January 1977
FORECASTING THE COMPOSITION
AND WEIGHT OF HOUSEHOLD
SOLID WASTES--An Executive Summary
JVIRONMENTAL 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
categories were established to facilitate further development and application
of environmental technology. Elimination of traditional grouping was con-
sciously 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
Information Service, Springfield, Virginia 22161.
-------
EPA-600/8-77-002
January 1977
FORECASTING THE COMPOSITION AND WEIGHT OF HOUSEHOLD SOLID
WASTES USING INPUT-OUTPUT TECHNIQUES
An Executive Summary
by
David Kidder
Ebon Research Systems
Silver Spring, Maryland 20901
EPA Contract No. 68-03-0261
Project Officer
Oscar W. 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 publica-
tion. Approval does not signify that the contents necessarily reflect the
views and policies of the U.S. Environmental Protection Agency, nor does men-
tion of trade names or commercial products constitute endorsement or recom-
mendation 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 solution
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, treatment, 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 publication is one of the products of
that research; a most vital contnunication link between the researcher of the
user community.
A new approach is explored for determining not only the current status
but also the probable effects of new government policies or other significant
economic developments on the quantity and composition of household solid waste
and the implications for resource recovery. The potential of the input-output
model for forecasting is discussed and results are compared with several other
studies. It is hoped that the methodology and results of this study will pro-
vide planners and policymakers with additional information for efficient
management of the increasing quantities of household solid waste.
Francis T. Mayo, Director
Municipal Environmental
Research Laboratory
111
-------
ABSTRACT
After a critique of previous methods for assessing household solid waste
generation, an improved input-output model based on transactions among indus-
tries and other sectors of the economy is presented. The various adjustments
and assumptions necessary in this model are explained along with its basic
concept of "path products" for long-term estimation of household solid waste.
The model is tested with industrial production data from earlier years and
projects the household waste producing inputs for 1985.
The integration of this method of manipulating industrial production
data with the INFORM model of economic growth is shown to be a module avail~
able for the Strategic Environmental Assessment System of the Environmental
Protection Agency that would have value for planning for resource recovery
efforts and management of household solid waste.
IV
-------
TABLE OF CONTENTS
Disclaimer ii
Foreword iii
Abstract iv
Acknowledgements vi
1. Introduction 1
2. Rationale for the Study 2
3. The Nature of Input-Output Analysis 2
4. Methodology 3
Minimal Adjustments 3
Further Adjustments 4
The Production Forecasting Model 5
5. Findings 7
6. Strengths of the System 8
7. Limitations of the Model 8
8. Possible Uses 10
Figures
1. Comparisons of IR&T Estimates of Household and Commercial Waste
(1971) with NCRR and EPA Studies 11
2. Waste Producing Inputs into the Household Sector:
Estimates (1971), Forecasts (1985) 12
v
-------
ACKNOWLEDGEMEINTS
The research reported in this publication was taken from a more exten-
sive report entitled, Forecasting the Composition and Weight of Household
Solid Wastes Using Input-Output Techniques, prepared by International Re-
search and Technology Corporation (IR&T) of Arlington, Virginia (EPA Contract
No. 68-03-0261, Stedman B. Nobel, project manager). The full report is
available in two volumes from the National Technical Information Service
(NTIS), U.S. Department of Commerce. Volume I, PB 257 499/AS; Vol II
(appendix), PB 257 500/AS.
Dr. David Kidder, assisted by Guy Hudgins, prepared this executive sum-
mary of the original report. Ebon Research Systems wishes to express thanks
for the support of the EPA Project Officer, Oscar Albrecht of the Municipal
Environmental Research Laboratory, Environmental Research Center, Cincinnati,
Ohio.
VI
-------
INTRODUCTICN
Effective planning for solid waste management and resource recovery has
been hampered by serious problems in prior models for predicting household
waste. Methodological limitations in past studies produced inadequate data
for long-term forecasting of waste levels. This particular weakness pre-
sented a major problem since accurate quantitative predictions of specific
materials is essential for useful planning. Furthermore, earlier studies
neglected to incorporate methods for controlling such variables as income
changes over time, family size, new packaging technologies and other phenom-
ena in calculating long-range solid waste estimates.
