COST K. (I) 4 >4 .V‘ .“=ls. ‘ " Nationa1 1*.193'Q Fae: ‘4 VJSelfn (F‘s J: ‘u . ‘- "; ‘ uh kc‘rtk v :41 P» ~C ‘- ~>-‘ ‘P'. ”3*“. h \“:«: \u" Crc. ~agn “ e ABSTRACT COSTS AND RETURNS OF ALTERNATIVE RICE PRODUCTION SYSTEMS IN NORTHERN GHANA: IMPLICATIONS FOR OUTPUT, EMPLOYMENT AND INCOME DISTRIBUTION BY Fred Everett Winch III In recent years the demand for food in Ghana has increased faster than the ability of the agricultural sector to meet domestic food requirements. As a consequence, Ghana is relying on imports to supple— ment domestic production, although there are local efforts to stem the crisis. The National Redemption Council, one such effort, has attempted to achieve self-sufficiency in food production through the ”Operation Feed Yourself" (OFY) Program, the focus of which is the rice industry in the Northern Region. The bottomlands in the region, well suited to the production of rain-fed paddy and rice acreage, expanded from 28,000 to 90,000 acres between 1968 and 1974. The principal reasons for this dramatic acreage expansion are: (1) access to idle bottomlands, (2) increased use of tractor mechanization for land preparation, (3) sub- sidized inputs such as improved seed, fertilizer, and a government Operated conbine harvesting service, (45 prestige associated with eXPanding farm size, and (5) high private returns from rice farming. There is a general lack of quantitative data on the costs, returns, OntPut, employment, and income distribution implications of alternative a'c graduation tech: 31%;: was to get sway of 161 ri £2.35 was used . LEI farmers from is: five bottom}; :‘z-zw. analyzed fr: xix of view, Financial r Efté‘zction system an: . ‘-«?&..39‘3.’h€nt wa: 45 ~ . 35'. .0 cerlve fir ‘J'N "rual emenA' t' ‘13: "r ‘ ‘ Q “ten. TL‘C‘ “ \ ta‘WA, any“ \- :‘m-AHA .‘g‘"" ‘01 R n S‘ Q‘ LEm ‘i:.a .t-Y -" " ti‘ A: ~_ v“ V . mductior :9" ~“":’ a. L-NH DI. u... ‘H Fred Everett Winch III production technologies for major fOod crops. The objective of this study was to generate such micro data on the rice industry through a survey of 161 rice farms in the Northern Region. The Cost Route Survey Method was used to collect farm level data by continuously interviewing 161 farmers from May, 1973 through February, 1974. The costs and returns for five bottomland production systems and one upland rice system were then analyzed from both the financial (private) and economic, or national, point of view. Financial rice enterprise budgets were prepared for each of the six production systems. A net cash return to operating capital, family labor and management was derived for each system. The budget data were also used to derive financial returns to (1) family labor, (2) operating capital expenditures, and (3) management, as well as cost of production. Each production system was then analyzed from an economic point of view. The unsubsidized costs of nonlabor inputs were estimated, and the economic costs and returns were calculated for each of six rice produc- tion systems. The analysis showed that when market prices (financial analysis) are used to value resources, the 119 acre, capital intensive production system.has the lowest cost of production (¢104 per ton). HOwever, when economic prices are used, this system has the highest cost of production, the highest capital/labor ratio, and the largest government income transfer via capital input subsidies. Moreover, the economic analysis revealed that four of the five bottomland production SYStems are generating economic losses from the national point of view. The next step was to compare the employment and income distribution implications of alternative rice production strategies. Our analysis revealed that there was a wide variation in the average man-hours per acre amng the f gez asre for t1”.- _:e: acre for Ch of the labor we! The emgloy. from hand harves 25:; revealed th. whereas carbine ‘r levels of comb in: sen'ices. Yet as Manually displ 31 the average, E 1131. we estima:.; f: “'39 acres ha: The Study a1 FEW ' M m the hot “210?. g. rategy 1 \‘- :A‘ “'Q1 3 ORA IJ'JU a ‘L ‘ :‘;‘*‘ U“. Jt ,. t. . 5‘s H \_ 3195;,“th 33?; ; ‘On Strater else . . «J‘shl‘ o ”S t ' a 1 “:2“ . .uin a 8%1 4:35 = Fred Everett Winch III acre among the five bottomland systems, from a high of 220 man—hours per acre for the 16 acre tractor hire system to a low of 38 man-hours per acre for the 119 acre tractor owner system. About three-fourths of the labor were employed in harvest activities. The employment and income distribution implications of a shift from hand harvesting to combine harvesting were analyzed in detail. Our data revealed that manual harvesting requires 142 man-hours per acre, whereas combine harvesting requires 10 man-hours per acre. At current levels of combine subsidies, farmers are encouraged to adopt combine services. Yet as combine harvesting expands, 22,000 man-days are potentially displaced per 1000 acres harvested by combine. Thus, if, on the average, 60 percent of the labor requirements for harvesting are hired, we estimated that about ¢12,700 would be lost by casual workers per 1000 acres harvested by combine. The study also illustrates that the production strategies being pursued in the bottomlands of Northern Ghana are providing rice producers with high financial returns, but at a high cost to the Ghanaian economy. The key policy issue now facing Ghana is how to develop a low cost pro- duction strategy in light of a growing fbreign exchange constraint and the need to reduce government subsidies to the rice industry. A hypo— thetical 35,000 acre Rice Production Campaign was used to illustrate the output, efficiency, employment, and income distribution implications 0f two production strategies: a small farmer strategy and a large farmer Production strategy. Our analysis showed that with current input/output relationships, a large farm strategy would produce about 17 percent more Ontput than a small farm strategy because of the higher yields on the large farms. Although there is only about a five percent difference in aggregate income itiications ar» been umn an av farm income of g: intensive stratel farm income for , ”‘13 Generate ac a large farm stra Fred Everett Winch III aggregate income between the two strategies, the income distribution implications are substantially different. A small farmer strategy based upon an average rice enterprise of four acres would provide a net farm income of ¢240 for 8700 small farmers, while the large-farm, capital intensive strategy of 100 acre farms would generate over ¢6000 in net farm income for each of 350 farmers. In addition, a small farm strategy would generate aggregate employment of about one million man—days, whereas a large farm.strategy would employ only 240,000 man-days, or 77 percent less labor. And a small farm strategy would generate about ¢55,000 in wages for casual labor compared with ¢200,000 in wages for the large farm strategy. Finally, under current subsidy policies about ¢2.8 million would be required to subsidize capital inputs for a 35,000 acre capital intensive strategy, whereas ¢O.9 million would be required for a small farm rice production campaign. The study concludes by recommending that the Ministry of Agriculture embark on a major Rice Production Campaign for small farmers; it dis- cusses ten recommended components of a small farm production campaign and points up the need fOr more research on the benefit incidence of agricultural development projects. in COSTS AND RETURNS OF ALTERNATIVE RICE PRODUCTION SYSTEMS ‘ IN NORTHERN GHANA: IMPLICATIONS FOR OUTPUT, EMPLOYMENT AND INCOME DISTRIBUTION BY Fred Everett Winch III A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Agricultural Economics 1976 © Copyright Fred Everett Winch III 1976 ACKNOWLEDGMENTS Many people have provided advice and encouragement during my gnaduate studies. I especially wish to express my appreciation to Chrl K. Eicher, my major professor and thesis supervisor for his inspira— tion and guidance throughout my graduate program. In addition, I wish to thank the members of my guidance and thesis committees: Drs. Lester Manderscheid, Glenn L. Johnson, Anthony Koo, Roy Black, Warren Vincent and Carl Liedholm. To the department of Agricultural Economics I am grateful for the generous financial support and several travel opportunities during my graduate program. I am indebted to many in the Ghana Ministry of Agriculture, especially Mr. M. S. 0. Nicholas the recent director of Agriculture for his constant support. Mallam Issah Seidu, Agricultural Economics, Economics and Marketing Division, provided many valuable insights for the field survey and served in his capacity as my counterpart very effectively. Among my fellow graduate students, I wish to pay special tribute to Tom Zalla and Merritt Sargent for assistance and intellectual stimu- lation. Thanks go to Ms. Janet Munn and Linda Sanders for typing. And a special note of appreciation to Janet for her good nature and assis- tance when deadlines seemed impossible. To my wife Frances and my daughters Dinah and Jill, I wish to exPress my deepest appreciation for their sacrifice and understanding. ii Chapter II III TABLE OF CONTENTS INTRODUCTION . . . . . . . . . . . . . . . . . . . . Problem Setting . . . . . . . . . . . . . . . . . Need for the Study . . . . . . . . . . . . . . . . Objectives of the Study . . . . . . . . . . . . . Scope of the Study and Research Approach . . . . OVERVIEW OF RICE PRODUCTION IN NORTHERN GHANA . . Introduction . . . . . . . . . . . . . . . . . . Physical Characteristics of the Rice Producing Areas . . . . . . . . . . . . . . . . . Climte O O O O O O O O O O O O O O O O O O 0 Soil and Vegetation . . . . . . . . . . . . . Agronomic Production System . . . . . . . . . The Number of Rice Producers and the Distribution of Farm Size . . . . . . . . . . . . . . . . . . Acreage Expansion and Production Estimates . . . Bottomland Production Practices . . . . . . . . . Land Clearing . . . . . . . . . . . . . . . . Land Preparation . . . . . . . . . . . . . . . Seed Varieties and Planting Practices . . . . Use of Inorganic Fertilizer . . . . . . . . . Weed Control . . . . . . . . . . . . . . . . . Harvesting Methods . . . . . . . . . . . . . . Harvesting Problems Arising From Climatic Features and Seed Varieties . . . . . . . . . . . . . . . Labor Shortage at Harvest . . . . . . . . . . . . Yield Estimates . . . . . . . . . . . . . . . . . Summary . . . . . . . . . . . . . . . . . . . . . RESEARCH METHODOLOGY . . . . . . . . . . . . . . . . Introduction . . . . . . . . . . . . . . . . . . Purpose and Scope of the Survey. . . . . . . . . Method of Primary Data Collection . . . . . . . . Survey Method . . . . . . . . . . . . . . . . Selection of Enumeration Areas . . . . . . . . iii Page H UIU'ibN \l \O 10 10 13 13 14 15 18 21 21 24 25 25 28 31 31 31 32 32 34 C'npter Chapter Sampling Procedure and Sample Size . . . Selection and Training of Field Enumerators Interviewing Procedures . . . . . . . . Supervision of Field Staff . . . . . . . The Nature and Measurement of Input—Output Data 0 C C O O C O O O O O O O O O O O I Data Tabulation, Coding and Storage . . . . Tabulation and Coding . . . . . . . . . Standard Data File . . . . . . . . . . . Principal Features of the Sample . . . . . Farms Excluded From the Sample . . . . . Sample Characteristics . . . . . . . . . Defining the Production Systems . . . . Summary . . . . . . . . . . . . . . . . . . IV .A FINANCIAL AND ECONOMIC ANALYSIS OF THE MAJOR PRODUCTION SYSTEMS IN NORTHERN GHANA . . . . . Introduction . . . . . . . . . . . . . . . Distinctions Between the Financial and Economic Analysis . . . . . . . . . . . . . Financial Analysis . . . . . . . . . . . Economic Analysis . . . . . . . . . . . A Financial Analysis of Six Rice Production systems 0 I O C O O O O O O O O I O O O O 0 Calculation of Budgets . . . . . . . . . Aggregation of Inputs and Factor Costs Derivation of the Costs of Land Prepara— tion Among Tractor Owners and Bullock Operators . . . . . . . . . . . . . . Derivation of Land Clearing Costs . . Costs and Returns of Six Production Systems System I: Farmers Hiring Tractor Services and Using Traditional Seed . . . . . System II: Farmers Hiring Tractor Services and Using Improved Seed . . System III: Farmers Hiring Tractor 0 Services and Using Mixed Seed Varieties . System IV: Tractor Owners Using Traditional Seed . . . . . . . . . . iv Page 35 39 40 42 43 45 45 46 48 48 49 54 55 57 57 57 57 58 59 59 59 60 61 61 62 64 66 69 C‘a - pte r go Chapter System V: Tractor Owners Using Improved Seed . . . . . . . . . . . . . . . . . . System VI: Upland Rice Farmers Using a Bullock Plow and Traditional Seed Varieties . . . . . . . . . . . . . . . Comparative Financial Analysis of Six Rice Production Systems . . . . . . . . . . . . Net Cash Income . . . . . . . . . . . . Return to Family Labor . . . . . . . . . Return to Operating Capital . . . . . . Return to Management . . . . . . . . . . Cost of Production . . . . . . . . . . . Summary of Financial Analysis . . . . . An Economic Analysis of the Six Rice Production system 0 I I O O O O I O O O O O O O O O O 0 Theoretical Framework . . . . . . . . . . . Calculation of Unsubsidized Factor Prices and the Percent of Subsidy for Each Factor Method of Calculating the Economic Cost of Rice Production . . . . . . . . . . . . . . A Comparative Analysis of the Economic Benefits and Economic Costs of Production of Six Rice Production Systems . . . . . . A Comparison of Financial and Economic Costs of Production of the Six Production Systems. Producer Income Support Derived from Capital Input SIJbSidieS O O O O O O O O O C O 0 Comparison Between Small Farm and Large Farm Rice Production Strategies . . . . . . . . . . . . A Large-Scale Capital Intensive Rice Production System Using Combine Harvesting . . . . . . Labor Utilization . . . . . . . . . . . Costs and Returns . . . . . . . . . . . A Small-Scale Labor Intensive Rice Production System Using Manual Methods of Harvesting . Labor Utilization . . . . . . . . . . . Costs and Returns . . . . . . . . . . . A Comparative Financial and Economic Analysis . Financial Analysis of the Two Systems . V Page 72 75 79 79 79 82 84 84 85 85 85 86 88 90 92 95 98 100 100 102 102 104 104 104 104 fiapter V Ba: .0. 5 .t r a. C mun Cw mmc be i ‘, V1. is 8 use U R. U C 00 Co E N .3 Chapter VI ‘ Economic Analysis of the Small Farm and. Large Farm Systems . . . . . . . . . . . . . Summary . . . . . . . . . . . . . . . . . . . . . EMPLOYMENT AND INCOME DISTRIBUTION IMPLICATIONS OF ALTERNATIVE RICE PRODUCTION SYSTEMS . . . . . . . . . Introduction . . . . . . . . . . . . - . . . . . . Labor Utilization Among Six Rice Production Systems . . . . . . . . . . . . . . . . . . . . . Five Bottomland Systems . . . . . . . . . . . . Pre-Harvest Activities . . . . . . . . . . . Harvest Activities . . . . . . . . . . . . . Upland Bullock System . . . . . . . . . . . . . Family Versus Hired Labor . . . . . . . . . Importance of Men, Women, and Children . . . Employment Implications of Expanding Combine Harvesting O O O O O O O O O O O O O O I O O O O 0 Potential Labor Displacement . . . . . . . . . Gains to Producers . . . . . . . . . . . . . . Who Benefits From the Current Combine Strategy? . . . . . . . . . . . . . . . . . Effects of Charging Farmers the Economic Cost of Combine Services . . . . . . . . . . Summary . . . . . . . . . . . . . . . . . . . . . SUMMARY AND POLICY PRESCRIPTIONS FOR IMPROVING THE ECONOMIC PROFITABILITY OF RICE PRODUCTION IN THE NORTHERN REGION OF GHANA . . . . . . . . . . . . . . Summary . . . . . . . . . . . . . . . . . . . . . Tentative Policy Recommendations to Improve the Performance of the Northern Region Rice Production Sector . . . . . . . . ... . . . . . . . . . . . . Major Policy Issues . . . . . . . . . . . . . . Recommended Policy Reorientation . . . . . . . Small Vs. Large Farmer Production Strategies: Output, Employment, and Income Distribution Implications of a 35,000 Acre Rice Production Campaign . . . . . . . . . . . . . . . . . . . vi Page 107 107 113 113 113 113 114 114 117 118 119 122 122 125 125 126 127 129 129 137 137 139 140 f“ xagter :EPEXDICES ('1 C) Chapter APPENDICES A Output, Employment, and Income Distribution Effects . . . . . . . . . . . . . . . . . . Income Distribution Implications . . . . . Employment Implications for Casual Workers. Summary . . . . . . . . . . . . . . . . . . Recommended Components of a Small Farm Production Campaign . . . . . . . . . . . . . Recommended Policy Changes for Large-Scale Rice Production . . . . . . . . . . . . . . ADDITIONAL INFORMATION ABOUT TRACTOR OWNERS, FARMERS HIRING PRIVATE TRACTOR SERVICES, AND BULLOCK me I I I I I I I I I I I I I I I I I I I I I I SELECTED ATTRIBUTES OF SAMPLE FARMS FOR SIX RICE PRODUCTION SYSTEMS IN NORTHERN GHANA, 1973-74 . . . CALCULATION OF LAND PREPARATION COSTS FOR TRACTOR OWNERS I I I I I I I I I I I I I I I I I I I I I I I CALCULATION OF LAND PREPARATION COSTS FOR BULLOCK OWNERS I I I I I I I I I I I I I I I I I I I I I I I ESTIMATED ECONOMICS OWNING AND OPERATING COSTS OF TWENTY-FOUR SELF-PROPELLED COMBINES . . . . . . . . ESTIMATED LAND CLEARING COSTS PER ACRE ON BOTTOMLAND RICE PAWS I I I I I I I I I I I I I I I I I I I I I ESTIMATED COST OF ONE BAG OF IMPROVED RICE SEED SOLD BY SEED MULTIPLICATION UNIT, MINISTRY OF AGRICULTURE NORTHERN REGION, GHANA AND THE RATE SUBSIDY, 1973-74 CALCULATION OF FERTILIZER COST PER TON AND RATES OF GOVERNMENT SUBSIDY AT OFFICIAL AND SHADOW FOREIGN E XCWGE “TE 8 I I I I I I I I I I I I I I I I I I I ESTIMATION OF THE IMPORT PARITY PRICE FOR RICE MILLED BY THE RICE MILLS UNIT AT TAMALE, NORTHERN GHANA . . BI BLI OGRAPHY I I I I I I I I I I I I I I I I I I I I I I I I I vii Page 140 142 ”144 144 146 153 157 168 176 181 183 185 186 187 188 191 A I 6.0 1.: in Table 2.1 2.2 2.3 2.4 2.5 2.6 2.7 3.1 3.2 3.3 3.4 305 3.6 4.1 LIST OF TABLES Distribution of Farm Size Among Rice Farmers in Northern Region, 1971 . . . . . . . . . . . . . . . . Number of Rice Farmers and Area Under Rice in Selected Districts of the Northern Region, 1971 . . . Estimated Acreage and Rice Production in Northern Region of Ghana Between 1968 and 1974 . . . . . . . . Purchase and Sale of Improved Rice Seed by the Northern Region Seed Multiplication Unit of the Ministry of Agriculture, 1970 and 1974 . . . . . . . Results of Ministry of Agriculture Farm Location Fertilizer Trials With IRr20, 1974 . . . . . . . . . Yield-Cut Estimates of Paddy in the Northern Region of Ghana . . . . . . . . . . . . . . . . . . . . . . Declared Yields of Paddy in the Northern Region for Farms 15 Acres and Above, 1971 . . . . . . . . . . . Desired Stratified Purposive Sample of Rice Farmers for the 1973-74 Production Season . . . . . . . . . . Classification and Listing of Activities Recorded on the Weekly Input—Output Record . . . . . . . . . . . Description of Resources Used in Rice Production and the Corresponding Variables Measured During the Field Sllrvey I I I I I I I I I I I I I I I I I I I I I I I Principal Features of the Stratified Sample of Rice Farms I I I I I I I I I I I I I I I I I I I I I I I I Sample Distribution of Farm Size and Source of Power for Land Preparation . . . . . . . . . . . . . . . . Method of Harvesting Used by All Sample Farms . . . . Rice Enterprise Budget for a 12.8 Acre Farm Based on Survey Data from Twenty-eight Farms in Northern Ghana Using Traditional Seed Varieties and Tractor Hire Services, 1973-74 System I . . . . . . . . . . . . . viii Page 11 11 12 16 20 27 27 38 44 44 50 51 52 63 L3 J— o ‘4 all ‘ - Cl.) J ‘ 9 IL) .4 - o 91.4 I...) Rice : Survey Using 1973-? Rice E Survey Improv. Hire (1) Rice E: SuIVey Tradit' Equipm. Rice E; Survey Using 1 Equigme Rice En Up0n 81; Ghana U Power f A COmPai PIOduct. Subsidiz “‘- Nortr A Compar Table 4I2 4.3 4.4 4.5 4.6 4.8 4.9 4.10 4.11 4.12 4.13 4.14 Rice Enterprise Budget for a 21.2 Acre Farm Based on Survey Data from Forty-Four Farms in Northern Ghana Using Improved Seed Varieties and Tractor Hire Services, 1973-74 (System II) . . . . . . . . . . . . . . . . . . Rice Enterprise Budget for a 16.9 Acre Farm Based on Survey Data From Eleven Farms in Northern Ghana, Using Improved and Traditional Seed Varieties and Tractor Hire Services, 1973-74 (System III) . . . . . . . . . . Rice Enterprise Budget for a 41.6 Acre Farm Based on Survey Data from Ten Farms in Northern Ghana Using Traditional Seed Varieties and Own Tractor and Equipment, 1973-74 (System IV) . . . . . . . . . . . . Rice Enterprise Budget for a 119.3 Acre Farm Based on Survey Data from Nineteen Farms in Northern Ghana Using Improved Seed Varieties and Own Tractor and Equipment, 1973-74 (System V) . . . . . . . . . . . . . Rice Enterprise Budget for a 1.1 Acre Rice Farm Based Upon Survey Data from Fourteen Farmers in Northern Ghana Using Traditional Seed Varieties and Bullock Power for Land Preparation (System VI) . . . . . . . . A Comparative Financial Analysis of Six Major Rice Production Systems in Northern Ghana . . . . . . . . . Subsidized and Unsubsidized Prices and Percent of Subsidy for Selected Inputs Used in Rice Production in Northern Ghana, 1973-74 . . . . . . . . . . . . . . A Comparative Economic Analysis of Six Rice Production Systems in Northern Ghana . . . . . . . . . Financial and Economic Costs of Production of Six Rice Production Systems in Northern Ghana, 1973-74 . . Financial and Economic Capital-Labor Ratios for Six Rice Production Systems in Northern Ghana, 1973-74 . . A Comparison of the Costs Per Acre for Capital Resources and the Capital-Labor Ratios of Six Rice Production Systems in Northern Ghana, 1973-74 . . . . . . . . . . Comparative Capital Subsidy Producer Income Support Among Six Rice Production Systems in Northern Ghana During the 1973-74 Production Season . . . . . . . . . Rice Enterprise Budget for a 287.5 Acre Farm Based on Survey Data from Four Selected Farms in Northern Ghana Using Improved Seed and Own Tractor and Equipment and Having Above Average Yield . . . . . . . . . . . . . . 1x Page 65 68 7O 73 76 80 87 91 93 94 96 97 101 Tile 1‘: 0.4 4&6 . 111 Rice 1 Survey Ghana Having A Com; and La Survey Table 4.15 4.16 4.17 5.1 5.2 5.3 5.4 5.5 5.6 6.1 Rice Enterprise Budget for a 3.9 Acre Farm Based on Survey Data from Five Selected Farms in Northern Ghana Using Improved Seed and Tractor Hire Services and Having Above Average Yield . . . . . . . . . . . . . . A Comparative Financial Analysis Between Small Farm and Large Farm Rice Production Strategies Using Survey Data . . . . . . . . . . . . . . . . . . . . . A Comparative Economic Analysis of Small Farm and Large Farm Rice Production Strategies Using Survey Data . . . . . . . . . . . . . . . . . . . . . Summary of Labor Utilization for Six Rice Production Systems in Northern Ghana . . . . . . . . . . . . . . The Relationship Between the Method of Harvesting and Average Labor Requirements for Five Bottomland Rice Production Systems in Northern Ghana . . . . . . Average Man-Hour Requirements Per Acre for Manual Harvesting Activities at Three Yield Levels . . . . . The Relative Importance of Men, WOmen, and Children as Sources of Labor for Field Activities Among Six Rice Production Systems in Northern Ghana . . . . . . Comparative Labor Requirements for Manual and Combine Harvesting and Estimated Labor Displacement for Combine Harvesting in Northern Ghana . . . . . . . . . Estimated Loss of Income to Casual WOrkers Resulting from a Shift from Manual to Combine Harvesting Under Three Assumptions Regarding the Proportion of Total Labor Requirement Hired . . . . . . . . . . . . . . . Projected Output, Employment, and Income Impacts of a 35,000 Acre Rice Production Campaign in Northern Ghana: Small Farm Vs. Large Farm Strategies . . . . . Producer Income Distribution Implications of a 35,000 Acre Rice Production Campaign in Northern Ghana: Small Farm Vs. Large Farm Approach . . . . . . . . . . Income Distribution Implications of a 35,000 Acre Rice Production Campaign for Casual workers in Northern Ghana: Small Farm Vs. Large Farm Approach . . . . . . Page 103 106 108 115 116 118 120 124 124 141 143 145 Al L2 Ad (:1 I... (”D o B) K] 9.1 Use Trai Esti a TL Duri SUbS 53th Trac' DUIi: SUbS EStii in N: ESti in N: Table A.1 B.l C.1 C.2 C.3 D.l D.2 E.1 F.1 G.l H.l Page Use of Selected Improved Practices Among 25 Sample Tractor Owners . . . . . . . . . . . . . . . . . . . . 160 Use of Selected Improved Practices Among 79 Farmers Using Tractor Hire Services for Initial Land Preparation . . . . . . . . . . . . . . . . . . . . . 163 The Relative Importance of Rice and Other Crops Among 79 Sample Rice Farmers Hiring Tractor Services for Initial Land Preparation During the 1973-74 Crop Season . . . . . . . . . . . . . . . . . . . . . 164 Selected Attributes of Sample Farms for Six Rice Production Systems in Northern Ghana, 1973-74 . . . . 168 Tractor and Equipment Performance Assumptions for Northern Ghana Tractor Owners . . . . . . . . . . . . 176 Estimated Financial Owning and Operating Costs of a Tractor and Associated Equipment in Northern Ghana During the 1973-74 Crop Season: The Case with Subsidies . . . . . . . . . . . . . . . . . . . . . . 177 Estimated Economic Owning and Operating Costs of a Tractor and Associated Equipment in Northern Ghana During the 1973-74 Cr0p Season: The Case Without Subsidies . . . . . . . . . . . . . . . . . . . . . . 179 Estimated Financial Cost Per Acre for Bullock Plowing in Northern Ghana, Based Upon Survey Data, 1973 . . . 181 Estimated Economic Costs Per Acre for Bullock Plowing in Northern Ghana‘ . . . . . . . . . . . . . . . . . . 182 Estimated Economic Owning and Operating Costs of Twenty-Four Self-Propelled Combines Operated by the Ministry of Agriculture in Northern Ghana, 1973-74: When Import Prices are Converted at the Shadow Rate of Exchange . . . . . . . . . . . . . . . . . . . . . 183 Estimated Land Clearing Costs Per Acre on Bottomland Rice Farms in Northern Ghana, 1973 . . . . . . . . . . 185 Estimated Cost of One Bag of Improved Rice Seed Sold by Seed Multiplication Unit, Ministry of Agriculture, Northern Region, Ghana and the Rate Subsidy, 1973—74 . 186 Calculation of Fertilizer Cost Per Ton and Rate of Government Subsidy at Official and Shadow Foreign Exchange Rates, 1973-74 . . . . . . . . . . . . . . . 187 xi Map LIST OF MAPS Location of Eleven Enumeration Areas in Northern Ghana for the 1973-74 Rice Production Survey . . . . . . . . xii Page 33 Ghana, a flat in 1957, : if the northerr south, the sou: te capital of PPICXimated at ;er year. Cocoa , w}: *0! Droduct liar. . CHAPTER I INTRODUCTION Ghana, a country on the West Coast of Africa which became indepen- dent in 1957, is divided into the following ecological zones: the savanna of the northern half of the country, the forest which covers much of the south, the southwest rain forest and a coastal savanna which surrounds the capital of Accra and extends eastward. The present population is approximated at 9.2 million and is growing at an estimated 2.8 percent per year. Cocoa, which is produced by smallholders in the forest zone, is the major product and foreign exchange earner of the country. Cocoa produc- tion expanded rapidly at the turn of the century and has been the engine of Ghanaian growth fOr over 75 years. During the 1960's about 20 percent of the total labor force was engaged in cocoa production either as farm. operators or as hired labor. Despite this labor concentration, there is a substantial diversity in crop production, farm size,and the degree of market orientation within Ghanaian agriculture.1 The typical farmer operates about five acres of land and has adapted his production practices to a relative abundance of land and labor, to meager capital resources,and to soils which become rapidly exhausted when farmed intensively.2 However, there are a growing 1For a good reference on the economy of Ghana and some aspects of social structure, see Walter Birmingham, I.Neustadt, and E. N. Omaboe, editors, Volumes One and Two, 1966 and 1967. 2Good references on Ghanaian agriculture include Wills, 1962; Killick, 1966. 1 timer of far acreages. Ca the form of t. extent. improx Over the pace with the POPulation gm since abOut 1, ?&Stities 0f number of farmers who produce one to two cash crops on relatively large acreages. Capital inputs on these farms are, for the most part, in the form of tractor mechanization (owned or hired) and, to a lesser extent, improved seed and fertilizer. Problem Setting Over the 1900-1958 period food production in Ghana generally kept pace with the growing demand for food which was largely a function of population growth and modest increases in per capita income. However, since about 1958, domestic production has been augmented with increasing quantities of imported food. To stem increasing foreign exchange require- ments for food imports and rising food prices, there was an attempt in the 1960's to expand domestic food production by means of opening new land through government tractor-hire services and public production units (e.g.,State Farms). However, the approach required large amounts of domestic capital and foreign exchange while contributing little to increased domestic food production. During the 1960's Ghana experienced a balance of payments crisis and import controls were introduced. However, over the 1968—72 period, import controls were liberalized. Ghana's debt burden expanded and food prices increased. In short, Ghana was living beyond its means. Then, in December 1971, the government devalued the cedi by 42 percent. The purpose of devaluation was to increase the domestic price of imported goods, but its magnitude was politically unsound. On January 13, 1972 a military coup d'etat took place. The new government, the National Redemption Council (NRC), immediately revalued the currency by 10 percent and imposed strict import licensing. Moreover, the NRC assigned agricul- ULnal development first priority and the government launched the '0peration W» haestic food The pUIE trace of agri: sufficiency" 1 wild be gener expansion by sr were establisne 0f the OH Rrog H129. livestoc 37:3 given to ya 35216 OFY Prog Hiasis since 3.5131 intensi - a. 4 :13 can be at 4" \ S‘VEH tn e PrOgr 23:94 "3“ EXChanc "Operation Feed Yourself" (OFY) Program, a national program to increase domestic fOOd production. The purpose of OFY was to create a national awareness of the impor— tance of agriculture and to stress the need for "self-reliance" and "self- sufficiency" in food production. OFY envisioned that increased production would be generated in the private sector, principally by means of acreage expansion by small farmers; hence, national and regional acreage targets were established. Import substitution was also an integral component of the OFY Program. The major commodities featured in OFY are rice, maize, livestock, sugar cane, oil palm,and cotton. In addition, emphasis was given to yam, sorghum and cassava production. Despite the reliance of the OFY Program on smallholders to increase food production, increased emphasis since 1973 has been placed on large farms which use relatively capital intensive production techniques. This shift to supporting large farms can be attributed to two factors. First, policy makers thought that output could be increased more rapidly on large famms. Second, given the progress in reducing the external debt burden and building up foreign exchange reserves over the 1972—74 period, the government believed it could afford to import more farm machinery and other capital inputs.3 chever, foreign exchange reserves have fallen drastically in 1975 because of falling world cocoa prices and rising prices of imported goods. Ghana again has a balance of payments deficit. Although domestic food production has increased since 1972, food prices have continued to rise. While the government is pushing for 3The reduction in the external debts was largely achieved, shortly after the 1972 coup, by a unilateral repudiation, of the debts arising out of the Nkruma era. The increase in foreign exchange reserves was due to an increase in the world cocoa and timber prices and reduced imports following import restrictions. 'self-suffici the basic foo s'llion, or 1 Rice pr Attaining to r71 ~13 00 rice Element of th- about 100 pro M iii: has favc i3?» . “1“an ts m. w ‘1.) \ and ‘ A. U o .' :de agr 'v~\ .‘ Is ‘ I ?:E ‘11:": ‘Iond c a ‘3‘ . $51ij -« v " N“ 12‘. ‘ “Arab ‘ 1 a” "self-sufficiency," to date there is an unsatisfied demand for many of the basic food crops. In 1973 Ghana's total food import bill was $80.9 million, or 18.5 percent of its total merchandise imports. Need for the Study Rice production has been given major emphasis in the OFY Program. According to government statistics, 54 percent of the rice produced in Ghana in 1973 was produced in the Northern Region. Other regions, in their order of importance, are the Upper Region (16 percent) and the Volta Region (13 percent). In the Northern Region about 80 percent of the 6,100 rice producers have rice farms of less than 10 acres; about three percent of the farmers have rice farms above 50 acres; and there are about 100 producers with rice farms larger than 100 acres. The OFY Pro- gram has favored the rice farmers with 50 acres and above through the provision of subsidized inputs (seed, fertilizer, land preparation, and combine harvesting services). Small rice fanmers have also been sub- sidized but, as will be shown in this study, to a far lesser extent. There is a dearth of information about the costs and returns of the major food crops produced in Ghana. Agriculture in Ghana is not supported by an effected applied research base which can generate output increasing technologies adapted to the various ecological zones of the country and to the financial managerial conditions of smallholders. To guide agricultural development in Ghana over the next five to ten Years, the government requires an agricultural deve10pment strategy which fixes beyond establishing acreage targets for specific agricultural crOps-- Eistrategy which encompasses production, income, and employment goals with Eishort-run and medium-term perspective. Without data on costs and returns, it will not b effects of a The obi it will not be possible for planners to evaluate the direct and indirect effects of a policy of self-sufficiency. Objectives of the Study The objectives of this study were the following: 1. To provide the Ministry of Agriculture with relevant data on the rice industry in northern Ghana; 2. To determine the relative land, labor,and capital requirements for the major rice production systems; 3. To determine the cost of production and farm incomes for the najor rice production systems; 4. To analyze direct and indirect effects of alternative rice production strategies, with particular emphasis on harvesting; 5. To identify rice production systems with high financial (private) and economic returns from the national point of view. Scope of the Study and Research Approach At the invitation of the Ministry of Agriculture, the author visited Ghana in January of 1971. During this visit the northern rice producing' areas were toured and discussions were held with regional and central lfinistry officials as well as with US/AID Mission personnel. The author submitted a report to the Ministry outlining applied economic research needs for the rice industry in northern Ghana. The main points of the report were: first, the Ministry did not have adequate economic data to fbrmulate sound recommendations for rice farmers on improved technology; second, to obtain such data, farms would have to be surveyed to determine LPIOduction costs and returns for the major production systems; third, an investigation of alternative mechanization strategies was required to determine tl‘h and fourth. program on or The Mir to undertake [3933“. Thus the field res research whi- for one farm uSli'lg 3 Fur? crop season PideJCtion r determine their feasibility from both a private and national point of view; and fourth, there were no data on the socio-economic effects of the rice program on community, regional, and national development. The Ministry reviewed the report and offered the author a contract to undertake the prOposed research as a Principal Agricultural Economist (OPEX). Thus, the author arrived in Ghana in December, 1971 and conducted the field research, in conjunction with other duties, until May, 1974, research which concentrated on collecting farm level input/output data for one farm enterprise-~rice. The study was conducted over two years using a purposive, nonrandom sample of rice farms; during the 1973-74 crop season 160 farms were included in the sample. The research on rice production reported in subsequent chapters estimates farm level resource requirements, costs of production, and net income of the major rice production systems in northern Ghana. The purg Production in briefly descri Ration on the ' P5731: CHAPTER II AN OVERVIEW OF RICE PRODUCTION IN NORTHERN GHANA Introduction The purpose of this chapter is to provide an overview of rice production in Northern Ghana. The salient features of the industry are briefly described to provide the reader with adequate background infor- mation on the industry to set the stage for the analysis which follows. Physical Characteristics of the Rice Producing Areas Climate The Northern and Upper Regions of Ghana are in the Savanna Zone which is characterized by a dry tropical climate with two distinct seasons. The rains build up from April/May to a peak in September and then decline in October. The dry season extends from November to May. In the Upper Region the average annual rainfall is between 35 to 40 inches, and in the Northern Region average annual rainfall is between 40 to 50 inches. In the Northern Region the rainfall between June and (knober, which is the growing period for paddy, has averaged about 32 inches over a period of about 60 rainy days. Noon-day temperatures vary between 75° F. in the rainy season to a maximum of 105° F in the dry season.' Average monthly temperatures ‘vary from 71° F.in the coldest month (December) to 92° F.in the warmest month (March). During the height of the dry season harmattan (December), early morning temperatures drop to 60° F.and below and noon-day tempera- ‘tures average 95° F. The average relative humidity during the months 7 June to Septt ée:lines to e The ra; the growing 5 creates seric section of t‘: V .