i

Egerton University

Tegemeo Institute Of Agricultural Policy And Development

Draft for Review-Working Paper No 08/2004

IMPROVING KENYA'S DOMESTIC HORTICULTURAL PRODUCTION AND MARKETING SYSTEM: CURRENT

COMPETITIVENESS, FORCES OF CHANGE, AND CHALLENGES FOR THE FUTURE

By

Kavoi Mutuku Muendo, David Tschirley, and Michael T. Weber

Tegemeo Institute Of Agricultural Policy and Development, Egerton University.

P.O Box 20498 Nairobi. Tel: (020) 2717818

Email: egerton@tegemeo.org and/or tschirle@msu.edu

Support for this research has been provided by the Tegemeo Agricultural Monitoring and Policy Analysis Project (TAMPA) between Tegemeo Insitute/Egerton University and the Department of Agricultural Economics at Michigan State University. Financial support for this project is provided by the Kenya Mission of the United States Agency for International Development. Additional support is also provided by MSU through resources from the Food Security III Cooperative Agreement.

ii

ACKNOWLEDGEMENTS

The authors thank Samuel K. Mburu for assistance in data collection, data analysis and editing, and Sindi Kirimi for assistance in data analysis and editing.

.

iii

TABLE OF CONTENTS

LIST OF TABLES .............................................................................................................iv

LIST OF FIGURES............................................................................................................vi

LIST OF ACRONYMS ………………………………………………………… ……..vii

EXECUTIVE SUMMARY ……………………………………………………… …..viii

1. INTRODUCTION AND OBJECTIVES ................................................................1

2. DATA AND METHODS ..........................................................................................4 2.1 RESEARCH DESIGN ............................................................................................................................. 4 2.2. DATA SOURCES ................................................................................................................................... 5

3. PRODUCTION TRENDS FOR DOMESTIC AND INTERNATIONAL MARKETS...............................................................................................................6

3.1 FRUIT PRODUCTION............................................................................................................................ 6 3.2 VEGETABLE PRODUCTION................................................................................................................ 10 3.3 INTERNATIONAL EXPORT AND DOMESTIC MARKET SHARES OF FRUITS AND VEGETABLES ........ 14

4. FRUITS AND VEGETABLES IN RURAL HOUSEHOLD LIVELIHOODS .18 4.1 MOST WIDELY GROWN AND SOLD HORTICULTURAL CROPS ........................................................ 18 4.2. INCOME SHARE ANALYSIS................................................................................................................ 20 4.3. CONCENTRATION OF HORTICULTURAL PRODUCTION AND SALES................................................. 20 4.4 SUMMARY.......................................................................................................................................... 23

5. DOMESTIC AND REGIONAL MARKETING CHANNELS AND PRODUCT FLOWS OF FRESH HORTICULTURAL PRODUCE ....................................27

5.1 MARKETING CHANNELS ................................................................................................................... 27 5.2. PRICE RELATIONSHIPS AND LIKELY RETAIL MARKET SHARES .................................................... 31 5.3. FORMAL BORDER POINT IMPORTS AND EXPORTS OF FRESH HORTICULTURAL PRODUCE .......... 32

5.3.1. Imports of Fresh Horticultural Commodities...................................................................... 33 5.3.2. Formal Exports of Fresh Horticultural Commodities ........................................................ 37

5.4. HANDLING OF AGRICULTURAL TRADE AT THE BORDER POINTS ................................................... 38 6. COMMODITY COMPETITIVENESS BETWEEN KENYA AND

TANZANIA/UGANDA .........................................................................................41 6.1. WHOLESALE MARKET ANALYSIS RESULTS .................................................................................... 42 6.2. BORDER POINT TRADERS OF FRESH FRUITS AND VEGETABLES .................................................... 49 6.3. FARM ANALYSIS RESULTS: COSTS OF PRODUCTION FOR ONION IN TANZANIA AND KENYA ....... 52 7. INPUT MARKETS FOR FRESH HORTICULTURAL PRODUCE IN

TANZANIA AND KENYA...................................................................................57 7.1. MARKET AND REGULATORY SYSTEM FOR VEGETABLE SEEDS ...................................................... 57 7.2 VEGETABLE RESEARCH AND DEVELOPMENT IN TANZANIA AND KENYA ...................................... 65 7.3 RESEARCH AND DEVELOPMENT ON FRUITS .................................................................................... 67

7.3.1 Citrus Fruits ................................................................................................................... 67 7.3.2 Bananas ................................................................................................................................. 68

7.4 FERTILIZER AND AGROCHEMICAL INPUTS...................................................................................... 69

iv

8. CONCLUSIONS, RECOMMENDATIONS, AND FURTHER RESEARCH..72 8.1 LEGAL/REGULATORY ENVIRONMENT ............................................................................................. 73 8.2 TECHNICAL PRODUCTION CONSTRAINTS ........................................................................................ 75 8.3 HARD AND SOFT MARKET INFRASTRUCTURE ................................................................................. 76 8.4 OTHER RECOMMENDATIONS ........................................................................................................... 78

8.5 FURTHER RESEARCH .............................................................................................78 REFERENCES …………………………………………………………………………….80 Appendix A Trends in Fruit Production in Kenya Appendix B Trends in Vegetable Production in Kenya Appendix C Enterprise Budgets for Onions in Kenya and Tanzania

v

LIST OF TABLES Table 3.1. Area and production shares of seven fruit crops in Kenya, in 1992 and 2001. 6 Table 3.2. Comparison of Fruit Yield (MT/Ha) Among Top World Producers ..……….10 Table 3.3. Area and production shares of vegetable crops in Kenya, in 1992 and 2001 .10 Table 3.4. Vegetable Yields (Mt/ha) Among Top World Producers .............................. 12 Table 4.1. Districts and sample sizes by zone in “high potential” and “low potential” samples............................................................................................................ 18 Table 4.2. Percent of households growing and selling horticultural crops, and average

value of horticultural production and sales, by geographical area in selected “high potential” areas of Kenya...................................................................... 19

Table 4.3. Most widely grown and sold horticultural crops, by geographical area in selected “high potential” areas of Kenya ........................................................ 19

Table 4.4. Most valuable horticultural crops in production and sales, by geographical area in selected “high potential” areas of Kenya....................... 20

Table 4.5. Income shares by geographical area in selected areas of Kenya .................... 21 Table 4.6. Concentration of horticultural sales: percent of total sales by quintiles of total

household horticultural sales value................................................................. 21 Table 4.7. Concentration of horticultural sales: distribution of households across quintiles of total household horticultural sales value, by zone ....................... 22 Table 4.8. Concentration of horticultural sales: selected household level indicators by

quintiles of total household horticultural sales value...................................... 25 Table 4.9. Concentration of horticultural production and sales by crop: indicators of

geographical and household level concentration ............................................ 26 Table 5.1. Upper- and lower- bound estimates of import market share for selected

horticultural crops in Kenya.............................................................................32 Table 5.2. Import of bananas (bunches) through the Main Cross-Border Posts in Kenya33 Table 5.3. Import of Tomatoes (crates) through the Main Cross-Border Posts in Kenya 34 Table 5.4. Import of Oranges (bags) through the Main Cross-Border Posts in Kenya.....35 Table 5.5. Red Onions Imports (bags) From Various Border Points................................36 Table 5.6. Trade Parameters for Oranges in the Various Border Points...........................37 Table 5.7. Trade Parameters for Oranges in the Various Border Points …..……………39 Table 6.1. Distribution of Wholesale Marketing Costs (Kshs/bunch) of a Bunch of

Bananas From Uganda and Kenya at Wakulima Market in Nairobi .............42 Table 6.2. Distribution of Marketing Margins, Trader Profit and Return on Capital …..43 Table 6.3. Distribution of Wholesale Marketing Costs (Kshs) of a Crate (35 kg) of .......44 Table 6.4. Distribution of Marketing Margins, Trader Profit and Return on Capital for a

Crate (35 kg) of Tomato at Kongowea Market in Mombasa (Kshs/Crate) .....44 Table 6.5. Distribution of Wholesale Marketing Costs of a Bag (100kg) of Oranges

from Tanzania and Kenya at Kongowea (Mombasa) and Wakulima Nairobi) Markets ............................................................................................................45

Table 6.6. Distribution of Marketing Margins, Trader Profit and Return on Capital for a bag of oranges from Tanzania and Kenya...............................................46 Table 6.7. Distribution of Wholesale Marketing Costs (Kshs./bag) of a Bag (100kg) onions from Tanzania and Kenya at Kongowea Market Mombasa..................46

vi

Table 6.8. Distribution of Marketing Margins, Trader Profit and Return on Capital for a Bag of Wholesale Onion (Kshs/bag) from Tanzania and Kenya at Kongowea Market in Mombasa .........................................................................................47

Table 6.9. Distribution of Wholesale Marketing Costs (Kshs./Net) of a Net (13 kg) of Onion from Tanzania and Kenya at Wakulima Market in Nairobi .................48

Table 6.10. Distribution of Marketing Margins, Trader Profit and Return on Capital for a net (13kg) of wholesale onion from Tanzania and Kenya at Wakulima .........48

Table 6.11. Summary marketing cost build-ups for bananas, tomatoes, oranges, and onions from Tanzania, Uganda, and Kenya to Wakulima and Kongowea markets....49

Table 6.12. Reasons for not being a Member of Self Help Group .....................................50 Table 6.13. Sources of Market Information........................................................................50 Table 6.14. Price comparison among countries ..................................................................51 Table 6.15. Quality Comparison of Commodities ..............................................................51 Table 6.16. How to Ensure Quality Commodities are Sold................................................51 Table 6.17. Challenges Traders are Facing.........................................................................52 Table 6.18. Opportunities for Interventions........................................................................52 Table 6.19. Distribution of Onion Production Costs (Kshs/Acre) in Tanzania and Kenya 54 Table 6.20. Percentage Distribution of Onion Production Costs in Tanzania and Kenya..55 Table 6.21. Efficiency Measures of Onion Production in Tanzania and Kenya.................55 Table 7.1. Sources of Vegetable Seed Inputs Sold by Various Companies in Kenya and

Tanzania ………………………………………………………………………63 Table 7.2. Vegetable Seeds Produced Locally and the Approximate Percentage of

Production by Companies ................................................................................63 Table 7.3. Average Prices of Various Vegetable Seeds in Kenya and Tanzania..............64 Table 7.4. Seed Import Charges in Kenya and Tanzania..................................................65 Table 7.5. Crop and Seed Varieties Produced in Tanzania...............................................66 Table 7.6. Average Sales Price (Kshs) of Common Chemicals Among Private Sector

Companies in Kenya and Tanzania..................................................................70 Table 7.7. Average Sales Price (Kshs) of Fertilizer (50 kg Bag) for Public Companies/

Producer Organizations and Private Sector Companies in Nairobi (Kenya) and Arusha (Tanzania) in 2003 …………………...……………………………...70

Table 7.8. Average Sales Price (Kshs/50 kg Bag) of DAP for Public Companies/ Producer Organizations in Various Towns of Kenya, Compared with Previous Private Sector Prices …………………...……………… ……………...71

Appendix C Table C.1. Onion Budgets (Per One Acre) for Mang’ola in Tanzania Table C.2 . Onion Budgets (Per One Acre) for Oloitoktok in Kenya Table C.3. Onion Budgets (Per One Acre) for Taveta in Kenya Table C.4 . Onion Synthetic Budget (Per One Acre) for Narok, Laikipia and Meru in Kenya

vii

LIST OF FIGURES Figure 3.1. Production of Bananas and Pineapples (Tons) in Kenya (1992-2001).................7 Figure 3.2. Production of Citrus, Mangoes and Pawpaw fruits (Tons)

in Kenya (1992-2001) .........................................................................................7 Figure 3.3. Production of Avocados, Passion Fruits and Other Fruits in

Kenya (1992- 2001 ..............................................................................................8 Figure 3.4. Yields (Tons/ha) of Bananas, Mangoes and Passion Fruits in Kenya (1992-

2001). ....................................................................................................................8 Figure 3.5: Yields (Tons/ha) of Avocados, Citrus and Pawpaw in Kenya..............................9 Figure 3.6. Yields (Tons/ha) of Pineapples in Kenya (1992-2001). .......................................9 Figure 3.7. Production of Cabbages, Tomatoes and Kales in Kenya (1992-2001) ...............11 Figure 3.8. Production of Onions, Carrots and Other Vegetables in Kenya (1992-2001). ...11 Figure 3.9. Production of French Beans, Traditional Vegetables and Garden Peas in Kenya

(1992-2001).........................................................................................................12 Figure 3.10. Yields (Tons/ha) of Cabbages, Tomatoes and Kales in Kenya (1992-2001). ....13 Figure 3.11. Yields (Tons/ha) of Onions, Carrots and Other in Kenya (1992-2001). ............13 Figure 3.12. Yields (Tons/ha) of French beans, Indigenous Vegetables and Garden Peas in Kenya (1992-2001). ...........................................................................................14 Figure 3.13. Fresh Vegetable Exports as Share of Total Production, by Value (1992-2001).16 Figure 3.14. Market channel shares (farm, local sales, export sales) of total vegetable production in Kenya, 1997-2001, valued at farm-gate prices.............................16 Figure 3.15. Total value added (and share) in farm, local sales, and export sales channels for

vegetables in Kenya, 1997-2001.........................................................................17 Figure 5.1. Domestic, Regional and International Marketing Channels for Fresh

Horticultural Produce in Kenya ..........................................................................28 Figure 5.2 Principal flows of horticultural products through physical market places in

Kenya …………………………………………………………………………..30 Figure 7.1. Structure of the Seed Industry in Kenya .............................................................58 Figure 7.2. Structure of the seed industry in Tanzania Before Liberalization ......................60 Figure 7.3. Structure of the Liberalized Seed Industry in Tanzania......................................61 Figure 7.4. Structure of Vegetable Seed Sub-Sector in Tanzania since Seed Sector

Reforms...............................................................................................................62 Appendix A Figure A.1. Area Trend Under Bananas in Kenya (1992-2001) Figure A.2. Area Trend Under Citrus, Mangoes and Pineapples in Kenya (1992-2001) Figure A.3. Area Trend Under Avocados, Pawpaw and Passion Fruits in Kenya (1992-2001) Appendix B Figure B.1. Area Trends Under Cabbages, Tomatoes and Kales in Kenya (1992-2001) Figure B.2. Area Trends Under Onions, , French Beans, Carrots in Kenya (1992-2001) Figure B.4. Area Trends Under Indegenous Vegetables, Garden Peas and Other Vegetables

in Kenya (1992-2001)

viii

LIST OF ACRONYMS CIDA Canadian International Development Agency COMESA Common Market for Eastern and Southern Africa EAC East African Community EU European Union FAO Food and Agriculture Organisation FPEAK Fresh Produce Exporters Association of Kenya GDP Gross Domestic Product HCDA Horticultural Crop Development Authority IBR Institute for Biotechnology Research ICIPE International Centre of Insect Physiology and Ecology IFAD International Fund for Agricultural Development IGAD Inter-Governmental Authority JKUAT Jomo Kenyatta University of Agriculture and Technology KARI Kenya Agricultural Research Institute KBS Kenya Bureau of Standards KEPHIS Kenya Plant Health Inspectorate Service KFA Kenya Farmers’ Association KFU Kenya Farmers Union KRA Kenya Revenue Authority KSC Kenya Seed Company MoALD Ministry of Agriculture and Livestock Development MRLs Maximum Residual Levels NCPB National Cereals and Produce Board NGOs Non-Government Organizations OPVs Open Pollinated Varieties PTA Preferential Trade Area QDS Quality Declared Seed SADC Southern African Development Community TAMPA I Tegemeo Agricultural Monitoring and Policy Analysis TFC Tanzania Fertilizer Company THRC Thika Horticultural Research Centre THRI Tengeru Horticultural Research Institute TOSCA Tanzania Official Seed Certification Agency TSC Tanzania Seed Company UK United Kingdom UNDP United Nations Development Programme USAID United States Agency for International Development WV World Vision

ix

EXECUTIVE SUMMARY Kenya’s horticultural sector (defined here to include fruit and vegetable production and marketing, but not flowers) has received a great deal of attention over the past decade due to the rapid and sustained growth of its exports to Europe. This impressive growth has undoubtedly contributed to increased rural incomes and reduced rural poverty in Kenya. Yet despite this growth, exports remain a small fraction of Kenya’s overall horticultural sector. For the past decade, over 90% of all fruit and vegetable production was consumed domestically, and the domestic market accounted for over 90% of the total growth in quantity of fruit and vegetable production. While over 90% of smallholder farmers in all but the arid regions of Kenya produce horticultural products, fewer than 2% do so directly for export. This overwhelming dominance of the domestic market, combined with slower growth experienced in the export sector over the past decade, the challenges that smallholders face to continue participating in the export sector, and the possibility of more rapid growth in domestic demand, all argue for a more active focus on the potentials and constraints of domestic horticulture in Kenya. Such a focus implies also the need to assess the competitiveness of local production and marketing against that of neighboring countries such as Tanzania and Uganda. This paper explores these key issues. Its overall objectives are to provide a broad diagnostic overview of the horticultural sector, to identify specific constraints that limit the system’s performance, to make suggestions for selected policy and programmatic changes, and to identify key research that needs to be done to guide further investments to improve sector performance. The paper is organized as follows. Chapter 2 presents the data and methods used in the report. Chapter 3 focuses on production and yield trends for seven fruit and nine vegetable crops, and estimates the international export market share for vegetables. Chapter 4 uses household survey data to examine the role of horticultural production and sales in smallholder livelihood strategies, and evaluates the structure of production and marketing at the farm level. Chapter 5 identifies the structure of horticultural marketing channels, and quantifies the imports from Tanzania and Uganda of two vegetable and two fruit crops. Chapter 6 develops marketing cost budgets for these same four crops and farm budgets for onions. Chapter 7 evaluates the input sector serving horticulture in Kenya and Tanzania, with a special emphasis on lessons from Tanzania’s experience for the horticultural seed sector. Finally, Chapter 8 presents conclusions, recommendations, and suggestions for further research. Production and Yield Trends: Official production and yield data indicate that the yields of fruits in Kenya over the past decade have been stagnant with the exception of bananas, mangoes and passion fruits, which have risen. Production of banana has recovered dramatically after overcoming disease problems in the mid-1990s. Production of pineapples, mangoes, avocados, and passion fruit has also trended upwards while production of citrus, pawpaw, and “other fruits” has stagnated. Citrus greening disease is a persistent problem contributing to poor performance in that sector. Production of cabbages and carrots has declined over the past decade, while kales, tomatoes and traditional vegetables show steady increases. Vegetable yields have been stagnant with the exception of French beans and indigenous vegetables, which have risen.

x

Using data from various sources for 1997-2001, we estimate that at least four- to five times more horticultural produce, by value, was sold in domestic markets than in international export markets. Value added in domestic markets (post farm gate) was at least three times that in the export sector. Horticulture in Smallholder Livelihood Strategies: Production and sales at the farm level of all the major horticultural crops in Kenya are quite concentrated. Fifteen percent of rural households account for about 80% of all horticultural sales. Concentration is higher for individual crops: in all but one of the top 10 crops, 5% of the rural population accounts for at least 50% of production and at least 70% of sales. Bananas and Sukuma wiki are the least concentrated both geographically and at the household level. Improvements in production and marketing of these two crops would have the broadest impacts on income levels and poverty rates. Carrots, French beans, macadamia nuts and oranges are the most concentrated. For these crops, a private sector led strategy of focused assistance to relatively few growers on production and marketing constraints could be most effective in boosting production and sales. Such a strategy would not be effective in oranges unless the citrus greening problem is first addressed. Households selling the most horticultural produce are better off than other households based on a wide range or indicators. Yet this group still earns a slightly higher income share off the farm than they do through horticultural sales, suggesting substantial continued income diversification. A potential implication is that, if marketing costs and market risk can be reduced and farm level productivity increased, this group of households may be well poised to take advantage of expanding market opportunities through greater specialization. Marketing Channels and Regional Trade Patterns: The traditional marketing system, including urban wholesale markets, continues to play the dominant role in FFV (fresh fruits and vegetables) marketing in the country. Based on retail price relationships between the traditional system and supermarkets, and patterns seen in Central and South America, where supermarket development began earlier, we estimate that the supermarket share of the FFV market in Nairobi is below 10%. Outside of Nairobi, it would certainly be lower. Supermarkets have, however, expanded their share rapidly since they first focused on this market three- to four years ago. The two major chains – Uchumi and Nakumatt – each carry upwards of 80 horticultural products in their Nairobi stores, and each has ambitious expansion plans. As supermarkets have done throughout the rest of the world, Uchumi and Nakumatt are attempting to bypass the wholesale markets in favor of direct procurement with an assortment of contracted commercial farmers and some organized small- and medium-sized farmers. At the present time, traditional wholesale markets are unattractive to buyers concerned with assuring high quality and food safety while reducing procurement cost. Information about the current market share and likely rate of growth of marketing channels that bypass traditional wholesale markets is crucial for understanding the potential evolution of and constraints to smallholder participation in the domestic market. New information is also needed about options for designing investment programs to facilitate continued smallholder participation in fruit and vegetable value chains, while reducing overall marketing costs and prices to final consumers. Banana and tomato imports from the region are estimated to have no more than a 7-8% share of the Kenyan market. Orange imports (nearly all from Tanzania) may exceed 20%, while the onion import share (also nearly all from Tanzania) may exceed half. Kenya exports almost no produce to regional markets.

xi

Regional Competitiveness: Collecting wholesaler budgets are consistent with these observed trade patterns: trader profits per unit of bananas and tomatoes are higher for Kenyan produce than for imports, profits per bag of oranges are higher for the commodity from Tanzania but returns to capital are comparable, and both profit per bag and returns to capital are higher for imported onions. Farm budgets for onion also demonstrate Tanzania’s advantage in this crop: costs of production in that neighboring country are lower by 20-50%, gross margins per acre of land are higher by 60-300%, and gross margins per bag are higher by 15-150%. These results are driven by yields that are 45-100% higher in Tanzania and seed costs that are one-tenth those in Kenya. Higher quality of irrigation in that country may also contribute to Tanzania’s advantage, and superior onion storage infrastructure at the farm level allows it to supply the Kenya market throughout the year. The Inputs Sector: Following liberalization in 1994, government in Tanzania has played a facilitating role in seed sector development, while the focus of Kenya’s authorities is primarily on regulation and “policing”. Community Based Seed Production, especially of the mang’ola red onion variety, and local varietal development more generally, appear to be major successes of this more flexible approach in Tanzania. Horticultural seed research in Kenya is plagued by understaffing. While Tanzania’s research center has five vegetable breeders, a number of seed technologists, and a fully operational seed production unit, Thika Research Center in Kenya has one full-time breeder, no seed technologist, and no operational seed production unit. Kenya’s performance dealing with horticultural diseases is mixed. Widespread distribution of clean banana planting material from the mid-1990s was a major success, reversing dramatic falls in yields and production. Similar success with citrus greening disease is complicated by the fact that it is a vector-borne disease in the mid- and high altitudes. Clean planting material exists, but eradication requires coordinated felling of infected orchards, not just planting of the new material. Kenya Farmer’s Association and National Cereals Produce Board (a parastatal) have both become active in the fertilizer sector since 2001. They charge lower prices than private companies and some suggest that their presence has reduced the prices these companies charge. This development needs to be watched quite closely to ensure that unsustainable government subsidies to NCPB or KFA do not undermine the notable success of fertilizer sector liberalization in Kenya. Conclusions, Recommendations, and Further Research: Fresh fruit and vegetable production and marketing value chains are becoming increasingly important to a broad array of Kenyan consumers. These also hold potential market opportunities for important segments of the smallholder farming community. But investments are needed to upgrade marketing infrastructure and facilitating services for traditional participants in the system. Important forces of change include the entry of supermarkets into the domestic horticultural market; this may extend the dualism currently seen between export and domestic systems into the domestic system itself. Both major supermarket chains are moving towards direct procurement through “preferred grower” programs. Traditional retail and wholesale markets will likely continue to supply a large share of the consumer market for many years to come, primarily because they are well adapted to the shopping habits and preferences of poor consumers. The risk is that this traditional system – and the small farmers and traders who supply it – will be confined to this end of the market, while commercial farmers and a small

xii

number of organized smallholder farmers dominate the expanding and more profitable direct procurement system of the supermarket chains. How to avoid this entrenched dualism, with its negative implications for smallholder incomes, rural poverty reduction, and the quality of the urban food supply, is a key public policy issue over the next five to ten years. Expanding domestic and regional markets for Kenyan horticultural produce and integrating the country’s smallholder farmers into profitable supply chains that satisfy these markets will require investment in three key areas: the legal and regulatory environment, technical production constraints, and “hard” and “soft” market infrastructure. The high level of investment needed means that active partnering by government with donors and private sector will be crucial. Key recommendations regarding the legal and regulatory environment touch on the current revision of the Seed and Plant Varieties Act of 1991, on the Horticulture Bill, and on issues of quality and food safety, farmer organizations, and intellectual property rights in seed varieties. Recommendations regarding technical production constraints focus on consolidating the country’s success in reducing banana diseases, dealing with citrus greening disease in a cost effective manner, and improving adaptive varietal research in the context of a revised seed law that encourages the production of Quality Declared Seed at the village level. Traditional wholesale markets should be the central but not exclusive focus of investments in three key types of hard and soft market infrastructure. First, improved logistical efficiency, especially for loading and unloading, is needed to reduce costs and improve hygiene in the markets. Second, improved hygiene combined with logistical improvements will make these markets more attractive options for a broader range of retail outlets. Third, improved grades and standards, and more easily available information on prices and volume by grade of product, will increase market transparency and further attract customers. Achieving these improvements will require that wholesale market management take on a business orientation while recognizing that it is providing a partial public good by integrating smallholder farmers into a more dynamic and competitive system while providing poor consumers with higher quality produce at competitive prices. Active partnering between government, private sector and donors will be crucial to mobilize the needed financial resources and knowledge to make these improvements. Government and donors could also play an important role partnering with supermarkets to reduce the cost to them of dealing directly with smallholder farmers. Improvement in secondary and tertiary roads is also key to modernizing the sector. To help guide investments to relieve bottlenecks in the production and marketing system, further research needs to provide information in the following areas: Urban Retailing, especially market shares for the full range of retail outlet types, the costs and standard operating procedures of each retailer’s procurement system, and key bottlenecks that, if relieved, could reduce costs and increase quality. Product quality: Understanding the degree and specific mechanisms of quality differentiation in the traditional system is fundamental to designing a more formal system of

xiii

grades and standards that is workable and that can increase transparency and create a dynamic of constant quality improvement. Urban Wholesaling: The behavior and performance of urban wholesale markets affects costs, prices, and the distribution of benefits throughout the production and marketing system. Identifying specific investments to improve logistics, hygiene, and market information requires applied research in close collaboration current and potential users. Links between urban markets and rural producers: To design programs that link small farmers more closely to market outlets, one needs to know more about the system wide “price discovery” process. One would also want to establish how many small farmers sell through associations, what cost and other marketing advantages these associations provide, and what if any price premia these organized farmers receive. Finally, it is important to know what the share of smallholder farmers vs larger commercial farmers is for the main horticultural crops. Rural marketing: We anticipate that many rural households will be net buyers of horticultural produce. If this is true, then the performance of the rural marketing system, including rural retailing, will affect the real incomes of net sellers and net buyers.

