United States         Office of Water        EPA 440/2 -80-085
             Environmental Protection     Regulations and Standards    December 1980
             Agency            Washington. O.C. 20460

             Water



v»EPA       Economic Impact Analysis of


             Proposed Effluent Limitations


             Guidelines, New Source Performance


             Standards, and Pretreatment


             Standards for the Coal Mining


             Point Source Category






             Volume III- Appendix  B,


             Methodology Description

-------
            ECONOMIC IMPACT ANALYSIS
  OF  PROPOSED EFFLUENT LIMITATIONS GUIDELINES,
        NEW SOURCE PERFORMANCE STANDARDS,
           AND PRETREATMENT STANDARDS
   FOR THE COAL MINING POINT SOURCE CATEGORY
VOLUME III — APPENDIX B,  METHODOLOGY DESCRIPTION
                  prepared for
      U.S. ENVIRONMENTAL PROTECTION AGENCY
    OFFICE OF WATER REGULATIONS AND STANDARDS
             WASHINGTON, D.C.  20460
                       by
             CONTRACT NO. 68-01-4466
                  DECEMBER 1980

-------
       This document is available through the U.S. Environmental Protection
       Agency, Economic Analysis Staff WH-586, 401 M Street, S.W.,
       Washington, D.C. 20460, 202-755-2484.

       This report has been reviewed by the Office of Water Planning and
       Standards, EPA, and approved for publication.  Approval does not
       signify that the contents necessarily reflect the views and policies
       of the Environmental Protection Agency, nor does mention of trade
       names or commercial products constitute endorsement or recommendations
       for use.
U;S. Em/iror.r.s.'-.^i! ppfrfcr/'on

-------
                               PREFACE


This volume is an Appendix to a contractor's study prepared for the
Office of Water Planning and Standards of the Environmental Protection
Agency (EPA).   The purpose of the study is to analyze the economic
impact which could result from the application of effluent standards
and limitations issued under sections 301, 304, 306 and 307 of the
Clean Water Act to the Coal  Mining industry.

The study supplements the technical study (EPA Development Document)
supporting the issuance of these regulations.  The Development Document
surveys existing and potential waste treatment control methods and
technology within particular industrial source categories and supports
certain standards and limitations based upon an analysis of the
feasibility of these standards in accordance with the requirements
of the Clean Water Act.  Presented in the Development Document are
the investment and operating costs associated with various control
and treatment technologies.   The attached document supplements this
analysis by estimating the broader economic effects which might result
from the application of various control methods and technologies.
This study investigates the effect in terms of product price increases,
effects upon employment and the continued viability of affected plants,
effects upon foreign trade and other competitive effects.

The study has been prepared with the supervision and review of the  Office
of Water  Planning and  Standards  of EPA.   This  Appendix was  submitted
in fulfillment of Contract No. 68-01-4466 by Arthur D. Little, Inc.,
and was completed in July, 1980.  The work was performed from June, 1977
through July, 1980; the data sources referred to in the report were
current at the time the work was performed.

This report is being released and circulated at approximately the same
time as publication in the Federal Register of a notice of proposed
rule making.  The study is not an official EPA publication.  It will
be considered along with the  information contained in the Development
Document and any comments received by EPA on either document before
or during final rule-making proceedings necessary to establish final
regulations.  Prior to final promulgation of regulations,  the accom-
panying study shall have standing in any EPA proceeding or court
proceeding only to the extent that it represents the views of the
contractor who studies the subject industry.   It cannot be cited,
referenced, or represented in any respect in any such proceeding as
a  statement of EPA's views regarding the Coal Mining  industry.

-------
                                APPENDIX  B
                            TABLE  OF  CONTENTS
          SECTION                                              PAGE NUMBER
  I.   INTRODUCTION                                                B-l
      A.   Main Sources  of  Data  Used  in  The  Imapct Analysis         B-l
      B.   Four Main Premises  Used  for the Analysis                B-2
 II.   IMPACT ANALYSIS METHODOLOGY                                  B-4
III.   REGIONALIZATION OK THE  MINE  FILE  (MODULE  1)                 B-8
 IV.   MINIMUM REQUIRED  PRICES (MODULE 2)                           B-8
      A.   Number of Different Sets of Model Mine  Parameters
          Required                                                B-10
      B.   Model  Mine Production Cost  Parameters                    B-10
      C.   Mine Flow Volumes  and Water Treatment Costs              B-15
      D.   Projection of Changes in Mining Conditions Between
          1976 and 1984                                           B-20
      E.   Minimum Required Price Calculation and  MRP Function
          Estimation                                              B-29
      F.   Details of the Minimum Required Price Calculation        B-31
      G.   Sample Output                                           B-33
      H.   Estimated Function  Parameter  Values for MRP,  Wages,
          Cashflow and  Required Investment  in Mining Equipment     B-33
  V.   REGIONAL SUPPLY CURVES  (MODULE  3)                            B-35
      A.   Separation into Major Coal  Markets                       B-35
      B.   Changes in the Mine Population                           B-38
      C.   Detailed and  Linearized Supply  Curves                    B-41
 VI.   COAL MARKET SIMULATION  (MODULE  4)                            B-52
      A.   Sulfur Content Distributions, Air Quality Control
          Standards and Coal  Utilization  Costs                     B-52
      B.   Transportation Costs                                     B- 56
      C.   Demand for Coal                                          B- 58

-------
                            Table of Contents
                               (continued)
         Section                                                Page No.
 VII.  LINEAR PROGRAMMING MODEL OF THE STEAM COAL MARKET         B-58

       A.  Market Clearing Algorithms                            B-64

       B.  Linear Programming Formulation of Contract
           Market (No Producer's Surplus)                        B-64

            1.  Specification of Constraints for
                the LP Model                                     B-66

            2.  Specification of Objective Function              B-70

       C.  Mixed Integer Programming Formulation of
           Spot Market (Producer's Surplus)                      B-70

VIII.  OUTPUT OF THE COAL MARKET SIMULATION MODEL (MODULE 6)     B-71

  IX.  LIMITS OF THE ANALYSIS                                    B-75

       A.  Summary                                               B-75
       B.  Statistical Significance of the Impact
           Estimates                                             B-77

       C.  Sensitivity of the Impact Estimates to
           Systematic Errors                                     B-87

       D.  Bias Resulting from Aggregation Errors                B-93
                                   n

-------
                                  APPENDIX B

                                LIST OF TABLES

 No.                                 Title                            Page No.

 1   Supply and Demand Regions Used for the Impact Analysis            B-9
 2   Number of Different Combinations of Mining Cost, Water            B-ll
     Treatment Cost and Fiscal Parameters Required for the
     Analysis of     One Control Level
 3   Example of Input Data for Model Mine Cost Analysis                B-13
 4   BPT Investment Cost for New Mines                                  B-19
 5   BAT Costs for Small Mines (0.025 MMT/Yr)  with Different Flow      B-21
     Volumes (Without Correction for Difference in Mining Region or
     Mine Type)
 6   BAT Costs for Large Underground Mines (0.6 MMT/YR) With           B-22
     Different Flow Volumes (Without Correction for Differences in
     Mining Region or Mine Type)
 7   BAT Costs for Large Strip Mines (2.0 MMT/Yr)  With Different       B-23
     Flow Volumes (Without Correction for Differences  in Mining
     Region or Mine Type)
 8   BAT Costs for Large Strip Mines (4.0 MMT/Yr)  With Different       B-24
     Flow Volumes (Without Correction for Differences in Mining
     Region or Mine Type)
 9   Multipliers to Correct for Regional  Differences in BAT Invest-    B-25
     ment Costs
10   Projected Escalation in Costs and  Labor Productivity Used in      B-27
     the Model Mine Cost Analysis
11   Results of Model  Mine Analysis                                    B-34
12   Estimated Function Parameters for  Large Underground Mines in      B-36
     Northern Appalachia
13   Contract and Spot Market Coal Supplied to Electric Utilities      B- 39
     in 1976
14   Attrition Rate Parameters                                         B-40
15   Mine Productivity Distributions                                   B-42
16   Existing Production of Contract Mines in  1976 in Different        B-44
     Size Ranges
17   Estimated Potential Increases in Mining Capacitv Until 1984       B-47
18   Mine Water Flow Distributions                                     B-48
                                      iii

-------
                            List of Tables
                              (continued)
No.                               Title                             Page No.
19      Detailed Supply Function for Pennsylvania, 1984 (BAT-4)     B- 49
20      Linearized Supply Curve for Pennsylvania                    B-51
21      Ratios of Coal Available in Different Sulfur Content Ranges B-53
22      Ratios of Coal Demand Subject to Different Air Quality
        Control Standards                                           B- 54
23      Coal Utilization Costs                                      B- 55
24      Parameter Values for Rail Transportation Cost Equations and
        Applicable Origins and Destinations for Contract Market     B- 57
25      Transportation Costs Not Based on Rail Rate Equations       B- 59
26      Transportation Costs for Links Between 27 Supply
        (Horizontal) and 35 Demand (Vertical) Regions (Contract
        Market)                                                     B- 60
27      Rail Transport Cost Equations and Applicable Origins and
        Destinations for the Spot Market                            B- 61
28      Bridge from Contract to Spot Market Regions                 B- 62
29      Coal Demand in 1984 (Billions of Btu's)                     B- 63
30      Nomenclature Used in the LP Model of  the Steam Coal
        Market                                                      B- 67
31      Coal Supply Summary - 1984 Level 4                          B-72
32      Coal Burned Cost Summary                                    B-73
33      Example of Calculation of Solution Supply for a Sampled
        Curve  for Pennsylvania  (under BPT)                          B-82
34      Results of Thirty Supply Samples  for  Pennsylvania           B-84
35      Estimated Supply Impact of BAT-4  on the  Contract Market
        Compared with  the Possible Variation  in  that  Estimate
        Due  to Uncertain Information on Mine  Water  Flows            B-85
                                  IV

-------
                                List of Tables
                                  (continued)


No.                                  Title                          Page No.

36   Possible Variation in Estimated BAT-4 Supply Impact Due         B-88
     to Uncertain Information on Mine Water Flows

37   Estimated Supply Impact of BAT-4 on the Spot Market             B-89
     Compared with the Possible Variation in that Estimate
     Due to Uncertain Information on Mine Water Flows
38   Possible Variation in Estimated BAT-4 Supply Impact
     Due to Uncertain Information on Mine Water Flows                 B-90

39   Results of Sensitivity Tests in Terms of Changes                 B-92
     in Estimated Supply Impacts

-------
                              APPENDIX B
                            LIST OF FIGURES

 No.                             Title                           Page No.
 1          Schematic of Impact Analysis Methodology               B-5
 2          Coal Supply-Demand Model                               B-6
 3          Coal Mine Water Flows -  Appalachia                      B-16
 4          Coal Mine Water Flows -  Midwest  and Central West        B-17
 5          Coal Mine Water Flows -  Great Plains and West          B-18
 6          Historical Mining Labor  Cost and Mining Equipment
           Cost Indices                                           B-26
 7          Historical Productivity  Indices  for Underground
           and Surface Mines                                      B-28
 8          Flow Diagram of Model Mine Cost  Analysis Program       B-30
 9          Main Function of Module  3:  Estimation of Regional
           Coal Supply by Price in  1984                           B-37
10          Typical Supply Curve Illustrating Alternate Market
           Clearing Mechanisms                                    B-65
11          Changes in the Regional  Supplies, Marginal Prices      B-74
           and User Costs
12          Illustrative Examples of Supply  Curves Based on        B-78
           Average versus Sampled Values for Mine Productivity
           and Mine Water Flows
13          Illustrative Example of  Indeterminate Range of the     B-79
           BAT Supply Curve Because of Uncertainty About Mine
           Water Acidity and Flows
14          Methodology to Derive Maximum Possible Error
           in Impacted Regions                                    B-80
15          Calculation of the Interaction Error                   B-86
                                  vi

-------
                             APPENDIX B

                        METHODOLOGY DESCRIPTION


I.  INTRODUCTION

This Appendix describes  the  coal  supply-demand model  that  was deve1-
oped to analyze the economic impact of proposed, more  stringent water-
quality control  standards ^or point sources in coal  production.  These
control standards will  require coal  producers  to  spend more for water
treatment at the mine  site,  thereby  increasing  their overall cos*s of
producing coal.

The  relative  increase  in  production  costs for  individual  mines is
influenced  by   numerous  factors  that  are  related  to  the different
environments in which  an operator  has to  deal, such  as:

    f    Meteorological:  relative rainfall;

    •    Topographical:  hilly  versus flat  terrain;

    •    Locational:   proximity to other  mines;

    •    Geological:    coal quality ind depth  of coal  seam,
         presence  of   acquifers,  chemical  composition  of
         overburdens;

    •    Financial:  cost of money, equipment and  labor;

    •    Microeconomic:  a  mine's market position  in  terms
         of location  and producer/user  relationships;

    •    Macroeconomic:  energy and  feedstock supply demand
         balances  and  the  costs  o*  other  fuels   and
         feedstocks.

This  analysis  is to  estimate  the  relative  dependence of  coal
production  costs  on   each  of these  factors in  sufficient  detail   to
allow  reliable  estimates  of the  direction  and  relative  magnitude  of
inter-regional  shifts  of coal production that can be expected  to take
place  in 1984 if more  stringent mine water  treatment requirements  are
imposed.

A.  Main Sources of Data Used in the  Impact  Analysis

The 1976 MESA Mine File  described  in Chapter II of  Appendix A  of this
report  is  used  as  a   starting  point  for  estimating mine  production
costs  of individual   mines.   This file  contains detailed  production
information for  individual  users.   Some of this  detail  is lost  or
                                 B-l

-------
smoothed out  through  averaging,  when regional supply  curves are  lin-
earized.

Transportation  costs  are  estimated  by  statistical  analysis  of  rail
rates and water transport costs, as described in Chapter V of Appendix
A.

Coal sulfur content distributions are derived from an  analysis of  coal
quality data for coal sales to electric utilities in 1976, as obtained
from  the FPC.

Coal  utilization  costs  are  developed  as  described  in  Chapter  VI  of
Appendix A  of  this  report.   These  costs  account, for the different  Btu
content and sulfur content of different coals, and for the varying  air
quality control standards in the different demand regions.

As will be  shown  in  later  sections,  the  limited  number of data  points
on water flow  volumes for coal treated at  individual mine sites  is  the
main limiting  factor  in  the level of regional detail achievable  in  the
impact estimates.

B.  Four Main  Premises Used for the Analysis

The  estimated  economic  impact of the  regulations  is obtained through
simulation  of  supply and demand in  steam  coal  markets  in  1984.   The
simulation  is  structured around the following four premises:

    (1)  Coal,  as a  fuel,  competes with  coal  rather  than
         oil,  gas  or  nuclear  energy.   This premise  is based
         on  the  current trend  of  increasing  oil   and  gas
         costs  -  as  discussed in Section VI.1.3 of  Appendix
         A  -  the  restrictions on the  use  of gas as utility
         boiler fuel,  and  the growing  resistance to further
         rapid  growth in the  use  of  nuclear energy.   The
         analysis  used  a fixed demand  estimate for  1984 and
         concentrated  on the  estimation  of inter-regional
         shifts  of coal supplies  which  can  be  expected to
         take  place  because of  relatively higher estimated
         pollution control costs in some regions (as result-
         ing  from differences  in  the  relative  wetness of
         mines  in  the various coal-producing  regions).

    (2)  Coal  producers, in  order  to continue to mine coal,
         will  have to recover  their operating expenses and
         investment  costs  (but  not  sunk  costs), including  a
         return on those investment costs  over the remaining
         life  of  the  mine.     In  this   analysis  this
         requirement  establishes the minimum price  at which
         coal  will  be sold.   If mine operators  cannot get
         that  minimum price, then they  will  close  the  mine
         and  cease to produce.
                                   D-2

-------
    (3)   It  Is  assumed  that the steam coal contract  and  spot
         markets  operate  independently  of each other.   Only
         two interactions  are  allowed in  the  analysis:

                  Large   mines  producing  more  than
                  50,000  tons  per year 'i.e., contract
                  market  mines)   are  assumed  to sell
                  some of their output  in  the spot
                  market  at variable costs.

                  Contract  market  mines  which   are
                  found   to   be  non-competitive   in
                  contract markets  are assumed to sell
                  to  the spot market  (i.e., a market
                  dominated  by  small  mines  producing
                  less than   50,000  tons of coal   per
                  year).

    (4)   The nature of transactions between producers  and
         users  in contract and  spot markets  is assumed to be
         completely  different.   In  the  contract  market  all
         producers are assumed to  sell  coal on a cost-plus
         basis;  the  large,  virtually  "infinite"   resource
         base  creates  a  very  competitive  situation on  the
         supply side,  always  allowing large  buyers to select
         among  a large   number  of potential   suppliers,
         thereby driving  prices down.

         In   the   spot  market  all   coal  from  a  region  is
         assumed  to  sell   at the  marginal  price.   Buyers  in
         this market are  more  concerned  with timely supply
         than  with  price.   Suppliers  in this  market  know
         what the cost  of coal  on the margin is and  are  able
         to  negotiate  prices  up  to that margin.    In  other
         words,  in  the  contract market  all  producers  are
         assumed  to  obtain the same return on their invest-
         ment;  in  the  spot market  producers with  low-cost
         coal  will have a higher return  than  producers  with
         high-cost coal.

Given these  four main  premises, the  analysis concentrates primarily on
specifying the  differences between costs  of  supply at different output
levels in the various  supply  regions,  allowing for:

    •    Differences in  coal   mining  conditions  'including
         mine wetness);

    •    Different transportation costs  because of different
         distances between supply and  demand regions;
                                  B-3

-------
    0    Differences in user costs because of different roal
         quality;

    t    Differences  in  user costs because  of the  differ-
         ences between  air  quality  control   regulations  in
         different demand regions.

The remainder of this chapter explains the above in more detail.

II.  IMPACT ANALYSIS METHODOLOGY

Figure  1  shows   a  schematic   of  the  impact  analysis  methodology.
Regional supply volumes and costs are projected for 1984 for  thp steam
coal markets (spot and contract^ and for the met coal market, allowing
for differentials  in  coal  prices  because  of  differences in production
costs, transportation costs and utilization costs.

As shown, ADL has made estimates of potential supply and related costs
for 27 different  supply  regions.   These supply estimates are combined
with  estimates  of demand  in  35 demand regions 'as  obtained from EPA
which  used  these demand projections  recently for an  analysis  of the
impact of air quality control regulations).

These  two  sets  of estimates, together  with  estimates  of coal   quality
differentials, and transportation and  utilization  costs,  are  used in
the coal market  simulation model  to  obtain  estimates  both beforp and
after  the  proposed  regulations.   This  allows  us to  derive the dif-
ferential  economic  impact  resulting   from  production  cost  increases
caused by more stringent water  clean-up requirements at the mine site.
As shown in Figure 1, the  primary impact measures are:

    t    Increased  user  costs  both  as an average  and on the
         margin  in different demand regions;

    •    Increases  in marginal  prices  in  supply regions;

    •    Lost production in  supply regions;

    •    Number  of closed-mines;

    •    Number  of jobs  lost;

    •    Resulting  decreases  in  regional  mine  workers'
         wages;

    •    Mine  cashflows   versus   increased   investment
         requirements  for  pollution control  equipment.

