EPA-600/8-77-012
AUGUST 1977
IMPACT OF THE
FEDERAL TAX CODE ON
E RECOVERY
A CONDENSATIO
PRICE
SUPPLY
(WITHOUT TAX
SUBSIDY)
SUPPLY
(WITH TAX
SUBSIDY)
QUANTITY
lunicipal Environmental Research Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, Ohio 45268
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology. Elimination of traditional grouping, was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
6. Scientific and Technical Assessment Reports (STAR)
7. Interagency Energy-Environment Research and Development
8. "Special" Reports
9. Miscellaneous Reports
This report has been assigned to the "SPECIAL" REPORTS series. This series is
reserved for reports targeted to meet the technical information needs of specific
user groups. The series includes problem-oriented reports, research application
reports, and executive summary documents. Examples include state-of-the-art
analyses, technology assessments, design manuals, user manuals, and reports
on the results of major research and development efforts.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.
-------
EPA-600/8-77-012
August 1977
IMPACT OF THE FEDERAL TAX CODE ON RESOURCE RECOVERY
A Condensation
by
Barbara J. Stevens
Columbia University
Graduate School of Business
New York, New York 10027
Grant No. R803362
Project Officer
Oscar Albrecht
Solid and Hazardous Waste Research Division
Municipal Environmental Research Laboratory
Cincinnati, Ohio 45268
MUNICIPAL ENVIRONMENTAL RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
CINCINNATI, OHIO 45268
-------
DISCLAIMER
This report has been reviewed by the Municipal Environmental Research
Laboratory, U. S. Environmental Protection Agency, and approved for pub-
lication. Approval does not signify that the contents necessarily reflect
the views and policies of the U.S. Environmental Protection Agency, nor
does mention of trade names or commercial products constitute endorsement
or recommendation for use.
11
-------
FOREWORD
The Environmental Protection Agency was created because of increasing
public and government concern about the dangers of pollution to the health
and welfare of the American people. Noxious air, foul water, and spoiled
land are tragic testimony to the deterioration of our natural environment.
The complexity of that environment and the interplay between its components
require a concentrated and integrated attack on the problem.
Research and development is that necessary first step in problem solu-
tion and it involves defining the problem, measuring its impact, and
searching for solutions. The Municipal Environmental Research Laboratory
develops new and improved technology and systems for the prevention, treat-
ment, and management of wastewater and solid and hazardous waste pollutant
discharges from municipal and community sources, for the preservation and
treatment of public drinking water supplies, and to minimize the adverse
economic, social, health, and aesthetic effects of pollution. This publica-
tion is one of the products of that research; a most vital communications
link between the researcher and the user community.
The extent to which federal tax subsidies to extractive industries re-
strict the recycling of secondary materials has been the subject of consider-
able controversy. This report analyzes the impacts of federal tax code on
supply curves for the major extractive industries, using econometric tech-
niques to obtain estimates of percentage changes in recycling which would
result from elimination of the tax advantages.
Francis T. Mayo, Director
Municipal Environmental Research
Laboratory
111
-------
ABSTRACT
This report assesses the extent to which a variety of federal tax sub-
sidies to extractive industries affect the materials flow in competing
secondary materials industries. The impacts of tax subsidies on virgin mate-
rial supply curves for the steel, paper, lead, copper, and aluminum indus-
tries are analyzed in detail. The flows of virgin and secondary materials
industries are characterized at points where the two materials substitute as
inputs to production and consumption. Econometric models specified at these
points of substitution are used to analyze the impacts of the tax subsidies
on the quantities of secondary materials recycled. In the short run, within
the limits of existing plant and equipment, it is estimated that elimination
of tax subsidies to virgin material industries would increase the flow of
scrap steel by 0.42 percent, wastepaper by 0.63 percent, lead by 0.75 per-
cent, copper by 0.35 percent, and aluminum by 1.0 percent. These estimates
make no allowance for the long run effects on investment which may arise
from the elimination of subsidies to the virgin materials industries. When
investment effects and other federal policies (for example, ICC regulation of
freight rates, labeling requirements for scrap-based products, and the free
access to minerals on federal lands) are also considered, the cumulative
impact on recycling may be considerably greater than the relatively modest
effects measured in this report.
The full report was submitted in fulfillment of Grant No. R803362 by
the Environmental Law Institute under the sponsorship of the U.S. Environ-
mental Protection Agency. This report, "Impact of the Federal Tax Code on
Resource Recovery" by R. C. Anderson and R. D. Spiegelman, is available from
the Department of Commerce, National Technical Information Service, Spring-
field, Virginia 22161, as NTIS PB-264 886/AS(EPA-600/2-77-009).
IV
-------
CONTENTS
Foreword iii
Abstract iv
Figures vi
Tables , vii
Acknowledgment viii
1. Summary and Conclusions 1
2. Theoretical Analysis 5
3. The Iron and Steel Industry 13
4. The Paper Industry 19
5. The Aluminum industry 23
6. The Copper Industry 26
7. The Lead Industry 30
References and Bibliography 35
Glossary. ..... 38
-------
FIGURES
Number Page
1 The impact of an excise tax on output and price 6
2 Materials flows and points of substitution 11
3 Primary and secondary supply and price determination in the
aluminum industry 24
4 Primary and secondary supply and price determination in the
lead industry 31
VI
-------
TABLES
Number Page
1 Impacts of Tax Subsidies on Virgin Materials Supply
Curves 2
2 Percentage Increase in Recycling of Secondary Materials
Expected from Tax Equalization 3
vn
-------
ACKNOWLEDGMENTS
The research reported in this publication was taken from a more ex-
tensive report Impact of the Federal Tax Code on Resource Recovery submitted
by the Environmental Law Institute to the U.S. Environmental Protection
Agency in fulfillment of Grant No. 803362.
The project director for that study was Robert Anderson and his asso-
ciate Richard Spiegelman. Many other persons contributed to portions of the
main report, including persons in private and public service. Oscar Albrecht
of the Municipal Environmental Research Laboratory, Environmental Research
Center, Cincinnati, Ohio served as the EPA's project officer for this con-
densation of the full report.
Vlll
-------
SECTION 1
SUMMARY AND CONCLUSIONS
While no specific legislation contains a comprehensive federal policy
regarding the use of virgin versus recycled materials, a de facto policy can
be discerned from a variety of sources. The present de facto federal policy,
as found in various pieces of legislation, appears to favor the use of vir-
gin materials rather than recycled materials. One of the most important
components of this policy is the federal tax code, which impacts on relative
prices and the use of virgin and secondary materials. Relevant components of
the tax code which impact on materials prices include:
*Percentage depletion tax deductions
*Expensing of exploration and development costs
*Domestic International Sales Corporation allowances
^Foreign tax credits
*Western Hemisphere Trade Corporation allowances
Each of these components has the effect of reducing the costs of pro-
ducing virgin materials, causing the supply curves for these materials to
be lower than would prevail in the absence of favorable taxation. The lower
supply curve leads to a lower market price for virgin materials and as a
consequence to a relatively greater use of virgin goods. As these components
of the tax code affecting the production of raw materials are available to
producers of virgin materials and not to producers of recycled materials,
the impact of the differentially favorable tax considerations appears to be
an increase in the use of virgin materials at the expense of recycled mate-
rials .
The purpose of, this report is positive rather than prescriptive. That
is, no attempt is made to determine or design the essential features of
future federal materials policies. The intent is to provide a comprehensive
analysis of one element of the present de facto materials policy--the
federal tax code--and its impact on virgin material use and recycling in
selected key industries: paper, steel, copper, aluminum, and lead. The
comprehensive analysis consists of two parts. First, a general theoretical
approach towards determining the empirical effect of the federal tax code
on virgin and recycled materials is developed. Second, on an industry by
industry basis, econometric estimates are made of the parameters necessary
-------
to determine the empirical impact of the tax policy on resource use and re-
covery. Together, the two parts allow an estimate to be made of the quanti-
tative impact of federal tax policies on materials use in each of the five
industries.
The empirical estimates of the short run impact of preferential tax
features for virgin material production on the volume of recycled material
production attempt to be optimistic in that where assumptions must be made
they err towards overestimating the increase in recycled materials produc-
tion. However, it should be remembered that the short run estimates contain
no allowance for long run impacts resulting from altered investment decisions
which can reasonably be expected if tax equalization were effected.-
Estimates of the increase in the supply curves for five virgin materials
which could be expected to result from elimination of preferential tax pro-
visions are displayed in Table 1. These figures can be interpreted as the
price increase which would be necessary in order for virgin material output
to be unchanged after removal of preferential tax treatment. The supply
curve is shifted up for various reasons by the removal of each type of
tax subsidy. Preferential taxation of capital gains probably should be
viewed as a subsidy to capital used in production; elimination of this sub-
sidy causes the supply curve to shift up in proportion to the capital inten-
sity of the industry. For the depletion allowance, which can be regarded as
a negative excise tax on capital input, supply is shifted up by the amount
of the subsidy.