1

Improving Kenya's Domestic Horticultural Production and Marketing System: Current Competitiveness, Forces Of Change,

and Challenges For The Future 1. Introduction and Objectives Kenya’s horticultural sector has received a great deal of attention from local and international researchers, government, and donors over the past decade, due to the rapid and sustained growth of its export sector (Jaffee 1994, Jaffee 1995, Swernberg 1995, Kimenye 1995, Stevens and Kennan 1999, Dolan et al. 1999, Kamau 2000, Thiru 2000, Harris et al. 2001, Minot and Ngigi 2002). From a very low base, Kenya’s horticultural exports (defined here to include fruit and vegetables but not flowers) grew 9% per year in the first decade after independence, then 17% per year from 1974-1983 (Minot and Ngigi 2002). Growth slowed over the 1980s and 1990s, but still averaged about 4% per annum over the past decade. By the year 2000, fruit and vegetable exports amounted to US$270m, or 15% of Kenya’s total export economy. This impressive growth has undoubtedly contributed to increased rural incomes and reduced rural poverty, through both direct production effects and linkage effects, as horticultural incomes from export are re-spent in rural areas. Yet despite its rapid and sustained growth, exports remain a small fraction of Kenya’s overall horticultural sector. For the past decade, over 90% of all fruit and vegetable production was consumed domestically, either on-farm or through domestic markets. Despite higher percent growth rates in the export sector, the absolute amount of growth has come overwhelmingly from the domestic sector: between 1992/93 and 2000/01, the domestic market accounted for 98% of the total growth in quantity of fruit production and 91% of the total growth in vegetable production. Even allowing for higher prices of export commodities, the dominance of the local market is clear. This dominance is reflected at the farm level. While over 90% of smallholder farmers in all but the arid regions of Kenya produce horticultural products, fewer than 2% do so directly for export (Bawden et al, 2002. Kenyan smallholders who have succeeded in producing for the export market also face a daunting set of challenges if they are to maintain their participation in the sector. These challenges are driven by increasing consumer demand for quality and food safety in the UK and continental Europe, and by the related rise of supermarkets in these areas. By the late 1990s, supermarkets’ share of the fresh fruit and vegetable market in the UK had surpassed 70%, and the share of chains among supermarkets had increased to nearly 80%. Consolidation in the retail sector has led to increasing market power for large retail concerns, and much more control by them over production practices. A focus on Maximum Residue Levels (MRLs) of pesticides on fresh produce, and the need to ensure that exports do not exceed these, has led to an increasing emphasis on the traceability of horticultural production; exporters want to be able to trace production back to the specific farm from which it came in order to ensure quality and safe production and handling procedures. Researchers, development practitioners, and governments are concerned that these changes in international supply chains for horticultural and other high-value agricultural products will make it increasingly difficult for smallholders to maintain their position in this trade (Dolan

2

et al. 1999; Dolan & Humphrey, 2001; Dolan & Sutherland, 2002; Harris et al, 2001; Jaffee 2003; Kamau and Sisule 2001). Estimates of changes in Kenyan smallholders’ share of the fresh horticultural export market vary widely. Most researchers seem to agree that shares were as high as 75% in the early 1990s (Harris 1992). The most optimistic current estimate is by Kenya’s Horticultural Crops Development Authority (HCDA), which places smallholder export market shares at 40% for fruit and 70% for vegetables, implying an overall horticultural share of 55-60%. Dolan and Sutherland (2002) provide the lowest estimate. Based on interviews with four leading exporters, they suggest that smallholder shares fell to 18% by 1998 and 11% by 2001. Minot and Ngigi (2003) suggest that this figure is probably too low, based on the small number of firms interviewed and on the tendency of exporters to underestimate smallholder shares “to satisfy European buyers who are suspicious of smallholder quality control.” Minot and Ngigi cite Jaffee (2003) as perhaps the most reliable current source. Based on interviews with several dozen exporters, he estimates smallholder export market shares of 27% for fresh vegetables and 85% for fresh fruit, for an overall horticultural share of 47%. Part of the reason for this much smaller estimated decline in smallholder participation in the export market (compared to Dolan and Sutherland) is that about 60% of Kenya’s fresh horticultural exports are sold, not to UK supermarkets, which have the strictest food safety and quality requirements, but to UK wholesalers and other European countries, whose standards are not as strict. Thus, outright pessimism about continued Kenyan smallholder participation in fresh horticultural export markets does not seem warranted. Yet their share does appear to have fallen substantially over the past 10 years, from about 75% to under 50%. In addition, Kenya’s horticultural export sector as a whole faces increasingly stiff competition from other African countries such as Cote d’ Ivoire, Morocco, Zimbabwe, South Africa and Cameroon. Kenya’s horticultural export expansion has been aided by the country’s preferential duty-free access to EU markets under the Lome Agreement, which currently runs through 2008. If this agreement is not renewed, or if other developing countries obtain similar benefits, Kenya can expect to face even stiffer competition in these markets. Thus, the continued growth of Kenya’s horticultural exports, and the ability of smallholder farmers to participate in any growth that does occur, cannot be taken for granted. Kenya’s economy is also changing, with continued high rates of urbanization expected to drive increases in demand for horticultural products. The rise of supermarkets in Kenya may over time bring similar pressures to the domestic market that we just described for the export market1. Finally, if the new government is able to reverse the country’s economic decline and stimulate private investment to generate renewed growth in per capita incomes, then the increase in domestic demand for horticultural products will accelerate.2 Responding to this growing demand will require increased productivity in both the production and marketing parts of the value chain; if productivity and quality remain low in either part of the chain, poor consumers will be faced with increasing prices, and small farmers may see little effective growth in the demand for their output. All of these factors – the overwhelming dominance of the domestic market, the slower growth experienced in the export sector over the past decade, the challenges that smallholders face to continue participating in the sector, the possibility of more rapid growth in domestic 1 See Weatherspoon, et al (2003) for one of the first attempts to assess the likely impacts of the rise of African supermarkets on smallholder ability to continue supplying domestic horticultural markets. 2 Income elasticities of demand for fruits and vegetables are generally high.

3

demand, and the need for productivity growth in both production and marketing to meet this demand and protect the real incomes of poor consumers – argue for a more active focus on the potentials and constraints that the domestic horticultural market faces in Kenya. A focus on the domestic market implies also the need to assess the competitiveness of local production and marketing against that of neighboring countries such as Tanzania and Uganda. In this paper we propose to explore these key issues. The overall objectives of this study are to provide a broad diagnostic overview of the horticultural sector, to identify specific constraints that limit the system’s performance, to make suggestions for selected policy and programmatic changes, and to identify key research that needs to be done to guide further investments to improve sector performance. The primary focus in this paper is on production and marketing of fruits and vegetables from the farm up to the wholesale level. We exclude flowers. Key issues at the retail level, including the potential emerging role of supermarkets, are examined briefly; new research currently underway will, among other things, provide more detail on the retail sector (both supermarkets and the broader traditional system) and how it relates to the rest of the system. The issues raised above will be addressed by fulfilling the following specific objectives: ! Examine production and yield trends and compare the relative sizes of domestic and

export horticulture in the economy; ! Assess the contribution of domestic horticulture to the livelihoods of rural agricultural

households; ! Determine the share of imports from Tanzania and Uganda in Kenya’s horticultural

markets; ! Investigate the competitiveness of Kenya’s horticultural produce in local and regional

markets; ! Compare the influence of input systems on the performance of the horticultural sectors in

Kenya and Tanzania, and ! Recommend steps that should be taken to place Kenya’s domestic horticulture in a

position to compete favorably in local and regional markets. The paper is organized as follows. Chapter 2 briefly presents the data and methods used in the report. Chapter 3 focuses on production and yield trends for seven fruit and nine vegetable crops, and identifies the international export market share for vegetables. Chapter 4 uses household survey data to examine the role of horticultural production and sales in smallholder livelihood strategies, and evaluates the structure of production and marketing at the farm level. Chapter 5 identifies the structure of horticultural marketing channels, and quantifies the imports from Tanzania and Uganda of two vegetable and two fruit crops. Chapter 6 develops marketing cost budgets for these same four crops and farm budgets for onions. Chapter 7 evaluates the input sector serving horticulture in Kenya and Tanzania, with a special emphasis on lessons from Tanzania’s experience for the horticultural seed sector. Finally, Chapter 8 presents conclusions, recommendations, and suggestions for further research.

4

2. Data and Methods 2.1 Research Design

This study adopted cross-sectional and panel forms of research designs (Frankford-Nachmias and Nachmias, 1996) with exploration of the existing secondary data on fruits and vegetables. Specifically the study focused on the following areas:- ! The status of domestic horticulture in relation to export horticulture. This was done

by obtaining aggregate national data i.e. area, quantities produced and value of production for the two sectors. The production and yield trends of various horticultural crops over the years were established.

! The contribution of domestic horticulture to the livelihoods of rural agricultural

households. This analysis used household level data from a 2000 rural income survey to assess the importance of different horticultural commodities in household production and sales, the structure of production and sales, and the contribution of horticultural production and sales to the income portfolios of rural families by geographical zones.

! Possibility of expanding local and regional markets for domestic horticulture. For

this aspect, four areas were explored. • The size of the domestic market taken up by neighboring countries e.g. Tanzania

and Uganda for commodities locally produced. This effort involved identifying the types of horticultural commodities coming into Kenyan markets, their volumes, prices and destinations. These items were compared with the local commodities in the respective destinations. A purposive sample of 32 collecting wholesalers was used to elicit information on horticultural trade patterns in the main wholesale open air markets in Kenya, i.e. Kongowea in Mombasa and Wakulima market in Nairobi.

• The second aspect examined the competitiveness of Kenya’s horticultural produce in domestic and regional markets. It involved preparation of commodity budgets of collecting wholesalers for a commodity from Kenya and Tanzania/Uganda and comparing them at a common market point e.g. Kongowea market in Mombasa. It also included the socio-economic characteristics and qualitative business aspects of fruits and vegetable retailers at the border point markets.

• The operations at the farm level influence quality, volumes and prices of the commodities at the market. Thus, farm level enterprise budgets for onions were prepared for Kenya and Tanzania.

• One of the factors affecting competitiveness at the regional level is the legal framework in place and how it shapes the structure and operations of any given sub-sector. Since most horticultural output markets in Kenya and Tanzania have been said to be competitive and to operate with very little government influence, this study explored horticultural input systems in both countries. Discussions were held with private and parastatal input companies and Government institutions dealing with vegetable seeds and seedling production in both Kenya and Tanzania. The information sought was in relation to their horticultural input system operations, the current position of the national seed/vegetable seed policy, and development of their planting materials.

5

2.2. Data Sources

To undertake this study, data were obtained from several sources. Between November 2002 and February 2003, secondary data on imports and exports of fresh horticultural produce were obtained from Customs Department records in seven cross-border points: Lunga Lunga, Taveta, Ilasit, Namanga, Isebania, Busia and Malaba. Also, 32 horticultural collecting wholesale traders sourcing produce from both Kenya and Tanzania/Uganda were interviewed in Mombasa at Kongowea market, Nairobi at Wakulima market and in the main market in Taveta. In addition, 51 fruit and vegetable retailers were interviewed using a structured questionnaire. Systematic method of sampling was used to select them. During these interviews, data were collected on various aspects of fresh horticultural trade mainly geared towards determining commodities imported and exported, their origin, destinations, quantities, costs, prices, values and comparison of different origins of a commodity at a common market point. Discussions were held with customs officials at the Kenyan and Tanzania/Ugandan border points on handling of horticultural and broader agricultural trade. Data on horticultural imports and exports collected during border point and city market visits were supplemented by the TAMPA II cross-sectional data set collected through a nation-wide household survey undertaken by Tegemeo Institute in 2000. The household survey covered a wide range of issues related to rural household economies i.e. household characteristics, agricultural production, off-farm activities, and others. The sample covered a total of 1,549 households in 24 Districts, some of which are important in horticultural production. The Districts are grouped into seven agro-ecological zones for this analysis. Further interviews were held with horticultural input suppliers (i.e. seed, fertilizer and chemicals) in Kenya and Tanzania regarding the structure of input systems, sources and prices of inputs, the current policy environment on horticultural inputs and the challenges input suppliers are facing in both countries. Secondary data on various aspects of domestic and export horticulture were gathered from Kenya Revenue Authority, Horticultural Crop Development Authority, Ministry of Agriculture Livestock and Rural Development-Horticulture Division, Ministry of Trade and Industry, Central Bureau of Statistics and various horticultural input suppliers.

6

3. Production Trends for Domestic and International Markets This chapter discusses the current status and trends in production of fruits and vegetables for domestic and export markets in Kenya in the last decade. It relies on official data from Ministry of Agriculture, Land, and Rural Development (MoALRD).3 3.1 Fruit Production

Fruits are grown for generation of food and income, as well as providing raw materials for processing firms. According to MoALRD, the top seven fruits in Kenya in terms of area and total production are bananas, citrus fruits, mangoes, avocados, passion fruits, pineapples and papaws. Each fruit’s proportion of total area and production between 1992 and 2001 are shown in Table 3.1, while production trends are presented in Figures 3.1 and 3.2. Area trends are shown in Appendix A. Table 3.1. Area And Production Shares Of Seven Fruit Crops In Kenya, In 1992

And 2001.

Fruits Area shares Production Shares 1992 2001 1992 2001 Bananas 63 55 58 49 Citrus Fruits 13 11 7 6 Mangoes 10 12 5 8 Avocados 1 3 1 2 Passion fruit 1 2 1 4 Pineapples 6 10 22 28 Pawpaw 4 5 4 4 Other fruits 2 2 2 1

Data Source: MoALRD Production of bananas predominates in total fruit production, with a total area five times its closest rival and total production six times all others except pineapple. Total output of bananas declined from 1992, reached a minimum in 1995 and has been increasing ever since (Figure 2.1). The decline in banana production has been attributed to widespread diseases e.g sigatoka and panama wilts and pests such as banana weevils and nematodes in the late 1980s and early 1990s (Kahangi, 1996). During the same period, tissue culture banana biotechnology research was in progress and a breakthrough was experienced in the early 1990’s. Thus, the increase in production from 1995 has been attributed to the use of disease free biotechnology products adopted by farmers as planting materials. A panoramic view of all fruits over the past 10 years (Figures 2.1-2.3) indicates that the output of pineapples, mangoes, avocados and passion fruits has increased most rapidly. Total growth for various fruits over the 10 year period were: bananas 10%, citrus fruits 13%, pawpaw 25%, pineapples 60%, mangoes 99%, avocados 175%, and passion fruits 200%.

3 Horticultural production data are difficult to collect and quality is thus difficult to determine. A careful assessment of MoALRD methods for estimating area and production would be a useful part of any overall attempt to improve the performance of the domestic horticultural production and marketing system.

7

Figure 3.1. Production of Bananas and Pineapples (Tons) in Kenya (1992-2001)

Figure 3.2. Production of Citrus, Mangoes and Pawpaw fruits (Tons) in Kenya (1992-2001)

-

200,000

400,000

600,000

800,000

1,000,000

1,200,000

1992 1993 1994 1995 1996 1997 1998 1999 2000 2001

Year

Prod

uctio

n (T

ons)

Bananas

Pineaples

-20,00040,00060,00080,000

100,000120,000140,000160,000180,000200,000

1992 1993 1994 1995 1996 1997 1998 1999 2000 2001

Year

Prod

uctio

n (T

ons)

Citrus

Mangoes

Pawpaw

8

Figure 3.3. Production of Avocados, Passion Fruits and Other Fruits in Kenya (1992- 2001)

The production of citrus and other fruits (apples, plums, pears, water melon, grapes, strawberries etc) has been relatively flat. The stagnation in production and low yields of citrus production in Kenya have been attributed to the use of infected planting materials, use of non-budded planting materials, low use of fertilizers and irrigation, inadequate use of chemicals for pest and disease control, and planting of unimproved cultivars of scions and root stocks (Obukosia and Waithaka 2000). As production has stagnated, imports from South Africa, Tanzania, and other countries have met the growing demand. The citrus greening disease has had the greatest adverse effects on orchards in Kenya. The spread of the disease is attributed to a consignment of fruits imported from South Africa in the 1950s when the Government was trying to establish citrus production. Figure 3.4. Yields (Tons/ha) of Bananas, Mangoes and Passion Fruits in Kenya (1992-

2001).

-

1 0 ,0 0 0

2 0 ,0 0 0

3 0 ,0 0 0

4 0 ,0 0 0

5 0 ,0 0 0

6 0 ,0 0 0

1 9 9 2 1 9 9 3 1 9 9 4 1 9 9 5 1 9 9 6 1 9 9 7 1 9 9 8 1 9 9 9 2 0 0 0 2 0 0 1

Y e a r

Prod

uctio

n (T

ons)

A v o c a d o e s

P a s s io n f ru it

O th e r f ru its

0.002.004.006.008.00

10.0012.0014.0016.00

1992 1993 1994 1995 1996 1997 1998 1999 2000 2001

Years

Yiel

ds (T

ons/

ha)

BananasMangoes Passion Fruits

9

Figure 3.5 Yields (Tons/ha) of Avocados, Citrus and Pawpaw in Kenya (1992-2001).

Figure 3.6. Yields (Tons/ha) of Pineapples in Kenya (1992-2001).

The yields of most fruits in Kenya have generally been stagnant with the exception of bananas, mangoes and passion fruits, which have risen (See Figures 3.4 – 3.6). Table 3.2 shows fruit yields in Kenya compared to those in the top five producing countries (by total production) in the world for each crop. Notably, the yield of avocados and pineapples is among the highest in this group, while the yield of mangoes and citrus is in the middle of the pack. The production of mangoes, pineapples, and avocados has been the fastest growing among the seven main fruit crops in Kenya (only passion fruit has shown higher percentage growth in production over the past 10 years). The technology and production system for each fruit differs, however, from country to country. For example, in Kenya production of pineapples is entirely on plantations and it is both a capital- and input intensive system (Jaffee, 1994). The yields would therefore be expected to be higher than those of the countries where the production system is by smallholder farms. Bananas, on the other hand, are produced under much less intensive systems in Kenya than in leading countries like Costa Rica and Egypt, and thus show much lower yields.

0.0010.0020.0030.0040.0050.0060.0070.00

1992 1993 1994 1995 1996 1997 1998 1999 2000 2001

Years

Yiel

d (T

ons/

ha)

0.002.004.006.008.00

10.0012.0014.0016.0018.00

1992 1993 1994 1995 1996 1997 1998 1999 2000 2001

Years

Yiel

d (T

ons/

ha)

AvocadoesCitrusPawpaws

10

Table 3.2. Comparison of Fruit Yield (MT/Ha) Among Top World Producers (by Total Production) and Kenya in 2001.

Bananas Citrus Mangoes Pineapples Avocados

Country Yield Country Yield Country Yield Country Yield Country Yield Kenya 14.6 Kenya 5.8 Kenya 10.9 Kenya 43.8 Kenya 12.2 Costa Rica 47.9 Japan 21.9 Sudan 20.5 Colombia 40.8 Dominican 13.1 Egypt 41.0 China 8.5 Congo 16.5 Phillipines 36.8 Mexico 10.0 Equador 33.0 Nigeria 4.5 China 11.1 USA 36.0 USA 7.8 Mexico 28.9 Guinea 5.1 Brazil 8.0 Thailand 22.4 Brazil 6.9 Brazil 11.7 Guatemala 4.2 Nigeria 5.8 China 22.3 China 5.5

Source: FAO Statistics 2001, www.fao.org 3.2 Vegetable Production

Estimates of annual per capita consumption of vegetables in Kenya are around 20 kilograms in rural areas and 40 kilograms in urban areas (National Development Plan 1994-1996). Cabbages, kales, tomatoes, onions, carrots, French beans, garden peas and traditional vegetables are prominent among the vegetables produced, in terms of area and total output (Table 3.3). However, cabbages, tomatoes and kales have predominated in vegetable production for at least the past decade. The total production trends over the past 10 years are presented in Figures 3.7 to 3.9 below whereas the area trends are shown in Appendix B. Trends in production for most vegetables show a slight increase (Figures 3.7-3.9). However, cabbages show a sharp drop in 1993 and stagnation since that time, while carrots show steady decline in output, with partial recovery in production in 2001. Kales, tomatoes and traditional vegetables show steady increases in output. Table 3.3. Area And Production Shares Of Vegetable Crops In Kenya, In 1992 And

2001 Fruits Area shares Production Shares 1992 2001 1992 2001 Cabbages 25 17 32 22 Kales 21 25 25 31 Tomatoes 17 18 22 24 Onions 6 6 5 5 Carrots 6 4 6 5 French Beans 8 6 2 2 Garden Peas 8 7 2 2 Traditional Vegetables 5 10 3 5 Other Vegetables 4 7 3 4

Data Source: MoALRD

11

Figure 3.7. Production of Cabbages, Tomatoes and Kales in Kenya (1992-2001)

Figure 3.8. Production of Onions, Carrots and Other Vegetables in Kenya (1992-2001).

-

50,000

100,000

150,000

200,000

250,000

300,000

350,000

400,000

1992 1993 1994 1995 1996 1997 1998 1999 2000 2001

Year

Prod

uctio

n (M

T)Cabbages.TomatoesKales.

-10,000

20,00030,00040,00050,000

60,00070,000

1992 1993 1994 1995 1996 1997 1998 1999 2000 2001

Ye ar

Prod

uctio

n (to

ns)

Onions.Carrots.Other veges.

12

Figure 3.9. Production of French Beans, Traditional Vegetables and Garden Peas in Kenya (1992-2001)

The yield of most vegetables has been stagnant (Figures 3.10 to 3.12). Only French beans and indigenous vegetables showed a slight increase. When compared to the top five producers of each crop in the world, the yield of vegetables in Kenya is the lowest in all crops. In most of the developed countries, production of vegetables is highly capital and technology intensive. It is usually characterized by use of newly developed technologies such as fertigation in green houses e.g. in Israel. Hence most weather factors are made controllable, production is evened out through the year and output per unit of land is relatively high. This is in contrast with production conditions in Kenya where it is usually rain-fed and takes place at almost one point in time. Most farmers also lack knowledge and skills on production techniques. This has resulted in low yields compared to other world producers as well as frequently low quality produce. Table 3.4. Vegetable Yields (Mt/ha) Among Top World Producers (by Total

Production) and Kenya in 2001 Cabbages Tomatoes Onions Carrots Green peas

Country Yield Country Yield Country Yield Country Yield Country Yield S. Korea 61.6 Canada 78.2 Japan 46.6 UK 54.1 France 15.0 Japan 40.5 USA 62.5 USA 46.4 Israel 53.6 Cyprus 11.3 Russian 24.0 Italy 52.7 Iran 30.1 USA 37.3 USA 10.4 India 17.9 Morocco 48.9 India 12.8 Poland 27.9 UK 9.8 China 18.9 China 25.8 China 20.8 China 17.8 China 8.1 Kenya 15.6 Kenya 16.7 Kenya 11.2 Kenya 13.6 Kenya 4.0

Source: FAO Statistics 2002, www.fao.org

0

10000

20000

30000

40000

50000

60000

70000

1992 1993 1994 1995 1996 1997 1998 1999 2000 2001Year

Prod

uctio

n (T

ons)

French BeansInd. VegetablesGarden Peas

13

Figure 3.10. Yields (Tons/ha) of Cabbages, Tomatoes and Kales in Kenya (1992-2001)

Figure 3.11. Yields (Tons/ha) of Onions, Carrots and Other in Kenya (1992-2001)

02

468

101214

1618

1992 1993 1994 1995 1996 1997 1998 1999 2000 2001

Years

Yiel

d (T

ons/

ha)

Onions.Carrots.Other vegetables

02468

1012141618

1992 1993 1994 1995 1996 1997 1998 1999 2000 2001

Year

Yiel

d (T

ons/

ha) Cabbages.

TomatoesKales.

