THE ECONOMIC IMPACT OF PUBLIC INVESTMENTS IN MAIZE RESEARCH AND EXTENSION IN GHANA, 1979-97. A THESIS SUBMITTED TO THE DEPARTMENT OF AGRICULTURAL ECONOMY AND FARM MANAGEMENT, UNIVERSITY OF GHANA, IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE AWARD OF DOCTOR OF PHILOSOPHY DEGREE IN AGRICULTURAL ECONOMICS SEPTEMBER, 1999. University of Ghana http://ugspace.ug.edu.gh (X ’MUU Ssw-E»$J)22 University of Ghana http://ugspace.ug.edu.gh DECLARATION I do hereby declare that, except for references cited which have been duly acknowledged, this work presented in this thesis was carried out by me as a student at the Department of Agricultural Economics, University of Ghana, Legon. This work has never been presented anywhere either in part or whole for the award of any degree. Awere Ansong Dankyi (Student) This work has been submitted for examination with our approval as supervisors. Dr. E. Kweku Andah Agricultural Economics Department (Major Supervisor) K Yerfi Fosu Agricultural Economics Department (Member of Supervisory Committee) Or. Yaw Asante Economics Department (Member of Supervisory Committee) Dr. Felix Y.M. Fiadjoe Agricultural Extension Department (Member of Supervisory Committee) University of Ghana http://ugspace.ug.edu.gh TABLE OF CONTENTS LIST OF TABLES.................................................................................................. vii LIST OF FIGURES................................................................................................. x LIST OF ACRONYMS........................................................................................... xi ACKNOWLEDGMENTS....................................................................................... xiii ABSTRACT............................................................................................................ xiv CHAPTER: 1 INTRODUCTION...................................................................................... 1 The Problem .......................................................................... 5 Objectives of the Study.................................................................. 7 Justification of the Study............................................................... 9 Limitations of the Study.................................................................. 10 Organisation of the Study................................................................. 11 2 LITERATURE REVIEW............................................................................ 12 Impact Assessment.......................................................................... 12 Impact Assessment Methods............................................................ 12 The Economic Surplus Method of Ex-post Evaluation.................... 15 Role of Economic Policy on Returns to Agricultural Research 20 Adoption and Difliision................................................................... 22 Determinants of Adoption............................................................... 24 Chapter Summary........................................................................... 26 3 MAIZE TECHNOLOGY DEVELOPMENT IN GHANA.......................... 28 Agricultural Research in Ghana...................................................... 28 The Ghana Grains Development Project.......................................... 29 Other Collaborating Institutions............................... L ................... 31 Maize Varieties.......................................................... \ .................. 34 Planting Densities........................................................................... 40 Row Planting................................................................................... 41 Fertilizer Application...................................................................... 42 The Maize Recommendations.......................................................... 44 Chapter Summary............................................................................ 45 4 THE PRODUCTION ENVIRONMENT.................................................... 46 Ecological Zones.............................................................................. 46 Rainfall .................................................................................. 48 Inputs............................................................................................... 51 Seed Production and Distribution........................................ 51 Fertilizer.............................................................................. 54 Marketing and Distribution .............................. 55 Trends in Fertilizer Imports.................................................. 55 Trends in Fertilizer Prices................................................... 57 Fertilizer and Maize Prices.................................................. 58 Page iii University of Ghana http://ugspace.ug.edu.gh Chapter Summary............................................................................ 61 5 RESEARCH METHODOLOGY................................................................. 62 Data Collection Activities............................................................... 62 The Survey Instrument.................................................................... 66 Economics of Maize Production...................................................... 66 Determinants of Adoption of Improved Maize................................ 66 Returns to Investment ............................................................. 67 Key Assumptions and Critical Parameters...................................... 71 Baseline Scenario............................................................................ 71 Alternative Scenarios...................................................................... 73 Scenario 2 (higher research cost)......................................... 73 Scenario 3 (higher extension cost)...................................... 74 Scenario 4 (higher research cost and higher extension cost) 74 Scenario 5 (lower projected research benefits).................... 74 Chapter Summary........................................................................... 75 6 THE ADOPTION OF MAIZE TECHNOLOGIES IN GHANA................. 76 Demographic Characteristics of the Farmers.................................. 76 Adoption of Maize Technologies..................................................... 77 Explaining Adoption....................................................................... 89 Extension and Farmers.............................. 89 Characteristics of Farmers................................................... 91 Other Factors Associated with the use of Improved Maize Technologies................................................................................... 92 Improved Variety................................................................ 92 Row Planting....................................................................... 93 Fertilizer.............................................................................. 94 Maize Utilisation and Storage......................................................... 96 Gender................................... 98 Decision-making in Maize Farming Activities................................ 99 Impacts of Improved Maize Technologies...................................... 100 Incomes of Farmers............................................................. 101 Nutrition.............................................................................. 103 Gender Effects..................................................................... 104 Productivity.......................................................................... 104 Factors Explaining the Adoption Pattern between the Ecological Zones............................................................................. 107 Population........................................................................... 108 Cropping Intensity............................................................... 109 Markets and Commercial Orientation.................................. 112 Economics of Maize Technology.................................................... 112 Risks.................................................................................... 115 Marginal Rate of Return to Maize technology as a Package. 118 CHAPTER Page iv University of Ghana http://ugspace.ug.edu.gh CHAPTER Page Logistic Regression Analysis.......................................................... 122 TOBIT Analysis.............................................................................. 126 Chapter Summary............................................................................ 129 7 RETURNS TO INVESTMENTS MAIZE RESEARCH AND EXTENSION 130 Scenario 1 (Baseline)...................................................................... 131 Scenario 2 (higher research cost)..................................................... 137 Scenario 3 (higher extension cost).................................................. 138 Scenario 4 (higher research cost and higher extension cost) 138 Scenario 5 (lower projected research benefits)................................ 139 Chapter Summary............................................................................ 139 8 POLICY IMPLICATIONS AND CONCLUSIONS.................................... 140 Policy Implications.......................................................................... 140 Research Policy Issues.................................................................... 140 Extension Policy Issues................................................................... 145 Input Policy Issues.......................................................................... 146 Concluding Remarks....................................................................... 147 BIBLIOGRAPHY................................................................................................... 149 APPENDIX 1: Maize Survey Questionnaire.......................................................... 155 2: Theoretical Model for Measuring Net Social Gains...................... 168 3: Empirical Parameter Estimates...................................................... 175 4: Derivation of Key Formulae for Net Social Gains and internal Rate of Return................................................................................. 184 5: The Logistic Regression and the TOBIT Model............................. 195 6: Logistic Growth Curve................................................................... 200 7: Financial Rate of Return (Baseline scenario).................................. 206 8: Economic Rate of Return (Baseline scenario)................................ 215 9: Financial Rate of Return with Higher Research Cost..................... 224 10: Economic Rate of Return with Higher Research Cost.................... 232 11: Financial Rate of Return with Higher Extension Cost ........ 240 12: Economic Rate of Return with Higher Extension Cost................... 248 13: Financial Rate of Return with Higher Research and Higher Extension Cost............................................................................... 257 14: Economic Rate of Return with Higher Research and Higher Extension Cost............................................................................... 265 15: Financial Rate of Return with Lower Projected Research Benefits.......................................................................................... 273 16: Economic Rate of Return with Lower Projected Research Benefits.......................................................................................... 281 17: Area Planted to Specific Improved Maize Varieties in the Ecological Zones of Ghana............................................................ 290 University of Ghana http://ugspace.ug.edu.gh APPENDIX 18: Principal Month for Selling Maize in the Ecological Zones of Ghana....................................................................................... University of Ghana http://ugspace.ug.edu.gh LIST OF TABLES 1.1 The Use of Improved Varieties over the Years............................................ 3 1.2 Comparison of Counterpart Funds Requested and Released in Ghana for Agricultural Research, 1992 to 1996 (in ‘000 cedis).................... 6 1.3 Number of Ghanaian Researchers and Person Years in Commodity Research, 1988............................................................................................ 10 3.1 Maize Varieties Developed and Released by CRI, 1954-1996..................... 32 3.2 The Sources of Released Maize Varieties in Ghana, 1979-1996.................. 33 3.3 Average Grain Yield of Full Season Maize Varieties, 1979-84 (in t/ha) 35 3.4 Average Grain Yield of Maize Varieties, 1981-84 (in t/ha)......................... 36 3.5 Average Grain Yield of Early Maturing Maize Varieties, 1983-86 (in t/ha).......................................................................................... 36 3.6 Streak resistance levels of some selected varieties........................................ 37 3.7 Grain Yield of Selected Medium Maturing Maize Varieties (in t/ha) 38 3.8 Protein and Tryptophan Contents of Obatanpa and Okomasa...................... 38 3.9 Grain Yield and other Parameters of Selected Maize Varieties (in t/ha) 39 3.10 Grain Yield of Local and Improved Maize Varieties (in t/ha)..................... 40 3.11 Grain Yield of Maize Varieties under Different Densities (t/ha).................. 40 3.12 Selected Characters for La posta at Three Densities in Ghana.................... 41 3.13 Effect of Time, Placement and Type of Fertilizer on Maize Yield (in t/ha)........................................................................................................ 43 3.14 Maize Grain Yield from Farmer-Managed Fertilizer Trials (in t/ha) 44 3.15 Demonstrated Technological Options for Maize Production in Ghana 45 4.1 Comparison of Mean Monthly Rainfall for 1967-1978 and 1979-1880 Periods for some Selected Districts in Ghana............................ 50 4.2 Inproved maize seed growers and dealers in Ghana since 1990.................. 52 4.3 Certified Seed Sales in Ghana (in mt)........................................................... 53 4.4 Fertilizer Imports since 1970 (in mt)............................................................ 56 4.5 Nominal Fertilizer Prices and Subsidy Rates in Ghana................................. 57 4.6 Fertilizer Prices and Subsidies..................................................................... 59 5.1 Sampling Procedure for Maize Technology Survey in Ghana..................... 63 5.2 Location of Survey Districts....................................................................... 64 5.3 Exchange Rates, 1979 to 1997................................................................... 70 5.4 Research and Extension Cost Scenarios....................................................... 75 6.1 Demographic Characteristics of Maize Farmers, Ghana.............................. 76 6.2 Access to Infrastructure by Survey Household............................................ 77 6.3 Adoption of Maize Technologies in Ghana, 1997......................................... 78 6.4 Farmers Stopping Maize Technology Use................................................... 78 6.5 Proportion of Area under Maize Technology Adoption.............................. 79 6.6 Area Planted to Specific Maize Varieties.................................................... 80 6.7 Interaction among Maize Technologies in Ghana, 1997....................... 80 Table: Page vii University of Ghana http://ugspace.ug.edu.gh 6.8 Adoption of Improved Varieties in the Political Regions of Ghana............. 81 6.9 Adoption of Row Planting in the Political Regions of Ghana ............ 84 6.10 Adoption of Fertilizer in the Political Regions of Ghana............................ 84 6.11 Extension Contacts of Adopters and Non-Adopters of Maize Technologies 90 6.12 Level of Education of Farmer and Adoption of Maize Technologies 92 6.13 Fanners’ Age and Adoption of Improved maize technologies..................... 92 6.14 Factors Associated with Improved Varieties............................................... 93 6.15 Factors Associated with Adoption of Row Planting.................................... 93 6.16 Factors Associated with Fertilizer Adoption................................................ 95 6.17 Household Source of Income from Maize................................................. 97 6.18 Principal Month for Selling Maize............................................................... 97 6.19 Maize Technology Adoption among Male and Female Farmers.................. 98 6.20 Male and Female Adopters oflmproved Maize Technology....................... 99 6.21 Gender and Decision-Making on Maize Farming Activities....................... 100 6.22 Use of Increased Income from Maize. ....................................................... 103 6.23 Descriptive Statistics of Farmers’ Estimates of Maize Harvests (in maxibags)............................................................................................... 106 6.24 Estimated Yield Increase Attributable to Improved Varieties and Fertilizer....................................................................................................... 106 6.25 Fanners’ Formal Education Level and Household Size............................... 109 6.26 Years of Continuous Maize Cropping of Major Maize Field....................... 110 6.27 Average Fallow Periods............................................................................... I l l 6.28 Economic Indices for Estimating Marginal Rate of Return to Maize Technologies............................................................................................... 113 6.29 Marginal Return Analysis of Adoption oflmproved Maize in the Ecological Zones......................................................................................... 114 6.30 Effect of Yield Reduction in Improved Maize on the Marginal Rate of Return (%)................................................................................................... 116 6.31 Effect of Lower Maize Grain Price on the Marginal Rate of return on Improved Maize (%)........................................................................... 116 6.32 Marginal Return Analysis On the Adoption of Fertilizer on Maize Production in the Ecological Zones.......................................................... 117 6.33 Effect of Yield Reduction and Increase in Fertilizer Price on the Marginal Rate of Return (%)....................................................................... 118 6.34 Marginal Rate of Return to Adopting Improved Maize, Fertilizer and Row Planting as a Package.................................................................................. 120 6.35 Changes in Yields on Marginal Rate of Return to Improved Maize Technology (%)................................................................................ 121 6.36 Marginal Rate of Return to Changes in the Price of Maize (%).................. 122 6.37 Definitions of Variables in the Logistic Model............................................ 123 6.38 Classification Table for MVUSER............................................................... 123 6.39 Indicator Variables for Factors of Adoption of Improved Maize Varieties in Ghana....................................................................................... 125 Table Page viii University of Ghana http://ugspace.ug.edu.gh Table: Page 6.40 Correlation Matrix of Variables in the Logistic Model................................. 125 6.41 Definitions of the Variables in the TOBIT Model......................................... 126 6.42 Estimated Results of Factors Affecting Area under Improved Maize Production using Farmer, Farm Specific Variables and Ecological Zones, Ghana............................................................................................... 127 6.43 TOBIT Marginal Effect for Factors of Maize Technology Adoption 128 7.1 Key Indicators Used in Calculating Internal Rate of Return......................... 130 7.2 Financial and Economic Rate of Return to Maize Research in Ghana 131 7.3 Social Gains and Financial Rate of Return................................................... 133 7.4 Economic and Financial Net Social Gains................................................... 135 ix University of Ghana http://ugspace.ug.edu.gh LIST OF FIGURES 1.1 Maize production indicators, Ghana, 1975-1997......................................... 5 2.1 Economic Surplus with Linear Demand and Supply Curves....................... 17 2.2 Benefit-Cost Approach................................................................................. 18 2.3 Akino-Hayami Approach............................................................................. 19 4.1 Ecological Zones of Ghana........................................................................... 47 4.2 Nitrogen-to-Maize Price Ratio, Ghana (1979-1997)..................................... 59 4.3 Kilograms of Ammonium Sulphate that 1kg of Maize can buy (1970-1996)....................................................... 60 5.1 Location of Survey Districts, Ghana Maize Technology Adoption Study... 65 5.2 Consumer and Producer Surplus................................................................. 68 6.1 Regional and District Boundaries Showing Adoption Rates for Improved Maize Varieties in the Regions (%)............................................. 83 6.2 Regional and District Boundaries Showing Adoption Rates for Row planting in the Regions (%)................................................................ 85 6.3 Regional and District Boundaries Showing Adoption Rates for Fertilizer in the Regions (%)....................................................................... 86 6.4 First Year Farmer Used Improved Varieties................................................ 88 6.5 First Year Farmer Used Fertilizer................................................................. 88 6.6 First Year Farmer Planted Maize in Rows................................................... 89 6.7 Source oflmproved Seed in 1997 and in Previous Years............................. 91 6.8 Percent of Farmers Indicating Maize as a First Source of Income................ 96 6.9 Percent of Farmers with Increases in Maize Production Indicators.............. 102 7.1 Real (Farmgate) Domestic and World of Maize......................................... 136 A2.1 Shift and Movement in Production Functions.............................................. 171 A2.2 Production Increases, Supply Shift and Input Expenditure........................... 173 A6.1 Logistic Regression....................................................................................... 204 A6.2 Logistic Growth Curve................................................................................. 205 Figure: Page X University of Ghana http://ugspace.ug.edu.gh ABBREVIATIONS AND ACRONYMS CIDA Canadian International Development Agency CIMMYT International Maize and Wheat Improvement Centre CN$ Canadian dollar CPI Consumer price index CRI Crops Research Institute CSIR Council for Scientific and Industrial Research EA Enumeration area FOB Freight on Board GGDP Ghana Grains Development Project GLDB Grains and Legumes Development Board GSIU Ghana Seed Inspection Unit IFAD International Fund for Agricultural Development UTA International Institute for Tropical Agriculture IRR Internal Rate of Return ISNAR International Service for National Agricultural Research MOFA Ministry of Food and Agriculture MRR Marginal rate of return NARP National Agricultural Research Project NARS National Agricultural Research Systems NGO Non-governmental organization NS Not significant NSS National Seed Service University of Ghana http://ugspace.ug.edu.gh ODI Overseas Development Institute OFY Operation Feed Yourself OFYI Operation Feed Your Industry PPMED Policy, Planning, Monitoring and Evaluation Division RER Real exchange rate SARI Savannah Agricultural Research Institute SG 2000 Sasakawa Global 2000 US United States of America USAID United States Agency for International Development USDA United States Department of Agriculture University of Ghana http://ugspace.ug.edu.gh ACKNOWLEDGMENTS Many people have helped in the development of this research. My major supervisor, Dr. E. Kweku Andah, deserves special recognition for his untiring guidance from the initial stage through several painful editing to the final copy. I thank my other members of the thesis committee: K Yerfi Fosu, Dr. Yaw Asante and Dr. Felix Y. M. Fiadjoe for their immense suggestions and time spent on the thesis. I owe a gratitude to Dr. (Mrs.) Ramatu Al-hassan for suggestions at the initial stage of my research. I likewise wish to make mention of Akwasi Mensah-Bonsu, a teaching assistant of the Department for reading through some of the drafts and making useful comments. I am very grateful to Dr. Michael L. Morris of CIMMYT, Mexico, who was very instrumental in obtaining funding for my visit to CIMMYT as a visiting scientist and for his numerous suggestions and brotherly reception. I also thank the staff of CIMMYT Economics Program especially Dr. Prabbu Pingali, the Director, for hosting me. My appreciation is due to E. A Addison, former Director of the Crops Research Institute for nominating me for the National Agricultural Research Project (NARP) award. My research would not have been possible without the financial support from the NARP. Ithank all the staff of the NARP secretariat. I am thankful to Professor John H. Sanders and Dr. Ahmed Mohamed for their suggestions and for playing host to me at Purdue University, USA. To the staff of the Department of Agricultural Economics, Purdue University, I am grateftil for their moral support. For many others whose names I cannot mention, I say thank you very much. Finally, special acknowledgment needs be given to my family for their patience and moral support. xiii University of Ghana http://ugspace.ug.edu.gh ABSTRACT Dankyi, Awere Ansong. PhD., University of Ghana. June 1999. The Economic Impact of Public Investments in Maize Research and Extension in Ghana, 1979-97. Major Supervisor; E. Kweku Andah Maize is the most important cereal in Ghana grown by some 1.74 million households and contributing 67 billion cedis or 14 million dollars to Ghana as at the 1991-92 crop season. Maize is eaten in various forms in most homes. In the late seventies, maize production declined. The Ghana Grains Development Project was established in 1979 to provide research and extension services for maize production in Ghana. Since the inception of the project, maize production has been increasing. These increases in the past two decades have been attributed to the availability and adoption of GGDP-generated improved maize technologies. This study measures the degree to which maize recommendations are taken up by fanners and determines the factors driving the adoption process. It assesses the economics of maize recommendations and analyzes the returns to public investments in Ghana’s research and extension activities for maize between 1979 and 1997. The study also makes suggestions for future maize research and extension work in Ghana. It is observed that fifty-four percent of the total area under maize production is cultivated to improved maize varieties and 55 percent of the area is row planted with maize. Only 26 percent of the area are fertilized. Furthermore, the main factors driving adoption of improved varieties are extension contact, farming maize in the coastal savannah zone, transition zone, Guinea Savannah zone, gender and total maize area planted. The financial rate of return on investment in maize research and extension is estimated between 50 and 79 percent and the economic rate of return is 25 to 33 percent. This is a clear indication that maize research and extension activities between 1979 and 1997 have been profitable and have contributed to national food security. This has a number of policy implications for research, extension, input delivery system, agricultural policy makers and donors. Funding, germplasm improvement, crop management recommendations, need for speeding up fertilizer technology dissemination and efficient input supply systems are some of the policy issues needing attention. xiv University of Ghana http://ugspace.ug.edu.gh CHAPTER 1 INTRODUCTION Background Agriculture is the largest and most important sector in the Ghanaian economy. The rate of economic development is closely associated with the performance of the agricultural sector. Agriculture, including forestry, accounts for about 55% of the Gross Domestic Product (GDP) and also employs about 59% of the labour force.1 The main activity in the rural areas is agriculture and about 70% of the population live in these areas. Poverty is basically a rural phenomenon and about 84% of the National hard-core poverty in the country in 1988 were from the rural areas2. Therefore what affects the rural folks has a direct impact on agriculture and consequently on the economy as a whole. Any improvement in agriculture is also likely to improve the welfare of the majority of the rural folks in particular and the whole population in general. The role of agricultural research in many countries such as Pakistan, Nepal and Japan has been well documented.3 These studies have shown that there are high rates of return to agricultural research investments. By 1997 there were a few of such studies in i Asenso Okyere,W., F. A Asante, L. Oware Gyekye,"Case Studies on Rural Poverty Alleviation in the Commonwealth. Ghana.” Food Production and Rural Development Division. Commonwealth Secretariat, London. (1992) pp 1-3. Boateng, E.O., K. Ewusi, R Kanbur, A McKay. "A Poverty Profile for Ghana, 1987-1988. Social Dimensions o f Adjustment in Sub-Sahara Africa". Working Paper No.5. The World Bank, Washington D. C. (1990). Byerlee, D,, and P. Moya."Impacts of International Wheat Breeding Research in the Developing World, 1966-90". MexicoD.F.:CIMMYT. (1993) pp 11-50. Morris, M.L., H J. Dubin and T. Pokhred. "Returns to Wheat Research in Nepal". CIMMYT Economics Working Paper 92-04. Mexico, D.F.: CIMMYT. (1992) pp 15-31 Akino Masakatsu and Yujiro Hayami. "Efficiency and Equity in Public Research: Rice Breeding in Japan's Economic Development" American Journal o f Agricultural Rcnnnmir.s Vol.57. (1975) pp 1-10. 1 University of Ghana http://ugspace.ug.edu.gh Africa4. The results of the 34 recent studies measuring the impact of agricultural research in Africa indicate that about 70% of these studies estimated a rate of return of more than 20% per year. These high rates of return indicate substantial contribution of agricultural research to national income and economic growth, thus justifying further investments in research in this area. The Ghana Grains Development Project (GGDP) of the Crops Research Institute (CRI) has been in existence since 1979. Through a systematic research process it has developed improved maize, cowpea and soybean varieties and disseminated recommendations for growing these crops in the country. Before 1979, some of the improved maize varieties available were the Composites, Diacol 153 and La posta. After 1979, several improved varieties were developed and made available to farmers throughout Ghana. The performance of the developed improved varieties in on-farm trials was better than the older or local materials. For example, over the 1979-85 period, the average yield of La posta, CRI an improved variety, was 4.3 tons/ha as against a local variety yield of 2.3 tons/ha in the same period - nearly a 100% yield increase5 (GGDP Annual Reports 1979-85). Given the sizeable yield gains and favourable production environment, it is not surprising that there is a dramatic increase in maize production and higher incomes for farmers who have adopted the technology.6 Oehmke, James F., and Eric Crawford.. "The Impact o f Agricultural Technology in Sub-Saharan Africa: A Synthesis o f Symposium Findings". MSU International Paper No. 14. East Lansing: Department of Agricultural Economics, Michigan State University. (1993) pp 2-12 ; Masters W.A., Touba Bedingar and James F. Oehmke. "The Impact o f Agricultural Research in Africa: Aggregate and Case Study Evidence". Paper presented at the 23rd International Conference o f Agricultural Economists. August 10-16,1997, Sacramento, California. (1997). Whole Document. Estimated by the author from data in various GGDP Annual Reports, 1979 to 1985. Dankyi, A A , V.M. Ancbinnah, AO . Apau.. " Profitability and Impact o f Extension Test Plot on Maize Production in Ghana". GGDP/SG2000/DAES Collaboration. GGDP Mimeo. (1995) pp 21-29. 2 University of Ghana http://ugspace.ug.edu.gh Farmers are generally cautious in using recommendations and they normally will test them gradually. This is a "step-wise" adoption or partial adoption. Farmers will continue to use them once the recommendations are appropriate. A 1987 survey7 presented in Table 1.1 reveals that a smaller proportion of farmers (44%) who were introduced to improved varieties in 1986 used them whereas for those introduced earlier (1982), a higher percentage (76%) continued to use the improved varieties. This suggests that when a new variety is introduced to formers, the rate of use of the variety increases over time leading to possible increases in production. Table 1.1: The Use oflmproved Varieties in Ghana over the Years First year to use improved varieties Farmers who plant at least half their maize area to improved varieties (%) 1986 44 1985 60 1984 69 1983 75 1982 or earlier 76 Source: Tripp R. et aL GGDP 1987, p. 18. A number of surveys conducted by GGDP have shown high rates of adoption of maize recommendations. For example, in 1984, the adoption rate of improved varieties in the transition ecology zone was 53%. In 1986, the adoption in the same zone was 81%8. In Ghana, the amount of money released by the government to finance agricultural research always falls short of the amount requested For example, between 1993 and 1996, only 12-48% of the counterpart funds requested for agricultural research was released. Nevertheless, the potential high yields of improved varieties among others Tripp,R., K. Marfo, A A Dankyi, and M. Read. "Changing Maize Production Practices o f Small-scale Farmers in the Brong-Ahafo Region, Ghana. Kumasi: Ghana Grains Development Project (1987) pp 17-19. 8 Tripp R , K. Marfo, AA . Dankyi and M Read. Op. Cit pp 17-19. 3 University of Ghana http://ugspace.ug.edu.gh ensured a steady flow of resources to maize research from the Governments of Canada and Ghana. Since limited resources are used to support research in agriculture, it is necessary to show evidence of the returns to public investments in maize research in Ghana to attract more funds. Therefore, it is important to establish whether past maize research has paid off and whether this pay-off is sustainable Maize is the most important cereal crop in Ghana. It is grown throughout the country, except in a small portion of the North-eastern part characterised by harsh climatic conditions and rocky soils. Although official production statistics on maize are less accurate than desired, general trends can be established in the pattern of maize production in Ghana (Figure 1.1). Following an extended decline, the area planted to maize and maize yields began to increase in mid-1980s. From 1975 to 1997, maize area grew at an average rate of 2.9% per year, while maize yields grew at an average rate of 2.3% per year. This increase in area planted and yields fuelled overall growth in maize production of 5.3% per year over the same period9. The data for 1983 and 1984 (Figure 1.1) are drastically different from the others. In 1983, there was drought coupled with bush-fires causing low yields and production. In the following year (1984), the area under maize production shot up because there were enough rains, vast bare lands due to bush-fires and jobless Ghanaian deportees from Nigeria who took to farming. Estimated by the author. 4 University of Ghana http://ugspace.ug.edu.gh Figure 1.1: Maize Production Indicators, Ghana, 1975-1997 Year Source: Compiled and estimated by the author based on data from PPMED, MOFA, Accra, Ghana. The Problem Agricultural research in Ghana has sometimes suffered a great deal of embarrassment from some policy makers. Some political-headed institutions in the sixties " publicly declared agricultural research as a waste of time and money, an irrelevant and unproductive activity associated with men who were only competent to talk and write about food but could not produce it themselves"10. This was probably because there was no study to show the returns to research and there was a communication gap between researchers and decision-makers. A recent study in Ghana on returns to maize research La-Anyane, S. " Research in Agriculture in Relation to National Economic Development". CSIR/Universities/U.S., Academy of Sciences Workshop, January 18. 1971 pp 1-17. 5 University of Ghana http://ugspace.ug.edu.gh investment indicated that the internal rate of return to maize research between 1968 and 1991 was 74%n . Bilateral grants and loans have contributed significantly towards research funding. However, in view of the usual excuse of financial constraints feeing the Central Government, the operating budget per scientist has been on the decline. In Ghana, the amount of money actually released to finance agricultural research always falls short of the amount requested. In Table 1.2, the counterpart funds requested and released for agricultural research are compared. Apart from 1992 when about 78% of the request was met, less than 50% of the budget request have been provided since 1993. Table 1.2: Comparison of Counterpart Funds Requested and Released in Ghana for Agricultural Research, 1992 to 1996 (in '000 cedis). Year Budget Requested Budget Released % Released 1992 129000 100000 78 1993 845900 100000 12 1994 753500 200000 27 1995 984900 244100 25 1996 1500000 724000 48 Source: World Bank (Aide-Memoire), 1997. pp.9-10. Sanders John H., Taye Bezuneh, and Alan C. Schroeder. (1994). ” Impact Assessment o f the SAFGRAD Commodity Networks." USAID/AFR/OAU/STRC-SAFGRAD. pp. 1-28. University of Ghana http://ugspace.ug.edu.gh In recent times, the demand for assessing agricultural research impact has been growing. Some of the reasons for high demand for agricultural research impact assessment are: 1. The steady decline in public and donor funds for agricultural research. 2. Increased pressure for accountability of research funds and better management of agricultural research. 3. Competing demands for available funds. 4. Increased request for evidence of accomplishments and people level impact from such research activities. 5. The shift to market-driven research. There is the need to continuously provide information on the returns to investments in agricultural research for donors, policy makers and research managers, to justify demand for increases in budget allocations to agricultural research. Research Questions: The important questions that therefore, arise are as follows: What is the adoption rate of maize technology in Ghana? What are the factors driving the maize adoption process in Ghana? What pay-off have investments in maize research and extension made on the Ghanaian society? These are the issues that the present study addresses. Objectives of the Study The benefits of farmers adopting a new technology are often ascribed to research investments only, with the implicit assumption that related investments of institutions are kept constant. Therefore it is necessary that in evaluating a research programme, 7 University of Ghana http://ugspace.ug.edu.gh information is provided on the role and sequencing of investments in complementary institutions to avoid over estimation of a particular investment. To disaggregate the impacts of complementary investments requires other quantitative methods that will demand high quality time series. Unfortunately, such data is not available in Ghana. The general objective of the study therefore, is to determine the economic impact of public investments in maize research and extension in Ghana associated with the GGDP from 1979 to 1997. The specific objectives of the research are: 1. To describe the biophysical, socio-economic, organisational and policy environment of maize production in Ghana. 2. To determine the factors associated with the adoption of improved maize varieties. 3. To estimate the economic returns to past public investments in maize research and extension in Ghana from 1979 to 1997. 4. To make policy recommendations for improving efficiency of future maize research. In order to achieve these specific objectives, the study describes and analyses the production environment in Ghana, traces the development of maize research, and measures the pattern and intensity of adoption of improved maize technologies between 1979 and 1997. 8 University of Ghana http://ugspace.ug.edu.gh Justification of the Study As indicated earlier, only one detailed study of returns to research in Ghana has been carried out. Donors12 are eager to find out the rate of return to their investment in maize research and the impact of that on the Ghanaian economy. Maize has been chosen as the focus for this study for two main reasons. First, maize is Ghana’s leading cereal crop and is grown by a large number of rural households. The economic importance of maize in Ghana cannot be overemphasized. During the 1991-92 cropping season, out of two million households engaged in harvesting staple grains and cash crops, 1.74 million grew maize.13 In that same period, the value of maize crop totaled approximately 61 billion cedis, or about US $ 14 million at the prevailing exchange rate. Since research-induced productivity gains in maize are translated directly into increased income for millions of producers and consumers, it is important to know whether maize research and extension efforts are producing tangible benefits. Second, in recent years maize has commanded a large share of Ghana’s limited agricultural research resources. The Ghanaian Council for Scientific and Industrial Research (CS1R) and the International Service for National Agricultural Research (ISNAR) have reported that there were many more researchers working on maize than on any crop in the early 1990s in Ghana14 Table 1.3 shows that there are many more researchers in maize than in cassava or rice. The number of researchers in maize development is more than five times that of rice research and three times that of cassava research. Time spent on maize research by researchers is about ten times that of rice researchers and about six times that of cassava researchers. Like the Canadian Development Agency (C3DA) and the National Agricultural Research Project (NARP). 13 Ghana Statistical Service. Ghana Living Standards Survey 3. 1995. pp. 67-68. CSIR and ISNAR,. "Review of the Ghana Agricultural System". Vol. 1: Report The Hague: ISNAR 1991. pp 54-57. 9 University of Ghana http://ugspace.ug.edu.gh Table 1.3: Number of Ghanaian Researchers and Person Years in Commodity Research, in Ghana,1988. Commodity No. of researchers involved Person years Maize 32 10.9 Cassava 10 1.8 Rice 6 1.1 Source: CSIR and ISNAR VoLl, 1991, p.55. The period 1979-1997 was chosen because it is associated with a specific research and extension project. The Ghana Grains Development Project (GGDP), a Canadian International Development Agency (CIDA) funded project, has provided significant support for developing and disseminating the maize technologies between 1979 and 1997. Thus, it is important to establish what impact the level of investment in maize research and extension have achieved so far. Limitations of the study As will be noted, a large number of varieties and agronomic recommendations were released by the GGDP between 1984 and 1997 for various ecological zones. Ideally, one should have identified the costs and benefits of each of the recommendations to measure the aggregate net social gains to society. Given the number of technologies developed, the long period of time and poor data availability particularly at the farmer level, it is practically impossible to obtain data on costs and benefits accruing to farmers. The study therefore, combines the range of technologies and estimates the rate of return. The impact of maize research and extension in Ghana might be more than the estimated internal rate of return (IRR) if other benefits such as nutritional benefits are captured. These are however given qualitative descriptions. Furthermore, several maize technology experiments were conducted in formers’ fields in the research phase of the Project (1979-1983). Hence, actual adoption of the technologies might start earlier than the assumed year (1984) and this will have a positive effect on the IRR. 10 University of Ghana http://ugspace.ug.edu.gh Research and extension costs and yield gains can affect the outcome of the IRR and since these, in some cases, are estimated, the issue is overcome with sensitivity analysis. The study realises that unless the effect of the improved technologies at the farm level are monitored and evaluated continuously, it is difficult to obtain reliable estimates for the IRR The study therefore uses conservative estimates for the assessment of the benefits. Although it would have been more revealing to ascertain farmers’ perception of the risks of the technologies relative to the traditional ones, this was not done because data on this was unavailable. Furthermore, the link between profitability and risk of technologies on one hand and, adoption has not been established for lack of data. Organisation of the Study The rest of the study is organised into eight chapters. Chapter Two reviews the literature on research impact. In Chapter Three, maize technology development in Ghana is discussed. This chapter traces the history of agricultural research in Ghana and describes the institutions that have collaborated in the development of maize technology in Ghana. The chapter further looks at how the maize technology was developed from 1979 to 1997, the period under study. It thus reviews the breeding and agronomic experiments underlying the release of varieties and production recommendations. This is to show that a real maize research has occurred; it also shows biological scientists' perceptions of research impact. The Fourth Chapter discusses the production environment and the Fifth Chapter focuses on the methodology of the study. The research results on maize adoption and the determinants of adoption are given in Chapter Six. Chapter Seven presents the results of the returns to maize research, and Chapter Eight summarises the study and its policy implications. Technical details are included in the appendices. University of Ghana http://ugspace.ug.edu.gh CHAPTER 2 LITERATURE REVIEW Impact assessment Impact assessment is a special form of evaluation that deals with the effects of a project output on target recipients. The basic principles of impact assessment are causality, attribution and increment. There are two broad categories of interpretation of impact. Biological scientists look at impact as a direct output of an activity. For example, the release of a new variety is an impact. Social scientists, on the other hand, examine the effects of the product on the ultimate users - people level impact. Impact begins when the behaviour of beneficiaries changes. Thus, impact deals with actual adoption of the research output and subsequent influence on production, income, welfare, the environment or any set development objectives. A project that has an impact must have some movements in the direction of the desired objective. Impact Assessment Methods Impact assessment may be done ex-post or ex-ante. The ex-post analysis deals with technologies already in use by farmers while ex-ante is for technologies not yet adopted. In both cases, some of the data required to measure impacts can be observed directly or estimated indirectly from other evidence. The benefits and costs are transformed into a single number, the internal rate of return (IRR), which is a discounted evaluation measure for a single or a set of projects. The IRR is defined as the discount rate that just makes the net present value of incremental net benefits stream, or incremental cash flow equal to zero. The internal rate of return thus represents the 12 University of Ghana http://ugspace.ug.edu.gh maximum interest that a project can pay for the resources used if the project is to recover its investments and operating expenses and still break even. A project or program is usually considered economically viable if the internal rate of return exceeds the opportunity cost of capital. The rate of return to a set of investments can be calculated as a marginal or average rate. The marginal rate of return (MRR) measures the return to the last cedi invested in each component. This can be done with econometric estimation of the relationship between the supply function and program expenditures and requires estimating an aggregate function including research and complementary investments as separate variables. However, estimating MRR requires good time series data. These are problematic in Ghana. The average rate of return (ARR) looks at the whole expenditure as given and calculates the return to a whole set of expenditures. The ARR indicates whether or not the allocation of resources between investment components is optimal.2 There are various methods for estimating the rate of return for ex-ante and ex-post analysis3. These include: A. Ex-ante analysis: i. Benefit cost analysis ii. Simulation models iii Mathematical programming models Gittinger Price J. Economic Analysis of Agricultural Projects. Johns Hopkins University Press. 