~ can a C ‘tiszr~1 q‘ fl Q“. "”3 . 5r v. 038 \ 9.x . .:\;N. June to September varies between 78-83 percent and then progressively declines to a low of about 26 percent in January. The rainfall pattern is adequate to support rice production during the growing period; however, the climate directly following the rains creates serious harvesting problems which are discussed in a later section of this chapter. Soil and Vegetation The northern savannas are underlaid with VOltain sandstones and granites. The soil classification of the rice lands is known as Ochrosol-Groundwater Laterite intergrades. Their origin is due to poor internal drainage. The valley—bottom soils consist of grey, porous, structureless, silty loams to clays, rather loose at the surface but becoming very firm'with depth. While these soils are low in humus and chemical nutrients, they are better provided with nutrients than adjoin- ing upland soils. The nature of the soil and the terrain gives rise to natural flooding during the rains; thus, their main use has been for rain-fed, flooded, bottomland rice production. The natural vegetation of the area is Guinean savanna character- ized by tall grasses and short trees often widely spaced. In the bottomlands or fadamas many of the trees have been removed, opening up large areas for rice production. Prior to the spread of lowland rice production, the bottomlands were not used for crop production as seasonal flooding conditions are not suitable for the production of other crops. In 1971 a Physical Land Survey was undertaken by the Ministry of Agriculture to estimate the acreage of bottomlands suitable for rice production. acres of bot pmduction. only minimal Villages as tiaral 1033 less desira‘: accessible, production. From this survey1 it was estimated that there were 150,000 acres of bottomlands which were most suited for development of rice production. These lands were well flooded during the rains, required only minimal land clearing, were readily accessible,and were close to villages as a source of casual labor. The survey identified an addi- tional 100,000 acres for rice production, but these lands were considered less desirable as they required more land clearing or were not readily accessible. Agronomic Production Systems There are two rain-fed agronomic systems of rice production in Northern Ghana: "upland" and "bottomland" (or "fadama") production systems. The upland system is dominant in the Upper Region and in parts of the Northern Region. The lighter upland soils are prepared using the handhoe or bullock plow. While upland soils are very wet during the growing season, water does not normally collect and stand for any extended period of time. Bottomland production is centered in the Northern Region where tractorplowing is widely used for initial land preparation since it has adapted to the heavy soils of the naturally concave bottomland and riverain areas which are subject to temporary flooding.2 In the bottom- lands there is generally no standing water on the rice fields until the plant is about six inches tall. When the plant is about ten inches tall, 1The Physical Survey was undertaken as part of a project identi- fication study which the Ministry of Agriculture initiated to develop alfice development project to be submitted to the IBRD for partial financing. 2In the Northern Region rice is produced only under rain-fed con- ditions; irrigation has been used on a pilot basis only. water is sta Edth as twe] has reached soil is dry. In 197 rice out of three Percen acres 01' 165 10 water is standing on the field, and, at full plant height, there is as much as twelve to sixteen inches of standing water. After the plant has reached full height, water recedes and rice is harvested when the soil is dry. The Number of Rice Producers and the Distribution of Farm Size In 1971 it was estimated that there were 6,100 holders producing rice out of a total of 61,200 holders in the Northern Region.3 Fifty- three percent of the rice farmers in 1971 were producing rice on five acres or less,and 90 percent were producing on 15 acres or less (Table 2.1). Approximately 10 percent of the farmers (670) who produced rice on more than 15 acres were, as a group, producing rice on about half of the total rice acreage. 0n the other hand, about 90 percent of the rice farmers (5,400) were producing paddy on about the same acreage (28,500 acres); see Table 2.2. Acreage Expansion and Production Estimates There has been a rapid increase in rice production in the Northern Region in recent years because of acreage expansion and to a lesser extent because of increases in yield per acre. Acreage expanded from 28,000 acres in 1968 to about 90,000 acres in 1974 (Table 2.3), an increase of about 220 percent over six years. During the same period it is estimated that average rice yields increased from 800 pounds per acre in 1968 to about 1,200 pounds per acre in 1974, an average yield 3As used by the Ghana Sample Census of Agriculture 1970, a "holder" is the person who has the responsibility for the agricultural "holdinga" Almflding is all the land which is used for agricultural production and iscmerated as one technical unit. A holding generally consists of several fields or "farms" in the Ghanaian context. We will refer to holders as farmers and will refer to their rice farms which may be part or all of a farmers holding. \ “mks (f) 0 / r. 11 Table 2.1. Distribution of Farm Size Among Rice Farmers in Northern Region, 1971 Acres No. of Farms Percent of Farms 0.1 - 2.0 1,200 20 2.1 - 5.0 2,000 33 5.1 - 10.0 1,600 26 10.1 — 15.0 700 11 15.1 - 50.0 400 7 50.1 - 100.0 100 2 More than 100 acres 100 .__1 6,100 100 Source: Ministry of Agriculture, Economics and Marketing Division, Accra, mimeo, 1972. Table 2.2. Number of Rice Farmers and Area Under Rice In Selected Districts of the NOrthern Region, 1971 Less Than More Than Total 15 Acres 15 Acres Number of Farmers Tamale District 2,500 405 2,905 Yendi District 1,200 118 1,318 Walewale 1,000 46 1,046 Other Districts 700 98 798 Total 5,400 667 6,067 Area Under Rice Tamale District 16,000 18,800 34,800 Yendi District 5,600 4,800 10,400 Walewale District 4,700 1,700 6,400 Other Districts 2,200 4,200 6,700 Total 28,500 29,500 58,300 Source: Ministry of Agriculture, Economics and Marketing Division, Accra, mimeo, 1972. increase of action in 1 from 10,003 of 480 perm Table Year The Pmd“¢*£l(3: 12 increase of 80 percent over six years. The increase in total annual pro- duction in the Northern Region is therefore estimated to have increased from 10,000 metric tons to 58,300 metric tons of paddy, or an increase of 480 percent over the six-year period 1968 to 1974. Table 2.3. Estimated Acreage and Rice Production in Northern Region of Ghana Between 1968 and 1974 Year Acres Average Yield Total Production Per Acre of Rice (lbs./acre) (long tons) 1968 28,000 800 10,000 1970 52,000 960 22,300 1972 65,000 1,100 ‘31,900 1973 70,000 1,200 37,500 1974 90,000 1,450 58,300 Source: Ministry of Agriculture estimates. The major reasons advanced for the impressive increase in rice production in the Northern Region are: 1. Easy access to free unutilized bottomlands not requiring a great deal of clearing; 2. Increased imports of tractors and associated equipment in recent years for sale to private farmers and individuals desiring to engage in private custom plowing: 3. Subsidized selling prices of tractors and associated equip- ment with resulting low custom plowing charges and land preparation costs for tractor owners; 4. An increasing guaranteed floor price for paddy as established by the Government Rice Mills Unit; 5. I fertilizer; 6. I i-put subsi rice farms; Ministry 01 13 5. Increased availability of subsidized improved seed and fertilizer; 6. Artificially high financial returns resulting from high input subsidies; 7. Prestige associated with land extension and large individual rice farms; 8. Subsidized combine harvesting services introduced by the Ministry of Agriculture. Bottomland Production Practices Land Clearing Land clearing, normally done during the dry season between the end of January through March, is carried out for the most part by hand methods using cutlasses and hand axes. However, mechanized land clearing has been undertaken on many large farms. The Ministry of Agriculture (MOA) operates a land clearing service and charges ¢20 per acre for this service.4 Small crawler tractors with conventional blades are used to push down trees and to push them to the farm boundaries. Official estimates on the amount of acreage mechanically cleared are not available; however, it is believed that no more than a quarter of the total acreage under cultivation in 1974 was mechanically cleared. In 1974 the Ghanaian-German Agricultural Development Project, Northern and Upper Regions, provided the Ministry of Agriculture a grant for a new land clearing unit. -The grant included two D9 Cater- pillar crawler tractors, chain and ball clearing equipment, land clearing 4A private contractor has recently offered land clearing services at ¢40 per acre using larger machines equipped with front-end rake blades. ___,‘.. blades, gezsrall rice or: Other (:1 b! hand. ~2. . ’I "K'aj'fi :J'n. l4 blades, two flat bed trucks to transport the tractors and equipment, plus associated equipment. During the 1975 land clearing season the unit cleared an estimated 5,000 acres. Land Preparation Data are not available on the number of acres plowed by tractor, prepared by hand hoe or prepared by Bullock plow. However, it is generally believed that about 90 percent of the land area devoted to rice production in the Northern Region is plowed by tractor.5 For other crops in the region about 95 percent of the acreage is prepared by hand. In the Upper Region no estimates have been made, but it is probable that no more than 25 percent of the rice lands are prepared by tractor or bullock. Again,whi1e there are no official estimates, three-fourths of the farms that are plowed with a tractor are harrowed once and perhaps a third are harrowed twice. A few farmers harrow their fields three times. Large operators tend to harrow twice, whereas smaller farms are usually harrowed only once. During the 1973-74 production season there were about 300-350 private tractors operating in the Northern Region.6 Most private tractor owners engage in custom plowing and harrowing services. Private contractors charge six cedis per acre for plowing, three to four cedis . . . . 7 for first harrowing,and two to three cedis per acre for second harrowing. The 10 percent not plowed by tractor is upland rice. Some uplands aue mechanically plowed; however, most are prepared by hand hoe. 6My estimate was arrived at by reviewing sales records of the Tamale nachinery dealers and by interviews with MOA personnel. 7The real financial charges are, however, greater as it was found onoEH 09 ON 0.00m I 0.0H comm + co>oumEH OB om 0.00m I o.oa came 0 Hmcoauacmue 09 ca 0.0m I o.m moaniou + co>oumEH was on 0.0m I o.m comm + oo>ouosH was om 0.0m I o.m comm o HmoowuHcMHB mmB ma o.m I m.o comm + cm>ouoEH xooHHom ma o.m I m.o comm o Hmoowuwcmua xooHHom 0H o.H I m.o comm + cm>oumEH comm oH o.H I m.o comm o Hmooauacmua comm mouoo ow manna mo .02 mufim sumo ca moomm mowumo>umm mo cocoa: m am: 32398.». 53.23 8 a m W853. 338 condom coauozcoum an Iman ecu uom mumeomm mowm mo mHmemm m>fimomusm cmamwumuum woodman ......m canoe 1111' SQI IEC ‘1‘ Ill. I , when». .... r“; ‘- . ch 39 Selection and Training of Field Enumerators The Economics and Marketing Division of the Ministry of Agricul- ture assigned 15 field assistants to our study who had previously served as enumerators for the 1970 Agricultural Census. All had received Middle School Certificates and were between the ages of 20 and 36. In the first year of study, they participated in a 10 day training course consisting of instruction in the purpose and importance of the survey, interviewing techniques, practice with using survey forms, identification of seed varieties and fertilizer types, adminis- trative procedures,and practice in field measurement. The course was organized on the basis of lectures, discussions, field trips.and tests. Each participant was required to prepare his own Field Manual during the course and submit new sections twice a week for approval. Each trainee was also required to pass a final written exam and have his Field Manual approved. During this first year process, one of the assigned field assistants did not appear for the course, one trainee did not complete the training,and two did not pass. About half way through the survey, one enumerator was dismissed because he was sub- mitting false data. Preceding the second year of the survey, 22 young men with Middle School Certificates were interviewed for new positions, and five were invited to a l4—day training course along with the 10 enumerators who were employed during the first year. -The course was designed along the same lines as that of the preceding year. All but two of the 15 (one from the old group) passed the training course. Thus, 13 enumeratos were employed during the second survey year. DI marily grade. their The re bicyc: away : given 40 The enumerators were paid between 36 and 38 cedis per month, depending on their grade in the civil service, which was based pri- marily on years of service. But in order to be promoted to a new grade, an enumerator needed a good performance record. In addition to their salary, enumerators received a transport and travel allowance. The rate for transport depended on whether they owned or hired a bicycle, and a fixed nightly travel rate was given for nights spent away from post. Enumerators were supplied with survey forms, field books, writing supplies,and a carrying case. In addition, at the time of Field Measurement, the enumerators were assigned to teams and each team was given a measuring wheel and a portable angle board. Interviewing Procedures Enumerators were required to interview farmers and to observe field activities on each farm a minimum of twice a week. It was learned that most enumerators interviewed their farmers on an average of three times a week, though there was not a general pattern among the field staff regarding the place of the interview. At some times farmers were interviewed in the village, while at others the farmers were interviewed at the farm. Generally, however, the enumerators preferred to interview at the farm because there they were able to observe field activities and to obtain more accurate information about the number of laborers working on the farm. The frequency of farm visits and interviews was increased during broadcasting and harvesting activities in order to collect more accur- ate data on labor utilization and total production. The enumeration Inc in 16 2h ‘ mid 1i Jkk RIO a: ...O. 41 period started in May (plowing time) and ended in February (towards the end of the selling period). The basic survey form used for data collection was a slightly modified version of the Weekly Input—Output Record used by Spencer in his study of rice production in Sierra Leone.5 Information acquired during the interviews was recorded in an enumerator's field book and then transferred by the enumerator to this "Weekly Input-Output Record." In an effort to minimize recall problems, this information was concerned with activities undertaken one to three days before the interview. The records were collected from the enumerators the following week. Part I of the Weekly Input-Output Record was organized for re- cording field data about family labor. Data were obtained on the basis of: (1) field activity, (2) labor description (men, women,and children), and (3) field hours. Part II required recording data on hired labor. These data included: (1) field activity, (2) labor description, (3) hours of field work, (4) wage rates, (5) total expenditureSIand (6) pay- ments in kind (food). Finally, the purpose of Part III of the Record was to record purchases of inputs (excluding labor) and sales of paddy, and Part IV was a blank page provided for enumerators to elaborate on preceding parts, if necessary, and to report any problems they might have.6 5See Dunstan S. C. Spencer, "The Efficient Use of Resources in the Production of Rice in Sierra Leone: A Linear Programming Study." Ph.D. dissertation, University of Illinois at Urbana-Champaign, 1973. 6All Weekly Input-Output Records were turned over to the Director of Agriculture before the author left Ghana in order that the Ministry would have all the raw data. It was agreed that the Ministry would also receive a copy of this dissertation and any other research reports which might be produced. vie for ‘3 i d an. We: 1' I) d v; .. r 1 \ nu 5v a C . “WI : . o me at LJ 0 .0 ..lillll 42 Supervision of Field Staff All enumerators (except those posted to the Upper Region) were visited weekly on an undetermined day by the author or his counterpart for the purpose of checking and collecting the field records. The enumerators in the Upper Region were visited twice a month. During these visits, we discussed any queries resulting from previous sub- missions, as well as field problems that might have arisen during the week. Field books were regularly checked to insure conformity between them and the Weekly Input-Output Records, and, periodically, sample farms were visited by the author to confirm the data being collected and to show interest in the individual enumerator's farms.7 Upon collection of the Weekly Input-Output Records, the data were transferred to a "Primary Tabulation Form" in the project office by the author and his counterpart. Tabulation forms were maintained on a farmrby-farm basis, and purchasing and field activities were periodically cross-checked for consistency. If discrepancies were found, queries were discussed with the appropriate enumerator the following week. At times, an enumerator was required to reinterview a farmer regarding particular information submitted. The purposes of the Primary Tabulation Form.were to greatly reduce the bulk of paper that had to be handled for each farm and to serve as a means of supervising the enumerators. The preparation of the Form on large paper (12 3/4" x'lS 3/4") facilitated the review of individual farm activities and expenditures in that a single page contained data for several weeks and showed at a glance all farm activities and expenditures for those weeks. 71t was found that farm visits were good for morale of the enumer- ators, and farmers often requested such visits. fan an: ti: of ti 43 The Nature and Measurement of Input—Output Data The classification and description of resources used in rice production and the corresponding variables measured during the field survey are given in Table 3.2. The method of classifying field activi- ties involved in rice production and a list of individual activities under each classification are given in Table 3.3. To measure labor utilization, data were collected separately for family and hired labor and were recorded on the basis of field hours and labor description (men, women,and children). Since rest and eating time was found to be virtually impossible to estimate, the measurement of field hours was based upon the time the worker entered (starting time) and left the field (finishing time). For hired labor information on wage rates (peswas per day), total expenditures and the estimated market value of payments in kind were collected. For each purchasing activity, data were obtained on the basis of a description of the item, quantity, unit price, total expenditure.and date of purchase, while for each field activity data were collected on the basis of labor utilization and/or mechanization services (owned or hired) used. Total production estimates were determined by daily counts of bags of paddy. Information on the disposition of the crop was acquired by asking farmers the reasons for which bags of paddy left the farm (sale or various storage purposes). For the quantities sold, informa- tion was collected on the location of.the sale, type of buyer, selling costs involved, unit selling price, and gross income. To estimate farm size, each farm was measured after harvest by the triangulation method by which enumerators,working in teams, drew a map of each farm, depicting the shape of the farm and how it was on. filo 44 Table 3.2. Classification and Listing of Activities Recorded on the Weekly Input-Output Record Purchasing Activities Field Activities Disposition Activities Mechanization services Land clearing Quantity sold; for each Hired Labor Land preparation sale: Seed 1 , f Compound fertilizer E OWIn? - ty?: 0 .buyer .. bigger... ““1 Tractors and equipment _ seed g gross income Bullocks and equipment Stored for later sale - compound Spare parts . . Stored for seed fertilizer Maintenance Fuel, oil, lubricants , Weeding - top dressing Stored fOr home consumption Gifts Harvesting cutting heaping threshing winnowing bagging Table 3.3. Description of Resources Used in Rice Production and the Corresponding Variables Measured During the Field Survey Resource Description Variables Measured land Acres Capital - Farm Produced or Owned - Purchased Seed: variety, quantity Power: days, area, variable input requirements Fertilizer and Mechanization Services: Description of item, quantity, price, total expgnditure Labor - Family: Men, women, children - Hired: Men, women, children Activity, field hours, estimated work done Same as family labor, plus wage rates, total cash expenditure, and payments in kind ~ q~w~ divil tri a: sure boar< 45 divided in triangles. The measurement of the base and height of each triangle was recorded on the farm map. The equipment used to so mea- sure the farms consisted of survey poles, a portable right angle board, and a measuring wheel. Data Tabulation, Coding and Storage Tabulation and Coding Initial data tabulation was undertaken during the survey in that data submitted on the Weekly Input-Output Record were continuously recorded on the Primary Tabulation Forms separately maintained for each sample farm. Upon return to Michigan State University, the data on the Primary Tabulation Forms were first tabulated and then coded for punching. All data were tabulated and coded in the same units as they were collected during the field survey. For example, the labor data were coded in terms of field hours and fertilizer data as 112 pound bags. The task of tabulation was broken into successive operations across sample farms rather than coding all data for each farm separately. It was known at the outset that tabulation would take a great deal of time because of the detailed labor utilization data that were collected. Therefore, it was believed that the tabulation procedures should be specialized with regard to the different types of data in order to increase the efficiency and accuracy of the tabulation process. 8The Primary Tabulation Forms were designed solely for the purpose of condensing raw field data and for checking the accuracy of field submdssions. They were not suitable for direct punching. As a conse- quence, a preliminary stage of tabulation was required before punching. c ...,“ (r 46 The tabulation of the labor data was very time-consuming. First, the labor data were aggregated on a field activity basis for each labor description category. For example, the total field hours of male family labor involved in the broadcasting of seed were aggregated along with female family labor and child family labor. In addition, the field hours for male, female,and child hired labor were separately aggregated along with the corresponding expenditures for each labor description category. The aggregated labor data were then coded on the basis of (l) farm identification number, (2) field activity code, (3) labor description code (family or hired labor),and (4) a sex code (men, women and,children). In summary, then, the tabulation and coding process involved several phases based upon the nature of the data. For each phase the appropriate data for all forms were tabulated and coded before pro- ceeding to the next phase. The phases were as follows: - total production; - mechanized land preparation; - purchase of seed and fertilizer; - labor utilization data for all field activities; - mechanized harvesting; - beginning and ending dates for all field activities. Standard Data File A Standard Data File was designed to organize and store data on tape for subsequent computer analysis. All data in the file were identified and organized by successive farm identification numbers and individual variable numbers. A Variable Code Book was developed for l J.- a\ i .a\ II\ III e 16 PL c CL t .I\ \ ‘1‘ h ‘rk AN.» ~ It AM i \H‘ 47 all variables which were expected to be used in the analysis. Each variable was given a number, name, and description. A large number of variables were defined, many of which were created from the raw data. For example, a computer was used to convert all labor data for each farm from field hours to man-hours per acre on an activity-by-activity basis.9 The same program was employed to calculate for each farm, on an activity-by-activity basis, the absolute number and relative per- centage of man-hours per acre for (1) men, women, and children and (2) family and hired labor. Mechanization expenditures for initial land preparation (plowing and harrowing) were re-calculated. Farmers who hired a private tractor operator were charged on the basis of an "unmeasured acre." During the field survey per acre charges were recorded as well as the total expenditure incurred by the farmer for each mechanization operation. Finally, at the end of the harvest season, all sample rice fields were measured. Using "measured acres," actual charges per acre were then calculated,10 and these charges, along with the total expenditure for each mechanized operation, were entered in the Standard Data File. Bags of seed and fertilizer were converted to pounds, and applica- tion rates for seed and fertilizer were then calculated. These appli- cation rates, in terms of pounds per acre, were entered in the Standard 9The coefficients used to convert field hours to man-hours were 1.0, 0.75, and 0.50 for men, women, and children, respectively, for broadcasting of seed and fertilizer, cutting, and heaping. For weeding, threshing, and winnowing the conversion factor for women was 1.0, or equal to men. 10After measuring all sample farms and randomly remeasuring about 20 farms to confirm our measurements, we found that rice farmers over- declared their acreage by about 30 percent. Custom tractor operators are probably in part responsible as it pays them to over-estimate the amount of plowing done since they charge on a per acre basis. 48 Data File. In addition, variables were created in terms of pounds of nutrients from the two fertilizers used by rice farmers, and these application rates, in terms of nutrient pounds per acre, were entered in the File. Principal Features of the Sample Farms Excluded From the Sample During the 1973-74 production season input-output data were collected for 161 rice farms. Of these, 143 farms were retained for analysis. The following 18 farms were excluded from the sample. 11 bullock farmers who hired tractor services for initial land preparation.11 5 bullock farmers who had mixed stands of paddy,12 1 tractor hire farmer whose farm was completely burned before harvest, - l farm inadvertently excluded. The 11 bullock farms were excluded from analysis because bullock power is limited to an upland rice production system, whereas tractor power is representative of a bottomland production system. It is assumed that land preparation by bullock and by tractor power is not the same, and, since the soil types are different for upland and bottom- land paddy production, these 11 farms were not consistent with either production system. Given the variability in seed variety and fertilizer llThese farmers did not intend, at the time the "Sample Identi- fication Form" was administered, to use tractor services for initial land preparation. 12These farmers indicated that they would plant a pure stand but planted other food crops in the stand after paddy had been planted. '18 to 49 use among these farms, the subsample of 11 was judged to be too small to be analyzed separately. The five bullock farms with mixed stands (paddy mixed with other food crops) were excluded because the production system was not typical of the study area and the subsample was too small. Sample Characteristics It will be recalled that five criteria were used to design a stratified, purposive sample based upon farmers' intended production practices; however, not all farmers followed their intended practices. Table 3.4 shows the number of sample units within each strata for the sample. The table not only indicates how many sample farms used each power source, seed variety, fertilizer,and method of harvesting included in the study, but is also organized to illustrate the distribution of sample farms among four bases of stratification. For example, of the 83 farms that hired a tractbr for initial land preparation, 28 used traditional seed (12 of these used compound fertilizer and no top dressing, and six harvested their crop by hand). Of the remaining 15 farms using traditional seed, in this power source group, 13 used no compound fertilizer and tOp dressing and eight harvested by hand methods. The sample distribution of farm size and power source for land preparation for each acreage classification is reported in Table 3.5. The table also shows that the total acreage of the 143 farms was 4,520 acres. If the total acreage of the bullock farms from the Upper Region is subtracted from this total, then the acreage in the sample from the Northern Region is 4,504. During the 1973-74 production season, it was estimated that there were about 70,000 acres in the Northern Region. .1 .1 1'3“ 3 . \u ‘1‘ it» 5() <- Table 3.4. Principal Features of the Stratified Sample of Rice Farms 8/ b/ Power Seed No. of Fertilizer Use— Method of Harvesting- Source Variety Obser. Compound Top Dressing No. of Hand Combine Other Obser. (No. of Observations) Hand' Traditional 4 X 0 l l -- -- 0 0 3 3 -- -- ( 4) ( 4) ( 4) Bullock Traditional 16 X 0 8 8 -- —- 0 0 8 8 —- -- Improved 5 X 0 2 2 -- -- 0 0 3 3 -- -- Mixed 2 0 O 2 2 -- -- (23) (23) (23) Tractor Traditional 28 X 0 12 6 -- 6 Hire 0 x 3 3 -- __ Service 0 0 13 8 -- 5 Improved 44 X X 12 4 2 6 X 0 27 20 2 S 0 0 5 3 -- 2 Mixed 11 X X 5 3 -- 2 X 0 2 l -- l 0 0 4 1 -- 3 (83) (83) (49) ( 4) (30) Tractor Traditional 10 X X l -- -- 1 Owners X 0 7 -- -- 7 0 X l -- 1 -- 0 0 l l -— ~- Improved 19 X X 9 2 5 2 X 0 5 1 2 2 0 X l -- -- l 0 0 4 l -— 3 Mixed 4 X X l —- -- 1 X 0 2 —- -- 2 0 0 1 l -- -- (33) (33) ( 6) ( 8) (19) Total 143 143 82 12 49 2/"X" - Fertilizer Used "0" - Fertilizer Not Used. b "/Hand: Farm entirely harvested by hand methods. Combine: Farm entirely harvested with a self—propelled combine. (3ther: Combination of hand and mechanized harvesting methods. Ta 51 Table 3.5. Sample Distribution of Farm Size and Source of Power for Land Preparation No. of Range in Average Total Power Source for Farms Farm Size Farm Size Acres Land Preparationl (Acres) H B THS TO 45 0.1 - 5.0 2.5 111.6 4 22 18 l 50 5.1 - 20.0 11.6 577.5 - 1 41 8 35 20.1 - 80.0 39.6 1,386.4 - -— 24 ll 7 80.1 - 150.0 92.8 649.5 - -- -- 6 150.1 + 299.2 1,795.2 - —- -— 6 143 4,520.2 4 23 83 33 1 Codes for the Power source for initial land preparation: H = Hand B = Bullock THS = Tractor Hire Service T0 = Own Tractor Therefore, 6.4 percent of the estimated total paddy acreage in the Northern Region was included in the study sample. Given the estimated number of rice farmers in the region (6,100), the sample of 143 farmers represents about 2.4 percent of the region's rice farmers. The inclu- Sion.of a disproportionately large number of tractor owners (33 farms, or 23 percent of the sample) in the study explains why 2.3 percent of the rice producers account fOr as much as 6.4 percent of the estimated regional acreage . With the introduction of the combine harvester and the large number of tractors in the study area, there were several combinations 0f harvesting methods used. Table 3.6 describes how paddy was harvested OD‘Ume sample farms. All of the 27 farmers who used either the handhoe or the; bullock plow for initial land preparation harvested their crOp 3.2.N 5...; AJWANE-Nuu H H< \AA4 absezxy 552.“. d mnmu>1~ .uv- hunt. hufivusnsfivz u m.» o R. my NAN-raps» .Hoczo uouomue n 09 “oofi>uwm mua: Houomua u was “Houmuomo xooaasm u on «comm u m 52 m .mcflzmmunu Houomua u 99 “Hmnmmwnu >Hmcofiumum m mm com: wcflnfiou n Emu «mawnfioo >3 omumm>uwm n 00 “was; >3 ooumm>uom u uma ...... mm m w 3.2 m... mm mm nu: nu: H u m.vm v.0 H Hm 1:: mm --- u- H m.oH 5.0 H HH mm Hm . nu: H u 0.04 8.0 H nu: on as nu: m n v.60m : m.~m v.¢mH ¢.H m mm as nu- on a- m m.vv n v.Hv m.mv v.H m nu: --- mm mm H H 0.48 a m.m¢ m.Hm v.H m mm --- vH Hm H H m.m0H u m.mv ~.ss v.H m Hm 6H we 5H m H m.mm u m.om o.cu H.~ m nun me :1: mm m H o.mmH u H.mm H.mm m.m v vm In: 1:: we m s m.sm . «.5 ~.om m.m m OOH m mH H.m¢H : m.m v.s~ o.vH om 00H H H m.soH u m.ov m.eOH .v.H m 00H m w s.mm¢ u o.m m.HmH v.m NH 00H 6 me >.~m u ~.m m.~H m.mm mm 00H 1. nu mm v m.0H n m.o h.H m.mH hm Anamouom. . as _ emu _ oo _ um 09 was on m Ammuoav Ammuoav 8 .oz «scrum: comm mcfimo wouzom umsom Huonumz sums sh sh emHMHmmmHo muHm sums muHm sums omumm>umm Show mo coauwomowm . maumm mo Hmnfisz cw omcmm ommu0>¢ mfiumm magma mHmEmm Had an com: mcflumm>umm no worse: .m.m mHnme 53 by hand methods. The farm size in this subsample ranged from 0.3 to 10.5 acres and averaged 1.7 acres. There were 55 farms (39 percent of the sample) utilizing tractors for land preparation that also used hand methods of harvesting exclusively. The average size of these farms was 12.8 acres, which is small compared with the size of farms which used tractor power for land preparation and combine harvesting. The average size of a farm on which only a combine was used for harvesting was 131.3 acres. Of the 12 sample farms in this category, eight were operated by farmers who owned tractors. On two farms, a self-propelled combine was used as a stationary thresher, all other harvesting activities being done by hand (i.e., cutting and heaping). On 20 farms (14 percent of the sample)"tractor threshing" was used for all.threshingactivities. The other harvesting activities (i.e., cutting, heaping, winnowing and bagging) were done by hand. The average size of farms in this group was 27.4 acres. The remaining 27 farms (19 percent of the sample) used various combinations of harvesting methods as shown in Table 3.6 which indicates the percentage of the total farm acreage harvested by each method within each harvesting classification. For example, the nine farms which used a combination of hand harvesting (HC) and tractor threshing (TT) harvested, on the average, 46 percent of the acreage completely by hand. On the remaining 54 percent of the acreage farmers used a tractor for threshing, although on this latter acreage cutting, heap- ing,winnowing and bagging were done by hand. In general, as farm size increases, the degree of mechanized harvesting increases. 