14

Figure 3.12. Yields (Tons/ha) of French beans, Indigenous Vegetables and Garden

Peas in Kenya (1992-2001)

3.3 International Export and Domestic Market Shares of Fruits and Vegetables

Fruits and vegetables produced in Kenya can be retained on the farm, or marketed through local fresh markets, local processed markets, fresh export markets, or processed export markets. Establishing what proportion of total production flows through each of these channels is hampered by lack of data, especially on processing, and by definitional issues. In this section we first use data from several sources to estimate the proportion of total vegetable production that is a) consumed on farm, b) marketed locally, and c) exported in fresh or processed form. For comparability, we value all flows at farm-gate prices. Next, we value flows in each channel at final prices in that channel to estimate total value added in each of these channels. Together, these two results provide a picture of the relative importance of local and export markets for Kenya’s horticultural sector. We focus on vegetables for two reasons. First, vegetables appear by all accounts to contribute most to horticultural export earnings. FAOStat data on all fresh and processed horticultural exports (not including flowers) show vegetables with about a 60% share. HCDA data, which are limited to fresh exports, show vegetables with an 80-85% share over the past five years. The difference in these shares is due to the overwhelming importance of canned pineapples and pineapple juice in fruit exports – about 85% of all fruit exports according to FAOStat. Second, pineapple production and exports in Kenya are dominated by Del Monte’s vertically integrated production, processing, and export operation: including Del Monte in fruit calculations would make them less applicable to the typical smallholder or commercial farm, and we lack data to make the calculations accurately without Del Monte. Vegetable production data come from MoALRD, and include smallholder and commercial production. MoALRD values production at “farm gate” prices that it collects. HCDA reports volume and value data for fresh vegetable exports (primarily French beans and Asian vegetables)4. These figures represent all exports regardless of whether they come from

4 “Asian vegetables” include eggplant, chillis, dudhi, karela, okra, and other vegetables used widely in South Asian cooking

0

1

2

3

4

5

6

7

8

1992 1993 1994 1995 1996 1997 1998 1999 2000 2001

Y ears

Yiel

ds (T

ons/

ha)

French B eansInd. V egetablesG arden peas

15

smallholder or commercial farms. HCDA values are based on FOB export prices; we revalued these fresh vegetable exports using MoALRD farm-gate prices to make the production and export figures comparable. This exercise shows that fresh vegetable exports rose from 4-5% of total vegetable production in the early 1990s to over 12% in 2000, before falling to about 7% in 2001 (Figure 3.13). The trend is clearly positive; the lower figure in 2001 is slightly higher than those for 1996 and 1997 and well above those of the early 1990s. Over the past 5 years (1997-2001), fresh vegetable exports averaged 9.3% of production, by value. Adding processed vegetable exports, which FAOStat data show to be about 1/3 as much, by value, as fresh exports, raises the total export share (processed plus fresh) of vegetables in Kenya between 1997 and 2001 to about 12%. In the absence of more detailed data, this final calculation assumes that the mix of processed vegetable exports is comparable to fresh, and that export prices for fresh and processed are also comparable. As a final step to calculate market channel shares, we use data from the 2000 Tegemeo/MSU Tampa smallholder income survey which show that 64% of total vegetable production that year was sold, and 36% retained on farm. This calculation provides a lower bound for marketed share if we assume, as is reasonable, that commercial producers sell nearly all their production.5 By combining all these data, we arrive at Figure 3.14, showing that the value of vegetable production sold and then consumed domestically over the past five years has been at least four-to-five times as large as the value exported in fresh and processed form (52% compared to 12%). Value added per unit of farm-gate production is higher in the export sector due primarily to higher quality and health standards. Comparing MoALRD farm-gate prices with HCDA export prices for French beans and Asian vegetables shows that export prices of these vegetables have exceeded farm-gate prices by a factor ranging from 2.7 to 6.2 since 1992, with an average of 3.9, or 290%. In contrast, mark-ups in domestic markets are typically about 100% from farm-gate to collecting wholesaler sales, and an additional 20-25% to retail.6 These figures imply a 150% total markup from farm-gate to retail in local markets. Applying these markup figures to the share of production flowing through the domestic and export channels, and continuing to value unsold production at farm-gate prices, shows that total value added in domestic vegetable markets is nearly three times that in vegetable export markets (Figure 3.15). These calculations show two things. First, vegetable exports are an extremely important component of the vegetable supply chain, absorbing about 20% of all sold production by value, and accounting for about one-quarter of all value added after the farm gate. Second, domestic markets nonetheless remain the primary outlet for vegetable production and generate much more value added than do export markets. This conclusion will hold even more for fruit, which has a higher total value of production and lower value of exports.

5 Unfortunately, MoALRD does not report production separately for smallholder and commercial farmers. This makes it impossible to calculate a more accurate marketed surplus figure. 6 See Tables 6.2, 6.4, 6.6., 6.8, and 6.10 for farm-gate to collecting wholesale markups. Mark-ups from collecting wholesale to retail are based on data collected in Wakulima market in November 2003. See Appendix XX for the original price data.

16

Figure 3.13. Fresh Vegetable Exports as Share of Total Production, by Value (1992-2001)

0.000

2.000

4.000

6.000

8.000

10.000

12.000

14.000

1992 1993 1994 1995 1996 1997 1998 1999 2000 2001

Expo

rt Sh

are,

% o

f Tot

al P

rod'

n

Figure 3.14. Market channel shares (farm, local sales, export sales) of total vegetable

production in Kenya, 1997-2001, valued at farm-gate prices

Production Ksh 14.5B

Retained on Farm

Ksh5.2B 36%

Sold then Consumed

Domestically

Ksh7.5B (52%)

Sold and Exported

Ksh1.7B (12%)

Source: Derived from Tegemeo/MSU 2000 household survey data, production data from MoALRD, and export data from HCDA

17

Figure 3.15. Total value added (and share) in farm, local sales, and export sales channels for vegetables in Kenya, 1997-2001

Value of Production Ksh 14.5B

Farm-gate Value, Retained on Farm

Ksh 5.2B

(17%)

Retail Value, Domestic Sales

Ksh18.8B

(61%)

FOB Value, Export Sales

Ksh6.8B

(22%)

Source: Derived from Tegemeo/MSU 2000 household survey data, production data from MoALRD, and export data from HCDA

18

4. Fruits and Vegetables in Rural Household Livelihoods Rural households in Kenya participate in a wide variety of economic activities to ensure their consumption and increase their incomes. This chapter uses data from the 2000 Tegemeo/ MSU Rural Household Survey to evaluate the role that horticultural production and sales play in rural livelihoods. The sample covered 1559 smallholder households in the relatively high potential areas of 24 districts. We examine seven areas of the country that were sampled in this survey, as show in Table 4.1. Surveyed areas not included in this analysis due to low sample size include Northern Arid and Marginal Rain Shadow. Table 4.1. Districts And Sample Sizes By Zone In “High Potential” And “Low

Potential” Samples Zone Districts Sample

Size Coastal Lowlands Kilifi and Kwale 80

Eastern Lowlands Taita Taveta, Kitui, Machakos, Makueni and Muingi 170

Western Lowlands Kisumu, Siaya 188

Western Transitional Bungoma and Kakamega 171

High Potential Maize Zone

Bungoma (higher elevations), Kakamega (higher elevations), Bomet, Nakuru, Narok, Trans Nzoia, Uasin Gishu

420

Western Highlands Kisii, Vihiga 163

Central Highlands Meru, Muranga, Nyeri 268

4.1 Most Widely Grown and Sold Horticultural Crops

Table 4.2 shows that, throughout areas of Kenya where cropping is practiced, nearly all households grow horticultural crops. The partial exception to this pattern is in areas of the Western Lowlands sampled in the 2000 survey. Western Lowlands also has the lowest mean production value among those growing, the lowest percentage of households selling, and the lowest mean sales value among those selling. Eastern Lowlands, Western Highlands, and Central Highlands stand out for high production values, high proportions of households selling, and high mean sales values among those selling. Tables 4.3 and 4.4 illustrate the importance of different horticultural crops in production and sales, by zone. Table 4.3 presents the most commonly grown and sold crops, while Table 4.4 presents those crops that are most valuable in production and sales. Table 4.3 highlights the importance of bananas throughout the country. In five of the seven zones, this crop is the most widely grown, and is second in one zone; only in the Western Lowlands is it not among the top three. In five of the zones it is the first or second most widely sold. Sukuma wiki also stands out, being among the three most widely grown in four of the zones and among the three most widely sold in all but one zone (Coastal Lowlands). Other widely grown crops include mangoes, pawpaw, and avocado. On the sales side, mangoes are among the three most widely sold in the Coastal and Eastern Lowland zones, while cabbage takes third place in two of the three higher elevation zones. Avocado takes second place in both Western and

19

Central Highlands. Surprisingly, tomatoes appear only once in the table, as the second most widely sold crop in the Coastal Lowlands. Table 4.2. Percent Of Households Growing And Selling Horticultural Crops, And

Average Value Of Horticultural Production And Sales, By Geographical Area In Selected “High Potential” Areas Of Kenya

Geographical Area % of Households Growing

Mean value of production among

those growing

% of Households

Selling

Mean value of sales among those selling

Coastal Lowlands 96.2 18,614 65.8 10,386

Eastern Lowlands 99.4 27,762 80.7 18,577

Western Lowlands 82.5 4,898 51.4 3,879

Western Transitional 100.0 13,972 87.3 7,328

High Potential Maize Zone 97.5 10,001 70.7 7,778

Western Highlands 100.0 19,730 90.1 12,673

Central Highlands 100.0 21,349 83.0 16,148

Table 4.3. Most Widely Grown And Sold Horticultural Crops, By Geographical

Area In Selected “High Potential” Areas Of Kenya Most widely grown Most widely sold

Geographical Area First Second Third First Second Third

-------------- % growing -------------- ------------- % selling ---------------

Coastal Lowlands Bananas (76)

Cowpea lvs (75)

Coconut (73)

Coconut (49)

Tomatoes (20)

Mangoes/ Lemons (14)

Eastern Lowlands Bananas (78)

Mangoes (64)

Pumpkin (61)

Bananas (34)

Suk. Wiki (32)

Mangoes (32)

Western Lowlands Cowpea lvs. (43)

Pawpaw (43)

Mangoes (41)

Mangoes (18)

Bananas (18)

Suk. Wiki (14)

Western Transitional Bananas (90)

Suk. Wiki (77)

Pawpaw (54)

Bananas (56)

Suk. Wiki (48)

Cowpea lvs.(23)

High Potential Maize Zone

Suk. Wiki (77)

Bananas (58)

Indig. Veg. (52)

Suk. Wiki (46)

Bananas (22)

Cabbage (21)

Western Highlands Bananas (94)

Avocado (77)

Suk. Wiki (68)

Bananas (63)

Avocado (52)

Suk. Wiki (44)

Central Highlands Bananas (91)

Suk. Wiki (82)

Avocado (71)

Suk. Wiki (41)

Avocado (35)

Cabbage (31)

Table 4.4 shows the three most valuable crops in production and sales, and their contribution to the total value of production and sales in each zone. Bananas maintain their importance in both production and sales, while sukuma wiki, being one of the lowest priced crops, stays among the top three in both production and sales in only two zones. Tomatoes appear more often, twice among the most valuable in production and three times among the most valuable in sales.

20

Table 4.4. Most Valuable Horticultural Crops In Production And Sales, By Geographical Area In Selected “High Potential” Areas Of Kenya

Most valuable in production Most valuable in sales Geographical Area

First Second Third First Second Third

------- % of total prodn value ------- ------- % of total sales value -------

Coastal Lowlands Coconut (30)

Mangoes (26)

Bananas (15)

Coconut (54)

Mangoes (20)

Bananas (16)

Eastern Lowlands Bananas (22)

Avocado (14)

Cabbage (9)

Bananas (18)

Avocado (13)

Macadamia Nuts (12)

Western Lowlands Bananas (27)

Mangoes (13)

Pawpaws (12)

Bananas (26)

Sugarcane (14)

Mangoes (13)

Western Transitional Bananas (51)

Suk. Wiki (10)

Pineapple (7)

Bananas (47)

Suk. Wiki (14)

Tomatoes (10)

High Potential Maize Zone

Bananas (19)

Pumpkin (14)

Tomatoes (14)

Tomatoes (21)

Bananas (16)

Cabbage (13)

Western Highlands Bananas (63)

Suk. Wiki (9)

Tomatoes (8)

Bananas (58)

Tomatoes (12)

Suk. Wiki (11)

Central Highlands Cabbage (27)

Bananas (23)

Carrots (10)

Cabbage (38)

Carrots (14)

Bananas (12)

Note: % of total production and sales values are within zones. 4.2. Income Share Analysis

Income shares indicate the proportion of total household income that comes from different types of economic activities, and thus reveal the importance of different types of activities in smallholder income and food security strategies. Income includes the value of all crop and livestock production, even if retained on farm, plus off-farm incomes and remittances. Table 4.5 shows mean household income shares of eight different economic activities during 2000. Off-farm labor (informal business and wages plus formal salaries and remittances) is a key contributor to overall income in all zones, with shares ranging from 27-29% in Western and Central Highlands to 60% in Coastal Lowlands. Horticultural income shares are highest in Western Highlands at 26%, even though total value of horticultural production and sales in this area are below those in Eastern Lowlands and Central Highlands (Table 4.2); the discrepancy is due to very low overall income levels in the Western Highlands. Other zones with relatively high horticultural shares are Coastal Lowlands and Eastern Lowlands. Only in Western Highlands does the horticultural share exceed that of cereals, tubers, and pulses, but horticulture exceeds livestock in four of the seven zones and exceeds industrial crops in five of the seven. Horticultural sales are equivalent to or greater than sales of cereals, pulses, and tubers in six of the seven zones, reflecting the fact that households tend to sell a greater proportion of their horticultural production than they do their staples. 4.3. Concentration of Horticultural Production and Sales

The previous results suggest that horticultural production and sales are an important but not predominant contributor to income for the average household. Yet averages can hide a great deal of variability. Examining levels of concentration of horticultural sales will allow us to peer behind these averages and assess differences in the role that horticultural production and sales play across households. In Table 4.6, we break all households in our sample into seven groups: households that did not produce any horticultural crops, households that produced but

21

did not sell, and five groups of equal size (quintiles) among those that did sell horticultural produce, ranging from those that sold the least (quintile 1) to those that sold the most (quintile 5). Results suggest that horticultural sales are highly concentrated in Kenya. Only 3% of households in the sample did not produce any horticultural crops and therefore also did not sell, but over 20% of those that did produce did not sell any. Among those that did sell, one-fifth (15% of the total population) sold extremely small amounts – only Ksh 234 on average. Nearly 80% of the total value of horticultural sales among all households was accounted for by the largest 20% of all sellers, who represent only 15% of the total population. Table 4.5. Income Shares By Geographical Area In Selected Areas Of Kenya

Geographical Area Cereals, tubers, pulses

Horticultural crops

Retained Sold Retained Sold

Indus-trial crop

sales

Live-stock

In-formal off-farm (wages+ business)

Salary and

Remit-tances

Total (%)

Total (Ksh per

capita)Coastal Lowlands 19.7 1.6 11.5 4.1 0.0 3.2 37.3 22.6 100 13,493

Eastern Lowlands 18.9 2.9 10.2 8.4 0.7 12.7 20.5 25.6 100 17,006

Western Lowlands 17.4 3.8 6.6 3.7 4.7 22.7 18.5 22.6 100 7,321

Western Transitional 15.7 4.5 5.8 5.0 33.2 7.3 16.0 12.5 100 17,865

High Pot’ Maize Zone 13.4 14.3 3.8 3.6 3.9 26.1 19.1 15.8 100 20,847

Western Highlands 20.8 3.4 12.4 13.9 16.0 6.6 8.3 18.6 100 12,716

Central Highlands 8.2 3.5 0.5 6.9 29.1 23.9 12.2 16.7 101 28,501

Table 4.6. Concentration Of Horticultural Sales: Percent Of Total Sales By

Quintiles Of Total Household Horticultural Sales Value Sales category % of

farmers Average value of

horticultural production per hh

(Ksh)

% of total prodn in sample

Average value of horticultural

sales per hh (Ksh)

% of total sales in sample

No production 3.2 — — — ---

Production, no sales 21.2 3,911 5% 0 0%

1 Lowest sales 15.1 3,475 3% 234 0%

2 15.1 5,927 6% 1,112 2%

3 15.2 8,953 9% 2,807 5%

4 15.1 15,496 15% 7,850 14%

5 Highest sales 15.1 61,995 61% 43,980 79%

Table 4.7 examines how these households are distributed across our seven zones. We see that Western Lowlands has by far the highest proportion of non-producers, and that it along with Coastal Lowlands has the highest proportion of non-sellers and the lowest proportion of sellers in the top sales quintile. Central Highlands, Eastern Lowlands, and Western Highlands stand out for their relatively large proportion of large horticultural sellers.

22

Table 4.7. Concentration Of Horticultural Sales: Distribution Of Households Across Quintiles Of Total Household Horticultural Sales Value, By Zone

Zones Sales category

National

Coastal Low-lands

East. Low-lands

West. Low-lands

Western Tran-

sitional

High Potential

Maize Zone

West. High-lands

Cent. High-lands

------------------------------ % of hhs in each category ------------------------------

No production 3.2 3.8 0.6 17.0 0.0 2.8 0.0 0.0

Production, no sales 21.2 30.4 18.6 31.3 12.7 26.6 9.9 17.0

1 Lowest sales q’tile 15.1 17.7 9.9 17.0 13.9 18.3 11.9 13.9

2 15.1 17.7 12.4 13.6 19.3 16.0 14.6 13.1

3 15.2 13.9 16.1 11.4 19.3 13.0 21.2 14.7

4 15.1 8.9 18.0 6.8 24.1 13.3 20.5 14.7

5 Highest sales q’tile 15.1 7.6 24.2 2.8 10.8 10.0 21.9 26.6

Table 4.8 examines the characteristics of households in each of these horticultural sales categories. Several patterns emerge. First, the mean level of education of the head of household steadily rises as the value of horticultural sales increases. The total increase in mean years of education of the household head over the categories is 2.2 years. Second, the largest horticultural sellers appear less likely than other households to be headed by females, and non producers of horticultural crops appear more likely to be female-headed. Third, cropped area also rises steadily through the quintiles of sellers, though producing non-sellers actually crop more total area than do the sellers in the bottom two sales quintiles. Non-producers have the lowest mean area. Fourth, horticultural sellers in the top sales quintile are clearly better-off than other households: total value of assets, total per capita income, and total per capita cash income all rise substantially in this last group. The fact that assets, and not just income, are higher in this group suggests that those households with high horticultural sales during the survey period have enjoyed higher incomes for some time, which they have used to increase their asset holdings. Non-producers are clearly worse-off than other households, with the lowest amount of cropped area, lowest per capita incomes, and lowest assets. Fifth, households that produce but do not sell horticultural crops are virtually indistinguishable from all but the largest sellers; the incomes and assets of the former are comparable to sellers in the top four sales quintiles. This suggests that these households have chosen to earn most of their cash incomes not from horticulture but from other agricultural and non-agricultural activities. Finally, the share of off-farm income in total income falls steadily as horticultural sales increase, being replaced largely by horticultural sales. It should be noted, however, that the actual value of off-farm income is highest among households in the top horticultural sales category, despite the off-farm income share being the lowest in this group. We close this section by examining the concentration of horticultural production and sales – both geographically and at the household level – by individual crop. Table 4.9 presents indicators of geographical and household level concentration for the top ten crops, based on

23

total sales value in our sample. We see that a majority of these 10 crops (banana, cabbage, tomato, sukuma wiki, avocado, onion, and orange) are sold in at least six of the seven zones, and are produced in all seven. Of these, sukuma wiki appears to be the least concentrated geographically, with the top zone (High Potential Maize Zone) accounting for less than 30% of total sales and just over 30% of total production. Orange, on the other hand, while produced in all seven zones and sold in six of them, shows greater geographical concentration, with about 50% of production and nearly 60% of all sales coming from the High Potential Maize Zone. Carrots are the most concentrated in sales, with 94% of production and 96% of sales coming from the Central Highlands. Nearly 80% of Macadamia Nut sales come from the Eastern Lowlands, and this crop is produced in only two of our seven zones. Table 4.6 already showed that the concentration of sales at the household level is high, with nearly 80% of sales coming from only 15% of the rural households. Table 4.9 shows that production and sales of sukuma wiki and banana are substantially less concentrated than other crops, while carrot, french bean, macadamia nut and orange show the highest concentration. Among all these latter crops, 5% of rural households account for nearly all sales, and for 90-100% of production. 4.4 Summary

This review has shown that the production and sales of all the major horticultural crops in Kenya is quite concentrated. In all but one of the top 10 crops, 5% of the rural population accounts for at least 50% of production and at least 70% of sales. Because the larger producers and sellers tend to specialize in one or two crops7, concentration at the crop level is higher than it is across all crops; in the latter case, 15% of rural households account for about 80% of all horticultural sales. Nevertheless, useful distinctions can be made between crops. Bananas and Sukuma wiki are the least concentrated both geographically and at the household level. Each is produced throughout the country and is actively marketed; in most areas at least one-third of all rural households sell these crops. Improvements in production and marketing of these two crops would have the broadest impacts on income levels and poverty rates. Returns to research and general extension on banana diseases may therefore be very high. Maintenance and expansion of use of disease free planting material, which was a major success of the mid-1990s period, is especially important. Key constraints to sukuma wiki production and marketing need to be better understood. Carrots, French beans, macadamia nuts and oranges are the most concentrated in production and sales. Five percent of the rural population accounts for nearly all sales of these crops, and except for oranges, they are produced in four or fewer of our seven zones. These characteristics suggest that a strategy of focused assistance to relatively few growers on production and marketing constraints could be effective in boosting production and sales. Such a strategy would likely not be effective in oranges unless the serious disease problems in that crop are first addressed (see Chapter 6 for more detail on citrus greening disease). Broader income gains through such focused activities would be achieved through re-spending of incomes by the relatively few direct beneficiaries. 7 The top quintile sells an average of six horticultural crops, but the most important of these accounts for 64% of the households’horticultural sales, on average.

24

Cabbage, tomato, avocado and onion fall between these two other groups in terms of concentration of sales. Each is produced in at least six of our seven zones, and 5% of the rural population accounts for 81-88% of sales. The review also showed that the largest horticultural sellers are better off than other households based on a wide range or indicators: they have more education, crop more land, are more likely to use fertilizer, have higher assets and incomes, and are less likely to be female-headed than other households. Interestingly, this group still earns a slightly higher income share off the farm than they do through horticultural sales, though their off-farm share is lower and horticultural sales share is higher than any other group. This pattern suggests that these households are still relatively diversified in their income strategies, as is typical of African smallholders. A potential implication is that, if marketing costs can be reduced, farm level productivity increased, and market outlets be made more reliable, this group of households may be able to specialize substantially more in horticulture and thus be well poised to take advantage of expanding market opportunities.

25

Table 4.8. Concentration Of Horticultural Sales: Selected Household Level Indicators By Quintiles Of Total Household Horticultural Sales Value

Horticultural Sales Category

% of farmers

Mean education of head of hh

(years)

% female headed

households

Cropped area (acres,

Main season)

% Using fertilizer on farm

Total value of assets

Total per capita income

(Ksh)

Total per capita cash

income (Ksh)

Off-farm income share

Horti-cultural

sales share

No production 3.2 5.4 26.7 3.4 52 74,700 10,562 8,332 54% 0%

Production, no sales 21.2 5.4 18.0 4.9 39 145,411 16,309 11,392 41% 0%

1 Lowest sales quintile 15.1 5.5 16.7 3.6 49 97,065 14,140 10,453 40% 1%

2 15.1 6.4 11.4 3.8 55 169,232 15,616 10,700 35% 2%

3 15.2 5.9 15.2 5.0 60 104,799 15,747 10,834 34% 4%

4 15.1 6.6 15.7 4.8 68 114,824 18,819 13,550 34% 12%

5 Highest sales quintile 15.1 7.4 8.1 5.9 83 204,038 32,611 23,192 24% 22%

26

Table 4.9. Concentration Of Horticultural Production And Sales By Crop: Indicators Of Geographical And Household Level Concentration Of Top 10 Horticultural Crops By Sales Value

Zone Banana Cabbage Tomato Suk. Wiki Avocado Onion Carrots French Beans

Maca-damia Nuts

Orange

Geographical Sales Indicators

Sales share of top zone 33.9 67.8 41.9 29.3 47.4 48.1 96.2 46.8 78.9 57.9

Name of top sales zone W. High-lands

W. High-lands

High Pot. Maize

High Pot. Maize

E. Low-lands

Central Highlands

Central Highlands

E. Low-lands

E. Low-lands

High Pot. Maize

Number of zones selling 7 6 7 7 6 6 4 4 2 6

Geographical Prod’n Indicators

Production share of top zone 28.3 62.1 38.6 31.5 44.2 41.0 94.0 46.0 74.0 50.2

Name of top production zone same same same same Same same same same same same

Number of zones producing 7 7 7 7 7 7 6 4 2 7

HH Level Sales Indicators

National Gini Coefficient, sales 0.909 0.951 0.949 0.869 0.948 0.961 0.978 0.982 0.997 0.983

National sales share, top 5% 69.7 83.6 81.4 56.5 81.0 87.7 98.1 100.0 100.0 99.1

HH Level Prod’n Indicators

National Gini Coefficient, prod’n 0.795 0.936 0.923 0.769 0.907 0.921 0.972 0.980 0.996 0.967

National prod’n share, top 5% 50.1 79.0 73.4 45.2 71.0 74.4 94.9 100.0 100.0 90.9

Marketed Surplus, % 43.9 81.2 72.4 58.8 50.0 72.1 84.7 92.3 92.4 65.3

27

5. Domestic and Regional Marketing Channels and Product Flows of Fresh Horticultural Produce Various regional cooperation initiatives have been put in place to boost intra-Africa trade, including the East African Community (EAC), the Southern African Development Community (SADC) the Inter-Governmental Authority on Development (IGAD), and the Common Market for Eastern and Southern Africa (COMESA). Expectations upon formation of these groups were that member countries would take advantage of these cooperation efforts to increase their trade within regional markets. Yet such trade remains low in both physical and value terms. For example, in 1992 the value of total Preferential Trade Area (PTA) country exports was $ 12.453 billion, and only $ 826 million (about 7%) was to member states. The total value of PTA imports was $17.496 billion of which $ 826 million (5%) was from member states (Preferential Trade Area, 1994). Horticulture has been identified as one of the main commodity areas with potential for increasing rural household incomes in Kenya. Hence, increasing the volume and value of traded fresh horticultural produce among rural agricultural households as well as between them and the rest of the domestic, regional and international economy would be in line with the objective of increasing rural incomes and reducing poverty. The purpose of this section therefore is to assess the magnitude of trade in fresh horticultural produce between Kenya, Tanzania, and Uganda through the main border points. First we examine the marketing channels that domestic and imported produce flow through. 5.1 Marketing Channels

Government assistance to the horticultural sector has been concentrated primarily on the export market; public influence on the domestic market is seen primarily through construction of public markets, which primarily serve urban areas and are dominated by the horticultural trade, and maintenance of roads. Concerns about traffic congestion and lack of hygiene in public markets have become increasingly pressing in recent years, while poor road infrastructure has imposed high costs on the marketing of all agricultural products. Thus, the size of the urban population, the degree of self sufficiency of rural households, the purchasing power of urban and rural households, and the costs of collecting, transporting, and selling horticultural products are the key determinants of the size of the horticultural market for Kenya smallholders. Figure 5.1 shows the various local, regional and international marketing channels for horticultural produce in Kenya, emphasizing the actors involved in the process. Figure 5.2 focuses on the principal physical market places through which domestic horticultural products pass8. The export market is served by a few large–scale own company farms, an increasing number of contracted commercial horticultural farms, and a declining but still significant number of contracted smallholder farms (Dijkstra and Magori, 1995). Independent smallholders produce the bulk of the vegetables and fruits for domestic markets. 8 The size of the boxes in both figures should not be taken to indicate the relative size of various channels or actors. The figures attempt only to show the types of channels and actors involved, and relations among them. So, for example, neither figure should not be taken to imply that rural consumption is substantially less than urban consumption.

28

Figure 5.1. Domestic, Regional and International Marketing Channels for Fresh Horticultural Produce in Kenya

T

Farmers and traders in Tanzania,

Uganda, Ethiopia etc.