2nd Ed- 1982 pp. 299-361. Oehinke James F.and Eric W, Crawford.”The impact of Agricultural Technology in Sub-Saharan Africa: A Synthesis of Symposium Findings.” Michigan State University International Development Papers, no. 14, 1993.pp 1-12. See Schuh Edward G. and Hellio Tollini.’’Costs and Benefits of Agricultural Research.” World Bank Staff Working Papers.The World Bank, no.360. 1979. pp 1-55. Anandajayaseram P. ,D R. Martella and M. Rukuni. A Training Manual on R and D Evaluation and Impact Assessment of Investments in Agricultural and Natural Resources Research. SACCAR1996.pp 51-59. 13 University of Ghana http://ugspace.ug.edu.gh iv. Scoring models B. Ex-post analysis: i. Production functions/Econometric method ii Surplus approach a. Benefit-cost approach b. Input saved approach c. Index number approach 1. Linear functions with parallel shifts. 2. Linear functions with non-parallel shifts. 3. Non-linear functions with parallel shifts. 4. Non-linear functions with non-parallel shifts. Ex-ante analysis looks at the future and therefore provides a more realistic and relevant guide for decision-making. Its disadvantage is that it is time consuming and complicated because a series of scenarios must be analysed. Secondly, the quality of data input has effect on judgement and decision-making. The strengths of ex-post approach are that it draws on prior knowledge of an empirical effect of research on the economy and how the economy functions. It also employs statistical procedures to separate the effect of research and can control other variables affecting research output. This permits a more precise evaluation of the contribution of research. The disadvantage is that a lot of data is required. Some writers4 have reviewed and compared approaches used to evaluate public agricultural research. They follow the procedure of categorising returns to research studies into ex-post and ex-ante evaluations. Ex-post studies fall into two main groups. Those studies Norton G. W. and Jeffrey S. Davis.” Evaluating Returns to Agricultural Research: A Review.” American Journal of Agricultural Economics 63(4) 1981. pp 685-699, 14 University of Ghana http://ugspace.ug.edu.gh using consumer and producer surplus analyses to estimate average rate of return to research and those using production function analyses to estimate the marginal rate of return to research. Ex-ante studies M into four categories. These are those using scoring models to rank research activities, those employing benefit-cost analysis, those using simulation models and those employing mathematical programming to select an optimal mix of research activities. They conclude that none of these approaches is superior in all situations. The Economic Surplus Method o f Ex-post Evaluation By far, the most widely used method for analysing research impact in many situations is a partial equilibrium framework based on economic surplus approach.3 The main advantage of the economic surplus approach is that it is relatively flexible. It can be modified to account for a number of side effects of technological change such as price policy and trade. It further provides the means of analysing the distribution of benefits from research to producers and consumers when the demand and supply curves are known. It can also be applied to both closed and open economies. Some simplifying assumptions are made in the economic surplus method. First, the supply and demand curves are assumed to be linear and to shift in parallel as a result of technological changes. With the assumption of parallel shifts, the choice of functional form has little effect on the size and distribution of benefits. In relation to total benefits, functional forms and elasticity are relatively unimportant compared with the nature of the supply shift. In relation to the distribution of benefits, functional forms are relatively unimportant compared with the sizes and the nature of the supply shifts6. Second, a single period (static) model is used and dynamic issues are not included. Third, competitive markets are assumed. Fourth, Alston Julian M .G eorge W. Norton and Philip G. Pardey. Science Under Scarcity: Principles and Practice for Agricultural Research Evaluation and Priority Setting. Cornell University Press. 1995. pp. 207-208 15 University of Ghana http://ugspace.ug.edu.gh standard surplus measures are used as measures of welfare change. With these assumptions, qualitative effects and formulae can be computed. In impact assessment analysis, the main point is to compare a situation without research against a situation with research. It must be based on carefully constructed scenarios of the situations of with and without research.7 The economic surplus method provides a relatively simple and flexible approach to valuing research in with and without situations. In order to turn agronomic data into economic values, the surplus approach employs the concepts of supply, demand and equilibrium. "Supply" represents producers' production cost, and "demand" is the consumers' consumption values. The interaction of the two leads to an "equilibrium". Economic welfare depends on the equilibrium price and quantity, producers' production costs and consumers consumption values. Generally, the social value of a given production and consumption level can be estimated using the economic surplus concept defined as the area between the supply and demand curves as in Figure 2.1. The equilibrium point is the optimal point because it provides the maximum amount of economic surplus. Total surplus is the sum of the consumer and producer surpluses. Figure 2.1 assumes a shift to the right in the supply functions from S0 to Sn that improved technology brings. The benefits from research and extension are estimated by finding the gains and losses in producer and consumer surpluses over time, giving a net gain represented by area BCDE. The shape and the size of this area depend on the functional form of the supply and demand curves and the magnitude of the supply shift. The cumulative benefits are then compared with cumulative research costs by calculating the average rate of 6 Alston Julian M, George W. Norton and Philip G. Pardey. Op. Cit.pp.207-234. Masters W.A.^Bakari Coulibaly^Diakalia Sanago, Mamadou Sidibe, Anne Williams, John H. Sanders and Jess Lowenberg-DeBoer. The Economic Impact of Agricultural Research: A Practical Guide. Department of Agricultural Economics. Purdue University. 1996. pp. 6-12. 16 University of Ghana http://ugspace.ug.edu.gh return to research expenditures. Figure 2.1: Economic Surplus with Linear Demand and Supply Curves. The index number method was used to estimate the impact of increased hybrid com yields on the net social surplus in the United States by Grilliches (1957). It was hypothesised that the essential impact of agricultural research was to raise productivity, causing the aggregate supply function to shift downward from So to Sn (Figure 2.1). Index number approach incorporates elasticity of demand and supply in the analysis. A simplified version of the economic surplus approach is the Benefit-Cost approach. The Benefit-Cost estimate assumes a perfectly inelastic supply curve and perfectly elastic demand curve. In the Benefit-Cost model, the welfare gains from research and extension are 17 University of Ghana http://ugspace.ug.edu.gh represented by area “ABCD” in Figure 2.2. This is estimated by multiplying the differences between outputs (Q„-Q„) by the price, P. A major empirical problem is how to obtain productivity gain that reflects only the output of research. This measure is required to know how much to shift the supply curve. Often, the supply shift is assumed constant but this is usually not the case. Figure 2.2: Benefit-Cost Approach Price S sn A B P D C Quantity Akino-Hayami Method: Another method that has been used widely in the assessment of research impact, particularly in developing countries, is the Akino and Hayami method8. This requires moderate data. The method assumes non-linear demand and supply curves and a pivotal shift of the Akmo Masakatsu and Yujiro Hayaim.’E fficiency and Equity in Public Research: Rice B reeding in Japan’s Economic Development.” American Journal o f Agricultural Economics. 57(1975).pp 1-10. 18 University of Ghana http://ugspace.ug.edu.gh supply curve as a result of changes in technology as shown in Figure 2.3. A kino and Hayami developed formulae to quantify the areas of the change in social surplus resulting from investments using Japanese rice breeding research. This method incorporates elasticity of demand and supply and specifies constant elasticity curves. Figure 2.3: Akino-Hayami Approach The total benefits (social gains) from research and extension is the area “COB” plus area “ABC.” Area COB is calculated as: COB = k-factor x Total Production Value, where: k-factor = Adoption rate x (Yield gains from improved varieties -*■ Yield of improved varieties). Or, k-factor = Adoption rate x {(Yield of improved variety - Yield of traditional variety) +Yield of improved varieties} 19 University of Ghana http://ugspace.ug.edu.gh Area ABC is calculated as: ABC = y 2 x area COB x {k-factor x (1 + price elasticity of supply)}2 -(price elasticity of supply + price elasticity of demand) The formula for the social gains generally does not include per unit input costs. It is therefore necessary to subtract them separately from social gains, as a separate step in the analysis. The estimated social gains from Akino-Hayami method are considered to be conservative because the social gains estimated with this approach are smaller than those estimated with linear curves. The mathematical relationships of the parameters in the Akino- Hayami approach are however, more complicated than with linear curves. The conservative nature of the results with linear curves can be overcome with sensitivity analysis. Role o f economic policy on returns to agricultural research A particular economic policy regime can have a major impact on the potential and actual returns to agricultural research. Economic policy affects the benefits and costs of research by distorting the price relatives. It can prevent people from adopting technologies that would have otherwise been adopted. An example is given on Brazil where fertilizer and hybrid com had low adoption because of government policy that lowered domestic agricultural prices relative to fertilizer prices9. When economic policy distorts relative price ratios, the contribution of a research programme can be either over- or under-stated. In this situation, shadow prices are used to estimate the benefit-cost ratios. The difference in research impacts between Africa and Asia results from the degree of Schuh Edward G, and and Helio Tollini. Op. Cit. pp 1-40. 20 University of Ghana http://ugspace.ug.edu.gh policy intervention,10 In Africa, restrictions on international trade and domestic marketing have taxed agriculture, transferring funds from farmers to urban dwellers. The implications of agricultural taxation are reduced adoption rates of new technologies and the transfer of whatever gams were achieved with the new technologies to the non-farm sector. Both effects make the returns to past research higher than those implied by conventional methods. Masters and Sanders (1994) further note that there is a more rapid population growth in Africa than elsewhere. This channels productivity gains into on-farm consumption. If the value of the on-farm consumption is not accurately valued, it may affect the impact results. Cross-border spillover is valuable and critical for investment decisions of donors and international agencies. Attractive rate of return to investment in wheat breeding in Nepal was due in part to her ability to capture spillover benefits from neighbouring countries and from international agricultural centres.11 While any price policy affects the distribution of the research benefits, the net national or world benefits may be reduced, left unchanged or increased, depending on the nature of the policy and the significance of the country in the world market for the commodity.12 Price distortions may affect research investments and thus economic efficiency. All forms of intervention modify the pattern of benefits from cost-reducing research relative to those obtained under free trade. Other factors affecting the results of impact assessment studies in Africa include climatic conditions, civil unrest, research system performance, policy affecting input and Masters William A. and John EL Sanders. “The Impact of Agricultural Research in West and Central Africa: Concepts and Evidence.” Paper prepared for discussion at the SPAAR/USAID Workshop on Regionalization of Agricultural Research. Banjul, The Gambia March 14-18,1994. pp. 1-26. Moms Michael L.Jfi. J. Dubin and Taneswar Pokhrel. “Returns to Wheat Breeding Research in Nepal” Agricultural Economics 10(1994) pp.269-282. Alston Julian M , Geoff W. Edwards, and John W. Freebaim. “Market Distortions and Benefits fromResearchAmerican Journal of Agricultural Economics. 9(1988). pp. 281-288. 21 University of Ghana http://ugspace.ug.edu.gh output prices and markets involving input supplies and output markets which tend to support adoption of technologies.13 A study of trends in food prices in Ghana between 1970 and 1993 showed that the real wholesale price of food was declining.14 One of the explanations of this decline was attributed to favourable climate and yield increases for crops such as maize. Another reason is that of falling marketing costs since 1984 because trucks, spare-parts and funds for road construction and repairs were made available. Evaluation of agricultural research is a complex task that provides a wide range of alternatives and challenges. Although there are various methods to evaluate agricultural research, none of them is ideal in all situations and one has to use the appropriate method suitable for one’s condition. Indeed, the method chosen to evaluate research should be shaped by the availability of appropriate data. Often, simple approaches which are less data demanding are more useful than more complicated procedures. Adoption and diffusion In the assessment of impact of agricultural research, it is imperative to know the extent of use of the developed technologies. This is needed to compute the benefits of research. In situations where data are unavailable, it is necessary to conduct adoption studies to know the level of adoption of the technologies. Agricultural research has little merit unless it leads to adoption and use of the derived technology by farmers15 The adoption of a new technology can be considered in several ways. Depending on 13 Oehmke James F. and Eric W. Crawford. Op.Cit pp 1-12. Alderman Harold and Gerald Shively. ‘'Economic Reform and Food Prices: Evidence from Markets in Ghana.” World Development. 24(1996) pp. 521-532. Ekpere J. A ^Towards a New Orientation to the Challenge of Technology Transfer Systems and Food Production in the 21 Century Africa. ’ Paper Presented at the Regional Workshop. Technology Options and Transfer Systems. Abidjan, Cote D’Ivoire. April 26-28,1995. pp, 1-26. 22 University of Ghana http://ugspace.ug.edu.gh one's objective, the adoption of a new technology such as improved variety, may range from using a little bit of the technology to using all the technology. Information about current and past technology practices should be collected to examine adoption history and diffusion. While adoption is measured at one point in time, diffusion is the spread of the new technology across population and over time. It is assumed that technology diffusion has an S-shaped curve showing a cumulative proportion of adoption. There is a slow initial growth in the use of the technology. This is followed by a more rapid increase in adoption and then a decrease as cumulative proportion of adoption gets close to a maximum. The logistic function is used to depict the diffusion curve. The curve can mathematically be described as: Yt = K/( 1 + e"a'ht) Where Yt = the cumulative percentage of adopters or area at a time K = the upper limit of adoption, b = a constant, related to the rate of adoption, a = a constant, related to the time when adoption begins. The logistic curve assumes that with the constants, a, and b, infrastructure, relevant price ratio and technology remain unchanged over the period. 23 University of Ghana http://ugspace.ug.edu.gh Determinants o f adoption The use of discrete choice models such as LOGIT, PROBIT and TOBIT in explaining adoption process is well documented.16 These studies have variously used farm and former characteristics and or fanner perceptions to explain the adoption process. The role of formers’ preferences in adoption decisions has received little attention in adoption studies. A TOBIT model used on sorghum in Burkina Faso and Guinea found that formers’ perception of technology significantly affect their adoption decisions.17 Several attempts have been made to explain the factors of adoption. Generally, three main issues for explaining adoption decisions can be identified. These are economic constraint, innovation-diffiision and “adopter perception.” The economic constraint model argues in favour of the distribution patterns of resource endowments as the major determinants of adoption. The innovation-diffusion model believes that access to information is the main determining factor in adoption decisions.18 The adopter perception model suggests that the 19perceived attributes of innovations are the key determinants of adoption behavior . Shively Gerald E, “Consumption Risk, Farm Characteristics and Soil Conservation Adoption among Low-mcome Farmers in the Philipines.’' Agricultural Economics. 17(1997)pp.l65-177. Nichola Tennassie and John H Sanders. “A Probit Analysis of the Determinants of Adoption when Inputs are Rationed; The Gezira Experience with Hybrid Sorghum.” Science, Technology and Development Vol,14.no.3, 1996 pp. 107-119. Adesina, A A. and Baidu-Forson, J. “Farmers’ Perception and Adoption of New AgriculturalrEvidence from Analysis in Burkina Faso and Guinea, West Africa.” Agricultural Economics 13(1995) pp 1-9. Adesina Akinwumi A and Moses M. Zinnah. ‘Technology Characteristics, Fanners’ Perceptions and Adoption Decisions: A Tobit Model Application in Sierra Leone.” Agricultural Economics 9(1993) pp.297- 331. Dankyi, A.A., H A Dakurah and J.N. Asafo-Agvei. “Cowpea Technology Adoption in Ghana” Paper presented at the OAU/SAFGRAD Workshop, Abidjan, Ivory Coast 1993 pp. 1-33 Dankyi, A A , V.M. Anchirinah and A.O. Apau ‘The Adoption of the Improved Rice Technologies in the Wetlands of the Ashanti Region of Ghana” Unpublished Paper. West Africa Rice Development Association (WARDA). Bouake, Ivory Coast, 1996 pp 1-39. i! Adesinah, A A. and Baidu-Forson, J. Op. Cit. pp 1-9 18 Rogers, EM. “Diffusion of Innovations. Free Press. Glencoe. Illinois. 1962 Kivlin, J.E. and F.C. Fliegel. “ Differentiated Perceptions of Innovations and Rate of Adoption.” Rural Sociology 32 (1967) pp78-91. 24 University of Ghana http://ugspace.ug.edu.gh In many of these studies, ecological zones are not included as some of the explanatory factors affecting the adoption of improved technologies. Farmers in different ecological zones may adopt improved technologies differently because of the environment in winch they live. If technologies developed are not compatible with the conditions prevailing in the ecological zones, adoption may be low. Thus, the type of improved crop varieties and technologies developed may be a response to the environmental conditions prevailing in the ecological zones. In Ghana, there are four ecological zones running approximately horizontal to each other. Beginning from the coast is the coastal savannah and moving northwards are the forest, transition and the Guinea Savannah in that order. In general, rainfall increases northwards from the coastal savannah zone to the forest zone but decreases from transition zone to the Guinea Savannah zone. Temperatures, on the other hand, increase from the coastal savannah zone to the Guinea Savannah zone. Thus while the Guinea Savannah has one rainfall season, the coastal savannah, the transition and the forest zone have two rainfall seasons. With this ecological distribution, farmers may adopt improved technologies differently. Hence, it is important to include ecological zones when explaining the adoption of improved technologies. Understanding adoption behaviour goes beyond looking at single relations to multivariate relationships. Multiple regression may be used to analyse such relationships but the dependent variable must be continuous. Since many adoption studies deal with adoption as a "yes" or "no" dependent variable, LOGIT, PROBIT and TOBIT analyses are the appropriate tools. Both LOGIT and PROBIT techniques use a series of characteristics of the farmer or the farm to predict the probability of adoption. Their analyses give similar results. The University of Ghana http://ugspace.ug.edu.gh difference between them is that while LOGIT assumes that the dependent variable follows a logistic distribution, PROBIT assumes a cumulative normal distribution. The estimated probability of adoption in the LOGIT model is given by: F(bX) = 1/ l+e"bx This is the cumulative probability distribution. The expression b'X is defined as: b'X = bo + bi Xi + biX2 +. . . .+ bkXk Where: b0 = constant bi, b2,....bk = other estimated coefficient. Xi, X2,....Xk = values of independent variables. With this expression the probability of adoption of a new technology can be calculated for farmers. The logistic regression gives the same results as LOGIT. The logistic regression can be rewritten in terms of the “odds” of an event occurring. The odds of an event occurring are the ratio of the probability that it will occur to the probability that it will not. When the logistic model is written in terms of the log of the “odds”, it is called a LOGIT. The TOBIT model measures both the probability of adoption and intensity of adoption. The theoretical aspect of the logistic regression and TOBIT are further discussed in Appendix 5. Chapter Summary There are several methods for assessing impact of agricultural research but the most widely used method is the economic surplus method. The advantages of this method he in its relative flexibility, simplicity and less data requirement. One of the parameters needed to estimate the benefits to research and extension activities are the adoption rates of the improved technologies. These can be estimated through adoption studies. A complementary 26 University of Ghana http://ugspace.ug.edu.gh analysis is to determine the factors affecting the adoption process. This can be done with logistic regression and TOBIT models. 27 University of Ghana http://ugspace.ug.edu.gh CHAPTER 3 MAIZE TECHNOLOGY DEVELOPMENT IN GHANA Agricultural Research in Ghana Organised agricultural research in Ghana did not exist until after the establishment of the Government Botanical Gardens at Aburi in 1890. In the same year, the Aburi Botanical Station became the Department of Agriculture. The original objective was to introduce plants found to be profitable in other countries and to search for other economic plants in Ghana. Serious agricultural research however, began in 1900. Regional agricultural stations were established between 1903 and 1918 to introduce and distribute new crops and provide training for agricultural officers. The Ministry of Agriculture (MOA), as known at that time, had eight divisions of which the Scientific Services Division was responsible for all research work. In 1962, the MOA was re-organised and the Ghana Academy of Sciences (GAS) took charge of all agricultural research. Other agricultural stations were given to the State Farms Corporation (SFC) and the United Ghana Farmers Co-operative Council (UGFCC) to produce food As a government policy, the CSIR was created to replace GAS in 1968. Together with the Universities they carried out agricultural research in Ghana. Presently, five institutes of the CSIR and the Universities carry out agricultural research in Ghana. The CSIR institutes are the Crops Research Institute (CRI), Soil Research Institute (SRI), Animal Research Institute (ARI), Savannah Agricultural Research Institute (SARI) and the Forestry Research Institute (FRI). The Cocoa Research Institute of Ghana (CRIG) and 28 University of Ghana http://ugspace.ug.edu.gh the Oil Palm Research Institute (OPRI) also carry out research in cocoa and oil palm respectively. Maize was first introduced in Ghana in the 16th century from the Portuguese1. In the colonial years, other maize varieties were introduced from Latin America and other African countries. By 1956, the Ghana Government had established maize improvement centre at Kwadaso that in 1963 became part of Crops Research Institute. The Ghana Grains Development Project The Ghana Grains Development Project (GGDP), under the CRI, is a bilateral aid venture between the Governments of Ghana and Canada. It was established in 1979 to improve maize and legume production in Ghana. The Crops Research Institute is the executing agency for the Ghana Government and the International Maize and Wheat Improvement Centre (CIMMYT in Mexico) represents the Canadian Government. Other participating institutions are the Grains and Legumes Development Board (GLDB), and the Ministry of Food and Agriculture (MOFA). The International Institute of Tropical Agriculture (DTA) in Nigeria since 1985, participated in the second and third phases of the project. Initial efforts of the GGDP in Phase 1 (1979-1984) concentrated on training staff procurement of basic equipment and identifying appropriate improved maize and legume varieties for farmers. Later, emphasis was shifted to crop management research on farmers' fields to develop improved production practices. Effective recommendations were formulated for maize mono-cropping system and later for inter-cropping systems. Edmeades, G., A. ADankyi, K. Marfo and R. Tripp. “On-farm maize research in the transition zone of Ghana” In Planned Change in Fanning Systems. Edited by R Tripp. 1991. P.66. Wiley-Sayce Co- Publication. 29 University of Ghana http://ugspace.ug.edu.gh In Phase 2 of the Project (1985-1990), the advances and methodologies of the previous research were expanded. The research priorities, extension strategies and maize recommendations were refined. There was more collaboration between research, extension and farmers. In the third and final Phase (1991-1995), the productivity increases in maize by fanners were the main concern of the project. Gender awareness was given priority. As at 1998, nine open-pollinated improved varieties and three hybrids have been released together with their agronomic recommendations. At the onset of the Ghana Grains Development Project in 1979, the major improved varieties recommended for use in all ecological zones were La posta CRI, Composite 4 and Golden Crystal. These varieties were used in both major and minor rainy seasons. They were very tall and therefore easily lodged, and were susceptible to diseases. Furthermore, on-farm agronomic data to support their planting were very scanty. The main objectives of maize breeding programme of the GGDP were to: i) Increase the grain yield and the yield components. ii) Improve lodging resistance. iii) Develop specific varieties for the different ecological zones and livestock feeding. iv) Develop maize variety for livestock feeding. v) Develop varieties for major and minor seasons. vi) Improve disease and insect resistance in maize. vii) Improve the protein quality of mai ze. 30 University of Ghana http://ugspace.ug.edu.gh In 1982, following a mid-term review of the GGDP, the breeding programme was modified in scope and direction for on-farm and extension activities. The varietal requirements for Ghana were determined and grouped into five types. These were based on ecology, season, former and consumer preferences2. The varieties were: i) 120-day white dent varieties for the transition, forest and Guinea Savannah zones in the major season. ii) 110-day yellow flint/dent varieties, for all parts of Ghana in major and minor seasons. iii) 105-day white dent varieties for transition, forest and coastal savannah zones in the major season. iv) 90-day soft white dent for maritime Volta Region. v) 90-day yellow flint varieties for early planting in the Guinea Savannah. Between 1979 and 1996, nine new varieties were released by GGDP. The released maize varieties by the CRI from 1954 to 1996 and their potential yields are shown in Table 3.1. Other collaborating institutions The international agricultural institutions such as CIMMYT, ETA, and the Semi- Arid Food Grain And Development (SAFGRAD) have played various roles in the human development, and the flow of information and materials among National Agricultural Ghana Grains Development Project. Fourth Annual Report. Research Results. 1982. pp. 2-7. 31 University of Ghana http://ugspace.ug.edu.gh Research Systems (NARS). The development of early and extra-early maize varieties was supported by SAFGRAD, Table 3.1: Maize Varieties Developed and Released by CRI, 1954-1996. Vatiety Year of Release Special Attributes Days to Maturity Grain Yield ton/ha Nyankariwana I 1954 Yellow 100 2.5 Nyankariwana II 1954 Yellow 100 2.5 GSI, GSIL GSffl 1955-61 White dent 120 3.5 Mexican 17 1961 White dent 110 3.5 Composite 2 1968 White dent 120 4.0 Composite W 1972 White dent 120 5.0 Composite 4 1972 White dent 120 5.0 La Posta CRI 1972 White dent 120 5.5 Golden Crystal 1972, 1984 Yellow Flint/dent 110 5.0 Kawanzie 1984 Yellow 95 4.0 Safita-2 1984 White dent 95 4.0 Dobidi 1984 White dent 120 5.5 Aburotia 1984 White dent 105 4.5 Okomasa 1988 White dent Streak resistant 120 6.0 Abeleehi 1990 White dent Streak resistant 105 5.0 Dorke SR 1990 White dent Streak resistant 95 4.5 Obatanpa 1992 White dent Quality Protein, Streak resistant 110 5.5 j Source: Adapted from Twumasi-Afriyie S. "Achievements of the Crops Research Institute.” CRI, Ghana, 1995. p.5 32 University of Ghana http://ugspace.ug.edu.gh CIMMYT is an internationally funded and non-profit scientific research and training organisation. It is involved in world-wide research on maize and wheat. It is one of the thirteen international centres supported by the Consultative Group on International Agricultural Research (CGIAR). CIMMYT has had a long collaboration with the CRI on maize research in Ghana. It played an important role in the establishment and management of the GGDP in 1979. Since 1979, CIMMYT has supported grains development in Ghana with maize germplasm which breeders in the GGDP have further worked on for the release of several maize varieties. The sources of some of the released maize varieties in Ghana are presented in Table 3.2. Table 3.2: The Sources of Released Maize Varieties in Ghana, 1979-1996. Released variety Germplasm source Institution source Dobidi CIMMYT population 43 (La posta) CIMMYT Safita 2 CIMMYT Pool 16 CIMMYT, SAFGRAD Aburotia CIMMYT population 49 CIMMYT Kawanzie CIMMYT population 31 CIMMYT Okomasa CIMMYT population 43 CIMMYT Abeleehi CIMMYT population 49 CIMMYT Dorke SR CIMMYT population 16 CIMMYT Obatanpa CIMMYT population 63 CIMMYT Source: Various GGDP Annual Reports. 1979 to 1996 The Savannah Agricultural Research Institute (SARI) was originally known as the Nyankpala Agricultural Experimental Station (NAES). Until 1995,it was part of the CRI. At the moment, it carries out crop research including maize in the northern sector of Ghana. The Ministry of Food and Agriculture has been one of the main collaborators of the GGDP. Together with the Grains and Legumes Development Board (GLDB), they 33 University of Ghana http://ugspace.ug.edu.gh carried out on-farm trials, maize technology demonstrations, training of farmers and other extension activities in the transfer of maize technologies. Sasakawa Global 2000 (SG2000) project in Ghana was started in 1986 in the Upper West and Northern regions of Ghana. Later it spread throughout the whole country. The SG2000 project focuses on teaching farmers how to use improved methods for growing crops like maize, sorghum and cowpea. The project arranges the necessary credit by bridging farmers and banks so that inputs like fertilizer are loaned to formers to be repaid after harvest. The SG2000 project carries out its extension activities through field demonstration called Production Test Plots (PTP). Maize Varieties Some of the very early maize varieties released by the CRI were Ghana Synthetics (eg. GSI, GSII and GSDI). CIMMYT introduced La posta into the country in 1968 and in 1972, it had been released as La posta CRI. Other improved varieties were composite 4 and Golden Crystal. Since 1979, several improved varieties have been evaluated against local/farmer varieties. These are full season, medium maturing and early maturing maize varieties. Others are streak resistant and protein quality maize. The average performance of the various recommended full season improved varieties against the local/farmer variety over the years in the various ecological zones are shown in Table 3.3. The full-season maize varieties take about 120 days to mature. 34 University of Ghana http://ugspace.ug.edu.gh Table 3.3: Average Grain Yield of Full Season Maize Varieties, 1979-84 (in t/ha). Variety Ecological zones Transition Forest Guinea savannah La posta CRI 4.21 4.37 3.08 Dobidi 4.05 4.38 3.15 Composite 4 3.98 3.98 2.99 Golden Crystal 4.05 3.54 3.08 Local 2.72 3.07 2.13 Source: Compiled by the author from GGDP Annual Reports, 1979 to 1984. The GGDP recommended varieties," Dobidi" and" Golden Crystal", generally out-yielded the local varieties and showed a good level of stability across ecological zones and years. "Dobidi''(originally Ejura-1-7843), was derived from CIMMYT population "La posta" from progeny selected at Ejura, and consistently performed welL For example, it out-yielded the local variety by 49% in the Transition zone, 43% in the forest zone and 48% in the Guinea Savannah zone. In addition, Dobidi has other desirable characteristics like reduced plant height and lodging. It was released in 1984. To satisfy farmers who plant in the minor season (September-November) or are late in planting in the major season, a medium maturing variety, Aburotia, which matures in 105 days, was developed. Its original name was Tuxpeno P.B. C16, a CIMMYT product. It performed better than the local varieties and was comparable with the full season maize like Golden Crystal. The grain yield from on-farm trials averaged across three years (1981 tol984) for Aburotia and other varieties are compared in Table 3.4. Aburotia yielded about 0.8 t/ha more than the local variety and has sim ilar yield as Golden Crystal, a full-season maize. 35 University of Ghana http://ugspace.ug.edu.gh Table 3.4: Average Grain Yield of Maize Varieties, 1981-84 (in t/ha). Variety Mean yield (t/ha) La posta CRI 2.47 Composite 4 2.40 Aburotia 2.32 Golden Crystal 2.38 Local 1.57 Source: Compiled by the author from GGDP Annual Reports (1981 - 1984). Farmers in the guinea savannah zone in the early years of the project, planted early maturing local maize varieties to fill the hunger gap. Similarly, in the maritime Volta Region, early white local varieties were planted to take advantage of the short and erratic rainfall. Based on these, the GGDP developed a 90-day yellow flint variety, Kawanzie, for early planting in the guinea savannah. In the maritime Volta Region which is a coastal savannah, another 90-day soft white dent variety called Safita 2 was developed and released. Both varieties were released in 1984, The performances of these varieties compared with other varieties in the savannah zones from on-farm trials are presented in Table 3.5. Both recommended varieties, "Kawanzie" and "Safita 2" out- yielded the local variety of about 0.8 t/ha representing about 35% yield increase. Table 3.5: Average Grain Yield of Early Maturing Maize Varieties, 1983-86 (in t/ha). Variety Mean yield (t/ha) Kawanzie* 3.00 Safita 2* 3.02 Pool 16 2.75 Local 2.23 Source: Compiled by t le author from GGDP Annual Reports, (1983-1986). In the early eighties, particularly in 1983, maize streak virus disease became severe and widespread throughout Ghana as a result of a long drought. Breeding for resistance to maize streak virus disease was considered as the most cost-effective control 36 University of Ghana http://ugspace.ug.edu.gh option. Streak resistant varieties available from CIMMYT and IITA were obtained for further research. These were to replace susceptible improved varieties like "Aburotia''. To insure against further epidemic of the disease, streak resistance germplasm were incorporated in all breeding pools and populations. The GGDP management further decided that only varieties that had a fairly good amount of streak resistance were to be released in Ghana. As a result of further research, "Okomasa", " Abeleehi" and "Dorke SR" which were all streak resistant varieties were developed to replace "Dobidi", "Aburotia" and "Safita 2", respectively. The new streak resistant varieties had similar yields as their predecessors. In Table 3.6, the results of streak resistance levels of the released varieties evaluated under artificially induced maize streak virus disease pressure are shown. The replacing varieties, Dorke SR, Okomasa and Abeleehi have high streak resistance levels. Table 3.6: Streak Resistance Levels of Some Selected Varieties. Non-resitant variety % Streak resistant plants (8 weeks after planting) Resistant repacement variety % Streak resistant (8 weeks after planting) Safita 2 8 Dorke SR 95 Dobidi 8 Okomasa 94 Aburotia 0 Abeleehi 86 Source: GGDP Annual Report, 1990, p. 7. Quality protein maize (QPM)3 research began in the early eighties by CIMMYT but was abandoned because of unacceptably soft endosperm In 1989, however, dMMYT made available to GGDP, QPM with grain type similar to normal maize. Normal maize, although has about 10% protein, lacks essential amino acids-lysine and tryptophan. QPM has reasonable amount of these essential amino acids and is thus better than normal maize grown in Ghana 37 University of Ghana http://ugspace.ug.edu.gh Research efforts by GGDP led to the release of a streak resistant medium maturing white dent QPM variety called "Obatanpa"4 in 1992. As presented in Table 3.7, Obatanpa compares favourably with normal maize counterparts across the ecological zones. Across all zones, it yields 3.7t/ha and is 0.5t/ha more than the farmer’s variety. Table 3.7: Grain Yield of Selected Medium Maturing Maize Varieties (in t/ha). Variety Ecological zones Mean Forest Transition Coastal savannah Guinea savannah Obatanpa 3.9 3.5 4.0 3.5 3.7 Abeleehi 3.2 3.7 3.7 4.0 3.7 Aburotia 2.1 3.0 3.5 3.7 3.1 Farmer variety 2.7 3.8 2.2 3.9 3.2 Source: GGDP Annual Report, 1991, p. 109 Quality protein analysis is presented in Table 3.8. Tryptophan, an essential amino acid is about 100% more in the protein and the endosperm of Obatanpa than they are in Okomasa. These make Obatanpa superior to Okomasa. Table 3.8: Protein and Tryptophan Contents of Obatanpa and Okomasa. Variety % Tryptophan in: % Protein Sample Protein Whole grain: Obatanpa 9.8 0.08 0.85 Okomasa 10.4 0.05 0.46 Endosperm: Obatanpa 8.0 0.06 0.76 Okomasa 10.0 0.38 0.38 Source: GGDP Annual Report, 1991, p. 14. “Obatanpa” is a vernacular word meaning “good nursing mother” which is given to maize variety because it has quality protein. Obatanpa was released to farmers in 1992. 38 University of Ghana http://ugspace.ug.edu.gh To further verify the superiority of the improved varieties over local variety, a farmer-managed variety experiment was conducted in 1985. In this trial, formers were given improved varieties to plant and manage themselves in whatever way they wished. The only conditions were that they also planted their own variety in the same field and under the same management and that researchers could observe and take records. The results of this experiment are presented in Table 3.9. Dobidi and Aburotia, the recommended varieties yielded 0.58t/ha and 0.45 t/ha respectively more than the former’s variety. The densities and the number of ears harvested from the recommended varieties were more than the former’s variety. This is evidence that the recommended varieties perform better than the farmers' variety. Table 3.9: Grain Yield and other Parameters of Selected Maize Varieties (in t/ha). Variety Harvest density Ears at harvest Grain yield Dobidi 27000 25400 2.49 Aburotia 27500 25200 2.26 Farmer's variety 25800 22200 1.81 Source: GGDP Annual Report, 1985, p. 133. Hie yield advantages of improved varieties are further shown in Table 3.10. The table shows that improved varieties are more responsive to fertilizer. For example, the improved variety yields more than the local variety by 0.76 tons/ha at zero fertilizer and by 1.20 tons/ha at higher levels of fertilizer. Thus improved maize gave more yields using fertilizer than local varieties. 39 University of Ghana http://ugspace.ug.edu.gh Table 3.10: Grain Yield of Local and Improved Maize Varieties (in t/ha). Maize variety Fertilizer level Zero (Difference) 90:60:60 (NPK) Improved 2.47 (1.50) 3.97 (Difference) (0.76) (1.20) Local 1.71 (1.06) 2.77 Source: GGDP Annual Report, 1980, p.43 Planting Densities Results from an exploratory survey in Mampong-Sekyedumasi area in 1979 showed that planting densities in farmers’ fields varied from 20,000 to 45,000 plants/ha. Using these as guidelines, variety by planting density trial was conducted in 1980. It involved some improved varieties and three planting densities - 25,000, 50,000 and 75,000 plants/ha. The results from the forest, transition and Guinea Savannah zones are presented in Table 3.11. The data indicate that as the density increased beyond 50,000 plants/ha the yield generally decreased as a result of competition for light and soil nutrients. This further means that the resources to the plant - nutrients and water in the soil and light are being used more intensively resulting in negative returns to scale. Table 3.11: Grain Yield of Maize Varieties under Different Densities (t/ha). Density (plants/ha) Variety 25,000 50,000 75,000 Forest Zone: La posta CRI 4.61 4.89 4.06 Aburotia 3.68 4.04 4.38 Transition Zone: La posta CRI 3.85 4.35 3.69 Aburotia 3.55 4.34 4.08 Guinea Savannah: La posta CRI 2.87 3.00 2.84 Aburotia 2.59 3.42 3.10 ource: GGDP Annual Report, 1980, p.62. 40 University of Ghana http://ugspace.ug.edu.gh Planting densities in excess of the optimum caused considerable lodging and barrenness particularly with full season varieties such as La posta. With medium maturing varieties such as Aburotia, total lodging and ears per plant decreased with increase in densities. These results are presented in Table 3.12. Bird damage also generally increased. Taking into consideration a 12% stand loss by farmers, a planting density of about 56,000 plants/ha was recommended for farmers growing full season (120-day) varieties and 62,500 plants/ha for medium varieties such as Obatanpa and also for early varieties. Table 3.12: Selected Characters for La posta at Three Densities in Ghana. Variety Density Total lodging (%) Bird damage (%) Ears per plant La posta CRI 25,000 15.6 2.9 1.01 50,000 24.6 7.2 0.87 75,000 28.8 10.2 0.72 Aburotia 25,000 16.2 4.0 1.02 50,000 15.8 7.8 0.92 75,000 8.1 2,0 0.83 Source: GGDP Annual Report, 1980, p.61 Row Planting In the early 1980s, two recommendations were given. On land free of stumps and projections, ropes were used to mark rows while the use of sighting poles (at least 3) was recommended for lands with stumps and projections. As at 1986,only sighting poles were recommended for making rows. One problem that arose was how to measure the distances in the field since farmers did not carry measuring tapes to their farms. A simple 41 University of Ghana http://ugspace.ug.edu.gh but very useful method was developed. The normal cutlass used by formers was used to measure the distance5. Fertilizer Application On taking off the GGDP embarked on some fertilizer trials using the existing fertilizers in the country. Between 1982 and 1984 it concentrated on on-farm timing and method of application. This was because farmers were not following the recommendation and usually waited until the emergence of the seedling. Table 3.13 shows some of the results obtained in 1982 and 1983. That experiment looked at applying starter fertilizer at planting or at 3 weeks after planting; burying it or placing it on the surface; using 15-15-15 (NPK) or tripple super phosphate (TSP) or sulphate of ammonia (S/A). The results show that there was no significant difference in yields between applying the starter fertilizer at planting time and delaying it until three weeks in the transition zone although it was significant in the forest zone. There was also no effect of placement of fertilizer on yield. Furthermore, there was a small but consistent advantage of applying phosphorus fertilizers compared with ammonium sulphate. On potassium, previous experiment had shown that the response of maize to it was very low or non­ existent in Southern and Central Ghana. The length of this cutlass was about 60cm. Using one-and-a-half times the length of the cutlass gave a distance of 90cm between rows and using two-thirds of the length gave die distance between plants in the row of 40cm apart. 42 University of Ghana http://ugspace.ug.edu.gh Table 3.13: Effect of Time, Placement and Type of Fertilizer on Maize Yield (in t/ha). Treatment 1983 Forest Transition Guinea savannah Time: At planting 3.46 3.68 2.33 NS6 NS At 3 weeks 3.34 3.87 2.38 Placement: Burried 3.42 3.92 2.27 NS NS NS Surface 3.38 3.62 2.44 Types: 15:15:15 (NPK) 3.54 4.14 2.56 Triple super­ phosphate 3.22 3.58 2.17* Sulphate of ammonia 3.45 3.60 2.34 Source: GGDP Annual Report, 1983, p. 74 These findings endorsed the farmer practices of applying compound fertilizer to the soil surface 2-3 weeks after planting. This avoided the loss of fertilizer if crop establishment foiled either because of poor seed or predator attack It also saved him some labour at planting time. Therefore it was recommended that farmers could apply starter fertilizer at planting or up to about 2 weeks after planting without yield loss. For fields that were recently in fallow, little or no fertilizer was required. Higher rates were recommended for more continuously cropped fields and for the guinea savannah zone because of low soil fertility. To further examine the effect of fertilizer on yield under farmers management, former-managed fertilizer experiments were conducted in 1985 in the forest, transition and savannah zones. In these trials, two plots were demarcated. All decisions on the NS =not significant 43 University of Ghana http://ugspace.ug.edu.gh maize variety to plant and all other production decisions were made by the former. The only difference between the two plots was that, on one, fertilizer was added at a rate of 250 kg/ha of compound and sulphate of ammonia. The other plot received no fertilizer. The results from those trials are presented in Table 3.14. The yield increase due to fertilizers was between 0.6t/ha in the forest zone and 0.9t/ha in the Guinea Savannah zone. Again, this shows that the Guinea Savannah is more responsive to fertilizer. Table 3.14: Maize Grain Yield from Farmer-Managed Fertilizer Trials (in t/ha). Treatment Ecologica Zone Across sites Forest Transition Savannah With fertilizer 1.97 1.61 2.03 1.89 Without fertilizer 1.37 0.88 1.12 1.11 Difference 0.60 0.73 0.91 0.78 Source: GGDP Annual Report, 1985, p. 137. The Maize Recommendations Putting all the agronomic trials together and subjecting them to economic analysis, maize production recommendations were made for Ghana. The maize production options are presented in Table 3.15. Option A is recommended for fields that have low soil fertility and option B for moderate soil fertility. These options were demonstrated to several thousands of farmers throughout Ghana by MOFA, GLDB and SG 2000. 44 University of Ghana http://ugspace.ug.edu.gh Table 3.15: Demonstrated Technological Options for Maize Production in Ghana Practice Option A Option B Variety Okomasa, Dobidi,Obatanpa or other improved varieties Okomasa, Dobidi, Obatanpa or other improved varieties Planting density 56,000 plants/ha for Okomasa, Dobidi; 62,500 plants/ha for Obatanpa and other early varieties 56,000 plants/ha for Okomasa, Dobidi; 62,500 plants/ha for Obatanpa and other early varieties Sowing method Rows using sighting poles Rows using sighting poles Between - row 90cm for Okomasa and 90cm for Okomasa and distance Dobidi; 80cm for Obatanpa and early varieties Dobidi; 80cm for Obatanpa and early varieties In -row distance 40cm 40cm Seeds per hill 2 2 Starter fertilizer 38:38:38fNPK) 19:19:19(NPK) Starter fertilizer application time At planting At planting Side-dressing with nitrogen 53:0:0 26:0:0 Side-dressing application time 4 weeks 4 weeks Hand weeding Twice at 3 and 6 weeks Once at 4 weeks. Source: Maize and Legumes Production Guide. GGDP, 1996. pp.5-16. Chapter Summary The chapter has shown that various institutions have collaborated in maize research in Ghana. In a well-planned research and extension activity, maize recommendations were developed and extended to farmers. This was to help increase maize production in Ghana. The chapter has also shown the efforts made by biological scientists in the development and promotion of maize technologies in Ghana. 