United States
              Environmental Protection
              Agency
           Clean Air Markets Division
           Office of Atmospheric Programs
           (6204N)
EPA430-R-01-002
March 2001
SrEPA
IMPACTS OF THE ACID
RAIN PROGRAM ON COAL
INDUSTRY EMPLOYMENT
                     U.S. EPA Headquarters Library
                        Mail code 3201
                     1200 Pennsylvania Avenue NW
                      Washington DC 20460


-------
EPA would like to acknowledge ICF Consulting for project management and analysis of the data and the issues
central to this report.

-------
                                                                   U.S. EPA Headquarters Library
                                                                         Mail code 3201
                                                                   1200 Pennsylvania Avenue NW
                                  Table  of  Contents       3ton DC  2046°
            I.     Executive Summary                                                  i
 I           II.    Purpose and Background                                               1
 b~          III.   Projections of Changes in Labor Demand                                 7
^                      A. Overview of Methodology                                     7
 ^/\
 L>                      B. Changes in Coal Demand as Projected by the IPM96 Analysis        7
 (^                      C. Translation of Coal Demand into Labor Demand                   8
-K
^          IV,   Demand Changes in Context                                            11
                        A. Historical Employment Levels and Trends                        11
                        B. National Title IV Employment Impacts in Context                  12
                        C. Employment Changes in the Regions Most Affected by Title IV       13
                        D. Programs to Assist Unemployed Coal Miners                      15
            V.    Limitations of the Analysis                                             17
            VI.   References                                                          19

            APPENDIX A: Description of the 1996 Integrated Planning Model          .         A-l
                  A.I Analytical Overview                                              A-l
                        A. 1.1 Macro Energy And Economic Assumptions                    A-3
                        A. 1.2 Electric Energy Cost And Performance Assumptions             A-3
                        A.1.3 Pollution Control Performance And Cost Assumptions            A-5
                  A.2 References                                                      A-6

            APPENDIX B: Coal Use Projections With and Without Title IV                     B-l
            APPENDIX C: Differences Between 1990/1992 and 1996 Analyses                  C-l
                  C.I Analyses Considered                                               C-l
                  C.2 Industry Changes Since The 1992 RIA                                C-2
                  C.3 Comparisons Of Coal Demand Projections                             C-6
                  C.4 Changes In Employment Impact Projections                            C-9
                  C.5 References                                                       C-12

            APPENDIX D: Results of Modeling and Analysis for 2000                         D-l
            APPENDIX E: List of Acronyms and Definitions                                 E-1
            APPENDIX F: Peer Review Process                                            F-l

-------
lv

-------
                      I.  Executive  Summary
 This report addresses the impacts of the Acid
 Rain Program (created under Title IV of the Clean
 Air Act Amendments of 1990) on coal  mining
 employment. It revisits the results of a study that
 was originally conducted in 1990, which com-
 pared the economic impacts of the acid rain pro-
 visions of several legislative scenarios being con-
 sidered at the time. The earlier study projected
 that Title IV could have a  substantial effect on
 coal mining employment. It predicted a gross loss
 of 13,000-16,000 coal miner job slots' by the year
 2001 as a result of Title IV.

 The current study revisits the original 1990 analy-
 sis and the differences in the results are quite sub-
stantial. The 2000 report projects that by the year
2010, Title IV of the Clean Air Act could result in
a gross loss  of approximately 7,700 job slots, or
about half the loss projected by the  1990 study.
The net loss would be only 4,100 coal miner job
slots because 3,600 new job slots would be creat-
ed. The extent to which Title IV affects coal min-
ing employment,  specifically job slots, is  the
focus of this paper.

These changes should be considered within the
context of  historical  trends  in  coal mining
employment. Employment for coal miners over
the last 50 years peaked in 1978 at approximately
250,000 workers and has steadily declined in sub-
 Exhibit 1: Job Slot Changes in the Mining Sector: Coal Mining Employment Changes With and
 Without The Acid Rain Program
          250
                                                                           0 With Title IV

                                                                           • No Title IV
               1970
                   2010
                          Year
Source: U.S. Department of Energy, Energy Information Administration (EIA) (2000). Energy Policy Act Transportation Rate Study:
Final Report on Coal Transportation.
1 The measure of coal mining employment used in this report is lob slots' the sum of which is equal to the number of coal mining
jobs or to the number of working coal miners in any one year. For details, see page 1.

-------
sequent years. Even without Title IV, jobs for coal
miners decreased as productivity improvements
and other economic factors eroded the need  for
large numbers of miners. In 1990, the number of
coal workers fell to 130,000. By 2010, approxi-
mately 50,000 coal miner jobs are projected to
remain. Ninety-five percent of the decline (over
75,000 jobs) is expected to be due to productivity
gains, and only  five percent of the loss in jobs
(4,100) is expected to be attributable to Title  IV.
These percentages are more dramatic compared
to the peak of coal mining employment in  1978,
where  98 percent of  the projected job loss
between  1978 and  2010 is related to productivity
gains and only two percent of the net decrease in
coal miner job slots is due to Title IV.

This analysis is complicated by changes in  the
demand for labor  in the coal  industry, resulting
from factors unrelated to Title IV. Worker produc-
tivity improvements and the increasing  share of
production from strip mining due to increased
substitutability of  coal  types  have allowed  the
demand for coal to be satisfied with fewer work-
ers. These differences in productivity are a result
of the  differences  in mine types and  the differ-
ences in the  kinds of technologies used in each
mine type. Regional shifts of coal production
result in decreased labor requirements, or miner
job slots.

Concerns over the effects of Title IV  on coal
employment  stem  from limits  the Acid  Rain
Program places on  the emissions of sulfur dioxide
(SO2),  especially those emissions from the elec-
tric power industry. The Acid Rain Program lim-
its the number of tons of SO2 that are emitted to
about half the number of tons plants would emit
without Title IV's limits. To  comply with  the
emissions limits mandated by Title IV, some util-
ities changed the type of coal that was used while
others installed control technologies. The switch
by utilities to lower-sulfur coals can reduce  the
demand for high-sulfur coal and the workers who
mine it. The employment consequences  of Title
IV compliance decisions are derived largely from
the fact that different regions of the country tend
to have different levels of sulfur in their coals.
Coal found west of the Mississippi, for example,
is generally lower hi sulfur than coal found in the
Midwest  or  the  Appalachians.  Within   the
Appalachians, sulfur content  varies; northern
Appalachian coal tends to have  a  higher sulfur
content than central  or  southern  Appalachian
coal. Likewise, the sulfur content of Midwestern
coal also tends to vary somewhat but it is prima-
rily high  in sulfur. As plant  operators  switch
among coals based on sulfur content (in addition
to other characteristics such as  heat and moisture
content) in response to Title IV, they can increase
or decrease the demand for the  coal mined in dif-
ferent regions of the country. In turn, demand for
miners  in different regions  can increase or
decrease. These changes are  further complicated
by the fact that the labor required to produce a ton
of coal differs across the  regions. Thus, shifts to
lower sulfur coal can, in some cases, reduce the
net demand for miners, rather than simply shifting
the locus of coal  mined.

The results of this 2000 report show that the Acid
Rain Program has  had a  limited  impact  on coal
employment and  that the program's  future
impacts on coal mining employment will be con-
siderably lower than originally  predicted.

-------
                II.  Purpose  and  Background
This  report  reviews the  impacts of the  U.S.
Environmental Protection  Agency's (EPA) Acid
Rain  Program, established under Title IV of the
1990 Clean Air Act Amendments, on coal mining
employment1 The estimated impacts are derived
from  modeling runs conducted in  1996 to esti-
mate  the costs of air regulations. By comparing
coal demand by region for two model runs, only
one of which took into account the SO2 reductions
required by Title IV, it was possible to calculate
the broad regional shifts in coal production that
could be attributed to Title IV. Productivity esti-
mates for the eastern and western portions of the
country were then used to translate these changes
in coal demand  into changes in coal industry
employment.

The breakdown of coal supply into regions is not
sufficiently detailed to capture all of the intrare-
gional and intrastate shifts that may be caused by
Title  IV. Nonetheless, EPA considers the broad
conclusions  of the analysis to be  sound. The
report includes Title IV's impact on the number of
miners'jobs and how those jobs shift between the
eastern and western U.S. coal production regions.

Future employment impacts are expressed in
terms of  incremental changes in  "job  slots,"
which are defined as the number of workers need-
ed to produce the industry's projected output of
coal at projected productivity levels.2 The analy-
sis focuses on the extent to which Title IV results
in fewer coal mining job slots, rather than on Title
IV's impact on miners' employment. As mining
productivity increases and demand for coal is
steady, there will be fewer job slots but not nec-
essarily miner lay-offs if these  changes can be
met through retirement or voluntary job changes.
Because the demand for coal miners has been
changing significantly as a result of factors unre-
lated to Title IV, one important component of this
report is to document the extent to which coal
miner job slots would be expected to decline in
the absence of the Acid Rain Program~i.e., in the
"baseline"  scenario.  Comparing  incremental
effects of EPA's program to this baseline reduc-
tion in job slots shows the incremental effects of
Title IV.

"Jobs slots" is used in this report as a measure of
employment and employment impacts, because it
is identified with numbers of workers. Because
productivity per hour varies  from  region to
region, and labor hours per worker per year vary
as well, concentrating on job slots can mask some
of the effects of a change in demand. For instance,
total labor hours by miners could rise, while job
slots decline, if there were an increase in demand
for western coal and a decrease in demand  for
eastern coal. The changes in job slots presented in
this report could have been divided into changes
caused by drops in coal demand,  shifts to regions
with higher hourly  productivity,  and shifts to
regions with higher hours per work year; howev-
1. This paper uses the E1A definition of coal employment, that "includes all employees engaged in production, preparation, pro-
cessing, development, maintenance, repairs, shop or yard work at mining operation. It excludes office workers but includes mining
operations management and all technical and engineering personnel." (EIA, 1995)

2. In any given year, the number of mining jobs, and the number of working coal miners, are expected to equal the number of "job
slots.* This term is used to avoid the impression that a miner loses his or her job whenever there is a reduction in the demand for
mine labor, given that no miners need to lose their jobs if the rate of attrition matches the rate of reduction in demand. Other
measures of employment demand use the total number of employees or the number of shifts completed by employees. However,
as the length of shifts and number of hours worked per miner change over time, using these factors prevents consistent compar-
isons over time. Thus, the calculations in this report are based on a 2,060 hour work year in the East and a 2,536 hour work year
in the West (For more detail, see Chapter III.C). Although the number of hours per slot could always change, it would be hard to
predict what effects this might have on job slots. For example, if miners increase the number of hours worked per year by 5 per-
cent but there is 4 percent less shifting of coal, the number of jobs in the East would still decline.
                                                1

-------
er, an analysis at this level of detail would be
cumbersome and unlikely to add significantly to
the outcomes presented in this report. Because the
annual output per worker in the East and West dif-
fer by about a factor of four, while labor hours per
year differ by only about 20 percent, it is clear
that the annual productivity differences between
East and West are due primarily to the differences
in  output per miner per hour. Furthermore,
because the regional shifts in coal use are so much
greater than the changes  in total coal production,
it is apparent that most of the changes in job slots
are due to the interregional shirts.

Concerns over the effects of Title IV on coal
employment stem from the limits the Acid Rain
Program places  on emissions of sulfur dioxide
(SO2) from utilities, which used almost 90 percent
of the coal produced in 1996 (EIA, 1998). These
limits reduce the demand for high-sulfur coal and
the workers who mine it. Media reports have cited
job losses in the  thousands due to restrictions
imposed by the Clean Air Act. The New York
Times, for  example, reported that coal  mine
employment  in   Illinois,   Indiana,  Ohio,
Pennsylvania, western Kentucky, and  northern
West  Virginia plunged  more than 50 percent
between  1990 and 1996. While the change in
employment is attributed largely to automation,
the Clean Air Act in general and the Acid Rain
Program In particular are also blamed by  the
media, political groups, and mining interests for
the loss of miners' jobs in the high-sulfur areas of
the East (New York Times, 1996).

West  Virginia and other states producing high-
sulfur  coal  have  been   identified  as  suffering
severe employment losses as a result of environ-
mental policies,  which  include air  regulations;
however, policies that affect land use, waste dis-
posal, and mining methods can also be important
Loss of mining employment has effects not only
on the miners,  but on other employees in  the
region, as these miners' demand for retail, enter-
tainment,  and related  services  declines  with  the
loss of  income.  As recently  as  1999,  the
Associated Press reported that "limits on sulfur ,
pollution were imposed during the 1990s by  the
Clean Air Act,  [subsequently] decimating  the
 high-sulfur coalfields of northern West Virginia,
 western Pennsylvania, Ohio, Indiana and Illinois"
 (AP, 1999).

 This report examines the validity of such employ-
 ment loss claims. The Acid Rain Program was
 developed pursuant  to  a  mandate contained in
 Title IV of the Clean Air Act as amended in 1990.
 Title IV sets two broad goals for EPA to reduce
 acid precipitation. First, SO2 emissions are to be
 reduced  by  10 million tons per year below the
 level in  1980. Second, NOx emissions  are to be
 cut (in combination with regulations from other
 Titles of the Clean Air Act, including Title n,
 which affects mobile sources) by 2  million tons
 per year below the 1980 level. Almost all of Title
 IV's SO2 and NOx reductions are to be made from
 coal-fired power plants operated by utilities.

 Efforts to meet the NOx reduction goals are not
 expected to have any significant impacts on coal
 use or miners' employment. Most plants, through
 modifications to the  power plant burners, can
 meet the moderate reductions in power plant NOx
 emissions required by the Title IV NOx program.
 Furthermore, the type of coal burned does not suf-
 ficiently affect either NOx emissions rates in the
 absence  of combustion controls or their control
 efficiency. Thus, the NOx  program does not dis-
 courage the use of any particular type of coal, and
 would have  minimal effects on the  total use of
 coal.