Regional Fresh markets

Rural-Urban Wholesalers and Brokers

Independent Smallholder

Farmers

Commercial Farmers

Contracted Smallholder

Farmers

Company Farms

Rural Retail

Centers

Open Air Retail

Markets

Urban Kenya

Kiosks (mid-class

Green Grocers)

High-end Green

Grocers

Super-markets

Hotels Inter-national markets

(EU)

Rural Assemblers &

Purchasing Agents

International Exporters (Fresh and Processed)

Local Processors

Rural Con-

sumers

Kenyan Urban Consumers

Inter-national Consum

29

The main traders in the regional markets are the wholesalers. Wholesalers as a group are divided into collecting wholesalers and distributing wholesalers. The former specialize in collecting produce from farmers in the region. They travel long distances to purchase commodities in spot markets from the producing areas and towns in Kenya, Tanzania, and Uganda. To facilitate operation, collecting wholesalers frequently employ purchasing agents who work in the production areas on their behalf. Purchasing agents reduce costs by identifying produce for sale, carrying out the negotiations, accumulating, assembling and carrying the produce to a nearby earth road for ease of collection. Hence, they streamline the procurement process (Dijkstra, 1996; 1997; 1999). Once enough product is obtained, collecting wholesalers then transport the commodities to the main cities/towns generally using lorries with a minimum of seven tons. These professional collecting wholesalers sell primarily in urban wholesale markets to distributing wholesalers9. For example, oranges are fetched from Tanga and Morogoro in Tanzania and sold in Moshi, Arusha, Mombasa, Nairobi and Kisumu; onions are obtained from Arusha/Mang’ola in Tazania and sold in Mombasa, Dar es Salaam and Nairobi. Coconuts are obtained from Mombasa and sold in Dar es Salaam, Moshi, Arusha, Nairobi and Kisumu. Bananas are obtained from Mbale in Uganda and and Kisii in Kenya and sold in Nairobi. Collecting wholesalers operate in such a way as to allow distributing wholesalers to focus entirely on their urban clientele. This is important in large regional urban centers such as Nairobi, Mombasa, Dar es Salaam and Kampala where wholesale and retail markets are operational six days a week. For such distributing wholesalers, being absent results in lost revenue and poor customer relations (Dikistra, 1997). The urban clientele that these distributing wholesalers serve are highly diverse. They include traders in traditional open-air retail markets, green grocers serving middle-class clientele in roadside kiosks, high-end green grocers mostly in established retail centers, supermarkets, and hotels. Supermarkets have expanded their participation in horticultural markets a great deal over the past three years, but their market share remains low (see section 5.2). The two major chains – Uchumi and Nakumatt – each carry upwards of 80 horticultural products in the produce section of their Nairobi stores, including fresh whole produce from Kenya, imported produce, and prepared vegetables ready for cooking. Each has ambitious expansion plans, with Uchumi planning to reach 50 stores within five years from 30 currently (Weatherspoon et al, 2003). Though quantitative data on the flow of produce through market channels is not easily available, it is clear that urban wholesale market places continue to play a key role in the domestic horticultural marketing system as the dominant source of supply for open-air retail markets and kiosks. The two largest supermarkets, however, appear to have completely by-passed these markets, relying instead primarily on brokers and secondarily on direct procurement with an assortment of contracted commercial farmers and some organized small- and medium-sized farmers (Weatherspoon et al). While brokers may obtain some of their produce in wholesale markets, it appears that the largest supermarket chains intend to phase out brokers over the next five years as they develop their “preferred grower” programs. At the present time, traditional wholesale markets are unattractive to buyers concerned with improving quality and food safety while reducing cost. Traffic congestion is often extreme, hygiene is poor, and no formal grading is practiced. A key question for the horticultural

9 Collecting wholesalers do also sell directly to professional retailers in the market. The distinction between the various actors is to some extent artificial because at the end of the day wholesalers often sale produce that they have left over directly to consumers, thus taking the role of retailers. The trading system is very flexible.

30

marketing system is the current market share and rate of growth of marketing channels that bypass these traditional wholesale markets. Such information is crucial for understanding the likely evolution of and constraints to smallholder participation in the domestic market, and for designing investment programs to facilitate their continued participation while reducing marketing costs. Figure 5.2. Principal flows of horticultural products through physical market places

in Kenya

T

Farms and assembly markets

in Tanzania, Uganda, Ethiopia

etc.

Urban Wholesale Markets

Smallholder Farms

Commercial Farms

Rural Retail

Centers

Open Air Retail

Markets

Urban Kenya

Kiosks (middle-

class Green Grocers)

High-end Green Grocers

Supermarkets Hotels

Rural Assembly Markets

Processing Plants

31

5.2. Price Relationships and Likely Retail Market Shares

Berdegué, et al, in a review of the rise of supermarkets in Central America, show that the supermarket share of the fresh fruit and vegetable (FFV) market lags well behind their penetration of the overall food market. While supermarkets’ overall share in food retail is 36% in the region, their share of the FFV market is only 10%. Similar patterns are found in South America. The authors suggest that the key reason for the lagging FFV share of supermarkets is that they offer similar quality produce for substantially higher prices. The primary advantages of supermarkets were convenience, safety, and cleanliness. Supermarkets’ move in that region (and throughout the developing world) towards preferred buyer programs and centralized procurement and distribution are attempts to reduce costs while maintaining or improving quality, thus enabling them to capture more of the FFV market. The rate at which these firms will be able to capture FFV market share, and the impact that key public and private investments in the traditional marketing system can have in maintaining its competitiveness against supermarkets, are key policy issues in Kenya and other developing countries. We will return to these issues in Chapter 8. For now, we focus on establishing the likely current FFV market share of supermarkets in Kenya by examining price relationships for FFV among alternative retail outlets in Nairobi (Table 5.1). In late April, 2003, researchers collected price data on all FFV items offered for sale in two supermarkets, three randomly selected City Market stalls, one high-end green grocer, and one typical roadside kiosk in Nairobi. The City Market tends toward the higher end of the “traditional marketing system”, while roadside kiosks primarily serve Nairobi’s middle- and lower-middle classes. Traditional open air markets serving the city’s poor were not visited. A more definitive assessment of price relationships among outlets will require collection over a longer time period and inclusion of open-air markets. However, in the absence of data on volumes flowing through each retail channel, these price relationships can generate some insight into how far supermarkets may have penetrated the FFV market in Nairobi. The table first presents information on all items offered by each outlet, then information only on the 39 items found in at least one supermarket and at least one City Market Stall (Subset 1), and finally on the 13 items found in at least one supermarket, at least one City Market stall, and the roadside kiosk. Four key patterns emerge. First, supermarkets and high-end green grocers have much greater variety than the roadside kiosk, with about five times more items on offer. Second, supermarkets and high-end green grocers offer more high-priced imported and semi-processed items than either kiosks or the City Market stalls. Kiosks offer few or none: the highest priced item in the kiosk was one-half to one-eighth the highest price item in the other outlets. Third, supermarkets (and green grocers) typically, but not always, charge higher prices than other outlets on the same items. In the head-to-head comparison with City Market stalls (subset 1), mean supermarket prices were about 13% higher, but Uchumi had the lowest (or tied for the lowest) price on 11 of their 32 items, while Nakumatt was lowest on four of their 32.10 In the head-to-head comparison among supermarkets, City Market stalls, and the roadside kiosk, the kiosk emerges as the lowest price option; mean kiosk prices on the 13 items were 35% below supermarkets and 26% below City Market stalls, and the kiosk had the lowest price for seven of the 13 items compared. Supermarkets were the lowest price option in only 2 cases. Based on clientele served, we would expect that head-to-head prices in open air markets would be lower than in kiosks. 10 While each supermarket individually had only 32 of the 39 items, together they had all 39.

32

Table 5.1. Product Availability And Prices Of Fresh Fruits And Vegetables In

Various Retail Outlets, Nairobi, 21-28 April 2003

Super-markets Three City

Market stalls

High-end Green Grocer

Roadside Kiosk

Uchumi Nakumatt All Items Total number of items offered 71 79 47 91 18 Mean price on all items (Ksh/kg) 91 94 66 87 38 Maximum price over all items (Ksh/kg) 540 399 680 320 80 Subset 1 Number of items offered 32 32 39 31 13 Mean price on these items (Ksh/kg) 69 70 60 72 40 Minimum price frequency 11/32 4/32 18/39 6/31 7/13 Subset 2 Number of items offered 10 12 13 9 13 Mean price on these items (Ksh/kg) 63 66 54 57 40 Minimum price frequency 0/10 2/12 5/13 1/9 7/13

Notes: 1) Two supermarkets: Uchumi Hyper on Ngong Road and Nakumatt Mega on Uhuru Highway, 2) Three randomly selected City Market stalls, 3) High-end green grocer is The Corner Shop in YaYa Center, 4) One roadside Kiosk on ??? road., 5) Subset 1 is the 39 items offered in at least one supermarket and at least one City Market stall, 6) Subset 2 is the 13 items offered in at least one supermarket, at least one City Market stall, and the roadside kiosk These price patterns are similar to those found in Central America by Berdegué, et al, where supermarket prices were 15-60% higher than those among traditional retailers. This, plus the fact that supermarket penetration of the FFV market started later in Kenya than it did in Central America, suggests that the supermarket share of the FFV market in Nairobi is likely below the 10% share found in Central America11. Outside of Nairobi, it would certainly be lower still. There is some evidence that Uchumi is attempting most aggressively to penetrate this market: in the head-to-head comparison with City Market, one-third of Uchumi’s items were the lowest price, and the company is aggressively expanding its number of stores inside and outside of the capital city. Following the supermarket chains’ progress in penetrating the FFV market – and identifying public and private actions that maintain and enhance the competitiveness of the traditional system – will be key policy issues for national and municipal policy makers concerned with the welfare of poor small farmers and urban consumers.

5.3. Formal Border Point Imports and Exports of Fresh Horticultural Produce

A survey on the main cross border points was carried out in the months of November and December 2002 to determine the formal flow of imports and exports of fresh horticultural produce between Kenya, Tanzania, and Uganda. These border posts were Lunga-Lunga, Taveta, Loitokitok, Namanga, Isebania, Busia and Malaba. Data on commodity, quantity imported/exported and the respective values on daily basis was extracted from Form 88 booklets and registers of the Customs Department from November 2001 to October 2002.

11 Njagi (1995) uses results from the CBS Urban Food Purchasing Survey of 1989 to indicate that, in that year, supermarket’s share of the total food market in Nairobi was only 2%. This report has some of the best detail currently available on food marketing in Nairobi.

33

Because informal border trade can often be larger than formally recorded trade, border agents were also asked to estimate the amount of informally traded produce that crosses the border. In between these major entry points, there are smaller posts that were not visited. Also, the whole of the Eastern, Northeastern and Northern frontiers were not surveyed because these areas are not important entry points for horticulture. However it is important to point out that some onions and cabbages have been reported to come from Ethiopia through the Northern frontier. These regions therefore may also be included in future studies. The results are summarized and discussed in the following sub-section. 5.3.1. Imports of Fresh Horticultural Commodities

Fresh horticultural commodities passing through the Kenya/Tanzania or Uganda border posts vary from one entry point to the next, but oranges, bananas, tomatoes and red onions predominate in volume and frequency. Imports of other fresh horticultural commodities were minimal. Table 5.1 presents information on production, marketed surplus, and imports of these commodities. Because import data is for the November 2001 through October 2002 period, we present mean production for each crop during the 2001 and 2002 production seasons. We estimate marketed surplus of each crop using data ffrom the 2000 Tegemeo/MSU rural household survey. To account for informal imports, we create lower bound estimates assuming that these imports are nil, and upper bound estimates assuming they are four times as large as formal imports. The latter scenario is two times as large as border agent estimates. The table suggests that, during the period of analysis, imports of bananas and tomatoes were a relatively minor portion of total domestic supply, while orange imports were more substantial, and onion imports may have captured more than half of the Kenyan market. Table 5.2. Upper- And Lower- Bound Estimates Of Import Market Share For

Selected Horticultural Crops In Kenya

Import Shares

Crop

Formal imports, Nov 01 --

Oct 02 (mt) Mean Production,

2001 and 2002 (mt)

Domestic MarketedSurplus, % of

Production Lower Bound Upper Bound Bananas 6,885 1,060,000 44 1.5 6.9 Tomato 3,255 262,500 72 1.7 7.9 Oranges 4,300 126,000 65 5.0 20.8 Onions 9,880 58,000 72 19.1 54.2 Notes: 1) Lower bound estimates assume no informal imports; upper bound assume informal imports are four times formal, based on border agent qualitative assessments. 2) marketed surplus percent is from Tegemeo/MSU 2000 household survey. This is for smallholder farms only, and thus provides a lower bound estimate on total marketed surplus. Bananas Kenya imports bananas mainly through Malaba and Busia border posts (Table 5.2). Approximately 70% of the recorded imports were from Uganda, with the balance coming from Tanzania through Taita Taveta and Namanga border posts. A key reason for importing bananas is that there are sweet varieties (bogoya) in Uganda which are not available in Kenya. The total recorded number of bunches imported during the period analyzed was 404,633 with a total value of Kshs. 42,139,618.

34

Table 5.3. Import Of Bananas (Bunches) Through The Main Cross-Border Posts In Kenya

Tanzania-Kenya Uganda-Kenya Month

Taita taveta Namanga Isebania Busia Malaba

Total (Bunches)

Total Value(Kshs)

November 01 145 11000 736 4525 20280 36686 3103550 December 01 63 9200 400 6538 13810 30011 2530550 January 02 0 5800 208 8068 18500 32576 2967600 February 22 8800 110 4436 16400 29768 2320100 March 5604 14200 244 7053 20200 47301 4853443 April 4353 0 0 7777 11600 23730 3025825 May 11955 0 138 6153 55400 73646 9153710 June 8952 0 140 7670 17710 34472 4783000 July 0 6200 0 7629 6175 20004 1690400 August 1788 6000 0 10567 11128 29483 2916500 September 1397 7200 250 4394 15345 28586 2695650 October 02 1775 0 132 3868 12595 18370 2099290 Total 36054 68400 2358 78678 219143 404633 42139618

Source: Kenya Revenue Authority, Customs Department.

Tomatoes The production of tomatoes in Kenya and Tanzania is throughout the year. Tanzania experiences surplus supply in May to August and November to December. During theses months, Kongowea market in Mombasa experiences shortages of local supply. Hence, there is some seasonal importation of tomatoes mainly from Tanzania during these periods (Table 5.3). Seventy-eight percent of the quantities imported go to the coastal region i.e. Kongowea market in Mombasa through Lunga-Lunga and Taita Taveta borders. Approximately 18% of tomato imports passed through Namanga border post to Wakulima market in Nairobi. For the period of November 2001 and October 2002, 92,737 crates of tomatoes valued at Kshs. 65,818,607 were imported from Tanzania and some minimal quantities from Uganda. In Tanzania, Kenyan collecting wholesalers source for tomatoes from Iringa and Lushoto, 915 km and 715 km respectively from Mombasa. The variety of tomatoes produced in Tanzania is moneymaker. It has a soft skin and a shorter shelf life when compared to cal-J, which is produced in Kenya. Kenyan consumers have a preference for Cal-J, which has a hard skin and a longer shelf life. Oranges Orange imports appear to be substantially more important than imports of either bananas or tomatoes, with upper bound estimates of imports reaching nearly 10% of local production and 15% of local marketed surplus (Table 5.1). The Kenyan market is a very important export market for Tanga region oranges from Tanzania. It has been estimated that 60% of oranges produced in this region are exported to the Kenyan market during the peak production season (Development Alternative Inc., 2003).

35

Table 5.4. Import of Tomatoes (crates) through the Main Cross-Border Posts in Kenya

Tanzania Uganda Month

Lunga lunga

Taita Taveta

Namanga Isebania Busia Malaba

Total Number of Crates

Total Value (Kshs)

November 01 2430 1432 530 70 0 85.5 4547.5 2815570 December 01 1000 859 260 38 0 0 2157 1204150 January 02 0 0 0 15 0 0 15 13500 February 0 0 0 5 0 0 5 4000 March 0 68 0 7 0 0 75 23525 April 0 28 0 0 0 0 28 7000 May 12265 1261 739 85 357 24 14731 11869467 June 0 2008 9460 73 958 0 12499 3870900 July 17400 2767 5416 8 529 0 26120 18281858 August 28109 863 358 100 819 0 30249 26141420 September 1350 95 80 130 196 10 1861 1405567 October 02 0 205 0 21 223 0 449 181650 Total 62554 9586 16843 552 3082 110 92737 65818607 Source: Kenya Revenue Authority, Customs Department.

For the period considered in the study, oranges were imported throughout the year (Table 5.4). Of the total recorded cross border imports of oranges, 59% passed through Lunga Lunga to Mombasa and 34% through Namanga to Nairobi. A total of 42,565 bags valued at Kshs. 25,509,575 were imported, about 95% coming from Tanzania. Wholesalers procure oranges from Mweza District in Tanzania, which is about 320 km from Mombasa. Although oranges are produced in most parts of Tanzania, the major areas of concentration are Tanga and Morogoro region. It is generally considered that orange production in Tanzania grew to a major economic importance during the late 1970s (Development Alternative Inc., 2003). The Ministry of Agriculture in Tanzania embarked on a plan to improve fruit production. Actions focused on the establishment of district nurseries for the production of planting materials, the introduction of new cultivars and the establishment of mother orchards. Accumulation of surplus fruits that could not be marketed and were pilling up along the roadsides was observed as early as 1980s. This led to establishment of processing plants in Korogwe, Mweza and Morogoro in the early 1980s. However, they all failed due to lack of working capital. At the time Tanzania was establishing the citrus sub-sector in 1970s, Kenya’s orchards were facing an attack from citrus greening disease. This disease thrives in mid-altitude areas of Kenya and has never been fully controlled in the country. Tanzanian production zones are primarily coastal, where the vector which spreads the disease is not active. Greening has therefore not been a major problem in Tanzania, which has allowed it to fill the gap in Kenya created by stagnant production over the past 10 years.

36

Table 5.5. Import of Oranges (bags) through the Main Cross-Border Posts in Kenya Tanzania Uganda Mothn

Lunga lunga

Taita taveta

Namanga Isebania Busia Malaba

Total (bags)

Total Value (Kshs)

November 01 0 148.5 2952 2 167 0 3270 3235000 December 01 2240 80 1008 0 0 0 3328 1849600 January 02 1680 0 732 0 16 0 2428 1308800 February 5383 0 624 14.5 0 0 6022 2459580 March 1735 31 0 286 0 0 2052 875821 April 1190 46 837 299 0 0 2372 1554700 May 980 257 1104 101.5 0 0 2443 1784398 June 1120 34 1008 79 0 0 2241 1456750 July 1680 40 1188 2 0 0 2910 1887700 August 140 110 1272 0 0 0 1522 1429600 September 1800 200 0 0 796 48 2844 1255426 October 02 7155 232 3672 0 0 76 11135 6412200 Total 25103 1178 14397 784 979 124 42565 25509575Source: Kenya Revenue Authority, Customs Department. Onions Onion imports appear to be very substantial in Kenya., reaching as high as half of local production and two-thirds of local marketed surplus (Table 5.1). Kenya imports onions from Tanzania throughout the year, nearly all passing through Namanga post to Nairobi and Taita Taveta border to Mombasa (Table 5.5). The yield of onions in Kenya is one of the lowest of the world producers (see table 3.4). Area under production has oscillated around 5,000 ha over the last ten years and total production (tons) has stagnated around 56,000 tons/yr. Thus, as demand for onions has continued to rise with population, Kenya has been forced to import from Tanzania.

In addition to horticulture, there were a number of other primary agricultural commodities also imported. These were as follows:- Lunga Lunga: Groundnuts, Yellow Grams, raw milk, Sorghum, Coconut cake, and Copra

cake; Taita Taveta: Beans, Maize, Rice, Dry peas, Tobacco, Timber, Green Grams, arrowroots,

Sorghum and Groundnuts; Namanga: Maize, Finger Millet, Peas (Dry), Wheat, Cloves, Beans, Fish, Coffee,

Cotton, Rice,Milk, Groundnuts, Spices, Prawns, Sunflower Cake, Maize germ, Sisal, Feed Barley, Cow peas, Pigeon peas;

Isebania: Fish, Ground nuts, Peas, Rice, Maize, Cassava, Sorghum, and Finger millet, SimSim and Cotton Seed Cake;

Busia: Finger Millet, Maize, Beans, Soya beans, Ghee (milk), Groundnuts, Eggs, SimSim, Cow peas, Sorghum, Fish, Ginger (Tangawizi), Cotton Seed Cake, Rice Jam, Cassava, Honey, Timber, Wheat bran;

Malaba: Maize, Finger millet, Eggs, Beans, Maize bran, G/nuts, Timber, Rice, Green Grams, peas, Arrow Roots, Wheat bran, Rice bran, Cotton and omena.

37

Table 5.6. Red Onions Imports (bags) From Various Border Points Tanzania Uganda Month

Taita Taveta Namanga Isebania Malaba

Total Number of Bags

Total Value (Kshs)

November 01 2903 11162 12 0 14076.5 12510300 December 01 2760 9192 2 0 11954 11036300 January 02 0 6685 1 4 6690 4015550 February 75 9631 3 0 9709 5930960 March 884 994 7 0 1884.5 2370665 April 1019 5280 63 0 6362 5263330 May 3233 4322 19 0 7574 9076490 June 1972 7493 16 0 9480.5 8447600 July 1513 6003 0.5 0 7516.5 6628300 August 2201 4297 4 0 6502 6983000 September 2068 125 0 0 2193 4211000 October 02 1565 9855 4 0 11424 9047640 Total 20192 75039 131 4 95366 85521135 Source: Kenya Revenue Authority, Customs Department. 5.3.2. Formal Exports of Fresh Horticultural Commodities

The export records of the Customs Department indicated that there is very little cross border export of fresh horticultural produce from Kenya to Tanzania and Uganda. However, there was a substantial amount of recorded fresh coconut exports to Tanzania through the Lunga Lunga and Taita Taveta borders. It was further observed at the Loitokitok border, that minimal unrecorded quantities of tomatoes, onions, cabbages, kales (sukuma wiki) and indigenous vegetables are sold to Tanzanian residents around Tarakea border point for local consumption. These commodities do not reach the main township areas in Tanzania. The customs officers said that there is very little export of fresh horticultural produce from Kenya to Tanzania and Uganda. Thus, most of the official trade flow of fresh horticultural produce and raw agricultural commodities is from the neighboring countries i.e. Tanzania and Uganda to Kenyan domestic markets. The direction of formal trade flow was assumed to be an indication of the direction of flow for informal trade for the same commodities. With this kind of scenario, it was concluded that Kenya’s fresh horticultural produce have not yet developed the required competitive advantage to claim space in the regional markets. Improved infrastructure and fewer regulations in Tanzania may be one key reason that it is able to export more successfully to Kenya than Kenya is to Tanzania. Unlike Kenya, Tanzania has continued to tarmac roads to its border posts. Four of its border points with Kenya are covered by the telecommunication network whereas in Kenya only one (Isebania) is covered. Tanzanian authorities have made issuance of trade permits administratively easy and cheap. Any interested trader can access these documents. In Kenya, procedures for issuing permits and import brokerage are complex, and only known brokers can clear goods. It is not clear what role relative exchange rate movements over the past decade have played in promoting imports into Kenya from Tanzania and Uganda. Since 1994, the Tanzanian Shilling has depreciated nearly 50% against the Kenyan Shilling, but official inflation data (from IMF) indicate that accumulated inflation in Tanzania has also been about 50% higher.

38

These patterns would suggest little change in real relative prices between the two countries, but may not accurately reflect dynamics affecting the products and areas of Tanzania under consideration. During this same period, the Ugandan Shilling has fallen about 20%, while accumulated inflation in that country has actually been about 20% less than in Kenya. These patterns do suggest falling real relative prices in Uganda. This may have contributed to the relatively high imports of banana from Uganda. 5.4. Handling of Agricultural Trade at the Border Points

Duty Concessions Import duty is the levy charged on any import cargo into the country. Generally 35% of Cost Insurance and Freight (CIF) price is charged on any imports from outside COMESA and members of the East African Community (EAC). However, tax concessions are given to member states at different percentages. For example, Kenya has given Tanzania and Uganda 90% concession on primary agricultural produce. Thus, import cargo within this category attracts a duty of 3.5%.. Import Declaration Form fees are charged on goods that have a value of more than $5,000 at a rate of 2.75% of CIF price. The interpretation on what rates to levy on each commodity is derived from the First Schedule of the Customs and Excise Act (Tariff Interpretation). In return for Kenya’s 90% concession, Tanzania and Uganda are expected to reciprocate by giving Kenya tax concessions on some categories of cargo e.g. primary agricultural produce. However, Kenya custom officials complain that Tanzania frequently resorts to suspended duty (anti-dumping) at 25% on Kenyan goods. Kenyan traders also complain of red tape with high tax rates imposed. When the Presidents of Uganda, Tanzania, and Kenya met in July 2003, one action they took was the creation of a task force to harmonize duties among the countries to facilitate trade. Horticultural Crop Development Authority Levy (HCDAL) The HCDA levy is a fresh horticultural import levy collected by customs department on behalf of HCDA. However, during the research visit, this charge was not being levied at Lunga-Lunga and Taveta. The charge is supposed to be one shilling (Ksh 1.00) per kg of produce. However, indications are that there is substantial under estimation of weights so as to minimize the cost. Measurements and Recording

It is not easy to get accurate measurements of what comes into the country through cross border posts because there are no weighing scales nor standardized packaging of commodities. The Custom offices resort to Direct Assessment using Form 88. Because none of the border points are properly equipped with computers, electronic weighing machines, weigh bridges, and other infrastructure, the officer assessing the cargo uses own judgment to estimate the quantity and price per unit and checks with the First Tariffs Schedule whether or not it is dutiable (Import Duty) and at what rate. Also, an assessment is made on whether or not to charge Import Declaration Form Fees (IDF). One would expect uniformity in all the trade parameters assessed for the same commodity in the various border points. This however, was not the case; different stations had different measurement units charged differently for same produce and others had different duty rates (Table 5.6). This inconsistency was observed for all agricultural commodities traded. In practice, these procedures appear to result in underestimation of measurements, inaccurate recording and

39

ultimately loss of revenue. It also results in low operational efficiency in trade due to the time and effort wasted by the cargo assessment officers in walking within the post.