45 University of Ghana http://ugspace.ug.edu.gh CHAPTER 4 THE PRODUCTION ENVIRONMENT Ecological zones Maize is the most important cereal crop in Ghana. It is grown throughout the country, with the exception of a small portion of the north-east characterized by harsh climatic conditions and extremely rocky soils. Generally speaking, the climate in Ghana grows hotter and drier as one moves northward and inland away from the Atlantic coast. Maize cropping systems and production technologies vary between the four ecological zones in which significant amounts of maize are cultivated (Figure 4.1). The coastal savannah zone includes a narrow belt of savannah that runs along the Atlantic coast, widening toward the east of the country. Farmers in this zone grow maize and cassava as their principal staples, often intercropped. Annual rainfall, which is bimodally distributed, totals only 800 mm per year, so most maize is planted following the onset of the major rains beginning in March or April Soils are generally light in texture and low in fertility, and productivity is low. Immediately inland from the coastal savannah lies the forest zone. Most of Ghana's forest is semi-deciduous, with a small proportion of high rain forest remaining only in the southwestern part of the country near the border with Cote d'Ivoire. Maize in the forest zone is grown in scattered plots, usually intercropped with cassava, plantain, or cocoyam as part of a bush fallow system Some maize is consumed in the forest zone, but it is not a leading food staple, and much of the crop is sold. Annual rainfall averages 46 University of Ghana http://ugspace.ug.edu.gh CO TE D' I VO IR E Figure 4.1: Ecological zones of Ghana BURKINA FASO darvSjavannah; Guinea Savannah 60 60 126 Kilometers 47 TO G O University of Ghana http://ugspace.ug.edu.gh about 1,500 mm and maize is planted both in the major rainy season (beginning in March) and in the minor rainy season (beginning in September). Moving farther north, the forest zone gradually gives way to the transition zone. The exact boundary between the two zones is subject to dispute, which is not surprising considering that the boundary area is characterized by a constantly changing patchwork of savannah and forest plots. What is certain, however, is that the transition zone is an important region for commercial grain production. Much of the transition zone has deep, friable soils, and the relatively sparse tree cover allows for more continuous cultivation and greater use of mechanized equipment. Rainfall is bimodally distributed and averages about 1,300 mm per year. Maize in the transition zone is planted in both the major and minor seasons, usually as a monocrop or in association with yam or cassava. The Guinea Savannah zone occupies most of the northern part of the country. Annual rainfall totals about 1,100 mm and falls in a single rainy season beginning in April or May. Sorghum and millet are the dominant cereals in the Guinea Savannah, but maize grown in association with small grains, groundnut and/or cowpeas is also important. Some fields are prepared by tractor, but most are prepared by hand. Maize is grown in permanently cultivated fields located close to homesteads, as well as in more distant plots under shifting cultivation. Rainfall Maize is generally grown throughout Ghana under rain-fed conditions. Based on rainfall regimes, the country may be divided broadly into two main regimes. The southern part has two rainy periods and the northern half has one long rainy period. 48 University of Ghana http://ugspace.ug.edu.gh The major rainy season for the south fells between April and July and the minor season is from September to November. Maize is mostly grown in the major season in the south. However, in some parts of Ghana particularly in the transition and some forest ecological zones, minor season maize has grown in importance lately because of changes in the rainfall distribution. In the northern half of Ghana, the rainfall season is from May to September. Since maize is grown under rainfall conditions, the importance of rainfall cannot be overemphasised. It is thus important to analyse the rainfall distribution without and with the establishment the GGDP. This is done in an attempt to isolate the effect of rainfall on production with and without the GGDP research. In this respect two periods are compared. The period of 1967 to 1978 representing "without" the GGDP is compared with 1979 to 1990 period as “with" GGDP. Using data from the Meteorological Services Department, Accra, the mean monthly rainfall is compared for the two periods for some of the selected districts for the maize adoption survey on which rainfall data were available. The main task is to test whether the mean of casewise differences in monthly rainfall between the two periods (1967-1978 and 1979-1990) differs from zero. Using the paired samples t-test procedure, the mean monthly rainfalls of the two periods are compared. The results are presented in Table 4.1. 49 University of Ghana http://ugspace.ug.edu.gh Table 4.1: Comparison of Mean Monthly Rainfall 1967-1978 and 1979-1990 Periods for Some Selected Districts in Ghana. Survey district Nearest meteorological station Mean monthly rainfall (mm) Paired difference t Significance (2-tailed) 1967-1978 1979-1990 Wa Wa 86.9 77.3 9.6 1.454 .174 Damango Dmango 97.1 105.1 -8.0 -0.487 .636 Salaga Salaga 88.1 97.2 -9.1 -0.815 .433 Nkoranza Ejura 115.3 108.7 6.6 0.760 .463 Amansie West Kumasi 116.4 107.0 9.4 0.891 .392 Adansi East Bekwai 108.3 94.1 14.2 1.437 .179 West Akim Prankese 124.0 94.1 29.9 1.722 .113 Adidome Ohawu 75.5 72.0 3.5 0.455 .658 Tema Tema 62.9 51.8 11.1 1.291 .222 Agona Agona Swedru 102.1 93.7 8.4 1.093 .298 Jasikan Hohoe 138.4 122.5 15.9 2.502 .029 Source: Estimated by the author based on data from the Meteorological Services, Accra. The probability values (p value) associated with the t-statistics in Table 4.1 show that apart from Hohoe which had significantly higher mean monthly rainfall in the 1967- 1978 period than the 1979-1990 period, for all the other districts, the mean monthly rainfall was not significantly different between the two periods. This means that the mean monthly rainfall for the periods 1967-1978 and 1979-1990 are apparently the same. The results thus seem to suggest that even if rainfall is an important factor in maize 50 University of Ghana http://ugspace.ug.edu.gh production in Ghana, it has basically played an unbiased role for maize production in the period prior to the establishment of the GGDP (1967-1978) and with the GGDP existence (1979-90). Inputs In maize factor of production trial conducted on-farm by the GGDP in 1980, fertilizer contributed 54% to yield, improved variety was 21%, appropriate plant density 18% and weed control 7%. Fertilizer and seed are thus important inputs in maize technology development. It is thus imperative that their availability, distribution and use are analysed for purposes of policy development. Seed Production and Distribution The seed industry is crucial to agricultural development in Ghana, and seed is one of the important forms in which technology can easily be transferred to farmers. The Seed Multiplication Unit of MOFA used to produce and market different kinds of seeds including maize before 1979. In 1979, however, the Ghana Seed Company (GSC) was established as a parastatal to take over the functions of the Seed Multiplication Unit. During the period of 1979 to 1989, the GSC became a major producer and marketer of certified seed in Ghana. The GSC received maize breeder seed from the GGDP of CRI. It, in turn, produced foundation seed which was then given to contract growers to produce certified maize seeds. The channels of distribution and sale of GSC seed were direct sales through their offices, sales kiosks, mobile sales, commission agents and extension agents from the Ministry of Food and Agriculture. The Grains and Legumes Development Board also 51 University of Ghana http://ugspace.ug.edu.gh produced some foundation and certified seeds, side by side, with GSC. It also obtained its breeder seed from the GGDP. la 1989, as a result of Ghana Government policy favouring privatisation of the seed industry, the GSC was closed down. Since 1989, the Ghana Government has restructured the whole Ghana seed industry. Breeder seed is produced by CRI and made available to GLDB for foundation seed production. Private seed growers then produce certified seeds that were sold either directly to fanners or through seed dealers. The Ghana Seed Inspection Unit (GSIU) and the National Seed Service (NSS) provide technical support of quality control and certification. Other international bodies such as the United States Agency for International Development (USAID), SG2000, International Fund for Agricultural Development (IFAD) and the World Bank have in various ways supported the whole seed industry. The establishment of the seed industry has provided employment for entrepreneurs and enterprises engaged in seed production and distribution. For example, the number of seed growers and dealers has increased since 1990. The number of growers has more than doubled from 1990 to 1996, and the number of dealers has increased sixteen folds since 1991(Table 4.2). These seed dealers are the retailers of improved Table 4.2: Improved Maize Seed Growers and Dealers in Ghana, 1990-1996 Year Number of growers Number of dealers 1990 52 0 1991 65 10 1992 80 80 1993 91 96 1994 108 0 1995 120 160 1996 120 160 Source: Compiled by the author from the National Seed Service records, Pokuase, Ghana. 52 University of Ghana http://ugspace.ug.edu.gh maize seed and they are found in most areas of Ghana. The quantity of certified seeds sold by erstwhile GSC, GLDB and seed growers since 1979 are presented in Table 4.3. Between 1979 and 1983, the main maize variety sold was La posta. Since 1984, the newer varieties such as Dobidi, Okomasa and Abeleehi developed and released by GGDP have been some of the major varieties officially on sale. Table 4.3: Certified Seed Sales in Ghana (in mt) Year GSC GLDB Seed growers Total Estimated area under cultivation (ha) 1979 992 25 0 1017 45200 1980 893 9 0 902 40089 1981 617 46 0 663 29467 1982 711 16 0 727 32311 1983 196 18 0 214 9511 1984 55 26 0 81 3600 1985 41 75 0 116 5156 1986 145 86 0 231 10267 1987 209 86 0 295 13111 1988 227 145 0 372 16533 1989 0 157 0 157 6978 1990 0 145 0 145 6444 1991 0 0 297 297 13200 1992 0 0 326 326 14489 1993 0 0 475 475 21111 1994 0 0 816 816 36267 1995 0 0 1082 1082 48089 1996 0 0 713 713 31689 Source: Compiled by the author from GSC, GLDB Warehouse and National Seed Service records. In the period between 1984 and 1995 a total of 4,393 metric tons of certified seed were sold to formers covering a cumulative area of 195,244 hectares. Thus, on the 53 University of Ghana http://ugspace.ug.edu.gh average 16,270 hectares were under improved maize over the 12 year period (1984 - 1995). MOFA has tried to encourage farmers to use improved seeds through seed promotional programmes since 1990. The Ministry advertises yearly in the daily newspapers the names of growers and dealers, their addresses, location and varieties of certified maize seed for sale. It also earmarks a day in each year as seed promotion day to remind farmers of the beginning of the maize growing season. As part of seed promotion, MOFA develops fact sheets about seed maize for interested people. The GSHJ further organises field days on seed growers' farms for neighbouring farmers. Certified seed sold by the seed growers and dealers are probably just a small proportion of the total improved maize grown in the country1. An important point to note is that a conscious effort has been made to make quality improved seeds available for farmers. Fertilizer Most of the fertilizers imported into Ghana are used in maize production. Other crops that use fertilizers are rice, cotton and some vegetables. The use of fertilizers in Ghana is generally low. In 1990, for example, Ghana used less than 5 kilograms per hectare of plant nutrients that only replenished about one-seventh of the plant nutrients .For example, in a maize seed management study in the Brong-Ahafo and Volta Regions of Ghana, 78% and 76% of farmers between 1992-94 farm-saved their seeds. (Tripp 1996. DRAFT REPORT). Of the fanners who acquired seed off-farm only 15% and 18% of the fanners in Brong-Ahafo and Volta Regions respectively purchased seed from a formal source between 1992 and 1994. A seed rate of 22.5kg/ha was used to estimate the area under certified maize seed 54 University of Ghana http://ugspace.ug.edu.gh removed by crops2. One can attribute the slow growth in fertilizer use to macro-economic policy in Ghana. Marketing and distribution MOFA was the sole importer of fertilizer in Ghana until 1988 when it distributed its fertilizer through its sales outlets dotted throughout Ghana. The system of marketing and distribution created many problems. In particular, delays in Government funding, untimely delivery of fertilizers to formers, and ineffective physical distribution at the district and sub-district levels became common. Other problems include the dual and conflicting role MOFA played in marketing fertilizer and carrying out extension activities. In 1988, Ghana Government privatised the fertilizer trade apparently to mitigate some of the problems mentioned earlier. The marketing and distribution of fertilizer have since been privatised. Trends in fertilizer imports In Ghana, fertilizers that are recommended for maize cultivation are ammonium sulphate, urea and compound fertilizer of nitrogen-phosphorus-potasium (NPK) combination. Since these are the recommended ones, analysis of the study is based on them Bumb B.L., J.F. Teboh, J.K Atta and W.K. Asenso-Okyere (1994).11 Ghana Policy Environment and Fertilizer Development.” IFDC. 1994. P.25 55 University of Ghana http://ugspace.ug.edu.gh The bulk of fertilizer consumed in Ghana is in the Guinea Savannah and the transition zones because of poor soils, as well as intensive and continuous cropping. Imports of fertilizers in Ghana since 1970 are presented in Table 4.4. Table 4.4: Fertilizer Imports to Ghana since 1970 (in mt) Year Ammonium sulphate Urea 15-15-15 (NPK) 20-20-0 (NPK) Total 1970 3140 3 3671 1087 7901 1971 1943 3 4923 544 7413 1972 3852 33 5787 2085 11757 1973 8100 8 14278 3418 25804 1974 4150 10 6000 1000 11160 1975 2258 3 16075 1475 19811 1976 2557 0 36626 4218 43401 1977 2900 0 4200 4000 11100 1978 13739 0 15896 6180 35815 1979 19000 0 39650 0 58650 1980 17980 0 39600 0 57580 1981 - 0 0 0 0 1982 14000 0 20000 8000 42000 1983 0 0 0 0 0 1984 13600 200 24550 0 38350 1985 5437 0 21362 3200 29999 1986 8500 0 0 9600 18100 1987 14650 0 4800 17620 37070 1988 20550 0 9400 3300 33250 1989 25711 6015 8000 15500 55226 1990 2000 20100 8500 34850 44950 1991 0 0 0 0 0 1992 11538 0 9575 4100 25213 1993 7600 0 10000 0 17600 1994 0 0 9040 0 9040 1995 8800 0 13040 0 21840 Source: Compiled by the author from MOFA records andBumb et al, 1993, pp38-40. Fertilizer imports are somewhat indicator of fertilizer consumption in Ghana. The increase in fertilzer imports and consequently use in the 1970s were the results of the rice boom in Northern Ghana and irrigated lands that were developed at that time. This was 56 University of Ghana http://ugspace.ug.edu.gh due to the Ghana Government’s " Operation Feed Yourself' (OFY) and “Operation Feed Your Industry" (OFYI) pohcies which placed more emphasis on food and industrial crops production. Another contributing factor was that fertilizers were heavily subsidised. Trends in fertilizer prices The pricing policies in Ghana in the seventies were to keep fertilizer prices low and constant over some time (1970 to 1976) so that formers could plan ahead and use them. For example, the level of subsidy on fertilizer was 80% in 1979 as in Table 4.5. Table 4.5: Nominal Fertilizer Prices and Subsidy Rates in Ghana Year Price (cedis/50kg bag) Subsidy rate (%) Ammonium sulphate 15-15-15NPK 1970 2.0 2.8 Not available (n.a.) 1971 2.0 2.8 n.a. 1972 2.0 2.8 n.a 1973 2.0 2.8 n.a 1974 2.0 2.8 n.a 1975 2.0 2.8 n.a 1976 2.0 2.8 n.a 1977 5.0 6.5 n.a 1978 6.0 7.5 n.a 1979 8.5 10.0 80 1980 12.0 15.0 65 1981 25.0 30.0 45 1982 25.0 30.0 45 1983 45.0 58.0 45 1984 295.0 440.0 45 1985 295.0 440.0 59 1986 490.0 780.0 36 1987 820.0 1380.0 42 1988 1600.0 2300.0 30 1989 2350.0 3350.0 15 1990 3100.0 4200.0 0 1991 3500.0 6000.0 0 1992 6000.0 8500.0 0 1993 7500.0 10000.00 0 Source: MOFA, Accra., Bumb et a! 1993. p.40. 57 University of Ghana http://ugspace.ug.edu.gh The price of fertilizer remained constant from 1970 to 1976 after which several drastic adjustments in prices were made. Several factors have accounted for these instabilities. One of them was a change in pricing policy that gradually removed subsidies on fertilizer from 80% in 1979 to zero percent in 1990. Consequently, fertilizer prices rose sharply. Furthermore, the exchange rate policy adversely affected fertilizer prices. As part of the Ghana Government's Economic Recovery Programme (ERP), the value of the Ghanaian Cedi was allowed to be market-determined. Thus the value of the cedi changed from 2.75 cedisAJSSl in 1980 to 350 cedis/US$l in 1990. In 1996, it rose to 1700 cedis/US$l. This led to increased import costs that were subsequently passed on to formers. Fertilizer and maize prices In order that the real cost of fertilizer to the farmer is understood, it is necessary to compare maize and fertilizer prices. This is because fertilizer and maize prices have both changed over time. The economic reform and privatization programme in 1983 led to the removal of subsidies on agricultural inputs. For example, subsidies on fertilizer were reduced from 80% in 1979 to zero in 1990. This policy induced increases in fertilizer prices (Table 4.6). The nitrogen-maize price ratio has risen from 0.6 in 1979 to 11 in 1997 (Figure 4.2). This policy has made fertilizer more expensive. 58 University of Ghana http://ugspace.ug.edu.gh Table 4.6: Fertilizer Prices and Subsidies in Ghana Year % subsidy Cost for recommended rates of compound and ammonium sulphate fertilizers (cedis/ha) ___________With subsidy________ Without subsidy 1979 80 46 83 1980 65 68 112 1981 45 138 200 1982 45 138 200 1983 45 258 374 1984 45 1838 2665 1985 59 1838 2922 1986 36 3175 4318 1987 42 5500 7810 1988 30 9750 12675 1989 15 14250 16388 1990 0 18250 18250 Source: Compiled by the author from MOFA records, Accra. Figure 4.2: Nitrogen-to-Maize Price Ratio, Ghana (1979-1997) N <0 _ E .2iO « ** >- ■c O)o Year Source: Updated MOFA data by the author. The real cost of fertilizer has changed considerably since 1970. The quantity of fertilizer that 1kg of maize can buy in the period, 1970 to 1996 are shown in Figure 4.3. 59 University of Ghana http://ugspace.ug.edu.gh Figure 4.3: Kilograms of Ammonium Sulphate Fertilizer that 1kg of Maize can buy in Ghana (1970-1996) 1975 1980 1985 1990 1995 Year 45 40 35 30 25 20 15 10 5 0 1970 Source: Estimated by the author from MOFA data. The quantity of fertilizer that a kilogram of maize could buy increased between 1970 and 1976. It further fluctuated but with increasing trend between 1977 and 1983. Since 1984, the amount of sulphate of ammonia that maize can buy has continuously decreased. This means that a farmer can now buy less of fertilizer for the same amount of maize. It again shows that the real cost of fertilizer has increased since 1984 and former’s capacity to buy fertilizer has decreased. For example, farmers could buy about 3kg of ammonium sulphate in 1970, 17kg in 1980, but 0.7kg of sulphate of ammonia in 1993 for the same lkg of maize. This increase in real cost of fertilizer means a reduced profitability in relative terms. This trend of affairs has worked against formers. 60 University of Ghana http://ugspace.ug.edu.gh Chapter Summary The chapter has described the production environment such as the ecological zones, rainfall and input supply for maize production in Ghana. Maize can be produced in all the four ecological zones. Rainfall is an important factor in the production of maize in Ghana but it has played an unbiased role for maize production prior to the establishment of the GGDP and with the existence of the GGDP. Inputs such as improved seeds and fertilizer have been made available to facilitate the adoption of the improved maize technologies. 61 University of Ghana http://ugspace.ug.edu.gh CHAPTER 5 RESEARCH METHODOLOGY Data collection activities A national survey of farmers was conducted from November 1997 through March 1998 by CRI in collaboration with CIMMYT, CIDA and Overseas Development Institute (ODI) to assess the adoption and impact of GGDP-generated maize technologies. The author participated in this survey. Unlike some maize adoption studies carried out in Ghana in 19871 and 19912 which focused on selected districts which were not necessarily representative of the whole country, this study sought to carefully sample the entire maize-producing population to generate information at the national level. The survey sample was drawn using a three-stage randomised selection procedure. Firstly, 20 agricultural districts were selected at random from among all the agricultural districts in Ghana using random numbers from a statistical table3. Since adoption rates are measured in terms of area under production in the assessment of economic impact of agricultural research, this production parameter is used as the main criterion for selecting the districts4. The probability of the selection of each district was proportional to the area planted to maize in the district. This self-weighting procedure led 1 GGDP. “A Study of Maize Technology in Ghana.” Mexico, D.F, GGDP. 1991. pp 3-4, Tripp R., K. Mario, A A Dankyi and M Read “Changing Maize Production practices o f Small-scale Fanners in the Brong-Ahafo Region, Ghana.” Ghana Grains Development Project Kumasi. 1987. pp5-7. Clarke G.M. and D. Cooke. "A Basic Course in Statistics" 2nd ed. Edward Arnold (publishers) Ltd. 1983. pp 404-405. This production parameter is based on data from the "Annual Sample Survey o f Agriculture, Ghana, 1996," o f the Policy Planning Monitoring and Evaluation Division o f the Ministry o f Food and Agriculture (MOFA), Accra. 1996. Various pages. 62 University of Ghana http://ugspace.ug.edu.gh to the selection of districts located in nine of the ten regions of Ghana and across all ecological zones (See Tables 5.1, 5.2 and Figure 5.1) Secondly, because maize production activities are carried out largely in rural and semi-urban areas, within each selected district, three enumeration areas (EAs) were chosen randomly from among all the EAs classified as rural or semi-urban.5 The advantage of using EAs instead of villages is that EAs contain roughly the same number of inhabitants, which removes the need to adjust for differences in village size. Thirdly, in each of the 60 selected EAs, seven farmers were picked at random from a sampling frame consisting of a list of all maize farmers in the EA. This procedure produced a total sample size of 420 maize farmers. The sampling procedure for the maize survey is summarised in Table 5. 1. Table 5.1: Sampling Procedure for Maize Technology Survey in Ghana Sampling stage Sampling unit Selection criterion Number at this unit Cumulative number 1 District Randomly selected. The probability of selection is proportional to the area under maize production in the district 20 20 2 Enumeration area Randomly chosen from among enumeration areas classified as rural or semi- urban 3 60 3 Farmer Randomly selected from among all maize farmers in the enumeration area 7 420 Source: Compiled by the aut tor The Ghana Statistical Service classifies enumeration areas into rural, semi-urban and urban. 63 University of Ghana http://ugspace.ug.edu.gh Table 5.2: Location of Survey Districts for Maize Adoption Study in Ghana, 1997 District Ecological zone Region Wa Guinea savannah Upper West Salaga Guinea savannah Northern Damango Guinea savannah Walewale Guinea savannah Nkoranza Transition Brong Ahafo Dorma-Ahenkro Forest Amansie West Forest Ashanti Adansi East Forest Sekyere West Transition Adidome Coastal savannah Volta Jasikan Forest Tema Coastal savannah Greater Accra Yilo Krobo Transition Eastern Suhum Kraboa Forest Fanteakwa Forest West Akim Forest Agona Coastal savannah Central Gomua-Assin-Ajumako Coastal savannah Mpohor Wassa Forest Western Sefwi Wiawso Forest Source: Compiled by the author. 64 University of Ghana http://ugspace.ug.edu.gh CO TE D' I VO IR E 64 64 128 Kilometers. C rMMYT Figure 5.1: Location of Survey Districts, Ghana Maize Technology Adoption Study, 1997 BURKINA FASO 65 University of Ghana http://ugspace.ug.edu.gh The survey instrument The questionnaire deals with farmer’s household characteristics, maize variety use and production practices and maize utilisation patterns. Other sections deal with maize variety preferences, maize seed acquisition and management, impacts of modem variety adoption, and information about the respondent. The questionnaire administered to farmers is presented as Appendix 1. Economics o f maize production Crop production recommendations should be updated regularly because of the changing economic environment. The results from such assessment can be used to plan further research, modify recommendations to farmers and provide information to policy­ makers regarding current policy on input supply. Results from the GGDP on-farm experiments are updated with current input prices using a partial budget technique. The marginal rate of return is then estimated. The analyses are done on the individual elements from the technological package such as fertilizer and improved variety and on the whole package of improved maize, fertilizer and row planting. Since farmers usually take risks into consideration when adopting technologies, the results are subjected to risk analysis. The sensitivity of the results to changes in risk factors (yields and prices) is tested. Determinants o f adoption o f improved maize In determining the main factors affecting the adoption of improved maize technology in Ghana, the Logistic regression model and the TOBIT model are used. While the Logistic regression measures the probability of adoption, the TOBIT model not 66 University of Ghana http://ugspace.ug.edu.gh only measures the probability but also the intensity of use. TOBIT model further gives better results if the dependent variable is continuous. In this study, the dependent variable in the TOBIT model is continuous and the dependent variable in the Logistic regression is dichotomous. Further details of these models are given in Appendix 5. Returns to investment Returns to investment in agricultural research can be calculated using an economic surplus approach whereby research costs are related to the benefits (measured in terms of changes in producer and consumer welfare) that result from technological innovations generated by the research program The basic theoretical model is described in Alston et al6. Applying the economic surplus approach to a commodity research program, research-induced technological innovations are hypothesized to increase production per unit of input (reduce per unit production costs), resulting in an outward shift in the aggregate supply curve (cost curve) for the commodity. Assuming linear supply and demand curves, the outward shift of the supply curve leads to a reduction in the market- clearing equilibrium prices and generates gains in producer and/or consumer welfare, with the distribution of these gains determined by the slopes of the supply and demand curves. In other words, producers benefit from research through reduced production costs. Consumers, on the other hand, benefit from research by being able to consume more at a lower price. The benefits generated by a commodity research program can be depicted graphically as shown in Figure 5.2. As formers use improved technologies generated by 67 University of Ghana http://ugspace.ug.edu.gh the research program, the aggregate supply curve shifts outward (downward) from S 0 to S,. Any increase in economic surplus is a measure of social gains attributable to the new technology. The change in total surplus caused by the shift of the supply curve is area AEF minus area ABC, which is equal to area FEBC. Thus area FEBC represents the social gains derived from maize research. It is these gains that must be measured. In order to utilize this model, it is necessary to obtain empirical estimates for three key parameters: • The J-parameter (which represents the productivity increase attributable to research- induced technological change), • The K-parameter (which represents the net reduction in production cost attributable to the new technology), and • The I-parameter (which shows the increase in per unit input production costs). Alston M Julian, George W. Norton and Philip G, Pardey. “ Science Under Scarcity.” Published by Cornell 68 University of Ghana http://ugspace.ug.edu.gh Calculation of these three key parameters is described in detail in Appendix 2. In addition to calculating the three key parameters, it is also necessary to estimate the change in output attributable to research (dQ). In the case of commodity research programs, this is simply calculated as the product of the yield increase attributable to the adoption of improved technology (dY) multiplied by the area over which adoption takes place (dA), while holding input levels constant. In Ghana as elsewhere, observed market supply and demand curves do not reflect all of the costs and benefits associated with maize production. Government policies (producer price supports, fertilizer subsidies, exchange rate controls) give rise to externalities and transfers that are not reflected in market prices. For this reason, it is advisable to calculate both financial and economic rates of return. In the financial analysis, production inputs and outputs are valued using actual market prices (domestic prices). This is because financial analysis takes the viewpoint of the individual participant of the project. In the economic analysis, production inputs and outputs are valued using economic prices that reflect their true scarcity value (social prices). Economic analysis thus takes the whole society's point of view into consideration. Prices are thus adjusted to reflect the type of analysis used. In the economic returns, the world market prices adjusted to farm-gate price is used. Since fertilizer subsidies can be considered as transfer payments and are costs to the society, the subsidies have been removed to reflect the true cost to society. The effects of other costs such as seed and labour are small relative to fertilizer costs. The distortions in their prices will have very little effect on the gains from research and therefore, they have University of Ghana http://ugspace.ug.edu.gh A real exchange rate is used for the maize price and other foreign component costs of research in the economic rate of return estimation, The real exchange rate (RER) is estimated from the formula: RER = eP*/P where, RER= real exchange rate e = nominal exchange rate P* = foreign price (proxied by the consumer price index for the United States) P = domestic price (proxied by the consumer price index for Ghana) The US dollar is used as a proxy because of the role it plays as an intemational currency and as a result of the stable national economy of the US. The base period is taken to be 1997 and the estimated real exchange rates are shown in Table 5.3. Table 5.3: Exchange Rates, 1979 to 1997 Year Consumer price Consumer price Nominal Real exchange index,Ghana index,United exchange rate rate ,RER, (1997=100) States (1997=100) (e) (1997=100) 1979 0.31 45.24 2.75 401.32 1980 0.46 51.34 2.75 306.92 1981 1.01 56.63 2.75 154.19 1982 1.22 60.13 2.75 135.54 1983 2.72 62.00 8.83 201.27 1984 3.81 64.69 35.99 611.07 1985 4.19 67.05 54.37 870.05 1986 5.23 68.27 89.20 1164.38 1987 7.31 70.79 162.37 1572.39 1988 9.61 73.64 202.35 1550.58 1989 12.03 77.22 270.00 1733.12 1990 16.50 81.37 326.33 1609.30 1991 19.49 84.92 367.78 1602.46 1992 21.44 87.39 437.09 1781.59 1993 26.79 89.99 649.06 2180.25 1994 33.46 92.27 956.71 2638.24 1995 58.34 94.87 1200.4 1952.04 1996 78.19 97.64 1637.2 2044.46 1997 100 100 2050.2 2050.20 Source: Intemational Financial Statistics, 1998 pp.438-441; 892-893 and estimates by the author. 70 University of Ghana http://ugspace.ug.edu.gh Tti the financial rate of return, domestic maize price and official exchange rates are used for cost estimation. Subsidies on fertilizers are also included. The world market price for maize is used for the economic rate of return. Since this is to reflect the price the society is paying, 30% transport and 20% marketing costs have been added to the Freight-On-Board (F.O.B) maize price. The farm-gate price for the world price is estimated as 70% of the landed price and this is used as the economic maize price to the firmer (see Appendix 3 for details). Key assumptions and critical parameters Key assumptions made in estimating the theoretical model are briefly described in the scenarios presented below. More detailed explanations of how values were calculated for critical parameters appear in Appendix 3. In Appendix 4, the details of the derivation of the various formulae used to estimate the returns to investment are presented. Baseline scenario (Scenario 1) Research phase: Maize research under the GGDP was initiated in 1979. The first improved maize variety by the GGDP was released in 1984, together with fertilizer application rates and row planting recommendations. The research phase thus lasted from 1979 to 1983, a period of five years. Adoption phase: Adoption of the GGDP-generated technologies began in 1984. The diffusion path was estimated based on the results of various adoption studies carried out during the course of the project. The diffusion path was modelled as a smooth logistic growth curve beginning in 1984. Based on historical adoption data, it is projected that complete adoption will be achieved after 30 years. 71 University of Ghana http://ugspace.ug.edu.gh Research costs: Maize research costs consisted mainly of GGDP costs. The rate of returns calculations also took into account modest research investments made by the Savannah Agricultural Research Institute (SARI), the National Agricultural Research Project (NARP), as well as direct project management costs incurred by CIMMYT. Under the baseline scenario, 90% of GGDP costs were assigned to maize research in accordance with the initial budget allocation of the GGDP. Extension costs: Extension cost for maize was borne mainly by MOFA MOFA extension works on several crops in Ghana but the Policy Planning Monitoring and Evaluation Division (PPMED) of MOFA keeps production records of about 6-8 crops. Therefore using the area under production for these crops, extension activities were assumed to be proportional to it. This assumption put maize extension cost between 20% and 27%. Thus, under the baseline scenario, 30% of MOFA costs were assigned to extension activities on maize. In addition, the rate of return calculations took into account extension costs incurred by the Sasakawa Global 2000 Project. Productivity gains: Productivity gains attributable to research were estimated from experimental data showing the yield gains associated with adoption of the three GGDP-generated maize technologies (improved varieties, fertilizer, row-planting). In on- farm trials managed by farmers, yields obtained through the use of improved technologies were compared to yields obtained through the use of farmers’ own practices. The difference between the two (expressed in percentage terms) was attributed to the adoption of improved technologies. 72 University of Ghana http://ugspace.ug.edu.gh Financial prices: Financial price for outputs (maize grain) and key inputs such as fertilizer were collected through surveys and also taken from several official government publications.7 (See also Appendix 3) Economic prices: Economic prices were calculated when these were considered to cause economic profitability to diverge significantly from financial profitability. In particular, economic prices were calculated for maize grain and for fertilizer. Since the cost of other inputs such as seed used in maize production are small relative to the cost of fertilizer, policy-induced distortions in the prices of these other inputs had very little influence on profitability and therefore were ignored (see Appendix 7). Exchange rate: In the financial analysis, international prices were converted to domestic prices using the official exchange rate. In the economic analysis, international prices were converted to domestic prices using the real exchange rate (see Appendix 8). Alternative scenarios Sensitivity analysis was carried out to test whether the rates of return are sensitive to changes in the values of key parameters such as research and extension costs, and yield gains. Four additional scenarios were modelled. Scenario 2 (higher research costs) Since the GGDP supported work on crops other than maize, in calculating returns to investment in maize research it is necessary to decide what proportion of GGDP costs should be assigned to maize and what proportion to other crops. The decision is not entirely straightforward, because many GGDP expenditures cannot be linked to specific crops. Under Scenario 2, 100% of the GGDP research costs were assigned to maize. Policy Planning Monitoring and Evaluation Division, PPMED, (Statistics Division), MOFA, Accra Various Issues of Market Price Surveys. 1996 73 University of Ghana http://ugspace.ug.edu.gh These assumptions are very conservative, because clearly some portions of the GGDP funds were used to support work on cowpeas and soybeans. Scenario 3 (higher extension costs) Few MOFA extension agents work exclusively with maize; most work with maize along with a range of other crops. For this reason, just as it is necessary to decide how to assign research costs among crops, it is also necessary to dedde what proportion of MOFA extension costs should be assigned to maize and what proportion to other crops. Under the baseline scenario, 30% of MOFA extension costs were assigned to maize. Under Scenario 3, this proportion was increased to 50% to reflect the feet that many MOFA extension agents in fact dedicate significantly a little more time to maize than to other crops. Scenario 4 (higher research costs and higher extension costs) To be even more conservative, 100% of the GGDP research costs and 50% of MOFA extension costs were assigned to maize. These represent extreme cases of research and extension expenditures on maize. This scenario assumes that all GGDP research costs were assigned to maize and that half of all MOFA extension costs were on maize. Scenario 5 (lower projected research benefits) In the baseline scenario, the productivity gains attributed to the adoption of improved technologies were set equal to the yield gains achieved in on-farm yield trials through use of improved varieties, fertilizer, and row planting. These productivity gains were relative to the yields achieved with local varieties and using farmers' management practices. Since all of the farmers who adopt the improved technologies are unlikely to achieve the same yields as those who participated in the on-farm trials, under Scenario 5 74 University of Ghana http://ugspace.ug.edu.gh the projected yield increases attributable to adoption of the improved technologies were reduced by 40%8. The details of the scenarios are presented in Table 5.4. Table 5.4: Research and Extension Costs Scenarios for Ghana, 1997 Scenario Research and extension costs Scenario 1 GGDP cost (actual) = 90%, MOFA cost (actual) = 30% Scenario 2 GGDP cost = 100%, MOFA cost = 30%. Scenario 3 GGDP cost (actual) = 90%, MOFA cost = 50%. Scenario 4 GGDP cost = 100%, MOFA cost = 50%. Scenario 5 Same as scenario 1 except that yield gain is reduced by 40%. Source: Compiled by the author. Chapter Summary A formal survey with prepared questionnaire is used to obtain information from randomly selected farmers for the adoption rates of improved maize technologies. Furthermore, the economics of maize production technologies are analyzed. The information obtained is used for planning research and modifying maize recommendations. The determinants of adoption of improved maize are estimated with the Logistic regression and the TOBIT models. The returns to investments in research and extension activities are estimated with the economic surplus method. CIMMYT, 1988. Op. Cit. pp.23-25. (CIMMYT recommends 5% to 30% yield adjustment to compensate for the difference between experimental yield and the yield fanners could expect from the same treatment). 75 University of Ghana http://ugspace.ug.edu.gh THE ADOPTION OF MAIZE TECHNOLOGIES IN GHANA CHAPTER 6 Demographic characteristics o f the farmers The demographic information about the farmers interviewed is presented in Table 6.1. The data are broken down into ecological zones to highlight possible geographic and demographic indicators that might affect farmers’ acceptance of improved maize technologies. Table 6.1: Demographic Characteristics of Maize Farmers, Ghana. Ecological zone Years of schooling (years) Average age (years) Average household size Gender of respondents Residence status (N) Male (%) Female (%) Native (%) Settler (%) Forest 6.7 44 8.0 70 30 55 45 189 Transition 6.5 45 9.8 65 35 90 10 63 Coastal Savannah 6.3 47 9.7 71 29 73 27 84 Guinea Savannah 2.3 41 15.4 98 2 74 26 84 All zones 5.7 44 10.1 75 25 68 32 420 Source: Author’s computations based on the!998 CRI/CIMMYT survey. On the average, farmers interviewed have less than seven years of formal education. Farmers in the Guinea Savannah Zone have the least number of years in formal education. The mean age of the farmers is forty-four years and this is within the age group of Ghana’s labour force. Another important note is that female farmers form a quarter (25%) of the survey respondents. However, there is a considerable variability in the proportion of female farmers between ecological zones. These range from as low as 2% in the Guinea Savannah zone to 35% in the transition zone. Restrictions on women’s access to land are common in the Guinea Savannah Zone, where inheritance is patrilinear and this 76 University of Ghana http://ugspace.ug.edu.gh could be one of the explanations for low representation of females in the sample from the Guinea Savannah. In Table 6.2, information on farmers’ access to infrastructure such as education, good drinking water and health services are presented. The data on infrastructure is important because infrastructure may affect the flow of information, goods and services that in turn may influence the adoption of improved technologies. The data in Table 6.2 shows that the Guinea Savannah Zone has the least of the infrastructure particularly electricity and health services. Farmers in the transition zone have better access to infrastructure. In general, however, the coastal savannah has the best of the infrastructure. Table 6.2: Access to Infrastructure by Survey Household Zone Percent of farmers who live in a village with: Elementary school Electricity Health post Pipe- borne water Tarred road Easy transport­ ation Market Forest 100 19 30 41 15 41 33 Transition 100 22 44 44 44 56 22 Coastal Savannah 100 50 33 67 58 92 46 Guinea Savannah 83 0 8 50 17 33 46 All zones 97 22 28 48 28 52 34 Source: Author’s computations based on thel998 CRI/CIMMYT survey Adoption o f maize technologies Three main recommendations whose adoption rates were measured are improved varieties, fertilizer and row planting. The recommendations have been widely accepted. Farmers in the ecological zones who are using one or more of the maize technologies on part of their firm in 1997 crop season are presented in Table 6.3 77 University of Ghana http://ugspace.ug.edu.gh Fifty-four percent of all farmers plant improved varieties and a similar proportion (54%) plant at least part of their maize fields in rows. The rate of fertilizer use is however low. Less than a quarter (21%) of the farmers use fertilizers. As explained earlier, the economic reform policy of the Ghana government made fertilizer more expensive and this made some farmers cease utilizing fertilizer. Furthermore, male fanners have adopted the improved technologies more than their female counterparts in all the ecological zones. Table 6.3: Adoption of Maize Technologies in Ghana, 1997 Zone Percent of fanners using techno ogies Improved variety Rowp anting (a Fertilizer Male Female Total Male Female Total Male Female Total Coastal savannah 42 27 69 31 22 53 24 5 29 Forest 23 15 38 23 16 39 8 1 9 Transition 41 27 68 35 24 59 24 5 29 Guinea savannah 40 26 66 43 28 73 30 6 36 All zones 54 54 21 (a) Row planting excludes ridge planting Source: Author’s computations based on the 1998 CRI/CIMMYT survey In Table 6.4, the percentage of farmers who have stopped using fertilizer (31%) is far more than formers who have stopped using improved varieties (9%) and row planting (13%). Table 6.4: Disadoption of Maize Technologies in Ghana Improved variety Fertilizer Row planting Number of farmers ever used recommendation 250 128 239 Number of fanners using in 1997 227 88 208 Difference 23 40 31 Percentage of those stopping technology use 9.2% 31.3% 13% Source: Author’s computations based on the 1998 CRI/CIMMYT survey 78 University of Ghana http://ugspace.ug.edu.gh The adoption rates of maize technologies (in terms of number of fanners using the technology) vary among the ecological zones. Generally, the adoption rates of all the three technologies are lowest in the forest zone. The adoption of technologies can also be measured based on the area under the technologies. In Table 6.5 the percent of maize land area planted with improved varieties, row planted and fertilized in both the major and minor seasons are shown. In general, the adoption is high in the coastal savannah, transition and the Guinea Savannah Zones; and low in the forest zone. It is interesting to note that the proportion of the area planted to the improved varieties is identical to maize farmers who have adopted the improved technologies (54%). Table 6.5: Proportion of Area under Maize Technology Adoption in Ghana, 1997 Zone Improved variety Row planting Fertilizer (%) (%) (%) Coastal Savannah 74 59 38 Forest 33 44 9 Transition 68 66 42 Guinea Savannah 60 68 32 All zones 54 55 26 Source: Author’s computations based on the 1998 CRI/CIMMYT survey Among the improved varieties released by the GGDP, the most popular is Obatanpa. In 1997, Obatanpa accounted for 16% of Ghana’s total maize area. In Table 6.6, the area planted to specific maize varieties in 1997 is presented. A significant proportion of the area is planted with improved varieties (18.5%) whose identities are not known and are simply called “Agric” by the farmers. The area under these specific varieties in the ecological zones is further presented in Appendix 17. 79 University of Ghana http://ugspace.ug.edu.gh Table 6.6: Area Planted to Specific Maize Varieties in Ghana Variety Year released Major season (ha) Minor season (ha) Total (ha) Total (%) Improved varieties: “Agric” Unknown 103.0 16.6 119.6 18.5 Obatanpa 1992 80.1 22.5 102.6 15.9 Dobidi 1984 33.6 7.8 41.1 6.4 Aburotia 1984 17.6 5.4 23.0 3.6 Abeleehi 1990 13.0 7.8 20.8 3.2 La Posta Pre-1980 19.6 0.4 20.0 3.1 Okomasa 1988 16.6 2.4 19.0 2.9 Golden Crystal 1984 0.8 0.8 1.6 0.2 Dorke SR 1990 0.2 0 0.2 0 All Improved varieties 284.4 63.4 347.8 53.8 Local varieties 246.9 51.0 297.9 46.1 Total 531.4 114.4 645.8 100 Source: Author’s computations based on the 1998 CRI/CIMMYT survey Fanners are generally known to adopt technologies in a step-wise manner and are selective of the available technologies depending on the farmers’ circumstances. The interactions in maize technologies among maize formers are presented in Table 6.7. Table 6.7: Interaction among Maize Technologies in Ghana, 1997 Practice Percent of farmers who: Planted improved variety Planted local variety Applied fertilizer No fertilizer Applied fertilizer No fertilizer Row planting 12% 23% 5% 11% Random planting 1% 10% 0.5% 38% Note: n = 392 (excludes ridge planting) Source: Author’s computations based on 1998 CRI/CIMMYT survey Only 12% of the formers adopted the three technologies at a time. The common combination of practices among these farmers is the use of inproved varieties and row planting. One of the explanations for the low adoption of all three technologies (improved variety, fertilizer and row planting) might be the cost involved especially with the use of 80 University of Ghana http://ugspace.ug.edu.gh fertilizer. The reason for adopting improved variety and row planting together by these farmers might be the way improved maize technologies were introduced to farmers in the past through technological packages. Another reason could be the relative low cost involved with improved maize-row planting combination while at the same time ensuring good plant density. Although maize recommendations by the GGDP are made on the basis of ecological zones, Ghana’s agricultural activities are based on political boundaries rather than on ecological zones. Frequently, policy makers want to know the adoption rates within the regions of Ghana. For this reason, the adoption of the improved technologies in Ghana1 is presented according to regions as in Tables 6.8 (and Figure 6.1), 6.9 (and Figure 6.2), and 6.10 (and Figure 6.3). Table 6.8: Adoption oflmproved Varieties in the Political Regions of Ghana Region % of sample within region % of sample within total Western 26.8 2.6 Eastern 48.8 9.8 Volta 47.6 4.8 Ashanti 59.7 8.8 Brdng Ahafo 54.8 5.5 Central 52.4 5.3 Greater Accra 85.7 4.3 Northern 58.7 8.8 Upper West 85.7 4.3 (Total) 54.2 54.2 Source: Author’s computations based on the 1998 CRI/CIMMYT survey In Table 6.8, the percent of farmers who have adopted the improved technologies are presented. This is also shown in Figure 6.1. In Tables 6.9 and 6.10, the adoption rates None of the districts in the Upper East fell into the sample apparently because not much maize is grown there. 