TABLE 1. IMPACTS OF TAX SUBSIDIES ON VIRGIN MATERIAL SUPPLY CURVES
Upward Shift in Supply Curves
Maximum Possible "Most Likely"
Industry Impact Impact
(%) (*)
Paper
Steel
Copper
Aluminum
Lead
4.2
3.0
6.0
2.2
3.0
1.0
2.0
5.0
—
-------
As can be seen in Table I, the copper industry is predicted to experi-
ence the greatest upward shift in the supply curve for virgin metal if pre-
ferential tax consideration of producers of virgin metal is eliminated. The
supply curves for the five industries considered are expected to increase
by amounts ranging from one to six percent from tax equalization.
To estimate the increase in recycled materials which would result from
the increase in the supply curves of virgin materials presented in Table 1,
econometric estimates of the supply and demand elasticities for virgin and
recycled materials in each industry were necessary. These elasticities
tended to bias the estimates because there are severe measurement errors as-
sociated with the reported virgin materials prices. In several markets, less
than ten percent of recyled materials is actually sold and thus available
data do not accurately reflect the market price. As an alternative to the use
of direct price data, it was assumed that virgin and recycled materials are
perfect substitutes for each other and that the quantity demanded of re-
cycled goods is small enough relative to the quantity demanded of both vir-
gin and secondary goods that the price elasticity of demand for recycled
goods may be considered infinite. This assumption was used for the aluminum
and the paper industries. Another alternative to the use of the direct
price data is to assume that the elasticity of demand for recycled materials
with respect to the price of virgin materials is infinite, implying that a
single market exists for recycled and virgin materials and that the price of
virgin materials equally reflects the price of recycled materials. This
assumption is used for the lead and copper industries.
Table 2 contains a summary of the estimated impacts of virgin materials
TABLE 2. PERCENTAGE INCREASE IN RECYCLING OF SECONDARY MATERIALS EXPECTED
FROM TAX EQUALIZATION
Secondary
Material
Demand
dependent on
virgin and
scrap prices
Demand
elastic with
respect to
all prices
(*)
Demand
dependent
virgin price
only
(*)
Wastepaper
Scrap steel
Copper
Aluminum
Lead
0.04 0.63
0.42
0.35
1.00
0.75
-------
tax subsidies on the quantity of secondary materials recycled. All figures
are based on the maximum supply curve shifts which could possibly result
from an equalization of the tax treatment for the virgin and secondary
materials producers (as presented in Table 1), and are expressed as a per-
centage of the quantity of secondary materials currently recycled. Tax
equalization is expected to result in an increase of approximately one per-
cent in the quantity of aluminum recycled; smaller increases are expected
for the other industries.
Tax subsidies to virgin material production are only one aspect of
existing federal policies which adversely affect the flow of recycled ma-
terials. Other policies which allegedly may affect recycling include freight
rate discrimination (which requires producers of a secondary material in
competition with a virgin material to pay higher rates for shipment than
the producer of the virgin material), labeling requirements for scrap based
products (which may make marketing difficult if consumers believe that the
virgin material is superior to the recycled material), mining laws (which
give away valuable mining rights), failure to price residential solid waste
collection and disposal, and in other ways pricing materials at less than
their full social cost. Although the percentage impact on secondary material
production estimated to result from equalization of tax provisions is rela-
tively small (assuming existing technologies), the cumulative long run impact
of all federal policies which affect virgin material use may significantly
reduce the flow of recycled materials.-
-------
SECTION 2
THEORETICAL ANALYSIS
TAX SUBSIDIES AND THE PRICE OF VIRGIN PRODUCTS
The theoretical basis for determining the relationship between price
of a virgin material and specific tax subsidies is considered below. Three
types of tax provisions are considered:
1. Taxation of net income
2. Taxation of gross income
3. Deduction of expenses from income.
The first type of tax affects the paper industry, which is allowed to
treat as a capital gain the increase in value of timber ready for harvest
over its initial cost, thereby incurring a lower tax liability on portions
of net income. The second type of tax provision reflects the impact of
percentage depletion allowances on the mining industries--iron ore, copper,
bauxite, and lead. This kind of tax subsidy acts to exempt a portion of
gross income from taxation. Finally, expensing of exploration and devel-
opment costs in many ways acts as an accelerated depreciation allowance
and in effect shifts the time profile of tax liability. This provision is
also available to the mining industries.
Taxation of Net Income
Economics defines profits as returns minus total costs, where costs
are defined to include a "normal" rate of return on capital used in the pro-
duction process. A tax on economic profits does not alter a producer's
profit maximizing output level or price. For two reasons, the corporate
profit tax, a tax on net corporate income, is not identical to a tax on
economic profit. First, the corporate profits taxed by the United States
governments include both economic profit and normal returns for the use
of capital goods. The corporate profit tax is therefore a tax on an input
to the production process as well as a tax on profits. Second, as the cor-
porate profit tax is levied on accounting profits, the impact of the tax
depends in part on how capital is obtained: where capital is borrowed,
interest is deductible and the tax liability for the return on this capital
is reduced by the amount of the interest paid.
Economists generally believe that since the corporate profit tax applies
-------
Price
o
S'=P =S + S.Q
5.0 1
S = P
D = P,=D +D Q
d o 1
Q
V
Quantity
Figure 1. The impact of an excise tax on output and price.
-------
not only to economic profits but also to the normal return to capital, it
can be considered, at least in part, as a tax on that factor of production.
The capital gains provision allowed the timber industry has the impact of
reducing the profit tax paid by timber companies, or, to the extent that the
subsidy falls on the capital good timber, of reducing its relative price in
the production of paper. A subsidy to one factor of production should
cause a substitution of that factor for other factors (such as labor). The
size of the substitution of capital for labor as a result of the capital
gains tax subsidy to timber would depend on the elasticity of substitution in
the paper industry.
Precise estimates of the elasticity of substitution are difficult to
obtain. One method is to calculate impacts based on an extreme case; for
example, elasticity of substitution of zero, implying fixed factor propor-
tions. With fixed factor proportions, the impact of the tax subsidy on the
price of paper is greatest; however, using this assumption tends to over-
estimate the impact of removal of the subsidy on paper prices.
Taxation of Gross Income
Percentage depletion and severance taxes are similar in that both im-
pact on the taxes which must be paid on gross revenues but their effects
are in opposite directions. The severance tax is a direct tax on gross
revenues. The percentage depletion tax, by exempting a given__percentage of
gross revenues from taxes (up to a maximum of 50% of net income for any one
firm) results in a tax subsidy equal to the taxes which would have been paid
on those exempted gross revenues. Producers who are constrained by the 50%
rule would be less inclined to produce at a high level of output than would
producers not constrained by the 50% rule.
The impact on production of removal of the percent depletion allowance
can be analyzed as the imposition of an excise tax. Figure 1 shows the reac-
tion of a competitive industry to imposition of an excise tax. Before the
tax, the industry is at equilibrium at the intersection of its supply curve
(S) and its demand curve (D) producing QQ units of output for sale at a per
unit price of PQ. Imposition of a tax of t percent of the price shifts the
industry supply curve upward to S', causing a new equilibrium at a lower
output, Q-L, and a higher price, P^. The change in the price of the good, AP,
depends on the slopes of the demand and supply curves facing the industry.
Algebraically, one can develop a general expression for the percentage
change in the price of a virgin material which would be expected to result
from a revocation of the percentage depletion allowance. This expression
will be used later in the individual estimates for each mining industry.
Let the industry demand function be D = D - D..Q where DX is equal to
the slope of the demand function and D is its°intercept. Let industry
supply (after imposition of the tax--o? after elimination of the percentage
depletion allowance) equal S1 = S + S-, Q where S, is the slope of the
supply function and where S is its intercept. Referring to Figure 1,
it is clear that the amount of the excise tax, T, may be defined according
to equation (1):
-------
(1) T = D ZQ + S1 ZQ
where AQ is the total change in industry output resulting from the imposi-
tion of the excise tax (i.e., the removal of the percentage depletion allow-
ance). Also, it is clear that the change in the market price, AP, resulting
from the imposition of the tax depends upon the slope of the demand function
and the reduction in output resulting from the tax increase.
