University of Florida Digital Collections Home
UFDC Home myUFDC Home  |   Help 
<%BANNER%>

Essays in Renewable Energy and Emissions Trading

University of Florida Institutional Repository
Permanent Link: http://ufdc.ufl.edu/UFE0022518/00001

Material Information

Title: Essays in Renewable Energy and Emissions Trading
Physical Description: 1 online resource (216 p.)
Language: english
Creator: Kneifel, Joshua
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2008

Subjects

Subjects / Keywords: electricity, emissions, energy, environment, environmental, pollution, renewable, utility
Economics -- Dissertations, Academic -- UF
Genre: Economics thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Environmental issues have become a key political issue over the past forty years and has resulted in the enactment of many different environmental policies. The three essays in this dissertation add to the literature of renewable energy policies and sulfur dioxide emissions trading. The first essay ascertains which state policies are accelerating deployment of non-hydropower renewable electricity generation capacity into a states electric power industry. As would be expected, policies that lead to significant increases in actual renewable capacity in that state either set a Renewables Portfolio Standard with a certain level of required renewable capacity or use Clean Energy Funds to directly fund utility-scale renewable capacity construction. A surprising result is that Required Green Power Options, a policy that merely requires all utilities in a state to offer the option for consumers to purchase renewable energy at a premium rate, has a sizable impact on non-hydro renewable capacity in that state. The second essay studies the theoretical impacts fuel contract constraints have on a electricity generating unit's compliance costs of meeting the emissions compliance restrictions set by Phase I of the Title IV SO2 Emissions Trading Program. Fuel contract constraints restrict a utility?s degrees of freedom in coal purchasing options, which can lead to the use of a more expensive compliance option and higher compliance costs. The third essay analytically and empirically shows how fuel contract constraints impact the emissions allowance market and total electric power industry compliance costs. This paper uses generating unit-level simulations to replicate results from previous studies and show that fuel contracts appear to explain a large portion (65%) of the previously unexplained compliance cost simulations. Also, my study considers a more appropriate plant-level decisions for compliance choices by analytically analyzing the plant-level decision-making process to show how cost-minimization at the more complex plant-level may deviate from cost-minimization at the generating unit level.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Joshua Kneifel.
Thesis: Thesis (Ph.D.)--University of Florida, 2008.
Local: Adviser: Kenny, Lawrence W.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2008
System ID: UFE0022518:00001

Permanent Link: http://ufdc.ufl.edu/UFE0022518/00001

Material Information

Title: Essays in Renewable Energy and Emissions Trading
Physical Description: 1 online resource (216 p.)
Language: english
Creator: Kneifel, Joshua
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2008

Subjects

Subjects / Keywords: electricity, emissions, energy, environment, environmental, pollution, renewable, utility
Economics -- Dissertations, Academic -- UF
Genre: Economics thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Environmental issues have become a key political issue over the past forty years and has resulted in the enactment of many different environmental policies. The three essays in this dissertation add to the literature of renewable energy policies and sulfur dioxide emissions trading. The first essay ascertains which state policies are accelerating deployment of non-hydropower renewable electricity generation capacity into a states electric power industry. As would be expected, policies that lead to significant increases in actual renewable capacity in that state either set a Renewables Portfolio Standard with a certain level of required renewable capacity or use Clean Energy Funds to directly fund utility-scale renewable capacity construction. A surprising result is that Required Green Power Options, a policy that merely requires all utilities in a state to offer the option for consumers to purchase renewable energy at a premium rate, has a sizable impact on non-hydro renewable capacity in that state. The second essay studies the theoretical impacts fuel contract constraints have on a electricity generating unit's compliance costs of meeting the emissions compliance restrictions set by Phase I of the Title IV SO2 Emissions Trading Program. Fuel contract constraints restrict a utility?s degrees of freedom in coal purchasing options, which can lead to the use of a more expensive compliance option and higher compliance costs. The third essay analytically and empirically shows how fuel contract constraints impact the emissions allowance market and total electric power industry compliance costs. This paper uses generating unit-level simulations to replicate results from previous studies and show that fuel contracts appear to explain a large portion (65%) of the previously unexplained compliance cost simulations. Also, my study considers a more appropriate plant-level decisions for compliance choices by analytically analyzing the plant-level decision-making process to show how cost-minimization at the more complex plant-level may deviate from cost-minimization at the generating unit level.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Joshua Kneifel.
Thesis: Thesis (Ph.D.)--University of Florida, 2008.
Local: Adviser: Kenny, Lawrence W.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2008
System ID: UFE0022518:00001


This item has the following downloads:


Full Text
xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID E20101212_AAAAAX INGEST_TIME 2010-12-12T13:13:05Z PACKAGE UFE0022518_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES
FILE SIZE 58180 DFID F20101212_AAATQE ORIGIN DEPOSITOR PATH kneifel_j_Page_093.pro GLOBAL false PRESERVATION BIT MESSAGE_DIGEST ALGORITHM MD5
eeb2f49f4c56cbc4c31d59b8c45e81eb
SHA-1
7144c00d4a6dc2f4b5d4c716f2efb51aa27f36ef
48610 F20101212_AAATPQ kneifel_j_Page_076.pro
29af2809fedf8b8ac1cc842a42c3504a
2f48adb11dadfdd4d2a42fca1e99f64efefbafa3
23922 F20101212_AAASMO kneifel_j_Page_148.QC.jpg
0c0b408e3e6ebd5b3eb449c0abb744bd
33e6694905e10f7dd062ff9578235f9358a3bf13
2304 F20101212_AAASNC kneifel_j_Page_050.txt
768187242152b2b1ff30de21249080fc
92563c7cb38f80d223aee8521dcda4dc7ec39790
41351 F20101212_AAATQF kneifel_j_Page_094.pro
2afee06896be280194e7003290be4b22
f42cd4ea8934ec849143be8b8efcbe7155ae8343
54822 F20101212_AAATPR kneifel_j_Page_077.pro
9467cd0bef47085625d524e51542ee0e
178b8fad4bcf585b429dbbd7237e990ebc55b0ef
19987 F20101212_AAASMP kneifel_j_Page_012.QC.jpg
5dd19cb697d6a82316abce601b9ef794
2aa75409cf304ec26344d63ac8d7093ceb912e50
86207 F20101212_AAASND kneifel_j_Page_214.jpg
7696879311dbd9d90f6081746a27f590
9368017fd82a90eca9ad115abccca320eb5fa631
49272 F20101212_AAATQG kneifel_j_Page_095.pro
ffbef97ce6f7e7b978907ee2b054d183
1b7af10a5eb9a7780a304c2c79a701c56964f7b2
57863 F20101212_AAATPS kneifel_j_Page_078.pro
c2abcc99e9d82dbc955ed8d07bb6e2e5
8236b8a2a6018fdb62dd5d764b437dd4ed0bdb46
5852 F20101212_AAASMQ kneifel_j_Page_106thm.jpg
4b9b576463969609d1937fc8422c80f6
964710705c305971109abb892c6d533de2d54e61
44296 F20101212_AAASNE kneifel_j_Page_009.pro
907774cd7e3d3d6ec30bd731acfd6d3c
dca442e116bac9b21383867f06039374790c5278
44842 F20101212_AAATQH kneifel_j_Page_096.pro
9667b0b50eb770791299992105d99e5c
5b0b370d97125c9095c255fe93c9c588f22e044e
56415 F20101212_AAATPT kneifel_j_Page_079.pro
542dabf3d4573319e597b941789e07aa
c1e1f7bf5e26e377efd7fd43b94aca8d1ba9ef54
91911 F20101212_AAASMR kneifel_j_Page_140.jpg
4d9c874908c3d03062ad725241272067
e20300e900a0121e005f2a0f906b504e8fc872c5
1051954 F20101212_AAASNF kneifel_j_Page_078.jp2
20f69d0f31a1ab5f2fda9590e1b13170
300510fe2614b4a3cf54a345c714c092338c2d4c
53437 F20101212_AAATQI kneifel_j_Page_098.pro
506e2faee443740a00cdb06ce738f74a
94fb6f367027ba686d66533f90e0d13f34f167b5
53597 F20101212_AAATPU kneifel_j_Page_080.pro
dfd94970419686f232e5c9e174839e42
1121de07411898bbf9bf184b3ddf937718a51dbd
25271604 F20101212_AAASMS kneifel_j_Page_069.tif
ff48c5784d74e9973c2a20f688d920ed
fc0c0bf835eac8abf1d9e3093833a99d953109f8
107583 F20101212_AAASNG kneifel_j_Page_176.jp2
d44a541bed02343da2884d0bb4babc56
4c5dc34630619e40ea8969b241d6df78ccf8b627
64553 F20101212_AAATQJ kneifel_j_Page_100.pro
e5939e8032bd356461ebf07a7ca4044d
9cd72940b4689cc448f52740066be3b0ca95a9ad
42247 F20101212_AAATPV kneifel_j_Page_081.pro
6f9daa4c5043a85f100345d6da7c1092
711a3907d6fe692d40da7d71f5bfb84c6dca32a1
44939 F20101212_AAASMT kneifel_j_Page_106.pro
6d9b2b36516722a472e855175b2cc691
d4a9685cef3dfbc63cfb6cb2cb0882477753123d
270120 F20101212_AAASNH kneifel_j_Page_197.jp2
5c3b24a9fc09f62391b4e2e9a2d3d121
c831f4e42985f66c036b4a7850ba1dc8763aa420
37846 F20101212_AAATQK kneifel_j_Page_101.pro
7a266901c43ee4696d1244d71df36611
5e58cc2d044572bd2e0d7b3862c62825b02f21f3
42127 F20101212_AAATPW kneifel_j_Page_082.pro
ed49bcde8f983a9f72f1b690a9fc12fe
656f962fb7b08513be461f3f2cd6c7dc2ff8cfb2
56161 F20101212_AAASNI kneifel_j_Page_177.pro
69917fd1872c5ca264dc3bfd9a7dd4e0
6c18f7f41560592ca37ce0625316e0e76ec992e4
F20101212_AAASMU kneifel_j_Page_107.tif
cae2631826beb369c943774c96c21662
af7c7d631a0e685349307f50bdbca335d9f34690
57167 F20101212_AAATRA kneifel_j_Page_127.pro
2ddabb2ff5bbdf57606fd30687103c35
9087f98d4a43b86a6eca514dc21631e50ad05d79
45388 F20101212_AAATQL kneifel_j_Page_102.pro
c783d1d478494a648342b4bf101f0142
00a0ad45def8bd42c3cf82a04ad7d3b94c8b9bcb
56691 F20101212_AAATPX kneifel_j_Page_083.pro
b977112f1034b6410f29a3b61246f720
0ded006022bf4f3025fa5a02db7ac6df8f723b2d
2583 F20101212_AAASNJ kneifel_j_Page_020.txt
5738c21e54ebfdd202114b375f856fb3
1a6fec150c6d52f64bce084c352179638ad426eb
F20101212_AAASMV kneifel_j_Page_101.tif
1edf5a46c01f1e0f395e7d2dcde48623
f969b7c86a3a609126d9656c08705f2ae3afb9ad
49767 F20101212_AAATRB kneifel_j_Page_128.pro
ea29d258c0b8326f9567e4e7492818aa
42e332942cef04a4413b068b67c2793680bd7e01
45779 F20101212_AAATQM kneifel_j_Page_104.pro
21891a382251d2afbcca1e7e733f0e1b
bb3adafb406be9691e99cb5539bfda5d7eeef5c2
60157 F20101212_AAATPY kneifel_j_Page_084.pro
0ab7bf4b569ea48f8f3a8fa1d8733167
20e08761e9b66f32692787bd1af77af1ada40820
11166 F20101212_AAASNK kneifel_j_Page_196.QC.jpg
eff88b0eb00b0762e689c94f69e024a8
0b844f421a7afb7c04322c82a7d980a4b0c554af
4293 F20101212_AAASMW kneifel_j_Page_114thm.jpg
2e72b4d3f69f7061986274fbdd9f6b13
d8c18520b83367ff2fa0ae3a53c5a9413945745d
48885 F20101212_AAATRC kneifel_j_Page_129.pro
2e502c56d1e830bb2b990e8a7c6b3a01
ce0756a4863e7a84223f1d265d4f38cfca3ae8e9
46312 F20101212_AAATQN kneifel_j_Page_105.pro
5e87eac3807572cf34d8bcd52c91425d
a13706be1eaff4c19ec9dbd36b7edd373af66e5b
54586 F20101212_AAATPZ kneifel_j_Page_085.pro
5d14f6ea7ab25605aa9455a4f37626b6
5971cb36772cb3b08367ef12421830178815aff1
123946 F20101212_AAASOA kneifel_j_Page_111.jp2
e503fb4899c79fc5091beffff2893039
8501aa2b57e29f9729489ae9f23516c02957ce30
913 F20101212_AAASNL kneifel_j_Page_002.pro
4bb27119ea2db114a6915b665bef5430
69bf5fdd9d1278bac06091db0019f43c1d36e5bd
65228 F20101212_AAASMX kneifel_j_Page_187.jpg
a8d8fec0a558de4a276f1013ccc32289
a36d348019e89e840bd338aede483e5229cedddf
58318 F20101212_AAATRD kneifel_j_Page_130.pro
f81dc4d9305566d97d38f507c18d30c3
bc199f2b000974755e426c0ddee946b4c8f79713
64626 F20101212_AAATQO kneifel_j_Page_108.pro
34183876f35932e55a645296d2fa9c67
6d554b600f33249a09211322a829bfb31cf3160a
20956 F20101212_AAASOB kneifel_j_Page_056.QC.jpg
fab58e58c33d7aa840ae0b56c90b1e1b
c28604e6c55f33e130ffa6ff902753c7968a2ad3
F20101212_AAASNM kneifel_j_Page_035.tif
9811371c5a7eff4593f22eaa00a25e38
a6a485642d9f47a961e3cb1faba0111bdc4bfaeb
1053954 F20101212_AAASMY kneifel_j_Page_136.tif
e179f99910f4b3c7b0693d592e6b4631
824645dc9b6b1dbe5cc0858688b1b26f146debed
56755 F20101212_AAATQP kneifel_j_Page_109.pro
58817d28fc1b93d7790a061776a5bd96
7e8ed80d92e8eaa0b7b49fa87feaa7cbd2105271
8995 F20101212_AAASNN kneifel_j_Page_124.QC.jpg
24f16b895305f77eb0ee3e48a946214b
dcd0aa256d49fa6c15c4585fa9462c7ef243d91f
F20101212_AAASMZ kneifel_j_Page_007.tif
a339d635a07a694da0c124cb7101b8d9
8ab3acfb261f8a8cf5b7607c43199f9040e13d3c
54122 F20101212_AAATRE kneifel_j_Page_131.pro
012ec1f1cd080c4f9af2b45444ea591a
d751d6e786dca936a30843629dfcde9078d9dc87
54101 F20101212_AAATQQ kneifel_j_Page_110.pro
8d6acd46781c480430a543247feb79fc
628fe7b7fcce70bf7b7d081f792f934360c9e520
67374 F20101212_AAASOC kneifel_j_Page_200.jpg
8c19d87a3962dafc0d6f77d64a492367
f8da65f5e17224cdcac52e8f20ac318a4f1686d3
6284 F20101212_AAASNO kneifel_j_Page_127thm.jpg
ce865d09418c8cbf47f2b3e4c50c0066
409e1845236bedccc474fe519cd6763c50ffa4f5
57504 F20101212_AAATRF kneifel_j_Page_132.pro
2616aac1d17a768691c31f208591a3e4
0e1b0df86ed9a42e9d606a465f751fbb237af218
61128 F20101212_AAATQR kneifel_j_Page_111.pro
34b59f99a1492bf6e994f02351cfbaf4
0bfa49f3e92990dfb43fd2d7fcd92a609444a95e
2229 F20101212_AAASOD kneifel_j_Page_170.txt
9fdfa32ed349e1e19e231b524c62d356
16095ca5707cd777386dbd1f6708c7ea57986f8e
50849 F20101212_AAASNP kneifel_j_Page_040.pro
11570654d5c6546393b2ec43f3acf5cc
53f49d43ce770301861658a9778a003e5e9cda12
54428 F20101212_AAATRG kneifel_j_Page_134.pro
28d9c726a63a2cf20b36161d1732f015
7dd83d609d27e7c1b0ae5e74f21c8300dac8b38b
31663 F20101212_AAATQS kneifel_j_Page_114.pro
08b36bda3b8f5865d4a84566cd9de240
52e2e5ae9f3c8f44b6634a1082406bc4fa8f9783
1348 F20101212_AAASOE kneifel_j_Page_002thm.jpg
7b558fe6a4db3c41c0ebada554a0cc12
d001a8c574613949eaced2f60bc05ba7cfc59121
40299 F20101212_AAASNQ kneifel_j_Page_179.pro
f26f47cbe658d6bd163bb3080c8fa9f0
3dc6ed153b543970a38ab9e475e73d6d715099cd
54700 F20101212_AAATRH kneifel_j_Page_135.pro
cd0de4793b3b273e85429017a53abac5
fff89625235116e72eb361bb47e2e634962b9c0e
3640 F20101212_AAATQT kneifel_j_Page_117.pro
d133d71d1d5a2528fb1ac4ed08d1de2d
ad2e8e78cd083dd1ea03124a038359328da4ee26
5277 F20101212_AAASOF kneifel_j_Page_182thm.jpg
86fc98cda1b1c1947cb39b4c7e24b3a6
7e1b725a4581583b9a51c5a033c610d669976545
25736 F20101212_AAASNR kneifel_j_Page_122.pro
96d2409b076520b282b99d89d0085568
6b8fa0d176183e355d0472ade150633d3561da40
53503 F20101212_AAATRI kneifel_j_Page_137.pro
3da37d69bd912a64cf7c73fb824afa7b
f15c17202e205990e7936301768849061179b221
17230 F20101212_AAATQU kneifel_j_Page_119.pro
d9fb411b516d80fa50cffa0444a28ef3
e95b42d3fd89f9b8a77d5b03467cc6da6dc4da0b
52471 F20101212_AAASOG kneifel_j_Page_026.pro
89a9ff9a9654f6c430be91f874a01f88
a09475358302d2566cded34a632d350ea65569dd
56747 F20101212_AAASNS kneifel_j_Page_009.jpg
bc5a201360442020b5e8b523aeac7724
7be49f216916a6c8ee364626649bd13e3b4c5f2e
52425 F20101212_AAATRJ kneifel_j_Page_138.pro
9974e9eec9bcdc9bae683a3dd1deff0a
888430449a0fd412594bc7dd8afed655f0b260ec
35651 F20101212_AAATQV kneifel_j_Page_120.pro
daba5f3d3334fc0135641fde8e0a72b2
bf586698de7475a59f6dbc5af756933a542a98de
6291 F20101212_AAASOH kneifel_j_Page_090thm.jpg
f66732adba681673c7d51d1c250a5d2a
deb7fdbff0b43ed3dc808d72437919386e2a1022
56047 F20101212_AAASNT kneifel_j_Page_057.pro
8a81a07f8db83f0a8c47ce4663feefbc
10a0c836899c53af38ff4c710b35d067afa004c5
47916 F20101212_AAATRK kneifel_j_Page_141.pro
abce724a6de8779b087d8cba4db12f15
90b559f553b53ef2142dfda49922a96ad6a8ec5e
16075 F20101212_AAATQW kneifel_j_Page_123.pro
c7841ce3903165c2f20ca28c8352eeef
57c433765ad3156ae51a04c6642cc8158866cf7d
1885 F20101212_AAASOI kneifel_j_Page_120.txt
960b6d69d7b6c394ecf920ac4d96b345
a9a7670e235f475c1e4e2b99d037e6e6869f34d3
2317 F20101212_AAASNU kneifel_j_Page_070.txt
2b323e040074d9041816c4886ea9e75f
ccac65ba4d9d08e1317454d99ce4b81792f586e8
50218 F20101212_AAATSA kneifel_j_Page_162.pro
70c567e48914f20f2d60242b790f854c
eb29f8e052c1f9737e3c562c84455aeb860a4cf2
40989 F20101212_AAATRL kneifel_j_Page_142.pro
6f4a85708d734cb79ec5a9bce923e109
abb9aa1380eb790bab40974d7fe6fbead07b4fdd
7111 F20101212_AAATQX kneifel_j_Page_124.pro
2ffaf086b4688b3b8714487095ec562d
046b978860d47988846b12a29c0517ce50b0fb24
394669 F20101212_AAASOJ kneifel_j_Page_196.jp2
6b0d9f40517b935bb0ff80c68f35569d
31f648309c67be8f5ad4917cdec8a47d40f2a1fb
6141 F20101212_AAASNV kneifel_j_Page_066thm.jpg
76ca9d59c54be586e8420131c4c21293
62667adfe1875fa735319fb64f789827a141621e
32780 F20101212_AAATSB kneifel_j_Page_163.pro
68ebdbab2db9c856f5983a130aa29df3
fe35b8fe266ae5f5e743a429186cf34f679e1b2c
54386 F20101212_AAATRM kneifel_j_Page_144.pro
c2b1eed16c88bb69282121707178c1d2
dd249e19ea97d150b3c7b3d59072f392cacc5fdd
11464 F20101212_AAATQY kneifel_j_Page_125.pro
83742ed57970423d52e6fc2c07d84bda
d032bb42d9f7b10f6d84afa7365198483484071b
F20101212_AAASOK kneifel_j_Page_014.tif
7b274d7ada1c7d66dab867a361506e0a
f1c441865f1afc3cbc7a659a8d1d7bff03a23f8f
22674 F20101212_AAASNW kneifel_j_Page_173.QC.jpg
6280a171529bad1c6ed9c81eb9d6ae9d
bab70a039571d914d77c2ac128f0cf2178a87043
40048 F20101212_AAATSC kneifel_j_Page_164.pro
3dfb25ea028b71999945064a436a9a79
fce4e323ffe1db93463a4de1d260955b4b4cb797
47394 F20101212_AAATRN kneifel_j_Page_145.pro
bc9d94469d06ab2f65c3dbc1df69fd21
50a65e4bcb8833cc53cbc15f9742a993e2e0e3f7
4896 F20101212_AAATQZ kneifel_j_Page_126.pro
8c6855a84e4c7f27150e382f91a6b619
349d7cf9a392f639b39628dfdda4e8d5776e6592
5440 F20101212_AAASOL kneifel_j_Page_002.jp2
c7afca1906a5145bb0034ac54c7ecefd
d0877fa8b2604455101ed961c1ca32700b2a4ca6
F20101212_AAASNX kneifel_j_Page_034.tif
3f3496db8a7d09f1ac300165a1ced5c0
3696427cc58f65d9601d4a754339e8777472ea0a
F20101212_AAASPA kneifel_j_Page_064.tif
933f93446d90db15adbc0e0db2f9aa18
71db715709f164dbe87be30a77ebea96d2d1515e
46597 F20101212_AAATSD kneifel_j_Page_165.pro
3ec19d4efa5a4416d047e2844148b507
ee2524171991ad806c5d86539c20c2925c26b833
36051 F20101212_AAATRO kneifel_j_Page_146.pro
826a82e1ae6100d55565e9a889dddf10
0ecb5bec090111ec93128e6c0f18275b9c8214df
65348 F20101212_AAASOM kneifel_j_Page_071.jpg
5a469d5615da566ce2855003b5ebb508
f19b968f0edca710749a5628de0943c9441a834e
23752 F20101212_AAASNY kneifel_j_Page_162.QC.jpg
be2ddbc4a31e992d28afe8f1a67978a9
177f0bc1aeeeb7ba5a5f63ca455a820a39332518
83968 F20101212_AAASPB kneifel_j_Page_057.jpg
e99e50561ae6bb7d576c61dfb5353f39
1c353dfb3297ac27546c92d2ebd1872e8255fe6e
52688 F20101212_AAATSE kneifel_j_Page_166.pro
82baf78fea3193ad45e1d7189e5cc605
3570bc716076dbad22b9b2b12a57960af6bdb7f7
49764 F20101212_AAATRP kneifel_j_Page_147.pro
959c56fc6863df515cdd25bce3ed5c2c
0ef7d35424fd608419086d31ae5b62e86ce7929f
F20101212_AAASON kneifel_j_Page_079.tif
5b4fcea89b1a69c259dce36d8b2c4e95
f126240376e95f29c1ccb9190c346c5aa3e208a3
45215 F20101212_AAASNZ kneifel_j_Page_063.pro
e64114680cdbe0e76abca82e2e2da181
ffba79539f5c4ec119a4e8249ad3915ab57d8e0a
2960 F20101212_AAASPC kneifel_j_Page_191thm.jpg
6c0ec2181da3eec42109e68e384770fb
86c42debe6bee65def2f311d9a115d57fab3fc36
57240 F20101212_AAATRQ kneifel_j_Page_148.pro
48d8a43c6c42d763740746bbb9761d43
a2f07e185d421378e5b79de9b953c3da4be89039
F20101212_AAASOO kneifel_j_Page_110.tif
9496e9f7f3be6628f46054d093233666
9a292484c7c21b0a0fa25c6a1b624371ab438c39
49306 F20101212_AAATSF kneifel_j_Page_167.pro
d0f59d54dab4ea2e598ba0c87a7f0ff5
342cf1ad7007356401f18e64e41e9b289c812b34
44214 F20101212_AAATRR kneifel_j_Page_151.pro
01c3f4963d099b433ee95e14c6631ca4
d5203661066f7c7b6459367d25b1f5eeb2c36a1a
F20101212_AAASOP kneifel_j_Page_181.tif
0e2e80166397ca26030360946a694b76
b6648ae41527f5da93662dea27dcec3a1cca7385
73329 F20101212_AAASPD kneifel_j_Page_165.jpg
0e9fe8f7c1eb7c47943f874e51881ac2
569eb64b251b927da51297907e9960d846ca1e2a
44860 F20101212_AAATSG kneifel_j_Page_169.pro
f82c47c5e64cfafca9597d568d4f554c
0b824b77095e479f8a6cab43d7a9f3e3eb979467
53347 F20101212_AAATRS kneifel_j_Page_152.pro
89a22588ba588ae2eeb5196744121b9a
6bf611ac5d44022fbf66b6efcf10ebc7224d5a3f
45905 F20101212_AAASOQ kneifel_j_Page_064.pro
2c1ef5f66cbc47d65569b915b67d199d
75f5b7903e125dc75d5c23c8852c53c7aa520992
1051970 F20101212_AAASPE kneifel_j_Page_159.jp2
d208d46758bdb3fa396cae8d5b8965f4
0351800ab1c8a16c7f2ff889375e4be8a3699626
43405 F20101212_AAATSH kneifel_j_Page_171.pro
b4697607f626859f64bebfb816bd84f0
489e4c1d9ea9dd71025f2d048da879023ed8e470
46571 F20101212_AAATRT kneifel_j_Page_153.pro
f02f8c596af380ef65927d041bbfe291
6926910238d651df05b5154ecbbf98fae363cd17
25149 F20101212_AAASOR kneifel_j_Page_026.QC.jpg
c825cdb984eec91f3595883b34eb8fde
aa784e05e465e4b7f2b8c0aaf5dd35224a608453
99158 F20101212_AAASPF kneifel_j_Page_153.jp2
02b78a1b7dd160b349b490ce04edd695
0a803e56d4eb301b578270a25f4d2d69047cee31
42443 F20101212_AAATSI kneifel_j_Page_172.pro
06c835c67dff994934e3e63e8cfa80a0
9728f8c234531c7dca569917e2e32fc32da0674e
52830 F20101212_AAATRU kneifel_j_Page_154.pro
7d07c9ac98635f26cb3a9f0c1a1a56d3
bdd11ca055b57751b45234f6575c1427e36a0be4
F20101212_AAASOS kneifel_j_Page_010.tif
2f7e5e55e79894b131254dcf4d3bd153
1a35c5ca5041090cf287930521ffc01c00430508
F20101212_AAASPG kneifel_j_Page_197.tif
93233dde30c8b05d4902d0296ecef867
4746cd6b1ef9e55e9e4307ccdadafd9fce696dd7
55871 F20101212_AAATSJ kneifel_j_Page_174.pro
bcce804f6c3d2d4111fd5266e1dde866
42c4f6d6a4b4d353243093d06f6af4e110d3a915
60600 F20101212_AAATRV kneifel_j_Page_155.pro
5d57e6b413bda13322b5e00453b90e90
6b84fea8ef6c90eb7cfcc3529c2f8a0a32bbbc70
22691 F20101212_AAASOT kneifel_j_Page_016.QC.jpg
fac62538978a2662bf1fb0d05d6cfef2
a46fb01386f99e886b858d71fc3ce20b55c049a2
1084 F20101212_AAASPH kneifel_j_Page_216.txt
3cf832fd4029cc4c1e8b4887485f3543
c7059091245c9dd587dd554357de73ebc2242075
51272 F20101212_AAATSK kneifel_j_Page_176.pro
89d89361df96ab6787840576577fc2ca
5b7139370ae880df55e401ed345c67c89d26f0d4
60236 F20101212_AAATRW kneifel_j_Page_156.pro
83648463a9629c017199e70eb67b0910
8d7774bf7b6bc647ea5afb6b329c3b05b8337998
80241 F20101212_AAASOU kneifel_j_Page_201.jp2
bd28edbf93689b18d88a658d789b63b0
165b940ae528d72e425bf5108ccb70b18475767b
F20101212_AAASPI kneifel_j_Page_134.tif
8ed35437c2bc61cde9129c6789aedb4a
37d473dccca90ded87405381c0c0ad07c6a05132
42524 F20101212_AAATTA kneifel_j_Page_200.pro
c9445a0ef794f6f1ae11e03a60c1dd1a
e8c81c36863cfb9fdf35d9798378864cf0272996
57516 F20101212_AAATSL kneifel_j_Page_178.pro
4834e696ed6b85c2171595600a5408e8
117a063001a2b6e5ad4fd3298b14018cdb09871b
55801 F20101212_AAATRX kneifel_j_Page_159.pro
67c74227bbfe32a7f4c2949680e92c21
fc2cce78b8e896d4432a4d2999a0d30e3be10b57
2139 F20101212_AAASOV kneifel_j_Page_129.txt
f5d1a880562175188e121c7f5a7250dc
1e52814f993899bbb781b10cb29399485c756158
F20101212_AAASPJ kneifel_j_Page_189.tif
bab781c6647e14300a1a351f3c29b3c0
c584ee644d2156212b9e371e0ca51b1e5cd435dd
50688 F20101212_AAATTB kneifel_j_Page_202.pro
9b262ad36ecbdf637e8f54b1487fc02a
f7c17e62dd8770b72884a3211cfe167475dc895b
58782 F20101212_AAATSM kneifel_j_Page_180.pro
10804abb8c82401e612c1196519ec6d5
830cabb56955a0414dd9e29773e7a6b4b6aec8d9
50274 F20101212_AAATRY kneifel_j_Page_160.pro
681fabe92abf6fbd905db19f63e37f7b
e1005c7ac7f002fc47c87f0232beb13681736e6d
48586 F20101212_AAASOW kneifel_j_Page_091.pro
ce9076d43169d305ecaa4e5ad630cadf
22164c6d1f8218f25bc86cc339f7d9511b331482
1931 F20101212_AAASPK kneifel_j_Page_088.txt
c80651ee8fa8fa6c4bb5e40b8db3d02f
1ae9f8a313566e576b485f30d41dc1d8994756db
15925 F20101212_AAATTC kneifel_j_Page_205.pro
3232033e1752e5bbd627938c4e2a0b7f
ccfc942be6acba089aa1295df0afb74ede783488
56456 F20101212_AAATSN kneifel_j_Page_181.pro
cf3dd97f0ebe2cdefb02f4fd289a9a72
da1b4e29f60cb198aa17f50aa7bde8e3c76a7feb
52726 F20101212_AAATRZ kneifel_j_Page_161.pro
baa3a7839a2aeaa876c887ff1aa7498b
38a15663abfe0af9227e815ff79ade75aa6b58c8
2232 F20101212_AAASOX kneifel_j_Page_029.txt
31a2d92b594615b6f579ba8466e356de
6e805547e50dfd380ae5d89ddd92dbc9992b89bd
24842 F20101212_AAASQA kneifel_j_Page_167.QC.jpg
7ba34ec1646c74383d76b82ef49f1ef8
59db982d6160769f68c1d720a39f8ddb09d89c2f
28835 F20101212_AAASPL kneifel_j_Page_215.jpg
c659717e412ba969608d633713da8018
a749e0632ac4b193794e40a0077cfebbe72521e3
53709 F20101212_AAATTD kneifel_j_Page_207.pro
bf97c3b9571abf625e1bf96aba073608
fd8f2b42891fd029b710c828eb69e897506c741a
36304 F20101212_AAATSO kneifel_j_Page_183.pro
051d14dbbe0e76fbeafcddc6e4328766
dc54551c15a55981d325a5547b3baf29a9aad455
24138 F20101212_AAASOY kneifel_j_Page_090.QC.jpg
397c3728d13c117050ecd4a78c0d8892
763c83c2c180323fd96e0b4293c2dedd4bc53fd3
6681 F20101212_AAASQB kneifel_j_Page_004.QC.jpg
b4762b486ab2423835d0f36131674993
334ca02e8435edc9fde482cc5dbe96775403bf7c
25133 F20101212_AAASPM kneifel_j_Page_004.jp2
70e0e2cddb624691ef1c07eb987b551c
6c916e18cb299683c48eb4c3aec718d8dfccd63a
53712 F20101212_AAATTE kneifel_j_Page_208.pro
a85aef450f8937ecdf339b1c3e7f6607
a2623a2da33b410358f201cf5ea1a45cf408dbb8
35730 F20101212_AAATSP kneifel_j_Page_184.pro
15c3219f36709fd70fdbdc793bee3e8e
93a07ccf41465fdec3270f506135e0c0ddd80842
95113 F20101212_AAASOZ kneifel_j_Page_062.jpg
91c1b44f925e082928fd58dac18e46d9
3a0672d3e22d13134c2bea80d1b0ec6b8cb03461
56435 F20101212_AAASQC kneifel_j_Page_170.pro
9c33d06aeac27555031a376adf7e4fe6
d3cd6824b55cec07974b538d3ee06b629a750175
6625 F20101212_AAASPN kneifel_j_Page_095thm.jpg
2ee1de514847c6bd579dbaa231b1f752
22fa69c9b1e53823d85e3a196e4ef2eb6fbee349
31984 F20101212_AAATTF kneifel_j_Page_209.pro
26556e002b67c07ea4943deb103aabe2
42abef23a6e207572b2762c99057941b4f30717f
31430 F20101212_AAATSQ kneifel_j_Page_185.pro
9a493c41bb02b5df3dbe7f611b04f429
e0ac4c26dbea4f90fa38e80546da23fffce09d2f
23741 F20101212_AAASQD kneifel_j_Page_037.QC.jpg
2ec79c8d1523360f4baa1463c5a66f09
90e4702df45eb845ff6efdb05d5c75cdfa9dbf2d
F20101212_AAASPO kneifel_j_Page_026.tif
cbf11c9b638210eb7cd23551a889e953
a48a3ea791a2f402e9bab34ecb9e0aaf25efd9c4
56901 F20101212_AAATSR kneifel_j_Page_188.pro
54e5b35d6a8a51185155e787f70c4f51
663b657a555c8e6aa9d2bf65e8009e611279c61f
F20101212_AAASPP kneifel_j_Page_015.tif
638faac0d71b711a85ba272a7c5536fd
4f51350fe84f429423fff81bbfdb5e4a7b23ac8f
6785 F20101212_AAATTG kneifel_j_Page_210.pro
af2d99696159256fcd068759ae80a75a
65e8567fd05a88de72c9cbd7f765806c7930dc2e
59267 F20101212_AAATSS kneifel_j_Page_189.pro
5739c91adeceee13e65c4e09538ffa60
5f4429fe352cc281768bcc4d6fcb8bcfe26abb6c
716 F20101212_AAASQE kneifel_j_Page_013.txt
923e5b35b4bf4b29aac8fa2f18dbce6d
fc08ff3756f79402516f03af1b2e53c5a9aca04f
16019 F20101212_AAASPQ kneifel_j_Page_119.QC.jpg
409ed3f210edad7376d90682b5fdacd6
a0da7c35394f0fd95f0a7fecb0bbe33228182b4a
62555 F20101212_AAATTH kneifel_j_Page_212.pro
993921c5c3c4f2ae0c5fd3fd5d555670
a68a650f11268871b31708d616f0cbe405264c78
47507 F20101212_AAATST kneifel_j_Page_192.pro
8f52d2b69afd8184a65bd6010caa05de
1bdeab70d2447debe63474dd6af31c56e6c98e7b
71026 F20101212_AAASQF kneifel_j_Page_016.jpg
639acb45ac60571fc2f26fc77bbcce1b
49d5335479b147f583f36caa2f8f88071d818594
F20101212_AAASPR kneifel_j_Page_031.tif
b75335ba1f476665899ce2e1f2947f65
5273c33e55afe130adcdfa4025a15e5d9524aa45
64792 F20101212_AAATTI kneifel_j_Page_213.pro
2dbd323e1e17ff5a3ab845ce80c9a2d7
9639e077a22e2d3be9a03e93eb947d3c263e8775
8683 F20101212_AAATSU kneifel_j_Page_193.pro
7f06b17ce7a9f5dcd4f414b6d6ef2f83
1cf34c80aaf8c9f55ea79d42e5b1ba04846e58c3
33702 F20101212_AAASQG kneifel_j_Page_043.jp2
b7b9c1fa5256b2703f94134d96bf3bb3
d93de7fed609b6ff52fcb28a5a607da03a18c360
38930 F20101212_AAASPS kneifel_j_Page_201.pro
24a57d97da82c59d67e98e72c1390773
86851962f8c072369b47886a6a4527985d2d5535
65974 F20101212_AAATTJ kneifel_j_Page_214.pro
538b5543fd6d84b744fdde6557dc7a9a
83225771f2229783f7a78898ff35a5191748d7f8
8474 F20101212_AAATSV kneifel_j_Page_194.pro
e8ee11e2c07854fbd135b00427d76a8b
b528fa34dadb711335d140e4f695414367c31c59
6073 F20101212_AAASQH kneifel_j_Page_141thm.jpg
55bbc8ed5f4389b67be3e4695a73ecec
98c863e29e475f91160681381edf2bc2acc67879
F20101212_AAASPT kneifel_j_Page_066.tif
58766f74ab4eab8f3e3f47ef5c1b57e5
514d25b11e2b1a176cc9c9b95dfe503a3a4e0ca7
16267 F20101212_AAATTK kneifel_j_Page_215.pro
9a67e62e30f9afd6e86480b6cf536580
f4cffb2c4e288ee8d6a565d48c2c63bae337b953
8679 F20101212_AAATSW kneifel_j_Page_195.pro
b202f5c4c454f6cf7b4a9a4da6bec323
183d5e50c13929dc5af62cb33a0a3a7c83691652
2398 F20101212_AAASQI kneifel_j_Page_111.txt
7dab881d0b7ff9100e5d3fda8f33339a
10dc087ea1d6d2e62b22a1d45932b9b4f607416d
2744 F20101212_AAASPU kneifel_j_Page_023.txt
f86cfdbd2963b5e68e67857803705124
a85ca56d7df35f7658b306c78dfa4d0a2ced9351
2426 F20101212_AAATUA kneifel_j_Page_021.txt
1e588d0b982787d0f30ba9cc2aa2b54f
bdfbd570f0683303ecce5e944dcf2fc918e7cd31
93 F20101212_AAATTL kneifel_j_Page_002.txt
259e9791cc1e91371d31306b7bc582b6
ac3b407ff2381a0e26aba7ce856a1c11bd0a0b8b
11199 F20101212_AAATSX kneifel_j_Page_196.pro
cbb3325db451970f370ef7f5a52897e2
7d172915c68ef33e09ef7c6c37d3d91bc3a6a8c4
60517 F20101212_AAASQJ kneifel_j_Page_022.pro
c8f36d942d338e3d26bd6817e7445712
1283ae01e3bb06b825f26c45c8addf022949b744
50340 F20101212_AAASPV kneifel_j_Page_097.pro
38ecca5e3c278e375bf3dc8238b8dc6d
62bfcd8e90f16af9eff4257c1ac29225ca3cf0f7
2470 F20101212_AAATUB kneifel_j_Page_024.txt
572c1ca1fae5d81f59bfc960709b4dc1
334452b04a599dbe14682536d80d3f1e34ed7a3f
144 F20101212_AAATTM kneifel_j_Page_003.txt
9da281fb10a8368186ce3b9a4301f31c
44c620fdbc14dc3bd0221fdbd72bca719175ab0e
4863 F20101212_AAATSY kneifel_j_Page_197.pro
deb9642cce59a37e86caf66f94449c31
e7c3cedcd8f570fd533f9d8a9f592a032e41ff6e
5821 F20101212_AAASQK kneifel_j_Page_069thm.jpg
3ee7bb8421791e4b428312dd8ae2df45
71701de18ae0c41d985ee58e7e82437ead5400f8
86605 F20101212_AAASPW kneifel_j_Page_078.jpg
f903179139855bc9e6584d537fcbed73
5d42339eb1fa598a61a4517bb5aba74eacd0cd84
2913 F20101212_AAATUC kneifel_j_Page_025.txt
f0742f2831f5b1df02481a05a38d7f87
268eb4490470b414f43720a35ef13e50b27760d0
432 F20101212_AAATTN kneifel_j_Page_004.txt
d9f546dd7f3645a3c7a2672683901203
0d3f902ccbba7cb00ee0f034ec20b33437e2f359
6181 F20101212_AAATSZ kneifel_j_Page_198.pro
5016238eb20878e5ae85b50deab4bb84
19da5e7c2b6a61a036777120672215b0d8d409a4
2115 F20101212_AAASRA kneifel_j_Page_136.txt
42282763b663f2a77c981fb115969cd7
5ebddce22428df1087d4fa5dd57f5fd8a6658ff4
18952 F20101212_AAASQL kneifel_j_Page_187.QC.jpg
76a929a20d0bdb80d811285b502f9eeb
fa239681076250d08c16009b9d9b92f15af843b8
F20101212_AAASPX kneifel_j_Page_194.tif
806400317bee07b73d60733c8e60ef97
5cb40026ac1d9b9933fb26272a05704b3e7a86a4
2077 F20101212_AAATUD kneifel_j_Page_026.txt
fc5e92d8c5a35879b616c8a349cce6b5
a83f02fb8d3a49896110dc82b432cdb0aed0cfe0
2866 F20101212_AAATTO kneifel_j_Page_005.txt
ba506db256762de3466a925a8f988b29
18356642c78be0180118126d7a3620f5b254e385
25552 F20101212_AAASRB kneifel_j_Page_054.QC.jpg
2819643cace13ed9f1f48c98a097d872
75f63282940b92052cb7e9296962436f90353c62
16504 F20101212_AAASQM kneifel_j_Page_044.QC.jpg
59324f5583f5b424c11925fb70d6d559
bf317fd602a5697c395b548892bdab71160c74f7
1070 F20101212_AAASPY kneifel_j_Page_121.txt
b9eb216ece9b9da77a76aa769e0e70e5
534008b959b60b32a434dbb1a101a40094925183
2358 F20101212_AAATUE kneifel_j_Page_027.txt
9354eda61486ead9d5efa5521d6eb95e
6e57add9b777c958e7c2ea060fd754d3a385f526
3588 F20101212_AAATTP kneifel_j_Page_006.txt
243135b15a071ac5744fc059ecf208a1
bdc90f6f07e2a250d34a35387381ce7634297e23
106089 F20101212_AAASRC kneifel_j_Page_202.jp2
0f35957f12701f4c5f48150173566561
8b900158f5f429c521e7da86fe30caa5a686c13c
2113 F20101212_AAASQN kneifel_j_Page_090.txt
5c0ba9d3b940b89abf6e3d29c91d7cd9
b4fa7dde6d12e09a40a578fe1654dab07e821b20
F20101212_AAASPZ kneifel_j_Page_024.tif
0fb3edc49ee07e46b3d241126f08f055
4e34331eb0f71943f679bcd9ddf102223c49a074
2215 F20101212_AAATUF kneifel_j_Page_028.txt
fe584ca03c2b815466d753a0a11083f7
6d36416cc608d94ffc806eee790bad9265486057
3611 F20101212_AAATTQ kneifel_j_Page_007.txt
71a642b5f9fd6fa66bad6e8e4027653c
8a3991811236db4c7ca68a078a79a5f5f23bd550
40651 F20101212_AAASRD kneifel_j_Page_044.pro
4cba22a214ea50969f24cbb84380db57
76c26c59f656a189ced11cd13b2217db6c701787
84354 F20101212_AAASQO kneifel_j_Page_017.jpg
e529788dda81a2c33669ea927a8f41d0
de6c16832c2e98e968ee53aebcf72d786384d6a8
17028 F20101212_AAAUAA kneifel_j_Page_009.QC.jpg
7e396329c158088b1877eea48d5fd95b
ec90689754a8f29c42f367772fcf9228cdc59aa4
2509 F20101212_AAATUG kneifel_j_Page_030.txt
f584a6091467c47c8bfc29fa862d6fb7
65889d8984754db5c97054502135ab984a37aafd
2603 F20101212_AAATTR kneifel_j_Page_008.txt
272a2835c1ce246d6eed79f767cafdaa
fcad6599a989f5a36c8e9800d333ffa5e3af0a5f
60713 F20101212_AAASRE kneifel_j_Page_157.pro
576c1a7625a0376dbc5f8052a7cc38e0
73578eb95833a426857a48eb4da3706f36b07570
1051960 F20101212_AAASQP kneifel_j_Page_175.jp2
dda977161b128f57622bd8ca0b0b5e16
acafe90c0ab16e2a6a7a7ba9034c212849bfd996
4312 F20101212_AAAUAB kneifel_j_Page_009thm.jpg
5c77a2afe677ded83931f3d11b89d063
472519124d77bc9ef6dbda67dfa318360c9b6cb3
1864 F20101212_AAATTS kneifel_j_Page_009.txt
22d855c4fb60bc8e9f99d61823294b5d
b37b687a255521cd93a716f90074dcec77a3d8f0
73487 F20101212_AAASQQ kneifel_j_Page_088.jpg
304d96d62fd0bc5a344b95f440d1e1be
def81561db24b4306d8be5183152077d3a571395
5630 F20101212_AAAUAC kneifel_j_Page_010thm.jpg
c0e19bba57d9861ea17b025b0ddfaf99
117eee0694140d6437577feab8e6889fc6a8a16d
2326 F20101212_AAATUH kneifel_j_Page_032.txt
9905cb15d4b3205ceed74421308c7d8c
c9cf34ada36c6ba0ffb3412373253f012b2374b1
2324 F20101212_AAATTT kneifel_j_Page_010.txt
41715fb60ff0e336258ad26b13240065
bb92f0e3fe361e90dbd091143ea0d552e70a698d
6560 F20101212_AAASRF kneifel_j_Page_214thm.jpg
1e45936d596bb68f4f0bb9bd13c6ef5a
783fea16548b0f616ba85c2900be53ab6b143973
F20101212_AAASQR kneifel_j_Page_209.tif
ba43b7712dea09b10c538d22187e4c5e
c1df439f0eb6577178de4de9fbc2b35583fb71e1
2448 F20101212_AAAUAD kneifel_j_Page_011thm.jpg
2fe7247ba2a13d7a387ed8037c943343
b5378dfb36916e3f39b72e9b63194dd204d69cd6
2214 F20101212_AAATUI kneifel_j_Page_033.txt
b454b0858d79e04ce3587bafd531e218
2426c7fe26c39ae986941e0f49720c08e7fad3f4
668 F20101212_AAATTU kneifel_j_Page_011.txt
9eab70fa5dc960aa48f39c54e25d82f9
854aa53953def7cb3db4436b68a2a1fd11efdd60
70642 F20101212_AAASRG kneifel_j_Page_066.jpg
0e3467dcf3170ba41c2dbf7fb18d3c58
e239cb1be04cf9ea3d0b18713d01ff9fea07e024
8423998 F20101212_AAASQS kneifel_j_Page_195.tif
711091a5f47fad88d6a9ecd21b2ebccd
e2685b546b8789c747a2246d8cc82d2c4f1dcd9e
5270 F20101212_AAAUAE kneifel_j_Page_012thm.jpg
9428f3b184ab0118d3f5179efcaefeb3
08a6da79a4b4a31b8cf3b103cd67174b42c56755
2243 F20101212_AAATUJ kneifel_j_Page_034.txt
8edd1c42b7c260539cf3529b176cfa1f
e63a24b97d1393b8f4788efc8efbd09b32ac4109
2237 F20101212_AAATTV kneifel_j_Page_014.txt
f6c9c8bd31d9d741d2d15c4bc8e09805
072795362dbdfc88672e5e3eebb5576b89ffe643
2279 F20101212_AAASRH kneifel_j_Page_109.txt
f98c58119b7ff9ae6cee7a4ad6637f1e
90dd43220d2236d816bd1c1a71c26ac9d397543a
19135 F20101212_AAASQT kneifel_j_Page_184.QC.jpg
27372b7100717f6ff0761aeea40fd0b5
32f13386a0291292917865ed846c778b7fba0d72
9504 F20101212_AAAUAF kneifel_j_Page_013.QC.jpg
e93c9cb6bb75b40946ae6f6af4984aed
33bf4b7265646f4b408d5b8f64930c2af45fb0b4
2378 F20101212_AAATUK kneifel_j_Page_036.txt
4ec251502e096e0be0fc29a19044ce41
0fe9d8907bd9a7981cb018e2a38c7272af70511d
2160 F20101212_AAATTW kneifel_j_Page_016.txt
3c1c9762d2716d34337ba04d116a68a8
b3f4b40581e224672feb2bfeecb8a68aed2e0c86
23676 F20101212_AAASRI kneifel_j_Page_095.QC.jpg
d6624e9e28711410143c117886d29956
b4ee30231ab0fd6afcad82787b7a8208de5c4ad9
2218 F20101212_AAASQU kneifel_j_Page_176.txt
5232590ecbf9f5f5dede5a9b46505836
439fc33c3bc3b215617f51813b91a257509e66b4
2673 F20101212_AAAUAG kneifel_j_Page_013thm.jpg
1d9dc8c76599edc8b55fc0986ab39f24
38f1d1b0d4265e6e17ad7eca00c56657dcfcbeb1
2288 F20101212_AAATVA kneifel_j_Page_055.txt
5576f8cceae2c132b98e8545c35d7273
4aab59adc8bf98cf58047deebb0e436aa49fca25
2184 F20101212_AAATUL kneifel_j_Page_037.txt
9bcf46f528016b3aa18b08663e2f5ebe
9d85509a5f1908f2dc92116f225b697ffca3db3a
2182 F20101212_AAATTX kneifel_j_Page_017.txt
f39a59a86b6fa3208815394121bf80c0
075c43c99e89b473b8f98c16f57758a172691a5f
50952 F20101212_AAASRJ kneifel_j_Page_185.jpg
fb998fc321890b133a9642659c017520
0a5cea7eeb4fcf0ffc06f41f1eeabcbeb7fcb474
88299 F20101212_AAASQV kneifel_j_Page_075.jp2
5e4601a080a6eda96342f88100311d05
a40b38e9fa6dc4d5d7d298cd89fc64cda13ec096
25332 F20101212_AAAUAH kneifel_j_Page_014.QC.jpg
26062e8a965c101edf35939e42cd008f
cb7f3fb94563e7c1294479ca288d9d6e23deafd9
1947 F20101212_AAATVB kneifel_j_Page_056.txt
b55c22930afd9eaeda691be85d591952
24a01219d40454c4219ecbc2bf75887a9083ded4
2322 F20101212_AAATUM kneifel_j_Page_038.txt
77ef2704aaa54669b43a71ac1e8571de
4c4ddb1803b98c947280f922cbab99cc0d83062e
2224 F20101212_AAATTY kneifel_j_Page_018.txt
a3804bf58c7e05ac3f1b78ba9227e488
eed7ceb6cc9b10857b7ca40d660a79c8e272ae9e
2035 F20101212_AAASRK kneifel_j_Page_106.txt
3b9dcb4c1ff75441fc9e4656a8a9d6f5
37aa64e5552cc3eadee45c5c45948394dbbafd6d
25675 F20101212_AAASQW kneifel_j_Page_156.QC.jpg
a7bda91aea4c7f31906312d1140ff590
939307ecd627945e718cbb890c7c0d5ab20fab04
6452 F20101212_AAAUAI kneifel_j_Page_014thm.jpg
57dab8ed27ea272dc5e7c18a0491a255
879aec9b5479a61c664684b07ddc4c0c8cbf1ca7
2217 F20101212_AAATVC kneifel_j_Page_057.txt
bd8cd5fdb3b86ccf1921b2d5bd5acd80
d4e4432c392242fa15acadd9ea1967f8ef319c03
F20101212_AAATUN kneifel_j_Page_039.txt
ab988c6e482247cbacbedf248c7c0ca8
8a4f735d8466824db418c79ec3f0fb010dce2fe9
2331 F20101212_AAATTZ kneifel_j_Page_019.txt
447bf9af6786a46eb2b1a447c79e7af2
e2061f340f3653f78b57ad0c575f8ec80a741f71
5838 F20101212_AAASRL kneifel_j_Page_138thm.jpg
a379c03f2886ed26ed17dc99f7426667
f533f88f272d1d8f49781475ae32c614fe544097
60573 F20101212_AAASQX kneifel_j_Page_036.pro
dd0e93c313a35a54447830fd590b2863
17d0ee4d2cf6a38195dca19dab22fb6352773297
1051913 F20101212_AAASSA kneifel_j_Page_040.jp2
0a38b6b746a048555f98148979e2e4ce
a34ac4bbc4ef5d06da1b9b82fd0245b4abde5781
6715 F20101212_AAAUAJ kneifel_j_Page_015thm.jpg
11e9f76e7006eb15a230ef3e8b4cfe77
fabb43bdb1efe483a0016eed765575b47ef3ea5e
2320 F20101212_AAATVD kneifel_j_Page_060.txt
188060d9dbfa689fb35f4d773a0355af
df5087ac15047cdae04fe9e7991b8732a6cb5560
2019 F20101212_AAATUO kneifel_j_Page_040.txt
fd5638d8a3c76edffc0a323b8a5ef23d
b0a04eb7628960aded88c9fe11c314b4449e543b
1051986 F20101212_AAASRM kneifel_j_Page_134.jp2
ba3e70d7d2f208b478ddbfa315e461b8
64000544753f9949b633dd864135676fed7beff5
6925 F20101212_AAASQY kneifel_j_Page_078thm.jpg
c77bed85222018711b889df12dd6ea0d
c13236661fe9c8b2b66b855309dcfac758f4ad84
F20101212_AAASSB kneifel_j_Page_058.txt
a7c33fd008d2aead49a92a762fa81834
41d5594ee6cf88afcc774f906dadaa2592e6c94d
5809 F20101212_AAAUAK kneifel_j_Page_016thm.jpg
17ad638b738ba9c78bcd1a20dafb5420
8b75dfe447c14acdbdff5c4cf74f82c78715d04f
2561 F20101212_AAATVE kneifel_j_Page_062.txt
04520a202e2ea3b974172dc23ea9fdec
cedaebe27b85e5942ea5c8606740a1b7a06b1cd0
2211 F20101212_AAATUP kneifel_j_Page_041.txt
4a0123da5160dd37eb6228f2f7ef6afd
8d3e482702c1f956265c0892e8b207492479401f
99829 F20101212_AAASRN kneifel_j_Page_192.jp2
c4cbc81c4e221541d307c94740d13545
67a6d19561bd1150a37e7acfb97729842506b9e2
F20101212_AAASQZ kneifel_j_Page_105.tif
963837e26bf666e238d0e34c377d504e
3f0eb2a95503801afe62bc430250c0214efe7f94
73543 F20101212_AAASSC kneifel_j_Page_178.jpg
6e221336fc33ec3335572df11e0bd0b7
e22e6e18ff8b284c015c96a3e3d8261e6f565caf
25911 F20101212_AAAUAL kneifel_j_Page_017.QC.jpg
4af3eab7a232b87d97f28740c7a782cb
5c483d7cb0dd1f7c49294b36399dd9fb96f3a0f1
2109 F20101212_AAATVF kneifel_j_Page_063.txt
d64d005f7bba27d749c85c18a5a10637
cf13a7885bbee43ed60655209e060c8d41d7f572
2175 F20101212_AAATUQ kneifel_j_Page_042.txt
3af7268d019800ed8873f1b38bf7460c
3f5a442f94eb4f13c6e887887091ddafe80492f4
F20101212_AAASRO kneifel_j_Page_216.tif
491d4a281781ea3038e47bbed6dad3aa
455f68c99e3d30640a27325632d001f3761ce294
2114 F20101212_AAASSD kneifel_j_Page_075.txt
f1fd827ce4a7fd92cdb1240da37cace1
2ebdf857b5cd5fb2768680167691ab8a05765a18
6826 F20101212_AAAUBA kneifel_j_Page_024thm.jpg
5888dcb7ee5b9a605d597f8b581de921
7e35e231e075a05fe8de363d715e1af3262ff977
6714 F20101212_AAAUAM kneifel_j_Page_017thm.jpg
447bfd444f9508f46886fc98298db665
4b890a4ca84f8f62136420817d459dcebceac129
1969 F20101212_AAATVG kneifel_j_Page_064.txt
b58a9e36a366e057cc77f3780a2fc209
7da19a218e746f06cdc151cd5d78c28c74b159d8
807 F20101212_AAATUR kneifel_j_Page_043.txt
3330c7bfefac45309cd067db7fc23ff6
0eb3bd5f6552e5e1e249ca27ae5fb750dc36099d
55223 F20101212_AAASRP kneifel_j_Page_041.pro
fbb171491e544aa023c51bb9a74f80bd
0d8c6ef102b1b758fcc691d0249bab5dfd8f5eca
22109 F20101212_AAASSE kneifel_j_Page_171.QC.jpg
99a4372ffb9064259383152e8e830248
dbd86e0ee5df8f55e0f9cb647dc4be2a259932b4
30291 F20101212_AAAUBB kneifel_j_Page_025.QC.jpg
48de27bfd6173427eb02d3f827ec18db
d5c728084d71175f3949db028f3a583a3ee4caf7
25630 F20101212_AAAUAN kneifel_j_Page_018.QC.jpg
054ba23d22ee07f15f9659c75eecee72
3fe9d5060371fd289bc8d970f5a5baed952886b4
2112 F20101212_AAATVH kneifel_j_Page_066.txt
1700aba386fdbc92387c1c867646b6bd
1884fcea77397eb0b6fa0b280af3c06f497c5f66
345 F20101212_AAATUS kneifel_j_Page_045.txt
c98ffb80a37bfb49730b27ea7ed6423c
e055b5fc98aa4ca982b5f05e2f1fb6a94e7e8d7c
49048 F20101212_AAASRQ kneifel_j_Page_206.pro
92a897fec4fe506ecb005ec3e29c7f34
b0ee9b89ae581da94fbd92161a0510bdb760491c
77759 F20101212_AAASSF kneifel_j_Page_189.jpg
c88e8500a02dcdd884155557b8e3b59e
7964bef1991d946f9e4780e9e709cf8e96e769db
6215 F20101212_AAAUBC kneifel_j_Page_026thm.jpg
2a5cddf915cfa7cf31146d13e74b85dd
d1ba68af4de288eb62dc3f1e643efdc4b3afad7b
6470 F20101212_AAAUAO kneifel_j_Page_018thm.jpg
606379a6d981561e77024162c56f2f74
569d95bc3443edfb60db8b2d7bdaf62367110c16
534 F20101212_AAATUT kneifel_j_Page_046.txt
a94ff09295f4fd28f4f74da64d95f21e
07887cba3d45ab0ad73a7669e239fddba2c1acbb
55356 F20101212_AAASRR kneifel_j_Page_037.pro
dcd0090cdcc56fac34349c69515b0455
5dfaa8c75493b4ae7fbba5da870218a777f40cdb
27821 F20101212_AAAUBD kneifel_j_Page_027.QC.jpg
d91ed3d028535c7a8202616580feae8d
400e2a57562b39ff84f28b86d6db27faf6a59c8c
27111 F20101212_AAAUAP kneifel_j_Page_019.QC.jpg
da58fd3c660578a34ea1e21d04c5fb3d
1112953628e8b2da39d9923237b4bdd3a9397efa
2084 F20101212_AAATVI kneifel_j_Page_067.txt
01727be5e845389a286e5e8004b37fff
a38e4cab04ccf8f89d9c194a834fd31bed215878
2183 F20101212_AAATUU kneifel_j_Page_047.txt
937fda1b536ff74427e51b1feb34ecbd
dad1af28c0e11b06e5ac6e1cf6945c8f9fefcf19
110284 F20101212_AAASSG kneifel_j_Page_098.jp2
766aaef45fb143e3804042baa1bfb3c5
e9ee5f1d0da46ef1209a1c083c94ec0917f0d3d1
83431 F20101212_AAASRS kneifel_j_Page_007.pro
e4db61b70ed5c551268d82aaa39633cf
957307a562ead47ff7f0a510168d7b0c603b8c24
26861 F20101212_AAAUBE kneifel_j_Page_028.QC.jpg
48f9011bce5620aa718775581244e754
e3d94a815d7b3d1a57f2cd5a7d6d63d0226ddb27
6748 F20101212_AAAUAQ kneifel_j_Page_019thm.jpg
fbc7d867b3a5e7aca07934d4bed7fa29
e90754fd7d8e74f6dd13fbe8151a4556f11a6753
2466 F20101212_AAATVJ kneifel_j_Page_068.txt
68fca2edeee29c1650032f1cd8679a56
e84fcd2f8103ec2752c6dd0d43074119a4dce0cc
2312 F20101212_AAATUV kneifel_j_Page_048.txt
673910929c0b3ad6d1a535d98ec04549
2941a4228c0aee2e8817a4377dc98669ba4484d6
F20101212_AAASSH kneifel_j_Page_089.tif
36a044b9dc5f58eb0d612695ce64544a
7c45a5285de2750313a58f9bf5ca85d6cad7877e
653000 F20101212_AAASRT kneifel_j_Page_119.jp2
f94ec664e27f0548eebc75510e5f1690
9984194da89adacb48fd085c26ec40ec021a8d73
6692 F20101212_AAAUBF kneifel_j_Page_028thm.jpg
83d7ff825f1b6b685bf0db46438b4add
8a02f83d9ef0215ec870af5d5f238dda03ec3b17
28455 F20101212_AAAUAR kneifel_j_Page_020.QC.jpg
dabd8be26af2687da1178925216d2988
9a0744b3567facec7364c76e1474415534a323d1
F20101212_AAATVK kneifel_j_Page_071.txt
c8c8fd0b015dc2c63742e67a1e714726
cb710f069b40bb3fa24648e0c9b1ac924fd17938
2152 F20101212_AAATUW kneifel_j_Page_049.txt
611d3cead3f395e558e51e987dd4d4bd
b290212cae65875a71e801fc5e5e573473203ba8
5982 F20101212_AAASSI kneifel_j_Page_176thm.jpg
e9f219129be186ce63431858c7faff92
604cbf24017a8789f4d1b199a4ebb0ad51161cc9
F20101212_AAASRU kneifel_j_Page_164.tif
248d85b705e3ec5d0419f8d55c7d0085
8fb52a32bc5d09b672bd6bbf691d621f5e5693cf
26201 F20101212_AAAUBG kneifel_j_Page_029.QC.jpg
977d5f3a8db30d7b3ca1807491819a4c
aacc609043b1ef539d6cf0363c08f451123e2ac3
1937 F20101212_AAATWA kneifel_j_Page_096.txt
d7425cdc99aeb01f495c0debd86b6a20
f9763c997671874d4a842322eecd9272dbf6bd60
1778 F20101212_AAATVL kneifel_j_Page_073.txt
849f53212b58f3c410c148caecedddab
aaca8144043a8fb4d0877dd415da99853c296f00
2126 F20101212_AAATUX kneifel_j_Page_051.txt
3f12c9475ef59f98b10d3e9a19f170f8
8b7fae2cca872ae5fa48bfd86c7b72c22feff48b
F20101212_AAASSJ kneifel_j_Page_088.tif
e152a9874aba900fbb68ae1db51df9bd
ce41650cc8d35ce19482963f66285230f6d4a322
F20101212_AAASRV kneifel_j_Page_051.tif
ad76adfe02ef415b514d53d7d02cc00a
1a564963822e30d8b51f0a68b0f78d7e6da2c29d
27620 F20101212_AAAUBH kneifel_j_Page_030.QC.jpg
a4fd7768b15639406d3f4fb6049de32f
8cc88cf6b7793e593ae5f9af98d7e7a62cebc339
7145 F20101212_AAAUAS kneifel_j_Page_020thm.jpg
fcae98c53d3d1ff36f9a8e13f7bd47ce
acc94e583dd52383cf93a003a8ed1da7869c8b50
2365 F20101212_AAATWB kneifel_j_Page_097.txt
d111f1d9adc00e0a2031f323c0cc6fdd
8cbd9dc3f8feea1810b2b60a0192bf74af3eabef
2133 F20101212_AAATVM kneifel_j_Page_074.txt
5190c6e86f0e41ab65d3bf87f66b8e50
73e8d63f859e932331dff060ca1077b45e4b0278
2381 F20101212_AAATUY kneifel_j_Page_053.txt
55d3bae18a5b90234b03e452dc92a57e
0d1e0d7daddc8318814bb546b01119e8a6c5cabe
23414 F20101212_AAASSK kneifel_j_Page_042.QC.jpg
89ec3197b46c28648bbb12bd69a45c83
84bb4e2b513097798458bafefc5e5d7b368e7e61
916684 F20101212_AAASRW kneifel_j_Page_072.jp2
fc119ae84141d8a29f33155772b9b5ce
435df77fdebbed5520643e5e080023736706bec6
6695 F20101212_AAAUBI kneifel_j_Page_030thm.jpg
ae588e99f7fff19afabfcec6d0fcbf76
0efba3ba14aef0f226bc2d4762285300ba87c2c0
26899 F20101212_AAAUAT kneifel_j_Page_021.QC.jpg
b70f7f63406db2154240c8190438926e
3a4e205987b7e4f739109fa2439b031396131c2d
2281 F20101212_AAATWC kneifel_j_Page_098.txt
4258e6d136b0cfafde5f7b6a1509adc6
7e087ee096cf953f349a4d4b16f1e4ec53f6f818
F20101212_AAATVN kneifel_j_Page_077.txt
d15b8da51cc58a8d137a5b4f3f723a14
f20036664d299d1332a1dc3542355fd7a79c9a8c
F20101212_AAATUZ kneifel_j_Page_054.txt
88719316e589afb61895830f55ad71c5
b162fbe15d5e22dbe4c4f228f8e6c3e479c4ba91
10587 F20101212_AAASTA kneifel_j_Page_112.pro
604c209de1ae0a01626cc39d3d42ccbb
43056a28c52a4f0033292e78fb88c6f00e01a860
4948 F20101212_AAASSL kneifel_j_Page_143thm.jpg
d170a117bdee5a431a7cedcce01e38de
552f1310c48e6b51a2fedc5794cfa0730c330639
55050 F20101212_AAASRX kneifel_j_Page_017.pro
9f505f1398764fdfd0a1dbd660f19e36
6ca47bc242e46b2b13482cf1ce5be4afc5acd797
24394 F20101212_AAAUBJ kneifel_j_Page_031.QC.jpg
ddecf9685d9b0ab91478819fecee1f40
60f2f41906d433cbe0c0a2da2aa32aa5edbe20e0
6791 F20101212_AAAUAU kneifel_j_Page_021thm.jpg
2f6f36c57d2d952c2139e53b51e64add
3dad8011e70bb72485004ef5877e3dd1b5322628
2563 F20101212_AAATWD kneifel_j_Page_099.txt
09e6e81b12a3df3627d2ec0372d2f4f1
53eeba43ba5e6a06f55998502a17649e14aff84e
2230 F20101212_AAATVO kneifel_j_Page_079.txt
46b4638bdef517dedfc1742883edce0d
5ed8f03e1e54da1f3fa3016653a2cc7b97c4b02d
64798 F20101212_AAASTB kneifel_j_Page_107.pro
624e7e84ef44d6c177d3cb10ccefd764
fabf7a43c9fefdec50e604d03a3008d693f7efff
1660 F20101212_AAASSM kneifel_j_Page_044.txt
238c5ea1bf9c1fa9fede893eee03060f
ac2a0338a83facc971e992169b26d7226371b77e
964872 F20101212_AAASRY kneifel_j_Page_171.jp2
cf5915481449c21f9bb266f47dd9f1b4
6b6a21c3bb828bcfc5398d4d33849fb580d924f8
6429 F20101212_AAAUBK kneifel_j_Page_031thm.jpg
b3243ba973c4a25e5845eb078ada7061
29bc9ca220ab9d1d1dbdf0d497a7b49dab8fe709
25776 F20101212_AAAUAV kneifel_j_Page_022.QC.jpg
702fc1cb0fd475a9430bbe89675fb051
2ee58e7928d540cac829a747b8553ece6acd183c
2652 F20101212_AAATWE kneifel_j_Page_100.txt
250f59960337cafa8d814da660aff25c
cd6c66c232f250e01866097b6b9eb2331371b974
2141 F20101212_AAATVP kneifel_j_Page_080.txt
e9daa562c864a2f3bbd145dd3e3a7991
6ddb828f95fa0e862018cdb309447400ccdc7702
5387 F20101212_AAASTC kneifel_j_Page_192thm.jpg
082285b650ba050c593aa73165384463
818d218f75977842347f160f01c57e42c0b8b064
5954 F20101212_AAASSN kneifel_j_Page_072thm.jpg
7781d4ff87b60951b0246f76371cfefb
3db3235d60582e9dcd97e84a70489946a7bde1a5
74093 F20101212_AAASRZ kneifel_j_Page_136.jpg
16356eed9f4b29b66a93758f08fc5461
29aec7fc3178d72b1988b38391f4440cf5cf9775
27627 F20101212_AAAUBL kneifel_j_Page_032.QC.jpg
8c4fca1c9ac1eb07fafbdb800386ec53
76ace53389fad7a77d20d0fda9a3990444bb78a7
6330 F20101212_AAAUAW kneifel_j_Page_022thm.jpg
63cc89ead4fdff0afbcf0a45e20c5042
a7a6e78285296a7bbe9ac43ebbcc45bda9ee9683
1657 F20101212_AAATWF kneifel_j_Page_101.txt
95af637a19d68ada5dcc5e2fc728819a
808c563f1bff1fce53a2ed7079548f823733729f
1967 F20101212_AAATVQ kneifel_j_Page_081.txt
bce5ed67293876cd4697c7f3aed96934
87b394ea58b954b7f7398715ca52713395690f4b
83942 F20101212_AAASTD kneifel_j_Page_058.jpg
6b066179033991a360fb2903083ff491
43fdf3897a4796bf17a38b43fd7ce2f95a45e589
2377 F20101212_AAASSO kneifel_j_Page_022.txt
37326b5ae13b894041ffb939ef7172aa
1af051f0997a42d568d6ab9e4a3896f13c523ac0
5859 F20101212_AAAUCA kneifel_j_Page_042thm.jpg
4b127b4d77c0fa4241af90399c2a8797
393c10d6799b8004d103ed9d02f118cadd49ba4b
6820 F20101212_AAAUBM kneifel_j_Page_032thm.jpg
4880a0f72720c103b07ee6caabaf7471
6674c229a1db12b30115ba91cd10ff3359bb5b36
28593 F20101212_AAAUAX kneifel_j_Page_023.QC.jpg
606bce5fcf3dd8b8deada95d7ce0bc52
38b45e1fdd5a9b49599defd2c68a198edcadad9f
2158 F20101212_AAATWG kneifel_j_Page_102.txt
f6bf5e2664a490c7bbaae5f0583d3fb7
e8943d3a8ad2041f92be313884b4389d050adf8a
1876 F20101212_AAATVR kneifel_j_Page_082.txt
2ba86eb47e3fbc423b42d629cfb21ebd
3168791b3e12887da5416fdccf8f87feb7026396
66376 F20101212_AAASTE kneifel_j_Page_020.pro
66b272e9f3050e19ba15984d95257f5b
d07749a79d90700ea3d83e3172634e50930acbe1
F20101212_AAASSP kneifel_j_Page_145.tif
517b1c640cbcb5e2cf0c1cd6a72e9daf
6ab70a8caec185dbe96cacbd4a6d920a3385fa77
7888 F20101212_AAAUCB kneifel_j_Page_043.QC.jpg
561d621dc29f1380febfef9e8160edb8
b89a74109eebcf2bd2e94ded8229f8926a621a27
26876 F20101212_AAAUBN kneifel_j_Page_033.QC.jpg
2bdd5499a60143a1dda550264cca55df
2c57065bbf51530932d745fc5ab93935bf36c225
6883 F20101212_AAAUAY kneifel_j_Page_023thm.jpg
4f7171ee4b541a3f077e4b4bd3e41cbb
f4d02334d9a5f34ab2b53188682fe24343b4a094
1835 F20101212_AAATWH kneifel_j_Page_103.txt
41b0433d60274eb8c26638e1b3ce7f30
7af0663828416490ac0cb331a794d6fb268fc10c
2242 F20101212_AAATVS kneifel_j_Page_083.txt
bbdd56d4ddc8b11ee348bae8ef8a45af
b4b33ead49ed76d7f4eba5e2e435e80ee5516561
9956 F20101212_AAASTF kneifel_j_Page_193.QC.jpg
35634a122548da3bab2da2049468ca15
606667b2392eb20c84ae1af4015763ba75255f82
2103 F20101212_AAASSQ kneifel_j_Page_161.txt
839401c63cfadca52a18035583c5d2db
eb75d97a2faa364e7b25e056c340871886daa1da
4499 F20101212_AAAUCC kneifel_j_Page_044thm.jpg
c2083a517fed74f286713e7e4aaad870
01c2899b309f15086d2784bd5bd7d156f1aaf1de
F20101212_AAAUBO kneifel_j_Page_033thm.jpg
c0e15bc0af704fa29f38c55a2fe982f3
a3fe42f8fa8444c6c70658b812cafbf12e630055
27920 F20101212_AAAUAZ kneifel_j_Page_024.QC.jpg
ec5b7698ea912f6e30e5811807cc807a
52604fa9ffdb57b3e2ccce9ee24027baf8929cff
2427 F20101212_AAATWI kneifel_j_Page_104.txt
3ac3491a361a739ccd3570b4dd36efd2
f6f5125954205cf33870852b618fb871db65fb99
F20101212_AAATVT kneifel_j_Page_084.txt
38727fc763838c28ebadd7601bee560c
7aa6757211ae03a67efcd3b57272ea8f5d0610bc
27025 F20101212_AAASTG kneifel_j_Page_050.QC.jpg
c5eaaa560493973770d165008c80ab29
d175be858ae2144258e81ffec42f7f404f44dc1c
58862 F20101212_AAASSR kneifel_j_Page_050.pro
ee19b2c87d0e9edbd59802e947dac31f
6d9b7afa34a7183fa2c3eba5c6f01b1324c1e6b5
5913 F20101212_AAAUCD kneifel_j_Page_045.QC.jpg
a8eef24ea3c0d721114dd273ccb4405a
c7c4f9db634f0a1b891b37e300155c28c27f38f6
27449 F20101212_AAAUBP kneifel_j_Page_034.QC.jpg
4ecf3ace266b7d3c1a74e122d8e4d149
8b862693e94058259c87623fc25c63a013db9b33
2159 F20101212_AAATVU kneifel_j_Page_085.txt
bae22124beb6b891ee694d28e477d63b
0370bc85eaa0c506fcb0f488858bea0c43177d5c
6001 F20101212_AAASSS kneifel_j_Page_091thm.jpg
2994941ef9f3418f23108286d275906e
feffb09c9de3d9b7075f4a2f7f74a7c834fa6125
1988 F20101212_AAAUCE kneifel_j_Page_045thm.jpg
e2217df3f8c8be3bcc617a6c645fdaa1
3545bb7a9d0116fb9953fd937de1e37308523cda
25278 F20101212_AAAUBQ kneifel_j_Page_035.QC.jpg
d27f75b274ab3275883c356e2f8f89ab
d176f204ada0661a8ab1793722759ad223dd4d72
2686 F20101212_AAATWJ kneifel_j_Page_107.txt
a507ee9f4eabba10219fc23f2d32d1aa
9851e6743f7597811f02de68fad94f5345204cea
2146 F20101212_AAATVV kneifel_j_Page_086.txt
9552197671bf81276469024a615817b2
4ea55de685efadd181fdc0b0c40918a04df258b2
23542 F20101212_AAASTH kneifel_j_Page_108.QC.jpg
d1eeece5cc87b5704612913520f69128
e50ab4a580af213e427c4a6581886bf18ecda857
111830 F20101212_AAASST kneifel_j_Page_110.jp2
456dda6e7bd1c61ed18c4b445ed54fed
0cf4b3ba699b0f7be2e57d8054758538cc003e10
11058 F20101212_AAAUCF kneifel_j_Page_046.QC.jpg
0a13d5e81ef2a14650e86e4e56f5346b
7dfb2d3cb70194adc7ca71e8cb13141dca15eb97
6507 F20101212_AAAUBR kneifel_j_Page_035thm.jpg
271606cfe4563e823d3b7088e8cac5cc
238c5196f7cf9fbf0e8ecb8598adfc6226ba9808
2792 F20101212_AAATWK kneifel_j_Page_108.txt
64e9a1674b670416551fb3abce47140f
8a08c6c31a6ca48e271d73159aa8991d5153c672
2300 F20101212_AAATVW kneifel_j_Page_089.txt
42b97563de916899353d9c2935047d9b
bd726df8aa782ccf7ead8c83df2e656fd8b852e5
F20101212_AAASTI kneifel_j_Page_008.tif
da97b7575b7b78bbcfb4adac51278c03
830dcd34a5bf419f05f46ce11be9852b0ebd959e
F20101212_AAASSU kneifel_j_Page_094.tif
cbaabf48fcb702c5ba45b0e49c7d49aa
a61100474e1457732240dd15111e29ebbae97d3c
3330 F20101212_AAAUCG kneifel_j_Page_046thm.jpg
ea8c9c200bd9038d7261150a1899a405
c9f51a20060aee14c3c6622e483e58873128a3f8
25264 F20101212_AAAUBS kneifel_j_Page_036.QC.jpg
5d53b454eb00a9c5935a0fe6fc3eaf90
569de05c4e854346c04d36af1a84ee18e8d2828d
2143 F20101212_AAATXA kneifel_j_Page_133.txt
f9d77ae524d40e7e00af14406b54a28e
35bea19f0dc381f79369388ebc920ffd86064846
2172 F20101212_AAATWL kneifel_j_Page_110.txt
755a1266a706fb82861b9deeb3b5766b
3db2a1ef8013b7e6887673835085da0c53c96698
2226 F20101212_AAATVX kneifel_j_Page_092.txt
bbdcc4f77eaf85099a95bb6ddb431c4d
8353d92346cfa3bd277ae67b085ada2efe4217a3
22172 F20101212_AAASTJ kneifel_j_Page_206.QC.jpg
64009ae2a7d6a57b6c4f2c20ad8bd652
16b9fb2f8d52c9551b08c9179e92e4412bd6e51b
43050 F20101212_AAASSV kneifel_j_Page_182.pro
c1f5fbcb56764c3e770294e2dbf8a03e
fb8e7488c7000ec996380b9508deb9a2a995272d
F20101212_AAAUCH kneifel_j_Page_047thm.jpg
53f200eec35b7b374d66d87ebe334ee8
3d79a3a47055ce1524f218e13a52e7fd2c7f40f0
2286 F20101212_AAATXB kneifel_j_Page_134.txt
80eb6feea47d34625b968c7a0004202b
d6235f4881ddfe21860acf2e480f4663f0bb5acb
427 F20101212_AAATWM kneifel_j_Page_112.txt
7514d084048a8f923c34187c1c38371f
0e58bd744dbddbc2c95aa48bd777cb55a0821c09
1938 F20101212_AAATVY kneifel_j_Page_094.txt
40272ea778f7236c87a74c3f090a953f
0053089ff6a5374fdbd8da8d56222515c98cf702
F20101212_AAASTK kneifel_j_Page_034.jp2
f00ac4977bf2abe4028c1f8b928f945d
0b2cc1b9d8f8f0c1829357c43fd5d3b740cd378b
5985 F20101212_AAASSW kneifel_j_Page_098thm.jpg
172c978111d90cd6f59f9ec2f59b6a06
22673a6fc8915bb71f75808594a7e11326e78f12
24472 F20101212_AAAUCI kneifel_j_Page_048.QC.jpg
824f028b46e08b458da1ba54bd17a768
42089be34e176128db3022661eae825b70e46dbf
6237 F20101212_AAAUBT kneifel_j_Page_036thm.jpg
1643e65c464d55d217f8119d18bb8548
55114b89803edd414adadff1deee89bf38d78a36
2262 F20101212_AAATXC kneifel_j_Page_135.txt
c4e8b28223c0be7d44ea5d8be6dbd551
3c695ad10906ff0eb74f986bc3ca7942997ae7d6
1406 F20101212_AAATWN kneifel_j_Page_114.txt
8d8cafbd893368510e97dafe5472aafd
91f1dfd3f190bb353b0ce1a4897c59c7a1f274c0
2335 F20101212_AAATVZ kneifel_j_Page_095.txt
73f7c7189d754293403804e09adb08a1
62841648e5a4a86d019a94a2555c4289412fe0b4
F20101212_AAASTL kneifel_j_Page_081.tif
b9159c1dd390c80fa19a0b7915d3b13a
70406256d24e1758682c2933fb47270c4dbb3993
68080 F20101212_AAASSX kneifel_j_Page_114.jp2
5de7b5c11b2378052323e1487b8f3352
ca0165b280a5c60beb65b440568c112e330ec1a7
83799 F20101212_AAASUA kneifel_j_Page_014.jpg
e61c1b0efbae21f66f213d95e68d775f
4595ec2b6b0a33097018eb79ef126c9dfd230e59
6278 F20101212_AAAUCJ kneifel_j_Page_048thm.jpg
3a1f5a1069d9c1b61c394c9f1382957d
8246cb168d8cd16fd21f00a1099c3c3b5c53b2f3
5928 F20101212_AAAUBU kneifel_j_Page_037thm.jpg
ee029858a9151c441e776cee2f0e5112
14e9069112a7c8b1586d32067fa2288d7d21ab60
2080 F20101212_AAATXD kneifel_j_Page_138.txt
554343ae2fef1949299c26e3c343bbdb
622a2eb314789fe432b9ab415fc131fbdb56183c
280 F20101212_AAATWO kneifel_j_Page_115.txt
b12b3abf1ea5a22125be227a8273bf08
b8fc5e5de83840db774819216dfd1914512050da
F20101212_AAASTM kneifel_j_Page_013.tif
0ad2b81d1581a0b250cc69a6a112212b
31a17378d7b6ea4d1e9bd690883eb1b1555c26e5
85320 F20101212_AAASSY kneifel_j_Page_186.jpg
e8c09a21fe94054c0147aa23ecb58944
7539982b4bc8074b35c5a041a260b809d70df111
6469 F20101212_AAASUB kneifel_j_Page_102thm.jpg
050208f68b30899fb126af0c117cde54
7f29e23f7fbfeddb09496b8ca84088a839e1e2e0
6375 F20101212_AAAUCK kneifel_j_Page_049thm.jpg
0846aadbdd2a2290a051c6e011ba5a07
562cd05fe4acee9aa47ff717a7f548e75ad461ef
23951 F20101212_AAAUBV kneifel_j_Page_039.QC.jpg
f3ff4a723c578dc0b34c085d112d95fa
9164c43d2421c6ce03828d9421170bcc6b76b78b
2161 F20101212_AAATXE kneifel_j_Page_139.txt
252b3a27b093444ec3614f2c0e3cdf00
681eb2f6ccee42c2c727804a31ab9a0c3df7f02f
938 F20101212_AAATWP kneifel_j_Page_118.txt
90d731d04c6ab1278739980d5435d7a3
09926ad86574c8df8ac59aa0fbfd62f422910678
1051957 F20101212_AAASTN kneifel_j_Page_050.jp2
0cd17d5ab96d14174c51d9606b6fa25e
2673beb88e7304d378cbc2ad94be18c802f5b1bf
6529 F20101212_AAASSZ kneifel_j_Page_029thm.jpg
1ea6b6fad64ee554d62471f39b9ccd50
4032d40be87e836921de164fc9245decff7352bb
66282 F20101212_AAASUC kneifel_j_Page_172.jpg
d5fd1c7ea10f218d5170f3d5ed43cfae
e2796fa272f8d5a6a5230db9f8cc5824532f4c3d
6904 F20101212_AAAUCL kneifel_j_Page_050thm.jpg
9bc8a3fed8c9f69dca1233857c98bf05
e6251ac1ef9a9b4f05a77f04590ae1e3106e679f
5950 F20101212_AAAUBW kneifel_j_Page_039thm.jpg
e5408d0818695e52312ade4cf2bc2f45
94cd976bcc81726a0975cd5678df3df7a1bab164
2409 F20101212_AAATXF kneifel_j_Page_140.txt
296eda87a66f6c27f43d0835c7dad346
9464634ad8102b5f588f119730353e773c1dd325
838 F20101212_AAATWQ kneifel_j_Page_119.txt
c73720f2bf721fe3cd638a1032355023
912baeb54c9e9d404f011cc2d752954988a40861
39191 F20101212_AAASTO kneifel_j_Page_187.pro
92b2d2426832ffa7095d47cf67e7bccd
6c82e7410bfc41f0bdee9de27b9b581e0e749f45
115421 F20101212_AAASUD kneifel_j_Page_150.jp2
78245b218866bfba096595f987f8b303
27b86b146dc395ed472384fe86ce174a2f31c468
26408 F20101212_AAAUDA kneifel_j_Page_062.QC.jpg
b8390d53748fa6ce081112f5bdded56a
5f1809d37843c39154803fd2aab3edf3c0991788
26222 F20101212_AAAUCM kneifel_j_Page_051.QC.jpg
09f78a30508284615cc115544edc429e
e495de697b745dec7d445eddca6eacc6ffa6f5d6
F20101212_AAAUBX kneifel_j_Page_040.QC.jpg
e96c9ae93104cc9b10f62a2e2d36a3d8
c88684ffc056e69b8652fc7f7eaba11c85e48726
1872 F20101212_AAATXG kneifel_j_Page_142.txt
3abaf31f91a22b392259fd6edbc999bf
0db470d5a398b2fcb4d63d701e5e68afad373522
1548 F20101212_AAATWR kneifel_j_Page_122.txt
e9f38b7349d24440eb1d60060b9ea36f
c2a4ce3585ba33b28bf90a3c57aede0c7e4a4b09
125634 F20101212_AAASTP kneifel_j_Page_036.jp2
5a31c73c5b9b4222d7878a510a39dc07
96f218a7368a0a9fcbe97e00855d2fb3294ec591
2257 F20101212_AAASUE kneifel_j_Page_150.txt
1dc6fe9d30cc0a53e190958b63aa766c
80a4d2fb423cab3118ddb9776e69294765e880dc
6891 F20101212_AAAUDB kneifel_j_Page_062thm.jpg
e5623d12670c80f38fc644e15efcb9c7
a34a1b1bd792745218180e3f4b46db29f0478699
6460 F20101212_AAAUCN kneifel_j_Page_051thm.jpg
96480f8b76930d629e085fe0886d246d
c7e746d29ff47c53343cee1b8be8fbb097aa3be1
23537 F20101212_AAAUBY kneifel_j_Page_041.QC.jpg
0e1cb4d32733811a7583ccace477b4b0
ae1c7c5a4c23af8acaece36b3e7c98401877d539
1996 F20101212_AAATXH kneifel_j_Page_143.txt
96bea1f51c441be90fd5f2b824900296
4fa18e0e8140622a9ba433d15b404178ee59740a
952 F20101212_AAATWS kneifel_j_Page_123.txt
0bc5f9aeca0ee24333cc710a4b4867ac
8701783a960b194ccce62ea0040cbc8ca07cdcea
1634 F20101212_AAASTQ kneifel_j_Page_184.txt
ff7de450e5b9c9707125c0860d0da069
b65a1161f0f098a1d7d75040d70dbc3ed050e5d5
49605 F20101212_AAASUF kneifel_j_Page_175.pro
b9d0af229d76eb23ed30ea034f41f84f
4980f1aa67aaa0a1510d1a9953f5eb52baf1efcd
21751 F20101212_AAAUDC kneifel_j_Page_063.QC.jpg
3d2a5e8b7575759b265c8c46094db995
8dc4a7c0690f272031c68c76e2e11793c320f20c
25766 F20101212_AAAUCO kneifel_j_Page_052.QC.jpg
cc6ff563f50e1ad51e017f897ae33674
2ba42db304360d8e9dbdfaf93750b999f329e563
6131 F20101212_AAAUBZ kneifel_j_Page_041thm.jpg
7ec90f458df07569c84df02579d8061f
47f3d6b056331ae03852f873ee09d83c170296cf
2373 F20101212_AAATXI kneifel_j_Page_144.txt
555fc337dddbf0c5684ef08b92346444
7de1532587475248c85d279ed573a9476c578fd5
555 F20101212_AAATWT kneifel_j_Page_125.txt
babd0bf3d65a9a008e41e314fd79bbcb
0f370101ed0406b6c909b3fd1db1d634c949bfb2
55057 F20101212_AAASTR kneifel_j_Page_031.pro
70ea05d227ccf352aeb48f7c93e8f3b3
5be5b3d43b64c410460fe4c94ac472454045ed13
73030 F20101212_AAATAA kneifel_j_Page_145.jpg
0559f5b99bf6f378b3bd617572acf92f
aa908580a29842080e65110653bdbc8f08ad6e3f
378337 F20101212_AAASUG kneifel_j_Page_194.jp2
ac098d525f213d180924fdc40715c296
b7765733c0d48ccb395889d67fc6851b0ecce608
5907 F20101212_AAAUDD kneifel_j_Page_063thm.jpg
6451f442733a1a1690c962b4cb27c8ef
1bc7cf3cdc3265b481c0e5e4b7ed6c6d87ad237e
6882 F20101212_AAAUCP kneifel_j_Page_052thm.jpg
c6003d94617d411eb8bec5a92d3b7396
f632da0d2fa84d7e9877e8647a64092391849ea0
F20101212_AAATXJ kneifel_j_Page_145.txt
d6b2872d0f8f3bfa621e0dc8badfa91c
a23466df53de11e7bf402d8dbf8e87e8483a8b6d
324 F20101212_AAATWU kneifel_j_Page_126.txt
38f30d35d5755c45814037c6409fcafc
b5dcc8112b957dc2db0c44a9f3ef226fe766a5ef
82893 F20101212_AAASTS kneifel_j_Page_086.jpg
9017dcd9fe4190357ebb4158ecbc45e2
75699df5ad65c8d1891adfe48ced14f3ca12bc7d
54494 F20101212_AAATAB kneifel_j_Page_146.jpg
b14b59793ad0cc8104e4efc98f2b60f6
ec57c8c0c58a548d3ace78d8e5a8685dda4e178b
26302 F20101212_AAASUH kneifel_j_Page_015.QC.jpg
df90180c0f75a5648070f73df310bb0d
f0ac6b0dd20156dbeb7d344db2c7286ae7d43b60
21319 F20101212_AAAUDE kneifel_j_Page_064.QC.jpg
76876a76a09d570e3285e18a196f3a34
614ef3159e74862377e687cff25eeafd86bf2091
25246 F20101212_AAAUCQ kneifel_j_Page_053.QC.jpg
f5c9b458ac88bed6a47542dcef671801
d71341859be911146c1c2ff6bbb985e31152c6e7
2333 F20101212_AAATWV kneifel_j_Page_127.txt
9065317169583ea89741e5c69e35f1fb
bba501fff2a785e5bcbf90e986d1aaf25c047697
4908 F20101212_AAASTT kneifel_j_Page_185thm.jpg
6d4282f6fdfbca0521692ca7e038b044
7e6ffab403f6083bf9bffa763f38741e6d171aad
74668 F20101212_AAATAC kneifel_j_Page_148.jpg
5883c09c0c3b1f3dcb40ac2bca771a4d
d82aabfd6b7ab4370f24ed24b693d644a6b2bd16
6062 F20101212_AAAUDF kneifel_j_Page_064thm.jpg
4daf569ddbf53ecb42cf3dcbd7143b52
eeb58a49ab633cb2fac5c7acd6fb4f95011642f0
6365 F20101212_AAAUCR kneifel_j_Page_053thm.jpg
2eec82543c382d2c552d49aa2051f214
a33689cbcb4e184ae571abba21a7f6f4a2ca41e1
1978 F20101212_AAATXK kneifel_j_Page_147.txt
779d11a0f9523e273d2760ecde7c13a5
e476f100de6963f704fbb05ccf979b9c29f23132
2166 F20101212_AAATWW kneifel_j_Page_128.txt
d649aec98e88e3b2aa7ab362715b91c3
bddd240bfef1486900d93ca34a9b1584fc3b66a4
516 F20101212_AAASTU kneifel_j_Page_117.txt
d82eea757c39462d3ea6d761080bc567
de0069b26e996e3ab7baae7e6128337c8cdd0806
86679 F20101212_AAATAD kneifel_j_Page_149.jpg
9cc996357cc11eaad9403284a615a948
c02258a5442e3ac7db5507e70299246cd4052274
9776 F20101212_AAASUI kneifel_j_Page_191.QC.jpg
6948eaf8f422a6f265ca9803c7b6bf77
6a6c8342a8648ecdeee7b2e476f392a9434c62db
24822 F20101212_AAAUDG kneifel_j_Page_065.QC.jpg
eae32578ed42f1fdc6ad527a804d0d0b
442a49150262fff8f9448cdbf00e0267bc2baecd
24629 F20101212_AAAUCS kneifel_j_Page_055.QC.jpg
75065f5ebb1c121ec64f988edbb14dcf
89953e66a41171a8fe61c510d258e723d335b55f
2107 F20101212_AAATYA kneifel_j_Page_167.txt
b43734c163f7b3e24cf52acc53da23fa
44fe8e1c39560c9bca3b2e2afef6223272bfee48
2259 F20101212_AAATXL kneifel_j_Page_148.txt
03f71e17fae7b176f6fbdcb3f13f1899
2096686877db72658a96ea9288359f3ec74f2244
2372 F20101212_AAATWX kneifel_j_Page_130.txt
06c17323eb8bb4864cc424342275a261
5e45b04dd6ee91485ccfbf36b109d64479668bcd
24147 F20101212_AAASTV kneifel_j_Page_057.QC.jpg
e918c44fcc1f8fca440ab0041f2c0e18
8199f86910c7c8833f3dcfb03c12fecadeb83f42
74012 F20101212_AAATAE kneifel_j_Page_150.jpg
59722d5ea8ab8e4335c7d0f90ab38c33
3a6faa0bfe1b034052f5d172102822c4551d5f89
6670 F20101212_AAASUJ kneifel_j_Page_112.QC.jpg
d83129777abbcad259e4884213669e54
8144529059631d67211e0978e4651f9dde35b8c1
6492 F20101212_AAAUDH kneifel_j_Page_065thm.jpg
48b5e68b57f61bed7cb65fdd900d7315
6785d1fea5d5859aca11b767ce29037904c25a8d
6238 F20101212_AAAUCT kneifel_j_Page_055thm.jpg
709a70efa44ce224e69341e9e1e55d0d
14a6f266ed580ecfdcc57ae1b57aacac3291c4b3
1989 F20101212_AAATYB kneifel_j_Page_168.txt
175e6e1a254113260671400497919433
9391af65d8074496f7ced63fbc09827fca381b35
2253 F20101212_AAATXM kneifel_j_Page_149.txt
831ea08472e6f27418afb6d08fc5ad4e
6dc6604be4742b666671c15bfceb876b1f3fea76
2170 F20101212_AAATWY kneifel_j_Page_131.txt
9434b4aaf02653824a447b728162cbb2
f2fbed41f581ba32704fd47997555b8c6abc7387
52759 F20101212_AAASTW kneifel_j_Page_068.pro
3ae1fbb8665c5e84222f05aaee6ddd98
9d518d8e01cf5e072cf6e37a4295ec530e7ed6ab
69458 F20101212_AAATAF kneifel_j_Page_151.jpg
79397752d4b49959a3bdec1f3822eb53
96a79af3f1472a5acb021c43ab8fa8b16856da55
47543 F20101212_AAASUK kneifel_j_Page_114.jpg
6a55e90233aef4727414170cdc9b13d8
986cb358aeb02d12aa3a5aa7c866cf5d1499bb3e
22077 F20101212_AAAUDI kneifel_j_Page_066.QC.jpg
7c0bfc85fa402a01710193d955684155
87b214f71c5d4cd534901689f5427f230eda20dc
1955 F20101212_AAATYC kneifel_j_Page_169.txt
5b4b0698686f273a552a0207b5547a4a
b4e55736ce5598faa698789c80ea4b43bf320baa
1770 F20101212_AAATXN kneifel_j_Page_151.txt
2f90664e60b47a36a53526e4134288c2
5a2bfb55daddaa7173b784a54d65fd6aa71d80aa
2397 F20101212_AAATWZ kneifel_j_Page_132.txt
5236e2e175ee93cd4da25dc1eb8be399
e60a2e6d8d0119fad2f067458d2959bcea5a200e
66478 F20101212_AAASTX kneifel_j_Page_069.jpg
17dbc05c04701afd250e637840c8b7eb
d26754e02392a16825455331004e49cf48e423cc
79700 F20101212_AAATAG kneifel_j_Page_152.jpg
fdbb4a9eff3824dd073d082f769effda
053865de6a998432604b7134a57dc2e6e92437e6
116242 F20101212_AAASVA kneifel_j_Page_181.jp2
097437301df72867f68be5397e5aa925
dace5450a6fe935ddf245b4ac97afe5aea7d96ac
25519 F20101212_AAASUL kneifel_j_Page_047.QC.jpg
247415ed1ef27e0f24ae8ce992b6aab4
e49dee6aa97dff9deb19b581994c6094262c4735
23569 F20101212_AAAUDJ kneifel_j_Page_068.QC.jpg
72ea44b03e51cbc2535fec4977d2c122
22ad323d5f1832a9446aad6057b654b709dda5c3
24791 F20101212_AAAUCU kneifel_j_Page_058.QC.jpg
cdff08c0b9de82d74d30f319e4e59988
4c96b3a9fb5e9c4bf2932590374937acc67bbc39
1998 F20101212_AAATYD kneifel_j_Page_171.txt
79e2514c464b8cc2ae0c6cd7a977ed67
27d4640fd610dccd7191b63b6afa00920f2b59f6
2099 F20101212_AAATXO kneifel_j_Page_152.txt
aff1196a6778d5bab72cc7d5226a89f5
5bbb6183c9e5addd3d07785b4cd0a72cbbbe7fa1
60460 F20101212_AAASTY kneifel_j_Page_027.pro
6e5585b6a6cf59490ad6f5c163cac25d
7d315924d29e704ac638fc2cbbce8df8c247ff53
65071 F20101212_AAATAH kneifel_j_Page_153.jpg
88bbd5eac6fdeac409a93e5b94eee560
e82f639982fe9fac6a370e0e48ee42873bb3ccdc
6817 F20101212_AAASVB kneifel_j_Page_085thm.jpg
3633c5399fb400b24ed6d50fa86e36fd
64e063ffae6a9206ff16e63638558da186eddc88
117241 F20101212_AAASUM kneifel_j_Page_135.jp2
50501e31e172cdcbbb54eae350374969
583243f76cd8c0cbf58b5ac0d239cfffcf4db82d
6627 F20101212_AAAUDK kneifel_j_Page_068thm.jpg
a68de8cedbdaae043af9bda2bc74fd8e
244e92495620bff6e35173ce9e8293f0a756e206
6361 F20101212_AAAUCV kneifel_j_Page_058thm.jpg
ec9c5008002a86596cfcc0409d213e42
77b7a0fa7bb5d75dc07152efec63aeb39f5a5e27
1922 F20101212_AAATYE kneifel_j_Page_172.txt
6c7d8ed41f2bd8e4f71aeff883de9b16
e5c971b699bcfefbb82cdad0f46e4eb126c56d0f
2088 F20101212_AAATXP kneifel_j_Page_154.txt
f5ff74bfea48b5d3b8b870fb795f3f93
6fbd212a95d9cabb5b123f5109fabc88d2bbd995
420 F20101212_AAASTZ kneifel_j_Page_194.txt
991c56b2876c881cb69e2d59c128d1aa
dbe8ccb0ed4f4567fb3b2f0e2dba91e63cb184e5
81026 F20101212_AAATAI kneifel_j_Page_154.jpg
b53be0041e3df9b3b337466482061066
14f596b522d37e083b92b902fe42e6924fb6b4c2
2456 F20101212_AAASVC kneifel_j_Page_043thm.jpg
17dd04133ad1fb266b8354d951c08e19
7d74b1228f63569ffde371dc935670cd3e8226d4
5728 F20101212_AAASUN kneifel_j_Page_200thm.jpg
be3db833fb4cf42585b6e6f3370c29e8
a3afe3470d88efccdd9d469fc12ae69769196088
21444 F20101212_AAAUDL kneifel_j_Page_069.QC.jpg
06c0ca7e48fc1088e5ad6bc760220d4d
577eebbcc4725ec6c435876627d213f5db68e66d
27227 F20101212_AAAUCW kneifel_j_Page_059.QC.jpg
771247c0735e191b56d1a4e292410879
b5a193be3372181120501a232fab8c9249be47fc
2028 F20101212_AAATYF kneifel_j_Page_173.txt
948fca7114a02c4dd79a0ac1610d9d3d
03b3ac3d63913e30c954aa16a755fd6c5769aacb
2387 F20101212_AAATXQ kneifel_j_Page_155.txt
cbf369b0190f41af24aa77a015ad3889
d3ba6377fc72ee45fbc95188f27eecdefd8a8e45
78580 F20101212_AAATAJ kneifel_j_Page_156.jpg
ef0b21b824a611d8c9dcb21178482f3a
3df6c1b3e16a6a0a8a08d0483273f9fced5f5305
78326 F20101212_AAASVD kneifel_j_Page_155.jpg
7079405d650d233592c78c20ae4d5bdc
51e85e36d51a5ac845e5c4376ba12758365dfcfe
53023 F20101212_AAASUO kneifel_j_Page_047.pro
58dac4204b6bf3edefbe43f90ce75f20
e119aeb435f527aeb48cee0713551d41f59d857b
27056 F20101212_AAAUEA kneifel_j_Page_078.QC.jpg
60d1d7e8c4e2d772c7d6ce9e530b5a3b
e01a2b4b6bcac156a20e18eb77e991f6469d39f3
23319 F20101212_AAAUDM kneifel_j_Page_070.QC.jpg
66047a955bd84f5ceec3752a0caf9045
467c5b7a58c86ea2d65fc2fb498d39c453917633
6838 F20101212_AAAUCX kneifel_j_Page_059thm.jpg
fcb2fb102fa948a7821a51a85133380b
e6729bad20ff29196022d6ca9bccb27d232466a5
2221 F20101212_AAATYG kneifel_j_Page_174.txt
fa2096b7ae5a35f2755bc6ce16f0d12c
ce5ff8f6ff73a4c6ba22cf691545a971668bb171
2359 F20101212_AAATXR kneifel_j_Page_156.txt
6011c31433497ce17f39fde0308fd041
fe427692e0f80b21938fa4264bdeb06e8f4bc0c9
78212 F20101212_AAATAK kneifel_j_Page_157.jpg
edd43f73865d585fbd0b07b0f88a8b95
c66b2ada59a8915a22e962f960dee51c18194e42
844139 F20101212_AAASVE kneifel_j_Page_183.jp2
39b19dbbe59d3e3692fbc80ef7caaa60
b1f7ee515918e06fb30b1550f8ed056f64500bd9
70776 F20101212_AAASUP kneifel_j_Page_168.jpg
445c7f750cd060709b190d4ed763a25f
a38114b42a817f0a2eaa85af020f01ef8001e294
26228 F20101212_AAAUEB kneifel_j_Page_079.QC.jpg
08bb3e6adfeac75025ad18366a278aa6
77448f610d407ee24659e5377c2eccf8b729ab05
6472 F20101212_AAAUDN kneifel_j_Page_070thm.jpg
c021a5e1dd3b68c6a36ea673881e9267
2c570d1b1a5301bf6e1b304b3222eb68fa48839e
25244 F20101212_AAAUCY kneifel_j_Page_060.QC.jpg
219f32329459a3207b67e1d27477b16e
75196526911e11a7725ee7fbeda45a3479595289
2656 F20101212_AAATYH kneifel_j_Page_175.txt
68c049d11152d03fe8aae16984299544
69b6ea653c3dded3121090e13889856df171da2c
F20101212_AAATXS kneifel_j_Page_157.txt
1763856c704f73e7469c4ee922f76539
0460ee63ec69328e272fad3770715f5c442edf66
78428 F20101212_AAATAL kneifel_j_Page_158.jpg
94bef7f026312d1de43b9fd0f3c33b63
d16c7ff83b81d51fa2032bde9ad99b838bbc456e
28907 F20101212_AAASVF kneifel_j_Page_194.jpg
3178703de93d2deff3001b2a0814028e
837afba6ab68ec045f08a1fc7b296d7405252926
6339 F20101212_AAASUQ kneifel_j_Page_067thm.jpg
88e1140ad4d7487b70d0e3ebb21d95df
bc57028ffaea85aade8365d736ccd8f1f510ffea
6684 F20101212_AAAUEC kneifel_j_Page_079thm.jpg
ba034a46b0931c8e181bcbaea7b2b54d
7c10bdf2ec26c5a39e7dcf13018e264dfb76293c
20605 F20101212_AAAUDO kneifel_j_Page_071.QC.jpg
801774e712999cfc2960e814b62fa6e1
262641c74438d07ea9188a42210d09873add6776
6565 F20101212_AAAUCZ kneifel_j_Page_060thm.jpg
3f17d5ca191256b7e5bdf6bcb855d49a
28351579f5e4883020c51233604a73a58d195d14
2210 F20101212_AAATYI kneifel_j_Page_177.txt
f8366630a8565a727c928ba6564f8d4f
bf09632dd0869b5f6af38d56dc967d8593c75769
2334 F20101212_AAATXT kneifel_j_Page_158.txt
9e316386716d50fcb5fc7594984b70d3
2248b833ab6d5251c25b35ff6490e4bfcbe98ed5
84224 F20101212_AAATAM kneifel_j_Page_159.jpg
1258bb3501dc0dd30531e99ffe7f6f06
812a533b0fce2a0ebe7e4168f67d283383d80e46
25115 F20101212_AAASVG kneifel_j_Page_130.QC.jpg
c8c76016dc0d5281d830cdda2e82a0aa
27c00e7a532aa430c813a774d4cdca6a861d1618
F20101212_AAASUR kneifel_j_Page_182.tif
a65f93505621964fb2caae1f34fc8b1c
a7c208f05bb4f69423a8b8c712b96c92632d2b14
67991 F20101212_AAATBA kneifel_j_Page_176.jpg
8c82a6fd23f17b432ea6110b754523e1
99903104644cb56abd194529bf896316953cff7e
26512 F20101212_AAAUED kneifel_j_Page_080.QC.jpg
f46f73aaba4927096468bc61ecd53b6f
8a8867ee7ff73ec6f3ec9f2dccb93f49812e700d
5628 F20101212_AAAUDP kneifel_j_Page_071thm.jpg
936155510516b5345e4b4ca86a3cff7c
ebec0ae6499ddf95d8983e4c0deabf166edae26a
2278 F20101212_AAATYJ kneifel_j_Page_178.txt
2717f9287fad853d49a727315af0f871
62fd50ae061f81288009d806525e76e28cb04887
2204 F20101212_AAATXU kneifel_j_Page_159.txt
940591a0aafc4cafe283977afba657b0
dd43a48595b8ccb325405d704cd3cc0eb4558275
77675 F20101212_AAATAN kneifel_j_Page_160.jpg
42217d048c5d2c4b87e8ae685b3ecfb5
86fdb52340625516415d01d945e38f4dd83d8a0b
7242 F20101212_AAASVH kneifel_j_Page_126.QC.jpg
91b19d35b95cfa65ddfcd1c42874d666
6016a1b7598d983b97ee802a0b6ee67df3a55ca6
3953 F20101212_AAASUS kneifel_j_Page_196thm.jpg
5184f0c5f6b24c3fbb04f3d9a63895dc
01609fa07311d55eca92e36c5e5d03a231dccb98
85302 F20101212_AAATBB kneifel_j_Page_177.jpg
0bd4563f98ac8de96ba0e34a0b0f6e2e
98375225e8001561e93878a51e2fbaa9004badd8
6864 F20101212_AAAUEE kneifel_j_Page_080thm.jpg
4f82ad842db9a17aa49d472ba295edca
d0947e56802bdf5dde9ec675a6843a7e58abc1e8
20043 F20101212_AAAUDQ kneifel_j_Page_073.QC.jpg
2c81bf9b159685c083df1cbe2b29851a
d6560f07a1f77be53a35d3ff1b4982c62364d7bd
F20101212_AAATYK kneifel_j_Page_180.txt
c4e6194caffabd718f45b24bb7174b5e
7e5169776446135c4ffe8ed16423c85d29f91e8a
2002 F20101212_AAATXV kneifel_j_Page_160.txt
1f6e2c4340926df613ede795d1442ded
c2a873340a05228666241d1cced3091c1cfd13ca
68550 F20101212_AAATAO kneifel_j_Page_161.jpg
0f0a905e2ff387f59db0db2b5ecc6027
ac0800f7c74dc1ea64629fd3e77979f9c1c77120
2693 F20101212_AAASVI kneifel_j_Page_116.pro
2fa60fa12c38e615641ebe15396795c4
dea51b1e1d6f30bc8334399665a0b431345464d7
F20101212_AAASUT kneifel_j_Page_162.tif
1aee8505c029b091f9b81b4d2680216a
cf34b81ffd6b46ee5fd6cb8111c957e1f427af47
77572 F20101212_AAATBC kneifel_j_Page_180.jpg
cd114c53dee4780b22dd14025f764805
3132a074247234506714073f06b898fedbe86de1
18535 F20101212_AAAUEF kneifel_j_Page_081.QC.jpg
74d02b06ad2c76ccdff304b5aa04dc7d
b67778efcf37e249bb321506784fe5eb27ae7a4e
5512 F20101212_AAAUDR kneifel_j_Page_073thm.jpg
86fe1dc575e5c77339e0b8d91150f4e8
f37508a7decdb6cd68ae3961607eb43e1463c473
2003 F20101212_AAATXW kneifel_j_Page_162.txt
0937271243131d49a33d42d68bd5077c
26613ca8c08ee1b179485641fc8217fca92a7ef6
76967 F20101212_AAATAP kneifel_j_Page_162.jpg
b3120e88fafe061bf9876a97c75af508
2bc0ecf3f40dd059a2b80d21b0b2d158ad19f4dc
40908 F20101212_AAASUU kneifel_j_Page_199.pro
2e1e602c6e331356ae51a348595875f9
6cc77fc0dc8138acbe87514c844b6aa9446ac72e
74260 F20101212_AAATBD kneifel_j_Page_181.jpg
31865cce5194dae2c3c04be3e889d315
274c1b55b36c2ca2321ef6005b286685ce0db397
20304 F20101212_AAAUEG kneifel_j_Page_082.QC.jpg
e396af938c7df99440b2e92678b901e4
39aaa03e04f6536b23104dbea5b1ceb1e7ce3afd
24808 F20101212_AAAUDS kneifel_j_Page_074.QC.jpg
91851ee8f68b009b10f30c45ef807b18
597ac988b4a6df0c2a0085436e6393d6de86eeda
F20101212_AAATZA kneifel_j_Page_201.txt
3f129acd097543546848a08cddea7f35
0a785f41af96f31e6ee906d9cd1c5e9eebd67c31
2225 F20101212_AAATYL kneifel_j_Page_181.txt
8e99584ef886c62a88a35e4db8c50d2a
1f3b35e310e59830776d9017f5d7eb7a758fac56
1705 F20101212_AAATXX kneifel_j_Page_164.txt
bad4c209d947492bbbb22fcad30921a9
d84cba5f62999da907beb68effadcc12e55ef1a4
60639 F20101212_AAATAQ kneifel_j_Page_163.jpg
417b155957b0a2dd541eaf603e6e4439
4f4b91e81921dfa86a2096a46f55440e94c1d221
2176 F20101212_AAASVJ kneifel_j_Page_072.txt
5793dcdee794b4fda06fa93a9d0aceac
e43a060816ee845224424486e208e7c1a355123d
1051984 F20101212_AAASUV kneifel_j_Page_102.jp2
ea53c5f7f9e278773afd990e87fcbfc2
4a2bc0f095a38c02ba2c797487b920e0aa189ffa
58824 F20101212_AAATBE kneifel_j_Page_182.jpg
bbf16571611d9f5f49490b7357621e6e
8b5b5a7f029a66e3d7cc5b497e2c1bb857bc2b1d
6009 F20101212_AAAUEH kneifel_j_Page_082thm.jpg
af5770a80f79ed71e4a837863d40c585
56cee7c92389df165511fc5b0c9070d119403240
6632 F20101212_AAAUDT kneifel_j_Page_074thm.jpg
b67fa619d7d56f6136eb669437f2d207
e1ef93a95a9e6ad11ba92c7da7251ee7cfb2e4b7
2261 F20101212_AAATZB kneifel_j_Page_202.txt
916914bb63d9d79db191b9b91e510312
3c0881bf201f418adfae7534c06a3777f1e1e6d1
1780 F20101212_AAATYM kneifel_j_Page_182.txt
f915b8c8cb4418f66d28539da0664e4a
0ad50f7f10fccdcd5b1891a78b1e1d1bdcd66b6f
1894 F20101212_AAATXY kneifel_j_Page_165.txt
823329caab0cb363f6607a19388bc0d5
9269a067a5eb144a2d69a2f8a5b30d96ad42375a
61317 F20101212_AAATAR kneifel_j_Page_164.jpg
f94159b119c90234a69d3da936502f1f
64d1e476783ac83262a93803b23c0fdb1f044a5d
22726 F20101212_AAASVK kneifel_j_Page_110.QC.jpg
507dd351f37dff567af8bb4205b67fca
30197066b7a3d7cc7d95fdc675eb389d38b1e573
5509 F20101212_AAASUW kneifel_j_Page_122thm.jpg
216050a8770c2bca4e56822db4e39ded
63e5a093e080c71069b3705daf8619e2f973d6b0
75620 F20101212_AAATBF kneifel_j_Page_188.jpg
b8184c4f3eafbcb37aea8ebe42185fd5
788a82f8b07c4475430b8bd156b882b8152ae340
25854 F20101212_AAAUEI kneifel_j_Page_083.QC.jpg
703b1fb51020fa7f086262716accce8e
cbd87f8f5522b72d2ebaf14b90cc87d4c0104029
18244 F20101212_AAAUDU kneifel_j_Page_075.QC.jpg
09032f2a8204e3ddd62b2fd39ce13349
03fafc28ec8abcd9febb92c9eaa94d7cb64a5eb8
2197 F20101212_AAATZC kneifel_j_Page_203.txt
6053a62ba42ad54a112aac90fad8b5c5
12d6f1aea371884201c49cb93e865a63aa147104
1777 F20101212_AAATYN kneifel_j_Page_183.txt
fefbda218c2508244a28c946117cf4f0
cbb550a406cf08e5a83c83824dcdf49a6a853bc2
2236 F20101212_AAATXZ kneifel_j_Page_166.txt
1222e54093f44994f49618c2582ae846
654c8a934092cea7a4b2d6c8f4ac41bb0e5ba3cf
79125 F20101212_AAASWA kneifel_j_Page_010.jpg
6ef3a0bfce6e4f4164f3fe42cec22a67
6fe3c3c51cbb9682adf186150089809184e4f294
78802 F20101212_AAATAS kneifel_j_Page_166.jpg
44860447fbeb6b3a23fa495973177ff1
1f52f5797759de94c361645d4ced852adb5b4de9
59479 F20101212_AAASVL kneifel_j_Page_158.pro
b20107d829f7a8d0a053dddb13b377eb
2a8b0c086d87489f1e38b645273f00339a149c9d
26411 F20101212_AAASUX kneifel_j_Page_089.QC.jpg
8dc787fb58107bc775045aa80dae0086
766dbf99a0c4b0967d054052d66f6f51fa6e05ea
44281 F20101212_AAATBG kneifel_j_Page_190.jpg
007c4fcd5b9e5db9a83916b98405a6c3
31edb11bb36128a238da929d20cc7830f976521e
6721 F20101212_AAAUEJ kneifel_j_Page_083thm.jpg
1b6f054e39bc4ef0843500c4dfa24f8c
332be90ebbe917dca88c8d40a62190f67d070fd3
1993 F20101212_AAATZD kneifel_j_Page_204.txt
4eb8184ec328083d59bb7f61c01685ae
b2fbf50fa517a7a7ba9be7ee21556cacb0f60ee8
1528 F20101212_AAATYO kneifel_j_Page_185.txt
6f9df4f2cb1e1edc1c7253b848080ce4
9b5981cbaa7960afb6593caa26607bf2aac64988
27874 F20101212_AAASWB kneifel_j_Page_011.jpg
7a9517f3f31b74ba5c923a4ba018e914
7afe113e56bc76ca009fb1dea0780d51cf9079d3
123486 F20101212_AAASVM kneifel_j_Page_130.jp2
b7841a868769c70b69156b8f18266cdd
a787a59ed54c688fdbf96f8fb4dd1a3735ea284f
68222 F20101212_AAASUY kneifel_j_Page_147.jpg
48b74e8a97cc5d6c469a7135d6d9b670
af88747bd7a06031fd9d11ec2b291d04564ea78f
34313 F20101212_AAATBH kneifel_j_Page_191.jpg
77116d4187eae4b64c4a9c38f1e69916
53d086fddc6384bc77756042c64f73e9f3369711
25497 F20101212_AAAUEK kneifel_j_Page_084.QC.jpg
c2996d34dfd40ac999ab81b852775790
e33e6eed64251e94a62fa58ed5d869c505576032
5301 F20101212_AAAUDV kneifel_j_Page_075thm.jpg
006b0859a94fd19005cab6f5f9a8b503
00fa228f9d6ab59eb491e3c03505a35a55e70f41
727 F20101212_AAATZE kneifel_j_Page_205.txt
996ef3ee8a7252a2828cdd89185530fe
365691ce21911784456238e3c614db5429b6a736
2240 F20101212_AAATYP kneifel_j_Page_186.txt
e208a266b1087114b6c5692afe111d10
ca49ba51fe8b776b369231571b19adfd06fd9d43
61549 F20101212_AAASWC kneifel_j_Page_012.jpg
1007f5172cbae053ff771ecc3f1a8f76
19daceeed20aae6521c841a916a2605fa46a433b
78021 F20101212_AAATAT kneifel_j_Page_167.jpg
f0ac9c86013b1229914910fa903e687c
35e521c995424948af99741d7139e32ccfdb58ef
21454 F20101212_AAASVN kneifel_j_Page_197.jpg
ee30560295ef69bf2f7b9b5fdc2cb42f
9bfc561ae6f571ade6c5e3dbf1df8e1588ae7569
83259 F20101212_AAASUZ kneifel_j_Page_035.jpg
d23ba5be9fdc2188fc03d212351326f8
1a0546d795870b2ff4277902df37050a1acc0849
32557 F20101212_AAATBI kneifel_j_Page_196.jpg
e47ff7ed1e1a6b2128089637c798bc84
a1d239b705328910a3308b971b43e7e9f8800f2f
6369 F20101212_AAAUEL kneifel_j_Page_084thm.jpg
7d27776488e3ce867bba9e4f29b30b04
1c41457fc3605376a11e005ed9bc95ea6429993c
22374 F20101212_AAAUDW kneifel_j_Page_076.QC.jpg
737a232a4fa05c5024c90bd568165848
7093c48f9004030d20e80245ad41008c4a295823
F20101212_AAATZF kneifel_j_Page_208.txt
5d813278037d0045de7baa155fd1d24c
e9ac57c64a3c67123d0a3a71ce6737428559c618
1743 F20101212_AAATYQ kneifel_j_Page_187.txt
455ab0ac4fc0fabdbf811c75d646f3bf
9964c598e89d84b132f39b478654bf96bf056fb9
82545 F20101212_AAASWD kneifel_j_Page_018.jpg
06bbeddc9bd356529b9967a887be73f2
e7ed7829495afddaabf17c888fb2d96bab884396
68294 F20101212_AAATAU kneifel_j_Page_169.jpg
9139a023ee380bdeb2caf9675c7a5114
c4e88dfca24a92bb225ae48ae51fdd420fee0e43
1031316 F20101212_AAASVO kneifel_j_Page_076.jp2
c8b8f021dd8fe738b31237046bde7c0b
76b0255446b5c72de7f542fc9972e53e1d017094
24530 F20101212_AAATBJ kneifel_j_Page_198.jpg
98080a2357eeddc4c85a21ad1e510e6e
ef95639b446ba9fa3da1616a2a3bbeb863e0debe
22094 F20101212_AAAUFA kneifel_j_Page_096.QC.jpg
01331907d27a306381067e2625163371
1f7c7bd65eeda2b459dfc8040806118bdbed05b3
26948 F20101212_AAAUEM kneifel_j_Page_085.QC.jpg
86d707fe645be367e3a47fb017bcb045
4417722891ac260aac668d016867d9cdf54eb2f5
6236 F20101212_AAAUDX kneifel_j_Page_076thm.jpg
d811195ef4016836be51edd5546e4288
93d349e3cae13fb0d1035a7773f1dfc5108353a5
1285 F20101212_AAATZG kneifel_j_Page_209.txt
f9740b47ca2347529cc2bc308eba2ff2
9ed612fc044450a20e85860f18534e0877836739
2271 F20101212_AAATYR kneifel_j_Page_188.txt
d6e852cfcb85d4769dc973c38111ad97
e8fbb7ad09adc29ce9590af2e46e887f6aa6a9ee
98541 F20101212_AAASWE kneifel_j_Page_020.jpg
8d182b00aeac10f0f0e39d3170437f82
876382d54bb830e429659e9adea916a0d384babc
86489 F20101212_AAATAV kneifel_j_Page_170.jpg
0fe61af651d4caf2ffc4482a531cbd9e
fcc1608776f94b22d90d069c8918f8515a846e9b
92218 F20101212_AAASVP kneifel_j_Page_021.jpg
2c44035da08269dada3ba3b3d0614b71
38c1a2d7c676d79fb35be5b640d1a19783a517c7
51057 F20101212_AAATBK kneifel_j_Page_201.jpg
55f24865f9b4c03bc633ce982d6a83b9
599d2c996632fdd87de2f9dd17ae57f4761c8372
5999 F20101212_AAAUFB kneifel_j_Page_096thm.jpg
98906ba97550cddbbbfd1ca27e9312f3
801d2644040ccb0edf5042a1f512054e4763c154
26296 F20101212_AAAUEN kneifel_j_Page_086.QC.jpg
1738996eb1d28cc6af57d444d65b008c
27413fcc602a23f108b21021434d397a24abbb4e
25641 F20101212_AAAUDY kneifel_j_Page_077.QC.jpg
87ce4ef30b98b70b4cd6ab88a9f554ae
1ba9f66bd034944611bce5d13ac73e7eefd580c5
2464 F20101212_AAATZH kneifel_j_Page_211.txt
fc61f32c86b343a926bfc937e8a3e920
3a0a81c2abece66e872e346b48376d36908d0e9e
2330 F20101212_AAATYS kneifel_j_Page_189.txt
1113b7e8b4fae4e342e2438732b8db0c
4dce7ddcb134896a1d69fced20de2abf9cb46fa6
80401 F20101212_AAASWF kneifel_j_Page_022.jpg
1837c8b2bfab330dcbb1352eea431c7b
d1b7bd50c02cf8831bd77d7c6b9b83a13055b33d
69689 F20101212_AAATAW kneifel_j_Page_171.jpg
86f87a5feadf1d3cc2b9454510ba51a1
a2223e5db758302eb4f8d4f7c9ea2c1ca4a9ead4
5341 F20101212_AAASVQ kneifel_j_Page_056thm.jpg
427703e055c830b6bcb7d5019e094d21
ecefdd812b5f759a55c9a99a644f5d47c375a0b0
69165 F20101212_AAATBL kneifel_j_Page_202.jpg
79c315c16444588faf6a4a87eb2156dd
02fbbe75a607c2857d5dad115d4e16d25f3d2243
21963 F20101212_AAAUFC kneifel_j_Page_097.QC.jpg
4ae9d9388ef2d7aeb4785a3fb7b59d0f
0ca107bf296abffe7164ff8fc67ba89771d63c98
6780 F20101212_AAAUEO kneifel_j_Page_086thm.jpg
8ee10ede8fb71c5af9237e158c3780e6
767277e620bfbf101c8f180b53ced317fe654fd2
6506 F20101212_AAAUDZ kneifel_j_Page_077thm.jpg
f323ade1861df4166de5c215eaa0748d
ef7ff175373513476b5a59b1dbe337dfa57cefba
2510 F20101212_AAATZI kneifel_j_Page_212.txt
dc341fa2ba5858999a08f885081d6ed3
48e60f4ca64a04c4f86873945f7b8b04728147c3
379 F20101212_AAATYT kneifel_j_Page_191.txt
f3da18a2f338c2573626e110864c037c
9845b29499121ce42095e2e16479c30fd7f97ba0
92999 F20101212_AAASWG kneifel_j_Page_024.jpg
24c2761fbc5d886d96f5b108d76fa031
dc684c8661e6044fdab50ad11d0807598cfc7030
70741 F20101212_AAATAX kneifel_j_Page_173.jpg
154738817907076a244aeb47a7a8c31a
53fa3f10ab1de76c07b420e9223777636ad38e64
322494 F20101212_AAASVR UFE0022518_00001.xml FULL
6c3acccb37b19d8f61b459310fed8831
ce7ec84d9110d0ef4fe7a094476e7cf99d8c3eea
1051968 F20101212_AAATCA kneifel_j_Page_008.jp2
02d1cb94eec24c74a2c0f7aca1a85867
0dcc1f02eead7a9b26080f01330e6ec4d9241adc
61658 F20101212_AAATBM kneifel_j_Page_203.jpg
afcbca614addcaf8e56138830e349ca7
6c3307d08672bb6f885274cd2ab2deb8c051588d
6029 F20101212_AAAUFD kneifel_j_Page_097thm.jpg
145ce5ec93bfc578595abbc18d86b8b6
5bfbca8f08d2c7d767b8f920faac219c091f7ce6
22345 F20101212_AAAUEP kneifel_j_Page_087.QC.jpg
bc18510dba05c4d9dc91d59ba160e506
7d31a9e68215b2eeb4643411d9eb1900c28c686a
2636 F20101212_AAATZJ kneifel_j_Page_214.txt
f42ee0ff6b936dd32e25532b5840d95d
50d8e34724d57f0f3672464825647ff4fddaaa88
F20101212_AAATYU kneifel_j_Page_192.txt
514a9e69fb640364c14c079861165ef5
c615d505ce40bedf0a74a89a8beec986c8731ffe
108497 F20101212_AAASWH kneifel_j_Page_025.jpg
343aa6c8d8684f0c1917e006e4d99d18
4dde38a238bc490310eec889b0cf3852a45a34df
86690 F20101212_AAATAY kneifel_j_Page_174.jpg
3404f3baeddef495c71b91ff81a64ecb
f748f6d4efcf45b79ec6b73edbe9fa68f3fdd1e0
1051980 F20101212_AAATCB kneifel_j_Page_009.jp2
fdf6fa45c1af8361c78dd8c4a54c79f6
c86df3411efc6c4726d8baefb00d43b6a8afe0ab
59491 F20101212_AAATBN kneifel_j_Page_204.jpg
d71fb6a946dce13f5d1413f4724ac83d
fe4892f98f148f21e4fdf4143b5938da7c8f8009
20796 F20101212_AAAUFE kneifel_j_Page_098.QC.jpg
ee49dffdc36c61134aaa46500190e00a
3ea5ab4f3d91a090d5f8adc92441870ae6314da1
6013 F20101212_AAAUEQ kneifel_j_Page_087thm.jpg
80cf1dae7e3c3c3f3053be525bbe5648
8b4d75ea09d8d800aa9ed262e9339f695c980ff9
663 F20101212_AAATZK kneifel_j_Page_215.txt
8a8d2bd7bfe662d9d1be7962d281c5cb
68beb8eb009016f05704f9db3dcd0d520fab9b73
511 F20101212_AAATYV kneifel_j_Page_195.txt
e3e01d0bd041e226be1f3823f8e529d7
dba7899ab3855823d05534461d9cf334a840d61b
81574 F20101212_AAASWI kneifel_j_Page_026.jpg
bc8c980748f7d85806d12cdf410e71aa
fc49946eef740f5f3142ef5a8c85c6c9ec0a140c
77775 F20101212_AAATAZ kneifel_j_Page_175.jpg
b6f58025370c5c92e3b3c8a51b6ec5af
8c229428be27d1901622385bd44add8660e7a48c
F20101212_AAATCC kneifel_j_Page_010.jp2
d40113c370390533ef0fc96e03bb3a37
fb0bc24fa19b08ee122204a92136c48a2f44588f
25833 F20101212_AAATBO kneifel_j_Page_205.jpg
acb8ff19d9f7f05251ace38013a3b4ec
497318e929b78cbc716ace3bbaab4cab18f834d1
24939 F20101212_AAAUFF kneifel_j_Page_099.QC.jpg
bde7a3d51291f42e914ebe496deecb9e
311ed523d42fcf170be1163d2460f56dff6c7d1b
22912 F20101212_AAAUER kneifel_j_Page_088.QC.jpg
dbcbd9b1fc93d550d8402356c534c9cb
f479038278d76a636503ac293efbd7bff4e053a0
2132 F20101212_AAATZL kneifel_j_Page_001thm.jpg
4b2a4e90cdc4d67de5982e4fb27ac3c1
964c6c48840d2113c1f4f4186c4796cb60752c91
603 F20101212_AAATYW kneifel_j_Page_196.txt
9da23bd42fc251a5a8d647c9750716cc
ec79aaeaf7c1018ccaf17f70472a7005dbbf4fcd
92111 F20101212_AAASWJ kneifel_j_Page_027.jpg
6e87c68f8837f5b29ea640173d91be65
a7191907c8c8474a7a4a63a2bea432959e37d0da
9898 F20101212_AAASVU kneifel_j_Page_002.jpg
51e5ad57291b9155474e95b7e582a160
344b2cc5d4d99f92f5ab4b96cc06843750a66de4
560645 F20101212_AAATCD kneifel_j_Page_011.jp2
ad2d6ef09bd43ac15b76fe767aa30421
9554acfd7191c1b5bbc884c14520448763e5e5aa
67524 F20101212_AAATBP kneifel_j_Page_206.jpg
a36ed75e93899c34a9fed5ef9546f8ef
76ab4771a7107b796e4570b8f98700fea4af05a8
6726 F20101212_AAAUFG kneifel_j_Page_099thm.jpg
e0afeab9844e14bef494b604cb5d88e8
5509160714a73534e2c90382fddfbefa6a34f455
F20101212_AAAUES kneifel_j_Page_089thm.jpg
a071e662e95d095ae837416f34ab2716
a3e09c66d5d2d0d7c534a3e7e44f971f7896cff1
268 F20101212_AAATYX kneifel_j_Page_197.txt
a3ef054670ef48c9974d901df6c3e3e2
42903c88ff446f97dbc40e09393c9c2c4ea296b0
11085 F20101212_AAASVV kneifel_j_Page_003.jpg
84edb06b01a5c48cc29cde08f43021d6
1e5a3802f97cfcd75ef14597877d571e8ae54a23
41426 F20101212_AAATCE kneifel_j_Page_013.jp2
834be3419b5e414799e137dfb57de7a5
b281f9c6b463c22ad0cbc64f2a9aa2aaeed52362
82949 F20101212_AAATBQ kneifel_j_Page_207.jpg
476b094481a621d4f085c8e091a082bd
0e586040c0260a3601e6ffee313b82b81ef48df4
7023 F20101212_AAAUFH kneifel_j_Page_100thm.jpg
89988736c5e63c19235bb6d83abc3090
4bf780615202d69d342fae92f1090680434f2e1a
22498 F20101212_AAAUET kneifel_j_Page_091.QC.jpg
3a08c82196e0b23a5517cf5ba660de15
50f7dc2d3c74c8f8a57140620d519f4ddb9af264
1609634 F20101212_AAATZM kneifel_j.pdf
e231069fe69e7a856b5dabb8e17cbf9a
8f7f52f10553a7e3e5b009778a606315571b7bf5
360 F20101212_AAATYY kneifel_j_Page_198.txt
128e29f639eb1c0db3338aa8153f904a
c655e8b49b6f9d110bde59ce20d2d3941ac9cfa7
87186 F20101212_AAASWK kneifel_j_Page_028.jpg
f5233f5e64b0b2a3cdafe2d188170fb9
d5a71b5ab66791b3b5e7d69c62e7be0fe4a3186d
20641 F20101212_AAASVW kneifel_j_Page_004.jpg
b2289a5dfb642c88af90f3e497f56d32
8abaf1bfe36fc4ed9d9f8491521c027448f6262b
1051953 F20101212_AAATCF kneifel_j_Page_014.jp2
76a729314c4523e99bee67b948b50562
c7a99347466c4d307d8c84ce8775f24d291bba7a
83755 F20101212_AAATBR kneifel_j_Page_208.jpg
e68f596928f7b22a4da6a880886a21ae
d9ca4cdddc6bfaa5b00d9525bd027b42a57e263c
5608 F20101212_AAAUFI kneifel_j_Page_101thm.jpg
7e8ce6189296072fa89f7fe23a45b5b3
99433f0ced5ba4ba55c7018931a692a54d905265
19750 F20101212_AAAUEU kneifel_j_Page_092.QC.jpg
0230c1db0f34922670fedcd0814b0f39
4fa6c7144bfec96c94d1ae98dc4afe1fbf822936
6809 F20101212_AAATZN kneifel_j_Page_001.QC.jpg
f82eba761d4c729f07b72b66257c7c7c
a8dd68239b3835bdfc69cee747cc410bd76d648f
F20101212_AAATYZ kneifel_j_Page_199.txt
157f438109fc1ff4747c563d82996116
4bf6dbe876349b30610aed81b6deba69ff91c009
87442 F20101212_AAASWL kneifel_j_Page_029.jpg
058a0a6af333354d56963cc07564c771
584a0178555a17528e5dec43141a5877ca792394
82967 F20101212_AAASVX kneifel_j_Page_005.jpg
4c28258ade18509d164e7d97af6bb363
47dec113c45fce748e07f870210e14fd971c5294
1051969 F20101212_AAATCG kneifel_j_Page_015.jp2
c975d0ad7878561e36ad036ef7d1816c
a16adb659c8d5f2a9ff647726c1a2b1c717c2d5a
83198 F20101212_AAASXA kneifel_j_Page_047.jpg
df90af31cc2038c5e2aed515c3e813e2
8a2cc2a043001d5d5d0e1c9c4b3672e2b3c0f907
45566 F20101212_AAATBS kneifel_j_Page_209.jpg
0749ee8503a1f2e08f107efab14a5339
9c790a909a712f9a1a74344db34c3ff1c3fb9df9
20237 F20101212_AAAUFJ kneifel_j_Page_103.QC.jpg
56093fa43d013d2e4d51d2d3ce37c63b
97363b331f9aa67fd5434213cc3ff835f5000551
5941 F20101212_AAAUEV kneifel_j_Page_092thm.jpg
3e99c53cde597b493d95f9cbd1d66c7d
174daefbf0dd6a91191e4c989ee8630a991c6335
3217 F20101212_AAATZO kneifel_j_Page_002.QC.jpg
df62fa0e2b7e584ebf180e77bc832ce0
98ba3fa3a6ec7232fc42f1b7c6390c7c7f4af64a
95092 F20101212_AAASWM kneifel_j_Page_030.jpg
c35883dd72fb1423898b51fd73a7fce9
2da2fbd59f9152bc317fb60bac79fb209cf5f18e
102226 F20101212_AAASVY kneifel_j_Page_006.jpg
1fb9a25f4ca3f5334c5bdcdc32ce58b3
ab47a5b17557eb23ed9f2623bed3ab95bb93b213
114995 F20101212_AAATCH kneifel_j_Page_016.jp2
dd253595b07d92dec4d8f2634208bd12
ca7eec19f73284c5814f05e02a9e412b49940dda
74124 F20101212_AAASXB kneifel_j_Page_048.jpg
bd15d83be6c198cbec08a0681efb7410
0b5ce66a98168edae0bd7a4c282007a2a2fee09f
26713 F20101212_AAATBT kneifel_j_Page_210.jpg
e76b38620f4b4bf12361600d85d38c7d
3c538b812aadc3ff6246f56a665296ad4c53a3da
20650 F20101212_AAAUFK kneifel_j_Page_104.QC.jpg
39d950d3a142233abaf1494c635508ce
3d2bcf2e570f85cf2b023a6a247384af7cd893d6
3647 F20101212_AAATZP kneifel_j_Page_003.QC.jpg
09b6a697dbff76a7801b07e712414d64
0b98e1508b7a02f18c6da65840ab7624e30ed987
83160 F20101212_AAASWN kneifel_j_Page_031.jpg
7313ab5b2291b5d28bf560fd6b316895
0275d8e858d0c824f9a975936097eead952f1140
76028 F20101212_AAASVZ kneifel_j_Page_008.jpg
8040c7405106124f7ac94c9426bfc0cf
4f33baadc2496c4de5cecface87243d2334632f2
1051959 F20101212_AAATCI kneifel_j_Page_017.jp2
ff1f9187f181f21e294ec7099f6595e0
45ed3ce259b79270995702e7a68a9e05e0bf6b99
81939 F20101212_AAASXC kneifel_j_Page_049.jpg
455d5e79d71d1afba90ebaa22e0de2a0
26429289beac98d697423c0c36d1c5bb065df55a
6143 F20101212_AAAUFL kneifel_j_Page_104thm.jpg
20c91221e759694beb7ddf4acb0be7a4
a584c4332fa61b430b43111c352d40997e15994d
27339 F20101212_AAAUEW kneifel_j_Page_093.QC.jpg
5d1020af87fe82bcaa6ec67165133266
d95c49d4fca4b5608961a3ea169de7486c78e198
1614 F20101212_AAATZQ kneifel_j_Page_003thm.jpg
e9ca2a138e6cda3c2431aa3552193a91
e84ef8466074c8d66a3f983af847178e2d593df0
1051962 F20101212_AAATCJ kneifel_j_Page_019.jp2
1a03a85d449d49804d1ed0d20648ea68
a46bce830d9b078b36f72272685f3439fcf72eb4
83266 F20101212_AAASXD kneifel_j_Page_051.jpg
0846276deafd0dadad0a3e3d83690d99
6e2c8e2fa27803a135f6a57a73a339e80d9f79ba
80531 F20101212_AAATBU kneifel_j_Page_211.jpg
e0e040435f7c46f0bd059783fc1ae6f4
61897b594efe069d86babb417d15ebdc07443463
91415 F20101212_AAASWO kneifel_j_Page_032.jpg
eabcfa723a581999275d84dc76556b33
6bdddc7fa71bb56c75afbe8849f1ad3e124e5ddb
8758 F20101212_AAAUGA kneifel_j_Page_115.QC.jpg
70ec33ffb56c967ae4aa6c5b11bebde6
e3bef68dc837a9649448d01a31c549006be826bb
21919 F20101212_AAAUFM kneifel_j_Page_105.QC.jpg
d2725d8328d9d2f2e83dee6549301763
6214cebfc1627de39691b474a565aaedae2e14bd
6776 F20101212_AAAUEX kneifel_j_Page_093thm.jpg
093497cd61a8bde31e9a4a48dd8a1d1e
23a79aa12a98db9dfa343b70ec78a3f9535a4c63
2178 F20101212_AAATZR kneifel_j_Page_004thm.jpg
fbf96573a99c58f3e52c8406bb4a07fe
ac5c04b58d008fddaef6b51760c7f1e89889fd6b
1051978 F20101212_AAATCK kneifel_j_Page_020.jp2
82683f25767cc1f49b4330400d5d0705
56eeae16ff6ab9da254d378b0ec493829ba544c5
84852 F20101212_AAASXE kneifel_j_Page_052.jpg
5eda6ceba3980146871be8d21a526763
cadac632375d24cc0c18e6654ebe320e6a87a545
85820 F20101212_AAATBV kneifel_j_Page_213.jpg
43eac0eee042e0f8cedb0e5d2b5d73e1
64fc541d97f216fc8c2bff8d4f9d0cff8c03d19a
87762 F20101212_AAASWP kneifel_j_Page_033.jpg
8707e4a3bb4ff1f54ca23977cf90d9e6
a1c1c568ecd75dd27bd6f4196ae613af9328a0c7
11525 F20101212_AAAUGB kneifel_j_Page_116.QC.jpg
79265c09f4849d633c659ea0549db880
004c86176ae2d59ce35e1a8c4e1167bf3c335473
6271 F20101212_AAAUFN kneifel_j_Page_105thm.jpg
6ae51c6efe8b57beb3bc0d1962ae00f5
9a30cc3d97c07510c4809b08ade70ede5a8095c5
19021 F20101212_AAAUEY kneifel_j_Page_094.QC.jpg
c2f08ca54f46b700e722b8724ed33985
f2ba9ab30cc71b146caa23dbdb4ad2b8abcd5653
21982 F20101212_AAATZS kneifel_j_Page_005.QC.jpg
206933fefe4811ce302f27ac3f418303
26eb96aeaff98ef6df880812b4e58cb9fbf06f25
1051888 F20101212_AAATCL kneifel_j_Page_021.jp2
e05dceae201b071a04a9e6df022c80b9
3965361f03ceeeb27fa1b44ae5dd61a29a73eb4e
78822 F20101212_AAASXF kneifel_j_Page_053.jpg
9a7e60edf286557dc4bfc8065ac9f9cc
c82f324b2bdbc63af2e3fd79f7754daca352aab9
41396 F20101212_AAATBW kneifel_j_Page_216.jpg
74a3aa0163ec3373ee6a885865d0ff19
93e268d7655dab317052a7a405d0ebed9edf6415
87979 F20101212_AAASWQ kneifel_j_Page_034.jpg
984fc204701443b1d0f9547508c34697
f2b3397d13d719f2b80157b74fb844e6299cec8f
3241 F20101212_AAAUGC kneifel_j_Page_116thm.jpg
244831cc2a09b33c10ed4c4aa833c225
cd69a3c9d00f3a71923d0d32a074324d63e3b075
26623 F20101212_AAAUFO kneifel_j_Page_107.QC.jpg
888ce57c2e43bf7135bbdc4a052a9a31
ced12e13f113c40f2b236c212c2c301511b1cb13
5634 F20101212_AAAUEZ kneifel_j_Page_094thm.jpg
9a81b8fe50c72e704d0de347753c37f9
97900561f33704a6b5130a24c6a74e79613031b6
5522 F20101212_AAATZT kneifel_j_Page_005thm.jpg
255392eaf912c995b5bf908be023f0f5
9736f5d79981f5a94d4fb77ad576307dfeba0d42
117284 F20101212_AAATDA kneifel_j_Page_041.jp2
2065be5bf1bddae6a37a39a572d427f2
4a8606c9ebf5e018783e9b2e7dea91e814050793
129658 F20101212_AAATCM kneifel_j_Page_022.jp2
d65577cf969946cc4b6f65a859b82d9a
51e9c7792db7320770a1bf08d9259d12e92f4f73
84489 F20101212_AAASXG kneifel_j_Page_054.jpg
3100bbeaddb56d2c9bd203303b3c328f
0bf1aab54949c0332b846322a80c1a54c9de4da6
1051983 F20101212_AAATBX kneifel_j_Page_005.jp2
fff4ae41f6ea7b02f7223d898da75cdb
1e5f3f38013e577582edbe3040fcc2ae68b131e4
74219 F20101212_AAASWR kneifel_j_Page_037.jpg
36695ad78a4ae14787f55c9651e2478c
b486a9bf6fff89b3efe402794ae773901b6fd0ad
10865 F20101212_AAAUGD kneifel_j_Page_117.QC.jpg
f5671e3dc860185de2007bfc4c200da2
2ab38b6b215d4475f802f61b77d22d5e5756a1f6
6998 F20101212_AAAUFP kneifel_j_Page_107thm.jpg
3554091afa3b9ecba49c1b227797db98
d49c04ee36a5ea846616ebd8ebb0be4c5818b883
26459 F20101212_AAATZU kneifel_j_Page_006.QC.jpg
86a0c464e548c29c2043d3668c8d91e3
256a8674368eee0efe1aced0f70baa5aeafd3b22
118334 F20101212_AAATDB kneifel_j_Page_042.jp2
4d7d00e3de7ca53025186be6f673dd09
9a591a637f246603bc5911da5d2a7b3fd39de269
1051973 F20101212_AAATCN kneifel_j_Page_023.jp2
831fa0cb9f2150d4d1353079c58f5acc
ed6bc839a1e821f6ac3302e129f44b154ccf0765
77183 F20101212_AAASXH kneifel_j_Page_055.jpg
cb5c7bf80ee0625eb6d6fe23d6729005
f2ae592cac1815020186334f85e5641bf1448883
F20101212_AAATBY kneifel_j_Page_006.jp2
be0304aed231666da94a5c4093ded868
7a7f3c34d15d6ecfda734c51b1d31e5440e40080
91426 F20101212_AAASWS kneifel_j_Page_038.jpg
974df29df64dd15d8093fea34671e2ea
951e1680bb0860ab5d2eed2c65190ce497f3954d
12672 F20101212_AAAUGE kneifel_j_Page_118.QC.jpg
244a15f7c0263ed0e933ae1a745ed3d2
087a814bc8438a8824c882a9ed3297557c2087eb
6489 F20101212_AAAUFQ kneifel_j_Page_108thm.jpg
6a71aa692a479c09ae9d574557d5923d
c638c62679a40719cdc0cee58ea343a23c959475
6491 F20101212_AAATZV kneifel_j_Page_006thm.jpg
9ce458beb75f793aee546436b6b4f8ed
6cc4b4ce8df071ac91058c3f23cc15fa5c8bcd8f
80750 F20101212_AAATDC kneifel_j_Page_044.jp2
e086c64f9a917c6088603a95c5fbef69
f4a88757b29f6ab3b0c862f80afaff07e0661584
1051979 F20101212_AAATCO kneifel_j_Page_024.jp2
c50e6dba12701397b4b357580acfb4ab
fb19afd352bda14448aedfe07ff994410b0dd2bc
66005 F20101212_AAASXI kneifel_j_Page_056.jpg
1576f85b422993de48b46d19213674d7
edbd18c1e7f1f0ebd04e20fff24a40936a4ec029
1051941 F20101212_AAATBZ kneifel_j_Page_007.jp2
0b1f359ee2335699f6f6a12da01fe342
88bad59ac23a46726aa9a0dd8e14859f63be844f
74841 F20101212_AAASWT kneifel_j_Page_039.jpg
5e8d6a03a1b813f04f0e8620b4366997
8172d6334e37e1877424117aede7c9151239270f
3670 F20101212_AAAUGF kneifel_j_Page_118thm.jpg
b0ed2311d074323c677a465851fb7670
762a61893f0d54f51d226f19e58ce183798d54e5
26667 F20101212_AAAUFR kneifel_j_Page_109.QC.jpg
ef8ff5a32ecfba8660f754fc0290714b
bc027f354fe4fe133f0a0ef00aa2a61fa9465d5f
25774 F20101212_AAATZW kneifel_j_Page_007.QC.jpg
31bae4bd0622a3100d6b535f4268b171
63c06aecf63ab93f6ef366420fc7a38070f0d6ab
19643 F20101212_AAATDD kneifel_j_Page_045.jp2
aedf4c2e6174b46f243a82791f8d89b6
4f4c3c9a9440bddea7d351cb011e4e5688a0367b
F20101212_AAATCP kneifel_j_Page_025.jp2
2647f20f2e54c806dd919815ab59351c
d248b822e696b37a6f3524c158cdb49c010b4776
100525 F20101212_AAASXJ kneifel_j_Page_059.jpg
4ec0e56c597136aaf89f881ffe336421
1611b6d34cf55b5379ef408615ed7ec57969d97a
79821 F20101212_AAASWU kneifel_j_Page_040.jpg
103f50b26807063cf81ae3a22e713f19
3349069688b3522be12c735c7e4ed28c26e55450
4878 F20101212_AAAUGG kneifel_j_Page_119thm.jpg
7c8bb0b1a4166017fdd1849d97f36be3
d5c7973906ea0026df92681ef298c32fb4bd6b19
6830 F20101212_AAAUFS kneifel_j_Page_109thm.jpg
3da0b6eb87447b6a5247c861db9db976
34ea8022a2700124e93684dae6176b42690723ff
6164 F20101212_AAATZX kneifel_j_Page_007thm.jpg
87c4ed8b3587a9fb412c949fc9a92e92
e1f1a8e0c51208e94c89ad78275a04eb304c6aab
54046 F20101212_AAATDE kneifel_j_Page_046.jp2
b9e068c74ff37b421a33518fac34deec
aa8b93f3ae48e5d7c4d79fb9997993536d731f23
1051937 F20101212_AAATCQ kneifel_j_Page_026.jp2
e17d47e5289e7cf861a393f2ecf557cf
b822349826db60f263cd022e871786a378c577d6
88144 F20101212_AAASXK kneifel_j_Page_060.jpg
f80f598ae6faa1d4ccb0e8687f08c531
b9f8f380f59c4ba3d19a48bfcb68f3dfad8790e7
73856 F20101212_AAASWV kneifel_j_Page_042.jpg
8b1afcad5e9d8b7af1456714a6cb4dea
2ff54818687645cd62408d858b204e84ae7f4e9e
18375 F20101212_AAAUGH kneifel_j_Page_120.QC.jpg
6a638ab9c4c8a8a61f1722bc9ed69666
4d9cc089144886e903e778408203072154d70f44
5685 F20101212_AAAUFT kneifel_j_Page_110thm.jpg
3de88524197508da32786c83e7a9d819
42a311e858d98c5534340b6954231b0979d4ad33
20708 F20101212_AAATZY kneifel_j_Page_008.QC.jpg
2db785d52896cc911a1146b7434e7461
61aefced3965ec6801aef8c4ac6a5dd479c9fd5d
119628 F20101212_AAATDF kneifel_j_Page_048.jp2
c132b0cd6840bedc9867af214f1de593
fa4b39185f4417244166380ffbed3e141df7cadc
F20101212_AAATCR kneifel_j_Page_027.jp2
8fe7d94a6fa9f2897cbe6479129fdbad
2f95ff521a3b693fb6593942955a1664edabe697
26503 F20101212_AAASWW kneifel_j_Page_043.jpg
601b63428e22432816111934790fbba4
7d2bbda5fb8359df777759e8852497e8066220fa
5539 F20101212_AAAUGI kneifel_j_Page_120thm.jpg
af3e5f41bbfdcb3d21470cad7ba1bed9
3f7c527a5e63747fd0488272a4dd281fad518215
25391 F20101212_AAAUFU kneifel_j_Page_111.QC.jpg
fc51b1b9aff360cd2af0d6b09c603597
853d4e9c4a0e960374f9cfc27b68907428d18bf3
5384 F20101212_AAATZZ kneifel_j_Page_008thm.jpg
81a8d66d300c5d5e10c378fb2569b033
242861b9a5582cd1d9b6393d10f33d7405d9c088
F20101212_AAATDG kneifel_j_Page_049.jp2
e05aa599ddddf074f8267d620639b5a0
b72129bafbc1537ebfc36205a7accf959081a772
78046 F20101212_AAASYA kneifel_j_Page_084.jpg
c9cba18a288db395a35c79cbd186b049
34ae90c9ea1a859399d29fa55db3f7bcdaeacee6
F20101212_AAATCS kneifel_j_Page_028.jp2
33886c1fb244433760a4e7f46b08311e
3a95bcbacf5b209e45dc6b9bc07a403e730b3f7d
88552 F20101212_AAASXL kneifel_j_Page_061.jpg
fd66aaf466415abc333afb9af1b3d740
d0254a936f4328f988470c0216876ff2473f8bb1
53109 F20101212_AAASWX kneifel_j_Page_044.jpg
8668a8b44565df4b4738b95ba2564455
746e0a166acd02f8bee423c8978b73482dc44ba1
16522 F20101212_AAAUGJ kneifel_j_Page_121.QC.jpg
ce00c52f836ca1949f79075eed18d0b0
cbd7a913b14421f160795d01e845b0ad48282f20
6438 F20101212_AAAUFV kneifel_j_Page_111thm.jpg
44d3484226b4dde8e4618ff2f962bdd1
12b38b85884616204c6417c1c3437a78981923c3
124683 F20101212_AAATDH kneifel_j_Page_053.jp2
851b2f89fad1ef3926f2648d23149713
17c0f38d5fc3ecb35d88ca868dd2af87af9f6ad2
84956 F20101212_AAASYB kneifel_j_Page_085.jpg
b88a1c0419d7ef809b81382281ddc6c7
1f81c2c2a831fd0f32081c6aa5ea28c379fea0b6
F20101212_AAATCT kneifel_j_Page_029.jp2
03f71a1ae817e1226d2275ccb38202ba
902cb970f678668ef1728d634b6e8f2fe5b85ec6
69038 F20101212_AAASXM kneifel_j_Page_063.jpg
f5cfe41b3e467a4e3a5f156f44aa4937
42fd427d00d1249801e65cdf0cebae049bacdf05
19046 F20101212_AAASWY kneifel_j_Page_045.jpg
0234c6c37e4ebdc76cf94a33e72b1705
dc520411eb39e4dad4a5d79bd54bb49fb5c3f190
16216 F20101212_AAAUGK kneifel_j_Page_123.QC.jpg
94b671ed25b5bdf9dd87c6f7cf1688b5
a3dca0fb3d242054dba0eb1a39071c3c663110b7
2125 F20101212_AAAUFW kneifel_j_Page_112thm.jpg
24cdc00eeda1c343227db9af2dca8789
618172b02c59cc082c2b0c8029e208893f1eee89
1051966 F20101212_AAATDI kneifel_j_Page_054.jp2
7495543fcdc617f981457b0d2c982745
9f91f72d2f87d7cc867b3163e52e7c7828269d34
71724 F20101212_AAASYC kneifel_j_Page_087.jpg
8c15bea7aa91f2f855cdf98640d6020c
05d9a5db77f1025c2e19fa8ecf978200279bf0c4
F20101212_AAATCU kneifel_j_Page_030.jp2
d7396c2cbc4a7d29a17832796f4833a8
fbdea0404bead5a9e214e7f1dcc4d5677bc36a2d
67980 F20101212_AAASXN kneifel_j_Page_064.jpg
981be0af5487248ccf36377aeb8a8033
96805444f793591b85afe7a43c308f6b925db491
39441 F20101212_AAASWZ kneifel_j_Page_046.jpg
42f0c9f27da339034afa2aef2a06ca38
3a01fb7747d5deeae445ad7331ce4ce211d7af01
5227 F20101212_AAAUGL kneifel_j_Page_123thm.jpg
ded8cf90f04fb236cc2364c099eb9729
076743cd9909cc2d9511e8586926355aad22ecc3
120830 F20101212_AAATDJ kneifel_j_Page_055.jp2
ded1a342806fb7776de7f5c2f3617185
60fa63a071804e6833bd0132a5bd2dec00706bc2
86255 F20101212_AAASYD kneifel_j_Page_089.jpg
b21e393b9f476b246e8082d161efc386
b013f93ef08d216be64d654ff67a3de83a01d66d
69960 F20101212_AAASXO kneifel_j_Page_067.jpg
465512d059ae7e1d5c3b2efe7c5dc1b1
af232ffa9c5b5ea1c1708f1d0f35686d143ba05f
23941 F20101212_AAAUHA kneifel_j_Page_135.QC.jpg
ca28aaebd2b7880392bc6ad1c0e8b611
69a8a81f7fdd281d1d51c16f6535fc5a01b4edec
3273 F20101212_AAAUGM kneifel_j_Page_124thm.jpg
5acb9bdd620c0c64ec6881bc0de62575
93909fd581e2f2d21bb73fc1ad65dc0e19c68f26
9842 F20101212_AAAUFX kneifel_j_Page_113.QC.jpg
6d904b77c3793379d2ce438f6c74633d
ed37d529db2d7e359144e0e0fbe55e76a87c2a6e
103478 F20101212_AAATDK kneifel_j_Page_056.jp2
ca619bb5fd149d19e3cc05288805a134
0f810e43b9c6fbf36b79cfb25376ce2c48eaf50a
75254 F20101212_AAASYE kneifel_j_Page_090.jpg
f3ffa6bbd88c05a894827be14cd7992d
300debda3d7ecc4bf8e2eb5d7661609e55aa020d
F20101212_AAATCV kneifel_j_Page_031.jp2
8f0261679f406fe1c13bccfc5bc2753d
197dea18cdb0bbd457f12be2aea2dcc047a70936
74490 F20101212_AAASXP kneifel_j_Page_068.jpg
89e7214e68492ecb86c42db6d047c167
565660542492748b9a0ebfc8068d6cb019be4085
23215 F20101212_AAAUHB kneifel_j_Page_136.QC.jpg
d320c8db72cf42a8adb97a5c26ad1c7f
24a23a2eef1e59c227d7c6b1903792c02201acbd
16422 F20101212_AAAUGN kneifel_j_Page_125.QC.jpg
8cb507f843dd90cc9afec78f0c072284
fe460a8ff57475127fc368ac9f35fb527a395762
3067 F20101212_AAAUFY kneifel_j_Page_113thm.jpg
6ee7e690a0c8338f3b71e85435551754
540d6a3bf95f83defb1bbd0b6000d8853fec955b
F20101212_AAATDL kneifel_j_Page_057.jp2
9844ac3633c8421cf6d40d4a6c070d11
cf154853406f924fe091b8118ed8dde9c3adf498
71595 F20101212_AAASYF kneifel_j_Page_091.jpg
449a70b18730c8ae6c0d8028fd9431f5
0394ebfff513e74cf03e2f918333bd413c045ac7
1051981 F20101212_AAATCW kneifel_j_Page_035.jp2
528f04d923d6a04f2d2a94bdae9d550e
09dfd06457df03529ea4532d5a55697b64350f40
73941 F20101212_AAASXQ kneifel_j_Page_070.jpg
8a9b8854b97065276ab3c6c9f7c76eab
165c8d41f0359bac8f5588853dc42334a215cf1d
6004 F20101212_AAAUHC kneifel_j_Page_136thm.jpg
8424daac0f0f5edd19475f5b7e9dadfd
5043fe0c7126d1b92478c9ba00cb5d1571e852b1
5561 F20101212_AAAUGO kneifel_j_Page_125thm.jpg
456aa4ddf451304fb8f83393dd3c02c2
f04d08f5e186310e4144e43a075973a56649bb22
13910 F20101212_AAAUFZ kneifel_j_Page_114.QC.jpg
aa29b77f6655e369512ef76637bff861
0940e2582bb196debfc91059a778bff4e442871c
61594 F20101212_AAASYG kneifel_j_Page_092.jpg
9958c6145ee6987b56ae86544bb7d014
19fdfa17aa9cdb9054abbc589c74a98ebe7fbfb1
118169 F20101212_AAATCX kneifel_j_Page_037.jp2
355ed019224649aab4cf9db0ccf3e346
91910b5225699f4e8f279d685d60fe63fdf55129
63954 F20101212_AAASXR kneifel_j_Page_072.jpg
46754b05863fcde31419074427467385
979e6fc8de6dea6677aaf3d984127ebdf1e43760
1051985 F20101212_AAATEA kneifel_j_Page_074.jp2
05c6a8eb2cd4d60ac4c7d622172dd79c
1303358c04efae7bf738d7f138b7ecad283891b9
F20101212_AAATDM kneifel_j_Page_058.jp2
2c88ee41ced921b6b5ec81a96077b63b
1a179fe2011a22284ebe0bc3213e2f7e79d2577d
22800 F20101212_AAAUHD kneifel_j_Page_137.QC.jpg
8bf2b28aee181fe6e13691dccf524326
4b65a17dad927922f3dccdde38a21526b219b98e
2526 F20101212_AAAUGP kneifel_j_Page_126thm.jpg
6134f57312c0c60c64ff87d7180c6101
db1adddc8ff92ff605d08089d8334ed1807bd579
88452 F20101212_AAASYH kneifel_j_Page_093.jpg
631e8e9b3d7a322b01add564d2a9ad50
bdaf43775f84964ee5c40679cb5e0885800cfc1c
F20101212_AAATCY kneifel_j_Page_038.jp2
3b4aca31022bb6fa9f19e18f93ded135
6448a1d876aef9dec293322a3271bb859b35d7be
57134 F20101212_AAASXS kneifel_j_Page_075.jpg
23ba17c2ea5df07cf3e891634909edb8
abd746eebe681dc7bd8faed38de3a89ce2218e21
F20101212_AAATEB kneifel_j_Page_077.jp2
aee8db22187a13ca8032fcef1667ce1b
514c17dba3a13e3c7aaefae0553419e71755a6db
F20101212_AAATDN kneifel_j_Page_059.jp2
a06024f8e797c95806431cdb01282809
dd13e4ea3503c31a52fd4e7fc935ed6b35fb812d
5737 F20101212_AAAUHE kneifel_j_Page_137thm.jpg
3c8b7b1bf59ba49c962957da8bf0d82a
cfbf1105b36abeabb32f6b0b7412aa1a8d2dbf71
24347 F20101212_AAAUGQ kneifel_j_Page_127.QC.jpg
5907253ebb44c24afd3de87b1a019753
e7a1924797b2035b8e93133babbccb539cba5695
58837 F20101212_AAASYI kneifel_j_Page_094.jpg
521ab99f354284b2bb620d4c386deb87
cb4365ca89d9827551048373675c9dcaf93f71f6
118227 F20101212_AAATCZ kneifel_j_Page_039.jp2
7776c8d8c914debe2056822c36c0127d
21eb21c5273127e0ec1ac83d4143771150f4a476
72448 F20101212_AAASXT kneifel_j_Page_076.jpg
b258f87a1c2825d00dcd0e1c71bacdec
dc4bbeaef25b77ee2c2e1286739b6a6435577bde
1051958 F20101212_AAATEC kneifel_j_Page_079.jp2
245806e5ddd45e1b8c35c86e5f51eed5
094a4335661943eb7be7c60f0f450e21b15d37be
1051982 F20101212_AAATDO kneifel_j_Page_060.jp2
253c2d87c6afac99fdba9844bec55849
24af1d428420a3cec0a01c6dc69e95694748d96d
22537 F20101212_AAAUHF kneifel_j_Page_138.QC.jpg
f4c5bc39bac284247f4b0338220fa554
9550432168ebe64479c71dc8f10009ad687d2d4e
21678 F20101212_AAAUGR kneifel_j_Page_128.QC.jpg
95ed922b879aff1f3f1c63325c3dc424
9acc5cfc4044c71657487f215e7b821414040376
78700 F20101212_AAASYJ kneifel_j_Page_095.jpg
71eb84c13261df82e9d6e7d069c7d09b
f2d3cae65bd4ca77b13e73f21259a17b3fbf65af
82651 F20101212_AAASXU kneifel_j_Page_077.jpg
55789bcc705981d2c0bf56b9273e7bb7
778def73453b424da95c504ec0a1436ae102605d
1051971 F20101212_AAATED kneifel_j_Page_080.jp2
d7264522283d54b941689a50805a2edd
f2d2b399ce736bd00ba3ca50443c870122ac73fc
F20101212_AAATDP kneifel_j_Page_061.jp2
ce73b2d61af1f0e28c29b7f52e901360
aada64aec5cbf0463cdd17683a14ea98be274630
24358 F20101212_AAAUHG kneifel_j_Page_139.QC.jpg
532513040eb3531e42463d91c611723f
4ad22fbe0a8d678bc7f5a2d9b9231a7e673c2ce6
5746 F20101212_AAAUGS kneifel_j_Page_128thm.jpg
0aff5f5ff9e3aee9e46ecd5dbaba782c
ec7896f8ae493b65ccd265be0dbf8a8da8412845
69222 F20101212_AAASYK kneifel_j_Page_096.jpg
938e24a54c8f4a9965115938f29e6947
2576c31b4aa09306dfa3698609e4717636e8253d
83640 F20101212_AAASXV kneifel_j_Page_079.jpg
8fc43e988c80bfcf55b0024c8de33f1d
d7c9923c94772055600d56919d01e81e3bc38c63
91711 F20101212_AAATEE kneifel_j_Page_081.jp2
9ff720691252841f885bb5d6379ee4e3
ac9f861b7b0828b9183e9387bb3f7ec731acd153
F20101212_AAATDQ kneifel_j_Page_062.jp2
b483765f5dc07bee251354d100feafa7
4e5e8ffde52eecb3b222494e5e1ee077a55982fa
6372 F20101212_AAAUHH kneifel_j_Page_139thm.jpg
1e912f12be1a4e6a18b98427a3504ff1
40167feb67f1902cce9b7fb23b288c490405c323
23956 F20101212_AAAUGT kneifel_j_Page_129.QC.jpg
97d5f26836d05cbe019c7a4c424cdc8b
7a00b8f8ba759c5c957fc2a0d8d362cabca348e6
69473 F20101212_AAASYL kneifel_j_Page_097.jpg
b97909460d79f6dc62e4e8a4f6dee34f
bb9a777ec6f03c7978a2fd4385486a780972acac
83987 F20101212_AAASXW kneifel_j_Page_080.jpg
3f2ff53985d804521dd46c17fe448082
8312e63c3a3f8852df52853cd64bf0888a8cb06f
911461 F20101212_AAATEF kneifel_j_Page_082.jp2
6df2ae1638a32747f6eaf41fec80344c
24e05e6be2fa8e6b761fceb2c7f31fe85a9ac80b
967142 F20101212_AAATDR kneifel_j_Page_064.jp2
595cd4e181af3e4b3886d99a799cb46e
c44b55a90dd07814d4ad6119eb47aaf3b89df7fd
28255 F20101212_AAAUHI kneifel_j_Page_140.QC.jpg
d20c3cf20f67a29ee8b9b17c54237d84
8cba141f08571927fb8a89e0b524960932fa6681
6326 F20101212_AAAUGU kneifel_j_Page_130thm.jpg
078c4711e83c0369a1a642ef4144d1c3
c5627b9e8fc6667b7b1e36ff9dc735b319c1f0f1
57290 F20101212_AAASXX kneifel_j_Page_081.jpg
44d49a0f32882a5bb43adcc531227de1
f74401db98b13af9fb3f97efd88f8ecbbc0f049b
F20101212_AAATEG kneifel_j_Page_083.jp2
b2c138b2610f8e150d6728c9d054a73c
5e8c48e53f0632dfdd8699dbdec61ff8415d97ca
33609 F20101212_AAASZA kneifel_j_Page_113.jpg
b4d453b69b16d7f946ea065ed043e95b
c82b82b9686cef092733f10c5d328d4ffb2a718f
1051923 F20101212_AAATDS kneifel_j_Page_065.jp2
7ed85294cff409d48b85934bb46be33d
816d2c555ea933d531e98d4ade90dca1930484ae
7087 F20101212_AAAUHJ kneifel_j_Page_140thm.jpg
d0575c22c8bc842cf4dd3299ca0eb5e4
2124727a4bb95f10b86571f4199d316eb673e985
23914 F20101212_AAAUGV kneifel_j_Page_131.QC.jpg
333e941450a2bd31a7360df601118b74
e2e78bca1276cf07594d10edbf72f3b33658f0ac
88419 F20101212_AAASYM kneifel_j_Page_099.jpg
fa9904722ec661bc32a6dfec6dcdf6db
bd545b6e396e4c8cfc2839f1919eff251f83da8f
63073 F20101212_AAASXY kneifel_j_Page_082.jpg
62a9391ee1253778ad25a9169a489534
1065a983bd06173181f7414e8c07a04dd775a614
125062 F20101212_AAATEH kneifel_j_Page_084.jp2
c67436260934c0972f6fdb81e5526777
dbfff3b0c8deac037ae26e25e5fac035053181f0
30330 F20101212_AAASZB kneifel_j_Page_115.jpg
e4aa8f934e8ae61e7e36804a2f5e4787
52caf772c39c0695d9f2f835aca4e5b6169b1876
987241 F20101212_AAATDT kneifel_j_Page_066.jp2
9f3ca606b8968e0c1bf44813f49ec92a
bcda70b9d4ffde6320c90ed76766e62c691633e9
23492 F20101212_AAAUHK kneifel_j_Page_141.QC.jpg
5205fe2ffd5ac6287bd36bb8909c3304
cf458d58abfa4d24e5872a83eabb22012724988e
24712 F20101212_AAAUGW kneifel_j_Page_132.QC.jpg
3649a327a8409f598c2fa43736837ada
73e3ab0f6d795f9bc277368fd7a88b3c3981dfda
93864 F20101212_AAASYN kneifel_j_Page_100.jpg
3986860d57614295b3324170ea50b87d
3188c6007e836398a6d8428b8d67025dcf9a5f48
83138 F20101212_AAASXZ kneifel_j_Page_083.jpg
42089bd98a1e7a3d190fdd91e4ddf27a
df59effbeadd6bae23ea55edaef58e5c85d6b89a
1051902 F20101212_AAATEI kneifel_j_Page_085.jp2
5c355017d21f8c5568ed0995b212481d
fef6bdaf6ac2131df4d37bf0691a42580ef0a4cc
36459 F20101212_AAASZC kneifel_j_Page_117.jpg
58569383651586ec17b94bfa795c09dc
222ce1a90f53c205b12691c1bccde4825cf4dbd2
F20101212_AAATDU kneifel_j_Page_067.jp2
6fe0578f47a63423cc28edfc81be77a7
2bccc064644f213a69fc4929fc51f0b02def8876
20249 F20101212_AAAUHL kneifel_j_Page_142.QC.jpg
c51f0d1c3ee528e110ebbba8427edf56
f84e2692e6cbe1412af676f0f5e06f9403d03cb1
18129 F20101212_AAAUGX kneifel_j_Page_133.QC.jpg
b449256c3a35f77cf7c451bf16c5facf
ef813c8ea5e5c381c39c98bde19719f205ecf347
59097 F20101212_AAASYO kneifel_j_Page_101.jpg
04fb6b1bd137e00abeab6b4744d9d778
4700a58f148b3501187a11045d787092f4d2d3ca
F20101212_AAATEJ kneifel_j_Page_086.jp2
87d819c7cc49b1865e19a846b558560f
b2746f9c515e8c6423fda94a7f24aa25a8768e9e
42709 F20101212_AAASZD kneifel_j_Page_118.jpg
7a025682079aff851914362a633bfb84
c563b2ac5148169092bf2eab1becce0973b14d17
F20101212_AAATDV kneifel_j_Page_068.jp2
a1896f5d4d49e9ee917eb31d041b952e
96b1272a7d11fda7af857aab40418c17a52aa24e
24540 F20101212_AAAUIA kneifel_j_Page_154.QC.jpg
5c5485c9f562414766cefb333581bca6
e7d4d3fcaf4aa438980ed225ad7de4e2c0d77572
6104 F20101212_AAAUHM kneifel_j_Page_144thm.jpg
0b8e7f6583b0e92682abd95c9b535097
f38348df8badd9b4e476d69d4d677ee5a638c31f
77330 F20101212_AAASYP kneifel_j_Page_102.jpg
4ea02bd8afbf0d7b0ced044794463867
d7df29a7441c1edc21f1e1f008d697aa5d6beb62
996108 F20101212_AAATEK kneifel_j_Page_087.jp2
8a126145e6a998887a9405efd5abfee1
b45b23f79282a8450dc3969f091a5cf242facf8c
48567 F20101212_AAASZE kneifel_j_Page_119.jpg
a77d8f0868f58a9448a82b1631ab2acc
c308143f0f2a234ce36114e891ff0b60d10e01d0
25496 F20101212_AAAUIB kneifel_j_Page_155.QC.jpg
eeccb991271801689f8d5ad8d02aec59
9439dcbe5ec322ae4e32be7ae567fd56d86161a4
21855 F20101212_AAAUHN kneifel_j_Page_145.QC.jpg
7712d8892025fcd06667572c02e4b36c
ea1dc77cb2ee886e2bb09d779e2ddb2612af4a05
5011 F20101212_AAAUGY kneifel_j_Page_133thm.jpg
bebe0f64953fea06362fbd61d182e792
a289735911673e52f3585c2c0d3d1651aeb49816
66830 F20101212_AAASYQ kneifel_j_Page_103.jpg
3801fa647319019153c5c022ce16faac
e2036e957478299c514287ebd56fc1afc743ea0b
1051963 F20101212_AAATEL kneifel_j_Page_088.jp2
86dbb10ec9b7ff8ea738485c24cb356d
af6625efaa9581f3e410dd1ec72d92f7e692a562
63326 F20101212_AAASZF kneifel_j_Page_120.jpg
87ec016cdc2184d6e17abfffe0fcddc4
f5399c4c731d79b98288d1cf112bf69a5ea98769
977909 F20101212_AAATDW kneifel_j_Page_069.jp2
db345c5bb4a77bf8cd54b5d6208d6913
ba33ea52c2ae4a6e1cb1f53f2bcdbea1c6b0e1df
6390 F20101212_AAAUIC kneifel_j_Page_155thm.jpg
40b50f5f087715175119fe532c8e7262
4290b45210251d1324d5e619106e8485fa2ebb36
17688 F20101212_AAAUHO kneifel_j_Page_146.QC.jpg
b476925faccd4ac8ec65fc263108791d
e8d9d3189e3919f2dc0596d905941b0b97f4da77
26244 F20101212_AAAUGZ kneifel_j_Page_134.QC.jpg
630d64e92895d7445cec3c0bf7995885
76b0870ed43acc614241255b4f9ba7c73ac860bf
67409 F20101212_AAASYR kneifel_j_Page_104.jpg
d299d57366e6d3390f982649e4a9f529
d49667cfaca8b03cb542bad7d403b4643a9e8d27
36880 F20101212_AAATFA kneifel_j_Page_115.jp2
f1267527890aebd5d47c8adc3e8343ca
c5853539dc45e8f1ce4d9d32c32f301f8cfecb0e
F20101212_AAATEM kneifel_j_Page_089.jp2
418c3f7979f364706b52111c27dfcc20
91631282062b379d37a778a2e3c8a36b1459423c
54721 F20101212_AAASZG kneifel_j_Page_121.jpg
a48a8c66f870a9583efb219b875ee6e8
09ebf01e19a739b93e513c4a02e348284379b07f
F20101212_AAATDX kneifel_j_Page_070.jp2
5db2aab25e2b5567df597b3d2e2b803b
bfe100b24c1c05f423d2a46215231f05baf22a19
6324 F20101212_AAAUID kneifel_j_Page_156thm.jpg
93ec578a92c9179f83ebbfdb7e52bc29
c3a6a1f60801e56f9c6c325043c3610d7e50c6e0
4883 F20101212_AAAUHP kneifel_j_Page_146thm.jpg
8480393520a66b85e78645dedcf796ca
9a9e594aba4e53279721f23a09910d71a22154c6
71567 F20101212_AAASYS kneifel_j_Page_105.jpg
4c23683af4e69afa039c9926e9a14cef
47505dafc387f73ad6f25d84bdeed0c020f78517
53849 F20101212_AAATFB kneifel_j_Page_116.jp2
3258fe0fba6fed96fda26a0d5ce37187
9343e6261e19252c8679b0f9b28b24edd2d637fb
F20101212_AAATEN kneifel_j_Page_090.jp2
1433513caa353bd6d1b21814086a68a4
3e7cc5a31862573f1ec580e6198062ae9070821a
57799 F20101212_AAASZH kneifel_j_Page_122.jpg
f1add216a6bba56c4f36fa15bc9034a4
ea2acaaf300abe095d1a3b38706325e435c347e0
102488 F20101212_AAATDY kneifel_j_Page_071.jp2
0451fd327dfe17a9f723619a4b669e5b
14c2adba179439ad73c5da850a33bbe8379e8bc5
25382 F20101212_AAAUIE kneifel_j_Page_157.QC.jpg
34f488cee5c399a0527f83451e21e4ed
ce72dee5c0f665641588fe035a780f8dc8ec256a
5779 F20101212_AAAUHQ kneifel_j_Page_147thm.jpg
dd6848e611a1e2e0d7c9c664b4cfe912
be0e3759185178c566dc2202baa22734f1f18942
66047 F20101212_AAASYT kneifel_j_Page_106.jpg
96e80bb63975100dbaa460bdce9c142f
7b231b28b607abd10773b8c232878f5608680947
50297 F20101212_AAATFC kneifel_j_Page_117.jp2
336a173fa4dafad6cd0c79e04c0a9152
468078e68ebb562092c9f83f59056b01b93512c6
1001775 F20101212_AAATEO kneifel_j_Page_091.jp2
0f638f746682e8ae0e0d8b6c1024e98c
8c12c64adc1315b7f4f22c27281fa41e88c6717c
54412 F20101212_AAASZI kneifel_j_Page_123.jpg
07ee38165791ae778387ebabf109fecd
865d6fa873af4efc0e1bdb37eb923fbe6d4d97f7
908540 F20101212_AAATDZ kneifel_j_Page_073.jp2
061e586823a6be40ba5b2104431dd2bf
23ce9c51f98c06f06d115efd73e45a8fedd793b3
25138 F20101212_AAAUIF kneifel_j_Page_158.QC.jpg
49a399d99cd8c5649d3c74b75c3a9c66
d68738b481f10c620d5a0a122ffc0b99c63c096c
5883 F20101212_AAAUHR kneifel_j_Page_148thm.jpg
b03e25f32bd0256d32c68635a46c115e
1394f68cc24e6789990a9871f40e227ec75a89ac
89045 F20101212_AAASYU kneifel_j_Page_107.jpg
790383210c60d0be0b4f16a86e8fda9a
7a8378b408e358b0d0c70c023abed6a52038dd24
63621 F20101212_AAATFD kneifel_j_Page_118.jp2
c425121efd71df77f4d439c26fce6080
9150ab5f4cc1ad96148ad92c899a3bdfac58de89
937308 F20101212_AAATEP kneifel_j_Page_092.jp2
c6cae82793939d4b5bbae8ec78bad5c8
25383803090890a5cc2ef4992a6aa57ad6631146
26585 F20101212_AAASZJ kneifel_j_Page_124.jpg
7dd8a618b4e4c592b8b324a07052275a
9b14e818aa4e60bf6280467945a715013846561e
25733 F20101212_AAAUIG kneifel_j_Page_159.QC.jpg
4035baedbb237e042a3769ca2bffdfff
210ba5e1bb02ddefbd359cf4e5485ae2b56b9821
26813 F20101212_AAAUHS kneifel_j_Page_149.QC.jpg
ec428432f2b8e93de31f676f28b0d745
706a4c47b545b376efa207b8a75f8987403deb01
80476 F20101212_AAASYV kneifel_j_Page_108.jpg
a9819ce1b7731bd891c1ce004156cfc7
f54d154134f16aa25fa0b9e78712d49c3e2bc696
F20101212_AAATFE kneifel_j_Page_120.jp2
b8dfc97042b584b6b7f6f0ddc61e2b1e
d02473e5c050c128ee7ff843e9f886be7bd28d90
1051967 F20101212_AAATEQ kneifel_j_Page_095.jp2
bbaca9c8c189c26da7d4b2692162f99e
811056a9a69436563e81549c79c3b97fb137f417
48592 F20101212_AAASZK kneifel_j_Page_125.jpg
d1e4907c05190f6787bff380ddef36a9
44ed274d7405d98f4891a4653e333334a2eaed69
6517 F20101212_AAAUIH kneifel_j_Page_159thm.jpg
d488a566e39431bc25570283387851ae
4d45db5e5b79d6f5a6682867f99490ef6ed1c3a6
6750 F20101212_AAAUHT kneifel_j_Page_149thm.jpg
021ea71f640685de41ed247fdd33abf1
1bbb152f57fafd8f79f547da44a91823c3083508
86454 F20101212_AAASYW kneifel_j_Page_109.jpg
b6d15738781419a2221df31ef8e23a49
af08c2074b97c59a06977f1fb11409379284df8d
F20101212_AAATFF kneifel_j_Page_121.jp2
587d574a300be631cd899bc7c46e947d
3ea780c3e7184208e27481f066d17d97d2141fec
965439 F20101212_AAATER kneifel_j_Page_096.jp2
1e991eeb203fb836becf57796f3873b8
c9a0110fed96a0a9bbed9245f781af4b490868dc
22778 F20101212_AAASZL kneifel_j_Page_126.jpg
d3bdd9a793d98e21646ebcd5579c15f1
d01ccc76c4a14d5f72b07d8852dc4e5504eb577b
23901 F20101212_AAAUII kneifel_j_Page_160.QC.jpg
9baec04939dfc9b20cd32156e6e47263
0a7a78b3ecca0d693067d7fef516125ba76998d7
23907 F20101212_AAAUHU kneifel_j_Page_150.QC.jpg
c5f95fafb6212734a36106ddd69d6710
19e7a5defd1c6b06f5a386880f380f92517650c5
70863 F20101212_AAASYX kneifel_j_Page_110.jpg
dc672892e4d6e40d304ff4150709fac7
cc7ffa39a3a9d71d128f3cf62c42b464a770c8fb
1051944 F20101212_AAATFG kneifel_j_Page_122.jp2
511c5e8e5dfa89a8d2ccb5d221cd675a
6c9f23bf60acfd7b773a4852952c1b4566e88403
1051951 F20101212_AAATES kneifel_j_Page_099.jp2
fda69550b17b403ef42fafcef34dfbcb
44215a63a74f99c2151439c353f18a04dfb2f1f5
77333 F20101212_AAASZM kneifel_j_Page_127.jpg
9daaa2aa71d3eb78748458fa18549708
3a3bb62c33a2922548bfece1ee8833f72674013d
6396 F20101212_AAAUIJ kneifel_j_Page_160thm.jpg
5ae5803c16404965d258e07a0e0c5d24
54dcebe11a2c4e6da578d75462e304fcdbb77e2d
5758 F20101212_AAAUHV kneifel_j_Page_151thm.jpg
595345231ef3ffeb594a7a1d8ae42883
a4c05895ad86fcabea2c6bb063eabbf0bd6c55e7
77801 F20101212_AAASYY kneifel_j_Page_111.jpg
e14e284bae527de2ec08fa8559d1a33a
231bd99d445ad526c12a8ea6bb00907d76d6fbe6
1051945 F20101212_AAATFH kneifel_j_Page_123.jp2
a38bda174000eb9e05639d8879a7658f
bfc513de44de7ca52b9499c7f6b4fc9a21336529
F20101212_AAATET kneifel_j_Page_100.jp2
809780c3741420bd49d532992fb49b19
04832077eec2a51e785f326f4a2a2e056ad5961a
22609 F20101212_AAAUIK kneifel_j_Page_161.QC.jpg
5e074e7e0e33da9e48a32ed5392f225b
a98c309fedb2ef781a6328e35980452916e2ebc6
24730 F20101212_AAAUHW kneifel_j_Page_152.QC.jpg
7ec17d53a09a1aaa2ec913434079c518
d7de6e3ea28ed93bce4632555a2c2b6836d05ce3
19975 F20101212_AAASYZ kneifel_j_Page_112.jpg
f650bed82350be546043d8cf29b64ae8
1610692ea05d781c6217a51a174469f58bde83f1
297889 F20101212_AAATFI kneifel_j_Page_124.jp2
b5a219061192b3b37bbd8c3c98855eea
741fe018ad016b0b6d0f55f77c5fcd8e5c0063b9
851271 F20101212_AAATEU kneifel_j_Page_101.jp2
68890f485e99c03e19de5b4ed69ae719
b8ec52ca936b33df0303ed3d2d49a2841fe5d084
66697 F20101212_AAASZN kneifel_j_Page_128.jpg
7e11cbdedc19e193dcc26b28e36ea109
7ad947f62bfd0dcc15264b90add430b99ebceb92
5807 F20101212_AAAUIL kneifel_j_Page_161thm.jpg
02646a1a4bfdd959407b73a2352db75e
7d783a7784e77f95aefc2d3d3e9f88b9ae27ce55
6584 F20101212_AAAUHX kneifel_j_Page_152thm.jpg
f1193000ab021eb99c94cbf53fc7eab2
f9b2adb7e7b8f6ca01330eb6434c5c89020f3638
700582 F20101212_AAATFJ kneifel_j_Page_125.jp2
5b845d2f28da3f7f7f98b2b55cc093d3
042c2447428b10f2bb88f11d4f6d59af0902609a
1009142 F20101212_AAATEV kneifel_j_Page_104.jp2
a415ea55a19d6c130324ddbc8ebccec1
14dec9475089b46b115d14a784a7703e81639c78
75803 F20101212_AAASZO kneifel_j_Page_129.jpg
a2da51c9157660db3b7b8a8506dbd32f
5b4301fed537580141fa56c1a6baf73be06e7e12
6827 F20101212_AAAUJA kneifel_j_Page_170thm.jpg
51762fff4465c5242ac2703d85505494
bb9c393a4b86aba7cb9ca3cd795e34d528753d5f
6190 F20101212_AAAUIM kneifel_j_Page_162thm.jpg
04efe75e7fdc1840d9e7a1183b9b54b9
2cd2aa0386543c8c3fe379d0cf503bd91d1d870a
20343 F20101212_AAAUHY kneifel_j_Page_153.QC.jpg
5bacf8eec7066704b6525ae63cf1e3c1
62d9b1cf112c673d5f299a8f5a849e4a98e0edb8
255016 F20101212_AAATFK kneifel_j_Page_126.jp2
156bd5549de32ae21514a93f4c4be9a4
1b40ea6535f093f1376133805e102d480cc72cb1
1016406 F20101212_AAATEW kneifel_j_Page_105.jp2
f9780501462dbef2f7f3b71213297b00
ab6ce8b16260ee5e01e20c788ae9b18243a2e6a5
73088 F20101212_AAASZP kneifel_j_Page_131.jpg
2ece389f2e031a21085a19c7cab9d8ea
5ccca8db1c2b828b70327027fae19df80758af47
6070 F20101212_AAAUJB kneifel_j_Page_171thm.jpg
bc387e3caa4f3dbb5d717b40f6b264d4
b05219e3a645c0da58e92c0ae17b98137d5d6221
18786 F20101212_AAAUIN kneifel_j_Page_163.QC.jpg
eee8e55b9f6851ba6f72cb3f10f97c53
ed716fc7fc9ec8acf8058063ad2875c735734d43
121187 F20101212_AAATFL kneifel_j_Page_127.jp2
f33bc8f72f972305242f01e703374cbc
462a594d9d2861b38eb3c46efb6adf4e2e1f9106
75558 F20101212_AAASZQ kneifel_j_Page_132.jpg
14b2ccd4909091ac0bed36edeef0800c
86add88302e349b170f02624e4ce728e9a87af14
20379 F20101212_AAAUJC kneifel_j_Page_172.QC.jpg
5491f19ef40516ccb23841347c172bb5
fa03a68eecd3f8735b71a5000b6527c969c4d05a
5212 F20101212_AAAUIO kneifel_j_Page_163thm.jpg
80b1d52991f5d07c62dc9c94200a6d29
243d37339411e91a6f1e971e4a8ac5b9d753dfd7
F20101212_AAAUHZ kneifel_j_Page_153thm.jpg
cd4de8bfd570f8e458d2992fa72f1312
5442839eda8140d4a501cfee30b6d04b7aa8a639
105583 F20101212_AAATFM kneifel_j_Page_128.jp2
9895aa6b7934afd4fc640d45d316baa4
b999178cc79f0876c4c51d72188102eb27fbee25
968252 F20101212_AAATEX kneifel_j_Page_106.jp2
b60606f31d017cec2bde6cd52230caca
618a7ade44fb1c6b019eaa4be5aea8e119eb407d
56077 F20101212_AAASZR kneifel_j_Page_133.jpg
0d84baa1ea87cb92a48d265b91963693
d94ce3a9531a6b84624a662742ca007b2907b11a
115724 F20101212_AAATGA kneifel_j_Page_148.jp2
928db9fadc3d8b53cfde1a5b10c95b8a
d1964ed1b98a08a0933f2fc4875711ee66c1aaff
5876 F20101212_AAAUJD kneifel_j_Page_172thm.jpg
51ad79ebe88750eaeacb0264f34fdae0
6a30e7627dcbe52a81a69d14279c3cf42020ebef
20004 F20101212_AAAUIP kneifel_j_Page_164.QC.jpg
9a888b2d73ecddc53fdf4df0d531eaf5
8376deacedbc75014400a91621694b4b610755d0
F20101212_AAATFN kneifel_j_Page_129.jp2
c24250e4e39ff840cc9d24c04f01a8b3
3113536d2c981bfbb717358f7c0394da2c7edb01
F20101212_AAATEY kneifel_j_Page_107.jp2
5e978e48cbcd9448c5c52d0a7d338ee0
5c14b86bc512b2dcce09a186824ff2777f223723
84049 F20101212_AAASZS kneifel_j_Page_134.jpg
3632d87a32cd824a601cf62073263ecd
cb43439e39e2a141e1e583e6291ac748e99e5e18
985401 F20101212_AAATGB kneifel_j_Page_151.jp2
6129ce9429a23236f7b3eee1169345a9
7212367387eaccc0fc7729f9e38e541ef2544ebb
6198 F20101212_AAAUJE kneifel_j_Page_173thm.jpg
e255c32b0a735c4ed1cb007bffe102f7
98f7ff195d018b0131b19ab41b028ccf8a517dcf
5391 F20101212_AAAUIQ kneifel_j_Page_164thm.jpg
0ba31f1ef450351e929aefa74b446abe
b093e9f6616cd566427f1d331dafbde731166627
119636 F20101212_AAATFO kneifel_j_Page_132.jp2
c34ec37c658c3cdf0f8844768b8fa081
1656cf7002002098fa746f67114a8f70200540bc
50819 F20101212_AAATEZ kneifel_j_Page_113.jp2
6699f9fd8667bbbe04b9b4c46f5acd44
387ccd8692294aeca388cfe0b8644b10db625469
74406 F20101212_AAASZT kneifel_j_Page_135.jpg
fdbc0f12feea6035c0dc76f9a5f916c0
e73002905c6ddf44428db340ee511040144b25c2
F20101212_AAATGC kneifel_j_Page_154.jp2
cff2e6dce2d0d9a5946a6022f9352240
fa09bc158193ebbd733533c9ae2090b7f9e3b592
26734 F20101212_AAAUJF kneifel_j_Page_174.QC.jpg
1be62b8932105c1fe945ba6d86182598
ef982b18dba2f18926ca433106e88bf436fb70be
23244 F20101212_AAAUIR kneifel_j_Page_165.QC.jpg
532fc2929e47a766785b3f0039a97295
85db6ba02613454908a82fc074a2f362f27a9c06
88536 F20101212_AAATFP kneifel_j_Page_133.jp2
f635f1df0e10b0b561b2df4110ff179f
a7cc89bfebd754305506c5ad8ac2de5623cd60fe
69414 F20101212_AAASZU kneifel_j_Page_138.jpg
f1214018510850123f9e3ba5e0984e8d
405e6e954bfb31d164dee30d5eb07d1f9d7537eb
123397 F20101212_AAATGD kneifel_j_Page_155.jp2
e140bdfe289d9ba9a8cec8f8fafb44d9
c40d34e3fb3f28567cf85772969494d2d1a29091
23102 F20101212_AAAUJG kneifel_j_Page_175.QC.jpg
01be4d5904d49f7d39e0125cc470f44a
f5c0d868fe120ff012c51a28acc1013fcf7be22c
5880 F20101212_AAAUIS kneifel_j_Page_165thm.jpg
ff55540356add0024d910a85727d1d0e
b1878aa62dc4e4648e225e22e457b9b769d8794c
113030 F20101212_AAATFQ kneifel_j_Page_136.jp2
90c8a9cce186d43bc10a9110d14b90b4
9e50e8107e5c90245a26c0b1f1499d5d0f2c835f
81013 F20101212_AAASZV kneifel_j_Page_139.jpg
be12af7b7a01b98ef826ab72fcd7f56e
5318f65e4f82fb2ddfca3e09a7af4b6858c46d7d
123123 F20101212_AAATGE kneifel_j_Page_156.jp2
53e23a2877cc6133dd5774e63943af35
557d7d620117d0bcbeb6b10d03f315f5abf4f9c9
6074 F20101212_AAAUJH kneifel_j_Page_175thm.jpg
87ea0e0af345cf098b341de7abbdcc58
4b705ddf0df4a29f35b9049b2de5de0a0fa30b61
6706 F20101212_AAAUIT kneifel_j_Page_166thm.jpg
facfb78d900b1b41dd8580673c0e75e1
10d7b14fd0e0a5b284d722bf05a415dedc273e33
F20101212_AAATFR kneifel_j_Page_139.jp2
387b27bac4331980c2f886ad7dac659d
95152415d5a4eeb06025c51d8462931fe619420d
73004 F20101212_AAASZW kneifel_j_Page_141.jpg
13567a5fa833c1535913c8407d3e0695
62612540ff83fdd0b3b61337dcff3b34a0804d5e
124678 F20101212_AAATGF kneifel_j_Page_157.jp2
13878f41a0a51719860923fd5becf6d6
253dba4aec9e591e69eed39519dd7b5d62ca3d33
21797 F20101212_AAAUJI kneifel_j_Page_176.QC.jpg
fbe54d9a4a06db5c543bf1d98f2b1690
1eeedca398bed17ff4828b4613f2ac6c453a2a02
6699 F20101212_AAAUIU kneifel_j_Page_167thm.jpg
d9ad5a4efb31c67c87f4e7998f961f49
7403536778295f58e6d0bb91537bfbfc20e930ac
1051947 F20101212_AAATFS kneifel_j_Page_140.jp2
6d809ae265c22a0b533497a39c93811a
e39d281e3a73209a3885d2e887121dde80860dec
62814 F20101212_AAASZX kneifel_j_Page_142.jpg
efcdf08095a31ce6965f61ba9e6b6292
ac6f3d0f88d7b9c016327e096b825993892f7eeb
122140 F20101212_AAATGG kneifel_j_Page_158.jp2
4c49cdb54518191de2ad4c74e5743e05
502df946f1534e07e9fad9d2b060601204ea9e33
26393 F20101212_AAAUJJ kneifel_j_Page_177.QC.jpg
7bdb161b1afbc6a800308ce0bc51e585
29549c59c375d06a5e2899fc520910968e950c2e
21404 F20101212_AAAUIV kneifel_j_Page_168.QC.jpg
a1872ca405c052d02234ab2169ec1773
96ad3db9d42770ca71b47a6e43ca0bb614a8faee
1051975 F20101212_AAATFT kneifel_j_Page_141.jp2
59b13970a858c30211605535baa14c9d
a107c3baacbfdd8a0a208c9129994454994b12cc
52607 F20101212_AAASZY kneifel_j_Page_143.jpg
68bf4ce88a6cc24be1c28d8294302083
acfa43f86b8cd62e7610411f5ee0ac6f3898fdfa
109534 F20101212_AAATGH kneifel_j_Page_161.jp2
18b93afdbbc4e76973316e83203c8ade
f3eddcc6a8273dece106b676cc7220fb1bd5271d
6412 F20101212_AAAUJK kneifel_j_Page_177thm.jpg
c2ede7ab0589f2bc2412fa4a58850585
184547eb381190c3a3296165a020ef83914887b4
6158 F20101212_AAAUIW kneifel_j_Page_168thm.jpg
66dd1d911d62f715236bc5e495ebe612
306b4cd4c7d949927c9a43062fc39135acda7dd4
887165 F20101212_AAATFU kneifel_j_Page_142.jp2
4629e11c7dfa53626347bf40191b2138
310dafe160b2d2b98e41f3b069a83a3c2a862298
70477 F20101212_AAASZZ kneifel_j_Page_144.jpg
7f015ad1ef4ba8e0f3127f3df667c24a
0abd65aedebb73b544b96a6e100e65a3b3f1d785
1051965 F20101212_AAATGI kneifel_j_Page_162.jp2
ddcc8c1bccd32d17a2d6fac56007ff04
912330cad32e432b5d45c4273e7a0c24b00b72f9
23603 F20101212_AAAUJL kneifel_j_Page_178.QC.jpg
b3a2f387e9589dd70a4d47ef01602727
5db8f4f49e30bcbf4aa134b056e5058fdf54c560
22639 F20101212_AAAUIX kneifel_j_Page_169.QC.jpg
e0370b1bbe2e00129d015044753ba17c
71cb4d20356ff99f293e16cbc817a6677b460d20
81144 F20101212_AAATFV kneifel_j_Page_143.jp2
e089a799b5ad02e64d2addbc1911becc
a453e162b4e28fbe9fb1c3d00f176606f9051589
838981 F20101212_AAATGJ kneifel_j_Page_163.jp2
931a8ea27aea94509bcbefec2926d691
527b97681f0fe9cd3e5582eccfbab462c4869784
24121 F20101212_AAAUKA kneifel_j_Page_188.QC.jpg
60033ef90df4319c27d1b8ef6043c668
5b82c7f7c1b65765ce70707928d21197b4a4cfb1
6117 F20101212_AAAUJM kneifel_j_Page_178thm.jpg
624de1461cb38719db8bc160b8b9414c
c0fefd78006eae7456f3003f4fbb4467c3619659
5975 F20101212_AAAUIY kneifel_j_Page_169thm.jpg
ddbbafa98cbc154dc48ad59554b530e6
3452e01971f6e2d2c4d4a2430902b70d16f1ec20
113135 F20101212_AAATFW kneifel_j_Page_144.jp2
12b51456a68a8f5e83ecf0326cde7d0e
129c9d95cf57c9c3cbead33c8ed50d00604e285c
863702 F20101212_AAATGK kneifel_j_Page_164.jp2
13b9da3ef7148727a1e6f1389d64ffe0
934234400a6596b549b0e0ac5b8fb89fd693f913
5956 F20101212_AAAUKB kneifel_j_Page_188thm.jpg
5587cfc18670c04f2e1826086131dde7
7ae5de0841822751442f1fade1580a04b3d2de01
19764 F20101212_AAAUJN kneifel_j_Page_179.QC.jpg
0a7ad8fb16d595cd8478edabb475eda6
87b9c58acdc0dfb1d330f894f047c90d23b1e060
27209 F20101212_AAAUIZ kneifel_j_Page_170.QC.jpg
67306280841e9b2d37b43a356edd2a00
0e5fa593f72b074b522b808bf7acce26b6673f5f
1010521 F20101212_AAATFX kneifel_j_Page_145.jp2
fe818c9233f8034cc66b5d345782ec81
4a8f4d78607cca00335ef03ddebd625c17a88682
F20101212_AAATGL kneifel_j_Page_166.jp2
81ceb94eae7b274bfd860633e088fffe
5f4a75703b29ac1ad5a1c386b38d82009f83c179
24628 F20101212_AAAUKC kneifel_j_Page_189.QC.jpg
7c42a9e77d0bcc8df60ffb260549e4a2
9c92f94e2b4ff6206bb2a41749fb480fba85b2ff
5774 F20101212_AAAUJO kneifel_j_Page_179thm.jpg
e8327e34dec1c9ad7c8b37393ed158b7
5eb2d8e9a77013b1f61ae01457ba485c09c8ef9f
887164 F20101212_AAATHA kneifel_j_Page_187.jp2
75439dc47996c0658635ec2e995180af
d9c39632632a82931c27d6c2ba8600ce984aa430
1051932 F20101212_AAATGM kneifel_j_Page_167.jp2
300d2884eec9442379b9348bf18a7aff
1e1467688068fc43e713b41aae6f8a7fb26acf94
6028 F20101212_AAAUKD kneifel_j_Page_189thm.jpg
ad53bfbe95b5f5afab2a5cb314ab2c9c
81d699de1e50690c8d9e5ccc13ab2a8b8bf50c33
6304 F20101212_AAAUJP kneifel_j_Page_180thm.jpg
baa3548a72c8722d9cd49b6ff5b949d7
d30e56759e298b0466a76feec72d64cf64cba26e
82572 F20101212_AAATFY kneifel_j_Page_146.jp2
4073427a607875b685788c28cdd6bbbd
8c96713c199d3c75daf527a9af14a8c840171c51
118265 F20101212_AAATHB kneifel_j_Page_188.jp2
854e049de07152f58de813a85ac02df4
6a46d1259fdd403a61439915768b9d14287c28f3
1004667 F20101212_AAATGN kneifel_j_Page_168.jp2
d71cb866ad3fca369458cf8ac744a1b5
e06fd5525f195ee7b59477408b96ff2d7e5d4a71
12518 F20101212_AAAUKE kneifel_j_Page_190.QC.jpg
e30ff7241e4578ecbaab29f2761a6d27
3fa257dad23d49c3418091d4c864ed4d51736008
23837 F20101212_AAAUJQ kneifel_j_Page_181.QC.jpg
cd8759034ffcb2a9f6cb5a8be4c51893
0157ab070ab117581e9272193374def86a1e6038
104114 F20101212_AAATFZ kneifel_j_Page_147.jp2
c9dfacc8e5fc9f2c6c169bb1e362b8fb
cb95f1ac6bed01b6d79031fcf680ba9b08e44fc8
121446 F20101212_AAATHC kneifel_j_Page_189.jp2
1f899e5249ac64191fb90e2b4a16a0c5
b37856d8e2db985139d62cb5d67a7a72d28b105c
1004962 F20101212_AAATGO kneifel_j_Page_169.jp2
7b9e0a266ad5c99ca49fb862b27d034a
3fb97f8fe70a9bde115b15ef3fd68a1ef4b165d0
3523 F20101212_AAAUKF kneifel_j_Page_190thm.jpg
8240c067816d912cf7e2d4f4bf56759c
ad24126e7cd59b24154e1e72a45451e175bbd413
6130 F20101212_AAAUJR kneifel_j_Page_181thm.jpg
a3ea4fe707451597209f8c44083d212e
233fe9d24be606402a48de456fa4c59d869dc452
55837 F20101212_AAATHD kneifel_j_Page_190.jp2
9e7300b64aa5f86edfa02b0655fe1a92
98f7fc3c5858666629e71a690fe0e39825359118
F20101212_AAATGP kneifel_j_Page_170.jp2
42b59b926906ccab45db5a6c4b586d46
7bed621adeebdd81919720e66397db2f961c93e8
20226 F20101212_AAAUKG kneifel_j_Page_192.QC.jpg
ca6a4075a92dd706fa901f2d1df1bb22
f1e0cbd849ca6416779d694fd1070b4856c61c0f
18953 F20101212_AAAUJS kneifel_j_Page_182.QC.jpg
46f175c73295d29fe4046c644c83c98c
e4b932b959e85f3bd357261d8c0526fb22e218fa
46708 F20101212_AAATHE kneifel_j_Page_191.jp2
6d3a3f8d821677c42bacd3c637c06c88
df4bbb7cfcdfe6df145474fed9534a69d7af9d8b
936513 F20101212_AAATGQ kneifel_j_Page_172.jp2
8cdafa83c7ec5b295a59284a550da044
9ef7a10e4e30b83f06e08116d0971678184462c6
3468 F20101212_AAAUKH kneifel_j_Page_193thm.jpg
5b0ec75f0b1cc7fe810cc5dc6e4c00d6
2bcb8de90ab71ef63b6a5a9ae023f293b50e1ae5
18084 F20101212_AAAUJT kneifel_j_Page_183.QC.jpg
3c68c97ce0b332af95b470bb3a8ef79b
766c8703ca05d6d980ea328c7b20fc65ee31e4d7
390341 F20101212_AAATHF kneifel_j_Page_193.jp2
6464ce3d0faf9b6a1b74a5a10b7aa0b1
e4e0ecc6a816a337a18de395db75590589d3034d
1017771 F20101212_AAATGR kneifel_j_Page_173.jp2
8687311dc6128f293eaf5b084ce77bdd
b7560110119576ad2acc54a0d848ff4e8238ae0e
9563 F20101212_AAAUKI kneifel_j_Page_194.QC.jpg
eeb0b9a6a86f9e061e2cf583eec8e843
2ebff983286b77fece276fe74c38dd4f3558ac9e
5475 F20101212_AAAUJU kneifel_j_Page_183thm.jpg
38cea36ea107dbf9dbb9e85e99a2b036
41730f256147e80376df4e5b44f7802de6626b6d
453175 F20101212_AAATHG kneifel_j_Page_195.jp2
30544e8f06c2dd9803f29f44fa4cd13e
55bb54061c87e960f79c72204a8ed5bea9005c3a
F20101212_AAATGS kneifel_j_Page_174.jp2
215e8820a1e0bef84d7132e66815e6b7
43e6c2b262fda3ae88c47b21540d71405d01ee28
3442 F20101212_AAAUKJ kneifel_j_Page_194thm.jpg
9e85532e9b1428e63f5bfcccbd9994fd
b69e1c683e11cbf3c7117a62542f14c0d5f86090
5471 F20101212_AAAUJV kneifel_j_Page_184thm.jpg
dae4b39acc63e547710203da7e735596
a68fbb386925ffbdfb952c49b667ef11b586c0ca
327167 F20101212_AAATHH kneifel_j_Page_198.jp2
946017f3a3ded49baefe608818928005
d189c621b1c1b7d61b617f55eb850d92d0581565
F20101212_AAATGT kneifel_j_Page_177.jp2
7f390599694adb0cfae29ff0636b176d
a26e3caeb22f0c9f8f71e07c651b599060c71e00
10659 F20101212_AAAUKK kneifel_j_Page_195.QC.jpg
251120e1266a1fd3395a1c40c5343a47
844046b63e019132f6bd6752ca76ab517a690cba
16183 F20101212_AAAUJW kneifel_j_Page_185.QC.jpg
63c5dd9b5756e8ac481d8728b48c98fe
635a7ce3355230aed8032e6a626048e2b95883ff
918209 F20101212_AAATHI kneifel_j_Page_199.jp2
1724ed4c102e7b5fb9d1d2f018e17009
e16e5fb7db989f5fdeb0d663d32ab4fad86728f2
111914 F20101212_AAATGU kneifel_j_Page_178.jp2
7a7b82cac439f6b3b9aafa059347f6f9
84f6278b69cd0a904188dd807591ed10dd595272
3385 F20101212_AAAUKL kneifel_j_Page_195thm.jpg
8b32c6c24592bacd442b493b3eaa8e07
a98ef14ee59afe415f9abcafa77a059b106d0491
26543 F20101212_AAAUJX kneifel_j_Page_186.QC.jpg
9ecfe9061713aca188a1cf7ed014adac
541d6b12febafd40da087c35ff151a32bd391a01
960475 F20101212_AAATHJ kneifel_j_Page_200.jp2
d7177c6b99be57982321f610af559f76
a1ef197485f5f651d0e2032a81a9c62dc3e7ddca
906179 F20101212_AAATGV kneifel_j_Page_179.jp2
6db26d4f5113450cfa700bd10068abd2
f80f03ee31f1c3c540ec712047f3e573305967cf
14772 F20101212_AAAULA kneifel_j_Page_209.QC.jpg
44183705a5e1e24146ed4236c584d9aa
4d4a6e112569e7043e691a86a4b67291814903c8
7371 F20101212_AAAUKM kneifel_j_Page_197.QC.jpg
e581dade82e3b68c6a40ee0b6ffdae81
84a20fd38f9a024e296fb7819ea08f7ef1b7cd96
6656 F20101212_AAAUJY kneifel_j_Page_186thm.jpg
0f751bad0998370489e3fedcf6df37f5
23bd9a3966aeba79e5900b6971a93cd6979b9284
96538 F20101212_AAATHK kneifel_j_Page_203.jp2
d58aaa2d0230cce5b1232041404d59e2
a13de24621abdbfab67eae355e4b73e147dc42fb
121336 F20101212_AAATGW kneifel_j_Page_180.jp2
1352ec165d3d70a0b560f7a12809eaf9
0636651e10120c9e348cc71fb3db6275d4b5dc50
23480 F20101212_AAAULB kneifel_j_Page_211.QC.jpg
4420a4095c2bd3cbb4b76a7ee663df0b
3c1d5633ceace9fc7339edadbc20c730450b9475
19490 F20101212_AAAUKN kneifel_j_Page_199.QC.jpg
1921227fd6fd22da83c3bba52331b4be
747e523f5fc4a88e45999d3c4dcd7f1b12fa110e
5733 F20101212_AAAUJZ kneifel_j_Page_187thm.jpg
0969322460bb4fa483461a0f445ffbae
af55d2665f13131866759377bdbdd911b422ecad
847680 F20101212_AAATHL kneifel_j_Page_204.jp2
6b55e1631264a9bfd97a13ea9fc1dbd6
0654148ca2da81cbd741dcd43d380611b4431988
837789 F20101212_AAATGX kneifel_j_Page_184.jp2
8afc42e69c007018495d49e2688811d2
a72767aaa5ed7c57a219d86951c048ecbe907f87
6212 F20101212_AAAULC kneifel_j_Page_211thm.jpg
b7e55978718ccaf18c0c531630e3bf1e
a0b9deaf1cfe74051fee9bb1f995151dd7f8b356
5739 F20101212_AAAUKO kneifel_j_Page_199thm.jpg
16e27f9103da49ad383a4055b5eeedff
a55be50b0add82d75ad9e4c2c5afae6a94ae36f0
F20101212_AAATIA kneifel_j_Page_006.tif
5ee3ef45bc1a2c93c315dc22ef10ad18
e10efdd59afda48b2124f4f1d0cdca2d7a564d58
33689 F20101212_AAATHM kneifel_j_Page_205.jp2
05d72f761689ee62dd86ef20fde201ab
fe9e5dd0039286b006cef4d4f04e05f2fd02e95b
739098 F20101212_AAATGY kneifel_j_Page_185.jp2
a16d0223105e599b503cff917486f738
55548cee30b3862c8a2687326de64dfa972b53b6
24240 F20101212_AAAULD kneifel_j_Page_212.QC.jpg
0187a71040dff7ab2d544b47276ec1a4
800ab2a12d581e4d6c71f031c77fe17275275bdc
16291 F20101212_AAAUKP kneifel_j_Page_201.QC.jpg
660cece1429dbb84c9cfce9546b7f749
b7c8b3629eaf568800ac59fbd71369bcd49fb4ab
F20101212_AAATIB kneifel_j_Page_009.tif
8bc52d81b484cf23256c5dc12be489d1
c9cd0b35da0b58b5e6e5977605fa008d88a962b8
106742 F20101212_AAATHN kneifel_j_Page_206.jp2
a75e2c942e28de81302d93214a6566af
d7ff0f4c65b558d74ccf5697e5536aff07f5af65
6101 F20101212_AAAULE kneifel_j_Page_212thm.jpg
491c4cc762b7ae88ed35313f1eb3a0bf
747fd19e97933edb606a2204c33bb0448d5fe7bc
21718 F20101212_AAAUKQ kneifel_j_Page_202.QC.jpg
d2db72a29dbd355d884f8a03a25bae43
819c5f699f1f9dfc2f373334331928377c9dd24c
F20101212_AAATIC kneifel_j_Page_011.tif
346d4a701201fb4b4efac54cdb03a75b
62c134201b55ec3adb86bff137b65dd930bb37a8
1051949 F20101212_AAATHO kneifel_j_Page_207.jp2
344bd0975f2b4d8cb614c3553b4b4dd9
dc34a5da53840c9a2a1fe9d94f1a4433d4f9d937
F20101212_AAATGZ kneifel_j_Page_186.jp2
68c2cfb9f4055ef0891c0a088408a3e1
4f9a968e43b6112e4203fd1f949f58f5ed9ccb31
24995 F20101212_AAAULF kneifel_j_Page_213.QC.jpg
d7ddbab70af995ce77e2f73e49eddcbc
f64c81e75ce075cd62c2d2e70c28d355b5add8c1
20172 F20101212_AAAUKR kneifel_j_Page_203.QC.jpg
6b7c4728d400f9b9ce7f8890b1db7bc9
09ed52b79a0986055e094e362a1d16bff05762b9
F20101212_AAATID kneifel_j_Page_012.tif
ad7b2eb93038ff5583a831134895cf32
ba16a08f664feb16848c45c4dd4403fd5894d74e
1051977 F20101212_AAATHP kneifel_j_Page_208.jp2
f671861affb1ff3c9a41780f89c2fdb9
40964685157e76211bb27bb9f5de0973e062580b
6376 F20101212_AAAULG kneifel_j_Page_213thm.jpg
8236855b11671f2d31d7d8e0733447d9
d502d6189ae63e0fb1daecf96076d22f7f9786bf
5703 F20101212_AAAUKS kneifel_j_Page_203thm.jpg
c9f56f22ad98cfe15132c6ad1ee78815
6fef434bb80457a49d38889be0a724bc3f4c1312
F20101212_AAATIE kneifel_j_Page_016.tif
2ccaf8b43b6dd89b4a4a000c7af4c34d
933f92f8088bfc61e1a8950e3e89ab9b9c8eeb3b
68683 F20101212_AAATHQ kneifel_j_Page_209.jp2
686dcf58dedf641e7e6bf50b9801246e
31dfd4b2ecfb2220f8e0a246385b4bde98a889c4
24635 F20101212_AAAULH kneifel_j_Page_214.QC.jpg
7650b5f7bb04e81355a3dcb34139bfcd
41e056aa08d3044309657f1f48da1440482fafad
18659 F20101212_AAAUKT kneifel_j_Page_204.QC.jpg
180620a8a02d938252567e78d6ab9fc7
743c81af976a68e300f4867330412651565a2e6e
F20101212_AAATIF kneifel_j_Page_017.tif
72bd63e4a703197f6b60a09122bbb259
ab681325a290b66e42332d9570834ade47955a6f
127005 F20101212_AAATHR kneifel_j_Page_211.jp2
848ac564568a9f5f9efe3de2be03d2ca
2a5cca76fbcbf207feaf142f288115fe0b0b722d
8450 F20101212_AAAULI kneifel_j_Page_215.QC.jpg
9db7a42b9a1f979d0c5260be70d0b4e0
eb0489ac45695afe32d17ad19c5e4ea2d1560183
5266 F20101212_AAAUKU kneifel_j_Page_204thm.jpg
8db8b4c3a3e53a7d3a93dc899599d2ee
0a9b9c5075ed85292e48ffb21dac7a2824960ee3
F20101212_AAATIG kneifel_j_Page_018.tif
9fb0d01576e79f89def450abf59be29e
eb6e831669e118232e4c505eb8c3112d255bea9a
133734 F20101212_AAATHS kneifel_j_Page_213.jp2
0b0db74513b1187a253701d5fbcd8e68
5d6872016b750fa8f21bf0cf738a8215c9ecd183
2512 F20101212_AAAULJ kneifel_j_Page_215thm.jpg
fdf19ef80701605023724eede17e25eb
3043efdd5cc779516620f651c6d8bfbe1ebcb67f
2830 F20101212_AAAUKV kneifel_j_Page_205thm.jpg
f835a9bbfad8d1462872b6ecea22eb94
00b0c937534af09c7bab22ea6adecb8ce922ba2d
F20101212_AAATIH kneifel_j_Page_020.tif
2a40974e29ecea2501192107703e9074
4d0e6ee74bc93e6b53d0f14d199b7c681a47308d
136348 F20101212_AAATHT kneifel_j_Page_214.jp2
34f542004391e740943f33ddd63f6052
f4a9372116dcb98ff2dc1fbc8a67d8d0054c8fb1
20395 F20101212_AAASFF kneifel_j_Page_106.QC.jpg
74a778bfa84cccec34f59821e53798c5
a64d2b945cba721222ca9698220c10b3bf40b257
13250 F20101212_AAAULK kneifel_j_Page_216.QC.jpg
56c540cdd997ea06915fd1925b30c9fb
1f7ddeec23c863893b3a2bcc1aa4ae9a27df289a
25491 F20101212_AAAUKW kneifel_j_Page_207.QC.jpg
7ab4df17064ad9e4a3ebb38855c06905
4af025379c9aa2a6b1043ce12e87825df2b03fbc
F20101212_AAATII kneifel_j_Page_021.tif
289f2f6565876f9ce43f0ca8c6abdf0c
a33e7631d29924f436617e19b928c0c0d31246f4
36657 F20101212_AAATHU kneifel_j_Page_215.jp2
28634f84d09748beb108f5a97e69f31c
dea042e2a0536df66ac64db96942ad8400f521ef
104795 F20101212_AAASFG kneifel_j_Page_023.jpg
ef5cac38f9c065e442ee448d645da288
3dece4234b9666f440fbd8d80100448bb0f0af7a
3554 F20101212_AAAULL kneifel_j_Page_216thm.jpg
84e6786ad10d0f3eff30a21894179793
a0af18fd7f741494e61b748ff08029544a0ebad0
6561 F20101212_AAAUKX kneifel_j_Page_207thm.jpg
ec60486e9df11c9a96c36078ae04a89c
8de012b37a287043c195725fcea8dd97e70b5fff
59023 F20101212_AAATHV kneifel_j_Page_216.jp2
656bcfebc670918358b3f68b1f87b2bb
09dee69f3a0d698b4567b30081033648ffac3efc
2296 F20101212_AAASFH kneifel_j_Page_078.txt
cfe3d343441ecbdc0bcd833ca3f77f90
0b46cad06c2b2fa8de3d572d0e99088cca9f0999
F20101212_AAATIJ kneifel_j_Page_023.tif
f097eb39877a206aec448e85796c895d
c6883514ce64e20c9656e1de1d89b2bebb59588d
248623 F20101212_AAAULM UFE0022518_00001.mets
262656a945cb69630f550cc6bfd92f88
cdca222900486a229b6b8414a33ffcae215f19a1
26484 F20101212_AAAUKY kneifel_j_Page_208.QC.jpg
6bbdb8acb3659becd43a671408a069f7
d97ac0e4125a265e6205fb11b44b5100d4aaa538
F20101212_AAATHW kneifel_j_Page_002.tif
db909af960e7093d603e91212693d86a
e34c86858c63af13f9faafab5d2de7f3d351de6a
22271 F20101212_AAASFI kneifel_j_Page_001.jpg
3597c5168daf1295f2a1a0e87cc96014
f0c02413f3704b19778cfd180738badf42d93d3f
F20101212_AAATIK kneifel_j_Page_025.tif
eb92fb15cacf82ec25864891f3c3458f
4f2fb100490fce8796237777e115229341405227
6606 F20101212_AAAUKZ kneifel_j_Page_208thm.jpg
9a2b6c3e7e6cd8ba117179dd45b9d349
4e0aa5c8d5547964a83037f90acd73dc81362803
F20101212_AAATHX kneifel_j_Page_003.tif
075820d1ca1a22f874962e2372c22401
607d526c07e61b8cddfa435d0d95c5759a2dd8a7
F20101212_AAASFJ kneifel_j_Page_036.tif
1e6bfbd8432a2c391b7caae8156a45c0
708cdeebb83e4a77bf6eff65871bf8a7c76b4c53
F20101212_AAATIL kneifel_j_Page_027.tif
65ccb43b4b44c5124a7b0a170cecfbe4
c3cd396a5202d61b2984cad2d066c0567f8945dd
F20101212_AAATHY kneifel_j_Page_004.tif
25771ca5cc35c137982235b04627e95f
1c32d84030c2eb3d717a289a3131f0f15ed954c8
F20101212_AAATJA kneifel_j_Page_048.tif
c81dfa17675465970fde377e2fb9e60d
461ca6184a73e3d8143a3bc8efc1ff68a793cc8d
F20101212_AAASFK kneifel_j_Page_109.tif
b94bb6b068970df94ceeba2ce151da68
aa82ee081df1417a92e46a7453aa88a6282198c9
F20101212_AAATIM kneifel_j_Page_028.tif
9266423c50dc3c4eb7738e1f9724bf7a
4363bcc5cc1032505ddf03f81ce3166121c5e004
F20101212_AAATHZ kneifel_j_Page_005.tif
f83a49fbdca541882ab94836c6e4ea01
ca5f945474d9b93a45b847cbcda84644b3dbf103
F20101212_AAATJB kneifel_j_Page_049.tif
722635a16fc052dc23affd3037eec7b1
bb9220d4a904b5cdff912dba2c039811ba6b9421
38831 F20101212_AAASFL kneifel_j_Page_116.jpg
d1594a14a032fe1f98af01c68a2ea18a
ab0e7bec73a63f51f1a2e9c79423a45080f07bf4
F20101212_AAATIN kneifel_j_Page_029.tif
f9bb0f4eec57c5862e7f4f24557cf897
2a0571bbbd8d465e082671f54416a3ff3aaf6d88
57101 F20101212_AAASGA kneifel_j_Page_149.pro
3e83d8a9a808832308a6849e26d471d7
a22774a8de1dbbd348f7f8ef29c8f20d17be1db6
F20101212_AAATJC kneifel_j_Page_050.tif
950211c6b7355060b2568b62bb90062a
3e5b1584061a590a42419be5965ecef62e9a6edd
F20101212_AAASFM kneifel_j_Page_032.tif
4b10859f4d0e89b731158bb946a7c47c
a524c23a1e1c1127b85bbb7185324e7c12aada64
F20101212_AAATIO kneifel_j_Page_030.tif
0a961e3f90a673025c9e9b9d630cd92b
dc3b89649fdedd1bce14d5994844d6927fdb5042
40279 F20101212_AAASGB kneifel_j_Page_204.pro
58c861173e1138bd69abf06b1d089485
a350a2f99fc22428a3f51c5ff2eecca9981ed432
F20101212_AAATJD kneifel_j_Page_052.tif
fc303578db0df698285fdb72506dd5f3
4e229cb3416bbcee411186869f9a57d3406cb04c
24095 F20101212_AAASFN kneifel_j_Page_102.QC.jpg
d9f8150e7c0cf6ee16db2c3638b8d9ee
7dcdfa36cefa3e24194aaeebeca7f60bd12ce850
F20101212_AAATIP kneifel_j_Page_033.tif
2f4ad8bd437d66435fa02f67049cbeac
d3d0ba92f3b56b1d64530827d56467422adcd599
F20101212_AAASGC kneifel_j_Page_033.jp2
a0952e6183c40b02472a50652d73ece4
ec8443f24888afbdbf1e85ec8c5324f5550c371d
F20101212_AAATJE kneifel_j_Page_053.tif
5b8eb0deecc0f2c8aee91329bab4ad13
bf3bc76c2b6ba6f60f574ddf164f00a7df29ec28
38649 F20101212_AAASFO kneifel_j_Page_143.pro
3d2e2b4c43f52d15bc45345e4732d3d3
7b17f9a9a0ce022269619e75e4f39e4ba8a6f7c6
F20101212_AAATIQ kneifel_j_Page_037.tif
aec1a333c3791b697f0ef5d7a6bb339b
f954be44ad6bc3240bbd76c2abf6dbc672bb5fc5
27309 F20101212_AAASGD kneifel_j_Page_038.QC.jpg
2c7a44cdb168d29cee8ebbe5f85d0241
b7200bd26e1f98e6a35c0efbc0d0c01b7ca4f641
F20101212_AAATJF kneifel_j_Page_054.tif
d5874a0bbf8e98770129385d6a700eec
cfd5af8bf3a6dff0dd38251a0d2ed97390c66ada
60988 F20101212_AAASFP kneifel_j_Page_179.jpg
77211648250332ece33f3f4c012fe389
471a6138138a632358bf185b6ab85b58287c9dd4
F20101212_AAATIR kneifel_j_Page_039.tif
da400dd7c695c5dcdc259778ac9947af
63782859070e9bfc2d9f93bf42e8f2cc27d02cfb
63756 F20101212_AAASGE kneifel_j_Page_199.jpg
84147f483a8f6402c7d30fbab4f762e9
ac38e8f9e1ed80376f08a910211282efdc1222fd
F20101212_AAATJG kneifel_j_Page_055.tif
ad343d6ebab381fe655779f2c6f81190
30e464cddce178b117fbf4b358f9cf41078aa5da
80111 F20101212_AAASFQ kneifel_j_Page_074.jpg
d487b8f79a72246d7a4ae9380563b4ee
30563f6b59f1e8cd78ddd915becdb01a52974bb7
F20101212_AAATIS kneifel_j_Page_040.tif
f229b53eda6eabb645fc0676ee08dc2d
efd65b43e105b5a7bc8fe5bb375f217b4f9c9429
42174 F20101212_AAASGF kneifel_j_Page_103.pro
c299e58d01696c665815f5fdac22777e
37d5595c154dfdbdc2756cc55340d46c51552bd0
F20101212_AAATJH kneifel_j_Page_056.tif
fd74dba82e37eb47226252bd8af3719c
b6b23f7ca48c0037c80fb8590833a5e110da136d
F20101212_AAASFR kneifel_j_Page_001.tif
48f7e89db85e07b6a993702381d0023d
f884ad30523dc94f9a667ce0d78fbdc1bab7d7e6
F20101212_AAATIT kneifel_j_Page_041.tif
76eff0b35487c1d08170250c8b784d57
5a8283e1f62b89e19852beeae7ec443d8c148890
5410 F20101212_AAASGG kneifel_j_Page_081thm.jpg
8934ed62b5e1119a3f4cdf896e918127
f0e43a4c21a02e0669c80c2f9972696bd77aa0db
F20101212_AAATJI kneifel_j_Page_057.tif
e551ad1f6b885716517d6b098fd43d54
44fcc13b55ddb1ea4637994adbdb7cbc5572b6d8
2062 F20101212_AAASFS kneifel_j_Page_141.txt
76121eb1190a156567ae7a39615a51a2
02f89e65e813b4fa48d07af591ff824f12515098
F20101212_AAATIU kneifel_j_Page_042.tif
dc76c3155494a9d2915003da84bd771c
d59fc08ececb919927d7840ec0f05de31df918fb
100648 F20101212_AAASGH kneifel_j_Page_007.jpg
7776351a1829cd82ebb947d7dee78788
5fe70aeba63985186a85fbd6797d9c89e1e5ed42
F20101212_AAATJJ kneifel_j_Page_058.tif
dde222e097bc1f2ea2649768fc559387
32bac0dcd48f49449f5b361d41d8741dc22d495a
25785 F20101212_AAASFT kneifel_j_Page_112.jp2
93b1a56697e72eebae3be14019d60ecd
b27813534c0e703c3f8ff23ac7ed34bee2165aba
F20101212_AAATIV kneifel_j_Page_043.tif
db8d953c7e9140607120a3400bbb7c19
8d540fa46d62d375f26a820b15ce4d1d1b015475
1916 F20101212_AAASGI kneifel_j_Page_146.txt
70a6c5abace25945b51cc0629460622e
a5e5043b66e96ab78d52d17ff3882aac0135633c
F20101212_AAATJK kneifel_j_Page_059.tif
7f7afd255dfd276510240a700f0cd5a0
d5b577392f9db0a6ee29732dd90102c4d843f430
F20101212_AAASFU kneifel_j_Page_022.tif
824cd3c550a280c454cbbec349ed109a
fbd7fc3991a63f4b55d99a64772a968a6cd84cef
F20101212_AAATIW kneifel_j_Page_044.tif
531f0741d398d2169b3acc6b18ab499f
189bf43a95bbb0e1b1617ae902a9d6bd63c3c925
F20101212_AAATJL kneifel_j_Page_061.tif
33d6024590e17d2f44b4f8b7700aba17
83d5a075120569482a9f041a3988dfbfbb48b080
1051964 F20101212_AAASGJ kneifel_j_Page_051.jp2
2b289ba402614aa3c0da2238b3156b48
d429e550e5a886b3aa669d58f290394a2768e4bb
9904 F20101212_AAASFV kneifel_j_Page_004.pro
f60cfb3b6e0031e3f27424fb4846c6c7
7bdb4930b58dcba72b510124f4439a456235150c
F20101212_AAATIX kneifel_j_Page_045.tif
a0500a4538ac44a3af70166ea33b795a
4378edcbb9a36ce9d70eb21aba7a69c4bef02ec0
F20101212_AAATKA kneifel_j_Page_082.tif
d0ef5e45ebac850a1fdaafe7f14e26c3
91408119faf6d34e2afac4c6d181d9ba430f1deb
F20101212_AAATJM kneifel_j_Page_062.tif
9799b520cfcde706debf62c68d633b4d
0c3e686aaaeeead6099f0a02d47f120093b1bf50
41161 F20101212_AAASGK kneifel_j_Page_133.pro
73d61588ae2c24e97b7b604b44f7b169
f749f642b27a8a70fde94cccac591761125323e4
46129 F20101212_AAASFW kneifel_j_Page_173.pro
603d1001c4cc6964b300949b246b65a8
5ccb6723d2b6019645fb7590b1b56bc51f7dd893
F20101212_AAATIY kneifel_j_Page_046.tif
ab96fd40a43e8182aafe31f2ab3aa896
9256ab6663b9140d0e73cccdbb564548fe9ebbe1
F20101212_AAATKB kneifel_j_Page_083.tif
ffae2bc8e9c35d02fe6423dcc9e45fb4
a794e7b5903b1db6bcd01f1f6916b7d73df5eb8f
F20101212_AAATJN kneifel_j_Page_063.tif
0b227f57cb0b8a251b142f935c94185f
cd4d3bc5a557e7037b6253fe2aa161124bb20fb1
6341 F20101212_AAASGL kneifel_j_Page_135thm.jpg
f8af8a7e727eee5f714c2b2c185d89f9
f24be4b4e0a4beb524a1a3187f2bae460c96332c
3323 F20101212_AAASFX kneifel_j_Page_210thm.jpg
a68f7eab43837319a2f16fa400d9cd77
c85389dfb89701266c7edbd3f27f010c29a83108
F20101212_AAATIZ kneifel_j_Page_047.tif
5b113dad76deed79a4467d64873ccb2e
da65a2502c68288f090dcbc16d6fefe513d80af2
F20101212_AAATKC kneifel_j_Page_084.tif
bdf174901415ff75a10f6c4de9b24854
9651b193be069504c2215c2adfa6870fa79558e8
F20101212_AAATJO kneifel_j_Page_065.tif
3bf340833795c0f29de9deda9fdd1d40
ae58f8c85cb2ed935ad2d1cd5c9dbb30094eae24
83713 F20101212_AAASHA kneifel_j_Page_212.jpg
634edaa8cc5d216d4aeaa93c59ab41f3
a2497f378c4bb8f3e2d427306e6f1b8a39bb4b81
6079 F20101212_AAASGM kneifel_j_Page_040thm.jpg
6038ba1cebeaebcd21298d161cc9f8d2
e23b28b4b39564c3cacb1dddb3089330923b264a
1024764 F20101212_AAASFY kneifel_j_Page_097.jp2
88ecdff805b5df505579456501d494b0
106d4f05c315d99f5034a87c9f26aa8adb26880c
F20101212_AAATKD kneifel_j_Page_085.tif
3e666b478cc6348b12457a6314428d52
7e4077ce205bb055c9a2b4cb3da84d7e2669b89b
F20101212_AAATJP kneifel_j_Page_067.tif
af1a26e4884e548db90a78226b8ba624
16de59acaddf906b2b1ede33dc47b954d0b8674e
5679 F20101212_AAASHB kneifel_j_Page_202thm.jpg
b91bddb3bbdbcdb6b010cb7f78834153
7165da0b447a4890fd8bd6029f1a8b25dc4f5172
2722 F20101212_AAASGN kneifel_j_Page_115thm.jpg
cd8168bb780b43bd7e901b8984fdbf6b
a16b84af10d5d4f31b761a11a1d814e06e8084ed
F20101212_AAATKE kneifel_j_Page_086.tif
7ffd26ba5fe5f5ed94fac4ab37403afa
5d3542b37d6f6cd74ee904892cd69b877465545a
F20101212_AAATJQ kneifel_j_Page_068.tif
4b3d021a9bf8f0f387e64a1b587db310
c5cc62efe0bc3bb65c08719b73126cae185e744e
8567 F20101212_AAASHC kneifel_j_Page_011.QC.jpg
cc87af3f667f4937e28227cb9f0d3f40
d2bbc6e2dd5449bca1b29193101cf436abc4f209
22461 F20101212_AAASGO kneifel_j_Page_067.QC.jpg
d8f8433c1d5653703e2dc6372ad8604c
6660ab9f222a17adbed91e70d9064ae81597386e
6948 F20101212_AAASFZ kneifel_j_Page_174thm.jpg
681ca1552d768f4192a3ad7c06981a6c
e3617d97bbb311dcaf2aa8a56a983aa01d819e2f
F20101212_AAATKF kneifel_j_Page_090.tif
e378b39c2d49963556af60d9443b7b0f
d44660f7b26bda0ed7626aba38e84277ebdd4a45
F20101212_AAATJR kneifel_j_Page_070.tif
b5279f697b86f2769b54bc47e01b3256
7107a40a3fd241c6ea17d201c504e19612a8e3af
1784 F20101212_AAASHD kneifel_j_Page_200.txt
84eeeffbf7ffd8cd87ad53ec6b1308cf
b3d80ef86b941aa7a651f7cad9b3fe54a1076b45
F20101212_AAASGP kneifel_j_Page_170.tif
9dc235ef451eef91196185b2e60b9d29
34af41f02006cda7430aa16bc65d2524501d1133
F20101212_AAATKG kneifel_j_Page_091.tif
c4928192119b64cc8d0624d472cf85ab
3f011ea2f13714826104ac2267572f909e9e5348
F20101212_AAATJS kneifel_j_Page_071.tif
e2c04c8780f1d6a7c762d309576e53cd
c03345fd08b4085fbc8194bbcf455bccc252cf46
5439 F20101212_AAASHE kneifel_j_Page_121thm.jpg
ef7bc2dea6616faa52a329aef8c64e1b
d4f297c4096467369d9b7ec0b9815798b99830f0
31661 F20101212_AAASGQ kneifel_j_Page_195.jpg
efbcafbe2488d05ef0dd72593c6dfcc7
d8762ecc25885122f5137d88b29e2ba2d79186a0
F20101212_AAATKH kneifel_j_Page_092.tif
3e1d4a58bcbfbd3c52f5272d0f1ecc30
4b449e314725a6a31578336a732914c311055eae
F20101212_AAATJT kneifel_j_Page_072.tif
b380c8d2dc3b7d23a533935c03e914d9
b6cba569bfca517398c0c9bcbf1e33a14b94ccd5
9101 F20101212_AAASHF kneifel_j_Page_210.QC.jpg
2e1e52774301509f0ee5a51dd8dd73b9
d1b0b1c350804d3e7a0621dd73351abd5033b1ad
F20101212_AAASGR kneifel_j_Page_191.pro
badc60429d9102e04af5b92a78898c1a
c3415f01491ce76486cd6717bb69b73018a55683
F20101212_AAATKI kneifel_j_Page_093.tif
f69e10f56937be3b683b33eef29e4caf
347979f0718a387f9c2d1c408d97d9d9a6d8be39
F20101212_AAATJU kneifel_j_Page_074.tif
fbbe0a156f9aaf5b86bc3e5f81683f45
870c56294322b2d8b9abd5ffa2f0b2e559890931
4002 F20101212_AAASHG kneifel_j_Page_209thm.jpg
8edb86e37c6c017ad7348970393de317
56651c0563c1520112ce632588a2feba6a269cad
51707 F20101212_AAASGS kneifel_j_Page_139.pro
594070adcd024bce1b7cc9be9724ac6a
4c8156430fd0c1edf70cf450ae6ee5d9b9b25413
F20101212_AAATKJ kneifel_j_Page_095.tif
ecbd30167a94570af68bf84c48d1d169
a43be1825fcb5ee1c6fde6be721112fd6cddf80a
F20101212_AAATJV kneifel_j_Page_075.tif
3a6781220c9bd1c4876d9982d905dd32
26947c8ffed694ea0ea87845266d59fbb5360897
5946 F20101212_AAASHH kneifel_j_Page_057thm.jpg
39d1ce05459147e098d143c30e59c0f9
a75163f68bc34af219dbf02dd5cf7ffc09916f43
28636 F20101212_AAASGT kneifel_j_Page_193.jpg
14c700dd726f4d392bcc6ae3443e310a
e20f5144c148c759d2f6434d89e5f92351516e83
F20101212_AAATKK kneifel_j_Page_096.tif
b4ccdb0bdce9aae8a5b5908cc4140e7f
d297250f7a2b7fc4c94662c2ebc026b75ab91b85
F20101212_AAATJW kneifel_j_Page_076.tif
83e2ff3d7ef09a281d65ac1808c01b7a
7af3b0480c6aecaa09d68873308f8e48e37edba8
8784 F20101212_AAASHI kneifel_j_Page_205.QC.jpg
913de9d6c560c3cce64c5ee68f8dc4b4
d7d1af585ce6eee2d2f21419075b029f25db829e
F20101212_AAASGU kneifel_j_Page_116.txt
a57e48518f4d6c716acb355c62d96469
a9ffd1c52e8b04c5bf17052855c594b09bc60e87
F20101212_AAATKL kneifel_j_Page_097.tif
c51297848081d43110ccb26665e8f46f
f755e1178526154d3b92968dde38f220a853a33d
F20101212_AAATJX kneifel_j_Page_077.tif
c734876ef082994e871b570b882a7f8c
3e78b6d7a73b0be88d4297fb7e953f6c842dd212
F20101212_AAASHJ kneifel_j_Page_038.tif
df799e42577c8a16043f2e9280d84d3e
073272c33dd92bbfe11711ddebdfb07da0a96db1
90649 F20101212_AAASGV kneifel_j_Page_182.jp2
094d6161d71e49157e246ba44021cf1f
f418b4f65c82a8fb83b8d4bcb1312cadddff559d
F20101212_AAATLA kneifel_j_Page_119.tif
81906425294e8444987330e08e603636
8c427a812a9ee93df8a78cd11603d89b75868232
F20101212_AAATKM kneifel_j_Page_100.tif
3136c4bd51430ea1cefc73194e5e41e1
6d2ea4949525e4acd7f95bf3a15f08be25cefc64
F20101212_AAATJY kneifel_j_Page_078.tif
dfd7ec6d81d46bc3b35942ab3fb1f08d
1424b33b0ce69b552087e7fe8a6f5129649201a2
28502 F20101212_AAASHK kneifel_j_Page_100.QC.jpg
4be51c9ea6627fd37ad183daab2122fa
d672b3bbb7ad80e4450ed966b88ed6541a112f97
F20101212_AAASGW kneifel_j_Page_152.jp2
550ac304fb988b881a524fe11fad4d03
1a80a02d03078a1b0a009c1f87c85397d49d925c
F20101212_AAATLB kneifel_j_Page_121.tif
eec5f449ebbc9f672edc833482040e8e
fb911ef746a63ff8945c20ad52dca5e208ef69b6
F20101212_AAATKN kneifel_j_Page_102.tif
c2fccaa2537ce4ae74611f177d119cda
6b698f1fdae71ea9e9f2d62d69198b89f1690f4b
F20101212_AAATJZ kneifel_j_Page_080.tif
f6d4bdcb48f8744db947609edd82b653
e96994183da1024a829369e59a0ada5676dac0b9
1353 F20101212_AAASHL kneifel_j_Page_163.txt
4eb4aec97ff42bde0e78502234881c9d
1662dac6d125ea700fdc44405685088adb324f07
2174 F20101212_AAASGX kneifel_j_Page_207.txt
d088855d7c18113eac8ec546f9fd6178
ec069bbd6ed9b30a97bb77b3127f7dadf3ac5062
F20101212_AAATLC kneifel_j_Page_122.tif
3460878f90ea51db1cd6db639a960812
f5137319bddeeac4c38bd60a0ca255180a37fe22
F20101212_AAATKO kneifel_j_Page_103.tif
90fffe173ddbce4645a74854147ca74b
b2ea18c4bbd6bf4406c2030c93cf69fc3216e6dc
6292 F20101212_AAASHM kneifel_j_Page_061thm.jpg
aeae3bd26aaa8c463ae449e106705186
639b8167afff79ba286bf9a88c2cdaee41c84675
2295 F20101212_AAASGY kneifel_j_Page_093.txt
778c8b060d5be7ad187ad949217a92cf
367a7805da8c7f1018f343fe841a701dc7297727
24915 F20101212_AAASIA kneifel_j_Page_166.QC.jpg
a4129fd9f64a77bd2009faa557bc10df
4edc37a6fa4c8d929e046ddd7738326d68e3eff6
F20101212_AAATLD kneifel_j_Page_123.tif
c3ace80d7d78349849b70e01c2d6cf5d
09f360d330be83ab37dba044a107aeba8afff83e
F20101212_AAATKP kneifel_j_Page_104.tif
8806ee1fd249b646e14291650a6b8d48
f9514a1f4764847f71f615a4da2046c5787fe446
1051907 F20101212_AAASHN kneifel_j_Page_032.jp2
fe3d6b6d842a3c29f89d792bedfefa6a
5bfcd46ef7a551737bf084995c1ebc23de1c50e1
6283 F20101212_AAASGZ kneifel_j_Page_132thm.jpg
0ebd14b8b9e961e6abe258b73b0b6180
3e2cba291a0381992bb92047a1bf4ff3196f6eca
F20101212_AAASIB kneifel_j_Page_140.pro
ae0f44e7bf82c957d078d4062704ded9
d044e9d2c7f55e06798397baa29bf43f28c43501
F20101212_AAATLE kneifel_j_Page_124.tif
3b26786b39e5fd46e3fdb87b119ec6cd
7d85183008658cc5031f17c0377bc3a6ccb07224
F20101212_AAATKQ kneifel_j_Page_106.tif
4aae9f2d0b426bf3b512237a46e4a3de
4d6b54f0ce007b628fe3d2911b5d9ac969f35a36
2118 F20101212_AAASHO kneifel_j_Page_076.txt
644af9aba845ddc3a73d7f7088525ff6
bdc766683cf24a501761494d3f8e5f31f426e4fa
449 F20101212_AAASIC kneifel_j_Page_193.txt
bb93b4c57077191c5ac13286b636f328
99715f2381583901ec11bb83668aa210539182aa
F20101212_AAATLF kneifel_j_Page_125.tif
33c266b0c401475877a7a8972887007e
f4ed729430a085c3fc18403369d15ad4824111b3
F20101212_AAATKR kneifel_j_Page_108.tif
f5489d66f0340334878d020781c4e727
a077c0cdf1ccc25c760698a81eda10efb497cfd1
40343 F20101212_AAASHP kneifel_j_Page_073.pro
12c9db9af82d303d0f6b2fb63eba3d04
e96fc7224edff6edbd3768b04fb8435c2362e6f8
5684 F20101212_AAASID kneifel_j_Page_103thm.jpg
be422687ef6f562529e1fef01f2a5611
38e6e1bfbcf589798e8289ccde5798718d758170
F20101212_AAATLG kneifel_j_Page_126.tif
16a2de57bd554b860b8c984f13394ecf
d0737cb354780a1f4c51706863150c7f2f581e27
F20101212_AAATKS kneifel_j_Page_111.tif
5760e67bae1495f862a4f85618708e31
7fb20b7cfbfdcdaa812c7bdbb32e239c7b9d9e56
56362 F20101212_AAASHQ kneifel_j_Page_186.pro
a84e41c978779bf1fd44acea516e043c
e50e1cd3b4b5c0da4fbed1ae117a7680aa63afd0
52029 F20101212_AAASIE kneifel_j_Page_074.pro
9f106e298524ddafe658ee04977ba79d
c4b3f15d9f622ff538e9d3ebedf4d02d576fbef6
F20101212_AAATLH kneifel_j_Page_127.tif
4b2bb6ef57089d20acb69a2f68aa4d61
b9a5fb24ffb6b64cc3b6b0b94044dbdbfe418a8f
F20101212_AAATKT kneifel_j_Page_112.tif
dbf55b971a1e79c45b950ed9c8f5b509
7dc644c99cbd141b923a1efc7c06b71104ce382d
57549 F20101212_AAASHR kneifel_j_Page_060.pro
80ca0168555ad68f3e0a9102d4c31dd7
6e1a48d6327519e7f4d16f87d970b39438fef58d
F20101212_AAASIF kneifel_j_Page_087.txt
1e1c20a30dba4d13a1529a11b2ba1c6c
a8daab76c5ba536fa67e3f87ae78b7ee06a659f4
F20101212_AAATLI kneifel_j_Page_128.tif
39b3ac938bbb41dae272bbab403bb6f7
31aed8b01264201b35d3cc97f640ccfa8948ff1a
F20101212_AAATKU kneifel_j_Page_113.tif
a6f855c6b99ca9b5d1c013418c85771a
a9ac828fc4c6decf2a646c9273323a8fb611be75
77770 F20101212_AAASHS kneifel_j_Page_130.jpg
9380bc9711d83e4236d844fe8b5d897e
0b88947e59d42d78751eedddb1aeee35c5c2a476
6401 F20101212_AAASIG kneifel_j_Page_157thm.jpg
d307804f33206b79c88ab8386b824c80
6160d65d5069325df63f2797ecaa59073ffe1a3d
F20101212_AAATLJ kneifel_j_Page_129.tif
714ce4bc5d9e4276afea4d689ac282b2
ae68e71e41cb8891aa6285d46136954c23f27e3b
F20101212_AAATKV kneifel_j_Page_114.tif
3838c832f5a656d91744707460ae5e42
b2a3215551440f75b685e7ddbc2544bf5a8d47c7
61326 F20101212_AAASHT kneifel_j_Page_021.pro
f2471c8656f7108aa8df7b4af06063b7
c3829686bee711c6b94918e6de01398b6492c44d
56105 F20101212_AAASIH kneifel_j_Page_089.pro
275853e9098d3e4605b62aa4bd33c7fb
482e230601ee81338c37e9401b1848e256f7d510
F20101212_AAATLK kneifel_j_Page_130.tif
b55e7eb9011b5a39be3583f76edc6585
25ed80dbe6dab0e0d0933c70cd21e0e17f87f13e
F20101212_AAATKW kneifel_j_Page_115.tif
60195fcdef4da494fc1b2c82d6c627a2
e534757803729a29e5cf02df1f4bbf48f7d1db8d
421 F20101212_AAASHU kneifel_j_Page_001.txt
491f1f9498e9e8390b843c99ac087ca3
b488b190190df5d611462ff19945968d75aaa87b
6254 F20101212_AAASII kneifel_j_Page_158thm.jpg
f628d259bc17a0700e149d72cdbe510e
cf2d06b891bfc6af6f77ffa7c85e599d139967d0
F20101212_AAATLL kneifel_j_Page_131.tif
7d0883ef1022df6ae46d339b6a8a1db2
98c7bcd418040d8dc5908ee0c5dba83d89461361
F20101212_AAATKX kneifel_j_Page_116.tif
be86d1b5552bb75d76d0f6e97a2ad0bf
c7504a33e89d3acb806647d7564b06c84370757a
3151 F20101212_AAASHV kneifel_j_Page_198thm.jpg
f324ccef924898a41cfd5ff2dc15c798
a6f57c62bd2ad0828624c3c4f4cf9e62339074eb
2587 F20101212_AAASIJ kneifel_j_Page_059.txt
f83eadf4ca90d954f3ebd2fbcbee475d
c0a3c56464a64fe18d433b7850cfde6139238b3e
F20101212_AAATMA kneifel_j_Page_150.tif
648c99c5a6e1fb5a0562cd95b4f9b1a9
39ccde093eb4b217a83b34e4ab02158989d867a6
F20101212_AAATLM kneifel_j_Page_132.tif
a16d06b58d4b4330f59dd621b87c0acd
c567d94554351863550b1e84ec9c4cf4a306c648
F20101212_AAATKY kneifel_j_Page_117.tif
02171af0e6e06ae7d60d7d3d39a1d106
2369066868efe97ee41be096a46bd215306f0d0b
846286 F20101212_AAASHW kneifel_j_Page_094.jp2
7b5969b9f65f2c56c5ae74f7a127c0ea
9e00441e99312f3a9d53ceb16e056e559843895e
58175 F20101212_AAASIK kneifel_j_Page_055.pro
3353293f130e23432a8942a4a6b322af
177abfb40a7378366d34cc0018122f9fe2583435
F20101212_AAATMB kneifel_j_Page_151.tif
c90544f6e2a4c3ba22c0b159a9cb5d86
6f4248f895ef9e44d239347da5093b92523fed14
F20101212_AAATLN kneifel_j_Page_133.tif
b64e985a800525fcb748e7035d8e3691
79f3e6e8650f5aecc094f43bbc63bb775bc34741
F20101212_AAATKZ kneifel_j_Page_118.tif
0850d412104965049976a829b5383581
ca592853cd417ebf6dd5770a7061aed360d6eb4f
25404 F20101212_AAASIL kneifel_j_Page_190.pro
c7b806ca29c842ae1c7985ae73baad3f
1b5eba6b9e423e01d209bf3462bacf2160d7b0aa
F20101212_AAASHX kneifel_j_Page_091.txt
dcac6270ac464827281cef7dcef700cb
3b49cf7394af44e65347cefd4590835a01bae2dc
F20101212_AAATMC kneifel_j_Page_152.tif
50629ef9e8bf31f0624243a130308701
946332f90a250a9f57e3316edcd7ae6aa091190c
F20101212_AAATLO kneifel_j_Page_135.tif
37aaa995a51a038e416c3e9b648134b5
4123deb94efa5f81060b925b890cd28f5d8b85c2
14929 F20101212_AAASJA kneifel_j_Page_118.pro
71a484ca0adb9e10626905c4c9ab6d5d
9f41b3c7320940caadd3dec06f8306cef6bddf76
6952 F20101212_AAASIM kneifel_j_Page_034thm.jpg
08223f839df586ab3a1e56d7d553edf5
8472730e4a1e93baa7cdd5392792244a29d43a86
85734 F20101212_AAASHY kneifel_j_Page_015.jpg
2a35e12eb0871c330c1bc4ca13e39fe7
a1a39ae0a156de0a6fdc8ba14f8f60d46cb0ca2d
F20101212_AAATMD kneifel_j_Page_153.tif
d464a250429389be576863b5d9abc32d
0520c4907e0b112c70fe45d2600264d6d3cb268a
F20101212_AAATLP kneifel_j_Page_137.tif
59d4bd2375712f84bdc4bd867945898d
1288417cfc6c57394d323abcb223130cae45ade8
F20101212_AAASJB kneifel_j_Page_019.tif
62587e22f13c5803de20cbd0b6ed51b9
e3022f7f688ea6b7d27890adc445a557255ae69c
F20101212_AAASIN kneifel_j_Page_099.tif
cad50926cebf37567c6c211ef5f90eb7
a9109b73fa29fafa0324e4a4bcddf71f763424b1
F20101212_AAASHZ kneifel_j_Page_031.txt
f95c382db98709a86494ca5ff2b12a44
f70b1bb98affe34e624af60f706d707304285b54
F20101212_AAATME kneifel_j_Page_154.tif
e373d1867788d482851a87e3dbd95661
c089bd0644761c96358350026e55cff9ad159205
F20101212_AAATLQ kneifel_j_Page_138.tif
cb82cccdd8e542c158ffd55f7d13fdf4
6bf71dc40f742d388223751854e9b117270a56fa
26320 F20101212_AAASJC kneifel_j_Page_216.pro
abb679375209f09b0b211c7cee31ffa0
a84dd8e631a19b6f6b3c98c8188b4d8e741f75f9
132588 F20101212_AAASIO kneifel_j_Page_212.jp2
8a190b1d71a27c4d59d84683043b9d86
8e501c0b5519ba48c8a99371c0cab89bd12f057c
F20101212_AAATMF kneifel_j_Page_155.tif
0f3af822befec63147fd9708298a9c50
6f1f676515ae744cc41237c4034a959be21f4806
F20101212_AAATLR kneifel_j_Page_140.tif
16b7cbfbcda7eacd3f484c131d8f4082
525bf46588f578de640ce86ea90a96893155db10
5381 F20101212_AAASJD kneifel_j_Page_142thm.jpg
cc6548b549a25dd1b57760ae93acd659
316b4bdc795e7ef5f30544bca83c34911fa15acd
22270 F20101212_AAASIP kneifel_j_Page_001.jp2
b7ff0bc0c5c1755800b73b6fd6d6e71b
3d7bb89f2662899135fc19d65cb1717c4058266e
F20101212_AAATMG kneifel_j_Page_157.tif
01d379bb04b53af6cb5f2c70ded528b2
3971733a574f41303652372ab14b3eb958f2776f
F20101212_AAATLS kneifel_j_Page_141.tif
592f3955094add4edd2184eddbeb5058
201760f7130619bf4121bb7a61cef8bcaff7caf1
46674 F20101212_AAASJE kneifel_j_Page_012.pro
337d53150dad72b8a460ef6221e33b81
f086c99aed5ecfd7047d71b7eb5d062e9cf61557
21948 F20101212_AAASIQ kneifel_j_Page_147.QC.jpg
3cb32195366a44c67c3211359d072556
b720dd9d6d7262f092e567622a27b5eefc10f05a
F20101212_AAATMH kneifel_j_Page_158.tif
ed751d11af451fd0336c4a71b685750e
6f95e2096d99dacad6f5a27cfe913d398a94db7c
F20101212_AAATLT kneifel_j_Page_142.tif
d29051204f44173278b6464c92ba0d35
e2109a861079349576d4ce27fa6d1913a5f78c84
6908 F20101212_AAASJF kneifel_j_Page_134thm.jpg
2f27f5c92e2ca7b6e15edcf19779b7ea
3df54559ed35cf1b594a3f623a8c96d030e069a1
21993 F20101212_AAASIR kneifel_j_Page_151.QC.jpg
6507a02f90b9cac909f03339b0fae96f
e03f38638079f8fbce634da64800c621cb3d6429
F20101212_AAATMI kneifel_j_Page_159.tif
21429d60f6d683d84614daf01ce7671f
dc820a066550f93b89c64b9289d2cbf49ee485b2
F20101212_AAATLU kneifel_j_Page_143.tif
9225dd2473a58612a206cda5f1c5ea5d
3c569a9f17f810ca877db5f36c19d76d9cc58133
59365 F20101212_AAASJG kneifel_j_Page_099.pro
89b59aca863459fde9d87839648f474b
5166ce2cac29ddbbc8c1e6adc917208ad2c263af
25174 F20101212_AAASIS kneifel_j_Page_180.QC.jpg
59448f52d648f3680b188d06d7ed5363
d6bd7c9b8d5e9667e47c0a2b72326f2cff79cdcf
F20101212_AAATMJ kneifel_j_Page_160.tif
e4d8560f928b8167a3e1751d805cd64d
c06cddeda863305ba72fdf368624ebf864a886c8
F20101212_AAATLV kneifel_j_Page_144.tif
ef6cd7a4dcedc5dae5b52de274d614de
59de842f02ac7cd285ff79879871026594f10cc6
131791 F20101212_AAASJH kneifel_j_Page_108.jp2
a6b424809df4a0aa07ab329209ad0889
fdf3d781d27f085c50775faf2154864e930b4934
2104 F20101212_AAASIT kneifel_j_Page_065.txt
0c61eefcea02a5cc9c703158b54400a6
d1ea611cbba2e076015dbea7ee43285675e3f578
F20101212_AAATMK kneifel_j_Page_161.tif
2cf29639bb1993aee0fe8724f2854f9a
a80fe425d0f20fe5e91026f93c41744b5d5a5564
F20101212_AAATLW kneifel_j_Page_146.tif
adfb9f9b3c3d0aa96a85c87b6cadf198
1765fdbee1886cfc523a58af8fe81c0c442f0b76
F20101212_AAASJI kneifel_j_Page_187.tif
d2fac4baa0839dc3851850979baef509
4d0d6c108cce6fadbc8fad73003ed5c901c897c4
F20101212_AAASIU kneifel_j_Page_139.tif
4152ec0befd1866645007a8f1e98a61f
675398e993bcea3db4a068142e8ad8faa5f50027
F20101212_AAATML kneifel_j_Page_163.tif
b71405081ded1fd4ce6f7b48f90a25a0
fa5b8ff0bdc584d22b3f4f4940cd8364859e70ee
F20101212_AAATLX kneifel_j_Page_147.tif
38930921630c5fbf1881e4c95228e067
ced50ecad112ad2dacc256326adbf7ebab49cd3f
1004229 F20101212_AAASJJ kneifel_j_Page_165.jp2
079fc31ee1ccc89c804239a327cfdb04
7bea45d916731cf5c38663d22b7c4eba27a0606e
F20101212_AAASIV kneifel_j_Page_174.tif
3338c62944a46eb791d9d27a2e720ded
ce7dfcf0d86281d16d2822c396d4157e1c02dd55
F20101212_AAATNA kneifel_j_Page_184.tif
f4fe88ad611b2ea33d0eb4274f0a5081
cadcacdb276771f27d577a3e2f12c2abc26ac2a1
F20101212_AAATMM kneifel_j_Page_166.tif
30df664934b8ba8fa8cc911d3a9fe216
e8c720d497f6566ecc8877ad964282bf814043bd
F20101212_AAATLY kneifel_j_Page_148.tif
b5a722ac7539a31efe2931843d08e9e4
ada7a06c5e7f812cd8631c2514546b7522a833e9
62457 F20101212_AAASJK kneifel_j_Page_073.jpg
770912a08134564932adfe116ec726f6
2c2aa01980e501467d561f1cae727b6f3eea79f8
6886 F20101212_AAASIW kneifel_j_Page_038thm.jpg
249505320cc06591e96044c2380c9158
413ab14d5b7359b1facc69bbd8d44a3f6205de85
F20101212_AAATNB kneifel_j_Page_186.tif
57b8e5617cd2005acf7765c086e18677
c825f4fb0967837693287b5b010dc9f5e4b75c33
F20101212_AAATMN kneifel_j_Page_167.tif
0ad671f749d6a58c1bb28c13e7bf1a20
2f9253b1e334c0d5847bb0dcb3fb407eb35c4b2e
F20101212_AAATLZ kneifel_j_Page_149.tif
668732848013739537184cc56e8de774
ddcf6db31aa8e82e963f66c4c9a974ee32c2a452
23701 F20101212_AAASJL kneifel_j_Page_061.QC.jpg
b9200c7cf84d0da436b799f0470adc7a
0c6252b2914b6ef51bdef8b0aa34a20149c2093b
2579 F20101212_AAASIX kneifel_j_Page_213.txt
ae4ba48a4427a8cd480564489cef06ca
fd8dc281db790a0597be4400a5afd5b6ced97488
F20101212_AAATNC kneifel_j_Page_188.tif
1edc04f652c63836618cab98f1280bf6
e9c16e007e4df691ed52b4d772f6408b7973d001
F20101212_AAATMO kneifel_j_Page_168.tif
9d49b9016ae4ab309db3db18cef97f68
71a16ddd23452992ce7b7584a29388669af670a8
23302 F20101212_AAASJM kneifel_j_Page_144.QC.jpg
385121598f2db064fb9116f40132d423
be3d716dd45ee3c6b666ca9e72ea86aaaeba34ad
5870 F20101212_AAASIY kneifel_j_Page_206thm.jpg
bcf95ca3bee8e184649c49c33419b00c
93ae4c03f05fc87b3b30b474a74276b103537998
F20101212_AAASKA kneifel_j_Page_120.tif
5a6f93a38450fcc0edeea22136f47880
3016c714ab46e326d7724ee519bac8e73662e5a0
F20101212_AAATND kneifel_j_Page_190.tif
2f3ca6c354a8ee4091dbc2b60190f28c
55b3136391bcb7a66a8fa57c2ce020fd6988e22c
F20101212_AAATMP kneifel_j_Page_169.tif
adfbecdc8900dac8c3c388e75a14af17
8a3b71f18f145052dd1ad93c55f647b3e3eeaa77
6092 F20101212_AAASJN kneifel_j_Page_131thm.jpg
9d1b4b0d65be380fb96a1d6b51d96767
d1f9ddf749026e3678b509dbda2b53662e5aa025
6379 F20101212_AAASIZ kneifel_j_Page_154thm.jpg
5fc51ffd327516712c8ac8797a4c210c
1bae2dd1dff795cdbd8ef1bd17d6395ab5ee8aa3
5637 F20101212_AAASKB kneifel_j_Page_145thm.jpg
bd1105cd00440dcb9b17ca96d6c7fa97
c9ff94b47781b64cf3f2d40d1ba3942c6e377acb
F20101212_AAATNE kneifel_j_Page_191.tif
039fd78544e1fca4045a617f2e3a9404
68be6a56391cdccaeaf79fc254f0af83762e571f
F20101212_AAATMQ kneifel_j_Page_171.tif
606f1a282915bef34f8d5b238b87b855
f47b419c3bb2ad05a79009b8cc059d50e55926b2
442 F20101212_AAASJO kneifel_j_Page_124.txt
b322d7141c78e7ef376632a08ec9e350
ff2746252794c78ae52d8b6a83f095a5efbbf0c9
20433 F20101212_AAASKC kneifel_j_Page_121.pro
28c52512a61f52cb616f0812a72569db
bd09cd5c8a70a75564bdd6ae56c2dce30330a72e
F20101212_AAATNF kneifel_j_Page_193.tif
ce00911bc310f480de7be0c4f701183c
f1ea2313357723e3b1e630ae5af75afbc4ce7483
F20101212_AAATMR kneifel_j_Page_172.tif
fd4fa1556162916e7a5693ef8bd9d318
ef77c1cb1ee3e98481d424a02aac6380b973d20d
F20101212_AAASJP kneifel_j_Page_087.tif
45c9d1f442e8abe518eb241429067f7c
561ec5ead6aff20bc2b4efb6ea382102529ac4fe
F20101212_AAASKD kneifel_j_Page_093.jp2
fc1126c9a37ea7d302cf59b173fa0e25
40237c33e9dd9e69da6d2a633b23c54df853386d
F20101212_AAATNG kneifel_j_Page_196.tif
988e69eea4adb9230e771b284e32f12f
c525ff45dcb302b60b957e594a6ec3adabf5e37e
F20101212_AAATMS kneifel_j_Page_173.tif
7be09b58b08efc34bc6e7849b6346f4b
cadc0023386342b581aa73ea1c7d9cb3e35620cd
71664 F20101212_AAASJQ kneifel_j_Page_137.jpg
767e599c9606bd0da3a1fef1a218380f
6ad41714ce03148538b058826e1b7898ce8fe470
113871 F20101212_AAASKE kneifel_j_Page_137.jp2
5a12fb417cb19983526aba1785b7048d
bdd268595612576e54aa1a7a1acebf9ad26c9968
F20101212_AAATNH kneifel_j_Page_198.tif
0d7997496a904acf77f2e8d0314ae324
d97bc137483a87f7ca587ea9b3d6316ccfd54f26
F20101212_AAATMT kneifel_j_Page_175.tif
ebbeb26d061737af1e038c30825fb5b0
1317c00464e6c68a624f723f44c6282525e03576
114047 F20101212_AAASJR kneifel_j_Page_131.jp2
71590a056ff0a50b3d507ad94ce85dc5
3ee3b7446a068047c70ee37a4673b470b61e7e25
43637 F20101212_AAASKF kneifel_j_Page_069.pro
a161b669b6a1a121ae38095863fccca1
2cc81ca72c632799f2e7c8923980f793ec34ab10
F20101212_AAATNI kneifel_j_Page_199.tif
710c3fbbc95c3dc7eeb6850f49a57440
8509df505c1af7a6ffacae73f69d2abe07fe0056
F20101212_AAATMU kneifel_j_Page_176.tif
9084f675543d303209f4f4bed1e4ff02
f28b24aa998767dbf92cd19edb867aa38aac3afe
6443 F20101212_AAASJS kneifel_j_Page_054thm.jpg
87c161beeacfcb4722a5ab1d07b5e533
ffd34dfbc6e9ea054a70c99fedf4adcd66edcef9
F20101212_AAASKG kneifel_j_Page_185.tif
62ee84e7ffca90df4b605f3b1ff77925
00d6e4c072e7cac068bead142b0985980831f393
F20101212_AAATNJ kneifel_j_Page_200.tif
e194c3b1ec0793dbbea2131f48e200f8
c2160037d718045b3713868f086b77796160c1ef
F20101212_AAATMV kneifel_j_Page_177.tif
f91e578f139967d890dd28622cb47673
6a19e00f488b11d81d23579c12fb849a98472a07
F20101212_AAASJT kneifel_j_Page_109.jp2
10dc6724b0ad0c755411f82d136f298e
457e99ef5748c58b824967824fda274e274866dd
7106 F20101212_AAASKH kneifel_j_Page_025thm.jpg
44471d53a4d083688fc2b1039efbfe82
02b510f3e610217bbd36e68e3131bde33ed059cf
F20101212_AAATNK kneifel_j_Page_201.tif
f795bee568b359d16d24d10719a690c9
09c3efe17952bf88515caa3540c930f9d44bc0df
F20101212_AAATMW kneifel_j_Page_178.tif
e48d8d6ba41647b2da384eb34a5a74d1
9232aedb924b654b04b46356af47c74409c98ed7
F20101212_AAASJU kneifel_j_Page_192.tif
dc5d52e1c28c8c7b061615aff4a49261
bf8a47409ebac40da755b99d049dfab3df02aa00
6311 F20101212_AAASKI kneifel_j_Page_150thm.jpg
48e443b8cf95cb0cfa87a0320e2e482c
00a21a89a66c924b1756aa1d3e919df11913f552
59103 F20101212_AAATOA kneifel_j_Page_010.pro
e0819f85f0e45bbc198337360dd86eef
a0a02824b8623c570345e3701f1f05dfe09fbf36
F20101212_AAATNL kneifel_j_Page_202.tif
8056099310502f661fb0cf1460208e7d
2e2f5fd05984d0c267b7656263ec040ae4469460
F20101212_AAATMX kneifel_j_Page_179.tif
09e533bae800435f0b6d51b7a9d54298
cb1ea5021978a3617ebc7cec9eb8516ced564664
52466 F20101212_AAASJV kneifel_j_Page_136.pro
fff9ad973d768786670e93445870a3a3
767811afbccb62c0151d63a36b805dc7663bb398
6959 F20101212_AAASKJ kneifel_j_Page_027thm.jpg
82ee5d15d85f066a3c6b960340788f46
72ab7fc8adc2aaef22490ce6b82e54001256940a
F20101212_AAATNM kneifel_j_Page_203.tif
095bcffc81148a1596f55dadedf8f375
c90ad6f7078da55f205c3743093d1aef309e71b9
F20101212_AAATMY kneifel_j_Page_180.tif
7320121c746f57bbe2d5f695a0736414
3b60b0c73ed48aa94c91be539813761671caecf6
F20101212_AAASJW kneifel_j_Page_098.tif
d304c173e8376b13fa6f8abb30a5fee9
fb16d303e23cf16ee146e7ac6c044c65ba2133bb
60408 F20101212_AAASKK kneifel_j_Page_184.jpg
3936314691fb433febca20b132272b20
633bc20df1e1b78bd3101e55a75bf5dc3368bb94
16859 F20101212_AAATOB kneifel_j_Page_011.pro
be206195e15754cad7f2a0d166aec702
5e96b5c3c1630e1a069c306fc17d80c11038e01a
F20101212_AAATNN kneifel_j_Page_204.tif
a0d8c00a1a02c065acae496fed18ede9
35901146c7e4722d45860d85c340e34bcf1e0737
F20101212_AAATMZ kneifel_j_Page_183.tif
eff16589ec31b8deaa789c3ca511f615
c96c7252683534deafeaa367107a257708fb0f26
F20101212_AAASJX kneifel_j_Page_149.jp2
5ed231ee4b669ec276e8a942ee5f9b75
64f549aaf9a40fb4540e8bae3cfa552816f1e284
17079 F20101212_AAASKL kneifel_j_Page_143.QC.jpg
65f8580b93ee26f7a1b1946e0c2e38ec
2e24a236b6f8b173c1875594beb37b94115e0c4c
17917 F20101212_AAATOC kneifel_j_Page_013.pro
7790226bcae107d48260140e90c76f10
bde6b217e8779484e62777b96643d70449d0dd07
F20101212_AAATNO kneifel_j_Page_205.tif
f85d4439aeaa1985d6baeb814a1c5108
98fa0eecfa0db0e7bc83baf6a1a240fb954c9acf
3175 F20101212_AAASJY kneifel_j_Page_117thm.jpg
2a441511d477139a5769fe100c61bb82
954e1c620cb1ad0c1848236f5dc5bcbcf60228c6
F20101212_AAASLA kneifel_j_Page_165.tif
32fe39421b3f8f0da8d5b9cab2ecea3f
bcd4802e82a98dcd36700ffb3478bd9110d99b51
2016 F20101212_AAASKM kneifel_j_Page_012.txt
970700a0654b2e027bf123feefbc5769
7becae39c9f42a858f249efa6056beb2426a4055
55033 F20101212_AAATOD kneifel_j_Page_014.pro
3b2e6344529da46d1571c07908a182a5
a17b32e5be60070a964c15458923e74cc5c1ea33
F20101212_AAATNP kneifel_j_Page_208.tif
78334f897c2efceba71d6c80365c5253
8faffcfb125f3f8782461827dcba3bcc095f17b9
23175 F20101212_AAASJZ kneifel_j_Page_010.QC.jpg
d83fee274fd74c86ccfb454bf5125978
3896ef85668418f800267b7a2fde3208ca4a09d1
1860 F20101212_AAASLB kneifel_j_Page_153.txt
f18a085afc658bce0fb1f4d25c0ecf39
023b66e9dc2458e4f47fe2fadcd3f9980cb7d30d
F20101212_AAASKN kneifel_j_Page_060.tif
68a1b0e200c5a8cedd796c45372f0786
65bc4317ff087af102b1dce42167f7c72fe7e0cc
55420 F20101212_AAATOE kneifel_j_Page_015.pro
a1085aa3faf16f5e81db3b96af3a0cbe
4303ecca1d31712e17c1ef9a2b52959d155b0ea7
F20101212_AAATNQ kneifel_j_Page_211.tif
39e75857703b12f1aa93d1b2e2c507d5
eba9f39578d70887071ffc0b41346e1a91557b77
1961 F20101212_AAASLC kneifel_j_Page_105.txt
4391fffb8046dabb57a657ee1c7bef6e
5c06cf514dbfa6299932b8caeec1dff1ef98251a
420982 F20101212_AAASKO kneifel_j_Page_210.jp2
b0d48a5000412b0fea61f54e18b11f3e
ff1eaa1d37c244fb0bd8624751bed3ca83a33cd7
53932 F20101212_AAATOF kneifel_j_Page_016.pro
adbf05c0f40f2e69dc9966b995b09795
5f175f62e4dc8bb8395b29bbd05837b595feaf5a
F20101212_AAATNR kneifel_j_Page_212.tif
8d96c6d0739f2bdbb3b6188b59ccf77d
211da44dae179f3f899156a1cd563df75e174963
18222 F20101212_AAASLD kneifel_j_Page_101.QC.jpg
baf66507ce1c240ca30f1a7fb4052891
46b36eb7da8c24f92848e3e61982dfff5860af8b
48807 F20101212_AAASKP kneifel_j_Page_203.pro
10a59da7ae923b7c5c4b21fedc72db32
876c1ed757d57c262b0c58930be69627d02d0b70
55531 F20101212_AAATOG kneifel_j_Page_018.pro
3900bd1e87e5769e9ad85bfda37744a3
c36be1d189a63957c15e5734f6e0acc53b3bf9c5
F20101212_AAATNS kneifel_j_Page_213.tif
e5cf5b522a7e0768cf6eb15ab0d7f376
48f6e34d69fafd8b6df531fce5f30a2a796c09b7
F20101212_AAASLE kneifel_j_Page_047.jp2
4f79a4b162e69cdca7ef11915e3cb692
bf4acf4279dadd8b6d681ad26846d8a88c20bf61
2188 F20101212_AAASKQ kneifel_j_Page_069.txt
df9ec741a2941a6d4f91963a516e74ee
3a2d3b3aabe0edad6ed5eb68e934b741eccfa94f
58877 F20101212_AAATOH kneifel_j_Page_019.pro
1e54db40b463d056e724e29dc29d7a37
b75ac749227e3b95fdb881ccafab91d0e0d7c1ac
F20101212_AAATNT kneifel_j_Page_214.tif
4446b08410c6e0a25a7e0cff3ebfbd3f
0854d4e43f438c81e75fe844fb54dced67d4bbce
2395 F20101212_AAASLF kneifel_j_Page_061.txt
1fa658936270bc97bb06f25fb55d441e
0a9a40c1bbc8c52452201dd476692ac23c547be4
4931 F20101212_AAASKR kneifel_j_Page_201thm.jpg
34e315e69118c283d1e046b8b258226e
b649c06e8dba42011f2a105b35b28b85c91851f5
70769 F20101212_AAATOI kneifel_j_Page_023.pro
dad80590e469cf76ef050fb9c769e14f
0ff97a6b5ae67970ca0076cdb10b8f3acfbd7047
F20101212_AAATNU kneifel_j_Page_215.tif
0dd5c724a005d098b6e1472166322ff1
4866d8f822d1ebdeea8723ad9f5fdc564fd5d77b
1051974 F20101212_AAASLG kneifel_j_Page_160.jp2
23e1371de8a125e3461bb7cb3b123ff8
338cdc30114269d1b17087e8bbe1826f2be93362
2068 F20101212_AAASKS kneifel_j_Page_206.txt
fcc17b90b2cfa421af6091fb674a6986
2c7146774dc18ede5df09ad3dd021679e4eac501
63336 F20101212_AAATOJ kneifel_j_Page_024.pro
9181fd4df26b2b6e781ccf676a8b0dc4
018604c0e3e9732fa3eeffe575b5a9a19171ebd0
7557 F20101212_AAATNV kneifel_j_Page_001.pro
1e0800307270dda541b9f711ba2d7082
e7b96d2af3d35d6930e8689ec4485473a59a50b8
1109 F20101212_AAASLH kneifel_j_Page_190.txt
7a2f3df06ed1796e7214167d15506d99
ceaa59685c5e6702bb91154f77781d0a920d07fe
F20101212_AAASKT kneifel_j_Page_206.tif
58faf882e1cce7a67dc294fc2a7c399e
c771174e72bb2db798e6dfc72601c18d9bba34ee
75576 F20101212_AAATOK kneifel_j_Page_025.pro
3cbbd0c6ed6ec3e314b18767e33f8390
809ebe4ba47f29e42b27424e94d9160d0fb96705
2289 F20101212_AAATNW kneifel_j_Page_003.pro
ed8f184c5825e5169794e97fb18a8e97
2e82fbd8fe0b3da493ddbfe785af7484d02b6d1d
79661 F20101212_AAASLI kneifel_j_Page_065.jpg
6d4776071a92a9bebe31f39b709dc00d
597ff52be2cd0a6252453c8e938ec8a2ead20b42
73994 F20101212_AAASKU kneifel_j_Page_041.jpg
fbb436c60cd4920f2e89ffe304a3c533
ba3a792ad7c8a5154c9bde1d99b484f689422b7b
53668 F20101212_AAATPA kneifel_j_Page_051.pro
f99568938526d49f2c23b057a0180e8f
3ae9bbea96e2b796a36a9bf0426a5d0f976c7b49
56138 F20101212_AAATOL kneifel_j_Page_028.pro
8ff6f428717160b933dbad57f545ea90
9c812fcef45541b50d0ca4da70a3d8d0441c4afc
67341 F20101212_AAATNX kneifel_j_Page_005.pro
301b25aa1eb4e604bb234d23186e97b3
8cd2c7d741ae46973669e8e7155d872017eb6de0
F20101212_AAASLJ kneifel_j_Page_210.tif
d13bad7b2d0c817d4ee3ae53aee4df0f
46fe7497270e15f7b74d39b9bcb37cc05aaef896
374 F20101212_AAASKV kneifel_j_Page_210.txt
99d07a31e782644203c3fda012e7b1da
d0cfb3c81e1352c2dec839cef23c4417614871c3
54734 F20101212_AAATPB kneifel_j_Page_052.pro
a2fd26ab28096d4aeabc4ba47370e066
0bb8afe49cc1e58c13834f0fb0bb253eb661f765
56797 F20101212_AAATOM kneifel_j_Page_029.pro
95dc8ff7173b67f6368687ccc50a14a2
776b6449d4152870f481dcdf096e095c20940870
79293 F20101212_AAATNY kneifel_j_Page_006.pro
026de22a167091ee2871cd18a3681aef
6fb41870e9d913de646b4d4dd859c09b8f74502d
2123 F20101212_AAASLK kneifel_j_Page_137.txt
afcb5b64b01de9c10c32606879b56ce7
013303cf15c985ffe949bcf22237bbaf1a81e142
29600 F20101212_AAASKW kneifel_j_Page_013.jpg
7dad9c8c219f16ddee8144382ec2e12f
e94330c89cd71e2d730205af6e80aa946dd374c5
64433 F20101212_AAATON kneifel_j_Page_030.pro
96f641b279511789ec6602c259973b3b
0c17161051d37b671ca3fb98d7e2419507e267e8
56540 F20101212_AAATNZ kneifel_j_Page_008.pro
34aabf5e4afd1b9c00d038ebb6dbf85b
8e5218a74dfb2f91abb8ca481292e81466639625
2827 F20101212_AAASLL kneifel_j_Page_197thm.jpg
1e7b3ff4fc07789b8c4ad2ca7db5f897
1ccf20b58bd08ef5b64ba1fe0d2cb297a92872d6
4432 F20101212_AAASKX kneifel_j_Page_115.pro
9e27c4d4a04dd6e5debeb772a9ea0f06
0c261acd080bc248c5b06ebf02de28021be87d1f
60452 F20101212_AAATPC kneifel_j_Page_053.pro
be83fbbcdf236658aa7471407c6fc999
b6b63d4cfc50ec8f875593f224e16e9fd1a1ae4c
59338 F20101212_AAATOO kneifel_j_Page_032.pro
ebfdb45f26bb296a97aef62ace0651da
80d9e266b24ede9128428bf2fdee6a6e1b6c12d9
66462 F20101212_AAASMA kneifel_j_Page_098.jpg
58213eba4bc7f31a9264daba09f386a0
04bc70d07fa0eb66edf3a2e2dce9ce0ca4b62efe
61106 F20101212_AAASLM kneifel_j_Page_211.pro
1e9efdabe3026dc907bef69cb915f591
d8cdb8e87356ec0c0870c203f65469ec2762379d
58004 F20101212_AAASKY kneifel_j_Page_183.jpg
6df6e837400c5489e0d57280ca449b4f
6ecd7a310b83fc24d4d2f4bf012099a56e66577a
56520 F20101212_AAATPD kneifel_j_Page_054.pro
6f8e55f5ba94ecb40012ee9e3c257b23
d837e9cf53aa77513f80dbcf01d730c30a293282
55736 F20101212_AAATOP kneifel_j_Page_033.pro
1a079f147b4cfe8796b804dc7de73e30
a0fb2536927b9b3f1d2687a56fa563f0a26570c8
2169 F20101212_AAASMB kneifel_j_Page_052.txt
6aa67da376c2e9d5361144f0837e4a15
c595bd57a89decc6bf6157857b13dab332d5f713
47160 F20101212_AAASLN kneifel_j_Page_087.pro
9c5c351f46955eea2716a2bb78283e8c
60a56a65e4ddc680eeef3375bffe71d7c5aa8353
F20101212_AAASKZ kneifel_j_Page_049.QC.jpg
c369535affd81df466dcf1aa6b22a229
42718f1c14324ad921907e8a24600673f2fdb4e9
49054 F20101212_AAATPE kneifel_j_Page_056.pro
7bfc643475fce0b1fef4cc23b913ef17
42da6e050bb40648e3aefdbddc6f0856fc4c7d9a
56756 F20101212_AAATOQ kneifel_j_Page_034.pro
4de2da7f4875056afcf98b24a51081ba
a6efd419c96bce8d1f19c7f8146a5b442a3a2ba1
2216 F20101212_AAASMC kneifel_j_Page_179.txt
845e077b27fcb17b3ac664f8a4b1ad24
defc1ea7bdfca5ac7f0ecc49c844faf1a7586112
F20101212_AAASLO kneifel_j_Page_129thm.jpg
3f37a05cd512bc41f5263b9fab22e244
3680a85b3f0c3cfa74b1ec9509a161b6d99f0bce
55483 F20101212_AAATPF kneifel_j_Page_058.pro
fe8da3c53701816c61993574a5de0b68
70fa92be43b3fd1374b6fc8f57cb8de0c7720d59
53850 F20101212_AAATOR kneifel_j_Page_035.pro
15f713528887cd8cf3db21666c4a8e93
9f7d5afbd8a6eeaf5f94d3eb67612c451d977faf
2130 F20101212_AAASMD kneifel_j_Page_035.txt
22892257975521a6a93a0902febf549b
5d5c45d3f1fe410d07fe51dc387c103403ef84f9
20505 F20101212_AAASLP kneifel_j_Page_072.QC.jpg
fcfd7ba18ab767842b08732430f44a76
30e8ebc39ac6fb64c7802105c7ff30f59478badc
66605 F20101212_AAATPG kneifel_j_Page_059.pro
40b8f91f6dbad56af56a31a97154626e
00d896f615eedc7bd884282640d76d66eca048cb
59289 F20101212_AAATOS kneifel_j_Page_038.pro
a641bce43e212d5abb3dc23aa1e01a7c
3a824458777398843153051eb52744ebd74183b7
99435 F20101212_AAASME kneifel_j_Page_012.jp2
0eecb925faf58b0a585fb19a9b96d17b
d8c60142a2960eec858e023e4ad544c30f45b943
67951 F20101212_AAASLQ kneifel_j_Page_192.jpg
6094830a2239e29d8379b0c11c8ddca9
013578f94861baae855e24684dca42d7b6105c9c
58475 F20101212_AAATPH kneifel_j_Page_061.pro
d711d996ebfd798cf775579e8ca734cc
6f561095887ca8c7c4430d5678198b9a535831d6
55183 F20101212_AAATOT kneifel_j_Page_039.pro
3d2f556ee46f259afbf31ff6c8aaad9b
96f0325318e24320b987c44830b9f689282b71f3
F20101212_AAASMF kneifel_j_Page_015.txt
146b17e258f8257377fec0da911a098f
b81d2a14a05570ad4bb1aea5cedbffc5f6589749
F20101212_AAASLR kneifel_j_Page_207.tif
d973b7e30b094709e4be931b38195887
0a862713b1abb93c8150aeb28963375659cba51e
63876 F20101212_AAATPI kneifel_j_Page_062.pro
945a8a67f49e3b5aee33bb73a6db9dfa
d1008db1f1aa4eadb6621188d53903d9cd70a82a
55019 F20101212_AAATOU kneifel_j_Page_042.pro
f1da4fd03f2747457a45a4d42d3baa04
ee6c97c8e810f9e60044542ff54f0fd3a3a6b812
5878 F20101212_AAASMG kneifel_j_Page_088thm.jpg
d149ee71c474a58d7714322f42f4f537
217b1fb5967f93bf99206c2636d03879830d0d76
88671 F20101212_AAASLS kneifel_j_Page_019.jpg
2f9fb6f873ad99bd85b5f861e3c4058a
3acf822b5df9aaef5cb67cf54e41a8db4ab4fead
49697 F20101212_AAATPJ kneifel_j_Page_065.pro
a4793a777379ed20cea8a8fe9ae8568b
90701fafaf556cf74c8751ddf4c05d58351d6afb
16000 F20101212_AAATOV kneifel_j_Page_043.pro
d2858269c533eea33c6d90abe1d13050
1381ea39184fd9316657cfcae70bf2a58bfbed01
F20101212_AAASMH kneifel_j_Page_073.tif
5a3d3b676b5fec10767a198646147c65
4bb05eb0876e656b099143d39951aa14760d515a
8349 F20101212_AAASLT kneifel_j_Page_198.QC.jpg
6d0d4b3c443a1591a182dfe609ba516c
6d0bd167a6990e9043cec3778741b3ff5f9347cb
43913 F20101212_AAATPK kneifel_j_Page_066.pro
f94a3676963b49e6095f98bed3c20876
8a91a79388c4885d5bbccde9bf9620007e7b742b
6829 F20101212_AAATOW kneifel_j_Page_045.pro
704fdf2e3f3337094dc64393339faee4
7004634e117797ff5d85b6d65f0fc1cf2a421e84
F20101212_AAASMI kneifel_j_Page_018.jp2
254693888b5ea315b86e94094f86a95f
61bc856afa31b72b1b9d92854eccda60f2555b20
57347 F20101212_AAASLU kneifel_j_Page_150.pro
9952bca7a7da3614d09eff40763f76da
5235448aa9b9fb1e8d3ccc254f3e8245dfddd2f9
53701 F20101212_AAATQA kneifel_j_Page_086.pro
46ee38d385815f1f9fd0c1961584b3a1
ac7140928cd2d5062eed8a94479b333439eb3793
43587 F20101212_AAATPL kneifel_j_Page_067.pro
2375b4e02c37d420eb7980231d19b095
7a6d3e5b91430cf37b3a5d3818af96c31f83f598
5504 F20101212_AAATOX kneifel_j_Page_046.pro
edb443afd4301393b904665a918557ae
2e808d2e16773bc333dbf6353505bc6912025605
111131 F20101212_AAASMJ kneifel_j_Page_138.jp2
f4f3a98d01d54d27f52a68858d1ee7d9
c7ba49c302276819c8b1269ae32c41ee11c5740e
17600 F20101212_AAASLV kneifel_j_Page_122.QC.jpg
1483fe9a7853441aa34bf23619c24efd
a8b899f3a8e09d9f4797b65cfe008113c33db890
46023 F20101212_AAATQB kneifel_j_Page_088.pro
7ecf96504b2d6fbbeec0d613b3b72546
c067d64ac38d24afca15a348a2ab193e1105eaf9
51428 F20101212_AAATPM kneifel_j_Page_070.pro
f323cb3ffe74635d9119e72a91417ff8
43f7a73cd051ae9a8af72c920df27a96657cf9a8
58109 F20101212_AAATOY kneifel_j_Page_048.pro
af7b248606faba59f854d952c222a19f
11fc5145873a22e09ad4cfdc23136bce9b5da575
21 F20101212_AAASMK processing.instr
8ac76025beac2c4059e6932d41d2f612
6eddee155652ab561b476a7536cb7c21096ff5e9
21825 F20101212_AAASLW kneifel_j_Page_200.QC.jpg
d98176e469c377d5535a136714d2eb9c
73e9e00a6978894d2d2b663f0e9f637a9faf7aa6
51121 F20101212_AAATQC kneifel_j_Page_090.pro
0b196f06d6a0adac13def438cdaf72c9
efcbdd73ee778429f5e9f1a24b23e5ab323f53fd
48341 F20101212_AAATPN kneifel_j_Page_071.pro
a3fab03546a604c13b720b12bf62eaa2
a754668801601175d9aa99bce117eaf150c1bd1a
54449 F20101212_AAATOZ kneifel_j_Page_049.pro
137dffca39473eeb5d1b5fd8df0b260d
dca4aff12c30ebb942ff178a313d264787496b34
F20101212_AAASML kneifel_j_Page_156.tif
2040f2094e346ce9ae589f8d91bd0d2b
ab7a533771aa1145fc39c918ffd203ce83a40a23
79004 F20101212_AAASLX kneifel_j_Page_036.jpg
a194937c0959df3439ee5946c378ecbe
59c6ca9865797ec0eee7b0e8362772e560976046
8796 F20101212_AAASNA kneifel_j_Page_003.jp2
44be89e258317d077912405003cbe7d6
855f201c334b4cabbd26ada5e1cf3699c84e3252
41406 F20101212_AAATPO kneifel_j_Page_072.pro
03a9e3f2c795163ec29ec799bdb7681f
a5954a2df80a7a9f8024bfaa538b5f132184b32b
88834 F20101212_AAASMM kneifel_j_Page_050.jpg
ed863df53920f69c1459ddfa395035be
e553f3a2dd9dde0cc87518928ca8ea8c019affcb
F20101212_AAASLY kneifel_j_Page_052.jp2
76e0cedb72c9b2fd6208ed4d9ead8b0b
fdc755a78f4b3ae4851bb2471c205e90d9dc7ec9
43767 F20101212_AAATQD kneifel_j_Page_092.pro
17b61608f8cc44608b12c1760b721722
4adc661ca0c4439730ef910d64c4aff11c265dac
42754 F20101212_AAATPP kneifel_j_Page_075.pro
45531a4159a5aae8b1174df03a849688
0c3bc4e8164a9654da5d5850d8b51672068ab9df
1014532 F20101212_AAASMN kneifel_j_Page_063.jp2
cd76d9a29d86a7f109b4046466b8da4a
a8b4c25e8571c1d9544736ca9f804221f7a538b3
44037 F20101212_AAASLZ kneifel_j_Page_168.pro
e9adf83409e76a52a2534f9a87387c8e
21a9558158f7c923c3fbd749d0cebc0efa2ade49
917697 F20101212_AAASNB kneifel_j_Page_103.jp2
b5af2bdcfbfea601393af049e63c1b1a
c754e8e51652e9c148a74b392a591d761e1c7f4d







