ANALYSIS OF RICE PROFITABILITY AND MARKETING CHAIN: THE CASE OF FOGERA WOREDA, SOUTH GONDAR ZONE, AMHARA NATIONAL REGIONAL STATE, ETHIOPIA M.Sc. Thesis ASTEWEL TAKELE January 2010 Haramaya University ANALYSIS OF RICE PROFITABILITY AND MARKETING CHAIN: THE CASE OF FOGERA WOREDA, SOUTH GONDAR ZONE, AMHARA NATIONAL REGIONAL STATE, ETHIOPIA A Thesis Submitted to the College of Agriculture Department of Agricultural Economics, School of Graduate Studies HARAMAYA UNIVERSITY In Partial Fulfillment of the Requirements for the Degree of MASTER OF SCIENCE IN AGRICULTURE (AGRICULTURAL ECONOMICS) By Astewel Takele January 2010 Haramaya University SCHOOL OF GRADUATE STUDIES HARAMAYA UNIVERSITY As members of the Examining Board of the Final M.Sc. thesis Open Defense, we certify that we have read and evaluated the thesis prepared by Astewel Takele, entitled " Analysis of Rice Profitability and Marketing Chain: The Case of Fogera Woreda, South Gonder Zone, Amhara National Regional State, Ethiopia" and recommend that it be accepted as fulfilling the thesis requirement for the degree of: Master of Science in Agriculture (Agricultural Economics). ---------------------------------- --------------------------- ------------------------- Name of Chairman Signature Date ---------------------------------- --------------------------- ------------------------- Name of Advisor Signature Date ---------------------------------- -------------------------- ------------------------- Name of Internal Examiner Signature Date --------------------------------- -------------------------- ------------------------- Name of External Examiner Signature Date Final approval and acceptance of the thesis is contingent upon the submission of the final copy of the thesis to the Council of Graduate Studies (CGS) through the Departmental Graduate Committee (DGC) of the candidate's major department. I hereby certify that I have read this thesis prepared under my direction and recommend that it be accepted as fulfilling the thesis requirement. ---------------------------------- ---------------------------- ------------------------- Name of Thesis Advisor Signature Date ii? ? DEDICATION I dedicate this thesis manuscript to my beloved Mother Alem Kassa, who had played major role in nursing and educating me, and my Brothers and Sisters that brought me to this success. iii? ? STATEMENT OF THE AUTHOR First, I declare that this thesis is my authentic work and that all sources of materials used for this thesis have been duly acknowledged. This thesis has been submitted in partial fulfillment of the requirements for an advanced M.Sc. degree at the Haramaya University and is deposited at the University Library to be made available to borrowers under rules of the library. I solemnly declare that this thesis is not submitted to any other institution anywhere for the award of any academic degree, diploma, or certificate. Brief quotations from this thesis are allowable without special permission provided that accurate acknowledgment of source is made. Requests for permission for extended quotation from or reproduction of this manuscript in whole or in part may be granted by the head of the major department or the Dean of the School of Graduate studies when in his or her judgment the proposed use of the material is in the interests of scholarship. In all other instances, however, permission must be obtained from the author. Name: Astewel Takele Signature------------------ Place: Haramaya University, Haramaya Date of Submission: January 2010 iv? ? ACRONYMS AND ABBREVIATIONS ACSI Amhara Credit and Saving Institute ANOVA Analysis of Variance ANRS Amhara National Regional State BOFED Bureau of Finance and Economic Development CC Contingency Coefficient CSA Central Statistical Authority GMM Gross Marketing Margin GO?s Government Organizations IAR Institute of Agricultural Research ILRI International Livestock Research Institute IPMS Improving Productivity and Marketing Success LDC Less Development Countries MEDaC Ministry of Economic Development and Co-operation MoARD Ministry of Agriculture and Rural Development NERICA New Rice varieties for Africa NRRDS National Rice Research and Development Strategy OLS Ordinary Least Squares PAs Peasant Administrations RDBOA Rural Development Bureau of Agriculture S-C-P Structure -Conduct and performance TLU Tropical Livestock Unit VIF Variance Inflation Factor v? ? BIOGRAPHICAL SKETCH The author was born to his father Ato Takele Dessie and his mother W/o Alem Kassa on 8 December 1973, in Bahir Dar of Amhara Regional State. He attended elementary school at Tserse and Fasilol from 1976-1982, and attended his secondary education at Tana Haik comprehensive high School from 1985-1988 at Bahir Dar. He then joined the then Alemaya University of Agriculture in September 1996 and graduated in B. Sc. Degree in Agricultural Economics in July 1999. After graduation, he was employed at various governmental offices at different positions for eight years. He joined the School of Graduate Studies at Haramaya University for his postgraduate studies in the field of Agricultural Economics in 2006/7. vi? ? ACKNOWLEDGEMENT I express my genuine gratitude to my major advisor Dr. Moti Jaleta for his constant advice, guidance, constructive and critical comments from the very beginning of this work, until the final date of submission. My equal special appreciation goes to Dr. Berhanu Gebremedhin for his unlimited encouragement and comments provided since the inception of the proposal in problem identification and research direction. Both have much contribution in spending their little spare time for improving my work. I also extend exceptional thanks to Tesfay Abebe, Adet Agricultural Research Center Director, for his encouragement and providing assistance in transportation facilities and other field materials to accomplish this work. I express my heartfelt gratitude to Amhara Agricultural Research Institute and Rural Development for giving me the opportunity to attend post graduate studies. I am also deeply grateful to Adet Agricultural Research Center and staff members for assisting and providing all the necessary materials for my work. Especial thanks goes to Yeshiwas Admasu, Merim Ali, Mulugojam Birhan and Mastewal Chanie for their unlimited dedication in assisting me in every aspect. I would also like to acknowledge IPMS-ILRI staff members particularly Birke Enyew, and Tilahun Gebey (Research and Development Officer at Fogera IPMS-ILRI pilot learning site) for their at ease facilitation of all logistical matters. Improving Productivity and Market Success (IPMS) of Ethiopian Farmers Project- International Livestock Research Institute (ILRI) is also impassable without giving thanks for granting research fund. I wish to extend my particular appreciation to Dereje Tilahun, Abayineh Shita,Yaze Chanie, and Melese Awoke for their assistance in data collection and data encoding to accomplish this study. Special thanks go to Zelalem Nega, Gezahegn Bekele, and Akalu Teshome for their assistance in software applications of program and computer matter, my Brother, Mehabaw Takele in assisting questionnaires? administration and data collection and my friends Mulken Bantayehu, and Kaleab Tensaye for their advice and idea support. I am beholden to appreciate, Abay Akalu and Daniel Tilahun for their comment on the questionnaire. Lastly, my special thanks go to my beloved church brothers, Pastor Amdebrihan Kassie, and Pastor Zelalem Engidaw for their wholehearted encouragement, love and prayer during my work. vii? ? TABLE OF CONTENTS STATEMENT OF THE AUTHOR iii ACRONYMS AND ABBREVIATIONS iv BIOGRAPHICAL SKETCH v ACKNOWLEDGEMENT vi LIST OF APPENDIX xii ABSTRACT xiv 1. INTRODUCTION 1 1.1. Background 1 1.2. Statement of the Problem 3 1.3. Objectives of the Study 5 1.4. Scope of the Study 6 1.5. Significance of the Study 6 1.6. Limitations of the Study 7 1.7. Organization of the Study 7 2. LITERATURE REVIEW 8 2.1. Definitions of Basic marketing Concepts 8 2.2. Fundamental Approach to the Study of Marketing 12 2.3. Review of Empirical Marketing Studies in Rice and Related Crops 13 2.4. Rice Research in Ethiopia 15 2.5. Rice Ecosystem and Production Trend 17 2.6. Structure-Conduct-Performance Approach 19 2.7. Market participation 23 3. RESEARCH METHODOLOGY 25 3.1. The Study Area 25 3.2. Methods of Data Collection 28 3.3. Sampling Procedure 29 3.3.1. Producers sampling 29 viii? ? TABLE OF CONTENTS (CONTINUED) 3.3.2. Traders` sampling 29 3.4. Methods of Data Analysis 30 3.4.1. Econometric analysis 30 3.4.1.1. Heckman?s two-stage selection procedure 33 3.4.1.2. Specification of variables 36 3.4.2. Descriptive analysis 40 3.4.2.1. Analysis of market structure 40 3.4.2.2. Analysis of market conduct 41 3.4.2.3. Analysis of market performance 41 4. RESULTS AND DISCUSSION 45 4 .1. Household and Farm characteristics 45 4.1.1. Household characteristics 45 4.1.1.1. Family size and age of the household 45 4.1.1.2. Sex and education of the household 46 4.1.2. Farm characteristics 48 4.1.2.1. Land holding 48 4.1.2.2. Crop production 49 4.1.2.3. Livestock production 51 4.1.2.4. Ownership and farm implements 52 4.1.2.5. Farming experience and Income 52 4.2. Access to Services 53 4.2.1. Location and infrastructure 53 4.2.2. Credit availability 55 4.2.3. Market information and extension service 56 4.2.4. Agricultural input use 58 4.2.4.1. Chemical fertilizer and seed 58 4.2.4.2 Herbicides and insecticides 59 4.2.4.3. Labour and machinery use 61 4.2.4.4. Storage facilities 61 4.2.5. Rice marketing of farmers 62 4.3. Profit Analysis of Rice production 64 ix? ? TABLE OF CONTENTS (CONTINUED) 4.3.1. Unit and conversion factors 64 4.3.2. Gross income of paddy production 65 4.3.3. Cost of production of paddy 67 4.3.3.1. Labor cost 67 4.3.3.2. Animal power cost 68 4.3.3.3. Material input cost 68 4.3.3.4. Other costs (land tax and interest rate) 69 4.3.4. Net income / profit 69 4.4. Analysis of Econometric Results 70 4.4.1. Heckman two step results 70 4.4.2. Tobit model results 77 4.5. Analysis of Structure-Conduct and Performance 79 4.5.1. Profile of rice traders in Fogera 79 4.5.2. Characterization of marketing actors 81 4.5.3. Rice market channels 84 4.5.4. Analysis of structure of the market 87 4.5.5.1. Purchasing strategy 89 4.5.5.2. Pricing strategy 89 4.5.6. Market performance 90 4.5.6.1. Degree of buyers and sellers concentration 90 4.5.6.2. Marketing cost and margin analysis of rice traders 90 4.5.6.3. Marketing costs, gross margin and profit margin of traders 97 4.6. Production and Marketing Constraints of Rice 98 4.6.1. Producers? constraints 98 4.6.2. Traders? constraints 101 5. SUMMARY AND CONCLUSIONS 104 5.1. Summary 104 5.2. Conclusions and Recommendations 108 6. REFERENCES 112 7. APPENDICES 119 x? ? LIST OF TABLES Table Page 1. Area covered, yield and productivity of rice in Ethiopia ....................................................... 2 2. Area and production of rice and participants of farmers ,2008 ............................................ 17 3. Rice production trends in Fogera Woreda of the Amhara region ........................................ 18 4. Land use pattern of Fogera Woreda ..................................................................................... 26 5. Age, family labour and family size of households ............................................................... 45 6. Demographic characteristics of sampled farmers ................................................................ 47 7. Land holding of household head in hectare ......................................................................... 48 8. Cultivated area and yield of paddy/rice crop per hectare, 2007/8........................................ 49 9. Production of rice by sample households in qt/ha, 2007/8 .................................................. 50 10. Number of livestock owned by sample households, 2007/8. ............................................. 51 11.Ownership and farm implements of the sampled farm households .................................... 52 12. Farming experience and farm income of a farmer ............................................................. 53 13.Traveling time required to the market center and development center (in hours) .............. 54 14. Means of transportation used by sample households in rice marketing............................. 55 15. Credit availability to the sample farm households ............................................................. 55 16. Credit giving institutions .................................................................................................... 56 17. Credit purpose for households ........................................................................................... 56 18. Source of information about supply, demand and price, 2007/8 ....................................... 57 19. Quality of source of information about supply and demand, 2007/8 ................................. 57 20. Frequency of extension contact .......................................................................................... 58 21. Input utilization of farmer for rice production ................................................................... 60 22. Source of labour employed in rice cultivation 2007/8 ....................................................... 61 23. Average storage duration in months to store paddy ........................................................... 62 24. Quantity of rice sales by kebeles in quintal (marketed surplus), 2007/8 ........................... 63 25. Use pattern of rice produce at a household level ............................................................... 63 26. Profit and Cost of production of rice per hectare ............................................................... 66 27. Average cost per hectare of rice production ...................................................................... 67 28. Average labor cost per hectare of rice production ............................................................. 68 29. Average animal power cost per hectare of rice production ................................................ 68 30. Agricultural input cost per hectare of rice production for household ................................ 68 31. Cost of land rent (tax) and interest rate per hectare of rice production. ............................. 69 32. Gross income, cost and profit of paddy production per hectare by kebele ........................ 70 33. Description of dependant and independent variables used in econometrics models ........ 72 34. Factors influencing the decision to sell rice (Probit results) .............................................. 73 35. Factors influencing the level of rice crop sales/ OLS/ results ............................................ 76 36. Maximum likelihood estimates Tobit model ..................................................................... 78 37. Personal profile of rice traders ........................................................................................... 79 38. Commercial profile of rice traders ..................................................................................... 80 39. Average value of asset for traders (in Birr) ........................................................................ 81 40. Persentage of rice market outlets ...................................................................................... 82 xi? ? LIST OF TABLES (CONTINUED) 41. Barriers to entry for rice market ......................................................................................... 88 42. Education level of wholesalers and millers ........................................................................ 88 43. Marketing cost and margin of farmers or producers .......................................................... 91 44. Marketing cost and margin of assemblers. ......................................................................... 92 45. Average total cost and margin of wholesalers ................................................................... 93 46. Average total cost and margin of millers/processors. ........................................................ 94 47. Average marketing cost of rice distributors ....................................................................... 95 48. Marketing cost and margin of retailers. ............................................................................. 96 49. Summary of marketing cost, margins and profit of farmers and traders ........................... 97 50. Production, marketing and institutional problems of farmers .......................................... 100 51. Problems of wholesalers and millers in rice market ........................................................ 101 52. Main problems of retailers ............................................................................................... 102 53. Problems to millers .......................................................................................................... 102 54. Problems of assemblers .................................................................................................... 103 xii? ? LIST OF APPENDIX Appendix Table Page 1. Amount of land size and Land rent payment in birr ...................................................... 120 2. Conversion factors to compute tropical livestock unit ...................................................... 120 3. Conversion factor used to estimate man equivalent .......................................................... 120 4. Type, quantity produced and productivity of crops in 2007/8 .......................................... 121 5. ANOVA analysis of gross income, cost and profit among rice producer ......................... 122 6. Contingency table for dummy independent variables (CC) .............................................. 122 7. Variance inflation factor (VIF) test ................................................................................... 123 8. Market concentration of rice wholesalers. ........................................................................ 123 9. Rice miller?s sales list per product handled ...................................................................... 124 10. Market concentration of rice wholesalers and millers ................................................... 124 11. Wholesalers purchase sources. ........................................................................................ 125 12. Millers/processors purchase sources ............................................................................... 125 13. Farmers? sampling distribution ....................................................................................... 126 14. Traders? sample ............................................................................................................... 126 15. Producers selected administrative kebeles ...................................................................... 127 16. Cultivated area of crops in upland and low land rice production system ....................... 128 17. Farmers sample selection ................................................................................................ 128 18. Rice production, area and number of participant farmers by Woreda and Region ......... 129 xiii? ? LIST OF FIGURES Figure Page 1. The study areas south Gondar Zone and Fogera Woreda .................................................... 27 2. Rice marketing channels ...................................................................................................... 86 3. Rice Production trend in Fogera woreda in years .............................................................. 130 4. National rice production trend (2007-2009) ...................................................................... 131 5.Trends in the amount of commercial rice import (1999 ? 2008) ........................................ 132 6. Distribution of rice potential areas in Ethiopia .................................................................. 133 xiv? ? ANALYSIS OF RICE PROFITABILITY AND MARKETING CHAIN: THE CASE OF FOGERA WOREDA, SOUTH GONDAR ZONE, AMHARA NATIONAL REGIONAL STATE, ETHIOPIA ABSTRACT This study examined the profitability and marketing chain of rice in Fogera Woreda, South Gondar zone of Amhara Regional State. From the woreda, 14 peasant associations (PAs) producing rice were selected purposively and it is stratified based on the existing rice production farming system (upland and lowland), from each farming system two PAs were selected randomly. Then samples of respondents were selected randomly proportional to its population size. A total of 165 sample farm households were selected from the four PAs for the interview. In addition, market related data were collected from 25 assemblers (20 rural and five urban marketers) and six wholesalers and 10 millers at Woreta market, 21 retailers and five urban distributors at Bahir Dar market and 29 retailers at Gondar market. Both econometrics and descriptive analyses consistency used in this study. Results from the descriptive analysis show that wholesalers and millers are the most important buyers of rice from producers, about 45% and 27%, respectively. Farmers travel, on average, 1.6hr to the woreda market to sell their rice produce. The market concentration ratio is 0.77, showing that the rice market is oligopsonsitic. High initial capital and prior control of farmers is a barrier to entry in rice trading. Results from the Heckman?s two step selection model show that, market information access, quantity of paddy produced, total value of livestock unit and extension contact with farmers increase household?s probability of selling rice. Household head?s education level and total quantity rice produced were positively affecting the level of rice sale. However increase in family size decrease the volume of rice supply to the market per household. The Tobit result also revealed that quantity produced is jointly affected both the probability of market participation and volume of supply. The cost benefit analysis of rice production shows that rice production is a profitable business for farmers. The net income obtained from production per hectare of rice is Birr 5006.48. The cost margin indicate that producers obtain on average 35.97 Birr per qt, assemblers get 139 Birr per qt, millers a profit of 5.4 Birr per qt, wholesalers 9 Birr per qt, urban distributors birr 3.88 Birr per qt and xv? ? retailers around 19 Birr per qt respectively. Though, assemblers get more profit, they also incur more marketing cost. The possible recommendations forwarded are strengthening market information and extension system, intervention to increase production and productivities by using improved agricultural inputs, promoting education and trainings about rice production and marketing and finally promoting family planning are the recommended policy implications. 1. INTRODUCTION 1.1. Background The economy of Ethiopia is largely dependent on agriculture. The sector contributes 43.2% of the country's Gross Domestic Product (GDP), and about 85% of the population is engaged in it (CSA, 2004). Ethiopia has a total land area of about 112.3 million hectares (CSA, 1998). Out of the total land area about, 16.4 million-hectares are suitable for the production of annual and perennial crops. According to Ministry of Economic Development and Cooperation (MEDaC), crop production is estimated to contribute on average about 60%, livestock accounts around 27% and forestry and other sub-sectors around 13% of the total agricultural value. Rice belongs to the family ?Gramineae? and the genus ?Oryza?. There are about 25 species of Oryza. Of these only two species are cultivated, namely Oryza sativa Linus and Oryza glaberrima Stead. The former is originated from North Eastern India to Southern China but has spread to all parts of the world. The latter is still confined to its original home land, West Africa. Rice (Oryza sativa Linu) is one of the main staple foods for 70% of the population of the world. Africa produces an average of 14.6 million tonnes of rough rice in the years 1989- 1996 on 7.3 million ha of land equivalent to 2.6 and 4.6 percent of the world total production and rice area respectively. Africa also consumes a total of 11.6 million tonnes of milled rice per year, of which 3.3 million tonnes (33.6%) is imported (FAO, 1996). Rice is among the important cereal crops grown in different parts of Ethiopia as food crop. The country has immense potentials for growing the crop. It is reported that the potential rice production area in Ethiopia is estimated to be about 5.4 million hectares. According to National Rice research and document strategy (2009), the trend in the number of rice producing farmers, area allocated and production shows high increase rate especially since 2006. The number of farmers engaged in rice production has increased from about 53 thousand in 2006 to about 260 thousand in 2008. Similarly, the area allocated has increased from about 18 thousand in 2006 to about 90 thousand ha in 2008 along with production 2? ? increase from about 150 thousand tones in 2006 to about 286 thousand tones in 2008. As presented in Table 1, there is an increased trend in area allocation and production of rice in Ethiopia (NRRDS, 2009). Table 1. Area covered, yield and productivity of rice in Ethiopia Season No farmers Area(ha) Production(ton) 2006/07 53,902 18,527 na 2007/08 149868 48,966 122,302 2008/09 260328 90,547 285,924 Note: na=data not available Source: NRRDS, 2009 Shahi (1985) also explains that Ethiopia does not grow rice at present, but around 250,000 ha in the near future and around 1 million ha in the distant future could come under rice cultivation. According to Tareke (2003), four rice ecosystems were identified in Ethiopia. These are: upland rice, rain fed lowland rice (Hydromorphic), irrigated lowland ecosystem, and paddy rice (with or without irrigation). Out of the total national production of rice in 2008, 40% is produced in the Amhara regional state, 1.14% in Tigray region, 0.41% in Benshangul-Gumz, 7.23 % in Oromia, and 1.55 % in Gambella ,13.33% in Somalia, 27.18% Southern region (NRRDS, 2009). Bull (1988) estimated that about 3.5 million hectares of vertisols is found in the Amhara region, which remains waterlogged for most of the year and possible to produce food crops in these soils through better water management (drainage) and use of water loving crops such as rice. The discovery of wild rice in the Fogera plain in Ethiopia was the cause for rice production activity in the Amhara region. The pilot production was promising when Jigna and Shaga farmer cooperatives (eye-opener and risk-taker PAs) located in Dera and Fogera woredas started large-scale production of rice with the technical support of North Korean experts. However, some technical and marketing problems hindered the production and rice production was ceased when farmers' cooperatives were dismantled in 1990 (Getachew, 2000 unpublished). 3? ? Due to the demand for food and improving farmer's awareness, Fogera and Metema woreda of the Amhara region, the number of households involved in rice production and its area coverage is also increasing. According to report of NRRDS (2008), in the Amhara regional state the estimated area and production of rice was 52985 ha and 140,235 tonnes, respectively. Attempts have been made to improve the rice varieties in the Fogera area. The popular upland rice variety in the Fogera plain was X-Gigna (N. KOREA) but now three rice varieties Kokit (IRAT-209), Tigabe (IREM-194) and Gumara (IAC-164) ) were released for Fogera and similar areas. Other introduced varieties like New Rice for Africa (NERICA) are being tested for adaptation trial (Sewagegne, 2005). According to IPMS (2005), rice is sold in too many regions in the country, including Dire Dawa, Somalia and Gambella. There is also a high potential for marketing this crop even beyond its current marketing area. However, there are problems associated with rice marketing. According to Tareke (2003), these marketing problems are related to knowledge of grading, market information, lack of group marketing options (coop/unions), use of storage as marketing strategy, excessive intermediaries, price seasonality, limited number of buyers, and lack of markets. This shows that without convenient marketing systems, boosting up of production does not stimulate farmers to increase outputs at household level in particular and at national level in general. Under traditional market structure which is characterized by failure to reflect market signals, absence of quality, excesses intermediaries and imperfect competition, it calls for studying the market structure from production up to the end consumers. This study therefore helps to identify the determinants of rice supply for possible interventions and policy implications. 