<%BANNER%>

Economic Analysis of Producing and Handling Francis Mango in Haiti to U.S. Market

Permanent Link: http://ufdc.ufl.edu/UFE0044768/00001

Material Information

Title: Economic Analysis of Producing and Handling Francis Mango in Haiti to U.S. Market A Financial Risk Management Approach
Physical Description: 1 online resource (77 p.)
Language: english
Creator: Pierreval, Isnel
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2012

Subjects

Subjects / Keywords: distribution -- economic -- financial -- haiti -- investment -- management -- mango -- risk -- stochastic -- uncertainty
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: Farmers and agribusiness firms around the world are facing a more complex decision-making process regarding investment in perennial crops than other short-term crops cultivation. Scattered smallholder Haitian farmers or entrepreneurs have competitive advantage to export francis mango during two additional seasons corresponding to higher prices in the U.S. market. A stochastic simulation model was used to incorporate the risk variables price, yield, fruit fly outbreak probability and rate of loss at packinghouse level. Farmers from two complementary seasonal harvesting regions were purposely surveyed to determine expected yield along different harvesting periods. Interviews with managers of export companies assisted in formulating packinghouse capital budget. Finally, stochastic dominance and other key financial output variables were used as decision criteria. The results demonstrated that an investor had a higher likelihood for success by first investing in the two complementary seasonal mango orchards, and exporting when orchards reached early maturity. The results also implied that enforcement of property rights by the Haitian Government along with easier access to capital could improve the likelihood of success
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 Isnel Pierreval.
Thesis: Thesis (M.S.)--University of Florida, 2012.
Local: Adviser: Vansickle, John J.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2012
System ID: UFE0044768:00001

Permanent Link: http://ufdc.ufl.edu/UFE0044768/00001

Material Information

Title: Economic Analysis of Producing and Handling Francis Mango in Haiti to U.S. Market A Financial Risk Management Approach
Physical Description: 1 online resource (77 p.)
Language: english
Creator: Pierreval, Isnel
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2012

Subjects

Subjects / Keywords: distribution -- economic -- financial -- haiti -- investment -- management -- mango -- risk -- stochastic -- uncertainty
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: Farmers and agribusiness firms around the world are facing a more complex decision-making process regarding investment in perennial crops than other short-term crops cultivation. Scattered smallholder Haitian farmers or entrepreneurs have competitive advantage to export francis mango during two additional seasons corresponding to higher prices in the U.S. market. A stochastic simulation model was used to incorporate the risk variables price, yield, fruit fly outbreak probability and rate of loss at packinghouse level. Farmers from two complementary seasonal harvesting regions were purposely surveyed to determine expected yield along different harvesting periods. Interviews with managers of export companies assisted in formulating packinghouse capital budget. Finally, stochastic dominance and other key financial output variables were used as decision criteria. The results demonstrated that an investor had a higher likelihood for success by first investing in the two complementary seasonal mango orchards, and exporting when orchards reached early maturity. The results also implied that enforcement of property rights by the Haitian Government along with easier access to capital could improve the likelihood of success
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 Isnel Pierreval.
Thesis: Thesis (M.S.)--University of Florida, 2012.
Local: Adviser: Vansickle, John J.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2012
System ID: UFE0044768:00001


This item has the following downloads:


Full Text

PAGE 1

1 ECONOMIC ANALYSIS OF PRODUCING AND HANDLING FRANCIS MANGO IN HAITI TO U S MARKET: A FINANCIAL RISK MANAGEMENT APPROACH By ISNEL PIERREVAL A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIA L FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2012

PAGE 2

2 2012 Isnel Pierreval

PAGE 3

3 To Therese and Rosine for their care and love.

PAGE 4

4 ACKNOWLEDGMENTS I am grateful to God I thank my family for uncondit ional support and love during my academic years I would like to thank my committee formed of Dr John VanS ickle and Dr Richard Weldon for their patience and guidance from the conception to the final writing of this research. My gratitude is extended to the United States Agency for Sciences (UF/IFAS) international program and Haiti WINNE R project staff for financial and other contribution s in the pursuit of my master degree at University of Florida. I carry in my heart all the friends that have made my journey unique in Gainesville.

PAGE 5

5 TABLE OF CONTENTS Page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 8 LIST OF FIGURES ................................ ................................ ................................ .......... 9 LIST OF ABBREVIATIONS ................................ ................................ ........................... 10 A BSTRACT ................................ ................................ ................................ ................... 11 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 12 Investment in Perennial Crop ................................ ................................ .................. 12 Problem Statement ................................ ................................ ................................ 14 Researchable Question and Study Objective ................................ ......................... 15 2 ASSESSMENT OF THE MANGO VALUE CHAIN IN THE HAITIAN ECON OMY ... 17 ................................ ................................ 18 Production ................................ ................................ ................................ ............... 19 M arket Channels and Economic Agents ................................ ................................ 22 Analysis of Mango Sector ................................ ................................ ....................... 24 Suppliers Bargaining Power ................................ ................................ ............. 24 High Buyers Bargaining Power ................................ ................................ ......... 25 Low Threat of Substitutes ................................ ................................ ................. 2 5 Low Threat of New Entrants ................................ ................................ ............. 27 Rivalry ................................ ................................ ................................ .............. 27 3 THEORETICAL FRAMEWORK ................................ ................................ .............. 29 Investment Decisions under Unce rtainty ................................ ................................ 29 The Expected Utility Framework ................................ ................................ ............. 30 Real Option Pricing Model ................................ ................................ ..................... 32 Simulation Model under Uncertainty ................................ ................................ ....... 33 Hypothesis ................................ ................................ ................................ .............. 35 4 EMPIRICAL IMPLEMENTATION ................................ ................................ ............ 37 Data Collection and Stochastic Spreadsheet Methodology ................................ .... 37 Farming Operations of Mango ................................ ................................ ................ 39 Hypothetica l Orchard Operations and Budget ................................ ........................ 40 Packinghouse Operations ................................ ................................ ....................... 42

PAGE 6

6 Yield ................................ ................................ ................................ ................. 43 Price ................................ ................................ ................................ ................. 44 Inflation ................................ ................................ ................................ ............. 45 Exchange Rate ................................ ................................ ................................ 46 Rate of Loss of Ma ngo ................................ ................................ ..................... 46 Fruit Fly Outbreak ................................ ................................ ............................. 46 Scenario Analysis ................................ ................................ ................................ ... 47 Scenario 1 I nvestment Only in a Packinghouse ................................ ............. 47 Scenario 2 Investing Simultaneously in a Packinghouse and One Complementary Seasonal Orchard ................................ ............................... 48 Scenario 3 Investing Simultaneously in a Packinghouse and Two Complementary Seasonal Orchards ................................ ............................. 48 Scenario 4 Investing in Two Complementary Seasonal Orchards and in a Packinghouse Fiv e Years Later. ................................ ................................ ... 48 Scenario 5 Investing in Two Complementary Orchards and in a Packinghouse Twelve Years Later. ................................ ............................... 49 Key Output V ariables ................................ ................................ .............................. 49 Net Present Value ................................ ................................ ............................ 49 Probability of Negative Cash Flow ................................ ................................ .... 50 Probability of Losing Real Net Worth ................................ ................................ 50 5 DISCUSSION ................................ ................................ ................................ ......... 51 Scenario Results ................................ ................................ ................................ .... 51 Results of Scenario 1 ................................ ................................ ....................... 51 Results of Scenario 2 ................................ ................................ ....................... 53 Results of Scenario 3 ................................ ................................ ....................... 54 Results of Scenario 4 ................................ ................................ ....................... 55 Results of Scenario 5 ................................ ................................ ....................... 57 Best Investment Strategy ................................ ................................ ................. 58 6 SUMMARY AND CONCLUSIONS ................................ ................................ .......... 61 Summary ................................ ................................ ................................ ................ 61 Conclusions ................................ ................................ ................................ ............ 62 Implications ................................ ................................ ................................ ............. 64 Future Research Needs ................................ ................................ .......................... 65 APPENDIX A HYPOTHETICAL PACKINGHGHOUSE BUDGET ................................ ................. 66 B ESTIMATED COST BUDGET OF HYPOTHETICAL COMMERCIAL ORCHARD PLAIN ARCAHAIE CABARET / LEOGANE ................................ ............................ 68 C DOMINANT STRATEGY ................................ ................................ ........................ 70

PAGE 7

7 LIST OF REFERENCES ................................ ................................ ............................... 72 BIOGRAPHICAL SKETCH ................................ ................................ ............................ 77

PAGE 8

8 LIST OF TABLES Table page 2 1 Distribution of Mango production seasonality according to regions in Haiti. ....... 21 2 2 Matrix of U.S. price and production seasonality in Haiti. ................................ ..... 26 4 1 Budget cost of harvesting for a dozen francis mango ................................ ......... 41 4 2 Establishment budget per hectare for a hypothetical commercial orchard in plain Arcahaie ................................ ................................ ................................ ..... 41 4 3 Assumed probability distribution of stochastic variables ................................ ..... 47 5 1 Summary of the five scenarios ................................ ................................ ........... 51 5 2 Summary of net present value cumulative distribution function of scenario 1 .... 52 5 3 Summary of net present value cumulative distribut ion function of scenario 2 .... 53 5 4 Summary of net present value cumulative distribution function of scenario 3 .... 55 5 5 Summary of net present value cumulative distribution function of scenario 4 .... 56 5 6 Summary of net present value cumulative distribution function of scenario 5 .... 58 5 7 Summary of investment strategy dominance ................................ ...................... 59 A 1 Estimated budget for building and machinery for the packinghouse ................... 66 A 2 Estimated variable cost to handle a francis mango box by the packinghouse .... 67 B 1 Estimated fixed cost of establishment of an orchard in plain Arcahaie or plain Leogane. ................................ ................................ ................................ ............ 68 B 2 Estimated cost of establishment of a commercial orchard in plain Leogane ...... 68 C 1 Summary of probability of losing real net wo rth by scenario at end of 40 years. ................................ ................................ ................................ .................. 70 C 2 Analysis of stochastic dominance with respect to a function .............................. 70

PAGE 9

9 LIST OF FIGURES Figure page 2 1 Contribution of mango to foreign earning among Haiti export commodity from 2002 to 2009 ................................ ................................ ................................ ....... 18 2 2 Volume of production of mango of Haiti from 1980 to 2008 ................................ 19 2 3 Map of mango production in Haiti ................................ ................................ ....... 20 2 4 s. .................. 28 4 1 Simulated yield of francis mango per tree from plain Arcahaie for the 40 years of planning horizon. ................................ ................................ .................. 44 5 2 Financial r isk of negative cash flow from investing in a packinghouse and a complementary orchard. ................................ ................................ ..................... 54 5 3 Financial risk of negative cash flow from investing in a packinghouse and two complementary orcha rds. ................................ ................................ ................... 55 5 4 Financial risks of negative cash flow from investing in two complementary orchards and a packinghouse in next five years ................................ ................. 57 5 5 Financial risks of negative cash flow from investing in two complementary orchards and a packinghouse in next twelve years ................................ ............ 58 C 1 Summary of the cumulative distribution function of the NPV for the five different investment scenarios. ................................ ................................ ........... 71 C 2 Summary of the distribution of negative cash flow probability for the five different investment scenarios over the planning horizon ................................ .. 71