ESSAYS IN RENEWABLE ENERGY AND EMISSIONS TRADING


By

JOSHUA D. KNEIFEL



















A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTORATE OF PHILOSOPHY

UNIVERSITY OF FLORIDA

2008


































2008 Joshua D. Kneifel


































To My Parents, who have supported me through all of my academic achievements.









ACKNOWLEDGMENTS

I thank my supervisory committee chair, Lawrence Kenny, along with my other

committee members (Sanford Berg, Jonathan Hamilton, and Jane Luzar) for their

comments and advise regarding my dissertation. Special thanks goes to Paul Sotkiewicz

for his interest, guidance, and support in my research. I also thank my parents for their

never-ending support.









TABLE OF CONTENTS


page

ACKNOW LEDGMENTS ................................. 4

LIST OF TABLES ....................... ............. 9

LIST OF FIGURES ....................... ........... 10

ABSTRACT . . . . . . . . . . 12

CHAPTER

1 EFFECTS OF STATE GOVERNMENT POLICIES ON ELECTRICITY
CAPACITY FROM NON-HYDROPOWER RENEWABLE SOURCES ..... 14

1.1 Introduction . . . . . . . .. 14
1.2 Literature Review ......... ...................... 16
1.3 M odel . . . . .. . . . 18
1.4 Variables and Data .......... ................... 21
1.4.1 Economic and Political Variables: Wit ....... ......... 21
1.4.2 Regulatory Policy Variables: Rit .................. 27
1.5 Statistical Specifications and Empirical Analysis ....... ....... 33
1.5.1 Economic and Political Variables: Wit ................ 34
1.5.2 Regulatory Policy Variables: Rit ............ .. .37
1.5.3 State Fixed-Effects Variables: Sit ........ . . 40
1.5.4 Year Variables: Tit .................. ........ .. 40
1.6 Conclusions ............... .............. .. 41

2 EFFECTS OF COAL CONTRACT CONSTRAINTS ON SO2 TRADING
PROGRAM COMPLIANCE DECISIONS ............... .. .. .. 47

2.1 Introduction .................. ................ .. 47
2.2 Policy Background ............ . . . ...... 48
2.2.1 Title IV of the Clean Air Act Amendment . . ..... 48
2.2.1.1 Phase I of Title IV ................. . .. 49
2.2.1.2 Phase II of Title IV ................. . .. 50
2.2.2 Clean Air Interstate Rule .................. .... .. 51
2.3 Literature Review .................. ............. .. 52
2.3.1 Title IV: Phase I ............ . . ...... 52
2.3.2 Utility-Level Models of Compliance Costs . . ..... 55
2.3.3 Long-Term Coal Contracts ................ .... .. 56
2.4 Inefficiencies Resulting from Coal Contract Constraints . .... 57
2.5 Model and Parameters ............. . . ...... 60
2.6 Generating Unit Level Decision-Making Process . . ..... 62
2.6.1 Generating Unit's Problem ...... .......... . .. 63
2.6.2 First-Order Conditions ................. .... .. 64









2.6.3 ('!, ii :terizing a Unit's Spot Market Fuel C('!i... and Marginal
Cost of Abatement from Fuel Switching . . . 65
2.6.3.1 Necessary conditions for using both high sulfur and low
sulfur coal ...... ........ ... .. ...... 65
2.6.3.2 Only high sulfur coal use: Necessary conditions . 65
2.6.3.3 Only low sulfur coal use: Necessary conditions ...... ..66
2.6.4 Coal Use Under a High Sulfur Coal Contract Constraint ...... 66
2.6.5 Coal Use under a Low Sulfur Coal Contract Constraint ...... ..68
2.6.6 Generating Unit-Level Compliance Costs . . ..... 70
2.6.7 Generating Unit's Net Allowance Position: Excess Demand
Correspondence .................. .. 72
2.6.7.1 Cost savings of fuel switching versus allowance purchases
when PA > MCA'.. .................. .. 74
2.6.7.2 Effects of high sulfur coal contracts on excess demand and
costs .................. .......... .. 74
2.6.7.3 Cost savings of allowance purchases versus fuel switching
when PA < MCA'.. ............... .. .. 80
2.6.7.4 Effects of low sulfur coal contracts . . ... 80
2.6.7.5 Fuel switching versus allowance purchases when PA
M CAf'" ...... . .. ... ..... 85
2.6.8 Generating Unit's Scrubber Installation C('!i i. . . 86
2.6.8.1 When will a generating unit install a scrubber? . 86
2.6.8.2 Different marginal costs of abatement . . .... 87
2.6.8.3 Excess demand correspondence . . ..... 87
2.6.9 Impact of Coal Contracts on Excess Demand Correspondence . 89
2.6.9.1 Impact of a binding high sulfur coal contract . ... 90
2.6.9.2 Impact of a binding low sulfur coal contract . ... 100
2.7 Possible Implications on the Allowance Market and Industry Compliance
C costs . . . . . . . . .. 109
2.8 Conclusions .................. ................ .. 110

3 THE EFFECT OF FUEL CONTRACTING CONSTRAINTS ON SO2 TRADING
PROGRAM COMPLIANCE: EMPIRICAL EVIDENCE . . 127

3.1 Introduction ............. .. ........ ..... 127
3.2 Review of Generating Unit Model ................ ... 128
3.2.1 Generating Unit Problem .................. ...... 128
3.2.2 Optimal Compliance C('!. ~ .. ............... 129
3.3 Allowance Market Equilibrium ..... . . .... 132
3.4 Comparative Statics: Effects on the Allowance Market . .... 134
3.4.1 Comparative Statics: Effect of Relative Fuel Cost on the Allowance
Market ................... ............... 134
3.4.2 Comparative Statics: Effect of Coal Contracts on the Allowance
Market Given the Scrubber C('! i... .................. 135
3.4.2.1 Impact of high sulfur coal contract on allowance market 135










3.4.2.2 Impact of low sulfur coal contract on allowance market 136
3.4.3 Comparative Statics: Effect of Coal Contracts on the Allowance
Market with Endogenous Scrubber Choice . . ..... 136
3.4.3.1 High sulfur coal contract binds . . .... 137
3.4.3.2 Low sulfur coal contract binds . . .... 138
3.5 Compliance Costs . . . .. . ... 139
3.5.1 Compliance Costs with Coal Contracts Relative to Compliance Costs
from Previous Studies .................. ....... 139
3.5.2 Total Industry Compliance Cost .... . . .. 141
3.5.3 Impact of Allowance Allocation on Compliance Costs Given Scrubber
the Ci('... .............. .............. 144
3.5.4 Impact of Allowance Allocation with Endogenous Scrubber ('C!i.. 146
3.6 Simulation M odel .................. ............. 147
3.6.1 Introduction .................. ............ 147
3.6.2 Data ............ .......... .............. 147


3.6.2.1 Fuel data.... . . .....
3.6.2.2 Allowance, actual emissions, and demand data .
3.6.2.3 Technical generator and scrubber data . .
3.6.3 Simulation Model Design... . .....
3.6.4 Simulation Results . . .
3.6.4.1 Total industry costs and allowance market results .
3.6.4.2 Industry and generating unit coal use . .
3.6.4.3 Generating unit scrubber installation choices .


. 147
. 150
. 151
. 152
. 152
. 153
. 157
. 158


3.6.4.4 Impact of allowance allocation on the allowance market
and compliance costs ............... .. 160
3.6.4.5 Summary of simulation results . . ..... 160
3.7 Plant Level Decision-Making Process .................. ... 161
3.7.1 Introduction .................. ............ .. 161
3.7.2 Plant-Level Problem .................. ..... .. 163
3.7.3 First-Order Conditions . . ..... .... 164
3.7.4 C(!i i :terizing a Unit's Spot Market Fuel C('!i. . ... 165
3.7.4.1 Case 1: Necessary conditions for using both high sulfur
and low sulfur spot market coal . . . 166
3.7.4.2 Case 2: Necessary conditions for only high sulfur spot market
coal use ...... ..... ........ 167
3.7.4.3 Case 3: Necessary conditions for only low sulfur spot market
coal use . . . . . .. .. 167
3.7.5 Excess Demand Correspondence ..... . . .. 168
3.7.6 C('! i :terizing a Generating Unit's Contract Fuel ('Ch!., .. .. 170
3.7.6.1 Case 1: Necessary conditions for high sulfur contract coal
use at Generating Unit i" ...... . . .... 170
3.7.6.2 Case 2: Necessary conditions for low sulfur contract coal
use at generating unit "i" . . . 171
3.7.7 "Non-Affected" Generating Units at an "Affected" Plant ..... 171









3.7.7.1 C'! i ,i.terization of "non-affected" generating units at an
il!, l. 1" plant ...... ... ... ... ..... 172
3.7.7.2 Nun-affected" generating units and high sulfur coal
contracts .......... ...... 173
3.7.8 Scrubber Installation C(!..... .. ............ ... 174
3.7.8.1 Marginal cost of abatement with and without a scrubber 174
3.7.9 A Plant's Preferred Order of Scrubber Installation . .... 176
3.7.9.1 At which generating units will a plant install a scrubber? 177
3.7.9.2 At what allowance price will a plant install a scrubber at
a given generating unit? .................. 178
3.7.9.3 Scrubber installation and high sulfur coal contracts . 179
3.7.9.4 Scrubber installation and low sulfur coal contracts ..... 181
3.7.10 Scrubber Installation Example: Plant with Two Affected Generating
Units ...... ...... ..... ... .......... 182
3.7.10.1 Case 1: Install no scrubbers ................ .. 183
3.7.10.2 Case 2: Install one scrubber . . 184
3.7.10.3 Case 3: Install two scrubbers . . . 185
3.7.11 Scrubber Installation Example: Plant with One Affected and One
Non-Affected Generating Units ............. . 186
3.7.12 Summary of Plant Level Results .............. .. 187
3.8 CONCLUSIONS .................. ............. 188

A CONTRACT IMPACTS ON COSTS AND SCRUBBER INSTALLATION
INDIFFERENCE PRICE .................. . ..... 199

A.1 Impacts on Total Costs and Compliance Costs from a Coal Contract
Constraint .................. ............ .. 199
A.2 Derivation of Cost-Minimizing Input Use to Find P . . .. .. 205

B MARKET EQUILIBRIUM AND SIMULATION DESIGN . . ... 206

B.1 Conditions for Existence of an Equilibrium ..... . . ..... 206
B.2 Technical Details of Simulation Model Design . . ..... 208

REFERENCES .................. ................ .. .. 211

BIOGRAPHICAL SKETCH .................. ............. .. 216










LIST OF TABLES


Tabl

1-1

1-2

1-3

1-4

2-1

2-2

2-3

2-4

2-5

2-6

2-7

2-8

3-1

3-2

3-3

3-4

3-5

3-6

3-7


Dependent and Control Variables . ......

Regressions Results . .............

Policy Variables . ... .. .. .. .. ... .

Variable Effects of Significant Variables . ..

Phase I Compliance Cost Estimates . ....

High Sulfur Coal Contract: Assumptions . .

High Sulfur Coal Contract: Results . ....

Low Sulfur Coal Contract Examples: Assumptions .

Low Sulfur Coal Contract Examples: Results .

Example Epsilon Magnitude: Case 1 . ....

Example Epsilon Magnitude: Case 2 . ....

Example Epsilon Magnitude . ........

Example: Contract Coal Distribution . ...

Sulfur Conversion by Fuel Type . ......

Simulation Results . ..............

Impact of Contract Constraint on Scrubber ('C! ..i .

Simulations with Engineering Data . .....

Impacts of a Reduction in the Allowance Allocation o

Math Example: Two Affected Units . ....


e


page

.. . 43

.. . 44

.. . 45

.. . 46

. .. . 13

.. . 114

.. . 114

. . 114

... . 15

. .. . 16

.. . 117

. .. . 18

.. . 190

.. . 190

.. . 19 1

... . 92

. .. . 9 2

f 10' . . 192

. .. . 9 2










LIST OF FIGURES


Figu

2-1

2-2


page


Savings


re

The SO2 Allowance Price . ..........

Excess Demand Correspondence and Compliance Cost
Over Allowance Purchasing .. ...........

High Sulfur Contract: Shift in Minimum Excess Dema

No Contract: Compliance Costs . ......

High Sulfur Contract: Compliance and Total Costs .

High Sulfur Contract: Relative Savings from Contract

Cost Savings from Using Allowances Over Fuel Switch


2-8 Low Sulfur Contract


No Contract: Compliance Costs . .......

Low Sulfur Coal Contract: MCA8'c . .

Low Sulfur Coal Contract: MCA8'c . .

Compliance Costs: PA = MCA8' . .

High Sulfur Coal Contract . ..........

Low Sulfur Coal Contract . ...........

Excess Demand Correspondence: MCA"'" < P ...

Excess Demand Correspondence: MCA"'" > PA .