 Title IV's SO2 program, on the other hand, could
 potentially  cause a  significant increase in  the
 costs for utilities that burn high sulfur coals. The
 program operates by distributing authorizations to
 emit SO2, called allowances, to the owners of fos-
 sil fuel fired power plants  that were  in operation
 by 1995. To be permitted to emit SO2, plant own-
ers must hold allowances; for each  ton emitted
from  a   plant, the owner must  surrender one
 allowance. Thus, the  Act controls the number of
 tons that are emitted by limiting the number of
allowances  that  are distributed. By  distributing
fewer allowances than the number of tons that
 plants would emit without Title IV's limits,  the
Act causes  SO2  emissions  to go down. This
reduction in SO2 emissions,  in turn, reduces the

-------
associated sulfates that occurs when SO2 is trans-
formed in the atmosphere and returns to earth as
acid rain and dry acidic deposition. Sulfates are a
form of  fine paniculate matter that adversely
affects  human health, visibility, the  ecology  of
lakes and streams and the aesthetics and durabili-
ty of bronze or marble structures and statues.

Though the  Act controls the  total  number  of
allowances, it does not control  which plants will
emit SO2. Instead, EPA lets the allowances that
are distributed circulate freely  among  power
plants through the  market. Allowances may be
sold by plants that are able to  reduce emissions
enough to be left with surplus allowances, and
may  be  bought  by  plants   that need more
allowances than their allocation to cover their
emissions. Utilities may also bank allowances for
future use or  sale. The flexibility introduced by
the allowance market has important implications
for coal use, in that it opens up many  more com-
pliance options for utilities. Rather than selecting
only among control options that meet a particular
emissions limit for each stack, plants can choose
to switch to a lower sulfur coal, buy allowances,
or  use  banked allowances.  Utilities  may also
choose to install a scrubber that might be expen-
sive to install and operate, but might allow a plant
to over-control emissions.3 If a plant's allowances
exceed its emissions, the excess allowances may
be sold or banked. The system  also establishes a
nationwide marginal  price for SO2 reductions,
equal to the market price for  allowances. The
emissions reduction goal is derived from a statu-
tory cap of 8.95 million tons of SO2  emissions
from utilities.

Title IV's SO2 program is being implemented  in
two phases. In Phase I, which took effect in 1995,
263 of the largest boilers in  the eastern United
States with high emission rates received enough
allowances for them to emit about 2.5 Ibs. of SO2
per mmfitu. In Phase n, which started in 2000,
the rest of the fossil-fueled utility units serving
generators larger than 25 MW have been brought
into the system, and all units have been provided
with allowances sufficient to emit at levels up to
1.2 Ibs. of SO2 per mmBtu (or less, if their exist-
ing permits required a lower limit). Thus, begin-
ning in  2000, there is both an expansion  in the
number of utilities affected by the limit  and a
tightening of the  aggregate SO2 emission limit.
Knowing that their plants would have fewer SO2
allowances in 2000, many operators covered by
Phase I over-controlled emissions and banked the
excess allowances for use during the early years
of Phase II. Because utilities have been over-con-
trolling and banking excess allowances, the full
impact of Title IV's SO2 program on coal employ-
ment is not  expected to be seen until about a
decade  into Phase II (by the year 2010). For this
reason,  this report concentrates on projections for
the year 2010.

The combination of the SO2 emission limits under
Title IV and the flexibility of control strategies
provided by  the allowance system interact with
the regional  distribution of coal types to affect
coal employment. Because of the allowance sys-
tem, utilities are not obligated to change the type
of coal they use or the way they use it—they can
continue to use coal that is high in sulfur if they
purchase enough allowances or choose to  scrub.
On the other hand, the allowance market provides
a clear signal that the use of high sulfur coal has
an additional cost. Therefore, power plant opera-
tors are  given a market incentive to switch to a
lower sulfur fuel  or to install scrubbers in the
stacks to reduce SO2 emissions. If the choice is
made to scrub, the plant might continue to use or
even switch  to a  higher  sulfur fuel, depending
upon relative fuel prices,  because the scrubber
reduces SO2  emissions from high sulfur coal. In
Phase I of the program, 16 plants employing 27
boilers  chose to  scrub. Because sulfur removal
rates were 90-95%, these few units accounted for
3 This report is based on modeling that does not account for the SO, reduction potential provided by coal washing. In a study of
SO, controls for China, coal washing was assumed to reduce SO2 emissions from coal-fired generation by a third. In that it
excludes a potentially cost-effective control measure, this report may overestimate tie impacts of Title IV on coal industry employ-
ment (Liu and Spofford, 1994).

-------
about half of the  SO2  reductions in Phase I
(Ellerman et al., 2000).

The employment consequences of these compli-
ance decisions are derived largely from the fact
that different regions of the country tend to have
different levels of sulfur in their coals.4 Coal pro-
duction  regions  as defined  in  this  report are
shown in Exhibits  1 and 2.5 Coal found west of
the Mississippi, for example, is generally lower in
sulfur than coal found in the Midwest or the
Appalachians.  Within  the Appalachians,  sulfur
content varies; northern Appalachian coal tends
to be higher in sulfur than central or southern
Appalachian coal.  As  plant operators  switch
among coals based  on the coal's sulfur levels in
response to Title IV, they can increase or decrease
the demand for the coal output in different regions
of the country. In turn, demand for miners in dif-
ferent regions can increase or decrease.  These
changes are complicated by the fact that different
regions need more or fewer miners to produce a
ton of coal. Coal miners' productivity is generally
higher in some of the most important low-sulfur
mining regions in the West than in most of the
eastern high-sulfur mines because the coal in the
West lies  in  thick layers close to the surface.
Thus, shifts to lower sulfur coal can reduce the
net demand for miners, rather than simply shifting
the locus of coal mined.
Exhibit 1. Map of Coal-Producing Regions in the United States.

                           Powder River
   Northwest (N. &,-^          Basin        North Dakota
                                              Lignite   '
                                                                              \J
Source: Energy Information Administration, nttp://www.eia.doe.gov/cnea(/ooal/ooaLtrans/Tig3.htrnt
4 Because the heat content per ton of coat can vary, it is more precise to state that different regions tend to have different ratios of
sulfur per mmBtu of heat content.

5 The regions in Exhibit 1 show the breakdown used in IPM, which closely track the regions used by EIA.

                                                 4

-------
  Exhibit 2. Coal-Producing States By Region
Region
Eastern Region
Northern
Appalachia
Central and
Southern
Appalachia
Midwest
States in Eastern Sub-
Region
Maryland
Ohio
Pennsylvania
Northern West Virginia
Alabama
Eastern Kentucky
Tennessee
Virginia
Southern West Virginia
Illinois
Indiana
Western Kentucky
Western Region
Alaska
Arizona
Arkansas
Colorado
Iowa
Kansas
Louisiana
Missouri
Montana
New Mexico
North Dakota
Oklahoma
Texas
Utah
Washington
Wyoming
Coal Mined in 1992
(1,000 short tons and
percent of total)
3,341 (0%)
30,403 (3%)
68,981 (7%)
50,022 (5%)
25,796 (3%)
119,382(12%)
3,476(0%)
43,024 (4%)
112,142(11%)
59,857 (6%)
30,466 (3%)
42,686 (4%)
Coal Mined in 1992
1,534 (0%)
12,512(1%)
58 (0%)
19,226 (2%)
289 (0%)
363 (0%)
3,240 (0%)
2,886 (0%)
38,889 (4%)
24,549 (2%)
31,744 (3%)
1,741 (0%)
55,071 (6%)
21,339(2%)
5,251 (1%)
190,172(19%)
Sulfur Content
Primarily High Sulfur
Primarily High Sulfur
Primarily High Sulfur
Primarily High Sulfur
Primarily Low Sulfur
Primarily Low Sulfur
Primarily Low Sulfur
Primarily Low Sulfur
Primarily Low Sulfur
Primarily High Sulfur
Primarily High Sulfur
Primarily High Sulfur
Sulfur Content
Primarily Low Sulfur
Primarily Low Sulfur
High and Low Sulfur
Primarily Low Sulfur
Primarily High Sulfur
Primarily High Sulfur
High and Low Sulfur
Primarily High Sulfur
Primarily Low Sulfur
Primarily Low Sulfur
Primarily Low Sulfur
Primarily High Sulfur
High and Low Sulfur
Primarily Low Sulfur
Primarily Low Sulfur
Primarily Low Sulfur
Source: Energy Information Administration, Coal Industry Annual. 1996. Table 1:Coal Production by State, 1987,1992-1996, p.4.

-------
                Organization of the Report

                Section  III  of this report  outlines  projected
                changes in coal use by utilities from a 1996 analy-
                sis and translates those demand changes into the
                expected  number of job slot changes in  2010.6
                Section IV presents a discussion of those job slot
                changes relative to  national coal employment
                trends, and Section V outlines limitations of the
                analysis. Appendix A provides a description of
                the 1996 Integrated Planning Model, which was
                used for  the analysis; Appendix  B  shows  the
model projections of coal production  under the
baseline and Title IV for 2000, 2005, and 2010;
Appendix C contains a comparison of results
between the 1996 analysis and the analysis creat-
ed for  the  1990 Clean  Air Act Amendments;
Appendix D contains a comparison of results of
modeling and analysis for 2000 based on the 1996
analysis and the 1992 analysis; Appendix E pres-
ents  definitions of the acronyms and  abbrevia-
tions used in the report; and Appendix F summa-
rizes the peer review process that  was undertaken
for this report.
                6IPM1$ a model that allows the calculation of coal and other energy input use, operating costs, emissions, and least-cost
                responses to emission limits for electric generating units in the United States. Appendix A contains a detailed description of the
                1996 version of the integrated Planning Model.

                                                                 6
IN

-------
                                                                 U.S. EPA Headquarters Library
                                                                         Mail code 3201
                                                                 1200 Pennsylvania Avenue NW
                                                                     Washington DC 20460
             Projections  of  Changes  in  Labor
                                     Demand
A. Overview of Methodology

The projections of changes in labor demand as a
result of Title IV are calculated using estimates of
changes in coal use by utilities and regional mine
labor productivity projections. The change in coal
use by utilities is a model-derived output that is
generated using a series of assumptions about  the
utility  sector. The  1996 analysis (TPM96) was
produced as part of a series of model runs con-
ducted by ICF in  support of EPA's Clean Air
Power Initiative (CAPI) and the prospective
analysis of the effects of the Clean Air Act (CAA)
under Section 812 using a utility model called  the
Integrated Planning Model (IPM).1  This  study
complements an earlier work for EPA using  the
same model. By comparing two of these model
runs, one with Title IV's SO2 program and one
without the SO2 program, changes in utilities' fuel
choices were projected. Productivity of miners in
2010 is estimated using E3A data and IPM-pro-
jected productivity growth rates. These inputs can
be translated into regional changes in the demand
for miners. This section describes those calcula-
tions for the 1996 IPM analysis.

Over the past ten years, labor productivity growth
in the mining sector has greatly exceeded the his-
toric productivity growth rate in manufacturing.
An analysis of the  technology and management
practices used in existing  mines  indicates that
considerable opportunities still exist to improve
productivity (ICF, 1996).

EPA's productivity growth assumptions for  2010
are based on an analysts  of historic productivity
by state, in union and non-union areas, and in sur-
face and deep mines. Historic rates of improve-
ment are projected to continue, but decline over
time toward a more typical U.S. manufacturing
labor productivity growth rate (2.5- 3.0 percent
per year). This rate of improvement is  applied to
both existing and new mines (ICF, 1996).

The majority of this paper focuses on projections
of coal use and employment in 2010 from  the
1996 IPM analysis. Appendix C compares  the
1992 analysis completed for EPA's Regulatory
Impact Analysis  (RIA)  of the  Acid Rain
Implementation Regulations with the  1996 IPM
analysis.
B. Changes in Coal Demand as Projected
   by the IPM96 Analysis

Changes in baseline coal use in favor of low-sul-
fur  western coal  can  be  expected to  have an
impact on coal use and coal mining employment.
The regional changes in coal demand predicted by
IPM for the year 2010 as a result of Title IV range
from a loss of 58 million tons (or about 32 per-
cent) in northern Appalachia to a gain of 63 mil-
lion tons (or almost 11 percent) in the West from
a baseline  that excludes Title IV. Nationwide,
consumption is expected to drop by 2 million tons
in 2010 (which is about a fifth of one percent) as
a result of Title IV's SO2 requirements.2 Exhibit 3
presents the baseline of utility coal use and the
projected impact of Title IV on utility coal use in
the  year  2010. The analysis of coal  use and
1 Both analyses are available on the internet EPA's Clean Air Power Initiative, Oct. 1996. Obtained from
http://www.epa.gov/capi/capifs3.html, September 1998.

2 Because tons of coal from different mines vary in their heat content, the percentage change in the use of energy from coal is not
the same as the percentage change in tonnage used. Much of the coal produced in the West is subbituminous, which has a lower
heat content than the bituminous coal mined in the East. Thus, a shift of coal production from the East to the West would be likely
to reduce the use of coal measured on the basis of total energy content even if total tonnage did not change.