(2)AP = D^
Using (1), equation (2) can be rewritten as:
AP =
Recalling that the elasticity of demand, E,, is defined as (AQ/Q)/(AP/P)
can
e supply
or equivalently as (P/Q)/D1 , and that the elasticity of supply, Eg, can
similarly be expressed as *- (P/QjS , we can express the slopes of th
,
and demand curves in terms of elasticities of supply and demand:
(4) Sx = (P/Q)/ES
DX = CP/Q)/ED.
Substituting (4) into (3) , dividing both sides of the equation by the
market price before imposition of the tax (removal of the subsidy) , and
simplifying yields the following general expression for the percentage in-
crease in the price of a virgin product expected to result from removal
of the percentage depletion allowance:
(5)
Po =0 + %
Thus, to calculate the percentage increase in the price of the virgin
material which would result from eliminating the depletion subsidy one
needs to know the elasticity of supply, the elasticity of demand, the amount
of the subsidy, and the current market price of the good. (This analysis
is only approximately correct for several reasons. First, the elasticity
of supply may vary along the supply curve. Second, imposition of an excise
tax causes the slope of the supply function to shift; T is only approximately
equal to D^ Q + S| Q where S,* is the slope of the supply curve prior to
imposition of the tax. The formula works accurately enough, however, where
the value of T is calculated from other data and the resulting figure merely
substituted into (5) to estimate an increase in market price.)
-------
Expensing of Exploration and Development
Expensing the exploration costs in the year in which they are incurred
has a counterpart in manufacturing such as expensing of research and devel-
opment costs. Expensing of development costs such as tunnels in mines, etc.,
however, has no counterpart in manufacturing. An equivalent would be a tax
provision allowing manufacturing plants to depreciate new plant and equip-
ment acquisitions in a single year.
The immediate expensing of outlays for exploration and development
results in a postponement in tax liabilities for firms making such invest-
ments. To the extent that profitable mines are located and put into pro-
duction, the expenditures are eventually taxed later in the form of taxes
paid on profits from the mines. Although capitalization of expenses is
allowed, if a firm wishes to utilize the favorable percentage depletion tax
provision the expenditures on exploration and development must be expensed.
The expensing of expenditures on exploration and development does not
result in a tax savings on a profitable mine. When a mine for which ex-
ploration outlays were previously expensed reaches the producing stage, the
deductions are recaptured through reductions in depletion deductions that
may be taken (or through addition to the adjusted basis if cost depletion is
used). Thus, the expensing provision serves mainly to provide an incentive
to develop new sources of supply by allowing for tax purposes the immediate
recognition of costs of acquiring new properties. A similar deduction is
not available to secondary materials producers who incur expenses developing
and constructing new types of recycling plants, for example.
MODELS OF TAXATION AND RESOURCE USE
Materials Flows
A general model of the relationship between taxation of virgin materials
and resource recovery to generate recycled materials must consider demand
and supply relationships in three markets. In the product market, where
the final good to which the virgin or scrap-recycled product is an input
is traded, demand can most generally be considered to be a function of the
price of the good, income, and other prices whereas supply can be modeled
as dependent upon the price of the virgin material input, the price of the
scrap material input, and the prices of all other inputs to the production
process. In the new materials market where the virgin material input is
bought and sold, demand depends upon the price of the virgin input, the price
of the final good, and the price of the scrap input, whereas supply depends
upon the availability of the virgin material and the costs of processing the
material. In the intermediate market where the scrap input, a substitute
for the virgin material input, is traded, demand is modeled as a function of
the price of the final good, the price of the scrap input, and the price of
the virgin input, while supply is functionally determined by the availability
of scrap and its processing co'st.
Using this general framework of analysis, the impact on scrap output of
-------
a change in taxes on the virgin input can be identified. An increase in
taxes on the virgin input would be expected first to shift up its supply
curve, leading to an increase in the market price for the virgin input. An
increase in the price of the virgin input leads to an upward shift in the
supply curve in the product market and a consequent increase in the price
of the final product. In response to an increase in the price of the virgin
material input, the demand for the scrap input would shift upwards. The ex-
tent to which scrap would be substituted for virgin material would depend
upon the elasticity of demand for scrap with respect to the price of virgin
input (the cross price elasticity of demand) and the supply elasticities for
scrap and virgin materials.
As the supply curve for final products shifts upwards, a lower pro-
duction of final products will occur and perhaps lower the demand for inputs.
Thus, the impact on the virgin input prices normally will be less than that
indicated in the first round of changes. Similarly, the demand for scrap
should rise somewhat less than predicted by estimated cross price elastici-
ties multiplied by first order effects on virgin input prices.
The general model indicates that six relationships must be estimated
in order to determine the effect on the recycled material of a tax increase
on its virgin substitute. Fewer relationships will suffice where the re-
cycled and virgin input are extremely close substitutes for each other,
allowing estimation of a single demand function for recycled and virgin in-
puts. A single demand curve is estimated for the lead and copper industries.
For the paper and steel industries, the virgin material supply functions are
assumed to be infinitely elastic.
Before discussing the particular industry estimates, a brief overview
of the relationship between the primary and the secondary markets is appro-
priate. Basically, there are five potential points in the production cycle
of a raw to a finished output where a secondary material might be substitu-
ted for a primary input. Before an econometric model can be estimated, the
relevant point of substitution for a particular industry must be established
so that data can be collected for the proper variables. It is the particu-
lar technology existing within each industry which determines the proper
point in the production process at which to model the substitution of sec-
ondary for primary inputs.
Figure 2 shows diagramatically the five possible substitution points.
The secondary input may substitute for the primary at the point just past
extraction and prior to processing, directly at the point of processing,
past processing and prior to manufacture, directly at the manufacturing
point, or directly at the point of consumption in the final product market.
As will be discussed in greater detail for each industry below, the point
of greatest substitutability between the primary and the secondary sectors
differs from industry to industry. For each industry, however, demand and
supply equations are estimated at that point of potential substitution
in the production cycle where the possibilities for substitution are in
actuality the greatest, given the technological constraints on production
in that particular industry.
10
-------
Primary
A
/ «v
Secondary
—->
Extraction
Collection
Processing Manufacturing
Consumption
Figure 2. Materials flows and points of substitution.
-------
Estimation Procedures
As both supply and demand functions are specified, though both may not
be estimated for each of the five industries considered below, two stage
least squares estimation techniques are used in all cases unless otherwise
noted. This procedure yields consistent estimates of the coefficients in the
equation. Where autocorrelation was evident, generalized least squares pro-
cedures were paired with the two stage least squares estimation technique.
The parameters and statistical tests for determining the significance
of estimated coefficients are less well known for the two stage least squares
technique than for the ordinary least squares technique. The method devel-
oped by Fisher (15), who suggests using a cutoff of 2.00 on the value of
the ratio of a regression coefficient to its estimated standard error, was
followed as a test of the statistical significance of individual regression
coefficients. These t ratios are reported beneath coefficients in parenthe-
ses.
12
-------
SECTION 3
THE IRON AND STEEL INDUSTRY
USE OF SECONDARY MATERIALS IN STEEL PRODUCTION
Steel can be produced from various combinations of virgin material and
scrap iron inputs. From the virgin inputs material consisting of an inter-
mediate product, "hot metal" or pig iron, is produced. Hot metal and scrap
steel are the two basic charges for steel production. In terms of Figure 2,
substitution occurs either at point 1 just after collection of scrap or,
where collected scrap requires more preliminary processing (which would nor-
mally be required in the case of obsolete scrap such as automobiles) at
point 2.
Depending upon the type of furnace process, scrap steel may be consid-
ered a perfect or a close substitute for pig iron as it is processed into
steel. The open hearth furnace can accept any proportion of scrap steel
and pig iron as a charge, allowing perfect substitution of scrap for pig
iron. The basic oxygen furnace, however, can accept a maximum of about 401
of its charge in the form of scrap steel, and the maximum can be utilized
only if the scrap is preheated. Thus, scrap and pig iron are only imperfect
substitutes for each other where the basic oxygen furnace is used.
In practice, the open hearth furnace operates with an average of 40-45
percent of its charge in the form of scrap steel, the basic oxygen furnace
with about 30 percent of its charge in the form of scrap steel, and the elec-
tric furnace, primarily designed to process scrap, with a charge of close to
100 percent scrap steel (5). Over half of total steel output is currently
produced in the basic oxygen furnace, and this trend is continuing. It is
predicted that by 1985 only 7 percent of all domestic steel will be produced
in the open hearth furnace (as compared to about 25% today). The share of
output attributable to the electric furnace is expected not to rise (26).