81 University of Ghana http://ugspace.ug.edu.gh for row planting and fertilizer are respectively presented. These are also depicted in Figures 6.2 and 6.3 respectively. The Western Region has the least adoption rates of improved varieties, row planting and fertilizer use (see Tables 6.8, 6.9, and 6.10). This is not surprising since the region is basically a forest zone where tree crops are predominant. The Ashanti, Volta, Eastern and Brong Ahafo Regions are of more than one ecological zone of transition and forest zones. The adoption of improved varieties in these regions is between 48% and 55%. The Greater Accra region is a coastal savannah while the Northern and Upper West Regions fill basically in the Guinea Savannah Zone. The adoption rates in these regions are fairly high compared to other regions. These are similar to the results of the analysis based on ecological zones for coastal savannah and Guinea Savannah. The Upper West Region and the Greater Accra Region have the highest adoption rates (85.7%) for improved varieties in Ghana. With row planting technology, the Northern and Upper West Regions have more adoption than the other regions (Table 6.9 and Figure 6.2). Three-quarters of sample farmers in these regions plant maize in rows. Within the country as a whole, however, the Northern Region formers have adopted more of the row planting technology than the other regions. About a fiflh of all sample farmers in Ghana who have adopted row planting technology come from the Northern Region sub-sample (10.2% out of 54.1%). 82 University of Ghana http://ugspace.ug.edu.gh CO TE D' I VO IR E Figure 6.1: Regional and District Boundaries of Ghana Showing the Adoption Rate oflmproved Varieties in the R eg ions (% ) 64 0 64 128 Kilometers University of Ghana http://ugspace.ug.edu.gh Table 6.9: Adoption of Row Planting in the Political Regions of Ghana Region % of sample within region % of sample within total Western 31.0 3.2 Central 50.0 5.2 Eastern 42.9 8.9 Greater Accra 66.7 3.5 Volta 66.7 6.9 Ashanti 41.9 6.5 Brong-Ahafo 66.7 6.5 Northern 75.9 10.2 Upper West 76.5 3.2 (Total) 54.1 54.1 Source: Author’s computations based on the 1998 CRI/CIMMYT survey For fertilizer technology, the adoption rates for all the regions are generally low (Table 6.10 and Figure 6.3). This could be due to the high price of fertilizer. The adoption rates are between about 9% and 43%. The Eastern Region has the least adoption rates while the Upper West Region has the highest adoption rate of 42.9%. These results are not surprising since Eastern Region has basically more forest ecology that might be fertile while the Upper West is savannah that usually might require additional fertility. Table 6.10: Adoption of Fertilizer in the Political Regions of Ghana Region % of sample within region % of sample within total Western 9.5 1.0 Eastern 8.3 1.9 Volta 31.0 3.1 Ashanti 15.9 2.4 Brong Ahafo 28.6 2.9 Central 19.0 1.9 Greater Accra 19.0 1.0 Northern 33.3 5.0 Upper West 42.9 2.1 (Total) 21.0 21.0 Source: Author’s computations based on the 1998 CRI/CIMMYT survey 84 University of Ghana http://ugspace.ug.edu.gh CO TE D' IV O IR E Figure 6.3: Regional and District:Boundaries of Gkaaa Showing tne Adoption Rate o f Fertilizer in the Regions (%) .UpperEast Region Upper West Region NorthemiSegion B r o n g - A h j r f o - R e g io n Volta Region A^h^htl (Sifliori ri is.9 ? / • v r . »ir-—J Eastern Region Westerh Region ;ra Region Central Ftealoo vITm P BURKINA FASO 64 o 64 128 Kllomotors University of Ghana http://ugspace.ug.edu.gh Thus, the adoption of the improved maize technologies seems to be related more to ecological zones than political regions. Maize technology recommendations based on political regions will be difficult to make since the recommendation domains are different even within regions. Thus, recommendations on ecological zones are more appropriate. In making any inference as to whether maize research and extension have indeed helped farmers in the adoption of improved maize technologies, it is important to trace the history of the use of the technologies. The history of the adoption of maize technologies in Ghana is presented in Figures 6.4, 6.5, and 6.6. Most farmers began to use maize technologies in the late eighties and nineties. In all, 95% of the farmers first adopted improved maize varieties from 1984 to 1997 with an incremental rate of 65.2%(Figure 6.4). The period was the active research and extension period of the GGDP. Similarly, 86.8% of the fanners used fertilizer in the same period as shown in Figure 6.5, giving an incremental rate of 42.8%. With row planting, 92.7% of the farmers adopted it between the 1984 to 1997 period as in Figure 6.6 resulting in an incremental adoption rate of 57.3%. the 1984 to 1997 period as in Figure 6.6 resulting in an incremental adoption rate of 57.3%. 87 University of Ghana http://ugspace.ug.edu.gh Figure 6.4: First Year Farmer Used Improved Variety 1990-1997 80% 1978 or earlier 2% 1979-1983 Source: Author’s computations based on the 1998 CRI/CIMMYT survey Figure 6.5: First Year Farmer Used Fertilizer Source: Author’s computations based on the 1998 CRI/CIMMYT survey 88 University of Ghana http://ugspace.ug.edu.gh Explaining adoption: Extension and the farmers: Although the adoption rate of the maize recommendations are fairly high, there are still about 37.5% of farmers who do not use any of the technologies. One way of explaining this situation is to examine farmers’ contact with extension. Access to information on technology is an important ingredient influencing technology adoption. From CRI/CIMMYT survey it was found that farmers who adopted any of the improved maize technologies had more contacts with extension agents. They have two to three times in a growing season extension officer visits to talk about maize technologies. These results are presented in Table 6.11. These significant differences of the number of the extension contacts suggest that if extension agents visited farmers many 89 University of Ghana http://ugspace.ug.edu.gh more times, they could demonstrate the benefits to be derived from the adoption of improved maize technologies and thereby convince farmers to use improved inputs. Table 6.11: Extension Contacts of Adopters and Non-Adopters of Maize Technologies Technology Number of extension contacts per growing season Significance Adopter Non-adopter Improved variety 3.3 1.1 <.001* Fertilizer 4.0 1.9 <001* Row planting 3.3 1.2 <.001* * t-test Source: Author’s computations based on the 1998 CRI/CIMMYT survey There is no doubt that extension has played an important role in introducing new technologies to maize growers in Ghana. Indeed, extension has assisted in the distribution of improved seeds to farmers. Data on the sources of imp rov ed seed provide evidence on this. Almost 47% of the survey respondents who grew improved maize in 1997 reported that the seed was originally obtained from an extension agent as in Figure 6.7. In previous years, this proportion was even higher (48%). Only 6% of the farmers acquired seeds from a recognized input dealer in earlier years, but in 1997, this number increased to 26%. This suggests that farmers are at least, becoming aware of where to obtain improved seeds. It could also be due to the privatization of the seed sector. Furthermore, only 5% of farmers now purchase seeds from the open grain market as against 12% in the previous years. In addition, fewer fanners currently obtain seeds from their neighbours than before suggesting that farmers are now getting aware of the correct source of improved seeds. 90 University of Ghana http://ugspace.ug.edu.gh Figure 6.7: Source oflmproved Seed in 1997 and in Previous Years dealer (□Acquired in 1997 ®Acquired previouysly Source: Author’s computations based on the 1998 CRI/CIMMYT survey Characteristics o f the farmers Another way of explaining adoption is to recognize that the characteristics of the farmers themselves may influence their acceptance to the efforts of extension. Improved technology adopters have more formal education than non-adopters except in the use of fertilizer as shown in Table 6.12. 91 University of Ghana http://ugspace.ug.edu.gh Table 6.12: Level of Education of Fanner and Adoption of Improved Maize Technologies Technology Years in school Significance level of difference Adopter Non-adopter Improved variety 6.3 5.0 <01* Row planting 6.3 5.3 <.05* Fertilizer 5.7 5.7 Not significant* * t-test Source: Author’s computations based on the 1998 CRI/CIMMYT survey The mean age of the adopters is not significantly different from those of non­ adopters. For example, the mean age of improved variety adopters is 45.1 years while that of non-adopters is 43.3 years. Furthermore, adopters of fertilizer have a mean age of 42.4 years and non-adopters are 44.7 years (Table 6.13). Table 6.13: Farmers’ Age and Adoption of Improved Maize Technologies Technology Farmers Age (years) Significance level of difference Adopter Non-adopter Improved variety 45.1 43.3 Not significant* Row planting 44.5 44.3 Not significant* Fertilizer 42.4 44.7 Not significant* * t-test Source: Author’s computations based on the 1998 CRI/CIMMYT survey Other factors associated with the use o f improved maize technologies Improved variety: Some of the factors that are associated with the use of improved varieties are presented in Table 6.14. Farmers who plant improved varieties own more land than those who do not plant improved varieties. In a similar manner, improved maize adopters significantly put more acreage under maize production than their non­ adopter counterparts. Although farmers who plant improved varieties sell more maize (that is, 7.6 maxi­ bags) than those who do not plant improved varieties (that is, 6.8 maxi-bags), the 92 University of Ghana http://ugspace.ug.edu.gh difference is not significant. This result fails to support the notion that when farmers are market-oriented, they are more likely to invest in improved varieties and productivity improvement technologies. Table 6.14: Factors Associated with the Adoption oflmproved Maize Varieties in Ghana Factor Improved variety Significance level of differenceAdopter Non-adopter Total land owned 2.3 hectares 1.4 hectares <.001* Maize area (maior season) 1.4 hectares 1.0 hectares <.001* Maize sales 7.6 maxibags 6.8 maxibags Not significant* * t-test Source: Author’s computations based on the 1998 CRI/CIMMYT survey Row planting: Row planting is closely linked with the use of improved varieties. Row planting allows for better management of maize fields such as weeding, fertilizer application and harvesting. Farmers who adopt row planting own more land and plant larger area to improved maize than those who do not adopt row planting as shown in Table 6.15. Table 6,15: Factors Associated with Adoption of Row Planting in Ghana Factor Row planting Significance level of difference Adopter Non-adopter Resources: Total land owned 2.1 hectares 1.4 hectares <001* Major season maize area 1.4 hectares 1.0 hectares <.001* Cropping period: Average period cropped 2.7 years 1.2 years <.001* Land preparation method: <001** Animal or tractor 81% of fanners 19% of fanners Manual 44% of farmers 56% of farmers Maize sales 8.4 maxibags 6.6 maxibags Not significant* * t-test ** chi-square test Source: Author’s computations based on the 1998 CRI/CIMMYT survey 93 University of Ghana http://ugspace.ug.edu.gh The kind of land preparation method employed seems to be associated with the management of plant population. For example, more farmers ( 8 1 % ) who prepare their fields with tractors or animal power plant in rows. This is not surprising since the use of tractors gives an indication that the fields have fewer stumps and more open to facilitate row planting. Complementing this finding is the feet that fields that have been row planted have been continuously cropped for 2 .7 years than those not row planted ( 1 .2 years). Furthermore, farmers who plant in rows sell a little more maize than those who randomly plant maize but the difference is not significant. Fertilizer: Fertilizer management is one of the maize recommendations difficult to adopt, because soil fertility is site specific. However, to make fertilizer recommendations for each maize field will be complex or almost impossible. Therefore, a more general fertilizer recommendation has to be made across ecological zones and cropping history. In this study, a former is said to adopt fertilizer if he or she uses some fertilizer at least in a portion of his or her field. Some of the factors that seem to be associated with the use of fertilizers are presented in Table 6.16. Farmers who apply fertilizers own more land and also cultivate more maize than those who do not apply fertilizer and these differences are significant. Furthermore, fertilizer adopters have, on the average, cropped their fields continuously for more years than their non-adopter counterparts. This finding is logical since continuous cropping generally may deplete the soil of its nutrients if it is not replenished. Cropping history makes a big difference in a former’s decision to use fertilizer. A former is more likely to use fertilizer on an older field than one that has been newly cleared (or virgin land). It is also interesting to note that fertilizer use seems to be related to land tenure. Farmers with ownership of land tend to apply fertilizer than formers with other 94 University of Ghana http://ugspace.ug.edu.gh land arrangements such as sharecropping and renting of land. This is probably so because investments in fertilizer are expensive for fanners and therefore would wish to receive the full benefits from such investments. The low adoption of fertilizer among farmers who sharecrop is understandable since production must be shared with the land-owner but costs are borne solely by the sharecropper. Table 6.16: Factors Associated with Fertilizer Adoption in Ghana Factor Fertilizer Significance level of difference Adopter Non-adopter Resources. Total land owned 2.6 hectares 1.7 hectares <001* Maior season maize area 1.5 hectares 1.2 hectares <01* Cropping period Average years cropped 4.1 1.7 <001* Cropping history: <01** More than 5 years cropping 45% 55% 1 - 5 years cropping 20% 80% 0 - 5 years fellow 14% 86% More than 5 years fellow 9% 91% Land tenure: <01** Own land 23% 77% Rent land 9% 91% Sharecrop land 4% 96% Marketing: Maize sales 8.8 maxibags 6.9 maxibags Not significant* * t-test ** chi-square test Source: Author’s computations based on the 1998 CRI/CIMMYT survey Commercial orientation seems to play a little role in the adoption of fertilizer. There is no significant difference between farmers who adopt fertilizer and those who do not in the sale of maize. This again negates the notion of market orientation as a factor for the adoption of improved maize technologies. 95 University of Ghana http://ugspace.ug.edu.gh Maize utilization and storage As pointed out earlier, maize is largely a commercial crop. It is the first or second most important source of income for 82% of the formers in the survey. From the ecological zones, two-thirds of the formers in the transition zone derive most of their cash income from maize production as compared with less than half of the formers in the other ecological zones as shown in Figure 6.8 and Table 6.17. This seems to confirm the foot that maize is a commercial crop in the transition zone of Ghana. Fewer farmers in the coastal savannah and the Guinea Savannah consider maize as first source of income probably because in these two areas maize is very important in the household diet. If this is taken as an opportunity cost, then maize is also a commercial crop for formers in the coastal and Guinea Savannah zones. Source: Author’s computations based on the 1998 CRI/CIMMYT survey. Figure 6.8: Percent of Farmers Indicating Maize as a First Source of Income Forest Transition Guinea All zones savannah savannah Ecological zones 96 University of Ghana http://ugspace.ug.edu.gh Table 6.17: Household Source of Income from Maize Zone Household in which maize is: First source of income Second source of income Coastal Savannah 39% 50% Forest 49% 35% Transition 67% 19% Guinea Savannah 45% 21% All zones 49% 33% Source: Author’s computations based on the 1998 CRiyClMMYT survey Although farmers sell maize throughout the year, only 5.4% of them sell immediately after harvest, that is usually in August for the major maize season. This does not support the notion that most farmers sell their maize immediately after harvests. Most farmers at least store for over a month before selling probably to have a higher price. Two-thirds of the farmers sell their maize after 1 - 4 months of storage. Twenty-three percent of the fanners use actellic, a recommended storage insecticide, for storing their maize. The percent of farmers who sell some maize in the year and the sale price of maize are presented in Table 6.18. Prices of maize are higher between January an d May with the peak period in May. About a third of the farmers sell their maize within this period. Table 6.18: Principal Month for Selling Maize in Ghana, 1997 Month Percent of farmers selling Mean price (cedis/maxlbag) (a) January 6.6 51,824 February 4.6 64,375 March 7.1 56,324 April 6.0 51,516 May 4.8 66,640 June 3.1 49,875 July 2.7 38,071 August 5.4 37,214 September 11.0 37,771 October 14.3 40.472 November 21.2 44,964 December 19.1 45,859 (a) Maxibag = 110kg Source: Author’s computations based on the 1998 CRI/CIMMYT survey 97 University of Ghana http://ugspace.ug.edu.gh Gender Gender issues are important in the generation and transfer of technologies. This is because females tend to be marginalised in the society but their household decisions directly affect children’s welfare. Maize technologies developed by the GGDP are gender- neutral and all sexes are expected to adopt them equally. The percentages of male and female formers who adopted the improved maize technologies is presented in Table 6.19. The adoption of inproved varieties and row planting is significantly higher among male formers than their female counterparts. These findings are contrary to earlier maize adoption studies in Ghana2 that showed no difference between male and female formers in the adoption of improved maize variety and row planting. There is no difference, however, in the use of fertilizer among the male and female farmers in the 1998 CRI/CIMMYT survey. Table 6.19: Maize Technology Adoption among Male and Female Farmers in Ghana Technology Male adopters Female adopters Significance level of difference(%) (%) Improved varieties 59 39 <.001 Row planting 59 38 <.001 Fertilizer 23 16 Not significant Source: Authors computations based on the 1998 CRI/CIMMYT survey Initially, there is no basis for these differences in the use of improved varieties and row planting among male and female formers. Some of the reasons that might explain the differences in the adoption rates are presented in Table 6.20. Male formers have better formal education than female formers. Education is often an important factor in the Tripp R., K.A. Maifo, A A Dankyi and M.D.Read, “Changing Maize Production Practices of Small-scale Farmers in the Brong Ahafo Region, Ghana.” GGDP. 1987. p 23. 98 University of Ghana http://ugspace.ug.edu.gh adoption of improved technologies since more educated fanners have greater ability to understand and use complex technologies. Male fanners who have adopted improved maize technology have significantly higher formal education, more extension exposure and attend more field days3 than their female farmers. These opportunities help to enhance the adoption of improved technologies and therefore appear to explain the observed differences in the adoption rates. Men fanners also have bigger maize farms. Large-scale farmers tend to have greater economic incentives to invest in learning and using new technologies than small-scale farmers and thus men have higher adoption rates. Female farmers, however, have more years’ experience in growing maize than male fanners. Table 6.20: Male and Female Adopters oflmproved Maize Technology in Ghana Circumstances Male adopters Female adopters Significance level Years in school 6.3 4.1 <.001* Years growing maize 13.6 16.3 <05* Cultivated maize area (major season) 1.4 hectares 0.8 hectares <001* Extension contacts (mean) 2.6 1.4 <05* Attendance in field days (number) 0.6 0.3 .01* Access to land (own land) 77.6% 76.0% Not significant * t-test Source: Author’s computations based on the 1998 CRI/QMMYT survey Decision-making in maize farming activities The survey tried to find out who takes decisions with respect to maize fanning activities. The results are presented in Table 6.21. The results show that as long as the farm belongs to the individual, irrespective of gender, he/she takes his/her own decisions 3 Field days are organized in farmers’ fields by extension agents to demonstrate the use of maize technologies. 99 University of Ghana http://ugspace.ug.edu.gh Table 6.21: Gender and Decision-Making on Maize Farming Activities Decision / Field Person responsible for decision Significance (chi- square)Respondent (%) Spouse (%) Both (%) Other (%) Selecting maize variety: <.001 Male field 85.0 1.9 9.6 3.5 Female field 68.6 11.4 18.1 1.9 When to plant: <001 Male field 87.2 6.0 10.2 1.9 Female field 69.5 13.3 17.1 0 Obtaining fertilizer: Not significant Male field 85.7 1.6 7.9 4.8 Female field 75.0 12.5 12.5 0 Harvesting: <001 Male field 85.4 1.6 12.0 1.0 Female field 71.4 11.4 17.1 0 Sale of grain: <001 Male field 82.9 2.1 14.2 0.7 Female field 70.7 14.1 15.2 0 Use of cash <001 Male field 81.4 1.8 16.4 0.4 Female field 62.0 12.0 26.1 0 Source: Author’s computations based on the 1998 CRI/CIMMYT survey in most cases but when decisions are made by spouses, then female formers significantly tend to rely more on their spouses’ decisions than their male counterparts. In other words, when forming activity decisions are to be made by the form femily, female farmers tend to rely more on their male counterparts. Impacts o f improved maize technologies Although a measure of the rate of return to investment in agricultural research gives some indications of the impact of the research, it can be supplemented with some descriptive or qualitative impacts such as former incomes, nutrition, productivity and gender effects. A more quantitative analysis of these impacts is not possible since the GGDP did not have baseline data at the onset of the project in 1979 and therefore makes it difficult for such an assessment. 100 University of Ghana http://ugspace.ug.edu.gh Incomes of farmers Income is an important welfare indicator because it is linked to the ability to acquire property such as clothing and shelter, and wealth. An increase in income can be translated either directly or indirectly into the betterment of the well being of the formers. One of the goals of maize technology development is to increase the welfare of formers. In the survey, formers were asked whether over the past ten years they had noticed any changes in their production of maize, sales and income levels. Their qualitative assessment of impact of maize technology is presented in Figure 6.9. Improved maize adopters have benefited more in the production indicators than local maize adopters. These suggest that formers who adopt improved maize technologies are likely to be better off than farmers who do not adopt the technologies. More than one-half of the adopters of improved maize technologies indicated that there were increases in production (56%), yields (57%), sales (50%), consumption (74%) and income (52%) from maize cultivation. These findings are important because increase in income is a measure of welfare gains. Thus these results suggest that maize research and extension have had some impact on household incomes and welfare of some of the maize formers. 101 University of Ghana http://ugspace.ug.edu.gh Figure 6.9: Percent of Fanners with Increases in Maize Indicators □ Improved maize adopters IB Local maize adopters Source: Author’s computations based on the 1998 CRI/CIMMYT survey 102 University of Ghana http://ugspace.ug.edu.gh Farmers who obtain their income from maize sales were asked to indicate what they have done with the extra money from maize sales. Ihe answers from improved maize technology adopters are presented in Table 6.22. Most of the farmers spend the extra money from maize on children’s education. A third of the farmers take care of the whole family. A girrrilar proportion has constructed houses from the income derived from maize sales. These are signals to improved welfare gains of the whole household. Table 6.22: Farmers Use of Increased Income from Maize Use of income Percent of farmers (a) Pay for children’s fees/education 42.1 Take care of the family 34.2 Build house 28.3 Expand farm 15.8 Purchase livestock 5.3 Trade 4.6 Household furniture/equipment 1.9 Purchase vehicle or bicycle 1.3 (a) Total percentage is more than 100% because of multiple answers. Source: Author’s computations based on the 1998 CRI/CIMMYT survey Nutrition One of the objectives of the GGDP was to improve the nutritional status of formers and consumers at large by making more maize available to the Ghanaian society through the development and subsequent adoption and production of improved maize. Maize consumption provides carbohydrate that is turned into energy in human beings. As discussed earlier (Figure 6.9), of about 77% of all the farmers who indicated increased consumption of maize, about 74% were adopters of improved maize technologies. Similarly, as maize is either consumed by farmers or eventually sold to the public, consumers also benefit from its consumption. In addition to the importance of maize as a widely consumed food, it is used as a weaning food for infants in Ghana. In 1992, the GGDP released protein-quality maize, 103 University of Ghana http://ugspace.ug.edu.gh Obatanpa. The feet that this variety is the most popular variety grown by farmers underscores the importance of Obatanpa. Although the actual impact of Obatanpa on Ghanaians is not documented, those on animals are documented by the GGDP. In the present survey, 51% of all farmers said they were aware of the nutritional enhancing ability of the Obatanpa. However, only 35.8% reported using Obatanpa or other improved maize varieties as weaning food. The reasons for this need further investigation but certainly, more education is needed on the nutritional prospects of Obatanpa. Gender effects As has been discussed earlier on, women are usually the disadvantaged in the society especially in decision making. The survey revealed that apart from the use of fertilizer in which men and women adopted equally, women have lower adoption rates for improved variety and row planting. From the survey, since women have grown some maize for a longer period than male farmers, there is need to ensure that they get equal access to the technologies. Productivity The importance of agriculture in rural Ghana cannot be overemphasized. This makes agricultural productivity a valid indicator of impacts of maize research and extension in Ghana. The feet that many Ghanaian farmers grow some maize, any technology that helps to increase productivity will bring about real income gains for producers and subsequently to consumers. This is because farmers can produce at a lower cost as productivity increases; thus, the cost of production will decrease. For consumers, 104 University of Ghana http://ugspace.ug.edu.gh increases in productivity are translated into lower prices for maize. Thus, consumers can consume more at a lower price. It has been shown earlier on that maize productivity (which is measured here as the yield per unit land area) has generally been increasing since 1985 (see Figure 1.1). Obtaining accurate historic yield gain data from farmers is difficult because most Ghanaian farmers do not keep quantitative records of their forming activities. Under this situation, the only way to obtain accurate information is to make crop yield cuts to measure the actual yields from farmers’ fields but this will be equally expensive on a large scale. In this study therefore, formers were asked to estimate the number of bags they would expect to harvest from their largest maize field using each of the following combinations: • Local variety without fertilizer • Previous improved variety without fertilizer • Current improved variety without fertilizer • Local variety with fertilizer • Previous improved variety with fertilizer • Current improved variety with fertilizer These are equivalent to experimental “treatments”. Farmers were further asked to give estimates only for technology combinations that they have actually used. The results are therefore based on formers’ direct experience. Farmers’ estimates of maize harvests in bags under “with fertilizer” and “without fertilizer” scenarios are presented in Table 6.23. With fertilizer, the yield from the improved varieties doubles. The local varieties are less responsive to fertilizer and give about a quarter yield increase. 105 University of Ghana http://ugspace.ug.edu.gh Table 6.23: Descriptive Statistics of Farmers’ Estimates of Maize Harvests (in maxibags) Treatment Mean Standard deviation Sum (harvest) (N) Without fertilizer: Local variety 8.4 10.6 2461.2 (292) Previous improved variety 10.8 9.2 520 (48) Current improved variety 11.8 11.0 2546 (216) With fertilizer: Local variety 12.3 16.5 1363.5 (111) Previous improved variety 23.7 37.2 733.5 (31) Current improved variety 21.8 28.3 2660.5 (112) Source: author’s computations based on the 1998 CRI/CIMMYT survey. Table 6.24: Estimated Yield Increase Attributable to Improved Varieties and Fertilizer Treatment Percent yield increase Without fertilizer With fertilizer Local 'variety to current improved variety 39.9 77.6 Previous improved variety to current improved variety 8.9 -7.8 Local variety to previous improved variety 28.5 92.7 Addition of fertilizer to local variety 45.7 Addition of fertilizer to current improved variety 85 Addition of fertilizer to previous improved variety 118.5 Contribution of both fertilizer and current improved variety 158.7 Contribution of both fertilizer and previous improved variety 180.7 Source: Estimated by the author from the 1998 CRI/CIMMYT survey In Table 6.24, the current improved variety alone increases yield by 39.9 percent. The yield difference between previous improved variety and current improved variety is 8.9%. This is not surprising since both are improved varieties. What is surprising is the fact that previous improved varieties respond more to fertilizer than the current improved variety. Newer improved varieties have some disease resistance incorporated in them than the earlier improved varieties and this is, however, expected to increase yields. Nevertheless, this assertion needs further experimental investigation. Fertilizer response in improved varieties under farmer’s estimate is about twice the response in local varieties. This is more than experimental results which put it about fifty 106 University of Ghana http://ugspace.ug.edu.gh percent4. Farmers further recognize that interaction of fertilizer and improved variety gives higher yield increment. These results give ample evidence that improved maize technologies have greatly increased productivity on forms where the adoption of maize technologies have taken place. Further evidence is given by the responses formers gave to the question of whether they have seen any changes in their maize yields in the past ten years. Nearly 60% of those who responded indicated that their maize yields have increased. Factors explaining the differences in adoption pattern between the ecological zones: The GGDP maize recommendations are based on ecological zones and it is important to examine some of the factors that explain the adoption pattern between the ecological zones. It is believed that the GGDP played down research and extension of inproved maize technologies in the forest zone because the forest zone is noted for growing tree crops. This explanation may not be wholly correct. The results of the maize survey show that formers in the forest zone had a fair share of extension contacts on maize. The number of extension contacts by formers in the forest zone is comparable with those of the coastal savannah and Guinea Savannah and is even more than the transition zone. For example, the number of extension contact in the forest zone is 2.1 and those of the Guinea Savannah and coastal savannah are 2.6 and 3.1 respectively. Farmers in the transition zone have 1.5 times extension contacts. In spite of comparable extension contacts, the adoption of maize technologies in the forest is low compared to the other zones. 4 GGDP Annual Report, 1980. p. 43 107 University of Ghana http://ugspace.ug.edu.gh Furthermore, the years fanners have acquired experience in growing maize are about the same for farmers in the forest, transition and coastal savannah. For example, formers have 15.2 years of experience in growing maize in the forest zone. It is 15 years in the transition, 16.2 years in the coastal savannah and 10 years in the Guinea Savannah. Therefore there seems to be other reasons for the differences in the adoption of maize technologies between the ecological zones other than those explained earlier. The factors or reasons for such differences in adoption patterns include differences in farmer characteristics, resource ownership and access to technology. In addition, there are others that are frequently neglected in explaining adoption patterns. Based on the Boserup hypothesis5, factors such as intensification of agriculture, market orientation and land-labour endowments are other important factors influencing the adoption process. Population density There appears to be some relationship between years of schooling and household size. Generally, across the ecological zones, as the years in school are high the household size is low. In Table 6.25, while farmers in the forest zone have the highest number of years in school, their household size is the lowest compared to the Guinea savannah whose formers have the lowest number of years in school but the highest household size. 5 Boserup, Ester. "The Conditions o f Agricultural Growth: The Economics o f Agrarian Change under Population Pressure. New York: Aldine Publishing. 1965. Ester Boserup developed a theory to explain the interaction between population growth, land settlement and technological change. Boserup challenged the classical Malthusian theory of population increasing faster than food production and thereby leading to disaster. Boserup argued that as a result o f increasing population pressure, a traditional agrarian society shifts from slash-and-bum fallow system to bush fallow, then to annual cropping, and finally to multiple cropping. Each sequence is an increasing intensification of the farming system. Technological change arises with increasing population pressure. 108 University of Ghana http://ugspace.ug.edu.gh Table 6.25: Fanners’ Formal Education Level and Household Size Ecological zone Years schooling (years) Average household size Forest 6.7 8.0 Transition 6.5 9.8 Coastal Savannah 6.3 9.7 Guinea Savannah 2.3 15.4 All zones 5.7 10.1 Source: 1998 CRI/CIMMYT survey Technological change comes as a response to adversity, which results from increasing population pressure. The higher the household size, the more mouths the subsistence former is likely to feed and to have some in stock for the lean season. The former may therefore be forced to adopt strategies for increasing his or her food supply. This is probably one of the main reasons why the adoption rate of maize technologies is high in the Guinea Savannah, transition and coastal savannah zones than there is in the forest zone. Probably because the forest zone has a lower household size, it could afiord extensive agriculture with traditional technologies. Cropping intensity Fertilizer use on maize is positively associated with high levels of cropping intensity. In general, maize fields that have received fertilizer have been continuously cropped for 4.1 years on the average compared to 1.7 years for fields that did not receive fertilizer. As population density increases, there is over-exploitation of land and other natural resources of agriculture6. The land continuously becomes degraded because without fertilizer or other fertility enhancing measures, the land is unable to sustain crop growth 6 Broekhuyse, J.T. and AM. Allen. "Farming Systems Research " Human Organization 47(4): 19S8 pp.. 330-342. 109 University of Ghana http://ugspace.ug.edu.gh once the Mow system breaks down. There will be, in the long run, yield decline and finally abandonment of land unless more intensive technologies are employed7. In Table 6.26, there is more continuous cropping of maize in the Guinea Savannah, the transition and the coastal savannah zones than there is in the forest zone. Table 6.26: Years of Continuous Maize Cropping of Major Maize Field Ecological zone Average years t-ratio Significance Guinea savannah 4.7 8.753 .0000 Coastal savannah 3.2 4.322 .0000 Transition 1.9 2.876 .0060 Forest 0.6 Source: 1998 CRI/CIMMYT survey For example, while farmers in the Guinea Savannah crop maize continuously for 4.7 years, fanners in the forest zone crop maize for less than one year. These differences are significant compared to that of the forest zone. This seems to suggest that fanners in the forest zone plant maize on new fields every year. Maize fields in the forest zones might thus be more fertile than maize fields in the Guinea Savannah zone. It is not surprising therefore, that the adoption of fertilizer is relatively high in the Guinea Savannah, coastal savannah and the transition zones than the forest zone. In a further explanation. Table 6.27 shows the average number of years that the major maize field was fellow. As expected, the fellow period for fields in the Guinea Savannah, coastal savannah and the transition zones are significantly shorter than the forest zone. For example, while the fellow period in the forest zone is 3.4 years, the fellow period in the Guinea Savannah is 0.6 years. This again provides evidence that there is less pressure on the land in the forest zone than the other ecological zones. Hence, the Guinea 1 Sanders John H., Barry I. Shapiro and Sunder Ramaswamy. The Economics o f Agricultural Technology in Semiarid Sub-Saharan Africa. Johns Hopkins University Press. 1996 pp 145-177. 110 University of Ghana http://ugspace.ug.edu.gh Savannah, the coastal savannah and the transition zones necessarily have to adopt some improved measures in order to get high yields. Table 6.27: Average Fallow Periods in Ghana Ecological zones Average No. of years t-ratio Significance Guinea Savannah 0.6 -10.825 .0000 Coastal Savannah 2.3 -3.004 .0040 Transition 2.3 -4.302 .0000 Forest 3.4 All zones 2.6 Source: 1998 CRI/CIMMYT survey Extensive crop systems are the expected former behaviour in the forest zone because potential good land is available. Coupled with a low population density, extensive agriculture is economically and socially sustainable in the forest zone8. Furthermore, row planting of maize is positively associated with cropping intensity. Continuous cropping, no doubt, de-stumps the fields and makes planting in rows easier. Hence, the adoption rate of row planting is higher in the Guinea Savannah zone (73%) than the forest zone (39%), for example. In addition, row planting is related to land preparation methods. Of the formers who cultivate their fields by hand, about 56% plant their maize at random. On the other hand, about 81% of formers who cultivate their fields with animals or tractors plant their maize in rows (1998 CRI/CIMMYT survey). In the transition, Guinea Savannah and coastal savannah, there is more tractor power or animal (in the Guinea Savannah) services than there are in the forest zone. These further explain the use of improved technologies in these areas than in the forest zone. Pingali, P., Y. Bigot and H. Binswanger. “Agricultural Mechanization and the Evolution of Fanning Systems in Sub-Saharan Africa.” Baltimore: Johns Hopkins University Press. 1987 pp 5-15 43-54 111 University of Ghana http://ugspace.ug.edu.gh Markets aud commercial orientation Given suitable soil conditions, areas with better access to markets may be more intensively cultivated. Access to urban centres is increased with reduced transport costs. Reduced transport cost is associated with good infrastructure. Access to infrastructure by the households in the present maize study has been presented in earlier section (see Table 6.2). Fanners in the coastal savannah have access to improved infrastructure. The intensity of maize technology adoption in the coastal savannah may partly be explained by the access to infrastructure such as tarred road network, easy transportation system and well organized markets. Similarly, farmers in the transition zone have reasonable access to transportation facilities and improved tarred roads but limited access to market. Farmers living in the coastal savannah have more electricity than other farmers in the other zones. Since maize is a major component of the diet of people in the coastal savannah, perhaps the use of electricity facilitates the processing of maize products that further induce marketing and subsequent adoption of improved technologies. Economics o f maize technology One of the important determinants of adoption of a technology is the profitability of the technology. Whether farmers sell little or most of their farm produce, they are interested in the economic returns. Farmers weigh the benefits to be gained from the use of the technology against the costs incurred in the form of labour and cash given up. The farmers are simply assessing the returns to investment in the technology. Of die three GGDP-generated maize technologies, the use of improved maize requires the least cash outlay per unit area. Farmers who decide to adopt improved varieties require few changes to their crop management practices. In Table 6.28, the 112 University of Ghana http://ugspace.ug.edu.gh average values used in estimating the marginal rate of return are presented. The profitability of the adoption of improved maize in the ecological zones is analyzed in Table 6.29. The marginal rate of return (MRR)9 to investment in improved maize in the forest zone is 3030% for Dobidi and that of Aburotia is 2250%. These are about twice as that of the transition zone (Dobidi: 1800% and Aburotia: 810%) or the savannah zones where the use of Dobidi gives a MRR of 1560% and that of Aburotia is 1290%. Thus, in the forest zone for example, for every one cedi invested in Dobidi, farmers can expect to recover the one cedi, and obtain additional 30.30 cedis. Although these returns are attractive in all the zones, the forest zone has the highest MRR Table 6.28: Economic indices for Estimating Marginal Rate of Return to Maize Technologies in Ghana, 1997 Item Unit and Price Option A Improved maize; Row planting; 5 bags of NPK fertilizer, 5 bags of Sulphate of ammonia Option B Improved maize; Row planting; 2.5 bags of NPK fertilizer, 2.5 bags of Sulphate of ammonia Option C Farmer practices Reid price of maize grain 300 cedis/ kg Improved maize seed 2,000 cedis/kg Local seed 1,500 cedis/kg Seed rate 20kg/ha NPK fertilizer 30,000 cedis/bag of 50kg Sulphate of Ammonia 36,800 cedis/bag of 50kg Transport cost for fertilizer 1,000/bag Labour for fertilizer application 35,000 cedis/ha Row planting cost 35,000 cedis/ha Source: Data is from personal interviews with MOFA staff, GLDB, farmers and various GGDP reports The Marginal Rate o f Return (MRR) is obtained as the change in net benefits between farmer practice and the improved practice divided by the change in costs o f the same practices, expressed as a percentage. 113 University of Ghana http://ugspace.ug.edu.gh Table 6,29: Marginal Return Analysis of Adoption z] Mean of Variable Constant -4.88674 0.5985 -8.165 0.0000 EXTVISrr 0.14602 0.36953E-01 3.952 0.0001 2.3000 HHSIZE 0.17974E-01 0.26549E-01 0.677 0.4984 10.1047 ZONECS 2.28518 0.45684 5.002 0.0000 0.2000 ZONETR 2.62006 0.49790 5.262 0.0000 0.1500 ZONEGS 1.42237 0.49998 2.845 0.0044 0.2000 GENDER 1.02497 0.43278 2.368 0.0179 0.7500 TENURE 0.88860 0.37861 2.347 0.0189 -1.6523 MZAREA 0.65899 0.51433E-01 12.813 0.0000 3.7049 Disturbance standard deviation Sigma12 3.06781 0.15564 19.713 0.0000 Log of likelihood function -694.7732 Source: Estimated by the author based on data from the 1998 CRI/CIMMYT survey. The results in Table 6.42 show that extension contact, coastal savannah, Transition, Guinea Savannah zones, gender, land tenure and maize area are all significant at 5% level and are positively related to the proportion of area planted to inproved varieties. The significance level of extension contact is remarkable because it is through contacts with extension officers that many formers become aware of inproved varieties and consequently acquire them for use. This finding confirms the result that inproved maize adopters had relatively more contacts with extension agents during a 12-month period of 1997 crop season. Thus, extension information plays an important role in the promotion of adoption of inproved technologies. 12 Sigma is the ancillaiy parameter 127 University of Ghana http://ugspace.ug.edu.gh The ecological zones are important factors explaining adoption. All the three zones have significant positive coefficients. The value of the coefficient of the coastal savannah is 2.2852 and that of the transition zone is 2.6201 while it is 1.4224 in the Guinea Savannah zone. Household size is, however, not significant but it is positively related to the dependent variable. The marginal analyses of the factors influencing the proportion of area planted with improved varieties are shown in Table 6.43. All the factors are significant except household size. The coefficients of the equation give indications of the marginal changes. Table 6.43: TOBIT Marginal Effect for Factors of Maize Technology Adoption Variable Coefficient Standard error b/St. Er P[I z I > z] Mean of Variable Constant -1.53086 0.4486 -3.413 0.0006 EXTVISIT 0.45744E-01 0.2013E-01 2.273 0.0230 2.3000 HHSIZE 0.56307E-02 0.86928E-02 0.648 0.5171 10.1047 ZONECS 0.71588 0.26716 2.680 0.0074 0.2000 ZONETR 0.82079 0.31942 2.570 0.0102 0.1500 ZONEGS 0.44558 0.22961 1.941 0.0523 0.2000 GENDER 0.32109 0.16454 1.951 0.0510 0.7500 TENURE 0.27837 0.30610E-01 9.094 0.0000 -1.6523 MZAREA 0.20644 0.72315E-01 2.855 0.0043 3.7049 Source: Estimated by the author based on data from the 1998 CRI/CIMMYT survey. For example, one standard deviation change in extension contact leads to a change of 0.046 or 4.6% change of proportion of area planted to improved varieties. Similarly, one standard deviation change in maize area, changes the proportion of improved maize area by 0.2 or 20%. The coastal and the transition zone have the highest marginal effect among the variables in the model. The marginal effect of the transition zone is 82% while that of the coastal savannah is about 72%. This shows that ecological zones are important variables to consider in the introduction and adoption of improved maize technologies in Ghana. 128 University of Ghana http://ugspace.ug.edu.gh The TOBIT analysis has examined the factors that affect proportion of area that is planted to improved maize varieties and the marginal effect of these factors. These are important factors that must be taken into consideration to guide fixture maize research and improve strategies for increased adoption of improved maize varieties. The analysis has also included ecological zones as separate independent variables in order to show their significance than if they were bulked together as ecological zone. Future adoption studies could include ecological zones as some of the explanatory variables in adoption decision models. Chapter Summary The adoption of improved maize and row planting has been relatively high However, fertilizer adoption is low. Over ninety percent of the farmers first adopted the improved maize technologies between 1984 and 1997 coinciding with the increased research and extension activities of the GGDP. Farmers who adopted improved maize technologies reported of increases in their maize production, yield, consumption and income. The marginal analysis of maize recommendation shows that when formers adopt inproved maize technologies in a piece-meal manner, the MRR are much higher than adopting the whole package of improved seed, fertilizer and row planting. The profitability of maize technology depends also on the ecological zones. The critical factors affecting the adoption of improved varieties are former contact with extension activities, being in the transition zone, Guinea Savannah zone and the coastal savannah zone. Gender and the size of the area cultivated to maize further affect the adoption of improved maize varieties. 