The economic model described in this study allows the user to make more
realistic predictions about the quantity and composition of household solid
waste than was previously available. In addition to showing the levels of
future household purchases and the quantities of particular materials em-
bodied in the products discarded by households, the report introduces a
methodology for accounting for other relevant variables in the model.
By applying the techniques of input-output analysis to the study of
solid waste generation by households, the study used data collected in 1958
to 1964 to prepare estimates of waste by product group for 1971. Forecasts
of solid waste generation for 1985 were made using the input-output type
coefficients and the production forecasts from the INFQRUM model of sectoral
growth.
The IRSeT estimates for 1971 showed that paper and metals were the com-
ponents of household and commercial wastes, contributing 41.0 and 37.5 mil-
lion metric tons respectively. Glass contributed 12.5 million, and the com-
bination of plastics, textiles, wood and rubber contributed approximately 18
million metric tons. [p.83 of IR&T Report]
The forecast of growth in consumer expenditures and resulting waste be-
tween 1971 and 1985 was considerable for most product categories included in
the study. For some product categories, the forecasts were subject to pre-
diction errors resulting from the aggregation of sectoral data and weaknesses
in productivity forecasts.
-------
RATIONALE FDR THE STUDY
This study tested the feasibility of using input-output analysis among
numerous industries to estimate and project the future volume of household
wastes. Moreover, it compared input-output to other methods for estimation
and projection of waste volume. The quantity and composition of household
solid waste were forecasted to 1985 using (1) a model for the economy de-
scribed in terms of materials embodied in products purchased by households as
a baseline, and (2) INFORUM model's extrapolations of inter-industry coeffi-
cients. If the input-output system documented by this study proves useful,
it can provide a "module" in the Strategic Environmental Assessment System
[SEAS] of EPA that would forecast solid waste production in the future.
THE NATURE OF INPUT-OUTPUT ANALYSIS
Input-output analysis looks at the inter-relationships among (1) indus-
tries as producers, and (2) users of raw materials and semi-finished goods.
It also shows inter-relationships among industries and users — households,
capital goods buyers, governments and foreign buyers — of final products.
The structure of inter-relationships among sectors is defined in a variety of
ways:
1. the dollar values of goods flowing from INDUSTRY A to INDUSTRY B or
from INDUSTRY A to FINAL USE;
2. the ratios of INDUSTRY A's product used to make a dollar's worth of
B's product. These are the "technical coefficients" that, along with
other industry input coefficients, comprise the components of INDUS-
TRY B's production technology and
3. the multiplier values that show the total direct and indirect impact
of a specific dollar value of final use (or a change in this value)
on production volumes. These values result from an "inversion" of
any array of technical coefficients that describe production tech-
nologies for an entire economy.
In order to illustrate input-output analysis, consider machine tools as
they relate to production. Machine tools are a component part of the manu-
facturing process in many industries. Along the "row" of machine tool tech-
nical coefficients, one usually finds high coefficients between the machine
tool and other industries, and a low coefficient relating machine tools to
household use. The direct effect of an increase in final demand by
households on machine tools would depend on the size of the coefficient:
"machine tools — household"
total machine tools production
But the total effective change in machine tool production would depend on all
the relationships specified as occurring among industries. For example,
-------
automobiles require more machine tools, according to the coefficient:
"machine tools — autos" .
total machine tools
Machine tools use steel; therefore, higher production levels in machine tools
will lead to more orders and increased production in the steel industry.
Other "ripple effects" that create business for machine tools add to a multi-
plied indirect effect on production. The final total effect of a dollar's
increase in household demand on machine tool production will include the dol-
lar value of additional tools bought by households and the total increase in
machine tool sales to other industires.
A complete set of technological coefficients and associated multiplier
values similar to those included in this study can provide the user with a
tool that satisfies two goals: (1) it shows how projected increases in final
demand expand production requirements in the related industries, and (2) it
gives a structure for analysis that is adaptable to adjustment and change if
necessary.
