 12,457.5 tons per year, and total emission reductions of approximately 18,321.9
90
-------
tons per year are anticipated. Cost effectiveness of this regulation is calculated by dividing the
annualized costs of the regulation by the estimated emission reductions. The cost effectiveness
of this regulation is estimated to be $3,714 per ton of total emissions reduced and $5,463 per ton
TABLE 4-2. ENVIRONMENTAL IMPACTS AND COST EFFECTIVENESS
Model
Plants
1
2
3
4
5
6
7
8
Total
Nationwide
Emission Reductions
by Plant (tons/yr)
Total
114.5
214.2
210.7
538.7
467.8
226.0
7.43
0
HAP
54.6
109.2
143.1
406.9
349.7
216.1
0
0
Emission Reductions
Nationwide (tons/yr)
Total
3.551.0
2,142.4
1,475.2
4,848.4
4,678.2
1,582.0
44.6
0
18,321.9
HAP
1,692.3
1,091.8
1,001.8
3,662.1
3,497.1
1,512.5
0
0
12,457.5
Cost Effectiveness
(1994$/ton)
All
Pollutants
$4,723
$2,679
$3,329
$2,233
$2,993
$9,476
$18,108
N/A
$3,714
HAP
$9,910
$5,257
$4,902
$2,957
$4,004
$9,912
N/A
N/A
$5,463
91
-------
of HAP reduction. Table 4-3 presents the total baseline emissions, controlled emissions, and
emission reductions by emission type.
TABLE 4-3. BASELINE EMISSIONS AND EMISSION REDUCTIONS
Emissions
Particulates
(tons/yr)
Hydrogen
Chloride (tons/yr)
Chlorine (tons/yr)
Total
Hydrocarbons
(tons/yr)
Dioxins/Furans
(lb/yr)
HAP Metals
(ton/yr)
Total (tons/yr)
Total HAPs
(tons/yr)
Baseline
Emissions
9,378
16,902
1,098
4,169
1.79
64.4
31,548
18,065
Controlled
Emissions
3,514
4,445
.24
24.1
13,226
5,569
Emission
Reductions
5,864
12,457
1.55
40.3
18,332
12,457
Percent
Reductions
62.5
73.7
86.6
62.5
58.1
69.2
4.4 ESTIMATE OF ECONOMIC COSTS
Air quality regulations affect society's economic well-being by causing a reallocation of
productive resources within the economy. Resources are allocated away from the production of
goods and services (secondary aluminum) to the production of cleaner air. Estimates of the
economic costs of .cleaner air require an assessment of costs to be incurred by society as a result
92
-------
of emission control measures. By definition, the economic costs of pollution control are the
opportunity costs incurred by society for productive resources reallocated in the economy to
pollution abatement. The economic costs of the regulation can be measured as the value that
society places on goods and services not produced as a result of resources being diverted to the
production of improved air quality. The conceptually correct valuation of these costs requires the
identification of society's willingness to be compensated for the foregone consumption
opportunities resulting from the regulation. In contrast to the economic cost of regulation,
emission control costs consider only the direct cost of emission controls to the industry affected
by the regulation. Economic costs are a more accurate measure of the costs of the regulation to
society than an engineering estimate of compliance costs. However, compliance cost estimates
provide an essential element in the economic analysis.
Economic costs are incurred by consumers, producers, and society at large as a result of
pollution control regulations. These costs are measured as consumer surplus and producer
surplus. Consumer surplus is a measure of well-being or the welfare of consumers of a good and
is defined as the difference between the total benefits of consuming a good and the market price
paid for the good. Pollution control measures will result in a loss in consumer surplus due to
higher prices paid for secondary aluminum and to the deadweight loss in surplus caused by
reduced output of secondary aluminum in the post-control market.
Producer surplus is a measure of producers welfare that reflects the difference between
the market price charged for a product and the marginal cost of production. Pollution controls
will result in a change in producer surplus that consists of three components. These changes
include surplus gains relating to increased revenues experienced by firms in the secondary
aluminum industry due to higher post-control prices, surplus losses associated with increased
costs of production for annualized emission control costs, and surplus losses due to reductions in
post-control output. The net change in producer surplus is the sum of these surplus gains and
losses.
The sum of the change in consumer surplus and producer surplus constitutes the
93
-------
economic costs of the regulation to society. Table 4-4 summarizes the economic costs associated
with the regulatory alternative. The economics costs of the secondary aluminum industry
NESHAP is $76.4 million (1994$).
TABLE 4-4. ANNUAL ECONOMIC COST ESTIMATES
FOR THE SECONDARY ALUMINUM REGULATION
Description
Loss in Consumer Surplus
Net Gain in Producer Surplus
Total Loss in Surplus or Estimate of Economic Costs
Amount (millions of 1994S)
$148.9
72.5
$76.4
94
-------
5.0 PRIMARY ECONOMIC IMPACTS AND FINANCIAL IMPACT ANALYSIS
5.1 INTRODUCTION
Estimates of the primary economic impacts resulting from implementation of the
secondary aluminum NESHAP and the results of the financial impact analysis are presented in
this chapter. Primary impacts include changes in the market equilibrium price and output levels,
changes in the value of shipments or revenues to domestic producers, and plant closures. The
financial impact analysis assesses the ability of affected firms to raise capital and the impacts of
control costs on firm profitability.
5.2 ESTIMATES OF PRIMARY IMPACTS
The partial equilibrium model is used to analyze the market outcome of the regulation.
As outlined in Chapter 3 of this report, the purchase of emission control equipment will result in
an upward vertical shift in the domestic supply curve for the secondary aluminum market. The
height of the shift is determined by the cash flow required to offset the per unit increase in
production costs. Since the control costs vary for each of the model plants, the post-control
supply curve is segmented, or a step function. Since the underlying production costs for each
facility are unknown, a worst case assumption was necessary. The facilities with the highest
control costs per unit of production were assumed to also have the highest pre-control per unit
cost of production. Thus, firms with the highest per unit cost of emission control are assumed to
be marginal in the post-control market.