54 Defining the Production Systems Since sample farmers did not in many cases follow intended production practices, we ended up with too few observations for many of our intended strata. As a consequence, we were forced to redefine our production systems on the basis of only two criteria: (1) power source for initial land preparation and (2) seed variety. These two criteria were used to delineate six rice production systems for the analysis: (a) (b) (c) (d) (e) (f) System I: THS - Traditional Seed This subsample consists of 28 bottomland rice farms where farmers hired private tractor services for initial land preparation and used traditional seed varieties. System II: THS - Improved Seed This subsample consists of 44 bottomland rice farms where farmers hired private tractor services for initial land preparation and used improved seed varieties. System III: THS — Mixed Seed This subsample consists of 11 bottomland rice farms where farmers hired private tractor services for initial land preparation and used a combination of improved and tradi- tional seed varieties. System IV: TO - Traditional Seed This subsample consists of 10 tractor owners who produced rice on bottomlands and used traditional seed varieties. System V: TO - Improved Seed This subsample consists of 19 tractor owners who produced rice on bottomlands and used improved seed varieties. System VI: BO - Traditional Seed This subsample consists of 14 upland rice producers who used traditional seed varieties and a bullock team and plow for land preparation. 55 Summary The purpose of the field survey was to obtain farm level data in order to estimate the financial costs and returns of the major rice production systems in current use in Northern Ghana. The Cost Route Survey Method was used to collect input—output data from 161 farmers during the 1973-74 crop season. The sample size was determined by a fixed budget, which per- mitted the hiring of 15 field enumerators, and by the number of farmers that a field enumerator could effectively interview. Enumera- tors had Middle School Certificates and were between the ages of 20 and 36,and all participated in a lO—day training course before the survey year. The training course concentrated on the purpose and importance of the survey, on interviewing techniques, on practice in using survey forms, on administrative procedures,and on practice in field measurement. Each enumerator was required to prepare his own Field Manual and to pass a final exam. Enumerators were also required to interview farmers a minimum of twice a week. The frequency of farm visits and interviews was increased during broadcasting and harvesting activities in order to obtain accurate data on labor utilization and output. A Weekly Input- Output Record was used to record farm data. Furthermore, enumerators were visited weekly for the purpose of checking and collecting input- output records. Upon collection of the field records, the data were transferred to Primary Tabulation Forms in the project office which were maintained on a farm-by-farm basis to facilitate supervision of field staff. Field data were periodically cross-checked for consistency. 56 Labor utilization data were collected on an activity-by-activity basis, separately for family and hired labor. Data were recorded on the basis of field hours and labor description (men, women, and children). For hired labor, information regarding wage rates, total labor expendi- tures,and the estimated value of payments in kind were also collected. Total production estimates were determined by physically counting bags of paddy on a daily basis. To estimate farm size each sample farm was measured after harvest by the triangulation method. Finally, five bottomland systems and one upland rice production system were defined on the basis of (1) power source used for initial land preparation and (2) seed variety. Three bottomland systems were based upon farmers hiring private tractor hire services (traditional, improved, and mixed seed varieties); two bottomland systems were based upon tractor owners (traditional and improved seed varieties); and one upland system was based upon farmers using traditional seed varieties and a bullock plow for land preparation. CHAPTER IV A FINANCIAL AND ECONOMIC ANALYSIS OF THE MAJOR RICE PRODUCTION SYSTEMS IN NORTHERN GHANA Introduction There are two principal objectives of this chapter. The first is to estimate the private costs and returns to rice farmers using current production technologies. Towards this end, rice enterprise budgets were developed for five tractor mechanized bottomland produc— tion systems and one upland rice enterprise system which used a bul- lock team for land preparation. The second objective of the chapter is to estimate the economic costs and returns of the major rice production systems from the point of view of the national economy. Distinctions Between the Financial and Economic Analysis Financial Analysis The financial analysis is based upon budgets for each of six rice production systems. All budgets were constructed from survey data drawn from a subsample of farms for each system. Moreover, each budget was developed by deriving mean farm estimates for (1) input quantities, (2) factor prices, and (3) physical output. Factors of production were priced or valued at 1973-74 market prices, namely, the prices actually paid for mechanization services, fertilizer, labor, etc. For the five bottomland production systems, output was valued at the floor price as established by the Government Rice Mills Unit (RMU). 57 58 For the upland system, output was valued at the average price which bullock farmers received for their output.1 For each enterprise budget a net cash return was computed for operating capital, family labor,and management. The budget data were used to derive financial returns to (1) family labor, (2) operating capital expenditures,and (3) management, as well as cost of production. Sconomic Analysis The factor prices of all resources used in rice production in northern Ghana, except hired labor, contain subsidies. As such, sub- sidized factors were not priced nor valued in the market at costs which reflect real scarcity values. To eliminate the factor-price distor- tions, subsidies were estimated and market prices were increased by the amount of the subsidy to arrive at real economic resource values, or unsubsidized costs. Two types of factor subsidies were computed in our analysis: implicit subsidies, resulting from an overvalued exchange ratez, and explicit or budgetary subsidies. After these subsidies were computed, 1About 75 percent of the total output of bottomland rice producers was sold to the RMU at the floor price of ¢12.00 per 180 lb. bag. Vir- tually all of the output of farms greater than 50 acres was sold to the RMU. Farmers with less than 50 acres sold part of their output to prie vate traders at prices ranging from ¢l3.00 to ¢15.00 per bag. The upland bullock farmers sold all of their output to private traders at an average price of about ¢l4.50 per 180 lb. bag. Private traders were purchasing for the small-scale millers who cater to the local market (as opposed to the Southern market for the RMU). In addition, private traders were purchasing paddy for neighboring country markets (illegal trade) in producing areas where the RMU did not have buying stations. 2The official exchange rate is GH¢1.15 = US$1.00. It is estimated by the Ministry of Finance and Economic Planning and the IBRD that the Ghanaian cedi is about 35 percent overvalued. The shadow exchange used in the economic analysis is therefore GH¢1.55 = US$1.00. the) pbw: 0" I used tior out; 151; POPE 59 they were added to financial costs to arrive at economic factor costs. Physical output was valued at its estimated import parity price.3 A Financial Analysis of Six Rice Production Systems Calculation of Budgets This study is the first attempt to quantify the physical resources used and the corresponding financial costs and returns of rice produc- tion systems in Northern Ghana. In order to estimate costs and returns of rice production, survey data were employed to derive enterprise budgets for the six rice production systems. Sufficient detail was incorporated in the budgets (1) to estimate physical and financial resource requirements on an activity-by-activity basis in order to compare resource use among activities and production systems (e.g., labor utilization in harvesting), and (2) to derive financial returns to selected factors of production (e.g., returns to family labor). Aggregation of Inputs and Factor Costs For each sample farm, specific resource quantities and costs were . . . . . 4 estimated on a per acre ba51s for each field act1v1ty. For each of the six production systems, mean acre input quantities and expenditures 3For the calculation of the import parity price of domestic farm output, see Appendix I. The import parity price is estimated taking into account projected world rice prices, domestic milling and trans- port costs, and the shadow rate of exchange. 4All factor costs are determined.on a per acre basis except seed and fertilizer and combine harvesting which are both calculated on a per unit bag basis. Seed is sold in bags of an average weight of 160 pounds; fertilizer is sold in 112 pound bags,and combine services are charged on the basis of 180 pound bags harvested. The average cost per 160 pound bag of seed was calculated among all sample farms and the re- sulting average cost was used in the budgets. The MOA fertilizer price and combine harvesting charges are the unit prices used in the budgets. The average cost of hiring a combine as a stationary thresher (CST) was VET! fan thax far: an rat} $111: one SUP] 903. [5”]? set and for 9911 tio tha wet bud ago the (The aSs Cos 60 were computed as a simple average of the individual farm means of all farms of a given production system. Thus, mean farm estimates rather than mean acre estimates were used. This procedure weighs individual farm resource requirements within each production system equally. As a result, for each production system an average farm budget is developed rather than an average acreage budget because the objective of this study is to estimate average farm resource use rather than to identify one or two production systems which would provide the greatest short-run supply response.5 This weighting procedure also permits the analyst and planner to identify farm level trade-offs among often conflicting goals of increased farm (1) output, (2) income, and (3) employment. Derivation of the Costs of Land Preparation AmonggTractor Owners and Bullock Operators The costs of land preparation for farmers hiring private contract services were treated as an operating expenditure item.6 The owning and operating costs of a tractor and associated equipment were estimated for an average tractor owner in Northern Ghana. The capital stock in equipment was reduced to a capital flow,and the cost of land prepara- tion has been expressed in terms of costs per acre:7 The resulting cost an average cost paid by sample farmers as our survey data revealed that the fixed price of ¢0.6O per bag was not accurate as many farmers were charged above the fixed price. 51f supply response were the sole objective, then average acreage budgets as opposed to average farm budgets would have been derived. 6Private contractors normally charge farmers on a per acre basis. Farm measurement revealed that private contractors over-declared acre- age by about 30 percent. During the process of tabulating and coding the farm data, per acre contract charges were adjusted by actual (measured) farm acreage. 7The derivation of the owning and operating costs of a tractor and associated equipment in Northern Ghana and the conversion to average costs per acre are reported in Appendix C. 61 peracre was operating capital costs which include an estimated value of the capital stock consumed or used during an average year over the life of the equipment. Similarly, the cost of owning and operating a bullock team and plow in Northern Ghana has been computed, and the derivation of the costs per acre for bullock plowing is reported in the . 8 Appendix. Derivation of Land Clearing Costs Since land clearing is a required activity for farmers establish- ing new rice farms, or extending their present farms, this cost was included in all budgets. During the 1973-74 production season about 30 percent of the sample farmers were engaged in land clearing. In all cases the clearing activity was for farm extension and not the establishment of a new farm. The estimated cost of land clearing by hand methods and by a combination of machine and hand methods was derived from sample data. These average per acre costs were amortized over a five-year period.and the annual cost has been included in the budgets.9 Costs and Returns of Six Production Systems The major features of each rice production system are briefly described in this section. The reader desiring more detail about a production system is referred to the enterprise budgets presented in the following pages. 8See Appendix D, "Calculation of Land Preparation Costs for Bullock Operators." 9See Appendix F, "Estimated Land Clearing Costs Per Acre on Bottomland Rice Farms in Northern Ghana, 1973." 62 System I: Farmers Hiring Tractor Services and Using Traditional Seed A financial budget representing the average physical resource utilization and corresponding expenditures for 28 sample farms using hired tractor services for land preparation and traditional seed is reported in Table 4.1. The average sample farm size for this produc- tion system is 12.8 acres. Pre-Harvest Activities The farmers in this subsample undertook,on the average, first harrowing and second harrowing on about 90 and 40 percent of their acreage, respectively. There was virtually no third harrowing under- taken. Traditional seed varieties were applied at slightly above the recommended rate of 70 pounds, or 0.44 bags per acre. The mean appli- cation rate of compound fertilizer was only 42 percent of the recom- mended one cwt. bag per acre. Harvest Activities Most harvesting activities were done by hand methods. Of the 10.3 acres cut by hand, 6.3 acres were threshed by hand, and a hired tractor was used for "tractor threshing" on 4.0 acres. In addition, 2.5 acres were completely harvested with a self-propelled combine. Labor Utilization The mean labor utilization per acre in all field activities was 116 man-hours per acre, of which 24 percent were fOr pre-harvest activities and 76 percent for harvesting activities. Thirty-eight percent of the total farm man-hours were hired labor; 81 percent of the hired labor were employed for harvest activities, and 14 percent were hired for weeding. 63 Table 4.1. Rice Enterpri 1se Budget for a 12. 8 Acre Farms need on Su urvey Data from Nanty-eight Farms in Northern Ghana I Using Traditional Seed Varieties and Tractor Hire Serv1ces. 1973-74 Sys sent 1 ActiviVy 1 Operating Expernitures and Labor Utilization by Activity Item Acres ‘ Non labor Expencutures labor Utilization and Expenditures Units Rate Total Cost Total anhouz‘s 383.3,: Expenditure Per Units Per Cost . l d ———_—P-- Total Acre Unit Per Total Fan-11v H M Per .r Activity labor Labor Hour Acre Acre : ¢ -——-————-——————— c -— ¢ —— ¢ — A. land Clearim‘y 12.8 acres 1.00 12.80 . PI ~f’tfl”st B ‘1‘ r;: 12.8 acres 9.00 115.20 lst rianwir» 11.6 acres ".44 21.50 2nd ltm'ywirg 5.3 acres. 4.16 12.05 Harms-11v 0.7 as »s 2.37 1.66 Seed 5 “0031 8 us 5 a 1:1 00 131021) 9 u 120 3 96 0 2b.} 0 1o 13 2 20 2. b s . . -.. . . . . _ -. Cor-1p. Fertilizer 12.8 mags 0.4? 5 u 2 80 10.80 1.; 13% 1;; ii; 33% .33 1.12 2 1 .uo 0. . . . . . . ‘5'”? 57mm *3? m... '09 l 2 00 2 1u.o 179.2 112.7 66.5 0.15 .7§ 9.73 1.: 0991...; L1'5 2.2 28.2 17.9 10.3 1.37 1 as 29.00 and ugmjfib—Tocai _. 913.130 ‘27.?! 355 9 2E9.7 1M2 5.36 3.00 3839 C. Harvest 1. Mechanical (1W ‘ ' .0 1.00 13.00 "41:11? :5 gigs 13 0.2 0.5 0.0 0.5 2.50 0.50 1.25 .1 ' , . . Tractor 11.0 acres a a 0 25 0 2’ 1.31 . W 14.09 16.36 4.5 18.0 1..8 5_.: . ., A’WSSt-Tl-Jtal 5). 53 15.5 12.3 5.7 L .56 Hand 1 '1 6.98 “ 114.1 “54.2 195.7 158.5 3.1“ 3.59 3 [0:11.15 1(0)} gaff: 2u.1 248.2 1117.2 mm 0.12 1. 2:: 12.77 'H'; J! “STEM 6.; 31;”;5 27.9 175.8 151-9 23-9 0.17 0. 66 3.97 Wimxmliru; & 10.3 ‘ acme 1 8 C 22.5 231.8 171.0 59. 0.09 0.52 . malt-70ml 109.51 1113.0 653.3 “6.2 0'1 5239.0 3. - o 50 33.30 sub- smmwliigmeflm 62.66 88.29 1‘812 :5 676'6 W13 0 1n n 62— 61.64 D. Total Expenditures filizatli‘q 36027 116.0 1559.4 920.3 558. l 0.10 7.52 100.0? Incane and Eggnditure Sm. SW! 0" Income 51mg of Eggnditums fl A diturca a. Totjl Production a. Non labor Expen bags/3cm x '12. 0 acres = 66.6 1. 1.3m! Clearirg ‘ 1:283 U. b. Value 01‘ llr'oduction ,) 3.15:2: 3223121111. 91”“) 6 611:1: 00 S c 799A 14. Mechanical Harve sting 29.33 c 360 27 c. less Total l-‘am mums 1160.70 5. Bags 32.} . (1. Farm Gate income 338.90 b. labor Expeniitur'es f/ l. Pie-harvest Activity 35.39 e. Less Estimated Selling Costs— 12.28 2 es: 7mm”. 61.64 g 100.0” E 1, 1‘. [Set mfitfloiorfiief;ili:finiapi a c 31.2.22 c. Total Pam Expenditures C 160.70 The cost of land clearing is the average annual amortized cost. See Appendix F for the calculation of land clearing costs. b -/'Tractor Threshing" involves driving over paddy several times with a tractor. s/’l'he 109.5 total manhours Cper acre is the labor requirement for 10.3 acres cut by hand. A weighted average man hours per acre (0 or one sub-a titvl ty, threshing, is included in thist tto t.al A weighted average is ed cause the “activity acres' for hand threeh1nq is not equal to the 10. 3 acres cut b han of 10. J acres of threshing is 18. 9 manhours which is based upon 6. 3 acres threshed by hand and 4.0 acres of" tractor thresh hing. c-l/Averav.;e total manhoura per acre used for mechanical and hand harvesting activities on l2. 8 acres. g/mtal expenditures per acre for labor used for hand and mechanical harvesting activities (¢61. 64 divided by 11.3 acres). SlCost of transporting paddy from {an to market at £0.30 per bag. s-/Cc:st of one bag is £1.00 and assumed to last two crop seasons. — av: «1 W6 ‘10 64 Costs and Returns The mean farm expenditures for this production system were ¢460. Land preparation was the largest expenditure item, accounting for 41 percent of the total farm expenditures, followed by hired labor (22 percent), and seed and fertilizer (21 percent). The mean yield for this system was 5.2 bags, or 936 lbs. per acre, the lowest of the six systems. Total production was 66.6 bags, or 5.35 metric tons. The gross income was ¢799, and the net return to operating capital, family labor,and management was ¢319. §y§tem II: Farmers Hiring_Tractor Services and Using Improved Seed A financial budget representing the average physical resource utilization and corresponding expenditures for 44 sample farms using improved seed varieties is reported in Table 4.2. The average sample farm size for this system is 21.2 acres. Pre-Harvest Activities First harrowing was undertaken by these farmers on virtually their entire farms and about half the farms were harrowed a second time. For all practical purposes, no third harrowing was done. Seed Fflas applied at the recommended rate for improved seed varieties, and tlie mean application rate of compound fertilizer was 1.27 bags per a~‘531I1‘e:as opposed to the recommended rate of 2.0 cwt. bags for improved SEEKS varieties. On the average, farmers applied ammonium sulfate as a tor> dressing at a quarter of the recommended rate. "" a.<““ Table 4.2. Rice Enterprise Budget for a 21.2 Acre Farm Based on Survey Data {rm Forty-Four Fans in Phrthern Giana Using Inpruved Seed Varieties and Tractor Hire Services, 1973-74 (System 11) and Labor Utilization by Activity “~3'."; Operating Expenditures 1' Mr 1:3" Labor Expenditures Labor Utilization ant. Expenditure ..Ln'r. . \ '..: II I ._ _ th" 33;. “Ural Frat Total Manhpurs Hare Expeniiture '” “‘ ‘. ' “" “' a Rfie Per UIJLS For post . . ‘ M Am” 1va it. For Total Pam 1y Hi nod Fe! . at .11 i 7' Activity Labor labor nor-2 Acre : ,3 c. c t ‘ W " - . u- 1 'i " A. I£JH UialPiW'E/ -i.p -NJ14 1.00 t1.-0 U. ITe—tuwrst V 1. T... . ~ um.» 9.05 191..-9 . Val, .40- - ‘ ‘ ‘ ‘ r “'vm-‘w‘w' 1. unz' 4.16 95.275 ‘n .A . A . .4 ‘ h A.‘ l ,. J”: "Lu." '51:; if...i ’i~"".." 9.‘- “to. H, ’Z‘i xi'u‘l‘m‘r-v' 1).." (“V‘s ' ’0.‘ j. 4: . . .. “a. v ...__ L. _-,, .x. r r . . ’ _. .1 ‘ .. u} 2 J1 M ~ tars 0 ‘4” ll 7) m :30 11.1 an.) 47.1: ~11 5 0.1:~ 3m x 1 7"“ -- . ‘57 n»?- . >7 a 75. 2,2 2.6 55.1 79.7 :35 u 0.4:. .35 l .1 mm. - .. - .« 1 “1 g 1n 5 12 7 0 a9 .66 4 'u I' ..‘ buff . . in. ', J 1.1.13 bl .’ u. ‘7. . ’ J . m 52’1“” 3.1 V. 13.7 NM 133.6 .4: i U-‘Jb 1.5. >'- ‘1 .111.| it‘s-f1. ,. )8 :lv;*,‘.5 {1.5 118:; :2”; 1:7 2-; .11. mu. 1”" .. ’ l - -.- asap)? 3;": '3". 5.5 377.3 5.1”? u w 1. 71,—) ,“J. .l..'-‘.-n71 , C. 7 ‘ Pg "D l ‘I 1'0" 9 "'7: " ‘ ' - -"_ V”" l l ' J l D".- 1,’< 3.0 2.0:} 1.173 331‘: I I I ‘I - w I.’ - If ‘\ L5" L L I“: 1‘ 16 1".) ‘.S -).2 (’13 1 a: 1.5“ .-.I" 'g’ ' “ ' ’ ‘ 1.2 2.0 0.0 5..) 0.4%: 3341 3.5:? a.) “ "I ’VH'“ ‘ '3 ‘r “I: 1 I, 7'}; 3.1 U,“ 3 3" 0.11 1': 0‘ "4‘ -‘\ as c. _. _ ' _ _T—' fi" —_h.- .— ‘ q' I‘M .'_.§ “0» il.l U.‘C/ ll-ll ‘7 i 2?.1 524.6 220.; ij.3 0.14 2,45 Lu.gy I . . A. r A]. ‘ . , i . .‘ . 1% " 1t.? 274.9 10%.1 lcy.7 O..5 ..52 iu.u2 33". 2a.?» man 176.3 A an o lo 1.39 :1.» . ... - —, 1 :m w" ‘1 an ‘ ' 1n.3 2".7 115.3 12..“ J J: l.v( .. H- ...-r43” ‘3.{ —- v———-._ r—M—r - - "Lv"~‘"“?'}" 5 l Eh‘ié was 2.2.0 sags, u id lam/7 [21)"."f" _ Cub—'1': 3.711 ’ , a Ekugs 0 EC 03.40 f/ .J.... ..u, 33 ,. , n/ —~-r . -‘_i- -v u nub ,H‘Ndljf'A l‘LJ‘Vt £1.16, l—ql? 7“ 72 1-. lr‘ii. (33.0 (105.3 0 13 it. U4 uxmxe._u\m .. b. "it". .; Lixpen ii: .: :3 3m 99—" :1; F3 ?71; m l_)___j m w T577 {arcr YIiLLJLiicfi Jud. ‘ ., . - Incwmlanifimpeniflzue firrnry "1.11:? »f In::;mte 81mm! of Exi‘eruitzu‘os i Toia‘ frol.c?13n a. Non Labor Expenditures ' ""r"c:'- " gin: amt-s = I‘M ‘ L” by)" 1 x A b 3‘ 1. mm sugar-mm; 2.90 ‘ ' ’ 7 ‘ r; "~«v , t: '« .Fl b. Viiu1t'Jl {TiKIJC'1LH} . ‘ .. riiii Vlff’u‘ftj‘l‘ :1, J 1-‘1 5. t )1 +1 93 = c1576.:u 3 seed a. remilizcr ..;;,.,ar . I‘ll-J v n a .1 n ' ) U. recnnnical Harvesting Q7.ub ’_ c. LLLC :ULli Echn Lxgqnditlres ?7§.41 5. igug; 6).,0 co»o..r ' ., —. . .. \ . .. 1. . 2:: d. Fm alsitr: LECU’I‘: 400.9r b Labor Apezhiituxxs -. . T "" " g r: " \ i “ ’ T c. Less Estimated boiling to 32.4t l rTe—hirvest nativity ‘(i.J/ ‘L 3. narvcst activity if7'91 61,- )b f.3ht pehxv.t0‘gergtht‘Capital, ‘ . A . . 6‘ [Yule-LN ...-L201" a“ PERI—.5ct": Del“ ‘ 2 C. A‘U‘UMLL" PW»: W ‘- “LUKE? 30‘2- - 9 The cost of land clearing is the average annual amortized cost. <=learing costs. 23/ CST represents a self-propelled combine used as a stationary thresher. E5a/ E/See mle 4.2. (mate 1:. 1/3e Table 4.1. W. 1:. total mnhour: 36.53cmcofthMrg s 5. ms for feedirg 21. 77 acres of and 36.5 for Ni: thresrad by hand and the pad acres of threshed dy 6. requimient 5 or 21. 7 acres of paddy threshed by a c The cost of land clearirg is the average annual amrtized cost. 1‘ acre 13 the labor mull-arch: for 36. 5 acres cut by hand. sub-activities, (i) threshixsg7 and (ii) winnowing and baggirg, axe incl SeeAppendix 1’. Unit costs of mechanized Land prepamtion are based upon cawuted mum an! apex-stir; costs of a tractor am associated equipment hemm See Appendix C, Tabiaz avenge mnhoun per acrere for two mnmursper 114. 9 acres or spaddy amine used u a stationary thre s/Avemge total mnmura per acre for mechnical and kind harvest“ activities on 1&1 6 acres yTUtll expenditure per act! for labor used for m and nechanical harvestirg activities (G 319.28 divided by “1.6 acres). B/;ivex'agecostoftranspcx‘tirgpaddyfrom{'amlnzoniarkiet:it00.301:at_-rt~.93. 71 Pre-Harvest Activities 0n the average, first harrowing was undertaken on the entire farm, and 60 percent was harrowed a second time. Traditional seed varieties were applied at the rate of 83 pounds per acre which is above the recommended rate of 60 to 70 pounds per acre for traditional varieties. Compound fertilizer was applied at 83 percent of the recommended rate of one cwt. bag per acre, and ammonium sulfate. which is not recommen- ded for traditional varieties, was applied at the rate of 0.23 bags per acre. As a result, the mean application rate of nitrogen was 19.2 pounds per acre which is about 14 percent above the recommended rate fbr traditional seed varieties. Harvesting Activities Of the mean farm size of 41.6 acres, five acres, or 12 percent of the farmmwas harvested with a self-propelled combine and 36.5 acres, or 88 percent of the farm was cut by hand. Very little hand threshing was undertaken. About 60 percent of acres cut by hand were threshed with a combine used as a stationary thresher; "tractor threshing" was employed for 35 percent of the acres cut by hand, and the remainder was threshed manually. Labor Utilization The mean labor utilization per acre for this system‘was 88.8 manrhours, of which 78 percent were used for harvest activities and about half were for cutting. Seventy-four percent of the total labor were hired labor; 82 percent of the hired labor were employed in harvest activities and about 10 percent in weeding activities. 72 Costs and Returns The mean farm expenditures for the system were €2,002. Land preparation accounted for 32 percent of the total expenditures, followed by seed and fertilizer (21 percent), hired labor (20 percent), land clearing (12 percent), and mechanical harvesting (nine percent). The mean farm yield was 6.5 bags, or 1,170 pounds per acre. Total production was 41.6 bags, or 21.73 metric tons. Gross income to the system was ¢3,245, and the net return to operating capital, family labor,and management was ¢1,l62. System V: Tractor OwnersUsing Improved Seed A financial budget representing the average physical resource utilization and corresponding expenditures for 19 sample farms on which tractor owners used improved seed varieties is reported in Table 4.5. The average sample farm size for this production system is 119.3 acres. Pre-Harvest Activities 0n the average, first harrowing was undertaken on the entire farm, and 82 percent of the farm‘was harrowed a second time. Twelve percent of the acres were harrowed a third time. Improved seed was applied at 0.53 bags, or 83 pounds per acre which is slightly above the recom- mended rate of 70 to 80 pounds per acre for improved varieties. Com- pound fertilizer was applied at 60 percent of the recommended rate for improved varieties, and the application rate of ammonium sulfate was about 40 percent below the recommended rate. 73 Table 4.5. Ritz Enterprise BLdget for a 119.3 Acre Pam Based on Survey Data frun Nineteen Farm; in lbrthern Giana Usmg Inproved Seed Varieties and Om Tractor and l-lquipmt, 1973—74 (systan V) ACT-Wit)? Opwratirnz Ext/arriitures and Labor Utilization by Activity Item Acres Non labor mi‘fljitms lab-er Utilization and Expemnums Unit 5 Rate Total Cost Tm al Harbours Wage Exp-end it 111"“ Per Uni t 3 Per Cost R3 t e Ac re Uni t Per Total Pami 1y Hi red Per PP? TO‘LBI Act. 1‘71 t 37 Labor Labor HO Aer“? AC re e i t — 6) Fr A. weir-aim? 119.3 acres 5.55 6.)»... B. Pre-harvvst Flown»? 110.3 3*. itu'mwirg 1139. 3 2nd Hm‘mw ing 97 . .8 3rd Kim‘mdrx, 6 . O Subfi'l'otal 2 la. 568.00 Seal 119.3 baits 0.5;“ 62.0 114.00 3.5 1183.3 155.0 398.3 0.16 .190 i47.7.." Gem's. Fcrt. 119.3 bags 1.90 195.: 3.80 £100.96 2.“ 28r.-.3 71.6 2124.7 1.2; .56 95.33 m. Sulfate 119.1, 0.39 73.9 2.80 1170.80 1.1 155.1 36.8 119.3 0.33 .52 33.1.; 1;: Heading 1127.3 7.5 906).? 155.0 751.7 3.1} 1.07 13.03 2110 91.97111}; 119.3 3.0 1387.9 35.9» 372.1 0.44 1.1 11.4}: Sub—Tomi 1402.7. 18.1 37.9.3 583.2 17"¢'u.1 o :2 1.35 331-.) C. Harvest l. Mectmdcal t‘lanim: 9.3.. bags 65, .2 1. 00 £53.20 t?6}ibb§ 99.0 5.; 9.8.u 215 3 263.1 0 in 0.07 87.‘~ «TS: ; 7.6.. 12333 574.0 0.80 113,3: 2.1 1.7.0 1.5 111.6. 0.19 C: '4 5.1.: 111.331.»: 7.6 l 1.9 19.9 0.7 13.7 1.31 2.3b 17.911 -. i'ICOI‘ ‘1, . ""r.:--~3.'ii.'tat‘—’ 6.13 { gen 5 3.215 27.99 2.1 l__-.."-' 1-9 11. 1.00 1.79 11.30 ;:.:.1r;_1bv.a1 . 7"”? 33.: :19." 113.? W 12’: >7 2. Harri (hitting 27.3 7 31.2 991.7 9.7.8 758.9 0.??? 6 2.? 160 31 11e:-.;1L~~.: 27.3 é.) uzu.1 101.1 3:12.11 0.35 «.10 811.8 11amcrlrg. 13.3 ‘ 19.2 18c.9 30.6 158.3 0.23 5.7% 5;.La Wirurrw‘iw: .‘1 31.1711; . . -. i. y. .i ‘ ' | H/ -1-r——- _1_? 1 ——,-T_-'W‘ bu’b-iot .11 r): . Er— .‘ _. O. 50 1153 . 50 '4 O N p—l :r b—1 ..AJ in .. ox A] (N l: r—l E! R . . C) r _. r. H *4 (I) M. I. 3. Bags Sub Total of Pamest 1er _'—‘T - , . —- _ _P—TT . “if—1" -T—‘ Expmiitums 1145.18 19,79 72:3,?) 7.80.1 1073.2 0.37 lec/ .50.:1 '7‘! 0' ‘1 ‘ ..' .J r . a. , 1‘. --::: .———_— .—r ._ - _. . _. if - D. .34.. Exp-.mitxmrs and 5170.13 fl. ‘ r—fl‘i. . .1 C {In} 3 j, $7.12 5.3.1.5; labor Utilimtim . . i , . _ Income 8111 frfimnditure Sammy 513171111 :7f‘ [7; 5171-; 31.17132; 01‘ i‘in4’flili?‘-b"?3 2;. Total Pr:‘xl:;c‘.icn a. N- n Labor l-bcr-eniitzires 7 7.1 l.fi:f.:;/ii.3:~:' x 11‘). -; acr‘vn’: = F1“. - J 5383 o {7' . a .1 J . 1am C 19511113.: 1am Frenzu‘atior'. 1 Seed 3. Fertilizer 1,909.76 :0» b. Valua- i’l‘fxl‘wT'l ion 614?.{1- brat:- x c 12.11? = c 10,161.00 3 J. c. 1133.: Tctal Pam E-‘xtwnditzimr: 5,979.99 14- Methylcal H’mxwtim: 74.00 5. Bags 4153.3”: it fx,l'-Y“..l’< (1. Farm Gate Inca??? '4"ng ————~ b . Int/or [ixgx‘rrzditunm e. Less E-‘stimated Sellim; Costs 39.10 1_ Wurm'vcst Activity 31.19.“ ’3 v '7 '1 “'9‘ I.“ '. ”til 37.". .1 A :4 “-3.7%, 1". HI". 8.5111; to Cte‘ratiw: tapital» “' tar." "V “A.1”... “—441 ‘ A" 7 ‘ Frill; incur am T-L'u'i:1.:sf=m:nt ‘ ‘4, 7.17.01 . - . . ' ~ [ . . m .. - i I '2'. - c. 1otal Pam rxuériiiamx. f g, 1 — I‘m: cost of land clezirimr is the average annual amortized cost. See Apron-31x F for the calculation of land clearing L . g/fixa unit cost of wrenmi'zm land preparation is based upon cmmuted ownim: and operating costs of a tructc * and associated «-"1111-.'nvnt in flyr'tirer'n Six-uh. Sm Arpeniix C, Table 2.“ E/Eee Table 4.2. footrntc b. yaw Table 4.1, footnote b. 9/ ’11.» 68.5 total murmurs per acre is the labor rum-liramnt for .:'7.3.acres cut by hand. A wigjted average rmii'xurs per 7133‘": fit-r '1-5' Elb-activitivs. f1) Circa-1mm and (ii) whmowim: and b'axrsimr. are 1.1-1.7111161 in this total. A weigf-‘t-fi avmn .2; is Lizwi m m "#3152117; there-.3" by hand imvestixg methods ”or these two Bib-activities are not equal to the 27.3 acres; 1:19. and heat-e: by turn. "Dir: "#31,".“9‘1 -.'.'v-."'1.*v- “Hummum per acre for 97.3 acres c“ tl‘ caning is 8.0 “WEOUI‘S which is base: 11 ..n 13.3. acres of mm threshing, 6.4 acres or~ tr’attcr firm. .‘7r- U157: mm the later requirqnmts for Yet-ding 7.6 acres. of maped paddy into a combine used a7?- a stationary tin-$21.02*. "Mo weigrrl .i‘"1;, ~551- rioqrs per acre for wizmnwir‘g and taggirg 27.3 acres of threshed paddy is 13.11 m'mnours. This is bridal 1mm. wlnr’vwir‘r 1:11 bylaw, 13.: 1'1"." 01‘ paddy threshed by hmrl and the margins? requirenmnts of 7.6 acres of paddy thres'rxad by a Ctlnffill‘if.‘ meal 31:: a static-5.111; EN‘CLZLNY‘. £/ verage total nanhours per acre and mechanical and mm harvesting activities on 119. '4. acres. a/Totzil expenliture per acre for labor for hand and mechanical harvestirg, activities (c £451.31 jivlimi t3: 151.3 '12:“: ‘1. . tn-‘ Har all 92 sta 9532 SEE 74 Harvest Activities This system was characterized by more mechanized harvesting than all the other systems studied. Of the mean farm size of 119.6 acres, 92 acres were harvested with a self-propelled combine. Of the 27.3 acres cut by hand, 13.3 acres were threshed manually, a combine as a stationary thresher was used for 7.6 acres,and "tractor threshing" was employed for 6.4 acres. Labor Utilization The mean labor utilization per acre for this system was 37.8 nan-hours which is the lowest labor utilization per acre among all of the production systems. Fifty-two percent of the total man-hours were used in harvesting activities and 28 percent for weeding. Seventy- five percent of the labor were hired labor; 50 percent of the hired labor were used in harvest activities, 40 percent for broadcasting seed and fertilizer,and 10 percent for weeding activities. Costs and Returns The mean farm expenditures among tractor owners using improved seed were ¢5,980. The largest proportion of total farm expenditures was land preparation which accounted for 32 percent of the total. This expenditure item was followed by seed and fertilizer (24 percent), .hired labor (14 percent), and mechanical harvesting (12 percent). The mean farm yield for the system was 7.1 bags, or 1,278 pounds Igor acre which is the second highest yield among the bottomland systems. Total production was 847 bags, or 68.1 metric tons. Gross income was ¢10,164, and the net return to operating capital, family labor,and management was ¢3 , 9 30 . 75 System VI: Upland Rice Farmers Using a Bullock Plow and Traditional Seed Varieties The vast majority of farmers in the Northern and Upper Regions produce crops on small, upland holdings cm? four to five acres. Among these farmers, rice is typically cultivated in pure stand on one-quarter to one acre of land. Most farmers producing upland rice are using traditional seed varieties and the handhoe to prepare the soil. However, in the Na, Navrongo, and Bawku Districts of the Upper Region and in the Yendi District of the Northern Region many farmers are utilizing bul— locks to prepare their crop land. The purpose of this section is to compare the costs and returns of bullock farmers producing rice on uplands with the five bottomland production systems. . The average size holding of 14 sample bullock farmers was 7.7 acres, of which 1.1 acres, or 14 percent of the holding was a rice enterprise. These farmers had bullock teams consisting of two West African shorthorns.‘ Also, among these farmers, the common implement was a small tool frame upon which a plow or a ridger is attached. The plow attachment is then used to prepare rice lands and the ridger attachment for other crops (e.g., sorghum, millet, and groundnuts). In the analysis which follows only the rice enterprise is investigated. A rice enterprise budget based upon 14 sample farmers in the Bawku and Yendi Districts using traditional seed and bullocks as a source of draft power for land preparation is reported in Table 4.6. Pre-Harvest Activities The 14 bullock farmers in this subsample plowed their rice farms with a bullock team, and of the mean farm size of 1.1 acres, 0.4 acres were hand-harrowed. A distinguishing features of this system is the - Q ~x; an“ 76 .wE .Sa and» S omega... 8 manna $328.38» oo 38 m .mcommmm no.8 03.. puma o» gamma urd 00:3 3 Mm: 650 .8 38> .m fictodg mom .umoo 85955 ESE vmegm as» 3 Waffle? 6:3 go umoo 28S 2.39 nonzflntaxm Emu 138. .o n| 22K 9 225332 Em .583 .325 Same mod .3232 98.5w: .m 4338 gauged 8 Esme $2 a fig 9 3233 pnwfifié A aim \mmumoo 9.53. Eugene 28 w mgfiucfid .893 A Esme 9:85 .38 E .u 8&2. was $8 .3 3.7... mgfifiofim E Egon. $3 0 5.: 52395 we Bum .m $8.; 9 n 8.3» x .38 Tm 56 5328.5 33 .m 8332 .8 .53, .n 24 e wfiéflo 13 A 3.8 Tm n megs... HA x 88x? m. a mfizfincii .523 E: .m 8388.5 H88. .m magical xm to m 8505 no “Em . m 23:: Em 98qu 1. _ 8 mm_ 8 om S 0 m5: fiSm m d; _ 50% 3.8 _ 833:3: 1!; H H I ll. - ,I III .893 us. 380 389 d mtm 8.0 \n mwmm d $5 Ed 2 o 33 team Tmom Heme 38.9.0. $6 $8 3.0 6.: in: 3mm .9? HA Wigwam _. . Em 565.53 26 3.0 35 in 0.5mm fimmm m 0.2m HA wing: EA 9.; 2.0 or. M 9mm. 9mm. , 9? HA wing: $6 sea. was . 3: _ TE 5.23 mdi H4 9:35 v . Long ummg .U 2.2 2.: F5 m Ti .. Ha: ASS RAH _ HSBLam Ea unfim , W: _ T: 1 Tom .. _ 1H 252% Em mo.m 9...... T: . Tum . Wm: _ a . H H 2:383 Em $6 mam PB _ 1% egg . .. _ HA ”.5683 92 | I .3 ...m mad em.m . 8.0 $6 1 mes H a .3238.» .88 l _ To me $6 8;: _ 8.0 _ $5 3.3 HA 80m 8.0 3.0 H; Tr. m 3. m R we m . 33:30. 8.0 Pro 35 H; a...“ 6 2 .28 m 3.0 wciofisi Hm: I I. mam m mm fem woe. 8... HA . may: a; mafia: _ ”3.52.6.5 m . u . 34 m 8; 38.. HA \m 8288 ES .< lu u Ill. u , _ . u l u . . . . 20¢ 8% .52: ~85 ~32 .3333. . . 135 .5 he . 68E 3E3 183 .5 ads ._ 22.. _ 35... unou Lou mug: . Lou . Expands .. _ $3854... H35 _ $8 E89 35.. _ 3?: awhxncmea can 538st LES mgzunvem .68: co: 35¢ Ed 3232 .3 8381.3: L83 new 8,3828%.“ waisted 3232.. 8, E395 Sauagm ES now .33.. .3033 65. $5ng comm 3833 053 «:55 50532 :4 mum—nah g 59¢ 38 E5 CREE Banding mow—what H. H a he owed wnnmuouafl 002 .w.v 0.33. FmIfO-m'c . .... ......I’IL... .. 77 high application rate of seed in that the mean application rate of the seed was 0.57 bags, or 91 pounds per acre which is about 20 pounds above the recommended application rate for traditional seed varieties. Compound fertilizer was applied at the rate of 9.83 bags per acre which is 17 percent below the recommended rate for traditional varieties. Labor Utilization The total labor utilization per acre in all field activities was 633 man-hours, the largest labor utilization of all systems studied. About 28 percent of the total man-hours were used for pre-harvest activities, including five percent for land preparation and 72 percent for harvest activities. About 70 percent of the total labor was used for three activities: threshing (33 percent), cutting (22 percent), and first weeding (15 percent). Unlike the other production systems, three separate weeding activities were undertaken here. The total labor-use involved in the three weedings was 169 man-hours per acre (27 percent of the total labor-use) which is more labor utilization for weeding than any other production system studied. An explanation is required for the higher labor utilization by the upland system as compared with the bottomland systems. An upland rice production system requires more weeding than a bottomland system because the uplands have been continuously cropped over a longer period of time. As a result, soil fertility is lower which is more . 