United States
Environmental Protection
Aaencv
Office of Air Quality
Planning and Standards
Research Triangle Park, NC 27711
FINAL REPORT
EPA-453/R-94-039
June 1994
Air
ERA ECONOMIC IMPACT ANALYSIS
OF THE
SECONDARY LEAD SMELTERS
NESHAP - FINAL
-------
Economic Impact Analysis
of the
Secondary Lead Smelters
NESHAP-- Final
Emissions Standards Division
Lisa Conner
U.S. Environmental Protection Agency
Office of Air and Radiation
Office of Air Quality, Planning and Standards
MD-13; Research Triangle Park, N.C. 27711
June 1994
-------
(Disclaimer)
This report is issued by the Emission Standards Division of the
Office of Air Quality Planning and Standards of the Environmental
Protection Agency. It presents technical data on the National
Emission Standard for Hazardous Air Pollutants (NESHAP), which is
of interest to a limited number of readers. It should be read in
conjunction with the Industry Profile for the Secondary Lead
Smelters NESHAP (June 1994). Both the Economic Impact Analysis
and the Industry Profile are in the public docket for the NESHAP
proposal. Copies of these reports and other material supporting
the proposal are in Docket A-92-43 at EPA's Air and Radiation
Docket and Information Center, Waterside Mall, Room M1500,
Central Mall, 401 M. Street SW, Washington, D.C. 20460. The EPA
may charge a reasonable fee for copying. Copies are also
available through the National Technical Information Services,
5285 Port Royal Road, Springfield, Virginia 22161. Federal
employees, current contractors and grantees, and non-profit
organizations may obtain copies from the Library Services Office
(MD-35), U.S. Environmental Protection Agency; Research Triangle
Park, N.C. 27711; phone (919)541-2777.
-------
TABLE OF CONTENTS
Page
List of Tables iii
List of Figures iv
1.0 INTRODUCTION 1
1.1 SCOPE 1
1.2 ORGANIZATION 2
1.3 SUMMARY .3
2.0 BACKGROUND 5
2.1 IMPACTED COMPANIES AND FACILITIES 5
2.2 CONTROL COSTS 10
3.0 THEORY AND METHODOLOGY 10
3.1 INDUSTRY-WIDE IMPACTS 10
3.1.1 Market Framework 10
3.1.1.1 Basic Model 10
3.1.1.2 Price Elasticities of SF and
SD.P 14
3.1.1.3 Alternative Model 18
3.1.2 Price Elasticity of Demand 20
3.1.3 Market Response 21
3.1.4 Methodology 26
3.2 PER-FACILITY IMPACTS . . . 27
4.0 ESTIMATION OF THE PRICE ELASTICITY OF SUPPLY 28
4.1 INDEPENDENT VARIABLES 32
4.2 REGRESSION RESULTS 34
5.0 ECONOMIC IMPACTS 37
5.1 INDUSTRY-WIDE 37
5.2 PER-FACILITY 42
5.2.1 Baseline Risk 42
5.2.2 Cost Absorption 43
-------
TABLE OF CONTENTS
(Continued)
5.2.2.1 Baseline Profitability 45
5.2.2.2 Calculations 48
5.2.3 Capital Availability 51
5.3 SMALL BUSINESSES • 53
5.4 SUMMARY AND CONCLUSIONS 58
6.0 REFERENCES 63
APPENDIX A - INPUTS TO THE REGRESSION ANALYSIS
-11-
-------
LIST OF TABLES
Page
TABLE 1. THE U.S. SECONDARY LEAD INDUSTRY, DECEMBER 1993 ... 6
TABLE 2. CAPITAL CONTROL COSTS 11
TABLE 3. TOTAL ANNUALIZED CONTROL COSTS 12
TABLE 4. REGRESSION MODEL 30
TABLE 5. RESULTS OF THE REGRESSION ANALYSIS 35
TABLE 6. ESTIMATED ANNUAL INDUSTRY-WIDE OUTPUT IMPACTS ... 41
TABLE 7. NUMBER OF MAJOR SOURCES WITH A PER-POUND IMPACT
OF TOTAL ANNUALIZED CONTROL COSTS EXCEEDING
0.25 CENTS 50
TABLE 8. NUMBER OF SIGNIFICANTLY IMPACTED FACILITIES .... 56
-111-
-------
LIST OF FIGURES
FIGURE 1. CONCEPTUAL FRAMEWORK FOR THE U.S. LEAD MARKET
FIGURE 2. CONCEPTUAL FRAMEWORK FOR THE U.S. LEAD MARKET,
ASSUMING PERFECTLY ELASTIC FOREIGN SUPPLY (SF)
FIGURE 3.
FIGURE 4,
THE EFFECTS ON THE U.S. LEAD MARKET OF CONTROL
COSTS ON SECONDARY LEAD PRODUCERS
Page
13
. 19
22
THE EFFECTS ON THE U.S. LEAD MARKET OF CONTROL
COSTS ON SECONDARY LEAD PRODUCERS, ASSUMING
PERFECTLY ELASTIC FOREIGN SUPPLY (SF) 23
FIGURE 5. THE EFFECT OF CONTROL COSTS ON U.S. SECONDARY LEAD
PRODUCTION IF PRICE DOES NOT CHANGE (P, = P2) . . .
38
-IV-
-------
1. 0 INTRODUCTION
1.1 SCOPE
This report evaluates the economic impacts of a
proposed National Emissions Standard for Hazardous Air
Pollutants (NESHAP) for secondary lead smelters. The report
seeks mainly to determine a) the market response to the
regulation — specifically, impacts on industry-wide output,
employment, and revenue; and b) the ability of each impacted
facility to absorb annual control costs and obtain financing
for capital control costs.
The NESHAP applies to, and therefore the analysis in
this report is conducted for, all 23 secondary lead smelters
in the U.S. Fifteen of these facilities are estimated to
emit more than 25 short tons (about 22.7 metric tons) per
year of hazardous air pollutants (HAPs) and are therefore
classified as "major sources" of HAP emissions. The other
eight facilities are classified as "area sources" because
they are estimated to emit fewer than 25 short tons of HAPs
per year.•
Five control options are evaluated. Control Option 1
consists of various Maximum Achievable Control Technology
(MACT) floor requirements, as well as requirements for
recordkeeping and reporting. Control Option 2 has the same
requirements as Control Option 1 but goes "above" the MACT
floor by requiring a temperature of 1600°F rather than
1300°F for blast-furnace afterburners. (If a facility does
not operate a blast furnace, there is no difference between
Control Option 1 and Control Option 2.) Control Options 3
through 5 add monitoring to the requirements of Control
Option 1. Control Option 3 adds to Control Option 1 the
requirement of continuous opacity monitoring (COM) for
baghouses. Control Option 4 adds to Control Option 3 the
requirement of continuous emissions monitoring (CEM) for
-------
total hydrocarbons (THC). Control Option 5 adds to Control
Option 4 the requirement of CEM for hydrochloric acid (HC1) .
Much of the information needed to perform the analysis
was available from publicly available sources, including
publications of the U.S. Bureau of Mines and the trade
literature, particularly American Metal Market. However, as
the U.S. secondary lead industry is relatively small and
most of the firms are privately owned, there still were data
gaps after reviewing the publicly available information.
Consequently, a telephone survey was initiated. Three firms
representing four facilities (two major sources, two area
sources) participated in the survey. Also responding to the
survey were the U.S. Bureau of Mines, a representative of an
industry trade group, and a lead industry consultant/
analyst.
1.2 ORGANIZATION
In Section 1.3, which follows, the findings of the
economic impact analysis are summarized. Background
information is provided in Section 2.0. This includes
information on the impacted companies and facilities in
Section 2.1 and a summary of control costs in Section 2.2.
In Section 3.0, some theoretical concepts are discussed and
the methodology for the analysis is laid out. The price
elasticity of supply, which is needed to calculate industry-
wide impacts, is estimated in Section 4.0. Economic impacts
are calculated and evaluated in Section 5.0. Industry-wide
impacts are addressed in Section 5.1 and per-facility
impacts are addressed in Section 5.2. Impacts on small
businesses are assessed in Section 5.3, as a Regulatory
Flexibility Analysis is conducted. In Section 5.4, the
economic impacts are summarized, industry-wide and per-
facility impacts are reconciled, and implications of the
impacts are discussed.
-------
1.3 SUMMARY
Lead is an internationally traded commodity whose price
is determined by global market factors. Consequently, U.S.
lead producers have little control over price — they are
essentially price-takers. Recognizing that if any price
increase is achieved in an attempt to recover control costs,
it is likely to be minimal, the economic impact analysis
assumes that no price increase will be achieved and control
costs will have to be absorbed. This will reduce net income
in an industry that appears to be currently (December 1993)
just about breaking even.
Depending on the price elasticity of supply (a measure
of the responsiveness of quantity supplied to a change in
price), annual industry-wide output is estimated to decline
by 1,127-1,948 metric tons under Control Option 1, the least
stringent control option, ranging up to 3,155-5,453 metric
tons under Control Option 5, the most stringent control
option. The upper end of the range under Control Option 5,
5,453 metric tons, represents only 0.61 percent of baseline
production. Industry-wide employment and revenue impacts
are likewise minimal.
In the per-facility analysis, four active (i.e.,
currently operating) major sources and one active area
source are "significantly impacted" by the NESHAP. This
means that, at the facility level, total annualized control
costs exceed 0.25 C/lb and/or, at the company level, capital
control costs are more than five percent of baseline total
assets and post-regulation total liabilities would exceed
two-thirds of baseline total assets if the capital control
costs are financed with debt. Two major sources are
significantly impacted under all five control options
(though the impacts increase in significance as the control
options become more stringent), one is significantly
impacted under Control Option 2, and one is significantly
-------
impacted under Control Option 5. The area source is
significantly impacted under Control Options 2, 4, and 5.
Under current market conditions in the U.S., none of
these facilities are likely to close as a consequence,
however. The impacts of total annualized control costs are
all less than 1 0/lb. In comparison, the price of lead has
risen by 2 or 3 C/lb in the past several months, and lead
supplies are "tight." If price turns back down, though, a
closure or two is possible. The most likely candidates for
closure are the two major sources that are significantly
impacted under all five control options. This is because
both are very possibly "marginal"-— i.e., among the
industry's highest-cost producers — in the baseline. The
likelihood of either facility closing of course increases as
the control options become more stringent.
By making marginal facilities even more marginal, the
NESHAP could also contribute to a closure or two if new.
secondary lead capacity comes on stream in the U.S. GNB,
RSR, and Asarco have all indicated an interest in building a
new secondary lead facility, though their plans are all
indefinite. The replacement of relatively small facilities
by larger, more efficient facilities has perhaps been the
dominant trend in the U.S. secondary lead industry since the
1970s. Over 100 facilities, mostly small, have closed since
1975. If new capacity comes on stream, it is likely to be
at the expense of some existing capacity. The NESHAP may
influence which existing facilities are marginal and
therefore most vulnerable to closure.
The NESHAP will also give all seven of the secondary
lead smelters in the U.S. that are shut down (four major
sources, three area sources) an additional incentive not to
reopen. The incentive not to reopen is greatest for
facilities that are "significantly impacted" (three under
Control Option 1, three under Control Option 2, four under
-------
Control Option 3, six under Control Option 4, and all seven
under Control Option 5), and increases as the control
options become more stringent.
The NESHAP impacts small companies, defined as having
500 or fewer employees, disproportionately. While the
number of significantly impacted smelters owned by small
companies ranges from five under Control Option 1 to 10
under Control Option 5, only one smelter owned by a company
that is not small is significantly impacted (under Control
Options 2, 4, and 5). Small companies are dispropor-
tionately impacted by the NESHAP for two main reasons: l)
they tend to own smaller facilities, which do not benefit
from the economies of scale inherent in the control costs
(i.e., per-unit control costs tend to decrease as facility
size increases), and 2) they have fewer capital resources.
It is also possible that, because of their greater
resources, larger companies have been able to control their
operations more tightly in the baseline. This would make it
easier, and less costly^ to comply with the NESHAP.
2 . 0 BACKGROUND
2.1 IMPACTED COMPANIES AND FACILITIES
The NESHAP will impact all 23 of the secondary lead
facilities in the U.S. These facilities are listed in Table
1. As the table indicates, seven facilities are large, 13
are medium-sized, and three are small. Only 16 of the 23
facilities are currently active; the other seven, including
all three small facilities, are presently shut down.*
* For purposes of this analysis, a secondary lead smelter is
considered "shut down" if its operating equipment has not
been sold. Such a facility has the potential to restart.