Impact of a High Sulfur Coal Contract: MCA"' < PA

Impact of a High Sulfur Coal Contract: MCA"'" > PS

Impact of a Low Sulfur Coal Contract: MCA"'j > P .

Impact of a low sulfur Coal Contract: MCA^s < P .

Excess Demand Correspondence . .......

Impact of High Sulfur Coal Contract . .....

Impact of Low Sulfur Coal Contract . .....

Excess Demand Correspondence with Scrubber ('C!I .,

Impacts of High Sulfur Coal Contract . ....


from Fuel


. 119

Switching


n





(

ii


d . . 120

. .. . 20

.. . 20

Coal . . ... 121

ig . . . 121

. .. . 121

. .. . 22

. .. . 22

. .. . 22

. .. . 23

. .. . 23

. .. . 23

. ... . 24

. ... . 24

. . . 125

. . . 125

. . . 125

. . . 26

. .. . 93

. .. . 93

. .. . 94

. . . 94

. .. . 95


2-9

2-10

2-11

2-12

2-13

2-14

2-15

2-16

2-17

2-18

2-19

2-20

3-1

3-2

3-3

3-4

3-5










3-6

3-7

3-8

3-9

3-10

B-1

B-2


Impacts of Low Sulfur Coal Contract . .........

Given Scrubber C('!. Shift from High Sulfur Contract .

Given Scrubber C('!... Shift from Low Sulfur Contract .

With Scrubber C('!. .. Shift from High Sulfur Contract .

With Scrubber C('!. Shift from Low Sulfur Contract .

Upper Semi-Continuous Correspondence . .......

Correspondence with Scrubber C('!i.... .........


.. 195

. . 96

. . 96

. . 97

. . 98

.. 210

.. 210









Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctorate of Philosophy

ESSAYS IN RENEWABLE ENERGY AND EMISSIONS TRADING

By

Joshua D. Kneifel

August 2008

C'!I i': Lawrence Kenny
Major: Economics

Environmental issues have become a key political issue over the past forty years and

has resulted in the enactment of many different environmental policies. The three e -(- Z

in this dissertation add to the literature of renewable energy policies and sulfur dioxide

emissions trading.

The first e-i c ascertains which state policies are accelerating deployment of

non-hydropower renewable electricity generation capacity into a states electric power

industry. As would be expected, policies that lead to significant increases in actual

renewable capacity in that state either set a Renewables Portfolio Standard with a

certain level of required renewable capacity or use Clean Energy Funds to directly fund

Sliil-i --- 1.1 renewable capacity construction. A surprising result is that Required Green

Power Options, a policy that merely requires all utilities in a state to offer the option

for consumers to purchase renewable energy at a premium rate, has a sizable impact on

non-hydro renewable capacity in that state.

The second e-;i- studies the theoretical impacts fuel contract constraints have on

a electricity generating unit's compliance costs of meeting the emissions compliance

restrictions set by Phase I of the Title IV SO2 Emissions Trading Program. Fuel contract

constraints restrict a utility's degrees of freedom in coal purchasing options, which can

lead to the use of a more expensive compliance option and higher compliance costs.









The third E-iv analytically and empirically shows how fuel contract constraints

impact the emissions allowance market and total electric power industry compliance costs.

This paper uses generating unit-level simulations to replicate results from previous studies

and show that fuel contracts appear to explain a large portion (i.' .) of the previously

unexplained compliance cost simulations. Also, my study considers a more appropriate

plant-level decisions for compliance choices by analytically analyzing the plant level

decision-making process to show how cost-minimization at the more complex plant level

may deviate from cost-minimization at the generating unit level.









CHAPTER 1
EFFECTS OF STATE GOVERNMENT POLICIES ON ELECTRICITY CAPACITY
FROM NON-HYDROPOWER RENEWABLE SOURCES

1.1 Introduction

Renewable energy has recently become an important aspect in the U.S. electricity

generation mix and a primary focus of government policy for environmental and energy

security/price volatility reasons. First, the public's growing concern for the environment

and progressively stringent regulation of emissions in the electric power industry has

driven policies to increase the amount of renewable energy in the electricity generation

portfolio. Electricity production from renewable resources creates little, and often zero,

emissions of the pollutants that result from traditional fossil fuel generating technologies.

More renewable energy use helps utilities in their emissions compliance obligations.

Moreover, the prospect of compliance with any future carbon emissions regulation

would further strengthen the incentive to shift toward cleaner electricity generating

technologies.

Second, recent uncertainty in the U.S. energy supply due to political concerns in the

Middle East countries and other foreign oil producing countries as well as volatility in oil

and natural gas prices have led to a push to increase U.S. energy independence through

a greater domestic energy supply and to decrease the impacts on the economy from any

price shocks in the fossil fuel markets, such as the natural gas price spikes in 2000-2001

and following the 2004 and 2005 hurricane seasons.2



1 Smith et al. (2000) estimates the displacement of emissions from the Massachusetts
Renewables Portfolio Standard.
2 Bird et al. (2005) explains the market factors behind wind power deployment, which
include the volatility of natural gas prices. GDS Associates (2001) supported this factor
as well in the reasoning behind the enactment of Hawaii's Renewables Portfolio Standard.
The delivered price of natural gas to electric utilities has risen from $2.62/million cubic
foot (\!CF) in 1999 to $8.45/ \!CF in 2005 (EIA Annual Energy Review 2005).









Complementing federal policies such as the production tax credit, state governments

have taken actions to increase renewable energy capacity and generation, with 41 of

the 50 states enacting policies to encourage the use of renewable energy in their state.

Individual state policies show a great deal of variance. The objective of this paper is to

determine which state policies have led to increased deployment of .-.-regate non-hydro

renewable energy capacity into a state's electric power industry.3 The literature on state

renewable energy policies consists mainly of case studies on policy effectiveness. Only one

previous paper uses econometric methods to estimate the effects of various state policies

on renewable capacity. Menz and Vachon (2006) measure the impacts on wind capacity

in 39 states for 1998-2002. In contrast, my paper uses panel data from all 50 states for

1996-2003 to estimate the effects on total nonhydro renewable capacity deployment, not

just wind power capacity deployment. It estimates the effects of additional policies, and

also controls for differences in the market and political environments.

Three distinctly different types of policies are found to be effective at expanding

non-hydro renewable capacity deployment: a command-and-control policy known as a

Renewables Portfolio Standard (RPS), a tax-and-subsidy scheme facilitated through a

Public Benefits Fund (PBF) or Clean Energy Fund (CEF), and a market-based policy

where consumers can express their preferences to buy power from renewable resources at a

premium price.

The command-and-control policy targets the utility by mandating a specified level

of capacity that must come from renewable energy, and is generally referred to as a

Renewables Portfolio Standard. The tax-and-subsidy scheme collects an additional charge

per unit of electricity consumed from all customers in a state and places the proceeds into

this Public Benefits Fund or Clean Energy Fund. Monies from the PBF/CEF are used to



3 The electric power industry accounted for ,i 1', of renewable energy production in
2003.









subsidize renewable capacity deployment through grants, loans, or production incentives.

The market-based policy creates a differentiated demand by mandating that utilities

must offer their customers the choice to purchase green power, which allows consumers to

express their preferences through p ,iing an extra, utility commission-approved charge for

green power.

The econometric results support many of the conclusions from various case studies

with respect to Renewables Portfolio Standard and Clean Energy Fund policies. Moreover,

the results presented here also show, unlike previous studies, that the potential for

offering consumers the option to purchase renewable electricity at a higher price than

conventionally produced electricity can increase renewable capacity in a state.

1.2 Literature Review

The bulk of the literature in this area uses case studies to determine the specific

characteristics of effective state renewable energy policies. There are two main types

of case studies: (1) analyses of a specific policy enacted in a particular state; and (2)

a summary of the general impacts of a specific policy mechanism used across multiple

states, including policy design characteristics that are effective across multiple states.

Langniss and Wiser (2003) analyze the Texas Renewables Portfolio Standard, including

the achievements of the policy mechanism and the design characteristics that allowed

the policy to be effective at increasing renewable energy capacity. It was found that the

clearly defined capacity requirements have been effective in increasing renewable capacity

in Texas.

Wiser et al. (2004) considered all Renewables Portfolio Standards and found the

pitfalls in the current policy designs. Some key problems in policy designs include

insufficient duration and stability of targets, weak enforcement, and narrow applicability

of the policy. Other conditions that may impact a policy's effectiveness are the presence of

long-term power purchasers and political and regulatory stability.









Petersik (2004) provides a non-econometric analysis of the effectiveness of different

types of Renewables Portfolio Standards as of 2003 for the United States Energy

Information Association (EIA). He finds that only Renewables Portfolio Standards

that mandate a certain level of capacity (number of megawatts) have had any significant

impact on renewable capacity deployment. Policies with renewable generation or sales

requirements as well as voluntary policy programs were found to have no significant effect.

C'!, i, et al. (2007) compares the results from 28 policy impact projections for state

or utilit,---1. ,l Renewables Portfolio Standards and finds that (1) the impact on electricity

prices is minimal, (2) wind power is expected to be the primary renewable used to

meet policy requirements, and (3) the benefit-cost estimates rely heavily on uncertain

assumptions, such as renewable technology costs, natural gas prices, and possible carbon

emissions policy in the future.

Bolinger et al. (2001) describe in detail 14 different state Clean Energy Funds,

enumerating the regulatory background, funding approaches, the current status of the

fund, and the resulting impacts on renewable energy. Programs that fund utility-scale

projects are found to be the most effective at increasing renewable capacity deployment.4

Bolinger et al. (2004, 2006) summarize the same 14 Clean Energy Funds. They find that

due to d-.1 i- and cancelled projects actual capacity often is much lower than initially

obligated capacity.

Wiser and Olson (2004) examine participation in 66 utility green power programs.

They find local green power programs have residential participation rates ranging from

0.02'. to 6.45'. and averaging 1.t:''. However, this study does not look at any state-level

Required Green Power Options that require all utilities in a state to offer consumers

the option to purchase renewable energy. The paper focuses on participation rates of



4 Funding is usually based on actual production, but it is paid in a lump sum once the
capacity has been constructed.









the utility-based programs, but does not analyze the impact of these local programs on

renewable energy generation or capacity.

Bird et al. (2005) summarize federal renewable energy policies, general market factors,

and state-specific factors, such as state policies, that are driving the deployment of wind

power. The key market factors are the volatility in natural gas prices during the early

2000s and the lowered wind energy generation costs due to larger wind turbines, which

have combined to make wind power more competitive with natural gas-fired generation.

Only one paper has attempted to econometrically estimate the effects of state

renewable energy policy on renewable capacity. Menz and Vachon (2006) use ordinary

least squares to estimate state policy effects on wind power capacity and generation with

a panel dataset for 39 states for 1998-2002 while controlling for wind power availability,

retail choice, and policy dummy variables for Public Benefits Fund, Renewables Portfolio

Standard, Required Green Power Option, and fuel mix disclosure.5 Renewables

Portfolio Standards, which require a minimum amount of renewable energy capacity

or generation, and Required Green Power Options, which require all utilities in a state

to offer renewable-based electricity to all consumers for a premium price, are found

to have a statistically significant effect on wind capacity deployment. No statistically

significant effects were found for Public Benefits Funds, which aid both the funding of

energy efficiency, and for Clean Energy Funds, which fund renewable energy programs and

projects.

1.3 Model

This paper uses an ordinary least squares approach as did Menz and Vachon (2006),

but differs in many aspects. This paper includes state fixed-effects, a larger sample,



5 Fuel mix disclosure is a policy that requires the fuel mix a power producer uses in its
electricity generation to be disclosed to the public. It is believed that consumers will use
this information to purchase electricity from power producers that use cleaner burning
fuels or alternative energy.









and additional and more detailed policy variables as well as control variables for a

state's electricity market and political environment. Without controlling for differences

in market size and political environments, omitted variables may bias the results and

lead to incorrect policy interpretations. State fixed-effects are used to control for

renewable availability and capacity constructed prior to 1996, which is in large part

due to the implementation of prior federal policy at the state level as well as the effects of

environmental preferences not captured by other variables.

Cit = ao + 3 Rit + Wit + Si + it
The model estimates total non-hydropower renewable capacity (Cit) for 1996-2003,

where subscript i is the state and t is the year of the specific observation. Rt is the

vector of seven regulatory policies (Clean Energy Fund, Renewables Portfolio Standard

with Capacity Requirements, Renewables Portfolio Standard with Generation/Sales

Requirements, Net A\ I i :- Interconnection Standards, State Government Green Power

Pui 1 i i- i- and Required Green Power Option) and Wi is the vector of eight political

and economic variables. Vector Si is the state fixed-effects dummy variables and vector Tt

are the year variables. The year variables, most of the control variables, and some of the

policy variables are interacted with each state's electricity generation level to control for

market size in each state.

The dependent variable is the total non-hydropower renewable nameplate capacity

in the electric power industry (Cit), which includes all nameplate capacity of utilities,

independent power producers (IPPs), and industrial or commercial combined heat and

power producers that use solar, wind, geothermal, or biomass as an energy source.6 The

sum of all non-hydropower renewable energy in a state is used instead of the capacity



6 Nameplate capacity is the amount of capacity the generator produces under ideal
conditions. Non-hydro renewable nameplate capacity is derived from EIA Historical State
Electricity Databases found on the EIA website in which solar, biomass, geothermal, and
wind nameplate capacity are combined into a single category labeled Other Renewables.









of one specific type of renewable energy because using only one type would preclude

any interesting cross-state comparison of policy effects of states with different available

renewable energy resources.7 For example, comparing the effects of a policy on Maine

and Texas using only wind power capacity excludes the policy effects on biomass capacity,

which is a more likely renewable choice for Maine. Both types of renewable resources must

be included to directly compare the effectiveness of policies across states.

The effects of state renewable energy policies are best estimated using total state

non-hydro renewable capacity as the dependent variable because several policies mandate

or fund a specific amount of renewable capacity. Policies that do not set specific renewable

capacity requirements can be measured in capacity terms by controlling for each state's

market size, which will be discussed in more detail in Section 4.

A large amount of renewable capacity created before 1996 originated from the Public

Utilities Regulatory Policy Act (PURPA), a federal policy passed in 1978 requiring

utilities to purchase electricity from Qualifying Facilities (QFs), which are IPPs that meet

specific requirements and include renewable-based facilities. For a variety of reasons, the

effects of PURPA varied from state to state. State dummy variables (Si) measure these

effects and other unchanging state factors, such as renewable resource availability.8



7 Hydropower is not included in the renewable energy capacity because most
hydropower was created well before the mid-1990s, with few changes in capacity or costs
over the time period being analyzed. These aspects allow hydropower to be considered
a type of current generating technology, which includes steam or gas turbines fired by
natural gas, coal, petroleum, or nuclear power. For hydropower to be a viable power
option there must be an available river or stream as well as a significant change in
elevation. Most of these sites in the U.S. already have hydropower capacity in place.
Removing hydropower from the dependent variable allows the focus of the paper to be
on the policy effects on the emerging technologies of wind, solar, biomass, and geothermal
power

8 (\ !, i, 2003). There is some concern that expiration and buyouts of PURPA
contracts during the 1990s have led to decreases in renewable capacity, especially in
California where deregulation in the early to mid-1990s created competition based on price









1.4 Variables and Data

1.4.1 Economic and Political Variables: Wi

Eight variables account for non-policy variability (Wit) in nameplate non-hydropower

renewable capacity in the electric power industry of each state for 1996-2003. The

economic variables measure the percentage of capacity from hydropower and nuclear

power in a state, net generation, retail prices, fuel costs, renewable energy costs, and

sugarcane production, while the political variable measures a state's preferences for

renewable capacity. These variables are interacted with generation to control for different

market sizes across states.9 Table 1-1 summarizes the data for the dependent variable

(RENEWABLE CAPACITY) and the control variables.10

Total generation (GEN) is the total amount of electricity generated (in terawatthours)

in a state for a given year." It is expected that more renewable capacity will be found

in states that generate more electricity to help meet the higher demand for electricity

found in those states.12 The other control variables as well as some of the policy variables

are interacted with generation to account for market size across states. For example,



without any consideration of costs or environmental impacts. Any capacity shut down due
to PURPA contract expiration after 1996 will decrease the positive effects of any enacted
policy. There is also the possibility of a state changing its interpretation and enforcement
of PURPA after 1996, which would not be captured in the model.

9 Electricity generation in a state is has been chosen to represent market size instead of
electricity sales in a state because some electricity sales originate from outside a particular
state.

10 Data on capacity, generation, and price are found in the Historical Databases of the
Electric Power Annual survey on the EIA website. Electricity summary data is available at
the state level from the EIA.

1 A terawatt-hour (TWh) is the same as 1,000 GWh or 1 billion kWh.

12 Total generation was chosen instead of total sales because some of the electricity
demand for a state's power producers may come from other states. Generation is not
contaminated with these interstate sales, which may otherwise inflate or deflate the market
size measure. Generation and sales are highly correlated (0.952).









an increase in fuel costs will have a larger impact on renewable capacity in California

than in Rhode Island. Larger states should have more funding to 1 iv for projects to

increase renewable capacity. Renewables Portfolio Standards with Sales Requirements set

requirements on the percent of generation that must originate from renewable sources.

States with more generation will have more total generation that is required to originate

from renewable resources, which should lead to more renewable capacity in those states.

The following three variables are included in the model to control for market

structure. Two of these variables are hydropower capacity (PCT HYDROPOWER)

and nuclear power capacity (PCT NUCLEAR) as a percentage of total capacity excluding

non-hydro renewables. Hydropower should lead to less non-hydro renewable capacity

because hydropower has low marginal production costs, and the capacity typically was

constructed many years ago. With lower marginal costs and sunk capital costs associated

with hydropower, hydropower will be the first renewable energy to be implemented

because it is more economically competitive than most non-hydropower renewables

available to the electric power industry. Consumer and/or policy driven demand for

renewable-based electricity may not differentiate between hydropower and other renewable

sources, which allows hydropower to be a substitute of non-hydro renewables.

Similar to hydropower, nuclear power has low marginal costs of producing base load

electricity, has sunk capital costs, and has no emissions. If non-hydro renewable capacity

is deploy, ,1 based on economic factors, given similar emissions profiles, greater nuclear or

hydropower capacity should decrease the amount of non-hydro renewable capacity.

An alternative possibility is that regulators in states with large amounts of nuclear

power encourage power producers to use other resource types to meet new demand.

Renewable energy may be used by utilities to alleviate pressure from environmentalists

over nuclear power, thus leading to greater deployment of renewable energy capacity in

states with large amounts of nuclear capacity. The sign of PCT NUCLEAR will depend on

which of these two factors has the larger effect on power producers.









A state's annual weighted average real fuel cost (in 2002 dollars) per million Btus

(FUEL COST) measures the impact of both a state's composition of fossil fuel mix and

a state's average costs for each fossil fuel type: coal, natural gas, and fuel oil.13 FUEL

COST captures the effects of all these variables, which may have offsetting effects on

renewable capacity. FUEL COST is used instead of creating separate variables for the

cost and capacity of each fossil fuel for several reasons. First, using one variable instead of

five variables simplifies the model. Second, data on specific fossil fuel costs are missing for

many states.14

Levelized cost of each renewable source is the estimated real cost of production

per kilowatt-hour of electricity over the lifetime of the equipment, including all federal

production incentives.15 It captures the economic competitiveness of each renewable



13 Fuel cost data can be found on the EIA website in the electricity databases section
under Monthly Cost and Quality of Fuels for Electric Plants Database (FERC Form-423).
The cost per unit, Btus per unit, and number of units purchased for every fuel purchase
made by all public utilities are used to obtain a nominal average fuel cost measure. The
data are ., i'--regated and deflated using the Consumer Price Index for all goods from the
Federal Reserve Bank of St. Louis to get the state's annual average real fuel cost per
million Btus in January 2002 dollars. FUEL COST has 30 missing observations for 8
different states. Idaho is the only state without any fossil fuel purchases. Estimates of
the fuel costs are used to fill in the missing data. The non-Idaho missing observations
are extrapolated from the existing data for a state from 1990-2003. Idaho's observations
are generated by using the average fuel costs of the states bordering Idaho. A missing
data dummy variable is included in the model to capture any bias created through the
extrapolation and approximation.
14 There are missing fuel cost observations for coal (69), natural gas (65), and fuel oil
(58). This might be due to no deliveries of a particular fuel to a state, or it could be the
missing observations are due to changes in data reporting requirements during the sample
period.
15 Levelized cost is calculated by a model that accounts for the initial capital costs of
constructing the capacity, expected lifetime of the equipment, interest rates on debt,
inflation rate, fuel costs, operational and maintenance costs, capacity factor of the
equipment, and federal production incentives. Read McVeigh (1999) for a more detailed
description of levelized cost used in this paper.









energy type. Renewable energy as well as nuclear and hydropower have little or no fuel

cost and very high capital costs, while current generating technologies based on fossil fuel

have large fuel costs but lower capital costs.16 As renewable energy has gotten cheaper to

produce, it has become more economically competitive. This implies that decreases in the

levelized cost of each type of renewable energy will lead to more renewable capacity. The

levelized cost also includes federal production incentive policies that vary over time.17

MiT _,, researchers have tried to estimate the levelized cost of energy for each

renewable source. Making such an estimate is beyond the scope of this paper, so the data

set being used for this variable is obtained from McVeigh et al. (1999). The estimated

levelized cost of energy in the U.S. for each renewable energy source is in real 2002 dollars

and is estimated for every five years, from 1980 to 2005.18 These data points are used to

interpolate a polynomial curve that had the best fit (highest r-squared value). Due to this

interpolation from estimated data, the cost of energy for each type of renewable energy

is a reasonable though imprecise estimate of the decreasing cost of renewable energy

over time in the U.S. The resulting trend lines for each type of renewable energy have a

high correlation. So a weighted average of the levelized costs of wind, solar, biomass, and

geothermal for the entire U.S. is used to create the new variable RENEW COST, which is

an average national trend for renewable energy costs.19



16 Fossil fuel costs are uncertain for current generating technologies, and technical
efficiency of capital equipment is uncertain for renewable energy generation.
17 The Renewable Energy Production Incentive (REPI) and Production Tax Credit
(PTC) were passed in the Energy Policy Act (EPACT) of 1992. The level of the REPI
is decided by Congress annually, while the PTC was reenacted in 1999 and 2001.

18 The data points for 1985, 1990, and 1995 were estimated based on actual cost
information while 2000 and 2005 were forecasts made in 1999. The polynomial curves
have an order of two for biomass and geothermal and three for solar and wind.

19 The weighted average for each year is based on the sources' share of total
non-hydropower net summer renewable capacity in the U.S in 2002. Net nameplate









If renewable capacity is being constructed on economic grounds, a rise in the retail

price of electricity makes renewable energy more profitable and should have a positive

effect on renewable capacity.20 However, retail prices in a state may be simultaneously

determined with renewable capacity because using more renewable capacity increases the

average costs of production, which could lead to higher prices. Using the state's retail

price could also lead to multicollinearity problems with fossil fuel costs because higher

fuel costs will lead to higher electricity prices. To control for this endogeneity and possible

multicollinearity, the model must use a proxy for a state's retail price. A proxy must be

correlated to the endogenous variable and have no impact itself on the dependent variable.

The weighted average real retail price per kilowatt-hour of the bordering states (BORDER

PRICE) is an ideal proxy for retail prices because it meets both of these requirements.21



capacity data for each type of renewable energy are not available from the EIA, making
net summer capacity the closest available alternative measure. Even though there is a cost
of energy estimate for both solar thermal and solar Photovoltaic, the solar capacity data
are not segregated into these two types. A non-weighted average of solar thermal and PV
is taken to get the levelized cost for total solar capacity. Since all solar power accounts
for less than 2.5'. of total non-hydro renewable capacity in the U.S., it is unlikely that
using some weighted average of solar thermal and solar PV would make any significant
difference. Summer capacity refers to the maximum output generating equipment is
expected to supply to a system demonstrated by tests at the time of summer peak
demand. Nameplate and summer capacity have high correlation, but are not identical
due to different operating conditions across utilities. Definition of nameplate is in Footnote
7.
20 Average retail price is based on all sales in the market: residential, commercial,
industrial, and other customers. Data are available from the EIA Historical Databases.
Average retail price data are originally in nominal terms for each month. Two steps have
to be taken to adjust the data into real terms for each year. First, the monthly data are
divided by the CPI for all goods to get the monthly data into real terms. Second, monthly
electricity sales are used to get a weighted average price for each year. The resulting
variable is the real average retail price for each state and year in January 2002 dollars.
21 Bordering states are all states that either share a border, such as Arizona and New
Mexico, or meet at a corner, such as Arizona and Colorado. The prices are weighted by
sales in the bordering states. The correlation of retail price to BORDER PRICE is 0.836.









Using BORDER PRICE instead of the state's retail price removes the possible collinearity

with FUEL COST as well.

Florida, Hawaii, Louisiana, and Texas use the byproduct of sugar production from

sugarcane as a biomass fuel. For example, in Hawaii sugarcane is one of the primary

sources of biomass. Due to market conditions most of the sugarcane farms in Hawaii were

shut down over the 1990s, removing the fuel source for much of the biomass capacity in

the state. ('!i ,n1, in sugarcane production are likely to have an impact on the amount of

biomass capacity in a state. The change in total tons of sugarcane production from 1996

levels (SUGARCANE PROD CHANGE) is included in the model to control for its impact

on renewable capacity. SUGARCANE PROD CHANGE is the only control variable not

interacted with generation.

A political variable is included to measure changes in renewable energy preferences

in a state. The League of Conservation Voters (LCV) rating is used to determine if policy

preferences for environmental protection increase renewable energy capacity independent

from its policy effects. The League of Conservation Voters (LCV) annually publishes the

National Environmental Scorecard, which rates all congressional votes on conservational

issues by each representative.22 For example, if there are ten total votes in a year on

environmental issues and a congressperson voted in favor of conservation six times, his or

her LCV rating would be 60.

An average of all the votes by a state's representatives is taken to get the average

House of Representatives score (LCV SCORE). The scores from the House of Representatives

are used instead of the Senate because representatives have a shorter term in office



22 Data from the National Environmental Scorecard is available from the League of
Conservation Voters website, www.lcv.org. The LCV rating has been used in prior studies,
including Baldwin and Magee (2000), Kalt and Zupan (1984), and Nelson (2002).









than senators, two years versus six years. The shorter term creates greater pressure on

representatives to act according to their constituents' preferences.

A high LCV rating for a state indicates that the state's constituents are environmentally

friendly and are more likely to demand electricity from renewable energy, all other things

being equal. Consumers or environmental groups in states with higher LCV ratings may

be more likely to pressure utilities to use greater amounts of renewable energy no matter

which, if any, policies have been enacted by the state.

Policies may be endogenous to higher LCV ratings because states with congresspersons

who vote for federal pro-environmental policies may be more likely to enact state

pro-environmental policies. The policy endogeneity issue is not addressed in the body

of this paper because LCV SCORE is not a strong enough predictor of state policies to be

a satisfactory instrument. Note also that removing LCV SCORE from the regression does

not change the other results.

1.4.2 Regulatory Policy Variables: Ri

Seven of the independent variables are policy variables capturing the effects of

different types of renewable energy regulation, either by a state's legislature or Public

Utility Commission (Rit).23 Most policies are enacted through state legislation, and then

enforced by the Public Utility Commission (PUC). There are a few instances, however,

in which a PUC adopts guidelines without state legislation. No legislation or PUC action

is required for state governors to use executive orders to create a state government green

power purchasing agreement or to set voluntary goals for generation.



23 Information on renewable policies is available on the Database of State Incentives
for Renewable Energy (DSIRE) website, www.dsireusa.org, which is a project of the
Interstate Renewable Energy Council and funded by the U.S. Department of Energy.
The information is compiled from many different sources, including federal and state
officials, public utility commissions, and renewable energy organizations. The source of
the information is included within each policy description. Bollinger et al. (2001) includes
additional information on the enactment and design of Public Benefits Funds.









Policy dummy variable values are determined by a policy's enactment date, zero

before enactment and one after enactment. The enactment date is the year that the policy

is passed by the state legislator, created through an executive order, or announced as a

mandate under new PUC guidelines. Some of these policies allow a grace period for power

producers to meet the new regulations. The effective date is the year that the policy

requirements must be met. The average lag from the enactment to effective date is a little

over one year, but can be longer for Renewables Portfolio Standards. The enactment year

is a better choice to determine when the policy begins to impact the power producers.

Once a power producer becomes aware of a future requirement, it may begin to construct

any necessary renewable capacity. These actions could lead to large amounts of renewable

capacity being constructed between the enactment date and effective date.

Regulatory policies described below include a Renewables Portfolio Standard with

a Capacity Requirement, Renewables Portfolio Standard with a Generation/Sales

Requirement, Clean Energy Fund, Net A ii.- Interconnection Standards, State

Government Green Power Purchasing, and Required Green Power Options. Table 3-2

summarizes the data for the policy variables.

The first policy that will be discussed is a Renewables Portfolio Standard, which

specifies an amount of a state's electricity production, sales, or capacity that must be

renewable-based. Renewables Portfolio Standards can be differentiated into three main

structural forms, policies that set (1) mandatory renewable generation or sales levels,

(2) voluntary renewable generation or sales goals, and (3) mandatory renewable energy

capacity requirements.

The first type of Renewables Portfolio Standard sets a percentage of total generation

or sales for each power producer/retailer that must originate from renewable sources,

usually increasing every year or every few years. For example, Arizona's tiered renewable

levels that have to be met began at 0.' in 2001 and increased by 0.' each year,









resulting in a requirement of one percent in 2005. Most other states' Renewables Portfolio

Standards have similar structures, but vary in percentage levels and enforcement dates.24

Iowa, Minnesota, Texas, and Wisconsin have mandated utilities to install a certain

level of megawatts of renewable capacity.25 As long as the requirements are implemented

effectively, renewable capacity requirements should increase renewable capacity by the

same number of megawatts required by the mandate. These capacity requirements will

make this type of Renewables Portfolio Standards more effective in this model because

they target actual capacity construction versus generation or sales based Renewables

Portfolio Standards.

The differences between Renewables Portfolio Standards can be accounted for in the

model by two variables: a variable that measures the size of the renewable generation

or sales requirement (RPS: SALES REQ) and a variable that measures the size of the

capacity requirement (RPS: CAP REQ).26 The capacity requirement size and date are

used to extrapolate the expected requirement for each year assuming a linear function,

where the power producers increase capacity by the same amount each year until they

meet the final requirements, to form the variable RPS: CAP REQ.



24 In 1998, Wisconsin introduced mandatory capacity levels before it enacted a
Renewables Portfolio Standard with mandatory generation or sales in 1999. Two states
(Illinois and Hawaii) with Renewables Portfolio Sales Goals (not requirements) are counted
as Renewables Portfolio Standards with a sales requirement. The only result that changes
when these non-mandatory goals are treated as a requirement of zero is the coefficient on
RPS: SALES REQ becomes smaller and becomes less significant for all specifications. All
other coefficients remain relatively unchanged.
25 Minnesota has had both types of Renewables Portfolio Standard since 2001. A
voluntary generation goal was enacted in 2001, while the capacity requirement was enacted
in 1994.

26 Capacity requirements range from 50 to 2000 MW, and generation/sales requirements
range from 0-31 1' .









RPS: SALES REQ is an even more complex variable. The generation/sales

requirement, which usually sets a target about five years after enactment, is linearly

interpolated backwards to the enactment date of the policy. For example, a policy enacted

in 1996 with a sales requirement of 1.0' beginning in 2000 would be linearly interpolated

to be 0.2'". in 1996 and increase by 0.2'". each year until it reaches 1.0'". in 2000. Although

the requirement is not enforced until 2000, it would be necessary for power producers

to begin construction at least several years before 2000 to get the necessary capacity

constructed in time to meet the sales requirement.

Although this policy does not directly require the construction of renewable capacity,

an increase in the required amount of renewable generation may lead to a need for

more renewable capacity. If current levels of renewable capacity cannot meet a future

generation/sales requirement, additional capacity will need to be constructed.27

A Clean Energy Fund is a state-level program that is often, but not albv-l- created

through the restructuring of the electricity market and is used to fund grants, loans,

and production incentives for both research and development and actual deployment of

alternative energy. T !,J' Clean Energy Funds focus on funding actual renewable capacity

deployment, which should lead to more renewable capacity in a state.

Clean Energy Funds are paid for through System Benefits C(i rges (SBCs), which are

additional charges paid by all consumers on their electricity consumption. SBCs can be



27 Some state Renewables Portfolio Standards a with generation/sales requirement
allow the use of some hydropower electricity to meet the requirement. However, there are
normally specific requirements as to which facilities will be eligible, including restrictions
on a unit's maximum capacity, type of hydropower, and year of installation. For example,
some states do not allow generating units greater than 30 MW to be eligible. One
state does not allow any hydropower to originate from dammed hydropower plants.
Another state only allows electricity from new hydropower capacity to be eligible. These
restrictions will decrease the effectiveness of these policies to increase renewable capacity
in a state. However, the complexities of the restrictions make it difficult to create an
appropriate measure for these effects.









considered a consumption tax on electricity to fund deployment of renewable capacity in

the industry. In Minnesota, a settlement with the electric utility Xcel Energy created a

similar fund that is p .iing for renewable energy research and deployment. Maine created

a voluntary fund similar to a Clean Energy Fund for the state's customers to donate

money. 28

Similar to Renewables Portfolio Standards, Clean Energy Funds must be differentiated

to understand how effective these policies are at increasing renewable deployment in a

state. The variable used in this model is a variable that measures the amount of capacity

that is being funded for iuil' ii--, i1.- projects from Clean Energy Funds (CEF: CAP

FUNDED).29

Some customers may prefer to build generating capacity to provide their home with

some of their own electricity. Net Metering (NET METERING) allows customers that

are able to produce more electricity than they consume in a given month to sell any

excess to the utility to offset the charges for electricity in months the customer is a net

purchaser. The effect of net metering is expected to be negative because if renewable

energy demanders produce their own renewable electricity through a solar PV system

or small wind turbine, they will demand less renewable capacity from power producers.

From a utility perspective, if it is required to reach a renewable capacity or sales target,

these customer-owned facilities may serve to offset a utility's needs to build renewable



28 Database of State Incentives for Renewable Energy (DSIRE) does not include New
Mexico as having a Clean Energy Fund, while Bolinger et al. (2001) verifies that New
Mexico does have a Clean Energy Fund.
29 The capacity obligations as of 2003 are interpolated backwards linearly to the
enactment year so that an equal amount of additional capacity obligations are made each
year and total the overall obligations as of 2003. The data for is variable originated from
the Database of Utility-Scale Renewable Energy Projects from the Clean Energy States
Alliance (CESA).









capacity. NET METERING is interacted with GEN to control for the policy's effect based

on market size.

Interconnection standards (INTERCON STANDARDS) are a set of guidelines used

to safely and effectively connect individual renewable generating units to the electric

utility power grid. Some have technical requirements, such as generator type and size

limits, mandatory safety and performance standards, and insurance requirements that

must be met before a net metering customer can connect to the utility's network.

Interconnection standards must be met by any commercial, industrial, residential, or

government customer that decides to connect to the grid. Without these state policies,

the net metering connections could cause 1n i, i problems for the grid, power producers,

and other purchasers. Interconnection standards increase the costs of hooking up to

the grid for net metering and may offset some of the negative effect from net metering.

INTERCON STANDARDS is also interacted with GEN to control for market size.30

State Government Green Power Purchasing policies require that some percentage

of a state government's electricity purchases be from renewable sources. These purchase

agreements range from 5'. to 50' of a state government's electricity purchases. Similar

to Renewables Portfolio Standards with Sales Requirements, a State Government Green

Power Purchasing agreement increases the need for renewable-based electricity generation.

As state government electricity use rises, the renewable generation needed to meet the

requirement increases. If the new generation needs cannot be met by current renewable

capacity, power producers will need to construct new renewable energy capacity. The

size of the State Government Green Power Purchasing requirement, in terms of a

percentage of the state government's electricity purchases, is interacted with GEN to



30 Since only four observations have interconnection standards and no net metering, the
interaction term measures the effect of interconnection standards on states that already
have net metering policies. Only 86 of the 187 observations ( I'.' ) with net metering also
have interconnection standards, which removes concerns of multicollinearity.









control for both the state's purchase requirement and the state's market size (PCT STATE

PURCHASING*GEN).

A Required Green Power Option requires utilities to offer customers the option to

purchase renewable power at a premium. There are two versions of how these options

are implemented. The most common type gives consumers the option to make voluntary

contributions, called voluntary renewable energy tariffs in return for the guarantee that

some of the consumer's electricity consumption is produced from renewable sources.

Consumers purchase electricity at the market price and then p i,- a premium for blocks of

green electricity, usually about $2 per 100 kWh. The second type allows the producers to

charge consumers a higher rate per kilowatt-hour, but only to cover the additional costs

for electricity from renewable sources. Both the premium block rate and premium per

kilowatt-hour rate must be approved by the state's Public Utilities Commission (PUC).

Required Green Power Options elicit customer preferences and a crude measure of

willingness to p iv for renewable energy by allowing consumers to voluntarily p i,- higher

prices for the knowledge that they are supporting renewable-based electricity. The creation

of this niche market for renewable energy generation should have a positive impact on

renewable capacity. The variable REQ GREEN POWER OPT is a dummy variable, which

is interacted with GEN in the model to measure the effect of the policy based on the

state's market size (REQ GREEN POWER OPT*GEN).

1.5 Statistical Specifications and Empirical Analysis

Ordinary Least Square regressions with state fixed-effects and robust standard errors

are used in this paper to estimate total non-hydro renewable capacity. Robust standard

errors are used to account for heteroskedasticity, which was found to exist in the model

by using a Breusch-Pagan/Cook-Wesiberg Heteroskedasticity Test.31 Table 3 reports



31 The result was a Chi-Sq 448 and P(*)>Chi-Sq 0.0000, so there is a significant
difference in the variance of the dependent variable, which creates heteroskedasticity.









the regression results. Specification 1 includes only the policy variables. Specification 2

includes the economic market and political control variables, and Specification 3 replaces

RENEW COST with year dummies interacted with GEN. The following subsections

describe the results using the coefficients from Specification 3.32

1.5.1 Economic and Political Variables: Wi

The coefficient for GEN is insignificant, which cannot be easily interpreted because of

how many different v--~i that a state's generation can impact a state's level of renewable

capacity. Although the coefficient is insignificant, generation levels do have effects through

other variables that are interacted with GEN, which are explained below.

The percentage of other capacity comprised of hydropower interacted with generation

(PCT HYDRO*GEN) is not statistically significant. However, the coefficient for the

percentage of other capacity comprised of nuclear power interacted with generation (PCT

NUCLEAR*GEN) is positive and statistically significant. A one standard deviation

(12., 11'.) increase in the percentage of non-renewable capacity comprised of nuclear power

leads to an increase of 2.09 MW per terawatt-hour of generation in a state. So this one

standard deviation change in a state with a median generation level (51.15 TWh) leads to

an increase of 107 MW. It is possible that utilities with more nuclear power are deploying

more renewable capacity because the utilities are focused on diversifying its generation

mix, either to decrease the utilities' use of fossil fuels and lower emissions or to alleviate

pressure from environmentalists who are upset about the use of nuclear power.

The coefficients on the average LCV score for the House of Representatives interacted

with generation (LCV SCORE*GEN) are positive and significant. A one standard

deviation increase (26.51 points) in a state's LCV score leads to an increase of 0.663 MW

per terawatt-hour of generation. A one standard deviation increase in a state with median



32 Results from Specification 2 are nearly identical to results from Specification 3 with
the same interpretations.









generation leads to an increase of 34 MW. Preferences for renewable energy capacity do

in fact lead to a small amount of deployment of some renewable capacity, holding policies

fixed.

As expected, renewable energy cost interacted with generation (RENEW COST*GEN)

has a negative and statistically significant coefficient. A one-cent per kWh decrease in

renewable energy cost leads to an increase of 0.712 MW per terawatthour of generation.

In a state with median generation, a one cent decrease in RENEW COST leads to

an increase of 36 MW. RENEW COST decreased by 1.79 cents from 1996 to 2003,

which implies an increase of 65 MW for a state with median generation. This effect

does not just include the technological changes in renewable energy. As mentioned in

Section 4, all federal production incentives are included in the costs of production for

each renewable source, capturing the federal policy changes as well as the technological

advances. The year variable coefficients, which explain the same impacts as RENEW

COST, are explained in detail in Section 5.4.

The coefficient on average border state retail price interacted with generation

(BORDER PRICE*GEN) is negative and is marginally statistically significant only

in Specification 3 of Table 3. Higher electricity prices do not appear to result in more

renewable energy capacity construction. In fact, the negative coefficient implies that

an increase of one cent in the price of electricity leads to a small decrease in renewable

capacity by 13 MW in a state with median generation. A one standard deviation (2.07

cents) increase in price leads to a decrease of only 27 MW. It is possible that consumers in

a state with high electricity prices have less of an appetite for further increases in prices

through more expensive renewable generation.33



33 High electricity prices may be one of the factors driving the enactment of state
renewable energy policies. However, high prices by themselves do not appear to lead to
renewable capacity deployment in a state.









The coefficient for average fuel cost interacted with generation (FUEL COST*GEN)

is statistically significant. It is difficult to interpret the meaning of the coefficient, since

the components of the variable may lead to opposite effects. Higher costs should make

renewable energy capacity more competitive. But the variable also reflects differences

in fuel type use in a state. To take one example, since natural gas is more expensive

than coal, more natural gas use would lead to a higher average fossil fuel cost and make

renewable energy more competitive in the market. On the other hand, natural gas results

in lower emissions than using coal or oil. All else equal, a state with more natural gas

capacity will have lower emissions than if a state had higher amounts of coal capacity,

which lowers the need for non-emitting renewable capacity to meet emissions reduction

goals. If this holds true, a higher average fossil fuel cost will be correlated with less

renewable capacity construction.

To alleviate any concerns about FUEL COST, an additional specification is estimated

replacing FUEL COST with five variables: average cost of coal, average cost of oil, average

cost of natural gas, percent of non-renewable capacity comprised of coal, and percent

of non-renewable capacity comprised of natural gas. Natural gas and coal capacity are

treated in the same manner as hydropower and nuclear power capacity in the model. Each

variable is interacted with GEN, just like the other control variables.

A higher percentage of total non-renewable capacity comprised of coal is correlated

with more renewable capacity, which gives some support that states with dirtier

conventional capacity use more renewable capacity. As the price of coal increases, less

renewable capacity is constructed. These two results support the idea that the emissions

requirements utilities must meet are a driving force to renewable deployment, while

economic competitiveness in the market does not have much of an impact. Caution is

necessary in interpreting the results with the additional set of fuel variables because

there are many missing observations that must be extrapolated. The dummy variables

that control for missing observations for coal and natural gas are both statistically









significant, which brings up concerns about the variable coefficients and any possible

biases due to the missing data. The most important result from this specification is that

the additional variables have no effect on the policy variable coefficients, which remain

relatively unchanged relative to the original model.

The coefficient on SUGARCANE PROD CHANGE is insignificant. The missing fuel

cost dummy variable interacted with generation (M\ISSING FUEL COST*GEN) controls

for any measurement error caused by the extrapolation of the 38 missing data points and

is insignificant as well.

1.5.2 Regulatory Policy Variables: Ri

Table 4 estimates the statistically significant effects from both the control variables

and the policy variables based on a state with median generation levels. Clean Energy

Funds, Renewables Portfolio Standards with Capacity Requirements, and Required Green

Power Options have statistically significant effects on renewable capacity in the electric

power industry. Renewables Portfolio Standards with Generation/Sales Requirements and

State Green Power Purchasing Programs are marginally significant in Specification 1, but

lose their significance once control variables are introduced into the model.

CEF: CAP FUNDED, which measures the amount of capacity that the fund has

agreed to help finance, has a marginally statistically significant coefficient. This includes

capacity that has been agreed upon, but has not yet been built, either because the project

has not been finished or the project is later canceled. For each megawatt of capacity that

the Clean Energy Fund has funded or agreed to fund in the near future, approximately

0.206 MW has been constructed. This is not significantly different than the fraction of

capacity that has actually been constructed as of 2003, which was 0.33 MW per 1 MW.

Even though actual renewable capacity is probably not constructed linearly over the

lifetime of the policy, the estimates from the linear interpolation seem to be representative

of actual capacity construction due to the Clean Energy Funds.









The coefficient on RPS: CAP REQ is positive and significant, and about the same

size as would be expected. For each megawatt of capacity required by the Renewables

Portfolio Standard, approximately 1.14 MW is constructed. The coefficient is not

significantly different than one. Similar to CEF: CAP FUNDED, the linear interpolation

approach taken in designing RPS: CAP REQ is effective at capturing the policy effects by

allowing variation in the timing of capacity construction.

The most interesting result from the model is the effect that Required Green

Power Options have on renewable capacity. The coefficient for REQ GREEN POWER

OPT*GEN has a positive and statistically significant coefficient and has some of the

largest effects on renewable capacity of any variable, where enactment leads to renewable

capacity increasing by 3.46 MW per terawatt-hour of generation. A state with a median

generation level (51.15 TWh) that enacts a Required Green Power Option would have

an increase of 177 MW. Washington has the largest electricity market of states that have

enacted a Required Green Power Option (100.1 TWh), which would lead to an increase in

renewable capacity of 346 MW. To give some perspective on these results, the estimated

impacts in total renewable capacity can be expressed in terms of a percentage of total

capacity in a state. Depending on the state, a Required Green Power Option leads to

an increase of about 1.2-1.,'. 34 The statistically and economically significant increase

implies that Required Green Power Options, which create a niche market for green power

by requiring states to offer renewable-based electricity to their consumers at a premium,

are very useful in increasing the amount of renewable energy capacity in a state.35



34 These estimates are made by taking the estimated effect of a Required Green Power
Option in a state on total renewable capacity in 2003, and then dividing that value by
total capacity in the state for 2003. The resulting impact is a measure of the change in
renewable capacity in percentage terms of total capacity.

35 This estimation is in the range of the average participation rate for all local green
power programs of 1.!' (Wiser and Olson, 2004).









The coefficients for both STATE PURCHASING: PCT REQ*GEN and RPS: SALES

REQ*GEN are marginally statistically significant in Specification 1. However, once

control variables are included in the regressions, the coefficients are no longer statistically

significant. Initial state government purchase levels are minimal and account for a

relatively small portion of the electricity market in a state. The low demand can still be

met by current renewable generation in a state. Given the size and statistical significance

of the coefficient in Specification 3, it is unlikely that state government purchases of

renewable energy will result in any significant increase in renewable capacity in a state.

There are multiple reasons for the insignificance of RPS: SALES REQ*GEN. First,

most Renewables Portfolio Standards with sales requirements have been enacted fairly

recently and currently have low requirements. Power suppliers may be able to meet their

low initial requirements by using current renewable capacity. Second, some of the states

implementing these policies have available hydropower capacity already in place that is

considered eligible to meet a portion of these currently low mandates, which decreases the

policies' effectiveness in encouraging new non-hydro renewable capacity deployment in a

state. Third, some states allow the purchase of Renewable Energy Credits (RECs), which

are certificates that represent the environmental rights of renewable electricity, instead of

actual generation. Many of these same states allow power producers/retailers to purchase

RECs from out-of-state power producers to meet their in-state requirements. All three of

these factors will decrease the policy's impacts on in-state renewable capacity deployment.

A few more years of data should result in RPS: SALES REQ*GEN to have statistically

significant impacts on renewable capacity in its state.

Neither net metering nor interconnection standards appear to have an impact on

renewable capacity. The coefficients on NET METERING and INTERCON STANDARDS

are statistically insignificant.









1.5.3 State Fixed-Effects Variables: Sit

Forty-nine state fixed-effects variables are included in Specification 3 in Table 3

to control for state interpretation of federal policies enacted prior to 1996 as well as

any time-invariant differences across states.36 Time-constant variation across states

includes the availability of renewable energy resources and the initial level of a state's

preferences for renewable energy use. The coefficients should be highly correlated

to the amount of renewable capacity that existed in 1996. The correlation between a

state's initial renewable capacity and its state-fixed effect coefficient is 0.978. The state

fixed-effects seem to effectively control for the impact of existing regulation and the

market environment prior to 1996.

1.5.4 Year Variables: Tit

The third specification replaces the national renewable energy cost trend variable

(RENEW COST) with year variables interacted with GEN. The coefficients are

statistically insignificant for all years except for 2003, where the impact is 0.530 MW per

terawatt-hour of generation in a state. A state with median generation has an increase of

27 MW from 1996 to 2003. These year variables measure the impact of renewable energy

becoming more economically viable as well as federal policy implemented to encourage

renewable energy use. The federal Renewable Energy Production Tax Credit (PTC) and

federal Renewable Energy Production Incentive (REPI) were enacted in 1992. The federal

PTC was renewed in both 1999 and 2001, while the funding for the REPI changes from

year to year based on congressional appropriations. Each year coefficient controls for the

impact of these policies on each state as funding for these production incentives change as

well as the improvements in the technology behind renewable energy.



36 One state must be dropped from the model to remove multicollinearity of the
fixed-effects variables.









1.6 Conclusions

States have enacted many policies to increase the deployment of non-hydro renewable

capacity into the electric power industry in that state. The literature evaluating the

effectiveness of these programs consists of case studies and one statistical study, which

explains the use of wind power. My statistical study utilizes a larger panel, more policies,

and more control variables to explain the deployment of total renewable capacity in a

state.

Three regulatory policies appear to be effective at increasing renewable capacity

deployment in a state. The significant results from these regulatory policies confirm many

of the findings from prior case studies, which find Renewables Portfolio Standards with

Capacity Requirements and Clean Energy Funds have increased renewable capacity. An

additional policy, Mandatory Green Power Options, is also found to increase capacity

deployment in a state as well.

The previous empirical study found Public Benefits Funds, which include any Clean

Energy Fund in a state, to be insignificant in their model. My paper finds that Clean

Energy Funds with ul il li--i 1 .J projects increase the deployment of renewable capacity in

a state. By using System Benefits C!i irges (SBCs) a state can effectively make consumers

p ,i for cleaner energy without creating a different market for renewable energy demand.

Similar to the case study findings by Bolinger et al. (2001, 2004, 2005), larger utility-scale

projects make Clean Energy Funds more effective at increasing renewable capacity

deployment in a state.

This paper finds that different types of Renewables Portfolio Standards have different

effects on renewable capacity. Each megawatt of capacity mandated by Renewables

Portfolio Standards with Capacity Requirements results in the deployment of one

megawatt of additional renewable capacity in a state. But recent Renewables Portfolio

Standards that mandate generation or sales levels appear not to have statistically

significant effects. These results mirror Petersik's case study in that only Renewables









Portfolio Standards with Capacity Requirements have increased renewable capacity, but

expand on the case study by finding evidence on the size of the policy effect holding other

policies fixed.

Statewide Required Green Power Options appear to have been as effective as any

other policy. Forcing utilities to offer customers the option to purchase renewable-based

electricity at a reasonable premium rate drastically increases renewable capacity in a state.

The policy has a greater impact in larger electricity markets and appears to be effective

regardless of a state's political environment.

There are i ii' ,r renewable policy implications if these results hold when additional

years are eventually included in the model. Only five states have currently implemented

Required Green Power Options even though creating a statewide green power market

appears to be as effective at increasing renewable energy capacity in a state as a

command-and-control scheme of a Renewables Portfolio Standards or tax-and-subsidy

scheme of a Clean Energy Fund. State government purchasing agreements of renewable

energy appear to be no more than window dressing for politicians to show their support to

the environmental community, and additional funding to renewable power producers.

The remaining policies in the model do not appear to impact renewable energy

capacity construction in the the electric power industry. State government green power

purchasing does not increase renewable electricity demand enough to drive capacity

construction. Net metering and interconnection standards target residential and

commercial capacity and do not impact electric power industry decisions.

The important policy implications that arise from the results indicate policymakers

have a wide array of tools at their disposal to promote renewable energy deployment

in a state to meet environmental and energy security policy goals. The array of policy

mechanisms will become even more useful to state governments if the prospect of U.S.

climate change/carbon emissions policy becomes a reality.











Table 1-1. Dependent and Control Variables


Variable
RENEWABLE CAPACITY (MW)
GEN (TWh)
PCT HYDROPOWER (Percentage)
PCT NUCLEAR (Percentage)
BORDER PRICE (2002 cents/kWh)
RENEW COST (2002 cents/kWh)
FUEL COST (2002 dollars/MMBtu)
LCV SCORE (0 to 100)
SUGARCANE PROD CHANGE


Mean.
348.7
73.82
14.14
11.13
7.58
6.93
2.086
43.06
88.95


Std. Dev..
827.6
64.98
20.68
12.46
2.07
0.585
0.987
26.51
599.33


Min.
0.00
4.95
0.00
0.00
4.82
6.00
0.601
0.0
-1707.0


Max.
6177.4
385.63
91.59
56.20
14.49
7.79
7.431
100.0
4882.0


Median.
178.5
51.15
6.26
7.54
6.68
6.94
1.861
38.0
0.0










Table 1-2. Regressions Results
Total Non-Hydro Renewable Capacity
CEF: CAP FUNDED (MW)

RPS: CAP REQ (MW)

RPS: EFFECTIVE GEN REQ

PCT STATE GREEN POWER PURCHASING*GEN

REQUIRED GREEN POWER OPT*GEN

NET METERING*GEN

INTERCON STANDARD*GEN

SUGARCANE PRODUCTION CHANGE (Tons)

GEN (1 TWh)

FUEL COST ($/mmBtu)*GEN

FUEL COST MISSING*GEN

BORDER PRICE (Cents/kWh)*GEN

RENEW COST (Cents/kWh)*GEN

LCV SCORE*GEN

PCT HYDROGEN

PCT NUCLEAR*GEN

YR1997*GEN

YR1998*GEN

YR1999*GEN

YR2000*GEN

YR2001*GEN

YR2002*GEN

YR2003*GEN

CONSTANT

Observations
State Fixed-Effects
R-squared
Robust Standard Errors in Parentheses;* significant at


(1)
0.198
(0.107):
1.245
(0.175):
0.153
(0.056):
0.026
(0.014):
3.539
(0.667):
0.093

-0.246
(0.211)


(2)
0.144
(0.145)
1.144
(0.168)'
0.108
(0.077)
0.013
(0.017)
3.318
(1.207)
-0.237
(0.214)
-0.241
(0.195)
-0.011
(0.023)
-0.184
(1.150)
-0.154
(0.107)
-1.261
(0.843)
0.199
(0.207)
-0.712
(0.211)
0.024
(0.011)
0.018
(0.028)
0.143
(0.043)'


309.103 413.745
(7.689)*** (55.340)**
400 400
50 50
0.598 0.634
10%;** significant at 5%;'


(3)
0.206
(0.124)*
1.142
(0.168)***
0.100
(0.071)
-0.002
(0.016)
3.457
(1.166)***
-0.033
(0.227)
-0.240
(0.210)
-0.007
(0.023)


-0.124
(0.109)
-1.167
(0.802)
-0.259
(0.117)**


0.025
(0.011)**
-0.027
(0.024)
0.168
(0.052)***
-0.048
(0.151)
-0.142
(0.186)
-0.187
(0.182)
-0.269
(0.199)
0.322
(0.247)
0.250
(0.199)
0.530
(0.226)**
267.888
(50.741)***
400
50
0.646
significant at 1%



































Table 1-3. Policy Variables
Variable
CEF: CAP FUNDED
RENEWABLES PORTFOLIO STANDARD
RPS: CAP REQ
RPS: SALES REQ
NET METERING
INTERCON STANDARDS
STATE PURCHASING: PCT REQ
REQ GREEN POWER OPT


States with Policy Non-Zero Observations
8 51








































































0000 CO
rq
co -r0o
co t^^-! a


_^ cb t-
cU aL-
O Or
0 -
tb 0




fc ^


0
Co


oo
C 0
ss k











Co-<




C0t 0]
(M-
0 H


'- ^
(M 03
C-i o6











010 (M
(MOO
06^ o


Cr















bic











b-C
.- 4









O0


Cr2
C- -4









4-3 4-









CHAPTER 2
EFFECTS OF COAL CONTRACT CONSTRAINTS ON SO2 TRADING PROGRAM
COMPLIANCE DECISIONS

2.1 Introduction

Title IV of the 1990 Clean Air Act Amendments (CAAA) introduced the first sulfur

dioxide (SO2) emissions cap-and-trade program in the United States (U.S.). The program

was claimed to be a success by the Clinton Administration due to a lower than projected

allowance prices, and total compliance costs well below the estimated costs under an

alternative command-and-control policy.1

A growing body of evidence sI:. -- -I that much of the potential cost savings have not

been achieved by the Title IV SO2 Trading Program.2 State public utility commission

(PUC) regulations, adjustment costs, and long-term coal contracts have all been cited as

leading to non-cost-minimizing actions taken by many electric utilities.3 There is a body

of evidence indicating state PUC regulation has led to compliance costs that are in excess

of least-cost compliance in cap-and-trade programs as had been previously conjectured.

However, these estimates appear not to account for much of the excess compliance costs

that resulted during Phase I of Title IV.4

No work has been done to date to show the effects long-term coal or other fuel

contracts have on the ability of cap-and-trade programs to achieve the least-cost

compliance solution. Fuel contract constraints decrease the degrees of freedom in



1 Burtraw et al. (2005); Sotkiewicz and Holt (2005)
2 Carlson et al. (2000); Ellerman et al. (1997); Sotkiewicz and Holt (2005)

3 Carlson et al. (2000), Ellerman et al. (1997), Ellerman et al. (2000), Bohi (1994);
Bohi and Burtraw (1997), Fullerton et al. (1997), and Swift (2001) all listed either state
regulations or long-term coal contracts or both as possible reasons for the apparent or
potential sub-optimal behavior.

4 Sotkiewicz and Holt (2005) found actual compliance costs to be much higher than
their estimated costs while controlling for the effects of state regulation.









compliance choices on which pollution markets rely to improve cost-effectiveness in utility

decision-making. It may well be the case the presence of long-term contracts are driving

part, or most, of the deviations from least-cost that have been simulated or estimated in

the literature for Phase I. If long term coal contracts did in fact lead to inefficiencies under

Phase I, contracts could have similar effects under the newly enacted Clean Air Interstate

Rule (CAIR) of 2005 that further restricts SO2 emissions.

In this paper a model of unit-level S2O compliance is constructed that incorporates

the presence of coal contracts to examine how long-term coal contracts affect utility

compliance choices and a unit's compliance costs. As expected, the presence of coal

contract constraints leads to compliance costs in excess of the hypothetical least-cost

solution. The presence of binding high sulfur contract constraints that were likely in Phase

I of the Title IV SO2 Program may explain the lower than expected allowance prices

in Phase I that accompanied compliance costs that were above the least-cost solution.

It is also found that the presence of binding low sulfur coal constraints that may exist

under CAIR, which may lead to allowance prices that are higher than without the binding

constraint. The effects of the contract constraint seem counter-intuitive: binding high

sulfur coal constraints leading to lower excess demand for allowances, which could reduce

the allowance market price. Binding low sulfur coal constraints leading to higher excess

demand for allowances, which could increase the allowance market price. The interaction

between the contract constraints and the discrete nature of the scrubber choice leads to

these unexpected results.

2.2 Policy Background

2.2.1 Title IV of the Clean Air Act Amendment

Under the Title IV S02 emission trading program, affected units are allocated

allowances, which permit the holder to emit one ton of S02 in the year in which the

allowance is issued or any year thereafter, and that may be traded (bought or sold) in the

market or banked for future use. At the end of each year, generating units are required to









hold at least enough allowances to cover their yearly emissions to be in compliance. The

program allows generating units several degrees of freedom in choosing how to best meet

its compliance obligations: switch from high sulfur to low sulfur fuels; install scrubbers;

and buy or sell allowances; or any combination thereof.5

2.2.1.1 Phase I of Title IV

Phase I of Title IV, which ran from 1995-1999, capped the initial level of emissions

at 8.7 million tons of SO2 per year for the 110 largest polluting plants, which included

263 generating units. The EPA allocated allowances gratis to these affected units

based on average heat input during 1985-1987 multiplied by an emissions rate of 2.5

lbs. S02/mmBtu.

An additional 168 units participated in Phase I in 1996 based on the rules established

by EPA allowing a plant to "opt-in" units (7 units), designate substitution units (160

units), or designate compensating units (1 unit) as part of their Phase I compliance plans.

The voluntary participation of these additional units resulted in a total of 431 affected

generating units under Phase I. A "substitution unit" is a unit that would eventually

be affected in Phase II that voluntarily enrolled into Phase I to meet some or all of the

required emissions reductions for a Phase I unit (Sotkiewicz and Holt, 2005). Substitution

units receive an allowance allocation based on its historical heat input. A utility may

decide to reduce its electricity production at a Phase I affected unit. To do so, the utility

must have a "compensation unit" from the Phase II units the utility operates to cover the

necessary additional electricity. This compensation unit is then brought into Phase I and

given an allowance allocation based on its historical heat input. Industrial sources of S02

emissions could use the opt-in provision and voluntarily enroll into Phase I and receive



5 There are additional compliance options, including shutting down the affected unit
and shifting dispatch away from the affected unit (Energy Information Association, 1997)









allowance allocations similar to substitution and compensation units. There were seven

units that entered the program through this opt-in provision (Ellerman et al., 2000).6

Allowance prices were low compared to initial marginal abatement cost estimates,

and relatively stable throughout Phase I and the beginning of Phase II. Initial marginal

cost estimates used by the EPA ranged from $199-$226 (Smith and Ellerman, 1998). The

market opened in 1995 at a price of $150, soon hit a low of $70 in early 1996, and then

slowly rose back to around $150. Other than a slight spike in 1999 as utilities positioned

themselves for the start of Phase II, the allowance price remained relatively stable around

$150 (Burtraw et al., 2005).

2.2.1.2 Phase II of Title IV

Phase II, which began in 2000 and will continue until the implementation of CAIR in

2010, includes all units over 25 MW in generating capacity. The more than 2,000 affected

generating units throughout the U.S. were allocated allowances based on an emissions rate

of 1.2 lbs. SO2/mmBtu of heat input, multiplied by the unit's baseline heat input during

1985-1987. New generating units were given no allowances and were required to purchase

any necessary allowances in the allowance market. Phase II allocations were capped at

10.0 million tons annually in 2000, have decreased to 9.5 in 2002 where it will remain until

2010, when it drops to 8.95 million tons.

The banking provision has allowed utilities to trade intertemporally with utilities

using the substantial allowance bank accumulated through Phase I for compliance in

Phase II leading to annual emission levels in excess of 10 million tons in each year from

2000 to 2005.



6 There was an incentive to opt-in generating units voluntarily if it is beneficial to
the utility. Opting Phase II units into Phase I give utilities additional v--i to decrease
emissions and sell allowances and, apparently more importantly, bank allowances for
future use in Phase II. Actual SO2 emissions by Phase I units were much lower during
Phase I than the total allowance allocation during Phase I, which allowed utilities to bank
additional allowances for use during Phase II (Ellerman et al., 1997).









The allowance prices remained fairly stable through the beginning of 2004. A large

spike up to over $700 took place in 2004, which according to Burtraw et al. (2005) has

been associated with several factors: an increase in natural gas prices, increased electricity

demand, and the proposal of future emissions control legislation now referred to as the

Clean Air Interstate Rule (CAIR).

2.2.2 Clean Air Interstate Rule

In 2005, the EPA issued the Clean Air Interstate Rule (CAIR) that further restricts

the emissions of S02 in 25 eastern states and the District of Columbia effective in 2010.

The states under CAIR are the same states that had affected units under Phase I.

Generating units still receive their allowance allocation as defined under Phase II.

Beginning in 2010, the emissions value of the allowances for units in the CAIR region is

cut in half from 1.2 to 0.6 lbs. SO2/-\I Ihtu of heat input which implies a unit must hold

two Title IV allowances for each ton of actual emissions. Starting in 2015 units must hold

2.86 Title IV allowances for every ton of emissions, which translates to an allocation of

approximately 0.4 lbs. SO2/\ iIhtu. Meanwhile generating units outside the CAIR region

will continue operate under the Title IV, Phase II Program with trades allowed to take

place between CAIR and Phase II units. Units under Phase II and CAIR participate in

the same allowance market and face the same market allowance price.

The spike in allowance prices up to over $1,600 in 2006 seen in Figure 2.8 may

have been a result of the proposal and enactment of CAIR. Utilities could have chosen

to bank allowances instead of selling allowances in the market to ensure their ability

to cover requirements at the beginning of CAIR. It is still uncertain what led to the

temporary spike in allowance prices, but it appears likely that it was a temporary reaction

to the policy environment or other market conditions as the allowance price has quickly

decreased to its current price of less than -i .11 l/ton.









2.3 Literature Review


2.3.1 Title IV: Phase I

There has been significant research done on the Title IV S02 Cap-and-Trade

Program, both for Phase I and Phase II (Ellerman et al. (2000); Burtraw et al. (2005)).

From Table 2-1, it can be seen that, in general, the compliance cost estimates before

Phase I took effect were higher than the estimates made after Phase I became effective

and actual data could be used in the estimates. The pre-policy estimates range as high

as $1.34 billion/year with most estimates at least 1.11 million/year. The actual cost

estimates are towards the lower end of this range between $730-$990 million/year.

There are several reasons for the differences between initial estimates and actual

.,.- -regate industry compliance costs. The most important factor was the decrease in

delivered low sulfur coal prices. At the unit level, lower low sulfur coal prices decreased

the marginal cost of reducing emissions through fuel switching, which was the compliance

option chosen by 5"' of all affected units, while 3:'. of affected units chose to purchase

allowances, 10'-. installed a scrubber, :'-. shut down, and :'-. chose other methods.7

Several of these studies have estimated the cost savings resulting from the allowance

trading system. Carlson, et. al. (2000) used an econometric-based simulation model to

estimate the potential cost savings from trading in the program compared to a uniform

emissions rate standard. The potential savings was estimated at -'>iI million, Of of

which is a result of switching from high to low sulfur coal and 211' from technical change,

such as improved scrubber technology (Burtraw et al., 2005).

Keohane (2002) simulates which generating units would have installed scrubbers

under a uniform emissions-rate standard and finds that the total number of scrubbers

would have been one-third higher than the actual number of installed scrubbers under the



7 Energy Information Association, The Effects of Title IV of the Clean Air Act
Amendments of 1990 on Electric Utilities: An Update"









cap-and-trade approach. Sotkiewicz and Holt (2005) find that due to PUC regulation, not

only is there a greater number of scrubbers actually installed at the beginning of Phase I

relative to the least cost solution (18 scrubbers), but only nine of those actually installed

are at units that install scrubbers under the least cost solution. An increase in the number

and inefficient location of scrubber installations increases the total costs of compliance

because installing a scrubber is the most expensive compliance option under Phase I.

In the initial years of Phase I, many firms were not active participants in the

allowance market, choosing to switch fuels and bank allowances or shift allowances

between only their own units (Hart (1998); Ellerman et al. (1998)). The firms that did

participate mainly traded allowances within the same utility company. Bohi and Burtraw

(1997) find that intra-utility trading accounts for two-thirds of the allowance transactions

while the remaining one-third were inter-utility trades. Since most trades were made

between units owned by the same company, trading between two generating units at

the same plant would be a common occurrence. Many studies -ii-.-. -I. 1 state public

utility regulations and other state laws as a reason for the inefficiencies resulting from this

self-sufficient behavior (Bohi (1994); Bohi and Burtraw (1997); Swift (2001)).

Arimura (2002) uses econometric approaches to study the impact of PUC regulation

on compliance choices, and finds that utilities that face PUC regulation are more likely to

switch fuels instead of purchasing allowances for compliance.

Winebrake et al. (1995) estimated the cost inefficiencies from state government

restrictions on a utility's allowance trading, and estimates the total cost estimates for the

first ten years of Title IV (1995-2005). A command-and-control approach was estimated to

result in compliance costs of $4.19 billion greater than in the unrestricted permit trading

system (-. n_- billion, or an average of A 2- million/year) and an estimated allowance price

of $143/ton.

Winebrake et al. (1995) simulates the additional costs from restrictions on between-state

trading that were under consideration by both New York and Wisconsin. Both states were









trying to minimize allowance sales to states whose emissions will eventually reach New

York/Wisconsin and result in hotspots, which are are areas with extreme emissions levels

that result in greater damages in a particular area relative to damages throughout the

rest of the region. Preventing utilities in New York and Wisconsin from selling allowances

to utilities outside their state would have resulted in more than double the compliance

costs in both states, and increased nationwide compliance costs. Some of the additional

costs from these restrictions would have been offset by lower costs for utilities not in New

York or Wisconsin that would have been able to sell more allowances due to the additional

demand no longer being met by allowances sales from New York and Wisconsin utilities.

Some studies further explain the inefficiencies by examining the actual lost cost

savings that are specifically a result of state PUC regulation under Phase I. Carlson et al.

(2000) find that the actual compliance costs were $339 million (59'-.) greater in 1996 than

the least-cost solution. The study concludes the difference between actual compliance costs

and the least-cost compliance may be attributable to ,Illustment costs associated with

changing fuel contracts and capital expenditures as well as regulatory policies." Sotkiewicz

and Holt (2005) find that PUC regulations resulted in $131 million of the additional

compliance costs relative to the least cost solution. However, there is a significant amount

of compliance costs that remained unexplained.8 My study conjectures long-term coal

contracts may be responsible for what appears to be inefficient behavior resulting in

additional compliance costs.



8 Sotkiewicz and Holt model the possibility of ex post "prudence", which assumes that
there is some cost, such as future state PUC cost disallowance, to the generating unit for
choosing a less cost-effective option. If PUC regulations allow for total pass-through of
costs without threat of ex-post prudence, then a generating unit is indifferent to costs and
may not make the lowest cost compliance option.









2.3.2 Utility-Level Models of Compliance Costs

Previous studies have estimated the compliance costs at the generating unit or

utiitv--1. v,-1, and focus on the impacts of state regulation on utility-level compliance

choices. Swinton (2002) calculates the shadow prices of emissions reductions for seven

Florida power plants from 1990 to 1998 and compares their optimal choices to their actual

actions. Several factors were discussed as the reasons for some utilities making sub-optimal

decisions: state PUC regulations, program learning curve, small magnitude of potential

gains from trade, and uncertainty over the program's longevity.

Swinton (2004) follows the same approach except it expands the study to 40 plants

with data from 1994 to 1998, and introduces the possibility that long-term coal contracts

may prohibit utilities from switching coal types, although it is not modeled. Both studies

find the actual utlit.-J-1. -v. compliance costs to be much higher than the estimated

least-cost solution.

Co--.ii'- and Swinton (1996) use an output distance function to estimate the shadow

price of SO2 emissions abatement for electric power plants in Wisconsin. The study

estimated the allowance shadow price to be greater than the observed allowance prices

at the time, which they assert may be partially explained by Wisconsin's strict state

regulations on SO2 emissions.

Several studies have shown analytically or through simulation models that state PUC

regulations can lead to inefficiencies at the utility-level. Bohi and Burtraw (1992) develop

a model of utility decision-making given two compliance options, purchasing allowances or

installing emissions control technology. Bohi and Burtraw derive two recommendations so

that state regulation does not result in inefficient compliance choices by a utility. First, if

a utility's allowed return is less than its cost of capital with respect to both compliance

options, symmetrical cost recovery rules are recommended as uneven treatment of cost

recovery may create incentives for a utility to make suboptimal compliance choices.

Second, if a utility is allowed to earn more than its cost of capital with regard to both









compliance options, Bohi and Burtraw recommend the more expensive be treated less

favorably. Fullerton et al. (1997) uses a numerical model to determine the impact state

regulation will have on a utility's compliance costs by modeling the cost-minimizing

utility compliance choices and a utility's compliance choices under its Public Utility

Commission rules. The study finds that .i-,iiiii', Ilical cost recovery rules can lead to

utility compliance costs much higher than the least-cost solution, and possibly higher than

a command-and-control approach.

2.3.3 Long-Term Coal Contracts

Joskow (1985, 1988) states that coal contracts decrease transaction costs in coal

purchasing that result from uncertainty and complexity in future coal markets. A utility

may be willing to lp .i more than the current spot market price for coal to protect itself

from unexpected higher rates in the future.

Joskow (1988, 1990) finds that during periods in which the spot market coal prices

were lower than the contracted prices, the contract prices failed to adjust downward. This

downward rigidity of coal prices can lead to utility coal costs being higher than is optimal

in the short run. Some renegotiation, breach of contract, and litigation has occurred,

but nearly all contracts appear to have continued unchanged. The main reason for the

constraints in altering these coal contracts is that less than 15' of coal consumed by

utilities is supplied by a coal company owned by the same utility (Joskow, 1987). Firms

have high legal or negotiation costs of breaking a coal contract when the agreement is

made with a firm that has no financial ties to the utility. Coal contracts may also be

a result of regulations protecting the local coal industry (Arimura 2002). Due to the

inability of contracted coal prices to decrease with spot market prices, large coal price









reductions can lead to significant differences between coal contract prices and spot market

coal prices.9

Ellerman and Montero (1998) found that investment and innovation in coal

production and delivery as well as greater competition between railroads due to the

Si .-.-. i Rail Act of 1980 created lower coal prices during the first year of Phase I,

especially for low sulfur coal from the Powder River Basin. These lower coal prices led to

lower marginal costs of abating SO2 emissions through fuel switching, which is reflected in

the lower than expected allowance price in 1995 (Burtraw et al., 2005). Considering the

downward rigidity of contract coal prices, the same coal price reductions also resulted in

lower spot market prices for both high sulfur and low sulfur coal relative to the coal prices

under contract.

2.4 Inefficiencies Resulting from Coal Contract Constraints

There are three plausible scenarios where binding contract constraints results

in sub-optimal compliance choices. First, during Phase I a utility with high sulfur

coal contracts may be unable to switch to low sulfur coal for compliance when it is



9 A utility cannot sell contracted coal on the spot market because of the transaction
costs involved in selling to another utility from both its contract for coal purchases and
its contract for coal transportation. A coal contract sets a given type and amount of coal
for an agreed upon price from a particular coal source. A transportation contract sets a
given price for delivery of coal purchases from a coal source. The combination of these
two contracts results in the delivered cost of a coal purchase. For one utility to sell coal
to another utility, it would need to either buy out its contract with a provision to deliver
the coal to the other utility or it would need to p ,li for the shipment of the coal from its
facility to the other utility. There are large additional costs associated with either of these
actions. As shown in Joskow, few contracts were bought out, breached, or renegotiated.
Also, there does not appear to be any sales of coal from one utility to another.









cost-effective to do so. Consequently, the utility is forced to sub-optimally install a

scrubber or purchase allowances.10

Second, during Phase I spot coal prices were declining and often were lower than

the price of contract coal of similar characteristics (heat and sulfur content). Contract

constraints may have prevented utilities from switching to lower priced spot market coal

alternatives of similar sulfur and heat content than was being utilized under contract.

Third, under CAIR some utilities may be locked into low sulfur coal contracts entered

into for Title IV compliance and may be unable switch back to high sulfur coal and scrub

if it is cost-effective to do so. Consequently, the contract constraint pushes a utility into

sub-optimal compliance choices such as allowance purchases or scrubber installation while

using low sulfur coal.

An examination of the data for Phase I affected units indicates that of the 26

scrubbers installed by the end of 1996 in response to the passage of the 1990 Clean Air

Act Amendments, 23 of those scrubbers were installed at facilities with 40 percent or

more of its coal deliveries by contract and with 20 of them having a weighted average SO2

emission rates greater than the Phase I allowed level of 2.5 lbs. S02/mmBtu (pounds

per million Btus of heat).11 Additionally, the 14 generating units with scrubbers and

greater than 7.'-. of coal under contract all have emissions rates over 2.5 lbs. S02/mmBtu.

This indicates the possibility that high sulfur coal contract constraints are driving some



10 The actual purchasing of allowances is not what increases iiil, li-i v---ide compliance
costs. It is the sub-optimality of the allowance purchase that increases the total
compliance costs in an industry.

1 A total of 29 scrubbers were installed between 1990 and 1996. Three of the scrubbers
are not considered to have been installed for compliance of Phase I. Two were installed
on Port Washington units in Wisconsin to meet New Source Review requirements while a
third was installed on a Yates unit as results of a pilot program.









compliance decisions.12 The idea that spot market purchases may have been preferred can

be indicated in the fact that many Phase I facilities in 1996 had allowed both long-term

high and low sulfur coal contracts to expire during the 1990-1995 period and replaced

those with spot market coal of an equivalent or lower sulfur content. This may also be

an indication of potentially binding low sulfur contract constraints as utilities face future

compliance decisions under CAIR.13

Contract constraints will continue to bind a generating unit's decisions until the

expiration of the contract. The average length of coal supply contracts (weighted by

tonnage) in 1996 was 16.5 years with 53.;:' of delivered coal under contracts signed for

greater than 10 years and 22.1 of greater than 30 years. The impacts of coal contract

rigidities will dissipate over time. However, the impacts of these contracts on compliance

decisions could continue to linger for years due to the length of many of these contracts.14



12 Data used is available from the EIA FERC-423 database and Electric Power Annual
1996.
13 Data from the EIA FERC-423 Database and Coal Transportation Rate Database
show the percentage of coal purchased through contract agreements. Of the 93 Phase
I Affected Units with contract expirations between 1990-1995, 62 did not sign any new
contracts, 13 replaced high sulfur contracts with low sulfur contracts, and only 18 replaced
old contracts with new contracts for the same sulfur content. Contracts were shifted from
11-30 years (decrease of 48 to :; ') to contracts of 5 or fewer years (increase from 13 to
2,!'). Low sulfur contracted coal deliveries increased by 3-'1. and high sulfur contracted
coal deliveries decreased by 50'. for Phase I affected units between 1988 and 1997 while
non-affected units saw an increase of _i'. and a decrease of !-'., respectively. From the
available FERC-423 data for 133 plants with at least one affected unit, 34 reduced the
percentage of coal under contract by greater than 25 percent: 12 switched from high
contract coal to high spot market coal, 12 switched from low sulfur contract coal to low
sulfur spot market coal, 8 switched from high sulfur contract coal to low sulfur spot
market coal, and 2 switched from low sulfur contract coal to high sulfur spot market coal.
14 The length of the contract does not represent the years remaining on the length of the
contract. There will be variation in the time frame under which these contracts expire.









2.5 Model and Parameters

The model is a static production cost model that draws heavily from Sotkiewicz

(2003) and Fullerton et al. (1997), which simulates production costs at the generating unit

level with constraints on demand for electricity and emissions levels, and introducing high

sulfur coal and low sulfur coal contract constraints. It would seem that adding contract

constraints to the model would not cause any in iir, disruptions. However, the model

results in rather complex interpretations due to how the contract constraints interact

with the non-convexities of a unit's scrubber choice. Let "i" be the index of units. The

parameters in the model are described below.

Technology Parameters:

* zi E {0, 1} represents a generating unit's discrete scrubber choice where zi = 1 if a
unit installs a scrubber and zi = 0 if a unit does not install a scrubber.

Pi represents the levelized yearly cost of a scrubber, which are the average annual
costs from depreciation and use of capital plus the operation and maintenance costs
of installing and operating a scrubber.15

ri E [0, 1] represents the scrubber emissions capture rate or emissions removal
efficiency rate, which is the fraction of emissions that the scrubber removes from the
exhaust stream. The removal rate is independent of the sulfur content of the coal
used by a utility because it removes some percentage of emissions after production.
Depending on the scrubber technology and vintage, it can remove 25-9' -. of SO2
emissions.16

Demand Parameter:

Di represents electricity demand, in million Btus of heat input, for a given
generating unit. Demand is derived by taking the total kilowatt-hours of electricity



15 The capital costs are assumed to be $260/kW under Phase I and $141.34/kW under
CAIR. Capital costs are discounted at a 10' rate based on a 20 year equipment lifespan
( (+di). The operation and maintenance costs are assumed to be 2.0 mills/kWh under
Phase I and 1.23 mills/kWh under CAIR.
16 Table 30: "Flue Gas Desulfurization (FGD) Capacity in Operation at U.S. Electric
Utility Plants as of December 1996" from the 1996 Electric Power Annual Vol. II









demand multiplied by the heat input required to generate one kilowatt-hour
of electricity. Demand at the unit level is assumed to be fixed. Modeling each
generating unit's hourly dispatch and hourly costs in the context of varying loads
and dispatch are not easily modeled, and would require arbitrary assumptions about
how units would be utilized. For these reasons, it is assumed in this paper that
utilities do not have the option to shift electricity production across generating units
to meet demand.

Coal Parameters:

* Ci,Ci are the quantities, in tons, of high sulfur and low sulfur spot market coal use
for a given unit, respectively.

C2,Cj are the quantities, in tons, of high sulfur and low sulfur contract coal use for
a given unit, respectively.

Pi,Pi are the delivered prices, in dollars/ton, of high sulfur and low sulfur spot
market coal for a given unit, respectively.

P,,P, are the delivered prices, in dollars/ton, of high sulfur and low sulfur contract
coal for a given unit, respectively. Delivered coal prices will differ across regions of
the U.S. due to the location of coal mines across the country.17 It is assumed that
generating units are price takers in purchasing coal.

H[,,H, are the heat content for high sulfur and low sulfur spot market coal for a
given unit, respectively. Heat content is the average amount of heat, in million Btus,
in one ton of coal. The delivered price is the dollars/mmBtu paid for coal.

H,,H, are the heat content for high sulfur and low sulfur contract coal for a given
unit, respectively. The heat content will differ across regions of the U.S. due to the
heat content of coal from different coal mines across the country.

Si,Si are the sulfur content for high sulfur and low sulfur spot market coal for a
given unit, respectively. Sulfur content is the percentage of a ton of coal comprised
of sulfur.



17 For example, a generating unit in Wisconsin will have different delivered coal prices
for a particular coal type relative to a unit in Georgia. Wisconsin is closer to the low
sulfur coal mines in the Powder River Basin in the Western U.S., which results in a much
lower delivered price to Wisconsin than to Georgia.









S2',Sj are the sulfur content for high sulfur and low sulfur contract coal for a given
unit, respectively. The sulfur content will differ across regions of the U.S. due to the
heat content of coal from different coal mines across the country.

represents the rate at which sulfur is transformed into SO2, which is assumed to
be a constant (1.9) for simplicity.18

C0, Ci represent the contract constraints for a given unit, which requires the use of
a minimum amount of each coal type.

Allowance Parameters:

* Ei represents a generating unit's tons of SO2 emissions.

* Ae represents a generating unit's initial allowance allocation in tons of SO2
emissions.

Ai represents a generating unit's net allowance position in tons of SO2 emissions,
which is the difference between the actual allowances used and a unit's initial
allowance allocation. A unit is a net buyer of allowances (positive excess demand)
if it uses more allowances than its initial allocation (Ai > 0), a net seller (negative
excess demand) if it uses fewer allowances than its initial allocation (Ai < 0), and
neither if it uses exactly the same amount of allowances as its initial allocation
(A =- 0).

PA is the allowance price, which is endogenously determined in the model by
the decisions of the utilities. Each allowance that is bought (sold) will increase
(decrease) the utility's production costs by PA. Each generating unit takes PA as
given.

2.6 Generating Unit Level Decision-Making Process

The model is a static model with decisions made at the generating unit level where

each generating unit chooses its coal use, net allowance position, and scrubber choice to



is Sulfur content is the tons of sulfur per ton of coal. In this paper, any coal that results
in emissions greater than 2.5 lbs. SO2/l\ I 1tu is considered high sulfur coal. Under Phase
II of Title IV and CAIR, the high sulfur-low sulfur cut-off value is reduced from 2.5 to
1.2 lbs./\ i\l Ibtu. Under CAIR the new allowance allocation is based on 0.6 lbs./\l\I Ibtu,
which cannot be met by fuel switching alone because low sulfur coal normally ranges from
0.7-1.2 lbs./\ lil Itu with few shipments of low sulfur coal resulting in emissions of 0.6
lbs./\ l ktu. m = 1.9 for bituminous and anthracite coal, m = 1.75 for subbituminous
coal, m = 1.5 for lignite coal. These differ due to each coal types composition.









minimize its costs based on its constraints for emissions, electricity demand, and coal use

for both high sulfur and low sulfur contract coal.

2.6.1 Generating Unit's Problem



mm zPiz + PAA, + FPC + P + PCi (2-1)
zi,AiChC, ill',C 2i
subject to...A' + A, > (1- zrs)(m)(CASf, + CS>1 + CilS + C>SI) Al (2-2)

(CIH + CH, + CtjHi8 + CH>) > Di Ai2 (2-3)

Ch h /ih (2-4)

C, pgi (2-5)

C _, Ci > 0 (2-6)

Zi E {0, 1} (2-7)

Equation (3-1) represents the unit's cost function. These costs include the cost of

scrubber installation (ziPi,), net costs of allowance purchases (PAAi), and costs of coal
purchases (PihCi + PiCIQ + PfhC, + PC: ). The emissions constraint is shown in (3-2),

where the number of allowances held (Ae + Ai) must be as large as the amount of total
emissions by the generating unit [(1 zri)(rm)(CSi[ + C hSh + CSi + CizSi)]. Total

emissions is a function of the amount of each coal type used as well as the emissions

reduction due to a scrubber, if one is installed. The Lagrange multiplier on the emissions

constraint is represented by A~i. The demand constraint requires that the amount of heat

input to generate electricity (ChHh + CchHch + CQ1H, + CczH) must cover the consumer

demand (Di) for electricity expressed as heat input, which is seen in (3-3). The Lagrange

multiplier on the demand constraint is represented by Ai2. Coal contract constraints
require the unit to use a specific amount of each contract coal type, Cc for high sulfur

coal in (3-4) and ci? for low sulfur coal in (3-5). A unit will use exactly the contracted

amount because (1) if the contract coal is more expensive than spot market coal, then a









unit will not want to use any more contract coal than is necessary and (2) if contract coal

is cheaper than spot market coal, the coal producer would prefer to sell any additional

non-contracted coal through the spot market. The Lagrange multiplier for each contract

constraint on each coal type is represented by /ih for high sulfur contract coal and /il for

low sulfur contract coal.

2.6.2 First-Order Conditions

The partial derivative with respect to Ai yields the impact of a one unit change in the

net allowances purchased on the unit's total costs.

PA Ai = 0 (2-8)

Since Ai can be either positive or negative based on the net allowance position, (2-8) will

hold with equality. The additional cost to the firm of emitting one more ton of emissions is

equivalent to the allowance price, Ail = PA

Let f E {h, 1} represent the type of coal and g E {s, c} represent the type of purchase.

Each coal type has its own sulfur content (S9,, heat content (H9), and delivered price

(Pi). The partial derivatives with respect to Cf represent the impact a one unit change

in "f" type sulfur coal (high or low) from a "g" type purchasing agreement (spot market

or contract) has on the unit's total costs.


Pi$s + Ai1( ziTir()(m)() Ai2 Hh > 0, = 0 if CQ > 0 (2-9)

Pj + Al(1 zir)(m)(Sf)) iA2H > 0,= 0 if CQ > 0 (2-10)

P' + Ai1(1 z riL()(Sm)(S) Ai2Hi h = 0 (2-11)

Pi + AiI(1- z-r)(m)(S) Ai2HA H i= 0 (2-12)

The cost of using one more unit of C0 can be di- _V.regated into four different

cost changes: P,[ is the additional costs of purchasing one more unit of coal, Ai (1 -

ziri)(m)(Sf) is the additional costs of the extra emissions from one more unit of coal,









Ai2Hif is the benefit from meeting the demand remaining from using the additional unit

of coal, and pif represents the reduction in costs from meeting the contract constraint. If

Cf > 0, then (3-37) or (2-10) holds with equality. Since the contract constraint holds

with equality, (2-11) and (2-12) .i.- li- hold with equality.

2.6.3 Characterizing a Unit's Spot Market Fuel Choices and Marginal Cost of
Abatement from Fuel Switching

A generating unit's choice of fuel type is based on its scrubber installation choice as

well as its marginal cost of abatement relative to the allowance price. In this section, the

scrubber choice is taken as given and the focus is solely on comparing the allowance price

to the marginal cost of abatement of switching from high sulfur coal to low sulfur coal.

2.6.3.1 Necessary conditions for using both high sulfur and low sulfur coal

If a generating unit uses both high sulfur and low sulfur spot market coal (Cf, >

0, C, > 0), the additional costs of using each coal type are equal and (3-37) and (2-10)

hold with equality, or

P4 + Ail(1 zr)) (m)(S), s) Pi + Ai I( z-ri) (m) (S) (2 )
Hh His

(2-13) can be rearranged to isolate the shadow price of allowances or emissions, Ail, to

derive the Marginal Cost of Abatement from Switching Fuels from high sulfur spot to low

sulfur spot market coal (MCA'"8) in (2-14). Exploiting (2-8), the allowance price equals

MCA"':


PA = Ail MCA8'8 = h ((2-14)
(1- zir)(Tn)(P ||- )

The shadow price is equal to the difference in price per unit of heat divided by the

difference in emissions per unit of heat.

2.6.3.2 Only high sulfur coal use: Necessary conditions

If a generating unit uses only high sulfur spot market coal (Cfi > 0), the additional

costs to the generating unit of using high sulfur spot market coal is weakly less than using









low sulfur spot market coal and (3-37) holds with equality while (2-10) holds with weak

inequality, or

P' + A (l Zi i)(mn) (Sh) < Pi Ail(l ziri)(m) (S) (25)
------ 4----- -----4----


(2-15) can be rearranged to show the allowance price is weakly less than MCA'8:


PA < MCA8/ = H1"1 h (2-16)
(1 ziri)(m)(j-h I ( )
Th Ti

2.6.3.3 Only low sulfur coal use: Necessary conditions

If a generating unit chooses to use only low sulfur spot market coal (CQ > 0), the

additional costs of using low sulfur spot market coal is weakly less than using high sulfur

spot market coal and (2-10) holds with equality while (3-37) holds with weak inequality,

or
Pi + Ai(l zir)(mr)(Sfh) P + AI (l zri)((m)(S) (27)


(2-17) can be rearranged to show that the allowance price is weakly greater than MCA'":


PA > MCA8'8 H= H7 (2-18)
(t z)ri)(m)( ||h 71


2.6.4 Coal Use Under a High Sulfur Coal Contract Constraint

Pih is the shadow price of the high sulfur contract constraint, which includes both the
change in costs due to fuel costs and emissions. If contract coal is more expensive than

spot market coal, then Pih > 0 and it increases fuel costs. If contract coal is cheaper than

spot market coal, then Pih < 0 and it decreases fuel costs.

Assume a unit has a high sulfur coal contract and uses only low sulfur spot market

coal (no high sulfur spot market coal). So (2-10) and (2-11) hold with equality. It has

been shown in (2-17) that if a unit uses only low sulfur spot market coal, PA > MCAA'j:

P+ + Ai1(1 ziri)(m)(Sh) ih Pi + Ai(1 ziri)()(S) (
H c(2 19)









By rearranging (2-19) and exploiting (2-8), it can be shown that the allowance price

is equal to Marginal Cost of Abatement of switching from high sulfur contract to low

sulfur spot market coal (MCA'") plus an additional term representing the benefits of

meeting the contract constraint, which is weakly greater than MCA'"8:
Pih
Ai = PA = MCA,'" + ih > MCAs's (2-20)


From (2-20), if MCAS'" MCAI'" > 0, then pih > 0 and high sulfur contract coal is

more expensive than spot market high sulfur coal. If MCA'"8 MCA'"8 < 0, then pih can

be positive or negative. If Pih < 0, then contract coal is cheaper than spot market coal and

the additional compliance costs due to emissions dominate the savings from the lower fuel

costs. If Pih > 0, then contract coal is cheaper than spot market coal and the savings from

the lower fuel costs dominate the increased compliance costs due to emissions.
Pih
-h > MCA8'" MCAf'" (221)
(1 ziri)(m)( -H )
ibh Tii

Now assume a unit has a high sulfur coal contract and uses only high sulfur spot

market coal (no low sulfur spot market coal). (2-11) and (2-12) hold with equality

while (2-10) holds with weak inequality. Also, it has been shown that when a unit uses

only high sulfur spot market coal, PA < MCA"'8. Using the same approach as in the

previous case, it can be found that the allowance price is weakly less than MCA'"8 plus an

additional term representing the benefits of meeting the contract constraint:
Pih
Th PA < MCA"'' + io Sk (2 22)






determine its sign.

Ps, + Ai1(l fri)(mn)(Suh) Pl + Ai1(l zrti)()(Sch) Uih
dHe Hh









p P5 + Ai1(1 z+ri) (m) (S2) PA + Ai (l zri) (m)(S) (2
(2 23)
He He H
ih ih ih
If the additional costs, both from fuel costs and emissions, of using high sulfur

contract coal is more expensive than using high sulfur spot market coal, then tih > 0. If

the additional costs, both from fuel costs and emissions, of using high sulfur contract coal

is less than using high sulfur spot market coal, then Pih < 0. If the additional costs, both

from fuel costs and emissions, of using high sulfur contract coal is is the same as using

high sulfur spot market coal, then ih = 0.

Now assume a unit has a high sulfur coal contract and uses both high sulfur and low

sulfur spot market coal. So (2-10) and (2-11) hold with equality. Also, it has been shown

in (2-15) that when a unit uses both high and low sulfur spot market coal, PA = MCA"'8:
Pih
H0
PA= MCA' = MCA<' + ih (2-24)
(l- z)r(m)( Tn))

By solving for Ph in (2-24), the sign of Pih can be determined. If MCA'" MCA' = 0,

then Pih = 0 and high sulfur contract coal has the same cost as high sulfur spot market

coal. If MCA'" MCA'"8 > 0, then Pih > 0 and contract coal is more expensive to use

than spot market coal. If MCA'" MCA'"8 < 0, then pih < 0 and contract coal is cheaper

to use than spot market coal.
P ih
(h Sc = MCA'8 MCAI" (2-25)



2.6.5 Coal Use under a Low Sulfur Coal Contract Constraint

pI- is the shadow price of the low sulfur contract constraint, which includes both the

change in costs due to fuel costs and emissions. If contract coal is more expensive than

spot market coal, then pil > 0 and it increases fuel costs. If contract coal is cheaper than

spot market coal, then pil < 0 and it decreases fuel costs.

Assume a unit has a low sulfur coal contract and uses only high sulfur spot market

coal. It has already been shown in (2-16) that when a unit uses only high sulfur spot









market coal, PA < MCA"'8. (3-37) and (2-12) hold with equality while (2-10) holds with

a weak inequality. Using the same approach as in the previous section yields:
hil
Ail PA= MCA'c ---- < MCAs" (2-26)
(t z)ri)(m)( h )
Th 77

Using (2-26), the sign of pd can be determined. If MCA'" MCA'"s > 0, then

pil > 0 and low sulfur contract coal is more expensive to use than low sulfur spot

market coal. If MCAs'" MCAc'" < 0, then pin can either be negative or positive. If

MCA"'" MCA'" < 0 and Pid > 0, then the higher fuel costs dominate the lower costs

from emissions. If MCA'"' MCA'" < 0 and pil < 0, then the the lower costs from

emissions dominate higher fuel costs.
ILil
H(1 S S- > MCA'c MCA's, (2-27)


Now assume a unit has a low sulfur coal contract and uses only low sulfur spot

market coal. So (2-10) and (2-12) hold with equality while (3-37) holds with a weak

inequality. It has already been shown in (2-17) that when a unit uses only low sulfur spot

market coal, PA > MCA'".
l{il
PA > MCA',c _-- (2-28)


Although the sign of Pi, cannot be determined by comparing MCA<'" and MCA"'.

The first order conditions for low sulfur contract and spot market coal can be used to

determine its sign.

Pi + Ai1(l zri) (m) (S) Pi + Ail(1 ziri)(mn) (S) pi
Hi H/,

P Pi + Ai1(1 zri) (m)(Si) + Ai (1 zri) (m)(S) (229)
--> -(2-29)
H, H i H s ,

If the additional costs, both from fuel costs and emissions, of using low sulfur contract

coal is more expensive than using low sulfur spot market coal, then Pi, > 0. If the









additional costs, both from fuel costs and emissions, of using low sulfur contract coal is

less than using low sulfur spot market coal, then phi < 0. If the additional costs, both from

fuel costs and emissions, of using low sulfur contract coal is is the same as using low sulfur

spot market coal, then pi = 0.

Now assume a generating unit has a low sulfur coal contract and uses both high sulfur

and low sulfur spot market coal. So (3-37), (2-10), and (2-12) hold with equality and

PA = MCA"':

PA = MCAf' = MCA8'" s s (2-30)
(1- zri)(m) (h H- ) (
ih ii
By solving for pil, the sign of pi, can be determined. If MCA>'c MCA'" = 0, then

pi = 0 and contract coal is as costly to use as spot market coal. If MCA' M MCA'"' > 0,

then Pih > 0 and contract coal is more expensive to use than spot market coal. If

MCA',c MCA''" < 0, then Pih < 0 and contract coal is cheaper to use than spot market

coal.

MCAf' = MCA8'c (231)
( z r)(n)($ S S-
h 77Ti

2.6.6 Generating Unit-Level Compliance Costs

Total compliance costs for a generating unit are the additional costs due to ,IH.lfying

the emissions constraint, including costs from switching fuels, the costs from its net

allowance position, and scrubber installation costs. Compliance costs may be positive

or negative depending on its compliance decisions and its initial allowance allocation.

The scrubber installation costs are represented by Pi,, and will only attribute to a unit's

compliance costs if a scrubber is installed (zi 1).

The costs of a unit's net allowance position is the difference between a generating

unit's initial allowance allocation and its actual emissions multiplied by the allowance price

(PAA,).

The costs of switching fuels is the larger of two values: (1) total costs of actual coal

purchases (PCihyh + P$C4i + Ph Ci + PcCc) minus the costs of purchasing only high sulfur









spot market coal given any contracted coal (Pi'M + PCA + P5C'), or (2) zero

where:
,MAX Di- C H C Ht (2 32)
ih
The latter will only occur if it is weakly cheaper for the generating unit to use low sulfur
ps ps
coal without the emissions restrictions (Q > -l ). Notice that the contracted coal will be

used regardless and will cancel out.

ziPi, + PAA, + max{(P Cf1 + P, Pzsh MA), 0} (2 33)


Combining each of the three cost components results in a unit's total net compliance costs.

Even though the contract coal has no direct affect, the contracts will indirectly affect

compliance costs through a unit's allowance position and scrubber choice. Proposition 1

shows the sufficient conditions under which a coal contract will either increase or decrease

compliance costs.

Proposition 1: Given the scrubber choice

(i) If PA > MCA"'? and the sulfur to heat content ratio of high sulfur contract coal
(h) is greater than the sulfur to heat content ratio of high sulfur spot market coal
("), then a high sulfur coal contract increases compliance costs.

(ii) If PA < MCA"'j and the sulfur to heat content ratio of high sulfur contract coal
(h) is greater than the sulfur to heat content ratio of high sulfur spot market coal
( ), then a high sulfur coal contract increases compliance costs.

(iii) If PA > MCA"'. and the sulfur to heat content ratio of low sulfur contract coal ( )
is greater than the sulfur to heat content ratio of low sulfur spot market coal (>),
then a low sulfur coal contract increases compliance costs.

(iv) If PA < MCA"'S, then a low sulfur coal contract decreases compliance costs.

See the Appendix A for detailed proofs of Proposition 1. Proposition l(iv) may seem

counter-intuitive, but it shows the importance of being careful about defining a unit's









compliance costs versus a unit's total costs, which is something to keep in mind for the

remainder of the paper.

2.6.7 Generating Unit's Net Allowance Position: Excess Demand Correspon-
dence

Assume there are no contract constraints (C, = 0, C, = 0). A generating unit's net

allowance position, or excess demand, is the difference between a unit's initial allowance

allocation and the unit's actual allowance use as governed by (3-2). From (3-2) and (3-3),

the minimum and maximum excess demand for allowances can be formally derived.

If PA < MCA"'8 a unit will use the maximum amount of high sulfur spot market coal

which can be derived from (3-3):
CMAX Di (2-34)
Hh (2 34)
1ih

The use of all high sulfur spot market coal leads to the maximum emissions level:


EMAX (1 Zii(m)(S)( ) (2-35)
H ih

Inserting C,,MAX in for h in (3-2) gives an expression for the maximum allowance excess

demand, which is the difference between the maximum emissions level (E MAX) and the

initial allowance allocation (A:):

AMAX MAX Ae (1 z Lr)(m)(St)) A' (2-36)
11ih

If a unit's initial allocation cannot cover its maximum possible emissions, then it will have

a positive net allowance position and be a net buyer of allowances.