PAGE 10

10 LIST OF ABBREVIATION S ANEM National Association of Mango Exporters BRH Ba nk of Republic of Hait i CSD Carbonated Sugar Drink IHSI Haitian Institute for Statistics and Informatics IR AM In stitute for Research and Appli cations of Methodologies of Development KOV Key Output Variable MARNDR Ministry of Agriculture, Natural Resources and Rural Development MCS Monte Carlo Distribution MEF Ministry of Economic and Finance NGO Non Governmental Organization NPV Net Present Va lue USAID United States Agency for International Development USDA United States Department of Agriculture WINNER Watershed Initiative for National Natural and Environmental Resources

PAGE 11

11 Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Master of Science ECONOMIC ANALYSIS OF PRODUCING AND HANDLING FRANCIS MANGO IN HAITI TO U S MARKET: A FINANCIAL RISK MANAGEMENT APPROACH By ISNEL PIERREVAL August 2012 Chair : J o h n VanSickle Major: Food and Resource Economics Farmers and agribusiness firms around the world are facing a more complex decision making process regarding investment in perennial crop s than other short term crops cultivation. Scattered smallholder Ha itian farmers or entrepreneurs have competitive advantage to export francis mango during two additional seasons corresponding to high er price s i n the U S market. A stochastic simulation model was used to incorporate the risk variables price, yield, fruit fly outbreak probability and rate of loss at packinghouse level. Farmers from two complementary seasonal harvesting regions were purposely surveyed to determine expected yield along different harvesting period s Interviews with managers of export companies assisted in formulat ing packinghouse capital budget. Finally, stochastic dominance and other key financial output variables were used as decision criteria. The results demonstrated that an investor had a higher likelihood for success by first investing in the two complementary seasonal mango orchards, and exporting when orchards reached early maturity The results also implie d that enforcement of property rights by the Haitian Government along with easier access to capital could improve the likelihood of s uccess

PAGE 12

12 CHAPTER 1 INTRODUCTION Investment in Perennial Crop Farmers and agribusiness firms around the world are facing a more complex decision making process regarding investment in perennial crop s The situation is even more complex for the individual fa rmer or entrepreneur in poor countries. They experience a lack of information, financial skills and tools to assess adequately existing agribusiness entrepreneurial opportunities. Success in a multi year commodity orchard demands capital and skilled labor. It also requires careful planning because it is a long term investment involving establishment costs and a fruit maturity period (Marini, 1997). In such a complex venture of modern agriculture it is fundamental that farmers have realistic knowledge about the income and costs of developing their agribusiness ( Araujo 2004). Budget analysis presents some key elements that help growers to determine profits from operations including cash requirements and break even prices (Muraro, 2005). Using a similar method ology, Bakhsh et al. (2006) estimated the costs of establishing a mango orchard and the long term profitability by using discounted cash flows from production and marketing operations. However, without taking into account of alternative opportunities, usin g traditional positive or negative net present value (NPV) as a decision rule may either eliminate an entrepreneurial opportunity or underestimate the risks in some cases (Dixit and Pyndick, 1994). Investing in a perennial crop and in the operation for tre atment, packaging and shipping of this crop offer opportunity for sequential investment either in the production or the handling facility.

PAGE 13

13 An investment in perennial crops with an alternative for including a packinghouse creates options for sequential inve stments. The model has to account for elements such as the recouped cost including irrigation and planting, risk associated with climate and the flexibility to monitor the timing of the investment. For example investing in francis mango in Haiti opens up the opportunity to market either with an orchard or a packinghouse. The investor may delay the lat t er investment and gather sufficient information to improve the likelihood of success. In this manner, t he entrepreneur could avoid some potential losses and gain better returns while minimizing inherent risks by delaying the investment (Brigham and Ehrhardt 2011). Furthermore, Brigham and Ehrhardt argued that the cost of capital estimated for future investments should be lower mainly because of the higher lik elihood of profitable returns or reduced risk from low and negative returns. Similarly, if risks increase during the delay period, the cost of capital will also increase. E lmer et al. (2001) also showed that the discount rate for a 25 year investment commi tment to a 20 acre grapefruit grove is 24% when they account for uncertainty. However, the same project would be feasible for a 6 per cent discount rate based on the conventional net present value rule. Studies of investment in perennial crops have been m ore feasible and accurate where researchers were able to access high quality time series data matching yields and Mapp (1976) have presented an applied simulation proc edure for financial risk management in agribusiness using various probability distributions. Similarly, s tochastic simulation methods have proved to be helpful in the case of developing countries like

PAGE 14

14 Haiti, which typically have inadequate records for agri cultural production and transactions. Problem Statement Mango export value for Haiti is estimated at about $10 million (BRH, 20 12 ) but it is the only major export country where production is relying on smallholder farms with random mango trees. T he coordi nation of output among thousands of farmers increases transaction costs. Therefore, production decline and postharvest losses contribut e to falling competitiveness of Haiti in the U.S. mango market. The country has been unable to take advantage of the clim ate to produce during a yearlong period. Export volume has continued to decline steadily on average by 2 % during the past decade. The Inter American Development Bank (2010) questions the sustainability of current practices and USAID (2011) estimated that p ostharvest loss es estimated at 40 % are detrimental for the prosperity of the industry. In addition to the mentioned problem, the packinghouses for exporting mangoes have suffered from a deficit in management. Following the breakout of fruit fly in 2007 disrupting the exportation to the U S market, price variability of inputs in the market and increasing cost of processing mango for export have caused six processing plants to go out of business (Wiener, 2009). Farmers and potential investors in Haiti kn ow little about the economics of growing mango under uncertainty. Existing reports on mango research in Haiti describe more the market value chain. Unfortunately, little research exists on examining the economics of growing and exporting mango and characte rizing financial risks. Even though new techniques for grafting and improved logistics have been suggested in current reports (USAID, 2010), their costs and returns have not been examined in a financial farm

PAGE 15

15 management framework. Beyond traditional risks s uch as yield and prices, because mango is a major perennial crop investment in mango is subject to other uncertainties. Bad weather such as a hurricane can disrupt a growing plantation. Disease outbreaks during the life span of the orchard could also disr upt production. Researchable Question and Study Objective This study aims to determine the best strategy and the likelihood of success for investment in producing and exporting mango from Haiti to the US market. It compares different investment strategies starting either with a commercial orchard, a packinghouse or both including different timing for each investment with the objective to export to the US market. earthquake of 2010 offers a momentum to undertake an ex ante investment feasibility study. The Mango industry was identified by both a national and multilateral agency as a priority for economic development of Haiti In addition, it responds to the need s of farmers and the average entrepreneur who wants to learn about the economics of growing mango but who are unable to access information about the cost and profitability of investment. Obstacles include absence of financial information and limited access by farmers to loans for financing their rural entrepreneurial initiatives. The capacity to build productive partnerships with the public and interested private sector is also limited by the general low education level of farmers and a deficit of organization. Therefo re, farmers continue to rely on random mango trees in their backyards inherited from past generations. Most observers agree that Haiti needs to modernize its agricultural practices in order to maintain its share in the U S market for mango. The main objec tive of this

PAGE 16

16 study is to provide economic information regarding investment in the business of exporting mango from Haiti to the U S market. Second, it will provide information regarding the establishment cost for a hypothetical orchard and packinghouse. T hird, it will compare different investment alternatives taking account of the introduction of risk and the flexibility of when to invest. Finally, the ranking of those investments will provide managerial insights for decision making. The following study is organized in three sections. The first section provides an assessment of the value chain for mangoes in the Haitian economy. The second section outlines the theoretical framework for the evaluation. The third presents the empirical results. Finally, we di scuss conclusive findings.

PAGE 17

17 CHAPTER 2 ASSESSMENT OF THE MA NGO VALUE CHAIN IN T HE HAITIAN ECONOMY The purpose of this chapter is to analyze the importance of mango in the economy of Haiti and to assess the overall agribusiness sector involving mango. The importance of this crop in the economy is a fundamental factor to determine potential government support or regulation of th is sector. Entrepreneurs must consider potential intervention of the government in a relatively small developing country dominated by agriculture. Moreover, they must scrutinize interactions of players in the value chain and the potential margin to be captured. In fact, from a broad point of view, an industry is a composite result of social tendencies, economic forces and circumstance s that have Porter (2008) emphasizes the importance of considering the five forces shaping the industry when considering a strategy. This approach underlines that the fundamenta l analysis of the overall industry is more important than focusing only on competitors or other factors such as growth. Similarly, we are considering the mango threat o f new entrants and substitutes, and the existing rivalry among competitors. and the production practices use d to grow mangoes in Haiti. The second part presents the market c hannel and the third part briefly assesses the mango industry regarding economic context.

PAGE 18

18 Mango is the single most valuable export has a market share of 2% of the volume of mango consumed in the U.S. market (USDA, 2010). Haiti is one of the top twenty most important producers of mango in the world and the sixth largest exporter to the United Stat es of America. In the Haitian agriculture, mango is the second largest volume of fruit produced after banana (FAO, 2012). One of the main contributions of francis mango is foreign earnings. Mango value chain employs around 15,000 workers. Mangoes provide f ood security as a main source of vitamin A and contribute to forestation. Figure 2 1 Contribution of m ango to foreign earning among Haiti export commodity from 2002 to 2009 Adapted from Haiti, Republic, Administration Gnrale des Douanes (AGD). 2010. Exportations par grande catgorie de 2002 a 2009. Unpublished data Mango has taken an increasing share in the primary commodity export basket of Haiti over the last ten years. Mango export value represented 22% of the total expor t earning s of the country in 2002. By the end of 2009, export earnings progressively increased to 41% (Figure 2 1 ). Between 2002 and 2009, while some export commodities either declined or stagnated, mangoes grew from $7.93 to $10.21 million in foreign earn ings. Export of mangoes from Haiti to the U.S. market has increased on

PAGE 19

19 average by 2 % in value since but it has decreased by 2 % in volume in the last 10 years with significant volatility in the annual based price comparison. Production The production trend for mango in Haiti has been declining for the last 20 years. Since 1998 production has not reached 250 ,000 metric tons (MT) (figure 2 2 ), which is largely under potential capacity of the country given production exceeded 363 000 MT in 1985. From the peak of 350,000 MT in 1986, production plummeted by more than 55 % between 1989 and 1993 due to politic al turmoil and an economic embargo from 1991 to 1993 imposed by the U S Government Between 2000 and 2010, the production volume fluctuated between 250,000 and 300,000 MT per year. Francis mango represents 15 % of total mango production in Haiti and less than 33 % of its total production is exported. The production system is one of the structural factors that has contributed to export decline. Figure 2 2. Vo lu me of production of m ango of Haiti from 1980 to 2008. Adapted from Food and Agriculture Organization of the United Nations. 2012. FAOSTAT. Rome, Italy: FAO. http://faostat3.fao.org/home/ind ex.html#compare

PAGE 20

20 The production of francis mangoes is scattered among small farms and a few farms of 1 hectare ( h a) or more. Mangoes are grown on plains with altitudes ranging between 0 and 400 meters, with rainfall levels attaining between 800 and 1200 m illimeter s per year. Francis mango production can be implemented in six different regions of Haiti. Climate diversity provides the country with a unique competitive advantage to spread the output along a yearlong calendar. In favorable conditions, the aver age mango yield reaches 10 metric tons per ha (MT /ha ) at a density of 100 trees per hectare (USAID, 2010 ). The seasonal production is spread between two and four months among the following four regions: N orth, West, C entral Plateau and Artibonite (T able 2 1 ). Figure 2 3 Map of mango production in Haiti Reprinted by permission of Buteau, J.M. 2011. Manager and co owner of JMB.SA. http://www.mango haiti.com/haitimap.htm