If a facility's operating equipment has been sold, it does
not have the potential to restart and is considered
"closed." Such facilities are not included in the
analysis.
-------
TABLE 1. THE U.S. SECONDARY LEAD INDUSTRY, DECEMBER 1993
Facility
Company
Location
Size"
Status
Delatte Metals,
Inc.
East Penn
Manufacturing Co.
Exide Corp.
General Smelting &
Refining Co.
GNB Battery
Technologies'5
Gopher Smelting &
Refining, Inc.
Gulf Coast
Recycling Inc.
Master Metals Inc.
PBX Inc.
Refined Metals
Corp.
Ross Metals, Inc.
RSR Corp.
Sanders Lead Co.
Schuylkill Metals
Corp.
Tejas Resources
Inc.
The Doe Run Co.
Ponchatoula, LA Small
Lyon Station, PA Medium
Muncie, IN Medium
Reading, PA Medium
College Grove, TN Medium
Columbus, GA Medium
Frisco, TX Medium
Vernon, CA Large
Eagan, MN Large
Tampa, FL Medium
Cleveland, OH
Norwalk, OH
Beech Grove, IN
Memphis, TN
Rossville, TN Small
City of Industry, CA Large
Indianapolis, IN Large
Middletown, NY Large
Troy, AL Large0
Baton Rouge, LA Larged
Forest City, MO Medium
Terrell, TX Medium
Boss, MO Medium
Shut down
Active
Active
Active
Active
Active
Active
Active
Active
Active
Small
Medium
Medium
Medium
Shut down
Shut down
Shut down
Shut down
Shut down
Active
Active
Active
Active
Active
Active
Shut down
Active
"Small = annual production capacity less than 20,000 metric tons,
medium = 20,000-75,000 metric tons, large = greater than
75,000 metric tons.
"Before October 1993, GNB Inc.
'Capacity is nominally "large," but early in 1993 production was
cut in half. It is not known if the cutback is temporary or
permanent. (Source: American Metal Market, May 12, 1993,
page 2.)
Capacity is nominally "large," but in May 1993 output was cu<:
by 23 percent. The company hoped that the cutback would be
temporary. (Source: American Metal Market, May 12, 1993,
page 2.)
-------
The 23 facilities are owned by 16 firms. GNB and RSR
have the most facilities — three each. GNB and Doe Run are
both subsidiaries of publicly held companies. GNB's parent
is Pacific Dunlop Limited, a highly diversified,
multinational company with headquarters in Australia (based
on the exchange rate on August 2, 1993, Pacific Dunlop's
sales in the fiscal year ended June 30, 1992 translated to
U.S. $4.0 billion). Doe Run is a subsidiary of Fluor
Corporation (sales in the fiscal year ended October 31, 1991
were $6.7 billion), which is concentrated in engineering
services, construction, and coal production more than in
lead production. Fluor has declared its intent to divest
Doe Run.1 All other secondary lead smelters in the U.S. are
owned by private companies. Typically these companies are
closely held (e.g., by several officers).
Of the 23 total facilities, 15 are major sources of HAP
emissions and eight are area sources. However, only 11
major sources and five area sources are currently active.
Four major sources and three area sources are shut down.
Three firms — East Penn, Exide, and GNB — are
integrated downstream into battery manufacture (i.e., their
secondary lead output is captively consumed in the
manufacture of batteries). This is an important variable in
the U.S. secondary lead industry. Competition for spent
batteries, which account for about 85 percent of all lead
scrap, is intense.* The battery distribution networks of
integrated producers give them an advantage in collecting
spent batteries. Spent batteries can be collected at the
* Lead scrap is the most important input in secondary lead
production. Some other important inputs include coke,
natural gas, electricity, alloying metals such as
antimony, and labor.
-------
retail outlets (e.g., auto parts stores, car dealerships,
service stations, discount stores) where new batteries are
delivered. Through important accounts such as Pep Boys and
Wal-Mart, GNB, for example, encourages consumers to trade in
used batteries when purchasing new ones.2 Exide recently
entered into an agreement to recycle spent batteries
collected by Montgomery Ward (342 stores nationwide), which
is paying consumers one dollar per battery.3
For integrated producers, battery production and
secondary lead smelting can be a "closed loop." For
example, one truck can deliver new batteries to retail
outlets, pick up spent batteries at these outlets (which may
acquire the spent batteries either by offering consumers
payment for them, or as a consequence of state laws
requiring retailers to accept spent batteries from consumers
when new ones are purchased), deliver the spent batteries to
a secondary lead smelter, and transport lead that has been
recycled at the smelter to a battery production facility.
GNB acknowledges that its integrated approach, which it
calls "Total Battery Management" ("manufacturing and
distribution of new lead-acid batteries, responsible
collection, storage and transportation of spent batteries,
safe reclamation of battery materials and use of the
materials in new batteries"), gives it a "significant
marketing edge."4-5 The U.S. Bureau of Mines (BOM) offers
that "the integrated producer with a retail collection
system has more control over secondary scrap supply than a
nonintegrated producer and is therefore better able to
influence the price he pays for lead scrap feedstock."6
Traditionally scrap dealers/battery haulers have been
the major suppliers of lead scrap to secondary lead
smelters. More and more, however, battery manufacturers
-------
have been controlling the flow of lead scrap for recycling.*
This applies to both integrated and nonintegrated battery
manufacturers. Both have special access to spent batteries,
and both, because they consume lead, have a vested interest
in seeing that batteries are recycled. While integrated
battery producers wish to ensure a flow of lead scrap to
their own lead smelting operations, nonintegrated battery
producers will direct shipments of lead scrap to independent
secondary lead smelters. Often this is done under a
"tolling" arrangement, whereby the provider of lead scrap
(i.e., the battery producer) retains title to the lead and
pays a fee to the smelter to have it processed (recycled).
An implication of the increasing dominance of battery
producers in the market for spent batteries is that it is
important for nonintegrated secondary lead smelters to have
connections with battery producers. Otherwise the supply of
lead scrap can be inadequate. RSR, in particular, is
regarded to have strong connections with battery
manufacturers.7 Therefore, despite not being integrated, it
is considered to be well positioned in "battery loops." The
number-one secondary lead producer in the U.S., RSR buys
spent batteries in all 50 states, handling about one-third
of all batteries scrapped in the country.8'9
* The increasing importance of battery producers and the
declining importance of scrap dealers/battery haulers in
the market for spent batteries has several causes,
including 1) increasing aggressiveness by battery
producers, 2) new state laws that encourage or require
consumers to return spent batteries to retail outlets that
sell new batteries, and 3) growing reluctance by scrap
dealers/battery haulers to participate owing to potential
Superfund liability at battery-breaking sites.
-------
In August 1993, RSR reached an agreement to be the lead
investor in the purchase of a 110,000-metric-ton-per-year
secondary lead business in England and France.10 As a
result, the company will become the first to have secondary
lead operations in both the U.S. and Europe.
Doe Run is unique as the only U.S. producer of both
secondary lead and primary lead. Aside from Doe Run and the
three producers integrated into battery manufacture, U.S.
secondary lead producers are typically not significantly
diversified away from secondary lead.
2.2 CONTROL COSTS
Capital control costs and total annualized control
costs are presented in Tables 2 and 3, respectively. Due to
confidentiality, the facility names are not disclosed. A
code scheme is used instead.
Capital costs reflect the up-front costs of pollution
control equipment. Total annualized cost is the sum of
annual operating and maintenance (O&M) costs and annualized
capital costs. Capital costs are annualized using the
"capital recovery factor" based on a 7 percent discount rate
and a useful life that varies depending on the type of
equipment.
Industry-wide capital costs range from $1.4 million
under Control Option 1 to $9.0 million under Control Option
5. Industry-wide total annualized costs range from $2.0
million under Control Option 1 to $5.4 million under Control
Option 5. See Section 1.1 for a description of the control
options.
3.0 THEORY AND METHODOLOGY
3.1 INDUSTRY-WIDE IMPACTS
3.1.1 Market Framework
3.1.1.1 Basic Model. Figure 1 presents a framework
for understanding the U.S. lead market. The figure shows
that there are three sources of supply: domestic primary
10
-------
TABLE 2. CAPITAL CONTROL COSTS
Facility
2
3
4
5
6
8
9
10
12
13
14
15
16
17
19
20
22
23
25
26
27
28
29
Major
or area
source?
M
H
A
M
A
A
M
M
A
M
M
M
M
M
M
A
A
A
M
M
M
H
A
Control
Option 1
$47,000
$54,000
$0
$310,000
$47,000
$0
$280,000
$0
$0
$110,000
$0
$150,000
$0
$0
$0
$0
$0
$0
$260,000
$54,000
$0
$54,000
$0
Control
Option 2
$610,000
$54,000
$0
$310,000
$610,000
$0
$280,000
$0
$0
$110,000
$0
$150,000
$250,000
$0
$0
$0
$0
$0
$260,000
$54,000
$360,000
$54,000
$0
Control
Option 3
$125,200
$132,200
$39,100
$349,100
$125,200
$39,100
$319,100
$78,200
$39,100
$188,200
$78,200
$189,100
$39,100
$39,100
$78,200
$78,200
$39,100
$78,200
$299,100
$132,200
$39,100
$132,200
$78,200
Control
Option 4
$268,800
. $275,800
$182,700
$492,700
$268,800
$182,700
$462,700
$221,800
$182,700
$331,800
$221,800
$332,700
$182,700
$182,700
$221,800
$221,800
$182,700
$221,800
$442,700
$275,800
$182,700
$275,800
$221,800
Control
Option 5
$395,700
$402,700
$309,600
$619,600
$395,700
$309,600
$589,600
$348,700
$309,600
$458,700
$348,700
$459,600
$309,600
$309,600
$348,700
$348,700
$309,600
$348,700
$569,600
$402,700
$309,600
$402,700
$348,700
Total $1,366,000 $3,102,000 $2,734,500 $6,037,300 $8,956,000
Control Option 1 MACT floor
Control Option 2 Above the MACT floor
Control Option 3 MACT floor + COM
Control Option 4 MACT floor + COM + THC CEM
Control Option 5 MACT floor + COM + THC CEM + HCl CEM
11
-------
TABLE 3. TOTAL ANNUAL I ZED CONTROL COSTS
Facility
2
3
4
5
6
8
9
10
12
13
14
15
16
17
19
20
22
23
25
26
27
28
29
Major
or area
source?
M
M
A
M
A
A
M
M
A
M
M
M
M
H
M
A
A
A
M
M
M
M
A
Control
Option 1
$88,000
$63,000
$56,000
$190,000
$58,000
$46,000
$220,000
$91,000
$46,000
$78.000
$90,000
$87,000
$46,000
$77,000
$46,000
$46,000
$51,000
$46,000
$230,000
$63,000
$140,000
$52,000
$46,000
Control
Option 2
$340,000
$63,000
$56,000
$350,000
$290,000
$46,000
$410,000
$91,000
$46,000
$78,000
$90,000
$87,000
$100,000
$77,000
$46,000
$46,000
$51,000
$46,000
$350,000
$63,000
$220,000
$52,000
$46,000
Control
Option 3
$121,000
$96,000
$72,500
$206,500
$91,000
$62,500
$236,500
$124,000
$62,500
$111,000
$123,000
$103,500
$62,500
$93,500
$79.000
$79.000
$67,500
$79,000
$246,500
$96,000
$156,500
$85,000
$79,000
Control
Option 4
$185,600
$160,600
$137,100
$271,100
$155.600
$127,100
$301,100
$188,600
$127,100
$175,600
$187,600
$168,100
$127,100
$158,100
$143,600
$143,600
$132,100
$143,600
$311,100
$160,600
$221,100
$149,600
$143,600
Control
Option 5
$245,900
$220,900
$197,400
$331.400
$215,900
$187.400
$361,400
$248,900
$187,400
$235,900
$247.900
$228,400
$187,400
$218,400
$203,900
$203,900
$192,400
$203,900
$371,400
$220,900
$281,400
$209,900
$203,900
Total
5================
Control Option 1
Control Option 2
Control Option 3
Control Option 4
Control Option 5
$1,956,000 $3.044,000 $2,533,500 $4,019,300 $5,406,200
MACT floor
Above the MACT floor
Mact floor + COM
MACT floor + COM + THC CEM
MACT floor + COM + THC CEM * HCl CEM
12
-------
FIGURE 1. CONCEPTUAL FRAMEWORK FOR THE U.S. LEAD MARKET
Price
p
t{
I I I
Or QO-, Q
Quantity
-------
lead (SD.P) , domestic secondary lead (SD.S) , and foreign lead
(SF) . All three sources of supply are shown in the figure
to diminish in price elasticity (a measure of the
responsiveness of quantity supplied to a change in price)
and approach perfect price inelasticity as quantity
increases. This is because primary lead is subject to
natural-resource constraints, and secondary lead is
constrained by the finite amount of recyclable lead
available. Domestic secondary lead and domestic primary
lead are assumed in Figure 1 to have similar fixed costs;
hence, SD.S and SD.P start at the same point on the price
axis. Foreign lead is assumed to have a transportation cost
disadvantage "t."