If PA > MCA"', a unit will use the maximum amount of low sulfur spot market coal,

which can be derived from (3-3):

CSMAX (2-37)
Hil
The use of all low sulfur spot market coal leads to the minimum emissions level:


E" = (1 zrj)(T)(S)( ) (2-38)
U,









Inserting C,,)MAX in (2 37) for C, in (3-3) gives an expression for the minimum allowance

excess demand, which is the difference between the minimum emissions level (EMIN) and

the initial allowance allocation (Ar):


AMIN MN AE (- z ()(S Ae (2-39)
11 Hli

If a unit's initial allocation can cover its minimum possible emissions, then it will have a

negative net allowance position and be a net seller of allowances.

If PA = MCA'8, a unit may use any combination of high sulfur spot market

coal and low sulfur spot market coal, which leads to any level of excess demand in the

range (E,1N A', EAX Ar). The allowance excess demand can be represented by

Ai (OEfAX (1 O)EMIN A) where the constant 0 [0, 1]. A unit that is indifferent

between fuel switching and allowances purchases could be either a net buyer or a net

seller.

Combining the excess demands for each of the three cases creates the Excess Demand

Correspondence:

A AX if PA > MCA'
A, = A AX (1 )AMIN if PA MCA'8 VO e [0, 1]

AIN if PA < MCA'"

A generating unit's excess demand correspondence can be seen graphically in Figure

2.8(i). High sulfur spot market coal use corresponds to the right-hand vertical line where

PA < MCA"'8. Low sulfur spot market coal use corresponds to the left-hand vertical line

where PA > MCA"'>. The case where a generating unit uses some combination of low

sulfur spot market coal and high sulfur spot market coal is represented by the horizontal

line at which PA = MCA"'.









2.6.7.1 Cost savings of fuel switching versus allowance purchases when
PA > MCAf'

Assuming that high sulfur spot market coal is cheaper than low sulfur spot market

coal ( < ), a generating unit that does not have an emissions constraint prefers to

use high sulfur spot market coal to meet electricity demand.19 Allowing generating units

to have a choice in their compliance options can lead to costs savings in several cases.

First, consider the case that PA > MCA"'j. Using low sulfur spot market coal leads to

the minimum number of allowances used by the unit (A"IN). Assuming that the initial

allocation by the firm is larger than the minimum possible allowance use, the unit will sell

its remaining allowances after meeting its allowance requirement. The excess demand for

such a unit can be seen in (2-39), where the excess demand will actually be negative.

The unit's cost savings from switching fuels over purchasing allowances is (PA -

MCAs's)(A AX AF"), which is the area "a+b" seen in Figure 2.8(ii). The dark-shaded

area (a) is the cost savings for the generating unit from abating emissions through fuel

switching instead of purchasing additional allowances. The light-shaded area (b) is the

cost savings from abating more than its initial allocation and selling the extra allowances.

2.6.7.2 Effects of high sulfur coal contracts on excess demand and costs

Now consider how a high sulfur coal contract will impact a generating unit's allowance

excess demand correspondence in Figure 2.8, which is summarized in Proposition 2.

Proposition 2: Given the scrubber choice,

(i) For the range of allowance prices PA > MCA"'8, a high sulfur coal contract will
weakly increase a unit's allowance excess demand.

(ii) For the range of allowance prices PA < MCA"'8, a high sulfur coal contract will
Sc S
weakly decrease excess demand if Sh < .h




19 The assumption that high sulfur coal is cheaper than low sulfur coal is supported by
actual coal prices.









(iii) For the range of allowance prices PA < MCA"'8, a high sulfur coal contract will
SC SS
weakly increase excess demand if S > .h
Hfh H Tih
Proof of Proposition 2(i):

For allowance prices PA > MCA"', a unit prefers to use all low sulfur coal, which

leads to the minimum allowance excess demand (AMIN). A high sulfur coal contract

forces some high sulfur coal use and decreases low sulfur coal use from CSMAX to C8,MAX

which is the maximum amount of low sulfur coal a unit will use given its high sulfur coal

contract:
sMAX .sMAX Di (20)
CIS,, > JihH (240)

The decrease in low sulfur spot market coal use increases emissions from EMIN to E IN

which is the minimum emissions given the high sulfur coal contract constraint:


MN MIN (1 zr)( m) (S : s,MAX + Sicj (-41)
Si < r+ 2^i


Higher emissions must be covered by additional allowances, which results in the minimum

excess demand with a high sulfur coal contract to be greater than the minimum excess

demand with no high sulfur coal contract:


= AMIN < Ai, v -MIN A (2-42)


Therefore, a high sulfur coal contract will increase excess demand for allowance prices

PA > MCA8'".E

Proof of Proposition 2(ii) and 2(iii):

For allowance prices PA < MCA"'8, a unit prefers to use all high sulfur coal, which

leads to the maximum allowance excess demand (AjMAX). A high sulfur coal contract

forces a unit to decrease its high sulfur spot market coal use from CMAX to CMAX

where:
Cis,MAX ^s,MAX Di -c H (24
ih >ih CH (2-43) i
H^sh i









SC SS
If < the sulfur content per unit of heat content is lower for high sulfur
ih Th
contract coal than high sulfur spot market coal and will decrease the maximum emissions

from EMAX to EMAX.

M EMAX < MAX t Z i m)AX + S ,~ ) (2-44)


Lower emissions result in a decrease in a unit's maximum excess demand from AMAX to

AMAX in Figure 2.8(i):

= AAX > i -. MAX E Ae (2-45)

Therefore, a high sulfur coal contract for coal with a lower sulfur to heat content ratio will

decrease excess demand for allowance prices PA < MCA'".f
SC SS
If > ', the sulfur content per unit of heat content is greater for high sulfur
Hfh Hh
contract coal than high sulfur spot market coal and will increase the maximum emissions

from EMAX to MAX.

E, E7MAX > FMAX (_ Ziri) (n) (S j- MAX + S~ (2 4
SUi +i ~ihih) 2-4)

Greater emissions result in an increase in a unit's maximum excess demand from AMAX to

AMAX in Figure 2.8(ii):

= AMAX i< Ai', EMAX AA (2-47)

Therefore, a high sulfur coal contract for coal with a higher sulfur to heat content ratio

will increase excess demand for allowance prices PA < MCA'".E

A binding high sulfur coal contract that restricts a unit's ability to switch to low

sulfur coal can force a net seller of allowances to decrease their allowance sales from AmIN

to AIN and abate fewer emissions than the generating unit would prefer as shown in

Figure 2.8(i). If the contract constraint is large enough it will shift AMIN to the right of

Ai = 0 and force a unit to be a net buyer. If the contract constraint forces the generating

unit to use all high sulfur coal, then the excess demand is a vertical line where the unit's

only choice is to purchase the maximum amount of allowances. This case may have









occurred during Phase I because some generating units purchased 1C(i' high sulfur coal

through contracts in 1996.20

High sulfur coal is assumed to be the preferred coal to use prior to Title IV as units

would wish to use the cheapest coal regardless of sulfur content, which was usually high

sulfur coal. In Figure 2.8, a unit prefers to use low sulfur coal because PA > MCA"'8, and

the compliance costs from switching fuels to abate emissions can be seen in area (a + b).

Area (b + c) is the revenue gained (negative cost) from selling the remaining allowances

that are available due to abating emissions below a unit's initial allowance allocation. The

net compliance costs for a unit will be the costs of switching fuels minus the revenues from

allowance sales, or (a + b) (b + c) = (a c). If (a c) < 0, then a unit will have negative

compliance costs.

Now consider how a high sulfur coal contract will impact the excess demand

correspondence, compliance costs, and total costs. As has already been shown above

and can be seen in Figure 2.8, a high sulfur coal contract will increase the minimum excess

demand from A"IN to A~If and may increase or decrease the maximum excess demand

depending on the relative sulfur to heat content ratio of contract to spot market coal.

These shifts in a unit's excess demand may have three distinct effects on a unit's

costs. The first cost impact is the additional compliance costs to a unit from lost

allowance sales, which is represented by area (d) in Figure 2.8. A unit must use some

high sulfur coal, which results in a unit covering additional emissions through more

expensive allowances instead of switching to low sulfur coal.

The second impact results from the difference in the sulfur to heat content ratio
SC S
between high sulfur contract and spot market coal. If h < then the maximum

emissions a unit can create will decrease and AjAX < AMAX. This will decrease the

compliance costs by (f) in Figure 2.8(i) for reducing emissions to the allowance allocation



20 Available from FERC-423 Data.









emissions level. If h > h then the maximum emissions a unit can create will increase
Th fih
and A\AX > AMAX. This will increase the compliance costs by area (f) in Figure 2.8(ii)

for reducing emissions to the allowance allocation emissions level.

Consider how these shifts in excess demand will impact a unit's net compliance

costs. The increase in the minimum excess demand will increase net compliance costs by

decreasing area (b) and area (c). If maximum excess demand decreases, net compliance

costs decrease by area (f) in Figure 2.8(i). Combining the two impacts results in net

compliance costs of (a c) in Figure 2.8(i). If maximum excess demand increases, net

compliance costs increase by area (f) in Figure 2.8(ii). Combining the two impacts results

in net compliance costs of (a + f c) in in Figure 2.8(ii).

The third cost impact results from different prices for high sulfur spot market and

contract coal, which is represented by area (e). Higher priced contract coal causes a unit

to have additional fuel costs to meeting electricity demand, which is an increase a unit's

total costs.

The graphical description of a high sulfur coal contract's impacts on excess demand

and costs can be seen in the example defined in Table 3-1. Given the coal characteristics,

demand, allowance allocation, and allowance price, and using a high sulfur contract coal

price of $1.50/mmBtu, the minimum and maximum allowance use and MCA"'8 can be

computed if a unit faces no high sulfur coal contract (C =- 0). For this example, high

sulfur contract and spot market coal are assumed to have the same characteristics, which

isolates the effect that high sulfur coal contracts have on excess demand for allowance

prices PA > MCA'".

The minimum allowance use based on the low sulfur coal characteristics is 11,400

tons while the maximum allowance use based on the high sulfur spot market coal

characteristics is 38,000 tons. The MCA'"8 is $270.68, which is lower than the allowance

price of $300.00. So a unit will switch to all low sulfur coal to minimize its total costs. In

doing so, it will use use 11,400 of its allowance allocation to cover its minimum emissions









level and sell the remaining 8,600 allowances at $300.00 each. Net compliance costs are the

additional costs from switching to low sulfur spot market coal plus the costs of allowance

purchases, which is $4.62 million. Total costs are the total cost of fuel purchases minus

allowance sales, which is :-, million.

Now consider the same unit with a high sulfur coal contract for 500,000 tons of coal,

which accounts for half of the required heat input. The maximum allowance use remains

at 38,000 tons, but the minimum allowance use decreases to 24,700 tons because of the

coal contract. The MCA8' is $90.23, which is much lower than the allowance price of

$300.00 and MCA'"8 of $270.68. The coal contract results in compliance costs (costs

of switching fuels minus allowance sales) of $5.01 million and total costs to meeting

electricity demand (low sulfur spot market coal purchases plus high sulfur contract coal

purchases plus allowance purchases) of $38.61 million. Compliance costs increased by

$390,000 because the unit could not switch all its coal use to low sulfur coal. Total costs

to the unit increased by $2.79 million because of the higher price for the contract coal.

Another possibility is that the high sulfur contract coal could actually be cheaper,

which is reasonable because the main reason for making a coal contract agreement is for

protection against future coal price fluctuations. This case can be seen in Figure 2.8,

where MCA'"8 > MCAj'" and there is actually a cost savings to the unit from using high

sulfur contract coal over high sulfur spot market coal. The net compliance costs remain

the same as in the previous example, area (a c).21 However, the total costs to the unit

will decrease relative to not having coal under the contracted price. Area (d + e) is the

additional compliance costs to the unit for not being able to switch to low sulfur coal.

Area (e) are the "cost 'i,-s,- of using lower priced high sulfur contract coal over high

sulfur spot market coal.



21 For simplicity, the sulfur to heat content ratio is assumed to be equal for high sulfur
contract and spot market coal.









By using the same assumptions as in Example 1 except changing the price of high

sulfur contract coal to $1.20/mmBtu, compliance costs and total costs to a unit when

contract coal is cheaper can be computed. MCA'"8 is ", 11.40. The change in net com-

pliance costs remain the same at $390,000 while the costs due to the lower priced coal

actually decrease by about $1.2 million. The unit actually has benefits by lowering its total

costs by $810,000 through the coal contract even though it must increase its compliance

costs.

2.6.7.3 Cost savings of allowance purchases versus fuel switching when
PA < MCAf',

Assuming that high sulfur spot market coal is cheaper than low sulfur spot market
p s S
coal ( < ), a generating unit that does not have an emissions constraint prefers

to use high sulfur spot market coal to meet electricity demand. Consider the case that

PA < MCA"'S. Using high sulfur spot market coal leads to the maximum number of

allowances used by the unit (AMAX). Assuming that the initial allocation to the unit

is smaller than the maximum possible allowance use, the unit will purchase additional

allowances to meet its allowance requirement. The excess demand for such a unit can be

seen in (2-36), where the excess demand will be positive.

Given that a unit preferred to use high sulfur coal before SO2 constraints, the unit's

cost savings from using allowances over switching fuels is (MCAf'" PA)(AAX AMIN),

which is the area "a+b" seen in Figure 2.8. The dark-shaded area (a) is the cost savings

for the generating unit from purchasing additional allowances instead of switching fuels.

The light-shaded area (b) is the cost savings from using the allocated allowances instead of

abating emissions and selling the extra allowances.

2.6.7.4 Effects of low sulfur coal contracts

Now consider how a low sulfur coal contract will impact a generating unit's allowance

excess demand correspondence in Figure 2.8, which is summarized in Proposition 3.

Proposition 3: Give the scrubber choice,









(i) For the range of allowance prices PA < MCA"'8, a low sulfur coal contract will
weakly decrease a unit's allowance excess demand.

(ii) For the range of allowance prices PA > MCA"'8, a low sulfur coal contract will
Sc Ss
decrease excess demand if < H.

(iii) For the range of allowance prices PA > MCA"', a low sulfur coal contract will
SC S
increase excess demand if > > .
H H
Proof of Proposition 3(i):

For allowance prices PA < MCA' a unit prefers to use all high sulfur coal, which

leads to the maximum allowance excess demand (AMAX). A low sulfur coal contract forces

some low sulfur coal use and decreases high sulfur coal use from Cs,'A to CAX, which

is the maximum amount of high sulfur spot market coal a unit will use given the low

sulfur coal contract constraint:

s,MAX js,MAX Di -c H, (248)
"ih > 1ih H- h H ,4h
ih ih

The decrease in maximum high sulfur coal use decreases a unit's maximum emissions from

EMAX to EAX.

SEMAX > EMAX (1 Zi i) () (S,CsMAX + S>C ) (2 49)


Lower emissions decrease the allowances used, which results in the maximum excess

demand with a low sulfur coal contract to be lower than the maximum excess demand

with no low sulfur coal contract:


SAMAX > i '" AX A7 (2-50)


Therefore, a high sulfur coal contract will increase excess demand for allowance prices

PA < MCAf".

Proof of Proposition 3(ii) and 3(iii):

For allowance prices PA > MCA"'8, a unit prefers to use all low sulfur coal, which

leads to the minimum allowance excess demand (A"IN). A low sulfur coal contract









decreases the maximum amount of low sulfur spot market coal use from Cs MAXt to
is,MAX

Cis,MAX > s,MAX Di c H (251)

SC Ss
If < :, the sulfur content per unit of heat content is lower for low sulfur contract

coal than low sulfur spot market coal and will decrease the minimum emissions from

EMIN to EIN:

SEMIN > EMIN ( )Qs (s,/1MAX +ScC) (2 52)
-^ ^i > 'i ( ,- ( ~ )i) ,., + (Sil) pc s( )


Lower emissions result in a decrease in a unit's minimum excess demand from AyAX to

AMAX in Figure 2.8(i):

SAIN > A' J EIN A7 (253)

Therefore, a low sulfur coal contract for coal with a lower sulfur to heat content ratio will

decrease excess demand for allowance prices PA > MCAj'".U
SC SS
If >HZ the sulfur content per unit of heat content is greater for low sulfur

contract coal than low sulfur spot market coal and will increase the minimum emissions
from EMIN to EMIN


= EMIN < EMIN (1- r)(m)( ,MAX + SC) (2-54)


Greater emissions result in an increase in a unit's minimum excess demand from AIN to

AIIN in Figure 2.8(ii):

SAmIN > A'''~ EN At (2-55)

Therefore, a high sulfur coal contract for coal with a higher sulfur to heat content ratio

will increase excess demand for allowance prices PA > MCAj'".E

A binding low sulfur coal contract that restricts a unit's ability to use allowance can

force a net buyer of allowances to decrease their allowance purchases from AMAX to AMAX

and abate more emissions than the generating unit would prefer as shown in Figure 2.8. If









the contract constraint is large enough it will shift A AX to the left of Ai = 0 and force

a unit to be a net seller. If the contract constraint forces the generating unit to use all

low sulfur coal, then the excess demand is a vertical line where the unit's only choice is to

purchase the minimum amount of allowances. This case may occur during Phase II if some

generating units purchased 101t' low sulfur coal through contracts.

In Figure 2.8, the net compliance costs can be seen in the shaded area (a) that

represents the costs of purchasing allowances to cover the unit's emissions above its

allowance allocation.

Now consider how a low sulfur coal contract will impact the excess demand

correspondence, compliance costs, and total costs. As has already been shown above

and can be see in Figure 2.8, a low sulfur coal contract will decrease the maximum excess

demand from A"AX to AAX and may increase or decrease the minimum excess demand

depending on the relative sulfur to heat content ratio of contract to spot market coal.

These shifts in a unit's excess demand may have two distinct effects on a unit's costs.

The first cost impact is the additional compliance costs to a unit from switching from high

sulfur spot market to low sulfur contract coal instead of purchasing allowances, which is

represented by area (c) in Figure 2.8. A unit must use some low sulfur coal, which results

in a unit abating additional emissions instead of purchasing allowances. Area (c) is the

difference between the costs of switching fuels to decrease emissions, area (b + c), and the

decrease in costs from allowance purchases, area (b). Net compliance costs increase from

(a+b) to (a + b +c).

The second cost impact results from different prices for low sulfur contract and spot

market coal, which is represented by area (d). Higher priced contract coal causes a unit

to have additional fuel costs to meeting electricity demand, which increases a unit's total

costs.

The graphical description of coal contract impacts on excess demand and costs

can be seen in the example defined in Table 2-4. Given the coal characteristics, demand,









allowance allocation, and allowance price, and low sulfur contract coal price of $1.80/mmBtu,

the minimum and maximum allowance use and MCA"'8 can be found if a unit faces no

low sulfur coal contract (C =- 0). For this example, low sulfur contract and spot market

coal are assumed to have the same characteristics, which isolates the effect that low sulfur

coal contracts have on excess demand for allowance prices PA < MCA<'.

The minimum allowance use based on the low sulfur coal characteristics is 11,400

tons while the maximum allowance use based on the high sulfur spot market coal

characteristics is 38,000 tons. The MCA'"8 is $270.68, which is higher than the allowance

price of $200.00. So a unit will purchase allowances to minimize its total costs. In doing

so, it will use use the entire 20,000 ton allowance allocation and purchase an additional

18,000 allowances to cover its maximum emissions level. Net compliance costs are the

additional costs from purchasing allowances, which is $3.6 million. Total costs are the total

cost of fuel purchases plus allowance purchases, which is $34.8 million.

Now consider the same unit with a low sulfur coal contract for 500,000 tons of

coal, which accounts for half the required heat input to meet demand. The minimum

allowance use remains at 11,400 tons, but the maximum allowance use decreases to

24,700 tons because of the coal contract. The MCA 'c is $451.13, which is much higher

than the allowance price of $200.00 and MCA'" of $270.68. The coal contract results

in net compliance costs (additional costs of allowance purchases and fuel switching) of

$4.54 million and total costs to meeting electricity demand (high sulfur spot market coal

purchases plus low sulfur contract coal purchases plus allowance purchases) of $38.14

million. Compliance costs increased by $940,000 because the unit could not use all high

sulfur coal. Total costs to the unit increased by $3.34 million because of the higher price

for the contract coal.

Another possibility is that the low sulfur contract coal could actually be cheaper than

low sulfur spot market coal, which is reasonable because the main reason for making a coal

contract agreement is for protection against future coal price fluctuations. This case can









be seen in Figure 2.8, where MCA"'8 < MCA'"8 and there is actually a cost savings to the

unit from using low sulfur contract coal over low sulfur spot market coal. The compliance

costs remain the same as in the previous example, area (a + b + c). However, the total

costs to the unit will decrease relative to not having coal under the contracted price. Area

(c + d) is additional costs to the unit for not being able to switch to low sulfur coal. Area

(d) is the cost savings of using lower priced high sulfur contract coal over high sulfur spot

market coal.

By making the same assumptions as in Example 1 except changing the price of low

sulfur contract coal to $1.50/mmBtu, we can solve for compliance costs and total costs to

a unit. MCA'c is $180.45. The change in compliance costs remain the same at $940,000

while the costs due to the lower priced coal actually decrease by $1.2 million.22 The

unit actually gains by lowering its total costs by $260,000 through the coal contract even

though it must increase its compliance costs.

2.6.7.5 Fuel switching versus allowance purchases when PA = MCA'8

In the knife-edge case a generating unit has no strict preference between purchasing

allowances and abating emissions because PA = MCA"'. The unit's excess demand may

be any value in the range [A -IN Ae, A AX- A(]. The generating unit has no preference

in compliance options because any combination of abatement and allowance purchases

result in the same compliance costs of area (a) in Figure 2.8.

A high sulfur coal contract will have the same impacts on the excess demand

correspondence when PA = MCA"' as in Section 5.7.2 where PA > MCA"'j. However,

the impacts on compliance costs and total costs will be different. The reasoning for this

is that the contract does not force a unit to use a more expensive compliance option. By

comparing Figure 2.8 to Figure 2.8, these differences can be derived. The shift in a unit's



22 For simplicity, the sulfur to heat content ratio is assumed to be equal for low sulfur
contract and spot market coal.









minimum excess demand has no impact on compliance costs. Total costs will still increase

by area (c) if the relative price of high sulfur contract coal is more expensive than high

sulfur spot market coal. The shift in maximum excess demand will still impact compliance

costs by area (f).

A low sulfur coal contract will have the same impacts on the excess demand

correspondence when PA = MCA'8 as in in Section 5.7.3 where PA < MCA"'8. However,

the impacts on compliance costs and total costs will be different. The reasoning for this

is that the contract does not force a unit to use a more expensive compliance option. By

comparing Figure 2.8 to Figure 2.8, these differences can be derived. The shift in a unit's

minimum excess demand has no impact on compliance costs. Total costs will still increase

by area (c) if the relative price of high sulfur contract coal is more expensive than high

sulfur spot market coal. The shift in maximum excess demand will not impact compliance

costs because the additional costs of abating emissions are offset by the decrease in

allowance purchases.

2.6.8 Generating Unit's Scrubber Installation Choice

It is not possible to completely characterize a generating unit's decision based on

MCA, and allowance price alone because of the non-convexities of scrubber installation.

In this section a unit's decisions with the option of installing a scrubber are examined, the

allowance price at which a unit will install a scrubber is derived, the change in the excess

demand correspondence shown, and the effect of coal contracts on the excess demand

examined.

2.6.8.1 When will a generating unit install a scrubber?

Let (Ch, C(, Ai) and (Chj, C, Ai) be the cost minimizing combination of spot coal

and allowances with and without a scrubber installed, respectively. A unit is indifferent to

installing a scrubber if the total costs with a scrubber installed are equal to the total costs

without a scrubber installed:


P. + Pfi + P, Ch + P,' = PA + PSh + P ,Q (2-56)









PS is the allowance price at which the unit is indifferent between installing a scrubber or

not. The amount of high sulfur and low sulfur contract coal will be the same both with

and without a scrubber and will cancel out, but the contract coal still affects the allowance

position.23

A generating unit's decisions will hinge on this allowance price. (2-56) can be used

to solve for PS, the minimum allowance price at which a generating unit will install a

scrubber:
pS > r. Zr Pih(vih ih) + Pil l ( Q81 Q8
PA + C )~8(C 0) (2 -57)
(A Ai)
A unit will prefer to install a scrubber at PA if the average costs of abatement from using

a scrubber is weakly less than the costs of purchasing an allowance.

2.6.8.2 Different marginal costs of abatement

The installation of a scrubber leads to an increase in the unit's marginal abatement

cost of switching from high sulfur spot market to low sulfur spot market coal relative to

the marginal abatement cost without a scrubber installed:
ps ps ps ps
il ih il ih
MCA '" iH ih H MCA'8 il ih (2-58)
(l- rj)(Tn)( ) (T)(|__- )
"Th 7T Hh 7T

Scrubber installation decreases the size of the denominator of MCA7'" by (ri)(m)(

$), which is due to the fact that only a fraction (based on the scrubber's reduction rate)

of the emissions reduction from switching fuels is realized.

2.6.8.3 Excess demand correspondence

A generating unit's excess demand correspondence becomes significantly more

complicated when a unit's scrubber installation choice is introduced into its decision-making

process. The excess demand correspondence is a combination of a unit's excess demand



23 See Appendix A for the derivation of this equation.









correspondences with and without a scrubber with a discontinuity representing the

discrete choice separating the two pieces.

A generating unit's excess demand correspondence can be derived from its optimal

compliance choices as the market allowance price changes. For allowance prices MCAj'" <

PA, a generating unit installs a scrubber because PS < PA, and switches fuels from high
sulfur spot market to low sulfur spot market coal because MCA'"8 < PA. So a generating

unit will have minimum excess demand when a scrubber is installed (AMIN), which has

already been derived in (2-36) given that zi 1.

For allowance prices PA < PA < MCA>', a generating unit installs a scrubber

because Pf < PA, and uses high sulfur coal because PA < MCA'A". So a generating

unit will have maximum excess demand when a scrubber is installed (AfMAX), which has

already been derived in (2-39) given zi 1.

For allowance prices MCAj'8 < PA < Pf, a generating unit does not install a scrubber

because PA < P, and switches fuels from high to low sulfur coal because MCA"'8 < PA.

So a generating unit will have minimum excess demand without a scrubber installed

(AMIN), which has already been derived in (2-36) given zi = 0.

For allowance prices 0 < PA < MCA"', a generating unit does not install a scrubber

because PA < Pf, and high sulfur coal because PA < MCA"'j. So a generating unit will

have maximum excess demand without installing a scrubber (AMAX), which has already

been derived in (2-39).

AAX if 0 < PA < MCA"' < P or 0 PA P< PX < MCA'
OAMAX _(1 )AMIN if (MCA = PA < P)V e [0,1]

A AIN if MCA>'" < PA< PA
Ai =S
AMAX if P < PA < MCAj'"
OASMAX (1- O)AMIN if MCA = PA V e [0,t1]

AfIN if MCA'" < PA









Using these four different excess demands for each allowance price range and the

knife-edge allowance prices, it is possible to mathematically derive the excess demand

correspondence.

The excess demand correspondence seen in Figure 2.8(i) includes a generating unit's

net allowance position under the four different allowance price ranges if (h < H ). If
1qh Ti
(H- > H ) a generating unit faces negative marginal abatement costs and will use only

low sulfur coal, which will result in a special case of only two price ranges where the only

choice a unit makes is whether to install a scrubber in Figure 2.8(ii).24

If a generating unit never prefers to switch fuels before installing a scrubber because

MCA'"8 > Ps, there is a special case in which the allowance price range MCAj'" <

PA < PAS does not exist for the excess demand correspondence. The excess demand

correspondence under this case can be seen in Figure 2.8.

2.6.9 Impact of Coal Contracts on Excess Demand Correspondence

Given the scrubber choice, a coal contract restricts the available coal use options,

which affects the excess demand correspondence as shown in Section 6.7. A contract

constraint also changes the allowance price at which a generating unit will optimally

install a scrubber, which alters the excess demand correspondence as well. Under Phase

I, a generating unit without a scrubber may prefer to use low sulfur coal, but when facing

a high sulfur coal contract constraint it would install a scrubber and prefer to use high

sulfur spot market coal. The opposite may occur under CAIR, where generating units may

face low sulfur coal contract constraints. A unit may wish to install a scrubber and use

all high sulfur coal, but the low sulfur contract constraint may result in no scrubber being

installed and allowance purchases occurring instead. The combination of these two effects

may change the excess demand correspondence for all allowance prices.



24 An example where > in the actual data is the delivered prices of coal to
generating units in the state of Wisconsin.
generating units in the state of Wisconsin.









First, it is important to generalize the indifference price at which a generating

unit will install a scrubber (Pf), which can be derived by setting a unit's costs when it

does not install a scrubber, which includes net allowance purchases (Ai) and fuel costs

(PijC + PiC,), to a unit's costs when it does install a scrubber, which includes net
allowance purchases (AMAX), fuel costs FCs s,MAX), and costs of a scrubber (P).

Assuming that high sulfur spot market coal is cheaper than low sulfur spot market coal, a

unit uses all high sulfur spot market coal if it installs a scrubber because MCA'"8 > PA. It

is uncertain if a unit will use high or low sulfur coal if it does not install a scrubber, and

will depend on the relationship between PA and MCA'".


Piz + PSASMAX + pih CMAX PAAi + P, i, + PiCi (2-59)


A unit that faces a coal contract will face a different indifference allowance price of

(PfA ) because parameters values on both sides of the equality will change. e could

be positive or negative depending on several conditions, including the type of coal under

contract. The new values that solve this equality are the contract constrained cost

minimizing parameter values. The fuel costs for the contract coal will be the same both

with and without a scrubber and will cancel out.

ASA Y -sMAX SP S
P. + (Pf AMA + PMh ( A s)AXA, + PsC + PC (2-60)

By solving for the constant P, in (2-59) and (2-60) and setting the two expressions equal

to each other, the sign and value of c can be derived.

2.6.9.1 Impact of a binding high sulfur coal contract

The impacts of a high sulfur coal contract on the two pieces of the excess demand

correspondence will be the same as in Section 6.7 where the scrubber choice is given

except there will be an additional impact on excess demand from the contract on the

allowance price at which a unit is indifferent to installing a scrubber.









There are two cases for which the value of c must be derived to determine the

contracts impact on the allowance indifference price, the first of which will have two

subcases. As will be shown for each of the cases, if ( > ), then c will be positive and

may decrease the allowance price at which a unit will prefer to install a scrubber.

In the first case, without a high sulfur coal contract, a unit prefers to use low sulfur

spot market coal without a scrubber and high sulfur coal with a scrubber. The two

subcases will be determined by the generating unit's characteristics and the size of the

contract constraint.

In the first subcase, both with or without a high sulfur coal contract, a unit prefers

to use low sulfur spot market coal if it does not install a scrubber and high sulfur spot

market coal it does install a scrubber. For this to hold, (MCA"'8 < PA < MCA7'8) and

(MCA'<" < PA < Pf).

If a unit that has no high sulfur contract coal (C =- 0), a unit uses the maximum

amount of high sulfur coal with a scrubber and the maximum amount of low sulfur coal

without a scrubber, and is indifferent to installing a scrubber at allowance price PS:

Pi + PASMAX I ps i,MAX S MIN + Pss,MAX (2 61)
4iz + P Ai + ii A i + il

(2-61) can be rearranged to find an expression for Pi:

SF = PS(A MIN ASMAX pS cs,MAX ps s,MAX (2 62)


A high sulfur coal contract will change the optimal values of the other parameters, which

will change the indifference allowance price by some value e:

S+ (PA e MAX + PM i PS ii} V psAY P iMh (2 ps MAX -63)
P. + (PjS )+ (P c)A;i p1 picc (2 63


(2-63) can be rearranged to find an expression for Pi:


(ptS (A' I ASMAX) + pssMAX s issMAX (2 64)
,^ ( ) \ A ~ ~) + i 'ilil ;ih ih.









Since Pi, is a constant, the two expressions for Pi, in (2-62) and (2-63) can be set equal

to solve for the value of e:

Ps AjIN AMIN + ASMAX ASMAX) + ps(s,MAX Cs,MAX) Ps sMAX c_ s,MAX
(A'MIN ASMAX)
(2-65)

Filling in for allowances and coal use parameters, it is possible to determine the sign

of C.





SMAX Di )((1 -A SMAX [c+, (D Cihy" ( (i)( e
H -h Ds C=hH, h Sih)(m A
A \~n( )(sih) (r)(1 r-)-A A [C\hSih+( H- Vi)h)] (rn)(1 r^) Ai
ih ih
s,MAX Di C ^s,MAX DI Chh As,MAX Di ^s,MAX DI ChH
Hih T Hs h Hi, H ils
iih "ih "i iii
By filling these expressions into (2-65), the expression for c can be simplified to parameters

for coal price, sulfur content, heat content, and scrubber capture rate.

0(C c S) (Tn r, h c Th ) +ch os' 1 os _s F D
SH (PsA(m) (rih h h I ) i) l ih
e = (2-66)
(AMIN ASMAX)

An interpretable form is derived by separating terms, multiplying though by (AIN

AMAX), and dividing through by m(| | ).
rS S( P Sh

v A S VnMAX) C hH sh SH PAS H H)
H H7 H 7 Hi

rs, ( ;sh SS
Ai i ih h S;7h ;7 + ( A ,
("V_ SMAX HcHh s S pS MCA') (2-67)
Th a 7T
From the initial assumption that a unit uses low sulfur coal when it does not install a

scrubber, it is known that Ps > MCAs'5. So if (; > ), then c > 0 and c increases as
ih Hih
the size of the high sulfur coal contract (C~h) increases. If (; < T), then the sign of c is
iuh ih
unknown.









An example reflective of Phase I data will help to explain which sign is most likely

for c. The example in Table 2-6 is based on data from the a unit that installed a scrubber

under Phase I.25 The first example will consider c given that a unit uses low sulfur

coal from the Southern Appalachian Mountains and high sulfur coal from the Northern

Appalachian Mountains. Southern Appalachian coal has a low average sulfur content

(0. 1.' .). In this example, high sulfur spot market coal is assumed to have both a lower

sulfur content and price than high sulfur contract coal. Given the assumptions in Table

2-6, a unit will prefer to use low sulfur coal if it does not install a scrubber because

MCA'8 = $244.26 and PA = -'0. If a scrubber is installed, a unit prefers to use

high sulfur spot market coal because the marginal cost of abatement increases above the

allowance price to MCA'"8 = $4, 885.20. The annualized scrubber costs are Pi = $15.886

million, which results in P' = $1,572.56. A high sulfur coal contract for 50'.- of coal use

results in a decrease in the indifference allowance price of $1, 210.61 to (PA c)

$382.00 (a 71.'- decrease in PA). Although a unit's compliance choices are not altered at

PA = -'0, the large decrease in c encourages scrubbing at much lower allowance prices
than without the coal contract. This subcase will hold for a contract constraint of less

than 73.5'. of coal use. This example shows that for many units, high sulfur coal contracts

would not have enough of an impact to result in scrubber installation in Phase I. Only

about 10' of units actually installed scrubbers to comply with Phase I, but these units

account for over half of total abatement by affected units.

In the second subcase, assume that without a high sulfur coal contract, a unit

prefers to use low sulfur spot market coal if it does not install a scrubber and high sulfur

spot market coal it does install a scrubber because (MCA''" < PA < MCAj'8) and

(MCA>'" < PA). However, with a high sulfur coal contract, a unit prefers to use high

sulfur spot market coal both with and without a scrubber because (PAF c < MCA"'j).



25 Gen. JM Gavin had 1011'- coal under contract.









If a unit that has no high sulfur contract coal (C~ = 0), a unit uses the maximum

amount of low sulfur coal if a scrubber is not installed, but uses the maximum amount of

high sulfur coal if a scrubber is installed. The unit is indifferent to installing a scrubber at

allowance price Pj:

z + PSMAX I ps sMAX p PAMIN psCMAX (2 68)
Piz+ Al i + Pis 7 h ACi + Vi [1-6)


(2-68) shows that the costs with and without a scrubber are equal for a given Pf, and can

be rearranged to find an expression for Pi:

SF, PS(AMIN- ASMAX) ps s, MAX ps sMAX (269)
A- ihih il'il

Since a unit uses low sulfur coal without a scrubber, a high sulfur coal contract will

change the coal use by a unit without a scrubber and change the indifference allowance

price by some value e:


PLi + (PS c)MAX + P ICDS AX + S (P i e i + MPAX Pich (2-70)
A 7,h (PA ihvh +lC1ih

(2-70) can be rearranged to find an expression for Pi,:


SPi = (PS )(Ail A AS MAX) (2-71)

The two expressions for Pi, in (2-69) and (2-71) can be set equal to solve for the

value of e:

P~(SMAX AMIN ASMAX SMAX) _ps s,MAX p s,MAX
Sih i (2 72)
SA(AMAX_ ASMAX)

By filling in for allowances and coal use, it is possible to determine the sign of e.

AMIN (D i(f)(m)- A iA ;'" (Di c A
n A/e;l Ah + Cc Sih)Tf -
Hisi Hsh ih

MAXih Di )() )- i MAX) (D ih- c H) ( )(- r_) A'
Hih ih









s,MAX Di s,MAX Di
ih "i

By filling into (2-72), the expression for c can be simplified to parameters for coal price,

sulfur content, heat content, and scrubber capture rate:


S A ) m ThH H) + Di ( D H )
(2-73)

The new form in (2-73) is easier to determine the sign of c because the sign of the

left-hand side is unchanged while simplifying the right-hand side. Now get the right-hand

side into a form that is interpretable by separating terms and dividing through by

Tn( $( h Lh
m(h 7H7)


V=>(A"A MAX) pS rrHc h h D H HT h (2-74
-- Ai ( I-P2D.- (2 74)
i (l i l(
ih T1 Hih if

The last two terms in (2-74) can be combined to get Di(P MCA"'8):

SMAX p Hcih Haih s i -A 75)
i i A riMACihX) ;H5ih SSh + D(PA MCA') (2 s5)
ih il

From our initial assumption that a unit uses all low sulfur spot market coal if it does not

install a scrubber, if it does not have a high sulfur coal contract, then PA > MCAj'". If

(; H> ), then the first term is non-negative, which makes c > 0. As in Case 1, an
increase in the size of the high sulfur coal contract (C. ) increases the value of e. However,

if (h < jh) then it is uncertain whether c is positive or negative.
ih Hih
Assuming the data in Table 2-6, this case will occur if the contract is greater than

73.5'. of coal use. For this case to occur, a high sulfur coal contract must result in a

unit preferring to use high sulfur coal without a scrubber installed when, without a

contract, it initially prefers to use low sulfur coal without a scrubber (PFA > MCA"'j and

PAS < MCA"'j). A unit must also prefer to use high sulfur coal with a scrubber both

with and without a high sulfur coal contract (PfA < P1s < MCA'8). The allowance









price at which a unit will initially install a scrubber is PA = $1, 592.61, which is much

higher than the assumed allowance price (PA = .-1-.00). As the size of a contract coal

increases from 50'- to 7".' of coal use, c increases from c = $1, 210.61 to c = $1, 349.82,

respectively. With a coal contract of 50' a unit still prefers to use low sulfur coal if it

does not install a scrubber because (PAS e) = $382.00 > MCA'8 = $244.26. However,

as the coal contract increases in size to 7,' a unit now prefers to use high sulfur coal if

it does not install a scrubber because (Ps e) = $242.79 < MCA' = $244.26. Also, a

unit will now install a scrubber because (PAS e < PA), which may have a large impact on

compliance costs.

In the third case, assume that a unit, both with and without a high sulfur coal

contract, prefers to use high sulfur coal both with and without a scrubber because

(PA < MCA"') and (PF < P < MCA'"). Given that a unit that has no high sulfur

contract coal (Ch = 0), a unit is indifferent to installing a scrubber at allowance price PAS:

P + PS MAX sMAX MAX + ps sMAX AX + P AX(2-76)
PAz i + FihAih A' i ihih (2

(2-76) can be rearranged to find an expression for Pi, where the only change in costs is

the change in a unit's net allowance position:


SP, P1A(AMAX ASMAX) (2-77)

A high sulfur coal contract will change the indifference allowance price by some value

c because it alters the costs from net allowance purchases and fuel costs:


P, + (P )nSMAX + P;sMA + P (PS )A" + Ps~M AX + Pit (2 78)

(2-78) can be rearranged to find an expression for Pi,:


SPi = (P e) (A MAX) (2-79)









The two expressions for Piz in (2-77) and (2-79) can be set equal to solve for the

value of e:
ps(AMAX AMAX + ASMAX_ ASMAX)
( ( i (2-80)
(AMAX ASMAX)

Filling in for allowances and coal use, it is possible to determine the sign of e.

AMAX D ( ) (S)( )- A '" (D- C~hHhh Si + Cc S~h) (m) Ae
iil H iH8 ihih)h i

SMAX sDi ))(1- S-MAX Di C hHS e+ c ()(I A
Ai [ ih)(T ) (t ri) -A' Ah %C \SQ(n)(tlrj)- A
i Hsh i Hsh ih i
ih ih
By filling into (2-80), the expression for e can be simplified to parameters for coal price,

sulfur content, heat content, and scrubber capture rate:


= (A" ASMAX) = P SmrTi, H0H( Sih ih) (2-81)
ih ih

If high sulfur contract coal has a weakly higher sulfur to heat content ratio than high

sulfur spot market coal (- > ), then e > 0 and as the amount of coal under contract
Tih ~- ih
(C h) increases, the size of c increases. If ( < ), then e < 0.
Hih ih
Under the initial assumptions, PAS < MCA"'8. The example in Table 2-7 uses data

representative of a unit that installed a scrubber under Phase I.26 A unit facing these coal

characteristics prefers to use high sulfur coal with or without a scrubber installed because

MCA S'" = $1,137.74 > PA= 250.00 and MCA'" = $1,137.74 > PA = 460.80. In this

case, e is much smaller than in the previous examples at e = $44.66 because a unit does

not prefer to use low sulfur coal without a scrubber. There are no additional compliance

costs from not being able to switch from high to low sulfur coal. Although the size of c is

smaller than in other examples, the initial size of P' is much smaller than in the previous

two cases. So a smaller value of c may still alter a unit's compliance choices.



26 BL England had 1 'I. of coal under contract.









Based on the above two cases, the sign of epsilon can be summarized in the following

Proposition 4.

Proposition 4: Given a high sulfur coal contract,

(i) > 0 if (h > s|).
ih ih

(ii) The sign of c is unknown if (- < ).
ih ih "
Given the impact a high sulfur coal contract has on excess demand defined in

Proposition 2 and the sign of c defined in Proposition 4, the impact of a high sulfur coal

contract is derived in Proposition 5(a) and 5(b).
SC S
Proposition 5(a): Assuming (- > ) and allowing for the scrubber choice

(i) For the range of allowance prices PA > MCA"', a high sulfur coal contract will
,. .i,, increase a unit's excess demand.

(ii) For the range of allowance prices PAS < PA < MCA'S8, a high sulfur coal contract
will ,.. r. ;, increase excess demand.

(iii) For the range of allowance prices (PAS ) < PA < PS, a high sulfur coal contract
will ., tl. /; decrease excess demand if A^IN > A;MAX and 'r.. I;, increase excess
demand if AMIN < AMAX

(iv) For the range of allowance prices MCA'"8 < PA < (PAS e), a high sulfur coal
contract will d, A,/; increase a unit's excess demand.

(v) For the range of allowance prices 0 < PA < MCA"', a high sulfur coal contract will
1. .li, increase a unit's excess demand.

Proof of Proposition 5(a):

(i) When a unit faces PA > MCAA'j, a unit prefers to install a scrubber and use all low
sulfur coal. From Proposition 2(i), a high sulfur coal contract increases the minimum
emissions level, which will weakly increase a unit's allowance excess demand.

(ii) When a unit faces PAS < PA < MCA7'8, a unit prefers to install a scrubber and use
all high sulfur coal. From Proposition 2(iii), given ( > ), a high sulfur coal
ffih Hih
contract increases the maximum emissions level, which will weakly increase a unit's
allowance excess demand.

(iii) From Proposition 4(i), when a unit faces (PA c) < PA < PS, a high sulfur
coal contract decreases a unit's indifference allowance price of installing a scrubber









below the allowance price by c > 0, which leads to a unit installing a scrubber
where it initially would not and decreases a unit's emissions and a unit's excess
demand. From Proposition 2(iii), given the scrubber choice and (- > t), a high
ih Hih
sulfur coal contract will weakly increase a unit's emissions and excess demand. If
A"IN > A~MAX, the combined net effect of the countering shifts is weakly negative
and weakly decreases excess demand. If A"IN < AMAX, the combined net effect is
weakly positive and weakly increases excess demand.

(iv) When a unit faces (S > h), a unit does not install a scrubber and prefers to use
Hih ih
low sulfur coal. From Proposition 2(i), a high sulfur coal contract increases a unit's
minimum emissions level and excess demand.

(v) When a unit faces 0 < PA < MCA'8 a unit does not install a scrubber and prefers
to use high sulfur coal. From Proposition 2(iii), given ( > ), a high sulfur coal
Hffh H- h
contract increases a unit's maximum emissions level and excess demand.E

Proposition 5(a) is shown graphically in Figure 2.8(i). Proposition 5(b) expresses the

impact a high sulfur coal contract will have on a unit's excess demand correspondence by

assuming that e > 0.

Proposition 5(b): Assuming (> < ), e > 0, and allowing for the scrubber choice

(i) For the range of allowance prices PA > MCA"'8, a high sulfur coal contract will
,. .i;, increase a unit's excess demand.

(ii) For the range of allowance prices PA < PA < MCA7' a high sulfur coal contract
will ;, i... ,/; decrease excess demand.

(iii) For the range of allowance prices (PAS e) < PA < PS, a high sulfur coal contract
will ; i..i. ,; decrease excess demand.

(iv) For the range of allowance prices MCAj'" < PA < (PAS c), a high sulfur coal
contract will ,. A,/; increase a unit's excess demand.

(v) For the range of allowance prices 0 < PA < MCA"', a high sulfur coal contract will
., l.; decrease excess demand.

Proof of Proposition 5(b):

(i) When a unit faces PA > MCA'"8, a unit prefers to install a scrubber and use all low
sulfur coal. From Proposition 2(i), a high sulfur coal contract increases the minimum
emissions level, which will weakly increase a unit's allowance excess demand.









(ii) When a unit faces P < PA < MCA7'8, a unit prefers to install a scrubber and use
all high sulfur coal. From Proposition 2(ii), given (, < ), a high sulfur coal
contract decreases the maximum emissions level, which will weakly decrease a unit's
allowance excess demand.

(iii) If c > 0, when a unit faces (PAS ) < PA < PA, a high sulfur coal contract decreases
a unit's indifference allowance price of installing a scrubber below the allowance
price, which leads to a unit installing a scrubber where it initially would not and
decreases a unit's emissions and a unit's excess demand. From Proposition 2(ii),
Given the scrubber choice and ( < ), a high sulfur coal contract will weakly
decrease a unit's emissions and excess demand. The combined net effect is weakly
negative and weakly decreases excess demand.

(iv) From Proposition 2(i), when a unit faces MCA"'8 < PA < (PA e), a unit does
not install a scrubber and prefers to use low sulfur coal. A high sulfur coal contract
increases a unit's minimum emissions level and excess demand.

(v) From Proposition 2(ii), when a unit faces 0 < PA < MCA'S a unit does not install
a scrubber and prefers to use high sulfur coal. Given (>h < h ), a high sulfur coal
Hih i- Hih
contract decreases a unit's maximum emissions level and excess demand.E

Proposition 5(b) is shown graphically in Figure 2.8(ii).

There are conditions under which some of these allowance price ranges do not exist.

For example, PA' < MCA'"8 in the third case described above. So there is no price

range (MCA'", PAS) in Figure 2.8. However, Propositions 5(a) and 5(b) still hold for the

price ranges that do exist. In Case 2, c is large enough to shift the allowance price from

PA > MCA"'8 to (PA c) < MCA"'j and causes the visual representation of the excess

demand correspondence to shift from Figure 2.8 to 2.8.

2.6.9.2 Impact of a binding low sulfur coal contract

Proposition 5 can be proven using the same approach that was used to determine the

sign of c with a high sulfur coal contract is used to show that c is alv--, less than or equal

to zero, and may increase the allowance price at which a unit is indifferent to installing a

scrubber. Once again there will be three case under consideration.

In the first case, assume that both with and without a low sulfur coal contract, a unit

prefers to use low sulfur spot market coal if it does not install a scrubber and high sulfur









spot market coal if it does install a scrubber because (MCA''8 < PA < MCA'8) and

(MCAf'" < PA).

If a unit has no low sulfur contract coal (C^i = 0), a unit uses the maximum amount

of high sulfur coal with a scrubber and the maximum amount of low sulfur coal without a

scrubber, and is indifferent to installing a scrubber at allowance price PS:


P + PSASMAX + Ps WMAX
Piz + PAsi + 7ihih


PSAMIN +PirsCs,MAX
Ai + ilil


(2-82)


(2-82) can be rearranged to find an expression for Pi,:


z = p M IN ASMAX) P h sMAX C ssMAX
P?, iz TAs (Ai A 7,h + ~, Ii i C IS,


(2-83)


A low sulfur coal contract will change a unit's coal use and allowance purchases, which

changes the indifference allowance price by some value e:

P + (P ) SMAX pi s ^s,MAX c _c /S i s i s ,MAX + cC (284)
(2z (P ih ih +rearra d to find an exprsion + (2-4)

(2-84) can be rearranged to find an expression for Pi,:


ASMAX) + P^sls,MAX _s is,MAX
Ai ) + 7, ~ ,h C1


(2-85)


The two expressions for Pi, in (2-83) and (2-85) can be set equal to solve for the value of


PAs(A"IN AmIN + AMAX


(AMIN ASMAX)
(2-86)


Filling in for allowances and coal use, it is possible to determine the sign of e.

AIN (D ( A D Cc," H ) (S)m) A
nil= j( l


AiDMAX (Di H i ,)(S)+sC s (m)(1- r)- A
H,)+


A X ( T ) (Sh) (T,) (t- ri)- A


=> C


= > z = (PA 0) (i" '










C8MAX C, MAX D, ?H
s,MAX D_ i s,MAX Di CizHH z Cs,MAX Di Cs,MAX D, -C H-
h "i ih ii ilH

By filling these expressions into (2-86), the expression for c can be simplified to parameters

for coal price, sulfur content, heat content, and scrubber capture rate.

C, H(P D(m [(r h h () H %S]H
[[ HH Hl ] HH \(2 87)
(AMIN ASMAX)

A more interpretable expression is derived by multiplying through by ((A" AfMAX))

and dividing through by m(h H ).

ihih



S OS
((H, WA i) $- (PA MCAn S' (289)
ih TI
From the initial assumption that a unit uses low sulfur coal when it does not install a

scrubber, it is known that (PS > MCA"'). Since (> < h ), then e < 0 and its
il ih
magnitude increases as the size of the low sulfur coal contract (Ci,) increases.

A binding low sulfur coal contract is more likely to impact units under CAIR. An

example using recent data reflective of the coal availability and delivered prices for a

unit in Alabama in 2000 will help to show e < 0.27 Under the assumptions, a unit

will prefer to switch fuels to abate emissions if it does not install a scrubber because

MCAs's = -'11.40, which is much lower than the assumed allowance price PA $700.00.

If a scrubber is installed, a unit prefers to use high sulfur spot market coal because the

marginal cost of abatement is much greater than the allowance price (MCA'"8 = $11, 228).

A unit will not install a scrubber because the allowance price at which a unit is indifferent

to installing a scrubber (PA = $731.46) is higher than PA. A low sulfur coal contract



27 The data can be found in Table 15.A of the 2000 Electric Power Annual Volume II.









for 5(1 of coal use results in e = -$42.00, which increases the indifference price to

(PA c) = $773.46. The example shows that some units will not be impacted by a

low sulfur coal contract because the increase in the indifference price does not alter the

scrubber choice.

In the second case, assume that without a low sulfur coal contract, a unit prefers

to use high sulfur spot market coal both if it does or does not install a scrubber because

(Pf < MCA"'8). However, with a low sulfur coal contract, a unit prefers to use low sulfur

spot market coal if it does not install a scrubber and high sulfur spot market coal if it

does install a scrubber because (MCA'" < PA e < MCA9).

If a unit has no low sulfur contract coal (Ci = 0), a unit uses the maximum amount

of high sulfur coal both with and without a scrubber, and the unit is indifferent to

installing a scrubber at allowance price PAS:

PLPS + SMAX + ps sMAX MS PAMAX + Ps P s,MAX (2 90)
?z + A'i C ihvih A' i ihih (2

(2-90) shows that the costs with and without a scrubber are equal for a given PA, and can

be rearranged to find an expression for Pi,:


Pz Ps(AAX ASMAX) (2 91)


Since a unit uses high sulfur coal both with and without a scrubber, a low sulfur coal

contract will change the coal use by a unit both with and without a scrubber and change

the indifference allowance price by some value e:


P + (P ) AMAX s sMAX pc PS VpsMAX +1ii (2 92)


(2-92) can be rearranged to find an expression for P,:


p i (PAj (A i I V -A MAX) + p MAX + FPsi (2-93)









The two expressions for PiF in (2-91) and (2-93) can be set equal to solve for the

value of e:

P S((AMIN AMAX + A7MAX _SMAX )+Ps SMAX ps sMAX
A i ~ t +A i ) ih 1 (2-94)
(AMIN AMAX)

By filling in for allowances and coal use, it is possible to determine the sign of e.


AfAX ( (Hs)(mn) -A AA D sH S+CS" S) n- An
--C

ASMAX ( (Sh)()(1- -A MAX D( c iHhAS+ A A S,rh)(m)(-ri) A7
ih Iihh
sMAX Di CzH, sMAX D, i C
^l -s sMAX
Hiil HI Hih
By filling into (2-94), the expression for e can be simplified to parameters for coal price,

sulfur content, heat content, and scrubber capture rate:


MA4 Aih MAX) H+ ri( -4 _H Sih Hc Jil c )


Di[mPf'( Si Si) (Pi f P ) (2-95)
Hfi Hi8 Hi8 HtJ
An interpretable form is derived by multiplying through by (AjIN AMAX) and dividing

through by mQ(, il ) and combining like terms:
Hih i
i ih i ih
Ais S, S, c (2 96)
e("1 MAX)i i + (DO- Hi C) PA h H (2-96)c
(Hh i H h 7i

( sa sh
( e '- 1AfMA) '- P r i + (D H }ICI)(PA MCA"'8) (2-97)
ih HT
From our initial assumption that a unit uses all high sulfur spot market coal if it does

not install a scrubber if it does not have a high sulfur coal contract, then PAS < MCA"'8.

The total amount of heat content from the contract coal (HicCic) is weakly less than the

total heat content needed to meet demand (Di). So the second term is weakly negative.
Meanwhile, the first term is negative because which means
Meanwhile, the first term is negative because (- < ) and ($ > ), which means
ff"l Tih T' h Tl' 1









e < 0 and an increase in the size of the high sulfur coal contract (C h) increases the

magnitude of e.

The second example in Table 2-8 uses data that reflects this case. Without a coal

contract, PA = -".-' :", which is less than MCA"'" -<.")7.90 and PA = $700.00, and

a unit will use high sulfur coal both with and without a scrubber installed. In this case,

a unit will install a scrubber because PA < PA = $700.00. Now consider a low sulfur

coal contract for 50'. of all coal use, which results in c = :".-' )9 and will increase the

indifference price above both MCA'" and PA to (PA c) = $873.34. With the low sulfur

coal contract, a unit will now not install a scrubber and prefers to use low sulfur spot

market coal.

In the third case, assume that a unit, both with and without a low sulfur coal

contract, prefers to use high sulfur coal with and without a scrubber because PA <

(Pf e) < MCA'8. Given that a unit has no low sulfur contract coal (C = 0), a unit is

indifferent to installing a scrubber at allowance price PS:


S+ PFAS SMAX I pscs,MAX rAS AMAX r+ s Cs,MAX (2 9)

(3-26) can be rearranged to find an expression for Pi, where the only change in costs are

the price of a scrubber and the change in a unit's net allowance position. PA is equal to

the average cost of reducing a unit of emissions from scrubber installation:

SP, = Ps(A AX AfMAX) (2-99)

A low sulfur coal contract will change the indifference allowance price by some value c

because it alters the costs from net allowance purchases and fuel costs:


P. + (PA c)MAX + Pish sAX+ PC (PS )A + PishMA P+c? (2-100)

(2-100) can be rearranged to find an expression for Pi,:


SP = (PS A)(A' MAX) (2-101)









The two expressions for PiF in (2-99) and (2-101) can be set equal to solve for the

value of e:
P~(AMAX AMAX + ASMAX AMAX)
( e = '- i i' (2-t10 2)
(AMAX ASMAX)

Filling in for allowances and coal use, it is possible to determine the sign of e.


AMAX D ) (Sih) () AD (DA-- eHs Sih + Sii) () A~
Hih ih


SMAX = )()(1-)-A MAX in the size of the coal contract () (-increases
iih iih
By filling into (2 102), the expression for e can be simplified to parameters for coal price,

sulfur content, heat content, and scrubber capture rate.


C e(A4" AiSMAX) = STnihHih ) (2103)


Since ( < t e < 0 and an increase in the size of the coal contract (i) increases

the magnitude of e.

The third example in Table 2-8 uses data reflective of delivered costs and coal

characteristics for a unit in Florida. A unit initially prefers to use high sulfur coal both

with and without a scrubber because (PA = 'i111.00) < (MCA"' = $3, 887.56). A unit will

prefer to install a scrubber in this example because the indifferent price is PS 1

A low sulfur coal contract for 50'. of coal use will increase the indifference price to

(Pjs e) = ".1 12.42, which will result in a unit not installing a scrubber.
Based on the above three cases, the sign of epsilon can be summarized in Proposition

6.

Proposition 6: Given a low sulfur coal contract, e < 0.

Given Proposition 3 and Proposition 6, the impact of a low sulfur coal contract is

derived in Proposition 7(a) and 7(b).
Proposition 7(a): Assuming ) and allowing for the scrubber choice...S
Proposition 7(a): Assuming (-> >-) and allowing for the scrubber choice...









(i) For the range of allowance prices PA > MCA'"8, a low sulfur coal contract will
,, ., /l.; increase excess demand.

(ii) For the range of allowance prices (Pf e) < PA < MCA'8 a low sulfur coal contract
will ;, '.it l,; decrease excess demand.

(iii) For the range of allowance prices PA' < PA < (PF e), a low sulfur coal contract will
,, .. I; increase excess demand.

(iv) For the range of allowance prices MCA"'8 < PA < PFA, a low sulfur coal contract will
.i,;, increase excess demand.

(v) For the range of allowance prices 0 < PA < MCA"', a low sulfur coal contract will
., l.; decrease excess demand.

Proof of Proposition 7(a):

(i) When a unit faces PA > MCA'A", a unit prefers to install a scrubber and use all
low sulfur coal. From Proposition 3(iii), given S > S a low sulfur coal contract
increases the minimum emissions level, which will weakly increase a unit's allowance
excess demand.

(ii) When a unit faces (Pf c) < PA < MCA' ", a unit prefers to install a scrubber and
use all high sulfur coal. From Proposition 3(i), a low sulfur coal contract decreases
the maximum emissions level, which will weakly decrease a unit's allowance excess
demand.

(iii) From Proposition 6, when a unit faces PA < PA < (PA e), a low sulfur coal
contract increases a unit's indifference allowance price of installing a scrubber above
the allowance price, which leads a unit to not install a scrubber where it initially
would have done so and increases a unit's emissions and a unit's excess demand.
SC S
From Proposition 3(iii), given the scrubber choice and S > 7 a low sulfur coal
contract will weakly increase a unit's emissions and excess demand. The combined
net effect is weakly positive and weakly increases excess demand.

(iv) When a unit faces MCAA'j < PA < Pf, a unit does not install a scrubber and prefers
to use low sulfur coal. From Proposition 3(iii), given > a low sulfur coal
contract increases a unit's minimum emissions level and excess demand.

(v) When a unit faces 0 < PA < MCA"'8, a unit does not install a scrubber and prefers
to use high sulfur coal. From Proposition 3(i), a low sulfur coal contract decreases a
unit's maximum emissions level and excess demand.E

Proposition 7(a) is shown graphically in Figure 2.8(i).
Proposition 7(b): Assuming that ( < and allowing for the scrubber choice...
Proposition 7(b): Assuming that (- < H) and allowing for the scrubber choice...









(i) For the range of allowance prices PA > MCA'"8, a low sulfur coal contract will
,, ., l.I; decrease excess demand.

(ii) For the range of allowance prices (Ps c) < PA < MCA'8 a low sulfur coal contract
will ;, '.it l,; decrease excess demand.

(iii) For the range of allowance prices PAs < PA < (PS e), a low sulfur coal contract
will increase excess demand ASMAX < AMIN and a ., l/; decrease excess demand if
ASMAX > AMIN

(iv) For the range of allowance prices MCAj'" < PA < PAS, a low sulfur coal contract will
a, '. I, decrease excess demand.

(v) For the range of allowance prices 0 < PA < MCA"', a low sulfur coal contract will
a, '. I, decrease excess demand.

Proof of Proposition 7(b):

(i) When a unit faces PA > MCA'8, a unit prefers to install a scrubber and use all
SC SS
low sulfur coal. From Proposition 3(ii), given S < T a low sulfur coal contract
decreases the minimum emissions level, which will weakly decrease a unit's allowance
excess demand.

(ii) When a unit faces (Ps c) < PA < MCAA'j, a unit prefers to install a scrubber and
use all high sulfur coal. From Proposition 3(i), a low sulfur coal contract decreases
the maximum emissions level, which will weakly decrease a unit's allowance excess
demand.

(iii) From Proposition 6, when a unit faces PS < PA < (PS e), a low sulfur coal
contract increases a unit's indifference allowance price of installing a scrubber
above the allowance price, which leads a unit to not install a scrubber where it
initially would have done so and increases a unit's emissions and a unit's excess
SC SS
demand. From Proposition 3(ii), given the scrubber choice and < a high
sulfur coal contract will weakly decrease a unit's emissions and excess demand. If
AMIN > AiMAX, the combined net effect of the countering shifts is weakly positive
and weakly increases excess demand. If AIN < AMAX, the combined net effect is
weakly negative and weakly decreases excess demand.

(iv) When a unit faces MCAO'" < PA < PA a unit does not install a scrubber and prefers
Sc SS
to use low sulfur coal. From Proposition 3(ii), given < HS, a low sulfur coal
contract decreases a unit's minimum emissions level and excess demand.

(v) When a unit faces 0 < PA < MCA"', a unit does not install a scrubber and prefers
to use high sulfur coal. From Proposition 3(i), a low sulfur coal contract decreases a
unit's maximum emissions level and excess demand.E









Proposition 7(b) is shown graphically in Figure 2.8(ii). As with a high sulfur contract,

some of the price ranges may not exist for a particular case. However, the remaining parts

of the propositions hold.

2.7 Possible Implications on the Allowance Market and Industry Compliance
Costs

Under Phase I of Title IV, there were many generating units facing high sulfur

coal contracts for at least a fraction of their total coal use. If a binding high sulfur coal

contract leads a unit to choose a suboptimal compliance choice, such as purchasing

additional permits or installing a scrubber instead of switching fuels, and results in

weakly higher compliance costs. A unit's suboptimal choices not only increases a unit's

compliance costs, but should also increase compliance costs for the industry as a whole.

As has been show in several examples, high sulfur coal contracts for a large fraction

(50-101i i.) of coal use can greatly reduce a unit's "indifference price" to installing a

scrubber. Some units under Phase I initially appear to have installed a scrubber when

it was not a unit's optimal compliance option, increasing a unit's compliance costs.

Additional scrubber installations should have resulted in greater emissions reduction,

which should simultaneously lower demand and increase supply of allowances as a unit

switches from a net demander to a net seller. In doing so, the equilibrium allowance

market price should be driven lower, which may explain the lower than expected allowance

prices realized during Phase I. Even though the allowance market price was lower than

expected, the inefficient unit compliance choices resulted in higher than expected total

industry compliance costs.

Under future CAIR regulation, a unit's compliance options may be restricted by

low sulfur coal contracts agreed upon during the 1990s to meet Title IV emissions

requirements. A unit may find installing a scrubber and using high sulfur coal to be it

best compliance option. However, low sulfur coal contracts may lead a unit to choose a

suboptimal compliance choice, such as switching fuels or installing a scrubber while using









low sulfur coal. Suboptimal compliance decisions will lead to higher compliance costs at

the unit-level and may lead to higher total industry compliance costs.

As has been show in several examples, low sulfur coal contracts for a large fraction

(50-101t i') of coal use can greatly increase a unit's "indifference price" to installing a

scrubber. Some units under CAIR may not install a scrubber when it is optimal for them

to do so, increasing a unit's compliance costs. Fewer scrubber installations would result

in greater emissions, which should simultaneously increase demand and decrease supply

of allowances as a unit would be a net demander instead of a net seller, and the allowance

market price should be driven higher than would be expected. In this case, a higher than

expected allowance market price would occur with higher than expected total industry

compliance costs.

2.8 Conclusions

This paper analytically derives the impacts that long-term fuel contract constraints

may have on a generating unit's compliance choices and compliance costs in meeting Sa2

emissions restrictions. There are five important results from this analysis.

First, given the scrubber choice, it is easy to determine how a coal contract will

impact a unit's excess demand for allowances. A high sulfur coal contract will weakly

increase a unit's excess demand while a low sulfur coal contract will weakly decrease a

unit's excess demand.

Second, some coal contracts may actually decrease a unit's total costs relative to

using only spot market coal while increasing a unit's compliance costs if a coal contract

allows a unit to lock in a lower price for a given type of coal. The coal contract will

restrict compliance choices, which may result in higher compliance costs. If the fuel cost

savings is greater than the increase in a unit's compliance costs, then the coal contract

lowers a unit's total costs. Since a unit only cares about its total costs, a unit's compliance

decisions do not necessarily minimize a unit's compliance costs.









Third, under certain conditions a unit's compliance costs will increase when a coal

contract results in a suboptimal combination of spot market coal use, allowance purchases,

and scrubber installation. A suboptimal combination may result from two situations:

(1) a coal contract alters a unit's compliance choice or (2) the compliance choice does

not change, but the coal under contract has a higher sulfur to heat content ratio than

the same type of coal available in the spot market, which increases a unit's allowance

purchases.

Fourth, coal contract constraints change the allowance price at which a unit will

prefer to install a scrubber. In the case of Phase I, a high sulfur coal contract may increase

a unit's "indifference price", which creates a greater incentive for a unit to scrub and

sell its extra allowances. This result may explain why some scrubbers were installed at

sub-optimal units while the allowance market price was much lower than expected during

Phase I. The opposite may occur under CAIR, where a low sulfur coal contract may

increase a unit's "indifference price" and lower the incentive for a unit to install a scrubber

even if it would be the optimal compliance choice. In either case, a suboptimal compliance

choice will be made and a unit's compliance costs will weakly increase.

Fifth, there is certainty how a coal contract will impact a unit's excess demand

for allowances even when the scrubber choice is considered for most allowance price

ranges. However, due to the discrete scrubber choice and the change in the "indifference

price" due to a coal contract, it is uncertain how a coal contract will alter a unit's excess

demand if the allowance market price falls in one particular price range. A high sulfur

coal contract will shift excess demand as derived in Proposition 2 except if the allowance

market price falls in the price range (P' e, P- ) and S >Sh If the allowance market

price falls in this range, there are two countering effects on excess demand, the decrease in

excess demand resulting from the scrubber installation and the increase in excess demand

from the higher sulfur content of high sulfur contract coal relative to high sulfur spot

market coal. A low sulfur coal contract will shift excess demand as derived in Proposition









3 except if the allowance market price falls in the price range (Ps, Ps e) and < .

If the allowance market price falls in this range, there are two countering effects on excess

demand, the increase in excess demand resulting from no scrubber installation and the

decrease in excess demand from the lower sulfur content of low sulfur contract coal relative

to low sulfur spot market coal.











Table 2-2. High Sulfur Coal Contract: Assumptions
Coal Type Price/mmBtu Hf Price/Ton
cQ $1.60 24 $38.40
Ch $1.30 24 $31.20
Ch (Ex. 1) $1.50 24 -:1, 10
Ch (Ex. 2) $1.20 24 -.- 0O

PA $300.00
A' 20,000 tons
Di 24,000,000 mmBtu


Table 2-3. High Sulfur Coal Contract: Results
Costs Compliance Total Costs
Costs (Ex. 1) (Ex. 1)
Unconstrained $4.62 million -2 million
Constrained $5.01 million $38.61 million
CI' ii,';, $390,000 $2.79 million


Allowance Use (Tons)
Minimum 11,400 tons
Maximum 38,000 tons
Constrained Min. 24,700 tons


Compliance
Costs (Ex. 2)
$4.62 million
$5.01 million
$390,000

MCA,
MCAf's
MCAf" (Ex. 1)
MCAf'" (Ex. 2)


Total Costs
(Ex. 2)
-< ". million
S;". ill million
-$810,000


$270.68
$90.23
- 1 ,1g.90


Table 2-4. Low Sulfur Coal Contract Examples: Assumptions
Coal Type Price/mmBtu HI Price/Ton SY
CQ $1.60 24 $38.40 0..',
Ch $1.30 24 $31.20 2.0'C
cQ (Ex. 1) $1.80 24 ;.20 0..'
CQ (Ex. 2) $1.50 24 ., 1)0 0.,' ,


$200.00
20,000 tons
24,000,000 mmBtu


Sif
0.i .' .
2.0' ,
2.0' ,
2.0' ,























o

-E C

00utC
* Co
0000r3~


o


007


n^ n
oo-o

O
S*0 e
00000


o





00
ee ee


0 e
a t





- 0 -O
0oo


CI O

Go-~


rO]
MM

v scv


M




E .2


2



























0 -4













coc
o | |

bc(O 030





















A Do C C O G O
oc
Am C 00














^ 0 bL 0 L0
Hl
cbc m O O



















o






cI

































_O OO
S- 1 0 1 o
4 --
00




























7 u -4 --
Sc0 0 0


'Or



























A >a c^oncc 3















1171
oi









(-1L D ^ i- 4-















02 0
















117






































u



o
Os

o o

o
i
0 ^u


a

O


0

OS
o

0
60

g^
1 0

ao ;





ic,

Ood


o
9
iz^
^ ^

OS -f


as

I t



fi







lu 0oo
*--' e/s e


01c


0


S00



0m aK
i0 CEN
^-i E ^
\E CM O


C
0


: c


C
0
U
O

a


0


U
8 ^

0 <
o- '^


0







o
2
0-





CO
a
0


0 ^
n o


C
0
C
0.
o
a
01

f-
If- v
9 S


U
0


: c
m
O


as


C
aa o
0

C o

z c 0
92 02



-Of t


01



izs


00C

,e
aCO
10 CM
CO ^
lit 01
*,o Or


C10 C

zM 0
\E a

lz E lz


C
O


01
u
u
E-
E6


60 d
CO C
O O
a a
c; -^ : c

C o a
OmmQ
o u o


o
0
a


ma



-S
KC
0.











$1,600.00


$1,400.00

$1,200.00
o
I- $1,000.00 -



$600.00
0
oa
$ 4L')0 .00)


$200.00 JO

1....01
Jan-95 Aug-96 Apr-98 Dec-99 Jul-01 Mar-03 Nov-04 Jul-06


Figure 2-1. The SO2 Allowance Price. Data was given to me by Dallas Burtraw of
Resources for the Future


Excess Demand ColTespondence (i)


MCA049


o .MAX


AMW


NET SELLER NET BUYER


COMPLIANCE COST SAVINGS (ii


1$ 0ii


EXTRA
ALLOWANCE
SALES


I
FEWER
ALLOWANCE
PURCHASES


NETSELLER NETBUYER
NET SELLER NET BUYER


Figure 2-2. Excess Demand Correspondence and Compliance Cost Savings from Fuel
Switching Over Allowance Purchasing

























.-


0 j .. .- -1 -ri
4!^


NET SELLER NET BUYER


(ii)


I *


-L _______________ 4--I--b


MIN 1 VL
4 4


M4X-I


NET SELLER NET BUYER


Figure 2-3. High Sulfur Contract: Shift in Minimum Excess Demand


NET REVENUE FROM
ALLOWANCE SALES




rr


.1,^^~\\\\ ,\\^


.i1-


COMPLIANCE COSTS
FROM SWITCHING FUELS


MAX
4i


Figure 2-4. No Contract: Compliance Costs


MCA,
DIFFERENCE IN
FUEL COST
DUE TO
CONTRACT
AMN -uMI


COMPLIANCE COSTS
FROM FUEL SWITCHING ;Mc,.S ,
COMPLIANCE
COST
SAVINGS 4,, -
FROM LOWER
CONTRACT DIFFERENCEIN

CONTRACT
A0 .M 4 AW
;r1 -1" -1 1


COMPLIANCE COSTS
ROM FUE SWITCHING
COMPLIANCE
IR THINGS
COST FROM
HIGHER
CONTRACT
PRICE


MAX A_


Figure 2-5. High Sulfur Contract: Compliance and Total Costs


(i)


P .--- ---- ------------


q~iN"" .i


P

CA'


MCa-S

MaQ S -


!


JAI
............................
. . . . . . . . .
..................................
1 1 . . . . . . . . .
................ ...................
................. .................
.............................. .... ... .....
. . . . . . .
A
................
.............
................
... 1.0













DIFFERENT E IN COMPLIANCE (i
COSTS E TO CONTRACT

ALLOWANCE SALES
COMPLIANCE COSTS
FROM FUEL SWITCHING


\
DIFFERENCE IN
FUEL COSTS DUE
TO CONTRACT



4Mr


MC42',

P,


MCAS3,s


S---DIFFER NCE IN FUEL
COSTS DU : TO CONTRACT

NET REVENUE FROM
o AsrE -AS T r -"
COMPLIANCE COSTS FROM
FUEL SWITCHING


MAT


Figure 2-6. High Sulfur Contract: Relative Savings from Contract Coal


USING ALLOWANCE
ALLOCATION
t^


ADDITIONAL
ALLOWANCE PURCHASES
/


Figure 2-7. Cost Savings from Using Allowances Over Fuel Switching


MCs, c

MCAss


AMW ArM 0 AMAY

NET SELLER NET BUYER


Figure 2-8. Low Sulfur Contract


MC4S, -

MCAf6s^


--b


- J I


I I I IU


NET SELLER NET BUYER


P^
C,4 -

MCASs


' _Jc4SS


AMW
A-


4MAX


(ii)


_._ -




















MCAcC


Fi 0


Figure 2-9. No Contract: Compliance Costs


4-


DIFFERENCE IN
COSTSDUE TO(
COMPLIANCE
COST S FROM
ALLOWANCE
PURCHASK4

^Mf ,,M


$ (i)
DIFFERENCE IN TOTAL
COSTS DUE TO CONTRACT
___-____ ___


MCA C


;1MAX 4MMY


DIFFERENCE IN
COSTS DUE TO
COMPLIANCE
COST S FROM
ALLOWANCE
PURCHASE*-


4S iB


(ii)
DIFFERENCE IN T OTAL
COSTS DUE TO CONTRACT
\


0 AAX Y MAX
IAI i


Figure 2-10. Low Sulfur Coal Contract: MCA"'


4-


(1)
DIFFERENCE IN FUEL
COSTS DUE TO CONTRACT
I


|I I1


M. ---------------
" DIFFERENCE IN
COSTS DUE TO

P-- .---------------
A COMPLIANCE
COSTS FROM
ALLOWANCE,
PURCHASES
I!


PA

MC40


-41 W A4AY
0 -iM4^ ,|M


COMPLIANCE
ALLOWANCE


g)iN AMI


$ (ii)
DIFFERENCE IN COMPLIANCE
COSTS DUE TO CONTRACTS




COSI S FROM
PURCHASE S DIFFERENCE
IN FUEL COSS
DUE TO
CONTRACT


o q ARY


Figure 2-11. Low Sulfur Coal Contract: MCA"'





122


MCAcS


m m


















PF = MCAP

MCA' -



-AM7N 0




Figure 2-12. Compliance Costs: PA



p $
COMPLIANCE
FROM EITH
ALLOWANCE PUl
OR SWITCHING
P =MC ^S


COMPLIANCE COSTS
FROM EITHER
ALLOWANCE PURCHASES
OR SWITCHING FUELS


A MAXY
Z'


MCA8,8


4MN 4MMA 0 40 'N 0 ^M4X


Figure 2-13. High Sulfur Coal Contract


COMPLIANCE COSTS


$(i)
ADDITIONAL FUEL COSTS

min,>


0 \ AM

MASC
PA


MCA5,s --
PA =MCAs'S


COMPLIANCE
COSTS -

A^~ jMW


$(ii)
ADDITIONAL FUEL COSTS








0 AMr MA4X
4 (ii


Figure 2-14. Low Sulfur Coal Contract


MC4sc -
PA = MC4'-











$ (i) (ii)





McAps -MC^ ~
!MS ; S M


I I o I
A 0






PA









MCA',_
A


S2 SMExcess Dmd Crres: MA '

Figure 2-16. Excess Demand Correspondence: MCA"' > PS





















4ML ;4-fMAX N A4 s a


Figure 2-17. Impact of a High Sulfur Coal Contract: MCA'" < PS


GAfigss


AMCA7's -

PA


PA

MC4Ass



MCA -
A -


Figure 2-18. Impact of a High Sulfur Coal Contract: MCA > P



A (i

AI '- MCf -



ACA'- ,, MCA4s
II
IA


A"~


WMAX A'WX


Figure 2-19. Impact of a Low Sulfur Coal Contract: MCAi'" > PA


_ _,dA-x SMSX A- 4 uMMy









































I 4-






A
I p' -4
I -
l l
J,


0 4"MA 4A


M ss .






pss
IA


4SMh 4SMAX 4SMAX 4 Q


Figure 2-20. Impact of a low sulfur Coal Contract: MCAs < P


i- 4i MAM









CHAPTER 3
THE EFFECT OF FUEL CONTRACTING CONSTRAINTS ON SO2 TRADING
PROGRAM COMPLIANCE: EMPIRICAL EVIDENCE

3.1 Introduction

The U.S. SO2 Trading Program created by Title IV of the 1990 Clean Air Act

Amendment led to lower compliance costs than what would have occurred under a

Command-and-Control approach. However, all compliance cost savings were not realized

in the early years of the program. There have been several conjectures as to why the

hypothetical outcome was not obtained, including short-run rigidities from fuel contracts.

C'! lpter 2 shows how fuel contracts could alter a generating unit's compliance decision

in the U.S. SO2 Trading Program, but ignores any .I -: regate allowance market and

iiillli-1i v-.1-ide compliance cost impacts. This paper expands on C!i lpter 2 by looking at

the allowance market equilibrium impacts and total industry compliance costs from fuel

contracts through analytics and empirical modeling.

Given the scrubber choice, an allowance market equilibrium will exist. Allowing

for the scrubber choice makes it impossible to guarantee an equilibrium, but one still

may exist. Binding fuel contracts may lead to altered unit-level excess demands and,

in so doing, the allowance market price (PA). Meanwhile, binding contracts can alter

compliance decisions and increase total industry compliance costs.

This paper uses generating unit-level simulations to replicate results from previous

studies and show that short-run fuel contracts appear to explain a large portion of the

previously unexplained excess compliance costs found in previous simulations. Simulating

the least-cost compliance choices without including fuel contract constraints results in

minimum annual industry compliance costs of _'-- 3 million, which varies greatly from

the actual compliance costs of $1.30 billion found in these simulations. Once fuel contract

constraints are introduced into the simulation, the minimum annual industry compliance

costs become $1.01 billion. Based on these results, fuel contract constraints explain -i.'1.1

million, or 6 !'. of the excess compliance costs realized in the program for 1996.









Also, this paper considers a more appropriate level for compliance decisions. The

literature has only considered compliance costs at the generating unit level. However,

actual compliance decisions would be made at the plant level because of the economic

relations and the physical proximity between generating units. Coal is delivered at

the plant level, where multiple generating units may be located. Since a plant is

only concerned about minimizing its total costs at the plant, the minimization at a

particular generating unit is not necessary for a plant to make optimal choices. This paper

analytically analyzes the plant level decision-making process to show why it is appropriate

to consider decisions as the plant level instead of the generating unit level, the interaction

of choices for economically related generating units, and how cost-minimization at the

plant level may deviate from cost-minimization at the generating unit level.

The paper will be structured in the following manner: Section 2 will look at

the conditions for an allowance market equilibrium while Section 3 will look at the

comparative statics of the allowance market. Section 4 will look at compliance costs,

both for an individual generating unit and the entire industry. Section 5 and Section

6 will explain the data used in the generating unit level simulations and the results

from the actual simulations, respectively. Section 7 will analyze the plant level decision

making process, how it differs from the generating unit level model, and what additional

information the model will add to the literature.

3.2 Review of Generating Unit Model

3.2.1 Generating Unit Problem




mm zA- + P A. + PC + PAA + + + P + P zi,Ai,Cg,C ,C',C
subject to...A7 + A, > (1 z)(m)(CS + CS + CS + CS>) Ai (3-2)

(CSH + C H + CQH8 + CHI) > D Ai2 (3 3)

C> /4 (3-4)









Cl = C pil (3-5)

Ch, Ci > 0 (36)

zi {0,1} (3-7)

Equation (3-1) represents the unit's cost function. These costs include the cost of

scrubber installation (ziPi,), net costs of allowance purchases (PAAi), and costs of coal

purchases (P 'hC + P"jC[ + P. C. + PFCi). The emissions constraint is shown in (3-2),

where the number of allowances held (A + Ai) must be as large as the amount of total

emissions by the generating unit [(1 ziri)(m)(C[2S/i + C2Sc + C["ii + CjSj)]. Total

emissions is a function of the amount of each coal type used as well as the emissions

reduction due to a scrubber, if one is installed. The Lagrange multiplier on the emissions

constraint is represented by Ail. The demand constraint requires that the amount of heat

input to generate electricity (C[,H, + CfhHch + C[H,[ + C5Hj,) must cover the consumer

demand (Di) for electricity expressed as heat input, which is seen in (3-3). The Lagrange

multiplier on the demand constraint is represented by Ai2. Coal contract constraints

require the unit to use a specific amount of each contract coal type, Cc for high sulfur

coal in (3-4) and Ci for low sulfur coal in (3-5). A unit will use exactly the contracted

amount because (1) if the contract coal is more expensive than spot market coal, then a

unit will not want to use any more contract coal than is necessary and (2) if contract coal

is cheaper than spot market coal, the coal producer would prefer to sell any additional

non-contracted coal through the spot market. The Lagrange multiplier for each contract

constraint on each coal type is represented by /ih for high sulfur contract coal and /il for

low sulfur contract coal.

3.2.2 Optimal Compliance Choices

The following is a summary of a generating unit's compliance choices derived in

C'! lpter 2 both with and without fuel contracts.









Given the scrubber choice, a generating unit that does not face any coal contract

constraints will make its optimal choices based on the relationship of the allowance

price (PA) and the marginal cost of abatement from switching fuels from high sulfur

spot market coal to low sulfur spot market coal (MCAj'"). If PA > MCA"'S, then

it is cheaper to meet the emissions constraint by decreasing emissions by switching

fuels than purchasing allowances. A unit will use all low sulfur coal and purchase the

minimum amount of allowances, which will result in the following compliance choices:

ch = Cf1 = s,MAX Di, and A, AMI. If PA < MCA'", then it is cheaper

to meet the emissions constraint by purchasing allowances than decreasing emissions by

switching fuels. A unit will use all high sulfur coal and purchase the maximum amount

of allowances, which will result in the following compliance choices: Ch= C, Di,

C, 0, and Ai = AAX.

Introducing a high sulfur coal contract constraint will restrict spot market coal use by

Cc and will alter excess demand (Ai). The maximum amount of high sulfur spot market

coal use decreases from Di = MAX to Di C = ,AX The maximum amount

of low sulfur spot market coal decreases from Di CI,MAX to Di Cc = ,MAX

If PA > MCA"', a unit will use it's contract constrained maximum amount of low

sulfur spot market coal (Cs,MAX) and excess demand will increase from AMIN to AMIN.

If PA < MCA"'8, a unit will use it's contract constrained maximum amount of high

sulfur spot market coal (,MAX) and excess demand will shift from AMAX to MAX. If
Sh < S then AAX < MAX. If i> i then AMAX > MAX
H H S s H H i
ih i- h ih ih
Introducing a low sulfur coal contract constraint will restrict spot market coal use by

Ci and will alter excess demand (Ai). The maximum amount of high sulfur spot market

coal use decreases from Di = CsMAX to Di Cc = sMAX The maximum amount
,MAX MAX
of low sulfur spot market coal decreases from Di C"SM to Di Ci= ",' f

PA > MCA8', a unit will use it's contract constrained maximum amount of low sulfur

spot market coal (iMAX) and excess demand will shift from AMIN to AI. If < H









then AMIN < AMN. If S > then AmN > AN If PA < MCA8, a unit will
s i HiH He' i i
use it's contract constrained maximum amount of high sulfur coal (C 8MAX) and excess

demand will decrease from AMAX to AAX.

Allowing the scrubber choice to be endogenous results in more complex impacts

of coal contracts on compliance decisions. A generating unit that does not face any

coal contract constraints will make its optimal choices based on two conditions: (1) the

relationship of PA relative to the allowance price at which a unit is indifferent to installing

a scrubber (PA) and (2) the relationship of PA relative to MCA8'" given the scrubber

choice.

If PAS > PA, a unit will not install a scrubber because total costs to the generating

unit will be lower without installing a scrubber. The resulting coal use and excess demand

will be the same as described above: C, = D, Q = 0, and A AMAX if PA < MCA '

and CQ 0, CQ = Di, and A, = A^IN if PA > MCA'8 where MCA'" is the marginal

cost of abatement from switching fuels without a scrubber.

If PAS < PA, a unit will install a scrubber because total costs will be lower with a

scrubber installed. As explained in C'! lpter 2 the marginal cost of abatement with a

scrubber (MCAj'") will be much higher than the marginal cost of abatement without a

scrubber (MCA"'j). The resulting coal use and excess demand will now be CQ Di,

CQ = 0, and A =- AfA if PA < MCA'" and CQ 0, CQ DA, and A = Af if

PA > MCA".

The result is a combination of the two excess demand correspondences where at some

allowance price (PA) where a scrubber will be installed there is a large non-continuous

decrease in a unit's excess demand.

Introducing a high sulfur coal contract constraint will have the same direct impacts

as when scrubbers are given. Coal use will be restricted by Cc. The excess demand

correspondence will be impacted in the same manner except there will be four shifts

instead of two. Minimum excess demand with a scrubber (AfMIN) and without a scrubber









(AMIN) will increase to AfMIN and A jJN, respectively. Maximum excess demand with

a scrubber (ASMAX) and without a scrubber (A AX) will shift in the same direction,

which depends on the relationship of -h to A high sulfur coal contract creates a
Hih Hfh
greater incentive to install a scrubber because of the higher minimum excess demand, and

decreases the allowance price at which a unit will install a scrubber from PA to P e. For

this price range, a unit will install a scrubber where it would not have without the high

sulfur coal contract constraint.

Introducing a low sulfur coal contract constraint will have the same direct impacts

as when scrubbers are given. Coal use will be restricted by C-. The excess demand

correspondence will be impacted in the same manner except there will be four shifts

instead of two. Minimum excess demand with a scrubber (AfMIN) and without a scrubber

(AMIN) will shift in the same direction, which depends on the relationship of to .
ii l
Maximum excess demand with a scrubber (ASMAX) and without a scrubber (A AX)

will decrease to AMAX and AAX, respectively. A low sulfur coal contract decreases

the incentive to install a scrubber because of the lower maximum excess demand, and

increases the allowance price at which a unit will install a scrubber from PA to PA + e. For

this price range, a unit will no longer install a scrubber where it would have without the

low sulfur coal contract constraint.

In summary, coal contracts will restrict a unit's coal use options, which may alter a

unit's compliance choices and excess demand for allowances. Less freedom in compliance

decisions may result in an increase in a unit's compliance costs.

3.3 Allowance Market Equilibrium

An allowance market equilibrium will exist under certain conditions. First, assume

that there is perfect information and that all generating unit's are price takers in the

allowance market. The definition of an equilibrium for the allowance market is as follows:

Definition 5.8.1: An equilibrium for the allowance market is a price PJ > 0,

allowance excess demands (A~), spot market coal fuel purchases (C',>* C*), and scrubber









installation choice (z*) such that: (1) For each i c {1,..., n}, A, z*, CC, CO,* solve...


mm z Pi, + P1 A + P+ C+ + PICfl + P Ch+ PiCi (3-8)
zi,Ai,C0^,C'
A, i,0f i
subject to...A7 + A, > (1 z (r)(m)(C S i + C Si + CiSSi + C' Si) (3-9)

(C, H,A + ChHc + COHil + OCH ) > Di (3-10)

Cc > U, (3-11)

c > (3-12)

iC, Cil > 0 (3-13)

zi C {0,1} (3-14)

Ai is Unrestricted (3 15)

(2) The allowance market clears. At P1,


A < 0 and P A 0 (316)
i= 1 i= 1

Theorem 5.1: Assuming the scrubber choice (z*) as given, an equilibrium exists for

the allowance market.

From Kakutani's Fixed Point Theorem, a fixed point exists if the market excess

demand correspondence upper semi-continuous, compact-valued, and convex-valued.

Each generating unit's excess demand correspondences are upper semi-continuous,

compact-valued, and convex-valued, which implies that the market excess demand is also

upper semi-continuous, compact-valued, and convex-valued. Given these conditions there

is a fixed point, and therefore, an equilibrium exists. Note that there may be multiple

equilibria, but there is at least one equilibrium. The Proof of Theorem 1 is in Appendix B.

The existence of an allowance market equilibrium is no longer guaranteed once the

discrete scrubber choice is introduced into the model. The non-convex nature of the

scrubber decision variable makes it impossible to guarantee that an equilibrium exists

because the excess demand is no longer convex for the range (A SMAX, A^IN). In such









a case, there will either be positive or negative market excess demand. In either case, it

could be considered a "quasi-equilibrium" because it can be assumed that an allowance

broker will sell any excess demand or buy any excess supply of allowances.

3.4 Comparative Statics: Effects on the Allowance Market

3.4.1 Comparative Statics: Effect of Relative Fuel Cost on the Allowance
Market

Since a generating unit's excess demand for allowances relies on the relative costs

of purchasing allowances (PA) compared to reducing emissions through switching from

high sulfur to low sulfur coal (MCAj'"). If PA > MCA"'S, then it is cheaper to switch to

low sulfur coal. If PA < MCA"'S, then it is cheaper to purchase allowances. Therefore,

changing the parameters of the marginal cost of abatement may alter the excess demand

and, in turn, impact the allowance price.

Proposition 1: Assuming the scrubber choice, a higher fuel price differential for low

sulfur spot market coal relative to high sulfur spot market coal will weakly increase PA.

Proof of Proposition 1: From C'! lpter 2, the marginal cost of abatement is derived as

(3-17).
HS Hh
MCASS H-- -hs (3 17)


Consider the impact of an increase in the price of low sulfur spot market coal on

the MCAj'". An increase in the price difference between fuels per unit of heat from

S- ) toh (o ) where < increases the numerator of (3 17) and results
T1 Th T1 4h iT iT
in an increase in the marginal cost of abatement from MCAj'" to MCAj'" + 6 where
pTl P,
H,
6 = H-- -_-__. A higher MCA8'" makes the relative cost to the generating unit
ih
purchasing allowances lower, and will lead to a weak increase in a unit's excess demand.

Excess demand will increase if MCA'"8 < PA and MCA>'" + 6 > PA and not change if

MCA'" +6 < PA or MCA"'j > PA. Greater excess demand in the allowance market results

in an increase in the total allowance market demand and possibly a decrease in allowance

market supply, both of which will weakly increase PA.









3.4.2 Comparative Statics: Effect of Coal Contracts on the Allowance Market
Given the Scrubber Choice

Assuming the scrubber choice as given, each binding coal contract will have an

unambiguous impact on the allowance price. A binding high sulfur coal contract weakly

increases the allowance price, ceteris paribus. A binding low sulfur coal contract weakly

decreases the allowance price in the market, ceteris paribus.

3.4.2.1 Impact of high sulfur coal contract on allowance market

A generating unit's excess demand for allowances may be altered by a binding high

sulfur coal contract constraint. From C'! Ilpter 2, a binding high sulfur coal contract

will result in more high sulfur coal use than would be optimal for a generating unit and

increase a generating unit's excess demand. An increase in a generating unit's excess

demand may increase the allowance market price.

Proposition 2: Assuming the scrubber choice as given and ( ), a high sulfur
Hih ih
contract results in a weakly higher allowance price.

Proof of Proposition 2: As has already been shown in C'!i lpter 2, given the

scrubber choice and S- a high sulfur contract results in a weak increase in a
ih ih
generating unit's excess demand.

If PA < MCA"', a generating unit prefers to use high sulfur coal, create its maximum

possible emissions, and have an excess demand of AMAX. Assuming the same relative

sulfur content for spot and contract coal, a high sulfur coal contract will not change total

emissions or excess demand (A AX= AMAX) and PA remains unchanged.

If PA > MCA"', a generating unit prefers to switch fuels from high to low sulfur coal,

create its minimum possible emissions, and have an excess demand of A"IN. However, the

high sulfur coal contract would force greater than the minimum amount of emissions and

weakly increase excess demand to AIN. A higher excess demand will lead to a decrease

in the allowance market supply or both a decrease in allowance market supply and an

increase in the allowance market demand. In both cases PA weakly increases.









3.4.2.2 Impact of low sulfur coal contract on allowance market

A generating unit's excess demand for allowances may be altered by a binding

low sulfur coal contract constraint. From C'! Ilpter 2, a binding low sulfur coal contract

will result in more low sulfur coal use than would be optimal for a generating unit and

decrease a generating unit's excess demand. A decrease in a generating unit's excess

demand may decrease the allowance market price.

Proposition 3: Assuming the scrubber choice as given and H, a low sulfur

contract results in a weakly lower allowance price.

Proof of Proposition 3: As has already been shown in C'!i lpter 2, given the

scrubber choice and H H a low sulfur contract results in a weak decrease in a
H: H77T
generating unit's excess demand.

If PA > MCA"', a generating unit prefers to switch fuels from high to low sulfur coal,

create its minimum possible emissions, and have an excess demand of AIN. Assuming

the same relative sulfur content for spot and contract coal, a low sulfur coal contract will

not change total emissions or excess demand (AIN = AFIN) and PA remains unchanged.

If PA < MCA"', a generating unit prefers to use high sulfur coal, create its maximum

possible emissions, and have an excess demand of AMAX. However, the low sulfur coal

contract would force less than the maximum amount of emissions and weakly decrease

excess demand to AAX. A lower excess demand will lead to an increase in the allowance

market supply or both an increase in allowance market supply and a decrease in the

allowance market demand. In both cases PA weakly decreases.

3.4.3 Comparative Statics: Effect of Coal Contracts on the Allowance Market
with Endogenous Scrubber Choice

Previously when the scrubber choice was taken as given, the impact of coal contracts

on the allowance price was unambiguously non-negative or non-positive. High sulfur

contracts may only lead to an increase in the allowance price while low sulfur contracts









may lead to only a decrease in the allowance price. However, taking into account the

scrubber choice causes the sign of the impact on the allowance price to become ambiguous.

When you consider the scrubber choice in the decision-making process, it is uncertain

how a coal contract will affect the allowance market because the contract may increase

or decrease excess demand depending on the allowance price ranges derived in ('! Ilpter

2. Although it is certain that if the coal contract binds, then there will be a shift to a

sub-optimal excess demand for three different allowance price ranges.

3.4.3.1 High sulfur coal contract binds

Each of the three price ranges derived in C'!i lpter 2 must be discussed to understand

how a high sulfur coal contract will effect the allowance market, both in terms of the

allowance market supply, allowance market demand, and the allowance price. Assume

ih ih

If a generating unit has a high sulfur coal contract, the unit's excess demand increases

for two allowance price ranges: (PA > MCA7'8) and (P' e > PA > MCA"'").

The increase in excess demand will be from AfMIN to AI when (PA > MCA"'j).

The increase in excess demand will be from AIN to AIN or AIN to AAX when

(PX e > PA > MCA,'"), depending on the relationship between (PS c) and MCA"'.

It will be the latter in the special case where the generating unit ah--,v- prefers to use high

sulfur coal.

When the market allowance price is in these two ranges that result in an increase in

a generating unit's excess demand, there will be either an increase in the market demand

for allowances, or both an increase in market demand and a decrease in market supply for

allowances. In both situations, the allowance price will be driven higher.

The third price range that has a shift in the excess demand is (Pf e, PA). A

generating unit decreases its excess demand from AMIN to ASMAX, which is a result of

a unit installing a scrubber for a price range for which it initially would not install a

scrubber.









When the market allowance price is in this range that results in a decrease in a

generating unit's excess demand, there will be either an decrease in the market demand

for allowances, or both an decrease in market demand and a increase in market supply for

allowances. In both situations, the allowance price will be driven lower.

3.4.3.2 Low sulfur coal contract binds

Each of the three price ranges derived in C'i plter 2 must be discussed to understand

how a low sulfur coal contract will effect the allowance market, both in terms of the

allowance market supply, allowance market demand, and the allowance price. Assume
ss SC
H' H?
If a generating unit has a low sulfur coal contract, the unit's excess demand decreases

for two allowance price ranges: (MCA>'8 > PA > Pf + e) and (min{Pf, MCAf'"} >
PA). The decrease in excess demand will be from ASMA to AAX when (MCAS'" >
PA > Pf + ). The decrease in excess demand will be from AAX to ~AX when

(min{PI, MCA'"} > PA).

When the market allowance price is in these ranges that result in a decrease in a

generating unit's excess demand, there will be either an decrease in the market demand

for allowances, or both an decrease in market demand and a increase in market supply for

allowances. In both situations, PA will be driven lower.

The third price range that has a shift in the excess demand is (Pf, PA + e). A

generating unit increases its excess demand from either ASMAX to AMIN or ASMAX to
-AMAX depending on the relationship between (PA + e) and MCA'". This increase is

a result of a unit not installing a scrubber for a price range for it initially would have

installed a scrubber.

When the market allowance price is in this range that results in an increase in a

generating unit's excess demand, there will be either an increase in the market demand

for allowances, or both an increase in market demand and a decrease in market supply for

allowances. In both situations, PA will be driven higher.









3.5 Compliance Costs

Total compliance costs for a generating unit are the additional costs due to the

li. -' ,ri including costs from switching fuels, the costs from its net allowance position,

and scrubber installation costs (seen in (3-18)).


zPi, + PjA, + max{(PCfi, + PFCf Pi MAX), 0} (3-18)

Compliance costs may be positive or negative depending on its compliance decisions

and its initial allowance allocation. The scrubber installation costs are represented by

P,, and will only attribute to a unit's compliance costs if a scrubber is installed in

response to the program (zi = 1). The costs of a unit's net allowance position is the

difference between a generating unit's initial allowance allocation and its actual emissions

multiplied by the allowance price (PAAj). The costs of switching fuels is the larger of

two values: (1) total costs of actual coal purchases (PCihh + Pji C + PFhC + PFCi)

minus the costs of purchasing only high sulfur spot market coal given any contracted
-s s ,M A X '. c cc
coal (P CA M + PFhC + PFC?), or (2) zero. The latter will only occur if it is weakly

cheaper for the generating unit to use low sulfur coal without the emissions restrictions

( > H7). It is important to consider that the contracted coal will be used regardless
ih ii
of the program and will have no direct impact on compliance costs. However, a contract

could have an indirect impact by altering compliance decisions.

3.5.1 Compliance Costs with Coal Contracts Relative to Compliance Costs
from Previous Studies

Previous studies have assumed no restrictions on coal use, which results in a different

estimation of compliance costs. The optimal compliance choices in this case will not

account for any coal use restrictions. The constant "c" is the minimum costs for a

generating unit to meet its electricity demand if emissions and coal use are not restricted.


Pi + PA + PC + P C c (3-19)









A binding high (low) sulfur coal contract constraint leads to two possible cost

inefficiencies, but only one of which increases compliance costs. First, a contract could lead

to a generating unit using more than the optimal amount of high (low) sulfur coal and

less than the optimal low (high) sulfur coal. In this case, actual compliance decisions are

altered and may lead to a sub-optimal coal mix. Second, a contract may force a unit to

use some high (low) sulfur coal that is "more exp' .i-,i. than the high (low) sulfur spot

market coal price, which increases total costs while leaving compliance costs unchanged. In

previous studies, both of these cost inefficiencies were identified as excess compliance costs

because contract restrictions were ignored in the baseline costs without the program (c).


