RAPID COMMODITY ASSESSMENT SERIES

PREPARED FOR THE USAID-FUNDED HAITI HILLSIDE AGRICULTURAL PROGRAM UNDER CONTRACT NO. 521-C-00-00-00035-00

JULY 2001

i

TABLE OF CONTENTS

INTRODUCTION vii

CHAPTER ONE COFFEE 1 INTRODUCTION ...... 1 PRODUCTION...... 4 Producers...... 4 Input and Technology ...... 4 Chemical Fertilizers...... 5 Returns ...... 6 Cost of Production ...... 7 Net Returns to Farmer...... 9 Constraints and Opportunities...... 10 Opportunities for Improvement ...... 11 POSTHARVEST ...... 12 Transport...... 12 Processing ...... 13 Processing Costs ...... 14 Constraints and Opportunities...... 15 MARKETING...... 16 Domestic Market...... 20 Export Market ...... 21 Constraints and Opportunities...... 22 POLICY ENVIRONMENT ...... 23 Policy Areas...... 23 Constraints and Opportunities...... 23

ANNEX I: THE U.S. SPECIALTY MARKET FOR COFFEE 25 ANNEX II: PERSONS CONTACTED, COFFEE SUBSECTOR STUDY 35

CHAPTER TWO CACAO 39 SUMMARY CONCLUSIONS ...... 39 RECOMMENDATIONS...... 40 INTRODUCTION ...... 41 PRODUCTION...... 41 Producers...... 41 Returns ...... 47 Constraints and Opportunities...... 53

ii

POSTHARVEST ...... 54 Storage ...... 54 Transport...... 55 Processing ...... 55 Costs...... 56 Constraints and Opportunities...... 56 MARKETING...... 56 Domestic Market...... 60 Export Market ...... 60 Market “Windows” ...... 62 Constraints and Opportunities...... 62 POLICY ENVIRONMENT ...... 62 Constraints and Opportunities...... 63

ANNEX I: CACAO: REPORT ON CLONAL GARDENS 65 ANNEX II: CACAO: TRADING CASH-FLOW PROJECTIONS 71 ANNEX III: CACAO: LABOR REQUIREMENTS IN COMMERCIAL PROJECTION 75 ANNEX IV: CACAO: SERVICOOP SALES RECORD 79

CHAPTER THREE MANGOS 83 SUMMARY CONCLUSIONS ...... 83 Phase I: Organizational...... 83 Phase 2: Expand and Create Orchards...... 85 Phase 3: Stabilize Mango Production and Hillsides ...... 86 BACKGROUND...... 89 PRODUCTION...... 89 Producers...... 89 Location ...... 89 Farm Size ...... 91 Land Types...... 91 Productive Infrastructure ...... 92 Inputs and Technology...... 92 Returns ...... 94 Constraints and Opportunities...... 96 POSTHARVEST...... 96 Storage ...... 96 Transport...... 98 Processing ...... 99 Constraints and Opportunities...... 99

Development Alternatives, Inc. (DAI) iii

MARKETING...... 100 Domestic Market (including Dominican Republic)...... 100 Export Market ...... 101 Constraints and Opportunities...... 102 United States Market Opportunities for Fresh Mangos ...... 102

ANNEX I: MANGO PRODUCTION AREAS 107 ANNEX II: NUMBER OF BOXES OF MANGOS SHIPPED FROM HAITI 111

CHAPTER FOUR ETHNIC PRODUCE 115 SUMMARY CONCLUSIONS ...... 115 Increase Volume and Channel Efficiency of Ethnic Products...... 115 Work with Exporters to Increase Trade in Yams and Pumpkin ...... 115 Select Lead Farmer Groups ...... 116 Provide Packhouse Development Support to Association...... 116 Provide Extensions Support to Farmers During Planting Season...... 116 Organize Yam and Pumpkin Collection in Jacmel...... 116 Support Planting of Exportable Varieties of Cassava in Cap Haitien Area...... 116 Support Increased Collection and Export of Genip ...... 117 Identify Demonstration Farms for Training Purposes ...... 117 BACKGROUND...... 117 TARGET ETHNIC CROPS ...... 117 PRODUCTION...... 118 Producers...... 118 Farm Size/Types ...... 119 Inputs...... 120 Returns ...... 121 Climatic Conditions and Seasonality...... 122 Yields/Quality...... 123 POSTHARVEST CONSTRAINTS AND OPPORTUNITIES ...... 124 TRANSPORT...... 125 MARKETING...... 125 Yams ...... 125 Cassava ...... 126 Avocados...... 127 Malanga...... 129 Calabaza...... 130 Scotch Bonnet Peppers ...... 132 Genip...... 133

ANNEX I: ETHNIC PRODUCE: REGIONAL MAPS 135 ANNEX II: ETHNIC PRODUCE: RAINFALL CHARTS 139

iv

APPENDIX: SCOPE OF WORK A-1

Development Alternatives, Inc. (DAI) v

LIST OF TABLES AND FIGURES

TABLE

1 Producer Price in Gourdes, FY 2000 ...... 7 2 Smallholder Cost of Production of 1 Marmite of Fresh Cherries and Café Pilé...... 8 3 Smallholder Net Returns from Coffee Production...... 10 4 Transportation Cost—Varied Locations...... 13 5 Processing Cost for Washed Coffee, 1999-2000 ...... 15 6 Farmer’s Gross Marketing Margin—Various Channels...... 20 7 Persons Contacted—Coffee Subsector Study...... 24 8 Specialty Coffee Categories...... 27 9 Gourmet/Rare Origin/Estate Coffees Offered at the Café Connection/ Coffee Masters...... 30 10 Retail Selling Prices of Fair Trade Coffee...... 32 11 Cacao-Growing Areas...... 42 12 Rainfall Precipitation...... 43 13 Summary of Estimated Returns from Alternative Tree Crops...... 52 14 Summary of Marketing Chains...... 58 15 Summary of Margins ...... 59 16 Number of Boxes of Mangos Shipped from Haiti, November 1999- September 2000...... 113 17 Farmer Groups in the Area of Jacmel...... 119 18 Farmer Groups in the Area of Cap Haitien...... 119 19 Total Hectares in Production for Selected Ethnic Crops in the Jacmel Area ...... 121 20 The Cost of Production Activities in the Jacmel Area...... 121 21 Planting and Harvesting Seasons for Jacmel Area ...... 122 22 U.S. Avocado Supply—Seasonality Chart ...... 128

FIGURE

1 Coffee Production by Department, 1999 ...... 1 2 Coffee Production and Export ...... 2 3 Area in Coffee Production, 1954-1999...... 3 4 Coffee Sub-sector Map...... 17 5 Trends in Coffee Sector, 1950-1995...... 21 6 E.U. Imports of Arabica Green Beans: 1993-1997...... 22 7 World Green Coffee Production, 1995-1999...... 28 8 U.S. Imports of Coffee...... 28 9 U.S. Import Market Share—Arabica Coffee ...... 29 10 U.S. Import Market Share—Robusta Coffee...... 29 11 U.S. Retail Sales: Gourmet/Specialty Coffee ...... 30 12 Illustration of Market Channels ...... 57 vi

13 Minimum Margins...... 59 14 Maximum Margins...... 59 15 International Cocoa Markets 30 Years ...... 61 16 Daily Cocoa Prices—3 Years ...... 61 17 Recent Haitian Mango Production...... 95 18 U.S. Imports of Fresh Mangos...... 103 19 U.S. Import Market Share, 1999...... 103 20 U.S. Monthly Imports of Mangos, 1999...... 104 21 Los Angeles Wholesale Prices for Mangos ...... 104 22 U.S. Imports of Fresh Yams ...... 125 23 U.S. Monthly Imports of Fresh Yams, 1999...... 126 24 U.S. Imports of Fresh Cassava...... 126 25 U.S. Monthly Imports of Fresh Cassava, 1999...... 127 26 Miami Wholesale Prices for Cost Rican Cassava...... 127 27 U.S. Imports of Fresh Avocados...... 127 28 New York Wholesale Prices for Greenskin Avocados ...... 129 29 U.S. Imports of Malanga...... 129 30 Miami & New York Wholesale Prices for Malanga...... 130 31 U.S. Imports of Fresh Pumpkin ...... 130 32 U.S. Monthly Imports of Fresh Pumpkin, 1999 ...... 131 33 U.S. Import Market Share: Fresh Pumpkin, 1999 ...... 131 34 New York Wholesale Prices for Jamaican Calabaza...... 132 35 U.S. Imports of Fresh Chillies ...... 132 36 Miami Wholesale Prices for Jamaican Scotch Bonnet Peppers ...... 133 37 Region 1 Rainfall Year 2000 ...... 141 38 Region 2 Rainfall Year 2000 ...... 141

Development Alternatives, Inc. (DAI) vii

INTRODUCTION

During October-November 2000, the Haiti Agricultural Program (HAP) conducted rapid commodity assessments of several high-value commodities, as input to the development of commodity-specific development strategies to maximize sales, profits, and farmer incomes.

A team of expatriate short-term consultants, together with long-term expatriate and local staff, undertook rapid assessments using a commodity chain approach, tracking the various commodities from farm to market. The commodities covered were coffee; cacao; mangos; and “ethnic produce,” which included niche export market products such as yam, cassava, avocados, malanga, calabaza, Scotch Bonnet peppers, and genip. These assessments provided HAP staff and stakeholders with up-to-date information on the location and structure of production and marketing channels and on the inputs, services, transactions, participants, and estimated costs and returns. The assessment also identified constraints and opportunities to improve the performance of the commodity chains. Specifically, each assessment looked at available data, analyses, and studies and gained information from interviews with PLUS and other development program staff, the business community, and industry experts. Following are the segments of the commodity chain that were reviewed in conducting the assessments:

Inputs: seed, agro-chemicals, goods and services, and credit needs and availability.

Production: crop varieties, location, productivity, seasonality, costs, and quality.

Postharvest: storage, transport, processing, marketing, and sales.

Domestic Market: consumer quality requirements, location, product form, product use, growth potential, and key buyers.

Export Market: consumer quality requirements, location, product form, product use, growth potential, and key buyers;

Business Participants: characteristics of growers, processors, buyers, coyotes, employment levels, investments, costs and revenues, productivity and efficiency, and leading institutions; and

Policy Environment: direct and indirect trade, other policies, subsidies, and taxes.

This report comprises the results of the assessments, categorized by the three major export commodities—coffee, cacao, and mango. This is followed by a chapter on ethnic produce. HAP staff thoroughly reviewed these documents and incorporated appropriate recommendations into the project work plans for 2001 and beyond. Furthermore, the reports continue to be used as reference works as questions arise. HAP’s quarterly and other reports will update the information contained in these reports as new concepts and ideas are tried and as the crop sectors develop with interventions and technical guidance from HAP staff. viii

HAP staff and consultant participation during the assessment process was as follows:

Team Leader—Tim Aston, HAP Chief of Party, oversaw the field activity to ensure the product reports were completed on time and were of high quality.

Coffee Analyst—Bagie Sherchand worked with resident staff to review the production to market channels, including postharvest practices, and drafted the Coffee chapter.

Cacao Analyst—John Finn worked with staff to review the production-to-market channels and drafted the Cacao chapter.

Mango Analyst—Tom Davenport worked with HAP staff to review mango production-to- market channels, including postharvest practices, and drafted the Mango chapter.

Ethnic Foods Advisor—Fernando Correa, HAP International Marketing Advisor, assisted the Team Leader in his duties and contributed substantively to the Ethnic Produce chapter.

Export Market Analyst—Adrian Baskey, export market researcher, conducted desk-top research and provided international market information to all commodity chapters.

Domestic Marketing Advisor—Zach Lea, HAP Domestic Marketing Advisor, contributed on-the-ground information on coffee and cacao and provided contacts to the overall effort.

Domestic Market Analyst—Junior Paul, HAP Market Coordinator, provided insight into the workings of local markets and contributed to the Ethnic Produce chapter.

Assistance from all the HAP and USAID staff in-country, particularly Tom Lenaghan, Elzadia Washington, and Mark Eddy Martin, is gratefully acknowledged.

Development Alternatives, Inc. (DAI) 1

CHAPTER ONE COFFEE

INTRODUCTION

Coffee remains one of Haiti’s most important export crops, and it continues to be a vital component of the Haitian economy. Agriculture generated about one-third of the country’s gross domestic product, and coffee exports accounted for 70-80 percent of the total agricultural exports, generating about $15 million in 1998/99 (EWW, 1999). In the past, the coffee industry has generated between 28 percent and 50 percent of Haiti’s export earnings.

Coffee is grown in all nine departments of Haiti. However, 65 percent of all Haitian coffee is produced in the southern part (Sud, Sud-Est, and Grande Anse) of the country (see map). The remaining 35 percent is grown in the west (Baptiste) and northern regions (Dondon). Of all the production zones within the country, Beaumont (in Grande Anse Department), Thiotte (Sud-Est), and Baptiste (Center) are considered as the areas with the highest production potential and promise. In 1999, for example, these three departments together produced 51 percent of the total coffee production for that year (Figure 1).

Approximately 90 percent of all coffee grown in Haiti is of the Arabica Typica variety; other strains and cultivars of the Typica variety—including Caturra, Catuai, Bourbon, Salvadoreno, Dllle Albora, Padang, and Mundo Novo—also have been tried. These cultivars, however, require more care and cost (in terms of fertilizer and soil preparation) than the traditional Typica variety used in Haiti, which is well adapted and yields comparably good beans. Figure 1: Coffee Production by Department, 1999 In Haiti, coffee begins to flower during January to March, but the cherries are harvested on hillsides in

30 the lower altitudes (400-900 meters) 25 22.7 20.2 between September and December, 20 15.1 with the peek harvest occurring 15 11.1 in % 9.4 8.4 10 during mid-November and mid- 5.4 4.7 5 3.0 December. In higher elevations (900 0 Grand' Sud-Est Nord Sud Ouest Centre Artibonite Nord-Est Nord- meters and above in Thiotte), coffee Anse Ouest 2 is harvested the following January through April, with the peak harvest season falling between February and March.

Coffee production in Haiti is largely a smallholder household activity. At present, Haiti is reported to have 250,000 smallholder households (1.5 million1 people, or approximately 20 percent of the total population) growing coffee on 110,000 to 120,000 hectares—an average of about 0.5 hectare of non-contiguous plots per farm-household. Farm unit size is continuing to decline in Haiti because of land fragmentation resulting from the tradition of farms passing across generations and being divided among all children, with each operating independently thereafter. Over 90 percent of Haiti’s coffee production is at this smallholder level, contributing to 30-40 percent and sometimes as high as 50 percent (in Thiotte, for example) of the total household cash income.

In addition to coffee production, coffee processing and transport engage a large segment of Haiti’s working population and provide a source of income for thousands of women and men who sort, grade, and haul coffee at warehouses and processing centers. It is estimated that 2.4 million people are directly and indirectly employed by the coffee sector (IICA, 1992). When considering all those involved in coffee production, processing, transportation, and related direct as well as indirect activities, approximately 1 out of every 3 persons2 is estimated to be involved in the coffee sector.

The employment figures used to be larger during the 1950s and even into the 1960s, when Figure 2: Coffee Production and Export coffee was a central driving force of the MT economy, generating as much as 65 percent of 38,000 total exports and engaging 600,000 farm- 33,000 Production households in coffee production alone. Export coffee production during that same time 28,000 Exports averaged 500,000 sacks (30,000 metric tons). 23,000 During the last two decades, however, coffee 18,000 exports have steadily spiraled to a low, in 1995, 13,000 of 150,000 bags or 9,000 metric tons (Figure 2). 8,000 1950 1960 1970 1980 1990 1995 More recent estimates for exports show the number falling even further. Association des Exportateurs de Café (ASDEC) reports that coffee export for the year 1999-2000 was 3,500 metric tons. This year, 2000-2001, ASDEC estimates both production and exports to slip by another 30 percent as a result the damages caused by the Coffee-Borer infestation.

1 Average household size is estimated at 6 persons per household; total population is estimated at 8 million. 2 It is estimated that an additional 1 million people outside of coffee production are involved directly and indirectly in the sector.

Development Alternatives, Inc. (DAI) 3

The decline in coffee production has been a direct consequence not only of low and unstable world market prices (New York “C” Market prices), but also of low yields (currently 200- 250 kilograms per hectare) and diminishing hectares devoted to coffee production. Land in coffee has dropped from 171,000 hectares in the 1950s to 110,000 hectares in 1999 (Figure 3). Moreover, coffee (a cash/export crop requiring very little management per Haitian practice) now has to compete with other more Figure 3: Area in Coffee Production, 1954-1999 “stable” alternative crops that can be domestically marketed. As a result, land that had been dedicated to coffee is increasingly Figure 3: Area in Coffee Production Ha 1954-1999 being cleared off of coffee (and its 180000 171000 accompanying shade trees) for the 170000 160000 production of alternative crops with ready 150000 137000 domestic markets. In the case of Haiti, many 140000 130000 115000 of the alternative crops favored by the 120000 110000 110000 farmers are such low-value, subsistence 100000 crops as maize, beans, yams, cassava, and 1954 1984 1994 1999 banana. In the face of increasing poverty and the precarious economic environment, these subsistence crops provide farmers the option of minimizing risks (as opposed to making choices to maximize returns) to meet food-security needs while being sellable in the local markets.

But despite the depressed world market prices, low yields, and other similar disadvantages contributing to increasing disincentives to cultivate coffee, some passive coffee production is occurring in the country. This is because coffee is a deeply entrenched part or “trump card” of a small farmer’s survival strategy. Traditionally, coffee has served as a resource (a fairly liquid collateral) that extends eligibility to access credit (from the speculateurs/traders) for a large majority of rural smallholder farmers. This traditional system of financing remains strong in many parts of Haiti, where the speculateurs through traditional coffee export houses extend loans (up to 60 percent against the speculateur-determined value of their coffee) to smallholders in advance or at times of need, in exchange for the coffee cherries. Speculateurs favor this practice because it guarantees a supply of the product at a satisfactory price. Needless to say, prices agreed to by smallholders under this practice are often lower than the going market price.

As such, in the absence of commercial or agricultural banks or their unwillingness to lend to small farmers, smallholders in Haiti have learned to use coffee as an insurance crop or a low- return savings account that provides, at the very least, a safety net in times of dire need for cash. Furthermore, coffee, being an off-season crop, offers the opportunity to improve cash flow at the farm level during lean times, and for this reason, coffee trees in much of Haiti have survived from being up-rooted and turned to charcoal. Consequently, coffee continues to grow in neglect.

4

PRODUCTION

Producers

The majority of coffee producers in Haiti are smallholders (with less than 0.5 hectare), who view coffee as a cash crop dependable enough to be stored without fear of spoilage and to be sold at a time of need for money, usually during December, when school-fees and Christmas expenditures require cash. Coffee in Haiti is generally grown in the back yard, with an average of 1,500 trees per smallholder household. Even though many smallholders regard coffee as an important crop, most coffee trees are treated as hedge/extractive crop with practically no pruning, fertilization, or regular caring and management. Coffee in Haiti is a hillside crop.

Input and Technology

Virtually all coffee trees grown are self-generated and germinated at random, mostly from coffee cherries that have either fallen to the ground or been spit out by rats (and called “Café rat”). Very few households manage their coffee trees or purchase coffee seedlings from nurseries to ensure higher yields or better crop. As a result, much of the coffee grown by smallholders is old, exhausted, and shaded too heavily. It is planted so densely it competes with at least 10 other annual and perennial crops, including banana, yams, breadfruit tree, citrus, and shade trees. Leguminous trees like Sucrain are most often used as shade trees and inter-planted with food crops as part of the traditional and virtually universal system of coffee production in Haiti.

In addition to poor, and in many cases non-existent, coffee cultivation practices, Haiti’s coffee producers practice very few preventive measures to maintain or control coffee quality, largely because of the cash expense required to purchase the insecticides or fungicides. The high cost and unavailability of agricultural chemicals have reduced and, in many instances, eliminated their use. Consequently, coffee growers regularly contend with such diseases as Leaf Rust and Root Rot, which affect coffee quality. Leaf Rust, or Hemilea vastratrix, is characterized by white spots that cause partial defoliation and occurs in humid areas, while Root Rot, or Rosellinia necatrix, occurs because of nematodes in the soil. Leaf rust is said to have steadily spread throughout Haiti, causing an estimated 10 percent destruction in coffee yield each year.

In 2000, for the first time, producers are visibly experiencing the ravages of the Coffee-Borer or Broca. The Broca is an insect that bores coffee cherries to feed and lay its eggs, which damage the quality of the coffee bean. Quality is reduced as a result of broken or unsightly beans, which cannot be exported even though their aroma and taste properties are said to remain intact. It is believed that the Broca made its way into Haiti from the Dominican Republic several years ago. Some industry experts estimate that the damage caused by the Broca this year could be as high as 30 percent. Already, farmers in the northeast around the Dondon area have lost their crops to Broca, and the fear of it spreading to other areas farther

Development Alternatives, Inc. (DAI) 5

south—for example in Thiotte, which thus far is free of infestation—is a griping concern for producers and exporters alike.

In response to the Broca infestation, northeastern farmers around the Dondon area belonging to the Reseau Des Cooperatives Cafeires de la Region Nord (RECOCARNO),3 an alliance of five cooperatives, have just begun to look into organic insecticide application. Tobacco leaves as an organic insecticide, which has proved successful in Mexico, is being promoted in the region by SEFADES, a local microcredit, nongovernmental organization. Needless to say, more needs to be done on controlling the Broca infestation problem.

SEFADES was established to provide financial services to agro-enterprises in the northern region of Haiti. In addition to financial assistance, SEFADES provides production, processing, marketing, and institutional/organization development assistance to cooperatives and its members. In essence, SEFADES is a technical assistance and business center that provides its services, which are subsidized, to agro-enterprises. To date, SEFADES has worked with only nine cooperatives. SEFADES has a membership base of 3,000, thus far. RECOCARNO is its major client. SEFADES’ services (in the form of program support) are subsidized by international NGOs and donors—mainly OXFAM International and the European Union, INTERMON (Spain), and NOVID (Oxfam family). SEFADES focuses on coffee. As such, SEFADES has quickly become the marketing arm and channel of the cooperatives in the northern region. Although not a federation or union of cooperatives, SEFADES is akin to FACN institutionally in terms of the marketing and financial services it provides.4 An important feature of SEFADES is that it not only provides assistance like FACN, but it also goes beyond FACN by providing more comprehensive assistance, covering production, processing, and institutional development issues as well.

Chemical Fertilizers

Similarly, chemical fertilizers for coffee are viewed as a luxury and therefore their application at the smallholder level is virtually absent. However, the majority of farmers recognize the value of coffee parch “pay” (organic wastes) and other by-products to increase the fertility of the soil. Conversations with several agronomists in both the Jacmel and the Cap Haitien areas indicate that Haitian producers need further guidance and training on how to produce organic fertilizers and compost. In the Dondon area, SEFADES has plans to help 3,000 coffee growers produce 7,000 kilograms (or 5 m3) of compost annually.

3 Five coffee-producer cooperatives in the northern region of Haiti around Dondon have come together to form RECOCARNO. RECOCARNO is an umbrella group set up to protect the cooperatives and its members from “exploitative” marketers, to take advantage of economies of scale in marketing activities, and to serve as a united voice for its coffee-growing members. RECOCARNO is made up of two representatives from each of the five cooperatives; this group behaves like a council and convenes every month to discuss and decide on production, processing, and marketing issues. At its annual meeting, this group decides which cooperative gets how much advance money to purchase coffee cherries from its members. SEFADES works with RECOCARNO to provide assistance to its cooperative members. Farmer-members sell to their area cooperatives, which then sell their coffee through RECOCARNO, assisted by SEFADES, to the international markets. 4 FACN is a federation of 32 coffee grower associations. 6

Although smallholder families produce the bulk (90 percent) of Haiti’s coffee, less than 10 percent of the production comes from commercial coffee operations. Such commercially oriented operations are more prevalent in the center and southeast departments, including Grande Anse, where “plantations” range in size from 10 to fewer than 100 hectares. These coffee farms are said to employ better production techniques, including fertilizers, insecticides, and testing of new cultivars, resulting in higher yields (about 400 kilograms of green beans per hectare in the Grande Anse area).

Even among smallholder coffee producers, farm size in these departments is said to be relatively larger than the national average of 0.5 hectare, particularly in the Beaumont and Jacmel area, where farm size can range from 1 to 2 hectares. But given variable yields (coffee production in Haiti follows a three-year cycle, with a good year followed by an average year, which is then followed by a poor year) and low returns from coffee, farmers all over the country are reluctant to adopt new production and processing techniques, even though once established coffee trees require very little management.

Returns

Coffee in Haiti is treated as a bush-crop, grown wild and generally neglected except during harvest time. Because no effort is placed on caring for and managing the coffee tree, pre- harvest costs of producing coffee cherries are virtually nil. Smallholder coffee production costs, therefore, mainly comprise harvesting costs, which can range from 25 to 36 gourdes (gds.)5 a day for hired labor, depending on location (rural versus urban). However, coffee harvesting is generally carried out by household members, mostly women and children, without pay. When additional labor is needed for harvesting larger crops, Haitian smallholders either hire labor or rely on a social, rotating labor-exchange pool/group called eskwad. Increasingly, the tendency is to hire labor.

Harvesting of coffee can be labor intensive, depending on the yield and harvesting practice used. Farmers estimate that a person can harvest 7-10 marmites of coffee cherries a day—for example, Jacmel farmers say they harvest 10 marmites a day, while Dondon farmers harvest 7 marmites.6 An average household in the Jacmel area produces 800 marmites of coffee cherries a season, of which 600 marmites are sold to the area’s producer association. The coffee producer, who sells his or her coffee cherries to the collection/wet-processing center, receives 10.5 gds. for a marmite of fresh cherries. Producers from about a radius of 7 kilometers, or 2 hours of walking distance, bring their coffee cherries to the nearest wet- processing center, where the cherries are processed. The remaining 200 marmites of cherries are reportedly used for production of café pilé—that is, natural or dry-processed coffee.

Producer price at the collection/processing centers varies marginally in other parts of the country, as can be observed from Table 1 below. In both Thiotte and the Cap Haitien/Dondon

5 $US=20 gds. 6 1 marmite=approximately 5 lbs.

Development Alternatives, Inc. (DAI) 7

areas, farmers receive 10-12 gds. a marmite for fresh coffee cherries bound for wet processing.

Coffee producers who belong and sell their cherries to an association or cooperative also receive an additional ristourne (dividend) at the end of the season depending on the financial performance of the association or cooperative. The ristourne, which serves to induce farmers to sell to the association or cooperative, is provided to members on a per-marmite-sold basis.

In addition to cherries, Haiti’s coffee producers market coffee in other forms: café coque and café pilé. Café coque is dried coffee cherries, whereas café pilé is coffee beans using the dry- processing method. A significant proportion of coffee produced is dried at home to be sold as café pilé; a small percentage ends up in the production of café coque, but trading in café coque is marginal. A small amount of coffee is also processed at home for personal use. About 10 percent of coffee is reportedly sold as cherries for the production of café lavé.