1.2. Statement of the Problem Agricultural marketing is the main driving force for economic development and has a guiding and stimulating impact on production and distribution of agricultural produce. The increasing 4? ? proportion of the population living in urban centers and rising level of income require more organized channels for processing and distributing agricultural products. The weak performance of agricultural markets (both input and output markets) in Ethiopia has been recognized in various studies as a major impediment to growth in the agricultural sector and the overall economy (Eleni et al., 2004, cited in Dawit, 2005). Wolday (1994) also explained that in Ethiopia the performance of agricultural marketing system is constrained by many factors such as: poor quality of agricultural produce, lack of market facilities, weak extension services which ignored marketing development and absence of marketing information. Dawit (2005) also explained that the flow of agricultural produce from the producer to the consumer involves a long chain of intermediaries, who, without creating value-added, merely keep on stretching the chain. He further pointed out; the involvement of these superfluous intermediaries has constrained the development of the sector and deprived the farmers of equitable returns. Mohammed (2007) also clearly states that the knowledge gaps in the crop sector in Ethiopia were inefficiency of the market system (which includes inefficient marketing chain, improper transmissions of price to producers and the type of product produced by farmers i.e. whether it satisfy the consumers taste and preference). Improving marketing facilities for agricultural crops in general and rice sector in particular enable farmers to plan their production more in line with market demand, to schedule their harvests at the most profitable times, to decide which markets to send their produce to and negotiate on a more even footing with traders. Besides, a proper rice marketing system is also enables, to increase production and market efficiency. Under the current situation of the rice sector in Ethiopia, the research and development gaps were identified in different producing regions of the country. Fogera Woreda is one the main producers of rice which contributes 58% of the region and 28% of the national production of rice. 5? ? In the Woreda rice is one of the food crop produced by the majority of the farmers, after teff, maize and finger millet. Study conducted by Gebremedhin and Hoekstra (2007), indicated that 72% of the households are producers of rice and about 50% of the farmers sell rice in the area. However, the nature of the product on the one hand and the lack of organized market system on the other have resulted in low producers? price. Besides, there are challenges associated with rice production and marketing mainly on Knowledge of grading, market information, excessive intermediaries, price seasonality, limited number of buyers, and lack of markets (Tareke , 2003). Despite the significance of rice in the livelihood of many farmers and income generating crop in the study area, it has not been given due attention. It is only recently that few studies have been done on rice. However, most of these studies have focused on production and were limited to a specific area and marketing aspects. Systematic and adequate information on the process of market competition, on market structure, conduct, performance; not well identified. Further more, rice marketing channels and their characteristics have not yet been studied. Hence, this study attempts to fill in these gaps. 1.3. Objectives of the Study The over all objective of this study is to analyse the rice marketing chains in Fogera woreda. The specific objectives are: 1. To examine the determinants of household?s rice supply to markets 2. To analyze the structure of rice productions costs and determine profitability of rice production in the study area. 3. To analyze the structure, conduct and performance of rice market. 4. To examine the support services (like extension, input supply, credit, and marketing services) in rice production and marketing. 5. To identify major constraints and opportunities in rice production and supply to market. 6? ? 1.4. Scope of the Study The study is limited to Fogera Woreda, ANRS, with specific crop category, rice. The commodity approach to market study will be followed to analyze the marketing chains of rice. It emphasized on different market levels, roles of market players in the market channels, price setting, the cost benefit analysis of production of rice, cost-margin for producers and traders buying and selling strategies, storage, transport and market information will be the center of the study. 1.5. Significance of the Study Marketing is the most important aspect in the development process. This is obviously due to the fact that development basically means larger size productive activities in the economy. But we can not have more of production unless the goods produced are actually sold out and selling depends on the proper marketing conditions (Prasad and Prasad, 1995). The importance of this study is to producers and to all actors in the marketing system. The performance of marketing of rice has impact on the income of producers, processors, traders and consumers too. This information could help farmers, consumers, traders, investors, and others, who need the information for their respective purposes. Since Fogera woreda is one of the selected growth corridor woreda in the region, (or rice basket of the region), detailed information on how the rice market is currently functioning and identifying the pros and cons of the marketing system helps governmental and non-governmental organizations to design appropriate intervention measures. Besides, the document also would serve as a reference for researchers to embark upon similar or related work in other parts of the country. Since Adet Agricultural Research Center was currently assigned or nominated to coordinate the national rice research work in the country, this study will also partially fill the gap in this regard. 7? ? 1.6. Limitations of the Study Collection of the traders? data was the most difficult task during the survey. Most of the time traders are reluctant to give appropriate information as they link it with tax fees. Besides, they are busy and time specific during interview. Some traders also appointed some more days to fill the questionnaire. Despite being aware of the effect of quality on price, we are able to examine its impact because the intermediaries purchase and sell rice based on their own criteria (this might be a problem in most of the agricultural markets in Ethiopia). 1.7. Organization of the Study With the above brief introduction, the remaining part of the thesis is organized as follows. Chapter 2 presents review of literature on marketing analysis from different sources. Subsequently, description of the study area and methodologies are presented in chapter 3. In chapter 4, both descriptive and econometric results are presented and discussed in detail. The last section, chapter 5, presents the summary, conclusion and policy implications of the findings of the study. 8? ? 2. LITERATURE REVIEW 2.1. Definitions of Basic marketing Concepts 2.1.1. Market and marketing The term market has got a variety of meanings. Abbott and Makeham (1979), defined market as an area in which exchange can take place. It also means the people living there who have the means and the desire to buy a product. Thus, there can be a ?local" market, a "domestic" market, and a ?world" market. The limits of this kind of market are set not by a physical boundary fence but by the ease of communication, transportation, political and monitory barriers to the free movement of goods and money. Mendoza (1995) also defined marketing as a system because marketing usually comprises several interrelated structures along the production, distribution and consumption units underpinning the economic process. According to Casavant et al. (1999), marketing encompasses all of the business activities performed in directing the flow of goods and services from the producer to the consumer or final user. These activities are usually classified into six stages. These are: production, assembly, processing, wholesaling, retailing and consumption. According to Kotler (2003), marketing is a social process by which individuals and groups obtain what they need and want through creating, offering, and freely exchanging products and services of value with others. For managerial definition, marketing has often been described as ?the art of selling products?, but people are surprised when they hear that the most important part of marketing is not selling, i.e., selling is only the tip of the marketing iceberg. 9? ? 2.1.2. Marketing channels According to Giles (1973), the term ?channels of distribution? refers to the system of marketing institutions through which goods or services are transferred from the original producers to the ultimate users or consumers. Most frequently a physical product transfer is involved, but sometimes an intermediate marketing institution may take title to goods without actually handling them. Kohls and Uhl (1990), cited in Duc Hai, (2003) define marketing channels as ?alternative routes of product flows from producers to consumers?. They focus on the marketing of agricultural products, as does this study. Their marketing channel starts at the farm-gate and ends at the consumer?s front door. The marketing channel approach focuses on firm?s selling strategies to satisfy consumer preferences. Kotler (2003) also explains marketing channels as a set of interdependent organizations involved in the process of making a product or services available for use or consumption. Most producers do not sell their goods directly to the final users; between them stands a set of intermediaries performing a variety of functions. These intermediaries constitute a marketing channel also called a trader channel or distribution channel. 2.1.3. Market chain, supply chain and value chain analysis According Harahap (2004), Undertaking a sub-sector or market chain analysis is a way of gaining insight into the (1) operations of specific market channels while focusing on their growth potential, (2) activities and efficiency of actors along the chain, (3) business support services involved, and (4) policy and regulatory frameworks. With the information from the analysis, opportunities and constraints can be identified within specific market chains, and ways can be seen to improve a defined client's capacity to compete more effectively. Lundy et al. (2004) also clearly stated that a market chain is used to describe the numerous links that connect all the actors and transactions involved in the movement of agricultural 10? ? goods from the farm to the consumer, it means agricultural goods and products flow up the chain and money flows down the chain. The term supply chain analysis is used to refer to the overall group of economic agents (a physical person such as a farmer, a trader or a consumer, as well as legal entities such as a business, an authority or a development organization) that contribute directly to the determination of a final product. Thus the chain encompasses the complete sequence of operations which, starting from the raw material, or an intermediate product, finishes downstream, after several stages of transformation or increases in value, at one or several final products at the level of the consumer (FAO, 2005). On the other hand, a similar terminology with a market chain is value chain. The term value chain has been used for more than twenty years. It refers to the full ranges of activities needed to bring a product or a service from conception, through production and delivery to final consumers. A value chain can be the way in which a firm develops competitive advantages and creates shareholder value. It can also demonstrate the interrelation and dynamic between individual businesses. A narrow economic-based definition of value chains involves identifying the serious of value-generating activities performed by an organization. A broader system approach looks of activities implemented by various actors, from primary producers, harvesters, processors, traders, service providers, and upstream suppliers to the down stream customers. Value chain analyses encompass issues such as organizational, coordination, power relationship between actors, linkages, and governance aspects. The value chain approach has been a very useful analytical tool for taking a more objective look at an organizations position in a market. It allows for examining the consequence of empowering one group (the producer) and identifying how to link them to importers and consumers. It enables analysis of the implication of who does what, at which stage in the chain, and what this means for risk, capital needed and margins. It can help to identify with whom to form partnership in the chain (Ingram, 2009). 11? ? 2.1.4. Marketing efficiency Market efficiency is defined as the movement of goods from producers to consumers at the lowest cost consistent with the provision of the service that consumers desire and are able to pay for. The efficiency of a market can be evaluated (one approach) through analyzing the existing channels according to price and service provided. The prevailing price should reflect cost plus a profit margin and the profit must be just sufficient to reward investment at the going rate of inters rate. The quality of service should be neither to high nor too low in relation to cost and consumers desire. Factors that count for efficiency can also be evaluated by examining marketing enterprises for structure, conduct and performance (Abbott and Makeham, 1981). The marketing efficiency model is stated from shepherd?s formula. Market efficiency of 100% is perfect efficiency. While above 100% is excess profit. Shepherds formula is given by (Oscar and Chukwuma, 2008). 1 I V E ?= , Where E = market efficiency, V = Value of marketed Rice (value added or profit), I= Total marketing cost. 2.1.5. Marketable Surplus According to Atteri et al. (2003), marketable surplus can be defined as the residual production of agricultural produce left with the producer after meeting his requirements of family consumption, farm needs (seed and feed), kind payments, etc. The importance of increasing marketable surplus for meeting the increasing demand for food, raw materials and other agricultural products by the non-farming population is well recognized. If the size of marketable surplus in an economy does not rise, it may well contribute a fundamental limiting factor on the tempo of development by reducing supplies available for urban consumption, for industries and exports. 12? ? 2.2. Fundamental Approach to the Study of Marketing Marketing economists have developed various approaches to study marketing that can be serving as the framework (Brason and Norvell, 1983; Mendoza, 1995). According to Mendoza (1995) marketing studies adopt different view points and approaches. For instance, the functional or marketing functions approach, the organizational or institutional approach covering all market participants, the commodity sub-system approach which combines the previous two approach; the post harvest approach which analyzes all harmful or loss- provoking elements and other causes in the transfer of products and mixed system approach. According to Casavant et al. (1999), the roll of marketing and marketing firms will be explained based on functional, intuitional, commodity system, and structural-evaluation approaches. They explained that each of these approaches is quite traditional and has evolved over time under the writings of various authors The functional approach is the study of activities performed in changing the product of the farmer into the product desired by the consumers. It involves the business activities performed by firms in the marketing system. The most common classification of the functions performed are exchange functions, physical and facilitating function. This approach allows easy identification of the utilities being created and serves to identify the activity being examined in the other approaches. Institutional approach is the second very common approach to studying marketing which emphasizing on who is doing the market function. The institutional approach identifies the business organization and managers that add utility to the product. These are the people often considered ?parasitic middlemen? by agricultural producers. This middlemen are classified as merchant middlemen (retailers, wholesalers), agent middlemen (broker and commission men), speculative middlemen (buy and sell on their own account but expect profit made from price movement), processors, manufacturers and facilitators. 13? ? Another approach receiving less emphasis in recent years is the commodity approach. This approach simply follows one product, such as cotton, and studies what is done to the commodity and who does it as it moves through the marketing system. This approach is quit simple and allows both functional and institutional approach to be combined. It is extremely useful to the person who is interested in only one product since it does allow in-depth analyses. However, this is also a disadvantage because it ignores between product and market alternative and also ignores multi-product firms. Indeed it is now rare to see a large, institutional, cultural marketing group handling only one commodity. A more recent approach to emphasize the system of marketing, dwelling on the interaction of subsystems rather than on individual function or firms is the system approach. This behavioral system allows systems to be identified with the particular problem being addressed. Systems type include input-output, which identifies motives and means of affecting the input?output ratio. The obvious disadvantage of this method is that it is abstract in nature and the reliance on intimate knowledge of individual?s firm characteristics and behavioral interactions. Such data and on intimate knowledge is seldom available. The last approach is the structural-evaluation approach. This approach evaluates the ultimate performance of the marketing system by examining the level of competition existing in the industry. The industry structure, including the number and size of firms, is combined with firm conduct, the price behavior, advertising and product development to denote a performance that can be evaluated as good or bad. This approach is used extensively by government regulatory agencies to achieve the goods of competition and avoid the evil of monopoly power. However, the lack of precise norm against which to judge performance has caused a minimal use of this approach by economists studying marketing. 2.3. Review of Empirical Marketing Studies in Rice and Related Crops Many studies conducted in analyzing the market participation and volume of sale in different crops. Abay (2005) and Rehima (2006) studied the market participation of vegetables and pepper marketing at Fogera and Siltie Zone, respectively. Their studies indicate that both 14? ? where used Heckman two step model to identifying the factors that affect the market participation and volume of sales. The results show that distance from main road, frequency of extension contact and number of oxen were found significant for onion while experience of the farmers and distance from road were significant for tomato. The identified variables found in pepper marketing study were pepper production, crop yield of the households and extension contacts. Similarly, Makhura (2001) determined the effect of transaction costs on market participation in the four commodities horticulture, livestock, maize and other field crops in South Africa. He estimated by following Heckman two-step procedure (heckit). The variables were household endowment, access to information, household characteristics and interaction factors. He also used Tobit model to answer the two questions by identifying the factors affecting the decision to participate and the level of participation at the same time. In connection to the above studies Gebremedhin and Hoekstra (2007) identified determinants of household?s market participation of three crops (teff, wheat and rice) from three districts of Ethiopia (Ada, Alaba and Fogera). For analysis, they used community level and household level data. At the household level, Probit model was used to analyse the determinants of household choice to produce these market oriented crops. Also Heckman two-steps estimation was applied for the two crops (due to data availability rice result was not given) and the result shows that distance to market place didn?t have effect on market orientation, there was a U- shaped relation between age of household head and market orientation of household in the cereal crops, availability of cultivated land, traction power, and household labour supply, are important factor that induces households to be market oriented. A survey by Tesfaye et al. (2005) identified the challenges of the rice production, utilization and marketing of rice at Fogera, Dera and Libokemke districts. The studies pointed out both production and market constraints and more recommendations were forwarded. On the same area, Wolelaw (2005) identifies the main determinants of rice supply at farm level. The study uses Cobb Douglas production function model to estimate the limiting factors. The result that identified were, the current price, one year lagged price, actual consumption in the household, total production of rice in the farm, distant to the market and weather variables were significant to influence the supply of rice. A similar study on production part, Moses and 15? ? Adebayo (2007), examined the factors determining rainfed rice production in Adamawa state (Nigeria). Production function analysis was used to analyze the factors. The result shows that two of the variables used (farm size and seed) were significantly affect the production. Also resource productivity analysis revealed that seed was over utilized, while land and herbicide were underutilized. Decreasing the quantity of seed use and increasing the size of land and quantity of herbicide respectively could increase efficiency. Duc Hai (2003) also studies the organization of the Liberalized rice market in Vietnam. The result shows that the major rice market places were competitive. That is (1) no barriers to entry are detected that influence the formation of prices; (2) there is no concentration of market shares in the hands of private companies; (3) product differentiation is not a major issue in the market; (4) information is accessible for traders. However, in the case of large- scale millers/ polishers, important barriers to entry concern access to capital, an unstable output market and proper milling technology. The study by Harahep (2004), Rice chain study in farmers? community in North Sumatra/Indonesia, shows that paddy/rice distribution was one factor that determines rice supply in consumer level. Main actors in conventional rice chains were the capital owner both in village level (small rice chain owner, and paddy retailer) and in outside village level (whole seller and big rice mill owner). These owners controlling the chains implement strategies such as a) giving credit to peasant for production and even living cost, and (b) developing human relationship with peasant. Within these strategies, the owner of chain structurally, made peasant in a high dependency to them. 2.4. Rice Research in Ethiopia The discovery of wild rice in the Fogera plain and Gambella areas in early 1970?s has initiated different governmental and non-governmental organizations to start adaptation trials on cultivated rice in different parts of the country such as Fogera plain, Chefa, Gambella, , Melka Werer, Lante, Pawe (Getachew, 2002). The Americans, Japan Oversea Cooperation Volunteers? (JOCV), Institute of Agricultural Research (IAR), Agricultural Development Department (ADD) of the Ministry of Agriculture, Tana Beles Project (TBP), Ethiopian Water Construction Authority (EWCA) International Institute of Tropical Agriculture (IITA) 16? ? and the North Korean agricultural experts were involved in rice research up to late 1980?s and they came up with encouraging results. On average, 6 tones per hectare grain yield was recorded under experimental station conditions (Getachew, 2002). Some ?improved varieties? had been released informally and extended in to the resettlement areas in Gambella and Pawe for demonstration and large scale production. In the Fogera plain of the Amhara Region also the Jigna and Shaga farmers? producers? cooperatives started large scale production of rice with the technical support of North Korean agricultural experts. The extension program of rice was very successful. However, due to the liquidation of farmers? producer?s co-operatives and the evacuation of rice producers from the resettlement areas around 1990?s, the rice research, extension and production activities were weakened. In 1993, the Ministry of Agriculture proposed a new rice research and extension program and Fogera plain was selected for its implementation. The program was handled by the Bureau of Agriculture of the Amhara National Regional State. The Bureau was conducting the research activity using the introduced rice varieties from IITA. Rice variety demonstration was also conducted in different potential areas of the region using the variety called X-Jigna, which was introduced and informally released by North Korean agricultural experts In the late 1990?s rice was first cultivated by farmers in Fogera Plain and Pawe with the support of North Korean Project and Tana-Beles Italian Project, respectively. After the phase out of these projects rice production in Fogera Plain has been continuously and enormously expanded and now becomes the most economical crop of the area. Following the introduction of rice the Fogera Plain has been transformed from year-after-year grain shortage and food insecurity to surplus grain producing one. There was initially one popular upland rice variety in the Fogera plain known as X-Gigna but now three rice varieties (Kokit, Tigabe and Gumara) were released by Adet Agricultural Research Center for Woreta and Metema areas in 1999/2000. New rice varieties (NERICA) were introduced in the region and were evaluated in the past and some of them are currently introduced in the farmer?s field (Sewagegne, 2005; Taddese, 2005). 17? ? 2.5. Rice Ecosystem and Production Trend Rice ecosystem: Rice is grown in the tropical and sub tropical regions of all continents. Because of its long history of cultivation and selection under diverse environments rice has acquired a broad range of adaptability and tolerance. Its cultivation extends over a wide range of climatic, soil and hydrological conditions. One of the main reasons for this wide range of climatic conditions is the genetic diversity of rice cultivars (Onwueme and Sinha, 1991). Rice produced in Africa in the following five ecosystems: (1) Dry land (rain-fed upland), (2) hydromorphic (rain-fed lowland), (3) Mangrove swamp, (4) Inland swamp, and (5) Irrigated ecology. The various ecosystems face many constraints. Some of these constraints are specific to particular ecosystems while others are general and cut across ecosystems and regions. Production trend: Due to the demand for food security and improving farmer's awareness, Fogera and Metema woreda of the Amhara region, the number of households involved in rice production and its area coverage is also increasing. According to Report of NRRDS (2009) in the Amhara regional state the estimated area and production of rice for farmers was 52985ha and 140135 tonnes respectively. Table 2. Area and production of rice and participants of farmers, 2008 Participant farmers Area (ha) % Production (ton) % Amhara region 211440 52985 58.52 140135 49.01 Tigray region 3600 1271 1.40 3286 1.15 Benshangual Gumuz 1474 362 0.40 1181 0.41 Oromiya region 22036 5200 5.74 20676 7.23 Somali region 5154 9920 10.96 38120 13.33 Southern region 15741 18,721 21 77,723 27.18 Gambella region 657 1314 1.45 4,456 1.56 Total 260328 90,547 100 285,924 Source: Report of NRRD, 2009 Rice is a unique food crop having several advantageous features: it grows under flooded and submerged conditions where other crops can not so, (2) because of its C 4 nature it has high 18? ? capacity of harvesting solar radiation, which is normally excess, and thereby it has high yield potential up to 50 quintals/ha under rain-fed and 100 quintals/ha under irrigation, (3) as contrast to many cereals rice is suitable for flood and furrow irrigation, (4) it is also one of the few crop plants that can grow on the same land year after year without serious soil problems, (5) it also grows under a wide range of altitude, temperature, soil acidity and alkalinity. Due to its comparative advantage of productivity from other food crops farmers in the Fogera woreda producing rice mainly for consumption and for market. Its productivity is more attracting to allocate more land for rice production. In the Woreda now there are 14 PAs which are currently major rice producing area. Table 3 shows the number of PAs, participant farmers, production trend and productivity for the last 15 years. Table 3. Rice production trends in Fogera Woreda of the Amhara region Cropping year Number of PAS Participants Total area(ha) Production (qt) Productivity (quintal/ha) 1993/1994 2 A 30 6 160 20 1994/1995 5 256 65 1625 25 1995/1996 5 494 130 1640 13 1996/1997 5 1374 487 14510 30 1997/1998 5 2957 1113 16127 15 1998/1999 11 4445 1670.5 41908 35 1999/2000 13 6158 1968 60411 35 2000/2001 14 9413 2907 66830 35 2001/2002 14 9796 3037 106295 35 2002/2003 14 11032 3346 117110 35 2003/2004 14 11583 4239 139300 35 2004/2005 14 12162 6378 288765 35 2005/2006 14 12770 6871 274860 45 2006/2007 14 12930 8014 344739 45 2007/2008 14 17300 9213 417735 45 Source: Fogera Woreda Agricultural and Rural Development Office, 2008 A - are Jigna and Shaga kebeles (cooperatives) which are an eye-opener and risk- taker PAs in the production of rice for the first time in the Fogera woreda. 