-------
Exhibit 3. Baseline Utility Coal Use Projections in the Year
2010 (millions of tons)
Coal Supply Region
East (Total of Eastern
Sub-Regions)
Northern Appalachia
Central & Southern
Appalachia
Midwest
West
Total US
Baseline
Utility Coal
Use
483
180
196
107
575
1,058
Utility
Coal Use
with Title
!V
. 418
122
228
68
638
1,056
Title IV
Impact
• 65
-58
+ 32
-39
+ 63
-2
Source: 1996 IPM tuns. Totals do not sum due to rounding.

changes in job slot  demands  is  conducted  by
region; refer to Exhibit 2 for a breakdown of the
states in the eastern and western regions. Though
the shift to western coal in response to Title IV is
associated with a  reduction in  demand .for coal
produced in the East, it could have a mitigating
effect on shifts in coal production within the East.
Because of its low sulfur  content, each ton of
western coal that is used in place of an equivalent
amount of high-sulfur eastern coal (from northern
Appalachia or the Midwest) could eliminate the
need to shift several tons of production from high-
sulfur to medium-sulfur eastern coal. This issue,
and  interaction  between  transportation cost
changes and  baseline demand patterns,  is dis-
cussed more fully in Appendix C.
            C. Translation of Coal Demand
               into Labor Demand

            As noted above, the difference in pro-
            ductivity  between the  eastern and
            western coal producing regions is pri-
            marily attributable to the difference in
            how the coal is mined.  Given  esti-
            mates  of  future coal  production,
            regional miner productivity rates are
            necessary to develop  projections  of
            mine labor demand in future years. In
            1994, the year in which the base case
            assumptions for the 1996 IPM analy-
            sis were developed, eastern coal min-
            ers produced an average of 3.28 tons
            of coal per hour while western miners
            produced  13.23 tons per hour (EIA,
2000c). Western miners also averaged more hours
of work per year — 2,536,3 versus an average of
2,060 per year for eastern miners (EIA, 1995).4

These two factors imply that each million tons of
coal mined per year in the East required 148 min-
ers:

     1,000,000 tons /(3.28 tons/hr * 2,060 hr/yr)
     or 148 miners,

while  each million  tons  mined  in the  West
required only 30 miners:

     1,000,000 tons /(13.23 tons/hr * 2,536 hr/yr)
     or 30 miners.5
3 Hours-per-job estimates are from the Coal industry Annual 1994, Energy Information Administration, October 1995, Table 1
(Production by region}. Table 39 (Average number of miners by region), and Table 38 (Productivity). Production (tons) is first divid-
ed by the number of miners to obtain tons/worker. This result is then multiplied by hours/ion to obtain hours/worker.

4 This means that any given shift in coal production from eastern to western mines wBI tend to decrease the number of job slots
because of higher productivity per unit of labor in the West and because the number of hours worked by a miner in the West is
higher than the number of hours worked by a miner in the East

5 Because of differences in heat contents among coals, a shift of a million tons from one region to another would not necessarily
leave the total energy content of the coal constant. Shifting a given amount of coal,  measured in terms of its heat content, from
East to West might require an increase in the West that exceeded the reduction in the East, slightly changing the job shift esti-
mates. In the modeling of the effects of Title IV, these differences in heat contents were accounted for.
                                                   8

-------
Thus,  shifting  output  from
eastern  to western  coalfields
will have employment impacts,
even given constant production
volumes. For example, a shift
of 30 million tons  of output
from eastern to western coal-
fields would require  148  * 30
or 4,400 fewer  miners in the
East, and only 30 *  30 or 900
more miners in the West, for a
net reduction  of  3,500  job
slots.6

Exhibit 4 shows the productiv-
ity rates that were used in the
1996IPM analysis.
Exhibit 4. Productivity Measures used in the Analysis7 (tons per
worker per hour and percentage growth per year)
Productivity Measure
1994 Base Productivity
Productivity Growth Rate: 1994-1999
2000 Productivity (calculated)
Productivity Growth Rate: 2000-2004
Productivity Growth Rate: 2005-2010
2010 Productivity (calculated)
East
3.28
4.5%
4.27
4.0%
3.5%
6.14
West
13.23
4.5%
17.23
4.0%
3.5%
24.77
Sources: U.S. Department of Energy, Energy Information Administration, Coal Industry
Annual 1994 and ICF analysis.
The  productivity  increases  for  coal mining
nationwide are incorporated into the IPM analysis
using the base year 1994 and the values listed in
Exhibit 4, above. Such changes  in productivity
would occur even in the absence of the Acid Rain
Program. At these rates of productivity increase,
output per hour in  the East would reach 3.28 *
1.045(2000 - 1994), or 4.27 tons per worker per
hour in the year 2000 and 6.14  in 2010, while
western output would reach 13.23 * 1.045(2000 -
1994), or 17.23 tons per hour in 2000 and 24.77
in 2010. This level of productivity represents an
87 percent increase over the productivity level in
1994 in both regions. Assuming no change in the
number of hours of work per year in each region
from the  1994 base  case assumptions, these
hourly productivity projections lead to estimates
of 79 miners per million tons per year in the East,
and  16 miners  per million tons per year  in the
West Using these values, the regional coal output
changes shown in the third column of Exhibit 3
                    can be translated  into the  labor requirement
                    changes shown in Exhibit 5.

                    Employment impacts of Title IV are reported in
                    terms of net and  gross changes.  The  net loss
                    (4,100 miner job slots) is the nationwide change
                  _ in jobs ~ balancing losses in some regions with
                    gains  in others. Gross losses (7,700 miner job
                    slots)  are the total job slot losses in regions that
                    show  decreases in the need for coal miners.8 As
                    discussed above, these projections can be seen as
                    indicating a fairly small decrease in net employ-
                    ment (of approximately 4,100 miner job slots in
                    2010,  which is approximately eight percent of the
                    baseline level of 54,000). At the same time,  the
                    figures suggest more substantial gross change in
                    job slots in individual regions: a total of 7,700
                    fewer miners (18 percent of the baseline level of
                    43,400 miners in these regions) would be needed
                    in northern Appalachia and  the Midwest in  the
                    year 2010; 3,500 more miners (34 percent of the
                    baseline level of 10,400 miners in these regions)
6 Thus, one job slot is the equivalent of one miner-year of work, or 2,538 hours in the West and 2,060 hours of labor in the East.

7 Actual growth rates could be sensitive to the demand shifts discussed in this report, especially if Title IV's effects on demand led
to more closures or cutbacks at the less efficient mines in each region. Differential cutbacks could increase the effects of coal
demand shifts (by increasing the number of job slots affected by each million ton shift), though this change might be offset in the
long run as higher eastern productivity led to greater competitiveness.

8 It should be noted that gross job loss within the mining industry does not fully capture all of the possible employment effects
within mining regions. As a result of multiplier effects on regional economies, additional job losses outside of the industry are also

-------
 would  be  needed  in central  and  southern
 Appalachia and the West

 Another aspect of these measures of gross job
 changes that should  be kept  in  mind  is  that
 intraregional shifts in coal demand (i.e., within
 northern Appalachia) may be even more impor-
 tant than the interregional shifts discussed above
 (i.e., between the East and West).9 Though the
 northern Appalachian region is described as pro-
 ducing  medium-to-high sulfur coal, for example,
 coals with a wide range of sulfur levels can be
 found in  that region.  Shifting from high-sulfur
 mines  to  low-sulfur  mines  within  northern
 Appalachia would cut average sulfur levels,  and
result in gross miner labor demand reductions in
the higher sulfur mines of the region, without reg-
istering as a net change in the region's output or
miner labor demand. To some extent, these shifts
in  intraregional coal shipments mean that the
measure of gross job slot changes presented in
this analysis understate the changes (both positive
and negative) in job  slots that  will result from
Title IV. On  the other hand, job slot shifts that
occur within smaller geographical regions are not
likely to be as disruptive  as those that involve
shifts in labor demand from the East to the West,
because many of the same miners might be able to
fill the job slots even after they have shifted.
9 Coal switching within regions in Phase I of Title IV (1995-2000) was four times as important as coat switching between regions.
   (Herman et al., 1997).

                                               10

-------
           IV.  Demand  Changes in  Context
The preceding sections have shown that Title IV
of  the Clean  Air Act  will reduce the overall
demand for miners in the coal industry and will
also induce regional shifts in demand for miners,
though neither of these changes is as great as had
been projected when the CAAA were enacted in
1990. They also touched on concerns over effects
of declining coal industry employment on  local
economies,  and  the  linkage  between  these
impacts and Title IV.

Before concluding that Title IV is or will  be a
major factor in these employment changes, how-
ever, it is important to place the projected effects
of Title IV in context. This section presents data
on  the sharp decline in  coal employment in the
past, along with projections of future changes in
employment expected even in the absence of Title
IV.  This baseline trend is then compared to the
projected effects of Title IV itself, both for the
nation as a whole and for regions that will  be
more heavily affected by Title IV due to the sul-
fur  content of their coal resources.  This compari-
son makes clear that it is the underlying trend
toward higher miner productivity, not the effects
of Title IV, that drives the  long-term change in
mining employment.
A.  Historical  Employment Levels  and
    Trends

The level of demand for coal miners can be ana-
lyzed by looking separately at the total amount of
coal  produced and miner productivity. Higher
coal production leads to increased miner demand,
while higher miner productivity (output per miner
per hour) reduces the mine labor required for a
given quantity of coal.
 From 1972 to 1990, annual coal production grew
 at an average rate of between three and four per-
 cent, almost doubling in that period to just over a
 billion tons in 1990. Since 1990, coal production
 has changed little. Thus, in the absence of signif-
 icant changes in miner productivity, demand for
 coal miners would be near its peak.1 Exhibit 6
 shows  annual coal production between 1972 and
 1995.
 Exhibit 6. Historical Trends in Annual Coal
 Production
Year
1972
1975
1980
1985
1990
1995
Annual Production
(millions of short tons)
602.5
654.6
829.7
883.6
1.029.1
1.033.0
Source: Annual Energy Review.
http://www.eia.doe.gov/emeu/aer/coai.html (EIA, 2000b).
 Miners'  productivity, however, did change dra-
 matically over this period. After rising rapidly
 and steadily from 1949 through the late 1960s,
 miner productivity  plunged  by about a  third
 through  the 1970s. This drop  in productivity
 resulted  from  a combination of labor unrest,
 aggressive implementation of mine safety regula-
 tions, and rapid entry of less experienced firms
 and miners in response to the oil shocks of the
 1970s (Darmstadter, 1997).

 Productivity (in terms of the hourly average tons
 of coal mined per miner) began climbing again at
11t should be noted that improvements in productivity have contributed to the increased use of coal over time. In the absence of
these changes in miner productivity, current levels of coat usage would be lower.
                                            11
             U.S. EPA Headquarters Library
                    Mail code 3201
             1200 Pennsylvania Avenue NW
                Washington DC 20460

-------
 the  end of the  1970s and grew  very rapidly
 through at least 1994: in contrast to the two-per-
 eent-per-year  drop in miner productivity from
 1970 to 1980, miner productivity rose by more
 than six  percent  annually  from  1980  through
 1994. Part of this rapid  gain resulted from
 improvements and  growing mechanization  in
 both surface mining (prevalent in the West) and
 underground mining (common among mines in
 the eastern U.S.)- Labor productivity at surface
 mines benefited from increases in the size of the
 excavating equipment (i.e., larger equipment can
 haul more coal per hour than smaller, more labor-
 intensive  equipment). At the same time, under-
 ground  mining shifted towards  highly  mecha-
 nized "longwall"  techniques in which mining
 machines ate into a coal seam along a wall a thou-
 sand feet long or more, while the unsupported
 roof of the mine was allowed to collapse behind
 the equipment as it moved. Miners' productivity
 at both underground and surface mines also bene-
 fited from  the  increasing  use of  sophisticated
 computer control systems (Darmstadter, 1997).

 Another factor leading to higher average produc-
 tivity among miners was the closure of some of
 the  smallest  and least efficient mines. These
 mines had been opened in response to the oil
 shocks and a resulting spike in the price of coal in
 the mid 1970s (EIA  1998,  p. 113). When coal
 prices declined, these mines were no longer prof-
 itable.

 Aggregate miner productivity was  also  boosted
 by the rapid increase in the West's share of coal
 production starting in the early  1970s.  Because
 miner productivity in the West is several times
 that  of the rest of the U.S., the nation's  average
 miner productivity increases as production shifts
 to the West. Western coal is transported to eastern
 utilities  by rail. Aiding the penetration of western
 coal into eastern markets, including the increased
use in the utility industry, has been a drop in rail
rates of more than a third between 1979 and 1993
(Darmstadter, 1997). Moreover, ambient air qual-
ity standards for SO2  favored low-sulfur coal
(even  before the promulgation  of regulations
under Title IV2).

Through the 1970s, the combination of increasing
coal production and declining output per miner
led   to   rapid  increases  in   employment.
Employment reached a peak in  1978 at about
250,000 workers. From that point, rapid increases
in productivity outstripped the rate of growth in
coal demand, and employment began to drop. By
1994, just before Title  IV's  SO2 program went
into effect, the number of coal workers had fallen
below 100,000, a loss of about  150,000 miner
jobs in less than 20 years.
 B. National Title IV Employment Impacts
    in Context

By projecting coal demand and miner productivi-
ty into the future, it is possible to project future
changes in  demand  for coal  miners. Under
assumptions that leave out the effects of Title IV,
modest increases are projected in coal demand by
utilities for both eastern and western coal through
at least 2010. These  estimates  are  presented in
Appendix B, which also  shows projections for
total coal demand under simplifying assumptions
that the ratio between utility demand and  total
demand is constant3 Over the same period, annu-
al increases are projected in miner productivity of
4.5 percent until 2000, falling to 4.0 percent per
year for the next five  years, and then to 3.5 per-
cent  in  2005 and  thereafter (EPA,  1996).
Combining region-specific estimates of coal pro-
duction for 2000, 2005,  and 2010 with estimates
of labor requirements  per million tons of produc-
2 The influence of air quality standards in existence before the Title IV regulations were promulgated may have been weakened by
the nature of the SO2 emissions rate targets, a large portion of which (perhaps over 90 percent), were non-binding. State
Implementation Plans generally are not the binding mechanisms that drive reduction in SO2 emissions and are only one way to
meet local air quality standards (Burtraw, 2000).

3 The numbers supporting these graphics are presented in Appendix B.

                                               12

-------
don (calculated using projected increases in min-
ing productivity from a 1994 baseline) yields pro-
jections of mining employment in the absence ofJ
Title IV. These projections are shown in Exhibit
7a. Exhibit 7b provides the data points shown in
Exhibit 7a.
Exhibits 7a and b. Job Slot Changes in the Mining Sector: National
Coal Mining Employment With and Without Title IV
  7a. 250
                                                 CT With Title IV  !