129 University of Ghana http://ugspace.ug.edu.gh RETURNS TO INVESTMENT IN MAIZE RESEARCH AND EXTENSION The key assumptions and scenarios used to estimate the internal rate of return to TTiaiye research and extension are given in Table 7.1. Five scenarios are described. These scenarios serve as sensitivity analysis. The results are summarized in Table 7.2, whereas in Table 7.3, the estimation of the financial rate of return for the baseline scenario is given. The details of the results from the scenarios are presented in Appendices 7 through to 16. CHAPTER 7 Table 7.1: Key Indicators Used in Estimating Internal Rate of Return Parameter Scenarios 1 2 3 4 5 Research kg (years) 5 5 5 5 5 Adoption lag (years) 30 30 30 30 30 Adoption level ceiling (%) 95 95 95 95 95 GGDP research cost included (%) 90 100 90 100 90 MOFA extension cost included (%) 30 30 50 50 30 Yield increase (%) 100 100 100 100 60 Source: Compiled by the author. Scenario 1 in Table 7.1 (see also Appendices 7 and 8) is the baseline scenario where actual research and extension costs for maize are included. Because of lack of continuous data on adoption rates for each year and the fact that these had to estimated in order to estimate the IRR, upon discussions with scientists and extension officers, the time to full adoption was assumed to be 30 years with adoption ceiling of 95%. These were used in the logistic function to estimate the adoption rates with the best fit. Scenario 2 to 5 serve as sensitivity analysis where research costs, extension costs and yields are changed. In scenario 2 (Appendices 9 and 10), research costs are increased from 90% to 130 University of Ghana http://ugspace.ug.edu.gh 100%. Higher extension costs are included in scenario 3 (Appendices 11 and 12) from MOFA baseline budget of 30% to 50%. In scenario 4 (Appendix 13 and 14), both research and extension costs are increased. The yield advantage of the improved maize varieties over farmers’ varieties is reduced by 40% in scenario 5 (Appendix 15 and 16). Scenario 1 (Baseline): Under the baseline scenario where the actual expenditures from research and extension were included, the financial rate of return to public investment in maize research and extension in Ghana during the period 1979-97 was 79% (Table 7.2). Table 7.2: Financial and Economic Rate of Return to Maize Research in Ghana Scenarios Financial rate of return Economic rate of return 1 Baseline 79% 33% 2 Higher Research Costs 73% 30% 3 Higher Extension Costs 71% 31% 4 Higher Research and Higher Extension Costs 68% 30% 5 Lower Projected Research Benefit 50% 25% Source: Estimated by the author. This internal rate of return is the discount or interest rate at which the Net Present Value is equal to zero. It thus represents the maximum interest rate that maize research and extension in Ghana could pay for all the resources employed from 1979 to 1997 after recovering its investment and operating cost. The financial rate of return is thus the return accruing to the maize formers from maize research and extension. The IRR of 79% means that for every cedi invested in research and extension activities in Ghana during 131 University of Ghana http://ugspace.ug.edu.gh the period of 1979 to 1997, seventy-nine pesewas was generated in return after recovering the cos. This result is similar to other studies in Ghana.1 The economic rate of return was 33%. This is the contribution of maize research and extension to national income relative to the resources used. Similarly, an IRR of 33% means that the Ghanaian society gains thirty-three pesewas for every cedi spent on maize research and extension after paying the cost. In estimating the economic rate of return, since economic prices were used for the social costs of the inputs and maize price, the analysis represents the whole society’s point of view. Hence, the economic rate of return is the benefit for the whole Ghanaian society. The financial and economic rates of return are very attractive by normal investment standards and suggest that there may have been under-investment in maize research. The difference in the financial and economic rates of return indicates that policy- induced distortions were relatively excessive. This suggests that markets for maize and inputs used in maize production are not competitive. In other words, the domestic prices of maize and inputs over the entire period of years were relatively lower than the world prices. 1 Sanders John H., Taye Bezuney, Alan C. Schroeder. Op. Cit pp. 1-28. 132 University of Ghana http://ugspace.ug.edu.gh Table 7.3: Social Gains and Financial Rate of Return Year Proportional production Increase Proportional cost increase Net sbift in supply Change in quantity Caused by research Parameter j = t x (dY/Ym) i = I/P = (dC x t)/(Ym x P) k = (j / E) - i dQ = (Q x e x E x k)/(e + E) Unit 1979 0 0 0 0 1980 0 0 0 0 1981 0 0 0 0 1982 0 0 0 0 1983 0 0 0 0 1984 0.0133 0.0019 0.0149 3,746 1985 0.0185 0.0035 0.0199 3,443 1986 0.0274 0.0047 0.0299 7,317 1987 0.041 0.0076 0.0443 11,598 1988 0.0547 0.0137 0.0555 14,577 1989 0.0684 0.0325 0.0541 16,906 1990 0.0889 0.0335 0.0791 19,119 1991 0.1163 0.0421 0.0151 42,843 1992 0.1505 0.0948 0.0957 30,589 1993 0.1915 0.1337 0.1087 45,724 1994 0.2325 0.1725 0.1218 50,107 1995 0.2804 0.2261 0.1288 58,296 1996 0.3283 0.2485 0.167 73,647 1997 0.3693 0.2895 0.178 77,881 133 University of Ghana http://ugspace.ug.edu.gh Table 7.3 Continued Social gains from research (In million cedis) SG = (k x P x Q) - l/2(k x P x dQ) Research, Extension and CIMMYT real cost (million cedis) NG= SG- (R+ Ext) Net gains (million cedis) Internal Rate of Return (%) 269 -269 168 -168 102 -102 138 -138 0 77 -77 3,022 1292 1,730 2,222 1995 227 6,308 1794 4,514 11,869 1579 10,290 14,553 3222 11,331 10,524 2819 7,705 14,115 2340 11,775 36,592 5209 31,383 20,441 3699 16,742 27,059 7587 19,472 29,892 3583 26,309 31,920 1978 29,942 36,040 1314 34,726 35,919 673 35,246 79% Source: Estimated by the author. Where, j = proportional production increase (incremental output) dY = yield increase due to the improved technology t = adoption rate (proportion of total area under improved technology) Ym = farmers’ yield (instead of national average yield because the yield gain is higher than the national average yield), i = proportional (per unit) input expenditure I = per unit input expenditure P = maize price dC — change in costs due to new technology k = net shift in supply E = elasticity of supply dQ = change in quantity caused by research SG = social gains. NG = net gains R = research cost Ex = extension cost 134 University of Ghana http://ugspace.ug.edu.gh The difference between the financial and the economic rates of return seems to have resulted from the differences between the real exchange rates and the official exchange rates used to estimate them These exchange rates had effect on the total costs and benefits from research and extension activities and hence the net social gains. There were initial higher negative values of the net social gains in the economic rate of return than in the financial rate of return (Table 7.4). In other words, the initial losses in the social gains in economic rate of return were higher than the financial rate of return. Since internal rate of return calculations are especially sensitive to net benefits accruing earlier in the period of analysis, the lower initial benefits had the effect of making the economic rate of return lower than the financial rate of return. Table 7.4: Economic and Financial Net Social Gains Year Net social gains (million cedis) Economic Financial 1979 -24,898 -269 1980 -11,895 -168 1981 -3,551 -102 1982 -2,238 -138 1983 -1,143 -76 1984 11,515 1,730 1985 7,594 227 1986 29,406 4,514 1987 42,571 10,289 1988 50,468 11,331 1989 85,497 7,705 1990 56,886 11,775 1991 97,231 31,383 1992 100,027 16,742 1993 148,728 19,472 1994 194,312 26,309 1995 120,896 29,942 1996 133,228 34,726 1997 53,549 35,246 Source: Extracted from Appendices 7 and 8 by the author 135 University of Ghana http://ugspace.ug.edu.gh As shown in Figure 7.1, both the real domestic and world maize price has been falling steadily. With the increase in national maize production and yield, the fall in maize price seems to suggest that the supply curve for maize has indeed shifted to the right. Consumers are able to consume more at a lower price. Thus, the consumer surplus has increased over the years. On the producer side, the increase in supply means a reduction in maize production cost. Thus, the producer surplus has increased and therefore, formers have also gained by spending less per unit of output. These are indicators of the benefits of maize research to producers and consumers. Figure 7.1: Real (Farmgate) Domestic and World Price of Maize Source: Estimated by the author As fiirther shown in Figure 7.1, since the real domestic price of maize is lower than the real world price, (using the rural price index as a deflator), the export of maize 136 University of Ghana http://ugspace.ug.edu.gh seems possible. Transportation and marketing costs were estimated as 50% of the F.O.B. price and assumed constant. Given improved transportation system and infrastructure, transportation cost and marketing margin are expected to decrease over time. This invariably affects the import parity price estimated because parity price is the border price of the input or product at the border and adjusted for expenses between the border and the project boundary. The value of the parity price for maize for example, may in turn, affect the estimation of economic rate of return. There is thus a need for better estimates for transportation and marketing margin over time Scenario 2 (Higher research costs): As discussed earlier, in estimating parameters for the net social gains, because of lack of reliable data, several assumptions were made. These assumptions could in some cases, have underestimated the cost with inherent overestimation of the benefits leading to higher IRR. The reverse of this situation is also possible. To assess the effect of these assumptions, sensitivity analyses were made. Sensitivity analysis measures the effects of risks and uncertainty in project analysis. By changing some of the values of some key parameters of the project, it is possible to evaluate the robustness of the IRR. Not only does sensitivity analysis have important implications for decisions, it also has implications for project management. In estimating the IRR for maize research and extension, total expenditures from research and extension could affect the results of the IRR Since some of these costs are based on estimates, it is important that some sensitivity analysis is conducted on research 137 University of Ghana http://ugspace.ug.edu.gh and extension costs to see how sensitive the results are to changes in their costs. If the proportion of GGDP research costs assigned to maize is increased from 90% to 100% and all other costs and assumptions remain the same, the financial rate of return goes down from 79% to 73% leading to a fell of 7.6%. The economic rate of return on the other hand, drops from 33% to 30% representing 9% drop. Scenario 3 (Higher extension cost): As pointed out earlier, obtaining the exact cost information for extension activities for maize was difficult. As such estimates were based on information from MOFA and in consultation with senior extension officers. Some costs may have been underestimated. In order to assess the sensitivity of the IRR to increased extension costs, the analysis was repeated with the proportion of MOFA extension costs assigned to maize increased from 30% to 50% while all other conditions remained the same. The financial rate of return decreased to 71% from the baseline value of 79% (see Table 7.2). This is a decrease of 10.1%. The economic rate of return is 31%. Scenario 4 (higher projected research and extension costs): The essence of this scenario is to test the IRR with further increased costs. Thus if the proportion of research costs is increased from 90% to 100% and that of MOFA extension costs also increased from 30% to 50%, the financial rate of return drops further to 68%, a decrease of 13.9%. The economic rate of return becomes 30%, felling by 9%. The result shows that as costs increase, the returns to investment also decline. Therefore, care should be taken not to over spend. 138 University of Ghana http://ugspace.ug.edu.gh Scenario 5 (lower projected research benefits) A sensitivity test of IRR to errors in estimated yield was conducted. A test to determine how sensitive the financial and economic rate of return are to lower yields not only provide useful information in deciding whether the project is worthwhile, but also emphasize the need for sufficient extension services if high yielding varieties are expected. If the projected yield gains attributable to the adoption of the research­ generated technologies are reduced by 40%, the financial rate of return falls to 50% and the economic rate of return is 25%. The drop in the financial rate of return is 36.7% and that of economic rate of return becomes 24.2%. These are substantial reductions of the IRR and indicate that the IRR analysis is sensitive to 40% drop in yield gain. This has implications for future maize research because the drop in the values of the IRR show how much yield loss will cost the nation in lost opportunities to generate wealth. Hence, researchers will have to ensure high-yielding and stable varieties. Chapter Summary The estimated IRR for maize research and extension investment in Ghana ranged between 25% and 79%. Under the most conservative assumptions, the IRR was 25%. This means that for every cedi invested, the society as a whole received twenty-five pesewas in return. This is a clear demonstration that past public investments in maize research and extension activities in Ghana from 1979 to 1997 have had a pay-off. 139 University of Ghana http://ugspace.ug.edu.gh CHAPTER 8 POLICY IMPLICATIONS AND CONCLUSIONS Policy implications During the past two decades, public investment in maize research and extension in Ghana clearly has been successful The analysis shows that the resources invested by the Ghana government, foreign donor agencies, international agricultural research centers, and non-governmental organizations (NGOs) have generated attractive returns. The estimated internal rate of return to investments in maize research and extension activities ranged from 25% to 79%. This means that for every cedi invested in maize research and extension activities, 25 to 79 pesewas are obtained as a return after recovering cost. This clearly demonstrates that past public investments in maize research and extension support in Ghana have been profitable. The payoffs to maize research have been reflected in increased output, decreased production costs, and lower prices for consumers. These findings lave a number of important policy implications. Research policy issues: Considering the role maize plays in the diet of most Ghanaians and in food security in general in Ghana, it is legitimate to focus on maize crop because it is such an important crop and thus a priority for research and extension. This is contrary to the notion that maize research and extension have had sufficient financial support and therefore do not need any more support. Indeed, some financial and material support is still needed to maintain and increase research and extension activities. 140 University of Ghana http://ugspace.ug.edu.gh Continued funding is absolutely essential to maintain the successes that have been achieved over the past years. Since research is a long-term process requiring sustained financial support, it is necessary that maize research, and extension activities in particular, is financed to maintain the momentum of adoption of the improved maize technologies. Agricultural research and extension expenditures in one year potentially affect agricultural productivity for many years to come. The effect of inadequate funding of agricultural research and extension activities in one period will thus be very difficult to overcome later. Germplasm improvement research has complemented crop management research. Adoption of fertilizer-responsive improved varieties is an important factor in enhancing the efficiency and thus the profitability of fertilizer use. Continued research on germplasm is therefore necessary to facilitate the increased maize adoption and subsequent production. With crop management activities, attention should be given to soil fertility issues particularly the use of leguminous plants in the cropping systems. Technology adoption is usually related to profitability. Therefore, crop management recommendations should take the step-wise adoption behaviour of farmers and profitability into consideration. In developing maize recommendations, there is the need to identify the important technological components and options and their effect on production systems and the use of resources. Recommendations for maize production should have alternatives for fanners to select and adjust to their needs and circumstances. 141 University of Ghana http://ugspace.ug.edu.gh The rate of return on recommended technological package of improved seeds, fertilizer, and row planting for maize is very sensitive to changes in yield. Even at 10% yield reduction, none of the technological options recommended for the ecological zones had acceptable minimum rate of return of 35%. Therefore, if farmers are to accept recommended maize technological package, then yields from varieties must be stable. Hence, breeders should develop stable varieties that can stand the test of weather and diseases. Recycling of maize seed is a common practice among farmers. There is need to undertake research on maize seed recycling for loss assessment. Perhaps, this information when made available to farmers may change their attitude to purchasing improved seeds from seed dealers. Maize technology recommendations were made based on past agronomic data. With changing economic and climatic conditions, some of the recommendations may have been outdated or inappropriate. Therefore, it is necessary to review and update the recommendations for growing maize in Ghana. This will also enable researchers to modify their research activities to produce competitive recommendations for formers. Impact assessment studies usually overlook many benefits generated by research projects because it is not easy assigning economic value to them. If all these were included, then the benefits would be much higher than expected. Therefore, even if such benefits cannot be measured, a descriptive assessment is a useful complement to these quantitative estimates. 142 University of Ghana http://ugspace.ug.edu.gh In interpreting these findings ^ it is important to remember that this study has focused only on economic benefits of research. Other benefits have not been taken into account, even though these have undoubtedly been considerable. These include: (a) Improved nutritional status: One of the objectives of the GGDP was to improve the nutritional status of rural households and maize consumers in general. This is because by increasing maize production, it is expected that there will be improved levels of food consumption and more so when maize is important in the diet of most Ghanaians and as a weaning food for children. The feet that protein-quality maize was released to farmers and that most farmers planted this variety emphasizes the importance farmers and researchers are paying to nutritional impact. (b) Improved research efficiency: The GGDP was instrumental in introducing an integrated national strategy for technology generation, testing and diffusion that involved a large number of research and extension organizations. This approach has been adopted by other research and extension projects. (c) Strengthened institutional capacity: The GGDP served as a model for collaboration between key players in maize technology development process, including national agricultural research and extension organizations (CRI, MOFA, GLDB), international agricultural research centres (CIMMYT, IITA) and donor agencies (CIDA). Therefore, the GGDP has had the experience to manage other new projects when available. (d) Better-trained human capital: During the life of the GGDP, hundreds of Ghanaian researchers and extension officers received various forms of training. The effects of this training will long outlive the Project. Crops Research Institute, MOFA and the 143 University of Ghana http://ugspace.ug.edu.gh national agricultural extension service will continue to benefit from having better-trained personnel and continue to have impact on Ghanaian agriculture. (e) Gender issues: The GGDP was one of the first agricultural research and extension project that incorporated gender issues in its research programmes. All CRI research projects and programmes have had spill-over effects of this and are following suit. (f) Increased employment: The GGDP also generated employment for those employed in the inputs distribution industries, such as seed growers, seed distributors, fertilizer importers and distributors. (g) Improved policy information: The GGDP generated a substantial amount of information that can and is still put to good use by policy makers, extension officials, research managers, CRI scientists, input distributors, grain traders and others in their day- to-day activities. One of the problems of impact assessment is with data particularly on national production figures and yield gain from the new technology. Since these can affect the estimation of internal rate of return, it is important that more attention and resources are given to them for quality data. For crop technology, it is imperative that a clear scenario of “with-and-without” technologies are continuously included in on-farm trials and should be farmer- managed as much as possible. This will provide a better assessment of yield gains. An evidence of the advantages of collaborating with intemational research centres is that the GGDP took advantage of CIMMYT and IITA germplasms and avoided a substantial portion of possible expenses in the development of improved varieties. This spill-over from international centres is necessary and collaboration of national research 144 University of Ghana http://ugspace.ug.edu.gh systems and the international centres should be encouraged. This collaborative effort has resulted in the high rates of return for maize research in a relatively short time. This experience goes to emphasize that if research methods or outcomes can be borrowed from other institutions at no or very little cost and are adaptive and user friendly, it will be better to do so. This will not only save time, resources and money but will provide rapid returns to investment. Furthermore, the high rate of return to maize research underscores the role of international cooperation in making National Research Systems (NARS) cost- effective. Extension policy issues: Information about improved technologies ought to reach more farmers. Improved seed has had a fairly extensive adoption in much of the maize-growing areas in Ghana, whereas fertilizer technology has been rather less successful There is need for extra effort to speed up dissemination of fertilizer technologies to farmers for ultimate adoption. Women farmers are not being reached as effectively as men farmers. Since women are usually the disadvantaged in the society, this calls for some review on the technology transfer strategy. About 36% of farmers actually use quality protein maize or other improved varieties as weaning food but because of its nutritional value, additional efforts are needed to publicize benefits of quality protein maize. With maize recommendations on planting practices, the adoption of maize practices as a package seems to be less profitable unless yields and fertilizer prices are stable. In particular, the adoption of all the three components of the package (improved 145 University of Ghana http://ugspace.ug.edu.gh variety, fertilizer and row p la n t i n g ) together is not profitable in the Guinea savannah zone. However, a piecemeal adoption of variety or fertilizer is profitable among all the ecological zones. This approach is consistent with small-scale farmers’ stepwise adoption behaviour. Thus, formers should be encouraged to select and adopt portions of the recommended package for growing maize appropriate to their conditions and circumstances. The adoption of improved maize technology is generally low in the forest zone but this zone has more agricultural districts than the other ecological zones. In addition, the forest zone has a reasonably good climate for growing maize. If adoption of improved maize is to be expanded, then the forest zone should receive equal attention as others. The marginal rate of return to investing in improved maize in the forest zone is high ranging from 672% to 3030%. Therefore, extension activities for the adoption of improved maize varieties in the forest zone should be given a priority. Input policy issues: Efficient supply of inputs is probably one of the key elements of policy environment necessary to facilitate technology transfer. When farmers begin to use purchased inputs, the public and private sectors have a role in promoting the use of the inputs by supplying them on timely basis. Support such as credit from the public sector in particular is necessary to enhance the adoption for purchased inputs such as fertilizer. Furthermore, most input stores are located in bigger towns farther away from the producing areas. It is necessary to improve the input delivery system by siting input shops 146 University of Ghana http://ugspace.ug.edu.gh closer to the producing areas for easy access to these inputs. In addition, the role of farmer organizations should be strengthened to improve access to input. Although the seed delivery system for improved varieties have shown significant improvements, there is room for further strengthening. A critical requirement for more widespread adoption of maize varieties is an effective improved seed delivery system. Although there is usually a fanner-to-farmer seed distribution, this facility is inadequate. A well-developed seed production and distribution system is important for smooth adoption of improved technologies. Additionally, formation of farmer organizations for acquisition of inputs and other agricultural services should be encouraged. The maize adoption study revealed that fertilizer adoption is generally low in all the ecological zones. Considering the feet that yields are still low by international standards, more continuous cropping will persist and fertilizer prices might continue to rise, there is the need to explore ways to reduce the cost of chemical fertilizer or new and profitable fertility improvement technologies. With continuing attention to fertilization, yields can be increased further. Concluding remarks Some conclusions emerge from the study. First, the adoption of improved maize varieties has maintained its momentum in Ghana. Second, GGDP-generated varieties and other recommendations continue to be used extensively and third, investment in maize research and extension has continued to provide high rate of return. The results of this study raise some issues that must be addressed by CRI. The fact that the improved varieties have been adopted widely places special responsibility on CRI 147 University of Ghana http://ugspace.ug.edu.gh breeders to maintain and widen the genetic diversity of improved maize. With more varieties being released from different crosses, diversity such as disease resistance increases. Efforts to promote greater diversity will require complete awareness by breeders and stronger extension support. Furthermore, improved seed systems are needed to promote a much fester replacements of varieties over time. Another issue is whether these gains can be maintained in the liiture. There are opportunities for the expansion of the area under improved varieties and accompanying recommendations in all the ecological zones particularly the forest zone. However, the expansion and adoption of improved technologies in the marginal areas will be slow and the impacts modest. More attention should be given to fertility issues in the Guinea Savannah zone, transition zone and coastal savannah zone. In more favorable areas, a major source of growth will be genetic yield gains in areas already sown to improved maize. Yields are still low by intemational standards and can still be increased. The gains in developing resistant varieties, increased yield potential and grain quality will make important contributions to yield stability and the overall food security for most Ghanaians. In Ghana, public investments in maize research and extension have generated attractive rates of return. Investment in agriculture may just be as profitable as investing in other sectors of the Ghanaian economy. Finally, this study has contributed to the growing body of literature on research impact assessment and evaluation. 148 University of Ghana http://ugspace.ug.edu.gh BIBLIOGRAPHY Adesina Akinwumi A and Moses M. Zinnah. ‘Technology Characteristics, Farmers’ Perceptions and Adoption Decisions: A TOBIT Model Application in Sierra Leone.” Agricultural Economics 9(1993). Adesina, A. A. and Baidu-Forson, J. ‘Farmers’ Perception and Adoption of New Agricultural Technology: Evidence from Analysis in Burkina Faso and Guinea, West Africa.” Agricultural Economics 13(1995). Akino Masakatsu and Yujiro Hayami. "Efficiency and Equity in Public Research: Rice Breeding in Japan's Economic Development" American Journal o f Agricultural Economics VoL57. (1975). Alderman Harold and Gerald Shively. “Economic Reform and Food Prices: Evidence from Markets in Ghana.” World Development. 24(1996). Alderman Harold and Paul Higgins. “Food and Nutritional Adequacy in Ghana. ” Cornell Food and Nutrition Policy Program. Working paper 27. 1992. Alston Julian M,. George W. Norton and Philip G. Pardey. Science Under Scarcity: Principles and Practice for Agicultural Research Evaluation and Priority Setting. Cornell University Press. 1995. Alston Julian M., Geoff W. Edwards, and John W. Freebaim. “Market Distortions and Benefits from Research.” American Journal o f Agricultural Economics. 9(1988). Anandajayasekeram,P, D. R. Martella and M Rukuni. A Training Manual for Biological Scientists on Impact Assessment o f Investments in Agricultural and Natural Resources Research. Southern African Center for Coorperation in Agricultural and Natural Resources and Training (SACCAR). 1996. Anandajayasekeram, P., DR. Martella, Rukumi M. A Training Manual on Research and Development Evaluation and Impact Assessments o f Investments in Agricultural and Natural Resources Research. SACCAR. 1996. Anandajayasekeram P .D. R. Martella and M. Rukuni. A Training Manual on R and D Evaluation cmd Impact Assessment o f Investments in Agricultural and Natural Resources Research. SACCAR. 1996. Anderson, J.R., and R.W Herd! “Reflection on Impact Assessment.” Proceedings of the ISNAR/Rutgers Agricultural Technology Management Workshop, 6-8 July, 1988. Rutgers University, New Jersey. Vol.2. Assessing the Impact of Agricultural Research, ISNAR. The Hague. The Netherlands. 1990. Asenso Okyere,W., F.A Asante, L. Oware Gyekye."Case Studies on Rural Poverty Alleviation in the Commonwealth. Ghana.” Food and Rural Development Division, Commonwealth Secretariat, London. (1992). 149 University of Ghana http://ugspace.ug.edu.gh Asiedu-Saforo Kwame. “Economic Reform Programmes and Agricultural Development: Macro Policy Sequencing in Ghana, 1983-88.” Food Policy. 14 (1989). Boateng, E.O., K Ewusi, R Kanbur, A McKay. "A Poverty Profile for Ghana, 1987-1988. Social Dimensions of Adjustment in Sub-Sahara Africa". Working Paper No.5. The World Bank, Washington D. C. (1990). Boserup, Ester. “The Conditions of Agricultural Growth: The Economics of Agrarian Change under Population Pressure. New York: Aldine Publishing. 1965. Bumb B.L., J.F. Teboh, J.K Atta and W.K. Asenso-Okyere (1994). " Ghana Policy Environment and Fertilizer Development.” IFDC. 1994. Byerlee, D , and P. Moya. Impacts o f International Wheat Breeding Research in the Developing World, 1966-90. Mexico D.F.: CIMMYT. 1993. Byerlee, D., with P, Anandajayasekeram, A Diallo, Bantanyu Gelaw, P. W. Heisey, M, Lopez Pereira, W, Mwangi, M. Smale, R Tripp, and S. Waddington. Maize Research in Sub- Saharan Africa: An Overview o f Past Impacts and Future Prospects. CIMMYT Economics Working Paper. 94-03. Mexico, D.F.: CIMMYT. 1994. CIMMYT Economics Program. The Adoption o f Agricultural Technology: A Guide for Survey Design. Mexico, D.F.: CIMMYT. 1993. Clarke G.M and D. Cooke. A Basic Course in Statistics. 2nd ed. Edward Arnold (publishers) Ltd. London. 1983. CSIR and ISNAR,. "Review of the Ghana Agricultural System". VoL 1: Report. The Hague: ISNAR 1991. Dankyi A A and AOApau. “Maize Marketing Channel Study” GGDP. 1990. Dankyi, A A , H A Dakurah and J.N. Asafo-Agyei “Cowpea Technology Adoption in Ghana” Paper presented at the OAU/SAFGRAD Workshop, Abidjan, Ivory Coast. 1993. Dankyi, A A , V.M. Anchirinah and AO. Apau “The Adoption of the Improved Rice Technologies in the Wetlands of the Ashanti Region of Ghana.” Unpublished Paper. West Africa Rice Development Association (WARDA). Bouake, Ivory Coast, 1996. Dankyi, A A , V.M. Anchirinah, AO. Apau.. " Profitability and Impact of Extension Test Plot on Maize Production in Ghana". GGDP/SG2000/DAES Collaboration. GGDP Memio. (1995). Edmeades, G., A ADankyi, K. Marfo and R Tripp. “On-Farm Maize Research in the Transition zone of Ghana.” In Planned Change in Farming Systems. Edited by R Tripp. 1991. Wiley-Sayce Co-Publication. 150 University of Ghana http://ugspace.ug.edu.gh Ekpere J. A. “Towards a New Orientation to the Challenge of Technology Transfer Systems and Food Production in the 21st Century Aftica.” Paper presented at the Regional Workshop. Technology Options and Transfer Systems. Abidjan, Cote DTvoire.April 26-28, 1995. pp, 1- 26. Evenson RE. “ Economic Impact Studies o f Agricultural Research and Extension. Yale University. 1999. GGDP Annual Report. Various Issues GGDP. A Study o f Maize Technology in Ghana. Mexico, D.F, GGDP. 1991. GGDP. Maize and Legumes Production Guide. 1996. Ghana Grains Development Project. Fourth Annual Report. Research Results. 1982. Ghana Living Standards Survey 3. 1995. Ghana Statistical Service. “Rural Communities in Ghana.” Report of the National Rural Community Survey. Ghana Living Standards Survey. 1993. _______ Quaterly Digest of Statistics. June 1995 Gittinger Price J. Economic Analysis o f Agricultural Projects. Johns Hopkins University Press. 2nd Ed 1982. Greene William, H Econometric Analysis. MacMillan Publishers. 1995. Griliches Zvi “Research Expenditures, Education and Aggregate Agricultural Production.” The American Economic Review. Vol, LTV (1964). Griliches Zvi. “Hybrid Com: An Exploration in the Economics of Technological Change.” Econometrica. 25 (1957). Griliches, Zvi “ Research Costs and Social Returns: Hybrid Com and Related Innovations.” Journal o f Political Economy. VoL 66. No. 5. October, 1958. Gujarati Damodar, N., Basic Econometrics. 3rd ed. McGraw-hill, Inc. 1995. Howard Julie, A., with George M. Chitalu, ans Sylvester M. Kalenge. The Impact o f Investment in Maize Research and Dissemination in Zambia. ” MSU Intemational Development Working Paper. No. 39/1. 1993 Intemational Monetary Fund. Intemational Financial Statistics Yearbook. Washington DC. 1993. Intemational Financial Statistics, 1998. Kivlin, J.E. and F.C. FliegeJ , “ Differentiated Perceptions of Innovations and Rate of Adoption.” Rural Sociology 32 (1967). 151 University of Ghana http://ugspace.ug.edu.gh La-Anyane, S. " Research in Agriculture in Relation to National Economic Development". CSIRAJniversitiesAJ. S. Academy of Sciences Workshop, January 18. 1971. Laker-Ojok Rita. “The Rate of Return to Agricultural Research in Uganda: The Case of Oilseeds and Mail®.” MSU International Development Working Paper. No.42. 1994. Masters W. A , Touba Bedingar and James F. Oehmke. "The Impact of Agricultural Research in Africa: Aggregate and Case Study Evidence". Paper presented at the 23rd International Conference of Agricultural Economists. August 10-16,1997, Sacramento, California. (1997). Masters W.A,Bakari Coulibaly,Diakalia Sanago, Mamadou Sidibe, Anne Williams, John H. Sanders and Jess Lowenberg-DeBoer. The Economic Impact o f Agricultural Research: A Practical Guide. Department of Agricultural Economics. Purdue University. 1996. Mazzucato Valentia and Samba Ly. “An Economic Analysis of Resource and Technology Transfer of Millet, Sorghum and Cowpeas in Niger.” MSU International Development Working Paper. No, 40.1994. Melinda Smale and Paul W. Heisey. “Maize of the Ancestors and Modem Varities: The Microeconomics of High-Yielding Variety Adoption in Malawi” Economic Development and Cultural Change. 1995. Morris, M.L., H,J. Dubin and T. Pokhred. "Returns to Wheat Research in NepaP'. CIMMYT Economics Working Paper 92-04. Mexico, D.F.: CIMMYT. (1992). Moshi A , P.Anandajayasekeram, A Kaliba, D. Martella, W.Mwangi, and F. Shao. Economic Impact o f Maize Research in Tanzania. Southern African Center for Coorperation in Agricultural and Natural Resources (SACCAR). 1997. Mudhara, M., P, Anandajayasekeram, D. Martella E. Mazhangara and M. Murata. Impact Assessment o f Soybean Research and Enabling Environment in Zimbabwe: 1965-2000. Department of research and Specialist Services, Ministry of Agriculture, Zimbabwe. SACCAR 1997. Nichola Tennassie and John H Sanders. “A Probit Analysis of the Determinants of Adoption when Inputs are Rationed: The Gezira Experience with Hybrid Sorghum” Science, Technology and DevelopmentNo\\4.no.3, 1996. pp. 107-119. Nkonya, E., P. Xavery, H Akonaay, W. Mwangi, P. Anandajayasekeram, H Verkuiijl, D. Martella, and A Moshi Adoption o f Maize Production Technologies in Northern Tanzania. Mexico D.F.: CIMMYT, The United Republic of Tanzania and the Southern African Center for Cooperation in Agricultural Research. 1998. Norton G. W.and Jeffrey S. Davis."Evaluating Returns to Agricultural Research: A Review.” American Journal o f Agricultural Economics 63(4) 1981. 152 University of Ghana http://ugspace.ug.edu.gh Oehmke James F.and Eric W, Crawford.’The impact of Agricultural Technology in Sub- Saharan Africa: A Synthesis of Symposium Findings.” Michigan State University International Development Papers. No. 14,1993. Oehmke, J., L. Daniels, J. Howard, M. Maredia, R. Bemsten, 1992. "The Impact of Agricultural Research: A Review of the Ex-post Assessment Literature with Implications for Africa. Staff Paper No. 92-38. Department of Agricultural Economics. Michigan State University. Oehmke, James F., and Eric Crawford.. "The Impact of Agricultural Technology in Sub- Saharan Africa: A Synthesis of Symposium Findings". MSU Intemational Paper No. 14. East Lansing: Department of Agricultural Economics, Michigan State University. (1993). Rahtn Michael, R and Wallace E. Huffman “ The Adoption of Reduced Tillage: The Role of Human Capital and Other Variables.” Australian Journal o f Agricultural Economics, 66(1984). Rogers, E.M. Diffusion o f Innovations. Free Press. Glencoe. Illinois. 1962 Sanders John H , Taye Bezuneh, and Alan C. Schroeder. (1994). " impact Assessment of the SAFGRAD Commodity Networks." US AID/AHR/O AU/STRC-SAFGRAD. 1994 pp 1-28. Schuh Edward G. and Hellio Tollini.’’Costs and Benefits of Agricultural Research.” World Bank Staff Working Papers. The World Bank, no.360. 1979. Schwartz, L.A, Stems. J. A., and Oehmke, J.F. “Economic Returns to Cowpea Research, Extension, and Input Distribution in Senegal.” Agricultural Economics. 8 (1993). Shively Gerald E. “Consumption Risk, Farm Characteristics and Soil Conservation Adoption among Low-income Farmers in the Philipines "Agricultural Economics. 17(1997). Stems James, A , and Richard H Bemsten. “Assessing the Impact of Cowpea and Sorghum Research and Extension in Northern Cameroun.” MSU Intemational Development Working Paper. No.43. (1994). Stryker Dirck, J. with assitance of E. Dumeau, Jennifer Wohl, P. Raymond, Andrew Cook, and Katherine Coon. “ Trade, Exchange Rate, and Agricultural Pricing Policies in Ghana.” The World Bank. Washington D.C. 1990. Tennasssie Nichola. “The adoption of Hybrid Sorghum at the Farm and Regional Levels in the Gezira Irrigated Scheme of Sudan”. Unpublished PhD Thesis. Department of Agricultural Economics, Purdue University. West Lafayette, USA Timothy J. Dalton. “Soil Degradation and Technical Change in Southern Mali.” Unpublished PhD Thesis. Department of Agricultural Economics, Purdue University, West Lafayette, USA Tripp Robert. “Maize Seed Management Study in the Brong-Ahafo and Volta Regions of Ghana.” GGDP. 1996. 153 University of Ghana http://ugspace.ug.edu.gh Tripp,R, K. Maxfo, A. A. Dankyi, and M. Read. Changing Maize Production Practices o f Small-scale Farmers in the Brong-Ahafo Region, Ghana. Kumasi: Ghana Grains Development Project. 1987. Twumasi-Afriyie S. ''Achievements of the Crops Research Institute.” CRI Ghana. 1995. Voon Jan P., and Geoff W. Edwards. “The Calculation of Research Benefits with Linear and Nonlinear Specifications of Demand and Supply Functions. ” American Journal o f Agricultural Economics. 73 (1991). West Africa Rice Development Association (WARDA). “An Introduction to Impact Assessment Research.” RETWA Methodology Series. 1997. White Fred, C. and Joseph Havficek. Jr. “Optimal Expenditures for Agricultural Research and Extension: Implications of Underfunding.” American Journal o f Agricultural Economics. 64 (1982). 154 University of Ghana http://ugspace.ug.edu.gh APPENDIX 1 MAIZE SURVEY QUESTIONNAIRE 155 University of Ghana http://ugspace.ug.edu.gh Adoption and impacts of improved maize technologies in Ghana Collaborating institutions: Crops Research Institute (CRI) International Maize and Wheat Improvement Center (CIMMYT) Canadian International Development Agency (CIDA) Overseas Development Institute (ODI) Survey questionnaire Enumerator #1 Enumerator #2 Interview date Questionnaire ID Region District Enumeration Area Village Ecological zone (i) (2) (3) (4) (5) (1 - forest, 2 - transition, 3 - coastal savannah, 4 - Guinea savannah) Respondent’s name 156 University of Ghana http://ugspace.ug.edu.gh A.l Household membership A. Farmer’s household characteristics 1. Size of household includes: Adult males (6) (7) Adult females (8) Children under 16 A.2 Farmer’s resource ownership 2. Land fanned (all crops) in 1997(ac): - owned (ac) (10) Bicycles (14) - sharecropped (ac) ( i i ) Motorcycles (15) - rented (ac) (12) Vehicles (16) - other (ac) (13) Tractors (17) Cattle Pigs Sheep/goats (21)3. Does your house have electricity? (1 =yes, 2 = no) A. 3 Farmer’s sources o f income 4. Of the crops that you grow, please list those that provide the most cash income (list in order of importance) Source 1 (22) Source 2 (23) Source 3 (24) 1 = maize, 2 = rice, 3 = sorghum, 4 = millet, 5 = cassava, 6 = yam, 7 = cocoyam, 8 = plantain, 9 = cowpea, 10 = soybean, 11 = groundnuts, 12 = cocoa, 13 = coffee, 14 = tobacco, 15 = cotton, 16 = vegetables, 17 = oil palm, 18 = citrus, 19 = other, 88 = N/A (9 ) (18) (19) (20) 5. Please list your most important sources ofnon-agricultural income (list in order of importance) Source 1 (25) Source 2 (26) 1 - wage labor in agriculture, 2 = wage labor elsewhere, 3 remittances from relatives, 4 = sales o f animals, 5 = trading, 6 = transport, 7 = tailoring, 8 = carpentry, 9 = other, 88 = N/A 6. What is your most important source of income? (check one) Agriculture (27) Non-agriculture (28) 157 University of Ghana http://ugspace.ug.edu.gh 7. Please indicate your use of agricultural credit during 1997: A.3 Use o f agricultural credit Crop Source of credit Type Use of credit (29) (30) (31) (32) (33) (34) (35) (36) (37) (38) (39) (40) (41) (42) (43) (44) (45) (46) (47) (48) (49) (50) (51) (52) (53) Crops: 1 = maize, 2 = rice, 3 = sorghum, 4 = millet, 5 - cassava, 6 = yam, 7 = cocoyam, 8 = plantain, 9 = cowpea, 10 = soybean, 11 = groundnuts, 12 = cocoa, 13 = coffee, 14 = tobacco, 15 = cotton, 16 = vegetables, 17 = oil palm, 18 = citrus, 19 = other, 88 =N/A Sources of credit: 1 = government program, 2 = private bank, 3 = family member, 4 = neighbor, 5 = money lender, 6 = credit society, 7 = other, 88 = N/A Types of credit: 1 = cash, 2 = kind, 88 = N/A Uses of credit: 1 = purchase seed, 2 = purchase fertilizer, 3 = purchase herbicide, 4 = purchase pesticide, 5 = hire labor, 6 = hire tractor, 7 = other, 88 = N/A A.4 Contact with extension services and technical assistance organizations Maize 8. Number of times in the last three years that you discussed your farming activities with an extension officer 9. Number of field days attended in the last three years 10. Ever gave land for a maize demonstration plot? (1 =yes, 2 = no) Other (54) (56) (58) (55) (57) B. Maize varietal use and production practices Bl Maize cropping activities 11. How many years have you been growing maize? 