Farm-gate prices are lower for café pilé and café coque (see Table 1), which are mostly sold to voltigeurs or speculateurs who collect the coffee at the homestead. It is estimated that a season’s harvest of fresh cherries (estimated by FACN at 600 marmites) can take anywhere from 2 weeks to 30 days to dry into café coque and another 2-4 days to decorticate and pound the dried cherries into café pilé.

Table 1: Producer Price in Gourdes, FY 2000

Cap Rouge/ Dondon/ Thiotte/ Speculateur/ SERVICOOP ASSO COOP COOP Exporter Coffee Cherries (1marmite) 10 11 10 12 - Café Pilé (1 lb. of green bean) - - 11 - 8.0 Café Coque - - - - 1.5 (1 lb. of dried cherries) Ristourne/Dividend (1marmite of fresh cherries sold to 10~15 5~7.5 3~6 3.2~3.5 - cooperatives/associations)

Cost of Production

The cost of producing coffee is virtually zero because no inputs, equipment, or pre-harvest labor goes into coffee production. Labor is required for harvesting, but usually unpaid household labor is used. There is clearly a market for harvesting labor, however, should the smallholder farmers choose to enter this market. In fact, the next-best alternative that is most likely to occur is for these farmers to work as wage laborers either for larger producers or in urban areas. In this labor market, hired-labor wage rates range from 25 to 36 gds. per day.7 The other option for smallholders is to divert their labor into the cultivation of other crops, but because coffee-harvesting is an off-season activity, household labor does not compete with other activities. The market wage is therefore a good estimate of the opportunity cost of

7 Wages for rural labor tend to be lower than the urban wage rate; however, rural wages for hired labor usually include a full meal plus the wage. 8 labor and also serves as the economic value of labor for many of those engaged in picking coffee cherries.

Other items included in coffee (green beans) production are the cost of transporting cherries from the farm to the wet-processing center and the cost of processing, dehulling, and shipping. Because smallholders do not carry out their own wet processing, the only other cost to the smallholder is the cost of transport to the wet-processing center. In the Jacmel area, small-scale farmers usually haul fresh cherries on a donkey (hired at 36 gds. a day to carry 90 kilograms a day) to the collection/wet-processing center.

For those producers selling fresh cherries, the cost of production of a marmite of cherries, when the opportunity cost of labor is ascribed to harvesting, can range from 4.5 to 7 gds., depending on harvesting/hauling efficiency (Table 2). In reality, however, given that household labor is unpaid, the actual, cash cost-of-production equals that of transport cost at approximately 0.91 gds. or $0.05 per marmite.

Table 2: Smallholder Cost of Production* of 1 Marmite (5 lbs.) of Fresh Cherries and Café Pilé

Tasks Fresh Cherries Café Pilé gds. $ gds. $ Planting 0 0 0 0 Material Inputs 0 0 0 0 Weeding 0 0 0 0 Pruning 0 0 0 0 Harvesting (imputed costs) 3.6 0.18 3.6 0.18 Drying (imputed costs) 0 0 1.5 0.08 Pounding (imputed costs) 0 0 0.24 0.01 Total Cost of Production 3.6 0.18 5.34 0.27 Hauling on Donkey to Processing 0.91 0.05 0 0 Center Total Producer Cost 4.51 $0.23 5.34 $0.27 Yield in Green Beans 0.91 lb. café lavé 1 lb. café pilé *US$1=20 gds. When opportunity cost of labor is ascribed to labor requirements.

Table 2 also presents cost-of-production figures for café pilé on the assumption that the required labor is actually paid. Because much of the production of café pilé uses unpaid family labor, the cost-of-production coffee using the dry-processing method is actually zero. This implies that whatever a farmer receives at the farm-gate is the returns to his or her labor and is perceived to be pure profit.

However, with the opportunity cost of labor, the cost of harvesting, hauling, drying, and pounding is estimated at 5.34 gds. or $0.27 per marmite, on the assumption that all 600 marmites of cherries are harvested and turned into café pilé. Table 2 assumes that it takes 25 days to dry 600 marmites of cherries and 4 days to pound. The estimation provided by various farmers and FACN experts was 2 weeks to 35 days to dry and 2-4 days to pound into café pilé.

Development Alternatives, Inc. (DAI) 9

Other factors contributing to the cost of coffee production are the distance and difficulty of hauling coffee cherries to the collection center and the harvesting practice used by the farmers. Few economies of scale can be achieved using the currently practiced method of hauling (usually by humans or a donkey). In fact, the method imposes limitations on any substantial expansion of production per individual farmer or groups of farmers, even when hired laborers are used.

Table 2 above presents a rough estimation of the cost of production of 1 marmite of cherries. The assumption is that 10 marmites of cherries are being harvested in one day. The labor requirement for harvesting depends on the harvesting practice applied. Haiti’s coffee- harvesting practice, unlike in most coffee-producing countries where coffee cherries are hand picked, is to strip the berries off the stem. Such harvesting practice is one key reason for the country’s low yield and poor bean quality. A simple change in harvesting practice alone— from stripping to picking—is expected to increase yield by 30-40 percent according to industry experts. Improving yield and quality is key to lowering the cost of production and increasing profitability.

FACN estimates that 1 marmite of cherries can yield from 0.75 to 0.95 of a pound of washed green beans. The green-bean yield supposedly depends on the elevation of production, with higher altitudes yielding more. For this exercise, we have applied a 0.91 pound conversion basis on the assumption that it takes about 5.5 pounds of fresh cherries to yield 1 pound of washed green beans (according to RECOCARNO). For café pilé, the conversion is estimated on a 1-to-1 basis (based on conversations with farmers and on a 1994 USAID study conducted by IICA) because all cherries, regardless of poor quality or insufficient ripeness, get dried and pounded to produce café pilé.

The estimated average yield for Haiti is 200-250 kilograms (green beans) per hectare. In Jacmel, the average 1992 estimate was 257 kilograms per hectare, and in Beaumont the same year, the yield was 270 kilograms per hectare (IICA, 1992). In 1999, the Dondon coffee yield was 300 kilograms per hectare of green beans. The yield in Haiti compared with other coffee-growing countries is extremely low.

The low yields also are attributed to exhausted coffee-stands that are sometimes more than 20 years old. The younger trees have regenerated more as a result of random acts of nature than because of deliberate planning. Low yields also result from other factors such as very limited or non-existent management, excessive shading, and extreme plant density because many farmers believe the more trees the better no matter how close together. This same attitude extends to the lack of understanding of the higher production capability of properly pruned trees with the optimal amount of shade (usually 40-50 percent shade).

Net Returns to Farmer

A large proportion (90 percent) of the coffee produced in Haiti follows the dry-processing method to produce natural coffee or café pilé. Table 3 shows that, for all the labor invested in 10

harvesting 600 marmites of cherries to dry and pound into 600 pounds of café pilé for sale, coffee producers, who market natural coffee through the traditional marketing structure, receive about $79.80 at the end of the season if they produce coffee on 1 hectare plots. If he or she is an average coffee producer, the land in coffee is estimated to be only 0.5 hectare and the returns in a season would be only $39.90. Such a low return is a disincentive to coffee production. At returns of this nature, it is no surprise that the majority of coffee producers are allowing their coffee production to dwindle, while seeking out higher-return alternative crops. And the fact that farmers are turning to other crops and have more than 10 products in a mixed plot indicates the low value of dry-processed coffee. This is also why farmers are keen on participating in the cooperative-driven wet-processed methods where, for picking fresh cherries, their returns are comparatively higher.

The 15 percent of coffee producers who sell their cherries to be wet processed do much better financially than those who sell them to be dry processed. Table 3 indicates that farmers selling cherries receive about $90 per season. The assumption is that of the 800 marmites of cherries farmers produce, they sell 600 while keeping 200 behind for processing into café pilé for personal use. Those farmers who are also members of the associations and cooperatives that wet process the coffee do even better at the end of the season, when each member receives a dividend/ristourne for the cherries they sold to the associations and cooperatives.

Table 3: Smallholder Net Returns from Coffee Production*

Fresh Cherries Café Pilé Farmers’ Cost of Production 2,705 gds. $135.27 3,204 gds. $160.20 600 marmites of fresh 600 lbs.* of pilé green Yields per Ha cherries beans Farmer Price (lbs.) 10.5 gds. $0.53 8 gds. $0.40 Value of Product 6,300 gds. $315.00 4,800 gds. $240 Farmer’s Net Returns per Ha. 3,595 gds. $179.73 1,596 gds. $79.80 Farmer Net. Returns @1/2 Ha. each 1,797 gds. $89.86 798 gds. $39.90 Net Returns from Coffee as a Percent of Total Household Income 36% 16% of $250/year US$1=20 gds. *Given the assumption that 1 marmite yields 1 pound of dry-processed green beans, a 600-marmite yield per hectare gives 600 pounds of café pilé.

Constraints and Opportunities

Haiti’s coffee sector is severely burdened by serious production constraints. Despite several decades of production assistance, these constraints continue to inhibit the sector from achieving its full potential. Topping the list of production constraints is the low priority placed on crop management despite the importance of coffee in the lives of the producers. Inadequate coffee culture practices, combined with poor quality stock and heavy shading of

Development Alternatives, Inc. (DAI) 11

the coffee trees, have also contributed directly to low productivity and have increased vulnerability to infestation.

Lack of soil fertility is another constraint on productivity and thereby profitability. However, restoring soil fertility requires the purchase of often expensive and sometimes hard to get chemical fertilizers. Organic fertilizer is an option most farmers recognize, but Haitian producers are new to this and therefore require technical guidance and training on how to compost and produce quality organic fertilizer. This area signals a need for further assistance.

The overcrowded and neglected coffee gardens testify to the lack of knowledge about proper cultivation methods. Coffee producers appear not to be receiving adequate technical assistance and support related to either production or postharvest issues. This is also evident from the harvesting practice employed by the farmers. The most prevalent method of harvesting coffee among smallholder producers is to strip the berries off the stem rather than pick each individual ripe cherry, as is done in many coffee-growing countries. A simple change in this harvesting practice from stripping to picking only ripe berries alone could increase yield by 30-40 percent.

Although not a constraint, Broca infestation is the biggest, most urgent production problem requiring immediate attention. Farmers in the Dondon area report that close to 50 percent of their crop has been destroyed by the Broca this year. The Broca was detected several years ago, but the smallholder coffee producers only now seem to be experiencing severe crop damage and loss.

Opportunities for Improvement

Coffee is a traditional crop, the trees are established and bearing fruit, smallholders are familiar with the crop, and it is a known commodity with an organized market. These factors give the smallholders a firm base on which improve overall profitability.

Improvement in areas that address the constraints listed above has the potential to more than double their coffee-derived income because it would greatly increase productivity and profitability.

Technical assistance is seriously needed, however. These areas should all be seriously considered as opportunities to help improve the coffee sector:

1. Improving agricultural/coffee cultivation practice; 2. Understanding what adequate shade requirement is and how to achieve it; 3. Learning the science of composting to prepare organic fertilizer; and 4. Learning a better harvesting method.

Furthermore, the Hillside Agricultural Program (HAP) should also consider collaborating with those organizations already working with smallholders—for example, producer groups, associations, and cooperatives. The idea is to tap those structures already set in place to bring 12

smallholders and technical experts together to help resolve production constraints facing the sector.

The serious problem of the Broca is yet another area where the project could immediately have an impact on crop quality and income of the producers, particularly in the northeast. HAP could play a central role in organizing or coordinating a workshop to bring local and international experts together to prepare an action plan on how best to control the Broca.

POSTHARVEST

In the postharvest handling of coffee, coffee cherries are either sold off to the wet-processing facilities or are sun dried promptly. Many smallholders dry coffee cherries on their farms. Once dried, some producers sell coffee immediately to make ends meet, although many farmers store it in the house to cash in later. Those with better economic standing are able to hold off sale because they do not need the cash right away, and they have the needed storage facilities. Storage is usually in sacks that are placed on a platform under the eaves of their houses or under a bed.

Transport

The number of participants involved in the transport of coffee depends on the processing method chosen by the producer for the coffee cherries. For example, if the producer chooses to sell ripe cherries for wet processing, the producer transports the cherries from the farm to the local processing center. The mode of transport is most likely to be a donkey. In Cap Rouge, for example, the center buys cherries from farmers within a radius of 7 kilometers, or 2 hours walking distance. Once the cherries are depulped and dried into café parche, the coffee is trucked to Tom Gateau by the Cap Rouge Producer Association, which operates the central processing center.

Café coque and café pilé are either sold at the farm-gate or in the local markets. When sold at the farm-gate, the voltigeur or speculateur collects the coffee at the farm-gate and hauls it either to a larger speculateur or directly to the exporters. Coffee sold at the local markets is transported on a draught animal or by humans. Once at the markets, the coffee is collected by the voltigeurs and trucked to the Madam Saras or speculateurs or their sub-agents.

Transport plays a major role in the structure of the coffee industry. Farmers and traders consistently cited as their highest priority improved road infrastructure to ensure that the roads from the local processing centers to the central processing center in Tom Gateau, for example, are reliable and accessible by trucks and other transport. The road from Cap Rouge to Jacmel is in serious disrepair and needs upgrading at the very least. Poor roads exemplify the risks involved in transporting such agricultural products as coffee. Many people feel that poor roads have discouraged competition among transporters, thus leading to higher transportation costs. The bad road conditions lead transporters to experience higher vehicle

Development Alternatives, Inc. (DAI) 13

repair costs and a much slower turnaround time, which end up increasing the marketing costs.

Green coffee is normally sold internationally at a FOB Haitian port price. Freight cost to Miami from Port-au-Prince is estimated at $0.10 per kilogram of green beans. Examples of local transportation costs provided to the study team are shown in Table 4.

Table 4: Transportation Cost—Varied Locations

gds./60 Kg. Location Transport Bag Cap Rouge to Jacmel Truck 15 Cap Rouge to Tom Gateau Truck 40 Tom Gateau to Port-au-Prince Container Truck 20 Thiotte to Tom Gateau Truck 55 Thiotte to Bell Anse Truck 45 Farm to Processing Center Donkey 36/day

Processing

Coffee-processing methods practiced in Haiti generally follow two tracks: (1) the traditional and widely used on-farm dry-processing method, which produces café naturel (natural coffee), also popularly known as café pilé; and (2) the wet-processing method, which produces washed coffee, or café lavé.

About 90 percent of coffee produced is processed on the farm using the dry method. The dry method entails sun drying ripe coffee cherries on cement patios called glacis. Those who cannot afford glacis dry the cherries on the ground. Because coffee harvest season coincides with the rainy season, the coffee dried on the ground takes on the odor of the wet soil and becomes vulnerable to mold. This method of drying a season’s harvest (estimated at 600 marmites per household) can take from 14 to 35 days. Drying beans is laborious; it requires spreading the beans in the sun, raking the beans six to eight times a day, and collecting the beans each evening for storage. This process is repeated again the next day until the cherries are sufficiently dried.

Once dried, many tend to sell it immediately as café coque. Many also store these dried coffee beans in the house. Others continue to process further the café coque by pounding it in a mortar to remove the bean’s glutinous outer shell to make café pilé. On average, it takes two to four days to pound café coque (produced from 600 marmites of fresh cherries) and winnow it to separate the chaff from the coffee bean to produce café pilé, also sometimes called café naturel. Café pilé is traditionally collected by voltigeurs and/or speculateurs either at the homestead or in the local market to begin the journey to domestic and international markets through wholesalers or exporters. Customarily, a small amount of coffee is retained for personal and social use and for sale in the local markets.

14

Approximately 10 percent of Haiti’s coffee is sold as cherries for the production of café lavé. Preparation of washed coffee is arduous because it requires organization and time keeping. For example, farmers have to deliver their ripe coffee cherries within 24 hours of harvest to the washing facility or else the coffee quality begins to deteriorate. Once in the wet- processing facility, cherries are washed in tanks to separate the false berries. The good cherries are then depulped using a hand- or motor-driven pulper.8 The depulped cherries are then placed in tanks to ferment for 20 to 24 hours. The fermentation process helps remove the mucilage coating from the bean. The complete removal of mucilage requires several washings. After the mucilage is finally off of the bean, the resulting parchment coffee bean is set out to dry on concrete platforms. On the average, it takes 10 days to dry depulped, parchment coffee.

Generally, at this stage, the parchment coffee is sold as café parche. As a result of the protective thin shell or parchment covering the coffee bean, café parche is the most conducive stage for shipping and storage until ready for use or export. The parchment helps protect quality and appearance of the coffee bean. Café parche is then hulled, graded, and sorted to finally transform it into café lavé. A crucial factor in the marketing of coffee beans is the grading to a set standard established by the buyers. Poor grading and sorting adversely affect the quality and therefore the price of the coffee. This aspect of coffee production and processing has historically been neglected, thus significantly eroding the quality of Haitian coffee and its image internationally.

Processing Costs

Coffee beans using the wet-processing method require significantly more resources, in terms of equipment, infrastructure, and water supply. Hence, wet-processing facilities are often off- farm entities owned and operated either by exporters who can afford it or by cooperatives and associations. For example, the cost of establishing a small-scale wet-processing facility can be as high as 600,000 gds., as was the case with the one established in Cap Rouge/ Jacmel. The study team visited this coffee producer association in Cap Rouge/Jacmel. The team also visited the central hulling/grading center at Tom Gateau belonging to FACN, which further processes the café parche that is delivered by the area’s coffee producer associations.

FACN is a federation of 32 coffee grower associations representing 20,000 coffee-producing households. There are 15 associations in the Jacmel area alone, each with a membership base of about 600 families divided among groups of 10-15 families each. Other associations are located in Thiotte, which has 1 to date; Beaumont has 13; and Nord Vallé, Saint Suzanne, and Baptiste each has 1. There are 3 planned for the Marmalade area in the north; this will bring the total number of associations under FACN to 35.

These associations each own and operate a wet-processing facility, which also serves as a central collection point for coffee producers in a 7-10 kilometer radius. Every harvest season, the associations buy coffee cherries at the facility from coffee producers who are members of

8 Hand-driven pulpers are more common in Haiti than motor-driven ones.

Development Alternatives, Inc. (DAI) 15

the associations. Membership is not a prerequisite for the association to purchase coffee cherries. However, membership is compulsory for producers to be eligible for a dividend/ ristroune at the end of the season. The amount received by a producer, as ristourne, depends on the number of marmites of cherries sold to the association.

Once the producers are paid for the fresh cherries, the association takes ownership of the cherries and immediately begins the processing activities. The processing of cherries to café parche can take 5-10 days. In the case of Cap Rouge, the final processing cost, from depulping cherries at the wet-processing facility to the production of green beans at the Tom Gateau plant, is estimated to be $0.26 per pound. The wet-processing cost for the same pound of café lavé/green beans in the Dondon area, as managed by RECOCARNO, an alliance of cooperatives, was estimated at $0.24 per pound. Table 5 illustrates the processing cost structure for the two processing facilities, one owned by an association in the south and the other by the alliance of cooperatives in the north.

Table 5: Processing Cost for Washed Coffee, 1999-2000

Association Cooperatives

(FACN) (RECOCARNO) $/lb. Café Lavé Depulping/Washing Cost 0.05 0.07 Decorticating/Hulling And Grading 0.08 0.05 Transportation Cost 0.02 0.03 Processing Facility Administration Cost 0.02 0.01 Interest Expense 0.04 0.03 Association/Coop. Service Fee 0.04 0.05 Total Processing Cost 0.26 0.24

Processing costs of producing café pilé are mainly for labor (required for drying, raking, and pounding the beans) because virtually nothing else is involved in processing café pilé. The numbers of $0.03-$0.06 per pound are an estimate, and the study team was unable to verify these figures during the field visits, given the shortness of its time in the field.

Constraints and Opportunities

A large percentage of coffee producers use the home-based, dry-method to process their cherries. Although this method requires minimum resources, the limited financial returns produce an attitude of disdain and thus reinforce the pattern of coffee crop neglect. A constraint is the inadequate capacity of processing centers, which cannot service an increasing number of members who want to switch processing methods to increase volume and enhance profitability. This situation could be improved by either increasing the capacity of currently operational processing centers or by adding new ones so more coffee producers can choose the option of wet processing to enhance their incomes.

16

Another constraint is the inability to maintain consistency in bean quality. As a result, Haiti’s market share has been slowly eroding. For example, the highly successful Haitian Bleu is reportedly suffering the same fate of declining quality. The lack of a set standard for quality combined with the inability of the association or cooperative processing the beans to enforce a quality standard contributes to low quality. Furthermore, associations and cooperatives involved in processing lack the institutional capacity to provide technical expertise on either production- or processing- related issues. Improvements to enhance this capacity could bring processor and producer into a closer partnership that benefits both.

The issue of poor roads in most coffee production areas also is an important constraint. Uphill ascents and blind curves combined with steep, narrow roads and extremely poor road conditions contribute to risks and to the high costs of transport. In hilly places where coffee is mostly grown, the transportation cost per unit of product could remain a sizable component of the marketing costs. For a commodity like coffee, the high cost of transport can cut profitability down significantly. Improvements in encouraging alternative modes of transport to eliminate risks and thereby costs faced by farmers will be as important as increasing productivity to lower the cost of production.

MARKETING

The traditional marketing system for coffee in Haiti contains three essential links: producers, intermediaries (voltigeurs/speculateurs), and exporters. This traditional structure operates alongside the “newer,” alternative marketing channels that have opened up for the coffee producers in the marketplace since the 1970s. These new marketing channels have been created by coffee-marketing cooperatives and associations, whose development was stimulated by development agencies (USAID, United Nations, European Union, and the like) to increase the competitiveness of the coffee market. These organizations have successfully demonstrated their capability to help reduce marketing margins (by reducing the number of intermediaries) and, more important, reduce risks that a lack of competition can engender.

Figure 4 depicts the channels available for marketing coffee. About 35 percent of all coffee produced in Haiti is for the export market, and 65 percent is consumed domestically, with some leaking across the border to the Dominican Republic. Of the 35 percent of export coffee, 98 percent passes through the traditional marketing structure (voltigeur, speculateur, and exporter) to be exported through the large export houses, which have now banded together to form the Association des Exportateurs de Café (ASDEC). ASDEC and its members also service the growing domestic market with their own particular blends and roasts.

Development Alternatives, Inc. (DAI) 17

Figure 4: Coffee Sub-sector Map

Smallholder Coffee Producers Medium-Sized Coffee Producers

Producer Cooperatives Voltigeurs/Intermediaries Marchandes Associations (on average passes hands 3 times at this level before going to the next)

RECOCARNO/ Speculateurs Madam Saras/ FACN KOPCAB SERVICOOP Wholesalers

Fair Trade/ Haitian Bleu Max Havelaar Traditional Distributor/ IMPORTERS/ Importers Exporters (eg., PRIMEX, Cross-Border Domestic Importer DISTRIBUTOR Wiener, Novella) Trade to Roasters S Dominican Republic U.S. Fair Trade E.U. E.U. Exporter Roasters/ Roasters/ Roasters/ Roasters/ Roasting Wholesalers Wholesalers Wholesalers Wholesalers Operation Dominican Republic Roasters/ Wholesalers Retailers Retailers Retailers Retailers Retailers

U.S. Specialty E.U. Fair Coffee Trade E.U. E.U. Domestic Dominican Domestic Consumers Consumers Consumers Consumers Consumers Republic Consumers Consumers

Washed coffee represents 10 percent of total coffee Natural coffee represents 90 percent of total coffee production and less than 2 percent of the total export production and 98 percent of the total export coffee market. coffee market.

The remaining 2 percent is exported through alternative marketing channels comprising coffee-producer associations and cooperatives, with each following its own marketing strategy. Among these new players, FACN is pursuing the high-priced specialty coffee markets in the United States. Thus far, FACN, with the assistance of Gary Tallboy, has successfully launched and marketed its coffee under the Haitian Bleu brand name. FACN estimates that less than 0.5 percent of all exported Haitian coffee is exported as Hatian Bleu. For example, in 1999 FACN exported only 237 bags (60 kilograms per bag) of Haitian Bleu coffee out of 87,000 bags. 18

RECOCARNO, KOPCAB, and APCAB are each an alliance or coalition of coffee-producer cooperatives formed to market their coffee beans. RECOCARNO, for example, is based in the north, near Dondon, and represents 3,000 producer members. KOPCAB is a coalition of coffee-producer cooperatives based in the Thiotte area and represents 1,200 members. These coalitions of coffee-producer cooperatives provide a marketing channel different from the one provided by FACN.

In the last several years, these cooperatives have successfully pursued the Fair Trade markets popular in Europe and budding in the United States and Japan. Fair trade markets offer access to products sold by suppliers, importers, and manufacturers and guarantee a “fair” price to the producers and workers in developing countries providing the products. Fair trade is essentially an innovative market penetration strategy that has helped achieve a higher price for small producers in developing countries. Fair trade coffee has been most successful in Europe, capturing between 2 percent (Germany) and 5 percent (Switzerland) of most markets and is moving from niche to mainstream outlets. For example, Fair Trade coffee is available in 90 percent of supermarkets in the United Kingdom (World Resources Institute, 1999). Products with a Fair Trade label ensure the buyer that the products were grown responsibly (implying the application of environmental friendly approaches) and that the producers who grew the products received a fair price.

Twenty NGOs worldwide (with most of them in Europe) certify Fair Trade products and provide the Fair Trade seal of approval. Max Havelaar Foundation is one of the largest Fair Trade certifying and monitoring organizations in Europe, and the commodities it promotes carry the “Max Havelaar” trademark or label. Max Havelaar coffee, for example, commands 3 percent of the market share in the Netherlands. The Max Havelaar trademark is reportedly recognized in 14 European countries. Transfair USA is similar to Max Havelaar but caters the U.S. market.

The Max Havelaar Foundation, a non-profit organization, works with developing country producers, developed country companies and importers, and consumers. The foundation does not sell commodities itself, but certifies, monitors, and maintains a database of registered producers-cooperatives and trade-license holders (companies and importers certified as honoring the Fair Trade practice) that carry its trademark. The foundation, like other Fair Trade organizations, also markets and promotes the Fair Trade brand to educate consumers and expand market share.

In Haiti, the coffee-producer cooperatives such as RECOCARNO and KOPCAB have begun to market some of their coffee to those importers and distributors licensed to sell Max Havelaar trademarked commodities. These cooperatives are being assisted by such international NGOs9 as Oxfam International/UK, Lutheran World Federation (FLM), and

9 International NGOs provide technical assistance primarily in marketing but also in production to the cooperatives (or alliance of cooperatives such as RECOCARNO, KOPCAB, and APCAB) through local NGOs like SEFADES. For instance, Oxfam International provides technical as well as programmatic assistance to RECOCARNO through SEFADES, the Lutheran World Federation, and Action-Aid work with KOPCAB in the Thiotte area. APCAB receives assistance from the ICF as well as Oxfam International. The European Union is also providing technical and marketing assistance through its STABEX program.

