<%BANNER%>

Evaluating Risks and Returns Associated with Altnernative Marketing Strategies for Processed Citrus

Permanent Link: http://ufdc.ufl.edu/UFE0021913/00001

Material Information

Title: Evaluating Risks and Returns Associated with Altnernative Marketing Strategies for Processed Citrus
Physical Description: 1 online resource (68 p.)
Language: english
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2008

Subjects

Subjects / Keywords: citrus, fcoj, hedging, marketing, risk
Food and Resource Economics -- Dissertations, Academic -- UF
Genre: Food and Resource Economics thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: The citrus industry is the largest agricultural sub-sector in Florida with 719,000 acres in use. Over 90 percent of the oranges produced in Florida are Valencia and early, midseason oranges grown primarily for processing. Volatility in the citrus industry caused by weather, disease, and other factors is a concern for citrus growers. Alternative marketing strategies are available to growers to manage this risk. This study provides an examination of alternative marketing strategies and forecasts potential outcomes. The results provide citrus growers with knowledge about these strategies and their forecasted returns and risks. This study examines the risks and returns associated with 4 scenarios for each of the two major types of citrus used in processing: 1) the pooling method, in which growers pool their crops and receive payment based on the average price received for the entire pooled crop; 2) the 1-week cash market, which assumes the producer sells his/her entire crop into the cash market in one week; 3) the 3-week cash market, which assumes the grower sells his/her entire crop into the cash market over the course of three weeks; and 4) hedging, which assumes that the producer hedges his/her crop utilizing the FCOJ contract in the futures market when the previous seasons crop is ending. Forecasted spot market and pooling prices are obtained using the forecasted all-orange prices from a recent study. Simetar? (Simulation for Excel to Analyze Risk), an Excel add-in that was developed explicitly for stochastic simulation modeling was used to create stochastic forecasts utilizing a pseudo-random number generator and observed probabilities. Using the stochastic forecasted prices along with stochastic forecasted yields in conjunction with a pro-forma financial statement model, the net present value (NPV) of discounted revenues through the year 2020 is determined. Pooling provided the highest standard deviation as well as the highest NPV for Valencia. The one-week cash market scenario yielded the second highest NPV. The hedging scenario provided the lowest NPV along with the lowest standard deviation. For early, midseason, the pooling baseline provided the highest NPV combined with the highest standard deviation, that is, the highest return with the highest risk.
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.
Thesis: Thesis (M.S.)--University of Florida, 2008.
Local: Adviser: Vansickle, John J.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2008
System ID: UFE0021913:00001

Permanent Link: http://ufdc.ufl.edu/UFE0021913/00001

Material Information

Title: Evaluating Risks and Returns Associated with Altnernative Marketing Strategies for Processed Citrus
Physical Description: 1 online resource (68 p.)
Language: english
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2008

Subjects

Subjects / Keywords: citrus, fcoj, hedging, marketing, risk
Food and Resource Economics -- Dissertations, Academic -- UF
Genre: Food and Resource Economics thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: The citrus industry is the largest agricultural sub-sector in Florida with 719,000 acres in use. Over 90 percent of the oranges produced in Florida are Valencia and early, midseason oranges grown primarily for processing. Volatility in the citrus industry caused by weather, disease, and other factors is a concern for citrus growers. Alternative marketing strategies are available to growers to manage this risk. This study provides an examination of alternative marketing strategies and forecasts potential outcomes. The results provide citrus growers with knowledge about these strategies and their forecasted returns and risks. This study examines the risks and returns associated with 4 scenarios for each of the two major types of citrus used in processing: 1) the pooling method, in which growers pool their crops and receive payment based on the average price received for the entire pooled crop; 2) the 1-week cash market, which assumes the producer sells his/her entire crop into the cash market in one week; 3) the 3-week cash market, which assumes the grower sells his/her entire crop into the cash market over the course of three weeks; and 4) hedging, which assumes that the producer hedges his/her crop utilizing the FCOJ contract in the futures market when the previous seasons crop is ending. Forecasted spot market and pooling prices are obtained using the forecasted all-orange prices from a recent study. Simetar? (Simulation for Excel to Analyze Risk), an Excel add-in that was developed explicitly for stochastic simulation modeling was used to create stochastic forecasts utilizing a pseudo-random number generator and observed probabilities. Using the stochastic forecasted prices along with stochastic forecasted yields in conjunction with a pro-forma financial statement model, the net present value (NPV) of discounted revenues through the year 2020 is determined. Pooling provided the highest standard deviation as well as the highest NPV for Valencia. The one-week cash market scenario yielded the second highest NPV. The hedging scenario provided the lowest NPV along with the lowest standard deviation. For early, midseason, the pooling baseline provided the highest NPV combined with the highest standard deviation, that is, the highest return with the highest risk.
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.
Thesis: Thesis (M.S.)--University of Florida, 2008.
Local: Adviser: Vansickle, John J.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2008
System ID: UFE0021913:00001