This trend in technological process which is less suitable for recycling is
due to the greater efficiency of the basic oxygen furnace as compared to the
open hearth furnace. Production of a batch of steel requires ten hours in
an open hearth furnace, while only forty-five minutes are required in the
basic oxygen furnace. In sum, both currently and in the near future, scrap
steel and pig iron are imperfect substitutes for one another. In the long
run,development of a new technology might allow greater substitution between
the alternative charges.
Of the three basic types of steel scrap, approximately 100 percent of
the home scrap and 90 percent of the prompt scrap is automatically recycled
13
-------
(26). Obsolete scrap is collected and processed by approximately 1,500 firms
with 2,000 facilities located throughout the United States. These dealers
engage in collection as well as upgrading processes such as burning to remove
organic contaminants and breaking or cutting up of large irregularly shaped
pieces into smaller more regularly shaped pieces. The major purpose of the
model described below is to determine the effect on recycling of the obsolete
scrap resulting from removal of tax subsidies to producers of virgin material.
THE MDDEL: USE OF SCRAP STEEL IN THE STEEL INDUSTRY
Although the general model described above specified six separate
functions which would ideally be estimated for each industry, only two
specific functions are estimated for the steel industry. To obtain a quan-
titative estimate of the impact of removal of income tax subsidies to produ-
cers of iron ore, limestone, and coal, the following approach is used. In-
formed judgments are used to determine the increase in the price of a virgin
substitute for scrap steel (i.e. pig iron) which could be expected if tax
subsidies were eliminated. The demand and supply for scrap steel are then
modeled econometrically and coefficients of these functions, together with
the estimated price increase in pig iron resulting from eliminating tax
subsidies are used to determine the increase in recycling of scrap steel
which could be expected.
Supply and Demand in the Scrap Steel Market
Most generally, the model described here estimates how the quantity of
scrap steel supplied responds to changes in its own price and how the quan-
tity of scrap steel demanded responds to the changes in the price of obtain-
ing pig iron and the price of scrap steel.
Equations and Data
The supply and demand functions are stated fully below.
> Avl> Av2> Ex.
where: Q = the quantity of scrap steel supplied, in total dollar
s receipts for obsolete and prompt scrap.
P rrar) = the price of scrap, measured by the Bureau of Labor
^ Statistics price indices for iron and steel, deflated
by the wholesale price index.
Av-. = availability of raw scrap (obsolete and low quality
prompt) proxied by an Almon lagged scrap price variable,
under the assumption that scrap availability is negative
ly associated with previous prices (and hence with
14
-------
collections).
Av = availability measure for automatically recovered scrap
2 (high quality prompt) measured as a lagged function of
the Federal Reserve Board's index of automobile produc-
tion, measured in thousands of short tons.
Ex = quantity of scrap exports, which is assumed to shift the
domestic supply curve by an amount equal to the dollar
value of exports.
T = the state of technology in the scrap industry, proxied
p by a time trend variable designed to reflect important
recent labor saving innovations such as the automobile
shredder.
Q, = the quantity of scrap demanded, in total dollar receipts
for obsolete and prompt scrap.
= the list price of pig iron ($/ton) which was used despite
the fact that most pig iron is not traded on the market-
place and thus the list price need not accurately reflect
the true equilibrium price, deflated by the Bureau of
Labor Statistics wholesale price index.
Q . , = the activity in the scrap consuming sectors, measured by
the American Iron and Steel Institute's series on raw
steel (4) production in thousands of short tons (treated
as an Almon lag of five to eight months) produced in
basic oxygen furnaces, a proxy for the ability of the
steel producing industry to demand scrap steel.
In both equations, scrap demanded (or supplied) is measured as total
dollar receipts for obsolete and prompt scrap. To the extent that 90 percent
of prompt scrap is automatically recycled irrespective of the price elasti-
city of supply of scrap, use of this series leads to underestimation of the
price elasticity of supply of scrap steel from obsolete sources.*
Estimated Equations
The linear supply and demand equations, as estimated using two stage
least squares with an autoregressive correction are shown below:
(6) Q = -763 + 52.9 Pqrrari - 10:2 Av +2.6 Av
s (9.9) (9.9) SCrap (1.58) X (4.9) 2
- .94 Ex + 5.0 T
(6.9) (5.6)SCrap
* As, by using this series, Q in the elasticity formula Eg=(AQ/AP)(P/Q)
is approximately twice as large as would be desired.
15
-------
(7) Q = 601.7 - 23.5 P,._rsm + 9.66 P . + . 30 Q, .
* (3.7) scraP ci.eg)?1* (ii.ofeel
The estimates used monthly data from 1962 through 1974. All signs of
the coefficients were as expected, the magnitudes seemed reasonable, and most
had a t ratio of two or more.
From equation (6), supply elasticities may be calculated. At the mean,
the price elasticity of supply is 1.4. This takes into account both Pscrap
and Avi, tne lagged price of scrap. As discussed above, however, the
data series used for scrap supply biases the estimate of price elasticity
downwards. Adjusting the estimated elasticity to reflect only those scrap
receipts attributable to obsolete and other "not automatically recycled"
scrap, one obtains a price elasticity of supply of 2.32.
Other variables in equation (6) tended to confirm expectations. High
prices for scrap in the recent past indicating low current availability of
scrap (Avl), are negatively associated with the quantity of scrap supplied.
The availability of prompt scrap (Av2) is positively associated with scrap
supply, as expected. However, the percent increase in supply resulting
from a one percent increase in availability of prompt scrap, calculated as
.2, seems a bit low. If prompt scrap is truly "automatically" recovered,
and truly represents approximately 50 percent of total recycled prompt and
obsolete scrap, an elasticity of approximately .5 would have to be expected.
The low estimate may be attributable to the fact that automobile production
fails to account for production of all prompt scrap. A one ton increase in
exports leads to a decrease in scrap supplied domestically or approximately
one ton (-94), as expected. The time trend, reflecting technological change
in the scrap industry, indicates that, all other things equal, the quantity
of scrap supplied increases by about one percent per year.
The estimated demand function indicates a price elasticity of demand
of .63 and the usual downward sloping demand curve shape. A one percent
increase in the price of pig iron is associated with a .28 percent increase
in the demand for scrap steel. The coefficient on which this cross"price
elasticity estimate is based, however, is not estimated with great precision
(t ratio computed as 1.69).
The Supply Price of Pig Iron and Tax Subsidies
Having estimated the cross price and own price elasticities of supply
and demand in the scrap steel industry, one additional data item is needed
to determine the effect on scrap steel output of elimination of tax sub-
sidies to producers of pig iron. This is the increase in the price of pig
iron which would arise were these subsidies no longer available to pig iron
producers.
Two tax subsidies for virgin inputs to pig iron are available: per-
centage depletion allowances of fifteen percent on iron ore, ten percent on
coal and fourteen percent on limestone and capital gains taxation of royal-
ties in the production of coal and iron ore. The analysis of price impacts
16
-------
will be confined to the depletion allowances. Using the input proportions
of coal (.77 tons), limestone (.20 tons), and iron ore (1.5 tons) cited by
Yaughan (29), assuming that the full depletion allowance is taken and that
the relevant corporate tax rate is .48, and taking the prices per ton for
inputs to pig iron production as of May 1974 ($30 for coal, $2 for limestone,
and $11 for iron ore), the maximum upward shift in the supply curve of pig
iron attributable to increased taxes resulting from elimination of depletion
allowances is calculated as the product of input tons x price/ton x deple-
tion allowance x corporate tax rate for each of the three virgin inputs to
pig iron. The maximum price increase, per ton of pig iron, is $2.33, of
which $1.11 is attributable to the loss of the depletion allowance on iron
ore, $.03 to the loss of the depletion allowance on limestone, and $1.19
>to the loss of the depletion allowance on coal. This price increase repre-
sents about 31 of the May 1974 price per ton of pig iron of $70. Thus, the
maximum effect of elimination of depletion allowances is a three percent
increase in the price of pig iron.
FEDERAL TAX IMPACTS ON THE RECYCLING OF SCRAP STEEL
The calculated elasticities of demand will be used to estimate the per-
centage increase in the recycling of scrap steel which would result from
elimination of percentage depletion allowances to producers of virgin mate-
rials. This supply increase, it should be emphasized, is an upper bound
estimate of the short run impact of tax equalization.