As  shown in Figure 2, the model  itself consists  of six modules,  each
composed of one  or  more data  files  and  computer programs,  which  use
those  data  as input.  These modules perform  the  following main opera-
tions:
                                   B-4

-------
                                 FIGURE 1
                  SCHEMATIC OF IMPACT ANALYSIS METHODOLOGY
                   ADL

               Coal  Supply

                  Model
     1984 Supply
   Estimates in 27
   Supply Regions
     AOL:

 Coal Quality

 Differentials
 National

Coal Supply

   Model
         ADL:

 Transportation Costs
         ADL:

 Utilization Costs
Coal Simulation

  Market Model
                            1984 Demand
                         Estimates in 35
                         Demand Regions
BEFORE REGULATION IN 1984 (by region):
Number of producing mines
Total  cost of coal  to the user
Tons of coal  produced
Coal prices
Number of mine workers employed
Wages  earned after tax cashflows
Required investment in mining equipment
            AFTER REGULATION IN 1984 (by region):
            Number of producing mines
            Total cost of coal to the user
            Tons of coal produced
            Coal prices
            Number of mine workers employed
            Wages earned after tax cashflows
            Required investment in mining equipment
                           PRIMARY IMPACT OF THE REGULATION (by region):

                           Increased cost to the user
                           Lost production
                           Number of closed mines
                           Number of jobs lost
                           Decrease in wages
                           Decrease in cashflows
                           Increased investment required
                                    B-5

-------
                                                                                FIGURE  2

                                                                    COAL  SUPPLY-DEMAND  MODEL
                                         MODULI 1
                                                                                                              Hfllllll I V
ta
           MODULE 3
           MODULE 4
           MODULE 5
                                                                                                /Model Mine
                                                                                               /Costs by Re-
                                                                                               (i|ion and Mine I
                                                                                               V   Type
                                                                          Mater

                                                                        Treatment

                                                                       Cost Data
                                                                                                 rPROGRAM 2:

                                                                                                (Price Analysis
                         \suppi
 /Spot Market /^
/and Met Coal  /
I Market Deei-  I
\£lon Rules   V
 /rapacity
/Growth and
(Mine Replace* |
Went Para-
 \meters
                                                                                                 Cost and
                                                                                               fPrice Functv
                                                                                                Parameters
                                   i
                        XlMne    f

                       ( Hater Flow   I

                       ypistributionsV
                                            /Hal
                                            /Tre<
                                   Mater
                                  Treatment
                                (   Cost
                                V Parameters
         I
i
/PROGRAM 3:\






                                I
                              Detailed
                         (nd Linearized/
                        (Supply Curves
                           by Major
                             Market
    Coal Sulfur/^
    Content    /
   Distribution
 S  Coal
/Utilization
  Cost Data
              V
I  Cost
/Air
I Stai
   Regional
    Control
Standards
c
                                                Transporta-
                                               tion Costs

 Demand
Forecast
                           /PROGRAM 4:

                          /   Market
                          \Fomulation
\-





                                                                         Estimated
                                                                       Supply and
                                                                         Cost of
                                                                         Supply
                                                                                        lUDULC  6

-------
   MODULE 1: Separates the mines contained in the MESA file
             into  27  supply  region files  which  serve  as
             inputs to Module 3.

   MODULE 2: Uses engineering estimates of production costs
             for  different  types  of model  mines  in  dif-
             ferent supply  regions.   It calculates minimum
             required  prices  (MRP^  for  these  mines  for
             different  labor  productivities,  as  well  as
             different  investment  costs  in  pollution
             control  equipment  resulting  from  different.
             mine  water  flow volumes.   These price  esti-
             mates are  used  to  calculate  a  functional
             relationship  between  minimum  required  price
             (the  dependent  variable^, mine  productivity
             and  required  investment  in  pollution control
             equipment per annual ton of coal produced 'the
             independent variables).

   MODULE 3: Uses  the MRP function resulting from Module 2,
             plus  the regional  mine  files,  to  calculate
             three  supply curves  for  each  of  the supply
             regions  for  the steam  coal  spot  and contract
             markets.    The  supply  functions,  specifying
             cumulative  potential  supply  at  increasing
             prices at  the  mine, are first calculated on a
             mine-by-mine  basis 'i.e., detailed)  and then
             linearized  for  use as  an  input  to  the linear
             program  used in Module  5 to balance supply and
             demand in coal markets.

   MODULE 4: Organizes  all  the  information  used  in  the
             market simulation  in a  linear program which is
             solved in  Module 5.  In  addition to the lin-
             earized  supply curves  for the  27  supply re-
             gions,  information on  sulfur  content distri-
             bution  and  transportation  and utilization
             costs  of  different  types  of   coal   with
             different sulfur content  is provided.

   MODULE 5: Provides  a  solution to  the linear program set
             up by  Module 4.

   MODULE 6: Takes  the  output  from  Module  5  and  organizes
             it  into   several  reports   to  allow  more  rapid
             evaluation of the  results  of the analysis.

The nature  of the  data and the  computations performed by the programs
used  in these modules  are described  in  the following sections.
                                 B-7

-------
III.  REGIONALIZATION OF THE  MINE  FILE  (MODULE 1)

The MESA file, which contains 1976 information  on  mine  location,  mine
type, status  (i.e.,  open,  temporarily  or permanently closed^,  number
of  hours worked,  and tons  of coal produced,  is organized  into  the  27
different supply regions.  These supply  regions, as  shown  in  Tablp  1,
coincide largely  with  the supply  regions used  in the National  Toal
Model.

IV.  MINIMUM REQUIRED PRICES  (MODULE  2)

The 1984 minemouth prices  used in the analysis  are based  on estimated
production costs.  Since actual production costs for  individual  mines
are  not  available,  engineering  estimates are made  of unit investment
and operating costs for mine  models,  categorized by:

    •    Location (region);

    •    Mine type (underground  or surface);

    •    Mine size (large or  small);

    •    Remaining producing  life  (new or existing  mine);

    •    Required water  treatment costs  for  compliance  with
         EPA regulations.

Within each mine category,  costs  are  specified to changp with:

    •    Mine productivity (tons  of  coal produced  per  mine
         shift};

    0    Level of water flow to be treated.

As  shown in  Figure 2,  the  input to the  computer programs  of  Module ?
consists of a set  of model  mine parameters  and  water treatment  costs;
the  output  consists of  a  set of  parameter  values,  relating minimum
required price  (MRP)  to productivity  and dollars  of  required invest-
ment in water treatment equipment.

The  calculation  of the  parameter values  which relate  the MRP  for a
mine  to productivity  and  treated water flows  is  performed  in  two
stages:

    •    Calculate  MRP's  for  each  mine  category   for
         different  mine productivities   and  for  different
         flow levels;

    •    Estimate  function  parameters  relating MRP's,
         remaining required  investment in mining  equipment,
                                 B-3

-------
                                                     TABLE  1


                                       SUPPLY AND DEMAND REGIONS USED
                                            FDR THE IMPACT ANALYSIS
                                       MCfl REGIONAL CtNTROIDS WITH
                                    rHEIGHT STATION ACCOUNTING CODES
Supply Region

Pennsylvania (PA)
Ohio (OH)
Maryland (HO)
West Virginia.  north (NV)
Nest Virginia,  Much (SV)

Virginia (VA)
Kentucky, cast (CK)
Tennessee (TM)
•Alabama  (AL)
Illinois (It)
Indiana  (IN)
Kentucky, vest (MCI
lo»a
    ourl (HO)
    •s (KS)
Arkansas IAR)
Oklahom (OK)
Toaas (TX)
North Dakota |ND)
Montana, east (EH)

Wyoming (wr) *^

Colorado        (CS)

Utah (irn
Arizona (AZ)
Hew Mexico (NH)
Washington (WA)
 Centroid

Johnstown
Cambridge
Lonaconing
Clar keburg
Blurfleld
Charleston
Appalechia
Ha sard
Clinton
Cordova
Ceniralia
Hunt Ingburg
Central City
OttiMwa

Clinton
Pittsbwrg
Russollviiio
TUlua
Con. icana
Hilton
Sidney

Casper

Carbondalc

Sunnyside
Nlm.low
Gallup
Cent ralia
Freight Station
Accounting Code

   05001151
   05001549
   83909238
   05000469
   SS0039S5
   OS003642
   72401230
   44402631
   7240S420
   72407457
   07624120
   72300344
   44404124
   07620241

   69300162
   40000130
   49406080
   69300839
   69401202
   0765769'
   076S9225

   07632236

   19702416

   19709106
   02210286
   022101 SB
   14004514
   Deasnd Region
Hew England
                      Centcoid
freight Station
Accounting Code
South Atlantic
                MV Concord, K.H.        0690012!
                MC Sprxnqfield,  Mass.   C2219122
Middle Atlantic NU Osweoo, N.Y.         62210667
                UP Fittsbur?,  Pa.       62204727
                PJ Trenton, H.J.        62200203
                VM Baltlnore,  Md.       05000129
                   Norfolk. Va.         55001001
                W Wheeling, H.Va.      05000161
                   Huntlnaton. H.Va.    05000768
                CA Charlotte,  H.C.      72402535
                T Atlanta, Ca.         71213699
                          ria. (Barge only)
                      Palm Beach, fla.  71226590
                                                              Eaat north
                                                              Central
                                                              Bast South
                                                              Central
Hast North
Central
                                West South
                                Central

                                Mountain
                                                              Pacific
                SP w.
                ON Cleveland.  O.        05002168
                OH Coluefeus. O.         05001640
                OS Marietta, O.         05001648
                MI Detroit. Mich.       62206874
                XL Peoria, 111.         62208884
                IN Indianapolis,  Ind.   62208463
                Ml Milwaukee,  Hie.      140C!<>«9

                EX Winchester, Ky.      44402033
                NX LouUville, Ky.      44401000
                ET Kiorville.  Tena.     44405683
                NT Memphis. Term.       44406380
                AH Biminahaa, Ala.     44407206

                DM Minneapolis. Minn.   14005898
                KM Topeka. Kan.         02202571
                XA Das Moinea, la.      14500759
                MO St. Louis.  No.       69300752

                AO Little Rock. Ark.    14507420
                TX Dallas, Tex.         02209354
                MM Billings. Mont.      07630841
                UN Salt Lake City, Utah 80200081
                CO Denver. Col.         07620977
                AN Phoenix, Aris.       02214194
                NO Seattle. Hash.       07602199
                CM Oakland, Calif.      72100013
                CS Los Angeles. Calif.  07612141
     Wyoming  has  been separated into Wyoming Powder River Basin  and Wyoming,  Other Regions.
                                                    B-9

-------
         annual wages,  and  cashflow  per  annual ton  of  coal
         produced (the  dependent variables)  with  productiv-
         ity  and  mine  water  flow  volumes  'the  independent
         variables).

A.  Number of Different Sets of Model Mine Parameters Required

As  shown  in Table 2,  model  mine specifications for both  old  and new
mines required estimates for the following parameters:

    •    Investment and operating costs;

    0    Fiscal   parameters,   such   as   royalty  payment,
         federal, state and local taxes;

    •    Water treatment costs.

A total of  50  different combinations of these three sets of parameters
resulted  after careful  analysis of available engineering estimates of
mine  operating and  investment  costs, state  taxes and  royalties and
available  information  on  mine water  flows in conjunction  with cost
estimates  for different water treatment  levels provided  to  us by the
EPA  (see  Table 2).   Each  pollution  control  level  required evaluation
of  such a  set  of  50  mines.  The impact analysis considered the effect
of  two  different  proposed  BAT treatment  levels relative to the exist-
ing  BPT treatment level.   Therefore, a  total  of 150  different mine
models  had  to  be  evaluated.

B.  Model  Mine Production  Cost  Parameters

Table  3 shows an example  of the  model  mine investment and operating
costs  used  for the calculations  described  in this  section.

Preliminary analysis showed that within  the mine  size ranges, judged
to  be  representative  for the  two major coal  markets  analyzed  (contract
and  spot), minimum required prices were  only very slightly  related to
mine  size.   Therefore, one representative mine size was used  for each
mine  category in  the  different  regions.

As  shown  in  Table  3,  the cost  parameters  are  divided  into  three
groups:

     •    Investment  cost  data  for the  mine facilities  and
          mining  equipment;

     •    Operating  cost   data,  together  with  the   average
          labor productivity for which they are specified;

     •    Operating  cost  ratios,  specifying by how many
          percentage  points  certain  categories of operating
                                   B-10

-------
                                                  NUMBER OF DIFFERENT COMBINATIONS OF MINING COST.
                                                 WATER TREATMENT COST AND FISCAL PARAMETERS REQUIRED
                                                        FOR THE ANALYSIS OF ONE CONTROL LEVEL'
                              TYPE OF PARAMETERS
Coal Mining Cost
REGIONS
1. NA Ug

Su

2M. CA & SA Ug

Su
co
i
^j
~* 4. MW Ug

Su

5. CM Ug

Su

6. Tx Ug

Su


L
S
L
S
L
S
L
S
L
S
L
S
L
S
L
S
L
S
L
S
OLD
Ao
Bo
Co
Do
Ao
Bo
Co
Do
Ao
Bo
Eo
Fo
Ao
Bo
Eo
Fo
Ao
Bo
Go
Fo
NEW
Al
Bl
Cl
01
Al
Bl
Cl
Dl
Al
Bl
El
Fl
Al
Bl
El
Fl
Al
Bl
Gl
Fl
Uater Treatment Cost
OLD
ao
bo
bo
bo
CO
CO
bo
bo
ao
ao
do
do
CO
CO
do
do
CO
CO
bo
bo
NEW
al
al
bo
bo
cl
cl
bo
bo
a2
a2
do
do
c?
c2
do
do
CO
CO
bo
bo
Royalties/
Taxes

1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
Z
2
2
?
Parameter Combination
Old Mines
Aoaol
Bobol
Cobol
Dobol
Aocol
Bocol
Cobol
Dobol
Aoaol
Boaol
Eodol
Fodol
Aocol
Bocol
Eodol
Fodol
Aoco2
Boco?
fiobo2
Fobo2
New Mines
Alall
Blall
Clbol
Dlbol
Alcll
Blcll
Clbol
Olbol
AlaZl
Bla?l
Eldol
Fldol
Alc21
Blc21
Eldol
Fldol
Alco2
Blco2
Glbo2
Flbo?
Model Mine #
Old
1
3
5
7
9
11
5
13
1
16
18
20
22
11
18
20
25
27
29
31
New
2
4
6
8
10
12
6
14
15
17
19
21
23
24
19
21
26
28
30
32
The control  level could be the existing BPT regulation or any of the  proposed  RAT  regulations  considered  by  the  EPA

-------
TABLE 2 - Cont'd
REGIONS
7. NGP
MA

ND/SD

Wy

^ Rocky M.
849. Nev/NMex.



10. WA

Coal Mining Cost
Old" New
Su

Su

Ug

Su
Ug

Su

Su

L
S
L
S
L
S
L
S
L
S
L
S
L
S
Ho
Fo
Io
Fo
Ao
Bo
Jo
Fo
Ao
Bo
Jo
Fo
Jo
Fo
Hi
Fl
11
Fl
Al
Bl
Jl
Fl
Al
Bl
Jl
Fl
Jl
Fl
Mater Treatment Cost
Old" ' '
bo
bo
bo
bo
CO
CO
bo
bo
CO
CO
bo
bo
bo
bo
New
bo
bo
bo
bo
CO
CO
bo
bo
cl
cl
bo
bo
bo
bo
Royalties/
Taxes
3
3
4
4
4
4
4
4
1
1
1
1
1
1
1'iiriiiiieter Combination
Old Mines
Hobo3
Fobo3
Iobo4
Fobo4
Aoco4
Boco4
Jobo4
Fobo4
Aocol
Bocol
Jobol
Fobol
Jobol
Fobol
New Mines
Hlbo3
Flbo3
Ilbo4
Flbol
Alco4
Blco4
Jlbo4
Flbo4
Alcll
Bocll
Jlbol
Flbol
Jlbol
.Flbol
Model Mine '
Old
33
35
37
39
41
43
45
39
9
11
47
49
47
49
New
34
36
38
40
42
44
46
40
10
12
48
50
48
50

-------
                                             TABLE  3

1 ..1 I.I 1.
iiti.it : ••
L
1
I
J
I
1
1
1
1
I
1
1
1
1
1
J
I
I
1
1
\
J
J
I
I
1
•L
\
I
1
PL A MM IN
....

EXAMPLE OF
.I'l.l t i.i ii inn 1. 1 f'l II
1 n|' K'l (M. I'd -1 INl I'1!
"Ui*iH PilU, it r.li'il
U U t< -J l . U 0 U 0
. 140'i i . r?.m
ir./n i (i'ioo
I'. ,'0
. J £. C f '
. .. ' •'.'
'"•••U
. :-5 -o
."7>n
. VJ v!j
. •"! J i
.-. 'lv>
.i "J J
.. J l,ll
. .VA
.'•'«.-.'!
1 1 "' *, V
r ,
rle. • "*
.r. ..,.5
. vfu
• .•
. i. •. .j
•t; n
.vi. •<
/•..-: 7
V.ir /
. 1 ""'i'j
•-• j ' r
• -yc •
•-'::;,:'
G_vfARS

1 . U A 1 0
1 . Uti^O
i . J in; 0
j . ; .-60
i . j»t' n
1 . I 7.70
i . i?:.u
l .nl.T?
1 . .'ll"^3
i , ''-..31
1 .::•/'.. 5
1. '.IVl
j . ^u: ,
i . ". f.' f
1 .'ii'Ol
1 .•''•Ml
j . l''..67
1 . 'v!4S->3
i . ".il"»ci
1. . !.''! t •*'
J . ti /i'ȣl
1 .t>([Z J
1 . dll OP
1 . v . ".-3
• J .7iW.fi
1 .7't09
1. 77i,7


INPUT DATA
VI 1 '1 ''I i.l
A'-l '•>
l 01 II PHI NT
1.0000
1 .032P
1 .t'l-'.O
l.p-cMO
1 . ii >?0
\. I'vjn
1 . liV t$
\.OT- .50
i . on,:u
l . HV.JI*
l . Ii i!V
1 . i n t1:
J . J HVi
i. t IVi
l .i.'il
l .1:",;
1 . r .-.s
l . i .••::n
- . i '•"•:>
1 . I1!'"1!-
l.ii-i-i
l.U.-T'
1 l..f.V
i . \ '.':>:,
1 .IVi.iJ
i . i P-I •>
\ . i -I..-
1 . 19bl
i.:io:>i


FOR MODEL MINE COST ANALYSIS
ml l
" " "1
I
J
1
1
j
1
J
1
J
I
1
1
1
I
1
1
1
1
j
1
1
1
1
)
1
1
I
1
1


o-Ti rr
TiWt)
.1710
. 37.-.0
.1900
.20 50
. '.MC'O
. '.-'3 *0
pil70
.'.!£ f 0
. :J". *o
. 2''"i'!
. -M'i<#
' 1M .0
. •I'm:;
. V .'/.~<
. 3V "3
. ^.i ' j
.'I'jfiO
.'i '.:»:
.'!•.•> 7
.•137.'
•r. •>.'
. M ' .: »
. iiin'j
. 'j"1 .'>
.ii'l/.V
. w">;o
. HI'1*!
.6031
. A.-:'3


Sr.'"..I VIVj TV PARAMETERS
    Tie K iT " I'LoNNINCrVL'ARS
NUMl'tR OF HCNE SIZES           1
HUH!."! fir MINE PROUUCTTVITIEIS
M'.I'LW Ol" MljlLJ LlVbS           3
. ".'n.. <: or r ois          i
NUnpuR CF n.i]U3          5

f3.iC'il. Milt F INAMUAl.  F'ARAMKTiTRS
f e'll K-;i" fn/' r--:.lV" "
S'V:-:KAKCL TAX  RAT:
\ii-i Ml l KEDIT
l.;",:
-------
                                   .uitiMji  nirJi     u v\ I.  'i    MM UN  I
                   MKJ:  irniiiK 1 1 vi i u '.
                   lji'ii|ii)i  ! IVMV v Mil '... MA
                                                                                              TABLE 3  -  (cont'd.)
                                                                                                                          •JO
CO
 I
                   •rii.'f " :.'!/[   uir.l/Yt.MO
                   :?."•:  IT  viMjj:  - EXPLORATION  cosrs  
                   w.i.i-J.if.;-,  r:r.
                   Pi;.!  iif  .-'.Llir    EIJfLDtNGS
                   i-'ji1!;-'!! ,'Ji     h'tir  \.n~\~
                   r -J".' :•! fl!   -  Jn Yl  A PS
                   Fi-j11. ' !• .1:       I., 'it Ik1-.-

                   Hi'  I r •  • '•-•  '.I1'", I OM'i
                   fi*'. ' ,.!  .  r'f'i-i .-  I j VI IY
                   ,'.'ll  ,  'Mil/ Yl iV-1 '
                    U  I i,l'!  H.i'S  Hl.nCH LUNG
                    I • i • --jii.i! (r;N  » : L
                    li'ln Kl :. I   I ••.'•: 1>
                    I'm, '." I  •'.'r.ki.Z
                            ,,14:-  J IjlJI ii i'CF
                    Pi" ..:if '-lil  LlJiiTIJ'ii. rOS]_S
                    ji"vt '.-fi 1 •*!  <:ii:"' I ;".", ~i«HT
                    or1: -.  i. '.•: :J>,  |:
.OiUlU
. ii :i P u
.211-8
. ''i?.'
. (•:.>> o
.0000
.0300
.0000
.ODOU
.0000
.0000
.iHiftO
.(Id no
. Dl'OO
. !} <} fl 0
. niuuj
illljO
. -K' I"1
. ll U fl '1
.0010
.00(111
.cnsio
, is ij n i.i
. fi fj U U
. f. o n ii
. prmo
.00f»0
• 7P'tJ
. 'L 7?:.'
,U.'99
.0000
.0090
. 0000
.0000
.0000
.coca
. orj'i.j
.0!>0i>
.OH II j
. 0 0 fl 0
. I'JIlllO
.00(10
.Ullfl'l
. (lilf;!1
• Ol'Uti
.0009
. I) II II (1
.0-00
.0000
.Oi-i'O
.00(10
. 0 ii 'J 0
-1.7-1
. iri.,1
.fi:.'r3
.0000
.0000
.0000
.0000
.0000
.onon
.once
.!>OuO
. (lUCO
.1)000
. ufiuO
.(IOI..O
.uoou
.PUOO
.Ofi li>)
. uaon
. r, ii i) o
.0"L-!l
. GDO'l
.O'SU'J
.0090
. CfiO't
~n. .- 'MI
. i it:.'
.o:1.,-!

