XvX
     Cost of
 Air and Water
Pollution Control
   1976-1985
Section One
Agency Draft
February 1976

-------
    This review draft of the proposed report, "Cost of Air and Water Pollution

                                  I.
Control 1976-1985," is a xerox copy produced from the WORD ONE automated word


processing system.  This system uses symbol and letter combinations to code


different type fonts for the photocomposition process that will ultimately pro-


duce the final report copy.  These symbol/letter codes, i.e., $a, $b, $c, $d,


$e, $s, and $t, appear in this interim copy at the beginning and ending of


chapter heads, paragraph titles, and tables.  These codes will, of course, be


removed when the final production process is accomplished.  The following


sample page has been included to serve as an example of the final report's for-


mat.

-------
                                              Section  Three

                     THE  ECONOMICS  OF  WATER  POLLUTION  CONTROL
Chapter  1
Summary
The economics of controlling water pollution encompasses
both the expected benefits ai:d the probable costs of control.
The  principal  findings in  the  control  cost  area  are
summarized below; benefits of water  pollution control are
discussed in the next chapter of this section.
       5.    INDUSTRIAL CONTROL  COSTS

Introduction

The extent of water pollution and the costs of treating it vary
significantly among industries and among the firms within
an industry; therefore,  it is  important  to  examine the
structure, production methods, sources of pollution, effluent
standards,  and wastewater control  technology for  each
industry.  The following  sections of this  chapter briefly
summarize the relevant characteristics of each industry and
report the estimated annual abatement costs attributable to
achieving full compliance  with  the  1977 (BPT) and  1983
(BAT) effluent standards.
MODELING AN INDUSTRY

The plants in an industry have various options by which to
comply with the water pollution standards. In general they
may:

  I.  Fully treat their effluents.

  2.  Pretreat their effluents and discharge to a municipal
     system.
                                                                             Table 24.
                                                               Pollution Control Costs as a 1'iTcentane
                                                                      of Alinon Scenario d'NP

Air Stationary
Source Costs
Capital Costs
O&M Costs
Water Industrial
Costs
Capital Costs
O&M Costs
Water Municipal
Costs
Capital Costs
O&M Costs
1975


0.60-3}
0.42%


0.26%
0.02 %

0.36%
0.03%
1977


0.47%
0.37%


0.36%,
0.02%

0.39%
0.05%
I'W)


0.30%
0.43%.


0.52%
0.23%

0.36%.
0. 1 2%
1983


0. IX^
0.3K9i


0.48%
0.27%

0.07%,
0.11%
1985


0. 1 1 '-,
0.34%


0.0d%
0.52%

0.02%
0. 1 1 %
                                                       ECONOMIC  FORECASTING  IN THE
                                                       SEAS SYSTEM

                                                       Economic forecasting within the SEAS system takes place in
                                                       a context which  meshes the features ofa traditional input-
                                                       output (I/O) model and econometric model-building. The
                                                       basic  structural   economic   component  ( of  SEAS,
                                                       INFORUM', is not a typical economic input-output model.
                                                       It is, to a substantial degree,  a combined econometric-I/O
                                                       approach to economic forecasting.
                                                                               N
                                                                                   1
3.  Change their manufacturing process.                                        (1
                                                    Table 10.
                                General Projections of the Reference Scenarios (SI),
                                                    1975-1985

Statistics
Population (Millions)
l.alxir Force (Millions)
Unemployment K.ile (% )
Disposable Income
Per Capita (l'"75 S)
Gross National Prixluc'l
(Trillion 1175 S)
Personal Consumption
Kxpcmlilinc
Investment
Ciovcnnncnl l:.\peiulllures
I'Vilcral •
Value
|<)75
:i.i.»
93. S
S.4

3.553

t.470

-133
222
.307
.117
In
19X5
235.7
107.7
4.4

5.753

2.365

1.563
.3X0
.414
.136

1975-85
0.97
1.40
-6.38

4.94

4.87

5.30
5.59
3 05
1.61
Annu.il
1 ''75-77
0.89
1.71
-X.07

5.16

6.43

4.87
13.69
2.99
.71
Percentage
t'177-80
0.97
1.62
-I0.6d

7.96

6.50

7.55
6.43
3.74
.70
Change
I9SO-83
1.02
1.25
-3.04

3.29

3.34

4.36
1.85
2.68
2.58

19X3-85
1.00
0.96
-2.94

2.74

3.18

3.79
2.31
2.64
2.43

-------
                             Table oi. Conl.enLs



Section One-OVERVIEW

Chapter                                                      .     Page

             Acknowledgements

             Executive Summary

  1          Introduction                                         1-1

  2          The Benefits of Pollution Control Programs           1-35

  3          Pollution Control Cost Reduction                     1-63
             Through Process Change

-------
                    SbACKOWLEDGEMENTSSR



Preparation of this combined air and water pollution control

cost analysis was an extensive effort made possible only

through coordination of the hard work of many different

dedicated Individuals.  The final product Is the result of

expert analysts In EPA and the private sector.   EPA

personnel and contractor personnel responsible for various

aspects of the report are listed below:
ScOVERALL REPORT MANAGEMENT.
INTEGRATION. AND REV1EWSR

  EPA:  Peter House.  Roger Don Shul1

  Control Data Corporation:  Rafael  Ubtco. Michael  Kranlas,

     Cheryl Herrln.  Thomas Germach,  Bradford W1ng.

     Consultants:  Matthew Barrett (Analytic Products Inc.).

     Lyman Clark (CONSAD Corporation).  Jeffrey Krischer

     (Johns Hopkins University)
ScWATER POLLUTION CONTROL
COST ANALYSIS$R

  EPA:  Donald H. Lewis.  Richard K.  Schaefer

  CONSAD Corporation:   Donald McCartney.  Samuel  Hadeed.

     Forrest Arnold, William Carlson,  Consultants:   Richard
                            1-1

-------
     Burrows (AWARE. Inc.). Ralph Lueken (Private

     Consultant. Washington,  O.C.)

  Vanderbtlt University:  Andrew Edwards
$CAIR POLLUTION CONTROL
COST ANALYSISSR

  EPA:   Philip D. Patterson,  WfHand Smith

  Battelle Columbus Laboratories:  Philip R. Beltz. Gabor

     Kovacs.  Ted Thomas
ScCOMPREHENSIVE ECONOMIC
ANALYSISSR

  EPA:  Peter House. Edward Williams.  Philip Patterson.

     Samuel Rattck.  Richard K. Schaefer, Richard H. Ball

  Control Data Corporation:  Rafael Ublco.  Cheryl Herrin,

     Kenneth Thompson,  B.  Scott Miller

  International Research and Technology Corporation: Sted

     Noble, Marc Narkus-Kramer. Richard Meyer. Martin Stern

  CONSAD Corporation:  Wilbur Steger,  Ronald Adonolfl. Naztr

  Dozzanl
                            1-2

-------
$CPROCESS CHANGE ANALYSISSR



  EPA:  Michael Hay

  International Research and Technology Corporation:  James

     Saxton, Richard Meyer, Thomas Jones. Robert Cape 11
ScPOLLUTION CONTROL BENEFIT
ANALYSISSR

  EPA:  Fred Abel. Thomas Waddel1.  Dennis P. Tlhansky

  Envlro-Control.  Incorporated: Alex Hershaft, Theodore

     Helntz. Or.,  Gerald Horak
$CPOPULATION-AT-RISK STUDYSR



  EPA:  Fred Abel. Thomas Waddel1

  Envlro-Control, Incorporated: Steve Takacs, G. Bradford

     Shea
ScPRODUCTION COMPOSIT10NSR



  Control Data Corporation:  Cathy Blank, Donna Cloutler,

     Linda Luehrs, Donna Selby, Dav Davlsson
                            1-3

-------
                   SaExecutlve SummarySR
This Is the first Agency report to combine the national
economic Impact analyses required by the Clean Air Act (P.L.
91-604) and the Federal  water Pollution Control  Act (P.L.
92-500) Into a single.  Integrated study.  The report Is
designed to facilitate comparison of the projected Impacts
of the two laws on a common basis, as well as to provide an
estimate of their combined effect on the economy.   It is
presented In four sections:  Overview. The Economics of  Air
Pollution Control, The Economics of Water Pollution Control
and A Comprehensive Assessment of Pollution Control.
                  SbAGGREGATED ESTIMATESSR
The report Indicates that If the nation's economy continues
to recover and grow at official Federal forecast rates,  the
total cost to meet requirements of the laws will be about
$480 billion over the decade beginning this year and ending
In 19B5.  This amount represents the decade sum of annual
debt retirement payments, plus operation and maintenance
costs associated with a capital Investment In plant and
equipment of about $220 billion.  The $460 billion (1975

-------
dollars) comprises about two and one-half percent of the
gross national  product (GNF>) summed over the same decade.

Pollution control expenditures are estimated to have a net
positive effect on total national employment over the 1976-
1985 decade.   During the period 1976-1979. when pollution-
control construction and capital equipment demands are at
their highest,  the Increase In tctal employment Is estimated
to range between one and two percent per year.   In
subsequent years, the Job requirements related to pollution
control shift toward operation and maintenance activities,
and the Increase In total employment Is estimated to decline
to one-quarter of one percent by 1985.  Beginning In 1984,
with the economy assumed to be at full employment, growth  In
pollution control employment comes at the expense of labor
1n production activities.

Other Investigations have indicated that premature closings
of older Industrial facilities due to pollution control
requirements Is probable In some cases, causing temporary,
localized unemployment.   Forecasting of this complex
economic process, however,  was not within the scope of this
study.

In developing these projections, e national cost calculation
procedure aas employed vafaleh allows future cost estimates  to

-------
be adjusted to conform with different forecasts of national
economic growth,  as Indicated by GNP.  This computer-based
adjustment capability Is useful  because of the uncertainty
and frequent changes associated with GNP forecasts.   For
example, the report examines estimated air and water
pollution control costs for a more conservative growth
forecast (called the "Low Productivity Scenario")  In which
the decade GNP Is about 90 percent of the official
"Reference Scenario." Another scenario, referred to as the
"Energy Conservation Scenario."  provides for a GNP
equivalent to the Reference Scenario, but with a different
mix of products due to the Implementation of certain
feasible energy conservation practices.  Table 1 presents
some summary Indications of the Impact of these different
forecasts of the future.  Although air and water pollution
control costs change with the economic forecasts,  the decade
costs remain very close to two and one-half percent of the
ONP.

-------
                         Sellable 1.
                Summary of Scenario Results
$t
Scenario
Reference
Reference
Abatement
Reference
Abatement
with $7 B/yr
Municipal
Appropriation
Low Productivity
Low Productivity
Abatement
Energy
Conservation
Energy
Conservation
Abatement
Decade GNP
(Tri Dion 1975$)
20.
20.
20.
IB.
18.
20.
20.
007
162
195
596
849
062
201
Decade
Pol 1 ut Ion .
Control Cost
(X GNP)
0.0
2.42
2.60
0.0
2.53
0.0
2.41
1985 Energy
Consumpt Ion
(Quad. Btu's)
109
112
113
102
105
95
99
.0
.9
.2
.0
.7
.7
.6
SO
100.0
37.9
38.0
90.3
34.1
89.9
34.6
Net
Part NO
100.0 100
12.6 91
12.7 91
90.6 88
11.3 81
93.9 91
11.9 84
Residual
HC
.0 100.0
.3 39.3
.6 39.3
.5 87.3
.5 35.1
.8 86.3
.5 34.8
s In 1985 (X of
CO BOD SS
100.0 100.0 100.
26.5 16.9 9.
26.5 14.5 7.
84.9 93.2 91.
23.0 15.7 8.
86.8 99.3 99.
24.4 16.9 9.
Reference)
DS Nut,
0 100.0
3 61 .4
9 61 .4
9 88.5
7 54.3
4 96.7
2 58.6
100.0
44.1
27.4
99.3
43.8
99.9
44.2
1.   See Tables 2. 4, 9. 12. 17. 21. and C-1. for
     composition of pollution control costs by year from
     1976 through 1985.

2.   Part • Partlculates, SS • Suspended Solids. OS =
     Dissolved Sottds. Nut. • Nutrients, SO • Sulfur Oxides,
     NO • Nitrogen Oxtdes.SR

-------
The calculation procedure also provides for consistent
estimation of the mass of pollutants discharged Into the
nation's air and water, based on the particular economic
forecast and the particular set of pollution control
measures assumed to be employed-  This report expresses
pollution control costs as Increments occasioned by the
passage and implementation of the two laws.  For this
purpose. It was assumed that without current Federal
statutes, pollution control efforts would not have advanced
beyond that practiced 1n 1971.  Costs and pollutant
discharges reported are thus Incremental to those that would
have occurred by maintaining 1971 levels of control
throughout the 1976-1985 decade.

Table 1 presents the results from pairs of future forecasts
or scenarios designed to demonstrate the Impact of the
Federal air and water quality statutes on pollution control
costs and pollutant discharges under different economic
forecast assumptions.  Each pair contains estimates of
control costs (as a percent of GNP) and pollutant discharges
(as a percent of Reference Scenario forecasts for 1985) with
and without the Federal controls.  The costs without the
controls are zero by definition.  A third scenario 1s
presented under the official Reference Case forecasts. In
which the municipal wastewater treatment construction grant
program Is assumed to be augmented by $42 billion ($7

-------
billion per year for six years) above current
authorizations.  Comparing the first two scenarios of the
Reference Case. It Is seen that,  with Federal controls and
official economic growth rates, the discharge of
parttculates Into the air and suspended solids into the
water (n 1985 are estimated to be only 13 and 9 percent.
respectively, of what probably would have occurred during
that year fn the absence of the controls.

The projected reductions In pollutant discharges are
translated Into Improved ambient  environmental quality,
which Is. In turn, recognized as  pollution control benefits
by society.  Discussions In Sections Two and Three of the
report describe the techniques which can be used to
calculate various types of benefits on a monetary basis,
with several examples from the recent literature.  Some
benefits, however, such as esthetic and other psychic
categories, are not yet reliably  quantifiable, even though
subjectively they appear very Important.  This lack of
Information, plus the lack of sufficient data for
calculating the more quantifiable benefits on a nationwide
basis, make any total national benefit figures somewhat
suspect.  For these reasons,  projected total national
pollution control benefits are not expressed In dollar terms
In this report.

-------
Table 2 shows the estimated capital  Investment schedule for
various air and water pollution control  categories
throughout the decade.  These entries represent only the
value of the capital plant and equipment purchased,  and do
not Include any Interest charges or  operations and
maintenance expenses.  Table 3 presents  the total  cost
calculation Information.  Total cost Is  defined as the sum
of anualtzed capital costs plus operation and maintenance
costs.  Annual!zed capital costs are derived by amortizing
capital Investment at an Interest rate of 10 percent over
the life of the equipment.  This can be  thought of as
repayment of a capital loan at 10 percent Interest over the
life of the capital plant and equipment.  All Investments
are amortized at the same Interest rate, so they can be
compared on an equivalent basis.  Both sets of costs are
significant:  the Investment represents  the value of plant
and equipment that must be produced  and  delivered by the
pollution control Industry and related suppliers.   The total
cost Is a close estimate of the actual costs which must be
borne by the purchaser of the pollution  control equipment
and fact titles.

Since pollution control equipment must be continually
operated and maintained and periodically replaced, and new
equipment must be purchased for new  Industrial or public
service growth, total annual costs are continually

-------
Increasing.  The nation must be willing to accept these
costs as a permanent, Increasing part of the national  budget
If desired environmental quality Is to be restored and
maintained.  Figure 1 shows how total annual costs grow from
$27 billion to $64 billion over the decade.

-------
                         Salable 2.
              Air and Water Pollution Control
                  Investments.  1976-1985
                    (B111 Ions of 1975 $)
St
                    1976
1977
                      5.1
                      6.6
                     11.7
  7.3
  8.1
 15.4
                                      1978
        1979
        1980
        1981
       1982
                                                                                   1983
                1984
                1985
3.7
5.5
0.14
9.2
5.5
6.3
0.0
11 .8
5.9
11 .6
0.0
17.5
3.4
11.7
0.0
15.1
1 .7
11 .8
0.0
13.5
1.4
12.0
0.0
13.4
1 .0
12.1
0.0
13.1
0.7
12.2
0.0
12.9
0.6
12.4
0.0
13.0
0.7
12.4
0.0
13.2
24.6
108.1
0.2
132.9
Atr
  Stationary
     Sources
  Mob)le Sources
  Transportation
     Controls  .
Totals

Water
  Industrial
  Mun1cI pa I
Totals

Combined Atr and
  Water Total
 Components may not sum to totals due to rounding.

 Value approximately 0.003,  or $3 mil 11 on.$R
 4.8
 8.1
12.9
 6.8
 5.6
12.4
 7.8
 2.9
10.7
7.2
1 .8
9.0
6.8
0.8
7.6
5.4
0.7
6.1
1 .4
0.7
2.1
1 .7
0.5
2.2
                                                                                 Decade
                                                                                 Total
54.3
35.8
90.1
                                                                                                              223.0

-------
                         Salable  3.
                Total  Cost  of  Air and Water
               Pollution  Control,  1976-1985
                    (Bl1 lions  of  1975 $)
St
                    1976
1977
Air
Stationary Source
Annual 1 zed
Capital
0AM
Totals
Mobl 1e source
Annual (zed
Capital
0AM
Totals
Transportat Ion
Control
Annual Ized
Capital
0AM
Totals
Air Total
Water
Industrial
Annual 1 zed
Capital
0AM
Totals
Municipal
Annual 1 zed
Capital
04M
Totals



4.0
3.5
7.5


5.4
5.6
11 .0



0.02
(0.06)
(0.04)




2.2
3.7
5.9


2.3
0.7
3.0



4.8
4.1
8.9


6.6
4.6
11 .2



0.02
(0.01)
(0.01)




3.1
4.9
8.0


3.2
0.9
4.1
1978
                                        5.6
                                        4.6
                                       10.2
                                        9.4
                                        4.2
                                       13.5
                                        3.7
                                        5.9
                                        9.6
                                        4.1
                                        1 .4
                                        5.5
                                               1979
                    6.1
                    5.0
                   11.1
                   11.9
                    4.0
                   15.9
                                        0.02     0.02
                                       (0.14)    (0.13)
                                       (0.12)    (0.11)
                    4.5
                    6.4
                   10.9
                    4.7
                    1-6
                    6.3
1980
                    6.3
                    5.3
                   11.6
                   14.1
                    3.9
                   18.0
                             0.02
                            (0.12)
                            (0.10)
                    5.5
                    6.9
                   12.4
                    5.0
                    1.8
                    6.8
1981
           6.5
           5.4
          11.9
          16.1
           3.8
          19.8
           6.4
           7.2
          13.6
           5.2
           1.8
           7.0
1982
           6.6
           5.4
          12.0
          17.8
           3.6
          21 .4
                             0.02      0.02
                            (0.12)    (0.11)
                            (0.10)    (0.09)
Mater Total

Air A Water  Total

 Components  may not sum to totals due to rounding.

 Values In parentheses Indicate savings due to Improved  fuel economy.$s$R
           7.3
           7.5
          14.8
           5.3
           1.9
           7.2
                                                      1983
           6.7
           5.4
          12.1
          19.2
           3.4
          22.6
           8.0
           7.8
          15.8
           5.4
           1.9
           7.3
                                                      1984
 6.8
 5.4
12.2
20.4
 3.3
23.7
                             0.02      0.02
                            (0.12)    (0.12)
                            (0.10)    (0.10)
 8.2
11 .0
19.2
 5.4
 2.0
 7.4
                                                                        1985
 6.9
 5.4
12.3
21 .3
 3.2
24.5
 8.4
11 .5
19.9
 5.5
 2.0
 7.5
                                                                       Decade
                                                                       Total
 60.3
 49.3
109.6
142.2
 39.5
181 .8
                                      0.02      0.02
                                     (0.12)    (1.0)
                                     (0.10)    (0.8)
                                             290.6
 57.3
 72.8
130.1
 46.1
 16.0
 62.1

192.2

482.8
                            10

-------
                                   Figure 1.
                        Total Annual Expenditures for
                       Air and  Water Pollution  Control
 31
•H U
  °
C o
O -H
 I O

 
 u c
 01 O
 O -H
eg oa
O  C
i-i M
65



60



55



50




45



40




35



30



25



20



15



10 i



5
                                                                                                 Grand Totals

                                                                                                 Air + Water
                                                                                                 Total Air
                                                                 Total Water
                                                  _L
               1976
               1977
1978
                                        1979
1980    1981

     Year  .
1982
                         1983
                                                                                   1984
                                                                                   1985

-------
                     SbUNCERTAINTIESSR

Whenever a new estimate of national pollution control costs
Is produced, there Is a natural tendency to compare it with
related estimates developed by other parties.  Such efforts
frequently show considerable dlscrepency between estimates
for ostensibly the same cost categories.  Such d1screpencles
are. In fact, to be expected, due to the vast number of
conditions which must be specified to assure that even the
category of cost analysis Is Identical between two different
estimates.  For example, costs reported for the "Animal
Feedlots" category can contain estimates for any number of
the following animal  production businesses:  beef cattle,
dairy cattle, hogs, sheep, chickens, turkeys, etc.

Since most pollution control regulations are In terms of
effluent or emission concentrations, assumptions must be
made as to what type of technology will be generally applied
                                      e/'/Vk <:*//• QSL 
-------
Economic variables are another cause of discrepancies;
dollar value deflators,  Interest rates, equipment  lifetimes,
wage rates, power and material costs,  economic growth rates,
capital availability, and other factors must  all be
estimated and projected Into the future to produce an
estimate of pollution control  costs over a period  of  years.

In addition, the particular purpose of the estimates  may not
be exactly the same.   The estimates In this report represent
the expenditures which would probably be incurred  If  all
parties met the regulations on schedule by Installing an
assumed particular type of equipment.   The resulting
forecasts are thus unrealistic to the extent  that  polluting
activities fall to meet all requirements on schedule.