PAGE 21

21 Table 2 1 Distribution of Mango production seasonality according to regions in Haiti Oct. Nov. Dec. Jan. Feb. Mar. Apr. May June July Aug. Sept. West Department Leogane x x x Cabaret x x x Arcahaie x x x Fond Blanc x x x x Croix des Bouquets La Plaine x x x x Center Department Saut d'Eau x x Mirebalais x x Artibonite Department Gros Morne x x x x St Michel de l'Attalaye Pont Sonde South East Department Jacmel x x Cayes Jacmel x x Margot x x South Department Auin St Louis du Sud x x x St Jean du Sud x x x Camp Perrin and Plaine des Cayes x x x Adapted from U nited S tates Agency for International Development (USAID). 2010. Mango Forum Report: Export 5 Million Cases of USDA Certified Mangoes by 2015. Haiti. http://pdf.usaid.gov/pdf_docs/PNADW226.pdf

PAGE 22

22 T he regions of Gros Morne and Plaine du Cul de Sac have the longest period of mango production and produce the highest yield. Production in the Plaine Cul de Sac reg ion has decreased due to urban pressure while the Gros Morne region remained prominent claiming one third of the total export of mango. There is only one important mango o rchard of more than 10 hectares in the Gros M orne region; others are small farms esta blished in mountainous areas where other crops form a diversified farming system. Transport of mangoes from small farms to collection centers is particularly difficult due to mountainous landscape and the absence of infrastructure Regions from the South an d West departments offer better opportunities for transportation and establishment of commercial orchards. The main areas of the West department for production of mango are Cabaret, Arcahaie, Leogane and P lain du Cul de Sac. Small farms in W est department are located closer to packinghouses and have better road access. The fruit company, CARIFRESH SA, has developed with the support of Inter American Development Bank (Inter American Development Bank, 2010) a commercial orchard in Plain of Cul de Sac. Another company, named JMB SA, has pla n ned to develop an orchard in the region of Zoranje of W est department ( Buteau 2011). Market Channels and Economic Agents The value chain of mango in Haiti includes three main operations: production, postharvest handling a nd marketing. Small farmers who could be either members of cooperatives or independent growers are the backbone of the production system. Haitian diasporas and two export companies own a few commercial orchards. Most of the production is organic T herefore, input suppliers provide mainly compost, seedlings,

PAGE 23

23 and labor from the local market but in some rare cases fertilizers and fruit fly traps are imported from the U.S. The postharvest operations provide economic opportunities for several economic agents. The independent growers are connected with the wholesalers who provide some cash advance for mango fruit. The main wholesalers are registered at the Ministry of Agriculture and they are directly linked with the manager of a packingh ouse. During harvesting peri od, growers contract at least one grafter to climb the mango tree and one receiver to catch the mangoes. Producers enable mango trees to grow very high for security a constraint which makes the task of harvesting more difficult. Following this process, th ey rent donkeys to transport harvested mangoes to be sorted at a hub for wholesaling. At this postharvest center, some agents wash, wipe and classify qualified mangoes that are bought either by cooperatives or wholesalers. Then, mangoes are transported by truck to the packinghouses in the cities. Truck drivers are independent and are contracted either by the packinghouse or by the wholesaler. The marketing operations are largely dominated by nine packinghouses situated in the metropolitan region in proximit y to the seaboard. Fruit & Legume SA is the only packinghouse that is located outside of the West department in a city named Saint Marc. Grading of mangoes is processed by employees of the packinghouse under the supervision of a USDA and Ministry of Agricu lture agent. M angoes that do not meet export standards are sold on the local market and the ones satisfying quality and sanitary norms for the U.S. market are handled and shipped to U S p orts (Miami, Tampa, New York) to be distributed by importers as fres h products.

PAGE 24

24 Analysis of Mango Sector from the point of view of a new packinghouse investor. We evaluate along the value chain : the bargaining power of suppliers including buyers then threat of substitutes, the rivalry among competitors and the threat of new entrants. Following this analysis the competitive advantage of Haiti in producing and handling mangoes for the US market is assessed. Suppliers Bargaining Power Investment i n a packinghouse has two principal inputs to secure for its successful operation: fresh mangoes and labor. The suppliers of labor for packinghouse operations or in the commercial orchards have low bargaining power due to an abundance of workforce and low q ualification required. The current salary of workers on a packinghouse line is r oughly a minimum wage of $ 5 per day. Part time labor in commercial orchards earns an estimated $ 3 per day. T he bargaining power of mango suppliers is also limited compared to current packinghouse managers. Generally, the packinghouse managers buy mangoes at the same price from all wholesalers and cooperatives. Following the required quality inspection, packinghouse managers pay wholesalers based on the volume of mangoes receiv ed In the case where a significant fruit fly infestation or post harvest issue is identified, the loss of fruit will be charged back to the mango wholesaler or other supplier (such as the cooperatives). As the volume of available mango declines, more mana gers of packinghouses use forward contracts by providing cash advances to secure their raw inputs. Moreover, the bargaining power of mango wholesalers to packinghouses is decreasing as small farmers are grouping themselves into cooperatives to deliver

PAGE 25

25 dir ectly to packinghouses. Farmers grouped in cooperatives may negotiate on higher volume including the ability to switch contracts to another company offering better terms. For example, IRAM reported in 2004 that the earn ings of a grower increased by 3% when joining a cooperative (Bellande, 2005) Because mangoes are perishable, wholesalers and packinghouses have to manage the supply with forward contracts and have to provide incentives to agents at the bottom of the value chain. Consequently, wholesalers of mangoes may have considerable bargaining power on a new packinghouse because the production is limited and they may have established long term contracts or may be loyal with other packinghouses. Based on the law of supply and demand, the cost of raw materi al may be higher for a new company who must buy their way into the supply chain. It is worth noticing that the suppliers bargaining power decreases as packinghouses are integrating backward by developing their own mango orchards. High Buyers Bargaining Pow er U S importers of fresh mangoes have significant bargaining power in buying mangoes fr o m the Haitian market. For example, a single U.S. wholesaler is the buyer of certified organic mango es Managers of packinghouses h ave more leverage dealing with non o rganic mango importers The Haitian exporters of mango can factor in the uniqueness of f rancis mango variety to increase their bargaining power during the high sale season in the U.S. market. Low Threat of Substitutes Francis mango from Haiti is unique in the U S market compared to other mango varieties imported from other countries in Latin America. In addition, Haitian exporters of mango can count on an optimistic outlook for the fruit industry in the U S economy

PAGE 26

26 against substitutes such as Carbonated S ugar Drinks (CSD). Policy programs could increase the consumption of fresh fruit in the U.S Panteva (2010) forecasted a 3% annual growth of revenue for fruit and peanut s d uring 2010 to 2015 period. His report also predicted that improvement of personal income would shift consumer preference to fresh fruit. The shift of consumer preference to fresh fruit would be an advantage over the traditional substitutes such as CSD. An attractive economic outlook will more likely attract new investors to the market. Table 2 2. Matrix of U.S. price and production seasonality in Haiti Jan. Feb. March Apr. May June July Aug. Sept. Oct. Nov. Dec. Area Low price Medium price Low price Medium price High price Medium price Leogane X X X Plain du Cul de Sac X X Plain arcahaie X X X Central Plateau X X Gros Morne X X X X South depart ment X X Adapted from U.S. Department of Agriculture, Agricultural Marketing Service (USDA AMS). 2010. Fruit & Vegetable Marke t News. Washington DC http://www.marketnews.usda.gov U nited S tates Agency for International Developmen t (USAID). 2010. Mango Forum Report: Export 5 Million Cases of USDA Certified Mangoes by 2015 Haiti http://pdf.usaid.gov/pdf_docs/PNADW226.pdf The Haitian export mango sector has a c ompetitive advantage compared to the other five main suppliers in the western hemisphere: Brazil, Mexico, Guatemala, Ecuador and Peru. Francis mango responds to the five first criteria that U.S. customers

PAGE 27

27 identify in buying mangoes: price, ripeness, appear ance, freshne ss and quality (Ward, 2011 ). Moreover, the country has the potentiality to produce f rancis mango during a yearlong season if necessary investments are made at the production level. Such initiatives will empower national exporters to capture hi gh and medium price markets in the U.S. market during low supply periods of international competitors including October to December and February to March periods ( T able 2 2). In addition, the country can count on the uniqueness of the f rancis mango variety (Lundahl, 2004). Low Threat of New Entrants Mango export c ompanies are rallying together under the umbrella of a sole organization to d efend their common interest. This organization, National Mango Exporter Association (ANEM) coordinates with the USDA re garding safety standards and decides how fees for USDA inspection are distributed bet ween the export companies. The members of ANEM also cooperate to import technical expertise from Latin America countries for equipment maintenance, which decreases the cos t per enterprise. These elements present a subtle barrier of entry. These barriers coupled with upfront intensive capital investment required to e nter this industry lower threat of new entrants. As the existing packinghouses do not always meet the potentia l demand, rivalry between existing firms is based more on access to fresh man goes and international clients. Rivalry Nine packinghouses are competing in Haiti for market share in the mango industry. The main challenge of those companies is to secure fresh mango during the high price season while the production keeps declining. Therefore, packinghouses have to compete fiercely to keep their suppliers loyal and are constantly seek to take

PAGE 28

28 additional share of the local production. In the weeks preceding harves ting season, packinghouse managers have to send key officers on small farms to assess maturity and quality of the mangoes. Consequently some companies are intending to integrate backward by developing orchards at their own to secure supply and to decrease transaction costs. Francis mango is unique in U.S. market, but as a small supplier Haiti can not compete on price with Latin American competitors during the peak season. Figure 2

PAGE 29

29 CHAPTER 3 THEORETICAL FRAMEWOR K Investment Decisions under Uncertainty Returns on investment in perennial crops and handling facilities are constrained by longtime horizon planning and uncertainty. In long term investments, after a strategic choice is ma de the future outcomes are not immediately observed. In an uncertain environment the decision maker may have difficulty assessing the probability of certain outcomes (Richardson, 2006) In this non probabilistic case the following decision criteria are use d: mini max criterion, maxi min criterion and mini max regret criterion (Groebner et al, 2008). Focused on the probabilistic framework, long term investment in agriculture has been studied from the expected utility framework, the option pricing model (Hild ebrandt and Knoke, 2011) and the financial simulation model (Clark, 2007). According to Zimmerman (2000), investment decisions under uncertainty should be modeled based on the quality and quantity of existing information (as cited in Hildebrandt and Knoke 2011). This study uses the financial simulation framework and the literature review includes the expected utility framework including the option pricing model. Expected utility theory will illustrate criteria to compare different sets of investments. The p robabilistic framework approaches compare estimated probability distributions of different alternative investments. Generally, they use repeated random samples of the outcomes to derive the probability distribution known as Monte Carlo Simulation Technique (MCS). For risk assessment in financial evaluations, MCS is mainly applied due to its ability to integrate different risky inputs (Hildebrandt and Knoke, 2011). Ferson and Ginzburg (1996) mentioned that input variables are more

PAGE 30

30 likely unknown in the MCS m odel and uniform distribution is generally used (as cited in Hilderbrandt and Knoke 2011). The Expected Utility Framework The expected utility model is based on the assumption that expected utility of outcome s is maximized (Richardson, 2006 ). According to Schoemaker (1982) expected utility models conduct a holistic evaluation of alternatives, then elaborate operations of outcomes distinctly from probabilities and use a multiplicative transformation to combine probabilities and outcomes. Von Neumann and Mor genstern (1944) have elaborated the axioms defining a cardinal utility function. They also formulated the mathematical rationality of maximization of expected utility as a criterion (Schoemaker, 1982). The expected utility framework helps to describe attit ude of the decision maker towards risk using different functions. First, a linear increasing utility function illustrates the risk neutral individual C onvex increasing utility function s describe risk seeking individual s and concave increasing utility func tion s define for risk averse individuals. A decision maker may change attitudes over the planning horizon and these mixed attitudes towards risks may be illustrated by a third degree polyn omial utility function behavior (Hildebrandt and Knoke 2011). In fa ct, Post s preferences might fluctuate over the planning horizon. Schoemaker (1982) has intensively discussed the utility functions, the probability and the measurement of outcomes variants. Introduction of stochastic dominance criteria has helped to rank investment alternatives taking account of the attitudes of the decision make towards risk.