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 E20101204_AAAAAD INGEST_TIME 2010-12-04T05:43:14Z PACKAGE UFE0021913_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES
FILE SIZE 23997 DFID F20101204_AAACUE ORIGIN DEPOSITOR PATH shinbaum_e_Page_19.QC.jpg GLOBAL false PRESERVATION BIT MESSAGE_DIGEST ALGORITHM MD5
93d0c5df87e8b488911d0c3af72102b0
SHA-1
135ad3d7a34feb3fd5f7b9053fb8bf4fa38d7f40
26923 F20101204_AAACFM shinbaum_e_Page_12.QC.jpg
ab1442f7fcf1d652f0c0ed6924c5da0b
487982c7fd50ac734e41da07bb57943e7641bd83
48164 F20101204_AAACPH shinbaum_e_Page_15.pro
274efcce4bcd01380763e4643e7df7cd
b19b42ef7d935115dab1586d35855c2bee9404d6
84903 F20101204_AAACKK shinbaum_e_Page_27.jpg
c6104d3b915cb196ee8ec5f68811f7a7
15010bede48aa1d7f92001caf0720f6be09c3e93
6929 F20101204_AAACUF shinbaum_e_Page_19thm.jpg
e156b9d65402696d05e3c11f3b63db80
dc20a13ef03056e35b7c4b56c145a80f52c5b56b
25271604 F20101204_AAACFN shinbaum_e_Page_34.tif
7c03caaf1c1b269daea083d3b701c864
66bdaa676666cc917fe0d0c8e4649c031a7e4b60
53882 F20101204_AAACPI shinbaum_e_Page_17.pro
f2e89df81cd98060a6132bdc7a458f75
dfd893c8fce4b35e2a9640c279ffc9d3ad1ccc8c
84657 F20101204_AAACKL shinbaum_e_Page_29.jpg
54deb3dcc2a3321acd1d35d7b628e586
2cc0bfb21ea761dcc7f5a9a8416746b7853bee7c
22087 F20101204_AAACUG shinbaum_e_Page_20.QC.jpg
17dea1cf149078ac6fa02e3b3776da4f
55bed57ed0d7f1da334b230e9f339d0417141dfe
F20101204_AAACFO shinbaum_e_Page_07.tif
bc8492426ef4271b9e9330f92a48cf27
54999b002b2bee0bc240df0bf8ddabcbdfc7de50
51717 F20101204_AAACPJ shinbaum_e_Page_18.pro
444d5c355b088e6a4ad591afc198cc4c
bf523ea52debf5f52a679d77dd6f04d794b8bd83
84570 F20101204_AAACKM shinbaum_e_Page_32.jpg
14eebfe0fc16eef3d7ea3ebe9d27cf7b
18a986be1a6a9203cd53f1e0361fc2bf7c784e38
24763 F20101204_AAACUH shinbaum_e_Page_21.QC.jpg
9e309d805c4d63982d7ceb8e441f3733
cc8ffa461a557509f47de5ecbb9971544e969d96
1230 F20101204_AAACFP shinbaum_e_Page_26.txt
b76a74c0821743dd58cec785f07f296d
43d0342ce5a22eccae9ba5c52323c66d2db09661
49607 F20101204_AAACPK shinbaum_e_Page_19.pro
db882f38eee1687b5f28ca242e0b0d3b
40223913e52fa827227a5fc07448ad20f6e6f8eb
28214 F20101204_AAACKN shinbaum_e_Page_36.jpg
92e9e2c9f03847335de86f4d1ddc1b26
fd03451d151b2874c83be51af96651b2b2048d7e
19003 F20101204_AAACUI shinbaum_e_Page_24.QC.jpg
c9244cb75d232a700130d46300907453
b6e5954254ad0cf751fef618f933c50f8d99bb9d
1051986 F20101204_AAACFQ shinbaum_e_Page_67.jp2
21fa41623426c2ea6fe284d2b9ff3960
dfe90f48ab63a74779f9c3bc9c67f572f0bbaf1a
47048 F20101204_AAACPL shinbaum_e_Page_20.pro
36fe82207a690c0f0a5632f66b47768a
8a201f83297b1be3444b4fb4101d7f12100b4259
85951 F20101204_AAACKO shinbaum_e_Page_37.jpg
281e6b9cd22f30fb2a8fb736b392c601
19b5f8a11cc2160b14a16418a81cb0c44a79f732
14404 F20101204_AAACUJ shinbaum_e_Page_26.QC.jpg
79a95c6b9a517190d080e924e23d0a60
395bbc9283af65300a41fa070f515f46e2e2fa02
53219 F20101204_AAACFR shinbaum_e_Page_32.pro
76d8300f944e06426f12af767dec6953
3b8fb73bcdb6841b1e428dcf0a478bee9ed1d9bc
55030 F20101204_AAACPM shinbaum_e_Page_22.pro
902e4a4eb1aa773bd0a8c4eb89658816
395d2cc4d9c715298d605a92163c3f23f73295bd
76875 F20101204_AAACKP shinbaum_e_Page_38.jpg
17646ad92100a53392f1cf3addf29b25
30db336b294879f6a39408af18a69f4b7b69946e
4818 F20101204_AAACUK shinbaum_e_Page_26thm.jpg
73bf154dc30b3d36ea49483010b72036
16a73ee6a93cd7856cd6af4da4093791587c7ddd
1051922 F20101204_AAACFS shinbaum_e_Page_31.jp2
65c2691ad17708e734f5dc704afc8718
fb100b280f882cc94d32e4dcc377755a9b066b8a
33905 F20101204_AAACPN shinbaum_e_Page_24.pro
7d980e73c04b61c0dfb0fb73e6c9a05b
e87c5b22e1720021bf69507c002c7877abc78e3b
80742 F20101204_AAACKQ shinbaum_e_Page_39.jpg
2674e8f4c12e0d00f2df78fd293fb089
4e8d5b599227b74043160fdf54e848681d62e00a
7255 F20101204_AAACUL shinbaum_e_Page_27thm.jpg
c3dbbd1d0bb8bac13a985b833f622ffd
39723f40053a9f698c12b723ffaf9f6500814471
F20101204_AAACFT shinbaum_e_Page_12.tif
b0936d80344519e310a9e85045f6ce41
41edf2dd686b8c6f60964166fb1342bd39b4dec8
21236 F20101204_AAACPO shinbaum_e_Page_25.pro
28620fbfe8e1057413d1117d0ee739a9
012c5e42871a642a533a90aad398ed71cc6ee1b9
26671 F20101204_AAACUM shinbaum_e_Page_29.QC.jpg
1d6ca41be7b6a57afea87a2733ba624c
862682fd65d15156fc12afd296e0b6d11905c634
F20101204_AAACFU shinbaum_e_Page_40.tif
ee5a8715216f74a65b993705f29ee012
ff65523001d8d1f0a3bcaf340a7020987605f086
53471 F20101204_AAACPP shinbaum_e_Page_27.pro
4a9c23a1b9e47d594dfd78a962eda419
434f423d01471b91d7e24fbe47640e1b04df1a7f
81218 F20101204_AAACKR shinbaum_e_Page_42.jpg
76ed11291bdd095eb992a2909a495d52
b9d73b359d5052dceae0fbf0be795467acb1c7d3
7187 F20101204_AAACUN shinbaum_e_Page_29thm.jpg
b666a1b0f0c18ecf7391f7ab0ab7a3a4
7a82005ddddbba20545c0a5b68a65a9af5a3c1a9
43073 F20101204_AAACFV shinbaum_e_Page_35.jpg
fd17e48e82099a737cca0d6f435e7a02
0471a5a5d2e0125d17a6efaa4d17fdfa3e54ae0a
53720 F20101204_AAACPQ shinbaum_e_Page_29.pro
3c2487611d762b501f1ae85b47bbb1bb
05bc93bc91b8e5b0f539952c3cb8758a41e5eb8e
51635 F20101204_AAACKS shinbaum_e_Page_45.jpg
174091e94a400c9e1682dd163acc868d
2437b3778690901ff3dbefefb045940332a511a4
7251 F20101204_AAACUO shinbaum_e_Page_31thm.jpg
2fb02b83b29f881e5f6092adcd99b5d0
02931d9ec272822dbab1a028affe127e2af35cb3
11563 F20101204_AAACFW shinbaum_e_Page_05.QC.jpg
c95e4a85f7f33fdf623b472b5b457fe3
023379da30ca898a68acbf4482978e150d6f3421
54575 F20101204_AAACPR shinbaum_e_Page_31.pro
cb5d624f49cecae9f702cc8b7ef4b4e2
0dc8cf1f449b5d52ffc24deb8d81a934037197ed
50128 F20101204_AAACKT shinbaum_e_Page_46.jpg
a4dd413b806a806bfe5c5ab54b9a5a36
4fc4bda786671e26657c9fe2a1538956338d07a6
7285 F20101204_AAACUP shinbaum_e_Page_32thm.jpg
71310f96d102fe3457aa4cc0d1d8909e
7650f4e6f8283c8689a7a89aa5d12a20bafeeea9
1965 F20101204_AAACFX shinbaum_e_Page_51.txt
8bdad784ced3262c8387c27dc371a973
f2a6196b34c847b6ad00ddf46167066e88804e67
50409 F20101204_AAACPS shinbaum_e_Page_34.pro
8ea218d4c4db77b91f88eb50ac8262e0
8aec9ffd930c4f4762b0f80aace2e9a44706280e
64506 F20101204_AAACKU shinbaum_e_Page_47.jpg
80b63144c624874920193d13f187da05
fe5907ff055cab5e1b4b8b5f00e4d04d24852c72
7192 F20101204_AAACUQ shinbaum_e_Page_33thm.jpg
2e7bb3f9f542f6d130c4ed38275c21c7
6910245fba91951a9c3c5355b7a6b1646909329d
6664 F20101204_AAACFY shinbaum_e_Page_50thm.jpg
3b111ee43967e7d699150cfaad940ad6
a8b8d62a9475de00c7426d5bf5cedea175ab6139
15559 F20101204_AAACPT shinbaum_e_Page_35.pro
fe12dcce46363e51ef663a3aa1163209
db5852363c31f061f6cd9728c6bee6b71d8718e0
79113 F20101204_AAACKV shinbaum_e_Page_48.jpg
ea837579646cfbe8c32fd7bb3bb1b59c
bab84ef3d39468cd30b282f6a772f3384e08517e
6993 F20101204_AAACUR shinbaum_e_Page_34thm.jpg
1e31148c7f6ef515bb000e00511e3664
a25d52adab0426ad532650b09598d1a03fc38939
1944 F20101204_AAACFZ shinbaum_e_Page_48.txt
176b5c251cfe8d49163a206a8909a538
cc397c015c2ce289336887acafd4491ea3a9c657
5602 F20101204_AAACPU shinbaum_e_Page_36.pro
f0badc9578de87db42b58cb4f9a2585a
4ad223abb9329de82e9f1e61aae8fe5edd76b2d7
72675 F20101204_AAACKW shinbaum_e_Page_49.jpg
fa6dfe6d1c1a0036be59a9968f3bfd94
679c12cd76eee0c701f0311e5140c8ba4c93e95c
14466 F20101204_AAACUS shinbaum_e_Page_35.QC.jpg
293394f4dfb93f7061aae2b4d3cd26c9
b53029396a67abffb048e0e4d763691d8585fd24
53406 F20101204_AAACPV shinbaum_e_Page_37.pro
6af1d341f532e53721e6610520675187
88b6492ff02fed1f399d16c678308b3b5c42b326
77729 F20101204_AAACKX shinbaum_e_Page_50.jpg
caa8f5c528309b0c372635a62a115d2d
f17e7da9bd848ce92b5988af9b6442e3a19f0ccd
24676 F20101204_AAACUT shinbaum_e_Page_39.QC.jpg
fc01f867151992b49aa8b1a7619a6435
be483044f078c7711ce87b0fd01d5a2f754a3767
2207 F20101204_AAACIA shinbaum_e_Page_37.txt
3691799945a58e56c7ca0b85f6a785d5
8f9922feec3d3759f4d149a64467d23c6d28e39f
76693 F20101204_AAACKY shinbaum_e_Page_51.jpg
684ecd9f4f22493314a8ee32035c3df6
0312011f80db43c5a058f40617d59d2d599773df
6709 F20101204_AAACUU shinbaum_e_Page_39thm.jpg
47c9fd64984b362fbc7405f8d500c8a8
6882e4ca84ba34860c264e8d401f65abbdfa94ba
49136 F20101204_AAACPW shinbaum_e_Page_38.pro
1ceaf98ce3ef8f22a9b863b9d1020f6e
654a2a65974886d494b4c58bf2a021b69dc43661
4465 F20101204_AAACIB shinbaum_e_Page_35thm.jpg
20d72dc15c67b9bdad308fe20c55601a
8bef119280c9b36f5e636c2ffa44e4ae806bfb37
63190 F20101204_AAACKZ shinbaum_e_Page_53.jpg
8a6dad36b7692b93c063c8d33b64592f
f70bdf586ee8bb4939e6fd08e12c0fe80a632d3f
26814 F20101204_AAACUV shinbaum_e_Page_40.QC.jpg
5728d9165b6e016cc37c76f7df9a7c03
7032805bc37a21a730b442c46da065036eacc493
52683 F20101204_AAACPX shinbaum_e_Page_39.pro
330763a67cf2bfa033023770fe75d6e5
fb095362b9e86e2e36071b8b4c1664945124dc52
F20101204_AAACIC shinbaum_e_Page_29.jp2
b3112c48980ef4b8bc40c83ab0dc28da
20f64ec116b2b47f51ac8cb343913e0a3e996b16
7323 F20101204_AAACUW shinbaum_e_Page_40thm.jpg
966fa76e126a5b4f2fdf0bc1ce6fc889
add285b4ac6e03654946f05da86898c44eaea7e1
57883 F20101204_AAACPY shinbaum_e_Page_40.pro
cdb4f2a7b39c6c91a26b735dc694a717
208829a92b54f8cc38a37d27c720cf5d21be1a94
331824 F20101204_AAACID shinbaum_e_Page_36.jp2
97f0235d64652abc1c5b5098d8e36845
6e77cd5ca94a1b1080f9d8cd9cfc05751e11b02d
210337 F20101204_AAACNA shinbaum_e_Page_61.jp2
ecbab603104dbceecbd396b5df7bd2a7
cb89103e92180a504554a0679a16bbb797672adb
26807 F20101204_AAACUX shinbaum_e_Page_41.QC.jpg
9f8992188d9640ca5b656bdb79eba404
1d6e724ecec2b965c2e1583b37b4adb2b47f37b7
56726 F20101204_AAACPZ shinbaum_e_Page_41.pro
dd4f57bb33f7bf2ef7bc7c41707c752a
60ed88ab7f16fac1a9b1653a87279a5b87aa394f
1051953 F20101204_AAACIE shinbaum_e_Page_39.jp2
e1a9b81d55c40ef106e5532da81bdd1c
d4e575e7333de6cffcc4ac442de915348adc6694
1051941 F20101204_AAACNB shinbaum_e_Page_62.jp2
94fab5c080b8d5f1f8a36f5a35af92d1
88e7a111090e3e4bfe689319e65a1344defe139d
25541 F20101204_AAACUY shinbaum_e_Page_42.QC.jpg
5f15e416184834d8334c1e730ee1734a
a3a95ee614ddf86f84607bb1d1d5ae8ef9cb87a9
6988 F20101204_AAACIF shinbaum_e_Page_06.QC.jpg
1ba99b49c2e2b78c5daa07c19687afa9
40a6c9152faf8842c4e06e0f040b2ba3cce05a44
1051977 F20101204_AAACNC shinbaum_e_Page_64.jp2
474088e887ade88354045931b987c26d
11f01f0a90b0033e37c8b49ca5fca87416562dc1
27185 F20101204_AAACUZ shinbaum_e_Page_43.QC.jpg
6d662501f30ff79a1a95f09eff4ecbd0
301b17bde12e7be4c10f12c526aa1265b3a9e60a
27737 F20101204_AAACIG shinbaum_e_Page_52.QC.jpg
8de7ed33a2d8a1ece3ef0e6f0bd26c44
b0afc2bb2025a2b9fd24dc70538c7faee9af9aff
347 F20101204_AAACSA shinbaum_e_Page_56.txt
fdbb588c79e3fdd611f05388179d4930
11e0c564f92c6b5d47c903f0ab8b69491227dbac
671225 F20101204_AAACND shinbaum_e_Page_65.jp2
9e63efec01dea00875a9dceb780c16e1
3fdf0b6ea01f74ed70e4f0cd0de77327cd645432
26812 F20101204_AAACIH shinbaum_e_Page_28.QC.jpg
eda239e242a0e8cfcb329fb3bdb0d7a6
1509c5a1aec6848596be41deb66ca1d8e884af6c
294 F20101204_AAACSB shinbaum_e_Page_57.txt
3ebb6282550cb41a1ab26cecfbef53e9
097ffc888ab5b7043215e20ae3572b1225e29a8b
306109 F20101204_AAACNE shinbaum_e_Page_68.jp2
b912ffcca3b088882cb2c5b1729fc7ca
04ecc2264bb3acc21623da5e92ed9af70cf7395c
386134 F20101204_AAACII shinbaum_e_Page_26.jp2
bea5f12158607bc3e5890749a511411f
c238224dedc91ce83c5969b21529b93715871397
293 F20101204_AAACSC shinbaum_e_Page_59.txt
38aa000f4d777f0d65c89121dcd594f2
faeb3e5c6ac4b2b741615dd9e9b6fa1e17c3be91
F20101204_AAACNF shinbaum_e_Page_01.tif
9f5ae47bd3449b83f8faf18ff13eb309
78fea836793f70b0c5dacd3f23f7846c64577ad5
2363 F20101204_AAACIJ shinbaum_e_Page_01thm.jpg
05de064f913bab3458278141a4eae65a
4612b899bbb41f0f0e96ec6bf26da50bd2d04f69
301 F20101204_AAACSD shinbaum_e_Page_60.txt
02ce7fec63498db37e2f24f3996d55a9
f16a2121846cf596099c0fd18601817268998c76
F20101204_AAACNG shinbaum_e_Page_02.tif
15aecd50529047b31472b0c6bb71ce80
97ad25b378e94a9487a831009f8e7571cbefdb75
78564 F20101204_AAACIK shinbaum_e_Page_33.jpg
41a13a931e11c1cacf134771759617f2
4b69d9de54bd10493e6e42daaadb493b84ea4dc3
217 F20101204_AAACSE shinbaum_e_Page_61.txt
11197888fd93ad56402c3f355131b640
54ac9e4a18f696a6cd59d6cacac33cc9c33aa401
F20101204_AAACNH shinbaum_e_Page_03.tif
27f633dd1df06ba47dd3d36dbc45094b
ba9ee9d03f4f9034b654d70753370534ac0c7e75
1883 F20101204_AAACIL shinbaum_e_Page_50.txt
01bff89136d839512abcf897de6decd9
6685e22e67cc86cf02c53f0b717a7fe8998f3eb6
2147 F20101204_AAACSF shinbaum_e_Page_62.txt
06515493ca71314a739bf8411712f4a8
1791089a9e650bd68e844641228f70415d57cf7c
F20101204_AAACDN shinbaum_e_Page_17.tif
b203cebab0dc0a10a3ea79493f406db6
750dfcbc82521312fbcce23b8c56f473e3de5b88
F20101204_AAACNI shinbaum_e_Page_06.tif
8a5d2bdb02c3aed23cffd07890da8ac7
bf7eb34e1b1403cad0bf8471fde794f02110ce68
F20101204_AAACIM shinbaum_e_Page_51thm.jpg
a3324b327ccdefe838f7036d51ecd636
c948aaf3b4ec5fd3904573bd347092a40791ccbb
2143 F20101204_AAACSG shinbaum_e_Page_63.txt
5c1ca5afa77b4f58c1c7d905ca0cf30c
01e69eb26b51ca573fdb6ddb2232240d3cd4bfa0
1643 F20101204_AAACDO shinbaum_e_Page_47.txt
fb340c6ff4f83ae7de53041695cc3528
02f076a181e2fb61debcbce1cbbe063fbbf5b51e
F20101204_AAACNJ shinbaum_e_Page_09.tif
0f681057dfdba0b27a9334bfadfa4c45
25b860c6749bbee9fc3bc9e7024d9b118529ef17
F20101204_AAACIN shinbaum_e_Page_53.tif
1c637bb9880d333780e80b1b01b88a42
e7acdd5f684a5435801bac9f49d4ed42ceff97d6
2226 F20101204_AAACSH shinbaum_e_Page_64.txt
c669f9023faf2ca6e399758af41cc48a
583a52ab2beb1c8d504cfbe4555049f02cb8a59e
6861 F20101204_AAACDP shinbaum_e_Page_21thm.jpg
8c81a54904c7dc962a2be66488d20b0a
98d9bedc6826a8be80f414d56fa8d2db4aa4a00a
F20101204_AAACNK shinbaum_e_Page_10.tif
ac3f33a409aae165d38734f8770f6aec
0b48cc6231fcb1cb3ae13141204793d1f3f805b5
1507 F20101204_AAACIO shinbaum_e_Page_24.txt
8cf06bc7b45809f2a4a000f240781722
4ba9bb73f99a96c483c843923535a10a865b7555
1208 F20101204_AAACSI shinbaum_e_Page_65.txt
27c803ec8b57c529acb791ed3a8219bb
76429a7058763582e317a43ffd5d997b5e860df5
4469 F20101204_AAACDQ shinbaum_e_Page_03thm.jpg
b256e3cf4b5d1f53b7f40866f67be813
81a1013a453447c9d347dcf31f2bb5accb594e18
F20101204_AAACNL shinbaum_e_Page_13.tif
f73882f2b58d6f724652e20958c0d89b
84254164b9e2ad39ced2c1b8909f7756b848fa3a
2292 F20101204_AAACSJ shinbaum_e_Page_66.txt
eb795b03b850d0277e1cd9936ee02432
31ddebc340627170841013aa902916ca31107d09
F20101204_AAACDR shinbaum_e_Page_08.tif
5aa8e66c4cc323d70cd9a195452445dc
1cedf5e2731b80ebdc924450bd57c2e825ca26e4
F20101204_AAACNM shinbaum_e_Page_14.tif
a329de28f251da811a36f5be658668ea
f621f368bdf0b89764e0757b3b19a7550a992a1e
F20101204_AAACIP shinbaum_e_Page_38.tif
80fcc5a1a3f50c65617cbffcc3d3473c
e5d209892721e56ec02aa26eccc237dc5d69bb24
2110 F20101204_AAACSK shinbaum_e_Page_67.txt
b896d351d79f22c0ba5ea492505ee018
76334629a0d09f7f57fe1c2e10a40302d364aef9
2087 F20101204_AAACDS shinbaum_e_Page_21.txt
e833b953d67c78e62bbb86b89eecbb12
fa883b2a102143dcb3ca136ccba3f2a13e6e7dae
F20101204_AAACNN shinbaum_e_Page_15.tif
c6d952db26e44cbee300d08eb882de92
581d2964707b7b21a4f89595d143eee71e66ee6f
25971 F20101204_AAACIQ shinbaum_e_Page_23.QC.jpg
dc15361da24bba56dde0f85224e4e2e2
38eaebcfc857166db7dcc0e62e45970db57e3243
544 F20101204_AAACSL shinbaum_e_Page_68.txt
759c7d9b9454929ce1a9bdc6d8473c2e
ec1dcfd21af2d0071be819010eaac20f47e9cfd1
1051944 F20101204_AAACDT shinbaum_e_Page_40.jp2
e17c8fc5393898bcd4b7f4086b0b65b1
78abfcc46245f7f903b163d70b1706b5d269a40e
F20101204_AAACNO shinbaum_e_Page_20.tif
01396f6261e0d9dda2f464fc384f9817
b4fa966f61468af58bc19ee49b489989efcd163f
1051976 F20101204_AAACIR shinbaum_e_Page_32.jp2
c6f9dd3ada2ed1c658001ff91ed2eacb
0f0757ebe8395efd234f3af8a37811eb30db9d14
5164 F20101204_AAACSM shinbaum_e_Page_25thm.jpg
d6fa4eb877094e89737ac008ae06caab
e1170374b6b84d7c8ce8678199eecebcfa5b5ed8
83302 F20101204_AAACDU shinbaum_e_Page_67.jpg
565077d27af83534e228e53c38070a4c
e341afbc27d56b3fd9187ea0fd467a1052a3a820
F20101204_AAACNP shinbaum_e_Page_21.tif
a236f1af2d3383164f6f6386e049012a
a489898014a21e2dc3dfe4b41a7452b776259f88
3501 F20101204_AAACIS shinbaum_e_Page_56thm.jpg
295f3b5c641025f4f1b473e2d46d5ed6
78247e84377fb1611e8f5493c20d8eb4e49ea461
1425 F20101204_AAACSN shinbaum_e_Page_02thm.jpg
6f959669399e6cdb3d64e6787bc1ac69
e39367dcdb8e440f76dc150195ec155fdbf035b2
F20101204_AAACDV shinbaum_e_Page_24.tif
f2e2a7d5bf72eee95f9fd67d7a6aab83
279769710391a766b1e8d53c56ad44447338cc75
F20101204_AAACNQ shinbaum_e_Page_22.tif
4a732c47ae18b829fb27f1d8b5397a83
19f0a16fde5628f8217c3ad5a01d20a3a260b1df
26086 F20101204_AAACIT shinbaum_e_Page_58.jpg
3e355a8b84914f08567638cfd21418df
7375cbae887ce09be563277f931b81522ec7618d
5220 F20101204_AAACSO shinbaum_e_Page_24thm.jpg
62ae2f08fe66ee4fde9d5da75d18b18c
6d11ab3afd9a751805fb7db6b7fdad826a671839
3586 F20101204_AAACDW shinbaum_e_Page_36thm.jpg
f875cfc8dbb8eb757f08ce49fdcc3edc
884b11c2b6c10b6b8ebb177cb0248b1737aaa6c1
F20101204_AAACNR shinbaum_e_Page_23.tif
9c66d553faca63da1e93c74a12c408b5
f302e7d927a5b0d1615feab58a26352f71f4969a
42573 F20101204_AAACIU shinbaum_e_Page_25.jpg
f44b4e6e5444f8f31e637f06f8c99338
b560c1239d3373f21237ad64b23806cd6b49ad06
26396 F20101204_AAACSP shinbaum_e_Page_63.QC.jpg
7b01625ba8750896429f7c27e91b5ba6
249133ee7dc9f576ec3df23ea7591f449004699c
26377 F20101204_AAACDX shinbaum_e_Page_62.QC.jpg
42e3615e7bda3ee68fdf56538ecccb04
d9dde96f36aa5f18b957105b35618207730a46f4
F20101204_AAACNS shinbaum_e_Page_25.tif
23a6ed0fec994316b9e0859872789fca
1a3431aa669f37281316eb72a229126d2cfb3984
26204 F20101204_AAACIV shinbaum_e_Page_44.QC.jpg
0e0d1e913bd81b84499279e08afa9edf
e4c319762ace417d30401a9589027f3ce831776f
26996 F20101204_AAACSQ shinbaum_e_Page_64.QC.jpg
aba5c27218d0569ab04c27b7ab8c206d
4dde0357e32723ae7f7c83a621efcf7f623c58d1
81548 F20101204_AAACDY shinbaum_e_Page_18.jpg
975ab15f149f5ea33ad3b03b05029a3c
d7d87505859d046e9934c61ce44e69c979c4a3e8
F20101204_AAACNT shinbaum_e_Page_26.tif
2a673ab18dc3534a419c20d234bdbddc
c66046d4ccb2c9d56efc0703f76fa233847d83b1
85788 F20101204_AAACIW shinbaum_e_Page_04.pro
1c6b89507f3377822ed4d69fab0733a2
22c4d6f6d802b9bda284c7ea321c584020302c95
7207 F20101204_AAACSR shinbaum_e_Page_28thm.jpg
2a0095c308cbd2956f44e31467075403
2f373f1037163218a9cbbf9082103ed6f8b9aabc
24331 F20101204_AAACDZ shinbaum_e_Page_38.QC.jpg
0b70c804fc578ff8ef618b2677127678
d3382e9d21f18122dfc0fac4d330f4a54f2ac200
7534 F20101204_AAACSS shinbaum_e_Page_44thm.jpg
23dd61238e8e81b7939eb1d1f98a783e
acd2972b63e67e7dbcc1a75883d698149180f8ae
F20101204_AAACNU shinbaum_e_Page_27.tif
457e6c0621cd72a214ed778552c49374
a533857cc877a7ddcff3a524d4a7a06b80c282bc
F20101204_AAACIX shinbaum_e_Page_37.tif
040fffc02b67d14a508081d1b0eda97a
d7c7841d278cfc7700171b26f6ca82dcd07a9592
7326 F20101204_AAACST shinbaum_e_Page_30thm.jpg
6fac87c3fc5bad5e3851a917fe0d2225
0a4536f154726bc8362867ac32eee55ac4084c81
F20101204_AAACNV shinbaum_e_Page_30.tif
e703fd6003f1420942be5e5182b7eea3
6aab113d65a1e748694dacaed6c0c0a71448732f
73814 F20101204_AAACGA shinbaum_e_Page_08.jpg
0a748bd2b7d45d29ddd7f88a1c65549a
e6dc8d3a3fada9f09e0920b2c498d71c6d7e8827
F20101204_AAACIY shinbaum_e_Page_05.tif
d55bd3d854753243fb273bd4ed7c1c2f
ae2f0f20e8211937a0450b2461369371bf63e52e
25083 F20101204_AAACSU shinbaum_e_Page_34.QC.jpg
80412241bf0a6dedd7ee6da988dc4442
c845263a4a0f9b4cd5ff24461ba979baf017de9e
F20101204_AAACNW shinbaum_e_Page_31.tif
9476cb7c8df4392fbc151bb46149fe6e
4b21e4ab1fbd9fe901ac7c25e3fdb754d425c4ed
7149 F20101204_AAACGB shinbaum_e_Page_62thm.jpg
15858859ac1454cfdca5143038dd0b86
63a771a81de8eae3071496ebcbbd987638ef805f
50841 F20101204_AAACIZ shinbaum_e_Page_03.jpg
41f4b778abe1298de31de5357fea0334
998177eceee767f8b6b9ca1831fc4fc1b77034dc
26357 F20101204_AAACSV shinbaum_e_Page_30.QC.jpg
57d00c50dedc686053ca7bd4574b60d8
ede4dc94db4e13226996b36a52c3fb70150fd005
F20101204_AAACNX shinbaum_e_Page_32.tif
369f73b92e8434f98f4e39384d8aad06
e59dfce0d1eab391c2c8c64238e772f46132d37a
23895 F20101204_AAACGC shinbaum_e_Page_51.QC.jpg
6fbb12b46a866f007a5871cd85991558
7a2c8e10728442d3d6abfe663fd897b66f9480f1
21065 F20101204_AAACSW shinbaum_e_Page_04.QC.jpg
7cd9cbcc18b6ba339d3fd64fe43da64e
2fd80e82e81c025d04abe0d7d089dc9fd8032cb1
66456 F20101204_AAACLA shinbaum_e_Page_54.jpg
36b8d79bf0c0957ee324323a492c0a88
d5b53498a52cf913589edcf8fff13898dcf241fb
F20101204_AAACNY shinbaum_e_Page_33.tif
f95a47410590c33c447b560625a8a59b
3d202a7c60cba67a75bbd841d3703387f76e2ca5
85146 F20101204_AAACGD shinbaum_e_Page_31.jpg
a321da5b132845f56c71a5cc051c6169
570fdd0f81e99c0405cc2183431dd8ebf134de72
7172 F20101204_AAACSX shinbaum_e_Page_22thm.jpg
8ff07ac67b367b13a2ce4c4d46dd5cad
9ee371ca811935aa51f0b4d7af6b5e3cc1924389
32980 F20101204_AAACLB shinbaum_e_Page_55.jpg
ab95fd6245fe7195cad5dae5af4a442a
2779b08ae23c681d73ca3f4e742649d5953e68f6
F20101204_AAACNZ shinbaum_e_Page_35.tif
d2c34a7e1acaeedf7b329db7166e7ab8
c0d8f3ca5b2a737466e9328a5cdfd0710b9ec9d2
54731 F20101204_AAACGE shinbaum_e_Page_28.pro
2f2dc2644fa41752c14d7add503fe22e
e3e3e146cd891d5160a252e1d39e9d60b7d6ce4f
6062 F20101204_AAACSY shinbaum_e_Page_08thm.jpg
152b10950212e32a12aec3d9081a895a
b427432994d6c53d8c212a0e17d0e6d3832594cd
27501 F20101204_AAACLC shinbaum_e_Page_59.jpg
0bb7ed8a6caebd508764f3f7341289d7
56acc023e7f0f6df52228ea5474be81364fc02f7
7460 F20101204_AAACGF shinbaum_e_Page_41thm.jpg
ab1482272e0d3b7dd43f799bc260d883
d91fc282f65ab484d29e26d34137e15503b15cb3
27816 F20101204_AAACLD shinbaum_e_Page_60.jpg
f3778bd5d18cd1775fe749073bc0d77d
cc9ab3bf5ba49efbbe763c63f1d1404e94d9eb5f
1014317 F20101204_AAACGG shinbaum_e_Page_49.jp2
e7fd23348d63e23a150a57b69d680382
fd257830f8da9dfff860e84ff1c88e5b32324678
51193 F20101204_AAACQA shinbaum_e_Page_42.pro
3c20d11fdeb191b97bef71d288a496c2
2ac5ba3da83ff1957afa43ca93d41f4f6c500b5a
7289 F20101204_AAACSZ shinbaum_e_Page_11thm.jpg
52239fa99e2294d676203664a30daefe
9e75036c449f904c23239274aeebc348a2e5e805
18436 F20101204_AAACLE shinbaum_e_Page_61.jpg
8608cadbc0129e9edb61b4c267fafca3
a2a46bff06f99c391e02dd64a3f04471cade98c0
2097 F20101204_AAACGH shinbaum_e_Page_13.txt
113655593970daa037c7a86c9d40039b
e24a302a4ae2d835c222cdae1594ad8788804630
54956 F20101204_AAACQB shinbaum_e_Page_43.pro
30efe0bb8a0c189bd69d84608893564e
91fef6fdcb935f1df81a338c31c702bd14de77ce
82214 F20101204_AAACLF shinbaum_e_Page_62.jpg
4adfc92a825423b6d0ef332265c6477d
97030f5a6b7b71cd1539651a19d8b1f40ab00c0e
51027 F20101204_AAACGI shinbaum_e_Page_21.pro
69bdf4f94f50d048af4731d688dc87e0
362bf1fafb6212b3ac32419fae6854ab5e482335
53597 F20101204_AAACQC shinbaum_e_Page_44.pro
fc18f00104d25980ebdc0b0800b1d860
f6f35f45e473e8a6902126c9bb4b6300e15c08bc
7315 F20101204_AAACVA shinbaum_e_Page_43thm.jpg
400d14ae3c5b7c025f5bdd0972d3cecf
5fc6f6a26a17760a9504ddb017fc3cf7fcdfb0be
84385 F20101204_AAACLG shinbaum_e_Page_63.jpg
9adcc20bb0b1e2b4efcba0f450220a8e
0bdf19154cc997c061c88622f82a84a940c13271
16454 F20101204_AAACGJ shinbaum_e_Page_65.QC.jpg
a9b0470c75e7b2c1c77470ee568bb76d
707d6195ea41769f15b0d77badd220ab73a33376
30758 F20101204_AAACQD shinbaum_e_Page_45.pro
e91c494585146663a8649e343651ff28
18a19193ef389adcdda938ea9cd3df8028e5ad74
16652 F20101204_AAACVB shinbaum_e_Page_45.QC.jpg
54196ee4eb5eb553b4725195ffa14537
88870d82b05a2da4b000d47151dac37e41352c3f
86033 F20101204_AAACLH shinbaum_e_Page_64.jpg
83ea9c05f9d35151ddccc071a77ab643
8fdcaaf0badc5845ec0a117d7adac4ab88759dfc
369 F20101204_AAACGK shinbaum_e_Page_58.txt
12d2321ab5cef44428d3289238f9ad80
2f9168076cd4cf02a8f980143c1ea7307f18813a
38234 F20101204_AAACQE shinbaum_e_Page_47.pro
c163aad938d0cbccec4d7ca046d22e22
8575c211987565cf3ed3c706d9e6dd9727bcbb7c
4609 F20101204_AAACVC shinbaum_e_Page_45thm.jpg
b370ab93b0a5badc31ac4be830fd1011
980bddae1f4a4553882e786abbe5906def246c46
50437 F20101204_AAACLI shinbaum_e_Page_65.jpg
d4dd1cc3060ffe3d4252670da36a56d0
8e08f7cdd155a9ad7bcbe339bda6f1b6b778dbe8
54436 F20101204_AAACGL shinbaum_e_Page_30.pro
8e23b00f863911b6ecad86076070915d
dbe097870f3503744e8d2fdfe33b2ba52081487e
48959 F20101204_AAACQF shinbaum_e_Page_48.pro
00e533c506df97bf73dea8ae0b9fc414
a823184530d84161eccc0032ada70780c10b3a48
13545 F20101204_AAACVD shinbaum_e_Page_46.QC.jpg
c6622e98017bb83d167909e2a0583a7b
3090b8f9ee1b1d4cf00e9ac29647b3b94a0004e0
87685 F20101204_AAACLJ shinbaum_e_Page_66.jpg
168b8097c7056d78337acaec9f7530da
2af395dcf241f138777f9328c8313e40a4346d13
55639 F20101204_AAACGM shinbaum_e_Page_66.pro
9538a72d86b704c9a57ee586b5bf6022
f363de52a3a0c6ba7beb30b8dbf2432a444dcbdd
45448 F20101204_AAACQG shinbaum_e_Page_49.pro
af60e35588b9e1330d8d5cae1827512d
2f198f2bc70df02727a76640a47b951c011bac96
3773 F20101204_AAACVE shinbaum_e_Page_46thm.jpg
c6c549dbaaefdeb1e0674fa62d7a2fcd
0566ece8e11baeffca0be4fbdd353432fe5bf551
28023 F20101204_AAACLK shinbaum_e_Page_68.jpg
ad66200bb8780e3d7a21644403de1d10
3305c0894e387e7fce9ff058674ed75429145a06
54550 F20101204_AAACQH shinbaum_e_Page_52.pro
ad50f6f37458bce5ba3f9bfc143cf270
6e0575df0d44d4350bef6ce757824dec5d142523
5729 F20101204_AAACVF shinbaum_e_Page_47thm.jpg
9d8ee0b0adfbdd2a50e0ce37691cc89f
eefa642b727007aaf7c1c726ed783308ab38720b
261116 F20101204_AAACLL shinbaum_e_Page_01.jp2
803f00f4802321b3fb921c46b5a41a7d
51ee304cef72ecf72a3ab663f839a647b264445d
10154 F20101204_AAACGN shinbaum_e_Page_02.jpg
789e78c6c4f43c3d48e6aa29c40b5893
003ed2bfffe4e0321e9a600cf411fc775a37b64b
38256 F20101204_AAACQI shinbaum_e_Page_53.pro
7e3648e82b150b151aa00d38fb860408
0864466a36b92b34c1dc17edaed7e695cbe113be
25100 F20101204_AAACVG shinbaum_e_Page_48.QC.jpg
9e12fd303b3d1c35f54666829a4cd050
c1a63b175a82e1d848043aae44e69de9ab9adc31
660358 F20101204_AAACLM shinbaum_e_Page_03.jp2
6de9f11f434ebaab0634f260c19350f2
4c3007a4ea2c3df6086371280e3fa55d409802ad
1051972 F20101204_AAACGO shinbaum_e_Page_16.jp2
480b053387d72941bca6fdf611c991d5
31f9123fe252fde6f44667e7d4047399488c06d3
50727 F20101204_AAACQJ shinbaum_e_Page_54.pro
272e79fe108ff03e22a37ff6c5ac7bf5
c1bc08440bf67246a1f981d91a9248a1d13e3ea9
23256 F20101204_AAACVH shinbaum_e_Page_49.QC.jpg
289080635660f4fa3afe35abe3e47e26
39107fbc6cee955b5f3ff69275e246fd878da317
1051966 F20101204_AAACLN shinbaum_e_Page_04.jp2
7dabd8f71f4beb95c495989be11c730e
a437fa63360fee1422c11567566cf3de22396290
863510 F20101204_AAACGP shinbaum_e_Page_47.jp2
350677310385ace2422ae3366030ad47
178ca9e428005913dc5dae99200141f8d07d64fa
12133 F20101204_AAACQK shinbaum_e_Page_55.pro
c6ee75ecac30de422eedab695a836ff4
bdf6056d400f026ba78a74c0274a4957ff99bb5f
6442 F20101204_AAACVI shinbaum_e_Page_49thm.jpg
be5c4bb9525ee656e765be81bb22fe46
45665649d626ff2754082e6c722ac460a4460cd0
1051949 F20101204_AAACLO shinbaum_e_Page_05.jp2
e00037382a2a028297daa117fbc37d77
8719d5975a64ecf3a116bd2adf067b06ead2c88d
5740 F20101204_AAACGQ shinbaum_e_Page_54thm.jpg
a2eaf6e2b312572b2fbf88d2bc8b9cb3
9e751ecfc1339165506a7d06b9d437444241ddf9
6793 F20101204_AAACQL shinbaum_e_Page_57.pro
db12b3cb250d83d1f381ca0275a959eb
fdd00a393ff94da001d4c8f36df3afe31539852e
19597 F20101204_AAACVJ shinbaum_e_Page_53.QC.jpg
75ffac2c5152c901758cf0f3961cf7b4
e7447d51541387caaa7b4a994e90cb434474a9f1
421772 F20101204_AAACLP shinbaum_e_Page_06.jp2
096edd97cccfa770f86fd8ab39f078b8
4b2114c353647ec584f903aec7bceddc3ea6766f
F20101204_AAACGR shinbaum_e_Page_16.tif
1f9d0a21bf45b066fa9d6b228f0f44da
69c1b16a67b73e1b3936cad37b705b07ad4f0832
7400 F20101204_AAACQM shinbaum_e_Page_58.pro
37f02f34a49fce8d7ac703eaf7e69314
00d468eaa5af14f95b63cef00c58266e49547726
5735 F20101204_AAACVK shinbaum_e_Page_53thm.jpg
bdb95b897fcb28c10e297e2284792888
c9f2ae717d66411ea1826c475029f842efe6b6d7
1051974 F20101204_AAACLQ shinbaum_e_Page_07.jp2
c4536f31633bd49ac80dead7b6d49213
89426fcf44a03af4b237616f5e73cd0fe2b1ec83
6597 F20101204_AAACGS shinbaum_e_Page_20thm.jpg
704bf72fb80ac1456a5a363057b6bf1c
84fb24675163bda1e482e2997d5d49a6d662942b
6496 F20101204_AAACQN shinbaum_e_Page_59.pro
6cecbe7f72672130c9d6416ecaca7a90
0d3b375655770edf47e46a91b662bb09dc660395
10235 F20101204_AAACVL shinbaum_e_Page_55.QC.jpg
e66132e9aee04780b0879a538e18ce89
5f74144b6e55ed3feb03dc2b299e4c73697eccc1
1009054 F20101204_AAACLR shinbaum_e_Page_08.jp2
af87e9d99b085ade9347faa1d1eef227
ff511e5c8b4a198f54e73a87b554a7ad67f7c788
78937 F20101204_AAACGT shinbaum_e_Page_34.jpg
68d16fc94c4b94e7537b3f34b923437c
4c4e9800165a1b090a498945a7109f4c8bf1fbd0
6734 F20101204_AAACQO shinbaum_e_Page_60.pro
6d337dbaebd7a457fd98c26880bc34bc
96c1c808e8f67ab0999e5c78a8925d002db9d54d
3389 F20101204_AAACVM shinbaum_e_Page_55thm.jpg
a2c3e6c737227798bc7921c14160f710
fc7c01714d0747c253fe18b1d819b96d4731273f
86127 F20101204_AAACGU shinbaum_e_Page_41.jpg
dd68f41665903b29a7ddbb54b65df606
39c5d81f37474e37301c88140b77c1579da4eb45
51716 F20101204_AAACQP shinbaum_e_Page_62.pro
1cf97ea77bde63c2d5688a61e464aac6
f5706eef1e0338acf104c79410713d4a7e394434
10387 F20101204_AAACVN shinbaum_e_Page_56.QC.jpg
652f96ed8f07e6835740dbcdc72638dd
6bc9ad6b770e2bb33a8690cba851bc1c6ee2beaf
605464 F20101204_AAACLS shinbaum_e_Page_09.jp2
1096f4ad1f6bf1f3cfea617afcfb5e45
eb724331d9096ec1203bc14dcb50ca74f6ff65b7
77223 F20101204_AAACGV shinbaum_e_Page_19.jpg
aecdb440a58eabf959003973ee775292
9c8be859d947066f49e862a6cc7591c5ad3875ef
30123 F20101204_AAACQQ shinbaum_e_Page_65.pro
99260650c8aacfef8d49cc9538ea99c4
cb60d92f9a4eaea0d86b2b2494e9b6af10bdc9c3
8723 F20101204_AAACVO shinbaum_e_Page_57.QC.jpg
69cceb1c24c5c98c7808b8904d2e2eec
68fd1ae5732e6a3b59661eca43e67decf6ca7cdf
26497 F20101204_AAACGW shinbaum_e_Page_57.jpg
af78996678c517740fe785ac95cb2ff4
97be72e6aa5e277c8c4ea934a2721368b7cc648a
51227 F20101204_AAACQR shinbaum_e_Page_67.pro
e26d871f9e3cec8a8aefffd1310667bb
0d4a2e4f6849d7163bd82a06bd0d686f4aab5ffa
F20101204_AAACLT shinbaum_e_Page_10.jp2
2999b06ad27fc2a45f25e729f2ad7300