Foreign demand and supply are assumed to have the same price elasticities as domestic
demand and supply, respectively. The United States had a negative net scrap export balance for
the year 1994 and, thus, a negative trade balance for secondary aluminum is assumed. Net
exports are therefore negative in the baseline market models. Foreign and domestic post-control
supply are added together to form the total post-control market supply. The intersection of this
post-control supply with market demand will determine the new market equilibrium price and
95
-------
quantity in the secondary aluminum industry.
Table 5-1 presents the primary impacts predicted by the partial equilibrium model. The
anticipated per ton price increase of $11.17 equates to less than a 1 cent per pound increase in the
price of secondary aluminum. The percentage increase in price anticipated as a result of this
regulation is 0.75 percent. Domestic production is expected to decrease by 64 thousand tons
annually or by 0.49 percent.
The value of domestic shipments, or revenues, for domestic producers is expected to
increase by 0.25 percent. Economic theory predicts that revenue increases are expected to occur
when prices are increased for products which have an inelastic price elasticity of demand,
holding all other factors constant. This revenue increase results because the percentage increase
in price is greater than the percentage decrease in quantity for goods with inelastic demand.
It is anticipated that there will be between zero to one closure as a result of this
regulation. Firms that have post-control supply prices that exceed the market equilibrium price
are assumed to close or cease to produce secondary aluminum. This assumption is consistent
with the theory of perfect competition which presumes that all firms in the industry are price
takers. In reality, firms with the highest per unit control costs may not have the highest
underlying cost of production as postulated in the analysis. This is a worst-case assumption that
is likely to bias the impact results and as a result, overstate the number of plant closures and other
adverse effects of the emission controls. A sensitivity analysis assuming that the market
experiences the average per unit cost of emission control costs for all model plants is presented in
Appendix A. All primary market impacts are diminished with this assumption. Specifically, no
facility closures are predicted with the average per unit cost of control assumption.
Of further note is the uncertainty associated with the estimates of facility-level production
quantities and national production. Model plant estimates were used in this analysis that were
based upon data obtained from the EPA information collection requests. The national secondary
aluminum production is based upon the model plant estimates and the estimated number of
96
-------
facilities nationwide. Published production data estimates were assumed to understate
production in this industry since much of the production is captive. Since much of secondary
aluminum production is not sold as an end-use product but rather used as an input to produce
other end-use products by the secondary aluminum producer, the EPA information collection
data are assumed to be a more accurate source of production data.
TABLE 5-1. SUMMARY OF PRIMARY ECONOMIC IMPACTS
OF THE SECONDARY ALUMINUM NESHAP
Amount
Percentage
Price
Increases'
11.17
0.75
Estimated Impacts4
Value of
Domestic
Production2 Shipments3
(0.064) 48.777
fO.49^ 0.25
Facility
Closures
0-1
NOTES: 'Prices are shown in dollars per ton (1994 dollars). Equates to a price increase of less than
one cent per pound.
2Annual production quantities are shown in millions of tons.
3Values of domestic shipments are shown in millions of 1 994 dollars.
"Brackets indicate decreases or negative values.
The estimated primary impacts reported for the secondary aluminum industry depend
upon the set of parameters used in the partial equilibrium model. Two of the parameters, the
price elasticity of demand and the price elasticity of supply, have some degree of estimation
uncertainty. For this reason, a sensitivity analysis was conducted. The results of these analyses
are presented in Appendix A. Sensitivity analyses were performed for low- and high-end
estimates of demand and supply elasticities, respectively. In general, the sensitivity analysis
shows that the estimated primary impacts are relatively insensitive to reasonable changes in the
price elasticity of demand and price elasticity of supply estimates.
97
-------
5.3 FINANCIAL IMPACT ANALYSIS
The financial impact analysis involves examining cost-to-sales ratios. Model plant
estimates of production and the 1994 average market price of Secondary Aluminum No. 380
alloy were used to estimate annual revenues per model plant. The emission control costs for
each model plant reported in Chapter 4 of this document are used in the analysis. Cost-to-sales
ratio refers to the change in annualized control costs divided by the sale revenues of a particular
good or goods being produced in the process for which additional pollution control is required.
It can be estimated for either individual firms or as an average for some set of firms such as
affected small firms. While it has different significance for different market situations, it is a
good rough gauge of potential impact. If costs for the individual (or group) of firms are
completely passed on to the purchasers of the good(s) being produced, it is an estimate of the
price change (in percentage form after multiplying the ratio by 100). If costs are completely
absorbed by the producer, it is an estimate of changes in pretax profits (in percentage form after
multiplying the ratio by 100). The distribution of costs to sales ratios across the whole market,
the competitiveness of the market, and profit to sales ratios are among the obvious factors that
may influence the significance of any particular cost-to-sales ratio for an individual facility.
Table 5-2 summarizes the cost-to-sales ratios for each model plant and for the average
facility affected by the regulation. The cost-to-sales ratios are less than 1 percent for each model
plant and range from 0.02 to 0.97 percent for model plant eight and model plant one,
respectively. The average cost-to-sales ratio for the industry is 0.37 percent. The magnitude of
these financial ratios indicates that the regulation will not impose a significant financial impact
on firms in the industry.
5.4 LIMITATIONS
Several qualifications of the primary impact results presented in this chapter are required.
98
-------
A single national market for a homogenous product is assumed in the partial equilibrium
analysis. Secondary aluminum is not a perfectly homogenous product market and product
differentiation does exist. There may also be some regional trade barriers that would protect
individual secondary aluminum producers. In the analysis, the assumption is made that the
TABLE 5-2. MODEL PLANT COST-TO-SALES RATIOS
OF THE SECONDARY ALUMINUM NESHAP
Model Plant
1
2
3
4
5
6
7
8
Average
Cost-to-Sales Ratio (%)
0.97
0.52
0.82
0.71
0.22
0.22
0.81
0.02
0.37
facilities represented by model plants with the highest per unit control costs are marginal in the
post-control market. This assumption presumes that the facility with the highest per unit cost of
emission controls also has the highest cost average per unit of production, and the cost to the
marginal facility is the cost that will influence the market outcome. The highest per unit control
costs are estimated to significantly exceed the average per unit cost of control for the proposed
regulation. The result of the foregoing list of qualifications is overstatement of the impacts of the
chosen alternative on the market equilibrium price and quantity, revenues, and plant closures.