Short run impacts on scrap steel production can be calculated from the
previously discussed upward shift in the supply curve of pig iron (3%), the
cross price elasticity of demand for scrap steel (.28), and the price elas-
ticities of demand (.63) and supply for scrap steel (2.32). The three per-
cent price increase in pig iron predicted from elimination of the percentage
depletion tax provision times the cross price elasticity of demand for scrap
steel (.28) leads to an upward shift of .841 in the demand for scrap function.
Using an analysis comparable to that developed in Figure 1 and equations
(1) through (5), the percent change in the market price resulting from this
.84% upward shift in the demand function can be calculated as ,84Ej)/(ED + Eg)
which is equal to .18%, where EQ and Eg refer to the elasticities of demand
and supply, respectively. The quantity of obsolete scrap steel recycled
would increase by the change in price multiplied by the supply elasticity,
or .18 x 2.32 = .42%.
An estimate of the increase in the quantity of obsolete scrap steel
which would be recycled in the long run can be obtained solely from the
estimated supply curve for scrap steel. To make this estimate, it must be
assumed that there is long run perfect substitutability between virgin and
scrap inputs and that the long run supply of virgin inputs to steel produc-
tion is highly responsive to price changes. With these assumptions, an upper
estimate on the increased quantities of obsolete scrap which would be recy-
cled should the tax subsidization of the inputs to pig iron be eliminated
can be obtained by multiplying the supply elasticity of scrap steel by the
change in the price of virgin based pig iron, to obtain a figure of 2.32% x
3% = 6.4%. This figure is biased upwards in large part due to the assumption
17
-------
of perfect substitutability between virgin and scrap inputs. Given current
technology, this assumption is clearly an over simplification of the iron
and steel industry.
18
-------
SECTION 4
THE PAPER INDUSTRY
USE OF SECONDARY MATERIALS IN PAPER PRODUCTION
The percentage of paper products recycled has declined over the last
twenty-five years. In 1950, about 28% of paper products were recycled, but
by 1973 the figure had dropped to 17%. (14)
As in the iron and steel industries, except in periods of exceptionally
weak demand, home and prompt paper scrap is generally recycled. In order of
importance for recycling, obsolete paper scrap originates in commercial,
industrial, and residential establishments.
Most recycled paper is sold as paperstock, a source of cellulose fibers
in direct competition with other fiber sources, principally virgin wood.
Paperstock and virgin wood may be used as substitutes in the production of
paperboard (boxboard, corrugating medium and linerboard, for example), tissue,
and newsprint. Substitution occurs, then, at flow point 2 in Figure 2, as is
the case in the iron and steel industry. Paperboard production accounts for
over three fourths of the consumption of paperstock. In modeling the demand
for paperstock, then, it must be considered as a demand derived largely
from the demand for paperboard.
THE MODEL: USE OF WASTE PAPERS IN THE PAPER INDUSTRY
Supply and Demand in the Wastepaper Industry
As for the iron and steel industry, only two of the six general equa-
tions in the model are specified and estimated. These are the equations of
the demand for and supply of wastepaper products, or, more specifically, the
demand for and supply of prompt and obsolete paperstock.
Equations and Data
In general, the demand for paperstock is hypothesized to depend on its
price, the price of virgin woodpulp, and activity in the paperboard industry.
The supply of paperstock is hypothesized to depend on its price and the
availability of wastepaper.
The estimated equations are presented below:
19
-------
Qd = W V QPbi> V
Qs = Q (Pps,
where:
Q = the quantity of paperstock demanded, measured as the raw
tonnage (in thousands of tons) of all grades of paperstock
consumed at user mills, Department of Commerce Series M26A
in the Current Industrial Reports.
Ps = the price of paperstock, a Bureau of Labor Statistics index
p divided by the Bureau of Labor Statistics wholesale price
index (times 1,000).
P = the price of woodpulp, measured by the Bureau of Labor Sta-
^ tistics composite of prices for a number of woodpulp grades
divided by the Bureau of Labor Statistics wholesale price
index (times 1,000).
Q_, . = a measure of activity in the paperboard industry, proxied
by the Department of Commerce Series M26A Current Industrial
Reports figures for output of construction paper and board,
in thousands of tons.
Q' , 2 = a measure of activity in the paperboard industry, proxied
P as production of combined paperboard in thousands of tons.
Together, construction and combined paperboard account for
the use of 851 of paperstock.
QQ = the quantity of paperstock supplied, measured as Q .
s d
= availability of inputs to the production of paperstock,
proxied as the Federal Reserve Board index of production of
converted paper products and paperboard carton production
(times 10).
= availability of wastepaper products for recycling, proxied
by lagged prices of paperstock, under the assumption that
recent high prices cause greater collection and recycling
and a consequent current scarcity of recyclable materials.
Estimated Equations
The linear supply and demand functions, estimated using two stage least
squares on monthly data from January, 1961 to December, 1972 are presented
below. The t ratios are in parentheses beneath the coefficients.
20
-------
(8) (
(9) (
3 = 49.1 - .13 P -
(2.37)
3s = 457.1 + .46 Pps
(3.67)
" -11 Pwp '
(1.94)
+ .31 Avl
(10.62)
t- .51 Q bl
(6.20)
- .40 A^
(3.03)
1.11
(13.30)
All coefficients in the demand function have the expected signs and most
have large t ratios. In addition, the magnitudes of the coefficients seem
reasonable. For example, McClenahan shows that approximately twelve percent
of paperstock consumed is used in construction paper, where it accounts for
two thirds of production (23). Since the measure used for Qp^l included
output which used no paper stock, its coefficient should be less than two
thirds; in fact, the coefficient of Q bl is .51.
The price elasticity of demand for paperstock is calculated as .16.
The cross price elasticity of demand for paperstock, the percent change in
paperstock demand resulting from a one percent change in the price of wood
pulp is .13.
All estimated coefficients in the supply function have the expected
signs and most have t ratios greater than 2.0. The elasticity of supply with
respect to the current price of paperstock is 0.40. The elasticity with
respect to both current and lagged prices can be thought of as representing
a longer range price elasticity; its value is .15.
The Supply Price of Wood Pulp and Tax Subsidies
Since 1944, timber manufacturers have been allowed to treat the increase
in value of timber from planting to harvest as a capital gain. The major
effect of eliminating this tax provision would be to cause marginal lands
planted with timber (called stumpage) to be removed from timber production,
causing an upward shift in the long run supply curve for timber (stumpage)
and wood pulp.
If an elastic supply of stumpage is assumed, with a unit of land cur-
rently earning $1.00 before taxes and $.70 after taxes, removal of the
capital gain tax privilege implies that the same land must earn $1.346 to
yield the same $.70 as before. Thus, the maximum long run impact of capital
gains taxation on the price of stumpage would be 34.6%. As stumpage prices
have historically averaged between 6 and 12 percent of market pulp prices,
the maximum long run impact from elimination of capital gains taxation on
market pulp prices would be 12 x 34.6 = 4.2% (16, p. 149).
FEDERAL TAX IMPACTS ON THE RECYCLING OF WASTEPAPER
Calculation of the increase in the quantity of wastepaper recycled as
a result of eliminating the tax provisions favoring use of virgin paper fol-
lows the same procedures as used for the scrap steel industry. In the short
21
-------
run, a 4.2% rise in the price of woodpulp causes a .551 rise in the demand
for paperstock function (the cross price elasticity of demand for paperstock,
0.13, times the percent increase in the price of woodpulp, 4.2%), which in
turn results in an increase in the market price for paperstock. This can be
calculated as .55 x En/(En + Eg) where En. and Eg refer to the elasticities of
demand and supply, respectively, for paperstock. The percentage increase in
paperstock price, .28, when multiplied by the price elasticity of supply of
paperstock, .15, gives the percentage increase in recycled paperstock, .04%,
which would result from equalizing tax treatment of primary and secondary
producers of inputs to paper manufacture.
If in the long run perfect substitutability between wood pulp and paper-
stock prevails, then the impact of tax equalization on paperstock production
can be calculated as the price increase in woodpulp from equalization (4.2%)
times the price elasticity of supply for paperstock (0.15) which gives 0.63
percent.
22
-------
SECTION 5
THE ALUMINUM INDUSTRY
USE OF SECONDARY MATERIALS IN THE ALUMINUM INDUSTRY
Aluminum is manufactured by processing alumina (aluminum oxide) bearing
ores (for example, bauxite), to obtain alumina and then electrolytically
reducing alumina in a molten bath of cryolite into aluminum ingot. The sub-
stitution of virgin and secondary aluminum supplies occurs primarily in
the casting industry, where aluminum scrap competes directly with primary
aluminum ingot. In Figure 2, this would occur at flow number 3.