03fcf832647549fc42e29fbd03b26894fb944087
3329 F20101204_AAACVP shinbaum_e_Page_57thm.jpg
fdf19b905e113d9451fb9cd22f960a45
6018a8f5ca5b440aa9c1a964ab85942f93199eab
1990 F20101204_AAACGX shinbaum_e_Page_05.txt
02017e2905d6d771cf80e055de49a5f3
07ca5f61ea2cad9c6253a1b67e766373fff68ec5
12764 F20101204_AAACQS shinbaum_e_Page_68.pro
a843a2ecf8f7786299453ce8a6ba3572
21392fb1520790d071132c6f5750b788c8803fb9
1051925 F20101204_AAACLU shinbaum_e_Page_11.jp2
a5ccb5a868cca2b1f15110043b817218
9a38d6e2ef0dfd4b4fd7f7f0ca8d5116e8436aa8
9060 F20101204_AAACVQ shinbaum_e_Page_58.QC.jpg
f167183ab37375bd37dfab876faffb3b
b229460122696548e30532a784d87e402f238ca5
F20101204_AAACEA shinbaum_e_Page_19.tif
6fa2d38162434f0991bd57680b21fc06
7188a5022a34a8d2c56629b01b483505e1e2d1bb
7369 F20101204_AAACGY shinbaum_e_Page_12thm.jpg
bae14b46f40af820faca1b91d8219817
7334b7b4967e771157a11c3d6632544a59455e34
484 F20101204_AAACQT shinbaum_e_Page_01.txt
0e9fd7b8cf258fda34036b5bb78c058f
508656db5fe00fa00e72763cd8ee029ccc84f6dc
F20101204_AAACLV shinbaum_e_Page_12.jp2
4e33ec97a7c18142cf4bdc89fdae0adc
c86c7a3ad5b65bc5f68c6b0655b5308f2a3355bc
8744 F20101204_AAACVR shinbaum_e_Page_59.QC.jpg
fc63fde03f58399498d6169b37f322fe
802487f0c242abdb72a397f202c015dd64b0b8e1
7165 F20101204_AAACEB shinbaum_e_Page_18thm.jpg
a86ebeb4fc83ad2247cc1388cd4e3a28
e016d7887d9c9cdb0f2879b2178a4dbba6251bae
94 F20101204_AAACGZ shinbaum_e_Page_02.txt
5ce2d0322ead883a5da9489ee8eca710
6e97ae62eabe0b74e3ad453abe585ec76d57fffe
1197 F20101204_AAACQU shinbaum_e_Page_03.txt
1fc187e3d819910430093d7acc4cce61
7593eae0d72f10e761a208b48c7c1918db6f4488
1051982 F20101204_AAACLW shinbaum_e_Page_13.jp2
837824c8bf797bef5fb22a8785686598
6d345710edff0470c2d5a60cadcddd3d6b5336c3
3488 F20101204_AAACVS shinbaum_e_Page_59thm.jpg
c62b1f5ccf963f806d08acf9da55a753
d193df469bc4c53ddedbd0d83eb89df5d5593a07
2030 F20101204_AAACEC shinbaum_e_Page_42.txt
32ca3b3f460f1fa50e399f4179bd3dc9
741f2f7f634e22f743d490b026a0905218f53721
3529 F20101204_AAACQV shinbaum_e_Page_04.txt
511e47614ba9eec1df895a8221a09b60
075dbdd3e43537b5870096c38f4f3930bdf319fa
1051980 F20101204_AAACLX shinbaum_e_Page_15.jp2
1393e5c30278983e2a0d210beecec94f
3619b1d59217ce9d299fc807793b0145c50e8827
9490 F20101204_AAACVT shinbaum_e_Page_60.QC.jpg
5e355db313bf1cf15005fd5399e90336
54e14688d88f3a3a22e3ad86eb0dc7aae0347586
24007 F20101204_AAACED shinbaum_e_Page_26.pro
3c57a503913118ca00a0f882ff63e6a5
d28f19ea3afdf2d95cc637321c1861f0e5beb1cf
2557 F20101204_AAACQW shinbaum_e_Page_07.txt
23c7796cae5b6928a986d2df81a273b5
fd7b375d317c7acdf541ec5d8edfdad1c7b01f4c
4111 F20101204_AAACJA shinbaum_e_Page_61.pro
6143657aca6cab78c7920d5afe5117f1
f9fc9b073b31eb3b32af22b92da6671a2f63e56d
F20101204_AAACLY shinbaum_e_Page_17.jp2
fa60e478a2d81edf779e0afbfc03188a
eeb2e8517387ac7c0e02f0ec1891a9e7428e3630
3190 F20101204_AAACVU shinbaum_e_Page_60thm.jpg
bc4b36fac0c01c8fe76ea5e52d868382
f56665396575f731ab2d5f6e5501d47291ea280c
1811 F20101204_AAACEE shinbaum_e_Page_49.txt
65d8cb0b08e2641a72e764c5273477d2
e7a42b40f86ea493c03d1d609b9c05545f5054e7
43154 F20101204_AAACJB shinbaum_e_Page_46.pro
78af448314018434071baa6cd7c4c8a4
c7365e318491ccff836635c12f721315fda0b516
F20101204_AAACLZ shinbaum_e_Page_18.jp2
1f6427b5a34a2f404a3f1d9b6b687d83
5c1d15676e1e28e51d0dbacbeb17d31d62818c49
7385 F20101204_AAACVV shinbaum_e_Page_63thm.jpg
c689a5daca505adc80b3afe2e0429835
6face2618e9549f3fb8977a778b78b2b841d5361
1051910 F20101204_AAACEF shinbaum_e_Page_33.jp2
93e247e4bc25d5915b9edee13ed08b14
2fe21e1a521ea79b72a837beced7e839e7a37d96
1946 F20101204_AAACQX shinbaum_e_Page_08.txt
4852ad0e25732f167da3f1314e204038
a0eeb6e4871a7be631c9163c94c8a542f2863f6e
F20101204_AAACJC shinbaum_e_Page_44.tif
bb9ac7746cf4f0c5feaecfda62105329
9a3e0d21c16c43e2dad101dc625fd0eb89d6b95d
4810 F20101204_AAACVW shinbaum_e_Page_65thm.jpg
af8af0e9f8748b3be5d8601a67238b4c
c89c63f014905cb920419254eb3d4f3d721dd020
F20101204_AAACEG shinbaum_e_Page_47.tif
249812a7cf5c2884cc58c7f85801c72a
7441a6d4956d8456bc14dd81744f7ee22f52a5fd
F20101204_AAACOA shinbaum_e_Page_36.tif
1b6ac96f869d983505a4da8e09571c3b
16c801fbb6467126d3dbd37f8eef9a081925834c
1051 F20101204_AAACQY shinbaum_e_Page_09.txt
95a1807276c7289f347b6491777ade01
0f85590bfc80e4196c356c2e29120e3e42b477da
52775 F20101204_AAACJD shinbaum_e_Page_23.pro
497bb0bfacad0a077380b8b1e16ea04b
e8c74317308142ef2b2f38dee4342b3e3d014ffb
25193 F20101204_AAACVX shinbaum_e_Page_66.QC.jpg
7525522a6967933dc83bc3e22689aa7c
53cc5d4257a8fc259453d3ede7a098f69ee92b15
F20101204_AAACOB shinbaum_e_Page_39.tif
675e939d581e05e7ace1b1cd83c5b99b
14f9e4fac806f1196197c30e120228bf4efa835c
2149 F20101204_AAACQZ shinbaum_e_Page_10.txt
17981ffe88345a8c7673ce99abe91511
c7b006e1745109beb756649b4ba9b667de770419
373 F20101204_AAACJE shinbaum_e_Page_36.txt
30b10227b45b3ed857baa3ed9b785981
bd6402d14ab6f494f5b6810006f9324f2f210604
7072 F20101204_AAACEH shinbaum_e_Page_42thm.jpg
4e030f43313c7a40f1bf19d95ea75d29
f0c3761ba3902f0a3c4df1559001f9c5ad9495ec
8835 F20101204_AAACVY shinbaum_e_Page_68.QC.jpg
0250ddcc65f58dfe4e7ca0f6f1b57f33
c94b0bdac7b7f0547230c139326d270d012e5441
F20101204_AAACOC shinbaum_e_Page_42.tif
9f7de2779e78366daa3be144091f3bfa
3e247a2c326b76b2d11c0bf697f0cd05522e554a
72512 F20101204_AAACJF shinbaum_e_Page_04.jpg
befdc380c82b9c9a5b8d570738bb5bed
62761612a876379e43197aa0642260ccde31be03
2270 F20101204_AAACEI shinbaum_e_Page_40.txt
46181a5f5612bc6d9707b3a85b110450
f7b3c6b1c474ff3d43e66011d841d207260a68f6
2682 F20101204_AAACVZ shinbaum_e_Page_68thm.jpg
132888dec9f5e80fdfa3fe900458c503
843e6d6388a287d3cc0706e7ff0c161e44c248c0
26827 F20101204_AAACTA shinbaum_e_Page_31.QC.jpg
3ab4a7580c11b5ca434b991567f944fb
b0c8ac36032f224656b6babf91bf074160cbbd8a
F20101204_AAACOD shinbaum_e_Page_43.tif
7a6b70c888421fdb341727ccbfcf6bad
de94bf3fa3524fb487453610f580a37df8e2edd0
2154 F20101204_AAACJG shinbaum_e_Page_28.txt
235f5929234df69a8b6ea9006eb12162
d709aa933583ac00864fd2baae786f4d2249f44c
1984 F20101204_AAACEJ shinbaum_e_Page_33.txt
f595572b77439fb6f02753f254a91210
5202345b94df3eab565eb40d0978a764912410d3
5650 F20101204_AAACTB shinbaum_e_Page_61.QC.jpg
38b907d57e53b9547e91c254c2354f9d
ad48709e4969c0ee49581b8892a07df518c0b6ba
F20101204_AAACOE shinbaum_e_Page_45.tif
3da0af0fb63ab4dabf1106eafc30a1cb
c9920f6b4e40e5b93f88a3891248f5291bb6c845
6689 F20101204_AAACJH shinbaum_e_Page_48thm.jpg
3d848949e6f40c18a9ffe269785646b8
4927ebc7fd9e7a7d03b860f54999361ff125dd69
82601 F20101204_AAACEK shinbaum_e_Page_44.jpg
4f2cc439f8c3d5fc9228ace4d8205070
eb2931606369762779c97c8ab360503d94cbeb2f
7730 F20101204_AAACTC shinbaum_e_Page_01.QC.jpg
02997c373e6537e7a776ee5c984dec19
6b4e73e198e21d898b027b907ca087e1345fae38
F20101204_AAACOF shinbaum_e_Page_46.tif
6e1276c41be41cc7f423853da52b6a93
6b18f078e2c55f31a1aa2638bb5206fe5f81aabe
F20101204_AAACJI shinbaum_e_Page_18.tif
4c644fdb44933a9304f7f6b49f4e215a
f3915ae3946ba1d7ffc5ceaa8dbed0610e403918
4265 F20101204_AAACTD shinbaum_e_Page_07thm.jpg
3a5e862fdc50a192229bd348dd1baed3
2040ac1c73b5443d8e38c0e6dec4ce9b96659cde
F20101204_AAACOG shinbaum_e_Page_48.tif
3a3a73e923066514f17c8aba08b4af1f
d839550fcae16e3db118d12dcdcea98e655d2882
F20101204_AAACJJ shinbaum_e_Page_30.txt
a9b739e95b700b8fcde922465df4306b
521db2710851e84163d2050cf54a41de4e5f0607
28809 F20101204_AAACEL shinbaum_e_Page_02.jp2
8960e7d6507e5f97f0116cb157786c66
3662b635cf40dc9765ede0a281a8ecb0824d6a5c
26403 F20101204_AAACTE shinbaum_e_Page_27.QC.jpg
0734093a395d5c7a0e9cb1b7b2943711
c080e0ef24dbdf2fba725726ad4c02136ee6537a
F20101204_AAACOH shinbaum_e_Page_50.tif
6396a1082dc7dc5ba2a03351819c7f9d
e8cf4c1bf821c6d018547d5c2de854b3f48d5d61
26865 F20101204_AAACJK shinbaum_e_Page_32.QC.jpg
c16a5fdc8faf83ba0b350e5d63046eb8
c8280c50ed04a13962b3a6b9970e805333ca088f
6611 F20101204_AAACEM shinbaum_e_Page_38thm.jpg
fb3174e85f802904045e5af7062b6820
6264166d5274e4e7c0ef494b35246767e693cf3e
23687 F20101204_AAACTF shinbaum_e_Page_67.QC.jpg
cc5a414f48446b35fbf7ff56d3d62af0
da4b5b3a2b0218575efcc488e66d3db8176885c3
F20101204_AAACOI shinbaum_e_Page_51.tif
751d7673f496cd05ed97ee8a5b6baad8
6799fa3e17bc7f8eb1ded180b1dced6ea1891f22
9761 F20101204_AAACJL shinbaum_e_Page_36.QC.jpg
9006e3473836985f764470255716eb8b
4f092ee3cb3552c4f5fee62c7eb1222e21b7ff06
1051957 F20101204_AAACEN shinbaum_e_Page_42.jp2
8745adff79202fd20b92afe458f41d70
f30f9e1768690c45ef836b882755f51c7358f22e
7118 F20101204_AAACTG shinbaum_e_Page_37thm.jpg
1aacf567cd9a070226892aa098da3d4e
f4752c1ea172a62eb28774fe6e457085c64e7d09
F20101204_AAACOJ shinbaum_e_Page_54.tif
0ab5b980180cbc5bb8c37ada29ae84ae
82e500e360a91b8ce0c284ea0bc7e4b402710c15
395171 F20101204_AAACJM shinbaum_e_Page_58.jp2
8bf0fa477e6292b49fb359d3d6404273
ceb89f08443d8f449f80627e3aaedae57ee1c3b9
7358 F20101204_AAACEO shinbaum_e_Page_23thm.jpg
17eddf33ed49fb22d810cbbc91afb37c
f93a70d48e035f545a98de688e87fec709bf85f2
3794 F20101204_AAACTH shinbaum_e_Page_14thm.jpg
98a13b03c2abcb26edd0c3f438dac0de
46f72e40f59f1cb5e501bc1e2bcd3502a2b10c06
F20101204_AAACOK shinbaum_e_Page_55.tif
c78cf9aee1204981f759884bd27a7e53
8427f40a20cabff3b76f1cb20c616e5a8b191102
482252 F20101204_AAACJN shinbaum_e_Page_14.jp2
af38b870e230c8e94cd0333ea5041348
9bb70d7b6573296fb3f8fbfab60d207564e38d33
3203 F20101204_AAACEP shinbaum_e_Page_58thm.jpg
215abdd2bb9f836cc40ed3ae5cc9f125
97c44f011081b8c572aac13965f3cc2847650213
4125 F20101204_AAACTI shinbaum_e_Page_09thm.jpg
da99016a8cb93299a36384a20de44f68
c8dc7211a7d40171de38b9537b6a0a1298f1a20d
F20101204_AAACOL shinbaum_e_Page_57.tif
2808709a705e99725a555942ac95066a
e60f35f4dcb87f0f0322cbae5afe9755a461a7fd
F20101204_AAACJO shinbaum_e_Page_11.tif
e510a3acda8fc18c7411e4f8858c90f5
4ccdfd05d569c1f086c4c0c0467a65f2043f2437
2216 F20101204_AAACEQ shinbaum_e_Page_61thm.jpg
10fbc11552fbe6ec54a83cac9dc8cdb7
a4b82baa3760560c47f556357d313f20be95a576
6706 F20101204_AAACTJ shinbaum_e_Page_67thm.jpg
8349a698789240c72e0741eb747be9db
620d0e7590833005a298aca90acd20aadb0b11b6
F20101204_AAACOM shinbaum_e_Page_59.tif
3c588eeca43d209873fcddc35791223e
f867e37c011a14ec7ba2b275b7298e393d21579d
81098 F20101204_AAACJP UFE0021913_00001.mets FULL
558e98b2a41f0973e010246f9f73884b
01a73058697808001869aff983f0a37fc2172e3a
402552 F20101204_AAACER shinbaum_e_Page_57.jp2
295b6406239c8d12d1119fdd505e09e3
597f3f200c26cbea088f21029bffa9b9c27f5682
7498 F20101204_AAACTK shinbaum_e_Page_64thm.jpg
26907c7a103617db9c5e51c0998c832b
48fc952cba35b31032453631247f91ae7ef9d1ca
F20101204_AAACON shinbaum_e_Page_60.tif
3494c247e8a4e22dece28c41e49a4726
feab345569516c3f8cd2b378233acb0c7272b8d0
1051967 F20101204_AAACES shinbaum_e_Page_63.jp2
d6b850cf65545d3aeb630dc0a849e2a3
6ec40325029d174481ed614b51842352b06c4457
19498 F20101204_AAACTL shinbaum_e_Page_54.QC.jpg
f20f170608e2d87a7ea3044f0d535307
fd254e408fb8803369f28853d6d71c2c2f204e0e
F20101204_AAACOO shinbaum_e_Page_61.tif
1405a58b50711b65a98c28cf0476645b
8a1ad333f49e05714d52276fed7bad72f182c47a
3128 F20101204_AAACET shinbaum_e_Page_02.QC.jpg
dbd41e44a7d526c1d2d0fecfb9e36e14
6f1916e889281ce44a67b4f9aa4bcfdffebb4737
12579 F20101204_AAACTM shinbaum_e_Page_14.QC.jpg
bf28f6d6b27058ed2683c24f8decb2e4
3dc545ae12c5f2b3a7e5d568f4fd804cd7c1ee6a
F20101204_AAACOP shinbaum_e_Page_62.tif
518d821d2aad52959d3cdcb3267417af
984a0eeda12ecaa2d4cfd5a2d38e12128e3913a4
F20101204_AAACEU shinbaum_e_Page_48.jp2
6012f1f710103afb5811ba04ceb35b19
5c97ff10f79c16eb8b3d452f789d33bf7e2c85f7
7314 F20101204_AAACTN shinbaum_e_Page_66thm.jpg
ddad03a6f06ccd3ca15d3a9250be19fc
ac513ea6c88f25f3531b10d9eeca9ee756f236eb
26652 F20101204_AAACJS shinbaum_e_Page_01.jpg
21aeeb8b9d177815e97ab7089d780d7c
48fe4d30cdd070dbfd6c898d4b6da7be3cc2a2ec
56812 F20101204_AAACEV shinbaum_e_Page_64.pro
41ec0e7e15e6d95df2ff895b0bb2f964
f8dca33a35bf90fa7a46c4101285ade1d0ff4c3e
F20101204_AAACOQ shinbaum_e_Page_63.tif
839a848b67f91ce6a79359b704632739
b48399d5f5f364d13fe518fb4b0116717caa2776
104840 F20101204_AAACTO UFE0021913_00001.xml
39878f60728afedc169953d2c0073a8b
d5302a65cdf6530f7c85d5c0fd34f553d2351ae0
43959 F20101204_AAACJT shinbaum_e_Page_05.jpg
b1e36a611eef447c8055401c8ad73f5a
53e7dfdf6924ab2e823c290598ac1ca98576974c
7029 F20101204_AAACEW shinbaum_e_Page_13thm.jpg
3d20432a8f9c4931fb43f7dee8e9bd47
273f8af705c1bbbc2c72b1e41a93c5d947d4fd25
F20101204_AAACOR shinbaum_e_Page_65.tif
528e4bac1e11652bd616a77f509a52fb
aba83c0a927372493233914cad588424785e69b3
15895 F20101204_AAACTP shinbaum_e_Page_03.QC.jpg
f62cf036b99c534e678df64b4fa9bd34
1ed0e37df4b4efb6fc2863608a3d2a41282184e9
22390 F20101204_AAACJU shinbaum_e_Page_06.jpg
09f59cb92e6cf346697975aaef6319b2
957047705d1081ed4ae2712872319c208a0cc9d8
955 F20101204_AAACEX shinbaum_e_Page_06.txt
5e8cb14ac4eb396739c7d77e309df7cd
bd18773fb77aaf408af0626670814de8753193ad
F20101204_AAACOS shinbaum_e_Page_66.tif
282688ec7df2270b20b0cd00ffacb6f8
35d3ec25293da065e2bd98a27dbe997e586d378e
5320 F20101204_AAACTQ shinbaum_e_Page_04thm.jpg
35e7eb7507eaadc9d840591040a3aaea
71804f1f6fed9aacd2a842a346a7b0934c84f065
59383 F20101204_AAACJV shinbaum_e_Page_07.jpg
4c7f76490a9499fddec33cf6d71a790e
653956886fb44df6b6e5568efa8fa76e54986d01
53378 F20101204_AAACEY shinbaum_e_Page_63.pro
484a3876a50a649f2b49a3d7cb435929
17ee06a5131dcb9591c53826942a21e09396e68c
F20101204_AAACOT shinbaum_e_Page_67.tif
fca7f824df08464c7652bca65f6f6f7e
1373c4045d5071adb756dce7adc6694eb0922432
3161 F20101204_AAACTR shinbaum_e_Page_05thm.jpg
75b3056ea61f3d3edb29ba119daab488
cd0191b247865e7494ecbaa7dd61294853c9d271
46175 F20101204_AAACJW shinbaum_e_Page_09.jpg
c4ef3163e469b552ec24b49ec7df80aa
4acd8baba16b262d57278bcda50c71172d65bdad
85733 F20101204_AAACEZ shinbaum_e_Page_43.jpg
8b5ed07c95bf68cb32eb7131c73ee69d
446ebb81b6bf0f6f923c587a0fea7eaef8223f98
F20101204_AAACOU shinbaum_e_Page_68.tif
0e8ae886f190661cbbf0e149d2eea29c
53be807cbae65d6950af5661c8f4432f17043904
2245 F20101204_AAACTS shinbaum_e_Page_06thm.jpg
3c305f5f652931a0d81e2ad7d3e169a8
ff8531807b25245e7b376dc6bf96c05eb70fca35
82477 F20101204_AAACJX shinbaum_e_Page_10.jpg
991bf721c2e6b1437fe048d930f65958
99573d1acd2ea05fb899073a1195addd5534cd65
22430 F20101204_AAACTT shinbaum_e_Page_08.QC.jpg
686a67bb4e2f164d5b230ac50b3cd8c3
c51c2b177f02b8da5b313a95bacc946c7bad9eed
F20101204_AAACHA shinbaum_e_Page_29.tif
8a5f53bb3a45ded22512c4fa509c9060
b85ce14fa116319c0d0fb3be065a256a48b45bd4
85848 F20101204_AAACJY shinbaum_e_Page_11.jpg
d686035400bab55d5295a5fcf46d0294
20a16973ae7ee484b26471dee6ab83a8a1c1c40f
8563 F20101204_AAACOV shinbaum_e_Page_01.pro
c78e17f85f163b78a9dc795ae4739a7c
6373f1d6f7dcca8020738065163a2a4c0eb91024
14860 F20101204_AAACTU shinbaum_e_Page_09.QC.jpg
caf1bd733e01efd23ab8aa40895e9a62
ca4a0f1c12e945cec9b8655f24a09e9aa446cb8b
86704 F20101204_AAACHB shinbaum_e_Page_40.jpg
2304f047b2a639e20fc78647fc286ec4
0c67d2927a30842dc2a5ce997cc4453cd1dc5f4f
85199 F20101204_AAACJZ shinbaum_e_Page_12.jpg
4f8517445930c8b1d92b98d6de005563
9d34966845650bbbc37d45b92c713224de281617
939 F20101204_AAACOW shinbaum_e_Page_02.pro
df73e46a0db7694a23c4c9f183004ed3
04a9fde4894bfdcbc15aeb64e66481348f587711
26255 F20101204_AAACTV shinbaum_e_Page_10.QC.jpg
125687704ac8f7d4308a075bde8d3fb8
1b37f4741207a84c955e0d9b572f747479addae1
2156 F20101204_AAACHC shinbaum_e_Page_31.txt
f73d11e2935aa0d413cf531987aa6ad8
2f54dce6c70a7389e0c7d702d879e72f89616bc1
28760 F20101204_AAACOX shinbaum_e_Page_03.pro
888a20085220a8e23a78655ab39f3d12
7c6b24a9b8f59f4f13648bf0530accfd0c3d5946
7074 F20101204_AAACTW shinbaum_e_Page_10thm.jpg
4b2123409b50d78a2b9758963c5644f3
6a5fbc597cb96c7e3470cba386c979afec8bd545
F20101204_AAACMA shinbaum_e_Page_19.jp2
26cacb88e74bc55c9881bcc85af4e476
9fb73cb213d573b545908d2722ba6d700fe80a10
48313 F20101204_AAACOY shinbaum_e_Page_05.pro
700ee85e78b2afe405c9beea68e67b77
2996355b60197f0193527a5768af82316383ef92
1997 F20101204_AAACHD shinbaum_e_Page_20.txt
f58c839182b29086ad0c9aa4260b39c8
62b414ca275b899b2e9358b2a25c577a7a1a25d4
27376 F20101204_AAACTX shinbaum_e_Page_11.QC.jpg
05295fdf840f16d57024e2d5b9f1d7bf
0a7686566f213ee897195b0405ff6cdd339c0f97
1004951 F20101204_AAACMB shinbaum_e_Page_20.jp2
de1404d9d472ad0193333032da238c91
61bf8d90569355cdeb66d7ecab293df7217b481d
19283 F20101204_AAACOZ shinbaum_e_Page_06.pro
a060bb79f27f11dcae8d70f664aa75df
58b66266b99538b6fc6fa5712b2ac20c2d260c7d
7310 F20101204_AAACHE shinbaum_e_Page_17thm.jpg
508fde6f378eefcce615f74394f8871b
e74fccc7be845ae72c15e185feb362e801247cac
26135 F20101204_AAACTY shinbaum_e_Page_13.QC.jpg
0a24d8505c56d0ff4900afb63d765265
9b14962aea9e7e5d317f7f05376765fbd99ea063
F20101204_AAACMC shinbaum_e_Page_21.jp2
e302c12e593ab904f9d26024ed141d31
0449a1b9cdde38c931b78da8b0881914ca78c6f8
17502 F20101204_AAACHF shinbaum_e_Page_07.QC.jpg
f118d3ab7bf7f2beb25ea6d6797f570d
f88e9b37aa1b85f6b77b976aba2038616cb7462d
25050 F20101204_AAACTZ shinbaum_e_Page_15.QC.jpg
1b4544e9f1328078fa91f44ca9129a7d
2760986c17992b6154a57e6a97f29b3367966a13
2168 F20101204_AAACRA shinbaum_e_Page_11.txt
d50f32f5652a2a9d7091b876a032eb63
152404c9996d8daec4ffb625a4573436d0267062
F20101204_AAACMD shinbaum_e_Page_22.jp2
951ad09048b38ab17132406f7dafb631
5d07edc388477a760f9f9c57188f0d392711638d
F20101204_AAACHG shinbaum_e_Page_52thm.jpg
5ea26f03e1b79b43f0ff6f3b275050c0
ed1f8c7e4b021c24b3045808ba234fc6e351fa1f
F20101204_AAACRB shinbaum_e_Page_12.txt
0ff8e8145138f5e4ee9363f04124f421
971fb1ba6ece693ca5d0418beacebca5a9e8e7e0
922602 F20101204_AAACME shinbaum_e_Page_24.jp2
4c5c0ae4470e1a33c6ea35c7e431f867
3d57cbf326a5c7495ebc061b31367c686be60897
312113 F20101204_AAACHH shinbaum_e.pdf
31107a129b7b19740b6302af291ef0a4
3ea59608e0bcf7e9d9d4c56ef4e439de988defcc
896 F20101204_AAACRC shinbaum_e_Page_14.txt
9b09da22d5da49c557cdbfda741a68a3
b5635e82845f639350c9d8be5414242de898ef67
393040 F20101204_AAACMF shinbaum_e_Page_25.jp2
83b75cfe2bb4c40642e0a1e8ba443b90
d548b7b07c805b140740985fc3930dd54f826f92
F20101204_AAACHI shinbaum_e_Page_66.jp2
0c79cd86e22c38147b2ce3a5e5f1525d
9244a51296f328dab977e434c42b64b89504d0e2
1994 F20101204_AAACRD shinbaum_e_Page_15.txt
07d82eac5f821f0137655c875625c45c
04d03cb327545cf78459a2613b67e566f3ffa901
F20101204_AAACMG shinbaum_e_Page_27.jp2
b1590473eca39ccbbe7cf73330282d55
fb18f9107037ce3278c2a5755c57a1baf9165aaa
2161 F20101204_AAACHJ shinbaum_e_Page_16.txt
7b6b8421dc419e491aa2c94460cfdadf
a4ee6a179775020e1ac3db7907360c9d05150d76
2120 F20101204_AAACRE shinbaum_e_Page_17.txt
6e23175006b36a9cd9ef8b50dfc05345
fcccceba1aa4bcb937e1fc673ef27f4d181ce770
1051985 F20101204_AAACMH shinbaum_e_Page_28.jp2
6699dc48fa49316eb57a7865a48a4e00
c218d70f849c2464663ca9ddc92b5b9c5da2efc1
F20101204_AAACHK shinbaum_e_Page_58.tif
ec288b6b5ada18287f3fa9ac9899f395
03b9013310a106005bfaa2d36e505191a3c3439a
2081 F20101204_AAACRF shinbaum_e_Page_18.txt
af0657025797e4c059648dd8f1a16020
ff08bb02fd5ef8c34bb736aab7c0aee21967b50f
1051975 F20101204_AAACMI shinbaum_e_Page_30.jp2
9404bd50394b7171aeef11359cfdac36
beea7bed0c10d2238d29a70e77ae4e9066bc6fc6
26336 F20101204_AAACHL shinbaum_e_Page_17.QC.jpg
aaa94b9bf8f58f02f05101a0269680c5
61cc4e961a6c163f0838510cfba1d55ed7231d14
2004 F20101204_AAACRG shinbaum_e_Page_19.txt
5b2ed29180f373640502a56844c20b36
cb1dd3690d42d85ce9d4e53fd7348962213da6df
1051984 F20101204_AAACMJ shinbaum_e_Page_34.jp2
92df730bf8bb40051ed15c2e101cf03e
6524c3f57bda21de958a4562e9b8926572ce96c8
47271 F20101204_AAACHM shinbaum_e_Page_50.pro
d4f7e151c67e269cbc83f47b94e2834e
4982f995f16530d3f17b33c410ef4e19b7ed7268
2171 F20101204_AAACRH shinbaum_e_Page_22.txt
9beb6765861827ba3452d7c003805210
8e1b3a0b446adb1a5644145b3f2811459688de5c
F20101204_AAACMK shinbaum_e_Page_37.jp2
4f5f7d9811d97227e1bdf164a3c35372
4ed1c0d2e5c7797a13fdcb88bbdc3798637eb12e
20327 F20101204_AAACHN shinbaum_e_Page_47.QC.jpg
4f681d576afddc7d567c24109b5ae11e
5034f2aa4a752ec7336071d1e31acc842b57665f
2082 F20101204_AAACRI shinbaum_e_Page_23.txt
7dccdf6e474599adbedd54ececabb79e
b1bbf53548aa33b06a037e88d486c779ef81b122
1051931 F20101204_AAACML shinbaum_e_Page_38.jp2
400105d4e6c6f871f546ebc4ff3e47c2
629b9e5395480338bcec40c40ddd79471106c4a2
1117 F20101204_AAACRJ shinbaum_e_Page_25.txt
d9175a1bc0c25e0ee0eae6b1b037ca6a
0be27f92d18373694805e24a3a121f0e2de1de62
1051926 F20101204_AAACMM shinbaum_e_Page_41.jp2
0d57584548d15bb0922974d1765e7eba
8b915d67a3a873d423522c621f61289ea9eeaefc
F20101204_AAACHO shinbaum_e_Page_23.jp2
e6331dbe3008167ed0018a5771841006
84aa04d58e0d9996ad5f483b69d8b6ac3fb6c84e
2182 F20101204_AAACRK shinbaum_e_Page_27.txt
f27c8899b5490eb25c3f1577864d082d
0e01c4c91f2bf36c8d4015af6a078e592681540a
1051943 F20101204_AAACMN shinbaum_e_Page_43.jp2
205ed7e43305f6a0060c57614fcdc7dd
2fe044279d065003790fa11de76b29a59eb5234f
53123 F20101204_AAACHP shinbaum_e_Page_16.pro
dad632b4d9816f71087a52fa4a6cb46d
0fc4ca969c35c93543f6e56fc47dc1625b4f9fdc
2126 F20101204_AAACRL shinbaum_e_Page_29.txt
7eb85a774be89c436a5b0d23059f0a92
54727c594a7159765454672343458a6022648897
1051969 F20101204_AAACMO shinbaum_e_Page_44.jp2
5ace983ae11e00f41cb82f13820da257
03fbf81e660ae238d8436a3c412d84a666b534e7
F20101204_AAACHQ shinbaum_e_Page_28.tif
2935060f61d02eff3cce487af5081f63
93ca8a5ae3da3518c5523764b9aea215abe36c35
675785 F20101204_AAACMP shinbaum_e_Page_45.jp2
cb7b68027df9fb2b82424c91da54bdde
6bf0c74eb6c4c5466ddb93435ccc5ce11fca7e02
27208 F20101204_AAACHR shinbaum_e_Page_22.QC.jpg
b9bb5644eafee2cdabf0a613cd6237e5
daa39b987592af91930c6f9c5ec416dd8d507ccc
2100 F20101204_AAACRM shinbaum_e_Page_32.txt
661ba0693530e466010e67964f386c78
3581d8f9f6637abd505bb5e11e8b4e266c7e3835
1051971 F20101204_AAACMQ shinbaum_e_Page_46.jp2
5b824ea0a6d09a2758ff8dcc26ab444a
d08001ec7815b35888604304ab96688b11d8ff70
F20101204_AAACHS shinbaum_e_Page_49.tif
413888d94666e8601a0e74225b64fb24
f0488367929045f960d1aea1ce44201c2d9a82ae
1993 F20101204_AAACRN shinbaum_e_Page_34.txt
9cba5107d3000c9a6e908d60a8bbca2f
1149ab70093bb849a368393be58345024bc086df
F20101204_AAACMR shinbaum_e_Page_50.jp2
899beec98132feff13e1a5773bf7e8d6
b08dd9cddc6917897199961223f8804de2f96f68
30944 F20101204_AAACHT shinbaum_e_Page_56.jpg
bc4ec995d881e8634e73efed55f557ac
10a524a337bdc3140b8e6c4c4414ab7ff4373ffe
720 F20101204_AAACRO shinbaum_e_Page_35.txt
80ec0b43f54fa635e45ad2d51505a6ce
00c438a104a16ba69b9988086e62a7359c177c96
1051965 F20101204_AAACMS shinbaum_e_Page_51.jp2
696f48a5d0e76a23d7e01dcbe097328c
c5b024abba0509e2f03321fb7b0db3676ec2ad9a
F20101204_AAACHU shinbaum_e_Page_04.tif
c38dddcdbdfe91c1f2af2a324c18af1a
b31691b33c3da1a3a9e0faacc48a713e7dbedfe7
1962 F20101204_AAACRP shinbaum_e_Page_38.txt
359b4f7c4e48ef698b051603116ffc8e
c1a5d15bd3c8b5c0f177195f67d9ca14a15b28da
83921 F20101204_AAACHV shinbaum_e_Page_16.jpg
36371c0b7e8be17efb736af63ef55091
4a8873e6c5c7c2708ce5ea514b56da54f00b13b4
2119 F20101204_AAACRQ shinbaum_e_Page_39.txt
a292f3093dfe9b31732baaa5ee16d8ad
3bbb5636ca9f6edb01af0b5cc76253901d039ff6
1051934 F20101204_AAACMT shinbaum_e_Page_52.jp2
4bcc3071d6c22a3e520ca15c83dd6b41
3b3fe2c38513f09d0b605881094e94a07af2d402
51832 F20101204_AAACHW shinbaum_e_Page_10.pro
18edebcde442da3d677611ed931ea136
8881e265eafa4d98ecfbd51e9bf768e173d165ea
2228 F20101204_AAACRR shinbaum_e_Page_41.txt
91cfb90eca5c2642813cf29410360f84
49ba70e33225461913abdf941ac97967d294b9d2
853359 F20101204_AAACMU shinbaum_e_Page_53.jp2
3ff203f58709de5fddeebb9ec9e1e81d
376e145ea565c6e45ad3369e2880e70253521f9f
F20101204_AAACHX shinbaum_e_Page_41.tif
a1aa17ef20a36dffe8829a7ed12fbcfc
8508dd872ea6152c478fdf4a0de30763a472b53c
F20101204_AAACRS shinbaum_e_Page_43.txt
491a95e3ede71e430c4afedfb17ec177
a0172cb1342ca69ebc97d6be473e3eae0eed7dff
847188 F20101204_AAACMV shinbaum_e_Page_54.jp2
f1927c5e442d93824066af79ba9b7252
c79cf8fd64b97ed8ca217299cd4315101a5bef5b
85152 F20101204_AAACFA shinbaum_e_Page_28.jpg
8a8333904f5c21c81f188ba21764a794
20cc2fd370b50f67ae1e2af3291dfc7de60ce6ff
48492 F20101204_AAACHY shinbaum_e_Page_51.pro
4e5bbe6469be6f6af7b36b920a2bacf2
519d855eba4e9e46957d56b70f369dd233c94d15
2114 F20101204_AAACRT shinbaum_e_Page_44.txt
cc81e09579383d1a94dcd38ce2419bba
4997610743c9278b17c6f27e6eb49796fecd30a7
547368 F20101204_AAACMW shinbaum_e_Page_55.jp2
964878ce73729677efac0ccc72331d06
9d395e24f701b8abfea5ed7d64147c79d70d8268
85641 F20101204_AAACFB shinbaum_e_Page_52.jpg
265bfb458848e4fd76483620abb76292
fa48510476a00f4e140e0810429659cd26153487
F20101204_AAACHZ shinbaum_e_Page_56.tif
9851f3b18cc8e158ed910373bf152f01
6caab8dc0848a975d9ada9d0f8bb9764913f30b2
1224 F20101204_AAACRU shinbaum_e_Page_45.txt
9ceac6f5e80d6c4177ea3438be57fb1d
16cefe216abb578856edfe08716d34cf180f118d
579950 F20101204_AAACMX shinbaum_e_Page_56.jp2
a119008f0957aaa343420c8d30f213ce
cab005dd5200dee48971b1c34490c9edc204fce9
25360 F20101204_AAACFC shinbaum_e_Page_33.QC.jpg
aa37b9402284c106b14d356c92a628cc
bcc82242abda68cbd2761da239a79bb3d6620d30
2427 F20101204_AAACRV shinbaum_e_Page_46.txt
99c8c5d5663c9a3a130e2ef8a3b761d2
0c7cebc3c30177df5592a8c5a2fbba123556c93e
421025 F20101204_AAACMY shinbaum_e_Page_59.jp2
fa8cc42971904789b22248f497f1dd32
acc607142243f756c32df224f2e7eb19d557b126
F20101204_AAACFD shinbaum_e_Page_52.tif
a792ee59e7646e50db281a45f1b0ebb9
ea5711de5b1fba1b87c3d04fdf1ab534d957b6b8
F20101204_AAACRW shinbaum_e_Page_52.txt
0eff3d16ea4e35ed20efe8ab1b46f965
f1c728155168c8a39c8e1fa4a603905a58f7001f
81407 F20101204_AAACKA shinbaum_e_Page_13.jpg
11570f86ae540720e084d6ed19a2a875
6bddec0829570cbb6053b04a2615504024f8d2e3
414893 F20101204_AAACMZ shinbaum_e_Page_60.jp2
3f6b9429450388b9c85f41ef7c5025df
6048090e2c1a5d42ef25ff0d4a7b004b16403b70
15601 F20101204_AAACFE shinbaum_e_Page_25.QC.jpg
5f192ed6c0aaa38a9ffa091a98284af9
a3b7a6c45e22cbb44b3303aa03100ba5bd9bf1c1
1537 F20101204_AAACRX shinbaum_e_Page_53.txt
f033ebd074ff1eaaae27736a97c0a44c
01b3e65c2263e1de918ebea5294679ee884d6e80
38713 F20101204_AAACKB shinbaum_e_Page_14.jpg
8163d60deec50aa1f82bea5b7ee2ca12
f7b37d0eb98545187ec8d3864de3b5fa85070305
608173 F20101204_AAACFF shinbaum_e_Page_35.jp2
f082184ec9e53d50df805951afea3bf9
70cb645027e1a4be030638deab57d88d3e8a4081
78470 F20101204_AAACKC shinbaum_e_Page_15.jpg
fdabeb520149e7f00180c8e969cb97b7
cfb489c4e9d3e478abf9d1d28da08b62db5fa77b
26587 F20101204_AAACFG shinbaum_e_Page_37.QC.jpg
e2dbdc5b710fc6db6a0718107ffb7650
643074a0dda8f37f753857f952f148ffb307875d
62207 F20101204_AAACPA shinbaum_e_Page_07.pro
39e6e26a317d8904e1a24a8564435773
81981172e93644230acde73c06d3e5f81bc3c3b0
2516 F20101204_AAACRY shinbaum_e_Page_54.txt
a1e2aad0029148f79af36d6dda760305
a9a1c925a6a59bed475798d30743303e2ba48e9b
84740 F20101204_AAACKD shinbaum_e_Page_17.jpg
3a942ac198a81f7beff9909c5deb4cbe
7e4ceb9e94ce1db2605f19f0fb8fbe02abe19c12
84340 F20101204_AAACFH shinbaum_e_Page_30.jpg
37c78f0fc8544f1369b45f6bd6cb68d1
9f514cf49b100dfc3cf9a42d54cf022d0e00625c
44177 F20101204_AAACPB shinbaum_e_Page_08.pro
1dfcf8725026a71e66107802e1cdc0e6
abcf1c8a02f75a444291c51f5fc3e2e178e2af0d
579 F20101204_AAACRZ shinbaum_e_Page_55.txt
527dd70dc2f790d1aa890ac3ef750fc0
83fd9def39eeb863849f724f0433f711b5f5eca2
72429 F20101204_AAACKE shinbaum_e_Page_20.jpg
02b6c86356fb790039ea3af69edb9586
74271e11b613b9c58f093ed6757e91fc0b5a1dd1
50094 F20101204_AAACFI shinbaum_e_Page_33.pro
260229bd14af256a092bf2441ec350c3
cdae34817e2ca3e164fecaa7c3ea6f5f849dd6a5
26167 F20101204_AAACPC shinbaum_e_Page_09.pro
dd54f3ac39d21175df06a5d7428aa1c2
a36f9ad3e76cdd28e7a53baa366299cd4a7fbd90
79950 F20101204_AAACKF shinbaum_e_Page_21.jpg
64de30356f506baee95eb6256c1ef757
7996b0987958e894a6edd62a79853ecc3449d28a
6527 F20101204_AAACUA shinbaum_e_Page_15thm.jpg
8cc078e5abed2259a497973cf8549410
8097ba4a83fd98f8ffc6ae97b6c404c7deb3bbd6
7270 F20101204_AAACFJ shinbaum_e_Page_56.pro
0d2475bf702232b05513d5abb584915b
145c5b3f48f46fc4c0f8474be231b4de3f33a2d1
55078 F20101204_AAACPD shinbaum_e_Page_11.pro
054df037fe2627f5ac88b04235bcafcf
b3eadf164af311e41f23f00659bea26b4c017962
85359 F20101204_AAACKG shinbaum_e_Page_22.jpg
cc26b734e0c771f18119ec3f223679ef
019b6b5e80b5a484931247ae522f9fbd178a419e
26530 F20101204_AAACUB shinbaum_e_Page_16.QC.jpg
aa24179043ce4d9d974f3bdaeb537562
3491c7c9ba91367a288a76c61b2a5de3cc2554ad
24355 F20101204_AAACFK shinbaum_e_Page_50.QC.jpg
d2e04e669dff3b42bb9978fc0283868b
5ca9f450311486e530eaa4ddca6fb6fbdbe481d4
54749 F20101204_AAACPE shinbaum_e_Page_12.pro
ec462c18cfbddf7bd154e7ba52efa826
a4d144998b0b6daf3edc01b5554412fff30b55e3
81857 F20101204_AAACKH shinbaum_e_Page_23.jpg
756a0a72a574d332ac14c759f111a719
c6c07bb0fe4c9d90f4b5e340314765656fe4d565
7259 F20101204_AAACUC shinbaum_e_Page_16thm.jpg
2d8fc246f28868b91809669ec17d209a
722714fa645f064328ca92025aa1455fd998a5b0
F20101204_AAACFL shinbaum_e_Page_64.tif
14b47bf0e69709a6cb1db54179b36c82
1b65454a09360b5f8aaa313e1fe768300d02b401
52129 F20101204_AAACPF shinbaum_e_Page_13.pro
6786b608dea8751603bf1a71add3251a
3aab3aa68a9bc56545f8bffafadbf5fe0b1b97f3
61516 F20101204_AAACKI shinbaum_e_Page_24.jpg
d37dd6b0794fa6e71de69566ebe7b96b
d24f34486a2fe1cb63273c66818638b5b437644e
26026 F20101204_AAACUD shinbaum_e_Page_18.QC.jpg
82d7f9fede4cd466334ad17e2e0d149d
00d6668f768219ff04e25f90e8a3dc0341e35a16
21353 F20101204_AAACPG shinbaum_e_Page_14.pro
8b7c1c354752444dd8c3ad480c7ea898
27b98553c88ee76ac501d280d0767d8f404e382c
40977 F20101204_AAACKJ shinbaum_e_Page_26.jpg
13bc463ecc01e06361557aafc0ce8517
013e096a5914c7a535717f35f0fb2c3837660d80