12. How many maize fields did you plant in 1997? Major season: Minor season: Number of fields Number of fields (59) (so) Total area (ac) (62) Total area (ac) (61) (63) 158 University of Ghana http://ugspace.ug.edu.gh B.2 Management o f maize fields (1997 cropping season) 13. Please provide the following information about your 1997 maize production practices. Major season (up to three fields) Minor season Maize field (largest fields) Field 1 Field 2 Field 3 Field 4 Size of field (ac) (64) (82) (100) (118) Before 1997 , how many years continuously cropped? (65) (83) (101) (119) Before 1997 , how many years Mow? (66) (84) (102) (120) Land tenure (1 = own field, 2 = share crop, 3 = rent for cash, 4 = other) (67) (85) (103) (121) First maize variety (I = local, 2 = agric, 3 = La Posta, 4 = Aburotia, 5 = Safita, 6 = Kawanzie, 7 = Golden Crystal, 8 = Okomasa, 9 = Dobidi, 10 = Abeleehi, 11 =Dorke, 12 = Obatanpa, 13 - other) (68) (86) (104) (122) Year in which this variety first planted (69) (87) (105) (123) Second maize variety (1 = local, 2 = agric, 3 = La Posta, 4 = Aburotia, 5 = Safita, 6 = Kawanzie, 7 = Golden Crystal, 8 = Okomasa, 9 = Dobidi, 10 = Abeleehi, 11 =Dorke, 12 = Obatanpa, 13 = other, 88 = N/A) (70) (88) (106) (124) Year in which this variety first planted (71) (89) (107) (125) Main intercrop (0 = none, 1 = cassava, 2 = sorghum, 3 = millet, 4 = plantain, 5 = cocoyam, 6 = rice, 7 = cowpea, 8 = groundnuts, 9 = other) (72) (90) (108) (126) Land preparation practice (1 = manual, 2 = animal, 3 = tractor, 4 = other) (73) (91) (109) (127) Planting date (fortnight/month) (74) (92) (110) (128) Planting pattern (1 = lines, 2 = ridge, 3 = random) (75) (93) (111) (129) First type of fertilizer used (0 = none used, 1 = urea, 2 =ammonium sulphate, 3 = compound, 4 = manure) (76) (94) (112) (130) Second type of fertilizer used (0 = none used, 1 = urea, 2 =ammonium sulphate, 3 = compound, 4 = manure) (77) (95) (113) (131) Herbicide used? (1 = yes, 2 = no) (78) (96) (114) (132) Pesticide used on maize in field? (1 = yes, 2 — no) (79) (97) (115) (133) Pesticide used on maize in storage? (1 = none, 2 =actellic, 3 = other) (80) (98) (116) (134) Number of maxi bags harvested (81) (99) (117) (135) Local varieties: Field 1: Field 3: (136) Field 2: (138) Field 4: 159 (137) (139) University of Ghana http://ugspace.ug.edu.gh B.3 Decision making responsibility fo r largest maize field (1997 cropping season) 14. Please provide the following information about your 1997 maize fields. Who took decision? Who performed task? Hired labor cost (c/day) • obtaining land (140) !p iSS !lp i • selecting maize variety (141) S i l II 8® 1 • obtaining seed (142) (143) (144) X X i p • obtaining fertilizer (145) (146) (147) • preparing land (148) (149) (150) (151) • planting (152) (153) (154) (155) • applying fertilizer (156) (157) (158) (159) • weeding (160) (161) (162) (163) • harvesting (164) (165) (166) (167) • selling grain (168) (169) (170) (171) • use of cash from maize sales (172) 181111! '.1 ix '/h : C-Jfy - &£& Person responsible for decision: 1 = respondent, 2 = respondent’s spouse, 3 = both, 4 = other, 88 = N/A Task performed by: 1 = respondent, 2 = respondent’s spouse, 3 = respondent and spouse, 4 = other family members, 5 = hired labor, 6 = exchange labor, 7 = other, 88 = N/A B.4 Fertilizer use on maize 15. What was the most recent year in which you used fertilizer on maize? _____ 073) (Never = 00) 16. For the most recent year in which you used fertilizer on maize, please indicate: Size of field (ac) (174) Years continuously cropped (before that year) (175) 17. Please describe your fertilizer practices in that year: Fertilizer type Days after planting Rate (bags/acre) Method (176) (177) (178) (179) (180) (181) (182) (183) (184) (185) (186) (187) Fertilizer type: 1 = urea, 2 = ammonium sulphate, 3 = compound, 4 = manure, 5 = other, 88 = N/A Fertilizer application method: 1 = applied on sioface, 2 - buried 160 University of Ghana http://ugspace.ug.edu.gh • In that year, how many acres of maize did you plant? • Of those acres, how many received fertilizer? • If you formerly used fertilizer but no longer use it, why did you stop? 18. What was the first year in which you used fertilizer on maize? (Never = 00) (189) (190) (191) (188) 1 = don't need (land fertile), 2 = too expensive, 3 = can’t get, 4 = credit not available, 5 = other If you have never used fertilizer on maize, why not? 19. What was the first year in which you row planted maize? • In that year, how many acres of maize did you plant? • Of those acres, how many were row planted? • If you have never row planted maize, why not? (192) (Never = 00) (193) (194) 20. What was the first year in which you planted improved maize? • In that year, how many acres of maize did you plant? • Of those acres, how many were planted to the improved variety? • What was the name of the improved variety? (195) (Never = 00) (196) (197) (198) 1 - local, 2 = agric, 3 = La Posta, 4 = Aburotia, 5 = Safita, 6 = Kawanzie, 7 = Golden Crystal, 8 = Okomasa, 9 = Dobidi, 10 = Abeleehi, 11 =Dorke, 12 = Obatanpa, 13 = other, 88 = N/A If you have never planted an improved maize variety, why not? 161 University of Ghana http://ugspace.ug.edu.gh C.1 Maize grain marketed in 1997 21. Please provide information about sales of the maize grain you harvested in 1997. C. Maize utilization patterns Code How many bags harvested? How many bags sold? Proportion sold (%) Variety 1: (199) (200) (201) (202) Variety 2: (203) (204) (205) (206) Variety 3: (207) (208) (209) (210) Variety 4: (211) (212) (213) (214) 1 = local, 2 = agric, 3 = La Posta, 4 = Aburotia, 5 = Safita, 6 = Kawanzie, 7 = Golden Crystal, 8 = Okomasa, 9 = Dobidi, 10 = Abeleehi, 11 =Dorke, 12 = Obatanpa, 13 = other, 77 = all varieties combined, 88 — N/A C.2 Household maize grain sales in 1997 22. Please provide information about your 1997 maize grain sales. Quantity sold (bags) Average price (cedis/bag) Buyer Distance transported (km) January (215) (227) (239) (251) February (216) (228) (240) (252) March (217) (229) (241) (253) April (218) (230) (242) (254) May (219) (231) (243) (255) June (220) (232) (244) (256) July (221) (233) (245) (257) August (222) (234) (246) (258) September (223) (235) (247) (259) October (224) (236) (248) (260) November (225) (237) (249) (261) December (226) (238) (250) (262) Buyer: 1 - neighbor, 2 = trader in your village, 3 = trader in other place, 4 = other 23. In 1997, did you feed maize to poultry? (263) No. of maxi bags (1 = yes, 2 = no) 24. In 1997, did you feed maize to pigs? (265) No. of maxi bags (264) (266) (1 = yes, 2 = no) 162 University of Ghana http://ugspace.ug.edu.gh 25. If you are tliinldng about planting a new maize variety, how important are the following characteristics? D. Maize varietal preferences D.l Farmer’s rating o f the importance o f varietal attributes (in general) Yield (267) Early maturity (268) Drought tolerance (269) Insect resistance (270) Lodging tolerance (271) Grain size (272) Grain texture (273) Milling quality (274) Taste (275) Chafiness (276) Nutritional value (277) Storage quality (278) Market price (279) Importance of characteristics: 1 = very important, 2 = somewhat important, 3 = not important D.2 Farmer’s rating o f specific varieties 26. Please rate the performance of the varieties you currently are growing or used to grow with regard to the varietal characteristics. Improved varieties Local varieties Now Now Used to Now Used to growing (1) growing (2) grow growing (1) grow Variety name Variety code (280) (294) (308) (322) (336) Yield (281) (295) (309) (323) (337) Early maturity (282) (296) (310) (324) (338) Drought tolerance (283) (297) (311) (325) (339) Insect resistance (284) (298) (312) (326) (340) Lodging tolerance (285) (299) (313) (327) (341) Grain size (286) (300) (314) (328) (342) Grain texture (287) (301) (315) (329) (343) Milling quality (288) (302) (316) (330) (344) Taste (289) (303) (317) (331) (345) Chafiness (290) (304) (318) (332) (346) Nutritional value (291) (305) (319) (333) (347) Storage quality (292) (306) (320) (334) (348) Market price (293) (307) (321) (335) (349) Varieties: 1 = local, 2 = agric, 3 = La Posta, 4 = Aburotia, 5 = Safita, 6 = Kawanzie, 7 = Golden Crystal, 8 = Okomasa, 9 = Dobidi, 10 = Abeleehi, 11 =Dorke, 12 = Obatanpa 13 = other, 88 = N/A Performance ratings: 1 = superior, 2 = acceptable, 3 = poor 163 University of Ghana http://ugspace.ug.edu.gh E. Maize seed acquisition and management practices E.l Maize seed acquisition practices 27. The following table refers to seed planted in 1997 (fields described in section B. 1): Note 1: I f two varieties were grown in the same field, enter in separate columns. Note 2: I f the same variety was planted in two seasons, enter in separate columns (if seed used in 1997 minor season was savedfrom 1997 major season, fill in only first three lines). Variety 1 Variety 2 Variety 3 Variety 4 Variety (1 = local, 2 = agric, 3 = La Posta, 4 = Aburotia, 5 = Dobidi, 6 = Abeleehi, 7 = Safita, 8 = Golden Crystal, 9 =Dorke, 10 = Kawanzie, 11 = Obatanpa, 12 = Okomasa, 13 = other, 88 = N/A) (350) (365) (380) (395) Season planted (1 = major, 2 - minor, 3 = other) (351) (366) (381) (396) Source of seed used in 1997 (1 =farm saved, 2 = mother farmer, 3 =extension,4 = SG2000, 5 = inputs shop, 6 = grain market, 7 = other) (352) (367) (382) (397) IF FARM SAVED: Maize seed selection practice (1 = infield before harvest, 2 = after harvest before storage, 3 = at home after storage) (353) (368) (383) (398) Selected by whom (1 = man, 2 = woman, 3 = both) (354) (369) (384) (399) Original source of seed (2 = another farmer, 3 = research or extension, 4 = SG2000, 5 = seed dealer, 6 = grain market, 7 = other, 8 = unknown) (355) (370) (385) (400) Year acquired this seed (not variety) (356) (371) (386 ) (401) FOR ALL SEED: Quantity seed (kg) originally acquired (Note: 1 bowl = 2.5 kg) (357) (372) (387) (402) How originally acquired? (I = purchase, 2 = gift, 3 = loan) (358) (373) (388) (403) If purchased, price (cedis/kg) (359) (374) (389) (404) Why did you acquire new seed? (1 = to try new variety, 2 = had planted variety previously [old seed available but needed fresh seed], 3 = no other seed mailable, 4 = participated in extension activity, 5 = other) (360) (375) (390) (405) Distance from seed source (in km) (361) (376) (391) (406) Was seed in a sealed bag? (1 = yes, 2 = no) (362) (377) (392) (407) Did you ask for this variety by name? (1 = yes, 2 = no) (363) (378) (393) (408) How did you first learn about variety? (2 =from another farmer, 3 =from research or extension, 4 =from trader, 5 = from radio, 6 = other) (364) (379) (394) (409) 164 University of Ghana http://ugspace.ug.edu.gh E.2 Past use o f improved maize seed 28. Have you ever wished to obtain improved maize seed from a dealer and discovered that none was available? • If yes, most recent year • If former answers 1996, most recent year prior to 1996 • How many dealers did you visit in your search for seed? (410) (1 = yes, 2 = no) Which variety were you looking for? (411) (412) (413) (414) 1 = local, 2 = agric, 3 = La Posta, 4 = Aburotia, 5 = Safita, 6 = Kawanzie, 7 = Golden Crystal, 8 = Okomasa, 9 = Dobidi, 10 = Abeleehi, 11 =Dorke, 12 = Obatanpa, 13 = other, 88 = N/A Were you able to get seed anyway from another source? (415) If no, which variety did you end up planting? (1 = yes, 2 = no, 88 = N/A) (416) 1 = local, 2 = agric, 3 = La Posta, 4 = Aburotia, 5 = Safita, 6 = Kawanzie, 7 = Golden Crystal, 8 = Okomasa, 9 = Dobidi, 10 = Abeleehi, 11 =Dorke, 12 = Obatanpa, 13 = other, 88 = N/A 29. If you buy maize seed from a dealer, what do you do to try to ensure good quality? Buy seed from known dealer (417) Buy seed from known seed producer Look for certification label (418) (1 = always, 2 = sometimes, 3 = never, 88 = N/A) (1 = always, 2 = sometimes, 3 = never, 88 = N/A) (419) (1 = always, 2 = sometimes, 3 = never, 88 = N/A) 165 University of Ghana http://ugspace.ug.edu.gh F. Impacts of MV adoption F.l Incremental production attributable to modern varieties (Refer to the respondent’s main maize field. Fill in the box on the first line based on the answers given in Section B.2. Then ask farmer to estimate amounts in the other two boxes.) 30. Without fertilizer: In a normal year, how many bags does_______________ yield without fertilizer? ( fa rm e r ’s c u r r e n t im p ro v e d v a r ie ty ) (420) In a normal year, how many bags did yield without fertilizer? ( fa rm e r ’s p r e v io u s im p ro v e d v a r ie ty ) (422) In a normal year, how many bags would______________ ( fa rm e r ’s c u r r e n t o r p r e v io u s lo c a l v a r ie ty ) (424) 31. With fertilizer : In a normal year, how many bags does yield with fertilizer? ( fa rm e r ’s c u r r e n t im p ro v e d v a r ie ty ) (426) In a normal year, how many bags did yield with fertilizer? ( fa rm e r ’s p r e v io u s im p ro v e d v a r ie ty ) (428) hi a normal year, how many bags would yield with fertilizer? ( fa rm e r ’s c u rr e n t o r p r e v io u s lo c a l v a r ie ty ) (430) yield without fertilizer? (421) (423) (425) (427) (429) (431) F.2 Nutritional awareness 32. Are any maize varieties particularly good for young children, to help them grow? IF YES: Name of variety that is particularly good for young children (433) Name of variety that is particularly good for young children (434) (432) (1 = y e s , 2 = n o ) 1 - local, 2 = agric, 3 = La Posta, 4 = Aburotia, 5 = Safita, 6 - Kawanzie, 7 = Golden Crystal. 8 = Okomasa, 9 = Dobidi, 10 = Abeleehi, 11 =Dorke, 12 = Obatanpa, 13 = other, 88 = N/A How did you learn that these varieties are particularly good for young children? Source of nutrition information: 1 — own experience, 2 — other farmer, 3 = research or extension worker, 4 = h e a lth w orker, 5 = rad io , 6 = new spap e r , 7 = o th e r (435) Do you use these varieties to prepare weaning foods? 166 _________ (436) (1 = yes, 2 = no) University of Ghana http://ugspace.ug.edu.gh 33. In the past ten years, have you noticed changes in any of the following: Maize yields in your fields (437) Area you plant to maize (438) Quantity of maize you produce (439) Quantity of maize consumed in your household (440) Quantity of maize you have sold (441) • Income from maize sales (442) 1 = increased, 2 = same, 3 = decreased, 4 = don’t know, 5 = other, 88 = N/A 34. If your income from maize sales has increased, what have you done with the extra money? G. Information about respondent 35. Gender (1 = male, 2 = female) 36. Age (443) (444) 37. Years of schooling (445) 38. Marital status (1 = single, 2 = married, 3 = widowed, 4 = divorced) 39. Residence status (1 = native, 2 = settler) (446) (447) H. Information about village 40. Please indicate if your village has the following infrastructure (1 =yes, 2 = no) : Electricity (448) Pipebome water Tarred road (449) Market Good feeder road (450) Health post Easy transportation (451) Elementary school (452) (453) (454) (455) 167 University of Ghana http://ugspace.ug.edu.gh Appendix 2 Theoretical Model for Measuring Net Social Gains 168 University of Ghana http://ugspace.ug.edu.gh APPENDIX 2 MODEL FOR MEASURING NET SOCIAL GAINS. Key parameters that must be estimated in order to run the model for estimating the net social gains include the J-parameter (which measures the production increases or the incremental output from the technology). The K-parameter (which measures the net reduction in production cost from the new technology), and the I-parameter (which measures the increase in per-unit input cost). Other parameters that must be estimated include the change in quantity caused by research (dQ'm equation 4), social gains and the net gains from research1. The J-parameter, which measures the total increase in production attributable to adoption of the new technology (holding prices and costs constant), is estimated from the formula: j = (dY xt) /Ym (A2.1) Where: j = proportional production increase (incremental output) dY = yield increase due to the improved technology t = adoption rate (proportion of total area under improved technology) Ym = farmers’ yield (instead of national average yield because the yield gain is higher than the national average yield). Masters William A., Bakari Coulibaly, Diakalia Sanogo, Mamadou Sidibe, Anne Williams, John a Sanders and Jess Lowenberg De-Boer. “ The Economic Impact of Agricultural Research- A Practical Guide. 1996. pp 6-29. 169 University of Ghana http://ugspace.ug.edu.gh The I-parameter measures the increase in per-unit input costs associated with the reduction in production cost attributable to adoption of the improved technology (k). In practice, input levels often increase with the adoption of new technologies. This per unit cost can be referred to as "input expenditure" (I-parameter) instead of adoption cost.2 A distinction must be made between costs associated with a shift in the production function and those associated with movement along a production function. This is shown in Figure A2.1. Let F„ be the production function of an old technology (say, cutlass) and let F, be the production function for a new technology (say, chainsaw). At an input level of L I5 production can be increased from A to B with a new technology. This is a complete shift between production functions but at the same input level. This shift is equivalent to the K-parameter or the net shift in production cost. It is per unit input cost associated with this shift that are to be estimated in impact assessment. Position C is a point on the same production function, F t as position B. This point can only be produced at a new input level of L2. This is movement along the same production function, FI and not a change in the production function. Costs associated with this movement from B to C are not the same as those associated with the change from A to B. Increase in output ( 0 10 0 ) is related to the shift of inputs (L1L 2 ). Moving from A to C involves not only the shift but also change in input use. In assessing per unit input expenditure, we are interested in per unit input costs associated with the shift in the production function or the net reduction of production costs. 2 Masters William A et al .Op. Cit,p. 16 170 University of Ghana http://ugspace.ug.edu.gh Figure A2.1: Shift and Movement in Production Functions In estimating per unit input cost (which Masters et aL call adoption costs, or the I- parameter,), I-parameter is defined as "the increase in per unit costs required to obtain a given production increase (J)." This definition dwells on production increase on the same production function curve instead of the actual shift of production function curves due to a shift from old to new technologies with fixed inputs. A clear distinction must therefore be made between production increase due to a shift of supply curve and one due to movement on the same supply function. The I- 171 University of Ghana http://ugspace.ug.edu.gh parameter, which will be called the "per unit input expenditure" is that cost associated with the shift from one supply curve to another given the same input. Alternatively, it is per unit of input cost associated with the net reduction of production cost (K-parameter) as illustrated Figure A2.2. The I-parameter may therefore be defined as the increase in per unit input expenditure associated with the reduction in production cost (K-parameter). It is not associated with the production increase (J-parameter). As shown in Figure A2.2, shifting from A to G involves changes in quantity from Q„ to Q' (which is similar to changes in input use in production junction as in Figure A2.1). Instead of keeping quantity constant and observing the costs associated with the actual shift of the curve, we rather observe changes in quantity. As in Figure A2.2, the input expenditure (EF) is the cost associated with the shift from A to E at the same quantity, Q„. Input expenditure should therefore be associated with the shift in supply curve and not with the production increase. The input expenditure (per unit) is estimated from the equation: i = Where I /P = (dCxt)/ (YmxP) i = proportional (per unit) input expenditure I per unit input expenditure P maize price dC = change in costs due to new technology t adoption rate Ym = farmer's yield (Yf) 172 University of Ghana http://ugspace.ug.edu.gh Figure A2.2: Production Increases, Supply Shift and Input Expenditure So The net supply shift (K-parameter) is obtained from the formula: k G / E ) - i Where k net shift in supply j proportional production increase E elasticity of supply i = proportional input expenditure The change in quantity caused by research, dQ is estimated with the formula: dQ = (Q x e x E x k) / (e + E) (A2.3) (A2.4) 173 University of Ghana http://ugspace.ug.edu.gh Where dQ = change in quantity caused by research e = elasticity of demand E elasticity of supply k net shift in supply Q = quantity supplied The social gain is estimated from: SG = (k x P x Q) - Vi (k x P x dQ) (A2.5) Where SG = social gains. Parameter k, P, Q, and dQ are as defined earlier on. The net social gain is obtained from: NG = SG - (R + Ex) (A2.6) Where NG = net gains R = research cost Ex = extension cost The estimated empirical model is derived from equation (A2.1) in which the J- parameter is estimated, equation (A2.2) in which estimates of the proportional increase in costs are made. In equation (A2.3), the proportional supply shift is estimated. Equation {A2A) is used to estimate the change in quantity caused by maize research and extension. 174 University of Ghana http://ugspace.ug.edu.gh The social gains are estimated using equation (A2.5) and the net gains are finally calculated with equation (A2.6). The IRR is then estimated fiom the streams of costs and benefits (net gains). 175 University of Ghana http://ugspace.ug.edu.gh APPENDIX 3 EMPIRICAL PARAMETER ESTIMATES The methods for estimating key parameters used in the internal rate of returns calculations are summarized below. • Adoption rate: The logistic growth function adopted from Griliches1 is used to estimate the adoption rates for 1984 to 1996. The adoption rate for 1997 is obtained from the CRI/CIMMYT survey conducted in 1998. In general, the logistic function is used to portray the diffusion curve. The curve shows the cumulative proportion of adoption of a technology. The diffusion curve follows an S-shaped curve in which there is a small initial growth in the use of the new technology, followed by a more rapid increase and then a decline as the cumulative proportion of adoption reaches its maximum. The curve can be described mathematically Yt = K / (1 + e -a-bt) (A3.1) Where, Yt - cumulative percent area adopted by time t K = upper bound of adoption (ceiling adoption level) b = rate at which adoption occurs, a = constant term t = year (1, 2,... N) 1 Griliches, Z. “Hybrid Com: An Exploration of the Economics of Technological Change.” Econometrica 1957 Vol. 25: 201-522. 176 University of Ghana http://ugspace.ug.edu.gh - In ( ( K /Y t)-1 ) = a + bt (A3.2) • Yield gains attributable to adoption of improved technology: Productivity gains associated with the adoption of improved maize technologies were estimated based on experimental data generated through GGDP on-farm trials. Since experimental data are available for two years only (1984 and 1985), the average percentage yield increases reported in these two years were used for subsequent years and were assumed to remain constant between 1986 and 1997. Prior to 1991, the mean yields registered in GGDP trials for farmer-managed treatments were higher than national mean yields. This indicates either that official statistical data are flawed or that the GGDP trials were carried out in more favorable locations. The productivity gains used to calculate the research benefits were estimated by applying the proportional yield gain observed in the GGDP trials to the formers’ yields instead of the national mean yields figure for reasons given above. For consistency, this method was used for the entire period, 1984-97. • Proportional production increase (j-parameter): This was obtained by multiplying production increase, dY/Ym by the adoption rate, t. • Fertilizer prices: The average fertilizer recommendation for maize calls for the application of compound fertilizer and sulphate of ammonium at a rate of 2.5 bags each per hectare. This function may be simplified as: 177 University of Ghana http://ugspace.ug.edu.gh This is equivalent to applying 30 kg/ha of nitrogen, 30 kg/ha phosphorus, and 45 kg/ha of potassium. Fertilizer price data were obtained from MOFA records. The financial analysis was conducted using the subsidized fertilizer prices actually paid by farmers. These subsidies were removed for the economic analysis because subsidies are simply transfer payments. • Fertilizer application cost: Based on GGDP trial data2, it was determined that five person days per hectare are needed to apply fertilizer. In 1997, agricultural wage rates averaged 2,925 cedis per day, so the cost of the labor required for applying fertilizer amounted to approximately 11% of the cost of fertilizer itself Based on these data, fertilizer application costs for all years were assumed to be 12% of total fertilizer cost. • Row planting cost: Based on GGDP trial data, it was determined that 10 person days per hectare are required for row planting (GGDP 1984). In 1997, agricultural wage rates averaged 2,925 cedis per day, so the cost of the labor required for row planting one hectare of maize amounted to approximately 22% of fertilizer cost. Based on these data, row planting cost for all years were assumed to be 30% of total fertilizer cost. 2 GGDP Annual Report. 1984. P.94. 178 University of Ghana http://ugspace.ug.edu.gh • Maize prices: Import parity prices for maize were calculated based on the standard international reference price for maize, No.2 yellow, F.O.B. US Gulf ports.3 This price was adjusted by adding international transport and handling costs to arrive at the important parity price in Accra. Domestic transport and handling costs were then subtracted to arrive at the import parity price valued at the farm gate. Domestic maize prices were obtained from the Policy Planning Monitoring Evaluation Division of MOFA. In the absence of farm-gate prices, it was necessary to begin with wholesale prices in regional markets; these wholesale prices were then adjusted downwards by 30% (representing domestic transport and handling charges) to arrive at estimated farm gate prices. The 30% figure was based on the results of a maize marketing study conducted in 19914. • Exchange rate (cedis: 1 US $): Official exchange rates were used for the estimation of the financial rate of return for 1984-97. In the economic rate of return estimation, real exchange rates were used. Data were obtained from the IMF International Financial Statistics Yearbook (various issues). 3 Obtained from United States Department of Agricultue (USDA) and FAO Food Outlook on the Webside, May/June, 1998. 4 Dankyi A.A. and A.O.Apau. “Maize Marketing Channel Study.” GGDP Annual Report 1990 pp. 112-126. 179 University of Ghana http://ugspace.ug.edu.gh • Price index: The rural price index from the Ghana Statistical Service is used as a deflator. In estimating the real prices, 1997 is taken as the reference year. • Supply and demand elasticity: The value for elasticity of Supply for maize was obtained from Stryker5 and the elasticity of demand from Alderman and Higgins6. The elasticity of demand for maize was assumed to be 0.981, and the elasticity of supply was assumed to be 0.79. • Ghana Grains Development Project (GGDP) research costs: Cost data for the GGDP were compiled from GGDP Annual Report (various issues). GGDP local costs include salaries and benefits, travel costs, direct research costs, vehicle costs, general administration charges, capital development costs, financial charges, and counterpart funds. In GGDP project documentation, the research costs from 1979-81 are combined. To obtain yearly cost figures for 1979, 1980, and 1981, the three-year total was broken down as follows: 30% for 1979, 30% for 1980, and 40% for 1981. In 1997, GGDP was funded from the National Agricultural Project (NARP). Therefore, GGDP cost data for 1997 were obtained from NARP documents. 5 Stryker Dirck, J. Emmanuel Dumeau, Jennifer Wohl, Peter Haymond, Andrew Cook and Katherine Coon. “Trade, Exchange Rate and Agricultural Pricing Policies in Ghana” The World Bank, Washington D.C. 1990. .pp 164-167. 6 Alderman Harold and Paul Higgins. “Food and Nutritional Adequacy in Ghana” Cornell Food and Nutrition Policy Program. Working paper 27. 1992. P 48. 180 University of Ghana http://ugspace.ug.edu.gh • Proportion of GGDP cost assigned to maize: Since the GGDP works on three main crops (maize, cowpea, soybean), it was necessary to decide the proportion of the total GGDP research budget that was allocated to maize research. The GGDP costs were assigned among these three crops in the following proportions: 1979-83: Maize 90%, cowpea 10% 1984 - 96: Maize 55%, cowpea 30%, soybean 15% • GGDP foreign costs: The Canadian International Development Agency (CIDA) paid for certain costs associated with the GGDP, including vehicles, equipment and supplies, professional services, management and administrative costs, consultants and visiting scientists, technical training, and miscellaneous local costs. The CIDA contribution was estimated using assumptions similar to those used in estimating GGDP costs. Thus, it was estimated that maize research consumed 90% of all expenditures from 1979 to 1983, and 55% from 1984 to 1996. Long-term postgraduate training expenses supported by CIDA were not included because training is assumed to be a long-term investment whose returns are unlikely to be fully captured in the period under study. • NARP-CRI cost: The National Agricultural Research Project (NARP) finances 7 projects of Crops Research Institute. The data from NARP is a bulk sum. Maize cost is therefore, assumed 181 University of Ghana http://ugspace.ug.edu.gh to be one-seventh (15%) of the total NARP-CRI costs. The data is compiled from the NARP secretariat. • SARI cost: Savannah Agricultural Research Institute (SARI) is the second research institute working on maize. It used to be part of Crops Research Institute until 1994 when it became autonomous. Its research costs from 1979 to 1993 are inclusive of the GGDP costs. Available data (NARP/CSIR, 1997), show that scientific time for maize research in Ghana is 8%. This has been used to apportion research costs to SARI from 1994-97. The estimates are obtained from the Ministry of Finance and Economic Planning, Ghana. • MOFA extension cost estimates: The Ministry of Food and Agriculture (MOFA) budget estimates is used. Data available shows that usually 60-70% of the estimates is released. In this calculation, 80% of the estimates is assumed released. The data is compiled from records of the PPMED of MOFA, Ghana. • MOFA extension cost on maize: MOFA extension works on several crops in Ghana but the Policy Planning Monitoring and Evaluation Division (PPMED) of MOFA keeps production records of about 6-8 crops. Therefore using the area under production for these crops, extension activities are assumed to be proportional to it. This assumption puts maize extension cost 182 University of Ghana http://ugspace.ug.edu.gh between 20% and 27%. In this analysis, maize extension cost is taken to be 30% of all MOFA costs. • Global 2000 costs The Sasakawa Global 2000 is an NGO providing additional extension for maize. It was established in 1988 to help disseminate maize and cowpea technologies in Ghana. Eighty percent of its budget was spent on maize extension. The data is from SG 2000 records. Data on extension cost was only available for 1995-97. The costs for the other years have been calculated as 15% drop from the preceding years budget. • CIMMYT costs These are estimated costs by CIMMYT management staff. The cost does not include basic research on maize. This "spill-over benefits" generated by CIMMYT global breeding research is assumed free. CIMMYT costs are obtained from CIMMYT, Mexico. 183 University of Ghana http://ugspace.ug.edu.gh DERIVATION OF KEY FORMULAE FOR NET SOCIAL GAINS AND INTERNAL RATE OF RETURN APPENDIX 4 184 University of Ghana http://ugspace.ug.edu.gh DERIVATION OF KEY FORMULAE FOR NET SOCIAL GAINS AND INTERNAL RATE OF RETURN APPENDIX 4 I. PRODUCTION INCREASES, J-PARAMETER The J-parameter is the total increase in production as a result of the adoption of the technology if prices or costs remain constant. Where dY = Yield increase in adopting the technology t = Adoption rate (proportion of area under new technology) A = Total area under maize production To simplify calculation, J can he expressed in proportional terms as j. This is the increase in quantity produced over the total quantity produced in the country. J = dY x t x A (A4.1) j = J/Q (A4.2) where Q = total quantity of maize produced or, j = (dY x t x A)/Q (A4.3) 185 University of Ghana http://ugspace.ug.edu.gh But Q can be expressed as Ym x A (A4.4) where Ym = overall national average production (yield) level Therefore, j = (dY x txA ) / (YmxA) (A4.5) = (dY x t) / Ym (A4.6) H. INPUT EXPENDITURE, [-PARAMETER The I-parameter is the increase in production costs (per unit costs) required to obtain the given Production increase (J) It can be calculated as: Where dC = adoption costs (change in costs by switching to new technology) t = adoption rate Ym = National average production level To avoid using different units in the calculation, I, can be expressed in a proportional terms, i I = (dCxt)/Ym (A4.7) 186 University of Ghana http://ugspace.ug.edu.gh , as a : Proportional cost increase, i = I / P (A4.8) where P = product price But, I = (dC x t) / Ym (from equation A4.7) Therefore, i = ((dC x t)/ Ym) / P (A4.9) = (dC x t) / Ym x P (A4.10) III. THE MET SHIFT IN SUPPLY CURVE, K-PARAMETER K-parameter is the vertical shift in the supply curve given the production increase, J, and the Per unit input cost, I It is the net reduction in production costs because of the new technology. K-parameter can be calculated from the slope (b) of the supply curve and combining the effects of Production increases (J) and the adoption costs (I). Thus : Because the slopes of the supply curve are associated with specific measurement units, the supply Elasticity (E) is used instead. This is devoid of units of measurement. K =( Jxb ) - I (A4.ll) 187 University of Ghana http://ugspace.ug.edu.gh Generally. E = %dQ / %dP (A4.12) Where dQ = change in quantity dP = change in price And, %dQ = dQ/Q (A4.13) %dP = dP / P (A4.14) Therefore. E = (dQ/Q)/(dP/P) (A4.15) = (dQ / Q) x (P / dP) (A4.16) = (dQ / dP) x (P / Q) (A4.17) But dQ / dP is the inverse of the slope (b) ie. 1/b Substituting dQ / dP with 1/b in equation (A4.17), E = (1/b) x (P/Q) (A4.18) Therefore, the slope, b = E x (Q / P) (A4.19) But from equation (A4.11), K = (J x b) - 1 Hence substituting for b, K = J / (EQ/P) - 1 (A4.20) = [(J x P) / (E x Q)] - 1 (A4.21) 188 University of Ghana http://ugspace.ug.edu.gh k = K /P (A4.22) Substituting for K using equation (A4.21), k = [((J x P) / (E x P)) - I ))] / P (A4.23) = [ ( J x P / E x Q x P ) ] - ( I / P ) (A4.24) Canceling out the Ps k = ( j / E ) - I (A4.25) IV. THE CHANGE IN QUANTITY CAUSED BY RESEARCH, dQ Generally, it is given by: dQ = ( Q x e x E x k ) / ( e + E ) (A4.26) Where, dQ = change in equilibrium quantity e = elasticity o f demand and E = elasticity o f supply Q = quantity supplied k = proportional supply shift The change in quantity actually caused by research, dQ, depends on the shift o f the supply Curve and the responsiveness of supply and demand. Without research, the equilibrium is: Qs =Qj Expressed linearly, Expressing K in proportional terms (k) to the price of the product, a +b„P = a rf+ b rfP (A4.28) 189 University of Ghana http://ugspace.ug.edu.gh Or a, - aa = b dP - b J.P Or a , - O j = p ( b4 ~ bs ) Or P - ° s ~ad!bd -b s (A4.29) (A4.30) (A4.31) With research, the equilibrium is the new supply curve that is shifted in the direction of the price increase(as in Figure A2.2): (A4.32) as +bsP' + bsK = ad+bdP' (A4.33) (A4.34) (A4.35) (A4.36) (A4.37) (A4.38) P' = (as - a d+bsK)/(bd -b s) The resulting change in price, dP = P - P' dP = P -P '= [(/-i u + Where: B ( = Benefit in each year C, = Cost in each year C0 = Initial investment t = 1 ,2,... n n = number of years i = discount (interest) rate. Internal Rate of Return is thus a measure of the maximum interest rate that a project could pay for resources used if the project is to recover its investment and operating cost 1 Gittinger Price, J. “ Economic Analysis o f Agricultural Projects. ” The World Bank Johns Hopkins University Press. 1982. pp. 332-343. 193 University of Ghana http://ugspace.ug.edu.gh and still break even. The selection criterion for IRR is to accept all independent projects having an IRR equal to or greater than the opportunity cost of capital. 194 University of Ghana http://ugspace.ug.edu.gh APPENDIX 5 THE LOGISTIC REGRESSION AND THE TOBIT MODEL 195 University of Ghana http://ugspace.ug.edu.gh APPENDIX 5 THE LOGISTIC REGRESSION The logistic regression model is suited for models in which the dependent variable is dichotomous. The probability of an event occurring can be estimated directly with the logistic regression. For a single independent variable, the logistic regression model can be written as : Probability of adoption = h — (A5.1) Probability of adoption = ----------------------------------------------------------------- (A5.2) where ba and bY are coefficients from the data, X is the independent variable, and e is the natural logarithm of value of approximately 2.718. For more than one independent variable, the model can be written as: Probability of adoption = ----— (A5.3) Probability of adoption = — - - (A5.4) where Z is a linear combination of explanatory variables, and Z = bB + bxX x + b2X 2 +... + bkX k (A5.5) The probability of the event not occurring is estimated as: Prob(no event) = 1 - Prob(event) (A5.6) 196 University of Ghana http://ugspace.ug.edu.gh The probability estimated is between 0 and 1. In logistic regression, the parameters of the regression are estimated using the maximum likelihood method. Thus, the coefficients that make the observed results most “likely” are selected. THE TOBIT MODEL A TOBIT model is used to determine the factors of adoption of improved maize in Ghana. Mathematically, the TOBIT model1 can be expressed as: Yi = P x+ /32X 2i + l^2i , if Right Hand Side > 0 (A5.7) = 0, otherwise where Yt — Limited dependent variable X i= Explanatory variable fa = Disturbance term /?! = Intercept P2j= Coefficient to be estimated The TOBIT model measures not only the probability of adoption but also the intensity of adoption. Rahm and Huffman2 developed a general model where formers are assumed to make adoption decisions based on utility maximization Define tx = New technology t2= Farmer technology The underlying utility function ranking the preference of the i'h farmer is given by: U { H M . 1 Gujarati Damodar, N., Basic Econometrics. 3rf ed. McGraw-hill, Inc. 1995.pp.570-575. 2 Rahm, Michael R. and Wallace E. Huffman, ‘The Adoption of Reduced Tillage: The Role of Human Capital and Other Variables.” Australian Journal of Agricultural Economics, 1984. 66:pp. 405-413. 197 University of Ghana http://ugspace.ug.edu.gh This means that utility is a function of Hh a vector of farm specific attributes such as farm size, and Nn which are technology-specific characteristics such as taste. Un = caF(Hn, Nn)+en (A5.8 ) where t = \,2\i = \...,N The utilities are random and the/* farmer adopts the new technology (t = 1) if Uu > U2i or if the non-observable random variable y'=uu-u2i >o. The probability that 7 * > 0, (i.e. that the ikrmer adopts the new technology) can be written as a function of the independent variables: Pi = PrCyi > 0) = Pr(t/u >U2i) + eu > aiFi{Hn, Nn)+ eu\ = Pr[cu - eji > N n\a 2 - ai)] = Pr(fa > F,iHu,Nu)p) = Fi(X,p) where Pt = Probability function Ht =eu - e2j is a random disturbance term P = a2 - a1 is a vector of parameters to be determined. X = (n x k) is vector of explanatory variables. F(XIP) = cumulative distribution function for the disturbance term, ju,, evaluated at x,p. (A5.9) (A5.10) (A5.ll) (A5.12) 198 University of Ghana http://ugspace.ug.edu.gh The dependent variable (Yt) is the proportion of maize area under improved varieties. The dependent variable is censored because information on the regressand is available only for some observations. TOBIT model can therefore be used. 199 University of Ghana http://ugspace.ug.edu.gh APPENDIX 6 LOGISTIC GROWTH CURVE 200 University of Ghana http://ugspace.ug.edu.gh APPENDIX 6 LOGISTIC GROWTH CURVE EXPERIMENT GHANA MAIZE STUDY - LOGISTIC GROWTH CURVE EXPERIMENT LOGISTIC GROWTH CURVE FORMULA (ADOPTED FROM GRILICHES, 1957): -LN[K/Yt>l] = A + BT WHERE: Yt = CUMULATIVE PERCENTAGE OF ADOPTERS BY TIME t K = UPPER BOUND ON PERCENTAGE ADOPTION b = RATE AT WHICH ADOPTION OCCURS a = CONSTANT TERM EXAMPLE 1: K = 95 30 YEARS TO TOTAL ADOPTION K = 95 a = -3.923 b = 0.3028 t Y(t) t - LN[K/Yt)-1] 3 5 3 -2.89 9 20 9 -1.322 14 54 14 0.2754 23 90 23 2.8904 25 93 25 3.8395 201 University of Ghana http://ugspace.ug.edu.gh APPENDIX 6 continued YEAR % ADOPTION (Predicted) 1 2 2 3 3 4 4 6 5 8 6 10 7 13 8 17 9 22 10 28 11 34 12 41 13 48 14 55 15 62 16 68 17 73 18 78 19 82 20 85 21 87 22 89 23 91 24 92 25 93 26 93 27 94 28 94 29 94 30 94 202 University of Ghana http://ugspace.ug.edu.gh APPENDIX 6 Continued SUMMARY OUTPUT Regression Statistics Multiple R 0.99854355 R Square 0.99708921 Adjusted R 0.99611895 Square Standard 0.1754122 Error Observations 5 ANOVA D f SS MS F Significance F Regression 1 31.62018283 31.62018 1027.649 6.6709E-05 Residual 3 0.092308321 0.030769 Total 4 31.71249115 Coefficients Standard Error tStat P-value Lower 95% Upper 95% Intercept -3.92325883 0.160314356 -24.4723 0.00015 -4.433451142 -3.413066527 X Variable 1 0.30282977 0.009446614 32.05696 6.67E-05 0.272766395 0.332893135 Lower Upper 95.0% Upper 95.0% 95.0% -4.43345 -3.413066527 -3.636447359 0.272766 0.332893135 0.325277624 Source: Estimated by the author 203 University of Ghana http://ugspace.ug.edu.gh Lo gi sti c fu nc tio n Figure A6.1: Logistic regression Source: Estimated by the author. 204 University of Ghana http://ugspace.ug.edu.gh Figure M 2 : Logistic growth curve Years Source: Estimated by the author 205 University of Ghana http://ugspace.ug.edu.gh tame pa: F inancial R ate o f R eturn (B ase lin e s c en a r io ) Adoption rate Yield of Yield of Change in Proportional Total quantity produced Total area planted National average (at K = 95%, improved farmer yield increase Year production Lag = 30 years) technology technology Unit (mt) (ha) (mt/ha) (mt/ha) (mt/ha) (mt/ha) Parameter Q A Ym t Yn Yf = Ym dY=Yn-Yf dY/Yf 1979 308600 313900 0.98 0 1980 354000 319900 1.11 0 1981 334200 315500 1.06 0 1982 264300 276300 0.96 0 1983 140800 279800 0.50 o' 1984 574000 723600 0.79 0.02 2.88 1.73 1.15 0.6647 1985 395000 405000 0.98 0.03 2.75 1.7 1.05 0.6176 1986 559100 472100 1.18 0.04 2.93 1.74 1.19 0.6839 1987 597700 548336 1.09 0.06 2.93 1.74 1.19 0.6839 1988 600000 500000 1.20 0.08 2.93 1.74 1.19 0.6839 1989 714614 595800 1.20 0.1 2.93 1.74 1.19 0.6839 1990 552549 464784 1.19 0.13 2.93 1.74 1.19 0.6839 1991 931478 610445 1.53 0.17 2.93 1.74 1.19 0.6839 1992 730617 606783 1.20 0.22 2.93 1.74 1.19 0.6839 1993 960920 636670 1.51 0.28 2.93 1.74 1.19 0.6839 1994 939908 629401 1.49 0.34 2.93 1.74 1.19 0.6839 1995 1034170 668600 1.55 0.41 2.93 1.74 1.19 0.6839 1996 1007610 664950 1.52 0.48 2.93 1.74 1.19 0.6839 1997 1000000 650000 1.54 0.54 2.93 1.74 1.19 0.6839 206 University of Ghana http://ugspace.ug.edu.gh I anie A / continued Change in Change in Improved Farmers Proportional Fertilizer prices fertilizer cost row planting seed seed production increase (comp. + S.Ammonia) including application cost Year (cedis/ha) (cedis/ha) (cedis/ha) (cedis/ha) (cedis/ha) i = t x (dY/Yf) 1979 1980 1981 1982 1983 1984 0.0133 1,838 2,059 551 400 300 1985 0.0185 1,838 2,059 551 1,000 750 1986 0.0274 3,175 3,556 953 1,000 750 1987 0.0410 5,500 6,160 1,650 2,000 1,500 1988 0.0547 9,750 10,920 2,925 2,000 1,500 1989 0.0684 14,250 15,960 4,275 3,000 2,250 1990 0.0889 18,250 20,440 5,475 3,000 2,250 1991 0.1163 23,750 26,600 7,125 5,000 3,750 1992 0.1505 36,250 40,600 10,875 5,000 3,750 1993 0.1915 43,750 49,000 13,125 9,000 6,750 1994 0.2325 58,750 65,800 17,625 9,000 6,750 1995 0.2804 92,500 103,600 27,750 12,000 9,000 1996 0.3283 107,500 120,400 32,250 20,000 15,000 1997 0.3693 132,500 148,400 39,750 30,000 22,250 207 University of Ghana http://ugspace.ug.edu.gh i acne /v/ conrinuea Change in Change in total Maize price: seed cost cost of technology Average wholesale Farmgate price Rural price index Real price (C/t) price (70% of wholesale) (1977= 100) (1997= 1) (cedis/ha) (cedis/ha) (cedis/mt) (cedis/mt) ^ Year dC 1979 279 1980 448 1981 938 1982 1,150 1983 0 0 2,640 0 1984 100 2,710 23,380 16,366 3,652 354,204 1985 250 2,860 20,380 14,266 3,974 283,737 1986 250 4,759 33,110 23,177 4,824 379,744 1987 500 8,310 53,870 37,709 6,591 452,205 1988 500 14,345 68,590 48,013 8,581 442,244 1989 750 20,985 53,000 37,100 10,637 275,674 1990 750 26,665 85,030 59,521 14,310 328,755 1991 1,250 34,975 116,050 81,235 16,784 382,551 1992 1,250 52,725 100,480 70,336 18,613 298,678 1993 2,250 64,375 110,720 77,504 23,092 265,280 1994 2,250 85,675 138,630 97,041 28,598 268,201 1995 3,000 134,350 200,000 140,000 44,880 246,557 1996 5,000 157,650 250,000 175,000 62,230 222,269 1997 7,750 195,900 300,000 210,000 79,039 210,000 208 University of Ghana http://ugspace.ug.edu.gh Table A 7 continued Year Real change In cost, Proportional cost increase Change in quantity caused by Supply elasticity Net shift in Demand elasticity research, Unit (cedis/ha) (E = 0.79) supply (e = - 0.981) dQ Parameter dC i = l/P = (dC . t)/(Yf. P) k = (j/E )~ i dQ = (Q x e x E x k) / (e + E) L 1979 0.79 0,981 1980 0.79 0.981 1981 0.79 0.981 1982 0.79 0.981 1983 0.0000 0.79 0.981 0 1984 58,651 0.0019 0.79 0.0149 0.981 3,746 1985 56,882 0.0035 0.79 0.0199 0.981 3,443 1986 77,966 0.0047 0.79 0.0299 0.981 7,317 1987 99,653 0.0076 0.79 0.0443 0.981 11,598 1988 132,131 0.0137 0.79 0.0555 0.981 14,577 1989 155,931 0.0325 0.79 0.0541 0.981 16,906 1990 147,280 0.0335 0.79 0.0791 0.981 19,119 1991 164,704 0.0421 0.79 0.1051 0.981 42,843 1992 223,894 0.0948 0.79 0.0957 0.981 30,589 1993 220,342 0.1337 0.79 0.1087 0.981 45,724 1994 236,788 0.1725 0.79 0.1218 0.981 50,107 1995 236,606 0.2261 0.79 0.1288 0.981 58,296 1996 200,233 0.2485 0.79 0.1670 0.981 73,647 1997 195,900 0.2895 0.79 0.1780 0.981 77,881 209 University of Ghana http://ugspace.ug.edu.gh Table A 7 continued Social gains from research GGDP (all local costs), Proportional GGDP GGDP (foreign costs), Exchange rate cedis maize cost (cedis) Canadian $ (CN$:1US$) (90% for 1979-83, Year SG = (k , P . Q) -1/2 (k . P.dQ) 55% for 1984-96) 1979 0 389,380 350,442 228,346 1,1714 1980 0 389,380 350,442 228,346 1.1692 1981 0 519,173 467,256 304,462 1.1989 1982 0 1,402,036 1,261,832 319,903 1.2337 1983 0 1,030,870 927,783 197,800 1.2324 1984 3,022,432,946 9,918,988 5,455,443 998,300 1.2951 1985 2,222,511,149 36,387,694 20,013,232 1,430,482 1.3655 1986 6,308,461,916 65,949,829 36,272,406 319,101 1.3895 1987 11,868,961,201 80,393,569 44,216,463 175,183 1.326 1988 14,553,035,743 98,738,058 54,305,932 2,023,483 1.2307 1989 10,524,457,658 187,897,196 103,343,458 365,767 1.184 1990 14,115,007,059 201,386,259 110,762,442 501,532 1.1668 1991 36,591,818,286 278,083,542 152,945,948 2,484,702 1.1457 1992 20,441,327,147 427,815,898 235,298,744 1,160,624 1.2087 1993 27,059,117,346 450,923,659 248,008,012 4,821,245 1.2901 1994 29,891,504,647 699,105,294 384,507,912 679,765 1.3656 1995 31,920,145,086 192,709,160 105,990,038 212,020 1.3724 1996 36,040,383,882 815,007,795 448,254,287 55,996 1.3635 1997 35,919,159,696 0 210 University of Ghana http://ugspace.ug.edu.gh Table A 7 continued Exchange rate, Cedis:CN$ All costs Proportional foreign Total GGDP NARP-CRI Cost NARP-CRI Maize (cedis) cedis: 1US$ (CN$ to cedis) cost (cedis) maize cost (cedis) (cedis) (15% of NARP-CRI) Year (90% for 1979-83, 55% for 1984-96) 1979 2.75 2.35 536,069 482,462 832,904 0 0 1980 2.75 2.35 537,078 483,370 833,812 0 0 1981 2.75 2,29 698,366 628,529 1,095,785 0 0 1982 2.75 2.23 713,085 641,777 1,903,609 0 0 1983 8.83 7.16 1,417,214 1,275,492 2,203,275 0 0 1984 35.99 27.79 27,742,118 15,258,165 20,713,608 0 0 1985 54.37 39.82 56,957,383 31,326,561 51,339,792 0 0 1986 89.2 64.20 20,484,929 11,266,711 47,539,117 0 0 1987 153.73 115.94 20,309,866 11,170,426 55,386,889 0 0 1988 202.35 164.42 332,698,290 182,984,059 237,289,991 0 0 1989 270 228.04 83,409,704 45,875,337 149,218,795 0 0 1990 326.33 279.68 140,268,202 77,147,511 187,909,953 0 0 1991 367.83 321.05 797,720,116 438,746,064 591,692,012 0 0 1992 437.09 361.62 419,704,761 230,837,618 466,136,362 0 0 1993 649.06 503.11 2,425,608,309 1,334,084,570 1,582,092,583 0 0 1994 956.71 700.58 476,228,744 261,925,809 646,433,721 0 0 1995 1200.43 874.69 185,452,615 101,998,938 207,988,976 1,361,979,250 204,296,888 1996 1637.23 1,200.76 67,237,500 36,980,625 485,234,912 1,535,228,229 230,284,234 1997 0 1,927,478,424 289,121,764 211 University of Ghana http://ugspace.ug.edu.gh Table A 7 continued Year SARI cost SARI maize (8%) GGDP ,SARI and MOFA Extension Actual releases MOFA Extesion cost (cedis) NARP-CRI (cedis) estimates (cedis) (80% of estimates) for maize at 30% (cedis) 1979 0 0 832,904 0 0 0 1980 0 0 833,812 0 0 0 1981 0 0 1,095,785 0 0 0 1982 0 0 1,903,609 0 0 0 1983 0 0 2,203,275 0 0 0 1984 0 0 20,713,608 156,631,040 125,304,832 37,591,450 1985 0 0 51,339,792 195,788,800 156,631,040 46,989,312 1986 0 0 47,539,117 244,736,000 195,788,800 58,736,640 1987 0 0 55,386,889 305,920,000 244,736,000 73,420,800 1988 0 0 237,289,991 382,400,000 305,920,000 91,776,000 1989 0 0 149,218,795 853,829,000 683,063,200 204,918,960 1990 0 0 187,909,953 857,208,000 685,766,400 205,729,920 1991 0 0 591,692,012 1,997,101,000 1,597,680,800 479,304,240 1992 0 0 466,136,362 1,516,825,000 1,213,460,000 364,038,000 1993 0 0 1,582,092,583 2,435,152,000 1,948,121,600 584,436,480 1994 535,742,408 42,859,393 689,293,114 2,272,267,000 1,817,813,600 545,344,080 1995 597,144,000 47,771,520 460,057,384 2,455,425,000 1,964,340,000 589,302,000 1996 1,171,434,025 93,714,722 809,233,869 533,000,000 426,400,000 127,920,000 1997 1,353,509,000 108,280,720 397,402,484 829,000,000 663,200,000 198,960,000 212 University of Ghana http://ugspace.ug.edu.gh i aDie A / continued Extension + Research Year SG 2000 80% of SG2000 Total extension (cedis) Extension + Research (Real cost in cedis) CIMMYT Costs (CN$) 1979 0 0 0 832,904 235,956,716 50,000 1980 0 0 0 833,812 147,106,421 50,000 1981 0 0 0 1,095,785 92,334,466 50,000 1982 0 0 0 1,903,609 130,834,226 50,000 1983 0 0 0 2,203,275 65,963,890 50,000 1984 0 0 37,591,450 58,305,058 1,261,876,635 50,000 1985 0 0 46,989,312 98,329,104 1,955,670,376 50,000 1986 0 0 58,736,640 106,275,757 1,741,278,930 50,000 1987 0 0 73,420,800 128,807,689 1,544,656,495 25,000 1988 20,789,424 16,631,539 108,407,539 345,697,531 3,184,196,145 25,000 1989 24,458,146 19,566,517 224,485,477 373,704,272 2,776,836,698 25,000 1990 28,774,290 23,019,432 228,749,352 416,659,305 2,301,351,139 25,000 1991 33,852,105 27,081,684 506,385,924 1,098,077,936 5,171,054,694 25,000 1992 39,826,006 31,860,805 395,898,805 862,035,167 3,660,581,184 25,000 1993 46,854,125 37,483,300 621,919,780 2,204,012,363 7,543,865,110 25,000 1994 55,122,500 44,098,000 589,442,080 1,278,735,194 3,534,161,514 25,000 1995 64,850,000 51,880,000 641,182,000 1,101,239,384 1,939,413,094 25,000 1996 84,305,000 67,444,000 195,364,000 1,004,597,869 1,275,950,682 25,000 1997 96,000,000 76,800,000 275,760,000 673,162,484 673,162,484 0 213 University of Ghana http://ugspace.ug.edu.gh i aoie « / continued Year CIMMYT Costs CIMMYT Real Costs Research, extension Net social gains (in cedis) (1997= 1) and CIMMYT real costs (cedisl) NG = SG - ( R + E ) 1979 117,381 33,253,297 269,210,013 -269,210,013 1980 117,602 20,748,051 167,854,472 -167,854,472 1981 114,688 9,664,031 101,998,497 -101,998,497 1982 111,453 7,660,140 138,494,367 -138,494,367 1983 358,244 10,725,475 76,689,365 -76,689,365 1984 1,389,468 30,071,786 1,291,948,420 1,730,484,526 1985 1,990,846 39,595,990 1,995,266,366 227,244,783 1986 3,209,788 52,590,881 1,793,869,811 4,514,592,105 1987 2,898,379 34,757,236 1,579,413,732 10,289,547,470 1988 4,110,466 37,861,215 3,222,057,360 11,330,978,383 1989 5,701,014 42,361,794 2,819,198,493 7,705,259,165 1990 6,991,987 38,619,122 2,339,970,261 11,775,036,799 1991 8,026,316 37,797,425 5,208,852,119 31,382,966,167 1992 9,040,498 38,389,939 3,698,971,123 16,742,356,024 1993 12,577,707 43,050,814 7,586,915,924 19,472,201,422 1994 17,514,463 48,406,378 3,582,567,892 26,308,936,755 1995 21,867,349 38,510,994 1,977,924,088 29,942,220,999 1996 30,018,885 38,127,313 1,314,077,995 34,726,305,887 1997 0 0 673,162,484 35,245,997,213 IRR = 79% Source: Estimated by the author. 