The total supply of lead in the U.S., ST, is derived by
summing SD.P, SD.S, and SF horizontally. Note that ST is more
elastic than any of its three components individually. The
market clears at the intersection of ST and demand (D) . At
the market-clearing price, P, quantity supplied is
distributed among the three sources of supply as QF, QD.P,
and QD.S. They add up to total quantity supplied, QT.
3.1.1.2 Price Elasticities of SF and SDP. Although SF
and SD.P will by necessity eventually approach the vertical,
and therefore perfect price inelasticity, there is good
reason to believe that they are highly elastic in their
current ranges (i.e., around QF for SF, and around QD.P for
SD.P) . To begin, the U.S. Bureau of Mines indicated in the
telephone survey that if, hypothetically, U.S. secondary
lead producers were to attempt to increase prices by, say, %
or 1 C/lb, it is a "safe assumption" that foreign lead and
domestic primary lead would "flood the market."11 Another
respondent in the telephone survey, a representative of an
industry trade group, said that in response to an attempt by
domestic secondary lead producers to increase prices, "it is
assumed that at some point lead imports would increase."12 A
14
-------
1992 study of the U.S. lead industry described the supply of
foreign lead as "typically ... highly elastic because of
large foreign reserves."13
The ready availability of foreign lead and perhaps
domestic primary lead, and the substitutability of both of
these sources of lead and domestic secondary lead, also
point to high price elasticities for SF and SD.P. The ready
availability of foreign lead is immediately suggested by the
current record-high world stocks. Stocks of lead on the
London Metal Exchange (LME), the world's clearing-house for
lead and many other metals, stood at 262,250 metric tons on
July 13, 1993, compared to, for example, 126,350 metric tons
on January 2, 1992.14>15 There is also substantial excess
world capacity. Excluding the U.S., world primary lead
capacity exceeded production by 1.75 million metric tons (4
million metric tons of capacity vs. 2.25 million metric tons
produced) in 1989.:6 Perhaps some of the idle world capacity
could be activated for export to the U.S. if prices of lead
in the U.S. were to increase. (Caveat: Outside the U.S.,
lead is mined primarily as a byproduct of zinc or silver.
Therefore, activating the idle capacity would depend greatly
on the profitability of these other metals.)
Further, foreign secondary lead producers have recently
become more aggressive in exporting to the U.S. In the
first 11 months of 1992, U.S. imports of refined lead were
173,672 metric tons (112,800 metric tons, or 65%, of which
were from Canada), compared to 116,473 metric tons in all of
1991.n Particular inroads were made by Mexico and Peru. -
From 1991 to the first 11 months of 1992, Mexico's exports
to the U.S. increased from 22,614 to 51,778 metric tons,
while Peru's increased from 500 to 8,997 metric tons.
Mineroperu, the Peruvian mining company, has indicated its
aim to increase exports to the U.S. by 20 to 30 percent in
1993.l8
15
-------
Importantly, Mexico and Peru have been able to ship to
the southwestern U.S. at a premium over the LME spot price
of only 2 cents per pound.19* In contrast, in 1993, Asarco's
announced (published) premium over the LME spot price has
ranged from 4.25 C/lb for the period January through May to
7.0 0/lb in November, and RSR's "four-corners" premium,
based on both spot and forward LME prices, has ranged from
5.0 C/lb for January-May to 7.75 0/lb in November. (Asarco
and RSR are, respectively, the top primary lead and
secondary lead producers in the U.S. Their premiums are
consistently published and therefore are benchmarks for the
U.S. lead industry.) Clearly, lead from Mexico and Peru is
very price-competitive.
Although foreign secondary lead producers are generally
less efficient than U.S. producers, they tend to have lower
overall costs owing to lower labor costs and less-stringent
environmental regulations.20-21 A respondent in the telephone
survey gave the example that in the U.S. market, the
Taiwanese are able to sell secondary lead for less than U.S.
producers despite procuring spent batteries in the U.S. and
shipping them back to Taiwan for processing.22
It should be mentioned that foreign secondary lead
producers are not immune to two of the biggest current
pressures on U.S. secondary lead producers: low lead prices
and environmental control costs. In August 1993, Immsa, one
of the two Mexican producers of refined lead (51,000 metric
tons in 1992) , indicated that it intends to shut down its
smelting and refining operations sometime in 1994.23-24 Low
lead prices and environmental considerations were cited.
* A "spot" price is for immediate delivery of a product. A
"forward" price, in contrast, is for delivery of a product
at some future date.
16
-------
The company will continue to mine lead and have its
concentrates tolled by the other Mexican lead smelter and
refiner, Penoles, however. Also, due to low lead prices and
the recent appreciation of the yen, MIM Holdings (Australia)
and its Japanese partners have put on hold plans to develop
in Japan a lead smelter with an annual capacity of 60,000
metric tons.25 The smelter was originally scheduled to start
up in the fall of 1995.
Lead may also be readily available from domestic
primary sources. A respondent in the telephone survey said
that domestic primary mines currently have excess capacity
that could be activated.26 The Missouri mines, for example,
are presently significantly curtailed. In the first five
months of 1993, U.S. lead mine production was down 19,000
metric tons, or 11 percent, from the same period in 1992.27
That SF and SD.P are highly price-elastic is also
suggested by the substitutability of foreign lead, domestic
primary lead, and domestic secondary lead. This is because
pure lead and antimonial lead, the alloy commonly used in
lead-acid batteries, are commodities. The specifications
for pure lead and antimonial lead are standardized, and they
can be met by any of the three sources of supply — domestic
secondary, domestic primary, or foreign. Furthermore, when
a customer has a special request (and there are about 300
common alloys of lead), the order can generally be filled
from any of the three sources. Mexican lead, for example,
is high-quality (highly refined), and good-quality lead is
also available on the LME (though LME lead is usually of
lower quality than U.S. and Mexican lead).28 While not all
LME lead is U.S. battery-grade, most is believed to be.29
Secondary lead smelters are differentiated from primary
lead smelters in that they alloy with antimony much more.
The secondaries do not have market power over antimonial
lead, however, because the primaries are able to produce it,
17
-------
if requested. Moreover, the battery manufacturers can
purchase pure lead and alloy it with antimony themselves, if
they wish.30
The value added by antimonial lead to pure lead is
minimal. One secondary smelter's antimonial lead is
typically 1% percent antimony in content and is not sold at
a premium over pure lead.31 Some of the smelter's antimonial
lead is up to six percent antimony in content, but still is
sold for only %-% C/lb more than pure lead.
Recently, with the growth of waterless or "maintenance-
free" batteries, calcium has been replacing antimony in the
lead alloy used in some of the plates. However, like
antimonial lead, calcium lead is basically a commodity. One
secondary lead smelter responding in the telephone survey
said that while calcium alloys generally add 1.5-2.5 C/lb to
pure lead, "these are common alloys today and can hardly be
classified as specialty products."32
Finally, the substitutability of primary and secondary
lead is underscored by the fact that primary and secondary
lead producers sometimes arrange to ship out of each others'
stockpiles (presumably to save shipping costs) ,33 (This
practice is more common among secondary lead producers,
though.)
3.1.1.3 Alternative Model. A highly price-elastic SF
or SD.P has significant implications for the model developed
in Figure 1. Figure 2 demonstrates the case of perfectly
elastic SF. Price is now constrained to the level at which
foreign producers are willing to supply any quantity.
Importantly, ST is perfectly elastic after it reaches the
price level. If instead of SF, SD.P had been assigned
perfect elasticity in Figure 2, ST would still have been
perfectly elastic, though in this case the perfect
elasticity would have begun on the price axis at the point
from which SD.P emanates. (Note: Because SF is perfectly
18
-------
FIGURE 2. CONCEPTUAL FRAMEWORKTOR THE U.S. LEAD MARKET,
ASSUMING PERFECTLY ELASTIC FOREIGN SUPPLY (SF)
Price
o-s
Q-
0-P ^F WD-S
Quantity
-------
elastic, it may not be immediately apparent, unlike in
Figure 1, how QF is determined in Figure 2. It is
determined as QT minus QD.S minus QD.P.)
As a rule, it takes only one perfectly elastic source
of supply to impart perfect elasticity to total supply. As
a corollary to this rule, it takes only one highly elastic
source of supply to impart high elasticity to total supply.
Therefore, if either SF or SD.P is highly elastic, ST is
highly elastic.
The different implications of the models in Figure 1
and Figure 2 for the impacts of the NESHAP on the U.S. lead
market will be seen in Section 3.1.3.
3.1.2 Price Elasticity of Demand
The demand for lead in the U.S. was shown in Figures 1
and 2 to be fairly steeply sloped, reflecting that it is
relatively price-inelastic. The price elasticity of demand
indicates the responsiveness of quantity demanded to a
change in price. It is measured as the percent change in
quantity demanded divided by the percent change in price.
Demand is relatively price-inelastic — the case for lead —
when the absolute value of the percent change in quantity
demanded is less than the absolute value of the percent
change in price. This is reflected in an elasticity
measurement between zero and -1.0.
In a study in the early 1980s, the price elasticity of
U.S. demand for lead in storage batteries, which currently
account for about 80 percent of U.S. secondary lead
consumption, was estimated at -0.23.34 This represents high
inelasticity. The U.S. demand for lead is highly inelastic
for three main reasons. First, there are few substitutes
for lead, in part because many nonessential uses of lead
have been weeded out by regulations in recent years. In
addition, there are presently no commercially successful
substitutes for lead-acid storage batteries. Alternative
20
-------
materials to lead have proven to be more costly and to have
less-favorable recycling economics. Secondly, demand for
lead's primary end use, storage batteries, is highly
inelastic, largely because storage batteries comprise a very
small portion of total cost in their applications (e.g., the
cost of an automotive battery is a small fraction of the
total cost of owning and operating an automobile). Thirdly,
many battery manufacturers need not be particularly
sensitive to the price of lead because they have sales
contracts allowing pass-alongs on their lead costs.35
3.1.3 Market Response
The effects on the U.S. lead market of the NESHAP's
control costs, which cause domestic secondary supply, SD.S/
to shift up and to the left, are shown in Figure 3 for the
model introduced in Figure l and in Figure 4 for the model
introduced in Figure 2. The shift in SD.S to SD.S' is the
same in Figure 3 and Figure 4. The vertical distance
between SD.S' and SD.S represents average unit control costs
for U.S. secondary lead smelters. It is assumed that unit
control costs are constant over all output levels. Hence
the vertical shift in SD.S to SD.S' is uniform, i.e., the
vertical distance between SD.S and SD.S' is constant over all
output levels.
In Figure 3, because SD.P and SF are unchanged, the
shift in ST (to ST') is less pronounced than the shift in SD.S
to SD.S'« (ST' equals the horizontal sum of SD.P, SD.S', and
SF.) Consequently, the increase in price, from P to P',
falls short of unit control costs. Due to the relatively
small price increase, the decrease in total quantity
supplied (and quantity demanded), from QT to QT', is muted.
The output mix changes significantly, however. While QD.S
decreases appreciably to QD.S'/ QD.P and QF gain in the market.
As long as SD.P and SF are not perfectly inelastic, the
contributions of QD.P and QF to total quantity supplied will
21
-------
FIGURE 3. THE EFFECTS ON THE U.S. LEAD MARKET OF
CONTROL COSTS ON SECONDARY LEAD PRODUCERS
Price
rv>
P'
P
Or Or
Quantity
-------
FIGURE 4. THE EFFECTS OlTTHE U.S. LEAD MARKET OF
CONTROL COSTS ON SECONDARY LEAD PRODUCERS,
ASSUMING PERFECTLY ELASTIC FOREIGN SUPPLY (SF)
Price
f\3
CO
'O-S
Quantity
-------
increase in response to the regulation, and QD.S will
decrease by more than the decrease in total quantity
supplied (QT minus QT')-
In Figure 4, as a result of the shift in SD.S to SD.S'f ST
shifts to ST', but this has no effect on ST beyond the point
at which it reaches the price level P. Beyond this point,
ST remains perfectly elastic. As a result, there is no
change in price (P' = P) , and control costs must be fully
absorbed by secondary lead producers. With no upward shift
in ST where it intersects demand, there is also no change in
total quantity supplied (and quantity demanded). However,
the redistributive effects are greater than in Figure 3.
