<%BANNER%>

Cost/Benefit Analysis of Public Investment into CMNP Registration and Development for Citrus Mechanical Harvesting

Permanent Link: http://ufdc.ufl.edu/UFE0025050/00001

Material Information

Title: Cost/Benefit Analysis of Public Investment into CMNP Registration and Development for Citrus Mechanical Harvesting
Physical Description: 1 online resource (76 p.)
Language: english
Creator: Blanco, German
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2009

Subjects

Subjects / Keywords: abscission, benefit, citrus, cmnp, cost, harvesting, mechanical, oranges, registration, valencia
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: Maintaining global competitiveness and offsetting cost increments due to the incursion of new management strategies to combat citrus greening and canker are two reasons why the Florida citrus industry needs to find ways to reduce costs. Since harvesting costs play such an important role in citrus production, the Florida Department of Citrus (DOC) has dedicated a portion of grower taxes (public funds) to develop and register an abscission agent compound, CMNP, with the U.S. Environmental Protection Agency as part of its effort to make citrus mechanical harvesting cost effective. Total cost of registration is estimated to be between $7 and $10 million. In addition, the University of Florida-IFAS has spent between $500 and $700 thousand of public dollars annually since 1996 to research abscission compounds and develop management strategies. The primary objective of this research was to compare a discounted stream of private benefits of mechanical harvesting with abscission agent application against the public costs of abscission development and chemical registration. The time period for this analysis spans 22 years, from 1996 to 2018. This research assumes that abscission agent application is the solution to harvesting late-season 'Valencia' oranges. Without abscission agent application, mechanical harvesting systems would have a negative impact on next year's yield due to the removal of young immature fruit at the time of harvesting during the present year. Two benefits of CMNP with mechanical harvesting late-season 'Valencia' are estimated: 1) harvest cost savings during the current year, and 2) preserving the value of fruit that would have been lost in the next season by mechanically harvesting during the current season without abscission. The net present value (NPV) of mechanically harvesting 25,000 acres (acreage limit by an anticipated Experimental Use Permit) through the 2018 season ranges between $30.62 and $50.67 million, depending on price, production, and fruit loss scenarios. For any of the scenarios considered, the discounted value of abscission benefits with mechanical harvesting fully pays for the costs of registration and development within four years. A subsequent analysis of minimum acreage of CMNP application during the late season and levels of savings indicates that there exists a threshold level under which the NPV of this investment is not longer positive. Results from this analysis should provide: 1) a measure of confidence that the pay-off to the general industry should be positive and significant and 2) information about threshold levels of acreage and cost differential between hand and mechanical harvesting options to assure a positive NPV for the public investment into the abscission compound, CMNP.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by German Blanco.
Thesis: Thesis (M.S.)--University of Florida, 2009.
Local: Adviser: Roka, Fritz M.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2011-08-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2009
System ID: UFE0025050:00001

Permanent Link: http://ufdc.ufl.edu/UFE0025050/00001

Material Information

Title: Cost/Benefit Analysis of Public Investment into CMNP Registration and Development for Citrus Mechanical Harvesting
Physical Description: 1 online resource (76 p.)
Language: english
Creator: Blanco, German
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2009

Subjects

Subjects / Keywords: abscission, benefit, citrus, cmnp, cost, harvesting, mechanical, oranges, registration, valencia
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: Maintaining global competitiveness and offsetting cost increments due to the incursion of new management strategies to combat citrus greening and canker are two reasons why the Florida citrus industry needs to find ways to reduce costs. Since harvesting costs play such an important role in citrus production, the Florida Department of Citrus (DOC) has dedicated a portion of grower taxes (public funds) to develop and register an abscission agent compound, CMNP, with the U.S. Environmental Protection Agency as part of its effort to make citrus mechanical harvesting cost effective. Total cost of registration is estimated to be between $7 and $10 million. In addition, the University of Florida-IFAS has spent between $500 and $700 thousand of public dollars annually since 1996 to research abscission compounds and develop management strategies. The primary objective of this research was to compare a discounted stream of private benefits of mechanical harvesting with abscission agent application against the public costs of abscission development and chemical registration. The time period for this analysis spans 22 years, from 1996 to 2018. This research assumes that abscission agent application is the solution to harvesting late-season 'Valencia' oranges. Without abscission agent application, mechanical harvesting systems would have a negative impact on next year's yield due to the removal of young immature fruit at the time of harvesting during the present year. Two benefits of CMNP with mechanical harvesting late-season 'Valencia' are estimated: 1) harvest cost savings during the current year, and 2) preserving the value of fruit that would have been lost in the next season by mechanically harvesting during the current season without abscission. The net present value (NPV) of mechanically harvesting 25,000 acres (acreage limit by an anticipated Experimental Use Permit) through the 2018 season ranges between $30.62 and $50.67 million, depending on price, production, and fruit loss scenarios. For any of the scenarios considered, the discounted value of abscission benefits with mechanical harvesting fully pays for the costs of registration and development within four years. A subsequent analysis of minimum acreage of CMNP application during the late season and levels of savings indicates that there exists a threshold level under which the NPV of this investment is not longer positive. Results from this analysis should provide: 1) a measure of confidence that the pay-off to the general industry should be positive and significant and 2) information about threshold levels of acreage and cost differential between hand and mechanical harvesting options to assure a positive NPV for the public investment into the abscission compound, CMNP.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by German Blanco.
Thesis: Thesis (M.S.)--University of Florida, 2009.
Local: Adviser: Roka, Fritz M.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2011-08-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2009
System ID: UFE0025050:00001


This item has the following downloads:


Full Text

PAGE 1

1 COST/BENEFIT ANALYSIS OF PUBLIC INVESTMENT INTO CMNP REGISTRATION AND DEVELOPMENT FOR CITRUS MECHANICAL HARVESTING By GERMAN BLANCO LOBO 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 2009

PAGE 2

2 2009 German Blanco Lobo

PAGE 3

3 To my family Fany and German Daniel

PAGE 4

4 ACKNOWLEDGMENTS First, I thank God for giving me life, and th e opportunity of completing my m asters degree. I thank my committee chair Dr. Fritz Roka for his mentoring, support, guidance and specially patience. I also want to thank the rest of my supervisory committee, Dr. Richard Kilmer and D r. Jackie Burns, for their time and support Tanks are due to the faculty and staff of the Food and Resource Economics Department of the University of Florida for their assistance. I would like to thank my parents, German Blanco and Norma Lobo, who have al ways supported me I also want to thank my brothers and sister, Juan, Aldo and Daniela. I thank all of my family members and friends who somehow helped me during this process. A special thank you to my wife Fany and my son German Daniel for being always en couraging, supporting and loving me.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS .................................................................................................................... 4 LIST OF TABLES ................................................................................................................................ 7 LIST OF FIGURES .............................................................................................................................. 9 ABSTRACT ........................................................................................................................................ 11 CHAPTER 1 INTRODUCTION ....................................................................................................................... 13 Problem Setting and General Problem ....................................................................................... 13 Research Question ....................................................................................................................... 16 Research Objectives .................................................................................................................... 16 2 REVIEW OF MECHANICAL HARVESTING SYSTEMS A ND THE ABSCISSION COMPOUND CMNP ................................................................................................................. 21 Current Mechanical Harvesting Systems ................................................................................... 21 Effect of Mechanically Harvesting Late -season V alencia O range ....................................... 22 Abscission Compounds and CMNP Studies ............................................................................. 25 Economics of Mechanization ..................................................................................................... 26 3 DEVELOPING A MODEL TO ANALYZE THE POTENTIAL EC ONOMIC BENEFIT OF CMNP APPLICATION ..................................................................................... 33 Outline of the Model ................................................................................................................... 34 Benefits ................................................................................................................................. 34 Costs ..................................................................................................................................... 35 NPV and Discount Factors .................................................................................................. 36 Benefit Equation Components .................................................................................................... 37 Value of Lost F ruit ............................................................................................................... 37 Percentage y ield reduction of late s e ason Valencia from mechanical h arvesting ...................................................................................................................... 37 Late season Valencia by harvest d ate ....................................................................... 38 Fruit p rice ...................................................................................................................... 39 Abscission (CMNP) a c reage c onstraint ....................................................................... 39 Savings in Harvest Cost ...................................................................................................... 41 Cost Equation Components ........................................................................................................ 41 Scenario Overview ...................................................................................................................... 42

PAGE 6

6 4 COST/BENEFIT ANALYSIS AND SCENARIOS RESULTS .............................................. 53 Scenario 1, High Price/Low Production and High % of Lost Fruit. ........................................ 54 Scenario 2, High Price/Low Production and Low % of Lost Fruit. ......................................... 55 Scenario 3, Low Price/High Production and High % of Lost Fruit. ........................................ 57 Scenario 4, Low Price/High P roduction and Low % of Lost Fruit. ......................................... 58 NPV Threshold. ........................................................................................................................... 59 5 SUMMARY IMPLICAT IONS AND CONCLUSIONS .......................................................... 69 Summary ...................................................................................................................................... 69 Implications and Conclus ions ..................................................................................................... 71 REFERENCE LIST ............................................................................................................................ 73 BIOGRAPHICAL SKETCH ............................................................................................................. 76

PAGE 7

7 LIST OF TABLES Table page 1 1 Compar ative costs in processing oranges in Florida and Sao Paulo, 200001. .................. 18 1 2 Total production and harvesting costs ($/box) for central Florida (Ridge) Valencia oranges under two cultural programs, 200607. ................................................................... 19 2 1 Average performance statistics of Coe Collier Trunk -Shake Catch (TSC) system, Oxbo Continuous Canopy Shake Catch (CCSC) system and Oxbo Tractor Drawn Canopy Shaker (T -CS) on early -mid and late season oranges, 200304 season. ............... 28 2 2 Spray concentrations of ABG 3030 required to reduce pull force to five pounds or less. .......................................................................................................................................... 30 2 3 Spray concentrations of ABG 3030 required to cause objectionable damage at various growth stages. ............................................................................................................ 30 3 1 List and definition of variables entering the benefit and cost equations. ........................... 44 3 2 Calculation of Valencia oranges as a percentage of the total Florida oranges for seasons 198586 through 200506, Florida production and processed Valenc ia oranges in thousand boxes. .................................................................................................... 45 3 3 Proportion of total Valencia production on the tree at each of the four harvest dates for seasons 198586 through 200506. .................................................................................. 46 3 4 FDOC all round orange projection and Valencia oranges production scenarios (million boxes) for seasons 200708 through 201718. ....................................................... 47 3 5 Projected late -season Valencia oranges by harvest dates, low and high production scenarios (million boxes), for seasons 200708 through 201718. ..................................... 47 3 6 Percentage difference between all round oranges and Valencia oranges on tree price per box ($/box) for seasons 198586 through 2005 06. ............................................. 48 3 7 Projected Valencia orange price scenarios ($/box), seasons 200708 through 201718 ............................................................................................................................................. 49 3 8 Percentage difference between all round oranges and Valencia oranges yield per acre for seasons 198586 through 200506. .......................................................................... 50 3 9 EUP acreage constraint converted to total boxes of Valencia orange for seasons 200708 through 201718 under two production scenarios (low and high). ...................... 51 3 10 Abscission registration and R&D expenditures (million $) from 1996 to 2008 discounted to 2008 dollars to determine initial inve stment in 2008. .................................. 51

PAGE 8

8 3 11 Scenario description in terms of price, production, percentage of lost fruit and EUP assumptions. ........................................................................................................................... 52 4 1 Scenario 1, high price/low production high fruit loss percentage. CMNP projected net cash flows (million $) from 200708 through 201718. ................................................ 62 4 2 Scenario 2, high price/low production low fruit loss percentage. CMNP projected net cash flows (million $) from 200708 through 201718. ................................................ 62 4 3 Scenario 3, low price/high production high fruit loss percentage. CMNP projected net cash flows (mil lion $) from 200708 through 201718. ................................................ 63 4 4 Scenario 4, low price/high production low fruit loss percentage CMNP projected net c ash flows (million $) from 200708 through 201718. ................................................ 63 4 5 Summary of scenario results, c umulative NPV ($), payback periods (years from 2008), and IRR (%), 1995 through 2018 .............................................................................. 67 4 6 Minimum acreage of CMNP application by selected unit harvest cost savings to achieve positive NPV for each scenario by the end of the 201718 season. ...................... 67 4 7 Minimum acreage of CMNP application to achieve a payback period within 6 years by selected unit harvest cost savings. .......................................................................... 68

PAGE 9

9 LIST OF FIGURES Figure page 1 1 Top citrus producing countries .............................................................................................. 17 1 2 Florida citrus harves ting season ............................................................................................ 17 1 3 Total production cost per acre ($) for processed oranges with and without the additional cultural management program for canker greening, from 200304 through 200607 ................................................................................................................................... 18 1 4 Citrus mechanically harvested acreage from 97 98 to 06 07 season .................................. 19 1 5 Presence of young growing fruit along with mature fruit on the same Valencia tree. .... 20 2 1 Trunk shake and catch s ystem, position of shaker (right) and receiver (left ) units ........... 28 2 2 Overhead view of the OXBO continuous canopy shake and catch s ystem. ....................... 29 2 3 Tractor drawn c an opy shake s ystem ..................................................................................... 29 2 4 Influence of date on subsequent yield of trees harvested with limb shaker measured as a percentage of the yield of hand-picked trees ................................................................. 30 2 5 The effect of CMNP and Ethrel on reduction of FDF in Valencia orange from 0 to 72 hours after abscission material application ................................................................. 31 2 6 Leaf a bscission induced by increasing concentrations of CMNP and Ethrel in calamondin 5 days after a pplication ...................................................................................... 31 2 7 Example of a verage total cost curve and economies of scale. ............................................ 32 2 8 Effect of canopy shaker ground speed and CMN P on percentage mature fruit removal in Valencia orange. Means followed by the same letter are not statistically different, p=0.05 ..................................................................................................................... 32 3 1 Yield reductions due to mechanically harvesting Valencia oranges, as compared to the handpicked control, by harvest date in 2004 and 2005 .................................................. 44 3 2 Confidence interval for the proportion of total Valencia production by harvest date. ... 46 4 1 Comp arison of CMNP net present values (2008 million $) under different percentage of lost fruit assumptions. ........................................................................................................ 64 4 2 Comparison of CMNP net present values (2008 million $) by price/production combination under different fruit loss assumptions. ............................................................ 64