PAGE 31

31 The theoretical and empirical extensions in economics and finance of the stochastic dominance framework, accordin g to Levy (1992) were developed after the publication of four independent papers (Hadar and Russell 1969; Hanoch and Levy 1969; Rothschild and Stiglitz 1970; and Whitmore1970). The establishment of preference between two investment alternatives X and Y wi th cumulative functions F and G may be determined respectively by first stochastic dominance (FSD), second stochastic dominance (SSD) and third stochastic dominance (TSD). The FSD is valid invariantly of the decision maker risk preference. FSD, SSD and TS D require the respective rule: FSD: F (x) is preferred to G (x) or F dominates G, where x is the variable for economic decision, if: G (x) x, G (x) F (x) > 0 for at least one x (3 1) SSD rule: (3 2) (3 3) The SSD criterion has more power because it assumes the decision maker is risk averse. Contrary to TSD, FSD and SSD are often used in investment analysis and are incorporated in simulation programming system s to compare a set of alternative investments. FSD and SSD criteria will be used to determine the best investment strategy in this study. In the following paragraph we will explain the flexibility of

PAGE 32

32 investment in a long term horizon planning which is not taken into account in the classical investment model. Real Option Pricing Mo del The classical framework of analyzing investment fails to take into account the flexibility of the decision maker along the planning horizon (Dyxit and Pyndick 1994). Dixit and Pyndick (1994) criticized the classical approach of investment using the s ole criteria of a positive net present value because this method assumes this a now or never proposition. By analogy to the financial option, the real option is a common way to quantify the economic value of real investment flexibility (Hildebrandt and Kno ke, 2011). Financial options are contracts providing the right to sell (put options) or buy (call options) a given asset at a given price during a time interval (so called American options) or at a determined time (so called European options). The real op tion approach accounts for uncertainty, irreversibility and ability to delay the investment and acquire information for the option to invest (Dixit and Pyndick 199 5 ). According to the Dixit Pyndick model, for example, we have to estimate the expected value function of the investment, F (V), for a given period of time, n: n t=1 R t / (1+ ) t I 0 ] (3 4) The right side of the equation is the expected returns from initial investment I 0 due to uncertainty in the variables used to estimates the a nnual net cash flow. There are three stochastic processes for V: the geometric Brownian motion (GBM) with and without drift, and a mean reverting process (Dixit and Pyndick, 1994). The result of ng uncertainty (Dixit and

PAGE 33

33 Pyndick, 1994). This rate is the minimum rate that an uncertain investment has to return to be accepted as profitable (Tozer, 2009). However, the Dixit Pyndick model is hard to estimate through required economic model due to rudim entary quality of existed data ( August and Purvis, 1995). Nevertheless this model gives insight to evaluate alternatives and taking account of the sequential investment options regarding this research. The flexibility of the investor to invest either in a commercial orchard or a packinghouse or both with different time intervals should be considered. In fact, in constraint of advanced historical data, Tozer (2009) suggested to use ex ante simulation approach for such economic analysis. Simulation Model unde r Uncertainty The purpose of simulation in risk analysis is to estimate distributions of economic returns for alternative strategies so the decision maker can make better management decisions (Richardson, 20 06 ). As mentioned by Clark (2007), stochastic com puter simulation models have been used to analyze capital expenditures and agricultural management scenarios under conditions of uncertainty (Richardson and Mapp 1976 ; Richardson et al., 2000; Crawford and Milligan, 1982 ; Featherstone et al., 1990; Griffi n & Thacker, 1994; Fumasi 2005). A ccording to Richardson and Mapp (1976), using a stochastic simulation approach provides advantages in reporting the feasibility of a project as a distribution of possible outcomes. Based on their recommendations, a financ ial simulation model may be developed in five stages. For clarity, the identification of the problem is added to the top of their stages. Richardson (20 06 ) present s the simulation approach in a pyramidal sketch. It starts from identifying the Key Output V research question. I have modified his approach by adding the identification of the

PAGE 34

34 problem at the top. Using this approach each investment scenario will determine and obability of negative cash flow and change in real net worth Following this stage, the identification of intermediate results refers to the different financials statements or tables required to calculate the PV are derived from the income statement, cash flow statement, financial ratios and balance sheet. Information for financial statement analysis is gathered from capital and production budgeting including receipts at firm level and is entered into the dif ferent accounting or econometric equations. Figure 3 1. Simulation p yramid A dapted from Richardson J.W., Schumann, K.D. & Feldman, P.A. (2006) Simetar Simulation for Applied Risk Management, Version 2006 Texas A&M University, an d College Station, Texas The stochastic spreadsheet methodology integrates input variables and core economic theories such effects of inflation on capital flows, tax implications, changes in productivity over time and changes in real prices (Malcolm, 2004 ). Exogenous variables, (for example, interest rate and inflation) are retrieved from reliable sources and

PAGE 35

35 endogenous variables (such demand or production costs) are estimated through the model. The past distributions of stochastic variables are estimated from historical data in order to project the pattern during the forward planning period for the project. For this research the planning horizon is 40 years and historical data for yields, prices of mango boxes in the U S market, exchange rate and average export volume by packinghouse will be simulated to be incorporated in the financial spreadsheet. Finally, experts and stakeholders of the given industry validate the model. Hypothesis In order to conduct this study the following assumptions were considere d: Productivity of francis mango trees from one to four years is considered null; maturity is reached between the seventh and twelfth years and finally full production is attained between the thirteenth to fortieth years. Life span of the project is assum ed to be 40 years. All equipment of the packinghouse and material inputs of the commercial orchard are considered brand new. The full capacity of the packinghouse is 10,000 boxes per day where each box contains 4.5 kg of fruit. The profitability of the i nvestment will depend on the U.S. market price, rate of loss at packinghouse, yield, the probability of fruit fly breakout, investment value and operation costs. Investors have the opportunity to invest in either only a packinghouse or in the packinghouse and commercial orchards from two distinct regions: Arcahaie for harvesting period from January to March and Leogane for harvesting period from October to December. They may also buy mango from traditional wholesalers during high season collected mainly fro m the region of Gros Morne. The analytical framework will help to test the following hypothesis: If an entrepreneur invests only in a packinghouse to export mango from Haiti to the U.S. market, under the current production system, the real net worth value of the firm will be negative at end of the planning horizon. Diversifying the production by region will increase the likelihood of positive net present value.

PAGE 36

36 Diversifying the production by region and starting an export processing facility during early maturity of the orchards will result in the most stochastic dominant investment strategy

PAGE 37

37 CHAPTER 4 EMPIRICAL IMPLEMENTA TION Data Collection and S tochastic S preadsheet Methodology The point of departure is to use financial simulation modeling to assess t he probability of Net Present Value (NPV) outcomes over a 40 year planning horizon. D ata was collected at the farm and packinghouse level because of lack of historical data regarding g rowing and handling operations. This information helps to determine the example observed yield at different age and historical export volumes will help in developing stochastic yield and stochastic export volume during high season for the model. The data collection process involve s collecting information at three levels: 1) operation of producing mango; 2) capital and operation budget of handling mango including packinghouse and 3) marketing information on the U.S. market and other exogenous variables. Direct and indirect sources w ere consulted to gather information to develop the model. In Haiti, 30 farmers who have at least 20 mango trees were purposely surveyed. Five of the ten packinghouse managers or owners were interviewed to assess capital investment, operations and productio n capacity of the industry. Stakeholders of the industries were consulted to establish and review the model. Those experts included professionals from the Ministry of Agriculture in Haiti, managers of packinghouses and the office responsible for francis ma ngo from United States Agency for International Development (USAID) funded project named Watershed Initiative for National Natural Environmental Resources (WINNER). Interviews with the 30 growers of mango were stratified in two different regions: plain Arc ahaie and plain of Leogane. The Ministry of Agriculture and WINNER project

PAGE 38

38 provided a list of Cooperatives in those regions from which they retrieved the 30 most important growers who have at least 20 trees. Some of the growers did not have the minimum num ber of trees; therefore we used wholesalers in the region to identify additional farmers to complete the sample size. Those two regions were surveyed because they complement the high season of April September to provide a quasi yearlong production. The un it analysis selected is a farm planted with at least twenty francis mango trees. Contrary to the conventional measure of kilogram per hectare ( Kg/h a), the yield is measured by dozens of francis mango produced per tree per year for normalization purpose. Th is measure allows for using different dimensions for farm size. Information was collected regarding the cost of production and transaction costs. The survey gathered information regarding the minimum, average and the maximum of mango yield per tree for the period s between 5 and 7 years of age 8 and 12 years of age and between 13 and 40 years of age Yield per tree during the period 0 and 4 years of age is considered null. The budget cost was established for an orchard equivalent of 100 trees planted on one hectare. Interviews with owners of the five packinghouses and observation of their facilities assisted in defining the required capacity and equipment for a packinghouse. The typical packinghouse working at full capacity handles 10,000 boxes of mangoes pe r day where each box weights 4.5 kilograms. An equipment and installation estimates was asked to two U.S. companies for a new facility with the same capacity from different US companies. The proposal of Agri Machinery Parts was retained because they assemb led at least two out of the five packinghouses surveyed and this proposal was competitive. This information was reviewed by packinghouse managers and provided

PAGE 39

39 the capital investment and operations budgets to establish a packinghouse (Appendix A). Additiona lly, we consulted some indirect sources for several exogenous variables. The Fruit Production office of Ministry of Agriculture of Haiti provided data regarding the volume of boxes of francis mango exported, the percentage of mangoes rejected at the packin ghouse and the cost of fruit fly treatment per orchard. The traps for detecting fruit fly are free of charge and paid by the fruit fly program of the Ministry of Agriculture. Financial information including inte rest rates, inflation rates and exchange rate s are retrieved from the World Bank Inter American Development Bank and the Centr al Bank of Haiti. The price of mangoes in the U.S. market came from USDA Agricultural Marketing Service. Th i s data are analyzed to determine stochastic variables and to estim ate a deterministic budget for production and the handling of mangoes for export. Farming Operations of Mango Geographical positions of the two complementary orchards are strategic. It is assumed that the packinghouse will be located along the coast of P ort au Prince to benefit from proximity with the port. In fact, the modeled site of the packinghouse is located 40 and 32 kilometers respectively from the production areas of plain Arcahaie and Leogane (IHSI, 2007). Arcahaie is a coastal county of 408.7 k ilometer s square dominated by vast plain. On average, 84.2% of the population of Arcahaie lives in rural areas. They enjoy a calm climate that become s cooler during the period of December to February. Its rivers (mainly the Matheux and Courjolle) facilita te irrigation in the plain but wells as water storage are frequently used in the community. Agriculture is the main economic activity of Arcahaie and it is known as the premier area of production for bananas. The