The aluminum industry, which in 1968 produced 418 thousand short tons
from virgin products and 817 thousand short tons from secondary metal, is
second only to steel in domestic metal output (3). In contrast to other
industries such as lead and copper, recycled obsolete scrap accounts for only
one fifth of recovered aluminum. In part this fact is due to the newness of
the aluminum industry and to the consequent lack of a large stock of metal
available for recycling. However, in large part the fact is also due to
a technological constraint in the production of aluminum, whereby it is
not possible to produce as pure a product from recycled metal as it is from
bauxite ore. Most recycled metal, then, comes from hone or prompt sources.
What obsolete scrap is recycled comes largely from the airframe and automo-
bile industries (9).
THE MODEL: USE OF SECONDARY ALUMINUM IN THE ALUMINUM INDUSTRIES
Supply and Demand in the Scrap Aluminum Market
Due to the relative abundance of bauxite ores in the earth, it is ex-
pected that the price of aluminum is determined by production costs. Given
this expectation, as Figure 3 shows, the demand elasticity is irrelevant for
determining the impact of tax equalization on recycling of aluminum.
In Figure 3, the Secondary Supply curve is upward sloping, reflecting
the fact that production costs of secondary aluminum rise as less desirable
scrap is recycled and more effort is expended in collection. The primary
supply curve is drawn as perfectly elastic. An equalization in taxes for
the primary sector would result in an increase" in the price of primary
(and secondary) aluminum equal to the value of the foregone tax subsidy.
(In Figure 3, P' as compared to Po.) Under such assumptions, the percentage
change in secondary production resulting from a price increase in the primary
23
-------
Price
Scrap Supply
Virgin Supply
with tax
equalization
Virgin Supply
Demand
Quantity
Figure 3. Primary and secondary supply and price
determination in the aluminum industry
24
-------
market (Q ' - QS)/QS, can be calculated using the percentage price increase,
(P' - P0)s/ PQ, and the price elasticity of secondary supply.
Thus, an equation for secondary supply is estimated using two stage
least squares on monthly data from 1962 through 1972.
(10) Q = 5.8 -i- 1.44 Pa
S 3-
(1.42)
where:
Q = the net metallic recovery from aluminum scrap and sweated
pig consumed at secondary smelters.
P = the price of a #12 secondary alloy ingot.
3.
The elasticity of supply is .86. This number must be interpreted with
caution as the coefficient from which it is derived is not precisely esti-
mated and is of only marginal significance.
The Supply Price of Aluminum Ores and Tax Subsidies
The major tax subsidy to the aluminum industry is the percentage de-
pletion allowance on bauxite ore. In 1973, Alcoa obtained a percentage
depletion allowance (as a percent of the value of ingot produced) of 1.41;
Kaiser Aluminum and Chemical, 1.9%; and Reynolds Aluminum, 2.3% (13 and 17).
Since Reynolds Aluminum controls most of the domestic production of bauxite,
and since the depletion rate for domestic production exceeds that for for-
eign output, it would be expected that the depletion allowance is more
important to Reynolds Aluminum as a percentage of final product price than
to the other producers.
The impact of the percentage depletion allowance in terms of lower
prices for final outputs is, at most, equal to the tax savings, or 48% of
the ratio of depletion to final product price. Percentage depletion would
thus reduce market price by somewhere between .48 x 1.4 = .7 and .48 x 2.3 =
1.1 percent.
FEDERAL TAX IMPACTS ON THE RECYCLING OF SCRAP ALUMINUM
Elimination of the percentage depletion allowance for aluminum would
result in a maximum price increase of 1.1 percent. Multiplying this figure
by the price elasticity of supply for scrap aluminum gives the maximum in-
crease in output of recycled aluminum to be expected from tax equalization.
This figure is 1.1 x .86 = 1.0%.
25
-------
SECTION 6
THE COPPER INDUSTRY
USE OF SECONDARY MATERIALS IN THE COPPER INDUSTRY
Copper ranks third in production by weight among metals in the United
States. The secondary copper industry is the largest of the secondary non-
ferrous metal industries. In 1972, secondary copper production totaled
1,479 thousand tons, of which 40% was attributable to old or obsolete scrap
and 601 to new or prompt scrap. Total output equaled about 42% of the do-
mestic copper supply produced in 1972 (22).
The largest source of obsolete copper scrap is electric wire, followed
by cartridge brass, automotive radiators, low grade scrap and residue,
railway car boxes, and magnet wire. Copper scrap is sold in dispersed mar-
kets. There are approximately 80 secondary smelters and about one dozen
primary producers (18). Copper scrap is consumed by each of these entities.
Some substitutions of scrap for virgin copper occur at each stage of the
flows diagrammed in Figure 2. Brass mills and primary producers each pur^
chase about one third of copper scrap, foundries buy about 6% of copper
scrap, and the remainder is bought by secondary smelters, ingot producers,
and, to a lesser degree, chemical plants.
THE MODEL: USE OF SCRAP IN THE COPPER INDUSTRY
As prompt scrap is generally substituted for primary copper ingots in
the brass industry, and as old or obsolete copper scrap substitutes for pri-
mary copper earlier in the production process, it is useful to disaggregate
the copper scrap supply function into two parts: that supplying new scrap
and that supplying old scrap. In addition to the two supply functions for
copper scrap, the model below presents a supply function for new copper, a
demand function for copper output, and an equation describing the net
exports of copper from the rest of the world to the United States.
Supply and Demand in the Copper Industry
An annual model, developed by Fisher, et. al., is postulated below (15),
Use of annual data means that estimated elasticities can be considered to
provide relatively long run estimates of supply and demand. The five equa-
tions in the model are:
Qsi
26
-------
QS3 - Qs3tct)
Ct ' CtCPc> Pa> IP, ID, IDt_lf Ctt.1)
Xf. = ^(PDIF, CDIF, D)
where:
Qsl = the total United States mine production of primary copper,
in thousands of metric tons.
PCI = the price of copper, from the Engineering and Mining
Journal, divided by the wholesale price index, expressed
as $ /me trie ton.
= a measure of copper availability, proxied by lagged values
of domestic output of primary copper.
QS2 = the supply of obsolete scrap, in thousands of metric tons.
Pc2 = ^ie Pr^ce °f cop?6*" scrap, obtained from the London Metal
Exchange, deflated by the wholesale price index.
= the availability of scrap for recycling, proxied by a
series generated by assuming a 1948 stock of scrap and
adjusting this stock annually to account for production,
net imports, and removals for recycling, measured as the
lagged scrap recycled in the preceding year.
= the supply of new scrap, in thousands of metric tons.
C = the total consumption of copper, in thousands of metric
tons. New scrap is assumed to be generated as copper is
used for other purposes.
P = the price of the substitute for copper, aluminum, lagged
a one year.
IP = the Federal Reserve Board's index of industrial produc-
tion.
ID = the change in inventories of consumer durable goods.
ID = the change in inventories of consumer durable goods lagged
t- one year.
Ct = the consumption of copper, lagged one year.
X = exports of copper, from the rest of the world to the
27
-------
United States.
PDIF = the spread between the United States and the London Metal
Exchange price of copper.
CDIF = the difference between domestic consumption and primary
domestic output (CL - CLi) •
D = a dummy variable taking the value of 1 if export controls
exist for copper and zero otherwise.
These equations were estimated, using two stage least squares and a
correction for autocorrelation, on annual data from 1950 to 1968, and are
reported below. The t ratios are in parentheses beneath the coefficients.
(11) Q T = -160.04 + 14.27 P p1 + .726 A,
51 (2.99) Ci (3.55) Vi
\
(12) Oc7 = -9.878 + .422 P - .373 A0 (log-log form)
sz (3.99) c (2.96) vz
(13) Qo^ = -275.2 + .3961 G
S'5 (7.56) t
(14) C* = -14.75 - 12.37 F + 8.29 P + 5.08 IP + 60.69 ID
r (7.06) c (1.79)a (5.56) (9.43)
- 44.40 IDt_-, + .79 Ctt ,
(6.25) (7.02) r"J'
(15) X = 795.5 + 1.397 PDIF + .934 CDIF + 145.8 D
t (2.21) (2.20) (2.64)
With the exception of the coefficient of the price of aluminum (the
substitute for copper), all coefficients have t ratios over two. Moreover,
all coefficients are of the expected signs. Fisher reports the elasticities^
calculated at the means (With the exception of those obtained from equation*
(12) , which was estimated in a log-log specification and whose coefficients
can, then, be directly interpreted as elasticities) for the various sections
of the model. The price elasticity of supply of primary copper is 1.67
(from equation (11)), that of obsolete scrap is .42 (based on current prices,
from equation (12)), and .32 when prices are lagged, allowing short run
adjustments to be elininated from the estimate. From equation (14), Fisher
calculates that the short run price elasticity of demand is .173 and the
long run price elasticity is .876. No elasticities are reported for the
exports to the United States in Fisher's model.