PAGE 10

10 4 3 Impact of mechanical harvesting acreage during the late season and unit cost savings under scenario 1, high price/low production and high % of lost fruit. ............................... 65 4 4 Impact of mechanical harvesting acreage during the late season and unit cost savings under scenario 2, high price/low production and low % of lost fruit. ................................ 65 4 5 Impact of mechanical harvesting acreage during the late season and unit cost savings under scenario 3, low price/high production and high % of lost fruit. ............................... 66 4 6 Impact of mechanical harvesting acreage during the late season and unit cost savings under scenario 4, low price/high production and low % of lost fruit. ................................ 66

PAGE 11

11 Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Masters of Science COST/BENEFIT ANALYSIS OF PUBLIC INVESTMENT INTO CMNP REGISTRATION AND DEVELOPMENT FOR CITRUS MECHANICAL HARVESTING By German Blanco Lobo August 2009 Chair: Fritz M. Roka Major: Food and Resource Economics Maintaining global competitiveness and offsetting cost increments due to the incursion of new management strategies to combat citrus greening and canker are two reasons why the Florida citrus industry needs to find ways to reduce costs. Since harvesting costs play such an important role in citrus production, the Florida Department of Citrus (DOC) has dedicated a portion of grower taxes (public funds) to develop and register an abscission agent compound, CMNP, with the U.S. E nvironmental P rotection A gency as part of its effort to make citrus mechanical harvesting cost effective. Total cost of registration is estimated to be between $7 and $ 10 million. In addition, the University of Florida IFAS has spent between $500 and $700 thousand of public dollars annually since 1996 to research abscission compounds and develop management strategies. The primary objective of this research was to compare a discounted stream of private benefits of mechanical harvesting with abscission agent application against the public costs of abscission development and c hemical registration. The time period for this analysis spans 22 years, from 1996 to 2018. This research assumes that abscission agent application is the solution to harvesting late -season Valencia oranges. Without abscission agent application, mechani cal harvesting systems would have a negative impact on next years yield due to the removal of

PAGE 12

12 young immature fruit at the time of harvesting during the present year. Two benefits of CMNP with mechanical harvesting late -season Valencia are estimated: 1) harvest cost savings during the current year, and 2) preserving the value of fruit that would have been lost in the next season by mechanically harvesting during the current season without abscission. The net present value (NPV) of mechanically harvesting 25,000 acres (acreage limit by an anticipated Experimental Use Permit) through the 2018 season ranges between $30.62 and $ 50. 67 million, depending on price, production, and fruit loss scenarios. For any of the scenarios considered, the discounted value of abscission benefits with mechanical harvesting fully pays for the costs of registration and development within four years. A subsequent analysis of minimum acreage of CMNP application during the late season and levels of savings indicates that there exists a threshold level under which the NPV of this investment is not longer positive. Results from this analysis should provide: 1) a measure of confidence that the payoff to the general industry should be positive and significant and 2) information about thr eshold levels of acreage and cost differential between hand and mechanical harvesting options to assure a positive NPV for the public investment into the abscission compound, CMNP.

PAGE 13

13 CHAPTER 1 INTRODUCTION Problem Setting and General Problem In the 200607 season, Florida citrus industry total produc tion was 162.1 million boxes1 and accounted for 70 percent of t he total U.S. citrus production. T otal preliminary value of the 200607 Florida crop e qualed $ 1.36 billion (FASS, 2008). As a country, the United States ranks third behind Brazil and China for total citrus production (Figure 1 1). In terms of the orange juice market, Brazil and United States a re the major producing countries. During the 200708 season, Brazil produced about 5 6 % of the world's orange juice, while the United States produced 34% (For eign Agricultural Service, USDA, 2008) Florida growers produce a variety of citrus fruits inclu ding oranges, grapefruit, temples, tangerines, and tangelos within a harvesting s eason of 10 months (Figure 1 -2 ). Floridas subtropical climate and abundant water provides a comparative advantage for citrus production. However, l ong term economic sustainab ility of Floridas citrus industry depends on finding ways to reduce costs of production. Total cost of production, for a typical cultural program for processed oranges has been increasing steadily from 200304 through 200506. With the inclusion of new ma nagement strategies to combat citrus greening and citrus canker diseases, the estimated average total cost to produce Valencia oranges jumped from $973.79 to $1462.52 per acre (Figure 1 3; Muraro, 2007). Offsetting these costs increases is pushing the Fl orida citrus industry to utilize new technologies that could reduce production costs. 1 A box is a standard unit of measure for citrus fruit. One box of oranges weighs 90 pounds (40.8 kg).

PAGE 14

14 Another important reason to reduce costs has to do with maintaining global competitiveness. A comparison between Florida and Sao Paulo (Brazil) for the 200001 citrus se ason suggested that the largest cost gap between the two states lies in harvest costs, a difference of $1.569 in cost per harvested box (Table 1 1). Brazil enjoys a significant advantage in lower labor costs (Muraro et al., 2003). Nearly all of Florida s 6 21,373 commercial citrus acreage (FASS, 2008) is hand harvested. H arvesting costs exceed production cost s under typical cultural programs, and represent more than 70% when compared to the production costs of programs including canker -greening management (T able 1 2 ). Since 1995 the Florida Department of Citrus (DOC) has been investing grower taxes in a mechanical harvesting program. M echanical harvesting increases harvest labor productivity, and when combined with economies of scale there is a potential to reduce the unit cost of harvesting (Roka, 2007). Although citrus mechanically harvested acreage is increasing (Figure 1 4), only 35,600 acres were mechanically harvested during the 200607 season representing about 7% of the total citrus acreage (UF/IF AS, 2008). Rok a (2007) lists several significant impediments to the widespread adopt ion of mechanical harvesting systems, of which the most significant obstacle is harvesting late -season Valencia oranges [Citrus sinensis (L.) Osb.] Late season Valencia harvest is defined as harvest occurring sometime after May 1st. During a typical year, Valencia trees flower and set fruit during March. This crop is harvested 12 to 15 months later between March and mid June the following year Sometime durin g May of the current year, next years growing fruit reaches one inch in diameter (Figure 1 5). P revious studies estimate the fruit loss from mechanical harvesting with young fruit greater than one inch diameter to be between 20% and 50% of next years cro p ( Hedden and Coppock, 1971; Coppock et al., 1981; Roka et al., 2005; Whitney, 1975; Whitney and Hedden, 1973). Therefore,

PAGE 15

15 mechanical harvesting systems are forced to shut down sometime during May, leaving an estimated 25% to 30% of Floridas mature Valen cia crop on the tree ( FASS, 2005) with hand crews as the only means of harvesting. Cost efficiencies of mechanical harvesting could be enhanced if equipment could operate throughout the Valencia season. This requires keeping the removal rates of mature fruit at high percentage while at the same time having little or no impact on next years yield. A bscission compounds offer one solution to the late -season problem when sprayed on pre harvested fruit. An effective abscission compound should selectively l oosen this years mature fruit. With a reduction in the required force to remove fruit, harvesting equipment should be able to operate with less force, thereby preserving the young fruits on the tree for next years crop. One compound 5 -chloro 3 -methyl 4 -nitro 1H -pyrazole (CMNP) has attracted a significant amount of attention. CMNP loosens fruit over wide ranges of concentrations It is non phytotoxic and does not affect tree health or yield (Burns, 2002). CMNP application combined with mechanical harv esting systems is expected to improve overall efficiency throughout the season. By extending the season through selectively loosening of mature fruit after early May harvesting capacity of the machines increase s. U nit cost of harvesting should decrease. A s unit harvest costs decrease, growers will have more incentive to implement mechanical harvesting and the transition from hand harvesting to mechanical systems should accelerate. On the basis of this anticipated benefit, the Florida Department of Citrus has begun the formal EPA ( U.S. Environmental Protection Agency) registration process to register CMNP in 2005. An experimental use permit (EUP) is expected by spring of 2010. If no problems are identified during the two years under EUP provisions, full regi stration should be completed by 2012.

PAGE 16

16 Research Question Since 1995 the Florida Department of Citrus (DOC) and the state of Florida through the University of Florida/IFAS have invested more than $7 million of public funds into the research and development of abscission compounds. In addition DOC has spent more than $750 thousand on the formal CMNP registration process a s of July 1, 2007. Grants from the federal government have provided $2.054 million toward CMNP registration. An estimated $6.5 million of a dditional funds will be required to complete the full registration process, which is anticipated by spring 2012. Under the assumption that abscission is the solution to mechanically harvest Valencia oranges during the late season this research examine s the following question: Does the private benefit of extending mechanical harvest ing into late May and June with the abscission compound CMNP more than offset the public cost s to develop and regist er CMNP ? Research Objectives The overall objective of thi s research is to estimate the future private economic benefit s f r o m the application of the abscission compound CMNP, which should allow mechanical harvesting to proceed through the Valencia orange late -season, specifically after mid May. In order to me et this general objective, the following specific objectives need to be achieved: 1 Estimate the value of the expected yield reduction i n next years crop if la te season Valencia oranges are mechanically harvested this year without CMNP application 2 Estima te the savings in harvest cost by extending the mechanic al harvesting season into late May and June with CMNP 3 Estimate the total cost of development and registration of CMNP. Results from objectives 1, 2 and 3 will be incorporated into a spreadsheet mode l S tatic future production and price expectations will be developed to determine the net present value (NPV) of abscission compound application for mechanically harvesting late season Valencia oranges (i.e. after early -May).

PAGE 17

17 Figure 1 1. Top citrus producing countries. Source: Foreign Agricultural Service, USDA, 2008. Figure 1 2 Florida citrus harvesting season. Source: Florida Agricultural Statistics Service, USDA, 2008.

PAGE 18

18 Figure 1 3. Total production cost per acre ($) for processed or anges with and without the additional cultural management program for canker greening, from 200304 through 200607. Source: Muraro R. P., 2004, 2005, 2006 and 2007. Table 1 1 Comparative cost s in processing oranges in Florida and Sao Paulo, 2000 01. Costs $/box Florida Sao Paulo Difference Total production 1.874 1.378 0.496 Total harvesting 2.103 0.534 1.569 Assessments and other 1.125 0.768 0.357 Total delivered in 5.102 2.680 2.422 Processing 1.322 1.015 0.307 Domestic N/A 0.175 0.175 Foreign N/A 2.514 2.514 Total FOB costs 6.424 6.384 0.040 Source: Muraro et al. 2003. $0 $200 $400 $600 $800 $1,000 $1,200 $1,400 $1,600 200304 200405 200506 200607 Cultural program Canker greening

PAGE 19

19 Table 1 2. Total production and harvesting costs ($/box) for central Florida (Ridge) Valencia oranges under two cultural programs, 200607. Without Canker Greening With Canker Greening Weed management 0.459 0.526 Pest management 0.368 1.069 Fertilizer 0.489 0.561 Pruning 0.066 0.076 Tree replacement 0.209 0.505 Irrigation 0.390 0.447 Mandatory canker decontamination/other 0.085 0.315 Total production/cultural costs 2.066 3.499 Total Harvesting Picking, roadsiding, Hauling and Canker decontamination 2.561 2.561 Total Costs 4.627 6.060 Source: Muraro R. P., 2007. Figure 1 4 Citrus mechanically harvested acreage from 97 98 to 0607 season. Source: UF/IFAS, 2008a.