Table 5.7. Trade Parameters for Oranges in the Various Border Points Border Points Unit Weight Price per

Unit (Kshs/Unit)

Import Duty Rate

(%)

IDF Rate

HCDA Levy

(Kshs/kg) Lunga Lunga Bag 20 kg 340 3.5 2.75 0 Taveta Bag 100 kg 1000 3.5 2.75 0 Namanga Ton 1000 kg 6000 3.5 2.75 1 Isebania Bag 100 kg 900 7 0 1 Busia Bag 18.75 kg 500 3.5 2.75 1 Malaba Bag 500 kg 6000 1 0 1

Source: Kenya Revenue Authority, Customs Department.

Infrastructure To enhance accurate measurements and recording of traded cargo in Tanzania, the Governments has improved the various forms of infrastructure. Every Tanzanian border point with Kenya is supplied with electricity and a plan to computerize them is in progress. In Kenya, on the other hand, only two (Taveta and Oloitoktok) of the five border points with Tanzania have electricity supply. The road network in some of the border points and their production areas is relatively poor. Yet some of these areas are among the main producers and suppliers of fresh horticultural produce to major towns and cities. For example, Taita Taveta region supplies fruits and vegetables to Kongowea market. Oloitoktok area also supplies fresh produce to Nairobi and Mombasa. Yet they are connected to the rest of the country with poor roads. Horohoro and Tarakea border points in Tanzania face a similar problem. The production areas have seasonal earth roads that are impassable during the rain season. Poor road network results in losses due to wastage in the farms and deterioration of quality of the produce during transportation to the market. It increases transportation costs for inputs and produce resulting in lower margins for farmers.

Border Point Agriculture Personnel Whereas several Government departments in Kenya, Tanzania and Uganda send officers to manage various aspects of trade at the border points e.g. Kenya Plant Health Inspectorate Services (KEPHIS), Kenya Bureau of Standards (KBS), Public Health etc, Tanzania in addition sends two Agricultural Extension Officers and two Veterinary Officers from the Ministry of Agriculture and Livestock to every border point. Their work is to inspect and record all types and quantities of agricultural trade leaving or coming into the country. The Ministry of Agriculture in Kenya would need to post Agricultural Officers at the border points to monitor and record agricultural trade, summarize the data on quarterly basis and send them to the headquarters. A decision had been made by the Director of Agriculture to send officers at the border point. However, it was not implemented. By the time of the survey there was only one officer posted at Busia border point, but he had left to go on study leave.

40

Telecommunication Telecommunication services in East African region are expensive, inadequate and unreliable. It has hampered quick and efficient flow of information between farmers and traders at the border points/production areas and the markets in the main cities and towns. However, out of the five border points of Kenya with Tanzania, only Isebania is not covered with telecommunication network; Horohoro, Holili, Tarakea and Namanga are all served by cell phone services. Indeed, Kenyan horticultural farmers/purchasing agents in Taita-Taveta area who can afford mobile phones have taken lines in Moshi region in Tanzania to enable them to receive market information from and transact business with partners or wholesalers at Kongowea market in Mombasa. Telecommunication has enhanced regional onion trade between distributing wholesalers at Wakulima market and collecting wholesalers at Kilombero market in Arusha. Traders in Arusha receive market information from their partners at Wakulima market in Nairobi early in the morning. This information forms the basis for price negotiation with traders in Nairobi and facilitates quick transaction of business before the commodity is shipped to Nairobi. It seems that telecommunication network is going to play an important role in business transactions of fruits and vegetables in the regional markets in the near future. Cargo Inspection Facilities (Verification Bays). These are the facilities that would assist Customs officials to inspect cargo in the transportation vehicles. Such a facility would be a structure with a roof and a pit where cargo can be offloaded for thorough inspection. These facilities are lacking or are not operational in all the border posts visited. This means that the officers cannot be effective in inspecting cargo on transit. Porous Borders Except for Malaba, which has a natural but limited barrier (a fast running river), the rest of the borders are quite open and difficult to patrol. Border patrols and arrests usually made by security agents of the would-be smugglers do not help to solve the problem. It is estimated that in some border points, as much as 90% of all commodities entering Kenya do so informally. Loitokitok border for example maintained they did not make a single official entry of any agricultural produce imported through that point because there was none. The data recorded is estimated at one-third of what actually is traded across the border point. It is difficult to ascertain the value of such informal trade. The most serious limitation is that none of the trade data recorded at the border posts is transmitted to KRA headquarters. What is remitted is the consolidated revenue by month. The implication is that national agricultural figures on trade are inaccurate because what is reported excludes unofficial cross-border trade activities. This means that there is underestimation of quantities of commodities imported, exported and consumed domestically. Thus, national food policy decisions are based on incomplete information.

41

6. Commodity Competitiveness Between Kenya and Tanzania/Uganda There are two commonly held views that explain how economies grow. The earlier view was that the economic success of a country is based on the available factors of production i.e., its comparative advantage. Comparative advantage is important because it is the basis of competitiveness of local production on both internal and external markets (World Bank, 2000). A more recent view explains the economic success of a country on the ability of the people to add value to available resources i.e., its competitive advantage. The two views result to radically different implications for a country’s economic policy. The growth of economies like Japan with little if any comparative advantage have also lead to the a conclusion that indeed, competitive advantage can be created. “Competitiveness is the degree to which a country can, under free and fair market conditions, produce goods and services that meet the test of international markets while simultaneously expanding the real incomes of its citizens (Castells, 1996)”. It is the requirement for staying in business on a sustained basis (World Bank, 2000). The fact that Kenya imports substantial quantities of some fresh horticultural produce from its regional neighbors while exporting very little to them suggests that Kenya is less competitive in these products than are her neighbors. The fact that the country is nevertheless very competitive in international produce markets suggests the dualistic nature of Kenya’s horticultural sector. The export sector, comprised of a relatively limited number of organized smallholder farmers along with medium- and commercial scale farmers, all of whom receive some level of production and marketing support from private export companies, competes effectively in the highly competitive and quality conscious markets of Europe. Meanwhile, the domestic sector is comprised of the vast majority of (largely unorganized) smallholder farmers, who receive little if any production or marketing support. The sector pays very little attention to productivity or quality, and struggles to compete with imported produce primarily from Tanzania. It is for this reason that this section analyzes marketing margins and the subsequent distribution of observable marketing costs as a measure of performance and efficiency in the production and domestic marketing of horticultural commodities. The section focuses on price spread at wholesale level, estimated wholesale profit and return to investment as indicators to measure competitiveness of commodities from Kenya and Tanzania/Uganda at a common market and at the farm level. Since the objective was to compare commodities, it was imperative to start the analysis at the wholesale level where the commodities from Kenya and Tanzania/Uganda meet at Kongowea market in Mombasa and Wakulima market in Nairobi.

For ease of comparison, this study focused on collecting wholesalers who purchase the commodities from farmers, transport them to the main markets and sell them either to retailers or distributing wholesalers. The analysis was based on cost buildup budgets for this group of traders of fresh horticultural commodities with a common market destination. Though few in number, these traders play a crucial role in the trading system by linking farmers and urban traders. They also have the broadest view of problems and opportunities in the system, due to the breadth of their contacts. Eight collecting wholesalers were interviewed for each commodity, 32 in total. Interviews in Mombasa (Kongowea market) and

42

Taveta were conducted in November 2002, while those in Nairobi (Wakulima market) were done in January 2003. The results are discussed by commodity in the following sub-sections. 6.1. Wholesale Market Analysis Results Bananas Bananas imported into Kenya’s domestic markets are obtained primarily from Uganda in Mbale region. Locally, bananas are produced and sold throughout the country (except for arid zones), but are most concentrated in Western and Central Highland areas, especially in Kisii and Murang’a/Nyeri in central province. Local and imported bananas are sold in Wakulima market in Nairobi. Some minimal quantities are imported from Tanzania and sold in Wakulima market also whereas others find their way to Kongowea market through Taita Taveta border point. Cost buildup budgets for bananas from Kenya and Uganda at Wakulima market are shown in Tables 6.1 and 6.2. The computations were based on a bunch of bananas. The statistics indicate that the cost of transportation takes the greatest proportion of marketing costs in both Kenya (77%) and Uganda (63%). It is also noticeable that transportation cost is higher for bananas from Kenya than for Uganda even though the distance covered is longer from Uganda. Table 6.1. Distribution of Wholesale Marketing Costs (Kshs/bunch) of a Bunch of

Bananas From Uganda and Kenya at Wakulima Market in Nairobi Kenya –Kisii/Nyeri Uganda-Mbale Cost Items

Marketing costs

(Kshs/bunch)

% share of marketing

costs

Marketing costs

(Kshs/bunch)

% share of marketing

costs Handling Costs at source: Assembly/Loading

2.70 5.70 1.00 2.51

Transport cost from source to market 36.60 77.22 25.00 62.66 Customs fees - - 10.00 25.06 Council fees per unit 7.50 15.82 3.50 8.77 Handling Costs at market: *offloading 0.60 1.27 0.40 1.00 Total marketing costs 47.40 100.00 39.90 100.00 Source: Authors Computation Examining the distribution of marketing margins, trader profits, and return on capital shows that profits per bunch of banana are more than 50% higher for the Kenyan commodity, but returns on capital are comparable for local and imported bananas (Table 6.2). The main advantage that bananas from Uganda have in Kenyan markets is the very low purchase price – 20 Ksh/bunch in Uganda vs. 60 Ksh/bunch in Kenya. Kenyan producers receive 43% of the selling price compared to only 25% in Uganda. Marketing costs from Uganda net of customs duties are also lower – about 30 Ksh/bunch vs. 47 Ksh/bunch. Thus, Ugandan bananas clearly have a substantial cost advantage over Kenyan bananas. Yet, despite the much more remunerative prices for Kenyan producers, and the somewhat higher local marketing costs, imported quantities as a percent of total supply appear to be very low under any reasonable scenario (Table 6.1); even if informal imports were 10 times formally recorded imports, total imports would only rise to 15% of locally marketed surplus. It would thus appear that Kenyan consumers prefer local varieties of banana.

43

Table 6.2. Distribution of Marketing Margins, Trader Profit and Return on Capital for a Bunch of Wholesale Bananas (Kshs/bunch) From Uganda and Kenya at Wakulima Market in Nairobi

Category of values Kenya-Kisii/Nyeri Uganda-Mbale (Kshs/bunch) (Kshs/bunch) Mean purchase price 60.00 20.00 Mean selling price 140.00 80.00 Marketing margin 80.00 60.00 Marketing costs (including customs duties) 47.40 39.90 Estimated wholesale profit 32.60 20.10 Share of selling price accruing to producers (%) 42.86 25.00 Share of selling price accruing to marketing costs (%) 33.86 49.88 Share of selling price accruing to wholesalers (%) 23.29 25.13 Estimated return on wholesale investment (%) 30.35 33.56 Source: Authors Computation Tomatoes Tomatoes consumed in Mombasa and most of the coastal region are obtained locally from Karatina , Loitoktok and Taveta areas. When there is a shortage, tomatoes are imported from Tanzania in Iringa. Table 6.3 shows the marketing costs from these production areas to Kongowea market in Mombasa. The computations are based on a crate (35 kg) of tomatoes. The results indicate that transportation takes the highest proportion (> 60%) of the marketing costs for Kenyan as well as Tanzanian tomatoes. The proportion is greater in all Kenyan regions than in Tanzania, though actual transport cost is highest from Tanzania, consistent with the longer distances traveled. The share of selling price accruing to producers in Kenya was about 50% in the regions considered (Table 6.4). In Tanzania, the farmers got a relatively small share (28%). The difference in actual purchase prices was even greater: Ksh 180/crate in Tanzania, compared to Ksh 350-400 in Kenya. With the exception of tomatoes from Taveta, trader profit per crate and return to capital were both substantially higher for locally produced tomatoes than for the imported product.

44

Table 6.3. Distribution of Wholesale Marketing Costs (Kshs) of a Crate (35 kg) of Tomatoes From Tanzania and Kenya at Kongowea Market in Mombasa.

Tanzania-Iringa Kenya-Karatina Kenya-Oloitoktok Kenya-Taveta Cost Items

Marketing costs

Kshs/crate)

% share of total

cost

Marketingcosts

Kshs/crate

% share of total

cost

Marketing costs

Kshs/crate

% share of total

cost

Marketing costs

Kshs/crate

% share of total

cost Distance from market(KM)

915 650 470 309

Handling cost at source: Assembly/ Loading/Market fee

12 4.3 12 9.8 10 7.6 29 18.2

Transport cost to Kongowea market

172 61.7 90 73.2 100 76.3 100 62.8

Custom fees 72 25.8 Any other payment 1.7 0.6 Council fees per unit

11 3.9 11 8.9 11 8.4 10 6.4

Handling costs at market:Offloading /storage/Brokerage

10 3.6 10 8.1 10 7.6 20 12.6

Total marketing costs

279 100.0 123 100.0 131 100.0 159 100.0

Source: Authors Computation Table 6.4. Distribution of Marketing Margins, Trader Profit and Return on Capital for

a Crate (35 kg) of Tomato at Kongowea Market in Mombasa (Kshs/Crate) Category of Values Tanzania-Iringa

(Kshs/crate) Kenya-Karatina

(Kshs/crate) Kenya-

Oloitoktok (Kshs/crate)

Kenya-Taveta (Kshs/crate)

Mean purchase price 180.00 400.00 400.00 357.14 Mean selling price 638.00 825.00 825.00 700.00 Marketing margin 458.00 425.00 425.00 342.86 Marketing costs 278.70 123.00 131.00 159.20 Estimated wholesale profit 179.30 302.00 294.00 183.66 Share of selling price accruing to Producers (%)

28.21 48.48 48.48 51.02

Share of marketing costs (%) 43.68 14.91 15.88 22.74 Share of selling price accruing to wholesalers (%)

28.10 36.61 35.64 26.24

Estimated return on wholesale investments (%)

39.09 57.74 55.37 35.57

Source: Authors Computation Oranges Imported oranges landing at Kongowea market in Mombasa are sourced from Tanga area in Tanzania, whereas local oranges come primarily from Shimba Hills, which lies 70 km from Kongowea market. The oranges sold in Wakulima market are obtained from Mweza District in Tanzania, about 700 km from the market. No Kenyan oranges were found in Wakulima during the period of research. Orange varieties from Tanzania are of better quality than those from Kenya, being consistently larger, juicier and with fewer surface blemishes. This quality superiority is based in part on improved varieties and in part on better control of citrus greening disease, which has affected orchards in Kenya since the 1970s. Although symptoms

45

can vary across varieties, the disease is known for producing small irregularly shaped fruit with a thick, pale peel. Because the vector which spreads the disease thrives primarily in mid-altitude areas, while orange production in Tanzania is concentrated in the coastal lowlands, Tanzania has had far fewer problems with this disease. The “matombo sweet variety”, produced only in Mweza district in Tanzania, is especially valued by consumers in Nairobi, and in fact has come to be called “Nairobi variety” because so much of its production is exported to Nairobi Tables 6.5 and 6.6 show the cost budget and selling price shares for wholesale traders in both markets. The key result from Table 6.11 is that total marketing costs from Tanga in Tanzania are lower than from Shimba Hills in Kenya, despite the transport distance being nearly 5 times longer from Tanga. At the time of data collection, transport costs for the 70 km route from Shimba Hills were about 90% higher than for the 320 km from Tanga. This reflects extremely poor road infrastructure in the main orange producing areas of the Coast, resulting in transport of oranges on head or in wheelbarrow for relatively long distances until reaching passable roads. While the problem is worse during the rains, costs are high throughout the year in these areas. Producer price and the share of selling price accruing to producers are relatively low for Kenyan oranges (29%) compared to Tanzanian oranges (44% -- 50%), reflecting both the higher quality of Tanzanian oranges and high marketing costs out of Shimba Hills. Oranges from Mweza appear to be of exceptionally high quality, based on their purchase and sales prices. Wholesaler profits per bag are higher for Tanzanian oranges, especially for those from Mweza, but returns to working capital are comparable across all purchase locations.

Table 6.5. Distribution of Wholesale Marketing Costs of a Bag (100kg) of Oranges from Tanzania and Kenya at Kongowea (Mombasa) and Wakulima (Nairobi) Markets

Tanzania-Tanga Kenya –Shimba hills Tanzania –Mweza Kongowea Kongowea Wakulima

Cost Items Marketing costs

Kshs/bag

% share of marketing

costs

Marketing costs

Kshs/bag

% share of marketing

costs

Marketing costs

Kshs/bag

% share of marketing

costs Distance from source to sale market (Km)

320 70 700

Handling cost at source Assembling/Loading

52 22.91 60 21.62 70.70 20.23

Market fee 6 2.64 Transport cost per unit from source to Kenyan market

100 44.05 186.50 67.21 147.80 42.30

Custom fees 40 17.62 60.00 17.17 Any other payment 4 1.76 6 2.16 50.15 14.35 Council fees per unit 6 2.64 6 2.16 8.00 2.29 Handling charge at sale market:

Offloading 6 2.64 6 2.16 10.00 2.86 Packing and weighing 11 4.85 11 3.96 0.00 Security 2 0.88 2 0.72 2.75 0.79 Total marketing costs 227 100.00 277.50 100.00 349.40 100.00

Source: Authors Computation

46

Table 6.6. Distribution of Marketing Margins, Trader Profit and Return on Capital for a Bag (100 kg) of Oranges (Kshs/Crate) From Tanzania and Kenya

Tanzania-Tanga

Kenya- Shimba Hills

Tanzania –Mweza

Category of values

Kongowea (Kshs/bag)

Kongowea (Kshs/bag)

Wakulima (Kshs/bag)

Purchase price 375.00 192.50 1008.75 Selling Price 850.00 660.00 2005.50 Marketing Margin 475.00 467.50 996.75 Marketing Costs 227.00 277.50 349.40 Estimated wholesalers profit 248.00 190.00 647.35 Share of selling price accruing to producers 44.12 29.17 50.30 Share of selling price accruing to marketing costs (%) 26.71 42.05 17.42 Share of selling price accruing to wholesalers (%) 29.18 28.79 32.28 Estimated return on traders’ investment (%) 41.20 40.43 47.66

Source: Authors Computation Onions Onion imported from Tanzania are mainly produced in Mang’ola in Arusha region, approximately 550 km and 450 km from Kongowea and Wakulima markets respectively. Locally, key areas of onion supply to the market are Karatina, Meru, Oloitoktok and Taveta. Tables 6.7 and 6.8 show the wholesale trade budgets for Tanzanian and Kenyan onions at Kongowea market whereas Tables 6.9 and 6.10 are for Wakulima market. The key result from Table 6.7 is that total marketing costs – driven by transport costs and “other transit costs” – are highest from Taveta in Kenya, which is the shortest distance from Kongowea market. Trader profits per unit and returns to working capital for onions sold in Kongowea are both about 2.5 times higher on onions from Tanzania, but are still low compared to other commodities analyzed. Table 6.7. Distribution of Wholesale Marketing Costs (Kshs./bag) of a Bag (100kg)

of Onions From Tanzania and Kenya at Kongowea Market in Mombasa Country –Source: Tanzania –Mangola Kenya–Karatina Kenya Taveta/ Loitokitok

Cost Items

Marketing costs

(Kshs/bag)

% share of marketing

costs

Marketing costs

(Kshs/bag)

% share of marketing

costs

Marketing costs

(Kshs/bag)

% share of marketing

costs Distance from source to sale market(KM)

550 650 373

Handling cost at Marketing fee/Brokerage source:Assembly/Loading

42 7.50 115 19.33 39 5.01

Transport cost to Kongowea market

217 38.67 235 39.50 390 50.13

Custom fees 28 5.06 Any other payment 10 1.78 32 4.11 Any other cost on transit 10 1.78 72 9.25 Council fees per unit 48 8.57 48 8.07 48 6.17 Handling at sales market: Offloading 20 3.57 20 3.36 19 2.40 Packing and weighing 35 6.31 27 4.54 30 3.86 Cost of nets/ gunny bags 150 26.77 150 25.21 148 19.07

Total marketing costs 560 100.00 595 100.00 778 100.00 Source: Authors Computation

47

Table 6.8. Distribution of Marketing Margins, Trader Profit and Return on Capital for a Bag of Wholesale Onion (Kshs/bag) from Tanzania and Kenya at Kongowea Market in Mombasa

Category of values

Tanzania –Mangola(Kshs/bag)

Kenya –Karatina (Kshs/bag)

Kenya –Taveta/ Loitokitok (Kshs/bag)

Mean Purchase price 1483 1560 1447 Mean Selling price 2400 2300 2367 Marketing margin 916.67 740.00 920.00 Marketing costs 559.67 595.00 777.33 Estimated Wholesale profit 357.00 145.00 142.67 Prod’r share of selling price (%) 61.81 67.83 61.13 Share of selling price accruing to marketing costs (%)

23.32 25.87 32.85

Wholesaler share of selling price (%) 14.88 6.30 6.03 Estimated return on traders investment (%)

17.47 6.73 6.41

Source: Authors Computation Onions sold at wholesale in Wakulima market show a similar pattern as the Kongowea market in terms of cost and distribution of the selling price (Tables 6.9 and 6.10). However, the sales unit is a net of 13 kg rather than a bag of 100 kg. Also, onion wholesalers at Wakulima in Nairobi purchase Tanzanian onion from traders in Kilombero market in Arusha, rather than purchasing at the farm level as wholesalers in Kongowea do. Thus, purchase prices per kg are about 50% higher for onions destined for Wakulima compared to those destined for Kongowea in Mombasa12. The collecting wholesalers in Arusha market receive market information (about quantities and prices) through telephone from their counterparts at Wakulima market in Nairobi every morning between 4.00-5.00 a.m. This information forms the basis for price negotiation with traders from Nairobi. The onions are then transported to Nairobi in ten-ton lorries. The cost budgets of wholesalers at Wakulima market show that the main cost component for both Tanzanian and Kenyan onion is transport and the proportion is higher for Kenyan onion than for Tanzanian onion (Table 6.9). The share of selling price accruing to farmers is lower for Tanzanian onion than for Kenyan onion (Table 6.10). However, the share of selling price accruing to wholesalers is higher for Tanzanian onion (27%) than for Kenyan onion (16%). The estimated trader profit and return on investment are also higher for Tanzanian onion than for Kenyan onion.

12 The reason for this differential purchasing behavior makes for an interesting story. Wakulima onion collecting wholesalers originally sourced their onion, as they do now, at Arusha point from Tanzanian traders who had delivered the product from Kilombero market. Later on, Wakulima traders realized they could get onions more cheaply at the farm gate at Mag’ola point. As a result, Tanzanian middle men at Arusha could not sell their onion as they used to, which lead to an onion “trade war”. It was finally agreed that Wakulima collecting wholesalers must stop going to the farms and instead purchase their onion at Arusha. For whatever reason, this agreement did not affect traders taking onion to Kongowea market through Moshi.

48

Table 6.9. Distribution of Wholesale Marketing Costs (Kshs./Net) of a Net (13 kg) of Onions From Tanzania and Kenya at Wakulima Market in Nairobi

Tanzania Kenya

Cost Items Marketing costs

(Kshs/net) % share of marketing

costs

Marketing costs

(Kshs/net)

% share of marketing

costs Handling Costs at source

*Assembling & loading 15.75 12.20 17 20.48 Transport cost to market per unit 66.65 51.61 44 53.01 Customs fees or other costs 17.95 13.90 Council fees per unit 20 15.49 20 24.10 Handling Costs at market:

Offloading 8.5 6.58 2 2.41 Security 0.3 0.23

Total marketing costs 129.15 100.00 83 100.00 Source: Authors Computation Table 6.10. Distribution of Marketing Margins, Trader Profit and Return on Capital

for a Net (13kg) of Wholesale Onion (Kshs/net) From Tanzania and Kenya at Wakulima Market in Nairobi

Category of values Tanzania (Kshs/net) Kenya (Kshs/net) Mean purchase price 273.00 168.00 Mean selling price 550.00 300.00 Marketing Margin 277.00 132.00 Marketing Costs 129.15 83.00 Estimated wholesale profit 147.85 49.00 Share of selling price accruing to farmers (%) 49.64 56.00 Share of selling price accruing to marketing costs (%) 23.48 27.67 Share of selling price accruing to wholesalers(%) 26.88 16.33 Estimated return on wholesale investment (%) 36.76 19.52

Source: Authors Computation Summarizing across all four commodities, results of the collecting wholesaler analysis are quite consistent with observed trade patterns (Table 6.11; see also Table 5.1 for estimated share of each commodity in the Kenyan market). Imported bananas and tomatoes take up a very small portion of the Kenyan market, and trader profits per unit of each of these commodities were shown to be higher for Kenyan produce than for imports. Return to capital on tomatoes was also higher for the local product, while returns on bananas were comparable between the imported and local commodity. Imported oranges take up a larger share of the Kenyan market, and the wholesaler analysis shows profits per bag to be higher for the commodity from Tanzania, especially the high quality oranges from Mweza. Returns to capital are similar for imported and local oranges. Finally, onions take up as much as two-thirds of the Kenyan market, and both profit per bag and returns to capital were shown to be higher for the imported commodity.

49

Table 6.11. Summary Maketing Cost Build-Ups For Bananas, Tomatoes, Oranges,

And Onions From Tanzania, Uganda, And Kenya To Wakulima And Kongowea Markets

Commodity and Origin

Producer Price

(Ksh/unit)

Sales Price

(Ksh/unit)

Marketing Costs

(Ksh/unit)

Trader Profit

(Ksh/unit)

Return to Trader

Working Capital

Bananas From Uganda (Mbale) 20 80 40 20 33.3% From Kenya (Kisii/Nyeri) 60 140 47 33 30.8%Tomatoes From Tanzania 180 638 279 179 39.0% From Kenya 386 783 138 260 49.7%Oranges From Tanzania Tanga 375 850 227 248 41.2% Mweza 1,009 2,006 349 647 47.6% From Kenya (Shimba Hills) 193 650 278 190 40.3%Onions From Tanzania (Mangola) To Kongowea (Mombasa) 1,483 2,400 560 357 17.5% To Wakulima (Nairobi) 273 550 129 148 36.8% From Kenya To Kongowea (Mombasa) 1,504 2,334 686 144 6.6% To Wakulima (Nairobi) 168 300 83 49 19.5%

6.2. Border Point Traders of Fresh Fruits and Vegetables This section deals with the socio-economic characteristics and business analysis of traders dealing with fresh fruits and vegetables at seven border points: Lunga Lunga, Taveta, Oloitoktok, and Namanga on the border with Tanzania, and Isebania, Busia, and Malaba on the border with Uganda. Fifty-one respondents were selected using systematic sampling. The survey was carried out in November 2002. Nearly three-quarters of the traders were female, and these tended to be younger (31 years average) than the males (36 years). Three-quarters of all traders were married. Educational levels are relatively high, compared to the general population, with two-thirds having some primary education, 18% some secondary, and 6% some post-secondary. Only 10% had no formal schooling. Most of the respondents (63%) are Kenyan citizens who reside at the border town. Twelve precent are foreign traders who also reside at the border town but on the Kenyan side, and 23% are Kenyan traders who live within 20 km of the town and commute daily to the market. The majority of the interviewed traders were retailers (53%) but others combine small scale wholesale and retail (39%). Only 8% were pure wholesalers. Transport to or from the border market to other market destinations is by lorries (69%), Public Service Vehicles (14%), bicycles and pickups (6% each, 12% total) and foot (4%).