-------
         costs will  be  higher because  of add-ons  'such  as
         indirect costs).

Working capital  requirements  are estimated  to  be  equivalent  to  one
month of wage payments (i.e.,  3% of  annual wage payments).

As  mentioned  previously,  the  model  mine costs  are  estimated  for  a
specific average labor productivity.   To allow for analysis of minimum
required prices as a  function  of mine  productivity,  four other values
for mine productivity  are  specified  for  each model mine  in the range
over which productivities  for  actual  mines had been found to vary 'see
Chapter II.4 of Appendix A of  this  report for a discussion of the mine
productivity analysis).

C.  Mine Flow Volumes and  Water Treatment Costs

The analysis  of  available  data on  coal mine  water  flows demonstrates
that,  when  expressed in gallons per  ton  of  coal produced,  they  are
generally  larger  for smaller  mines.   This is  shown  in  Figures 3, 4,
and 5.

Average flow  levels  obtained  from  the  data  are  highest  in  the three
Appalachian  regions,  somewhat less  in  the Midwest and  Central West,
and much less (by a factor of      about  10  )  in the  rest of the coal
mining regions (Northern Great Plains, Texas, Rocky Mountains,  Nevada,
New Mexico,  Washington).   Also, the  range of  possible  flows   is much
narrower  in  these  "dry   areas"  than   in the  "wetter"  Appalachian
regions, the Midwest and the Central  West.

The  increase  in  the  MRP as a  function  of water treatment costs for a
given  model  mine  is  calculated  for  five different  volumes,   ranqing
from  1  to  10,000 gallons  per  ton for  the wet Appalachian regions and
the  regions  in  the Midwest and Central  West  and ranging  from 0.1 to
1,000 gallons per ton for the other drier  areas.

It  is  assumed   that by  1984  all   existing   mines   will  have water
treatment  installations  in  compliance with BPT  standards.   The
increase in  required price  for these  mines  will be  through costs to
upgrade treatment  levels  from BPT  standards to  BAT standards.
Investment  costs  in  BPT equipment  are  treated as sunk  costs   and are
not considered in the calculation of the minimum  price per ton  of coal
required by the mine to stay in production.  However,  investment costs
for BPT equipment are considered in the MRP calculations  for new mines
since  the  investment has to be  incurred  to open up  the new mine.  The
estimated  BPT  investment  costs for  different model mines used in the
analysis  are shown  in  Table  4.   Operating  costs  are   judged to be
negligible  for BPT.

The  investment and operating  costs  for  the two different possible BAT
treatment  standards,  for  which the  potential  economic   impact  is
estimated,  are derived from data provided  by the  EPA.
                                 B-15

-------
o




3
I







   10
                             Mine Production, Tons/Year
              FIGURE  3     COAL MINE WATER FLOWS - APPALACHIA
                                  B-16

-------
   10
     ,5 -
tnzr 	
' 	
r*


__ _

ET==
1 'V^ •

—
kJ
L_.

	
	


	
	
=
i —


^x



-

f
; 	

I


1 	

' 	 M—
1 	 i — i—


—



r^:

_t-
-
\ 	 1
_^—_
1 	

^

— m — i
— i- — — —
i — ,_ — , — |

I
3
I
I
10°
   10   t

                          " "" '  *~	'	' •-*"• ••*••'- - V?  — *—.__.-    -  ^ •
                v-£^:^4^4.t±==
                                                                       I

       10"
                          10=
10C
10'
                              Mine Production, Tons/Year
      FIGURE  4      COAL MINE WATER FLOWS-MIDWEST AND CENTRAL WEST
                                   B-17

-------
   10=
                                          I  I I T I i
   10"
0
i
I

                                                                m

       10"
10°                  10C
 Mine Production, Tons/Year
10'
         FIGURE 5    COAL MINE WATER FLOWS - GREAT PLAINS AND WEST
                                        B-18

-------
                                  TABLE 4


                  BPT Investment Cost for New Mines (SMM)

                                 (1978 S)
                    Small      Large     Large     Large     Large     Small
                      UG        SU        UG      Texas      NGP        SU
Mine size MMT/yr    0.025     0.600     0.600     2.00      4.00      0.025


NA.CA.SA
   $MM              0.015     0.090     0.366                         0.00375
   S/TPY            0.610     0.150     0.610                         0.150


MW,CW,TX
   $MM                                  0.036     0.120
   S/TPY                                0.060     0.060


NGP,ROCKY,NEV.N.MEX
   SMM              0.0015               0.054                         0.360
   S/TPY            0.060               0.090                         0.090
                                      B-19

-------
The BAT  investment  and operating costs for  different flow  levels  as
calculated  for the  different  representative  mine  sizes  in the
different  regions are  shown  in  Tables  5,  6, 7,     and  8.   The  costs
shown  in  these  tables  are not  corrected  for  construction cost
variations between different  regions.   The  multipliers  used for  this
correction are shown in Table 9.

D.  Projection of Changes in Mining Conditions  Between 1976 and 1984

Mining conditions  are expected  to  change significantly  between  1976
and 1984 because of:

    •    Changes  in  constant dollar  unit  costs  of  labor,
         power and equipment;

    t    Changes in labor productivity.

As  shown  in  Figure  6, constant dollar  labor and  equipment costs have
increased  significantly since the early 1970s.   Mining equipment costs
have  increased  at a  very high average rate of 11%  per  year  between
1973  and  1976,  slowing  to  1% per  year between 1977 and 1979.   The
labor cost index  has  increased  relatively steadily  at an  average rate
of  about 2%  per year between 1968 and 1979.

Given the  required continued growth in coal  supply, labor costs can be
expected to  continue to  increase in real  terms.   As shown in Table 10
a  continued  annual  growth rate  of  2%  is used for the period from 1978
to  1984.   The actual  average  increase of  labor  costs between 1977 and
1978,  resulting  from the  end-1977 settlement  of  wage  negotiations
between  the mine  workers'  union and  coal  mine management,  is
calculated to be 14% over and above inflation.

Real  increases in  equipment  costs  (including   pollution  control
equipment) are  judged to  gradually  decrease  from  about  11% per year
between  1975 and  1976 to about 0.5%  per  year  between  1978  and 19R4.
The high  increase in equipment costs - 11% per year - between 1973 and
1976  are  judged  to  have been created by  temporary  bottlenecks due to
the sudden renewed  interest  in  coal  as  a  fuel.   This renewed interest
was brought  about by  increases in imported crude oil  prices  at the end
of  1973;  these  bottlenecks are expected to gradually  disappear.

Power  costs  are projected to increase  at an  average of  2% per year,
reflecting continued  real  increases  in the costs  of  all  fuels (see
Table  10).

As  shown  in Figure  7,   labor  productivity in  both underground  and
surface  mines  has  decreased dramatically  during  the  last  decade.
These  decreases  are  mainly attributable to  regulations  protecting the
mining  environment  and mine  workers'  health and safety.  This decline
is  expected  to continue  until  1981,  especially for  surface mines in
                                   B-20

-------
                                                          TABLE  5
to
i
to
BAT COSTS
(Without
Flow G/T
Flow MMG/D
Level 2
Log of Flow
INV $MM
OPC $MM/Y
Level 4
Log of Flow
INV $MM
OPC $MM/Y
FOR SMALL MINES (0.025 MMT/yr) WITH
Correction for Differences in Mining
104
0.685

-.1643
.0350
.0115

-.1643
.2575
.0421
103
0.0685 0.

-1.1643 -2.
.0350
.0063

-1.1643 -2.
.1494
.0271
DIFFERENT
Region or
102
00685

1643
0350
0060

1643
1179
0270
FLOW VOLUMES
in Mine Type
10
0.000685

-3.1643
.0350
.0060

-3.1643
.1100
.0270

1.0
0.0000685

-4.1643
.0350
.0060

-4.1643
.1100
.0270

-------
TABLE 6
BAT COSTS FOR LARGE UNDERGROUND MINES (0.6
(Without Correction for Differences in
Flow G/T
Flow MMG/D
Level 2
7 Log of Flow
to
INV $MM
OPC $MM/Y
Level 4
Log of Flow
INV $MM
OPC $MM/Y
104
16.438
1.2158
.0350
.0485
1.2158
.9039
.1979
103
1.6438
.2158
.0350
.0162
.2158
.3433
.0580
MMT/yr) WITH DIFFERENT FLOW VOLUMES
Mining Region or in Mine Type
102
.16438
-.7841
.0350
.0075
-.7841
.1772
.0299
10
.016438
-1.7841
.0350
.0060
-1.7841
.1236
.0273
1.0
.0016438
-2.7841
.0350
.0059
-2.7841
.1187
.0250

-------
                                                      TABLE  7
bo
t
BAT COSTS FOR LARGE STRIP MINES (2.0 MMT/yr) WITH DIFFERENT FLOW VOLUMES
(Without Correction for Differences in Mining Region or Mine Type)
Flow G/T
Flow MMG/D
Level 2
Log of Flow
INV $MM
OPC $MM/Y
Level 4
Log of Flow
INV $MM
OPC $MM/Y
103
5.479

.7387
.0350
.0278

.7387
.5481
.1030
102
.5479

-.2613
.0350
.0107

-.2613
.2410
.0393
10
.05479

-1.2613
.0350
.0061

-1.2613
.1441
.0267
1.0
.005479

-2.2613
.0350
.0060

-2.2613
.1171
.0260
0.1
.0005479

-3.2613
.0350
.0050

-3.2613
.1100
.0250

-------
                                                          TABLE 8
w

to
*.
BAT COSTS FOR LARGE STRIP MINES (4.0 MMT/yr) WITH DIFFERENT FLOW VOLUMES
(Without Correction for Differences in Mining Region or Mine Type
Flow G/T
Flow MMG/D
Level 2
Log of Flow
INV $MM
OPC $MM/Y
Level 4
Log of Flow
INV $MM
OPC $MM/Y
103
10.96

1.0398
.0350
.0392

1.0398
.7455
.1535
102
1.096

.0398
.0350
.0138

.0398
.2988
.0495
10
0.1096

-.9602
.0350
.0068

-.9602
.1629
.0283
1.0
0.01096

-1.9602
.0350
.0062

-1.9602
.1207
.0282
0.1
0.001096

-2.9602
.0350
.0060

-2.9602
.1200
.0280

-------
TABLE 9
Region
NA
CA
SA
MW
CW
Texas
NGP
Rocky Mtns.
Nev., N. Mexico
MULTIPLIERS TO CORRECT FOR REGIONAL
DIFFERENCES IN BAT INVESTMENT COSTS
Multipliers
All Surface
and New Underground
1.32
1.32
1.28
1.12
1.08
1.00
1.00
1.36
1.26

Existing
Underground
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
  B-25

-------
FIGURE 6
:i
•i
• • 1 .• 6 •-
• • • I ; l-.
r
-r.2-
s
T
	 1
1
N
It
£ D'8
X
' ' D'.i
Oil






i
^r-

i
4 	
50 51
1

-


y*-*^
*^~-



™ S3
HIS'



*/
|**>S
•*——"*"



5* c-.-
55
FORICAL


z~*~
^
~-* —
^•"


lj u ,1
5A 57
MINING LABOR C



^..i.. 	



J 	 1 	
56 ay

& 	 £,
V""----V 	
*-— -•*_. J



	 .!„
60 61
OST AND



~ 	 a... ...j
* ~---~



	 i . - r
62 63
YEAR
MINING EQUIPMENT COST

! :

t^-r. -ja~.'...T3
	 -H
'"""


* M1N1N
A .TOTAL
V UNlifc.h
64 65
^•-"~i-^
^— __^"


INDICE



jS_— «•


", LABOR COST 1NȣX
.MIMING -EQUIPHEW-T .CU51 .1
iKQIIND ri;IN]NG fciUUlPMEMT
66 67 ^ 69
70 71
B (1972=


.~*-^
~^i


iJ!>tX . . .
COS! JUT
1 	 .1. ._
72 73
=100)

"?/
../•./..
. /
i/ ^
ft _-"<
(i--"




L"X
	 1 —
74 75
•
>7
. V
r-"""*'

_...>——-*.
*• 	 -*'"

:

•
.. .-J_ .L 	 L. .J
76 77 i78 79





	
60

-------
                                            TABLE 10
PROJECTED ESCALATION IN COSTS AND LABOR PRODUCTIVITY
USED IN THE MODEL MINE COST ANALYSIS
(1977 is 1.00)
Year
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986 and
following
year
Wages
1.000
1.140
1.157
1.157
1.180
1.203
1.228
1.253
1.277
1.02x
Power
Costs
1.000
1.020
1.040
1.061
1.081
1.104
1.126
1.149
1.172
(1.02x)
Mining
Equipment
Cost
1.000
1.032
1.048
1.056
1.061
1.067
1.072
1.077
1.083
(1.005x)

Underground
1.000
0.990
0.985
0.980
0.990
1.000
1.010
1.020
1.030
(xl.Ol)
Productivity
Surface
East. Regions
1.00
0.90
0.85
0.80
0.808
0.816
0.824
0.832
0.841
(xl.015)

Surface
Other Regions
1.00
0.95
0.92
0.90
0.914
0.927
0.941
0.955
0.970
(xl.015)
(1)  Surface other includes  strip mines  in the West, Midwest and Texas

-------
                                                          FIGURE  7
                       HISTORICAL  PRODUCTIVITY INDICES FOR UNDERGROUND AND SURFACE MINES  (1967=100)
ea

N5
00
                                                                * .UNDERiROUND.HINE.PROI JC1IVITY

-------
the  Eastern  regions  (the  relatively  small  surface  mines  in these
regions  are  expected  to  suffer   most   from  the   Surface  Mine  and
Reclamation Act; the  difficult  terrain will  make it costly to  restore
mined areas to  conditions  required  by  the Act).  Beyond 1981,  produc-
tivity is expected to rise again at  an average  rate of 1.5*.  per year,
mostly due to a maturing work force.

E.  Minimum Required Price Calculation and MRP Function Estimation

Given the  input parameter  values  for mining,  productivity  and water
treatment  costs  discussed  in  the  previous  sections,   the  minimum
required  prices could  be calculated  for different  mine  categories.
The  general  algorithm  of  the  computer  program  developed  for this
purpose  is  shown  in  Figure 3.   Given  input parameter  values for  a
specific  type  of  mine  model,  the  program  can   calculate  minimum
required prices for:

    •    Different planning years in the  future;

    t    Different treatment cost levels;

    •    Different remaining lives for the mine;

    •    Different mine labor productivities.

Originally the computer program allowed investment  costs  and  operating
costs to change with  mine size.   However, analysis  of the  sensitivity
of MRPs  to  changes in  mine size and  productivity  revealed  that MRP
variations  are  dominated by  productivity variations.   Mine size was
subsequently dropped  as an  independent varirable in the  calculations.
This significantly reduced the number of  computations  required  for the
analysis.  As shown in Figure 8, the results  are ranges of  values for:

    •    Minimum required price (in  dollars per ton);

    •    Annual  wage  payments Mn dollars per annual ton^;

    •    Cashflow generated in the planning year (in dollars
         per annual ton);

    t    Required  investment  in  mining  equipment  in   the
         planning year  (in dollars per annual ton>.

These calculated  values (i.e.,  the   dependent variables*  are retained
in relation to  the  pre-specified  values  of the independent variables,
which are:

    •    Labor  productivity (in tons per  miner  shiftV,

    •    Mine  water  flow volumes  (in  million  gallons  per
         year).
                                  B-29

-------
                   FIGURE  8
FLOW DIAGRAM OF MODEL MINE  COST ANALYSIS PROGRAM
                Read data
          Next water flow
                   I
           Next planning


                    I
          Next  remaining


                    I
   Calculate investment requirements
   Project after- tax operating costs

     for different productivities
   Calculate minimum required  price,

   wages, cashflow,  required  remain-

   ing investment,   treatment  opera-

      ting and investment  costs
              Store  results


                    f
          Next remaining  life?


                    |
           Next planning  year?


                    I
          Next water flow  level?


                    *
                    NO


                    *
                   STOP
                    B-30

-------
These data are used to estimate parameter values  for  MRP functions as
described in Section H below.
F.  Details of the Minimum Required Price Calculation
The minimum  required  prices  are  calculated subject  to  the following
two conditions:
CONDITION 1:  After tax revenues should cover all  costs plus a return
1)  Sum (PV (INV. (l-credit}+ WC + OC . (1-t) - DEPP  . t^ =
    Sum (PV (PROD . MRP . (1-roy - sevtax^ . 'l-'l-depP  . t ^
CONDITION 2:  In each year of the planning period
2)  [PROD  . MRP  .  (1-roy -  sevtax) . (1-fl-depl) . O  -  OC .  'l-t> +
    D'EPR . t"1 = T(PROD . MRP  . (1-roy - sevtax)]  . CONSTANT
                              where,
    MRP     the minimum  required  price in a given year
            of the mine's life.
    INV     the  investment  in a  given year,  including
            investment in water treatment equipment.
    OC      the • operating  costs  in  a  given  year,
            including water treatment  costs.
    DEPR    the depreciation  allowance  in a  given year.
    depl    the depletion allowance  rate.
    PROD    the production  in a given year.
    credit  the  investment  credit  rate.
     t      the  federal  corporate  income  tax rate.
     roy    the  royalty  rate.
     sevtax  the  state  and  local severance  tax  rate.
     PV      the  present  value operator for  year t.
     Sum    summation  over  the  remaining  mine  life.
     WC      the  working  capital requirement (treated  as
            an investment  occurring  in the first  year
            of the mine's  life).
                                   B-31

-------
The actual value of  the  "CONSTANT" will  result  from the calculations.
For example, if the remaining mine life is  30 years, then we will  have
30 equations  of type  (2)  and  one equation of  type '1^, or ?l  equa-
tions, which  in  order to have a  unique  solution will  need  to  have 31
unknowns:  30 MRPs (one MRP for each  year)  and the "CONSTANT."

The  calculation  of  the  MRPs  and the  "CONSTANT"  is broken  down  into
four steps:

    STEP 1:   Calculate:  P*t I*, d*, where:

              P* = PROD . (1 - roy -  sevtax^

              I* = Sum . fPV (INV .  '1 - credit) + WO)

              d* = (i - (1 - depl) .  t )

    STEP 2:   Calculate:   A*  = Sum  'PV  (OC  '1 -  O  -
              DEPR .  t ))

    STEP 3:   Calculate:   CONSTANT =  d*  -  d* .  A*/ (A*
              + I*)

    STEP 4:   Calculate:  MRP =  'OC  .  '1 - t.)  - DEPR .
              t)/  (P*  . (d* - CONSTANT) =

              A/ (P*  . {D* - CONSTANT, WHERE:

              A =  OC  . (1 - t) - DEPR . t

              Substitute:  (d* -  CONSTANT)  .  =  A*  . d*/
              (A*  +  I*)

              MRP  =  A . (A* + I*) / P* . A* . d*

To  relate changes in  costs with changes in  productivity,  a standard
production function  is used:

    C/C* =   (PRTY*/PRTY)F, where:

    C     =   the unit operating or investment costs 'in
             dollars   per  ton)  related  with   a  produc-
             tivity PRTY (in tons  per miner shifO.

    C*   =   the unit  operating  or  investment  costs
             related  with  the benchmark productivity
             PRTY*.