Although an input-output table identifies money flows among sectors, by
a series of adjustments it is possibe to convert the model for use in esti-
mating flows of physical product weights and the technical coefficients that
reflect these flows. In this form, waste volume estimation for industries,
households and the commercial sector becomes feasible.
METHODOLOGY
This study used a "minimal adjustment" approach to convert the basic
input-output system from a description of market value relationships to a
description of flows of product by weight among sectors. Further adjust-
ments, beyond the minimum, were subsequently made to increase estimating
accuracy and realism in the model's structure.
MINIMAL ADJUSTMENTS
The analysis identified 16 two-digit industries (classified by the
Department of Commerce's SIC codes) plus two one-digit industries (wool and
cotton) as major producers of outputs that would become household waste.
"Final use" waste included waste from the commercial sector, wholesale and
retail trades. The primary goal of minimal adjustment was to show how much
poundage of each product group becomes embodied in products that households
buy and discard.
The input-output bench mark figures used in computations were from the
INFORUM model, developed at the University of Maryland, that included 182
Separately-identified industries. The model's technical coefficients were
adjusted to calculate physical flows in pounds or tons (all values were
divided by unit prices). Input flows not embodied in the weight of the
-------
products used by households and flows to users other than households or the
commercial sector were removed. For example, service inputs that have no
effect on the disposal of a final product that used them were excluded.
From the adjusted technical coefficients, multiplier values were com-
puted by summing (for each industry) all of the possible direct and indirect
ways the industry's products could be embodied in the weight of the products
used by households. This approach was referred to as computation by path
products. Utilizing this technique, various paths by which a product could
enter final consumption were specified. Fiber, for instance, could enter
final consumption by the path as input embodied in clothing or by the path as
input in automobile upholstery.
Using the truncation principle, the study computed 1700 significant
paths. However, it was found that 214 paths accounted for approximately 75%
of all output finally purchased by households. These 214 major paths were
subjected to analysis for proration errors, and it was found that about 85
had no substantial error from this source. The major proration errors en-
countered were in plastics, fibers and non-ferrous metals.
The study stressed, throughout the analysis of errors, that thoroughly
accurate corrections were not currently possible, given the form in which
industry production and sales data were collected and presented. It was
important to attempt corrections that could be done with some precision and
to stress the type and direction of bias to be expected when quantitative
changes proved impossible.
FURTHER ADJUSTMENTS
The minimally-adjusted model previously described provided an appropri-
ate basis for the analysis, but contained a number of weaknesses that were
either corrected or noted in the report. Whenever possible, the necessary
corrections were made before arriving at final estimates in the study.
Many of the problems that arise in using input-output analysis for any
purpose result from excessive aggregation of industry data. When describing
waste generation, this is particularly true. For instance, at the two-digit
level an input from one industry enters production of another in fixed pro-
portion. In actuality, the purchasing industry may make a variety of prod-
ucts which use different proportions of the input it purchases. Consequent-
ly, fixed proration may cause inaccuracy in forecasting, expecially if the
goods sold by the input-purchasing industry should change. Metal cans pro-
vide an example of this situation. Cans used for some purposes include a high
proportion of aluminum, but for other uses different metals are preferred.
Furthermore, different users of an input may not pay the same unit prices.
Again, because data relative to input are too gross (e.g. "steel" rather than
steel by specific grade), an adjustment that uses a single unit price of
steel to convert money values to physical values will lead to misrepresenta-
tions. (A full adjustment incorporates information about price differentials
when they are available.)
Assumptions about the manner in which a product is wasted may also lead
4
-------
to inaccuracies in the model. For instance, a car battery may be saved even
though the owner scraps the car. The same kind of problem occurs in the in-
dustrial sector when some inputs sold are embodied in final product weight
and some are not embodied in the final product weight. Two-digit aggregation
forces the user to discard these industries or to use them and accept the
resulting lack of accuracy. Informed judgment is required to adjust for
errors resulting from misleading representations by the model regarding the
way final products are wasted.
In developing final estimates, this study recognized the existence of
such aggregation problems and attempted to make adjustments where data per-
mitted the correction. Some problems, such as the proration of highly spe-
cialized inputs, were intractable (given the way current data are gathered).