PAGE 20

20 Figure 1 5. Presence of young growing fruit along with mature fruit on the same Valencia tree Source: UF/IFAS, 2008b.

PAGE 21

21 CHAPTER 2 REVIEW OF MECHANICAL HARVESTING SYSTEMS A ND THE ABSCISSION COMPOUND CMNP Problems associated with harvesting costs and labor shortages to meet steadily increasing citrus production motivated the Florida citrus industry to establish a mechanical harvesting and abscission research project Early r esearch and development efforts into citrus mechanical harvesting spanned four decades, from the late 1950s into the early 1980s (Coppock and Hedden, 1977; Wilson et al., 1977; Whitney and Sumner, 1977; Sumner and Churchill, 1977; Hedden and Coppoc k, 1977). Mechanic al harvesting R&D (research and development) efforts stopped during the mid 1980s for several reasons such as limited production due to freezes, abundant la bor supply at a reasonable cost and relatively hig h fruit prices. Also numerous problems were encou ntered with mechanical harvesting, such as low harvesting efficiencies with Valencia oranges (Whitney, 1995). In 1994, the Florida Department of C itrus initiated a second mechanical harvesting program Harvesting costs were a major concern because of high production levels, low labor availability and low fruit prices due to foreign competition. During the mid 1990s harvesting costs were almost equal to the cost of production. These conditions motivated the Florida citrus industry to renew mechanical h arvesting R&D efforts to determine if future crops could be harvested at a competitive level (Whitney et al., 1996). Current Mechanical Harvesting Systems Roka and Hyman (2004) evaluated the performance of three citrus mechanical harvesting systems for t he 200304 season (Table 2 1). The data collected represents more than 2,750 acres and nearly 1.25 million boxes of harvested oranges. The average performance by machine was calculated for: removal percentage, recovery percentage, harvest speed, machine pr oductivity,

PAGE 22

22 labor productivity, and percent runtime. The three systems evaluated in this study (Roka and Hyman, 2004) were: Trunk shake and catch (TSC Figure 2 1). This system shakes the trunk of a tree and removes on average 95% of the fruit. Between 85% and 93% of the available fruit is collected by the catch frame and labor productivity averages nearly 100 boxes/ hour per person2. Continuous canopy shake and catch (CCSC Figure 2 2 ). Two harvesting units work in tandem on either s ide of the tree row. The s haker heads remove 95% of the fruit, and overall recovery averages 91%. Although each set of harvesters requires 6 people compared to 3 for TSC, labor productivity for CCSC systems average close to 100 boxes/hour per person. The tractor drawn canopy shak er (T CS Figure 2 3) unit is similar to the CCSC only that it is not self propelled, but instead drawn by a tractor and harvest one side of the tree at a time. The harvesting unit removes nearly 95% of the fruit. Fruit falls to the ground where a pick up crew recovers almost a 100% of the fruit available. Labor productivity is much lower than the other systems with catch -frames. However, given the fact that much of the data collected on T SC systems came from harvesting large, widely spaced trees, average labor productivity of 12 boxes per hour may represent a 2 -fold increase in hand -harvest productivity in the same tree conditions (Roka, 2008) For a more de tailed description of the above mentioned mechanical harvesting systems you may visit the UF/IFAS Citrus Mechanical Harvesting web page at http:// citrusmh.ifas.ufl.edu/index Effect of Mechanically Har vesting Late -season Valencia O range In the early 1960s, r esearch was mainly directed toward s harvesting processed oranges which em ployed the vast majority of the harvesting labor. Development of limb, air, trunk, and foliage (canopy) shakers began without the fruit loosening aid of abscission chemicals. A five year study by Hedden and Coppock (1968) showed that early and mid-season v arieties yields were not affected significant ly by a 15 -cm stroke limb shaker. Y ields of Valencia oranges however, were reduced up to 40% in the subsequent year when limb shaking occurred during late May and early June 2 Average productivity of hand harvesters is between 8 and 10 boxes per hour (Roka and Emerson, 1999).

PAGE 23

23 With no abscission chemicals avai lable to loosen mature Valencia oranges, shaking strokes of 10, 15 and 20 cm at different frequencies were investigated as shaker adjustments were made to minimized young fruit removal without compromising mature fruit removal efficiencies. Subsequent yi eld reductions over the harvest season ranged from 8 % to 28% at an average mature fruit rem oval of 81% (Coppock, 1971). The s e field trials concluded that Valencia oranges can be mechanically harvested without abscission until the young fruit is about 0.5 0 i nch in diameter, without significant reduction in yield during the subsequent season Coppock (1972) evaluated the influence of date on the diameter of young fruit for three harvest seasons. He determined that the diameter of young fruit incr eased a ccording to the linear equation Y = 0.472X 0.69, (2 1 ) where X is the number of days after April 1 Around May 15 or when the young fruit reach 22 mm in diameter (0.8 inches ), several changes begin to occur in young fruit specifically you ng fruit weight and detachment force, which have a marked influence on harvest ing with limb shakers. In his study Coppock also estimated that the subsequent yield of shaken trees ra n ged from 8 0% percent at the beginning of the season to 94% percent about May 15 after which it decreased rapidly (Figure 2 4 ). Several experiments were conducted to determine the eff ect of mechanical harvesters on subsequent Valencia orange yield with the use of abscission chemicals Abscission chemicals affect the fruit removal percentage by loosening mature fruit, thus improving mechanical harvesting efficiency. Mechanical harvesting efficiency was defined by Whitney (1976) as the percentage of crop that could be recovered mechanically compared to manual harve sting. For Valencia oranges the combination of sufficient mature fruit removal with large yield reductions

PAGE 24

24 resulted in low harvesting efficiencies. In 1976, Whitney calculated harvesting efficiencies of the AREC (Agricultural Research and Education Cente r) air shaker with 2 different concentrations of the absc ission compound Release (5 -chloro 3 -methyl 4 nitro 1H pyrazol) on Valencia oranges when the young fruit diameter averaged 1.2 inches. Harvesting efficiency of the shaker was 80 percent with 250 ppm and 78 percent with 375 ppm Release. No significant reduction in next years yield was calculated when spraying 250 ppm Release. Coppock et al. (1985) conducted an experiment to determine the effect of three shaker strokes (15.2, 20.3 and 25.4 cm) with and without the Release abscission compound Subsequent fruit yields were reduced an average of 15 percent over the corresponding initial yield. Although not statistically significant, there was an indication that fruit yield reduction was less with the l ess aggressive shaker s trokes. M ature fruit removal with abscission averaged 96.5 percent versus 93.1 percent for treatments without the compound. T hey also concluded that the abscission chemical had little effect on selectivity of mature fruit when combin ed with the more aggressive shaker stroke treatment. Overall, the most efficient treatment was combining abscission compound application with the 20.3 cm stroke ; a harvesting efficiency of 87.5 percent was calculated for this treatment. Burns et al. (2006) suggested that the appropriate equipment operation parameters, in combination with CMNP application significantly influenced mature fruit removal. Differences were greater and statistically significant (p>0.01) for the treatment with 4 seconds shake durat ion at half of the full throttle (4.8 Hz.). Such treatment significantly decreased immature fruit removal and did not signif icantly reduce subsequent yield Therefore it was concluded that the applic ation of the abscission compound CMNP combined with low f requency trunk shaker

PAGE 25

25 harvesting can achieve a high percentage of mature fruit removal with no significant impact on yield of next year crop. Abscission Compounds and CMNP S tudies Several abscission chemicals were developed and screened dur in g the early abscission program (1960s to 1980s ). A few compounds emerged as viable abscission agents and after years of testing, one compound, ABG 3030, was identified as a superior compound Initially identified by Abbott Laboratories, North Chicago, IL, USA, this compound was formulated as Release. The activity of ABG 3030 was first detected in January, 1971 (Wilson, 1973) and additional testing during the Valencia orange season was conducted to demonstrate the effectiveness of ABG 3030 as an abscissi on agent. Kenney et al. (1974) suggested that early and mid season varieties are more sensitive to the activity of ABG 3030 than the Valencia variety. This wa s reflected in the concentration range required to reduce the pull force to 5.0 pounds or less (Table 2 2). Unfortunately, m ature fruit burn occurred on all varieties. With respect to the phytotoxicity, Kenney et al. (1974) summarized the effect of ABG 3030 on the various stages of development foliage and immature fruit of the Valencia variety (Table2 3) and concluded that the margin between the effective concentration and the concentration that causes significant adverse effect wa s large. Several experimental use permits were granted by EPA for the active ingredient in Release 5 -chloro 3 -methyl 4 nitro 1H pyrazole (CMNP, initially called ABG 3030) Although the material was tested in Florida for several years (Biggs and Kossuth, 1980; Wilson et al., 1977), the product was never formally registered as a fruit abscission agent. S everal reasons explain why enthusiasm waned: high cost for full registration of Release, high fruit prices, adequate labor availability, and initial indications of problematic toxicology for the activ e ingredient.

PAGE 26

26 Burns (2002) screen ed several abscission materials for citrus including CMNP. Her work tested different concentrations of CMNP to measure selectiveness and phytotoxicity. Applications of 200 mg.L1 (ppm) of CMNP to Valencia trees resulted in a reduction in FDF (fruit detachment force) 24 hours a fter treatment. FDF continued to decline afte r the first 24 hours (Figure 2 5 ). She also documented that leaf drop in Valencia was generally unaffected by CMNP treatment at recommended rates. Leaf drop was low at 0 to 1000 mg.L1 of CMNP, but rapidly inc reased at concentrations greater than 2000 mg.L1 (Figure 2 6 ). S everal recent studies (Burns, 2002; Burns et al., 2005; Burns et al. 2006) confirmed that CMNP is a selective abscission agent and that it loosens mature fruit within 3 to 5 days after appli cation. Despite the high concentration of CMNP applied during these tests, tree s did not exhibit phytotoxicity; however, fruit removal rates significantly increased when combining CMNP and mechanical harvesters. Although there are clear advantages to using CMNP, Burns et al. (2005) indicated that more work is needed to determine the concentr ation needed to maximize removal and recovery and hence grower advantage Economics of Mechanization According to Samuelson (1970) i ncreasing returns to scale, or so -called economies of mass production, are often associated with one of the following advances: (1) the use of nonhuman and non animal power sources; (2) the use of automatic self adjusting mechanism; and (3) the specialization of function and division of labor. Unit cost savings accrue only if a sufficient number of units are being produced. Consider the general shape of a long run average total cost curve as shown in Figure 2 7. When the long run average cost curve declines as output level increases from Q1 to Q2, there are said to be economies of scale, the average cost will go down from C1 to C2 which it can be translat ed, in the case of the citrus industry, into lowering the unit cost per box

PAGE 27

27 Economies of scale in citrus mechanical harvesting systems can be achieved by increasing their capacity. Capacity is measured as boxes per season. As capacity in a harvest season increases, fixed costs are spread over more harvested boxes, thus, lowering the unit cost to harvest. With CMNP application, the capacity of a mechanical harvester can be improve (Burns et al., 2005). The primary effect of CMNP application is extending the mechanical harvesting season beyond mid May, thus, allowing these systems to access a significant portion of the Valencia crop. More than 40% of the Valencia crop is harvested after May 1st. CMNP could also have a positive impact on boxes per hour (Bx /hr). Boxes per hour can be improved by increasing recovery percentage. An abscission compound can influence the Bx/hr by allowing the equipment to harvest faster without compromising fruit removal rates. Burns et al. (2005) calculated that harvest speeds can be increased by 25% with CMNP while maintaining high percent mature fruit removal (Figure 28). With CMNP application, less machine force will be necessary to achieve high rates of mature fruit removal, thus minimizing the possibilities of machine wea r and breakdown (Roka, 2006). Decreasing machine breakdowns has a positive impact on runtime percentage which in turn, improves equipment efficiency. Higher system efficiencies will result in higher labor productivity or boxes per hour per crew member, the refore, increasing the number of boxes harvested in 8 hours/day.

PAGE 28

28 Table 2 1. Average performance statistics of Coe Collier Trunk Shake Catch (TSC) system, Oxbo Continuous Canopy Shake Catch (CCSC) system and Oxbo Tractor Drawn Canopy Shaker (T -CS) on early -mid and late season oranges, 200304 season. TSC (Coe Collier) CCSC (Oxbo) T CS (Oxbo) Measure Units Early/ mids Late season Early/ mids Late season Early/ mids Late season Removal % 95 97 94 95 91 94 Recovery % 86 88 88 91 99.5 99 System efficiency (% runtime) % 66 51 60 78 Harvest speed tree/hr 209 261 481 527 200 450 Machine productivity boxes/hr 420 512 968 1223 Crew size # 2.7 2.6 6 6 23 27 Labor productivity boxes/hr 108 101 100 169 7 12 Source: Roka and Hyman, 2004. Figure 2 1 Trunk shake and catch s ystem, position of shaker (right) and receiver (left) units. Source: UF/I FAS 2008a

PAGE 29

29 Figure 2 2 Overhead view of the OXBO continuous canopy shake and catch s ystem. Source: UF/IFAS, 2008a Figure 2 3 Tractor drawn canopy shake s ystem. Source: UF/IFAS, 2008a

PAGE 30

30 Figure 2 4 Influence of date on subsequent yield of trees harvested with limb shaker measured as a percentage of the yield of hand-picked trees Source: Coppock, 1972. Table 2 2. Spray concentrations of A BG 3030 required to reduce pull force to five pounds or less. Variety Concentration (ppm) Hamlin 50 100 Parson Brown 50100 Pineapple 25 75 Valencia 150 200 Source: Kenney et al. 1974. Table 2 3. Spray concentrations of ABG 3030 required to cause objectionable damage at various growth stages. Growth stage (Valencia) Concentration (ppm) Feather flush >800 Six inch flush >800 3 month leaves 800 1 year or older leaves 400 Pin head bloom >800 Full bloom >800 Immature fruit (all stages) >800 Source: Kenney et al. 1974.

PAGE 31

31 Figure 2 5 The effect of CMNP and Ethrel on reduction of FDF in Valencia orange from 0 to 72 hours after abscission material application. Source: Burns, 2002. Figure 2 6 Leaf a bscission induced by increasing concentrations of CMNP and Ethrel in calamondin 5 days after application. Source: Burns, 2002.

PAGE 32

32 C 1 C 2 Q 1 Q 2 Average Total Cost (ATC) (Unit cost per box) Output (Number of boxes) Economies of Scale Long run ATC Figure 2 7. Example of a verage total cost curve and economies of scale. Figure 2 8.Effect of canopy shaker ground speed and CMNP on percentage mature fruit removal in Valencia orange. Means followed by the same letter are not statistically different, p=0.05. Source: Burns, et al., 2005.