50

There are no permanent market structures in the border sites. Most traders (57%) have put up simple structures made of wood, iron sheets and polyethylene where they place their commodities. Others place them on the open ground (43%). The traders complained of huge losses they incur especially during the rainy season. In Taveta, Isebania and Busia, the council had not allocated any area for traders, who therefore occupied private plots, from which they had instructions to vacate anytime the owners would decide. In Malaba, traders sold by the roadside. Forty-five percent of the traders indicated that they were members of trading groups. The majority of those who did not belong to any trading group (60%) felt that their businesses were too small to join the group (See Table 6.12). Others were willing to join but said they were not aware of any self-help group in the area. Few had fears of mismanagement of the groups. Table 6.12. Reasons For Not Being A Member Of Self Help Group Possible Reasons Frequency Percent No self help group in this market 6 21.4 Have not attained the requirement for various groups 3 10.7 Not had need to join one 17 60.7 Fear of previous experience related to mismanagement 2 7.1 Total 28 100.0

Source: Interviews with Traders, November 2002. Eighty-three percent of those who are members of groups indicated that the groups provided small loans to their members. Most of the traders are small-scale and hence their contribution to the groups is low. There was no collateral required and the loans were given on a rotational basis. Seventeen percent of the members indicated that their groups buy household items for the members. The majority of the traders (90%) used their own savings while setting up their businesses. The rest borrowed money from relatives and/or friends. Only 22% tried to get credit for purposes of horticultural trading. All of them got the credit through the self-help groups. It seems that business around the borders remains low partially because of limited credit facilities. Market information around the border markets is mainly by word of mouth (92%) from friends, relatives and business colleagues (Table 6.13). The border point traders rely on cross-border traders for market information. Few rely on the media such as newspapers. The traders said they are not aware of any market information program on the radio. Table 6.13. Sources of Market Information. Information Sources Frequency Percent From newspapers 2 3.9 By word of mouth from friends/ relatives/brokers/traders 47 92.2 Do not need market information for my business 2 3.9 Total 51 100.0

Source: Interviews with Traders, November 2002. Prices of Kenyan and Tanzania/Uganda commodities appear to be comparable. However, 47% of traders interviewed indicated that Kenyan commodities are more expensive (Table 6.14). Overall, they preferred selling commodities from Tanzania/Uganda since they were more profitable unless they were out of stock. The traders complained of high transport charges on the Kenyan side. Also, 55% of traders indicated that commodities from

51

Tanzania/Uganda are of better quality than Kenyan commodities (see Table 6.15). This has led to greater demand for these commodities coupled with their affordability. Traders seem to prefer them as they sell faster. Table 6.14. Price Comparison Among Countries Price comparison Frequency Percent Kenya's commodities are cheaper 19 37.3 Kenya's commodities are more expensive 24 47.1 There is no difference in price 5 9.8 Do not Know 3 5.9 Total 51 100

Source: Interviews with Traders, November 2002. To ensure that they sell quality produce to customers, the traders selected good quality produce (86%), and maintained clean sales sites (39%). Others packed the produce for the customers (Table 6.16). When asked to list the four main challenges that they felt constrain their businesses (Table 6.17), most traders indicated low demand for fruits and vegetables (37%), lack of market space (35%), competition from neighboring country traders (22%), and harassment by town council authorities (18%). High charges of cess and licenses were mentioned by over 10%. Some of the markets (e.g. Taveta) charged sellers when bringing in produce and also charged buyers when carrying it out of the market to be taken to other market destinations. Table 6.15. Quality Comparison of Commodities Quality Comparison of Commodities Frequency Percent Kenya's commodities are better 14 29.4 Tanzania's/ Uganda's commodities are better 29 54.9 They are bout the same quality 5 9.8 Do not Know 3 5.9 Total 51 100.0

Source: Interviews with Traders, November 2002. Table 6.16. How To Ensure Quality Commodities Are Sold Action to Ensure Quality Frequency Percent By packing the produce 11 21.6 By maintaining clean sales site 20 39.2 By doing selection of good produce 44 86.3 Others 2 3.90

Source: Interviews with Traders, November 2002.

52

Table 6.17. Challenges Traders are Facing Challenges Traders are facing Frequency Percent Competition from neighboring country traders 11 21.6 Lack of market space 18 35.3 High charges of cess and licenses 6 11.8 Harassment by town council authorities 9 17.6 Low demand 19 37.3 Hostility from traders across the border 1 2.0 Tribalism 1 2.0 Poor exchange rate 1 2.0

Source: Interviews with Traders, November 2002. To tackle these challenges, nearly all traders referred to physical conditions in the markets: three-quarters felt that the township councils could assist them by constructing market facilities, and nearly one-quarter referred to the need to improve coordination and allocation of space within the market (Table 6.18). This could help reduce losses incurred through damage and boost security in the market. Lack of credit facility was another serious limitation to expanding business operations among traders and micro-finance institutions could intervene to uplift their businesses. Some traders complained of high transport charges due to the poor state of roads and asked if the government could intervene to construct new roads and maintain the existing ones. They complained of harassment by custom officials when they intend to sell across the borders and asked if the government could help to allow efficient cross border trading. Table 6.18. Opportunities for Interventions Possible Areas for Intervention Frequency Percent Storage facilities at the border 1 2 Improved roads 2 4 Transport availability 7 14 Constructing a market facility 38 76 Provision of credit facilities 14 28 Supply of electricity to the border market 2 4 Effective market co-ordination and allocation of space 12 24 Lower commission 4 8 Cooperation by customs officials 1 2 Allow more cross-border trade 1 2 Improve prices 1 2

Note: Traders referred to more than one action, thus percentages sum to more than 100% Source: Interviews with Traders, November 2002. 6.3. Farm Analysis Results: Costs of Production for Onion in Tanzania and Kenya

The production and marketing costs of commodities drive their competitiveness in markets. Having just analyzed the marketing costs of four key horticultural commodities, we now turn to farm-level production costs. Due to resource limitations, we focus on only one commodity: onions, for which trade flows suggest Kenya is least competitive in regional markets.

53

Synthetic crop budgets were developed for four production areas – three in Kenya and one in Tanzania – through group interviews of farmers. Tanzanian onions reaching Kongowea market in Mombasa and Wakulima market in Nairobi are produced at Mang’ola, approximately 200 km from Arusha town (450 km from Nairobi). Locally, Kongowea market gets onions from Taveta, Oloitoktok (Kimana) and also Karatina whereas Wakulima market gets onions primarily from Oloitoktok, Laikipia, Narok, Meru, and Karatina. Production of onions in these areas is by irrigation. To facilitate comparison, it was important to categorize areas with similar production systems, climate and a common market for the commodities. Hence separate onion enterprise budgets were developed for Mang’ola in Tanzania, and for Taveta and Oloitoktok in Kenya. The budgets for Narok, Laikipia and Meru were averaged to get one representative synthetic budget for production areas far removed from Mang’ola and Taveta/Oloitoktok region. In each case, 5-7 farmers were interviewed as a group to develop an average farm budget for the commodity in that area. Results are shown in Tables 6.19 to 6.21 below. Table 6.19 presents detailed costs by item, Table 6.20 expresses these in percentage terms, and Table 6.21 summarizes the information to show cost of production and gross margin for farmers per bag. Cost of Production Results in the three tables indicate that the onion production system in Tanzania has a higher cost per acre than in Kenya, but lower costs and higher returns to farmers per bag. These lower unit costs of production in Tanzania are driven by seed costs one-tenth of those in Kenya, and average yields that are more than 50% higher. Across the board, the main cost components are weeding, irrigation and seeds (Tables 6.19 and 6.20). Weeding cost in Tanzania accounted for 17% of TVC whereas in Kenya it ranged between 12% and 14%. Onion farmers in Tanzania weed each basin four times while in Kenya it is done three times (See Appendix C). The cost component for irrigation was 31% in Tanzania compared to 14% in Narok and 16% in Taveta. Irrigation quality may be higher in Tanzania, however. Unlike in Kenya, a specialized group of people in Tanzania have sharpened the skill to irrigate farms. These people can quite a number of farms simultaneously, working on the basis of well organized irrigation schedules of the farms they manage. Farmers hire the services of these irrigation specialists when they plant onion, and typically pay them an equivalent of 10 bags per acre. Kenyan farmers also typically purchase irrigation services (except in Oloitoktok, where farmers do it themselves and hence the cost is accounted for as part of return to farmers’ resources), but the providers have not developed as organized a system as in Tanzania. Hence the payments are relatively lower. A key question is whether the higher cost in Tanzania reflects better quality service resulting in higher yields. Yields in Tanzania average 120 bags per acre, compared to 60-85 bags/acre in Kenya, but it is not clear what role the quality of the irrigation service plays in this difference.

54

Table 6.19. Distribution of Onion Production Costs (Kshs/Acre) in Tanzania and

Kenya Cost Items

Tanzania Mang'ola

Kenya Oloitoktok

Kenya: Narok/Laikipia/Meru

Kenya Taveta

Number of basins per acre 250 300 300 300 Land Own Own Own Own Nursery operations ------------------------------ Ksh/acre ------------------------------ Ploughing and preparation 750 900 1567 800 Seed 640 6250 6400 4400 Irrigation-Nursery/Field 14830 0 6933 6340 Polytrin 3200 3200 2647 1800 Foliar feed 380 750 Spraying labour 420 400 420 450 Main field Ploughing 1000. 2000 1467 2000 Making of basins & furrows 2083 1800 2643 1800 Lines for seed placement 1042 600 Planting/Fertilizer labor 3125 1800 3200 4800 Chemicals 4125 6650 5170 2100

4000 4800 2500 3000 Fertilizers-Nursery/Field 4000 3300 1305 -

Weeding/Fertilizer labor 8333 5400 6000 5400 2083 3000 3125 3500 Harvesting: Uprooting

Cutting 2083 3000 3125 3500 Total variable cost (TVC) 51714 43100 46882 40640

Source: Authors Computation Usually, the recommended rate of fertilizer application on onions is 200 kg (4 bags) of DAP or 17:17:17 (NPK) for planting and 150 kg (3 bags) of ASN per acre. In Tanzania, farmers use 4 bags of CAN for planting and 4 bags of ASN for top dressing. In Kenya, onion production regions used less than the recommended rates except Oloitoktok (see Appendix C). Seed cost in Kenya is higher by a factor of nearly 10 compared to Tanzania. This huge price difference can be attributed to the way farmers obtain seed in the two countries. In Kenya, farmers purchase imported seed, either Red Creole or Bombay Red from the various companies. The cost ranges between Kshs. 2,200 to 3,000/kg. In Tanzania, local farmers produce quality declared seed that is of higher quality and is relatively cheap. See Chapter 7, section 7.1 for more detail on the quality declared seed program in Tanzania. Table 6.21 shows yield and information related to unit cost of production and gross margins for farmers. In Tanzania, the yield of onion is approximately 120 bags/acre, compared to an average of 73 bag/acre in Kenya. The cost of production is about Kshs 405/bag in Tanzania and an average of Kshs 623/bag in Kenya. Thus, the yield is higher and the cost of production is lower in Tanzania than in Kenya. Both gross margin/acre and gross margin per bag are much higher in Tanzania than in Kenya also. Together with the marketing cost buildup results, these farm measures help explain why Tanzanian onion has a competitive advantage over Kenyan onions.

55

Table 6.20. Percentage Distribution of Onion Production Costs in Tanzania and

Kenya Tanzania

Mang’ola Kenya

Oloitoktok Kenya

Narok/Likipia/Meru Kenya Taveta

% cost of TVC

% cost of TVC

% Cost of TVC % Cost of TVC

Nursery operations Ploughing and preparation 1.45 2.09 3.34 1.97 Seed 1.24 14.50 13.65 10.83 Irrigation 28.68 0.00 14.79 15.60 Polytrin 6.19 7.42 5.65 4.43 Folier feed 0.00 0.81 1.85 Spraying labour 0.81 0.93 0.90 1.11 Main field Ploughing 1.93 4.64 3.13 4.92 Making of basins & furrows 4.03 4.18 5.64 4.43 Lines for seed placement 2.01 1.39 Planting 6.04 4.18 6.83 11.81 Chemicals 7.98 15.43 11.03 5.17

7.73 11.14 5.33 7.38 Fertilizers 7.73 7.66 2.78 7.38

Weeding 16.11 12.53 12.80 13.29 4.03 6.96 6.67 8.61 Harvesting: Uprooting

Cutting 4.03 6.96 6.67 8.61 Total variable cost (TVC) 100.00 100.00 100.00 100.00

Source: Authors Computation Table 6.21. Efficiency Measures of Onion Production in Tanzania and Kenya Items Tanzania

Mang’ola Kenya

Oloitoktok Kenya

Narok/Laikipia/Meru Kenya Taveta

Yield (bags) 120 85 60.0 75 Price(Kshs/bag) 1,483 1,447 1,241 1,400 Gross output (Kshs/acre) 177,960 122,995 74,460 105,000 Gross Margin (Kshs/acre) 129,398 79,895 24,745 65,080 Cost per bag 405 507 829 532 Gross Margin/bag 1,078 940 412 868 Source: Authors Computation Onion Storage Storage is another aspect that adds a competitive advantage to Tanzanian onions. Farm storage is primarily concerned with making onion available at the desired time in the market. It becomes more useful by being held from periods of relative plenty to periods of relative scarcity. Most farmers in Tanzania have good post harvest management practices. They have built well aerated stores that can keep onions in good condition for a period of six months. This ensures continuous supply of onions to the markets all the year round and reduces the risk of price instabilities to the farmers. Most onion-producing households have a storage structure. Some of the structures can hold up to 120 bags of onion each weighing 100-120

56

kgs. The development of the storage system for onions has been a result of local initiatives and donor support particularly from the Spanish Government. Mang’ola area in Tanzania has only one rainy season, from November to March, and receives less than 500mm. During the rainy season, farmers plant food crops. Very few of them (15%) plant onions (Kamau, 2001). The main onion planting season is the dry period from late February to the end of September. Most of the onion planted in July to September is stored after harvest and sold up to end of May when the onion planted in February is harvested. Thus Tanzania has developed a competitive capability to supply onion to the regional markets throughout the year. Kenya on the other hand is not yet able to store onions, so what is produced tends to be sent to the market straightaway.

57

7. Input Markets for Fresh Horticultural Produce in Tanzania and Kenya The Kenyan and Tanzanian governments have not intervened to any significant degree in horticultural output markets to buy, sell, export, or set prices. The main regulatory body of the horticultural sub-sector in Kenya is the Horticultural Crop Development Authority (HCDA), which was established in 1967. HCDA was originally given authority to fix prices, regulate trade, and operate processing facilities and market horticultural goods. For example, HCDA maintained a monopoly on onion marketing and export briefly, then competed with private onion traders and finally, by 1986, withdrew its buying and selling functions from the market. Based on unsuccessful experiences, these functions were pared back to regulation, provision of market information and advisory services. Hence, horticultural commodities are bought and sold in a private, competitive market environment with very little government regulation and no direct government participation in commercial activities (Dijkistra, 1999). Horticultural input markets, on the other hand, have faced various types of interventions in each country. This section explores the policy environment and how it has shaped the system of input markets for horticultural commodities in each country, ultimately resulting in a greater competitive advantage of horticultural commodities in the regional markets for Tanzania than for Kenya as realized in the previous section. 7.1. Market and Regulatory System for Vegetable Seeds

Vegetable seed production and distribution in Kenya is governed by the Seed and Plant Varieties Act of 1991. Under the Act, Kenya Plant & Health Inspectorate Services (KEPHIS) is the main regulatory body of the seed industry. KEPHIS is a parastatal in which Ministry of Agriculture (MoALRD), Kenya Agricultural Research Institute (KARI), Kenya Seed Company (KSC), Horticultural Crop Development Authority (HCDA), Kenya Forestry Research Institute (KEFRI), Kenya Farmers Association (KFA), Kenya Farmers Union (KFU) and a member who may be co-opted to represent interests of the industry sit in the Seed Regulation Committee (SRC). The SRC is mandated to formulate and recommend developmental policies for the growth of the industry. It is also supposed to make standards, recommend registration of seed merchants, recommend certification fees and act as moderator in cases of appeals by aggrieved industry players. KEPHIS has the responsibility of administering the Act and is empowered to inspect, test, certify, control imports and enforce the seed law. Figure 7.1 shows the structure of the seed industry in Kenya. The seed law in Kenya is comprehensive and very stringent, and KEPHIS has concentrated most of its efforts on enforcing this law -- policing the activities of the private sector -- rather than acting in their interests to offer quality, timely and satisfactory services. The private sector has complained of the way KEPHIS has adopted an attitude of ‘policing’ the industry and exercising regulatory monopoly powers over the private sector rather than providing services, co-ordination and leadership in the seed industry. They suggest that this approach has been an impediment to private sector innovation & growth and has usurped the developmental role in the industry.

58

Figure 7.1. Structure of the Seed Industry in Kenya

Breeder seed Breeder seed Foundation seed Certified seed Sale of certified seed Sale of certified seed Administration KEPHIS regulations Breeder seed Foundation seed Certified seed

KARI Breeder material development

Seed Production Unit (through informal groups)

Kenya Seed Company (KSC) PRIVATE COMPANIES

AGENTS/STOCKIST

FARMERS

CONTRACTED FARMERS

Seed bulking to produce Certified

Seed

MINISTRY OF AGRICULTURE

ADC: Bulking Foundation

59

In light of these difficulties, the recent process of reform in the Tanzanian seed industry may hold some useful lessons for Kenya. The Tanzanian seed sector had also been tightly controlled particularly for cereals. Figure 7.2 shows the seed industry structure before liberalization. The main regulatory body of the Tanzanian seed industry was the Tanzanian Official Seed Certification Agency (TOSCA). Its mandate was to ensure that seed produced in the country met the required standard of quality and the rules and regulations were followed. It was empowered to inspect, test, certify and control seed imports. Breeding of plant varieties, production of the seed and seed treatment was mainly done by the Government. The law mandated the research stations to do the breeding work, then to pass the pre-basic seed to five Government seed multiplication farms under the Ministry of Agriculture, which were required to multiply and produce foundation seeds that would be sold at a very cheap price to the Tanzania Seed Company (TSC). TSC was a Government parastatal and was mandated to take all the foundation seeds that the five farms in the country multiplied. It then contracted farmers to grow the certified seed, which the company would process, package and then sell to companies/stockists for distribution to farmers. Retail sales prices were heavily subsidized. This put the company in a difficult financial situation since it could not raise enough operational funds. Thus, due to lack of capital and the Ujamaa policy and government price controls, TSC collapsed in the early 1990’s. By then, farmers had also lost confidence in the quality of hybrid seeds that the company was selling and turned to Open Pollinated Varieties (OPV’s) or imported hybrid seeds. In 1994, the Tanzanian Government liberalized the economy and the seed industry took a different structure and operations (see Figure 7.3). TOSCA and the five state farms remained, but government monopolies were eliminated. Research stations are now free to sell breeder seed to private companies or to the government seed multiplication farms, or to contract capable farmers to undertake multiplication of foundation seeds for them. They can also do seed bulking to produce certified seeds or contract farmers to do this. The five state farms took over the place of TSC. They can sell certified seed directly to the farmers or sell the foundation/certified seed to the private companies. Currently the research stations and the state farms are supposed to partially fund their programs from profits accrued from their operations. The current system appears to be competitive and dynamic. Following liberalization in Tanzania, the Seed Act was reviewed to reflect the current government policy of a free economy, and is currently pending in Parliament. A key innovation in the Act is that it allows seed to be produced at village level as Quality Declared Seed (QDS), which has its rules and regulations printed and applied by TOSCA. This process of village level seed production is operationalized as Community Based Seed Production (CBSP). The Tanzanian Parliament passed the Plant Breeders Rights Bill in 2002 and this was seen as a positive move for the seed industry. Liberalization of the seed industry in Tanzania had major impacts on the vegetable seed industry. Prior to liberalization, permits and licenses were difficult to obtain for importers. Large amounts of vegetable seed were brought in through the Ministry of Agriculture by NGOs for sale to farmers at subsidized prices. Some few established private companies (Pop Vriend, Sluis Brothers, Rotian Seed, and others) were given permits to import vegetable seeds, but the process was quite tedious and the quantities imported were relatively small. Competition from NGOs undoubtedly hindered the development of a competitive private industry. Liberalization streamlined the import process, and now most vegetable seed imports in Tanzania take place through commercial channels (see Figure 7.4

60

Figure 7.2. Structure of the seed industry in Tanzania Before Liberalization Free breeder material Foundation seed: Sell at a low price Administration TOSCA regulations Breeder seed Foundation seed Certified seed

RESEARCH STATIONS Breeder material development

FIVE STATE FARMS Multiplication of the breeder

Material to produce foundation Seeds

TANZANIA SEED COMPANY.

AGENTS/STOCKIST

FARMERS

CONTRACTED FARMERS

Seed bulking to produce

TOSCA

MINISTRY OF AGRICULTURE

COMMUNITY BASED SEED

61

Figure 7.3. Structure of the Liberalized Seed Industry in Tanzania Breeder material Foundation Certified Foundation/ Certified Sell Certified seed Administration TOSCA regulations Breeders seed Certified seed Foundation seed Foundation/Certified seed

RESEARCH STATIONS Breeder material development Multiplication of foundation seeds.

Also produces certified seed

FIVE STATE FARMS Multiplication of foundation

seed Multiplication of certified seed

Packaging of certified seed

PRIVATE COMPANIES Multiplication of foundation

seeds Multiplication of certified seeds

AGENT/STOCKISTS

FARMERS

MINISTRY OF AGRICULTURE

COMMUNITY BASED SEED PRODUCTION

CONTRACTED FARMERSTo multiply foundation seedand produce certified seeds

for research stations

62

for the current structure of the vegetable seed industry in Tanzania). Testing for seed viability remains quite stringent, to avoid dumping. Local vegetable seed production is also very active in Tanzania. This began prior to liberalization, but has taken on a new dynamic since that time. The Ministry of Agriculture recently started a seed production unit under the National Vegetable Seed Program. The unit is situated at Tengeru Horticultural Research Institute (THRI) and is mandated to produce vegetable seeds for the country. It has joined with companies such as Alpha Seeds to help in distribution of the seeds produced locally. FAO started a vegetable seed-producing program that has been handed over to locals and is functioning successfully. There is also an on-farm Figure 7.4. Structure of Vegetable Seed Sub-Sector in Tanzania since Seed Sector

Reforms

Administration TOSCA regulations Breeders seed Certified seed

Ministry of Agriculture National Seed Production Program

Tengeru Horticultural Research Station Vegetable Seed Production Unit

Seed Companies e.g Alpha seeds Seed importers

Stockists/Agents

F A R M E R S

Community Based Seed Production (CBSP)

TOSCA

63

seed production program in which extension agents train farmers in the production of quality-declared seeds, and inspect the seeds for quality. This is done on about seven vegetable crops. There was a similar program in Mangola where farmers were trained to produce quality onion seeds for themselves and for companies. This program developed the Mangola Red variety from Bombay Red, with higher yields and a longer storage period – a key characteristic for onions. To determine the status of vegetable seed markets a survey of the major horticultural seed companies was undertaken in Nairobi, Kenya and Arusha, Tanzania. Table 7.1 shows the sources of the seed inputs sold by these companies. Table 7.1. Sources of Vegetable Seed Inputs Sold by Various Companies in Kenya

and Tanzania Source of seeds Kenya Tanzania Number of Main Companies selling horticultural seeds

6 7

Imported seed only 3 1 Local seed only 0 1 Both local/imported 3 5

Overall local seed share 5% 8% Source: Authors computation

In Tanzania only one of the seven companies interviewed operated exclusively with imported seed. In Kenya, three of the six companies interviewed do so (Table 7.1). One company in Tanzania trades exclusively locally produced vegetable seeds while in Kenya no firms operate in this way. Five of the seven companies in Tanzania and three of the six in Kenya deal with a mixture of both local and imported seeds. Over all companies and all types of vegetable seeds, the share of local seeds in Kenya is 5%, and in Tanzania is 8%. Table 7.2 shows the vegetable seeds that are produced locally, the percentage production and the number of companies doing it. Table 7.2. Vegetable Seeds Produced Locally and the Approximate Percentage of

Production by Companies Company Commodity Local Seed Share (%)

Kenya Simlaw, E.A seed (2) Pepper, okra, beans 5.0 Tanzania: Suba Agro Chem Watermelon, okra, eggplant 2.5 E.A seed, Kibo (2) Pepper, okra, beans 5.0 Pop Vriend Ltd Cabbages, tomatoes 80 Alpha Seed Ltd Black night shade, okra, cucumber

watermelon, amaranthus 100

Source: Authors’ Interviews with Seed Companies.

Table 7.3 shows prices per kg of various vegetable seeds in Kenya and Tanzania. Neither country has systematically lower prices than the other. The results show that there is no clear-cut trend in price differences between companies in Kenya and Tanzania. However, for those seeds produced locally (either in Community Based Seed Production programs or by private companies), the prices are relatively lower. The most notable price difference is the very low price for Mang’ola red variety of

64

onion in Tanzania produced under CBSP, whose price is one-third and 40% that of the Red Bombay in Tanzania and Kenya, respectively. Table 7.3. Average Prices of Various Vegetable Seeds in Kenya and Tanzania.

Kenya Tanzania Seed Variety

Imported Seed Imported Seed Locally Produced Seed

(CBSP and private Companies)

------------------------------ Ksh/kg ------------------------------ Onions Red Creole 2958 3250 Red Bombay 1500 1813 1250 Mang'ola red - 600 Tomatoes Cal-J 4700 4580 Money Maker 1550 4200 2500 Cabbages Gloria 1869 1392 Copenhagen 1309 1755 Drumheads - 1321 Kales Thousand headed 1350 1189 Eggplant 2000 1667 1667 Amaranth 800 - 740 Watermelon 1500 1800 1250 Source:Authors’ Interviews with Seed Companies

Table 7.4 shows the various seed import charges in Kenya and Tanzania. Though neither country has an import duty on seed, other charges substantially increase prices in each country. Overall, these charges add about 18% to the CIF price of seed in Kenya; they add about 15% in Tanzania, plus a fixed charge of TzSh10 per import. Thus, financial costs of importation in Tanzania are slightly lower; combined with the greater bureaucratic difficulties placed on imports in Kenya, seed importation becomes a somewhat more costly and time-consuming activity in Kenya than in Tanzania.