.# ^
Finally, some facilities may find it profitable to expand production in the post-control market.
99
-------
This would occur when a firm found its post-control incremental unit costs to be smaller than the
post-control market price. Expansion by these firms would result in a smaller decrease in output
and increase in price than would otherwise occur.
This analysis does not fully recognize the unique characteristics of the interrelationship
between the production of secondary aluminum and primary aluminum. Due to data, time, and
resource constraints, the secondary aluminum market analysis does not directly model the market
impact of the secondary aluminum NESHAP on the scrap, primary aluminum, and secondary
aluminum markets and the interrelationship between these markets. As discussed in the industry
profile section of this report, secondary aluminum and primary aluminum are very close demand
substitutes, with each product being interchangeable for end-use market purposes in many cases.
The interrelationship between secondary and primary aluminum is further complicated by the
fact that the production of aluminum directly impacts the supply of secondary aluminum since
recycled aluminum or scrap is a significant input into the production of secondary aluminum.
From the demand perspective, the existence of primary aluminum is likely to make demand more
elastic for secondary aluminum, ceteris paribus. The proposed regulation will lead to a decrease
in the supply of secondary aluminum and an increase in the demand for primary aluminum, all
other factors held constant. This would result in an increase in the market equilibrium price of
primary aluminum and an increase in the quantity demanded. Since scrap originating from
primary and secondary aluminum products are recycled, the market impact of an increase in the
market equilibrium quantity of primary aluminum is to increase the supply of recycled
aluminum. In contrast, a decrease in the market equilibrium quantity supplied of secondary
aluminum potentially decreases the supply of scrap or recycled aluminum. These factors will
have offsetting effects, and the timing and magnitude of the impact of the proposed regulation on
the scrap market is uncertain. In order to estimate the market consequences of secondary market
impacts for the scrap market, more detailed information concerning the origin of scrap and the
vintage of recycled scrap is needed. Such data are not readily available. Since the primary
market impacts of the regulation for the secondary aluminum industry are estimated to be
minimal with price and quantity changes of less than 1 percent, secondary impacts on the scrap
100
-------
and primary aluminum markets are likely to be minimal also.
The results of the sensitivity analysis of demand and supply elasticities are reported in
Appendix A. For the price elasticity of demand, an upper bound of-0.80 is used in the
sensitivity analysis. This elasticity is the estimate found in the literature for the overall
aluminum metal (primary and secondary aluminum). An estimate of-0.16 is used as a lower
bound price elasticity of demand in the sensitivity analysis, and this estimates represents a
decrease of one standard deviation from the estimate used in the analysis. These results show
less adverse impacts on producers when demand is less elastic, or when supply is less elastic, in
terms of reduction in market output and reduction in value of domestic shipments. The results of
the economic analysis are relatively insensitive to reasonable variations in the price elasticity of
demand or the price elasticity of supply inputs.
Baseline data limitation also exist in the analysis and lend uncertainty to the primary
market impacts reported. The baseline market price used in the analysis is the price of No. A-
380 alloy, and this price is used as a surrogate for the market price of all secondary aluminum.
No. A-380 alloy is a mid-range priced secondary aluminum alloy and a representative estimate of
the market price of the product. However, secondary aluminum products are differentiated and
the market price of different types of aluminum do differ. Market consumption and production
data reported in public data sources do not reflect captive production in the market and thus
could not be used for the analysis. Data from the EPA information collection request were used
to estimate annual production in the market based upon the assumption that these data are more
reflective of production in the market than published sources of data. Since there is uncertainty
about the baseline market equilibrium price and quantity in the secondary aluminum market,
there is also uncertainty about annual revenues for the industry that are a function of the market
price and production sold in the market during the base year.
The financial inipact analysis also has limitations. Model plant data were used in the
analysis due to the lack of publicly available financial data. To the extent that annual sales
101
-------
revenue of firms in the industry differs from the estimates based on model plant production and
the price of No. A-380 secondary aluminum alloy, the cost-to-sales ratios may be understated or
overstated.
5.5 SUMMARY
In general, the primary market impacts of this regulation are expected to be minimal with
price increases and production decreases of less than one percent. A market price increase of
0.75 percent and domestic production decrease of 0.49 percent are predicted. Revenues or the
value of domestic shipments for the industry are expected to increase by 0.25 percent. The
increase in the value of shipments results because the price elasticity of demand for secondary
aluminum is inelastic. Products that demonstrate inelastic price elasticity of demand are
characterized by larger percentage price increases than production percentage decreases
occurring when prices are increased. For products with inelastic demand, a price increase leads
to increases in revenue or value of shipments. Individual facilities within the industry may
experience revenue increases or decreases, but on average the industry revenues are anticipate to
increase slightly with this regulation. Potentially, one facility may possibly close as a result of
the regulation. Financial impacts, as measures by cost-to-sales ratios for the firms in the
industry, are minimal and estimated to be less than one percent for all model plants.
102
-------
6.0 SECONDARY ECONOMIC IMPACTS
6.1 INTRODUCTION
In addition to impacts on price, production, and revenue, implementation of emission
controls is likely to have secondary impacts on the labor market, international trade, substitute
markets such as the primary aluminum industry, and regional effects. The potential changes in
employment, balance of trade, and substitute market impacts for industries such as the primary
aluminum industry, and regional impact distribution are presented individually below.
6.2 LABOR MARKET IMPACTS
The estimated labor impacts associated with the NESHAP are based on the results of the
partial equilibrium analyses of the secondary aluminum industry, and are reported in Table 6-1.