The Bureau of Economic Analysis (BEA)  of the  Department  of
Commerce conducts periodic surveys of Industries  to estimate
actual pollution control expenditures.  These estimates  are
                           /A/ so/»»e  ca-t-esories ; #ivd ov/y  o/v^ -//
nearly twice as high  as EPA estlmatesjln others.   These
differences can be attributed to variations In industry
category definition,  slower or faster equipment Installation
schedules, different  judgements of the amount of  Industrial
expenditures for process modification which can be properly
attributed to pollution reduction, end the probable
statistical errors In BEA's Industrial questionnaire
sampling process.  Chapter 3 In this Section  discusses the

                            13

-------
Impact that process modification can have on pollution
reduction and the difficulties Involved in apportioning cost
between pollution control  and production cost accounts.   A
discussion of BEA estimates Is presented In Section Four.

The closest parallel to the estimates in this report are the
estimates for the cost of  water pollution control  recently
prepared by the National Commission on Water Quality.   The
Commission's estimates Involve the same effluent
limitations, and many of the same economic assumptions and
Industrial category definitions.  Some of the Industrial
category cost estimates compare very closely, but  there are
still categories which differ significantly.  These
differences are attributed primarily to (1) uncertainties 1n
plant Inventories In those Industries characterized by large
numbetfof small plants. (2) differences In professional
judgement on what process  would most likely be applied to
achieve the required effluent quality, (3) Industrial  growth
rates and plant size trends over the decade, and (4)
different assumptions about the current status of  pollution
control in the Industries  (capltal-In-place).

To summarize, variations in estimates of national  activities
of this level of complexity are to be expected,  but detailed
examination of the data and calculation procedures can
usually explain the reasons for the variations.   The general

                            14

-------
economic assumptions used In this report are explicitly
stated in Sections One and Four of the report,  and Industry
descriptions and pollution control process descriptions are
described In considerable detail In Sections Two and Three.
Information In greater detail  In support of the entire
document ts maintained on ftta In the Agency.
                            15

-------
                        SbHIGHLIGHTS
Combined EffectsSR

The five most severely Impacted Industrial  categories based
on the total control costs for both air and water pollution
as a percentage of total  output for the category are
Fabricated Metals and Electroplating.  Pulp  and Paper.
Leather Tanning. Iron and Steel, and Steam  Electric Power.
all having combined expenditures In excess  of 4 percent of
their total output.  The  high ranking of the first and third
categories Is due entirely to water pollution control
expenditures.  The other  three show significant expenditures
In both air and water pollution controls.

In terms of pollution control Investment relative to total
Investment projected In the various sectors, the top five
categories become Leather Tanning,  Fabricated Metals and
Electroplating, Canned and Frozen Foods. Pulp and Paper, and
Grain Milling and Feed Mills.  This parameter projects a
more severe Impact, with  Leather Tanning being required to
spend nearly 70 percent of Its total decade Investment on
pollution control.  The lowest of the five. Grain Milling
and Feed Mills, 1s anticipated to spend over 15 percent of
Its total Investment on pollution control,  about 95 percent

                            16

-------
of which would be directed toward air pollution control.
The percentages on which these Impacts are based for the
entire Industrial community are shown In Table 4.   Chapter 4
of Section Four (Tables 8 and 9) presents the actual values
used to develop these percentages.
                            17

-------
                         Sellable 4.
Ranking of Impacted Sectors by Total Abatement Expenditures
          as Percentages of Investment and Output
                        (1976-1985)
$t

                                         Investment
Fabricated Metals 6
  Electroplating
Pulp & Paper
Leather Tanning
Iron & Steel
Steam Electric
Cheml cals
Nonferrous Metals
Grain Milling. Feed Mills
Asbestos. Clay, Lime &
  Concrete
Canned & Frozen Food
Petroleum 8 Asphalt
Plastics & Synthetics
Machlnery
FertI 11zers
Furnlture
Dal ry
Transportation Equipment
Meat & Poultry
     A Cane Sugar
       & Wood Products
                                                                                Output
Beet
Lumber
Paints
Glass
Natural Gas
BuiIder's Paper
Rubber Products
Text!les
Wholesale A Retal1
  (Grain HandlIng)
Mining
Printing
Agriculture
Services (Dry Cleaning)
Air
Rank X
.
5
.
6
2
10
3
1
7
.
4
15
20
8
9
-
18
.
.
.
12
.
14
.
-
17
13
11
16
.
19
.
5.05
.
4.69
6.28
0.99
5.98
14.49
2.98
.
5.37
0.20
0.04
1 .83
1 .36
-
0.10
.
.
.
0.54
-
0.37
-
-
0.12
0.53
0.55
0.15
.
0.04
Water
Rank X
2
4
1
10
13
5
22
20
23
3
1 1
9
8
12
-
6
14
7
15
21
-
18
-
16
19
17
,
-
-
24
-
22
10
69
5
4
9
0
0
0
15
4
7
7
4

9
3
7
2
0

1

2
1
1



0

. 11
.56
.31
.12
.23
.64
.79
.81
.25
.69
.87
.56
.67
.57
-
.59
.45
.93
.99
.80
-
.21
-
.34
.09
.33
.
'.
-
.20
•
Both
Rank X
2
4
1
9
7
6
14
5
17
3
8
12
13
15
21
10
16
11
18
24
26
22
28
19
23
20
27
25
30
29
31
22.11
15.61
69.31
9.81
10.51
10.71
6.77
15.31
3.22
15.69
10.24
7.77
7.72
6.40
1 .36
9.59
3.55
7.93
2.99
0.80
0.54
1 .21
0.37
2.34
1 .09
1 .44
0.53
0.55
0.15
0.20
0.04
A<
Rank
.
1
-
4 .
2
9
3
6
5
-
7
13
18
10
8
-
16
-
-
-
11
-
12
-
-
19
14
15
17
-
20
r
X
.
3.21
-
2.61
3. 16
0.64
2.70
1.75
1 .76
-
1 .28
0.07
0.02
0.61
0.88
-
0.55
. -
-
-
0.21
-
0. 18
-
-
0.02
0.06
0.06
0.04
-
0.01
Water
Rank X
1
4
2
6
9
3
14
21
23
5
13
7
8
12
-
10
11
15
16
17
-
18
-
19
20
22
.
-
-
24
-
5.30
2.05
4.54
1.55
0.93
3.34
0.37
0.10
0.07
1.78
0.39
1 .35
1 .07
0.43
-
0.73
0.67
0.29
0.25
0.24
-
0. 19
-
0.15
0.13
0.10
.
-
-
0.02
-
Both
Rank %
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
5.30
5.26
4.54
4.16
4.09
3.98
3.07
1 .86
1 .82
1 .78
1 .67
1 .42
1 .09
1 .04
0.88
0.73
0.73
0.29
0.25
0.24
0.21
0.19
0.18
0.15
0.13
0.12
0.07
0.06
0.04
0.02
0.01$R$S
                            IB

-------
$bA1r$R

By comparing total atr pollution control  costs to total
Industrial category output,  the ftve most severely Impacted
categories are found to be Pulp and Paper, Steam Electric
Power. Nonferrous Metals.  Iron and Steel, and Asbestos.
Clay, Lime and Concrete, all  five having  relative costs
below 3  percent.  Comparing air pollution control
investment to total project  investment in the categories,
the most severely Impacted Industries become Grain Milling
and Feed Mills,  Steam Electric Power. Nonferrous Metals,
Petroleum and Asphalt, and Pulp and Paper, their relative
pollution control Investments ranging between 5 and 15
percent.  Table 4 presents these rankings for the entire
Industrial community.

Since It Is now apparent that many stationary sources did
not meet the 1975 target for installet ion of controls. It 1s
no longer logical to assume  that all sources meet all
requirements according to specified schedules In the State
Implementation Plans.  Instead. It Is now assumed that
stationary controls will be  Installed according to a more
realistic schedule as proposed In a 1975  EPA report on "The
Economic Impact  of Pollution Control:  Macroeconomtc and
Industry Reports,• wherein all sogrces are brought within
                            19

-------
compliance by 1981.   This change shifts Investments into
later years 1n the decade as reported in Section Two.

Chapter 4 of Section Two discusses mobile source emission
control costs.  Costs in this category are significantly
different than earlier estimates due to a variety of
factors, Including changes In standards, changes In lead-
content phase-out schedules, changes In projected average
age and weight of the automobile population,  and other
conditions related to the recent energy shortage problems.

In metropolitan areas where mobile emission control devices
are not sufficient to guarantee achievement of ambient air
quality standards, additional efforts are required, such as
transportation control plans.  An Important factor In  these
plans Is Inspection and maintenance.programs, wherein  all
vehicle owners are required to have periodic Inspection and
required maintenance for both engine performance and
emission controls performance.  Such maintenance, while
costly, results In better gas mileage, which translates Into
an economic benefit to the owner.  Estimates of fuel savings
have been calculated in the analysis this time.  Using
assumed fuel cost projections, the Inspection and
maintenance programs actually result In a net economic
btntflt to the nation In addition to Improving air quality.
                            20

-------
Details of these estimates are shown at the end of Section
Two.
SbWaterSR

The five Industrial categories most severely Impacted by
water pollution control  regulations on a total  cost basis
are Fabricated Metals and Electroplating.  Leather Tanning,
Chemicals,  Pulp and Paper, and Canned and Frozen Foods,
their expenditures ranging between 1  and 5 percent of their
output.  On a relative Investment basis, the ranking Is
Leather Tanning, Fabricated Metals and Electroplating,
Canned and Frozen Foods.  Pulp and Paper, and Chemicals,
whose water pollution Investments range from 9 to nearly 70
percent of total projected Investments.  The rankings for
all water-using Industrial categories are presented In Table
Because of the recently completed municipal  wastewater
treatment Needs Survey, and the overriding constraints of
the Construction Grant Program on the rate of expenditure
for Municipal facilities,  costs were not calculated based
upon meeting requirements, but rather,  consist of the $20.8
billion of Federal funds scheduled for outlay during the
decade, required matching funds from local governments (25

                            21

-------
percent of total investment).  $600 million estimated to be
spent Irrespective of Federal  grants,  plus operation and
maintenance (O&M), and finance charges associated with the
aforementioned Investments.  Other Agency documents have
discussed the need for a greater amount of Federal  grant
funds.  Section Two.  accordingly. Includes an analysis of
the Impact of a potential augementatton of the Construction
Grant Program.

Because of the rapidly changing situation with regard to
nonpolnt source pollution from urban and rural stormwater
run-off, and because  control of this type of pollution is
primarily dependent upon plans Issuing from the P.L. 92-500
"208" planning process In late 1978. no estimate of the
types or levels of these controls Is currently available.
Therefore, there are  no estimates for costs of urban
stormwater pollution  nor agricultural  run-off pollution
control In this report.
SBComprehensIve AnalyslsSR

Section Four describes the analytical  procedures used to
develop pollution control cost forecasts on a year-by-year
basis over the decade.  By using the "Input/output" national
economic analysis technique of the Strategic Environmental

                            22

-------
Assessment System. Individual  Industrial  category growth
rates are derived which are totally consistent with the
aggregate national GNP growth  forecasts of each scenario.
This approach eliminates the problem of obtaining Individual
growth rates from a number of  different sources with varying
accuracies and differing base  assumptions.  The projected
growth rates are used In the cost analysis to provide an
Indication of the number of facilities subject t° ne* source
performance standards In both  air and water pollution
control.

The analysis of pollution control expenditures over the
decade, as presented In this report, provides the Agency
with a longer-range view of Its programs and a better
understanding of the Interrelationships and time phasing of
Its many different pnograms.  In addition, the ten-year
forecast  horizon should be sufficient to accommodate most
delays In compliance and still provide stable estimates of
total costs and related impacts.  The assessment system also
provides an estimate of the total amount  of residuals
discharged to air, water,  and  land disposal, rather than
studying any single environmental medium in Isolation.   This
allows for an analysts of potential secondary pollution
caused by primary pollution control processes, a concept
brought to the fore by the National Environmental Policy
Act.
                            23

-------
Finally, the existence of.a rout1n1zed, fully documented set
of data and calculation procedures provides for more
defensible cost estimates,  and quickly pinpoints particular
problem areas whenever the aggregate cost estimates are
challenged.  Negotiation and resolution of these detailed
areas of conjecture continually carries us toward improved
estimates which should result In fewer and fewer areas of
disagreement as the cost estimation exercise Is reiterated.
                            24

-------
 SaChapter  1$R
 Saintroduction$R
                        $bfURPOSE$R
 Standards  for  the control of air and water pollution have
 been  developed to implement the requirements of the Clean
 Air Act  (PL  91-604) and  the Federal Water Pollution Control
 Act (PL  92-500).  Under  the provisions of Sections 312(a)
 and 516(b) of  these acts, respectively, the Environmental
 Protection Agency (EPA)  has the responsibility of submitting
 regular  reports to Congress regarding the cost of pollution
 controls necessary to achieve the  legislated standards.

 Two series of  six reports each have been previously
 submitted  to Congress on the Economics of Clean Water and
 the Cost of  Clean Afr.   The last submlttals were a 1973
 biannual report on water and a 1974 annual report on air.
 For 1975,  the  two reports have been combined Into this
 single  Integrated report to provide a comprehensive
.assessment of  both air and water pollution control.
                            1-1

-------
                 SbSCOPE AND ASSUMPTIONSSR
This report presents the best available EPA estimates of the
national costs of complying with the Clean A1r Act and the
Federal Water Pollution Control  Act over the next decade,
1976-85.  It begins with an overview section which presents
a discussion of those Issues common to the study of both air
and water pollution control.  The next two sections present
the costs of pollution control for air and water.
respectively, together with estimates of the reduction in
environmental pollution effected by the controls-  The
fourth and final section presents an analysis of the
economic Impacts and tradeoffs associated with these costs.
Particular emphasis Is placed on Illustrating how these
Impacts and tradeoffs might change under alternative sets of
assumptions about future economic activity and energy
conservation policies.

Included In the overview (Section One) Is a presentation of
the basic assumptions and general approach taken in the
development of control costs and In the analysis of the
consequent Impacts of these costs.  This Is followed by a
discussion of the concept of benefits as applied to the
economic analysts of pollution control.  Finally, the
economic advantages of controlling pollution through process

                            1-2   <

-------
changes are presented.   Five maj.or Industries are used as



examples In this analysis:  copper, aluminum,  pulp and paper,



petroleum refining, and Inorganic chemicals.







Both Sections Two and Three begin with a brief summary



followed by a discussion of the estimated types of damages



resulting from pollution.   In Section Two,  the cost of



controlling air pollution  is presented in terms of



government program expenditures. Industry and utility



control costs, and transportation control costs.  Section



Three, on the.cost of controlling water pollution, also



Includes a presentation of  government program expenditures,



followed by municipal and  Industrial  cost estimates.








A comparative analysis approach 1s taken (n Section Four to



examine the relative Impact of pollution control under



alternative futures or scenarios.  included In this



presentation is an examination of the gains and losses



experienced by consumers and by individual  Industries which



spend and/or receive funds  for pollution abatement.








Wherever possible, the national pollution abatement costs,



the economic impacts and tradeoffs, and the associated



environmental changes that  have been estimated and presented



In this report are those that would not have occurred



without Federal legislation.  Specifically. 1t is assumed






                            1-3

-------
that. In the absence of the two laws,  the amount of
pollution discharged per unit of production (or per person
for sewage, per mile for vehicles) would have remained the
same as In 1971.  A pre-1egtslat ion baseline, defined In
terms of 1971 pollution control technology levels,  is thus
established,  and all costs. Impacts, tradeoffs, and
environmental changes are measured as differences from that
basel me.
SbProblem OvervlewSR

Both the comprehensive assessment of pollution control and
also the Industry-by-Industry estimates of pollution control
expenditures and pollution reduction are presented at the
national level.  Although more detailed Information 1s
provided In many instances at state or local levels, or for
typical sizes of Industrial plants, this Information 1s
presented primarily to enhance an understanding of the bases
established for the national aggregated estimates.

Estimating the control costs and the quantities of
pollutants produced on a national basis is a complicated
process.  Not only are there a large number-of pollution
sources, but each source could emit a number of pollutants
that can be controlled separately or Jointly by several

                            1-4

-------
alternative control  technologies.   Conversely,  each specific
pollutant can be traced to a considerable number of
different sources.   The costs of control  are most
conveniently estimated by source,  even though they will
usually cover more than one pollutant for each source.   On
the other hand, levels of pollution are more easily examined
by pollutant: these levels are estimated by aggregating
emissions by pollutant across all  sources of that pollutant.

A general overview of the relationships among sample
sources, pollutants,  effects, and control technologies Is
presented 1n Table 1.  Discussions of these relationships
are found for each Industry affected by Federal pollution
control legislation In Sections Two and Three of this
report.
                            1-5

-------
                         SdTable 1 .
     Overview of Sample Pollution Control  Relationships
Medium

Air
Source

Automoblles
           industry
              Sulfur1c Acid
              Petroleum
Pollutant

NOx.  HC.CO
           Electric Utllltles   SOx
                                Partlculates
                     SOx
                     HC
Effects

Smog.  Lung Damage


Respiratory Problems
                                          Sol 1 ing.  Reduced
                                          Vlstbl11ty
                     Respiratory Problems
                     Smog
Control Technology

Engine Modification,
Catalysts

Scrubbers.  Fuel
Switching

Electrostatic
Precipltators,. Filters
                        Absorption
                        Floating Roof Tanks
Water      Municipal Sewers
           Industry
                     BOD

                     Suspended Solids

                     Pathogens

                     BOD

                     Suspended Sol Ids

                     Dissolved Sol Ids

                     Acids

                     Toxics
                     Dissolved Oxygen        Oxidation. Adsorption

                     Materials, Fish Damage  Sedimentation, Filtration

                     Infection               Disinfection

                     Dissolved Oxygen        Oxidation. Adsorption

                     Materials. Fish Damage  Sedimentation. Filtration

                     Materials Damage        Ion Exchange

                     Materials Damage        Neutralization

                     Poisoning               Adsorption
                            1-6

-------
SbAssumptlons$R

The Federal pollution control  legislation ultimately
requires Industries, consumers (transportation vehicles),
and municipalities to lessen or completely eliminate their
discharges of pollutants Into the nation's atmosphere and
waterways.  Hence, these pollution contributors must spend a
portion.of their money resources for pollution abatement
regardless of the state of the economy.   However,  pollution
control expenditures are not independent of the state of the
economy because the level of economic activity affects the
level  of production, which In turn affects the amount of
pollution generated by Industries, consumers,  and
municipalities.  Consequently, the forecasts of pollution
control expenditures are based on corresponding forecasts of
national economic activity.

Forecasts of pollution control expenditures must aiso be
based upon explicit assumptions about the rate of  compliance
with pollution control legislation.  The assumed timetables
for Installing pollution abatement equipment are given later
In this Introduction as part of the compliance assumptions
for this report.  All cost estimates presented In  this
report are expressed 1n 1975 dollars unless otherwise noted.
In addition, annual costs apply to calendar years  unless
specified differently.

                            1-7

-------
SbEconomtc AssumpttonsSR

A consistent set of economic assumptions is the basis for
the cost estimates presented In this report.   These
assumptions were used to produce a "Reference Case" forecast
of the U.S. economy and are summarized In Table 2.   An
alternative set of economic assumptions Is presented In
Section Four; the pollution control  cost and pollutant
discharge estimates corresponding to this alternative
scenario enable us to evaluate possible variations  from the
Reference Case estimates introduced  by different economic
assumptions.
SbEnergy AssumptlonsSR

The energy assumptions for Reference Case pollution control
forecasts are taKen from the Federal Energy Administration's
"Business as Usual" scenario in the November 1974 Project
Independence Report where the Import price for oil  is $7 per
barrel; they are summarized In Table 3.
                            1-8

-------
                          Table .2.
            Reference Case Economic Assumptions
                             St
Economic
Assumption
Population-Series E
Projections
(Millions of People)

Labor Force
(Ml 1 lions of People)
Labor Productivity
Gross National
(Trl1 lions of
1975 Dollars).
Product
Forecast Time Period
Unemployment Rate In
1985 (Full Employment
Economy)

Nominal Interest Rates
Federal Expenditures In
1980 and 1985 Excluding
Transfers and Pollution
Control Progress.
(Ml 11 tons of 1975
Oollars)

Federal Expenditures
for Pollution ControlSR
           Government
           Agency

           Bureau of the
           Census
           Bureau of Labor
           Statistics
•Bureau  of  Labor
 Statistics

 Ford/Councl1  of
 Economic Advisors
 (1975-1980)  Bureau  of
 Labor Statistics
 (1980-1985)
Values

1975-213.9
1980-224.1
1985-235.7

1975-  93.8
1980-101.8
1985-107.7

Varies by
Industry

1975-1.47
                                   1976-
                                   1977-
                                   1978-
                                   1979-
                                   1980-
      .57
      .69
      .81
      .85
      .99
                                                  1985-2.40
           EPA
           Bureau of Labor
           Statistics
           Office of Manage-
           ment and Budget

           Department of
           Commerce, Bureau
           of Economic
           Analysis
           EPA
                      dan.  1.  1976
                     Dec.  31.  1985

                            4.5%
                      Public   -  10%
                      Private  -  10%

                   1980  -  $156.400
                   1985  -  $173.400
                            1-9

-------

1972
12.495
32 . 966
23.125
576
2.946
$t
1977
1 6 . 854
37.813
21.558
2.830
3.543

1980
18.074
41 .595
22.934
4.842
4.014

1985
19.888
47.918
23.947
12.509
4.797
                          Table 3.
          United States Total  Gross Consumption of
       Energy Resources (In Trillions of BTU's/Year)
    (Buslness-as-Usual Without Conservatlon-$7/Bbl 011)
  Fuel


Coa 1

Petroleum

Natural Gas

Nuclear Power

Other
TOTALS             72.108      82,598      91,459     109.059
Source: ProJect_Independence_Report,  Federal
        Energy Administration,  Appendix A1,  p.37.
        November 1974.$R
SbAIr Compliance AssumptlonsSR



EPA regulations and Federal  legislation related to the Clean

Air Act of 1970 apply different levels and modes of air

pollution controls to these  specific pollution source

categories:  mobile sources (transportation vehicles).

existing stationary sources  of air pollution,  new stationary

sources of air pollution,  and sources of hazardous

pollutants.   The Clean Air Act and the cost estimates
                           1-10

-------
presented in this report are based on the principle that
pollutant emissions win be brought under whatever level of
control Is necessary to achieve national  ambient air quality
standards.  However, for many different reasons, many
Industries have not met the July 1. 1975, compliance date
originally set for existing stationary sources.   Similarly,
the original dates and standards established for
transportation vehicles have been changed.  The specific
assumptions for each source category are described below:

  1. Mobile Sources (Transportation Vehicles).   The
emissions standards and the compliance schedule which must
be met by mobile sources are presented In Section Two of
this report (see Mobile Sources and State Transportation
Control Plans).  The assumed compliance dates reflect the
delayed Implementation of standards for reduced hydrocarbons
and carbon monoxide emissions from light-duty vehicles from
model year 1977 to model year 1978.