                                                 • No Title IV
           1970
                     Year
                                            (IMS)
Sources: 1) Data from the U.S. Department of Energy,
Energy Information Administration, Coal Industry Annual
1994; 2) ICF analysis.
7b.
Year
1970
1975
1980
1985
1990
1994
2000
> 2005
2010
Thousands of miners
With Title
IV
150
175
230
169
131
98*
67
60
50
Without
Title IV
150
175
230
169
131
98
69
63
54
  Sources: 1) Data from the U.S. Department of Energy,
  Energy Information Administration, Coal Industry Annual
  1994; 2) ICF analysis.
 Exhibit 7a  also  shows  projected  coal mining
 employment if the effects of Title  IV are taken
 into account. This projection, shown as the light-
 colored bar, was calculated based on the coal out-
 put projections shown in Appendix B under the
 heading "Projections Including Effects of Title
                       IV," and identical produc-
                       tivity estimates.  Exhibit
                       7a  shows  clearly that,
                      . although mining employ-
                       ment has fallen dramati-
                       cally since its peak in the
                       late 1970s, and is expect-
                       ed to continue falling,
                       most of  this drop  has
                       occurred and would con-
                       tinue to occur independ-
                       ently of  Title IV. This
                       same point  is made in  a
                       different way in Exhibit
                       8. Exhibit 8  compares the
                       cumulative  loss in min-
                       ers'  job  slots  through
                       2010 from the 1978 peak
                       and from 1990 (the year
	     the CAAA  was enacted)
                       to the  miners'  job slot
 losses attributable to Title IV. As shown, the loss-
 es attributable to Title IV are small compared to
 the drop in miner job slots that was already under-
 way and continues to occur due to increases in
 coal miner productivity.
 C. Employment Changes in the Regions
    Most Affected by Title IV

 The preceding section concentrated on the effects
 of Title IV on total nationwide coal employment,
 netting out the reductions in miners' job slots in
 some regions with the gains in others. It is also
 useful to focus specifically on the regions that are
 projected to lose miner job slots as a result of Title
 IV, keeping in mind that there will be offsetting
4 There may have been some effect on employment in 1994, even before Title IV took effect, as a result of utilities anticipating the
regulations. This possible effect was not modeled in EPA's analyses.

                                                13

-------
 gains in job slots for miners in other
 regions.

 In addition  to projected output by
• region and for the entire U.S., Exhibit
 3 also shows total output and output
 changes in northern Appalachia and
 the Midwest. Coal  employment  in
 these  two  regions  is  expected  to
 decline under Title IV because of the
 higher  sulfur levels in  their  coal
 resources. To put these changes in the
 context  of  historical  employment
 changes  in  these  areas, Exhibit 9a
 shows  total  employment  in these
 regions  with and  without  Title  IV.
 Exhibit 9b provides the data points
 shown  in Exhibit  9a.  Though  the
 effects of Title IV are more evident in
 Exhibit 9a than in  Exhibit 7a, which
 shows national mining job losses, it is
 again true that most of the  lost job  slots came
 before Title  IV, and  that mining  employment is
 expected  to continue to  decline in these regions
 even if Title  IV did not exist. Exhibit  10 displays
 the cumulative loss in miners' job slots through
 2010 from the 1978 peak and from 1990 (the year
Exhibit 8. National Coal Mining Employment Losses
With and without Title IV

200 i
iftn
ifin
1dfl
•8-
H ion .
9 inn .
O an

Afk
Oft
0.
















1













-,










-
1985 1990 1994 2000 2005 2010
Year
|
•Total Reductions in Job Slots. Relative to 1978
O Total Reductions in Job Slots, Relative to 1 990
•Total Difference in Job Slots, Relative to Baseline Without Title IV
Sources: ICF analysis of and data from the U.S. Department of Energy,
Energy Information Administration, Coal Industry Annual 1994.
         the CAAA was enacted) to the miners' job slot
         losses attributable to Title IV. By the year 2010,
         the loss in job slots attributable to Title IV will
         represent only about ten percent of the reduction
         from the peak in 1978.
 Exhibit 9a. Coal Miners' Job Slots With and Without Title IV in Northern  Appalachia and the
 Midwest
                                                                                         1
              1970
                                                                                 2010
                  Year
                                               14

-------
 D. Programs to Assist Unemployed Coal
    Miners

 Mine and other coal industry workers facing job
 losses attributable  to the requirements of the
 Clean Air Act are eligible for grants from the
 Federal government for job training, educational,
 and relocation assistance.  Operating  between
 1992 and 1993, the Clean Air Employment and
 Training Act (CAETA) was designed as a special
 appropriations through the U.S. Department of
 Labor to assist such workers. The program was
 discontinued after 1993, but the Department of
 Labor maintained the authority to provide grants
 through a discretionary fund in Part B of Title III
 of the Job Training Partnership Act, as amended
 by  the 1990  CAAA in Title XI  "Clean Air
 Employment Transition Assistance," (29 U.S.C.
 1501).5

 As  of 1998, over $82 million was granted by the
 Federal government  between 1992 and 1996 to
 coal mining companies, states, and the United
Exhibit 10. Coal Miners' Job Slot Losses With and Without Title IV in
Northern Appalachia and the Midwest
   Exhibit 9b. Coal Miners' Job Slots With
   and   Without  Title  IV  in  Northern
   Appatachia and the Midwest
Year
1970
1975
1980
1985
1990
1994
2000
2005
2010
Thousands of miners
WithTitielV
70
, 90
105
.71
52.3
35.8
28.6
23.5
18.1
Without Tide IV
70
90
105
71
52.3
35.8
34
31
26
Sources: ICF analysis of and data from the U.S. Department of
Energy. Energy Information Administration, Coat Industry
Annuan994.
inn -,
i on -
1 au
• 80
W.
£ on .
8 50
3 W
1 £ 40 .
1 K 40
30 •
20
m .
0.
























1i











385











1<











390











1i











39











4
Yea











2(
r








^ —
£F

)00











2(








^l
*!—
yr
m
>05











2(







-]
s|_
§f~
I
)10














D Total Reductions in
Inh mnte Relative !
to 1978 {
a Total Reductions in
Job Slots, Relative
to 1990

•Total Difference in
Job Slots, Relative
to Baseline
Without Title IV |


|
Sources: ICF analysis of and data from the U.S. Department of Energy, Energy Information
Administration, Coal Industry Annual 1994.
  Mine Workers Union for provision of job coun-
  seling,  vocational  and  occupational training,
  needs  related payments,  and  related services/
  Workers  in eastern  and  midwestern States
                          received  the majority
                          of grants and funding,
                          with    Illinois    (12
                          grants,  $32.5 million)
                          and  West  Virginia
                          (seven  grants, $23.5
                          million), receiving the
                          largest  share of the
                          assistance.  The aver-
                          age  grant   served
                          approximately   182
                          workers and provided
                          nearly $2.4 million in
                          dislocated    worker
                          services. Exhibit  11
                          shows,  by  State, die
                          number  and value  of
                          CAETA and Title  m
                          grants directed to dis-
                          located coal miners
                          between  1992  and
                                                                           1996.
5 During 1992 and 1993, almost $25 million was allocated through CAETA. An additional $58 million was awarded through the dis-
cretionary Title III program. Nearly $31 million was granted through Title Hi in 1995.
                                               15

-------
                Exhibit 11. CAETA and Title III Grants: Distribution by State 1992-1996 Program Years
Location of Grantees)
Illinois
West Virginia
Indiana
United Mine Workers Union
Pennsylvania
Ohio
Kentucky
Missouri
Idaho
Kansas
Texas
Total
Number of Grants
Awarded
12
7
1
1
1
8
1
I
l_ 1
1
1
35
Number of Workers
Served
2,119
1,258
682
625
543
495
225
192
164
33
30
D93OD
Total Amount of Grants
(Undiscounted)
$ 32,508,695
. $ 23,482,784
$ 6473,467
$ 2,000,000
$ 1,400,000
$ 8,103,152
$ 5,249,890
$ 1,000,000
$ 485,027
$ 900,000
$ 842,189
$ 82,545,204
                 Source: Brian Deaton, US Department of Labor, February 27,1998. Table titled 'Projects Funded to Assist Dislocated Workers
                 from Clean Air Amendments Impacts.*
                                                                  16
*\

-------
             V. Limitations  of the Analysis
The analyses conducted for this report incorporat-
ed a fine-grained representation of the behavior of
a large number of industrial  entities; it covers
both a long period of time and a wide geographi-
cal area. As with any similar  attempt to project
the future in detail, it is subject to limitations and
uncertainties. Thus, several factors could lead to
cost and emission impacts above  or below the
reported impacts.  Those factors are shown in
Exhibit 12.1
Exhibit 12. Limitations of the Analysis
Limitation
This study is based on the results of two energy/economic models, IPM and CEUM, both of which
necessarily make various simplifying assumptions. For example, these models assume that utilities
act so as to minimize the total cost of producing a given quantity of electricity, setting aside other
motives such as a desire to preserve local mining jobs. The models also assume risk neutrality and
perfect foresight. This may affect job loss because, for example, utility planners might expect
higher allowance prices than are actually seen, and over-invest in scrubbers, which use higher sulfur
coal that would further limit job loss. Even with entirely correct information on the current
population of generating units, however, the difficulty of predicting changes in that population (as
well as in other factors) in the future would add uncertainty to the projections.
Pollution Control Costs and Performance - EPA used estimates of SOj scrubber costs and
performance that reflect the current state of the ait. However, technological progress stimulated by
competition could lead to improvements in the performance and cost of pollution control technology
in the future. These improvements could in turn lead to greater reliance on scrubbers relative to coal
switching, and therefore to smaller changes in coal mining employment. For this reason, the
Agency's estimates of future job impacts could be overstated.
The analyses used for this report relied on a database that consists of information on virtually every
utility generator in the U.S. The Agency has assembled the best information on each boiler and
generator that is publicly available. Inevitably, when working with information on such a large
number of facilities, some units might not be represented correctly. Improvements to die database
could lead to changes in estimated impacts.
Though most of the attention given to the job impacts of Title IV relates to coal miners, Title IV
positively affects employment in railroads and pollution control equipment manufacturing,
installation, and operation. These secondary employment effects have not been examined in this
analysis. It is important to note instances in which environmental programs will generate new jobs
to offset jobs that are lost The crux of the controversy over Title IV, however, appears to be
centered on the gross number of mining jobs lost in high-sulfur coal mining regions, rather than on a
measure of net jobs created. As no number of jobs created in the transportation or pollution control
industries would have an effect on the gross reductions in labor demand due to coal miner job
losses, no estimate of jobs created was produced for this report
Because coal-mining regions were the geographical unit of analysis, this report did not pick up
potential shifts between states in these regions, or within individual states. This limitation probably
has little effect on estimates of the net changes in labor demand, but might have resulted in
underestimates of gross labor demand shifts.
To ensure consistency with related analyses, the most recent model runs used for this report were
conducted in 1996. and therefore did not use the most up-to-date projections of the growth in labor
productivity, other regulatory initiatives, or fuel prices.
Potential Influence
on the Findings
Decreased job loss
Decreased job loss
Mixed effects on job
loss
Decreased net job
loss
Increased gross job
loss
Mixed effects on job
loss
1 In addition to those limitations discussed, it should also be noted that this analysis does not take into consideration non-utility
use of coal, nor does it consider whether any difficulties might arise for displaced workers in finding new employment opportuni-
ties.
                                           17

-------
18

-------
                                 VI.   References
 Associated Press (1999). West Virginia Coal Industry Under Pressure. January 1999.

 Burtraw, Dallas (2000). Phone conversation with E. Holler, ICF Consulting. October 6,2000.

 Deacon, Brian (1998). US Department of Labor. Table titled "Projects Funded to Assist Dislocated Workers from Clean
 Air Amendments Impacts," February 27,1998.

 Darmstadter, Joel (1997). Productivity Change in U.S. Coal Mining. Resources for the Future. July 1997.

 Ellennan, A. Denny, Schmalensee, Richard, Joskow, Paul L., Montero, Juan Pablo, and Bailey, Elizabeth M. (1997).
 Emissions Trading Under the U.S. Acid Rain Program: Evaluation of Compliance Costs and Allowance Market
 Performance. MIT Center for Energy and Environmental Policy Research, Cambridge, MA, October 1997.

 Ellerman, A. Denny, Schmalensee, Richard, Joskow, Paul L., Montero, Juan Pablo, and Bailey, Elizabeth M. Markets
 for Clean Air: The U.S. Acid Rain Program. New York: Cambridge University Press, 2000.

 ICF Consulting (1996). Coal Supply Assumptions in the Clean Air Power Initiative. Attachment F-5-5.

 Liu, Feng, and Spofford, Walter O. (1994). Air Pollution Control in China: Current Status, Policy Issues, and
 Prospects. Mimeographed. Washington, D.C.: Quality of the Environment Division, Resources for the Future. August
 1994.

 The New York Times (1996). East's Coal Towns Wither In the Name of Cleaner Air. February 16,1996, page Al.

 U.S. Department of Energy, Energy Information Administration (EIA) (2000a), Coal Glossary.
 http://www.eia.doe.gov/OTeafrcoal/page/glossiitmIfe

 U.S. Department of Energy, Energy Information Administration (EIA) (2000b), Annual Energy Review.
 http://www.eia.doe.gov/emeu/aer/coal.html

 U.S. Department of Energy, Energy Information Administration (EIA) (2000c). http://www.eia.doe.gcrv/cneai/coai/cia/
 p6p01.txt

 U.S. Department of Energy, Energy Information Administration (EIA) (1995). Coal Industry Annual 1994. October
 1995. Table 1 (Production by region), and Table 38 (Productivity), Table 39 (Average number of miners by region) and
 note to Table 39.

 U.S. Department of Energy, Energy Information Administration (EIA) (1996). Coal Industry Annual, 1996. Table 1:
 Coal Production by State, 1987,1992-1996, p.4.

 U.S. Department of Energy. Energy Information Administration (EIA) (1998). Impacts of the Kyoto Protocol on U.S.
Energy Markets and Economic Activity. October 1998. Table B16, p. 202. This analysis did not assume compliance
with the Kyoto Protocol.

U.S. Environmental Protection Agency, Office of Atmospheric and Indoor Air Programs, Acid Rain Division, (1992).
Regulatory Impact Analysis of the Acid Rain Implementation Regulations.

U.S. Environmental Protection Agency, Office of Air and Radiation (1996). Analyzing Electric Power Generation
Under the CAAA. July 1996, page A2-A-2.