PAGE 33

33 CHAPTER 3 DEVELOPING A MODEL TO ANALYZE THE POTENTIAL ECONOMIC BENEFIT OF CMNP APPLICATION This chapter present s a model, its key components ( equations ) data sources and methodology to de rive cost / benefit values for developing and registering the a bscission compound CMNP. Effective use of CMNP should allow citrus mechanical harvesting to extend through the Valencia season without adversely affecting fruit yields in the following season. Increasing the hours of equipment operation should increase harvest capacity of the harvesting s ystems and thus, lower unit costs of harvesting. The model compares projections of future private benefits with known and projected public expenditures The model evaluates four scenarios to examine the sensitivity of future benefits under various levels of price, production, and fruit loss The economic success for the CMNP public investment is defined as a positive net present value (NPV) of net cash flows over a 10 -year planning horizon, discounted at an interest rate of 10% (Moss et al., 1991) The cost/benefit analysis proceeds in five steps: 1) identify relevant costs and benefits; 2) measure past and future costs related to abscission compound registration and development; 3) estimate value of future benefits from abscission agent application during the late season Valencia harvest; 4) calculate the performance of this public investment by comparing cost and benefit streams accruing during the pre -defined lifetime for this project (10 years) in constant dollars (NPV); and 5) conduct a sensiti vity analysis of key assumptions embedded in the cost/benefit analysis. The model consists of two equations, one estimating economic benefits and the other quantifying cost of developing and registering CMNP for commercial use. All the variables included in these equations are explained in detail in the following sections. Economic viability of CMNP will be evaluated by calculating the NPV of accrued costs since 1995 and anticipated

PAGE 34

34 benefits starting in 2010 and continuing through the 2018 season. A paybac k period and the internal rate of return (IRR) will be calculated as additional measures of financial performance. Outline of th e Model Benefits Two benefits of abscission are quantified in the model. First, t he dollar value of fruit that would have been l ost in the next season if the crop were to be mechanically harvested in the current season without abscission compound application. Second, the harvest cost savings earned by a mechanical system in the current season with abscission compound application. A ll the variables used in the following equations are listed and defined in Table 3 1. Equation 3 1 estimates a dollar value of lost fruit ( VA) in the season following mechanical harvesting of late season Valencia oranges without abscission compound appl ication This value equals the expected future fruit price ( Pt+1) multiplied by the projected quantity of Valencia oranges during the specified harvest dates ( vls tQ1 ). This value, in turn, is multiplied by the percentage of lost fruit ( ), to account for the proportion that would be adversely affected by mechanically harvesting during the previous Valencia late season without using abscission (i.e. after early May). The ( t ) represents the area (acres) of abscission application, and hence the area of late season Valencia that will be mechanically harvested. t vls t t T a t AQ P V *1 1 (3 1) A second benefit of CMNP is the cost savings (SA) generated by mechanically harvest ing the additional boxes of late season Valencia oranges preserved due abscission application compound. H arvest cost savings (Equation 3 2) equals the unit cost savings multiplied by the

PAGE 35

35 projected number of preserved boxes mechanically harvested with abscission during ear ly -May, mid -May, early June and mid June ) ( *t t t vls t t T a t AA MH HH Q S (3 2) where ( *vls t tQ ) corresponds to the number of additional boxes harvested during the current season as a result of abscission compound application. T he dollar per box unit cost savings is calculated as the difference between the cost per box to hand harvest ( HHt) and the combined cost to mechanically harvest ( MHt) with abscission compound application ( At). T he calculation of total economic benefit of abscission (BA) (Eq uation 3 3), for a given year t is the sum of the value of lost fruit (3 1) with the total harvest cost savings generated by mechanically harvesting Valencia oranges with abscission after the specified dates (3 2). T a t A A AS V B ) ( (3 3) )) ( ( ) (/ cos / cos / cos 1 1 bx t bx t bx t vls t t t vls t t T a tA MH HH Q Q P Costs Equation 3 4 calculate s the cumulative CMNP expenditures (CA) over 22 years from 1996 to 2018. Abscission expenditures are divided into two categories. The first category includes expenditures from both the Florida Department of Citrus (FDOC) and federal grants to register the compound CMNP with the U.S. EPA ( Regt). The second category includes abscission research and development expenditures by FDOC ( FDOC tD R & ), and UF/IFAS ( UF tD R& ). For the purposes of this analysis, abscission R&D and CMNP registration costs were totaled from 1996 and are expected to continue until CMNP is fully registered for commercial

PAGE 36

36 use, expected in 2012. In addition, research and development costs included only direct ex penditures and do not include faculty salaries and university overhead costs. UF t FDOC t t T a t T a t t AD R D R Reg C C & & (3 4) where, time t=a represents the period of analysis that is from 1996, where a= -12, through 2018, where a= 10, with 2008 as the base year, where a=0 NPV and Discount Factors A net present value (NPV) was calculated for each scenario evaluated in this study. The calculation of this appraisal method is given in equation 3 5 (Brigham and Ehrhardt, 2005): T a t t tr CF NPV ) 1 ( (3 5) where, CFt i s the expected net cash flow at period t r is the discount rate and n corresponds to the projects planning horizon. CFt is calculated as the difference between benefits of CMNP (inflows, equation 3 3) and total costs of CMNP (outflows, equation 3 4), during the tth year. Incorporating equation 3 5, benefits and costs are rewritten as: T a t t A A Ar S V B ) 1 ( (3 6) 1 / cos / cos / cos 1 1 1) 1 ( ) ( ) 1 ( *t bx t bx t bx t vls t t t t vls t t T a tr A MH HH Q r Q P and t UF t FDOC t t T a t T a t t t Ar D R D R Reg r C C ) 1 ( & & ) 1 ( (3 7)

PAGE 37

37 Benefit Equation Components Value of Lost F ruit The value of lost fruit is calculated as the dollar value of the anticipated yield reduction on next years Valencia crop from mechanicall y harvesting this years crop without abscission. The value of lost fruit is estimated by: 1) percent yield reduction in next years crop, 2) forecasted quantity of next years late season fruit and 3) forecasted on -tree fruit price during next season. Th e calculated value of lost fruit was constrained by the area over which CMNP will be applied. All the data were tabulated and organized within an Excel spreadsheet. Averages, proportions, mean values and their respective confidence intervals were calculate d. Percentage y ield r eduction of l ate s eason Valencia from m echanical h arvesting For the purposes of this paper, late season Valencia is defined as oranges harvest ed after May 1. This definition corresponds to a field trial that was conducted in a commercial grove south of Immokalee from 20032005 during four harvest periods (earlyMay, mid May, early June, and mid June). Two years of yield impact data were collected during the trial and a description of the trial is reported in Roka et al. (2005). Treatments included both canopy and trunk shakers operating at three intensities and without abscission compound application Treatments were applied in the first and second year of the trial. In the second and third year of the trial, yield s were compared across all harvest dates and among mechanical ly harvest ed and handpicked control plots Results from this trial are summarized in Figure 3 1. Significant production losses in 2004 were attributable to the mechanical harvesting treatments applied during 2003. Losses increased from 15% for early May harvest date to nearly 50% for the mid June harvest date. In 2005, however, yield reductions resulting from mechanical harvesti ng during the two May 2004 harvest dates were not significantly different from hand harvested plots. Yields from trees

PAGE 38

38 mechanically harvested during early and mid June 2004 were 15 and 20 percent lower than hand harvested plots. Two cost/benefit scenarios are developed around these results. A high loss scenario (H%) assumes that yield reductions occur at each harvest date, starting with a 15 % yield reduction from an early May harvest and increasing to 20 % 45% and 50 % for late May, early June, and mid June dates, respectively. A low loss scenario (L%) assumes no yield reductions during the May harvest dates. Yield reductions of 15% and 20% occur for the early and mid June harvest dates, respectively Late s eason Valencia by h arvest d ate The expected quantities of Valencia oranges harvested during the four late -season dates were estimated by analyzing the historical proportion of the total Florida orange crop that represents processed Valencia oranges, and then determining the proporti on of Valencia production harvested in early May, mid May, early June, and mid June. The harvest date percentages were used to apportion forecasted Valencia production between 2008 and 2018 by the four late season harvest dates. The percentage of tota l Valencia processed boxes harvested during a typical year was estimated using the historical FASS data from 1985 86 through 200506 (Table 3 2). On average Valencia oranges for the processed market represent 40.74% of the total orange crop in Florida. A 95% confidence interval reflects a lower and upper bound of 39.31% and 42.17%, respectively. Table 3 3 utilizes historical data from crop years 198586 through 200506 to estimate the proportion of Valencia oranges that have been harvested by early M ay, mid May, early June, and mid June (Florida Agricultural Statistics Service (FASS)). Figure 3 2 presents the mean value and the 95% confidence interval for these proportions. On average 43.50% of total

PAGE 39

39 Valencia crop is still on the tree during the la te season period; 30.65% is allocated during early and mid May leaving the final 12.85% to be harvested during June harvest dates. Florida Department of Citrus (FDOC) economists annually project the next ten years of citrus production (Brown, 2007). High and low production trends reflect varying assumptions about the future affect of pests and diseases on total fruit production. FDOC production projections are for total oranges, grapefruit, and specialty fruit. In order to differentiate late season Vale ncia oranges, the previously calculated percentages of Valencia oranges were applied to the FDOC estimates of Florida orange production for 2007 08 through 201718 (Table 3 4). Future Valencia production is allocated to one of the four harvest dates c onsidered in this study (Table 3 5) according to the percentages presented in Table (3 3). Fruit p rice During most harvest seasons, Valencia oranges command a higher delivered-in price than early and mid season oranges. A percentage differe ntial between Valencia and all round oranges on -tree price per box was estimate d from historical price data (FASS) from 1985 86 through 200506 season (Table 3 6 ). ANOVA was used to determine the mean value of the price differential and its respective co nfidence intervals. A mean value of 16.9 % reflects the historical price premium paid for Valencia oranges as compared to all round oranges. Future ontree Valencia prices were estimated over a 10 -year planning horizon starting in the 200708. Future prices for all Florida oranges were estimated by the Florida Department of Citrus ( Brown 2007) from 200708 through 201718. The h istorical pric e differential w as applied to the FDOC projected on tree price per box fo r all Florida oranges (Table 3 7 ). Abscission ( CMNP ) a creage c onstrain t FDOC will submit an application package to the U.S. EPA for the registration of CMNP as an abscission compound to be used as part of citrus harvesting. As part of that package, FDOC

PAGE 40

40 will be asking for an experimental us e permit (EUP) that will allow an expansion of the available acreage to receive CMNP material. FDOC is expected to request an EUP for 25,000 acres to be mechanically harvested with CMNP (Roka, 2008). If the EPA review process finds little or minimal risk f rom CMNP use, the EPA should grant an EUP by spring of 2010. The acreage amount of the EUP is at the discretion of the EPA. A major assumption of this model is that the entire 25,000 acre request is granted and utilized immediately for CMNP application dur ing the late -season Valencia harvesting period. It is important to note that once CMNP reaches full registration (perhaps as early as 2012), the EUP acreage is no longer going to constrain usage of this compound. For the purposes of this analysis, howeve r, it was assumed that only late season Valencia orange acreage will be treated with CMNP and total area will be limited by the EUP over the next 10 years. An average yield per acre was calculated for Valencia oranges. This value was applied to the acreage constraint to project the number of boxes that will be available for mechanical harvesting with abscission during the late season ( vls t tQ ). FASS production data from 198586 through 200506 season was used to calculate an average yiel d per acre for Valencia oranges. Valencia blocks averaged 305.09 boxes per acre, a 12.61% reduction in average yield per acre from all round oranges (Table 3 8). This percentage differential was applied to the FDOC projected yield per acre for all roun d oranges for seasons 200708 through 201718 and, converted to boxes by multiplying the number of acres allowed by EUP (Table 3 9). The model allocates CMNP on the basis of one harvest date at a time. That is, if the entire EUP acreage allotment, converte d to boxes, can be consumed during the first harvest date where abscission is needed to eliminate a subsequent reduction in yield (i.e. early May in scenarios with high

PAGE 41

41 percentage of yield reduction), then no mechanical harvesting with CMNP will be applied during the rest of the season. Savings in Harvest Cost Preserving the value of next years fruit with abscission is a benefit that accrues next year. Savings in the unit cost of harvesting with mechanical systems is a benefit that accrues during the curre nt season. Unit cost savings were calculated by subtracting the mechanical harvesting and CMNP application cost from hand harvesting rates per box, $1.50, $0.20 and $2.50 respectively (Roka, 2008). Total savings in harvesting cost were calculated by applyi ng this cost differential of $0.80 to the total number of preserved boxes of Valencia oranges resulting from mechanical harvesting with abscission during the late season once the EPA grants an EUP ( *vls t tQ ). Cost Equation Components Expenditures on abscission include costs to register CMNP with the U.S. EPA and expenditures on research and development. Expenditures on registration began in 2004 with federal grants. Federal government funding stopped in 2006 and provided a total of $2. 054 million toward CMNP registration. FDOC expenditures on CMNP registration for the 200607 season were $759,642 and they have projected that an additional $6.5 million will be required to complete the full registration process. Even though, more than $500,000 were expended between 1960 and 1975 during an earlier abscission program (King and Roka, 2008), this model only considers FDOC funding on abscission R&D that started in 1996. By 2005, almost $4.00 million have been funded by the FDOC on R&D. The Flor ida legislature passed the Citrus Initiative (CI) that allocated funds to the University of Florida/IFAS for mechanical harvesting, abscission research and education