Although QD.P does not change (because there is no change in
price) , QD.S falls by more and QF increases by more than in
Figure 3.
While in Figure 3 the price effect is muted> in Figure
4 there is no price effect at all. The weight of anecdotal
evidence, both from responses to the telephone survey and
from the trade literature, suggests that if any price
increase is achieved, it will be minimal. It is even very
possible that no price increase will take place. U.S. lead
producers have little control over price. Lead is an
internationally traded commodity whose price is determined
by global supply and demand, which are mostly external to
the U.S. In 1990, for example, U.S. supply of refined lead,
both primary and secondary, was about 22 percent of global
supply, while U.S. demand was about 22 percent, as well, of
global demand.36 In this global context, the lead market is
highly competitive. U.S. producers act as price-takers,
constrained to set prices that do not exceed the price of
lead in Europe (the LME price) plus the cost of moving lead
from Europe to the U.S. (the basis for the U.S. producer
premiums over the LME price).
24
-------
Neither the LME price nor the U.S. producer premiums,
the constituents of P (and P') in Figure 4, are likely to be
significantly impacted by the NESHAP. The LME price is not
likely to be significantly affected because it is influenced
by lead supplies from all over the world, of which U.S.
secondary lead is only one component (15.5% in 1990). The
U.S. producer premiums — which reflect withdrawal fees from
LME warehouses in Europe, ocean freight from Europe to the
U.S., the cost of insurance, tariffs, the cost of delivery
in the U.S., and a small "grade premium" to ensure that LME
lead is U.S. battery-grade — are not likely to budge because
none of their determinants will be affected by the
regulation.37 The premiums are not arbitrary; rather, they
are disciplined by arbitrage in a competitive market. If
U.S. producers were to attempt to increase their premiums,
lead traders could make a profit by arranging to have lead
delivered to U.S. buyers from LME stocks in Europe. Mexican
lead — which has been selling in the U.S. for only 2-3 C/lb
above the LME price — might also be available.38 It is
telling that if foreign lead can be delivered in the U.S.
more cheaply than domestic lead, U.S. producers are willing
to purchase it and sell it at a profit, even if this
cannibalizes their own production.39
Finally, Asarco, a primary lead producer, is considered
to be the U.S. price leader (i.e., it takes the lead in
establishing a premium over the LME price).40 This suggests
further that U.S. secondary lead producers would not be able
to increase their premiums, as primary lead is not subject
to the NESHAP.
It can be concluded that SF and/or SD.P are sufficiently
elastic that if U.S. secondary lead producers can achieve a
price increase, it will be negligible. Therefore, Figure 4,
which posits no price increase, will be the model for
assessing industry-wide impacts. At a minimum, this will
25
-------
result in conservative (worst-case) output and employment
impacts. However, the impact calculations are liable to
also be more accurate than calculations based on some kind
of price increase.
3.1.4 Methodology
The most important consequence of the model in Figure 4
is that there is no price increase in the U.S. lead market.
So, no "market price increase" is estimated in the economic
impact analysis.* A second important consequence is that
there is no change in total supply in the U.S. lead market.
This is because ST does not shift up at the point where it
intersects demand (D). Because, as a result, no change in
quantity demanded — dictated by the price elasticity of
demand —is traced along D, the price elasticity of demand
has no influence on QT or on the changes in any of the
components of QT, such as QD.S.
Instead, the change in U.S. secondary lead output, from
QD-S to QD-S' in Figure 4, is a function of the magnitude of
control costs and the price elasticity of supply. The
magnitude of control costs is reflected in the magnitude of
the shift in SD.S to SD.S', while the price elasticity of
supply is reflected in the slope of SD.S (and SD.S').
Accordingly, using regression analysis, the price elasticity
of supply is estimated in Section 4.0. Then, in Section
5.1, after calculating the shift in SD.S to SD.S' , the
industry-wide change in output is estimated. Impacts on
industry-wide employment and revenue are in turn estimated
from the change in output.
* The "market price increase," defined as the average
industry-wide price increase, is typically estimated in
analyses of the economic impacts of regulations.
26
-------
3.2 PER-FACILITY IMPACTS
Without a price increase, control costs cannot be
passed along to customers. Furthermore, there is limited
downward flexibility in the cost of lead scrap. This makes
it uncertain that control costs can be passed back to scrap
suppliers. Therefore, it is assumed that control costs will
have to be absorbed. A calculation of the resultant percent
reduction in net income is not meaningful because the U.S.
secondary lead industry appears to be currently (December
1993) only just about breaking even. (Earlier in 1993, when
the price of lead was lower, the industry was probably
losing money.) Instead, in Section 5.2.2.2, for all 23
secondary lead facilities in the U.S., the per-pound impact
of total annualized control costs is calculated. This
represents the amount per pound by which net income will
decline as a result of the NESHAP. The possibility that any
facility may not be able to sustain the impact and therefore
is at risk of closure is evaluated. Although discussed, the
impacts calculated in Section 5.2.2.2 are not disclosed so
that the facility identities are not revealed. This is
necessary to protect confidentiality.
In Section 5.2.3, for all companies that own one or
more of the 23 secondary lead facilities in the U.S., the
availability of capital to finance capital control costs is
assessed by calculating the ratio of capital control costs
to baseline total assets and by comparing post-NESHAP total
liabilities — assuming debt is issued to finance capital
control costs — to baseline total assets. The possibility
that any company may not be able to obtain capital and
therefore one or more of its facilities is at risk of
closure is evaluated. Again, the calculations are not
disclosed to protect confidentiality.
27
-------
4.0 ESTIMATION OF THE PRICE ELASTICITY OF SUPPLY
The price elasticity of supply (e) indicates the
responsiveness along a supply curve of quantity supplied to
a change in price. It is measured as the percent change in
quantity supplied divided by the percent change in price.
For a normal upward-sloped supply curve, there is a positive
relationship between the change in quantity supplied and the
change in price, and e :> 0. If e < l, supply is considered
to be relatively price-inelastic. Conversely, supply is
relatively prdce-elastic if e > 1.
As was seen in Figure 4, the impact of the NESHAP on
U.S. secondary lead output (QD.S) is determined by the change
in the point at which the shifting industry supply curve
(SD_S) intersects the unchanged price level. The extent to
which QD.S declines will depend both on the magnitude of
control costs and on e. The magnitude of control costs is
reflected in the magnitude of the shift in SD.S to SD.S' ,
while e is reflected in the slope of SD.S (and SD.S'). The
greater are control costs (i.e., the greater is the shift in
SD.S) , and the more SD.S is price-elastic (i.e., the lower the
slope of SD.S) , the greater will be the decline in QD.S.
The price elasticity of supply is therefore a necessary
input for determining the impact of the NESHAP on QD.S (and,
in turn, industry-wide employment and revenue). There is no
known estimate, such as in the economics literature, of e
for the U.S. secondary lead industry. However, e can be
estimated by regressing the natural log of price, as an
independent variable, against the natural log of quantity
supplied (output), as the dependent variable. The price
elasticity of supply is represented in this framework by the
coefficient obtained for the natural log of price.
There are of course factors other than price that
influence supply. For example, technology and the costs of
inputs (e.g., labor, raw materials, energy, capital)
influence the willingness and/or ability of firms to
28
-------
produce. The supply curve is normally presented as the
relationship between price and quantity supplied. A change
in price therefore results in a movement along the supply
curve. Changes in other factors such as technology and the
costs of inputs, on the other hand, result in a shift in the
supply curve.
In determining the influence of price on quantity
supplied, and in turn estimating e, it is important to
control for the other factors that significantly influence
quantity supplied. Otherwise the influence of price and the
estimate for e may be distorted.
The regression model used to estimate e is specified in
Table 4. In addition to the price of lead, four other
independent variables — the cost of lead scrap, the producer
price index (PPI) for secondary nonferrous metals, a dummy
variable indicating whether or not state battery recycling
laws were in place, and a time trend — were included for
their influence on U.S. secondary lead production, the
dependent variable. The variables were entered in the
regression as a time series from 1970 to 1991. The values
of the variables are given in Appendix A.
Normally, price is determined by the interaction of
supply and demand and therefore a two-stage regression is
required to estimate the supply function. In the case of
secondary lead, however, U.S. supply has minimal influence
on price. This is because the market for lead is global,
and global supply and demand dictate price. For example,
even though total U.S. output of .refined lead fell by about
36,000 metric tons in 1992 (primary lead output was off
about 41,000 metric tons from 1991 while secondary lead
output increased by about 5,000 metric tons) in the face of
relatively stagnant U.S. demand, the price of lead plummeted
largely due to heavy shipments from the former East Bloc and
weak worldwide demand, especially in Europe.41 This means
29
-------
TABLE 4. REGRESSION MODEL
Variable*
Variable description
Source
Expected
sign of the
coefficient
("+" or "-")
Dependent variable:
Secondary lead
production
Independent variables:
Price of lead
u>
o
Natural log of U.S.
secondary lead
production, in metric
tons.
Natural log of the real
spot price of lead, in
C/lb, on the London Metal
Exchange (LME).
Cost of lead scrap Natural log of the real
dealer's buying price, in
C/lb, for heavy soft
scrap lead.b
PPI for secondary
nonferrous metals
Natural log of the real
producer price index
(PPI) for secondary
nonferrous metals.
U.S. Department
of the Interior,
Bureau of Mines
U.S. Department
of the Interior,
Bureau of Mines
"Metal
Statistics"
(annual),
Diversified
Publishing
Group, New York,
NY
U.S. Department
of Labor, Bureau
of Labor
Statistics
N.A.
-------
TABLE 4. (CONTINUED)
Variable"
Variable description
Source
Expected
sign of the
coefficient
(«+» or «_«)
State battery
recycling laws
Time trend
Dummy variable indicating
whether or not state
battery recycling laws
are in place (no = 0, yes
= 1).
Series of increasing
integers representing the
passage of time (1970 =
1, 1971 =2, ... 1991 =
22) .
Battery Council
International,
Chicago, IL
'See Appendix A for values.
bNew York prices 1970-1986, Chicago prices 1987-1991. Chicago prices adjusted to New
York basis based on the relationship between the two bases in 1986, the year in
which the two series overlap.
N.A. Not applicable.
Note: All "real" variables are adjusted to 1982 dollars using the producer price
index for finished goods. Source: U.S. Department of Commerce, Bureau of
the Census, "Statistical Abstract of the United States 1992."
-------
that the price of lead in the U.S. is determined
exogenously, not by the interaction of domestic supply and
demand, and the U.S. secondary lead industry (not just the
individual producers) can be modeled as facing perfectly
elastic demand. As a result, it is possible to estimate
U.S. secondary lead supply, SD,S, with a one-stage
regression.
4.1 INDEPENDENT VARIABLES
The price of lead is represented in the model by the
spot price on the LME. This is more appropriate than the
U.S. quoted price (e.g., "North American Producer Price")
because the actual transaction price in the U.S. is closely
linked to the LME price, differing only by the relatively
constant cost of transporting lead from Europe to the U.S.
The U.S. quoted price, on the other hand, is a list price
off of which varying discounts are allowed. The
relationship between the price of lead and U.S. secondary
lead production (i.e., the sign of the coefficient for the
price of lead) was, of course, expected to be positive, in
accordance with an upward-sloped supply curve.
The cost of lead scrap, on the other hand, was expected
to have a negative effect on U.S. secondary lead production.
The higher the cost of a factor input, the less the
willingness to produce, ceteris paribus. This is manifested
in a shift in the supply curve to the left when the cost of
a factor input increases, and a shift to the right when the
cost of a factor input decreases. However, it was
recognized that this relationship is liable to be obscured
in the regression model. While the cost of lead scrap can
have an independent negative influence on secondary lead
production, secondary lead production can have an
independent positive influence on the cost of lead scrap.
For example, if production increases independently, say
because demand increases or the cost of energy decreases,
the price of lead scrap, the supply of which is limited, is
32
-------
likely to be bid up. This applies to the market for lead
scrap because no industry, other than perhaps the disposal
industry, competes with the secondary lead industry for lead
scrap. As a result, the secondary lead industry has
significant buying power over lead scrap. This would not
apply to an input such as unskilled labor, for which all
industries compete.