  2. Stationary Sources (Existing).  Stationary sources of
air pollution (Industrial plants, electric utilities) which
existed at the time of passage of the Clean Air Act are
regulated by approved State Implementation Plans (SIP's).
The standards assumed for each Industry and for utilities
are given In the industry summaries In Section Two of the
report.  Most SIP's require compliance by duly 1. 1975, but

                           1-11

-------
achievement of this goal  would imply a peaking of Investment
which did not occur in 1974 and 1975.   Hence,  except for
sulfur dioxide control by electric utilities,  all existing
stationary sources are assumed to be moving toward full
compliance at an extended expenditure rate, as given In  the
Summary for Section Two.   A compliance date of January 1,
1981. Is assumed for sulfur dioxide from utilities.

  3. Stationary Sources (New).  New sources of air pollution
Include new Industrial plants built since the  passage of the
Clean Air Act and also existing plants which have made
certain modifications in their facilities.   These sources
are assumed to comply with EPA New Source Performance
Standards (NSPS) except where such standards have not yet
been developed or where SIP standards are more stringent.
In these latter two cases. SIP standards are assumed.  New
pollution sources are assumed to be in compliance with these
standards when they go Into operation.  The exact standards
being assumed are given In the appropriate sections in
Section Two.
                           1-12

-------
SbWater Compliance Assumpt1ons$R

UnlIKe the Clean Atr Act.  the 1972 Amendments to the Federal
Water Pollution Control  Act prescribe full  Federal
regulation of water pollutant sources,  except as rede legated
to specified states.  In addition to setting ambient water
quality standards to be  met by 1983, the act specifies the
levels of control technology to be utilized by industrial
and municipal pollution  sources by July 1,  1977 and by July
1, 1983.   EPA has defined these technologies for most major
Industrial pollution sources in effluent guidelines
documents.  It enforces  the act through permit programs In
40 states, the remaining 10 having been delegated authority
for state enforcement.   The provisions of the act and the
compliance assumptions for this report are enumerated below.

  1. Industrial Sources.

     a. Industries discharging pollutants Into the Nation's
        waters In 1972 will adopt the best practicable
        pollutIon control  technology (BPT) by January 1,
        1978, and the best aval'able technology (BAT) by
        January 1, 1984.  These dates have been pushed back
        six months from those specified 1n the act to allow
        the analysis for this report to be done on a
        calendar year basis.

                           1-13

-------
b.  Industries for which BPT and BAT are not defined In
   EPA guidelines are assumed to adopt control
   technologies similar to those of related industries
   covered by the guidelines.  Specific control
   technology assumptions for water polluting
   Industries Investigated by this report are provided
   In Section Three.

C.  Industries discharging their wastewater Into
   municipal treatment plants must (and It 1s assumed
   they do) pretreat  their effluents so that Industrial
   pollutants do not  Interfere with plant operation and
   do not pass through the treatment process without
   adequate treatment.  Pretreatment technology  must be
   operating by January 1, 1978.  Pretreatment  Is
   assumed to be unnecessary for those Industries for
   which pretreatment guidelines have not been
   prepared.

d.  All new sources of water pollution (usually plants
   constructed since  1974) are assumed to comply with
   EPA NSPS guidelines.
                      1-14

-------
2. Municipal Sources.







   Compliance with the Federal  Water Pollution Control  Act



   by all  publicly-owned sewage treatment plants in



   existence on duly 1.  1977.  would require them to



   achieve a secondary treatment level  for all effluents.



   Because of the current economic recession and the



   corresponding difficulty facing the  municipalities In



   raising capital,  treatment  plants.cannot be built at a



   fast enough rate  to assure  compliance with the act.



   Instead, It Is assumed In this report that new plants



   Mill only be built  as rapidly as permitted by Federal



   appropriations and  state matching funds, which are



   Proposed as shown In Table  4.
                         1-15

-------
                          Table 4.
         Direct Capital  Outlays for Construction cf
           Publicly Owned Sewage Treatment  Plants
                 (Federal.  State,  & Local)
               (In Millions of  1973 Dollars)

                             $t
                    Fiscal  Year          Calendar Year

1975                   4.190                4.190
1976                   4.182                4.190
Transition             1.260                N/A
1977                   5.462                4.769
1978                   6.212                5.691
1979                   5,538                5,865
1980                   3.925                5,017
1981                   1.850                3.349
1982                 .  1.100                1.829
1983                     600                1.006
1984                     600                  662
1985                    --                     380

  This "transition period"  represents the months of
  July through September 1976;  all  subsequent  Fiscal
  Years wl11  run from October 1 through September 30
  of the following year.SR

     Section  Three discusses In depth the relationship

     between  these appropriated funds and the  expenditures

     which would be necessary to comply with the act.   It

     also considers the possibility of a $5 billion

     continuing appropriation after current appropriations

     exptre.
     These economic,  energy,  and compliance assumptions and

     other less quantifiable  policy variables are further

     discussed In Section Four.
                           1-16

-------
  3.  Elimination of Discharge.

     Although Elimination Of Discharge (EOD)  Is specified as
     the goal of the Water Pollution Control  Act.  It  Is not
     currently required by regulations except for  those
     Industries where BAT Is the same as EOD.  Consequently.
     EOD Is not assumed for the pollution control  cost
     estimates appearing In this report.
                 SbPOLLUTION CONTROL COSTS:
           DEFINITIONS AND CALCULATIONS METHODSSR
The various costs presented 1n this report are described
below, and the general  approach used to estimate costs in
each of three major categories 1s discussed.   The three
categories are direct costs,  government program
expenditures, and Indirect costs.
                           1-17

-------
SbDIrect CostsSR

The expenditures associated with acquiring,  owning,  and
operating the buildings and equipment needed to control
                                                       •
pollution are direct costs.  These costs are directly
Incurred by Industries and municipalities to reduce
pollutant levels: they include Investment costs, operating
and maintenance costs, and the costs Incurred to borrow the
necessary capital funds.
Sblnvestment CostsSR

These costs Include all expenditures for pollution control
equipment and associated modifications or additions to
buildings.  They are the actual cash outlays used to
purchase and Install the equipment and to construct the
buildings or building changes.  In the case of municipal
treatment plants, the cost of building the whole plant Is a
Investment cost for pollution control.  These costs do not
Include those charges made by a lending Institution for
borrowing the money, nor do they take Into account the
Income tax writeoff benefits which accrue to an Industry due
to depreciation.
                           1-18

-------
SbOperattng and Malntenance$R
$b(08M) CostsSR

The annual costs of operating ana maintaining ths pollution
control equipment and plant Include expenditures for:

  1. Materials used by the equipment (e.g..  chemicals)

  2. Labor for maintenance and repairs

  3. Energy

  4. Materials for repairs

  5. Overhead

  6. Monitoring  (labor)

  7. Byproduct credits.
                            1-19

-------
SbTotal Annual CostsSR

Total annual costs are those costs Incurred each year by
Industry or government (municipalities)  in owning and
operating pollution control  equipment and plants.   They are
the sum of the 0AM costs for the year and the annual!zed
capital costs for the year.   Note that annualIzed capital
costs are not the same as the Investment costs discussed
above.  AnnualIzed capital  costs are derived by amortizing
the initial investment over the life of  the facility, and
can be thought of so the annual amount needed to repay the
loan with interest over « specific time  period.
SbCosting Me.thodologySR

The direct costs of air and water pollution control  are
reported separately for each source and source category.
For air pollution, the major source categories are:
(1JAstationary sources, comprising industries, power
utilities, and space heating: and (2)Amob1le sources.
namely, automobiles, trucks, and aircraft.  The major  source
                        t
categories for water pollution are: (1)Apo1nt sources, which
include municipalities, industries, and power utilities;  and
(2)Anonpotnt sources, primarily run-off from urban areas  and
from mining and drilling operations, and agricultual crop

                           1-20

-------
production activities.   Because nonpotnt-source pollution

control Is a far more complex problem and an established

regulatory procedure such as effluent permits ts not yet

developed and Implements! costs for these sources could not
                                                       •
be reliably estimated,  and hence,  are not reported in this

document.



The details of calculating costs differ among the major

source categories.  In general, the procedure for each

source Is:



  1. Examine the regulations to determine the emission or

     effluent standards to be met.
                                          i


  2. Select from the alternative technologies those

     pollution control  methods that are likely to be

     employed.



  3. Estimate the cost of using these methods for

     representative units (plants, vehicles, or run-off

     areas)-.      .



  4. Multiply these unit estimates by the total number of

     such sources In the nation that are anticipated to

     require control 1n the appropriate year.  Thus, for

     automobile emission controls, the cost of an individual


                           1-21

-------
     control system ts multiplied by the total  nurr.ber of
     automobiles estimated to be sold In the appropriate
     year with that system.

This procedure, which 1s more complicated for Industrial
sources. Is outlined below and Is discussed more thoroughly
In Section Four of this report:

  1. Total Industry production capacity 1s Inventoried or
     estimated.

  2. Unique production processes within the Industry which
     emit differing levels of pollutants and/or require
     different control techniques are Identified.

  3. For each production process, the applicable abatement
     control technologies are Identified and the percentage
     of plants using each technology Is specified.

  4. For each control technology associated with a  given
     production process, the percentages of plants  covered
     by different state Implementation plans are estimated
     (for air control cost calculations only);

  5. Usually from one to three typical plant sizes  for each
     given Implementation plan,  control technology,  and

                           1-22

-------
     production process combination within the industry are
     defined.   (This combination Is hereafter referred to as
     an Industry segment.)

  6. The capacity for the Industry segment Is allocated
     among the plant sizes,  and capital  and O&M costs are.
     developed for a typical  plant of each size 1 r» the
     segment,  depending on  the standard it must meet.  (This
     depends In part on whether It Is a hew or existing
     plant).

  7. The costs are applied  to all  plants of the same size
     within the segment: then costs for the different size
     Classes are summed to  obtain  total  capital and O&M
     costs for the segment.   This  1s done for each segment
     of each production process within the Industry.
     Control costs for the  Industry are obtained by
     totalling all the capital and O&M costs computed for
     the Industry's segments.

The costs associated with building and operating municipal
wastewater treatment plants for this report are directly
related to the Federal appropriations and state matching
funds available to build new plants.  These costs have,
however, been reported In five "Needs" categories.  These
categories relate to the Municipal Needs Survey (Final

                           1-23

-------
Report to the Congress. "Cost Estimates for Construction of
Publicly-Owned Wastewater Treatment Facilities",  revised May
6. 1975) which was conducted In 1974 by EPA to determine the
physical facilities needed by municipalities to adequately
handle their sewage treatment problems: the categories are:
     Category I
Secondary Treatment Required.
     Category II    -   More stringent treatment  required
                       by water quality.

     Category IIIA  -   Correction of sewer
                       Infiltration/inflow.
              11 IB  •   Major sewer rehabilitation.
     Category IVA
              IVB
Collector sewers.
Interceptor sewers.
     Category V
Correction of combined sewer
overflow.
              VI     -   Treatment  and/or  control  of
                       stormwaters.
                       Combined Sewers.
                           1-24

-------
            SbGOVERNMENT PROGRAM EXPENDITURESSR
Program costs which are Incurred by governmental  agencies In
carrying out pollution control legislation Include
expenditures for planning, administration, enforcement,  and
research grants.  These costs are Incurred at all  three
levels of government: Federal, state, and local.   The costs
of constructing, operating, and maintaining control
equipment owned by these governments are direct costs, and,
as such, are Included In the air and water program costs
discussions 1n Sections Two and Three, respectively.
$bA1r Program CostsSR

Government program costs for air pollution control have been
estimated separately for Federal and non-Federal programs.

Federal programs Involve two types of funds: grant funds,
which are passed on to state and local governments; and In-
house funds, which are expended by a Federal agency or by
Its contractors.  Es.tlmates of projected grant expenditures
are obtained from the relevant agencies, primarily from EPA,
which accounts for the vast majority (1973) of grant funds.
and from the Appalachian Regional Commission and the

                           1-25

-------
Department of Transportation,  which account for most of the
remainder.  Estimates of projected in-house expenditures are
based upon Fiscal Year 1974 outlays.  Fiscal Year 1974-76
obligations,  and forecasts of  trends supplied by the
relevant Federal agencies.

The basic procedure used for estimating program expenditures
by state and local  governments makes use of available data
for 15 representative states.   The estimated ratios of
expenditures for various functional areas, such as
enforcement and engineering, are first derived for these
states and are then applied to all other states based on the
similarity of Industrialization, geography, population, and
general air pollution control  policies.

In general, sources of data for projecting government
program costs for air pollution abatement beyond 1979 were
not available.  Instead, extrapolations were made from
baseline data on the basis of  several reports that provided
forecasts of future government expenditures for specific
program components.
                           1-26

-------
SbWater Program CostsSR








The 10-year Federal  water program expenditure projections



are essentially the  EPA's response to the Congressional



Budget and Impoundment Act of 1974 (P.L.  93-344).   The major



assumptions underlying the 10-year projections are:







  1. Future year estimates are a continuation of the budget



     year (Fiscal  Year 1975)  program level,  except for



     statutory or  other provisions which  make the  future



     year size of  the program uncontrollable, and



     legislation or  other provisions which clearly add a new



     component.








  2. No new major  legislative amendments  will be nade to the



Federal Water Pollution Control Act.








As with the air program expenditures. Federal water program



expenditures are divided Into two general categories:



Assistance Programs, which administer Federal grants: and



Regulatory Programs, which Include all  other Federal



administration and enforcement expenditures.








The 10-year state program expenditure projections  are



derived from the requirements under the 1972 Amendments of



the states to Issue  permits,  review construction grants, and






                           1-27

-------
 monitor  compliance.   Permit  costs are developed  for each
 major  category  of  activity.   State  agencies perform a
 variety  of  additional  activities over and above  those needed
 to comply with  Federal  requirements: the expenditures for
 these  activities are  not  Included here.  In addition, there
 (8 no  provision for program  expenditures for nonpolnt-source
 control  activities.
 Sblndlrect  CostsSR

 Indirect  costs  are  those experienced by government,
 business, or  consumers as a  result of having  to bear  the
 direct  costs  of pollution control.  The added industrial
 costs  for pollution control  must  either be passed on  to the
 consumer  1n the form of  Increased prices or be absorbed by
 Industry  In reduced profits.   Where Investment requirements
 are  high  and  profits are already  low. some marginal plants
 might  find  It Impossible to  continue operation in the face
 of pollution  control  requirements.  The resulting plant
•closures  may  thus result In  local unemployment problems.

 This report examines some of  the  Indirect macroeconomlc
 effects of  pollution control  at the national  level.   Thus,
 Section Four  presents an analysis of the impact of control
                            1-28

-------
costs on aggregate production.  Investment,  employment  and
other national accounts.

EPA's Office of Planning and Evaluation 1s  currently engaged
In a series of detailed economic Impact analyses for six
major industries:  steel,  electric utilities,  non-ferrous
metals,  petroleum refining,  chemicals,  and  pulp-and-paper.
These studies, to be completed during Fiscal  Year 1976,  will
cover the effects of current and proposed emission and
effluent standards on prices, profits,  production.
productivity, plant closures, and employment  for each
Industry, at both national  and regional levels.
                SbCOMPREHENSIVE ASSESSMENTS!)
The primary reason for assessing the costs,  benefits,  and
Impacts of air and water pollution control resulting from
Federal legislation and regulations In the same report Is to
make possible analysts of total Impacts on the economy,
Including changes In the Interrelationships among the
various elements and sectors of the economy.  Another
consequence of the combined report Is the capability of
estimating the total pollution control costs for a single

                           1-29

-------
Industry and their likely Impact on that Industry.   For this
report, a comprehensive.  Impact estimation and analysts
system has been used to examine the comprehensive impacts of
pollution control, at both national and Industry levels.
This system, the Strategic Environmental Assessment System
(SEAS), Is summarized In Section Four.

alternative assumptions about the future.   A comparative
analysis procedure Is then used to assess the. results.
Scenario assumptions, scenario run results,  and comparisons
among scenarios are presented In Section Four.

As noted earlier, pollution control expenditures are not
Independent of the state of the economy.  Similarly, the
Impact of these expenditures on the economy, the
environment, and energy consumption depend on the Initial
assumptions made about the future In each of these  areas.
Hence, the objective In this report Is  not to predict
exactly what the Impacts of pollution control will  be over
the next 10 years, but rather to conditionally forecast
their relative magnitude and interrelationships.  The
analysis focuses on how Impacts vary as basic assumptions
about future economic activity and energy policy are
differentially changed.
                           1-30

-------
 The  comparative  analysis scheme used to assess the economic
 end  environmental  Impacts of pollution control in this
 report  takes  Into  account that various experts may hold
 differing  views  about  future U.S. economic growth, economic
 composition,  and energy consumption.  By exploring the
 Impacts of a  range of  reasonable assumptions about the
 future, one  Is able, by this approach, to determine how
 sensitive  the economy, the environment, and energy budgets
 are  to  alternative actions.
                  SbALTERNATIVE  FUTURESSR
 Assumptions  for  several alternative  futures or scenarios are
 defined  In Section  Four of  this  report.  These scenarios
 provide  the  basis for  the comprehensive assessment of
 pollution control Impacts on  the economy and the environment
 also presented  In Section Four.  Although one forecast has
 been termed  the  Reference Case.  It should not necessarily be
 Interpreted  as a prediction of the most realistic future.
.Rather, -.It is the benchmark or reference against which the
 comparative  analysis was conducted.  Assumptions for the
 Reference Case are  essentially those enumerated earlier In
 this Introduction.  They describe a  high productivity/high

                            1-31

-------
growth-oriented economy where full  employment  Is reached In

the early 1980's.


Other alternative scenarios considered In Section Four  are

briefly described below.


  1. The Low Productivity Scenarios.   These scenarios are

based on time series projections of labor productivity  from

1952 to 1971 made by the developers of the INFORUM input-

output model of the economy used in SEAS.  They reflect a

slowing down of productivity because of shifts toward

service Industries In the pattern of final demand, and

because of a slowing down of the productivity  increase  rates

In other Industries.  GNP estimates which correspond to

these assumptions are shown In Table 5 compared with those

for the Reference Case.
                          Table 5.
      Comparison of GNP Estimates for Low Productivity
                and Reference Case  Scenarios
               (In Trillions of 1975 Dollars)

                       Low Productivity  Reference Case
                             GNP               GNP

        1975                 1.53              1.47
        1977                 1.65              1.69
        1980                 1.84              1.99
        1983                 1.99              2.23
        1985                 2.08              2.40
                           1-32

-------
  2. The Energy Conservation Scenarios.   These scenarios
comprise a variation of the Reference Case In which energy
consumption is reduced through selected conservation
measures.  It Is based on the Federal Energy
Administration's "Buslness-as-Usual  with Conservation"
scenario where the Import price of oil Is $11 per barrel.
(See Appendix A1,  page 46 of the November 1974 Project
Independence Report.) The energy usage composition projected
by Project Independence Is not exactly matched because  of
differences in energy demand resulting from the
redistribution of  monetary savings to consumers.

Two scenarios are  run and analyzed for each set of economic
and energy-related assumptions-  The first scenario tn  each
case Is used to develop a set of forecasts on the economy,
Industry output, environmental residuals, and energy budgets
given no.Increase  In pollution control beyond that present
In 1971.  The same parameters are then forecast In a second
scenario, with pollution controls, costs, and equipment
punchases superimposed on the original economic assumptions
as necessary to comply with Federal  legislation.   This
procedure results  In six major scenarios:
                           1-33

-------
                       Without Incremental    With
                       Incremental Abatement Costs       Abatement
                       Costs
Reference Case         Scenario 1             Scenario 2 Low
Productivity        Scenario 3            Scenario A Energy
Conservation     Scenario 5       •    Scenario 6
The scenarios are then par led for a comparative analysis of
relative Impacts and tradeoffs In the following manner:
(1.2) (1.3) (1.5) (2.4) (2.6) (3.4) (5,6).   A subset of
Scenario 2, which assumes a continuing appropriation of  $5
billion a year for municipal  sewage treatment facilities,  is
also compared with Scenarios 1 and 2.  In addition to these
analyses, which are presented In Section Four,  Section One
Includes a study of the cost  savings resulting from process
change as compared with Scenario 2 control  costs.
                           1-34

-------
SaChapter 2$R
SaThe Benefits of$R
SaPoHutton Control ProgramsSR
Pollution control legislation has traditionally favored
rigid standards, either to control the discharge of
pollutants Into air and water or to maintain ambient quality
levels.   While It was not possible to base such legislation
on an analytical estimation of the full  benefits that would
result.  Its enactment reflected the judgement that the
overall  benefits to society were great enough to justify the
necessary costs.  Federal legislation also recognized the
need for more elaborate and more accurate assessment of the
costs and benefits of such programs, both for their
Implementation and for future consideration of additional
legislation.

The purposes of such an assessment transcend the emphasis
often given to the techniques for quantifying benefits and
their numerical results,  Important though they may be.  The
purpose of cost-benefit analysis Is to provide the type of
information on the value of public investments that the
market system provides on the value of private Investments.
However, public Investments usually have many objectives In
addition to those easily measured In dollars and cents.

                           1-35

-------
Still, the process of logical and systematic scrutiny that
Is Inherent In the accepted methods of cost-benefit analysis
can contribute greatly to society's ability to improve Its
well-being by allocating more efficiently Its limited
resources.