19? ? 2.6. Structure-Conduct-Performance Approach Structure, conduct, and performance (SCP) analysis was developed by Bain (1968). This theory tells us that the market structure (the environment) determines market conduct (the behavior of economic agents within the environment) and thereby sets the level of market performance. It is an attempt to compromise between formal structures of economic theory and empirical observations of organizational experience in imperfect markets. It is a standard tool for market analysis (Duc Hai, 2003). According to Kizito (2008), SCP is an analytical approach or framework used to study how the structure of the market and the behavior of sellers of different commodities and services affect the performance of marketing, and consequently the welfare of the country as a whole. The definition of structure, conduct and performance differs from one author to the other, depending on the sector and region being studied and the perception of the researcher. A. Market structure: Bain (1968) as cited in Duc Hai (2003) says market structure is defined as ?the characteristics of the organization of a market which seem to influence strategically the nature of the competition and pricing within the market. Abbott and Makeham (1979) define market structure as the market behavior of the firms. In what way they compute? Are they looking for new techniques and do they apply them as early as practicable? Are they looking for new investment opportunity or they disinvesting and transforming funds elsewhere? In general, market structure can be studied in terms of the degree of seller and buyer concentration, the degree of product differentiation, the existence of entry and exit barriers, and the power distribution (Scott, 1995; Duc Hai, 2003). 20? ? Structural characteristics may be used as a base for classifying markets may be perfectly competitive, monopolistic or oligopolistic Perfect computation is an economic model of market possessing the following characteristics: each economic agent?s acts as if price is given, i.e., each acts as a price taker; the product being sold is considered a homogenous good. Product differentiation does not exist. There is free mobility of and exit of firms. And all economic agents in the markets possess complete and perfect knowledge. Pure monopoly exists when there is only one seller (producer) in the market, barriers to entry to other potential competitors from selling in this market. Oligopoly is said to exist when more than one seller is in the market but when the number is not so large as to render negligible the contribution of each. A typical oligopoly exists when, for example, three firms control over 50% of all sales of a particular good in a particular market and certain barriers prevent potential competitors from entering the market (Tomek and Robinson, 1990). B. Market concentration: refers to the number and relative sizes of buyers /sellers in a market many studies indicate that the existence of some degree of positive relation between market concentration and gross marketing margins. It is generally believed that higher market concentration implies non-competitive behavior and thus inefficiency. But studies warn against the interpretation of such relationships in isolation from other determinant factors, like barriers to entry and scale economics (Scott, 1995). Kohls and Uhl (1985) suggest that as a rule-of-thumb, a four largest enterprises concentration ratio of 50% or more is an indication of a strongly oligopolistic industry, 33-50% shows weak oligopoly, and less than 33% shows un concentrated industry. The problem associated with this index is the arbitrary selection of r (the number of firms that are taken to calculate the ratio). The ratio does not indicate the size of distribution of the firm. In most LDC, where firms? records are usually not available publicly, it would be difficult to determine such ratios. Koch (1980) lists two kinds of partial concentration indeces: The Gini Coefficient and Herfindahl Index (HHI). Both utilize market shares to determine the extent of market concentration. The Herfindahl Index is given as: 21? ? 1,2,3....ni,SHHI r 1i 2 i == ? = Where S i is the percentage market share of i th firm and the total number of firms and n, the total number of firms. The index takes into account all points on the concentration curve. It is also considers the number and size distribution of all firms. In addition, squaring the individual market share gives more weight to the shares of the larger firms which is an advantage over concentration ratio. Avery small index indicates the presence of many firms of comparable size whilst an index of one or near one suggests the number of firms is small and/or that they have very unequal share in the market. The method is limited in its application for it imposes burden in so far a more data must be collected (Admasu, 1998). C. Market conduct: Refers to the patters of behavior that trader and others market participants adapt to affect or adjust to the markets in which they sell or buy. These include price setting behavior, and buying and selling practices (Kizito, 2008). On the definition market conduct is the condition which makes possible exploitive relationships between sellers and buyers this is done via unfair price setting practice. D. Market performance: Kizito (2008) defines the market performance as the extent to which markets result in outcomes that are deemed good or preferred by society. Market performance refers to how well the market fulfils certain social and private objectives. These includes price levels and price stability in long and short term, profit levels, cost, efficiency and qualities and quantity of food commodities? other scholars defines market performance as to the impact of structure and conduct as measured interns of variables such as price , costs, and volume of output, by canalizing its level of marketing margin and their cost components, it is possible to evaluate the impact of structure and conduct characteristics on market performance (Bain, 1968; Bressler and King, 1970, cited in Pomery and Trinidad, 1995) . 22? ? The two major indicators of market performance are net returns and marketing margins. Estimating net returns and marketing margins provide indication of an exploitive nature when net returns of buyer are much higher than his fair amount. Net returns can be calculated by subtracting fixed and variable costs from gross returns. The mathematical formulation is )VC(FCVPNR II +?= ? , where, NR is Net Return, i P is price, i V , is amount, FC is fixed cost and VC is variable cost. E. Marketing cost and margin One way of defining costs is that they are all of the expenses incurred in organizing and carrying out marketing process. Another definition is the charge which should be made for any marketing activities. Assembling transport, storage, grading, processing, wholesaling and retailing, which can all be stages in the marketing chain, involves expenses. People are often ignorant of the true cost of marketing because many of these costs are hidden, and only come to light with the patient investigation of the whole marketing process. To calculate the true cost of marketing, estimates have to be made of all these implicit cost of items. We use the economist?s concept of opportunity cost for this purpose. This is defined as the benefit foregone by not using a resource in its best alternative use (Smith, 1992). According to Tomek and Robinson (1990), marketing margin is defined as a difference between price paid by consumers and that obtained by producers or the price of collection of marketing services. Menduoza (1995) also explained that marketing margin measures the share of the final selling price that is capturing by particular agent in the marketing chain. It includes costs and typically, though not necessarily, some additional income. Many researchers applied the SCP method for conducting their study on agricultural markets in developing countries. However, the SCP method has been subject to criticism; the SCP model is too deterministic to understand the functioning of imperfect markets. As most agricultural markets are imperfect markets, there is a need to develop more dynamic models showing how structure, conduct and performance interact. It means that market structure and market conduct determines market performance. In turn, market performance will influence market structure and market conduct in the long run (Duc Hai, 2003; Admasu, 1998). 23? ? 2.7. Market participation According Reardon et al. (2005), also argue that market participation is both a cause and a consequence of economic development. Markets offer households the opportunity to specialize according to comparative advantage and thereby enjoy welfare gains from trade. Recognition of the potential of markets as engines of economic development and structural transformation gave rise to a market-led paradigm of agricultural development during the 1980s. He explained further as households? disposable income increases, so does demand for variety in goods and services, thereby inducing increased demand-side market participation, which further increases the demand for cash and thus supply-side market participation. Similarly Christopher (2007), explain the answer for why smallholder market participation so important to economic growth and poverty reduction. He traces its origin to Adam Smith and David Ricardo. He explained that given a household?s desire for a diverse consumption bundle, it can either undertake production of all such goods and services for auto consumption, or it can specialize in production of those goods in which it is relatively skilled i.e., holds comparative advantage?consuming some portion and trading the surplus for other goods and services it desires but for which it holds no comparative advantage in production. Another scholar also explains that the poorest people in the world are farmers with low market participation and low agricultural productivity. Increasing either one could help to improve the other, and both could boost living standards: higher market participation could drive productivity by providing incentives, information and cash flow for working capital, while higher productivity could drive market participation since households with higher productivity are more likely to have crop surpluses above their immediate consumption needs (Ana et al., 2008). Ana et al (2008), defined market participation in terms of sales as a fraction of total output, for the sum of all agricultural crop production in the household; this includes annuals and perennials, locally-processed and industrial crops, fruits and agro-forestry. This ?sales index? would be zero for a household that sells nothing, and could be greater than unity for 24? ? households that add value to their crop production via further processing and/or storage. The measure is intended to measure market orientation or commercialization in a scale-neutral manner, independently of the household?s wealth or productivity. Its definition is ? ? ? > ?= == ? ? = = seller 0 sellernon 0 production crop sales crop index Sale J 1j ji, J 1j ji, i Study by Bellmare et al (2005) about market participation in Kenya and Ethiopia on livestock indicated that rural households had made sequential decision making rather than simultaneous decision making in market participation. Iddo Kans (2006) also examined that endowments and resource allocation decisions determines farm out put and non-farm income, and these intern determine market participation. Analysis was also conducted by Rios et al (2008) on the direction of causality between market participation and productivity on multi-county farm households. Result indicates that households with productivity tend to participate in agricultural markets regardless of market access factors. In contrast having better market access doesn?t necessary lead to productivity. The finding suggests that investment in markets access, infrastructure provide minimal, if any, improvement in agricultural productivity; whereas programs targeted enhancements in farm structure and capital have the potential to increase both productivity and market participation. Stanton et al (2000) on their study of the roll of agribusiness, explain that with increasing efforts to promote free markets, one must ask whether the impact on some agricultural producers may be less than desirable. They argue that small producers with limited access and competitive buyers may be unable to participate in new marketing opportunities. They recommended that development policy be enlarged to encompass agribusiness enterprises, however this may require, a different governmental roll, primarily in provision of basic infrastructure, transportation policies and emphasis on availability of capital and technology 25? ? 3. RESEARCH METHODOLOGY 3.1. The Study Area Based on the CSA (2007), Amhara Region has a population of 17,214,056 of which 8,636,875 were men and 8,577,181 were women. Urban inhabitants were 2,112,220 or 12.27% of the total population. With an estimated area of 159,173.66 square kilometers, this region has an estimated population density of 108.15 people per square kilometer. For the entire region 3,953,115 households were counted. This results to an average of 4.3 persons per household. The average family size in urban and rural area is 3.3 and 4.5 persons, respectively. Fogera Wereda is one of the 106 Woredas of the Amhara Regional State and found in South Gondar Zone. It is situated at 11 0 58 N latitude and 37 0 41 E longitude. Woreta is the capital of the Woreda and is found 625 km from Addis Ababa and 55 km from the Regional capital, Bahir Dar. The woreda is bordered by Libo Kemkem woreda in the North, Dera woreda in the South, Lake Tana in the West and Farta woreda in the East. The Woreda is divided into 29 rural Peasant Associations (PAs) and 5 urban Kebeles (RDBOA, 2007/8). The total land area of the Woreda is 117,414 ha. The current land use pattern includes 44 percent cultivated land, 24 percent pasture land, 20 percent water bodies and the rest for others. The total population of the Woreda is 251,714. The rural population is estimated at 220,421. The proportion of male and female population is almost similar in both rural and urban areas. The number of agricultural households is 44,168. The mean annual rainfall is 1216.3 mm, with Belg and Meher cropping seasons. Its altitude ranges from 1774 up to 2410 masl allowing a favorable opportunity for wider crop production and better livestock rearing (IPMS, 2005). 26? ? Most of the farm land was allocated for annual crops where cereals covered 51,472 hectares; pulses cover 9819.98 hectares; oil seeds 6137 hectares; root crops 1034.29 hectares; and vegetables 882.08 hectares (CSA, 2003). The major crops include teff, maize, finger millet and rice, in order of area coverage. According to IPMS (2005), average land holding was about 1.4 ha with minimum and maximum of 0.5 and 3.0 ha, respectively. The study area is one of the surplus crop producing areas and has a good potential for rice production. The area gets much of the flood water that accumulates around Lake Tana and the two big rivers, i.e., Rib and Gumara. The rivers bring eroded soil from up hill and deposit on the low land plain. The soil seems relatively deep and fertile. In the study area, rice is planted at lower slopes of an undulating landscape where the water table moves to the surface for substantial period during cropping season. In addition, rice is irrigated with water, which is diverted from the streams at the upper part of a drainage system. However, the irrigated water is usually not substantial. In Fogera and the nearby woredas water supply to rice plants is principally provided by rainfall, run-off water, and under-ground water. Bunds are usually used for rain fed rice production. The bunds serve to retain flood water, as well as rain water, which fall during the growing season (Tesfaye et al., 2005; Abaye, 2007; IPMS, 2005). Table 4. Land use pattern of Fogera Woreda Land use Area coverage/ha/ % coverage Land planted with annual crops 51472 44% Grazing Land 26999 24% Area covered with water (wet land ) 23354 20% Infrastructure including settlement 7075 6% Un productive land (hills) 4375 3.70% Forest land 2190 1.80% Swamp land 1698 1.40% Perennial crops 2190 0.20% Total 117414 100% Source: ILRI /IPMS, 2008 27? ? Figure 1. The study areas south Gondar Zone and Fogera woreda 28? ? 3.2. Methods of Data Collection The data for this study were collected from both primary and secondary sources. Primary data were collected from samples of the respondents. Sources of primary data were smallholder farmers, traders, brokers, retailers and rice millers. The data collected through a questionnaire survey includes the following: a) Data on quantity of rice marketed, price of rice supplied, total acreage of rice cultivated, expenditure on factors of production, distance from market, size of output, access to market, market information, livestock ownership, land holding, extension service contact, credit access, family size, were collected and these were used to analyse factors determining marketable supply of rice. b) Data on output produced and sold, production costs, input costs, and marketing costs were collected and used to analyse the net returns (profitability) of rice production and the cost and price information used to construct marketing costs and margins. c) Data on market information system, exchange arrangements, system of storage, transport facilities, price setting strategy, purchasing strategy, selling strategy, barriers to entry and capital were collected from sample informants using a questionnaire, and these were used to investigate the structure and conduct of the market. d) Data on input usage, credit facilities, agriculture extension service, marketing information, and institutional support activities were collected and used to analysis production and marketing support services. In addition to primary data on the above issues, secondary data like population number, agricultural inputs and output prices, land use pattern, agro-ecology, list of licensed and non- licensed traders, marketing agents and their role, marketing directions, conversion factors were collected from different sources. Secondary data sources were Woreda office of 29? ? Agriculture Rural Development, Research centers, Cooperatives at different levels, Office of Trade and Industry, and other bureaus, different publications, research studies, websites, etc. 3.3. Sampling Procedure For this study, a multi-stage random sampling technique was employed. The sampling covered farmers, traders on proportional to size basis. 3.3.1. Producers sampling For producers, a multistage sampling technique was used to draw sample units. In the selection process both Woreda agricultural office experts and IPMS experts were consulted. In the Fogera woreda, there are 5 urban and 29 rural kebeles. Out of 29 rural kebeles, 14 administrative kebeles are producing rice. These were selected purposively and is stratified based on the existing rice production farming system (up land and low land rice producing system). From each farming system two PAs were selected randomly (a total of 4 PAs were selected). Then samples of respondents from each farming system were selected randomly proportional to its population size. The sample frame of the study is the list of household obtained in the Fogera woreda of agricultural office. Hence, a total number 165 farmers were selected and interviewed for the study (Appendix Table 13,15and 17). 3.3.2. Traders` sampling According to Mendoza (1995) researchers do not agree on sample size and procedure that should be used in each segment of the marketing chain. The decisions involved were partly a function of information currently known, time and resources available, accessibility to and openness of the marketing participants as well as the estimated size of the trading population. At first in order to have the possible level of representative traders, secondary information from and discussion was made with the Woreda Trade and Industry Office, Woreda agricultural office experts and IPMS experts (Since there was a new structural change of rural 30? ? kebeles and urban kebeles). Rural assemblers were selected from two local small markets points (Maksegnt from Nabega) and Hodgebya (from Kidist Hanna) during main market days. And urban assemblers were selected from the main city Woreta during marketing days. There was no recorded data for neither rural assemblers nor urban assemblers in the trade and industry office of the Woreda. Consulting other traders, information was gathered (counting) and size of assemblers was determined by developing a sample frame. Hence, 20 rural assemblers and 5 urban assemblers, a total of 25 assemblers were selected out of 75 and interviewed by administering structured questionnaire randomly. In the case of wholesalers, milers and retailers, sample respondents were selected from the sample frame obtained from the trade and industry office of the Woreda. Based on the list of sample frame, 6 wholesalers, 10 millers and 10 retailers were selected randomly at Woreta. Similarly retailers and distributors samples were also collected from different main towns. Hence, 5 distributors and 21 retailers from Bahir Dar, 29 retailers from Gondar were selected randomly and information was gathered by administering structured questionnaire. A total of 60 retailers were selected randomly. The distributor?s data were also collected at Bahir Dar town. They are all 5 in number and all were interviewed purposively. Since there were only three brokers at Woreta, only one broker was interviewed and information was gathered through discussion (Appendix Table 14). 3.4. Methods of Data Analysis In this study, both descriptive and econometric methods were used in analyzing data from farmers and market survey. 3.4.1. Econometric analysis To look at factors that increase the level of participation in the market ideally, the OLS model is applicable when all households participate in the market. In reality, all households may not participate. Some households may not prefer to participate in a particular market in favor of another; while others may be excluded by market. If the OLS regression is estimated 31? ? excluding the non-participants from the analysis, the model would have sample selectivity bias problem (Gujarati, 2003). If only the probability of selling is to be analyzed, Probit or Logit models would be adequate techniques for addressing it. But if one is interested to know factors that influence the level of sales, at the same time, there is a need for a model that is a hybrid between the Logit or Probit and the OLS. The appropriate tool for such is the Tobit model that uses maximum likelihood regression estimation. According to Gujarati (2003) a sample in which information on the regressand is available only for some observations are known as a censored sample. The Tobit model is also known as a censored regression model originally developed by James Tobin. Some authors call such models limited dependent variable regression models because of the restriction put on the values taken by the regressand. Hence, a Tobit model answers both factors influencing the probability of selling and factors determining the magnitude of sale. Following the Tobit model specified in Maddala (1992), the maximum likelihood Tobit estimation (Tobin, 1956) with left-censoring at zero is specified as: ii m 1i i0 * i ?X??Y ++= ? = .m1,2.3,....i= (1) ,0)(Ymax Y and 0Yif, 0Y ,0Y if,YY * ii * ii * i * ii =?= >= (1a) termedisturbanc Unobserved? e,t variablindependenX et variablindependan i of cofficient? intercept an? supply quantity Y Where, 0 i th i 0 * = = = = = 32? ? The model parameters are estimated by maximizing the Tobit likelihood function of the following form: ? )X?( F ? )X?(y f ? 1 L ii 0Y ii 0Y ** ?? = ?? ?> (2) error. standard theis ? and estimates likelihood maximum tobit ofavector is ? 0y for which i over those product themeans , 0y and 0yfor which i over thoseproduct themeans 0,yY Y offunction on distributi ecummulativ andfunction density thely,respective are F and f Where i i * i ** i * i * * i ? ?>> ?? Since Tobit model has some notable limitations, it can be remedied with the use of a sample selection model in its place. Firstly, in the Tobit model, the same set of variables and coefficients determine both the probability that an observation will be censored and the value of the dependent variable. Secondly, this does not allow a full theoretical explanation of why the observations that are censored are censored (Blaylock and Blisard, 1993). Sample selection models address these shortcomings by modifying the likelihood function. According to Heckman (1979), sample selection bias may arise in practice for two reasons, first there may be self selection by an individual or data units being investigated. Second sample selection decision by analysts or data processors in much the same fashion as self selection. Selective samples may be the result of rules governing collection of data or the outcome of economic agent?s own behavior. The latter situation is known as self-selection. Statistical analyses based on those non-randomly selected samples can lead to erroneous conclusions and poor policy (Heckman, 2008). The Heckman's correction, a two-step statistical approach, offers a means of correcting for non-randomly selected samples. The first stage formulates a model for the probability of participation used to predict the probability for each individual and then in the second stage, removing the part of the error term correlated with the explanatory variables and avoiding the bias. 33? ? Though the Heckman procedure was easy to apply and it yields consistent estimates of the parameters, they are not as efficient as the ML estimates (Gujarati, 2003). Hence, in this analysis Tobit used for comparison purpose and will be discussed when ever needed. Study by Makhura (2001), Rehema (2006) also used Tobit for comparisons for market participation. Scott (1995) explained that if majorities (95%) the sampled households are market participant?s i.e. potential suppliers, then it is advisable to apply OLS model. For this study, therefore since out of 165 rice producers, 24% of the sampled households did not participate in the rice marketing, employing the Heckman?s two stage model was appropriate. Many market studies also used this model, for the study of market participation, for instance, Rehima (2006) on pepper marketing, Abay (2007) on vegetable marketing, Zelalem (2008) on poultry marketing, Woldemichael (2008) on dairy marketing, and Makhura (2001) on transaction cost barriers to market participation in south Africa. 3.4.1.1. Heckman?s two-stage selection procedure James Heckman has proposed an alternative to the ML method, which is comparatively simple. This alternative consists of a two-step estimating procedure. In the first stage, a ?participation equation?, attempts to capture factors affecting participation decision. The second stage provides heckit analysis that determines the level of participation. The probability of participation was modeled by Maximum Likelihood Probit, from which the inverse Mill?s ratios will be estimated. The specifications for Heckman?s two-stage models are as follows: i. The participation Equation: The Probit model is specified as: 1,2,.ni , ??XY iiii =+= (3) = * i y 0 Y if 0 0Y if, 1 * i * i ? > (3a) Where, * i Y is the latent dependent variable which is not observed and Y i is a binary variable that assumes 1 if household ,i sells rice and 0 otherwise. 34? ? i ? is a vector of unknown parameters in participation equation . i X is a vector of explanatory variables in the Probit regression model. i ? is random error term that are assumed to be independently and normally distributed with zero mean and constant variance. ii. Regression (OLS): Selection model is specified as: iiiii ?????Q ++= (4) Where: i Q is the volume of rice supplied to market i ? is a vector of unknown parameters to be estimated in the quantity supply equation i ? is a vector of explanatory variables determining the quantity supplied ? is the parameter that helps to test whether there is a self selection bias in market participation i ? is the error term. Lambda, which is related to the conditional probability that an individual household will decide to participate (given a set of independent variables) is determined by the formula.. () ()?? ?? ? F f i ? = 1 (5) Where, )(??f is density function and )(1 ??F? is distribution function. Econometric Software known as ''LIMDEP'' were employed (Maddala, 2001) to run the model (Heckman two-stage selection). Before fitting important variables in the models, it was necessary to test multicolinearity problem among the variables which seriously affects the parameter estimates. Several methods of detecting the problem of multicollinearity have been used in various studies. Two measures are often suggested in the discussion of multicollinearity are the variance ?inflation (VIF) factor and the condition number. VIF is defined as: 35? ? jR 2 j 1 1 )? ? (VIF ? = (6) We can interpret VIF ( j ? ? ) as the ratio of the actual variance of j ? ? to what the variance of j ? ? would have been if X i were to be uncorrelated with the remaining X?s, it compares the actual situation with the ideal situation. The conditional number is supposed to measure the sensitivity of the regression estimates to small change in the data (Maddala, 1992). As a rule of thumb, the values of VIF greater than 10 (that is, R j 2 exceeding 0.90) are often taken as a signal that the model have multicollinearity problem .The measure of tolerance can also be used, alternatively, to detect multicolinearity. The inverse of the VIF is called tolerance (TOL). That is, j 2 jj VIF 1 )R(1TOL =?