The Supply Price of Copper and Tax Subsidies
Copper producers have available to them many income tax subsidies:
percentage depletion allowances, expensing of exploration and development
28
-------
foreign tax credits, and investment tax credits. As the impact of expensing
provisions are difficult to quantify, and as the investment tax credit is
available jointly to primary and secondary producers of copper, and as for-
eign tax credits (since the nationalization of the copper industry in Chile)
have a minimal impact on the copper industry, quantification of tax impacts
on the supply curve of primary copper producers is limited to an estimate
of the effect of the percentage, depletion allowance on copper prices.
Copper producers may take a 15% depletion allowance on revenues from
copper concentrates. Copper concentrates, in general, are not traded; how-
ever, the revenues attributable to them are obtained by allocating costs
to other operations involved in transforming copper concentrates to copper
(primarily smelting and refining), and treating the residual as the value
of copper concentrates. While determination of the tax impact at this level
is difficult, one can-estimate that, given that 5/6 of the price of blister
copper is not attributable to refining and smelting of concentrates (and is,
then, allocated to concentrates), the depletion rate is 15%, and the tax
rate is .48, that the price of blister copper is reduced by the tax savings
attributable to percentage depletion by an amount equal to (5/6) x .15 x
.48 = 6%. Actually, 5% seems to be a better estimate of the tax impact of
depletion on blister copper prices as a Treasury Depletion Survey (25, p.309)
showed that rates of less than 15% are common for the industry.
FEDERAL TAX IMPACTS ON THE RECYCLING OF SCRAP COPPER
The following estimated effect of tax equalization on the recycling of
scrap copper assumes that elimination of tax subsidies to primary copper
producers will cause their supply curve to shift up by the full amount of
the tax subsidy foregone. As the Fisher model does not provide an elas-
ticity for foreign suppliers of copper with respect to our domestic price,
the estimate will be biased upwards to the extent that increased imports of
copper resulting from an increase in our domestic price are ignored.
Assuming that tax equalization will cause a 5% increase in the price of
blister copper, the supply curve for final copper products will shift up-
ward by a weighted average of the price elasticity of supply of the primary
and secondary producers. Weighting the primary elasticity (1.67) by its
share of the market (.45) and the scrap sector's elasticity (.32) by its
share of the market (.185), (the remaining supply coming from new scrap or
imports), the industry supply curve for final copper products is predicted
to increase by 4.05% as a result of tax equalization. The supply curve for
final output is assumed to shift upward by 5% times the share of primary
copper (0.45) in the total domestic industry supply, or 2.25%. From equa-
tion (5), the percentage increase in the equilibrium price of copper can
then be calculated as: (2.25% x .81)/(.876 + .81) = 1.08%, where .81 is
the (weighted) price elasticity of supply and .876 is the price elasticity
of demand. Based on the price elasticity previously calculated, a 1.08%
increase in the price of copper will result in a .35% increase in the supply
of obsolete copper recycled.
29
-------
SECTION 7
THE LEAD INDUSTRY
.#
USE OF SECONDARY MATERIALS IN THE LEAD INDUSTRY
The major source of primary lead is galena, a lead sulfide, which con-
tains desirable antimony as well as sulfur. Smelting of galena burns off
the sulfur and reduces lead ores to metallic lead. Although antimony is
present in lead ores, most such ores are refined into soft lead products
such as gasoline additives, cable sheathing, lead oxides, pigments, and a
variety of lead alloys.
Secondary lead supplied the major portion of domestically consumed lead
in 1973: 42% used came from secondary sources, 39.5% from primary output
and 18.5% from imports. Unlike the steel industry, virgin and scrap lead
are processed by separate firms, and the majority of scrap lead is obsolete.
Only two percent of scrap lead is home scrap and from fifteen to eighteen
percent is prompt scrap; the remainder is obsolete scrap. Automobile and
other lead-based batteries provide about sixty percent of the obsolete scrap
lead.
Scrap lead substitutes for primary lead at point flows three and four
in Figure 2. Battery lead is processed in either a blast or a reverberating
furnace, with some antimony added, to produce antimonial lead which is used
directly in the manufacture of grids for new battery plates. Other scrap
lead is processed in furnaces to yield a less refined product which must
then proceed through flows two and three before being offered for final
consumption.
THE MODEL: USE OF SCRAP LEAD IN THE LEAD INDUSTRY
Supply and Demand in the Lead Industry
The model assumes, as shown in Figure 4, that the primary and secondary
lead products constitute perfect substitutes for one another, implying that
the separate supply curves for each may be added horizontally to form
a single supply function for the industry as a whole. Equalizing of tax
subsidies to the primary and the secondary producers of lead results in an
upward shift of the primary lead supply function to Sp', leading to a
consequent decrease in the quantity of primary lead demanded (-AQp), an
upward shift in the industry supply function for lead (to Si1), and an in-
crease in both the market price for lead (to Pi'") and the quantity of scrap
30
-------
Price
AP
AQs
AQp
Qi Qo
Si'
Si = Ss+Sp
Quantity
Figure 4. Primary and secondary supply and price determination in the lead industry.
-------
scrap lead provided (AQs).
Equations and Data
The model contains equations for the supply of primary and secondary
lead as well as for the demand for lead. The supply of primary lead is
hypothesized to depend on the price of lead and on recent outputs of the
lead production industry. The supply of scrap lead is hypothesized to de-
pend on the price of lead and the availability of two year old batteries for
recycling. Consumption of lead is modeled as dependent on the price of lead
and the general level of industrial activity. The equations and data are
described below.
where :
Q = output of primary lead in thousands of short tons, obtained
11 from the Mineral Yearbook.
p = the producer price of lead in cents /pound deflated by the
Bureau of Labor Statistics wholesale price index.
A = the availability of lead, measured as the one year lagged
v output of primary lead.
= the output of secondary lead, measured in thousands of
short tons.
B = the Federal Reserve Index of replacement of storage bat-
tery production, a proxy for availability of scrap lead
for recycling.
C, = domestic consumption of lead, measured in thousands of
short tons.
I = the index of industrial production, from the Bureau of
1 Mines Mineral Yearbook.
Estimated Equations
Estimates of the preceding equations, based on annual data from 1949
to 1967 using two stage least squares with a correction for autocorrelation,
are shown below:
(16) Q.. = 4.19 + 11.58 P + .45 A
11 (4.21) 1 (3.70)vl
32
-------
(17) Q12 = 27.29 + 18.23 P, + 2.98 B
(7.49) (18.10)
(18) Cx = 711.3 - 17.31 P, + 10.23 L
(2.62) (4.94) L
All coefficients are of the expected signs and significant at the .05
level or higher. From equation (16) it is calculated that the short run
price elasticity of primary lead supply (using information from the lagged
variable) is about 1.00. The estimated price elasticity of supply of scrap
lead is .48.
^From equation (18) the price elasticity of demand was estimated as .21,
indicating that consumption is largely unresponsive to changes in market
price. This is to be expected for an input which is low in value relative
to the final output price (as it is in pigments, gasoline additives, and
bearings), or has few or no substitutes at present price levels (as in stor-
age batteries).
The Supply Price of Primary Lead and Tax Subsidies
Primary lead producers may use a 221 percentage depletion allowance
on domestically produced lead concentrates, a commodity which, like copper
concentrates, is not traded on any organized market. In 1972, however, the
proportion of lead concentrates in refined lead was estimated as 7.6/15 (28).
In calculating the effect of tax equalization on the supply curve for
primary lead it is assumed that primary and secondary lead are perfect sub-
stitutes for one another. In this case, the subsidy attributable to the pro-
ducers of primary lead at a maximum is equal to the product of the corporate
tax rate (-48), the percentage depletion allowance (.22) and the proportion
of the value of lead attributable to lead concentrate (7.6/15), or 5.31.
However, firms producing lead claim less than the allowed 22% in depletion
(25). On average, they claim only about half the maximum rate, or 111.
Thus, the price of primary lead is decreased about 2.5 to 31 by the presence
of the percentage depletion allowance.