Thus, a major purpose of this discussion and assessment of
the national benefits of air and water pollutIon control  1s
the achievement of a more precise understanding of the
nature, sources,  and approximate magnitude of such benefits.
Such en understanding, when shared by legislators, program
managers, and the public, may well be of greater value than
the numerical results themselves.
                 SbDEFINlTION OF BENEFITSSR
Benefits of controlling air -and water pollution derive from
the reduction of damages caused by air and water pollution.
The measurement of benefits 1s performed In terms of the
damages that would otherwise be incurred.   A baste concept
In benefit evaluation Is willingness to pay. which can be
defined as the highest price that Individuals would be
Hilling to pay to obtain the improvement in air or water

                           1-36

-------
 quality  resulting  from  a  given  pollution control  program.
 Benefits are  evaluated  whenever possible In monetary  terms
 because  It  provides  a common measure of all the  types of
 benefits and  costs.  The  corresponding economic  damages
 result  In out-of-pocket losses  caused by Increasing the
 costs of using air and  water, by decreasing the  level of use
 Of  the  resource, and by Increasing  costs of avoiding  or
 repairing the effects of  pollution.

 Many types  of benefits  are  not  amenable to quantification in
 monetary terms because  of their nature and the state  of the
 art of available measurement methods.  This Is the case with
 "psychic* damages, so labeled because they relate to  the
 pleasure or displeasure associated  with the use  of the air
 and water In  our environment.   Psychic damages include
 decrease or loss of  pleasure from the use of air  or water
 that has become polluted, and fhe Increased experience of
 displeasure,  pain, and  anxiety,  as  well as'the so-called
 option,  preservation, and vicarious values experienced by
 non-users.

 Option values arise  because people  are willing to pay to
.ensure  the  availability of  clean air and water,  even  If they
 are uncertain'when or how they  would actually use It.
 Preservation  values  arise In a  similar fashion,  when  people
 are willing to pay for  the  preservation of a resource, even

                            1-37

-------
when they are certain that they will never use It directly.
Both preservation and option values are frequently
associated with a unique environmental  resource,  for which
no substitute exists.  Preservation value can also be
associated with risk aversion. In which a value ts placed on
the reduction tn the probability of the loss of an
environmental resource through extinction of a species or
collapse of an ecological system.

Finally- the term vicarious satisfaction has been used to
describe the motivation of people who are willing to pay to
provide benefits for their fellow citizens rather than for
themselves, and bequest value describes the similar benefit
derived by Individuals preserving an environmental resource
for future generations.  Although all these psychic values
and the corresponding damages caused by. pollution are
currently not easily measured, they apparently account for a
significant portion of the total value of pollution control
to society.

In general, estimation of .benefits resulting from
alternative pollution control programs calls for four steps:

     Estimate the amounts of pollutants produced by
     projected economic activity.
                           1-38

-------
     Estimate the remaining discharge of pollutants to the
     environment after imposition of specified control
     measures.

     Estimate the ambient air or water quality that results
     from the diffusion and assimilation of pollutants by
     the environment.

     Estimate the nature and magnitude of resultant reduced
     damages and the corresponding benefits.

The first two steps Involve the projection of a suitable
economic scenario and evaluation of the cost-effectiveness
of various administrative and technological pollution
controls.  The third step requires the use of complex models
of the diffusion and assimilation of specific pollutants.
The lest step relies on the development and interpretation
of dose-effect factors op damage functions, which are
discussed in the next  section.

Finally to the extent  that they were developed for specific
cases, estimates must  eventually be aggregated over the
pollutant/effect combinations, geographic regions, and time
periods of interest.
                           1-39

-------
                  SbPHYSICAL  AND  ECONOMICSR
                     SbDAMAGE FUNCTIONSSR
 A  damage  function  Is  the quantitative expression of a
 relationship  between  exposure  to  specific pollutants and  the
 type  and  extent  of  the associated effect on a  target
 population.   Exposure 1s typically measured 1n terms of
 ambient concentration levels and  their duration, and It may
 be expressed  as  "dosage" or "dose".  The former Is the
 Integral  of  the  function defining the relationship between
 time  and  ambient level to  which the  subject has been
 exposed.   Dose,  on  the other hand, represents  that portion
 of the dosage that  has been instrumental In producing the
 observed  effect  (e.g., the amount  of pollutants actually
 Inhaled In the case of health  effects of air pollution).

 The effect can'become manifest  In a  number of  ways and can
 be expressed  In  either physical and  biological  or economic
 terms.  If the effect is physical  or biological, the
 resultant  relationship Is  known as a physical  or biological
 damage function, or a dose-effect  function.  On the other
.hand, an  economic damage function Is expressed in monetary
 terms.  Economic damage functions can be developed by
 assigning  dollar values to the effects of a physical or
 biological  damage function, or by direct correlation of

                           1-40

-------
economic damages with ambient pollutant levels.   A
representative economic damage function,  showing the
benefits corresponding to a given improvement In
environmental quality. Is presented tn Figure 1.
                           1-41

-------
                         Figure 1.
                       Damage Function
                                    823
                                    82U
in
o
o
LU
                                                   DAMAGE FUNCTION
  THRESHOLD
 WITH
CONTROL
PROGRAM
WITHOUT
CONTROL
PROGRAM
CONCENTRATION OF
POLLUTANT IN THE
AMBIENT ENVIRONMENT
                                                              703

-------
The S-shaped damage function ts rather characteristic of the
relationships between pollutant exposure and resultant
effect.   The lower portion of the curve suggests that, up to
certain pollutant ambient values, known as threshold levels,
there are no measurable damages, while the upper portion
Indicates that there is a saturation level (e.g..  death of
the target population), beyond which Increased pollutant
levels do not produce additional damages.  Between these
segments ts a range where damages are roughly proportional
to the concentratton of pollutants.

In reporting a damage function, one must specify the
pollutant, the dose rate, the effect, and the target
population, or the population at risk.  Dose rate, or the
rate at which ambient concentration varies with time, has a
major influence on the nature and severity of the resultant
effect.   Long-term exposure to relatively low concentrations
of air pollutants may result In manifestations of chronic
disease, characterized by extended duration of development.
delayed detection, and long prevalence.  On the other hand,
short-term exposure to high concentration levels may produce
acute symptoms characterized by quick response and ready
detection.  Characterization of the population at risk 1s
considered in'more detail In subsequent paragraphs of this
discussion.
                           1-43

-------
The two principal techniques for analyzing the relationship
between exposure and effect Indices necessary to construct a
damage function are Known as mult1varlate regression and
nonparametr1c or distribution-free estimation.  Multivarlate
regression Is by far the favored technique because it
provides a rapid Indication of the degree of association
between a large number of Independent and dependent
variables and 1s readl 1y. programmable for'computer
operation.  However, Its validity Is heavily contingent on a
fairly precise a priori  definition of the relationship
between each independent and dependent variable, and on
precise measurement of the Independent variables.   Thus,
this technique Is especially vulnerable to the poor
precision in measurement and reporting of air pollution
levels for a. given segment of population.  Nonparametric
estimating Is free of these assumptions,  but It calls  for
laborious data reduction for each of the many pa'rs  of
Independent and dependent variables, and expert judgement to
guide each step of the process-   Moreover, this technique
requires sufficient data for each Independent variable to
Isolate and remove the Influence of likely Interfering
factors.

The data required to construct damage functions can  be
obtained by the following approaches:
                           1-44

-------
     Epldem1ologlcal  or field studies and observations

     Toxtcological or laboratory Investigations

     Market studies

     Delphi method
                                                       •
     Public opinion surveys

     Legislative decisions

     Litigation surveys.




The first two approaches are attuned to physical  damage

functions, while the remaining ones are directed toward

derivation of.economic relationships.




The first approach involves the comparative examination of

the effects of pollutants on large segments of population

exposed to different  levels of pollution In order to deduce

the nature and magnitude of the likely effect.  Field

studies and observations represent the same approach to

assessment of effects on animals,  vegetation,  and materials,

and they are characterized by similar analytical  techniques

and concerns.  Toxlcologlcal studies Involve deliberate.

administration of controlled doses of pollutants to animal

subjects, followed by observation of the resulting effects.

Laboratory studies represent essentially the same approach

for determining effects of pollutants on plants and

materials.
                           1-45

-------
 Two  considerations  need  to be noted about epidemic log teal



 and  field  studies.   First. It 1s very  important to remove or



 control  the  Influence of  factors other than pollution that



 may  be  responsible  for the different effects observed.   In



 the  cose of  health  effects,  for example, these  include



 physiological,  genetic,  and  other characteristics of  the



 population under  observation, such as age. sex. race, family



 medical  history,  occupational exposure, medical care, state



 of health, and  nutrition.  When these characteristics cannot



 be factored  out.  It  Is frequently assumed that  their



 distribution 1s sufficiently uniform in the populations



 under observation that the basic results are not affected



 significantly.  Secondly, eptdemlological and field studies



 and  observations  can only Indicate an association between



 exposure to  pollution and the observed effect,  though the



 Impact  of  an association can be strengthened considerably



 through evidence  of  consistency and specificity of the



 relationship.   A  causal  relationship can be demonstrated, or



 made plausible  by toxlcologlcal and laboratory  studies, or



 by the  construction  of a plausible connective mechanism.








 Market  studies, such as  those Investigating differences  In



.property value  or Income, employ prices or wages as an'



 Indication of'the values affected by pollution, and their



 usefulness has  been  demonstrated In a number of cases.  This



 approach is  heavily  dependent on the investigator's ability






                           1-46

-------
.to  Identify and Isolate the many other factors that affect
the value of property, or other Indicators used.  In the
Delphi method, the knowledge of a diverse group of experts
1s pooled for the task of quantifying variables that are
either Intangible or shrouded 1n uncertainty.  This method
provides an efficient way to obtain subjective, but
Informed. Judgements.  Thus, m a recent project, the
California Air Resources Board under EPA sponsorship
constructed a number of dose-response functions based on
expert opinions submitted by a group of clinicians and other
health effects researchers.

Surveys of public opinion focus on estimating  tndtvudal
preferences and demands.  Such surveys have been
particularly helpful'In understanding how attitudes about
pollution are formed and affected by changes 1n
environmental quality.  They can also provide an indication
of what people may be willing to pay for enhancement of
environmental quality, or perhaps, what their preference
might be for the reduced risk of experiencing certain
adverse effects.  Surveys of legislative decisions or
litigation awards can also provide some Insight Into the
perceived value of pollution abatement.
                           1-47

-------
                   SbPOPULATION .AT RISKSR
In the past. It Mas customary to assess the severity of air
pollution In terms of point-source emissions,  and later'  in
terms of ambient concentrations.  These Indicators reflected
the progression In the state of the art from visual
assessment of smoke plumes to Increasing availability of  air
quality monitoring stations and associated data processing
capabilities.  However,  the real significance of air
pollution lies In Its physical, economic,  and social Impact
on the affected population.

Beyond this, characterization of the population at risk in
terms of its potential susceptibility to various levels of
air pollution can provide useful Indications for allocation
of resources and setting of priorities in air pollution
abatement.  For example, a higher clean-up priority  could be
assigned to an area containing a large population of older
people or those exposed to high occupational pollution than
to another area with a smaller population of relatively
healthy people not otherwise exposed to harmful pollutants.
This procedure can be refined further through control of
specific pollutants.
                           1-48

-------
Since the Importance of characterizing the population at
risk to various levels of air pollutants became recognized.
there have been several attempts to obtain such a
characterization through crude regional  estimates.   The
first comprehensive, national assessment was only recently
completed.  The major assumptions and findings of this study
are summarized here.

The specific objective of the population at risk study was
to calculate the number of people in selected demographic
and soctoeconomlc classes Mho are exposed to various levels
of several air pollutants.  This was accomplished in six
steps:

     Select air quality Indices
     Select population indices
     Select air quality and population coverage units
  •  Obtain and process air quality data
     Obtain and process census data
  •  Calculate population at risk.

The pollutants selected were total suspended participates,
sulfur dioxide, nitrogen dioxide, carbon monoxide,  and
photochemical'oxldants.  The air quality Indices were
expressed In terms of the relationships of pollutant ambient
levels to their corresponding short-and long-term primary

                           1-49

-------
standards.  They were divided Into four classes:  0-75

percent. 75-100 percent.  100-125 percent,  and above 125

percent of the corresponding primary standard.   In the case

of short-term standards,  the 90th and 99th percent lies of

the observed values were found to be more  useful  Indicators

than the maximum values.



Human susceptibility and resultant response to toxlcologlcal

and physical  stress produced by air pollutants 1s determined

somewhat by certain Intrinsic traits, such as age. race,

sex. and general health,  as well as by such extrinsic

characteristics as employment. Income, educational level.

and general environmental conditions.  The population

classes selected for our study are listed  below:
$t
        Age:                              -   Employment:

        - Under 19 years                    - Manufacturing
        - 20-64 years                       - Other
        - 65 years and over

        Race:                            •   Family income:

        - Whlte                             - Under $5,000
        • Negro                             - $5.000-524,999
        - Other                             - $25.000 and
                                               ovcr$R
Although population Information from the U.S.  Bureau of the

Census 1s available for the entire country,  air quality

data, stored In EPA's National Aerometrlc Data Bank, are
                           1-50

-------
not.  The gaps occur in the fo^m of specific pollutants,  the

short-term or long-term values,  or missing stations.

Consequently, this study dealt with 241  standard

metropolitan statistical areas (SMSAs),  which cover 68.6
                                                       •
percent of the population and 11.0 percent of the land area

of the United States.   Pollutant ambient levels In these

areas were derived by  plotting Isopleths (equal

concentration contours) between air quality monitoring

stations and by superimposing this display over maps of the

SMSAs.  The year of coverage for air quality data was 1973.

though the population  information was based on the 1970 U.S.

census.



Finally, the population at risk was computed within each

pollutant and population class,  and aggregated to state,

regional, and national levels.  The results are displayed In

tables of population versus air quality  classes for

different combinations of pollutants and geographic

locations.  The national aggregations for all five

pollutants are presented In Tables 1 through 5.




The study concluded that the exposure of the U.S. population

surveyed to short-term partlculate, short and long-term

sulfur dioxide, and short-term carbon monoxide levels was

within the respective  permissible primary air quality

standards.  On the other hand, significant portions of the



                           1-51

-------
population surveyed were exposed to excessive long-term
partlculate (31 percent),  long-term nitrogen dioxide (24
percent), and short-term oxldant (58 percent) levels.
                           1-52

-------
                                   t-i        .                             |  101B
                               Table 1.                                  j  1019
Popvl.it ion character ir Total suspend ^articulates                 |
                           (1,000  persons)                               |  1022
                                                                                  1020
Area: United Slates

Air Pollutant: Tola! Suspended Partlculates

Air (Vuality Index: Short Term - 90th percentlle of 24 hour data
             Long Term - Annual geometric mean
Population
Characteristics
A. General
Age: 0-19
20-64
65 and over
Race: White
Negro
All other
B. . Economic
Annual family Income:
(thousands of families)
$0-54,999
SS.OOO-J24.9W
$25. 000 and over
C. Labor Force
Percentage In
manufacturing
D. Total Population

Air Quality Level Classes - jlg/m3
Short Term
< 200

47. 859
68.487
11,644
110,531
14.836
1,672



9,271
25.093
1,876


25.4%
127.990
201-260

2,004
2,717
462
4,261
859
63



265
963
SO


26.6%
5,183
261-320

471
631 '
116
904
301
13



66
224
8


28.5%
1,218
321-450

142
205
44
325
58
8



22
68
3


24.8%
391
>451

105
139
19
214
47
2



13
52
1


19.1%
263
Long Tonn
< 60

19.889
28,434
.4,442
47,564 '
4,537
664



1,768
10,644
954


26.2%
52,765
61-76

10.631
16,684
2.874
25,309
4.553
327



1,313
6.027
408


25.1%
30. 189
76-90

7,054
8.350
1,650
13,763
3,005
286



812
3.220
196


27.2%
17,054
91-120

4,640
6,802
1.195
10.385
1.992
260



621
2.187
141


26.9%
12.637
>121

2. 768
4,006
670
6.443
808
184



324
1,644
111


24.0%
7.435
                               I-
                                                                           I  703

-------
                      .      .•;:• ' •• ••••-»'»'»' i-'i'^X*-"'••-•*V-1'   ••  :  • •                   |  1026
'.•'Ponul.ir.ion Characterized by Sbcioeconomic and Demographic Factors   |  1027
                      Expori-vi to sulfur  Dioxide                          |  1028
                             (1,000 persons)                                i  1029
 Area:  fnltt-'l States
 Air Pollutant: Sulfur Dioxide
 Air Quality Index: Short Term - 90th pcrccntlle of 24 hour data
              Long Term - Annunl arithmetic mean
Population
Characteristics
A. General
Age: 0-19
20-04
65 and over
Race: White
Negro
All other
B. Economic
Annual family Income:
(thousands of families)
$0-$4,999
$5,000-$24,999
$25, 000 and over
C. Labor Force
Percentage In
manufacturing
D. Total Population

Air Quality Level Classes - jjg/m3
Short Term
< 280

•19,243
70,412
12, 174
113,073
16,032
1,755



5,457
25,848
1.932
25.4%
131,869
281-365

62
93
19
161
12
1



6
37
2
43.0%
174
366~I20

2
3
1
3
3
0



1
1
0
14.5%
6
>420















Long Term
<60

37,403
53,534
8,887
83,921
13.133
1,760



4,057
18,394
1,554
25.2%
99,824
61-80

399
566
118
928
145
. 10



43
205
14
23.7%
1,083
' 81-100

167
214
29
383
15
2



13
80
6
26.2%
410
>100

1
1
0
2
0
0



0
0
0
38.9%
*
                                                                                703

-------
                              Table  3.                                |   10 J3
Population  Characterized  by Socioeconomic and  Demographic  Factors   J
                    Exposed  to Carbon Monoxide                      |   1035
                          (1,000 persons)                             j   1036
103<*
Area: United States

Air Pollutant: Carboo Monoxide

Air Quality Index: Short Term - 99th percenttle of one hour data
Population
Characteristic*
A. General
Age: 0-19
20-64
65 and over
Race: White
Negro
AD other
B. Economic
Annual family Income:
(tbouaands of families)
$0-*4,999
$5,000-»24.999
f 25, 000 51

191
292
60
452
47
44



32
84
3

.
31.7%
543
                                                                          703

-------
                             Table U.                                 |  1QUO
Population Characterizel by Socioeconomic  and Demographic Factors   |  10«1
                  Expose;! to  Nitrogen Dioxide                      |  10U2
                         (1,000 persons)                              i  io<*J
 Area: I'nili-d Stales

 Air Pollutant: Nitrogen Dioxide

 Air Qualln Index: Long Term - Annual arithmetic mean
Population
Characteristics
A. General
Age: 0-19'
20-64
65 and over
Race: White
Negro
All other
B. Economic
Annual family Income:
(thousands of families)
$0-84.999
$5,000-524,999
$25, 000 and over
C. Labor Force
Percentage In
manufacturing
D. Total Population

Air Quality Level Classes - pg/m3
<80

6,268
9, 034
1,421
15,024
1.195
504



659
3,302
226


23.1%
16,723
81-100

1,697 .
2,470
472
3,649
966
.24



193
905
63


30.1%
4,639
101-125

2,223
3,374
554
5,364
577
210



243
1.100
104


26.9%
6,151
>126

220
460
90
535
183
62



35
139
21


27.0%
770
                                                                      I  703

-------
                             Table 5.                                 !  10U7
Population Characterized by Socioeconomic and Demographic Factors
                       Exposed to Oxidants                           |  1
                        (1,000 persons)Ss                            j  1050
                                            1CU8
 Area: United State*

 Air Pollutant: Oxtdaati

 Air Quality Index: Short T»rm - 99th perceotlle of <
> hour data
Population
Characteristic*
A. General
Age: 0-19
20-64
65 and over
Race: White
Negro
All other
B. Economic
Annual family Incomes
(thousands of famtlle*)
IO-J4.999
*5.000-»24.999
J25, 000 and over
C. Labor Force
Percentage In
manufacturing
D. ' Total Population

Air Quality Level Claases -j/g/m3
< 120

11,262
15.900
2,630
24,909
4,149
734



1.116
6,911
458


24.0%
29,792
121-160

6,011
7.128
1.205
11,690
1,576
178



561
2.592
167


21.2%
13,444
161-200

.- 11,326
16,945
3,061
26,378
4,540
413



1,357
6,074
530


22.0%
31,333
>200

10,726
15. 30G
2.424
25,008
2,889
560



1,073
5,616
421


21.3%
28,455
                                                                          703

-------
                SbPROBLEMS OF MEASUREMENTSR
Assessment of benefits of pollution control  ts beset by a
number of major difficulties that have a profound effect on
the accuracy and reliability of the benefit  estimates.   Some
of these difficulties can be largely overcome with the  aid
of available ancillary information, while others require the
expenditure of much additional  effort.  Still others must be
dealt with by Indirect estimation and other  Imprecise
techniques.  The more Important problem areas may be listed
as follows:

     Collection of reliable ambient quality  data
     Selection of exposure indices and identification of
     synergtsttc effects
     Selection of representative populations
  -  Measurement of effects
     Establishment of causal relationships
     Presentation of non-quantifiable Information
     Regional, demographic, and temporal extrapolation
  •  Consistent classification of damages
     Double-counting. and omission of damages
     Assessment of damage reductions.     *
                           1-58

-------
Collection of sufficient air and water ambient quality data
requires a very large number of measuring stations and a
commitment to measurement and data handling well  in excess
of the present level.  because the problem concerns numerous
point and nonpolnt sources of pollutants discharging at
Irregular intervals into air and water.  Consequently, the
available data seldom reflect hourly,  or even diurnal
variations that may be Important.

Collection of useful data on damages and their proper
attribution to exposure to specific levels of various
pollutants suffers from several handicaps.  One is the
problem of selecting the proper exposure Index for each
pollutant In terms of level, duration, and presence of other
pollutants, pr Influence of meteorological and nydrologlcal
factors.  Another Is the need to select sample populations
that are representative of the population at large In  terms
of susceptibility to detectable levels of damage.   In  the
case of health effects, this involves segregation  based on
demographic and socloeconomlc makeup of the population at
risk.      .

A third difficulty lies in measuring the resultant effects.
This Is especially problematic In the case of psychic
damages, such as those associated with health, recreation,
aesthetics, option, and preservation values.  Such damages

                           1-59

-------
are not adequately assigned costs by the market system
because they are aspects of environmental use that are not
owned privately or exchanged.   Thus, estimation of the
corresponding benefits requires development of proxy or
surrogate measures.