EVALUATING RISKS AND RETURNS ASSOCIATED WITH ALTERNATIVE
MARKETING STRATEGIES FOR PROCESSED CITRUS





















By

EVAN MARC SHINBAUM


A THESIS PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF SCIENCE

UNIVERSITY OF FLORIDA

2008


































2008 Evan Marc Shinbaum









ACKNOWLEDGMENTS

I express my sincere thanks and appreciation to Dr. John VanSickle, the chairman of my

committee. He has provided me with a great opportunity in graduate school that I would

otherwise be unable to financially afford, as well as much of his time and effort during my

preparation of this thesis. I would also like to thank Dr. Richard Weldon for his efforts and

inclusion within my committee.

I would also like to show my appreciation for the instructors in the department that have

helped build the foundation of knowledge needed for this journey, including Richard Kilmer,

Ronald Ward, Lisa House, John VanSickle, Richard Weldon, James Stems, and Evan

Drummond. A recognition is needed also for Ron Muraro, for his assistance to me in acquiring

necessary data for my analysis. A particular thanks is needed for Jennifer Clark who donated

much time and effort in assisting me throughout the thesis process, and extraordinary help with

my data analysis.

I wish to also thank my parents for their support in helping me get through tough times

and reach this incredible goal.









TABLE OF CONTENTS

page

A CK N O W LED G M EN TS ................................................................. ........... ............. 3

LIST OF TABLES ......... ........... .............................................. 6

LIST O F FIG U RE S ................................................................. 7

ABSTRAC T .........................................................................................

CHAPTERS

1 PROBLEM STATEMENT AND OBJECTIVE ...........................................10

P rob lem Statem en t ............................................ ........................................10
O b je ctiv e ...... ...................... .................................................................1 3
O organization of Study ............................................................................................. .................. 14

2 BACKGROUND AND PROBLEM SETTING ...................................... 15

C itru s In d u story .................................................................................................................. 1 5
H isto ry ................................................................1 5
Citrus Industry in Florida ................................... .. .. ..... .. ............15
F C O J In d u story ......................................................................................................................... 1 6
F utu res M ark ets ................................................................16
Options ................................................ 18
Risk M management ................................................................ ..... ..... ......... 19
B asis......................................................... ......... 20
B asis in th e C itru s In du stry ............................................................................................... 2 1
H edging Exam ples.................................................... 21

3 REVIEW OF THE LITERATURE ............................................................. ............. 27

4 METHODOLOGY AND DATA................ .................. .......................................37

T h e M o d e l ...........................................................................................3 7
Forecasting M market Prices ....................................................39
Forecasting Futures Prices for the Hedge Model ................. ................. ..... 43
F orecasted Y field s ................................................................44

5 RESULTS AND DISCUSSION .................................................................. 47

R e su lts .................................................................................................................4 7
V a le n c ia ...........................................................................................................................4 7
Baseline scenario................................... ..................... 47
One-week cash market scenario .......................................................................48



4









Three-w eek cash m market scenario ........................................ ......................... 48
Hedging scenario...................................... ........... 49
E a rly -m id .................................................................................................................... 4 9
B aseline scenario ............. .............................................................. .......... .. .. 49
O ne-w eek cash m market scenario...................................................... ...... ......... 50
Three-w eek cash m market scenario ........................................ ......................... 50
Hedging scenario...................................... .............
D iscu ssio n ................... ...................5...................1..........
Valencia ............... ........................................51
E arly-m id ......... ...... ... ........................................................... ...... 52
C changes in basis .................................................................. 53

6 SUMMARY AND CONCLUSIONS .................. ......... .............. 62

Sum m ary ............. .... .... .. ...................................... ............... 62
C onclu sions..... ..........................................................63
Im p lic a tio n s ................................................................................................................ 6 4
Further Research Needs ....................................................................... ......... .................. 65

LIST OF REFEREN CES ............... ........ ...................................................... ....... 66

BIOGRAPHICAL SKETCH ............... ......... ................... 68









LIST OF TABLES

Table page

5-1 Scenario R results (V alencia)........................................................................... .................... 54

5-2 Scenario R results (E arly-m id).......................................................................... ................... 54

5-3 Critical Points (V alencia) ...................... ...................... .. .. ............. .......... 54

5-4 C critical P points (E arly-m id) ............................................................................ ....................54

5-5 Average Y early Standard D eviations ............................................ .............................. 55









































6









LIST OF FIGURES

Figure page

2-1 F C O J C contract Inform action ............................ ........................................... ............................24

2-2 H edging Outcom es with Increasing Prices......................................... ......................... 25

2-3 Hedging Outcom es with Decreasing Prices ........................................ ........................ 25

2-4 Hedging Outcomes with Strengthening Basis................................................................26

2-5 Hedging Outcom es with W eakening Basis ........................................ ........................ 26

3-1 Seasonality of B asis R esiduals ........................................................................ .................. 35

3-2 B asis R esiduals over T im e ............................................................................ ....................36

4-1 Valencia Spot Index............ ...... ........................ ........ .............. 46

5-1 All Valencia Scenarios .................................. ... .. .......... ........... .... 55

5-2 A ll Early-m id Scenarios .............................................. .. .. ............. .......... 56

5 -3 V alen cia P o o l ..........................................................................................5 6

5-4 Valencia Scenario 1 .................................... .. .... ...................... 57

5-5 Valencia Scenario 2 .............. ................. ........... ................. ........... ..57

5-6 Valencia Scenario 3 .................................... .. .... ...................... 58

5-7 V alencia C change in B asis PD F ............................................................................... ........ 58

5-8 E early -m id P ool ............................................................................... 59

5-9 Early-mid Scenario 1 ..................................... ................. ............ .. 59

5-10 Early-mid Scenario 2............................. .................... ............ 60

5-11 Early-mid Scenario 3 ..................................... ................. ....... ..... 60

5-12 E arly-m id C change in B asis ............................................................................. .............. 61









7









Abstract of Thesis Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Master of Science

EVALUATING RISKS AND RETURNS ASSOCIATED WITH ALTERNATIVE
MARKETING STRATEGIES FOR PROCESSED CITRUS

By

Evan Shinbaum

May 2008

Chair: John VanSickle
Major: Food and Resource Economics


The citrus industry is the largest agricultural sub-sector in Florida with 719,000 acres in

use. Over 90 percent of the oranges produced in Florida are Valencia and early, midseason

oranges grown primarily for processing. Volatility in the citrus industry caused by weather,

disease, and other factors is a concern for citrus growers. Alternative marketing strategies are

available to growers to manage this risk. This study provides an examination of alternative

marketing strategies and forecasts potential outcomes. The results provide citrus growers with

knowledge about these strategies and their forecasted returns and risks.

This study examines the risks and returns associated with 4 scenarios for each of the two

major types of citrus used in processing: 1) the pooling method, in which growers pool their

crops and receive payment based on the average price received for the entire pooled crop; 2) the

1-week cash market, which assumes the producer sells his/her entire crop into the cash market in

one week; 3) the 3-week cash market, which assumes the grower sells his/her entire crop into the

cash market over the course of three weeks; and 4) hedging, which assumes that the producer

hedges his/her crop utilizing the FCOJ contract in the futures market when the previous seasons

crop is ending.









Forecasted spot market and pooling prices are obtained using the forecasted all-orange

prices from a recent study. Simetar@ (Simulation for Excel to Analyze Risk), an Excel add-in

that was developed explicitly for stochastic simulation modeling was used to create stochastic

forecasts utilizing a pseudo-random number generator and observed probabilities. Using the

stochastic forecasted prices along with stochastic forecasted yields in conjunction with a pro-

forma financial statement model, the net present value (NPV) of discounted revenues through the

year 2020 is determined.

Pooling provided the highest standard deviation as well as the highest NPV for Valencia.

The one-week cash market scenario yielded the second highest NPV. The hedging scenario

provided the lowest NPV along with the lowest standard deviation. For early, midseason, the

pooling baseline provided the highest NPV combined with the highest standard deviation, that is,

the highest return with the highest risk.









CHAPTER 1
PROBLEM STATEMENT AND OBJECTIVE

Problem Statement

The citrus industry is the largest agricultural sub-sector in the state of Florida, with more

land being used to grow citrus than any other state in the country. With 719,000 acres used,

citrus accounts for over 40 percent of all land used to farm crops in the state (NASS). In 2005,

Florida accounted for 67 percent of the total U.S. production for oranges at roughly $843 million

(Florida Agriculture). Out of that, over 95 percent of all oranges grown in Florida are processed.

The three major types of oranges grown in Florida include the early-mids, Valencias, and

naval oranges. As implied by the name, the early-mids are harvested earlier in the crop season.

The Valencia orange, also known as Murcia orange, is a late season fruit. Valencia oranges have

zero to six seeds per fruit, however the excellent taste and internal color make it desirable for the

fresh market as well as for processing juice. Naval oranges, also known as the Washington,

Riverside, or Bahie naval, develop a second orange at the base of the original fruit, opposite the

stem. Because these are seedless oranges, the only means available to cultivate more is to graft

cuttings onto other varieties of citrus trees.

Of the 140 million boxes of citrus that were utilized through processing in the 2005-2006

season, early-mids and Valencias accounted for over 95 percent. While over 93 percent of all

early-mids, and over 96 percent of all Valencias were processed, less than one-third of all naval

oranges were processed. Early-mids and Valencias accounted for over 97 percent of all citrus

grown in the state (NASS).

With ever increasing volatility in the citrus industry, citrus growers are faced with crucial

decisions regarding the marketing strategies they use for their products. Supply is constantly

affected by threats of weather such as hurricanes and freezes, diseases such as canker, greening,









and tristeza, and even urban development (Brown).While the southern movement of the citrus

industry in Florida has helped decrease the exposure to freeze damage, the threat of hurricanes

comes every year. Because of these supply shocks, volatility is likely to remain high.

The futures market at the ICE (Intercontinental Commodity Exchange, formerly known as

the New York Board of Trade) in New York City provides a common method of pricing frozen

concentrated orange juice (FCOJ). While the market presents processors and growers with a

place to buy and sell FCOJ, only a small amount of actual FCOJ is sold in these markets. The

markets much more common purpose is to be used as a hedging and price discovery tool for both

processors and growers while speculators provide the necessary liquidity while hoping to

capitalize on price changes. With the FCOJ futures market, citrus growers can lock in a floor

price for selling their product. This enables them to protect themselves against an adverse price

change. While this elimination of risk with respect to adverse price movements can be helpful,

the ability to take advantage of favorable price movements is greatly diminished in hedging

programs (less so with the use of options as opposed to direct futures positions).

There is considerable price risk in FCOJ as evidenced in the futures markets. The seasonal

coefficient of variation of the average weekly closes for the past 5 seasons (2002-2007) averaged

over 12 percent. Also, volatility can be higher in the FCOJ market because the market for the

FCOJ contract is not as liquid as others such as corn. A large number of FCOJ contracts traded at

any time can cause a movement in price.

There are many other factors that can cause high volatility within the FCOJ futures

markets. Hurricane season serves as a great example of this. Because of the uncertainty caused

by hurricane seasons, volatility within the FCOJ can increase before a hurricane affects it. As the

hurricane season begins, the price in the market will generally increase, as more and more









speculators take long positions to prepare to capitalize on any formed storms that can damage the

orange crop. As storms develop and trend toward or away from the state of Florida, prices

fluctuate dramatically as information on the storms evolve. This is a great example of a situation

in which the market price reflects potential for occurrences rather than current supply and

demand.

Diseases pose an especially large risk to citrus growers because they can impact the

production of a grown crop for several seasons. There are certain production practices that are

used by growers to reduce the risk or uncertainty of these diseases. Tristeza, also known as CTV

(citrus tristeza virus), is a viral species that causes the most economically damaging disease to

oranges. It has accounted for the death of millions of citrus trees, and was coined the term

"tristeza" from farmers in Brazil and South America, as it is Spanish and Portuguese for sadness.

Citrus canker is an infection that causes lesions on the leaves, stems, and fruit of the trees. This

causes leaves and fruit to drop prematurely. While these fruit are still eatable, they are generally

blemished and harder to sell to consumers. The other main disease is greening, which is

distinguished by the common symptoms of yellowing of the veins and adjacent tissues, followed

by yellowing or mottling of the entire leaf and premature defoliation. The trees will have stunted

growth, bear multiple off-season flowers, and produce small, irregularly-shaped fruit and the

disease may often lead to the death and replacement of trees.

Perhaps the most common, easy to use, and easy to understand strategy of marketing is the

use of pools. Pooling is simply a process where the proceeds from many sales of a particular

commodity are averaged and growers all receive the average price after costs have been

deducted. Pooling still has some volatility as well, though not as much. With early-mids, over the

past 15 seasons the mean return for pooling was $1.01 with a standard deviation of 26 cents,









while the mean return for Valencias was $1.157 with a standard deviation of over 22 cents. Both

pooling and hedging in the futures market are risk management strategies, as opposed to growers

selling straight in the cash market at the current price at the time of harvest.

The option of selling directly into the cash market is generally perceived to hold the largest

risk for growers, and the most volatility. Selling directly into the cash market means that you

accept the current price at the time of harvest, which can differ significantly from the price at the

beginning of the production season. This makes budgeting more difficult, but does allow the

farmer to capitalize on the market when the price rises during the production season. In situations

where disease or weather hurts the crop, farmers who still hold their product are in great shape to

capitalize on the favorable increased price that results from the decreased supply. Many farmers

take out crop insurance to protect themselves from production risk that can result in a damaged

crop.

One final method of price risk management is the use of forward contracts. A forward

contract is an agreement between two parties to buy or sell an asset (in this case citrus) at a pre-

agreed future point in time. The trade date and delivery date are therefore separated. Most

forward contracts don't have standards and aren't traded on exchanges. A farmer would use a

forward contract to "lock-in" a price for his oranges for the upcoming harvest.

Objective

The general objective of this study is to enhance the citrus growers' knowledge of

marketing strategies available, and help to provide information regarding the differences in risks

and returns associated with the alternative strategies. The specific goal of this study is to create a

model which will demonstrate the forecasted situations, allowing a prediction of the best

marketing strategy available, under a given set of assumptions.









It is the goal of this study to become a tool for citrus growers as well as pool operators to

generate new ideas for alternative marketing strategies. It should be of interest to all growers to

be informed of, and understand the options available to them regarding their marketing

strategies, and be aware of the risks associated with them.

Organization of Study

The following chapter will give a background on the citrus industry and FCOJ, the futures

and options market, risk management, and basic hedging strategies. Chapter 3 reviews literature

related to the citrus industry and hedging. Chapter 4 presents the methodology, including data

collection, the model used, and analysis. A discussion of the results is presented in Chapter 5,

before summarizing and drawing conclusions in the final chapter.









CHAPTER 2
BACKGROUND AND PROBLEM SETTING

Citrus Industry

History

The oldest reference to citrus comes from Sanskrit literature. Jambhila is the name applied

to citron and lemon in the White Yahir-veda, specifically in the Vajasaneyi Samhita, part of the

Brahmin sacred book which dates prior to 800 B.C. The Citron was sanctified in India, and

consecrated to Ganesh, the God of wisdom and knowledge (Scora). In non-rabbinic Jewish

Tradition, a citron in the house keeps the Karines (bad spirits) away, while the Romans used

citrus as a basic body oil. Citron was the first of all citrus to reach the west and also the first

citrus fruit to grab European attention. This was all mostly due to its relative imperishability

combined with pleasant odor and appearance that made it suitable for long travel. In Medieval

times, visiting dignitaries would be entitled to a certain amount of sweet orange slices that were

detailed in cookbooks. As an expensive item, citrus became a fashion of the rich merchants

(Scora).

Citrus Industry in Florida

Citrus has been produced commercially in Florida since the mid-1800's. The environment

of Florida provides a comparative advantage for citrus production due to natural resources such

as the subtropical climate (Hodges). Also, the abundance of rainfall and water causes the soil to

be light and infertile, so frequent and heavy fertilization is necessary. This helps the quality of

oranges to be very good (Crist). Several freezes in north central Florida in the 1980's caused

crop loss, permanent tree damage, and large price fluctuations (Ranney). The industry responded

by moving into the southern region of the state (Andre). Florida citrus production followed an

upward trend from 1990 to 2000, increasing 46 percent; oranges and grapefruit being the top









commodities (Hodges). The Florida citrus industry contains much more processed oranges than

California, the second largest citrus producing state, mostly due to the better taste of the Florida

oranges, while the better look and texture of California's oranges cause them to be used as fresh

fruit (Andre). In fact, 94 percent of Florida's round-oranges are processed (NASS).

FCOJ Industry

Research that led to the development of Frozen Concentrated Orange Juice (FCOJ) in 1947

allowed Florida to expand orange production. Florida produced 57.5 million boxes of oranges in

1947 with as much as 76 percent of the round-orange crop in the state processed. In 1948, there

were only three processors, but by 1952 there were eighteen. The FCOJ contract was created in

1966, trading under the newly formed Citrus Associates of the New York Cotton Exchange. By

1980, the states record crop of over 206 million boxes of round oranges was produced, and over

83 percent of that crop was processed into FCOJ (NASS).

FCOJ stocks vary seasonally, with inventories approaching their maximum levels in the

latter months of the harvest season. "Freeze bias" refers to the bidding up of futures relative to

spot prices prior to the end of a freeze / hurricane risk period due to the potential for speculative

windfall gains that could be realized if a freeze causes considerable fruit loss. Large swings in

FCOJ prices caused by freeze and hurricane damage to trees are unique among commodities

traded in the futures markets (Malick).

Futures Markets

Futures markets are organized exchanges of derivative markets where many buyers and

sellers meet to trade futures contracts on an ever-expanding list of commodities (Salnars).

Futures markets can be traced back as early as the Middle Ages, when markets were developed

to meet the needs of farmers and merchants. The first futures-type contract was known as a to-

arrive contract. The Chicago Board of Trade (CBOT) was established in 1848 to bring farmers









and merchants together, and in 1874, the Chicago Produce Exchange was established, beginning

with markets for butter, eggs, poultry and other perishable agricultural products (Hull).

The futures contract helped provide a way for both the producing farmer and the

purchasing company to eliminate the risk it faces due to uncertainty of future prices. The main

task in originally developing the markets was standardizing the qualities and quantities of the

grains being traded (Hull). The futures market can be used by citrus growers to manage price risk

while making sure they have somebody to purchase their product, while the processors can use it

to manage price risk and make sure they have the ability to get the product. Speculators use

various strategies in the markets in an attempt to make profit off of price movements.

Speculative traders provide the necessary level of trading activity or "liquidity" in the market to

prevent the occurrence of added risk from failure to establish or terminate a contract (Ward,

1974). The final function of the futures markets is price discovery. Futures prices change as

buyers and sellers judgments about a commodities worth at a particular point changes. These

judgments are subject to supply and demand developments, and the availability of current

information. This continuous process is known as price discovery (Salnars).

Traditional futures contracts have been traded by what is known as the open-outcry

system, in which the traders physically meet on the floor and indicate the trades they would like

to carry out using a complicated set of hand signals and open outcry. This is still used during

regular trading hours, however in recent years, electronic trading has grown largely as an

alternative (Hull). There are certain terminologies that go along with the "futures language". To

go long on a futures contract is to buy the contract that would allow taking delivery of the

specified commodity; to go short on a futures contract is to sell the side of the contract that









would allow making delivery of the specified commodity. In its simplest terms, to go long is to

be the buyer of a commodity and to go short is to be the seller of the commodity.

Options

In addition to futures contracts, option contracts are offered for many commodities, giving

the holder (buyer) the right, but not the obligation to take a position with a futures contract.

There are two types of options, the call option and the put option. A call option gives the holder

the right to buy a futures contract by a certain date for a certain price, while a put option gives

the holder the right to sell a futures contract by a certain date for a certain price. This certain

price is referred to as the 'strike price'. The buyer of an option (frequently referred to as the

holder) is long on that option, while the seller (frequently referred to as the writer) is short the

option. An American option can be exercised at any point during its life, while a European

option can only be exercised on its maturity date (Hull).

The first trading in puts and calls began in Europe and the United States as early as the

eighteenth century; however corrupt practices gave the market a bad name in its early years. The

Put and Call Brokers and Dealers Association was set up by a group of firms in the early 1900s,

and in April of 1973, the Chicago Board of Trade set up the Chicago Board of Options Exchange

(CBOE) which is the largest exchange in the world for trading stock options (Hull). Options first

appeared in America in the early 19th century, and were known as "privileges". In 1934, the

Investment Act legitimized options, and put it under the eye of the recently formed Securities

and Exchange Commission. On April 26, 1973, the Chicago Board of Options Exchange began

trading listed call options. By the end of 1974, average daily volume exceeded 200,000. The

FCOJ options became available in the New York Board of Trade (now the ICE) in 1985.









Risk Management

There are various ways of managing risk for the citrus grower. Pooling, perhaps the most

common, involves gathering the citrus from several growers and paying the growers the average

sale price for the product in the specific pool. There are also forward contracts, which are similar

to a futures contract in that it is an agreement to buy or sell product at a certain time in the future

for a certain price. However, while futures contracts are traded on exchanges, forward contracts

trade over-the-counter (Hull). This means that forward contracts are created by two parties who

agree on a delivery date. These are not set quantity and quality amounts traded like they are on

the futures exchange.

Hedging is a risk management tool that generally uses the futures and options markets.

When a citrus grower takes a short position in the futures market equivalent to the quantity of

citrus they are producing, they are lessening their price risk. With a short position in the futures

market, the grower is putting themselves in a position to offset losses in their sale value of citrus

if the price drops with a positive gain in the futures market. If the price of citrus rises, the loss in

the futures market is offset by the higher price for which they can sell their citrus. The

transaction costs become the only truly lost value, which is generally small relative to the price

risk associated with the commodity.

There are multiple transfers of risk taking place in this type of risk management. The

transfer from price risk to basis risk for a hedger, and the acceptance of price risk by speculators

provides the necessary liquidity for the market to work. Using this method of risk management,

the basis risk becomes the main point of interest for hedging. Understanding basis and being able

to forecast it is the key issue in hedging (Salnars).









Basis

Basis refers to the difference between the price of a given commodity in the nearby futures

contract and the price that same commodity could be bought or sold for in the spot market (also

known as cash market). The equation used is: basis equals cash price minus futures price.

B =CP FP (2-1)

Where B represents basis, FP represents futures price, and CP represents cash price

In reality, basis is the connecting link between the present and future (Malick). Cash and

futures prices generally move together and react to continuous changes in supply and demand

factors (Salnars). Basis residuals are a calculated deviation from the average basis (Malick), and

can be represented as

BR = BM ABS (2-2)

Where BR represents the basis residual, BM represents the basis for a specific month, and

ABS represents the average basis for a season. Basis residuals refer to the difference between

expected basis and actual basis.

Basis can change based on many factors. An increase in basis is referred to as a

strengthening basis, when the spot price increases by more than the futures price. In a normal

market where the futures price is higher than the relative cash price due to carrying costs (such as

time, place, and quality) a strengthening basis is a narrowing basis. A weakening basis is just the

opposite, in which the futures and cash prices diverge in a normal market (Hull). The hedging

term used for someone that is planning on selling goods is a short hedge, because their initial

position taken in the futures market is a short position. A long hedge is used for managing the

risk when purchasing a commodity, and is accomplished by first taking a long position in the

futures market.









Basis in the Citrus Industry

Basis levels in general can be affected by product quality, transportation, location,

insurance, time, storage, delivery method, or any combination thereof (Salnars). Basis risk in

FCOJ is hypothesized to be affected by six major variables: a measure of risk premium,

convenience yield, market liquidity, freeze / hurricane bias, bias adjustments, and an actual

freeze / hurricane event (Ward, 1977). Basis risk decreases as the time to the hedge expiration

(delivery month) decreases. If the asset to be hedged and the asset underlying the futures contract

are the same, the basis should be zero at the expiration of the futures contract (Hull).

If nearby prices (the price of the nearest futures contract traded on the exchange at any

given time) reflect a current shortage in cash product, some convenience yield exists for having

at least a minimal stock held for future consumption. This yield offsets at least part of the

carrying cost of holding the stocks in inventory (Ward, 1977). Lower stocks help push cash

prices up relative to futures encouraging stock holders to draw down existing inventories

(Malick).

Hedging Examples

The benefit of implementing a short hedge for a producer can be demonstrated as shown in

figure 2-2. The basis in figure 2-2 remains constant over the period of holding the hedge to show

the general objectives of hedging with futures.

The current (June) spot market price, per pound solid FCOJ is $2.05 in figure 2-2. The

current futures price is $2.25. The basis therefore is -0.20. A grower who plans to sell 75,000

pounds worth of FCOJ in September, would need to take a short position with 5 contracts (each

contract contains 15,000 pounds) to offset any losses in the spot market between the time the

hedge is implemented and the time the FCOJ is sold in the cash market (which should be about

the same time as the contract maturity date).









In Figure 2-2, the prices in the spot and futures markets rise to $2.30 and $2.50,

respectfully, by September. While there has been a 25 cents increase in the spot price, this gain is

offset by a loss of 25 cents in the futures position held by the hedger after liquidation. Therefore

the realized price that the grower receives for his crop remains the original $2.05 due to a lack of

change in the basis ($2.30 for the cash market combined with a 25 cents loss in the futures

contract). At 75,000 pounds of goods, the total sale value is $153,750 (note: this simple scenario

neglects transaction costs).

In a second scenario (shown in Figure 2-3), the prices in the spot and futures markets fall

to $1.65 and $1.85, respectfully, by September. In this situation, the 40 cents per pound solid lost

in the spot market is offset by an equal gain in the short futures position after liquidation. Again,

the realized price for the grower would remain equal to the original spot price of $2.05, due to no

change in the basis (ignoring transaction costs).

The above examples hold basis constant and disregard transaction costs, neither of which

generally occurs in the market. Figure 2-4 demonstrates the outcome of a hedging transaction

when basis strengthens. Again, the current spot market price is assumed to be $2.05 per pound

solid when the hedge is implemented, and the futures price is assumed to be $2.25. The current

basis in this scenario therefore is -0.20. A grower who plans to sell 75,000 pounds worth of

FCOJ is long 75,000 pounds of FCOJ in the cash market. He would take a short position with 5

contracts to manage the price risk held with his current cash position.

In this third scenario shown in Figure 2-4, the prices in the spot and futures markets rise to

$2.40 and $2.55, respectfully, at the time of maturity. In this example the basis has gone from

-0.2 to -0.15, thus the basis has strengthened. In this case, the grower has gained 35 cents per

pound solid from the increase in spot market price, and has lost 30 cents per pound solid in the









futures position after liquidation. With basis strengthening by 5 cents per pound solid, the grower

has now gained 5 cents per pound solid, and the realized price is the original $2.05 plus the 5

cents gained, resulting in a realized price of $2.10. At 75,000 pound solids worth, this is a net

gain of $3,750.

Figure 2-5 demonstrates the outcome of a hedging transaction when prices in the spot and

futures markets fall from $2.05 and $2.25 to $1.25 and $1.55, respectfully. The basis in this

scenario has weakened from -0.2 to -0.3. The outcome of the weakening basis results in an 80

cent loss per pound solid in the spot market and a 70 cent per pound solid gain on the short

position in the futures market. Subtracting the 10 cent net loss between the two positions on the

original $2.05, the new realized price that the grower receives has decreased from $2.05 to $1.95.

With 75,000 pounds of product to sell, the grower lost $7,500.

While the transaction costs were ignored in these examples, it is not hard to see how

important basis is in hedging. It is easy to see how understanding basis, and being able to predict

it's behavior can be important to the grower. Taking a futures position at the right time and

gaining additional returns from strengthening in the basis can be the difference between a

successful harvesting season and a failure.

To account for transaction costs, we can assume the cost of taking a futures position round

trip (meaning the cost for taking to original position, plus the opposite position to liquidate) is

$30 per contract. In the third example, the realized price increased from $2.05 to $2.10, a 5 cent

increase per pound solid. With a $30 round trip transaction cost for the futures position, the per

pound solid transaction cost is $0.002 per pound. Now the realized price is the $2.10, minus the

00.002 of transaction costs, and the realized price is $2.098 per pound solid. At 75,000 pound

solids, this is a $150 transaction cost loss.










The spread between the futures and spot market price can be explained in part by the net

marginal outlay for storage, which is defined as the marginal outlay for physical storage plus a

marginal risk-aversion factor minus the marginal convenience yield on stocks (Ward, 1977). This

is consistent with storage theory, which holds that the basis equals the marginal costs of carrying

stocks through time (Malick). Historically, the futures basis has been judged by how well it

reflects the market fundamentals; however the basis may yield discounts and premiums if it

reflects anticipation of future events or the potential for such events (Ward, 1977).


StotDci 02


Exc bngk


New Vark Board of Trade


Descriptio FCOJ A (Frozen Cane 0 BZ&FL)
Cal Symbol: 01
Put SymrbW: OJ
Amoun, 15.000
Uniti Pounds (Ibs)
Stait Timl IO00 EST
End Tim.i 13:30 EST
Last Day Start ibms 10:00 EST
Last Day End Tirme 13:30 EST
Opbon Start TVfOW 10O00 EST
Option End Timi 13:30 EST
iptIt Last Day Stat TUei 10:00 EST
Option Last Day End Time 13.30 EST
Unit String Dollars pr Pound
Caemesion Faetor: 0.0100
Convrenms Factor Unit Dollars per Pound
Ptrdut TVype Commodity
FCOJ-A contract specifications provided by FACTSim
(www.FACTSIM.org)


Figure 2-1 FCOJ Contract Information














June -0.2

Cash market price of $2.05 Sell 5 contracts at $2.25

September -0.2

Sell 75,000 Ibs at $2.30 Buy 5 contracts at $2.50


Gain of $0.25


Cash


Loss of $0.25


IFutures


Basis


June -0.2

Cash market price of $2.05 Sell 5 contracts at $2.25

September -0.2

Sell 75,000 Ibs at $1.65 Buy 5 contracts at $1.85


Loss of $0.40


Gain of $0.40


Figure 2-3 Hedging Outcomes with Decreasing Prices


Gains
Losses


Realized
Price
SSold 75,000 Ibs at $2.05

Figure 2-2 Hedging Outcomes with Increasing Prices


Date


Gains
Losses


Realized
Price
S Sold 75,000 Ibs at $2.05





Date


I Futures |


Basis


Cash |














June -0.2

Cash market price of $2.05 Sell 5 contracts at $2.25

September -4.15

Sell 75,000 Ibs at $2.40 Buy 5 contracts at $2.55


Gain of $0 35


a 4


Realized
Price
S Sold 75,000 Ibs at $2.10


Loss of $0.30


Figure 2-4 Hedging Outcomes with Strengthening Basis


Cash


SFutures


Basis


June -0.2


Cash market price of $2.05 Sell 5 contracts at $2.25

September -0.3


Sell 75,000 Ibs at $1.25 Buy 5 contracts at $1.55


Loss of $0.80


.1 &


Realized
Price
Sold 75,000 Ibs at $1.95


Gain of $0.70


Figure 2-5 Hedging Outcomes with Weakening Basis


Gains
Losses


Date


Gains
Losses


Date


I Cash I


I Futures I


Basis









CHAPTER 3
REVIEW OF THE LITERATURE


This chapter reviews the most highly related studies to risk management in citrus.

Previous studies in this field are generally divided into two themes: basis and hedging in the

citrus industry. Basis refers to the difference between the spot market price and the price of the

same commodity in the futures market. Specifically, basis is calculated as the cash price less the

futures price. Hedging refers to the risk management processes of citrus growers.

Malick and Ward (1987) studied the stock and seasonality effects of FCOJ. They used a

Constant Period from Maturity (CPM) model to measure seasonality. They concluded that when

stocks are relatively low, cash prices are pushed up relative to futures prices, encouraging stock

holders to draw down existing inventories. Seasonality became more pronounced in the later

months of the citrus harvesting season when inventories approached their maximum levels.

According to these authors, one of the largest seasonal effects that arise is very unique to the

citrus industry. This is the large swings in FCOJ prices due to freeze losses.

"Freeze bias", as termed by Frank Dasse and Ronald Ward (1977), refers to the bidding up

of the futures contract relative to spot prices before the end of a freeze period due to potential

speculative gains that could be achieved in the event of a freeze. These authors used bulk FCOJ

wholesale price (that is the price after processing the oranges into Frozen Concentrated Orange

Juice) as the spot price for basis calculating (futures cash), as opposed to the spot market price.

Current stock levels play a large part in determining the level of price adjustments; indirectly

related, larger stock levels result in smaller price adjustments because of the ability to

compensate for losses.

Using the Malick and Wards CPM model set to 5 months, a graph of the seasonal nature of

FCOJ basis residuals is shown in Figure 3-1 using 3 different levels of stock,. 1.0, 1.2, and 0.8









stock levels were used, where 1.0 is the 'norm' or average. At stock level 1.2 for example, the

stock level is 120 percent of the average stock. The basis residuals are a calculated deviation

from the average season basis. For example, with stock levels set at 1.0, the basis residual in

Figure 3.1 in April is just below 1. This means that the seasonality effect at that moment causes

basis to be higher by 1 than would otherwise be predicted without seasonality.