PAGE 42

42 efforts. Since July 1, 2005, the first year of the CI, UF/IFAS has expended $1.32 million on R&D of abscission compounds. A summary of actual R&D and registration costs and the models anticipated costs through 2012 are presented in Table 3 10. The cost / benefit analysis in this paper uses the 200708 season as the initial reference point, that is t = 0 during 200708 season. Funds spent from 1996 through 2007 were added and appropriately inflated to reflect the time -value of money. The model applied the same interest rate (r), used to discount future cash flows, as well as to calculate the present value of past costs Scenario Overview Previous sections describe how benefits and costs of extending the Valencia late season with mechanical harvesting and abscission compound application were determined. Four scenarios were developed to determine if the private benefit of extending mechanical harvesting into late May and June with abscission compound application more than offset its public cost of development and registration The scenarios, described below, examined changes in price, production, p ercentage of lost fruit and EUP acreage constraint (Table 3 11). In scenarios 1 and 2, high price levels are linked to low production levels, and their projections suggest an increase in price from $7.88 /box in 200708 season to $12.13 /box in 201718 se ason (Table 3 7). Production levels (Low) for 200708 season start at 74.55 million boxes and decrease to 61.92 million boxes of Valencia orange by 201718 (Table 3 4). On average, 43.50% of the total projected boxes of Valencia oranges will be allocat ed in the late season and apportioned during four harvest dates (Table 3 5). Scenarios 3 and 4 link low price levels with high projected production levels. Prices increase from $7.56 /box in 200708 season to $8.90 /box in 201718 season (Table 37). P roduction levels (High) for 200708 season start in 76.18 million boxes and decrease to 70.89 million boxes of Valencia orange by 201718 (Table 3 4). The proportion of total Valencia

PAGE 43

43 used to calculate the projected production during the late season remains the same for these scenarios, 43.50% out of the total projected boxes of Valencia oranges. Late season Valencia acreage that can be mechanically harvested with abscission agent applic ation is constrained by the EUP to 25,000 acres. Low and High production scenarios, however, imply different per acre yields. Hence, the number of boxes mechanically harvested will vary by production scenario. Under low production levels (scenarios 1 and 2) the number of boxes mechanically harvested ranges from 8.6 1 to 7.38 million boxes for seasons 200910 through 201718 (Table 3 9). Under high production levels (scenarios 3 and 4) mechanically harvested boxes ranges from 8.59 million boxes in 200910 to 7.67 million boxes in 201718. Based on the Valencia proje cted production by harvest date depicted in Table 3 4, early May production decreases from 11.99 million boxes in 200708 season to 9.96 million boxes in 201718 season. Therefore, scenarios assuming a high percentage of lost fruit will utilize the entire EUP allotment during early May. Scenarios assuming a low percentage of lost fruit will allocate the EUP allotment during early June and for part of the mid June harvest date. Projected net cash flows over a 10-year planning horizon for CMNP were discounte d using a 10% rate. Registration and R&D expenditures from 1996 to 2007 were combined and discounted using Equation 3 5, to 200708 dollars for all four scenarios, totaling an initial investment of $19.10 million (Table 310). Also, expenditures on registr ation and R&D from 200809 through 201112 will not vary from one scenario to another.

PAGE 44

44 Table 3 1. List and definition of variables entering the benefit and cost equations. Variable Description Variable Description V A Value of abscission C A CMNP total expenditure P t + 1 On tree price Regt Registration cost VLS tQ1 Boxes of Valencia late season FDOC tD R & R&D sponsored by FDOC Percentage of loss fruit UF tD R & R&D sponsored by UF/IFAS t acreage mechanically harvested r Discount rate HHcost/bx Hand harvesting unit cost t Time period MHcost/bx Mechanical harvesting unit cost Acost/bx CMNP application unit cost Figure 3 1 Yield reductions due to mechanically harvesting Valencia oranges, as compared to the handpicked control, by harvest date in 2004 and 2005. Source: Roka, et al. 2005.

PAGE 45

45 Table 3 2. Calculation of Valencia oranges as a percentage of the total Florida oranges for seasons 198586 through 200506, Florida production and processed Valencia oranges in thousand boxes. Season Florida production (1,000 boxes) Processed Valencia (1,000 boxes) Valencia Proportion (%) 1985 86 119,200 51,161 42.92 1986 87 119,700 50,297 42.02 198788 138,000 55,850 40.47 1988 89 146,600 58,246 39.73 1989 90 110,200 41,029 37.23 1990 91 151,600 57,615 38.00 199192 139,800 52,628 37.65 1992 93 186,600 68,534 36.73 1993 94 174,400 64,040 36.72 1994 95 205,500 82,040 39.92 199596 203,300 78,728 38.73 1996 97 226,200 88,485 39.12 1997 98 244,000 100,625 41.24 1998 99 186,000 69,304 37.26 199900 233,000 96,112 41.25 2000 01 223,300 91,823 41.12 2001 02 230,000 99,043 43.06 2002 03 203,000 88,060 43.38 200304 242,000 111,722 46.17 2004 05 149,800 67,706 45.20 2005 06 147,900 70,450 47.63 Lower limit 39.31% Mean value 40.74% Upper limit 42.17% Source: FASS, 2007.

PAGE 46

46 Table 3 3. Proportion of total Valencia production on the tree at each of the four harvest dates for seasons 198586 through 200506. Season Early May proportion (%) Mid May proportion (%) Early June proportion (%) Mid Jun proportion (%) 1985 86 18.84 15.31 9.85 4.09 1986 87 19.56 18.01 12.31 4.03 1987 88 21.21 18.46 14.13 11.19 1988 89 19.28 16.90 12.74 7.34 1989 90 11.09 8.90 0.48 0.23 1990 91 13.27 9.59 1.08 0.27 1991 92 15.62 8.90 2.65 0.94 1992 93 17.00 13.78 8.15 4.99 1993 94 18.43 17.24 11.30 7.96 1994 95 14.51 9.54 4.48 2.09 1995 96 17.33 20.59 6.26 1.81 1996 97 16.91 18.37 6.98 5.31 1997 98 13.62 9.65 6.53 3.46 1998 99 14.37 9.17 3.51 1.06 1999 00 16.11 13.96 10.33 13.68 2000 01 15.58 22.04 9.42 6.72 2001 02 14.65 19.30 8.25 3.47 2002 03 15.60 16.82 7.62 1.56 2003 04 14.72 12.31 7.83 6.16 2004 05 15.59 14.06 6.38 8.48 2005 06 14.43 13.17 10.89 13.94 Average ( % ) 16.08 14.57 7.67 5.18 Source: FASS, 2007. Figure 3 2. Confidence interval f or the proportion of total Valencia production by harvest date. Source: FASS, 2007. 14.98 % 12.67 % 5.97 % 3.31 % 63.07 % Lower limit 16.08 % 14.57 % 7.67 % 5.18 % 56.50 % Mean 17.18% 16.48% 9.38% 7.05% 49.91% Upper limit Early May Mid May Early June Mid Jun Before May

PAGE 47

47 Table 3 4. FDOC all round orange projection and Valencia oranges production scenarios (million boxes) for seasons 200708 through 201718. All round oranges a Valencia oranges b Season Low production (million boxes) High production (million boxes) Low production (million boxes) High production (million boxes) 2007 08 183 187 74.55 76.18 2008 09 179 185 72.92 75.37 2009 10 174 181 70.89 73.74 2010 11 169 178 68.85 72.52 2011 12 163 175 66.41 71.30 201213 159 174 64.78 70.89 2013 14 156 173 63.55 70.48 201415 154 172 62.74 70.07 2015 16 152 173 61.92 70.48 201617 152 174 61.92 70.89 2017 18 152 174 61.92 70.89 Source: a All round oranges production forecasted by Brown (2007). b Valenci a oranges production equals A ll round or anges forecasted production multiplied by 40.74% ( from Table 3 2). Table 3 5. Projected late-season Valencia oranges by harvest dates, low and high production scenarios (million boxes), for seasons 200708 through 201718. Season Early May (million boxes) Mid May (million boxes) Early June (million boxes) Mid June (million boxes) Low High Low High Low High Low High 2007 08 11.99 12.25 10.86 11.10 5.72 5.84 3.86 3.95 2008 09 11.73 12.12 10.63 10.98 5.59 5.78 3.78 3.90 200910 11.40 11.86 10.33 10.74 5.44 5.66 3.67 3.82 201011 11.07 11.66 10.03 10.57 5.28 5.56 3.57 3.76 2011 12 10.68 11.46 9.68 10.39 5.09 5.47 3.44 3.69 201213 10.42 11.40 9.44 10.33 4.97 5.44 3.36 3.67 2013 14 10.22 11.33 9.26 10.27 4.87 5.41 3.29 3.65 2014 15 10.09 11.27 9.14 10.21 4.81 5.37 3.25 3.63 201516 9.96 11.33 9.02 10.27 4.75 5.41 3.21 3.65 2016 17 9.96 11.40 9.02 10.33 4.75 5.44 3.21 3.67 2017 18 9.96 11.40 9.02 10.33 4.75 5.44 3.21 3.67 Source: Valencia forecasted production from Table 3 4 multiplied by average values from Table 3 3.

PAGE 48

48 Table 3 6. P ercentage difference between all round oranges and Valencia oranges on tree price per b ox ($/box) f o r seasons 198586 through 200506. Season All round oranges ($/box) Valencia oranges ($/box) Price differential (%) 1985 86 3.94 3.97 0.8 1986 87 5.22 6.02 15.3 1987 88 7.58 8.73 15.2 1988 89 7.41 8.41 13.5 1989 90 6.21 6.53 5.2 1990 91 5.89 6.58 11.7 1991 92 5.93 6.65 12.1 1992 93 3.48 3.88 11.5 1993 94 4.09 4.61 12.7 1994 95 3.74 4.41 17.9 1995 96 4.40 5.57 26.6 1996 97 3.54 4.07 15.0 1997 98 3.69 4.88 32.2 1998 99 4.84 5.58 15.3 1999 00 3.67 4.33 18.0 2000 01 3.21 4.02 25.2 2001 02 3.47 4.20 21.0 2002 03 3.17 3.85 21.5 2003 04 2.89 3.64 26.0 2004 05 3.49 4.24 21.5 2005 06 5.46 6.42 17.6 Lower limit 13.60% Mean value 16.90% Upper limit 20.20% Source: FASS, 2007.

PAGE 49

49 Table 3 7. Projected Valencia orange price scenarios ($/box), seasons 200708 through 201718 All round oranges a ($/box) Valencia oranges b ($/box) Season Low price High price Low price High price 2007 08 6.47 6.74 7.56 7.88 2008 09 6.61 7.11 7.73 8.31 2009 10 6.99 7.77 8.17 9.08 2010 11 7.26 8.40 8.49 9.82 2011 12 7.51 9.04 8.78 10.57 2012 13 7.64 9.56 8.93 11.18 2013 14 7.72 9.98 9.02 11.67 2014 15 7.73 10.25 9.04 11.98 2015 16 7.69 10.38 8.99 12.13 2016 17 7.61 10.38 8.90 12.13 2017 18 7.61 10.38 8.90 12.13 Sources: a All round oranges price forecasted by Brown (2007). b Valencia oranges price equals A ll round oranges forecasted price increased by 16.90% from mean price differential calculated in Table 3 6.

PAGE 50

50 Table 3 8 Percentage difference between all round oranges and Valencia oranges yield per acre for seasons 198586 through 200506. Season All round oranges (boxes/acre) Valencia oranges (boxes/acre) Differential (%) 198586 324 297 8.3 1986 87 319 289 9.4 198788 363 327 9.9 198889 377 334 11.4 198990 276 229 17.0 199091 360 324 10.0 199192 315 280 11.1 1992 93 381 321 15.7 199394 341 281 17.6 199495 365 326 10.7 199596 342 293 14.3 199697 362 311 14.1 199798 401 356 11.2 1998 99 304 251 17.4 199900 387 335 13.4 2000 01 369 315 14.6 200102 392 339 13.5 200203 345 299 13.3 200304 428 389 9.1 200405 276 242 12.3 200506 301 269 10.6 Average yields 348.95 305.09 Lower limit 11.34% Mean value 12.61% Upper limit 13.99% Source: FASS, 2007

PAGE 51

51 Table 3 9 EUP acreage constraint converted to total boxes of Valencia orange for seasons 200708 through 201718 under two production scenarios (low and high). All oranges avg. yield (boxes /acre ) Valencia avg. yield (boxes /acre ) Valencia (million boxes) Season Low High Low High Low High 2007 08 388.04 388.29 339.09 339.31 8.48 8.48 2008 09 389.05 389.72 339.97 340.56 8.50 8.51 2009 10 394.29 393.22 344.55 343.62 8.61 8.59 2010 11 400.66 399.82 350.12 349.39 8.75 8.73 2011 12 388.10 390.02 339.14 340.82 8.48 8.52 2012 13 375.80 380.49 328.39 332.50 8.21 8.31 2013 14 365.43 371.24 319.33 324.42 7.98 8.11 2014 15 355.99 361.95 311.08 316.30 7.78 7.91 2015 16 345.14 356.48 301.60 311.51 7.54 7.79 2016 17 337.63 350.88 295.04 306.62 7.38 7.67 2017 18 337.63 350.88 295.04 306.62 7.38 7.67 Source: Yield differential calculated in Table 3 7 applied to All round oranges yield per acre forecasted by Brown (2007). Converted to boxes in 25,000 acre of EUP. Table 3 10. Abscission registration and R &D expenditures (million $) from 1996 to 2008 discounted to 2008 dollars to determine initial investment in 2008. Year FDOC R&Da (million $) UF/IFAS R&D (CI) (million $) FDOC Registration (million $) Federal Grants (million $) Nominal total (million $) Inflation factor Nominal totals inflat ed to 2008 dollars (million $) 1996 0.05 0.00 0.00 0.00 0.05 3.14 0.16 1997 0.08 0.00 0.00 0.00 0.08 2.85 0.21 1998 0.43 0.00 0.00 0.00 0.43 2.59 1.11 1999 0.61 0.00 0.00 0.00 0.61 2.36 1.44 2000 0.49 0.00 0.00 0.00 0.49 2.14 1.06 2001 0.67 0.00 0.00 0.00 0.67 1.95 1.30 2002 0.51 0.00 0.00 0.00 0.51 1.77 0.90 2003 0.46 0.00 0.00 0.00 0.46 1.61 0.74 2004 0.50 0.00 0.00 0.07 0.57 1.46 0.84 2005 0.20 0.38 0.05 0.65 1.28 1.33 1.70 2006 0.00 0.42 0.71 1.33 2.46 1.21 2.98 2007 0.00 0.53 3.92 0.00 4.44 1.10 4.89 2008 0.00 0.48 1.30 0.00 1.78 1.00 1.78 Total in 2008 dollars $19.10 So urce: King and Roka 2008. Notes: a FDOC R &D expenditures include only abscission. Does not include funds spent on equipment design.