The fourth and most formidable problem Involves Identifying
and documenting a causal relationship between exposure to a
given dose and production of a specific effect and deriving
the corresponding damage function.   The existing literature
contains estimates for only a  few discrete points on the
many damage functions of Interest.   In order to produce
national benefit estimates. It Is frequently necessary to
make major assumptions about the shape of the damage curve
on the basis of these few points.

Most studies leading to the evaluation of damages resulting
from exposure to various pollutants address a specific
geographic area, population, and time frame.  Extension to
the national level and a more  recent time frame requires
extrapolations of ambient levels, population at risk.
personal Income, and Increases In costs of resultant damages
due to Inflation.  The classification of damages, for  which
the data are collected. Is often dictated by availability of
sources and analytical expediency,  rather than a uniform and
self•consistent framework.  Consequently, different studies

                           1-60

-------
evaluate damages that are not necessarily additive OP even
comparable, and any effort to reconcile or aggregate the
results of such studies must apply careful Interpretive
techniques to prevent gross overlaps or omissions of damage
estimates.  Moreover, In aggregating such fractional
results, tt Is not currently possible to reflect the
potential Impacts of changes In one pollutant or one region
on the damages caused by other pollutants or 1n other
regions, nor has tt been possible to reflect the Impact of
the general adjustments the economy would make to pollution
control programs and the resulting reduction In damages.

Finally, with effective abatement, the estimate of benefits
associated with a given level of pollution control can be
expressed In terms of the corresponding reduction of
damages.  This step. In turn, requires the definition of a
quantitative relationship between reduced emissions and
resultant ambient levels, as well as between these Improved
ambient levels and reduced damages.  Development of
pollutant transport and dispersion models describing the
first set of relationships has been only partly successful
because of the many 111-defined variables Involved.  Thus,
It Is commonly assumed that the fractional decrease in
ambient levels 1S essentially proportional to the fractional
reduction of emissions.  The second set of relationships Is
defined by the damage functions discussed earlier.  The unit

                           1-61

-------
damages obtained from a damage curve are converted to total
damages through multiplication by the number of units at
risk and the cost-per-untt damage,  as appropriate.

Thus, assessment of benefits associated with a given level
of pollution control is still  most  assuredly an art,  which
permits divergent interpretation of available data that may
lead to widely differing results.

For this reason, although certain studies on air and water
pollution damages are cited in Section Two and Three.
national aggregate damage estimates are not presented in
this report.
                           1-62

-------
SaChapter 3
Pollution Control Cost Reduction
Through Process ChangeSR
                      $bINTRODUCTION$R
Opportunities for air and water abatement cost reduction
through process change were identified for 40 industries.
Five Industries were examined In detail:  copper,  aluminum,
pulp and paper, petroleum, and Inorganic  chemicals.   Using
Reference Case abatement costs developed  in Sections II  and
III as a baseline, the extent of reduction achleveable
through specific process change candidates In each industry
was determined.  The relative savings In  accumulated capital
expenditures through 1985 In the five Industries  are:  14.5
percent. 9.6 percent, 10.1 percent.  12.0  percent,  and 2.5
percent.  The analogous savings In annual)zed costs  for  1985
are: 35.0 percent. 11.0 percent, 28.5 percent. 24.0  percent,
and 25.0 percent.  When these savings are assessed 1n terms
of their applicability to opportunities In the other 35
Industries, the total capital and annual!zed cost  reductions
for all 40 industries relative to reference case  abatement
costs are estimated to be (In minion dollars and  percentage
reductions):  $197, (-1.2 percent) and $211. (-9.9  percent).

                           1-63

-------
 $bImpact  of  Process  Change  Upon
 the  Cost  of  a  Clean  EnvlronmentSR

 Pollution control  legislation and associated effluent
 guidelines require that  Industry attain specific  levels of
 pollutant control.   The  mechanism for achieving  these  levels
 1s  left to the discretion of each Industry.  The  simple
 approach  is  to add treatment steps  to the process at the
 points of waste emission, which are  termed end-of-pipe
 control.   The  costs  associated with  these end-of-pipe  (EOP)
 steps furnish  an economic motive for waste-reduction process
 changes.   If a net abatement cost reduction can be achieved
 through process change relative to  that process or a
 competing process  employing end-of-p1pe treatment, an
 incentive for  process change exists.  This concept is
 evidenced in the generic types of process change  in the more
 advanced  standards (BAT. BPT, NSPS).  For example, process
 changes designed to  reduce  water requirements, permit
 greater water  reuse, and minimize leaks and spills are
 Included  in  the compliance  strategies recommended In the EPA
 effluent  guideline development documents; considerable
 evidence  exists to indicate such potential.  Exemplary
.plants In many industries do operate at much higher
 efficiencies than  the corresponding  typical plants, and
 plant modernizations have been able  to substantially Improve
 abatement efficiency at  a reasonable cost.  In this

                            1-64

-------
discussion, emphasis ts placed upon assessing the cost
reductions achievable through process changes other than
those not Included In the Reference Case of Section Four.

A number of Important distinctions must be made.   There are
Important differences between what can be achieved in a new
plant as compared with the upgrading of an existing
facility.  in some Instances. It is less costly to abandon
an ex 1st tng fact 11ty and but Id a new one than 11  is to
convert the older facility.   In such a case,  nearly all
capital associated with the abandoned plant must  be
forfeited.  When conversion of the existing facility Is
reasonable, the capability to do so may be unevenly spread
across the Industry.  The larger firms have both  greater
technological capability and financial reserves than the
smaller firms.  Thus, even a technologically-feasible
retrofit process change may have considerable economic
Impact.

Such economic considerations are well  Known.   They are
restated here to emphasize their Importance-In assessing
process change as a method of reducing end-of-ptpe treatment
requirements.  A final  general comment of this type pertains
to tax considerations.   If a tax benefit Is granted EOP-type
Investment and not those related to process change, there is
an Incentive to pursue the former course.

                           1-65

-------
This discussion Identifies the type of savings that may be
achieved through process change.   The estimates made are
Intended to be Indicative rather  than exact;  I.e..  the
analysts objective Is to establish reasonable bounds between
                                                       •
which the Impact of process change can be evaluated.  The
reference cases for comparison are the Industry costs
established In Sections Two and Three of this report.  The
measure of the economic benefit from the process change Is
the extent to which the pollution control savings relative
to the Reference Case exceed the  costs incurred tn trie
process change.  The Industry-wide savings are .derived by
Identifying the extent of Industry acceptance of the
designated process change.
SbEffect of Environmental Standards
on the Rate of
Process ChangeSR

In considering the effect of environmental regulations on
Industry's acceptance of process change, ft must be
remembered that this relationship takes place within the
framework of Industry's overall Investment decisions.  Most
Industries have a tacitly expressed, minimum acceptable rate
of return.  Below this level, Investment is not believed to
enhance a company's financial position, and other

                           1-66

-------
const derations, such as liquidity,  may predominate.   Whether
or-not sufficiently lucrative opportunities exist often
depends on the Investment climate,  which In turn rcay be
heavily Influenced by Interest rates,  current market
behavior,  etc.  Even under favorable Investment conditions,
corporations .have limited capital  resources.   Consequently,
they must  select among investment  options,  seeking the
opportunity most likely to bring a high, reliable return on
venture capital.                    :

Comparison of Investment opportunities Is conducted on the
basis of comparative profitability.  A piece of equipment,
like a furnace. Mill process a given product throughput over
a specified lifetime.   The value of this production, based
on projected prices. Is compared against the capital outlays
required to build and operate the unit; ancillary costs and
benefits must be Included In this comparison.  An existing
furnace has an established set of  operating specifications:
energy requirements, recovery efficiency, etc.  If the
challenging process can reduce energy needs,  the operating
savings that result are Included In the profitability
comparison.

In addition, an attempt should be made to assess the
"venture risk" involved In the Investment:  an example Is a
shoe manufacturer's Investment in a line of ski coots.  The

                           1-67

-------
Investor understands that an unseasonally warm winter might
cut his sales prospects In half.   This estimate of risk 1s
taken Into account In determining the desirable rate of
return.   Venture risk similarly applies to the introduction
of new processes, where the firm takes a risk that the
process win not live up to expectations.

In a highly competitive market, the costs associated with
end-of-ptpe control may be so high that firms cannot pass
them on as higher prices without  losing competitiveness.
These plants must either develop alternative control
strategies that can be Implemented at an acceptable level of
cost, or close their doors,  in these cases, tn-plant
controls can truly be said to be environmentally Inspired.

However, environmental regulations can also Indirectly
affect Investment decisions by altering the profitability of
certain options.  Existing facilities will have additional
capital  and operating costs associated with end-of-pfpe
treatment of its wastes, assuming compliance with
environmental standards.  In-plant changes that reduce
treatment costs will be treated like any other benefit In
profitability calculations.

Abatement costs can affect the process trends that would
have developed In the absence of environmental

                           1-68

-------
considerations In a variety of ways.   The additional  cost

can tip the scales In favor of a project that was formerly

less profitable.   Alternatively, It can further Improve the

profitability of  an already preferred Investment
                                                       •
opportunity, thereby accelerating Its rate of acceptance  by

the Industry.  It Is important to realize that in both these

events the environmental  regulations are only one of  several

motivating factors; the abatement savings are not usually

sufficient to Justify investment unless other advantages  are

gained as well.   This fact becomes relevant when allocating

the portion of cost savings attributed to the environmental

regulations.



On the other hand, environmental Investments do Involve one

special circumstance that vitally Increases their

Importance.  Traditional  decisions on an investment,  such as

capacity expansion, offer a firm three choices: expansion

using proven technology,  expansion using a challenging

process, or no expansion.  By law. abatement decisions do

not permit the third path of Inaction to be taken; either an

alternative abatement strategy must be found, or the present

plant abated through end-of-ptpe methods.  Furthermore.

expenditures on equipment with the sole function of control

yield no direct economic return to the corporations.

Consequently, firms may be receptive to strategics that can
                           1-69

-------
attain abatement objectives while In some way Improving the
processing efficiency of the plant.

Before proceeding, two cases should be noted In which
environmental regulations do not affect general process
trends.  The first case Is where little difference exists
between the abatement costs for the two processes.  If e
relatively new process Is only marginally more profitable
after subtracting venture risk than the established
technology, most companies will retain the proven profit-
maker.  This Is Important In the present discussion because
the time frame In which alternatives to EOP treatment can be
undertaken Is very short.  In the second case, a process
that has to pay much higher abatement costs may remain more
profitable than its competitor.  In this case, the "dirtier"
process will continue to be substituted for the "cleaner"
process.  This process change will have the effect of
Increasing total Industry abatement costs.
SbTypes of Process ChangeSR

A survey was conducted within 29 polluting Industries to
Identify those process changes that have significant
pollution treatment Implications.  Three major categories of
process change were found: material changes, process

                           1-70

-------
modifteat Ion,  and process substitutions.   An additional  and
Important type of change exists that,  while not associated
with a specific process, affects the control costs for each
process.  These are plant-wide changes,  such as
housekeeping,  coordinated water usage by a set of processes
to achieve a net reduction In water usage, etc.  In the
following discussion,  plant-wide changes are addressed In
terms of their effect  on Individual processes.
ScMATERIAL CHANGESSR

Material changes Include modifying the nature or quality of
raw materials employed or adjusting the specifications of
the product produced.   For example, use of natural  Trona as
a source of sodium carbonate obviates the large quantities
of waste generated by Solvay process synthesis of sodium
carbonate from salt and limestone.  Likewise, use of rutlle
rather than llmenlte in the production of titanium dioxide
significantly reduces waste quantities.  Alternatively,
synthetic rutlle can be generated by pretreatment of
llmenlte.  Recycled or secondary material Inputs are also
Important.  For example. Increased aluminum recycling
circumvents the waste produced during bauxite processing.
An example of a product specification change Is the
                           1-71

-------
Incorporation of a portion of process waste sludge In paper
products not requiring high brightness.

Often, material changes are made on the basis of economic
                                                       •
considerations related to materials availability.   For
example, domestic bauxite Is of lower quality than bauxite
Imported from Jamaica. Surinam, or Australia: hence,  the
majority of bauxite consumed In the United States each year
Is Imported.  However, If the countries of origin are able
to establish a higher bauxite price, the domestic
alternative will appear more desirable.   Such a. change In
material Input will affect the nature and Increase the
quantity of wastes generated.  Another example relates to
the use of ruttle In titanium dioxide production,  as already
discussed.  Ruttle. possessing a higher tltanta content.  1s
predominantly Imported from Australia, while large
quantities of tlmenlte ore exist In the United States.  An
adjustment could be made In the event rutlle became either
hard to obtain or highly priced.  Again, the nature of the
waste stream would change.

Crude oil quality also varies with  Its point of geographic
origin.  For example. Middle Eastern crudes have a higher
sulfur content than domestic crudes, and the percentage
usage of the former is increasing.  Meanwhile, restrictions
on the sulfur content of fuel oil for consumer use require

                           1-72

-------
that the refinery,  which Is now dealing with additional
sulfur In its primary raw materials,  reduce the sulfur
content in Its final  product.   This change in product
specifications directly affects the amount of processing
required, and. hence, the pollution-control related costs.

Environmental considerations are only one of the factors
Impinging upon the selection of raw material type.
Nevertheless, the nature of the raw material utilized  can
have a direct effect  upon the costs of pollution abatement.
ScPROCESS MOOIFICATIONSSR

Three types of process modification were Identified:  revised
process operation, byproduct recovery, and process-specific
waste treatment.

  1. Revised Process Operation.  This category includes
     those process modifications made In an effort to
     Improve process economics.  The principal attribute Is
     that In some way the efflcency of the central reaction
     Is Improved, I.e.. greater quantities of the desired
     products'and lower quantities of pollution are
     generated per unit of Input material.  This may be
     accomplished by changing the temperature or pressure of

                           1-73

-------
   the reaction,  extending or shortening the residence
   time,  improving reactant mixing.  Introducing a more
   stable catalyst,  Increasing recycle quantities.
   reducing water use.  or invoking real-time computer
   control.  In some cases, optimal  process operation when
   pollution control 1s required will  differ from that
   when such control Is not required.   Usually, a complex
   linear programming scheme Is required to balance the
   many factors involved In Identifying .the optimal
   performance, and this determination Is strongly
   affected by the character of the Input materials,  as
   previously discussed.

2. Byproduct Recovery.   The recovery of a salable material
   from the process waste stream Is an obvious and often
   mentioned method of  simultaneously  reducing the waste
   load and at least partially compensating for the costs
   Involved.  However,  the opportunity to profitably sell
   such recovered materials Is sometimes elusive.  An
   extreme example Is the recovery of  sulfur and Its
   various compounds as pollution control.  The
   marketplace may be unable to accommodate the quantities
   of sulfur to be made available.  Hence, extraction of
   sulfur from the air  and water waste stream could merely
   serve to transform the sulfur into  a more readily-
   controlled solid waste.  Attempts are underway to

                         1-74

-------
     expand the market for sulfur compounds by Identifying
     new applications, but there may be limits to the amount
     of market expansion possible.

     In many cases, recovered material can be1put to
     profitable use.  Frequently, the application Is an 1n-
     plant use of the recovered material to perform a
     function that previously required a purchased Input,
     (e.g., heat and fiber reuse in the paper Industry).   In
     addition. Industrial complexes are beginning to
     cooperate In using each others waste streams when a
     desired attribute is present.

  3. Process- Spec 1fIc Treatment.  This process modification
     Is the treatment of process waste prior to merging it
     with the waste streams of other processes for end-of-
     plpe treatment.  In general, the process waste must
     have some specific attribute that necessitates a unique
     treatment step; otherwise, the economies of scale
     associated with end-of-plpe treatment prevail.

Process Substitution. Process substitution Is differentiated
from process modification 1n that a fundamental change Is
made to the central reaction step.  For example, going from
the mercury cell to the diaphragm cell In chlorine
production and from the open-hearth to the basic-oxygen

                           1-75

-------
furnace In steel production are process substitutions.   For
comparison, changing the reaction conditions,  enlarging the
reactor, or adding ancillary process equipment are process
modifications.

Process 'substitutions are an extremely Important process
change category in terms of their effect upon pollution
control requirements.  A recent study of solid waste
generation  showed that for 17 of the 34 largest producers
of process solid waste among industrial chemicals, a process
substitution was underway or had already taken place.   In
each case, the amount of solid waste generated was reduced.
As process efficiencies are improved, the yield of the main
product goes up and the quantity of waste generated
correspondingly goes down.  In addition, the remaining
wastes tended to be easier to treat.  Usually, wastes
associated with the raw materials can be segregated with
comparative ease.  The ones produced during the principal
reaction, however, are generally closely associated with the
main product, and hence, are more difficult to separate.
                           1-76

-------
                  SbCOSTING METHODOLOGYSR
The Industry Survey analysis,  which appears later In this
discussion, disclosed a number of promising process change
opportunities.  These opportunities were evaluated to
determine the extent to which  such process changes can be
expected to reduce pollution control costs relative to the
Reference Case, primarily an.end-of-p1pe approach.  To do
this,  five representative Industries were selected that
together Illustrate the various modes of process change.
Specific procass change candidates, ranging from the
modification of a single processing step to replacement of
an entire process, were examined.  for each challenging and
defending process, total unit  costs (process + end-of-pipe)
were calculated.  The capital  requirements and annualIzed
costs of the changed operation were compared with costs
developed for the Reference Case discussed 1n Section Four.
SbCostlng at the Unit LevelSR

A new process may be related to existing operations In one
of three ways:
                           1-77

-------
  1. It can be basically Interchangeable with part of  the
     existing plant,  with potential  for both retrofit  and
     new plant applications.   (Examples: Continuous and
     batch digesters, oxygen  paper processes, flash and
     reverberatory furnaces.)

  2. It can be basically Incompatible with In-place
     facilities and represent an alternative for new
     capacity only.  (Examples:  Hydrometallurgy, dry forming
     of paper.)

  3. It can be basically additive In nature, with no unit
     Serving a comparable function In the present process
     scheme.  (Examples: Byproduct recovery units, spin
     containment systems.)

Each of these relationships calls for a different type of
comparison of baste process costs.  Table 1 diagrammatically
represents the basis for comparison 1n each of these
situations.
                           1-78

-------
 $t
                           Table  1.
              Nature  of  Process Cost Comparisons
                                    Relevant Cost  Parameters

1 .
2.
3.

Inter-
changeable
Processes
Al ternat 1 ve
Processes
Add! ttve
Processes

Retrofit
Appl teat ion
New Plant
Appl i cat Ion


Old vs
Process
06M
Capital .
OSM
Capital.
OSM
None
New
Process
Capital
OSM
Capital
OSM
Cap) tal
OSM
Capl tal
O&MSR
 Values for  capital, operating and maintenance  (OSM) and
 annual(zed  costs  were obtained  from available  engineering
 cost  estimates.   Capital  costs  represent  the Installed costs
 of  process  equipment; this  figure Includes actual component
 costs plus  expenditures  for engineering plans,  site
 preparation,  and  construction of necessary auxiliary
 facilities.   Startup costs  and  penalties  for plant shut-down
 time  have not been  Included, because  these values tend to be
 very  plant  specific.  The operating and maintenance category
'Includes: materials, taxes  and  Insurance, direct and
.Indirect  labor, and maintenance.  AnnuaHzed costs are
 defined as  OSM costs + depreciation on capita)  Investment
 (calculated at 10 percent of the unpaid principal per year
 and normalized over the  capital  lifetime.)  All costs are
                            1-79

-------
developed for specific plant configurations,  or model
plants.  Where competing units exhibit different economies
of scale, more than one model  size was used.

Sources of process cost estimates Included technical
journals, EPA economic Impact  studies, and other government
publications, such as Bureau of Mines Information Circulars.
The available materials frequently had to be  converted to a
form applicable to cost comparison at the unit level.   In
some cases,  simplifying assumptions were employed.   For
example:

     Operating and maintenance figures are frequently
     available only at the plant level.   In these Instances,
     allocations between processes were constructed on the
     basis of Information contained 1n the source
     literature.  In the copper Industry, for example,
     operating costs were provided for a typical smelter.
     For some of the operating expense Items, such as
     electric power, chemicals, etc., the significant  1n-
     plant users were delineated; costs could therefore be
     attributed to those specific sources.  For materials
     where detailed Information was not available,  and for
     general expenditures (labor costs,  maintenance),  costs
     were distributed according to the fraction of total
     capital Investment represented by each unit process.

                           1-80

-------
     for some units, estimates of capital  and O&M

     requirements are simply not available.   This is

     particularly true for old defending process

     technologies, like the open-hearth steel furnace,  where
                                                       •
     the last new unit of its type was built many years ago.

     Cost estimates for these processes were related

     directly to estimates obtained for challenging

     processes.   The comparison between hydrotreating and

     drying and sweetening, included in the  representative

     industry evaluation of petroleum refining,  is a case In

     point.  Operating costs for drying and  sweetening can

     be expected to be lower than those attributed to a

     hydrotreatlng unit, due to the large hydrogen

     requirements of the latter process.  Where  operational

     differences could be clearly Indicated  in this manner.

     costs were estimated In accordance with these

     deviations; otherwise, costs were presumed  to be

     roughly comparable.
$bEnd-of-Pipe CostsSR



The reference case for abatement costs is the set of costs

provided for each industrial  sector in Sections Two and

Three of this report.  These estimates were developed for

current and projected plant inventories using treatment cost


                           1-81

-------
curves which are Included In Appendix F.   This material  was
supplemented by information obtained from EPA development
documents, technical and trade journals,  and other recent
studies on the costs of pollution control.   To make this
data base responsive to the specific needs of the process
change Investigation, methods had to be devised for the
allocation of reference case costs between specific unit
processes, and the translation of waste load reductions
possible through process change Into a revised estimate  of
end-of-plpe costs.
ScALLOCATlON OF REFERENCE CASE COSTSSR

Much of the Information concerning abatement costs has been
developed only at the plant level, while process changes
frequently affect a single phase of the production process.
Where this dichotomy exists,  some technique for apportioning
treatment costs among the processes within a plant Is
necessary.  The demands on this allocation method Increase
with the complexity of the control problem.

In the simplest case, each piece of pollution control '
equipment in the treatment scheme can be associated with the
abatement of a particular pollutant generated at a single
source within the plant: e.g.. a baghouse for control of

                           1-62

-------
participates from process A.  and a wet scrubber for control
of sutfuroxtde from process B.   In this Instance,  the only
Information required for cost allocation Is the breakdown of
total abatement costs Into the expenditures required for
each control component.