0 conducive to weed growth than a newer field, or newly cleared field.1 10In fact, some bottomland rice producers have abandoned their farms after three to five years because of weed infestation. It is known that soil fertility on these farms at the time of abandonment is low principally because many farmers did not apply fertilizer until the second year when it was observed that fertility was declining. 78 In addition, given the small size of the holding, farmers may believe that they can better cope with weeds than the large-scale farmers who do not have much hope of hand-weeding their entire farm well. Upland farms require more man-hours in cutting for two reasons. First, traditional varieties are tall-stalked and susceptible to lodging if fertilizer is applied above the recommended rate. Second, many upland farmers do not use the sickle for cutting the entire stalk. Rather, many either cut bunches of paddy with a long knife or out only the pinnacles which contain the grain and not the entire stalk. Farmers who use this method argue that it requires more time than using a sickle, but higher recovery is achieved because less shattering occurs. Farmers utilizing traditional seed varieties are also likely to require more labor for threshing than farmers with improved varieties because tradi- tional varieties are more difficult to thresh than improved varieties as the grain is not easily released by threshing. In addition, bullock farmers have labor requirements associated with team driving and hand- harrowing which farmers using tractor services do not. Finally, this author has observed that small holders utilize labor less productively than the more capital intensive bottomland producers. Hired labor accounted for 25 percent of the total labor utiliza- tion; about 73 percent of the hired labor”wereemployed for cutting (44 percent) and first weeding (29 percent). Costs and Returns The mean farm expenditures for the upland system were ¢44, of which hired labor was the largest expenditure item (50 percent), followed by seed and fertilizer (26 percent), and land preparation (13 percent). The mean farm yield of this systemnwas 7.5 bags, or 79 1,350 pounds per acre which is the second highest yield per acre among the six production systems studied. The mean output of the system was 8.3 bags, and the gross income was ¢120. The net return to operating capital, family labor,and management was ¢74. Comparative Financial Analysis of Six Rice Production Systems The purpose of this section is to compare the financial returns to the production systems. Five measures of economic efficiency have been computed for each system,and the results are analyzed to identify production strategies with the highest financial returns and lowest cost of production. Net Cash Income11 Among the five bottomland systems the variation in net cash income was from ¢319 to ¢3,930, or, on a per acre basis, from ¢25 for System I to ¢48 for System III. Net Cash Income for the 1.1 acre bullock system was ¢73.70; on a per acre basis, System VI had the highest cash income (Table 4.7). Return to Family Labor In order to compute the return to family labor, an opportunity cost must be assigned to operating capital expenditures. WOrking capital is defined as that portion of capital investment (stock) which 11The operating expenditures for land preparation for the tractor owners (Systems IV and V) and the bullock farmers (System VI) include as a cost a proportion of the stock of physical assets (tractor and associated equipment; bullock team and plow) consumed in an average year. Namely, capital stocks have to be converted to a flow of services in which the average depreciation of the capital stock is included as a fixed cost per acre. "nm‘ EBC) Table 4.7. A Comparative Financial Analysis of Six Haior Rice Production Systems in Northern Ghana Item Production Systems System I System II System III System IV System v System VI A. General Characteristics Number of Farms 28 44 ll 10 19 14 Agronomic System Bottomland Bottomland Bottomland Bottomland Bottomland Upland Power Source TBS 4 THS THS T0 TO B? Seed Variety Traditional ImprOVPd Mixed Traditional Improved Traditional Average Farm Size (Acres) 12.8 21.2 16.9 41.6 119.3 1.1 . , , . 7 . Thtal Production (18) lb. bags) 06.6 131.4 140.3 -76.4 84..0 8 3 . 7.LI Average Yield Per Acre (180 lb. bags) 5.2 6.2 8.3 6.5 7.1 u. Summary Financial Information 1. Cross Income1 c799 c1577 c1684 c3245 c10166 c110.35 . ' ’6.63 2. Operating ExpendituresZ LOO 915 871 2083 6214 a 3. Opportunity Costs a) Family laborJ 167; 136‘ 2436 1465 i215 66::96 b) Upctdting capital 67 7 128* 122 19a 3.) 4. Total Costs] 714 1179 1236 2423 7100 120.32 C. Measures of Efficiency 6 1. Net Cash Income ~ 3) Firn 319 662 813 1162 3910 :j.3? . . , , ‘J b) Pct acre 25 31 a 28 13 67 J 9 2. Return to Family Labor 3) Total 252 534 691 968 1347 shill b) Per man-hour 0.27 0.59 0.14 1.01 1});0 0.78 c) Per man-day 1.62 3.34 2.04 6.06 . . 3. Return to Operating Capital a) Iotallo 155 532 575 1079 ism 6.77 11 ' 8 S 66 4 53.4 63.2 14.9 b) Percent of costs 32.5 5 . . 12 4. Return to Management a) Total 85 398 448 21 3064 0:01 b) Per acre 6.60 18.70 26.50 19.60 25.70 . 13 S. opportunity Cost 01 Production 3) Per 180 lb. bag 10.70 9.00 8.80 9i30 2;:0 1:450 b) Per metric ton 133 112 110 1 l , . Output for the five bottomland systems IS valued at (12.00 per bag. For the upland system output is valued at £14.50 per bag. 3 Total farm expenditures plus estimated Selling costs as reported in the rice enterprlfifi budgets. where family labor is valued at the average wage rate for hired labor as reported {or each enterpriSr budget. For the throw THS systems. the opportunity cost of operating capital is calculated at the rate of 15 purcont of total farm expenditurts excluding land :learing costs as the latter include oprortunity costs. Fifteen iwrwent 0! total farm expenditures, excluding land r‘learing and land preparation costs, plus 15 percent of that portion of land preparation costs which are opirating expensvs (25 percent). Opvrating expenses were not charged an opportunity cost in deriving the per acrv estimates for land preparation costs for tractor owners (Sew Appendix C, Table 2). 6 For thv upland bullock system, the opportunity cost of oivrating capital is calculated at the rate of 15 percent of total farm oximndi- tures excluding land clearing and plowing costs, as the latter two already include opportunity costs. 7 The sum of operating expenditures plus the sum of the opportunity costs of family labor and operating capital. 8 Gross Incoum (Boll less Operating Expenditures (8-2). 3Gross Income loss the sum of 1) Operating Expunditures (8-2) and 2) the opportunity cost of operating capital (B-l-b). 10 . The YNJdflf will recall that certain capital stocks were converted to flows rice enterpriSH hudiers. In order to do this. assumptions about the source was charged an oppurtunity cost of 15 percent. Theixartion of fixed assets the subsidized bank lending rate. and those costs were, in turn, USvd in the preparation of th! of financing were made. The undeperiatud value of owners equity flnflnund by bank credit was charged an interes' rate of SIX jmrwwnt, The oiantunity cost of owners equity in fix~d assets is pirt of hln normal rvturn to curital and as J rrnilt, it must be deducted from operating expenditures before computing the total return to oierating capital (assumrd to ho lfifl prrtvnt rquity). For all six yroduction systnns, 100 percent of land clearing costs are assumod to be financed out of owner's rqnity. For 'ni bullock systom, out— half of the average undrpreciated value of fixed assets used in plowing and harrowing is assumed to be owner vquity. All fix-d ripital it'ms underlying plowing and harrowing activities of the tractor owners are as;umed to bu financed entirely from bank :rviit. Th« total amount of ogiortunity cost or return to equity included in theororating oxpenditutes‘summarizcd in B-J IS as follows: System I (1.49; System II €5.79; System III f4.61; Syutum IV €63.03; System V €180.76: System VI £1.16. The return to OPNYACIHQ capital, therefore, is Gross Income (B-l) less the sum of 1) Operating Expenditures (0-2) and 2) the opiortunity cost of family labor (B-J-a), plus the adjustments itemized above. 11 . . . . . Calculated as total return to operating capital diVided by operating nxpvndlturns reVised according to footnntv 010. 12 , . Gross Income less Total Costs (3-4) where total costs are operating expenditures plus the oppirtunity costs of family labor and ni«xating capital. 13 . . . Total Cost (8-4) diVided by total physical output. - -_. -‘ 81 is required to finance operating expenditures (flows) over an accounting period. Normally,operating expenditures are treated as inputs having no opportunity costs in the accounting period. However, since the rice enterprise budgets cover one accounting period (an average year) and since most operating inputs or operating capital items are tied up for a period of six to ten months in practice and a full year in effect, they effectively become operating capital expenditures. Hence, the budget expenditures for nonlabor items and hired labor are treated as capital expenditures which have an opportunity cost. In our analysis the private opportunity costs of operating capital expenditures are assumed to equal 15 percent, which is the social opportunity cost of capital as estimated by the Ministry of Finance and Economic Planning.12 The return to family labor for the five bottomland systems ranged from ¢252 to ¢3,347. For all bottomland systems the return per man-hour of family labor was significantly greater than the average wage rate paid to hired labor; for the upland bullock system, however, the return to family labor per man-hour was equal to the wage rate. The return to family labor must be considered in light of its relative importance as a component of the total cost of production. The relative importance of labor (and family labor) is largely explained by the importance of manual harvesting and the hours of weeding under- taken.13 For the systems where family labor is a relatively small 12The social opportunity cost of capital in Ghana is estimated to be 15 percent. See Romer, Michael and Stern, Joseph J. "Project Appraisal: Notes and Case Studies." Unpublished paper. Accra, Ghana: Ghana Institute of Management and Public Administration. May, 1972. 13For the two most labor-intensive bottomland systems (Systems I and III), the opportunity cost of family labor accounts for, on the average,21 percent of total costs: family labor accounts for 12 percent 82 component of total cost, the returns to family labor are, of course, high, given the financial profitability of the systems. A return to family labor as such is not as meaningful as the return to management. Hewever, the returns to family labor per man-day indicate that family labor can earn more per day when employed on the family rice farm than it can when employed elsewhere. Family labor in all systems, except System VI, earned more than the minimum wage rate for unskilled labor employed in the public sector.14 Thus, there is little financial ad- vantage in family members seeking wage employment on other rice farms or in urban areas, except for the days or weeks when family labor is not required on the family farm.15 Return to gperating Capital In computing a return to operating capital a value must be assigned to family labor. In the analysis which follows it is assumed that the opportunity costs of family laobr are equal to the local agricultural 16 . . . wage rate. The return to operating capital among the five bottomland of the total cost of System II and, on the average,five percent of total is family labor for Systems IV and V; for the upland bullock system, 57 percent of the total cost is the opportunity cost of family labor (Table 4.7). 14In 1974 the minimum wage rate for unskilled laborers employed in the public sector was ¢l.00 per day. 15Exceptions would include family members who do not share in the distribution of the return to family labor and when immediate cash is desired, as the return to family labor is not realized until the sale of the harvest. 6For each production system, the wage used in the computation is the overall system mean wage rate for all field activities as reported in the enterprise budgets. The variation in mean wage rates among pro- duction systems is from ¢O.12 (System III) to ¢O.25 (System V) per man- hour. This variation is due to (a) a variation in nominal wage rates and payments in kind (food), and (b) the composition of labor in terms of men, women, and children, each of which has different coefficients for the con- version from field-hours to man-hours. 83 production systems ranged from 32.5 to 63.2 percent of total operating expenditures: for the upland system the return was 14.9 percent. Except for the latter system, the percentage return to operating capital for all production systems was significantly greater than the estimated private opportunity cost of capital, which is estimated to be 15 percent. For the upland bullock system the return to operating capital was essentially equal to the estimated private opportunity cost of capital. The case of the upland bullock farmers appears to be the system closest to a state of equilibrium. There has been virtually no expansion of rice acreage among the sample bullock farmers over the past two years. That the return per man-hour of family labor equals the wage rate, and that the rate of return to capital is equal to the estimated opportunity cost of capital are both indicative of the fact that capital and labor resources have been correctly valued. The assumptions that (l) the opportunity cost of family equals the average wage rate for hired labor17 and (2) the opportunity cost of capital resources is equal to the social opportunity cost of capital, in combination, exhaust the total value of the product. Consequently, therewas a zero return to nanagement as would be anticipated from an equilibrium condition. System VI provided us with the only objective basis upon which to estimate the private opportunity cost of capital resources. As a consequence, our estimated opportunity cost of private capital appears reasonable. 17It is assumed that the average wage for hired labor is equal to the average value of marginal product of hired labor. 84 Return to Management After opportunity costs were assigned to both family labor and operating capital, all bottomland production systems had a high return to management. For the upland bullock system, however, there was zero return to management (Table 4.7). Cost of Production Among the five bottomland systems, there was a 28 percent varia- tion in the financial cost of production. The 119.6 acre tractor owner system using improved seed had the lowest cost of production (¢104 per ton), while the 12.8 acre tractor—hire, traditional seed system had the highest cost of production (¢133 per ton). There was little difference in the financial cost of production among the other three bottomland systems where the average cost was ¢lll per ton. Finally, the upland bullock system had the highest cost of pro- duction of all systems (¢l79 per ton). The high cost of this system was due to the large quantity of labor inputs.18 Even if the oppor- tunity cost of family labor for the upland system was half the bullock system mean wage rate, then the estimated cost of production would be ¢10.40 per bag, or about equal to the highest cost bottomland production system (System I). 18The mean wage rate paid for hired labor by bullock farmers is ¢0.13 per man-hour which is the second lowest of the six systems (the mean wage rate for System I is ¢O.12 per man-hour). The opportunity cost of family labor for the bullock system is 57 percent of total costs. For System I to V the proportion that opportunity cost of family labor is of total cost is, respectively, 23, 12, 20, 6, and 4 percent. n“ 85 Summary of Financial Analysis The returns to management among the six production systems give the planner a good indication of the shifts farmers are likely to make if present subsidy policies are continued. Specifically, the following shifts can be expected to occur: 1) Widespread adoption of combine harvesting by farmers with access to combine services; 2) A shift toward very large farms (100 acres or more) by producers with the required equity and access to combine harvesting services; 3) Increased use of fertilizer; 4) Increased adoption of improved seed or a combination of improved and traditional seed varieties among small to medium sized rice farmers. An Economic Analysis of the Six Rice Production Systems The purpose of this section is (l) to determine the economic costs of the resources used by rice farmers in Northern Ghana, (2) to determine the economic costs and benefits for each of the six production systems, and (3) to compare the production systems in order to identify rice production strategies with high economic returns to the Ghanaian economy. Theoretical Framework In an economy with no factor price distortions, prevailing market prices for factors and the real economic costs of the factors are equal. However, in Ghana factor price distortions exist because of various sub- sidies, tariffs, duties, taxes, and an over-valued exchange rate. In . s 86 . an economic analysis factors of production are valued at costs which reflect real scarcity values. In Ghana the factor price distortions facing rice producers are budget subsidies on selected inputs (e.g., fertilizer) and an over- valued exchange rate (an implicit subsidy).19 These factor-price dis- tortions or subsidies increase the demand for artificially cheap capital resources over and above what the demand would be if factors were priced at their higher economic costs. The overall effect is that northern rice producers are encouraged to adopt production techniques which are more capital intensive than they would be if factors were priced at their real economic costs. Calculation of Unsubsidized Factor Prices and the Percent of Subsidy for Each Factor Farmers in Ghana pay subsidized prices for all capital inputs used in production. Rice farmers who have adopted capital using tech- nologies (e.g., mechanical plowing, combine harvesting, and fertilizer) are paying artificially low prices for these purchased inputs (Table 4.8). Fertilizer is illustrative of the implicit and explicit subsi- dies embodied in rice production. The explicit subsidy on fertilizer is a budgetary subsidy administered by the Ministry of Agriculture. The financial cost to the Ministry of importing and transporting compound fertilizer to the Central Fertilizer Depot in Tamale was in 1973, ¢234 per ton, or ¢1l.70 per cwt. bag. The.Ministry of Agriculture, however, sold compound fertilizer to farmers at a fixed price of ¢S6.00 per ton, 19The official exchange rate is GH¢1.15 = US$1.00 and the shadow rate of exchange is GH¢1.55 = US$1.00. 87 Table 4.8. Subsidized and unsubsidized Prices and Percent of Subsidy for Selected Inputs Used in Rice Production in Northern Ghana, 1973-74 § Unsubsidized Capital Input Unit Subsidized Percent of Prices Paid price2 Subsidy by Farmers 1. Fertilizer a. 15-15-15 112 lbs. 2.80 15.63 82 b. 20-0-0 112 lbs. 2.00 9.89 80 2. Improved Seed 160 lbs. 12.00 20.60 42 3. Land Preparation 3 a. Tractor Owners 1) Plowing acre 8.10 10.76 25 2) lat narrowing acre 5.16 6.77 24 3) 2nd narrowing acre 3.13 4.08 24 b. Contract Charges4 1) Plowing acre 9.36 2) let narrowing acre 4.48 3) 2nd narrowing acre 4.07 4. Mechanized Harvesting a. Combine 180 lbs. 1.00 4.20 76 b. Combine as Sta-‘ tionary Thresher 180 lbs. 0.80 2.55 69 c. Tractor Threshing acre 3.78 4.84 22 1 2 Actual prices paid by farmers during the 1973-74 production season. Computed. See the appendices for calculations of the economic costs of each factor. 3Based upon computed owning and operating cost for Northern Region tractor owners. See Appendix C. 4 per measured acre for 83 sample farms hiring tractor services. The actual financial cost-price of contract charges is the computed average charge We did not have the required data to estimate the unsubsidized cost of private contract plowing. As a consequence, the estimated unsubsidized cost of land preparation for tractor owners is used in the economic analysis. 88 or ¢2.80 per bag. Thus, the Government was directly subsidizing fer- tilizer at a rate of 76 percent during the 1973-74 production season. There is also an implicit subsidy on fertilizer due to the over- valued official exchange rate. When the implicit subsidy is removed, the cost of one ton of compound fertilizer is ¢313 per ton, or ¢12.80 per bag. When the over-valued official exchange rate is taken into account, the total subsidy for compound fertilizer is 82 percent.20 The factor price distortions arising from an overvalued exchange rate and budget subsidies were taken into account when computing the unsubsidized costs of factors of production.21 The unsubsidized cost- prices and rates of subsidy of all factors used by northern rice farmers during the 1973-74 production season are reported in Table 4.8. Method of Calculating the Economic Cost of Rice Production In the subsequent analysis economic costs rather than market prices are used to value resources engaged in rice production. The economic benefits to be assessed are the alternative net benefits realized by each of the six rice production systems.22 20In 1973 about 90 percent of the fertilizer sold in the north was imported by the Ghanaian-German Agricultural Development Programme (GADP). The c.i.f. price is a grant by German aid. The GADP pays for the inter- nal distribution of the fertilizer imported under the program. As a consequence, the actual budget subsidy applies only to the quantities imported by the Ghana Government. However, if the Government imported all of the fertilizer used in the north, compound fertilizer would be subsidized at the rate of 76 percent at the official rate of exchange, or 82 percent at the shadow rate of exchange. 21 . . . . See Appendix C for the computation of the unsubSidized factor cost of mechanized land preparation, improved seed, fertilizer,and the Ministry of Agriculture's combine harvesting service. 22However, if a production system realizes an economic loss (e.g., System I), one can state that the returns to the factors employed in the 89 For the subsequent economic analysis the rice enterprise budgets are based on the physical resources used by each production system. However, the prices or values used in the economic analysis were derived in the following manner: Nonlabor Costs. Nonlabor items are valued by the unsubsidized prices reported in Table 4.8, except for the land preparation charges of the three THS production systems. For the three THS systems the 23 unsubsidized cost of land preparation for a tractor owner is used. Hired Labor Costs. The financial cost incurred for hired labor is used in the economic analysis of each production system. The wages paid to casual labor working on northern rice farms are free market wages,and it is assumed that the MVP of hired labor in rice production is equal to its wage rate.24 A lower shadow wage rate is not justified because there is not a regional surplus of labor available to work on rice farms during the harvest period. Qpportunity_Costs of Capital. The opportunity costs of capital resources engaged in rice production are determined on the basis of the economic opportunity costs of capital in Ghana. The Ministry of Economic Finance and Planning has estimated that the economic opportunity c(fists of capital in Ghana are 15 percent. This means that capital z‘esources directed toward investment opportunities with the greatest \ Production system not only do not cover total economic costs, but a51-30, if diverted to some alternative employment, may earn higher net re'turns. - 23Private contract charges contain a profit component, which is a re‘turn to a resource used in rice production and therefore is not a cfost to be included in the economic costs. Since the profit component in private contract charges is not known, the estimated land preparation costs per acre for tractor owners are used. 24There are no minimum wages for casual labor working on private farms. A minimum wage does apply for publicly owned farms. 9O economic use or benefit would earn at least a 15 percent economic rate of return over the life of the investment. we assume in the analysis that the opportunity costs of capital resources in rice production are at the rate of 15 percent of the unsubsidized factor costs. All capital stocks have been converted to flows and expressed in terms of annual operating capital expenditures,25 all of which have been assigned a 15 percent opportunity cost. Opportunity Costs of Family Labor. The economic opportunity cost of family labor is assumed to equal the average wage rate paid to hired labor working on rice farms. Total Economic Cost of Production. For each system the total economic costs are estimated by adding (a) nonlabor costs, (b) the cost of hired labor,and (c) the opportunity costs of operating capital and family labor. The economic cost of production for each system is reported both as a cost per metric ton and per 180 pound bag (Table 4.8). A Comparative Analysis of the Economic Benefits and Economic Costs of Production of Six Rice Production Systems Given the rates of subsidy on capital resources reported in Table 4.8, the economic costs of production were substantially higher than the financial costs when economic prices were used to value the factors of production. From a national point of view, five of the six production systems generated economic losses (Table 4.9). In fact, the 16.9 acre tractor- hire, mixed seed system (System III) was the only system which generated 25See Appendix C, Table 3 for the calculation of the land pre- paration costs per acre fbr a tractor owner. Table 4.9. A Comparativo Economic Analysis of Six Ell Rice Production Systems in Northern Ghana Item Production Systems System I System II System III System IV System V System VI THS THS THS T0 TO 80 Traditional Improved Seed Mixed Seed Traditional Improved Seed Traditional Seed Seed Seed (12.8 Acres) (21.2 Acres) (16.9 Acres) (41:6 Acres) (119.3 Acres) (1.1 Acres) A. Cross Economic Benefit1 €799 €1,577 €1,684 €3,245 €10,164 €120.35 8. Resource Costs 1. Nonlaborz a. Land Clearing 13 21 17 283 811 1.10 b. Land Preparation 241 417 332 840 2,533 10.25 c. Seed and Fertilizer 136' 679' 481' 969' 4,212' 24.00' d.:Mechanical Harvesting 74 118 97 558 2,912 ---- e. Bags 43' 84' 90* 123‘ 542‘ 5.30' r. Sub Total "39'? 1,37; 1,017 2,773 11,010 40.65 2. Hired Labor3 a. Pro-Harvest Activities 38 71 48 76 400 12.40 b. Harvest Activities 62 118 154 319 450 9.70 c. Sub Total mm “"1733- 202- 395° 850‘ 22.16- 3. Opportunity Costs a. Operating Capital 61 161 131 270 1,019 7.70 b. Family Labor5 in) 136 243 146 283 68.10 c. Sub Total 228 _~_29_7 174 W W —7_§.—83 c 4. Total Economic Costs) €985 €1,825 €1,593 €3,584 €13,162 €138.60 5. Economic Profit or 1.0557 r-as e- 248 e 91 r— 339 €-2.998 ¢-18.25 6. Economic Cost of Production Per Metric Ton €165 t 173 f 141 Z 165 ¢ 193 €207 Per 180 1b. Bag 13.30 13.90 11.40 13.30 15.50 16.70 l . . . . . . . . . . . . Gross Economic Benefit is toatl phySical production times the estimated economic import parity priCe of domestic production (See Appendix I). 2 , , u . . The estimated economic costs per unit for nonlabor items are reported in Table 4.8. 3 . u . . . . Thv economic opportunity cost of labor and the market wage rate for hired labor are equal as explained in the text of this chapter. 4 . . . . . . . The opportunity c03t of operating capital (the sum of nonlabor and hlred labor costs) 15 calculated in the following manner. First, the starred (’1 items are summed and multiplied by 15 percent since they do not includD any opportunity costs. The nonstarred items are treated differently SlnCe they already include an opportunity cost for that portion of the itemized cost arising from fixed assets. Only that portion aricing from Operating costs remains to be charged an opportunity cost. Operating costs represent 25 percent of land preparation costs (Appendix C, Table 3) and 19 percent of COMbinu harvesting costs (Appendix E). None of the land Clearing costs is an operating cost. Thortfore, both the 25 percent of land preparation costs and the 19 percent of combine harvesting costs are charged 15 percent opportunity cost. This is added to the 15 percent computed for the starred items to arrive at the economic opportunity costs of operating capital not yet included in expenditures. r.‘ . . ‘Thc hours of family labor times the average wage rate paid to hired labor by the farmers in the system. The soc1al economic oplortunity cost of family labor is assumed to equal the wage rate for hired labor. 6The sum of the costs of (l) nonlabor items, (2) hired labor and (3) opportunity costs. \ 7Gross Economic Benefit less Total Economic Costs. 92 economic profits from the national point of View. The 119 acre tractor owner, improved seed system, the one with the greatest returns to management from a private point of View, had the greatest economic loss (¢-2998). A.Comparison of Financial and Economic Costs of Production of the Six Production Systems The comparative costs of production per ton from both the finan- cial and economic point of view are reported in Table 4.10. The economic costs of production for the five bottomland systems were, on the aver- age, 47 percent higher than the financial costs. This increase in economic costs over financial costs for the five bottomland systems (Systems I-V) ranged from 24 to 86 percent and 16 percent for the upland bullock system. Following is a discussion of the reasons for the increased costs. Capital-Labor Ratios The reason why the economic costs of production were signifi- cantly greater than the financial costs was mainly due to the high rates of subsidy on capital resources.26 Further, the economic costs were greater than the financial costs depending on the mix of capital resources and the relative rates of subsidy among the various capital resources. The financial and economic capital-labor ratios for the six systems are reported below (Table 4.11). The capital-labor ratios show that the capital intensities of the two tractor owner systems were relatively high, particularly for 26Where capital resources are defined as the operating capital expenditures for land preparation, seed and fertilizer,and mechanized harvesting. 93 Table 4.10. Financial and Economic Costs of Production of Six Rice Production Systems in Northern Ghana, 1973-74 Production System Financial Costs1 Economic Costs2 Percent Increase (¢/Ton) Rank3 (¢/Ton) Rank System I: THS - Traditional Seed 133 5 165 3 24 (12.8 acres) System II: THS - Improved Seed 112 4 173 4 54 (21.2 acres) System III: THS - Mixed Seed 110 2 141 l 28 (16.9 acres) System IV: TO - Traditional Seed 112 4 165 3 47 (41.6 acres) System V: T0 - Improved Seed 104 l 193 5 86 (119.3 acres) Average for Systems I-V4 114 167 47 System VI: BO - Traditional Seed 179 , 6 207 6 16 (1.1 acres) 1Factors of production are priced at market prices. The financial costs of production are drawn from Table 4.7. 2Drawn from Table 4.9, where resources are valued at their real economic costs. 3Ranking is from one or lowest cost of production to six, the highest cost of production. 4Average for the five bottomland production systems. 94 Table 4.11. Financial and Economic Capital-Labor Ratios for Six Rice Production Systems in Northern Ghana, 1973-74 Production System Capital-Labor Ratio1 Relative Change Financial Economic Percent System I: THS - Traditional Seed 1.2 1.9 58 (12.8 acres) System II: THS - Improved Seed 1.9 3.8 100 (21.2 acres) System III: THS - Mixed Seed 1.2 2.0 67 (16.9 acres) System IV: To - Traditional Seed 2.3 4.4 91 (41.6 acres) System V: TO - Improved Seed 3.6 8.5 136 (119.3 acres) System VI: 30 - Traditional Seed 0.2 0.4 100 (1.1 acres) lThe capital—labor ratios are computed as follows: "Capital" includes the operating capital expenditures for land preparation, seed and fertil- izer,and mechanical harvesting. Labor includes the cost of hired labor plus the opportunity cost of family labor. A capital-labor ratio of 1.9 means that for each ¢l.00 of labor costs there is ¢1.90 of costs for capital resources. 95 the 119.6 acre tractor owner system (System V). The reader will note that the tractor owner systems also had the highest costs of production among the bottomland systems. On the other hand, the upland bullock system had the lowest capital-labor ratio because of very high labor utilization by this system. However, it also had the highest cost of production of all systems studied. Underlying Reasons for Variation in Economic Costs Table 4.12 is designed to identify the underlying reasons for variation in the capital-labor ratios and economic costs of production among the six production systems. The large capital inputs in mechanical harvesting were the main reason for the increase in economic costs com- pared to financial costs. The two tractor owner production systems had the highest capital-labor ratios and the greatest cost per acre for mechanical harvesting, as well as the lowest cost per acre for labor. As one would expect, there was also a direct relationship between the cost per acre for land preparation and the ratio, with the underlying reason for the variation in the costs of land preparation among the systems being the amount of harrowing done. Finally, the overriding cause of a high capital-labor ratio among bottomland rice production systems in Northern Ghana was the use of combine harvesters on large farms. Producer Income Support Derived from - Capital Input Subsidies Rice farmers are receiving substantial income transfers from the government as a result of subsidized (1) land preparation, (2) seed and fertilizer,and (3) combine harvesting services (Table 4.13). The pro- portion of financial net income provided by these subsidies in 1973-74 96 .MOQMH xawsmm mo uwoo >qusuuommo on» monoaoaHm .mfioumam cowuonpoum onwasouuon Ham macaw meow man who mcfisouumn pom mcflsoam uom mumoo oasosoom whom you .wcop mcflzouumn mo unsofim on» on map aum> mamumam OGMHEOpuon on» 0coam codumummoum Gama Mom whom mom umoom .mumOO UflBOGOUO SCAMS UmmflmH 00.Nm 0H.Hm III 0m.m 00.NH 00.0 v.0 H.H comm HonoHuwomua I om HH>AEmum>m 0m.0 00.00 0v.vm 0N.Hm 00.vm 0h.0H m.m m.maa comm pmboumEH I OH u> Emumam 00.MH 00.0w 0v.ma 0m.0~ . 0m.mH 0.0 v.¢ 0.Hv comm Hmcoauflpmua I 08 “>H swumhm 0m.mm 05.nm 0h.m 00.0H 0H.FH 0m.HH 0.m 0.0H ooww.pmxfiz I mma ”HHH Emumhm 0m.ma 0m.mm 0m.0 0h.ma 0m.~m 05.0 0.m N.HN comm 00>0HQEH I was “HH swumzm 00.0w 0H.mm 00.m 00.0H 0m.b 00.5 0.H m.mH comm HMCOfiuwomua I mmB «H Emumaw IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII mHGOUIIIIIIIIIIIIIIIIIIIIIIIII ocflumm>nmm cowuoummmum mound Mom Hmuoa Hmowsmnomz m pawn Hmuwafluumm comm Oflumm hogan no N . ‘1 Honda umoo Hopes ammousommm Hmuwmoo now muod mom mumou IHMUMQMU monoa Emumxm newuospoum thmhma .mcmnw cumnuuoz ca mswumhm coauosooum ooflm xwm mo moflumm uoanIHmuammo on» new mwousommm Hmuwmmo How whoa Hum mumou may no comwummaou m .NH.v wands 97 Table 4.13. Comparative Capital Subsidy Producer Income Support Among Six Rice Production Systems in Northern Ghana During the 1973-74 Production Season Production System Capital Net Income Total Subsidy Subsidies As A Proportion of Net Farm Farm1 Per Acre Farm2 Per Acre Income3 -------------Cedis -------------- --Percent ------ System I: THS - Traditional Seed (12.8 acres) 197 15.40 319. 24.90 62 System II: THS - Improved Seed (21.2 acres) 652 30.80 662 31.20 99 System III: THS - Mixed Seed (16.9 acres) 391 23.10 813 48.10 48 System IV: T0 - Traditional Seed (41.6 acres) 1,167 28.00 1,162 27.90 100 System V: T0 - Improved Seed (119.3 acres) 5,880 49.30 3,930 32.90 150 System VI: 80 - Traditional Seed (1.1 acres) 19 17.30 74 67.30 26 lCapital subsidies are computed by subtracting total financial nonlabor expenditures (enterprise budgets) from total economic nonlabor resource costs (Table 4.9). 2Financial net farm income is the net return to operating capital, family labor, and management as reported in the rice enterprise budgets. 38y computing what proportion capital subsidies per farm are of net return to operating capital, family labor, and management. 