Development Alternatives, Inc. (DAI) 19

Action-Aid not only to comply with a set of criteria to become registered with the foundation but also to market and ship coffee. Once registered, the cooperatives have a better chance of being contacted by Max Havelaar-licensed importers and distributors interested in purchasing Fair Trade labeled coffee beans, which carry a premium. Fair trade coffee (primarily washed coffee because of its good taste and high quality) is fetching as much as $1.26 a pound FOB/Haiti Port, while specialty coffee under the Haitian Bleu label is receiving as much as $2 per pound FOB/Port-au-Prince.10 However, some cooperatives in Haiti have negotiated to sell their season’s green beans at lower Fair Trade prices; KOPCAB negotiated a $1.24 per pound for its coffee, and APCAB reportedly received $1.19 per pound. Once negotiated, the price remains fixed for the contracted amount. Nevertheless, with prices such as these, there is great interest among producers to increase their sale of coffee through these channels.

Those coffee beans not sold through the Fair Trade channel follow the traditional structure, ending up with exporters. For example, RECOCARNO sold some coffee in 1999 to Fair Trade importers and others to Novella, a traditional exporter based in Cap Haitien.

SERVICOOP provides a separate marketing channel for the producer-members of the older cooperatives around the country. Of the numerous older cooperatives created all over Haiti in the last decade or so, only about 15 remain operational. In the past, SERVICOOP marketed its members’ washed and dry-processed coffee beans either directly overseas or sold it to the traditional exporters (for example, in 1999 SERVICOOP sold 3,000 bags of washed coffee to PRIMEX). In 1999, for the first time SERVICOOP reportedly exported 1,600 bags of washed green beans under the Max Havelaar banner through a Max Havelaar-certified and -licensed European importer and distributor. Because it is not clear that SERVICOOP will pursue this channel as a normal practice, the link between the Fair Trade channel and SERVICOOP is not explicitly marked in Figure 4. In addition, SERVICOOP is planning to focus only on washed coffee

Natural coffee, or café pilé, dominates Haiti’s coffee bean market with 90 percent of all coffee beans produced following the dry-processing method. Café lavé, or washed coffee, makes up the remaining 10 percent, with a large proportion sent to international markets. For example, all the green beans exported through the FACN, RECOCARNO, KOPCAB, APCAB, and SERVICOOP (for 2000 and forward) channels are washed coffee. A significant amount of washed coffee produced and processed around the Thiotte area is also said to cross annually into the Dominican Republic. Last year, an estimated 68,000 bags of coffee (mostly washed but some natural as well) supposedly entered the Dominican Republic marketing channel from Haiti.

Although 15 percent of the farmers sell their cherries through their cooperatives and associations, most farmers (approximately 68 percent) sell their cherries and café pilé to voitigeurs and/or speculateurs (APROMA, 1996).

10 FACN receives a $2 pound FOB/ Port-au-Price for its coffee. This price is fixed, based on a contracted period (three years, at present) with the FACN roaster. 20

Table 6 shows the farmer’s average gross marketing margin by marketing channels. The gross marketing margins show the percent share of the FOB price that is captured by the farmers. The table shows that farmers receive higher returns for their coffee from the “newer” channels as opposed to from the traditional structure. Under the traditional marketing circuit, coffee changes hands three to five times among the voltiguers and speculateurs before reaching the exporters. The voltigeurs and speculateurs are said to each extract marketing margins of 12.5-20 percent. Consequently, the longer the marketing chain, the lower the returns to the producers, as is evident from the low (13 percent) farmer gross marketing margin shown in Table 6.

Table 6: Farmer’s Gross Marketing Margin—Various Channels

Farmer’s FOB/PAP Farmer’s Gross Price/Lb. Farmer’s Gross Marketing Received by Ristourne Price* Marketing Margin Assoc./ Margin with Coops. Ristourne FACN/Haitian Bleu $2.00 $0.53 26.5% $0.71 65% RECOCARNO/Fair Trade $1.26 $0.53 42% $0.29 65% KOPCAB/Fair Trade $1.24 $0.53 43% $0.30 67% APCAB/Fair Trade $1.19 APCAB/Marché Mondial $0.77 Exporter/ASDEC $0.78 $0.10 13% US$1 = 20 gds. * Price received by farmers at the cooperative/association. Excludes the ristourne at the end of the season.

Domestic Market

On the domestic front, coffee generally follows two marketing chains. One channel is through the marchandes and Madam Saras/wholesalers (women traders) to the local and regional markets. The second channel is through the voltigeur/speculateur and exporter circuit, to the local roasters (often owned by the big export houses themselves), which produce locally blended and roasted packaged coffee designed for the more sophisticated Haitian domestic consumers and for tourists. A large proportion of the coffee consumed in the domestic market is natural coffee, although a small amount of washed coffee is also marketed through the local roasters and exporters.

Among the most influential domestic roasters supplying the domestic coffee market are PRIMEX, Rebo, Cluadja, and Tasha. PRIMEX, the biggest local roaster, is an alliance of Haiti’s five biggest export coffee houses (Madsen, Brandt, Dufort, Kersaint, and Baptiste). PRIMEX sells its coffee under the “Ti Moulin” and “Le Café Primo” brand names to boutiques, supermarkets, hotels, and restaurants. Other smaller roasters also supply the domestic market with their own brand names and labels, such as Rebo, Claudja, Tacha, and Pidy.

Development Alternatives, Inc. (DAI) 21

Together, these roasters and traders Figure 5: Trends in Coffee Sector, 1950-1995 supply all the major regional cities and towns of Haiti to keep up with the Kgs country’s rapidly rising domestic 40,000,000 35,000,000 demand. Domestic consumption is Production reported to absorb as much as 65 30,000,000 Exports. 25,000,000 percent of the country’s total coffee 2.5 Kgs per cap. dom.consmp. production (Figure 5). However, the 20,000,000 15,000,000 2.6 Kgs per cap. increase in consumption is an outcome dom.consmp. of a rapid rise in population than from 10,000,000 an increase in per capita consumption. 5,000,000 0 Haiti’s annual per capita consumption of 1950 1960 1970 1980 1990 1995 coffee remains stable at only 2.6 kilograms compared with 4.3 kilograms for Brazil and Costa Rica. Overall, the picture of the domestic coffee industry is characterized by growth of domestic consumption, stagnation of coffee production, and steady decline in coffee exports (Figure 5).

Approximately 15-20 percent of coffee production is said to leak across the border into the Dominican Republic. For example, in 1999, 68,000 bags crossed the border to Dominican Republic from such places as Thiotte. Thiotte is a prime example of an area where coffee production is an important primary activity. There are an estimated 10 Dominican buyers, but one buyer (Mr. Kalo) alone is said to purchase about 80 percent of the coffee produced in the area. Farmers receive approximately 11 gds. for 1 pound of café parche, and because the buyers from the Dominican Republic are not too particular about quality, Haitian farmers take more cash home. Coffee growers around the Thiotte area sell frequently to these for several reasons:

1. Demand for quality is low; 2. Producer price is higher than what the intermediaries will pay; 3. The Dominican Republic is a ready market for Haitian coffee; and 4. Transportation cost to Port-au-Prince, the next big market, is generally higher.

Export Market

Almost all of Haiti’s export coffee is sold as green beans in European markets by a handful of exporters. PRIMEX, a consortium of five exporters, is by the largest exporter with a market share of 55 percent. Other exporters such as Weiner, which buys only washed coffee, command 20 percent of the export market, while Novella, operating out of Cap Haitien, controls about 15 percent of the export coffee market share (EWW, 1999). The traditional countries making up the E.U. market for Haitian coffee are Italy, France, Germany, and Belgium.

The coffee producer marketing organizations like FACN, RECOCARNO, and KOPCAB export less than 2 percent of Haiti’s green beans into promising niche markets of Europe and 22

the specialty and gourmet markets of the United States. In fact, FACN exports Haitian Bleu only to the United States. The other alliances and groups of coffee producers such as RECOCARNO are contemplating the production of organic coffee as well because the increase of the organic market shows no signs of slowing down.

The coffee market for Haitian green Figure 6: E.U. Imports of Arabica Green Beans, beans even in the European Union has 1993-1997 (ECU per kg.) been weak at best. Based on past performance of Haitian green beans in the E.U. market (Figure 6) and combined with the 20.00 continuing quality issues facing bean production, 10.00 green bean prices are expected to remain modest. 0.00 1993 1994 1995 1996 1997 Anticipating weak bean Haiti 1.17 1.65 2.50 2.07 2.87 prices and hoping to tap the Dom. Repub. 1.59 2.16 2.80 2.28 3.35 rapidly rising organic and Costa Rica 1.44 2.24 2.82 2.17 3.31 specialty coffee markets, Jamaica 9.94 7.41 10.40 17.88 13.86 ASDEC in collaboration with IICA and Agrotechnic is Source: ICO, 2000 gearing up to enter the organic coffee market. The goal of this collaboration is to begin production of organic coffee on 300 hectares in Haiti’s three best production zones: Beaumont, Dondon, and Thiotte. This pilot effort is eventually to induce local farmers to collaborate in the production of organic coffee. ASDEC and its collaborators envision creating IAA, an institute, to work closely with smallholder producers both on and off the pilot farms. IAA will be mandated to provide better crop management, introduce new coffee trees, promote the application of organic fertilizer, and practice effective pest management. IAA’s target is to produce 900,000 bags of green beans to enter the organic market.

Constraints and Opportunities

A key constraint clouding the alternative marketing channels opened up by the FACN and SEFADES, representing the marketing cooperatives following the Fair Trade market, is the issue of their sustainability. These organizations, which are several years old, operate with the assistance of a larger, international donors or projects. Although established to serve as the central marketing body for their respective members, these organizations were not created to operate as commercial entities in a highly competitive, demand-driven marketplace. So far, these organizations have successfully organized farmer-groups for processing and marketing purposes, serving as catalyst and guide for progressive change. This model deserves to be expanded in other parts of the country to increase volume and income for the producers. However, improvements would be required to transform these organizations into commercial entities operating on sound business principles and driven by

Development Alternatives, Inc. (DAI) 23

sustainability. The danger of not doing so could make the alternative marketing channels vulnerable to the vagaries of the market forces.

POLICY ENVIRONMENT

Policy Areas

Recognizing the limiting factor of taxation on coffee production, the Government of Haiti removed all taxes levied on coffee beginning September1987. Since then, no taxes have been imposed. A subsidy on fertilizer is provided to those smallholders using fertilizer for coffee. The fertilizer subsidy provided by the Government of Haiti is solely dependent on what the Japanese government donates toward the agricultural sector. Once this pipeline from the Japanese is stopped, it is unlikely that the Government of Haiti will continue this practice on its own, given its constrained resources. Furthermore, it is unclear whether the E.U.-funded STABEX program, with its 80 million gds. of program funds, is slated to cover costs of subsidizing fertilizer.

Government involvement in the coffee sector is nominal even in relevant policy areas. Many recognize the necessity of government investment in roads, research and development, extension, and pest management and in the provision of access to lines of credit to help develop and commercialize the agriculture sector. The government, however, remains uninvolved in these areas, preferring to pass on the issues to the international donor agencies, such as the European Union, USAID, World Bank, and UNDP/FAO as well as the international NGOs.

Haiti’s laws with regard to the coffee sector are encoded in the Le Code du Café, which dates back to the 1950s. However, the laws in practice are neither followed nor enforced.

Constraints and Opportunities

The absence of a national vision unifying the sector has given way to confusion and overlap among those that work in the coffee sector. The lack of a policy to support the sector has greatly limited its development because public investments are not being made in important areas such as research and development, extension services, and the like. This has also encouraged a dependency mentality on the donor community to provide all necessary support.

24

Development Alternatives, Inc. (DAI) 25

ANNEX I

THE U.S. MARKET FOR SPECIALTY COFFEE

Annex I—Coffee 26

Development Alternatives, Inc. (DAI) 27

THE U.S. MARKET FOR SPECIALTY COFFEE

The U.S. coffee market comprises multiple segments and product formulations. The majority of sales are of ground regular and instant/soluble coffee products, which are sold primarily under brand names such as Maxwell House, Folgers, and Nescafé. Whole bean and specialty coffee sales make up a small but increasing share of the retail market. Each of these markets can be further subdivided into additional product formulations—such as caffeinated/ decaffeinated, flavored/unflavored, Arabica/Robusta, and type of roast (dark, medium, etc.).

Contrary to common perceptions, U.S. per capita consumption of coffee has declined over the last 30 years. In 2000, per capita consumption is projected at 1.66 cups per day. This compares to 2.77 cups a day in 1960 and 2.02 cups a day in 1980. Although per capita consumption is decreasing, demand is increasing for specialty and gourmet coffees that often pay premium prices to producers.

The specialty coffee market is highly differentiated with no universally accepted definition as to what constitutes specialty coffee. For the purposes of this survey, “specialty coffee” is defined as coffee that is marketed in one of the four following categories: gourmet/rare origin/estate, organic, shade grown (also called “bird friendly”), and Fair Trade.

Table 8: Specialty Coffee Categories

Gourmet/Rare Coffee selling at premium prices because of the production Origin/Estate location (country, region, or estate) and perception of high quality. Examples include Jamaican Blue Mountain, Hawaiian Kona Fancy, Haitian Bleu, Papua New Guinea Sigri A, Ethiopia Yirgacheffe, and Kenya AA. Organic Coffee grown and processed without using chemicals (including chemical pesticides, herbicides, and fertilizers). Shade Grown Coffee grown under a natural forest canopy, providing a habitat for birds (hence also referred to as “bird friendly”), insects, other animals, and plants. Fair Trade Coffee purchased from Fair Trade certified cooperative members at generally higher prices than those offered by traditional marketing channels (i.e., middlemen).

Flavored coffees and gourmet blends are not included here because they are generally not a product of a developing country producer but a product of a roaster located within the market.

Annex I—Coffee 28

Production

World coffee production in the 2000/2001 crop year is expected to hit a record of 6.5 million metric tons, up 2 percent from the previous year. The majority of coffee is produced organically, but only a small portion is actually certified organic by independent organizations. Approximately 85 percent of Ethiopia’s production is thought to be organic, yet very little of that is actually certified. Although Mexico is the largest producer of organic coffee, only 2 percent of its production is currently certified. In Guatemala, registered organic coffee production accounted for about 5 percent of total production in 1999. This level is expected to increase in a few years as transitional farms become fully certified.

FigureFigure 7: World1: World Green Green Coffee Coffee Production, Production, 1995-1999 7,000

6,000

5,000

4,000

3,000 MT 000s

2,000

1,000

0 1995/96 1996/97 1997/98 1998/99 1999/00

North America South America Africa Asia & Oceania Source: USDA/FAS

Fair trade coffee constitutes a small percentage of total world production. Of the 240,000 metric tons of green coffee produced by Uganda in 1999, only 307 metric tons were produced by Fair Trade certified producers. However, this is an increase from 170 metric tons in 1997. Tanzania produced 833 metric tons of Fair Trade coffee in 1999, down from 888 metric tons in 1997 and representing only a small portion FigureFigure 8: U.S.2: US ImportsImports of of Coffee Coffee of its 46,000 metric ton harvest. 1,600,000

Prices for standard varieties of 1,200,000 coffee fluctuate widely from year to year, mainly because of supply 800,000 factors. Current high production MTs and high stock levels have 400,000 continued to place downward pressure on prices. Over the short 0 term, prices are expected to 1996 1997 1998 1999 remain at or under $0.90 per Arabica Robusta Other pound. Source: US U.S. Census Census

Development Alternatives, Inc. (DAI) 29

Imports

U.S. coffee imports increased over 1996-1999 in volume, but declined in value. In 1999, the U.S. imported 1.3 million metric tons of coffee worth $2.7 billion. Green, unroasted, and caffeinated coffee accounts for the majority of U.S. imports: 1.1 million metric tons ($2.3 billion). Figure 9: U.S. ImportFigure Market3: US Import Share—Arabica Market Share Coffee (MTs) Arabica Coffee (MTs) The United States imported 543,000 Colombia metric tons ($1.2 billion) of unroasted Guatemala 19% Arabica coffee in 1999. Seventy 14% different countries have exported Costa Rica unground Arabica coffee to the U.S. 4% market over the last four years. Brazil is Mexico 21% the largest supplier, with an import Peru market share of 22 percent by volume in 4% 1999. Other major suppliers include Mexico (21 percent), Colombia (19 El Salvador 4% percent), and Guatemala (14 percent). Brazil Other 22% 12% Imports of unground Robusta coffee Source: U.S. Census reached 601,000 metric tons ($1.1 billion) in 1999. The U.S. imported unground Robusta coffee from 84 different countries during the last four years. Brazil is also the largest supplier, with an import volume share of 26 percent in 1999. Other major suppliers include Colombia (13 percent), Vietnam (12 percent), Mexico (10 percent), and Guatemala (9 percent).

Specialty Coffee Market Figure 10: U.S. Import Market Share— FigureRobusta 4: US Import Coffee Market (MTs) Share Robusta Coffee (MTs) The definition of specialty coffee is vague, Vietnam so it is difficult to estimate the exact size 12% Mexico of the U.S. market. The Specialty Coffee 10% Association of America (SCAA) defines Colombia Guatemala specialty coffee simply as quality coffee. 13% 9% Specialty coffee may be marketed by Indonesia country of origin; by method of production 4% (organic, shade grown, etc.); and by Peru region, estate, roast color, and/or flavor. In 4% reality, many niche varieties fall under the Brazil Ecuador 26% specialty category. The primary common Other 4% characteristics of specialty coffee are that 18% they (1) usually are of very good quality; Source: U.S. Census (2) are differentiated in some way from other coffees; and (3) command higher prices than the more common commercial blends. Specialty coffee is also usually of the Arabica variety, known for it good flavor, high quality, and lower caffeine content.

Annex I—Coffee 30

According to the SCAA, gourmet/specialty Figure Figure5: US Retail 11: Sales:U.S. Retail Gourmet/Specialt Sales: y coffee sales exceeded $5 billion in 1999. This Gourmet/SpecialtyCoffee Coffee figure is higher than other estimates of gourmet 7.0 coffee sales, at $1 billion in 1990 and $2.5 billion in 1995. Although gourmet coffee will 6.0 account for only 8 percent of the volume of 5.0 beans roasted in 2000, it will make up one-third 4.0 of the $18 billion in total retail coffee sales. 3.0 US$ Billions 2.0 Despite the lack of detailed sales figures, the 1.0 specialty/gourmet/quality coffee market is 0.0 growing rapidly. Furthermore, its share of the 1990 1995 1999 2000 entire market will continue to increase with the expansion of the retail coffee house sector. Source:Source: VariousVarious industrIndustryy estimates estimates

Gourmet—Rare Origin—Estate Coffees

Table 9: Gourmet/Rare Origin/Estate Coffees Offered at The Cafe Connection/Coffee Masters ($/lb. for 5-lb bag)

Haitian Bleu 11.50 Jamaican Blue Mountain 39.95 Jamaican High Mountain 19.95 Hawaiian Kona Fancy 29.00 Tanzanian Peaberry 10.50 Tanzanian Kiliminjaro AA 10.00 Ethiopia Harrar 11.50 Ethiopia Yirgacheffe 12.00 Kenya AA 14.50 PNG Sigra A 11.00 India Monsoon Malabar 11.50 India Mysore Nuggets 12.50 Celebes Kalossi Toraja 13.00 Java Jampit Estate 10.50 Sumatra Mandheling 12.25 Brazil Santos 9.50 Colombian Supremo 9.50 Costa Rican “Doka” Estate 10.75 Costa Rican Tarrazzu 11.50 Guatemala Antigua 10.50 Mexican Altura Coatepec 9.50 Source: www.thecafeconnection.com, October 2000

Imports of rare origin coffee remain low. This is mainly because available volumes are so small. In 1999, the U.S. imported only 16 metric tons of coffee from Jamaica (the majority of Blue Mountain coffee is sold to Japan) and only 36 metric tons from Haiti. Estate volumes vary depending on the size of the estates, although the number of estate brands marketed continues to increase.

Development Alternatives, Inc. (DAI) 31

Gourmet, rare origin, and estate coffees generally retail at prices above $9.00 per pound. For the rarest coffees such as Jamaican Blue Mountain, prices can reach $40 per pound or more.

Organic Coffee

Organic coffee is thought to make up only 1 percent of all U.S. coffee sales and only 5 percent of the gourmet market, although some estimates are much higher. According to a survey conducted by the Specialty Coffee Association of America and Gourmet Retailer magazine, specialty coffee shop owners report that 10 percent of sales are organic. Some of the higher estimates of 7-10 percent of the market may be justified, considering that much of the organically grown coffee is not sold as such because it is not been certified (also called “passive organic”).

What is clear is that organic coffee is one of the fastest growing niches in the specialty coffee market, with growth estimated at 20 percent per year, according to industry sources. Indicative of this was the announcement in April 2000 by Seattle’s Best Coffee, one of the largest specialty coffee companies in the United States, that it was expanding its organic line from four to six blends because its organic sales had increased from 55,687 pounds in 1998 to 74,881 pounds in 1999.

Certified organic coffee fetches a premium of 10-15 percent above standard gourmet coffee. The main constraints to smallholders entering the organic market are obtaining, paying for, and maintaining organic certification. Producers need to obtain certification from independent third-party organizations, have a three-year history of not using chemicals, have a five-year farm plan, keep written records for an audit trial, and agree to annual inspections. Furthermore, the product must be maintained organic throughout the marketing chain. As of 1999, only 10 percent of U.S. roasters were certified organic, according to Tea and Coffee Trade Journal.

Shade-Grown Coffee

The shade-grown coffee “movement” is a result of the increased production of technified (sun-grown) coffee in Mexico and Latin America. Through the early 1990s, an estimated 40- 50 percent of coffee production in the region was transformed from shade grown to sun grown. Increased sun-grown production has resulted in the clear-cutting of natural forests, increased use of agrochemicals, and a decrease in bird habitat. Latin American production regions are habitats for a large number of wintering U.S. bird species, and advocates point to coffee technification as the main reason for sharp drops in the number of U.S. songbirds. For this reason, shade-grown coffee is also referred to as “bird friendly.”

No reliable market information is available on the current U.S. sales volumes of shade-grown coffee. However, based on the increased number of offerings by coffee roasters and retailers, sales are expanding. Organic coffee is often also shade grown because production under shade reduces the need for agrochemicals.

Annex I—Coffee 32

Fair Trade Coffee

The Fair Trade coffee market has evolved as Fair Trade Criteria part of a worldwide movement to better the terms of trade for producers in developing For an importer to become certified, it must countries. For coffee, this typically means meet the following requirements: removing the middleman from the transaction by having certified cooperatives sell directly to Pay a minimum price to small farmers included in the International Fair Trade importers and roasters at or above a “fair” Coffee Register. ($1.26 a pound FOB for minimum price. Some advocates point to the washed Arabica coffee, or $0.05 a pound common example of producers receiving just above the world price if it exceeds $1.26 $0.30-$0.50 per pound for coffee that retails a pound); for $8 per pound or more. Pay an additional $0.15 per pound FOB premium to farmers for certified organic coffee; An estimated 85 percent of Fair Trade certified Purchase beans from democratically coffee is shade grown and either passive or organized small growers; certified organic. Currently, over half of all Provide pre-harvest credit to growers; organic coffee is produced by Fair Trade and Agree to purchase on long-term and not certified cooperatives. Although U.S.-based one-time basis. organizations such as Equal Exchange (sales of $5.5 million in 1998) followed Fair Trade standards as far back as 1986, there was no national Fair Trade organization until 1996, when TransFair USA was founded. TransFair USA estimates that, although U.S. Fair Trade coffee sales totaled only 1.5 million pounds in 1999, this figure should reach 12 million pounds per year by 2002.

Table 10: Retail Selling Prices of Fair Trade Coffee

Roaster/Retailer Retail Price (per lb.) Starbucks $11.45 Peet’s $10.95 Tully’s (organic certified) $13.95 Equal Exchange $8.95 Source: Fintrac survey conducted in October 2000

The Fair Trade coffee movement in the United States received a big boost in 2000 when Starbucks Coffee began offering the product in its 2,300 company-owned retail stores (under pressure of protests organized by Global Exchange). It is being offered alongside Starbuck’s Shade Grown Mexico (sold in partnership with Conservation International) and organic Costa Rica coffees (certified by the Costa Rica agency, Eco-LOGICA). Starbucks plans to sell Fair Trade coffee for one year to assess demand. Other large roasters and retailers are also entering into licensing agreements with TransFair USA—such as Peet’s Coffee (57 stores), Green Mountain Coffee, and Dean’s Beans.

Research by TransFair USA shows that 50 percent of U.S. consumers will to pay $1.00-2.00 more per pound for Fair Trade coffee. TransFair USA estimates that Fair Trade coffee should be feasible if retailing in the range of $8.00-10.00 per pound, although most of the larger

Development Alternatives, Inc. (DAI) 33

roasters and retailers are pricing their Fair Trade coffee higher than this in anticipation of increased demand. The trend is being further supported through promotional activities of groups such as TransFair USA, which is planning campaigns in Seattle, Minneapolis, and Boston in 2000, with 12 more campaigns planned over the next three years.