PAGE 40

40 habitants also practice fishing and animal husbandry. Farmers reported that mango production in the area has been declining because short term crops such bananas provide cash faster and their harvesting period conflict with the packinghouse export season. Leogane is located to the south of Port au Prince and covers an area of 385.23 kilometer s square. Four rivers cross the plain of Arcahaie and they are used to irrigate sugarcane that has been the main agricultural activity besides fishing. Leogane has experienced development of mango orchards durin families claiming ownership over the properties. Recent efforts by some international NGOs have helped farmers to plant some mango t rees in the regions. Hypothetical Orchard Operations and Budget The hypothetical orchard has five main operations: planting, composting, irrigation, pruning, harvesting and transportation. Planting consists of preparing the soil, digging holes and plantin g the trees on an equal average distance. Farmers apply compost after planting and they either buy or prepare themselves the compost from organic materials. Irrigation varies for different settings. Two orchards of more than 5 hectares were observed that h ave irrigation canals, others relied on the rain y season and the use of watering cans at their early age. Orchard operations require continual maintenance. Pruning starts at the age of four and costs on average $ 0.63 per tree. Traditional farmers do not p rune their trees for security reasons because higher trees are less exposed to robbery. Harvesting operations are therefore more difficult for traditional farmers. The harvesting cost per tree of a dozen of mango including transport to the postharvest cent er is estimated at $ 0.29 (T able 4 1). The variable cost per hectare for

PAGE 41

41 establishment of a hypothetical orchard in Plain Arcahaie is estimated at $ 3,408 .50 and the drip irrigation system represent s more than 57% of this cost (Table 4.2). The variable cos t for the orchard in Leogane is slightly less expensive due to access of sugarcane raw compost (Appendix B). The fixed cost for establishment of the mango orchard is estimated at $ 75,950.00 (Appendix C). Table 4 1 Budget cost of harvesting for a dozen f rancis mango Table 4 2. Establishment budget per hectare for a hypothetical commercial orchard in plain Ar cahaie Item Unit Price ($) Quantity Amount ($) Direct Expenses Fertilizer Bag 37.50 3 112.50 Mango trees Each 2.50 100 250.00 Labor for land preparation Tractor rental /ha 250.00 1 250.00 Labor of hole for tree Tree/ha 0.25 100 25.00 Labor fo r land clearing Ha 125.00 3 375.00 Labor for planting trees Tree 0.08 100 7.50 Labor to apply fertilizer Tree 0.05 100 5.00 Labor for pruning T ree 0.63 100 62.50 Drip irrigation system Each 2 000.00 1 2 000.00 Hand hoe Each 5.00 2 10.00 Pruning kn ife Each 3.75 2 7.50 Machete Each 1.25 2 2.50 Shears Each 0.50 2 1.00 Ax Each 18.75 2 37.50 Nebulizer Each 87.50 3 262.50 Total variable cost /ha 3 408.50 Item Cost/dozen ($) Labor Picking 0.05 Labor Holding 0.04 Labor Washing 0.05 Transport to postharvest center 0.15 Total harvesting cost 0.29

PAGE 42

42 Packinghouse Operations The budgeted packinghouse has a capacity to handle 10, 000 boxe s of mangoes per day and it is connected to nine main operations. Transport of harvested mangoes to the packinghouse is the first operations influencing loss of fruit. At the packinghouse all the mangoes go through the washing machine where employees cut e xcessively long peduncle and c u ll out spoiled fruits. At the end of the sorting machine, worker s put mangoes of size 8, 9, 10 and 12 in a metal crate and immerse it in hot water. The hot water treatment is managed by electronic software according to USDA r equirements. A crane takes the metal crate to a second basin to lower the temperature of the fruits. The refreshed mangoes are then taken through another sorting conveyor for sorting, grading and packaging. The mangoes are marketed in box es of 4.5 kilogram s and may contain 8, 9, 10 or 12 units of francis mangoes. In this study, each box is assumed to contain 12 mangoes. The cost of the mango box is estimated to $ 3.81 including all exportation transport fees. The pallets contain 216 mango boxes and 26 palle ts fill a shipping container of forty feet. The container including the pallets are disinfected by a local pest control company and conditioned according to USDA standards. A cold storage facility with the capacity of four containers is maintained at 55 de gree Fahrenheit to condition the packaged mangoes until shipment to a U S port. Equipment in the packinghouse is capital intensive compared to other operations of the mango value chain. The budget f or equipment including building the facility is estimate d to be $ 1,445,497.54 The capital budget of investment in the packinghouse is detailed in Appendix A.

PAGE 43

43 Yield The yield of a mango orchard is dependent o n the age of the trees. Yields were estimated from growers data based on their experience in respecti ve targeted areas. Growers provided estimates for minimum, average and maximum yields of their trees between 0 and 4 years, 5 and 7 years, 8 and 12 years, finally between 13 and 40 years of age A uniform distribution was assumed for yield and each stochas tic variable is simulated according to a GRKS distribution. Gray, Richardson, Klose and Schuman (Richardson, 2006) developed the GRKS distribution to simulate subjective probability distributions based on minimal input data. It is a closed distribution for m based on minimum, midpoint, and maximum points provided by a business manager or a farmer (Richardson, 2006) In order to simulate the yield over the planning horizon, for each interval ( between 0 and 4 years of age, 5 and 7 years age, 8 and 12 years of ag e, 13 and 40 years of age ), the following formula was used in the simulation software, Simetar : X= GRKS (Min, midpoint, Max) + USD (4 1) where uniform standard deviation (USD) is a random number generated by the simulation software Simetar (Richardson et al, 2006) The yield of the orchard contributes to determining orchard revenue s and it also helps to define the expected availability of mangoes for the packinghouse. In the case of a vertically integrated firm, the orchards production may fail, meet or exceed the packinghouse capacity. The volume of mango handled by the company per year is given by: V =X O (1 p i ) + Xg*(1 p i (4 2) where X O + Xg X c and Xg = 0, if and only if X O X c (4 3)

PAGE 44

44 X c : the maximum capacity of mango handled by the packinghouse X O Xg: production available from wholesalers or cooperatives p i : rate of loss at farm gate for new orchard and packinghouse p i Figure 4 1. Simulated yield of f ran cis mango per tree from p lain Arcahaie for the 40 years of planning horizon. Price The price of mangoes on the U.S. market is derived from historical data published by USDA Agricultural Marketing Service of United States Department of Agriculture (USDA AMS 2010) The expected price of mango is high during the October to November period and medium during the months September, December, and February Finally, the price is low for the month of January and the April to August period ( Table 3

PAGE 45

45 2 ). The low price of the kilogram of mango ranges from $ 2.11 to $2.55, the medium price fluctuates from $2.56 to $2.98 and the high price ranges from $2.99 to $3.41 (USDA AMS 2010; Castaeda et al, 2011). Statistics from historical prices for exported francis mango box fro m March 2009 to August, 2011 were used to develop a GRKS distribution for the planning horizon. In order to reflect the seasonal price for each year, a premium was added to the price of high season. For example if P i is the simulated price for a 4.5 kg bo x of francis mango per year i for the high season month s (March to September), then the price for the period January to March will be: I = P i + x (4 4) and the price for period from October to December will be : I = P i + y (4 5) The constants x an d y are the average premium s f or the January to March and October to December periods respectively compared to the price in the April to September period. For the reference year of 2010, it was estimated that the premium for the January to March period w as $ 2.82 and the premium for the October to December period was $ 1.95 (USDA AMS, 2010). A firm that exports during the yearlong season would receive three different prices. A GRKS distribution is estimated to forecast P i using historical price s from Apri l I I are applied respectively to export volume during the periods of January to March and October to December. Inflation Inflation is introduced to some variable inputs to approximate the effective cost throu gh the planning horizon. The inflation rate s for the base year 2011 for fuel and labor was retrieved from the Ministry of Finance of Haiti (MEF, 2011) A conservative

PAGE 46

46 approach was taken by assuming the life of the project to be 40 years Therefore, it is a ssumed that the inflation rate for fuel a nd labor will continue at 1% and 0.2% respectively, over the planning horizon. Exchange R ate The annual exchange rate between Haiti and United States is taken from the Bank of Republic of Haiti (BRH, 2011) for the p eriod 1995 to 2011. If the exchange rate is devaluated, exporters of mango from Haiti will be more competitive i n the U S market. The exporter will need fewer U.S. dollar s to pay expenses (wages, buying fresh mango and operation expenses ) when the currenc y is devaluated. In contra st the firm will earn less from fresh mangoes sold in local currency to national consumers. A normal distribution is used to simulate the exchange rate over the planning horizon. Rate of L oss of M ango The rate of loss of mango m easures the percentage of mango that is culled in the handling process before being exported. If t he rate of loss is high it will have a negative impact on the profit of the company. The office of Ministry of Agriculture of Haiti and the USDA office in Ha iti are charged with monitoring sanitary compliances at packinghouses. They estimated the average rate o f loss of the industry to be 38% in 2009. The minimum and maximum rates of loss for the industry were estimated between 28.33% and 51 % respectively, fo r the same year. The rate of loss of the hypothetical firm has an empirical distribution based on these industry historical figures. Fruit F ly O utbreak Fruit fly outbreaks in the grove for mangoes results in loss of exports to the U.S. market. In 2007, whe n fruit fly first became a problem for Haitian mangoes the industry lost 40% of the export earnings from the U.S. market and 10 % the following year (Pierre

PAGE 47

47 2011). A Bernoulli probability was incorporated in the stochastic spreadsheet with a 10% probabilit y for occurrence of a fruit fly outbreak. In the case of a fruit fly out break the company is charged 40% on the current annual revenue and 10 % in the following year. Table 4 3 Assumed probability distribution of stochastic variables Variable Distribution Yield GRKS Price GRKS Exchange rate Normal Rate of loss of mango Empirical Outbreak fruit fly Bernoulli Export volume in high season Empirical Scenario Analysis In this study five different scenarios were simulated related to the possibility of investment in the value chain of mango. Three main cases are considered: 1 ) investing only in the packinghouse; 2 ) investing simultaneously in an orchard and in a packinghouse and 3 ) investing simultaneously in two complementary seasonal orchards and a pa ckinghouse. The final two scenarios involved establishing the packinghouse at two different times (5 th and 12 th years) after implementation of the complementary orchards. Scenario 1 Investment O nly i n a P ackinghouse Scenario 1 presents the case where an investor decides only to invest in a packinghouse starting in year 2012. It is assumed that mangoes are supplied either by cooperatives or wholesalers during the peak season. T he packinghouse would export mangoes only during th e peak season from April to A ugust. The export capacity of the packinghouse is assumed to be limited at the level of the industry at average for the peak season However as a new company entry the market, it is assumed that the new

PAGE 48

48 packinghouse for the first five years operates wit h 70% of the average export capacity of existing packinghouses. Scenario 2 I nvesting S imultaneously i n a P ackinghouse and O ne C omplementary S easonal O rchard In scenario 2, investment is made simultaneously in the packinghouse and in an orchard in the pl ain of Arcahaie. The packinghouse will start exporting during the January to March period upon the maturity of the orchard. In addition, the packinghouse will continue to export during the regular the peak season from April to August by buying mangoes from wholesalers or cooperatives as in scenario 1. Scenario 3 Investing S imultaneously in a P acking h ouse and T wo C omplementary S easonal O rchards S cenario 3 is the same investment as scenario 2 with an additional simultaneous investment in an orchard in Leog ane. Therefore, upon maturity the investor will be able to export from Leogane region for October to December period. In this scenario the firm exports during a yearlong season The price of mangoes during the January to March and October to December perio d s correspond respectively to high and medium price s i n the U.S. market. Scenario 4 I nvesting in T wo C omplementary S easonal O rchards and i n a P ackinghouse Five Y ears L ater. S cenario 4 assumes investment in the two complementary seasonal orchards located in Arcahaie and Leogane. In addition, five years after the orchard establishment, an investment in the packinghouse is made to start exportation.