    F     =   a factor with  a value between zero  and  one.
                                   B-32

-------
Higher labor  productivity within  the same  mine  implies  lower  unit
labor costs and higher mining equipment costs.   Therefore,  a positive
value of F relates labor costs with productivity and  a  negative valup
of F  relates  equipment costs  with productivity.   In the  analysis  a
value of  +0.8 was  used to  relate unit  equipment  costs  with  labor
productivity;  a value of -0.8 was used to  relate unit labor costs with
labor productivity.

G.  Sample Output

A detailed example of the results of the calculations described in the
two  previous  sections is shown  in Table  11.   The results  are  for  a
large underground mine  in Northern Appalachia,  where  mine  water flows
are  mainly acid  (see  Table  3  for  input  parameter values).    The
treatment costs are for level 4.

Table 11 shows the results of the calculations for control  cost levels
corresponding with  a high,  medium and  low flow  for  an  existing mine
and for a new mine.  The price difference, because of different flows,
can be as high as $.50/ton.   The  price difference because of different
productivities can be as high as  $18.9/ton.

H.   Estimated Function Parameter Values for MRP, Wages. Cashflow and
    Required Investment in Mining Equipment

The  last  step in the  calculations performed by  Module  2  consists of
the estimation of function parameters relating the dependent variables
- minimum  required  price, wages,  cashflow  and  required  investment in
mining equipment  (all  in  dollars per annual ton  produced)  - with the
independent variables -  productivity (in  tons  per miner  shift)  and
water flow level   (in gallons per  annual ton of coal produced^.

The  following  functional  formula  provided  a  good correlation between
the  values of  dependent  variables  calculated  by the program described
in the previous  section,  and the pre-specified  values of the  indepen-
dent variables:

    DV    = LogL) + b  . Log(FLOW) + c . Log'PRTY)
                a

    DV    = the   dependent   variable  (MRP,   wages,
            cashflow,  investment  in  mining equipment,
            all in dollars per ton)

    PRTY  = productivity (in tons  per miner shift^

    FLOW  = mine  water  flow  volume (in million  gallons
            per year)

     a,b,c = estimated parameter  values

The  parameter values  a, b, and  c  are  estimated using  a standard,
ordinary least-squares  regression  program.


                                   B-33

-------
                                    TABLE 11
                         RESULTS OF MODEL MINE ANALYSIS
                 LARGE UNDERGROUND MINES IN NORTHERN APPALACHIA:        m
           REQUIRED  PRICE. WAGES, CASHFLOW. INVESTMENT. TREATMENT COSTSu;
HIGH FLOW (10,000 G/T)




Price
Wages
Cashflow
(2)
Investment* '
Treatment Cost:
Operating
Investment

Price
Wages
Cashflow
(2)
Investment '
Treatment Cost:
Operating
Investment
NEW
Labor
Low
7
43.1
13.4
6.7
35.2

0.38
2.1

42.6
13.4
6.6
35.2

0.06
0.42
MINE
Productivity
Medium
10
35.0
10.1
6.7
43.6

0.38
2.1
Medium Flow
34.4
10.1
6.6
43.6

0.06
0.42
OLD MINE
(T/MSH)
High
24
24.2
5.0
8.6
77.7

0.38
2.1
(100 G/T)
23.8
5.0
8.5
77.7

0.06
0.42
Labor
Low
7
41.5
13.4
6.0
15.1

0.38
1.6

41.2
13.4
5.9
15.1

0.06
0.32
Productivity
Medium
10
33.3
10.1
5.9
20.1

0.38
1.6

33.0
10.1
5.9
20.1

0.06
0.32
Low Flow (1 G/T)
Price
Wages
Cashflow
(2)
Investment* '
Treatment Cost:
Operating
Investment
42.6
13.4
6.6
35.2

0.05
0.28
34.4
10.1
6.6
43.6

0.05
0.28
23.8
5.0
8.5
77.7

0.05
0.28
41.2
13.4
5.9
15.1

0.05
0.21
33.0
10.1
5.9
20.1

0.05
0.21
                                                                                   High
                                                                                    24
                                                                                    22.8
                                                                                     5.0
                                                                                     7.9
                                                                                    40.4

                                                                                     0.38
                                                                                     1.6
                                                                                    22.^
                                                                                     5.0
                                                                                     7.8
                                                                                    40.4

                                                                                     0.06
                                                                                     0.32
                                                                                    22.4
                                                                                     5.0
                                                                                     7.8
                                                                                    40.1

                                                                                     0.05
                                                                                     0.21
1
 All  costs are in dollars per annual ton.
"Required investment in mine equipment (and facilities) to open the mine or
 to keep the old mine producing.
                                        B-34

-------
Table 12 shows the estimated values for a,  b,  and c for the same mine
model used  as an example  in  the previous  section.  As  shown by the
values of the  correlation  coefficients (see Table  12\ the values  of
MRP, wages, and  required investment  in mining equipment, estimated  by
the  functions,  deviated less  than  !<&  from the  original  values; the
estimated cashflow  values   deviated  less  than 7%  from the orgininal
values.

V.  REGIONAL SUPPLY CURVES  (MODULE 3^

Module 3 operates on  the regional mine files with  production  and pro-
ductivity information  for  individual  mines  in  1975 obtained  from the
MESA  file.   As  shown  in  Figure 9,  Module 3  performs  six main  func-
tions:

    •    Organizes the mines in each region  into  three  major
         categories  (met  coal,  contract  and  spot market
         steam coal);

    •    Changes  the  population on these  regional   files  by
         retiring old mines and opening new mines;

    •    Estimates mine waterflows  for the individual  mines
         and  calculates the required  investment and opera-
         ting costs to  treat these flows;

    •    Calculates  estimates   for  minimum  required price,
         wages,  cashflow and  required  investment  for  these
         mines;

    •    Rank orders  the resulting  regional mine  population
         by increasing  price;

    •    Fits a  linearized  function to  the  resulting list  of
         mines,  relating  increasing  price  with  increasing
         production from the region.

A.  Separation into Major Coal  Markets

As  explained  in  the introduction it  is  assumed that the steam coal  and
the metallurgical coal  markets  operate  completely independently  of
each  other.   The steam coal market consists  of the  spot  and  contract
markets;  these markets  are assumed to  have only  limited  interaction
with  each other.

In  order to  create three  different mine  files  for each  region,  two
decision rules are used:

     •   All  underground mines with  productivities  of less
         than  nine tons per miner shift are assigned to the
         met  coal file;
                                   B-35

-------
                                   TABLE 12
                        ESTIMATED FUNCTION PARAMETERS FOR
                          LARGE UNDERGROUND MINES ON
                             NORTHERN APPALACHIA
New Mines
Dependent Variable
Minimum Required Price
Wages
Cashflow
Investment In
Mining Equipment

Existing Mines
Dependent Variable
Minimum Required Price
Wages
Cashflow
Investment In
Mining Equipment
Function
a
104.9542
63.6449
4.2516
10.0138
Function
a
105.2204
63.6449
3.603
3.1785
Parameters
b
0.001538
0.0
0.001842
0.000
Parameters
b
0.001104
0.000
0.000431
0.000

c
-0.4731
-0.800
0.2086
0.6419

c
-0.4934
-0.800
0.2316
0.800
Correlation Coefficient
    0.99474
    1.000
    0.9327
    0.99966
Correlation Coefficient (R )
  0.99409
  1.000
  0.93239
  1.000
 1
  The function parameters a,b, and c relate the dependent variables (MRP, wages,
  cashflow or investment) with the independent variables, mine labor productivity
  (PRTY) and log of flow.
              DV = log  x log of flowb x PRTYC
                      a
                                         B-36

-------
                               FIGURE 9


                       MAIN  FUNCTION OF MODULE  3:

          ESTIMATION  OF REGIONAL  COAL SUPPLY BY PRICE  in  1984
Met Coal  Mines
   in 1976
                     Production  Data  for  Existing
                            Mines  in  1976
Spot Marker Mines
    in 1976
                      Old &  New  Miies  in  1984
                                O
                      Old & New Mir
                                                  Separation  Rule
•Contract Market Mines
        in 1976
                                                   Mine  Attrition  Rate
                                                   New Productive  Capacity


                                                   New Mine  Parameters:
                                                   Size,  Productivity
                           Flow Distribution  (T
           es with Flows
                                                  .Price and Cost Function  (4)
                                                   Parameters
                      Old & New Mines  with  Price
                           and  Cost Data
                                                  -Rank Order by Ascending Price
                      File with Mines  in  1984  by
                          Ascending Price
                                                  •Linearize
                      Linearized Supply Curves
                                       B-37

-------
    •    All small mines  -  mines  with  less than 51,000 tons
         of  production in  1976  - are assigned  to the spot
         market file.

These decision rules are  admittedly crude.  However, the  1976  supplies
from  these  met coal  mines  obtained  in this  manner for  the different
supply regions corresponded  remarkably well  with the available  infor-
mation on actual supplies from the same regions.

The hypothesis that underground mines  with low  productivities  are  most
probably  metallurgical  coal  mines  is based  on  two   considerations.
Firstly, metallurgical coal  is a  relatively scarce  resource, occurring
mostly in  deeper  and thinner seams that  are  generally more difficult
to mine.   Generally,  therefore,  met coal   mines  have lower labor  pro-
ductivities  and  correspondingly  higher  production  costs.   Secondly,
underground  mines  with productivities  'as measured  in this analysis^
below nine  tons per man-day  are not competitive  at  today's  prices.

As  shown in  Table 13,  the total  amount of  coal  ourchased  in  spot
markets  by  utilities  in  1976  is about double  the amount  produced  by
small mines  assigned to  the spot  market  file, using the  decision  rule
mentioned above.  The  differential amount  of coal  supplied  to  the  spot
market  in  the  analysis   is  taken  from  contract  mines,   using   the
percentages  shown in Table  14 to  estimate  the relative  volumes of  spot
market  coal from  the  different  supply   regions.   These  additional
volumes  are assumed  to  be  available   at  variable  costs  because fixed
costs are already paid for  by contract sales.

B.  Changes  in the Mine Population

Steam  coal   supply  is  expected  to grow   at  an  annual   rate  of  4-5*,
between  1979 and  1984.   In the analysis it is  assumed  that all  of the
increased  production will come from large mines.   Small  mines  'i.e.,
those with  less  than  50,000 tons  per year  production)  will  be  more
affected by  the various environmental   regulations  'including the water
control  standards  analyzed here).  Also,  the majority of  coal  use  in
the future  is expected to be by electric  utilities,  which will be  more
interested  in  longer  term contracts  because   of  their emphasis  on
security of  supply.

The  estimates  of  the  number of  large  mines  closing between  1976  and
1984  are calculated  using  annual  attrition  rates  obtained   from  the
MESA  Mine  File analysis.   The resulting  attrition rates, compounded
for the  total  period of nine years between 1976  and  1984, are  shown  in
Table 14.

In  the  simulation  of  changes   in  regional   mine populations,   the
smaller,  least  productive mines  from  the 1976  mine file  are retired
first.   For mine  sizes   smaller  than  200,000  tons, the size of  the
replacement mine  is assumed  to be twice the size of  the old mine.   The
                                   B-38

-------
                                      TABLE 13
Origin

Pennsylvania
Ohio
Maryland
Northern WV
Southern WV
Virginia
E. Kentucky
W. Kentucky
Tennessee
Alabama
Illinois
Indiana
Iowa
Missouri
Kansas
Oklahoma
Texas
North Dakota
Montana
Wyoming
Utah
N. Colorado
S. Colorado
Ari zona
New Mexico
Washington
  TOTAL
CONTRACT AND SPOT MARKET COAL SUPPLIED TO ELECTRIC UTILITIES IN 1976
(000 Tons)
Total
ia 40,940
40,135
2,412
IV 26,981
IV 17,558
13,642
:y 52,722
,y 49,440
7,225
13,940
49,140
23,817
601
4,056
2,317
2,503
11,867
ita 10,031
24,958
25,781
4,632
lo 1 ,487
lo 4,174
10,258
i 8,465
i 3,600
452,682
Contract
29,058
31,265
902
23,402
15,883
11,589
39,225
44,644
6,154
9,967
45,897
20,078
444
4,019
1,898
1,832
11,855
10,011
24,858
24,956
4,391
1,041
4,053
10,258
8,465
3,600
389,745
Spot
11,882
8,870
1,510
3,579
1,675
2,053
13,497
4,796
1,071
3,973
3,243
3,739
157
37
419
671
12
20
100
825
241
446
121
0
0
0
62,937
Percent
Spot
29.0%
22.1%
62.6%
13.3%
9.5%
15.0%
25.6%
9.7%
14.8%
28.5%
6.6%
15.7%
26.1%
0.9%
18.1%
26.8%
0.1%
0.2%
0.4%
3.2%
5.2%
30.0%
2.9%
0.0%
0.0%
0.0%
13.9%
Spot as Percent
of
Total Spot
18.9%
14.1%
2.4%
5.7%
2.7%
3.3%
21.4%
7.6%
1.7%
6.3%
5.2%
5.9%
0.2%
0.1%
0.7%
1.1%
0.02%
0.03%
0.20%
1.3%
0.40%
0.70%
0.20%
0.00%
0.00%
0.00%
100.00%
    :  FPC Form 423
                                          B-39

-------
                               TABLE  14

                       ATTRITION RATE PARAMETERS
     The following tables specify the attrition rates used for mine retire-
ment and replacement simulation when projecting changes in the contract mine
population between 1976 and 1984.

     The data are organized in three sets of four records (a record is a
line in the table).  The first set specifies attrition rate parameters for
underground mines and the last set of records is for surface mines.

     The first line in each set of records has the attrition rate for the
corresponding size range; the second and third records give, respectively,
the lower and upper limits of the mine size range; and the fourth record gives
the maximum mine size for replacement mines.  The last or thirteenth record
in the table specifies the increase in mine size of replacement mines:  mines
with sizes of up to half a million tons, which were closed between 1976 and
1984, were assumed to be replaced with mines twice the size of the closed
mines.
     Annual  attrition  rate:
     Mine  size:  lower  limits
     Mine  size:  upper  limits
     Maximum Mine  Sizp:

     Annual  attrition  rate:
     Mine  size:  lower  limits:
     Mine  size:  upper  limits:
     Maximum Mine  Size:
     Replacement mine  size
     Multiplier:
0.52  0.35
0.05  0.10
0.10  0.20
in.o  in.o
0.19   0.13   0.131
0.20   0.50   1.00
0.50   1.00  10.00
10.0   10.0   10.0)
underground mines
0./5
0.05
0.10
10.0
2.00
0.59
0.10
0.20
10.0
2.00
0.35
0.20
0.50
10.0
2.00
0.25
0.50
1.00
10.0
1.00
0.19"
1.00
10.00
10.0
1.00
                               .surface mines
  Notes mine  size  in millions of tons per year.
                                   B-40

-------
productivity of the replacement mine  is  obtained  by  the sampling of a
productivity  distribution  derived  from  analysis of  the MESA  tape.
These productivity distributions are shown in Table 15.

The simulation of the closing  of  old  mines  is  continued until all are
retired.    Simultaneously,   the  opening  of  replacement,  mines  is
simulated,  insuring that the  sum  total  of productive capacity for thp
individual mines does not exceed  total  projected  productive rapacity.
If, after  the  retirement  of  old mines  and  the  opening of replacement
mines, additional productive capacity  is  still  needed,  the simulation
continues to add new mines.   The simulation of new mines starts at thp
high end of the mine size range and  gradually works toward the low end
until   all  projected  new  capacity  for the  region  is  realized.   The
percentage distribution of mines  over  different mine size categories,
determined  for  the  1976 mine  population (see  Table 16^,  is  used  to
allocate  new  mines  to  different  size classes.   The  result of  the
simulation  is  a  file  containing  old,   replacement,  and new  mines.
Table 17  shows  the  projected  increases in capacity  for the differpnt
regions.

For each  mine  added  to the  mine  file  the  annual   water flow  to  be
treated is  estimated.   This is  done by calculating  an average flow as
a  function  of  the  size of  the mine   and by sampling  a  distribution
which specifies the spread around that average.   The parameter values
for, and functional  relationships  between, average flow and mine size,
and the distributions of  the  spreads   around the  average  are shown in
Table 18.

Given  the  size  and  productivity  of  the  mine  and the  watpr  flow
estimates, the  minimum  required  price, wage  payments, cashflow,
required  investment  in mining equipment,  and  additional  required
investment  in water treatment equipment  can be  calculated.   This  is
done with  the functions estimated  in  Module 2  by modelling  the mine
cashflow analysis,  as  discussed in Section TV.

C.  Detailed and Linearized Supply Curves

The result  of this  sequence  of calculations is a list  of old and new
mines in the region  under analysis.   This list,  after being ordered by
ascending  price, is  stored  for future  use in the impact analysis.

Table 19  shows  an  example of  this mine file  for   Pennsylvania  with
estimated  water treatment costs specified for BAT level 4.  This list
of mines,  specifying  cumulative potential  coal  supply  in  1984  by in-
creasing  price,  is  essentially the projected "detailed"  supply  curve
for that region  in  1984.

As  shown  in Table  20,  linear  approximation of  the detailed  supply
curve  is   used   in  the  coal  market  simulation  model.   The  linear


   'including replacement mines
                                   B-41

-------
                               TABLE 15
                    MINE PRODUCTIVITY DISTRIBUTIONS

     The following tables (15.1,  15.2,  15.3,  15.4)  contain data  on  mine
productivity distributions, used  to obtain  replacement and new mine pro-
ductivity estimates as discussed  in Chapter I.
     The first three lines in the tables contain parameter values to cal-
culate average productivity (in tons per miner  shift)  as a function of
mine size for underground mines, a dummy mine type and surface mines,
respectively.  The next three lines contain, .nine fractile distributions
of the ratio of the actual  versus average calculated productivity.   The
fractiles   for these distributions are-0.0,  0.01,  0.05, 0.25, 0.50,
0.75, 0.99, and 1.0.

     The first parameter in each of the first three lines of the tables
specify the average tons per mineshift of mines with an annual production
(in million tons per year) as specified by  the  second parameter value.
The third and fourth parameter values (P3 and P4)  allow calculation of the
average productivity (PRODTY) for a mine with a size of S million tons per
year by

                      PRODTY = P3 + P4 . In (In S)
                                  B-42

-------
                 TABLE  15.1:  LARGF UNDERGROUND AND SURFACE MINES III IMF EAST
(1)
(2)
(3)
(1)
(2)
(3)


(1)
(2)
(3)
(1)
(2)
(3)


(!)
2)
(3)
1
(2)
(3)


0)
(2)
(3)
(1)
(2)
(3)



0.60
0.00
0.35





0.65
0.00
0.39





0.65
0.00
0.85





0.65
0.00
0.85
12.0
0.00
25.0
0.60
0.00
0.40


15.
0.
30.
0.71
0.00
0.41


15.
0.
100.
0.71
0.00
0.85


15.
0.
60.
0.7]
0.00
0.85
200. -2.72 9.50 6.0
0. 0.00 0.00 0.0 average produr. i« i LV para'-flrr-.
200. -1.00 15.79 10.0
0.650.75 0.95 1.201.90 1.95 2.00 Drodllcll », ty <"s - ,-,!,.-Li0-s ( ,< ,
o.ooo.oo ooo o.ooo.oo o.oo o.oo „? «r .,L , '. «H, Z7i, , •
0.460.67 0.87 1.041.30 1.50 1.50 ratio of «,vei ,•-••„, oducliviL-, ,
TABLE 15.2: LARGE UNDERC-WiD W2 SUT/'CE MINES
IN THE MI QUEST AND CENTRAL WEST
200. -2.72 11.12 6.0
0. 0.00 0.00 0.0 average product i>-i 'y parameters
200. -1.00 18.50 10.
0.84 0.94 1.00 1.C3 1.16 1.29 1.35
0.00 0.00 0.00 0.00 0.00 0.00 0 00 productivity diM t ibutions (as a
0.47 0.67 0.88 1.03 1.33 1.48 1.50 ratio of average productivity)
TABLE 15.3: LARGE UNDERGROUND AND SURFACE MINES
IN THE NORTHERN GREAT PLAINS
200. -2.72 11.12 6.0
0. 0.00 0.00 0.0 averanc productivity parameters
5000. .-226 .060 152.25 30.
°:« S-S J:§S i:!S 0:Jo i:3 J:™ ^^^ distributions {u .
0.870:920.99 K03 1.06 1.17 1.20 ratio of average productivity)
TABLE 15.4. LARGE UNDERGROUND AND SURFACE MINES
IN THE ROCKY MOUNTAINS, SCUTHl.'EST AND TEXAS
200. - -2.72 11.12 6.0
0. 0.00 0.00 0.0 average productivity parameters
5000. -24.76 39.94 20.
0.840941.00 1.031.16 1.29 1.35 _ ., _t ,„, tu ., tf.hllt.nn, / „ .
n nn n nn n nn n 11 n n« n nn n nn productivity distributions Us a
0.00 0.00 0.00 0.10 O.Ow 0.00 0.00 ratio of averagc productivity)
0.86 0 93 0.99 1.03 1.06 1.17 1.20
(1) Underground mines
(3) Surface mines
                                           B-43

-------
                                      TABLE 16
 1. ALA
 3. ARI
 4. ARK
 5.  COLO
 6.  ILL
 8.  IOWA
 7.   INO
 9.  KAN
14.  EKY
15.  WKY
12.  MD
Underground
Surface
Underground
Surface
Underground
Surface
Underground
Surface
Underground
Surface
Underground
Surface
Underground
Surface
Underground
Surface
Underground
Surface
Underground
Surface
Underground
Surface
EXISTING PRODUCTION OF CONTRACT MINES IN 1976
IN DIFFERENT SIZE RANGES
(Size
0.05-0.1
nd 0.0
2.706
nd

nd 0.
0.371
nd 0.
0.0
-------
           Table 16 cont'd



0.05-0.1    0.1-0.2    0.2-0.5    0.5-1.0    1.0-10.0    TOTAL
• j. MISS

16. NMEX

19. OKLA

18. OHIO

20. PENN

28. MONT-E

'" ND

SD

22. TENN

23. TEX

24. UTAH

25. VA

10. WVA-N

Underground
Surface
Underground
Surface
Underground
Surface
Underground
Surface
Underground
Surface
Underground
Surface
Underground
Surface
Underground
Surface
Underground
Surface
Underground
Surface
Underground
Surface
Underground
Surface
Underground
Surface
0.
0.