More often, however, a pollutant is generated at a number of
sources within a plant.   In the case of copper smelting,
sulfur oxide off-gases,  are produced In various proportions
during each of the major processing steps (roasting.
furnaclng, or conversion).  Some,  or all. of these streams
may be combined and sent to the same treatment sequence.  If
a control device handles wastes from several plant sources,
some portion of the related costs of control should be
assigned to each of these process sources on the basis of
the fraction of total pollutant loading each contributes.
To calculate these fractions, emissions factors that
establish a general ratio between pollution and output must
be obtained for each relevant process.  These factors, when
multiplied by the model  plant unit capacity, provide an
estimate of plant waste loads.   If roasting Is found to
contribute 55 percent of plant sulfur oxides. It is presumed
that It can be assigned 55 percent of the reference case
costs Incurred In controlling that waste stream by means of
a scrubber, acid plant,  etc.   This assumed one-to-one
correspondence is not entirely accurate, due to the fact

                           1-83

-------
that wastes classified In the same general  pollutant



category (TSS. participates) can have widely-divergent



strengths and treatabl11t1es.  Nonetheless,  the relationship



ts a generally accepted  rule of thumb which has been



employed In other recent  abatement cost studies.







In a single plant, many  different types of  pollutants are



generated,  and must be controlled by the same abatement



facilities..  Ideally, some portion of the total costs should



be allocated to each of  the pollutants removed by the



treatment system.  Formulas of thts type have been developed



for the Inorganic and organic chemicals Industry.  There



were serious limitations, however. In the application of



this type of analysis to the representative industry



examples.  Detailed breakdowns of model plant waste loads to



the subprocess level are only available for a limited number



of pollutants.  Similarly, references on waste reductions



resulting from process changes often confined their



discussion to one or two major parameters.   Consequently, It



was frequently necessary to designate one pollutant as the



dominant concern of Industry abatement standards-  In the



petroleum Industry, for  example, BOD removal was concluded



to be the compelling force behind BPT standards:  costs for



Installation of the required biological treatment systems



were therefore allocated between the various 1n-plant



sources of that pollutant.





                           1-84

-------
 $CWASTE  REDUCTIONS  AND  REVISED
 ABATEMENT  COSTSSR

 The  relationship between  the  reduction  in waste  load and  the
 reduction  in  treatment  costs  is  not proportional-  A 10
 percent  dlmunltton  of plant wastes might result  In a 5. 8,
 or even  12 percent  savings on control expenditures,
 depending  on  factors  like .the economies of scale  Involved,
 the  degree to which control systems are modular,  etc.  Two
 approaches were utilized  to determine the cost reduction
 associated with a given level of waste  reduction.  Where
 Information estimating  this relationship was provided  In  the
 literature, this material was employed.  An example of this
 type of  Information Is  the study by McGovern  on  waste
 reduction  in  the petroleum Industry.  In the absence of
 specific analysis,  end-of-plpe cost savings were  measured by
 moving down the treatment cost curves which are  Included  In
 Appendix F. to a facility size consistent with the waste
 load reduction achieved.  The difference between  this
 revised  value and reference case costs  represents the
 savings.   After, the revised level of end-of-plpe
 expenditures  Is determined, allocation  of these costs among
.unit processes is again undertaken in the manner  outlined
 above.
                            1-85

-------
The example of substituting hydrotreatIng for drying and
sweetening can be used as an 11 lustration of these two
procedures.  A model plant configuration was chosen  which
Included drying and sweetening.   Using BOD as a surrogate
Indicator, the contribution of this process to the total
refinery waste burden was 45 percent.   This fraction was
then applied to the estimated total for plant end-cf-plpe
expenditures, to determine the costs attributable to drying
and sweetening.  For the same plant, waste loads were
recalculated, utilizing lower polluting hydrotreatfrig
processes In place of drying and sweetening.  The resulting
reduction In waste (42 percent)  was converted Into Its
equivalent effect on end-of-plpe costs (23 percent), using
materials generated by the McGovern study.  The percentage
of new total BOO coming from hydrotreatIng was calculated,
with this fraction applied to the revised cost estimate.
SbEconomlc and Environmental
Motivations for Process Change
and the Allocation of Cost EffectsSR

In addition to indicating the substitution potential  of new
processing concepts and the pollution control cost savings
resulting from their Implementation, the unit cost
comparisons can serve as a basis for speculation about the

                           1-86

-------
motivating force behind a process change decision.   In some



cases, e.g.. spill containment In the paper Industry.



process changes are adopted that provide no economic return



on Investment, the only benefit being a reduction in eno-of-



plpe costs.  Changes of this type can truly be sa'd to be



environmentally Inspired.  Therefore, the costs for



Installing and operating the containment system should be



charged to pollution control.   Conversely, some concepts.



like the Bayer-Alcoa aluminum process,  have processing



advantages that are sufficient to Insure their adoption



before end-of-ptpe savings are taKen into account.   An



approximate line of demarcation beyond which process changes



are economically motivated Is an Industry's minimum



acceptable rate of return.  Since pollution control savings



are Incidental to the decision maker In cases providing



greater rates of return, 1t Is Inappropriate to attribute



these costs to pollution control.








In between these two clear cases lies a substantial gray



area.  Recovery and sale of byproduct H2S and NH3 1n a large



petroleum refinery  results In a return of about 3.6 percent



a year; this profit margin would not In Itself be sufficient



to justify the Investment.  However, when environmental



savings are Included and the revised treatment system is



contrasted with a pure end-of-p1pe approach, the process
                           1-87

-------
 becomes  very  desirable.   It would be  logical to charge only
 part  of  the process change costs to pollution control.

 This  concept,  although  Important to recognize, can not be
 accurately  implemented  given  the present data base.  Minimum
 acceptable rates of return vary by several percentage points
 among companies  In the  same Industry.  A more detailed
 analysis of Industry  Is required to delineate these
 variances.  Similarly,  there  is a degree of  Inaccuracy In
 the estimations of process cost effects.  Even a slight
 error can negate the  accuracy of a carefully-constructed
 allocation algorithm.   Since  only a few of the changes
 examined In the representative  Industry studies lay  in this
 gray  area, none of the  savings  In basic process costs
 stemming from process change  were Included in the estimates
 of control cost reductions.   It should be emphasized.
 however, that the resulting estimates represent the  lower
 boundary of possible  savings.
 SbCostlng  at  the  Industry  leve'SR

.Even  though a particular process change may be shown  to be
 economically  profitable on  the basts of the-unit  level
 comparison, the opportunities for  Its application may not be
 fully exploited.   It  Is necessary  to establish the  industry

                            1-88

-------
context Into which process change variables are Introduced



because certain characteristics of the Industry environment



will constrain or encourage adoption of new process ideas.



Table 2 presents a partial representative list of  elements



In the contextual picture that were examined for their



possible Influence on the rate of penetration.  These



limiting factors can be physical or financial, and not all



of these factors are applicable to each Industry considered



In the representative evaluations..
                           1-89

-------
     Factor Considered

1 •    Industry growth rate

2.
                                        SdTable  2.
                     Some Factors  Affecting  Process  Change  Potential

                                       Reason  for  Consideration
Rate of plant obsolescence/
replacement
3.   Availability of Input materials
     (Ex.  -  higher grade ores,  low
     sulfur  fuels, rutlle)

4.   Availability of markets for
     recovered byproducts.

5.   Industry attitude toward process
     change
6.   Size distribution of plants
7.   In-place end-of-plpe
     abatement facilities
6.   Availability of capital
Taken together,  these estimations measure the amount of
new capacity being built.  If the process change being
considered is an option for new plants only,  the
possibilities for penetration are highly controlled by
these variables.

External (outside the Industry) market conditions
frequently constrain the ability of the Industry to
employ particular options.  This Is especially true
of raw material  changes and byproduct recovery
and sale.
                                       The historical  recept1veness of  the  Industry  to  new
                                       process  Ideas is a general  Indicator  of  the  time
                                       frame  required  for the Industry  to Implement  new
                                       methods  on a  large scale.

                                       The profitability of  process change  Is  frequently
                                       linked to economies of scale.

                                       For a  plant with an already Installed treatment  system
                                       capable  of meeting environmental  standards,  the  utility
                                       of  Installing process  change measures designed  to  reduce
                                       waste  load Is greatly  diminished.

                                       If  a particular process change requires  a  large  Initial
                                       capital  investment.  Its application may  be restricted to
                                       those  firms with higher profit margins  and favorable
                                       liquidity position.$s
                           1-90

-------
 Because  the  possibilities  for process substitution  In a
 given  Industry are dependent on  the complex  Interactions of
 several  variables, a scenario approach was utilized to
 Indicate the range of possible results.  Two basic scenarios
 were defined: a maximum, and a best-guess estimate of
 process  change penetration.  In  the aluminum industry, for
 example,  the best-guess market share for the Alcoa smelting
 process  In 1985 Is 8 percent of  primary aluminum capacity;
 If maximum penetration  is  assumed, the share increases to 12
 percent.  The difference between  these scenarios 1s the
 assumption of price or other constraints on  the availability
 of bauxite and energy inputs.  In some cases, the maximum
 and best-guess penetrations are  equivalent.

 For each alternative, pollution  control costs as modified by
 process  change were calculated and compared against the
 reference case estimates.  To aggregate costs to the
 Industry level, a size distribution of existing and future
 plants was estimated.  The cost  studies In Sections Two and
 Three  of this report assign existing facilities 1n the plant
 Inventory to various size  classes.  For future growth, plant
 capacities were developed  from Information on Known
.expansion plans and extrapolation of recent  size trends.
                            1-91

-------
                    SblNDUSTRY SURVEYSR
This section summarizes the results of an Industry survey to



Identify those process change opportunities having



Implications for pollution control costs.  Each Industry



considered in the cost studies of Sections Two and Three In



this report were assessed to determine the answers to two



questions:  Is the Industry a significant contributor to air



and water pollution, and are there opportunities for process



change that can reduce the total cost of abatement?  By



comparing estimates of current Industry effluent levels with



corresponding national totals, a general measure of



significance was developed.  If an industry contributed more



than 1 percent of the national total  for any major pollutant



parameter,  it was considered to be a significant polluter.



In the case of air emissions, this analysis was supplemented



by comparing industry abatement costs to total abatement



expenditures.  Additionally, sectors were judged significant



If they were responsible for highly toxic emissions



(mercury, asbestos,  etc.)  that pose special abatement



problems.







If the answer to the first questions was affirmative, the



Industry was further Investigated for process change



potential.   Trade journals and other magazines. EPA






                           1-92

-------
development documents,  previous Cost of Clean Air and Water
reports, and other reports on the subject of Industrial
pollution control formed the base from which the survey
results were developed.   A process change was considered a
viable alternative only If It had at least been tested at
the pilot plant level.

The results of the Industry survey are summarized m Table
3. with additional Information provided In the industry
profiles presented In Appendix F of this report.  All
process changes discussed In these profiles have been
classified according to the type of process change Involved.
the media affected,  and whether or not the change was
Included as part of  the reference case abatement strategy.
This material is presented in Table 1A of Appendix F.
                           1-93

-------
                          Table 3.
                 Summary of Survey Results
                             $t
                                         Pollution Reduction
                       Significant       Potential Through
Industry Category      Polluter?         Process Change?

Fossil Fuels Group

  Coal Cleaning           	                  	
  Natural Gas
    Processing            —                  	
  Petroleum Refining      A,W                    W
  Steam Electric
    .Power                 A.W                  A

Foods Group

  Feedlots                  W                    W
  Meat Products
    Processing              W                    w
  Dat ry Products
    Processing              W                    W
  Seafood Processing        W                  No
  Canned & Frozen
    Fruits and
    Vegetables              W                    w
  Feed Mills              A                    NO
  Grain HandlIng          A                    No
  Beet Sugar                W                  No
  Cane Sugar                W                  No
  Fertilizer/
    Phosphates            —                  —

Construction Materials
  Group

  Cement                  A                    No
  Lime                    ••'                  "'
•  Asphalt                 A                    No
  Asbestos                A.W                  No
  InsulatIon
    Fiberglass            •••                  ...

Metals Group  •

  Aluminum                A.W                  A.W$R
                            1-94

-------
                    Table 3.  (Continued)
                 Summary of Survey Results
                             $t
                                      StPollutton Reduction
                    Significant       Potential  Through
Industry Category   Polluter?   .      Process Change?
Metals Group
 (con't)

  Copper               A                    A
  Iron and Steel       A.W                  A.W
  Lead                 A                    No
  Zinc                 ---                  ---
  Other Non-
    Ferrous Metals     A.W                  No
  Electroplating         W                    W

Chemicals Group

  Inorganic
    Chemicals          A.W                    W
  Organic
    Chemicals            W                  No
  Mtscellaneous
    Chemicals          —                  —
  Plastics &
    Synthetics           W                  No

Consumer Product
 Inputs Group

  Timber Products
    Processing         —                  —
  Pulp & Paper
    Mills              A.W                    W
  ButIders Paper
    and Board
    Mills              •--.
  Textlles               W                    W
  Soaps and
    Detergents         —                  —
  Leather Tanning        W       .           No
  Glass                —                  —
  Rubber               —                  —

Consumer and
 Government Services
                           1-95

-------
GroupSR
                          1-96

-------
                    Table 3.  (Continued)
                 Summary of Survey Results
$t
  Dry Cleaning         A                .    A
  Municipal
    Sol id Waste
    Disposal           A                    A
  Sewage Systems       —                  —

Key:  A-Atr: W-Water.

 Sectors are listed If they either pay more than 1X of total
national abatement expenditures ,  or generate more than 1%
of the national total  of parttculates.  hydrocarbons.  502,
NOX, BOD. COD. TSS. or oils and greases.

 Sectors generating highly toxic emissions.

 Sectors found to be nonsignificant polluters were not
Investigated further.$R
For several reasons, process Ideas now being considered will

not exert the same degree of Influence over an industry's

future planning.  Some processes,  though  promising 1n

theory, may encounter operational  difficulties that

substantlal1y.reduce currently anticipated economic

benefits: other changes may be restricted In application to

plants of a certain type, size, or age.  Therefore, twenty-

two "candidates" for further study were selected from the

Initial list of opportunities as best prospects  for

Implementation within the time frame and  at a level where

they could seriously influence the abatement cost outlook

for an Industry.  These changes are categories by type of

process change and by Industrial sector 1n Table 4.
                           1-97

-------
Industry

Municipal
  Refuse
  Disposal

Paper
Industrial
  Chemicals

Paints

Petroleum
  Refining

Iron and
  Steel

Copper

A1 utnl num
Electric
  Utilities

Dry Cleaning
                                      SdTable 4, Sheet 1 of 3.
                      Process Change Opportunities by Industry and Type Change
                                        A. Materials Changes
      Raw Materials

Old Process    New Process
Salt/LI me
   IImeni te
Trona
   Rut lie
Solvent Base   powder Base
Sulflde Ores   Oxide Ores
Bauxlte
High Grade
Bauxlte
Recycled
Aluminum
Low Grade
Bauxite
                        Pre.treatment

               Old Process    New Process
               Ilmenlte
                              Released
                              Fines
                              Stack Gas
                                 Scrubbing
Synthetic
   Rut 11e
                              Pellet
                              Agglomeration
                                                                             Product Specification

                                                                          Old Process    New Process
                                                            High Bright-
                                                            ness
                                                            Lower
                                                            Brightness
                                                            Downstream
                                                            Control
                                                            Low Sulfur
                                                            Fuel Oil
                              Fuel Desulfur-
                                 Izatlon
                           1-98

-------
Industry

Municipal Refuse
  Disposal

Paper
Industrial
  Chemicals

Paints

Petroleum
  Refining
Iron and
  Steel

Copper
A1umt num

Electroplating

Electric
  Utltitles

Dry Cleaning

Text)les
Fruits &
  Vegetables
                                       Table 4, Sheet 2 of 3.
                      Process Change Opportunities by Industry and Type Change
                                      B.  Process Modifications
   Byproduct Recovery

Old Process       New Process
Disposal
01sposal
Disposal
Disposal
D1sposa1
Sawdust for Pulping
Flber/Chem/Heat
Sulfur/NH3
                  SulfuHc Acid
Chemicals Recycling
Recovery of grease
from wool scouring/PVA
Rec1amat1 on/Latex
Recovery

Sol Ids Recovery
Revised Process Operations

Old Process       New Process




Batch Digesting   Continuous Digesting
Barometric Con-
densers 1n Vac.
Dlstt1latlon
                        Reverberatory
                        Furnace
Once Through
Surface Condensers
                  Flash/Electric
                  Furnace/Hydrometal •
                  lurgy
                  Counterflow Washing
Water Recycle
                           1-99

-------
                                       Table 4.  Sheet 3 of 3.
                      Process Change Opportunities by Industry and Type Change
                                      C.  Process Substitutions
Industry

Municipal Refuse
Disposal

Paper
Industrial
Chemicals
Paints

Petroleum Refining.


Iron and
Steel

Copper

Aluminum


Electroplating

Electric
Utilities

Dry Cleaning

Text)les

Fruits & Vegetables
Old Process

Incineration


Kraft Process
Wet Forming

C12: Mercury Cel1
Na2C03: Solvay Process
T102: Sulfate Process

Solvent Suspension

Catalytic Cracking
Open-Hearth/Electric-Arc
Blast Furnace
Hal 1 Process
Bayer-Hall Process
Petroleum Solvents
Mechanical Peeling
New Process

Landfi1 Is/Mlnef11 Is


Dry Forming
Diaphragm Cell Trona Process,
Chloride Process
Electrostatic Suspension

      Hydrotreat1ng.
      Hydnocracklng

Basic-Oxygen Furnace,
Direct Reduction
Alcoa Process,
Monochlorlde Process
Synthetic Solvents
Dry Caustic PeellngSs
                           1-100

-------
At this point, five Industries were selected for In-depth
study: copper, aluminum, pulp and paper,  petroleum refining.
and inorganic chemicals.  These Industries were chosen
because they were Industries in which'two or more process
changes are concurrently being considered, and collectively,
they contained examples of all the major  types of process
change.  Additionally,  It was felt that the data base of
process change Information in these areas was rich enough to
permit detalled analysts.  Short summaries of these
representative evaluations are provided in the next section.
The full text containing the bulk of the  assumptions,
calculations, and documentation that form the basis for
these conclusions Is presented In Appendix F.
           SbREPRESENTATWE INDUSTRY EVALUATIONS

CopperSR

The main environmental problem facing the copper industry Is.
the control of sulfur dioxide contained In the off-gases
from reverberatory furnaces used In primary smelting
operations.  Because of the very weak concentration of these
gases (usually less than 1  percent sulfur dioxide by
volume), they cannot be treated effectively through

                           1-101

-------
conversion Into sulfuric acid.  .The costs of abatement are
consequently very substantial: expenditures on control
measures In • recent year, for example, represented 22
percent of total Capital Investment.   As a result, U.S.
producers have greatly increased their Interest 1n
processing Innovations that have the potential to reduce the
Industry's control burden.  Research efforts have been
directed In support of three main process alternatives:
flash furnaces, electric furnaces, and hydrometallurglcal
smelting.  The first American commercial  scale example of
each technology has either been installed within the past 5
years or Is currently under construction.
ScPROCESS CHANQESSR

Plash smelting is a commercially proven technology that has
been employed extensively in Europe and Japan for over a
decade.  Off-gases from the furnace attain sulfur dioxide
concentrations of 10-14 percent and can be easily handled by
an acid plant.  By combining treatment of all plant
emissions in a single facility, a 1.500-ton of concentrates
per day smelter can achieve an estimated 11 percent
reduction In capital requirements, and a 27.2 percent
reduction In the annualized costs of pollution control.
Although process costs for the flash furnace are somewhat

                           1-102

-------
higher due to additional slag processing requirements.
overall unit costs figure to be 10-20 percent less than
those estimated for a reverberatory furnace of comparable
size.  On this basis. It Is projected that up to 50 percent
of new pyrometallurglcal capacity requirements in 1975-80,
and 75 percent In 1981-85,  will be supplied by flash
smelters.  If recently proposed new source performance
standards   are promulgated which would specify more
stringent and much more costly controls on reverberatory
furnaces, the rate of penetration by the challenging
technology will be further  accelerated.

Electric furnaces claim a dual advantage over their
reverberatory counterparts; they Increase the sulfur dioxide
concentration of off-gases  by eliminating combustion gases
within the furnace, and they exhibit a higher thermal
efficiency.  Two existing U.S. smelters have already made
the switch to this technology as part of their abatement
strategy.  The smelting site must be close to a source  of
cheap electric power If the process Is to be economically
competitive.  This fact alone will seriously restrict
application of this technology In some of the remote and
arid Western mining areas.   In addition. Industry spokesmen
have frequently expressed doubts   about the operating
reliability of electric furnaces.  Consequently, the option
Is viewed as a less preferred alternative, with Its

                           1-103

-------
substitution possibilities limited to areas where the cost
of power Is low enough to override other concerns.

Two hydrometallurgtcal smelting techniques, the Arbiter and
Cy-Met processes, are (n advanced stages of development.
Major questions affecting evaluation of the substitution
potential of these concepts concern the time frame In which
successful  scale-up can occur, and the extent to which
current process cost estimates will accurately represent
commercial  scale results-  If the operating economics
achieved during pilot plant operations can be maintained,
hydrometallurgy can reduce annual process costs by up to 25
percent; in addition, pollution control costs are
practically zero, requiring only some form of storage or
disposal for the sol fate solid waste which Is produced.
Even after  successful scale-up, substitution will proceed
slowly; hydrometal1urgy will  constitute no more than 4
percent total  primary capacity by 1980, and 12 percent by
1985.