= (7) When 1R 2 j = (i.e., perfect collinearity), 0TOL j = and when 0R 2 j = (i.e., no collinearity what so ever), 1TOL = Because of the intimate connection between VIF and TOL, one can use them interchangeably (Gujarati, 1995). I used VIF test for the analysis. Similarly, the Contingency Coefficient is employed as one of the means to check for association among discrete variables. It is a measure of association from cross-classification data and is computed as 2 2 ?n ? C + = (8) Where, E E)(O ? 2 2 ? = and n =Total sample size. The contingency coefficient is relatively easy to compute and satisfies the condition that it equals 0 when there is no association between the variables. However, it does have some disadvantages as a measure of association. For detecting both multicollinarity tests for continuous and dummy variables, Statistical package SPSS version 12 was used to compute both VIF and CC. 36? ? 3.4.1.2. Specification of variables Dependent variables Market participation decision (MPD): The dummy participation decision variable is the dependent variable in the first stage of the Heckman two stage estimation procedures. For the respondents who participate in rice market it is = 1, and = 0 for the respondents who did not participate in the market in the year 2007/8. Market supply (MS): It is a continuous variable which represents the actual amount of rice supplied to the market by the farm household. Independent variables Different variables are expected to determine a farmer?s decision to participate in the market and supply a certain volume of output. A number of studies revealed that farmer?s decision to participate in a market could be determined by a number of socio-economic and demographic factors. The following are hypothesized to influence market participation decision (Kinde, 2007; Rehima, 2007; Abay, 2007). Age of the household head (AGE): Age is continuous variable and measured in years. The expected influence of age was assumed positive taking the presumption that as farmers? gets older they could acquire skills and hence produce much and developed skills to participate to a market. It is also a proxy measure of farming experience. Gebremedhin and Hoekstra (2007) in their study showed that there is a U-shaped relation between age of household head and market orientation of household in the cereal crops. On the other hand, Tshiunza et al. (2001) found that younger farmers tended to produce and sale more cooking banana for market than older farmers. 37? ? Sex of the household head (SEX): This is a dummy variable. No sign could be expected a priori for this variable. It could take positive or negative signs. A study by Makhura (2001) on the households? participation process in livestock markets indicated that women are more inclined to sell their livestock than men. A study by Lewis et al. (2008) on gender difference and the marketing styles at Oklahoma wheat producers showed that men tend to sell grain more frequently then women (men trade more than women) and women tend store longer and receive 1.4 cents/bushel less than men. Family size (FS): This is the total number of family members that can be taken as a proxy for the level of consumption. This continuous variable is expected to influence participation decision and supply negatively. Study by Chauhan and Singh (2002) in India, indicated that the marketed surplus is negatively related with the size of family and level of consumption. Education level of the household head (EDU): This variable hypothesized to affect marketable supply positively. It has dummy values. Extension frequency (EXC): This is a dummy variable indicating the extension service farmers were getting. This variable was expected to influence participation and supply positively. Obviously, as farmers learned more and knew much it would be obvious that they would produce much and ultimately participated in a market. Distance from market (MRD): This is a variable used to measure access to markets measured in travel hours for a feet single trip. It is a continuous variable and expected to influence participation and supply negatively. Again Makhura (2001) explained that those households located closer to market centers will experience lower costs since they can get information more easily. The study by Sirak et al. (2007) on the analysis of cattle marketing participation in South Africa shows that distance to the preferred market channel is negatively related with the probability of selling. Also Shilpi et al. (2007) found that the likelihood of sales at the market increases significantly (positive) with an improvement with market facilities and a decrease in travel time from the village to the market. 38? ? Market price (MRP): This variable is measured in Birr per quintal. Tomek and Robinson (1985) argued that the product price has direct relations with marketable supply and hence it was expected to affect the household marketable supply of rice positively. But they argued that in the short run prices could not stimulate market supply due to the biological nature and time lag requirement of production. Lagged market price (LMP): This is also the variable measured in Birr per quintal and is expected to affect the marketable supply of rice positively. Because, lagged prices can stimulate production and thus marketable supply of rice for the next year. According to Myint (2003) explains that if prices in one year are bad, farmers will often respond by planting less in the next year. This will lead to lower production and higher prices, so encouraging more plantings in the following year and a consequent fall in prices. This cyclical nature of production and prices is quite common. Successful farmers are sometimes those who do the opposite to what is being done by other farmers. Boughton (2007) also discussed that local maize prices had a strong positive and highly significant effect on the probability of market participation as a seller on his study on maize market participation in Mozambique. Quantity produced (TQP): It is a continuous variable. A marginal increase in rice production has obvious and significant effect in volume of rice supply. The volume of production of rice is expected to have positive relation to market participation and marketable surplus. Study by Chauhan and Singh (2002) also showed that, marketed surplus of paddy is positively related to the volume of production as well as with area under crop. Total land size (TLS): The total size of farm land owned by a farmer is among the variables that could influence both participation and supply. If a farmer owns more land, the probability of allocating land for rice crops would increase. It is a continuous variable expected to influence participation and supply decision in similar direction. The study by Boughton (2007), the coefficients on available land area are highly significant for both the linear (positive) and quadratic (negative) terms, indicating a diminishing marginal effect on maize market participation as land area increases over the whole range of the data. On another study 39? ? also land holding has an indirect positive effect on market participation, though it is positive effect on farm output (Indo kan et al., 2006). Number of oxen owned (OXN): Being a power for plowing, rice supply would increase as farmers increased their number of oxen ownership. The expected influence is positive on supply. It is a continuous variable Labor (FL): It is a continuous variable, measured in man equivalent. This variable had a positive influence on market supply. As farmers own more number of labor power the interest to farm more size of land would increase. Access to market information (MINF): This is a dummy variable taking a value of 1 if the farmer had access to market information and 0 otherwise. It is hypothesized to affect rice marketable supply of the farm households positively. Because, producers that have access to market information are likely to supply more rice to the market. Obtaining information through extension contacts increased the chance of household selling rice. Study by Makhura (2001) implies that getting information through extension contacts has a considerable marginal effect on increasing the probability of selling horticultural crops. Credit Access (CREDIT): This is a dummy variable, which credit indicates taken for rice production. Access to credit would enhance the financial capacity of the farmer to purchase the necessary inputs. Therefore, it is hypothesized that access to credit would have positive influence on market participation and volume of sale. Study by Black and Knutson, (1985) in Texas survey showed credit users showing better production and market participation among cooperative members. Access to credit would enhance the financial capacity of the farmer to purchase the bird. Therefore, it is hypothesized that access to credit would have positive influence on level of production and sales. Non-farm income (NFINC): It is a continuous variable that obtained from non-farming activities by the household head. A study by Iddo et al. (2006) confirmed that non-farm 40? ? income has affected the decision of farmers to sell their farm out put (market participation) negatively in the study of rural Georgia. Total livestock unit (TLU): This is a continuous variable defined in terms of tropical livestock unit (TLU). Farmer could sell more rice when he/she produces more. On the other hand, when the household has less production; it must either borrow money or sell his livestock to meet household needs. Farmers who have low production of rice need to specialize in livestock production and hence it has an inverse relationship with crop production and marketable surplus. Study by Rehima (2006) on pepper marketing at Alaba and Siltie in SNNPRS of Ethiopia showed that TLU showed a negative sign on quantity of pepper sales. On the other hand, study by Makhura (2001) on maize market participation suggests that an increase in the value of livestock owned leads to an increase in maize sale. Therefore, it is expected to have positive and negative relationship with market participation. 3.4.2. Descriptive analysis In this section descriptive statistics analyses were employed to analyse the S-C-P model for rice market. 3.4.2.1. Analysis of market structure The perfect competition market model is often used in economics as a standard by which structure and conduct of markets can be compared and evaluated. Knowledge regarding structure can give indications about competitiveness. The variables used to explain market structure are the degree of concentration, vertical and horizontal integration, condition of entry in the market and magnitude of product differentiation (Nambiro et al., 2001). a. Concentration Ratio (C): A market concentration ratio is a measure of the percentage share of the market controlled by a specified percentage of firms ranked in order of market share from the largest to the smallest (Karugaia, 1990). 41? ? ? = r i i SC (9) Where: C= Concentration ratio S i = Percentage share of th i th firm r= Number of largest firms for which the ratio is going to be calculated. ? = i i i V V MS (10) buyer by the handledproduct ofAmount V Where i = ibuyer of shareMarket M i S = firms.r by the handledproduct ofamont TotalV i = ? b) Barriers to entry: A barrier to entry is simply any advantage held by existing firms over those firms that might potentially produce in a given market. Potential entry barriers will be investigated based on demand conditions, product differentiation and price elasticity, control over input supplies, legal and institutional factors. 3.4.2.2. Analysis of market conduct Conditions that are believed to express the exploitative relationship between producers and buyers was analyzed based on a) Pricing behavior analysis. Who sets prices? (e,g. one buyer or many buyers , factors considered in price setting (e.g. basic supply and demand conditions or artificially price restraint ?) and b) Buying and selling practices analysis (e.g. source of product, distribution channels used, formal and informal producer and marketing groups), were used for the study (Scot, 1995). 3.4.2.3. Analysis of market performance To analyze the performance of rice markets, margin analysis was used to address the second objective. The cost and price information were used to construct marketing cost and margin. Many studies used market margin than net returns for the analysis to compute profit. Rehima (2006) used marketing margin analysis to calculate profit of pepper marketing and Abay 42? ? (2007), also applied marketing margin analysis for vegetables. The two most common methods are a) Marketing margin: It is calculated as the difference between producers and retail prices. The producers share is the commonly employed ratio calculated mathematically as, the ratio of producer?s price to consumer?s prices. Mathematically, producers share can be expressed as: rr x P MM 1 P P PS ?== (11) Share ProducersPS: Where = rice of price Producers'P X = rice of Price RetailP r = margin marketing TotalMM = 100 Pricebuyer End Priceseller First Pricebuyer End TGMM ? ? = (12) margin marketing gross TotalTGMMWhere, = The producer margin also estimated by introducing the idea of ?farmer?s portion?, or ?producer?s gross margin? (GMMp) which is the portion of the price paid by the consumer that goes to the producer. It is calculated by using the following formula: 100 Pricebuyer End margin gross MarketingPricebuyr End GMMp ? ? = (13) priceconsumer in share producers' TheGMMp , Where = The net marketing margin (NMM) is the percentage of the final price earned by the intermediaries as their net income after their marketing costs are deducted. The percentage of net income that can be classified as pure profit (i.e. return on capital), depends on the extension to such factors as the middlemen?s own (working capital) costs. 43? ? PriceBuyer End Costs Marketingmargin Gross NMM ? = (14) b) Profitability analysis Nuru et al. (2006) also used the profitability analysis of processing crude honey. To estimate the profitability of crude honey at farm gates, local markets of the study areas were considered. Processing equipment and expenses were estimated based on current market price. The net profit of processing of crude honey was calculated by considering all inputs and expenses required to purchase and process the crude honey and also the output. Dejene (2008) studied the profitability of extension package inputs for wheat and barley in Ethiopia. He employed simple calculation of value-cost-ratio (VCR). The unit of analysis is hectare of land. The model takes the usual gross profit formula. Hence, for this study the gross profit and the cost margin analysis were adopted to analyse the profitability of rice production in the study area. i n i i qpPQCVProfit Gross ? ?=?= ( 15) production ofcost TotalC production of ValueV hectareper production TotalQ iinput ofQuantity q iinput of PriceP produce theof PricePWhere, i i = = = = = = The limitation of financial profit analysis is that it does not consider the economic costs and benefits. The financial analysis estimates the profit accruing to the project entity or to participant, where as economic analysis measures the effect of the project on national economy. The major difference lying in the definition of costs and benefits. In financial analysis all expenditures incurred under the project and revenues resulting from it are taken 44? ? into account where as in economic analysis attempts to assess the overall impact of on improving the welfare of the society. Moreover the price measurement is different, shadow price is used for economic analysis and market price is used for financial analysis. It measures simply the accountants cost and profits. Generally, an implicit cost is not considered in the calculation of the financial profit analysis. 45? ? 4. RESULTS AND DISCUSSION This chapter deals with the findings, descriptive statistics and econometric models, on rice marketing in Fogera Woreda especially, on marketing channels, and the marketing agents. It also deals with the analysis of cost and profit of paddy production. It quantifies costs and margins for key traders, identifies factors affecting rice market participation and volume of sales in the study area. This chapter, in addition, examines the support services (extension services, input supply, credit, and marketing services) in rice production and marketing. It also identifies major constraints and opportunities in production and marketing of rice. 4 .1. Household and Farm characteristics 4.1.1. Household characteristics This section discusses the socio-economic characteristics of the sample households in the study area. These socio economic variables include sex, age, religion, marital status, education level, family size and labor. 4.1.1.1. Family size and age of the household In the study area, the average family size was 5.72 with a minimum of 2 and maximum of 13. In upland rice production system the average family size was 5.74, it was also similar for low land rice production system. The t-test shows that there is no significant difference in family size between the two rice production systems at 5% level of significant. Table 5. Age, family labour and family size of households Characteristics N Mean St. Dev Min Max t-value Age of household head 165 42.69 12.301 22 75 1.197 Family labor (man-equivalent) 165 2.67 0.881 1 6.15 2.295** Family size 165 5.72 1.91 2 13 0.021 Source: Survey data, 2008/9 ** significant at 5% level 46? ? The family labor is the main input for rice production. The study shows that the farmers average family labor force was 2.67 in man-equivalent and 6.15 maximum (Low land rice production system and 1 minimum (in up land system). The mode was 1.8 man-equivalents. The t-test also indicates there was a significant difference in family labor force between up land and low land rice production systems at 5% level of significant. The age of the household is considered a crucial factor, since it determines whether the household benefits from the experience of an older person, or has to base its decisions on the risk-taking attitude of a younger farmer. Based on the Table 5, the age of the respondents ranges from 22 to 75 with the median of 41 and multiple mode of 35 respectively. The youngest head is 22 years old, while the eldest is 75 years of age. The mean age of heads of households are about 42.69 years of age for all kebeles that is 40.43 for Kuhar Michael, 44.34 for Nabega, 43.55 for Kidst Hanna and 42.59 for Diba Sifatira respectively. There is no significant difference in ages of the sampled households between upland and low land rice production system. 4.1.1.2. Sex and education of the household Normally the head of the household is responsible for the co-ordination of the household activities. As such it is pertinent to include some attributes such as sex and education of the head in the specification of market participation decisions. Of the 165 sampled respondents about 99% were male headed. Another attribute of importance is the level of education attained by the heads of the household, who, normally, are the decision-makers. Education also enables the person with ability to do basic communications for business purpose. From all household heads 38.8% were found to be illiterate, 26.1% were able to read and write (adult education and religious school), 33.3% attained primary school education and the rest 1.8% was found to be in secondary school education. These groups are able to interpret market and other information better than those who have less or no education. 47? ? Table 6. Demographic characteristics of sampled farmers Up land production Low land production Total Factors Quhar Michael Diba Sifatira Nabega Kidist Hanna 2 ? /t Sex of household head Male 37 53 44 29 163 Female 1 1 0 0 2 Total 38 54 44 29 165 0.369 Religion of households Orthodox Christian 38 54 44 29 165 Age of households ?18 - - - - - 19-59 33 50 36 27 146 ?60 5 4 8 2 19 Total 38 54 44 29 165 1.197 Family size of house holds ? 5 24 13 25 9 71 5-10 14 40 19 20 93 ?10 0 1 0 0 1 Total 38 54 44 29 165 0.021 Education level of households Illiterate 19 15 19 11 64 Read and write 11 13 15 4 43 primary school 8 24 10 13 55 secondary school 0 2 0 1 3 total 38 54 44 29 165 0.205 Marital status Married 35 53 43 29 160 Divorced 2 0 0 0 2 Windowed 1 1 1 0 3 Total 38 54 44 29 165 Family Labor(man equivalent) 1-2.0 17 12 6 11 46 2.1-4.0 19 26 19 42 106 4.1-6.0 2 4 4 1 11 ?6.0 0 1 0 0 1 Total 38 43 29 54 164 2.295** Note: ***, ** and * show the values statistically significant at 1%, 5% and 10% probability level respectively Source: Survey data, 2008/9 48? ? 4.1.2. Farm characteristics 4.1.2.1. Land holding According to CSA (2003), farm holdings is referred to all land or livestock holdings which are mainly used for both crop and livestock production. Depending on the type of activities, and agricultural holders engaged with farm holding has been categorized into three groups. These are crop only, livestock only and both crop and livestock. In Amhara Region, most of the agricultural holders (30.5%) had a total size of land hold that ranges from 1 to 2 hectare. Similarly, 13.6% of agricultural holders that are involved in crop production has under 0.1 hectare of agricultural holdings. On this study, the average land holding for households was 1.21 ha. About 52% rice farmers has land that ranges between 1 to 2 hectare and 6.1% of the farm households have an area above 2 hectare of land. In the study area farmers try to get access to additional land for production of rice through renting. There is a significant difference in land holding, private pasture land and cultivated land among the four sampled kebles at 1% and 10% level of significance. Table 7. Land holding of household head in hectare land use N average Std. Deviation F-value Land holding 164 1.21 0.6 4.338*** Cultivated land 162 0.93 0.43 2.567* Private pasture land 77 0.12 0.18 9.895*** Fallow land 2 0.004 0.04 - Home stead 92 0.11 0.12 0.959 Source: Survey data ,2008/9 49? ? 4.1.2.2. Crop production A total of 165 household were interviewed from 4 administrative Kebeles and all of them were producers of paddy /rice during main cropping season. The major reasons for growing rice are home consumption and sale. Rice straw also is used for animal feed and roof thatching. In terms of land utilization, Table 8 shows that, on average, 0.60 hectares of land per household is allocated to rice as compared to 0.36 and 0.31 hectares for teff and Maize, respectively. Table 8. Cultivated area and yield of paddy/rice crop per hectare, 2007/8 Types of crops Cultivated area in(ha) Productivity(q/ha) N Mean Std. Deviation N Mean Std. Deviation Teff 89 0.36 0.25 86 7.14 5 Maize 144 0.31 0.19 142 19.96 13.87 Wheat 14 0.21 0.13 14 13.67 6.18 Barley 9 0.22 0.13 9 12.36 5.74 Chick pea 102 0.29 0.23 101 12.96 7.51 Lentil 15 0.19 0.1 15 7.56 3.93 F. millet 92 0.31 0.2 91 14.28 8.23 Niger seed 5 0.26 0.16 5 8 4.9 Field pea 27 0.46 0.33 25 7.93 7.7 Grass pea 45 0.36 0.22 42 9.86 7.35 Tomato 16 0.15 0.08 13 62.22 24.24 Pepper 36 0.11 0.07 36 35.71 33.98 Onion 23 0.23 0.2 23 74.33 83.28 Potato 3 0.07 0.05 2 72 22.63 Em.wheat 24 0.21 0.13 24 21.62 17.98 Spice 7 0.24 0.24 7 10.19 8.47 Rice total 164 0.6 0.33 164 32.73 19.76 Own land 154 0.48 0.25 154 36.06 20.98 Rented-in 60 0.38 0.26 59 22.93 14.74 Rented-out 6 0.3 0.17 6 14.17 9.81 Source: Survey result, 2008/9 In addition to rice, sample farmers cultivate other crops like, teff, maize, finger-millets chickpea, grass pea and vegetables during the off rice season. There was no any double cropping of rice by using irrigation (or supplement irrigation). 50? ? The mean production of milled rice is 13 quintal per household. Out of this 8.6 quintal is used for consumption purpose and 1.11 quintal is used for seed and the remaining 2.9 quintal of rice was marketed. As described in Table 9. The average production of rice per hectare was higher in Kidist Hanna than other kebeles and almost the same in other three kebeles (12 qt). The one way ANOVA analysis shows that there is a significant difference in rice production among four kebeles at 5% level of significant (F-value is 3.564 and P<0.016). Table 9. Production of rice by sample households in qt/ha, 2007/8 Source: Survey result, 2008/9 ** significant at 5% level According to Tesfaye et al. (2005), rice can locally be prepared and consumed in a variety of traditional ways. In terms of importance and priority farmer utilize rice by making the following food types. Pancake or Engera which is prepared independently on its own or by mixing with teff or finger millet depending on the wealth status of the farmer, Dabo or bread which is prepared by mixing it with other cereal such as wheat and maize on different proportions. Kinche (boiled split rice mixed with either oil or butter) meals and local beer is also prepared from rice for home mainly for home consumption purposes. Utilizing rice by mixing up with other crops (mixing rice with crops like teff and finger millets) is common for urban consumers and hotels. In Woreta (capital of the woreda) town, one farmers? multipurpose cooperative association, was established and is giving service currently. The main function is to collect rice from cooperatives member producers and sell it to different consumers (including other Name of PAs N Mean Minimum Maximum Sum % of Total Sum F-value Kuhar Michael 38 11.725 1.4 31.5 445.6 21.30% 3.564** Diba Giorgies 54 11.848 0.7 42 639.8 30.60% Nabega 44 11.558 2.8 40.6 508.6 24.30% Kidist Hanna 29 17.114 1.4 28.7 496.3 23.70% Total 165 12.668 0.7 42 2090 100.00% 51? ? cooperatives). The advantage is price stabilization mechanism for grain producers of farmers who are members of the cooperative association. The cooperative has different milling machines used to prepare different forms of rice products. For instance, it can prepared rice used for hotels, for consumers, and for enjera. 4.1.2.3. Livestock production Livestock production is an integral component of the farming system in the study area and contributes very much to rice production in particular and to crop production in general. Important animals kept by the sample farmers are cattle, sheep, goats, mule, horses, donkey and poultry (Table 10). Oxen are the main source of farm power for plowing, short haulage, harrowing, and threshing. About 51% of the respondents owned one pair of oxen, 29.9% owned one, 11.5% owned three, 5.1% owned four, and the rest percent owned 5-6 respectively. The sample respondents have, on average, a pair of oxen (1.91) with standard deviation of 1.04. There is significant difference in number of yearling, sheep, oxen, Goats and in monitory value of livestock among 4 kebeles. Table 10. Number of livestock owned by sample households, 2007/8 Types N Mean Std. Dev F-value Cow 165 1.62 1.299 0.854 Oxen 165 1.91 1.041 3.470** Heifer 165 0.97 1.05 0.945 Yearling 165 0.67 0.871 4.238*** Calves 165 0.85 0.945 1.729 Bulls 165 0.07 0.391 2.036 Mature Sheep 165 1.04 2.288 3.615*** Lamb Sheep 165 0.33 0.932 2.274* Mature Goats 165 0.28 1.136 2.477* Kids Goat 165 0.05 0.336 1.224** Mature Donkey 165 0.59 0.634 1.509 Kids Donkey 165 0.19 0.412 0.849 Horses 165 0 0 Mules 165 0.02 0.134 2.006 Total livestock unit (TLU) 165 5.465 3.44433 1.861 Total monitory value (birr) 165 13849 9122.55 2.138* Source: Survey data, 2008/9 ***, ** and * show the values statistically significant at less than 1%, 5% and 10% respectively 52? ? 4.1.2.4. Ownership and farm implements The implements used in rice cultivation are generally traditional. Light hand-ploughs, drawn by oxen, are most frequently used. About 99 % of the respondents had plowing tools and 7 % farmers owned cart. About 73 % of the households also had grass roofed house and 65% had iron sheet roofed houses respectively. To assess the livestock holding TLU and birr were employed to calculate resource ownership per households. The average livestock owned was about 5.47 tropical livestock unit (TLU). In terms of monetary value it was about 13848 birr. There is significant difference in animal cart ownership only among the sampled kebeles. Table 11.Ownership and farm implements of the sampled farm households Ownership N Minimum Maximum Mean Std. Deviation F-value Grass roofed house 165 1 2 1.27 0.44 1.108 Iron sheet roofed house 165 1 2 1.35 0.47 0.850 Plowing tools (moffer, kenber, maresha ) 165 1 2 1.01 0.07 1.580 Animal cart ownership 165 0 2 1.92 0.30 2.492* Total livestock ownership 165 0.01 23.24 5.45 3.44 1.861 Source: survey data, 2008/9. * show the values statistically significant at 10% 4.1.2.5. Farming experience and Income The average year of farming experience for the rice producer?s households was 22.54 and the non-farming experience was 1.7 years (Table 12). The survey results indicate that farmers from low land rice production had more experience in farming rice. Almost all the households in the study area depend on farming income. The average amount of income earned form farming activities was 12,029 birr per year and from non-farm activities was birr 460.40 per year. Non-farm income can be used to finance marketing activities and also accessing on-farm income has a bearing on market participation. The t-test shows that there is a significant difference in non-farm income (p<0.025) and non-farm experience (p<0.033) between the two rice production systems at 5% level of significant. 