U.S. Environmental Protection Agency, Acid Rain web site - http://www.epa.gov/airmarkets
                                                 19

-------

-------
        Appendix A:  Description  of the  1996
                  Integrated Planning  Model
 This appendix discusses the use of the Integrated
 Planning Model (JPM) for the analysis, including
 assumptions about the baseline and about tech-
 nologies for power generation and emission con-
 trol.
A.I. Analytical Overview

The actions of electricity generators under a set of
regulations were projected using IPM, which is a
detailed computer model of the electric power
industry. IPM is designed to find the most effi-
cient (that is, the least-cost) way to satisfy the
demand for electricity under a series of limita-
tions or constraints. The constraints under which
IPM "produces" electricity can include a limit on
tons of SO2 emissions, and it is by setting this
constraint that the effects of Tide IV can be mod-
eled. Running IPM without a limit on tons of SO2
emissions produces a picture of the baseline situ-
ation in which the Title IV is not in effect. A rerun
of the IPM after adding a constraint that limits
SO2 emissions to a specified number of tons
shows what the industry would do to comply with
Title IV while keeping its costs as low as possible.
More detail on how IPM operates is provided
below   and  in  Analyzing  Electric Power
Generation Under the CAAA, Office of Air and
Radiation,  U.S.   Environmental  Protection
Agency, July  1996  (www.epa.gov/capi/IPM/
update.htm).

IPM is an  optimization model that uses a linear
programming  formulation to  select investment
options and to dispatch generating and load man-
agement resources  to meet overall electricity
demand and energy requirements. The model
selects the investment options based on the cost
and performance characteristics of the  available
options, forecasts of customer demands for elec-
 tricity, and reliability criteria. System dispatch,
 which involves the determination of proper and
 most efficient use of the existing and  new
 resources  available to  utilities and  their  cus-
 tomers, is determined using the resource mix, unit
 operating characteristics, and fuel and other costs.
 Unit and system operating constraints provide the
 system-specific reality to the model's simulations.
 The model has the capability of using forecasts of
 conditions, requirements, and option characteris-
 tics to make present decisions and is thus termed
 as dynamic. The model tries to represent the per-
 spective of utility managers,  regulatory person-
 nel, and the investing public in reviewing impor-
 tant financial options for the utility industry and
 electricity consumers. The model's objective is to
 minimize the discounted sum of capital and oper-
 ating costs over the full planning horizon.

 IPM has been used for over ten years by electric
 utilities, trade associations, and government agen-
 cies both in the U.S. and abroad to address a wide
 range of electric power market issues. The appli-
 cations have included capacity planning, environ-
 mental policy and compliance planning, whole-
 sale price  forecasting, and asset valuation.  EPA
 has used IPM extensively for environmental poli-
 cy and regulatory analysis. In particular, EPA has
 used IPM to analyze NOx emission policy and
 regulations as part of the Clean Air Power
 Initiative (CAPI) in 1996 and as a tool to analyze
 alternative trading and banking programs during
 the OTAG process in 1996 and 1997.

 IPM has undergone extensive review and valida-
 tion over this ten-year period. In April 1996, EPA
requested  participants in the CAPI  process  to
comment on the Agency's new approach to fore-
casting electric power generation and selected air
emissions. EPA received many helpful comments
and made a series of changes in its methodology
and assumptions based on CAPI participants' rec-
                                           A-1

-------
 ommendations. Recently, IPM and EPA's model-
 ing assumptions were reviewed as part of the
 OTAG process. Again, changes were made to the
 methodology and assumptions to accommodate
 participants' recommendations.

 The version of IPM used by EPA represents the
 U.S. electric power market in  21 regions, as
 depicted in Exhibit A-l. These regions corre-
 spond in most  cases  to the  regions and  sub-
 regions used by the North American  Electric
 Reliability  Council  (NERC).  IPM models  the
 electric demand, generation,  transmission, and
 distribution within each region  as well as the
 transmission grid that connects the regions.
cation of the model has focused heavily on under-
standing the future operations of coal-fired units,
which will have the greatest air emissions among
the fossil-fired units. The operation of other types
of  non-fossil fuel-fired  generation capacity,
including nuclear and renewables, are also simu-
lated but at a higher degree of aggregation.

Working with these existing model plants and
representations of  alternative new power plant
options, IPM determines the least-cost means for
supplying electric demand  while limiting air
emissions to remain below specified policy limits.
Multiple air emissions policies can be modeled
simultaneously. For example, IPM is used in this
 Exhibit A-1. Integrated Planning Model Regions in the Configuration Used by EPA
                                                                           MAAC
                                                                           •East
                                                                           •West
                                                                           •South
 Source: ICF Consulting.
The model includes existing utility power plants
as  well as independent power producers  and
cogeneration facilities that sell firm capacity into
the wholesale market. Data on the existing boiler
and generator population, which consists of close
to  8,000 records, are  maintained in EPA's
"National Electric Energy Data System (NEEDS).
In order to make the modeling more time and cost
efficient, the individual boiler and generator  data
are aggregated into "model" plants. EPA's appli-
study to  simulate  compliance  with existing
CAAA  Title IV  SO2  emission requirements.
While determining the  least-cost solution,  IPM
also determines the optimal compliance strategy
for each model plant.

A wide range of compliance options are evaluat-
ed, including the following:

 *J» Fuel Switching - For example, switching
                                             A-2

-------
    from high sulfur coal to low sulfur coal.
 <• Repowering - For example, repowering an  .
    existing coal plant to a gas combined-cycle
    plant.

 »J» Pollution Control Retrofit - For example,
    installing selective catalytic reduction (SCR),
    selective non-catalytic reduction (SNCR), or
    gas reburn (to reduce NOx emissions), or
    flue gas desulfurization (to control SO2 emis-
    sions).

 •J* Economic Retirement - For example, retiring
    an oil or gas steam plant.

 •> Dispatch Adjustments - For example, running
    high NOx cyclone units less often, and low
    NOx combined-cycle plants more often.

IPM provides estimates of air emission changes,
incremental electric power system costs, changes
in fuel use, and other impacts for each air pollu-
tion policy analyzed.

The model is not limited in scope to facilities
owned by electric utilities, but also includes inde-
pendent power producers (IPP) that provide elec-
tricity to the power grid on a firm-contract basis,
as well as IPP facilities larger than  25 MW that
provide power on a non-firm basis.

IPM simultaneously  models over an extended
time period, and reports results for selected years.
These analyses  provide  results  for  2000,  2005,
and 2010.

In applying IPM to analyze EPA emission policy,
EPA has developed a set of data and assumptions
that reflect the best available information on the
electric market and operating factors. The rele-
vant data and assumptions can be grouped into the
following categories:

*>   Macro Energy and Economic Assumptions -
    These assumptions are  related  primarily to
    electricity demand projections, fuel prices,
    power plant  availability, capacity  factors,
     lifetimes, and heat rates. Heat  rate data on
     individual coal plants are used in construct-
     ing the model plants. Also included in this
     category  are discount rate and year dollar
     assumptions.

 * Electric Technology Cost and Performance -
  " - These assumptions are related to electric tech-
    nology cost and performance for existing and
    new plants, as well as for existing plant refur-
    bishment and repowering.

 •t*  Pollution Control Performance and Costs -
    These assumptions primarily cover the per-
    formance and unit costs of pollution control
    technologies  for NOx and SO2.

Each of these sets of data and assumptions are
briefly discussed below. More detail can be found
in  EPA's report entitled "Analyzing Electric
Power Generation under the CAAA."
 A.I.I Macro Energy and Economic
      Assumptions

 EPA made assumptions about major macro ener-
 gy and economic factors, as shown in Exhibit A-
 2.  See  Appendix  No.  2* of EPA's  report
 "Analyzing Electric Power Generation under the
 CAAA" for details on most of the macro energy
 and economic factors.
 A.1.2. Electric Energy Cost and Performance
       Assumptions

In order to simulate the  electric power market
under  base case  conditions, assumptions were
made on the cost and performance of new power
plants  as well as for repowering existing power
plants. These characterizations of new power
plant costs and performances were used in IPM to
determine the least cost means for meeting pro-
jected  future  electricity requirements subject to
the base case emission restrictions and  the SO2
emission limits specified for the analysis.
                                            A-3
               U S  EPA Headquarters Library
                      Mail code 3201
               1200 Pennsylvania Avenue NW
                  Washington DC 20460

-------
Exhibit A-2. Key Baseline Assumptions for Electricity Generation
               Factor
           Assumption
    Source/Comments
   Discount Rate
            6 percent
ICF estimate
   Electricity Demand Growth Rate
   (% per Year)
        1993-2000=1.8%
        2001-2010=1.7%
        After 2010 =1.3%
NERC and DRI estimates
   Power Plant Lifetimes
Fossil Steam = 65 years if > 50 MW

           = 45 years if < 50 MW

    Nuclear = 40 year license length
                                                                         ICF estimate
   U.S. Nuclear Capacity (GW)
          1993 = 99 GW
          2000 = 97 GW
          2005 = 94 GW
          2010 = 90 GW
          2020 = 48 GW
ICF estimate based on
current utility plan and
assumed plant shutdowns;
No new nuclear construction
assumed
   Nuclear Capacity Factors (%)
           2000 = 75%
           2005 = 74%
           2010 = 73%
           2020 = 74%
                                                                         El A estimate
   U.S. Imported Crude Oil Prices
   (1995 $/BBL)
           2000 = $20
           2005 = $22
           2010 = $24
           2020 = $28
1996 EIA estimate, post
2010 extrapolation
   Wellhead Natural Gas Price
   (1995$ per MMBtu)
          2000 = $1.80
          2005 = $1.92
          2010 = $1.92
Model forecast of future gas
prices in the Base Case
   Fossil Steam Plant Availability (%)
                                                1994 = 80%
                                                2000 = 82%
                                             2005/10/20 = 85%
                                    ICF estimate based on
                                    historic trends
   Coal Mining Productivity Increases
   (% per year)
        1995-1999 = 4.5%
        2000-2004 = 4.0%
        2005-2009 = 3.5%
        2010-2014 = 3.0%
        2015-2025 = 2.5%
                                                                         ICF estimate based on long
                                                                         run trends
   Coal Transportation Rates
   (% change/year 2001-2010)
                                     Current rail, truck, and barge costs,
                                     declining 2% per year
                                    ICF estimate using unit cost
                                    information and FERC form
                                    423 delivery data and long
                                    run trends
Sources: As noted in the right-hand column.
                                                A-4

-------
 Power plant cost and performance assumptions
 were developed for the following new conven-
 tional and unconventional power plant types:
 New Conventional Power Plants
 *J« Conventional Pulverized Coal
 <• Advanced Coal (Integrated Gasification
    Combined Cycle - IGCC)
 «!* Combined Cycle
 *J* Combustion Turbine

 New Renewable/Nontraditional Options
 * Biomass IGCC
 •t* Solar Photovoltaics
 <• Solar Thermal
 *> Geothermal
 * Wind

 In order to  capture changes in technology  over
 time, cost  and  performance  projections were
 developed for 2000, 2005, and 2010. In general,
 the year 2000 estimates reflect current technolo-
 gy, the  2010 estimates reflect advancements in
 costs and performance, and the 2005 estimates
 represent midpoints between  these values.  The
 approach was adopted from work that the Energy
 Information Administration (EIA) did in support
 of the 1995 and 1996 Annual Energy Outlooks
 (AEO95 and AEO96). EIA had its approach peer-
 reviewed during its development.

 In addition to the AEO, key data sources used to
 develop the cost and performance assumptions
 are as follows:

 EPRI,  TAG  Technical  Assessment  Guide,
 Electricity  Supply - 1993, EPRI  TR-102276-
 V1R7, June 1993;

 SERI, The Potential  of Renewable Energy: An
 Intel-laboratory White Paper, SERI/IP-260-3674,
 March 1990; and

 TVA, Integrated Resource Plan Environmental
 Impact Statement, Volume Two,
Technical Documents, July 1995.

In addition to these assumptions on new power
plants, EPA also developed assumptions on the
cost and performance  of repowering existing
 power plants. The following three types of repow-
 ering options were considered:

  <• Repowering Coal Steam to Integrated
    Gasification Combined-Cycle
  *> Repowering Coal Steam to Gas Combined-
    Cycle
  •> Repowering Oil/Gas Steam to Gas
    Combined-Cycle

 The key sources  of data for this section are the
 repowering  studies   conducted  by  Bechtel
 Corporation  and  the  TVA Integrated Resource
 PlanEIS.

 For more details on the assumptions made about
 the cost and performance of new power plants and
 repowering  of  existing  power plants,  see
 Appendix  No. 3 of  EPA's report, "Analyzing
 Electric Power Generation under the CAAA."
 A.1.3. Pollution Control Performance and
       Cost Assumptions

 Exhibit A-3 contains the scrubber cost assump-
 tions used in the model. These assumptions were
 developed from a review of Phase I experience in
 the  1989  base case and 1992 SO2 RIA, updated
 for  1995 dollars and to reflect a 95 percent SO2
 removal rate (consistent with actual planned per-
 formance for  announced Phase' I  scrubbers).
 Actual costs will need to be  adjusted upwards to
 Exhibit A-3. Scrubber Cost Assumptions
 (1995$, excludes retrofit factor) (Based on estimates for medi-
 um sulfur coal (2%S))
Capital ($/kW)
Fixed O&M ($/kW/yr)
Variable O&M (mills/kWh)
Capacity Penalty (%)
Energy Penalty (%)
% Removal
172
6.2
1.0
2.1%
2.1%
95%
Source: U.S. Environmental Protection Agency (July 1996),
Office of Air and Radiation, Analyzing Electric Power
Generation Under the CAAA.
                                             A-5

-------
account for the difficulty  of the retrofit.  This
report assumes a retrofit factor of 1.1  for  large
power plants. For plants with installations below
500 MW,  these factors should  be scaled  by
(500/x)0.6, where x is the MW size of the power
plant.
For more details on the assumptions made about
pollution control cost and performance see
Appendix No. 5 of EPA's report, Analyzing
Electric Power Generation under the CAAA.
                                        A.2  References

    Bechtel Power Corporation (February 1996), Cost Estimates for NOx Control Technologies Final Report

    Bechtel Power Corporation (June 1996), Draft technical study on the use of gas rebum to control NOx at coal-fired
    electric generating units.