The number of workers employed by firms in this industry is estimated to decrease by up to 117
workers as a result of the emission controls. These job losses are considered transitional in
nature. The estimated loss in number of workers results primarily from projected reductions in
levels of production reported in Chapter 5 for the secondary aluminum industry. Gains in
employment anticipated to result from operation and maintenance of control equipment have not
been included in the analysis due to the lack of reliable data. Estimates of employment losses do
not consider potential employment gains in industries that produce substitute products such as
the primary aluminum industry. Similarly, losses in employment in industries that use secondary
aluminum as inputs or in industries that provide complement goods are not considered. The
changes in employment reflected in this analysis are only direct employment losses due to
estimated reductions in domestic production of secondary aluminum.
103
-------
TABLE 6-1. SUMMARY OF SECONDARY IMPACTS OF
THE SECONDARY ALUMINUM INDUSTRY
Estimated Impacts'
Secondary Aluminum Labor Input2 Foreign Trade3
Imports Exports
Amount (117) 0.027 (0.003)
Percentage (0.49%) 1.75% (0.25%)
NOTES: 'Brackets indicate decreases or negative values.
^Indicates estimated reduction in number of jobs.
3Reduction in import and exports in millions of tons
The loss in employment is relatively small in terms of number of jobs lost. The
magnitude of predicted job losses directly results from the relatively small estimated decrease in
production and the relatively low labor intensity in the secondary aluminum industry.
6.3 FOREIGN TRADE
The implementation of the NESHAP will increase the costs of production for domestic
secondary aluminum producers relative to foreign producers, all other factors being equal. This
change in the relative price of imports will cause domestic imports of secondary aluminum to
increase and domestic exports of secondary aluminum to decrease. The overall balance of trade
for secondary aluminum is currently negative (imports exceed exports). (Note trade balances are
estimated based upon the ratio of scrap imports and exports, respectively.) The NESHAP is
likely to cause the balance of trade to become more negative. The estimated impacts on net
exports for the secondary aluminum industry is to decrease net exports by 0.030 million tons per
year. This includes an increase in imports of approximately 1.75 percent or 0.027 million tons
and a decrease in exports of approximately 0.25 percent or 0.003 million tons per year. The
predicted changes in imports and exports are reported in Table 6-1.
104
-------
6.4 REGIONAL IMPACTS
No significant regional impacts are expected to result from implementation of the
NESHAP. Since the estimated economic impacts of the regulation are anticipated to be minimal,
the effect of the regulation is not expected to adversely impact one region of the country relative
to another.
6.5 SUBSTITUTE PRODUCT MARKETS
The industry profile section of this report outlines substitutes for aluminum that include
glass, plastics, magnesium, steel, copper, wood, titanium, graphite, paper, and fiber epoxies. The
degree of substitutability of these products for aluminum depends upon end-use product. For
example, plastic and glass are substitutes for aluminum when used as beverage containers. To
the extent that the price of secondary aluminum increases as a result of this regulation, the
substitute markets may experience increases in demand as consumers of aluminum substitute
these products for aluminum. As discussed in Chapter 5, the primary impacts of the secondary
aluminum NESHAP are anticipated to be minimal with predicted price increases of less than 1
percent. Thus, impacts on secondary substitute markets are likely to be minimal as well.
The aluminum market consists of primary aluminum and secondary aluminum
production. As previously discussed, the interrelationship between these products is somewhat
unusual. Primary and secondary aluminum are substitute products from a demand standpoint for
most end-use applications. Additionally, primary and secondary aluminum production directly
impacts the stock of scrap aluminum which is an input into the production of secondary
aluminum. Increases in the price of secondary aluminum are predicted to occur as a result of this
regulation. To the extent secondary aluminum prices do increase, an increase in the demand for
primary aluminum will likely occur. This increase in demand will lead to a higher market
equilibrium price and an increase in the quantity of primary aluminum produced and sold. As
discussed in Section 5.4, the impact of this regulation on the market for scrap is uncertain. The
105
-------
increase in the market equilibrium quantity of primary aluminum produced will potentially
increase the quantity of scrap available for recycling and secondary aluminum production, ceteris
paribus. However, the decrease in market equilibrium quantity of secondary aluminum may
potentially decrease the supply of scrap. These impacts are offsetting and the ultimate market
impacts are unknown. The magnitude of the primary market impacts of the regulation to the
secondary aluminum market are minimal with price and quantity changes of less than one
percent predicted. Based upon these findings, it is reasonable to assume that the secondary
market impacts to the primary aluminum and scrap markets will be minimal also.
6.6 LIMITATIONS
The estimates of the secondary impacts associated with the emission controls are based
on changes predicted by the partial equilibrium model for the secondary aluminum industry. The
limitations described in Section 5.4 of the previous chapter are also applicable to the secondary
economic impacts reported in this chapter. As previously noted, the employment losses do not
consider potential employment gains for operating the emission control equipment. It is
important to note that the potential job losses predicted by the model are only those which are
attributable to the estimates of production losses in the secondary aluminum industry. Likewise,
the gains or losses in markets indirectly affected by the regulations, such as substitute product
markets, complement products markets, or markets that use secondary aluminum as an input to
production, have not been considered, except as specifically discussed previously. Baseline trade
balance data for the secondary aluminum market were unavailable in the literature. In lieu of this
information, scrap import and export ratios were assumed for the secondary aluminum market.
To the extent that the scrap trade ratios do not reflect baseline imports and exports of secondary
aluminum, the estimated trade impacts may be overstated or understated.
6.7 SUMMARY
The estimated secondary economic impacts are relatively small. As many as 117 job
106
-------
losses may occur nationwide. A decrease in exports of secondary aluminum of 0.25 percent and
an increase in imports of 1.75 percent annually is predicted. The secondary aluminum NESHAP
will likely result in an increase in the demand for primary aluminum and other substitutes. This
increase in demand for primary aluminum will result in an increase in market equilibrium price
and output of primary aluminum, ceteris paribus. No significant regional impacts are anticipated.