PAGE 52

52 Table 3 1 1 Scenario description in terms of price, production, percentage of lost fruit and EUP assumptions. Scenario Price Production % lost fruit EUP acres Date of CMNP application % preserved fruit 1 High Low High 25,000 Early May 15% 2 High Low Low 25,000 Early and mid June 15% and 20% 3 Low High High 25,000 Early May 15% 4 Low High Low 25,000 Early and mid June 15% and 20%

PAGE 53

53 CHAPTER 4 COST/BENEFIT ANALYSI S AND SCENARIOS RESULTS As outlined in Table 3 11, four scenarios are developed and analyzed to determine if the private benefit of extending mechanical harvesting into late May and June with abscission agent application more than offset the public costs of development and registrat ion. These scenarios examined changes with respect to four variables: price, production, percentage of lost fruit and the amount of fruit allotted by the experimental use permit (EUP) to be sprayed with CMNP. Net present value (NPV) is the appraisal method used to evaluate the public investment into abscission agent application for citrus mechanical harvesting. The NPV calculation compared discounted benefits and costs between 1995 and 2018 for the development and registration of the abscission compound CMN P. Florida Department of Citrus (FDOC) projections link high price scenarios with low levels of production; conversely, a low price scenario implies a higher level of production. Therefore this paper will not consider high price/high production and/or low price/low production scenarios. The acreage treated with CMNP is assumed to equal the requested amount in the EUP. While the EUP acreage of 25,000 is the same across all scenarios, the model converts acreage into boxes by multiplying yield per acre. Thus, total boxes that can be mechanically harvested within that acreage vary according to production levels. These differences are related to low and high production differences in yield per acre. Scenarios assuming low production implies low yield per acre, an d scenar ios with high production imply high yield per acre (Table 3 9). The allocation of the EUP allotment depends on the amount of late season Valencia by harvest date and the percentage of lost fruit. In the case for scenarios with a high fruit loss percentage from mechanical harvesting without abscission, the entire EUP allotment is allocated to the early May harvest date. During the 200910 season and under a high production

PAGE 54

54 scenario, 11.86 million boxes are expected to be harvested in early May ( Table 3 5). The EUP under high production would allow only 8.59 million boxes to be mechanically harvested with CMNP ( see Table 3 9). Thus, the entire EUP allotment would be utilized during early May. Alternatively, in a scenario with low fruit loss perce ntages, yield reductions would not start to occur until the June harvest dates. In this case, the EUP allotment would be spread across both the early and mid June harvest dates. For example, in high production scenarios, 5.66 and 3.82 million boxes of Va lencia are expected to be harvested during early and mid June respectively. The model allocates 5.66 out of 8.59 million boxes in early June to be mechanically harvested with abscission and the remaining 2.93 million boxes allowed by the EUP would be util ized in mid June. Scenario 1, High Price/Low Production and High % of Lost Fruit. Net cash flows for Scenario 1, High Price/Low Production and High % of Lost Fruit, are presented in Table 4 1. CMNP benefits start accruing during spring 2010 when the EUP is first available. Mechanical harvesting operations in 2009 10 during early May with CMNP results in preserving an estimated 1.31 million boxes for the 201011 season. The dollar value of 1.31 million boxes in 201011 is estimated to be $12.89 millio n. In subsequent years the amount of preserved boxes decreases slightly, reflecting decreases in FDOC projected yields per acre. Higher prices ( see Chapter 3, Table 3 7), however, offset production declines and the dollar value of preserved fruit continues to increase from $13.44 million in 201112 to a maximum of $13.98 million in 201415. Afterward the value of preserved fruit decreases until 201718 when it totals $13.43 million. Similarly, s avings generated by mechanically harvesting 25,000 acres du ring the Valencia late season start accruing in season 201011. Under a low production scenario, the EUP allow for 8.61 million boxes to be mechanically harvested with CMNP during the 200910

PAGE 55

55 season out of those boxes, 1.31 million will be preserved d ue to CMNP application Given a net unit cost savings of 80-cents (see Chapter 3), total savings from mechanically harvesting the EUP allotment in 2010 equals $1.05 million. Number of preserved boxes start decreasing for the following seasons, thus, result ing in a steadily decline in savings. The cumulative harvest cost savings between 200910 and 201718 total s $7.62 million. Net benefits in season 201011 are $ 13.50 million, $12.89 million in value of preserved boxes plus $1.05 million in savings minus $0.53 million in R&D and registration costs for CMNP. The combined benefits of abscission value of lost fruit and mechanical harvesting cost savings, result in a positive net benefit of $14.02 million by the end of 2012. This net value increases to $14.93 million during the 201314 season. Estimated inflows for the following years start de creas ing until they reach a net value of $ 14. 31 million by 201617. T he cumulative net present value of net cash flows for CMNP equals $ 42.49 million The period of time r equired for the net cash flows to repay the total investment on R&D and registration of CMNP under this scenario is 4.5 4 years. An internal rate of return (IRR) is also calculated as a measurement of efficiency for this investment and equaled 34.74% (Table 4 5). Scenario 2, High Price/Low Production and Low % of Lost Fruit. Price and production assumptions for Scenario 2 are the same as for Scenario 1, except that Scenario 2 considers a low percentage of lost fruit due to mechanical harvesting without abscission. Recall that fruit losses could be high one year and low the next season, which could be a function of the diameter of young fruit present at the time of harvest (see Chapter 2). Under an assumption of low percentage fruit loss yield reduction s of 15 and 20 percent will occur during early and mid June, respectively. Harvested boxes during early and mid June of 2010 are estimated to be 5.44 and 3.67 million boxes, respectively ( see Chapter 3, Table 3 5). The entire

PAGE 56

56 early June harvest will be mechanically harvested with CMNP and the remaining portion of the EUP allotment (2.93 million boxes) will be utilized in mid June. Net cash flows are presented in Table 4 2. Mechanical harvesting with CMNP applications in 20091 0 results in preserving 1.49 million boxes, 15% of the projected boxes in early June total 0.79 million and 20% of the projected boxes in mid June total 0.70 million. Note that the 20% is applied to the remaining portion of the EUP allocated in mid June. I n 201011, the value of Valencia preserved fruit totals $14.60 million. As in the previous scenario, the amount of preserved boxes slightly decreases, but the dollar value continues to increase from $15.23 million in 201112 to a maximum of $15.78 million in 201314. The total value of preserved fruit from 201011 through 201718 total $122.39 million. Mechanical harvesting operations with CMNP application during June 2010 resulted in preserving 1.49 million boxes for the following season. Therefore, h arv est cost savings of $1. 19 million will be accrued during the 201011 season. As in the previous scenario, the number of preserved boxes decreas e for the following seasons resulting in a steadily decline in savings. The cumulative harvest cost savings betwe en 200910 and 201718 totals $8.59 million. While holding price/production assumptions constant, changes in yield reduction percentages increase the cumulative total number of preserved boxes from 9.52 to 10.73 million, a 12.71% increase across the eight years time periods. Increases in the value of preserved boxes are due to differences in the percentage of lost fruit, specifically the portion harvested during mid June that yields a 20% instead of 15%. When changing from high to low yield reduction percen tages, the time when the EUP allotment is shifted from early May to early and mid June. Due to the fact that the full EUP production allotment will be utilized, total boxes that are mechanically harvested with abscission from 200910 through 201718 are th e same. CMNP net

PAGE 57

57 benefit with Scenario 2 increases $13.78 million over the study time period, as compared to the cumulative net benefits in Scenario 1. In Scenario 2, the NPV of cumulative net cash flows for CMNP equals $ 50.67 million, 19.25% higher than t he NPV in Scenario 1 (Figure 4 1). The period of time required for the net cash flows to repay the total investment on R&D and registration of CMNP under this scenario is 4.36 years with an IRR of 38.13% (Table 4 5). Scenario 3, Low Price/High Production a nd High % of Lost Fruit. In Scenario 3, a high production assumption translates into higher yields per acre. Given the same EUP constraint of 25,000 acres, total boxes available for abscission agent application and mechanical harvest from 2009 10 through 2 01718 season are 73.29 million boxes, 1.18 million boxes higher than in Scenarios 1 and 2 under low production forecasts. This amount of EUP allotment is less than the volume of projected Valencia harvested during early May of any year through the 2018 season (Table 3 5). Hence, for Scenario 3 and the high percentage fruit loss assumption, the entire EUP allotment will be utilized during the early May harvest period. While total volume of preserved fruit is slightly higher, low price assumptions embed ded in Scenario 3 result in a lower value of preserved fruit. In 201011, the dollar value of 1.31 million boxes of preserved Valencia oranges total $11.12 million. The value of preserved fruit reaches a maximum of $11.22 million in 201112. Afterwards t his value starts to decrease until 201718, totaling $10.23 million accrued to abscission application. Net cash flows are presented in Table 4 3. Higher yielding blocks positively impact harvest cost savings. When summed across the entire planning horizon, savings total $ 7 76 million, value slightly greater than savings accrue under the first scenario. While differences in production are relatively small, on average prices in scenario 3 are 18.01% lower than prices in Scenarios 1 and 2 (high price scenarios ). For Scenario 3, the estimated net present value of net cash flows for CMNP result s in a positive $30. 62 million This NPV is 38. 77% lower than the

PAGE 58

58 NPV under Scenario 1; this difference can be attributed, mainly, to the effect of low price assumptions (F igure 4 2). The period of time required for the net benefits of CMNP to repay its total investment on R&D and registration under Scenario increases slightly to 4.7 9 years. An internal rate of return (IRR) is calculated to be 29.74% (Table 4 5). Scenario 4, Low Price/High Production and Low % of Lost Fruit. Price and production assumptions on this scenario are the same as those for Scenario 3, except that low yield reduction percentages due to mechanical harvesting are assumed. Similarly to Scenario 2, these reductions will occur during early and mid June, 15% and 20% respectively. Net cash flows are presented in Table 4 4. Mechanical harvesting operations in 200910 with applications of CMNP results in preserving 1.47 million boxes, 15% of the projected boxe s in early June total 0. 83 million and 20% of the projected boxes in mid June total 0. 64 million. The total value of preserved fruit from 201011 through 201718 totals $95.52 million. When compared to scenario 2 ($122.39 million), this reduction in the va lue of lost fruit can be attributed to differences in price assumptions. Savings generated by mechanically harvesting with CMNP during June totaled $8.61 million, value slightly greater than in previous scenarios Although, total boxes that are mechanical ly harvest with abscission from 200910 through 201718 are the same in Scenarios 3 and 4, changing from high to low yield reduction percentages increase the total number of preserved boxes from 9.71 to 10.76 million, a 10.81% increase. These results are consistent with the results from the comparis on between Scenarios 1 and 2. The value of preserved boxes in scenario 4 is higher than in scenario 3, increases are due to differences in the percentage of lost fruit for the portion harvested during mid June (20%) which is the case in the former The NPV of net cash flows for CMNP results in a positive $ 36.40 million, 11. 89% higher than the NPV in scenario 3, due to changes in yield reduction percentages (Figure 4 1). The effect of changes in price/production a ssumptions can be measured