( z +P +P +P +PC c +PC- c) > (zP P A* +PC,* +P, C* c)

(3-20)

The minimum compliance cost choices with no contract constraint for previous studies

are denoted by "*" while the contract constrained minimum compliance cost choices are

denoted by "A" in (3-20). The assumed "c" is the total costs of meeting demand assuming

no contract constraint or emissions constraint. The cost difference between these two sets

of choices will be the combination of changes in compliance decisions and fuel costs. Some

of these :; -- compliance < n-I may not be compliance costs, which makes the derived

compliance costs too high. So it is important to include the contract constrained coal use

in the baseline "c" to get the appropriate compliance costs. Assuming that contracted coal

is at least as expensive as spot market coal, the compliance costs will be greater for the

contract constrained case as shown in (3-20) and as expressed in (3-21).


Sz( z4) + P(A, Af+P(Csf ( Csf) + P(C PCi*P) C + P c > 0 (3-21)


Consider a simple example where high sulfur contract coal is more expensive than

high sulfur spot market coal, low sulfur coal is preferred over high sulfur coal, and no

scrubber would be installed if a generating unit had total freedom in its coal use choices.

Now compare a generating unit's compliance costs to the compliance costs a unit would









face with a high sulfur coal contract for 10t' of its coal use. The change in compliance

costs seen in (3-22) include the additional costs from an increase in allowance purchases

and a decrease in coal costs from using cheaper, but dirtier fuel.

P,(AMAX AAMIN) ps ,MAX + pcX >0 (3 22)
i 'ilvil 'I ih

Since the contract coal would have to be used regardless of enactment of the program, the

additional costs for high sulfur contract coal use instead of high sulfur spot market coal

(PICMAx PC'jMAX) should not be included in compliance costs. However, previous

studies would recognize these fuel costs as compliance costs based on the assumed cost

minimization without contract constraints.


PizP + PIA, + P

The appropriate unit-level compliance costs will take the form shown in (3-23) where the

baseline total costs will take the value "i", which represents the costs of meeting electricity

demand given the coal contract constraint. These costs "F' will be different than "c"

because the spot market coal use will be altered if the contracted coal differs relative the

spot market coal it replaces in production.


PlCil" + PihCih + iC + P Pici I PgiCil' + PiChih (3-24)


3.5.2 Total Industry Compliance Cost

The total compliance costs to the industry are the sum of the combined scrubber

installation costs and fuel switching costs for all affected generating units. The total

industry costs from net allowance purchases are zero because each allowance sold by one

generating unit at PJ is purchased by another generating unit at PA. Assuming that high

sulfur coal is cheaper than low sulfur coal, then the total industry compliance costs are

(3-25).

S[z Pi + max{(PiCf + Ph C P CtMAX), 0}] (3-25)
i=i









The minimum unconstrained total industry compliance costs to the industry is

derived from the following problem. Each generating unit chooses its optimal A Ci >

0, C[* > 0, z* E {0, 1} based on the market equilibrium allowance price, P1. The

total compliance costs as shown in the objective function in (3-26) are the sum of the

compliance costs for each generating unit. Similar to the generating unit's problem in

C'!i lpter 2, there are two constraints that must be met. First, the sum of allowance

allocations must cover the sum of emissions that are emitted by all affected generating

units. Second, each generating unit must produce enough electricity to cover its electricity

demand requirement.


mmin z Pi + max{(P ClI + P Ch PCMAx), 0} (3-26)
zi,Ai',Ch'C i 1

s.t. (t zr) (Ci Si + CSi) (m) < Ae
i= 1 i= 1
(C$H C + CfHf) >D, Vie {1,...,n}

ze {0, 1} Vie {1,...,n}

If all generating units' cost-minimizing decisions result in an equilibrium in the

allowance market, then total industry compliance costs are minimized. We can prove this

by showing the first-order conditions for the minimized total industry compliance costs

match up with the first-order conditions from the generating unit's problem in ('! Ilpter 2,

and the optimal choices will be the same for both problems.

Solve for the first-order conditions given zi. For high sulfur coal:


Pih + A i(l zriTi)()(Sih) Ai2Hfh > 0, 0 if Cih > 0 (3-27)

For low sulfur coal:


Pai + A1(1 z r)(m)(Sjt) Ai2H > 0, 0 if Ca > 0 (3-28)









The first-order conditions are identical to those of the individual generating unit in

C'! lpter 2 when neither coal contract constraints bind. So a generating unit's cost

minimizing choices of C[*, C,* at PJ for each generating unit also minimize total industry

compliance costs at Pj.

These first-order conditions can be used to solve for Ail, the allowance shadow price,

which is also the same as for the generating unit problem. Since the compliance choices

that minimize the costs for all generating units result in the equilibrium P!, the allowance

choices also minimize the total industry costs.
PS" P5"
Ail MCA' --7Hl h Th (3-29)
(t Z r) (m)( |-h -| )
ih 7T

The same approach can be used under the iiili i-I i v--i.de compliance cost problem

with coal contract constraints to show that contract constrained minimum total industry

compliance cost will differ from the unconstrained costs.


n
+ (p p, s, MAX 0 1
z ~ P + max{(P + Phi -h Pfih CM )0
Zi,Ai,Gi,G^ T^
n n

i=i i=1
(C H, + Ch Hh + CizH + CiH,) > Di

Cih, Cl >_ 0

z e {0, 1}


The first-order conditions. For high sulfur spot market coal...


(3-30)


Vi e { ,...,n}

Vi e {1,..., n}

Vi e {1,..., n}


1i + Ail(1 ziri)(m )(S,')) Ai2H,8 > 0,= 0 if C67 > 0


(3-31)


For low sulfur spot market coal...


Pil + Ai1(t zi)(m)(Sf) 2Hf, > 0, 0 if C'i, > 0


(3-32)










The first-order conditions are identical to those of the individual generating unit in

C'!i pter 2.
ps ps
il ih
H H (333)
(t )iri)(m)( | -h )
Th i

These first-order conditions can be used to solve for Ail, the allowance shadow price,

which is also the same as for the individual generating unit problem. If a generating

unit uses high sulfur coal, PA < MCA"'8. If a generating unit uses low sulfur coal,

PA > MCA"'8. If a generating unit uses both high and low sulfur coal, PA = MCA'8.

So a generating unit's cost minimizing choices of z*, C>, Ci*, Af* for each generating unit

also minimize total industry compliance costs. If the contract constrained compliance

choices that minimize the costs for a generating unit results in an equilibrium P!, the

choices also minimize the contract constrained total industry costs. Note that the optimal

parameter combination for the contract constrained case will not be the same as under the

unconstrained case.

It has already been shown that any binding contract constraint for an individual

generating unit will result in a sub-optimal combination of coal use, which will weakly

increase compliance costs for the contract constrained generating unit. Any sub-optimal

choices made by one generating unit will weakly increase the costs for the entire industry.

3.5.3 Impact of Allowance Allocation on Compliance Costs Given Scrubber
the Choice

There are several studies that have considered the impact of the allowance allocation

distribution on cost efficiencies of allowance trading systems. Montgomery (1972) showed

that the allowance allocation distribution does not affect compliance costs, but it does

not consider the possible effects of contract constraints or PUC regulation. Stavins (1995)

has shown that transaction costs in allowance trading markets with continuous marginal

abatement costs can cause the initial allowance allocation distribution to matter for

efficiency. Montero (1997) extends Stavins (1995) by introducing non-continuous marginal









abatement costs through discrete technology choices and introducing uncertainty into

the Stavins' model. Montero finds that in an allowance trading system with transaction

costs, non-continuous marginal abatement costs resulting from discrete technology choices

can cause the initial allocation of allowances to matter for efficiency, even with constant

marginal costs of abatement and certainty.

The importance of the allowance allocation in these previous studies relies on the

existence of transaction costs. However, this model has assumed no allowance transaction

costs, which allows it to show that even with coal contract constraints the allowance

allocation distribution will not impact a generating unit's compliance choices or the total

industry compliance costs.1

Consider the first-order conditions for both high sulfur and low sulfur coal, which are

independent of a generating unit's allowance allocation. The type of coal a unit will use

will not depend on Ae.


Pih + AiI(l ziri)(m)(Si%) Ai2H > 0 (3-34)

Pi4 + Ai(li zjrj)(m)(Sjf) Ai2H^I > 0 (3-35)

The choice between switching fuels or purchasing allowances is based solely on a

generating unit's relative marginal cost of purchasing the next allowance compared to the

effective marginal cost of abating the next unit of emissions. As can be seen, Ae has no

impact on a generating unit's compliance decisions. A generating unit will use allowances



1 It is assumed that generating units are unable to break their contracts. Generating
units could realistically break their contracts at a very high price and create the additional
freedom in its compliance choices. In such a case, contract constraints would result in very
high transaction costs for trading those additional allowances.









when PJ < MCA"'8 and switch fuels when PJ > MCA"'8.
p pF


ib ii
P | MCh'A H (3-36)



3.5.4 Impact of Allowance Allocation with Endogenous Scrubber Choice

Consider the expression for the allowance price at which a generating unit is

indifferent to installing a scrubber (PF), which has already been derived in C'i .Iter

2.
Pi, + Pih Cih Cih) + PilCUil Cil) ( gS
(A, Aj)
It is possible to fill in for Ai = (Ei A') since excess demand is the difference between

actual emissions and initial allowance allocation to see how Ae will impact PA. In the

denominator, Ae and -Ae cancel out, leaving an expression for PAS that is unaffected by

Aj. So a generating unit's scrubber choice will be made independent of Aj.

Pi, + Pih(Cih Cih) + Pil ( Cil)
-< PA (3-37)
(E? + Ae- ?, A<)

Now consider the compliance choices made after the scrubber installation decision.

It was shown in ('! Ilpter 2 that a generating unit's compliance choices are based on the

relationship between the PA and MCAj'" or MCA '", depending on the unit's scrubber

choice, for a particular unit of coal. Similar to what was shown above, a generating unit's

MCA"'" and MCA"'" for a given unit of coal are not affected by A' because allowances are

not a parameter in the marginal cost of abatement.

Piz Pih



PHl Pih
MCA' Hil Hih
MC-jHih H,,h









3.6 Simulation Model


3.6.1 Introduction

It has been shown analytically that contracts may alter compliance choices by altering

coal use and scrubber installation and lead to greater compliance costs. These results may

be able to explain a large portion of the excess compliance costs found in previous studies.

This section will show how much of these excess compliance costs can be explained by

contract constraints.

The first portion of this section will look at the data used to parameterize the

simulation model, which includes a description of where the data was obtained, the

techniques used to create the parameters, and some issues regarding the data. The second

portion will look at the model design and approach. The last portion will summarize

the simulation results in terms of the total industry, individual states, and individual

generating units.

3.6.2 Data

All data used in the simulation, with an exception for the coal contract data,

were obtained from Dr. Paul Sotkiewicz. Using the same data allows for direct result

comparisons to determine the coal contract impacts. Dr. Sotkiewicz originally hand-collected

the the data from the EIA's "Cost and Quality of Fuels 1996," the EIA's "Electric Power

Annual 1997," and the EPA's "1996 Compliance Report."

Up to this point all generating units are assumed to be coal-fired units because it is

the primary fuel options for electricity generation in the U.S. However, there are 24 units

of the 431 units that used fuel oil or natural gas for electricity generation. The treatment

of these units is described in Section 6.2.1.

3.6.2.1 Fuel data

Fuel data on heat content, sulfur content, and delivered price were obtained from the

EIA's "Cost and Quality of Fuels 1996", which compiled information from FERC Form

423. The coal contract data was gathered directly from the FERC Form 423 database for









1996, which describes the characteristics of all fuel delivered to utility plants. Information

on the heat content, sulfur content, purchasing agreement, and delivered price for each

shipment are available, which can differ greatly from utility to utility and delivery to

delivery for each plant depending on location.

The heat content of delivered coal is in Btus of heat per ton of coal, barrel of oil,

or 1,000 cubic feet of natural gas. All fuel parameters are converted in terms of a heat

content baseline to allow for comparisons across units.

The purchasing agreement is labeled as two possible parameters: (1) spot market

and (2) contract purchases. An agreement is considered a spot market purchase if the

agreement is for less than 2 years. Purchasing agreements of two years or greater are

considered to be under contract. The amount of fuel under contract is in tons, barrels, or

1,000 cubic feet depending on the fuel. The data is manipulated into total heat content by

multiplying the heat content per unit of fuel by the amount of fuel delivered.

Since coal deliveries are made at the plant level, the size of the contract constraint is

not easily derived for each generating unit. The parameter for the amount of contract coal

for each generating unit was derived in the following manner. First, any unit at a given

plant that actually installed a scrubber is allocated as much high sulfur contract coal as

possible while receiving as little low sulfur contract coal as possible. The reasoning for this

is that a unit with a scrubber will have a larger marginal cost of abatement from switching

fuels (MCAj'" < MCAf'"). Second, any remainder is spread out evenly throughout the

remaining generating units without installed scrubbers. Third, if the equal distribution

results in too much coal for a unit to use in production, the excess coal will be shifted to

a unit that requires a greater total heat input. Fourth, if a plant's coal contracts result

in more heat input than was actually required in production, coal use is capped at the

heat input needed to meet demand. Any remaining coal is considered be remain stored for

future use.









Consider the following example for a plant with three generating units each requiring

200,000 mmBtu of heat input to cover their electricity demand and one unit has a

scrubber. Contracts account for 101' i. of coal purchasing agreements where there are

contracts for 300,000 mmBtu for both high sulfur contract coal and low sulfur contract

coal. Based on the distribution approach described above, Unit 1 will use as much of the

high sulfur coal as possible (200,000 mmBtu) and the remaining high sulfur coal will be

distributed equally among the remaining two units (50,000 mmBtu each for Unit 2 and

Unit 3). The low sulfur coal will be distributed to Unit 2 and Unit 3 equally because

neither has a scrubber installed (150,000 mmBtu each). Since contracts account for 101'.

of coal use, there is no need to purchase any coal on the spot market. If there had been

any excess coal, such as an extra 10,000 mmBtu of low sulfur contact coal, it would not

change the allocation in Table 3-1 and would be considered coal stored at the plant for use

the following year.

Sulfur content is the percentage of each ton of coal, barrel of oil, or 1,000 cubic

feet of natural gas that is sulfur. The data must be manipulated to create the desired

variable, which is pounds of sulfur dioxide per million Btus of heat. Phase I of Title IV

distributes allowance allocations based on 2.5 pounds of SO2 per million Btus of heat. So

fuel is considered "high sulfur" if contains greater then 2.5 lbs. of SO2 per mmBtu and

"low sulfur" if it contains less than 2.5 lbs. of SO2 per mmBtu. Any coal use that has a

higher (lower) sulfur content will increase (decrease) emissions above (below) its allocation

allows.2

Getting the data in terms of pounds of sulfur dioxide per million Btus of heat requires

an emissions factor, which is the amount of sulfur dioxide emissions that will result from

a unit of sulfur. Emissions factors were found in the EIA's "Electric Power Annual 1997".

By taking the sulfur content multiplied by the emissions factor divided by the millions of



2 Coal sulfur content can vary significantly within each category (high or low sulfur).









Btu per unit (ton, 1,000 barrels, or million cubic feet), it results in a parameter in pounds

of sulfur dioxide per million Btu (lbs. SO2/mmBtu).

Although nearly all generating units affected by Phase I were coal-fired units, 24 units

use fuel oil or natural gas for electricity generation. For these units, fuel oil is designated

as high sulfur fuel option while natural gas is designated as the lower sulfur fuel option for

non-coal fired units.

The delivered price is the cents per ton, barrel, or 1,000 cubic feet paid at the time

of delivery. The price includes the purchasing of the fuel and the transportation costs of

shipment. The price is converted to the price of fuel in cents per million Btus. A separate

delivered price is required for both high sulfur and low sulfur fuel, which is a weighted

average of all deliveries of each category of coal. So the high (low) sulfur price used in the

simulations is the weighted average delivered price for all high (low) sulfur coal deliveries.

The parameter used in the simulations is actually dollars per mmBtu, or cents/mmBtu

divided by 100.

Another issue is that not all generating units purchased both high sulfur and low

sulfur fuel. Proxies were required to determine the fuel characteristics facing each unit,

and were first taken from other generating units that were owned by the same company

where available. If a unit did not have another unit under the same company, a proxy

was taken from the geographically closest plant because shipping costs are an important

factor in determining the delivered price. Regional variation in prices will be much smaller

variation in prices across the U.S.

3.6.2.2 Allowance, actual emissions, and demand data

Data for allowance allocation, actual emissions, and electricity demand was found in

the EPA's "1996 Compliance Report". Electricity demand is measured in terms of the

heat input instead of actual megawatt-hours of electricity. For a given unit, there is a

given amount of heat input required to produce one unit of electricity. This linear nature

of electricity production per unit of heat input allows for this simple conversion. For









example, if it requires 1 mmBtu to create 1 MWh of electricity, a unit facing a demand

of 100 MWh would require 100 mmBtu of heat input. Allowance allocations and actual

emissions can be used directly without any manipulation. Actual emissions are used as

the initial allowance allocation allowance banking caused the amount of allowances used

during 1996 be less than the total amount of allocated allowances.

3.6.2.3 Technical generator and scrubber data

Generator nameplate capacity and heat rate data was found in the EIA's "Annual

Generator Data 1996". The heat rate is the amount of heat required to produce one

kilowatt-hour of electricity. Heat rate data unavailable in the 1996 report were gathered

from "Annual Generator Data 1992-1995".

Scrubber cost data was found in the EIA's "Electric Power Annual 1997". The data

was the historical scrubber cost data, including state-by-state average installed costs

per kilowatt of capacity, average operation and maintenance costs in terms of mills per

kilowatt-hour, megawatts of capacity for units with scrubbers installed, and sulfur removal

efficiency.3

The scrubber installation choice is multi-year decision. However, the simulation model

is a one-year model, which makes it necessary to annualize the costs of installing and

operating a scrubber. Scrubber installation costs will differ depending on the scrubber

technology and the size of the generating unit. Total scrubber costs are the discounted

annualized cost of scrubber installation and the operation and maintenance costs for the



3 Engineering-based estimates for scrubber installation and operation costs are an
alternative to historical-based estimates.









given electricity production.4 The chosen discount rate of ten percent and a twenty year

lifespan for capital equipment are the same as in Sotkiewicz and Holt (2005).5

3.6.3 Simulation Model Design

The simulation requires a mixed integer linear program because of the discrete

scrubber choice. The simulation model is coded in Matlab and the linear program is solved

using the open source LP Solve mixed integer linear programming solver. If the scrubber

choice is taken as given, the simulation becomes a simple linear programming model,

which guarantees an equilibrium.

An endogenous scrubber choice creates a non-convexity that may not result in

an allowance market equilibrium. In such a case, there will be either a positive or a

negative excess demand for allowances and an allowance price at which a generating unit

is indifferent to installing a scrubber. This could be considered a "quasi-equilibrium"

where the excess demand is either bought from or sold to an allowance broker at the

"quasi-equilibrium" allowance price. The technical explanation of the bisection iterative

process used to converge to an equilibrium allowance price is described in detail in

Appendix B.

3.6.4 Simulation Results

Seven different generating unit level simulations were run to determine compliance

costs under several sets of conditions depending on the assumptions on the emissions

constraint, contract constraint, and scrubber choice. These simulations are used to derive

the impacts on not only the industry, but also on individual states and generating units.



4 To annualize capital costs, the following equation is used to determine the present
value of the costs: P(+pl l where p is the discount rate and "t" is the number of years the
lifetime of the equipment.

5 The use of historical scrubber cost data will result in different results than engineering
cost estimates because the historical costs for scrubbers in 1996 dollars are cheaper than
engineering estimates for 316 out of 431 units.









3.6.4.1 Total industry costs and allowance market results

The simulation results are used to analyze three key issues: (1) replication of results

from previous studies, (2) the impact contract constraints have on the results from

previous studies, and (3) the "true" compliance costs resulting from the Title IV program.

Simulations 1, 3, and 4 recreate the results from previous studies while Simulations 2, 5,

and 6 introduce the contract constraints to the model. Simulation 7 derives industry costs

based on actual compliance choices.

First, consider the ability of the simulation models to recreate the results from

previous studies. Simulation 1 recreates the unconstrained cost-minimization results

from Sotkiewicz (2003) and Sotkiewicz and Holt (2005). There were 17 scrubbers that

were installed as a result of the New Source Performance Standards (NSPS) and were

unrelated to Title IV. Assuming these scrubbers as given, the total costs to the industry

of meeting electricity demand were $7.69 billion. These costs will be used as the baseline

total industry costs to determine total industry compliance costs comparable to previous

studies.

Simulation 3 recreates the emissions constrained results from Sotkiewicz and Holt

(2005) assuming the 46 scrubbers that were installed in 1996 as given. The simulation

results in an allowance price of $149.64, which exactly replicates the allowance price found

by Sotkiewicz (2003) and Sotkiewicz and Holt (2005). Total industry costs are $8.23

billion where 29 scrubbers were actually installed in response to Title IV. The difference

between Simulation 3 and Simulation 1 is the total industry compliance costs of meeting

the emissions constraint, which is $541 million and similar to the results found in previous









studies.6 Carlson et al. (2000) and Sotkiewicz and Holt (2005) found the least cost

outcome to be $571 million and -. '7 million, respectively.7

Simulation 4 recreates the emissions constrained results from Sotkiewicz (2003) that

allows a generating unit's scrubber choice to be endogenous. The model assumes the 17

scrubbers installed to meet the NSPS, but allows generating units to make the scrubber

installation choice. The simulation results in an allowance price of $155.58. Total industry

costs are $7.97 billion, where 27 scrubbers were installed in response to Title IV (42

scrubbers total). Allowing generating units to make their scrubber installation choice

results in two fewer installed scrubbers, which may explain the slightly higher PJ because

fewer scrubbers will decrease supply and increase demand for allowances. The difference

between Simulation 4 and Simulation 1 is the total industry compliance costs of meeting

the emissions constraint, which is lower than the costs found in Simulation 3 at --'%N

million. The results are comparable to the $340 million found in Sotkiewicz (2003). As

expected, freedom in the scrubber choice improves efficiency and lowers total industry

compliance costs by -"-'" million, or 47'. lower than in Simulation 3.

The allowance market does not clear in Simulation 4 because generating units are

allowed to make their scrubber choice. The discrete nature of scrubber installation leads

to a non-negative excess demand. However, the excess supply of 53,576 allowances in

Simulation 4 account for less than 1 of the 5+ million allowance market and could

be assumed to be banked for future use or sold to an allowance broker, which could be

considered a quasi-equilibrium.



6 The value slightly differs relative to the results from Sotkiewicz (2003) and Sotkiewicz
and Holt (2005) because the scrubber cost estimates in this study have higher estimated
variable operating costs.

7 Burtraw et al. (2005) and Sotkiewicz and Holt (2005)









Second, consider how fuel contract constraints will impact the total industry

compliance costs. Simulation 5 introduces fuel contract constraints to the the emissions

constrained model assuming the 46 scrubbers installed as in Simulation 3. The simulation

results in a higher allowance price of I_'il. 70, which could be the result of a greater

demand for allowances due to more high sulfur coal use resulting from high sulfur coal

contracts. Total industry costs are $8.76 billion, where 29 scrubbers were installed in

response to Title IV. Assuming these scrubber choices, the minimum compliance costs

relative to the unconstrained model in Simulation 1 are $1.07 billion, or $531 million

('-I'.) higher than if contracts are not taken into consideration.

Simulation 6 introduces fuel contract constraints to the the emissions constrained

model while allowing generating units to make their scrubber choice. The simulation

results in an allowance price of $210.74. Total industry costs are $8.63 billion where 44

scrubbers are installed in response to Title IV. The minimum compliance costs when

compared to Simulation 1 are $939 million, or -.'I,1 million higher than if contracts are not

taken into consideration. The minimum compliance costs are close to the actual cost found

in Sotkiewicz and Holt (2005) at $990 million and Carlson et al. (2000) at $910 million.

Allowing generating units to choose whether to install a scrubber allows the industry

to lower its total costs by $132 million. As would be expected, introducing the contract

constraint results in more scrubber installations due to Title IV from 27 to 44 because

high-sulfur fuel contracts increase the incentive for a constrained generating unit to install

a scrubber. As in Simulation 4, the allowance market does not clear due to the discrete,

endogenous scrubber choice. However, the excess demand of 7,797 allowances account for

less than 0."' of the 5+ million allowance market, and could be assumed to be bought

from an allowance broker.

The actual total industry compliance costs are found in Simulation 7. By using

the actual emissions and electricity production for each unit, it is possible to determine

each unit's actual coal mix. These actual decisions resulted in an allowance price range









of 8.'1.33-4'1.38 during 1996, total industry costs of $8.98 billion, and total industry

compliance costs of $1.30 billion relative to Simulation 1. If the contract constraints

are excluded from the model and the actual costs are compared to Simulation 3 and

Simulation 4, the excess compliance costs are $757 million and $1.01 billion, respectively.

However, including the contract constraint into the model results in excess compliance

costs of $226 million and :.8 million. Contract constraints explain -'. ;1 million and ,

million of excess compliance costs, respectively.

Third, the "true" compliance costs will differ from these estimates because the

appropriate baseline was not used. As shown analytically in Section 4, comparing

Simulation 5 and Simulation 6 to Simulation 1 is not the most appropriate measure

of compliance costs. The contract constraints should be included in both the baseline

simulation and the policy-restricted simulation. Simulation 2 runs the same model as

in Simulation 1 except that it includes contract constraints and results in total industry

costs of $8.27 billion. The difference between total industry costs in Simulation 1 and

Simulation 2 are the additional costs due to contract constraints, which are $582 million.

These additional costs would have resulted with or without the SO2 Trading Program

and should not be considered compliance costs. This is a key result because these costs

were labeled compliance costs by previous studies even though these costs are a result of

generating units locking in prices to protect from the uncertainty of higher coal prices in

the future.

By comparing results in Simulation 5 to Simulation 2, the minimum compliance costs

considering contracts and given the scrubber choice are found to be $490.1 million. "True"

minimum compliance costs are much lower ("1 10 million lower) once this additional

constraint is included in the model. In Simulation 6, the "true" minimum compliance costs

are 7- ;:..8 million, or $581.6 million less than if contracts are excluded from the model.

These are similar to the least-cost results found by Sotkiewicz (2003) at $340- 2-'7 million,

Sotkiewicz and Holt (2005) at $423-$553 million, and Carlson et al. (2000) at $571 million.









Although actual industry compliance costs are higher than the least-cost results at $716

million, these compliance costs are lower than the compliance costs found in Sotkiewicz

(2003) and Sotkiewicz and Holt (2005) at $990 million and Carlson et al. (2000) at $910

million. Contract constraints appear to explain some of the excess compliance costs found

in previous studies, which implies that generating units' decisions were more cost-effective

than previously stated in the literature.

3.6.4.2 Industry and generating unit coal use

By comparing coal use in Simulation 5 to that in Simulation 3, the impact of contract

constraints assuming scrubbers as given can be derived. Simulation 3 uses a total of

2,607,025,040 mmBtu of high sulfur coal (40.7' ) and 3,794,557,150 mmBtu of low sulfur

coal (59.;:'. ). Simulation 5 uses a total of 2,041,933,640 mmBtu of high sulfur coal (31.9' .)

and 4,359,648,550 mmBtu of low sulfur coal (68.1 .). Introducing contract constraints into

the model results in a" '. decrease in high sulfur coal relative to Simulation 3. Contract

constraints results in less high sulfur coal than would have otherwise been preferred.

Contract constraints led to 16 units (3.7'. of affected units) using a suboptimal coal

combination. 15 of the 16 units had 1(1C' of its coal use altered. Only 2 of these units had

an increase in high sulfur coal.

Simulation 4 uses a total of 2,651,410,732 mmBtu of high sulfur coal (41. !',) and

3,750,168,009 mmBtu of low sulfur coal (58..', ). Simulation 6 uses a total of 2,614,773,732

mmBtu of high sulfur coal (411 -',) and 3,786,804,109 mmBtu of low sulfur coal (59.2'.).

Including contract constraints into the model results in a 0..' decrease in high sulfur

coal relative to Simulation 4. Overall coal use does not appear to have been significantly

altered. However, this does not tell the whole story. Contract constraints led to 27 units

(6.;:'. of affected units) using a suboptimal coal combination. 5 of the 27 units had a

change of at least '- of coal use and 15 of the 27 had at least a 50', change in coal use.

14 of the 27 units had an increase in high sulfur coal use while 13 units had an decrease

of high sulfur coal use. The concern with the contracts is not necessarily that the entire









industry choices are shifted, but that individual generating units are not able to make

their cost-minimizing choices.

3.6.4.3 Generating unit scrubber installation choices

There are three issues to consider regarding generating units' scrubber choices: (1)

impacts from the endogenous scrubber choice without contract constraints, (2) impacts

from the endogenous scrubber choice with contract constraints, and (3) impact of contract

constraints on scrubber choices. First, comparing scrubber installations in Simulation 4 to

Simulation 3 will show how scrubber decisions would be altered if the scrubber decision

is made endogenous and contract constraints are excluded from the model. Allowing the

scrubber choice to be endogenous results in 48 generating units altering their scrubber

choice, including 25 scrubbers to be removed and 23 to be added for a total of 2 fewer

installed scrubbers. 21 generating units maintain the same scrubber choices, but 17 of

those were installed for NSPS. So only 4 scrubber choices remained the same.

Second, comparing scrubber installations in Simulation 6 to Simulation 5 will show

how scrubber decisions would be altered if the scrubber decision is made endogenous

and contract constraints are included in the model. Allowing the scrubber choice to be

endogenous results in 53 generating units altering their scrubber choice, including 19

scrubbers to be removed and 34 to be added for a total of 15 more installed scrubbers. 27

generating units maintain the same scrubber choices, but 17 of those were installed for

NSPS. So 10 scrubber installations remained the same.

Finally, it is important to consider the impacts contract constraints have on a

unit's endogenous scrubber installation decision, which can be determined by comparing

Simulation 6 to Simulation 4. 33 scrubber choices are altered as a result of contract

constraints, including 25 units that will now install a scrubber and 8 units that no longer

install a scrubber. Of the 25 units that chose to install a scrubber, 20 of them had a high

sulfur coal contract. Of the 8 units that chose to not install a scrubber, all 8 of them

had a low sulfur coal contract. So 28 of 33 scrubber choices appear to have been directly









impacts by contract constraints. As can be seen in Table 3-4, contract constraints increase

scrubber installations in Ohio, Alabama, Florida, Indiana, Mississippi, and Missouri and

decrease scrubber installations in West Virginia, New York, Pennsylvania, Wisconsin,

Georgia, Kentucky, and New Jersey.

An unexpected result in Simulation 6 is that scrubber installations actually increase

relative to Simulation 5. These are likely a result of the assumed scrubber cost estimates

because the historical scrubber cost data used in the simulations is lower than the

engineering cost estimates. Scrubbers appear cheaper to install than the engineering

estimates state. Of the 404 generating units with at least a 911'. removal rate, 31 historical

capital cost estimates are higher, 84 are the same, and 289 are lower than the estimated

engineering capital costs.8 The higher cost estimates are not much higher than the

engineering costs with a difference of $11/kW. However, the historical capital cost

estimates that are lower than the engineering capital cost estimates range from $50/kW

to $216/kW higher, which could have some significant impacts on scrubber installation

choices. For example, historical capital cost estimates for generating units in Mizzouri are

assumed to be $50/kW while the engineering cost estimates are assumed to be $266/kW,

or a 43 difference.

As can be seen in Table 3-5, there is a significant difference in scrubber installations.

The use of engineering cost estimates for scrubber installation results in a decrease in

scrubber installations from 44 to 25 in Simulation 4 and from 61 to 38 in Simulation 6

because of the higher costs involved. Notice that introducing contract constraints results

in additional scrubber installations from 25 to 38, which is close to the actual installations

of 46 scrubbers. Compliance cost interpretations remain similar to simulations using



8 27 of the generating units are not directly comparable because historical data (50'".
removal rate) assumes a different scrubber technology than the engineering data (9, '.
removal rate).









historical scrubber cost estimate data, with compliance costs slightly higher for all

simulations.

3.6.4.4 Impact of allowance allocation on the allowance market and compli-
ance costs

The impacts of altering the initial allowance allocation are as would be expected.

Fewer available allowances forces the industry to produce fewer emissions, which will

result in a higher allowance price, more scrubber installations, and higher total industry

compliance costs. A decrease in the initial allowance allocation by 1(1'. leads to an

increase in the number of installed scrubbers by 17 in Simulation 4 and 12 in Simulation

6, an increase in the equilibrium allowance price of 33.85 in Simulation 4 and $11.48 in

Simulation 6, and an increase in the total industry compliance costs increase by $93.6

million (:-':) in Simulation 4 and $118.8 million ( :'.,) in Simulation 6.9 Contract

constraints restrict compliance options, which will lead to fewer compliance choice changes,

which leads to less of an impact on both scrubber installations and the allowance market.

However, restricting compliance options also leads to the higher compliance costs to meet

the additional emissions reduction.

3.6.4.5 Summary of simulation results

Contract constraints appear to explain a large portion of the "excess" compliance

costs found in the previous studies. Much of these excess costs would have occurred with

or without the program due to the contract restrictions, and therefore are not actually

compliance costs. These results find that firms' compliance choices appear to be more

cost-effective than previously thought because previous work has ignored a key restriction

on a generating unit's freedom in compliance options.



9 The results are not reliant on how the 1(1'. reduction in initial allowance allocation
occurs because the distribution of the allocation will not alter compliance decisions.









The "true" compliance costs to the industry appear to be lower relative to the

previous compliance cost estimates. Actual compliance decisions resulted in total

industry compliance costs that were higher than the least cost choices given the contract

constraints ($226 million compared to Simulation 5 and :.S million compared to

Simulation 6).

3.7 Plant Level Decision-Making Process

3.7.1 Introduction

The generating unit model in C'! plter 2 does not account for the fact that two

or more generating units are often owned and operated by one firm at the same plant.

A plant's decision-making process may not minimize costs for each generating unit

because a plant's concern is based on the combined costs of all generating units under

its operational control. This model derives the plant level problem, which allows us to

analyze differences in high versus low sulfur coal, spot versus contract coal, allowance

excess demand, scrubber installation, and the positioning of contract coal use based on the

different generating unit characteristics at a particular plant.

A plant level decision-making model is more realistic than a generating unit level

model to determine compliance decisions because coal deliveries are made at the plant

level where there are often multiple generating units. All generating units at that plant

facing the same coal use options, including sulfur contents, heat contents, and delivered

prices for both spot market and contract coal. The contract constraints become more

complex in this model where the sum of contract coal use for all generating units at a

plant must cover the contract requirement, 7 1 Cf > Cf Vi E {1, 2, ..., }. A plant

with multiple units has greater degrees of freedom in it's choices as to what fuel types to

purchase, in what quantities, and at which unit to burn the fuel based on their emissions,

demand, and coal contract constraints. The improved freedom in choice variables should

lead to lower compliance costs.









Some anecdotal evidence of these cost savings is seen in a comparison of the 46

generating units that had scrubbers installed in 1996. Of those units, 37 units are each

located at a plant that did not operate a generating unit that did not have a scrubber

installed. The remaining 9 generating units were each located at a plant that operated

at least one other generating unit that did not install a scrubber. By comparing the

average excess compliance costs (in percentage terms), it is possible to see how having at

least one additional generating unit at a plant improves its abilities to lower compliance

costs. For the generating units located at a plant without an additional generating unit

without a scrubber had average excess compliance costs of 6 '' above the unit's minimum

compliance costs. Generating units at a plant with at least one additional generating unit

that did not install a scrubber had average excess compliance costs of 2 '. above the

unit's minimum compliance costs.

Three additional factors will be of great importance at the plant level. First, plants

are able to trade allowances between generating units as needed to cover emissions

at no additional costs. The literature has stated that trading allowances between its

units at the same plant has been a common occurrence, which supports the use of a

model that considers decisions at the plant level instead of the generating unit level.10

Second, another aspect to consider is the choice to install a scrubber, which is based on

the characteristics of all the units at a plant, not just the unit at which the scrubber

may be installed. Third, a plant's decisions become more complex if it has one or more

"unaffected" units.

A plant may have one or more generating units at its location that are not affected

by Phase I and does not have to meet any emissions requirements. Owning a non-affected



10 Montero (1996)










unit allows greater freedom in coal use and may greatly alter a plant's preferred

compliance options.11

Plants must consider the total combined costs over all generating units, which

may lead to plant level decisions that are contrary to a specific generating unit's

cost-minimizing choice. A plant may be able to lower total costs by increasing the costs at

one generating unit to save money at other units.

3.7.2 Plant-Level Problem

A plant solves the following problem. The subscript "i" represents a specific

generating unit for each operating plant while "n" represents the number of generating

units owned by the plant.


mm n
zACmin ,PCCC
i]1
subject to...A + Ai > (1


i= 1
n




'I > C
il
i= 1
zi E {0, 1}

Ch, ,'i > 0

i G {1,..., n}


The Lagrange multiplier on a generating unit's emissions constraint is represented by Ail.

The Lagrange multiplier on a generating unit's demand constraint is represented by Ai2.

Coal contract constraints require a plant to use a minimum amount of each contract coal




11 Due to the design of the policy with "compensation" provision, a plant can shift
coal use, but not electricity production to a non-affected unit without the unit being
incorporated into the program.









type, C- for high sulfur coal and CU for low sulfur coal. The amount of contract coal at

a generating unit is a choice variable because a plant is able to choose which generating

unit to use its allotted contract coal. The Lagrange multiplier of each contract coal type is

represented by Ph for high sulfur coal and p, for low sulfur coal.

3.7.3 First-Order Conditions

The partial derivative with respect to Ai yields the impact of a one unit change in its

net allowance position on the unit's total costs.


PA Ail 0 (3-46)


Since Ai can be either positive or negative based on the net allowance position, (3-46) will

hold with equality. The additional cost to the firm of emitting one more ton of emissions is

equivalent to the allowance price, Ai = PA, which is the same result as from the previous

models.

The partial derivative with respect to C[ represents the impact a one unit change in

high sulfur spot market coal has on the unit's total costs.


Ph + Ai(li zri)(m)(S)() Ai2H, > 0, 0 if Ch > 0 (3-47)


If the generating unit uses some amount of high sulfur spot market coal (C[ > 0), then

(3-47) holds with equality.

The partial derivative with respect to C1 represents the impact a one unit change in

low sulfur spot market coal has on the unit's total costs.


PI + Ail(l zzri)(m)(SS ) Ai2 H > 0, 0 if Ci > 0 (3-48)

Similar to high sulfur spot market coal, if the generating unit uses some low sulfur spot

market coal (Cu > 0), then (3-48) holds with equality.









The partial derivative with respect to C represents the impact a one unit change in

high sulfur contract coal has on the unit's total costs.


c + A(il ziri)(m)(S) Ai2h- > 0, 0 if C > 0 (3-49)


Unlike with high sulfur spot market coal, high sulfur contract coal will be impacted by the

high sulfur coal contract constraint (ph). If the generating unit uses some amount of high

sulfur contract coal (Cc > 0), then (3-49) holds with equality.

The partial derivative with respect to Cf represents the impact a one unit change in

low sulfur contract coal has on the unit's total costs.


Pf + A1(1 ziri)(rm)(S) Ai2fHi- 1 > 0, 0 if Cf, > 0 (3-50)

Unlike with low sulfur spot market coal, low sulfur contract coal will be impacted by

the low sulfur coal contract constraint (pi). Similar to high sulfur contract coal, if the

generating unit uses some low sulfur contract coal (C, > 0), then (3-50) holds with

equality.

3.7.4 Characterizing a Unit's Spot Market Fuel Choices

A plant's choice of fuel type for a given generating unit is not only based on the

plant's marginal cost of abatement relative to the allowance price, but also the plant's

scrubber installation choice for each individual unit. For this section, we assume that the

scrubber choice is given and focus solely on a generating unit's marginal cost of abatement

and the allowance price, excluding the use of contract coal. The interaction of choices

for the use of contract coal, the consideration of multiple generating units, and scrubber

choice will be discussed later. As in previous models, three cases must be considered:

a generating unit uses both high and low sulfur spot market coal, only high sulfur spot

market coal, and only low sulfur spot market coal.









3.7.4.1 Case 1: Necessary conditions for using both high sulfur and low sulfur
spot market coal

Assume that demand for each generating unit is given and a generating unit chooses

to use both high sulfur and low sulfur spot market coal (Cf, > 0 and C, > 0), which

implies that both (3-47) and (3-48) hold with equality. The additional costs to the

generating unit of using one more unit of low sulfur spot market coal is equal to the

additional costs of using one more unit of high sulfur spot market coal.


Ps + A(l r)(m ) -zh(S) -2H Pl + Ail(1 ziri)(n)(S/) i2H (3 51)

Since both (3-47) and (3-48) hold with equality, it is possible to find an expression

for (Ai2) in each equation and set the two expressions equal to each other and derive the

allowance shadow price (Ail) in (3-52).


= PA il MCA s (3 52)
(1- zir)(Tn)( 18 (3-52)
Hh VI

As discussed earlier, the allowance shadow price (Ail)is an expression for a generating

unit's Marginal Cost of Abatement from Switching from high sulfur spot market to low

sulfur spot market coal (MCA'"8). From (3-46), we also know that the actual allowance

price equals the allowance shadow price (PA Ail). As can be seen in (3-52), the

allowance price (PA) equals to the marginal cost of abatement (MCA"'8), which is the

increase in coal costs from switching from high sulfur spot to low sulfur spot market coal

per unit of reduced emissions.

The marginal costs of abatement may or may not differ across generating units. If a

plant does not install any scrubbers, then MCA'" = MCA7'" V{i,j} for all generating

units and the plant's choices will be identical to the generating units' decisions in Section

5. This can be seen in (3-52) where the only way that the marginal costs of abatement

can differ across generating units is through the scrubber choice and a scrubber's emissions

capture rate. If all generating units at a given plant both install a scrubber where ri = rj,









then MCA'" = MCA"'". Once again the plants' choices will be identical to the generating

units' choices made in Section 5.

Marginal costs of abatement will only differ across generating units at a plant if

(1) a plant installs a scrubber at some but not all of its generating units, or (2) a plant

installs a scrubber at all its generating units but the scrubbers have different capture rates

(ri / rj 3{i,j}). We will introduce the scrubber choice, and derive its impact on a plant's
choices later.

3.7.4.2 Case 2: Necessary conditions for only high sulfur spot market coal use

If a generating unit chooses to use only high sulfur spot market coal (Cj' > 0), then

(3-47) holds with equality and (3-48) holds with weak inequality resulting in (3-53).


Ps + Ai(l zri)(m)(S) AH < Pl + Ail ( zr) m)r(ST) AH (3-53)


Equation (3-53) states that additional costs to the generating unit of using one more

unit of high sulfur spot market coal is weakly less than or equal to the additional costs

of using one more unit of low sulfur spot market coal inclusive of emissions and demand

requirements.

Following Section 6.4.1 we can derive the relation of PA to MCA7'". By comparing

these two expressions and solving for Ai2, you get an inequality comparing Ail = PA and

MCA"'". Since the generating unit uses high sulfur spot market coal, the allowance price

is weakly less than the marginal cost of abatement (PA < MCA"'j) as shown in (3-54).
ps ps
SPA < MCA"s H-f (3-54)
(1 ziri) (m) ( )


3.7.4.3 Case 3: Necessary conditions for only low sulfur spot market coal use

Similarly to the above case, if a generating unit chooses to use only low sulfur spot

market coal (Cf > 0), then (3-47) holds with equality and (3-48) hold with weak

inequality resulting in (3-55). The additional costs to the generating unit of using one









more unit of low sulfur spot market coal is weakly less than or equal to the additional

costs of using one more unit of high sulfur spot market coal inclusive of emissions and

demand requirements.


P i + A,(l ziri)(m)(S) Ai2H > P8 + Ai(l zir)(m)(S) AX2H (3-55)

Once again we can derive the relation of PA to MCA'"8. By comparing these two

expressions and solving for Ai2, you get an inequality comparing Ai = PA and MCA'".

Since the generating unit uses low sulfur spot market coal, the allowance price is weakly

greater than the marginal cost of abatement (PA > MCA"'8) as shown in (3-56).
ps ,ps

= PA > MCA's -Hf (3-56)



3.7.5 Excess Demand Correspondence

When only considering the use of spot market coal (no contract constraints), a unit's

excess demand correspondence will look nearly identical to the correspondence in ('!i plter

2. The minimum and maximum excess demand for allowances can be derived for each of

the three cases described above in the same manner as in C'!i plter 2.

First, if a generating unit faces PA < MCA"'j it will use the maximum amount of

high sulfur spot market coal. The maximum amount of high sulfur spot market coal is

expressed in (3-57).

C8 MAX (3 57)
h ih

Replacing CQ in (3-41) with the expression in (3-57) for Ck',MAX gives an expression

for the maximum allowance excess demand in (3-58). A generating unit's maximum

excess demand must cover the difference between its initial allowance allocation (Ar)

and the amount of allowances needed to cover the unit's maximum actual emissions
[EMAX (1 X)(m)(SnZ)S)( )].


A AX = EAX Ae (358)









If a generating unit faces PA > MCA"'8, it will use the maximum amount of low

sulfur spot market coal, which can be found from (3-41). Assuming only low sulfur spot

market coal use to meet demand, the maximum amount of low sulfur coal is expressed in

(3-59).
S- (3-59)

Replacing C i in (3-41) with the expression in (3-59) for C~,M gives an expression
for the minimum allowance excess demand in (3-60). A generating unit's minimum

excess demand must cover the difference between its initial allowance allocation (Ar)

and the amount of allowances needed to cover the unit's minimum actual emissions

[EJMIN ziri) (m)(Sl) ( )]. If a unit's initial allocation can cover its minimum

possible emissions, then it will have a negative net allowance position and be a net buyer

seller of allowances.
AMIN EMIN A (3-60)

If PA = MCA"'8, a generating unit may use any combination of high sulfur spot

market coal and low sulfur spot market coal and lead to any level of excess demand in the

range (E MIN A', EMAX A). The allowance excess demand can be represented by

A = (pEMAX (1 p)EIN A) where the constant p [0,1]. A unit that is indifferent

between fuel switching and allowances purchasing could be either a net buyer or a net

seller.

Combining the excess demands for each of the three cases creates the excess demand

correspondence shown below.

A AX if PA > MCAj',
Ai = pAAX (1 p)AMIN if PA MCA' Vp e [0,1]

A^IN if PA < MCA7',

A generating unit's excess demand correspondence can be seen graphically in Chapter 2.

High sulfur coal use corresponds to the right-hand vertical line where PA < MCA"'8. Low









sulfur coal use corresponds to the left-hand vertical line where PA > MCA"'8. The case

where a generating unit uses some combination of low sulfur coal and high sulfur coal is

represented by the horizontal line at which PA = MCA"'.

3.7.6 Characterizing a Generating Unit's Contract Fuel Choices

Unlike in the generating unit-level model, contracted coal is not treated the same

as spot market coal. It is not possible to simply require a certain amount of high or

low sulfur contract coal use because the use of contract coal is a choice variable at the

generating unit level.

The use of contract coal will be based on the relative marginal costs of using contract

coal at each generating unit operated by a plant. A plant will choose to use contract coal

at the generating unit that will result in the lowest increase in the plant's total costs. To

make these cost comparisons, it is necessary to derive the marginal costs of abatement

for the given combination of spot and contract coal and compare them across generating

units. Initially we will ignore a plant's scrubber choice for each generating unit and a

plant makes the same choice for all its generating units (zi = zj = 0, or zi = zj = 1 and

ri = rj). Under these conditions, the marginal costs of abatement are identical across all

generating units, and it does not matter at which of these generating units the contract

coal is used. This will be shown separately for both a high sulfur and low sulfur coal

contract because the use of high sulfur contract coal is independent of low sulfur contract

coal and visa versa.

3.7.6.1 Case 1: Necessary conditions for high sulfur contract coal use at
Generating Unit "i"

We first consider a plant with a high sulfur coal contract. We can derive the marginal

cost of abatement of switching fuels from high sulfur contract coal to low sulfur spot

market coal in (3-61). The marginal cost of abatement from switching from high sulfur

contract coal to low sulfur spot market coal (MCAf'") will be the same for all generating

units that make the same scrubber installation choice (zi), including the same scrubber









technology with the same capture rate (ri).


Ph + Ail( iri) (m)(Sh) + Ai2H P =- Pl + Ail ( ziri) (m) (S) + Ai2HS
Ph
= Ail MCAc, + h S- SS (3-61)
(t )iri)(mT)(H I^ -
All units at which a plant does not install a scrubber have the same marginal cost

of abatement of switching from high sulfur contract coal to low sulfur spot market coal

(MCA'" = MCAS'"). Also all units at a plant that install a scrubber with the same

capture rate will have identical MCAc'".

3.7.6.2 Case 2: Necessary conditions for low sulfur contract coal use at
generating unit "i"

Second, derive the marginal cost of abatement of switching fuels from high sulfur

spot market coal to low sulfur contract coal in (3-62). The MCA"'c will be the same for

all generating units that make the same scrubber installation choice, including the same

scrubber technology with the same capture rate.


P + AI(1 zr)(m)(S) + A2H= Pc + Ai1( z)(m)(S) + A2H -
II1
= A A MCAs -( H_- (3-62)
(tl-S)(n)( S

All units at which a plant does not install a scrubber have the same marginal cost

of abatement of switching from high sulfur spot market coal to low sulfur contract coal

(MCA',c = MCA`'). Also all units at a plant that install a scrubber with the same

capture rate will have identical MCA'.

3.7.7 "Non-Affected" Generating Units at an "Affected" Plant

So far it has been assumed that all generating units operated by a plant are all

"affected" units, meaning that all units face an emissions compliance constraint as a result

of Phase I. However, there are 27 plants (out of 160 total plants) that have both i1!. i ., I"









and "non-affected"units.12 Now we will introduce generating units that are not "Phase I

affected" into a plant's decision-making process.

The emissions constraint does not bind (Ail = 0) for these units, which means that

"non-affected" generating units will prefer to use the coal type with the lowest delivered

price. For most plants this will be high sulfur coal, particularly those located in the

Eastern U.S.

3.7.7.1 Characterization of "non-affected" generating units at an "affected"
plant

The first-order conditions from Section 7.3 will no longer be dependent on the

emissions for a non-affected unit because the emissions constraint will never bind (Ail

0), as can be seen in (3-63) and (3-64).


OL
C P, \A H > 0, 0 if C, > 0 (363)

and
AL
Oct P' Ai2H > 0, 0 if Ci, > 0 (3-64)

Assuming that a generating unit uses both high and low sulfur spot market coal, we

use (3-63) and (3-64) to derive (3-65). If a generating unit uses both high and low sulfur

spot market coal, then the price per unit of heat input is equal for both coal types. A

plant will no longer make it's generating unit level fuel choices based on its marginal cost

of abatement because it no longer is concerned about abating emissions.


h 8 (3-65)
Hh8 H8

By using the first-order conditions, we can determine when a generating unit will only

use one type of spot market coal. If a generating unit uses only high sulfur spot market



12 Source: www.eia.doe.gov, "Existing Generating Units in the United States by State,
Co_,l_, iw,, and Plant, 2003" database.









coal, then (3-63) holds with equality and (3-64) remains weakly greater than zero and low
ps ps
sulfur coal is weakly more expensive (Hh < H). If a generating unit uses only low sulfur

spot market coal, then (3-64) holds with equality and (3-63) remains weakly greater than
p s s
zero and high sulfur spot market coal is more expensive ( > 1).

3.7.7.2 "Non-affected" generating units and high sulfur coal contracts

Now consider how these non-affected units will impact a plant's coal use choices for

its generating units. This is one issue that will only impact plants under Phase I because

only the dirtiest units are affected while under Phase II and CAIR all generating units

in the U.S. are affected units. There is only a concern with Phase I where most units

preferred to switch from high to low sulfur coal. So assume a plant faces a high sulfur coal

contract.

First, assume that two generating units (Unit "i" and Unit "j") are both affected

units. A plant will be indifferent to using high sulfur contract coal at either unit when the

first-order conditions for using high sulfur contract coal are equal across generating units.


PhC + AiA(1 z ri)(Tm)(SA) A2 P + A(1 zr)(m)(S) hj2H -


= Ai(l ziri)(mn)(S) A\i2H= Ai(l- zr)(m)(S) Aja2H (3-66)

Now assume that Unit "i" is an affected unit and its emissions constraint binds (Ail > 0)

while Unit "j" is not affected and the emissions constraint will not bind (Aj = 0). Under

this condition, the additional costs of using high sulfur contract coal at an affected unit is

greater than at a non-affected unit.


Ph + Ai(l ziri)(m)(S) ) Ai2H Ph > Ph Aj2H Ph


= Ai( ziri)()(SA) Ai2Hh > -Aj2H

== Ai(1 ziri)(Tm)(S) > 0 (3-67)









Consider a simple example where a plant operates two generating units, one that

is affected (Unit "i") and another that is not affected (Unit "j") by Phase I of Title IV.

Assume P < which is normally the case under Phase I.13 Furthermore, the plant
h I
prefers to switch from high sulfur coal to low sulfur coal to meet its emissions requirement

for Unit i because PA > MCA"'8. But the plant also has a high sulfur coal contract for

a small amount of high sulfur coal (Cc < ) that has identical characteristics to high

sulfur spot market coal. The plant must choose at which generating unit to use the high

sulfur contract coal. Since plant prefers to use low sulfur coal to lower its emissions at

Unit "i" and prefers to use high sulfur coal at Unit "j" to minimize its coal costs, then

the plant's contract will not bind and will not increase a plant's total costs. Alternatively,

if Unit "j" was an affected unit and PA > MCA"'7, a high sulfur contract would bind

and result in higher total costs for the plant. A non-affected unit will weakly decrease a

plant's total costs relative to if the unit WAS affected. High sulfur coal can be shifted to

the unaffected unit to relax the emissions constraint.

3.7.8 Scrubber Installation Choice

Up to this point, we have assumed a plant's scrubber choice for a generating unit

as given. Now consider a plant's scrubber installation choice for a given generating unit.

A plant chooses to use contract coal in different generating units depending on whether

it has installed a scrubber at the particular generating unit. The choice is based on the

scrubber choices because it is the only variable that can change the marginal cost of

abatement across a plant's generating units.

3.7.8.1 Marginal cost of abatement with and without a scrubber

Finding the allowance price at which a generating unit will install a scrubber is not as

simple as in the generating unit model because a plant must take into account the costs of



13 Over i .'-. of generating units had higher delivered prices for low sulfur coal than high
sulfur coal on the spot market.









all units when choosing to install a scrubber. The marginal cost of abatement for all coal

types has already been derived. Now the MCAi for each coal combination both with and

without a scrubber can be derived. The only way the marginal cost of abatement can vary

across generating units is through the scrubber choice (zi and ri).

First, consider the marginal cost of abatement of switching from high sulfur spot

market coal to low sulfur spot market coal (MCA7'") with a scrubber in (3-68), and

compare it to the marginal cost of abatement without a scrubber in (3-69). As in '!i lpter

2, a scrubber decreases the savings from switching spot market fuels because a scrubber

causes the emissions reduction to be smaller than without a scrubber.
P Ph'
h I
Ail MCAs- I h (3-68)
(tl- r )(T )( S1
P18 Ph,

AiR MCAs ? H h (3-69)
(mn)() )

Second, consider the marginal cost of abatement of switching from high sulfur

contract coal to low sulfur spot market coal (MCAA'") with a scrubber in (3-70), and

compare it to the marginal cost of abatement without a scrubber in (3-71). As in the case

of only spot market coal, installing a scrubber decreases the savings from switching fuels.

In the case of a high sulfur contract, it is less costly to a plant to use high sulfur coal at a

particular generating unit if it has a scrubber.
Ph
H
= Ail MCAS + H s (3 70)
(1- rt)(m)(' ITT

Ph
H
= Al = MCAs + sh (3-71)
h I
Third, consider the marginal cost of abatement of switching from high sulfur spot

market coal to low sulfur contract coal (MCA'c) with a scrubber in (3-72), and compare

it to the marginal cost of abatement without a scrubber in (3-73). As in the two cases

above, installing a scrubber decreases the savings from switching fuels. In the case of a low









sulfur contract, it is more costly to a plant to use low sulfur coal at a particular generating

unit if it has a scrubber.


SAi =- MCA'c- Hi (3-72)



SAil MCAc H (3-73)
h I
For each combination of coal types, installing a scrubber decreases the size of the

denominator, which increases MCAi. As in C'! lpter 2, a scrubber greatly decreases the

savings from switching fuels. A higher MCA, increases the range of allowance prices at

which a plant will prefer to purchase allowances instead of switching fuels by lowering the

price at which a generating unit will be indifferent between purchasing allowances and

switching from high sulfur to low sulfur coal. Any unit operated by a plant that has a

scrubber will have a larger marginal cost of abatement than any unit without a scrubber

(MCAi > MCAj).

3.7.9 A Plant's Preferred Order of Scrubber Installation

Given a plant chooses to install a scrubber, it will install the scrubber at the

generating unit at which it will get the greatest "bang-for-the-buck", which will be the

generating unit with the lowest average cost of abatement through scrubber installation

(ACAi). A generating unit will have the lowest ACAi if the cost of a scrubber per ton of
emissions reduction (ACAi -= P (sc s scs cc) is lower than all other generating
Tmri (ssgqs+Shq+sI Chc +sC )
units operated by a plant. Give the same coal use ratio (percent of coal use for each coal

category), the price per unit of heat or price per unit of reduced emissions only vary based

on total demand (D{ = Ch + Cih + Ci + Ch) because "m" and "r," are constants. In other

words, scrubber installation will be based on the average cost for scrubber installation per

unit of demand (ACAi = -).
Di /
The decision is based solely on a unit's ACAi because the marginal costs of

abatement are equal across generating units with no scrubbers installed at the plant.









Following along the same thought process, if a plant chooses to install a scrubber at two

generating units, the scrubbers will be installed at the generating units with the two

lowest ACAi. The above condition will only hold for affected generating units. A plant

will not install a scrubber at a non-affected generating unit because reducing emissions at

a non-affected unit does not relax any emissions constraints.14

3.7.9.1 At which generating units will a plant install a scrubber?

The order a plant will install scrubbers at its generating units and a plant's marginal

costs of abatement for both with and without a scrubber are known. Now it must be

defined when a plant will install a scrubber at a given generating unit by finding the

allowance price at which a plant is indifferent to installing a scrubber. This is tough to

analytically show because a plant has multiple choices to minimize its total costs through

scrubber installation.

Assume that all generating units operated by a plant are affected by Phase I, each

unit uses its cost minimizing combination of coal and allowances based on its scrubber

choice, and the units are sorted by ACA, from smallest to largest (ACA1 < ACA2 < ... <

ACA,_1 < ACA,). A plant's scrubber choices will be based on the relative total costs of

each possible combination of scrubber installation. A plant will not install a scrubber at

any generating units if its total costs are lower with no scrubbers installed than installing

a scrubber at the generating unit with the lowest ACAi (C(zi = 0, z2 = 0, ..., z = 0) <

C(zi = 1, 2 = 0,..., z = 0)) where C(*) is the total costs for a given scrubber choices

and optimal coal and allowance choices. We already know that any other combination of

scrubber installation must result in higher costs.



14 Some plants voluntary enrolled units into Phase I, which are labeled as substitution
units or compensation units. Although these units were not initially affected, they were
enrolled into the program and face the same types of requirements as the original units.
For this reason, they are considered affected units.









If total costs are lower with a scrubber installed at the generating unit with the lowest

ACAi than without a scrubber at any units (C(zi 1, z2 = 0, ..., z, = 0) < C(zl = 0, z2

0,..., z, = 0)) and lower than with a scrubber installed at the two generating units with

the lowest ACA, (C(zi = z2= 0,..., = 0) < C(z 1, z2 1,3 = 0,..., z, 0)), then

a scrubber is installed only at the generating unit with the lowest ACA,.

Generalizing this condition, a plant's decision to install a scrubber at generating

unit "m". A plant will install "m" scrubbers at the "m" largest generating units if

(C(zi = 1,Z2 1,...,Zm, 1,Zm+ = 0, ..., z, = 0) < C(Zi 1,z2 1,..., m- 1 1,Zm =
0,..., zn 0)) and (C(zi = 1,Z2 = ,..., 1,Zm+1 = 0,..., = 0) < C(zz = 1,Z2

1, ..., Zn+1 1, Znm+2 = ..., Zn = 0)).
3.7.9.2 At what allowance price will a plant install a scrubber at a given
generating unit?

Finding the allowance price at which a plant is indifferent to installing a scrubber

at a given generating unit is the allowance price at which the total costs to the plant are

the same both with and without the scrubber (PF). Solving for this value for the plant's

scrubber choice for each generating unit requires an assumption on the scrubber choices

of all other generating units. Luckily, the scrubber installation choices have already been

ordered above.

At first glance, it appears difficult to derive P' because a plant compares its total

costs over all of its generating units, and a plant must choose whether to install a scrubber

at each unit. However, it has been shown that a plant will base its scrubber installation

on a single factor, relative ACA,. There are several requirements for a plant to install

a scrubber at a given affected generating unit "i": (1) a plant installs a scrubber at all

affected generating units with a lower ACA, than unit "i", (2) it is cheaper for a plant

to install a scrubber than to not install a scrubber at unit "i". This requires a simple

comparison between the costs of installing a scrubber versus not installing a scrubber.

Let "X" represent the parameters without a scrubber installed, and "X" represent the









parameters with a scrubber installed. The costs for all generating units "j 7 i" will not

change and will cancel out, leaving the same expression for PAS as derived in C'! Ilpter 2.

i-1
(PI, + PAAJ + Ph jh + P I C+ h Ch + P c)
j=1
+P, + PAA, + Ph + PfC1 + PhCh + PC>
n
+ (PAAk + PCs + PjS + PCh + P 1)
k=i+1
i-1
(PJ + P + PJCsh + P + Ch + Pc)
j=1
+PAA, + P"Ch + P1C8 + PhCh + PlC
n
+ (PAA + PhCh + PCS + PhCh + PCs
k=i+1
SPi + PAA, + P" h + P1SC + PhCh + P/C

SPAA, + PhC + PC + PhCh + Po (3-74)


Assuming no contract coal, we can derive the allowance price at which a plant is

indifferent to installing a scrubber for a given generating unit in (3-75).

P i phSr +^ \(( ps(cr^ 's\
S + ih ) + l il il (3-75)
(Ai Ai)

A special case exists for (3-75) in which the generating unit is a non-affected unit, which

results in PA -- o because the unit does not have to cover its emissions with allowances

and gains nothing from installing a scrubber. An indifference price of infinity implies that

a scrubber will never be installed at Unit "i" if it an unaffected unit.

P s P1 + h( 01) + Pis(C ) F, + P(C h) + PT(C Q)
SPA OO
(Ai A) 0
(3-76)

3.7.9.3 Scrubber installation and high sulfur coal contracts

So far we have ignored any coal contract constraints. We must determine at which

generating unit(s) a plant will use contracted coal, which can be derived from the









following three conditions: (1) given a plant installs no scrubbers, a plant is indifferent to

using contracted coal at any generating unit because each unit has the same MCAi, (2)

the marginal costs of using high sulfur contract coal at a generating unit with a scrubber

are lower than using high sulfur contract coal a generating unit without a scrubber, and

(3) any generating unit with a scrubber will have a lower ACA, than a generating unit

without a scrubber. We can simplify a plant's contract coal use choices from these three

conditions by using any high sulfur contract coal at the generating unit with the lowest

ACA,, which is also the generating unit at which a plant will first install a scrubber. By

doing so, we can determine at which generating unit to use high sulfur contract coal before

the scrubber choice is actually made by the plant. If all high sulfur contract coal will not

cover the entire heat input needed to meet demand at the generating unit with the lowest

ACA, (HCMAX > HC ), then all high sulfur coal will be used at that unit. If there

is more high sulfur contract coal than can be used at the generating unit with the lowest

ACAi (HCMAX < HC), then the remainder will be used at the unit with the second

lowest ACA,. The same process will be used if there remains any additional contract coal

after meeting the heat input demand for the generating unit with the second lowest ACA,.

A plant with a binding high sulfur coal contract has a greater incentive to install

a scrubber at each of its generating units because it must use some high sulfur coal,

even if it would prefer to use low sulfur coal at all generating units. A high sulfur coal

contract decreases the indifference price at which a unit will install a scrubber from PA to

(PA e) for each generating unit. The incentive may not be large enough to result in a

scrubber installation at its generating unit with the lowest ACA,. In such as case, a plant

is indifferent to using high sulfur contract coal at any of its generating units because all

units have the same MCA 8.

A binding high sulfur coal contract results in a decrease in PA.

At PAS where...


Pi, + PAA, + Ph h + PlC = + PC + PlC, (3-77)









If high sulfur contract coal is at least as expensive as high sulfur spot market coal, we

know that...


Pi + PAA + + + < AA + hh + gP0C + PP K FAi + Ph h (3-78)

This is the same result as in C'! plter 2 where a high sulfur coal contract results in an

inefficient coal use combination of high and low sulfur coal. The indifference price at which

a plant will install a scrubber at a given generating unit will weakly decrease when a plant

chooses to use high sulfur contract coal at that unit.

3.7.9.4 Scrubber installation and low sulfur coal contracts

It can be determined at which generating unit(s) a plant will use low sulfur

contracted coal from the following three conditions: (1) given a plant installs no scrubbers,

a plant is indifferent to using contracted coal at any generating unit because each unit has

the same MCAi, (2) the marginal costs of using low sulfur contract coal at a generating

unit with a scrubber are higher than using low sulfur contract coal a generating unit

without a scrubber, and (3) any generating unit with a scrubber will have a lower ACAi

than a generating unit without a scrubber. We can simplify a plant's contract coal use

choices from these three conditions by using any low sulfur contract coal at the generating

unit with the highest ACAj, which is also the generating unit at which will be the last

that a plant will install a scrubber. By doing so, we can determine at which generating

unit to use low sulfur contract coal before the scrubber choice is actually made by the

plant. If all low sulfur contract coal will not cover the entire heat input needed to meet

demand at the generating unit with the highest ACAi (HCil, ,MAX > HCl), then all

low sulfur coal will be used at that unit. If there is more low sulfur contract coal than can

be used at the generating unit with the highest ACA, MAX < H ), then the

remainder will be used at the unit with the second highest ACA,. The same process will

be used if there remains any additional contract coal after meeting the heat input demand

for the generating unit with the second highest ACAi.









A plant with a binding low sulfur coal contract has a smaller incentive to install a

scrubber at each of its generating units because it must use some low sulfur coal, even if

it would prefer to use high sulfur coal at all generating units. A low sulfur coal contract

increases the indifference price at which a unit will install a scrubber from PA to (PA + c)

for each generating unit. A plant is indifferent to using low sulfur contract coal at any of

its generating units if no scrubbers are installed because all units have the same MCA"'.

A binding low sulfur coal contract results in an increase in PS.

At Pf where...


P, + PAA, + PQ +P + Pl PAA + PC + PlC (3-79)

If low sulfur contract coal is at least as expensive as low sulfur spot market coal, we know

that...


Pi + PAA + PhCh + PC4 + P1C > PAA, + Ph CJ + P8C4+/c V (3-80)

This is the same result as in C'i plter 2. A low sulfur coal contract results in an

inefficient coal use combination of high and low sulfur coal. The indifference price at which

a plant will install a scrubber at a given generating unit will weakly increase when a plant

chooses to use low sulfur contract coal at that unit.

3.7.10 Scrubber Installation Example: Plant with Two Affected Generating
Units

For a simple example, consider the scrubber installation choices for a plant with only

two generating units. Assume that a plant will install a scrubber at Unit 1 before it will

install a scrubber at Unit 2 because (ACA1 < ACA2). There are three cases that may

result, each of which is described below with their own indifferent price for installing a

scrubber.









3.7.10.1 Case 1: Install no scrubbers

In the case of a plant choosing not to install any scrubbers, two conditions must hold.

First, the total costs of not installing any scrubbers must be less than installing a scrubber

at the generating unit with the lowest ACA1.

PAA, + PS C + PASC1 + PC h + PAC11

+PAA2 + PCh + P + P + PC C2

< Pi + PAA, + P + PiSC 1 + PhC + P~/CT

+PAA2 + PC h + P 1 + + P C2

From this equation, we can solve for PA for this condition to hold in (3-81). The amount

of contract coal is the same no matter the scrubber choice at each generating unit made

by a plant.

s < Pi + Ph(C0h h) + P'(C0' CO(3
P(A A)1)
(A, A,)

Second, the total costs to the plant of not installing any scrubbers must be less than

the total costs of installing scrubbers at both generating units.


PAA1 + PhClh + PiC1 + PhClh + P C11,
+PAA2 + P"C'h + Pf"C' + P/C h + P C'C2

< Pi, + PAAj + PIC'h + P1C11 + PhCIh + PfCi P2

+PAA2 + PhC h + PCs2 + PhCCh + P 1C'C

From this equation, we can solve for P, for this condition to hold in (3-82).

Ps + P[(Clh Ch) + (Ch Cl)] + P2z + PI[{1- C 1) + (2 Cs )]
PA < -----(3-82)
(A A) + (A2 A2)

The minimum of the two PA' will be the allowance price at which a plant is indifferent

to installing a scrubber and not installing any scrubbers. Notice that the allowance price









for installing one scrubber will be weakly less than the allowance at which a plant would

install scrubbers at both it's generating units.

3.7.10.2 Case 2: Install one scrubber

We know that a plant will install a scrubber at the generating unit with the lowest

ACAi, which in this case is assumed to be Unit 1. For a plant to install one scrubber, two

conditions must hold. First, the total costs of installing one generating unit must be lower

than the total costs of installing no scrubbers.



Pi1 + PAA, + PCr + PC + PS + P + CI + AA2 + PSCs + PI C + PhIC


From this equation, we can solve for Ps for this condition to hold in (3-83).

Piz + P (Csh Ch) + P1 (C1o Ci)
A < A1 (383)

Second, the total costs of installing one generating unit must be lower than the total

costs of installing two scrubbers.



Piz + PAA1 + PhC' + PlC1 Ph + P/cCt

+PAA2 + P + P1 1 + P h + PI 1

SP + + PAAl + PhC + PC11 + PCh PC( + P2z

+PAA2 P + P1S + PhC C P'C

From this equation, we can solve for Pc for this condition to hold in (3-84).

D PTss PAs I ns( nds s_ Csl) S A r

(A2 A2)
The allowance price must be between these two indifference prices, which results in

PAS < PA for Unit 1 and PAS > PA for Unit 2.









3.7.10.3 Case 3: Install two scrubbers

For a plant to install a scrubber at both generating units, two conditions must hold.
First, the total costs of installing two generating units must be lower than the total costs
of installing one scrubber.



Pi, l PAA1+ PCi + P1 + C1 + PC(h + P CT1 + P,2

+PAA2 + PhC + PC1 + Pl C2 + Pl 1C

< + PAAI + P T + Pih + C + PhC h + P

+PAA2 + Ph + PIfCS + PC h + PC C02

From this equation, we can solve for PA for this condition to hold in (3-85).

D I psi cs _rs \sj Is s1s Cs S
P2 + rPhs(h ~h) + P1 W(2l C1) < PA (3-85)
(A2 A2)
Second, the total costs of installing a scrubber at both generating units must be lower
than the total costs of installing no scrubbers.



P, + PAAI + P l + PPCI + P Cih + P C1 + P2,

+PAA, + P"' P PCh 1 + PhC( + Pl FC

< PAA + PI'C;h + PfCl1 + PhCoh + PcC11

+PAA2 + P Ch + P1C1c + PC h + P C21

From this equation, we can solve for PA for this condition to hold in (3-86).

P1 + P (C'h CO) + P(Ci Ci) P2 + P(2h Ch) + P T Cs) Ps 86)
(A1 A) + (A2 A2)

The allowance price must be greater than both of these two indifference prices for two
scrubbers to be installed.









We can monotonically order these indifference prices at which a plant will be

indifferent to installing a scrubber in each of the above cases. A simple mathematical

example using data based on data from the Colbert generating units can be used to show

the monotonic nature of these indifference prices in Table 3-7.

A plant with no contract coal will prefer to install no scrubbers if the allowance price

is below $194.48. An allowance price between $194.48 and $230.89 will lead a plant to

install a scrubber at Unit 1, but not at Unit 2. A plant will install scrubbers at both units

if the allowance price is greater than $230.89.

Now consider the plant has a high sulfur coal contract for 28.92' of coal use

(13,855,269 mmBtu), which will alter the indifference at which a plant will install a

scrubber at Unit 1. We have already shown that high sulfur contract coal will be used at

the generating unit with the lowest ACAi, which is Unit 1. The required use of high sulfur

coal will lower the allowance indifferent price to installing a scrubber at Unit 1 to $140.69.

There will be no change in the indifference price to installing a scrubber at Unit 2 because

the high sulfur coal contract does not bind for that generating unit.

Now consider the plant has a high sulfur coal contract for all of its coal use in both

generating units (47,913,973 mmBtu). The indifference prices for both generating units

will both decrease as a result. The allowance price at which the plant will install a

scrubber at Unit 1 will again be $140.49. The indifference price for Unit 2 will decrease to

$153.61. A high sulfur coal contract for all of a generating unit's coal requirements results

in large reductions of the allowance price at which a plant will be indifferent to installing a

scrubber for both Unit 1 ($54 or 2-',) and Unit 2 ($77 or 3 ;',).

3.7.11 Scrubber Installation Example: Plant with One Affected and One
Non-Affected Generating Units

The P for the non-affected unit (Unit 2) will be infinity because there is no price at

which a plant would want to install a scrubber at a non-affected unit. The non-affected









unit will not buy or sell any allowances either since it has no emissions constraint. The

only possible choice is to install a scrubber at the affected unit (Unit 1).
Case 1: Install No Scrubber

The total costs of not installing a scrubber must be lower than the total costs of

installing a scrubber at the affected unit.

P4AA + PC'h + Pl Cl + PCh + P Cc + PhsC2 + P1 C2 + PJCjh + P C7l

< Pl + P4Ai + PhClh + PlfCPl + PhClh + P/C'l + PhAC + PtlCj2 + PCh + PIF l

From this equation, we can solve for P, for this condition to hold in (3-87).

ps 1 + Ph(C' Ch) + Pi'(C CO-)
(A1 A1)

Case 2: Install One Scrubber

The total costs of installing a scrubber must be lower than the total costs of not

installing a scrubber at the affected unit.

Pz + PAA t + CISlh + Pl + PFClh + P CT + P ChS + PFC + PhFCh + P CIl


From this equation, we can solve for PAs for this condition to hold in (3-88).

ps z + Ph(Ci C +) P(C'1 O-8))
P F > F Cit (3 88)
A (A1- A1)


3.7.12 Summary of Plant Level Results

A plant's decision-making process may not minimize costs for each generating unit

because a plant's concern is based on the combined costs of all generating units under its

operational control. The choice to install a scrubber is based on the characteristics of all

the units at a plant, not just the unit at which the scrubber may be installed. Once the









order of preferred scrubber installations is determined based on the ACAi, a plant is able

to make its cost minimizing fuel choices.

A plant level model is more realistic than a generating unit level model because coal

deliveries are made at the plant level where there are often multiple generating units.

Operating multiple units allows a plant to relax its contract constraint because there are

additional degrees of freedom in fuel use depending on each unit's emissions, demand, and

coal contract constraints. Also, plants are able to trade allowances between generating

units as needed to cover emissions at no additional cost. A plant may also have one

or more "non-affected" generating units, which face no emissions constraint. Any high

sulfur contract coal can be used at these non-affect units without any negative financial

repercussions due to the policy. These factors can lead to plant-level choices that do not

minimize each generating unit's total costs.

3.8 CONCLUSIONS

The U.S. SO2 Trading Program led to lower compliance costs than what would have

occurred under a command-and-control approach. However, all compliance cost savings

were not realized in part due to short-run fuel contract rigidities, particularly during

the first years of the program. This paper considers the allowance market equilibrium

impacts and total industry compliance costs from fuel contracts both through analytics

and empirical modeling.

An allowance market equilibrium will exist only if the discrete scrubber choice is

given. Allowing for an endogenous scrubber choice makes it impossible to guarantee an

equilibrium, although one may still may exist. Binding fuel contracts may alter a unit's

compliance decisions and excess demand. Altering compliance decisions could lead to both

an altered allowance market price and an increase total industry compliance costs.

Generating unit-level simulations were able to effectively replicate the results from

previous studies and show that fuel contracts can explain a portion of the previously

unexplained excess compliance costs. Simulating the least-cost compliance choices without









including fuel contract constraints results in minimum annual industry compliance costs of

--I' I million, which varies greatly from the actual compliance costs of $1.30 billion found

in these simulations. Once fuel contract constraints are introduced into the simulation,

the minimum annual industry compliance costs become $939 million-$1.07 billion. Based

on these results, fuel contract constraints appear to explain -.' ~1.1 million, or i-.'.- of the

excess compliance costs realized in the program for 1996. These impacts should slowly

decrease over time as firms adjust to the policy environment and binding contracts expire.

However, the impacts on compliance costs should linger for years due to the long length of

coal contract agreements. Also, contracts appear to remain prevalent in coal purchasing

agreements and could lead to some issues resulting from further SO2 emissions restrictions.

A plant's decision-making process may not minimize costs for each generating unit

because a plant's concern is based on the combined costs of all generating units under its

operational control. The choice to install a scrubber is based on the characteristics of all

the units at a plant, not just the unit at which the scrubber may be installed. Once the

order of preferred scrubber installations is determined based on the ACAi, a plant is able

to make its cost minimizing fuel choices.

A plant level model is more realistic than a generating unit level model because coal

deliveries are made at the plant level where there are often multiple generating units.

Operating multiple units allows a plant to relax its contract constraint because there are

additional degrees of freedom in fuel use depending on each unit's emissions, demand, and

coal contract constraints. Also, plants are able to trade allowances between generating

units as needed to cover emissions at no additional cost. A plant may also have one

or more "non-affected" generating units, which face no emissions constraint. Any high

sulfur contract coal can be used at these non-affect units without any negative financial

repercussions due to the policy. These factors can lead to plant-level choices that do not

minimize each generating unit's total costs.









Table 3-1. Example: Contract Coal Distribution
Plant-Level Constraint Total mmBtu
High Sulfur Contract 300,000
Low Sulfur Contract 300,000
Demand 600,000


Unit Demand
1 200,000
2 200,000
3 200,000


Coal Dist'n by Unit
Unit 1
Unit 2
Unit 3


High Sulfur Coal
200,000
50,000
50,000


Low Sulfur Coal
0
150,000
150,000


Table 3-2. Sulfur Conversion by Fuel Type


Fuel
Bituminous
Sub-bituminous
Anthracite
Lignite
Fuel Oil #2
Fuel Oil #6
Natural Gas


Emissions Conversion Factor
(38 Sulfur Content)/(mmBtu per ton)
(35 Sulfur Content)/(mmBtu per ton)
(39 Sulfur Content)/(mmBtu per ton)
(30 Sulfur Content)/(mmBtu per ton)
(144 Sulfur Content)/(mmBtu per 1,000 bbl.)
(162 Sulfur Content)/(mmBtu per 1,000 bbl.)
(0.60 lbs. S02/mmCF)/(mmBtu per mmCF)


Baseline
lbs. S02/mmBtu
bs. S02/mmBtu
bs. S02/mmBtu
bs. S02/mmBtu
l Ibs. SO2/mmBtu
l Ibs. SO2/mmBtu
l Ibs. S02/mmBtu


Scrubber
Yes
No
No














zzzz
< < <


Z6c


1-
C
Vn


Co

1-oL


Co
06~


-6
0rc


V
1-

CIA
Ge,


1- 1- Co V Co r Co
rr~~Co~


~
~ 0~0c~
0
0 ~

ooo0o0~


oC
0 H..
^ ^


OOW
UZl


SCl VC DC D






191


0
n o







c
S-&









. m
0 0
0o
UU..










Table 3-4. Impact of Contract Constraint on Scrubber C('!i. ..


State Decrease
WV -4
NY -2
PA -2
WI -2
GA -1
KY -1


3-5. Simulations
Ai PA


NA
NA
0
1,972
0
2,611
NA


NA
NA
$149.64
$214.83
$207.60
$238.55
NA


with Engineering Data
Scrubbers Industry Costs


Installed
17
17
46
25
46
38
46


$7,685,800,000
$8,267,400,000
$8,239,000,000
$8,010,200,000
$8,770,900,000
$8,686,300,000
$8,995,600,000


Comp. Costs
(vs. 1)
NA
NA
-".. : 1)0,000
$324,400,000
$1,085,100,000
$1,000,500,000
$1,309,800,000


Table 3-6. Impacts of a Reduction in the Allowance Allocation of 10'-.
Simulation 4 Initial Allocation Allocation Minus 10'.(
Industry Compliance Costs -2-',300,000 $381,900,000


Scrubbers Installed
Allowance Price


Simulation 6
Industry Compliance Costs
Scrubbers Installed
Allowance Price


44
$155.58


61
$189.43


Initial Allocation
. ;.7,800,000
61
$210.74


Allocation Minus 1(1'.
$476,600,000
73
$222.22


Comp. Costs
(vs. 2)
NA
NA
NA
NA
$503,500,000
$418,900,000
$728,200,000


Increase
$93,600,000
17
$33.85

Increase
$118,800,000
12
$11.48


Table 3-7. Math Example: Two Affected Units
Coal Type Price Sulfur Content
Low Sulfur Spot $1.20 i., -".'.
High Sulfur Spot $1.10 2 -,'.
High Sulfur Contract $1.15 2 -,,'.


Generating Unit
Unit 1
Unit 2


Piz
$2,719,320
$7,298,412


Heat Content
24 mmBtu
24 mmBtu
24 mmBtu

Demand (mmBtu)
13,855,269
34,058,704


Scrubber Installation
None
Unit 1
Unit 2


P : No Contract
< $194.48
($194.48, $230.89)
$230.89


P': Unit 1 Demand
< $140.69
($140.69, $230.89)
$230.89


P': Unit 1 and Unit 2 Demand
< $140.69
($140.69, $153.61)
$153.61


Increase


State
MO
MS
IN
FL
AL
OH


Table
Sim.

1
2
3
4
5
6
7

















MCAss


S


SMIN 0 A M



NET SELLER NET BUYER

Figure 3-1. Excess Demand Correspondence


MCI' f


AMIN "'V-'


(i)


4


4-


0 A -l i "i-


'V ~-


NET SELLER NET BUYER


NET SELLER NET BUYER


Figure 3-2. Impact of High Sulfur Coal Contract


(ii)


MAX


I PC.


!


I __


P, .---.





















MCAsC -

MCAss


4-


I-


(i)


R N
NET SELLER NET BUYER


Figure 3-3. Impact of Low Sulfur Coal Contract


SI




s
MC4" I
T-_



MC.t" ~~-


NET SELLER NET BUYER
NET SELLER NET BUYER


AsM A ASM A; 'AV 0
I A, /i


Figure 3-4. Excess Demand Correspondence with Scrubber ('C!1. .,


Ar

























Figure 3-5. Impacts of High Sulfur Coal Contract









PA (i) PA (ii)



PAS T I I
MCg4u '-- Imacsof-- MCA4,-- --r--

: 11 I 10 A I
A 4"" a T~ X O 4MM 4 < AM 4M 0


Figure 3-6. Impacts of Low Sulfur Coal Contract




















MCAf 's


- --- --------- -


a --I


A A4X
^f


Figure 3-7. Given Scrubber C'!,i..


MCA Ss


Shift from High Sulfur Contract


------- --- -------


I -


E- m -


i


"MAX
A,


Figure 3-8. Given Scrubber C('! i..


mmm


~ABLi~kT
i


Shift from Low Sulfur Contract



























AMCAsA'


MCi


pS
A

- I
_e --


4----


ASAMI A SMAX 4MLV '?"
A A"' As- A Ar'


Figure 3-9. With Scrubber Choice: Shift from High Sulfur Contract





















P

MrA5
AJOA,4S's


MCI


I
I

S --u

I
-A

I

I
pj _


- U


1 14-


I


i


A i


Figure 3-10. With Scrubber C'!h. Shift from Low Sulfur Contract


A4SMIN
il


" MAX
A,


SMA


-i. T-*. *^---^


i









APPENDIX A
CONTRACT IMPACTS ON COSTS AND SCRUBBER INSTALLATION
INDIFFERENCE PRICE

A.1 Impacts on Total Costs and Compliance Costs from a Coal Contract
Constraint

First, consider a unit's total costs without a coal contract constraint. Without

the program restrictions on emissions, a unit will simply minimize its costs of meeting

electricity demand by using the coal with the lowest price per unit of heat content.

pstsl I pS1Csl (Al)
ihPi' + Pis(A-)

With the emissions constraint from the program, a unit will minimize its costs of coal use,

net allowance purchases, and scrubber installation.


z P + PA C + P* + P8Q* (A-2)

The difference between (A-1) and (A-2) is the total compliance costs resulting from the

program, which includes the change in coal costs, change in the net allowance purchases,

and scrubber installation costs.


zP, + P IA + P(q* CQh) + P(Cq* C1') (A-3)

Second, consider a unit's total costs with a coal contract constraint. Without the

program restrictions on emissions, a unit will simply minimize its costs of meeting

electricity demand by using all the the coal under contract, and cover the remainder of its

coal demand with the coal with the lowest price per unit of heat content.


PiC + pil + PhQ + PQ' (A-4)

With the emissions constraint, a unit will minimize its costs for coal use, net allowance

purchases, and scrubber installation given its emissions and coal contract constraint.


z Pz + PjA i + P;hC + PC~i + P C* + PC* (A-5)









The difference between (A-4) and (A-5) is the total compliance costs resulting from the

program, which includes the change in spot market coal costs, net allowance purchases,

and scrubber installation. The costs from contract coal cancel out because contract coal

use will be the same both with and without the program.


P, + PA* + P ( C- ) + P C C') (A-6)

The sufficient conditions under which a coal contract constraint will increase or

decrease a unit's compliance costs can be derived from the difference in compliance costs

with and without a coal contract ((A-6) minus (A-3)).


[, + P *A + P s rs- +C8) + Pis,* CI)

z[P, + PIA: + P, (Ch h) + P,(Q* C)] (A-7)

For simplicity, assume that the scrubber choice as a given and high sulfur spot

market coal is relatively cheaper than low sulfur spot market coal. So without an

emissions constraint a unit will prefer to use the cheaper high sulfur coal. Proposition

1 can be proven by considering the change in compliance costs in (A-8) resulting from

a coal contract. First consider a high sulfur coal contract to show Proposition 1(i) and

Proposition l(ii) hold.

Proof of Proposition 1(i):

If PA > MCA"', a unit prefers to switch from high to low sulfur coal to meet its

emissions requirement because it is the least-cost compliance option. Without a high

sulfur coal contract, a unit will use all low sulfur coal (C* = Ci' and C = 0) and

require the fewest allowances to cover the minimum emissions level (A"IN). With a high
sulfur coal contract, a unit will use less low sulfur coal (,,MAX < C: ), which will









increase the emissions level and require additional allowances (AMIN > AMIN).

P~iIV ( is, MAX\ ,s( s,MAX 0)] [ *AMIN+ ps / -s,MAX\ srs,MAX)
[PI -i T1 Pish (0- C"A +pis Cil i^ I ~{J \
(A-8)

The change in compliance costs will be:

J/IV- AMIN) S (C MAX is,MAX r s s,MAX s,MAX) (A 9)
A MN ) hih Cih + ~ il ) 9)


The first term is positive because AMIN AMIN. The second term is also positive because
~s,MAX -s ,MAX. sMAX < CIs,MAX
hMAX > MAX. The third term is negative because ",MAX < ,MAX

Now fill in for coal use:

is,MAX Di ^sMAX D, CHh Cs,MAX Di ^sMAX Di Cfh Hh
ih H i H8 i H i -
h h Hih i His

The change in compliance costs resulting from a high sulfur coal contract is the

increase in net allowance purchases minus the cost savings from not switching fuels from

the high sulfur contract coal.


Pj4 V AIN) ) hH- (A-10)
Hih His

Now fill in for the net allowance position:

MIN D I/ I V D ch ih cs c A 'e
IN Srm A Ai D s i + C
His, His, M i




By adding and subtracting mPA ,h combining like terms, and dividing through by

m( H), the expression can be simplified to:
r50 szc ps ps
ih ih ih ih





itHCh [P>( SMh ) + PA CA (A11)
i h i h
iCHh PI ( h ) + PA MCA"] (A- t)
;57h ;571









For this case, it is assumed the PJ > MCA7'8. So if > compliance costs will
ih ih
increase, and the increase will be get larger as the coal contract gets larger.E

Proof of Proposition l(ii):

If PA < MCA"'>, a unit prefers to use all high sulfur coal and purchase allowances to

meet its emissions requirement because it is the least-cost compliance option. Without a
MAX
high sulfur coal contract, a unit will use all high sulfur spot market coal (Ci = CiM

and Ci* = 0) and require the most allowance use to cover its maximum emissions level

(AMAX). With a high sulfur coal contract, a unit will use less high sulfur spot market coal

(C AX < AX ). If high sulfur contract coal has a higher sulfur to heat content ratio

(; > ), a unit will generate additional emissions, which will require a unit to purchase
1ih Hih
more or sell fewer allowances (A AX > AMAX). Now fill in for known values and combine

like terms:

P*(A" '' AjMAX) PC TTH( I)ih (A-12)
Hih Hih
All coal use remains the same, which means the change in compliance costs will be the

increase in costs from additional allowances. If h > a unit's compliance costs will
Hih Hih
increase. U

Proof of Proposition l(iii):

If PA > MCA"', a unit prefers to switch to all low sulfur use because it is the

least-cost compliance option. Without a low sulfur coal contract, a unit will use all low
sulfur spot market coal (C* = CMX and C* = 0) and require the least possible

allowances to cover its minimum emissions level (AmIN). With a low sulfur coal contract,

a unit will use less low sulfur spot market coal (CAX < C ). If low sulfur

contract coal has a higher sulfur to heat content ratio (; > ;), a unit will generate

additional emissions, which will require a unit to purchase more or sell fewer allowances

(AMIN > AMIN). Coal use will remain the same with and without an emissions constraint,









and will cancel out. Fill in for known values and combine like terms:

AMIN Di -I Ae I r 'V Di /17 r~ ,Y c A-
Ai Sm A il A Hiiil i


P(A1i1 AMIN) PCiH l( ) (A-13)
i ( Ai A i I HI H i H isi l
The change in compliance costs will be the increase in costs from the increase in a unit's

net allowance position. If > S a unit's compliance costs will increase.E

Proof of Proposition 1(iv):

If PA < MCA"'8, a unit prefers to use all high sulfur coal and purchase allowances

instead of switching fuels to meet its emissions requirement because it is the least-cost

compliance option. Without a low sulfur coal contract, a unit will use all high sulfur coal

(C7 = CMAX and C"* 0) and require the largest net allowance position to cover the
maximum emissions level (AMAX). With a low sulfur coal contract, a unit will use less

high sulfur coal (CiA < sih A), which will decrease the emissions level and requires

fewer allowances (AAX > AMAX). Since a unit prefers to use high sulfur coal with and

without the emissions constraint, coal use will remain the same. Fill in for known values

and combine like terms:

D/ iI'V AMIN)Ps\ CS,MAX _s,MAX s ,MAX s s,MAX it i, V AMIN) (A 1)
SA~i~i Aih\h ~ih hi (iC ) C C ~ A A (A

The change in compliance costs is the change in costs from the change in net allowance

position. Now fill in for the net allowance position:

AMAX SDi 8 e I ; _i (HD -s ,+ CSc c Ae
S- m Ai Ai + C,-Sm Aii
Hih ih

CcC HP [i ihA5
HzPm sh < 0 (A15)
SC S
Since ~ < compliance costs will decrease.E

Another way of looking at the impacts of coal contract constraints on compliance

costs is to find the change in total costs for a unit facing an emissions constraint with










and without a coal contract ((A-5)-(A-2)) and split it into two components, the change

in compliance costs and the change in fuel costs. For simplicity, assume the conditions in

Proposition l(i) hold.


4P, + PI(A* A*() C+ P +P ( -C + P(' C}*) (A-16)


Assume that a unit faces a high sulfur coal contract, and prefers to switch to low sulfur

coal use to meet its emissions requirement instead of purchasing allowances or installing a

scrubber.
P/ ,iV AMIN) + p ps(s,MAX I,MAX) (A7)
i js ilc (il

To be able to interpret this expression, it is necessary to add and subtract (P ,hJi ).


C (P PH ) P( V AMIN) + Ps h + P s,MAX s,MAX) (A 18)
-- ich (P ~h i h T + p (A i )+ Ai ihrr 1H isil ~ pi ) "-,;)
ih ih

The first term is the change in high sulfur coal costs from using the contract coal instead

of spot market coal. These are not changes in compliance costs because they will occur

with or without the program. The remaining terms are the change in compliance costs

resulting from the program.

H h ,AIN), Al Pc -9)
ch(Pich Ph ) + P( i" V AMIN )rCH (A 19)
hh ihrs Hsh Hi (iA 1

Change in Fuel Costs Change in Compliance Costs

By filling in for the coal use, the last terms give the same expression for the change in

compliance costs as in the proof of Proposition l(i).

ih ih
Ch Hh [P( ) + PA MCA] (A 20)
H H
ih il
SC S
A unit's compliance costs increase if > H-.
Hih ih









A.2 Derivation of Cost-Minimizing Input Use to Find PA

Assuming no scrubber, the cost-minimizing combination of coal and allowances solves

the problem below.

miP ihip AMAX ps i s,MAX p AMIN s,MAX (A21)
n ,,C (PAAi + ihih A PA + PilCil ) (A )

Assuming a scrubber is installed, the cost-minimizing combination of inputs is expressed

below.
miniChC A (l ASMAX + pis s,MAX P ASMIN + ps s,MAX) (A22)
mTnAi,c-ih, t, A +ih ih A~i + .il.U ) (A 22)

Notice that contract coal use and the cost of installing a scrubber can be ignored because

all are constants.









APPENDIX B
MARKET EQUILIBRIUM AND SIMULATION DESIGN

B.1 Conditions for Existence of an Equilibrium

Theorem 5.1: Assuming the scrubber choice (z*) as given, a market equilibrium exists.

Proof of Theorem 5.1: To prove that a market equilibrium exists, it is necessary to

apply Kakutani's Fixed Point Theorem.
Kakutani's Fixed Point Theorem that states: If X is a non-c>mi'l; compact, convex

subset of R" and if f is an upper semi-continuous correspondence from X into itself such

that (Vx c X) the set f(x) is non-e(t ,ji; and convex, then f has a fixed point (there is an

x f(x)).
Assuming all other parameters as given for all generating unit "i", a generating units

excess demand is a function of the allowance price, Ai(PA). Consider a generating unit's

excess demand correspondence from Section 5.

SAAX if PA > MCAj',
A pAAX (1 p)AfIN if PA = MCAf Vp e [0,1]

AfIN if PA < MCAj',

To find the market excess demand, you must sum all generating unit excess demands,

which gives you Am = Yi Ai. It is necessary to show that the market excess demand

is compact. The market excess demand is bounded above by the sum of the maximum

allowance excess demands for each generating unit ( 1i AAX) and below by the sum

of the minimum allowance excess demands for each generating unit ( Ai AfIN). So the

market excess demand lies in a closed interval [Ei 1 AfIN, 1 AMAX] X. X being a

closed interval on the real line is non-empty, compact (closed and bounded), and convex

(any point on the line connecting any two points in the set is also in the set).
It is necessary to show that the set of prices is compact (closed and bounded). Define

the lower bound on price to be zero because it is necessary to have a positive price.

Define the upper bound on price to be PA argmin Ai(PA), or the allowance









price that results in the smallest possible excess demand for a particular generating unit

multiplied by two. This ensures a upper bound that will be higher than any possible price.

The closed interval for prices is [0, PA], which is non-empty by design. Since there is an

upper bound, a lower bound, and both bounds are included in the set (closed), the set is

compact.

Since the set of prices and market excess demands are compact and the market

excess demand depends on the allowance price, the market excess demand (Am(PA)) is a

mapping from [0, PA] into X. It is necessary to show that excess demand is non-empty,

convex valued at each PA E [0, PA], and upper semi-continuous. It is sufficient to show

that each generating unit's excess demand correspondence is non-empty, convex, and

upper semi-continuous because the sum of a finite number of non-empty, convex, upper

semi-continuous correspondences is also non-empty, convex, and upper semi-continuous.

A generating units' excess demand (Ai(PA)) are defined to be mappings from

[0, PA] into X, which makes them non-empty by construction. A generating unit's
excess demand correspondence is closed and bounded in the interval [A"IN A AX],

which makes it a compact set. The set of excess demands is convex because the average

of any two excess demand values is also in the set. For a correspondence to be upper

semi-continuous, the convergence excess demand value of any price sequence must also

be in the correspondence. Since every possible price sequence converges to a value that

is in the excess demand correspondence (see Figure B.2), each unit's excess demand

correspondence is upper semi-continuous.

Now we must define the mapping from X into the set of prices [0, PA] as pl(x) where...



( PA E [0,PA PA [, = maxQ[ PA Q x if x / 0
PA E [0, PA :PA = 0 if otherwise

Since [0, PA] is non-empty, p(x) must be non-empty as well. The mapping is convex

valued since for market excess demand equal to zero (x = 0), p(x) = [0, PA]. If there is a









positive excess demand (x > 0), p(x) =PA. If there is a negative excess demand (x < 0),

or excess supply, p(x) = 0. Since [0, PA] is compact, the graph of p(x) is closed, which

implies p(x) is upper semi-continuous.

Now define F(x, PA) i(x) x Am(PA). Since p(x) and Am(PA) satisfy all properties

needed to apply Kakutani's Fixed Point Theorem, there exists a fixed point (x*, Pj) such

that PA E [0, PA] and x* c X such that PA E p(x*) and x* e Ar(PA).E

In English...There is a market excess demand correspondence for which each excess

demand value can only result from only one allowance price while each allowance price will

result in at least one market excess demand value.

Once the scrubber choice is introduced into the decision-making process, the

correspondence becomes more complex, as seen in Figure B.2. An equilibrium may in

fact exist, but there is no way to guarantee an equilibrium because the excess demand

correspondence is no longer a convex set. The average of the two excess demand values

(ASMAX and A"IN) is not in the excess demand correspondence.

B.2 Technical Details of Simulation Model Design

The equilibrium allowance market price is solved by using a bisection iterative

process. An upper limit ($1,000) and lower limit ($0) for the allowance price are chosen.

The initial allowance price is set to the upper limit and the simulation solves for each

generating unit's cost-minimizing choices. Then it checks if the allowance market is in an

equilibrium.

If market excess demand is positive, the allowance price is too low and the allowance

price is increased by one-half the difference between the upper and lower limits. The old

price now becomes the new lower limit while the upper limit remains the same. If the

market excess demand is negative, the allowance price is too high and the allowance price

is decreased by one-half the difference between the upper and lower limits. The old price

becomes the new upper limit and the lower limit remains the same. In this case, the upper









limit is set high enough to guarantee the price is too low. So the price will decrease by

$1,000-($1000-$0)/2=$500 to a new price of $500.

The program is then run again with the new allowance price, each iteration decreasing

the difference between the upper and lower limits by half until the program converges to

an allowance price. Once the program converges, it must ensure a market excess demand

of zero. A concern is that the program tends to push a unit's choices towards a corner

solution where the market may not clear.

Given the scrubber choice, the allowance price will converge to a value equal to at

least one generating unit's MCA'", which allows those firms choices to be shifted to

an interior solution to clear the market without altering the unit's total costs or total

industry costs because the unit is indifferent to purchasing allowances or switching fuels

from high to low sulfur coal at the equilibrium allowance price (PJ).

Allowing for the scrubber choice will result in the convergence of PA at an allowance

price where a generating unit is indifferent to installing a scrubber. In this case, there is

not a true equilibrium and must consider it a quasi-equilibrium as described above.











L -- a p


PI


MCAdI"3

Figure B-1. Upper Semi-Continuous Correspondence


A,(PA)





Not in A 1
Set ,.? _


Non-Convexity


MICAT.s Ps MCs_ PA


Figure B-2. Correspondence with Scrubber ('C! ,,,


A (PA)

MAXM








A MIN


PA









REFERENCES


[1] T. Arimura. An empirical study of the SO2 allowance market: effects of PUC
regulations, J. Environ. Econ. Manage., 44 (2) (2002) 271-289.