53? ? Table 12. Farming experience and farm income of a farmer Rice production system Farming experience Non- farm experience Annual income from farming Annual income from non -farming Upland N 90 91 91 91 Mean 21.8 2.47 12063.7 647.31 Std. Deviation 11.37 5.96 9402.31 1426.01 Low land N 73 73 73 73 Mean 23.52 0.91 11985.8 227.39 Std. Deviation 11.76 3.06 10847.1 942.7 Total N 163 164 164 164 Mean 22.61 1.78 12029 460.4 Std. Deviation 11.54 4.93 10039.2 1248.66 t-value 0.903 2.157** 0.217 2.232** Source: own survey result, 2008/9, ** shows the value statistically significant at less than 5% level. 4.2. Access to Services 4.2.1. Location and infrastructure Location: Agricultural production is affected by the availability and utilization of inputs and service used such as credit, agricultural extension, and market information. Road accessibility and facility of transportation are also needed to market agricultural outputs. In the study area, rice producing farmers travel a maximum of 4 hrs and a minimum of 0.08 hour to reach the nearest market center (woreda capital Woreta). The average distance needed for farmer to travel to the market was about 1.6 hours per trip. The distance to the local extension office (developmental centers) is an important factor since the interaction of the farmers with the extension office is crucial in making information available. The mean distance required to travel to the development (extension office) was about 0.57 hours. So, since the distance to this centre has a bearing on farmers? access to markets, proximity in walking hours will be included in the specification of the model for market participation. The 54? ? analysis of ANOVA indicted that there is a significant difference in distance to travel to the market center at 1% level of significant (p<0.001) but there was no any difference in distance to travel to development centers among the sampled Kebeles. Table 13.Traveling time required to the market center and development center (in hours) PA Distance in hour to Market center Development office Quahar Micheal N 38 38 Mean 1.12 0.46 Std. Deviation 0.56 0.44 Minimum 0.08 0.08 Maximum 2 2 Nabega N 44 44 Mean 1.58 0.43 Std. Deviation 1.25 0.35 Minimum 0.08 0.08 Maximum 4 1.5 Kidist Hanna N 29 29 Mean 1.62 0.38 Std. Deviation 1.25 0.29 Minimum 0.08 0.08 Maximum 3.5 1.5 Diba Sifatira N 54 53 Mean 1.97 0.86 Std. Deviation 0.81 0.87 Minimum 0.25 0.08 Maximum 3 3 Total N 165 164 Mean 1.60 0.57 Std. Deviation 1.03 0.61 Minimum 0.08 0.08 Maximum 4 3 F-Value 5.621*** 0.884 Source: Own survey, 2008/9 Infrastructure: Fogera woreda has about 17 kms asphalt road, 30 kms all weather gravel road, and much dry weather road. In the harvest season, vehicles and carts could travel to the direction they wish. However, about 91% of producers transport rice to local markets by packing animals and 4.2% by head load, they also use animal carts and some few producers 55? ? also use vehicles. The average market transportation cost is about 9.50 Birr per quintal. There is one bank service at Woreta, and there is also credit giving institution, ACSI, with wider service coverage. Mobile telephone worked in all 4 kebeles. All rural Kebeles had a telephone line. Table 14. Means of transportation used by sample households in rice marketing Frequency Percent Head /back loading 6 4.2 Animal carts 5 3.5 Vehicles 2 1.4 Pack animals 130 90.9 Total 143 100 Source: Own survey, 2008/9 4.2.2. Credit availability The survey result indicated that about 62% of the sampled farmers need credit but the majority of them did not take credit both on-cash and in-kind to purchase inputs like, fertilizer (Dap and Urea), seed, chemicals and sprayer. This is because fearing of interest rate and defaulters (to make grouping as means of collateral). There is a high significant difference in getting credit among sampled kebeles. Table 15. Credit availability to the sample farm households Did you take credit? Name of peasant Administration Percent of households with credit access F-Value Kuhar Michael Nabega Kidist Hanna Diba Giorgies Total Yes 4 4 16 24 48 29.1 12.226*** No 34 40 13 30 117 70.9 Total 38 44 29 54 165 100 Source: own survey, 2008/9. ***, shows significant level at 1% level of significance. With regard to credit source out of 48 sampled farmers, 26.1% of the farmer get credit from Amhara Credit and Saving Institute (ACSI), 3% get credit from service cooperatives. 56? ? Table 16. Credit giving institutions Credit giving organizations No. of sample households received Percent (%) ACSI 43 89.6 Cooperatives 5 10.4 Total 48 100 Source: Survey data 2008/9 From a sample of 48 credit users about 96% used the obtained credit to purchase animals either for fattening or plowing purpose or to purchase pump for irrigation of vegetables. About 2.1% used for grain seed purchase and food grain production purpose. Table 17. Credit purpose for households Credit purpose No. of sample households Percent (%) To purchase animals for fattening , plowing or to purchase pump 46 95.8 To purchase grain Seed 1 2.1 To rent-in land for food grain production 1 2.1 Total 48 100 Source: Survey data, 2008/9 4.2.3. Market information and extension service The distribution of market information refers to the availability of relevant market information to farmers, about demand, supply and price of the crops. The survey result indicates that 79.2% of the households had price information before they sale their produce to the nearby market but 20.3% of the interviewed farmers do not have access to any information. a) Supply, demand and price information As indicated in Table 18, out of 133 farmers, 42.9% obtained information about rice supply by using other rice traders from previous market days and through personal observation 57? ? during their market visits. A sample respondent of 139 rice farmers also revealed that 40.3% of them get information about rice market demand from other trader and their personal observation. On the same manner 42.9% of the sampled households obtained price information from another farmers and their personal observation Table 18. Source of information about supply, demand and price, 2007/8 Source of information Information Supply Demand Price Frequency % frequency % frequency % Personal observation 15 11.3 24 17.3 31 22.1 Rice traders 18 13.5 14 10.1 8 5.7 Another Farmer and personal observation 43 32.3 45 32.4 60 42.9 Other rice traders and personal observation 57 42.9 56 40.3 39 27.9 Radio - - - 2 1.4 Total 133 100 139 100 140 140 Source: Survey data 2008/9 b) Quality of source of information With regard to quality of source of information from a total of 129 respondents, 42.6% were indicated that the information quality was adequate, 21.7% also responded both reliable and adequate, and 20.9% responded only reliable and only 2.3% was recorded as quality of information is timely. Table 19. Quality of source of information about supply and demand, 2007/8 Quality of information Supply Demand Frequency % frequency % Reliable 27 16.4 35 26.1 Adequate 9 5.5 10 7.5 Timely 3 1.8 3 2.2 Reliable* and Adequate** 28 17 25 18.7 Reliable and Timely 3 1.8 4 3 Adequate and Timely 4 2.4 4 3 Reliable ,adequate and Timely 55 33.3 53 39.6 Total 129 78.2 134 100 Source: Survey data, 2008/9, *means accuracy of the information &** means the amount and availability of enough information. 58? ? C) Extension service The average number of contacts farmers have with extension officers is about four times per month. The distance to the extension office affects the cost of searching for information. On average a household takes 1.57 hour per trip to reach the agricultural development offices. The study shows that 65.7% of respondents had a weekly contact with extension agents and 19.6% had contact once in two weeks. About 18.2% of the sampled respondents get advice on production and animal feeding, 16% on production only, and 15.4% got advice on production of crops, marketing, credit and health aspect. There is a significance difference in extension contact among sampled kebeles at 1 % level of significant (F=5.018 and p<0.002). Table 20. Frequency of extension contact Extension contact frequency Frequency Percent Weekly 94 65.7 Once in two weeks 28 19.6 Monthly 12 8.4 Twice in a year 2 1.4 Once in a year 2 1.4 Any time when I ask them 5 3.5 Total 143 100 Source: survey data, 2008/9 4.2.4. Agricultural input use 4.2.4.1. Chemical fertilizer and seed It is evident that chemical fertilizer could boost both production and productivity. Despite this fact, rice producer at Fogera Woreda used very small amount of fertilizer on their rice field. According to IPMS (2005), the reason is that due to flooding and fertile alluvial soil (washed soil from highland area). As shown in Table 21, only 3% of the sampled households used urea, 1.2% use Diamonium phosphate (DAP) and 4.9% used organic fertilizer for rice production. 59? ? In general, the farmers use two types of seed variety known as X-Jigna (local) and Gumara (IAC-164.) the improved one. The mean of the seed rate is 258.61 kg per ha. A bout 96% of the sampled household used X-Jigina variety (mostly popularized by farmers) and 56 % in the upland and 37% low land rice production system used this variety. The survey result also showed that about 25% of the sampled households used Gumara variety (the improved one). However, since it is red in color it is less demanded and used for consumption purpose compare to the white seed X-Jigina variety which has high market demanded. 4.2.4.2 Herbicides and insecticides In the study area farmers used little type of herbicides, namely 2-4-D and Malatainne for the rice cultivation. The survey result indicates that out of the sampled households 3% of them used insecticide, and 16.5% used herbicides for rice production. The 2 ? show that there is a highly significant difference in utilization of insecticides, herbicides at 1% level of significant in up land and lowland rice production system. 60? ? Table 21. Input utilization of farmer for rice production Rice production farming system Upland rice Lowland rice Inputs Kuhar Michael Diba Sifatira Nabega Kidist Hanna Total 2 ? urea yes 1 3 0 1 5(3%) * 1.222 No 37 51 43 28 159 Total 38 54 43 29 164 DAP yes 0 2 0 0 2(1.2%) * 1.602 No 38 52 43 29 162 Total 38 54 43 29 164 Organic fertilizer 3.242* yes 1 1 4 2 8(4.9) * No 37 53 39 27 156 Total 38 54 43 29 164 Insecticides 6.519*** yes 0 0 2 3 5 (3%) * No 38 54 41 26 159 Total 38 54 43 29 164 Herbicides 16.801*** yes 6 0 17 4 27 (16.59) * No 32 54 26 25 137 Total 38 54 43 29 164 X-Jigina variety 9.244*** yes 38 54 36 29 157(95.2%) * No 0 0 7 0 7 Total 38 54 43 29 164 Gumara Variety 50.205*** yes 1 3 18 19 41(25%) * No 37 51 25 10 123 Total 38 54 43 29 164 Note: 1. Chi-square shows between the two rice production systems 2. ***, **, and * are significant levels at 1%, 5%, 10% respectively 3. Figures in parentheses are percentages. Source: survey result, 2008/9 61? ? 4.2.4.3. Labour and machinery use Labour demand for rice farming is more than the other crops (Tesfaye et al., 2005). The labour is employed in rice cultivation from soil preparation to harvest. The family labour force (owned labour) consists of the highst percent in rice cultivation. About 44% of the labour is used from owned and very small part 7.9% obtained from hired and shared labour. The analysis of variance shows that there is significant difference in sources of labour among the sampled kebeles (F- value is 3.076 at p<0.005). With regard to farming implements the survey result shows that, 99.4% of the farmers had plowing tools for rice cultivation and 92.7% of the farmer also had two carts while the percentage varies among the Administrative Kebeles. Table 22. Source of labour employed in rice cultivation 2007/8 Source Frequency Percent Owned labour 72 43.6 Owned + hired labour 59 35.8 Owned + shared labour 13 7.9 Hired +shared labour 1 0.6 Owned+hired+shared 20 12.1 Total 165 100 Source: own survey result, 2008/9 4.2.4.4. Storage facilities According to De Lucia and Assennato (1994), post harvest loss is defined as a measurable quantitative and qualitative loss in a given product .The loss can occur at any point during harvest, threshing, drying, storage or transport. An estimated 10-37 % of total rice production is lost due to post harvest factors (Saunders, 1979). During harvest, depending on the type of machinery or manpower used, small amounts of the grain will be left in the field. Similarly, losses may occur during the drying process, which in developing countries commonly takes place on the road side. Further losses are incurred during the storage process due to molds, 62? ? insects and rodents. Estimates from Sub-Saharan Africa have shown rodents can consume or contaminate up to 20% of a stored harvest (FAO, 1994). Storage services helps for smooth and continuous flow of products to the market and create time utility. The survey result shows that all sampled farmer?s store rice in local granaries called Gottera or Gota which is made of bamboo tree plastered with mud and 3.6% of them used sack. The duration ranges from 3-24 months. The average month identified was 9.24 month. There is statistically significant difference at 1% level among the sampled kebeles in storage duration (F-value is 0.128 at p<5.012) The purpose of storage rice is 68.5% for sale and consumption and 31.5% for consumption purpose only. However, the motive behind storage was 60% of sample households respond that it is for saving and expecting higher future price. However, farmers reported that there was weight loss in rice during storage (change in quantity and quality). Table 23. Average storage duration in months to store paddy Name of peasant Administration N Mean Std. Deviation Kuhar Michael 38 9.18 3.56 Nabega 43 8.12 2.91 Kidist Hanna 29 11.06 3.79 Diba Sifatira 51 9.19 2.68 Total 161 9.24 3.29 Source: Survey result, 2008/9 4.2.5. Rice marketing of farmers Out of the total 165 sampled farmers 75.8% of the households sold their produce to the market and 24.2% of the respondents did not sell to the market. It is believed that these farmers consume what they produce and stored their produce for seed use. Quantity of rice marketed by sample households is presented in Table 24. Total supply of rice that is marketed per household in 2007/8 was on average 479.6 quintal. 63? ? Table 24. Quantity of rice sales by kebeles in quintal (marketed surplus), 2007/8 Name of Kebele Administration N Sum % of Total Sum Kuhar Michael 38 76.3 15.90% Nabega 44 123.4 25.70% Kidist Hanna 29 123.2 25.70% Diba Sifatira 54 156.7 32.70% Total 165 479.6 100.00% Source: Survey result, 2008/9 Among the two Rice production system, in upland rice production system 98.57 % of the rice sold went to Woreta market and in lowland production system 70.4% of the rice marketed to Woreta and 30% of the produced quantity went to local or rural market points. Table 25. Use pattern of rice produce at a household level Descriptive measures Rice produced (qt) Consumption ( qt) Seed (qt) Rice sold (qt) N 165 164 164 165 Mean 12.67 8.66 1.12 2.91 Std. Error 0.64 0.50 0.06 0.30 Minimum 0.70 0.00 0.00 0.00 Maximum 42.00 37.80 5.60 21.00 Sum 2090.20 1420.20 183.40 479.60 Proportions (%) 100.00 0.68 0.09 0.23 t-value 0.134 .638 .096* .036** Source: Survey result, 2008/9 64? ? 4.3. Profit Analysis of Rice production 4.3.1. Unit and conversion factors After harvesting, rough rice or paddy rice is dried, either mechanically or by open-air. Dried rice is then milled to remove inedible hull. Hulled rice is also called "brown" rice and consists of an average weight of 6-7% bran, 90% endosperm and 2-3 % embryo (Chen et al., 1998). Further milling removing the bran layer yields white rice. On average, paddy rice produces 25% hulls, 10% bran, and 65% white rice (Saunders, 1979). There are several degrees of milling which can take place, depending on consumer preferences and desired degree of whiteness or opacity. Milled rice is referred to as polished or whitened and there are various degrees or fractions of polishing. White rice implies 8-10% bran removal. Before proceeding to the calculation of profit and margins, the underlying assumptions must be explicit. In the present calculation we will try to estimate and fix the conversation rate that is used to convert from paddy to milled rice. For example the commonly conversation factor of paddy in Philippines is 0.65 but it applies to dry paddy also, however, most paddy hauled to mills is wet, for which the conversion factor of 0.58 were assumed. In this study based on farmers respond 0.70 was taken as the conversion factor for paddy yield. Hence the following points were considered in the calculation of profit and margin. 1. The conversion factor of paddy yield is 0.70. That is 0.30 is Husk yield. Husk yield is 30 percent of the grain yield. 2. Average selling price of a kilogram of husk is 25 cents per kg. 3. A straw yield is measured in ?shekim?, i.e. the amount of straw which is tied up with a rope having two meter circumference from one ?timad? (=0.25ha). About 10-30 number of ?shekim? (head /backload) of straw will be obtained 4. Since each farmer has plots with different soil fertility, flooding status, the opportunity cost of each farm will vary so the opportunity cost given by each farmer was considered as it is. 65? ? 5. Transportation cost by donkeys? from farm to farmer?s house was calculated based on the amount of quintal to be transported per day. 5.1. If it is from1-10 quintal, it requires 1 donkey at a price of 10-15 birr/day 5.2. If it is from 10-15 quintal, it requires 2 donkeys at a price of 10-15 birr/day. 5.3 If it is above 15 quintal, it requires 3 donkeys at a price of 10-15 birr/day 6. Labour cost is estimated based on the price or wage of labour in each locality. 7. The Price of a pair of oxen per day is estimated based on the rental value in the each locality. 8. A 10% interest rate per month is considered for the interest rate calculation which is available for loans or credits from Amhara Credit and Saving Institute (ACSI). 4.3.2. Gross income of paddy production The mean paddy rice production was 42.19 quintal per hectare with a standard deviation of 19.79. if it is converted to milled rice, the mean production (0.70% of paddy produced) was 30 quintal per ha. Rice producers generate income from sales of paddy alone or sales of polished (milled) rice. It has two by-products. These are straw yield and husk yield. Straw yield used for construction of house and husk yield (cover rice) also used for cattle feeding and fattening purpose for farmers. Husk yield is also used for making chip wood. Usually farmers do not use the husk yield. It will be left for millers during milling of their paddy. In this study, straw yield is also considered to calculate the gross income of farmers. The gross income of paddy production was 17549.21 birr per hectare and the standard deviation was 9741.43 (Table 26). 66? ? Table 26. Profit and Cost of production of rice per hectare Items Average Stedv 1. Revenue Paddy yield (qt/ha) 42.19 19.79 Price of paddy(birr/qt) 387.63 79.46 Straw yield (shekim/ha) 88.05 53.70 Price of straw (birr/shekim) 13.76 5.46 Value of paddy /ha (1) 16930.56 10021.58 Value of straw/ha (4) 1126.08 590.24 Total revenue (1+4) 17549.21 9741.43 2. Cost A. Opportunity cost of land (birr/ha) 4937.0 3009.34 B. labour cost Labor cost for plowing (birr/ha) 333.05 102.58 Frequency of plowing 4.15 1.02 Person required to plow (person day/ha) 16.73 4.20 Labor wage to plow (wage/person) 20.24 5.07 Labor cost for weeding(birr/ha) 2939.41 2171.96 Frequency weeding 2.76 0.62 Person required to weed (person day/ha) 144.19 93.86 Labor wage to weed (wage/person) 20.56 7.44 Labor cost for harvesting (birr/ha) 438.56 177.03 Person required to harvest (person day/ha) 22.47 8.33 Labor wage to harvest (wage/person day) 19.98 6.48 Labor cost for trashing and winnowing (birr/ha) 365.32 140.08 Frequency trashing and winnowing 1.55 0.51 Person required to trash &winnowing (person day/ha) 18.58 7.38 Labor wage for trashing and winnowing (wage/person day) 20.50 6.94 Total labour cost 4049.24 2263.06 C. Animal power cost Animal power cost for plowing (birr/ha) 903.93 408.47 Oxen required to plow hectare (oxen day/ha) 16.75 4.19 Rental rate of pair oxen (price/oxen day) 52.69 20.22 Animal power cost for trashing (birr/ha) 447.30 157.54 Rental rate of oxen required to trash hectare (ox day/ha) 20.41 3.33 Price of one ox for trashing (price/ox day) 22.54 8.14 Animal power cost for transport to home (birr/ha) 89.28 48.63 Total animal power cost 1424.19 480.68 D. Material input cost Amount of seed (kg/ha) 248.59 105.94 Seed cost (birr/ha) 1041.9 469.89 Amount of herbicides (litter/ha) 4.10 1.26 Herbicide cost (birr/ha) 242.21 51.28 Manure cost (birr/ha) 938.40 918.30 67? ? Table 26 (continued) Total input cost 1114.22 521.59 2.5 Other cost Land rent (birr/ha) 25.00 0.00 Interest rate (birr/ha) 1004.65 759.60 Total other cost 213.75 510.38 Total cost /2.1+2.2+2.3+2.4+2.5/ 11688.23 4010.39 Profit/1-2/ 5006.48 10040.62 Source: Own survey 2008/9. 4.3.3. Cost of production of paddy Table 27 gives expenditure per hectare on various inputs used in the production of rice. The Table reveals that the total cost per hectare was 11688.23 Birr on samples households. Opportunity cost of land (rental value of land), was the item taking maximum share in total cost (40.23%) followed by labour cost (34.65%) and animal power cost (13.11%). Material input cost like manure, herbicides, seed (10.26%) and value of other costs like value land rent/tax and interest in capital (1.75%) consists of the minimum share of production cost. Table 27. Average cost per hectare of rice production Type of costs Cost/birr % share Opportunity cost (land rent) 4937.00 40.23 Labor cost 4049.24 34.65 Animal power cost 1424.19 13.11 Input cost 1114.22 10.26 Other costs 213.75 1.75 Source: Own computation from survey, 2008/9 4.3.3.1. Labor cost Rice crop is a labor intensive crop, therefore, weeding labor ranked first. Weed is a major problem. About 67% of the cost expenditure goes for weeding purpose. Besides, harvesting, threshing and winnowing costs rank second and third in the cost component for rice production. 68? ? Table 28. Average labor cost per hectare of rice production Activities Cost/birr % share Plowing 333.05 9.81 Weeding 2939.41 67.39 Harvesting 438.56 12.58 Threshing and winnowing 365.32 10.63 Source: Own computation from survey, 2008/9 4.3.3.2. Animal power cost Similarly, the share of animal power cost used was highest for plowing and it is about birr 903. It ranked 60% of the total animal power cost available. Table 29. Average animal power cost per hectare of rice production Activities Cost/birr % share Plowing 903.93 60.39 Trashing 447.3 32.86 Transporting 89.28 6.75 Source: Own computation from survey, 2008/9 4.3.3.3. Material input cost Percentage share of seed to the total input cost was about 61.47% and it was indicated that the total input utilization from the total cost of production of paddy is very low (10.6 %). Farmers do not use inputs, even fertilizer, because their land is fertile (alluvial soil) and there is flooding problem. Table 30. Agricultural input cost per hectare of rice production for household Inputs used cost/birr % share Seed 1041.90 61.47 Herbicides 242.21 31.87 Manure 938.22 6.66 Source: Own computation from survey, 2008/9 69? ? 4.3.3.4. Other costs (land tax and interest rate) Land rent payment for farmer is calculated based on the available standard given by the bureau of finance. Its payment is based on the amount of hectare a farmer owned (appendix-1) Similarly the interest rate (cost) of credit users of the sampled farmers was about birr 1004 per hectare. Farmers are obtained credit from Amhara credit and saving institute (ACSI) and the interest rate was about 10% per month. Table 31. Cost of land rent (tax) and interest rate per hectare of rice production. Items Cost/birr Land rent/ha 25 Interest rate /ha 1004.65 Source: Own computation survey, 2008/9 4.3.4. Net income / profit The cost benefit production of paddy per hectare bases shows that production of paddy was profitable. The average net income for production of paddy per hectare obtained was 5006.48 birr with a standard deviation of 9899.71. Table 32. indicates that there is a significant difference between four kebeles in terms of gross income and profit at 1% significant levels. But there is no significant difference in terms of cost of production of paddy (Appendix Table 5). The least significance difference or mean difference (LSD) shows that Kuhar Micheal administrative kebele has a significant difference from other three kebeles? (Nabega, Kidist Hanna and Diba Sifatira.) in terms of gross income, cost and profit. 70? ? Table 32. Gross income, cost and profit of paddy production per hectare by kebele PAS Gross income Total cost Profit Quahar Micheal N 37 38 38 Mean 12513.46 11648.26 535.89 Std. Deviation 5914.36 3796.09 7102.64 Nabega N 43 44 44 Mean 17390.23 11175.91 5819.09 Std. Deviation 9981.52 4004.05 10123.80 Kidist Hanna N 29 29 29 Mean 20131.31 11710.77 8420.54 Std. Deviation 10403.46 3877.62 11083.35 Diba Sifatira N 54 54 54 Mean 17778.52 12121.68 5656.83 Std. Deviation 8982.38 4283.12 10343.97 Total N 163 165 165 Mean 16899.55 11688.23 5006.48 Std. Deviation 9235.34 4010.39 10040.62 F-value 4.434*** 0.447 0.009*** Source: Own computation survey, 2008/9 4.4. Analysis of Econometric Results 4.4.1. Heckman two step results In this study, those factors that influence the decision to participate as well as the volume of rice supplied to market would be determined. About 15 variables were hypothesized to determine household level decision to participate in rice market and the volume of marketed surplus. The Probit and Heckman selection model results are depicted in Table 34, 35 and 36. 