Annex I—Coffee 34

Development Alternatives, Inc. (DAI) 35

ANNEX II

PERSONS CONTACTED, COFFEE SUBSECTOR STUDY

Annex II—Coffee 36

Development Alternatives, Inc. (DAI) 37

Persons Contacted—Coffee Subsector Study

Company/ No. Date Person Address Organization WEEK 1 Oct. 23 Monday Arrive in PAP/Tim Aston HAP/DAI PAP 1 Oct. 24 Tuesday Zach Lea SECID PAP 2 Oct. 25 Wednesday George Conde Private Sector PAP Federation de Association du 3 Oct. 25 Wednesday Stephane Jean-Pierre PAP Café Natural (FACN) 4 Oct. 25 Wednesday Jean Claude Fignole Private sector PAP 5 Oct. 26 Thursday USAID Library USAID PAP 6 Oct. 26 Thursday Martial N. Bailey USAID PAP 7 Oct. 27 Friday Pierre Yves Boulain STABEX/EU PAP 8 Oct. 27 Friday Hubert Dufort & C. Weets ASDEC PAP WEEK 2 9 Oct. 30 Monday Ag. Entianne Ministry of Agriculture PAP 10 Oct. 30 Monday Alfredo J.Mena Pantoleon IICA PAP 11 Oct. 30 Monday Marc-Eddy Martin USAID/CTO PAP Cap Rouge coffee growers associations (3) and 15 12 Oct. 31 Tuesday Cap Rouge, Jacmel Jacmel association members and Stephan Jean Pierre Mario Mora; and FACN Tom 13 Oct. 31 Tuesday EWW, FACN processing plant manager Gateau Villa 14 Nov. 1 Wednesday Davenport briefing HAP/DAI Creole

15 Nov. 2 Thursday Henry Christopher Louis SERVICOOP PAP

Ag. Ettianne, Zach Lea, Jose 16 Nov 2. Thursday MOA, SECID, EWW PAP Gameil Coffee and cocoa production Dondon and Grande Riviere Cap 17 Nov. 3 Friday cooperatives du Nord Haitien Cap 18 Nov. 4 Saturday B. Dean Teadwell Bosco Foundation Vincent Haitien Lulu Leon, and Buddy, Prod. SEFADES and Cap 19 Nov. 4 Saturday Manager RECOCARNO Haitien WEEK 3 20 Nov. 6 Monday/ Mr. Wierner Weiner Exporter PAP Thiotte/ Jean-M. Vital 21 Nov. 7 Tuesday KOPCAB/Oxfam met in Depart PAP

Annex II—Coffee

39

CHAPTER TWO CACAO11

SUMMARY CONCLUSIONS

The prospects for the Haitian cocoa industry are directly linked to the international price of the commodity. The price is strongly depressed because of reduced demand from Russia as a result of its weak economy and because of the ISO 2000 standards ruling in the European Union that allows dilution of cocoa in chocolate. There has also been increased production from Côte d’Ivoire and from the Far East. The stocks of cocoa held by the industry, although dwindling, also serve to depress prices for as long as these stocks last. Therefore, this is no time to expand cocoa production in Haiti with new plantings or by using purchased inputs. It is rational to continue with a zero input production system, which minimizes exposure to financial risk, and to delay any replanting program until the stocks are depleted.

Alternative tree crops give considerably better returns to both labor and land, and several of them have far greater potential for profitable production under intensified management than does cocoa with the current prices. Annual field crops also give better returns to land than cocoa.

The international price has been depressed and, no doubt, will recover again at some time— particularly as stocks are used up. It is not therefore advisable to remove the plantations, particularly as the farmers are already used to managing them. The cost of retaining the plantations, at current production levels, is the opportunity cost of the land, which, although comparatively high, does not justify destroying a potentially productive stand.

In this case, it is rational to ensure that output from the existing orchards is optimized so long as returns from management are positive, as indeed they are. Existing stands should be pruned, topped to 5 meters, and protected from pests through crop sanitation. Shade trees should also be trimmed to limit shade of the cacao canopy to 50 percent. Increased output from existing stands would bring with it economies in both production and marketing. Programs with these aims should nonetheless be aware that there are likely to be calls on labor capacity that would give higher returns than under the current cocoa price regime.

Returns to farmers would be improved by standardization of the weights and measures used in the marketing chain and by a sustained challenge to the exporting cartel such as SERVICOOP has initiated.

Small increases in the international price can be reflected by much greater price rises at farm- gate because the costs in the marketing chain remain absolute, although there would be a tendency for greater profit taking in the informal sector. For example, a 10 percent increase in a New York market price of $800 per ton could result in a 20 percent increase on 4 gds.

11 In this report, the word “cacao” refers to the trees and the word “cocoa” refers to the product.

Chapter Two—Cacao 40

per pound and a 40 percent increase on 2 gds. per pound if there was no increase in profit taking by intermediaries.

If the New York market price were to triple, and there was no increase in profit taking in the market chain, the return to 1 hectare of well-managed cacao would rise from 3,300 gds. to 27,000 gds. Although profitable, this price is still less than the potential earnings from plantain or mango. However, the open-ended international market, the familiarity of the crop to growers, and its suitability to production in outlying areas would make the prospects for expanded production attractive under such a price regime.

The multiplying effect of international prices at farm-gate is greater when there is efficiency in marketing, a condition enhanced by the presence of competitors who can maintain the rigidity, or at least minimize the elasticity, of the cost chain between farmer and international buyer. Programs to support components of the marketing structure should be wary of distorting the competitive forces of the players because this could ultimately undermine the basic arena for streamlined marketing.

RECOMMENDATIONS

1. Investigations into the status of cacao plantations and nurseries in Haiti should be carried out to determine their ownership, location, extent, age, and health. This information would provide a useful baseline for programs to enhance output.

2. A database on regional producer prices, intermediate trading prices, exports revenues, and destinations of coca should be maintained by the project. This would be useful information for efforts to streamline market revenue to the farmers and counteract the formation of cartels.

3. HAP should aim to provide support for capacity building in extension services to growers that would aim to:

Maintain existing cacao orchards and discourage their destruction;

Improve the productivity of existing orchards on zero or low inputs by promoting removal of unproductive trees, pruning, topping, shade control, ventilation, drainage, and vertebrate pest control;

Standardize weights and measures used in the marketing chain;

Sustain technical assistance to marketing channels that challenge exporters’ cartels; and

Streamline the flow of market revenues to the grower, thereby increasing his or her share.

Development Alternatives, Inc. (DAI) 41

4. The program should also prepare to provide support to the replanting or establishment of new cacao plantations in the event that market analysis indicates an imminent resurgence of international prices such as may transpire within the coming few years as cocoa stocks are depleted. HAP should keep an ear to the ground in the international market by subscribing to market intelligence reports and by periodically asking the opinion of cocoa marketing specialists. The project should prepare a strategy for the expansion of cocoa production to be employed in the event of clear prospects for an upturn in the market. Such a strategy would include maintenance of nurseries with potential for expansion and would identify actual production sites and their owners to be targeted with assistance.

5. The project should promote the updating of the Cacao Act by the Ministry of Agriculture by providing legal assistance and a review of the Aproma recommendations. This would establish a clear legal environment for commercial initiatives when the cacao industry enters a renewal period.

INTRODUCTION

This assessment of the Haitian cocoa industry provides information to develop a strategy for maximizing farmer incomes from cocoa. The study was conducted by visiting production areas in the regions of Jacmel, Cap-Hatien, and Jeremie and by interviewing stakeholders on the land, in farming cooperatives and marketing chains, and in other cooperatives and private companies as well as officials in the Ministry of Agriculture, donor agencies, and NGOs.

PRODUCTION

Producers

Farmers typically produce a range of crops, and cacao is one of them. Usually, a farmer would aim to produce subsistence crops and cash crops, and cacao has proved to be dependable as a cash crop, requiring little attention or skill once it has been established.

The farmers often are illiterate and therefore lack the skills to easily acquire further knowledge of production and markets. Their numeracy skills are also low.

About 10-15 farmers form a group, and the groups make up the membership of grower associations and cooperatives, which have up to 600 members. The cacao industry shares communication mechanisms with the coffee industry because many of the cacao-growing areas are shared with coffee. There is no significant institutional structure beyond the associations and cooperatives to serve as a channel between government or any other development agency and the farmers themselves.

Chapter Two—Cacao 42

Cocoa-producing cooperatives in the Grande Riviere du Nord area are:

Jean Babtiste Chavanne; Port Margot—CAPUP; Borgne—KAFUBO; Anse a Foleur; Tibo; and Babon—KOTAM.

Areas in the same region producing cocoa but without cooperatives are:

Dondon; Limonade; and Milot.

Areas of production in Grande Anse are:

Dame Marie; Moron; Chambellan; Anse d'Hainault; and Les Trois.

There has been no recent survey of the extent of production, but it is generally assumed that there are 10,000 hectares distributed as shown in Table 11. Total production is 3,060 tons per annum, which at 1 kilogram per tree implies 3.06 million trees. If these were all planted at the recommended spacing of 3 by 3 meters, this crop would cover an area of only 3,060 hectares. But because the plantations are not monocropped and the spacing usually is much wider and the yield lower, cacao is distributed among 10,000 hectares. The majority of production is from Creole Gardens and contains a varied menagerie of tree and root crops.

Table 11: Cacao-Growing Areas

District Area in 1995 Ha Grande Anse 6000 South 50 South East 100 North East 200 North 2600 North West 1000 Centre 50 Total 10,000 Source: MARNDR, 1996

Development Alternatives, Inc. (DAI) 43

Rainfall in most cacao-growing areas of the world is between 1,200 and 3,000 millimeters, and at the lower end, the distribution becomes increasingly critical. As a result, some of the areas listed in Table 12 are in cocoa production are already marginal.

Table 12: Rainfall Precipitation

Median Precipitation Region Zone Months with > 50 mm (mm) Grande Anse Miragoane 1,389 4 Anse d'Hainault 1,566 1 Beaumont 2,207 2 South Les Anglais 1,287 1 Tiburon 1,363 1 Camp Perrin 2,173 0 South East Jacmel 950 4 Thiotte 1,755 3 Cayes Jacmel 1,783 2 North East Saint Suzanne 1,529 1 Valliere 1,893 2 North Limbe 1,610 1 Pilate 1,716 1 Plaisance 1,786 2 Dondon 2,372 0 North West Port de Paix 1,163 2 Saint Louis 1,950 0 Centre Savanette 1,183 1 Cerca La Source 1,198 5 Maissade 11,793 3 Lascahobas 1,907 3

Farm Size/Type

In 1950, 33 percent of the farms were between 0.33 and 0.63 hectares, and 55 percent were under 1 hectare. Less than 1 percent of the farms were larger than 6 hectares.

There is a steady tendency under Napoleonic law, which prevails in Haiti, to divide land holdings among the offspring on the death of the owner, so farm sizes have further diminished and land parcels belonging to single individuals have become fragmented.

There has also been a tendency for workers to become squatters and to claim ownership of land such that the registered owner may not have access to it for production. It is said that certificates of ownership or land titles do not carry the authority needed to remove sitting tenants. This poses a constraint to the development of commercial plantations, which is further exacerbated by persistent crop theft.

Chapter Two—Cacao 44

The squatters may, however, feel secure enough to plant annual crops but may not be as willing to plant perennial crops without official clarification of their ownership status.

In Grande Riviere, where 70 percent of the land in production is in cacao, farmers of the Jean Babtiste Chavanne Cooperatives claim about 300 trees per family and an average output of 400 pounds per family. Ten families are said to own as much as 1,000 trees each. When asked to confirm these statistics, however, they stated there was no way of knowing.

Land Types

Land used for cacao production is not the prime land used for food crop production. The land under cacao tends to be sloping to steep and usually also is occupied by forest trees or other domesticated trees such as mango and avocado.

Shaded cacao is not demanding of fertilizer, so this is as expected, although the intensity of shade is usually too great for optimal zero-input production conditions (50 percent at cacao canopy level). The small plantations are often found in ravines, partly because other crops are not suited to them but also because cacao is sensitive to strong winds and the ravines afford some protection. Light wind, in contrast, helps distribute the midges that affect pollination and provides an atmosphere that is conducive to growth.

Productive Infrastructure

No infrastructure is associated with the production of cacao, and the production system relies on gathering the pods only three times per year. The pods are broken open by hand, and the beans scooped onto banana leaves or into marmites or baskets where they are left as the mucous dissolves from the beans. This phase is far from what could be called a fermentation process, and no infrastructure is in place to standardize and regulate it.

The cocoa beans are carried away in baskets or buckets on animals and dried further on the ground or on patios before sale. Where patios do exist, they are the only form of infrastructure related to cocoa production and usually are used for several other agricultural and social functions as well.

Inputs and Technology

No inputs are applied to cacao production in Haiti. The system is a low input-low output one with minimal exposure to financial risk.

If shade were to be reduced and inputs applied, yield could be increased, but this is not justified in the context of the current international price or the technical skills of the planters.

Development Alternatives, Inc. (DAI) 45

Input suppliers are:

Agrisupplies; Agriboutique; Cabinet Nereget; and Agrotechnique.

Varieties

Improvement to plant stock is made either by the selection of promising material to be grafted onto existing root-stock, which requires skills are not commonly found in Haiti, or by the selection of strains for hybridization.

Improvements to the national cacao stock were inhibited by the loss of the planting plan for the clonal gardens near Grande Riviere. (A report on establishment of clonal gardens is included as Annex I.) A new clonal garden has now been created, and trials continue under PADF.

In 1983, 21,000 hybrid seedlings were produced from seeds donated by Dominican Republic, and 60,000 seedlings were planted in 11 regional nurseries for distribution to farmers under MENA.

Agro-Chemical Use

No inputs are used in cacao production under the current system in Haiti. The challenges posed to the crop by Black Rot (Phytophthora) in particular are better met by thorough harvesting of the top pods than by any attempts to spray with fungicides. The problem is that pods at the top of high trees are left because they are out of reach. They are attacked by Phytophthora as they rot, and the spores drip down onto the maturing fruit lower on the trees, infecting them as well. Pruning the tree to 5 meters in height and ensuring that all pods are harvested when ripe give better yield from the orchard by improving the light and ventilation and by removing the source of infestation.

Woodpeckers cause extensive damage by making holes in pods to gain access to the fleshy layers, thereby providing access by rats and bats to the beans. The only measures taken are to destroy woodpeckers with catapults, but there is no systematic prevention of rats from trees by using tin-sheet collars and it is doubtful that such measures would be cost-effective.

Access to chemical fertilizers in rural areas is limited. Private sector distributors are threatened by subsidized fertilizer distribution and are unlikely to extend their own distribution networks while the donor-aided programs may undermine their margins. Fertilizer has been distributed at 50 percent of the market value in previous programs, and Japanese donations of fertilizer have continued for 10 years.

Chapter Two—Cacao 46

Cultural Practices

Cacao should be produced either under shade with no fertilizer for low cost, giving low output, or without shade but with fertilizer for higher yields.

The most sustainable system requires that farmers minimize exposure to financial risk and satisfy themselves with small cash surpluses from subsistence production, rather than aim for higher returns using purchased inputs. The shaded production system is therefore the best choice of small-scale farmers.

However, the depth of shade is too great in most of the production areas and no effort is made to control either shade trees or cacao trees. The normal practice in Haiti is not to prune the cacao trees to improve their shape, thereby enhancing access to light and air, and not to remove dead or unproductive trees or branches.

The trees are generally too high or too shaded so top fruits are not harvested, leading to high incidence of pod Black Rot and providing a constant source of food to vertebrate pests. Although there are many orchards have 3 by 3 meter spacing, there are many more where plantation structure tends to be random and plant spacing is far from optimal so that return per unit of land is by no means optimized.

Agricultural Credit

Because no inputs are used in cacao, there is no need for credit unless replanting or rehabilitation is planned. However, the farmers' immediate or emergency cash needs might be met by selling a portion of the cocoa crop at a low price while it is still on the tree.

Since credit structures in the informal agricultural sector command interest rates of 100 percent over only one or two months, the forward sale of a crop can provide an alternative to informal credit for the cash-strapped farmer.

IICA has initiated a credit program for women that relies on group accountability for repayment in lieu of collateral. The program encourages repayment through peer pressure because new loans to group members are made only on repayment of the first loan. The scheme charges 24 percent interest (standard commercial rates are 26 percent), and the borrowers lend on to others at 5 percent per month, illustrating the demand for cash in rural communities.

Labor

Although the minimum wage for labor is 36 gds. per day, agricultural labor in more remote areas is sometimes paid as little as 10 gds. per day. However, much of the work is paid on a task basis and amounts vary widely. Similarly there is wide variation in the duration of a working day; at times, it is less than four hours per day, which is a tiring day's work when applied to tree pruning for instance.

Development Alternatives, Inc. (DAI) 47

Labor shortage is cited by growers as the constraint to plantation improvement, but an appreciation by owners of the returns to tree management would, in all likelihood, result in the work being accomplished by them without hired labor.

Returns

Cooperatives in the north have reported that farmers are so dissatisfied with the returns from cacao that they have wanted to cut down their plantations. The cooperatives have had to mount an education campaign against the impulse, which they say has proved successful so far.

Product Prices (Farm-Gate)

Prices paid to farmers at the farm-gate are often on a volumetric basis rather than by weight. Fresh wet beans from the pod are expected to dry to only 40 percent of their weight, so the degree of processing by fermentation or simple drying has a crucial bearing on their value. The “dry” crop can also range in moisture content from the 6-8 percent required for safe storage and shipping to more than 40 percent, which would result in a 30 percent weight loss in drying. It is easy enough to dry a crop down to 6 percent in four to five days in the sun, providing the greatest solid matter ratio at the time of purchase by the exporter. When such dried beans are exposed to moist air at sea ports, they can be hydroscopic and gain moisture and therefore weight by 2 percent.

Tests carried out on cocoa beans by SERVICOOP show a non-linear relationship between moisture content and weight, and indicate a weight loss of 25 percent when moisture content is reduced from 13 percent to 6 percent. Although this defies the logic of the definition of moisture content, it does confirm the claim by exporters that total weight loss in the drying and sorting process amounts to 25 percent. The SERVICOOP experiment was not scientific, and its moisture meter does not read outside the 6-13 percent range.

The dilemma faced by the marketing chain is whether to buy wet or dry. If buying wet, the minimum price can be paid on the assumption that no drying has taken place. The drying can then be effectively managed to correct standards by the buyer. Buying wet carries the risk that the crop will become moldy before it can be properly dried, but at least this risk is under the management of the buyer.

The drying of the beans while in possession of the farmer is his cost and thus to the advantage of the buyer. However, because the beans are sold by weight, some farmers resort to the destructive practice of pouring water on the beans to increase their weight.

Thus, if the buyer buys “dry,” the risk is that the beans are actually wet to varying degrees and will become either over or under dried in the buyer's drying process, which will be done at the buyer’s expense. The farmers will profit by minimizing the drying process (and

Chapter Two—Cacao 48

consequent weight loss) while passing beans off as dry—but with increased mold because drying has been retarded.

The compromise is to pay according to moisture content, which is fine so long as this can be accurately quantified at every transaction point. A moisture meter costs about $300 and is calibrated on the basis of the electrical conductivity of the whole beans, reflecting their water content. This is also a function of the contact points between the beans, which, for a given volume, is smaller (and therefore less accurate) than for bigger grains or beans. Thus, the only significant result of the SERVICOOP experiment with weight loss on drying is to demonstrate the inconsistency of readings from the moisture meter.

In view of the expense, unreliability, and dubious accuracy of quantifying moisture content, the market depends on the judgment of the buyer and the seller about the actual moisture content and quality of the beans in determining transaction prices. Part of the portfolio of skills of a successful speculateur will be his or her professed long experience with beans and the mystique surrounding the determination of moisture content from the feel and sound of the tumbling beans, a sufficiently nebulous set of criteria to create a comfortable cushion for profit taking by gifted players in the trade.

Margins are made by distortion of the conversion rates and of the volumetric measures. Some of these distortions are applied with skill to the advantage of the voltigeur or speculateur, but sometimes are also applied, no doubt inadvertently, to the advantage of the farmers.

For example, intermediate traders usually buy in kilograms and make the conversion to pounds for the purposes of negotiation at only 2 pounds per kilogram (it should be 2.2046 pounds per kilogram), thereby gaining 10 percent advantage, or 204.6 pounds per ton. Scales marked with both pounds and kilograms are used to measure in kilograms only, and the conversion is then made to pounds simply by multiplying the kilograms by 2, despite the reading on the pound scale. SERVICOOP, in contrast, has been making conversions at the correct rate, but the advantage is not appreciated by the farmers because their lack of numeracy skills has prevented them from being aware of the loss through other channels.

Voltigeurs without scales use either gotés (soup cans) or marmites (gallon cans). A goté of cocoa of 10 percent moisture content weighs approximately 300 grams or 0.66 of a pound, and 1 gd. was being paid in the first week of November 2000 in remote areas of Grande Anse for this quantity. This is equivalent 1.5 gds. per pound.

Larger quantities are bought in a marmite for 8 gds. This weighs 2 kilograms, which is converted to 4 pounds using the Haitien traders’ conversion rate and implies a price of 2 gds. per pound. However, at the farm level a marmite is said to be equivalent to 8 gotés—in fact, 2.4 kilograms and actually 5.29 pounds. Therefore a voltigeur claiming to pay 2 gds. per pound is in fact receiving 5.29 pounds for 8 gds. when bought in gotés, resulting in an actual price to the farmer of only 1.5 gds. per pound.

Voltigeurs selling to speculateurs in the market on a volumetric basis claim to be selling at 2 gds. per pound. They make their margin because in the market there are only 6 gotés to the

Development Alternatives, Inc. (DAI) 49

marmite. This provides the voltigeur with 2 gds. per marmite for transport and commission— that is, 33 percent.

If, which is more likely, the voltigeur is selling by weight, having bought by volume, his or her margin is ensured because the marmite was filled on the assumption it contained 4 pounds when actually it contained 5.9 pounds—a 32 percent margin so long as the beans have not dried out.

Speculateurs buying at village markets judge moisture content and pay between 2 and 3 gds. accordingly. One speculateur claimed to be able to buy 100-200 sacks per market day in Grande Anse, and another said he could buy 1,000 per month if not constrained by working capital.

In Grande Riviere du Nord, in 1999 farmer prices were 2-4 gds. per pound. A farmer and association leader in Cap Rouge claims to achieve 6 gds. per pound and 8 gds. at times of shortage in Jacmel, but the quantities are small and the beans are used for the local production of cocoa butter.

The major buying season for cacao in 2000 has not started in the north at the time of writing because SERVICOOP has been delayed on the market by its own structural adjustment and exporters have not yet stepped in with any force. SERVICOOP has decided to pay 4.25 gds. for cocoa of less than 8 percent moisture content and less than 4 percent impurities, and 2 gds. for higher moisture content. In some cases, its regional officers are paid a commission of 1 gd. per pound for cocoa purchased.

Farmers in JBC in Grande Riviere du Nord report receiving 5-7.50 gds. per pound some years ago when a better international price regime prevailed.

The impact of inflation on farm-gate prices tends to be more devastating in Haiti than elsewhere in the world because farmers are isolated as compared with other sectors of the community with more frequent contact with the commercial world.

Transport Costs

A donkey (with its owner) that can carry up to 90 kilograms for 25 kilometers costs 30-36 gds. per day.

Transport by boat is 20 gds. per 60 kilogram bag from Grande Anse or seemingly anywhere else to Port-au-Prince—a distance of 192 kilometers as the crow flies. Speculateurs pay workers 4 gds. per bag to load, but the transport to Port-au-Prince is paid by the exporter. By truck, it is 50-60 gds. per sack and the journey can take up to nine hours. Cap Hatien is only 138 kilometers as the crow flies to Port-au-Prince but can take seven hours to drive from there by car.

Chapter Two—Cacao 50

The state of the roads in rural areas increases the real cost of transport because of wear and tear on vehicles and time taken to reach destinations. The bad roads also increase the isolation of the productive areas and limit the scope for production to crops with tolerance of bumpy rides and with reasonable value/weight ratios.

Seasonality and Variability

Variability in the farm-gate price relates to the international market price and, for small quantities only, the demand for local processing. Variability is also related to distance from the selling point, the moisture content, the value added by processing, the quality, and the bargaining power over the farmer.

Flowering takes place as the temperature rises from its lowest point, and unless it is dry, harvest takes place six months later. The main harvesting season in Haiti is from August through February; however, some farmers harvest only three times in this period and many of the pods are not therefore picked at optimal ripeness. Speculateurs say that cocoa trading continues throughout the year.

Yields

Yields range from insignificant amounts per tree to more than 1 kilogram of dried cocoa per tree per year. This depends not only on the viability of the tree but also on its position in the orchard or garden in relation to other trees, sunlight, wind, and rainfall.

It is expected there will be 20-50 beans per pod, which, when dried, should not exceed 110 per 100 grams of 0.9-1 gram each. Smaller beans are not popular on the international market although M&M Mars has stated that the bean size is not a concern to the company.

Some orchards are planted up with spacing of 3 by 3 meters to yield 1,100 per hectare, while others are placed haphazardly through the plot at random density. Many trees within cacao orchards are almost completely unproductive and should be replaced, and most are not producing to their full potential.

Grande Riviere growers claim 300 trees per family but an average output of 400 pounds, suggesting a yield of 1.33 pounds per tree.

Tree populations are between 200 and 240 trees per hectare, whereas in a monocropped situation there should be 1,100 trees, giving a yield in excess of 1 metric ton per hectare.

Development Alternatives, Inc. (DAI) 51

Production Costs

The only production costs faced by the farmers under the zero input system currently practiced are for pruning and harvesting. The majority of the farmers are not currently pruning because of low prices.

However, in early November pruning of an experimental plot in Grande Riviere was done under supervision of PADF staff with the aim of maintaining a tree height of 5 meters and allowing free air movement and no conflicting branches. This took 2.5 hours with six workers on 0.2275 hectare of trees at 3 meter spacing, giving a rate of 65 person-hours per hectare, which at 36 gds. per day implies a cost for hired labor of 292 gds. per hectare for an 8-hour day. This should optimally be done twice per year, therefore costing 584 gds. per hectare, although there would be less work per cycle on frequently pruned orchards.

The yield benefit from pruning should be at least 50 percent over non-pruned trees and more likely 100 percent. If this is equivalent to an improvement of 0.5-1 pound per tree at 1,100 trees per hectare, the benefit in yield would be 550-1,100 pounds, or 1,100-2,200 gds. at 2 gds. per pound, and the margin over the cost would be 550-1,100 gds. per hectare.

High-yielding crops in well-maintained orchards can be harvested at 1,500 pods per labor- day, and an additional labor-day is required to open them. At an average of 40 beans per pod and 1 gram per bean, this amounts to 60 kilograms per 2 labor-days, or 30 kilograms each. This amounts to 1.2 gds. per kilogram. There is less margin at the lower density of pods in the low yielding orchards.

One farmer in Cap Rouge reports that three workers harvest 40 marmites in a year, taking one day in each of the three seasons. This amounts to only 4.4 marmites per labor-day, which would dry to about 10 kilograms of beans valued at 40-80 gds. The returns could be low enough only to just cover the harvesting costs.

Most of the work is usually done by family labor, but occasionally a work-day is organized with neighbors and who are paid with food and rum.

See Annex III for labor requirements for operations in commercial cacao orchards.

On-farm Processing

The only on-farm processing carried out now with cocoa is leaving the beans in a basket for two or three days to let them drain off before drying for two to five days and some sorting. However, the marketing system does not favor this practice by farmers because it reduces the weight and the volume of the crop that can be sold. If a farmer is ensured of being paid for good quality dried beans, he or she will spend more time preparing them, but there is no institutionalized mechanism for quality control at the farm level or indeed in the marketing chain except the standards laid down by the ultimate buyer of the beans.

Chapter Two—Cacao 52

Fermentation adds value to the beans by enhancing the taste but only if the market requires it. This is not the case with M&M Mars, which pays a premium of $90 per ton over New York daily prices and prefers un-fermented beans from Haiti. Other markets, however, pay a premium of $110-$120 for fine and flavor cocoa that has been fermented correctly in other countries.

Correct fermenting requires that cocoa beans be collected into lots of at least 80 kilograms and preferably 100-140 kilograms, fresh from the pods, and that they are turned every one or two days, usually through a cascade of fermentation boxes, for up to five days.