    Electric Power Research Institute (June 1993), TAG Technical Assessment Guide, Electricity Supply -1993, EPRITR-
    102276-V1R7.

    SERI (March 1990), The Potential of Renewable Energy: An Intel-laboratory White Paper, SERI/TP-260-3674.

    Tennessee Valley Authority (July 1995), Integrated Resource Plan Environmental Impact Statement, Volume Two,
    Technical Documents.

    U.S. Environmental Protection Agency (July 1996), Office of Air and Radiation, Analyzing Electric Power Generation
    Under the CAAA. For more about the Clean Air Act Amendments, visit the Clean Air Power Initiative (CAPI) main
    page at http://www.epa.gov/capi/capi.html.
                                                A-6

-------
 Appendix B: Coal  Use  Projections  With and
                          Without  Title  IV
This appendix presents the projections of coal
production by region made in the course of mod-
eling the effects of Title IV. Under assumptions
that do not include Title IV, only modest increas-
es were projected in coal demand by utilities for
both eastern and western  coal through at least
2010. These estimates are presented in Exhibit B-
1, which also shows projections for total coal
demand under simplifying assumptions that the
ratio between utility demand and total demand is
constant. Exhibit B-2 shows  the change  in
demand for coal from 1994 to 2000, 2005, and
2010. These demand estimates, along with pro-
jections of miner productivity, were used to proj-
ect future changes in demand for coal miners.
Exhibit B-1. Coal Use by Utilities and Total Production With and Without Title IV (millions of tons per
year)
Source: IPM 1996 runs.

Year
Region
Northern Appalachia
Central & Southern Appalachia
Midwest
West
Total
Northern Appalachia + Midwest

Year
Region
Northern Appalachia
Central & Southern Appalachia
Midwest
West
Total
Northern Appalachia + Midwest

Year
•legion
Northern Appalachia
Central & Southern Appalachia
vlidwest
West
Total
Northern Appalachia + Midwest
Baseline Demand Projections
Demand by
Utilities
2000

170
164
94
525
953
265
Estimated Total
Demand
2000

194
187
108
598
1,086
301
Projections Including Effects
of Title IV
Demand by
Utilities
2000

139
188
76
549
952
215
Estimated Total
Demand
2000

159
214
86
625
1,084
245

2005

183
180
96
569
1,028 .
279

2010

180
196
107
574
1,058
287
2005

208
205
109
648
1,171
318

2010

205
223
122
654
1,205
327
2005

133
212
70
612
1028
204

2010

122
228
68
637
1055
190
2005

152
242
80
696
1,170
232

2010

138
260
78
725
1,202
216
                                       B-1

-------
Exhibit B-2. Coal Use Changes Due to Title IV

Year
Region
Northern Appalachia
Central & Southern Appalachia
Midwest
West
Total
Northern Appalachia + Midwest
Change
(in millions of tons)
Demand by
Utilities
2000

-31
24
-18
24
-1
-50

Year
Region
Northern Appalachia
Central & Southern Appalachia
Midwest
West
Total
Northern Appalachia + Midwest

Year
Region
Northern Appalachia
Central & Southern Appalachia
Midwest
West
Total
Northern Appalachia + Mid west
2005

-50
32
-26
43
0
-75

2010

-58
32
-39
63
-3
-97
Estimated Total
Demand
2000

-35
27
-22
27
-2
-56
Percent Change
Demand by
Utilities
2000

-18.2%
14.6%
-19.1%
4.6%
-0.1%
-18.9%

2005

-56
37
-29
48
-1
-86
2005

-27.3%
17.8%
-27.1%
7.6%
0.0%
-26.9%

2010

-67
37
-44
71
-3
-111
2010

-32.2%
16.3%
-36.4%
11.0%
-0.3%
-33.8%
Change in Utility
Demand Due to
Title IV
2000

-18.0%
14.4%
-20.4%
4.5%
-0.2%
-18.6%

2005

-26.9%
18.0%
-26.6%
7.4%
-0.1%
-27.0%

2010

-32.7%
16.6%
-36.1%
10.9%
-0.2%
-33.9%
Source: IPM 1996 runs
                                             B-2

-------
                      Appendix  C:  Differences  Between
                          1990/1992 and  1996  Analyses
i
            Several analyses have been conducted over the
            years bearing on the magnitude of the Title IV-
            induced shifts in coal and labor demand. Changes
            in the models and assumptions used to conduct
            these analyses have raised the issue of whether
            the  earlier  conclusions  about  labor  market
            impacts have changed as well. This appendix dis-
            cusses the changes in assumptions, and compares
            the projected coal and labor demand impacts of
            Title IV  in the context of these changes. The
            assessment  is complicated  by several circum-
            stances: many different analyses of the impacts of
            Title IV have been conducted; there are multiple
            factors affecting job impacts; projection time-
            frames differ from study to study; there are gaps
            in the data that are required to make consistent
            comparisons; and  there are different ways  to
            interpret the same results. Still, it is possible to
            infer that the more recent analyses take changes in
            the industry into account, and are thereby correct
            in predicting lower overall labor demand impacts.

            This  appendix is divided into four sections, cov-
            ering the  following: the analyses that are com-
            pared; industry changes since the 1992 analysis;
            differences in impacts on coal demand; and dif-
            ferences in impacts on labor demand.
            C.I. Analyses Considered

            This document  examines comparisons  of the
            impact in the year 2010 (and, to a lesser extent,
            2000) of Title IV from two studies: ICFs analysis
            for the EPA's Regulatory Impact Analysis (RIA)
            of the Acid Rain Implementation  Regulations;
            and ICFs Integrated Planning Model (IPM) runs
for the Clean Air Power Initiative (CAPI) and
Section 812 Prospective analyses (which are the
focus of the body of this report). The analysis for
the RIA was conducted in  1991 using the Coal
and Electric Utilities Mode! (CEUM), with base-
line assumptions prepared in 1989. Outputs of the
model that are available include coal output by
region with and without the SO2 program, under
two baseline scenarios: a low emissions growth
case, and a high emissions growth case. As dis-
cussed  in  the  body  of this  report,  the
CAPI/Section 812 analyses were conducted in
1996  using  an improved  and updated utility
model called IPM. Available outputs include util-
ity coal use by region, with and without the Title
IV SO2 program; the runs use a single set of base-
line assumptions dating from 1996.'

Three related studies were examined, but the
detailed comparisons made between the 1992 and
1996 studies were not also done for these studies.
These related studies include the following: ICFs
1990 analysis of proposed  acid rain programs,
using CEUM and 1989 base case assumptions
(which was a precursor to the 1992 ICF analysis
used for the RIA); ICFs 1995 analysis of the Title
IV SO2 program using CEUM and 1993  base case
assumptions; and a recent study conducted at MIT
of the actual effects of Phase I of the SO2 program
against a backdrop  of changes prior to the pro-
gram's effective date. The 1990 analysis is impor-
tant in that it explicitly calculated and presented
coal  industry employment  changes associated
with changes in coal supply.  The regulatory
regimes that it analyzed—competing Senate and
House acid rain proposals—were similar to the
final proposals analyzed in 1991 and presented in
the 1992 Regulatory Impact Analysis. The 1995
            1 Though the 1998 version of IPM was run for the SIP call analysis (see http://wnww.epa.gov/ttn/rto/sip), those results were not
            used in this report because a no-Title IV case was not run using 1998 data. A comparison of the job slot impacts of Tide IV
            requires a comparison of the model results with and without Title IV.
                                                       C-1
                                                                           U S EPA Headquarters Library
                                                                                  Mail code 3201
                                                                           1200 Pennsylvania Avenue NW
                                                                              Washington  DC 20460

-------
ICF analysis differed in that it used a single set of
base case assumptions, which were closer to the
"low"  1989 base case, and recognized the possi-
bility of using subbituminous western coal from
the  Powder River Basin  in  eastern  boilers
designed to use bituminous coal. The MIT analy-
sis employed a questionnaire,  interviews with
power plant operators, data on allowance prices,
and actual use of coal and scrubbers in complying
with Phase I of the Title IV SO2 program. The
review of  actual data  in the MIT analysis was
used to estimate the actual cost of the program
and help explain the utilities' responses to the pro-
gram.  These three analyses are referred to in the
text for illustrative purposes.
C.2. Industry Changes Since the 1992 RIA

Of the numerous changes in the coal and electric
utility industries that have  occurred  since the
1992 analysis was conducted, three are of partic-
ular importance to the question of employment
impacts.2 These three changes—coal mining pro-
ductivity projections, the suitability of subbitumi-
nous  coal for eastern boilers, and shipping cost
reductions—are discussed in turn in the following
sections.

Coal  mining productivity. The trend  toward
improved coal-mine productivity appeared much
stronger by 1996 than had been assumed in the
analyses of  the  late  1980s  and early   1990s.
Exhibit C-l  compares the  productivity  growth
rates mat were used in the 1992 and  1996 analy-
ses.

Labor productivity growth in the mining sector
has greatly exceeded the  historic productivity
growth  rate in  manufacturing over  the last ten
years. An analysis of the technology and manage-
ment  practices  used in existing mines indicates
that  considerable  opportunities still exist  to
improve productivity.
EPA's future productivity growth assumptions are
based on an analysis of historic productivity by
state, in union and non-union areas, and in surface
and deep mines. The historic rates of improve-
ment are then projected to continue, but to decline
toward a more typical U.S. manufacturing labor
productivity growth rate (2.5 - 3.0 percent per
year) over  time. This  rate of improvement  is
applied to both existing and new mines.

A review of 1985-94 average labor productivity
growth in each region indicates that growth has
been about 5.6 percent per year (SAB,  1997).
Furthermore, this rate seems to be similar in all
parts of the country, regardless of union member-
ship or mine type. Rates vary from one year to the
next, but, overall, no significant differences can
be identified by mine type or union affiliation in
the last ten  years. Although the 1993 coal strike
has affected the trend,  the rate of improvement
has been declining over time.

Exhibit C-l shows that the hourly output that
would have been predicted in the 1992 analysis
for the year 2010 was considerably lower than in
the 1996 analysis. The more recent projections,
used in the 1996 analysis, set the rate of produc-
tivity growth for the years  1996 through 2000 at
4.5 percent per year, dropping to 4.0 percent and
then 3.5 percent over the next ten years. This rate
translates into substantially higher projected pro-
ductivity for the year 2010. This gain in projected
labor productivity might appear to favor western
coal producers, whose projected output per work-
er has risen more than  the output per worker at
eastern mines (i.e., an increase from 2.0 percent in
the 1992 analysis, to between 3.5 and 4.5 percent,
instead of the smaller increase from 3.0 percent to
between  3.5  and  4.5  percent for  the  East).
Because  labor is a much larger component of the
cost of eastern coal than western coal, however,
the unexpectedly large  improvements in eastern
productivity are actually more significant eco-
nomically than the even larger percentage gains in
western  productivity. Thus, labor productivity
2 An additional change resulting from restructuring since 1998 is that plants have changed from being EIA classified as traditional
utilities to non-utilities. Because this analysis uses data from 1996, this issue does not affect the outcomes.

                                               C-2

-------
 Exhibit C-1. Comparison  of Productivity Measures used in the 1992
 and 1996 Analyses (tons per worker per hour and percentage growth per year)
Productivity Measure
1989 Productivity (calculated)
1994 Base Productivity
2000 Productivity (calculated)
2010 Productivity (calculated)
Growth Rate to back calculate
1994 Rate to 1989 Rate for 1992
Analysis
Growth Rate: 1994-1999
Growth Rate: 2000-2004
Growth Rate: 2005-2010
1992 Analysis
East
2.50
. 3.28
3.46
4.65
5.6%
3%
3%
3%
West
10.07
13.23
12.52
15.27
5.6%
2%
2%
2%
1996 Analysis
East
NA
3.28
4.27
6.14
NA
4.5%
4.0%
3.5%
West
NA
13.23
17.23
24.77
NA
4,5%
4.0%
3.5%
Source: ICF analysis.

improvements might have tended to shift utilities
toward the use of eastern coal to some degree.

Substitutabitity of Coal. A second important
change in assumptions is the recognition that sub-
bituminous coal from the Powder River Basin in
the West can be used in eastern boilers that were
designed for bituminous coal. The ability to sub-
stitute Powder River Basin coal is  important for
two reasons: it is less expensive to mine than the
eastern coal it can replace,  and it is generally
much lower in sulfur than eastern coal. Thus, it is
potentially a very attractive fuel source both for
economic reasons and for compliance with regu-
lations requiring lower SO2  emissions. In early
analyses  of the effects of Title IV, coal from the
western Powder River Basin had been assumed to
play a minor role in utilities' compliance strate-
gies because of questions about the costs of trans-
porting it to the East, and about its suitability for
boilers designed for higher-heating-value/lower
ash bituminous coal. The 1990 and 1992 analyses
assumed  that eastern bituminous boilers would
not switch to western subbituminous coal, though
the text mentioned that this  kind of switch was
potentially important By 1993, the possibility of
this type  of coal switching was well established,
                        though it was assumed
                        that it would involve a
                        significant capital  cost
                        (ICF,  1993). Coal  pur-
                        chasing patterns show
                        the growth in the num-
                        ber of power plants that
                        are switching to  sub-
                        bituminous coal, which
                        primarily  comes  from
                        western  regions.  For
                        example, during a peak
                        electricity usage month
                        in 1999, approximately
                        136 of 244 coal burning
                        power plants east of the
                        Mississippi River, or 58
                        percent  of the  plants,
                        purchased  subbitumi-
                        nous  coal. This com-
                        pares  with 110  of 263
                        plants, or 42 percent of
plants, during the same month in 1993 (EIA, 1993
and 1999). The effects of this change in assump-
tions  can be seen in the  1995 analyses of the
effects of Title IV, which showed a significantly
larger shift toward western  coal at the expense of
lower-sulfur  central and southern  Appalachian
coal (ICF, 1995).