98 ranged from 26 percent for the upland bullock system (System VI) to 150 percent for the 119 acre tractor owner system (System V). In fact, the latter system would realize a negative financial income (¢-l950) if there were zero factor subsidies. The income distribution impact of the current subsidy policy is well illustrated berable 4.13. The absolute subsidy for the 119.3 acre system is ¢5,880. About 100 of the 6,100 rice producers in the Nbrthern Region were receiving these transfers as factor subsidies. The smaller systems received a much smaller transfer. In fact, it is important to note that the bottomland production system with the lowest economic cost of production (System III) had the lowest proportion of its financial income derived from capital input subsidies (48 percent), whereas the bottomland system with the highest economic costs of pro- duction (System V) had the highest proportion of its financial income (150 percent) derived from subsidies. Comparison Between Small Farm and Large Farm Rice Production Strategies The economic analysis convincingly points to the need to re-evalur ate the current capital intensive rice production strategy being pur— sued in Northern Ghana which stresses heavy capital-input subsidies. By indirectly subsidizing land preparation services and directly sub- sidizing mechanical harvesting, farmers are encouraged to expand farm size, thereby using land extensive, capital intensive,and labor-saving production practices as opposed to land and labor intensive production practices. These subsidies result in artificially high incomes which, in turn, provide incentives for farmers to adopt uneconomic production practices. The current capital intensive, labor-saving production I I- II ‘ i I.’ ill?) ‘I 99 systems are thus yielding economic losses as shown in the economic analysis. Ghana's rice production strategy in the future should be based, in part, on a production system which would foster high yields, low economic costs of production, and attractive private returns to rice farmers. It is assumed that Ghana would find it advantageous to identify rice production systems with relatively low capital-labor ratios and high economic payoffs since the country faces (1) a critical foreign exchange gap, (2) employment problems,and (3) an agricultural produc- tion which is not keeping pace with the increased demand for food. In order to augment production and farm income for a majority of the 6,100 rice producers and to generate rural employment, large numbers of producers need to have available improved output increasing and labor intensive production technologies which are consistent with their managerial and financial capacity. In order to contrast farm level trade-offs more clearly, two production systems were selected for special study. The objective wasto identify from the sample of farms (1) small farms using only manual methods of harvesting and (2) large farms using only combine harvesting. In addition, we required that for both systems sample farms use improved seed and have above average yields. The two production systems--small farms and large farms--were analyzed from both a financial and economic point of view in order to illustrate the differences in output, employment,and returns to society. lll‘l‘i'llv’lllll'lllllvil‘lti II .IIIIIIIII‘ . [i I ll'l'l'lli III 100 A Large-Scale Capital Intensive Rice Production System Using Combine Harvesting From the sample of 19 tractor owners using improved seeds, farms were selected that (a) had above average yield per acre and (b) were completely harvested by a self-propelled combine.27 Four farms met these criteria and were thus selected as models from which to identify the production practices which resulted in high yields and to illustrate the level of capital use, labor utilization, and income received by large rice producers using these production techniques. A financial budget representing the average physical resource utilization and corresponding expenditures for this special case of four high yielding farms as defined above is reported in Table 4.14. The average farm size of these four farms is 287.5 acres. Labor Utilization The total mean labor utilization per acre among the four farms was 40.5 man-hours per acre. Sixty-six percent of the total man-hours were used in pre-harvest activities (and 58 percent in weeding acti- vities). The labor in harvesting was used for bagging combined paddy. Seventy-five percent of the total labor were hired labor: of these, 45 percent were used in weeding, 28 percent for harvesting, and 27 percent for broadcasting seed and fertilizer. 27The average yield of the 19 farms was 7.1 bags per acre. Ten farms had yields above the mean, of which seven harvested with a com- bine, one used hand harvesting methods, and three used a combination of hand and mechanical methods of harvesting. Of the seven farms harvesting with a combine, two did not use fertilizer and were con- sidered unrepresentative of the group and were excluded. '101 Table 4.14. Rice Enterprise Budget for a 287.5 Acre Farm Based on Survey Data from Four Selected Farms in Northern Ghana Using Imprownl Seed and Own Tractor and Equipment and Having Above AVeraqc Yield ‘ , ‘ _ I, , n . ., . , fitt‘nltv k-Iz‘xti .I' _rih "In: IuIJ‘ ”.1 '21‘._\I.ty M 11 it» - _ ., z . _ ... [tin ”er“; ...I; -7 II. ..zI.-«.‘-r‘ .‘v. i'vz' Po.” 0 ’Fa' :11 ~‘<‘::r Trial FYI-.711: "Ir‘f 'm'w- F‘x: wtsiTI. na- i'f‘f‘ UI.‘.I.:-‘ Pr‘I' I‘:‘:.-"‘ 55:1».- ~——- I'L’IV.‘ UNIT 1"“? 'f'It,I1 ram; 1y I! 1 red Tv‘!‘ EI '3,“ I1 A«?'1‘.’1’»‘j Labor Labor ‘1 s L." fur- H'T‘“ : : : — z : ... A. Law C.L'.'.'U‘)Y3'£J (337.1") 'iflTIV‘S 5.5 l.‘:‘:";.'.5 i U. Pf‘bf.‘!."‘v’-‘;'f. ig‘h’lta' 7?, 1‘." .1" 7. 1:21 Henri-air»: 25 .5 £15" .5 ‘41': l.-«“ C.2-1 IL‘U‘I' wire: Ila‘Tp, :32“): ’fi,‘ " - «. 9 / gut—II. 1.211 ., .'l . I ~‘—' 3*:er 397.5 121,7.- IZu‘iiI 1%.", Linn"? ”(3’1 I ”.1 r I;,-; Jim-'6 “I'd. I =.10 1.11 511."! 1““?- i“""“~ - 397.5 (“I/1.; <1 I. '4. L fill 13“: ‘u (I U. a ' 7 0. r: 71. "‘3 Anmon. Sulfate 95x.5 :xnm 05x, 37.,1 .,:’ ugn,(- 34> r.u: 13;.1, 13?. ‘,-.".. .‘I.~’ 11.1, 1._'-' 3 ’.“r II . «n I . . . . .~ ‘ “'11 m” 11?! o ) -' ,~ . ’ . . ‘ - n ""'_.‘T- -"T . .. . . .» .. Y— ‘T ”‘f‘f -.\QK:"2~':".’1‘L 'J,L"7. ‘0 "p.“ ["_“,'.t 1.7-“Ho? Q, ’4’..4 .u;.‘ H.’ “ 1,1 ‘A- l . I'Ir musical Cum 111': 231' . F,- ‘.“'I.;:. 3‘4 3.7 . '2» l. 00 .937? . ‘3'.) Erwin: 1m 354 .5 1,563.6 .‘r,uun.7 I:.:-1 0.1? 3.1 ‘ IL-. “1:11 I __ 2. 9.1.2: £2.50 1'45: , 35, Sat 'l‘otil of awning :;iI*‘:.i:e.g..._~, 14,393.?5 15.9 3,906.3 1,550.6 9.3%.? c.11 0.97 "7%.39 D. Total F'L¥;‘-4-ndi';uz~3s and .____ __ __ _.___ __. __ .____._. -_ -... T __ -.1 .-___. Larlc‘!‘ 31i11.3t1‘)’l 1_"-.— "'f1 41).!) 11,043.} L,‘l:; ‘03 ’74‘. 9"? S.~)l 1’5”)“ "5). Ini‘or'r an". Dignizxure SUIT m“; 8:121:15 of 9.1 ‘4st .iI‘t'C - I a To? :11 E».IIII:'-,,1:-,r-. a. Non Labor ix; «niitiu‘es V ' r r" ‘ l V ‘ ‘ I“ "-3 = " N) )r ". N I" " '~-’-~ 5*“. ’ " “MW 4‘" b ‘93") ‘33 1. [and L Lear-1m : 1,3,").6.’ b. Valaw . E‘r-IJJ‘H'II. I‘. [indi‘rvxemtion 4,713.1 : 155.5 (“M-‘- X 3 i- «9'17 ' C ,t‘bdfilflmi 3. 89641 i. Fertilizer 4,176.23 1) - r~‘.e-:!‘wiical meestim' ~ ,i . , ., _.. . ;. Bags 1 U€K.?C t 1H,99‘,71 ’1. FEE-'1 (3.120 .rIrr’m: Jj‘UHJN-j ‘ _L__._ 1). Labor hm‘ufiitur‘eg .- , ~ ‘ "A“ _ 4 . 1" r' 6:. r.r:1n’ItI_xi..f-Ilt.;t -u.-L.~ . A. P 1‘ Ftp-h'uvo'st Activitv 1 fund -:.-. Lit-43:3. T 1.3; .'..Vf[:I'-!lii’f.I_U"!.‘ 16, 525.11 375.05 1 unn,un Fin-"4W labor t't—'13 c. Total Penn Ejmnditunes ¢ L; Wumyvnvnt + 1'? 'I‘i’fll}. ‘0 f. .‘k". Path walxrrk ta» II‘-',>-“I'IY;1.:JT "" ital, 3- “31""?31 Ai‘tiVii‘J \A‘ 1;: a' . . . . —/The cost of land clearing is the average annual amortized cost. See Appendix F for the calculation of land clearing costs. b/ . . . . . . - Unit costs of mechanized land preparation are based upon computed owning and operating costs of a tractor and assoc1ated equ1pment in Northern Ghana. See Appendix C, Table 2. 102 Costs and Returns The mean total farm expenditures among these farms were ¢l6,323. The largest expenditure item was land preparation (29 percent), followed by seed and fertilizer (26 percent), combine harvesting (18 percent), land clearing (10 percent), bags (9 percent), and hired labor (9 percent). The mean fanm yield was 10.2 bags, or 1,836 pounds per acre. Total production was 2,933 bags, or 235.65 metric tons. Gross income was ¢35,l90, and the net return to operating capital, family labor, and management was ¢17,987. A Small-Scale Labor Intensive Rice Production System Using Manual Methods of Harvesting From the sample of 44 farms utilizing tractor-hire services and improved seed, farms were selected that (a) had above average yield per acre and (b) were completely harvested by hand methods.28 Five farms met these criteria and were thus selected for special analysis as models from which to identify the production practices which resulted in high yields and to illustrate the yield and income that can be obtained by farmers using improved production techniques on relatively small rice farms. A financial budget representing the average physical resource utilization and corresponding expenditures for the special case of five high yielding small rice farms as defined above is reported in Table 4.15. The average farm size of these five farms is 3.9 acres. 28The average yield of the 44 farms was 6.2 bags per acre. Thirteen (13) farms had yields above the average, of which 11 farms applied fertilizer. Among these 11 farms, three harvested with a combine, five harvested by hand, and three used a combination of hand and mechanical methods of harvesting. Table 4. 15. 103 Using lnproved Seed and Tractor Hire Services and Having Above Average Yield Rico Enterprise Budget for a 3.9 Acne Farm Based on Survey Data firm Five Selected Fame in Northern Ghana it: I 1‘." lt y lireratlmj; Capital and Labor Utilize? ion by I'Ictlvlty Ac has labor Mamm r5 wage It": 1'. 0? Per Total Family Hired Per A: t ivity Q A M": {‘9 labor Labor How" Lm'il Clr firing, 3.. , P1 ‘0-liU’V‘? 17.: P iQH'l F'Jj 3'. 1') 13!. Narrowing 3, 4 Sab—Tfltfil Eli-W1 Cramp. er‘f. Ammn. Sulfate lit L-Jeuii 1:3; 8:11 Wtf'ffililli‘; Sub-Total w~l~ c») L: xi) d l {31'V0xfi gut: '12.: 3-9 H . , . - ‘\ mapi L 3- ‘: 'I‘Ir'mhlr Ir? 3‘. . ‘7) Win. -'x :‘~'l:’;fif‘.’e§ i- 9 3 ’:I I35 ‘l‘otil b;:.v..-r;;ii:.ur2:-s mi Luuul‘ Otllld'illffll E O t" C.) *AL) Ln! 0‘ 117 C“ we've LA I: Y'v...‘ 5C . 30 29-1 11Jo) 11.2 II: uh C )5 1.99.0 :W . H r—J I . r LA. ._ 1| 3 6.2 .- .u 7.1I .— .9 3.” - .2 32.14 .3}? 0.1!: .I win new 0.13 2.10.1 913.19” 12.). 0.14' ’3‘. 0.11 xflkuflkn a --I ..— - \A P—-‘ \1 ‘0 O‘ T) 4 A‘ U} lg k}. J7 . 0.: 0.1 0 8 .7 0.09 n 0": Q O H Ft- .a‘l U . C: .3 P—l . .— l (‘2 O ya ...a Per Acre Expand it}. Ir'v: Total ¢———¢——¢— 2.43 12.30 1b.“? _‘I n.1,... 'l: Y!‘,'I“!r\‘/VI ., .11! . l ._ - TI ~L’I‘L {‘1‘ dual- 1:"er 5.] In}; x 3.3 ,11'TI"‘Z' - “Still..- -:I‘ Pl"')ll2<'}l.l(‘!“. my; Isa- x ,3 = y. . _ m . [I ' lel‘k‘,‘) ‘1‘ 1.1 J 1. OM11.)— 1" ul '1!) "0'." l--‘l . . ... . .1 ' . . v'll‘l'J‘ ! , i' ‘AY‘L‘A ..j lili' -.‘ ’Ulil i-lr;'I,’= 70:111. LIII'rrnrr 1:51 E;XE‘-~.I:.1ll" / u it Cost of land clearing is the average annual amortized cost. 1)," — Average cost of hiring a transpo Appendix F .-. .Z‘fi‘ .‘lt'l 3.1"? C‘ . . .>I Irma g; a . :‘iC-Yl I: a 0‘ l. T) of mm: Id 10:119.“. Labor .-.‘-:pen:litures land Clearlrnr v. Land Phrrnration Seed and li‘er‘tilizcr' V V _ ’~ {5:31 i . "Ibor‘ Pm-lammst I‘Ictivlt 14.x; ”ammst .ficti ‘."1"T1"‘.'“I c. Total l-a m Expen'litmm rter to convey paddy from farm to market is (20.30 per bag. 3.90 55.514 36.96 lF-g-‘I‘F, 16.142 v.18 _— ¢113. 55 t 49.6 $16..“ for the calculation of land clearing costs. 104 Labor Utilization The mean labor utilization per acre for this group of farms was 184.4 man-hours per acre. Seventy percent of the total labor were employed in harvest activities and 28 percent in weeding. The mean labor employment in weeding was 51 man-hours per acre which is second highest to System III (62 man-hours per acre) analyzed previously. Sixty-three percent of the labor were hired; of these, seventy-three percent were used for harvesting activities and the remaining in weeding activities. Costs and Returns The mean total farm expenditures among the five farmers were ¢162, of which 34 percent was for land preparation, 30 percent for hired labor,and 23 percent for seed and fertilizer. The mean farm yield was 8.7 bags, or 1,566 pounds per acre. Total production was 33.9 bags, or 2.72 metric tons. Gross income was ¢407, and the net return to operating capital, family labor,and man- agement was ¢235. A Comparative Financial and Economic Analysis The most salient results of an analysis of the two contrasting systems are described. The reader desiring more detail is referred to the enterprise budgets reported above and the tables reported in this section. Financial Analysis of the Two Systems From a financial point of view, both the labor intensive and capital intensive systems had higher returns to management per acre 105 and lower costs of production than the five bottomland systems pre- viously analyzed (Table 4.16). The greatest contrasting features of the two budgets are (1) the relative nonlabor expenditures per acre and (2) the average labor utilization per acre. The large-scale capital intensive system had expenditures of ¢52 per acre for nonlabor items, whereas the labor intensive system had nonlabor expenditures of ¢29 per acre. This difference can be explained,for the most part, by combine harvesting and additional harrowing done by the large scale tractor owner system. On the other hand, the greater labor utilization per acre by the small-scale system can be explained by manual harvesting and the greater labor utilization in weeding. As a result of the contrasting factor proportions between the two systems and the high factor subsidies as previously described, the relative proportion of net income derived from subsidies was 45 per— cent of the net income29 realized by the labor intensive system and 129 percent for the large scale, capital intensive system. ,The absolute amount of capital subsidy support for the labor intensive system was ¢lOS,whereas the level of subsidy for the capital inten- sive system was ¢23,l95.30 29Where net income is net return to operating capital, family labor and management as returned in the enterprise budgets. 30The absolute level of subsidy and the relative proportion of income support from factor subsidies are calculated in the same manner as reported earlier in the chapter; see Table 4.13. Table 4.16. 106 A Comparative Financial Analysis Between Small Farm and Large Farm Rice Production Strategies Using Survey Data Item Production Systems Small-Scale Large-Scale Labor Capital Intensive Intensive A. General Characteristics Number of Farms 5 4 Agronomic System Bottomland Bottomland Power Source THS TO Seed Variety Improved Improved Acres in Rice 3.9 287.5 Total Production (180 1b. bags) 33.9 2932.5 Average Yield Per Acre (bags) 8.7 10.2 B. Summary Financial Information 1. Gross Income ¢407 ¢35,190 2. Operating Expenditures 172 17,203 3. Opportunity Costs a) Family Labor 29 493 b) Operating Capital 20 679 4. Total Costs 221 18,375 C. Measures of Efficiency1 1. Net Cash Income a) Farm 235 17,987 b) Per Acre 60 63 2..Return to Family Labor a) Total 215 17,308 b) Per Man-Hour 0.80 5.97 c) Per Man-Day 4.80 35.82 3. Return to Operating Capital a) Total 206 17,494 b) Percent of Op. Exp. 120 102 4. Return to Management a) Total 186 16,815 b) Per Acre 48 58 5. Opportunity Cost of Production a) Per 180 lb. Bag 6.50 6.30 b) Per Metric Ton 81 78 6. Man-hours Per Acre 184 40 refer to the footnotes accompanying Table 4.7. 1For the methods used to calculate the measures of efficiency, 107 Economic Analysis of the Small Farm and Large Farm Systems The economic capital-labor ratio of the capital intensive system was 17.7, or six times greater than the ratio of the labor intensive system (2.5).31 And it is the relative economic costs of production and economic profits generated by the two systems which reflect the con- trasting ratios. From a financial point of view, there is only a four percent difference in the costs of production between the two systems. However, when economic prices are used to value resources,the relative costs of production diverge. The cost per bag by the large farm, capital intensive system*was ¢l4.80, which is 53 percent greater than ¢9.70,the cost of the small farm, labor intensive system (Table 4.17). The economic profit generated by the labor intensive system was ¢79. On the other hand, the large farm, capital intensive system generated an economic loss of ¢-8,261 from society's point of view. Moreover, the benefits of the capital intensive system not only do not cover the opportunity costs of operating capital and family labor, but also do not cover roughly ¢4300 in unsubsidized factor costs. Summary The purpose of this chapter was to estimate both financial (private) and economic costs and returns for six rice production systems in current use in Northern Ghana. Financial rice enterprise budgets were constructed from survey data for five bottomland production systems 31Calculated in the same manner as reported earlier in the chapter; see Table 4.11. Table 4.17. 108 A Comparative Economic Analysis of Small Farm and Large Farm Rice Production Strategies Using Survey Data 1 Item Production Systems Small-Scale Large-Scale Labor 'Capital Intensive Intensive (3.9‘Acres) (287.5 Acres) A. Gross Economic Benefits ¢407 ¢35,l90 B. Resource costs 1. Nonlabor a. Land clearing 4 1,955 b. Land preparation 68 6,213 c. Seed and fertilizer 124* 15,687* d. Mechanical harvesting -—- 12,317 e. Bags 22* 1,906* f. Sub Total 218 38,078 2. Hired Labor a. Pre-Harvest activities 16 1,162 b. Harvest activities 32 279 c. Sub Total 48* 1,441 3. Opportunity Costs a. Operating capital 32 3,439 b. Family Labor 30 493 c. Sub Total 62 3,932 4. Total Economic Costs 328 43,451 5. Economic Profit or Loss ¢ 79 ¢-8,261 6. Economic Cost of Production Per metric ton 121 184 Per 188 lb. bag 9.70 14.80 1For the method used to caluclate economic costs and benefits, the reader is referred to the footnotes accompanying Table 4.9. llll III Ill‘ l ' 'I‘llil I II. IIIIIIIIII II I III III! I ll Ill-III ill II If F1“ ‘ 109 which used tractor mechanization for initial land preparation and for one upland system in which the bullock plow was used. Among the five bottomland systems, there was a 28 percent varia- tion in the financial costs of production. The 119-acre tractor owner system using improved seed had the lowest financial cost of production (¢104 per ton), while the 12.8-acre tractor-hire traditional seed system had the highest cost of production (¢133 per ton). The upland bullock system had the highest financial cost of production of all systems studied (¢179 per ton). A net cash return to operating capital, family labor,and manage- ment is reported for each of the six systems. The budget data were used to derive financial returns to (1) family labor, (2) operating capital expenditures, and (3) management. The returns to family labor among the five bottomland systems ranged from.¢252 for System I to ¢3347 for System V. For all bottomland systems, the financial return per man-hour of family labor was greater than the average wage rate paid to hired labor. For the upland bullock system, however, the return to family labor was equal to the wage rate. (The return to operating capital varied from 15 percent of total operating expendi- tures (System VI) to 66 percent (System III). All five bottomland production systems had a high return to management after opportunity costs were assigned to family labor and operating capital- However, for the upland bullock system, there was a zero return to management. An economic analysis of each production system was undertaken to determine the economic profitability of each system from the na- tional point of view. All nonlabor cost items were valued by esti- mated unsubsidized cost-prices, and the cost of hired labor was valued 110 at the actual financial cost incurred. The economic opportunity cost of all operating capital expenditures was valued at 15 percent, and the opportunity cost of family labor was assumed to be equal to the local agricultural wage rate. The economic costs of production for the five bottomland systems were, on the average, 47 percent higher than the financial costs. This relative increase in economic costs of production among the systems is due to the mix of capital resources used in production and the relative rates of subsidy among nonlabor resources. When market prices were used to value resources (financial analysis), the large-scale capital intensive system (System V) had the lowest cost of production (¢104 per ton). However, when economic prices were used, this system had the highest cost of production (¢193). System V also had the second highest yield among the bottome land systems, the highest C/L ratio, the largest acreage, the greatest degree of combine harvesting, and the lowest labor inputs per acre. In contrast, System III had the lowest economic cost of production (¢14l per ton), the highest yield per acre, the lowest C/L ratio, the second lowest acreage, and the highest average labor utilization, per acre. The upland bullock system had the highest cost of production from both the financial and economic points of view, primarily because of the large labor inputs of this system. Our analysis showed that small number of the 6,100 rice farmers under present policies are receiving substantial income transfers in the form of subsidized (l) combine harvesting services, (2) seed and fertilizer, and (3) land preparation. The variation in income support from factor subsidies ranges from 26 percent (System VI) to 150 percent lll fer the large-scale, capital intensive system (System V). The system with the lowest economic cost of production has the lowest govern- ment transfer in terms of factor subsidies, while the system with the highest economic costs of production has the highest government transfer by way of factor subsidies. The economic analysis also illustrated that all production systems, except the 16.9-acre THS-system (System III), generated economic losses from the national point of view. The large-scale, capital intensive system (System V) generated very high economic losses from the national point of view. The analysis in this chapter thus convincingly points to the need to re-evaluate the government's production strategy. Current policies encourage farmers to expand farm size, thereby using sub- sidized capital intensive and labor-saving production practices which are financially profitable but uneconomic from the national point of view. In order to identify more clearly the trade-offs among (1) capital requirements, (2) farm employment, (3) producer income,and (4) output, two contrasting production systems were analyzed in detail. One system was based upon tractor ownership, high yields, and combine harvesting. This system is typical of the large-scale, capital intensive production approach. The second system was based upon the hiring of mechanized land preparation services, high yields, and manual harvesting. This system represents small farm, labor intensive rice production. The analysis of these two systems showed that, from a financial point of view, the returns to management were high and the cost of production about equal. However, from an economic point of view, the 112 costs of production differed substantially. The small farm, labor intensive system produced rice at ¢121 per ton as compared with ¢184 per ton for the large-scale, capital intensive system. Hence, the small farm approach produced substantial output and income to farmers and generated economic profits, from the national point of view, while the large farm system generated substantial economic losses. Our analysis thus demonstrates the need for reorienting the government's rice production strategy to include a substantial small farm emphasis. CHAPTER V EMPLOYMENT AND INCOME DISTRIBUTION IMPLICATIONS OF ALTERNATIVE RICE PRODUCTION SYSTEMS Introduction Planners in Ghana are in need of data on the efficiency, output, employment, and income distribution implications of alternative pro- duction systems.1 The purpose of this chapter is to describe the relative labor requirements for six rice production systems. Labor supply is analyzed in terms of (a) family and hired labor, and (b) men, women, and children. We compare the employment and income dis- tribution implications of a hypothetical expansion of combine harvest- ing in two production systems; one is a small farm, labor intensive production approach and the second is a large farm, capital intensive approach representing the present harvesting strategy in the study area. Labor Utilization Among Six Rice Production Systems Five Bottomland Systems There was a wide variation in the average man-hours per acre among the five bottomland systems. The variation ranged from a high of 220 man-hours per acre for the 16.9 acre tractor-hire system to a low of 38 man-hours per acre for the 119.3 acre tractor owner system. The principal reasons for the marked differences are the intensity of weeding and the degree of mechanical harvesting. 1There is a growing awareness of the need for more data on the direct and indirect implications of alternative production systems. See Byerlee, 1973 and Steele and Mabey, 1973. 113 114 Pre-Harvest Activities Pre-harvest activities consisted of the broadcasting of seed and fertilizer and weeding. The variation in man-hours for pre-harvest activities ranged from a low of 18 man-hours per acre for the 119.3- acre tractor owner system to 70 man-hours per acre for the 16.9-acre tractor-hire service system (Table 5.1). For broadcasting activities, all production systems used about the same man-hours per acre. Excluding System, III, which used 12 man-hours per acre, the average man-hours per acre in broadcasting activities (seed and fertilizer) was about eight man-hours. Weeding activities for all systems accounted for the greatest proportion of labor utilization in pre-harvest activities. The varia- tion ranged from 58 percent of the total pre-harvest hours for the THS-Traditional Seed System to 89 percent for the THS-Mixed Seed System. The two tractor owner systems used only 11 man-hours per acre for weeding; the 16.9-acre THS system employed 62 man-hours, and the 12.8- and 21.2-acre THS systems utilized 16 and 24 man-hours per acre, respectively. With the exception of System I, as farm size increased, the man-hours per acre in weeding declined. Harvest Activities Harvest activities accounted for the greatest proportion of the total labor utilization among the five bottomland systems, all of which used a combination of manual and mechanized methods of harvesting. Table 5.2 reveals that the variation in labor utilization in harvesting was most directly related to the degree of mechanical harvesting. Labor utilization in harvesting ranged from a high of 149 man-hours for the 16.9-acre system to a low of 20 man-hours per acre for the 119.3-acre system. JLLS .0:0pow3 000:» can ocoomm 000 com: nonm0 mocs0oc0 0 uOumuwmo 00000sm u Om pcm unoczo uOucmue n 09 qu0>uwm 000: uOuomue n 0190 .0000030 om0uououcm 000m 5000 00050500 0 00 0.00 00 0.00 00 0.00 00 0.00 00 0.00 00 0.00 00 00 00 III 00 00 00 000 0.000 000 0.00 000 0.00 000 0.000 000 0.00 000 0.00 00 0.00 00 0.00 00 0.00 00 0.00 00 0.00 00 0.0 00 00 00 00 00 00 000 0.000 000 0.00 000 0.00 000 0.00 000 0.00 000 0.00 0w 0.000 00 0.00 00 0.00 00 0.00 00 0.00 00 0.00 00 mN 0N MMI 0c No mmw 0.00m 000 0.00 000 0.00 000 0.000 000 0.000 000 0.000 000 0.000 00 0.00 00 0.00 00 0.000 00 0.00 00 0.00 .uun --- 00 0.0 0 0.0 0 0.0 0 0.0 0 0.0 000 0.000 000 0.00 000 0.00 000 0.000 000 0.00 000 0.00 00 00.00 00 0.0 00 0.0 00 0.0 00 0.0 00 0.00 00 0.00 00 0.0 0 0.0 0 0.0 00 0.0 0 0.0 we 0.00 00 0.0 00 0.00 00 0.00 00 0.00 00 0.00 000 0.000 000 0.00 000 0.00 000 0.00 000 0.00 000 0.00 00 0.000 00 0.00 00 0.00 00 0.000 00 0.00 00 0.00 00 0.0mw 00 0.00 00 0.00 00 0.00 00 0.00 00 0.00 000 0.000 000 0.00 000 0.00 000 0.000 000 0.000 000 0.000 v 0.004 .00“ N muu< .0000 N .wuu< hwm N NHU< 00% N OHU< uwm N 0.00.0.0 me 9050:0062 muDOLCQX thOSGQ—L mhdoncmz mHDOSCNX muDOLGGZ 0mco0u0omue 0~>ouae0 0mco0u0nmue 00x0z vu>ouas0 0mco0u0vmu0 00 00 00 0:0 0:0 mxe 0> syum>m > 500000 >0 Emummw 000 Ewum>n 00 aoumxm 0 500000 mu0u0>0uo< a mc0u0>00o< umm>u mw0u0>0uo< 00m 0000: 000200 mo>umm .0 pcu0z >00Emm mmeum .0 emu0z 000500 00 00¢ .0 uobm0 0c0umm>umz c MCOumo>umz 0mo0cmc mm0u0>0uu< o 00:00 0C0cwmz coo 0:00003 um mo0u0>0uu< umm>u umw> umm>ummI mu0u0>0uu< 00m 00 cousom .0 cm: .b act .0 mm>umx .0 00< .u com .0 000 .m «=Imum .0 um: .b 000 .0 0m 00¢ 00 000>0 comm 000500 00300 zu uo< 00000 .< condom pom 0>0uo< gonna 0 fiasco cuwnuuoz :0 meoum>m c00uospOum 000m X0m new 0530000005 .00qu .00 \Cmgm .0.m m0ndP 116 Table 5.2. The Relationship Between the Method of Harvesting and Average Labor Requirements for Five Bottomland Rice Production Systems in Northern Ghana Production Acres Method of Harvesting Total Man- System Hours Per Combine Cut by Mechanically Acre in Har- Harvesting Hand Threshed1 vesting ---Percent of Total Acreage ----- System I: THS-Traditional Seed 12.8 20 80 31 88 System II: THS-Improved Seed 21.2 14 86 33 72 System III: THS-Mixed Seed 16.9 3 97 47 149 System IV: TO-Traditional Seed 41.6 12 88 83 70 System V: TO-Improved Seed 119.3 77 23 12 20 1Includes using a combine as a stationary thresher and tractor threshing. Since many farmers used a combination of manual and mechanized methods of harvesting, it is difficult to directly estimate the labor requirement for manual and combine harvesting from the enterprise budgets. As a consequence, we estimated the average man-hour require- ments using labor data from 112 bottomland rice farms. We assumed that man-hour requirements per acre depend in part on yield. To estimate man-hour requirements per acre and the variation in yields, we sorted the sample farms into three yield levels and calculated the 117 mean man-hours per acre for each harvest subactivity. For manual harvesting techniques, we found that the man-hour requirements increased with yield per acre: however, the man-hours for labor associated with mechanized techniques were not related to change in yields. At a yield level of 6.0 to 8.9 bags per acre,2 it is estimated that 142 man-hours are required to harvest one acre of paddy using hand harvest- ing methods (Table 5.3). However, if a combine is hired to harvest paddy, only 10 man-hours per acre are required (for bagging only). Upland Bullock System Labor utilization per acre for the upland bullock system.was three times greater than that for the most labor intensive bottomland system (System III) for the following reasons. First, about 57 man- hours per acre were used for land preparation in association with bullock plowing and hand harrowing, whereas tractor mechanization was employed for land preparation among the bottomland systems. Second, about 169 man-hours per acre were utilized in weeding activities on the upland system which is about two and one-half times the man—hours employed in weeding by the bottomland system with the greatest number of man-hours in weeding. Third, about three times as much labor was used for manual harvesting activities as compared with the most labor intensive bottomland system. The reasons for the greater labor utili- zation in weeding and harvesting have been previously described. 2One bag of paddy is 180 pounds. 118 Table 5.3. Average Man-Hour Requirements Per Acre for Manual Harvesting Activities at Three Yield Levels Activity Bags Per Acre 3.0-5.9 6.0-8.9 9.0-12.0 ------ (Man-Hours Per Acre)------- Cutting 33.9 1 49.8 43.5 (2.8) (7.6) (7.3) Reaping 15.9 24.4 37.8 (1.8) (3.6) (7.6) Threshing 28.9 41.2 79.4 (4.5) (9.0) (21.1) Winnowing and Bagging 19.5 26.8 35.9 (2.0) (2.7) (6.8) Total 98.2 142.2 196.6 1Figures in parentheses are one standard deviation. The Composition of the Labor Force Family Versus Hired Labor All five bottomland production systems relied on hired labor to supplement family labor. For all field activities, the proportion of the man-hours hired to total labor requirements varied from a low of 38 percent for the 12.8-acre system (System I) to a high of 75 percent for the 119.3-acre system (System V). For all systems more labor was hired for harvest activities than pre-harvest activities (Table 5.1). A review of the individual enterprise budgets reveals that the greatest proportion of hired labor was utilized for cutting (34 percent), fo11owing by weeding (22 percent), and heaping (15 percent). For the bullock system, 23 percent of the total labor requirements were hired. About 45 percent of the hired labor were employed in cutting, 119 followed by 43 percent in weeding and 11 percent in other harvesting activities. Igportance of Men, Womenpand Children As one would expect, there is a wide variation in the relative importance of men, women,and children in various activities in rice farming. Despite this variation, it is possible to identify the relative overall importance of each and the activities which are dominated by men or women. For all bottomland production systems the average proportion of the total man-hours supplied by men, women,and children was, respectively, 57, 33, and 10 percent (Table 5.4). For pre-harvest activities3 women provided a greater proportion of the total labor requirements for the 119.3-acre system than for the other four bottomland production systems. For this large-scale system the prOportion of the total labor utilization supplied by men, women, and children was 44, 49,and 7 percent, respectively. Also for this system, men provided 58 percent of the labor for seed broadcasting, whereas 39 percent of the man-hours were supplied by women and three percent by children. Women provided about 50, S4, and 51 percent, respectively, of the total labor in the broadcasting of compound fertilizer, the broadcasting of top dressingland weeding, while children supplied about five percent of the total man-hours for fertilizer broadcasting and about nine percent of the man-hours for weeding. For the other four bottomland production systems men, women,and 3The coefficients used to convert field hours to man-hours were 1.0, 0.75, and 0.50 for men, women, and children, respectively, for broadcasting of seed and fertilizer, cutting, and heaping. For weeding, threshing, and winnowing the conversion factor for women was 1.0, or equal to men. 120 Table 5.4. The Relative Importance of Men, Women, and Children as Sources of Labor for Field Activities Among Six Rice Production Systems in Northern Ghana Production Systems Total All Activities Pre-Harvest Activities1 Harvest Activities2 Manhours * ‘ Men Women Children Men 1 Women L Children Men J Women[ Children System I: THS-Traditional Seed 1484 (12.8 acres) Man-hours 868 407 209 257 15 83 611 392 126 Percent 59 27 14 72 5 23 54 35 11 System II: THS-Improved Seed 2195 (21.2 acres) Han—hours 1032 838 326 365 144 155 667 694 171 Percent 47 38 15 55 22 23 44 45 11 System III: THS-Mixed Seed 3711 (16.9 acres) Man-hours 2267 1189 253 908 122 159 1360 1067 95 Percent 61 32 7 76 10 13 54 42 4 System IV: TO-Traditional Seed 3693 (41.6 acres) Man—hours 2499 956 238 537 241 21 1963 715 217 Percent 68 26 6 68 26 6 68 26 6 System V: TO-Improved Seed 4513 (119.3 acres) Man-hours 2281 1847 347 954 1050 155 1327 798 229 Percent 51 41 9 44 49 7 56 34 10 System VI: Bullock System Traditional Seed 741 Man-hours 362 295 84 190 39 10 172 257 74 Percent 49 40 ll 80 16 4 34 51 15 1Pre—harvest activities include the broadcasting of seed and fertilizer and weeding. In addition. for the bullock system only, bullock plowing and hand harrowing are included. 2 . . . . . . . . . . . Harvest activ1ties include manual harvesting requirements (cutting, heaping, threshing, winnowing and bagging) and labor associated with mechanized techniques. 