PAGE 49

49 Scenario 5 Investing in T wo C omplementary O rchards and in a P ackinghouse Twelve Y ears L ater. S cenario 5 ass umes investment in the two complementary seasonal orchards located in the Arcahaie and Leogane regions. In addition, twelve years after establishment of the orchards, an investment in the packinghouse will take place. Key Output Variables Net Present Valu e The net present value (NPV) is a capital budgeting method that accounts for the time value of money. It uses the discounting formulas for a uniform or non uniform series of payments to value the projected cash flows for each alternative investment at one point in a time (Barry et al 2000). In order terms, it summarizes the net returns for a multi year investment into a single variable and the net change in real net worth is incorporated in the NPV (Richardson 2006 ). In the case of a large capital outlay investment and non uniform series of returns, Richardson recommends using the following formula for NPV: NPV = I 0 + (4 6) where I 0 : is the initial cash outlay or net worth after purchasing the business NR t : is the annual net return withdrawn from the business, NW t : is nominal net worth in the last year (T) of the planning horizon, i: is the annual discount rate and t : is the time

PAGE 50

50 Probability of N egative C ash F low Risk is best assessed, for agricultural entrepreneurs, through the probabilistic ability to meet cash flows (Schoney and Moeller, 2005). Negative cash flows occur when the difference between cash inflows and cash outflow is less than zero. This is the amount of the operating loan, which must be refinanced or carried over Negative cash flow may lead to overhead costs associated with obtaining additional working capital or interest payments on extra short term borrowing ( Lowe and Whitworth, 1996). Negative cash flow may hurt profits of an export mango company if it is unab le to provide forward cash to secure mangoes from wholesalers during the period from April to September. A counter variable is used to identify when the ending cash balance is negative (counter = 1) and positive (counter = 0) in the simulation model. The model is simulated 500 times .The probability of negative cash flow is the number of times that the counter is equal to unity divided by the number of simulated results (500). Probability of Losing Real Net Worth The change in real net worth is the differe nce between beginning net worth of the firm and the present value of ending net worth. Therefore, the firm loses real net worth when the latter is less than the beginning net worth. When this occurs, the counter marked 1 for a loss in net worth and 0 other wise. Th e number of iteration for the model is 500 Then, the probability of losing real net worth is obtained from the sum of negative real net worth change divided by the number of iterations

PAGE 51

51 CHAPTER 5 DISCUSSION Scenario Results This chapter discusse s the results for each of the five scenarios (T able 5 1). The results of each scenario include the outcomes for each of the key output variables. Table 5 1 Summary of the five scenarios Scenarios Investment strategy Initial capital investment ($) 1 In vestment only in a packinghouse 1,445,497.54 2 Investing simultaneously in a packinghouse and one complementary seasonal orchard 1,832,078.50 3 Investing simultaneously in a packing house and two complementary seasonal orchard 2,100,000.00 4 Investing i n two complementary seasonal orchards and in a packinghouse in the next 5 years later 2,100,000.00 5 Investing in two complementary orchards and in a packinghouse in the next 12 years later. 2,100,000.00 Results of S cenario 1 The first scenario (invest ment only in a packinghouse) represents the general investment strategy of the nine current packinghouses in the Haitian fruit industry. Those investors entered the market by establishment of the packinghouse and purchase fresh mangoes from the different r egions during the high season. The initial investment assumed for this scenario was $ 1,445,497 million and the initial debt asset ratio 0.45

PAGE 52

52 The results suggest that this strategy is very risky. The net present value had a 39 % probability of be ing less than zero (Table 5 2) Moreover, the probability of negative cash flow increased over the entire planning horizon. The first 10 years of the simulated planning horizon had between 51% and 67.6% probability of hav ing a negative cash flow. This probability c ontinued to increase between years 11 % and 20 % from 70.6 % to 8 4 % Starting at the 21 st year of the planning horizon, the probability of negative cash flow remained close to 86 % This investment strategy had a probability of 99.8 % of los ing real net worth. Table 5 2 Summary of net present value cumulative distribution function of scenario 1 Probability Net Present Value ($) (less than or equal to) 39 % 0 50% 187 763.85 75% 584 966.00 100% 1 026 300.32 Figure 5 1 Financial risk of negative cash flow from investing only in packinghouse

PAGE 53

53 Results of S cenario 2 Scenario 2 (packinghouse and one complementary orchard) requires additional capital for establishment of an orchard in order to take advantage of the harvesting season between January to Marc h. Investment capital for this scenario was assumed to be $ 1,832,078.50 and the debt asset ratio was 0. 71. The additional capital was necessary to purchase land and farm equipment. The addition of the complementary orchard to the packinghouse improved the likelihood of success of the investment. In fact, the net present value of scenario 2 had only a 6 % probability to be less or equal to zero. The level of net present value of this investment strategy had a 50 % and 75 % chance respectively to be less or equa l to $ 1,900,000.00 and $ 2,139,398. 23. The cash flow probability deficit is quasi certain for the first ten years of the simulated planning horizon. Between the projected 11 and 24 years of the investment period the probability of neg ative cash flow star ted at 89.2% and decreased to 50 % After this period, this probability decreased slowly and stagnated around 38% Contrary to the strategy to invest only in a packinghouse, the probability to lo se real net worth is only 22.60% Table 5 3 Summary of net present value cumulative distribution function of scenario 2 Probability Net Present Value ($) (less than or equal to) 6 % 0 25% 1,44 6 ,252.00 50% 1,900,000.00 75% 2,139,398.23 100% 2,647,706.16

PAGE 54

54 Figure 5 2 Financial risk of negative cash flow from investing in a packinghouse and a complementary orchard. Results of S cenario 3 Scenario 3 (a packinghouse and two complementary orchards) adds to the second scenario another complementary orchard to take advantage of the harvest period from September to December. The establishment of the orchard increased the initial required investment to a total of $ 2,100,000.00 The diversification of production to two complementary seasonal orchards improved the net incomes compared to the two previous investment strategies. The NPV had a certain probability to be positive. The value of the discounted NPV had a probability of 50 % to be less or equal to $ 3,749,449.43 and a probability of 75% to be less or equal to $ 3,393,500.00 The financial cash flow position im proved steadily along the planning horizon. The probability of negative cash flow was close to 90 % from the first to twelfth year s ( f igure 5 3). However, during the period of years thirteen and twenty four, the probability of negative cash flow d ecreased continuously from 71.6% to 5 % Starting from the 25 th year of planning horizon, the company had less than 5% chance to have a cash flow

PAGE 55

55 deficit. The change in real net worth has a probability close to unity to be positive at the end of the 40 years plannin g horizon Table 5 4 Summary of net present value cumulative distribution function of scenario 3 Probability Net Present Value ($) (less than or equal to) 3% 2,870,941.27 25% 3,406,085.12. 50% 3,749,449.43 75% 3,393,500.00 100% 4,629,259.39 Figure 5 3 Financial risk of negative cash flow from investing in a packinghouse and two complementary orchards. Results of S cenario 4 Scenario 4 is investing simultaneously in two complementary orchards and a packinghouse after five years of operation. Contrary to the scenario 3, in vestment in the packinghouse facility is undertaken at the year corresponding to early maturity of the

PAGE 56

56 orchard. The capital investment remained identical including the debt asset ratio. The disbursement of capital for the pac kinghouse facility is done at the end of the fourth year. This strategy improved the financial position of the business in the short term between 1 and 5 years by anticipating costs of packinghouse equipment debts and incurred interest and principal liab ilities. However, the probability to have a negative cash flow in the first 10 years of operation was on average 98 % After the 14th year of the planning horizon the probability of negative cash flow was less than 38 % After the 19th year of operation, the results showed that the probability to have a negative cash flow was less than 5% The discounted NPV of scenario 4 was positive for the 500 times iterated cumulative distribution fu nction. The NPV had a 25% chance to be less or equal to $ 3,533,816.82 an d 50% to be more than $ 3,749,449.42 The change in real net worth for t his investment strategy had 100% probability of being positive. Table 5 5 Summary of net present value cumulative distribution function of scenario 4 Probability Net Present Value (less than or equal to) 3% 2, 301,193.91 25% 3, 533,816.82 50% 3,749,449.42 75% 3,965,082.04 100% 4,504,163.57

PAGE 57

57 Figure 5 4 Financial risks of negative cash flow from investing in two complementary orchards and a packinghouse in next fi ve ye ars Results of S cenario 5 Scenario 5 had the same capital investment requirement and an identical debt asset ratio as the third and fourth scenario. Investment capital in the packinghouse is disbursed in the 11th year of the planning horizon taking accou nt of full maturity of the orchard. The probability of cash flow deficit is more risky on the planning horizon compared to the third and fourth scenarios (figure 5 6) This result came from increasing liabilities for operating cash between the sixth and el eventh years of operation. The negative cash flow probability was quasi certain for the first 15 years of the planning horizon. The probability of a negative cash flow balance decreased from 73 6 % at year sixteen to 5.8% at year twenty six. During the rest of the periods, the probability of avoid ing nega tive cash flow was more than 95% The investor had a 50% and 75% probability respectively, to have a NPV less than or equal to $ 2 ,81 0 092 6 0 and $ 3,035,542.76

PAGE 58

58 Table 5 6 Summary of net present value cu mulative distribution function of scenario 5 Probability Net Present Value (less than or equal to) 3% 2,301,193.10 25% 2,663,196.48 50% 2 ,81 0 092 6 0 75% 3,035,542.76 100% 3,557,137.40. Figure 5 5 Financial risks of negative cash flow from inv esting in two complementary orchards and a packinghouse in next twelve years Best Investment Strategy The following results proved that investing simultaneously in a complementary orchard and starting handling of fresh mangoes at the fifth year as a domi nant investment strategy. The stochastic dominance framework compares the results from net present value for each of the five scenarios. The first stochastic dominance (FSD)

PAGE 59

59 assumes that the investor prefers more money than less when comparing the outcomes of the distribution of each risky scenario. Scenario 2 dominates scenario 1 by first degree stochastic dominance. This result assumes the two following conditions: 1) the investor has positive marginal utility of wealth for all levels of net present valu e and ; 2) for all net present value the cumulative distribution function of the scenario 1 is less than or equal to the cumulative distribution function of the scenario 2. This condition also includes strict inequality for some level of NPV. Graphically, t he cumulative distribution function of NPV from scenario 2 is to the right of scenario 1 and there is no intersection between the two distribution functions. In addition, scenario 3, 4 and 5 dominate scenario 2 by first stochastic dominance. However, the p reference s for scenario s 3, 4 and 5 were dependent on the risk a verse coefficient (RAC) of the decision maker. The intersection of cumulative distribution function of NPV for scenario 3, 4 and scenario 5 prevents a conclusion about the first stochastic dom inance (F igure 5 7). Table 5 7 Summary of investment strategy dominance Dominated investment strategy by: Dominating investment strategy First Degree Dominance Second Degree Dominance Scenario 1 Scenario 2 Scenario 1 Scenario 1 Scenario 3 Scen ario 1, Scenario 2 Scenario 1, Scenario 2 Scenario 4 Scenario 1, Scenario 2, Scenario 5 Scenario 1, Scenario 2, Scenario 3 Scenario 5 Scenario 5 Scenario 1, Scenario 2 Scenario 1, Scenario 2 The second stochastic dominance is a more important crite rion because it assumes the investor is risk a verse. S cenario 4 dominates scenarios 1, 2, 3 and 5 b y