65

Table 7.4. Seed Import Charges in Kenya and Tanzania Charges Kenya (Kshs) Tanzania (Kshs)

Clearing charges Negotiable (5 %) 6.50% Inspection fees 6.4 % 1.7% Import Declaration Form (Plant Import permit-TZ)

2.75% $10

Port charges 2.25% 4% Agency fees 1% 1.5% Import cargo handling 1% 1.5% Total Charges 18.4% 15.2% + $10 Source: Authors’ Interviews with Seed Companies

The Tanzanian experience would seem to hold at least two lessons for Kenya. First, a heavy focus on policing by the seed regulatory body is likely to hinder the development of a flexible system capable of providing quality seed from a variety of sources to farmers. Thus, the on-going review of the Seed Act in Kenya should as a matter of necessity address the issues of how KEPHIS and the regulation committee can play a facilitating role to coordinate the development of a seed regulatory framework, which can spur innovation and growth without necessarily compromising the integrity of the industry. Second, community based production of Quality Declared Seed can be a cost effective way of making more seed of higher quality available to many small farmers at more affordable prices. Tanzania’s strong support to local seed production, and the emergence of the much improved Mang’ola Red variety of onion from one of these efforts, appears especially noteworthy. 7.2 Vegetable Research and Development in Tanzania and Kenya

Horticultural research in Tanzania is mainly done in Tengeru Horticultural Research Institute and to some extent in Sokoine Agricultural University. Research is almost entirely on breeding and other farm-level technical issues. Post harvest handling, packaging, and storage, receive little if any attention. Tengeru is one of the two research institutes in Northern Zone. The institute began work in the 1980s, and was mandated to do both research and training in horticulture. After 1996, the training role was withdrawn to enable the institute to concentrate on research. The Government and donors such as World Bank fund the institute. It is also involved in collaborative research work with other organizations. The research work undertaken is driven by the needs and challenges that farmers and NGOs are facing, and includes vegetable breeding, mushroom growing, banana research, root and tubers, flower pest control and fruit trees among others. The vegetable seed production unit, which is part of the National Vegetable Seed Program under the Ministry of Agriculture, is also based in Tengeru Institute. The unit is mandated to produce vegetable seeds in the country. The unit has plant breeders who have specialized in vegetable seed breeding and has seed production technologists. There is seed production for indigenous vegetables also but very few are purified and the process is on-going. Research work in this area is at the initial stages of collecting germplasm for the various indigenous vegetable varieties. There is also a regional research program of indigenous vegetables in

66

Tanzania carried out multilaterally for Southern Africa Development Community (SADC) countries. Through research, the institute is trying to enhance production and utilization of quality vegetable seeds. The farmers’ response to the new seeds has been good. Most farmers like the indigenous vegetables especially women as it forms a good source of income. The crops are commonly referred to as “women’s crops”. Most people have accepted the indigenous vegetables as a rich diet hence the demand is growing. So far, the vegetable breeding section in Tanzania has developed fourteen varieties, which are commercially produced for the market by the Seed Production Unit. They are shown in Table 7.5 below. The Seed Unit has forward linkages with some companies e.g. Alpha Seeds to help in marketing and distribution of the seeds produced locally. Table 7.5. Crop and Seed Varieties Produced in Tanzania. Crop Varieties Tomatoes Tengeru 97, Money Maker (Roma), Tanya Onion Red Bombay, Mang’ola Red Okra Pusa Sawani Cucumber Ashley Cabbage Sugar Loaf Water Melon Sugar Baby Amaranth-Mchicha Wa-Unga and Nyeupe, Black Nightshade Mnavu Loshuu Local Ngowe Tengeru White African Eggplant Black Beauty. Source: Tengeru Horticultural Research Intitute. In Kenya, vegetable research work including breeding is undertaken by the Thika Horticultural Research Centre (THRC). The centre carries out basic and strategic research while other centers specialize in adaptive horticultural research work. The institute is a parastatal funded mainly by the Government though it has also received funding from organizations such as UNDP, FAO, USAID, CIDA, JICA, Netherlands Government and World Bank. The private sector has supported THRC on pest control research. Some of the work has been undertaken on tomatoes, French beans, runner beans, some Asian vegetables and indigenous vegetables. Research work on fruits has been mainly on citrus and bananas. Early 1990s the Center started involving stakeholders in planning research programs at the beginning of each year. With this participation in planning, it is now easy to approach the private sector for research funding. The institute hopes that by and by the private sector will support research in horticulture. KARI has developed an Agricultural Research Investment Scheme (ARIS), which has a Business Development Unit that publicizes research activities to the industry and finds out ways through which they can play a role in research programs. It has produced a brochure detailing the various research activities and services that can be offered. It is hoped that in the long run, up to 30% of research funding will come from the private sector through the Business Development Units. Understaffing seriously limits the scope of what THRC can do. While Tengeru in Tanzania has five breeders, a number of seed technologists, and a fully operational seed production unit, THRC has two breeders (one of whom is the Director and thus has little time for

67

breeding work), no seed technologist, and no operational seed production unit. The Centre developed and released the Kutuless variety of French beans, but focuses most of its breeding work on maintaining the varieties already bred. There is no strategic basic breeding research being done to develop new varieties. THRC purchased a seed production unit but has not installed it due to lack of personnel in the area of seed technology. As a further point of comparison, there are eight maize breeders in KARI distributed in various research centers and targeting different agro-ecological zones in the country. 7.3 Research and Development on Fruits

7.3.1 Citrus Fruits

Citrus yields in East and Southern Africa range from 10-40 tonnes per hectare (Kolade and Olaniyan, 1998), while potential is up to 75 tonnes per hectare under high density planting. In Kenya, yields between 1992 and 2001 averaged from 8 to 11 tonnes/ha. As early as the 1970s it was noted that citrus production in Kenya was declining. Consequently, in 1982 the Director of Agriculture formed a committee to investigate the status of citrus production in the country and the constraints that farmers were facing. It was expected to highlight and give recommendations on how to eradicate and/or control the citrus greening disease, which caused low yields in orchards. It was realized that the disease was caused by a Citrus Psyllide;- Trioza erytrae vector, prevalent in the highlands. It was also observed that the disease was prevalent in most of the citrus growing areas in Kenya and that most nurseries in the high and midlands where the pest is rampant were diseased. The committee noted that the disease could be controlled through citrus tissue culture to produce disease free planting materials, but stressed that the vector Citrus Psyllide must be controlled to avoid contaminating disease free materials once they are planted in the field. Following these recommendations, University of Nairobi and Thika Horticultural Research Centre started programs in an attempt to combat this disease. THRC identified Matuga, Kibwezi, Garissa and Pekerra as sites that were not prone to the greening disease. The sites were mostly below 800m where the Citrus Psyllide does not thrive.13 Disease free planting materials from Corcocica in USA and France were introduced in these areas. In 1997, University of Nairobi received full funding from Biotechnology Trust Africa for a tisuse culture project for citrus. As a result, a tissue culture laboratory was set up in the Department of Crop Science and research work has been going on ever since. The scientists have managed to produce disease free planting materials, which have been adopted by farmers. The department currently sells about 20,000 seedlings per year at Kshs. 60 and Kshs. 100 per seedling to project and non-project farmers respectively. However, the current level of supply is far below the quantities demanded by farmers. Most farmers who receive the plantings are developing new orchards since they can start with disease free planting materials.

13 The citrus greening disease is of two types; the African and Asia greening disease. The African greening is prevalent in areas above 800m elevation while the Asian greening is prevalent from sea level up to 800 meters. It is most common in South East Asia. However, the vector can transmit both diseases.

68

A key potential problem is that the insect vector has not yet been controlled and can still infect disease free biotechnology seedlings in the field. This problem is especially serious in the midlands and highlands where the vector thrives. University of Nairobi has been working to find ways of introducing Integrated Pest Management (IPM) to bring this vector under control. The vector has several endemic parasites that can be used to control it. However, funding for this initiative has been the main bottleneck. As a result, most nurseries and orchards in the high and midlands where the Trioza erytrae is rampant continue to be diseased. Research work on fruit planting materials in Tanzania especially for citrus fruits is carried out in Sokoine University of Agriculture. Because most citrus production areas in this country are in low altitudes, and officials there – as well as in coastal and other low-lying areas of Kenya -- report no problems with the disease. Research in Tanzania can thus focus on other issues. The Horticultural Department has a supply of good quality mother plants and has a commercial nursery that can supply high quality young trees. Citrus fruits are mostly grown in Tanga and Morongoro regions. Some farmers in these areas have mastered the budding (grafting) technique for the vegetative propagation of citrus. Young trees for replacing old or diseased trees as well as trees required for expanding orchards are either produced by the farmers themselves or purchased from other farmers that have small nurseries. Local orange varieties are raised from seedlings. Unlike in Kenya, Tanzania’s competitive advantage in citrus lies in maintaining relatively large potential areas that are free from diseases and relatively high yielding varieties that have been developed. 7.3.2 Bananas

The Banana industry in Kenya faced major challenges that led to reduced yields, widespread loss of banana orchards and lack of expansion in the 1970s and 1980s. The main difficulty in the industry was lack of clean planting materials (Kahangi 1996). The conventional method of generating banana planting material through suckers is not able to produce enough for sustained expansion of existing varieties or new improved clones, and also contributes to the spread of disease such as Sigatoka and Panama wilts and pests such as banana weevils, nematodes, and streak virus. As a result, yields in Kenya between 1992 and 2001 ranged between 10 and 15 tones per hectare as opposed to a potential yield of over 40 tonnes/ha (Kahangi 1996). Many farmers in Kenya have had to uproot their plantations due to disease epidemics. These problems caused a major decline in the banana industry in Kenya and in the late 1980s there was a public outcry to do something to save the industry. The decline in production resulted in high banana prices in the local markets as well as attracting imports from Uganda and Tanzania. It was against this background in 1987 that Jomo Kenyatta University of Agriculture and Technology (JKUAT) through Institute for Biotechnology Research (IBR) started research in tissue culture of bananas. The Institute submitted a proposal to establish a pilot commercial tissue culture laboratory to UNESCO and the World Bank. The overall aim was to enable multiplication of adequate disease free seedlings for the small-scale farming sector. A banana tissue culture laboratory was set up and IBR started by producing 10,000 biotechnology suckers per year which it sold to farmers at Kshs. 60 per sucker. UNESCO funding helped expand this capacity to 500,000 plantlets per year. Following this technology breakthrough in bananas and adoption of clean planting materials by farmers, there was an observed upward swing in the quantities produced from 1995 and an increase in the area under bananas from 1996 (see Figure 2.1 and Appendix A, Figure A.1). Note that because

69

there is no vector transmission of the banana diseases, the problem could be solved by progressively introducing clean planting material, which did not face the risk of infection through vector transmission. In citrus, vector transmission of the citrus greening pathogen makes the challenge much greater. See section 7.4 for more on this issue. 7.4 Fertilizer and Agrochemical Inputs

In 1993, the Kenya Government withdrew from fertilizer distribution and since then, it has relied on the private sector and cooperatives to meet the fertilizer needs for farmers (Wanzala et al, 2001). Approximately 95% of the fertilizer consumed in Kenya is imported and distributed by the private sector. The remaining 5% is donor-sourced by the Ministry of Agriculture-KRII program, which imports fertilizers, and sells to private traders via an open tendering system. After the revival of Kenya Farmers Association (KFA), it also joined the market. In the 2001/2 financial year, National Cereals and Produce Board (NCPB) also entered the market to supply fertilizers to farmers. Hence, Kenya’s fertilizer and agrochemical industry is quite competitive. The use of inputs such as fertilizers and agrochemicals contribute significantly to increased horticultural output. The cost of these inputs makes up a sizable component of the cost of horticultural production. In the case of onions, the cost share of fertilizers and other chemicals was 17%, 34%, 21% and 15% in Mang’ola-Tanzania, Oloitoktok, Narok/Laikipia /Meru and Taveta in Kenya respectively (see Appendix C). In a bid to lower the cost of inputs, the Kenya Government in 2002/03 eliminated duties on all raw materials not produced locally (previous duties were from 3% to 5%). The agricultural sector is now able to obtain all capital goods, fertilizer and chemicals, and other input requirements duty free. However, interviews with seven agricultural input companies indicated that the tax concession benefit was not passed over to farmers. Some companies indicated that the tax exception was too minor to be passed over. Others reasoned that since they are incurring some costs to provide extension services to farmers, this exception boosts their agricultural service provision budget. Hence the need to reduce the cost of these inputs for farmers persists. Interviews with public institutions, producer organizations and private sector companies supplying chemical inputs other than fertilizers indicated that the prices for chemicals tended to be higher in Tanzania than in Kenya and it is only the private sector which imported and supplied these inputs in both countries (Table 7.6). However, in both countries, public companies (such as NCPB in Kenya and Tanzania Fertilizer Company in Tanzania) and producer associations (KFA in Kenya and various associations in Tanzania) import and supply fertilizers at slightly lower prices than the private Sector (Table 7.8). At the present time, fertilizers appear to be more available through this channel in Tanzania than in Kenya, and the price difference between this channel and private traders is also larger in Tanzania. Table 7.6. Average Sales Price (Kshs) of Common Chemicals Among Private Sector

Companies in Kenya and Tanzania. Chemicals Units Kenya Tanzania Ambush Liter 1296 1280 Sancozeb Kilogram 159 485 Foliar Feed Liter 145 152 Dimethoate Liter 378 620

Sources: Authors’ Interviews with Chemical Suppliers in Nairobi and Arusha.

70

There is some evidence that the entrance of NCPB and KFA into the fertilizer input markets in Kenya has been associated with price reductions by private companies (Table 7.7). KFA and NCPB appear to be selling at about 10-15% less than previous private company prices in many areas of the country, and indications are that these private prices have come down in the past year. Although KFA and NCPB started with the commonly used fertilizers i.e. DAP and CAN, there is a plan to expand and supply all other fertilizers used in the country. The main objective of KFA and NCPB is to provide “one-stop” agricultural input shops for farmers and to supply quality inputs at competitive prices. Table 7.7. Average Sales Price (Kshs) of Fertilizer (50 kg Bag) for Public

Companies/ Producer Organizations and Private Sector Companies in Nairobi (Kenya) and Arusha (Tanzania) in 2003. Kenya Tanzania

Fertilizer Type

Public Companies and Producer Organizations

Private Sector

Public Companies and Producer Organizations

Private Sector

DAP 1131 1150 880 1116 CAN 1050 990 920 948 TSP - 1190 880 1024 NPK 1120 1180 1302 1308 UREA - 1050 840 1004

Sources:Authors’ Interviews with Fertilizer Suppliers in Nairobi and Arusha. To achieve the objective of lowering fertilizer costs for farmers, these institutions have undertaken following business plans: i) To negotiate for large volumes of fertilizer consignments so as to realize purchase

economies and ultimately lower prices; ii) Negotiate with shipping and transport companies so as to realize shipping and

transportation economies; iii) Form a closer working relationship with companies (both local and international) who

supply these institutions with fertilizer and iv) Start dealing with fertilizer future markets. This initiative needs to be watched very closely and also better understood. Current market share and level and type of government support need especially to be established. In recent years Kenya has registered one of the few fertilizer market liberalization success stories in SSA, with availability to farmers increasing after liberalization. In many countries, elimination of large subsidies but only partial liberalization of the sector has resulted in decreased fertilizer use. It is imperative that NCPB and KFA activities in the fertilizer market not undermine private sector profitability through unsustainable subsidies in its sales of fertilizer.

71

Table 7.8 Average Price (Kshs/50 kg Bag) of DAP for Public Companies/Producer

Organizations in Various Towns in Kenya, compared with Previous Private Sector Prices

Town Depot Public Institution/Farmer Association Prices

Previous Private Sector

Prices

% Reduction

Nakuru 1,090.00 1,200.00 9.2 Nyahururu 1,118.00 1,200.00 6.8 Kipkelion 1,118.00 1,250.00 10.6 Fort Ternan 1,118.00 1,300.00 14.0 Narok 1,110.00 1,250.00 11.2 Ziwa 1,110.00 1,300.00 14.6 Moi's bridge 1,120.00 1,250.00 10.4 Kitale 1,120.00 1,250.00 10.4 Kipkarren- Salient 1,110.00 1,300.00 14.6 Lugari 1,120.00 1,250.00 10.4 Meru 1,120.00 1,250.00 10.4 Maua 1,120.00 1,350.00 17.0 Kibwezi 1,090.00 1,350.00 19.3

Source: Interviews with Public Institutions & Farmer’s Association.

72

8. Conclusions, Recommendations, and Further Research This report has shown that, despite very high growth rates in export horticulture in Kenya, the domestic market continues to absorb at least 4-5 times more produce, by value, than does the export market. We have also shown that value added after the farm gate is at least three times greater in the domestic than in the export supply chain. At the same time, the domestic horticultural system is relatively uncompetitive in regional markets: while the country imports a substantial share of some horticultural crops, its exports to the region are negligible. We have thus referred to the dualistic nature of the current system, with an export sector of commercial farmers and some organized smallholder farmers closely linked to export companies, competing successfully in the highly competitive and quality conscious European market, while the domestic sector is dominated by smallholder farmers receiving little if any assistance and struggling in some instances to compete with imports. The domestic horticultural system is also subject to strong forces of change at the present time. Continued high rates of urbanization are expected to drive increases in demand; if per capita incomes begin once again to rise, total demand growth could exceed 5% per year. Satisfying such increases in demand year after year would be a major challenge for any commodity supply chain. The entry of supermarkets into the domestic horticultural market may extend the dualism currently seen between export and domestic systems into the domestic system itself. Both major supermarket chains are moving towards “preferred grower” procurement systems which eliminate or greatly reduce procurement from brokers in favor of direct procurement with a limited number of commercial and some organized smallholder farmers.14 This trend is driven by the chains’ concern with safety and quality, echoing similar concerns in the export market. We estimated that the traditional retail and wholesale market system supplies at least 90% of the FFV market in Nairobi, and more than that outside of the city. We also suggest that the traditional system will continue to supply a very large share of the consumer FFV market for many years to come, primarily because it is well adapted to the shopping habits and preferences of poor consumers. Yet supermarkets, especially Uchumi, are expanding rapidly and can be expected over time to compete more aggressively on price in the FFV market. The risk, then, is that the traditional system – and the small farmers and traders who primarily supply it – will be increasingly confined to the low income end of the market, with low value added, high costs, and limited profits, while commercial farmers and a small number of organized smallholder farmers dominate the expanding and more profitable direct procurement system of the supermarket chains. How to avoid this entrenched dualism, with its negative implications for smallholder incomes, rural poverty reduction, and the quality of the urban food supply, is a key public policy issue over the next five to ten years. Initiatives which help reduce this dualism will also be likely to increase the domestic system’s competitiveness in regional markets. Expanding domestic and regional markets for Kenyan horticultural produce, integrating the bulk of the country’s smallholder farmers into profitable supply chains that satisfy these markets, and ensuring consumers of a growing supply of horticultural produce with falling real prices and improving quality will require investment in three key areas: the legal and regulatory environment, technical production constraints, and “hard” and “soft” public market infrastructure. The level of investment needed in each of these areas to enhance the competitiveness of the traditional supply system is well beyond what the government alone

14 Little is know of the procurement systems of hotels and high-end green grocers, but we hypothesize that they also engage in substantial and increasing direct procurement.

73

could finance. Active partnering with donors and private sector will thus be crucial. In some cases, effective partnering will require a fundamental change of orientation on the part of government. In the rest of this chapter we briefly lay out the key issues that need to be addressed, and the knowledge gaps that need to be filled, in each of these broad areas. 8.1 Legal/Regulatory Environment

Addressing the horticultural sector’s critical constraints will require government to adopt an overarching vision of partnering with private sector and donors to expand demand and value added within the horticultural sector and facilitate greater smallholder farmer participation in this growth. Government must see its role as a facilitator and not a controller of economic activity. Seed Legislation: The seed sector is a clear example of the need for updating government’s view of its proper role in development. Kenya’s seed input sector is still strictly regulated when compared to its liberalized output markets. In Tanzania, liberalization of output markets was followed by revisions to the Seed Act to harmonize it with the new policy approach. The Tanzanian law allows and encourages seed to be produced at village level under what is termed Quality Declared Seed (QDS). This approach appears to have resulted in lower prices to farmers for some horticultural seeds, greater availability, and in at least one case (Mang’ola Red onion variety), development of a variety that has substantially improved Tanzanian competitiveness in regional markets. Kenya’s present seed legislation, the “Seed and Plant Varieties Act of 1991”, is undergoing review to harmonize it with a liberalized environment and also with regional and international treaties. However, the current draft bill may increase controls and rigidities in the regulatory environment by proposing to pass over the functions of the Seed Regulation Committee to KEPHIS Board of Management (see Section 7.1 for more detail on KEPHIS and the SRC). By so doing, the Act risks excluding private sector and farmer stakeholders from industry decision making, precisely the opposite of what the sector needs to do. There is need therefore for revisions based on more feedback from stakeholders before the bill is finalized. The objective should be to broaden the decision making process in the industry by allowing greater participation and representation of the private sector and farmers. The bill should also allow and encourage broader participation in seed production. KEPHIS and the industry players can learn from the experiences of the Tanzanian Official Seed Certification Agency (TOSCA) in coordinating and regulating the production of Quality Declared Seed (QDS) at the village level. The Horticulture Bill: In the view of many stakeholders, the 2001 Horticulture Bill is not well conceived to help the industry face its main challenges. One option which should be explored is shifting Thika Horticultural Research Centre from its present location in KARI to a position within the industry. This approach would be consistent with that in other sub-sectors of agriculture e.g. Tea Research Foundation of Kenya (TRFK), Coffee Research Foundation of Kenya (CRFK) and more recently Kenya Sugar Research Foundation (KSRF). The objective of moving the Centre out of KARI would be to give it greater flexibility and incentives to conduct responsive research in both domestic and export horticulture. Workable mechanisms to finance its activities, possibly including In levies on local markets, processors, exporters and importers, need to be established.

74

Quality and Food Safety: Future growth in the size and value of Kenya’s domestic horticultural sector requires ever greater attention to improving quality and food safety at reasonable cost. Traditional wholesale and retail markets currently pay almost no attention to quality and safety issues, and thus limit their ability – and that of smallholders – to contribute to and participate in future growth. Government can play a key role in this area if it works collaboratively with private sector traders and farmers to establish commonly accepted and workable “rules of the game”. Quality grades and standards and food safety regulations need simultaneously to recognize a) the constraints that the traditional system faces in adhering to rigorous standards and b) the increasing need for it to do so if it is not to be marginalized. Grades and standards and food safety regulations thus need to be flexible enough to acknowledge the diversity in the sector while encouraging all actors progressively to adhere to improved standards. Farmer Organizations: Smallholder farmers have both advantages and disadvantages in competing with larger growers for agricultural markets. Smallholders’ primary advantage is that, by using primarily family labor, they face negligible labor monitoring costs. Larger commercial farms which depend primarily on hired labor often have to incur substantial monitoring costs to prevent shirking. Also, because smallholder farmers tend to have fewer economic alternatives than large farmers, they will accept lower returns to their labor than will those larger producers. Both these factors tend to reduce smallholder production costs. Yet these advantages come at the cost of a small scale of operation, which increases marketing costs either for the farmer or for the trader purchasing at the farm gate. Smallholder farmers also tend to be spatially dispersed, which further increases marketing costs. Finally, the sheer number of smallholder farmers that a large enterprise such as a supermarket would have to deal with to satisfy its needs creates major logistical and informational challenges for the buyer, especially if they are focused on quality and food safety. To reduce these costs and allow smallholder farmers to better exploit their labor advantages, government must work collaboratively with donors, civil society, and private firms to enhance smallholder farmer ability to act cooperatively in response to commercial opportunities. Partnering with supermarkets to create more groups able to meet these firms’ increasingly stringent quality and food safety requirements could be especially valuable. Government need not and probably should not be directly involved in creating farmer organizations. It does, however, need to ensure that its legal and regulatory framework facilitates commercially oriented cooperation among smallholders; this cooperation can range from formally constituted groups to more strategic collaboration that may fall short of the creation of self-governing farmer organizations.15 Government could also support NGOs that have a comparative advantage in creating viable farmer groups. Intellectual Property Rights (IPRs) in Seeds: Kenya signed the 1978 International Convention for the Protection of New Varieties of Plants (UPOV), but has not signed the 1991 convention. The stated purpose of UPOV is “to provide and promote an effective system of plant variety protection, with the aim of encouraging the development of new

15 The literature on farmer organizations is vast and cannot be summarized here. However, one key criticism of devoting large amounts of public and donor resources to forming autonomous, self-governing farmer organizations is that many of the governance costs associated with such organizations do not contribute directly to exploiting commercial opportunities, and farmers thus have few incentives to bear them. On the other hand, farmers do have incentives to cooperate selectively and strategically to take advantage of commercial opportunities. The implications of this view need to be carefully considered in the design of legislation affecting smallholder organizations.