4.4.1.1. Market participation determinants Heckman two step estimates was analyzed using LIMDEP software. Both continuous and discrete explanatory variables were checked for the existence of multicollinearity. Variance Inflation Factor (VIF) was computed for continuous variables and contingency coefficients for dummy variables to see the existence of multicollinearity among variables. It was found that there is no problem of multicollinearity (Appendix Table 6 and 7). Moreover, explanatory 71? ? variables like market information access, land holding quantity produced were tested and only market information access were found to be endogenous variable. The problem of endogeneity occurs when an explanatory variable is correlated to the error term in the population data generating process, which causes, the ordinary least squares estimators of the relevant model parameters to be biased and inconsistent. Consequently, taking these variables their actual value can introduce endogeneity problem. The source of endogeneity could be omitted variables, measurement error and simultaneity (Maddala, 2001). This problem can be overcome by using two stages least square (2SLS) method. The method involves two successive applications. The first stage is made by regressing the suspected endogenous variables over the pre-determined or pure exogenous variables to get their predicted values. Then the predicted values of the endogenous variables in the first stage are used to estimate the supply equation. The Heckman model was estimated by using a two-step procedure. In the first step the Probit model was estimated to identify factors affecting decision to participate. In the second step the OLS adjusted for selectivity bias (heckit) model was estimated to identify the significant factors of level of participation or volume sold. The model is specified as: Pr (MPD) = f (AGE, SEX, EDU, FS, FL, EXC, MRD, TLS, TQP, OXN, MINF, CREDIT, NFINC, LMP, TLU) The Probit model estimation indicates that 4 variables were found to be the significant factors affecting the household market participation decision (Table 34). These variables are quantity of paddy produced, market information access, extension contact frequency and total Livestock value (TLU) respectively. Four of the variables had coefficients significantly different from zero. These significant variables increased the chance of household selling of rice to the market positively. More over all the significant variables had the expected signs. Market information access significantly affect the probability of selling at 5% (P<0.049) level of significant. Those farmers with better market information are in a better position to supply their surplus production to the market. Goetz (1992), in his study of household food marketing behavior found that better information significantly raised the probability of market 72? ? Table 33. Description of dependant and independent variables used in econometrics models (the Heckman and Tobit models) Variables Description Expected sign Type of variable MS Total quantity (volume) of rice supplied to the market in quintal Continuous AGE Age of households head in years +/- Continuous FS Family size in number +/- Continuous MKD Access to market distance (Hr/trip) - Continuous TLS Total land holding of household head in ha. + Continuous TQP Total quantity of paddy produced (qt/ha) + Continuous NFINC Annual income obtained from non- farming activities in Birr - Continuous LMP Lagged market price of wet paddy in price /qt + Continuous OXN Number of oxen owned in number + Continuous TLU Total livestock of households in TLU - /+ Continuous FL Family labor of household head in man-equivalent + Continuous MPD Market participation decision Dummy EDU Education level + Dummy SEX Sex of household head +/- Dummy EXC Extension contact frequency + Dummy CREDIT Credit access for farm households + Dummy MINF Market information access + Dummy 73? ? participation for potential selling households. Also quantity of rice produced has highly affected market participation positively at 1% significant level (p<0.000). This shows that the higher the output, the higher is the farmer willing to participate in the market. Study by Marcel et al. (2005), on coffee producers indicate that selling to the market is more likely when the quantity sold is large and the market is closed by. It is also found that extension contact with extension agents is positively and significantly influence to the probability of selling rice at 5% (P<0.022) level of significant. This suggests that access to extension service improved market participation and farmers could be aware of the various aspects of the production and selling of rice. The study by Abay (2005) on vegetable marketing in Fogera woreda of South Gondar Zone of ANRS shows that extension contact with farmers has positive influence in the onion market participation decision. Similarly, another variable which affect market participation is the total livestock value (TLU). It is significant at 10%. This indicates that as livestock value increase the income of farmers also increase, since the area is wet land (bordered by Lake Tana ), both crop and livestock production are integrated activities and are connected each other. Hence, owning of more of livestock helps to increase to purchase agricultural inputs for production and this indirectly increase the production and market participation of rice. Study by Makhura (2001) on maize market participation suggests that an increase in the value of livestock owned leads to an increase in maize sale. 74? ? Table 34. Factors influencing the decision to sell rice (Probit results) VARIABLES COEFF. (STD.ERR.) T-RATIO MARGIONAL EFFECT CONSTANT 1.044 (1.746) 0.598 .23255192 AGE -0.009 (0.014) -0.656 -.00207295 SEX 0.934 (1.090) 0.857 .29906007 EDU -0.097 (0.310) -0.314 -.02199104 FS -0.097 (0.094) -1.036 -.02167105 FL -0.100 (0.182) -0.547 -.02219815 EXC 1.206** (0.526) 2.291 .37863589 MKD -0.128 (0.137) -0.939 -.02859262 TLS 0.163 (0.284) 0.573 .03623407 TOP 0.070*** (0.017) 4.010 .01560285 OXN -0.257 (0.185) -1.390 -.05722679 CREDIT 0.079 (0.302) 0.262 .01730087 NFINC -0.237 (0.182) -1.307 -.05285606 LMP 0.000 (0.001) 0.893 .00010750 TLU 0.119* (0.069) 1.735 .02650310 MINF B 1.108** (0.563) 1.967 .15778388 Number of observations = 165 Prob [Chi Sq > value] = 0.3414741E-03 Log likelihood function = -69.82410 Prediction Success = 80.606% Restricted log likelihood = -90.23058 Chi squared = 40.81295 B = Predicted MINF (endogenous variable) Note: ***, ** and * show the values statistically significant at 1%, 5% and 10% probability level respectively 75? ? 4.4.1.2. Market supply determinant /volume/ The model seeks to identify factors that influence the level of rice sales or volume marketed. The model is specified as MS = f (SEX, EDU, FS, MAD, TQP, MINF, CREDIT, NFINC, LMP, LAMDA) This means that the quantity supply or sales depends on the set of factors indicated. The second stage of the selectivity model (heckit or OLS accounting for bias) is estimated to determine factors influencing the level of rice sales. Table 35 presents the results of the determinants regarding the quantity of (level of) sales. For the second-stage OLS results, the inverse mills ratio (lambda) for the level of rice sales was significant, implying that selection bias would have been resulted if the level of sales in rice had been estimated without taking into account the decision to participate. That is selection effects become important, the IMR is significant at the 5 percent level (P<0.056). Two of the significant variables were positively associated with the level of rice sales meaning that the factors were important only among those who were selling rice to the market. Quantity produced is significant at 1% (p<0.000) and Education level at 10% (p<0.065) level of significant. A study by Wolday (1994) on output of food grains (wheat, teff and maize) and Rehima (2007) on pepper market also found that quantity produced has positive effect on quantity supplied to the market. Study by Chauhan and Singh (2002) showed that, marketed surplus of paddy is positively related to the volume of production as well as with area under crop. On the same manner, Abay (2007) on the study of Vegetable marketing in Fogera woreda indicated that quantity produced has positive effect on tomato market supply. 76? ? The interpretation of the marginal effect is straight forward like any OLS interpretation. The results suggest that a one quintal increase in quantity of paddy production leads to an increase of about 0.12 quintal of sales. On average, if paddy producer gets educated, the amount of paddy supplied to the market increases by 0.96 quintal. This suggests that education improves level of sales that affects the marketable surplus. On the other hand, if a family size increased, the amount of paddy supplied to the market would be decreased by 0.25 quintal. Table 35. Factors influencing the level of rice crop sales/ OLS/ results VARIABLES COEFF. (STD.ERR.) T-RATIO CONSTANT -1.949 (2.970) -0.656 SEX 2.123 (1.817) 1.168 EDU 0.960* (0.520) 1.84636 FS -0.253** (0.133) -1.90631 MKD -0.088 (0.260) -0.336 TQP 0.128*** (0.023) 5.70103 CREDIT 0.024 (0.586) 0.040 NFINC 0.070 (0.299) 0.233 LMP 0.002 (0.001) 1.609 MINF B 0.795 (0.770) 1.032 IMR 0.719** (0.376) 1.91454 R-squared = 0.2493484 F[ 9, 155] (prob) = 4.92 (.0000) Adjusted R-squared = 0.2006048 Log likelihood = -414.9390 Rho = cor[e,e(-1)] = 0.1104117 Restricted(b=0) = -438.6011 Chi-sq [ 10] (prob) = 47.32 (.0000) B = Predicted MINF (endogenous variable) Note: ***, ** and * show the values statistically significant at 1%, 5% and 10% respectively 77? ? 4.4.2. Tobit model results Tobit model tends to answer the two questions by identifying the factors affecting the decision to participate and the level of participation at the same time. Table 36, presents Tobit model results. The result indicates that quantity of paddy produced jointly affected both the probability of market participation and volume of supply. Quantity produced is significant at1% level (p<0.000). This analysis reveled that applying Heckman two step model is appropriate because there was selection bias but if we had been using OLS model instead of Heckman two step model, the coefficients would have been inefficient. One of the weaknesses of Tobit model is that it assumes all producers are potential suppliers of a good and that volume of supply and market participation are influenced by the same variables in the same way (Blaylock and Blisard, 1993). 78? ? Table 36. Maximum likelihood estimates Tobit model VARIABLES COEFF. (STD.ERR) T-RATIO Change in probability /participation/ Change among rice sellers /intensity/ Total marginal effect CONSTANT -12.337 (6.939) -1.778 -3.25471 -.00123 -.00123 AGE -0.016 (0.056) -0.288 -.00423 .00000 .00000 SEX 8.254 (5.264) 1.568 2.17742 .00083 .00083 EDU 1.618 (1.263) 1.282 .42697 .00016 .00016 FS -0.339 (0.407) -0.834 -.08953 -.00003 -.00003 FL 0.008 (1.062) 0.007 .00201 .00000 .00000 EXC 0.998 (1.896) 0.527 .26340 .00010 .00010 MKD -0.328 (0.566) -0.579 -.08643 -.00003 -.00003 TLS 0.012 (0.820) 0.015 .00326 .00000 .00000 TQP 0.255*** (0.053) 4.780 .06721 .00003 .00003 OXN -0.607 (0.965) -0.628 -.16002 -.00006 -.00006 CREDIT -0.465 1.362) -0.341 -.12255 -.00005 -.00005 NFINC 0.307 (0.579) 0.531 .08107 .00003 .00003 LMP 0.003 (0.002) 1.241 .00072 .00000 .00000 TLU 0.158 (0.303) 0.524 .04181 .00002 .00002 MINF B 1.789 (2.332) 0.767 .47189 .00018 .00018 Log likelihood function = -60.1125 LM test [df] for tobit = 103.664[ 16] Number of observation = 165 B = Predicted MINF (endogenous variable) Note: ***, ** and * show the values statistically significant at 1%, 5% and 10% respectively 79? ? 4.5. Analysis of Structure-Conduct and Performance In this part of the thesis, rice marketing participants and market structure, conduct and performance will be discussed. 4.5.1. Profile of rice traders in Fogera The survey result showed that, wholesalers are fairly young average 36.6 years old and millers it is about 38.6 years (Table 37). On average, a wholesale trader household consists of five to six members and in millers about 6. Often family members are also involved in the business and usually act as accountant; or managers. It is indicated that most owners/managers in the wholesale market are male: about 93.3 percent. This also holds true for millers. In general male, dominate in the rice trade (wholesaling, milling, distributing, assembling and retailing). Table 37. Personal profile of rice traders Characteristics Wholesalers Millers Urban Distributors Assemblers Retailers Age of trader Std. Deviation 36.6 (4.87) 38.66 (12.64) 43.4 (10.47) 32.52 (6.63) 29.5 (8.05) Sex All male All male 4-male 1-female All male 7-male 1-female Number of family size Std. Deviation 5.5 (2.38) 6.25 (1.83) 5.4 (3.36) 4.6 (1.41) 5.67 (3.38) Number of persons employed Std. Deviation 8 (6.74) - - - - - - 2.13 (0.835) Family members employed Std. Deviation 1.8 ( 0.447) 1.67 (1.225) 2.2 (0.837) 1.56 (0.651) 1.88 (0.835) Non-family members employed Std. Deviation 6.2 (6.6) 2.11 (1.36) 1.2 (1.78) 0 0 0.25 (0.46) Source: Survey result, 2008/9 80? ? As shown in Table 38, rice wholesalers have 4.6 years experience in rice trading. On average, the rice millers and distributors just have 9 and 5 years experience and the rice assemblers and retailers have 7and 5 years of experience respectively. Table 38. Commercial profile of rice traders Characteristics of respondents Wholesalers Millers Urban distributors Assemblers Retailers Years of experience 4.6 (1.94) 9.33 (9.0) 5.00 (3.082) 7.20 (2.70) 4.88 (3.78) Permanent male employees 3.2 (1.48) 3.33 (2.0) 1.8 (1.30) 1.52 (0.714) 2.13 (0.84) Permanent female employee 0.40 (0.548) 0.33 (0.5) 1.6 (1.94) 0 0 0 0 Temporary employees 0.40 (0.89) 0.56 (0.8) 0.40 (0.89) 0.040 (0.20) 0 0 Source: Own survey result, 2008/9. Numbers in parenthesis are standard deviations. Table 39 shows the current asset of the rice traders. The average value of assets is much higher among rice wholesalers, 515,943.6 Birr, while it is 333,927.27 Birr for rice millers and 66,283.8 Birr for rice distributors and only 12,879.6 for Assemblers. The initial working capital for wholesalers was high, for millers fairly low and for assemblers very low. 81? ? Table 39. Average value of asset for traders (in Birr) Characteristics Wholesalers Milers Urban distributors Assemblers (n=5) (n=10) (n=5) (n=25) Residence house 206000 77777.8 20447.4 12060 Separate store 155000 166600 10179.4 0 Store residence 0 5000 0 Mobile telephone 2413.6 2033.6 1059.4 703.2 Fixed line telephone 1130 450 219.4 116.4 Vehicle /personal truck / 100000 18500 30059.4 0 Bicycle 1000 615 259.4 0 Motor bicycle 0 0 0 0 Milling machine 50400 62950 0 0 Total value of shop shed in birr currently 0 0.9 4059.4 0 Total value 515944 333927 66283.8 12879.6 Amount of initial working capital 26720 20909.4 12300 4188 Amount of working capital currently (2007/8) 1092500 76666.7 44650 16560 Source: Survey result, 2008/9 4.5.2. Characterization of marketing actors In the study area there are no traders who specialized in rice trading but they are grain traders in general. According to urban trade and industry office of the woreda there are 9 licensed grain wholesalers, 66 grain retailers and 26 rice millers or processors in 2008/9. Most grain traders are licensed and some are trading rice with out license, for instance, assemblers and some times brokers. Market participant (traders) can be characterized from the point of rice trading into different groups: 1. Producers: Producers are the first link in the marketing chain. Farmers produced paddy and sold to Woreta market or to local village market like (Hod Gebeya and Makisegnit). Out of 113 respondents 68.5% of the sample households answered that they sold to Woreta market (capital of Fogera Woreda) and the rest to local village market points. Farmers sell their rice through different channels or roots. The main four channels are wholesalers and millers (71.9%), rural assemblers (14.1%), urban assemblers (11.9%) and consumers (2.2%) respectively. 82? ? Rural assemblers are traders who collect rice from farmers at local markets during market days and sell it to wholesalers or millers. The markets are placed in remote areas which are open once in a week usually to satisfy some farmers need. There are two main local markets these are Hod Gebeya and Makisegnit Gebeya market points. Urban assemblers are few in number and purchase rice from producers during market days. They used to sell to wholesalers? only to get better price. Table 40. Percentage of rice market outlets Outlets Frequency Percent Rural Assemblers 19 14.1 Urban Assemblers 16 11.9 Consumers 3 2.2 wholesalers and millers 97 71.9 Total 135 100 Source: survey results, 2008/9 Farmers transport rice to the nearest markets (village market or Woreda market) using pack animals (90.9%), and the smaller percentage used head/ backload, animal carts and vehicles. Large amount of grains is sold and purchased in the months of production season (December through March,) which is the month?s immediately after harvest. Supplies of rice decrease in the months of May through October and reach the lowest level. The study shows that 21.7% of rice producers sales their out put immediately after harvest followed by 20.4% after three months and 11.8% after two and four months respectively. 2. Wholesalers: These are licensed grain wholesalers who store large bulk and assemble grains in either direction. Wholesalers don't move form one market to another like that of petty grain traders. They rather, permanently reside in town with their permanent store and collect rice grains brought by farmers, assemblers (rural and urban) and processors. They are few in numbers and most of the time they sold rice to Addis Abeba . 3. Millers (processors): Theses millers were licensed for both milling machine and retail trade. Millers, who is the owner of milling machine, have double participation in rice trading, 83? ? firstly they have involved in milling the paddy rice, secondly, they will purchased this milled rice for themselves to sold. They stored and sold rice to Addis Abeba, to locally available urban distributors and to consumers. Most of the time millers distribute regional wise. The distribution centers are Addis Abeba, Wollo, Bahir Bar, Gondar and Woldia. They collect rice from farmers, and rural assemblers. Except brokers almost all traders owned rice milling machine. Informal interview with brokers also told that there are 14 traders having with 2 milling machine and 8 traders having with one milling machine in Woreta Town. 4. Brokers: These are unlicensed legally but in reality they are doing like wholesaling activity. They don?t have warehouse. Informal interview with traders indicated that currently only three main brokers are available at the Woreda town. They facilitate buying and selling other traders and sometimes their own purchase. No broker activities were reported from farmers in buying and selling activities. 5. Assemblers: These are also unlicensed assemblers of rice. They are rural and urban assemblers. The numbers of assemblers in the selected administrative kebeles were estimated to be 75 in Nabega, 25 in Kidist Hanna and 20 in Diba Sifatra respectively. They collect rice during main market day at local market points. 6. Urban distributors: These are grain traders which reside in towns or cities regionally and distribute grains including rice in ether direction, incase of rice they receive and transmit to consumers and retail shops. Discussion with traders indicates that there are 5 distributors at Bahir Dar, 10 at Gonder, 10 at Woldia and around 20 urban distributors at Addis Abeba. 7. Retailers: These are shop retailers who has legally licensed for retailing different products they are not specialized to sell rice only but used as a complement to other grain products for customers. They purchase smaller quantity and it takes a longer time to finish selling. They usually purchase from distributors incase of Bahir Dar, Gondar and Woldia but incase of Woreta they have alternatives, to purchase either from millers, wholesalers, farmers or assemblers. 84? ? 4.5.3. Rice market channels The analysis of channel is intended to provide a systematic knowledge of the flow of the goods and services from their origin to the final destination (consumer). The rice market channel drawn based on the data collected from different sources. The total quantity produced by farmers was about 2090.2 quintal and the total quantity supplied to the market is 479 quintal from sampled farmers. Twenty four lines of market channels were identified. Five of these went outside the region and the rest sixteen ran inside. As can be understood from Figure 1, the main receivers from farmers were, wholesalers, Millers, Rural assemblers, Urban assemblers with an estimated percentage share of 44.9, 26.9,14.1 and 11.9 percent in that order. Besides, the volume that passed through each channel was compared and based on the result the channel that went out of region consisting 95 quintal hosted the largest, followed by channels that stretched from Farmer-? Wholesalers? Retailers? Consumers hosted 81.98 qt respectively. There are 9 main channels of rice marketing based on the volume (channel-3, 6, 7, 8, 19, 20, 21, 22, 23). 1. Farmer? Assemblers (urban) ? Wholesalers ? Out of region = 13.50Q 2. Farmer? Assemblers (urban) ? Wholesalers ? Consumers =16.30Q 3. Farmer? Assemblers (urban) ? Wholesalers ? Retailers? Consumers =21.72Q 4. Farmer? Assemblers (urban) ? Wholesalers ? Distributors (urban) ? Consumers= 13.05Q 5. Farmer? Assemblers (urban) ? Wholesalers ? Distributors (urban) ? Retailer ? Consumer = 8.70Q 6. Farmer? Wholesalers ? Out of region = 50.97Q 7. Farmer? Wholesalers ? Consumers = 61.53Q 8. Farmer? Wholesalers ? Retailers? Consumers = 81.98Q 9. Farmer? Assemblers (rural) ? Wholesalers ? Out of region = 7.20Q 10. Farmer? Assemblers (rural) ? Wholesalers ? Consumers = 18.23Q 11. Farmer? Assemblers (rural) ? Wholesalers? Retailers ? Consumers = 11.58Q 85? ? 12. Farmer? Assemblers (rural) ? Wholesalers? Distributor (urban) ? Consumers= 6.96Q 13. Farmer? Assemblers (rural) ? Wholesalers? Distributors (urban) ? Retailer ? Consumers =4.64Q 14. Farmer? Assemblers (rural) ? Millers? Out of the region= 6.83Q 15. Farmer? Assemblers (rural) ? millers? Distributors (urban) ? Consumers = 6.49Q 16. Farmer? Assemblers (rural) ? Millers? Distributors (urban) ? Retailers? Consumers = 4.32Q 17. Farmers? Assemblers (rural) ? Millers? Retailers? Consumers = 8.66Q 18. Farmer? Assemblers (rural) ? Millers ? Consumers = 10.82Q 19. Farmer? Millers? Out of the region = 23.70Q 20. Farmer? Millers? Distributors (urban) ? consumers =22.52Q 21. Farmer? Millers? Distributors (urban) ? Retailers? consumers = 51.54 22. Farmer? Millers? Retailers? consumers = 30.04Q 23. Farmer? Millers? consumers=37.54Q 24. Farmer? consumers=10.53Q 86? ? Figure 2. Rice marketing channels Rural assemblers Farmers (479QL) Millers/processors (2 0 ) Wholesalers (1 0 ) Urban Distributors Retailers Out of the Region ??14.1%11.9% 44.9% 23.32% 29.14% 23.7% 18.4% 26.9% 60% 40% 100% 29.14% 38.12% 38.18 % 28.61% 100% Urban assemblers 55% 45% Consumers Intermed ia ries 42.1% 87? ? 4.5.4. Analysis of structure of the market According to Pender et al. (2004), the structure of the marketing system should be evaluated in terms of the degree of market concentration, barrier to entry (licensing procedure, lack of capital and know how, and policy barriers), and the degree of transparency. The structure analysis of rice market will be based on the above two points. 4.5.4.1 Barriers to entry into the rice market The barriers to entry into the market reflect the competitive relationships between existing traders and potential entrants. If the barriers to entry are low, new traders can easily enter into rice markets and compete with established traders. However, with the presence of very high barriers to entry, established firms are difficult to stay longer in business. 4.5.4.1.1. Capital investment The survey result indicated that various barriers to entry into the rice business were identified by the traders (wholesalers and millers): lack of investment capital, high competition with prior control of farmers, information asymmetry and severe competition among none-licensed traders were the main ones. For wholesalers and millers, the most important barrier to entry was high competition with prior control of farmer and lack of investment capital. To enter in the market more capital is needed because they have to purchase more rice while his regular customers are coming during harvesting (peak purchase) time. They did not allow farmers go without purchase the available amount of paddy they brought, if they do so they will loss his customer at least in the short period of time. 88? ? Table 41. Barriers to entry for rice market Number of response on sampled rice wholesaler and millers (n=15) Barriers to entry Frequency Percent Capital and high competition to control farmers 5 33.3 Capital 3 20 High competition to control farmers 1 6.7 Information asymmetry and quality of rice 1 6.7 Information and high competition to control farmers 1 6.7 High competition and lack of working place 1 6.7 Capital, information asymmetry and high competition 1 6.7 Capital, competitions among traders and high competition to control farmers 1 6.7 Capital, high prior to control farmers and lack of working place 1 6.7 Total 15 100 Source: Survey result, 2008/9 4.5.4.1.2. Experience and education levels of rice wholesalers and millers The survey result indicate that about 47% of the respondents have experience in rice trading between 2-5 years, 40% of them had experience of 6-10 years and 6.7% had 11-20 and the remaining 6.7 % had above 21 years of experience respectively. With regard to education level, about 64 % were in secondary education and the rest are in primary education level. This indicate that education is not a barrier to rice traders because majority of rice traders had formal education Table 42. Education level of wholesalers and millers Education level of trader Frequency Percent Primary school education 5 35.7 Secondary school education 9 64.3 Total 14 100 Source: Own survey, 2008/9 89? ? 4.5.5. Conduct of rice traders Market conduct refers to the set of competitive strategies that a trader or a group of traders use to run their business. In other words, market conduct focuses on traders? behavior with respect to various aspects of trading strategies such as buying, selling, transport, storage, information and financial strategy. In line with the literature on institutional economics, these are called the rules that define the play of the game. 4.5.5.1. Purchasing strategy The survey result indicated that 66.7 % the wholesalers and millers were no any purchasing relationship based on ethnicity, family linkage and cloth relatives. Only about 7% purchase based on close relatives, socially meeting and some either combination was used. Most of the time wholesalers and millers buy 80% of rice from Woreta (on their ware house) and 20% from village market. The reason to stay more in that area was due to high supply and better quality of rice than to look for other markets. The purchasing strategy for wholesalers revealed that 13.33% of the sampled wholesaler purchase based on the long term client establishment, infra family link and spontaneous purchasing, 6.7% purchase with out median agent. The remaining percent were used to purchase on contract, broker and a combination of either methods. Convenient time of day preferable to purchase rice in terms of price was before 12a.m. 4.5.5.2. Pricing strategy About 53% of sample traders indicated that price is set by the market. But 27% of them are setting prices by themselves, 13% set by negotiation of buyer and traders and the rest was by marketing experts from Woreda Agriculture office. 90? ? 4.5.6. Market performance 4.5.6.1. Degree of buyers and sellers concentration The degree of buyer and seller concentration refers to the number of rice traders in the rice market. This concentration ratio can be interpreted as an indicator for the degree of competitiveness among rice traders. The study indicates that the rice market is dominated by few wholesalers. The CR4 ratio is about 77%. That means 77% of the market volume is occupied by few wholesalers (Appendix Table 8). The calculation of the concentration indices for both wholesalers and millers together is about 82.32%. This indicates the market is strongly oligopsonistic (Appendix Table 9). Black (2002), defined oligopoly is a market situation with only a few sellers, each anticipating the other reaction, where as oligopsony as a situation where there are only a few buyers in the market. 4.5.6.2. Marketing cost and margin analysis of rice traders The marketing margin refers to the difference between prices at different levels in the marketing system. The total marketing margin is the difference between what the consumers pays and what the producer/farmer receives for his paddy or rice, in other words it is the difference between retail price and farm price. A wide margin means usually high prices to consumers and low prices to producers. The total marketing margin may be subdivided into different components; all the costs of marketing services and profit margins or net returns. An analysis of marketing costs would estimate how much expenses are incurred for each marketing activity. It would also compare marketing costs incurred by different actors in the channel of distribution. 91? ? 4.5.6.2.1. Marketing cost and margin of producers Marketing cost of farmers are cost incur in transportation, loading and unloading and cost of milling for those farmers who sold after polishing (polished rice) which is summarized in (Table 43). Table 43. Marketing cost and margin of farmers or producers Cost items Paddy /rice Cost per unit (birr/qt) % Production cost /qt 332.43 94.53 Transporting cost /qt 9.33 2.65 Milling cost/qt 9.9 2.82 Total cost 351.66 100 Average selling price (paddy/rice) 387.63 Profit/Q 35.97 Source: Own computation, 2008/9 4.5.6.2.2. Marketing cost and margin of assemblers The marketing cost of rice for rural and urban assemblers is summarized in Table 44 below. The study indicates that the main cost of rural assemblers are transport cost, personal travel cost and sorting and milling costs which is consisting of 14-33% of the total cost, while in urban assemblers the main cost components are sorting cost, information cost and personal travel costs which ranges from 13-19 % of the total cost. 92? ? Table 44. Marketing cost and margin of assemblers. Cost Items Rural assemblers (N=21) Urban assemblers (N=5) Total (N=25) Average cost/qt % of total cost Average cost/qt % of total cost Average cost/qt % of total cost Cost of packaging material 4.5 6.37 5.2 10.23 4.85 7.98 Labor cost to fill the bag and stitch 1.8 2.54 2 3.93 1.9 3.12 Transport cost 23.35 33.05 3 5.9 13.17 21.69 Cost of storage loss 4.82 6.82 4.49 8.83 4.65 7.66 Cost of loss in transport and handling 4.41 6.24 6.12 12.04 5.26 8.67 Sorting cost /milling/ 10 14.15 10 19.68 10 16.46 Information cost 3.1 4.38 8 15.74 5.55 9.14 Market search cost /fee 2.65 3.75 5 9.84 3.82 6.29 Personal travel cost 15 21.23 7 13.77 11 18.11 Other overhead cost 1 1.41 0 0 0.5 0.82 Total cost per qt 70.63 100 50.81 100 60.72 100 Average selling price 554.75 630 592.37 Average buying price 378 408 393 Margin 176.75 222 199.37 Profit /Q 106.12 171.19 138.65 Source: Own survey result, 2008/9 4.5.6.2.3. Marketing cost and margin of wholesalers The marketing cost of rice wholesalers in the study area are summarized in Table 45. On average, the total marketing cost of rice wholesalers are 29.24 Birr per quintal. Cost of storage loss, cost of packaging material, cost of loss in transportation and handling and employer?s salary are highest cost items (14-28 percent of the total cost). 