214 University of Ghana http://ugspace.ug.edu.gh Table A8: Economic Rate of Return ( Baseline scenario) Yield of improved Yield of Change in Proportional Year Total quantity produced Total area planted National average Adoption rate technology farmer yield increase production (at K = 95%, technology Lag = 30 years) (mt/ha) (mt/ha) (mt/ha) Unit (mt) (ha) (mt/ha) Parameter Q A Ym t Yn Yf = Ym dY=Yn-Yf dY/Yf 1979 308,600 313,900 0.98 0 1980 354,000 319,900 1.11 0 1981 334,200 315,500 1.06 0 1982 264,300 276,300 0.96 0 1983 140,800 279,800 0.50 0 1984 574,000 723,600 0.79 0.02 2.88 1.73 1.15 0.6647 1985 395,000 405,000 0.98 0.03 2.75 1.7 1.05 0.6176 1986 559,100 472,100 1.18 0.04 2.93 1.74 1.19 0.6839 1987 597,700 548,336 1.09 0.06 2.93 1.74 1.19 0.6839 1988 600,000 500,000 1.20 0.08 2.93 1.74 1.19 0.6839 1989 714,614 595,800 1.20 0.1 2.93 1.74 1.19 0.6839 1990 552,549 464,784 1.19 0.13 2.93 1.74 1.19 0.6839 1991 931,478 610,445 1.53 0.17 2.93 1.74 1.19 0.6839 1992 730,617 606,783 1.20 0.22 2.93 1.74 1.19 0.6839 1993 960,920 636,670 1.51 0.28 2.93 1.74 1.19 0.6839 1994 939,908 629,401 1.49 0.34 2.93 1.74 1.19 0.6839 1995 1,034,170 668,600 1,55 0.41 2.93 1.74 1.19 0.6839 1996 1,007,610 664,950 1.52 0.48 2.93 1.74 1.19 0.6839 1997 1,000,000 650,000 1.54 0.54 2,93 1.74 1.19 0.6839 215 University of Ghana http://ugspace.ug.edu.gh Table A8 continued. Proportional production increase Fertilizer prices (j-parameter) (comp. + S.A, Change in fertilizer costs Change in Improved Farmers Change in cedis/ha) including application row planting seed seed seed cost ! Year (Without subsidy) (cedis/ha) cost (cedis/ha) (cedis/ha) (cedis/ha) i = t x (dY/Yf) (cedis/ha) 1979 0.0000 1980 0.0000 1981 0.0000 1982 0.0000 1983 0.0000 1984 0.0133 2,665 2,985 800 400 300 100 1985 0.0185 2,922 3,273 877 1,000 750 250 1986 0.0274 4,318 4,836 1,295 1,000 750 250 1987 0.0410 7,810 8,747 2,343 2,000 1,500 500 1988 0.0547 12,675 14,196 3,803 2,000 1,500 500 1989 0.0684 16,388 18,355 4,916 3,000 2,250 750 1990 0.0889 18,250 20,440 5,475 3,000 2,250 750 1991 0.1163 23,750 26,600 7,125 5,000 3,750 1,250 1992 0.1505 36,250 40,600 10,875 5,000 3,750 1,250 1993 0,1915 43,750 49,000 13,125 9,000 6,750 2,250 1994 0.2325 58,750 65,800 17,625 9,000 6,750 2,250 1995 0.2804 92,500 103,600 27,750 12,000 9,000 3,000 1996 0.3283 107,500 120,400 32,250 20,000 15,000 5,000 1997 0.3693 132,500 148,400 39,750 30,000 22,250 7,750 216 University of Ghana http://ugspace.ug.edu.gh Table A8 continued. Real Freight and Value of maize Exchange rate Value of maize Product price marketing cost ($/ton) ($/ton) (cedis: 1 US$) (Cedis/ton) No.2 yellow maize Change in total Export price (F.O.B) (30% + 20% = 50% (ie.100% + 50%) Year cost of technology U.S. Gulf ($/ton) of export price) (cedis/ha) dC 1979 1980 1981 1982 1983 0 0 1984 3,884 139.1 69.55 208.65 611.07 127,500 1985 4,399 116.8 58.4 175.2 870.05 152,433 1986 6,382 88.7 44.35 133.05 1164.38 154,921 1987 11,590 75.6 37.8 113.4 1572.39 178,309 1988 18,499 107 53.5 160.5 1550.58 248,868 1989 24,021 111.3 55.65 166.95 1733.12 289,344 1990 26,665 109.7 54.85 164.55 1609.3 264,810 1991 34,975 107 53.5 160.5 1602.46 257,195 1992 52,725 105.1 52.55 157.65 1781.59 280,868 1993 64,375 101.9 50.95 152.85 2180.25 333,251 1994 85,675 107.5 53.75 161.25 2638.24 425,416 1995 134,350 123.7 61.85 185.55 1952.04 362,201 1996 157,650 165.1 82.55 247.65 2044.46 506,311 1997 195,900 117.1 58.55 175.65 2050.2 360,118 2 1 7 University of Ghana http://ugspace.ug.edu.gh Table A8 continued. Year Real change in cost, Proportional cost increase Net shift in supply Unit (cedis/mt) Real price (C/t) Rural price index (Farmgate) dC i = l/P = (dC . t)/(Yf. P) Supply elasticity k = ( j / E) - i (1977 = 100) (1997= 1) ( E = 0.79) 1979 279 0.79 1980 448 0.79 1981 938 0.79 1982 1,150 0.79 1983 2,640 0.79 1984 3,652 1,931,604 84,067 0.0000 0.79 0.0168 1985 3,974 2,122,218 87,497 0.0000 0,79 0.0234 1986 4,824 1,776,813 104,559 0.0001 0.79 0.0345 1987 6,591 1,496,792 138,989 0.0003 0.79 0.0517 1988 8,581 1,604,615 170,388 0.0005 0.79 0.0687 1989 10,637 1,504,996 178,489 0.0009 0.79 0.0857 1990 14,310 1,023,846 147,280 0.0019 0.79 0.1106 1991 16,784 847,825 164,704 0.0040 0.79 0.1431 1992 18,613 834,882 223,894 0.0080 0.79 0.1825 1993 23,092 798,454 220,342 0.0130 0.79 0.2294 1994 28,598 823,034 236,788 0.0203 0.79 0.2740 1995 44,880 446,515 236,606 0.0709 0.79 0.2840 1996 62,230 450,149 200,233 0.0966 0.79 0.3189 1997 79,039 252,082 195,900 0.2412 0.79 0.2263 218 University of Ghana http://ugspace.ug.edu.gh Tab le A 8 continued. Change in quantity Social gains from research Research cost caused by research, dQ GGDP (all local costs), Proportional GGDP dQ = (Q x e x E x k) / (e + E) SG = (k . P . Q) -1/2 (k . P . dQ) cedis maize cost (cedis) (90% for 1979-83, Demand elasticity dQ 55% for 1984-96) Year (e = - 0.981) 1979 0.981 389,380 350,442 1980 0.981 389,380 350,442 1981 0.981 519,173 467,256 1982 0.981 1,402,036 1,261,832 1983 0.981 1,030,870 927,783 1984 0.981 4,221.3 18,564,547,348 9,918,988 5,455,443 1985 0.981 4,047.9 19,530,468,891 36,387,694 20,013,232 1986 0.981 8,452.0 34,058,854,442 65,949,829 36,272,406 1987 0.981 13,515.9 45,707,766,058 80,393,569 44,216,463 1988 0.981 18,044.8 65,172,721,789 98,738,058 54,305,932 1989 0.981 26,785.1 90,393,025,287 187,897,196 103,343,458 1990 0.981 26,741.6 61,052,934,824 201,386,259 110,762,442 1991 0.981 58,345.9 109,501,384,670 278,083,542 152,945,948 1992 0,981 58,339.2 106,859,423,256 427,815,898 235,298,744 1993 0.981 96,472.4 167,189,237,112 450,923,659 248,008,012 1994 0.981 112,697.1 199,252,301,157 699,105,294 384,507,912 1995 0.981 128,543.7 123,010,986,604 192,709,160 105,990,038 1996 0.981 140,624.8 134,563,115,856 815,007,795 448,254,287 1997 0.981 99,030.7 54,222,550,318 219 University of Ghana http://ugspace.ug.edu.gh Table A8 continued. Real Real ■ GGDP (foreign costs), Exchange rate Exchange rate, Exchange rate All costs Proportional foreign Canadian $ (CN$:1US$) cedis: 1US$ (Cedis:CN$) (CN$ to cedis) cost (cedis) Year (90% for 1979-83, 55% for 1984-96) 1979 228,346 1.1714 401.32 342.60 78,231,020 70,407,918 1980 228,346 1.1692 306.92 262,50 59,941,802 53,947,621 1981 304,462 1.1989 154.19 128.61 39,156,723 35,241,051 1982 319,903 1.2337 135.54 109.86 35,146,026 31,631,424 1983 197,800 1.2324 201.27 163.32 32,303,802 29,073,422 1984 998,300 1.2951 611.07 471.83 471,030,176 259,066,597 1985 1,430,482 1.3655 870.05 637.17 911,454,313 501,299,872 1986 319,101 1.3895 1164.38 837.98 267,401,815 147,070,998 1987 175,183 1.326 1572.39 1,185.81 207,734,538 114,253,996 1988 2,023,483 1.2307 1550.58 1,259.92 2,549,420,874 1,402,181,481 1989 365,767 1.184 1733.12 1,463.78 535,403,803 294,472,092 1990 I 501,532 1.1668 1609,3 1,379.24 691,734,185 380,453,802 1991 2,484,702 1.1457 1602.46 1,398.67 3,475,286,346 1,911,407,490 1992 1,160,624 1.2087 1781.59 1,473.97 1,710,727,320 940,900,026 1993 4,821,245 1.2901 2180.25 1,689.99 8,147,833,045 4,481,308,175 1994 679,765 1.3656 2638,24 1,931.93 1,313,256,600 722,291,130 1995 212,020 1.3724 1952.04 1,422.35 301,567,707 165,862,239 1996 55,996 1.3635 2044.46 1,499.42 83,961,556 46,178,856 1997 0 2050.2 0 220 University of Ghana http://ugspace.ug.edu.gh Table A8 continued. Total GGDP NARP-CRI Cost NARP-CRI Maize SARI cost SARI maize (8%) GGDP ,SARI and MOFA Extension maize cost (cedis) (in cedis) (15% of NARP-CRI) (cedis) NARP-CRI (cedis) estimates (cedis) Year (in cedis) (cedis) 1979 70,758,360 0 0 0 0 70,758,360 0 1980 54,298,063 0 0 0 0 54,298,063 0 1981 35,708,307 0 0 0 0 35,708,307 0 1982 32,893,256 0 0 0 0 32,893,256 0 1983 30,001,205 0 0 0 0 30,001,205 0 1984 264,522,040 0 0 0 0 264,522,040 156,631,040 1985 521,313,104 0 0 0 0 521,313,104 195,788,800 1986 183,343,404 0 0 0 0 183,343,404 244,736,000 1987 158,470,459 0 0 0 0 158,470,459 305,920,000 1988 1,456,487,413 0 0 0 0 1,456,487,413 382,400,000 1989 397,815,550 0 0 0 0 397,815,550 853,829,000 1990 491,216,244 0 0 0 0 491,216,244 857,208,000 1991 2,064,353,439 0 0 0 0 2,064,353,439 1,997,101,000 1992 1,176,198,770 0 0 0 0 1,176,198,770 1,516,825,000 1993 4,729,316,187 0 0 0 0 4,729,316,187 2,435,152,000 1994 1,106,799,042 0 0 535,742,408 42,859,393 1,149,658,435 2,272,267,000 1995 271,852,277 1,361,979,250 204,296,888 597,144,000 47,771,520 523,920,684 2,455,425,000 1996 494,433,143 1,535,228,229 230,284,234 1,171,434,025 93,714,722 818,432,100 533,000,000 1997 1,927,478,424 289,121,764 1,353,509,000 108,280,720 397,402,484 829,000,000 221 University of Ghana http://ugspace.ug.edu.gh Table AS continued. Year Actual releases MOFA Extesion cost SG 2000 80% of SG2000 Total extension Extension + Research (80% of estimates) for maize at 30% (cedis) (cedis) (cedis) (in cedis) Cost (in cedis) (cedis) 1979 0 0 0 0 0 70,758,360 1980 0 0 0 0 0 54,298,063 1981 0 0 0 0 0 35,708,307 1982 0 0 0 0 0 32,893,256 1983 0 0 0 0 0 30,001,205 1984 125,304,832 37,591,450 0 0 37,591,450 302,113,490 1985 156,631,040 46,989,312 0 0 46,989,312 568,302,416 1986 195,788,800 58,736,640 0 0 58,736,640 242,080,044 1987 244,736,000 73,420,800 0 0 73,420,800 231,891,259 1988 305,920,000 91,776,000 20,789,424 16,631,539 108,407,539 1,564,894,952 1989 683,063,200 204,918,960 24,458,146 19,566,517 224,485,477 622,301,026 1990 685,766,400 205,729,920 28,774,290 23,019,432 228,749,352 719,965,596 1991 1,597,680,800 479,304,240 33,852,105 27,081,684 506,385,924 2,570,739,363 1992 1,213,460,000 364,038,000 39,826,006 31,860,805 395,898,805 1,572,097,575 1993 1,948,121,600 584,436,480 46,854,125 37,483,300 621,919,780 5,351,235,967 1994 1,817,813,600 545,344,080 55,122,500 44,098,000 589,442,080 1,739,100,515 1995 1,964,340,000 589,302,000 64,850,000 51,880,000 641,182,000 1,165,102,684 1996 426,400,000 127,920,000 84,305,000 67,444,000 195,364,000 1,013,796,100 1997 663,200,000 198,960,000 96,000,000 76,800,000 275,760,000 673,162,484 222 University of Ghana http://ugspace.ug.edu.gh Table A8 continued. Extension + Research CIMMYT Costs) CIMMYT Costs CIMMYT Real Extension, Research NET SOCIAL GAINS Year (Real cost in cedis) (CN$) (in cedis) cost and CIMMYT real costs (1997= 1) (in cedis) NG = SG - (R + E) (cedis) (cedis) 1979 20,045,412,220 50,000 17,129,930 4,852,804,793 24,898,217,013 -24,898,217,013 1980 9,579,608,548 50,000 13,125,214 2,315,633,427 11,895,241,975 -11,895,241,975 1981 3,008,900,708 50,000 6,430,478 541,853,460 3,550,754,168 -3,550,754,168 1982 2,260,739,187 50,000 5,493,232 377,547,429 2,638,286,616 -2,638,286,616 1983 898,206,534 50,000 8,165,774 244,475,234 1,142,681,769 -1,142,681,769 1984 6,538,540,012 50,000 23,591,615 510,585,329 7,049,125,341 11,515,422,007 1985 11,302,983,049 50,000 31,858,294 633,630,517 11,936,613,566 7,593,855,325 1986 3,966,369,118 50,000 41,899,244 686,499,663 4,652,868,781 29,405,985,661 1987 2,780,830,406 25,000 29,645,362 355,505,958 3,136,336,364 42,571,429,694 1988 14,414,139,625 25,000 31,497,928 290,125,246 14,704,264,871 50,468,456,919 1989 4,624,052,912 25,000 36,594,595 271,918,789 4,895,971,702 85,497,053,585 1990 3,976,615,007 25,000 34,481,059 L 190,450,625 4,167,065,632 56,885,869,192 1991 12,106,093,214 25,000 34,966,833 164,665,364 12,270,758,578 97,230,626,092 1992 6,675,819,064 25,000 36,849,301 156,478,370 6,832,297,434 100,027,125,822 1993 18,316,141,504 25,000 42,249,632 144,611,495 18,460,752,999 148,728,484,113 1994 4,806,516,735 25,000 48,298,184 133,486,263 4,940,002,998 194,312,298,160 1995 2,051,883,936 25,000 35,558,875 62,623,394 2,114,507,330 120,896,479,274 1996 1,287,633,455 25,000 37,485,515 47,610,761 1,335,244,216 133,227,871,640 1997 673,162,484 0 0 0 673,162,484 53,549,387,834 IRR = 33% Source: Estimated by the author. 223 University of Ghana http://ugspace.ug.edu.gh Table A9: Financial Rate of Return with Higher Research Cost Adoption rate Yield of Yield of Total quantity Total area National average (at K = 95%, improved farmer Change in Proportional Prportional Year produced planted production Lag = 30 years technology technology yield increase production increase Unit (mt) (ha) (mt/ha) (mt/ha) (mt/ha) (mt/ha) Parameter Q A Y t Yn Yf = Ym dY=Yn-Yf dY/Yf j = t x (dY/Yf) 1979 308600 313900 0.98 0 1980 354000 319900 1.11 0 1981 334200 315500 1.06 0 1982 264300 276300 0.96 0 1983 140800 279800 0.50 0 1984 574000 723600 0.79 0.02 2.88 1.73 1.15 0.6647 0.0133 1985 395000 405000 0.98 0.03 2.75 1.7 1.05 0.6176 0.0185 1986 559100 472100 1.18 0.04 2.93 1.74 1.19 0.6839 0.0274 1987 597700 548336 1.09 0.06 2.93 1.74 1.19 0.6839 0.0410 1988 600000 500000 1.20 0.08 2.93 1.74 1.19 0.6839 0.0547 1989 714614 595800 1.20 0.1 2,93 1.74 1.19 0.6839 0.0684 1990 552549 464784 1.19 0.13 2.93 1.74 1.19 0.6839 0.0889 1991 931478 610445 1.53 0.17 2.93 1.74 1.19 0.6839 0.1163 1992 730617 606783 1.20 0.22 2.93 1.74 1.19 0.6839 0.1505 1993 960920 636670 1.51 0.28 2.93 1.74 1.19 0.6839 0.1915 1994 939908 629401 1.49 0.34 2.93 1.74 1.19 0.6839 0.2325 1995 1034170 668600 1.55 0.41 2.93 1.74 1.19 0.6839 0.2804 1996 1007610 664950 1.52 0.48 2.93 1.74 1.19 0.6839 0.3283 1997 1000000 650000 1.54 0.54 2.93 1.74 1.19 0.6839 0.3693 224 University of Ghana http://ugspace.ug.edu.gh Table A 9 continued. Change in Fertilizer prices Change in fertilizer costs row planting Improved Farmers Change in Change in total (comp. + S.Ammonia) including application cost seed seed seed cost cost of technology Year (cedis/ha) (cedis/ha) (cedis/ha) (cedis/ha) (cedis/ha) (cedis/ha) (cedis/ha) dC 1979 1980 1981 1982 1983 1984 1,838 2,059 551 400 300 100 2,710 1985 1,838 2,059 551 1,000 750 250 2,860 1986 3,175 3,556 953 1,000 750 250 4,759 1987 5,500 6,160 1,650 2,000 1,500 500 8,310 1988 9,750 10,920 2,925 2,000 1,500 500 14,345 1989 14,250 15,960 4,275 3,000 2,250 750 20,985 1990 18,250 20,440 5,475 3,000 2,250 750 26,665 1991 23,750 26,600 7,125 5,000 3,750 1,250 34,975 1992 36,250 40,600 10,875 5,000 3,750 1,250 52,725 1993 43,750 49,000 13,125 9,000 6,750 2,250 64,375 1994 58,750 65,800 17,625 9,000 6,750 2,250 85,675 1995 92,500 103,600 27,750 12,000 9,000 3,000 134,350 1996 107,500 120,400 32,250 20,000 15,000 5,000 157,650 1997 132,500 148,400 39,750 30,000 22,250 7,750 195,900 225 University of Ghana http://ugspace.ug.edu.gh Table A 9 continued. Maize price: Average wholesale Farmgate price Rural price index Real price Real change Proportional cost increase Year price (70% of wholesale) (1977 = 100) (1997= 1) in cost Supply elasticity (cedis/t) (cedis/t) (cedis/t) (cedis/ha) (E = 0.79) dC i = l/P = (dC . t)/(Y f. P) 1979 279 0.79 1980 448 0.79 1981 938 0.79 1982 1,150 0.79 1983 0 0 2,640 0 0.0000 0.79 1984 r 23,380 16,366 3,652 354,204 58,651 0.0019 0.79 1985 20,380 14,266 3,974 283,737 56,882 0.0035 0.79 1986 33,110 23,177 4,824 379,744 77,966 0.0047 0.79 1987 53,870 37,709 6,591 452,205 99,653 0.0076 0.79 1988 68,590 48,013 8,581 442,244 132,131 0.0137 0,79 1989 53,000 37,100 10,637 275,674 155,931 0.0325 0.79 1990 85,030 59,521 14,310 328,755 147,280 0.0335 0.79 1991 116,050 81,235 16,784 382,551 164,704 0.0421 0.79 1992 100,480 70,336 18,613 298,678 223,894 0.0948 0.79 1993 110,720 77,504 23,092 265,280 220,342 0.1337 0.79 1994 138,630 97,041 28,598 268,201 236,788 0.1725 0.79 1995 200,000 140,000 44,880 246,557 236,606 0.2261 0.79 1996 250,000 175,000 62,230 222,269 200,233 0.2485 0.79 1997 300,000 210,000 79,039 210,000 195,900 0.2895 0.79 226 University of Ghana http://ugspace.ug.edu.gh Table A9 continued. Social gains from research Change in quantity caused by GGDP (all loca Proportional GGDP Year Net shift in supply Demand elasticity research, cedis maize cost (cedis) (e = - 0.981) d Q (100% for 1979-96) k = ( j / E) - i dQ = (Q x e x E x k) / (e + E) SG = (k . P . Q) -1/2 (k . P . dQ) 1979 0.981 0 389,380 389,380 1980 0.981 0 389,380 389,380 1981 0.981 0 519,173 519,173 1982 0.981 0 1,402,036 1,402,036 1983 0.981 0 0 1,030,870 1,030,870 1984 0.0149 0.981 3,746 3,022,432,946 9,918,988 9,918,988 1985 0.0199 0.981 3,443 2,222,511,149 36,387,694 36,387,694 1986 0.0299 0.981 7,317 6,308,461,916 65,949,829 65,949,829 1987 0.0443 0.981 11,598 11,868,961,201 80,393,569 80,393,569 1988 0.0555 0.981 14,577 14,553,035,743 98,738,058 98,738,058 1989 0.0541 0.981 16,906 10,524,457,658 187,897,196 187,897,196 1990 0.0791 0.981 19,119 14,115,007,059 201,386,259 201,386,259 1991 0.1051 0.981 42,843 36,591,818,286 278,083,542 278,083,542 1992 0.0957 0.981 30,589 20,441,327,147 427,815,898 427,815,898 1993 0.1087 0.981 45,724 27,059,117,346 450,923,659 450,923,659 1994 0.1218 0.981 50,107 29,891,504,647 699,105,294 699,105,294 1995 0.1288 0.981 58,296 31,920,145,086 192,709,160 192,709,160 1996 0.1670 0.981 73,647 36,040,383,882 815,007,795 815,007,795 1997 0.1780 0.981 77,881 35,919,159,696 227 University of Ghana http://ugspace.ug.edu.gh Table A9 continued. GGDP (foreign costs), Exchange rate Exchange rate, Exchange rate All costs Proportional foreign Total GGDP NARP-CRI Cost Year (Canadian $) (CN$:1US$) cedis: 1US$ Cedis:CN$ (CN$ to cedis) cost (cedis) maize cost (cedis) cedis (100% for 1979-96) 1979 228,346 1.1714 2.75 2.35 536,069 536,069 925,449 0 1980 228,346 1.1692 2.75 2.35 537,078 537,078 926,458 0 1981 304,462 1.1989 2.75 2.29 698,366 698,366 1,217,539 0 1982 319,903 1.2337 2.75 2.23 713,085 713,085 2,115,121 0 1983 197,800 1.2324 8.83 7.16 1,417,214 1,417,214 2,448,084 0 1984 998,300 1.2951 35.99 27.79 27,742,118 27,742,118 37,661,106 0 1985 1,430,482 1.3655 54.37 39.82 56,957,383 56,957,383 93,345,077 0 1986 319,101 1.3895 89.2 64.20 20,484,929 20,484,929 86,434,758 0 1987 175,183 1.326 153.73 115.94 20,309,866 20,309,866 100,703,435 0 1988 2,023,483 1.2307 202.35 164.42 332,698,290 332,698,290 431,436,348 0 1989 365,767 1.184 270 228.04 83,409,704 83,409,704 271,306,900 0 1990 501,532 1.1668 326.33 279.68 140,268,202 140,268,202 341,654,461 0 1991 2,484,702 1.1457 367.83 321.05 797,720,116 797,720,116 1,075,803,658 0 1992 1,160,624 1.2087 437.09 361.62 419,704,761 419,704,761 847,520,659 0 1993 4,821,245 1.2901 649.06 503.11 2,425,608,309 2,425,608,309 2,876,531,968 0 1994 679,765 1.3656 956.71 700.58 476,228,744 476,228,744 1,175,334,038 0 1995 212,020 1.3724 1200.43 874.69 185,452,615 185,452,615 378,161,775 1,361,979,250 1996 55,996 1.3635 1637.23 1,200.76 67,237,500 67,237,500 882,245,295 1,535,228,229 1997 0 0 1,927,478,424 228 University of Ghana http://ugspace.ug.edu.gh Table A9 continued. NARP-CRI Maize SARI cost SARI maize (8%) GGDP ,SARI and MOFA Extension Actual releases MOFA Extesion cost Year (15% of NARP-CRI) NARP-CRI estimates (80% of estimates) for maize at 30% (cedis) (cedis) (cedis) (cedis) (cedis) (cedis) 1979 0 0 0 925,449 0 0 0 1980 0 0 0 926,458 0 0 0 1981 0 0 0 1,217,539 0 0 0 1982 0 0 0 2,115,121 0 0 0 1983 0 0 0 2,448,084 0 0 0 1984 0 0 0 37,661,106 156,631,040 125,304,832 37,591,450 1985 0 0 0 93,345,077 195,788,800 156,631,040 46,989,312 1986 0 0 0 86,434,758 244,736,000 195,788,800 58,736,640 1987 0 0 0 100,703,435 305,920,000 244,736,000 73,420,800 1988 0 0 0 431,436,348 382,400,000 305,920,000 91,776,000 1989 0 0 0 271,306,900 853,829,000 683,063,200 204,918,960 1990 0 0 0 341,654,461 857,208,000 685,766,400 205,729,920 1991 0 0 0 1,075,803,658 1,997,101,000 1,597,680,800 479,304,240 1992 0 0 0 847,520,659 1,516,825,000 1,213,460,000 364,038,000 1993 0 0 0 2,876,531,968 2,435,152,000 1,948,121,600 584,436,480 1994 0 535,742,408 42,859,393 1,218,193,431 2,272,267,000 1,817,813,600 545,344,080 1995 204,296,888 597,144,000 47,771,520 630,230,182 2,455,425,000 1,964,340,000 589,302,000 1996 230,284,234 1,171,434,025 93,714,722 1,206,244,251 533,000,000 426,400,000 127,920,000 1997 289,121,764 1,353,509,000 108,280,720 397,402,484 829,000,000 663,200,000 198,960,000 229 University of Ghana http://ugspace.ug.edu.gh Table A 9 continued. SG 2000 80% of SG 2000 Total extension Extension + Research Extension + Research CIMMYT Costs CIMMYT Costs Year (cedis) (cedis) (cedis) cost (cedis) (Real cost in cedis) (CN$) (cedis) 1979 0 0 0 925,449 262,174,129 50,000 117,381 1980 0 0 0 926,458 163,451,579 50,000 117,602 1981 0 0 0 1,217,539 102,593,851 50,000 114,688 1982 0 0 0 2,115,121 145,371,363 50,000 111,453 1983 0 0 0 2,448,084 73,293,211 50,000 358,244 1984 0 0 37,591,450 75,252,556 1,628,665,592 50,000 1,389,468 1985 0 0 46,989,312 140,334,389 2,791,114,686 50,000 1,990,846 1986 0 0 58,736,640 145,171,398 2,378,565,951 50,000 3,209,788 1987 0 0 73,420,800 174,124,235 2,088,090,643 25,000 2,898,379 1988 20,789,424 16,631,539 108,407,539 539,843,887 4,972,464,862 25,000 4,110,466 1989 24,458,146 19,566,517 224,485,477 495,792,377 3,684,021,219 25,000 5,701,014 1990 28,774,290 23,019,432 228,749,352 570,403,812 3,150,534,375 25,000 6,991,987 1991 33,852,105 27,081,684 506,385,924 1,582,189,582 7,450,827,120 25,000 8,026,316 1992 39,826,006 31,860,805 395,898,805 1,243,419,464 5,280,106,967 25,000 9,040,498 1993 46,854,125 37,483,300 621,919,780 3,498,451,748 11,974,455,557 25,000 12,577,707 1994 55,122,500 44,098,000 589,442,080 1,807,635,511 4,995,933,392 25,000 17,514,463 1995 64,850,000 51,880,000 641,182,000 1,271,412,182 2,239,107,564 25,000 21,867,349 1996 84,305,000 67,444,000 195,364,000 1,401,608,251 1,780,197,888 25,000 30,018,885 1997 96,000,000 76,800,000 275,760,000 673,162,484 673,162,484 0 o' 230 University of Ghana http://ugspace.ug.edu.gh Table A 9 continued. CIMMYT Real Costs Research, extension Net social gains (1997= 1) and CIMMYT real costs NG = SG - ( R + E ) (cedis) (cedis) Year 1979 33,253,297 295,427,426 -295,427,426 1980 20,748,051 184,199,630 -184,199,630 1981 9,664,031 112,257,882 -112,257,882 1982 7,660,140 153,031,503 -153,031,503 1983 10,725,475 84,018,686 -84,018,686 1984 30,071,786 1,658,737,377 1,363,695,569 1985 39,595,990 2,830,710,676 -608,199,527 1986 52,590,881 2,431,156,832 3,877,305,085 1987 34,757,236 2,122,847,879 9,746,113,322 1988 37,861,215 5,010,326,078 9,542,709,665 1989 42,361,794 3,726,383,013 6,798,074,645 1990 38,619,122 3,189,153,497 10,925,853,563 1991 37,797,425 7,488,624,544 29,103,193,742 1992 38,389,939 5,318,496,906 15,122,830,241 1993 43,050,814 12,017,506,371 15,041,610,974 1994 48,406,378 5,044,339,770 24,847,164,877 1995 38,510,994 2,277,618,558 29,642,526,528 1996 38,127,313 1,818,325,201 34,222,058,681 1997 0 673,162,484 35,245,997,213 IRR = 73% Source: Estimated by the author. 231 University of Ghana http://ugspace.ug.edu.gh Table A10: Economic Rate of Return with Higher Research Cost Yield of Yield of Change in Proportional Proportional Year Total quantity Total area National average Adoption rate Improved Farmer yield increase production increase produced planted production (at K = 95%, technology technology (j-parameter) Lag = 30 years) (mt/ha) (mt/ha) (mt/ha) Unit (mt) (ha) (mt/ha) Parameter Q A Y t Yn Yf = Ym dY=Yn-Yf dY/Yf j = t x (dY/Yf) 1979 308,600 313,900 0.98 0 0.0000 1980 354,000 319,900 1.11 0 0.0000 1981 334,200 315,500 1.06 0 0.0000 1982 264,300 276,300 0.96 0 0.0000 1983 140,800 279,800 0.50 0 0.0000 1984 574,000 723,600 0.79 0.02 2.88 1.73 1.15 0.6647 0.0133 1985 395,000 405,000 0.98 0.03 2.75 1.7 1.05 0.6176 0.0185 1986 559,100 472,100 1.18 0.04 2.93 1.74 1.19 0.6839 0.0274 1987 597,700 548,336 1.09 0.06 2.93 1.74 1.19 0.6839 0.0410 1988 600,000 500,000 1.20 0.08 2.93 1.74 1.19 0.6839 0.0547 1989 714,614 595,800 1.20 0.1 2.93 1.74 1.19 0.6839 0.0684 1990 552,549 464,784 1.19 0.13 2.93 1.74 1.19 0.6839 0.0889 1991 931,478 610,445 1.53 0.17 2.93 1.74 1.19 0.6839 0.1163 1992 730,617 606,783 1.20 0.22 2.93 1.74 1.19 0.6839 0.1505 1993 960,920 636,670 1.51 0.28 2.93 1.74 1.19 0.6839 0.1915 1994 939,908 629,401 1.49 0.34 2.93 1.74 1.19 0.6839 0.2325 1995 1,034,170 668,600 1.55 0.41 2.93 1.74 1,19 0.6839 0.2804 1996 1,007,610 664,950 1.52 0.48 2.93 1.74 1.19 0,6839 0.3283 1997 1,000,000 650,000 1.54 0.54 2.93 1.74 1,19 0.6839 0.3693 232 University of Ghana http://ugspace.ug.edu.gh Table A 10 continued. Product price: Year Change in fertilizer costs Change in Improved Farmers Change in Change in total Export price (F.O.B) Fertilizer prices including application row planting seed seed seed cost cost of technology U.S. Gulf (comp. + S.A, (cedis/ha) cost (cedis/ha) (cedis/ha) (cedis/ha) (cedis/ha) ($/ton) cedis/ha) (cedis/ha) (Without subsidy) dC 1979 1980 1981 1982 1983 0 1984 2,665 2,985 800 400 300 100 3,884 139.1 1985 2,922 3,273 877 1,000 750 250 4,399 116.8 1986 4,318 4,836 1,295 1,000 750 250 6,382 88.7 1987 7,810 8,747 2,343 2,000 1,500 500 11,590 75.6 1988 12,675 14,196 3,803 2,000 1,500 500 18,499 107 1989 16,388 18,355 4,916 3,000 2,250 750 24,021 111.3 1990 18,250 20,440 5,475 3,000 2,250 750 26,665 109.7 1991 23,750 26,600 7,125 5,000 3,750 1,250 34,975 107 1992 36,250 40,600 10,875 5.000 3,750 1,250 52,725 105.1 1993 43,750 49,000 13,125 9,000 6,750 2,250 64,375 101.9 1994 58,750 65,800 17,625 9,000 6,750 2,250 85,675 107.5 1995 92,500 103,600 27,750 12,000 9,000 3,000 134,350 123.7 1996 107,500 120,400 32,250 20,000 15,000 5,000 157,650 165.1 1997 132,500 148,400 39,750 30,000 22,250 7,750 195,900 117.1 233 University of Ghana http://ugspace.ug.edu.gh Table A 10 continued. Freight and Value of maize marketing cost Real (Cedis/ton) No.2 yellow maize Value of maize exchange rate Real price Proportional cost increase (30% + 20% = 50% (cedis: 1 US$) Rural price (Farmgate) Real change in Year of export price) index (1997= 1) cost ($/ton) ($/ton) (1997 = 100) (cedis/ton) (cedis/t) i = l/P = (dC. t)/(Yf. P) dC 1979 279 1980 448 1981 938 1982 1,150 1983 0 2,640 1984 69.55 208.65 611.07 127,500 3,652 1,931,604 84,067 0.0000 1985 58.4 175.2 870.05 152,433 3,974 2,122,218 87,497 0.0000 1986 44.35 133.05 1164.38 154,921 4,824 1,776,813 104,559 0.0001 1987 37.8 113.4 1572.39 178,309 6,591 1,496,792 138,989 0.0003 1988 53.5 160.5 1550.58 248,868 8,581 1,604,615 170,388 0.0005 1989 55.65 166.95 1733.12 289,344 10,637 1,504,996 178,489 0.0009 1990 54.85 164.55 1609.3 264,810 14,310 1,023,846 147,280 0.0019 1991 53.5 160.5 1602.46 257,195 16,784 847,825 164,704 0.0040 1992 52.55 157.65 1781.59 280,868 18,613 834,882 223,894 0.0080 1993 50.95 152.85 2180.25 333,251 23,092 798,454 220,342 0.0130 1994 53.75 161.25 2638.24 425,416 28,598 823,034 236,788 0.0203 1995 61.85 185,55 1952.04 362,201 44,880 446,515 236,606 0.0709 1996 82.55 247.65 2044.46 506,311 62,230 450,149 200,233 0.0966 1997 58.55 175.65 2050.2 360,118 79,039 252,082 195,900 0.2412 234 University of Ghana http://ugspace.ug.edu.gh Table A 10 continued. Net shift in supply Change in quantity Research cost: caused by research Social gains from research GGDP Year (all local cost) Supply elasticity Demand elasticity dQ = (Q x e x E x k) / (e + E) SG = (k . P . Q) -1/2 (k . P . dQ) (cedis) ( E = 0.79) (e = - 0.981) dQ 1979 0.79 0,981 389,380 1980 0.79 0.981 389,380 1981 0.79 0.981 519,173 1982 0.79 0.981 1,402,036 1983 0.79 0.981 1,030,870 1984 0.79 0.0168 0.981 4,221.3 18,564,547,348 9,918,988 1985 0.79 0.0234 0.981 4,047.9 19,530,468,891 36,387,694 1986 0.79 0.0345 0.981 8,452.0 34,058,854,442 65,949,829 1987 0.79 0.0517 0.981 13,515.9 45,707,766,058 80,393,569 1988 0.79 0.0687 0.981 18,044.8 65,172,721,789 98,738,058 1989 0.79 0,0857 0.981 26,785.1 90,393,025,287 187,897,196 1990 0.79 0.1106 0.981 26,741.6 61,052,934,824 201,386,259 1991 0.79 0.1431 0.981 58,345.9 109,501,384,670 278,083,542 1992 0.79 0.1825 0.981 58,339.2 106,859,423,256 427,815,898 1993 0.79 0.2294 0.981 96,472.4 167,189,237,112 450,923,659 1994 0.79 0.2740 0.981 112,697.1 199,252,301,157 699,105,294 1995 0.79 0.2840 0.981 128,543.7 123,010,986,604 192,709,160 1996 0.79 0.3189 0.981 140,624.8 134,563,115,856 815,007,795 1997 0.79 0.2263 0,981 99,030.7 54,222,550,318 235 University of Ghana http://ugspace.ug.edu.gh Table A 10 continued. Real Proportional GGDP GGDP (foreign costs), Exchange rate Exchange rate, Cedis:CN$ All costs Proportional foreign maize cost (cedis) Canadian $ (CN$:1US$) cedis; 1US$ (CN$ to cedis) cost (cedis) Year (90% for 1979-83, (100% for 1979-96) 55% for 1984-96) 1979 389,380 228,346 1.1714 401.32 342.60 78,231,020 78,231,020 1980 389,380 228,346 1.1692 306.92 262.50 59,941,802 59,941,802 1981 519,173 304,462 1.1989 154.19 128.61 39,156,723 39,156,723 1982 1,402,036 319,903 1.2337 135.54 109.86 35,146,026 35,146,026 1983 1,030,870 197,800 1.2324 201.27 163.32 32,303,802 32,303,802 1984 9,918,988 998,300 1.2951 611.07 471.83 471,030,176 471,030,176 1985 36,387,694 1,430,482 1.3655 870.05 637.17 911,454,313 911,454,313 1986 65,949,829 319,101 1.3895 1164.38 837.98 267,401,815 267,401,815 1987 80,393,569 175,183 1.326 1572.39 1,185.81 207,734,538 207,734,538 1988 98,738,058 2,023,483 1.2307 1550.58 1,259.92 2,549,420,874 2,549,420,874 1989 187,897,196 365,767 1.184 1733.12 1,463.78 535,403,803 535,403,803 1990 201,386,259 501,532 1.1668 1609.3 1,379.24 691,734,185 691,734,185 1991 278,083,542 2,484,702 1.1457 1602.46 1,398.67 3,475,286,346 3,475,286,346 1992 427,815,898 1,160,624 1.2087 1781.59 1,473.97 1,710,727,320 1,710,727,320 1993 450,923,659 4,821,245 1.2901 2180.25 1,689.99 8,147,833,045 8,147,833,045 1994 699,105,294 679,765 1.3656 2638.24 1,931.93 1,313,256,600 1,313,256,600 1995 192,709,160 212,020 1.3724 1952.04 1,422.35 301,567,707 301,567,707 1996 815,007,795 55,996 1.3635 2044.46 1,499.42 83,961,556 83,961,556 1997 • 0 2050.2 0 236 University of Ghana http://ugspace.ug.edu.gh Table A10 continued. Total GGDP NARP-CRI Cost NARP-CRI Maize SARI cost SARI maize (8%) GGDP ,SARI and MOFA Extension maize cost (cedis) (15% of NARP-CRI) NARP-CRI estimates Year (cedis) (cedis) (cedis) (cedis) (cedis) (cedis) 1979 78,620,400 0 0 0 0 78,620,400 0 1980 60,331,182 0 0 0 0 60,331,182 0 1981 39,675,896 0 0 0 0 39,675,896 0 1982 36,548,062 0 0 0 0 36,548,062 0 1983 33,334,672 0 0 0 0 33,334,672 0 1984 480,949,164 0 0 0 0 480,949,164 156,631,040 1985 947,842,007 0 0 0 0 947,842,007 195,788,800 1986 333,351,644 0 0 0 0 333,351,644 244,736,000 1987 288,128,107 0 0 0 0 288,128,107 305,920,000 1988 2,648,158,932 0 0 0 0 2,648,158,932 382,400,000 1989 723,300,999 0 0 0 0 723,300,999 853,829,000 1990 893,120,444 0 0 0 0 893,120,444 857,208,000 1991 3,753,369,888 0 0 0 0 3,753,369,888 1,997,101,000 1992 2,138,543,218 0 0 0 0 2,138,543,218 1,516,825,000 1993 8,598,756,704 0 0 0 0 8,598,756,704 2,435,152,000 1994 2,012,361,894 0 0 535,742,408 42,859,393 2,055,221,287 2,272,267,000 1995 494,276,867 1,361,979,250 204,296,888 597,144,000 47,771,520 746,345,274 2,455,425,000 1996 898,969,351 1,535,228,229 230,284,234 1,171,434,025 93,714,722 1,222,968,308 533,000,000 1997 1,927,478,424 289,121,764 1,353,509,000 108,280,720 397,402,484 829,000,000 237 University of Ghana http://ugspace.ug.edu.gh Table A 10 continued. Extension + Research Actual releases MOFA Extesion cost SG 2000 80% of SG 2000 Total extension Extension + Research (Real cost) (80% of estimates) for maize at 30% Year (cedis) (cedis) (cedis) (cedis) (cedis) (cedis) (cedis) 1979 0 0 0 0 0 78,620,400 22,272,680,245 1980 0 0 0 0 0 60,331,182 10,644,009,498 1981 0 0 0 0 0 39,675,896 3,343,223,008 1982 0 0 0 0 0 36,548,062 2,511,932,430 1983 0 0 0 0 0 33,334,672 998,007,260 1984 125,304,832 37,591,450 0 0 37,591,450 518,540,614 11,222,599,004 1985 156,631,040 46,989,312 0 0 46,989,312 994,831,319 19,786,228,637 1986 195,788,800 58,736,640 0 0 58,736,640 392,088,284 6,424,184,474 1987 244,736,000 73,420,800 0 0 73,420,800 361,548,907 4,335,679,572 1988 305,920,000 91,776,000 20,789,424 16,631,539 108,407,539 2,756,566,472 25,390,543,918 1989 683,063,200 204,918,960 24,458,146 19,566,517 224,485,477 947,786,476 7,042,596,154 1990 685,766,400 205,729,920 28,774,290 23,019,432 228,749,352 1,121,869,796 6,196,468,680 1991 1,597,680,800 479,304,240 33,852,105 27,081,684 506,385,924 4,259,755,813 20,059,988,064 1992 1,213,460,000 364,038,000 39,826,006 31,860,805 395,898,805 2,534,442,023 10,762,357,658 1993 1,948,121,600 584,436,480 46,854,125 37,483,300 621,919,780 9,220,676,484 31,560,412,637 1994 1,817,813,600 545,344,080 55,122,500 44,098,000 589,442,080 2,644,663,367 7,309,306,520 1995 1,964,340,000 589,302,000 64,850,000 51,880,000 641,182,000 1,387,527,274 2,443,600,005 1996 426,400,000 127,920,000 84,305,000 67,444,000 195,364,000 1,418,332,308 1,801,439,294 1997 663,200,000 198,960,000 96,000,000 76,800,000 275,760,000 673,162,484 673,162,484 238 University of Ghana http://ugspace.ug.edu.gh Table A 10 continued. CIMMYT cost CIMMYT Cost CIMMYT Real Costs Extension , Research NET SOCIAL GAINS (1997= 1) and CIMMYT real cost NG = SG - (R + E) Year (CN$) (cedis) (cedis) (cedis) 1979 50,000 17,129,930 4,852,804,793 27,125,485,037 -27,125,485,037 1980 50,000 13,125,214 2,315,633,427 12,959,642,925 -12,959,642,925 1981 50,000 6,430,478 541,853,460 3,885,076,469 -3,885,076,469 1982 50,000 5,493,232 377,547,429 2,889,479,859 -2,889,479,859 1983 50,000 8,165,774 244,475,234 1,242,482,495 -1,242,482,495 1984 50,000 23,591,615 510,585,329 11,733,184,333 6,831,363,015 1985 50,000 31,858,294 633,630,517 20,419,859,154 -889,390,263 1986 50,000 41,899,244 686,499,663 7,110,684,137 26,948,170,305 1987 25,000 29,645,362 355,505,958 4,691,185,530 41,016,580,528 1988 25,000 31,497,928 290,125,246 25,680,669,164 39,492,052,626 1989 25,000 36,594,595 271,918,789 7,314,514,943 83,078,510,344 1990 25,000 34,481,059 190,450,625 6,386,919,305 54,666,015,519 1991 25,000 34,966,833 164,665,364 20,224,653,428 89,276,731,242 1992 25,000 36,849,301 156,478,370 10,918,836,028 95,940,587,228 1993 25,000 42,249,632 144,611,495 31,705,024,132 135,484,212,980 1994 25,000 48,298,184 133,486,263 7,442,792,784 191,809,508,374 1995 25,000 35,558,875 62,623,394 2,506,223,399 120,504,763,205 1996 25,000 37,485,515 47,610,761 1,849,050,055 132,714,065,801 1997 0 0 0 673,162,484 53,549,387,834 IRR = 30% Source: Estimated by the author. 239 University of Ghana http://ugspace.ug.edu.gh Table A l l : Financial Rate of Return with Higher Extension Cost. Year Total quantity produced Total area planted National average Adoption rate Yield of improved Yield of farmer Change in Proportional production (at K = 95%, technology technology yield increase Unit (mt) (ha) (mt/ha) Lap = 30 years) (mt/ha) (mt/ha) (mt/ha) Parameter Q A Ym t Yn Yf dY=Yn-Yf dY/Yf 1979 308,600 313,900 0.98 0 1980 354,000 319,900 1.11 0 1981 334,200 315,500 1.06 0 1982 264,300 276,300 0.96 0 1983 140,800 279,800 0.50 0 1984 574,000 723,600 0.79 0.02 2.88 1.73 1.15 0.6647 1985 395,000 405,000 0.98 0.03 2.75 1.7 1.05 0.6176 1986 559,100 472,100 1.18 0.04 2.93 1.74 1.19 0.6839 1987 597,700 548,336 1.09 0.06 2.93 1.74 1.19 0.6839 1988 600,000 500,000 1.20 0.08 2.93 1.74 1.19 0.6839 1989 714,614 595,800 1.20 0.1 2.93 1.74 1.19 0.6839 1990 552,549 464,784 1.19 0.13 2.93 1.74 1.19 0.6839 1991 931,478 610,445 1.53 0.17 2.93 1.74 1.19 0.6839 1992 730,617 606,783 1.20 0.22 2.93 1.74 1.19 0.6839 1993 960,920 636,670 1.51 0.28 2.93 1.74 1.19 0.6839 1994 939,908 629,401 1.49 0.34 2.93 1.74 1.19 0.6839 1995 1,034,170 668,600 1.55 0.41 2.93 1.74 1.19 0.6839 1996 1,007,610 664,950 1.52 0.48 2.93 1.74 1.19 0.6839 1997 1,000,000 650,000 1.54 0.54 2.93 1.74 1.19 0.6839 240 University of Ghana http://ugspace.ug.edu.gh Table A11 continued. Proportional Fertilizer prices Change in fertilizer costs Change in Improved Farmers Change in Change in total Year production increase (comp. + S.A, including application row planting cost seed seed seed cost cost of technology (j-parameter) cedis/ha) (cedis/ha) (cedis/ha) (cedis/ha) (cedis/ha) ^cedis/ha) (cedis/ha) j = t x (dY/Yf) dC 1979 1980 1981 1982 1983 1984 ■“ 0.0133 1,838 2,059 551 400 300 100 2,710 1985 0.0185 1,838 2,059 551 1,000 750 250 2,860 1986 0.0274 3,175 3,556 953 1,000 750 250 4,759 1987 0.0410 5,500 6,160 1,650 2,000 1,500 500 8,310 1988 0.0547 9,750 10,920 2,925 2,000 1,500 500 14,345 1989 0.0684 14,250 15,960 4,275 3,000 2,250 750 20,985 1990 0.0889 18,250 20,440 5,475 3,000 2,250 750 26,665 1991 0.1163 23,750 26,600 7,125 5,000 3,750 1,250 34,975 1992 0.1505 36,250 40,600 10,875 5,000 3,750 1,250 52,725 1993 0.1915 43,750 49,000 13,125 9,000 6,750 2,250 64,375 1994 0.2325 58,750 65,800 17,625 9,000 6,750 2,250 85,675 1995 0.2804 92,500 103,600 27,750 12,000 9,000 3,000 134,350 1996 0.3283 107,500 120,400 32,250 20,000 15,000 5,000 157,650 1997 0.3693 132,500 148,400 39,750 30,000 22,250 7,750 195,900 241 University of Ghana http://ugspace.ug.edu.gh Table A11 continued. Product price: Average wholesale Farmgate price Real change in cost Proportional cost increase Year price (70% of wholesale) I = (dC . T)/Yf (cedis /t) (cedis/t) Rural price index Real price (cedis/t) (cedis/t) (1977 = 100) (1997= 1) dC i = l/P = (dC . t)/(Yf. P) 1979 279 1980 448 1981 938 1982 1,150 1983 2,640 0 0.0000 1984 23,380 16,366 3,652 354,204 58,651 0.0019 1985 20,380 14,266 3,974 283,737 56,882 0.0035 1986 33,110 23,177 4,824 379,744 77,966 0.0047 1987 53,870 37,709 6,591 452,205 99,653 0.0076 1988 68,590 48,013 8,581 442,244 132,131 0.0137 1989 53,000 37,100 10,637 275,674 155,931 0.0325 1990 85,030 59,521 14,310 328,755 147,280 0.0335 1991 116,050 81,235 16,784 382,551 164,704 0.0421 1992 100,480 70,336 18,613 298,678 223,894 0.0948 1993 110,720 77,504 23,092 265,280 220,342 0.1337 1994 138,630 97,041 28,598 268,201 236,788 0.1725 1995 200,000 140,000 44,880 246,557 236,606 0.2261 1996 250,000 175,000 62,230 222,269 200,233 0.2485 1997 300,000 210,000 79,039 210,000 195,900 0.2895 242 University of Ghana http://ugspace.ug.edu.gh Table A11 continued. Research costs: Net shift in supply Change in quantity caused by Social gains from research GGDP Year research, (ail local costs) Supply elasticity Demand elasticity dQ = (Q x e x E x k) / (e + E) SG = (k . P . Q) -1/2 (k . P . dQ) (cedis) ( E = 0.79) oXUJII (e = - 0.981) dQ 1979 0.79 0.981 0 389,380 1980 0.79 0.981 0 389,380 1981 0.79 0.981 0 519,173 1982 0.79 0.981 0 1,402,036 1983 0.79 0.981 0 0 1,030,870 1984 0.79 0.0149 0.981 3,746 3,022,432,946 9,918,988 1985 0.79 0.0199 0.981 3,443 2,222,511,149 36,387,694 1986 0.79 0.0299 0.981 7,317 6,308,461,916 65,949,829 1987 0.79 0.0443 0.981 11,598 11,868,961,201 80,393,569 1988 0.79 0.0555 0.981 14,577 14,553,035,743 98,738,058 1989 0.79 0.0541 0.981 16,906 10,524,457,658 187,897,196 1990 0.79 0.0791 0.981 19,119 14,115,007,059 201,386,259 1991 0.79 0.1051 0.981 42,843 36,591,818,286 278,083,542 1992 0.79 0.0957 0.981 30,589 20,441,327,147 427,815,898 1993 0.79 0.1087 0.981 45,724 27,059,117,346 450,923,659 1994 0.79 0.1218 0.981 50,107 29,891,504,647 699,105,294 1995 0.79 0.1288 0.981 58,296 31,920,145,086 192,709,160 1996 0.79 0.1670 0.981 73,647 36,040,383,882 815,007,795 1997 0.79 0.1780 0.981 77,881 35,919,159,696 243 University of Ghana http://ugspace.ug.edu.gh Table A11 continued. Proportional GGDP GGDP (foreign costs), Exchange rate Exchange rate, Cedis:CN$ All costs Proportional foreign Year maize cost (cedis) Canadian $ (CN$:1US$) cedis: 1US$ (CN$ to cedis) cost (cedis) (90% of total) (90% of total) 1979 350,442 228,346 1.1714 2.75 2.3476 536,069 482,462 1980 350,442 228,346 1.1692 2.75 2.3520 537,078 483,370 1981 467,256 304,462 1.1989 2.75 2.2938 698,366 628,529 1982 1,261,832 319,903 1.2337 2.75 2.2291 713,085 641,777 1983 927,783 197,800 1.2324 8.83 7.1649 1,417,214 1,275,492 1984 8,927,089 998,300 1.2951 35.99 27.7894 27,742,118 24,967,906 1985 32,748,925 1,430,482 1.3655 54.37 39.8169 56,957,383 51,261,645 1986 59,354,846 319,101 1.3895 89.20 64.1958 20,484,929 18,436,436 1987 72,354,212 175,183 1.3260 153.73 115.9351 20,309,866 18,278,880 1988 88,864,252 2,023,483 1.2307 202.35 164.4186 332,698,290 299,428,461 1989 169,107,476 365,767 1.1840 270.00 228.0405 83,409,704 75,068,734 1990 181,247,633 501,532 1.1668 326.33 279.6795 140,268,202 126,241,381 1991 250,275,188 2,484,702 1.1457 367.83 321.0526 797,720,116 717,948,104 1992 385,034,308 1,160,624 1.2087 437.09 361.6199 419,704,761 377,734,285 1993 405,831,293 4,821,245 1.2901 649.06 503.1083 2,425,608,309 2,183,047,478 1994 629,194,765 679,765 1.3656 956.71 700.5785 476,228,744 428,605,870 1995 173,438,244 212,020 1.3724 1,200.43 874.6940 185,452,615 166,907,353 1996 733,507,016 55,996 1.3635 1,637.23 1,200.7554 67,237,500 60,513,750 1997 0 0 244 University of Ghana http://ugspace.ug.edu.gh Table A11 continued. Total GGDP NARP-CRI Cost NARP-CRI Maize SARI cost SARI maize (8%) GGDP ,SARI and MOFA Extension Year maize cost (cedis) (15% of NARP-CRI) (cedis) NARP-CRI (cedis) estimates (cedis) (cedis) (cedis) 1979 832,904 0 0 0 0 832,904 0 1980 833,812 0 0 0 0 833,812 0 1981 1,095,785 0 0 0 0 1,095,785 0 1982 1,903,609 0 0 0 0 1,903,609 0 1983 2,203,275 0 0 0 0 2,203,275 0 1984 33,894,995 0 0 0 0 33,894,995 156,631,040 1985 84,010,569 0 0 0 o' 84,010,569 195,788,800 1986 77,791,282 0 0 0 0 77,791,282 244,736,000 1987 90,633,092 0 0 0 0 90,633,092 305,920,000 1988 388,292,713 0 0 0 0 388,292,713 382,400,000 1989 244,176,210 0 0 0 0 244,176,210 853,829,000 1990 307,489,014 0 0 0 0 307,489,014 857,208,000 1991 968,223,292 0 0 0 0 968,223,292 1,997,101,000 1992 762,768,593 0 0 0 0 762,768,593 1,516,825,000 1993 2,588,878,771 0 0 0 0 2,588,878,771 2,435,152,000 1994 1,057,800,634 0 0 535,742,408 42,859,393 1,100,660,027 2,272,267,000 1995 340,345,597 1,361,979,250 204,296,888 597,144,000 47,771,520 592,414,005 2,455,425,000 1996 794,020,765 1,535,228,229 230,284,234 1,171,434,025 93,714,722 1,118,019,722 533,000,000 1997 1,927,478,424 289,121,764 1,353,509,000 108,280,720 397,402,484 829,000,000 245 University of Ghana http://ugspace.ug.edu.gh Table A11 continued. Year Actual releases MOFA Extesion cost SG 2000 80% of SG 2000 Total extension Extension + Research Extension + Research (80% of estimates) for maize at 50% (Real cost in cedis) (cedis) (cedis) (cedis) (cedis) (cedis) 1979 0 0 0 0 0 832,904 235,956,716 1980 0 0 0 0 0 833,812 147,106,421 1981 0 0 0 0 0 1,095,785 92,334,466 1982 0 0 0 0 0 1,903,609 130,834,226 1983 0 0 0 0 0 2,203,275 65,963,890 1984 125,304,832 62,652,416 0 0 62,652,416 96,547,411 2,089,542,949 1985 156,631,040 78,315,520 0 0 78,315,520 162,326,089 3,228,508,245 1986 195,788,800 97,894,400 0 0 97,894,400 175,685,682 2,878,528,328 1987 244,736,000 122,368,000 0 0 122,368,000 213,001,092 2,554,300,301 1988 305,920,000 152,960,000 20,789,424 16,631,539 169,591,539 557,884,252 5,138,633,424 1989 683,063,200 341,531,600 24,458,146 19,566,517 361,098,117 605,274,327 4,497,534,789 1990 685,766,400 342,883,200 28,774,290 23,019,432 365,902,632 673,391,646 3,719,371,231 1991 1,597,680,800 798,840,400 33,852,105 27,081,684 825,922,084 1,794,145,376 8,448,966,658 1992 1,213,460,000 606,730,000 39,826,006 31,860,805 638,590,805 1,401,359,398 5,950,789,526 1993 1,948,121,600 974,060,800 46,854,125 37,483,300 1,011,544,100 3,600,422,871 12,323,481,003 1994 1,817,813,600 908,906,800 55,122,500 44,098,000 953,004,800 2,053,664,827 5,675,907,905 1995 1,964,340,000 982,170,000 64,850,000 51,880,000 1,034,050,000 1,626,464,005 2,864,395,911 1996 426,400,000 213,200,000 84,305,000 67,444,000 280,644,000 1,398,663,722 1,776,458,009 1997 663,200,000 331,600,000 96,000,000 76,800,000 408,400,000 805,802,484 805,802,484 246 University of Ghana http://ugspace.ug.edu.gh Table A11 continued. Year CIMMYT Costs CIMMYT Costs CIMMYT Real Costs Research, extension Net social gains (cedis, 1997 = 1) and CIMMYT real cost NG = SG - ( R + E ) (CN$) (cedis) (cedis) 1979 50,000 117,381 33,253,297 269,210,013 -269,210,013 1980 50,000 117,602 20,748,051 167,854,472 -167,854,472 1981 50,000 114,688 9,664,031 101,998,497 -101,998,497 1982 50,000 111,453 7,660,140 138,494,367 -138,494,367 1983 50,000 358,244 10,725,475 76,689,365 -76,689,365 1984 50,000 1,389,468 30,071,786 2,119,614,734 902,818,212 1985 50,000 1,990,846 39,595,990 3,268,104,235 -1,045,593,085 1986 50,000 3,209,788 52,590,881 2,931,119,209 3,377,342,708 1987 25,000 2,898,379 34,757,236 2,589,057,538 9,279,903,663 1988 25,000 4,110,466 37,861,215 5,176,494,639 9,376,541,103 1989 25,000 5,701,014 42,361,794 4,539,896,583 5,984,561,075 1990 25,000 6,991,987 38,619,122 3,757,990,353 10,357,016,707 1991 25,000 8,026,316 37,797,425 8,486,764,083 28,105,054,203 1992 25,000 9,040,498 38,389,939 5,989,179,464 14,452,147,683 1993 25,000 12,577,707 43,050,814 12,366,531,817 14,692,585,528 1994 25,000 17,514,463 48,406,378 5,724,314,283 24,167,190,364 1995 25,000 21,867,349 38,510,994 2,902,906,905 29,017,238,182 1996 25,000 30,018,885 38,127,313 1,814,585,322 34,225,798,560 1997 0 0 0 805,802,484 35,113,357,213 IRR = 71% Source: Estimated by the author. 