107
-------
7.0 POTENTIAL SMALL BUSINESS IMPACTS
7.1 INTRODUCTION
The Regulatory Flexibility Act (RFA) requires an analysis of the potential effects of
proposed regulations on small business entities. Specifically, the RFA requires that a
determination be made as to whether the subject regulation will significantly impact a substantial
number of small entities. The RFA was amended on March 29, 1996 when the President signed
the Small Business Regulatory Enforcement Fairness Act (SBREFA) into law. The SBREFA
strengthened the RJFA's analytical and procedural requirements in making these determinations.
7.2 SMALL BUSINESS CLASSIFICATION
The Small Business Administration (SBA) sets size standards for small entities at the 4-
digit Standard Industrial Code (SIC) level. These standards are set based on either company-
wide employment or annual receipts. The appropriate size standard used to determine whether a
particular firm is identified as a small entity is based on the SIC code that defines the firm's
activities. For the secondary aluminum rule, SIC's 3341 and 3353 will be regulated. The size
standard for a small entity in SIC 3341 is less than 500 employees, and for SIC 3353, less than
750 employees. Since many firms engage in more than one SIC, the limit of 750 employees was
applied as the small business standard to all firms that will be affected by this regulation.
The EPA received responses to an information collection request from 135 facilities
producing products in SIC's 3341 (secondary smelting and refining of nonferrous metals) and
3353 (aluminum sheet, plate, and foil)124; however-, it is thought that there are in excess of 400
facilities which produce these products. To define the small business entities, the 135 facilities
108
-------
were matched to their parent companies. It was determined that 33 of these companies meet the
SBA definition of a small business entity (less than 750 employees).125126
7.3 METHODOLOGY
The analysis of small business impacts for the secondary aluminum industry will focus on
a comparison of compliance costs as a percentage of sales (cost/sales ratio). Other methods that
can be used to determine whether impacts will be significant are the comparison of compliance
costs to production costs or to capital available to small firms, and an analysis on the potential
for closure. However, the information necessary to make such comparisons are generally
considered proprietary by small business firms.
Cost-to-sales ratio refers to the change in annualized control costs divided by the sales
revenues of a particular good or goods being produced in the process for which additional
pollution control is required. It can be estimated for either individual firms or as an average for
some set of firms such as affected small firms. While it has different significance for different
market situations, it is a good rough gauge of potential impact. If costs for the individual (or
group) of firms are completely passed on to the purchasers of the good(s) being produced it is an
estimate of the price change (in percentage form after multiplying the ratio by 100). If costs are
completely absorbed by the producer it is an estimate of changes in pretax profits (in percentage
form after multiplying the ratio by 100). The distribution of costs to sales ratios across the whole
market, the competitiveness of the market, and profit to sales ratios are among the obvious
factors that may influence the significance of any particular cost-to-sales ratio for an individual
facility.
Due to the number of facilities and variety of processes used in this industry, model
plants were developed to categorize facilities based on possible combinations of processes that
are performed. Eight model plants were created and annual compliance costs were calculated for
each one.127 The individual facilities were then assigned to the model plant that most closely fit
109
-------
their process structure, and the annual compliance cost for that model plant was used in
calculating the company's cost/sales ratio.
Two alternative approaches were used to estimate the sales revenues for the affected
small businesses. If actual sales data were available, these data were used to compute cost-to-
sales ratios for affected entities. In cases where the actual sales data were unavailable, model
plant revenues were estimated based upon the estimated model plant annual production and the
average 1994 price of secondary aluminum alloy A-380128129130131
The individual facilities were then assigned to the model plant that most closely fit their
process structure, and the annual compliance cost for that model plant was used in calculating the
company's cost/sales ratio. If actual sales data were available at the company level, this figure
was divided by the model plant annual compliance cost to get a cost/sales ratio for the company.
If actual sales data were not available for a company, the model plant cost/sales ratio was used.
It was determined that there would be no annual compliance costs associated with model plant 9.
therefore, the regulation would have no affect on the facilities that were allocated to this model
plant.
7.4 SMALL BUSINESS IMPACTS
Cost-to-sales data were developed using actual revenue data where available and model
plant estimate revenues for each of the 33 small businesses. As shown in Table 7-1, cost-to-sales
ratios based on model plant data yield ratios of less than 1 percent for each model plant and range
from 0.02 percent to 0.97 percent for model plant 8 and model plant 1, respectively. A summary
of the cost data and actual company annual revenues is shown in Table 7-2 below. As depicted
in this table, the majority of affected small businesses had cost-to-sales ratios below 1 percent.
Ten companies had cost-to-sales ratios above 1 percent. Of these ten companies, only one had a
cost-to-sales ratio abotve 3 percent. A cost-to-sales ratio above 3 percent is an indicator that this
small business may experience a significant economic impact as a result of this regulation.
110
-------
Therefore, only one of the 33 small entities is anticipated to experience significantly adverse
economic impacts as a result of this regulation.
TABLE 7-1. MODEL PLANT COST/SALES RATIOS
Model Plant Number*
1
2
3
7
8
Cost/Sales Ratio (%)
0.97
0.52
0.82
0.81
0.02
*Only model plants in which small business entities were identified are shown.
TABLE 7-2. COMPANY-SPECIFIC COST/SALES RATIOS
Cost/Sales Ratio
0.00%-0.99%
1.00%- 1.99%
2.00%-2.99%
}3.00%
Mean cost/sales ratio = 0.919%
Number of Small Companies in Range*
23
7
2
1
Total 33
* Includes 5 companies assigned model plant cost/sales ratios
111
-------
APPENDIX A
SENSITIVITY ANALYSIS
A-l INTRODUCTION
The market model used to estimate primary and secondary market impacts of the
secondary aluminum NESHAP is subject to uncertainties. Two of the model parameters, the price
elasticity of demand and the price elasticity of supply, have some degree of estimation
uncertainty. Additionally, the per unit cost of emission controls that will be marginal, and thus
influence market price and output changes is also not known with precision. The purpose of the
sensitivity analyses contained in this Appendix is to quantify these uncertainties in terms of
market impact results.