In addition to these basic process changes, It Is necessary
to assess the market opportunities for sale of the byproduct
sulfurlc acid generated during the control process.   If
there are profitable opportunities present, some of  the
estimated costs of pollution control can be defrayed:
contrarlly, if no opportunities exist, the costs of

                           1-104

-------
neutralization and disposal  of the byproduct  should be
counted as an additional  abatement expense.   Competition for
markets will be very strong, and smelter acids face one
major disadvantage by being far from their primary users.
                                                       •
However, smelters can take advantage of opportunities within
the Industry to use H2S04 as a leaching agent to extract
copper from oxide ores and mine tailings:    they can also
Increase their marketability by selling acid  at a price well
betow the going market rate.  Based on these  parameters.
four possible price/market opportunity scenarios were
examined.  In.the combination of circumstances deemed most
likely to occur, it was assumed that 12 of the primary
smelters with acid plants will be able to sell their add at
an average price well below market rate, resulting m
revenue of over $30 million per year.
Sblndustry EffectsSR

The overall reduction In pollution control  costs resulting
from Implementation of the process changes  discussed above
Is summarized In Table 5.   The bulk of the  savings attained
through 1980 Is the result of byproduct acid sales;  the
major Increase 1n savings estimated for 1985 1s attributable
to greater application of flash and hydrometallurglcal
smelting technologies.

                           1-105

-------
                          Table 5.
             Copper Industry -- Abatement Cost
              Reduction Through Process Change

               (In Millions of 1975 Dollars)
                             $t

                       Reference Case    Revised Abatement
                       Abatement Costs   Costs (with Process
                                         Change)

                       1980     1985     1980     1985

Cumulative Investment  1.405.1  1.433.8  1.292.6  1.174.2
(from 1972)                                (-8.OX)  (-18.1%)

Annual Costs             420.1    423.4    354.6    311.2
                                          (-15.6%)  (-26.5X)

  From Scenario 2 - air control costs only.SR
JbAlumlnumSR



The two pollutants of primary concern to the aluminum

industry are red mud from the refining of bauxite,  and

fluorides from the reduction of alumina to aluminum.   Red

mud Is usually Impounded In an evaporation pond,  and  it  Is

thus possible to achieve zero discharge.   Fluoride Is

associated with the Hall reduction process and Is about  70

percent controlled to date.  Existing facilities  may  have to

Install expensive secondary roof scrubbers to achieve the

proposed standards of 90 percent capture.
                           1-106

-------
A new source performance standard of 95.5 percent removal is
achievable by the Alcoa Dry Scrubbing Process   but used m
conjunction with vertical stud soderberg aluminum reduction
cells.  Other types of cells will require expensive
secondary scrubbers.  Thus, pollution control factors are
prompting consideration of alternative technologies.
ScPROCESS CHANGESSR

Three process substitutions may have an effect on pollution
control costs In the aluminum Industry.  The most direct
factor Mould be an Increase In the capacity to recycle scrap
aluminum.  Substitution of the Bayer-Alcoa   process for the
Bayer-Hall would decrease the unit pollution control costs
from primary smelting by 73 percent.  Non-electrolytic
processes, like the Monochlorlde   process would probably
Increase the unit cost of pollution control by 13 percent.
Such a technology might be considered In the future because
of energy and bauxite constraints.
                           1-107

-------
$CINDUSTRY-WIDE COST REDUCTIONSR

The penetration of new technologies Is related to the growth
rate of the Industry,  which in turn Is related to the
                                                       •
Industry's pollution control  cost.   The absence of
constraints on raw material availability or pollution output
tends to preserve the present technology.   Moderately-
constrained growth tends to encourage the search for less-
costly alternatives.  Three scenarios based on these growth-
penetration assumptions are presented 1n Appendix F.
However, for purposes of comparison, a 7 percent growth
scenario with moderate penetration  of recycling and the
Bayer-Alcoa process Is presented here.

The costs resulting from a Bayer-Hal1/Bayer-Alcoa/recyclIng
mix of 77 percent/1 percent/22 percent In 1980 and 68
percent/ 8 percent/24 percent mix In 1985 are shown in Table
6: note the lower capital and annual)zed operating figures
for the process change case.   The large Increase 1n savings
from 2 percent In 1980 to 11.3 percent In 1985 is due to  the
Increased coverage of recycling and the Bayer-Alcoa process.
Other scenarios with different growth and technology
coverage assumptions can be found In Appendix F.
                           1-108

-------
                           Table 6.
      Aluminum  Industry  Abatement Cost Reduction Through
                        Process Change
                (In Millions of 1975 Dollars)
                             $t

                  Reference Case    '   Revised Abatement Cost
                  Abatement Cost       (with process change)

                  1980      1985            1980      1985

 Cumulative        2.357.4   2.431.4         2.314.9   2.197.9
   Investment                               (-1.8X)  (-9.6X)
   (from 1972)

 Annual             787.5    728.6          716.6    663.0
   Costs                                   (-9.0%)  (-9.0%)

   From  Scenario 2 - air and water control costs.$R


 SbPulp  and  Paper industrySR
 The  paper  Industry discharged 2.47 billion gallons of water

 In 1972. even  though  It was  recycled over three times during

 processing.    About 60 percent of that water was used in

 direct  process contact, higher than any other  Industrial

 activity.   This leads to a discharge of about  2.2 million

 tons per year  each of BOO and of suspended solids.   The

 Industry spent 30 percent of  Its capital  Investment, the

 largest percentage of any manufacturing industry. In an

'effort  to  meet pollution control standards.
                            1-109

-------
ScPROCESS CHANGESSR

The pulp and paper Industry has several  short-term and
several long-term water pollution control  savings
opportunities through process change.   In  the short-term
(1975-1980), process modifications and product
specifications changes can have a significant effect.
Process modifications designed to contain  spills,  recover
fiber, process chemical and energy have some savings
Involved.   They range from a 20 percent savings per ton to a
65 percent savings per ton where applicable.   The increased
use of lower brightness papers can result  In a  67 percent
saving In-pollution control costs where applicable.
Unfortunately, the applicability of these  changes Is limited
to moderately old plants and the industrial  tissue market,
BO that the overall savings potential  Is decreased.

The long-term (1980-1985) process substitutions of oxygen
processes and dry forming, appear to have  a  substantial
effect on the cost of pollution control.  ,     The use of
oxygen for bleaching, waste treatment, and process liquor
recovery result In a 53 percent savings in water pollution
control costs.  Dry forming of paper eliminates water
pollution control costs where applicable.   These process
substitutions appear to have a wide range  of applicability.
but are limited to new capacity Implementation.

                           1-110

-------
ScINDUSTRY-WIDE COST REDUCTIONSR



If It Is assumed that 30 percent existing capacity and all

of the new capacity before 1980 will  take advantage of the
                                                       •
near-term savings,  and that 50 percent of new capacity after

1980 wt11 take advantage of the long-term savings, the paper

Industry can achieve the aggregate water pollution control

costs savings shown in Table 7.
                          Table 7.
      Abatement Cost Reduction Through Process Change
       Pulp and Paper Industry. Water Pollution Costs
               (In Millions of 1975 Dollars)
                             $t
                 Reference Case
                 Abatement Cost
                  Revised Abatement  Cost
                  (with process change)
Cumulative
  Investment
  (from 1972)

Annualized
  Cost
  (by year)
1980     1985     1980        1985
2.869.9  5.905.5        2,656.7  5.055.1
                    (-7.4X)     (-14.4X)
  606.0  1.386.9
       499.3  1.006.9
(-17.6X)     (-27.
  From Scenario 2 -  water control  costs only.SR
                           1-111

-------
 SbPetroleum  ReflntngSR

 The  petroleum  Industry has made a number of  tn-p'ant
 Improvements In  the past designed to  Improve water effluent
 characteristics  and Increase water  reuse and recycle rates.
 These efforts  have been fruitful, with the water reuse ratio
 In the  Industry  almost doubling In  the last 20 years:
 nonetheless, refineries face substantial future outlays  for
 pollution control systems.  In-plant  process changes
 designed to  minimize end-of-plpe treatment requirements  are
 likely  to be a major part of the overall abatement strategy
 selected.
 ScPROCESS CHANGESSR

 Many proposed  changes affect operations at  the subprocess
 level, and can achieve  substantial  reductions In plant waste
 loadings for a fairly small  Initial capital outlay.  An
 example of this  type of process modification is the  recovery
 of phenols produced during catalytic cracking.  Removal of
 this pollutant can reduce total p'lant  BOO by 7 percent and
.end of pipe costs by. 5  percent.  Additionally, there are
 economic advantages arising  from the recovery of free oils
 entrained In the wastewaters from  the  cracker.  Analysis of
 the effects of Installing such a unit  In three model

                           1-112

-------
refinery configurations  .     Indicates that this change
could be profitable for a group of refineries comprising 65
percent of current total capacity.

Recovery of byproduct sulphur and ammonia from refinery sour
waters has been a widely practiced technique In recent
years, and Is recommended In the EPA Development Document
as part of BPT abatement strategy.  Analysis In this section
of the study focused on estimation of the cost-offsetting
benefits achievable through sales of recovered materials.
Available process cost data on typical  stripping and
recovery facilities   demonstrates a potential for returns
on Investment of up to 20 percent per year,  provided that
all byproduct can be sold.   For both sulphur and ammonia, a
detailed analysis (refer to Appendix F) was  made of market
conditions: and an assessment given of  the competitive
opportunities available to refinery producers.  Results of
this Investigation indicated that sales of the ammonia and
sulphur generated at current production levels could
translate Into revenues of  $62 and $50  million.
respectively, provided that maximum sour water recovery was
practiced using dual stage  stripping techniques.  In
addition, maximum processing resulted In 45  percent
reductions in typical refinery BOD loadings, with a
corresponding pollution control savings of 25 percent.


                           1-113

-------
Greater use of hydrocracKIng has often been suggested as a
way to reduce atr and Mater pollution problems resulting
from catalytic cracking operations.  Although hydrocracklng
units offer greater operational flexibility and Increased
                                                       •
product yields In addition to reducing pollution problems.
Industry adoption of the process since its development In
the 1960's has been very cautious.   The major obstacle to
Implementation has been the higher costs associated with the
challengtnlng processes: this gap has recently widened due
to sharp increases In hydrocrack1ng Input prices-   As a
result, a great deal of effort has been funneled Into
modification and Improvement of the defending process.
Major developments Include use of new catalysts requiring
less frequent regeneration, and the Installation of carbon
monoxide waste heat boilers.  These recent events Indicate a
resurgence of expansion to catalytic cracking, with a
resulting increase In end-of-plpe requirements and costs.

The use of hydroprocesstng techniques has been rapidly
Increasing over the past decade, growing at an average of  B
percent per year.  The addition of hydro-desulfurizatIon
steps to refinery operations reduces the waste burden of
sulphur, nitrogen, and metals requiring end-of-plpe
treatment, and concentrates these const 1tutents In sour
water streams which can be readliy processed for byproduct
recovery.  In other areas of refinery, hydrotreating

                           1-114

-------
processing can replace older,  dirtier processes like acid
treating, or drying and sweetening.   Although the impetus
for greater use of the processes is still  strong,  there are
definite limitations on further extension  of these processes
(n refineries which have already exhausted their inplant
hydrogen surplus, since hydrogen production facilities are
an expensive capital cost Item.   Further  penetration by
this process Is likely to occur at a slower rate.
ScINDUSTRY-WIDE COST EFFECTSSR

It was very difficult to quantify the pollution cost savings
possible In the petroleum refining sector.   If all  process
changes discussed In this chapter were implemented  in a
specific refinery, waste load reductions of up to 60 percent
could be achieved.  There are many limitations restricting
the substitution possibilities which exist; and, given the
diverse structure of the Industry, It was hard to determine
the number of plants that were actually constrained.
Nonetheless, tt is believed that these various concepts
could be Introduced at a level sufficient to reduce average
waste loadings of BOD by 20 percent.   This corresponds to
about 12 percent reduction in end-of-pipe capital and O&M
costs.  Additional revenue 1s added from byproduct  recovery.
These estimates are summarized In Table 8.

                           1-115

-------
                          Table 8.
    Petroleum Refining-Abatement Cost Reduction Through
                       Process Change
               (In Millions of 1975 Dollars)
                             $t
                 Reference Case    Revised Abatement Costs
                 Abatement Costs   (with process change)

                 1980     1985     1980        1985

Cumulative       938.8    1,898.5  826.1       1.557.1
  Investment                      (-12.0%)      (-12.OX)
  (from 1972)

Annual Costs     204.8      412.4   128.4        274.6(-
                                   37.3X)       (-33.4X)

  From Scenario 2 - water control costs only.SR
Sblnorganic Chem1c»ls$R



Chemical and. allied products rank first In Industrial  water

consumption, with Inorganics accounting for over one-fifth

of this use.   The vast majority (72.3 percent) of water

Intake by Inorganic chemicals 1s for cooling,  with only 11.1

percent used as process water.   The principal  wastes are

Inorganic salts Including chlorides, sulfates, carbonates,

etc; other significant wastes Include ore tailings and

metals, such as chromium, mercury,  lead and Iron-   In EPA's

evaluation of water-borne pollution from 25 major

Inorganics, over 99 percent of  the  waste load was  attributed

to five products: sodium chloride (38.3 percent),  sodium

carbonate (35.6 percent), titanium  dioxide (17.1 percent).


                           1-116

-------
and the coproducts chlorine/ sodium hydroxide (8-5
percent).   Each of these large waste products was evaluated
for process change potential.
ScSODIUM CHLORIOESR

Sodium chloride waste Is usually deep-we 1 led or stored,  and
does not pose a difficult water pollution  problem.
ScSODIUM CARBONATESR

There are two manufacturing processes for sodium carbonate
(or soda ash).  The older Solvay process synthesizes sodium
carbonate from salt and limestone, with ammonia serving as a
chemical Intermediary.   Approximately 1.5 kilograms of
dissolved solid wastes are generated per kilogram of
product.   The dissolved solids are about two-thirds calcium
chloride, with the remainder mainly unreacted salt.  The
solids have slight market value and are usually discharged
Into surrounding water bodies.  In contrast, the newer
process utilizes natural ore, called Trona,  or lake brines
containing burkelte.  Neither of these alternatives
generates a troublesome waste, since ore tailings and brine
wastes can be returned to the mine or lake.

                           1-117

-------
The Solvay process has been steadily losing ground.   No
Solvay plants have been built since 1935.   From 1960 to 1972
Solvay plan participation In soda ash production declined
from 85 percent to 58 percent.   The one advantage still  held
by the Solvay plants Is geographic location.   The Trona and
lake brine deposits are concetrated 1n Wyoming and
California, whereas market concentrations  lie In the East.
As a result, the natural ores have only gradually displaced
the Solvay plants: pollution control requirements promise to
speed this displacement.  Partially due to such
considerations, two Solvay plants closed between 1972 and
1974. further reducing process participation  to 46 percent.
The extent of Solvay process participation Is the principal
factor determining the aggregate water pollution control
cost for sodium carbonate production.  The anticipated
Closing of two of the smaller plants by 1977  will cut Solvay
capacity by one-quarter, and reduce abatement capital and
annuallzed costs by 28 percent (BPT and BAT costs are the
same for this product).

Another important consideration Is whether to recover a
portion of the waste calcium chloride for  byproduct sale.
Assuming there Is a sufficient market, recovery and sale of
20 percent of'the calcium chloride would lead to an 81
percent reduction In annuallzed costs, but would necessitate
a 206 percent Increase In capital requirements.

                           1-118

-------
ScTITANIUM DIOXIDESR

There is competition both among processes and raw materials
for the production of titanium dioxide.   The older,  sulfate
process utilizes a more abundant,  less pure ore.  called
llmentte.  Until recently, the newer chloride process has
been restricted to the use of the  purer rutlle ore.   Since
the reserve of the latter Is quite limited. 20 to 25 years
at present consumption rates, raw  material costs  have played
a large role In process selection.  In spite of the  rutlle
constraint, process efficiencies achieved with the chloride
process have enabled It to increase Its production share to
46 percent since Its Introduction  In the mld-1950's.  No new
sulfate plants have been built since 1956.

Recent sharp Increases In rutlle and chlorine prices have
tended to slow the encroachment of the chloride process.
However, envl ronmen'tal considerations are lending a  new
competitive edge to the chloride process.  The sulfate
process generates 4 to 5 times the amount of waste per
kilogram as compared with only 1.2 times for the  chloride
process.  .    A significant aspect of the difference In
waste load Is the use of a purer raw material by  the
chloride process.  The sulfate waste is mainly spent
sulfuric acid and ferrous sulfate  (copperas).  The waste
from the chloride process is primarily ferric chloride.

                           1-119

-------
Abatement capital requirements for the chloride process are
only 56 percent of those for the sulfate process for BPT,
and 65 percent for BAT.   Similarly,  annual!zed costs for the
chloride process are 40 percent of those for the sulfate
                                                       •
process for BPT, and 59 percent for  BAT.

Byproduct recovery Is an Important aspect of the pollution
control opportunities for titanium dioxide.   Ferric chloride
from the chloride process.Is already being recovered and
sold for water treatment by some companies,  and can
a1ternat1vely.be converted to chlorine for recycling and to
Iron oxide for sale.  Sulfate process waste acid can either
                                                  t
be recovered and recycled or converted to gypsum,  and then
sold.  Acid recovery and recycle in  the sulfate process
alone enables a 22.4 percent reduction In the total  titanium
dioxide accumulated capital expenditures for abatement
through 1985. and a 23.2 percent reduction In annuallzed
abatement costs In 1985.
ScCHlORINESR

Environmental considerations have acted to reverse an
ongoing shift among process alternatives for chlorine
production.  Worldwide usage of the mercury cell
electrolysis process for chlorine substantially exceeds that

                           1-120

-------
of the competing diaphragm cell;  in the United States,  the
latter has always been predominant.  Nevertheless,  mercury
cell participation In U.S. chlorine manufacture had been on
the rise, increasing from 4.3 percent of production m  1946.
to 28.6 percent In 1968.   At that point, concern regarding
mercury emissions to the environment surfaced.  Since then,
some existing plants have converted from mercury cells  to
diaphragm ceils, and little new mercury cell  capacity is
betng added.  By 1973, mercury cell participation had
declined to 24.6.

The wastes from the mercury and diaphragm cells are similar:
brine Impunities,  unreacted salt, weak caustic, waste
sulfurlc acid, sodium hydrochlorate and sodium bicarbonate.
However, the mercury cell waste also contains a limited
quantity of mercury.  The need for strict control of the
mercury content causes significant abatement  cost
differences between the two cell-types.  The  diaphragm  cell
abatement capital  requirements for BPT and BAT are only 13
percent and 36.4-percent, respectively, of those for the
mercury cell.  Likewise,  the annual)zed capital cost
comparison Is 25.7 percent and 44.9 percent for BPT and
DAT.   As a result, the ongoing shift from the mercury  cell.
If no new mercury cell plants are built, will reduce the
accumulated capital expenditures through 1985 by 16.4
percent, relative to the reference case, and  1985 annualized

                           1-121

-------
costs for pollution control  by 12.8 percent.   It should be
noted that a great deal  of developmental  work Is underway to
bring mercury cell control costs into line with those of the
diaphragm eel 1.
ScINDUSTRY-WIOE COST REDUCTIONSR

The Industry-wide Implications of the process change
opportunities for the four chemicals, sodium carbonate.
titanium dioxide, and chlorine/caustic,  are presented In
Table .9.  The four chemicals account for more than half  the
abatement capital and annual)zed cost' requirements for the
entire Industry.  Presuming other chemicals have similar
process change opportunities,  a 38.1 percent reduction tn
abatement annuallzed costs In 1980 can be achieved and a
25.0 percent reduction In 1985.  A slight (2.5 percent)
reduction can be made tn cumulative capital expenditures.
                           1-122

-------
                          Table -9.
        Inorganic Chemicals Abatement Cost Reduction
                   Through Process  Change
               (In Millions of 1975 Dollars)
                             $t
                 Reference Case
                 Abatement Cost
Cumulative
  Investment
  (from 1972)

Annual Costs
  Cost (by year)
1980

419.4




166.7
1985

  727.4



  219.0
                     Revised Abatement Cost
                     (with process chapge)
 1980

 419.3
  (OX)
                                                  1985
709.2
(-2.5%)
 103.2      164.3
(-38.1%)  (-25.0%)
  From Scenario 2 -  water control  costs only.

  •The revised abatement costs were scaled from the costs
  determined for sodium carbonate, titanium dioxide,  and
  chlorine.   These three chemicals account for between 50%
  and 70% of the Reference Case values for capital and
  annuallzed costs.$R
                           1-123

-------
                     SbGENERALlZATIONS

Range of Pollution
Control SavtngsSR

The range of pollution control  savings through process
modifications varies among industry and category types of
process change.  This variation Is to be expected if  one
considers the specific Implementation limits on any given
process change.  Financial,  technical, and physical
constraints to process change vary considerably between
Industries and within each Industry.  The highly-specific
nature of process changes and the varied nature of the
Industrial climate In which  they are Imbedded inhibits
generalIzat1on.