Assumptions about  Transportation Costs. A
third  change over the past six  years has  been a
very significant drop in the per-mile costs of ship-
ping coal by rail. This  decrease is attributed to
several  factors.  Railroads  were able  to  reduce
costs  in part because deregulation of the industry
in 1980 allowed the industry to restructure itself
by  abandoning  unprofitable lines and merging
with other companies. Costs were further reduced
through work force and equipment downsizing.
Some of these cost savings  were eventually
passed on to customers. Utilities were also able to
secure lower rates by increasing use of leased or
utility-owned rail-cars, rather than  railroad-
owned cars.

While the cost to ship coal between the various
supply-and-demand-regions fell overall, the  dif-
ferences were not always uniform. For example,
                                              C-3

-------
between 1988 and 1993, the cost to ship a ton of
coal from western  supply regions to the  east
North Central region declined by 40 percent on
average. At the same time, the rail costs to ship a
ton of coal from eastern supply regions to the east
North Central region declined only by about 25 to
30 percent. Because of this difference in the rates
of cost  decline, and the great distance from the
western coal  fields to the East, the cost to ship a
ton of low-sulfur western coals to eastern markets
has decreased relatively more than the cost to  ship
eastern coals to the eastern markets (EIA,  1995).
This relative change in transportation costs has
apparently outweighed,  by a considerable degree,
the advantages of increased labor productivity in
the East, leaving a substantial net incentive to use
western coals.
 The effects of this increased cost-effectiveness
 are illustrated in Exhibit C-2. The diagram is a
 schematic picture of the interplay of mining and
 transportation costs in determining the relative
 sales of western and eastern coals.  Costs are
 shown vertically, while space is shown horizon-
 tally. Western coal costs considerably less to
 mine, so point A (showing the cost of the western
 coal with no transportation  cost) is lower than
 point B. Not all customers would necessary find
 western coal  less expensive, however,  because
 shipping it  eastward is costly: each  additional
 mile that it must be shipped toward the East rais-
 es its delivered cost, as shown by the line sloping
 upward to the right. (The slope of the line is relat-
 ed to the cost of shipping in cents per ton-mile.)
 Eastern coal must also be shipped if it is to com-
 pete for the business of coal users  between the
Exhibit C-2. Schematic Explanation of Effects of Rail Transportation Rates on Coal Competition
      35
     =E
     •s »
      is
                                                                       .0*'
                                                                             B
                                   33
   West

Source: tCF analysis.
             Geographical Range Over Which Western Coal's
             Delivered Price is Lower than Eastern Coal's
    Geographical Range Over Which
i    Eastern Coal's Delivered Price is
|    Lower than Western Coal's
                                                                                 East
                                                C-4

-------
              eastern and western coal-fields: the delivered cost
              of eastern coal is represented by the line sloping
              upward to the left.

              At a certain point,  the delivered costs of the two
              coals are equal (shown as point C). There, the
              delivered cost of the western coal at that point
              will be mostly  transportation cost, while  the
              delivered cost of the eastern coal will  be more
              heavily based on mining cost. For all points east
              of this dividing  line, eastern  coal will have  an
              economic advantage; for all points to the west,
              western coal will be less expensive.

              A decline in the per-mile cost of transportation
              can be shown on the diagram as a flattening of the
              delivered-cost lines, as shown in Exhibit C-3: at a
              lower  cost  per ton-mile, each incremental dis-
                                                    tance adds less to the delivered cost than at the
                                                    higher cost. As a result, the delivered-cost lines
                                                    cross at a point further to die right (i.e., further to
                                                    the East),  and the dividing line between the area
                                                    served by western and eastern coal regions moves
                                                    as well. These diagrams illustrate the concept
                                                    that, as the transportation costs hampering the
                                                    spread of  inexpensive western coal come down,
                                                    the area of the country in which it is economical-
                                                    ly competitive expands.

                                                    The  net increase in the attractiveness of using
                                                    western coal, described above, would exist even
                                                    in the absence of Title  IV. In other words, coal
                                                    users in the Midwest and East would find western
                                                    coal  relatively more  cost-effective even if its
                                                    lower sulfur content were  of  no importance.
                                                    Therefore, part of the reduction in SO2 emissions
              Exhibit C-3. Schematic Explanation of Effects of Rail Transportation Rates on Coal Competition
\
                                                                                              Less with Haance
                  •5 «
                                  Reduction in Rail Roes
                                  Means Delivered Costs Rise
                                  Less with Distance
                                                                                                     n Rail Rates
                                                                                                     vend Costs Rise
                                                                                                             £
                                                                                                             g
                                                                                             S
                                                                                             ' a
                     ag     Geographical Range Over Which Western Coal'
                     O     rviiwrpd Price k Lnaier. at Hioh Rail Rates
                           Delivered Price is Lower, at High Rail Rates
West
           Wider Geographical Range Over Which Western
           Coal's Delivered Price is Lower, at Low Rail Rates
                                                                                      East
                Source: ICF analysis
                                                             C-5

-------
 required by Title IV  would  already  be accom-
 plished in the baseline.3 Because Title IV's limits
 are specified in  terms of a  nationwide cap on
 emissions, the shift toward  low-sulfur western
 coal in the baseline means that less incremental
 SO2 reduction will be needed  to meet the Title IV
 requirements. The MTT study of Phase I compli-
 ance strategies suggests that the shift to western
 coal for economic reasons cut baseline SO2 emis-
 sions by 425,000 tons  annually—a reduction that
 accounted for more than  ten  percent of the total
 emissions reductions under Phase I (Ellerman et
 al., 1997).
 C.3. Comparisons of Coal Demand Projections

 The baseline shift in coal use discussed above
 appears to have been captured by ICFs IPM
 model. This tentative conclusion is based on the
 fact that the baseline results from 1996 show a
 relative increase in the use of western coal and a
 drop in eastern coal use, compared to the 1992
 results. Because of differences  between IPM and
 the earlier CEUM, though, this conclusion is not
 certain. This section presents and discusses these
 results and the reason for the  uncertainty in the
 conclusion.

 Exhibit C-4 compares two different coal projec-
 tions in (he absence of Title IV: the 1992 RIA's
 projection of total coal production, and the 1996
 analyses' projection  of coal  use  by utilities.41
 Because these two  projections differ  (by  the
 amount of coal exported or used by  industry), it is
 not possible to show conclusively how projec-
 tions of utility coal use differ between the 1992
 and 1996 analyses. The sizes and directions of the
 differences between the two projections for east-
 em coals versus western coals, however, strongly
 suggest that the  more  recent analysis projects
 more western coal use by utilities.

 As seen in Exhibit C-4, ICFs more recent analy-
 sis projects utility use of eastern coal in both 2000
 and 2010 to be considerably below earlier projec-
 tions of eastern coal production. These differ-
 ences do not necessarily mean that projected util-
 ity use  of eastern coal' fell between 1992 and
 1996, because the differences shown  in the table
 could be explained by the quantity of coal taken
 by industry and exporters. The 1996 projection of
 utility use of western coal, however, is now high-
 er than the 1992 projection of western  coal pro-
 duction for the year 2000. Because utility use of
 western coal is only one component of total pro-
 duction, it must be true that the year 2000 projec-
 tion of utility use of western coal was higher in
 1996 than in 1992. A similar pattern is seen for
 the year 2010: the difference between 1992 and
 1996 projections for western coal is much small-
 er than  for eastern coal.  The differences between
 the 1992 and 1996 projections for both 2000 and
 2010 are easily explained by a shift from  eastern
 to western coal by utilities. Nevertheless, it is true
 that these patterns could also have resulted from
 other shifts (e.g., changes in total utility demand
and/or changes in  exports or industrial use).

Exhibit C-5 compares Title IV's projected impacts
on coal use for the year 2010 for the 1992 and
 1996 analyses. The 1996 analysis predicts greater
shifts between eastern and western coal than the
 1992 analysis, and a much smaller  shift from
high-sulfur Midwestern coal to low-sulfur central
3 It may well be that some of the changes in technology that induced the shift in favor of using more western coat were actually
induced by the existence of Title IV and the development of the Acid Rain Program. For example, the prospect of a need for sub-
stantial cuts in utility SO2 emissions could have given utilities an added incentive to develop ways to use tow-cost, low-sulfur sub-
bituminous coal in boilers originally designed for bituminous coal. To the extent that this is the case, the baseline discussed here
would not be a true non-Title IV baseline, and the effects of Title IV presented in this report would be understated compared to
their true values. No attempt was made to determine the extent to which the changes favoring the use of subbituminous coal were
endogenous in this way, though if the attempt had been made, the results of the analysis might fall somewhere between those
shown for the 1992 analysis and 1996 analysis.

4 Projections for the year 2000 are included in the table because their implications for baseline coal use are less ambiguous than
are the projections tor 2010.

                                                 C-6

-------
 Exhibit C-4. Basefine Coal Industry Output and Use Projections in the Years 2000 and 2010 (millions
 of tons per year)

Coal Supply Region
East
Northern Appalachia
Centra] & Southern
Appalachia
Midwest
West
Total
2000
1992 RIA
(total
production;
high base
case)
648
197
302
149
491
1,138
1996
Analysis
(use by
utilities)
428
170
164
94
524
952
Difference
(1996
analysis
minus 1992
RIA)
-220
- 27
-138
- 55
33
-186
2010
1992 RIA
(total ,
production;
high base
case)
955
290
359.
306
675
1,630
1996
Analysis
(use by
utilities)
483
180
196
107
575
1,058
Difference
(1996 analysis
minus 1992
RIA)
412
-110
-163
-199
-100
-572
  Source: ICF analysis. Totals do not add due to rounding.
and southern Appalachian coal was projected in
1996 than in 1992.

A greater projected shift toward the West in the
1996 analysis might seem counterintuitive, given
the expectation  that  more  recent  projections
should take into account the greater baseline pen-
etration of western coal due to lower rail rates and
expanded opportunities for using subbituminous
coal in the East. By reducing baseline sulfur emis-
sions, this shift should reduce the need for further
coal switching, and might have been expected to
reduce net job slot losses.
From another perspective, however, die changes
that led to a baseline increase in the use of west-
em coal would also  make  it a more attractive
option for the additional reductions in SO2 emis-
sions that will be needed in Phase II. Because the
baseline shifts toward the use of low-sulfur west-
em coal for economic reasons did not meet the
goals of Title IV's Phase II program by them-
selves, there will still be a need for incremental
sulfur reductions. The impacts of these incremen-
tal reductions will depend on the choices made by
power plant operators from  among their compli-
ance options. If western coal is less expensive to
Exhibit C-5. Comparison of Projected Effects of Title IV on Coal Use in the Year 2010 (millions of
tons per year)
Coal Producing Region
East
Northern Appalachia
Central & Southern Appalachia
Midwest
West
Total
1992 RIA
Analysis, High
Base Case
-35
-41
135
-129
38
3
1996 Analysis
-65
-58
32
-39
63
-2
Difference (1996
Analysis minus
1992 RIA)
-30
-17
-103
90
25
-5
           Source: 1992 SO2 RIA, Appendix 4b, 1996IPM runs. Totals do not add due to rounding.
                                               C-7

-------
ship than before and still has a substantial cost
advantage, it might still be the best compliance
choice for a large segment of the industry.5 Thus,
it would not be surprising to see greater shifts in
job slots from eastern to western coal producing
regions in a model in which western coal is more
accessible. The only question would be whether
the baseline shift to the West was large enough to
preempt the need for further shifts  in response to
regulations.

This question can be addressed using the analyti-
cal framework presented in Exhibits C-2 and C-3,
Exhibit C-6 shows the combined effects of (1) the
need to purchase allowances on the cost of using
high sulfur coal, and (2) the reduction in trans*
portation  costs in the baseline. The need to pur-
chase additional allowances  for every additional
ton of high-sulfur coal is illustrated  by the upward
shift in the costs for the eastern coal. Minemouth
    cost B shifts upward to B" to show the true cost of
    using a ton of high-sulfur coal under Title IV. The
    delivered cost curves sloping upward to the left
    shift as well; the resulting curves are shown as
    dotted lines. If there had been no change in trans-
    portation costs, the  breakeven point  between
    western and eastern coal would be at D" (based on
    intersection point C"). Point D" is further to the
    East than the original breakeven point D, indicat-
    ing that Tide IV would cause western coal's mar-
    ket share to gain relative to eastern coal.

    As discussed above, however, a change in trans-
    portation costs would cause part of the market
    share change to take place in the baseline. In
    Exhibit  C-6, this  baseline shift is shown as the
    distance from D to D1. It might seem, then, that
    instead of Title IV causing an incremental market
    share change from D to D", it would cause only
    the small shift from D1 to D".
Exhibit C-6. Combined Effects of Transportation Rates and Allowance Price on Coal Competition
 West

Source: ICF analysis.
D   D'D'
East
5 In Ms simplified analysis, no distinction is made between the eastern and western coals in terms of heat content per ion. If the
heat content per ton of lower sulfur western coal is lower than that of the coal it is replacing, this difference would have to be
reflected by expressing the costs of purchasing and transporting the coals in terms of equivalent amounts of energy: point A would
be somewhat higher, and the slopes of lines A-C" and A-C"1 would be somewhat steeper. The main point of the graph'would not
be affected.

                                                C-8

-------
\
 The effects  of the  transportation cost change,
 though,  interact with the effects of allowance
 prices to cause an additional increase in the mar-
 ket share of western coal. With lower transporta-
 tion costs, the effect of a given allowance price,
 and hence a given upward shift in the  cost of
 using eastern high-sulfur coal (shown by the shift
 from B to B") would be to move the breakeven
 point from D' to D"'(based on  intersection point
 C"'). This eastward shift in  the  zone in which
 western  coal is competitive  is potentially large
 enough to outweigh the baseline effect noted in
 the paragraph above. In other words, even though
 the transportation cost reduction led to a baseline
 shift in western coal's market share that reduced
 the incremental  effects of Title IV, the net effect
 of Title  IV on  western  coal use could  still be
 greater with lower transportation costs. In the dia-
 gram,  lower transportation costs would increase
 Title FV's net effect so long as the distance from
 D' to D"1 exceeds the distance from D to D'.