The PPI for secondary nonferrous metals was included to
capture the costs of factor inputs other than scrap.
Because, in the long run at least, cost changes tend to be
reflected in price changes, it was reasoned that the PPI is
correlated with the generic costs, such as the costs of
labor and energy, incurred by all secondary nonferrous metal
producers, including secondary lead producers. As a
representative of the costs of factor inputs, the PPI was
expected to be negatively correlated with secondary lead
production. Separate independent variables representing the
costs of labor, natural gas, and electricity were tried in
the regression equation. Although collectively they had
more explanatory power (higher R-squared) than the PPI, they
were not used because in most of the runs, the coefficients
for two of the three variables had the unexpected (and
presumably wrong) sign.
Since 1989, states have been active in passing laws
promoting lead-acid battery recycling. These laws typically
include "take-back provisions" that require retailers to
accept spent batteries from consumers, and battery
manufacturers to accept spent batteries from retailers, when
new batteries are purchased.42 This has no doubt positively
influenced the battery recycling rate and therefore
secondary lead production. Nine states were first to enact
battery recycling laws in 1989; by the end of 1992, 37 had
done so.43 In order to account for this development, a dummy
variable indicating whether or not battery recycling laws
33
-------
were in place was used in the regression equation. The
variable's value is zero for all years prior to 1989, and
one for 1989-1991. The expected correlation with secondary
lead production was positive.
The final independent variable is a time trend. It
runs in increments of one, from one to 22, for the period
1970-1991. This variable was included to capture in
particular the cumulative effects of two forces: regulatory
(e.g., environmental, safety and health) compliance costs
and technological improvements. While regulatory compliance
costs shift the supply curve to the left and therefore tend
to have a negative effect on output, technological
improvements shift the supply curve to the right and
therefore tend to have a positive effect on output. Since
it is not known which of these forces has been stronger in
the past two decades, it was not known what sign to expect
for the time trend variable.
4.2 REGRESSION RESULTS
The key results of the regression analysis are
summarized in Table 5. In the first run, supply is
estimated to be very price-inelastic, as e is only 0.143.
However, this estimate cannot be taken with much confidence
because its T-statistic is only 0.571. There is also a
potential autocorrelation problem as the Durbin-Watson
statistic is only 0.662. (Autocorrelation is indicated by a
Durbin-Watson statistic less than 2.) Note, though, that
all independent variables have the expected sign.
The second run yields improvements. The price
elasticity of supply is now 0.548 and is significant at the
99 percent confidence level (T-statistic = 2.941). The
predictive power of the model improves as the adjusted R-
squared increases to 0.722. The Durbin-Watson statistic
also improves, though 0.922 may still indicate an
autocorrelation problem. The time trend.has turned out to
34
-------
TABLE 5. RESULTS OF THE REGRESSION ANALYSIS
Independent variable coefficients
(T-statistics in parentheses)
Run i
Run 2
Run 3
Price of
lead
0.143
(0.571)
0.548
(2.941)
0.317
(1.948)
Cost Of
lead
scrap
0. 134
(0.800)
-0.078
(-0.626)
0.061
(0.566)
PPI for
secondary
nonferrous
metals
-0.388
(-1.029)
-0.294
(-0.968)
-0.237
(-0.981)
State
battery
recycling
laws
0.387
(4.866)
0.206
(3.336)
Time
trend
0.025
(6.781)
0.018
(5.100)
Durbin-
• Watson
statistic
0.662
0.922
0.964
Adjusted
R-squared
0.569
0.722
0.826
F-
statistic
7.941
14.626
20.901
Mean square
error
0.013
. 0.008
0.005
-------
be significant (T-statistic = 6.781) and the coefficient is
positive. This suggests that technological improvements
(including the replacement of old plants with new ones) have
compensated for regulatory compliance costs over the past
two decades. This was perhaps suggested when a respondent
in the telephone survey said that recent process changes,
including the installation of environmental controls, have
improved its efficiency.44 Note that the sign for the cost
of lead scrap has changed. This probably reflects, as
discussed earlier, the reverse positive effect of secondary
lead production on the cost of lead scrap.
The third run includes the dummy variable for state
battery recycling laws and the time trend together, and both
are significant. The price elasticity of supply, 0.317, is
still significant, now at the 93 percent confidence level.
The explanatory power of the model is good, as the adjusted
R-squared is 0.826.
The price elasticities of supply obtained in the second
and third runs are both considered to be reasonable. Their
indication of relative inelasticity is consistent with the
observation that industry supply has been less variable than
price over the past two decades (though at the facility
level supply has been volatile as, for example, many small
facilities have shut down to be replaced by larger
facilities) . Both of these estimates for e will be used in
Section 5.1 to calculate industry-wide impacts. This will
not only result in a range of industry-wide impacts, but
will also allow for the sensitivity of the impacts to a
change in e to be tested. The price elasticity of supply of
0.548 will yield larger impacts than the price elasticity of
supply of 0.317.
36
-------
5.0 ECONOMIC IMPACTS
5.1 INDUSTRY-WIDE
As discussed in Section 3.1.4, industry-wide impacts
are calculated assuming that the NESHAP does not have an
effect on the price of lead. Figure 5 shows that, as a
result, the change in U.S. secondary lead output, from Q, to
Q2, is equal to the horizontal shift in the industry supply
curve, SD.S, to SD.S', represented at the unchanged price
level P, = P2 as AB. The horizontal shift in SD.S in turn
depends on 1) the magnitude of control costs, which is
reflected in the vertical shift in S^, and 2) the price
elasticity of supply (e), which is related to the slope of
SD-S (and SD.S'). (Note: SD.S and SQ.S' are shown for
simplicity to be linear. In reality, if they are constant-
elasticity functions and e < 1, they will be concave-
upward .)
AB~, or the change in Q, can be represented by the
horizontal displacement between A and C, the point directly
below B on SD.S. Recall from Section 4.0 that e is measured
for a supply curve as the percent change in quantity
supplied divided by the percent change in price:
e =
This can of course be rearranged as:
%AQ = 6 X %AP
For segment AC of SD.S, this can be expressed as
37
-------
FIGURE 5. THE EFFECT OF CONTROL COSTS ON U.S. SECONDARY LEAD
PRODUCTION IF PRICE DOES NOT CHANGE (P, = P2)
Price
p,=
P*
GO
CO
0-S
K
Q
0.
Quantity
-------
(Q2-Q,)/Q, = e x (P'-PJ/P, *
All but Q2 in this equation are already known. Two
alternative estimates of e, 0.548 and 0.317, were offered in
Section 4.0. Q,, as represented by U.S. secondary lead
output in the most recent year, 1992, is 888,500 metric
tons. P, is established as 25 C/lb, the approximate current
(November 1993) net price of lead in the U.S.45 Finally, P*
equals P, minus BC, where, assuming that unit control costs
are constant over all output levels (i.e., the vertical
shift in SD.S is uniform) , BC is equal to the vertical
distance between SD.S and SD.S' above the baseline
equilibrium, A, which is the average industry-wide unit
(per-pound) control cost. The average industry-wide unit
control cost is calculated as industry-wide total annualized
control costs — $1,956,000 under Control Option 1,
$3,044,000 under Control Option 2, $2,533,500 under Control
Option 3, $4,019,300 under Control Option 4, and $5,406,200
under Control Option 5 (see Table 2) — divided by 1992
production, 1.959 billion pounds (888,500 metric tons x
2,204.62 Ibs/metric ton). This comes to 0.10 C/lb under
Control Option 1, 0.16 C/lb under Control Option 2, 0.13
C/lb under Control Option 3, 0.21 C/lb under Control Option
4, and 0.28 C/lb under Control Option 5. P* is therefore
25.00 C/lb minus 0.10 C/lb = 24.90 C/lb under Control Option
1, 25.00 C/lb minus 0.16 C/lb = 24.84 C/lb under Control
Option 2, 25.00 C/lb minus 0.13 C/lb = 24.87 C/lb under
Control Option 3, 25.00 C/lb minus 0.21 C/lb = 24.79 C/lb
under Control Option 4, and 25.00 C/lb minus 0.28 C/lb =
24.72 C/lb under Control Option 5.
* P* will never actually be realized. It is only an
analytical construct that permits solving for %AQ along
39
-------
With these inputs it is possible to solve for Q2 and,
in turn, the change in annual industry-wide output (Q, minus
Q2) . The results are shown in Table 6. Under the least
stringent Control Option, 1, industry-wide output falls by
an amount ranging from 1,127 to 1,948 metric tons, depending
on e. Under the most stringent Control Option, 5,
industry-wide output falls by an amount ranging from 3,155
to 5,453 metric tons, again depending on e. Even under
Control Option 5, the impacts are not considered to be
significant. At the most (e = 0.548), output would decline
by only 0.61 percent. There is sufficient unused capacity
in the U.S. secondary lead industry (capacity utilization in
early 1993 was about 89%) that the decline in output should
not have an effect on the battery recycling rate.46 Ensuring
that there is adequate battery recycling capacity is an
important component of EPA's overall "lead strategy," "both
to prevent batteries from being discarded in the
environment, and to reduce the need to mine and smelt new
lead."47
The impacts in Table 6 represent permanent changes in
U.S. secondary lead output. However, as Figure 3 made
clear, the total quantity of lead supplied to the U.S.
market (QT) will not decline by nearly as much (and will not
decline at all if, as in Figure 4, foreign lead supply
and/or domestic primary lead supply are perfectly price-
elastic) . The quantity supplied of foreign lead (QF) and
domestic primary lead (QD.P) will increase to largely offset
the decrease in U.S. secondary lead output (QD.S) . The
extent to which foreign lead and domestic primary lead
increase their shares of the U.S. lead market will depend on
their price elasticities of supply. In the short run,
foreign lead is likely to gain more market share — the
record-high stocks of foreign lead (e.g., in LME warehouses
in Europe) suggest that its supply is highly elastic. In
the long run, foreign lead may also be the main beneficiary
40
-------
TABLE 6. ESTIMATED ANNUAL INDUSTRY-WIDE OUTPUT IMPACTS
Price
elasticity
supply (e) = 0.
Control
Option 1
Control
Option 2
Control
Option 3
Control
Option 4
Control
Option 5
Baseline
output
(Q,>, in
metric
tons
888,500
888,500
888,500
888, 500
888,500
Post-
regulation
output (Q2) ,
in metric
tons
886,552
885,384
885,968
884,410
883,047
Change in
output,
in metric
tons
-1,948
-3,116
-2,532
-4,090
-5,453
of
548
Percent
change in
output
-0.22%
-0.35%
-0.28%
-0.46%
-0.61%
Price
elasticity
supply (c) = 0.
Post-
regulation
output (Q2) ,
in metric
tons
887,373
886,697
887,035
886,134
885,345
Change in
output,
in metric
tons
-1,127
-1,803
-1,465
-2,366
-3,155
of
317
Percent
change in
output
-0.13%
-0.20%
-0.16%
-0.27%
-0.36%
-------
if 1) foreign producers continue to become more aggressive
in the U.S., 2) the general trend toward lower trade
barriers (e.g., NAFTA) continues, and 3) the domestic
primary lead industry is itself more strictly regulated.
In 1991, total employment at U.S. secondary lead
smelters and refineries was estimated to be 1,700.48
Assuming that employment is proportionate to output (i.e., a
fixed labor-output ratio) , the industry-wide employment
impact under the least stringent Control Option, 1, ranges
from -2 to -4, depending on e. Under the most stringent
Control Option, 5, the range is -6 to -10, again depending
on €.. These impacts are less than one percent and are
considered minimal.
Revenue equals output times price. With no change in
price, the percent change in revenue is equal to the percent
change in output. This is shown in Table 6. At the most
(Control Option 5, e - 0.548), industry-wide revenue
declines only by 0.61 percent.
5.2 PER-FACILITY
5.2.1 Baseline Risk
Irrespective of the NESHAP, there is tremendous risk in
the U.S. secondary lead industry. This has been reflected
in a staggering rate of attrition over the past two decades
or so. Compared to the 16 secondary lead smelters currently
in operation, there were 103 in 1980 and 156 in 1975.49
Nevertheless, output increased from 675,578 metric tons in
1980 to 888,500 metric tons in 1992. Clearly, the average
facility size has increased substantially, and small
smelters have been the main victims of the industry shake-
up.