PAGE 59

59 by comparing scenarios 2 and 4 (Figure 4 2). The NPV in scenario 4 is 39. 20 % lower than the NPV under scenario 2. The period of time required for the net cash flows to repay the total investment on R&D and registration of CMNP under this scenario is 4.60 years with an IRR of 32.58% (Table 4 5). NPV Threshold. The four scenarios described in previous sections evaluate how different price and production assumptions combined with percentage levels of lost fruit from mechanical harves ting late season Valencia oranges without abscission affect the NPV of a public investment to develop and register the abscission compound CMNP. NPV results by scenario provide a basis to evaluate minimum threshold levels for unit harvest cost savings an d EUP acreage. Figures 4 3 t hrough 4 6 depict NPV surfaces for each scenario. The vertical axis shows the NPV in million dollars for ranges of unit cost savings due to mechanical harvesting and EUP acreage limits. The bottom right axis represents different levels of unit cost savings that range from $0.00 to $0.80 per box the unit cost differential used in this analysis. The bottom left axis represents different levels of EUP acreage ranging from 0 to 25,000 acres, the amount being requested in the EUP application with U.S. EPA. For each scenario the NPV surface decreases as both EUP acreage and unit cost savings decrease. The surfaces allow threshold points to be identified where the NPV is no longer positive. The NPV surfaces for scenarios 1 and 2 are depicted in Figures 4 3 and 44 respectively. Points on the threshol d represent the combined effect on NPV from reductions in EUP acreage and unit harvest cost savings. These NPV surfaces suggest that the NPV is highly influenced by the EUP acreage constraint. Annual mechanically harvest areas of less than 8 403 acres, wit hout changing the unit cost savings, will result in negative values of NPV. On the other hand, if unit

PAGE 60

60 cost savings are $0.10 per box, then mechanically harvested areas with CMNP application of less than 8, 922 acres will also result in negative values of N PV. Under these scenarios, if unit cost savings are $0.40 or $0.20 per box, the minimum mechanically harvested area below which the cost of CMNP is greater than its benefits (negative NPV) equaled 8,692 and 8 844 acres respectively (Table 4 6). These two s urfaces yield the same NPV for acreage below 16,000 acres, for scenarios assuming low percentage of lost fruit this is the minimum acreage under which the entire EUP allotment can be consumed in early June, and therefore, the value of preserved fruit under scenarios 1 and 2 are the same for each year. Recall a 15% reduction in early May for high percentage of lost fruit and 15% reduction in early June for low percentage of lost fruit scenarios. In more conservative scenarios, points on the NPV threshold wer e similar. NPV surfaces for scenarios 3 and 4 are depicted in Figures 4 5 and 4 6 respectively. Annual mechanically harvest areas of less than 10,3 16 acres, without changing the unit cost savings, will result in negative values of NPV. If unit cost savings are $0.10 per box, then mechanically harvested areas with CMNP application of less than 1 1,123 acres will also result in negative values of NPV. Based on these NPV surfaces, when unit cost savings are $0. 4 0 or $0. 2 0 per box, positive values of NPV will re sult only if mechanically harvested area can be maintains above 10,762 and 11,000 acres respectively (Table 4 6). To add more clarity to overall sensitivity of unit costs, Table 4 7 depicts the maximum acreage of CMNP application to achieve a payback perio d within 6 years by selected unit cost harvest savings (i.e. Cumulative NPV=0 by 201314). If unit cost savings are $0.10 per box annual mechanically harves t areas of less than 13,581 acres will result in negative values of NPV in high price/low production scenarios. On the other hand, low price/high production scenarios

PAGE 61

61 will yield negative NPV if unit cost savings are $0.10 per box and mechanically harvested areas with C MNP application are under 16,234 acres

PAGE 62

62 Table 4 1 Scenario 1, hig h price/low production high fruit loss percentage. CMNP projected net cash flows (million $) from 200708 through 201718. Season 200708 200809 200910 201011 201112 201213 201314 201415 201516 201617 201718 Years 0 1 2 3 4 5 6 7 8 9 10 Registration Cost (9.12) (0.99) (0.09) (0.09) (0.09) R&D by IFAS/UF (2.07) (0.48) (0.45) (0.35) (0.35) R&D by FDOC (7.92) Preserved boxes 1.31 1.27 1.23 1.20 1.17 1.13 1.11 1.11 Value of Lost Fruit 0 12.89 13.44 13.76 13.97 13.98 13.72 13.43 13.43 Savings 0.00 1.05 1.02 0.99 0.96 0.93 0.90 0.89 0.89 CMNP net benefit (19.10) (1.47) (0.54) 13.50 14.02 14.75 14.93 14.91 14.63 14.31 14.31 Discount factor 0.909 0.826 0.751 0.683 0.621 0.564 0.513 0.467 0.424 0.386 Net present value (19.10) (1.33) (0.45) 10.14 9.57 9.16 8.42 7.65 6.83 6.07 5.52 Cumulative NPV (2008 $) (19.10) (20.43) (20.88) (10.74) (1.17) 7.99 16.41 24.06 30.89 36.96 42.49 Table 4 2 Scenario 2, high price/low production low fruit loss percentage. CMNP projected net cash flows (million $) from 200708 through 201718. Season 200708 200809 200910 201011 201112 201213 201314 201415 201516 201617 201718 Years 0 1 2 3 4 5 6 7 8 9 10 Registration Cost (9.12) (0.99) (0.09) (0.09) (0.09) R&D by IFAS/UF (2.07) (0.48) (0.45) (0.35) (0.35) R&D by FDOC (7.92) Preserved boxes 1.49 1.44 1.39 1.35 1.31 1.27 1.24 1.24 Value of Lost Fruit 0 14.60 15.23 15.57 15.78 15.75 15.42 15.02 15.02 Savings 0.00 1.19 1.15 1.11 1.08 1.05 1.02 0.99 0.99 CMNP net benefit (19.10) (1.47) (0.54) 15.35 15.94 16.69 16.87 16.81 16.43 16.01 16.01 Discount factor 0.909 0.826 0.751 0.683 0.621 0.564 0.513 0.467 0.424 0.386 Net present value (19.10) (1.33) (0.45) 11.53 10.89 10.36 9.51 8.62 7.67 6.79 6.18 Cumulative NPV (2008 $) (19.10) (20.43) (20.88) (9.35) 1.53 11.90 21.41 30.03 37.71 44.49 50.67

PAGE 63

63 Table 4 3 Scenario 3, low price/high production high fruit loss percentage. CMNP projected net cash flows (million $) from 200708 through 201718. Season 200708 200809 200910 201011 201112 201213 201314 201415 201516 201617 201718 Years 0 1 2 3 4 5 6 7 8 9 10 Registration Cost (9.12) (0.99) (0.09) (0.09) (0.09) R&D by IFAS/UF (2.07) (0.48) (0.45) (0.35) (0.35) R&D by FDOC (7.92) Preserved boxes 1.31 1.28 1.25 1.22 1.19 1.17 1.15 1.15 Value of Lost Fruit 0 11.12 11.22 11.14 10.98 10.72 10.50 10.23 10.23 Savings 0.00 1.05 1.02 1.00 0.97 0.95 0.93 0.92 0.92 CMNP net benefit (19.10) (1.47) (0.54) 11.73 11.80 12.13 11.95 11.67 11.44 11.15 11.15 Discount factor 0.909 0.826 0.751 0.683 0.621 0.564 0.513 0.467 0.424 0.386 Net present value (19.10) (1.33) (0.45) 8.81 8.06 7.53 6.74 5.99 5.34 4.73 4.30 Cumulative NPV (2008 $) (19.10) (20.43) (20.88) (12.07) (4.01) 3.52 10.26 16.25 21.59 26.32 30.62 Table 4 4 Scenario 4, low price/high production low fruit loss percentage CMNP projected net cash flows (million $) from 200708 through 201718. Season 200708 200809 200910 201011 201112 201213 201314 201415 201516 201617 201718 Years 0 1 2 3 4 5 6 7 8 9 10 Registration Cost (9.12) (0.99) (0.09) (0.09) (0.09) R&D by IFAS/UF (2.07) (0.48) (0.45) (0.35) (0.35) R&D by FDOC (7.92) Preserved boxes 1.47 1.43 1.39 1.35 1.31 1.29 1.26 1.26 Value of Lost Fruit 0 12.47 12.56 12.42 12.20 11.86 11.57 11.22 11.22 Savings 0.00 1.18 1.14 1.11 1.08 1.05 1.03 1.01 1.01 CMNP net benefit (19.10) (1.47) (0.54) 13.20 13.26 13.53 13.28 12.91 12.60 12.23 12.23 Discount factor 0.909 0.826 0.751 0.683 0.621 0.564 0.513 0.467 0.424 0.386 Net present value (19.10) (1.33) (0.45) 9.91 9.06 8.40 7.49 6.62 5.89 5.19 4.72 Cumulative NPV (2008 $) (19.10) (20.43) (20.88) (10.97) (1.91) 6.50 13.99 20.61 26.50 31.68 36.40

PAGE 64

64 Figure 4 1 Comparison of CMNP net present values (2008 million $ ) under di fferent percentage of lost fruit assumptions Figure 4 2. Comparison of CMNP net present values (2008 million $) by price/production combination under different fruit loss assumptions. 0 10 20 30 40 50 60 2 and 1 4 and 3 50.67 36.40 42.49 30.62 NPV million $ Scenarios Low % lost fruit High % lost fruit 0 10 20 30 40 50 60 70 80 1 and 3 (High fruit loss) 2 and 4 (Low fruit loss) 72.28 79.93 60.79 66.09 NPV million $ Scenarios High price/low production Low price/high production

PAGE 65

65 Figure 4 3. Impact of mechanical harvesting acreage during the late season and unit cost savings under scenari o 1, high price/low production and high % of lost fruit. Figure 4 4. Impact of mechanical harvesting acreage during the late season and unit cost savings under scenario 2, high price/low production and low % of lost fruit. (30.00) (20.00) (10.00) 0.00 10.00 20.00 30.00 40.00 50.00 25.0 22.5 20.0 17.5 15.0 12.5 10.0 7.5 5.0 2.5 0.0 Unit cost savings ($) NPV million $ EUP acreage (1,000 acres) (30.00) (20.00) (10.00) 0.00 10.00 20.00 30.00 40.00 50.00 60.00 25.0 22.5 20.0 17.5 15.0 12.5 10.0 7.5 5.0 2.5 0.0 Unit cost savings ($) NPV million $ EUP acreage (1,000 acres)

PAGE 66

66 Figure 4 5. Impact of mech anical harvesting acreage during the late season and unit cost savings under scenario 3, low price/high production and high % of lost fruit. Figure 4 6. Impact of mechanical harvesting acreage during the late season and unit cost savings under scenario 4, low price/high production and low % of lost fruit. (30.00) (20.00) (10.00) 0.00 10.00 20.00 30.00 40.00 25.0 22.5 20.0 17.5 15.0 12.5 10.0 7.5 5.0 2.5 0.0 Unit cost savings ($) NPV million $ EUP acreage (1,000 acres) (30.00) (20.00) (10.00) 0.00 10.00 20.00 30.00 40.00 25.0 22.5 20.0 17.5 15.0 12.5 10.0 7.5 5.0 2.5 0.0 Unit cost savings ($) NPV million $ EUP acreage (1,000 acres)

PAGE 67

67 Table 4 5. Summary of scenario results, c umulative NPV ($), payback period s (years from 2008) and IRR (%) 1995 through 2018. Acreage receiving CMNP 25,000 and unit cost savings of 80 -cents per box to mechanically harvest. Scenario Total NPV (2008 million $) Payback period (Years) IRR (%) S1: high price/low production high fruit loss 42.49 4.54 34.74 S2: high price/low production low fruit loss 50.67 4.36 38.13 S3: low price/high production high fruit loss 30.62 4.79 29.74 S4: low price/high production low fruit loss 36.40 4.60 32.58 Table 4 6. Minimum acreage of CMNP application by selected unit harvest cost savings to achieve positive NPV for each scenario by the end of the 201718 season. Scenario Unit harvest cost savings 80 cents 40 cents 20 cents 10 cents Acreage S1: high price/low production high fruit loss 8,403 8,692 8,844 8,922 S2: high price/low production low fruit loss 8,403 8,692 8,844 8,922 S3: low price/high production high fruit loss 10,316 10,762 11,000 11,123 S4: low price/high production low fruit loss 10,316 10,762 11,000 11,123

PAGE 68

68 Table 4 7. Minimum acreage of CMNP application to achieve a payback period within 6 years by selected unit harvest cost savings. Scenario Unit harvest cost savings 80 cents 40 cents 20 cents 10 cents Acreage S1: high price/low production high fruit loss 12,740 13,207 13,452 13,581 S2: high price/low production low fruit loss 12,740 13,207 13,452 13,581 S3: low price/high production high fruit loss 15,042 15,700 16,048 16,234 S4: low price/high production low fruit loss 15,042 15,700 16,048 16,234

PAGE 69

69 CHAPTER 5 SUMMARY IMPLICATIONS AND CONCLUSIONS Summary There are two main reasons why the Florida citrus industry needs to find ways to reduce costs. The first reason has to do with offsetting cost increases due to combating citrus greening and citrus canker diseases. A second important reason has to do with maintaining global competitiveness. A compa rison of citrus production costs between Florida and Sao Paulo (Brazil) for the 2000 01 season suggested that the largest cost gap between the two states lies in harvest costs and Brazil mains advantage is labor cost (Muraro, 2003). More than 93% Flori das citrus crop is hand harvested. Furthermore, harvesting costs exceed production costs under typical cultural programs. With greening management costs included, costs to hand harvest still represent more than 70% of total production costs (Muraro, 2007) Since harvesting costs play such an important role in the long -term economic sustainability in citrus production, the Florida Department of Citrus (DOC) has been investing grower taxes into a mechanical harvesting program since 1995. Roka (2006) lists s everal significant impediments to the widespread adopt ion of mechanical harvesting systems, of which the most significant obstacle is harvesting late -season Valencia oranges. The late season begins sometime during May of the current year, when next ye ars growing fruit reaches one inch in diameter. P revious studies estimate the fruit loss from mechanical harvesting with young fruit greater than one inch diameter to be between 20% to 50% of next years crop. An a bscission agent application offers one so lution to the late season problem. The current thinking is that the abscission compound CMNP can improve overall e fficiency of m echanical harvesting systems throughout the season. Increasing harvest labor productivity and capitalizing on the principles o f scale economies associated with