The purpose of fermenting is to remove the mucilage, prevent germination of the embryo, enhance the flavor, and initiate the drying process. This has obvious advantages if there is sufficient scale of production and the market rewards a homogeneous product with specific flavor. It requires concerted harvesting and a system for consolidation of the product on a weekly basis until volumes justify further fermentation cycles.

Net Returns to Farmer

Returns from cocoa production are at a low that is not unprecedented but may be more devastating because of the exacerbating effects on farmer income of poor returns to coffee and the deterioration of roads. Increased poverty leads to more intermediaries in the market chain, trying to create a living, and probably more exploitation of the poor skills of farmers in negotiation and their desperation in eking a living from the land.

Income from comparable tree crops is more promising, but the nature of perennial cropping implies there are time and capital barriers to entry that preclude the option to switch with ease between the fluctuating markets but also preserve the standing crops for better times.

The table below ranks alternative tree crops and illustrates why cacao is not enjoying current popularity.

Table 13: Summary of Estimated Returns from Alternative Tree Crops

CropProducer Price Yield per Spacing Output Comment per Unit Tree per Meters per Ha. Annum (gds.) Plantain 2 gds./kg 36 3.5 x 3.5 58,776 Quick to establish, market not satisfied Banana 1.5 gds./kg 20 2.5 x 2.5 48,000 Poorly developed market Coconut 5 gds./nut 35 7 x 7 35,714 Basic market Mango 0.8 gds./kg 300 10 x 10 24,000 Improvements for export. Avocado 0.9 gds./kg 200 9 x 9 22,222 Improvements for export. Litchi 2.5 gds./kg 70 9 x 9 21,605 Poorly developed market Cashew 25 gds./kg 18 15 x 15 20,000 Processed kernel Coffee 37 gds./kg 0.2 1.5 x 2.5 19,733 Under intensified management Cacao 4.4 gds./kg 1 3 x 3 4,889 Under intensified management

Development Alternatives, Inc. (DAI) 53

The choice offered to growers by alternative annual crops is one of where to apply their energy and financial resources because removing trees is not a sensible option when the world market price is depressed.

The annual crops that present significant possibilities for both subsistence and cash returns are maize, beans, yams, malanga, and pineapple.

Constraints and Opportunities

Production-related Constraints and Impact on Profitability

The low market prices that prevail as a result of world market conditions leave little scope for profitability. However, the trees are standing and returns should be optimized with minimum input. It is unwise to remove the trees unless they are unproductive because the world market price will improve as stocks are diminished and as the Russian market recovers.

Productivity of orchards is severely constrained by poor tree management and poor orchard management with regard to shade and crop sanitation. Trees are generally too high (well over the recommended 5 meters), creating too much shade on lower branches and putting many of the pods beyond easy reach where they are left to rot and harbor Phytophthora. Ventilation is inhibited by too much top growth, and light does not penetrate sufficiently into the tree.

Shade trees are too high and reduce sunlight access to the top cacao canopy by more than the recommended 50 percent.

A high proportion of the trees are unproductive and could be removed for firewood without loss of production from the orchard, In fact, the remaining trees would increase productivity in a more open environment.

Opportunities for Significant Improvement

Pruning the trees is an investment in future yields and insurance against infestation by pest and disease. Although the returns are not immediate and the benefits not strikingly apparent, farmers would be well advised to maintain their orchards to improve the little returns they currently make and to ensure good response in the event of improved prices.

The prunings from cacao and shade trees provide valued firewood or charcoal, thus mitigating the costs of removal.

Farmers cannot be advised in the current economic climate to increase their plantings or to apply fertilizers to their cacao.

Chapter Two—Cacao 54

POSTHARVEST

Storage

Storage of well-dried cocoa does not pose a problem in the short term, but the only reason to store is as a short-term savings device or in anticipation of the arrival of a buyer or a price rise. Generally, farmers do not store their cocoa for longer than necessary for the arrival of a buyer or a market day.

When the crop is not well dried, storage poses a problem because mold causes rapid and drastic deterioration. This is a major consideration in the marketing policy and the incentives paid for dry crops.

Drying capacity is an important component of the viability of a trader of any scale and may consist of a space on clean earth, a cemented patio, a sliding tray with roof, or a force- ventilated bulk store. The last two were most seen in Haiti by the consultant, but polythene- covered frames are under construction.

Losses resulting from mold contribute to deterioration in overall quality and hence of price at all transaction points in the marketing chain. The defense against such losses is simply effective drying with minimal delay.

Some aspects of losses from drying are discussed under Farm-gate Prices, page 47. In principle, the reduction in moisture content should coincide precisely with weight losses and values should be directly correlated, although drying costs will not be linear in relation to moisture content because drying rates decrease at lower moisture content levels.

The conventional wisdom is that there will be a 20-25 percent loss of weight between the farm-gate, and export, resulting from drying and sorting. If 25 percent reduction is applied to 1 pound of cocoa bought at farm-gate for 2 gds. per pound, the cocoa costs 2.667 gds. per pound at point of export, excluding value-added costs such as transport, bagging, drying, and sorting. However, there are many compensating factors associated with conversion rates and volumetric measurements.

Security and Availability of Financing

Availability of working capital is a constraint to marketing capacity and increasingly more so at levels closer in the chain to the farmer. Commercial interest rates are 26 percent, but access may be limited because of lack of collateral. Informal interest rates can be as much as 100 percent per month and are not therefore a viable alternative for cocoa trading.

Security is a concern, and measures to safeguard consignments against accidental loss are not apparently considered within country. One speculateur in Jeremie lost a large consignment of

Development Alternatives, Inc. (DAI) 55

cocoa when a boat was sunk and has had no compensation. He also professes to have lost an undisclosed amount because his assistant consistently cheated him over a period of months.

Transport

(see under Returns, page 47).

Processing

Drying and Sorting

Farmers minimize processing because they are not generally or consistently rewarded for anything other than drying, and even this is often not accounted for in the assumptions. Drying at farm level often results in contamination with soil and plant debris and is done to varying degrees. Consolidated consignments therefore display wide degrees of variability.

Drying by cooperatives and traders is complicated by the variability from farmers and results in some beans being over-dried and others insufficiently dried and therefore susceptible to mold. This leads to the need for hand sorting, which adds to the costs at one and possibly three levels in the marketing chain before export.

Fermentation

Some fermentation has been done on a trial basis by JBC, but there was no response to samples sent to an overseas buyer and no follow-up has been made. There is no facility in the country to handle fermentation on an appreciable scale, and the dispersed pattern of producers and the difficulties imposed on transport by the state of the roads militate against an easy transition to fermentation, despite the advantages to be had from the market.

The international market price for processed cocoa can be $120 or more per ton above the New York exchange price. However, this is subject to there being a buyer or buyers and to their willingness to pay the premium on the samples sent. Since M&M Mars is already paying a premium of $90 per ton for un-fermented cocoa, it may be unnecessary to go to the lengths of establishing the system and doing the processing.

Local Manufacture

There is a market for locally manufactured cocoa powder, and it is reported that a small-scale enterprise has started the manufacture of chocolate in a remote area.

Chapter Two—Cacao 56

Costs

The costs in the processing of cocoa as it is currently done in Haiti are included under the section on Marketing, below.

Constraints and Opportunities

Postharvest-related Constraints and Impact on Profitability

Processing is limited to drying and sorting, and the impact on profitability results from the variability in the product and the fact that the fermenting process has not been done to enhance the flavor.

There are no processing facilities, and there is no concerted system for consolidation of freshly harvested cocoa for fermentation. A poorly maintained road network and the topography impose logistic difficulties in establishing a system.

Opportunities for Improvement

Most of the work currently done on the crop after harvesting would be carried out in bulk in a fermentary, and the product would have far greater consistency. Eventually, the sorting work may fall away to a negligible amount and the variable quality product would be replaced by a more consistent one of better quality and value because of enhanced flavor. A fermentary would accept wet beans on specified days so harvest could be coordinated. The price paid would be for a freshly harvested crop only, and no debate would enter into the price negotiation on the basis of humidity or moldiness.

These developments in a processing system would reveal their advantages in marketing efficiency and value added to the product. But investment would not be justified until there is positive movement in the price, at which time the prospects could be studied in detail and with regard to specific areas of production and for a specific market.

MARKETING

Players traditionally making up the market channels include voltigeurs and speculateurs who move the crop from the farmers to exporters. In recent years, the system has been challenged by SERVICOOP, which aims to generate more competition in the market to provide a better flow (60-70 percent of CIF price) toward the farmers. It appears that the aim of SERVICOOP in improving the share of international prices that reach the farmers has been successful because it intends to pay 4.25 gds. per pound for good quality cocoa at 8 percent moisture content. Despite the hiatus caused by its financing difficulties, its throughput of cocoa has been steadily increasing from 300 to 500 tons and last year to 900 tons of exports. The

Development Alternatives, Inc. (DAI) 57

structure of the company has not yet settled into a sustainable format. (See Annex II for analysis of SERVICOOP costs, and Annex IV for its sales record.)

SERVICOOP also aims to provide services to the farmers that are not provided by either the other exporters or the government. Services like input supply and extension would be contracted out. Its activities are not to be confined to cocoa and coffee trading but will extend into other export crops and processing.

None of the exporters have mechanized sorting, drying, or grading systems but rely on hand labor and sun drying.

A schematic illustration of the flow of produce is given below.

Figure 12: Illustration of Market Channels

Farmers in North Farmers in Grande Anse

Cooperatives Cooperatives

Voltigeurs Voltigeurs SERVICOOP

Speculateurs Speculateur

Exportateurs Exportateurs Novella Weiner, Primex

USA E.U. E.U. M&M Mars USA

The functioning of the market and the price changes are discussed under Product Prices, page 47. There are direct channels for movement of cocoa via cooperatives to exporters, and there are more circuitous routes via intermediaries in a traditional system where margins are made by distortions of conversion rates. This traditional system is more relevant where the growers are dispersed and a network of intermediaries is required to consolidate loads.

Improvement of infrastructure will favor the development of more direct links between farmers and exporters and hence provide a greater proportion of export earnings directly to the growers. However, there is little enthusiasm on the part of the exporters for greater

Chapter Two—Cacao 58 involvement with the grower community. Exporters are unwilling to involve themselves with the organizational structures of the growers or the development of infrastructure. These communication channels that exist between the two are also used by the coffee trade so there are economies of scale that would benefit from streamlining and consolidation.

Speculateurs claim not to know about price movements in the international market and tend to be price takers from the exporters. Speculateurs state that there is a solid cartel operating among the exporters who call one another and inform their colleagues about which speculateur is in town and what price he or she has been offered. This claim is plausible, given the small number of exporters and the financial benefits to be gained from presenting a cohesive front to the sellers.

Table 14: Summary of Marketing Chain

Cocoa Marketing Chain Assumption 22 Gds./US$

Quality Gds. US$ Cents US$ per Lb. per Lb. per Ton Voltigeur Price to Farmer 25% Moisture 1.50 6.82 150.31 equivalent at 6% 1.13 5.11 112.74 SERVICOOP Price to Farmer 6% 4.25 19.32 425.89 Harvesting costs 0.60 2.73 60.13 Farmer margin Min. 0.20 0.89 19.54 Max. 3.65 16.59 365.76 Transport from farm to mkt. 30 gds./90kg. 0.33 1.50 33.07 Speculateur Price to Voltigeur 25% 2.00 9.09 200.42 10-6% 3.00 13.64 300.63 Voltigeur margin Min. 2.48 54.61 Max. 7.02 154.82

Exporter Price to Speculateur 15% 3.50 15.91 350.73 6% 4.25 19.32 425.89

Sorting & drying 0.25 1.14 25.05 0.50 2.27 50.10 Speculateur margin Min. 0.00 0.00 0.00 Max. 2.00 9.09 200.42

CIF Price to Exporter 7.78 35.38 780.00

Transport in Haiti 20 gds./60kg. 0.33 1.50 33.07 50 gds./60kg. 0.83 3.77 83.17 Shipping 1.50 6.80 150.00 Sacks 0.17 0.75 16.60 Sorting & drying 0.25 1.14 25.05 0.50 2.27 50.10

Margin to exporter Min. 0.29 1.32 29.18 Max. 2.04 9.28 204.55

Development Alternatives, Inc. (DAI) 59

Table 15: Summary of Margins (gds. per lb.)

Min. Max. Farmer 0.89 3.65 Voltigeur 2.48 7.02 Speculateur 0.00 9.09 Exporter 1.32 9.28

Figure 13: Minimum Margins

Farmer

Voltigeur Speculateur Exporter

Figure 14: Maximum Margins

Farmer Voltigeur Speculateur Exporter

Chapter Two—Cacao 60

Domestic Market

There is a local market for cocoa, but no statistical information is available on its extent.

Export Market

Price Trends and Seasonality

The international market price is determined on the London Cocoa Terminal Market and the New York Coffee, Sugar and Cocoa Market Exchange. Typically, there is a 15-year cycle when the rising price induces fresh plantings and an expansion in production to keep pace with ever-increasing world demand.

However, two other factors impinged on the natural cycle in the latter half of the 1990s:

The strong demand from the Russian market faded away with the demise of that economy; and

The European Union modified its ruling on ISO standards applicable to manufacturing chocolate so that cocoa butter could be diluted with other products, implying reduced cocoa content.

At the same time, there was extensive planting in West Africa, most notably Côte d’Ivoire and in the Far East, which increased supply onto diminishing demand.

Hopes that emerging markets would absorb reduced demand from Russia have not been realized, and the result has been a decline in price to levels unprecedented since 1971 as illustrated by the graph below.

Development Alternatives, Inc. (DAI) 61

Figure 15: International Cocoa Markets for 30 Years

International Cocoa prices

200

180

160

140

120 annual average 100

80

60 US$ cents per pound, cents US$ 40

20

0

2 3 8 3 4 8 9 4 9 71 7 7 76 77 7 81 82 8 8 87 8 8 92 93 9 97 98 9 9 9 9 9 9 9 9 9 9 9 1 19 19 1974 1975 1 1 19 1979 1980 1 1 19 19 1985 1986 1 19 19 1990 1991 1 1 19 1995 1996 1 1 19

Source: ICCO

Figure 16: Daily Cocoa Prices—3 Years

Chapter Two—Cacao 62

Daily International Prices 3 Years

90 80 70 60 50 40 30 20 US$ cents per pound per cents US$ 10 0

l r r 7 er 8 ri ly e 9 ril ly e '9 '9 p '9 p y April July A Ju ry A Ju ar y ua nuar Octob nu Octob n Oct ob Ja Ja Ja

Market “Windows”

The cocoa market has no particular window in terms of the seasons, but there are specialized sections or niche markets.

Haitian cocoa falls within the product classification of “Sanchez,” which means that it is not fermented but simply dried and is used for making cocoa butter only. This product undergoes only the simplest form of processing between harvest and export—that of drying, sorting, and bagging. The lack of fermenting implies that the full flavor is not brought out, but the customer that has signed a large contract, M&M Mars, is not concerned and demands simply consistency in the product.

The premium paid over the New York daily price is $90 per ton, whereas Côte d’Ivoire cocoa gains a premium of $300 or more.

Prices paid in the past by ED&F Man in United Kingdom have been so low that Novella has lost money on a consignment.

Buyers in Europe may pay more for the fermented product, but the quantities involved do not warrant changing the process at present because this would require investment in facilities, creation of a new market and building up a reputation to go with it, and a consistency in supply and quality.

Development Alternatives, Inc. (DAI) 63

Constraints and Opportunities

Lack of fermenting imposes the greatest constraint to market access, but this is mitigated by the contract with M&M Mars for non-fermented beans.

Opportunities for significant market expansion would arise in the event of a significant increase in price such as to induce an increase in volumes of production.

POLICY ENVIRONMENT

No taxes are imposed on the cocoa industry, and no subsidies are applied. Government involvement in the industry is minimal, and the burden of applying what support there is has fallen to NGOs and donor agencies.

Non-revenue trade restrictions are limited to those imposed by quality considerations, which reflect back to the lack of development of handling infrastructure in the industry.

The cocoa act dates back to the 1950s and has many irrelevancies and inadequacies. There is no legally enshrined body or entity charged with responsibility for either coffee or cacao. Detailed proposals for updating the act have been made in a report by Aproma, but no progress has been made in implementation. In mitigation, there is no significant machinery for applying the regulations to be included in an updated act. As a result, little harm is done by its being out of date because the de facto implementation of logical changes has not been hampered by officialdom.

Constraints and Opportunities

Transparency and dependability within the marketing chain cannot be instilled without standardization of the weights and measures used.

Variability in the exported product is tolerated by buyers, but if variability were to be reduced, the door to price negotiation could be reopened.

It is to the benefit of the farmers that weights and measures be standardized and that the moisture content be accurately determined or standardized at completely wet or completely dry. This could be determined by the act, and moves to standardize the product could be initiated.

Chapter Two—Cacao

65

ANNEX I

CACAO: REPORT ON CLONAL GARDENS

Annex I—Cacao 66

Development Alternatives, Inc. (DAI) 67

CACAO

REPORT ON CLONAL GARDENS

(Author unknown)

“Clonal Gardens” and the Cacao Program of the 1980s: information shared by Gelson Lormeis

The following information was collected during an interview with Gelson Lormeis in the town of Grande Rivière du Nord, February 10, 2000. He was closely involved with cacao activities from about 1980 to 1988, first with the Jean-Baptiste Chavannes cooperative (JBC), and then throughout and beyond the current CYIP area.

In the early '80s, Gelson was working with both MARNDR (the Ministry of Agriculture) and MCC (a Mennonite organization involved in the health and agriculture sectors). Along with Paul Derstein and Gordon Patterson, Gelson began researching cacao. At first they began working on pruning and renewal of older cacao trees, and then they expanded into planting of new trees. In 1982, Gelson and 5 others (only one of whom is now still alive) received technical training in the Dominican Republic for cacao planting and care. Lepido Batista was one of the people who provided that training to them. When they returned to Grande Rivière, they worked on establishing sources of quality seeds. Unlike fruit trees such as mango or apple, cacao does not have well-defined varieties that produce different-tasting beans. However, numerous strains of cacao have been developed to produce more pods per tree, or to resist particular diseases. These strains can be multiplied through grafting. The preferred method of producing good cacao trees, though, is not by grafting (which requires time and skill) but by hybridizing the strains to produce high-quality seeds for planting. To this end “clonal gardens”, consisting of the named strains, are established to provide seeds for cacao nurseries. In Haiti, MARNDR had two such gardens (see the section “MARNDR Clonal Gardens”), and JBC established one as well (see the section “JBC Clonal Garden”).

When the cacao program began expanding into nurseries, it planted the JBC garden (which would begin producing three years later) and also worked on renewal of the old MARNDR gardens (which then began providing quality seeds much earlier). The first nurseries, however, required the importation of quality seeds from the Dominican Republic. The first imported seeds were a gift, but after that Dominican seeds were purchased until the MARNDR gardens began producing the necessary quantities. In the North, numerous nurseries were established in Limbé, Port Margot, out to Petit Bourg de Borgne, Robillard, Tibò, and of course Bahon and Grande Rivière itself. In the South, the same program established nurseries in Chambellan, Dame Marie, and all of the surrounding area, as far out as Les Irois. The cacao program reached out into rural sections as well as villages. In each area the program established a nursery, then actually planted the seedlings for farmers, performing the work of spacing the seedlings, providing fertilizer, and so on. In addition the farmers and cooperatives in each area were trained in cacao production techniques. We did not ask Gelson for details, but by this time the program was run by MEDA, financed through

Annex I—Cacao 68

USAID, and it continued until about 1987 or 1988 when funding was cut. It involved marketing activities and cooperative formation as well as the production aspect. Gelson mentioned that they tried fermenting cacao and then returned to selling just the regular dried beans.

MARNDR Clonal Gardens

In the 1980s, MARNDR owned at least two badly-neglected Clonal Gardens, one at Grand Prix (or Grand Pre), on the road leading toward Grande Rivière and Dondon, and one at Marfranc in the Grande Anse. The cacao program worked in both of these gardens, pruning and renewing old trees, to improve their condition. These gardens soon began providing quality seed for cacao nurseries, Grand Prix for all of the North and Marfranc for the Grande Anse.

The Grand Prix cacao garden was destroyed by the people of Souverin in a bout of dechoukaj– apparently they took the word literally. I believe this was in the early '90s, but we did not ask the date. Apparently the land is still owned by MARNDR, and is now used for citrus fruits.

Gelson believes that the Marfranc garden still exists. He suggests that it could be found by asking in Chambellan for a man whose name is pronounced “Olipsyal”, or just generally asking for a MARNDR/BAC agronomist.

JBC Clonal Garden

Following their training in the Dominican Republic in 1982, Gelson Lormeis, the others who had gone with him, and Agronome Jacques Dolcy (now deceased) began work on the JBC Clonal Garden. The cooperative had bought land, at that time a rubber plantation, from a private owner. First they removed the rubber trees, and then they worked the land with a tractor to pull up stumps. They planted plantain/banana as temporary shade, and then they planted cacao seeds in a nursery. When the seedlings were old enough to graft, Lepido Batista came from the D.R. to deliver budwood of several different strains of cacao. He did not stay long, but two other Dominicans, Manuel Ela Gusto and Frikson Chantal, did stay to graft the budwood onto the seedlings. The various strains of cacao were planted in mostly alternating rows in the garden, with cement markers at the end of each row to show their names (such as “UFC 27”). The garden was established in 1984. Gelson credits this activity with revitalizing (“reveye”, literally, “waking up”) the cooperative.

The JBC garden began producing in about 1987, but by this time the MEDA cacao project was shutting down, and Gelson says the JBC technicians at that time were not interested in continuing to work, so the garden was never used as a source of seed for nurseries. Currently, the harvest from the Clonal Garden is mainly used to provide income for JBC. However, I did notice a small cacao nursery at JBC on a previous visit, and I am told that technicians from other cooperatives in the North recently took advantage of their time at a PADF

Development Alternatives, Inc. (DAI) 69

seminar (held at JBC) to collect seed to bring back to their own area. It should be noted that the seed now collected from the Clonal Garden has not been produced under controlled conditions, and that knowledge of hybridization has been lost. Chris Stevenson is researching the properties of the JBC strains to find out how they interact. With the new cooperative leadership installed during the past year, JBC may be more interested in re-learning necessary techniques, and it seems that Gelson will be willing to work with them. His younger brother, Henry Lormeis, is the new technical director for JBC.

Annex I—Cacao

71

ANNEX II

CACAO: TRADING CASH-FLOW PROJECTIONS

Annex II—Cacao 72

Development Alternatives, Inc. (DAI) 73

CACAO

TRADING CASH-FLOW PROJECTIONS

Cocoa Production and Marketing Chain Exchange Rate Based on SERVICOOP Cash-Flow Projection 19 Gds./$

U.S.$/MT US $/Lb. Gds./Lb. International futures price 845 0.38 7.28 Price premium 90 0.04 0.78 Shipping cost 150 0.07 1.29

FOB Price 785 0.36 6.77

Gross Profit 113.96 0.05 0.98

Cost of Exportable Cocoa 671.04 0.30 5.78 Handling fee to client coops GRD 0.5 58.02 0.03 0.50 Cost of exportable cocoa 561.94 0.25 4.84 Handling/storage loss 10% 51.09 0.02 0.44

Processing Expenses 510.86 0.23 4.40 Boat and truck to Servicoop 37.20 0.02 0.32 Sacks for handling 6.32 0.00 0.05 Labor 1.35 0.00 0.01 Misc. 1.86 0.00 0.02 Raw Material from Farms 464.13 0.21 4.00 US$ pm Fixed Costs 2,055 Salaries 1,026 Per diem 263 Rent 400 Depreciation & Insurance 50 Maintenance 53 Misc. 263 Sensitivity Assumed exports MT per month 50 100 150 Operating profit U.S.$ 3,643 9,341 15,038

Farm gate price Gds./lb. 2.5 3.5 4.5 Breakeven Futures Price U.S.$/MT 543 682 821 at 100 MT per month Breakeven throughput MT/month 6.4 18.0 46.3 at US$ 845/MT Int. Futures

Annex II—Cacao

75

ANNEX III

CACAO: LABOR REQUIREMENTS IN COMMERCIAL PRODUCTION

Annex III—Cacao 76

Development Alternatives, Inc. (DAI) 77

CACAO

LABOR REQUIREMENTS IN COMMERCIAL PRODUCTION

Cacao Laour Requirement

FIRST YEAR Labor Days per Ha. Labour Days per Ha. Total Clearing Under Natural Shade Survey 5 5 Underbrush 8 11.5 Felling 4 14 Burning 0.25 Stacking & reburning 12.5 Lining 6 31 Temp. Shade Leuceana 2.5 Temp. Shade Plantain 7.5 Perm shade 8 2.5 Cacao holes 12 12 Fertiliser in holes 2 2 Mulching 4 4 Transport 9 Seedlings/liming 2.6 9 Drainage 17 17.5 Roads 33 33 Weeding 34 17.5 Herbicide 8 6 Disease & Pest control 9 9.5 Fertiliser cacao 2.5 2.5 Shade adjustment 11 Replant 4 4

Young Cacao Labor Days per Ha. Labour Days per Year Y2 Y3 Y4++ Hand weed 32 29 36 Herbicide 3.5 3.5 0.6-3 Clean circle 7.5 Fertilize 3.25 3.8 4 Prune 8.25 8 10 Resupply 2.25 Shade adjustment 9 9 14 Drains 5 5 5 Roads 5 5 5 Spraying 6.6 6 3

Annex III—Cacao

79

ANNEX IV

SERVICOOP SALES RECORD

Annex IV—Cacao 80

Development Alternatives, Inc. (DAI) 81

CACAO

SERVICOOP SALES RECORD

DateQuantity FOB Future's Invoice Value Container Monthly Annually Price Price Container Annually (mt) (mt) (mt) ($/mt) ($/mt) ($) ($)

04-Dec-97 20.00 1,519 30,371.60 11-Dec-97 20.00 40.00 1,575 31,494.46 16-Jan-98 20.00 1,449 28,989.46 27-Jan-98 20.00 40.00 1,493 29,859.60 27-Feb-98 20.00 20.00 1,480 29,600.00 11-Mar-98 20.00 20.00 1,480 29,590.66 08-Apr-98 20.00 1,475 29,500.00 20-Apr-98 20.00 40.00 1,475 29,500.00 02-May-98 20.00 1,570 31,402.68 22-May-98 20.00 40.00 1,568 31,362.60 15-Jun-98 20.00 20.00 1,493 29,859.60 10-Jul-98 20.00 1,513 30,260.40 27-Jul-98 19.70 39.70 1,524 30,022.80 07-Aug-98 20.05 1,466 29,397.70 Aug-98 20.04 40.09 1,497 30,000.00 First 12 Months 299.79 451,212