PAGE 60

60 second stochastic dominance (T able 5 7). After applying the second stochastic dominance rule, the preference for scenario 3 and 5 co ntinued to depend on th e risk a verse coefficient of the decision maker (Appendix C )

PAGE 61

61 CHAPTER 6 SUMMARY AND CONCLUSI ONS Summary Haiti is the sixth largest exporter of mango to the U S market and is the only exporter where the production relies on smallholders with random trees. The production va lue has continued to decrease 2% per year while the penetration of mango es in U.S. households is increasing based on recent studies (Ward, 2011) The mango industry in Haiti has suffered a management deficit and struct ural issues. The fru it fly out break and management issues led six packinghouses to close operations. Haiti has, however, the following competitive advantage: a unique mango variety and potential to produce and export during a yearlong period. The export season for Haiti to t he U.S. market can be divided into three categories. First, the high season during the April to September period that coincides with low prices on U.S. market. T wo other potential export seasons occur in the periods January to March and September to Decemb er. The plain of Arcahaie and Leogane can provide mangoes during these two complementary harvesting periods. The months of those two export periods, except for January, also correspond to medium and high prices on the U.S. market. Five alternative investm ent strategies were considered for the investor. The first strategy assumes the entrepreneur invests only in a packinghouse and buys mangoes from wholesalers during the high season period to handle and export. The second strategy considers simultaneous inv estment in a packinghouse and a complementary orchard to export during the season from January to March The third strategy assumes investment simultaneously in the packinghouse and two complementary orchards to

PAGE 62

62 export yearlong. Finally, scenarios number 4 and 5 differ from the third scenario by timing the establishmen t of the packinghouse at the fifth and twelfth years respectively. The investment capital for the packinghouse is estimated at $ 1,445,497.54 and the establishment of two hypothetical commerc ial orchards of 50 ha required $ 79,350 each. This research determined the likelihood of success for investing in producing and handling mango from Haiti to the U.S. market and identified the most dominant strategy. The analysis took account of risk elemen ts such yield, rate of loss at packinghouse, price, and probability of fruit fly outbreak to evaluate different investment strategies. Using the financial management approach we tested the following hypotheses: 1. If an entrepreneur invests only in a packing house to export mango from Haiti to U.S. market, under the current production system, the real net worth value of the firm will be negative at end of the planning horizon. 2. Diversifying the production by complementary seasonal region will increase the likel ihood of positive net present value. 3. Diversifying the production by region and starting an export processing facility during early maturity of the orchards will result in stochastic dominant investment strategy. Conclusions The financial results of this re search showed that investment in producing and handling mango from Haiti to the U.S. market can be profitable. The probability of positive net present value is quasi certain for investment in two complementary orchards and a pa ckinghouse (scenarios 3, 4, 5 ). Financial risk s of negative cash flow are very high in the first 12 years of operations but they decrease steadily to less than 10% after 23 years of operation Exceptions t o this pattern are the first and second

PAGE 63

63 scenario s. Investing only in a packingho use has an increasing probability o f a neg ative cash flow; starting at 50% and increasing to 88 % The results confirmed the hypothesis that investing only in a packinghouse is not a sustainable strategy. The net present value for this strategy had a proba bility of 40 % to be less than or equal to zero. Besides the annual increasing risk of negative cash flow, if a n investor adopt s this strategy, it had a 99.8% chance of los ing real net worth in the long r scenario 1, a manager would h ave difficulties access ing fresh mango e s because of preferential relationships built over the years between wholesalers, farmers and already existing packinghouses. Financial analysis comparisons showed that diversification o f the mango production would increase the likelihood of success for an investor. The strategy to invest in two complementary seasonal orchards and a packinghouse dominated by f irst stochastic dominance the strategy to invest in one complementary orchard T he latter dominates by FSD the strategy to invest only in a packinghouse. The probability of positive cash flow had a better prospect for the more diversified mango production area strategy. Finally, diversification of production by region and starting an export processing facility after the fifth year maturity of the orchards dominated all the other strategies b y second stochastic dominance. The first five years will be used to exploit the mango orchards and gaining necessary information of the mango marke t. In addition, this strategy demonstrated the best cash flow outlook. This result confirmed the hypothesis suggesting that diversification of harvesting season and exportation to early maturity is a dominant strategy.

PAGE 64

64 Implications This study identified a strategy for expanding exports of francis mangoes from Haiti to the U.S. market. It highlights the importance of extending the production in complementary seasonal regions to produce and export during a yearlong period. Initiatives to establish new orchar ds imply enforcement of property rights. This study revealed that the investor would need important capital to cover liabilities including high probability of cash flow deficit in the midterm. Therefore, entrepreneurs should have flexible mechanisms to acc ess to capital. Previous studies in Haiti did not find any relationship between land tenure and successful farmers or technology adoptions (Wiens and Sobrado, 1998 ; Smucker et al, 2000). Nevertheless, Haitian government and international agencies or inves tors have conceded t o cadastral surveys as a principal component of agro industry investment (Smucker et al ., 2000). Therefore, transparent law and targeted procedures may be signals that could attract investors. The Inter American Development Bank Multil ateral Investment F und (2010) intended to support investors who use social entrepreneurship model. The social entrepreneurship model provides jobs to farmers and aligns personal interests of the community with the success of the orchards (Reimers, 2012). F inally, the most successful strategy is capital intensive. Therefore, the investor will need access to capital not only for the initial investment but also financial credit access that would support operational loan s in the first five years. The company sh ould also establish a proactive marketing plan i n the U S market to secure market share.

PAGE 65

65 Future R esearch N eeds Other scientific inquiries need to explore suitability of the soil s and relationships with yield in the different production regions More data should be collected when available to improve estimation of the stochastic variable distributions. Finally, it would be very effective to compare exportation of mangoes from Haiti to U.S. and Canadian markets taking account that the latter does not require hot water treatment.

PAGE 66

66 APPENDIX A HYPOTHETICAL PACKINGHGHOUSE BUDGET Table A 1 Estimated budget for building and machinery for the packinghouse Item Unit Price ($) Quantity Total ($) Building 800,000.00 Hot water treatment system 30" x 12 ft. long roller elevator with manual dump ledge, 6 ft incline, 6 ft flat for off grade pickout, motor and drive. 18,631.00 1.00 18,631.00 36" x 12 pin brush washer with 3 lines spray hood. Last four brushes have flicker bars to assist water removal, moto r and drive 19,349.00 1.00 19,349.00 15" wide x 10 ft long gravity roller on legs, with end stop 1,827.00 2.00 3,654.00 15" wide x 20 ft long powered roll conveyor, motor and drive 9,141.00 2.00 18,282.00 30" x 20 ft. long distribution belt, for manual size selection and manual crate fill by size, end chute to crate over run. Motor and drive 12,213.00 1.00 12,213.00 15 x 30" pack stands for plastic crates "UHMW" top surface. 580.00 14.00 8,120.00 Generator 250 KW 39,184.00 1.00 39,184.00 Pallet wrap machine 7,800.00 1.00 7,800.00 Regeneration tank for mangos amp "cool dip" 104,343.00 1.00 104,343.00 Amp single tank hot water dip treatment system for mangos 4 position treatment/ ref. drwg. #09 05 182,325.71 1.00 182,325.71 Forklift 28,000.00 2.00 56,000.00

PAGE 67

67 Table A 1 Continued Item Unit Price ($) Quantity Total ($) MANGO BOX PACKING AREA 18" Wide x 30 ft. sloped top crate set on table for box packing PVC top 3,218.00 1.00 3,218.00 15" x 35 Ft. powered roller packed box conveyor motor and drive. 13,069.00 1.00 13,069.00 15" x 10 Ft. gravity roller conveyors with end stops. 913.50 1.00 913.50 Vehicle 37,100.00 Tax on Equipment 121,295.33 Total Equipment (stainless materi als) 645,497.54 Total building and packinghouse equipment 1,445,497.54 Table A 2 Estimated variable cost to handle a francis mango box by the packinghouse ITEM Unit Amount ($) Flat box Mango box 0.70 Treated Pallet Mango box 0.18 P allet Mango box 0.50 Container treatment Mango box 0.06 Transport by boat to U.S. Mango box 1.16 Transport from farm to plant processing Mango box 0.09 Transport from plant processing to Port Mango box 0.90 Labor Mango box 0.07 Energy c ost Mango box 0.15 Total Variable cost Mango box 3.81

PAGE 68

68 APPENDIX B ESTIMATED COST BUDGE T OF HYPOTHETICAL CO M MERCIAL ORCHARD PLAI N ARC AHAIE CABARET / LEOGANE Table B 1 Estimated fixed cost of establishment of an orchard in plain Arcahaie or plain Leogane. Item Unit Unit Price ($) Quantity Total Fixed Cost Generator Each 13,000.00 1 13,000.00 Electric pomp Each 3,750.00 1 3,750.00 Crate Each 6 700 4,200.00 Hangar for material and harvesting Each 5,000.00 1 5,000.00 Well for irrigation Each 50,000.00 1 50,000.00 Total orchard fixed cost 75,950.00 Table B 2 Estimated cost of establishment of a commercial orchard in plain Leogane Item Unit Unit Price ($) Quantity Total Compost (sugar cane derivatives) Bag 12.50 6.00 75.00 Mango trees Each 2.50 100.00 250.00 Land preparation Tractor/ha 250.00 1.00 250.00 Hole for tree Hole /ha 0.25 100.00 25.00 Land Clearing Ha 125.00 3.00 375.00 Planting trees Tree 0.08 100.00 8.00 Apply compost Tree 0.05 100.00 5. 00 Pruning T ree 0.63 100.00 63.00 Drip Irrigation system Each 2,000.00 1.00 2,000.00 Small Materials Inputs Hand hoe Each 5.00 2.00 10.00 Pruning knife Each 3.75 2.00 7.50 Machete Each 1.25 2.00 2.50 Shears Each 0.50 2.00 1.00 Ax Each 18.75 2.00 37.50 Nebulizer Each 87.50 3.00 262.50 Total variable cost / ha 3,372.00

PAGE 69

69 Figure B 1 Simulated yield of mango per tree Leogane over planning horizon

PAGE 70

70 APPENDIX C DOMINANT STRATEGY Table C 1. Summary of probab ility of losing real net worth by scenario at end of 40 years Investment strategy Probability of losing real net worth (%) Scenario 1 99.80 Scenario 2 22.60 Scenario 3 0.00 Scenario 4 0.00 Scenario 5 0.00 Table C 2. Analysis of stochastic domin ance with respect to a function Efficient Set Based on SDRF at lower risk averse coefficient 0 Efficient Set Based on SDRF at upper risk averse coefficient 0.00001 Investment strategy Level of Preference Investment strategy Level of Preference S cenario 4 Most Preferred Scenario 4 Most Preferred Scenario 3 2nd Most Preferred Scenario 5 2nd Most Preferred Scenario 5 3rd Most Preferred Scenario 3 3rd Most Preferred Scenario 2 4th Most Preferred Scenario 2 4th Most Preferred Scenario 1 Least Preferred Scenario 1 Least Preferred

PAGE 71

71 Figure C 1 Summary of the cumulative distribution function of the NPV for the five different investment scenarios. Figure C 2 Summary of the distribution of negative cash flow probability for the five di fferent investment scenarios over the planning horizon.