93? ? Table 45. Average total cost and margin of wholesalers Cost Items (N=6) Average cost/qt Stdev % of total cost Cost of storage loss 8.36 7.68 28.59 Cost of packaging material 5.8 0.45 19.84 Cost of loss in transportation and handling 4.18 6.25 14.29 Labor cost for loading 3 0 10.26 Labor cost for unloading 3 0 10.26 Labor cost to fill the bag and stitch 1.4 0.89 4.78 Employers salary 1.043 0.53 3.57 Cost for brokers commission 1 0 3.42 Cost for store rent 0.7 0.45 2.39 Market search cost/fee 0.34 0.51 1.17 Tax (1) 0.16 0.12 0.56 Watching and warding cost (2) 0.13 0.16 0.46 Interest rate /cost 0.07 0.16 0.24 Personal travel cost 0.03 0.06 0.09 License cost 0.02 0.02 0.07 Total cost per qt 29.24 13.84 100 Average selling price 708 4.47 Average buying price 670 28.06 Margin 38 24.14 Profit /Q 8.76 36.05 Note: (1) Tax fee is taken as based on proportion to grain volume hold. (2) Watching and warding cost for rice wholesalers are taken as 10% of the total amount of grain volume cost. (3) Transport cost is covered by the rice purchaser not by wholesalers Source: own survey result, 2008/9 4.5.6.2.4. Marketing cost and margin of millers Table 46 shows the marketing cost of rice millers. The major cost items are storage costs which are 9.53 birr per quintal, loose in transporting and handling, 6.63 Birr per quintal, cost of packaging material, 5.2 Birr per quintal, and loading and unloading 2.5-2.9 Birr per quintal, employers salary, 1.10 Birr per quintal, electricity used for operating the machines 5 cents per quintal and maintenance costs 0.48 cents per quintal. On average milling of one quintal of paddy costs 9.9 Birr per quintal, processing cost for enjera or consumption costs 8.75 Birr per 94? ? quintal and processing cost for hotels is 10 Birr per quintal. The total cost per quintal is 32.68 Birr. Table 46. Average total cost and margin of millers/processors. Items Average cost/qt Stdev % of the total cost Cost of storage loss 9.53 5.46 29.17 Cost of loss in transportation and handling 6.63 8.67 20.29 Cost of packaging material 5.27 0.68 16.13 Labor cost for loading 2.91 0.30 8.90 Labor cost for unloading 2.55 1.04 7.79 Employers salary 1.11 0.94 3.40 Labor cost to fill the bag and stitch 1.09 0.30 3.34 Cost for store rent 0.89 0.71 2.73 Cost for brokers commission 0.86 0.32 2.64 Electricity 0.5 0.20 1.53 Maintenance cost 0.49 0.00 1.49 Transport cost of Head/back load 0.45 0.93 1.39 Watching and warding cost (2) 0.15 0.17 0.49 Tax (1) 0.10 0.06 0.3 Market search cost/fee 0.08 0.11 0.24 Personal travel cost 0.05 0.09 0.15 License cost (3) 0.03 0.02 0.08 Total cost per qt. 32.682 10.66 100 Average Selling price 656.64 87.67 Average buying price 619.55 84.54 Margin 37.09 32.61 Profit/Qt 4.408 28.35 Note: (1) Tax fee, is taken as based on proportion to grain volume hold. (2) Watching and warding cost, are taken as 10% of the total grain volume cost. Millers sell other crops also, there is no specialization of selling rice only. (3) Milling cost usually covered by farmers, millers receive charges for their milling service. The advantage of having a milling service is to collect more rice and also to get milling charges. (4) Electricity cost, fuel cost, and maintenance cost are estimated from 3500-4000 Birr/year. Source: Own survey, 2008/9. 95? ? 4.5.6.2.5. Marketing cost and margin of urban distributors Compared to rice wholesalers, rice distributors and retailers incur more marketing cost (82.10 and 79.31 Birr/q) respectively. The most important cost item is store rent, storage loss and sorting cost respectively. Table 47. Average marketing cost of rice distributors Items Average cost/qt Std. dev % share of the total Cost for store rent 29.6 21.45 36.05 Cost of storage loss 14.19 3.32 17.28 Sorting cost 10 0 12.18 Transport cost of vehicle 9.2 1.10 11.21 Cost of packaging material 4.5 6.84 5.48 Cost of loss in transportation and handling 3.95 5.50 4.81 Labor cost for loading 3 0 3.65 Labor cost for unloading 3 0 3.65 Personal travel cost 2.033 1.92 2.48 Market search cost/fee 1.03 0.96 1.25 Tax (1) 0.84 0.75 1.02 Watching and warding cost (2) 0.49 0.17 0.60 Labor cost to fill the bag and stitch 0.2 0.45 0.24 License cost (3) 0.10 0.13 0.09 Total cost per qt 82.10 21.44 100 Average Selling price 782 Average buying price 696 Margin 86 Profit/Qt 3.898 Note : (1) The tax for rice distributors are taken as 10- 20% proportion to the total tax levied for the grain volume hold. (3) License fee cost is taken as 10% of the total amount of grain volume cost. 4.5.6.2.6. Marketing cost and margin of retailers The marketing cost of retailers at Bahir Dar, Gondar and Woreta are summarized in Table-48. The result shows that the marketing cost of rice were 79.3, 75.34 and 52.34 Birr per quintal respectively. The marketing margin for Gondar is highest among all markets. Besides, the cost in Gondar per quintal of rice is 75.34 Birr which is very low compare to the tree market places. 96? ? Table 48. Marketing cost and margin of retailers. Bahir Dar (N=22) Woreta N=(10) Gondar (N=29) Cost Items Cost Birr/qt STDEV (% ) Cost Birr/qt STDEV (% ) Cost Birr/qt (% ) Cost of packaging material 18.88 10.48 23.81 15.93 9.8 21.14 8.3 15.85 Labor to fill the bag and stitch 1.38 4.38 1.74 2.22 1.3 2.94 0 0 Labor cost for loading 0.15 0.56 0.2 2.51 1.04 3.34 0 0 Labor cost for unloading 0.25 0.68 0.31 2.46 1.74 3.26 0 0 Cost for brokers commission 0 0 0 Transport cost of vehicle 5.81 2.83 7.33 8.75 7.75 11.62 0 0 Head/backload transport cost 0.22 0.75 0.28 3.6 6.87 Cost for store rent 7.02 9.34 8.85 2.2 0.44 2.91 0 Cost of storage loss 20.19 8.7 25.46 16.18 5.65 21.47 10.2 19.48 Cost of loss in transport& handling 1.82 4.7 2.3 12.15 7.56 16.13 11.75 22.44 Sorting cost 6.79 6.48 8.56 14.88 14.62 19.76 0 Other cost arrangement 0 0 0 12.5 10.6 16.59 0 Tax 5.56 13.31 7.01 1.82 2.39 2.42 8.66 16.54 License cost 0.44 0.78 0.56 1.36 2.06 1.8 3.83 7.31 Cost/ interest rate 0 0 0 0.002 0.012 0.003 0 Market search cost/fee 3.8 7.77 4.79 8.88 22.76 11.79 4.5 8.59 Watching and warding cost 1.89 3.38 2.39 4.62 11.66 6.13 1.5 2.86 Personal travel cost 5.02 16.09 6.34 11.26 22.68 14.95 0 Total cost/qt 79.3 31.75 100 75.34 50.09 100 52.34 100 Average selling price 814.61 106.01 947.65 105.34 770 Average buying price 726.25 106.53 747.67 81.73 699 Margin 88.36 52.35 199.98 97.18 71 Profit/Q 9.05 44.75 124.63 125.81 18.66 Note: (1) The tax cost, license fee cost, market search fee, watching and warding cost for rice retailers are taken as 2-15% of the total that grain volume cost. Source: own survey, 2008/9 97? ? 4.5.6.3. Marketing costs, gross margin and profit margin of traders Table 49 gives an overview of distribution of marketing margin among different actors in the channel. Assemblers (rural and urban) get the highest gross marketing margin (value added), which is 199 birr per quintal. Rice millers and wholesalers got almost equal gross margin (around 40 Birr/quintal). But millers get the lowest margin (37.09 Birr/qt). Table 49. Summary of marketing cost, margins and profit of farmers and traders Cost Items Cost and prices (birr/q) Gross marketing margin (1) (1) Total marketing cost (2) Profit margins (birr/q) (3)=(1)-(2) Amount 3 As % of cost price I Farmers 55.2 19.23 35.97 10.22 1. Production cost /qt 332.43 2. Total marketing cost 19.23 3. Cost price (3=1+2) 351.66 4. Average selling price 387.63 II. Assemblers 199.37 60.72 138.65 30.55 1. Average buying price 393 2. Total marketing cost 60.72 3. Cost prices (3=1+2) 453.72 4. Average selling price 592.37 IV. Millers 37.09 31.69 5.4 0.83 1. Average buying price 619.54 2. Total marketing cost 31.69 3. Cost prices (3=1+2) 651.24 4. Average selling price 656.63 V. Wholesalers 38 29.23 8.77 1.24 1. Average buying price 670 2. Total marketing cost 29.23 3. Cost prices (3=1+2) 699.23 4. Average selling price 708 VII. Urban distributors 86 82.10 3.9 0.50 1. Average buying price 696 2. Total marketing cost 82.10 3. Cost prices (3=1+2) 778.10 4. Average selling price 782 98? ? Table 49(continued) VIII. Retailers 119.58 68.99 50.59 6.37 1. Average buying price 724.50 2. Total marketing cost 68.99 3. Cost prices (3=1+2) 793.49 4. Average selling price 844.08 Note: (1) Gross marketing margin (value added) =Average selling price ?Average buying price. (2) Average selling and /buying price at different level was based on the own survey of this study, 2008/9. (3) The time dimension for profit margin is one year (2008/9) It can be observed that although rice assemblers get the highest marketing margin, they also incur the highest marketing cost (60.72 Birr/qt). Wholesalers got the lowest marketing cost (among traders excluding farmers) and urban distributors the lowest profit margin. The last column of Table 48 also indicates that among the different rice traders, rice assemblers obtain a relatively large profit as a percentage of the cost price (30.55%) and the lowest one is obtained by urban distributors (0.50%). 4.6. Production and Marketing Constraints of Rice 4.6.1. Producers? constraints ? Shortage of land: Shortage of land is the primary problem of the sample Pas. It is about 77% of the farmers respond for this problem. This situation reduces directly rice production .and forces the farmers to produce rice by renting land. ? Improved varieties: As indicated in Table 50, lack of improved varieties was responded positively by 76.1 per cent of the farmers. Most farmers cultivate local Variety X-Jigina (local variety) and the improved once are not yet widely disseminate and used by farmers. Only one variety called Gumara (IAC-164) which is released by Adet Agricultural Research Center is currently used but the color is red produces red enjera and is not accepted by farmers for marketing. It needs attention to look for early maturing and better yielding variety. 99? ? ? Diseases and pests: About 22 percent of the farmers also respond facing with problem of diseases and pests. According to IPMS (2005), the identified Diseases/pests for rice were wave worm, shoot fly, rice hispid (weevil) and rice blast. ? Shortage of seed supply: This is another problem as 36.2 per cent of farmers perceived it. It is also observed that 14.1 per cent of the farmers are lacking of improved post harvest management technologies such as storage and storage facilities. ? Lack of polishing technology: Problems of threshing machine or polishers were responded positively by 55.8 per cent of the farmers. This has an effect on the quality of rice for marketing. ? Malpractice in selling method (Scaling or Weighing): About 45 percent of the respondents were complaining various malpractices such as scaling or weighing, deduction, and quoting of lower prices than actual. ? Lack of market: About 33% also respond that there were market problems associated with low output price, maintenance of standards and grades. For Example, during husking, grains are broken in to pieces (farmer usually used traditional threshing i.e. by beating with stick and using ox) and this broken grain decreases market demand. ? Lack of information exchange: Poor contact or communication was also one of the problems of farmers. Information on market price, demand and supply is also mentioned as a problem by sample households. ? Transportation problem: About 47% of the sampled farmers were responding positively about transportation problem. During raining seasons as the area is near to Lake Tana, excessive flooding is a common problem and transportation is difficult especially in this period. 100? ? ? Lack of capital and credit availability: About 46% for capital shortage and 40% for credit availability of the sample producers respectively have responded these problems. Farmers have an urgent need for money immediately after harvest. Even if the price of paddy is always at lowest during that period, farmers badly needed cash during this period in order to pay their rent and debts as well as to buy certain necessities. Most of the time, lack of post-harvest credit forces farmers to sell their produce immediately after harvest, when prices are low. Table 50. Production, marketing and institutional problems of farmers No Description Number of respondents Percentage (%) A Production aspect 1 Problems of availability of improved rice variety (lack of improved and high yielding varieties) 163 14.7 2 Problems of fertilizer supply for rice production 163 14.7 3 Chemical supply problem 163 11 4 Seed supply problem 163 36.2 5 Shortage of land 126 77.3 6 Disease problem 163 22.3 7 Problems of farm implement 163 9.2 8 Problems of post harvest technology /storage loss/ 163 14.1 B Marketing aspect 1 Lack of market 163 33.1 2 Problem of price setting 163 27 3 Malpractice in selling method (scaling or weighing ) 163 44.8 4 Information exchange problem 163 21.5 5 Problem of storage facilities 163 19 6 Problems of threshing machine or miller /quality/ 163 55.8 C Financing and institutional aspect 1 Loan repayment problem 163 22.7 2 Lack of capital availability 163 45.4 3 Problems of credit facility 163 39.9 4 Transport problem 163 47.2 5 Lack of institutional support 163 13.5 6 Problem of theft 163 33.7 7 Problem of tax or double taxing 163 32.5 8 Problems of excess water (flooding) 163 8 Source: own survey, 2008/9 101? ? 4.6.2. Traders? constraints a) Wholesalers and millers As indicated in Table 51, the major problem of wholesalers and millers is capital shortage. This is responded by 53.7% followed by lack of information and high tax payment (20%). Usually millers as well as wholesalers pay tax based on the number of milling machine they have and their licensed trading. Another problem which was responded for wholesalers and millers were prior control of farmers (handling and attracting farmers to be a client supplier before other competitors handled) followed by lack of reliable information and competition. It is responded by 20% of the sampled millers and wholesalers. Table 51. Problems of wholesalers and millers in rice market Number of response on different levels (n=15) Problems frequency percent Lack of capital 8 53.3 Lack of Information and competition 3 20 High tax rate 2 13.3 License procedure 1 6.7 Lack of information and high prior to control of farmers 1 6.7 Total 15 100 Source: Own survey, 2008/9 b) Problems associated with retailers The common problem perceived by sample retailers at Bahir Dar, Gondar and Woreta are shortage of capital, quality, adulteration, and credit. The problem associated with retailers especially related to rice crop is quality. About 90% of sampled retailers at Bahir Dar responded that the quality of rice produced from Fogera is low as compare to the imported one. The common imported rice type available in shops and supper markets are Basmati rice (Pakistan), Ponte rice (Italy) and Dana rice (Pakistan). Similarly the problems of retailers at Gondar related to rice were capital shortage, tax payment, quality of rice, storage problems and competition with unlicensed traders. 102? ? Table 52. Main problems of retailers Bahir Dar Gondar Problems Yes % Total Yes % Total Taxation and other fees 16 59.3 27 16 55.2 29 Shortage of supply of rice 5 18.5 27 14 48.3 29 Storage 10 37 5 17.9 28 Quality 24 88.9 27 14 48.3 29 Adulteration 21 77.8 27 10 34.5 29 Information flow 6 22.2 27 12 41.4 29 Capital shortage 18 66.7 27 17 58.6 29 Access to credit 15 55.6 27 11 37.9 29 Too much competition with unlicensed traders 16 59.3 27 11 39.3 28 Un availability working place in the market 18 66.7 27 6 29 Source: Survey result, 2008/9 C) Problems of millers About 25% of the respondents complain lack of market facilities, low quality of farmers? rice due to problem of threshing, improper handling and harvesting of farmers (spoilage). Storage facilities are also a problem which is responded by 25% of the mille owners. Table 53. Problems to millers Problems Number of respondents (yes) Percentage (%) Low quality of rice 3 25 Lack of improved rice storage facilities 3 25 Lack of appropriate market facilities 3 25 Lack of improved rice threshing machines 2 16.7 Lack of improved rice huller or polisher 2 16.7 Source: Survey result, 2008/9 D) Rice assemblers The main problem associated with assemblers are road accessibility specially during flooding, lack of market , storage problems , capital shortage, credit access, farmers reluctant to sell rice do to low price are the main one. Quality problem of rice, absence to support and improve rice marketing is also responded positively by 72% of the respondents. 103? ? Table 54. Problems of assemblers Problems Number of respondents (yes) Percentage (%) A. Market problem Storage and lack of market 25 100 Capital shortage and credit access 25 100 Farmers reluctant to sell due to low price 25 100 Quality problem 23 92 Absence of support to improve rice marketing 18 72 Adulteration 15 60 Information flow 15 60 Competition with licensed traders 1 4 Competition with unlicensed trader 1 4 B. Institutional problem Road and electricity access problem 25 100 Technical training 9 36 Theft 8 32 Business management 3 12 Telephone, tax, water availability 1 4 Source: survey result, 2008/9 104? ? 5. SUMMARY AND CONCLUSIONS 5.1. Summary Rice is a main stay of Fogera farmers and it is the only "Rice basket of the region ". The main objective of the study is to analyze the profitability rice production and marketing chain of rice in Fogera woreda. The study specifically has focused on the profitability of rice production of farmers and traders, structure and conduct of the rice markets. And it investigates factors contributing towards household?s market participation in rice market and volume of rice supplied to market. The study also assesses the support inputs services, and constraints and opportunities of rice market in the study area. The data were generated by using pre-tested structured questionnaires. Data were obtained both from primary and secondary sources. The primary information was collected by interviewing farm households. Secondary data were obtained from different sources like Rural and Development office, Trade and industry office the Woreda, IPMS, agricultural research centers, Inland Revenue offices, publications and research studies, CSA, websites and agricultural magazines. A total of 165 farmers, 6 wholesalers, 10 millers, and a total of 60 retailers (from Bahir Dar, Gondar, Woreta) and 25 assemblers, 5 urban distributors were interviewed and the analyses were made using SPSS and LIMDEP. Summary of results obtained was the following. The descriptive analysis shows that the average family size of all households was 5.72 and with minimum 2 and maximum 13. The farmer?s average family labor force was 2.67 in man- equivalent with 6.15 maximum and 1 minimum. Rice producers are private farmers who produced paddy during main cropping season. The major reason for growing rice is for consumption and sale. In terms of land utilization rice is planted approximately on 0.6 hectares of land as compared with 0.36 and 0.31 hectares planted in Teff and Maize. 105? ? The production inputs used were seed and to some extent herbicides and pesticides. only 3% of the sampled households used urea, 1.2% use DAP and 4.9% used organic fertilizer for rice production The application of fertilizer was very minimum, because of flooding and the soil is fertile alluvial soil (Abay,2006; IPMA,2005). The common types of rice varities are X-Jigna (local) and Gumara (IAC-164.) the improved one. About 96% of the sampled household used X-Jigina variety (local and mostly popularized by farmers). However Gumara variety used less. Since it is red in color it is less demanded and used for consumption purpose as compare to the white seed X-Jigina variety which has high market demanded. From a total of sampled producers of households about 24% of rice producers were found to be non-sellers of rice mainly for different factors. Farmers have different market outlets and traveled 1.6 hour per trip to sell their product. Twenty four lines of market channels were identified. Five of these went outside the region and the rest sixteen ran inside. The main receivers from farmers were wholesalers, Millers, Rural assemblers, urban assemblers with an estimated percentage share of 44.9, 26.9, 14.1 and 11.9 percent respectively. Besides, the volume that passed through each channel was compared and based on the result the channel that went out of region consisting 95 quintal hosted the largest (42.1%) , followed by channels that stretched from Farmer? Wholesalers? Retailers? Consumers hosted 81.98 qt respectively. The central question for this study is "What will influence farmers' decisions to sell rice and what will stimulate them to sell more?" many variables were hypothesized for analysis. In order to test the above hypothesis, different methods were followed. The selectivity models encompass two steps to estimate factors on market participation and volume of sale. The result of the Heckman two step model indicates that market information access, quantity of paddy produced, extension contact with farmers and total livestock value increased the likelihood of households decision to sell rice. And education level and quantity of rice produced affects volume of rice sales positively but family size determines volume of sale 106? ? negatively. The Tobit result also revealed that quantity produced was jointly affected both the probability of market participation and volume of supply. The SCP model analyses also showed that the important entry barrier in rice market was high competition with prior control of farmer and lack of investment capital. They had fewer problems with taxes and license procedures. The survey result indicate that 46.7% of the respondents have 2-5 years of experience in rice trading and about 40% of them had 6-10 years of experience. Their educational status also indicates 64.3% were in secondary education and the rest are in primary education level. Regarding to pricing strategy 53.3% of sampled traders set price by the market, 26.7% set price by themselves, 13.3% set by negotiation of buyer and traders and the rest was by marketing experts. The four-firm Concentration Ratio (CR4) indicated that the rice market is dominated by few wholesalers. The CR4 ratio is about 77%. That means 77% of the market share going to major four wholesalers. This indicates the rice market is strongly oligopsonistic. The profitability analysis of rice production shows that, the gross income obtained from paddy production was birr 17549.21 per hectare and the total cost per hectare was 11688.23 Birr on samples households. Opportunity cost of land (rental value of land) , was the items occupying maximum share in total cost (40.23%) followed by labour cost (34.65%), animal power cost (13.11%). Material input cost like manure, herbicides, seed (10.26%) and other costs like land rent/ tax and interest rate (26.86%) consists of the minimum cost share. The cost benefit analysis of rice production shows that rice production is a profitable business for farmers. The net income obtained from production per hectare of rice is 5006.48 Birr. The cost margin indicates that producers obtain on average a profit of 35.97 Birr per qt with the market margin of 55.2 Birr per qt, assemblers get 139 Birr per qt, millers a profit of 5.4 Birr per qt, wholesalers 9 Birr per qt, urban distributors birr 3.88 Birr per qt and retailers around 19 Birr per qt. Though, assemblers get more profit, they also incur more marketing cost. 107? ? Constraints associated with farmers can be classified based on three categories, this are production constraints, marketing and institutional aspect. Shortage of land is the primary problem of the sample farm households in which 77% of households were respond it. The lack of improved varieties (disease resistant, high yield and early mature) was also a constraint in production which is responded positively by 76.1 per cent of the farmers. Most farmers cultivate local variety X-Jigina (local variety) than the improved variety Gumara (IAC -164). Marketing is the second main constraints of farmers. Problems of threshing machine or polishers to its marketing quality of rice were responded positively by 55.8 per cent of the farmers. And also 45% of the respondents were complaining various malpractices such as scaling or weighing, deduction and quoting of lower prices than actual. Moreover, about 33% also respond that there were market problems associated with low output price, maintenance of standards and grades. The last constraints for farm households are the institutional and financing aspect. The main problems were transportation facilities, capital and credit availability. About 47% of the sampled farmers were responding positively for transportation problem and 40% to 46 % for capital and credit respectively were perceived these problems. The major problems of wholesalers and millers are limitation of capital. This is responded by 53.7% followed by tax payment. Usually millers as well as wholesalers pay tax based on the number of milling machine they have and their licensed trading. Another problem which was responded for wholesalers and millers were prior control of farmers followed by information asymmetry and competition. It is responded by 20% of the sampled millers and wholesalers. The problems associated with assemblers are road accessibility, lack of market; storage problems, capital shortage, and credit access were the main once. With regarding to retailers, the common problems were shortage of capital, quality, adulteration, and shortage of credit. 108? ? 5.2. Conclusions and Recommendations Rice is a newly introduced crop in Ethiopia. However; it is increasing in production and area coverage. Rice is an exceptional crop due to its water loving nature and its higher productivity than other field crops. Though Ethiopia has tremendous area suitable for rice production little has been used until recently while many tones of imported rice are consumed in Africa as well as in Ethiopia. Hence, increasing production and productivity of this crop may contribute to food security. In Fogera and the nearby Woredas, rice is becoming a strategic crop for the livelihood of many farmers. In the past, the study area was very food insecure due to flooding problem. However, after the introduction of this crop, it is considered to be one of the surplus producing Woredas in South Gondar zone. The production trend shows that rice production increased from 160 qt in 1993/94 to 417,735 qt in 2007/08. Similarly, the area coverage of rice increased from 6 hectare in 1993/94 to 9,213 hectare in 2007/8. A number of factors may have affected market participation decision and volume of sales of rice in the country. In the case of Fogera district, the identified factors are access to market information, quantity of paddy produced, extension contact and livestock value were the main determinants of market participation decision for a household positively. For the volume of supply, household head?s education level (positively), quantity produced (positively), and family size (negatively) were the important variables that determines volume of rice sale in the market. Findings based on the results of the study (Heckman two-stage model), to promote rice market participation in a sustainable way, some policy implications are suggested to be addressed. 109? ? 1. Strengthening the existing price and market information system Generally, commercial farmers are capable of sourcing price and buyer information from different sources whereas poor farmers rely on other farmers and government extension staff for the same information. There is therefore, a great need to make information available to farmers at the right time and place. In response to this challenge, it is good to develop an integrated agricultural marketing information system that will be linked to Woreda information center, and to link them to government?s program. 2. Intervention to increase production and productivity of rice The quantity of rice produced at the farm level affected marketable supply of rice positively and significantly. However, farmers are working under limited plots of land by natural as well as socio-economic factors without using improved technologies and agricultural inputs. Rice producers in Fogera Woreda used little inputs (like improved seeds, pesticides and insecticides and modern technologies). Hence, increasing production and productivity of rice per unit area of land is better alternative to increase marketable supply of rice. Introduction of improved varieties, application of chemical fertilizers, using of modern technologies, controlling disease and pest practices should be promoted to increase production. 3. Facilitating extension services The results of the study indicates provision of extension service improve market participation of rice. Farmers have to linking production with marketing. And also it is good to enlightening farmers to produce based on market signals, consumer preferences and to direct or advice on the proper methods of handling, storing, transporting, and above all improving quality of rice. Hence, it is recommended to assign efficient extension system, updating the extension agent?s knowledge and skills with improved production and marketing system. 110? ? 