14-Sep-98 12.00 12.00 1,527 18,324.00 05-Oct-98 13.20 1,462 19,298.40 20-Oct-98 13.20 26.40 1,477 19,496.40 26-Nov-98 12.72 12.72 1,420 18,062.40 07-Dec-98 12.72 12.72 1,387 17,642.64 26-Jan-99 13.20 13.20 1,275 16,830.00 23-Feb-99 13.20 13.20 1,256 16,579.20 03-Mar-99 13.20 1,215 16,038.00 10-Mar-99 13.20 1,187 15,668.40 15-Mar-99 13.20 1,173 15,483.60 24-Mar-99 13.20 1,138 15,021.60 29-Mar-99 13.20 66.00 1,138 15,021.60 06-Apr-99 13.20 1,159 15,298.80 15-Apr-99 13.20 1,111 14,665.20 19-Apr-99 13.20 1,111 14,665.20 27-Apr-99 13.20 960 12,672.00 27-Apr-99 12.90 65.70 960 12,384.00 05-May-99 21.00 950 19,950.00 12-May-99 13.20 930 12,276.00 19-May-99 13.20 941 12,421.20 25-May-99 13.20 925 12,210.00 25-May-99 13.20 73.80 925 12,210.00 02-Jun-99 13.20 862 11,378.40 06-Jun-99 20.52 965 19,801.80 15-Jun-99 13.20 1,020 13,464.00 22-Jun-99 21.54 1,063 22,897.02 27-Jun-99 14.04 82.50 1,045 14,671.80 05-Jul-99 14.04 1,018 14,292.72 Jul-99 14.04 980 13,759.44 Jul-99 14.04 945 13,269.12 Jul-99 14.04 56.16 989 13,882.08 Aug-99 12.84 1,046 13,430.64 Aug-99 14.04 929 13,042.80 Aug-99 23.04 920 21,196.80 Aug-99 14.04 900 12,636.00 Aug-99 13.98 77.94 882 12,330.06 Second 12 Months 512.34 552,271

Annex IV—Cacao 82

Sep-99 20.04 918 18,392.94 Sep-99 20.04 937 18,771.20 Sep-99 20.04 937 18,771.00 Sep-99 14.04 937 13,160.50 Sep-99 20.04 935 965 18,737.40 Sep-99 14.04 108.24 935 965 13,127.40 Oct-99 14.04 935 965 13,127.40 Oct-99 20.04 935 965 18,737.40 Oct-99 20.04 935 965 18,737.40 Oct-99 14.38 935 965 13,445.30 Oct-99 25.66 992 1,022 25,454.72 Oct-99 20.04 930 960 18,637.20 Oct-99 20.04 933 963 18,697.32 Oct-99 20.04 933 963 18,697.32 Oct-99 20.04 933 963 18,697.32 Oct-99 20.04 933 963 18,697.32 Oct-99 14.14 208.50 933 963 13,192.62 Nov-99 13.94 825 11,500.50 Nov-99 26.06 848 22,098.88 Nov-99 12.94 803 10,390.82 Nov-99 7.06 803 5,669.18 Nov-99 11.22 770 8,639.40 Nov-99 20.00 91.22 800 16,000.00

Dec-99 14.04 763 823 10,712.52 Dec-99 20.04 776 836 15,541.02 Dec-99 20.04 54.12 753 813 15,090.12 1999 Sales in Metric Tons 910.58

Jan-00 14.04 809 869 11,358.36 Jan-00 14.04 28.08 790 850 11,091.60 Feb-00 12.84 12.84 721 781 9,257.64 Mar-00 12.24 738 798 9,033.12 Mar-00 14.04 26.28 762 822 10,698.48 Apr-00 14.04 743 803 10,431.72 Apr-00 21.00 760 820 15,960.00 Apr-00 14.04 704 764 9,884.16 Apr-00 13.20 62.28 704 764 9,292.80 May-00 21.00 735 795 15,435.00 09-May 42.00 735 795 30,870.00 23-May 21.00 785 845 16,485.00 30-May 21.00 105.00 785 850 16,485.00 06-Jun 14.04 780 845 10,951.20 12-Jun 14.04 780 845 10,951.20 Jun-00 14.04 42.12 790 855 11,091.60 622,001

Development Alternatives, Inc. (DAI) 83

CHAPTER THREE MANGOS

SUMMARY CONCLUSIONS

Two ingredients are essential before the following recommendations can be implemented. The first ingredient is already in place—the organization of active grower groups. The consultant observed the propagation of nursery plants for distribution to grower cooperatives in Jacmel and Casal, for example, and lectured to 30 grower cooperatives representatives, representing 500 landowners, in Gros Morn.

The second essential ingredient has begun, which is more direct grower-exporter interaction. More than $1 million have been injected into the same region from mango sales as a result of grower-exporter interaction, and this must be continued and expanded.

Following are suggestions that are economically attractive so that producers and exporters should gladly respond.

The recommendations are broken into three phases. Innovative change is always slow to catch on, but given the fairly dramatic returns that should be realized from each phase, widespread replication is almost ensured.

Phase I: Organizational

Utilize Exporters as Trainers and Partners

HAP should work with the exporters to convince them that increased sales begin with improved direct communications and more exporter investment at farmer association level, including the provision of extension services to the growers and the establishment of direct payment procedures. The project should facilitate sustainable partnership through cost- sharing mechanisms and advisory field support.

Educate the Growers on Fungicide Applications

The first action to be done this flowering season should be to educate the growers on the benefits of fungicidal sprays. Exporters should be given the opportunity to start this process by purchasing backpack sprayers and fungicides for several grower cooperatives in areas of the Artibonite District because these areas are most susceptible to anthracnose. This should be done in a way that avoids conflicts among exporters, and the project might consider coordinating the initial extension phase with them.

Chapter Three—Mangos 84

The exporters should send technicians to demonstrate proper spray techniques and schedules by spraying the lower canopies of trees during flowering in areas that are affected by rain during the flowering season. The first spray should be applied to expanding panicles just when the flowers first begin to open. The first treatment should be a systemic fungicide, such as either Benlate (E.I. DuPont de Nemours & Co.) or Abound (also named Bankit, manufactured by Zeneca). The second spray should consist of a contact fungicide, such as Daconyl or Bravo (Zeneca), applied two weeks after the first. The contact fungicide not only provides additional protection, but it also prevents proliferation of systemic fungicide- resistant fungi. Two weeks later, repeat the systemic product followed again two weeks later with the contact fungicide. By this time, the young fruit will be completely protected from fungal infections. Cocide, a copper-based fungicide, should be sprayed with a sticker on the fruit when they are about half size, especially if rainy conditions persist. Growers will see the immediate benefit of the sprays by comparing the number of fruit hanging from the treated branches versus the untreated sections of the upper canopy. The yield in the treated branches can be as great as 10-fold higher than untreated sections in severely affected anthracnose areas.

Arrange for Direct Exporter Payments to Growers Delaying Harvesting

At the same time as the introduction of fungicide applications, the exporters should convince growers they can reap the most benefits from increased production by waiting for the fruit to mature before harvesting for sale directly to the exporters. By doing so, farmers will not only realize increased harvest but will also receive up to 25 gds. per dozen (13) with the increased yield, resulting in significantly increased income over that received from fourniseurs.

Provide Harvest and Postharvest Support

Provide instruction in the proper use of picking poles with catch bags and in the use of proper field boxes on donkeys, to ensure protection of the increasingly valuable crop. Participating growers will see an immediate financial reward in the first year. Obviously, the exporter's motivation in this process is increased pack out and better quality fruit for increased market prices and returns. Moreover, the exporter builds producer loyalty and opens lines of communication for educating the producers about the rewards of producing export quality fruit and how to achieve it. Use of transit bins when transferring fruit to the packinghouses will ensure good quality fruit for export.

Systematize the Spraying and Investment Process

Most growers will buy into the system once they witness the immediate economic benefits of protective fungicidal sprays, protection of the crop from physical damage, and selling directly to exporters. In subsequent seasons, therefore, members of grower cooperatives will share in the purchase of sprayers and fungicides for their mango trees. The exporters would

Development Alternatives, Inc. (DAI) 85

continue to invest with cost-sharing arrangement—perhaps selling the backpack sprayers, for example, at half-cost to participating cooperatives.

Continued education by exporters will also be essential to ensure the sprays are properly applied, fruit are harvested at maturity, and they are protected. In the meantime, there should be interest in increasing the number of productive trees on the communal plantations.

Phase 2: Expand and Create Orchards

Plant New Orchard Trees

Each cooperative should plant a sufficient number of Francique seeds to satisfy the needs of the group. A price of one Haitian gourde is deemed reasonable by cooperatives that have already begun propagating nursery trees. Ideally, all seeds should come from healthy, productive Francique type trees. The husks of each seed should be removed to expose the white embryos. The single, bilaterally symmetric sexual embryo, located at one end of the seed, must be separated from the asexual embryos in each seed (see diagram at left).

Only the asexual embryos should be planted in two-liter black plastic nursery bags with drain holes, thus propagating only Francique seedlings that will be genetically identical to the mother trees from which they came. Seedlings should be transferred to the field at the beginning of the next rainy season to facilitate rapid growth. This consultant recommends that orchard trees be planted on a 10 by 4 meter spacing while including the pre-existing large trees, which will eventually be cut in Phase 3, in the overall orchard layout.

mango

breadfruit Typical Cooperative Plantation in

coconut palm Phase 2

seedling orchard mango

Chapter Three—Mangos 86

Maintain Income through Existing Trees and Other Crops

Growers should continue to spray and harvest the large trees to maintain income for three years while the orchard trees are growing. Other cash crops, especially those desired for the export market—such as pumpkin, chayote, yams, beans, malanga, cassava, pigeon peas, corn, snow peas, papaya, and cherry tomatoes or any of numerous other crops—should be interplanted between the rows of mango trees to sustain income throughout the year. Hillside farms require special considerations with regard to interplanted crops, as discussed later. Continuous vigilance must be exercised to prevent goats or other animals from grazing on the mango trees among vegetable crops.

Prune New Orchard Trees

Once the orchard trees begin to grow, they should be tip pruned regularly (about every two months) to encourage rapid, branching growth. For example, make the first tip prune when the plants are 60-70 centimeters high. Remove about 3 centimeters from the tip. The lateral buds will soon spout lateral stems. Tip prune those stems as soon as the new leaves become hardened. Repeat this process for each stem until the trees stand about 2 meters high (see diagram at right). A sharp machete is the best tool for this purpose.

The last tip prune should be done about six months prior to the normal first flowering in the area. A 4-percent potassium nitrate solution can be sprayed on the trees six months after the last prune to stimulate synchronous flowering throughout the trees. Thoroughly wet the undersides of all the leaves using backpack sprayers. Repeat spray if flowering buds do not develop within two weeks. The buds should flower before the need to repeat the spray for the third time.

Phase 3: Stabilize Mango Production and Hillsides

Cut Back Pre-Existing Trees

Fruit production in the first year of flowering of these orchard trees will likely surpass that of the one or two large pre-existing mango trees on each property within the new plantation orchard. Each producer should cut the large trees to a height of 1 meter and sell the wood. Cut trees that are not Francique should be top worked (grafted) to Francique and repeatedly tip pruned as described above until they are the same size as the other orchard trees before stopping the frequent tip pruning. If the trees are already a Francique type, they should be tip pruned frequently as described above. They too will come back in production about two

Development Alternatives, Inc. (DAI) 87

years after they are cut. Breadfruit trees should also be cut to about 2 meters to provide better light for the mango trees. They will come back into production at a lower level as will the previously large mango trees.

Systematize Pruning of All Mango Trees

The orchard trees should be tip pruned each year about six months prior to the desired flowering time. This pruning event resynchronizes growth and flowering each year, allowing for better anthracnose control with the fungicides. Once the trees reach a height of 4 meters, they should be pruned to the same height and width each year by making the prune cuts a bit deeper than the previously used tip prune, which allowed the trees grow to the desired size. In this way, the grower will always have room to raise the other crops planted between the rows.

Rehydrating and Stabilizing the Hillsides

Hillside mango orchards in the above-described program will require a mechanism to retain sufficient water in the mountain bedrock to sustain the trees throughout the year. Mangos naturally grow in ravines on hillsides because seepage water from periodic rains tends to collect there. Trapping water to facilitate recharge can be facilitated by installation of several check dams up these ravines. Many such small check dams have been constructed in Haiti where deforestation has removed the ability to retain rainwater.

Lateral ditches extending around the perimeter of hillsides from check dams could help distribute periodic rainwater to sections of the hills that do not normally retain water efficiently (see diagram below). Such a series of structures should be useful on slopes with inclines of up to 45 percent.

A cooperative-producer group owning land up such a hillside could organize itself to purchase cement and steel rod and, with the help of an engineering advisor, build the bottom check dam structure during the dry season. The mango trees, leguminous ground cover, and other crops should be ready to transplant onto the hillside by the beginning of the rainy season. It is critically important that this be done to restabilize the hillside soil to prevent erosion. A perennial, forage legume, ground cover should be planted all over the hillside for this purpose. Other commercial crops can include anything except root crops. These should be avoided to minimize soil disturbance during harvest. The process should be repeated at the

Chapter Three—Mangos 88

next level up in the following year. Over time, the lateral ditches, the mango trees, and cover crops will rehydrate and stabilize the entire hillside.

Produce Mangos Year-Round

Once an overall system is in place wherein high production and exportation of Francique- type mangos are being consistently achieved, it will be necessary to select and develop additional high-quality cultivars for export. Producers can begin to modify the timing of flowering as they learn how to manage their orchards so Haiti can produce mangos in high quantity and quality every month of the year.

Development Alternatives, Inc. (DAI) 89

Develop a Molded Plastics Industry

Field boxes should be used instead of woven baskets to transport mangos, especially on the backs of donkeys, to prevent physical damage to the fruit when transported from hillsides. It is recommended that a mechanism be made to provide assistance so that boxes can be manufactured in Haiti for this purpose. Molded plastic boxes utilizing recycled plastic would be ideal way to provide a needed product, create jobs, and help clean the island in one operation. An exporter has already expressed interest in building such a manufacturing plant. Each grower could purchase the boxes not only for mangos but also for all their crops.

BACKGROUND

The current status of Haitian mango production is primitive. Unlike other areas of mango production where the growers develop a packinghouse to sell the fruit, the exporters in Haiti have simply tapped into a natural, pre-existing resource of trees growing throughout the region. The notion that producers in Haiti are simply tapping this natural resource that happens to be on their land has prevailed among producers, fourniseurs, and exporters. There has been no economic incentive to change this attitude because the poor economy and infrastructure would not permit increased income for producers in the domestic market, and exporters could obtain sufficient fruit for a sizable harvest simply by using established chains of commerce to purchase fruit from fourniseurs. NGOs have provided opportunities for improvement, but the producers were unable to connect these improvements with increased income.

The goal of HAP with regard to the mango sector should be to increase the flow of exportable fruit by adapting measures at farm and postharvest levels. If growers begin to take charge of their trees, the benefits of increased productivity will be immediate. Increased foreign exchange will mutually benefit producers and exporters and thus also allow HAP to tap into the exporters for more investment and technical support of the small farmers.

PRODUCTION

Producers

Mango producers in Haiti are land owners on whose property one to several mango trees exist, usually by chance. Few in Haiti have sufficient land to plant more than a few trees, especially considering the large size of over 15 meters to which they typically allow them to grow (see photo above). The number of producers is unknown and difficult to characterize because they are typically opportunistic, taking advantage of the fact that a volunteer tree exists on the property.

Chapter Three—Mangos 90

A producer may have a primary income other than farming and use the plants grown on the property for subsistence. If they are farmers, they use the trees to augment income from other more economically viable crops, such as bananas, vegetables, and agronomic crops (see photo to left). Mangos have been considered to be of such low value as a crop that many producers have, until recently, cut large trees to sell the wood for use in making lumber and/or charcoal. One large tree can be sold for up to 100 gds. Such decisions are also driven by the fact that cutting a tree brings immediate income at times when money is most needed for family survival, whereas waiting for a tree to produce an unreliable crop may be too little and too late to get the family through the year.

Location

Mango trees are found in virtually every area where there is adequate water for them to germinate and grow without cultivation. This is usually in areas highly populated by people as they discard seeds after consumption. Annex I provides a map that indicates locations where mangos grow and are sought for harvests destined for export and domestic sales. About 50 percent of the commercial export production comes from the Gros Morne region, located in the northeast area of the Artibonite District; 25 percent comes from the other areas in the Artibonite District; and the remaining 25 percent of the annual exported crop comes from the rest of the indicted areas of the map. The districts located in the south generally have ample water as evidenced by lush green vegetation and mesophytic type trees that are typical of climates with ample rainfall.

The Centre, Artibonite, and Nord-Ouest districts are generally dry, as evidenced by the presence of xerophytic vegetation, but numerous rivers and streams drain the mountain water into the Golfe de la Gonave. The commercial mango production areas usually thrive in valleys and lower hillsides along these waterways. During this consultant’s visit, the mountains in Gros Morne exhibited evidence of more-lasting rainfall, as evidenced by extensive green vegetation on the hillsides, but it may have been a result of growth that occurs only during the rainy season. Although the Nord District was not toured during this assignment, the presumption is that it is a wetter area because anthracnose, a disease of mango that thrives in humid conditions, is most prevalent in that region.

Development Alternatives, Inc. (DAI) 91

Although the northern area around Cap Haitien produces Francique-type mangos, they suffer excessive anthracnose and fruit fly damage. Additionally, travel distances across poor-quality roads to packinghouses in Port-au-Prince make it difficult to transport mangos for export.

Farm Size

Few producers own a parcel of land occupying more than a fraction of a hectare. Although this consultant did view four holdings of 5-15 hectares in area, most of the producers own far less than 1 hectare because land has been split among family members with the passing of inheritance over the years. Except for four larger holdings observed, none of the producers has trees planted in an organized way. The mango trees are typically very large, growing among tall breadfruit trees, coconut palm, or other tall trees used in production of wood in moist areas. Intermediate-sized under-story plants may consist of banana, cassava, corn, or pigeon pea. Malanga, beans, chayote, pumpkin, or yams usually occupy the balance of the property. Goats, chickens, and occasionally cattle that are allowed to graze as much as possible on available land provide meat protein. Only crops that are deemed of value to a producer are protected from grazing animals. Traditionally, mango has not been a protected crop.

Land Types

From 50 to 60 percent of commercial mango production is on hillsides. This can range from slight slopes on the edges of the hills to about 45 percent slope on hillsides. The trees are usually concentrated in or near ravines. The rest are on alluvial planes and in river valleys (see photo to right). Gros Morne and the other mango production areas in the Artibonite District, noted by “H” on the map in Annex I, are sites where nearly all of the hillside agriculture in Haiti occurs.

Hillside volunteer mango trees appear to originate from discarded seeds near the tops of hills that spread downhill in a fan-shaped pattern in ravines where rainwater can best seep into the soil. Animals usually graze these areas, but some seasonal vegetables like corn and yams may be planted on the hillsides where the soil can retain enough water to sustain the plants to harvest.

The soils observed in most of the mango-growing areas appear to be fertile, with excellent, loose texture to facilitate unrestricted root growth and health. Nutritional health with regard

Chapter Three—Mangos 92

to nitrogen, phosphorus, and potassium is good, as evidenced by the density and dark green color of the mango tree canopies. Minor element deficiency symptoms in the leaves are rare. Nitrogen and potassium deficiency symptoms are apparent in banana trees, which require higher levels of these minerals than does mango. Information on soil pH was impossible to maintain, but would appear to vary from slightly acidic in highly organic top oils and more basic in soils closely associated with limestone bedrock on the hillsides. This assessment is by observation of tree foliage only, in lieu of any leaf or soil analyses.

Productive Infrastructure

There did not appear to be any area that utilized wells or any form of irrigation unless it was provided by aquaducts constructed by NGOs and supplied by local rivers. The two aquaduct systems observed in Jacmel and Casales were both functioning and were being utilized by farmers. How much of this system is being utilized for the benefit of mango production is difficult to surmise, but this consultant assumes that the irrigation systems are used primarily for vegetable production, with interplanted mango trees receiving some benefit. Producers simply do not have sufficient money to invest in any sort of sophisticated mechanical equipment such as pumps. All work is accomplished with hand tools, usually machete.

PADF has been responsible for construction of check dams in ravines on hillsides to help retain water to augment groundwater supplies and improve seepage along the hillsides. This is an important step in developing agricultural sustainability, especially in the drier areas.

Inputs and Technology

Varieties

Technically, there is no “variety” or “cultivar” grown in Haiti. Identified as a superior Haitian mango in the 18th century, the variously named, ‘Francis,’ ‘Madame Francis,’ or ‘Francique’ is properly termed a “Francique-like mango type”. This is because one out of several seedlings that emerge from each of the polyembryonic seeds of this Southeast Asian type of mango is derived from a “sexual” embryo that is a result of a sexual cross during pollination either by selfing in the same tree or from an adjacent tree that may or may not have the characteristics of Francique. These give rise to a tree that has different characteristics from those of the mother trees. This difference may be great or small, depending on the pollen source and the amount of genetic diversity it provides to the resulting seedling tree. The other “asexual” embryos in the same seeds produce plants that are genetically identical to the mother trees from which they were borne and, thus, propagate many trees identical to the mother tree. Hence, the germplasm of the originally named Francique tree has been diluted for nearly 300 years as a result of germination of sexual crosses- and self-pollinations of the offspring. These seedlings have formed a population of millions of trees, some of which have sufficient Francique characteristics so as to look and

Development Alternatives, Inc. (DAI) 93

taste like Francique but also possess altered characteristics that may affect production performance, disease resistance, or storage characteristics.

Numerous other types, such as ‘Bautist’ and ‘Corn,’ have been identified with the same caveats in their uniqueness as that described for Francique. The types display different shapes, sweetness, and flavor qualities that make them identifiable by Haitians, but they are simply groups of types with similar characteristics.

Regardless of this technical point, Francique was identified as and continues to be the favored mango type for export to the United States from Haiti. It has good-eating quality when harvested mature and properly ripened. It withstands the heat of quarantine treatment and ships well. The type is available throughout Haiti because of its favored status among the population.

Seed Supplies

Mango seeds do not store well, so seed supplies are non-existent. Historically, seeds have been discarded after being eaten and the successful volunteer seedlings allowed to grow. More recently, thousands of mango seeds have been gathered for planting in small 2-liter plastic bags by producer organizations for grafting on Francique-type scions and planting in organized orchards.

Agro-Chemical Use

Fertilizers are not used in Haiti on any crop, especially mango. Apparently, some people apply manure to a degree but it is rare. Pesticides are not used on mango for two reasons. First, producers are not aware of the need for pesticides on mango or how pesticides can benefit mango. Second, if producers were aware of the benefits, they could not afford pesticides with the current pay structure for mangos.

Cultural Practices

Haitian producers are ignorant of technologies that are universally used to produce mangos throughout the world. Their perspective, mostly as a result of the low income derived from fruit harvests, is that the tree can provide a minimal supplemental income, which augments that derived from the primary vegetable crops.

Agricultural Credit

This consultant came across none.

Chapter Three—Mangos 94

Labor

Producers usually harvest the fruit, although there are a few with sufficient numbers of trees to require assistance by local laborers. Otherwise, the entire crop on each tree is sold for a nominal price and the fourniseurs take the responsibility of harvesting the fruit.

Returns

Product Prices

The price received for mangos by producers varies greatly as determined by a complicated variety of factors negotiated between fourniseur and producer. Price varies most according to area. For example, fruit harvested in Gros Morne reportedly returns only 5-8 gds. per dozen, whereas fruit from the Cul de Sac near Port-au-Prince may get 7-10 gds. per dozen. Fruit harvested in Jacmel, Leogane, or Cabaret can get 12-17 gds. per dozen. This variation in price likely stems from the amount of anthracnose infection visible on the fruit and perhaps distance to packinghouses. A dozen mangos is a negotiated variable in itself, ranging from 13 to 18 or more fruit making up a dozen, depending mostly on size at harvest but also on fruit quality (anthracnose).

Seasonality and Variability

Production seasons can be divided into three basic harvest groups determined by the date fruit begins to reach horticultural maturity. Early-season harvest extends from November through January. Mid-season harvests occur from June through August, and late-season harvests range from June to September. Some areas have one major production season, whereas other areas may have two peaks of production, usually producing early- and late- season fruits (Annex I). This variation is caused by asynchronous growth patterns within major sections of individual tree canopies, resulting in different flowering times in those sections. Rainfall patterns in each area of the country contribute to this variation. In general, the early season fruit comes first from Leogane. Production proceeds northeastward and southwestward to Cabaret and to Jacmel and Les Cayes, respectively. The late-season fruit comes from the Artibonite and Nord area districts.

Based on volume, the two major seasons of export to the United States were November 1999 to September 2000, peaking in May-June (~600,000 boxes a month) and August (~550,00 boxes), respectively. The overall export production in 2000 was more than 2.65 million boxes.

Development Alternatives, Inc. (DAI) 95

Figure 17: Recent Haitian Mango Production

700,000

600,000 BO XE 500,000 S EX PO 400,000 RT ED TO 300,000 US A 200,000

100,000

0 Nov. 99 Dec. 99 Jan. 00 Feb. 00 Mar. 00 Apr. 00 May. 00 Jun. 00 Jul. 00 Aug. 00 Sep. 00 Oct. 00 MONTH

Yields

Individual tree yields are impossible to assess because no records are kept. Rejection can be as high as 90 percent because of anthracnose infection, fruit damage when striking the ground from high in the trees, transport damage, and lack of sufficient fruit maturity. These factors, along with the great variety on tree sizes and heights and the lack of synchrony of production within trees, make estimation of average annual yields per tree impossible. Based on the known export production of about 2.6 million boxes, a yield of 50 boxes per tree, and an actual export of perhaps 10 percent of the existing mangos, there is an estimated minimum of 520,000 productive Francique trees and about 260,000 producers in Haiti, if one assumes an average of 2 trees per producer. These numbers should be considered a gross estimate of the minimum at best. Others have estimated the numbers to be in the millions, but the current estimate considers only the Francique type of mango.

Production Costs

There are no production costs because producers provide no inputs to improve or maintain tree production.

Net Returns to Farmer

There are generally no costs to produce mangos, so the net returns are the same as gross returns.

Chapter Three—Mangos 96

Constraints and Opportunities

Production-related Constraints and Impact on Profitability

The primary constraints to high yields and profitability are ignorance of improved horticultural methods and the inability to purchase the most rudimentary equipment to facilitate improvement of tree production.

Opportunities to Increase Profitability and Reduce Variability of Returns

Recent trends in forming producer and exporter organizations has begun a process that must be accomplished to put growers in direct contact with exporters. Producer cooperatives or organizations facilitate sharing of information among growers. More important, such organization of land cooperative owners creates a plantation atmosphere of production with each landowner protecting and managing his or her portion. Costs for supplies and use of equipment can be shared among the members of each group to economize on the costs needed to improve yields, quality, and income. Working with exporters, the farmers can begin to know what is required to receive maximum returns on their mangos by dealing directly with exporters in the sale of fruit. A many-fold increase in income from mangos can be realized in the first season of cooperation just by being paid directly by the exporters for their fruit. Improvements in disease control, harvest, and transport techniques can further increase annual income from mango sales by as much as 10-fold or more. Moreover, adoption of improved planting skills and tree maintenance can facilitate even greater returns from mangos alone. (See Summary Conclusions at the beginning of this chapter for guidance to achieve these improvements.)