PAGE 72

72 LIST OF REFERENCES Araujo, J.L.P., E.P. Araujo, R. C. Correira, and J.A Guimaraes. 2004. Production Cost Analysis and Commercialization of Mangos for Exporting Produced in the Sub Middle Sao Franci sco Region, Brasil. Acta Hort. International Society for Horticultural Science 645: 379 383. August, W. G., and A Purvis. 1995. Technology Adoption Decisions Under Irreversibility American Journal of Agricultural Eco nomics pp. 541 551. Bakhsh, K., I. Hassan, and M.S. Akhterh. 2006. Profitability and Cost in Growing Mango Orchards. Journal of Agriculture & Social Sciences 2(1):46 50. Barry, P.J., P.N. Ellinger, C.B. Baker, and J. A. Hopkin. 2000. Financial Management in Agriculture 6th. ed. Illinois: Interstate Publisher Inc. Bellande, Alex. 2005. Les filires fruits, tubercules, lgumes, plantes ornmentales et du Dveloppement Rural. B anque Interamricaine de Dveloppement. Brigham, E.F., and M.C. Ehrhardt. 2011. Financial Management: Theory and Practice 13 th ed. Ohio: South Western Cengage Learning. Bukowski, J., L. Korn, and D. Wartenberg.1995. Correlated Inputs in Quantitative Risk A ssessment: The Effects of Distributional Shape. Risk Analysis 15: 215 219. Buteau, J.M. 2011. Manager and co owner of JMB.SA. http://www.mango haiti.com/haitimap.htm Castaeda, N.P., F. Rodrguez, an d M. Lundy. 2011. Assessment of Haitian Mango Value Chain, a Participatory Assessment of Mango Chain Actors in the Southern of Haiti Working paper. Catholic Relief Services (CRS) Maryland. Clark, J.L. 2007. A Shrimp Aquaculture Investment by Grapefruit Producers in the Indian River Production Region of Florida: A Risk Assessment. MS thesis, University of Florida. Crawford, E.W., and R.A. Milligan. 1982. A Multi Year, Stochastic Farm Simulation Model for Northern Nigeria: An Experimental Design Approach. American Journal of Agricultural Economics 64(4): 728 737. Dixit A.K., and R.S. Pyndick.1994. Investment Under Uncertainty. New Jersey: Princeton University Press. Dixit, A.K., and R.S. Pyndick. 1995. The Options Approach to Capital Investment. Harvard Bu siness Review, May June, pp.105 115.

PAGE 73

73 Elmer, N.A., A. P. Thurow, J.L. Johnson, and R.C. Parr 2001. An Ex Ante Assessment of Investments in Texas Grapefruit Under Uncertainty. Journal of Agricultural and Applied Economics 33(3):391 401. Featherstone, A.M., P.V. Preckel, and T.G. Baker. 1990. Modeling Farm Financial Decisions in a Dynamic and Stochastic Environment. Agricultural Financial Review 50:80 99. Ferson, S., and L.R. Ginzburg. 1996. Different Methods are Needed to Propagate Ignorance and Variabilit y. Reliability Engineering & System Safety 54: 33 144. Food and Agriculture Organizat ion of the United Nations. 2012 FAOSTAT. Rome, Italy: FAO. Available online: http://faostat3.fao.org/home/i ndex.html#compare Fumasi, R. J. 2005. Estimating the Impacts of Differing Price Risk Management Strategies on the Net Income of Salinas Valley Lettuce Producers: A Stochastic Simulation Approach. MS thesis, California Polytechnic State University. Gitt inger, J.P. 1982. Economic Analysis of Agricultural Projects. 2nd ed. Maryland: The John Hopkins University Press. Griffin, W.L. and S.G. Thacker.1994. Indoor Intensive Red Drum Aquaculture. Journal of the World Aquaculture Society 25(1): 86 100. Groebner, D.F., P.W. Shannon, P.C. Fry, and K.D. Smith. 2008. Business Statistics: A Decision Making Approach 7th ed. New Jersey: Prentice Hall Inc. Hadar, J., and W.R. Russell, 1969. Rules for Obtaining Uncertain Prospects. American Economic Review 59:25 34. Hait i, Republic of, Bank Republic of Haiti (BRH). 2012. Exportation par produits. Direction of International Affairs. http://www.brh.net/exportproduits.pdf Haiti, Republic of, Haitian Institute for Statist ics and Informatics (IHSI). 2007 Inventaires Haiti, Republic of, Ministry of Economic and Finance (MEF). 2011. Srie de prix de Haiti, Republic of, Bank Republic of Haiti. 2011. v olution du taux de change http://www.brh.net/tauxdechange_chrono.pdf Haiti, Republic, Administration Gnrale des Douanes (AGD). 2010, Exportations par grande catgorie de 2002 a 2009. Unpublishe d data. Hanoch, G and H. Levy 1969 The efficiency analysis of choices involving risk. Rev iew of Economics Studies 36:335 346

PAGE 74

74 Hildebrandt, P., and T. Knoke. 2011. Investment Decisions Under Uncertainty A Methodological Review on Forest Science Studies. For est Policy and Economics 13(1):1 15. Inter American Development Bank Multilateral Investment Fund. 2010. Haiti Mango as an Opportunity for Long term Economic Growth. http://idbd ocs.iadb.org/wsdocs/getdocument.aspx?docnum=35191701 Levy, H. 1992, Stochastic dominance and expected utility: survey and analysis Management science, (38): 555 593. The Roy al Institution of Chartered Surveyors (RICS). http://www.rics.org/site/download_feed.aspx?fileID=2643&fileExtension=PDF Lundahl, M.2004. Sources of Growth in the Hai tian Economy. Economic and Sector Study Series RE2 04 004, Washington DC: Inter American Development Bank Malcolm, L.R. (2004 ). Farm Management Analysis: A Core Discipline, Simpl e Sums, Sophisticated Thinking. AFBM Journal 1:45 56. Marini, R.P.1997. Growi ng Peaches and Nectarines in Virginia, Cooperative Extension. Horticulture Publication 422 519. Muraro, R.P., F.M. Roka, and R.E. Rouse. 2005. Budgeting Costs and Returns for Southwest Florida Citrus Production, 2004 05.Food and Resource Economics Departm ent, Florida Agricultural Experiment Station. EDIS FE 631, University of Florida Panteva, N.2010. Fruit & Nut Farming in the U S. IBIS World Report 11135. Pierre Charlemagne, Ministry of Agriculture and Natural Resources of Haiti, personal communication. June 2011. Direction de la Protection Vgtale. Port au Prince, Hati. Porter, M. 2008. The Five Competitiv e Forces That Shape Strategy. Harvard Business Review Posts, T., and P. Van Vliet. 2006. Downside Risk and Asset Pricing. Journal of Banking & Fi nance 30:823 849 Purvis, A., W.G. Boggess, C.B. Moss, and J. Holt, 1995. Technology adoption under irreversibility and uncertainty. American. Journal of Agricultural Economics 77 (3): 541 551 Reimers, W. March s president.

PAGE 75

75 Schoney ,R. ., and C.Moeller. 2005. A Comparison of Risk and Return Characteristics of Efficient Crop Portfolios for the Brown Soil Zones Saskatchewan and Mecklenburg, Germany. Research in Agricultural and Applied Economics International Farm M anageme n t Association Richardson, J.W. and H.P. Mapp, Jr.1976. Use of Probabilistic Cash Flows in Analyzing Investment Under Conditions of Risk and Uncertainty. Southern Journal Agricultural Economics. Richardson, J. W. 2006. Simulation for Applied Risk Management. Department of Agricultural Economics, Agricultural and Food Policy Center, Texas A&M University. College Station, TX, January 2006 Richardson J.W., Schumann, K.D. & Feldman, P.A. (2006) Simetar Simulation for Applied Risk Management, Version 2 006 Texas A&M University, College Station, Texas Richardson, J.W., S.L. Klose, and A.W. Gray. 2000. An Applied Procedure for Estimating Multivariate Empirical (MVE) Probability Distributions in Farm Level Risk Assessment and Policy Analysis. Journal of Ag riculture and Applied Economics 32:299 315. Rothschild M and J. E.Stiglitz.1970. Increasing risk I. A definition. 1970. Journal of Economic Theory (2): 225 243. Schoemaker, P.J. H. 1982. The Expected Utility Model: Its Variants, Purposes, Evidence and Limitations Source. Journal of Economic Literature 20(2): 529 563. Smucker, G.R.T., A. White, and M. Bannister. 2000. Land Tenure and The Adoption Of Agricultural Technology In Haiti. Working Paper International Food Policy Research Institute, CAPRiX. To zer, P. R. 2009. Uncertainty and Investment in Agriculture: Is it Worth the Money? Agricultural System 100(1 3): 80 87. U nited S tates Agency for International Development (USAID). 2010. Tanzania Agriculture Productivity Program Washington D C. U nited S ta tes Agency for International Development (USAID). 2010. Mango Forum Report: Export 5 Million Cases of USDA Certified Mangoes by 2015 Haiti http://pdf.usaid.gov/pdf_docs/PNADW226.pdf U nited S tate s Agency for International Development (USAID). 2011. Frontlines, Mangoes and Tees: the Next Phase of Haitian Recovery? Focus on Agriculture and Manufacturing Offers Opportunities for a New Generation Washington D C January February.

PAGE 76

76 U.S. Department of Agriculture, Agricultural Marketing Service (USDA AMS) 2010. Fruit & Vegetable Marke t News. Washington DC. http://www.marketnews.usda.gov Von Neumann, John and Oskar Morgenstern. 1944. Theory: of Games and E conomic Behavior. Princeton: Princeton University Press. Demand for Mangos. Food and Resource Economics Department, University of Florida. Whitmore, G.A. 1970. Third degree st ochastic dominance. American Economic Review 60: 457 459 Wiener, M. 2009. Mango Value Chain Assistance Plan. Petion Ville, Haiti: CNFA. Unpublished Wiens, T., and C. Sobrado.1998. Rural Poverty in Haiti: The Challenges of Poverty Reduction. Poverty Reduct ion and Economic Management Unit and Caribbean Country Management Unit, Latin American and the Caribbean Region. Report No. 17242 HA. Washington D C World Bank Zimmermann, H.J. 2000 An Application Oriented View of Modeling Uncertainty. European Journal of Operations Research 122:190 198.

PAGE 77

77 BIOGRAPHICAL SKETCH Isnel Pierreval earned a Higher Diploma in Applied Economics from the Center of Technics for Planning and Applied Economics (CTPEA) in 2005. His thesis outlined the role of agriculture in capital tr ansfers to secondary and tertiary market economy. In 2007, he pursued a certification in Management, y outh p rograming and l eadership from the University of South Florida and he received the Septima Clark Award from Georgetown University. He then served as Youth Project Coordinator and Livelihood Specialist at the Education Development Center Inc. (EDC) for a U nited States Development Agency International Development (U SAID ) workforce development project which became in 2010 a local Non Government al Organization (NGO) where he has continued to serve as a board member. In 2008, he was appointed as Caribbean Community (CARICOM) Youth Ambassador for Haiti to focus on youth development and their participation in Caribbean regional economy. In August 20 10, he received a fellowship from University of Florida (UF) to undertake a M aster of S cience in food and resource e conomics. In summer 2012, he was awarded membership of the honor society of agriculture, gamma sigma delta in recognition of outstanding scholarship achievement by the University of Florida chapter. Isnel plans to continue research in agribusiness and to return to Haiti to contribute to its reconstruction.