75

varieties of plants, for the benefit of society” (UPOV 2003). UPOV pursues this goal by working through the WTO and WIPO (World Intellection Property Organization) to promote IPRs focused on plant breeders. This approach has been criticized by some for not recognizing farmer and community rights to plant varieties that they have developed through traditional means over many years.16 Kenya needs to move ahead carefully in this arena to encourage farmer access to improved seed varieties while protecting farmer rights to save and replant seeds as they have for many generations. 8.2 Technical Production Constraints

Control of banana diseases through introduction of disease-free planting material has been one major success of the past 10 years in Kenya’s horticultural sector. Given the importance of bananas in rural production and marketing systems, continued progress replacing diseased orchards with clean material would seem to be a low-cost means of protecting and expanding smallholder income earning opportunities with this crop. Lessons from the experience with banana are partially relevant for citrus greening disease, a key production constraint for oranges in Kenya. The disease also reduces the quality of oranges and thus further undermines the domestic system’s competitiveness. Because the vector does not thrive at low altitudes, a phased replacement of infected orchards in those areas of the country should be a high priority. Recent biotechnology breakthroughs should make this possible (see section 7.3.1 for more detail). In mid- altitudes, eradication of the disease would require coordinated felling and disposal of all infected orchards along with planting of new disease free seedlings. This process will be much more costly and logistically complex than a phased replacement in coastal regions. Before such a program is undertaken, therefore, careful consideration must be given to its feasibility given current administrative resources, costs if successfully executed, and expected benefits. More generally, the horticultural sector suffers from diseased and limited choice of planting material for fruits, and a near absence of locally developed vegetable seeds. Planting materials from many local nurseries are diseased, low yielding and not true to type, and vegetable seeds are almost entirely imported. This situation is a result of two factors. First, underinvestment in varietal development means that Kenya’s agricultural research system has few vegetable breeders and no vegetable seed production technologists. Though seed production equipment was purchased, it has never been installed due to lack of personnel (see section 7.2 for more detail). Second, the country’s regulatory environment stresses control over facilitation. Local seed/varietal development in Tanzania has been an important part of that country’s success in capturing significant shares of Kenya’s onion and orange markets. HCDA and THRC should carry out registration of newly established nurseries, set guidelines/standards for operations and offer general advisory services to regulate and enhance quality. KEPHIS should undertake inspection and certification of planting materials from these nurseries. A combination of government, donor, and private sector funds needs to be mobilized to provide more resources for adaptive varietal research in the context of a revised seed law that encourages the production of Quality Declared Seed at the village level. 16 See, for example, Grain 2003, Business Line 2002, and Krishnan 2002.

76

8.3 Hard and Soft Market Infrastructure

Traditional wholesale and many retail markets in Kenya are congested and unsanitary. In their current condition, these markets are not an effective vehicle for expanding domestic and regional demand for Kenyan horticultural products by increasing quality and safety and reducing costs. Moreover, if these markets do not participate in this process, then supermarkets and other large actors will increasingly bypass them in favor of direct arrangements with large commercial growers and a limited number of organized smallholders. Such a development path will impose large direct and indirect costs on the Kenyan economy. Most visibly, the large mass of smallholder farmers and small traders will be confined to a slowly growing system with little value added, high unit costs, and low profit potential, while poor urban consumers will pay higher prices for a more limited range of poor quality produce. Less visibly but of great importance, market transparency will be reduced and large players will have more opportunity to exercise market power. Wholesale markets in tropical countries play key roles in “price discovery” by concentrating large quantities of produce from many sellers and then distributing it to many buyers, all in public space. This public balancing of supply and demand results in efficient pricing and public availability of price information. By contrast, private price negotiation between supermarkets or other large buyers and their large and organized suppliers does not result in publically available price information. Over time, large actors operating in this less transparent system may develop market power and use it to increase prices to consumers and lower them to farmers. Avoiding this situation will require intelligent partnering between government, donors and private sector to make selected improvements in the hard and soft infrastructure of the traditional marketing system. Traditional wholesale markets would be the central but not exclusive focus of these investments. Improvements in three key areas will help wholesale markets integrate small farmers into a growing and profitable horticultural marketing system while providing higher quality produce to poor- and middle-income consumers. First, improved logistical efficiency, especially for loading and unloading, will reduce costs and improve hygiene in the markets. Second, improved hygiene combined with logistical improvements will make these markets more attractive options for a broader range of retail outlets. Third, improved grades and standards, and more easily available information on prices and volume by grade of product, will increase market transparency and further attract customers.17 Achieving these improvements will require that wholesale market management take on a business orientation while recognizing that it is providing a partial public good by integrating smallholder farmers into a more dynamic and competitive system while also providing poor consumers with higher quality produce at competitive prices. These markets will have to attract business by providing loading and unloading services that reduce the time it takes for traders and farmers to arrive, load or unload, and depart. They will have to work with private

17 For a good example of the type of strategic marketing information that can be provided by a proactive public agency in the service of the traditional marketing system, see Costa Rica’s Mercanet web page at http://www.mercanet.cnp.go.cr/default.htm. Michigan State University has also done a great deal of work over the past 30 years on improving traditional agricultural marketing systems to reduce costs, improve quality, and thus facilitate demand expansion. See especially Harrison et al, 1987. Finally, FAO’s agricultural marketing division has done a great deal of work on linking improvements in market infrastructure to more fundamental "market facilitating" services. See http://www.fao.org/waicent/faoinfo/agricult/ags/AGSM/infrastr.htm, Tracey-White 2003, and Tollens 1997.

77

sector to develop a workable system of grades and standards. Information technology could allow potential buyers and sellers to access real-time information on prices and volumes by product grade. Cooling and washing services would improve quality and hygiene. Simple value-added services such as slicing, dicing, and simple packaging would provide a broader range of produce that, along with the improved hygiene and efficient logistics, could attract supermarkets and green grocers. This approach would be based on recognition of the potential complementary relationship between improvements that serve the traditional marketing value chain, and the small farmers and lower income consumers that use it, and those that serve the middle and upper income groups. Investments in the former can improve quality and reduce costs for the whole value chain, thereby avoiding the splintering into a dualistic system serving different clientele with vastly different qualities, choice, and costs. Growers and marketing agents who do a better job serving such an improved traditional system will also be in better positions to upgrade to meet the requirements of the higher end of the system. Active partnering between government, private sector and donors will be crucial to mobilize the needed financial resources and knowledge to make these improvements. Existing wholesale market places will need substantial physical improvements, and may need to be moved to achieve these. In many instances throughout the world, improved or new wholesale markets have not been used by the private sector for a complex set of reasons. It is thus imperative that the decision on market location be part of a broader process that focuses on modernizing and improving FFV wholesaling, and creating better links back to the assembly and on-farm production processes (Tracey-White 2003; Tollens 1997). Government and donors could also play an important role partnering with supermarkets to reduce the cost to them of dealing directly with smallholder farmers. Using government and donor funding, the extension service and national and international NGOs could bear the cost of developing viable smallholder farmer organizations and nurturing the relationships between these organizations and supermarkets. Once the organizations have developed sufficiently and the relationship with the supermarket has been stabilized, the assistance can move on to other areas. On the other hand, if investments in the traditional system are successful in improving quality and reducing costs, then traditional wholesale markets could become attractive once again to supermarkets, at least for a range of basic items. Thus, improvements in the traditional system and efforts to increase smallholder access to the direct procurement systems of supermarkets should be seen as complements, not substitutes. Macro infrastructure is also key to modernizing the sector. Ahmed and Rustagi (1987) estimate that more than half of the substantial difference in marketing costs between Africa and Asia is attributable to poor physical infrastructure (as quoted in Tollens 1997). The poor condition of most of Kenya’s roads implies high vehicle maintenance costs and consequently high transportation costs (Kamau, 2000). Similarly, lack of passable roads during rain seasons renders them inaccessible to collecting wholesalers and results in delayed transportation of perishable commodities. Improvement in secondary and tertiary roads is especially important. As the Ministry of Works allocates funds for the development of rural roads, a horticultural task force should be formed to lobby for targeting of areas that would most benefit horticulture. All of these efforts – improved hard and soft infrastructure in the traditional system, improved direct links between supermarkets and small farmers, and improved secondary and tertiary roads -- will have very positive implications for rural economic growth and poverty alleviation.

78

8.4 Other Recommendations

Duty Concessions: Unequal duties at border points may hinder the access of Kenyan produce to Tanzanian markets. Hence there is need for harmonization of duties in EAC. Kenya Revenue Authority should also harmonize the levies charged on various horticultural commodities traded across the border points. Border Point Infrastructure: All Tanzanian border points with Kenya have electricity whereas in Kenya, only two of the five border points have electricity. Electricity supply work to the remaining border points (e.g. Shimon junction to Lunga Lunga, Bisill to Namanga and Migori to Isebania) should now be completed so as to enhance computerization and installation of communication network. Telecommunications: Telecommunication services in Kenya are expensive and unreliable, imposing real costs on trade. In Tanzania, due to an open sky policy, low cost communications are available at most border points. Kenya needs to liberalize its telecommunications services so as to increase competition and reduce costs. 8.5 Further Research

This report has provided a relatively detailed assessment of Kenya’s horticultural production and marketing system from the farm up to the wholesale level. It has also identified the range of actors operating at the retail level, and has addressed the small but emerging role of supermarkets. In so doing, it has attempted to place these new entrants in the context of the broader FFV marketing system. The challenge now is to understand in more detail the behavior and performance of the traditional FFV production and marketing system, and how it can be improved to bring down the cost to a majority of consumers yet present better opportunities for growers. An additional and important question is to understand how the traditional system relates to and might be affected by emerging supermarkets. Specifically, national policy makers and municipal authorities need to understand the bottlenecks that, if relieved, could substantially reduce costs and allow this system to compete more effectively with what might be an emerging parallel and more "modern" system driven by supermarkets. If this process of identifying and relieving key bottlenecks in the traditional system is successful, then it will be possible to avoid creating an entrenched dualistic system and the ills of high costs and poor options for both consumers and farmers that such a system implies. To help guide investments in pursuit of this goal, more information is needed in the following areas: Urban Retailing: A recently completed consumer survey in Nairobi, conducted by Tegemeo Institute, will soon allow calculations of market shares for the full range of retail outlet types. In addition to these share data, one needs to establish the costs and standard operating procedures of each retailer’s procurement system, and key bottlenecks that, if relieved, could reduce their costs and increase the quality of what they offer. We have estimated that the supermarket FFV share is less than 10% in Nairobi, but will have data that directly inform this issue with the new Tegemeo survey results.

79

The role of product quality in the traditional marketing system: Formal quality standards play little if any role in the traditional FFV marketing system. Is there evidence of price differentiation based on informal grading? Is there evidence that quality differences affect the allocation of commodities to alternative retail channels? Understanding the degree and specific mechanisms of quality differentiation in the traditional system is fundamental to designing a more formal system of grades and standards that is workable and that can increase transparency and create a dynamic of constant quality improvement. Urban Wholesaling: The behavior and performance of urban wholesale markets affects costs, prices, and the distribution of benefits among farmers, traders, and consumers throughout the production and marketing system. We have noted how improvements in logistics, hygiene, information technology, and simple value added services could generate positive knock-on effects throughout the system (Chapter 7). Identifying useful specific investments in these areas will require applied research in close collaboration with retail and wholesale traders who use the current wholesale market or who may potentially use an improved market Links between urban markets and rural producers: We know relatively little about the specific mechanisms that rural assemblers and collecting wholesalers use to procure produce for urban markets. We do know (from Chapter 4) that the production and marketing of fruits and vegetables, even among smallholder farmers, is quite concentrated. To design programs that link small farmers more closely to market outlets and level the playing field among farmers and traders, one needs to know more about the system wide “price discovery” process: how are prices negotiated at key points in the supply chain, what information do farmers and traders have access to in these negotiations, and what additional information might they need to plan more market oriented production for both the traditional and emerging supermarket markets? One would also want to establish how many small farmers sell through associations, what cost and other marketing advantages these associations provide, and what if any price premia these organized farmers receive. Finally, it is important to know what the share of smallholder farmers vs larger commercial farmers is for the main horticultural crops. Rural marketing: We anticipate that many rural households will be net buyers of horticultural produce. If this is true, then the performance of the rural marketing system, including rural retailing, will affect the real incomes of net sellers and net buyers. A forthcoming rural survey by Tegemeo will be designed to shed some light on this issue.

80

REFERENCES Ahmed, R. & Rustagi, M. 1984. “Marketing and price incentives in African and Asian

countries: A Comparison”, in Dieter, E. (ed.), Agricultural Marketing Strategy and Pricing Policy. The World Bank; pp. 104-118.

Bawden, R., P. A. Sterns, S. Harris and J Berdegue (2002). Increasing Rural Household Incomes in Kenya Through Horticulture. A Design Proposal. A Report Prepared for USAID Kenya.

Berdegué, Julio A., Fernando Balsevich, Luis Flores, and Thomas Reardon (2003). “The Rise of Supermarkets in Central America: Implications for Private Standards for Qualtiy and Safety of Fresh Fruits and Vegetables”. Draft report for USAID-RAISE/SPS project.

Castells, M. , (1996). The Rise of the Network Society. Oxford, Blackwell Publishers Inc. Development Alternative Inc. (2003). “Orange Sub-Sector Study”. Private Enterprise Support

Activities Project, Tanzania. Final Report, 2003. Dijkstra, T. , and T.D. Magori ,(1995). ‘Flowers and French Beans from Kenya; A Story of

Export Success’. In S. Ellis & Y.A. Faur’e (eds) Enterprises et entrepreneurs africains. Paris: Karthala, pp. 435-444.

Dijkstra, T. , (1996). ‘Food Assembly Markets in Africa: Lessons from the Horticultural Sector of Kenya’. In British Food Journal, Vol. 98, no. 9, pp. 26-34.

Dijkstra, T. , (1997). Trading the Fruits of the Land: Horticultural Marketing Channels in Kenya. Ashgate Press.Aldershot, UK.

Dijkstra, T. , (1999). Horticultural Marketing in Kenya: Why Potato Farmers Need Collecting Wholesalers. In H. Laurens van der Laan, H., T. Dijkstra and Aad van Tilburg (eds) Agricultural Marketing in Tropical Africa. Contributions from the Netherlands. African Studies Centre, Research Series 15/1999.

Dolan, C., J. Humphrey, and Harris-Pascal, (1999). ”Horticultural Commodity Chains: The Impact of the UK Market on the African Fresh Vegetable Industry.” IDS Working Paper No. 96. Institute for development Studies. Sussex, UK.

Dolan, C., and J.Humphrey (2001). “Governance and Trade in Fresh vegetables. The Impact of UK Supermarkets on the African Horticultural Industry.” In Globalization and Trade: Implications for Exorts from Marginalized Economies, edited by O. Morrissey and I. Filatotchev. London:Frank Cass, 2001.

Dolan, C., and K. Sutherland, (2002). Gender and Employment in the Kenya Horticultural Value Chain. Discussion Paper 8. School of Development Studies and Overseas Development Group, University of East Anglia, Norwich NR47TJ, United Kingdom.

Frankford-Nachmias, C. and D. Nachmias, (1996). Research Methods in the Social Sciences. Fifth Edition. St. Martins Press Inc, Arnold, London.

GRAIN (2003). “One global patent system? WIPO's Substantive Patent Law Treaty”. Genetic Resources International, http://www.grain.org/publications/wipo-splt-2003-en.cfm.

Harris, C. , P.V. Hegarty, M. Kherallah, C. Mukindia, J.Ngige, P. Sterns, and J. Tatter, (2001). “The Impacts of Standards on the Food Sector of Kenya.” Report Prepared by the Institute for Food and Agricultural Standards , Michigan State University, for the U.S. Agency for International Development . East Lansing, Michigan.

Harris, S. , (1992). “Kenya Horticultural Sub-sector Survey”, Nairobi: Kenya Export Development Support Project.

Harrison, Kelly, Donald Henley, Harold Riley and James Shaffer (1987). “Improving Food Marketing Systems in Developing Countries: Experiences from Latin America”.

81

International Development Working Paper Reprint No. 9. Department of Agricultural Economics, MSU. East Lansing. http://www.aec.msu.edu/agecon/fs2/papers/older/idprp9.pdf

Jaffee, S. (1994). “Contract Farming in the Shadow of Competitive Markets: The Experience of Kenyan Horticulture.” In Little , P. and M. Watts. Living Under Contract: Contract Farming and Agrarian Transformation in Sub-Saharan Africa. University of Wisconsin Press. Madison, Wisconsin.

Jaffee, S. (1995). “The Many Faces of Success: The Development of Kenyan Horticultural Exports.” In Jaffee, S and J. Morton (eds) Marketing Africa’s High Value Foods. The World Bank . Washington, D.C.

Kahangi, E (1996). Biotechnology to benefit Small Scale Banana Producers in Kenya. Jomo Kenyatta University of Agriculture and Technology and Technology.

Kamau, M. (2000). The Way Forward in Export Oriented Smallholder Horticulture. Background Paper for Stakeholder Consultation Meeting , Norfolk Hotel, Nairobi.

Kamau, M. and T. Sisusle (2001). A Study of the Domestic Horticulture Sub-Sector in Kenya.

Kenya, GoK, (2002). National Development Plan, 2002-2008. Government Pinter. Nairobi. Kenya, GoK, (1994). National Development Plan, 1994-1996. Government Pinter. Nairobi. Kenya, GoK, (1991).Laws of Kenya. “The Seed and Plant Varieties Act, Chapter 326.”

Government Pinter. Nairobi. Kimenye, L. , (1995). “Kenya’s Experience Promoting Smallholder Production of Flowers

and Vegetables for European Markets.” African Rural and Urban Studies 2 (2-3): 121-141.

Kolade, J.A. and A.A. Olaniyan, (1998). Performance of 12 Sweet Orange Cultivars in South –Western Nigeria. Fruits-Paris 53:175-182.

Krishnan, B.J. (2002). “Move in join UPOV — Farmers' rights in jeopardy”. The Hindu Business Line, October 2, 2002. http://www.thehindubusinessline.com/bline/ 2002/10/02/stories/2002100200230900.htm .

Minot, C. and M. Ngigi, (2002). “Horticulture Development in Kenya and Ivory Coast”. A Paper Prepared for the IFRI Workshop on “Successes in African Agriculture”, Lusaka, June 10th –12th, 2002.

Njagi, Seraphin (1995). “A Study of Food Marketing in Nairobi”. Conducted for the Food and Agriculture Organization of the U.N. by Lusen’s Consultants.

Obukosia, S.D and K. Waithaka, (2000). Nucellar Embryo culture of Citrus Sinensis L and Citrus Limon L. African Crop Science Journal: The journal of tropical crop science and production: Vol 8; No.2 pp 109-116.

PTA, PTA project Cross-Border Trade in Food and Agricultural products Towards a Common Agricultural Policy for Eastern and Southern Africa. Ninth Meeting of the Committee on Agricultural Cooperation, Kampala, Uganda, August 25-26, 1994.

Stevens, C. and J. Kennan, (1999). “Will Africa,s Participation in Horticulture Chains Survive Liberalization ?”, IDS Working Paper No, 106. Institute of Development Studies. Sussex, UK.

Swernberg, K. , (1995). “Horticultural Exports from Kenya.” Horticultural Trade Journal 3: 3-5.

Thiru, A. (2000). International Trade in Fruits and Vegetables:Opportunities and Constraints for Kenya”, Presented at the Technical Woprkshop on Fruit and Vegetable Standards and Grades . Naro Moro River Lodge. 15th –17th August 2000.

Tracey-White, John (2003). “Planning and Desigining Rural Markets”. Marketing Extention Guide No. 4. FAO, Rome.

82

Tollens, Eric (1997). “Wholesale Markets In African Cities: Diagnosis, Role, Advantages, And Elements For Further Study And Development”. FAO Food Into Cities Collection (AC/05-97E). http://www.fao.org/waicent/faoinfo/agricult/ags/agsm/sada/DOCS/DOC/AC0597E.DOC

UPOV (2003). http://www.upov.int/en/about/mission.html . Weatherspoon, David, David Neven, Hikuepi Katjiuongua, Richard Fotsin, Thomas Reardon

(2003). “Distributional impacts of supermarkets in Southj Africa, Kenya, Zambia and Uganda”. World Bank Report. June 30.

World Bank,IFRI, (2000).Agriculture in Tanzania Since 1986: Follower or Leader of Growth? Washington D.C.

83

Appendices

84

Appendix A. Trends in Fruit Production in Kenya. Figure A.1. Area Trend Under Bananas in Kenya (1992-2001)

Figure A2: Area Trend Under Citrus, Mangoes and Pineapples in Kenya (1992-2001)

-2,0004,0006,0008,000

10,00012,00014,00016,00018,000

1992 1993 1994 1995 1996 1997 1998 1999 2000 2001

Year

Are

a (h

a)

CITRUSMANGOESPINEAPPLES

-

10,000

20,000

30,000

40,000

50,000

60,000

70,000

80,000

90,000

1992 1993 1994 1995 1996 1997 1998 1999 2000 2001

Years

Are

a (h

a)

85

Figure A.3. Area Trend Under Avocados, Pawpaw and Passion Fruits in Kenya (1992-2001)

-1,0002,0003,0004,0005,0006,0007,0008,000

1992 1993 1994 1995 1996 1997 1998 1999 2000 2001

Year

Are

a (h

a)

AvocadosPawpawPassion fruit

86

Appendix B. Trends in Vegetable Production in Kenya. Figure B.1. Area Trends Under Cabbages, Tomatoes and Kales in Kenya (1992-2001)

Figure B.2. Area Trends Under Onions, , French Beans, Carrots in Kenya (1992-2001)

-

5,000

10,000

15,000

20,000

25,000

30,000

1992 1993 1994 1995 1996 1997 1998 1999 2000 2001

Year

Are

a (H

ecta

res)

Cabbages.TomatoesKales.

-1,0002,0003,0004,0005,0006,0007,0008,000

1992 1993 1994 1995 1996 1997 1998 1999 2000 2001

Years

Are

a (h

a)

Onions.French beansCarrots.

87

Figure B.3. Area Trends Under Indegenous Vegetables, Garden Peas and Other Vegetables in Kenya (1992-2001)

-1,0002,0003,0004,0005,0006,0007,0008,0009,000

10,000

1992 1993 1994 1995 1996 1997 1998 1999 2000 2001

Year

Are

a (h

a)

Ind. VegetablesGarden peasOther vegetables

88

Appendix C : Enterprise Budgets for Onions in Kenya and Tanzania

Table C.1 . Onion Budgets (Per One Acre) for Mang’ola in Tanzania Production Items Specifications Actual costs

(in Tz shs) Cost in Kshs % cost of

TVC

Number of basins per acre 250 Land Own land

Nursery operations Ploughing and preparation 30 basins@300 9000 750.00 1.54 Seed 20 cups @ 384 7680 640.00 1.32 Irrigation (Nursery & Field) 177960 14830.00 30.54 Polytrin 4 times @10000/0.5ltr 1600 133.33 0.27 spraying labour 1000 per spray; 4 times 4000 333.33 0.69

Main field Ploughing Once @12000 12000 1000.00 2.06 Making of basins & furrows 250 @ 100 per basin 25000 2083.33 4.29 Lines for seed placement 50 per basin * 250 12500 1041.67 2.15 Planting 150 per basin *250 37500 3125.00 6.44 Chemicals Selcron 3 L @ 16500/L 49500 4125.00 8.49

CAN 4 bags @12000 48000 4000.00 8.24 Fertilizer SA 4 bags @12000 48000 4000.00 8.24

Weeding 4 times @100/basin 100000 8333.33 17.16 Uprooting @ 100/basin 25000 2083.33 4.29 Harvesting Cutting @ 100/basin 25000 2083.33 4.29

Total variable cost (TVC) 582740 48561.67 100.00 Yield (bags) 120 120 Price/bag 17796 1483 Gross output 2135520 177960 Gross Margin/acre 1552780 129398.33 Cost per bag 4856.17 404.68 Gross Margin/bag 12939.83 1078.32

Source: Authors’ Computation.

89

Table C.2. Onion Budgets (Per One Acre) for Oloitoktok in Kenya

Production Items Specifications Actual costs (in Kshs)

% Cost of TVC

Number of basins per acre 300 Land Own land 0 Nursery operations Ploughing and preparation 30 basins@30 900 2.09 Seed 2.5kg @2500 6250 14.50 Polytrin 2 litres@1600 3200 7.42 Spraying labour 100 per spray; 4 times 400 0.93 Main field Ploughing Once @2000 2000 4.64 Making of basins & furrows 300 @ 6 per basin 1800 4.18 Lines for seed placement 2 per basin * 300 600 1.39 Planting 6 per basin *300 1800 4.18

Polytrin 1400 3.25 Fungicide 2100 4.87

Chemicals

Foliar feed 750 1.74 Spraying labour 600 ; 4 times 2400 5.57

CAN 4 bags @1200 4800 11.14 Fertilizer SA 3 bags @1100 3300 7.66

Weeding 3 times @6/basin 5400 12.53 Uprooting @ 10/basin 3000 6.96 Harvesting Cutting @ 10/basin 3000 6.96

Total variable cost 43100 100.00 Yield (bags) 85

Price/bag 1447 Gross output 122995

Gross Margin/acre 79895 Cost per bag 507.06

Gross Margin/bag 939.94 Source: Authors’ Computation

90

Table C.3. Onion Budgets (Per One Acre) for Taveta in Kenya

Production Items Specifications Actual costs (in

Kshs)

% Cost of TVC

Number of basins per acre 300 Land own land 0

Nursery operations Ploughing and preparation 1 Manday 80 0.20 Seed 2kg @2200 4400 11.02 Irrigation (Nursery & Field) 6340 15.88 Polytrin 1 litre 1800 4.51 Folier Feed 750 1.88 Spraying labour 450 1.13

Main field Ploughing once @2000 2000 5.01 Making of basins & furrows 300 basins @ 6 per basin 1800 4.51 Planting 300 basins @ 6 per basin 1800 4.51 Chemicals fungicide 2100 5.26

CAN 3 bags @1000 3000 7.52 Fertilizer (Nursery & Field) Application 300 basins @10/= 3000 7.52

Weeding 3 times @6/basin 5400 13.53 uprooting 500 nets@ 7/= 3500 8.77 Harvesting cutting 500 nets @ 7/-- 3500 8.77

Total Variable Costs 39920 100.00 Yield (bags) 75 Price (Kshs/bag) 1400 Gross Output (Kshs/acre) 105000

Gross Margin/acre 65080 Cost/Bag 532.27

Gross Margin/bag 867.73 Source: Authors’ Computation

91

Table C.4. Onion Synthetic Budget (Per One Acre) for Narok, Laikipia and Meru in Kenya

Production Items Specifications Cost

(Kshs/acre) % cost

No. of basins 300 Seed 6400 12.87 Nursery Ploughing and Preparation 100 0.20 Irrigation 2767 5.57 Polytrin 2647 5.32 Foliar feed 380 0.76 Spraying labour 420 0.84 Main field Ploughing 1467 2.95 Preparation of basins & Furrows 2643 5.32 Transplanting 5pple@100*4days 3200 6.44 Fertilizer DAP 100kg (2 bags) 2500 5.03 CAN 1 bag 1305 2.62 Polytrin 2 litre 3000 6.03 Dithane 5 kg 2000 4.02 Foliar feed 2 litre 170 0.34 Irrigation labour 4167 8.38 Spraying labour 4300 8.65 Weeding labour 5 pple@100*8 6000 12.07

Uprooting 3125 6.29 Harvesting labour Cutting 3125 6.29

Total variable cost 49715 100.00 Yield (bags) 60.0

Price/bag 1241 Gross output 74460 Gross Margin/acre 24745 Cost per bag 828.58 Gross Margin/bag 412.42

Source: Authors’ Computation