[2] R. E. Baldwin, C.S. Magee, Is trade policy for sale? Congressional voting on recent
trade bills, Public C'! i... 105 (2000) 79-101.

[3] L. Bird, M. Bolinger, T. Gagliano, R. Wiser, M. Brown, B. Parsons. Policies and
market factors driving wind power development in the United States, Energy Pol. 33
(11) (2005), 1397-1407.

[4] D.R. Bohi. Utilities and state regulators are failing to take advantage of emission
allowance trading, Electricity J., 7 (2) (1994) 20-27.

[5] D.R. Bohi, D. Burtraw. Utility investment behavior and the emission trading market,
Resources and Energy, 14 (1992) 129-153.

[6] D.R. Bohi, D. Burtraw. SO2 allowance trading: How experience and expectations
measure up, Discussion Paper 97-24, Resources for the Future, Washington, D.C.,
1997.

[7] M. Bolinger, R. Wiser. 2004. The impact of state clean energy fund support for
utility-scale renewable energy projects, Lawrence Berkeley National Laboratory and
Clean Energy States Alliance.

[8] M. Bolinger, R. Wiser, G. Fitzgerald. 2006. The impact of state clean energy fund
support for iul iii ---I i1.- renewable energy projects, Lawrence Berkeley National
Laboratory and Clean Energy States Alliance.

[9] M. Bolinger, R. Wiser, G. Fitzgerald. An overview of investments by state renewable
energy funds in large-scale renewable generation projects, Electricity J. 18 (1) (2005)
78-84.

[10] M. Bolinger, R. Wiser, L. Milford, M. Stoddard, K. Porter. Clean energy funds: An
overview of state support for renewable energy, Energy Efficiency and Renewable
Energy, Office of Power Technologies, Department of Energy, 2001.

[11] D. Burtraw. Cost savings sans allowance trades? Evaluating the SO2 emission trading
program to date, Discussion Paper 95-30, Resources for the Future, Washington,
D.C., 1995.

[12] D. Burtraw, Evans, A. Krupnick, K. Palmer, R. Toth. Economics of pollution trading
for SO2 and NOx, Discussion Paper 05-05, Resources for the Future, Washington,
D.C., 2005.

[13] C. Carlson, D. Burtraw, M. Cropper, K. Palmer. Sulfur dioxide control by electric
utilities: What are the gains from trade?, J. of Political Economy, 108 (6) (2000)
1292-1326.









[14] C. C'!, i, R. Wiser, M. Bolinger, 2007. Weighting the costs and benefits of state
renewables portfolio standards: A comparative analysis of state-level policy impact
projections, Lawrence Berkeley National Laboratoy.

[15] J.S. Coc-:iii-, J.R. Swinton. The price of pollution: A dual approach to valuing SO2
allowances, J. Environ. Econ. Manage., 30 (1996) 58-72.

[16] Database of State Incentives for Renewable Energy. State renewable energy policy
summaries, Website, Accessed March 2004-March 2006, http://www.dsireusa.org.

[17] Database of Utility-Scale Renewable Energy Projects. Lawrence Berkeley National
Laboratory and Clean Energy State Alliance (CESA), Website, Accessed December
2006, http://eetd.lbl.gov/ea/ems/cases/Large_Renewables-Database.xls.

[18] A.D. Ellerman, R. Schmalensee, P.L. Joskow, J.P. Montero, E.M. Bailey. Emissions
trading under the U.S. acid rain program: Evaluation of compliance costs and
allowance market performance, Center for Energy and Environmental Policy
Research, Massachusetts Institute of Technology.

[19] A.D. Ellerman, et al. Market for clean air: the U.S. acid rain program, Cambridge
University Press, New York, 2000.

[20] A.D. Ellerman, J.P. Montero. The declining trend in sulfur dioxide emissions:
Implications for allowance prices, J. Environ. Econ. Manage., 36 (1998) 26-45.

[21] Energy Information Administration. Electric power annual 1996: Vol.2,
DOE/EIA-0348(96)/2, February 1996.

[22] Energy Information Administration. The effects of Title IV of the Clean Air Act
Amendments of 1990 on electric utilities: An update, DOE/EIA-0582(97), March
1997.

[23] Energy Information Administration. Energy Policy Act transportation rate study:
Final report on coal transportation, DOE/EIA-0597(2000), October 2000.

[24] Energy Information Administration. EIA historical state electricity databases
and FERC Form-423 database, Website, Accessed March 2004 March 2007,
http://www.eia.doe.gov/fuelelectric.html.

[25] Energy Information Administration. Annual Energy Review 2005.
Report #DOE-EIA-0384(2005). Website, Accessed February 2007,
www.eia.doe.gov/emeu/aer/contents.html.

[26] Environmental Protection Agency. Reducing power plant emissions for cleaner
air, healthier people, and a strong America. Office of Air and Rediation,
www.epa.gov/cair/chartsifiles/cair-finalpresentation.pdf, March 2005.









[27] Environmental Protection Agency. Acid rain program progress report,
EPA-430-R-06-015, Clean Air Markets Division Office of Air and Radiation, October
2006.

[28] Environmental Protection Agency. Information on the Clean Air Interstate Rule.
Website, Accessed September 2007, http://www.epa.gov/interstateairquality/.

[29] Federal Reserve Branch of St. Louis. Consumer price index data. Website, Accessed
December 2004, http://stlouisfed.org.

[30] D. Fullerton, S. McDermott, J.P. Caulkins. Sulfur dioxide compliance of a regulated
industry, J. Environ. Econ. Manage., 34 (1) (1997) 32-53.

[31] GDS Associates, Inc. Analysis of renewables portfolio standard options for Hawaii,
The State of Hawaii Department of Business, Economic Development, and Tourism,
2001

[32] P.L. Joskow. Vertical integration and long-term contracts: The case of coal-burning
electric generating plants, J. of Law, Econ., and Org., 1 (1) (1985) 33-80.

[33] P.L. Joskow. Contract duration and relationship-specific investments: Empirical
evidence from coal markets, American Economic Review, 77 (1) (1987) 168-185.

[34] P.L. Joskow. Price adjustment in long-term contracts: The case of coal, J. of Law and
Economics, 31 (1) (1988) 47-83.

[35] P.L. Joskow. Asset specificity and the structure of vertical relationships: Empirical
evidence, J. of Law, Econ., and Org., 4 (1) (1988) 95-117.

[36] P.L. Joskow. The performance of long-term contracts: Further evidence from coal
markets, RAND J. of Econ., 21 (2) (1990) 251-274.

[37] J.P. Kalt, M.A. Zupan. Capture and ideology in the economic theory of politics,
American Economic Review 74 (3) (1984), 279-300.

[38] N.O. Keohane. Environmental policy and the choice of abatement technique:
Evidence from coal-fired power plants. Yale School of Management, 2002.

[39] O. Langniss, R. Wiser. The renewables portfolio standard in Texas: An early
assessment, Energy Pol. 31(6) (2003), 527-535.

[40] League of Conservation Voters. Description and data from the LCV "National
Si. .', i 1.1 Website, Accessed March 2004, www.lcv.org.

[41] J. McVeigh, D. Burtraw, J. Darmstadter, K. Palmer. Winner, loser, or innocent
cictim? Has renewable energy performed as expected?, Discussion Paper 99-28,
Resources for the Future, Washington, D.C., 1999.

[42] F.C. Menz, S. Vachon. The Effectiveness of different policy regimes for promoting
wind power: Experiences from the states, Energy Pol. 34 (2006) 1786-1796.









[43] J.P. Montero. Marketable pollution permits with uncertainty and transaction costs,
Resource Energy Econ., 20 (1997) 27-50.

[44] D.W. Montgomery. Markets in licenses and efficient pollution control programs, J.
Econ. Theory, 5 (1972) 395-418.

[45] G. Morris. The status of biomass power generation in California. Subcontractor
Report. National Renewable Energy Laboratory. Green Power Institute, Pacific
Institute for Studies in Development, Environment, and Security, 2003.

[46] J.P. Nelson. Green voting and ideology: LCV scores and roll-call voting in the U.S.
Senate, 1988-1998, Review of Econ. and Stat. 84 (3) (2003), 518-529.

[47] T. Petersik. State renewable energy requirements and goals: status through 2003.
Energy Information Association, 2004.

[48] B.G. Rabe. Race to the top: The expanding role of U.S. state renewable portfolio
standards, Pew Center on Global ('Ci i. C. .('!i ,-, 2006.

[49] A.E. Smith, J. Platt, A.D. Ellerman. The costs of reducing utility SO2 emissions- Not
as low as you think, C'!i !l I River Associates, Electric Power Research Institute, and
Massachusetts Institute of Technology.

[50] D.C. Smith, R.C. Grace, R. Wiser. Massachusetts renewable portfolio standard: Cost
analysis report, La Capra Associates and Sustainable Energy Advantage, LLC under
contract to Massachusetts Division of Energy Resources, 2000.

[51] P.M. Sotkiewicz. The impact of state-level public utility commission
regulation on the market for sulfur dioxide allowances, compliance
costs, and the distribution of emissions. Doctoral Dissertation, Dept.
of Economics, University of Minnesota, January 2003 posted at
http://bear.cba.ufl.edu/centers/purc/primary/documents/PS-Dissertation.pdf.

[52] P.M. Sotkiewicz, L. Holt. Public utility commission regulation and cost-effectiveness
of Title IV: Lessons for CAIR, Electricity J., October 2005, 68-80.

[53] R. Stavins. Transaction costs and tradeable permits, J. Environ. Econ. Manage., 29
(1995), 133-148.

[54] B. Swift. How environmental law work: An analysis of the utility sector's response
to regulation of nitrogen oxides and sulfur dioxide under the Clean Air Act, Tulane
Environmental Law J., 14 (1) (2001) 309-426.

[55] J.R. Swinton. Phase I completed: An empirical assessment of the 1990 CAAA,
Environmental and Resource Econ., 27 (2004) 227-246

[56] J.R. Swinton. The potential for cost savings from the sulfur dioxide allowance market:
Empirical evidence from Florida, Land Econ., 78:3 (2002) 390-404.









[57] J.J. Winebrake, A.E. Farrell, M.A. Bernstein. The Clean Air Act's sulfur dioxide
emissions market: Estimating the costs of regulatory and legislative intervention,
Resource Energy Econ., 17 (1995) 239-260.

[58] R. Wiser, S. Olson. Utility green pricing programs: A statistical analysis of program
effectiveness, Environmental Energy Technologies Division, Lawrence Berkeley
National Laboratory, 2004.

[59] R. Wiser, K. Porter, R. Grace. Evaluating experience with renewables portfolio
standards in the United States, Environmental Energy Technologies Division,
Lawrence Berkeley National Laboratory, 2004.









BIOGRAPHICAL SKETCH

Joshua David Kneifel was born in 1981 in North Platte, N. ,i i-1: i. He grew up

in North Platte, graduating salutatorian from Hershey High School in 1999. Joshua

received his bachelor's degrees in economics and mathematics in 2003 from Doane College

in Crete, N. i i-1;: i. He received his Master of the Arts in economics in 2005 from the

University of Florida, where he specialized in industrial organization, public economics,

and econometrics. From Fall 2006 through Spring 2008, Joshua instructed four semesters

of a course in environmental economics. His classwork and research allowed him to obtain

his PhD in economics from the University of Florida.

Upon completion of his PhD program, he will take an economist position at the

National Institute of Standards and Technology (NIST) in Gaithersburg, Maryland. His

responsibilities at NIST will include research on the life-cycle cost and environmental

impacts of individual products used in the construction industry.





PAGE 1

1

PAGE 2

2

PAGE 3

3

PAGE 4

Ithankmysupervisorycommitteechair,LawrenceKenny,alongwithmyothercommitteemembers(SanfordBerg,JonathanHamilton,andJaneLuzar)fortheircommentsandadviseregardingmydissertation.SpecialthanksgoestoPaulSotkiewiczforhisinterest,guidance,andsupportinmyresearch.Ialsothankmyparentsfortheirnever-endingsupport. 4

PAGE 5

page ACKNOWLEDGMENTS ................................. 4 LISTOFTABLES ..................................... 9 LISTOFFIGURES .................................... 10 ABSTRACT ........................................ 12 CHAPTER 1EFFECTSOFSTATEGOVERNMENTPOLICIESONELECTRICITYCAPACITYFROMNON-HYDROPOWERRENEWABLESOURCES ..... 14 1.1Introduction ................................... 14 1.2LiteratureReview ................................ 16 1.3Model ...................................... 18 1.4VariablesandData ............................... 21 1.4.1EconomicandPoliticalVariables:Wit 21 1.4.2RegulatoryPolicyVariables:Rit 27 1.5StatisticalSpecicationsandEmpiricalAnalysis ............... 33 1.5.1EconomicandPoliticalVariables:Wit 34 1.5.2RegulatoryPolicyVariables:Rit 37 1.5.3StateFixed-EectsVariables:Sit 40 1.5.4YearVariables:Tit 40 1.6Conclusions ................................... 41 2EFFECTSOFCOALCONTRACTCONSTRAINTSONSO2TRADINGPROGRAMCOMPLIANCEDECISIONS ..................... 47 2.1Introduction ................................... 47 2.2PolicyBackground ............................... 48 2.2.1TitleIVoftheCleanAirActAmendment .............. 48 2.2.1.1PhaseIofTitleIV ...................... 49 2.2.1.2PhaseIIofTitleIV ...................... 50 2.2.2CleanAirInterstateRule ........................ 51 2.3LiteratureReview ................................ 52 2.3.1TitleIV:PhaseI ............................ 52 2.3.2Utility-LevelModelsofComplianceCosts ............... 55 2.3.3Long-TermCoalContracts ....................... 56 2.4InecienciesResultingfromCoalContractConstraints ........... 57 2.5ModelandParameters ............................. 60 2.6GeneratingUnitLevelDecision-MakingProcess ............... 62 2.6.1GeneratingUnit'sProblem ....................... 63 2.6.2First-OrderConditions ......................... 64 5

PAGE 6

................ 65 2.6.3.1Necessaryconditionsforusingbothhighsulfurandlowsulfurcoal ........................... 65 2.6.3.2Onlyhighsulfurcoaluse:Necessaryconditions ...... 65 2.6.3.3Onlylowsulfurcoaluse:Necessaryconditions ....... 66 2.6.4CoalUseUnderaHighSulfurCoalContractConstraint ....... 66 2.6.5CoalUseunderaLowSulfurCoalContractConstraint ....... 68 2.6.6GeneratingUnit-LevelComplianceCosts ............... 70 2.6.7GeneratingUnit'sNetAllowancePosition:ExcessDemandCorrespondence ....................... 72 2.6.7.1CostsavingsoffuelswitchingversusallowancepurchaseswhenPA>MCAs;si 74 2.6.7.2Eectsofhighsulfurcoalcontractsonexcessdemandandcosts .............................. 74 2.6.7.3CostsavingsofallowancepurchasesversusfuelswitchingwhenPA
PAGE 7

.. 136 3.4.3ComparativeStatics:EectofCoalContractsontheAllowanceMarketwithEndogenousScrubberChoice .............. 136 3.4.3.1Highsulfurcoalcontractbinds ............... 137 3.4.3.2Lowsulfurcoalcontractbinds ................ 138 3.5ComplianceCosts ................................ 139 3.5.1ComplianceCostswithCoalContractsRelativetoComplianceCostsfromPreviousStudies .......................... 139 3.5.2TotalIndustryComplianceCost .................... 141 3.5.3ImpactofAllowanceAllocationonComplianceCostsGivenScrubbertheChoice ................................ 144 3.5.4ImpactofAllowanceAllocationwithEndogenousScrubberChoice 146 3.6SimulationModel ................................ 147 3.6.1Introduction ............................... 147 3.6.2Data ................................... 147 3.6.2.1Fueldata ........................... 147 3.6.2.2Allowance,actualemissions,anddemanddata ....... 150 3.6.2.3Technicalgeneratorandscrubberdata ........... 151 3.6.3SimulationModelDesign ........................ 152 3.6.4SimulationResults ........................... 152 3.6.4.1Totalindustrycostsandallowancemarketresults ..... 153 3.6.4.2Industryandgeneratingunitcoaluse ............ 157 3.6.4.3Generatingunitscrubberinstallationchoices ........ 158 3.6.4.4Impactofallowanceallocationontheallowancemarketandcompliancecosts ..................... 160 3.6.4.5Summaryofsimulationresults ................ 160 3.7PlantLevelDecision-MakingProcess ..................... 161 3.7.1Introduction ............................... 161 3.7.2Plant-LevelProblem .......................... 163 3.7.3First-OrderConditions ......................... 164 3.7.4CharacterizingaUnit'sSpotMarketFuelChoices .......... 165 3.7.4.1Case1:Necessaryconditionsforusingbothhighsulfurandlowsulfurspotmarketcoal ............... 166 3.7.4.2Case2:Necessaryconditionsforonlyhighsulfurspotmarketcoaluse ............................ 167 3.7.4.3Case3:Necessaryconditionsforonlylowsulfurspotmarketcoaluse ............................ 167 3.7.5ExcessDemandCorrespondence .................... 168 3.7.6CharacterizingaGeneratingUnit'sContractFuelChoices ...... 170 3.7.6.1Case1:NecessaryconditionsforhighsulfurcontractcoaluseatGeneratingUnit\i" .................. 170 3.7.6.2Case2:Necessaryconditionsforlowsulfurcontractcoaluseatgeneratingunit\i" .................. 171 3.7.7\Non-Aected"GeneratingUnitsatan\Aected"Plant ...... 171 7

PAGE 8

........................ 172 3.7.7.2\Non-aected"generatingunitsandhighsulfurcoalcontracts .................. 173 3.7.8ScrubberInstallationChoice ...................... 174 3.7.8.1Marginalcostofabatementwithandwithoutascrubber 174 3.7.9APlant'sPreferredOrderofScrubberInstallation .......... 176 3.7.9.1Atwhichgeneratingunitswillaplantinstallascrubber? 177 3.7.9.2Atwhatallowancepricewillaplantinstallascrubberatagivengeneratingunit? ................... 178 3.7.9.3Scrubberinstallationandhighsulfurcoalcontracts .... 179 3.7.9.4Scrubberinstallationandlowsulfurcoalcontracts ..... 181 3.7.10ScrubberInstallationExample:PlantwithTwoAectedGeneratingUnits ................................... 182 3.7.10.1Case1:Installnoscrubbers ................. 183 3.7.10.2Case2:Installonescrubber ................. 184 3.7.10.3Case3:Installtwoscrubbers ................ 185 3.7.11ScrubberInstallationExample:PlantwithOneAectedandOneNon-AectedGeneratingUnits ..................... 186 3.7.12SummaryofPlantLevelResults .................... 187 3.8CONCLUSIONS ................................ 188 ACONTRACTIMPACTSONCOSTSANDSCRUBBERINSTALLATIONINDIFFERENCEPRICE .............................. 199 A.1ImpactsonTotalCostsandComplianceCostsfromaCoalContractConstraint ................................. 199 A.2DerivationofCost-MinimizingInputUsetoFindPSA 205 BMARKETEQUILIBRIUMANDSIMULATIONDESIGN ............ 206 B.1ConditionsforExistenceofanEquilibrium .................. 206 B.2TechnicalDetailsofSimulationModelDesign ................ 208 REFERENCES ....................................... 211 BIOGRAPHICALSKETCH ................................ 216 8

PAGE 9

Table page 1-1DependentandControlVariables .......................... 43 1-2RegressionsResults .................................. 44 1-3PolicyVariables .................................... 45 1-4VariableEectsofSignicantVariables ....................... 46 2-1PhaseIComplianceCostEstimates ......................... 113 2-2HighSulfurCoalContract:Assumptions ...................... 114 2-3HighSulfurCoalContract:Results ......................... 114 2-4LowSulfurCoalContractExamples:Assumptions ................ 114 2-5LowSulfurCoalContractExamples:Results ................... 115 2-6ExampleEpsilonMagnitude:Case1 ........................ 116 2-7ExampleEpsilonMagnitude:Case2 ........................ 117 2-8ExampleEpsilonMagnitude ............................. 118 3-1Example:ContractCoalDistribution ........................ 190 3-2SulfurConversionbyFuelType ........................... 190 3-3SimulationResults .................................. 191 3-4ImpactofContractConstraintonScrubberChoice ................ 192 3-5SimulationswithEngineeringData ......................... 192 3-6ImpactsofaReductionintheAllowanceAllocationof10% ............ 192 3-7MathExample:TwoAectedUnits ......................... 192 9

PAGE 10

Figure page 2-1TheSO2AllowancePrice .............................. 119 2-2ExcessDemandCorrespondenceandComplianceCostSavingsfromFuelSwitchingOverAllowancePurchasing ............................. 119 2-3HighSulfurContract:ShiftinMinimumExcessDemand ............. 120 2-4NoContract:ComplianceCosts ........................... 120 2-5HighSulfurContract:ComplianceandTotalCosts ................ 120 2-6HighSulfurContract:RelativeSavingsfromContractCoal ............ 121 2-7CostSavingsfromUsingAllowancesOverFuelSwitching ............. 121 2-8LowSulfurContract ................................. 121 2-9NoContract:ComplianceCosts ........................... 122 2-10LowSulfurCoalContract:MCAs;ci 122 2-11LowSulfurCoalContract:MCAs;ci 122 2-12ComplianceCosts:PA=MCAs;si 123 2-13HighSulfurCoalContract .............................. 123 2-14LowSulfurCoalContract .............................. 123 2-15ExcessDemandCorrespondence:MCAs;siPSA 125 2-19ImpactofaLowSulfurCoalContract:MCAs;si>PSA 125 2-20ImpactofalowsulfurCoalContract:MCANSi
PAGE 11

........................ 195 3-7GivenScrubberChoice:ShiftfromHighSulfurContract ............. 196 3-8GivenScrubberChoice:ShiftfromLowSulfurContract ............. 196 3-9WithScrubberChoice:ShiftfromHighSulfurContract ............. 197 3-10WithScrubberChoice:ShiftfromLowSulfurContract .............. 198 B-1UpperSemi-ContinuousCorrespondence ...................... 210 B-2CorrespondencewithScrubberChoice ....................... 210 11

PAGE 12

Environmentalissueshavebecomeakeypoliticalissueoverthepastfortyyearsandhasresultedintheenactmentofmanydierentenvironmentalpolicies.Thethreeessaysinthisdissertationaddtotheliteratureofrenewableenergypoliciesandsulfurdioxideemissionstrading. Therstessayascertainswhichstatepoliciesareacceleratingdeploymentofnon-hydropowerrenewableelectricitygenerationcapacityintoastateselectricpowerindustry.Aswouldbeexpected,policiesthatleadtosignicantincreasesinactualrenewablecapacityinthatstateeithersetaRenewablesPortfolioStandardwithacertainlevelofrequiredrenewablecapacityoruseCleanEnergyFundstodirectlyfundutility-scalerenewablecapacityconstruction.AsurprisingresultisthatRequiredGreenPowerOptions,apolicythatmerelyrequiresallutilitiesinastatetooertheoptionforconsumerstopurchaserenewableenergyatapremiumrate,hasasizableimpactonnon-hydrorenewablecapacityinthatstate. Thesecondessaystudiesthetheoreticalimpactsfuelcontractconstraintshaveonaelectricitygeneratingunit'scompliancecostsofmeetingtheemissionscompliancerestrictionssetbyPhaseIoftheTitleIVSO2EmissionsTradingProgram.Fuelcontractconstraintsrestrictautility'sdegreesoffreedomincoalpurchasingoptions,whichcanleadtotheuseofamoreexpensivecomplianceoptionandhighercompliancecosts. 12

PAGE 13

13

PAGE 14

1 14

PAGE 15

Threedistinctlydierenttypesofpoliciesarefoundtobeeectiveatexpandingnon-hydrorenewablecapacitydeployment:acommand-and-controlpolicyknownasaRenewablesPortfolioStandard(RPS),atax-and-subsidyschemefacilitatedthroughaPublicBenetsFund(PBF)orCleanEnergyFund(CEF),andamarket-basedpolicywhereconsumerscanexpresstheirpreferencestobuypowerfromrenewableresourcesatapremiumprice. Thecommand-and-controlpolicytargetstheutilitybymandatingaspeciedlevelofcapacitythatmustcomefromrenewableenergy,andisgenerallyreferredtoasaRenewablesPortfolioStandard.Thetax-and-subsidyschemecollectsanadditionalchargeperunitofelectricityconsumedfromallcustomersinastateandplacestheproceedsintothisPublicBenetsFundorCleanEnergyFund.MoniesfromthePBF/CEFareusedto 15

PAGE 16

TheeconometricresultssupportmanyoftheconclusionsfromvariouscasestudieswithrespecttoRenewablesPortfolioStandardandCleanEnergyFundpolicies.Moreover,theresultspresentedherealsoshow,unlikepreviousstudies,thatthepotentialforoeringconsumerstheoptiontopurchaserenewableelectricityatahigherpricethanconventionallyproducedelectricitycanincreaserenewablecapacityinastate. Wiseretal.(2004)consideredallRenewablesPortfolioStandardsandfoundthepitfallsinthecurrentpolicydesigns.Somekeyproblemsinpolicydesignsincludeinsucientdurationandstabilityoftargets,weakenforcement,andnarrowapplicabilityofthepolicy.Otherconditionsthatmayimpactapolicy'seectivenessarethepresenceoflong-termpowerpurchasersandpoliticalandregulatorystability. 16

PAGE 17

Chenetal.(2007)comparestheresultsfrom28policyimpactprojectionsforstateorutility-levelRenewablesPortfolioStandardsandndsthat(1)theimpactonelectricitypricesisminimal,(2)windpowerisexpectedtobetheprimaryrenewableusedtomeetpolicyrequirements,and(3)thebenet-costestimatesrelyheavilyonuncertainassumptions,suchasrenewabletechnologycosts,naturalgasprices,andpossiblecarbonemissionspolicyinthefuture. Bolingeretal.(2001)describeindetail14dierentstateCleanEnergyFunds,enumeratingtheregulatorybackground,fundingapproaches,thecurrentstatusofthefund,andtheresultingimpactsonrenewableenergy.Programsthatfundutility-scaleprojectsarefoundtobethemosteectiveatincreasingrenewablecapacitydeployment. WiserandOlson(2004)examineparticipationin66utilitygreenpowerprograms.Theyndlocalgreenpowerprogramshaveresidentialparticipationratesrangingfrom0.02%to6.45%andaveraging1.39%.However,thisstudydoesnotlookatanystate-levelRequiredGreenPowerOptionsthatrequireallutilitiesinastatetooerconsumerstheoptiontopurchaserenewableenergy.Thepaperfocusesonparticipationratesof 17

PAGE 18

Birdetal.(2005)summarizefederalrenewableenergypolicies,generalmarketfactors,andstate-specicfactors,suchasstatepolicies,thataredrivingthedeploymentofwindpower.Thekeymarketfactorsarethevolatilityinnaturalgaspricesduringtheearly2000sandtheloweredwindenergygenerationcostsduetolargerwindturbines,whichhavecombinedtomakewindpowermorecompetitivewithnaturalgas-redgeneration. Onlyonepaperhasattemptedtoeconometricallyestimatetheeectsofstaterenewableenergypolicyonrenewablecapacity.MenzandVachon(2006)useordinaryleastsquarestoestimatestatepolicyeectsonwindpowercapacityandgenerationwithapaneldatasetfor39statesfor1998-2002whilecontrollingforwindpoweravailability,retailchoice,andpolicydummyvariablesforPublicBenetsFund,RenewablesPortfolioStandard,RequiredGreenPowerOption,andfuelmixdisclosure. 18

PAGE 19

Thedependentvariableisthetotalnon-hydropowerrenewablenameplatecapacityintheelectricpowerindustry(Cit),whichincludesallnameplatecapacityofutilities,independentpowerproducers(IPPs),andindustrialorcommercialcombinedheatandpowerproducersthatusesolar,wind,geothermal,orbiomassasanenergysource. 19

PAGE 20

Theeectsofstaterenewableenergypoliciesarebestestimatedusingtotalstatenon-hydrorenewablecapacityasthedependentvariablebecauseseveralpoliciesmandateorfundaspecicamountofrenewablecapacity.Policiesthatdonotsetspecicrenewablecapacityrequirementscanbemeasuredincapacitytermsbycontrollingforeachstate'smarketsize,whichwillbediscussedinmoredetailinSection4. Alargeamountofrenewablecapacitycreatedbefore1996originatedfromthePublicUtilitiesRegulatoryPolicyAct(PURPA),afederalpolicypassedin1978requiringutilitiestopurchaseelectricityfromQualifyingFacilities(QFs),whichareIPPsthatmeetspecicrequirementsandincluderenewable-basedfacilities.Foravarietyofreasons,theeectsofPURPAvariedfromstatetostate.Statedummyvariables(Si)measuretheseeectsandotherunchangingstatefactors,suchasrenewableresourceavailability. 7 20

PAGE 21

1.4.1EconomicandPoliticalVariables:Wit 1-1 summarizesthedataforthedependentvariable(RENEWABLECAPACITY)andthecontrolvariables. withoutanyconsiderationofcostsorenvironmentalimpacts.AnycapacityshutdownduetoPURPAcontractexpirationafter1996willdecreasethepositiveeectsofanyenactedpolicy.ThereisalsothepossibilityofastatechangingitsinterpretationandenforcementofPURPAafter1996,whichwouldnotbecapturedinthemodel.9 21

PAGE 22

Thefollowingthreevariablesareincludedinthemodeltocontrolformarketstructure.Twoofthesevariablesarehydropowercapacity(PCTHYDROPOWER)andnuclearpowercapacity(PCTNUCLEAR)asapercentageoftotalcapacityexcludingnon-hydrorenewables.Hydropowershouldleadtolessnon-hydrorenewablecapacitybecausehydropowerhaslowmarginalproductioncosts,andthecapacitytypicallywasconstructedmanyyearsago.Withlowermarginalcostsandsunkcapitalcostsassociatedwithhydropower,hydropowerwillbetherstrenewableenergytobeimplementedbecauseitismoreeconomicallycompetitivethanmostnon-hydropowerrenewablesavailabletotheelectricpowerindustry.Consumerand/orpolicydrivendemandforrenewable-basedelectricitymaynotdierentiatebetweenhydropowerandotherrenewablesources,whichallowshydropowertobeasubstituteofnon-hydrorenewables. Similartohydropower,nuclearpowerhaslowmarginalcostsofproducingbaseloadelectricity,hassunkcapitalcosts,andhasnoemissions.Ifnon-hydrorenewablecapacityisdeployedbasedoneconomicfactors,givensimilaremissionsproles,greaternuclearorhydropowercapacityshoulddecreasetheamountofnon-hydrorenewablecapacity. Analternativepossibilityisthatregulatorsinstateswithlargeamountsofnuclearpowerencouragepowerproducerstouseotherresourcetypestomeetnewdemand.Renewableenergymaybeusedbyutilitiestoalleviatepressurefromenvironmentalistsovernuclearpower,thusleadingtogreaterdeploymentofrenewableenergycapacityinstateswithlargeamountsofnuclearcapacity.ThesignofPCTNUCLEARwilldependonwhichofthesetwofactorshasthelargereectonpowerproducers. 22

PAGE 23

23

PAGE 24

16 24

PAGE 25

25

PAGE 26

Florida,Hawaii,Louisiana,andTexasusethebyproductofsugarproductionfromsugarcaneasabiomassfuel.Forexample,inHawaiisugarcaneisoneoftheprimarysourcesofbiomass.DuetomarketconditionsmostofthesugarcanefarmsinHawaiiwereshutdownoverthe1990s,removingthefuelsourceformuchofthebiomasscapacityinthestate.Changesinsugarcaneproductionarelikelytohaveanimpactontheamountofbiomasscapacityinastate.Thechangeintotaltonsofsugarcaneproductionfrom1996levels(SUGARCANEPRODCHANGE)isincludedinthemodeltocontrolforitsimpactonrenewablecapacity.SUGARCANEPRODCHANGEistheonlycontrolvariablenotinteractedwithgeneration. Apoliticalvariableisincludedtomeasurechangesinrenewableenergypreferencesinastate.TheLeagueofConservationVoters(LCV)ratingisusedtodetermineifpolicypreferencesforenvironmentalprotectionincreaserenewableenergycapacityindependentfromitspolicyeects.TheLeagueofConservationVoters(LCV)annuallypublishestheNationalEnvironmentalScorecard,whichratesallcongressionalvotesonconservationalissuesbyeachrepresentative. Anaverageofallthevotesbyastate'srepresentativesistakentogettheaverageHouseofRepresentativesscore(LCVSCORE).ThescoresfromtheHouseofRepresentativesareusedinsteadoftheSenatebecauserepresentativeshaveashorterterminoce 26

PAGE 27

AhighLCVratingforastateindicatesthatthestate'sconstituentsareenvironmentallyfriendlyandaremorelikelytodemandelectricityfromrenewableenergy,allotherthingsbeingequal.ConsumersorenvironmentalgroupsinstateswithhigherLCVratingsmaybemorelikelytopressureutilitiestousegreateramountsofrenewableenergynomatterwhich,ifany,policieshavebeenenactedbythestate. PoliciesmaybeendogenoustohigherLCVratingsbecausestateswithcongresspersonswhovoteforfederalpro-environmentalpoliciesmaybemorelikelytoenactstatepro-environmentalpolicies.ThepolicyendogeneityissueisnotaddressedinthebodyofthispaperbecauseLCVSCOREisnotastrongenoughpredictorofstatepoliciestobeasatisfactoryinstrument.NotealsothatremovingLCVSCOREfromtheregressiondoesnotchangetheotherresults. 27

PAGE 28

RegulatorypoliciesdescribedbelowincludeaRenewablesPortfolioStandardwithaCapacityRequirement,RenewablesPortfolioStandardwithaGeneration/SalesRequirement,CleanEnergyFund,NetMetering,InterconnectionStandards,StateGovernmentGreenPowerPurchasing,andRequiredGreenPowerOptions.Table 3-2 summarizesthedataforthepolicyvariables. TherstpolicythatwillbediscussedisaRenewablesPortfolioStandard,whichspeciesanamountofastate'selectricityproduction,sales,orcapacitythatmustberenewable-based.RenewablesPortfolioStandardscanbedierentiatedintothreemainstructuralforms,policiesthatset(1)mandatoryrenewablegenerationorsaleslevels,(2)voluntaryrenewablegenerationorsalesgoals,and(3)mandatoryrenewableenergycapacityrequirements. ThersttypeofRenewablesPortfolioStandardsetsapercentageoftotalgenerationorsalesforeachpowerproducer/retailerthatmustoriginatefromrenewablesources,usuallyincreasingeveryyearoreveryfewyears.Forexample,Arizona'stieredrenewablelevelsthathavetobemetbeganat0.2%in2001andincreasedby0.2%eachyear, 28

PAGE 29

ThedierencesbetweenRenewablesPortfolioStandardscanbeaccountedforinthemodelbytwovariables:avariablethatmeasuresthesizeoftherenewablegenerationorsalesrequirement(RPS:SALESREQ)andavariablethatmeasuresthesizeofthecapacityrequirement(RPS:CAPREQ). 29

PAGE 30

Althoughthispolicydoesnotdirectlyrequiretheconstructionofrenewablecapacity,anincreaseintherequiredamountofrenewablegenerationmayleadtoaneedformorerenewablecapacity.Ifcurrentlevelsofrenewablecapacitycannotmeetafuturegeneration/salesrequirement,additionalcapacitywillneedtobeconstructed. CleanEnergyFundsarepaidforthroughSystemBenetsCharges(SBCs),whichareadditionalchargespaidbyallconsumersontheirelectricityconsumption.SBCscanbe 30

PAGE 31

31

PAGE 32

Interconnectionstandards(INTERCONSTANDARDS)areasetofguidelinesusedtosafelyandeectivelyconnectindividualrenewablegeneratingunitstotheelectricutilitypowergrid.Somehavetechnicalrequirements,suchasgeneratortypeandsizelimits,mandatorysafetyandperformancestandards,andinsurancerequirementsthatmustbemetbeforeanetmeteringcustomercanconnecttotheutility'snetwork.Interconnectionstandardsmustbemetbyanycommercial,industrial,residential,orgovernmentcustomerthatdecidestoconnecttothegrid.Withoutthesestatepolicies,thenetmeteringconnectionscouldcausemajorproblemsforthegrid,powerproducers,andotherpurchasers.Interconnectionstandardsincreasethecostsofhookinguptothegridfornetmeteringandmayosetsomeofthenegativeeectfromnetmetering.INTERCONSTANDARDSisalsointeractedwithGENtocontrolformarketsize. 32

PAGE 33

ARequiredGreenPowerOptionrequiresutilitiestooercustomerstheoptiontopurchaserenewablepoweratapremium.Therearetwoversionsofhowtheseoptionsareimplemented.Themostcommontypegivesconsumerstheoptiontomakevoluntarycontributions,calledvoluntaryrenewableenergytarisinreturnfortheguaranteethatsomeoftheconsumer'selectricityconsumptionisproducedfromrenewablesources.Consumerspurchaseelectricityatthemarketpriceandthenpayapremiumforblocksofgreenelectricity,usuallyabout$2per100kWh.Thesecondtypeallowstheproducerstochargeconsumersahigherrateperkilowatt-hour,butonlytocovertheadditionalcostsforelectricityfromrenewablesources.Boththepremiumblockrateandpremiumperkilowatt-hourratemustbeapprovedbythestate'sPublicUtilitiesCommission(PUC). RequiredGreenPowerOptionselicitcustomerpreferencesandacrudemeasureofwillingnesstopayforrenewableenergybyallowingconsumerstovoluntarilypayhigherpricesfortheknowledgethattheyaresupportingrenewable-basedelectricity.Thecreationofthisnichemarketforrenewableenergygenerationshouldhaveapositiveimpactonrenewablecapacity.ThevariableREQGREENPOWEROPTisadummyvariable,whichisinteractedwithGENinthemodeltomeasuretheeectofthepolicybasedonthestate'smarketsize(REQGREENPOWEROPT*GEN). 33

PAGE 34

Thepercentageofothercapacitycomprisedofhydropowerinteractedwithgeneration(PCTHYDRO*GEN)isnotstatisticallysignicant.However,thecoecientforthepercentageofothercapacitycomprisedofnuclearpowerinteractedwithgeneration(PCTNUCLEAR*GEN)ispositiveandstatisticallysignicant.Aonestandarddeviation(12.46%)increaseinthepercentageofnon-renewablecapacitycomprisedofnuclearpowerleadstoanincreaseof2.09MWperterawatt-hourofgenerationinastate.Sothisonestandarddeviationchangeinastatewithamediangenerationlevel(51.15TWh)leadstoanincreaseof107MW.Itispossiblethatutilitieswithmorenuclearpoweraredeployingmorerenewablecapacitybecausetheutilitiesarefocusedondiversifyingitsgenerationmix,eithertodecreasetheutilities'useoffossilfuelsandloweremissionsortoalleviatepressurefromenvironmentalistswhoareupsetabouttheuseofnuclearpower. ThecoecientsontheaverageLCVscorefortheHouseofRepresentativesinteractedwithgeneration(LCVSCORE*GEN)arepositiveandsignicant.Aonestandarddeviationincrease(26.51points)inastate'sLCVscoreleadstoanincreaseof0.663MWperterawatt-hourofgeneration.Aonestandarddeviationincreaseinastatewithmedian 34

PAGE 35

Asexpected,renewableenergycostinteractedwithgeneration(RENEWCOST*GEN)hasanegativeandstatisticallysignicantcoecient.Aone-centperkWhdecreaseinrenewableenergycostleadstoanincreaseof0.712MWperterawatthourofgeneration.Inastatewithmediangeneration,aonecentdecreaseinRENEWCOSTleadstoanincreaseof36MW.RENEWCOSTdecreasedby1.79centsfrom1996to2003,whichimpliesanincreaseof65MWforastatewithmediangeneration.Thiseectdoesnotjustincludethetechnologicalchangesinrenewableenergy.AsmentionedinSection4,allfederalproductionincentivesareincludedinthecostsofproductionforeachrenewablesource,capturingthefederalpolicychangesaswellasthetechnologicaladvances.Theyearvariablecoecients,whichexplainthesameimpactsasRENEWCOST,areexplainedindetailinSection5.4. Thecoecientonaverageborderstateretailpriceinteractedwithgeneration(BORDERPRICE*GEN)isnegativeandismarginallystatisticallysignicantonlyinSpecication3ofTable3.Higherelectricitypricesdonotappeartoresultinmorerenewableenergycapacityconstruction.Infact,thenegativecoecientimpliesthatanincreaseofonecentinthepriceofelectricityleadstoasmalldecreaseinrenewablecapacityby13MWinastatewithmediangeneration.Aonestandarddeviation(2.07cents)increaseinpriceleadstoadecreaseofonly27MW.Itispossiblethatconsumersinastatewithhighelectricitypriceshavelessofanappetiteforfurtherincreasesinpricesthroughmoreexpensiverenewablegeneration. 33 35

PAGE 36

ToalleviateanyconcernsaboutFUELCOST,anadditionalspecicationisestimatedreplacingFUELCOSTwithvevariables:averagecostofcoal,averagecostofoil,averagecostofnaturalgas,percentofnon-renewablecapacitycomprisedofcoal,andpercentofnon-renewablecapacitycomprisedofnaturalgas.Naturalgasandcoalcapacityaretreatedinthesamemannerashydropowerandnuclearpowercapacityinthemodel.EachvariableisinteractedwithGEN,justliketheothercontrolvariables. Ahigherpercentageoftotalnon-renewablecapacitycomprisedofcoaliscorrelatedwithmorerenewablecapacity,whichgivessomesupportthatstateswithdirtierconventionalcapacityusemorerenewablecapacity.Asthepriceofcoalincreases,lessrenewablecapacityisconstructed.Thesetworesultssupporttheideathattheemissionsrequirementsutilitiesmustmeetareadrivingforcetorenewabledeployment,whileeconomiccompetitivenessinthemarketdoesnothavemuchofanimpact.Cautionisnecessaryininterpretingtheresultswiththeadditionalsetoffuelvariablesbecausetherearemanymissingobservationsthatmustbeextrapolated.Thedummyvariablesthatcontrolformissingobservationsforcoalandnaturalgasarebothstatistically 36

PAGE 37

ThecoecientonSUGARCANEPRODCHANGEisinsignicant.Themissingfuelcostdummyvariableinteractedwithgeneration(MISSINGFUELCOST*GEN)controlsforanymeasurementerrorcausedbytheextrapolationofthe38missingdatapointsandisinsignicantaswell. CEF:CAPFUNDED,whichmeasurestheamountofcapacitythatthefundhasagreedtohelpnance,hasamarginallystatisticallysignicantcoecient.Thisincludescapacitythathasbeenagreedupon,buthasnotyetbeenbuilt,eitherbecausetheprojecthasnotbeennishedortheprojectislatercanceled.ForeachmegawattofcapacitythattheCleanEnergyFundhasfundedoragreedtofundinthenearfuture,approximately0.206MWhasbeenconstructed.Thisisnotsignicantlydierentthanthefractionofcapacitythathasactuallybeenconstructedasof2003,whichwas0.33MWper1MW.Eventhoughactualrenewablecapacityisprobablynotconstructedlinearlyoverthelifetimeofthepolicy,theestimatesfromthelinearinterpolationseemtoberepresentativeofactualcapacityconstructionduetotheCleanEnergyFunds. 37

PAGE 38

ThemostinterestingresultfromthemodelistheeectthatRequiredGreenPowerOptionshaveonrenewablecapacity.ThecoecientforREQGREENPOWEROPT*GENhasapositiveandstatisticallysignicantcoecientandhassomeofthelargesteectsonrenewablecapacityofanyvariable,whereenactmentleadstorenewablecapacityincreasingby3.46MWperterawatt-hourofgeneration.Astatewithamediangenerationlevel(51.15TWh)thatenactsaRequiredGreenPowerOptionwouldhaveanincreaseof177MW.WashingtonhasthelargestelectricitymarketofstatesthathaveenactedaRequiredGreenPowerOption(100.1TWh),whichwouldleadtoanincreaseinrenewablecapacityof346MW.Togivesomeperspectiveontheseresults,theestimatedimpactsintotalrenewablecapacitycanbeexpressedintermsofapercentageoftotalcapacityinastate.Dependingonthestate,aRequiredGreenPowerOptionleadstoanincreaseofabout1.2-1.6%. 34 38

PAGE 39

TherearemultiplereasonsfortheinsignicanceofRPS:SALESREQ*GEN.First,mostRenewablesPortfolioStandardswithsalesrequirementshavebeenenactedfairlyrecentlyandcurrentlyhavelowrequirements.Powersuppliersmaybeabletomeettheirlowinitialrequirementsbyusingcurrentrenewablecapacity.Second,someofthestatesimplementingthesepolicieshaveavailablehydropowercapacityalreadyinplacethatisconsideredeligibletomeetaportionofthesecurrentlylowmandates,whichdecreasesthepolicies'eectivenessinencouragingnewnon-hydrorenewablecapacitydeploymentinastate.Third,somestatesallowthepurchaseofRenewableEnergyCredits(RECs),whicharecerticatesthatrepresenttheenvironmentalrightsofrenewableelectricity,insteadofactualgeneration.Manyofthesesamestatesallowpowerproducers/retailerstopurchaseRECsfromout-of-statepowerproducerstomeettheirin-staterequirements.Allthreeofthesefactorswilldecreasethepolicy'simpactsonin-staterenewablecapacitydeployment.AfewmoreyearsofdatashouldresultinRPS:SALESREQ*GENtohavestatisticallysignicantimpactsonrenewablecapacityinitsstate. Neithernetmeteringnorinterconnectionstandardsappeartohaveanimpactonrenewablecapacity.ThecoecientsonNETMETERINGandINTERCONSTANDARDSarestatisticallyinsignicant. 39

PAGE 40

40

PAGE 41

Threeregulatorypoliciesappeartobeeectiveatincreasingrenewablecapacitydeploymentinastate.Thesignicantresultsfromtheseregulatorypoliciesconrmmanyofthendingsfrompriorcasestudies,whichndRenewablesPortfolioStandardswithCapacityRequirementsandCleanEnergyFundshaveincreasedrenewablecapacity.Anadditionalpolicy,MandatoryGreenPowerOptions,isalsofoundtoincreasecapacitydeploymentinastateaswell. ThepreviousempiricalstudyfoundPublicBenetsFunds,whichincludeanyCleanEnergyFundinastate,tobeinsignicantintheirmodel.MypaperndsthatCleanEnergyFundswithutility-scaleprojectsincreasethedeploymentofrenewablecapacityinastate.ByusingSystemBenetsCharges(SBCs)astatecaneectivelymakeconsumerspayforcleanerenergywithoutcreatingadierentmarketforrenewableenergydemand.SimilartothecasestudyndingsbyBolingeretal.(2001,2004,2005),largerutility-scaleprojectsmakeCleanEnergyFundsmoreeectiveatincreasingrenewablecapacitydeploymentinastate. ThispaperndsthatdierenttypesofRenewablesPortfolioStandardshavedierenteectsonrenewablecapacity.EachmegawattofcapacitymandatedbyRenewablesPortfolioStandardswithCapacityRequirementsresultsinthedeploymentofonemegawattofadditionalrenewablecapacityinastate.ButrecentRenewablesPortfolioStandardsthatmandategenerationorsaleslevelsappearnottohavestatisticallysignicanteects.TheseresultsmirrorPetersik'scasestudyinthatonlyRenewables 41

PAGE 42

StatewideRequiredGreenPowerOptionsappeartohavebeenaseectiveasanyotherpolicy.Forcingutilitiestooercustomerstheoptiontopurchaserenewable-basedelectricityatareasonablepremiumratedrasticallyincreasesrenewablecapacityinastate.Thepolicyhasagreaterimpactinlargerelectricitymarketsandappearstobeeectiveregardlessofastate'spoliticalenvironment. Therearemajorrenewablepolicyimplicationsiftheseresultsholdwhenadditionalyearsareeventuallyincludedinthemodel.OnlyvestateshavecurrentlyimplementedRequiredGreenPowerOptionseventhoughcreatingastatewidegreenpowermarketappearstobeaseectiveatincreasingrenewableenergycapacityinastateasacommand-and-controlschemeofaRenewablesPortfolioStandardsortax-and-subsidyschemeofaCleanEnergyFund.Stategovernmentpurchasingagreementsofrenewableenergyappeartobenomorethanwindowdressingforpoliticianstoshowtheirsupporttotheenvironmentalcommunity,andadditionalfundingtorenewablepowerproducers. Theremainingpoliciesinthemodeldonotappeartoimpactrenewableenergycapacityconstructioninthetheelectricpowerindustry.Stategovernmentgreenpowerpurchasingdoesnotincreaserenewableelectricitydemandenoughtodrivecapacityconstruction.Netmeteringandinterconnectionstandardstargetresidentialandcommercialcapacityanddonotimpactelectricpowerindustrydecisions. Theimportantpolicyimplicationsthatarisefromtheresultsindicatepolicymakershaveawidearrayoftoolsattheirdisposaltopromoterenewableenergydeploymentinastatetomeetenvironmentalandenergysecuritypolicygoals.ThearrayofpolicymechanismswillbecomeevenmoreusefultostategovernmentsiftheprospectofU.S.climatechange/carbonemissionspolicybecomesareality. 42

PAGE 43

DependentandControlVariables RENEWABLECAPACITY(MW)348.7827.60.006177.4178.5GEN(TWh)73.8264.984.95385.6351.15PCTHYDROPOWER(Percentage)14.1420.680.0091.596.26PCTNUCLEAR(Percentage)11.1312.460.0056.207.54BORDERPRICE(2002cents/kWh)7.582.074.8214.496.68RENEWCOST(2002cents/kWh)6.930.5856.007.796.94FUELCOST(2002dollars/MMBtu)2.0860.9870.6017.4311.861LCVSCORE(0to100)43.0626.510.0100.038.0SUGARCANEPRODCHANGE88.95599.33-1707.04882.00.0

PAGE 44

RegressionsResults CEF:CAPFUNDED(MW)0.1980.1440.206(0.107)*(0.145)(0.124)*RPS:CAPREQ(MW)1.2451.1441.142(0.175)***(0.168)***(0.168)***RPS:EFFECTIVEGENREQ0.1530.1080.100(0.056)***(0.077)(0.071)PCTSTATEGREENPOWERPURCHASING*GEN0.0260.013-0.002(0.014)*(0.017)(0.016)REQUIREDGREENPOWEROPT*GEN3.5393.3183.457(0.667)***(1.207)***(1.166)***NETMETERING*GEN0.093-0.237-0.033(0.205)(0.214)(0.227)INTERCONSTANDARD*GEN-0.246-0.241-0.240(0.211)(0.195)(0.210)SUGARCANEPRODUCTIONCHANGE(Tons)-0.011-0.007(0.023)(0.023)GEN(1TWh)-0.184(1.150)FUELCOST($/mmBtu)*GEN-0.154-0.124(0.107)(0.109)FUELCOSTMISSING*GEN-1.261-1.167(0.843)(0.802)BORDERPRICE(Cents/kWh)*GEN0.199-0.259(0.207)(0.117)**RENEWCOST(Cents/kWh)*GEN-0.712(0.211)***LCVSCORE*GEN0.0240.025(0.011)**(0.011)**PCTHYDRO*GEN0.018-0.027(0.028)(0.024)PCTNUCLEAR*GEN0.1430.168(0.043)***(0.052)***YR1997*GEN-0.048(0.151)YR1998*GEN-0.142(0.186)YR1999*GEN-0.187(0.182)YR2000*GEN-0.269(0.199)YR2001*GEN0.322(0.247)YR2002*GEN0.250(0.199)YR2003*GEN0.530(0.226)**CONSTANT309.103413.745267.888(7.689)***(55.340)***(50.741)***Observations400400400StateFixed-Eects505050R-squared0.5980.6340.646 RobustStandardErrorsinParentheses;*signicantat10%;**signicantat5%;***signicantat1%

PAGE 45

PolicyVariables CEF:CAPFUNDED851RENEWABLESPORTFOLIOSTANDARD1466RPS:CAPREQ427RPS:SALESREQ639NETMETERING31187INTERCONSTANDARDS2190STATEPURCHASING:PCTREQ614REQGREENPOWEROPT516

PAGE 46

VariableEectsofSignicantVariables VariableMeanStd.Dev.IncreasePerUnitperTWhStd.Dev.inMedianGen.State RENEWCOST(2002cents/kWh)6.930.59-0.725MW-36MWBORDERPRICE(2002cents/kWh)7.582.07-0.235MW-27MWPCTNUCLEAR11.1312.460.170MW107MWLCVSCORE(0-100)43.0626.510.025MW34MW EectofEnactingaPolicyMeanStd.Dev.PerUnitEectImpactinMedianGen.State PBF:CAPFUNDED8.2061.210.206MWper1.0MW20.6%ofFundedCapacityRPS:CAPREQ14.0578.301.142MWper1.0MW114%ofReq.Cap.REQGREENPOWEROPT0.020.153.457MWperTWh177MW Note1:ResultsareforSpec.3excludingRENEWCOST,whichisfromSpec.2.Note2:Allothervariablesarestatisticallyinsignicant.

PAGE 47

47

PAGE 48

Inthispaperamodelofunit-levelSO2complianceisconstructedthatincorporatesthepresenceofcoalcontractstoexaminehowlong-termcoalcontractsaectutilitycompliancechoicesandaunit'scompliancecosts.Asexpected,thepresenceofcoalcontractconstraintsleadstocompliancecostsinexcessofthehypotheticalleast-costsolution.ThepresenceofbindinghighsulfurcontractconstraintsthatwerelikelyinPhaseIoftheTitleIVSO2ProgrammayexplainthelowerthanexpectedallowancepricesinPhaseIthataccompaniedcompliancecoststhatwereabovetheleast-costsolution.ItisalsofoundthatthepresenceofbindinglowsulfurcoalconstraintsthatmayexistunderCAIR,whichmayleadtoallowancepricesthatarehigherthanwithoutthebindingconstraint.Theeectsofthecontractconstraintseemcounter-intuitive:bindinghighsulfurcoalconstraintsleadingtolowerexcessdemandforallowances,whichcouldreducetheallowancemarketprice.Bindinglowsulfurcoalconstraintsleadingtohigherexcessdemandforallowances,whichcouldincreasetheallowancemarketprice.Theinteractionbetweenthecontractconstraintsandthediscretenatureofthescrubberchoiceleadstotheseunexpectedresults. 2.2.1TitleIVoftheCleanAirActAmendment 48

PAGE 49

Anadditional168unitsparticipatedinPhaseIin1996basedontherulesestablishedbyEPAallowingaplantto\opt-in"units(7units),designatesubstitutionunits(160units),ordesignatecompensatingunits(1unit)aspartoftheirPhaseIcomplianceplans.Thevoluntaryparticipationoftheseadditionalunitsresultedinatotalof431aectedgeneratingunitsunderPhaseI.A\substitutionunit"isaunitthatwouldeventuallybeaectedinPhaseIIthatvoluntarilyenrolledintoPhaseItomeetsomeoralloftherequiredemissionsreductionsforaPhaseIunit(SotkiewiczandHolt,2005).Substitutionunitsreceiveanallowanceallocationbasedonitshistoricalheatinput.AutilitymaydecidetoreduceitselectricityproductionataPhaseIaectedunit.Todoso,theutilitymusthavea\compensationunit"fromthePhaseIIunitstheutilityoperatestocoverthenecessaryadditionalelectricity.ThiscompensationunitisthenbroughtintoPhaseIandgivenanallowanceallocationbasedonitshistoricalheatinput.IndustrialsourcesofSO2emissionscouldusetheopt-inprovisionandvoluntarilyenrollintoPhaseIandreceive 49

PAGE 50

ThebankingprovisionhasallowedutilitiestotradeintertemporallywithutilitiesusingthesubstantialallowancebankaccumulatedthroughPhaseIforcomplianceinPhaseIIleadingtoannualemissionlevelsinexcessof10milliontonsineachyearfrom2000to2005. 50

PAGE 51

GeneratingunitsstillreceivetheirallowanceallocationasdenedunderPhaseII.Beginningin2010,theemissionsvalueoftheallowancesforunitsintheCAIRregioniscutinhalffrom1.2to0.6lbs.SO2/MMBtuofheatinputwhichimpliesaunitmustholdtwoTitleIVallowancesforeachtonofactualemissions.Startingin2015unitsmusthold2.86TitleIVallowancesforeverytonofemissions,whichtranslatestoanallocationofapproximately0.4lbs.SO2/MMBtu.MeanwhilegeneratingunitsoutsidetheCAIRregionwillcontinueoperateundertheTitleIV,PhaseIIProgramwithtradesallowedtotakeplacebetweenCAIRandPhaseIIunits.UnitsunderPhaseIIandCAIRparticipateinthesameallowancemarketandfacethesamemarketallowanceprice. Thespikeinallowancepricesuptoover$1,600in2006seeninFigure 2.8 mayhavebeenaresultoftheproposalandenactmentofCAIR.UtilitiescouldhavechosentobankallowancesinsteadofsellingallowancesinthemarkettoensuretheirabilitytocoverrequirementsatthebeginningofCAIR.Itisstilluncertainwhatledtothetemporaryspikeinallowanceprices,butitappearslikelythatitwasatemporaryreactiontothepolicyenvironmentorothermarketconditionsastheallowancepricehasquicklydecreasedtoitscurrentpriceoflessthan$600/ton. 51

PAGE 52

2.3.1TitleIV:PhaseI 2-1 ,itcanbeseenthat,ingeneral,thecompliancecostestimatesbeforePhaseItookeectwerehigherthantheestimatesmadeafterPhaseIbecameeectiveandactualdatacouldbeusedintheestimates.Thepre-policyestimatesrangeashighas$1.34billion/yearwithmostestimatesatleast$860million/year.Theactualcostestimatesaretowardsthelowerendofthisrangebetween$730-$990million/year. Thereareseveralreasonsforthedierencesbetweeninitialestimatesandactualaggregateindustrycompliancecosts.Themostimportantfactorwasthedecreaseindeliveredlowsulfurcoalprices.Attheunitlevel,lowerlowsulfurcoalpricesdecreasedthemarginalcostofreducingemissionsthroughfuelswitching,whichwasthecomplianceoptionchosenby52%ofallaectedunits,while32%ofaectedunitschosetopurchaseallowances,10%installedascrubber,3%shutdown,and3%choseothermethods. Keohane(2002)simulateswhichgeneratingunitswouldhaveinstalledscrubbersunderauniformemissions-ratestandardandndsthatthetotalnumberofscrubberswouldhavebeenone-thirdhigherthantheactualnumberofinstalledscrubbersunderthe 52

PAGE 53

IntheinitialyearsofPhaseI,manyrmswerenotactiveparticipantsintheallowancemarket,choosingtoswitchfuelsandbankallowancesorshiftallowancesbetweenonlytheirownunits(Hart(1998);Ellermanetal.(1998)).Thermsthatdidparticipatemainlytradedallowanceswithinthesameutilitycompany.BohiandBurtraw(1997)ndthatintra-utilitytradingaccountsfortwo-thirdsoftheallowancetransactionswhiletheremainingone-thirdwereinter-utilitytrades.Sincemosttradesweremadebetweenunitsownedbythesamecompany,tradingbetweentwogeneratingunitsatthesameplantwouldbeacommonoccurrence.Manystudiessuggestedstatepublicutilityregulationsandotherstatelawsasareasonfortheinecienciesresultingfromthisself-sucientbehavior(Bohi(1994);BohiandBurtraw(1997);Swift(2001)). Arimura(2002)useseconometricapproachestostudytheimpactofPUCregulationoncompliancechoices,andndsthatutilitiesthatfacePUCregulationaremorelikelytoswitchfuelsinsteadofpurchasingallowancesforcompliance. Winebrakeetal.(1995)estimatedthecostinecienciesfromstategovernmentrestrictionsonautility'sallowancetrading,andestimatesthetotalcostestimatesforthersttenyearsofTitleIV(1995-2005).Acommand-and-controlapproachwasestimatedtoresultincompliancecostsof$4.19billiongreaterthanintheunrestrictedpermittradingsystem($5.02billion,oranaverageof$502million/year)andanestimatedallowancepriceof$143/ton. Winebrakeetal.(1995)simulatestheadditionalcostsfromrestrictionsonbetween-statetradingthatwereunderconsiderationbybothNewYorkandWisconsin.Bothstateswere 53

PAGE 54

SomestudiesfurtherexplaintheinecienciesbyexaminingtheactuallostcostsavingsthatarespecicallyaresultofstatePUCregulationunderPhaseI.Carlsonetal.(2000)ndthattheactualcompliancecostswere$339million(59%)greaterin1996thantheleast-costsolution.Thestudyconcludesthedierencebetweenactualcompliancecostsandtheleast-costcompliancemaybeattributableto\adjustmentcostsassociatedwithchangingfuelcontractsandcapitalexpendituresaswellasregulatorypolicies."SotkiewiczandHolt(2005)ndthatPUCregulationsresultedin$131millionoftheadditionalcompliancecostsrelativetotheleastcostsolution.However,thereisasignicantamountofcompliancecoststhatremainedunexplained. 54

PAGE 55

Swinton(2004)followsthesameapproachexceptitexpandsthestudyto40plantswithdatafrom1994to1998,andintroducesthepossibilitythatlong-termcoalcontractsmayprohibitutilitiesfromswitchingcoaltypes,althoughitisnotmodeled.Bothstudiesndtheactualutility-levelcompliancecoststobemuchhigherthantheestimatedleast-costsolution. CogginsandSwinton(1996)useanoutputdistancefunctiontoestimatetheshadowpriceofSO2emissionsabatementforelectricpowerplantsinWisconsin.Thestudyestimatedtheallowanceshadowpricetobegreaterthantheobservedallowancepricesatthetime,whichtheyassertmaybepartiallyexplainedbyWisconsin'sstrictstateregulationsonSO2emissions. SeveralstudieshaveshownanalyticallyorthroughsimulationmodelsthatstatePUCregulationscanleadtoinecienciesattheutility-level.BohiandBurtraw(1992)developamodelofutilitydecision-makinggiventwocomplianceoptions,purchasingallowancesorinstallingemissionscontroltechnology.BohiandBurtrawderivetworecommendationssothatstateregulationdoesnotresultininecientcompliancechoicesbyautility.First,ifautility'sallowedreturnislessthanitscostofcapitalwithrespecttobothcomplianceoptions,symmetricalcostrecoveryrulesarerecommendedasuneventreatmentofcostrecoverymaycreateincentivesforautilitytomakesuboptimalcompliancechoices.Second,ifautilityisallowedtoearnmorethanitscostofcapitalwithregardtoboth 55

PAGE 56

Joskow(1988,1990)ndsthatduringperiodsinwhichthespotmarketcoalpriceswerelowerthanthecontractedprices,thecontractpricesfailedtoadjustdownward.Thisdownwardrigidityofcoalpricescanleadtoutilitycoalcostsbeinghigherthanisoptimalintheshortrun.Somerenegotiation,breachofcontract,andlitigationhasoccurred,butnearlyallcontractsappeartohavecontinuedunchanged.Themainreasonfortheconstraintsinalteringthesecoalcontractsisthatlessthan15%ofcoalconsumedbyutilitiesissuppliedbyacoalcompanyownedbythesameutility(Joskow,1987).Firmshavehighlegalornegotiationcostsofbreakingacoalcontractwhentheagreementismadewitharmthathasnonancialtiestotheutility.Coalcontractsmayalsobearesultofregulationsprotectingthelocalcoalindustry(Arimura2002).Duetotheinabilityofcontractedcoalpricestodecreasewithspotmarketprices,largecoalprice 56

PAGE 57

57

PAGE 58

Third,underCAIRsomeutilitiesmaybelockedintolowsulfurcoalcontractsenteredintoforTitleIVcomplianceandmaybeunableswitchbacktohighsulfurcoalandscrubifitiscost-eectivetodoso.Consequently,thecontractconstraintpushesautilityintosub-optimalcompliancechoicessuchasallowancepurchasesorscrubberinstallationwhileusinglowsulfurcoal. AnexaminationofthedataforPhaseIaectedunitsindicatesthatofthe26scrubbersinstalledbytheendof1996inresponsetothepassageofthe1990CleanAirActAmendments,23ofthosescrubberswereinstalledatfacilitieswith40percentormoreofitscoaldeliveriesbycontractandwith20ofthemhavingaweightedaverageSO2emissionratesgreaterthanthePhaseIallowedlevelof2.5lbs.SO2/mmBtu(poundspermillionBtusofheat). 58

PAGE 59

12 59

PAGE 60

60

PAGE 61

61

PAGE 62

62

PAGE 63

subjectto...Aei+Ai(1ziri)(m)(CsihSsih+CcihScih+CsilSsil+CcilScil)i1 (CsihHsih+CcihHcih+CsilHsil+CcilHcil)Dii2 (2{6) Equation( 3{1 )representstheunit'scostfunction.Thesecostsincludethecostofscrubberinstallation(ziPiz),netcostsofallowancepurchases(PAAi),andcostsofcoalpurchases(PsihCsih+PsilCsil+Pcih 3{2 ),wherethenumberofallowancesheld(Aei+Ai)mustbeaslargeastheamountoftotalemissionsbythegeneratingunit[(1ziri)(m)(CsihSsih+ 3{3 ).TheLagrangemultiplieronthedemandconstraintisrepresentedbyi2.Coalcontractconstraintsrequiretheunittouseaspecicamountofeachcontractcoaltype, 3{4 )and 3{5 ).Aunitwilluseexactlythecontractedamountbecause(1)ifthecontractcoalismoreexpensivethanspotmarketcoal,thena 63

PAGE 64

SinceAicanbeeitherpositiveornegativebasedonthenetallowanceposition,( 2{8 )willholdwithequality.Theadditionalcosttothermofemittingonemoretonofemissionsisequivalenttotheallowanceprice,i1=PA. Letf2fh;lgrepresentthetypeofcoalandg2fs;cgrepresentthetypeofpurchase.Eachcoaltypehasitsownsulfurcontent(Sgif),heatcontent(Hgif),anddeliveredprice(Pgif).ThepartialderivativeswithrespecttoCgifrepresenttheimpactaoneunitchangein\f"typesulfurcoal(highorlow)froma\g"typepurchasingagreement(spotmarketorcontract)hasontheunit'stotalcosts. ThecostofusingonemoreunitofCgifcanbedisaggregatedintofourdierentcostchanges:Pgifistheadditionalcostsofpurchasingonemoreunitofcoal,i1(1ziri)(m)(Sgif)istheadditionalcostsoftheextraemissionsfromonemoreunitofcoal, 64

PAGE 65

3{37 )or( 2{10 )holdswithequality.Sincethecontractconstraintholdswithequality,( 2{11 )and( 2{12 )alwaysholdwithequality. 3{37 )and( 2{10 )holdwithequality,or ( 2{13 )canberearrangedtoisolatetheshadowpriceofallowancesoremissions,i1,toderivetheMarginalCostofAbatementfromSwitchingFuelsfromhighsulfurspottolowsulfurspotmarketcoal(MCAs;si)in( 2{14 ).Exploiting( 2{8 ),theallowancepriceequalsMCAs;si: (2{14) Theshadowpriceisequaltothedierenceinpriceperunitofheatdividedbythedierenceinemissionsperunitofheat. 65

PAGE 66

3{37 )holdswithequalitywhile( 2{10 )holdswithweakinequality,or ( 2{15 )canberearrangedtoshowtheallowancepriceisweaklylessthanMCAs;si: 2{10 )holdswithequalitywhile( 3{37 )holdswithweakinequality,or ( 2{17 )canberearrangedtoshowthattheallowancepriceisweaklygreaterthanMCAs;si: Assumeaunithasahighsulfurcoalcontractandusesonlylowsulfurspotmarketcoal(nohighsulfurspotmarketcoal).So( 2{10 )and( 2{11 )holdwithequality.Ithasbeenshownin( 2{17 )thatifaunitusesonlylowsulfurspotmarketcoal,PAMCAs;si: 66

PAGE 67

2{19 )andexploiting( 2{8 ),itcanbeshownthattheallowancepriceisequaltoMarginalCostofAbatementofswitchingfromhighsulfurcontracttolowsulfurspotmarketcoal(MCAc;si)plusanadditionaltermrepresentingthebenetsofmeetingthecontractconstraint,whichisweaklygreaterthanMCAs;si: From( 2{20 ),ifMCAs;siMCAc;si>0,thenih>0andhighsulfurcontractcoalismoreexpensivethanspotmarkethighsulfurcoal.IfMCAs;siMCAc;si<0,thenihcanbepositiveornegative.Ifih<0,thencontractcoalischeaperthanspotmarketcoalandtheadditionalcompliancecostsduetoemissionsdominatethesavingsfromthelowerfuelcosts.Ifih>0,thencontractcoalischeaperthanspotmarketcoalandthesavingsfromthelowerfuelcostsdominatetheincreasedcompliancecostsduetoemissions. Nowassumeaunithasahighsulfurcoalcontractandusesonlyhighsulfurspotmarketcoal(nolowsulfurspotmarketcoal).( 2{11 )and( 2{12 )holdwithequalitywhile( 2{10 )holdswithweakinequality.Also,ithasbeenshownthatwhenaunitusesonlyhighsulfurspotmarketcoal,PAMCAs;si.Usingthesameapproachasinthepreviouscase,itcanbefoundthattheallowancepriceisweaklylessthanMCAc;siplusanadditionaltermrepresentingthebenetsofmeetingthecontractconstraint: ThesignofihcannotbedeterminedbycomparingMCAs;siandMCAc;si,buttherst-orderconditionsforhighsulfurcontractandspotmarketcoalcanbeusedtodetermineitssign.Psih+i1(1ziri)(m)(Ssih)

PAGE 68

Iftheadditionalcosts,bothfromfuelcostsandemissions,ofusinghighsulfurcontractcoalismoreexpensivethanusinghighsulfurspotmarketcoal,thenih>0.Iftheadditionalcosts,bothfromfuelcostsandemissions,ofusinghighsulfurcontractcoalislessthanusinghighsulfurspotmarketcoal,thenih<0.Iftheadditionalcosts,bothfromfuelcostsandemissions,ofusinghighsulfurcontractcoalisisthesameasusinghighsulfurspotmarketcoal,thenih=0. Nowassumeaunithasahighsulfurcoalcontractandusesbothhighsulfurandlowsulfurspotmarketcoal.So( 2{10 )and( 2{11 )holdwithequality.Also,ithasbeenshownin( 2{15 )thatwhenaunitusesbothhighandlowsulfurspotmarketcoal,PA=MCAs;si: Bysolvingforihin( 2{24 ),thesignofihcanbedetermined.IfMCAs;siMCAc;si=0,thenih=0andhighsulfurcontractcoalhasthesamecostashighsulfurspotmarketcoal.IfMCAs;siMCAc;si>0,thenih>0andcontractcoalismoreexpensivetousethanspotmarketcoal.IfMCAs;siMCAc;si<0,thenih<0andcontractcoalischeapertousethanspotmarketcoal. Assumeaunithasalowsulfurcoalcontractandusesonlyhighsulfurspotmarketcoal.Ithasalreadybeenshownin( 2{16 )thatwhenaunitusesonlyhighsulfurspot 68

PAGE 69

3{37 )and( 2{12 )holdwithequalitywhile( 2{10 )holdswithaweakinequality.Usingthesameapproachasintheprevioussectionyields: Using( 2{26 ),thesignofilcanbedetermined.IfMCAs;siMCAc;si>0,thenil>0andlowsulfurcontractcoalismoreexpensivetousethanlowsulfurspotmarketcoal.IfMCAs;siMCAc;si<0,thenilcaneitherbenegativeorpositive.IfMCAs;siMCAc;si<0andil>0,thenthehigherfuelcostsdominatethelowercostsfromemissions.IfMCAs;siMCAc;si<0andil<0,thenthethelowercostsfromemissionsdominatehigherfuelcosts. Nowassumeaunithasalowsulfurcoalcontractandusesonlylowsulfurspotmarketcoal.So( 2{10 )and( 2{12 )holdwithequalitywhile( 3{37 )holdswithaweakinequality.Ithasalreadybeenshownin( 2{17 )thatwhenaunitusesonlylowsulfurspotmarketcoal,PAMCAs;si. AlthoughthesignofilcannotbedeterminedbycomparingMCAs;siandMCAs;ci.Therstorderconditionsforlowsulfurcontractandspotmarketcoalcanbeusedtodetermineitssign.Psil+i1(1ziri)(m)(Ssil) Iftheadditionalcosts,bothfromfuelcostsandemissions,ofusinglowsulfurcontractcoalismoreexpensivethanusinglowsulfurspotmarketcoal,thenil>0.Ifthe 69

PAGE 70

Nowassumeageneratingunithasalowsulfurcoalcontractandusesbothhighsulfurandlowsulfurspotmarketcoal.So( 3{37 ),( 2{10 ),and( 2{12 )holdwithequalityandPA=MCAs;si: Bysolvingforil,thesignofilcanbedetermined.IfMCAs;ciMCAs;si=0,thenil=0andcontractcoalisascostlytouseasspotmarketcoal.IfMCAs;ciMCAs;si>0,thenih>0andcontractcoalismoreexpensivetousethanspotmarketcoal.IfMCAs;ciMCAs;si<0,thenih<0andcontractcoalischeapertousethanspotmarketcoal. Thecostsofaunit'snetallowancepositionisthedierencebetweenageneratingunit'sinitialallowanceallocationanditsactualemissionsmultipliedbytheallowanceprice(PAAi). Thecostsofswitchingfuelsisthelargeroftwovalues:(1)totalcostsofactualcoalpurchases(PsihCsih+PsilCsil+Pcih 70

PAGE 71

Thelatterwillonlyoccurifitisweaklycheaperforthegeneratingunittouselowsulfurcoalwithouttheemissionsrestrictions(Psih Combiningeachofthethreecostcomponentsresultsinaunit'stotalnetcompliancecosts.Eventhoughthecontractcoalhasnodirectaect,thecontractswillindirectlyaectcompliancecoststhroughaunit'sallowancepositionandscrubberchoice.Proposition1showsthesucientconditionsunderwhichacoalcontractwilleitherincreaseordecreasecompliancecosts. (i) IfPA>MCAs;siandthesulfurtoheatcontentratioofhighsulfurcontractcoal(Scih (ii) IfPAMCAs;siandthesulfurtoheatcontentratioofhighsulfurcontractcoal(Scih (iii) IfPAMCAs;siandthesulfurtoheatcontentratiooflowsulfurcontractcoal(Scil (iv) IfPA
PAGE 72

3{2 ).From( 3{2 )and( 3{3 ),theminimumandmaximumexcessdemandforallowancescanbeformallyderived. IfPAMCAs;si,aunitwillusethemaximumamountoflowsulfurspotmarketcoal,whichcanbederivedfrom( 3{3 ): Theuseofalllowsulfurspotmarketcoalleadstotheminimumemissionslevel: 72

PAGE 73

2{37 )forCsilin( 3{3 )givesanexpressionfortheminimumallowanceexcessdemand,whichisthedierencebetweentheminimumemissionslevel(EMINi)andtheinitialallowanceallocation(Aei): Ifaunit'sinitialallocationcancoveritsminimumpossibleemissions,thenitwillhaveanegativenetallowancepositionandbeanetsellerofallowances. IfPA=MCAs;si,aunitmayuseanycombinationofhighsulfurspotmarketcoalandlowsulfurspotmarketcoal,whichleadstoanylevelofexcessdemandintherange(EMINiAei;EMAXiAei).TheallowanceexcessdemandcanberepresentedbyAi=EMAXi(1)EMINiAeiwheretheconstant2[0;1].Aunitthatisindierentbetweenfuelswitchingandallowancespurchasescouldbeeitheranetbuyeroranetseller. CombiningtheexcessdemandsforeachofthethreecasescreatestheExcessDemandCorrespondence:Ai=8>>>><>>>>:AMAXiifPA>MCAs;siAMAXi(1)AMINiifPA=MCAs;si82[0;1]AMINiifPAMCAs;si.ThecasewhereageneratingunitusessomecombinationoflowsulfurspotmarketcoalandhighsulfurspotmarketcoalisrepresentedbythehorizontallineatwhichPA=MCAs;si. 73

PAGE 74

2{39 ),wheretheexcessdemandwillactuallybenegative. Theunit'scostsavingsfromswitchingfuelsoverpurchasingallowancesis(PAMCAs;si)(AMAXiAMINi),whichisthearea\a+b"seeninFigure 2.8 (ii).Thedark-shadedarea(a)isthecostsavingsforthegeneratingunitfromabatingemissionsthroughfuelswitchinginsteadofpurchasingadditionalallowances.Thelight-shadedarea(b)isthecostsavingsfromabatingmorethanitsinitialallocationandsellingtheextraallowances. 2.8 ,whichissummarizedinProposition2. (i) FortherangeofallowancepricesPAMCAs;si,ahighsulfurcoalcontractwillweaklyincreaseaunit'sallowanceexcessdemand. (ii) FortherangeofallowancepricesPAMCAs;si,ahighsulfurcoalcontractwillweaklydecreaseexcessdemandifScih 74

PAGE 75

FortherangeofallowancepricesPAMCAs;si,ahighsulfurcoalcontractwillweaklyincreaseexcessdemandifScih ThedecreaseinlowsulfurspotmarketcoaluseincreasesemissionsfromEMINitobEMINi,whichistheminimumemissionsgiventhehighsulfurcoalcontractconstraint: Higheremissionsmustbecoveredbyadditionalallowances,whichresultsintheminimumexcessdemandwithahighsulfurcoalcontracttobegreaterthantheminimumexcessdemandwithnohighsulfurcoalcontract: Therefore,ahighsulfurcoalcontractwillincreaseexcessdemandforallowancepricesPA>MCAs;si. 75

PAGE 76

Loweremissionsresultinadecreaseinaunit'smaximumexcessdemandfromAMAXitobAMAXiinFigure 2.8 (i): Therefore,ahighsulfurcoalcontractforcoalwithalowersulfurtoheatcontentratiowilldecreaseexcessdemandforallowancepricesPA
PAGE 77

2.8 ,aunitpreferstouselowsulfurcoalbecausePA>MCAs;si,andthecompliancecostsfromswitchingfuelstoabateemissionscanbeseeninarea(a+b).Area(b+c)istherevenuegained(negativecost)fromsellingtheremainingallowancesthatareavailableduetoabatingemissionsbelowaunit'sinitialallowanceallocation.Thenetcompliancecostsforaunitwillbethecostsofswitchingfuelsminustherevenuesfromallowancesales,or(a+b)(b+c)=(ac).If(ac)<0,thenaunitwillhavenegativecompliancecosts. Nowconsiderhowahighsulfurcoalcontractwillimpacttheexcessdemandcorrespondence,compliancecosts,andtotalcosts.AshasalreadybeenshownaboveandcanbeseeninFigure 2.8 ,ahighsulfurcoalcontractwillincreasetheminimumexcessdemandfromAMINitobAMINiandmayincreaseordecreasethemaximumexcessdemanddependingontherelativesulfurtoheatcontentratioofcontracttospotmarketcoal. Theseshiftsinaunit'sexcessdemandmayhavethreedistincteectsonaunit'scosts.Therstcostimpactistheadditionalcompliancecoststoaunitfromlostallowancesales,whichisrepresentedbyarea(d)inFigure 2.8 .Aunitmustusesomehighsulfurcoal,whichresultsinaunitcoveringadditionalemissionsthroughmoreexpensiveallowancesinsteadofswitchingtolowsulfurcoal. Thesecondimpactresultsfromthedierenceinthesulfurtoheatcontentratiobetweenhighsulfurcontractandspotmarketcoal.IfScih 2.8 (i)forreducingemissionstotheallowanceallocation 77

PAGE 78

2.8 (ii)forreducingemissionstotheallowanceallocationemissionslevel. Considerhowtheseshiftsinexcessdemandwillimpactaunit'snetcompliancecosts.Theincreaseintheminimumexcessdemandwillincreasenetcompliancecostsbydecreasingarea(b)andarea(c).Ifmaximumexcessdemanddecreases,netcompliancecostsdecreasebyarea(f)inFigure 2.8 (i).Combiningthetwoimpactsresultsinnetcompliancecostsof(ac)inFigure 2.8 (i).Ifmaximumexcessdemandincreases,netcompliancecostsincreasebyarea(f)inFigure 2.8 (ii).Combiningthetwoimpactsresultsinnetcompliancecostsof(a+fc)ininFigure 2.8 (ii). Thethirdcostimpactresultsfromdierentpricesforhighsulfurspotmarketandcontractcoal,whichisrepresentedbyarea(e).Higherpricedcontractcoalcausesaunittohaveadditionalfuelcoststomeetingelectricitydemand,whichisanincreaseaunit'stotalcosts. Thegraphicaldescriptionofahighsulfurcoalcontract'simpactsonexcessdemandandcostscanbeseenintheexampledenedinTable 3-1 .Giventhecoalcharacteristics,demand,allowanceallocation,andallowanceprice,andusingahighsulfurcontractcoalpriceof$1.50/mmBtu,theminimumandmaximumallowanceuseandMCAs;sicanbecomputedifaunitfacesnohighsulfurcoalcontract( Theminimumallowanceusebasedonthelowsulfurcoalcharacteristicsis11,400tonswhilethemaximumallowanceusebasedonthehighsulfurspotmarketcoalcharacteristicsis38,000tons.TheMCAs;siis$270.68,whichislowerthantheallowancepriceof$300.00.Soaunitwillswitchtoalllowsulfurcoaltominimizeitstotalcosts.Indoingso,itwilluseuse11,400ofitsallowanceallocationtocoveritsminimumemissions 78

PAGE 79

Nowconsiderthesameunitwithahighsulfurcoalcontractfor500,000tonsofcoal,whichaccountsforhalfoftherequiredheatinput.Themaximumallowanceuseremainsat38,000tons,buttheminimumallowanceusedecreasesto24,700tonsbecauseofthecoalcontract.TheMCAc;siis$90.23,whichismuchlowerthantheallowancepriceof$300.00andMCAs;siof$270.68.Thecoalcontractresultsincompliancecosts(costsofswitchingfuelsminusallowancesales)of$5.01millionandtotalcoststomeetingelectricitydemand(lowsulfurspotmarketcoalpurchasesplushighsulfurcontractcoalpurchasesplusallowancepurchases)of$38.61million.Compliancecostsincreasedby$390,000becausetheunitcouldnotswitchallitscoalusetolowsulfurcoal.Totalcoststotheunitincreasedby$2.79millionbecauseofthehigherpriceforthecontractcoal. Anotherpossibilityisthatthehighsulfurcontractcoalcouldactuallybecheaper,whichisreasonablebecausethemainreasonformakingacoalcontractagreementisforprotectionagainstfuturecoalpriceuctuations.ThiscasecanbeseeninFigure 2.8 ,whereMCAc;si>MCAs;siandthereisactuallyacostsavingstotheunitfromusinghighsulfurcontractcoaloverhighsulfurspotmarketcoal.Thenetcompliancecostsremainthesameasinthepreviousexample,area(ac). 79

PAGE 80

2{36 ),wheretheexcessdemandwillbepositive. GiventhataunitpreferredtousehighsulfurcoalbeforeSO2constraints,theunit'scostsavingsfromusingallowancesoverswitchingfuelsis(MCAs;siPA)(AMAXiAMINi),whichisthearea\a+b"seeninFigure 2.8 .Thedark-shadedarea(a)isthecostsavingsforthegeneratingunitfrompurchasingadditionalallowancesinsteadofswitchingfuels.Thelight-shadedarea(b)isthecostsavingsfromusingtheallocatedallowancesinsteadofabatingemissionsandsellingtheextraallowances. 2.8 ,whichissummarizedinProposition3. 80

PAGE 81

FortherangeofallowancepricesPAMCAs;si,alowsulfurcoalcontractwillweaklydecreaseaunit'sallowanceexcessdemand. (ii) FortherangeofallowancepricesPAMCAs;si,alowsulfurcoalcontractwilldecreaseexcessdemandifScil (iii) FortherangeofallowancepricesPAMCAs;si,alowsulfurcoalcontractwillincreaseexcessdemandifScil Thedecreaseinmaximumhighsulfurcoalusedecreasesaunit'smaximumemissionsfromEMAXitobEMAXi: Loweremissionsdecreasetheallowancesused,whichresultsinthemaximumexcessdemandwithalowsulfurcoalcontracttobelowerthanthemaximumexcessdemandwithnolowsulfurcoalcontract: Therefore,ahighsulfurcoalcontractwillincreaseexcessdemandforallowancepricesPA
PAGE 82

IfScil Loweremissionsresultinadecreaseinaunit'sminimumexcessdemandfromAMAXitobAMAXiinFigure 2.8 (i): Therefore,alowsulfurcoalcontractforcoalwithalowersulfurtoheatcontentratiowilldecreaseexcessdemandforallowancepricesPA>MCAs;si. Greateremissionsresultinanincreaseinaunit'sminimumexcessdemandfromAMINitobAMINiinFigure 2.8 (ii): Therefore,ahighsulfurcoalcontractforcoalwithahighersulfurtoheatcontentratiowillincreaseexcessdemandforallowancepricesPA>MCAs;si. 2.8 .If 82

PAGE 83

InFigure 2.8 ,thenetcompliancecostscanbeseenintheshadedarea(a)thatrepresentsthecostsofpurchasingallowancestocovertheunit'semissionsaboveitsallowanceallocation. Nowconsiderhowalowsulfurcoalcontractwillimpacttheexcessdemandcorrespondence,compliancecosts,andtotalcosts.AshasalreadybeenshownaboveandcanbeseeinFigure 2.8 ,alowsulfurcoalcontractwilldecreasethemaximumexcessdemandfromAMAXitobAMAXiandmayincreaseordecreasetheminimumexcessdemanddependingontherelativesulfurtoheatcontentratioofcontracttospotmarketcoal. Theseshiftsinaunit'sexcessdemandmayhavetwodistincteectsonaunit'scosts.Therstcostimpactistheadditionalcompliancecoststoaunitfromswitchingfromhighsulfurspotmarkettolowsulfurcontractcoalinsteadofpurchasingallowances,whichisrepresentedbyarea(c)inFigure 2.8 .Aunitmustusesomelowsulfurcoal,whichresultsinaunitabatingadditionalemissionsinsteadofpurchasingallowances.Area(c)isthedierencebetweenthecostsofswitchingfuelstodecreaseemissions,area(b+c),andthedecreaseincostsfromallowancepurchases,area(b).Netcompliancecostsincreasefrom(a+b)to(a+b+c). Thesecondcostimpactresultsfromdierentpricesforlowsulfurcontractandspotmarketcoal,whichisrepresentedbyarea(d).Higherpricedcontractcoalcausesaunittohaveadditionalfuelcoststomeetingelectricitydemand,whichincreasesaunit'stotalcosts. ThegraphicaldescriptionofcoalcontractimpactsonexcessdemandandcostscanbeseenintheexampledenedinTable 2-4 .Giventhecoalcharacteristics,demand, 83

PAGE 84

Theminimumallowanceusebasedonthelowsulfurcoalcharacteristicsis11,400tonswhilethemaximumallowanceusebasedonthehighsulfurspotmarketcoalcharacteristicsis38,000tons.TheMCAs;siis$270.68,whichishigherthantheallowancepriceof$200.00.Soaunitwillpurchaseallowancestominimizeitstotalcosts.Indoingso,itwilluseusetheentire20,000tonallowanceallocationandpurchaseanadditional18,000allowancestocoveritsmaximumemissionslevel.Netcompliancecostsaretheadditionalcostsfrompurchasingallowances,whichis$3.6million.Totalcostsarethetotalcostoffuelpurchasesplusallowancepurchases,whichis$34.8million. Nowconsiderthesameunitwithalowsulfurcoalcontractfor500,000tonsofcoal,whichaccountsforhalftherequiredheatinputtomeetdemand.Theminimumallowanceuseremainsat11,400tons,butthemaximumallowanceusedecreasesto24,700tonsbecauseofthecoalcontract.TheMCAs;ciis$451.13,whichismuchhigherthantheallowancepriceof$200.00andMCAs;siof$270.68.Thecoalcontractresultsinnetcompliancecosts(additionalcostsofallowancepurchasesandfuelswitching)of$4.54millionandtotalcoststomeetingelectricitydemand(highsulfurspotmarketcoalpurchasespluslowsulfurcontractcoalpurchasesplusallowancepurchases)of$38.14million.Compliancecostsincreasedby$940,000becausetheunitcouldnotuseallhighsulfurcoal.Totalcoststotheunitincreasedby$3.34millionbecauseofthehigherpriceforthecontractcoal. Anotherpossibilityisthatthelowsulfurcontractcoalcouldactuallybecheaperthanlowsulfurspotmarketcoal,whichisreasonablebecausethemainreasonformakingacoalcontractagreementisforprotectionagainstfuturecoalpriceuctuations.Thiscasecan 84

PAGE 85

2.8 ,whereMCAc;siMCAs;si.However,theimpactsoncompliancecostsandtotalcostswillbedierent.Thereasoningforthisisthatthecontractdoesnotforceaunittouseamoreexpensivecomplianceoption.BycomparingFigure 2.8 toFigure 2.8 ,thesedierencescanbederived.Theshiftinaunit's 85

PAGE 86

AlowsulfurcoalcontractwillhavethesameimpactsontheexcessdemandcorrespondencewhenPA=MCAs;siasininSection5.7.3wherePA
PAGE 87

2{56 )canbeusedtosolveforPSA,theminimumallowancepriceatwhichageneratingunitwillinstallascrubber: (bAieAi)(2{57) AunitwillprefertoinstallascrubberatPSAiftheaveragecostsofabatementfromusingascrubberisweaklylessthanthecostsofpurchasinganallowance. ScrubberinstallationdecreasesthesizeofthedenominatorofMeCAs;siby(ri)(m)(Ssih 87

PAGE 88

Ageneratingunit'sexcessdemandcorrespondencecanbederivedfromitsoptimalcompliancechoicesasthemarketallowancepricechanges.ForallowancepricesMeCAs;si>>>>>>>>>>>>><>>>>>>>>>>>>>>:AMAXiif0
PAGE 89

TheexcessdemandcorrespondenceseeninFigure 2.8 (i)includesageneratingunit'snetallowancepositionunderthefourdierentallowancepricerangesif(Psih 2.8 (ii). 2.8 89

PAGE 90

Aunitthatfacesacoalcontractwillfaceadierentindierenceallowancepriceof(PSA)becauseparametersvaluesonbothsidesoftheequalitywillchange.couldbepositiveornegativedependingonseveralconditions,includingthetypeofcoalundercontract.Thenewvaluesthatsolvethisequalityarethecontractconstrainedcostminimizingparametervalues.Thefuelcostsforthecontractcoalwillbethesamebothwithandwithoutascrubberandwillcancelout. BysolvingfortheconstantPizin( 2{59 )and( 2{60 )andsettingthetwoexpressionsequaltoeachother,thesignandvalueofcanbederived. 90

PAGE 91

Intherstcase,withoutahighsulfurcoalcontract,aunitpreferstouselowsulfurspotmarketcoalwithoutascrubberandhighsulfurcoalwithascrubber.Thetwosubcaseswillbedeterminedbythegeneratingunit'scharacteristicsandthesizeofthecontractconstraint. Intherstsubcase,bothwithorwithoutahighsulfurcoalcontract,aunitpreferstouselowsulfurspotmarketcoalifitdoesnotinstallascrubberandhighsulfurspotmarketcoalitdoesinstallascrubber.Forthistohold,(MCAs;si
PAGE 92

2{62 )and( 2{63 )canbesetequaltosolveforthevalueof: (bAMINibASMAXi)(2{65) Fillinginforallowancesandcoaluseparameters,itispossibletodeterminethesignof.AMINi=Di 2{65 ),theexpressionforcanbesimpliedtoparametersforcoalprice,sulfurcontent,heatcontent,andscrubbercapturerate. Aninterpretableformisderivedbyseparatingterms,multiplyingthoughby(bAMINibASMAXi),anddividingthroughbym(Ssih Fromtheinitialassumptionthataunituseslowsulfurcoalwhenitdoesnotinstallascrubber,itisknownthatPSA>MCAs;si.Soif(Scih 92

PAGE 93

2-6 isbasedondatafromtheaunitthatinstalledascrubberunderPhaseI. 2-6 ,aunitwillprefertouselowsulfurcoalifitdoesnotinstallascrubberbecauseMCAs;si=$244:26andPA=$250.Ifascrubberisinstalled,aunitpreferstousehighsulfurspotmarketcoalbecausethemarginalcostofabatementincreasesabovetheallowancepricetoMeCAs;si=$4;885:20.TheannualizedscrubbercostsarePiz=$15:886million,whichresultsinPSA=$1;572:56.Ahighsulfurcoalcontractfor50%ofcoaluseresultsinadecreaseintheindierenceallowancepriceof=$1;210:61to(PA)=$382:00(a76%decreaseinPSA).Althoughaunit'scompliancechoicesarenotalteredatPA=$250,thelargedecreaseinencouragesscrubbingatmuchlowerallowancepricesthanwithoutthecoalcontract.Thissubcasewillholdforacontractconstraintoflessthan73.5%ofcoaluse.Thisexampleshowsthatformanyunits,highsulfurcoalcontractswouldnothaveenoughofanimpacttoresultinscrubberinstallationinPhaseI.Onlyabout10%ofunitsactuallyinstalledscrubberstocomplywithPhaseI,buttheseunitsaccountforoverhalfoftotalabatementbyaectedunits. Inthesecondsubcase,assumethatwithoutahighsulfurcoalcontract,aunitpreferstouselowsulfurspotmarketcoalifitdoesnotinstallascrubberandhighsulfurspotmarketcoalitdoesinstallascrubberbecause(MCAs;si
PAGE 94

( 2{68 )showsthatthecostswithandwithoutascrubberareequalforagivenPSA,andcanberearrangedtondanexpressionforPiz: Sinceaunituseslowsulfurcoalwithoutascrubber,ahighsulfurcoalcontractwillchangethecoalusebyaunitwithoutascrubberandchangetheindierenceallowancepricebysomevalue: ( 2{70 )canberearrangedtondanexpressionforPiz: ThetwoexpressionsforPizin( 2{69 )and( 2{71 )canbesetequaltosolveforthevalueof: Byllinginforallowancesandcoaluse,itispossibletodeterminethesignof.AMINi=Di

PAGE 95

2{72 ),theexpressionforcanbesimpliedtoparametersforcoalprice,sulfurcontent,heatcontent,andscrubbercapturerate: Thenewformin( 2{73 )iseasiertodeterminethesignofbecausethesignoftheleft-handsideisunchangedwhilesimplifyingtheright-handside.Nowgettheright-handsideintoaformthatisinterpretablebyseparatingtermsanddividingthroughbym(Ssih (Ssih Thelasttwotermsin( 2{74 )canbecombinedtogetDi(PSAMCAs;si): Fromourinitialassumptionthataunitusesalllowsulfurspotmarketcoalifitdoesnotinstallascrubber,ifitdoesnothaveahighsulfurcoalcontract,thenPSA>MCAs;si.If(Scih AssumingthedatainTable 2-6 ,thiscasewilloccurifthecontractisgreaterthan73.5%ofcoaluse.Forthiscasetooccur,ahighsulfurcoalcontractmustresultinaunitpreferringtousehighsulfurcoalwithoutascrubberinstalledwhen,withoutacontract,itinitiallypreferstouselowsulfurcoalwithoutascrubber(PSA>MCAs;siandPSA
PAGE 96

Inthethirdcase,assumethataunit,bothwithandwithoutahighsulfurcoalcontract,preferstousehighsulfurcoalbothwithandwithoutascrubberbecause(PA
PAGE 97

2{77 )and( 2{79 )canbesetequaltosolveforthevalueof: (bAMAXibASMAXi)(2{80) Fillinginforallowancesandcoaluse,itispossibletodeterminethesignof.AMAXi=Di 2{80 ),theexpressionforcanbesimpliedtoparametersforcoalprice,sulfurcontent,heatcontent,andscrubbercapturerate: Ifhighsulfurcontractcoalhasaweaklyhighersulfurtoheatcontentratiothanhighsulfurspotmarketcoal(Scih Undertheinitialassumptions,PSA
PAGE 98

(i) (ii) Thesignofisunknownif(Scih GiventheimpactahighsulfurcoalcontracthasonexcessdemanddenedinProposition2andthesignofdenedinProposition4,theimpactofahighsulfurcoalcontractisderivedinProposition5(a)and5(b). (i) FortherangeofallowancepricesPAMeCAs;si,ahighsulfurcoalcontractwillweaklyincreaseaunit'sexcessdemand. (ii) FortherangeofallowancepricesPSAPAMeCAs;si,ahighsulfurcoalcontractwillweaklyincreaseexcessdemand. (iii) Fortherangeofallowanceprices(PSA)PAPSA,ahighsulfurcoalcontractwillweaklydecreaseexcessdemandifAMINibASMAXiandweaklyincreaseexcessdemandifAMINibASMAXi. (iv) FortherangeofallowancepricesMeCAs;siPA(PSA),ahighsulfurcoalcontractwillweaklyincreaseaunit'sexcessdemand. (v) Fortherangeofallowanceprices0PAMCAs;si,ahighsulfurcoalcontractwillweaklyincreaseaunit'sexcessdemand. WhenaunitfacesPAMeCAs;si,aunitpreferstoinstallascrubberandusealllowsulfurcoal.FromProposition2(i),ahighsulfurcoalcontractincreasestheminimumemissionslevel,whichwillweaklyincreaseaunit'sallowanceexcessdemand. (ii) WhenaunitfacesPSAPAMeCAs;si,aunitpreferstoinstallascrubberanduseallhighsulfurcoal.FromProposition2(iii),given(Scih (iii) FromProposition4(i),whenaunitfaces(PSA)PAPSA,ahighsulfurcoalcontractdecreasesaunit'sindierenceallowancepriceofinstallingascrubber 98

PAGE 99

(iv) Whenaunitfaces(Scih (v) Whenaunitfaces0PAMCAs;si,aunitdoesnotinstallascrubberandpreferstousehighsulfurcoal.FromProposition2(iii),given(Scih 2.8 (i).Proposition5(b)expressestheimpactahighsulfurcoalcontractwillhaveonaunit'sexcessdemandcorrespondencebyassumingthat>0. (i) FortherangeofallowancepricesPAMeCAs;si,ahighsulfurcoalcontractwillweaklyincreaseaunit'sexcessdemand. (ii) FortherangeofallowancepricesPSAPAMeCAs;si,ahighsulfurcoalcontractwillweaklydecreaseexcessdemand. (iii) Fortherangeofallowanceprices(PSA)PAPSA,ahighsulfurcoalcontractwillweaklydecreaseexcessdemand. (iv) FortherangeofallowancepricesMeCAs;siPA(PSA),ahighsulfurcoalcontractwillweaklyincreaseaunit'sexcessdemand. (v) Fortherangeofallowanceprices0PAMCAs;si,ahighsulfurcoalcontractwillweaklydecreaseexcessdemand. WhenaunitfacesPAMeCAs;si,aunitpreferstoinstallascrubberandusealllowsulfurcoal.FromProposition2(i),ahighsulfurcoalcontractincreasestheminimumemissionslevel,whichwillweaklyincreaseaunit'sallowanceexcessdemand. 99

PAGE 100

WhenaunitfacesPSAPAMeCAs;si,aunitpreferstoinstallascrubberanduseallhighsulfurcoal.FromProposition2(ii),given(Scih (iii) If>0,whenaunitfaces(PSA)PAPSA,ahighsulfurcoalcontractdecreasesaunit'sindierenceallowancepriceofinstallingascrubberbelowtheallowanceprice,whichleadstoaunitinstallingascrubberwhereitinitiallywouldnotanddecreasesaunit'semissionsandaunit'sexcessdemand.FromProposition2(ii),Giventhescrubberchoiceand(Scih (iv) FromProposition2(i),whenaunitfacesMeCAs;siPA(PSA),aunitdoesnotinstallascrubberandpreferstouselowsulfurcoal.Ahighsulfurcoalcontractincreasesaunit'sminimumemissionslevelandexcessdemand. (v) FromProposition2(ii),whenaunitfaces0PAMCAs;si,aunitdoesnotinstallascrubberandpreferstousehighsulfurcoal.Given(Scih 2.8 (ii). Thereareconditionsunderwhichsomeoftheseallowancepricerangesdonotexist.Forexample,PSAMCAs;sito(PA)
PAGE 101

Ifaunithasnolowsulfurcontractcoal( ( 2{82 )canberearrangedtondanexpressionforPiz: Alowsulfurcoalcontractwillchangeaunit'scoaluseandallowancepurchases,whichchangestheindierenceallowancepricebysomevalue: ( 2{84 )canberearrangedtondanexpressionforPiz: ThetwoexpressionsforPizin( 2{83 )and( 2{85 )canbesetequaltosolveforthevalueof: (bAMINibASMAXi)(2{86) Fillinginforallowancesandcoaluse,itispossibletodeterminethesignof.AMINi=Di

PAGE 102

2{86 ),theexpressionforcanbesimpliedtoparametersforcoalprice,sulfurcontent,heatcontent,andscrubbercapturerate. Amoreinterpretableexpressionisderivedbymultiplyingthroughby((bAMINibASMAXi))anddividingthroughbym(Ssih (Ssih (m)(Ssih (Ssih Fromtheinitialassumptionthataunituseslowsulfurcoalwhenitdoesnotinstallascrubber,itisknownthat(PSA>MCAs;si).Since(Scil AbindinglowsulfurcoalcontractismorelikelytoimpactunitsunderCAIR.AnexampleusingrecentdatareectiveofthecoalavailabilityanddeliveredpricesforaunitinAlabamain2000willhelptoshow<0. 102

PAGE 103

Inthesecondcase,assumethatwithoutalowsulfurcoalcontract,aunitpreferstousehighsulfurspotmarketcoalbothifitdoesordoesnotinstallascrubberbecause(PSA
PAGE 104

2{91 )and( 2{93 )canbesetequaltosolveforthevalueof: Byllinginforallowancesandcoaluse,itispossibletodeterminethesignof.AMAXi=Di 2{94 ),theexpressionforcanbesimpliedtoparametersforcoalprice,sulfurcontent,heatcontent,andscrubbercapturerate:)(bAMINibASMAXi)=mPSAHcilCcilh(Ssih Aninterpretableformisderivedbymultiplyingthroughby(bAMINibASMAXi)anddividingthroughbym(Ssih (Ssih (Ssih Fromourinitialassumptionthataunitusesallhighsulfurspotmarketcoalifitdoesnotinstallascrubberifitdoesnothaveahighsulfurcoalcontract,thenPSA
PAGE 105

ThesecondexampleinTable 2-8 usesdatathatreectsthiscase.Withoutacoalcontract,PSA=$520:35,whichislessthanMCAs;si=$657:90andPA=$700:00,andaunitwillusehighsulfurcoalbothwithandwithoutascrubberinstalled.Inthiscase,aunitwillinstallascrubberbecausePSA
PAGE 106

2{99 )and( 2{101 )canbesetequaltosolveforthevalueof: (bAMAXibASMAXi)(2{102) Fillinginforallowancesandcoaluse,itispossibletodeterminethesignof.AMAXi=Di 2{102 ),theexpressionforcanbesimpliedtoparametersforcoalprice,sulfurcontent,heatcontent,andscrubbercapturerate. Since(Scil ThethirdexampleinTable 2-8 usesdatareectiveofdeliveredcostsandcoalcharacteristicsforaunitinFlorida.Aunitinitiallypreferstousehighsulfurcoalbothwithandwithoutascrubberbecause(PA=$600:00)<(MCAs;si=$3;887:56).AunitwillprefertoinstallascrubberinthisexamplebecausetheindierentpriceisPSA=$565:60.Alowsulfurcoalcontractfor50%ofcoalusewillincreasetheindierencepriceto(PSA)=$642:42,whichwillresultinaunitnotinstallingascrubber. Basedontheabovethreecases,thesignofepsiloncanbesummarizedinProposition6. GivenProposition3andProposition6,theimpactofalowsulfurcoalcontractisderivedinProposition7(a)and7(b). 106

PAGE 107

FortherangeofallowancepricesPAMeCAs;si,alowsulfurcoalcontractwillweaklyincreaseexcessdemand. (ii) Fortherangeofallowanceprices(PSA)PAMeCAs;si,alowsulfurcoalcontractwillweaklydecreaseexcessdemand. (iii) FortherangeofallowancepricesPSAPA(PSA),alowsulfurcoalcontractwillweaklyincreaseexcessdemand. (iv) FortherangeofallowancepricesMeCAs;siPAPSA,alowsulfurcoalcontractwillweaklyincreaseexcessdemand. (v) Fortherangeofallowanceprices0PAMCAs;si,alowsulfurcoalcontractwillweaklydecreaseexcessdemand. WhenaunitfacesPAMeCAs;si,aunitpreferstoinstallascrubberandusealllowsulfurcoal.FromProposition3(iii),givenScil (ii) Whenaunitfaces(PSA)PAMeCAs;si,aunitpreferstoinstallascrubberanduseallhighsulfurcoal.FromProposition3(i),alowsulfurcoalcontractdecreasesthemaximumemissionslevel,whichwillweaklydecreaseaunit'sallowanceexcessdemand. (iii) FromProposition6,whenaunitfacesPSAPA(PSA),alowsulfurcoalcontractincreasesaunit'sindierenceallowancepriceofinstallingascrubberabovetheallowanceprice,whichleadsaunittonotinstallascrubberwhereitinitiallywouldhavedonesoandincreasesaunit'semissionsandaunit'sexcessdemand.FromProposition3(iii),giventhescrubberchoiceandScil (iv) WhenaunitfacesMeCAs;siPAPSA,aunitdoesnotinstallascrubberandpreferstouselowsulfurcoal.FromProposition3(iii),givenScil (v) Whenaunitfaces0PAMCAs;si,aunitdoesnotinstallascrubberandpreferstousehighsulfurcoal.FromProposition3(i),alowsulfurcoalcontractdecreasesaunit'smaximumemissionslevelandexcessdemand. 2.8 (i). 107

PAGE 108

FortherangeofallowancepricesPAMeCAs;si,alowsulfurcoalcontractwillweaklydecreaseexcessdemand. (ii) Fortherangeofallowanceprices(PSA)PAMeCAs;si,alowsulfurcoalcontractwillweaklydecreaseexcessdemand. (iii) FortherangeofallowancepricesPSAPA(PSA),alowsulfurcoalcontractwillincreaseexcessdemandASMAXibAMINiandweaklydecreaseexcessdemandifASMAXibAMINi. (iv) FortherangeofallowancepricesMeCAs;siPAPSA,alowsulfurcoalcontractwillweaklydecreaseexcessdemand. (v) Fortherangeofallowanceprices0PAMCAs;si,alowsulfurcoalcontractwillweaklydecreaseexcessdemand. WhenaunitfacesPAMeCAs;si,aunitpreferstoinstallascrubberandusealllowsulfurcoal.FromProposition3(ii),givenScil (ii) Whenaunitfaces(PSA)PAMeCAs;si,aunitpreferstoinstallascrubberanduseallhighsulfurcoal.FromProposition3(i),alowsulfurcoalcontractdecreasesthemaximumemissionslevel,whichwillweaklydecreaseaunit'sallowanceexcessdemand. (iii) FromProposition6,whenaunitfacesPSAPA(PSA),alowsulfurcoalcontractincreasesaunit'sindierenceallowancepriceofinstallingascrubberabovetheallowanceprice,whichleadsaunittonotinstallascrubberwhereitinitiallywouldhavedonesoandincreasesaunit'semissionsandaunit'sexcessdemand.FromProposition3(ii),giventhescrubberchoiceandScil (iv) WhenaunitfacesMeCAs;siPAPSA,aunitdoesnotinstallascrubberandpreferstouselowsulfurcoal.FromProposition3(ii),givenScil (v) Whenaunitfaces0PAMCAs;si,aunitdoesnotinstallascrubberandpreferstousehighsulfurcoal.FromProposition3(i),alowsulfurcoalcontractdecreasesaunit'smaximumemissionslevelandexcessdemand.