The closure of so many small secondary lead smelters in
the 1980s is attributable to two main forces: competition
for a limited supply of lead scrap and the costs of
pollution controls.50 Previously, small, decentralized
42
-------
smelters enjoyed transportation advantages from being near
customers and sources of inputs. However, these advantages
have come to be outweighed by the costs of pollution
controls, which have been proportionately higher — i.e.,
higher per unit of output — for small facilities than for
large facilities (this is an example of an "economy of
scale").51
The industry is apparently not done restructuring.
Thus far in 1993, five relatively small (two "small" and
three "medium") smelters have shut down, and two
nonintegrated, regional smelters have cut back production.
William Woodbury, the former lead specialist for the U.S.
Bureau of Mines, says "you're going to see the big companies
getting bigger. Market ups and downs are hard on smaller
companies, but the battery manufacturers want to deal with
the smelters that are more or less impervious to the
market."52 The smaller smelters are disadvantaged because,
lacking scale economies, they are likely to be less
efficient; because they generally have less clout with the
battery manufacturers in procuring spent batteries; and
because they generally have less access to capital.53
Smelters that are not well positioned in a battery
recycling loop, either because they are not integrated into
battery manufacture or because they do not have good
connections with the battery manufacturers, are also
generally disadvantaged. Lead scrap supplies for such
smelters can be uncertain.
5.2.2 Cost Absorption
Without a price increase, control costs cannot be
passed along to customers. Furthermore, although there may
be some downward potential for the cost of lead scrap, it is
limited.54 This makes it uncertain that control costs can be
passed back to scrap suppliers.
43
-------
Like the market for lead, the market for lead scrap is
not closed — there are export outlets. In 1992, the U.S.
exported about 60,000 metric tons of lead scrap.55 Already
there is significant incentive to export lead scrap because
potential Superfund liability in the U.S. can be avoided and
because many countries have less-stringent battery recycling
laws than the U.S. Because of the reduced risk, one scrap
broker describes exporting lead scrap to Canada as a
"marketing tool" for procuring scrap.56 The availability of
export outlets limits downward flexibility in the cost of
lead scrap. In April 1993, a metals trader said that spent
battery prices (about 6 C/lb on average) were "just high
enough to forestall any significant return to the export
market. "57
The U.S. lead scrap market is not disciplined just by
foreign trade — it has an internal disciplining mechanism as
well. A certain minimum amount must be paid for lead scrap
in order to cover the costs of the successive stages in its
collection and shipment. The lower the price, the less the
incentive of agents in the recycling chain — consumers,
retail outlets, scrap dealers/battery haulers, et. al. — to
move lead scrap along for recycling (though battery
manufacturers, who as explained in Section 2.1 are more and
more controlling the flow of lead scrap for recycling, may
not be dissuaded from moving lead scrap along for recycling
because, as consumers of lead, they have a vested interest
in seeing that lead scrap is recycled). For example,
retailers will have less incentive to offer payment or a
discount to consumers for their used batteries or to even
accept them at all. Scrap dealers may hold on to
inventories, hoping for price to rebound. Also, with the
risk of Superfund liability at battery-breaking sites, scrap
dealers may refuse to handle lead scrap if it is not
sufficiently remunerative. In May 1993, one secondary lead
44
-------
smelter responding in the telephone survey said that its
cost of lead scrap — from 6 to 8 C/lb for delivered spent
batteries, depending on where they are collected — is
probably close to the threshold at which middlemen in the
recycling chain have sufficient monetary incentive to
participate.58
Seen another way, relatively low lead scrap prices can
prompt secondary lead smelters to tighten their "catchment
areas," or the areas in which they buy lead scrap (because
it is not worth incurring the high transportation costs of
going afar for spent batteries).59 This can leave pockets of
uncollected batteries.
In summary, if the price of lead scrap is too low, the
flow of lead scrap to secondary lead smelters, i.e., the
whole lead recycling chain, can be disrupted.
Without being able to pass control costs along to
customers, and with an uncertain prospect that they can be
passed back to scrap suppliers, it is assumed that control
costs will have to be absorbed. This will reduce net income
in an industry that appears to be currently (December 1993)
only just about breaking even.
5.2.2.1 Baseline Profitability. That the U.S.
secondary lead industry is apparently just about breaking
even can be appreciated by comparing costs and price. It
has been estimated that it costs secondary lead producers
between 23 and 27 cents to make a pound of lead (it is
believed that this accounts only for operating costs, not
for capital costs, which are represented in the income
statement by depreciation)-60 This comprises 12 C/lb for
lead scrap (based on 6 C/lb for spent batteries, which are
about half lead in content) and 11-15 C/lb for processing
(this ignores the small credits that some smelters get from
recovering plastic and/or sulfuric acid). In mid-1993, lead
scrap supplies began to ease somewhat in response to several
45
-------
factors, including production cutbacks in the U.S. and
Canada, hot weather (which influences the rate at which
automobile batteries wear out), and the release of scrap
supplies by dealers who had been holding out in hopes of an
increase in the price of lead.61 As a result, the cost of
spent batteries in the eastern U.S. has fallen to 5-6 C/lb
on a picked-up basis.62 (In the western U.S., cost has
remained fairly steady at 5%-6 C/lb on a delivered basis.)
Even in the case of 5 C/lb picked-up, and assuming % C/lb
for delivery, the cost of lead scrap is 11 C/lb (5% C/lb x
2), which with processing costs yields a total operating
cost of 22-26 C/lb.
Meanwhile, in November 1993, lead was selling in the
U.S. for about 25 C/lb.63 This price — which reflects an LME
spot price of about 18 C/lb ($395-400 per metric ton) and a
premium over the LME spot price of 7 C/lb, as announced by
Asarco — falls within the range of total operating cost,
suggesting that the U.S. secondary lead industry is just
about breaking even.
Although as a whole the U.S. secondary lead industry is
just about breaking even, some facilities are probably
making a profit. Just a few months ago, in the spring and
summer of 1993, the price of lead in the U.S. was only about
22 or 23 C/lb.64 At that time, the industry was probably
losing money. Yet, one respondent in the telephone survey
said that a few "big, modern" facilities might be making a
small profit.65 Additionally, a positive return can be had
from tolling, which is becoming more customary as the
battery producers increasingly control the lead scrap
market. Tolling offers a positive return, albeit perhaps a
small one, because generally a mark-up over processing costs
is charged. Facilities that do a lot of tolling may be
making a small profit.
46
-------
Although price has rebounded by two or three C/lb since
the summer, it is still a far cry from the price a little
over a year ago. Compared to about 18 C/lb in November
1993, the LME spot price of lead was 30 C/lb in September
1992. By the start of 1993, it had plummeted to 18.9 C/lb.
Later in 1993 it reached its lowest real (i.e., adjusted for
inflation) level in history. The main causes of the drop in
price were increased exports from the former East Bloc,
China, and North Korea, and slack demand worldwide,
especially in Europe. Net lead exports to the West from the
former East Bloc (particularly Eastern Europe, Russia, and
Kazakhstan), China, and North Korea were 155,000-160,000
metric tons in 1992, up from only 60,000-65,000 metric tons
in 1991.w This increase more or less accounted for the
world production surplus in 1992. The International Lead
and Zinc Study Group projects that exports to the West will
again be "high" in 1993.67 Meanwhile, in the first six
months of 1993, European lead consumption was down 5.5
percent from the same period in 1992.68
Profit margins have been squeezed over the past year
because the cost of lead scrap has not fallen in step with
the price of lead. There has been a relative scrap
shortage, in part due to the absence of extreme weather
fluctuations (which can cause automotive batteries to
expire). Excess capacity in the'U.S. secondary lead
industry, at least vis-a-vis the limited supply of lead
scrap, has also helped to prevent scrap costs from falling.69
While the U.S. secondary lead industry as a whole may
be currently breaking even, some facilities may be losing
money. Moreover, for much of 1993, before the recent upturn
in price, the majority of facilities in the industry may
have been losing money. Firms cannot stay unprofitable
indefinitely. In the long run, an adequate return on
investment must be generated in order to justify staying in
47
-------
business. In theory, the rate of return cannot fall short
of the rate of return on the best alternative investment.
Otherwise, operations will be terminated and capital will be
redeployed to the alternative investment.
In the short run, however, firms may continue to
operate despite negative returns. In practice, this may be
because an adequate long-run return is still expected (or
hoped for). In theory, rather than close, it makes sense
(i.e., it is profit-maximizing) to stay in business as long
as variable costs do not exceed revenue. Fixed costs do not
factor into this equation because they are, by definition,
incurred whether or not operations are maintained. Relative
to the option of shutting down temporarily (temporary
shutdowns are common in the U.S. secondary lead industry),
it makes sense to stay in operation as long as variable
costs do not exceed revenue by more than the costs to shut
down and later restart.
5.2.2.2 Calculations. Per-pound impacts of full-cost
absorption were calculated by dividing total annualized
control costs by annual production, in pounds. Facility
production data are not known; therefore, production was
approximated by multiplying production capacity by the
average industry capacity utilization rate. This assumes
uniform capacity utilization from facility to facility. To
the extent that capacity- utilization is below the industry
average, the calculated impacts are understated; to the
extent that capacity utilization is above the industry
average, the calculated impacts are overstated.. The
production estimates for Sanders Lead Company and for the
Schuylkill Metals Corp. facility in Baton Rouge, La. reflect
the cutbacks earlier this year (see Table 1) . If production
were to be restored to the previous level, the impacts would
be lower than calculated. For the seven facilities that are
shut down (see Table 1) , production is assumed to be at full
48
-------
capacity. The impacts for these facilities apply only if
the facilities are restarted.
Impact calculations for major sources are summarized in
Table 7. A threshold of 0.25 C/lb, about one percent of
baseline operating costs, is used to indicate a "significant
impact." This is conservative, considering that a five
percent threshold — approximately 1.25 C/lb in this case —
is typically used in regulatory impact studies. In
addition, 0.25 C/lb represents only about two percent of the
much more significant 12 C/lb drop in the LME price since
September 1992, which, because there has been little
compensating decrease in the cost of lead scrap, has hit the
bottom line' like a cost increase. However, a conservative
threshold is preferred, considering that the current
economic environment is disinflationary and that, with the
numerous facility closures over the past couple decades, the
U.S. secondary lead industry has demonstrated its
sensitivity to pressures on profitability.
The number of major sources that are significantly
impacted ranges from four under Control Option 1 to seven
under Control Option 5. Some of these facilities are
inactive, i.e., shut down. For these facilities (and for
the inactive facilities that are not significantly impacted,
as well) it can be said that the incremental costs of the
NESHAP will give them an additional incentive not to reopen.
Two active major sources are significantly impacted
under all five control options. A third active major source
is significantly impacted under Control Option 2, while a
fourth active major source is significantly impacted under
Control Option 5. (Though significant, the impacts are all
less than 1 C/lb.) Under current market conditions in the
U.S., with price having rebounded by 2 or 3 C/lb since the
summer, none of these facilities are likely to close as a
49
-------
TABLE 7. NUMBER OF MAJOR SOURCES WITH
A PER-POUND IMPACT OF TOTAL ANNUALIZED
CONTROL COSTS EXCEEDING 0.25 CENTS
Active Inactive Total
Control Option 12 2 4
Control Option 2 3 2 5
Control Option 32 3 5
Control Option 42 4 6
Control Option 53 4 7
50
-------
consequence. Demand has picked up and lead supplies are
said to be "tight."70 If price turns back down, however, a
closure or two is possible. The most likely candidates are
the two facilities that are significantly impacted under all
five control options. This is because both are very
possibly "marginal" — i.e., among the industry's highest-
cost producers — in the baseline. The likelihood of either
facility closing of course increases as the control options
become more stringent. The other two facilities are less
likely to close because they are less likely to be marginal
in the baseline and because their impacts are only
marginally above 0.25 C/lb.
Several area sources are also significantly impacted:
one under Control Option 1, two under Control Option 2, one
under Control Option 3, three under Control Option 4, and
four under Control Option 5. All but one of these
facilities are shut down. These facilities will be less
likely to reopen as a result of the NESHAP. The other
facility is significantly impacted under Control Options 2,
4, and 5 (though the impacts are less than 1 C/lb). It
could conceivably have to close if the market turns back
down. One argument against closure, though, is that the
facility will reportedly be overhauled and expanded.71
Perhaps the NESHAP would not derail such major plans. -
5.2.3 Capital Availability
The ability to finance capital control costs was
assessed by comparing the capital control costs to baseline
capital structure (i.e., liabilities and equity in relation
to assets). For the publicly owned companies, GNB and Doe
Run, information on capital structure was available from
public financial statements (e.g., annual reports). For
most of the private companies, information on capital
structure was available from Dun & Bradstreet Information
Services (Murray Hill, NJ). The analysis was conducted at
51
-------
the company level, not the facility level. This is because
capital activities such as borrowing are conducted by the
parent company, not an operating facility. Similarly, the
capacity to borrow is a function of the parent company's
finances, not an individual facility's. For companies that
operate more than one impacted facility, the per-facility
capital control costs were aggregated.