Substantial savings have been demonstrated In the
representative Industry studies.  These savings vary
considerably from industry to Industry as shown in Table 10.
On the capital side, they range from a savings of 2.5
percent In inorganics to 14.5 percent In copper.  The
annuallzed savings are somewhat larger than capital savings,
ranging '.from 11 percent In the aluminum Industry to 30'
percent In the petroleum industry.  The advantages accrued
through process change within the representative Industry
studies may serve as an Indication of the range of potential

                           1-124

-------
savings In a similar situation In another Industry.   It  1s
worthwhile to emphasize the approximative nature of  the
following generalizations;  they are made to facilitate
estimation of the overall  effects of process change,  and
they do not represent precise assessments of the situation
in a given Industry.
                           1-125

-------
                                            Sellable 10.
                           Abatement  Cost  Impact  of Process Change -  1985
                (% Change from Reference Case:  Cumulative Cap)taI/Annual)zed Cost)
Process Industry
Change (Media)
• Materials Change
Raw Materials
Copper
(Air)
• •
--
Aluminum
(A1r/Watei
-S.2/-3.3
Recycl mg
  Product Spedf1 cat ton
-  Process
  Modification

  Revised Operations
Byproduct
Recovery
- Process Substl•
  tut Ion
0/-21.8
Sulfur1c
Acid
                       Hydro-      Alcoa
                       metallurgy/
                       flash
                       furnace/
                       electric
                       furnace
                                               Pulp & Paper
                                               •0.1/-0.3
Reduced
brightness

-5.0/-21.5
                        Spi 11
                        Conta i nment
                                          Petroleum      Inorganics
                                          (Water)        (Water)

                                                         -2.2/-2.1

                                                         Trona (Na2C03)
                                                         Rut lie (T102)
-7.0/-25.0
                  Phenol
                  Recycle
Flber/Chem/Heat   Sulfur/Ammon
-14.5/-13.1 -B.3/-7.7   -5.0/-7.0
                        Oxygen
                        processlng
                                   Monochlo-
                                   ride
Industry Totals
                        Dry Forming


•14.5/-34.9 -9.5/-11.0  -10.1/-28.8
•M3.9/-9.9
               Calcium chloride
               (Na2C03)
               Sulfurlc acid (T102)
                  -5.0/-5.0      -14.2/-13.0
                  Hydrotreatlng  Chloride process
                                 (T102)
                  Hydrocracklng  Diaphragm Cell (C12)


                  -12.0/-30.0    -2.5/-25.0
  Mean 9.72/25.9$s
                           1-126

-------
 SbVarlattons  Within
 Process  Change  TypesSR
 The  variation  across  Industries  within a process change
 category  is  quite pronounced.
 ScMATERULS  CHANGESSR

 The  least  variation and  least -sizable  savings appears  In  the
 Materials  Change  category.   Paper  specification changes are
 limited  to a 0.1  percent  savings  In  total  Industry capital,
 and  a  0.3  percent  savings In industry  annual)zed costs
 because  of limited market acceptability.   Sodium carbonate
 and  titanium dioxide raw  material  changes  account for  a
 cap)tal/annualIzed cost  savings of 2.4 percent/2.1 percent.
 respectively because of  low  profit rr-arglns (low  change
 Incentive).   Increased aluminum recycling  (scrap as a  raw
 material)  results In a 3.2 percent and 3.3 percent savings
•In capital and  annualIzed costs,  respectively.  These
 relatively low  savings are due primarily to supply
 constraints  on  consumer  scrap and  the  quality limitations of
 secondary  aluminum which  limit recycling penetration.
                            1-127

-------
In considering other Industry material  change opportunities.
metallic ores provide varied cases of pollution control  cost
Impact.  Oxide ores of copper may be leached using acid  from
the acid plant at a copper smelter.   For such process
change, an overall decrease (n pollution control on the
order of 2 percent for the whole Industry might be expected.
In contrast, the substitution of lower grade bauxite or
alunlte.tn aluminum production as higher grade ores become
expensive Mill probably Increase total  industry pollution
control costs by about 2 percent.  Ilmenlte processed to
synthetic rutlle (Inorganics) may cause a slightly Increased
(1 percent) pollution control cost.   Product specification
changes toward low-sulfur-petroleum-derived products will
also tend to Increase the cost of the cleanup m the
petroleum Industry.
ScPROCESS MODIFICATIONSSR

Opportunities for both revised process operations and
byproduct recovery were evaluated In the five industry
studies.  There were two examples of revised operations
examined quantitatively: spill containment (paper) and
phenol recycle (petroleum refining).  Both examples affected
process efficiency primarily by reducing the waste load and
water use associated with process operations, while their

                           1-128

-------
effects on product yields are negligible.   These changes can

be usually implemented for small  capital  outlays,  but their

savings potential 1s small as well. In the range of 2 to 5

percent of capital and annual!zed abatement costs.  Other
                                                       •
changes of this type likely to achieve similar savings

Include: the use of counterflow washing 1n textile

manufacturing, the increased recycling of water used In the

processing of fruits and vegetables., and the Installation of

surface condensers In petroleum refinery vacuum stills..




Other operational modifications may have a greater effect on

product yield.  Improvement of the catalysts used In the

cracking of petroleum,  for example. Increases product yield

by 13 percent while simultaneously reducing process wastes.

Changes of this type seem likely to achieve somewhat greater

control cost savings than those modifications discussed

above. If only because the Increased economic benefits will

make the change attractive to a greater portion of the

Industry.




Byproduct recovery opportunities were found to exist In four

of the five Industries studied.  Estimated savings 1n

annual)zed abatement costs from application of these

recovery processes range from 9.9 percent to 23 percent.

The savings In capital  costs, on the other hand, are very

slight, and In the case of acid recovery during titanium



                           1-129

-------
 dioxide  manufacture  (organic chemicals), addition of
 byproduct  processes  substantially  increases total capital
 requlrements.

 Byproduct  options can be divided Into two basic categories:
 those  products  which are reused within  the recovering plant,
 and  those  products which are sold  1n the competitive
 marketplace.  Examples  In  the  first category  Include the
 recovery of  heat, fiber, chemicals  (paper, and sulfurlc add
 (Inorganic chemicals).  These  byproducts reduce the need for
 virgin Input materials  In  the  process.  The savings
 potential  of such measures Is  primarily dependent on whether
 the  material recovered  Is  a significant operating expense  .
 Item for the plant.  The latter category Is represented 1n
 the  Industry evaluations by the recovery of sulfurlc acid
 (copper),  sulfur and ammonia (petroleum refining), and
 calcium  chloride (inorganic chemicals).  Market
 considerations  are the  controlling  factor In  the
 determination of the cost  savings  achievable  through
 Implementation  of these processes.  Primary questions that
•are  of concern: Whether the supply-demand situation can
 accommodate  the new  Influx of  supply, or whether contrarlly.
 market opportunities are limited;  and whether byproduct
 producers  can Increase  their market share by  selling their
 goods  at prices below tha  prevailing market prices.  A third
                            1-130

-------
factor restricting savings potential  can be a low unit-price
for the recovered material.

Based on the factors discussed above,  estimates have been
made of the savings potential  of byproduct recovery in other
surveyed Industries.  In the first category, recycle of the
chemicals used In electroplating should reduce plant
materials costs substantially, with annual savings In the 16
to 23 percent range possible.   In contrast, three byproduct
recovery operations In the textile Industry (PVA
reclamation, latex recovery, and caustic recovery) will
affect materials requirements only for specific segments of
the Industry; consequently,  overall savings can be expected
to fall In the lower 10 to 15 percent range.  In the second
category, demand for whey (dairy products) and recovered
solids to be used In animal  feeds (fruits and vegetables) 1s
fairly weak: possibilities for market expansion seem to be
limited.  Producers of byproduct sawdust (timber product
processing) and fly ash (electric utilities) were In similar
circumstances a few years ago; however, both Industries have
extended their market opportunities by finding various new
applications for their products.  In addition, all of these
examples are products with very low unit prices.
Consequently, It Is projected that recovery of whey and
solids can achieve savings somewhat below those encountered
In the Industry evaluations, or about 6 to 9 percent of

                           1-131

-------
annual costs, while use of sawdust and fly ash can achieve
10 to 15 percent savings.   For all cases,  no capital  savings
ware assumed.
ScPROCESS SUBSTITUTIONSSR

As with other process change opportunities,  the abatement
savings achievable through process substitution are limited
by the applicability range of the change and the rate of
substitution.. Process substitutions are usually introduced
during capacity expansions, rather than as retrofit
conversions.  Thus they are related to the rates of industry
growth and equipment obsolesence.  Applicability range means
the fraction of industry able to incorporate a change, the
Impact on overall  costs of any single change,  and the number
of complementary or competitive changes; the effect of the
number of opportunities is reflected in Table 10.  Two
process changes were identified a« already underway in the
copper and inorganic chemicals industries--thetr capital  and
annual!zed cost savings fell between 13.0 and 14.5 percent.
In the.other three industries, only one process change was
found to be significant: the resulting capital and
annualized cost savings fell between 5.0 and 7.7 percent.
                           1-132

-------
It Is useful to consider how process substitution
possibilities In other industries relate to those In the
five representative Industries.   The displacement of
Incineration by landfill or mineflll In the disposal of
municipal waste Is a process substitution affecting air
emissions.  The high cost of Incineration equipment 1s
countered by the high land cost  In most urban settings.   In
addition, a great deal of effort Is underway to Incorporate
the recovery of metallic and thermal resource values in
municipal waste during the Incineration step.  Likewise, new
techniques of sanitary landfill  are being developed that
facilitate subsequent productive use of that land.  Overall.
some displacement of Incineration by landfill Is envisioned.
leading to abatement costs In the 5.0 to 7.7 percent range.
Another process substitution Is  the displacement of sol vent -
based paints by electrostatically-suspended paints.  This
substitution 1s occurring rapidly, motivated both by
environmental considerations and concern regarding future
shortfalls In solvent supply.  The abatement cost savings
should lie In the high range shown In Table 10 between 13.0
and 14.5 percent.  A similar type of process substitution 1s
the displacement of petroleum solvents by syntehtlc solvents
In the drycleanlng Industry.  Here the opportunities ar*e
more 11ml ted,'leading to possible abatement cost savings in
the 5 to 7.7 percent range.  A final example Is the
substitution of the basic-oxygen furnace for the open-hearth

                           1-133

-------
furnace In the Iron and steel  Industry.   This substitution
has been underway for some time;  1t Is estimated that  the
remaining possibilities for process substitution will  only
permit an abatement cost savings  of approximately 2 percent.
$CSUMMARY$R

The greatest opportunities for abatement  cost  savings,  as
reflected In reduced annual)zed costs,  lie In  the area  of
process modifications.   In the copper,  pulp and paper,  and
petroleum Industries, process modifications can lead to
annual!zed cost reductions of more than 20 percent,  and In
the Inorganic chemicals Industry the potential  savings  are
10 percent.  Process substitution offers the second  largest
opportunity for abatement cost savings.  In both the copper
and the Inorganics Industries, two significant process
substitutions were Identified that lead to abatement cost
reductions of 13 percent for eacy industry.   In the
aluminum,  pulp and paper, and petroleum Industries,  single
process changes were Identified that lead to a range of
annualized cost reductions of 5.0 to 7.7 percent for the
three Industries.  The process change opportunities  that
lead to the least abatement cost reduction are those
associated with materials changes; the range In annual)zed
cost reductions is 0.3 to 3.3 percent.   This apparently

                           1-134

-------
reflects a high degree of optimisation in the section of the
raw materials now being used.

Table 11 presents the estimate savings potential  of process
change through comparison with the Reference Case estimates
of abatement cost requirements.   In addition to the results
from the five Industries surveyed in-depth,  other kno*n
process change opportunities with readily-quantifiable cost
effects were Incorporated 1n the comparison.  This latter
group Includes greater use of  recycling in the metals-
producing sectors, application of subprocess modifications
to textile manufacturing, and  changes In consumer demand
patterns for paper products.  Where process change trends
that are primarily or partially Inspired by environmental
concern have been included in  the reference case economic
forecast (e.g.. hydrometallurglcal smelting of copper or the
Bayer-Alcoa process), adjustments have been made to the
baseline cost estimates.  Where process changes are an
Integral part of the Reference Case control  strategy, as In
the case of waste heat boilers for petroleum refinery carbon
monoxide control, these values have not been adjusted
because of the absence of a costed-out alternating strategy.
The results Indicate that savings of almost $2 billion In
capital expenditures and $1 billion 1n annual costs can be
attained by 1985 through application of process change In
these Industries.  In addition.  It should be recognized that

                           1-135

-------
the technologies considered do not fully exhaust the
possibilities within the surveyed Industries.   These
Industries represent 15.8 percent of total  capital
expenditures on abatement by industrial  point  sources
(excluding mobile sources of in emissions,  municipal water
treatment and waste Incineration, etc.), and 14.2 percent  of
total annual costs.
                           1-136

-------
                  Sellable 11  Sheet 1  of 2.
                 Abatement Cost Comparison:
End-Of-Plpo Controls vs.  Combined EDP/Process Change Strategies

             Air Pollution Control Costs -  1980




Industry
Aluml num
Copper
Inorganic
Chemicals
Lead
Petroleum
Ref tntng
Pulp £ Paper
Textt les
Zinc
Totals
Cumulative
Investment
1972-1980
(Reference
Case)
2.265.1
1,405.1 .


81 .1

1 .003.0
2.891 .0

77.4
7.722.7
Cumulati ve
Investment
1972-1980
(Revised
Value)
2.142.8
1 ,292.6


79.4

1.003.0
2.590.3

74.4
7 . 1 82 . 5


% Change
-5.4%
-8.0%

NOT CONSIDERED
-2.1%

0%
-10.4%
NEGLIGIBLE
3.9%
-6.9%
Annual
Costs 1980
(Reference
Case)
734.2
420.1


19.2

230.3
807.6

31 .2
2.242.6
Annual
Costs 1980
Revised
Value)
420.1
354.6


18.8

230.3
720.4

30.0
2.026.6



% Change
-4.7%
•18.1%


•1 .8%

0%
•10.8%

•3.9%
-10.0%
  Reference case value adjusted to reflect effects of process change.
  Reference case value includes process change effects that cannot be estimated.

            Water Pollution Control Costs • 1980




Industry
Aluminum
Copper
Inorganic
Chemicals
Lead
Petroleum
Refining
Pulp & Paper
Textt les
Zinc
Cumulative
Investment
1972-1980
(Reference
Case)
92.3
27.7

419.3
5.6

938.8
2,869.9
302.2
16.5
Cumulative
Investment
1972-1980
(Revised
Va 1 ue )
89.9
27.9

419.3
5.4

826.1
2.056.7
274.7
1.5.7




% Change
-2.6%
+ .8%

0%
-2.8%

-12.0%
-7.4%
-9.1%
-4.6%

Annual
Costs 1980
(Reference
Case)
53.3
18.6

166.7
5.9

204.8
606.0
65.8
6.0

Annual
Costs 1980
(Revl sed
Va 1 ue )
51.5
19.0

103.2
5.7

128.4
499.3
59.8
5.7




X Change
' -3.4%
+1 .9%

-38.1%
•2.8%

•35.4%
-17.6%
-9.1%
-5.2%
Totals
4.616.7
3.715.5
•7.2%
1 .127.1
872.6
•21.6%
  Reference case value adjusted to reflect effects of process change.
                           1-137

-------
                   Table 11  Sheet  2 of  2.
               Abatement of  Cost  Comparison:
End-Of-PIpe Controls vs. Combined EDP/Process Change  Strategies

             Air Pollution Control  Costs -  1985




Industry
Aluminum
Copper
Inorganic
Chemicals
Lead
Petroleum
Ref inlng
Pulp & Paper
Textiles
Ztnc
Totals
Cumulative
Investment
1972-1985
(Reference
Case)
2.281 .8
1.433.8


84.2

1 .099.5
3.147.5

83.9
8.130.7
Cumulatl ve
Investment
1972-1985
(Revised
Value)
2.010.3
1 .210.1

NOT
79.4

1 .099.5
2.672.2

78.7
7 . 1 50 . 2




X Change
-11.9%
- 1 5 . 6%

CONSIDERED
-5. 736

0% .
-15.1%
NEGLIGIBLE
-6.2%
-11.7%

. Annual
Costs 1985
(Reference
Case)
660.3
425.5


20.3

251 .7
423.5

39.4
2.320.7

Annual
Costs 1985
(Revised
Value)
600.2
312.1


19.3

251 .7
772.0

36.8
1 .992.7




% Change
-9.1%
-26.5%


-4.8%

0%
-16.4%

-6.5%
-15.6%
  Reference case value adjusted to reflect  effects of  process change.
  Reference case value includes process change effects that  cannot  be  estimated.

            Water Pollution Control  Costs -  1985
Cumulative Cumulative
Investment Investment
1972-1985 1972-1985
(Reference (Revised
Industry
Al uml num
Copper
Inorganic
Chemicals
Lead
Petroleum
Refining
Pulp & Paper
Textl les
21nc
Case)
149.3
33.4

727.4
6.4

1.899.0
5.905.5
461 .5
20.8
Value)
138.8
34.0

709.2
5.9

1.557.1
5.055.1
414.4
19.3
% Change
-7.0%
+ 1.9%

. -2.5%
-7.6%

- 1 2 .0%
-14.4%
•10.2%
-7.2%
Annual
Costs 1985
( Reference
Case)
68.3
25.6

219.0
5.5

412.4
1.386.9
91 .3
14.3
Annual
Costs 1985
Revised
Va 1 ue )
63.4
26.1

164.3
5.1

274.6
1.006.9
82.0
13.3


% Change
-7.2%
+1 .8%

-25.0%
-7.6%

- 33 . 4%
-27.4%
-10.2%
-7.2%
Totals
9.203.3
7.933.8
-11.9%
2.223.3
1.635.7
  Reference case value adjusted to reflect effects of  process change.$s
-25.6%
                           1-138

-------
SbFootnotesSR
1.   Saxton. j. and Kramer,  M..  "Industrial  Chemicals Solid
     Waste Generation," Environmental  Protection Technology
     Series. EPA-670/22-74-078.  November 1974.

2.   Bennett. H. J.,  "An Economic Appraisal  of  the Supply of
     Copper from Primary Domestic Sources."  Bureau of the
     Mines information Circular  8598.  1973,  pp.  25-28, 139-
     146.

3.   Catalytic. Inc.. Capabilities and Costs of  Technology
     for the Organic and Inorganic Chemicals Industry to
     Achieve the Requirements and Goals of the  Federal Water
     Pollution Control Act Amendments of 1972  2 vols..
     1975.                                   ~

4.   Ibid.

5.   McGovern. Joseph H., "Strategy for Industrial
     Wastewater Control Programs", 77th National Meeting -
     AICHE. Pittsburgh. Pa., dune 2-5. 1974.

6.   The Cost of Clean Water. Volume III_ Industrial  Waste
     Profile No_ 5: Petroleum Ref1nlng_ Federal  Water
     Pollution Control Admlnlstrati on,"November  1967.
     Appendix A, Table 5.

7.   Gould. G. D.  (Chevron Oil  Co.), telephone  conversation,
     January 30, 1975.

8.   This analysis Is based on the results 1n the 1974 Cost
     of Clean Air and Water Reports.

9.   This analysis Is based on data contained 1n the
     national residuals generation (RESGEN)  module of the
     Strategic Environmental Assessment System  (SEAS).

10.  "Facing the Change In Copper Technology",  Chemical
     Engineering.  April 6. 1973.  p. 94A-HHH.

11.  "Proposed New Source Standards -for the  Copper
     Industry", Federal Register. October 15. 1974.

12.  Wall Street Journal. October 16,  1974;  National  Journal
     Reports_'0ctober 19. 1974;  Chemical Engineering. April
     6. 1973.

13   Wall Street Journal. February 6.  1973.  p.  10.
                           1-139

-------
14.  "Development Document for Proposed Effluent Limitations
     Guidelines and New Source Performance Standards for the
     Bauxite Refining Subcategory of the Aluminum Segment of
     the Non-Ferrous Metajs Manufacturing Point Source
     Category", U.S. Environmental Protection Agency.  EPA-
     440/1-73-019.

15.  WIcKes. H.G., dr.  and Whltchurch,  J. G..  "Flowing
     Consumption Trends/ The Aluminum Industry (Alcoa  398
     Dry Scrubbing Process)",  AIME Paper 73-H-50:  given In
     Chicago. March 5.  1973.

16.  "Alcoa to Build Smelter with Its New Process at East
     Texas Town". Wall  Street  Journal.  May 24. 1973.

17.  Peacey. J. G.. and Davenport. W. G., "Evaluation  of
     Alternative Methods of Aluminum Production. Journal of
     Metals. July 1974. pp. 25-28.

18.  "Water Use in Manufacturing". Census of  Manufacturers.
     1972.

19.  Ibid. Section II,  paragraph c.3.b of this report.

20.  "Changes In Store for Pulping Technology",
     Environmental Science and Technology. January 1975.

21.  Main. Charles T..  "Spill  Containment In  the Pulp  and
     Paper Industry" and "In Process Effluent Containment In
     the Pulp and Paper Industry" Chas. T. Main. Inc.,
     Boston. Mass. January 1975.

22.  Bower, T. B.. et at.  "Residuals Management In the Pulp
     and Paper Industry".  Resources for the Future_ 1755
     Massachusetts Avenue, N.W.. Washington.  D. C. 20036.
     January 1972.

23.  "Chesapeake Moves to Use of Oxygen in Effluent
     Treatment of BLO". Paper  Trade Journal,  July 22,  1974.

24.  lannazt. Fred P.,  "Greater  Use of Secondary Fiber by
     Application of Dry Forming". Paper Trade Journal.
     October 25. 1971.

25.  Cost of Clean Water. Volume III. Industrial Waste
     Profile No. 5: Petroleum Refining. Federal Water
     Pollution Control  Administration.  November 1967.
     Appendix A, Table 4.

26.  Petroleum Refining Industry: Technology  and Costs of
     Wastewater Control. Engineering Science,  Inc., 1975.
                           1-140

-------
27.  "Development Document of Effluent Limitations
     Guidelines and New Source Performance Standards for the
     Petroleum Refining Point Source Category", U.S.
     Environmental Protection Agency. April 1974, p. 95.

28.  Klett, R. J.. "Treat Sour Water for Profit".
     Hydrocarbon Process>1ng_ October 1972, pp. 97-99.

29.  Op.clt. McGovern, Joseph H.. Reference 5.

30.  Oil and Gas Journal, March 22. 1974.

31.  "Water Use In Manufacturing: 1968," The 1967 Census of
     Manufacturers_ Vol. I. Cha.7.

32.  "Development Document for Effluent Limitation
     Guidelines and New Source Performance Standards for the
     Major Inorganic Products Segment of the Inorganic
     Chemicals Manufacturing Point Source Category", U. S.
     Environmental Protection Agency. EPA-440/1-74-007-a.
     March 1974.

33.  Ibid.

34.  'Cleaner Units for T102 Still Leave DuPont at Sea,"
     Chemical Week. January 1, 1975. pp.26-29.

35.  Op.clt. Reference 32.

36.  Ibid.

37.  "North American Chlor-Alkall Industry Plants and
     Production Data Book," The Chlorine institute. January
     1974.

38.  Op.clt. Reference 32. (-38.1%)    (125.OX)
                           1-141

-------
 END OF DOCUMENT  Sect tonI



 LINES PRINTED 04190



' PAGES 0143
                   >*«»***#»**»»»**»*»************+****»******»



  »*****»***»»•****»***»*«»**»»•***»*******•********»*»»*******



  *************************************************************



  »•*»•»*»•***•*****»«»»»**»»***»*********************»**»***»»








  »•*«••«»*»**•***»*«*****»**»»»»****»****»*********»*********»



  ••**••«**•»***#**»*****»»***»»**»****************************



  ft***********************************************************



  *************************************************************



  ***********************************.****************«*********








  CUSTOMER a4690 OPERATOR 068                                 469  068  SectlonI

-------