The basis residual turns positive in the October period for this model. This is generally

attributed to the potential for freezes. Since the migration of the citrus industry to the south,

lower freeze risk can attribute to the stock levels deviating slightly at the December peak as

opposed to the merge seen in Figure 3-1 The decline in the basis residual during the summer

months through August can be attributed to diminishing stock levels, which is why basis is so

accentuated with the lowest stock level in the example of 80 percent.

Hurricane seasons now have a very similar effect on basis due to their ability to destroy

large portions of a crop. Also, since Dasse and Ward published their work, Florida has played a

much smaller role in the overall market for FCOJ, causing the effects of weather to be less

elastic. There are many contributing factors to this, which include a loss of Florida crop due to

hurricanes, canker, greening, and tristeza, while Brazil and Mexico have increased production to

compensate. This has caused a decrease in the price sensitivity to freeze / hurricane destruction

possibilities.

Malick and Ward reference the basis turning to zero at maturity, but there have been

changes since they published their work that might suggest otherwise. For instance, the delivery

points were all in Florida when Malick and Ward conducted their research, while FCOJ can now

be delivered to many other areas of the United States. This in combination with the lowering

proportion of citrus that comes out of Florida compared to other areas of the country and world









can result in much higher delivery costs than in the past for producers. This would tend to have

the effect of keeping basis from disappearing at maturity.

Ronald Ward and Frank Dasse present the uniqueness of each commodity, and how it

deviates from a generalized theory for storage and basis. In this particular article, the commodity

being analyzed is FCOJ. Storage theory suggests that a futures basis should reflect the marginal

cost of physical product plus a risk premium minus a convenience yield. Ward and Dasse bring

out ideas that might disrupt this theory, such as premiums and discounts as it reflects anticipation

of future events, such as the case with freezes and hurricanes.

Ward and Dasse bring up other interesting points that affect the market. While the season

officially begins December 1st, the first crop estimate by the United States Department of

Agriculture (USDA) is released in October. However the December to February time period has

the higher probability for possible freeze and the late crop harvest isn't until July. Towards the

end of the season there is also added speculation about the next year's crop. This may be

reflected in contracts that span over these months.

Ward and Dasse show the basis residuals over a 7 year timeframe in Figure 3-2. The July

contract was selected by Ward and Dasse for analysis in their basis study because it extends over

the complete harvesting period and therefore should reflect a storage cost while being far enough

from the next season to not be heavily influenced by the expected crop.

Ward and Dasse use 6 major variables that are hypothesized to contribute to the FCOJ

basis residual. These include: measure of risk premium, convenience yield, market liquidity,

freeze bias, bias adjustments, and actual freezes. Ward and Dasse concluded that "The basis

model developed clearly supports both the theory of storage and establishes the necessity for

measuring market bias." (page 79). While Ward and Dasse say that much of the basis theory









cannot be clearly related to the practicing hedger nor is it necessary, they do feel that it is

essential that traders using the market have confidence in its economic performance.

It should be noted that like Ward and Malick, the article does not reference other crop-

destroying variables such as hurricanes and diseases. This is most likely due to when the article

was written, i.e., prior to many of the hurricanes and diseases seen in recent years.

Behr (1981) aims to "develop the conceptual net benefit model characterizing futures

market behavior in terms of hedging open interest, speculation open interest, and volume." In

this dissertation, Behr examines the cross-sectional, as well as time-based influences that can

affect market behavior. Included is market structure, government involvement in commodity

markets, price risk, and commodity perishability.

Salnars (2004) addresses the idea of capitalizing on price volatility as contracts approach

maturity. This can provide great opportunities for hedging. This high reward comes with an

equally large risk, as the small volume of trading can lead an unprepared hedger into an

unwanted hedging position. Salnars discusses the convergence of the futures and cash prices as

the contracts approach maturity. One of the reasons for this convergence (also known as

narrowing of the basis, or strengthening) is due to available information. As the maturity of a

contract approaches, the amount of information increases and the uncertainties about the future

that cause futures and cash prices to differ become fewer. Also, as the contract approaches

maturity, if there is not convergence between the cash and futures prices, speculators may

arbitrage between the two, causing the prices to merge while they take in a profit. In Salnars'

example of this, a scenario where futures prices are higher than cash prices as the contract

approaches maturity; profit could be made by buying the commodity in the cash market and

selling it in the futures market, making delivery of the commodity.









Ward and Niles prepared a report for the Futures Committee of the Florida Department of

Citrus, which contains a very well rounded approach to the varieties of hedging strategies that

were available to citrus growers (as of 1975). The report includes background information about

the futures markets, the contract specifications, types of hedging, price data and statistics, and

margin requirements. It then goes into a "Hedging Decision" section, in which it includes

information about risks, types of goods that can be hedged, basis patterns, and broker selection.

The final section discusses alternative hedging programs. In this section Ward and Niles

include eleven different "Hedging Cases". Included are: Cash Grower With Opportunity Cost,

Grower With Product Uncertainty, Premature Termination of Hedge, and Forward Contracting

With Foreign Importers. This report provides a foundation for growers that are not familiar with

hedging strategies.

One of Wards (1974) most emphasized concepts is how speculation creates the liquidity

necessary to hedge. According to Ward, trading volume from speculative traders is necessary "to

facilitate establishing futures positions without realizing a cost from market entry or exit." It is

argued that if a trade cannot be easily executed at any point in time, then the risk from futures

trading is greatly increased. While many of these liquidity points may be considered common

knowledge, Ward not only discusses the causes and effects of speculation that create liquidity,

but also the effects of other scenarios, such as questioning whether there can be excess

speculation and what potential consequences could arise.

Ward also addresses lack of complete identification of the composition of open interest,

such as the affect on liquidity when the composition of traders is altered. For example, a report

might proclaim that a major portion of the open interest is held by small traders; however the

intent of small traders is not always identifiable. Ward takes a mathematical approach to testing









the liquidity within the FCOJ market. According to the conclusions of Ward, "The speculative

index and its relationship to price distortion provide a clear signaling mechanism to alert policy

makers of any needed structural changes in this futures market. Likewise, the methodology

reveals the need for improvements in the present system for monitoring futures commitments in

general."

Ranney (1993) discusses the level of inefficiency in the FCOJ option market "because the

market has low trading volumes". Ranney also provides examples with the Black model and

standard arbitrage condition tests. According to Ranneys' thesis, the Black model test "indicated

inefficiency in every month due to the violation of put-call parity." The arbitrage conditions test

also indicated that there was inefficiency due to high percentages of violations of arbitrage

conditions for every contract month. Ranney found that levels of inefficiency were higher than

those found for any previously studied commodity option market, and that the level of

inefficiency is significant and may adversely affect citrus industry hedging operations.

Staying with the subject of efficiency as a necessity for a good hedging strategy, Ward and

Behr (1983a) take a look at liquidity in the futures market in general, not within a specific

commodity. The article deals with two main questions: (1) What criteria are to be used to

measure liquidity, and (2) what level of liquidity is considered optimum. One of the methods that

is employed is using the spreads between bid and ask prices as input data for calculating a

liquidity index. An index of liquidity is developed by Ward and Behr and factors leading to

adjustments in the index are evaluated.

Ward and Behr (1983b) also provide an all around summary on futures commitments. Both

the problem with lack of commitment information and the costly and impractical reasoning for

the lack of information are presented. The ability to increase the accuracy of market performance









from knowledge of open-interest allocation is proved undeniable. At the time Ward and Behr's

article was written, the CFTC had begun providing a month-ending classification of futures

traders that hold large (excess of 25 contracts) positions. The data included the numbers of long

and short commitments reported by hedgers and speculators, and also included the long and short

position of non-reported open interest.

Ward and Behr present 4 alternative schemes for allocating non-reported short and long

positions to hedging and speculation. Two of the procedures, known as the Larson and the

Rutledge models, are based on estimates from data obtained through special market surveys. The

other two, denoted in the article as "Scheme 1" and "Scheme 2", are based on numerical

weighting procedures with the weights derived theoretically rather than from econometric

estimates.

Ward and Behr show that all but the Larson's model give very similar allocations over a

broad range of commodities, and also note that both the Larson's and Rutledge models

occasionally "lead to allocations that were clearly impossible." All four models did share 2

things of common interest: first, a general procedure is suggested that is to be applied across all

commodity futures, and, second, the procedures are based on some weighting of the reported

futures data.

Rolls (1984) provides an in-depth look at the relationship between the interaction of prices

and "a truly exogenous determinant of value, the weather". Due to the highly centralized and

concentrated area of orange groves in central Florida, the ongoing weather in a particular spot

market can have enormous consequences on the value of the product.

Rolls explains that not only is weather a huge determinant in value because of the

centralized location, but there is so little affect on price from other supply and demand factors.









The demand may have very subtle movements due to price changes in substitutes such as apple

juice or grape juice, but national income and tastes generally will not fluctuate enough to explain

any significant daily movements. On the same note, short term variations in supply caused by

planting decisions would be very low because oranges grow on trees that require five to fifteen

years to mature.

Rolls also went into detail about daily movements that are allowed within the FCOJ

contracts daily fluctuations, and how extreme circumstances, such as freezes and hurricanes, can

cause a necessity for price movements beyond the daily limits. In this case, the limits cause price

inefficiency.

Muraro and Oswalt (2003) published an extensive report on Florida citrus in September

2003. The report is statistically based with only a short introduction and mentions of data

collection methods as well as necessary assumptions. While the study holds many points and

lays a great foundation for understanding the percentages of risk, it appears to serve much better

as a reference tool.

Spreen, et al (2006) prepared a report on the current status and future prospect of the

Florida citrus industry. The article explains the cause and effects of current citrus problems such

as citrus canker as well as citrus greening. Both have become an increasing concern regarding

tree health despite the decline of tristeza.

The report covers four main sections: 1) analysis of the economic impact of the industry on

the Florida economy; 2) assessment of the future structure of the Florida citrus tree nursery

industry; 3) analysis of the future prospects for new investment in Florida citrus; and 4) long-run

production and price forecasts for Florida citrus under varying assumptions that relate to supply

issues. According to Spreen et al, the estimated economic impact of the citrus industry on the










Florida economy for 2003-2004, was over $9.25 billion, and accounted for over 76,000 jobs in

the state.

This significant impact of the citrus industry in Florida provides strong motivation for

work related to helping citrus growers. Previous research helps provide an understanding of

basis, its patterns, causes and effects. It also provides a fundamental knowledge regarding

hedging, however it falls short of analyzing alternative marketing strategies employed in a risk

management.


a







4I



m -m













Figure 3-1 Seasonality of Basis Residuals
Figure 3-1 Seasonality of Basis Residuals



























cQc.-JuY I O&C.-Jniy I t.-Ju y M-,*
/967-60 1965 -69 1P6970 1970-77




.JO









,05



!977-2 t972-73 1973-74


Figure 3-2 Basis Residuals over Time









CHAPTER 4
METHODOLOGY AND DATA

The Model

The model used in this study is developed by modifying the orange production model

developed by Weldon, et al. (2004). An analysis of the impact of the marketing efforts for citrus

producers must explicitly examine the variability or risk of future returns. Simulation is a

technique that analytically models or replicates an actual system and is used in many areas of

study. In agriculture it has been used to model plant growth, market conditions, animal growth

and reproduction, environmental and ecological systems, intergenerational farm transfers, and

many other situations. Simulation modeling can be used to examine farm income under

alternative risk scenarios and provides insights into how various firm decisions influence firm

survival. Simulating the production and economic activities over a multi-year planning horizon

also allows for examination of the impact of federal policy or programs, such as the Florida

citrus disaster program, on the profitability and survival of the citrus operation. In this study, the

simulations examine the risks and returns associated with various methods of marketing citrus.

Simetar@ (Simulation for Excel to Analyze Risk), an Excel add-in that was developed

explicitly for stochastic simulation modeling by Dr. James Richardson, is used to simulate risks

and returns to a Valencia and early, midseason (hereon referred to as early-mid) orange grower.

The risk and return framework for a representative 'Valencia' orange and early-mid orange

grower are modeled to simulate financial results over a twelve-year time horizon ending in 2020.

The model is based on annual cash flows including operating and fixed costs, and forecasted

prices and yields. It generates an annual income statement, pro forma cash flow statement, and

pro forma balance sheet statement. Since this is a stochastic simulation each iteration, the

pseudo-random prices and yields are drawn from a multivariate empirical distribution using a









Latin Hypercube sampling method. The forecasted random data is correlated based on historical

correlations.

For each iteration, the simulator provides the ending net worth at the end of the twelve-

year time horizon. The Net Present Value (NPV) is calculated as

NPV = We for year 2020 / (1+ i)13 Wb for year 2008 (4-1)

where i is the real discount rate, We is the ending net worth, and Wb is the beginning net worth.

The NPV is a measure of the amount of wealth that is acquired over the twelve-year period in

present dollars and assumes that all excess cash flows remain in the business and are reinvested

with a return equal to the discount rate. For each simulation, 500 iterations are run to yield a

distribution for the NPV. The mean NPV is the average of these 500 iterations, or the expected

wealth generated or lost for that particular scenario. Because the objective of this study is to

contrast the reward with the risk inherent in the scenario, the standard deviation for NPV is used

to assess the risk associated with that reward. The standard deviation measures the dispersion of

a probability distribution.

In the simulations, we estimate the mean NPV, the standard deviation for the NPV, and a

Yearly Average standard deviation. The NPV standard deviation measures the variation of the

net present value. The yearly standard deviation measures the volatility from year to year

associated with the income stream in each iteration of the simulation. The yearly average

standard deviation is the average of the yearly standard deviation's measured from the

simulation.

For each scenario, the acreage of the representative grower is assumed to be the average

acreage of a citrus farm in Florida in 2002 (Guci, 2007). It is assumed that the farmer owns the

entire acreage. It is also assumed that the farmer is not in debt.









The following comparisons are made for each citrus variety. First, an NPVbaseline is

calculated as the expected gain in wealth projected for the next twelve years assuming the

producer enters their entire harvest into a pool. Three alternative NPVscenarios are simulated for

both early-mid and Valencia citrus varieties. The results of the simulations for the three

alternative scenarios are compared to the pooling NPVbaseline to evaluate the risk and return of

the alternative marketing strategies.

The first alternative scenario that is included in the analysis assumes that the producer sells

the entire crop in a single week in the cash market when the oranges are harvested. The second

alternative scenario assumes that the producer sells the crop in the cash market equally

throughout three consecutive weeks following harvest. The third and final alternative scenario

assumes the producer takes a short position in the futures market for the next crop at the end of

the current harvest season, and liquidates that position when the following crop is harvested.

Forecasting Market Prices

The simulations require a forecast of price and yield for each season in simulation. Pooling

return price forecasts are calculated using an index ratio of Valencia pooling price and early-mid

pooling price relative to the all orange value1. Pool data for Valencia ranges from $0.982 to

$1.71, while pool data for Early-mids ranges from $0.789 to $1.63. This data covers the seasons

1987-1988 to 2002-2003.

Pooling returns represent the annual value received as opposed to a weekly price, so an

annual index ratio is used to forecast pooling returns instead of a weekly index. A stochastic

normal distribution of the index ratio (derived from the historical annual index) is used with the




1 Historical pool price data was obtained via interview process from Dr. John J. VanSickle, professor at the
University of Florida who originally obtained this data through Florida's Natural.









forecasted all orange value obtained through Spreen et al. (2006) to derive stochastic forecasted

Valencia and Early-mid pool prices.

A discrete empirical formula is used to select a harvest week for the first alternative

scenario, where the entire crop is sold in the cash market in a certain week within a season. The

Simetar program randomly selects a harvest week within the projected season based on the

probability distribution for harvest weeks in prior seasons2. Using those probabilities, Simetar

chooses a particular harvest week and the stochastic price of that week to calculate receipts and

all the other necessary components of the forecasted financial model. Because of the large

number of stochastic calculations involved, it is necessary to use a large number of iterations to

get a clearer picture of the probabilities associated with the final value. Simulating these data

with a large number of iterations yields NPV and standard deviation statistics that allow an

examination of the risk-return framework for the marketing strategy's.

The same methodology is used for selecting the harvest week for the other alternative

scenarios with changes in the model to reflect the alternative marketing scenarios. In scenario

two the crop is sold equally over three weeks. The simulation is the same as for scenario one,

except that when Simetar chooses a harvest week based upon the probability distribution for

harvest weeks, it uses the average of the price for that week plus the next two weeks to use as the

average selling price. In the event that the week chosen is the second-to-last week of the season,

the price is an average of the last two weeks. In the event that the week chosen is the last week

within a season, the price for that week alone is used.

A hedge scenario with the futures market requires specification of a time to implement the

hedge and a time to liquidate, or offset, the hedge. It is assumed that a hedge is initiated when the

2 These data were collected from "Utilization of Florida Citrus Fruit" reports from the Citrus Administrative
Committee and can be found through their website, www.citrusadministrativecommitee.org.









prior crop season is closing and lifted when the crop is sold in the cash market. A decision is

added in this scenario in which if at the time of implanting the hedge, the spot price multiplied

by the per-yield acre is greater than the production costs per acre, the hedge is implemented,

otherwise the grower takes their chances, and sells directly into a single week in the cash market

at harvest. For this scenario, a random week based on an empirical distribution is used for

liquidation of the hedge. The gain/loss from the futures position is taken in conjunction with the

spot market price at the time of harvest to determine net receipts for that year.

Forecasted spot market prices are required to simulate the returns for the three alternative

scenarios. To do this, historical "seasonal index" is calculated from the weekly spot market

prices for Valencias and early-mids relative to the average all-orange spot market price of that

season. Spot market data for both Valencias and early-mids were obtained through the USDA

National Agricultural Statistics Service (NASS)3. The Valencia spot market prices range from

$0.550 to $2.15 per pound solid, with an average of $1.193 for the years 1983 to 2006. Early-mid

spot market prices range from $0.425 to $1.75 per pound solid, with an average of $0.992 for the

years 1982 to 2006.

The all-orange price per box on tree and all-orange concentrate yield were also obtained

from NASS. The on tree price per box ranges from $2.89 to $7.58 with an average of $4.722. A

standard conversion rate of 4.156 (Citrus Reference Book, 41) is used to derive an all-orange

price per pound solid by dividing the all-orange on tree price per box by the product of the all-

orange concentrate yield and the conversion rate.

A seasonal index is derived for Valencias and early-mids prices by calculating the ratio of

the weekly Valencias or early-mid price over all-orange annual price. This product gives a

3 These data are available through their website: http://www.nass.usda.gov/Statistics_by_State/Florida, as well as
through the hard copies of their annual Citrus Summary reports.









distribution of weekly index values for each product. For example, in week 13 the mean seasonal

index value of the Valencia spot price over the all-orange spot price from the years 1983 to 2004

is 1.599 with a standard deviation of 0.358 (Figure 4-1).

The distribution of weekly seasonal index values is assumed to be normally distributed,

and a stochastic function is used to create a random index for each week in each season to be

simulated. The series of weekly seasonal index values are then correlated with a correlation

matrix to insure continuity of prices within a season.

Using forecasted all-orange spot market prices, these stochastic seasonal indexes (along

with the forecasted price) are used to obtain weekly forecasted stochastic spot prices for valencia

and early-mids over the projected seasons. The forecasted all-orange spot price used in this

analysis is derived from Spreen, et al (2006). The forecasted all-orange spot price (converted into

per pound solid units) covers the seasons of 2007-2008 to 2019-2020 and ranges from $0.846 to

$1.472 with an average of $1.263.

Using weekly historical futures price data and weekly historical spot market price data, a

historical Valencia basis and an early-mids basis are calculated. This yields a weekly Valencia

basis which is used to calculate a mean and standard deviation. These two statistics are used to

produce a normally distributed basis for each week. These values are also used to derive a

correlation matrix for the weekly basis to insure seasonal consistency from week-to-week. For

example, with the correlation in place, there is a much smaller probability of the basis swinging

from one tail of the normal distribution to the other tail over the course of a week.

When the forecasted stochastic Valencia spot price in a given week is combined with the

stochastic Valencia basis, a stochastic forecasted Valencia based futures market price can be









developed. This process is repeated with the early-mid spot market price and basis data to

develop the forecasted futures market price.

For the final marketing strategy, hedging using futures, historical nearby contract data was

collected. Nearby futures contract data was received from the New York Board of Trade

(NYBOT), which is now the Intercontinental Exchange (ICE). The period of the data run from

October, 1982 to May, 2007, with prices ranging from $0.551 to $2.072.

To develop a discrete empirical distribution for the harvesting time, data was collected

through the Citrus Administrative Committee. This data contains the past 3 years of quantity of

data processed weekly for both Valencia and early-mid. Using this data, the percentages were

broken into the necessary number of categories needed, 10 for Valencia, and 17 for early-mid.

Using the Dempirical and Hlookup functions, a system is set in place for the model, in any

given year, to choose a harvesting week based on historical probabilities, and the stochastic

forecasted price from that week for the necessary year is used. For the scenario involving a

distribution of harvesting over three consecutive weeks, a system is set in place for any given

price to be chosen in which an average of the given week plus the next two weeks is used. In the

event that the 2nd to final week is chosen, it retrieves an average of the last two weeks. In the

event that the final week is chosen, that value is used. This process is done for both Valencia and

early-mid prices.

Forecasting Futures Prices for the Hedge Model

The hedging scenario, for both Valencias and early-mids requires a few key assumptions. The

first assumption is that the hedge is implemented at the end of the previous year's harvest season.

It is at the end of the previous harvest season that the following crop development begins and the

grower incurs production costs. The growers hedged amount comes from an average of his

forecasted yield for the upcoming year, and the realized yield of the two previous years. The









basic hedge formula uses the spot price when the hedge is implemented, and adds the change in

basis and subtracts transaction costs (brokerage fees) for trading futures contracts. The

transaction cost (each way) is assumed to be $20 per contract (at 15,000 pound solid per

contract), a citrus industry norm.

Like reality, the process of hedging an amount based on previous years yields leaves the

chance of over and under hedging. In the situation in which the crop is under-hedged, the excess,

un-hedged portion of the crop is assumed to be sold into the cash market at the week at harvest.

In the event that the crop is over-hedged, the gains or losses accumulated from the excess futures

position are added or subtracted from the final revenue for the given year.

Forecasted Yields

Historical farm-level yield data (boxes per acre) from southwest Florida were used to

calculate an average yield for a given year across all farms, as well as a standard deviation of the

yield from year-to-year within a single farm. The standard deviation of the yearly averages is

used as an 'annual standard deviation' while forecasting yields. This annual standard deviation is

calculated to be 48.6 for Valencia and 63.9 for early-mid.

Within each farm, from year-to-year a deviation from that particular years average yield is

calculated, and an average of those standard deviations is taken as a 'farm level deviation'. This

is done with both Valencia and early-mid data. This farm level deviation for Valencia is 70.3,

and 85.1 for early-mid. The farm level deviations were expected to be much higher than the

annual standard deviation because this annual standard deviation averages the yields across all

farms, lowering the volatility.

Using historical Valencia, early-mid, and all-orange yield data, an average ratio of

Valencia to all-orange, and early-mid to all-orange yields are calculated. Using the average and

normal distribution of these data, combined with the low-level greening forecasted yields found









in Spreen, et al. (2006), stochastic forecasted yields for Valencia and early-mid are derived. A

stochastic normal distribution for historical Valencia and early-mid concentrate yields, are used

with the conversion rate to convert from boxes per acre to pound-solids per acre.

The final yield numbers are calculated using the historical ratio of Valencia to all-orange

yield, combined with the forecasted yields provided by Spreen, et al. By using a stochastic

normal distribution of the Valencia to all-orange ratio, combined with the calculated 'annual

standard deviation' as the standard deviation, we arrive at our forecasted yields across all farms.

To add in the final volatility associated with farm level yield, one more stochastic normal

distribution is added. This normal distribution is calculated with an average of zero, and the

'farm level deviation' as the standard deviation. This last value adds in the farm-level volatility

necessary for this model. This process is repeated for the early-mid yield. Finally, the correlation

matrix based on historical prices and yields, provides the price-yield correlation necessary to

keep the values practical.













Valencias Spot / All
Orange value
Year!Week 10 11 12 13 14 15
1982
1983 1.346 1.346 1.358
1984 1.514 1.1 1.514 1.61 1. 1 1.631
1985 1.349 1.329 1.268 1.268
1986 1.127 1.127 1.090 1.090 1.134 1.163
1987 1.471 1.591 1.621 1.621 1.651
1983 1.256 1.235 1.278 1.278 1.299
1989 1.228 1.163 1.185 1.185 1.336 1.400
1990 1.600 1.600 1.579 1.579 1.538 1.764
1991 1.206 1.180 1.154 1.180 1.180 1.205
1992 1.562 1.590 1.590 1.0 1.50 1.73 1.573
1993 1.037 1.037 1.131 1.179 1.320 1.348
1994 1.752 1.752 1.752 1.673 1.673
1995 1.74 1.7146 1.7E8 1.804 1.77T 1.82R
1996 1.977 2.013 2.049 2.049 2.049 2.013
1997 1.568 1.661 1.568 1.568 1.568 1.568
1998 1.954 1.999 2.176 2.221 2.176 1.954
1999 1.543
20W4 1.577 1.577
2001
2002 1.733 1.705 1.705 1.705 1.686 1.705
2003 2.015 2.015 2.015 2.015 2.015 2.015
2004 1.567 1.567 1.567 1.7 67 1567 1.567
2005 2.357 2.357 2.357 2.357 2.357 2.357
20_,_I ; :2- I -::
n 15 18 19 20 22 23
V Sp.:-l Al Orarnte AVG i i I I .t i i I :-
id' 0 l II :s : sE ':. : z ':. ::1 : 1.
c.:>rrelatin:-r, It _I. 1.: 1 .: 11. '. _41 : I '1 .
ni:-.rrnral .-J,_Ir 1 :6 1.: 1 1 :i:., 1.456 1 411.


Figure 4-1 Valencia Spot Index









CHAPTER 5
RESULTS AND DISCUSSION

Results

The results from the simulations can be seen in Tables 5-1 and 5-2. A probability

distribution function (PDF) graph of the Valencia baseline and scenarios can be seen in Figure 5-

1. A similar PDF displaying the results of the early-mid baseline and scenarios be seen in Figure

5-2.

Valencia

Baseline scenario

The baseline scenarios for the Valencia and early-mid oranges assume the entire crop is

sold through a pooling program. The mean net present value (NPV) for the Valencia baseline

scenario is $845,722, with a standard deviation of 1,032,299. The minimum NPV obtained

during any single iteration is less than $-1.3 million, while the maximum NPV obtained is over

$4.5 million. A graph of this baseline scenario showing the probability distribution function

(PDF) for Valencia can be seen in Figure 5-3. As seen in Table 5-5, the average yearly standard

deviation (the average of the yearly standard deviation's measured from the simulation) for this

scenario is 77,182.

Critical NPV values can be seen in Table 5-3. This table shows that the Valencia baseline

scenario holds a 79 percent chance of producing a NPV of 0 (breakeven) or better. There is a 10

percent chance of losing at least $500,000, and a 4 percent chance of losing at least $750,000. On

the positive side, there is a 44 percent chance of making at least $750,000, and a 15 percent

chance of making at least $1.5 million.









One-week cash market scenario

In the scenario involving the entire harvest being sold into the cash market in a single

week (hereby referred to as 'Scenario 1'), the mean NPV is $835,567 with a standard deviation

of 853,462. The minimum NPV obtained during any single iteration is less than $-1.25 million,

while the maximum NPV obtained is almost $3.5 million. A PDF graph of Scenario 1 for

Valencia can be seen in Figure 5-4. As seen in Table 5-5, the average yearly standard deviation

for this scenario is 74,482.

As seen in Table 5-3, Valencia Scenario 1 holds an 83 percent chance of producing a NPV

of at least 0 (breakeven). This scenario holds a chance of losing $500,000 or more of 13 percent,

and a 3 percent chance of losing $750,000 or more. On the positive side, Scenario 1 has a 53

percent chance of producing a NPV of at least $750,000, and a 22 percent chance of making at

least $1.5 million.

Three-week cash market scenario

In the scenario involving the entire harvest being sold in the cash market the three weeks

following harvest (hereby referred to as 'Scenario 2'), the mean NPV is $800,552, with a

standard deviation of 841,486. The minimum NPV obtained during any single iteration is less

than $-1.25 million, while the maximum NPV obtained is over $3.25 million. A PDF graph of

Scenario 2 for Valencia can be seen in Figure 5-5. As seen in Table 5-5, the average yearly

standard deviation for this scenario is 52,164.

This Valencia Scenario holds an 82 percent chance of producing at least a breakeven NPV

(Table 5.3). This scenario has a chance of losing at least $500,000 of 7 percent, and a 3 percent

chance of losing $750,000 or more. On the positive side, this Valencia scenario holds a 52

percent chance of producing a NPV of at least $750,000, and a 20 percent chance of making at

least $1.5 million.









Hedging scenario

In the scenario involving hedging in the futures market (hereby referred to as 'Scenario

3'), the mean NPV is $547,506, with a standard deviation of 742,264. The minimum NPV

obtained during any single iteration is just under $-1.2 million, while the maximum NPV

obtained is over $2.7 million. A PDF graph of this scenario can be seen in Figure 5-6. The

average yearly standard deviation for this scenario is 65,594 (Table 5-5).

This Valencia scenario holds a 77 percent chance of producing at least a breakeven NPV

(Table 5-3). This scenario has a 10 percent chance of losing at least $500,000, and a 3 percent

chance of losing $750,000 or more. On the positive side, Valencia Scenario 3 holds a 39 percent

chance of producing a NPV of at least $750,000, and a 10 percent chance of making at least $1.5

million. A PDF graph of the change in basis associated with Valencia Scenario 3 can be seen in

Figure 5-7.

Early-mid

Baseline scenario

The baseline scenario for early-mid assumes the entire crop is sold through a pooling

program. The mean NPV for the early-mid baseline scenario is $633,507 with a standard

deviation of 844,480. The minimum NPV obtained during any single iteration is less than $-1.3

million, while the maximum obtained is over $4.5 million. A PDF graph of the early-mid

baseline scenario can be seen in Figure 5-8. The average yearly standard deviation for this

scenario is 65,393 (Table 5-5).

The early-mid baseline scenario holds a 75 percent chance of producing at least a

breakeven NPV (Table 5-4). This scenario has a 10 percent chance of losing $500,000 or more,

and a 4 percent chance of losing $750,000 or more. On the positive side, the early-mid baseline









scenario produced a 44 percent chance of producing a NPV of at least $750,000, and a 15

percent chance of producing a NPV of at least $1.5 million.

One-week cash market scenario

The mean NPV of the early-mid Scenario 1 is $364,364, with a standard deviation of

674,524. The minimum NPV obtained during any single iteration is less than $-1.2 million, while

the maximum is over $2.3 million. A PDF graph of Scenario 1 for early-mids can be seen in

Figure 5-9. The average yearly standard deviation for this scenario is 65,393 (Table 5-5).

The early-mid Scenario 1 has a 70 percent chance of producing at least a breakeven NPV

(Table 5-4). This scenario has a 13 percent chance of producing a NPV of negative $500,000 or

less, and a 4 percent chance of producing a NPV of negative $750,000 or less. On the positive

side, early-mid Scenario 1 has a 28 percent chance of producing a NPV of at least $750,000, and

a 5 percent chance of producing a NPV of at least $1.5 million.

Three-week cash market scenario

The mean NPV of the early-mids Scenario 2 is $369,187, with a standard deviation of

674,213. The minimum NPV obtained during any single iteration is under $-1.2 million, while

the maximum obtained is almost $2.4 million. A PDF graph of Scenario 2 for early-mids can be

seen in Figure 5-10. The average yearly standard deviation for this scenario is 56,993 (Table 5-

5).

This early-mid Scenario 2 holds just a 70 percent chance of producing at least a breakeven

NPV (Table 5-4). This scenario also holds a 13 percent chance of losing at least $500,000, and a

4 percent chance of losing at least $750,000. On the positive side, early-mid Scenario 2 holds a

29 percent chance of producing a NPV of at least $750,000, and a 5 percent chance of making at

least $1.5 million.









Hedging scenario

In the early-mid scenario involving hedging in the futures market (hereby referred to as

'Scenario 3'), the mean NPV is $169,851, with a standard deviation of 761,105. The minimum

NPV obtained during any single iteration is under $-1.5 million, while the maximum NPV

obtained is over $2.7 million. A PDF graph of Scenario 3 can be seen in Figure 5-11. The

average yearly standard deviation for this scenario is 69,199 (Table 5-5).