0.
0.48B
0.087
3.476
0.835
5.956






0.422
0.540


0.
0.
3.213
3.076
.100
0.500

0.

0.101

1.022
0.305
4.646
2.461
8.551


0.
0.128


0.936
1.743


0.960

2.330
3.109
.100
0.

0.

0.

0.222
0.320
6.329
6.602
11.120

0.288

0.747


0.516
1.419


1.289

4.2C7
2.097
.300
0.

3.058
0.830
0.865

0.
5.649
5.507
6.111
8.129

0.

0.


(0.844)


0.282
4.364

0.692
0.514
.80
.45
0.
2.636 5.797
0.830
7.979 8.947
0.
1.618 3.292
4.211 10.4
7.701 27.1
9.613 25.05
2.183 34.67
0.
25.792 25.795
0.
11.389 12.269


1.881
3.349
0. 0.
19.216 14.2
1.038 7.6R7
0.
9.989
8.463
15.00 16.715
6.350 7.026
                 B-45

-------
                                     Table 16 conc'd
11.  WVA-S







26.  WASH







29.  WYO





  Total

Underground
Surface
Underground
Surface
Underground
Surface

0.05-0.1 0.1-0.2 0.2-0.5
.150 .050 .300
.100 .100 .200
0.
0.
0.
0.
34.465 50.877 69.953
0.5-1.0
0.500
.550



2.417
83.577
1.0-10.0
25.40
9.90

4.084

26.479
301.289
TOTAL
26.779
10.991
0.
4.084
0.203
28.985
540.151
                                         B-46

-------
                                              TABLE_J_7
                     ESTIMATED POTENTIAL  INCREASES  IN MINING  CAPACITY UNTIL 1984
 1-PENN
 2-OHIO
 3-MARYLAND
 4-rf. VA. -N
 5-iJ. VA. -S
 6-VIRGINIA
 7-E. KY.
 8-TENNESSEE
 9-ALABAMA
10-ILLINOIS
11-INDIANA
12-W. KY.
'    OWA
It-HlSSOURl
15-KANSAS
16-ARKANSAS
17-OKLAHOMA
18-TEXAS
19-N. DAK
2CI-S. DAK
21-MONTANA E.
22-MONTANA U.
23-UYOMING
24-COLO - N
25-COLO - S
26-UTAH
27-ARIZONA
28-N. MEXICO
29-WASHINGrON

NEW
UG
15.9
16.6
1.8
1.5
13.8
3.1
10.4
0
4.0
22.6
2.0
3.5




2.0

MINES
SURFACE
0.4
2.6
1.0
8.3
1.6
0
21.5
0
4.3'
9.6
10.6
4.5




3.1
MHT Per
Year
REPLACEMENT MINES f
UG
4.6
1.5
0.05
2.2
3.5
3.4
6.8
1.0
0.41
4.2
0.13
3.1





SURFACE
15.3
10.4
0.26
1.7
2.3
5.0
14.3
0.50
7.3
6.6
6.1
7.1
0.13
1.3
0.13
0.28
1.4
                         0.39
                         1.3
                      1.4
                                                                        NEW + REPLACEMENT MINES
                                                       UG
                                                       20.5
                                                       18.5
                                                        2.0
                                                        4.0
                                                       17.5
                                                        6.5
                                                       17.5
                                                        1.0
                                                        4.5
                                                       27.0
                                                        2.5
                                                        6.5
                                                        2.0
                               7.0
                               4.5

                              18.0
       SURFACE
         16.5
         13.5
          1.5
         10.5
          4.0
          5.0
         35.0
          0.5
         11.5
         16.5
         17.0
         11.5
          0.5
          1.5
          0.5
          0.5
          4.5
         40.0
         70.0
          0.0
         75.0
          0.0
        220.0
         11.0
          0.0
          7.5
          2.5
         35.5
  ,U1AL
97.2
67.5
32.58  80.7
15.95   611.5
                                            B-47

-------
                       TABLE 18

             MINE WATER FLOW DISTRIBUTIONS
      Parameter Values to Calculate Average Flow (Flow in
      MMG/day) as a Function of Mine Size (Size in MTons/year)
                                  FLOW = A .  SIZE8

            REGIONS;                 A          B

            NA, SA, CA           0.00078    0.4754

            MW, CW, TX           0.01951    0.1651

            NGP, RM,  NEV,
             NMEX                0.06364    0.0
         Frequency Distributions of Multipliers to Allow
         for the Variation Around Average Flow Frequency
                                                  NGP.RM,
Multiplier        NA.CA.SA          MW.CW.TX      NEV.  NMEX

  45.2548          0.0              0.018           0.0
  22.6274          Q.027            0.070           0.0
  11.3137          0.072            0.053           0.0
   5.6569          0.162            0.140           0.222
   2.8284          0.072            0.140           0.111
   1.4142          0.144            0.158           0.223
   0.7071          0.135            0.088           0.111
   0.3536          0.126            0.088           0.222
   0.1768          0.135            0.070           0.111
   0.0884          Q.054            0.070           0.0
   0.0442          0.018            0.070           0.0
   0.0221          0.018            0.035           0.0
   0.0110          o.009            0.0             0.0
   0.0055          o.028            0.0             0.0
                          B-48

-------
                                                            TABLE 19
        NUMBER    PRODUCTION
                    COST
 DbTAILEU  SUPPLY  FUNCTION FOR PENNSYLVANIA,  1984
       Supply  Function for Mine Type:   1
                Pennsylvania. 1984

PRICE    NO. MINES    EMPLOYMENT    WAGES    FLOW    CASH FLOW    MINE EOPT    CONTROL EQPT
           442
           443
w
           445
           446
           44 T
           44B
           449
           450
45?
453
454
455
4S6
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
4T4
475
476
477
478
479
480
481

(1)
44. ?n
44.30
44.40
44.50
44.60
44.70
44.80
44.90
45. no
45.10
45.20
45.30
45.40
45.50
45. f>0
45.70
45. "0
45.90
46.00
46.10
46.20
46.30
4ft. 40
46.50
46.60
46.70
46. HO
46.90
47.00
47.10
47.20
47.30
47.4fl
47.50
47. hO
47.70
47. BO
47.90
46.00
48.10
976.4$
979.42
98?. 30
985.35
988.32
991. £9
994.77
997. <>5
1000.24
1001.23
100*. ?2
1004.22
1012. ?4
1015. 76
lOlfl.?*
1021.10
1024.32
1027.34
1030.36
1033. JB
103A.40
1031.4?
104?*44
1045.46
1040.48
1051.30
1054.'i2
10«7.S5
\Of-O.Sfi
1063.60
1066.63
10W. '.5
107?. 68
1075.71
I07S.74
1081.77
1084.30
1087.64
1090.88
1093.91
29.66
29.06
29. '.6
29. ^6
29.72
<>9.72
29. MS
29.^5
29. *9
29.i?9
21.79
29. *9
so. n
30.21
30.21
30. ^1
30.^1
30.21
30.21
30.21
30. ?1
30.21
30.21
30.21
30.21
30. •'I
30.r?6
30. -:6
30. £*>
30.26
30. ?6
30.29
30. £9
30.i?9
30.29
30.29
30.37
30.37
30.37
30. J7
180
180
180
180
181
181
18?
18?
183
183
184
184
IBS
185
185
185
185
185
185
185
185
185
18'j
185
185
185
186
186
186
186
186
187
187
187
187
188
189
189
189
189
25425.48
25425.48
25425.48
7S4P5.48
25719.48
2VM9.48
25877.48
2587 f. 48
2590B.-J3
25908.93
26117.93
261 17. 93
26149.51
26149.51
26149.51
26)49.51
26149. SI
26149.51
26149.5]
26149.51
2614Q.51
26149.51
2614«J.51
26149.51
26149.51
?614<).S|
272SB.SI
2725/1.51
27258. SI
27258.51
27258.51
27280.29
2f».nn.?9
27280.^9
27280.29
27339. ?9
27411.07
27411.07
27411.07
27411.07
301.42
301.42
301.42
301.'i(!
303. Cl
303.1} 1
305.72
30S.72
307.22
307. ?Z
308. S4
308.^4
321.13
321.13
321.13
321. U
321.13
321.13
321.13
321.13
3?1.13
321.13
321.13
321.13
321.13
321.13
3?4.H5
3?4.'J5
3?4.A5
3P4.85
324. i'j5
3?7.f,0
32r.>-0
327.^0
327. f>0
329. 06
335.15
335.15
335.15
335.15
40.54
40.54
40.54
40.54
40.75
40.75
40.«(S
40.86
40. H9
40. «9
41.01
41.01
41.19
41.19
41.19
41.19
41.19
41.19
41.19
41.19
41.19
41.19
41.19
41.19
41.19
41. 19
41.'
24-J.06
243.06
241. Ofi
?45.67
245.67
245.67
2'«5.67
246.25
251.97
251.97
251.97
251.97
1209.81
1209.P1
1209.81
1209.81
1221. 51
1221.51
1226.89
1226.89
1231,28
1231.28
1233.89
1233.89
1269.57
1269.57
1269.57
1269.57
1269.57
1269.57
1269.57
1269.57
1269.57
1269.57
1269.57
1269.57
1269.57
1269.57
12^5.43
1295.41
1295.43
1295.43
1295.43
1314. <>0
1314.60
1314. hO
1314.60
1317.40
1359.33
1359.33
1359.33
1359.33
44.33
44.33
44.33
44.33
44.33
44.33
44.62
44.62
44. »6
44. A6
44.86
44.86
45.08
45.08
45.08
45.08
45.08
45.08
45.08
45.08
45.08
45.08
45.06
45.08
45.08
45.01
45. 0"
45. OH
45.08
45.08
45.08
45.55
45.55
45.55
45.55
45.55
45.55
45.55
45.55
45.55
              Number of increments of  0.1  million tons of production  capacity

-------
W
I
     NUMBER    PRODUCTION
482
483
484
485
486
4fl7
488
489
490
491
492
493
494
495
496
497
498
499
SOO
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
                    COST
PRICE
NO.
48. ?0
48.30
48.40
48.50
48.60
48.70
48.80
48.90
49.00
49.10
49.20
49.30
49.40
49.50
49.60
49.70
49.80
49.90
50.00
50.10
50.20
50.30
50.40
50.50
bO.60
50.70
50.80
50.90
51.00
51.10
51.20
51.30
51.40
51.50
51.60
51.70
54.80
51.90
52.00
52.10
1006.95
1099.99
1103.02
1106.06
1109.10
1112.14
1115.18
1118.21
1121.75
1124.29
1127.33
1130. J7
1133.40
1136.44
1139.V8
1142.32
1145.56
1140.60
1151.65
1154.69
1157.74
1160.80
1161.85
116«-.90
1169.96
1171.01
117'-. 07
117Q.12
118?. 18
11«A.23
lin«.29
1191.34
1194.39
1197.45
1200.50
1201.56
1206.62
1209.69
1212.76
1215.83
30. 1/
30.37
30.17
30.37
30.18
30. JB
30. 38
30.. IB
30.18
30.18
30.- 18
30. IB
10. 18
30. JB
30. 38
30.- 18
30.36
30.46
30.<«6
30.46
30.55
30.55
30.55
30.55
30. S5
30. SS
30. 5b
3P.55
30.55
30.55
30.55
30.55
30.55
30.55
30.55
30. b5
30.67
30.67
30.69
30.69
TflRIF 1Q front 'H }
SUPPLY
MINES
18«>
189
189
189
190
190
190
190
190
190
190
190
190
190
190
190
190

192
192
193
193
193
193
193
193
193
193
193
193
193
193
193
193
193
193
194

195
195
FUNCTION FOR
Pennsylvania
EMPLOYMFNT
27411.07
27411.07
27411.07
27411.07
27442.98
27442.98
27442.98
27442.98
27442.98
27442.98
27442.98
27442.98
27442.98
274«2.9B
27442.98
27442.98
27442.98
27631.98
27811.98
27811.98
27831.76
27833.76
27833.76
27833.76
27833.76
27831.76
27833.76
27833.76
2783 ». 76
27833.76
2783.1.76
27833.76
27833.76
27833.76
27833.76
27833.76
28198.76
28190.76
28231.03
28231.03
MINE TYPE: 1
, 1984
WAGES FLOW CASH FLOW MINE EOPT
335.15
335.15
335.15
335.15
3*6.12
346.12
346.12
346.12
346.12
346.12
346.12
346.12
346.12
346.12
346.12
346.12
346.12
347.60
350.02
350.02
362.15
362.15
362.15
361.15
36?,15
36?. lb
362.15
36;-. 15
36r'.lb
36?. 15
36/7.15
362.15
365
43. ?5
43.8?
4.1.8?
43.82
43. H?
43.0?
43. H?
4.1.8?
43.82
43.8?
43.82
43. H?
43.82
43.8?
43.96
44.17
44.17
45.65
45.65
45.65
45.65
45.65
45.65
45.65
4?,. 65
4f,.6S
45.65
45.65
45.65
45.65
45.65
45.65
4S.65
45.66
45.66
45.70
45.70
251.17
251.97
251.97
251.97
260.7.9
260.29
260.29
260. ?9
260. ?9
260. ?9
260. ?9
260.29
260.29
260.29
260. ?9
260.29
260.29
260.87
261.81
261. «3
273.08
273.08.
273.08
271.08
273.08
271.0H
273.08
27.1. P8
P73.08
27.1.08
273.08
273.08
273.08
273.08
273.08
273.08
274.88
274.88
277.96
277.96
1359.33
1359.33
1359.33
1J59.33
1390.00
1390.00
1390.00
1390.00
1390.00
1390.00
1390.00
1390.00
1390.00
1390.00
1390.00
1390.00
1390.00
1392.79
1397.41
1397.41
1479.46
1479.46
1479.46
1479.46
1479.46
1479.46
1479.46
1479.46
1470.46
1479.46
1470.46
1479.46
1479.46
1479.46
1479.46
1479.46
149?. 52
1492.52
1503.71
1503.71
CONTROL EOPT
45.55
45.55
45.55
45.55
45. «5
45.85
45.85
45.85
45.85
45.85
45.85
45.85
45.85
45.85
45.85
45.85
45.85
45.85
45.85
45.85
46.23
46.23
46.23
46.23
46.23
46.23
46.23
46.23
46.23
46.23
46.23
46.23
46.23
46.23
46.23
46.23
46.42
46.42
46.62
46.62
        (1)
    Number of increments  of 0.1  million tons  production capacity

-------
TABLE 20
LINEARIZED SUPPLY. CURVE FOR PENNSYLVANIA


1
2
3
4
5
6
7
8
9

Supply
(1012Btu's)
2.28
15.96
31.92
79.80
134.52
335.16
679.41
971.22
1424.90
(BAT-4)
Cumulative
Costs
($MN)
0.98
7.82
17.08
50.11
94.98
289.61
688.94
109 .63
1289.13

Unit Price
($/MMBtu)
.43
.50
.58
.69
.82
.97
1.16
1.38
1.61

Number
Of Hines
1
3
7
13
23
59
131
170
243
      B-51

-------
approximation  is  obtained  by ordinary  least-squares regression  of a
series of  linear  functions between cumulative  cost  (i.e., production
times  price)  and  cumulative potential production;  the  length  of the
linear  segments   aTid  tfienumber  67segments  follow  from a  minimum
accuracy requirement.

For the impact analysis, minimum  required  price  values  specifying the
beginning and the end of linear segments are not allowed to deviate by
moffe  than  10% from the estimated  highest  or  lowest  values  of the
previous segment.

Ml. COAL MARKET SIMULATION (MODULE 4)

Module 4 organizes  the results of the 1984 projections (i.e., regional
supply, demand, transportation and utilization costs) in a linear pro-
gram, which balances supply and demand in coal  markets.

A.  Sulfur Content  Distributions, Air Quality Control  Standards and
    Coal Utilization Costs

Sulfur  content  distribution for  coals  from the different  supply  re-
gions  is  derived through  analysis  of  information on 1976  coal  pur-
chases by large utilities, obtained from the Federal Power Commission.
The results of this analysis are  shown in Table 21.

The  cost  of  coal  use  for  a specific plant  is  a  function  of the  air
emission  control  standard with  which  the  plant must  comply.    Two
different sets of standards  are  required  for the analysis.  The  first
set  (the  so-called  SIP),  issued  by  the individual states, establishes
permissible emission limits for plants constructed prior to 1975.   The
second  set  (the  so-called NSPS), which  applies  to  all  plants  cons-
tructed between 1976 and 1984, is issued by  the federal government.

The  estimated ratios  of  demand  by  old  plants subject  to  different
sulfur  emission  limits set by SIP  standards,  (in  terms  of pounds  of
sulfur  found  in stack  gas  per million Btu of coal  burned)  are shown in
Table 22.  These  ratios  are  essentially  the samp as those used in  the
National Coal  Model.   The  sulfur emission limit set by the NSPS  stan-
dards  is 0.6  pounds per million Btu of coal  burned.