247 University of Ghana http://ugspace.ug.edu.gh Table A12: Economic Rate of Return with Higher Extension Cost Yield of Yield of Change in Proportional Proportional Year Total quantity Total area National average Adoption rate Improved Farmer yield increase production increase produced planted production (at K = 95%, technology technology Lag = 30 years) (mt/ha) (mt/ha) (mt/ha) (j-parameter) Unit (mt) (ha) (mt/ha) Parameter Q A Ym t Yn Yf = Ym dY=Yn-Yf dY/Yf j = t x (dY/Yf) 1979 308,600 313,900 0.98 0 0.0000 1980 354,000 319,900 1.11 0 0.0000 1981 334,200 315,500 1.06 0 0.0000 1982 264,300 276,300 0.96 0 0.0000 1983 140,800 279,800 0.50 0 0.0000 1984 574,000 723,600 0.79 0.02 2.88 1.73 1.15 0.6647 0.0133 1985 395,000 405,000 0.98 0.03 2.75 1.7 1.05 0.6176 0.0185 1986 559,100 472,100 1.18 0.04 2.93 1.74 1.19 0.6839 0.0274 1987 597,700 548,336 1.09 0.06 2.93 1.74 1.19 0.6839 0.0410 1988 600,000 500,000 1.20 0.08 2.93 1.74 1.19 0.6839 0.0547 1989 714,614 595,800 1.20 0.1 2.93 1.74 1,19 0.6839 0.0684 1990 552,549 464,784 1.19 0.13 2.93 1.74 1.19 0.6839 0.0889 1991 931,478 610,445 1.53 0.17 2.93 1.74 1.19 0.6839 0.1163 1992 730,617 606,783 1.20 0.22 2.93 1.74 1.19 0.6839 0.1505 1993 960,920 636,670 1.51 0.28 2.93 1.74 1.19 0.6839 0.1915 1994 939,908 629,401 1.49 0.34 2.93 1.74 1.19 0.6839 0.2325 1995 1,034,170 668,600 1.55 0.41 2.93 1.74 1.19 0.6839 0.2804 1996 1,007,610 664,950 1.52 0.48 2.93 1.74 1.19 0.6839 0.3283 1997 1,000,000 650,000 1.54 0.54 2,93 1.74 1.19 0.6839 0.3693 248 University of Ghana http://ugspace.ug.edu.gh Table A12 continued. Fertilizer prices Product price: (comp. + S.A, Year cedis/ha) Change in fertilizer costs Change in Improved Farmers Change in Change in total Export price (F.O.B) (Without subsidy) including application row planting seed seed seed cost cost of technology U.S. Gulf ($/ton) (cedis/ha) cost (cedis/ha) (cedis/ha) (cedis/ha) (cedis/ha) (cedis/ha) dC 1979 1980 1981 1982 1983 0 1984 2,665 2,985 800 400 300 100 3,884 139.1 1985 2,922 3,273 877 1,000 750 250 4,399 116.8 1986 4,318 4,836 1,295 1,000 750 250 6,382 88.7 1987 7,810 8,747 2,343 2,000 1,500 500 11,590 75.6 1988 12,675 14,196 3,803 2,000 1,500 500 18,499 107 1989 16,388 18,355 4,916 3,000 2,250 750 24,021 1113 1990 18,250 20,440 5,475 3,000 2,250 750 26,665 109.7 1991 23,750 26,600 7,125 5,000 3,750 1,250 34,975 107 1992 36,250 40,600 10,875 5,000 3,750 1,250 52,725 105.1 1993 43,750 49,000 13,125 9,000 6,750 2,250 64,375 101.9 1994 58,750 65,800 17,625 9,000 6,750 2,250 85,675 107.5 1995 92,500 103,600 27,750 12,000 9,000 3,000 134,350 123.7 1996 107,500 120,400 32,250 20,000 15,000 5,000 157,650 165.1 1997 132,500 148,400 39,750 30,000 22,250 7,750 195,900 117.1 249 University of Ghana http://ugspace.ug.edu.gh Table A12 continued. Freight and marketing cost ($/ton) Real No.2 yellow maize Value of maize Exchange rate Value of maize Real change in cost, (30% + 20% = 50% dC (cedis/mt) Year of export price) ($/ton) (cedis: 1 US$) (Cedis/ton) Rural price index Real price (Cedis/ton) (1997 = 100) (1997= 1) 1979 279 1980 448 1981 938 1982 1,150 1983 0 2,640 1984 69.55 208.65 611.07 127,500 3,652 1,931,604 84,067 1985 58.4 175.2 870.05 152,433 3,974 2,122,218 87,497 1986 44.35 133.05 1164.38 154,921 4,824 1,776,813 104,559 1987 37.8 113.4 1572.39 178,309 6,591 1,496,792 138,989 1988 53.5 160.5 1550.58 248,868 8,581 1,604,615 170,388 1989 55.65 166.95 1733.12 289,344 10,637 1,504,996 178,489 1990 54.85 164.55 1609.3 264,810 14,310 1,023,846 147,280 1991 53.5 160.5 1602.46 257,195 16,784 847,825 164,704 1992 52.55 157.65 1781.59 280,868 18,613 834,882 223,894 1993 50.95 152.85 2180.25 333,251 23,092 798,454 220,342 1994 53.75 161.25 2638.24 425,416 28,598 823,034 236,788 1995 61.85 185.55 1952.04 362,201 44,880 446,515 236,606 1996 82.55 247.65 2044.46 506,311 62,230 450,149 200,233 1997 58.55 175.65 2050.2 360,118 79,039 252,082 195,900 250 University of Ghana http://ugspace.ug.edu.gh Table A 12 continued. Proportional cost Change in quantity Year increase caused by research, dQ. Unit I = (dC . t)/Yf Net shift in supply dQ = (Q x e x E x k) / (e + E) Parameter i = l/P = (dC . t)/(Yf. P) Supply elasticity 1 UJ’■»-II Demand elasticity dQ ( E = 0.79) (e = - 0.981) 1979 0.79 0.981 1980 0.79 0.981 1981 0.79 0.981 1982 0.79 0.981 1983 0.79 0.981 1984 0.0000 0.79 0.0168 0.981 4,221.28 1985 0.0000 0.79 0.0234 0.981 4,047.91 1986 0.0001 0.79 0.0345 0.981 8,452.03 1987 0.0003 0.79 0.0517 0.981 13,515.88 1988 0.0005 0.79 0.0687 0.981 18,044.83 1989 0.0009 0.79 0.0857 0.981 26,785.11 1990 0.0019 0.79 0.1106 0.981 26,741.63 1991 0.0040 0.79 0.1431 0.981 58,345.87 1992 0.0080 0.79 0.1825 0.981 58,339.16 1993 0.0130 0.79 0.2294 0.981 96,472.40 1994 0.0203 0.79 0.2740 0.981 112,697.06 1995 0.0709 0.79 0.2840 0.981 128,543.75 1996 0.0966 0.79 0.3189 0.981 140,624.80 1997 0.2412 0.79 0.2263 0.981 99,030.72 251 University of Ghana http://ugspace.ug.edu.gh Table A12 continued. Research cost Real Social gains from research 3GDP (all local costs), Proportional GGDP GGDP (foreign costs), Exchange rate exchange rate, cedis maize cost (cedis) (CN$:1US$) Cedis: 1US$ SG = (k . P . Q) -1/2 (k . P . dQ) (in Candian $) (90% of total) Year 1979 0 389,380 350,442 228,346 1.1714 401.32 1980 0 389,380 350,442 228,346 1.1692 306.92 1981 0 519,173 467,256 304,462 1.1989 154.19 1982 0 1,402,036 1,261,832 319,903 1.2337 135.54 1983 0 1,030,870 927,783 197,800 1.2324 201.27 1984 18,564,547,348 9,918,988 8,927,089 998,300 1.2951 611.07 1985 19,530,468,891 36,387,694 32,748,925 1,430,482 1.3655 870.05 1986 34,058,854,442 65,949,829 59,354,846 319,101 1.3895 1164.38 1987 45,707,766,058 80,393,569 72,354,212 175,183 1.326 1572.39 1988 65,172,721,789 98,738,058 88,864,252 2,023,483 1.2307 1550.58 1989 90,393,025,287 187,897,196 169,107,476 365,767 1.184 1733.12 1990 61,052,934,824 201,386,259 181,247,633 501,532 1.1668 1609.3 1991 109,501,384,670 278,083,542 250,275,188 2,484,702 1.1457 1602.46 1992 106,859,423,256 427,815,898 385,034,308 1,160,624 1.2087 1781.59 1993 167,189,237,112 450,923,659 405,831,293 4,821,245 1.2901 2180.25 1994 199,252,301,157 699,105,294 629,194,765 679,765 1.3656 2638.24 1995 123,010,986,604 192,709,160 173,438,244 212,020 1.3724 1952.04 1996 134,563,115,856 815,007,795 733,507,016 55,996 1.3635 2044.46 1997 54,222,550,318 0 2050.2 252 University of Ghana http://ugspace.ug.edu.gh Table A 12 continued. Real exchange rate All costs Proportional foreign Total GGDP NARP-CRI Cost NARP-CRI Maize SARI cost Cedis:CN$ (CN$ to cedis) cost (90% of total) maize cost (cedis) (cedis) (15% of NARP-CRI) (cedis) Year (in cedis) (cedis) 1979 342.60 78,231,020 70,407,918 70,758,360 0 0 1980 262.50 59,941,802 53,947,621 54,298,063 0 0 1981 128.61 39,156,723 35,241,051 35,708,307 0 0 1982 109.86 35,146,026 31,631,424 32,893,256 0 0 1983 163.32 32,303,802 29,073,422 30,001,205 0 0 1984 471.83 471,030,176 423,927,158 432,854,248 0 0 1985 637.17 911,454,313 820,308,882 853,057,806 0 0 1986 837.98 267,401,815 240,661,634 300,016,480 0 0 1987 1,185.81 207,734,538 186,961,084 259,315,296 0 0 1988 1,259.92 2,549,420,874 2,294,478,787 2,383,343,039 0 0 1989 1,463.78 535,403,803 481,863,423 650,970,899 0 0 1990 1,379.24 691,734,185 622,560,767 803,808,400 0 0 1991 1,398.67 3,475,286,346 3,127,757,712 3,378,032,899 0 0 1992 1,473.97 1,710,727,320 1,539,654,588 1,924,688,897 0 0 1993 1,689.99 8,147,833,045 7,333,049,740 7,738,881,034 0 0 1994 1,931.93 1,313,256,600 1,181,930,940 1,811,125,705 0 535,742,408 1995 1,422.35 301,567,707 271,410,936 444,849,180 1,361,979,250 204,296,888 597,144,000 1996 1,499.42 83,961,556 75,565,401 809,072,416 1,535,228,229 230,284,234 1,171,434,025 1997 0 1,927,478,424 289,121,764 1,353,509,000 253 University of Ghana http://ugspace.ug.edu.gh Table A 12 continued. SARI maize (8%) GGDP ,SARI and MOFA Extension Actual releases MOFA Extesion cost SG 2000 80% of SG2000 Year NARP-CRI (cedis) estimates (cedis) (80% of estimates) for maize at 50% (in cedis) 1979 0 70,758,360 0 0 0 0 0 1980 0 54,298,063 0 0 0 0 0 1981 0 35,708,307 0 0 0 0 0 1982 0 32,893,256 0 0 0 0 0 1983 0 30,001,205 0 0 0 0 0 1984 0 432,854,248 156,631,040 125,304,832 62,652,416 0 0 1985 0 853,057,806 195,788,800 156,631,040 78,315,520 0 0 1986 0 300,016,480 244,736,000 195,788,800 97,894,400 0 0 1987 0 259,315,296 305,920,000 244,736,000 122,368,000 0 0 1988 0 2,383,343,039 382,400,000 305,920,000 152,960,000] 20,789,424 16,631,539 1989 0 650,970,899 853,829,000 683,063,200 341,531,600 24,458,146 19,566,517 1990 0 803,808,400 857,208,000 685,766,400 342,883,200 28,774,290 23,019,432 1991 0 3,378,032,899 1,997,101,000 1,597,680,800 798,840,400 33,852,105 27,081,684 1992 0 1,924,688,897 1,516,825,000 1,213,460,000 606,730,000 39,826,006 31,860,805 1993 0 7,738,881,034 2,435,152,000 1,948,121,600 974,060,800 46,854,125 37,483,300 1994 42,859,393 1,853,985,098 2,272,267,000 1,817,813,600 908,906,800 55,122,500 44,098,000 1995 47,771,520 696,917,588 2,455,425,000 1,964,340,000 982,170,000 64,850,000 51,880,000 1996 93,714,722 1,133,071,373 533,000,000 426,400,000 213,200,000 84,305,000 67,444,000 1997 108,280,720 397,402,484 829,000,000 663,200,000 331,600,000 96,000,000 76,800,000 254 University of Ghana http://ugspace.ug.edu.gh Table A12 continued. CIMMYT Real Costs Extension + Research CIMMYT Costs CIMMYT Costs (1997 = 1) Total extension Extension + Research (Real cost in cedis) (CN$) (cedis) ( in cedis) Year cost (in cedis) 1979 0 70,758,360 20,045,412,220 50,000 17,129,930 4,852,804,793 1980 0 54,298,063 9,579,608,548 50,000 13,125,214 2,315,633,427 1981 0 35,708,307 3,008,900,708 50,000 6,430,478 541,853,460 1982 0 32,893,256 2,260,739,187 50,000 5,493,232 377,547,429 1983 0 30,001,205 898,206,534 50,000 8,165,774 244,475,234 1984 62,652,416 495,506,664 10,724,083,020 50,000 23,591,615 510,585,329 1985 78,315,520 931,373,326 18,524,110,801 50,000 31,858,294 633,630,517 1986 97,894,400 397,910,880 6,519,584,999 50,000 41,899,244 686,499,663 1987 122,368,000 381,683,296 4,577,130,338 25,000 29,645,362 355,505,958 1988 169,591,539 2,552,934,579 23,514,904,574 25,000 31,497,928 290,125,246 1989 361,098,117 1,012,069,016 7,520,252,230 25,000 36,594,595 271,918,789 1990 365,902,632 1,169,711,032 6,460,712,105 25,000 34,481,059 190,450,625 1991 825,922,084 4,203,954,984 19,797,211,508 25,000 34,966,833 164,665,364 1992 638,590,805 2,563,279,702 10,884,815,147 25,000 36,849,301 156,478,370 1993 1,011,544,100 8,750,425,134 29,950,842,375 25,000 42,249,632 144,611,495 1994 953,004,800 2,806,989,898 7,757,943,721 25,000 48,298,184 133,486,263 1995 1,034,050,000 1,730,967,588 3,048,439,108 25,000 35,558,875 62,623,394 1996 280,644,000 1,413,715,373 1,795,575,275 25,000 37,485,515 47,610,761 1997 408,400,000 805,802,484 805,802,484 0 0 0 255 University of Ghana http://ugspace.ug.edu.gh Table A12 continued. Extension, Research Net social gains and CIMMYT real costs (in cedis) NG = SG - (R + E) Year 1979 24,898,217,013 -24,898,217,013 1980 11,895,241,975 -11,895,241,975 1981 3,550,754,168 -3,550,754,168 1982 2,638,286,616 -2,638,286,616 1983 1,142,681,769 -1,142,681,769 1984 11,234,668,349 7,329,878,999 1985 19,157,741,317 372,727,574 1986 7,206,084,662 26,852,769,780 1987 4,932,636,295 40,775,129,763 1988 23,805,029,820 41,367,691,970 1989 7,792,171,019 82,600,854,267 1990 6,651,162,731 54,401,772,094 1991 19,961,876,873 89,539,507,797 1992 11,041,293,517 95,818,129,739 1993 30,095,453,871 137,093,783,241 1994 7,891,429,984 191,360,871,173 1995 3,111,062,502 119,899,924,102 1996 1,843,186,035 132,719,929,821 1997 805,802,484 53,416,747,834 IRR = 31% Source: Estimated by the author 256 University of Ghana http://ugspace.ug.edu.gh University of Ghana http://ugspace.ug.edu.gh Table A13: Financial Rate of Return with Higher Research and Higher Extension Cost Adoption rate Yield of Yield of Total quantity Total area National average (at K = 95%, improved farmer Change in Proportional Year produced planted production Lag = 30 years) technology technology yield increase Proportional Unit (mt) (ha) (mt/ha) (mt/ha) (mt/ha) (mt/ha) production increase Parameter Q A Y t Yn Yf = Ym dY=Yn-Yf dY/Yf j = t x (dY/Yf) 1979 308600 313900 0.98 0 1980 354000 319900 1.11 0 1981 334200 315500 1.06 0 1982 264300 276300 0.96 0 1983 140800 279800 0.50 0 1984 574000 723600 0.79 0.02 2.88 1.73 1.15 0.6647 0.0133 1985 395000 405000 0.98 0.03 2.75 1.7 1.05 0.6176 0.0185 1986 559100 472100 1.18 0.04 2.93 1.74 1.19 0.6839 0.0274 1987 597700 548336 1.09 0.06 2.93 1.74 1.19 0.6839 0.0410 1988 600000 500000 1.20 0.08 2.93 1.74 1.19 0.6839 0.0547 1989 714614 595800 1.20 0.1 2.93 1.74 1.19 0.6839 0.0684 1990 552549 464784 1.19 0.13 2.93 1.74 1.19 0.6839 0.0889 1991 931478 610445 1.53 0.17 2.93 1.74 1.19 0.6839 0.1163 1992 730617 606783 1.20 0.22 2.93 1.74 1.19 0.6839 0.1505 1993 960920 636670 1.51 0.28 2.93 1.74 1.19 0.6839 0.1915 1994 939908 629401 1.49 0.34 2.93 1.74 1.19 0.6839 0.2325 1995 1034170 668600 1.55 0.41 2.93 1.74 1.19 0.6839 0.2804 1996 1007610 664950 1.52 0.48 2.93 1.74 1.19 0.6839 0.3283 1997 1000000 650000 1.54 0.54 2.93 1.74 1.19 0.6839 0.3693 257 University of Ghana http://ugspace.ug.edu.gh Table A13 continued. Change in Farmers Change in Maize price: Fertilizer prices Change in fertilizer cost row planting Improved seed seed cost Change in total Average wholesale (comp. + S.Ammonia) including application cost seed cost of technology pnce Year (cedis/ha) (cedis/ha) (cedis/ha) (cedis/ha) (cedis/ha) (cedis/ha) (cedis/ha) (cedis/t) dC 1979 1980 1981 1982 1983 0 1984 1,838 2,059 551 400 300 100 2,710 23,380 1985 1,838 2,059 551 1,000 750 250 2,860 20,380 1986 3,175 3,556 953 1,000 750 250 4,759 33,110 1987 5,500 6,160 1,650 2,000 1,500 500 8,310 53,870 1988 9,750 10,920 2,925 2,000 1,500 500 14,345 68,590 1989 14,250 15,960 4,275 3,000 2,250 750 20,985 53,000 1990 18,250 20,440 5,475 3,000 2,250 750 26,665 85,030 1991 23,750 26,600 7,125 5,000 3,750 1,250 34,975 116,050 1992 36,250 40,600 10,875 5,000 3,750 1,250 52,725 100,480 1993 43,750 49,000 13,125 9,000 6,750 2,250 64,375 110,720 1994 58,750 65,800 17,625 9,000 6,750 2,250 85,675 138,630 1995 92,500 103,600 27,750 12,000 9,000 3,000 134,350 200,000 1996 107,500 120,400 32,250 20,000 15,000 5,000 157,650 250,000 1997 132,500 148,400 39,750 30,000 22,250 7,750 195,900 300,000 258 University of Ghana http://ugspace.ug.edu.gh Table A13 continued. Farmgate price Rural price index Real price Real change in cost, Proportional cost Year (70% of wholesale) (1977 = 100) (1997= 1) increase Supply elasticity (cedis/t) (cedis/t) (cedis/ha) (E = 0,79) dC i = l/P = (dC . t)/(Yf. P) 1979 279 0.79 1980 448 0.79 1981 938 0.79 1982 1,150 0.79 1983 0 2,640 0 0.0000 0.79 1984 16,366 3,652 354,204 58,651 0.0019 0.79 1985 14,266 3,974 283,737 56,882 0.0035 0.79 1986 23,177 4,824 379,744 77,966 0.0047 0.79 1987 37,709 6,591 452,205 99,653 0.0076 0.79 1988 48,013 8,581 442,244 132,131 0.0137 0.79 1989 37,100 10,637 275,674 155,931 0.0325 0.79 1990 59,521 14,310 328,755 147,280 0.0335 0.79 1991 81,235 16,784 382,551 164,704 0.0421 0.79 1992 70,336 18,613 298,678 223,894 0.0948 0.79 1993 77,504 23,092 265,280 220,342 0.1337 0.79 1994 97,041 28,598 268,201 236,788 0.1725 0.79 1995 140,000 44,880 246,557 236,606 0.2261 0.79 1996 175,000 62,230 222,269 200,233 0.2485 0.79 1997 210,000 79,039 210,000 195,900 0.2895 0.79 259 University of Ghana http://ugspace.ug.edu.gh Tabie A13 continued. Year Net shift in supply Demand elasticity Change in quantity caused by Social gains from research (e = - 0.981) research, GGDP (all local costs), k = ( i / E) - i dQ = (Q x e x E x k) / (e + E) SG = (k . P , Q) -1/2 (k . P . dQ) (cedis) 1979 0.981 0 389,380 1980 0.981 0 389,380 1981 0,981 0 519,173 1982 0,981 0 1,402,036 1983 0.981 0 0 1,030,870 1984 0.0149 0.981 3,746 3,022,432,946 9,918,988 1985 0.0199 0.981 3,443 2,222,511,149 36,387,694 1986 0.0299 0.981 7,317 6,308,461,916 65,949,829 1987 0.0443 0.981 11,598 11,868,961,201 80,393,569 1988 0.0555 0.981 14,577 14,553,035,743 98,738,058 1989 0.0541 0.981 16,906 10,524,457,658 187,897,196 1990 0.0791 0.981 19,119 14,115,007,059 201,386,259 1991 0.1051 0.981 42,843 36,591,818,286 278,083,542 1992 0.0957 0,981 30,589 20,441,327,147 427,815,898 1993 0.1087 0.981 45,724 27,059,117,346 450,923,659 1994 0.1218 0.981 50,107 29,891,504,647 699,105,294 1995 0.1288 0,981 58,296 31,920,145,086 192,709,160 1996 0.1670 0.981 73,647 36,040,383,882 815,007,795 1997 0.1780 0.981 77,881 35,919,159,696 260 University of Ghana http://ugspace.ug.edu.gh Table A 13 continued. I Proportional GGDP GGDP (foreign cost. Exchange rate Exchange rate, Cedis:CN$ Ail costs Proportional foreign Year maize cost (cedis) Canadian $ (CN$:1US$) cedis: 1US$ (CN$ to cedis) cost (cedis) (100% for 1979-96) (100% for 1979-96) 1979 389,380 228,346 1.1714 2.75 2.35 536,069 536,069 1980 389,380 228,346 1.1692 2.75 2.35 537,078 537,078 1981 519,173 304,462 1.1989 2.75 2.29 698,366 698,366 1982 1,402,036 319,903 1.2337 2.75 2.23 713,085 713,085 1983 1,030,870 197,800 1.2324 8.83 7.16 1,417,214 1,417,214 1984 9,918,988 998,300 1.2951 35.99 27.79 27,742,118 27,742,118 1985 36,387,694 1,430,482 1.3655 54.37 39.82 56,957,383 56,957,383 1986 65,949,829 319,101 1.3895 89.2 64.20 20,484,929 20,484,929 1987 80,393,569 175,183 1.326 153.73 115.94 20,309,866 20,309,866 1988 98,738,058 2,023,483 1.2307 202.35 164.42 332,698,290 332,698,290 1989 187,897,196 365,767 1.184 270 228.04 83,409,704 83,409,704 1990 201,386,259 501,532 1.1668 326,33 279.68 140,268,202 140,268,202 1991 278,083,542 2,484,702 1.1457 367.83 321.05 797,720,116 797,720,116 1992 427,815,898 1,160,624 1.2087 437.09 361.62 419,704,761 419,704,761 1993 450,923,659 4,821,245 1.2901 649.06 503.11 2,425,608,309 2,425,608,309 1994 699,105,294 679,765 1.3656 956.71 700.58 476,228,744 476,228,744 1995 192,709,160 212,020 1.3724 1200.43 874.69 185,452,615 185,452,615 1996 815,007,795 55,996 1.3635 1637.23 1,200.76 67,237,500 67,237,500 1997 0 0 261 University of Ghana http://ugspace.ug.edu.gh Table A 13 continued. r Total GGDP NARP-CRI Cost NARP-CRI Maize SARI cost SARI maize (8%) GGDP ,SARI and MOFA Extension Year maize cost (15% of NARP-CRI) NARP-CRI estimates (cedis) (cedis) (cedis) (cedis) (cedis) (cedis) (cedis) 1979 925,449 0 0 0 0 925,449 0 1980 926,458 0 0 0 0 926,458 0 1981 1,217,539 0 0 0 0 1,217,539 0 1982 2,115,121 0 0 0 0 2,115,121 0 1983 2,448,084 0 0 0 0 2,448,084 0 1984 37,661,106 0 0 0 0 37,661,106 156,631,040 1985 93,345,077 0 0 0 0 93,345,077 195,788,800 1986 86,434,758 0 0 0 0 86,434,758 244,736,000 1987 100,703,435 0 0 0 0 100,703,435 305,920,000 1988 431,436,348 0 0 0 0 431,436,348 382,400,000 1989 271,306,900 0 0 0 0 271,306,900 853,829,000 1990 341,654,461 0 0 0 0 341,654,461 857,208,000 ( 1991 1,075,803,658 0 0 0 0 1,075,803,658 1,997,101,000 1992 847,520,659 0 0 0 0 847,520,659 1,516,825,000 1993 2,876,531,968 0 0 0 0 2,876,531,968 2,435,152,000 1994 1,175,334,038 0 0 535,742,408 42,859,393 1,218,193,431 2,272,267,000 1995 378,161,775 1,361,979,250 204,296,888 597,144,000 47,771,520 630,230,182 2,455,425,000 1996 882,245,295 1,535,228,229 230,284,234 1,171,434,025 93,714,722 1,206,244,251 533,000,000 1997 1,927,478,424 289,121,764 1,353,509,000 108,280,720 397,402,484 829,000,000 262 University of Ghana http://ugspace.ug.edu.gh Table A13 continued. Actual releases MOFA Extesion cost SG 2000 80% of SG 2000 Total extension Extension + Research Extension + Research Year (80% of estimates) for maize at 50% cost (Real cost) (cedis) (cedis) (cedis) (cedis) (cedis) (cedis) (cedis) 1979 0 0 0 0 0 925,449 262,174,129 1980 0 0 0 0 0 926,458 163,451,579 1981 0 0 0 0 0 1,217,539 102,593,851 1982 0 0 0 0 0 2,115,121 145,371,363 1983 0 0 0 0 0 2,448,084 73,293,211 1984 125,304,832 62,652,416 0 0 62,652,416 100,313,522 2,171,051,606 1985 156,631,040 78,315,520 0 0 78,315,520 171,660,597 3,414,162,536 1986 195,788,800 97,894,400 0 0 97,894,400 184,329,158 3,020,147,666 1987 244,736,000 122,368,000 0 0 122,368,000 223,071,435 2,675,063,445 1988 305,920,000 152,960,000 20,789,424 16,631,539 169,591,539 601,027,887 5,536,026,472 1989 683,063,200 341,531,600 24,458,146 19,566,517 361,098,117 632,405,017 4,699,131,349 1990 685,766,400 342,883,200 28,774,290 23,019,432 365,902,632 707,557,092 3,908,078,617 1991 1,597,680,800 798,840,400 33,852,105 27,081,684 825,922,084 1,901,725,742 8,955,582,753 1992 1,213,460,000 606,730,000 39,826,006 31,860,805 638,590,805 1,486,111,464 6,310,684,144 1993 1,948,121,600 974,060,800 46,854,125 37,483,300 1,011,544,100 3,888,076,068 13,308,056,658 1994 1,817,813,600 908,906,800 55,122,500 44,098,000 953,004,800 2,171,198,231 6,000,746,100 1995 1,964,340,000 982,170,000 64,850,000 51,880,000 1,034,050,000 1,664,280,182 2,930,994,682 1996 426,400,000 213,200,000 84,305,000 67,444,000 280,644,000 1,486,888,251 1,888,512,944 1997 663,200,000 331,600,000 96,000,000 76,800,000 408,400,000 805,802,484 805,802,484 263 University of Ghana http://ugspace.ug.edu.gh Table A 13 continued. CIMMYT Costs (CN$) CIMMYT Costs CIMMYT Real Costs Research, extension Net social gains (1997= 1) and CIMMYT real cost NG = SG - ( R + E ) (CN$) (cedis) (cedis) (cedis) Year 1979 50,000 117,381 33,253,297 295,427,426 -295,427,426 1980 50,000 117,602 20,748,051 184,199,630 -184,199,630 1981 50,000 114,688 9,664,031 112,257,882 -112,257,882 1982 50,000 111,453 7,660,140 153,031,503 -153,031,503 1983 50,000 358,244 10,725,475 84,018,686 -84,018,686 1984 50,000 1,389,468 30,071,786 2,201,123,391 821,309,555 1985 50,000 1,990,846 39,595,990 3,453,758,526 -1,231,247,377 1986 50,000 3,209,788 52,590,881 3,072,738,547 3,235,723,370 1987 25,000 2,898,379 34,757,236 2,709,820,682 9,159,140,519 1988 25,000 4,110,466 37,861,215 5,573,887,688 8,979,148,055 1989 25,000 5,701,014 42,361,794 4,741,493,143 5,782,964,515 1990 25,000 6,991,987 38,619,122 3,946,697,739 10,168,309,321 1991 25,000 8,026,316 37,797,425 8,993,380,177 27,598,438,108 1992 25,000 9,040,498 38,389,939 6,349,074,083 14,092,253,064 1993 25,000 12,577,707 43,050,814 13,351,107,472 13,708,009,873 1994 25,000 17,514,463 48,406,378 6,049,152,478 23,842,352,169 1995 25,000 21,867,349 38,510,994 2,969,505,676 28,950,639,410 1996 25,000 30,018,885 38,127,313 1,926,640,256 34,113,743,625 1997 0 0 0 805,802,484 35,113,357,213 IRR = 68% Source: Estimated by the author 264 University of Ghana http://ugspace.ug.edu.gh Table A.14:Economic Rate of Return with Higher Research and Higher Extension Cost Yield of Yield of Change in Proportional Proportional Year Total quantity Total area National average Adoption rate Improved Farmer yield increase production increase produced planted production (at K = 95%, technology technology G-parameter) Lag = 30 years) (mt/ha) (mt/ha) (mt/ha) Unit (mt) (ha) (mt/ha) Parameter Q A Ym t Yn Yf = Ym dY=Yn-Yf dY/Yf j = t x (dY/Yf) 1979 308,600 313,900 0.9831 0 0.0000 1980 354,000 319,900 1.1066 0 0.0000 1981 334,200 315,500 1.0593 0 0.0000 1982 264,300 276,300 0.9566 0 0.0000 1983 140,800 279,800 0.5032 0 0.0000 1984 574,000 723,600 0.7933 0.02 2.88 1.73 1.15 0.6647 0.0133 1985 395,000 405,000 0.9753 0.03 2.75 1.7 1.05 0.6176 0.0185 1986 559,100 472,100 1.1843 0.04 2.93 1.74 1.19 0.6839 0.0274 1987 597,700 548,336 1.0900 0.06 2.93 1.74 1.19 0.6839 0.0410 1988 600,000 500,000 1.2000 0.08 2.93 1.74 1.19 0.6839 0.0547 1989 714,614 595,800 1.1994 0.1 2.93 1.74 1.19 0.6839 0.0684 1990 552,549 464,784 1.1888 0.13 2.93 1.74 1.19 0.6839 0.0889 1991 931,478 610,445 1.5259 0.17 2.93 1.74 1.19 0.6839 0.1163 1992 730,617 606,783 1.2041 0.22 2.93 1.74 1.19 0.6839 0.1505 1993 960,920 636,670 1.5093 0.28 2.93 1.74 1.19 0.6839 0.1915 1994 939,908 629,401 1.4933 0.34 2.93 1.74 1.19 0.6839 0.2325 1995 1,034,170 668,600 1,5468 0.41 2.93 1.74 1.19 0.6839 0.2804 1996 1,007,610 664,950 1.5153 0.48 2.93 1.74 1.19 0.6839 0.3283 1997 1,000,000 650,000 1.5385 0.54 2.93 1.74 1.19 0.6839 0.3693 265 University of Ghana http://ugspace.ug.edu.gh Table A14 continued Fertilizer prices (comp. + S.A, Product price: Year (Without subsidy) Change in fertilizer costs Change in Improved Farmers Change in Change in total Export price (F.O.B) cedis/ha) including application row planting seed seed seed cost cost of technology U.S. Gulf ($/ton) (cedis/ha) cost (cedis/ha) (cedis/ha) (cedis/ha) (cedis/ha) ($/ton) (cedis/ha) dC 1979 1980 1981 1982 1983 0 1984 2,665 2,985 800 400 300 100 3,884 139.1 1985 2,922 3,273 877 1,000 750 250 4,399 116.8 1986 4,318 4,836 1,295 1,000 750 250 6,382 88.7 1987 7,810 8,747 2,343 2,000 1,500 500 11,590 75.6 1988 12,675 14,196 3,803 2,000 1,500 500 18,499 107 1989 16,388 18,355 4,916 3,000 2,250 750 24,021 111.3 1990 18,250 20,440 5,475 3,000 2,250 750 26,665 109.7 1991 23,750 26,600 7,125 5,000 3,750 1,250 34,975 107 1992 36,250 40,600 10,875 5,000 3,750 1,250 52,725 105.1 1993 43,750 49,000 13,125 9,000 6,750 2,250 64,375 101.9 1994 58,750 65,800 17,625 9,000 6,750 2,250 85,675 107.5 1995 92,500 103,600 27,750 12,000 9,000 3,000 134,350 123.7 1996 107,500 120,400 32,250 20,000 15,000 5,000 157,650 165.1 1997 132,500 148,400 39,750 30,000 22,250 7,750 195,900 117.1 266 University of Ghana http://ugspace.ug.edu.gh Table A 14 continued Freight and marketing cost No.2 yellow maize Real (30% + 20% = 50% Value of maize exchange rate Value of maize Rural price index Real price Proportional cost increase Year of export price) (1997 = 100) (Farmgate) Real change ($/ton) ($/ton) (cedis: 1 US$) (cedis/ton) (1997 = 1) in cost i = l/P = (dC. t)/(Yf. P) (cedis/ton) (cedis/ton) dC 1979 279 1980 448 1981 938 1982 1,150 1983 0 2,640 1984 69.55 208.65 611.07 127,500 3,652 1,931,604 84,067 0.0000 1985 58.4 175.2 870.05 152,433 3,974 2,122,218 87,497 0.0000 1986 44.35 133.05 1164.38 154,921 4,824 1,776,813 104,559 0.0001 1987 37.8 113.4 1572.39 178,309 6,591 1,496,792 138,989 0.0003 1988 53.5 160.5 1550.58 248,868 8,581 1,604,615 170,388 0.0005 1989 55.65 166.95 1733.12 289,344 10,637 1,504,996 178,489 0.0009 1990 54.85 164.55 1609.3 264,810 14,310 1,023,846 147,280 0.0019 1991 53.5 160.5 1602.46 257,195 16,784 847,825 164,704 0.0040 1992 52.55 157.65 1781.59 280,868 18,613 834,882 223,894 0.0080 1993 50.95 152.85 2180.25 333,251 23,092 798,454 220,342 0.0130 1994 53.75 161.25 2638.24 425,416 28,598 823,034 236,788 0.0203 1995 61.85 185.55 1952.04 362,201 44,880 446,515 236,606 0.0709 1996 82.55 247,65 2044.46 506,311 62,230 450,149 200,233 0.0966 1997 58.55 175.65 2050.2 360,118 79,039 252,082 195,900 0.2412 267 University of Ghana http://ugspace.ug.edu.gh Table A14 continued Research cost: Net shift in supply Change in quantity caused GGDP Year by research. Social gains from research (all local cost) Supply elasticity k = ( i / E) - i Demand elasticity (in cedis) ( E = 0.79) (e = - 0.981) dQ = (Q x e x E x k) / (e + E) SG = (k . P . Q) -1/2 (k . P . dQ) dQ 1979 0.79 0.981 389,380 1980 0.79 0.981 389,380 1981 0.79 0.981 519,173 1982 0.79 0.981 1,402,036 1983 0.79 0.981 1,030,870 1984 0.79 0.0168 0.981 4,221.3 18,564,547,348 9,918,988 1985 0.79 0.0234 0.981 4,047.9 19,530,468,891 36,387,694 1986 0.79 0.0345 0.981 8,452.0 34,058,854,442 65,949,829 1987 0.79 0.0517 0.981 13,515.9 45,707,766,058 80,393,569 1988 0.79 0.0687 0.981 18,044.8 65,172,721,789 98,738,058 1989 0.79 0.0857 0.981 26,785.1 90,393,025,287 187,897,196 1990 0.79 0.1106 0.981 26,741.6 61,052,934,824 201,386,259 1991 0.79 0.1431 0.981 58,345.9 109,501,384,670 278,083,542 1992 0.79 0.1825 0.981 58,339.2 106,859,423,256 427,815,898 1993 0.79 0.2294 0.981 96,472.4 167,189,237,112 450,923,659 1994 0.79 0.2740 0.981 112,697.1 199,252,301,157 699,105,294 1995 0.79 0.2840 0.981 128,543.7 123,010,986,604 192,709,160 1996 0.79 0.3189 0.981 140,624.8 134,563,115,856 815,007,795 1997 0.79 0.2263 0.981 99,030.7 54,222,550,318 268 University of Ghana http://ugspace.ug.edu.gh Table A 14 continued Real Real Proportional GGDP GGDP (foreign costs), Exchange rate Exchange rate, Exchange rate All costs Proportional foreign maize cost (cedis) _(CN$:1US$) Cedis: 1US$ Cedis:CN$ (CN$ to cedis) cost (cedis) (90% for 1979-83, (Canadian $) (100% for 1979-96) Year 55% for 1984-96) 1979 389,380 228,346 1.1714 401.32 342.60 78,231,020 78,231,020 1960 389,380 228,346 1.1692 306.92 262,50 59,941,802 59,941,802 1981 519,173 304,462 1.1989 154.19 128.61 39,156,723 39,156,723 1982 1,402,036 319,903 1.2337 135.54 109.86 35,146,026 35,146,026 1983 1,030,870 197,800 1.2324 201.27 163.32 32,303,802 32,303,802 1984 9,918,988 998,300 1.2951 611.07 471.83 471,030,176 471,030,176 1985 36,387,694 1,430,482 1.3655 870.05 637.17 911,454,313 911,454,313 1986 65,949,829 319,101 1.3895 1164.38 837.98 267,401,815 267,401,815 1987 80,393,569 175,183 1.326 1572.39 1,185.81 207,734,538 207,734,538 1988 98,738,058 2,023,483 1.2307 1550.58 1,259.92 2,549,420,874 2,549,420,874 1989 187,897,196 365,767 1.184 1733.12 1,463.78 535,403,803 535,403,803 1990 201,386,259 501,532 1.1668 1609.3 1,379.24 691,734,185 691,734,185 1991 278,083,542 2,484,702 1.1457 1602.46 1,398.67 3,475,286,346 3,475,286,346 1992 427,815,898 1,160,624 1.2087 1781.59 1,473.97 1,710,727,320 1,710,727,320 1993 450,923,659 4,821,245 1.2901 2180.25 1,689.99 8,147,833,045 8,147,833,045 1994 699,105,294 679,765 1.3656 2638.24 1,931.93 1,313,256,600 1,313,256,600 1995 192,709,160 212,020 1.3724 1952.04 1,422.35 301,567,707 301,567,707 1996 815,007,795 55,996 1.3635 2044.46 1,499.42 83,961,556 83,961,556 1997 0 2050.2 0 269 University of Ghana http://ugspace.ug.edu.gh Table A 14 continued Total GGDP NARP-CRI Cost NARP-CRI Maize SARI cost SARI maize (8%) GGDPjSARI and MOFA Extension maize cost (15% of NARP-CRI) NARP-CRI estimates (cedis) (cedis) (cedis) (cedis) (cedis) (cedis) (cedis) Year 1979 78,620,400 0 0 0 0 78,620,400 0 1980 60,331,182 0 0 0 0 60,331,182 0 1981 39,675,896 0 0 0 0 39,675,896 0 1982 36,548,062 0 0 0 0 36,548,062 0 1983 33,334,672 0 0 0 0 33,334,672 0 1984 480,949,164 0 0 0 0 480,949,164 156,631,040 1985 947,842,007 0 0 0 0 947,842,007 195,788,800 1986 333,351,644 0 0 0 0 333,351,644 244,736,000 1987 288,128,107 0 0 0 0 288,128,107 305,920,000 1988 2,648,158,932 0 0 0 0 2,648,158,932 382,400,000 1989 723,300,999 0 0 0 0 723,300,999 853,829,000 1990 893,120,444 0 0 0 0 893,120,444 857,208,000 1991 3,753,369,888 0 0 0 0 3,753,369,888 1,997,101,000 1992 2,138,543,218 0 0 0 0 2,138,543,218 1,516,825,000 1993 8,598,756,704 0 0 0 0 8,598,756,704 2,435,152,000 1994 2,012,361,894 0 0 535,742,408 42,859,393 2,055,221,287 2,272,267,000 1995 494,276,867 1,361,979,250 204,296,888 597,144,000 47,771,520 746,345,274 2,455,425,000 1996 898,969,351 1,535,228,229 230,284,234 1,171,434,025 93,714,722 1,222,968,308 533,000,000 1997 1,927,478,424 289,121,764 1,353,509,000 108,280,720 397,402,484 829,000,000 270 University of Ghana http://ugspace.ug.edu.gh Table A 14 continued Extension + Research Actual releases MOFA Extesion cost SG 2000 80% of SG2000 Total extension Extension + Research (Real cost) (80% of estimates) for maize at 50% (cedis) (cedis) (cedis) (cedis) (cedis) (cedis) Year 1979 0 0 0 0 0 78,620,400 22,272,680,245 1980 0 0 0 0 0 60,331,182 10,644,009,498 1981 0 0 0 0 0 39,675,896 3,343,223,008 1982 0 0 0 0 0 36,548,062 2,511,932,430 1983 0 0 0 0 0 33,334,672 998,007,260 1984 125,304,832 62,652,416 0 0 62,652,416 543,601,580 11,764,985,018 1985 156,631,040 78,315,520 0 0j 78,315,520 1,026,157,527 20,409,276,487 1986 195,788,800 97,894,400 0 0 97,894,400 431,246,044 7,065,766,189 1987 244,736,000 122,368,000 0 0 122,368,000 410,496,107 4,922,652,374 1988 305,920,000 152,960,000 20,789,424 16,631,539 169,591,539 2,817,750,472 25,954,105,528 1989 683,063,200 341,531,600 24,458,146 19,566,517 361,098,117 1,084,399,116 8,057,706,284 1990 685,766,400 342,883,200 28,774,290 23,019,432 365,902,632 1,259,023,076 6,954,012,922 1991 1,597,680,800 798,840,400 33,852,105 27,081,684 825,922,084 4,579,291,973 21,564,743,697 1992 1,213,460,000 606,730,000 39,826,006 31,860,805 638,590,805 2,777,134,023 11,792,934,835 1993 1,948,121,600 974,060,800 46,854,125 37,483,300 1,011,544,100 9,610,300,804 32,894,013,738 1994 1,817,813,600 908,906,800 55,122,500 44,098,000 953,004,800 3,008,226,087 8,314,119,229 1995 1,964,340,000 982,170,000 64,850,000 51,880,000 1,034,050,000 1,780,395,274 3,135,487,123 1996 426,400,000 213,200,000 84,305,000 67,444,000 280,644,000 1,503,612,308 1,909,754,350 1997 663,200,000 331,600,000 96,000,000 76,800,000 408,400,000 805,802,484 805,802,484 271 University of Ghana http://ugspace.ug.edu.gh Table A14 continued Extension , Research NET SOCIAL GAIN CIMMYT Costs CIMMYT Costs CIMMYT Real Costs and CIMMYT real costs 1997= 1) (in cedis) NG = SG - (R + E) (CN$) (cedis) (cedis) Year 1979 50,000 17,129,930 4,852,804,793 27,125,485,037 -27,125,485,037 1980 50,000 13,125,214 2,315,633,427 12,959,642,925 -12,959,642,925 1981 50,000 6,430,478 541,853,460 3,885,076,469 -3,885,076,469 1982 50,000 5,493,232 377,547,429 2,889,479,859 -2,889,479,859 1983 50,000 8,165,774 244,475,234 1,242,482,495 -1,242,482,495 1984 50,000 23,591,615 510,585,329 12,275,570,347 6,288,977,001 1985 50,000 31,858,294 633,630,517 21,042,907,004 -1,512,438,112 1986 50,000 41,899,244 686,499,663 7,752,265,852 26,306,588,590 1987 25,000 29,645,362 355,505,958 5,278,158,332 40,429,607,726 1988 25,000 31,497,928 290,125,246 26,244,230,774 38,928,491,016 1989 25,000 36,594,595 271,918,789 8,329,625,073 82,063,400,214 1990 25,000 34,481,059 190,450,625 7,144,463,547 53,908,471,278 1991 25,000 34,966,833 164,665,364 21,729,409,062 87,771,975,608 1992 25,000 36,849,301 156,478,370 11,949,413,204 94,910,010,052 1993 25,000 42,249,632 144,611,495 33,038,625,233 134,150,611,878 1994 25,000 48,298,184 133,486,263 8,447,605,492 190,804,695,665 1995 25,000 35,558,875 62,623,394 3,198,110,517 119,812,876,087 1996 25,000 37,485,515 47,610,761 1,957,365,111 132,605,750,745 1997 0 0 0 805,802,484 53,416,747,834 IRR = 30% Source: Estimated by the author. 272 University of Ghana http://ugspace.ug.edu.gh Table A15: Financial Rate of Return with Lower Projected Research Benefit Adoption rate Yield of Yield of Change in Proportional Total quantity produ Total area planted National average (at K = 95%, improved farmer yield increase Year production Lag = 30 years) technology technology (reduced by 40%) Unit (mt) (ha) (t/ha) (t/ha) (t/ha) (Mia) Parameter Q A Ym t Yn Yf = Ym dY=Yn-Yf dY/Yf 1979 308600 313900 0.98 0 1980 354000 319900 1.11 0 1981 334200 315500 1.06 0 1982 264300 276300 0.96 0 1983 140800 279800 0.50 0 1984 574000 723600 0.79 0.02 2.88 1.73 0.69 0.3988 1985 395000 405000 0.98 0.03 2.75 1.7 0.63 0.3706 1986 559100 472100 1.18 0.04 2.93 1.74 0.714 0.4103 1987 597700 548336 1.09 0.06 2.93 1.74 0.714 0.4103 1988 600000 500000 1.20 0.08 2.93 1.74 0.714 0.4103 1989 714614 595800 1.20 0.1 2.93 1.74 0.714 0.4103 1990 552549 464784 1.19 0.13 2.93 1.74 0.714 0.4103 1991 931478 610445 1.53 0.17 2.93 1.74 0.714 0.4103 1992 730617 606783 1.20 0.22 2.93 1.74 0.714 0.4103 1993 960920 636670 1.51 0.28 2.93 1.74 0.714 0.4103 1994 939908 629401 1.49 0.34 2.93 1.74 0.714 0.4103 1995 1034170 668600 1.55 0.41 2.93 1.74 0.714 0.4103 1996 1007610 664950 1.52 0.48 2.93 1.74 0.714 0.4103 1997 1000000 650000 1.54 0.54 2.93 1.74 0.714 0.4103 273 University of Ghana http://ugspace.ug.edu.gh Table A 15 continued. Change in Change in Improved Farmers Change in Change in total Proportional Fertilizer prices fertilizer cost row planting seed seed seed cost cost of technology production (comp. + S.Ammonia) including application cost Year increase (cedis/ha) (cedis/ha) (cedis/ha) (cedis/ha) (cedis/ha) (cedis/ha) (cedis/ha) j = t x (dY/Yf) dC 1979 1980 1981 1982 1983 1984 0.0080 1,838 2,059 551 400 300 100 2,710 1985 0.0111 1,838 2,059 551 1,000 750 250 2,860 1986 0.0164 3,175 3,556 953 1,000 750 250 4,759 1987 0.0246 5,500 6,160 1,650 2,000 1,500 500 8,310 1988 0.0328 9,750 10,920 2,925 2,000 1,500 500 14,345 1989 0.0410 14,250 15,960 4,275 3,000 2,250 750 20,985 1990 0.0533 18,250 20,440 5,475 3,000 2,250 750 26,665 1991 0.0698 23,750 26,600 7,125 5,000 3,750 1,250 34,975 1992 0.0903 36,250 40,600 10,875 5,000 3,750 1,250 52,725 1993 0.1149 43,750 49,000 13,125 9,000 6,750 2,250 64,375 1994 0.1395 58,750 65,800 17,625 9,000 6,750 2,250 85,675 1995 0.1682 92,500 103,600 27,750 12,000 9,000 3,000 134,350 1996 0.1970 107,500 120,400 32,250 20,000 15,000 5,000 157,650 1997 0.2216 132,500 148,400 39,750 30,000 22,250 7,750 195,900 274 University of Ghana http://ugspace.ug.edu.gh Table A15 continued. Maize price: Average wholesale Farmgate price Rural price index Real price Real change in Proportional cost Year price (70% of wholesale) (1977= 100) (1997= 1) cost increase Supply elasticity (cedis/mt) (cedis/mt) (cedis) (cedis/ha) (E = 0.79) dC i = l/P = (dC . t)/(Yf. P) 1979 279 0.79 1980 448 0.79 1981 938 0.79 1982 1,150 0.79 1983 0 0 2,640 0 0.0000 0.79 1984 23,380 16,366 3,652 278,876 46,178 0.0019 0.79 1985 20,380 14,266 3,974 223,395 44,785 0.0035 0.79 1986 33,110 23,177 4,824 298,985 61,385 0.0047 0.79 1987 53,870 37,709 6,591 356,036 78,460 0.0076 0.79 1988 68,590 48,013 8,581 348,194 104,031 0.0137 0.79 1989 53,000 37,100 10,637 217,047 122,769 0.0325 0.79 1990 85,030 59,521 14,310 258,839 115,958 0.0335 0.79 1991 116,050 81,235 16,784 301,195 129,677 0.0421 0.79 1992 100,480 70,336 18,613 235,159 176,279 0.0948 0.79 1993 110,720 77,504 23,092 208,863 173,482 0.1337 0.79 1994 138,630 97,041 28,598 211,164 186,431 0.1725 0.79 1995 200,000 140,000 44,880 194,122 186,288 0.2261 0.79 1996 250,000 175,000 62,230 175,000 157,650 0.2485 0.79 1997 300,000 210,000 79,039 165,340 154,239 0.2895 0.79 275 University of Ghana http://ugspace.ug.edu.gh Table A15 continued. Social gains from research Year Change in quantity caused by GGDP Proportional GGDP Net shift in Demand elasticity research, (all local cost) maize cost (cedis) supply (e = - 0.981) dQ (cedis) (90% for 1979-83, k = ( i / E) - i dQ = (Q x e x E x k) / (e + E) SG = (k . P . Q) -1/2 (k . P . dQ) 55% for 1984-96) 1979 0.981 0 389,380 350,442 1980 0.981 0 389,380 350,442 1981 0.981 0 519,173 467,256 1982 0.981 0 1,402,036 1,261,832 1983 0.981 0 0 1,030,870 927,783 1984 0.0082 0.981 2,055 1,307,554,723 9,918,988 5,455,443 1985 0.0105 0.981 1,821 927,495,195 36,387,694 20,013,232 1986 0.0161 0.981 3,929 2,674,725,832 65,949,829 36,272,406 1987 0.0236 0.981 6,164 4,989,130,832 80,393,569 44,216,463 1988 0,0278 0.981 7,304 5,776,092,252 98,738,058 54,305,932 1989 0.0194 0.981 6,078 3,001,605,119 187,897,196 103,343,458 1990 0.0341 0.981 8,234 4,834,211,445 201,386,259 110,762,442 1991 0.0462 0.981 18,847 12,841,042,255 278,083,542 152,945,948 1992 0.0195 0.981 6,233 3,335,029,962 427,815,898 235,298,744 1993 0.0118 0.981 4,953 2,357,871,674 450,923,659 248,008,012 1994 0.0041 0.981 1,682 810,726,281 699,105,294 384,507,912 1995 -0.0132 0.981 -5,955 -2,649,338,607 192,709,160 105,990,038 1996 0.0008 0.981 358 143,001,579 815,007,795 448,254,287 1997 -0.0090 0.981 -3,947 -1,494,063,460 276 University of Ghana http://ugspace.ug.edu.gh Table A 15 continued. GGDP (foreign costs), Exchange rate Exchange rate, Exchange rate All costs Proportional foreign Total GGDP Year Canadian $ (CN$:1US$) cedis: 1US$ Cedis: CN$ (CN$ to cedis) cost (cedis) maize cost (cedis) (90% for 1979-83, 55% for 1984-96) 1979 228,346 1.1714 2.75 2.35 536,069 482,462 832,904 1980 228,346 1.1692 2.75 2.35 537,078 483,370 833,812 1981 304,462 1.1989 2.75 2.29 698,366 628,529 1,095,785 1982 319,903 1.2337 2.75 2.23 713,085 641,777 1,903,609 1983 197,800 1.2324 8.83 7.16 1,417,214 1,275,492 2,203,275 1984 998,300 1.2951 35.99 27.79 27,742,118 15,258,165 20,713,608 1985 1,430,482 1.3655 54.37 39.82 56,957,383 31,326,561 51,339,792 1986 319,101 1.3895 89.2 64.20 20,484,929 11,266,711 47,539,117 1987 175,183 1.326 153.73 115.94 20,309,866 11,170,426 55,386,889 1988 2,023,483 1.2307 202.35 164.42 332,698,290 182,984,059 237,289,991 1989 365,767 1.184 270 228.04 83,409,704 45,875,337 149,218,795 1990 501,532 1.1668 326.33 279.68 140,268,202 77,147,511 187,909,953 1991 2,484,702 1.1457 367.83 321.05 797,720,116 438,746,064 591,692,012 1992 1,160,624 1.2087 437.09 361.62 419,704,761 230,837,618 466,136,362 1993 4,821,245 1.2901 649.06 503.11 2,425,608,309 1,334,084,570 1,582,092,583 1994 679,765 1.3656 956.71 700.58 476,228,744 261,925,809 646,433,721 1995 212,020 1.3724 1200.43 874.69 185,452,615 101,998,938 207,988,976 1996 55,996 1.3635 1637.23 1,200.76 67,237,500 36,980,625 485,234,912 1997 0 0 277 University of Ghana http://ugspace.ug.edu.gh Table A15 continued. Year NARP-CRI Cost NARP-CRI Maize SARI cost SARI maize (8%) GGDP ,SARI and MOFA Extension Actual releases (cedis) (15% of NARP-CRI) (cedis) NARP-CRI (cedis) estimates (cedis) (80% of estimates) (cedis) (cedis) 1979 0 0 0 0 832,904 0 0 1980 0 0 0 0 833,812 0 0 1981 0 0 0 0 1,095,785 0 0 1982 0 0 0 0 1,903,609 0 0 1983 0 0 0 0 2,203,275 0 0 1984 0 0 0 0 20,713,608 156,631,040 125,304,832 1985 0 0 0 0 51,339,792 195,788,800 156,631,040 1986 0 0 0 0 47,539,117 244,736,000 195,788,800 1987 0 0 0 0 55,386,889 305,920,000 244,736,000 1988 0 0 0 0 237,289,991 382,400,000 305,920,000 1989 0 0 0 0 149,218,795 853,829,000 683,063,200 1990 0 0 0 0 187,909,953 857,208,000 685,766,400 1991 0 0 0 0 591,692,012 1,997,101,000 1,597,680,800 1992 0 0 0 0 466,136,362 1,516,825,000 1,213,460,000 1993 0 0 0 0 1,582,092,583 2,435,152,000 1,948,121,600 1994 0 0 535,742,408 42,859,393 689,293,114 2,272,267,000 1,817,813,600 1995 1,361,979,250 204,296,888 597,144,000 47,771,520 460,057,384 2,455,425,000 1,964,340,000 1996 1,535,228,229 230,284,234 1,171,434,025 93,714,722 809,233,869 533,000,000 426,400,000 1997 1,927,478,424 289,121,764 1,353,509,000 108,280,720 397,402,484 829,000,000 663,200,000 278 University of Ghana http://ugspace.ug.edu.gh Table A15 continued. Extension + Research Year MOFA Extesion cost SG 2000 80% of SG2000 Total extension Extension + Research (Real cost) for maize at 30% (cedis) (cedis) (cedis) (cedis) (cedis) (cedis) 1979 0 0 0 0 832,904 185,776,470 1980 0 0 0 0 833,812 115,821,715 1981 0 0 0 0 1,095,785 72,697,957 1982 0 0 0 0 1,903,609 103,010,082 1983 0 0 0 0 2,203,275 51,935,536 1984 37,591,450 0 0 37,591,450 58,305,058 993,516,909 1985 46,989,312 0 0 46,989,312 98,329,104 1,539,763,503 1986 58,736,640 0 0 58,736,640 106,275,757 1,370,966,078 1987 73,420,800 0 0 73,420,800 128,807,689 1,216,158,779 1988 91,776,000 20,789,424 16,631,539 108,407,539 345,697,531 2,507,022,180 1989 204,918,960 24,458,146 19,566,517 224,485,477 373,704,272 2,186,294,712 1990 205,729,920 28,774,290 23,019,432 228,749,352 416,659,305 1,811,929,318 1991 479,304,240 33,852,105 27,081,684 506,385,924 1,098,077,936 4,071,341,156 1992 364,038,000 39,826,006 31,860,805 395,898,805 862,035,167 2,882,095,764 1993 584,436,480 46,854,125 37,483,300 621,919,780 2,204,012,363 5,939,532,709 1994 545,344,080 55,122,500 44,098,000 589,442,080 1,278,735,194 2,782,561,406 1995 589,302,000 64,850,000 51,880,000 641,182,000 1,101,239,384 1,526,963,611 1996 127,920,000 84,305,000 67,444,000 195,364,000 1,004,597,869 1,004,597,869 1997 198,960,000 96,000,000 76,800,000 275,760,000 673,162,484 530,002,927 279 University of Ghana http://ugspace.ug.edu.gh Table A15 continued. Year CIMMYT Costs CIMMYT Real Costs Research, extension Net social gains CIMMYT Costs (in cedis) (1997= 1) and CIMMYT real costs (CN$) (cedis) (cedis) NG = SG - ( R + E ) 1979 50,000 117,381 26,181,413 211,957,883 -211,957,883 1980 50,000 117,602 16,335,622 132,157,338 -132,157,338 1981 50,000 114,688 7,608,809 80,306,766 -80,306,766 1982 50,000 111,453 6,031,080 109,041,163 -109,041,163 1983 50,000 358,244 8,444,519 60,380,055 -60,380,055 1984 50,000 1,389,468 23,676,504 1,017,193,413 290,361,309 1985 50,000 1,990,846 31,175,223 1,570,938,726 -643,443,531 1986 50,000 3,209,788 41,406,527 1,412,372,605 1,262,353,226 1987 25,000 2,898,379 27,365,513 1,243,524,292 3,745,606,540 1988 25,000 4,110,466 29,809,378 2,536,831,558 3,239,260,694 1989 25,000 5,701,014 33,352,832 2,219,647,544 781,957,575 1990 25,000 6,991,987 30,406,103 1,842,335,421 2,991,876,024 1991 25,000 8,026,316 29,759,153 4,101,100,309 8,739,941,946 1992 25,000 9,040,498 30,225,659 2,912,321,423 422,708,539 1993 25,000 12,577,707 33,895,319 5,973,428,029 -3,615,556,355 1994 25,000 17,514,463 38,111,931 2,820,673,337 -2,009,947,056 1995 25,000 21,867,349 30,320,970 1,557,284,581 -4,206,623,187 1996 25,000 30,018,885 30,018,885 1,034,616,754 -891,615,174 1997 0 0 0 530,002,927 -2,024,066,387 IRR = 50% Source: Estimated by the author. 280 University of Ghana http://ugspace.ug.edu.gh Table A16: Economic Rate of Return with Lower Projected Research Benefit Yield of Yield of Change in Proportional Year Total quantity Total area National average Adoption rate Improved Farmer yield (and increase produced planted production (at K = 95%, technology technology 40% reduction) Proportional Lag = 30 years) (t/ha) (t/ha) (t/ha) production increase Unit <»> - (ha) (Mia) Parameter Q A Ym t Yn 3 ii 3 dY=Yn-Yf dY/Yf j = t x (dY/Yf) 1979 308,600 313,900 0.98 0 0.0000 1980 354,000 319,900 1.11 0 0.0000 1981 334,200 315,500 1.06 0 0.0000 1982 264,300 276,300 0.96 0 0.0000 1983 140,800 279,800 0.50 0 0.0000 1984 574,000 723,600 0.79 0.02 2.88 1.73 0.69 0.3988 0.0080 1985 395,000 405,000 0.98 0.03 2.75 1.7 0.63 0.3706 0.0111 1986 559,100 472,100 1.18 0.04 2.93 1.74 0.714 0.4103 0.0164 1987 597,700 548,336 1.09 0.06 2.93 1.74 0.714 0.4103 0.0246 1988 600,000 500,000 1.20 0.08 2.93 1.74 0.714 0.4103 0.0328 1989 714,614 595,800 1.20 0.1 2.93 1.74 0.714 0.4103 0.0410 1990 552,549 464,784 1.19 0.13 2.93 1.74 0.714 0.4103 0.0533 1991 931,478 610,445 1.53 0.17 2.93 1.74 0.714 0.4103 0.0698 1992 730,617 606,783 1.20 0.22 2.93 1.74 0.714 0.4103 0.0903 1993 960,920 636,670 1.51 0.28 2.93 1.74 0.714 0.4103 0.1149 1994 939,908 629,401 1.49 0.34 2.93 1.74 0.714 0.4103 0.1395 1995 1,034,170 668,600 1.55 0.41 2.93 1.74 0.714 0.4103 0.1682 1996 1,007,610 664,950 1.52 0.48 2.93 1.74 0.714 0.4103 0.1970 1997 1,000,000 650,000 1.54 0.54 2.93 1.74 0.714 0.4103 0.2216 281 University of Ghana http://ugspace.ug.edu.gh Table A 16 continued. Change in fertilizer costs Change in Improved Farmers Change in Change in total cost of Fertilizer prices including application row planting seed seed seed cost technology (comp. + S.A, (cedis/ha) cost (cedis/ha) (cedis/ha) (cedis/ha) (cedis/ha) Year cedis/ha) (cedis/ha) (Without subsidy) dC 1979 1980 1981 1982 1983 1984 2,665 2,985 800 400 300 100 3,884 1985 2,922 3,273 877 1,000 750 250 4,399 1986 4,318 4,836 1,295 1,000 750 250 6,382 1987 7,810 8,747 2,343 2,000 1,500 500 11,590 1988 12,675 14,196 3,803 2,000 1,500 500 18,499 1989 16,388 18,355 4,916 3,000 2,250 750 24,021 1990 18,250 20,440 5,475 3,000 2,250 750 26,665 1991 23,750 26,600 7,125 5,000 3,750 1,250 34,975 1992 36,250 40,600 10,875 5,000 3,750 1,250 52,725 1993 43,750 49,000 13,125 9,000 6,750 2,250 64,375 1994 58,750 65,800 17,625 9,000 6,750 2,250 85,675 1995 92,500 103,600 27,750 12,000 9,000 3,000 134,350 1996 107,500 120,400 32,250 20,000 15,000 5,000 157,650 1997 132,500 148,400 39,750 30,000 22,250 7,750 195,900 282 University of Ghana http://ugspace.ug.edu.gh Table A 16 continued. Freight and Product price: marketing cost Real Export price (F.O.B) No.2 yellow maize Value of maize Exchange rate Value of maize Rural price inde) Real price Year U.S. Gulf ($/ton) (30% + 20% = 50% ($/ton) (cedis: 1 US$) (cedis/ton) (1997= 100) (Farmgate) of export price) (1997= 1) ($/ton) cedis/ton 1979 279 1980 448 1981 938 1982 1,150 1983 2,640 1984 139.1 69.55 208.65 611.07 127,500 3,652 1,931,604 1985 116.8 58.4 175.2 870.05 152,433 3,974 2,122,218 1986 88.7 44.35 133.05 1164.38 154,921 4,824 1,776,813 1987 75.6 37.8 113.4 1572,39 178,309 6,591 1,496,792 1988 107 53.5 160.5 1550.58 248,868 8,581 1,604,615 1989 111.3 55.65 166.95 1733.12 289,344 10,637 1,504,996 1990 109.7 54.85 164.55 1609.3 264,810 14,310 1,023,846 1991 107 53.5 160.5 1602.46 257,195 16,784 847,825 1992 105.1 52.55 157.65 1781.59 280,868 18,613 834,882 1993 101.9 50.95 152,85 2180.25 333,251 23,092 798,454 1994 107.5 53.75 161.25 2638.24 425,416 28,598 823,034 1995 123.7 61.85 185.55 1952.04 362,201 44,880 446,515 1996 165.1 82.55 247.65 2044.46 506,311 62,230 450,149 1997 117.1 58.55 175.65 2050.2 360,118 79,039 252,082 283 University of Ghana http://ugspace.ug.edu.gh Table A 16 continued. Year Rea! change in cost Proportional cost increase Net shift in supply dC (cedis/mt) Change in quantity caused by research Parameter i = l/P = (dC . t)/(Yf. P) Supply elasticity k = ( j / E) - i Demand elasticity dQ = (Q x e x E x k) / (e + E) ( E = 0.79) (e = - 0.981) 1979 0.79 0.981 1980 0.79 0.981 1981 0.79 0.981 1982 0.79 0.981 1983 0.79 0.981 1984 84,067 0.0000 0.79 0.0101 0.981 2,530.43 1985 87,497 0.0000 0.79 0.0140 0.981 2,426.22 1986 104,559 0.0001 0.79 0.0207 0.981 5,063.14 1987 138,989 0.0003 0.79 0.0309 0.981 8,081.59 1988 170,388 0.0005 0.79 0.0410 0.981 10,771.23 1989 178,489 0.0009 0.79 0.0510 0.981 15,956.33 1990 147,280 0.0019 0.79 0.0656 0.981 15,856.78 1991 164,704 0.0040 0.79 0.0843 0.981 34,350.38 1992 223,894 0.0080 0.79 0.1063 0.981 33,982.34 1993 220,342 0.0130 0.79 0.1325 0.981 55,701.21 1994 236,788 0.0203 0.79 0.1563 0.981 64,271.75 1995 236,606 0.0709 0.79 0.1421 0.981 64,292.15 1996 200,233 0.0966 0.79 0.1527 0.981 67,335.28 1997 195,900 0.2412 0.79 0.0393 0.981 17,202.73 284 University of Ghana http://ugspace.ug.edu.gh Table A16 continued. Real Research costs: Proportional GGDP GGDP (foreign cost: Exchange rate Exchange rate, Social gain from research GGDP (all local costs), maize cost (cedis) Canadian $ (CN$:1US$) cedis: 1US$ Year cedis (90% for 1979-83, SG = (k . P . Q) -1/2 (k . P , dQ) 55% for 1984-96) 1979 389,380 350,442 228,346 1.1714 401.32 1980 389,380 350,442 228,346 1.1692 306.92 1981 519,173 467,256 304,462 1.1989 154.19 1982 1,402,036 1,261,832 319,903 1.2337 135.54 1983 1,030,870 927,783 197,800 1.2324 201.27 1984 11,144,907,262 9,918,988 5,455,443 998,300 1.2951 611.07 1985 11,730,231,816 36,387,694 20,013,232 1,430,482 1.3655 870.05 1986 20,465,056,973 65,949,829 36,272,406 319,101 1.3895 1164.38 1987 27,455,852,643 80,393,569 44,216,463 175,183 1.326 1572.39 1988 39,141,982,605 98,738,058 54,305,932 2,023,483 1.2307 1550.58 1989 54,264,381,552 187,897,196 103,343,458 365,767 1.184 1733.12 1990 36,567,519,046 201,386,259 110,762,442 501,532 1.1668 1609.3 1991 65,324,735,077 278,083,542 152,945,948 2,484,702 1.1457 1602.46 1992 63,325,898,817 427,815,898 235,298,744 1,160,624 1.2087 1781.59 1993 98,687,794,274 450,923,659 248,008,012 4,821,245 1.2901 2180.25 1994 116,748,672,838 699,105,294 384,507,912 679,765 1.3656 2638.24 1995 63,562,775,074 192,709,160 105,990,038 212,020 1.3724 1952.04 1996 66,951,845,061 815,007,795 448,254,287 55,996 1.3635 2044.46 1997 9,824,501,845 0 2050.2 285 University of Ghana http://ugspace.ug.edu.gh Table A16 continued. Real Exchange rate All costs Proportional foreign Total GGDP NARP-CRI Cost NARP-CRI Maize SARI cost SARI maize (8%) Year Cedis:CN$ (CN$ to cedis) cost (cedis) maize cost (cedis) (15% of NARP-CRI) (cedis) (90% for 1979-83, (cedis) (cedis) 55% for 1984-96) 1979 342.60 78,231,020 70,407,918 70,758,360 0 0 0 0 1980 262.50 59,941,802 53,947,621 54,298,063 0 0 0 0 1981 128.61 39,156,723 35,241,051 35,708,307 0 0 0 0 1982 109.86 35,146,026 31,631,424 32,893,256 0 0 0 0 1983 163.32 32,303,802 29,073,422 30,001,205 0 0 0 0 1984 471.83 471,030,176 259,066,597 264,522,040 0 0 0 0 1985 637.17 911,454,313 501,299,872 521,313,104 0 0 0 0 1986 837.98 267,401,815 147,070,998 183,343,404 0 0 0 0 1987 1,185.81 207,734,538 114,253,996 158,470,459 0 0 0 0 1988 1,259.92 2,549,420,874 1,402,181,481 1,456,487,413 0 0 0 0 1989 1,463.78 535,403,803 294,472,092 397,815,550 0 0 0 0 1990 1,379.24 691,734,185 380,453,802 491,216,244 0 0 0 0 1991 1,398.67 3,475,286,346 1,911,407,490 2,064,353,439 0 0 0 0 1992 1,473.97 1,710,727,320 940,900,026 1,176,198,770 0 0 0 0 1993 1,689.99 8,147,833,045 4,481,308,175 4,729,316,187 0 0 0 0 1994 1,931.93 1,313,256,600 722,291,130 1,106,799,042 0 0 535,742,408 42,859,393 1995 1,422.35 301,567,707 165,862,239 271,852,277 1,361,979,250 204,296,888 597,144,000 47,771,520 1996 1,499.42 83,961,556 46,178,856 494,433,143 1,535,228,229 230,284,234 1,171,434,025 93,714,722 1997 0 1,927,478,424 289,121,764 1,353,509,000 108,280,720 286 University of Ghana http://ugspace.ug.edu.gh Table A16 continued. GGDP ,SARI and MOFA Extension Actual releases MOFA Extesion cost SG 2000 80% of SG2000 Total extension Year NARP-CRI (cedis) estimates (cedis) (80% of estimates) for maize at 30% (cedis) (cedis) 1979 70,758,360 0 0 0 0 0 0 1980 54,298,063 0 0 0 0 0 0 1981 35,708,307 0 0 0 0 0 0 1982 32,893,256 0 0 0 0 0 0 1983 30,001,205 0 0 0 0 0 0 1984 264,522,040 156,631,040 125,304,832 37,591,450 0 0 37,591,450 1985 521,313,104 195,788,800 156,631,040 46,989,312 0 0 46,989,312 1986 183,343,404 244,736,000 195,788,800 58,736,640 0 0 58,736,640 1987 158,470,459 305,920,000 244,736,000 73,420,800 0 0 73,420,800 1988 1,456,487,413 382,400,000 305,920,000 91,776,000 20,789,424 16,631,539 108,407,539 1989 397,815,550 853,829,000 683,063,200 204,918,960 24,458,146 19,566,517 224,485,477 1990 491,216,244 857,208,000 685,766,400 205,729,920 28,774,290 23,019,432 228,749,352 1991 2,064,353,439 1,997,101,000 1,597,680,800 479,304,240 33,852,105 27,081,684 506,385,924 1992 1,176,198,770 1,516,825,000 1,213,460,000 364,038,000 39,826,006 31,860,805 395,898,805 1993 4,729,316,187 2,435,152,000 1,948,121,600 584,436,480 46,854,125 37,483,300 621,919,780 1994 1,149,658,435 2,272,267,000 1,817,813,600 545,344,080 55,122,500 44,098,000 589,442,080 1995 523,920,684 2,455,425,000 1,964,340,000 589,302,000 64,850,000 51,880,000 641,182,000 1996 818,432,100 533,000,000 426,400,000 127,920,000 84,305,000 67,444,000 195,364,000 1997 397,402,484 829,000,000 663,200,000 198,960,000 96,000,000 76,800,000 275,760,000 287 University of Ghana http://ugspace.ug.edu.gh Table A16 continued. Extension + Research Extension , Research Extension + Research (Real cost in cedis) CIMMYT Costs CIMMYT Costs CIMMYT Real Costs and CIMMYT real costs Year (CN$) (cedis) (cedis, 1997= 1) (in cedis) 1979 70,758,360 20,045,412,220 50,000 17,129,930 4,852,804,793 24,898,217,013 1980 54,298,063 9,579,608,548 50,000 13,125,214 2,315,633,427 11,895,241,975 1981 35,708,307 3,008,900,708 50,000 6,430,478 541,853,460 3,550,754,168 1982 32,893,256 2,260,739,187 50,000 5,493,232 377,547,429 2,638,286,616 1983 30,001,205 898,206,534 50,000 8,165,774 244,475,234 1,142,681,769 1984 302,113,490 6,538,540,012 50,000 23,591,615 510,585,329 7,049,125,341 1985 568,302,416 11,302,983,049 50,000 31,858,294 633,630,517 11,936,613,566 1986 242,080,044 3,966,369,118 50,000 41,899,244 686,499,663 4,652,868,781 1987 231,891,259 2,780,830,406 25,000 29,645,362 355,505,958 3,136,336,364 1988 1,564,894,952 14,414,139,625 25,000 31,497,928 290,125,246 14,704,264,871 1989 622,301,026 4,624,052,912 25,000 36,594,595 271,918,789 4,895,971,702 1990 719,965,596 3,976,615,007 25,000 34,481,059 190,450,625 4,167,065,632 1991 2,570,739,363 12,106,093,214 25,000 34,966,833 164,665,364 12,270,758,578 1992 1,572,097,575 6,675,819,064 25,000 36,849,301 156,478,370 6,832,297,434 1993 5,351,235,967 18,316,141,504 25,000 42,249,632 144,611,495 18,460,752,999 1994 1,739,100,515 4,806,516,735 25,000 48,298,184 133,486,263 4,940,002,998 1995 1,165,102,684 2,051,883,936 25,000 35,558,875 62,623,394 2,114,507,330 1996 1,013,796,100 1,287,633,455 25,000 37,485,515 47,610,761 1,335,244,216 1997 673,162,484 673,162,484 0 0 0 673,162,484 I . . .... - 288 University of Ghana http://ugspace.ug.edu.gh Table A16 continued. NET SOCIAL GAINS Year NG = SG - (R + E) (cedis) 1979 -24,898,217,013 1980 -11,895,241,975 1981 -3,550,754,168 1982 -2,638,286,616 1983 -1,142,681,769 1984 4,095,781,921 1985 -206,381,749 1986 15,812,188,192 1987 24,319,516,279 1988 24,437,717,734 1989 49,368,409,851 1990 32,400,453,414 1991 53,053,976,500 1992 56,493,601,383 1993 80,227,041,275 1994 111,808,669,840 1995 61,448,267,743 1996 65,616,600,845 1997 9,151,339,362 RR = 25% Source: Estimated by the author. 289 University of Ghana http://ugspace.ug.edu.gh University of Ghana http://ugspace.ug.edu.gh APPENDIX 17 Table A. 17: Area Planted to Specific Improved Maize Varieties in the Ecological Zones of Ghana Variety Area (hectares) Forest Transition Coastal Savanna Guinea Savannah Total “Agric” 16.9 34.6 37.7 30.4 119.6 Obatanpa 14.5 29.7 32.3 26.2 102.6 Dobidi 5.8 11.9 13.1 10.3 41.1 Aburotia 3.2 6.6 7.2 6.0 23.0 Abeleehi 2.9 6.0 6.6 5.3 20.8 La Posta 2.8 5.8 6.3 5.1 20.0 Okomasa 2.7 5.5 6.0 4.8 19 Golden Crystal 0.2 0.5 0.5 0.4 1.6 Dorke SR 0 0.1 0.1 0 0.2 All improved 49.0 100.5 109.6 88.7 347.8 Source: Author’s computations based on the CRI/CIMMYT survey 290 University of Ghana http://ugspace.ug.edu.gh APPENDIX 18 Table A18: Principal Month for Selling Maize in the Ecological Zones of Ghana Month Percent of fanners selling maize Coastal Savannah Forest Transition Guinea Savannah January 6.0 7.5 9.5 9.5 February 3.6 6.9 4.8 7.1 March 8.3 11.6 6.3 4.8 April 3.6 10.1 0 10.7 May 8.3 7.4 4.8 1.2 June 1.2 3.7 9.5 2.4 July 2.4 4.8 4.8 0 August 7.1 6.3 11.1 2.4 September 10.7 16.0 23.8 2.4 October 14.3 19.8 30.2 4.8 November 15.5 31.2 47.6 10.7 December 16.7 30.2 34.9 9.5 Source: Author’s computations based on the CRI/CIMMYT survey 291 University of Ghana http://ugspace.ug.edu.gh