This early-mids Scenario 3 holds a 55 percent chance of producing at least a breakeven

NPV. This scenario has a chance of losing at least $500,000 of 23 percent, and a 7 percent

chance of losing $750,000 or more. On the positive side, this scenario holds a 22 percent chance

of producing a NPV of at least $750,000, and a 5 percent chance of making at least $1.5 million.

A PDF graph of the change in basis associated with Valencia Scenario 3 can be seen in Figure 5-

12.

Discussion

Valencia

Preliminary expected results would lead us to believe that the baseline scenario with

pooling would provide the lowest volatility, that is, the lowest standard deviation. However,

contrary to this belief, the baseline scenario in fact provided the highest NPV with the highest

level of volatility. Scenario 1 yielded the second highest NPV, with the second-to-least amount

of volatility. Scenario 2 yielded the third highest NPV, with slightly less volatility than scenario

1. Scenario 3, the hedging scenario, in fact provided the lowest NPV by a large margin,

combined with the lowest standard deviation.

While return vs. risk is stated to be determined by NPV vs. standard deviation, the

minimum and maximum potential values are relevant as well. The Valencia baseline scenario

yielded the highest maximum at over $4.5 million, while scenario 1 reached $3.4 million, and









scenario 2 fell about $150,000 short of scenario 1. Scenario 3 had the lowest maximum, falling

short of $3 million. The most favorable minimum was obtained by scenario 3 with negative $1.2

million, followed by scenario 2 at just under negative $1.25 million, while scenario 1 fell about

$3,000 below that, and the baseline scenario fell over $100,000 below that of scenario 1.

Having the highest chance of not being able to break even, at 23 percent, the Valencia

scenario 3 held the second highest chance of losing at least $500,000. Scenario 3 however fell

short only to the Valencia baseline which held a 5 percent chance of losing at least $750,000.

This, however, can be attributed to risk, as this same baseline scenario held the highest chance of

making $1.5 or 2.5 million at 25 percent and 6 percent respectively. This baseline scenario

clearly produces the higher risk complimented by a higher reward. The most risk averse Valencia

grower would be much better off utilizing Scenarios 2 or 3, which produced the highest chances

of at least breaking even (83 and 82 percent, respectively), as well as the highest chances of

making at least $750,000 (53 and 52 percent). Also, these scenarios generated the lowest chances

of losing $500,000 at 3 percent for each. In no categories shown in Table 5-3 did Scenario 1

perform the worst. Scenario 2 held the lowest year-to-year volatility with a standard deviation of

56,993, almost 10,000 lower then the next closest scenario.

Early-mid

Again, we would expect the baseline scenario of pooling to yield the lowest volatility,

and in fact, it again held the highest NPV with the highest level of volatility. As far as early-mid

oranges go, the baseline scenario yielded the most favorable results in other areas as well, having

the both the highest maximum and highest minimum. Scenario 3 yielded the second highest

maximum at over $2.7 million, followed by scenario 2 and lastly scenario 1. The minimums

obtained also favored the baseline scenario, with scenario 2 being the second best option.









The early-mid baseline scenario clearly outperformed the other scenarios for all critical

values. In no category was any other scenario favorable to the baseline (Table 5-4). In fact, many

categories are not even close. The early-mid baseline scenario holds a 15 percent better chance of

producing an NPV of at least $750,000. The next closest scenario (Scenario 2) was 3 times more

likely to produce a NPV of at least $1.5 million than any other scenario. Only the baseline and

scenario 3 showed any chance of reaching $2.5 million, at 2 and 1 percent, respectively. The

early-mid scenario 4 was not competitive in any category, and was the least-favored scenario in

all but a single category, obtaining a NPV of at least $2.5 million.

Changes in basis

A change in basis model, shown in Figure 5-7 and Figure 5-12 could provide some

insight into the detriment of the early-mid scenario 4. After 1,000 iterations, with an average of -

0.0907, this negative change in basis(i.e., weakening) hurts the citrus growers' net returns. The

Valencia change in basis, averaging a positive 0.0039, helps the citrus grower. This

strengthening of the basis helps move the net price higher for the producer.

The early-mid change in basis also produced a much higher variation than that of the

Valencia. While the Valencia iterations provided a positive change in basis 48 percent of the

time, the early-mid change in basis was only positive 30 percent of the time.










Table 5-1 Scenario Results (Valencia)
Scenario V Pool V 1-week V 3-week V Hedge
Mean NPV $845,722 $835,567 $800,552 $547,506
Std Dev 1,032,299 853,462 841,486 742,264
Min $-1,371,947 $-1,264,171 $-1,261,862 $-1,205,965
Max $4,536,464 $3,401,561 $3,251,220 $2,788,583

Table 5-2 Scenario Results (Early-mid)
Scenario EM pool EM 1-week EM 3-week EM Hedge
Mean NPV $633,507 $364,364 $369,187 $169,851
Std Dev 844,480 674,524 674,213 761,105
Min $-1,252,346 $-1,262,639 $-1,261,849 $-1,541,493
Max $3,429,396 $2,376,623 $2,397,002 $2,780,505



Table 5-3 Critical Points (Valencia)
Scenario Baseline Scenario 1 Scenario 2 Scenario 3
NPV of $-750,000 or 5 percent 3 percent 3 percent 3 percent
less
NPV of $-500,000 or 9 percent 7 percent 7 percent 10 percent
less

Breaking even or 79 percent 83 percent 82 percent 77 percent
better
NPV of At Least 51 percent 53 percent 52 percent 39 percent
$750,000
NPV of At Least 25 percent 22 percent 20 percent 10 percent
$1,500,000
NPV of At Least 6 percent 3 percent 3 percent 1 percent
$2,500,000





Table 5-4 Critical Points (Early-mid)
Scenario Baseline Scenario 1 Scenario 2 Scenario 3
NPV of $-750,000 4 percent 4 percent 4 percent 7 percent
or less
NPV of $-500,000 10 percent 13 percent 13 percent 23 percent
or less
Breaking even or 75 percent 70 percent 70 percent 55 percent
better
NPV of At Least 44 percent 28 percent 29 percent 22 percent
$750,000
NPV of At Least 15 percent 5 percent 5 percent 5 percent
$1,500,000
NPV of At Least 2 percent 0 percent 0 percent 1 percent
$2,500,000











Table 5-5 Average Yearly Standard Deviations
Scenario V V V V
Pool Scenario Scenario Scenario
1 2 3


EM
Pool


EM EM EM
Scenario 1 Scenario 2 Scenario 3


Average 77,182 74,482 52,164 65,594 65,393 65,140 56,993 69,199
Yearly
Std
Devs


-2000000 -1000000


0.00 1000000. 2000000. 300000. 4000000. 5000000.


V pool 1-week spot V 3-week V Hedge
Figure 5-1 All Valencia Scenarios




















0.00 1000000.0 2000000.0 3000000.0 4000000.0


-EM Pool -EM 1-week -EM 3-week -EM Hedge
Figure 5-2 All Early-mid Scenarios


-2000000. -1000000.



Figure 5-3 Valencia Pool


0.00 1000000. 2000000.00. 3000 4000000. 5000000.


-V pool


-2000000. -1000000.



















0.00 1000000.0 2000000.0 3000000.0 4000000.0


- 1-week spot


Figure 5-4 Valencia Scenario 1


-2000000. -1000000.


0.00 1000000.0 2000000.0 3000000.0 4000000.0


-V 3-week


Figure 5-5 Valencia Scenario 2


-2000000.0 -1000000.0



















0 00 500000 100000 150000 2DD000 25D000 3DD000


-V Hedge


Figure 5-6 Valencia Scenario 3


PDF Approximation


-0.60


-0.40 -0.20 0.00 0.20 0.40


0.60


-V change in basis

Figure 5-7 Valencia Change in Basis PDF


-150000-100000-5000DDDDD00




















0.00 1000000.0 2000000.0 3000000.0 4000000.0


- EM Pool


Figure 5-8 Early-mid Pool


-2000000.00 -1000000.00


Figure 5-9 Early-mid Scenario 1


0.00 1000000.00 2000000.00


3000000.00


- EM 1 -week


-2000000. -1000000.





















-2000000.00 -1000000.00


Figure 5-10 Early-mid Scenario 2


0.00 1000000.00 2000000.00


3000000.00


- EM 3-week


-2000000.00 -1000000.00


0.00


1000000.00 2000000.00 3000000.00


-EM Hedge


Figure 5-11 Early-mid Scenario 3









PDF Approximation


-1.00 -0.80 -0.60 -0.40 -0.20 0.00 0.20 0.40 0.60

-EM change in basis


Figure 5-12 Early-mid Change in Basis









CHAPTER 6
SUMMARY AND CONCLUSIONS

Summary

With the citrus industry being the largest agricultural sub-sector in the state of Florida at

over 719,000 acres in use, the state of Florida accounts for over 67 percent of the total U.S.

production for oranges. Over 95 percent of the oranges grown in Florida are grown for

processing, and over 90 percent of that comes from two types of citrus, Valencia and early,

midseason oranges. Volatility in the citrus industry caused by weather, disease, and other factors

causes concern to citrus growers whose goal is to secure profitable returns. With various

marketing strategies available to growers, an impartial examination of the various strategies and

forecasted outcomes provides citrus growers with the tools to further their knowledge on

available marketing strategies including their forecasted returns and the risks associated with

those strategies.

This study examines the risks and returns associated with 4 scenarios for each of the two

types of citrus. The baseline scenario is the pooling method, in which growers pool their crops

and receive payment based on the average price received for the entire pooled crop. This

scenario is an estimate of the average cash market price received in a single harvest season. The

first alternative scenario compared to this baseline assumes the producer sells the entire crop into

the cash market at once. The second scenario assumes the grower sells the entire crop into the

cash market over the course of three weeks. The final scenario assumes that the producer hedges

the crop utilizing the FCOJ contract in the futures market when the previous harvest season ends.

Simetar@ (Simulation for Excel to Analyze Risk), an Excel add-in that was developed

explicitly for stochastic simulation modeling was used to create stochastic forecasts for price and

yield, utilizing observed probabilities. Using the stochastic forecasted prices along with









stochastic forecasted yields in conjunction with a pro-forma financial statement model, we are

able to view the net present value (NPV) of discounted revenues through the year 2020. By

running a large number of iterations, we can see the expected return (NPV) and volatility

associated with the various marketing methods.

The baseline scenario of pooling for Valencia oranges provided the highest volatility (as

measured by standard deviation) along with the highest NPV. The one-week cash market

scenario (scenario 1) yielded the second highest NPV, with the least amount of volatility. Selling

the crop into three consecutive cash market weeks (scenario 2) yielded the second lowest NPV,

with slightly more volatility than scenario 1. Scenario 3, the hedging scenario, provided the

lowest NPV by a large margin, and the highest volatility.

For early-mid oranges, the baseline was clearly the most successful scenario, providing the

highest NPV combined with the lowest volatility. This baseline scenario also yielded the highest

maximum NPV obtained for early-mid oranges and the most favorable minimum as well.

Conclusions

Florida citrus growers are challenged every year to choose a marketing strategy that will be

the most favorable to them. As opposed to other commodities such as corn and wheat, the

physical area that is designated to growing citrus is very concentrated. This high concentration

causes much higher price volatility than is observed for most other commodities, in turn causing

citrus growers to spend large amounts of time and energy looking for marketing strategies that

can benefit them. This study provides growers information about various marketing alternatives

available to them, as well as an estimate on the forecasted risks and rewards associated with

these various methods.

For the Valencia grower, pooling is expected to provide the highest net present value along

with the highest volatility. This is the essence of a high risk to reward relationship. Scenario 1,









selling directly in the cash market in a single week, yielded the second highest net present value

with the lowest volatility. Despite contrary preliminary expectations, this scenario provided a

great lower risk alternative to the pooling strategy. The final scenario, hedging, proved to be an

unfavorable strategy, providing high volatility with a low net present value.

For the early, midseason grower, the baseline scenario of pooling provided the most favorable

alternative in almost every way. This scenario outperformed the three alternatives with a higher

net present value, a lower volatility, a higher maximum, and the most favorable minimum. The

low risk and high reward of the baseline scenario makes pooling a clear choice for the early-mid

citrus grower.

Implications

Marketing strategies continue to evolve for citrus growers. This study assumes that a citrus

grower will utilize a single marketing strategy for 100 percent of their crop. The information for

this study can be used as a base even though some citrus grower will utilize multiple strategies,

each regarding portions of their harvested crop. There are also other issues in the analysis that

can influence the results. For example, a higher discount rate will result in lower net present

values and vice versa. With various risk adversity levels of a particular farmer calculated, the

ability to utilize multiple strategies for various amounts of their harvested crop is available.

Recent hurricane events in Florida as well as the presence of greening, canker, tristeza, and

urban development of farm land serve as reminders that the production risk associated with

Florida citrus growers is valid and must be addressed. The financial implications including

potential losses due to these factors indicate great loss potential for citrus growers who do not

utilize some type of risk management. For instance, a farmer that takes a short position in the

futures market to offset the downside price risk they face from plant to harvest, is at risk to be

hurt twice as hard if a hurricane takes out his crop. A hurricane can cause a farmer to lose his









crop and to lose on a short position in a futures market that could see dramatic increases in price

following the hurricane. A season of losses at the level brought upon by these two consequences

of one event could put many farmers out of business.

Further Research Needs

There are many areas of this study that are left open for further research. The availability

of utilizing a combination of strategies can potentially add another level of practicality on this

study. An econometric study could provide great insight into the optimized combinations of

these marketing strategies for citrus growers with various risk adversity levels.

A study including the use of options along with futures contracts for a hedging could also

yield interesting results that would be useful for the citrus grower. However, the 'random walk'

of futures prices renders great problems in the hedging scenario. Market theory would suggest

that if futures prices were able to be forecast with any accuracy, the speculator arbitrage of the

information would in fact distort the futures price away from the predicted price and render the

forecasted price useless.









LIST OF REFERENCES

Andre, M. "The U.S. Citrus Growers: Competitiveness and Performance." M.S. thesis.
University of Florida, May 1996.

Behr, R. "A Simultaneous Equation Model of Futures Market Trading Activity." Ph.D.
dissertation. University of Florida, June 1981.

Brown, M., et al. "Florida Citrus Production Trends." Economic and Market Research
Department Report. Gainesville, Florida: 2007.

Citrus Administrative Committee. Utilization ofFlorida Citrus Fruit Reports. Internet Site:
(http://www.citrusadministrativecommittee.org/). No copyright, Fruit and Vegetable
Division. Lakeland, Florida. (Accessed February 21, 2008).

Crist, R. "The Citrus Industry in Florida." American Journal of Economics and Sociology
1(1955):1-12.

Florida Agriculture. Overview ofFlorida Agriculture. Internet Site:
(http://www.florida-agriculture.com/). Copyright 2004-2007, Florida Department of
Agriculture and Consumer Services. Tallahassee, Florida. (Accessed January 15, 2008).

Florida Department of Citrus, "Citrus Reference Book." Economic and Market Research
Department. May 2007.

Guci, L., and M. Brown. "Changes in the Structure of the Florida Processed Orange Industry and
Potential Impacts on Competition." Florida Department of Citrus, Gainesville. FL. Staff Report.
2007.

Hodges, A., et al. "Economic Impact of Florida's Citrus Industry." Economic Information Report
1999-2000. Gainesville, Florida: 2001.

Hull, J. Fundamentals of Futures and Options Markets. Upper Saddle River, NJ: Pearson, 2002.

Malick, W., and R. Ward. "Stock Effects and Seasonality in the FCOJ Futures Basis." The
Journal of Futures Markets 2(1987): 157-167.

Muraro, R., and W.C. Oswalt. "Budgeting Costs and Returns for Central Florida Citrus
Production." IFAS Report 2002-2003. Gainesville, Florida: 2003.

New York Board of Trade. "Orange Juice Statistics. Nearby Futures Contract Prices, 1982 -
2007." 2007.

Ranney, J. "Pricing Efficiency of Options on FCOJ Futures Contracts." M.S. thesis. University
of Florida, August 1993.

Roll, R. "Orange Juice and Weather." The American Economic Review 74(1984):861-80.










Salnars, C. "Evaluation of the Effect of Captive Supply and Market Liquidity on Basis Levels as
Contracts Approach Maturity in the United States Live Cattle Futures Markets." M.S.
thesis. University of Florida, May 2004.

Scora, R. "On the History and Origin of Citrus." Bulletin of the Torrey Botanical Club
6(1975):369-375.

Spreen, T., et al. "An Economic Assessment of the Future Prospects for the Florida Citrus
Industry." IFAS Report. Gainesville, Florida: 2006.

Spreen, T., M. Brown, and R. Muraro. "The Projected Impact of Citrus Greening in Sao Paulo
and Florida on Processed Orange Production and Price." Economic Research Department
Report. Gainesville, Florida 2005.

USDA's National Agricultural Statistics Service, Florida Field Office (NASS). Florida State
Agricultural Overview 2006. Internet Site: (http://www.nass.usda.gov/). Orlando,
Florida (Accessed January 15, 2008).

USDA's National Agricultural Statistics Service, Florida Field Office (NASS). Florida Citrus
Summaries 1982 2006. Internet Site: (http://www.nass.usda.gov/). Orlando, Florida.
(Accessed January 15, 2008).

Ward, R. "Market Liquidity in the FCOJ Futures Market." American Journal ofAgricultural
Economics 1(1974):150-154.

Ward, R., and J. Niles. "Hedging Strategies in FCOJ Futures." Economic Research Department
Report. Gainesville, Florida: 1975.

Ward, R., and F. Dasse. "Empirical Contributions to Basis Theory: The Case of Citrus Futures."
American Journal ofAgricultural Economics 1(1977):71-80.

Ward, R., and R. Behr. "Futures Trading Liquidity: An Application of a Futures Trading Model."
The Journal of Futures Markets 3(1983a):207-224.

Ward, R., and R. Behr. "Allocating Nonreported Futures Commitments." The Journal of Futures
Markets 4(1983b):393-401.

Weldon, R., R. Hinson, J. VanSickle, and R. Muraro. "The Economic Impact of the 2004 Citrus
Disaster Program on Florida Citrus Producers." IFAS Report. Gainesville, Florida: 2004.









BIOGRAPHICAL SKETCH

Evan Marc Shinbaum, the son of Amy and Kyle Shinbaum, was born in Fort Myers, FL,

on March 9th, 1983. After graduating from Fort Myers High School in 2002, he spent time

studying at Florida International University in Miami before transferring to the University of

Florida. After completing the Bachelor of Science (B.S.) degree in food and resource economics,

he entered the graduate program in the same field to pursue the Master of Science (M.S.) degree.





PAGE 1

1 EVALUATING RISKS AND RETURNS A SSOCIATED WITH ALTERNATIVE MARKETING STRATEGIES FOR PROCESSED CITRUS By EVAN MARC SHINBAUM A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2008

PAGE 2

2 2008 Evan Marc Shinbaum

PAGE 3

3 ACKNOWLEDGMENTS I express m y sincere thanks and appreciation to Dr. John VanSickle, the chairman of my committee. He has provided me with a great opp ortunity in graduate school that I would otherwise be unable to financially afford, as well as much of his time and effort during my preparation of this th esis. I would also like to thank Dr Richard Weldon for his efforts and inclusion within my committee. I would also like to show my appreciation for the instructors in the department that have helped build the foundation of knowledge needed for this jour ney, including Richard Kilmer, Ronald Ward, Lisa House, John VanSickle, Richard Weldon, James Sterns, and Evan Drummond. A recognition is needed also for Ron Muraro, for his assistance to me in acquiring necessary data for my analysis. A particular thanks is needed for Jennifer Clark who donated much time and effort in assisting me throughout th e thesis process, and extraordinary help with my data analysis. I wish to also thank my parents for their support in helping me get through tough times and reach this incredible goal.

PAGE 4

4 TABLE OF CONTENTS page ACKNOWLEDGMENTS ............................................................................................................... 3LIST OF TABLES ...........................................................................................................................6LIST OF FIGURES .........................................................................................................................7ABSTRACT ...................................................................................................................... ...............8 CHAP TERS 1 PROBLEM STATEMENT AND OBJECTIVE ..................................................................... 10Problem Statement ............................................................................................................. .....10Objective ..................................................................................................................... ............13Organization of Study .............................................................................................................142 BACKGROUND AND PR OBLEM SETTING ..................................................................... 15Citrus Industry ........................................................................................................................15History .............................................................................................................................15Citrus Industry in Florida ................................................................................................15FCOJ Industry .........................................................................................................................16Futures Markets ......................................................................................................................16Options ....................................................................................................................... .............18Risk Management ...................................................................................................................19Basis ......................................................................................................................... ...............20Basis in the Citrus Industry .....................................................................................................21Hedging Examples .............................................................................................................. ....213 REVIEW OF THE LITERATURE ........................................................................................274 METHODOLOGY AND DATA ........................................................................................... 37The Model ...............................................................................................................................37Forecasting Market Prices ..................................................................................................... .39Forecasting Futures Prices for the Hedge Model ............................................................ 43Forecasted Yields ............................................................................................................ 445 RESULTS AND DISCUSSION ............................................................................................. 47Results .....................................................................................................................................47Valencia ...................................................................................................................... .....47Baseline scenario ...................................................................................................... 47One-week cash market scenario ............................................................................... 48

PAGE 5

5 Three-week cash market scenario ............................................................................ 48Hedging scenario ......................................................................................................49Early-mid ..................................................................................................................... ....49Baseline scenario ...................................................................................................... 49One-week cash market scenario ............................................................................... 50Three-week cash market scenario ............................................................................ 50Hedging scenario ......................................................................................................51Discussion .................................................................................................................... ...........51Valencia ...................................................................................................................... .....51Early-mid ..................................................................................................................... ....52Changes in basis .............................................................................................................. 536 SUMMARY AND CONCLUSIONS .....................................................................................62Summary ....................................................................................................................... ..........62Conclusions .............................................................................................................................63Implications .................................................................................................................. ...64Further Research Needs ...................................................................................................65LIST OF REFERENCES ...............................................................................................................66BIOGRAPHICAL SKETCH .........................................................................................................68

PAGE 6

6 LIST OF TABLES Table page 5-1 Scenario Results (Valencia) .....................................................................................................545-2 Scenario Results (Early-mid) .............................................................................................. .....545-3 Critical Points (Valencia) ................................................................................................ ........545-4 Critical Points (Early-mid) ............................................................................................... .......545-5 Average Yearly Standard Deviations ......................................................................................55

PAGE 7

7 LIST OF FIGURES Figure page 2-1 FCOJ Contract Information .....................................................................................................242-2 Hedging Outcomes w ith Increasing Prices ..............................................................................252-3 Hedging Outcomes w ith Decreasing Prices ............................................................................252-4 Hedging Outcomes with Strengthening Basis ......................................................................... 262-5 Hedging Outcomes with Weakening Basis ............................................................................. 263-1 Seasonality of Basis Residuals ............................................................................................ ....353-2 Basis Residuals over Time ................................................................................................. .....364-1 Valencia Spot Index ....................................................................................................... ..........465-1 All Valencia Scenarios .................................................................................................... ........555-2 All Early-mid Scenarios ..........................................................................................................565-3 Valencia Pool ...........................................................................................................................565-4 Valencia Scenario 1 ....................................................................................................... ..........575-5 Valencia Scenario 2 ....................................................................................................... ..........575-6 Valencia Scenario 3 ....................................................................................................... ..........585-7 Valencia Change in Basis PDF .............................................................................................. ..585-8 Early-mid Pool .........................................................................................................................595-9 Early-mid Scenario 1 ...................................................................................................... .........595-10 Early-mid Scenario 2 .............................................................................................................605-11 Early-mid Scenario 3 .............................................................................................................605-12 Early-mid Change in Basis ................................................................................................ ....61

PAGE 8

8 Abstract of Thesis Presen ted to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science EVALUATING RISKS AND RETURNS A SSOCIATED WITH ALTERNATIVE MARKETING STRATEGIES FOR PROCESSED CITRUS By Evan Shinbaum May 2008 Chair: John VanSickle Major: Food and Resource Economics The citrus industry is the larg est agricultural subsector in Florida with 719,000 acres in use. Over 90 percent of the oranges produced in Florida are Valencia and early, midseason oranges grown primarily for processing. Volatil ity in the citrus industry caused by weather, disease, and other factors is a concern for citrus growers. Alternative marketing strategies are available to growers to manage this risk. This study provides an examination of alternative marketing strategies and forecasts potential outco mes. The results provide citrus growers with knowledge about these strategies and th eir forecasted retu rns and risks. This study examines the risks and returns associ ated with 4 scenarios for each of the two major types of citrus used in processing: 1) the pooling method, in which growers pool their crops and receive payment based on the average pri ce received for the enti re pooled crop; 2) the 1-week cash market, which assumes the producer sells his/her entire crop into the cash market in one week; 3) the 3-week cash market, which assume s the grower sells his/he r entire crop into the cash market over the course of three weeks; and 4) hedging, which assumes that the producer hedges his/her crop utilizing the FC OJ contract in the futures mark et when the previous seasons crop is ending.

PAGE 9

9 Forecasted spot market and pooling prices ar e obtained using the forecasted all-orange prices from a recent study. Simetar (Simulation for Excel to Analyze Risk), an Excel add-in that was developed explicitly for stochastic si mulation modeling was used to create stochastic forecasts utilizing a pseudo-random number gene rator and observed probabilities. Using the stochastic forecasted prices along with stochastic forecasted yields in conjunction with a proforma financial statement model, the net presen t value (NPV) of discount ed revenues through the year 2020 is determined. Pooling provided the highest standard devia tion as well as the highe st NPV for Valencia. The one-week cash market scenario yielded th e second highest NPV. The hedging scenario provided the lowest NPV along with the lowest standard devia tion. For early, midseason, the pooling baseline provided the highest NPV combined with the highest standard deviation, that is, the highest return with the highest risk.

PAGE 10

10 CHAPTER 1 PROBLEM STATEMENT AND OBJECTIVE Problem Statement The citrus in dustry is the largest agricultural su b-sector in the state of Florida, with more land being used to grow citrus than any ot her state in the country. With 719,000 acres used, citrus accounts for over 40 percent of all land us ed to farm crops in the state (NASS). In 2005, Florida accounted for 67 percent of the total U. S. production for oranges at roughly $843 million (Florida Agriculture). Out of that, over 95 percent of all oranges grown in Florida are processed. The three major types of oranges grown in Florida include the early-mids, Valencias, and naval oranges. As implied by the name, the early -mids are harvested earlier in the crop season. The Valencia orange, also known as Murcia orange, is a late season fruit. Valencia oranges have zero to six seeds per fruit, however the excellent taste and internal color make it desirable for the fresh market as well as for processing juice. Naval oranges, also known as the Washington, Riverside, or Bahie naval, devel op a second orange at th e base of the original fruit, opposite the stem. Because these are seedless oranges, the only means available to cult ivate more is to graft cuttings onto other vari eties of citrus trees. Of the 140 million boxes of citrus that were utilized through processing in the 2005-2006 season, early-mids and Valencias accounted for over 95 percent. While over 93 percent of all early-mids, and over 96 percent of all Valencias were processed, le ss than one-third of all naval oranges were processed. Early-mids and Valenc ias accounted for over 97 percent of all citrus grown in the state (NASS). With ever increasing volatility in the citrus industry, citrus growers are faced with crucial decisions regarding the marketing strategies they use for their products. Supply is constantly affected by threats of weather such as hurricanes and freezes, diseases such as canker, greening,

PAGE 11

11 and tristeza, and even urban development (Brown ).While the southern movement of the citrus industry in Florida has helped decrease the exposure to freeze da mage, the threat of hurricanes comes every year. Because of these supply s hocks, volatility is likely to remain high. The futures market at the ICE (Intercontinen tal Commodity Exchange, formerly known as the New York Board of Trade) in New York C ity provides a common met hod of pricing frozen concentrated orange juice (FCOJ). While the ma rket presents processors and growers with a place to buy and sell FCOJ, only a small amount of actual FCOJ is sold in these markets. The markets much more common purpose is to be used as a hedging and price discovery tool for both processors and growers while speculators provide the necessary liquidity while hoping to capitalize on price changes. With the FCOJ futu res market, citrus grower s can lock in a floor price for selling their product. This enables them to protect themselves against an adverse price change. While this elimination of risk with respect to adverse price movements can be helpful, the ability to take advantage of favorable pr ice movements is greatly diminished in hedging programs (less so with the use of options as opposed to direct futures positions). There is considerable price risk in FCOJ as evidenced in the futures markets. The seasonal coefficient of variation of the average weekly cl oses for the past 5 seasons (2002-2007) averaged over 12 percent. Also, volatility can be higher in the FCOJ mark et because the market for the FCOJ contract is not as liquid as others such as corn. A large number of FC OJ contracts traded at any time can cause a movement in price. There are many other factors th at can cause high volatility within the FCOJ futures markets. Hurricane season serves as a great exam ple of this. Because of the uncertainty caused by hurricane seasons, volatility within the FCOJ can increase before a hurricane affects it. As the hurricane season begins, the price in the market will generally increase, as more and more

PAGE 12

12 speculators take long positions to prepare to capita lize on any formed storms that can damage the orange crop. As storms develop and trend toward or away from the state of Florida, prices fluctuate dramatically as information on the storms evolve. This is a grea t example of a situation in which the market price reflects potential fo r occurrences rather than current supply and demand. Diseases pose an especially large risk to citrus growers because they can impact the production of a grown crop for several seasons. There are certain producti on practices that are used by growers to reduce the risk or uncertainty of these diseases. Tristeza, also known as CTV (citrus tristeza virus), is a vira l species that causes the most economically damaging disease to oranges. It has accounted for the death of millions of citrus trees, and was coined the term tristeza from farmers in Brazil and South Ameri ca, as it is Spanish and Portuguese for sadness. Citrus canker is an infection that causes lesions on the leaves, stems, and fruit of the trees. This causes leaves and fruit to drop prematurely. While these fruit are still eatab le, they are generally blemished and harder to sell to consumers. The other main disease is greening, which is distinguished by the common symptoms of yellow ing of the veins and adjacent tissues, followed by yellowing or mottling of the entire leaf and premature defoliation. The trees will have stunted growth, bear multiple off-season flowers, and pr oduce small, irregularly-shaped fruit and the disease may often lead to the death and replacement of trees. Perhaps the most common, easy to use, and easy to understand strategy of marketing is the use of pools. Pooling is simply a process wher e the proceeds from many sales of a particular commodity are averaged and growers all receive the average price after costs have been deducted. Pooling still has some volatility as well, though not as much. With early-mids, over the past 15 seasons the mean return for pooling wa s $1.01 with a standard deviation of 26 cents,

PAGE 13

13 while the mean return for Valencias was $1.157 with a standard deviation of over 22 cents. Both pooling and hedging in the futures market are risk management st rategies, as opposed to growers selling straight in the cash market at the current price at the time of harvest. The option of selling directly into the cash mark et is generally perceived to hold the largest risk for growers, and the most volatility. Sellin g directly into the cash market means that you accept the current price at the time of harvest, which can differ significantly from the price at the beginning of the production season. This makes budgeting more difficult, but does allow the farmer to capitalize on the market when the pric e rises during the producti on season. In situations where disease or weather hurts the crop, farmers w ho still hold their product are in great shape to capitalize on the favorable increas ed price that results from th e decreased supply. Many farmers take out crop insurance to protect themselves fr om production risk that can result in a damaged crop. One final method of price risk management is the use of forward contracts. A forward contract is an agreement between two parties to buy or sell an asset (in th is case citrus) at a preagreed future point in time. The trade date and delivery date are th erefore separated. Most forward contracts dont have stan dards and arent traded on exchanges. A farmer would use a forward contract to lock-in a price fo r his oranges for the upcoming harvest. Objective The general objective of this study is to enhance the citrus growers knowledge of m arketing strategies available, a nd help to provide information re garding the differences in risks and returns associated with the al ternative strategies. The specific goal of this study is to create a model which will demonstrate the forecasted si tuations, allowing a prediction of the best marketing strategy available, unde r a given set of assumptions.