PAGE 70

70 mechanical equipment constitutes a potential to reduce the unit cost of harvesting. The DOC initiated a process to formally register CMNP with the U.S. EPA (Environmental Protection Agency) as an abscission agent in citrus harvesting. A EUP (experimental use permit) is expected to be approved by spring of 2010. Under the pre sumption that abscission and specifically CMNP, solves the late -season problem to mechanically harvest Valencia or anges, this research examined the following question: Does the private benefit of extending mechanical harvesting into late May and June with CMNP, an absc ission compound, more than offset its public cost of development and registration? Cost/benefit analysis of CMNP was carried out thr ough a s preadsheet model with static future production and price expectations to deter mine the net present value (NPV) of CMNP in terms of mechanically harvesting late season Valencia oranges Two benefits of abscission were quantified in the model. Firs t, the dollar value of fruit that would have been lost in the next season if the crop in the current late season (after early May) was mechanically harvested without CMNP. Second, the unit cost harvest savings generated by mechanically harvesting after ear ly May. CMNP expend itures we re broken down by registration expenditures from both the Florida Department of Citrus (FDOC) and federal grants FDOC sponsored research & development (R&D) and UF/IFAS fund ed R&D The economic viability of CMNP was evaluated by calculating the NPV of accrued costs since 1995 and anticipated costs through 2012, the year CMNP is expected to be fully registered for commercial use. Benefits start accruing in 2010 with the approval of the E UP and continue through the 2018 season, corresponding to the last year of the DOCs most current production forecast. Costs and benefits were discounted at a 10% rate to express annual net cash flows in terms of 2008 dollars Further,

PAGE 71

71 the model examined f our scenarios to determine the sensitivity of the NPV calculations to changes in price and production forecasts, as well as percentage of lost fruit and EUP acreage. Implications and Conclusions Uti lizing CMNP should allow mechanical harvesting systems to achieve a high percentage of mature fruit removal during the Valencia orange late season with no significant impact on yield for the following season crop (Burns et al., 2006). The estimated NPV of CMNP publ ic investment from 1995 through 2018 under high price and low production expectations ranged from $ 42 to $51 million Under low price and high production expectations, NPV estimates over the same time frame ran ged between $ 3 0 and $37 million. In any of the four sce narios considered, the NPV of CMNP investment with 25,000 of treated late -season acreage tur ned positive by the 2013 season, five years from 2008, the reference year of this study. Results by scenario suggest that on -tree price will have a bigge r influence on CMNP returns than production levels. The effect of changing production assumption is relatively small compared to the effect of price changes. In high production scenarios, the total number of boxes mechanically harvest in the late season wi thin 25,000 acres of EUP is greater than in low production scenarios, resulting in higher savings. Low price assumption embedded w i th high production level s, however, result ed in a lower value of preserved fruit, offsetting gains in harvest savings. The e ffect of changes in yield reduction rates, from a high to a low loss scenario, on CMNP returns is not greater than the effect of on -tree price. More preserved boxes will result from mechanical harvesting with abscission application during the late season w hen one inch diameter fruit is present on the tree during early and mid June (low loss scenarios). Recall that in high loss scenarios the entire EUP allotment is allocated during early May whereas in low loss scenarios a portion is allocated in early June and the remainder in mid June, increases in the

PAGE 72

72 value of preserved boxes are due to differences in the percentage of lost fruit, specifically the portion harvested during mid June that yields a 20% instead of 15%. Results by scenario provide a basis to evaluate minimum threshold levels for two critical variables: unit harvest cost savings and EUP acreage. These threshold levels are useful to visualize the implications of changes in these two variables and their effect in the outcome of this investment. Red uctions in the gap between hand harvesting cost per box and mechanical harvesting plus the application of CMNP cost per box, referred to as the unit harvest cost savings, have a negative impact on the NPV. If this gap is reduced to zero, the NPV of CMNP wi ll still be positive, but the industry will have no incentive to implement mechanical harvesting. The overarching objective of abscission agent application and mechanical harvesting systems are to reduce harvesting costs and increase growers on tree retur ns. The EUP constrained the model by setting a limit in terms of the acreage that CMNP with mechanical harvesting will have access during the late season. If this constraint is removed and more acres receive CMNP application with mechanical harvesting, then more acres realize the estimated benefits. Returns will increase as well as the NPV of CMNP. On the other hand, a threshold exists for total acreage receiving CMNP, below which the NPV of CMNP is negative. The likelihood that the public investment int o CMNP will be positive is very strong. The minimum acreage threshold to achieve a positive NPV on CMNP investment within 6 years is only 15,000 acres, 10,000 acres less than the requested experimental use permit. This minimum acreage further assumes that net savings from mechanical harvesting will only be 10-cents per box less than hand harvesting The ability that CMNP has in enhancing efficiencies of mechanical harvesting can be tied back with the fact that with economies of scale unit savings should inc rease, and thus, the overall benefit of CMNP keeps expanding.

PAGE 73

73 REFERENCE LI S T Biggs R.H. and S.V. Kossuth. 1980. Modeling the Effectiveness of Release as a Citrus Harvest Aid for Valencia Fruits. Proc. Fla. State Hort. Soc. 93:301305. Brigham E.F., and M.C. Ehrhardt. 2003. Financial Management Theory and Practice. Thomson South Western. Brown, M.G. 2007. Florida Citrus Production Trends 2007 08 Through 201718. Economic and Market Research Department, Florida Department of Citrus Gainesville, Florid a. Burns, J.K. 2002. Using Molecular Biology Tools to Identify Abscission Materials for Citrus. Hort -Science 37:459464. Burns, J.K., Buker R.S., and F. M Roka. 2005. Mechanical Harvesting Capacity in Sweet Orange is Increased with an Abscission Agent. Hort Technology 15:758765. Burns, J. K. Roka, F. M. Li, K., Pozo, L. and R.S. Buker. 2006. Late -season Valencia Orange Mechanical Harvesting with an Abscission Agent and Low -frequency Harvesting. HortScience. 41(3): 660663. Coppock G.E. 1971. Harves ting Valencia oranges with a limb shaker. Proc. Fla. State Hort. Soc. 84:8488. Coppock G.E. 1972. Properties of Young and Mature 'Valencia' Oranges Related to Selective Harvest by Mechanical Means Amer. Soc. of Agric. and Bio. Eng. 15(2):235238. Cop pock, G.E. and S.L. Hedden. 1977. Citrus Harvesting in Florida International Society of Citriculture. 2:393 397. Coppock, G.E., Churchill, D.B. and S.L. Hedden. 1985. Shaker Stroke Affects Selective Removal of Valencia Oranges Amer. Soc. of Agric. and Bio. Eng. 28(4):10941096 Coppock, G.E., Sumner, H.R., Churchill, D.B., and S.L. Hedden. 1981. Shaker Methods for Selective Removal of Oranges Amer. Soc. of Agric. and Bio. Eng. 24(4):902904 Foreign Agricultural Service, USDA, 2008. Washington, D.C. Website: http://www.fas.usda.gov accessed on 04/2008. Florida Agricultural Statistic Service (FASS). 2005. Citrus Summary, 20042005. National Agricu ltural Statistics Service, USDA, Orlando, Florida. Florida Agricultural Statistic Service (FASS). 2006. Citrus Summary, 20052006. National Agricu ltural Statistics Service, USDA, Orlando, Florida. Florida Agricultural Statistic Service (FASS). 2007. Citrus Summary, 20062007. National Agricu ltural Statistics Service, USDA, Orlando, Florida.

PAGE 74

74 Florida Agricultural Statistic Service (FASS). 2008. Citrus Summary, 20072008. National Agricu ltural Statistics Service, USDA, Orlando, Florida. Hedden S.L. and G.E. Coppock. 1968. Effects of The Tree Shaker Harvest S ystem on Subsequent Citrus Yields. Fla. State Hort. Soc. 81:4852. Hedden, S.L. and G.E. Coppock. 1971. Comparative Harvest Trials of Foliage and Limb Shakers in 'Valencia' Oranges Fla. State Hort. Soc. 84:88 92. Kenney D.S., Clark R.K., and W.C. Wilso n. 1974. ABG 3030: an Abscission Chemical for Processing Oranges: Biological Activity. Fla. State Hort. Soc. 87:34 36. King, D. and F.M. Roka 2008 Annual Florida Department of Citrus expenditures on a bscission agent r egistration along with budget and t imeline of anticipated expenditures Personal communication. Muraro R. P. 2004. Summary of Citrus Budget for the Southwest Florida Production Region. UF/IFAS CREC, Lake Alfred, Fl. Website: http://www.crec.ifas.ufl.edu/extension/ economics/index.htm acces sed on 03/2008. Muraro R. P. 2005. Summary of Citrus Budget for the Southwest Florida Production Region. UF/IFAS CREC, Lake Alfred, Fl. Website: http://www.crec.ifas.ufl.edu/extension/ economics/index.htm accessed on 03/2008. Muraro R. P. 2006. Summary of Citrus Budget for the Southwest Florida Production Region. UF/IFAS CREC, Lake Alfred, Fl. Website: http://www.crec.ifas.ufl.edu/extension/ economics/index.htm accessed on 03/2008. Muraro R. P. 2007. Summary of Citrus Budget for the Southwest Florida Production Region. UF/IFAS CREC, Lake Alfred, Fl. Website: http://www.crec.ifas.ufl.edu/extension/ economics/index.htm accessed on 03/2008. Muraro, R.P., Spreen T.H., and M Pozzan. 2003. Comparative Costs of Growing Citrus in Florida and So Paulo (Braz il) for the 200001 Season. Electronic Data Informatio n Source (EDIS) FE364. UF/IFAS FRED, Gainesville, FL. Moss, C.B., Weldon, R.N., and R.P. Muraro. 1991. The Impact of Risk on the Discount Rate for Different Citrus Varieties. Agribusiness: An Internati onal Journal. 7: 327 38. Roka, F.M. 2008. Unit Cost Harvest Savings due to Mechanical Harvesting and CMNP compound Application. Personal communication. Roka, F M. 2007. Potential Economic Value of Abscission, Citrus Mechanical Harvesting Field Day & W or kshop. UF/IFAS SWFREC, Immokalee, FL Website: http://citrusmh.ifas.ufl. edu/index.asp accessed on 01/2008.

PAGE 75

75 Roka, F.M. and R.P. Emerson. 1999. Piece Rates, Hourly Wages, and Daily Farm Worker Income. Citrus and Vegetable Magazine 63(8):1013. Roka, F. M. Burns, J. K. and R. S. Buker. 2005. Mechanical Harvesting without abscission Agents Yield Impact on Late Valencia Oranges. Fla. State Hort. Soc. 118: 25 27. Roka, F. M. and B. Hyman. 2004. Report to the Citrus Harvesting Research Advisory Council Lakelan d, Fl. UF /IFAS S W REC Immokalee, Fl Samuelson, P. A. 1970. Economics. McGraw Hill. Sumner, H.R. and D.B. Churchill. 1977. Collecting and Handling Mechanically Removed Citrus Fruit International Society of Citriculture. 2:413418. UF/IFAS. 2008a. Citr us Mechanical Harvesting website: http://citrusmh.ifas.ufl.edu accessed on 01/2008. UF/IFAS. 2008b. Citrus Research and Education Center website: http://www.crec.ifas.ufl.edu/ crec_websites/abscission/mhs.htm accessed on 01/08. Whitney, J.D. 1975. Orange yield and removal studies with air and trunk shakers using two abscission chemicals. Proc. Fla. State Hort, Soc. 88: 120 124. Whitney J.D. 1976. Air Shaker Harvest Trials in Valencia Oranges With Two Rates of Abscission Chemical. Fla. State Hort. Soc. 89:4143. Whitney, J.D. (1995) A review of citrus harvesting in Florida. Proc. Citrus Engineering Conf., CREC, 700 Expt. Sta. Rd., Lake Alfred, FL 33850 USA. p 3360. Whitney J.D. and S.L. Hedden. 1973. Harvesting Valencia Oranges with a Vertical Foli age Shaker. Fla. State Hort. Soc. 86: 41 48. Whitney, J.D. and H.R. Sumner. 1977. Mechanical Removal of Fruit from Citrus Trees. Proc. Int. Soc. Citriculture. 2:407 412. Whitney, J.D., Wheaton, T.A., Castle, W.S. and D.P.H. Tucker. 1996. Orange Grove Fac tors Affect Manual Harvest Rate Amer. Soc. of Agric. and Bio. Eng. 39:399405 Wilson, W.C. 1973. A Comparison of Cycloheximide with a new Abscission Chemical. Proc. Fla. State Hort. Sci. 98:337 340. Wilson W.C., Holm R.E., and R.K. Clark. 1977. Abscission Chemicals Aid to Citrus Fruit Removal. Proc. Int. Soc. Citriculture. 2:404 406.

PAGE 76

76 BIOGRAPHICAL SKETCH German Blanco Lobo was born in Tegucigalpa, Honduras. One year after he graduated from the Instituto Salesiano San Miguel in 1998 he started st udding Agribusiness in the Pan American School of Agriculture, Z amorano, where he obtained his bachelor s degree in 2003. In 2006, he entered the University of Floridas Food and Resource Economics graduate program to pursue his Master of Science, where he is currently enrolled in the d octoral program.