4. Promoting education and trainings in production and marketing Changing the attitudes of farmers is a crucial factor in improving the marketing performance of households. If farmers have awareness about the benefit of the specialty market, they do not need only immediate economic advantages from the sale of their product. In case of production, household heads with very limited education encounter in successfully managing, fertilizer and pesticide applications, and also what to produce inline with taste and preference of consumers demand, especially in the presence of ineffective extension services. So stakeholders? and Agricultural and Rural Development Offices have to create awareness about the specialty of market. Continuous education and training on production and marketing will have a positive impact on their attitudes. 5. Promoting potentially collective organizations (cooperatives) Cooperatives are assumed to play important role in improving the bargaining position of the producers and creating, lowering transaction costs, reducing the level of oligopolistic market type by creating competitive market. 6. Improving the quality of rice Most attributes for rice is its quality. The Fogera rice has poor quality as compared to imported ones (Basmati, Ponte, and others types) both in kernel size and in color. This results from, its poor post harvest handling, spoilage during harvesting, hulling and threshing problems all together reduces the quality of rice in the market upon its selling price. Hence, especial attention should be given to improve quality so as to satisfy consumer?s desire, and farmer?s market price return. 7. Licensing the traders Traders should have license to operate at any level of trade, some of the traders have continued to operate with no license. Assemblers and brokers (though few) are with no 111? ? licensing. Also no clear demarcation of trading (fore instance, millers are acting as wholesaler). This has put the legal traders at a disadvantage when competing in the market. Therefore, public authorities in collaboration with representatives of traders should devise means of controlling those engaged in illegal trade. 8. Promoting family planning Family size is one of the significant demographic variables that affect volume of supply. With limited production, supporting a larger and extended family size would have been difficult for the farmers. This can be possible through the intervention of integrating family planning with health extension service and with respective concerned bodies. 112? ? 6. REFERENCES Abbott,J.C., and J.P.Makeham, 1981. Agricultural Economics and Marketing in the tropics. Abaye Akalu, 2007. Vegetable market chain analysis in Amhara National Regional State: the case of Fogera woreda, South Gondar zone. An M.Sc Thesis Presented to the School of Graduate Studies of Haramaya University. Admasu Shibru, 1998. Performance evaluation of coffee marketing in Sidama Zone. An M.Sc Thesis Presented to the School of Graduate Studies of Alemaya University. Agarwal D.R., 2001. Elementary athematics and statistics for Economist, Virinda publications (P) Ltd. Delhi. Ana, R.R, W. A. Masters and G. E. Shively, 2008. Linkages between Market Participation and Productivity: Results from a Multi-Country Farm Household Sample. Atteri, B. R., and G.Bisaria (2003). Marketable Surplus of Rice and Wheat and Benefits of Storage to the Farmers in India. Indian Agricultural marketing journal. Vol. 46 (1). Bain, J.S. ,1968. Industrial Organization, 2nd Edition, John Wiley, New York. Barnard C.S., J.S., Nix, 1979. Farm Planning and Control, Cambridge University Press, Cambridge. Belllemare, Marc F. and Barrett, Christopher B., 2005.An orderd Tobit model of market participation: Evidence from Kenya and Ethiopia. Evidence from Kenia AND Ethiopia, Working papers 14748, Cornell University, Department of Applied economics and management. Benfica, R., D. Tschirley and D. Boughton. 2006. Interlinked Transactions in Cash Cropping Economies: The Determinants of Farmer Participation and Performance in the Zambezi River Valley of Mozambique. Contributed paper presented at the 26 th International Association of Agricultural Economists Conference, Gold Coast, Australia, August 12-18, 2006. Berhanu Gebremedhin and D.Hoekstra, 2007. Cereal marketing and household market participation in Ethiopia: the case of teff, Wheat and rice. Proceeding of AAA conference. Addis Abeba, Ethiopia, 2007 (243-252), International livestock Research Institute (ILIR) Benfica, R., D. Tschirley, and L. Sambo. 2002. The Impact of Alternative Agro- Industrial Investments on Poverty Reduction in Rural Mozambique. MINAG/DE/DAP Research Report No. 51E. (http://www.aec.msu.edu/fs2/mozambique/wps51e.pdf) Black, J., 1997. Oxford dictionary of Economics, 2nd Edition, Oxford University press Inc., New York. Page. 331. 113? ? Black, W.E. and R.D. Knutson, 1985. Attitudes and opinions of Texas agricultural cooperative members. Report B-1483. Texas Agricultural Extension Service. USA. Blaylock, J. R. and Blisard, W. N., 1993. Women and the demand for alcohol: Estimating participation and consumption. Journal of Consumer Affairs. 27(2): 319-324. Branson R.E. and D.G. Norvell, 1983. Introduction to agricultural marketing, Mcgraw-Hill Inc, USA. Bull, T.A. 1988. Agro-ecological assessment of Ethiopian vertisols. PP.89-10. Proceedings of a conference on Management of Vertisols in sub-Saharan Africa. Addis Ababa, Ethiopia, 31 August-4 September, 1988. Chantal, P., Nielsen, 2002. Vietnam in the International Rice Market. A Review and Evaluation of Domestic and Foreign Rice Policies, Rapport 132. Chauhan,V.K and R. Singh, 2002. Marketed surplus of paddy- A regression analysis. National Level Quarterly Journal of Agricultural marketing. Vol. 45 (2). Christopher Winship, 1992. Models for sample selection bias. Department of Sociology, Northwestern University, 1810 Chicago Avenue Evanston, Illinois 6020. Chen, H., Siebenmorgen, T. and Griffin, K. 1998. Quality Characteristics of Long-Grain Rice Milled in Two Commercial Systems Journal of Cereal Chemistry. 75 (4):560-565. Central Statistics Authority (CSA, 2003). Statistical Report on Area and production of crops. part II-A. Addis Ababa, Ethiopia Central Statistical Authority (CSA, 2004). Ethiopia, Statistical Abstract, Addis Ababa Dawit Alemu, 2005. The status and challenges of agricultural marketing in Ethiopia. PP.1 Paper presented at a panel discussion organized by the Ethiopian Association of Agricultural Professionals (EAAP), April 22, 2005. Ethiopian Agricultural Research organization (EARO). Dejene Legese, 2008. Profitability of extension package inputs for Wheat and Barley in Ethiopia.the case of Limuna Bilbilo Wereda (Arsi Zone). Rural Development and environment in Ethiopia; prospect and challenges. Proceeding of the 10th Annual Conference of Agricultural Economics Society of Ethiopia, Addis Abeba, Ethiopia, may 2008. De Lucia, M. and Assennato, D., 1994. Agricultural engineering in development Post-harvest operations and management of food grains. FAO Agricultural Services Bulletin No. 93 Food and Agriculture Organization of the United Nations, Rome. Available online at http://www.fao.org/docrep/T0522E/T0522E00.htm. Accessed May 11, 2006. Doumach,W., C. Kodikara, and H. Maas,2004. Analysis of rice chain; towards sustainable (P) Rice, PP. 4-5, Report on an international workshop. Polonnaruwa, Sirlanka.11-3 March, 2004. 114? ? Duc Hai, L.T., 2003. The Organization of the Liberalized Rice Market in Vietnam. FAO, 2005.Addressing Marketing and processing constraints that inhibit Agri-food Experts: a guide for policy analysts and planners. FAO agricultural service Bulletin 160.Rome, Italy. FAO. 1996. Production year book, Vol.50.Rome FAO, 1994. African experience in the improvement of post-harvest techniques. Synthesis Based on the Workshop held in Accra, Ghana, 4-8 July 1994 Agricultural Support Systems Div. Food and Agriculture Organization of the United Nations, Rome Italy. Gauraha, A.K., K.N.S.Banafar, P.K.Verma, V.K. Choudhary, and B.C. Jain,(2002). Marketing Strategies of Rice in Chhattisgarh-A Case Study, 2002. Indian Journal of agricultural marketing. Vol. 45(3). Gebremeskel Dessalegn, T.S. Jayne and J.D. Shaffer, 1998. Market structure, conduct, and performance. PP.12 .Constraints on performance of Ethiopian grain markets. Working paper 8, Grain market research project Ministry of economic development and cooperation, January, 1998. Addis Ababa. Getachew Afework, 2000. Rice adaptation in Metema Woreda North Gondar one of the Amhara Regional State. Bureau of Agriculture, Bahir Dar. (Unpublished). Giles G.B., 1973. Marketing the English language book society and Macdonald, Evans Ltd. London. Goetz, S.J., 1992. A selectivity model of Household food marketing behaviors in sub-Saharan Africa. America Journal of Agricultural Economics. 74(2):444. Gujarati, D.N., 2003. Basic Econometrics. 4 th Edition. McGraw-Hill, New York. pp. 563-636 Harahap H,, 2003. Rice chain analysis in North Sumatra, Indonesia. Institute.(eds.). Rice improvement in Eastern, Central and Southern Africa. Proceedings of the International Rice Workshop at Lusaka, Zambia, April 9-19, 1984. Heckman, J.J., 1979. Sample selection bias as a specification error Journal of Econometrica. Vol. 47(1). Heckman?s correction, the free encyclopedia. htt://en.Wikipedia.org, 2008. Access on Nov. 25, 2008. Iddo K., Ayal K. and Zveleman, 2006. Farm out put, non-farm income and commercialization in rural Georgia, The electronic Journal of Agricultural and development Economics, Food and Agriculture Organization of the United States.Vol.3 (2): pages 276-286. IPMS, 2005. Fogera Wereda pilot learning Site Diagnosis and program Design. 115? ? Irini ,M., and Aysen Tanyeri-Abur, (2005).Transaction Costs, Institutions and Smallholder Market Integration: Potato Producers in Peru .Working Paper No. 05-04 Kinde Aysheshm, 2007. Sesame market chain analysis: the case of Metema Woreda, North Gondar zone, Amhara National Regional State. An M.Sc Thesis Presented to the School of Graduate Studies of Haramaya University. Kizito A., 2008. Famine early warning systems net work (FEWS NET) market Guidance No.2. Structure ?Conduct ?Performance and food security. Koch, J.V.180. Industrial organization and prices 2 nd Edition. London,UK; Prentice/hall International. Kohls, R.L. and J.N. Uhl, 1985. Marketing of Agricultural Product. Fifth Edition. McMillian Publishing Company, New York, USA. Kotler P., 2003. Marketing management. Eleven Editions. Pearson Education, Inc. USA. Lampkin N., Measures M (Eds) (1994, 2001). Organic Farm Management Handbook, Organic Farming Research Unit, Institute of Rural Studies, University of Wales. Organic Advisory Service, Elm Farm Research Centre. Aberystwyth, Newbury. Lawerence, D.S., 1992. Marketing and Agribusiness Development Paper No.1. Department of political Economy. University of Glasgow. Lewis, T. C., B. Wade Brorsen, Kim B. Anderson, EmilioTostao, 2008. Gender difference in marketing styles. Journal of Agricultural Economics. Vol. 38: page 1-7. Maddala, G.S., 2001.Introduction to Econometrics, third edition, John Wiley and Sons, Ltd, the Atrium, Southern Gate ,Chichester, West Sussex P019 8SQ, England. Makhura, T., 2001. Ph.D dissertation on overcoming transaction costs barriers to market Participation of smallholder farmers in the Northern Province of South Africa. University of Pretoria. June, 2001. Marcel, F. and R.V. HILL, 2005. Selling at the Farm gate or travelling to market. America Journal of Agricultural Economics. Vol. 87(3):717-734. Mark, L., M. V. Gottret, W. Cifuentes, C. F. Ostertag, R. Best, D. Peters and S. Ferris, 2004. Increasing the Competitiveness of Market chains for Smallholder producers Manual 3: Territorial Approach to Rural Agro-enterprise Development project. Menduoza, G., 1995. A primer on marketing channels and margins. Prices, Products and People: Analyzing Agricultural Markets in Developing Countries. Lynne Reinner Publishers, Boulder, London. 498p. 116? ? Ministry of Economic Development and Co-operation, 1999. Survey of Ethiopian Economy, Addis Ababa, Ethiopia Mohammed Jabbar, 2007. Agricultural Market Development in Ethiopia: Problems and Issues. International Livestock Research Institute. Moses, J., and Adebayo, E. F (2007). Efficiency of Factors Determining Rainfed Rice Production in Ganye Local Government Area, Adamawa State. Journal of Sustainable Development in Agriculture & Environment Vol. 3:20-30, Sept. 2007. Myint U. Kyaw, 2003. Agricultural Marketing Information System in Myanmar. Journal of agricultural marketing. Vol. 45(4): PP -24. Nombiro E., G. Hugode and O. Wills, 2001. Market structure and conduct of the hybrid maize seed industry a case study of the Trans NZOIA district in western Kenya. Nuru Adgaba, Eddosa Negara, Sebsibe Zuber, 2006. Profitability of processing crude honey. PP.133. Reversing Rural Poverty in Ethiopia; Dilemmas and critical issues. Proceedings of the 9th Annual conference of the agricultural Economics society of Ethiopia, Addis Abeba, Ethiopia, July 2006. Obasi, I.O., and E. Njpkuoma Chukwuma, 2008. Performance of rice market in Ebonyi State. Journal of economic theory. 2(1):22-23. Onwueme, I.C., and Sinha T.D., 1991. Field crop production in Tropical Africa. Technical Center for Agriculture and Rural Development (CTA), Wageningen, Netherlands. Osterteg, G., 2005. Identifying Market opportunities for Rural Smallholder Producers. Colombia. Pomery, R.S., and A.C Trinidad, 1995. Industrial organization and market Analysis: Fish marketing, U.S.A. Prasad, J. and A. Prasad, 1995. Indian Agricultural marketing Emerging trends and Prospective, Mital publications, New Delhi, India. Reardon, T., and C.P. Timmer. 2005. Transformation of Markets for Agricultural Output in Developing Countries Since 1950: How Has Thinking Changed? In R.E. Evenson, P. Pingali and T.P. Schultz (Eds). Handbook of Agricultural Economics Volume 3. Agricultural Development: Farmers, Farm Production and Farm Markets. Amsterdam: Elsevier. Rehima Musema, 2006. Analysis of red pepper marketing: the case of Alaba and Siltie in SNNPRS of Ethiopia. An M.Sc Thesis Presented to the School of Graduate Studies of Hararnaya University. 117? ? Saunders, R. and Betschart, A., 1979 in Inglett, GE and Charalambous, G (Eds) Tropical Foods: Chemistry and Nutrition Academic Press NY, USA. P. 191-216. Scott, G.J., 1995. Prices, Products and People: Analyzing Agricultural Markets in Developing Countries. Lynne Reinner Publishers, Boulder, London. 498p. Sewagegne Tariku, 2005. Status of Rice Improvement Research, Achievements, Major Constraints and Future Research Direction in the Amhara Region. PP.6-10. Proceedings of the National Workshop on the Status of Rice Research and Promotion in Ethiopia. Bahir-Dar, Ethiopia, 3-4, June 2005, Amhara Regional Agricultural Research Institute. Shahi, B.B., 1995. Potential rice varieties for East Africa. Proceedings of the International Rice Workshop on Rice improvement in Eastern, Central and Southern Africa. Lusaka, Zambia, 9-19 April, 1995. Shilpi, Forhad and Umali-Deininger, Dina, 2007 Where to sell? Market facilities and agricultural marketing. Policy research working paper series 4455, The World Bank. Singh, J.P., (2001). Strategy for effective agricultural Marketing extension to meet the Challenges in 21st century. Manage Extension Research Review July - December, 2001. Sirak,T., Bahta, S. Bauer, 2007. Analysis of the determinants of market participation within the South African small-scale livestock sector. Tropentag paper, October 9-11, 2007, Witzenhausen: "Utilisation of diversity in land use systems: Sustainable and organic approaches to meet human needs" Solomon Tilahun, 2004. Performance of Cattle Marketing System in Southern Ethiopia with Special Emphasis on Borena Zone. An M.Sc Thesis Presented to the School of Graduate Studies of Alemaya University. Stanton, Julie V., 2000. The roll of Agribusiness in development: replacing the Diminished Role of the Government in raising rural incomes, Journal of Agribusiness. Agricultural Economics Association of Georgia. Vol.18 (2). Taddese Yeneneh, 2005. Rice crop production and Extension package (Amharic version), Addis Abeba ,Ethiopia. Tareke Berhe, 2003. Rice: a high potential emergency and food security crop for Ethiopia Report no.2. Tenagne Kidane, 2005. Rice Processing and Utilization. Proceedings of the National Workshop on the Status of Rice Research and Promotion in Ethiopia. Bahir-Dar, Ethiopia, 3- 4, June 2005, Amhara Regional Agricultural Research Institute. Timmer, C.P. 1988. The Agricultural Transformation. In H. Chenery and T.N. Srinivasan, Eds., Handbook of Development Economics. Vol. 1. Amsterdam: North-Holland, 275-331. 118? ? Tshiunza, M., Lemchi, J., and Tenkouano, A. 2001. Determinants of Market Production of Cooking Banana in Nigeria. African Crop Science Journal. 9 (3): pp. 537-547. Tomek, W.G. and K.L. Rrobison, 1990. Agriculture product prices.3rd Edition. Cornel University Press. Ithaca and London. Verina, I., 2009. Magazine on low external input and sustainable agriculture (LEISA). June 2009.Volume 25, no.2 Wolday Amha, 1994. Food grain marketing development in Ethiopia after the market reform 1990: a case study of Alaba Sirarao district. Ph.D Dissertation. 1-Aufl-Berlin: Koster. Germany. 292p. Wolelaw Sendeku, 2005. Factors determining supply of rice: A study in Fogera district of Ethiopia. An M.Sc Thesis Presented to the School of Graduate Studies of Hararnay University. Zelalem Nega, 2008. Household poultry production and marketing: the case of Ada?a Wereda. An M.Sc Thesis Presented to the School of Graduate Studies of Haramaya University. 119? ? 7. APPENDICES 120? ? Appendix Table 1. Amount of land size and Land rent payment in birr Land size (ha) Amount in birr 2007/8 2008/9 0.0-0.5 20 40 0.6-1.0 25 55 1.1-1.5 30 75 1.6-2 35 100 2.1-2.5 40 130 2.6-3 45 170 Source: Fogera Woreda Trade and Industry Office, 2008/9 Appendix Table 2. Conversion factors to compute tropical livestock unit Source: Storck et al., 1991. Appendix Table 3. Conversion factor used to estimate man equivalent Labour category Sex Age ME Child M/F < 7 0 Child M/F 7-14 0.4 Adult M 15-64 1 Adult F 15-64 0.8 Elders M/F ? 65 0.5 Source: Bezabih, 2008/9. Farm management course Animal category TLU Calf 0.25 Weaned calf 0.34 Heifer 0.75 Cow or ox 1 Horse/mule 1.1 Donkey adult 0.7 Donkey young 0.35 Camel 1.25 Sheep or goat adult 0.13 Sheep or goat 0.06 Chicken 0.013 Bull 0.75 121? ? Appendix Table 4. Type, quantity produced and productivity of crops in 2007/8 Source: Survey result, 2008/9 Cultivated area ( ha) Quantity produced(q) Productivity(q/ha) Types of crops N Minimum Maximum Mean Std. Deviation N Minimum Maximum Mean Std. Deviation N Minimum Maximum Mean Std. Deviation Teff 89 0.06 1.5 0.36 0.25 87 0.15 10 2.3 1.97 86 0.6 24 7.14 5 Maize 144 0.03 1.5 0.31 0.19 142 0.5 40 5.99 5.64 142 4 112 19.96 13.87 Wheat 14 0.06 0.5 0.21 0.13 16 0.5 10 3.16 2.76 14 4 26.67 13.67 6.18 Barley 9 0.06 0.5 0.22 0.13 10 0.7 8 3.07 2.49 9 4 24 12.36 5.74 Chick pea 102 0.06 1.5 0.29 0.23 51 0.1 2 0.59 0.36 51 0.5 8 2.07 1.49 Lentil 15 0.03 0.38 0.19 0.1 15 0.3 4 1.32 1.01 15 1 1 1 0 F. Millet 92 0.06 1.25 0.31 0.2 91 0.5 25 4.44 4.01 91 3 50 14.28 8.23 Niger seed 5 0.06 0.5 0.26 0.16 6 0.5 8 2.92 2.76 5 4 16 8 4.9 Field pea 27 0.13 1.5 0.46 0.33 27 0.5 8 3.06 2.08 26 2 40 8.08 7.59 Grass pea 45 0.06 1 0.36 0.22 42 0.5 14 3.13 3.06 42 0.67 32 9.76 7.35 Tomato 16 0.06 0.25 0.15 0.08 13 3 12 8.06 3.22 13 12 96 62.22 24.24 Pepper 36 0.03 0.25 0.11 0.07 36 0.3 15 3.52 3.27 36 4.8 192 35.71 33.98 Onion 23 0.06 1 0.23 0.2 23 1 70 13.02 15.16 1 432 432 432 . Potato 3 0.03 0.13 0.07 0.05 2 5.5 7 6.25 1.06 2 56 88 72 22.63 Emmer wheat 24 0.04 0.63 0.21 0.13 24 1 12 3.81 2.69 24 8 96 21.62 17.98 Spice 7 0.06 0.75 0.24 0.24 8 1 5 1.63 1.38 7 1.33 24 10.19 8.47 Rice total 164 0.13 2 0.6 0.33 165 1 60 18.1 11.81 164 4 120 32.72 19.76 Own land 154 0.1 1.5 0.48 0.25 155 1 58 16 10.47 154 4 120 36.02 20.98 Rented-in 60 0.13 1.75 0.38 0.26 60 1 35 8.02 6.23 59 4 72 22.93 14.74 122? ? Appendix Table 5. ANOVA analysis of gross income, cost and profit among rice producer kebeles 2007/8 Sum of Squares df Mean Square F Sig. Gross income Between Groups 1066754920 3 3.56E+08 4.434 0.005 Within Groups 12750461515 159 80191582 Total 13817216434 162 Total cost Between Groups 21769895.08 3 7256632 0.447 0.72 Within Groups 2615873917 161 16247664 Total 2637643812 164 profit Between Groups 1149387028 3 3.83E+08 4.01 0.009 Within Groups 15384101960 161 95553428 Total 16533488988 164 Source: Owen computation, 2008/9 Appendix Table 6. Contingency table for dummy independent variables (CC) sex education level extension contact market information credit sex 1 0.005 0.053 0.009 0.087 education level 1 0.102 0.039 0.169 extension contact 1 0.203 0.132 market information 1 0.016 credit 1 Source: Owen computation, 2008/9 123? ? Appendix Table 7. Variance inflation (VIF) factor test Source: Survey result, 2008/9 Appendix Table 8. Market concentration of rice wholesalers. Amount of rice purchase Name wholsalers qt/month % share Rank 4 -firms Main Destinations 1 Tegegne Gizachew 2200 24.58 1st * Addis Abeba, Wollo 2 Habite Wolde Adamtie 1350 15.08 4th * Addis Abeba, Wollo 3 Hashim Hussien 1750 19.553 2nd * Addis Abeba, 4 Wokiel Ahimed 1650 18.44 3rd * Addis Abeba, 5 Mohamednur Hassen 1250 13.97 Addis Abeba 6 Tadesse Mihretie 750 8.38 Addis Abeba Total 8950 100 Concentration ratio (CR4 in %) 77.65 Source: Survey result, 2008/9 Collinearity Statistics Variables Tolerance(1/VIF) VIF(1-R 2 ) -1 AGE 0.633 1.58 FS 0.501 1.997 FL 0.525 1.905 MRD 0.908 1.102 TLS 0.533 1.875 TQP 0.511 1.957 OXN 0.354 2.822 NFINC 0.88 1.136 MRP 0.746 1.34 LMP 0.66 1.514 MS 0.73 1.37 TLU 0.313 3.191 124? ? Appendix Table 9. Rice miller?s sales list per product handled Name Qt/month % share rank four firms Main Destination 1 Ato Adamitie MengeshA 800 13.07 3 rd * AdisAbeba, BahirDar, Gondar,Wollo 2 Ato Kedir Ismael 1000 16.33 1 st * Adis Abeba ,Wollo 3 Ato Henok Getnet 470 7.67 AdisAbeba,Woldeya 4 Ato Adane Baye 600 9.80 Addis Abeba 5 Ato Takele Tesfaye 550 8.98 Addis Abeba 6 Ato Adigo Taye 600 9.80 4 th * Addis Abeba 7 Ato Abrarawu Ayal 900 14.70 2 nd * Addis Abeba 8 Ato Tsegawu Nibiret 450 7.35 Addis Abeba 9 Ato Selomon Mershaw 450 7.35 Addis Abeba 10 Ato Fekadu Teka 300 4.90 Addis Abeba Total 6120 100 Source: Survey result, 2008/9 Appendix Table 10. Market concentration of rice wholesalers and millers Market Name wholesalers Amount in qt/year % share Rank The 1 st 4- firms Woreta Henok Getnet 4160 2.54 5 th Woreta Tegegne Gizachew 76800 47.04 1st * Woreta Habitte Wold Adamitie 9600 5.88 3 rd * Woreta Adane Bayilie 2560 1.56 9 th Woreta Tsegaw Nibiret 1024 0.62 14 th Woreta Hashim Hussien 1600 0.98 11 th Woreta Solomon Mulusew 3200 1.96 6 th Woreta Adamtie Mengesha 1280 0.78 12 th Woreta Addis Ahimed 1280 0.78 13 th Woreta Adigo Taye 9600 5.88 3 rd * Woreta Kedir Esmaiel 3200 1.96 7 th Woreta Mohamednur Hassen 2560 1.56 10 th Woreta Takele Tesfaye 4800 2.94 4 th Woreta Kuhar Multi Purpose Coop 3200 1.96 8 th Woreta Zewdu Delalaw 38400 23.52 2 nd * Total sum 163264 100 Concentration ratio (CR 4 in %) 82.32% Source: Survey result 2008/9 125? ? Appendix Table 11. Wholesalers purchase sources. Monthly amount in quintal No Name of trader Farmers Rural assemblers Processors Total 1 Tegene Gizachewu 900 700 600 2200 2 Habitewold Adamite 450 450 450 1350 3 Hashim Hussien 750 700 300 1750 4 Wokeil 900 750 - 1650 5 Mehamed Nur Hassen 650 600 - 1250 6 Taddesse Mihiretie 450 300 - 750 Total 4100 3500 1350 8950 Source: Survey result, 2008/9 Appendix Table 12. Millers/processors purchase sources Name of trader Monthly purchase per quintal Monthly sale per quintal Far mers Rural Assem blers Proces sors Total AA Wollo Woldia Bahirdar, Gondar, Wollo Adamitie Mengesha 600 200 - 800 600 66.6 66.6 66.6 Kedir Ismael 600 400 - 1000 700 300 Henok Getnet 450 200 - 470 400 250 Adane Baye 300 300 - 600 600 Takele Tesfaye 400 150 - 550 550 Adigo Taye 400 200 - 600 600 Abrarawu Ayal 500 400 - 900 900 Tsegawu Nibiret 250 200 - 450 450 selomon Mersha 200 250 - 450 450 Fekadu Teka 150 150 - 300 300 Total 6120 5550 300 200 66.6 Source: Survey result, 2008/9 126? ? Appendix Table 13. Farmers? sampling distribution Characteristics of the rice produced kebeles Name of the rice produced Kebele Farming system population Distance from the main city Sample Selected 1 Woreta Zuria Low land 5475 Near 2 Kuhar Abo Lowland 6635 Near 3 Tiha Zekena Lowland 5632 Near 4 Shaga Lowland 7346 Middle 6 Shina Lowland 9743 middle 7 Nabega Lowland 10917 Very far 44 8 Wagetera Lowland 9556 Middle 9 Kidist Hanna Lowland 7333 Far 29 10 Kuhar Micheal Upland 6338 Near 38 11 Diba Upland 8422 Middle 54 12 Woji Upland 9670 Middle 13 Rib Gebireal Upland 7574 Far 14 Adis Betechristian Upland 9112 Far Total 165 Source: Survey result 2008/9 Appendix Table 14. Traders? sample Types of traders Population Sampled selected 1 Wholselares(grain) 9 6 2 Millers (grain) 26 10 3 Retailers(grain ) 3.1 Woreta 66 10 3.2 Bahir Dar 226 (39**) 21* 3.3 Gondar 251 29 4 Assmblers 4.1 Rural Assemblers 70 20 4.2 Urba Assemblers 5 5 4 Brokers 3 1 5 Urban distributers 10 5 *indicates one super market. and ** indicates licensed grain retailers. Source: Survey result 2008/9 127? ? Appendix Table 15. Producers selected administrative kebeles kuhar micheal Nabega Kidist Hana Diba Sifatira kebeles samle size kebeles sample size kebeles sample size kebeles sample size Ada beas 1 Abu Dir 5 Aba Dirok 1 Billa 6 Ada bet 8 Baboatie 4 Abaro 3 Deldalit 4 Ajafeji 1 Boakissa 1 Abir Degu 2 Diba 6 Aqua bet /warka mnder 6 Daga 1 Bursi 1 Fisashi 9 Baragie 8 Debir Mender 2 Bursie 3 Genet mender 8 Deqie micheal 2 Deqie Bet 1 Dingiz 1 Giedion 8 Luwalua 5 Fogerie bet 1 Dinjet 1 Gomibil 1 Messino 4 Fota 2 Gaba 1 Kiero mender 5 Nura mender 2 Girargie 4 Gaba Goti 1 Lahida 1 shiwenze 1 Kubaza 3 Girar 3 Shewana 1 Loha biet 1 Hudi Gebiya 3 Tachi Gulitochi 1 Luabit 1 Kidist Hana 4 Tinish Terara 4 Rieq 13 Maje 1 Rieq Fota 1 Tseyo Sariqo 3 Yemushira Dingay 2 Tigrie mender 1 Zifnie 1 sum 38 44 29 54 Source: Survey result 2008/9 128? ? Appendix Table 16. Cultivated area of crops in upland and low land rice production system Upland rice production system Low land rice production system Total Types of crop N Mean Std. Deviation N Mean Std. Deviation N Mean Std. Deviation Teff 52 0.35 0.23 37 0.37 0.29 89 0.36 0.25 Maize 79 0.27 0.13 65 0.36 0.24 144 0.31 0.19 Wheat 6 0.19 0.08 8 0.23 0.16 14 0.21 0.13 Barley 7 0.23 0.13 2 0.16 0.13 9 0.22 0.13 Chick pea 56 0.32 0.24 46 0.26 0.21 102 0.29 0.23 Lentil 1 0.25 . 14 0.19 0.1 15 0.19 0.1 Niger seed 5 0.26 0.16 5 0.26 0.16 Tomato 14 0.16 0.08 2 0.09 0.04 16 0.15 0.08 Onion 15 0.22 0.1 8 0.27 0.33 23 0.23 0.2 Finger millet 71 0.35 0.21 21 0.18 0.09 92 0.31 0.2 Field pea 17 0.41 0.27 10 0.54 0.42 27 0.46 0.33 Grass pea 36 0.35 0.2 9 0.42 0.27 45 0.36 0.22 Pepper 34 0.11 0.07 2 0.07 0.01 36 0.11 0.07 potato 2 0.04 0.03 1 0.13 3 0.07 0.05 E. wheat 18 0.2 0.14 6 0.23 0.05 24 0.21 0.13 Spice 3 0.13 0.11 4 0.33 0.29 7 0.24 0.24 Rice 92 0.49 0.27 72 0.74 0.33 164 0.6 0.33 Source: Survey result, 2008/9 Appendix Table 17. Farmers sample selection Name of PAs Number of Households No. Sample Size Kuhar Micheal 1506 38 Diba Sifatira 1266 54 Nabega 1779 44 Kidist Hanna 1151 29 Total 6602 165 129? ? Appendix Table 18. Rice production, area and number of participant farmers by Woreda and region /2006-2008/ Source: Zewdie G/Tsadik, 2009 (unpublished) Region Woreda /site 2006 2007 2008 No. of farmers Size (ha) No. of farmers Size (ha) Production (ton) No. of farmers Size (ha) Production (ton) Amhar Region Metema 351 117 3840 1280 3840 9500 2500 9250 Fogera 23616 7872 46800 15600 39000 116000 29000 81200 Libo-kemkem 12567 4189 27600 9200 18400 48800 12200 28060 Dera 8148 2716 15000 5000 10000 29380 7345 16159 Sekela 1338 446 2700 900 1800 6400 1600 4480 Achefer 208 52 360 120 240 1360 340 986 Sub-total-1 46228 15392 96300 32100 73280 211440 52985 140135 Tigray Region L/koraro 2880 720 1800 Tsegede 492 217 651 Tselemt 228 334 835 Welqayit Humera Sub-total-2 3600 1271 3286 Benshangul Gumz Bambasi 688 172 516 Kurmuk 786 190 665 Sub-total-3 1474 362 1181 Oromiya Region Chewaqa 740 185 5400 1800 6300 10248 2928 11126 Dedessa 859 359 2085 695 2085 4740 1185 3555 Borecha 291 77 960 320 800 3000 75 2850 Bedelle 126 60 345 115 230 1520 380 1064 Darimu 45 2 75 25 50 248 62 143 Shebe 2280 570 1938 Sub-total-4 2061 683 8865 2955 9465 22036 5200 20676 Somali Region Gode 70 15 5940 1980 5940 1734 3120 10920 Kelafo 80 13 7650 2550 7875 3420 6800 27200 Sub-total-5 150 28 13590 4530 13815 5154 9920 38120 Southern Region Yeki 150 75 450 150 300 1020 255 765 Boreda 100 50 336 112 224 1000 250 750 Gura-ferda 4515 2257 30000 10000 25000 12857 18000 75600 Gimbo 68 34 288 96 192 804 201 563 Shashego 12 3 18 6 12 24 6 16 Misha 18 4.5 21 7 14 36 9 29 Jinka/Bilate Sub-total-6 4,863 2,424 31,113 10,371 25,742 15,741 18,721 77,723 Gambella region Gambella 240 479 1,533 Abobo 417 835 2,923 Sub-total-7 657 1314 4,456 Grand Total (1-7) 53,902 18,527 149,868 48,966 122,302 260,328 90,547 285,924 130? ? Figure 3. Rice Production trend in Fogera Woreda in years Source: Computed from using data on Fogera Woreda Agricultural and Rural Development Office, 2008 131? ? Figure 4. National rice production trend (2007-2009) Source: Zewdie G/Tsadik, 2009 (Unpublished) 132? ? Figure 5.Trends in the amount of commercial rice import (1999 ?2008) Source: Ethiopian Customs and Revenue Agency for imports, estimated using data from Zewdie G/Tsadik, 2009 (Unpublished) 133? ? Figure 6. Distribution of rice potential areas in Ethiopia Source: Report of NRDS Draft, 2009