PAGE 14

14 It is the goal of this study to become a tool for citrus growers as well as pool operators to generate new ideas for alternative marketing strategies. It should be of interest to all growers to be informed of, and understand the options av ailable to them regarding their marketing strategies, and be aware of the risks associated with them. Organization of Study The following chapter will giv e a background on the citrus industry and FCOJ, the futures and options market, risk management, and basic hedging strategies. Chapter 3 reviews literature related to the citrus industry and hedging. Chapter 4 presents the methodology, including data collection, the model used, and analysis. A discu ssion of the results is presented in Chapter 5, before summarizing and drawing conc lusions in the final chapter.

PAGE 15

15 CHAPTER 2 BACKGROUND AND PR OBLEM SETTING Citrus Industry History The oldest reference to citrus com es from Sanskrit literature. Jambhila is the name applied to citron and lemon in the White Yahir-veda, specifically in the Vajasaneyi Samhita, part of the Brahmin sacred book which dates prior to 800 B.C. The Citron was sanctified in India, and consecrated to Ganesh, the God of wisdom a nd knowledge (Scora). In non-rabbinic Jewish Tradition, a citron in the house keeps the Karines (bad spirits) away, while the Romans used citrus as a basic body oil. Citron was the first of all citrus to reach the west and also the first citrus fruit to grab European attention. This was all mostly due to its relative imperishability combined with pleasant odor and appearance that made it suitable for long travel. In Medieval times, visiting dignitaries would be entitled to a certain amount of sweet orange slices that were detailed in cookbooks. As an expensive item, c itrus became a fashion of the rich merchants (Scora). Citrus Industry in Florida Citrus has been produced commercially in Fl orida since the m id-1800s. The environment of Florida provides a comparative advantage for citrus production due to natural resources such as the subtropical climate (Hodges) Also, the abundance of rainfall and water causes the soil to be light and infertile, so freque nt and heavy fertilization is necessary. This helps the quality of oranges to be very good (Crist). Several freezes in north central Flor ida in the 1980s caused crop loss, permanent tree damage, and large pri ce fluctuations (Ranney). The industry responded by moving into the southern region of the state (Andre). Florida citrus production followed an upward trend from 1990 to 2000, increasing 46 percent; oranges and grap efruit being the top

PAGE 16

16 commodities (Hodges). The Florida citrus industry contains much more processed oranges than California, the second largest citr us producing state, mostly due to the better taste of the Florida oranges, while the better look and texture of Californias oranges cau se them to be used as fresh fruit (Andre). In fact, 94 percent of Flor idas round-oranges are processed (NASS). FCOJ Industry Research that led to the developm ent of Frozen Concentrated Orange Juice (FCOJ) in 1947 allowed Florida to expand orange production. Florida produced 57.5 million boxes of oranges in 1947 with as much as 76 percent of the round-orange crop in the state processed. In 1948, there were only three processors, but by 1952 there were eighteen. The FC OJ contract was created in 1966, trading under the newly formed Citrus Associ ates of the New York Cotton Exchange. By 1980, the states record crop of over 206 million boxes of round oranges was produced, and over 83 percent of that crop was pr ocessed into FCOJ (NASS). FCOJ stocks vary seasonally, with inventor ies approaching their maximum levels in the latter months of the harvest season. Freeze bias refers to the bidding up of futures relative to spot prices prior to the end of a freeze / hurricane risk period due to the potential for speculative windfall gains that could be rea lized if a freeze causes considerable fruit loss. Large swings in FCOJ prices caused by freeze and hurricane damage to trees are unique among commodities traded in the futures markets (Malick). Futures Markets Futures m arkets are organized exchanges of derivative markets where many buyers and sellers meet to trade futures contracts on an ever-expanding list of commodities (Salnars). Futures markets can be traced back as early as the Middle Ages, when markets were developed to meet the needs of farmers and merchants. The first futures-type contract was known as a toarrive contract The Chicago Board of Trade (CBOT) wa s established in 1848 to bring farmers

PAGE 17

17 and merchants together, and in 1874, the Chicag o Produce Exchange was established, beginning with markets for butter, eggs, poultry and othe r perishable agricultu ral products (Hull). The futures contract helped provide a way for both the producing farmer and the purchasing company to eliminate the risk it faces due to uncertainty of future prices. The main task in originally developing the markets was standardizing the qualitie s and quantities of the grains being traded (Hull). The futures market can be used by citrus growers to manage price risk while making sure they have somebody to purchase their product, while the processors can use it to manage price risk and make sure they have the ability to get the product. Speculators use various strategies in the markets in an atte mpt to make profit off of price movements. Speculative traders provide the necessary level of trading activity or liquidity in the market to prevent the occurrence of added ri sk from failure to establish or terminate a contract (Ward, 1974). The final function of the futures markets is price discovery. Futures prices change as buyers and sellers judgments about a commodities worth at a particular point changes. These judgments are subject to supply and demand deve lopments, and the ava ilability of current information. This continuous process is known as price discovery (Salnars). Traditional futures contracts have been trad ed by what is known as the open-outcry system, in which the traders physically meet on th e floor and indicate the trades they would like to carry out using a complicated set of hand si gnals and open outcry. This is still used during regular trading hours, however in recent years, electronic trading has grown largely as an alternative (Hull). There are certain terminologies that go along with the f utures language. To go long on a futures contract is to buy the contra ct that would allow taking delivery of the specified commodity; to go short on a futures contract is to sell the side of the contract that

PAGE 18

18 would allow making delivery of the specified comm odity. In its simplest terms, to go long is to be the buyer of a commodity and to go shor t is to be the seller of the commodity. Options In addition to f utures contracts, option contracts are offered for many commodities, giving the holder (buyer) the right, but not the obligation to take a pos ition with a futures contract. There are two types of options, the call option and the put option. A call option gives the holder the right to buy a futures contract by a certain date for a certai n price, while a put option gives the holder the right to sell a futu res contract by a certain date fo r a certain price. This certain price is referred to as the strike price. The buyer of an option (frequently referred to as the holder) is long on that option, whil e the seller (frequently referred to as the writer) is short the option. An American option can be exercised at any point during its life, while a European option can only be exercised on its maturity date (Hull). The first trading in puts and calls began in Europe and the United States as early as the eighteenth century; however corrupt practices gave the market a bad name in its early years. The Put and Call Brokers and Dealers Association was set up by a group of firms in the early 1900s, and in April of 1973, the Chicago Board of Trade set up the Chicago Board of Options Exchange (CBOE) which is the largest exchange in the world for trading stock options (Hull). Options first appeared in America in the early 19th century, and were known as privileges. In 1934, the Investment Act legitimized options, and put it under the eye of the r ecently formed Securities and Exchange Commission. On April 26, 1973, the Chicago Board of Options Exchange began trading listed call options. By the end of 1974, average daily volume exceeded 200,000. The FCOJ options became available in the New Yo rk Board of Trade (now the ICE) in 1985.

PAGE 19

19 Risk Management There are various ways of m anaging risk for the citrus grower. Pooling, perhaps the most common, involves gathering the citrus from seve ral growers and paying the growers the average sale price for the product in th e specific pool. There are also forw ard contracts, which are similar to a futures contract in that it is an agreement to buy or sell produc t at a certain time in the future for a certain price. However, while futures cont racts are traded on exchanges, forward contracts trade over-the-counter (Hull). This means that forward contracts are cr eated by two parties who agree on a delivery date. These are not set quantity and quality amounts traded like they are on the futures exchange. Hedging is a risk management tool that gene rally uses the futures and options markets. When a citrus grower takes a short position in th e futures market equivalent to the quantity of citrus they are producing, they are lessening thei r price risk. With a short position in the futures market, the grower is putting themselves in a position to offset losses in their sale value of citrus if the price drops with a positive gain in the future s market. If the price of citrus rises, the loss in the futures market is offset by the higher pri ce for which they can sell their citrus. The transaction costs become the only truly lost value, which is generally small relative to the price risk associated with the commodity. There are multiple transfers of risk taking place in this type of risk management. The transfer from price risk to basi s risk for a hedger, and the accept ance of price risk by speculators provides the necessary liquidity for the market to work. Using this method of risk management, the basis risk becomes the main point of intere st for hedging. Understandi ng basis and being able to forecast it is the key issue in hedging (Salnars).

PAGE 20

20 Basis Basis refers to the d ifference between the pri ce of a given commodity in the nearby futures contract and the price that same commodity could be bought or sold for in the spot market (also known as cash market). The equation used is: basis equals cash price minus futures price. B = CP FP (2-1) Where B represents basis, FP represents futu res price, and CP re presents cash price In reality, basis is the connec ting link between the present an d future (Malick). Cash and futures prices generally move together and r eact to continuous changes in supply and demand factors (Salnars). Basis residuals are a calculate d deviation from the aver age basis (Malick), and can be represented as BR = BM ABS (2-2) Where BR represents the basis residual, BM represents the basis for a specific month, and ABS represents the average basis for a season. Ba sis residuals refer to the difference between expected basis and actual basis. Basis can change based on many factors. An increase in basis is referred to as a strengthening basis, when the s pot price increases by more than the futures price. In a normal market where the futures price is higher than th e relative cash price due to carrying costs (such as time, place, and quality) a strengthening basis is a narrowing basis. A weakening basis is just the opposite, in which the futures and cash prices diverge in a normal market (Hull). The hedging term used for someone that is planning on sell ing goods is a short hedge, because their initial position taken in the futures market is a short position. A long hedge is used for managing the risk when purchasing a commodity, and is acco mplished by first taking a long position in the futures market.

PAGE 21

21 Basis in the Citrus Industry Basis levels in general can be affected by product quality, transportation, location, insurance, tim e, storage, delivery method, or any combination thereof (Salnars). Basis risk in FCOJ is hypothesized to be affected by six ma jor variables: a measure of risk premium, convenience yield, market liquidit y, freeze / hurricane bias, bias adjustments, and an actual freeze / hurricane event (Ward, 1977). Basis risk d ecreases as the time to the hedge expiration (delivery month) decreases. If the asset to be hedged and the asset underl ying the futures contract are the same, the basis should be zero at the expiration of th e futures contract (Hull). If nearby prices (the price of the nearest futures contract traded on the exchange at any given time) reflect a current shor tage in cash product, some convenience yield exists for having at least a minimal stock held for future consum ption. This yield offsets at least part of the carrying cost of holding the stocks in inve ntory (Ward, 1977). Lower stocks help push cash prices up relative to futures encouraging stock holders to draw down existing inventories (Malick). Hedging Examples The benefit of i mplementing a short hedge for a producer can be demonstrated as shown in figure 2-2. The basis in figure 2-2 remains consta nt over the period of holding the hedge to show the general objectives of hedging with futures. The current (June) spot market price, pe r pound solid FCOJ is $2.05 in figure 2-2. The current futures price is $2.25. The basis theref ore is -0.20. A grower who plans to sell 75,000 pounds worth of FCOJ in September, would need to take a short position with 5 contracts (each contract contains 15,000 pounds) to offset any losse s in the spot market between the time the hedge is implemented and the time the FCOJ is sold in the cash market (which should be about the same time as the contract maturity date).

PAGE 22

22 In Figure 2-2, the prices in the spot and futures markets rise to $2.30 and $2.50, respectfully, by September. While th ere has been a 25 cents increase in the spot price, this gain is offset by a loss of 25 cents in the futures posi tion held by the hedger af ter liquidation. Therefore the realized price that the grower receives for his crop remains the original $2.05 due to a lack of change in the basis ($2.30 for the cash market combin ed with a 25 cents loss in the futures contract). At 75,000 pounds of goods, the total sale value is $153,750 (note: this simple scenario neglects transaction costs). In a second scenario (shown in Figure 2-3), th e prices in the spot and futures markets fall to $1.65 and $1.85, respectfully, by September. In this situation, the 40 cents per pound solid lost in the spot market is offset by an equal gain in the short futures posi tion after liquidation. Again, the realized price for the grower would remain equal to the original spot price of $2.05, due to no change in the basis (igno ring transaction costs). The above examples hold basis constant and di sregard transaction costs, neither of which generally occurs in the market. Figure 2-4 dem onstrates the outcome of a hedging transaction when basis strengthens. Again, the current s pot market price is assumed to be $2.05 per pound solid when the hedge is implemented, and the fu tures price is assumed to be $2.25. The current basis in this scenario therefore is -0.20. A grower who plans to sell 75,000 pounds worth of FCOJ is long 75,000 pounds of FCOJ in the cash market. He would take a short position with 5 contracts to manage the price risk held with his current cash position. In this third scenario shown in Figure 2-4, the pr ices in the spot and futures markets rise to $2.40 and $2.55, respectfully, at the time of maturity. In this exam ple the basis has gone from -0.2 to -0.15, thus the basis has strengthened. In this case, the grower has gained 35 cents per pound solid from the increase in s pot market price, a nd has lost 30 cents per pound solid in the

PAGE 23

23 futures position after liquidation. With basis stre ngthening by 5 cents per pound solid, the grower has now gained 5 cents per pound solid, and the r ealized price is the original $2.05 plus the 5 cents gained, resulting in a realized price of $2.10. At 75,000 pound solids worth, this is a net gain of $3,750. Figure 2-5 demonstrates the outcome of a hedgi ng transaction when pric es in the spot and futures markets fall from $2.05 and $2.25 to $1.25 and $1.55, respectfully. The basis in this scenario has weakened from -0.2 to -0.3. The outcome of the weakening basis results in an 80 cent loss per pound solid in the spot market and a 70 cent per pound solid gain on the short position in the futures market. Subtracting the 10 cent net loss between the two positions on the original $2.05, the new realized price that the grower receives has decreased from $2.05 to $1.95. With 75,000 pounds of product to se ll, the grower lost $7,500. While the transaction costs were ignored in these examples, it is not hard to see how important basis is in hedging. It is easy to see how understanding basis, a nd being able to predict its behavior can be important to the grower Taking a futures position at the right time and gaining additional returns from strengthening in the basis can be the difference between a successful harvesting season and a failure. To account for transaction costs, we can assu me the cost of taking a futures position round trip (meaning the cost for taking to original position, plus the opposite position to liquidate) is $30 per contract. In the third example, the real ized price increased from $2.05 to $2.10, a 5 cent increase per pound solid. With a $30 round trip transaction cost for the futures position, the per pound solid transaction cost is $0.002 per pound. No w the realized price is the $2.10, minus the 00.002 of transaction costs, and the realized price is $2.098 per pound solid. At 75,000 pound solids, this is a $150 transaction cost loss.

PAGE 24

24 The spread between the futures and spot market price can be explained in part by the net marginal outlay for storage, which is defined as the marginal outlay for physical storage plus a marginal risk-aversion factor minus the margin al convenience yield on stocks (Ward, 1977). This is consistent with storage theor y, which holds that the basis equals the marginal costs of carrying stocks through time (Malick). Historically, th e futures basis has been judged by how well it reflects the market fundamenta ls; however the basis may yield discounts and premiums if it reflects anticipation of future events or the potential for such events (Ward, 1977). Figure 2-1 FCOJ Contract Information

PAGE 25

25 Figure 2-2 Hedging Outcomes with Increasing Prices Figure 2-3 Hedging Outcomes with Decreasing Prices

PAGE 26

26 Figure 2-4 Hedging Outcomes with Strengthening Basis Figure 2-5 Hedging Outcomes with Weakening Basis

PAGE 27

27 CHAPTER 3 REVIEW OF THE LITERATURE This chapter reviews the most highly relate d studies to risk management in citrus. Previous studies in this field are generally di vided into two themes: basis and hedging in the citrus industry. Basis refers to the difference between the spot ma rket price and the price of the same commodity in the futures market. Specificall y, basis is calculated as the cash price less the futures price. Hedging refers to the risk management processes of citrus growers. Malick and Ward (1987) studied the stock and seasonality effects of FCOJ. They used a Constant Period from Maturity (CPM) model to measure seasonality. They concluded that when stocks are relatively low, cash prices are pushed up relative to futures prices, encouraging stock holders to draw down existing inventories. Se asonality became more pronounced in the later months of the citrus harvesting season when inventories approached their maximum levels. According to these authors, one of the largest seasonal effects th at arise is very unique to the citrus industry. This is the large swings in FCOJ prices due to freeze losses. Freeze bias, as termed by Frank Dasse and R onald Ward (1977), refers to the bidding up of the futures contract relative to spot prices before the end of a freeze period due to potential speculative gains that could be achieved in the event of a freeze. These authors used bulk FCOJ wholesale price (that is the pric e after processing the oranges into Frozen Concentrated Orange Juice) as the spot price for basi s calculating (futures cash), as oppos ed to the spot market price. Current stock levels play a large part in determining the level of price adjustments; indirectly related, larger stock levels result in smaller price adjustments because of the ability to compensate for losses. Using the Malick and Wards CPM model set to 5 months, a graph of the seasonal nature of FCOJ basis residuals is shown in Figure 3-1 using 3 different levels of stock,. 1.0, 1.2, and 0.8

PAGE 28

28 stock levels were used, where 1.0 is the norm or average. At stock level 1.2 for example, the stock level is 120 percent of the average stock. The basis residuals are a calculated deviation from the average season basis. Fo r example, with stock levels se t at 1.0, the basis residual in Figure 3.1 in April is just below 1. This means that the seasonality effect at that moment causes basis to be higher by 1 than would otherw ise be predicted without seasonality. The basis residual turns positive in the October period for this model. This is generally attributed to the potential for freezes. Since the mi gration of the citrus industry to the south, lower freeze risk can attribute to the stock levels deviating sli ghtly at the December peak as opposed to the merge seen in Figure 3-1 The d ecline in the basis resi dual during the summer months through August can be attributed to diminishing stock levels, which is why basis is so accentuated with the lowest stock le vel in the example of 80 percent. Hurricane seasons now have a very similar eff ect on basis due to thei r ability to destroy large portions of a crop. Also, since Dasse and Ward published th eir work, Florida has played a much smaller role in the overall market for FCOJ, causing the effects of weather to be less elastic. There are many contributing factors to this, which include a loss of Florida crop due to hurricanes, canker, greening, and tristeza, while Brazil and Mexico have increased production to compensate. This has caused a decrease in the price sensitivity to freeze / hurricane destruction possibilities. Malick and Ward reference the basis turning to zero at maturity, but there have been changes since they published their work that might suggest otherwise. For instance, the delivery points were all in Florida when Malick and Ward conducted their research, while FCOJ can now be delivered to many other areas of the United States. This in combination with the lowering proportion of citrus that comes out of Florida compared to other areas of the country and world

PAGE 29

29 can result in much higher delivery costs than in the past for producers. This would tend to have the effect of keeping basis from disappearing at maturity. Ronald Ward and Frank Dasse present th e uniqueness of each commodity, and how it deviates from a generalized theory for storage and basis. In this particular article, the commodity being analyzed is FCOJ. Storage theory suggests th at a futures basis should reflect the marginal cost of physical product plus a risk premium minus a convenien ce yield. Ward and Dasse bring out ideas that might disrupt this theory, such as premiums and discounts as it reflects anticipation of future events, such as the case with freezes and hurricanes. Ward and Dasse bring up other interesting po ints that affect the market. While the season officially begins December 1st, the first crop estimate by the United States Department of Agriculture (USDA) is released in October. Ho wever the December to February time period has the higher probability for possible freeze and the late crop harvest isnt until July. Towards the end of the season there is also added speculation about the next years crop. This may be reflected in contracts that span over these months. Ward and Dasse show the basis residuals ove r a 7 year timeframe in Figure 3-2. The July contract was selected by Ward a nd Dasse for analysis in their basis study because it extends over the complete harvesting period and therefore should reflect a storage co st while being far enough from the next season to not be heav ily influenced by the expected crop. Ward and Dasse use 6 major variables that are hypothesized to contribute to the FCOJ basis residual. These include: measure of risk premium, convenience yield, market liquidity, freeze bias, bias adjustments, and actual freezes. Ward and Dasse concluded that The basis model developed clearly supports both the theory of storage a nd establishes the necessity for measuring market bias. (page 79). While Ward and Dasse say that much of the basis theory

PAGE 30

30 cannot be clearly related to the practicing hedger nor is it necessa ry, they do feel that it is essential that traders using the market have confidence in its economic performance. It should be noted that like Ward and Malic k, the article does not reference other cropdestroying variables such as hurricanes and diseases This is most likely due to when the article was written, i.e., prior to many of the hurricanes and diseases seen in recent years. Behr (1981) aims to develop the conceptu al net benefit model characterizing futures market behavior in terms of hedging open intere st, speculation open interest, and volume. In this dissertation, Behr examines the cross-sectiona l, as well as time-based influences that can affect market behavior. Include d is market structure, government involvement in commodity markets, price risk, and commodity perishability. Salnars (2004) addresses the idea of capitaliz ing on price volatility as contracts approach maturity. This can provide great opportunities for hedging. This high reward comes with an equally large risk, as the small volume of tr ading can lead an unprepared hedger into an unwanted hedging position. Salnars di scusses the convergence of the futures and cash prices as the contracts approach maturity. One of th e reasons for this convergence (also known as narrowing of the basis, or strengthening) is due to available information. As the maturity of a contract approaches, the amount of information increases and th e uncertainties about the future that cause futures and cash prices to differ b ecome fewer. Also, as the contract approaches maturity, if there is not c onvergence between the cash and fu tures prices, speculators may arbitrage between the two, causing th e prices to merge while they take in a profit. In Salnars example of this, a scenario where futures prices are higher than cash prices as the contract approaches maturity; profit could be made by buying the commodity in the cash market and selling it in the futures market, making delivery of the commodity.

PAGE 31

31 Ward and Niles prepared a report for the Futures Committee of the Florida Department of Citrus, which contains a very we ll rounded approach to the varieties of hedgi ng strategies that were available to citrus growers (as of 1975). The report includes background information about the futures markets, the contract specifications types of hedging, price data and statistics, and margin requirements. It then goes into a He dging Decision section, in which it includes information about risks, types of goods that can be hedged, basis patter ns, and broker selection. The final section discusses alternative hedging programs. In this section Ward and Niles include eleven different Hedgi ng Cases. Included are: Cash Grower With Opportunity Cost, Grower With Product Uncertainty, Premature Te rmination of Hedge, and Forward Contracting With Foreign Importers. This report provides a f oundation for growers that are not familiar with hedging strategies. One of Wards (1974) most emphasized concepts is how speculation creates the liquidity necessary to hedge. According to Ward, trading volume from speculative traders is necessary to facilitate establishing futures positions without real izing a cost from market entry or exit. It is argued that if a trade cannot be easily executed at any point in time, then the risk from futures trading is greatly increased. While many of thes e liquidity points may be considered common knowledge, Ward not only discusses the causes and effects of speculation th at create liquidity, but also the effects of other scenarios, such as questioning whether there can be excess speculation and what potential consequences could arise. Ward also addresses lack of complete iden tification of the compos ition of open interest, such as the affect on liquidity when the compos ition of traders is altere d. For example, a report might proclaim that a major portion of the open in terest is held by small traders; however the intent of small traders is not always identifiable. Ward takes a mathematical approach to testing

PAGE 32

32 the liquidity within the FCOJ ma rket. According to the conclusi ons of Ward, The speculative index and its relationship to pr ice distortion provide a clear signaling mechanism to alert policy makers of any needed structural changes in this futures market. Likewise, the methodology reveals the need for improvements in the presen t system for monitoring futures commitments in general. Ranney (1993) discusses the level of inefficien cy in the FCOJ option market because the market has low trading volumes. Ranney also provides examples with the Black model and standard arbitrage condition tests. According to Ranneys thesis, the Black model test indicated inefficiency in every month due to the violation of putcall parity. The arbitrage conditions test also indicated that there was in efficiency due to high percentage s of violations of arbitrage conditions for every contract mont h. Ranney found that levels of inefficiency were higher than those found for any previously studied commod ity option market, and that the level of inefficiency is significant and may adversely affect citrus industry hedging operations. Staying with the subject of efficiency as a necessity for a good hedging strategy, Ward and Behr (1983a) take a look at liqui dity in the futures market in general, not within a specific commodity. The article deals with two main quest ions: (1) What criteria are to be used to measure liquidity, and (2) what le vel of liquidity is c onsidered optimum. One of the methods that is employed is using the spreads between bid an d ask prices as input data for calculating a liquidity index. An index of liquidity is deve loped by Ward and Behr and factors leading to adjustments in the index are evaluated. Ward and Behr (1983b) also provide an all around summary on futures commitments. Both the problem with lack of commitment informati on and the costly and impractical reasoning for the lack of information are presented. The abilit y to increase the accuracy of market performance

PAGE 33

33 from knowledge of open-interest allocation is proved undeniable. At the time Ward and Behrs article was written, the CFTC had begun providing a month-ending classification of futures traders that hold large (excess of 25 contracts) positions. The data included the numbers of long and short commitments reported by hedgers and specu lators, and also included the long and short position of non-reported open interest. Ward and Behr present 4 alternative scheme s for allocating non-re ported short and long positions to hedging and speculation. Two of the procedures, known as the Larson and the Rutledge models, are based on estimates from data obtained through special market surveys. The other two, denoted in the article as Schem e 1 and Scheme 2, are based on numerical weighting procedures with the weights derived theoretically rather than from econometric estimates. Ward and Behr show that all but the Larson s model give very similar allocations over a broad range of commodities, a nd also note that both the Larsons and Rutledge models occasionally lead to allocations that were cl early impossible. All four models did share 2 things of common interest: first, a general procedure is suggested th at is to be applied across all commodity futures, and, second, the procedures are based on some weighting of the reported futures data. Rolls (1984) provides an in-depth look at the relationship between the interaction of prices and a truly exogenous determinant of value, the weather. Due to the highly centralized and concentrated area of orange groves in central Florida, the ongoing weather in a particular spot market can have enormous consequences on the value of the product. Rolls explains that not only is weather a huge determinant in value because of the centralized location, but there is so little affect on price from ot her supply and demand factors.

PAGE 34

34 The demand may have very subtle movements due to price changes in substitutes such as apple juice or grape juice, but national income and tast es generally will not fluctuate enough to explain any significant daily movements. On the same note, short term variations in supply caused by planting decisions would be very low because ora nges grow on trees that re quire five to fifteen years to mature. Rolls also went into detail about daily movements that are allowed within the FCOJ contracts daily fluctuations, and how extreme circumstances, such as freezes and hurricanes, can cause a necessity for price movements beyond the daily limits. In this case, the limits cause price inefficiency. Muraro and Oswalt (2003) published an extens ive report on Florida citrus in September 2003. The report is statistically based with only a short introd uction and mentions of data collection methods as well as necessary assu mptions. While the study holds many points and lays a great foundation for understa nding the percentages of risk, it appears to serve much better as a reference tool. Spreen, et al (2006) prepared a report on the current status and future prospect of the Florida citrus industry. The articl e explains the cause and effects of current citrus problems such as citrus canker as well as citrus greening. Both have become an increasing concern regarding tree health despite the decline of tristeza. The report covers four main sections: 1) anal ysis of the economic im pact of the industry on the Florida economy; 2) assessment of the future structure of the Florida citrus tree nursery industry; 3) analysis of the future prospects for new investment in Florida citrus; and 4) long-run production and price forecasts for Florida citrus u nder varying assumptions that relate to supply issues. According to Spreen et al, the estimated economic impact of the citrus industry on the

PAGE 35

35 Florida economy for 2003-2004, was over $9.25 bill ion, and accounted for over 76,000 jobs in the state. This significant impact of the citrus indus try in Florida provides strong motivation for work related to helping citrus growers. Previ ous research helps provi de an understanding of basis, its patterns, causes and effects. It also provides a fundamental knowledge regarding hedging, however it falls short of analyzing alternative marketing strategies employed in a risk management. Figure 3-1 Seasonality of Basis Residuals

PAGE 36

36 Figure 3-2 Basis Residuals over Time

PAGE 37

37 CHAPTER 4 METHODOLOGY AND DATA The Model The m odel used in this study is develope d by modifying the orange production model developed by Weldon, et al. (2004). An analysis of the impact of the marketing efforts for citrus producers must explicitly examine the variability or risk of future returns. Simulation is a technique that analytically models or replicates an actual system and is used in many areas of study. In agriculture it has been used to model plant growth, market conditions, animal growth and reproduction, environmental and ecological syst ems, intergenerational farm transfers, and many other situations. Simulation modeling can be used to examine farm income under alternative risk scenarios and provides insights into how various firm decisions influence firm survival. Simulating the production and economic activities over a multi-year planning horizon also allows for examination of the impact of federal policy or programs, such as the Florida citrus disaster program, on the profitability and survival of the citrus operation. In this study, the simulations examine the risks and returns associat ed with various methods of marketing citrus. Simetar (Simulation for Excel to Analyze Risk), an Excel add-in that was developed explicitly for stochastic simula tion modeling by Dr. James Richardson, is used to simulate risks and returns to a Valencia and early, midseason (her eon referred to as early-mid) orange grower. The risk and return framework for a representa tive Valencia orange and early-mid orange grower are modeled to simulate financial re sults over a twelve-year time horizon ending in 2020. The model is based on annual cash flows including operating and fixed costs, and forecasted prices and yields. It generates an annual income statement, pro forma cash flow statement, and pro forma balance sheet statement. Since this is a stochastic simulation each iteration, the pseudo-random prices and yields are drawn from a multivariate empirical distribution using a

PAGE 38

38 Latin Hypercube sampling method. The forecasted ra ndom data is correlated based on historical correlations. For each iteration, the simulator provides th e ending net worth at the end of the twelveyear time horizon. The Net Presen t Value (NPV) is calculated as NPV = We for year 2020 / (1+ i)13 Wb for year 2008 (4-1) where i is the real discount ra te, We is the ending net worth, and Wb is the beginning net worth. The NPV is a measure of the amount of wealth that is acquired over th e twelve-year period in present dollars and assumes that all excess cash flows remain in the business and are reinvested with a return equal to the discount rate. Fo r each simulation, 500 iterati ons are run to yield a distribution for the NPV. The mean NPV is the average of these 500 itera tions, or the expected wealth generated or lost for that particular s cenario. Because the objective of this study is to contrast the reward with the risk inherent in the scenario, the st andard deviation for NPV is used to assess the risk associated with that reward. The standard deviation measures the dispersion of a probability distribution. In the simulations, we estimate the mean NP V, the standard deviation for the NPV, and a Yearly Average standard deviation. The NPV sta ndard deviation measures the variation of the net present value. The yearly standard deviati on measures the volatility from year to year associated with the income stream in each iteration of the simulation. The yearly average standard deviation is the average of the year ly standard deviations measured from the simulation. For each scenario, the acreage of the representa tive grower is assumed to be the average acreage of a citrus farm in Flor ida in 2002 (Guci, 2007). It is assu med that the farmer owns the entire acreage. It is also assumed that the farmer is not in debt.

PAGE 39

39 The following comparisons are made for each ci trus variety. First, an NPVbaseline is calculated as the expected gain in wealth projected for the next twelve years assuming the producer enters their entire harv est into a pool. Three alternative NPVscenarios are simulated for both early-mid and Valencia citrus varieties. The results of the simulations for the three alternative scenarios are compared to the pooling NPVbaseline to evaluate the risk and return of the alternative marketing strategies. The first alternative scen ario that is included in the analys is assumes that the producer sells the entire crop in a single week in the cash market when the oranges are harvested. The second alternative scenario assumes that the producer sells the cr op in the cash market equally throughout three consecutive weeks following harvest. The third a nd final alternative scenario assumes the producer takes a short position in the futures market fo r the next crop at the end of the current harvest season, and liquidates that position when the following crop is harvested. Forecasting Market Prices The sim ulations require a forecast of price a nd yield for each season in simulation. Pooling return price forecasts are calculated using an in dex ratio of Valencia pooling price and early-mid pooling price relative to the all orange value1. Pool data for Valencia ranges from $0.982 to $1.71, while pool data for Early-mids ranges from $0.789 to $1.63. This data covers the seasons 1987-1988 to 2002-2003. Pooling returns represent the annual value r eceived as opposed to a weekly price, so an annual index ratio is used to forecast pooling re turns instead of a weekly index. A stochastic normal distribution of the index ratio (derived from the historical annual in dex) is used with the 1 Historical pool price data was obtained via interview process from Dr. John J. VanSickle, professor at the University of Florida who originally obtained this data through Floridas Natural.