The  derivation of utilization costs for different  coals  is  extensively
discussed is  Chapter VI  of Appendix A of this  report.  Table 23  shows
the  utilization  costs  which are  used  in  the analysis for new  and  old
plants  subject   to  different  sulfur  emission  limit   standards  and
burning  coals with different  sulfur  contents.   The utilization cost
differentials shown in Table  23  are  required  because  of  significant
differences   in other  coal quality  characteristics, Such  as Btu  per
pound  and ash and moisture content.
                                   B-52

-------
                                  TABLE 21


       RATIOS OF COAL AVAILABLE  IN DIFFERENT SULFUR  CONTENT RANGES
ESCALAIUR
               1.062
                                                                       L8/HM8Tu
COAL
           F COAL
        LB/MMflHj
    3 MU
         0.03 0.6b 	
              0.14 O.S4
         0000.0.10
 0.
 .0
 .12
 <5 *A
         0000.0.05
         OOOO.OOOU
         oooo.oooo
         OOOO.OCOU
         0000.0000
         onoo.oooo
         OOOO.OOOU
         onoo.oooo
         0000.0000
         cooo.oooo
         ).01 0.02
         >.01 0.02
                 §0
                 0
               .00
               .00
               ,00
               .00
                oo
 0.60
 0.2P
.0000.
.0.00
.0000.
.0000.
,0100.
,0000.
,0000.
 oooo.
,0.4S
 0.39
 0.39
 0.00
 0.00
 0.00
 0.00
 o.oc
 o.oo
  S.oo
  .00
                         O.?o 0.03 OOUO.OOOO.
                         O.i?7 n.05 00.03 0.02
0.16 0.0A 0.04 _
0.23 0.16 0.14 0
      ,00
                         0.40 0
     ..Ob
0.00 0.00
                         0.00 n.QO 0.00 0.00
                         0000.0000.0.00 0.00
                         oooo.oooo.oooo.oooo
                         0000.0000.0000.0000,
                         OOOO.noOO.OOOO.0.00
                         0000.0000.0000.0.00
                         OOOO 0000
-------
                         TABLE 22

         RATIOS OF COAL  DEMAND SUBJECT TO DIFFERENT
               AIR QUALITY CONTROL STANDARDS
 ••::.,? L^r^'oT ;%sKbR  u;TuTbL7.J *?."**? r.sNDAPU
 1  !.            ."7               *5J
 2 •••-   • -'  '"3            ,,                   .37
 3 '"'                      b6
 5 VJ    *16       .'J1-               *''b
 o V-        -SB      •«                        .28
 7 *V              !j           .69               '16
 9 Ot-             «33      >ei7                    |.
10  ;^
11  ^      »;B
13  L'S       ».
!«,  Ml      .19   .f*1                     ,,,
Ib  IL            «hl
1»,  IN       .96      .0<»

 IB  tl            1%P           l'°
 19  -r.        .l'c  .f"'               ^
 20  tT        ."fc                   '7,.
 21  »T        .*"   ^^                B;B          .11

 23 OM             .lf<   •*!  •*l
 /"• N.«                 lt                1.0

 .) / .\l)            I."
                          1.0
                .ov  .•»:»
        .b?  .-3
 31 (-0   1.
 3? AN   1.
 33 t>0            J«
 3*. CN   1.
 34 CS   1.
                             B-54

-------
                               TABLE  23

                        COAL UTILIZATION COSTS
     Cost of Burning  as a Function of SIP  Limit and Coal  Sulfur  Content
 C'JbT OF Bi-i'M1..
COAL S Ltl/fMliTU
  SIl  <0.3
  SI- <0.h
  SIP <1.S
  blr- M.'S

  b' ' ••<•.. 0
iJ C*!rjTI.1G PLANTS N3/^.-dTU)»COAL S LU/MMBTU  HOR.t  SIP  LIMIT
 0.0  0.6  0.
  .V)
  .-P
  .'-P
                   I.1-"
                         .58
       .59
       .SH
       .5*
                              .54
.25
.b3
. /4
.67
.b9
.S4
• ^2
.b2
.52
«b2
.^2
.SO
.h4
.74
• <0
.56
.b6
.56
.56
3.00 '
.90 i
.66
.79
.74
.70
.67
.64
.64
.64
.64
*.00 5.00
?.00 2.11
.96 2.G9
.92 2.05
.86 2.00
.82 1.95
.79 1.91
.tS 1.78
.7b 1.78
.75 1.78
.7S 1.78
cSCALATCP      1.062
  COST Of ritiKMMG F0'4 NFW PLANTS AND ULO PLANTS SUBJECT TO SIP'S < 0.3 LB/MMRTU
 COAL 1> LlVKMbTU     (1.3  n.6  0.9  1.3  l.b I.7S  2.0  3.0  4.0  5.0
        CO"         l.SR l.i". 1.67 1.74 1.74 1.7S 1.76 1.86 1.96 2.09
    Cost of.Burning  Differentials  Due to Differences in  Moisture,
    Ash  and Btu Content of Different Coals
I
3
<•
1
S
6
r
8
4
] 1
On
M!i
^'J
PA
vS
VA
CK
TN
AL
J\L
I'.OO
•).0(l
O.CO
0.01.
O.JO
0.00
0.00
r.oo
O.PO
.0 «2
                            12**
                            lilA
                            1—«0
          le-TX

          iobl)
          PbCS
          2buT
                                       03<>
                                      .a-.-
                                      .ic-
                                      .011
                                      .074
                                      .074
                                      .011
                                      .Jll
                                      .011
                                      .053
                                      .053
                                      .051
                                  B-55

-------
B.  Transportation Costs

The transportation costs  for  appropriate  origin-destination links  are
estimated from statistical analysis of rail rates and water transport-
ation costs developed in Chapter V of Appendix  A.

The basic  determinant of transportation  costs  for  all  links  is dis-
tance.  The distances  between  coal-producing  and  coal-using  regions
developed for the  National  Coal  Model  are  used.   These distances  are
between  centroids  of production and consumption  in the producing  and
demand  regions.  The actual  locations  of  mines and major use centers
are taken  into account.   The  resulting  distance is  a weighted average
among  all   points  of  production  within  a producing  region  and  all
points  of consumption within a demand region.   The  weighting factor is
the quantity of coal shipped.

The second  factor  in determining transportation costs is the number of
rail  line  changes required  over  a  specific  link.   The line changes
have  been  developed  as part  of the  rate-qathering  exercise.   For  an
individual  point-to-point movement, the number  of line changes must be
an  integer.   The  transportation  cost  estimated  for the  model  is  an
average  of  all  the  movements  expected  to  take place.   The number of
line changes required for the average movement  between a producing  and
consuming region may be a non-integer value because a different  number
of line changes may  be required to move between different  parts  of  the
origin  and  destination  regions.  The line change values used for  the
model  are  rough  estimates and  are  applied to  those links where line
change  is a factor in the transportation cost equation.

The  statistical   analysis of  rail   rates   shows  that  there  are  two
separate  cost  estimation  equations  depending on the  area  of  the
country  in  which  the coal  moved.   Western  rates are statistically
different from Eastern  rates; therefore,  the estimation equation used
must  apply  to specific origin-destination  pairs.

The data base for  the statistical  analysis  of rates excluded rates  for
coal moving from the Western  plains to the  East Coast because  there is
little  current  movement  on  these  links.    However, it  is  felt that
higher  costs of  Eastern  rail  operations  should  be  taken into  account.
Therefore,  the cost  for  a combined  Eastern and Western  railroad  move-
ment  is estimated  by  an  equation  whose intercept  and  distance  coef-
ficients  are  the average of  the  Eastern  and Western cost equations.
These  cost  equations  and the  range of  origin-destination  pairs  to
which they  are applied are shown in Table  24.

The data and  equations   described  above   are  used  to  calculate  rail
transportation  costs  for all  origin-destination  links for  which  a
distance  is specified.   The  program sets  the  transportation  cost to
zero  for any link where  the distance is  specified as  zero.   A  zero
transportation cost  prevents any  coal  from  being  shipped  over  that
                                  B-56

-------
                               TABLE  24

       PARAMETER VALUES FOR RAIL TRANSPORTATION COST EQUATIONS
      AND APPLICABLE ORIGINS AND DESTINATIONS FOR CONTRACT MARKET
        Equation   Coefficients
Intercept      Distance        Line Change         Origins      Destinations



  3.288          13.568             -               1-12         1-35

  2.379          11.262           1.072             21-26         1-22

  1.529           8.956           1.072           M3-20,27-30      1  -  35
II3-20,27-
l   21-26
                                                               23  -  35
                                 B-57

-------
link.  The  transportation costs for coal  movements  within  production
regions are specified separately.  These transport, costs are  generally
smal 1.

There are specific  links  in  thp system which  arp  not  servpd by  rail-
roads.  For these links externally determined  costs are rpad  in by  thp
computer  program.   Table  25 shows  the origin-destination  links  for
which the rail rate estimated costs are replaced.

The  final transportation  costs  used in the  linear program  simulation
of  the  contract  coal  market   are  shown   in  Table  26.    The origins
(supply regions)  are  columns,  the  destinations are rows  (thres  rows
per demand region), and the  values are  in  dollars  per ton.

The  transportation  costs  used  for  the simulation  of the  spot  coal
market are higher than  the costs  used  for the  contract market because
of  the  smaller  volumes shipped.   These  higher  transport  costs  are
estimated by  using greater  distance coefficients  for the  estimation
equations as  shown  in Table  27.

Certain  regions, which  have   significant  contract coal   poroduction
and/or demand,  produce and/or  consume negligible  quantities of  spot
market coal.  These regions  are therefore  omitted  from the  spot  markPt
simulation.   The computer program which  generates the transport  file
estimates only those  transportation costs  required  for the  spot  markPt.
origins and destinations.   The  specific spot  and contract market  sup-
ply  and demand regions  used  in  the anlysis are shown in Table 28.   Thp
final transportation  costs for  spot  market coal are shown  in Appendix
IX.

C.  Demand for Coal
The  EPA Office  of  Air  Quality Standards  recently  published a  pro-
jection of  the  demand for coal  as part of an  anlysis of the impact of
air  quality standards.   The  demand  for  coal  within  the  35  demand
regions used in this  study  is taken  directly  from  this EPA air quality
study.  The projected  demands are  shown  in Table 29.

The  use of these  demands   implies  that  the water  effluent  control
standards do not  significantly  alter the total demand  for  coal  in any
of the  35  demand  regions.  The demand projections would be altered if
increased  costs  for water  treatment  are so  large  as to  render  other
fuels cost  competitive with coal  in certain demand regions.   As shown
by the  impact estimates,  this is not  the case.

VIL. LINEAR PROGRAMMING MODEL OF THE  STEAM COAL  MARKET

A fundamental  assumption  in the ADL  coal model  is that  the  market is
cleared on  the basis  of minimum total  cost.   Two distinct formulations
are  used depending on whether or  not  a  producer's surplus  of coal is
                                    B-58

-------
             TABLE 25

        TRANSPORTATION COSTS
  NOT BASED  ON RAIL RATL EQUATIONS
            Destination         Cost

 2              3            $ 3.50
 4              3              3.10
 5              3              5.10
 5              7              3.10
 5              13              3.70
 5              15              7.35
 2              19              7.70
 4              19              7.75
 5              19              5.40
 7              21             10.95
 7              23             14.75
 4              26             10.85
 5              26              6.93
 5              27              8.00
 7              27             12.50
12              23              7.53
11              17             17.48
21              13             19.43
22              13             20.43
23              13             18.43
22              14             16.13
21              20             19.15
22              20             20.15
23              20             18.15
27              32              2.55
28              32              2.34
                 B-59

-------
                                                          TABLE 26


                                  TRANSPORTATION COSTS FOR LINKS  BETWEEN  27 SUPPLY  (HORIZONTAL)

                                       AND 35 DEMAND (VERTICAL)  REGIONS  (CONTRACT MARKET)
10. Hb912.9.1?ll. 34512.-
 o.n ._ o.o   n.o   o.o
                   _ _ _

                 1  0.0
                 910.7
 9.<>0<> 0.0   9.57910.7(^12.3
 C.O   0.0   0.0   0.0   0.0
24.767U/  4. |9o h
	  /.Slrt 0
0.0

6:5(i>

  '.»*0
  9
  0
U.O
0.0
*}•!)•*' *• • •**f *
9.92610.694
                         0.0
                         0.0
                              0.0
                              o.n
                                    u.o
                                    Q.Q
                                           u.o
                     \*   tf«U   UV1'   HVV
                     059 6.039 9.054 8.926
                                .24011.734
                                      .0
  ..
 t.140 3l«44
 b./22 H.7U8
                  b.57? S.JfaO 7.?H/  /.3bl
                  9.081 9.H4«10.?9310.77^
                                 tf "» f
                          .?93
                   U.O  19.4/0??. 133 0.0
                   4.31B 3.491 7.^4J H
                  10. 32M1.09J11. 34912
4.31B 3.491 7.S4J H.017
»).32M1.09Jil.349l:
 (>
 U.
 <>.
IS.
 u.
 0.
u.
 0.
 u.
16.
 0.
 U.
Ib.
6,{911.<,H1 9.629 0,
/03 9.^87 0.0  10,
lOJJ 6.097 0.0   b,
/4bl4.2b311./45 0,
n   0.0   0.0   0,
0   0.0   0.0   /,
^t/21b.p4713.262 U,
/Ib U.O  19.80618,
n   o.n   o.o   /,
J/«Mb.95713.372 0,
0   0.0   0.0   0,
0   U.O   0.0   0,
b04 3.349 6.08010,
0   U.O   0.0   0,
0   0.0   0.0   U.
S69 9.1HO 9.99'. 7
  -  o.n
    o.o
                                                                           .     0.0
                                                                      040 H.Q61  6.046
                                                                      "    '  ~    "  '
                                                                          U.O
                                                                          U.O
                                                                               0.0
                                                                               0.0
                                                                                     0
                                                                          .      .
                                                                     41/17. ;,/9]>>.*3f  o
                                                                     o    u.o    o.o    o
                                                                     0    O.n    0.0   0.0
            n   u.o
            330 U.b4?
            229 0.0
            0   U.O
            1?9 6.417
            397 O.U
            0   0.0
            91515.070
            764 0.0
            0   0.0
            7131J.V
            H05 0.0
            0   0.0   I
            137U.48J I

8M2liU219 Olo   Olo   I
51810.001 9.361 9.617 I
960 6.7/9 /.Jl'J U.O   I
76815.105 0.0   0.0   I
54711.85711./92IO.244 I
394 4.125 b.900 0.0   I
0   0.0   0.0   0.0   I
?4616.10716.68314.4941;
760 6.48U 5.857 3.954 I
                0.0   I
                0.0   I
                0.0
                        .735 <5
                        .0   0
                        ..in no
                        . 8;.n o
                        .0   n
                        .?S9 fl
                        .O'j5 7
                        .0   0
                        .1401?
                        .01611
                        .0   0
                        .27614
                        .732 7
                        .910 0
                        91913
            . 00') \ 1 .
            .0    0.
            . IVilO.
            .0    /.
            .0    0.
            .njoio.
            .834  V.
            .0    0.
            .45411.
            .018  9.
            .0    0.
            .3b1U.
            .1U7  b.
            .0    0.
            .06713.
                                                                           0   0.0   U.O
                                                                           0   0.0   0.0
                                                                           o   n.o   o.o
                                                                           76310. 745 0.0
                                                                           0   0.0   0.0
                                                                                 -----  --
                                                                                                              0.0
                                                                                                              0.0
                                                                                                              - -
                                                                                              0  16.41514.216 0.0
                                                                                              PHI 4.62913.04/13.679
                                                                                              ??7?|.40620.76621.0?2
                                                                                              0  11.426 9.652 0.0
                                                                                              954 7.0/810.207 9.115
                                                                                              n   o.o   o.o   u.o
                                                                                              o   o.o   o.o   o.o
                                                                                              03712.698 2.406 2.208

                                                                                              ?«?j:*b:iz?:i56 8:8
                                                                                              0/211.245 0.0   0.0
                                                                                              o   n.o   o.o   o.o

                                                                                              SSSfWIH:!118:
                                                                                              o   n.o   o.o
                                                                                              ?7119.24?17.400
                                                                                              91811.666. H.It
                                                                                                                      _.
                                                                                                                    o.o
                                                                                                                      523

-------
                               TABLE 27
                   RAIL TRANSPORTATION COST EQUATIONS
       AND APPLICABLE ORIGINS AND DESTINATIONS  FOR THE  SPOT MARKET
        Equation  Coefficients
Intercept
Distance
Line Change
Origins
Destinations
  3.288
  2.379
  1.529
 21.030
 17.456
 13.882
  1.072
  1.072
 1 - 12
21 - 26
13 - 20
21 - 26
   1 - 35
   1 - 22
   1 - 35
  23 - 35
                                 B-61

-------
                    TABLE  28

   BRIDGE FROM CONTRACT TO SPOT  MARKET  REGIONS
SUPPLY REGIONS
DEMAND REGIONS
Contract
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
Spo
1
2
3
4
5
6
7
8
9
10
11
12
_
_
_
-
13
-
-
-
14
-
-
-
15
_
_
-
-
_
                                 Contract

                                     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
                                    32
                                    33
                                    34
                                    35
           Spot

             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
            32
                      B-62

-------
                             TABLE 29

                       COAL DEMAND IN 1984

                       (Billions of Btu's)
Region
Number    Region Name                 Contract  Market    Spot Market

   1       Vermont                          31.4             0.0
   2       Massachusetts,  Connecticut       206.0             0.0
   3       Pennsylvania West               514.3           238.1
   4       Pennsylvania East,  N.Y.,  N.J.    420.7           177.4
   5       New York Upstate                190.3           106.0
   6       Virginia, Maryland, Delaware    330.0           155.0
   7       Virginia West                   675.4           150.8
   8       Carolina, North & South         720.3.          154.5
   9       Georgia, Florida North          972.0           125.6
  10       Florida South                    18.0             0.0
  11       Ohio North                      224.0           125.8
  12       Ohio Central                    173.7            81.1
  13      Ohio South                      568.1           231.9
  14      Michigan                        545.1           150.3
  15       Illinois                        717.8           164.2
  16      Indiana                         919.0           251.7
  17       Wisconsin                       381.1            41.3
  18      Kentucky East                   123.2            22.4
  19      Kentucky West                   567.7            98.7
  20      Tennessee East                  181.3            41.4
  21       Tennessee West                  273.5            53.2
  22      Alabama, Mississippi            699.2           267.3
  23      Dakota, North & South           412.9            32.9
  24      Kansas, Nebraska                271.2           113.0
  25      Iowa                            265.0            29.7
  26      Missouri                        J87.7            63.5
  27      Arkansas, Oklahoma              548.6            50.1
  28      Texas                         1,260.2           125.0
  29      Montana, Wyoming,  Idaho         445.5            19.0
  30      Utah, Nevada                    303.7            27.9
  31      Colorado                        266.7            21.6
  32      Arizona                         391.7            20.4
  33      Washington, Oregon              123.6            22.8
  34      California  North                 16.5              0.0
  35      California  South                 27.0              0.0

                                       14,272.4          3,162.6
                                B-63

-------
assumed.   In  either  case,  the market clearing  mechanism is described
as minimization  of a  linear  function subject  to a  series  of linpar
inequality  constraints.    Such  a  formulation   is   called   a  "linear
program" and  well-developed mathematical  techniques are available for
its solution.   This  section describes the  details   of  thp  two market
clearing algorithms and their formulation as a linear program.

A.  Market Clearing Algorithms

Figure 10 illustrates the distinction between the two assumed clearing
mechanisms.   The  solid  line  represents the  estimated  supply  curve
based on minimum required  price  as discussed in  Sections  TIT  and TV.
Essentially, two interpretations of the curve are possible.

First,  if  most  of  the coal  is  sold via long-term contracts,  it is
reasonable  to expect  that each  mine will  bid  near to  its  minimum
required price  in  order to be  as competitive  as possible.   Tn  this
case, the  cost  of coal  supplied  is represented  by the cross-hatched
area in Figure 10.

On the other hand, if the market is dominated by  short-term  purchases,
mines are  likely  to charge whatever price  the market  will  bear.  In
this  case,  the  total  cost is represented  by  the  cross-hatched  area
plus  the  shaded  area  since  all  coal is  sold  at   the  highest   price
realized.   Producers  in  this case  will  reap an  economic  rent,   i.e.,
the producers' surplus S represented by the shaded area  in Figure  in.

The ADL coal  model  has been developed to accomortate both formulations
in order to simulate the behavior of both  the  spot and contract  mar-
kets.

B.  Linear Programming Formulation of Contract Market (No Producer's
    Surplus)

In order to develop  the  model,  it is necessary to  express  the market
clearing algorithm as a minimization of a  linear  objective  function
subject  to  a series  of  linear  constraints.    The  constraints,   which
represent the supply-demand balance  required of the  steam coal market,
are of the following four types:

    t    Constraints representing the total  available supply
         of coal from the different  supply regions;

    •    Constraints insuring the  balance  between coal   sup-
         plied and coal which is shipped;

    •    Constraints describing the  coal  demand in different
         demand regions subject to applicable SIP standards;
         and
                                   B-64

-------
                      FIGURE 10

TYPICAL SUPPLY CURVE ILLUSTRATING ALTERNATE MARKET
                 CLEARING MECHANISMS
    MRP
      MRP'
                                       MRP = F IQ)
              Q mln
                         B-65

-------
    0    Constraints describing  the  sulfur  balance  between
         coal  available from the supply regions and the coal
         which is burned in the different demand regions.

The  objective  function assembles  the  total burned  cost of  coal  and
consists of three components.

    t    The cost  of coal   supplied  from a  specific  supply
         region;

    •    The cost  of transporting coal  from  its  source to
         its point of consumption; and

    •    The cost  of burning coal based on  its  sulfur con-
         tent,  type  of consumer  and  applicable  pollution
         control standards.