Plastics, fibers and metals estimates appeared to suffer most from the as-
sumption of rigid proration.
THE PRODUCTION FORECASTING MODEL
To generate forecasts of solid waste weight and composition, IR&T con-
sidered a number of economic models constructed by governmental agencies and
private firms. It was decided that INFORUM and the Inter-Agency Growth Model
used by the Federal Government presented the details needed for analysis of
household waste. The following were offered as the advantages of the INFORUM
model:
1. INFORUM was part of the EPA's long-range forecasting tool, the SEAS;
2. INFORUM was free to projects sponsored or supported by EPA;
3. INFORUM forecasted 133 separate consumption categories compared with
82 by the Inter-Agency Growth Model. In addition, INFORUM's input-
output table details 182 sectors compared with 129 for the Inter-
Agency Model;
4. the interdependence of income, desire to work and spend, and produc-
tivity was spelled out in the INFORUM model;
5. the INFORUM model has been "purified" to eliminate inconsistencies
caused by aggregation of sectors;
6. since the assumptions INFORUM1s builders used were explicit, it
would be possible for another user to alter the assumptions and check
the sensitivity of predictions for a variety of ways of looking at
the world and
7. INFORUM made projections of input-output coefficients over time.
Some large coefficients in INFORUM were projected individually. The
coefficients used to describe the use of some metal in auto production is
representative of an individual projection. Other projections were made
"across the row". An example of the latter projection is that it seemed
generally true that less cotton would be used in production, and therefore
the cotton input coefficients for all industries were reduced by a constant
proportion that generally extrapolated a past trend. The extrapolations
permit an increase or decrease in the rate of change, but no trend reversal.
The weakness of this approach is apparent when one considers that in the
long-run material substitutions do occur and consumption patterns also
change.
-------
INFORUM1s basic data on personal consumption expenditures came from
Census of Manufactures' yearly figures from 1958 to 1971. Data from 1958 to
1964 were used to estimate statistically how consumption should relate to
other "explaining" factors.
The assumed relationships were basically of two types. In one instance,
changes in consumption were related to earlier changes in income, assuming
that it takes some time to adjust spending patterns to new living standards.
The second type of relationship added the assumption that installment buying
could break down the income-expenditure link and cause some over-reaction in
spending to changes in income. These relationships, based on time series
spending-income data, were checked against cross-section budget data that
looked at how families at different income levels made expenditures. In the
final analysis, predictions for specific spending categories were based on
(1) information derived from time series and cross-section data and (2) from
a careful reading of "what works" in a particular forecasting exercise.
The study attempted to take into account productivity forecasts using
techniques developed by the U.S. Department of Labor, Bureau of Labor
Statistics in The Structure of the U.S. Economy in 1980 and 1985 (1975).
This publication was based on the Inter-Agency Growth Model projections. The
underlying rationale for productivity estimates was the supply of workers
available to the economy over the relevant time period. Steps used by the
Labor Department included the following:
1. estimation of base line numbers of U.S. Citizens of working age and
Census projections of populations to 1985;
2. estimation of base line labor force participation rates, for the
appropriate age categories. Projections of these rates were more
difficult to obtain than projections of the working age population
because participation habits have changed for both men and women;
3. adjustments for normal unemployment. Recent years have demonstrated
that a "normal" level of unemployment, consistent with a reasonable
level of mobility among jobs and an acceptable break-in period for
inexperienced new job seekers (e.g. teenagers) is difficult to de-
fine and
4. projection of average wages. What the employed worker in 1985 will
have available for spending depends, in part, on productivity since
incomes that the economy makes available derive basically from pro-
ductive activity. Productivity forecasting is a difficult art, how-
ever, the long-run trend of 3-4% a year has been used often as a
bench mark. INFORUM projections tended to be pessimistic about pro-
ductivity growth and, by implication, slower growth in the income-
spending-waste generation system in the near future. This was based
on the INFORUM assumption that two well-established trends would
continue:
(1) more spending for services rather than consumer and producer
durables. Productivity was difficult to measure in services
but all indications suggest that productivity will grow
slowly in this very labor-intensive sector and
(2) labor productivity increased at a decreasing rate in all
-------
sectors in the 1970's after a rapid increase in the 1960's.