 The net effect of lower transportation costs on
 Title TV's impact on western  coals depends on a
 number of factors. One crucial factor, not shown
 in Exhibit C-6, is the extent to which the baseline
 shift in western coal use causes allowance prices
 to be  lower (than if the  baseline shift had not
 occurred). As noted above, Exhibit C-6 was con-
 structed as though allowance prices, and thus the
 shift from B  to  B", would be the same whether
 transportation costs were high or low. In fact, the
 baseline shift to greater use of western low-sulfur
 coal could reduce the  price of  allowances  by
 reducing the need for incremental SO2 reductions.
 Though  it would be cumbersome  to  illustrate,
 lowering B" would shift D'" to the left, showing a
 smaller net effect of Title IV on western coal use.
 In the extreme, if the baseline increase in western
 coal due to transportation cost reductions (D to
D') were large enough to accomplish all of Title
 IV's goals, the allowance price would drop to zero
 (because mere would be no need for, and thus no
 value in, further SO2 reductions). In that case, the
effects of Title  TV on western coal use would
clearly be smaller as a result of the transportation
cost change.

In actuality, emissions consequences of the base-
                                                               line shift in coal use was apparently only a frac-
                                                               tion of the required emission reduction in Phase I
                                                               of Title  IV  and manifestly did not drive  the
                                                               allowance price to zero (Ellerman et al., 1997). It
                                                               is,  therefore,  difficult to tell a  priori  whether
                                                               transportation  cost  reductions  increased  or
                                                               decreased the effects of Title IV on western coal
                                                               use. It  does  not  seem unreasonable  that  the
                                                               CEUM and IPM modeling results suggest that the
                                                               effects on western coal use increased.

                                                               Another consequence of the increasing attractive-
                                                               ness of western coal, both in the baseline and as a
                                                               compliance strategy, is that  it would definitely
                                                               reduce the need for coal shifting within  the East.
                                                               .In the earlier analysis, much of the needed reduc-
                                                               tion in sulfur had to come from shifts from high-
                                                               sulfur eastern coal to lower-sulfur eastern coal. In
                                                               part, these  changes could come within a given
                                                               region or state. To a substantial degree, however,
                                                               they also came from interregional  shifts within
                                                               the East: particularly, from high sulfur coal in  the
                                                               Midwest region to medium or low sulfur coal in
                                                               the  central and southern Appalachian region. To
                                                               the extent that the 1996 analysis projected greater
                                                               shifts  to western coal, it also projects a smaller
                                                               need  for shifts within the East. Furthermore,
                                                               because the differences in sulfur levels between
                                                               eastern coals is often not as great as between east-
                                                               ern and western coal, more total tons of coal must
                                                               be shifted in order to cut SO2  emissions  by a
                                                               given  amount. Thus,  because both the  baseline
                                                               and post-Tide IV shifts toward western coal  are
                                                               very good substitutes for  shifts between eastern
                                                               coal regions,  it is reasonable to expect that there
                                                               are fewer tons of coal demand moving among  the
                                                               eastern regions in the 1996 analysis.
C.4.   Changes  in   Employment  Impact
       Projections

The different productivity and baseline assump-
tions  used in  the  1992 RIA and  1996 analysis
result in somewhat different employment change
projections. The 1992 RIA did not quantify the
employment impacts of Title IV. It did, however,
provide a qualitative assessment, stating that there
                                                          C-9

-------
would be both a small net decline in nationwide
employment  and significant  shifts from one
region to another. Corresponding to the coal min-
ing changes shown in Exhibit C-5, the RIA pro-
jected  decreased   employment   in  northern
Appalachia and the Midwest.6

Similarly, employment gains were projected for
central  and southern Appalachia and the  West.
These employment gains, however, were  not
expected  to completely offset the employment
losses in northern Appalachia and the Midwest.
The effects of projected coal switching shown in
Exhibit C-5 on coal mining employment can be
calculated using projected labor  productivity.
Understanding the impacts on employment, how-
ever, is complicated by the fact that projections of
mining productivity have changed since the late
1980s and early  1990s.  As discussed above,
whereas the base cases prepared in 1989 (and
used in the 1992 analyses) projected coal miner
productivity increases of only two to three percent
per year, the 1996 analysis recognizes that actual
productivity has been rising twice as fast. Thus,
the employment implications of changes in coal
use projections from 1992 to 1996 depend on
which productivity assumptions are used.7

To focus on the effects of differences in coal use
       ,r
projections, we first show the projected effects on
employment holding the  productivity assump-
tions constant from the 1992 to the 1996 analysis.
Exhibit C-7 shows  the  incremental  impacts of
Title IV on coal labor requirements for the 1992
and 1996 analysis, using the labor productivity
assumptions from the 1992 analysis (shown in the
first two columns of Exhibit C-l). The projected
job slot changes in Exhibit C-7 are based on the
productivity factors  of 4.65 tons per  miner-hour
in the East and 15.27 tons per miner-hour in the
West. The differences in employment impact esti-
mates between the 1992 and 1996 analyses result
solely from differences in their projections of coal
use.
Exhibit  C-7. Labor  Demand  Comparisons  for  2010 Using  Constant Productivity Growth
Assumptions (from (he 1989 Base Cases) (numbers of job slots)
Coal Supply Region
East
Northern Appalachia
Central & Southern Appalachia
Midwest
West
Net* Change
Gross** Change (in regions with
reduced demand)
1992 RIA
( high base case)
-3,700
- 4,300
+ 14,100
- 13,500
+1,000'
- 2,700
- 17,800
1996 Analysis
-6,800
-6,100
+ 3,300
- 4,100
+ 1,600
- 5,200
- 10,100
Difference
(1996 analysis
minus 1992 RIA)
- 3,100
- 1,800
- 10,800
+ 9,400
600
-2,600
+ 7,600
  * Net is nationwide job slot change, netting the gains and losses in regions listed above.
  ** Gross is the total regional losses for those regions that have losses.

  Source: ICF analysis. Totals do not add due to rounding.
8 Because it takes fewer workers to mine a given quantity of coal in the West than in the East, shifting coal production from the
East to the West will add fewer workers in the West than it will subtract in the East.

7 As a concrete example, a drop in coal demand of 10 million tons will have twice the effect on labor demand if it took 200 workers
to produce each million tons than if it only took 100 workers.
                                              C-10

-------
Applying the lower labor productivity growth
rates used in the 1992 analysis to the coal demand
impacts projected by both the 1992 and 1996
analyses, the 1992 analysis projected a 2,700 job-
slot-loss in 2010, compared to the 1996 analysis's
5,200 job slots. However, from a difference per-
spective, the gross reductions in labor demand in
the regions most affected by Title IV are consid-
erably smaller (10,100  job slots  vs.  17,800)
according to the more recent analysis. This result
occurs because the 1996 analysis projected less
coal switching within the East than did the 1992
analysis.

Exhibit C-8 shows effects of combining the more
recent  coal demand estimates  with newer labor
productivity projections. The net effects of Title
IV are projected to be greater by only about 2,700
job slots. Gross employment impacts of Title IV
are now projected to be substantially smaller than
previously projected. The 1992 RIA analysis pro-
jected (implicitly) that almost 18,000 job slots
would be eliminated by 2010; now, projections
using the 1996 analysis are that less than half that
number—7,700—will  be lost. This  change in
gross job slot losses results from a dramatic drop
in projected job slot losses in the Midwest (from
over 13,500 down to about 3,000). As discussed
above, the increased shift to western coal was in
place of shifts to eastern low-sulfur coal
Exhibit C-8. Labor Demand Comparisons for 2010 Using Productivity Growth Assumptions from
the 1989 Base Cases for the 1992 Analysis, and Productivity Growth Assumptions from the 1996
Base Case for the 1996 Analysis (numbers of job slots)
Coal Supply Region
East
Northern Appalacnia
Central & Southern Appalacnia
Midwest
West
Net* Change
Gross** Change
(in regions with reduced
demand)
1992 RIA
(high-base case)
-3,700
^,300
+14,100
-13,500
-1-1,000
-2,700
-17,800
1996 Analysis
-5,100
-4,600
+2,500
-3,100
+1,000
-4, 100
-7,700
Difference
(1996 analysis minus
1992 RIA)
-1,500
-300
-11,600
+10,400
0
-1,500
+10,100
   * Net is nationwide job slot change, netting the gains and losses in regions listed above.
   ** Gross is the total regional losses for those regions that have losses.

   Source: ICF analysis. Totals do not add due to rounding.
                                             C-11

-------
                                       C.5.  References

Ellerman, A. Denny, Schmalensee, Richard, Joskow, Paul L., Montero, Juan Pablo, and Bailey. Elizabeth M. (1997).
Emissions Trading Under die U.S. Acid Rain Program: Evaluation of Compliance Costs and Allowance Market
Performance. MIT Center for Energy and Environmental Policy Research, Cambridge, MA, October, 1997.

ICF Consulting (1993). Memo from ICF Resources to Ann Watkins at EPA on base case revisions. August 20,1993.

ICF Consulting (1995). ICF briefing. September 1995.

Science Advisory Board (SAB) (1996). Minutes Of The Advisory Council Of Clean Air Compliance Analysis.
Attachment F-5-5 "Coal Supply Assumptions in the Clean Air Power Initiative,'' July 31,1996. Referenced at
http://www.epa.gov/sciencel/minut4.hbn

U.S. Department of Energy, Energy Information Administration (E1A) (199S). Energy Policy Act Transportation Rate
Study: Interim Report on Coal Transportation. October 1995, p. 21.

U.S. Department of Energy, Energy Information Administration (EIA) (1999 and 1993). Annual Energy Review
"Monthly Cost and Quality of Fuels for Electric Plants". Data from July 1999 and July 1993.
http://www,eia.doe.gov/cneaf/electritity/page/ferc423.htrnl
                                           C-12

-------
                 Appendix  D:  Results of  Modeling and
                                     Analysis for  2000
\
           As noted in the body  of this
           report, the analysis focuses  on
           2010 because the full effects of
           Title IV  are not expected to be
           seen until about that year. For pur-
           poses of comparison to  analyses
           that discuss Title TV's impacts in
           2000, and to show how projec-
           tions of coal and labor demands
           change over time, it may be useful
           to see projections for 2000 based
           on both  the 1990/1992  analysis
           and the  1996  analysis.  This
           appendix consists of four tables:
           Exhibits D-l and D-3 show coal
           use and employment projections
           based on the 19% IPM runs, and
           Exhibits D-2 and D-4 show simi-
           lar projections based on the earli-
           er 1990/1992 analysis.
          Tables D-l and D-2 show generally comparable
          coal shifts from the East to the West, with very lit-
          Exhibit D-2. Projections of Coal Industry Output in the Year
          2000. (Using 1990/1992 model runs) (millions of tons)
Exhibit D-1. Projections of Coal Use by Utilities in the Year
2000. (using 1996 IPM runs) (millions ol tons)
Coal Supply Region
East
Northern Appalachia
Central & Southern
Appalachia
Midwest
West
Total
Utility Coal
Use in
Baseline
428
170
164
94
524
952
Utility
Coal Use
with Title
rv
403
139
188
76
549
952
Title IV
Impact
-25
-31
24
-18
25
0
 Source: 1996 IPM runs. Totals do not sum due to rounding.

Coal Supply Region
•
East
Northern Appalachia
Central & Southern
Appalachia
Midwest
West
Total
Coal
Production
in Baseline

648
197
302

149
491
1,138
Coal
Production
with Title
IV
615
166
364

85
521
1,134
Title IV
Impact

-33
-31
62

-64
30
-4
             tie net change in total tons of coal either produced
             or used by utilities. The more recent analysis,
             however, shows a much smaller shift within the
             East. In 1996, utilities were projected to increase
             their year 2000 use  of central  and southern
                         Appalachian coal by 24 million
                         tons,  and to decrease their use of
                         Midwest coal by 18 million tons.
                         In 1990/1992, the equivalent fig-
                         ures were an increase of 62 million
                         tons  for  central and  southern
                         Appalachian coal, and a drop of 64
                         million tons in the Midwest.
          Source: 1992 SOg RIA. Appendix 4b (high base case). Totals do not sum due to
          rounding.
                          These different projections of
                          interregional   coal   shifts   are
                          reflected in the employment pro-
                          jections: Exhibit D-3  shows a
                          much smaller  shift  of job  slots
                          between the Midwest and central
                          and southern  Appalachia  than
                          does  Exhibit D-4.  Largely  as a
                          result of this difference, Exhibit
                                                     D-1

-------
D-3 shows a smaller gross and net reduction in
job slots.1 These patterns are similar to those seen
in  the analyses for the year 2010, with smaller
shifts within the East leading to smaller gross job
losses.

Exhibit D-4. Projected Effects of Title IV on
Coal Industry Employment in the Year 2000.
(Using 1990/19% model runs)
Coal Producing Region
East
Northern Appalachia
Central & Southern
Appalachia
Midwest
West
Net* Job Slot Change
Gross** Job Slot Change:
Change in Job
Slots
-4,000
-3,700
7,400
-7,700
700
-3,200
- 11,400
* Net is nationwide job slot change, netting the gains and
losses in regions listed above.
** Gross is the total regional losses for those regions that
have losses.

Source: 1992 SO2 RIA, Appendix 4b (high base case).
Totals do not sum due to rounding.
Exhibit D-3. Projected Effects of Title IV on
Coal Industry Employment in the Year 2000
(Using 1996 IPM runs)
Coal Producing Region
East
Northern Appalachia
Centra! & Southern
Appalachia
Midwest
West
Net* Job Slot Change
Gross** Job Slot Change •
Change in Job
Slots
-2,800
-3,500
2,700
-2,000
600
-2,300
-5,600
                                                         • Net is nationwide job slot change, netting the gains and
                                                         losses in regions listed above.
                                                         ** Gross is the total regional losses for those regions that
                                                         have losses.

                                                         Source: ICF analysis, using 1996IPM runs. Totals do not
                                                         sum due to rounding.
1 Here and elsewhere in this report, "gross job slot changes' are the sums of job slot reductions across all of the regions with loss-
es, whereas "net job slot changes' are the sums of job slot reductions across all regions, net of job slot increases.
                                                    D-2

-------