A-2 SENSITIVITY OF THE MODEL ESTIMATES OF THE PRICE ELASTICITY OF
DEMAND AND THE PRICE ELASTICITY OF SUPPLY
The sensitivity analysis contained in this Appendix explores the degree to which the
results presented earlier in this report are sensitive to the estimates of the price elasticities of
demand and supply which were used as inputs to the model. The analysis of the price elasticity of
demand will presume that the supply elasticity is 2.33 as hypothesized in the partial equilibrium
model. Alternatively, the sensitivity analysis of the price elasticity of supply will assume that the
demand elasticity estimate -0.34 postulated in the model of remains unchanged for each.
The results presented in this Appendix are based upon a low-end price elasticities of
demand estimate that differs by one standard error from that used in the model. For the high-end
estimate, a published estimate of the price elasticity of demand for aluminum (primary and
secondary aluminum) is utilized. Table A-l presents the alternative measures of price elasticities
112
-------
of demand and the price elasticity of supply .
TABLE A-l. PRICE ELASTICITY OF DEMAND
AND PRICE ELASTICITY OF SUPPLY ESTIMATES
Demand
Supply
Elasticity Measure
-0.34
2.33
High Estimate
-0.80
3.33
Low Estimate
-0.16
1.33
TABLE A-2. SENSITIVITY ANALYSIS FOR ESTIMATED PRIMARY IMPACTS: LOW-
END AND HIGH-END PRICE ELASTICITY OF DEMAND AND SUPPLY SCENARIOS'
Market Market
Price Change Output Change
(%) (%)
Change in the
Value of
Shipments (%)
Facility
Closures
Price elasticity of
demand
Low estimate
High estimate
0.80
0.64
(0.37)
(0.74)
0.43
(0.11)
0-1
0-2
Price elasticity of
supply
Low estimate
High estimate
0.68
0.78
(0.37)
(0.6H
0.31
0.16
0-1
0-2
NOTES: 1 Brackets indicate decreases or negative values.
The results of the sensitivity analysis relative to the demand elasticity estimates and
supply elasticities are presented in Table A-2. The results of the low-end and high-end demand
113
-------
elasticity scenarios differ very little from the reported model results presented in Chapter 5. The
signs of the changes in price, quantity, and value of shipments are unchanged, and the relative size
of the changes are not significantly different. The results of this analysis tend to present
relatively more favorable results for the affected industries with less elastic demand and less
favorable with more elastic demand. Similarly, the results with the low-end and high-end
elasticity of supply estimates do not differ significantly from the model results discussed in
Chapter 5. More favorable results occur for affected industries when the price elasticity of supply
is less elastic and less favorable with less elastic supply. In summary, the results of the sensitivity
analyses of elasticity estimates do not indicate that the model results are sensitive to reasonable
changes in the price elasticities of demand or supply. This conclusion provides support for
greater confidence in the reported model results.
A-3 SENSITIVITY ANALYSIS OF THE PER UNIT COSTS OF EMISSION CONTROLS
A partial equilibrium model is used to estimate primary and secondary market impacts of
the secondary aluminum NESHAP. The assumption is made in the model that the firm
experiencing the highest per unit cost of emission control also has the highest pre-control cost of
production. As the highest cost producer, this firm is marginal in the industry or the firm likely to
have the most significant impacts as a result of the regulation. This assumption leads to greater
price and quantity changes estimates in the model than alternative assumptions. In reality, this
assumption may not be the case. To determine the sensitivity of the primary market impacts to
this assumption, an additional analysis was conducted. This analysis postulates that the average
cost of emission controls (or per unit cost) is the costs that is relevant in the marketplace. The
results of this analysis are reported in Table A-3.
In general, the primary and secondary market impacts are significantly lowered when the
assumption is made that each facility faces the same industry average per unit emission control
costs. The primary market impacts and the secondary market impacts of this alternative average
cost secondary model are presented in Table A-3. No facility closures are predicted when
114
-------
identical average control costs are assumed. Impacts on price, output, and domestic value of
shipment (or revenue) decreases for the industry are less than 1 percent. Employment losses
decline to 70 for this industry and trade effects are minor. Based upon the results of this analysis,
it is reasonable to conclude that the regulatory impacts are minimal when the assumption is made
that all producers face identical average per unit emission control costs.
TABLE A-3. SENSITIVITY ANALYSIS FOR ESTIMATED PRIMARY AND SECONDARY
MARKET IMPACTS RELATIVE TO THE INDUSTRY AVERAGE
COSTS OF EMISSION CONTROLS
Primary Market Impacts Amount
Change in Market Price (%) 0.29
Change in Domestic Output (%) (0.18)
Change in the Value of Shipments (%) 0.10
Facility Closures (number of facilities) 0
Secondary Market Impacts
Labor Market Decreases (number of workers) 70
Increases in imports (%) 0.67
Decreases in exports(%) (0.10)
115
-------
REFERENCES
1. Ambrosia, J. "Aluminum Tries to Recycle Its Past Success." Iron Age. Vol. 8, No. 6.
June 1992. Page 29.
2. The Aluminum Association. Aluminum Recycling Casebook. Washington, DC.
December 1985. Page 11.
3. Standard and Poor's, Inc. Industry Survey: Metals—Nonferrous. McGraw-Hill. Inc.
New York, NY. December 9, 1993. Page M79.
4. Farin. P. and G. Reibsamen. Aluminum: Profile of an Industry. Metals Week. 1969.
Page 35.
5. Jolly, J.L.W., J.F. Papp, and P.A. Plunkert. Recvcling-Nonferrous Metals. 1991. U.S.
Department of the Interior, Bureau of Mines. May 1993. Page 3.
6. Guntermann, K.L. "Air Pollution Control in the Secondary Aluminum Industry." Land
Economics. Vol. 49, No. 3. August 1973. Page 287.
7. Infotrac-General Business File Database. Tart. P. Merrill Lynch Capital Markets.
March 23, 1993. Page 4.