The analysis was triggered by calculating the ratio of
capital control costs to baseline total assets. This ratio
measures the percent impact on company-wide total resource
requirements. Alternatively, it measures the percentage
point increase in the share of baseline assets that total
liabilities would comprise if debt is issued to finance the
capital investment. (Normally for an investment in
pollution controls, if external financing is needed, it is
assumed that debt, not equity, is issued because the
investment does not add to the firm's productive capacity.)
For example, if total liabilities to assets is 50 percent in
the baseline, a five percent ratio would indicate that total
liabilities to baseline assets increases to 55 percent.
A ratio exceeding five percent was taken to indicate a
significant impact on a company's balance sheet. If the
ratio exceeded five percent, it was in turn judged that
capital may be difficult to obtain if post-NESHAP total
liabilities — assuming capital control costs are financed
with debt — would exceed two-thirds of baseline total
assets. Generally, the higher the ratio of liabilities to
assets (i.e., the greater the "financial leverage"), the
more difficult it is to obtain financing because the risk of
default is greater.
Under Control Options 4 and 5, two facilities may have
difficulty obtaining capital. While one is a major source
and one is an area source, both are shut down. Therefore,
52
-------
under Control Options 4 and 5, difficulty in obtaining
capital could prevent the facilities from reopening.
The case for two other facilities is ambiguous because
no up-to-date information on capital structure is available.
The ratio of capital control costs to baseline total assets
exceeds five percent under Control Option 5 for the first
facility and under Control Options 4 and 5 for the second
facility. Whether it may in turn be difficult under these
control options to obtain capital cannot be said without
knowledge of baseline capital structure. The first facility
is a major source that is shut down. The second facility is
a major source that is in operation. This facility is one
of the two facilities that was found in Section 5.2.2.2 to
be significantly impacted by total annualized control costs
(> 0.25 0/lb) under all five control options.
5.3 SMALL BUSINESSES
In accordance with the Regulatory Flexibility Act of
1980, it is necessary to perform a Regulatory Flexibility
Analysis (RFA) if a proposed rule (in this case the NESHAP)
will have "a significant economic impact on a substantial
number of small entities." The EPA's "Revised Guidelines
for Implementing the Regulatory Flexibility Act" (1992)
interpret this broadly by considering any economic impact to
be significant and any number of small entities to be
substantial.
The chief emphases of an RFA are to distinguish
economic impacts on small entities from economic impacts on
other entities and to consider alternative regulatory
options that may minimize the impacts on small entities
(while still achieving the statutory objectives). Five
regulatory options have been established for the NESHAP.
They are the five control options described in Section I.I.
It is of course necessary to define a "small entity."
For the present NESHAP, a "small entity" can be equated with
53
-------
a "small business" because no small government jurisdictions
or small not-for-profit organizations are affected. The
U.S. Small Business Administration (SBA) definition for a
small business in SIC 3341, Secondary Smelting and Refining
of Nonferrous Metals, is 500 employees or fewer. Granted,
SBA's definitions were established primarily to give small
businesses a fair share in the awarding of government
contracts, not to accommodate regulatory impact analysis.
Nevertheless, the SBA definition seems very appropriate for
the U.S. secondary lead industry. Five companies in the
industry have more than 500 employees: East Penn
Manufacturing Company, Exide Corporation, Fluor Corporation
(the parent of The Doe Run Company), Pacific Dunlop Limited
(the parent of GNB Battery Technologies), and RSR
Corporation. All enjoy market advantages due to scale or
structure. RSR is the largest secondary lead producer in
the U.S. and is regarded to have strong connections with
battery manufacturers. East Penn, Exide, and GNB are all
vertically integrated into battery manufacture. This makes
it easier to procure lead scrap, which, as explained in
Section 2.1, is important for success in the U.S. secondary
lead industry. Doe Run is horizontally integrated in that
it produces both secondary lead and primary lead. This may
afford the opportunity to shift production between primary
lead and secondary lead depending on comparative market
conditions, and may also make it possible to offer customers
a wider range of lead products. As multibillion-dollar
public companies, Fluor and Pacific Dunlop no doubt have
better access to capital than the other U.S. secondary lead
producers.
In contrast, none of the U.S. secondary lead producers
with 500 or fewer employees are vertically or horizontally
integrated. And none are the subsidiary of a significantly
larger organization. Thus, the 500-employees threshold
54
-------
conveniently bifurcates the U.S. secondary lead industry.
The spirit of the Regulatory Flexibility Act's purpose for
distinguishing relatively small businesses is met as above
the threshold the businesses have distinct market advantages
and below the threshold the businesses are "generally ...
independently owned and operated and not dominant in (their)
field."
While five companies in the U.S. secondary lead
industry are "not small" (more than 500 employees), 11 are
"small" (500 or fewer employees). The five companies that
are not small own 10 of the 23 secondary lead smelters in
the U.S., and 10 of the 16 that are presently active (i.e.,
not shut down). The 11 small companies own 13 smelters, six
of which are active.
All 11 small companies are impacted by the NESHAP. Six
of these companies own the seven smelters that are shut
down. Because of its incremental costs, the NESHAP will
give these smelters an additional incentive not to reopen.
The other five small companies own six active smelters. The
NESHAP imposes control costs on all secondary lead smelters
in the U.S., including these six smelters. As discussed in
Sections 3.1.3 and 5.2.2, control costs will have to be
absorbed because neither a compensating increase in the
price of lead nor decrease in the cost of lead scrap is
likely.
Furthermore, as demonstrated in Table 8, of the 13
secondary lead smelters owned by small companies, the number
that are "significantly impacted" by the NESHAP ranges from
5 under Control Option 1 to 10 under Control Option 5. In
contrast, only one secondary lead smelter owned by a company
that is not small is significantly impacted (under Control
Options 2, 4, and 5). As set forth in Sections 5.2.2.2 and
5.2.3, a facility is "significantly impacted" by the NESHAP
if, at the facility level, total annualized control costs
55
-------
TABLE 8. NUMBER OF SIGNIFICANTLY IMPACTED FACILITIES
en
en
Facility type
Control Control Control Control Control
Option 1 Option 2 Option 3 Option 4 Option 5
Owned by a "small"
company
Active
Major source
Area source
Total
Shut down
Major source
Area source
Total
Total
Owned by a company
that is "not small'
Active
Major source
Area source
Total
Shut down
Major source
Area source
Total
Total
Total
2
1
3
5
3
3
2
1
3
6
2
2
3
1
4
6
1
1
1
7
4
2
6
8
1
1
1
9
4
3
7
10
1
1
1
11
-------
exceed 0.25 C/lb (about 1% of baseline operating costs)
and/or, at the company level, capital control costs are more
than five percent of baseline total assets and post-
regulation total liabilities would exceed two-thirds of
baseline total assets if the capital control costs are
financed with debt.
Small companies are disproportionately impacted by the
NESHAP for two main reasons: 1) they tend to own smaller
facilities, which do not benefit from the economies of scale
inherent in the control costs (i.e., per-unit control costs
tend to decrease as facility size increases), and 2) they
have fewer capital resources. It is also possible that,
because of their greater resources, larger companies have
been able to control their operations more tightly in the
baseline. This would make it easier, and less costly, to
comply with the NESHAP.
The impacts of the NESHAP on active facilities are of
more concern than the impacts on facilities that are shut
down. This is because the facilities that are shut down may
not reopen, regardless of the NESHAP. From the standpoint
of active facilities, the impacts of the NESHAP on small
businesses are minimized under Control Options 1, 3, and 4
because only two active facilities owned by small companies
(both major sources) are significantly impacted. In
contrast, under Control Options 2 and 5, three such
facilities are significantly impacted. The impacts are
lowest under Control Option 1 because control costs are
lowest. Control costs are slightly higher under Control
Option 3, and then again higher under Control Option 4.
As discussed in Section 5.3, the facilities that are
significantly impacted by the NESHAP are not likely to close
as a consequence under current lead market conditions. If
the price of lead turns back down or if new secondary lead
capacity comes on stream in the U.S., however, one or two
57
-------
significantly impacted facilities may have to close. For
the reasons discussed in Section 5.2.2.2, the most likely
candidates for closure are the two facilities discussed
above that are significantly impacted even under Control
Options 1, 3, and 4.
5.4 SUMMARY AND CONCLUSIONS
Four active (i.e., currently operating) major sources
and one active area source are "significantly impacted" by
the NESHAP. This means that, at the facility level, total
annualized control costs exceed 0.25 C/lb and/or, at the
company level, capital control costs are more than five
percent of baseline total assets and post-regulation total
liabilities would exceed two-thirds of baseline total assets
if the capital control costs are financed with debt. Two
major sources are significantly impacted under all five
control options (though the impacts increase in significance
as the control options become more stringent), one is
significantly impacted under Control Option 2, and one is
significantly impacted under Control Option 5. The area
source is significantly impacted under Control Options 2, 4,
and 5.
Seven secondary lead smelters (four major sources,
three area sources) in the U.S. are currently shut down.
Because the NESHAP imposes incremental costs, all will have
an additional incentive not to reopen, especially those that
are "significantly impacted" — three under Control Option 1,
three under Control Option 2, four under Control Option 3,
six under Control Option 4, and all seven under Control
Option 5.
The NESHAP impacts small companies, defined as having
500 or fewer employees, disproportionately. While the
number of significantly impacted smelters owned by small
companies ranges from five under Control Option 1 to 10
under Control Option 5, only one smelter owned by a company
58
-------
that is not small is significantly impacted (under Control
Options 2, 4, and 5).
Under current market conditions in the U.S., with price
having rebounded by 2 or 3 C/lb in the past several months
and with lead supplies "tight," none of the four active
major sources and one active area source that are
significantly impacted by the NESHAP are likely to close as
a consequence. If price turns back down, however, a closure
or two is possible. The most likely candidates for closure
are the two facilities that are significantly impacted under
all five control options.
The economic impacts of-the NESHAP are therefore
contingent on the course of the price of lead. Ultimately
the price of lead is determined by the interaction of global
supply and demand. The past year has been characterized by
world production in excess of world consumption and
therefore a depressed price. There are signs, though, that
price may rebound (indeed, at the time of writing, early
December 1993, the LME spot price, which was in the range of
$395-400/metric ton in November, was on the rise). To
correct excess supply (manifested in the U.S. secondary lead
industry primarily as excess demand for lead scrap),
capacity is being rationalized worldwide, including in the
U.S. (to wit the five secondary lead smelter shutdowns in
1993) . As evidence, "Western-World" lead (primary and
secondary) production is projected to decline in 1993 by
three percent from 1992.72 Moreover, exports from the former
East Bloc, China, and North Korea — which soared in 1992,
single-handedly accounting for the world production surplus,
and which are believed to be similarly high in 1993 — are
forecast by the International Lead and Zinc Study Group
(ILZSG) to decline in 1994.73 The ILZSG also expects the
demand for lead in Europe to turn around in 1994, permitting
a "limited recovery" of +3 percent in worldwide demand.74
59
-------
Although the price outlook may be generally sanguine,
there is downside potential (forecasting commodity prices is
never unambiguous). For one, there still is excess world
supply.75 This means that further capacity rationalizations
and/or an increase in demand are needed to bring the world
market into balance. Secondly, it is not for sure that the
recent price recovery in the U.S. is permanent. The fall is
perennially when the demand for lead from battery
manufacturers is highest, as battery production is stepped
up in anticipation of winter demand (battery replacements
are highest in the winter) . Therefore, it is possible that
the recent price increase has been at least in part
seasonal.
In addition to the price of lead, there is another
contingency that bears on the economic impacts of the
NESHAP. This is the possibility that new secondary lead
capacity will come on stream in the U.S. The U.S. has been
characterized in recent years by excess demand for lead
scrap. In the wake of recent shutdowns (e.g., five in
1993) , the U.S. secondary lead industry is in better supply-
demand balance. (Early in 1993, before two of the year's
shutdowns, the industry's estimated capacity utilization
rate was 89%.) If substantial new capacity comes on stream,
though, the industry will again have excess capacity. This
could adversely affect profitability by depressing price and
by increasing the demand for, and therefore the cost of,
lead scrap. In turn, marginal producers could be forced out
of the industry. The facilities that are significantly
impacted by the NESHAP, particularly the two facilities that
are significantly impacted under all five control options,
might be particularly vulnerable.