FINDINGS
The IR&T study provided estimates of 1971 solid waste and forecasts of
1985 waste. In addition, "path products" that related materials to house-
hold waste were computed. The forecasts by INFORM and the Inter-Agency
Growth Model were used to make projections of the weight and composition of
1985 waste.
Figure 1 compares 1971 estimates of household and commercial waste with
estimates resulting from two other studies. In order to make this comparison
possible, some of the industry-level detail in the IR&T study had to be sac-
rificed. There was considerable agreement among the three studies on waste
estimates by industry, except for a sizeable difference in the "metals" sec-
tor. IR&T's estimate of 37.5 million metric tons of scrap metal was consid-
erably higher than the 8.1 and 10.8 million metric tons shown by the other
studies. Paper scrap was estimated at 41 million metric tons by IR&T,
compared with 34.6 and 35.5 by NCRR and EPA. There was general agreement on
glass, plastic, textiles and rubber that contributed less to waste, but some
divergence on wood (IR&T's 9.9 million metric tons, compared with EPA's
4.2).
Figure 2 shows waste producing input estimates for 1971 and forecasts
for 1985. The estimates of "waste" as defined took into account the fact
that time from purchase to discard differed among products. Thus, not all
inputs in 1985 would become waste in that year.
The most rapid proportional growth in inputs was projected for plastics
(1.7 to 5.2 million metric tons), glass (9.4 to 19.2 million metric tons),
and other non-ferrous metals (0.2 to 0.4 million metric tons). Metals and
paper are expected to remain the largest contributors to solid waste (35.5
million metric tons for all metals and 47.3 million metric tons for paper in
1985).
The study incorporated INFORUM's forecasts of consumer expenditure from
1971 to 1985. Many of the expenditure categories in the INFORUM list were
not relevant to the IR&T study since they represented consumer services that
do not generate household solid waste. However, some of the expenditure cat-
egories where increases were projected comprised durables made largely of
metal; consequently, the projections had a direct impact upon the generation
of metal in household waste. In examining the small numbers of categories
that grew rapidly and had a direct effect upon the generation of household
waste, it was found that most of the categories increased at a similar rate.
In comparing specific consumer expenditure increases from 1971 to 1985 among
the relevant categories that grew most rapidly, aircrafts led the list with a
7.5 proportion of increase, followed by communication equipment with 3.6 and
guns with a 3.4 proportion of increase, [p.136 of the IR&T Report]
-------
STKENGTHS OF THE SYSTEM
The strength of input-output analysis comes from the clarity with which
inter-relationships among producers and users can be presented. The model
provides greater industrial detail than other models. In addition, since
technical coefficients are delineated, it is possible to test the sensitivity
of waste production results with assumed changes in production technology.
The study noted the difficulties involved in criticizing or correcting the
techniques used in other studies:
1. composition studies of waste that suffer from choice of "atypical"
time and location for sampling;
2. adjustments of production estimates that measure waste as "input"
used by households. Though broadly similar in approach, most pro-
duction-based studies failed to account for the full direct and in-
direct production impact of changes in the level and pattern of
final demand and
3. adjustments of consumer spending estimates that treat waste as an
"output" of the household sector. These share the lack of detail and
clear structure common to other techniques that do not use input-
output methodology.
LIMITATIONS OF THE MODEL
Errors in model forecasting can arise from the following two reasons:
(1) the model may not have been properly built (relating consumption to in-
come and even price may fail to capture the variety of social and personal
factors that can explain why people buy more or less of a particular pro-
duct) and (2) the information imbedded in the data may have been culturally
or historically specific. In a sense, this point reflects poor model spec-
ification, but the limitations of time series and cross-section data are in-
herent in every statistical experiment in which the entire economy is the
laboratory. Time series data portray behavior over time of people with vastly
different social and economic characteristics. Unique historical forces
affect this behavior, and to the degree that these forces cannot be
"controlled", the relations estimated for one period in history will not
forecast accurately the results for a succeeding time period. Cross-section
data, on the other hand, "freeze" history and examine the income-spending
link at a particular point in time. Moreover, incomes vary with age, and
different income classes have different historical experiences that condition
their behaviors. Consequently, these may not easily translate into accurate
forecasts for various income classes in the future.