8. Reference 4. Page 34.
9. The Aluminum Association. Aluminum Statistical Review, 1993. Washington, DC.
1994. Page 31.
10. Reference 4. Page 39.
11. Reference 2. Page 8.
12. Reference 4. Page 39.
13. Reference 2. Page 9.
14. Reference 2. Pages 8-10.
15. Reference 2. Page 10.
16. Reference 2. Page 7.
17. Reference 4. Page 39.
116
-------
18. Reference 2. Page 10.
19. Reference 2. Page 8.
20. U.S. Department of Commerce, Bureau of the Census. Aluminum Ingot and Mill
Products: Summary 1995. Current Industrial Reports. M33D. Table 3B. Issued
May 1996.
21. The U.S. Department of the Interior. U.S. Bureau of Mines. Mineral Facts and
Problems. 1985 Edition. Page 11.
22. Reference 2. Page 6.
23. Gale Research Inc., Kevin Hillstrom, Editor. Encyclopedia of American Industries.
Volume 1: Manufacturing Industries. 1994. Page 751.
24. Reference 6. Page 287.
25. Reference 9. Page 27.
26. The Aluminum Association. U.S. Aluminum Industry Plant Directory. Washington, DC.
March 1992. Page iii.
27. Carlin, J.F., Jr., D. Edelstein, J.H. Jolly, J.L. Jolly, J.F. Papp, and P.A. Plunkert.
Recvcling-Nonferrous Metals, 1992. U.S. Department of the Interior, Bureau of
Mines. 1994. Page 1064.
28. The Aluminum Association. Aluminum Statistical Review. 1991. Washington, DC.
1992. Page 27.
29. Reference 23. Pages 751 and 755.
30. U.S. Department of Commerce, Bureau of Census. "Concentration Ratios in
Manufacturing." Census of Manufactures-Subject Series. 1987 and 1992.
Washington, DC.
31. Reference 30.
32. Reference 23. Page 760.
33. Reference 20.
34. Gale Research Inc. Arsen J. Darnay, Editor. Manufacturing USA—Industry Analyses.
Statistics and Leading Companies. Vol.2. SIC 3312-3999. 1994. Page 1235.
117
-------
35. Fischman. L.L. World Mineral Trends and U.S. Supply Problems. Resources for the
Future. 1980. Page 316.
36. Plunkert, P.A. and E.D. Sehnke. Aluminum. Bauxite, and Alumina—1992. U.S.
Department of the Interior, Bureau of Mines. January 1994. Page 187.
37. Reference 27. Page 1064.
38. Reference 2. Pages 5 and 6.
39. Reference 9. Page 22.
40. Reference 7. Page 3.
41. References. Page 3.
42. Reference 7. Page 4.
43. Reference 9. Page 23.
44. The Aluminum Association (www.aluminum.org). Industry Profile: Aluminum—Facts at
a Glance. June 1998.
45. Reference 9. Page 31.
46. Plunkert, P.A. Aluminum--!997. U.S. Geological Survey . Minerals Information. 1998.
p. 7.
47. The Aluminum Association. Aluminum Statistical Review. 1994. Washington. DC.
1995. Page 30.
48. Reference 21. Page 21.
49. Reference 23. Page 765.
50. Reference 21. Page 18.
118
-------
51. Carlin, J.F., D.L. Edelstein, S.M. Jasinski, J.F. Papp, P.A. Plunkert, and G.R. Smith.
Recvcline-Nonferrous Metals. 1994. U.S. Department of the Interior. Bureau of
Mines. December 1995. Page 12.
52. Reference 9. Page 36.
53. Reference 47. Pages 31, 36, and 37.
54. Minerals Information Team. Recvcline-Metals. 1996. U.S. Geological Survey, Minerals
Information. 1997. Pages 9-14.
55. Reference 35. Pages 183-185.
56. Reference 1. Page 28.
57. Steuteville. R. "The Economics of Aluminum Recycling." BioCvcle. January 1993.
Page 74.
58. Reference 1. Page 30.
59. Reference 57. Page 75.
60. Reference 5. Page 3.
61. Reference 57. Page 77.
62. References. PageM79.
63. Reference 51. Page 1.
64. Reference 47. page 31.
65. Plunkert, P.A., N. Picciototta, and H. Willis. Aluminum Annual Review. 1994. U.S.
Department of the Interior. Bureau of Mines. May 1995. Page 7.
66. Reference 57. Page 75.
67. Reference 65. Page 9.
68. Reference 21. Pages 26 and 27.
69. Reference 51. Page 11.
70. Reference 4. Page 42.
71. Reference 2. Page 45.
119
-------
72. Plunkert, P.A. "The Scrap Content of New Aluminum Continues to Grow." American
Metal Market. March 15, 1993. Page 14a.
73. Reference 7. Page 9.
74. "Metal Statistics-1988." American Metal Market. Pages 25, 30.
75. "Metal Statistics--1992." American Metal Market. Pages 116-118.
76. "Metal Statistics--1994." American Metal Market. Pages 120-121.
77. "Metal Statistics--1995." American Metal Market. Page 443.
78. Reference 51. Page 2.
79. Plunkert, P.A. Aluminum--1995. U.S. Geological Survey, Minerals Information. 1996.
Page 4.
80. Plunkert, P.A. Aluminum--1996. U.S. Geological Survey, Minerals Information. 1997.
Page 3.
81. Reference 46. Page 7.
82. Reference 35. Page 437.
83. Reference 78. Page 2.
84. Reference 57. Page 75.
85. Reference 51. Page 2.
86. Carlin, J.F., Jr., D. Edelstein, S.M. Jasinski, J.F. Papp, P.A. Plunkert, and G.R. Smith.
Minerals Yearbook. Recycling-Nonferrous Metals. 1993. U.S. Department of
the Interior, Bureau of Mines. 1994. Page 916.
87. Duns Analytical Services. Industry Norms and Key Business Ratios: Desk-Top Edition.
1992-93. Dun and Bradstreet Corporation. 1993. Page 91.