As recently as a few months ago, GNB and RSR ostensibly
each had plans (the plans were announced to the public) to
build a large smelter in the southeastern U.S.76-77 Combined
60
-------
annual capacity of the two plants would be close to 200,000
metric tons. The companies' commitments to these plans are
questionable, however. In November 1993, GNB said its plans
to build a facility in Waynesboro, Ga. are "on hold
indefinitely."78 While RSR did announce in November 1993
that a site (in Montmorenci, S.C.) had been selected, in
August it had said it was taking a "short respite" from its
plans.79-80 Even if RSR proceeds with its plans, operating
permits must still be obtained. This suggests that start-up
is several years away. Asarco, the primary lead producer,
is considering entering the secondary lead business by also
building a secondary lead smelter in the Southeast.81 No
decision has been made, though.82
If price turns back down and/or new capacity comes on
stream, and as a result the NESHAP contributes to a closure
or two in the U.S. secondary lead industry, the output lost
at these one or two facilities would exceed the industry-
wide output loss estimated in Section 5.1 — 1,127-1,948
metric tons under Control Option 1, ranging up to 3,155-
5,453 metric tons under Control Option 5. Surviving
facilities, if not new facilities, can be expected to make
up the difference. Secondary lead smelters can alter
production levels fairly easily. One smelter responding in
the telephone survey said that it could double its current
capacity'(which is already "large") at very little capital
cost.83
The employment impact at these one or two facilities
would also exceed the industry-wide employment loss — 2-4
under Control Option 1, ranging up to 6-10 under Control
Option 5. Again, however, expansions or start-ups should
generate new employment to make up the difference.
Finally, the U.S. secondary lead industry would be more
concentrated after a closure or two. This should not
61
-------
significantly affect competition, however, because the lead
market is global.
62
-------
6. 0 REFERENCES
1. "Fluor to Spin Off Lead Operations." American Metal
Market. December 1, 1992, p. 1.
2. "GNB Draws 320M Ibs. of Lead from Batteries."
American Metal Market. April 8, 1993, p. 6.
3. "Exide, Ward Recycle Batteries." American Metal
Market. June 8, 1993, p. 10.
4. Reference 2.
5. Pacific Dunlop Limited. Annual Report 1992.
Melbourne, Australia. Page 28.
6. U.S. Department of the Interior, Bureau of Mines.
"The Minerals Related Implications of a Direct Tax on
U.S. Primary Lead Production and Primary Lead
Imports." Washington, D.C., April 1993. Page 29.
7. Telephone survey conducted by T. Scherer and A.
Jenkins, JACA Corporation, Fort Washington, Pa., with
three secondary lead firms, the U.S. Bureau of Mines,
a representative of a secondary lead industry trade
group, and a lead industry analyst/consultant. April
21 - June 24, 1993.
8. "New Secondary Lead Smelter Set." American Metal
Market. August 14, 1992, p. 1.
9. "Lead-Hungry RSR Pledges Protection." American Metal
Market. February 24, 1992, p. 1.
10. "RSR to Buy Billiton Lead Business." American Metal
Market. August 10, 1993, p. 1.
11. Reference 7.
12. Reference 7.
13. Mathtech, Inc. "Economic Impact Analysis of
Alternative Lead National Ambient Air Quality
Standards." Prepared for the U.S. Environmental
Protection Agency, Office of Air Quality Planning and
Standards, Economic Analysis Branch. Princeton, N.J.,
September 30, 1992. Page 3-7.
14. "Lead, Zinc Production Eases." American Metal Market.
July 16, 1993, p. 6.
63
-------
15. "Lead Prices Seen Notching Small Gains." American
Metal Market "Midyear Outlook," July 2, 1993, p. 2A.
16. Reference 6, p. 23, footnote 33.
17. U.S. Department of the Interior, Bureau of Mines.
"Mineral Industry Surveys — Lead in December 1992."
Washington, D.C., March 23, 1993. Table 1.
18. "Lead Imports Hit U.S. Secondary Smelters." American
Metal Market. March 22, 1993, p. 1.
19. Reference 18.
20. Reference 7.
21. U.S. Environmental Protection Agency, Office of
Policy, Planning and Evaluation. "States' Efforts to
Promote Lead-Acid Battery Recycling." Prepared for
the U.S. Environmental Protection Agency, Office of
Solid Waste. Washington, D.C., November 19, 1991.
Page 21.
22. Reference 7.
23. "Immsa Restarts its Monterrey Lead Refinery."
American Metal Market. August 10, 1993, p. 6.
•
24. "Immsa Plans Lead Mines Only." American Metal Market.
August 11, 1993, p. 16.
25. "Suddenly, Lead Prices Come Alive." American Metal
Market. August 6, 1993, p. 2.
26. Reference 7.
27. "But U.S. Lead Producers Seen Unlikely to Fare Better
Until Demand Starts to Rebound in Far East, Europe."
American Metal Market. August 6, 1993, p. 5.
28. Reference 7.
29. Reference 7.
30. Reference 7.
31. Reference 7.
32. Reference 7.
33. Reference 7.
64
-------
34. Reference 13, Appendix A, p. A-2.
35. "Battery Producers Hope Lead Holds on to Price
Strength." American Metal Market. July 1, 1992, p. 1.
36. U.S. Department of the Interior, Bureau of Mines.
"Annual Report 1990 — Lead." By William D. Woodbury.
Washington, D.C., April 1992. Table 1 (p. 14), Table
16 (p. 25).
37. Reference 7.
38. Reference 7.
39. Reference 7.
40. "Lead on LME Drops Below 20 C/lb. Mark." American
Metal Market. January 15, 1993, p. 1.
41. Reference 17.
42. Reference 21, p. 4.
43. Telephone conversation, T. Scherer, JACA Corporation,
Fort Washington, Pa., with Ann Noll, Battery Council
International, Chicago, II. June 4, 1993.
44. Reference 7.
45. "Nov. Lead Premium Lifted 1C on Tight Refined Lead
Supply." American Metal Market. November 1, 1993, p.
1.
46. Reference 7.
47. U.S. Environmental Protection Agency. "Strategy for
Reducing Lead Exposures." Washington, D.C., October
3, 1990. Page 29.
48. U.S. Department of the Interior, Bureau of Mines.
"Mineral Commodity Summaries 1992." Washington, D.C.,
January 1992. Page 100.
49. "Mixed Views on State of Lead Scrap Industry."
American Metal Market, August 6, 1993, p. 6.
50. Reference 13, p. 5-3.
51. Reference 13, p. 5-3.
52. Reference 49.
65
-------
53. Reference 7.
54. Reference 7.
55. "Lead Big Loser in Export Fall." American Metal
Market. March 12, 1993, p. 9.
56. "May Lead Scrap Exports Off 43%." American Metal
Market, August 11, 1992, p. 9.
57. "Lead Exports Off 13.6% in Jan." American Metal
Market. April 12, 1993, p. 11.
58. Reference 7.
59. Reference 7.
60. "Scrap Metal Slips on LME Lead's Tail." American
Metal Market. February 2, 1993, p. 2.
61. "Scrap Battery Supply Eases on East Coast." American
Metal Market. May 28, 1993, p. 2.
62. "Used Battery Prices Off 1 C/lb. on Supply." American
Metal Market. July 13, 1993, p. 10.
63. Reference 45.
64. Reference 45.
65. Reference 7.
66. Reference 15.
67. "Two Lead Concerns Cut Output." American Metal
Market. June 3, 1993, p. 12.
68. "LME Lead Tags Rebound After Hitting 7-Year Low."
American Metal Market. September 27, 1993, p. 20.
69. Reference 7.
70. Reference 45.
71. Telephone conversation, T. Scherer, JACA Corporation,
Fort Washington, Pa., with George Streit, U.S.
Environmental Protection Agency, Office of Air Quality
Planning and Standards, Emission Standards Division.
November 23, 1993.
66
-------
72. "See Less Lead, Zinc Out of East Europe." American
Metal Market. October 25, 1993, p. 4.
73. Reference 72.
74. Reference 72.
75. "Prospects for Nonferrous Mixed." American Metal
Market. March 10, 1993, p. 4.
76. "Battery Recycling Plant Set for Ga." American Metal
Market. September 25, 1992, p. 1.
77. Reference 8.
78. "GNB Battery Rethinks Building Lead Smelter."
American Metal Market. November 10, 1993, p. 2.
79. "South Carolina RSR Lead Site." American Metal
Market. November 3, 1993, p. 16.
80. "RSR Lead Smelter Plans for Southeast Still Alive."
American Metal Market. August 27, 1993, p.. 2.
81. "Lead Smelter Process Slowed." American Metal Market,
April 5, 1993, p. 5.
82. Reference 80.
83. Reference 7.
67
-------
APPENDIX A
INPUTS TO THE REGRESSION ANALYSIS
-------
TABLE A-l. INPUTS TO THE REGRESSION ANALYSIS
1970
1971
1972
1973
1974
1975
3, 1976
l
•- 1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
U.S. secondary
lead production
(metric tons)
541,943
541,405
559,368
593,559
633,853
597,342
659,132
757,592
769,236
801,368
675,578
641,105
571,276
503,501
633,374
615,695
624,769
710,067
736,401
LME spot
price of lead
(C/lb)
13.8
11.5
13.7
19.5
26.8
18.7
20.5
28.0
29.9
54.5
41.2
33.3
24.7
19.3
20.1
17.8
18.4
27.0
29.7
Price of heavy soft scrap
lead (C/lb)
Chicago basis New York basis
7.13
4.36
5.65
7.66
11.44
8.94
12.35
13.91
19.23
33.33
23.25
16.50
11.77
9.34
9.52
7.11
7.25 5.58
14.50
15.57
Producer Price Index
(PPI) for secondary
nonferrous metals
51.5
45.0
45.0
55.9
85.5
71.7
74.8
83.2
88.3
114.7
123.0
113.8
100.0
105.4
103.5
92.4
91.0
107.7
128.2
Producer Price Index
(PPI) for all
finished goods*
2.545
2.469
2.392
2.193
1.901
1.718
1.645
1.546
1.433
1.289
1.136
1.041
1.000
0.984
0.964
0.955
0.969
0.949
0.926
-------
TABLE A-l. (Continued)
Price of heavy soft scrap
lead (C/lb)
1989
1990
1991
U.S. secondary
lead production
(metric tons)
891,341
922,911
883,700
LNE spot
price of lead
(C/lb)
30.6
37.1
25.3
Producer Price Index
(PPI) for secondary
Chicago basis New York basis nonferrous metals
14.63
16.80
12.23
131.0
125.6
108.7
Producer Price Index
(PPI) for all
finished goods*
0.880
0.839
0.822
*Used to adjust all monetary inputs to 1982 dollars.
Sources: see Table 4.
i
INJ
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
EPA-453/R-94-039
3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
Economic Impact Analysis of the Secondary Lead
Smelters NESHAP—Final
5. REPORT DATE
June 1994
6. PERFORMING ORGANIZATION COOE
7. AUTHOR(S)
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Office of Air Quality Planning & Standards
U. S. Environmental Protection Agency (MD-13)
Research Triangle Park, North Carolina 27711
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
12. SPONSORING AGENCY NAME AND ADDRESS
Director
Office of Air Quality Planning and Standards
U. S. Environmental Protection Agency
Research Triangle Park. North Carolina 27711
13. TYPE OF REPORT AND PERIOD COVERED
Final
14. SPONSORING AGENCY CODE
15. SUPPLEMENTARY NOTES
16. ABSTRACT
Under the authority of the 1990 Clean Air Act Amendments, a National Emission
Standard for Hazardous Air Pollutants (NESHAP) is proposed to control emissions
from Secondary Lead Smelters. This document presents the economic impacts and
small business impacts associated with alternative regulatory options proposed in the
NESHAP.
17.
KEY WORDS AND DOCUMENT ANALYSIS
a.
DESCRIPTORS
b.lDENTlFIERS/OPEN ENDED TERMS
c. COS AT I Field/Group
Air Pollution
Secondary Lead Smelters
Hazardous Air Pollutant
Emission Controls
Economic Impact
13B
18. DISTRIBUTION STATEMENT
19. SECURITY CLASS /TIlis Report)
Unclassified
21. NO. OF PAGES
2O. SECURITY CLASS (This page/
Unclassified
22. PRICE
EPA Form 2220-1 (Rev. 4-77) PREVIOUS EDITION is OBSOLETE
------- |