Weaknesses of input-output analysis come from (1) the nature of the
model itself and (2) the gaps in the data used to construct the model. An
input-output table is, basically, a measure of value flows. The adjustments
needed to convert INFORM or any other existing table to the requirements of
solid waste estimation are, at best, crude efforts to adapt a system to a use
8
-------
for which it was not intended. The technical coefficients of an input-output
system are necessarily fixed at any one point in time. It is, however, pos-
sible to project changes in these coefficients over time to improve the accu-
racy of production forecasts. It is not feasible, however, to build in an
adjustment for one-time changes brought about by decisions to substitute
among existing methods or materials. In practice, both the forecast and
explanation of specific technology choices in an input-output framework (and
elsewhere) rely heavily on intuition.
Current input-output models use data gathered by the Department of Com-
merce at a fairly high level of aggregation. For some uses, two-digit indus-
trial data leave many important flows of waste-producing activity submerged
and require laborious industry-by-industry adjustment of input-output coef-
ficients to achieve usable results.
Forecasting waste volumes accurately depends not only on the availabil-
ity of complete and accurate input-output coefficients, but also on forecasts
made outside the system that must be used. Contemporary predictions of per-
sonal consumption expenditure, whether done by simple trend extrapolation or
by complex statistical models that relate several factors to consumption,
have not performed well over long periods for disaggregated types of expendi-
tures. Thus, major improvements in designing an input-output type system
will contribute little to planning if components of final demand cannot be
accurately forecast. Since personal consumption spending depends on income
and many other economic and non-market factors, the accuracy of consumption
forecasting depends on what is assumed will happen to these factors in the
future. A crucial determinant of consumer's income is the level of produc-
tivity for the economic system. If productivity rises, the income that
society's resources can produce at full employment will also rise.
Productivity forecasting is, however, a difficult job. It is not always
possible to measure accurately the "state-of-the-art" levels of productivity
in particular industries, because productivity indexes move as the rate of
capacity utilization changes. Thus, a system coming out of a recession usu-
ally registers increased labor productivity because output in the initial
recovery increases faster than the employed workforce. This growth in
productivity will slow down as additional workers are employed to fill new
orders and as longer hours and multiple shifts are instituted. These meas-
ures of productivity changes do not adequately reflect full-capacity resource
use or the technical efficiency of productive resources for projecting
secular income changes.
Bench mark productivity data are hard to find because cyclical changes
distort the figures; and limitations in technological forecasting seriously
restrict long-range productivity forecasts. These limitations are clearly
valid for long-range projections of waste coefficients as well. Although
technology has been applied consistently in the past to lowering production
costs and altering the use characteristics of final products, it is reason-
able to expect that in the future, as disposal costs rise, innovations will
work toward reducing the volume and expense of waste.
-------
POSSIBLE USES
This study can be used by policymakers and planners to evaluate not
only the likely effects of a changing technology on future volumes of waste
and the costs of disposal, but also the implications of governmental policies
designed to control waste disposal. The current model offers promise for
policy-makers with a national perspective. With modification and some sup-
plementary data inputs, the approach used in this study can be applied to
regional solid waste disposal problems as well.
Researchers and planners can utilize the computations in this study as a
basis for improving estimates and forecasts of household solid wastes or for
constructing estimates of waste volume for other sectors. How a potential
user will apply the technique depends on what questions are to be answereed
and what data resources are available. The minimal adjustment method, de-
scribed as a crude modification of the input-output technique for use in
studying waste production, is perhaps the least costly approach. But because
of the many problems discussed in this report, it is also the least satis-
factory, if the user chooses to make the full list of adjustments, retaining
the bench mark input-output coefficients computed by the minimal adjustment
technique, the cost of analysis will increase. Presumably, the accuracy of
waste volume estimates will also increase.
An ambitious re-working of all the coefficients for a limited number of
industries might be the best way to produce a system oriented toward the
needs of waste volume analysis. Direct collection of volume data, or at
least greater disaggregation of value data, would improve the structural
realism of the system.