POSTHARVEST

Storage

Facilities—On-Farm, Off-Farm

There are no storage facilities. Even refrigerated storage of freshly harvested fruit in the field with the technologies available today would not be economically or operationally feasible, considering that the fruit would have to warmed to ambient temperate before being exposed to heat treatment to satisfy quarantine regulations.

Development Alternatives, Inc. (DAI) 97

Pest and Disease Problems

Caribbean fruit flies—Anastrepha striata and A. obliqua—are the primary insect pests of mango in Haiti. Although mango is not a preferred host of these species, they do infest the fruit to some degree, especially in the northern districts of Haiti. Control is achieved by use of baits and by harvesting fruit before fruit softening begins. The APHIS hot water quarantine treatment ensures virtually 100 percent kill of any larvae that may be in fruit reaching the packinghouses.

Anthracnose, caused by the fungus Colletotrichum gloeosporioides, has the greatest impact on mango yield and fruit quality. Anthracnose is always associated with rain events and with high humidity conditions sufficient to cause night dew formation. Inflorescences are the most susceptible organ to the disease, causing them to turn black, resulting in virtually 100 percent loss of all open flowers following a rain event. This loss of flowers drastically reduces yield. Small fruit exposed to the same conditions become infected and will display black spots or streaks early or late in fruit development. Even mildly infected fruit are unacceptable to exporters because of serious decay problems occurring in shipping boxes during transit.

Losses

Losses from anthracnose infection affect producers by decreased salable fruit and exporters by latent infections ruining a portion or perhaps the entire shipment when inspected at the port of entry. Exporters who reject fruit displaying anthracnose symptoms minimize the extent of these losses. Rates of rejection at the packinghouse door are generally 30 percent but can range greatly depending upon the care that fourniseurs have used in selecting fruit from producers and how they were transported.

Losses caused by damage from physical impact occur when producers shake limbs in the upper canopy of large trees and allow the fruit to fall to the ground. Additional losses occur when producers or fourniseurs load fruit in woven straw baskets mounted on the backs of donkeys for transport down hillsides. The movement of the animals, coupled with the rough baskets used for mangos, causes unacceptable damage to the skin of the fruit. Additional damage occurs when fruit is transported long distances over rough roads without the protection provided by transport bins. The fruit usually end up on the bottom of pickup truck beds overlaid with other heavy cargo and people.

Chapter Three—Mangos 98

Additional losses may occur as a result of the hot water treatment of immature fruit, which results in collapse of the pulp in the shoulder of the fruit and by jelly seed (soft nose), a disorder usually attributed to overly mature fruit. Although impossible to put an accurate figure on overall losses, they are substantial when one considers the transport chain from tree to shipping container.

Storage Costs

There are no storage costs because the fruit is not stored for any purpose.

Transport

Participants

The producers usually transport fruit from picking areas to a collection point, but fourniseurs may participate in this when entire trees are sold. Fourniseurs may buy fruit and sell directly to an exporter or to a street vendor in the local market. Most of the time, fruit is resold to another fourniseur in a pyramidal chain of organization involving up to five transfers before being sold to exporters. Once the fruit reaches the packinghouses, it is selected, washed, and hot water treated followed by temperature equilibration before being cooled in cold rooms in preparation for loading into refrigerated containers. The intermodal containers are loaded onto a ship and delivered to Miami or New York. Canadian-bound fruit are not treated with hot water.

Vehicles

Producers rarely have a vehicle other than a donkey to transport fruit. Fourniseurs may load fruit into any type of available vehicle that will reach the next point of sale. Little care is taken to protect fruit in these transits. Some exporters now pick up fruit directly from producers in organized areas and transport it to the packinghouse in large bins transported by heavy-duty trucks. Greater care is thus taken to reduce losses in transit with a greater reward for those exporters who expend the time and money to do it.

Cost Estimates

Costs vary greatly for each level of transportation, making it difficult to obtain cost quotes.

Development Alternatives, Inc. (DAI) 99

Processing

Participants

Only one exporter, Nancy Frombrun, expressed interest in processing cull mangos for production of juice and puree.

Facilities

Ms. Frombrun has built a factory to receive and process the fruit and is encouraging other exporters to sell rejects to her at a discounted price.

Location

Ms. Fronbrun’s factory is in Port-au-Prince.

Constraints and Opportunities

Postharvest-related Constraints and Impact on Profitability

All of the above-named postharvest factors that contribute to fruit losses negatively impact profitability for producers, fourniseurs, and exporters.

Opportunities to Increase Profitability and Reduce Variability of Returns

Education at every level of postharvest handling—starting with use of proper picking poles equipped with cutting blades and catch bags by producers and ending with use of protective transport bins by fourniseurs and exporters—will immediately reduce fruit losses and profitability at all levels. Direct interaction with producers and producer organizations will greatly facilitate the changes necessary to induce greater profits. It is feasible that profits could more than double simply by protecting the fruit from damage during harvest transport.

A revolutionary new shipping container is poised to enter the produce transport market. It is the VacuFresh system invented by Dr. Stanley Burg and manufactured by Welfit Oddy in South Africa. The system is described on its web site http://www.oddy.co.za/vacufresh/vacufres.asp. The intermodal container provides a hypobaric atmosphere that can store mangos for up to two months with no deterioration of quality even of fully mature fruit. It may also provide an alternative quarantine treatment, replacing the hot water bath. Vastly expanded market areas will thus become available, such as oriental markets in the northwest United States and in

Chapter Three—Mangos 100

Europe using this container. (Other shipping container systems, such as controlled or modified atmosphere, do not work well for mangos.)

MARKETING

Domestic Market (including Dominican Republic)

Prices in Alternate Markets

Although it was difficult to obtain specific prices, the estimate of prices in alternate markets is 2-4 gds. per fruit based on supply and demand.

Participants

Farmers sell fruit on highways and in towns located near the production area.

Fourniseurs (Madame Saras, farm buyers/sellers) purchase fruit from the producers and likely transport it into larger towns and villages far removed from production areas to avoid the competition by local producers.

Quality Requirements

Consumers in the local markets will accept any level of quality. These are hungry people, and they know that a black-streaked mango usually still tastes good.

Location

Street venders are located on sidewalks and on the side of the highways carrying traffic through towns and villages; the vendors can be found anywhere people frequent.

Product Form

Mango fruit of all types is displayed fresh in large bowls or other convenient apparatus.

Development Alternatives, Inc. (DAI) 101

Product Use

Mango seems to be eaten mostly out of hand by consumers walking the streets, although it is also purchased to restock home supplies.

Growth Potential

If mangos were available every day of the year, the sale of local fruit would increase. Continuous availability would not likely increase the sale price, however.

Export Market

Participants

Exporters make up the export market.

Current Markets

“Red,” Indian-type mangos consisting of the cultivars Tommy Atkins, Haden, Keitt, Irwin, and Sensation are imported from most of the tropical belt countries in the Western Hemisphere, ranging from Mexico to Peru and Brazil and from Puerto Rico. The most popular imported cultivar is Tommy Atkins. Consumer markets apparently distinguish these various red mango cultivars from the Francique, which has yellow skin and a longer shape to it. Interestingly, market prices of the red mangos appear to change independently of prices of the Haitian mango. For example, June and July are usually the months of lowest returns for red mangos because of the high volume imported from Mexico. Haitian mangos still commanded a respectable return of $6-8 per box over the same period. The primary consumers of Francique are Haitian and Asian immigrants and a mixture of other ethnic groups distributed along the eastern seaboard of the United States and Canada where Francique is imported.

Potential Markets

Only 10 percent of the people in the United States know about and are familiar with the taste of mango, and much fewer are familiar with Francique. The potential therefore for expanded U.S. markets is tremendous through advertising programs to promote the high quality of this Haitian fruit.

Chapter Three—Mangos 102

Product Form

The product is sold fresh in units of one fruit.

Product Use

The fruit is eaten fresh.

Price Trends and Seasonality

According to two exporters, early-season mangos, shipped from November through January, receive $4-9 per box, mid-season mangos (March through May) receive $4-6.50 per box, and late-season fruit (June through September) receive $6-8 per box. The lower, $4 value is usually the result of shipping immature fruit. The first container shipment of mature fruit that arrived in Miami in 2000 from Sunshine Packinghouse in the last week of October received $12 per box.

Presence/Absence of Market “Windows”

The consistently high returns enjoyed by exporters of Francique suggest that the marketing window for this type of mango is less critical than the February-March window for red-type mangos.

Constraints and Opportunities

Opportunities to Penetrate/Expand Markets, Increase Profitability, and Reduce Variability of Returns

Haitian mangos should be promoted through aggressive advertising campaigns and marketed by Haitian representatives of the industry. Increased market share will become even more important to sustain the anticipated increases in export production once growers become aware of the substantial financial returns mangos can generate.

U.S. Market Opportunities for Fresh Mangos

The U.S. market for fresh mangos continues to expand, with imports accounting for almost all supply. At present, the market is supplied primarily by Mexico, Brazil, Peru, and Ecuador. Guatemala and Haiti supplied lower volumes to the market in 1999.

Development Alternatives, Inc. (DAI) 103

Market Size

The U.S. market continues to expand as more Figure 18: U.S. Imports of Fresh Mangos consumers become familiar with the fruit and 250,000 150,000 learn it is available year-round. Imports account for 99 percent of total sales, with 200,000 120,000 Florida accounting for minimal (and declining) domestic production. In 1999, the 150,000 90,000

U.S. imported 219 metric tons of mangos MTs 100,000 60,000 (value estimated at $142 million), an increase US$000s of 11 percent over 1998 volumes. One Los 50,000 30,000 Angeles-based importer predicts steady growth of 7-10 percent annually over the next 0 0 few years, with higher demand during the 1995 1996 1997 1998 1999 Volume Value winter months. Source: US Census Source: U.S. Census

Suppliers

Figure 19: U.S. Import Market Share Mexico remains the dominant supplier with an (1999 by volume) 81 percent market share in 1999. Mexico’s Brazil 6% relative share has decreased marginally over the Mexico Peru last few years (from 88 percent in 1994), 81% 5% primarily because of increased shipments from Ecuador Brazil, Ecuador, and Peru during the Mexican 5% counter season. Mexico’s main advantage over Guatemala 4% these other suppliers is its proximity to the Haiti United States, which it shares with rival 4% Source: U.S. Census Others Source: US Census suppliers Haiti and Guatemala. Like Mexico, 1% Haiti is a significant supplier during the summer months. Despite intense competition from Mexico, Haiti is able to compete by shipping mangos directly to Miami. Brazil is the main supplier during mid-September through February. Other suppliers include Ecuador and Peru, both of which gained significant market share in 1999, and Nicaragua and Costa Rica. Shipments from Venezuela have declined since 1995-96.

Seasonality

Nearly 80 percent of all mango imports arrived during April-August in 1999, a time when Mexico supplies the market with large volumes at low prices. Mexico’s main competitors during the spring are Guatemala and Haiti. Reliable supplies from Brazil, Peru, and Ecuador from September to March have made mangos available on a year-round basis.

Chapter Three—Mangos 104

Figure 20: U.S. Monthly Imports of Mangos, 1999 (volume in metric tons)

50,000

40,000

30,000

MTs 20,000

10,000

0 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Source:Source: US U.S. Census Census Mexico Others

Prices

Mango prices tend to be volatile because of the seasonal change in suppliers. During the summer months, when the Mexican season is in full swing, prices average as low as $4.00 per carton (4.5 kilograms net). Figure 21 provides a sampling of fresh mango wholesale prices in the Los Angeles terminal market from key supplying countries. Because of lower transport costs vis-à-vis competitors, Mexico is able to sustain these lower prices and effectively dominate the market. As the Mexican season ends, prices increase. In 1999, these averaged $7-10 per carton from September to December.

Figure 21: Los Angeles Wholesale Prices for Mangos ($ per 4.5-kg. carton)

20.00

16.00

12.00

8.00

4.00

0.00 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec

Source: USDA/MNS 1997 1998 1999

Importers note that prices fluctuate greatly from year to year as a result of supply and weather conditions. Because of sporadic changes in the market, it is difficult to predict a trend in the prices of fresh mangos. In general, however, prices have shown a decrease over the last few years, both during the Mexican season and in the fall and winter. The overall

Development Alternatives, Inc. (DAI) 105

drop in prices can be attributed to increased volumes sourced from Brazil and other South American suppliers.

Preferences

The majority of U.S. consumers remain generally unfamiliar with mangos, despite their growing availability at many large-scale supermarket chains. Importers note that Tommy Atkins remains a popular variety because of its red blush, despite it being fibrous and less sweet than others. Retailers contend that this stems from the propensity of consumers to select fruit based on their appearance, rather than taste; hence, produce managers are reluctant to stock supermarket shelves with less attractive varieties that taste better than Tommy Atkins. Of the major imported varieties, Haden possesses superior best taste. A disadvantage to the Haden is its limited shelf-life relative to the Tommy Atkins and Kent varieties. Although red is currently the color of choice among U.S. mango consumers, some buyers believe that sweeter, yellow-skinned varieties such as Mexico’s Ataulfo mango may soon develop a following. In addition to the large presence of poor quality varieties, the appearance, texture, and taste of mangos found in the United States are inferior to what consumers in other countries enjoy because most imported mangos are treated with hot water to meet phytosanitary requirements.

Chapter Three—Mangos

107

ANNEX I

MANGO PRODUCTION AREAS

Annex I—Mango Production Areas 108

Development Alternatives, Inc. (DAI) 109

MANGO PRODUCTION AREAS

Prevailing winds

50% L-1-H L-1-H L-1-H

L-1-H

E = Early-season harvest (Nov.-Jan.) M = Mid-season harvest (Mar.-May) Poor M-1-P roads M-1-H/P L = Late-season harvest (June-Sep.) 25% 1 = One primary season Paved 109 2 = Two primary seasons roads P = Trees grow on plains M-1-P E/L-2-P H = Trees grow on hillsides E/L-2-P

E-1-P Poor pave roads d E/L-2-H/P roads M-1-P M-1-P/H M-1-P

Annex I—Mango Production Areas 110

Development Alternatives, Inc. (DAI) 111

ANNEX II

NUMBER OF BOXES OF MANGOS SHIPPED FROM HAITI

Annex I—Mango Production Areas 112

Development Alternatives, Inc. (DAI) 113

MANGOS

Table 16: Number of Boxes of Mangos Shipped from Haiti (November 1999-September 2000)

Plant Name Nov. 99 Dec. 99 Jan. 00 Feb. 00 Mar. 00 Apr. 00 May. 00 Jun. 00 Jul. 00 Aug. 00 Sep. 00 TOTAL A&G - 540 1,220 - 6,770 21,844 12,739 24,759 23,160 8,000 - 99,032 Caribbean Produce - 770 200 - 1,740 44,255 51,557 32,729 31,680 70,476 1,421 234,828 La Finca - 223 - 270 - 35,717 119,274 88,615 7,416 121,760 26,056 399,331 Germain Paul - 3,009 - - - 30,300 129,802 97,003 13,104 45,761 - 318,979 Haiti Sunshine ------Harold Bussenius - 558 2,449 1,096 6,163 10,410 560 4,441 5,385 4,796 - 35,858 HTM (Agro-Pak) - 1,132 2,672 3,650 33,180 88,851 40,800 85,005 45,066 69,034 14,720 384,110 Jean Jacques Sylvain ------8,982 1,860 - 26,486 2,000 39,328 JMB - 8,996 3,126 507 14,566 88,313 80,427 88,396 39,660 68,860 6,420 399,271 M&M - - 690 1,762 7,123 23,980 14,340 28,028 20,400 29,840 402 126,565 Rainbow 1,682 4,300 3,889 3,921 30,386 60,750 34,350 70,051 48,802 86,280 7,176 351,587 Ralph Perry 1,800 3,610 - - - 35,884 88,872 87,750 16,141 25,050 - 259,107 Sagrimpex 1,547 - - 3,100 - - - - - 4,647 TOTAL 3,482 23,138 15,793 11,206 99,928 443,404 581,703 608,637 250,814 556,343 58,195 2,652,643 Source: APHIS/USDA

Annex II—Number of Boxes of Mangos Shipped from Haiti

115

CHAPTER FOUR ETHNIC PRODUCE

SUMMARY CONCLUSIONS

This assessment provides estimates of hillside production for ethnic produce and determines the percentage potentially available for sales beyond family food security. The assessment compares the two project regions of Jacmel and Cap Haitien and selects products from each that have immediate export potential. The chapter profiles hillside farmers, crop mixes, soil and rainfall conditions, production costs, transportation infrastructure, and market opportunities.

Based on the findings of this analysis, which included farm and association visits and numerous interviews with farmers, agronomists, exporters, and related agribusiness operators, the following activities are recommended for HAP implementation within the ethnic produce sector.

Increase Volume and Channel Efficiency of Ethnic Products

The volume and channel efficiency of ethnic products should be increased: all the recommendations that follow support this one. Current sales of ethnic produce are piecemeal in Haiti, thus not allowing for any type of stable trade to develop and grow. A lack of transparency and efficiency translates into lower returns for fewer farmers. The first step will be to start the product flow, so the project will have the basic information to evaluate the constraints and the advantages of one product vis-à-vis another and thus to determine which crops present the best opportunities for increased value of exports.

Work with Exporters to Increase Trade in Yams and Pumpkin

Begin monitoring and assisting with farmer collection and sales of yam to target exporters this season.

Work with exporters to set quantity preferences, supply timetables, and quality standards for yams next season.

Work with exporters to set quantity preferences, supply timetables, and quality standards for pumpkins in June-September 2001 season.

Assist exporters as necessary with buyer contacts.

Track U.S. prices for yam and pumpkin, disseminating weekly price reports to participating exporters (and farmers’ groups) during season.

Chapter Four—Ethnic Produce 116

Negotiate with exporters to obtain optimal prices for farmers, to provide technical support, and to make available containers where possible and appropriate for loading product in major regions.

Select Lead Farmer Groups

Determine if all the farmer associations in the Jacmel and Cap Haitien regions that grow ethnic produce can absorb HAP technical assistance in the short term; if not, select a subset of “lead” groups for focused initial attention (that can be expanded in Year 2).

Provide Packhouse Development Support to Association

Provide immediate assistance in packhouse development to selected associations, all of which have buildings, with some ventilation available, that can be converted to feature basic packhouse functions (all that is necessary for ethnic produce) with a minimum cost.

Provide Extensions Support to Farmers During Planting Season

HAP agronomists, augmented by short-term specialists as necessary, should work via target associations in Jacmel and Cap Haitien to attain any increased quantity goals (as articulated by exporters) via the planting season in January-March 2001. Closely monitor and report on field progress.

Organize Yam and Pumpkin Collection in Jacmel

Work with the same farmer associations to more effectively and efficiently organize yam and pumpkin collection as necessary, simultaneously introducing simple and appropriate postharvest technologies. Improve volume flow to targeted exporters. Closely monitor and report on field progress, inclusive of prices received by farmers.

Support Planting of Exportable Varieties of Cassava in Cap Haitien Area

Coordinate with Cap Haitien farmer associations to import exportable varieties of cassava (or obtain from Beaumont area). Ensure that an ethnic produce specialist advises HAP extensionists on the optimal approaches to planting and harvest. Consult with exporters to determine volume preferences.

Development Alternatives, Inc. (DAI) 117

Support Increased Collection and Export of Genip

Consult with exporters to determine volume and any other preferences with regard to genip. Verify any increased projections in market demand with Miami-based importers. Determine how best that HAP can support increased collection and export of product.

Identify Demonstration Farms for Training Purposes

For example, consider utilizing the Fondation Vincent farm (approximately 800 hectares) in the Cap Haitien area as a demonstration farm for that region. Although it is a very large farm by Haitian smallholder standards, there appears to be a willingness to ensure that small farmers in the area can have access seeds and other planting materials.

Additionally, HAP could assist Fondation Vincent in planting an optimal mixture of market- led target products, inclusive of organic scotch bonnet peppers, demonstrating appropriate production and postharvest technologies in lowland areas with replicability to hillside farming, as well as in reporting on subsequent returns (which if high enough will ensure adoption throughout the region quicker than most other activities).

BACKGROUND

HAP’s geographic focus is in two regions: Jacmel and Cap Haitien, which are located in Haiti’s northern and southern regions. The goals of the project include the replication of best agricultural practices among small farmers, an increase in their yields, an improvement in disease control, the introduction of appropriate storage and packing methods to lower postharvest losses and ensure better quality, new and expanded markets, and overall increases in farmer incomes.

The key to success is to establish a market-led, systematic approach to the provision of technical interventions that encourages widespread adoption and replication; only through the efficient combination of more and better product can Haitian hillside farmers achieve the economy of scale necessary to penetrate export markets. This need for substantially increased volume is also a reason to consider regions other than just Jacmel and Cap Haitien, if the production of “target” crops is notable elsewhere.

TARGET ETHNIC CROPS

Crops that are currently produced in Haiti and that are referred to in world markets as “ethnic” because of their appeal to primarily immigrant groups of West Indians and Latin Americans (and in other product cases, Africans and Asians) are the following:

Yams; Pumpkin;

Chapter Four—Ethnic Produce 118

Cassava; Avocado; Malanga; and Genip.

The second half of this chapter contains a U.S. export market analysis for these six products, plus scotch bonnet peppers. Please note that scotch bonnet peppers were added because farms near Cap Haitien indicated an interest. This product lends itself to organic production, which addresses concerns expressed earlier about the negative potential for pesticide introduction.

PRODUCTION

Increasing yields and collection efficiencies is essential if any economy of scale is going to be reached to achieve improved returns to hillside farmers. Jacmel should receive initial HAP emphasis over Cap Haitien in the area of ethnic produce, given that pumpkin and yams have the highest immediate export potential in this commodity category and are being produced in substantially higher amounts in Jacmel. By the same token, although the cassava being grown in Cap Haitien is not an export variety, farmers in that region do know how to grow it and the soil and climate conditions are proper for growing the right variety, which can be brought over from the Dominican Republic. The Jacmel climate is not ideally suited to export cassava. Another product that can be exported from the Cap Haitien area is genip, near Gros Morne, where there is reportedly sufficient quantity starting next year to reach 500,000 pounds in exports (100,000 pounds were exported to Miami in 2000 from July to October according to the exporter José Sylvain).

Producers

Haitian hillside farmers are market-based, net food buyers with access to very little land. One major common denominator with implications for HAP or any other agricultural support effort in this country is that, no matter how small the amount of land, Haitian smallholders tend to plant a combination of crops to meet both family needs and market demand. For example, hillside farmers will often have the seven target crops planted, in addition to coffee, cocoa, beans, oranges, and pidgeon peas, on less than 1 hectare.

The two regions have different characteristics in terms of farmer and farmer groups. In Jacmel, associations/centers are located in Blokus, Fond Jean Noel, Bas Fond Jean Noel, Marigot, Cap Rouge, Macarie, and Seguin. These associations/centers are located within a radius of about 25 kilometers from Jacmel and produce coffee, corn, sorghum, green beans, yams, pumpkins, malanga, and pidgeon peas.

The strongest farmer associations in Jacmel are featured, along with the products grown in their areas, in the following chart:

Table 17: Farmer Groups in the Area of Jacmel

Development Alternatives, Inc. (DAI) 119

Group Name Abbreviation Members Location Products Asosyason Peyizan APF 1,775 Bas fon Yams, plantain, honey, Fon Janwel Janwel oranges Asosyason Plante APKF 1,000 Fon Janwel Coffee, beans Kafe Fon Janwel Asosyason Plante APKP 307 Cap Rouge Yams, coffee, pumpkin, Kafe Kapye cassava, cacao Komite Presten Bon KPAB 401 Platon Pumpkin, cassava, cacao, Aksyon Desira avocado, pidgeon pea Oganizason pou ODM 1,300 Meyer Cassava, cacao, malanga Devlopman Meyer Total 5 Groups 4,783

The following are farmer groups in Cap Haitien:

Table 18: Farmer Groups in the Area of Cap Haitien

Group Name Abbreviation Members Location Products Couers Unis N/A 15 Balan Cassava, corn, coffee Agricole Cafetiere CACGAVA 575 Dondon Coffee Gabart Coop Agric. Cafetiere COOPPACVOD 502 Dondon Coffee, pigeon pea, yam Vincent Oge COOP Petit Planteurs CPPHM 181 Plaisance Coffee, yam Haut Martieau Coop. Agr. Jean JBC 455 Grande Cacao, cassava, pineapple, Baptiste Riviere du pigeon peas Nord Coop. Agric. Bedoret KAB 82 Bedoret Coffee, cassava, yams

Coop. Agric. De Gobe KADG 195 Gobe, Coffee, yam, pigeon peas Plaisance Coop. Tet Ansanm KOTAM 70 Bahon Mango, cacao, pumpkin, avocado Konfederasyon KOTATPOP 60 Pilate Yam, pigeon pea, sugar Teknisyen Agricol cane Komite de Pilate MODEKOP 1,500 Pilate Cacao, yam, cassava Total 10 groups 3,635

Farm Size/Types

In Haiti, there is an active land market. All siblings inherit land, and women can own land. In a 1996 impact survey done by PADF, 40 percent of plots were purchased by farmers and another 23 percent were sharecropped or rented. The following is an assessment of average farm size for one Jacmel-based association:

Farmer No. of Hectares 1 2.6

Chapter Four—Ethnic Produce 120

Farmer No. of Hectares 2 1.3 3 3.2 4 2.6 5 1.3 6 2.3 7 3.1 8 4.2 9 1.3 10 2.6 11 2.2 12 3.2 13 3.8 14 2.1 15 3.1 16 2.6 17 1.9 Total 37.9

As illustrated in the above table, in the case of this one association, 17 members own a total of 37.9 hectares. On average, this implies that each farm-household owns about 2.2 hectares in Jacmel. In Cap Haitien, however, the average is estimated at 0.5 hectare per farmer- household. Sixty-five percent of the farmers there have less than 1 hectare of agricultural land. About 30 percent were found to own 1 hectare, while only 5 percent own more than 3 hectares. In Jacmel, 15 percent have an area of more than 3.5 hectares, 35 percent have less than 2.2. Following is the crop breakdown for an average 3 hectare farm in the area of Fond Jean Noel in Jacmel:

Product Mª % Banana 2,900 9.6 Coffee 3,000 10.0 Corn 3,800 12.6 Pidgeon Peas 3,300 11.0 Beans 5,000 16.0 Yam 2,700 9.0 Avocado 1,200 4.0 Malanga 1,100 3.6 Sorghum 4,000 13.0 Cassava 1,200 4.0 Oranges 1,800 6.0 Total 30,000

Inputs

It is a rare exception for a Haitian hillside farmer to have any experience using fertilizers or pesticides on his or her land because they require spending cash, which is in short supply for traditional farm-households.