PAGE 109

2.8 (ii).Aswithahighsulfurcontract,someofthepricerangesmaynotexistforaparticularcase.However,theremainingpartsofthepropositionshold. Ashasbeenshowinseveralexamples,highsulfurcoalcontractsforalargefraction(50-100%)ofcoalusecangreatlyreduceaunit's\indierenceprice"toinstallingascrubber.SomeunitsunderPhaseIinitiallyappeartohaveinstalledascrubberwhenitwasnotaunit'soptimalcomplianceoption,increasingaunit'scompliancecosts.Additionalscrubberinstallationsshouldhaveresultedingreateremissionsreduction,whichshouldsimultaneouslylowerdemandandincreasesupplyofallowancesasaunitswitchesfromanetdemandertoanetseller.Indoingso,theequilibriumallowancemarketpriceshouldbedrivenlower,whichmayexplainthelowerthanexpectedallowancepricesrealizedduringPhaseI.Eventhoughtheallowancemarketpricewaslowerthanexpected,theinecientunitcompliancechoicesresultedinhigherthanexpectedtotalindustrycompliancecosts. UnderfutureCAIRregulation,aunit'scomplianceoptionsmayberestrictedbylowsulfurcoalcontractsagreeduponduringthe1990stomeetTitleIVemissionsrequirements.Aunitmayndinstallingascrubberandusinghighsulfurcoaltobeitbestcomplianceoption.However,lowsulfurcoalcontractsmayleadaunittochooseasuboptimalcompliancechoice,suchasswitchingfuelsorinstallingascrubberwhileusing 109

PAGE 110

Ashasbeenshowinseveralexamples,lowsulfurcoalcontractsforalargefraction(50-100%)ofcoalusecangreatlyincreaseaunit's\indierenceprice"toinstallingascrubber.SomeunitsunderCAIRmaynotinstallascrubberwhenitisoptimalforthemtodoso,increasingaunit'scompliancecosts.Fewerscrubberinstallationswouldresultingreateremissions,whichshouldsimultaneouslyincreasedemandanddecreasesupplyofallowancesasaunitwouldbeanetdemanderinsteadofanetseller,andtheallowancemarketpriceshouldbedrivenhigherthanwouldbeexpected.Inthiscase,ahigherthanexpectedallowancemarketpricewouldoccurwithhigherthanexpectedtotalindustrycompliancecosts. First,giventhescrubberchoice,itiseasytodeterminehowacoalcontractwillimpactaunit'sexcessdemandforallowances.Ahighsulfurcoalcontractwillweaklyincreaseaunit'sexcessdemandwhilealowsulfurcoalcontractwillweaklydecreaseaunit'sexcessdemand. Second,somecoalcontractsmayactuallydecreaseaunit'stotalcostsrelativetousingonlyspotmarketcoalwhileincreasingaunit'scompliancecostsifacoalcontractallowsaunittolockinalowerpriceforagiventypeofcoal.Thecoalcontractwillrestrictcompliancechoices,whichmayresultinhighercompliancecosts.Ifthefuelcostsavingsisgreaterthantheincreaseinaunit'scompliancecosts,thenthecoalcontractlowersaunit'stotalcosts.Sinceaunitonlycaresaboutitstotalcosts,aunit'scompliancedecisionsdonotnecessarilyminimizeaunit'scompliancecosts. 110

PAGE 111

Fourth,coalcontractconstraintschangetheallowancepriceatwhichaunitwillprefertoinstallascrubber.InthecaseofPhaseI,ahighsulfurcoalcontractmayincreaseaunit's\indierenceprice",whichcreatesagreaterincentiveforaunittoscrubandsellitsextraallowances.Thisresultmayexplainwhysomescrubberswereinstalledatsub-optimalunitswhiletheallowancemarketpricewasmuchlowerthanexpectedduringPhaseI.TheoppositemayoccurunderCAIR,wherealowsulfurcoalcontractmayincreaseaunit's\indierenceprice"andlowertheincentiveforaunittoinstallascrubberevenifitwouldbetheoptimalcompliancechoice.Ineithercase,asuboptimalcompliancechoicewillbemadeandaunit'scompliancecostswillweaklyincrease. Fifth,thereiscertaintyhowacoalcontractwillimpactaunit'sexcessdemandforallowancesevenwhenthescrubberchoiceisconsideredformostallowancepriceranges.However,duetothediscretescrubberchoiceandthechangeinthe\indierenceprice"duetoacoalcontract,itisuncertainhowacoalcontractwillalteraunit'sexcessdemandiftheallowancemarketpricefallsinoneparticularpricerange.AhighsulfurcoalcontractwillshiftexcessdemandasderivedinProposition2exceptiftheallowancemarketpricefallsinthepricerange(PSA;PSA)andScih 111

PAGE 112

112

PAGE 113

PhaseIComplianceCostEstimates SourceMCAi/tonPred.CostsLeast-CostActualCost ICF(1989,1990)$199-$226$450-860millionN/AN/AEPRI(1993,1995)$879-$1238$900-1,340millionN/AN/AGAO(1994)$299$1,170millionN/AN/AWinebrake,etal.(1995)$143$502millionN/AN/APre-PolicyEstimates$143-$1238$450-$1340millionN/AN/AEllermanetal.(1997)$200-300N/AN/A$730millionSotkiewiczandHolt(2005)$150-$180N/A$423-$553million$990millionCarlsonetal.(2000)$71N/A$571million$910millionPost-PolicyEstimates$71-$800N/A$423-$571million$730-$990million Sources:BohiandBurtraw(1997);Carlsonetal.(2000);SotkiewiczandHolt(2005);SmithandEllerman(1998)

PAGE 114

HighSulfurCoalContract:Assumptions CoalTypePrice/mmBtuHgifPrice/TonSgif Table2-3. HighSulfurCoalContract:Results CostsComplianceTotalCostsComplianceTotalCostsCosts(Ex.1)(Ex.1)Costs(Ex.2)(Ex.2) Unconstrained$4.62million$35.82million$4.62million$35.82millionConstrained$5.01million$38.61million$5.01million$35.01millionChange$390,000$2.79million$390,000-$810,000 AllowanceUse(Tons)MCAi Table2-4. LowSulfurCoalContractExamples:Assumptions CoalTypePrice/mmBtuHgifPrice/TonSgif 114

PAGE 115

LowSulfurCoalContractExamples:Results CostsComplianceCosts(Ex.1)TotalCosts(Ex.1)ComplianceCosts(Ex.2)TotalCosts(Ex.2) Unconstrained$3.6million$34.8million$3.6million$34.8millionConstrained$4.54million$38.14million$4.54million$34.54millionChange$940,000$3.34million$940,000-$260,000 AllowanceUse(Tons)MCAi

PAGE 116

ExampleEpsilonMagnitude:Case1 ScrubberCharacteristicsGeneratingUnitCharacteristics CapitalCost$260/kWCapacity365.3MWOandMCost2mills/kWhHeatRate10,000Btu/kWhPiz$15,886,295Cap.Factor75%ri95% ParametersCharacteristicsValueResults LowSulfurCoalPrice$1.50/mmBtu50%Contract(Spot)HeatContent24mmBtu/tonPSA=$1,592.61=$1,210.61SulfurContent0.65%PSA=$382.00Pct.#PSA=76%HighSulfurCoalPrice$1.30/mmBtu75%Contract(Spot)HeatContent24mmBtu/tonPSA=$1,592.61=$1,349.82SulfurContent2.5%PSA=$242.79Pct.#PSA=85%HighSulfurCoalPrice$1.50/mmBtu(Contract)HeatContent24mmBtu/tonSulfurContent4.0%

PAGE 117

ExampleEpsilonMagnitude:Case2 ScrubberCharacteristicsGeneratingUnitCharacteristics CapitalCost$260/kWCapacity365.3MWOandMCost2mills/kWhHeatRate10,000Btu/kWhPiz$15,886,295Cap.Factor75%ri95% ParametersCharacteristicsValuesResults LowSulfurCoalPrice$2.79/mmBtuBLEngland(Spot)HeatContent24mmBtu/ton50%ContractSulfurContent0.5%PSA=$460.80=$44.66HighSulfurSpotCoalPrice$1.52/mmBtuPA=$416.14Pct.#PSA=10%(Spot)HeatContent24mmBtu/tonSulfurContent1.91%HighSulfurCoalPrice$1.72/mmBtu(Contract)HeatContent24mmBtu/tonSulfurContent2.32%

PAGE 118

ExampleEpsilonMagnitude CapitalCost$141.34/kWCapacity365.3MWOandMCost1.23mills/kWhHeatRate10,000Btu/kWhPiz$9,016,616.85Cap.Factor75%ri95% ParametersCharacteristicsValuesResults LowSulfurCoalPrice$1.50/mmBtuCase1(Alabama)(Alabama)HeatContent24mmBtu/tonPSA=$731.46=-$42.00SulfurContent0.7%PA=$773.46Pct:"PSA=6%HighSulfurSpotCoalPrice$1.10/mmBtu(Alabama)HeatContent24mmBtu/tonSulfurContent1.6%HighSulfurSpotCoalPrice$2.00/mmBtuCase2(CAIR)(CAIRExample)HeatContent25mmBtu/tonPSA=$520.35=-$352.99SulfurContent1.0%PA=$873.34Pct."PSA=68%LowSulfurCoalPrice$2.20/mmBtu(CAIRExample)HeatContent25mmBtu/tonSulfurContent0.6%HighSulfurSpotCoalPrice$2.30/mmBtuCase3(Florida)(Florida)HeatContent24mmBtu/tonPSA=$565.60=-$76.82SulfurContent1.4%PA=$642.42Pct:"PSA=14%LowSulfurCoalPrice$3.50/mmBtu(Florida)HeatContent25mmBtu/tonSulfurContent0.7%

PAGE 119

TheSO2AllowancePrice.DatawasgiventomebyDallasBurtrawofResourcesfortheFuture Figure2-2. ExcessDemandCorrespondenceandComplianceCostSavingsfromFuelSwitchingOverAllowancePurchasing 119

PAGE 120

HighSulfurContract:ShiftinMinimumExcessDemand Figure2-4. NoContract:ComplianceCosts Figure2-5. HighSulfurContract:ComplianceandTotalCosts 120

PAGE 121

HighSulfurContract:RelativeSavingsfromContractCoal Figure2-7. CostSavingsfromUsingAllowancesOverFuelSwitching Figure2-8. LowSulfurContract 121

PAGE 122

NoContract:ComplianceCosts Figure2-10. LowSulfurCoalContract:MCAs;ci LowSulfurCoalContract:MCAs;ci

PAGE 123

ComplianceCosts:PA=MCAs;si HighSulfurCoalContract Figure2-14. LowSulfurCoalContract 123

PAGE 124

ExcessDemandCorrespondence:MCAs;si
PAGE 125

ImpactofaHighSulfurCoalContract:MCAs;siPSA ImpactofaLowSulfurCoalContract:MCAs;si>PSA

PAGE 126

ImpactofalowsulfurCoalContract:MCANSi
PAGE 127

Giventhescrubberchoice,anallowancemarketequilibriumwillexist.Allowingforthescrubberchoicemakesitimpossibletoguaranteeanequilibrium,butonestillmayexist.Bindingfuelcontractsmayleadtoalteredunit-levelexcessdemandsand,insodoing,theallowancemarketprice(PA).Meanwhile,bindingcontractscanaltercompliancedecisionsandincreasetotalindustrycompliancecosts. Thispaperusesgeneratingunit-levelsimulationstoreplicateresultsfrompreviousstudiesandshowthatshort-runfuelcontractsappeartoexplainalargeportionofthepreviouslyunexplainedexcesscompliancecostsfoundinprevioussimulations.Simulatingtheleast-costcompliancechoiceswithoutincludingfuelcontractconstraintsresultsinminimumannualindustrycompliancecostsof$288.3million,whichvariesgreatlyfromtheactualcompliancecostsof$1.30billionfoundinthesesimulations.Oncefuelcontractconstraintsareintroducedintothesimulation,theminimumannualindustrycompliancecostsbecome$1.01billion.Basedontheseresults,fuelcontractconstraintsexplain$651.1million,or64%oftheexcesscompliancecostsrealizedintheprogramfor1996. 127

PAGE 128

Thepaperwillbestructuredinthefollowingmanner:Section2willlookattheconditionsforanallowancemarketequilibriumwhileSection3willlookatthecomparativestaticsoftheallowancemarket.Section4willlookatcompliancecosts,bothforanindividualgeneratingunitandtheentireindustry.Section5andSection6willexplainthedatausedinthegeneratingunitlevelsimulationsandtheresultsfromtheactualsimulations,respectively.Section7willanalyzetheplantleveldecisionmakingprocess,howitdiersfromthegeneratingunitlevelmodel,andwhatadditionalinformationthemodelwilladdtotheliterature. 3.2.1GeneratingUnitProblem subjectto...Aei+Ai(1ziri)(m)(CsihSsih+CcihScih+CsilSsil+CcilScil)i1 (CsihHsih+CcihHcih+CsilHsil+CcilHcil)Dii2 128

PAGE 129

(3{6) Equation( 3{1 )representstheunit'scostfunction.Thesecostsincludethecostofscrubberinstallation(ziPiz),netcostsofallowancepurchases(PAAi),andcostsofcoalpurchases(PsihCsih+PsilCsil+PcihCcih+PcilCcil).Theemissionsconstraintisshownin( 3{2 ),wherethenumberofallowancesheld(Aei+Ai)mustbeaslargeastheamountoftotalemissionsbythegeneratingunit[(1ziri)(m)(CsihSsih+CcihScih+CsilSsil+CcilScil)].Totalemissionsisafunctionoftheamountofeachcoaltypeusedaswellastheemissionsreductionduetoascrubber,ifoneisinstalled.TheLagrangemultiplierontheemissionsconstraintisrepresentedbyi1.Thedemandconstraintrequiresthattheamountofheatinputtogenerateelectricity(CsihHsih+CcihHcih+CsilHsil+CcilHcil)mustcovertheconsumerdemand(Di)forelectricityexpressedasheatinput,whichisseenin( 3{3 ).TheLagrangemultiplieronthedemandconstraintisrepresentedbyi2.Coalcontractconstraintsrequiretheunittouseaspecicamountofeachcontractcoaltype, 3{4 )and 3{5 ).Aunitwilluseexactlythecontractedamountbecause(1)ifthecontractcoalismoreexpensivethanspotmarketcoal,thenaunitwillnotwanttouseanymorecontractcoalthanisnecessaryand(2)ifcontractcoalischeaperthanspotmarketcoal,thecoalproducerwouldprefertosellanyadditionalnon-contractedcoalthroughthespotmarket.TheLagrangemultiplierforeachcontractconstraintoneachcoaltypeisrepresentedbyihforhighsulfurcontractcoalandilforlowsulfurcontractcoal. 129

PAGE 130

Introducingahighsulfurcoalcontractconstraintwillrestrictspotmarketcoaluseby Introducingalowsulfurcoalcontractconstraintwillrestrictspotmarketcoaluseby 130

PAGE 131

Allowingthescrubberchoicetobeendogenousresultsinmorecompleximpactsofcoalcontractsoncompliancedecisions.Ageneratingunitthatdoesnotfaceanycoalcontractconstraintswillmakeitsoptimalchoicesbasedontwoconditions:(1)therelationshipofPArelativetotheallowancepriceatwhichaunitisindierenttoinstallingascrubber(PSA)and(2)therelationshipofPArelativetoMCAs;sigiventhescrubberchoice. IfPSA>PA,aunitwillnotinstallascrubberbecausetotalcoststothegeneratingunitwillbelowerwithoutinstallingascrubber.Theresultingcoaluseandexcessdemandwillbethesameasdescribedabove:Csih=Di,Csil=0,andAi=AMAXiifPAMCAs;siwhereMCAs;siisthemarginalcostofabatementfromswitchingfuelswithoutascrubber. IfPSAMeCAs;si. Theresultisacombinationofthetwoexcessdemandcorrespondenceswhereatsomeallowanceprice(PSA)whereascrubberwillbeinstalledthereisalargenon-continuousdecreaseinaunit'sexcessdemand. Introducingahighsulfurcoalcontractconstraintwillhavethesamedirectimpactsaswhenscrubbersaregiven.Coalusewillberestrictedby 131

PAGE 132

Introducingalowsulfurcoalcontractconstraintwillhavethesamedirectimpactsaswhenscrubbersaregiven.Coalusewillberestrictedby Insummary,coalcontractswillrestrictaunit'scoaluseoptions,whichmayalteraunit'scompliancechoicesandexcessdemandforallowances.Lessfreedomincompliancedecisionsmayresultinanincreaseinaunit'scompliancecosts. 132

PAGE 133

minzi;Ai;Csih;CsilziPiz+PAAi+PsihCsih+PsilCsil+PcihCcih+PcilCcil subjectto...Aei+Ai(1ziri)(m)(CsihSsih+CcihScih+CsilSsil+CcilScil) (3{9) (CsihHsih+CcihHcih+CsilHsil+CcilHcil)Di (3{13) (3{15) (2)Theallowancemarketclears.AtPA, FromKakutani'sFixedPointTheorem,axedpointexistsifthemarketexcessdemandcorrespondenceuppersemi-continuous,compact-valued,andconvex-valued.Eachgeneratingunit'sexcessdemandcorrespondencesareuppersemi-continuous,compact-valued,andconvex-valued,whichimpliesthatthemarketexcessdemandisalsouppersemi-continuous,compact-valued,andconvex-valued.Giventheseconditionsthereisaxedpoint,andtherefore,anequilibriumexists.Notethattheremaybemultipleequilibria,butthereisatleastoneequilibrium.TheProofofTheorem1isinAppendixB. Theexistenceofanallowancemarketequilibriumisnolongerguaranteedoncethediscretescrubberchoiceisintroducedintothemodel.Thenon-convexnatureofthescrubberdecisionvariablemakesitimpossibletoguaranteethatanequilibriumexistsbecausetheexcessdemandisnolongerconvexfortherange(ASMAXi;AMINi).Insuch 133

PAGE 134

3.4.1ComparativeStatics:EectofRelativeFuelCostontheAllowanceMarket 3{17 ). ConsidertheimpactofanincreaseinthepriceoflowsulfurspotmarketcoalontheMCAs;si.Anincreaseinthepricedierencebetweenfuelsperunitofheatfrom(Psil 3{17 )andresultsinanincreaseinthemarginalcostofabatementfromMCAs;sitoMCAs;si+where= Hsil HsihSsil Hsil).AhigherMCAs;simakestherelativecosttothegeneratingunitpurchasingallowanceslower,andwillleadtoaweakincreaseinaunit'sexcessdemand.ExcessdemandwillincreaseifMCAs;siPAandnotchangeifMCAs;si+PA.Greaterexcessdemandintheallowancemarketresultsinanincreaseinthetotalallowancemarketdemandandpossiblyadecreaseinallowancemarketsupply,bothofwhichwillweaklyincreasePA. 134

PAGE 135

IfPAMCAs;si,ageneratingunitpreferstoswitchfuelsfromhightolowsulfurcoal,createitsminimumpossibleemissions,andhaveanexcessdemandofAMINi.However,thehighsulfurcoalcontractwouldforcegreaterthantheminimumamountofemissionsandweaklyincreaseexcessdemandtobAMINi.Ahigherexcessdemandwillleadtoadecreaseintheallowancemarketsupplyorbothadecreaseinallowancemarketsupplyandanincreaseintheallowancemarketdemand.InbothcasesPAweaklyincreases. 135

PAGE 136

IfPA>MCAs;si,ageneratingunitpreferstoswitchfuelsfromhightolowsulfurcoal,createitsminimumpossibleemissions,andhaveanexcessdemandofAMINi.Assumingthesamerelativesulfurcontentforspotandcontractcoal,alowsulfurcoalcontractwillnotchangetotalemissionsorexcessdemand(bAMINi=AMINi)andPAremainsunchanged. IfPA
PAGE 137

Whenyouconsiderthescrubberchoiceinthedecision-makingprocess,itisuncertainhowacoalcontractwillaecttheallowancemarketbecausethecontractmayincreaseordecreaseexcessdemanddependingontheallowancepricerangesderivedinChapter2.Althoughitiscertainthatifthecoalcontractbinds,thentherewillbeashifttoasub-optimalexcessdemandforthreedierentallowancepriceranges. Ifageneratingunithasahighsulfurcoalcontract,theunit'sexcessdemandincreasesfortwoallowancepriceranges:(PA>MeCAs;si)and(PSA>PA>MCAs;si).TheincreaseinexcessdemandwillbefromASMINitobAMINiwhen(PA>MeCAs;si).TheincreaseinexcessdemandwillbefromAMINitobAMINiorAMINitoAMAXiwhen(PSA>PA>MCAs;si),dependingontherelationshipbetween(PSA)andMCAs;si.Itwillbethelatterinthespecialcasewherethegeneratingunitalwayspreferstousehighsulfurcoal. Whenthemarketallowancepriceisinthesetworangesthatresultinanincreaseinageneratingunit'sexcessdemand,therewillbeeitheranincreaseinthemarketdemandforallowances,orbothanincreaseinmarketdemandandadecreaseinmarketsupplyforallowances.Inbothsituations,theallowancepricewillbedrivenhigher. Thethirdpricerangethathasashiftintheexcessdemandis(PSA;PSA).AgeneratingunitdecreasesitsexcessdemandfromAMINitoASMAXi,whichisaresultofaunitinstallingascrubberforapricerangeforwhichitinitiallywouldnotinstallascrubber. 137

PAGE 138

Ifageneratingunithasalowsulfurcoalcontract,theunit'sexcessdemanddecreasesfortwoallowancepriceranges:(MeCAs;si>PA>PSA+)and(minfPSA;MCAs;sig>PA).ThedecreaseinexcessdemandwillbefromASMAXitobAMAXiwhen(MeCAs;si>PA>PSA+).ThedecreaseinexcessdemandwillbefromAMAXitobAMAXiwhen(minfPSA;MCAs;sig>PA). Whenthemarketallowancepriceisintheserangesthatresultinadecreaseinageneratingunit'sexcessdemand,therewillbeeitherandecreaseinthemarketdemandforallowances,orbothandecreaseinmarketdemandandaincreaseinmarketsupplyforallowances.Inbothsituations,PAwillbedrivenlower. Thethirdpricerangethathasashiftintheexcessdemandis(PSA;PSA+).AgeneratingunitincreasesitsexcessdemandfromeitherASMAXitoAMINiorASMAXitobAMAXidependingontherelationshipbetween(PSA+)andMCAs;si.Thisincreaseisaresultofaunitnotinstallingascrubberforapricerangeforitinitiallywouldhaveinstalledascrubber. Whenthemarketallowancepriceisinthisrangethatresultsinanincreaseinageneratingunit'sexcessdemand,therewillbeeitheranincreaseinthemarketdemandforallowances,orbothanincreaseinmarketdemandandadecreaseinmarketsupplyforallowances.Inbothsituations,PAwillbedrivenhigher. 138

PAGE 139

3{18 )). Compliancecostsmaybepositiveornegativedependingonitscompliancedecisionsanditsinitialallowanceallocation.ThescrubberinstallationcostsarerepresentedbyPiz,andwillonlyattributetoaunit'scompliancecostsifascrubberisinstalledinresponsetotheprogram(zi=1).Thecostsofaunit'snetallowancepositionisthedierencebetweenageneratingunit'sinitialallowanceallocationanditsactualemissionsmultipliedbytheallowanceprice(PAAi).Thecostsofswitchingfuelsisthelargeroftwovalues:(1)totalcostsofactualcoalpurchases(PsihCsih+PsilCsil+Pcih 139

PAGE 140

(bziPiz+PAbAi+PsilbCsil+PsihbCsih+Pcih Theminimumcompliancecostchoiceswithnocontractconstraintforpreviousstudiesaredenotedby\"whilethecontractconstrainedminimumcompliancecostchoicesaredenotedby\^"in( 3{20 ).Theassumed\c"isthetotalcostsofmeetingdemandassumingnocontractconstraintoremissionsconstraint.Thecostdierencebetweenthesetwosetsofchoiceswillbethecombinationofchangesincompliancedecisionsandfuelcosts.Someofthese\excesscompliancecosts"maynotbecompliancecosts,whichmakesthederivedcompliancecoststoohigh.Soitisimportanttoincludethecontractconstrainedcoaluseinthebaseline\c"togettheappropriatecompliancecosts.Assumingthatcontractedcoalisatleastasexpensiveasspotmarketcoal,thecompliancecostswillbegreaterforthecontractconstrainedcaseasshownin( 3{20 )andasexpressedin( 3{21 ). Considerasimpleexamplewherehighsulfurcontractcoalismoreexpensivethanhighsulfurspotmarketcoal,lowsulfurcoalispreferredoverhighsulfurcoal,andnoscrubberwouldbeinstalledifageneratingunithadtotalfreedominitscoalusechoices.Nowcompareageneratingunit'scompliancecoststothecompliancecostsaunitwould 140

PAGE 141

3{22 )includetheadditionalcostsfromanincreaseinallowancepurchasesandadecreaseincoalcostsfromusingcheaper,butdirtierfuel. Sincethecontractcoalwouldhavetobeusedregardlessofenactmentoftheprogram,theadditionalcostsforhighsulfurcontractcoaluseinsteadofhighsulfurspotmarketcoal(PcihCc;MAXihPsihCs;MAXih)shouldnotbeincludedincompliancecosts.However,previousstudieswouldrecognizethesefuelcostsascompliancecostsbasedontheassumedcostminimizationwithoutcontractconstraints. Theappropriateunit-levelcompliancecostswilltaketheformshownin( 3{23 )wherethebaselinetotalcostswilltakethevalue\bc",whichrepresentsthecostsofmeetingelectricitydemandgiventhecoalcontractconstraint.Thesecosts\bc"willbedierentthan\c"becausethespotmarketcoalusewillbealteredifthecontractedcoaldiersrelativethespotmarketcoalitreplacesinproduction. 3{25 ). 141

PAGE 142

3{26 )arethesumofthecompliancecostsforeachgeneratingunit.Similartothegeneratingunit'sprobleminChapter2,therearetwoconstraintsthatmustbemet.First,thesumofallowanceallocationsmustcoverthesumofemissionsthatareemittedbyallaectedgeneratingunits.Second,eachgeneratingunitmustproduceenoughelectricitytocoveritselectricitydemandrequirement. minzi;Ai;Csih;CsilnXi=1ziPiz+maxf(PsilCsil+PsihCsihPsihCs;MAXih);0g Solvefortherst-orderconditionsgivenzi.Forhighsulfurcoal: Forlowsulfurcoal: 142

PAGE 143

Theserst-orderconditionscanbeusedtosolvefori1,theallowanceshadowprice,whichisalsothesameasforthegeneratingunitproblem.SincethecompliancechoicesthatminimizethecostsforallgeneratingunitsresultintheequilibriumPA,theallowancechoicesalsominimizethetotalindustrycosts. (3{29) Thesameapproachcanbeusedundertheindustry-widecompliancecostproblemwithcoalcontractconstraintstoshowthatcontractconstrainedminimumtotalindustrycompliancecostwilldierfromtheunconstrainedcosts. minzi;Ai;Csih;CsilnXi=1[ziPiz+maxf(PsilbCsil+PsihbCsihPsihbCs;MAXih);0g] (3{30) Forlowsulfurspotmarketcoal... 143

PAGE 144

(3{33) Theserst-orderconditionscanbeusedtosolvefori1,theallowanceshadowprice,whichisalsothesameasfortheindividualgeneratingunitproblem.Ifageneratingunituseshighsulfurcoal,PAMCAs;si.Ifageneratingunituseslowsulfurcoal,PAMCAs;si.Ifageneratingunitusesbothhighandlowsulfurcoal,PA=MCAs;si.Soageneratingunit'scostminimizingchoicesofzi;Csih;Csil;Asiforeachgeneratingunitalsominimizetotalindustrycompliancecosts.IfthecontractconstrainedcompliancechoicesthatminimizethecostsforageneratingunitresultsinanequilibriumPA,thechoicesalsominimizethecontractconstrainedtotalindustrycosts.Notethattheoptimalparametercombinationforthecontractconstrainedcasewillnotbethesameasundertheunconstrainedcase. Ithasalreadybeenshownthatanybindingcontractconstraintforanindividualgeneratingunitwillresultinasub-optimalcombinationofcoaluse,whichwillweaklyincreasecompliancecostsforthecontractconstrainedgeneratingunit.Anysub-optimalchoicesmadebyonegeneratingunitwillweaklyincreasethecostsfortheentireindustry. 144

PAGE 145

Theimportanceoftheallowanceallocationinthesepreviousstudiesreliesontheexistenceoftransactioncosts.However,thismodelhasassumednoallowancetransactioncosts,whichallowsittoshowthatevenwithcoalcontractconstraintstheallowanceallocationdistributionwillnotimpactageneratingunit'scompliancechoicesorthetotalindustrycompliancecosts. Thechoicebetweenswitchingfuelsorpurchasingallowancesisbasedsolelyonageneratingunit'srelativemarginalcostofpurchasingthenextallowancecomparedtotheeectivemarginalcostofabatingthenextunitofemissions.Ascanbeseen,Aeihasnoimpactonageneratingunit'scompliancedecisions.Ageneratingunitwilluseallowances 145

PAGE 146

(3{36) (bAieAi)PSA (cEi+AeifEiAei)PSA(3{37) Nowconsiderthecompliancechoicesmadeafterthescrubberinstallationdecision.ItwasshowninChapter2thatageneratingunit'scompliancechoicesarebasedontherelationshipbetweenthePAandMCAs;siorMeCAs;si,dependingontheunit'sscrubberchoice,foraparticularunitofcoal.Similartowhatwasshownabove,ageneratingunit'sMeCAs;siandMCAs;siforagivenunitofcoalarenotaectedbyAeibecauseallowancesarenotaparameterinthemarginalcostofabatement.MeCAs;si=Pil 146

PAGE 147

3.6.1Introduction Therstportionofthissectionwilllookatthedatausedtoparameterizethesimulationmodel,whichincludesadescriptionofwherethedatawasobtained,thetechniquesusedtocreatetheparameters,andsomeissuesregardingthedata.Thesecondportionwilllookatthemodeldesignandapproach.Thelastportionwillsummarizethesimulationresultsintermsofthetotalindustry,individualstates,andindividualgeneratingunits. Uptothispointallgeneratingunitsareassumedtobecoal-redunitsbecauseitistheprimaryfueloptionsforelectricitygenerationintheU.S.However,thereare24unitsofthe431unitsthatusedfueloilornaturalgasforelectricitygeneration.ThetreatmentoftheseunitsisdescribedinSection6.2.1. 147

PAGE 148

TheheatcontentofdeliveredcoalisinBtusofheatpertonofcoal,barrelofoil,or1,000cubicfeetofnaturalgas.Allfuelparametersareconvertedintermsofaheatcontentbaselinetoallowforcomparisonsacrossunits. Thepurchasingagreementislabeledastwopossibleparameters:(1)spotmarketand(2)contractpurchases.Anagreementisconsideredaspotmarketpurchaseiftheagreementisforlessthan2years.Purchasingagreementsoftwoyearsorgreaterareconsideredtobeundercontract.Theamountoffuelundercontractisintons,barrels,or1,000cubicfeetdependingonthefuel.Thedataismanipulatedintototalheatcontentbymultiplyingtheheatcontentperunitoffuelbytheamountoffueldelivered. Sincecoaldeliveriesaremadeattheplantlevel,thesizeofthecontractconstraintisnoteasilyderivedforeachgeneratingunit.Theparameterfortheamountofcontractcoalforeachgeneratingunitwasderivedinthefollowingmanner.First,anyunitatagivenplantthatactuallyinstalledascrubberisallocatedasmuchhighsulfurcontractcoalaspossiblewhilereceivingaslittlelowsulfurcontractcoalaspossible.Thereasoningforthisisthataunitwithascrubberwillhavealargermarginalcostofabatementfromswitchingfuels(MCAs;si
PAGE 149

3-1 andwouldbeconsideredcoalstoredattheplantforusethefollowingyear. Sulfurcontentisthepercentageofeachtonofcoal,barrelofoil,or1,000cubicfeetofnaturalgasthatissulfur.Thedatamustbemanipulatedtocreatethedesiredvariable,whichispoundsofsulfurdioxidepermillionBtusofheat.PhaseIofTitleIVdistributesallowanceallocationsbasedon2.5poundsofSO2permillionBtusofheat.Sofuelisconsidered\highsulfur"ifcontainsgreaterthen2.5lbs.ofSO2permmBtuand\lowsulfur"ifitcontainslessthan2.5lbs.ofSO2permmBtu.Anycoalusethathasahigher(lower)sulfurcontentwillincrease(decrease)emissionsabove(below)itsallocationallows. 149

PAGE 150

AlthoughnearlyallgeneratingunitsaectedbyPhaseIwerecoal-redunits,24unitsusefueloilornaturalgasforelectricitygeneration.Fortheseunits,fueloilisdesignatedashighsulfurfueloptionwhilenaturalgasisdesignatedasthelowersulfurfueloptionfornon-coalredunits. Thedeliveredpriceisthecentsperton,barrel,or1,000cubicfeetpaidatthetimeofdelivery.Thepriceincludesthepurchasingofthefuelandthetransportationcostsofshipment.ThepriceisconvertedtothepriceoffuelincentspermillionBtus.Aseparatedeliveredpriceisrequiredforbothhighsulfurandlowsulfurfuel,whichisaweightedaverageofalldeliveriesofeachcategoryofcoal.Sothehigh(low)sulfurpriceusedinthesimulationsistheweightedaveragedeliveredpriceforallhigh(low)sulfurcoaldeliveries.TheparameterusedinthesimulationsisactuallydollarspermmBtu,orcents/mmBtudividedby100. Anotherissueisthatnotallgeneratingunitspurchasedbothhighsulfurandlowsulfurfuel.Proxieswererequiredtodeterminethefuelcharacteristicsfacingeachunit,andwerersttakenfromothergeneratingunitsthatwereownedbythesamecompanywhereavailable.Ifaunitdidnothaveanotherunitunderthesamecompany,aproxywastakenfromthegeographicallyclosestplantbecauseshippingcostsareanimportantfactorindeterminingthedeliveredprice.RegionalvariationinpriceswillbemuchsmallervariationinpricesacrosstheU.S. 150

PAGE 151

ScrubbercostdatawasfoundintheEIA's\ElectricPowerAnnual1997".Thedatawasthehistoricalscrubbercostdata,includingstate-by-stateaverageinstalledcostsperkilowattofcapacity,averageoperationandmaintenancecostsintermsofmillsperkilowatt-hour,megawattsofcapacityforunitswithscrubbersinstalled,andsulfurremovaleciency. 151

PAGE 152

Anendogenousscrubberchoicecreatesanon-convexitythatmaynotresultinanallowancemarketequilibrium.Insuchacase,therewillbeeitherapositiveoranegativeexcessdemandforallowancesandanallowancepriceatwhichageneratingunitisindierenttoinstallingascrubber.Thiscouldbeconsidereda\quasi-equilibrium"wheretheexcessdemandiseitherboughtfromorsoldtoanallowancebrokeratthe\quasi-equilibrium"allowanceprice.ThetechnicalexplanationofthebisectioniterativeprocessusedtoconvergetoanequilibriumallowancepriceisdescribedindetailinAppendixB. 152

PAGE 153

First,considertheabilityofthesimulationmodelstorecreatetheresultsfrompreviousstudies.Simulation1recreatestheunconstrainedcost-minimizationresultsfromSotkiewicz(2003)andSotkiewiczandHolt(2005).Therewere17scrubbersthatwereinstalledasaresultoftheNewSourcePerformanceStandards(NSPS)andwereunrelatedtoTitleIV.Assumingthesescrubbersasgiven,thetotalcoststotheindustryofmeetingelectricitydemandwere$7.69billion.Thesecostswillbeusedasthebaselinetotalindustrycoststodeterminetotalindustrycompliancecostscomparabletopreviousstudies. Simulation3recreatestheemissionsconstrainedresultsfromSotkiewiczandHolt(2005)assumingthe46scrubbersthatwereinstalledin1996asgiven.Thesimulationresultsinanallowancepriceof$149.64,whichexactlyreplicatestheallowancepricefoundbySotkiewicz(2003)andSotkiewiczandHolt(2005).Totalindustrycostsare$8.23billionwhere29scrubberswereactuallyinstalledinresponsetoTitleIV.ThedierencebetweenSimulation3andSimulation1isthetotalindustrycompliancecostsofmeetingtheemissionsconstraint,whichis$541millionandsimilartotheresultsfoundinprevious 153

PAGE 154

TheallowancemarketdoesnotclearinSimulation4becausegeneratingunitsareallowedtomaketheirscrubberchoice.Thediscretenatureofscrubberinstallationleadstoanon-negativeexcessdemand.However,theexcesssupplyof53,576allowancesinSimulation4accountforlessthan1%ofthe5+millionallowancemarketandcouldbeassumedtobebankedforfutureuseorsoldtoanallowancebroker,whichcouldbeconsideredaquasi-equilibrium. 154

PAGE 155

Simulation6introducesfuelcontractconstraintstothetheemissionsconstrainedmodelwhileallowinggeneratingunitstomaketheirscrubberchoice.Thesimulationresultsinanallowancepriceof$210.74.Totalindustrycostsare$8.63billionwhere44scrubbersareinstalledinresponsetoTitleIV.TheminimumcompliancecostswhencomparedtoSimulation1are$939million,or$651millionhigherthanifcontractsarenottakenintoconsideration.TheminimumcompliancecostsareclosetotheactualcostfoundinSotkiewiczandHolt(2005)at$990millionandCarlsonetal.(2000)at$910million. Allowinggeneratingunitstochoosewhethertoinstallascrubberallowstheindustrytoloweritstotalcostsby$132million.Aswouldbeexpected,introducingthecontractconstraintresultsinmorescrubberinstallationsduetoTitleIVfrom27to44becausehigh-sulfurfuelcontractsincreasetheincentiveforaconstrainedgeneratingunittoinstallascrubber.AsinSimulation4,theallowancemarketdoesnotclearduetothediscrete,endogenousscrubberchoice.However,theexcessdemandof7,797allowancesaccountforlessthan0.2%ofthe5+millionallowancemarket,andcouldbeassumedtobeboughtfromanallowancebroker. TheactualtotalindustrycompliancecostsarefoundinSimulation7.Byusingtheactualemissionsandelectricityproductionforeachunit,itispossibletodetermineeachunit'sactualcoalmix.Theseactualdecisionsresultedinanallowancepricerange 155

PAGE 156

Third,the\true"compliancecostswilldierfromtheseestimatesbecausetheappropriatebaselinewasnotused.AsshownanalyticallyinSection4,comparingSimulation5andSimulation6toSimulation1isnotthemostappropriatemeasureofcompliancecosts.Thecontractconstraintsshouldbeincludedinboththebaselinesimulationandthepolicy-restrictedsimulation.Simulation2runsthesamemodelasinSimulation1exceptthatitincludescontractconstraintsandresultsintotalindustrycostsof$8.27billion.ThedierencebetweentotalindustrycostsinSimulation1andSimulation2aretheadditionalcostsduetocontractconstraints,whichare$582million.TheseadditionalcostswouldhaveresultedwithorwithouttheSO2TradingProgramandshouldnotbeconsideredcompliancecosts.Thisisakeyresultbecausethesecostswerelabeledcompliancecostsbypreviousstudieseventhoughthesecostsarearesultofgeneratingunitslockinginpricestoprotectfromtheuncertaintyofhighercoalpricesinthefuture. BycomparingresultsinSimulation5toSimulation2,theminimumcompliancecostsconsideringcontractsandgiventhescrubberchoicearefoundtobe$490.1million.\True"minimumcompliancecostsaremuchlower($610millionlower)oncethisadditionalconstraintisincludedinthemodel.InSimulation6,the\true"minimumcompliancecostsare$357.8million,or$581.6millionlessthanifcontractsareexcludedfromthemodel.Thesearesimilartotheleast-costresultsfoundbySotkiewicz(2003)at$340-$527million,SotkiewiczandHolt(2005)at$423-$553million,andCarlsonetal.(2000)at$571million. 156

PAGE 157

Simulation4usesatotalof2,651,410,732mmBtuofhighsulfurcoal(41.4%)and3,750,168,009mmBtuoflowsulfurcoal(58.6%).Simulation6usesatotalof2,614,773,732mmBtuofhighsulfurcoal(40.8%)and3,786,804,109mmBtuoflowsulfurcoal(59.2%).Includingcontractconstraintsintothemodelresultsina0.6%decreaseinhighsulfurcoalrelativetoSimulation4.Overallcoalusedoesnotappeartohavebeensignicantlyaltered.However,thisdoesnottellthewholestory.Contractconstraintsledto27units(6.3%ofaectedunits)usingasuboptimalcoalcombination.5ofthe27unitshadachangeofatleast98%ofcoaluseand15ofthe27hadatleasta50%changeincoaluse.14ofthe27unitshadanincreaseinhighsulfurcoalusewhile13unitshadandecreaseofhighsulfurcoaluse.Theconcernwiththecontractsisnotnecessarilythattheentire 157

PAGE 158

Second,comparingscrubberinstallationsinSimulation6toSimulation5willshowhowscrubberdecisionswouldbealteredifthescrubberdecisionismadeendogenousandcontractconstraintsareincludedinthemodel.Allowingthescrubberchoicetobeendogenousresultsin53generatingunitsalteringtheirscrubberchoice,including19scrubberstoberemovedand34tobeaddedforatotalof15moreinstalledscrubbers.27generatingunitsmaintainthesamescrubberchoices,but17ofthosewereinstalledforNSPS.So10scrubberinstallationsremainedthesame. Finally,itisimportanttoconsidertheimpactscontractconstraintshaveonaunit'sendogenousscrubberinstallationdecision,whichcanbedeterminedbycomparingSimulation6toSimulation4.33scrubberchoicesarealteredasaresultofcontractconstraints,including25unitsthatwillnowinstallascrubberand8unitsthatnolongerinstallascrubber.Ofthe25unitsthatchosetoinstallascrubber,20ofthemhadahighsulfurcoalcontract.Ofthe8unitsthatchosetonotinstallascrubber,all8ofthemhadalowsulfurcoalcontract.So28of33scrubberchoicesappeartohavebeendirectly 158

PAGE 159

3-4 ,contractconstraintsincreasescrubberinstallationsinOhio,Alabama,Florida,Indiana,Mississippi,andMissourianddecreasescrubberinstallationsinWestVirginia,NewYork,Pennsylvania,Wisconsin,Georgia,Kentucky,andNewJersey. AnunexpectedresultinSimulation6isthatscrubberinstallationsactuallyincreaserelativetoSimulation5.Thesearelikelyaresultoftheassumedscrubbercostestimatesbecausethehistoricalscrubbercostdatausedinthesimulationsislowerthantheengineeringcostestimates.Scrubbersappearcheapertoinstallthantheengineeringestimatesstate.Ofthe404generatingunitswithatleasta90%removalrate,31historicalcapitalcostestimatesarehigher,84arethesame,and289arelowerthantheestimatedengineeringcapitalcosts. AscanbeseeninTable 3-5 ,thereisasignicantdierenceinscrubberinstallations.Theuseofengineeringcostestimatesforscrubberinstallationresultsinadecreaseinscrubberinstallationsfrom44to25inSimulation4andfrom61to38inSimulation6becauseofthehighercostsinvolved.Noticethatintroducingcontractconstraintsresultsinadditionalscrubberinstallationsfrom25to38,whichisclosetotheactualinstallationsof46scrubbers.Compliancecostinterpretationsremainsimilartosimulationsusing 159

PAGE 160

160

PAGE 161

3.7.1Introduction Aplantleveldecision-makingmodelismorerealisticthanageneratingunitlevelmodeltodeterminecompliancedecisionsbecausecoaldeliveriesaremadeattheplantlevelwherethereareoftenmultiplegeneratingunits.Allgeneratingunitsatthatplantfacingthesamecoaluseoptions,includingsulfurcontents,heatcontents,anddeliveredpricesforbothspotmarketandcontractcoal.Thecontractconstraintsbecomemorecomplexinthismodelwherethesumofcontractcoaluseforallgeneratingunitsataplantmustcoverthecontractrequirement,Pni=1Ccif 161

PAGE 162

Threeadditionalfactorswillbeofgreatimportanceattheplantlevel.First,plantsareabletotradeallowancesbetweengeneratingunitsasneededtocoveremissionsatnoadditionalcosts.Theliteraturehasstatedthattradingallowancesbetweenitsunitsatthesameplanthasbeenacommonoccurrence,whichsupportstheuseofamodelthatconsidersdecisionsattheplantlevelinsteadofthegeneratingunitlevel. AplantmayhaveoneormoregeneratingunitsatitslocationthatarenotaectedbyPhaseIanddoesnothavetomeetanyemissionsrequirements.Owninganon-aected 162

PAGE 163

minzi;Ai;Csih;Csil;Ccih;CcilnXi=1ziPiz+PAAi+PshCsih+PslCsil+PchCcih+PclCcil subjectto...Aei+Ai(1ziri)(m)[CsihSsh+CcihSch+CsilSsh+CcilScl]i1 (CsihHsh+CcihHch+CsilHsl+CcilHcl)Dii2 (3{44) TheLagrangemultiplieronageneratingunit'semissionsconstraintisrepresentedbyi1.TheLagrangemultiplieronageneratingunit'sdemandconstraintisrepresentedbyi2.Coalcontractconstraintsrequireaplanttouseaminimumamountofeachcontractcoal 163

PAGE 164

SinceAicanbeeitherpositiveornegativebasedonthenetallowanceposition,( 3{46 )willholdwithequality.Theadditionalcosttothermofemittingonemoretonofemissionsisequivalenttotheallowanceprice,i1=PA,whichisthesameresultasfromthepreviousmodels. ThepartialderivativewithrespecttoCsihrepresentstheimpactaoneunitchangeinhighsulfurspotmarketcoalhasontheunit'stotalcosts. Ifthegeneratingunitusessomeamountofhighsulfurspotmarketcoal(Csih>0),then( 3{47 )holdswithequality. ThepartialderivativewithrespecttoCsilrepresentstheimpactaoneunitchangeinlowsulfurspotmarketcoalhasontheunit'stotalcosts. Similartohighsulfurspotmarketcoal,ifthegeneratingunitusessomelowsulfurspotmarketcoal(Cil>0),then( 3{48 )holdswithequality. 164

PAGE 165

Unlikewithhighsulfurspotmarketcoal,highsulfurcontractcoalwillbeimpactedbythehighsulfurcoalcontractconstraint(h).Ifthegeneratingunitusessomeamountofhighsulfurcontractcoal(Ccih>0),then( 3{49 )holdswithequality. ThepartialderivativewithrespecttoCcilrepresentstheimpactaoneunitchangeinlowsulfurcontractcoalhasontheunit'stotalcosts. Unlikewithlowsulfurspotmarketcoal,lowsulfurcontractcoalwillbeimpactedbythelowsulfurcoalcontractconstraint(l).Similartohighsulfurcontractcoal,ifthegeneratingunitusessomelowsulfurcontractcoal(Ccil>0),then( 3{50 )holdswithequality. 165

PAGE 166

3{47 )and( 3{48 )holdwithequality.Theadditionalcoststothegeneratingunitofusingonemoreunitoflowsulfurspotmarketcoalisequaltotheadditionalcostsofusingonemoreunitofhighsulfurspotmarketcoal. Sinceboth( 3{47 )and( 3{48 )holdwithequality,itispossibletondanexpressionfor(i2)ineachequationandsetthetwoexpressionsequaltoeachotherandderivetheallowanceshadowprice(i1)in( 3{52 ). Asdiscussedearlier,theallowanceshadowprice(i1)isanexpressionforageneratingunit'sMarginalCostofAbatementfromSwitchingfromhighsulfurspotmarkettolowsulfurspotmarketcoal(MCAs;si).From( 3{46 ),wealsoknowthattheactualallowancepriceequalstheallowanceshadowprice(PA=i1).Ascanbeseenin( 3{52 ),theallowanceprice(PA)equalstothemarginalcostofabatement(MCAs;si),whichistheincreaseincoalcostsfromswitchingfromhighsulfurspottolowsulfurspotmarketcoalperunitofreducedemissions. Themarginalcostsofabatementmayormaynotdieracrossgeneratingunits.Ifaplantdoesnotinstallanyscrubbers,thenMCAs;si=MCAs;sj8fi;jgforallgeneratingunitsandtheplant'schoiceswillbeidenticaltothegeneratingunits'decisionsinSection5.Thiscanbeseenin( 3{52 )wheretheonlywaythatthemarginalcostsofabatementcandieracrossgeneratingunitsisthroughthescrubberchoiceandascrubber'semissionscapturerate.Ifallgeneratingunitsatagivenplantbothinstallascrubberwhereri=rj, 166

PAGE 167

Marginalcostsofabatementwillonlydieracrossgeneratingunitsataplantif(1)aplantinstallsascrubberatsomebutnotallofitsgeneratingunits,or(2)aplantinstallsascrubberatallitsgeneratingunitsbutthescrubbershavedierentcapturerates(ri6=rj9fi;jg).Wewillintroducethescrubberchoice,andderiveitsimpactonaplant'schoiceslater. 3{47 )holdswithequalityand( 3{48 )holdswithweakinequalityresultingin( 3{53 ). Equation( 3{53 )statesthatadditionalcoststothegeneratingunitofusingonemoreunitofhighsulfurspotmarketcoalisweaklylessthanorequaltotheadditionalcostsofusingonemoreunitoflowsulfurspotmarketcoalinclusiveofemissionsanddemandrequirements. FollowingSection6.4.1wecanderivetherelationofPAtoMCAs;si.Bycomparingthesetwoexpressionsandsolvingfori2,yougetaninequalitycomparingi1=PAandMCAs;si.Sincethegeneratingunituseshighsulfurspotmarketcoal,theallowancepriceisweaklylessthanthemarginalcostofabatement(PAMCAs;si)asshownin( 3{54 ). 3{47 )holdswithequalityand( 3{48 )holdwithweakinequalityresultingin( 3{55 ).Theadditionalcoststothegeneratingunitofusingone 167

PAGE 168

OnceagainwecanderivetherelationofPAtoMCAs;si.Bycomparingthesetwoexpressionsandsolvingfori2,yougetaninequalitycomparingi1=PAandMCAs;si.Sincethegeneratingunituseslowsulfurspotmarketcoal,theallowancepriceisweaklygreaterthanthemarginalcostofabatement(PAMCAs;si)asshownin( 3{56 ). First,ifageneratingunitfacesPA
PAGE 169

3{41 ).Assumingonlylowsulfurspotmarketcoalusetomeetdemand,themaximumamountoflowsulfurcoalisexpressedin( 3{59 ). ReplacingCsilin( 3{41 )withtheexpressionin( 3{59 )forCs;MAXilgivesanexpressionfortheminimumallowanceexcessdemandin( 3{60 ).Ageneratingunit'sminimumexcessdemandmustcoverthedierencebetweenitsinitialallowanceallocation(Aei)andtheamountofallowancesneededtocovertheunit'sminimumactualemissions[EMINi=(1ziri)(m)(Ssil)(Di IfPA=MCAs;si,ageneratingunitmayuseanycombinationofhighsulfurspotmarketcoalandlowsulfurspotmarketcoalandleadtoanylevelofexcessdemandintherange(EMINiAei;EMAXiAei).TheallowanceexcessdemandcanberepresentedbyAi=(EMAXi(1)EMINiAei)wheretheconstant2[0;1].Aunitthatisindierentbetweenfuelswitchingandallowancespurchasingcouldbeeitheranetbuyeroranetseller. Combiningtheexcessdemandsforeachofthethreecasescreatestheexcessdemandcorrespondenceshownbelow.Ai=8>>>><>>>>:AMAXiifPA>MCAs;siAMAXi(1)AMINiifPA=MCAs;si82[0;1]AMINiifPA
PAGE 170

Theuseofcontractcoalwillbebasedontherelativemarginalcostsofusingcontractcoalateachgeneratingunitoperatedbyaplant.Aplantwillchoosetousecontractcoalatthegeneratingunitthatwillresultinthelowestincreaseintheplant'stotalcosts.Tomakethesecostcomparisons,itisnecessarytoderivethemarginalcostsofabatementforthegivencombinationofspotandcontractcoalandcomparethemacrossgeneratingunits.Initiallywewillignoreaplant'sscrubberchoiceforeachgeneratingunitandaplantmakesthesamechoiceforallitsgeneratingunits(zi=zj=0,orzi=zj=1andri=rj).Undertheseconditions,themarginalcostsofabatementareidenticalacrossallgeneratingunits,anditdoesnotmatteratwhichofthesegeneratingunitsthecontractcoalisused.Thiswillbeshownseparatelyforbothahighsulfurandlowsulfurcoalcontractbecausetheuseofhighsulfurcontractcoalisindependentoflowsulfurcontractcoalandvisaversa. 3{61 ).Themarginalcostofabatementfromswitchingfromhighsulfurcontractcoaltolowsulfurspotmarketcoal(MCAc;si)willbethesameforallgeneratingunitsthatmakethesamescrubberinstallationchoice(zi),includingthesamescrubber 170

PAGE 171

Allunitsatwhichaplantdoesnotinstallascrubberhavethesamemarginalcostofabatementofswitchingfromhighsulfurcontractcoaltolowsulfurspotmarketcoal(MCAc;si=MCAc;sj).AlsoallunitsataplantthatinstallascrubberwiththesamecaptureratewillhaveidenticalMCAc;si. 3{62 ).TheMCAs;ciwillbethesameforallgeneratingunitsthatmakethesamescrubberinstallationchoice,includingthesamescrubbertechnologywiththesamecapturerate.Psh+i1(1ziri)(m)(Ssh)+i2Hsh=Pcl+i1(1ziri)(m)(Scl)+i2Hcll Allunitsatwhichaplantdoesnotinstallascrubberhavethesamemarginalcostofabatementofswitchingfromhighsulfurspotmarketcoaltolowsulfurcontractcoal(MCAs;ci=MCAs;cj).AlsoallunitsataplantthatinstallascrubberwiththesamecaptureratewillhaveidenticalMCAs;ci. 171

PAGE 172

Theemissionsconstraintdoesnotbind(i1=0)fortheseunits,whichmeansthat\non-aected"generatingunitswillprefertousethecoaltypewiththelowestdeliveredprice.Formostplantsthiswillbehighsulfurcoal,particularlythoselocatedintheEasternU.S. 3{63 )and( 3{64 ). @Csih=Pshi2Hsh0;=0ifCsih>0(3{63) and @Csil=Psli2Hsl0;=0ifCsil>0(3{64) Assumingthatageneratingunitusesbothhighandlowsulfurspotmarketcoal,weuse( 3{63 )and( 3{64 )toderive( 3{65 ).Ifageneratingunitusesbothhighandlowsulfurspotmarketcoal,thenthepriceperunitofheatinputisequalforbothcoaltypes.Aplantwillnolongermakeit'sgeneratingunitlevelfuelchoicesbasedonitsmarginalcostofabatementbecauseitnolongerisconcernedaboutabatingemissions. Byusingtherst-orderconditions,wecandeterminewhenageneratingunitwillonlyuseonetypeofspotmarketcoal.Ifageneratingunitusesonlyhighsulfurspotmarket 172

PAGE 173

3{63 )holdswithequalityand( 3{64 )remainsweaklygreaterthanzeroandlowsulfurcoalisweaklymoreexpensive(Psh 3{64 )holdswithequalityand( 3{63 )remainsweaklygreaterthanzeroandhighsulfurspotmarketcoalismoreexpensive(Psh First,assumethattwogeneratingunits(Unit\i"andUnit\j")arebothaectedunits.Aplantwillbeindierenttousinghighsulfurcontractcoalateitherunitwhentherst-orderconditionsforusinghighsulfurcontractcoalareequalacrossgeneratingunits.Pch+i1(1ziri)(m)(Sch)i2Hchh=Pch+j1(1zjrj)(m)(Sch)j2Hchh NowassumethatUnit\i"isanaectedunitanditsemissionsconstraintbinds(i1>0)whileUnit\j"isnotaectedandtheemissionsconstraintwillnotbind(j1=0).Underthiscondition,theadditionalcostsofusinghighsulfurcontractcoalatanaectedunitisgreaterthanatanon-aectedunit.Pch+i1(1ziri)(m)(Sch)i2Hchh>Pchj2Hchh)i1(1ziri)(m)(Sch)i2Hch>j2Hch 173

PAGE 174

174

PAGE 175

First,considerthemarginalcostofabatementofswitchingfromhighsulfurspotmarketcoaltolowsulfurspotmarketcoal(MCAs;si)withascrubberin( 3{68 ),andcompareittothemarginalcostofabatementwithoutascrubberin( 3{69 ).AsinChapter2,ascrubberdecreasesthesavingsfromswitchingspotmarketfuelsbecauseascrubbercausestheemissionsreductiontobesmallerthanwithoutascrubber. Second,considerthemarginalcostofabatementofswitchingfromhighsulfurcontractcoaltolowsulfurspotmarketcoal(MCAc;si)withascrubberin( 3{70 ),andcompareittothemarginalcostofabatementwithoutascrubberin( 3{71 ).Asinthecaseofonlyspotmarketcoal,installingascrubberdecreasesthesavingsfromswitchingfuels.Inthecaseofahighsulfurcontract,itislesscostlytoaplanttousehighsulfurcoalataparticulargeneratingunitifithasascrubber. Third,considerthemarginalcostofabatementofswitchingfromhighsulfurspotmarketcoaltolowsulfurcontractcoal(MCAs;ci)withascrubberin( 3{72 ),andcompareittothemarginalcostofabatementwithoutascrubberin( 3{73 ).Asinthetwocasesabove,installingascrubberdecreasesthesavingsfromswitchingfuels.Inthecaseofalow 175

PAGE 176

Foreachcombinationofcoaltypes,installingascrubberdecreasesthesizeofthedenominator,whichincreasesMCAi.AsinChapter2,ascrubbergreatlydecreasesthesavingsfromswitchingfuels.AhigherMCAiincreasestherangeofallowancepricesatwhichaplantwillprefertopurchaseallowancesinsteadofswitchingfuelsbyloweringthepriceatwhichageneratingunitwillbeindierentbetweenpurchasingallowancesandswitchingfromhighsulfurtolowsulfurcoal.Anyunitoperatedbyaplantthathasascrubberwillhavealargermarginalcostofabatementthananyunitwithoutascrubber(MeCAi>MCAj). Thedecisionisbasedsolelyonaunit'sACAibecausethemarginalcostsofabatementareequalacrossgeneratingunitswithnoscrubbersinstalledattheplant. 176

PAGE 177

AssumethatallgeneratingunitsoperatedbyaplantareaectedbyPhaseI,eachunitusesitscostminimizingcombinationofcoalandallowancesbasedonitsscrubberchoice,andtheunitsaresortedbyACAifromsmallesttolargest(ACA1
PAGE 178

Generalizingthiscondition,aplant'sdecisiontoinstallascrubberatgeneratingunit\m".Aplantwillinstall\m"scrubbersatthe\m"largestgeneratingunitsifC(z1=1;z2=1;:::;zm=1;zm+1=0;:::;zn=0)
PAGE 179

Assumingnocontractcoal,wecanderivetheallowancepriceatwhichaplantisindierenttoinstallingascrubberforagivengeneratingunitin( 3{75 ). (bAieAi)(3{75) Aspecialcaseexistsfor( 3{75 )inwhichthegeneratingunitisanon-aectedunit,whichresultsinPA!1becausetheunitdoesnothavetocoveritsemissionswithallowancesandgainsnothingfrominstallingascrubber.AnindierencepriceofinnityimpliesthatascrubberwillneverbeinstalledatUnit\i"ifitanunaectedunit. (bAieAi)=Piz+Psh(eCsihbCsih)+Psl(eCsilbCsil) 0=1(3{76) 179

PAGE 180

Aplantwithabindinghighsulfurcoalcontracthasagreaterincentivetoinstallascrubberateachofitsgeneratingunitsbecauseitmustusesomehighsulfurcoal,evenifitwouldprefertouselowsulfurcoalatallgeneratingunits.AhighsulfurcoalcontractdecreasestheindierencepriceatwhichaunitwillinstallascrubberfromPAto(PA)foreachgeneratingunit.TheincentivemaynotbelargeenoughtoresultinascrubberinstallationatitsgeneratingunitwiththelowestACAi.Insuchascase,aplantisindierenttousinghighsulfurcontractcoalatanyofitsgeneratingunitsbecauseallunitshavethesameMCAc;si. AbindinghighsulfurcoalcontractresultsinadecreaseinPSA. AtPSAwhere... 180

PAGE 181

ThisisthesameresultasinChapter2whereahighsulfurcoalcontractresultsinaninecientcoalusecombinationofhighandlowsulfurcoal.Theindierencepriceatwhichaplantwillinstallascrubberatagivengeneratingunitwillweaklydecreasewhenaplantchoosestousehighsulfurcontractcoalatthatunit. 181

PAGE 182

AbindinglowsulfurcoalcontractresultsinanincreaseinPSA. AtPSAwhere... Iflowsulfurcontractcoalisatleastasexpensiveaslowsulfurspotmarketcoal,weknowthat... ThisisthesameresultasinChapter2.Alowsulfurcoalcontractresultsinaninecientcoalusecombinationofhighandlowsulfurcoal.Theindierencepriceatwhichaplantwillinstallascrubberatagivengeneratingunitwillweaklyincreasewhenaplantchoosestouselowsulfurcontractcoalatthatunit. 182

PAGE 183

3{81 ).Theamountofcontractcoalisthesamenomatterthescrubberchoiceateachgeneratingunitmadebyaplant. (bA1eA1) (3{81) Second,thetotalcoststotheplantofnotinstallinganyscrubbersmustbelessthanthetotalcostsofinstallingscrubbersatbothgeneratingunits. 3{82 ). (bA1eA1)+(bA2eA2) (3{82) TheminimumofthetwoPSAwillbetheallowancepriceatwhichaplantisindierenttoinstallingascrubberandnotinstallinganyscrubbers.Noticethattheallowanceprice 183

PAGE 184

3{83 ). Second,thetotalcostsofinstallingonegeneratingunitmustbelowerthanthetotalcostsofinstallingtwoscrubbers. 3{84 ). (bA2eA2)>PSA Theallowancepricemustbebetweenthesetwoindierenceprices,whichresultsinPSAPAforUnit2. 184

PAGE 185

3{85 ). (bA2eA2)
PAGE 186

3-7 Aplantwithnocontractcoalwillprefertoinstallnoscrubbersiftheallowancepriceisbelow$194.48.Anallowancepricebetween$194.48and$230.89willleadaplanttoinstallascrubberatUnit1,butnotatUnit2.Aplantwillinstallscrubbersatbothunitsiftheallowancepriceisgreaterthan$230.89. Nowconsidertheplanthasahighsulfurcoalcontractfor28.92%ofcoaluse(13,855,269mmBtu),whichwillaltertheindierenceatwhichaplantwillinstallascrubberatUnit1.WehavealreadyshownthathighsulfurcontractcoalwillbeusedatthegeneratingunitwiththelowestACAi,whichisUnit1.TherequireduseofhighsulfurcoalwilllowertheallowanceindierentpricetoinstallingascrubberatUnit1to$140.69.TherewillbenochangeintheindierencepricetoinstallingascrubberatUnit2becausethehighsulfurcoalcontractdoesnotbindforthatgeneratingunit. Nowconsidertheplanthasahighsulfurcoalcontractforallofitscoaluseinbothgeneratingunits(47,913,973mmBtu).Theindierencepricesforbothgeneratingunitswillbothdecreaseasaresult.TheallowancepriceatwhichtheplantwillinstallascrubberatUnit1willagainbe$140.49.TheindierencepriceforUnit2willdecreaseto$153.61.Ahighsulfurcoalcontractforallofageneratingunit'scoalrequirementsresultsinlargereductionsoftheallowancepriceatwhichaplantwillbeindierenttoinstallingascrubberforbothUnit1($54or28%)andUnit2($77or33%). 186

PAGE 187

3{87 ). (bA1eA1) (3{87) 3{88 ). (bA1eA1) (3{88) 187

PAGE 188

Aplantlevelmodelismorerealisticthanageneratingunitlevelmodelbecausecoaldeliveriesaremadeattheplantlevelwherethereareoftenmultiplegeneratingunits.Operatingmultipleunitsallowsaplanttorelaxitscontractconstraintbecausethereareadditionaldegreesoffreedominfuelusedependingoneachunit'semissions,demand,andcoalcontractconstraints.Also,plantsareabletotradeallowancesbetweengeneratingunitsasneededtocoveremissionsatnoadditionalcost.Aplantmayalsohaveoneormore\non-aected"generatingunits,whichfacenoemissionsconstraint.Anyhighsulfurcontractcoalcanbeusedatthesenon-aectunitswithoutanynegativenancialrepercussionsduetothepolicy.Thesefactorscanleadtoplant-levelchoicesthatdonotminimizeeachgeneratingunit'stotalcosts. Anallowancemarketequilibriumwillexistonlyifthediscretescrubberchoiceisgiven.Allowingforanendogenousscrubberchoicemakesitimpossibletoguaranteeanequilibrium,althoughonemaystillmayexist.Bindingfuelcontractsmayalteraunit'scompliancedecisionsandexcessdemand.Alteringcompliancedecisionscouldleadtobothanalteredallowancemarketpriceandanincreasetotalindustrycompliancecosts. Generatingunit-levelsimulationswereabletoeectivelyreplicatetheresultsfrompreviousstudiesandshowthatfuelcontractscanexplainaportionofthepreviouslyunexplainedexcesscompliancecosts.Simulatingtheleast-costcompliancechoiceswithout 188

PAGE 189

Aplant'sdecision-makingprocessmaynotminimizecostsforeachgeneratingunitbecauseaplant'sconcernisbasedonthecombinedcostsofallgeneratingunitsunderitsoperationalcontrol.Thechoicetoinstallascrubberisbasedonthecharacteristicsofalltheunitsataplant,notjusttheunitatwhichthescrubbermaybeinstalled.OncetheorderofpreferredscrubberinstallationsisdeterminedbasedontheACAi,aplantisabletomakeitscostminimizingfuelchoices. Aplantlevelmodelismorerealisticthanageneratingunitlevelmodelbecausecoaldeliveriesaremadeattheplantlevelwherethereareoftenmultiplegeneratingunits.Operatingmultipleunitsallowsaplanttorelaxitscontractconstraintbecausethereareadditionaldegreesoffreedominfuelusedependingoneachunit'semissions,demand,andcoalcontractconstraints.Also,plantsareabletotradeallowancesbetweengeneratingunitsasneededtocoveremissionsatnoadditionalcost.Aplantmayalsohaveoneormore\non-aected"generatingunits,whichfacenoemissionsconstraint.Anyhighsulfurcontractcoalcanbeusedatthesenon-aectunitswithoutanynegativenancialrepercussionsduetothepolicy.Thesefactorscanleadtoplant-levelchoicesthatdonotminimizeeachgeneratingunit'stotalcosts. 189

PAGE 190

Example:ContractCoalDistribution Plant-LevelConstraintTotalmmBtuUnitDemandScrubber HighSulfurContract300,0001200,000YesLowSulfurContract300,0002200,000NoDemand600,0003200,000No CoalDist'nbyUnitHighSulfurCoalLowSulfurCoal Unit1200,0000Unit250,000150,000Unit350,000150,000 Table3-2. SulfurConversionbyFuelType FuelEmissionsConversionFactorBaseline Bituminous(38*SulfurContent)/(mmBtuperton)=lbs.SO2/mmBtuSub-bituminous(35*SulfurContent)/(mmBtuperton)=lbs.SO2/mmBtuAnthracite(39*SulfurContent)/(mmBtuperton)=lbs.SO2/mmBtuLignite(30*SulfurContent)/(mmBtuperton)=lbs.SO2/mmBtuFuelOil#2(144*SulfurContent)/(mmBtuper1,000bbl.)=lbs.SO2/mmBtuFuelOil#6(162*SulfurContent)/(mmBtuper1,000bbl.)=lbs.SO2/mmBtuNaturalGas(0.60lbs.SO2/mmCF)/(mmBtupermmCF)=lbs.SO2/mmBtu 190

PAGE 191

SimulationResults Sim.EmissionsContractScrubberScrubbersPAAiTotalCostsCompl.CostsComp.CostsConstraintConstraintChoiceInstalledvs.(Sim.1)vs.(Sim.2) (1)NONOGiven17NANA$7,685,800,000NANA(2)NOYESGiven17NANA$8,267,400,000NANA(3)YESNOGiven46$149.640$8,226,500,000$540,700,000NA(4)YESNOChosen44$155.58-53,576$7,974,100,000$288,300,000NA(5)YESYESGiven46$206.700$8,757,500,000$1,071,700,000$490,100,000(6)YESYESChosen61$210.747,797$8,625,200,000$939,400,000$357,800,000(7)--Actual46NANA$8,983,500,000$1,297,700,000$716,100,000

PAGE 192

ImpactofContractConstraintonScrubberChoice StateDecreaseStateIncrease WV-4MO2NY-2MS2PA-2IN3WI-2FL4GA-1AL7KY-1OH10NJ-1 Table3-5. SimulationswithEngineeringData Sim.AiPAScrubbersIndustryCostsComp.CostsComp.CostsInstalled(vs.1)(vs.2) 1NANA17$7,685,800,000NANA2NANA17$8,267,400,000NANA30$149.6446$8,239,000,000$553,200,000NA41,972$214.8325$8,010,200,000$324,400,000NA50$207.6046$8,770,900,000$1,085,100,000$503,500,00062,611$238.5538$8,686,300,000$1,000,500,000$418,900,0007NANA46$8,995,600,000$1,309,800,000$728,200,000 Table3-6. ImpactsofaReductionintheAllowanceAllocationof10% Simulation4InitialAllocationAllocationMinus10%Increase IndustryComplianceCosts$288,300,000$381,900,000$93,600,000ScrubbersInstalled446117AllowancePrice$155.58$189.43$33.85 Simulation6InitialAllocationAllocationMinus10%Increase IndustryComplianceCosts$357,800,000$476,600,000$118,800,000ScrubbersInstalled617312AllowancePrice$210.74$222.22$11.48 Table3-7. MathExample:TwoAectedUnits LowSulfurSpot$1.200.625%24mmBtuHighSulfurSpot$1.102.580%24mmBtuHighSulfurContract$1.152.580%24mmBtu GeneratingUnitPizriDemand(mmBtu) Unit1$2,719,32090%13,855,269Unit2$7,298,41290%34,058,704 ScrubberInstallationPSA:NoContractPSA:Unit1DemandPSA:Unit1andUnit2Demand None<$194.48<$140.69<$140.69Unit1($194.48,$230.89)($140.69,$230.89)($140.69,$153.61)Unit2$230.89$230.89$153.61

PAGE 193

ExcessDemandCorrespondence Figure3-2. ImpactofHighSulfurCoalContract 193

PAGE 194

ImpactofLowSulfurCoalContract Figure3-4. ExcessDemandCorrespondencewithScrubberChoice 194

PAGE 195

ImpactsofHighSulfurCoalContract Figure3-6. ImpactsofLowSulfurCoalContract 195

PAGE 196

GivenScrubberChoice:ShiftfromHighSulfurContract Figure3-8. GivenScrubberChoice:ShiftfromLowSulfurContract 196

PAGE 197

WithScrubberChoice:ShiftfromHighSulfurContract 197

PAGE 198

WithScrubberChoice:ShiftfromLowSulfurContract 198

PAGE 199

Withtheemissionsconstraintfromtheprogram,aunitwillminimizeitscostsofcoaluse,netallowancepurchases,andscrubberinstallation. Thedierencebetween( A{1 )and( A{2 )isthetotalcompliancecostsresultingfromtheprogram,whichincludesthechangeincoalcosts,changeinthenetallowancepurchases,andscrubberinstallationcosts. Second,consideraunit'stotalcostswithacoalcontractconstraint.Withouttheprogramrestrictionsonemissions,aunitwillsimplyminimizeitscostsofmeetingelectricitydemandbyusingallthethecoalundercontract,andcovertheremainderofitscoaldemandwiththecoalwiththelowestpriceperunitofheatcontent. Withtheemissionsconstraint,aunitwillminimizeitscostsforcoaluse,netallowancepurchases,andscrubberinstallationgivenitsemissionsandcoalcontractconstraint. 199

PAGE 200

A{4 )and( A{5 )isthetotalcompliancecostsresultingfromtheprogram,whichincludesthechangeinspotmarketcoalcosts,netallowancepurchases,andscrubberinstallation.Thecostsfromcontractcoalcanceloutbecausecontractcoalusewillbethesamebothwithandwithouttheprogram. Thesucientconditionsunderwhichacoalcontractconstraintwillincreaseordecreaseaunit'scompliancecostscanbederivedfromthedierenceincompliancecostswithandwithoutacoalcontract( A{6 )minus( A{3 ).hbziPiz+PAbAi+Psih(bCsihbCs0ih)+Psil(bCsilbCs0il)i Forsimplicity,assumethatthescrubberchoiceasagivenandhighsulfurspotmarketcoalisrelativelycheaperthanlowsulfurspotmarketcoal.Sowithoutanemissionsconstraintaunitwillprefertousethecheaperhighsulfurcoal.Proposition1canbeprovenbyconsideringthechangeincompliancecostsin( A{8 )resultingfromacoalcontract.FirstconsiderahighsulfurcoalcontracttoshowProposition1(i)andProposition1(ii)hold. 200

PAGE 201

Thechangeincompliancecostswillbe: ThersttermispositivebecausebAMINiAMINi.ThesecondtermisalsopositivebecauseCs;MAXihbCs;MAXih.ThethirdtermisnegativebecausebCs;MAXilCs;MAXil. Nowllinforcoaluse:Cs;MAXih=Di Nowllinforthenetallowanceposition:AMINi=Di 201

PAGE 202

Allcoaluseremainsthesame,whichmeansthechangeincompliancecostswillbetheincreaseincostsfromadditionalallowances.IfScih 202

PAGE 203

Thechangeincompliancecostswillbetheincreaseincostsfromtheincreaseinaunit'snetallowanceposition.IfScil Thechangeincompliancecostsisthechangeincostsfromthechangeinnetallowanceposition.Nowllinforthenetallowanceposition:AMAXi=Di SinceScil 203

PAGE 204

A{5 )-( A{2 ))andsplititintotwocomponents,thechangeincompliancecostsandthechangeinfuelcosts.Forsimplicity,assumetheconditionsinProposition1(i)hold. Assumethataunitfacesahighsulfurcoalcontract,andpreferstoswitchtolowsulfurcoalusetomeetitsemissionsrequirementinsteadofpurchasingallowancesorinstallingascrubber. Tobeabletointerpretthisexpression,itisnecessarytoaddandsubtract(Psih Thersttermisthechangeinhighsulfurcoalcostsfromusingthecontractcoalinsteadofspotmarketcoal.Thesearenotchangesincompliancecostsbecausetheywilloccurwithorwithouttheprogram.Theremainingtermsarethechangeincompliancecostsresultingfromtheprogram. {z }ChangeinFuelCosts+PA(bAMINiAMINi)+(Psih {z }ChangeinComplianceCosts(A{19) Byllinginforthecoaluse,thelasttermsgivethesameexpressionforthechangeincompliancecostsasintheproofofProposition1(i). Aunit'scompliancecostsincreaseifScih 204

PAGE 205

Assumingascrubberisinstalled,thecost-minimizingcombinationofinputsisexpressedbelow. Noticethatcontractcoaluseandthecostofinstallingascrubbercanbeignoredbecauseallareconstants. 205

PAGE 206

Theorem5.1:Assumingthescrubberchoice(zi)asgiven,amarketequilibriumexists. Kakutani'sFixedPointTheoremthatstates:IfXisanon-empty,compact,convexsubsetofRmandiffisanuppersemi-continuouscorrespondencefromXintoitselfsuchthat(8x2X)thesetf(x)isnon-emptyandconvex,thenfhasaxedpoint(thereisanx2f(x)). Itisnecessarytoshowthatthesetofpricesiscompact(closedandbounded).Denethelowerboundonpricetobezerobecauseitisnecessarytohaveapositiveprice.Denetheupperboundonpricetobe 206

PAGE 207

PA],whichisnon-emptybydesign.Sincethereisanupperbound,alowerbound,andbothboundsareincludedintheset(closed),thesetiscompact. Sincethesetofpricesandmarketexcessdemandsarecompactandthemarketexcessdemanddependsontheallowanceprice,themarketexcessdemand(Am(PA))isamappingfrom[0; PA]intoX.Itisnecessarytoshowthatexcessdemandisnon-empty,convexvaluedateachPA2[0; PA],anduppersemi-continuous.Itissucienttoshowthateachgeneratingunit'sexcessdemandcorrespondenceisnon-empty,convex,anduppersemi-continuousbecausethesumofanitenumberofnon-empty,convex,uppersemi-continuouscorrespondencesisalsonon-empty,convex,anduppersemi-continuous. Ageneratingunits'excessdemand(Ai(PA))aredenedtobemappingsfrom[0; PA]intoX,whichmakesthemnon-emptybyconstruction.Ageneratingunit'sexcessdemandcorrespondenceisclosedandboundedintheinterval[AMINi;AMAXi],whichmakesitacompactset.Thesetofexcessdemandsisconvexbecausetheaverageofanytwoexcessdemandvaluesisalsointheset.Foracorrespondencetobeuppersemi-continuous,theconvergenceexcessdemandvalueofanypricesequencemustalsobeinthecorrespondence.Sinceeverypossiblepricesequenceconvergestoavaluethatisintheexcessdemandcorrespondence(seeFigure B.2 ),eachunit'sexcessdemandcorrespondenceisuppersemi-continuous. NowwemustdenethemappingfromXintothesetofprices[0; PA]as(x)where... PA]:PAx=maxQ2[ PA]:PAx=0ifotherwise Since[0; PA]isnon-empty,(x)mustbenon-emptyaswell.Themappingisconvexvaluedsinceformarketexcessdemandequaltozero(x=0),(x)=[0; PA].Ifthereisa 207

PAGE 208

PA]iscompact,thegraphof(x)isclosed,whichimplies(x)isuppersemi-continuous. NowdeneF(x;PA)=(x)Am(PA).Since(x)andAm(PA)satisfyallpropertiesneededtoapplyKakutani'sFixedPointTheorem,thereexistsaxedpoint(x;PA)suchthatPA2[0; PA]andx2XsuchthatPA2(x)andx2Am(PA). Oncethescrubberchoiceisintroducedintothedecision-makingprocess,thecorrespondencebecomesmorecomplex,asseeninFigure B.2 .Anequilibriummayinfactexist,butthereisnowaytoguaranteeanequilibriumbecausetheexcessdemandcorrespondenceisnolongeraconvexset.Theaverageofthetwoexcessdemandvalues(ASMAXiandAMINi)isnotintheexcessdemandcorrespondence. Ifmarketexcessdemandispositive,theallowancepriceistoolowandtheallowancepriceisincreasedbyone-halfthedierencebetweentheupperandlowerlimits.Theoldpricenowbecomesthenewlowerlimitwhiletheupperlimitremainsthesame.Ifthemarketexcessdemandisnegative,theallowancepriceistoohighandtheallowancepriceisdecreasedbyone-halfthedierencebetweentheupperandlowerlimits.Theoldpricebecomesthenewupperlimitandthelowerlimitremainsthesame.Inthiscase,theupper 208

PAGE 209

Theprogramisthenrunagainwiththenewallowanceprice,eachiterationdecreasingthedierencebetweentheupperandlowerlimitsbyhalfuntiltheprogramconvergestoanallowanceprice.Oncetheprogramconverges,itmustensureamarketexcessdemandofzero.Aconcernisthattheprogramtendstopushaunit'schoicestowardsacornersolutionwherethemarketmaynotclear. Giventhescrubberchoice,theallowancepricewillconvergetoavalueequaltoatleastonegeneratingunit'sMCAs;si,whichallowsthosermschoicestobeshiftedtoaninteriorsolutiontoclearthemarketwithoutalteringtheunit'stotalcostsortotalindustrycostsbecausetheunitisindierenttopurchasingallowancesorswitchingfuelsfromhightolowsulfurcoalattheequilibriumallowanceprice(PA). AllowingforthescrubberchoicewillresultintheconvergenceofPAatanallowancepricewhereageneratingunitisindierenttoinstallingascrubber.Inthiscase,thereisnotatrueequilibriumandmustconsideritaquasi-equilibriumasdescribedabove. 209

PAGE 210

UpperSemi-ContinuousCorrespondence FigureB-2. CorrespondencewithScrubberChoice 210

PAGE 211

[1] T.Arimura.AnempiricalstudyoftheSO2allowancemarket:eectsofPUCregulations,J.Environ.Econ.Manage.,44(2)(2002)271{289. [2] R.E.Baldwin,C.S.Magee,Istradepolicyforsale?Congressionalvotingonrecenttradebills,PublicChoice105(2000)79{101. [3] L.Bird,M.Bolinger,T.Gagliano,R.Wiser,M.Brown,B.Parsons.PoliciesandmarketfactorsdrivingwindpowerdevelopmentintheUnitedStates,EnergyPol.33(11)(2005),1397{1407. [4] D.R.Bohi.Utilitiesandstateregulatorsarefailingtotakeadvantageofemissionallowancetrading,ElectricityJ.,7(2)(1994)20{27. [5] D.R.Bohi,D.Burtraw.Utilityinvestmentbehaviorandtheemissiontradingmarket,ResourcesandEnergy,14(1992)129{153. [6] D.R.Bohi,D.Burtraw.SO2allowancetrading:Howexperienceandexpectationsmeasureup,DiscussionPaper97{24,ResourcesfortheFuture,Washington,D.C.,1997. [7] M.Bolinger,R.Wiser.2004.Theimpactofstatecleanenergyfundsupportforutility-scalerenewableenergyprojects,LawrenceBerkeleyNationalLaboratoryandCleanEnergyStatesAlliance. [8] M.Bolinger,R.Wiser,G.Fitzgerald.2006.Theimpactofstatecleanenergyfundsupportforutility-scalerenewableenergyprojects,LawrenceBerkeleyNationalLaboratoryandCleanEnergyStatesAlliance. [9] M.Bolinger,R.Wiser,G.Fitzgerald.Anoverviewofinvestmentsbystaterenewableenergyfundsinlarge-scalerenewablegenerationprojects,ElectricityJ.18(1)(2005)78{84. [10] M.Bolinger,R.Wiser,L.Milford,M.Stoddard,K.Porter.Cleanenergyfunds:Anoverviewofstatesupportforrenewableenergy,EnergyEciencyandRenewableEnergy,OceofPowerTechnologies,DepartmentofEnergy,2001. [11] D.Burtraw.Costsavingssansallowancetrades?EvaluatingtheSO2emissiontradingprogramtodate,DiscussionPaper95{30,ResourcesfortheFuture,Washington,D.C.,1995. [12] D.Burtraw,Evans,A.Krupnick,K.Palmer,R.Toth.EconomicsofpollutiontradingforSO2andNOX,DiscussionPaper05{05,ResourcesfortheFuture,Washington,D.C.,2005. [13] C.Carlson,D.Burtraw,M.Cropper,K.Palmer.Sulfurdioxidecontrolbyelectricutilities:Whatarethegainsfromtrade?,J.ofPoliticalEconomy,108(6)(2000)1292{1326. 211

PAGE 212

C.Chen,R.Wiser,M.Bolinger,2007.Weightingthecostsandbenetsofstaterenewablesportfoliostandards:Acomparativeanalysisofstate-levelpolicyimpactprojections,LawrenceBerkeleyNationalLaboratoy. [15] J.S.Coggins,J.R.Swinton.Thepriceofpollution:AdualapproachtovaluingSO2allowances,J.Environ.Econ.Manage.,30(1996)58{72. [16] DatabaseofStateIncentivesforRenewableEnergy.Staterenewableenergypolicysummaries,Website,AccessedMarch2004{March2006,http://www.dsireusa.org. [17] DatabaseofUtility-ScaleRenewableEnergyProjects.LawrenceBerkeleyNationalLaboratoryandCleanEnergyStateAlliance(CESA),Website,AccessedDecember2006,http://eetd.lbl.gov/ea/ems/cases/Large Renewables Database.xls. [18] A.D.Ellerman,R.Schmalensee,P.L.Joskow,J.P.Montero,E.M.Bailey.EmissionstradingundertheU.S.acidrainprogram:Evaluationofcompliancecostsandallowancemarketperformance,CenterforEnergyandEnvironmentalPolicyResearch,MassachusettsInstituteofTechnology. [19] A.D.Ellerman,etal.Marketforcleanair:theU.S.acidrainprogram,CambridgeUniversityPress,NewYork,2000. [20] A.D.Ellerman,J.P.Montero.Thedecliningtrendinsulfurdioxideemissions:Implicationsforallowanceprices,J.Environ.Econ.Manage.,36(1998)26{45. [21] EnergyInformationAdministration.Electricpowerannual1996:Vol.2,DOE/EIA{0348(96)/2,February1996. [22] EnergyInformationAdministration.TheeectsofTitleIVoftheCleanAirActAmendmentsof1990onelectricutilities:Anupdate,DOE/EIA{0582(97),March1997. [23] EnergyInformationAdministration.EnergyPolicyActtransportationratestudy:Finalreportoncoaltransportation,DOE/EIA{0597(2000),October2000. [24] EnergyInformationAdministration.EIAhistoricalstateelectricitydatabasesandFERCForm{423database,Website,AccessedMarch2004{March2007,http://www.eia.doe.gov/fuelelectric.html. [25] EnergyInformationAdministration.AnnualEnergyReview2005.Report#DOE{EIA{0384(2005).Website,AccessedFebruary2007,www.eia.doe.gov/emeu/aer/contents.html. [26] EnvironmentalProtectionAgency.Reducingpowerplantemissionsforcleanerair,healthierpeople,andastrongAmerica.OceofAirandRediation,www.epa.gov/cair/charts les/cair nal presentation.pdf,March2005. 212

PAGE 213

EnvironmentalProtectionAgency.Acidrainprogramprogressreport,EPA{430{R{06{015,CleanAirMarketsDivisionOceofAirandRadiation,October2006. [28] EnvironmentalProtectionAgency.InformationontheCleanAirInterstateRule.Website,AccessedSeptember2007,http://www.epa.gov/interstateairquality/. [29] FederalReserveBranchofSt.Louis.Consumerpriceindexdata.Website,AccessedDecember2004,http://stlouisfed.org. [30] D.Fullerton,S.McDermott,J.P.Caulkins.Sulfurdioxidecomplianceofaregulatedindustry,J.Environ.Econ.Manage.,34(1)(1997)32{53. [31] GDSAssociates,Inc.AnalysisofrenewablesportfoliostandardoptionsforHawaii,TheStateofHawaiiDepartmentofBusiness,EconomicDevelopment,andTourism,2001 [32] P.L.Joskow.Verticalintegrationandlong-termcontracts:Thecaseofcoal-burningelectricgeneratingplants,J.ofLaw,Econ.,andOrg.,1(1)(1985)33{80. [33] P.L.Joskow.Contractdurationandrelationship-specicinvestments:Empiricalevidencefromcoalmarkets,AmericanEconomicReview,77(1)(1987)168{185. [34] P.L.Joskow.Priceadjustmentinlong-termcontracts:Thecaseofcoal,J.ofLawandEconomics,31(1)(1988)47{83. [35] P.L.Joskow.Assetspecicityandthestructureofverticalrelationships:Empiricalevidence,J.ofLaw,Econ.,andOrg.,4(1)(1988)95{117. [36] P.L.Joskow.Theperformanceoflong-termcontracts:Furtherevidencefromcoalmarkets,RANDJ.ofEcon.,21(2)(1990)251{274. [37] J.P.Kalt,M.A.Zupan.Captureandideologyintheeconomictheoryofpolitics,AmericanEconomicReview74(3)(1984),279{300. [38] N.O.Keohane.Environmentalpolicyandthechoiceofabatementtechnique:Evidencefromcoal-redpowerplants.YaleSchoolofManagement,2002. [39] O.Langniss,R.Wiser.TherenewablesportfoliostandardinTexas:Anearlyassessment,EnergyPol.31(6)(2003),527{535. [40] LeagueofConservationVoters.DescriptionanddatafromtheLCV\NationalScorecard".Website,AccessedMarch2004,www.lcv.org. [41] J.McVeigh,D.Burtraw,J.Darmstadter,K.Palmer.Winner,loser,orinnocentcictim?Hasrenewableenergyperformedasexpected?,DiscussionPaper99{28,ResourcesfortheFuture,Washington,D.C.,1999. [42] F.C.Menz,S.Vachon.TheEectivenessofdierentpolicyregimesforpromotingwindpower:Experiencesfromthestates,EnergyPol.34(2006)1786{1796. 213

PAGE 214

J.P.Montero.Marketablepollutionpermitswithuncertaintyandtransactioncosts,ResourceEnergyEcon.,20(1997)27{50. [44] D.W.Montgomery.Marketsinlicensesandecientpollutioncontrolprograms,J.Econ.Theory,5(1972)395{418. [45] G.Morris.ThestatusofbiomasspowergenerationinCalifornia.SubcontractorReport.NationalRenewableEnergyLaboratory.GreenPowerInstitute,PacicInstituteforStudiesinDevelopment,Environment,andSecurity,2003. [46] J.P.Nelson.Greenvotingandideology:LCVscoresandroll-callvotingintheU.S.Senate,1988{1998,ReviewofEcon.andStat.84(3)(2003),518{529. [47] T.Petersik.Staterenewableenergyrequirementsandgoals:statusthrough2003.EnergyInformationAssociation,2004. [48] B.G.Rabe.Racetothetop:TheexpandingroleofU.S.staterenewableportfoliostandards,PewCenteronGlobalClimateChange,2006. [49] A.E.Smith,J.Platt,A.D.Ellerman.ThecostsofreducingutilitySO2emissions-Notaslowasyouthink,CharlesRiverAssociates,ElectricPowerResearchInstitute,andMassachusettsInstituteofTechnology. [50] D.C.Smith,R.C.Grace,R.Wiser.Massachusettsrenewableportfoliostandard:Costanalysisreport,LaCapraAssociatesandSustainableEnergyAdvantage,LLCundercontracttoMassachusettsDivisionofEnergyResources,2000. [51] P.M.Sotkiewicz.Theimpactofstate-levelpublicutilitycommissionregulationonthemarketforsulfurdioxideallowances,compliancecosts,andthedistributionofemissions.DoctoralDissertation,Dept.ofEconomics,UniversityofMinnesota,January2003postedathttp://bear.cba.u.edu/centers/purc/primary/documents/PS-Dissertation.pdf. [52] P.M.Sotkiewicz,L.Holt.Publicutilitycommissionregulationandcost-eectivenessofTitleIV:LessonsforCAIR,ElectricityJ.,October2005,68{80. [53] R.Stavins.Transactioncostsandtradeablepermits,J.Environ.Econ.Manage.,29(1995),133{148. [54] B.Swift.Howenvironmentallawwork:Ananalysisoftheutilitysector'sresponsetoregulationofnitrogenoxidesandsulfurdioxideundertheCleanAirAct,TulaneEnvironmentalLawJ.,14(1)(2001)309{426. [55] J.R.Swinton.PhaseIcompleted:Anempiricalassessmentofthe1990CAAA,EnvironmentalandResourceEcon.,27(2004)227{246 [56] J.R.Swinton.Thepotentialforcostsavingsfromthesulfurdioxideallowancemarket:EmpiricalevidencefromFlorida,LandEcon.,78:3(2002)390{404. 214

PAGE 215

J.J.Winebrake,A.E.Farrell,M.A.Bernstein.TheCleanAirAct'ssulfurdioxideemissionsmarket:Estimatingthecostsofregulatoryandlegislativeintervention,ResourceEnergyEcon.,17(1995)239{260. [58] R.Wiser,S.Olson.Utilitygreenpricingprograms:Astatisticalanalysisofprogrameectiveness,EnvironmentalEnergyTechnologiesDivision,LawrenceBerkeleyNationalLaboratory,2004. [59] R.Wiser,K.Porter,R.Grace.EvaluatingexperiencewithrenewablesportfoliostandardsintheUnitedStates,EnvironmentalEnergyTechnologiesDivision,LawrenceBerkeleyNationalLaboratory,2004. 215

PAGE 216

JoshuaDavidKneifelwasbornin1981inNorthPlatte,Nebraska.HegrewupinNorthPlatte,graduatingsalutatorianfromHersheyHighSchoolin1999.Joshuareceivedhisbachelor'sdegreesineconomicsandmathematicsin2003fromDoaneCollegeinCrete,Nebraska.HereceivedhisMasteroftheArtsineconomicsin2005fromtheUniversityofFlorida,wherehespecializedinindustrialorganization,publiceconomics,andeconometrics.FromFall2006throughSpring2008,Joshuainstructedfoursemestersofacourseinenvironmentaleconomics.HisclassworkandresearchallowedhimtoobtainhisPhDineconomicsfromtheUniversityofFlorida.UponcompletionofhisPhDprogram,hewilltakeaneconomistpositionattheNationalInstituteofStandardsandTechnology(NIST)inGaithersburg,Maryland.HisresponsibilitiesatNISTwillincluderesearchonthelife-cyclecostandenvironmentalimpactsofindividualproductsusedintheconstructionindustry. 216