1.  Specification of Constraints for the IP Model

The  methodology  used to estimate  the  cost  and  constraint, parameters
has  been  described  in the  previous  sections,  so this discussion will
be confined to mathematical  formulation of the problem.
                    nomenclature used  in  the following sections which
                                      the LP model.   It is assumed  in
                                              with  the methodology  of
Table 30  shows  the
describe in detail each component  of
the  discussion  that  the  reader  is familiar
the discussion  that the  reader  is familiar  with the  methodology
supply curve development, specification of  costs  and  specification
demand as discussed in Sections TV and V.

a.  Constraints Representing Total Available Coal  Supply
                                                                     of
These  constraints  are  included  to ensure that the total  coal  supplied
from  any region (sum of coal supplied  at each  price  level) does  not
exceed  the  total  available resources.   Mathematically,  this  is
expressed for each  region  as  follows:
                                             (VI.1)
Additionally,  since only a  fixed  amount of  coal  is available  at  any
given  price level,  the following constraints  are  included for  each
region  and  each  price  level:
                                             (VI.2)
                                   B-66

-------
     Symbol
                      TABLE 30
 NOMENCLATURE USED IN THE LP MODEL OF THE STEAM COAL MARKET


	                                   Definition
      'i L
                           Cost of transporting coal from supply
                           region i to demand region j (see section
                           V.2).

                           Cost of supplying coal from region i at
                           the Ji'th price level (see section IV.3).
      C8jkn
Supply Parameters
      SL
        ik
                           Cost of burning coal with sulfur content
                           k in demand region j subject to SIP
                           standard n (see section V.I).
                           Total amount of coal available in supply
                           regioi i.

                           Amount of coal available in region i at
                           price level a.

                           Amount of coal supplied by region i at
                           price level SL.

                           Amount of coal shipped from supply region
                           i to demand reg-on j.

                           Fraction of coal mined in region i at
                           sulfur level k.
                                     B-67

-------
                                  TABLE 30 (cont'ri.)

             NOMENCLATURE USED IN  THE  LP MODEL OF THE STEAM COAL MARKET  (cont.)
Demand Parameters
              Definition
       jkn
Amount of coal subject to air quality
control standard n which is required by
region j.

Amount of coal of sulfur level k which
is burned in region j subject to standard
n.
   Indices
      J

      k
                                       Supply  region  number
Demand region number
Sulfur level number
                                       Air quality control  standard index number
                                      B-68

-------
b.  Constraints Ensuring Balance Between Coal Supplied and Coal  Which
    is Shipped

In order to accurately represent coal  pricing, it is necessary to have
two descriptions for each unit  of  coal.   First,  the coal must be des-
cribed according to its region of origin and its price 'Y.£>.  Second-
ly, in order to model  the cost of transportation, it mus^be described
by its origin and  destination  (X..).   This dual  representation neces-
sitates the  introduction  of  constraints to  ensure  that  the  amount of
coal  supplied from each region is the  same in each representation.  To
accomplish this,  the  following constraints  are  included in  the model
for each supply region:
This  simply  states  that the  total   coal  leaving  region  i   'to  all
destinations  j)  is  equal to the  total  coal  supplied  in  region i  (at
all price levels a ).

c.  Constraints Describing Coal  Demand as a Function of Demand  Region
    and Applicable Air Quality Control Standards

The model is  designed to  accomodate up  to  four different cost burning
conditions per  region.   These  costs  of burning  conditions  allow  for
consideration of different air quality control  standards  in different
demand  regions  in  combination  with   the  different burning  costs  for
different vintages of coal burning plants.'  This requires additional
constraints to  ensure  satisfaction of  demand  for each  set  of condi-
tions.  The following constraints express this  requirement:
                      k

                    k'U.ikn  '   Djn          

This  states  that  the total  coal  supplied at  all sulfur  levels k to
region j,  under cost  of  burning  conditions  n,  must  equal  the total
demand for the region j  under the cost of burning conditions n.

d.  Constraints Describing the Sulfur  Balance

It  is  necessary  to  ensure that  the  volumes  of coal  with  different
sulfur content that are  burned  do not  exceed  the total volumes
Consisting of  the  State's  "SIP"  standards and  the "NSPS"  and  "New
NSPS" standards, promulgated by the EPA.
2
 Consisting of  estimated  handling,  burning  and  clean-up costs  for
plants  subject to,  respectively,  "SIP",  "NSPS"  and  "New  NSPS"
standards.
                                   B-69

-------
of  coal  with  different  sulfur  contents  that  are  availablp.   This
condition is imposed by the following constraints:
                     SLik   -  r Ujkn
This  set  of constraints  specifies that  the volumes  of coal  U with
sulfur  level  k burned  under all  the  different cost  conditions n  in
demand  region  j has  to  be equal  to the  volumes of  coal  with  sulfur
content k shipped to  demand region j from all supply  regions  i.

2. Specification of Objective Function

The  objective  function  used  in the coal  model represents  the  total
burned  cost of  coal.   This cost contains three components:   a  trans-
portation cost, which is proportional to the coal flow between  any  two
regions; the cost of  supply, which is a function of the  volume  of coal
supplied  by  each   region;  and  finally the cost  of  burning,  which
depends  on  the demand  region,  air  quality control  standard,  sulfur
content  and  plant  type.    This   cost  function   has   the   following
mathematical formulation:
     «•   •   1.J      U       MUU     j.n CBjknUjkn

The  total cost C,  represented by this objective  function,  is minimized
subject  to  the  constraints  given  by expressions  VII.1  to VII.5.   *\s
previously mentioned,  this  is a standard mathematical  problem that can
be  solved on  the computer.   For  the  impact  analysis  the IBM  MPSX
mathematical programming  system is  used.

C.   Mixed Integer Programming Formulation  of Spot Market (Producer's
     Surplus)

In order to  clear the market with  a producer's surplus,  'see  Section
VILA),  it  is necessary  to modify  the  steam coal market formulation
described  in the  previous  sections to ensure  that all  coal  supplied
from a given  region  is  sold at the  highest price paid  for  coal  from
that region.   This requires that the model  is  formulated  as  a "mixed
integer"  program  in  which some  variables  can  assume   only  integer
values.

The  above is accomplished by modifying Yij,  to  be  the total  volume of
coal  supplied up  to  the price level  «.,  instead  of the  total  volume
                                  B-70

-------
supplied ^t level j, .  This ensures that the price coefficient  Hi 2.  will
apply to all volumes of  coal  up to and including the specified  level.
Given this formulation, it is necessary to ensure that only  one  Y^fc  is
non-zero for  each region i.   This is  accomplished by introducing  an
integer variable  6^ corresponding with  each Y^.   Equation VI.?  is
then replaced by the following:
where:  S..*   =   the total  amount of coal  available  in region  i  at  or
        below price level  .

Since Y^  >_  0,  this  ensures that only  one  Y^ ^  can  be greater  than
zero when the following constraints are added  to the model:

                                                (VI. 8)
                     .0
                     1 J6
                         integer
Thus,  constraint  VI 1. 8
variables  6^   is equal
zero.   This  constraint
desired conditions: only
                          ensures  that  exactly  one  of  the  integer
                          to one and  that  all the others  are equal  to
                         combined  with  constraint  VII. 7  ensures  the
                         one Y^ represents  the  total  coal flow out of
the supply region.  The  remainder  of  the model is the  same  as for the
contract market clearing algorithm.

VIII.  OUTPUT OF THE COAL MARKET SIMULATION  MODEL  (MODULE  61

Module 6  organizes  the results of  the coal market simulation into two
reports, which are  shown in Tables 31 and 32.  The  first  report shows
the amount  of coal  supplied  by the  different supply  regions and the
marginal  cost  of  these regional supplies  (see Table 31 1.   The second
report  shows  the average  user cost  of  burning  coal  in the different
demand  regions.   These average user costs are specified separately by
plant  type  - each  plant type  being  subject to different sets of air
quality  control   standard: -   and  as  a  weighted  average   using  the
relative amounts of coal consumed  by  those plant  types as  weights.

As  shown in  Figure 11,  changes  in  the  regional   supplies, marginal
prices,  and  user  costs are derived by subtracting the results of Coal
Market  Model  runs made with and without the  cost increases resulting
from EPA  regulations.
                                   B-71

-------
                                       TABLE 31


Region
PA
OH
PL)
WVAN
WVAS
VA
EKY
TEN
ALA
ILL
IND
<*> WKY
*** lOitf A
f^
MlbS
KAW
ARK
UKLA
TEX
NDAK
SDAK
MYW0
WMON
WY-P
NCOL
SCOL
UTAH
ARIZ
NHEX
MASH
Total Coal
Supplied
(MTon)
47299.93
332GU.01

1^.99
9.3-3
9.38
1 7 t^'^
17.20
ifO«97
15.61
14.72
15.18

-------
                              TABLE 32
Region

VERMONT
MASS.CO
 FA te
PE.NY,N
NY UPST
W VA
N, S CA
GAtFL N
FLA STH
OHIO N
OHIO C
0«IO S
v.ICH
ILL
wise

KY-w
TENS  E
TFNN  W
ALA,MIS
DAKNtSt
KAN,NEB
IOWA
MISSOUR
ARK.OKL
TEXAS
MA.WYtl
UT* NEV
COLO
ARI.NME
MASH*OR
CAL  N
CAL  S
COAL
BURNED COST
Burned Cost (MMBtu)
Average
J.OU
-'.at1
2, jb
3,66
3.66
3.64
J.39
3.60
3.57
o.o
3.49
3.56
3,bl
J. 37
3s2
-------
                            FIGURE 11


        CHANGES IN THE REGIONAL SUPPLIES.  MARGINAL  PRICES
                        AND USER COSTS
       RUN;

Coal Market Model
without Regulations
       RUN:

Coal Market Model
with Regulations
Regional Supplies, Prices,
User Costs
Regional Supplies, Prices,
User Costs^
                  Changes in Regional  Supplies,
                  Prices, User Costs
Detailed regiona1_
supply curves
Regional economic
baseline
Regional changes in number of
producing mines, mine workers
employed, wages paid, invest-
ment required
                                                     I
Changes in regional earnings,
employment, economic growth;
energy prices to users; balance
of payments effect; effect on
growth of the coal industry
                             3-74

-------
The  changes  in  other  primary  impact  measures  follow  from  the
comparison of  the detailed  regional  supply curves  that are  used to
derive the linearized supply curves used in the Coal  Market Model  'see
Figure 11).

Specifically, with  the  decrease or  increase  in a regional  coal  sup-
ply -  obtained  by comparison of  the Coal  Market  Model  runs  with and
without  additional   pollution  control  costs  (see  Table  19)   -  the
changes in the number of producing mines, mine workers employed, wages
paid, and investment required to keep mines producing is derived.

IX.  LIMITS OF THE ANALYSIS

A.  Summary

The impact,  as  measured by  the  decrease  in  the  consumption  of  coal
from  an  impacted  supply  region,  will  have  been  underestimated or
overestimated  if  the "demand elasticity"  of  coal from  the  impacted
supply  region(s) was  respectively  under-  or overestimated relative to
the "demand elasticity" of coal  from the other regions.

The "demand elasticity" specifies the decrease in  demand for coal  from
a supply region in response to an  increase  in  the  cost, to the user of
an  incremental  unit  of coal  from  that supply  region.    This "demand
elasticity" is increased by?the  incremental  compliance cost estimated
to  result  from  regulations."   The increase  in  the total user cost of
coal  will  cause  a  large  decrease in  the  use  of that  coal  if the
"demand elasticity"  is  high and  it will  cause  a small decrease  if the
elasticity is relatively low.
lun    j n  4.- •*. .•    A(Demand) / A(Cost per Ton)
 "Demand Elasticity" = -  Demand /  Cost per Ton

2
 The lowest end of the coal  supply curve of an impacted supply region
is  made  up by mines with  low  production  costs  and  negligible mine
flows and, therefore, negligible compliance costs  (see Figure 12).  *\s
a result, the lowest part of the supply curve of a given supply region
will  not  change  when  compliance  costs  resulting  from  stricter
standards in mine water  treatment  are  added.   However, the higher end
of the supply curve is shifted upward when compliance costs are added,
resulting in a higher cost  for the same supply from the region.  This
higher cost  per incremental  unit  of  supply  will  cause  a relatively
larger decrease  in  the  demand  of  that coal  per unit  increase in the
cost of that coal: the "demand elasticity" of demand for that coal has
increased.


                                 B-75

-------
The  total  user  cost  of an  incremental  unit  of coal  from a  supply
region will  consist of the  sum of:

    •    Production costs;

    •    Compliance costs;

    •    Transportation costs; and

    •    Utilization  costs  (handling,  burning  and  clean-up
         costs).

An  under-  or  overestimation of  the  "demand  elasticity"  can  occur
because of a systematic under- or overestimation  in  any of these four
different costs.   This  systematic  error  in the  different types  of
costs can be  caused  by aggregation  errors  in  the  'non-sampled)  data
used in the analysis.

The  use  of  sampled data for  labor  productivities of new  mines,  mine
water flows and  mine  water  acidity  establish  a  range within which the
impact  estimate  cannot  be determined:  the impact estimate  is
statistically   insignificant  within  this  range.    The  "demand
elasticity"  is  indeterminate within that range  because  the underlying
sampled  data  for mine  water acidity,  mine  water flows and new mine
productivities are indeterminate within a corresponding range.

The  impact estimates  for BAT-4 are  generally  significant  in a  statis-
tical  sense:  the  estimated  impact  exceeded  the  range within  which
impact estimates are  indeterminate  because of the use of sampled data.

The  extent of  systematic errors  possibly existing in  the  data  cannot
be  estimated.    Sensitivity  tests demonstrate that  the supply  impact
estimate is  relatively insensitive  to systematic  errors   in the  user
cost  of  the coals  from the different  supply regions.   However,  the
impact estimate  for the impacted supply regions - regions where supply
decreases because  of relatively high  compliance costs  -  is  highly
sensitive to an  underestimate of the compliance costs (but relatively
insensitive to an overestimate of the compliance costs^.

Because  data  on  mine  water  flows  are only  available   for  highly
aggregated supply regions - the Appalachians, the Midwest plus  Central
West, and the  rest of the-U.S.  - the  water  treatment  cost estimates
are  the limiting factor in  the impact analysis.

The  use of average cost data for mine production costs, transportation
costs, and  utilization costs in the  impact analysis has  most  likely
resulted in  an  overestimate of the  decrease  in the use of coal  from
impacted regions (caused by increased compliance costsK
    'Including  replacement mines
                                  B-76

-------
B.  Statistical  Significance  of the  Impact  Estimates

The model  mine analysis shows that the minimum  price  required  for new
mine openings is highly sensitive  to  the  value  of average annual  labor
productivity.   Since  the labor productivities  for new  mines  is  not
known, sampled  values  obtained from mine  labor  productivity  distri-
butions for existing mines  are used.

The compliance cost is highly dependent  on  the  volume  of  acid water to
be treated.   To  allow for  the wide range  over  which  flow volumes for
different  mines  can  vary,  a  sample of  flow  values,  obtained  from
regional  mine flow distributions,  is assigned  to the different  mines
in each supply region.  The acidity of those mine flows  is determined
through sampling of a distribution  that indicates what  percentage of
mines can be expected  to have acid mine  water in that  region.

The use of  these  sampled  values for productivity and acid mine  water
flow  results  in a   better representation of the  "true"  supply curves
than  if,  for example, average  productivity  in  mine flows  were  used.
This  is illustrated by Figure 12.

Using  an  average  productivity to  estimate  the  minimum required  price
for new mines would result  in too  low an estimate of the  elasticity of
supply because  new mines with  higher and  lower  than average  produc-
tivities would  have been mis-specified,  resulting in  a  flatter supply
curve  than the  "true"  supply  curve.    Use of these flatter supply
curves for  the  impacted regions results in  a  systematic underestima-
tion  of  the  true   "demand elasticity"  and  this  'other  things  being
equal) would cause  a systematic overestimation  of the impact.

Given  that  sampled  values  are used  for  the  flow,  acidity and  produc-
tivity  data, it  is  impossible to  specify the  "true"   supply  curve
exactly: the  probabilistic or uncertain  nature  of these data result in
an  indeterminate range for the "true" supply curve.

As  illustrated  by  Figure 13,  the width of  the range,   caused by the
uncertainty  about  acidity  and flows, determines  whether the  shift in
the  supply curve  resulting  from  increased  compliance costs is
statistically  significant;  the  impact  estimate is  statistically
insignificant  if  the  BPT supply  curve  lies within  the   indeterminate
range of  the BAT  supply  curve.    Figure  13   shows  an   example  of  a
statistically significant shift of the BPT supply curve.

The  possible variation due to the use of  sampled values for  new  mine
labor productivity will  be  the same in both  the BPT and  BAT supply
curve and,  therefore, cancel  when the impact is calculated.

In  Figure  14,  it   is   shown  how an  estimate  of  the  maximum  possible
variation  resulting from the use  of sampled  values  for water quality
and water  flow  is  obtained.
                                    B-77

-------
                                     FIGURE  12

                   ILLUSTRATIVE  EXAMPLES  OF  SUPPLY CURVES BASED
                    ON AVERAGE VERSUS SAMPLED VALUES FOR MINE
                        PRODUCTIVITY  AND  MINE WATER FLOWS
         t
 Minimum

Required

  Price

  (S/T)
                                       /•
 /
/
    "Sampled"
    FiAf Sup,.iy i in ve
,   "Average'1
   BAT Supply Curve
 /"Sampled"
/  f;PI Supply Curve
                                                            "Averaqc"
                                                            BPT  Supply  Curve
                                                Supply  (MMT)
                                       B-78

-------
                             FIGURE 13

            ILLUSTRATIVE EXAMPLE OF INDETERMINATE RANGE
                  OF THE BAT SUPPLY CURVE BECAUSE	
                     OF UNCERTAINTY ABOUT MINE	
                     "WATER ACIDITY AND FLOWS '
Minimum
Required
Price
($/Ton)
"Sampled" BAT
Supply Curve
Range over which BAT
Supply Curve is
indeterminate due to
uncertainty about
mine water acidity
and flows.
                                                BPT
                                            »'   Supply Curve
                                           Supply  (Millions  of Tons)
                             B-79

-------
                        FIGURE  14

         METHODOLOGY  TO DERIVE  MAXIMUM POSSIBLE
               ERROR  IN IMPACTED REGIONS
     30 Samples
  of Pennsylvania
        BPT
    Supply Curve
Calculate Mean Supply
     and Maximum
 Deviation from Mean
 Due to Productivity
      Sampling
            30 Samples
         of Pennsylvania
               BAT4
           Supply Curve
       Calculate Mean Supply
            and Maximum
        Deviation from Mean
Due to Productivity, Water Quality
         and Flow Sampling
                  Derive Maximum Error
                  Due to Flow Sampling
                   for Pennsylvania
                Derive Maximum Variation
              Due Directly to Flow Sampling
               for Other Impacted Regions
                Total Possible Variation
                 Due to Interaction for
                  All Impacted Regions
                            1
                Total Possible Variation
                    in Impact Results
                  Due to Flow Sampling
                          B-30

-------
Thirty samples  of  the Pennsylvania BPT  and BAT-4   supply curves are
obtained.  These samples  are taken with  different  starting seeds for
the  random  number  generator  to  obtain  a different  sequence  of
productivity and flow samples for  each  supply  curve.  (Thirty samples
are considered  to  be adequate  to  obtain a  statistically  significant
sample of the underlying distribution.)

The mean supply  and the  maximum  deviation from  this  mean due  to
productivity sampling  is  calculated for  the BPT case.   In a similar
manner,  the mean supply  and  the maximum  deviation  from this mean due
to both  productivity  and flow sampling  is calculated from  the results
of the thirty samples in the BAT-4 case.

The supply  taken from  the  thirty different supply curves  represents
the volumes  that would have  resulted  in the  linear program solution
under the assumption  that no  changes  had occurred in all  other  supply
regions.  (This  last assumption implies  that  the  total  cost of coal
supplied  from  Pennsylvania  would  have remained  the same  in all the
thirty different solutions.)

Table 33  gives  an  example  of how  the  "sampled"  supply is  calculated.
The first supply curve  shown  in Table 33 is the one which  is obtained
in the original  solution used for  the  impact analysis.  The  amount of
coal  supplied  by  Pennsylvania  in  that   solution  is 1,093.89 million
Btu's  at a  price of  $1.61  per million  Btu.  The  volume which  would
have resulted - using the second supply curve shown  in Table 33  in the
solution  - is  obtained  by calculating  the incremental  supply from the
last  linear  segment  -  at  that  linear  segment's minimum required
price  -  that  equates  the  cumulative  costs of  coal  supplied  in thp
original  solution (i.e., $1,071.80 million).