Development Alternatives, Inc. (DAI) 121

Returns

Information in Haiti is very hard to get because farmers are not used to keeping records of any kind for their various activities. Associations and a PADF agronomist estimate the following percentage of hectares in production for each crop.

Table 19: Total Hectares in Production for Selected Ethnic Crops in the Jacmel Area

Malanga 378 hectares Yams 946 hectares Cassava 420 hectares

An estimated 50 percent of farmed crops are used in the household with surplus quantities bartered or sold in local markets near the growing areas. Madame Saras (traditional wholesalers) often take the products from there to farther markets that have the highest demand and therefore price. Haitians rely on maize, pidgeon peas, green beans, pumpkin, malanga, yams, and yucca for their daily subsistence, although even a certain amount of “cash crops” (that is, sugar cane, cacao, and coffee) are kept for household use.

Farm families generally work the plots of land they own, but in all tasks they perform, they also use hired hands who receive 50 gds. per day plus food for the days they work. Thus, for certain “economic” costs such as planting and weeding, the workers can be hired labor or family labor, but it always has a cost that has to be considered.

Table 20: The Cost of Production Activities in the Jacmel Area (gourdes/hectare)

Product Planting Weeding Fertili- Seeds Harvest Total Yield Cost Selling Total Profit* % zation Cost (kg.) /kg. Price/kg. Sales Profit Corn 1,000 1,000 200 120 350 2,675 1,500 1.66 2.5 3,750 1,250 33.0 Beans 1,300 100 200 1,800 525 3,925 1,000 3.00 4.0 4,000 1,025 27.0 Sorghum 1,000 1,000 0 50 350 2,400 1,300 1.80 2.7 3,500 1,100 31.0 Malanga 3,000 1,000 0 2,000 700 6,700 3,000 2.30 4.0 12,000 5,300 44.0 Yam 4,000 1,200 0 10,000 850 16,050 6,000 2.70 4.5 27,000 10,950 40.0 Cassava 2,800 1,200 0 2,200 1,600 7,800 5,000 1.56 2.1 10,140 2,340 22.8 Pidgeon 1,300 1,000 0 250 350 2,900 1,500 1.93 3.0 4,500 1,600 35.0 Peas Pumpkin 1,000 1,000 0 50 200 2,250 3,500 0.64 1.5 5,250 3,000 57.0 * = Figure based on 1 hectare of harvested crop at current market prices.

Table 20 demonstrates the different variables that exist in the production of the products farmers around the Jacmel area are used to planting for their own consumption and for selling to the market. Some farmers have more than 1 hectare. The total average income for a farmer with 8 hectares of crop is estimated at 3,652 gds. per year. Pumpkin gives the highest returns, mainly because it is a crop that has become popular with the exporters, who are interested in increasing the flow of this product. Thus far, in 2000 200,000 pounds of pumpkin have already been exported to the United States. The pumpkin can be exported year-round.

Chapter Four—Ethnic Produce 122

Production of such export products contributes significantly to the household income of farmers, while helping them to shift to a more commercial orientation. HAP should start assisting farmer associations to more efficiently produce these crops in adherence to international market standards, which will help increase yields and decrease losses even for household and local consumption purposes.

This is the first time that a cost /benefit analysis has been done for farmers in the Jacmel area. The information gathered from the different farmer groups is not 100 percent accurate, mainly because the farmers are not used to sharing or tracking this kind of information and are also cautious about sharing actual figures with competitors. It gives a guideline, however, with which to make more responsible assessments.

Climatic Conditions and Seasonality

Annexes I and II comprise regional maps and rainfall charts. Note that soil and water analyses should be considered as basic and vital information for sound planting decisions, and these analyses still need to be consulted. From the field analysis, however, it is apparent that the predominantly clay Haitian soil (60 percent) produces very good yield and that the pH is generally neutral or basic at 7.0-8.2. There are slightly acid soils in some areas, but these are rare.

In terms of rainfall, the rainiest season for Jacmel is June until September/October, with the February to May the driest months. In Cap Haitien, April and May are the rainiest months. Table 21 provides a related view of planning and harvesting seasons for selected products in Jacmel, including most of the ethnic products tentatively targeted by HAP. One of the project’s activities in the ensuing months will be to compile more comprehensive seasonality charts for commercial distribution (and web posting).

Table 21: Planting (P) and Harvesting (H) Seasons for Jacmel Area

Jan. Feb. Mar. Apr. May June July Aug. Sept. Oct. Nov. Dec. Jan. Feb. Mar.

Corn P P P H H

Beans P P P P P H H

Malanga P P P H H H

Yam P P P H H H

Pidgeon Pea P P P H H H

Pumpkin P P P H H H H

Cassava P P H H H

Sorghum P P P H H H

Development Alternatives, Inc. (DAI) 123

Yields/Quality

Pumpkin

The Jacmel area produces relatively good quality pumpkin, and farmers already have experience packing for export with good results. During the past 14 months, 450,000 pounds have been exported for a total profit of $1 million gds. Pumpkin is an easy product to grow and has a long shelf-life without the need for cold storage, and can thus be exported year- round. In addition, exporters have a good tolerance for different shapes and sizes, and the export price is much better than the local market. Domestic consumption is high, however, which is an advantage because rejects for the export market can be sold to Madame Saras or other local buyers.

Yams

In the region of Macarie (Jacmel), the area in yam production is large (200 hectares) and farmers know how to grow the crop. In addition, the yield per hectare is good (5-6 tons) and production is thriving even without the use of any fertilizer whatsoever. The varieties grown in this area can be exported and are produced in sufficient quantities. Two varieties are being grown: the yellow variety (which makes up 30 percent of total yam production) and the white variety (70 percent of total). One could reasonably anticipate about 30 percent of yam production fit for packing and exporting in the first year of the project. In addition, a packing “center” for cleaning, selecting, and packing of yams is ready for the late November to February harvest.

Genip

This fruit (also called guenip or quinep), which resembles rambutan or litchi, is a product loved by Haitians in Miami, where it is known as kenepas. Genip is not farmed but grows in the wild. The largest quantities are found near Port-au-Prince, in Leogan, and toward Cap Haitien in the northwest, near Gros Morne. Last season (June/July-September), 100,000 pounds were exported to the United States. Next season, it is reportedly possible to export 500,000 pounds. Farmers have learned to pick the fruits when ripe and pack them in 10- kilogram boxes for the Miami market.

Cassava

The cassava variety currently grown in the Cap Haitien area (in Grande Rivière du Nord) is not the right variety for export, but the soil and climate conditions are good and farmers know how to grow it. Cassava promotion should be encouraged only with the right variety,

Chapter Four—Ethnic Produce 124 which is available from the Dominican Republic. Cassava can then be easily grown and packed for export.

Other Crops/Opportunities

Sufficient quantities of sweet and sour cassava, with weight and dimensions in adherence to export standards, will reportedly be available this coming season near Beaumont. Around the same region (Jeremie), malanga is grown in sufficient quantities for export, although Port-au- Prince is more than 12 hours away by road.

Finally, although not an ethnic product, pineapple is cultivated in the area of Grande Rivière du Nord (near Cap Haitien) in very large quantities. The variety is very different from the one the market demands, but the right conditions exist, as with cassava (that is, good soil and climate conditions and farmers familiar with its production), for new exportable varieties to be planted, so HAP might want to assess this crop further in the next few months.

Perhaps even more important, some exporters are interested in making pineapple juice. La Famosa, a mango exporter, has the installations necessary to make various kinds of juices, and pineapple can be one of them. HAP should therefore take this under consideration and ensure that any commercial technologist it brings on to review processing opportunities and operations for mangos also considers pineapple.

POSTHARVEST CONSTRAINTS AND OPPORTUNITIES

The major goal in this point of the chain from farm to market is to decrease rejects and loss of product. Every association toured in both regions had at least one building at its center (one had five). These buildings serve a variety of purposes, including as churches, schools, and meeting centers, and operate only primitively in a packhouse capacity. Therefore, it would be premature at this point to do a hard analysis of their costs and returns.

Nevertheless, the buildings do serve their purpose of helping get product to market, and generally they have enough ventilation so with a minimal investment they could be transformed into basic packhouses. It is strongly recommended that, because the foundation infrastructure is already in place, the project in the very short term should assist in this transformation, which would have immediate benefit in terms of improved grading, sorting, and packing for all seven crops.

Packaging costs in Haiti are $0.65 per 50-pound box and $0.50-0.75 per box for labor to sort and pack the product. If fetching product from Jeremie, for example, which is 12 hours by truck from Port-au-Prince, the total charges for the exporter can be as high as $1,000 a container.

Development Alternatives, Inc. (DAI) 125

TRANSPORT

Jacmel is located 121 kilometers from Port-au-Prince by road. The road is fairly good, and the amount of time needed to cover the distance by truck is about three hours, which translates to little bruising (if products are packed properly). The condition of the roads from the main growing areas to Jacmel proper (about 25 kilometers) is very poor. Mostly mountainous rocky dirt tracks, they require the use of a four-wheel drive vehicle even during the dry season.

The road from Cap Haitien to Port-au-Prince is good for 60 percent of the time, although for 35 kilometers the road near Gonaïves is very bumpy. The time needed to cover the distance is about six hours, but products can still arrive at port in acceptable condition, again if they have been packed properly.

Most product is moved from the associations to nearby markets by small trucks and from the smaller plots to the associations by donkeys or small trucks. It is possible to truck containers to the associations if there is enough volume to warrant the expense, and exporters are currently looking into this option in select cases. The product would then be hauled to Port- au-Prince for final sorting and packaging. This option would cost them about $100 per truck.

Most ethnic produce will be shipped out of Haiti by sea, although some product like scotch bonnet peppers could go by air given the relatively low cost ($0.60 per kilogram), available cargo space, and daily flights to Miami.

MARKETING

Many Haitian crops are raised specifically for the market; they are usually sold or bartered in small amounts in markets near to the growing areas, where Madame Saras take the crops to those markets with high demand and prices. This lack of volume not only stems from food security issues but also from very small plots of land and poor agricultural practices that lead to substantially lower yields and greater postharvest losses than need be the case. Following Figure 22: U.S. Imports of Fresh Yams are brief export market analyses for the six targeted ethnic products (and scotch bonnet peppers). 35,000 $30,000 30,000 $25,000 25,000 Yams $20,000 20,000 $15,000 U.S. imports of fresh yam have increased MTs 15,000 $10,000 US $000s since 1995, gaining particularly strong 10,000 growth in the last three years. After a slight 5,000 $5,000 drop in imports between 1996 and 1997, - $- growth has been steady, increasing 18 percent 1995 1996 1997 1998 1999 between 1998 and 1999 to 28,910 metric Source:Source: U.S. US CensusCensus Volume Value

Chapter Four—Ethnic Produce 126

tons. Despite the gains in volume, the corresponding value of fresh yam imports fell from $24.68 million in 1998 to $21.89 million in 1999 (but was still 23 percent higher than four years prior).

Yams are imported into the United Figure 23: U.S. Monthly Imports of Fresh Yams, 1999 States on a year-round basis. Both 3,500 production and demand fall during 3,000 warmer weather, resulting in a 2,500 decline in imports in the spring MT2,000 until an increase at summer’s end. 1,500 1,000 Costa Rica claims the majority of 500 the U.S. market, particularly - during the first four months of the Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec year, while Jamaica and Ghana Costa Rica Jamaica have significant market share as Others well. Costa Rica’s market share Source: U.S. Census increased from 37 percent in 1994 to 60 percent in 1999. Over the same period, Ghana’s share increased slightly from 4 percent to 7 percent, whereas both Jamaica and Colombia have lost market share. Other suppliers in 1999 included the Dominican Republic, Japan, and Panama.

Yam prices vary sharply by variety. The Jamaican Yellow variety consistently sells 50-100 percent higher than the more plentiful Costa Rican Blanca. In the most recent season, the Jamaican Yellow yams commanded strong premiums over Blanca, peaking at $80 per 50- pound carton. This was because of drought and a recent U.S. Department of Agriculture ban on a chemical long used by Jamaican producers. The price of Jamaican Yellow has since fallen and is trading in the mid-$40s, whereas the Blanca is trading around $20 per 50-pound carton.

Cassava Figure 24: U.S. Imports of Fresh Cassava U.S. imports of fresh cassava have been steady 30,000 18,000 for the past five years, averaging 25,000 metric 25,000 15,000 tons per annum. Import value has also been 20,000 12,000 steady, with the exception of a 1999 drop from 15,000 9,000 $15.55 million to $8.8 million in spite of the MTs volume increase. Sources indicate that the 10,000 6,000 US $000s decline in value is because of the substantial 5,000 3,000 oversupply by the main supplier of cassava, - 0 Costa Rica. 1995 1996 1997 1998 1999 Source:Source: U.S. US Census Census Volume Value The U.S. imports cassava year-round, with supply fluctuating between 1,900 metric tons and 2,700 metric tons per month in 1999. Costa Rica’s domination of the cassava market represents a challenge for new suppliers because

Development Alternatives, Inc. (DAI) 127

Costa Rica regularly maintains market share of over 96 percent and last year supplied 99.6 percent of U.S. imports.

Five other countries shipped minimal volumes to the United States: Figure 25: U.S. Monthly Imports of Fresh Cassava, 1999 Dominican Republic (which exported 3,000 21 metric tons), Nicaragua, Ghana, 2,500 Panama, and Ecuador. 2,000 1,500

MTs Prices have remained low throughout 1,000 the last year, spurred by continuing 500 Costa Rican oversupply. The Miami - n n l eb ar pr ay Ju ug ep ct ov ec wholesale market price for 50-pound Ja F M A M Ju A S O N D containers of cassava was $9-11 for Costa Rica Others most of the previous 12 months. Source: U.S. Census

Figure 26: Miami Wholesale Prices for Costa Rican Cassava (price per 50-lb. carton)

20.00 16.00 12.00 8.00

US Dollars 4.00 0.00

9 9 9 0 0 0 0 0 0 0 0 0 -9 -9 -9 -0 -0 -0 -0 -0 -0 -0 -0 -0 t v c n b r r y n l g p c o e a e a p a u Ju u e O N D J F M A M J A S

Avocados

The most commonly imported avocado in Figure 27: U.S. Imports of Fresh Avocados the United States is the Hass variety, which is smaller, oilier, and more flavorful than 80,000 80,000 the West Indian, or “Greenskins,” variety 60,000 60,000 produced in the Caribbean. Examples of Greenskin varieties include Lula, Monroe, 40,000 40,000 Hall, and Nesbitt. However, Greenskins MTs US $000s remain particularly popular among 20,000 20,000 Caribbean and Cuban consumers—the Hass variety turns a deep purple or black - 0 when it is ripe, whereas the Greenskins do 1995 1996 1997 1998 1999 Source: US Census Volume Value not, leading these consumers to assume that Source: U.S. Census a black avocado is rotten.

Chapter Four—Ethnic Produce 128

U.S. imports of fresh avocado increased gradually between 1995 and 1997 and then leapt by over 100 percent during the following year, both in volume and in value, because of the first- time entry of imports from Mexico. Imports in 1999 fell slightly from the previous year but still finished at 55,185 metric tons, well above prior levels. Import value has increased annually since 1995, reaching $72 million in 1999.

The bulk of U.S. imports comes from Chile and Mexico and is almost exclusively the Hass variety. Imports from the Caribbean have seen a slight decline in volume from 1996 to 1998, followed by a substantial increase in 1999 with 8,522 metric tons.

There are notable changes in product seasonality throughout the year. Chile’s main months of supply are September through December, whereas Mexican shipments are restricted to 19 northern U.S. states from November 1 to February 28. The California season takes over from March through September. Total U.S. production per annum is around 150,000 metric tons, roughly three times the level of all imports.

Table 22: U.S. Avocado Supply—Seasonality Chart

Jan. Feb. Mar. Apr. May Jun. Jul. Aug. Sep. Oct. Nov. Dec. California Florida Mexico Chile Dominican Rep.

little to no supply* minor supply* major supply*

* Producers are evaluated relative to their own exports to the United States, not to worldwide totals Source: U.S. Census and importer interviews

The Dominican Republic holds 90 percent of import market share among Greenskin suppliers, and its export season lasts from June until late February. The Bahamas is the only other significant competitor. Jamaica and Dominica have not exported avocado to the United States in several years. Haiti did not export avocados until 1999, when it shipped 3 metric tons valued at $7,200. Although the volume of Caribbean imports has rarely fluctuated, its relative market share in the United States has fallen sharply because of increased imports from Chile and Mexico.

There are notable differences in the wholesale price of Hass and Greenskin avocados, with Hass normally commanding higher figures. Double-layer packages (24 pounds) of Hass range from $40 during the height of the California season to $63 at other times, while double-layer packages of either Florida or Dominican Greenskins rarely broke the $20 mark in 1999-2000, often at $13-15.

Development Alternatives, Inc. (DAI) 129

Figure 28: New York Wholesale Prices for Greenskin Avocados (price per 24-lb. container) 25.00 20.00 15.00 10.00

US Dollars 5.00 0.00

9 0 9 00 0 0 -99 v-99 r-00 - g-00 ct o ec- p Jul u O N D Jan-00 Feb-00 Mar- A May-00 Jun-0 A Sep-00

Source: USDA/MNS Dominican Republic Florida

Malanga Figure 29: U.S. Imports of Malanga A root crop and staple of Hispanic and Cuban 6,000 cuisine, malanga, comes in four main varieties: 5,000 Blanca (white or yellow flesh), Lila (pink flesh), 4,000 Coco (rounder and fatter), and Eddoes (a type of taro root commonly referred to as malanga). 3,000 2,000 Official import statistics do not exist for malanga; MTs per year 1,000 however, Figure 29 provides approximate annual 0 import levels based on interviews with Florida Blanca Lila Coco Eddos Source: Fintrac Importer Surveys importers. Imports of the Blanca variety are estimated at 4,000-5,000 metric tons per year; of the Lila variety, 2,500-3,500 metric tons; of the Coco variety, 4,300-5,000 metric tons; and of the Eddoes variety, 225-300 metric tons. Imports are thought to be growing steadily as ethnic populations within the U.S. increase in size.

Costa Rica is by far the largest supplier of malanga to the United States. It produces all four varieties and exported a total of $8.85 million of the product worldwide in 1999. The Dominican Republic exports primarily the Coco variety and is reported to ship far smaller quantities than Costa Rica. Malanga is also grown in small quantities in southern Florida.

Chapter Four—Ethnic Produce 130

Because of the critical lack of supply and the poor quality of recent shipments from Costa Rica, prices for good Blanca and Lila product have surged to three times the normal rates. Coco and Eddoes are not encountering such volatility. Importers note that prices are less volatile in times of consistent supply.

Figure 30: Miami & New York Wholesale Prices for Malanga (price per 50-lb box) 50.00

40.00

30.00

20.00 US Dollars 10.00

0.00

Jul-00 Oct-99 Nov-99 Dec-99 Jan-00 Feb-00 Mar-00 Apr-00 May-00 Jun-00 Aug-00 Sep-00 Oct-00 CR Blanca (NY) CR Lila (NY) Source: USDA/MNS CR Coco (Miami) DR Coco (Miami)

Calabaza

Variously called Tropical Pumpkin, West Indian Figure 31: U.S. Imports of Fresh Pumpkin, Calabash, Auyama (Dominican Pumpkin Figure 11: US Imports of Fresh Pumpkin Republic), and Joumou (Haiti), the calabaza is a pumpkin-like squash found in a variety of sizes. 40,000 $16,000 Official U.S. Census import figures do not break imports down beyond the general “pumpkin” 30,000 $12,000 category, which also includes jicama and 20,000 $8,000 breadfruit. Imports in this larger category from MTs US $000s Caribbean and Latin American sources have 10,000 $4,000 risen by 23 percent since 1995, from 26,400 metric tons to 32,500 metric tons in 1999. - $- Import value dropped slightly between 1995 and 1995 1996 1997 1998 1999 1997, but by 1999 had increased 118 percent to Source:Source: U.S.US Census Census Volume Value $15 million.

Calabaza imports tend to be seasonal. From January to May 1999, U.S. imports ranged 3,300 to 3,900 metric tons per month. Between June and December, they dropped to 1,500-2,200 metric tons per month.

Development Alternatives, Inc. (DAI) 131

Mexico controls a slim majority of the market, with 51 percent of total market share by volume. Other suppliers include Panama, Trinidad and Tobago, Costa Rica, Dominican Republic, Honduras, and Jamaica (these figures exclude Canadian pumpkins, which flood the U.S. market for a two-month period in September and October).

Prices for 50-pound sacks of Jamaican calabaza have remained stable over the last 12 months, ranging from $13 to $18. The exception was a surge in price at the Miami wholesale market from November 1999 to February 2000 that led to prices as high as $29 per 50-pound sack.

Figure 32: U.S. Monthly Imports of Fresh Pumpkin, 1999 Figure 12: US Monthly Imports of Fresh Pumpkin, 1999 5,000

4,000

3,000

MTs 2,000

1,000

0

r eb un Jul Jan F Ma Apr May J Aug Sep Oct Nov Dec

Source: U.S. Census

Figure 33: U.S. Import Market Share: Fresh Figure 13: US Import Market Share: Pumpkin,Fresh Pumpkin 1999 (1999 (by by volume)volume) Panama 19%

Costa Rica 7% Mexico Trinidad 55% 8%

DR Others 5% Source:Source: U.S. US Census Census

Chapter Four—Ethnic Produce 132

Figure 34: New York Wholesale Prices for Jamaican Calabaza Figure 14: New York Wholesale Prices for Jamaican Calabaza (price per 50-lb. sack) (Price per 50-lb sack)

35.00 30.00 25.00 20.00 15.00 10.00 US Dollars 5.00 0.00

9 0 9 0 0 -00 l-00 - v-99 b-00 ar-00 y g-0 o Jan-00 Ju Oct-99 N Dec- Fe M Apr-00 Ma Jun-00 Au Sep

Source: USDA/MNS

Scotch Bonnet Peppers

Native to Jamaica, other Caribbean Figure 35: U.S. Imports of Fresh Chillies islands, and Belize, the Scotch Bonnet Figure 15: US Imports of Fresh Chillies pepper is related to the Habanero and is 160,000 $120,000 140,000 $100,000 the most popular of the Caribbean 120,000 varieties. It is also referred to as 100,000 $80,000 Bahamian, Bahama Mama, Jamaican 80,000 $60,000 Hot, and Martinique Pepper. MTs 60,000 $40,000 US $000s 40,000 $20,000 Official census figures do not break 20,000 - $- down imports of chili peppers by 1995 1996 1997 1998 1999 variety. Total U.S. imports of fresh hot Source:Source: U.S. US CensusCensus Volume Value peppers have shown consistent growth in recent years, increasing over 50 percent by volume and value since 1995. In 1999, imports totaled 135,000 metric tons and were valued at $106 million, although 99 percent of these came from Mexico. Several others countries, including the Dominican Republic and Jamaica, vied for the remaining market share.

Mexico exports fresh hot peppers year-round, with the primary season stretching from August to November. Import volumes hovered around 10,000 metric tons in 1999, jumping to a peak of 19,500 metric tons in September. Imports from other suppliers reach their maximum of 30 metric tons per month during May and June.

Wholesale prices for scotch bonnet peppers are consistent throughout the year, occasionally varying a few dollars per box. Prices during the last 12 months were at $14-20 per 8-pound box.

Development Alternatives, Inc. (DAI) 133

FigureFigure 36: 16: Miami Miami Wholesale Wholesale Prices Prices for for Jamaican Jamaican Scotch Scotch Bonnet BonnetPeppers Peppers (Price(price per p 8-lb.er 8-lb. box) box) 30.00 25.00 20.00 15.00 10.00

US Dollars 5.00 0.00

99 00 -00 00 Jul-00 Oct- Nov-99 Dec-99 Jan-00 Feb- Mar-00 Apr-00 May Jun-00 Aug-00 Sep- Source: USDA/MNS

Genip

Sometimes referred to as the Spanish lime, genip is imported into the United States in small quantities, based on demand from Caribbean consumers. Small quantities are also produced in Florida. Although U.S. Census import statistics do not exist for genip, less-reliable USDA/APHIS statistics show sporadic import levels, most arriving through Miami.

The leading U.S. suppliers are the Dominican Republic, Haiti, and Jamaica. U.S. imports in 1999 reportedly totaled 242 metric tons, representing a decrease from the 509 metric tons in 1998 and the 580 metric tons in 1997. The Dominican Republic’s 211 metric tons represented 87 percent of the U.S. import market in 1999. Haiti supplied the United States with 27 metric tons (or 11 percent of the import market), and Jamaica shipped 5 metric tons (2 percent).

Genip is handled by a small number of importers who specialize in niche Caribbean and Latin American produce. The product is distributed almost exclusively to small ethnic retailers. Those who deal with genip report only occasional imports, which usually arrive during the late summer and early fall. The fruit is normally shipped loose in 15-25 pound boxes, and sells for $12-20 per box wholesale.

Consumers prefer a green skin and will stay away from genips that have turned brown. Buyers state that Haitian genip is larger and has a higher sugar content than that sourced from the Dominican Republic. Dominican genip tends to be sour by comparison, yet has a longer shelf life because of the decreased sugar content. Genip is also produced in Puerto Rico and tends to yield higher prices.

Chapter Four—Ethnic Produce

135

ANNEX I

ETHNIC PRODUCE: REGIONAL MAPS

Annex I—Ethnic Produce 136

Development Alternatives, Inc. (DAI) 137

ETHIC PRODUCE

REGIONAL MAPS

Annex I—Ethnic Produce 138

Development Alternatives, Inc. (DAI) 139

ANNEX II

ETHNIC PRODUCE: RAINFALL CHARTS

Annex II—Ethnic Produce 140

Development Alternatives, Inc. (DAI) 141

ETHNIC PRODUCE

RAINFALL CHARTS

Rainfall for the area has to be analyzed to know exactly when the crops will be planted and to what extent the harvest can be accomplished in each part of the year. The first chart shows the levels of rainfall so far for year 2000 in Region 1. It is important to note that the rainfall in this area peeks in September.

Figure 37: Region 1 Rainfall, 2000

Region 1 Rainfall Year 2000 mm

600

500

400

300

200

100

0 Jan Feb Mar Apr May Jun Jul Aug Sept Oct Nov Dec

The following chart shows the precipitation for the area of Grande Rivière du Nord in Region 2 until August 2000.

Figure 38: Region 2 Rainfall, 2000

Region 2 Rainfall Year 2000 mm

105 90

75 60 45 30 15 0 JA N FEB MA R A PR MA Y JUNE JULY A UG

Annex II—Ethnic Produce

143

Annex II—Ethnic Produce