121 children supplied 68, 16, and 16 percent of the labor for pre-harvest activities, respectively. On the average, men accounted for 75, 75, 67, and 64 percent of the man-hours involved in the broadcasting of seed, broadcasting compound fertilizer, broadcasting top dressing,and weeding, respectively. For these four production systems there was too much variation in the labor utilization of women and children to generalize about their relative importance in specific pre-harvest activities. Hand harvesting activities consisted of cutting, heaping, thresh- ing, and winnowing and bagging. Mechanized harvest techniques requir- ing labor consisted of (1) bagging combine harvested paddy, (2) feeding paddy into a combine, which is used as a stationary thresher, and then bagging the machine-threshed paddy and (3) "tractor threshing" which involves turning paddy and removing the straw after a tractor has driven over the paddy. For all harvesting activities the average pro- portion of the total labor provided by men, women,and children among the five bottomland systems was 55, 36,and 8 percent, respectively (Table 5.4). For the hand harvesting activities among the five systems, men provided on the average 83, 40, 29, and 19 percent of the total labor requirements fer cutting, heaping, threshing, and winnowing and bagging, respectively. In the same order of activity, women provided 9, 41, 65, and 80 percent of the total labor.requirements and children 8, 20, 6, and 1 percent. Thus, cutting was mainly the work of men; heaping cut paddy was about equally shared between men and women; and threshing and winnowing and bagging were predominantly the work of women. Children contributed most to heaping activities. For mechanized harvesting, men 122 provided the bulk of the labor. However, for tractor threshing opera- tions, women supplied about 30 percent of the labor and children about seven percent. For the upland bullock system, 49 percent of the total labor was provided by men, 40 percent by women, and 11 percent by children. Pre- harvest activities were dominated by men who provided 80 percent of the total labor in these activities. Women were the most important source of labor for harvest activities,providing 51 percent of the total labor requirements; however, men dominated the cutting activity, followed by women (34 percent) and children (15 percent). Employment Implications of Expanding Combine Harvesting At the time this study was conducted, no more than an estimated 10 percent (9000 acres) of the total rice acreage in the NOrthern Region was harvested by combine. However, the Ministry of Agriculture through "Operation Green Harvest" had a short—run goal to increase combine harvesting and to harvest 35,000 acres by combine within three years. The purpose of this section is to identify the direct employment implications of a shift from hand harvesting to combine harvesting.4 Potential Labor Displacement The analysis which follows assumes an average yield of eight bags, or 1440 pounds of paddy per acre.5 The labor requirements for manual and combine harvesting are based upon.survey data. we have estimated 4The indirect employment implications are not considered in this analysis. 5As shown, labor requirements for manual harvesting depend on yield per acre; the higher the yield, the greater the labor displacement by a shift to combine harvesting. 123 that at a yield level of eight bags per acre, 142 man-hours are required to harvest one acre of paddy manually and 10 man-hours per acre for bagging combined grain. For every 1000 acres harvested by combine, 22,000 man-days of labor are potentially displaced. Given a harvesting period of 60 days, the labor of 367 man—equivalents per day for 60 days is displaced. If the Ministry's goal of 35,000 combined acres is achieved, then 770,000 man-days of labor would be displaced by a shift from manual to combine harvesting (Table 5.5). The loss in income to casual laborers seeking employment on rice farms depends on the going wage rate and the relative use of hired and family labor. The average wage rate in harvest activities among the five bottomland systems was ¢0.95 per man-day. At this wage rate and with the assumption that 50 percent of the total labor requirement is hired and the remaining is provided by family labor, ¢10,450 in income would be the loss by casual laborers per 1000 acres harvested by combine. Thus, if the government goal of harvesting 35,000 acres by combine were achieved, the loss in income by casual workers would be ¢3S6,750 if 50 percent of the labor requirement were hired and ¢S49,000 if 75 percent of the labor requirement were hired (Table 5.6). Given a 60-day harvest- ing period, 577,500 man-days (75 percent of the labor requirement potentially displaced by shifting to combine harvesting) would provide work for 9,625 workers working six hours per day for 60 days. If this labor were displaced, then the loss in income per worker over 60 days would be ¢57. I'll II II I' I'll ll u.|lla| ‘II. 124 Table 5.5. Comparative Labor Requirements for Manual and Combine Harvesting and Estimated Labor Displacement for Combine Harvesting in Northern Ghana 1 Number Labor Requirements by Harvest Method Estimated of 2 Labor Acres Manual Combine Displacement Man-Hours Man-Days3 Man-Hours Man-Days3 Man-Days3 l 142 23.7 10 1.7 22 1,000 142,000 23,700 10,000 1,700 22,000 35,000 4,970,000 829,500 35,000 59,500 770,000 1Based upon survey data; an average yield is assumed to be eight, lBO-pound bags per acre. 2For bagging combined paddy. 3One man-day is defined as six field hours. Table 5.6. Estimated Loss of Income to Casual workers Resulting from a Shift From Manual to Combine Harvesting Under Three Assumptions Regarding the Proportion of Total Labor Requirement Hired Number Estimated Estimated Loss of Income2 of Labor 1 Acres Displacement Proportion of Total Labor Requirement Hired (Man-Days) 50 Percent 75 Percent 100 Percent l 22 ¢10.45 ¢15.7o ¢20.90 1,000 22,000 10,450 15,700 20,900 35,000 770,000 365,750 549,500 731,500 1From Table 5.5 . 2Calculated under the assumption that the average wage rate is ¢0.95 per day which is the average wage rate in harvest activities among the five bottomland systems. 125 Gains to Producers Since the Ministry of Agriculture's combine rate is heavily sub- sidized, farmers are encouraged to adopt combine harvesting because it costs substantially less than to hire labor. The Ministry charges ¢l.00 per bag for combine harvesting, as compared with an estimated economic cost of about ¢4.00 per bag. With a yield of eight bags per acre, the farmer is charged ¢8.00 for combine services and pays labor ¢O.806 for bagging combined paddy if 50 percent of the bagging requirement is hired. Thus, a farmer's total cost for combine harvesting is ¢8.80 per acre. If, on the other hand, the farmer harvests his crop manually and hires 50 percent of his labor requirement, his total cost for harvesting is {211.257 per acre. Thus, it costs the farmer 20 percent less to harvest with a combine. Moreover, if farmers hire 75 percent of their labor requirements, it costs 45 percent less to harvest by combine. Thus, a farmer not only reduces his harvesting costs, but he does not have to recruit and supervise as much labor if he harvests with a combine. Who Benefits From the Current Combine Strategy? The benefits of a major increase in combine harvesting would accrue to the relatively few--about 100 of the 6,100 rice producers--who_ are fortunate enough to be able to hire combines. These farmers are typically the larger, wealthier producers who receive a subsidy of about ¢3.00 for each bag that is combine harvested. 0n the other hand, the principal losers of subsidized combine harvesting are the displaced 6Bagging combined paddy requires 1.7 man-days per acre. 1.7 X .50 X ¢0.95 = ¢O.80. 7The total labor requirement for manual harvesting is 23.7 man-days per acre. 23.7 X .50 X ¢0.95 = ¢ll.25. 126 workers who lose about ¢0.95 per man-day, or between ¢30—90 during the harvesting period, depending on the number of days worked. Loss of income of this magnitude significantly reduces the welfare of the rural poor along with their capacity to provide adequate food, clothing, and shelter for their household. This loss in cash income also reduces the capacity of small farmers to purchase improved seed, fertilizer, and other inputs. Effects of Charging_Farmers the Economic Cost of Combine Services If farmers were charged the estimated economic cost of combine harvesting (¢4.30 per bag), the cost to combine harvest an acre with a yield of eight bags would be ¢34.40 as opposed to ¢8.00 now charged. The total cost, including hired labor for 60 percent of the bagging requirements, would be ¢35.35. Comparing ¢35.35 per acre for combine harvesting with ¢13.45 for manual harvesting, the hand method would thus cost 62 percent less. If 100 percent of the labor required by both systems were hired, hand harvesting would cost about 38 percent less than combine harvesting (¢36.00 vs. ¢22.40). At the present average wage rate, there is a shortage of labor for harvesting. This has been the situation in the region in recent years and was one of the reasons for introducing the combine harvester. With the introduction of the combine, wages have not been permitted to rise because increasing numbers of farmers have substituted subsidized combine harvesting for hired labor. However, if combine charges were to increase so that the cost of hired labor was competitive with combine charges, there would be an increase in the demand for labor, because manual harvesting would be cheaper than combine services and wages for 127 labor would increase. While we do not have the data to estimate the elasticity of supply of labor for rice harvesting, wage increases would augment the supply of harvest labor, particularly labor mi- grating from other regions where there is seasonal unemployment. Summary In this chapter the employment and income distribution implica- tions of alternative production systems were studied. The variation in labor requirements among production systems was analyzed,followed by an analysis of the composition of the labor supply in terms of (1) family and hired labor and (2) men, women, and children. The employment and income distribution implications of a major shift from manual to combine harvesting were also examined. The average man-hours per acre for all field activities among the five bottomland systems ranged from a high of 220 man-hours for the 16.9-acre THS system to a low of 38 man—hours for the 119-acre tractor owner system. The amount of weeding and the method of harvest- ing were found to be principal determinants of the variation in labor utilization among all systems. With the exception of one system (System I), as farm size increased, man-hours per acre in weeding declined. With the exception of the most capital intensive system (System V), about 73 percent of the total labor were employed in harvest activities. Further, our analysis showed that the total labor utiliza- tion by the upland bullock system is three times greater per acre than the most labor intensive bottomland system. This is due to greater labor utilization in weeding and harvesting and the labor requirements associated with bullock plowing and hand harrowing. 128 Family labor was supplemented by hired labor for all six produc- tion systems. As acreage increased, the proportion of hired labor also increased from a low of 38 percent (System I) to a high of 75 percent (System V) of total labor requirements. Most hired labor was employed for cutting, followed by weeding and heaping activities. There was considerable variation in the role of men, women, and children on rice farms. For all five bottomland systems the percent— age of total man-hours supplied by men, women, and children was 57, 33, and 10 percent, respectively. Men performed most of the pre- harvest activities, particularly broadcasting of seed, and among harvest activities, men did most of the cutting and heaping. Thresh- ing, winnowing, and bagging were predominantly the work of women. Our analysis showed that manual harvesting requires 142 man-hours per acre and combine harvesting 10 man-hours, under average yield condi- tions. Under the Ministry of Agriculture‘s subsidized combine service policy, farmers are encouraged to hire combine services as it costs between 20 to 50 percent less than it would to hire labor for manual harvesting. Nevertheless, for every 1,000 acres harvested by combine, 22,000 man-days of casual labor are potentially displaced, and if 50 percent of the labor requirements are hired and the remaining labor is provided by family labor, our analysis revealed that these casual workers could lose ¢10,450 in wages for every 1,000 acres harvested by combine. CHAPTER.VI SUMMARY AND POLICY PRESCRIPTIONS FOR IMPROVING THE ECONOMIC PROFITABILITY OF RICE PRODUCTION IN THE NORTHERN REGION OF GHANA seem Like most developing countries, Ghana has had to increasingly rely on food imports over the past decade to augment domestic pro- duction. In 1972 the National Redemption Council attempted to increase food production through its "Operation Feed Yourself" program by creating a national awareness of the importance of agriculture among Ghanaians and stressing self-reliance and self-sufficiency in food production. In particular, government has given high priority to increasing rice production. However, there is a lack of basic data on the costs and returns of present and alternative rice production systems. This study generates farm level data to estimate the costs and returns for six rice production systems in northern Ghana. In addition, the study analyzes the efficiency, output, employment, and income distribution implications of the six production systems. In Chapter II the Northern Rice Production Industry was reviewed. The physical conditions in the region are adaptable to the production of rain-fed paddy. In 1971 there were an estimated 6,100 farmers pro- ducing rain-fed paddy, 50 percent of whom were producing rice on five acres or less. About 90 percent were producing paddy on 15 acres or less, and there were about 100 farmers with rice farms larger than 100 acres. 129 130 The Northern Region Rice Industry has been characterized by a rapid acreage expansion. The acreage of rice expanded from about 28,000 acres in 1968 to about 90,000 acres in 1974. During the same period average yields increased from about 800 to 1,200 pounds per acre. The major factors that have contributed to such a rapid increase in rice production in the Northern Region have been: 1. Easy access to free, unutilized bottomlands not requiring a great deal of clearing; 2. Increased imports of tractors and associated equipment in recent years for sale to private farmers and individuals desiring to engage in private custom plowing; 3. Subsidized selling prices of tractors and associated equip- ment with resulting low custom plowing charges and land preparation costs for tractor owners; 4. An increasing guaranteed floor price for paddy as established by the Government Rice Mills Unit; 5. Increased availability of subsidized improved seed and fer- tilizer; 6. Artificially high financial returns resulting from high input subsidies; 7. Prestige associated with land extension and large individual rice farms; 8. More recently, the introduction of subsidized combine har- vesting services by the Ministry of Agriculture. Except for mechanized land preparation, all field activities, including the application of seed and fertilizer, weed control,and harvesting, have been undertaken manually by the vast majority of rice 131 farmers. Few combines were in operation in the region before 1973, although in 1973 the MOA imported and operated in the Northern Region 31 self-propelled combines, machines which were hired to farmers at a heavily subsidized rate. An analysis of the Ministry's combines did indicate that the service encountered organizational problems and heavy losses in its first year of operation. With the rapid expansion of rice production, the demand for casual labor to work on rice farms has dramatically increased in recent years. However, a shortage of labor during the harvest has recently developed and, as a consequence, combine harvesting was introduced into the region. The research methodology employed for this study was described in Chapter III. The Cost Route Survey Method was used to collect farm data by continuously interviewing a sample of farmers over the May 1973 to February 1974 period. During the 1973-74 crop season, 161 farmers were interviewed. The purpose of the field survey was to obtain farm level input/output data in order to estimate the relative financial costs and returns of the major rice production systems in Northern Ghana. The sample size was determined by a fixed budget, which permitted the hiring of 15 field enumerators, and by the number of farmers that an enumerator could effectively interview. All enumerators partici- pated in a lO-day training course prior to the survey. Enumerators interviewed farmers a minimum of twice a week over a 10 month period. To measure labor utilization, data were collected on an activity-by- activity basis, separately for family and hired labor. These data were recorded on the basis of the number of field hours and the type of labor (men, women, and children). For hired labor information was also 132 collected on wage rates, total labor expenditures, and the estimated value of payments in kind. Total production was estimated by a physi- cal count of the number of bags harvested by each farmer. To estimate farm size, each sample farm was measured after harvest by the triangular method. Five major bottomland systems and one upland system were defined fer analysis on the basis of power source used for land preparation and seed variety. Three of the bottomland systems were based upon farmers hiring private tractor-hire services (traditional, improved, and mixed seed varieties); two bottomland systems were based upon tractor owners (traditional and improved seed varieties); and one upland system was based upon farmers using traditional seed varieties and the bullock plow for land preparation. In Chapter IV the private and economic costs and returns were estimated for each of the six major rice production systems. Financial rice enterprise budgets were constructed from survey data, and for each rice enterprise budget a net cash return to operating capital, family labor,and management was derived. The budget data were then used to derive financial returns to (1) family labor, (2) operating capital expenditures and (3) management, as well as costs of production. The returns to family labor among the five bottomland systems ranged from ¢252 for System I to ¢3347 for System V, while the return to operating capital varied from 15 percent of total operating expenditures (System VI) to 66 percent (System III). The five bottomland systems had a high return to management after opportunity costs were assigned to family labor and operating capital. However, for the upland system (System VI) there was a zero return to management, primarily due to 133 the large input of family labor. Among the five bottomland systems, there was a 28 percent varia- tion in the financial costs of production. The 119-acre tractor owner system using improved seed (System V) had the lowest financial cost of production (¢104 per ton), whereas the upland bullock system had the highest cost of production (¢179 per ton). Following a comparative financial analysis of the production systems, an economic analysis was undertaken. In order to evaluate the production systems from an economic point of view, unsubsidized costs of nonlabor inputs were estimated. The resulting economic costs of production among the bottomland systems were,on the average, 47 percent greater than the financial estimates, the variation depending upon the mix of capital items and the relative rates of subsidy among nonlabor resources. Capital-labor ratios were computed and showed that the capital intensity of the two tractor owner systems was high, particularly in the case of the 119-acre system. The analysis revealed that the high C/L ratio was due to two complementary factors: as farm size increased, the use of mechanized harvesting increased and the labor utilization per acre decreased in harvesting and weeding. The method of accounting used to evaluate the production systems altered the relative ranking of the system. When market prices were used to value resources (financial analysis), the large-scale, capital intensive system (System V) had the lowest cost of production (¢104 per ton). However, when economic prices were used, this sytem had the highest cost of production (¢193). This high cost system was character- ized by the highest C/L ratio and the largest acreage, the greatest 134 degree of combine harvesting, and the lowest average labor utilization per acre. In contrast, System III had the lowest economic cost of production (¢l41) and was characterized by the highest yield and labor utilization per acre, the lowest C/L ratio, and the second smallest acreage among bottomland systems. The upland bullock system had the highest cost of production from both the financial and economic point of view. Our analysis showed that rice farmers under present policies are receiving substantial income support through subsidized combine harvest- ing services, seed, fertilizer, and land preparation. The variation in income support from factor subsidies ranged from 26 percent of the net return to operating capital, family labor, and management (System VI) to 150 percent for the large-scale, capital intensive system (System V). The system with the lowest economic cost of production had the lowest proportion of its financial income derived from factor subsidies, whereas the system with the highest economic costs of production had the highest proportion of its income derived from subsidies. The economic analysis further illustrated that all production systems, except the 16.9-acre THS system (System III), generated economic losses from the national point of view. Current policies are encouraging farmers to expand farm size, thereby using subsidized capital intensive and labor-saving production practices. The resulting artificially high incomes provide incentives for farmers to adopt production practices which are financially profitable but uneconomic from the national point of view. Two contrasting production systems--small scale versus large scale--were analyzed in detail in order to identify more clearly the 135 trade-offs among (1) capital requirements, (2) farm employment, (3) pro- ducer income, (4) output,and (5) economic profitability. One production system represented a small-scale, labor intensive approach to rice production and a second represented a large-scale, capital intensive approach to production. The analysis showed that, from a financial point of view, the returns to management were high for both systems and the costs of production were about equal were the lowest of all systems. studied. However, from an economic point of view the small-scale approach to rice production had substantially lower capital requirements and costs of production than did the large-scale system. The small farm system was thus economically profitable, whereas the large-scale pro- duction approach generated substantial economic losses. In Chapter V the employment and income distribution implications of alternative rice production systems were analyzed. The composition of the labor supply in terms of (1) family and hired labor and (2) men, women, and children was identified. Our analysis illustrated a wide variation in the average man-hours per acre among the bottomland production systems, from a high of 220 man-hours for the 16.9-acre tractor-hire system to a low of 38 man-hours for the 119-acre tractor owner system. The principal cause for such variation in employment among the production systems was the method of harvesting and the intensity of the weeding operation. We found that, with the exception of one system, as farm size increased, man-hours per acre in weeding declined. With the exception of the most capital intensive production system, about 73 percent of the labor were employed in harvest acti- vities. For all field activities among all production systems, family 136 labor was supplemented with hired labor. Our analysis also showed that as the acreage of the system increased, the proportion of hired labor increased from a low of 38 percent (System I) to a high of 75 percent (System V) of total labor requirements. In general, the greatest pro— portion of hired labor was employed for cutting, followed by weeding and heaping activities. For all bottomland systems the average propor- tion of the total man-hours supplied by men, women,and children was 57, 33, and 10 percent, respectively. Among pre-harvest activities, men dominated all field activities, particularly the broadcasting of seed. With regards to harvest activities, cutting and heaping were mainly performed by men, while women dominated threshing, winnowing, and bagging. The employment and income implications of a shift from hand harvesting to combine harvestingwere also analyzed. We estimated that at current average yield levels manual harvesting requires 142 man-hours per acre, whereas combine harvesting requires 10 man-hours per acre for bagging. Our analysis showed that at current subsidized charges for combine services, farmers are encouraged to adOpt combine services as it costs less to harvest with a combine than to hire labor. Our esti- mates also revealed that as combine harvesting expands, 22,000 man-days are potentially displaced per 1,000 acres harvested by combine. Thus, if,on the average, 60 percent of the total labor requirements for manual harvesting are hired, we estimate that casual workers would as a group experience a loss in income of about ¢12,700 per 1,000 acres harvested by combine. 137 Tentative Policy Recommendations to Improve the Performance of the Northern Region Rice Production Sector This study provides planners in Ghana with some of the data required to identify the trade-offs among efficiency, output, employ- ment, and income distribution for alternative rice production strate- gies. The survey data have permitted a detailed analysis of the farm- level trade-offs for six rice production systems. However, we do not have regional and macro-economic data to rigorously trace the direct and indirect implications of these alternative production systems for the Northern Region and the national economy. With this caveat, this dissertation will pose major policy issues facing Ghana and then con- clude with tentative recommendations as to how to improve the per- formance of the rice industry in Northern Ghana. Major Policy Issues The present study indicates that the production strategies being pursued in the bottomlands of NOrthern Ghana are providing rice pro- ducers with high financial returns, but at a high cost to the Ghanaian economy. This study has also identified the key policy issues which Ghanaian policy makers should consider: Capital Intensity. The current policies of subsidized land pre- paration and combine harvesting encourage farmers to (l) expand farm size, (2) tolerate low yields per acre, (3) use increasing amounts of imported capital, and (4) reduce labor inputs, especially in harvesting. HOwever, with declining foreign exchange reserves and an anticipated foreign exchange gap in the years ahead, it is unlikely that the present capital intensive approach can be sustained. Thus,it will be necessary for planners to identify production strategies requiring less foreign 138 exchange. One way to reduce the presently high capital-labor ratio in harvesting would be to identify national policies which could stimulate an internal migration of labor at harvest time to ease the present har- vesting constraint and reduce the demand for combine harvesting. Income Distribution and Employment. Capital input subsidies are providing large income transfers to about 100 large scale, capital intensive rice producers. Yet subsidized combine harvesting is dis- placing large numbers of casual workers, and the loss of income to these workers is substantial. To improve income distribution and employ- ment in the northern rice production subsector, a major small fanm rice production campaign should be designed and implemented as soon as possible. Land Expansion. Although land expansion has been rapid among bottomland producers, there are a number of unfavorable medium and long term consequences of this strategy. The land extensive strategy is resulting in low average yields, increasing weed infestation, and declining soil fertility on farms with a low record of fertilizer use. As soil fertility declines on farms with a low record of fertilizer use, farm abandonment will likely increase. A less land extensive strategy could foster soil maintenance and increased yields and, thereby, reduce land expansion and the requirement for an increasing stock of tractors and associated equipment for land preparation. Foreign Exchange Constraint and Factor Subsidy Burden. Our study has shown that the rates of subsidy on capital inputs (mechanized land preparation, combine harvesting, improved seed,and fertilizer) are very high. As increasing numbers of farmers adopt and/or increase their use 139 of subsidized capital resources, there will be an increasing federal budgetary burden to finance subsidies and growing pressure on constrained foreign exchange reserves for the importation of equipment, spare parts, fuel, and fertilizer. Hence, one of the major challenges facing policy makers in the immediate future will be to identify means to reduce both foreign exchange requirements and input subsidies while at the same time providing farmers with incentives to increase rice production. Recommended Policy Reorientation Policy makers undoubtedly believe that there is a need for large- scale, capital intensive farms on the northern bottomlands because of their demonstrated ability to achieve dramatic short-run increases in output. Under the current focus, however, the small rice farmer has been neglected by the government's output strategies. This study illustrates that it is the small farm production campaign which would (1) be a lower cost approach, (2) generate more employment, (3) improve income distribution, and (4) require less foreign exchange and input subsidy support. As a consequence, this author believes Ghanaian policy makers should give careful consideration to a major reorientation of the government's rice production strategy. He recommends that at the current time less emphasis be placed on achieving self-sufficiency by assisting only about 200 large-scale farms, and more emphasis be given to achiev- ing increases in production by large numbers of small farms. A further recommendation proposes that government planners design and implement a small farm rice production campaign. While Ministry of Agriculture personnel are in a far better position to judge the desirable scope 140 of a small farmer campaign, this author would maintain that it is probable that the Ministry of Agriculture could include 500 farmers or 2,500 acres in the first year and expand to about 2,000 farmers or 10,000 acres over five years. The following section will trace the output, employment, and income distribution implications of both small farmer and large farmer rice production strategies. Small Vs. Large Farmer Production Strategies: Output, Employment, and Income Distribution Inplications of a 35,000 Acre Rice Production Campaign In this section we will analyze the employment and income dis- tribution implications of two hypothetical production strategies.1 In the analysis whidh follows we are assuming that a production campaign is focused on 35,000 acres of bottomland in Northern Ghana. The input-output data used in the analysis is drawn from the rice enterprise budgets in Chapter IV.2 The small farm system used tractor- hire services for land preparation, improved seed, above average fer- tilizer-use, and manual harvesting. The large farm system was composed of tractor owners who used improved seed, fertilizer, and combine harvesting. Output, Emplgyment, and Income Distribution Effects If a rice production campaign were focused upon labor intensive, small farms, we estimate that about 24,500 tons of paddy would be 1For a good example of this approach and the need for empirical farm level data, see Marsdan, 1969. 2See "Comparison Between Small Farm and Large Farm Rice Production Strategies" Chapter IV. 141 produced on 35,000 acres. If the campaign were focused upon large-scale, capital intensive farms, however, about 28,700 tons of paddy would be produced (Table 6.1). The large farm strategy would thus produce about 17 percent more total output; but, as our analysis in Chapter IV shows, the output by the capital intensive strategy would be produced at a high cost to society. Table 6.1. Projected Output, Employment, and Income Impacts of a 35,000 Acre Rice Production Campaign in Northern Ghana: Small Farm Vs. Large Farm Strategies Indicator Units Small Farm Large Farm Per 35,000 Per 35,000 Acre Acres Acre Acres Output Bags 8.71 304,500 10.22 357,000 Tons 24,470 28,688 Employment Man-Days 30.61 1,071,000 6.82 238,000 Man-Months 44,625 9,917 Net Farm Income Cedis 601 2,100,000 632 2,205,000 Input Subsidies Cedis 273 945,000 813 2,835,000 1Derived from Table 4.15. Man-hours are converted to man-days by assuming six man-hours per man-day. Man-days are converted to man-months by assuming 24 working days per man-month. 2Derived from Table 4.14. 3Derived from subtracting nonlabor financial expenditures (Rice Enterprise Budgets) from nonlabor economic costs (Table 4.17) and con- verting each to a per acre basis. Aggregate employment would be markedly different for the two pro- duction strategies. The small farm strategy would generate employment of 1,071,000 man-days, whereas the large scale, capital intensive strategy would employ only 238,000 man-days, or realize 77 percent less 142 aggregate employment.3 In terms of net farm income, bo£h_strategies would generate about equal aggregate income. However, under present policies, the large- scale, capital intensive system‘would require ¢2.8 million in government support to producers in the ferm of capital input subsidies. On the other hand, a small farm production campaign would require only ¢0.9 nullion, or about 66 percent less from the government budget. Income Distribution Implications Table 6.2 reveals the dramatic difference in the number of pro- ducers between small farm and large farm strategies. If the average size of the rice enterprise were four acres, then about 8,750 producers would be producing rice on 35,000 acres. If, on the other hand, the average farm size were 100 acres,then only 350 would be required to exhaust 35,000 acres. If a small farm production campaign were pursued and the average farm size were four acres, then our estimates show that 8,750 rice producers would each receive about ¢240 of net income from the produc- tion of rice. If, on the other hand, a production campaign focused on one hundred acre farms, 350 farmers would each receive about ¢6,300 in net income. The latter approach concentrates high producer incomes among a few farmers and would greatly aggravate income distribution in Northern Ghana. Under a small farm campaign, given current subsidy policies on capital inputs, ¢945,000 in factor subsidies would be 3As shown in Chapter IV,the primary reason for the difference in labor utilization between these two strategies is employment in harvest- ing and weeding. 143 .mumsumm no woman: 0:» >3 AH.0 magnet uwmmcmuu ucmficum>om mummmuomm onu mo moam> on» mcflpflbwm >3 ombaumo m .muoahmm mo Hones: 0:» ha AH.m mammav maoocfi Show um: mumooummm ocwcfl>flo an om>fiuwoa Hmm.vm mem.ma baa com com mmmtm @ oom.m~ oom.ma mba com own omh.m v ooa.w oom.w 0mm ooa oma oom.ha m IIIIIII IIIIImHUmOIIIIIIIIIII: InllllunnllnmwcmonlIIIIIIIII: umeumm mom Ammuodv umfiumm Hem Amwuodv Numsumm new A mEoocH muweumm Ewen «seamen Hem mEoocH muosumm Ewen Hemmcmue .u.>oo ammo umz mo .02 Mo ouem umwmcmua :mmo uwz mo .oz mo chem Shem mmumq Ehmm HHMEm somoummd Such comma .m> finch Hamsm "mango cumsuuoz CH cmemmEmu cofluosooum mofim whom ooo.mm m mo mcoflumoHHmEH cowusnwuumwo oEOOCH Hmosooum .N.® mHQMB 144 required to support 8,750 four acre farmers in terms of subsidized factor prices. On the other hand, ¢2.8 million would be required to support 350, one hundred acre farmers under a large farm campaign. The government support would thus be ¢108 and ¢8,lOO per producer under a small and large farm production canpaign, respectively. The resulting distribution of government funds favors high income as opposed to low income members of society. Employment Implications for Casual Workers The potential income received by casual workers depends in part on the degree to which family labor is supplemented by hired labor. If, under a small farm campaign, farmers on the average hired 50 per- cent of their labor requirement, then about ¢535,500 would be paid to hired laborers. If large-scale, capital intensive producers hired 90 percent of their labor requirement, then only £214,200, or 40 percent less would be paid to casual laborers. Thus, a small farm production campaign would generate more employment and income for casual workers (Table 6.3). Summary Aggregate employment would be markedly different depending on whether a small farm or a large farm, capital intensive strategy is pursued by Ghana in the future. We estimate that a small farm produc- tion strategy would generate one million man-days of employment, whereas a large-scale, capital intensive strategy would employ only 240,000 man-days, or 77 percent less labor. In terms of net farm income, both strategies would generate about the same net farm income in the aggregate. A small farm strategy would generate an estimated 145 Table 6.3. Income Distribution Implications of a 35,000 Acre Rice Production Campaign for Casual WOrkers in Northern Ghana: Small Farm Vs. Large Farm Approach Proportion of the Income Paid to Labor Requirement Casual Workers Hired . (Percent) Small Large Farm Farm 30 ¢321,300 ¢ 71,400 50 535,500 119,000 70 749,700 166,600 90 963,900 214,200 1Derived by multiplying man-days from Table 6.1 by the appropriate proportion of the labor requirement hired and multiply- ing the resulting sum by ¢0.95 per man-day. ¢240 in net income for 8,750 small farmers, whereas a large farm strategy would concentrate very high producer incomes (over ¢12,000) among only 200-400 rice producers. With the government's present subsidy policies, a large farm strategy would require an estimated ¢2.8 million to subsidize factor prices, whereas a small farm strategy would require only ¢0.9 million. The employment of casual workers would be dramatically different under the two production strategies, and the amounts of aggregate income paid to casual workers would be more than double under a small farm produc- tion campaign. 