Three different  types of variation result from the  sampling of produc-
tivity,  water  quality,  and  flow  distributions.   The  first  type of
variation is caused  by  productivity  sampling.   The two other types of
variation result from sampling  of flow distributions for  the impacted
supply region  considered.   They are the  "direct" and  the  "interaction
variation."

The  procedure,  previously  described,  allows  the  derivation  of an
estimate of  the  "direct  variation."   The other type of  variation, the
"interaction variation,"  results from variation  in estimated volumes
supplied in  any of  the  impacted regions because  of variation  in the
supply curves  of the  other impacted regions.
      estimated  impact of  BAT-2  compliance  costs was  found  to  be
 negligibly small  and therefore  statistically  insignificant.
                                   B-81

-------
                             TABLE  33

             EXAMPLE OF  CALCULATION  OF  SOLUTION SUPPLY
      FOR A SAMPTED SUPPLY  CURVE  FOR PENNSYLVANIA  (UNDER BPT)
  Pennsylvania  Supply Curve Used In Contract Coal Market Simulation
              MMBtu's

                2.28
               15.96
               31.92
               79.80
              134.52
              335.16
              679.41
              971.22,
             1093.89^
             1424.90
   MM$
   0.98
   7.82
  17.08
  50.11
  94.98
 289.61
 688.9f.
1091.63
1289.13
1822.06
$/MMBtu

  0.43
  0.50
  0.58
  0.69
  0.82
  0.97
    16
    38
    61
  1.61
    Sample of Pennsylvania Supply Curve Obtained With a Different
            Starting Seed for the Random Number Generator
2.28
15.96
31.92
79.80
102.60
321.48
636.09
994. 02, -^
1071. 80u;
1454.54
0.98
7.82
17.08
50.11
68.81
283.31
654.55
1159. 23, 9x
1289. 13u;
1928.30
0.43
0.50
0.58
0.69
0.82
0.98
1.18
1.41
1.67
1.67
   Supply from Pennsylvania in contract coal  market solution is 1093.89

^Supply calculated for this  case is  1071.80:

                1071.80 = (1289.13 - 1159.23)/!.67  + 994.02
                               B-82

-------
The variation caused by sampling of the productivity distributions can
be ignored because of its systematic nature.   Using the same starting
seed for  the random  number  generator  ensures  that the  same  produc-
tivity numbers are used for  the same mines in the three different cost
scenarios (i.e., the  BPT, the  BAT-2  and  BAT-4 case^.   Therefore, the
variation  in  the  BAT-4  and  BPT  supply  estimates  -  caused  by
productivity  sampling -  are  systematic and cancel  in  the calculation
of the impact estimate, when  BAT-4 and  BPT supplies are subtracted.

Having obtained mean  and standard deviation  estimates  for the  BPT and
BAT-4  cases,  the maximum  variation due  to water  acidity and  flow
sampling is derived for  the  Pennsylvania supply curve.   From  this an
estimate of  the maximum variation due directly to  water quality and
flow  sampling  for  other  impacted  regions  and   due  to  interaction
between  impacted  regions is estimated.    Comparing  the  sum total  of
these  two  types of  variation  for  individually impacted  regions  with
the estimated supply impact  shows whether or not the change in  supply,
attributed to the regulations,  is statistically significant.

The results from the thirty  supply curve  samples obtained for Pennsyl-
vania  are  shown  in  Table 34.    Pennsylvania  is  used  in  this  case
because it is shown  to have  the largest impact in the BAT-4 case.  The
mean supply  under the BPT  and  BAT-4 case  are, 47.3 million tons and
46.8 million  tons.   The difference in maximum deviation due to water
quality  and  flow sampling under BAT-4  is  found  to  be approximately
0.44 million tons or about  1% of the mean.

A  region  with a smaller number of mines  than Pennsylvania,  such as
Ohio,  will  have a  larger possible variation.   A region with a larger
number  of  mines than   Pennsylvania   will   have   a  smaller  possible
variation.   The ratio  of  the  maximum possible variation  of  the two
different regions will be the inverse of the square  root of the number
of mines producing in the different regions.

Table 35 shows  the derived  values for these possible direct variations
in  supplies  for what has been  called  the "impacted regions"  and the
"balancing  regions."   The  "impacted regions"  in  this  particular case
are  the  regions which  have  a   decrease  in supply  volumes  because of
increases in treatment costs.  The "balancing  regions" are the regions
which  have an increase in supply.

The second type of variation, the interaction  variation,  is calculated
as  shown  in  Figure  15.   The  total  possible variation in the supply of
balancing  regions   that resulted  from  the  sampling  of flows  is
allocated to the impacted regions and vice versa.

In the case of  the "impacted regions,"  the total possible variation in
the  supply  of  the  "balancing  regions"  supply   is   calculated  and
assigned  to  "impacted regions"  using  the  decrease  in  supply  volumes
                                   B-83

-------
                              TABLE 34

               RESULTS OF THIRTY SUPPLY CURVE SAMPLES
                        FOR PENNSYLVANIA?!)	
Compliance
   Costs
            Solution
             Supply
Mean Supply
   (MMT)
                 Difference
                  in Max
Deviation
  (MMT)
                          (2)
(Max.  Dev
 T  Mean)
    BPT
    BAT-4
              48.0
              45.1
   47.3
   46.8
                                                 1 0.44
                                                              1 0.01
(1)

(2)
Pennsylvania has 189 contract nines.

I.e., the maximum deviation due to water quality  and  flow
sampling under BAT-4.
                               B-84

-------
                                                                TABLE 35

                                   ESTIMATED SUPPLY IMPACT OF BAT-4 ON THE CONTRACT MARKET COMPARED WITH

                        THE POSSIBLE VARIATION IN THAT ESTIMATE DUE TO UNCERTAIN INFORMATION ON MINE WATER FLOWS
oo
            THE IMPACTED REGIONS
POSSIBLE VARIATION
     IN IMPACT
     ESTIMATES
     (in MMT)
                      POSSIBLE VARIATION
                           IN IMPACT
                           ESTIMATES
THE BALANCING REGIONS      (in MMT)
123 BY 4
:GION MINES SUPPLY LOSS DIRECT INTER- TOTAL
MMT MMT ACTION
PA 189 48.00 2.90 0.5 0.60 1.10
UT 1 0.33 0.33 0.0 0.07 0.07



REGION
OH
WV(S)
VA
AL
MT
1
MINES
103
117
90
88
15
2
SUPPLY
MMT
31
26
14
22
75
.2
.2
.2
.6
.9
3
GAIN
MMT
0
0
0
0
0
.4
.2
.3
.1
.5
BY
DIRECT INTER-
ACTION
0
0
0
0
0
.4
.3
.2
.2
.0
0
0
0
0
0
.07
.03
.05
.02
.06
4
TOTAL
0.47
0.33
0.25
0.22
0.06
                                                              WY(P)    26   201.4    2.8   0.0    0.04   0.40

-------
                             FIGURE 15

                CALCULATION OF THE INTERACTION ERROR
TNTFRAfTTON FRRnR - fLoss  or Gain  *n  SuPPly*   (  Total  Possible  Error  in   x
INTERACTION ERROR - (   Total  LQSS  Qr  ^ ')   ("Balancing  Regions"  Supply)
INTERACTION ERROR = Maximum possible  deviation  in  supply  due  to  interaction.
   Loss or Gain in Supply   =    Loss or  Gain  in  Impacted  Region
   Total  Loss or Gain
Total  Loss or Gain of All Impacted
Regions with a Loss or a
Gain
   Balancing Regions
All Regions with a Gain
or a Loss Offsetting the Loss or Gain
in Impacted Regions
                                 B-06

-------
due to impact as weights.  In the case of the "balancing  regions," the
total possible variation due to flow sampling in "impacted  regions"  is
calculated  and  assigned  to  individual  balancing  regions  using the
increase  in  supply  volumes  due to  impact  as  weights.    As  shown  in
Table 35, for the  heaviest  impacted  regions 'e.g., Pennsylvania^, the
maximum  possible   variation  due  to  interaction   is  larger  than the
maximum possible "direct" variation.

As shown in the right hand column of Table 36,  for a number of  regions
with  relatively small  impact  estimates, the  sum  of  the  direct and
interaction variation is larger than the estimated impact.  Therefore,
the  impact estimate  for these regions is inconclusive.   However, the
contribution by these regions  to  the total  impact estimate calculated
for larger regions (e.g., Northern Appalachia)  is relatively  small.

The  same  analysis  is done  for the  impact  estimates  obtained  for th»
spot  market,  as  shown  in  Tables  37  and  38.   The direct  variation
resulting from  the flow sampling  are  estimated to be  about twice  as
large for the  small  spot market jnines as the  variation obtained for
the larger contract market mines.

The results of this error analysis for the spot market show again  that
the  impact  estimate  for  region(s)  with  relatively small  estimated
impacts is inconclusive.

C.  Sensitivity of the Impact Estimates to Systematic Errors

To test the sensitivity  of  the impact  estimates to systematic  errors,
it  is estimated  how the  supply  impact  changes  if large systematic
errors occur in the different costs  (i.e., production,  transportation,
burning and clean-up costs) used in  the analysis.  As discussed in the
introduction,  systematic  errors  in  these  cost   estimates   for the
impacted regions result  in  an  over-  or underestimation  of  the  "demand
elasticity" for coal  from  those regions and  this  results  in an  over-
or  underestimation of  the  supply   impact  attributable   to  increased
compliance costs.

It is not  possible to determine what  actual  systematic  error may  be
present in the  different cost  estimates  used  in the analysis.   There-
fore, two extreme cases are constructed by:
 The analysis  of  mine  water flows  shows  water flows  'in gallons per
ton  produced)  to  be significantly  higher for smaller  mines than for
larger mines.   As a result, errors in estimates  of minimum required
prices,  including  an  estimate  for  water  treatment costs, for  smaller
mines  will  be  larger,  resulting  in  large  possible errors  in  total
supply available at a given total cost.
                                 B-87

-------
                               TABLE  36
          POSSIBLE VARIATION IN ESTIMATED BAT-4 SUPPLY IMPACT
           DUE TO UNCERTAIN INFORMATION ON MINE UATER FLONS U)
                            (Contract Market)
Pennsylvania
Ohio
W. Virginia South
Virginia
Alabama
Montana
Wyoming,
Powder River
Utah
Supply
With
BPT

48.0
31.2
26.2
14.2
22.6
75.9
201.4
Impact
With
BAT-4
MMT/Y
-2.9
+0.4
+0.2
+0.3
+0.1
+0.5
+2.8
Maximum
Possible
Variation
MMT/Y
± 1-1
+ 0.47
+ 0.33
+ 0.25
+ 0.22
+ 0.66
+ 0.4
Inconclusive
Impact Estimate


Yes
Yes

Yes


0.33
-0.33
+ 0.07
(1)
   Concerning acidity and alkalinity and volume of flow
                                 D-88

-------
                                                            TABLE 37
   REGION
at
00
*   PA

    OH
ESTIMATED SUPPLY IMPACT OK BAV-4 ON
THE SPOT MARKET COMPARED WITH
THE POSSIBLE VARIATION IN THAT ESTIMATE DUE 0) THE UNCERTAIN INFORMATION ON MINE WATER FLOWS


THE




IMPACTED


MINES SUPPLY

401
89

MMT
7.25
1.84



REGIONS


LOSS
MMT
0.44
0.43

POSSIBLE VARIATION
IN IMPACT
ESTIMATES
(in MMT)
BY
DIRECT INTER- TOTAL
ACTION
0.07 0.03 0.10
0.15 0.04 0.19

POSSIBLE VARIATION

THE

1
REGION

MD
KY(E)
OK



BALANCING REGIONS

2
MINES

14
366


3
SUPPLY
MMT
0.27
6.15
0.33


GAIN
MMr
0.04
0.43
0.08
IN IMPACT
ESTIMATES
(in MMT)
BY
DIRECT INTER-
ACTION
0.02 0.02
0.08 0.20
0.00




TOTAL

0.04
0.28
0.03

-------
                               TABLE 38
          POSSIBLE VARIATION IN ESTIMATED BAT-4 SUPPLY IMPACT
          DUE TO UNCERTAIN INFORMATION ON~MINE WATER FLOWS (0
                             (Spot Market)
Pennsylvania
Ohio
Maryland
Kentucky East
Oklahoma
Supply
With
BPT
MMT
7.25
1.84
0.27
6.15
0.33
Impact
With ,.i
BAT=4u;
MMT/Y
-0.44
-0.43
+0.04
+0.43
0.08
Possible
Error
MMT/Y
+_ 0.10
+ 0.19
+ 0.04
+ 0.28
+ 0.03
Inconclusive
Impact
Estimate?



Yes


(1)
(2)
Concerning acidity versus alkalinity and volume of flow
Net losses in total coal supply are expected to be made up
by supplies from large contract mines in the Wyoming Powder
River Basin.
                                 B-90

-------
    t    Increasing or decreasing the  production  and utili-
         zation costs of coal  from  the most impacted region
         (i.e., Pennsylvania) by approximately  10% for both
         the BPT and BAT-4  case;

    •    Increasing or  decreasing  the  transportation costs
         from all  regions by 10% for  both  the BPT and BAT-4
         case;

    •    Increasing or  decreasing  the estimated  BAT-4 com-
         pliance costs by approximately 30%.

The  changes  in  production,  transportation,   and   utilization  costs
result in a lower  and a  higher  estimate of the "user cost elasticity"
of demand relative to the cost  elasticity used  in the impact analysis.

The  impact  estimate obtained with  the lower estimate  of the "demand
elasticity" for coal  from Pennsylvania  in  combination with the higher
estimate of  compliance  costs  demonstrates what  the impart estimates
had been if user cost increases at the margin 'per incremental unit of
coal  from  Pennsylvania)  would  have been  systematically  overestimated
in the analysis while compliance costs were underestimated.  Likewise,
the  impact  estimate obtained with the  higher "demand elasticity" for
Pennsylvania  coal   in  combination  with  the   lower   compliance  costs
demonstrates what the impact estimate  had  been if the impact analysis
had  underestimated  the  marginal user  cost increase for Pennsylvania
coal while overestimating the compliance costs.

As  shown  in  'Table  39, if  the  marginal  user  cost  for   coal  from
Pennsylvania  in  1984  is 10% higher than  as  estimated for the impact
analysis, then the  projected demand for  coal  from this region in 1984
will  be  approximately 6.7  million tons per year  (or 14%) lower.  If,
in  addition,  the  compliance costs  for mines  in  Pennsylvania  in 1984
are  30% lower  than  as  estimated  in  the  impact analysis,  then the
impact of the BAT-4 mine water treatment standards will be an increase
in  coal  supply from  Pennsylvania  of  2.7  million tons  to  offset the
decline in supply of coal from Ohio (by 4.1 million  tons  per yearN and
Alabama (by 3.2 million tons per year).
If  the  marginal  user cost  of  coal  from  Pennsylvania  in  1984  is
lower than  as  estimated for the  impact  analysis,  then  the  projected
demand  for  coal  from this  region  will  be  approximately 13.5 million
tons  per year  (or  27%)  higher   than  as  estimated  for  the  impact
analysis.   If  the BAT-4  compliance  costs  for mines in Pennsylvania  is
30% higher  than  as  estimated in the  impact analysis,  then the impact
in  terms of  reduced  coal  supply for Pennsylvania  in  1984 will  be 2.5
million tons per year,  or  approximately 13% less than as  estimated  in
the impact analysis.  Apparently,  the compliance costs are so small  that
a 30% increase of these costs for one region (PA) will allow  the model
to find a computer- initiative solution:   the answers given by the model
become nonsensical.
                                 B-91

-------
                                                       TABLE  39
03

O
ro
RESULTS OF SENSITIVITY TESTS IN
TERMS



OF CHANGES IN ESTIMATED SUPPLY IMPACTS
Marginal User Cost Marginal User Costs
for PA Coal 10% for PA Coal 10% lower;
ORIGINAL
•^ 	 ESTIMATES 	 ^

PA
OH
WV (S)
VA
AL
WY (P)
UT
ILL
TX
BPT
Supply
48.0
31.0
26.2
14.2
22.6
201.4
0.3
41.4
17.7
BAT-4
(MMT)
45.1
31.6
26.2
14.5
22.7
204.1
0.0
41.4
17.7
Impact
(MMT)
-2.9
+0.4 •
+0.2
+0.3
+0.1
+2.9
-0.3
0.0
0.0
Higher; Compliance Compliance Cost
*- Cost 30% Lower — *• *«-— 30% Higher 	 *
BPT
Supply
42.3
31.2
22.6 .
14.2
22.6
228.9
0.0
41.4
11.9
BAT-4
(MMT)
45.1
27.1
23.9
14.5
19.4
228.9
0.0
41.4
17.7
Impact
(MMT)
+2.7
4.1
+1.3
+0.3
-3.2
0.0
0.0
0.0
+6.8
BPT
Supply
62.5
31.2 -
26.2
19.8
22.6
166.8
0.3
45.6
17.7
BAT-4
(MMT)
60.0
31.6
26.4
19.8
22.7
170.5
0.0
44.6
17.7
Impact
(MMT)
-2.5
+0.4
+0.2
0.0
+0.1
+3.7
-0.3
-1.0
0.0

-------
These  results   demonstrate   that  the  impact  estimate   for   specific
regions is highly sensitive to any systematic underestimation  that may
have occurred in compliance costs resulting from the misspecifications
of the treatment costs  and/or  the mine water quality and flow volumes
for  individual  regions.  ^  systematic  underestimation by  30^ of the
compliance costs of one region relative to the compliance costs of th*>
other  impacted  supply regions will  result  in a  complete missper.ifi-
cation of the supply  impact  for  that region; i.e., the supply in that
region will  go up rather than down as a result of the BAT-4 standards.
However,  a  systematic overestimation of the  compliance  costs for one
supply  region  relative to  the estimated  compliance costs  for other
supply regions will  result  in a relatively  small  error in the supply
impact for that region.

D.  BIAS RESULTING FROM AGGREGATION ERRORS

Use  of average values  for  the  different  costs  used for  the impact
estimate  probably  results in  an  overestimation of  the  supply impact
attributable to an increase in compliance costs.

As discussed earlier,  the impact will have  been  overestimated if the
"demand  elasticity"   for  coal   supply from  the  impacted  regions   is
overestimated.  The  impact  analysis  shows  that  the main supply impact
is expected to occur  in Pennsylvania,  resulting  in a decrease in coal
supplies  from that region.  The overestimation of the "demand  elastic-
ity" for  coal supply from Pennsylvania results from:

    •     The assumption that all mines  use  a discount rate
          of 10% per year  to  estimate their minimum  required
          price;

    •     The use  of  average  mine  investment  and  operating
          costs to calculate minimum  required prices;

    •     The use of average transportation costs;

    •     The use of average coal quality characteristics  for
          coal produced  in Pennsylvania; and

    •     The use  of average burning  and  cleaning costs  for
          Pennylvania coal.

In the impact  analysis,  it  is assumed that  mines  will  close if the
compliance costs render it impossible to make an average  (DCF) rate  of
return on investment  of 10% per year.   In  reality,  the  more heavily
impacted  mines,  i.e.,  mines  incurring higher compliance costs, will
most probably continue  to operate  even if their  rate of return drops
below  10%  per  year.    Therefore,   the  supply   impact   is   probably
overestimated.
                                   B-9:

-------
 The  use of  average mine investment and  operating costs  to  calculate
 minimum required  prices (even  if appropriate  ffTlowance"" is made  for
 mine type  and mine  labor productivity^  and  the^use- of. average trans-
 portation  costs, coal quality characteristics  and  average burning  and
 clean-up costs will  result  in  an overestimation o^ the  "demand elas-
 ticity."  In reality, these costs are  different for the mines that, in
 the  analysis, are assumed to have  the  same  production, transportation
 and  burning  costs.    As  a  result  of  these  differences  in  costs,  the
 actual  "demand elasticity"  is lower than the  "demand  elasticity"  used
 in  the  analysis;  i.e., a  coal   production  cost increase  in  a supply
 region  because  of  increased compliance  costs  in  reality will  cause
 less of a  change in demand  for  coal from that region than shown by the
 impact analysis.  Therefore, the aggregation errors resulting from the
 use  of  average costs probably  result  in   a  conservative  (i.e.,  too
 high)  estimate of the supply impact.
<»U. S. GOVERNMENT PRINTING OF F I CE I I 98 1 -3« I -085 /46 4 1
                                   B-94

-------