FEDERAL TAX IMPACTS ON THE RECYCLING OF SCRAP LEAD
As the relationships estimated in the model described above did not
yield any significant relationships between domestic prices and the import
of lead, and as estimates of the reaction of imported lead suppliers to an
increase in the domestic price of lead can consequently not be quantified,
the following estimates of the impact of tax equalization on resource re-
covery in the lead industry will tend to overstate the probable increase
in recycling of lead. Tax equalization would no doubt raise lead prices,
causing an increase in lead imports and thus lessening the impact on lead
recycling discussed below.
The price elasticity of total lead supply is equal to the weighted
33
-------
average of primary, secondary, and import elasticities of supply, or .6. The
supply curve for final lead outputs is shifted by .4% for every one percent
increase in the supply curve for primary lead. The maximum impact of percen-
tage depletion is estimated at 5.3% of the value of lead concentrates; thus
elimination of the depletion allowance would result in an increase in the
lead industry supply curve of approximately 2.1%.
The equilibrium price of lead rises by a percentage equal to the pro-
duct of the percent shift in final lead output (2.1%) and the estimated
supply elasticity (0.6), divided by the sum of the supply and demand elas-
ticities (0.6 and 0.21). Assuming that elimination of the percentage de-
pletion allowances results in a shift in primary supply of 5.3%, as stated
above, and the supply curve for final lead outputs shifts by 0.4% for
every one percent shift in primary supply, the equilibrium price rise would
amount to 1.6 percent. The estimated increase in scrap lead recycling is
then 1.6 percent times the scrap supply elasticity (0.48) or 0.75 percent.
34
-------
REFERENCES AND BIBLIOGRAPHY
1. Adams, R.L. Economic Analysis of the Junk Automobile Problem. U.S. De-
partment of the Interior, Bureau of Mines. Washington, D.C. Information
circular 8596, 1973.
2. Albrecht, 0. W. and R. G. McDermott. Economic and Technological Impedi-
ments to Recycling Obsolete Ferrous Solid Waste. U.S. E. P. A. National
Environmental Research Center, U.S. Department of Commerce. Washington,
D.C. NTIS PB-223-134. October, 1973.
3. Aluminum: Mineral Facts and Problems. U.S. Department of the Interior,
Bureau of Mines. Washington, D.C. 1970.
4. American Iron and Steel Institute. Annual Statistical Reports. Wash-
ington, D.C.
5. American Iron and Steel Institute, Yearbook. Washington, D.C. 1970.
6. American Paper Institute. Statistics of Paper and Paperboard, 1974.
New York, New York, July, 1974.
7. An Econometric Analysis of the Aluminum Industry. Charles River Asso-
ciates. U. S. Department of Commerce. NTIS PB-199-789, 1971.
8. Lead Industries Association, Inc., Annual Review 1973: U.S. Lead
Industry. New York, 1973.
9. Battelle Memorial Institute. A Study to Identify Opportunities for
Increased Solid Waste Utilization. Columbus, Ohio. Volumes II to VII.
1972.
10. Bennett, Harold J., L. Moore, L.E. Welborn, and J.E. Toland. An Economic
Appraisal of the Supply of Copper from Primary Domestic Sources. U. S.
Bureau of Mines Information Circular #8598. Washington, D. C., USGPO
1973.
11. Booz, Allen and Hamilton. An Evaluation of the Impact of Discriminatory
Taxation on the Use of Primary and Secondary Raw Materials. Final Re-
port to U.S. E.P.A. Washington, D.C. 1974.
12. Copper: Mineral Facts and Problems. U.S. Department of the Interior,
Bureau of Mines. Washington, D.C. 1970.
13. Corporation 10-K Reports.
35
-------
14. Darnay, Arsen, and W.F. Franklin. Salvage Markets for Materials in
Solid Wastes. Midwest Research Institute, Washington, B.C. CPE 69-3,
1972.
15. Fisher, F., P. Cootner, and N. Baily. An Econometric Model of the World
Copper Industry. Bell Journal of Economics and Management Science, Au-
gust 1972, pp. 568-600.
16. U.S. Department of Agriculture Forest Service. The Outlook for Timber
in the United States. Forest Resource Report No. 20. Washington, D.C.
USGPO. July, 1974.
17. Gillis, M., and C. McLure, Jr. The Incidence of World Taxes on Natural
Resources with Special Reference to Bauxite. American Economic Review
6_5 (2), May, 1975, pp. 389-396.
18. Gordon, R. Effective Systems of Scrap Utilization: Copper, Aluminum,
and Nickel. U.S. Department of the Interior, Bureau of Mines. 1972.
19. Higgins, C. Econometric Description of the U.S. Steel Industry, Uni-
versity Microfilms, Ann Arbor, Michigan, 1968.
20. Institute of Scrap Iron and Steel. Facts (Yearbook). Washington, D.C.
21. Johnson, W. R. The Supply and Demand for Scrap Automobiles. Western
Economic Journal IX (4) December, 1971.
22. U.S. Department of the Interior, Bureau of Mines. Minerals Yearbook
1972, Volume I. Washington, D.C.
23. McClenahan, W.S. Consumption of Paper Stock by United States Mills in
1969 and 1970. From TAPPI, Volume 55, No. 11, November, 1972,
24. Paones, James. Mineral Facts and Problems (Bulletin 650). U.S. De-
partment of the Interior, Bureau of Mines. Washington, D.C., USGPO,
1970.
25. President's 1963 Tax Message.
26. Regan, W.J.; R.W. James; and T.J. McLeer. Identification of Opportuni-
ties for Increased Recycling of Ferrous Solid Waste. Institute of
Scrap Iron and Steel, Inc., Washington, D.C. U.S. Department of Com-
merce, NTIS PB-212 729, 1972.
27. Russell, C.C. and W. J. Vaughan. Environmental Quality Problems in
Iron and Steel Production. Resources for the Future, Inc., Washington
D.C., Draft, August, 1974. '
28. U.S. Department of Commerce, Census of Mining Industries: Lead and
Zinc (preliminary report). Washington, D.C., USGPO, 1975.
29. Vaughan, William R., and Clifford S. Russell. Environmental (Duality
36
-------
Problems in Iron and Steel Production. Resources for the Future, Inc.,
Washington, B.C., 1974. (preliminary draft).
37
-------
GLOSSARY
ELASTICITY. In general, elasticity indicates the percentage change in one
variable which results from a one percent change in the value of another
variable. Many different types of elasticities occur so frequently in
economics as to have their own names. Those used in the text are sum-
marized below.
Price elasticity of demand: the percentage change in the quantity of a
demanded resulting from a one percent increase in its price.
Cross price elasticity of demand: the percentage change in the quantity
of a good demanded resulting from a one percent increase in the
price of a substitute good.
Elasticity of substitution: the percentage change in the ratio of
inputs (e.g. capital to labor) resulting from a one percent change
in the ratio of their relative prices (price of capital to wages).
SCRAP DEFINITIONS.
Home scrap: the name given to secondary materials generated in the
production processes and then reused at the same location or within
the plants of the same company, ususally in the same production
process.
Prompt scrap: the category of materials which originate as the by
products of one production process and then shipped off to other
plants to be used as inputs to other production processes.
Obsolete scrap: the category of materials consisting of discarded in-
dustrial equipment and consumer goods which are no longer in use.
It is also referred to as old scrap (as distinguished from new
scrap which also includes home and prompt).
TAX PROVISIONS.
Percentage depletion allowance: a provision available to the mineral
industries which allows exemption of a certain percentage of gross
revenues from income taxes.
Severance taxes: those levied by states on mining industries, usually
as a percentage of the value of output or quantity of output.
38
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
REPORT NO.'
EPA-600/8-77-012
3. RECIPIENT'S ACCESSION-NO.
. TITLE AND SUBTITLE
Inpact Of The Federal Tax Code On Resource Recovery:
A Condensation
5. REPORT DATE
August 1977 (Issuing Date)
6. PERFORMING ORGANIZATION CODE
AUTHOR(S)
"Barbara J. .Stevens
8. PERFORMING ORGANIZATION REPORT NO.
PERFORMING ORG \NIZATION NAME AND ADDRESS
Columbia University
Graduate School of Business
fiew York, I".I. 1002?
10. PROGRAM ELEMENT NO.
1D1312
11. CONTRACT/GRANT NO.
Grant Fo. 3303362
12. SPONSORING AGENCY NAME AND ADDRESS
Municipal Environmental Research Laboratory-- Gin. ,OH
Office of Research and Development
I .3. E-'-ironmenta] Protection
------- |