146 Recommended Components of a Small Farm Production Campaign Policy changes are identified for the evaluation of Ghanaian policy makers. For each, a tentative policy action is recommended for the consideration of government decision makers who determine national agricultural policy and planners involved in the rice development program. 1. Reorienting the Extension Service to Focus on Improving the Production Practices of Small Farmers. In the Northern Regiontfluafocus of the Extension Service has been almost exclusively on large scale, capital intensive rice producers. The proposed policy reorientation will require extension officers to focus upon the production problems of small farmers which will, in turn, demand the retraining of extension personnel in order for them to be effective change agents among the new target group. Since small farmers are also engaged in the production of other crops such as maize, sorghum, groundnuts, and yams, the training program should not be ex- clusively centered on the production problems and recommended cultural practices of rice. Extension officers will have to be equally effec- tive change agents for other crops in order to gain the confidence of farmers and in order to assist with multiple enterprise production problems. Extension officers should be trained in methods for establishing effective demonstration plots to show farmers how to use improved cultural practices and what the tangible benefits of improved pro- duction techniques can be. The training program should also teach extension officers the need and benefits of (l) retarding land exten- sion, (2) increasing yields per acre, (3) maintaining soil fertility, 147 and (4) relying on manual methods of harvesting. Our analysis shows that most farmers are not following recommended production practices; hence, there is a great potential for increasing rice production by encouraging the farmers to apply recommended rates of seed and fertil- izer, to undertake timely field activities, to intensify labor use in weeding and manual harvesting,to shorten the harvesting duration, and to plant a combination of traditional and improved seeds, thereby ndnimizing shattering problems. Extension officers should also be trained to assist farmers to obtain improved seed, fertilizer, hired labor, and credit to purchase improved inputs. 2. Identify and Promote Small-Scale Paddy Threshers. To remove some of the drudgery associated with manual harvesting and to shorten the duration of the harvesting period, the MOA should embark on a search fer proven, low-cost hand threshers.r The Inter- national Institutes, such as IITA and IRI, as well as aid donors could be asked to assist with the identification and provision of appropriate small-scale threshers for trial purposes. The MOA in collaboration with the Ghanaian Society of Agricultural Engineers should undertake on-farm performance trials to evaluate the relative effectiveness and durability of a range of hand threshers. Farmers, and particularly women, should be involved in these trials to identify the machine(s) they prefer. When one to two effective lowbcost threshers are identified, the MOA should import about 50-100 hand threshers to be sold to farmers. If sales and performance then prove to be good, larger quantities should be imported or,preferably, manufactured locally. 148 3. Foster an Internal Migration of Seasonal Labor to the Northern Region. To encourage an internal migration of unemployed and underemployed labor to the Northern Region to harvest rice during October through December, each September the government should launch a major propa- ganda campaign to inform the public of the employment opportunities on northern rice farms during the harvesting season. The government should also consider establishing low-cost labor stations where workers seeking farm employment can congregate and farmers can recruit contract labor. Encouraging internal migration as a source of supply of labor for harvesting should be the immediate priority. However, it is likely that weeding requirements will become increasinly critical as the harvesting constraint is minimized. Therefore, the MOA should deter- mine to what extent migratory labor will be required and, if necessary, launch a propaganda campaign to encourage labor migration to the bottom- lands to engage in weeding activities. 4. Encourage A Combination of Seed Varieties to Ease the Harvest- ing Constraint. To help ease the harvesting constraint, the Extension Service should encourage farmers to plant part of their farms to short matur- ing improved varieties and part to longer maturing traditional seed varieties. If the shorter maturing, higher yielding improved seed varieties are harvested first, followed by the longer maturing tradi- tional varieties, the harvesting period can be prolonged and the poten- tial shattering losses minimized. The proportion of the farm planted to improved varieties will depend upon farm size and the supply of labor individual farmers can realistically expect to engage during the harvesting operation. 149 Longer maturing varieties would significantly ease the shatter- ing losses and thereby reduce a critical seasonal harvesting bottleneck. Moreover, a shift to longer maturing varieties would complement hand harvesting methods and reduce the potential benefits of early combine harvesting. Other desirable seed characteristics should be incorpor- ated as a selection criteria for the screening trials. These would include (1) resistance to blast, (2) resistance to shattering, and (3) yield response to low application rates of fertilizer. Given current and anticipated world fertilizer prices, it would be highly advantageous to identify medium yielding paddy varieties which do not require high fertilizer application rates in order to achieve desirable yield levels. There is an important economic trade- off between (1) yield and (2) fertilizer requirement which, given Ghana'S' foreign exchange position, probably should be weighted more heavily toward medium yield varieties requiring less fertilizer. 5. PrOposed Seed Sales Policy. Improved seed varieties are multiplied and sold to farmers by the Seed Multiplication Unit of the MOA. Farmers purchase seed with cash or with credit vouchers from lending institutions and are not required to purchase fertilizer when they purchase improved seed. In order to foster increased yields per acre, it is recommended that improved rice seed be sold to farmers only under a condition that they present evidence (sales receipt) that they have purchased the appro— priate complement of fertilizer. If farmers were required to purchase recommended quantities of fertilizer, this would encourage them to shift away from a land extensive approach to a yield intensive pro- duction system. 150 6. Seed Selection Trials The Crops and Soils Research Institute at its NOrthern Region Nyanpala Station should undertake screening trials to identify high yielding seed varieties which are longer maturing than the 115-day improved varieties presently being promoted. Varieties with about l30-day maturity would be more appropriate for the Northern Region ecological zone. The West African Rice Development Association (WARDA) and IITA at Ibadan should be encouraged to suggest and supply appropriate seed stock for the screening trials. 7. Expanded Soil Testing Program. The MOA has a soil sampling program whereby farmers and Exten- sion Officers can bring to the Regional Headquarters soil samples for testing pH, organic matter, and nutrient content for the purpose of determining appropriate cultural practices. Given problems of declin- ing soil fertility and weed infestation on some farms and the reports of rice farm abandonment, we recommended that this program be greatly expanded. A systematic soil sampling campaign should be undertaken, taking care to obtain, in addition to soil samples, a history of cul- tural practices (mechanization practices, seed varieties, fertilizer use) and estimated yields for each sample farm. It is recommended that about 200—300 rice farms drawn from major bottomland areas throughout the region be established as an ongoing rice land soil testing sample. The soils of these sample farms should be tested annually over five years. Each year data should be obtained about cultural practices and paddy yield. From this program the MOA can objectively determine what is happening to soil fertility and the relative nutrient status of bottomland rice soils over time. The 151 analysis will be useful in evaluating current recommended cultural practices, particularly fertilizer recommendations. 8. Weed Control Techniques. The present study has identified problems of (l) weed infesta- tion, (2) late or prolonged weeding activities,and (3) low labor utilization in weeding on large farms. At the present time the MOA recommends two to three mechanical harrowings and manual weeding to control weeds. The MOA has undertaken aerial spraying on a pilot basis to determine if the technique is effective for weed control on large farms. For the majority of farmers, however, this control technique is not a practical solution. As a short term policy objective, we encourage a migration of labor to assist with weeding activities. In the longer term.we believe that simple, low-cost intermediate technology will be required. To this end we recommend field trials to identify low-cost hand sprayers which can be locally manufactured. In collaboration with these trials, low; cost, effective, easy to use, and safe chemicals need to be identified. 9. Reorient Credit Eligibility Requirements and Substantially Increase the Stock Funds for Production Loans to Small to Medium Sized Farms. In order to focus a production campaign on small rice farms, the credit eligibility policies of the Agricultural Development Bank (ADB) should be changed. Credit priority should be given to rice farmers with less than 10 acres, and farmers with more than 50 acres should be excluded from eligibility for ADB loans.4 Larger farmers should be required to use other commercial lending institutions. 4Present lending policies of the ADB favor rice farmers with above 50 acres. Among rice farmers receiving credit, the majority are tractor owners. 152 ‘ To support the proposed production strategies, the ADB should establish two types of lending policies for rice farmers. One group of farmers would be provided with credit for land preparation, seed and fertilizer, and hired labor. No farmer would receive credit for combine harvesting. The current policy of providing credit in kind for seed and fertilizer should be continued. Further, credit policies should be designed in such a manner that a high proportion of the pro- duction loans are used for the purchase of improved seed and fertilizer (at recommended rates) and the contracting of hired labor for manual harvesting and weeding. The second group of farmers would receive seasonal credit only for the purpose of hiring labor for harvesting. It is envisioned that as much as 25 percent of the farm loans should be in this latter category, thus providing added Sipport to a labor intensive harvesting strategy. 10. Improvement in the Input Distribution and Product Market- ing Systems. The present plan to establish a network of small, rural, low- cost input supply depots in major production centers in the north should be expanded to improve the distribution of improved seed and fertilizer to small farmers.5 In addition to supplying production inputs, it is recommended that these depots be used at harvest time as paddy buying centers of the Rice Mills Unit of the MOA. It is further recommended that "supply depots/buying centers" be 5The Ghanaian-German Agricultural Development Project [for the] NOrthern and Upper Regions has conceived of this plan and is presently financing a small network of rural input supply depots in the Northern and Upper Regions. 153 used as an institutional arrangement to provide credit to farmers as a means to increase the adoption of improved seed and fertilizer. Farmers with less than 10 acres who meet minimum standards6 should be supplied in May and June with appropriate quantities of improved seed and fertilizer from the rural supply depots. Farmers would not be required to pay for these inputs at the time of delivery, but rather would receive the inputs as credit-in-kind to be repaid at harvest time. Farmers receiving such credit would then repay their loansby delivering to the supply depot at harvest time the equivalent value of paddy, the value being determined by the going market price. The paddy would be turned over to the Rice Mills Unit, and the Mill would pay the Ministry of Agriculture for the paddy received. Recommended Policy Changes for Lagge-Scale Rice Production Several of the above policy recommendations for a small farm production campaign are equally appropriate for the large-scale farms. The three policy recommendations which follow are proposed to reduce the economic losses? resulting from large-scale rice production. 1. Retard the Expansion of Combine Harvesting. Over the next two crop seasons, the MOA should increase its custom rates from the present charge of ¢l.00 per bag to the estimated 61t is envisioned that District Extension Committees on the recommendation of local extension officers would screen and certify farmers who are eligible for credit. During certification the quantity of seed and fertilizer to be provided as credit-in-kind would be stipulated. 7Economic losses from the national point of view. Input sub- sidies are reduced to zero and output is valued at the economic import parity price. See Chapter IV for the procedure adopted to calculate economic costs and benefits. 154 economic cost of ¢4.00 per bag. This policy change would have several desirable spin-offs. Private combine owners engaged in custom harvest- ing would follow the lead of the MOA and increase their contract charges so that the average cost of combine harvesting would be above the cost of manual methods of harvesting, thus making labor competitive with the combine. This would retard future labor displacement and increase the demand for casual labor in harvesting. Given the short- age of casual workers in the bottomland areas, wage rates would increase. Expanded employment opportunities coupled with increased wage rates would then stimulate the migration of unemployed and under- employed labor to the Northern Region to seek work during the harvest- ing period. The Ministry should discontinue its combine harvesting services after the present combines have deteriorated. In addition, the govern- ment should discontinue importing combines and import only the required stock of spare parts to maintain the present privately owned machines. It is the author's opinion that combines should be used only on the large rice farms over 50 acres. Caution should be exercised so as not to create an excess capacity of combine harvesting services because of the major labor displacement that would be realized as documented by our analysis. Farms of less than 50 acres should be required to harvest manually with the assistance of small-scale paddy threshers. 2. Increase the Cost of Land Preparation by Imposing a Tariff on Imported Tractors and Associated Equipment. In order to retard land expansion, the cost of land preparation must be increased. This can be achieved by placing a tariff on imported tractors and associated equipment. It is thus recommended that the 15S government place a tariff of about 50 percent on imported tractors and associated equipment. The increased tariff, which should be phased over two to three years, will increase the cost of land preparation among tractor owners and force private contractors to increase their custom rates. This higher cost of land preparation will then dis- courage extensive production methods and encourage yield increasing techniques. We believe that farmers will be encouraged to increase yields per acre to maintain their present returns from rice production. Given the yield response to fertilizer on field trials, increased yields of 30-50 percent are within reach of farmers applying fertilizer at recommended rates and undertaking other recommended cultural practices. 3. Inventory Tractors and Associated Equipment to Determine if Tractor Imports Should be Temporarily Suspended. In addition to a tariff on imported machinery and equipment, the MOA should immediately take an inventory of the number of tractors operating in the region and identify the approximate acreage that can be mechanically prepared with the present stock of tractors and asso-- ciated contract services. The purpose of such an inventory should be to determine whether tractor imports should be temporarily suspended and to ascertain the estimated import requirements of associated equip- ment and spare parts. A temporary shortage of plowing services would assist with increasing contract charges,thereby discouraging farmers from expanding farm size. There would be an additional benefit in that tractor owners would be encouraged to increase the income earning utilization of their machinery. 156 4. Create Land Values by Imposing a Land Tax. In order to discourage farmers from expanding farm size too fast, the government should consider the merits of imposing a land tax on bottomland rice producers. This imposition, in collaboration with the tariff on imported machinery, would discourage land expansion. Free access to bottomland is a major contributing cause of land expansion. The current cost of land, which is embodied only in the cost of clearing, does not reflect the economic value of land. A land tax would increase the cost of land and thereby create land values. The added cost of land would then retard extensive land use, and farmers would be encouraged to increase their yield in order to pay for the tax and to maintain their farm income. To obtain the desired results from a land tax, the tax should be imposed on a per acre basis. The result would be that large farmers would have a greater tax bill than small farmers. Policy makers might consider a graduated tax whereby farms of less than 10 acres pay a tax at a low rate, farms of between 10 and 50 acres pay a higher rate, and farms of above 50 acres pay the highest rate. Such a graduated tax would thus place a higher tax burden on large farmers who utilize a greater quantity of the limited stock of bottomlands. It is esti- mated that the average tax rate to be effective should be equal to the financial value of one-half bag of paddy per acre. At the current guaranteed floor price this would amount to an average tax rate of ¢6.00 per acre. APPENDICES APPENDIX A ADDITIONAL INFORMATION ABOUT TRACTOR OWNERS, FARMERS HIRING PRIVATE TRACTOR SERVICES, AND BULLOCK FARMERS APPENDIX A ADDITIONAL INFORMATION ABOUT TRACTOR OWNERS, FARMERS HIRING PRIVATE TRACTOR SERVICES, AND BULLOCK FARMERS Introduction The objective of this appendix is to provide the reader with additional information about tractor owners, farmers hiring tractor services for land preparation, and bullock farmers. Tractor Owners Introduction There were 33 farms in the sample classified as farms where the source of power for initial land preparation was obtained from tractor ownership.1 Of these farms, 3O tractor owners operated 33 farms: the information which follows relates to 25 of these tractor owners.2 Occupational Background3 Most tractor owners, in addition to producing rice, were under- taking private contract plowing and harrowing for other rice farmers. A relatively large proportion of the tractor owners (60-80 percent) 1Improved seed was used by 19 farmers, traditional seed by 10, and a combination of improved and traditional seed was used by four farmers. ' 2 , . . Since rice owners were out of the region at the time of the survey form upon which this data is based was administered, it was not possible to obtain the information from them. 3This section is based upon informal interviews during the course of the survey because it was found during pre-testing that it was not possible to secure these data through a formal questionnaire. 157 158 had occupations other than rice farming and private contract work. Most of these were retail and wholesale traders, professionals (e.g., businessmen, lawyers, and doctors), and civil servants. Many tractor owners believed that they could pay for their equipment within two years by producing rice on their own farm and undertaking custom-hire work for other farmers. Number of Years Producing Rice and Farm Size The 25 tractor owners had been producing rice for an average of about six years. These farmers had been tractor owners from one to five years which implies that they relied upon tractor hire services for initial land preparation before purchasing their own tractors. The range in farm size among the tractor owners ranged from eight to 469 acres, and the average farm size was 107 acres. The author indirectly estimated that the tractor owners were undertaking the equivalent of 200 acres of contract plowing. Acreage Expansion Over the 1971-74 crop seasons, 20 tractor owners reported that they expanded their farm size by an average of 104 percent. Over the 1972-74 production seasons, 25 tractor owners expanded their acreage by 76 percent.4 Farm Abandonment In 1973-74 the 25 tractor owners had been producing paddy on the farm for 3.7 years, and they had been rice farmers for 6.4 years. Six 4These estimates are based on the farmers estimates rather than field measurements over time. 159 of the 25 farmers had previously abandoned a farm upon which they had operated, on the average, for four years. The reasons given for aban- doning the farms were as follows: two complained that their farm had developed too many weeds; three indicated that they were unable to expand their farms because of too many trees to be removed; one farmer complained of a loss of soil fertility. Use of Improved Practices Table A.1 reports the use of selected improved practices among the 25 tractor owners. During the survey year 64 percent of the tractor owners were using improved seed. The percentage using compound fer- tilizer, ammonium sulfate, and combine harvesting was, respectively, 80, 36, and 52 percent. The rate of adoption of combine harvesting and compound fertilizer was greatest among the five improved practices. During the 1971-72 production season 11 farmers were using compound fertilizer, and three farmers hired combine harvesters. However, by the survey year 20 tractor owners were using compound fertilizer and 13 were hiring combine harvesters. Labor Recruiting Practices Since many farmers complain of difficulty in recuriting labor to work on their farms, we asked the tractor owners from where they recruited their labor and how difficult it was to obtain labor to work on their farms. Tamale, the regional capital, is an important source of farm labor for tractor owners; however, villages surrounding rice farms are also important sources of labor. 160 Table A.1. Use of Selected Improved Practices Among 25 Sample Tractor Owners Improved Number of Number of Tractor Owners Using Practice Observations Improved Practices 1971/72 1972/73 1973/74 No. % No. No. % Improved Seed 15 13 48 l4 16 64 Compound Fer- tilizer 18 ll 44 19 20 80 Ammonium Sulfate 7 7 28 12 9 36 Tractor Plowing 24 21 84 25 25 100 Combine Harvesting l3 3 12 5 13 52 1Data were not obtained from one tractor owner regarding his first year of using improved practices. In addition, one farmer could not recall when he started using compound fertilizer. Source of Funds for Rice Farming Farmers were willing to reveal their source of funds for rice farming but not the absolute amount of funds withdrawn from savings and other business enterprises, or borrowed in the form of credit. In 1973-74 10 of the tractor owners obtained their funds from their own savings, 11 from savings and bank credit, and four from bank credit only. Crops Produced in Addition to Rice Of the 25 tractor owners, 14 produced crops in addition to rice. Of those producing other crops, the average acreage in other crops was 11 acres. The major crOps produced were maize and sorghum followed by yams and millet. 161 Farmers Hiring Tractor Services for Initial Land Preparation Introduction There were 83 sample farms where private tractor services were hired for initial land preparation. The information which follows relates to 79 farmers.5 Number of Years Producing Rice and Farm Size The number of years these farmers had been producing rice varied from one year (three farmers) to 17 years. The average number of years the 79 farmers had been producing rice was about four years, or about two years less than the tractor owners. The variation in farm size among the 79 farmers ranged from 2.8 acres to 74.0 acres, and the average farm size was 18.3 acres; the average farm size among sample tractor owners was 107 acres. Acreage Expansion Over the 1971-74 crop seasons, the last being the survey year, 53 farmers, on the average, expanded their farm size by 148 percent. Over the past two crOp seasons 41 farmers expanded their farm size, on the average, by 137 percent, 24 farmers had no change in farm size, and nine farmers reduced their farm size by 42 percent. Farm Abandonment The mean number of years that these farmers had been producing rice on the farms they were operating during the survey year was three years. Twenty-six of the 79 farmers (or 33 percent) indicated that 5One farmer had three farms in the sample, and we were unable to interview two farmers after harvest with the questionnaire upon which these data are based. 162 they had abandoned a rice farm since they started producing rice. Among the respondents, the variation in the number of years they pro- duced rice on the farm before abandoning it ranged from one to five years; the average number of years of producing rice before abandon- ment was 3.2 years. The main reasons given for abandonment, in order of frequency, were as follows: declining fertility; increasing weed problems; too much water; farms too small for extending; and farms too far from the village. Adoption of Improved Practices During 1973-74, 66 percent of the sample farmers using tractor hire services were using improved seed and 71 percent were using com- pound fertilizer. The rate of adoption for five improved practices over three years was greatest for compound fertilizer followed by improved seed. The use of ammonium sulfate and combine harvesting was relatively unimportant among these farmers. Labor Recruiting Practices The most important source of hired labor for this group of rice farmers is their own village and surrounding villages. Only 11 of the 74 responding farmers, or 15 percent, obtained at least part of their labor requirements from Tamale, the regional capital. Fifty-two per- cent of the tractor owners described earlier obtained at least part of their labor from the regional capital. Source of Funds for Rice Farming For the crop season surveyed, 59 (75 percent) of the farmers in this category obtained their funds from savings, l4 (18 percent) obtained 163 credit from a lending institution, three financed their costs from personal savings and bank credit, one farmer obtained a loan for part of his costs from a trader, and two farmers did not respond. Table A.2. Use of Selected Improved Practices Among 79 Farmers Using Tractor Hire Services for Initial Land Preparation Improved Number of No. of Farmers Using Improved Practice Observations Practice 1 2 1971/72 1972/73 1973/74 No. % No. No. % Improved Seed 53 21 37 39 53 66 compound Fer- tilizer 57 15 26 24 57 71 Ammonium Sulfate l9 8 14 ll 19 24 Tractor Plowing 79 54 95 76 79 100 Combine Harvesting 8 1 2 2 ll 14 1Based upon the response of 53 out of 57 farmers who were produc- ing rice in 1971-72; thus, the percentage is computed on the basis of 57 farmers. 2Seventy-six of the 79 farmers were producing rice in 1972—73; thus, the percentage is computed on the basis of 76 farmers. Crops Produced in Addition to Rice Fifty-five of the 79 farmers (70 percent) said that they produced other crops in addition to rice. We found that a greater proportion of the farmers with small rice farms produced other crops than did farmers with large rice farms. Eighty-eight percent of the farmers with rice farms less than five acres in size produced other cr0ps, and only 52 percent of the farmers with rice farms over 25 acres produced other crops (Table A.3). 164 Table A.3. The Relative Importance of Rice and Other Crops Among 79 Sample Rice Farmers Hiring Tractor Services for Initial Land Preparation During the 1973-74 Crop Season Acres of Rice Number of Farmers Average Acres of —- Farmers Producing Other Crops Range Average Other 1 2 Crops Declared Adjusted t" I No. % E Less than i 5.0 3.8 17 15 88 9.2 6.3 I 5.0-15.0 10.0 29 21 72 8.7 6.0 E 15.1-25.0 17.9 10 7 70 13.9 9.6 25.0— 39.8 23 12 52 15.8 10.9 lAs declared by farmers. 2On the average, this sample of rice farmers over-declared their rice farms acreage by 31 percent. The declared acreage of other crops is assumed to be equally over—declared. The other crOps produced by these farmers were maize, sorghum, millet, groundnuts, yams, cassava, and beans. Of these crops, 46 of the farmers produced maize, 41 yams, 39 sorghum, 33 millet, and 21 groundnuts. Only three farmers reported producing cassava and two beans. Smallholders Using the Bullock Plow for Initial Land Preparation Introduction There were 14 sample farms where farmers were using traditional seed and a bullock plow for initial land preparation. Since two of the sample farmers operated two rice farms, the information which follows relates to 12 farmers. 165 Number of Years Producing Rice The number of years bullock farmers had been producing rice varied from four years to 22 years, and the average number of years was about 9 years. Among these farmers, farm size varied from 0.3 to 4.6 acres, and the average size of the rice farm was 1.1 acres. Acreage Expansion Over the last three crop seasons, the last being the survey year, the 12 bullock farmers, on the average, only expanded the size of their rice farms by one percent. Of the 12 farmers, four reduced their farm size, on the average, by 38 percent; five farmers had no change in farm size, and three farmers expanded their rice farms, on the average, by 54 percent. Farm Abandonment The mean number of years the sample bullock farmers had been producing rice on their farms was about eight years. Only one of the bullock farmers indicated that he had abandoned a rice farm since he started producing rice. Unlike the bottomland rice producers, aban- donment of upland rice farms does not occur often among upland bullock farmers. We are puzzled by the comparative results on farm aband- donment between this subsample of farmers and the bottomland farmers reported in previous sections. Is it that traditional seed varieties with little or no fertilizer are better suited to upland soils than improved varieties on bottomland soils with declining soil fertility? 166 Use of Improved Practices All bullock farmers were using traditional seed during the survey year. The variation in the number of years farmers had used the bullock plow for initial land preparation ranged from one to 16 years, and the average number of years was nine. Five of the 12 farmers had used bullocks for, on the average, 2.6 years longer than they had been rice farmers (i.e., they used bullocks for land preparation of other crops before they started producing rice). Five of the bullock farmers, on the average, started using the bullock 4.6 years after they had been producing rice, and one farmer started using a bullock team the same year he started producing rice. Labor Recruiting Practices Bullock farmers operating on relatively small acreages, like bottomland rice farmers, hired labor for field activities. Nine of the 12 bullock farmers hired labor, three of which recruited labor from their own village, three from their own village and surrounding villages, and three from surrounding villages only. Source of Funds for Rice Farming During the survey year, nine of the 12 bullock farmers obtained their funds for rice farming from their own savings, two obtained credit from the Agricultural Development Bank, which was arranged by a local church mission for the purchase of a bullock team and plow, and one farmer obtained credit from a local trader. 167 Crops Produced in Addition to Rice The twelve sample bullock farmers all produced crops in addition to rice, and the total acreage of other crops was greater than the acreage of rice. Among the other crops produced, eight farmers pro- duced groundnuts; five a mixture of sorghum and millet; six millet in pure stand and five sorghum in pure stand, two a mixture of sorghum and beans. Other crops produced though less important among the sample farmers were a mixture of sorghum and beans, maize in pure stand, yams, and a mixture of groundnuts and beans. APPENDIX B SELECTED ATTRIBUTES OF SAMPLE FARMS FOR SIX RICE PRODUCTION SYSTEMS IN NORTHERN GHANA, 1973-74 mma mha moa when .mxz m.H >.ma o.m~ o.mH mxwmz coflumm susouo >pomm .mxz m.m mumccmh\a anmcuomm\w nonsmomo\H cuc02\xwms mcwumm>umm no open mcwpcm .mxz m.H Hmosm>02\m Honsmowo\m Honou00\v cu:02\xmez mcwumm>umm mo oumo mcflccwoom waco mumshmm 039 mcfiomwz cam .mxz H.v HmoEmummm\H HonouoO\v um5094\m nuc02\xom3 mcwpmmz umH mo mama mcwocm .mxz m.a umsms<\m umnouooxa senn\a gucozxxmmz Amsumm HNV mcwpooz umH mo mono maficcwmom m.HH H.m m.hm 0.0 once Hem .mnq cwmowuez mo wumm coaumowammm am we one mm whoa umm .mnu mums mcacmmm .mxz w.m >Hch\m umomc¢\m mm2\m cucoz\xowz mumo mcfiucmam .3st 382? F2} 555x83 195mm 3.3 mcflsowumm uma mo mumo ocsb\m umcmc<\a Heum¢\m nuc02\xm03 mcwsoam «0 euro e.m ~.m m.mH v.0 momm .na owe whoa Hem name» H.H m.m o.m o.H mumow Show anemone co monotone moem memo» mo .02 m.v 0.5H o.m mummy comm omocoowm mm: weaken whee» mo .02 m.va m.ma o.v> m.m memos chem such Amsumm mmv comm HecOfluwcmua I mme H Swuwwm Edsexmz acaflcfiz coaumH>wo «1 mousoflnuuc_5umm cuspcmum cums wmcmm hues: cam Emummm cofluoccoum vnnmhma .mcmco cuwcuu02ce msmummm coeuocpoum comm xflm you wanna mamsmm mo mousoflubum cwuomaom .H.m wanes 168 169 .mx3 m.m umcm:<\v uonm>OZ\H madn\m cuc02\xmmz maepmwz and mo mumo mcflpcm .mxz e.m umsmsa\~ umnouooxe saunxm nucoz\xmmz imsumm oec mcfipomz uma mo mumo mcflccflomm m.Hm 0.5m m.mm o.o whoa umm .mnu cwoouuaz mo mumm coaumoeammc em on mom mm whoa uwm .mna 00mm mahommm .mxz m.m >H56\N bmcmcm\m oscb\a cuc02\x003 wumo mcflucmHm Awaken NV mcwzouumm cum mo oumo .mxs H.m sasn\a umsmsa\m 0::6\H cucoz\x003 Amsnmm may mcfi3ouumm com mo mumo .mxz m.m wasn\v umsm5¢\m >m2\¢ cuc02\xwmz ”mason mvv mow3owumm uwa no open .mxz o.m mech\m >HDb\v >m2\m npc02\xm03 mcwzoam mo mama m.e m.m e.- H.H meme .nu owe whoa umm came» ©.H m.m o.m o.H muse» such ucmwwum co couscoum comm name» mo .02 m.v o.bH o.H mummw OOHm UmOSCOHm mam weaken meme» «0 .oz 0.5H N.Hm m.mo b.