PAGE 40

40 forecasted all orange value obtained through Spreen et al. (2006) to derive stochastic forecasted Valencia and Early-mid pool prices. A discrete empirical formula is used to sel ect a harvest week for the first alternative scenario, where the entire crop is sold in the ca sh market in a certain week within a season. The Simetar program randomly selects a harvest we ek within the projec ted season based on the probability distribution for ha rvest weeks in prior seasons2. Using those probabilities, Simetar chooses a particular harvest week and the stochastic price of that week to calcula te receipts and all the other necessary component s of the forecasted financial model. Because of the large number of stochastic calculations involved, it is necessary to use a large number of iterations to get a clearer picture of the probabilities associat ed with the final value. Simulating these data with a large number of iterations yields NPV and standard deviat ion statistics that allow an examination of the risk-return fram ework for the marketing strategys. The same methodology is used for selecting th e harvest week for the other alternative scenarios with changes in the model to reflect th e alternative marketing s cenarios. In scenario two the crop is sold equally over three weeks. The simulation is the same as for scenario one, except that when Simetar chooses a harvest w eek based upon the probability distribution for harvest weeks, it uses the average of the price for that week plus the next two weeks to use as the average selling price. In the event that the week chosen is the second-to-las t week of the season, the price is an average of the last two weeks. In the ev ent that the week chosen is the last week within a season, the price for that week alone is used. A hedge scenario with the futures market requires specification of a time to implement the hedge and a time to liquidate, or offset, the hedge. It is assumed that a hedge is initiated when the 2 These data were collected from Utilization of Florid a Citrus Fruit reports from the Citrus Administrative Committee and can be found through their webs ite, www.citrusadministrativecommitee.org.

PAGE 41

41 prior crop season is closing and lifted when the crop is sold in the cash market. A decision is added in this scenario in which if at the tim e of implanting the hedge, the spot price multiplied by the per-yield acre is greater than the produc tion costs per acre, the hedge is implemented, otherwise the grower takes their chances, and sells directly into a single w eek in the cash market at harvest. For this scenario, a random week based on an empirical distribution is used for liquidation of the hedge. The gain/loss from the fu tures position is taken in conjunction with the spot market price at the time of harvest to determine net receipts for that year. Forecasted spot market prices are required to simulate the returns for the three alternative scenarios. To do this, historical seasonal index is calculated from the weekly spot market prices for Valencias and early-mids relative to th e average all-orange spot market price of that season. Spot market data for both Valencias and early-mids were obtained through the USDA National Agricultural Statistics Service (NASS)3. The Valencia spot market prices range from $0.550 to $2.15 per pound solid, with an average of $1.193 for the years 1983 to 2006. Early-mid spot market prices range fr om $0.425 to $1.75 per pound solid, with an average of $0.992 for the years 1982 to 2006. The all-orange price per box on tree and all-orange concentrat e yield were also obtained from NASS. The on tree price per box ranges from $2.89 to $7.58 with an average of $4.722. A standard conversion rate of 4.156 (Citrus Reference Book, 41) is us ed to derive an all-orange price per pound solid by dividing the all-orange on tree price per box by the product of the allorange concentrate yield and the conversion rate. A seasonal index is derived for Valencias and ea rly-mids prices by calculating the ratio of the weekly Valencias or early-m id price over all-orange annu al price. This product gives a 3 These data are available through their website: http:// www.nass.usda.gov/Statistics_b y_State/Florida, as well as through the hard copies of their annual Citrus Summary reports.

PAGE 42

42 distribution of weekly index values for each product. For example, in week 13 the mean seasonal index value of the Valencia spot price over the all-orange spot price from the years 1983 to 2004 is 1.599 with a standard devi ation of 0.358 (Figure 4-1). The distribution of weekly s easonal index values is assume d to be normally distributed, and a stochastic function is us ed to create a random index for each week in each season to be simulated. The series of weekly seasonal index values are then correlated with a correlation matrix to insure continuity of prices within a season. Using forecasted all-orange spot market pric es, these stochastic seasonal indexes (along with the forecasted price) are used to obtain weekly forecasted stochastic spot prices for valencia and early-mids over the projected seasons. The for ecasted all-orange spot price used in this analysis is derived from Spreen, et al (2006). The forecasted all-ora nge spot price (converted into per pound solid units) covers the seasons of 2007-2008 to 2019-2020 and ranges from $0.846 to $1.472 with an average of $1.263. Using weekly historical futures price data and weekly historical spot market price data, a historical Valencia basis and an early-mids basis are calculated. This yiel ds a weekly Valencia basis which is used to calculate a mean and stan dard deviation. These two statistics are used to produce a normally distributed basis for each wee k. These values are also used to derive a correlation matrix for the weekly basis to insu re seasonal consistency from week-to-week. For example, with the correlation in place, there is a much smaller probability of the basis swinging from one tail of the normal distribution to the other tail over th e course of a week. When the forecasted stochastic Valencia spot price in a given week is combined with the stochastic Valencia basis, a stochastic forecas ted Valencia based futures market price can be

PAGE 43

43 developed. This process is repeat ed with the early-mid spot market price and basis data to develop the forecasted futures market price. For the final marketing strategy, hedging using futures, historical near by contract data was collected. Nearby futures contract data was received from the New York Board of Trade (NYBOT), which is now the Intercontinental Exch ange (ICE). The period of the data run from October, 1982 to May, 2007, with prices ranging from $0.551 to $2.072. To develop a discrete empiri cal distribution for the harves ting time, data was collected through the Citrus Administrative Committee. This data contains the past 3 years of quantity of data processed weekly for both Valencia and ea rly-mid. Using this data, the percentages were broken into the necessary number of categories needed, 10 for Valencia, and 17 for early-mid. Using the Dempirical and Hlookup functions, a system is set in place for the model, in any given year, to choose a harvesting week based on historical probabilities, and the stochastic forecasted price from that week for the necessary year is used. For the scenario involving a distribution of harvesting over three consecutive weeks, a system is set in place for any given price to be chosen in which an average of the given w eek plus the next two weeks is used. In the event that the 2nd to final week is chosen, it retrieves an average of the last two weeks. In the event that the final week is chosen, that value is used. This process is do ne for both Valencia and early-mid prices. Forecasting Futures Prices for the Hedge Model The hedging scenario, for both Valencias and early-m ids requires a few key assumptions. The first assumption is that the hedge is implemented at the end of th e previous years harvest season. It is at the end of the previous harvest season that the following crop development begins and the grower incurs production costs. The growers hedged amount co mes from an average of his forecasted yield for the upcoming year, and the r ealized yield of the two previous years. The

PAGE 44

44 basic hedge formula uses the spot price when th e hedge is implemented, and adds the change in basis and subtracts transaction costs (brokerage fees) for tr ading futures contracts. The transaction cost (each way) is assumed to be $20 per contract (at 15,000 pound solid per contract), a citrus industry norm. Like reality, the process of hedging an amount based on previous years yields leaves the chance of over and under hedging. In the situation in which the crop is un der-hedged, the excess, un-hedged portion of the crop is assumed to be sold into the cash market at the week at harvest. In the event that the crop is over-hedged, the ga ins or losses accumulated from the excess futures position are added or subtracted from the final revenue for the given year. Forecasted Yields Historical farm -level yield da ta (boxes per acre) from southwest Florida were used to calculate an average yield for a gi ven year across all farms, as well as a standard deviation of the yield from year-to-year within a single farm. The standard deviat ion of the yearly averages is used as an annual standard deviation while for ecasting yields. This annual standard deviation is calculated to be 48.6 for Vale ncia and 63.9 for early-mid. Within each farm, from year-to-year a deviati on from that particular years average yield is calculated, and an average of those standard deviations is taken as a farm level deviation. This is done with both Valencia and early-mid data. This farm level deviation for Valencia is 70.3, and 85.1 for early-mid. The farm level deviations were expected to be much higher than the annual standard deviation because this annual sta ndard deviation averages the yields across all farms, lowering the volatility. Using historical Valencia, early-mid, and al l-orange yield data, an average ratio of Valencia to all-orange, and early -mid to all-orange yields are calculated. Using the average and normal distribution of these data, combined with the low-level greening forecasted yields found

PAGE 45

45 in Spreen, et al. (2006), stochas tic forecasted yields for Valencia and early-mid are derived. A stochastic normal distribution for historical Valencia and early-mid concentrate yields, are used with the conversion rate to convert fro m boxes per acre to pou nd-solids per acre. The final yield numbers are calculated using the historical ratio of Va lencia to all-orange yield, combined with the forecasted yields provided by Spreen, et al. By using a stochastic normal distribution of the Valencia to all-ora nge ratio, combined with the calculated annual standard deviation as the standa rd deviation, we arrive at our fo recasted yields across all farms. To add in the final volatility associated with farm level yield, one more stochastic normal distribution is added. This norma l distribution is calculated with an average of zero, and the farm level deviation as the standard deviation. This last value adds in the farm-level volatility necessary for this model. This process is repeated for the early -mid yield. Finally, the correlation matrix based on historical prices and yields, provides the price-yield correlation necessary to keep the values practical.

PAGE 46

46 Figure 4-1 Valencia Spot Index

PAGE 47

47 CHAPTER 5 RESULTS AND DISCUSSION Results The results from the simulations can be seen in Tables 5-1 and 5-2. A probability distribution function (PDF) graph of the Valencia baseline and scen arios can be seen in Figure 51. A similar PDF displaying the results of the earl y-mid baseline and scenarios be seen in Figure 5-2. Valencia Baseline scenario The baseline scenarios for the Valencia and early-m id oranges assume the entire crop is sold through a pooling program. The mean net pr esent value (NPV) for the Valencia baseline scenario is $845,722, with a standard de viation of 1,032,299. The minimum NPV obtained during any single iteration is less than $-1.3 million, while the maximum NPV obtained is over $4.5 million. A graph of this baseline scenario showing the probability distribution function (PDF) for Valencia can be seen in Figure 5-3. As seen in Table 5-5, the average yearly standard deviation (the average of the yearly standard de viations measured from the simulation) for this scenario is 77,182. Critical NPV values can be seen in Table 5-3. This table shows that the Valencia baseline scenario holds a 79 percent chance of producing a NPV of 0 (breakeven) or better. There is a 10 percent chance of losing at le ast $500,000, and a 4 percent chance of losing at least $750,000. On the positive side, ther e is a 44 percent chan ce of making at least $750,000, and a 15 percent chance of making at least $1.5 million.

PAGE 48

48 One-week cash market scenario In the scenario involving the entire harvest being sold into the cash market in a single week (hereby referred to as Scenario 1), the mean NPV is $835,567 with a standard deviation of 853,462. The minimum NPV obtained during any si ngle iteration is less than $-1.25 million, while the maximum NPV obtained is almost $3.5 million. A PDF graph of Scenario 1 for Valencia can be seen in Figure 5-4. As seen in Table 5-5, the av erage yearly standard deviation for this scenario is 74,482. As seen in Table 5-3, Valencia Scenario 1 holds an 83 percent chance of producing a NPV of at least 0 (breakeven ). This scenario holds a chance of losing $500,000 or more of 13 percent, and a 3 percent chance of losing $750,000 or more. On the positive side, Scenario 1 has a 53 percent chance of producing a NPV of at least $750,000, and a 22 percent chance of making at least $1.5 million. Three-week cash market scenario In the scenario involving the entire harvest be ing sold in the cash market the three weeks following harvest (hereby referred to as Scenario 2), the mean NPV is $800,552, with a standard deviation of 841,486. The minimum NPV obtained during any single iteration is less than $-1.25 million, while the maximum NPV obtained is over $3.25 million. A PDF graph of Scenario 2 for Valencia can be seen in Figure 5-5. As seen in Table 5-5, the average yearly standard deviation for this scenario is 52,164. This Valencia Scenario holds an 82 percent chance of producing at least a breakeven NPV (Table 5.3). This scenario has a chance of losi ng at least $500,000 of 7 pe rcent, and a 3 percent chance of losing $750,000 or more. On the positive side, this Valencia scenario holds a 52 percent chance of producing a NPV of at least $750,000, and a 20 percent chance of making at least $1.5 million.

PAGE 49

49 Hedging scenario In the scenario involving hedging in the futu res market (hereby refe rred to as Scenario 3), the mean NPV is $547,506, with a standard deviation of 742,264. The minimum NPV obtained during any single iteration is ju st under $-1.2 million, while the maximum NPV obtained is over $2.7 million. A PDF graph of this scenario can be seen in Figure 5-6. The average yearly standard deviation fo r this scenario is 65,594 (Table 5-5). This Valencia scenario holds a 77 percent chance of producing at least a breakeven NPV (Table 5-3). This scenario has a 10 percent ch ance of losing at least $500,000, and a 3 percent chance of losing $750,000 or more. On the positive side, Valencia Scenario 3 holds a 39 percent chance of producing a NPV of at least $750,000, a nd a 10 percent chance of making at least $1.5 million. A PDF graph of the change in basis associat ed with Valencia Scenario 3 can be seen in Figure 5-7. Early-mid Baseline scenario The baseline scenario for early-m id assumes the entire crop is sold through a pooling program. The mean NPV for the early-mid baseline scenario is $633,507 with a standard deviation of 844,480. The minimum NPV obtained dur ing any single iterati on is less than $-1.3 million, while the maximum obtained is over $4.5 million. A PDF graph of the early-mid baseline scenario can be seen in Figure 5-8. The average yearly standard deviation for this scenario is 65,393 (Table 5-5). The early-mid baseline scenario holds a 75 percent chance of producing at least a breakeven NPV (Table 5-4). This scenario has a 10 percent chance of losing $500,000 or more, and a 4 percent chance of losing $750,000 or more. On the positive side, the early-mid baseline

PAGE 50

50 scenario produced a 44 percent chance of producing a NPV of at least $750,000, and a 15 percent chance of producing a NPV of at least $1.5 million. One-week cash market scenario The mean NPV of the early-mid Scenario 1 is $364,364, with a sta ndard deviation of 674,524. The minimum NPV obtained during any single iteration is less than $-1.2 million, while the maximum is over $2.3 million. A PDF graph of Scenario 1 for early-mids can be seen in Figure 5-9. The average yearly standard devia tion for this scenario is 65,393 (Table 5-5). The early-mid Scenario 1 has a 70 percent chance of produci ng at least a breakeven NPV (Table 5-4). This scenario has a 13 percent chance of producing a NPV of negative $500,000 or less, and a 4 percent chance of producing a NPV of negative $750,000 or less. On the positive side, early-mid Scenario 1 has a 28 percent ch ance of producing a NPV of at least $750,000, and a 5 percent chance of producing a NPV of at least $1.5 million. Three-week cash market scenario The mean NPV of the early-mids Scenario 2 is $369,187, with a sta ndard deviation of 674,213. The minimum NPV obtained during any si ngle iteration is under $-1.2 million, while the maximum obtained is almost $2.4 million. A PDF graph of Scenario 2 for early-mids can be seen in Figure 5-10. The average yearly standard deviation for this scenario is 56,993 (Table 55). This early-mid Scenario 2 holds just a 70 percent chance of producing at least a breakeven NPV (Table 5-4). This scenario also holds a 13 percent chance of lo sing at least $500,000, and a 4 percent chance of losing at least $750,000. On the positive side, early-mid Scenario 2 holds a 29 percent chance of producing a NPV of at least $750,000, and a 5 percent chance of making at least $1.5 million.

PAGE 51

51 Hedging scenario In the early-mid scenario involving hedging in the futures market (h ereby referred to as Scenario 3), the mean NPV is $169,851, with a standard deviation of 761,105. The minimum NPV obtained during any single iteration is under $-1.5 million, while the maximum NPV obtained is over $2.7 million. A PDF graph of S cenario 3 can be seen in Figure 5-11. The average yearly standard deviation fo r this scenario is 69,199 (Table 5-5). This early-mids Scenario 3 holds a 55 percen t chance of producing at least a breakeven NPV. This scenario has a chance of losing at least $500,000 of 23 per cent, and a 7 percent chance of losing $750,000 or more. On the positive side, this scenario holds a 22 percent chance of producing a NPV of at least $750,000, and a 5 percent chance of making at least $1.5 million. A PDF graph of the change in basis associated w ith Valencia Scenario 3 can be seen in Figure 512. Discussion Valencia Prelim inary expected results would lead us to believe that the baseline scenario with pooling would provide the lowest volatility, that is, the lowest standard deviation. However, contrary to this belief, the baseline scenario in fact provided the highest NPV with the highest level of volatility. Scen ario 1 yielded the second highest NP V, with the second-to-least amount of volatility. Scenario 2 yielded th e third highest NPV, with slightly less volatility than scenario 1. Scenario 3, the hedging scenario, in fact provided the lowest NPV by a large margin, combined with the lowest standard deviation. While return vs. risk is stated to be de termined by NPV vs. standard deviation, the minimum and maximum potential va lues are relevant as well. Th e Valencia baseline scenario yielded the highest maximum at over $4.5 millio n, while scenario 1 reached $3.4 million, and

PAGE 52

52 scenario 2 fell about $150,000 short of scenario 1. Scenario 3 had the lowest maximum, falling short of $3 million. The most favorable minimum was obtained by scenario 3 with negative $1.2 million, followed by scenario 2 at just under negative $1.25 million, while scenario 1 fell about $3,000 below that, and the baseline scenario fell over $100,000 below th at of scenario 1. Having the highest chance of not being able to break even, at 23 percent, the Valencia scenario 3 held the second highest chance of losing at least $500,000. S cenario 3 however fell short only to the Valencia baseline which held a 5 percent chance of losing at least $750,000. This, however, can be attributed to risk, as this same baseline scenario held the highest chance of making $1.5 or 2.5 million at 25 percent and 6 percent respectively. This baseline scenario clearly produces the higher risk complimented by a higher reward. The most risk averse Valencia grower would be much better off utilizing Scenar ios 2 or 3, which produced the highest chances of at least breaking even (83 a nd 82 percent, respectively), as well as the highest chances of making at least $750,000 (53 and 52 percent). Also, these scenarios generated the lowest chances of losing $500,000 at 3 percent for each. In no ca tegories shown in Table 5-3 did Scenario 1 perform the worst. Scenario 2 held the lowest yea r-to-year volatility with a standard deviation of 56,993, almost 10,000 lower then the next closest scenario. Early-mid Again, we would expect the baseline scenario of pooling to yield the lowest volatility, and in fact, it again held the hi ghest NPV with the highest level of volatility. As far as early-mid oranges go, the baseline scenario yielded the most favorable results in other areas as well, having the both the highest maximum and highest mini mum. Scenario 3 yielded the second highest maximum at over $2.7 million, followed by scenario 2 and lastly scenar io 1. The minimums obtained also favored the baseli ne scenario, with scenario 2 being the second best option.

PAGE 53

53 The early-mid baseline scenario clearly outpe rformed the other scenarios for all critical values. In no category was any other scenario favorab le to the baseline (Table 5-4). In fact, many categories are not even close. The early-mid baseli ne scenario holds a 15 pe rcent better chance of producing an NPV of at least $750,000. The next closest scenario (Scenario 2) was 3 times more likely to produce a NPV of at least $1.5 million than any other scenario. Only the baseline and scenario 3 showed any chance of reaching $2.5 m illion, at 2 and 1 percent, respectively. The early-mid scenario 4 was not competitive in any category, and was the least-favored scenario in all but a single category, obtaining a NPV of at least $2.5 million. Changes in basis A change in basis model, shown in Figur e 5-7 and Figure 5-12 could provide some insight into the detriment of th e early-mid scenario 4. After 1,000 ite rations, with an average of 0.0907, this negative change in basis( i.e., weakening) hurts the citr us growers net returns. The Valencia change in basis, averaging a positive 0.0039, help s the citrus grower. This strengthening of the basis helps move the net price higher for the producer. The early-mid change in basis also produced a much higher variation than that of the Valencia. While the Valencia it erations provided a pos itive change in basi s 48 percent of the time, the early-mid change in basis was only positive 30 percent of the time.

PAGE 54

54 Table 5-1 Scenario Results (Valencia) Scenario V Pool V 1-week V 3-week V Hedge Mean NPV $845,722 $835,567 $800,552 $547,506 Std Dev 1,032,299 853,462 841,486 742,264 Min $-1,371,947 $-1,264,171 $-1,261,862 $-1,205,965 Max $4,536,464 $3,401,561 $3,251,220 $2,788,583 Table 5-2 Scenario Results (Early-mid) Scenario EM pool EM 1-we ek EM 3-week EM Hedge Mean NPV $633,507 $364,364 $369,187 $169,851 Std Dev 844,480 674,524 674,213 761,105 Min $-1,252,346 $-1,262,639 $-1,261,849 $-1,541,493 Max $3,429,396 $2,376,623 $2,397,002 $2,780,505 Table 5-3 Critical Points (Valencia) Scenario Baseline Scenario 1 Scenario 2 Scenario 3 NPV of $-750,000 or less 5 percent 3 percent 3 percent 3 percent NPV of $-500,000 or less 9 percent 7 percent 7 percent 10 percent Breaking even or better 79 percent 83 percent 82 percent 77 percent NPV of At Least $750,000 51 percent 53 percent 52 percent 39 percent NPV of At Least $1,500,000 25 percent 22 percent 20 percent 10 percent NPV of At Least $2,500,000 6 percent 3 percent 3 percent 1 percent Table 5-4 Critical Points (Early-mid) Scenario Baseline Scenario 1 Scenario 2 Scenario 3 NPV of $-750,000 or less 4 percent 4 percent 4 percent 7 percent NPV of $-500,000 or less 10 percent 13 percent 13 percent 23 percent Breaking even or better 75 percent 70 percent 70 percent 55 percent NPV of At Least $750,000 44 percent 28 percent 29 percent 22 percent NPV of At Least $1,500,000 15 percent 5 percent 5 percent 5 percent NPV of At Least $2,500,000 2 percent 0 percent 0 percent 1 percent

PAGE 55

55 Table 5-5 Average Yearly Standard Deviations Scenario V Pool V Scenario 1 V Scenario 2 V Scenario 3 EM Pool EM Scenario 1 EM Scenario 2 EM Scenario 3 Average Yearly Std Devs 77,182 74,482 52,164 65,594 65,393 65,140 56,993 69,199 Figure 5-1 All Valencia Scenarios

PAGE 56

56 Figure 5-2 All Early-mid Scenarios Figure 5-3 Valencia Pool

PAGE 57

57 Figure 5-4 Valencia Scenario 1 Figure 5-5 Valencia Scenario 2

PAGE 58

58 Figure 5-6 Valencia Scenario 3 Figure 5-7 Valencia Change in Basis PDF

PAGE 59

59 Figure 5-8 Early-mid Pool Figure 5-9 Early-mid Scenario 1

PAGE 60

60 Figure 5-10 Early-mid Scenario 2 Figure 5-11 Early-mid Scenario 3

PAGE 61

61 Figure 5-12 Early-mid Change in Basis

PAGE 62

62 CHAPTER 6 SUMMARY AND CONCLUSIONS Summary With the citrus industry being the largest agricu ltural sub-sector in th e state of Florida at over 719,000 acres in use, the state of Florida accounts for over 67 percent of the total U.S. production for oranges. Over 95 percent of the oranges grown in Florida are grown for processing, and over 90 percent of that comes fr om two types of citrus, Valencia and early, midseason oranges. Volatility in th e citrus industry caused by weathe r, disease, and other factors causes concern to citrus growers whose goal is to secure profitable returns. With various marketing strategies available to growers, an im partial examination of th e various strategies and forecasted outcomes provides citrus growers with the tools to further their knowledge on available marketing strategies including their forecasted returns and the risks associated with those strategies. This study examines the risks and returns associ ated with 4 scenarios for each of the two types of citrus. The baseline s cenario is the pooling method, in which growers pool their crops and receive payment based on the average pri ce received for the entire pooled crop. This scenario is an estimate of the average cash market price received in a si ngle harvest season. The first alternative scenario compared to this baseline assumes the producer sells the entire crop into the cash market at once. The second scenario assu mes the grower sells the entire crop into the cash market over the course of th ree weeks. The final scenario assumes that the producer hedges the crop utilizing the FCOJ contract in the future s market when the previo us harvest season ends. Simetar (Simulation for Excel to Analyze Risk), an Excel add-in that was developed explicitly for stochastic simulation modeling was used to creat e stochastic forecasts for price and yield, utilizing observed probabilities. Using the stochastic forecasted prices along with

PAGE 63

63 stochastic forecasted yields in conjunction with a pro-forma financial st atement model, we are able to view the net present value (NPV) of discounted revenues through the year 2020. By running a large number of iterations, we can see the expected return (NPV) and volatility associated with the various marketing methods. The baseline scenario of pooling for Valencia oranges provided the highest volatility (as measured by standard deviation) along with the highest NPV. The one-week cash market scenario (scenario 1) yielded th e second highest NPV, with the l east amount of volatility. Selling the crop into three consecutive cash market weeks (scenario 2) yielded the second lowest NPV, with slightly more volatility than scenario 1. Scenario 3, th e hedging scenario, provided the lowest NPV by a large margin, and the highest volatility. For early-mid oranges, the baseline was clearly the most successful scenario, providing the highest NPV combined with the lo west volatility. This baseline scen ario also yielded the highest maximum NPV obtained for early-mid oranges and the most favorable minimum as well. Conclusions Florida citru s growers are challenged every year to choose a marketing strategy that will be the most favorable to them. As opposed to ot her commodities such as corn and wheat, the physical area that is designated to growing citrus is very conc entrated. This high concentration causes much higher price volatility than is obser ved for most other commodities, in turn causing citrus growers to spend large amounts of time an d energy looking for marketing strategies that can benefit them. This study provides growers info rmation about various ma rketing alternatives available to them, as well as an estimate on th e forecasted risks and rewards associated with these various methods. For the Valencia grower, pooling is expected to provide the highest net present value along with the highest volatility. This is the essence of a high risk to reward relationship. Scenario 1,

PAGE 64

64 selling directly in the cash ma rket in a single week, yielded th e second highest net present value with the lowest volatility. Despite contrary preliminary expectations, this scenario provided a great lower risk alternative to the pooling strategy. The final scenario, hedging, proved to be an unfavorable strategy, providing high volat ility with a low net present value. For the early, midseason grower, the baseline sc enario of pooling provided the most favorable alternative in almost every way. This scenario outperformed the three alte rnatives with a higher net present value, a lower vola tility, a higher maximum, and the most favorable minimum. The low risk and high reward of the baseline scenar io makes pooling a clear choice for the early-mid citrus grower. Implications Marketing strategies continue to evolve for ci trus growers. This study assum es that a citrus grower will utilize a single mark eting strategy for 100 percent of their crop. The information for this study can be used as a base even though some citrus grower will utilize multiple strategies, each regarding portions of their harvested crop. There are also other issues in the analysis that can influence the results. For example, a higher discount rate will result in lower net present values and vice versa. With vari ous risk adversity levels of a particular farmer calculated, the ability to utilize multiple strate gies for various amounts of their harvested crop is available. Recent hurricane events in Florid a as well as the presence of greening, canker, tristeza, and urban development of farm land serve as reminde rs that the production risk associated with Florida citrus growers is valid and must be addressed. The fi nancial implications including potential losses due to these f actors indicate great loss potentia l for citrus growers who do not utilize some type of risk management. For instan ce, a farmer that takes a short position in the futures market to offset the downside price risk they face from plant to harves t, is at risk to be hurt twice as hard if a hurricane takes out his crop. A hurricane can cause a farmer to lose his

PAGE 65

65 crop and to lose on a short position in a futures ma rket that could see dram atic increases in price following the hurricane. A season of losses at the level brought upon by th ese two consequences of one event could put many farmers out of business. Further Research Needs There are many areas of this study that are left open for further rese arch. T he availability of utilizing a combination of strategies can potentially add another level of practicality on this study. An econometric study could provide great in sight into the optimized combinations of these marketing strategies for citrus grow ers with various risk adversity levels. A study including the use of options along with futures contracts for a hedging could also yield interesting results that w ould be useful for the citrus gr ower. However, the random walk of futures prices renders great problems in th e hedging scenario. Market theory would suggest that if futures prices were able to be forecas t with any accuracy, the speculator arbitrage of the information would in fact distort the futures pr ice away from the predicted price and render the forecasted price useless.

PAGE 66

66 LIST OF REFERENCES Andre, M. The U.S. Citrus Growers: Competitiveness and Performance. M.S. thesis. University of Florida, May 1996. Behr, R. A Simultaneous Equation Model of Futures Market Trading Activity. Ph.D. dissertation. University of Florida, June 1981. Brown, M., et al. Florida Citrus Producti on Trends. Economic and Market Research Department Report. Gain esville, Florida: 2007. Citrus Administrative Committee. Utilization of Florida Citrus Fruit Reports Internet Site: ( http://www.citrusadministrativecommittee.org/) No copyright, Fruit and Vegetable Division. Lakeland, Flo rida. (Accessed February 21, 2008). Crist, R. The Citrus Industry in Florida. American Journal of Economics and Sociology 1(1955):1-12. Florida Agriculture. Overview of Florida Agriculture Internet Site: ( http://www.florida-agriculture.com/). C opyright 2004-2007, Florida Departm ent of Agriculture and Consumer Services. Tallaha ssee, Florida. (Accessed January 15, 2008). Florida Department of Citrus, Citrus Reference Book. Economic and Market Research Department. May 2007. Guci, L., and M. Brown. Changes in the Structur e of the Florida Processed Orange Industry and Potential Impacts on Competition. Florida Departme nt of Citrus, Gainesville. FL. Staff Report. 2007. Hodges, A., et al. Economic Impact of Florida s Citrus Industry. Economic Information Report 1999-2000. Gainesville, Florida: 2001. Hull, J. Fundamentals of Futures and Options Markets. Upper Saddle River, NJ: Pearson, 2002 Malick, W., and R. Ward. Stock Effects a nd Seasonality in the FCOJ Futures Basis. The Journal of Futures Markets 2(1987):157-167. Muraro, R., and W.C. Oswalt. Budgeting Cost s and Returns for Central Florida Citrus Production. IFAS Report 2002-2003. Gainesville, Florida: 2003. New York Board of Trade. Orange Juice Sta tistics. Nearby Futures C ontract Prices, 1982 2007. 2007. Ranney, J. Pricing Efficiency of Options on FCOJ Futures Contracts. M.S. thesis. University of Florida, August 1993. Roll, R. Orange Ju ice and Weather. The American Economic Review 74(1984):861-80.

PAGE 67

67 Salnars, C. Evaluation of the Effect of Captiv e Supply and Market Liquidity on Basis Levels as Contracts Approach Maturity in the United St ates Live Cattle Futures Markets. M.S. thesis. University of Florida, May 2004. Scora, R. On the History and Origin of Citrus. Bulletin of the Torrey Botanical Club 6(1975):369-375. Spreen, T., et al. An Economic Assessment of the Future Prospects for the Florida Citrus Industry. IFAS Report. Gainesville, Florida: 2006. Spreen, T., M. Brown, and R. Muraro. The Proj ected Impact of Citrus Greening in Sao Paulo and Florida on Processed Orange Production an d Price. Economic Research Department Report. Gainesville, Florida 2005. USDAs National Agricultural Statistics Service, Florida Field Office (NASS). Florida State Agricultural Overview 2006 Internet Site: ( http://www.nass.usda.gov/) Orlando, Florida (Accessed January 15, 2008). USDAs National Agricultural Statistics Service, Florida Field Office (NASS). Florida Citrus Summaries 1982 2006 Internet Site: ( http://www.nass.usda.gov/). Orlando, Florida. (Accessed January 15, 2008). W ard, R. Market Liquidity in the FCOJ Futures Market. American Journal of Agricultural Economics 1(1974):150-154. Ward, R., and J. Niles. Hedging Strategies in FCOJ Futures. Economic Research Department Report. Gainesville, Florida: 1975. Ward, R., and F. Dasse. Empirical Contributions to Basis Theory: The Case of Citrus Futures. American Journal of Agricultural Economics 1(1977):71-80. Ward, R., and R. Behr. Futures Trading Liquidity : An Application of a Futures Trading Model. The Journal of Futures Markets 3(1983a):207-224. Ward, R., and R. Behr. Allocating Nonreported Futures Commitments. The Journal of Futures Markets 4(1983b):393-401. Weldon, R., R. Hinson, J. VanSickle, and R. Mura ro. The Economic Impact of the 2004 Citrus Disaster Program on Florida Citrus Producers. IFAS Report. Gainesville, Florida: 2004.

PAGE 68

68 BIOGRAPHICAL SKETCH Evan Marc Shinbaum the son of Amy and Ky le Shinbaum, was born in Fort Myers, FL, on March 9th, 1983. After graduating from Fort Myer s High School in 2002, he spent time studying at Florida International Un iversity in Miami before transferring to the University of Florida. After completing the Bachelor of Scienc e (B.S.) degree in food and resource economics, he entered the graduate program in the same fiel d to pursue the Master of Science (M.S.) degree.