This study should be viewed not as an end in itself, but as a test
application of input-output techniques for forecasting household solid
wastes. The major barrier to full use of input-output in this area is not
methodological, as both the theory and technique of inter-industry analysis
are fairly well understood. Instead, lack of detailed data on production
volume and poor consumption forecasts limit its ffective use. If it is de-
termined that detailed industry-level production data are well worth their
additional costs, the input-output method described in this study provides a
powerful tool for analyzing and forecasting the quantity and composition of
household solid wastes.
10
-------
NCRR
EPA
IR&T
I
Paper
Glass
Metals
Plastic
Textiles
Wood'
Rubber"
Source: Adapted from IR&T Final Report; Table 4.4, P. 83.
•No estimate for NCRR
"EPA estimate combines rubber and leather
Comparison studies:
— "Municipal Solid Waste" NCRR Bulletin, Vol. Ill, No. 2, Spring, 1973.
—"Estimates of Household and Commercial Solid Wastes Based on Production Statistics." Draft Report. Resource Recovery
Division, Smith, Fred L., Jr. Office of Solid Waste Management Programs, U.S.E.P.A.
Figure 1. Comparisons of IR&T Estimates of Household and Commercial
Waste (1971) with NCRR and EPA Studies
-------
500
1971
Estimate
1985
Forecast
I
SIC Industries
41 — Lumber and Wood Products
48 — Paper and paperboard mills
50 _ Wall and building paper
62 — Plastic materials
63 — Synthetic rubber
78 — Glass
83 — Steel
84 ^- Copper
85 — Lead
86 — Zinc
87 — Aluminum
88 — Other primary non-ferrous metals
4. 64. 65 — Fibers
Source: Adapted From IR&T Final Report,
Table 5.7, P 164
Figure 2: Waste Producing Inputs into the Household Sector:
Estimates (1971), Forecasts (1985)
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
T. REPORT NO.
EPA-600/8-77-OQ2
2.
3. RECIPIENT'S ACCESSION-NO.
4. TITLE AND SUBTITLE
FORECASTING THE COMPOSITION AND WEIGHT OF HOUSEHOLD
SOLID WASTES USING INPUT-OUTPUT TECHNIQUES
An Executive Summary
5. REPORT DATE
January 1977 (Issuing Date)
6. PERFORMING ORGANIZATION CODE
7. AUTHQR(S)
David Kidder
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Ebon Research Systems
10108 Quinby Street
Silver Spring, Maryland 20901
10. PROGRAM ELEMENT NO.
1DB314
11. CONTRACT/EJWNX NO.
68-03-0261
12. SPONSORING AGENCY NAME AND ADDRESS
Municipal Environmental Research Laboratory - Gin., OH
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, Ohio 45268
13. TYPE OF REPORT AND PERIOD COVERED
Executive Summary
14. SPONSORING AGENCY CODE
EPA/600/14
15. SUPPLEMENTARY NOTES
Full report is available from the National Technical Information Service. Vol. I is
EPA-600/3-76-071a, PB 257 499/AS; Vol. II (Appendix) is EPA . . . 071b, PB 257 500/AS.
16. ABSTRACT
An input-output model based on transactions among industries and other sectors of the
economy is presented for assessing household solid waste composition and quantity.
The various adjustments and assumptions to exercise the model are explained along
with its basic concept of "path products" for estimation of household solid waste.
The model is tested with industry production data and projects the household waste
producing inputs for 1985.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS
c. COS AT I Field/Group
Decision making
Waste disposal
Materials estimates
Public administration
Economic forecasting
Economic models
Forecasting
Household solid waste
Solid waste
Input-output
Modeling
Predicting
5A
5C
8. DISTRIBUTION STATEMENT
RELEASE TO PUBLIC
19. SECURITY CLASS (ThisReport)
UNCLASSIFIED
21. NO. OF PAGES
19
20. SECURITY CLASS (Thispage)
UNCLASSIFIED
22. PRICE
EPA Form 2220-1 (9-73)
13
S. GOVERNMENT PRINTING OFFICE: 1977-757-056/5563 Region No. 5-11
------- |