88. Duns Analytical Services. Industry Norms and Key Business Ratios: Desk-Top Edition.
1993-1994. Dun and Bradstreet Corporation. 1994. Page 83.
89. Dun and Bradstreet, Inc. Dun's Census of American Business. 1994. Dun and Bradstreet
Corporation. 1993. Page 10.
90. Reference 1. Page 29.
120
-------
91. Reference 57. Page 77.
92. Reference 57. Page 75.
93. Reference 33. Pages 1235 and 1245.
94. Kuster, T. "Scrap Aluminum Ingot Shakeout Seen Waning." American Metal Market.
May 11,1992. Page 8.
95. Reference 89. Pages 91-92.
96. Reference 5. Page 6.
97. Journal of Metals. "Roundtable: Previewing the Next 10 Years in the Aluminum
Industry." JOM. Vol. 43, No. 5. May 1991. Page 61.
98. Reference 36. Page 187.
99. Dialog Business Database. McGraw-Hill Publishers. "GM Aluminum Consumption Up
40%, But Volatility a Concern." Metals Week. Vol. 237., No. 2. June 17, 1996.
Page 8.
100. References. Page 5.
101. Reference 7. Page 6.
102. Dialog Business Database. McGraw-Hill Publishers. "IMCO, Alchem to Build
Aluminum Recycling Plant." Metals Week. November 6, 1995. Page 4.
103. Dialog Business Database. McGraw-Hill Publishers. "Ravenswood Doubling Plate
Ouptut, Shifting Product Mix." Metals Week. Vol. 67, No. 45. November 11,
1996. Page 11.
104. Reference 105. Page 11.
105. Dialog Business Database. McGraw-Hill Publishers. "Auto Industry Adds to Rising
Aluminum Consumption." Metals Week. Vol. 65, No. 40. October 3, 1994.
Page 10.
106. Dialog Business Database. McGraw-Hill Publishers. "Metals Sellers Waiting to Gain
from U.S. Recovery." Metals Week. Vol. 65. No. 2. January 17, 1994.
Page 2.
107. Hyman, David N. Economics. Homewood, IL: Irwin Publishing. 1989. Pp. 214-215.
121
-------
108. Reference 76.
109. Compiled from responses by each plant in the industry to the Information Collection
Requests issued under Section 114 of the Clean Air Act Amendments.
110. Reference 47.
111. U.S. Office of Management and Budget. "Economic Analysis of Federal Regulations
Under Executive Order 12866" January 11, 1996.
http:/www. whitehouse.gov/WH/EOP/OMB/html/miscdoc/riaguide.html.
112. Reference 112.
113. Banks. R. E. Bauxite and Aluminum: An Introduction to the Economics of Nonfuel
Minerals. Lexington, MA. 1979. Page 90.
114. Reference 113.
115. Pindyck, Robert S. and Daniel L. Rubinfeld. Econometric Models and Economic
Forecasts. 2nd edition. McGraw Hill Publishing. 1981. .p. 174-201.
116. Reference 108.
117. Reference 9.
118. U.S. Department of Commerce, Bureau of the Census. Annual Survey of Manufactures.
Washington, D. C. 1958-1994.
119. U.S. Department of Commerce. Bureau of the Census. Bureau of Economic Analysis.
Statistical Abstract of the United States. 1994. 114th Edition. Washington, D. C.
1994 p. 706.
120. Memorandum from Juan Santiago to Linda Chappell. Model Plants Production Capacity.
June 6, 1996.
121. Reference 109.
122. Wright, Milton. Regulatory Impacts for Secondary Aluminum MACT Standards. Memo
from Milton Wright of Research Triangle Institute to Juan Santiago of the EPA.
September 17, 1998.
123. Reference 122.
124. Compiled from responses by each plant in the industry to the Information Collection
Request issued under section 114 of the Clean Air Act Amendments.
122
-------
125. Dialog Business Database. Dun and Bradstreet. Duns Financial Records Plus (Dialog®
File 519).
126. Dialog Business Database. Dun and Bradstreet. Duns Market Identifiers (Dialog® File
516).
127. Reference 122.
128. Plunkert, P. A.. Minerals Yearbook-Aluminum. 1994. U.S. Department of the Interior,
U.S. Geological Survey. 1995. Page 2.
129. Reference 79.
130. Reference 80.
131. Reference 46.
123
-------
TECHNICAL REPORT DATA
(Please read Instructions on reverse before completing)
i. REPORT NO.
EPA-452/R-99-002
3. RECIPIENT'S ACCESSION NO
4. TITLE AND SUBTITLE
Economic Impact Analysis for the Proposed Secondary Aluminum
Production National Emission Standard for Hazardous Air
Pollutants
5. REPORT DATE
November 1998
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
Linda M. Chappell/ Jan King, Innovative Strategies and
Economics Group ^^^^^^^^
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
U.S. Environmental Protection Agency
Office of Air Quality Planning and Standards
Research Triangle Park, NC 27711
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
12. SPONSORING AGENCY NAME AND ADDRESS
Director
Office of Air Quality Planning and Standards
Office of Air and Radiation
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
13 TYPE OF REPORT AND PERIOD COVERED
14. SPONSORING AGENCY CODE
EPA/200/04
15. SUPPLEMENTARY NOTES
16. ABSTRACT
This report evaluates the impacts of the proposed rule for controls of hazardous air pollutants (HAPs) in
the Secondary Aluminum industry. Social costs are estimated by evaluating costs of compliance with the
rule and associated market impacts including: potential price changes in secondary aluminum, possible
changes in the quantity of secondary aluminum produced, small entity impacts, and employment impacts.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b. IDENTIFIERS/OPEN ENDED TERMS
c. COSATI Field/Group
economic impacts
small entity impacts
Air Pollution control
Economic Impact Analysis
Regulatory Flexibility Analysis
18. DISTRIBUTION STATEMENT
Release Unlimited
19. SECURITY CLASS (Report)
Unclassified
21. NO. OF PAGES
152
20. SECURITY CLASS (Page)
Unclassified
22. PRICE
EPA Form 2220-1
------- |