~ mound wuwm such images was 600m em>oumsH u mes HH Emumwm sosfixmz scams“: cofiumfl>mo mousofluuum.aumm pHMUcMum cmmz emcmm nudes pom Emumww coeuocoowm cmscwucou .H.m magma 170 h.mm m.mm 0.mn 0.0 muoa uma .moa ammouuez mo mumm coflumoflamma 0v mm mom so mama Hma .mau mumm mawommm .mxz H.m >H50\N umcmcm\a mccb\m auc02\xmmz mama mcwucmaa >H90\m umcmcc\a mach\m >Hca\xmm3 AmEHma 0HV 0cfizouumm umH mo mama mcch\m >ch\v >mz\m aucoz\xmm3 ocwzoam mo muma e.m m.m o.mH m.H mama .nq owe whom umm mama» m.m v.0 0.0 0.H Hmm» Ehmm ucmmmua co pmocpoua mowm mummy mo .oz m.m 0.0a 0.m mummw mowm cmoacowa mm: HmEHmm mummy mo .02 h.NH 0.0a m.mv H.m mmuoa mNflm Ehmh Amsumm HHV ommm pmxwz I was HHH smumsm Nma Hmm moa muma m.m m.ma o.mm o.mH mxmwz seamen cuaouo momma .mxz 0.N HmoEmoma\¢ noum2\H umaSm>OZ\H aua02\xmm3 maflumm>nmm mo muma mawpam .mxz 0.m umoEm>02\N >umcuamh\m Hmaou00\m auc02\xmm3 assumm>umm mo muma madcaaomm .mxz H.H Hmafimuamm\m umaou00\H cua02\xmm3 maapmmz mam mo mama ocwpcm .mx3 m.H HmoEmuamm\H Hmnou00\H unams¢\m au:02\xmm3 Amshmm vav mcwpmmz cam mo muma mGACCflmma sceflxmz EDBHGHZ cowumfl>ma mmucaflwuua Emma pumocmum cmmz mmamm muwca 0cm Emumhm cowuoapoum omacfiuaou .H.m mHQMB 171 .mxz o.m sasn\a sass\e smz\m cacoz\xomz Amsnmm 0H0 amazowamm ama mo mama wasn\m sase\e Haemaxa cacoz\xmmz mcazoam mo mama H.e m.m e.ea m.H mama .na oma whoa new mamas m.H m.m 0.m 0.~ mamm» Euma acmmmaa co omoamoaa moam mummy mo .02 0.m 0.0a 0.m mummy moam omoapoaa mmm amEamm mummy mo .02 .mm 0.Hv m.0ha 0.5 mmuod muam eamm Amsamm 0H0 cmmm Hmcoaaaomue I 09 >H smamsm ema m>ma m.H ~.0H mammz oceama aa3ouo Somme .mxz m.m aamcamh\m mamaaamm\m umaEm>OZ\v nac02\xmm3 mawamm>umm mo mama mawpcm .mxz N.N amaEm>o2\m umosmoma\m amofim>OZ\H cac02\xmm3 mcaamm>amm mo mama acacaammm amnsmaamm\m amoEmammm\m amaEmammm\N aacoS\xmm3 moammmz mam mo mama acaccm umnsmam0m\a umnsmammm\~ amsm=a\~ ca:02\xm03 Amsamm HMO mcaommz pom mo mama moaccwomm amasmaamm\m amoEm>OZ\H >H50\v aac02\xmm3 moaommz ama mo mama weapon amsmsa\m umnoauo\q magn\m cuqoz\x003 Amsumm Hal mcaomm3 amH mo mama mcwccwmma Enaaxmz ESEacaz coaama>ma mmacaauaaa Emma mammcmam cmmz mmcmm mawca mam amawam c0aaocmoaa pmacaaaoo .H.m mHamB 172 m.H m.m 0.0 0.m mmmmw Emma acmmmma co nmosmoma mom“ mmmmw mo .02 m.m 0.m 0.N mmmmw momm pmoapoma mmm mmEmma mmmm» mo .02 m.0¢m n.0ma n.n0v ~.m mmmo< mnam Emma Amsmma may pmmm pm>omaEH I 09 > Emamxm NmH mama o.mm _ mxmmz mommmm basomo momma .mx3 m.m >mmacm0\m mmmamama\m mmasmoma\H aaco£\xmm3 mamamm>mmm mo mama mumpca .mxz m.m mmaEm>OZ\m mmaEmoma\v mmaoaoO\v aaa02\xmm3 mamamm>mmm mo mama mcmcammma .mxz m.m mmaEmaamm\N mmaoa00\v amsmsm\m cac02\xmm3 mampmmz and mo mama mcamca .mx3 m.~ amamc¢\m mmoEmaamm\m >H50\~ aacoz\xmm3 Hmsmma 0v mamomm3 amm mo mama mamccmmma H.0H 0.0a 0.mm 0.0 mmoa mmm .mba ammomamz mo mama comamoamaaa we mm 0mm on mace mom .mnu mama mammmmm m.H >H90\m amsma¢\m mcab\¢ aaa02\xmm3 mama mamaamaa Ammao game He 0cm3ommmm omm mo mama .mxz m.m smas\m amsmsm\m mcse\m sac02\xmmz imsmmm m0 mcm3ommmm mam mo mama EDmemz Easmcmz c0mamm>ma mmacammaaa Emma mmmmcmam cmmz mmcmm maaca mam Emamwm acmaocmoma omDCmacOO .H .m magma. 173 0mm mama .mxz m.m m.mm memes mommmm sazomo semen .mxz m.m mmmacmh\m wmmsmama\m mmoEm>02\N cacoz\xmm3 mcmamm>mmm mo mama anarcm .mxz m.m mmaEm>OZ\m >mmacm0\m mmaoa00\m cacoz\xmm3 mamamm>mmm mo mama acaccmmmm mmaEmaamm\m mmaoa00\H mmaEmaamm\H cacoz\xmm3 memommz pom mo mama ocmocm amama¢\m mmoamaamw\m amamc<\m aacoz\xmm3 Amsmma my mammmmz mam mo mama acaccammm amamaa\v mmaoaoO\v amaoca\a aac02\xmm3 mawomm3 amH mo mama mamccm amsasa\m umnsmummm\m sm5n\m s0202\xmm3 imammm mmc memommz and mo mama acaccmmmm m.mm m.mm m.0m 0.0 mmoa mma .mna ammomamz mo mama c0mamommaaa em mm mmm um mace mmm .mnn mama oammmmm .mxz N.m >H50\H am505£\a maab\m aaaoz\xmm3 mama.madaamaa aa:02\xmm3 Ammao Emma Ha mam30mmmm omm mo mama m:90\v amaaoa\m mc50\m aacoz\xmm3 msmma may 0cm30mmma mam mo mama 0::e\e amsmsaxm mmz\m sa:02\xmmz Amsmmm mmc mamzommmm ama mo mama .mmz m.m 0::n\m wasn\e smzxm raco2\x0m3 mcmzomm mo mama m.m m.s e.em m.~ mmmm .nq omm mane mmm mama» easmxmz ESEmcwz c0mamm>ma I. mmasammaam_smma cmmmcmam cmmz momma mamas mam Emamhm comaocmoma omaamaaou .H.m mHQmB 174 mmasmaamm\v mmnEm>OZ\H amcmaa\v cacoz\xmm3 mammmmz 0mm mo mama asapca mmnsmaamm\m amno000\m amsmmm\m macozxxmmz mammmms mum mo mama mmmmmmmmm .mxz m.m mmoEmaamm\H mmaoa00\m amamad\m ca202\xmm3 mammmm3 mam mo mama 0amccm .mxz s.m amsmsaxm mmnoaUO\m sman\e sac02\xmmz mammmmz mam mo mama acmccmmmm .mxz m.0 >Hsb\v mmaEmaamw\m >H50\H aac02\xmm3 mammmm3 and mo mama mammca .mxz m.e mmsn\m amsmsa\m mcssxm maa02\xmmz imsmmm mac 0cm0m03 ama mo mama acaccmmmm o.mm o.mm o.mm o.o mace mmm .mnm cmmomamz mo mama comamommmma mm N0 00m 50 mmod mma .maa mama mammmmm .mxz m.m >H50\H mach\m mach\m ca:02\xmm3 mama mamacmaa .mmz a.~ mmsnxm smsnxm massxm mamoz\mmmz amsmmm m0 mmm3ommmm meme .mxz m.m wasn\m masnxm mcse\m mamozxxmmz amazomm «0 mama s.~ m.s m.mm o.e mmmm .nq 0mm mmom mmm mmmmm m.v 0.5 0.0H 0.m mmmm» Emma aammmma aO mmocmoma momm mmmmw «0 .oz v.0 mm N mmmmw momm mmocmoma mmm mmemma mmmm» mo .02 H.H m.a 0.e m.0 mmmom mnam Emma Amamma may mmmm Hmcomammmma I om H> smammm saamxmz EUEmCHE commmm>mo mmasnmmaaa sums mmmccmam cmmz mmcmm mamas 0cm Emam>m comaosmoma cmacmacou .H.a maame 175 emm mama m . a m . mm o . mm o . mm 3003 6030a 530mm momma .mxz m.m mmafimoma\m ammccmb\v mmaoa00\m aaco2\xmm3 mamamm>mmm mo mama mammcm .mxz ¢.m mmaEm>02\H mmasmoma\m mmaoa00\m cac02\xmm3 mamamm>mmm Mo mama mammammmm Eaamxmz adsmcfiz comamm>ma mmaaammaaa Emma mmmmcmam cmmz mmcmm mamas mam Emamwm comaosmoma omscmacoo .H.m mHamB APPENDIX C CALCULATION OF LAND PREPARATION COSTS FOR TRACTOR OWNERS I'llallllllllllill am mmpcmamo Hmmcamaommmm maa cm>m0 mammmao cmo moaomma Hmmaz as mm m amaa mommmom cmamamama cm acmammmao mmmcmcome mo ommmcmam cam chmammcoo mammm maa mamaommmmm mmamfimamm m>mam>mmmcoo .mmmm 55am maa m .mmm>mmp 03a aamz mammcm 03a mo hmp xmoz moo: NH m mmESmma m .mmmm macaw maa H Aoovmv 50mm Enema mma mmcom moaomma Hmaoe .wmwl acmomma 0H am mama 0cmccam mcomcmammommz "mama 00mm eased mma mmsom moaomme 0cmxmoz Hmaoe .mmml 00 xmoz mmmmmma 0mm mm mm. MI 03 o .3 om . m .. .. .. amazommma mmm mmm mm 00m 0.0m 00.H. 30mmma memo VHINH 0amsommmm ama mmm mm 0mm 0 . a ma . 0 33a 88% amazoma mmaoa mmmw mhma moo: mma moaomme Hmcccd mma mama mmmoa \mmmom H mmmom acmEmHaEH acmammmao mmmczo moaomme mamao cmmaamoz mom mGOHaQEcmma mocmsmoamma acmfiamcoa mam moaommB .H.O mHamB 176 1f77 Table C.2. Estimated Financial Owning and Operating Costs of a Tractor and Associated Equipment in Northern Ghana During the 1973-74 Crop Season: The Case with Subsidies A. ESTIMATED OWNING AND OPERATING COSTS PER HOUR OF A 65 H.P. WHEEL TRACTOR2 l. Owning Costs Assumption: Anticipated life = 7000 hours (1400 hrs./yr. x 5 yrs.) a. Purchase Price of a Tractor Purchased in Tamale, Northern Region 1. Purchase price 3 ¢ 7250 2. Plus, interest at 6\ 1088 3. Price plus interest 8338 4. Plus, cost of repairs at 100‘ of initial cost 7250 5. Plus, interest on repairs at 6% (¢7250 x 0.06) 4 435 6. Cost of tractor, plus repairs over life of the tractor ¢16023 b. Owning Cost per Hour ¢16,023 = ¢2 29 7000 hours ' c. Annual Ownin Cost 16,023 g £L_______‘ ¢3,205 5 years 2. Operating Costs Assumptions: Tractor works 12 hrs./day; Tractor operates 1400 hrs. per annum: Average fuel consumption = 1.5 ga1./hr. a. Fuel:5 Average consumption rate = 1.5 gallons/hr. 1. Pump price = ¢O.SO per gallon 2. Annual Fuel requirements (1.5 gal. x ¢O.50 x 1400 hrs.) = $1050.00 b. Engine oil: 14 oil changes at 1 1/2 gallons per change 1. Price per gallon - ¢2.50 2. Annual requirement (1.5 gal. x 14 x €2.50) = ¢52.50 c. Grease: 24 lbs. for tractor and equipment 1. Price = €0.40 per lb. 2. Annual requirement (24 lbs. x ¢0.40) = ¢9.60 d. Wages for two tractor drivers 1. At ¢45.00 per month (¢45 x 2 x 12 mos.) = le80.00 e. Total Operating Costs Per Annum - ¢2192.10 f. Annual Operating Costs Per Hour: ¢2192 1400—HSEFE €1.57 per hour 3. Summary of Owning and Operating Costs: a. Owning Costs Per Hour: ¢2.29 b. Operating Costs Per Hour: ¢1.S7 Total £3.86 8. ESTIMATED OWNING COST OF A PLOW AND HARROW6 ASSUMPTIONS: ANTICIPATED LIFE = 4000 HRS. PURCHASE PRICE - ¢1225 1. Owning Costs: a. Purchase Price in Tamale ¢lZ25 b. Plus, Interest at 6‘ 184 c. Total Cost Without Repairs 1409 d. Plus, Repairs at 100‘ of Initial Costs 1225 e. Interest on Reparis at 6! (¢184 x .06) 74 f. Total Costs 2708 g. Owning Cost Per Hour £2708 a “—" 4000 hours ¢°'68 per h°“’ h. Annual Owning Costs - Plow ¢2708 -——————-= 2226 12 years i. Annual Owning Cost - Harrow £2708 = ¢271 10 years C. ESTIMATED OWNING COSTS PER HOUR FOR A TRAILER: ASSUMPTION: ANTICIPATED LIFE 3 4000 HOURS 1. Owning Costs a. Purchase Price in Tamale ¢2000 b. Plus Interest at 6\ 300 c. Total Cost Excluding Repairs 2300 d. Plus Repairs at 50‘ of Initial Costs 1000 e. Interest on Repairs at 6‘ (leOO x .06) 60 f. Total Cost ¢3360 g. Owning Cost Per Hour E3360 = ¢0.84 ‘—"_— 4000 hours 178 Table C.2. Cbntinued D. OHNING AND OPERATING COSTS OP UNDERTAKING FIELD ACTIVITIES 1. Tractor Owning Coats Associated with Field Activities Activity Total Owning Adjustment Adjusted Total Hours costs Factor8 Owning Owning Per Hour Costs Costs Per Hour Per Annum Plowing 336 ¢2.29 1.60 23.67 (1233 lst Narrowing 252 2.29 1.20 2.75 693 2nd Narrowing 168 2.29 1.00 2.29 385 Sub-Total 756 2311 Other 631 2.29 0.62 1.42 894 Total 1400 3205 2. Owning and Operating Costs of Undertaking Field Activities Activity Tractor Owning Costs Total Cost Acres Costs 0 of Implements Per Hour Per Hour Per Acre Owning Operating Costs Costs Plowing £3.67 £1.57 £0.84 €6.08 0.75 €8.10 lst Narrowing 2.75 1.57 0.84 5.16 1.00 5.16 2nd Narrowing 2.29 1.57 0.84 4.70 1.50 3.13 E. OWNING AND OPERATING COSTS ASSOCIATED WITH TRAILER WORK AND MISCELLANEOUS RUNNING 1. Trailer Work a. Tractor 1. Owning Cost: £1.4211 2. Operating Costs l.5_7 3. Total COStS €2.99 b. Trailer 1. Owning Costs 0.84 C. Total owning and operating costs £3.83 2. Miscellaneous Running12 a. Tractor 1. Owning Costs (1.42 2. Operating Costs _}.57 3. Total Costs 2.99 1The cedi equivalent of the c.i.f. price of capital imports Is artificially low since at the official exchange rate the GhanAIan Cedi is 35 percent OVervalued. The official exchange rate is GNC1.15 T US$1.00. ZBasnd on two manufacturer models, Ford 4000 and Massey Ferguson Model 165 diesel tractors. the technical coafficients and the spare parts and maintenance requirements are based upon a report prepared by G. M. Wylie (1972), an F.A.O. Agricultural Engineer who has worked in Ghana several years and was statIonnd in Northern Ghana during 1972-74. 3 . . . . . . The estimated private opportunity cost of a capital for owners of capItal in Ghana. This figure 15 6 percent of the undepreciated value of the asset over its five year life. 4A zero salvage value is assuned after 7000 hours. I )During the 1973-74 production season it was estimated that the National RedemptIon Council was subsidizing petroleum products by aLout 20 percent. It is assumed that 70 percent of the retail price 0! petroleum products is import content. 6Based upon a report by Wylie (1972), the anticipated life of both the plow and harrow is 4000 hours. The average retail price of a J-disc mounted plow and a 12-disc mounted harrow are equal. Thus, the above calculations are for either item 0f equipment. No operating costs are assumed to be associated with the plow and harrow. Repairs are treated as an owning cost. 7 - ' Based on Appendix C, Table C.l. "Tractor and Equipment Performance Assumptions for Northern Ghana Tractor Owners." 8Adjustment factor is a relative estimate of the rate of wear and tear or user cost associated with the different tractor operations. An adjustment of 1.50 for plowing means there is an estimated 50 [wrcent greater wear and tear on the tractor associated with plowing compared to 2nd harrowing. The adjustmrnt {actor for nonfield (Other) is treated as a reSIdual; it was determined by subtracting the total annual (»mIng costs for field work (€2311) from the total annual owning costs ((3205) and then calculating the adjusted owning costs per hour required for 631 hours to equal €894. 9Based upon report prepared by Wylie (1972). 1oTotal cost per hour divided by acres per hour. llSee adjusted owning costs per hour in Part D, above. nghe use of a tractor as a standard four—wheel vehicle, generally in nondirectly productive use. ~ was- 1 . s\.wu 179 Table C.3. Estimated Economic Owning and Operating Costs of a Tractor and Associated Equipment in Northern Ghana During the 1973-74 Crop Season: The Case Without Subsidies1 A. ESTIMATED OWNING AND OPERATING COSTS PER HOUR OF A 65 H.P. WHEEL TRACTOR2 l. Owning Costs Assunption: Anticipated Life - 7000 hours (1400 hrs./yr. x 5 yrs.) a. Real cost Price of a Tractor Purchased in Tamale, Northern Region 1. Purchase price 3 ¢7250 2. Plus, 35 percent of c.i.f. price (¢507S.00 x 0.35) 1776 3. Purchase Price without overvalued currency ¢9026 4. Plus, interest at 15\4 3385 5. Real cost price excluding repairs '12411 6. Plus, cost of repairs at 100‘ of initial costs 9026 7. Plus, interest on repairs at 15% (¢9026 x 0.15) - . 1354 8. Real cost price of a tractor plus repairs over the life of a tractor5 ¢22791 b. Owning Cost Per hour %%%%g%_ : ¢3.25 r. c. Annual Owning Costs £32412; ' £4558 5 years 2. Operating Costs Assumptions: Tractor works 12 hrs/day; Tractor operates 1400 hrs. per annum; Average fuel consumption = 1.5 gal./hr. a. Fuel:6 Average Consumption Rate - 1.5 gallons/hr. 1. Pump price = €0.50 per gallon 2. Pimp price, plus subsidy e (20.50 x 1.20 a ¢O.6O 3. 35‘ of the estimated c.i.f. price (¢0.3S x 0.35 8 ¢0.12) 4. Real cost of fuel without subsidy and overvalued exchange rate (¢0.60 + ¢0.12 = ¢0.72/gallon) 5. Annual fuel requirements (1.5 gal. x ¢0.72 x 1400 hrs.) - ¢1512.00 b. Engine Oil: 14 Oil Changes at 1 1/2 Gallons Per Change 1. Price = ¢2.50 per gallon 2. Price, plus subsidy (¢2.50 x 1.20 = £3.00) 3. 35% of estimated c.i.f. price (¢l.75 x 0.35 - ¢0.61) 4. Real cost without subsidy and overvalued exchange rate (€3.00 + ¢0.61 s €3.6l/gallon) 5. Annual oil requirements (1.5 gal. x 14 x ¢3.61) - 75.81 c. Grease: 24 Lbs. for Tractor and Equipment 1. Price = ¢0.40 per 1b. 2. 35\ of estimated c.i.f. price (¢0.28 x 0.35 - £0.10) 3. Real cost without overvalued exchange rate (¢0.40 + £0.10 = ¢0.SO) 4. Annual requirements (24 lbs. x ¢0.50) t ¢12.00 d. Wages for Two Tractor Drivers 1. At ¢45 per month (¢45 x 2 x 12 mos.) = ¢1080.00 e. Total Operating Costs Per Annum = ¢2679.81 f. Annual Operating Costs Per Hour ZEKKEL.__._ = ¢1.91 per hour 1400 hours 3. Summary of Owning and Operating Costs a. Owning Costs Per Hour ¢3.25 b. Operating Costs Per Hour 1.91 Total ¢5.16 B. ESTIMATED OWNING COSTS OF A PLOW AND HARROW7 ASSUMPTIONS: ANTICIPATED LIFE 8 4000 HOURS PURCHASE PRICE I ¢1225 1. Owning Costs a. Real Cost Price of Plow Purchased in Tamale, Northern Region 1. Purchase Price in Tamale 2. Plus, 35% of c.i.f. (¢858 x 0.35) 3. Purchase price without overvalued currency 4. Plus, interest at-lSt8 5. Plus repairs at 100\ of initial costs 6. Plus interest on repairs at 15 percent (¢1525 x 0.15) = ¢1225 300 1525 572 1525 229 7. Total costs ¢3851 b. Owning Costs Per Hour ¢3851 -——-—-——-= .96 4000 hrs. ¢0 I! fiii- “III£* 180 Tabla c. 3. manna-fl C. mI‘fID m CUB VII m C, I mxul “MINI MICIPITBO LI?! ' 6000 D008 l. mm mot. a. bal Out Drica o! a Trailer Menu“ in Tan-la. Iorthom Raglan l. Purcnaaa price (2000 2. Fina m or c.i.L triaoo I 0.35) 190 1. Purcnaao price without ovarvaluod currancy 2490 G. ’lm lntaroat at l5\ 9" 5. Ital coat price without repairs W 6. Ph.- repaira at 50! of initial cost: l245 7. Plus intarnt on ropairl at 15! (“MS I 0.15) la? I. ”(.1 coat 4856 b. Ming Coat Par ibur “'56 «mo hrs. ' “"1 D. ("RUIZ AID OVEIATIW C1575 0F UNIIR'I'AKIM 7121.0 ACTIVITIES l. Tractor thrninq Costa Msociated with Field Activities ICthlty Total Owning Adjustment Ad)ustod Total Hours10 Cost Fartor” Owning Ming Par Hour Coats Cents Per Hour For Anni- rlm-inq ”F. (L25 1.60 €5.20 (17‘? lat Marta-ing 252 3.25 1.20 3.90 98] 2nd mirror-ling 166 LZS 1.00 ).25 546 Sub-Ntnl 756 1276 Other 631 3.25 0.62 2.0) 1201 Total 1400 4558 2. (inning and (pt-rating Cont: of Unértakinq Field ActiVitiva ActiVity Tractor Ming Cost! Total Cost Acre: Par Coats Per u! Iaplmntl For Hour Mu!” ALTO Owning operating Costs Costs Flown-K; £5.20 (1.91 £0.96 (0.07 0.15 (10.76 lat Narrowing 1.90 1.91 0.96 6.77 1.00 6.7? 2nd Harrowinq ).25 1.91 0.96 6.l2 1.50 4.08 I. MIN”. AND OPERATING COSTS USCI‘IATED HIT" TRAILER ‘0” Am HISCILU‘EO'JS RWNIfll l. Trai ler tion a . Tractor 1. Gaming Costa [2.031‘ 2. Oprratinq Costa L3; C).‘M b. Trailnr . 1. Mine; Coats Li} C. Ntal Chninq and Operating Costs 5.15 2. Mincellanooua Dunning” a. Tractor l. uminq Costs 12.0) 2. (berating lfl 1.9! I IT?”- oHicul ioreiqn exchange rate in Giana in GflflJS - US$1.00. It is oath-ated by the LNJ‘. that the local currency is overvalued by 35 porcvnt; the amtwptod shadow rntr of orrhanqc is 113nm)!» - 5H¢LH - US$1.00. Tho import component of Lht' purchanm price of a tractor and anociatad r-quirnnt is; immunised by 35 [‘QI'Ci‘n' to reduce to rum the lwlimt swisidy of thd: overvalued crvrhmqv ratv. Th0 «xvii-tit subsidy on petroleu products has alvo bum rodmxd to arm in the calculation of Uu: ourfdllflq C'fltitx. 135306 on two mnufncturar mdcln, ford 4000 and Hassoy Ferguson 165 diesel tractors. Tm- tnchnirxl m-Hiciuntn and thr- swish party. and unintunancu roquirmonrs aro- bmmd upon a rem-rt prop-arm! by 0.7!. flyliu (W72). an F.A.O. aqricultural urnqinewr who has worked in rRun: snvoral yoar-i and was a‘tationed in Northern Ghana during 1972-74. 1 . . harm! upon diacussions with ropreoentativws ol the tvo rumor tam Mchint-ry dealers in Ghana. it in “guard that the retail price 0! tractors. anocxatad equipnant and spare parts is Connemara o! a 70 purrcr-nt ”Ivor! content. ‘T'hv social opportunity cost of capital in Ghana is null-ultlrd to be 15 percent by the Niniutn or Econmic l‘inancle and Planninq. 1hr.- figure i9 l5 percent 0! this undeprocutrrd value 0! NW asset mmr itIi five year life. 5 A znro salvage value is auu-ed aftor 7000 hours. tinted that the National floor-ration (‘mmril mil 6 , 97 )-'N roduct ion noanon it. was on . WH’W U” l P It in rumpled that 70 percent of "w r-‘Hul P'lf‘“ sulmidizino petrol-um prod-it"s by about 20 percent. of pr-trolem products in ”mart cont-mt. 7 W ‘ - r ~ lil'n of both the plot: and harrau is “00 ~96 . n a re rt. 13' II': ”97.). thq anti.iratrd 3‘3 “5'0 P0 I W m-mtuvj plan: and a lJ-disc muntod narrow are vqual. Thus, hours. Thur average retail prices of .I J-Jisc . the abow calculation: are for either ital of aquiysvnr. Mn operatinq casts are “mat-d to be ansnciatnd with the plan and harrw. Repairs aru treated an an ounlnq cost. in Ghana Iii mutilated to b1.- 15 percent by this Hiniutry of its 8 The social opportunity cost of capital Scone-1c Finance and Planning. no (iqure is 1% pcrmnt or the undaprociated value a! the “not over fun you: lift. I"The liquro in l5 percent of the undnpraciatrd value 0! the nsset over its (no year lifts. 10Based on App-audi- C. Tablo C.l. “tractor and [guitar-at Perforunco Assn—Pllons for northern Ghana Tractor Mers.‘ ”Adjuat-nt (actor is a relative estimato- of the rate or wear and tear of usi-r coat aswi':ot:rd with On ditto-rent tractor operations. In adjust-mt of 1.50 for planing means then: i'i an F‘i!|l\l.1f:hc 50 percent qreatar war and tear on the tractor a'mociated with plowing Lowanrd tu 2m! nnrro;i1-ZLUM; adjust-tit factor for nonfield work (Other) is treated as a residual; it v.2"- n!t-t.-rn.incd by 2-8- nu "w" the total annual owning contn for f‘upid work ((1:76) (ran the total annual uvninq coats (2.35 I an . . calculating the adjusted owning coats per hour required [or all houru to equal rum. ”Ia-ad was report praparad by fill. (“72). ”m1 coat par hair avid-d by acna par hour. “500 Min-tad min coats par hour in Part D. above. “The no of a tractor aa a atandard four-chad which. qanoral ly in nondiractly product in: use. APPENDIX D CALCULATION OF LAND PREPARATION COSTS FOR BULLOCK OWNERS Table D.l. Estimated Financial Cost Per Acre for Bullock Plowing in Northern Ghana, Based Upon Survey Data. 1973 I. Owning Cost of a Bullock Team1 A. Purchase Price of two,three year old West African shozrhorns €270. B. Plus. opportunity cost at 10.5 percent2 156. 426. C. Less salvage value after four years at 170 percent of initial purchase price3 473. D. Total owning cost -€ 47. E. Annual gain from owning team: -€47 % 4 years. - -€ 11.75 F. Allocation of gain to rice enterprise:4 ~€ll.75 x 0.14 I —€ 1.65 11. Equipment A. Bullock Plow 1. Purchase Price € 70.00 2. Plus, opportunity cost at 10.5 percent5 ' 74.00 3. Plus, maintenance at 50 percent of initial cost 35.00 4. Total Owning cost €179.00 5. Annual Owning cost: €179 -I 20 yrs. I m 6. Allocation to rice enterprise: €8.95 x 0.516 = € 4.56 B. Yolk, Harness. Chains, Nose Ring 1. Purchase Price € 22.00 2. Plus, opportunity cost at 15 percent7 4.95 3. Total purchase cost € 26.95 4. Annual owning cost: €26.95 % 3 yrs. = . 5. Allocation to rice enterprise: €8.98 x 0.14 - C 1.26 III. Total Cost of Bullock Plowing A. Annual cost of plowing 1.1 acres of rice land 1. Bullock Team -€ 1.65 2. Plow 4.56 3. Yolk. etc. 8 1.26 4. Supplemental Peed 1.70 E 5.87 B. Cost Per Acre: €5.87 % 1.1 acres I € 5.34 1Based upon the 14 bullock farmers for whom the average rice enterprise budget in Chapter Iv is computed. The mean total farm acreage of the 14 farmers was 7.7 acres, of which 1.1 acres was rice and 6.6 acres were other crops. 2Assumes on the average, that half the undepreciated value of the team is held as equity with an opportunity cost of 14 percent and half is in the form of bank credit at an interest rate of six percent. The undepreciated value of the team increases with time. and its not salvage value is 175 per- cent of its net acquisition value at the end of its four year useful life. It is assumed that at the end of four years the team is sold for meat. 31h is common experience in Ghana and other West African countries that bullocks appreciate over 3 to 5 years of use by 170 percent of the initial purchase price. 4 . Since 14 percent of the acres (1.1 of 7.7 acres) are rice, 14 percent of the gain is allocated to the rice enterprise. SAssumes that half the undepreciated value of the plow is held as equity with an opportunity cost of 15 percent and half is in the form of bank credit at an interest rate of six percent. Assumed salvage valuevis zero. 6Farmers purchase a bullock frame which has a plow attachment and a ridger attachment. The frame is €40.00 and the plow attachment is €30.00. Fourteen percent of the cost of the frame (1.1 of 7.7 acres) is assigned to the rice enterprise, plus the total cost of the plow attachment (€40 x 0.14 = €5.60; €5.60 = €30 = €35.60; €35.60 % €70.00 - 0.51). Fifty-one percent ’ of total owning cost is assigned to the rice enterprise. 7 . Assumes that these items are purchased with cash reserves which have an opportunity cost of 15 percent. Salvage value after threes is zero. 8Upland soils are light, sandy soils. Farmers. on the average, plow 1 acre per 6 hour day. Supplemental feeding is practiced only during the days in which bullocks are plowing (rice) or ridging (other crops). On the average. farmers feed foodstuffs valued at €1.55 per working day (sorghum. maize, plus. in some cases. purchased sorghum mash from local beer making). Supplemental feeding for 1.1 acres costs €1.70 (€1.55 x 1.1 days). 181 182 Table D.2. Estimated Economic Costs Per Acre for Bullock Plowing in Northern Ghana I. Economic Owning Cost of a Bullock Team1 A. Purchase price of two, three year old West African shorthorns B. Plus, opportunity cost at 15 percent C. Less salvage value after four years at 170 percent of initial purchase price D. Total Owning Cost E. Annual owning cost: €20 % 4 yrs. = F. Allocation of gain to rice enterprise: €5.00 x 0.14 = II. Equipment A. Bullock Plow 1. Purchase Price 2. Plus, 35 percent of c.i.f. price: €56 x 0.35 = 3. Purchase price without overvalued currency 4. Plus, opportunity cost at 15 percent 5. Plus, maintenance at 50 percent of initial cost 6. Total economic owning cost 7. Annual owning cost: €259 % 20 yrs. = 8. Allocation to rice enterprise: €12.95 x 0.51 = B. Yolk, Harness, Chains, Nose Ring 1. Annual owning cost 2. Allocation to rice enterprise III. Total Economic Cost of Bullock Plowing A. Annual Cost of Plowing 1.1 Acres of Rice Land 1. Bullock Team 2. Plow 3. Yolk, etc. 4. Supplemental Feed B. Economic Cost Per Acre: €10.26.% 1.1 = ¢270 o 223. 473. € 20. € 5.00 ¢ 0.70 € 70.00 19.60 89.60 M 224.00 35.00 €259.00 ¢ 12.95 ¢ 6.60 Fifi ...: N 0‘ ¢ 0.70 6.60 1.26 1.70 €10.26 (Z 9.33 _‘I' 1For the procedures used to calculate the owning cost of a bullock team, see companion table, Appendix D, Table D.l. APPENDIX E ESTIMATED ECONOMICS OWNING AND OPERATING COSTS OF TWENTY-FOUR SELF-PROPELLED COMBINES TflfleEJ. Estimated Economic Owning and Operating Costs of Twenty-Pour Self-Propelled Combines Operated by the Ministry of Agriculture in Northern Ghana, 1973-74: When Import Prices are Converted at the Shadow Rate of Exchange 1 INVESTMENT COST OF TWENTY-FOUR COMBINES A. Machine Number Unit Total Import Costs Manufacturer/ of c.i.f. Model Units Price Official Shadow Exchange Exchange Rate Rate Gloria-C12 13 €17,215 €223,795 €302,123 MF-400 6 21,800 130,800 176,580 MF-520 5 27,400 137,000 182,250 24 €491,595 €660,953 B. TOTAL OWNING COSTS I Machine Number Purchase Plus Plus, Plus, Real Cost of Price Interest Repairs at Interest of Combines Units at 15%2 40‘ of on Repairs Plus Initial Cost at 15‘ Repairs Gloria-C12 13 €302,123 €181,274 €120,849 €18,127 € 622,373 MF—4OO 6 176,580 105,948 70,632 10,595 363,755 MF-S2O 5 182,250 109,350 72,900 10,935 375,435 24 €660,953 €396,572 €264,381 €39,657 €1,361,S63 C. ANNUAL OwNING COSTS ASSUMPTIONS: Anticipated machine life is 8 years No Salvage Value Machine Number Total Annual of Owning Owning Units Costs Costs Gloria-C12 13 € 622,373 € 77,797 MF—400 6 363,755 45,469 MP-SZO 5 375,435 46,929 Total 24 €1,361,S63 €170,195 D. OPERATING COSTS FOR THE 1973-74 HARVEST SEASON Expenditure Item Machines Thirteen Six Five Total Gloria MF-400 MF-520 Expenditure 1. Fuel3 € 1,532 €5,322 €1,044 € 3,898 2. Lubricants 698 94 145 937 3. Operators' Salaries 10,633 3,825 3.431 17.889 4. Operator Travel and TranSport 744 273 295 1,342 Total €13,637 €5,514 €4,915 €24,066 E. TOTAL ANNUAL OWNING AND OPERATING COSTS Machine Owning Operating Total Costs Costs Annual Costs Type No. Gloria 13 € 77,797 €13,637 € 91,434 MP-400 6 45,469 5,514 50,983 MF-SZO 5 46,929 4,915 51.844 Total 24 €170,195 €24,066 €194,261 F. NET OPERATING POSITION Machine Total Gross Net Costs Revenue Return Type No. Gloria 13 € 91,434 €12,937 € -78,497 MF-4OO 6 50,983 12,721 -38,262 w-Szo _5 M M 1:21.020. Total 24 €194,261 €36,510 ¢-157,751 183 184 Table 3.1. Continued G. AVERAGE HARVESTING COSTS PER BAG DURING 1973-74 CROP SEASOI Machine Costs Per Bag Total Bags Fixed Variable Total Gloria 14,439 I 5.39 €0.94 €6.33 HP-400 15.461 2.94 0.36 3.30 ar-szo 11,415 4.11 0.43 4.54 ”sighted Average Cost Per Bag 41.315 €12.44 €1.73 €4.70 H. ADJUSTED COSTS4 Machine Adjusted Adjusted Adjusted Cost Bags5 Total Costs Per Bag Gloria €18,049 € 94,843 MF-400 19,326 52,362 HF-SZO 14,269 53,073 €51,644 €200,278 £3.88 1. AVERAGE LOSS PER BAG HARVESTED 1. Average Total Cost Per Bag €3.88 2. Average Charge 0_.?_()_ 3. Average less per bag €2.98 4. Average rate of subsidy 77 Percent J. BREAK-EVEN CHARGES FOR COHBINE HARVESTING 1. Bags harvested7 a. Complete combine harvesting 41,315 b. Combine as stationary thresher 10.329 c. Total bags 51,644 2. Estimated Break-Even Charges8 Method of Bags Charge Total Harvesting Harvested Per Bag Revenue Complete Combine 41,315 €4.20 €173,523 Harvesting Combine Used As A Stationary Thresher 10,329 2.55 26,399 9199,862 3. Current Rates of Subsidy Method of Current Actual Rate of Harvesting Charge Coat Subsidy (Percent) Complete Combine Harvesting €1.00 €4.20 76 Combine Used As A Stationary Thresher 0.60 2.55 76 1Shadow Rate of Exchange is GH€1.55 - US$1.00. This table is based upon flinch, 1974. 2Figures are calculated by taking 15 percent of the undepreciated value of the asset over the assumed life of the asset (8 years). 3The cost of fuel and lubricants is increased by the extent of the implicit and explicit subsidies. Given a 20 percent explicit subsidy on petroleum products and a 35 implicit subsidy, due to the overvalued exchange rate, on 70 percent of the retail price (import content), the total subsidies on these products is 44 percent. ‘The above costs require an adjustment to be made for the following reasons. The above costs are based upon the first year of operation of the combines. and many of the combines were not in the region at the beginning of the harvesting season. Owing to first-year organizational problems, inexperienced operators, and excessive down-time because of inadequate maintenance and service support, the combines operated at only 25 percent of rated capacity. It is assumed that some of these problems will be partially corrected. and. as a result, that annual utilization of the machines will be increased. It is assumed that the average increase in annual utilization in terms of bags harvested over the remaining life of the machines will be 25 percent. Part of the increase will come from antiCLpated increasing average farm yields, and part will come from an increase in the number of'days the machines are operated as a result of iflproved organization (deployment) and experience. The following are the adjusted figures for annual bags harvested and total costs. SActual bags harvested are increased by 25 percent. 6The operating costs in Part E above are increased by 25 percent. Thus, €170,195 0 €30,083 - €200,273; where €30,083 or total operating costs - €24,066 x 1.25. 7Assumes that 80 percent of the total bags harvested annually are done by complete combine harvesting and 20 percent by combines used as stationary threshers. aAssumption: The relative charge per bag for hiring a combine as a stationary thresher is 60 percent of the charge per bag for complete combine harvesting. The sum of the total revenue from each harvesting method should equal total owning and uoperating costs of the 24 combines as estimated above. APPENDIX F ESTIMATED LAND CLEARING COSTS PER ACRE ON BOTTOMLAND RICE FARMS Table F.1. Farms in Northern Ghana, 1973 Estimated Land Clearing Costs Per Acre on.Bottomland Rice 1 I. Machine and Labor Method A. Financial Costs Per Acre 1. Machine Costs a. Average contract charge2 2. Labor Costs at 13.4 Man-hours Per Acre a. Family labor3 5.0 man-hours G €0.32 - €1.60 b. Hired labor 8.4 man-hours @ €0.32 - €2.69 c. Total labor costs 3. Total Costs 4. Plus, Opportunity Cost at 15$ of the Undepreciated Balance Over Five Years4 5. Total Costs Including Opportunity Costs 6. Average Annual Cost Per Acre €27.83 % 5 Yrs. 8 Economic Costs Per Acre 1. Machine Costs a. Financial contract charge = €15.95 b. Adjustments €15.95 x 0.80 x 0.35 = 4.47 c. Estimated economic contract charge I 2. Labor Costs 3. Total Costs 4. Plus, interest at 15 Percent 5. Total Costs Plus Interest 6. Average Annual Cost Per Acre €33.95 % 5 Yrs. 8 6 II. Hand Labor Method D. E. P. G. Costs Per Acre at 30.8 Man-hours Per Acre Family Labor 9.5 Wan-Hours @ €0.12 - €1.14 Hir Labor 21.. flan-Hours 9 0.12 a 2.56 Total Labor Costs Plus, Interest at 15 Percent Total Costs, Plus Interest Average Annual Cost Per Acre €5.09 % 5 Yrs. - and labor methods of land clearing during 1973. ¢15.95 €20.40 4.29 24.69 9.26 €33.95 € 6.80 ¢ 3.70 _i-_2_9. ¢ 5.09 LL92 1Based upon survey data from 9 farms using a combination of machine 2Average contract charge per acre after adjusting for an assumed 30 percent overhdeclaration of acreage. The machines used were medium- size bulldozers with conventional blades of various manufacturers and models owned by the Ministry of Agriculture. 3The opportunity cost of family labor is assumed to be equal to the wage rate paid to casual workers for land clearing activities. at the end of year five. 5 content . 4Assuming straight line depreciation with a salvage value of zero It is estimated that 80 percent of the contract charge is import The import content of the financial charge is increased by 35 percent to reduce to zero the implicit subsidy of the over-valued exchange rate . of land clearing. 6 costs are measured using opportunity costs. 185 Based upon 1973 survey data from 37 farms using only hand methods Financial and economic costs are the same since all APPENDIX G ESTIMATED COST OF ONE BAG OF IMPROVED RICE SEED SOLD BY SEED MULTIPLICATION UNIT, MINISTRY OF AGRICULTURE, NORTHERN REGION, GHANA AND THE RATE SUBSIDY, 1973-74 186 Table 0.1. Estimated Cost of One Bag of Improved Rice Seed Sold by Seed Multiplication Unit, Ministry of Agriculture, Northern Region, Ghana and the Rate Subsidy, 1973-74 A. Cost of One 160 lb. Bag of Improved Seed1 1. Purchase Price (180 1b. Bag)2 €17.40 2. Cost of Mechanical Cleaning and Treating 2.20 3. Cost of Bagging and Handling 0.55 4. Overhead Charges 0.45 5. Total Costs for One Bag3 €20.60 B. Rate of Subsidy 1. Total Cost of One Bag €20.60 2. Less Selling Price 12.00 3. Subsidy ¢ 8.60 4. Rate of Subsidy: 42.0 Percent lCost data were Obtained from records of the Seed Multiplication Unit of the Ministry of Agriculture, Northern Region. 2The Seed Multiplication Unit purchases seed from Registered Seed Growers, whom it supervises. 3During the process of cleaning and treating and natural drying, the volume and weight of 180 lbs. of seed is reduced to 160 lbs. Seed is stored and sold in 160 lb. bags. APPENDIX H CALCULATION OF FERTILIZER COST PER 'ION AND RATES OF GOVERNMENT SUBSIDY AT OFFICIAL AND SHADOW FOREIGN EXCHANGE RATES 187 Table H.1. Calculation of Fertilizer Cost Per Ton and Rate of Government Subsidy at Official and Shadow Foreign Exchange Rates, 1973-74 Per Ton: 15-15-15 Ammonium Sulfate Official Shadow Official Shadow Exchange Exchange Exchange Exchange Rate Rate2 Rate Rate ----------------- €/Ton- = -- 1. c.i.f., Port of Tema 195.00 263.25 110.00 148.50 2. Port and Clearance Charges 2.90 2.90 2.90 2.90 3. Bank Charges 0.30 0.30 0.30 0.30 4. Transport to Tamale 36.00 46.08 36.00 46.08 5. Total Cost 234.20 312.53 149.20 197.78 6. Sales Price 56.00 56.00 40.00 40.00 7. Government subsidy 178.20 256.53 109.20 157.78 8. Percent subsidy 76 82 73 80 Per 1 cwt. bag: Total Cost 11.71 15.63 7.46 9.89 Sales Price 2.80 2.80 2.00 2.00 Subsidy 8.91 12.83 5.46 7.89 1 GH€1.15 = US$1.00 2GH¢1.55 = US$1.00 3It is assumed that the foreign exchange component of the transport is 80%. 28.80 x .35 = 10.08; 36.00 + 10.08 = 46.08). Thus, transport costs are increased by (€36.00 x 0.80 = 28.80; APPENDIX I ESTIMATION OF THE IMPORT PARITY PRICE FOR RICE MILLED BY THE RICE MILLS UNIT AT TAMALE, NORTHERN GHANA APPENDIX I ESTIMATION OF THE IMPORT PARITY PRICE FOR RICE MILLED BY THE RICE MILLS UNIT AT TAMALE, NORTHERN GHANA The average 1973 Ghanaian c.i.f. price fer milled rice was €329. or . $287 per metric ton.1 The 1973 price f.o.b. Bangkok for 25 percent broken milled rice was $200 per metric ton. The average expected price fir between 1973 and 1980 f.o.b. Bangkok in 1973 prices is $196.2 Given the relationship between the 1973 Ghana c.i.f. price and the f.o.b. Bangkok price and the projected f.o.b. price for 1980, we assume no change in the average Ghana c.i.f. price for imported rice in real terms. I. Estimation of Import Parity A. Milling Rates for the Tamale Rice Mills unit The average milling rate for the Rice Mills Unit is 55 percent. . . . 3 The output of milled rice in terms of grades 15 as follows: Grade Milling Rate Estimated Import Parity Value As a Percentage of 25 Percent Broken 25% Broken 30 Percent 100 40% Broken 15 Percent 75 100% Broken 10 Percent 50 Milling Loss 45 Percent O 100 Percent 1Personal communication from IBRD county representative. 2Rice price projection developed in 1973 by the IBRD and used in a project document. 3Prom Goodwin , 1975 . 188 189 The weighted average import parity value of one tone of domestic milled rice is therefore: [(30/50 x 1.00) + (15/55 x .75) + (10/55 x .50)] x $287. = $241. B. Adjustment for Processing and Internal Transport 1. Internal Transport Assumption: Fifty percent of the Tamale milled rice is transported and sold in Accra in competition with imported rice and must be reduced by transport costs to get import parity. The other 50 percent goes to distribution centers, on the average, equidistant between Tamale and Accra and has equal parity with imported rice. Therefore, the import parity price of Tamale rice marketed in Accra is equal to the import parity price less half the cost of transportation. The estimated economic transport cost in Ghana for a seven ton truck is $0.15 per ton mile.4 Thus, the transport cost from Tamale to Accra is $0.15 x 410 miles + $62.00 2. Domestic Milling Cost The estimated cost of milling one ton of rice at the Tamale Rice Mills is $44.00 per ton.5 3. Total Domestic Costs Total domestic costs are, therefore, 50 percent of the internal transport costs from Tamale to Accra, plus total milling costs, or $44.00 + $31.00 = $75.00. C. Net Foreign Exchange Value Per Ton of Domestically Milled Rice The net foreign exchange value or net economic value per ton of domestically milled rice is, therefore, the ex—mdll import parity price, 4 From Goodwin, 1975. 5Ibid. 190 less estimated processing and transport costs, or $241.00 - $75.00 = $166.00' D. Domestic Parity Value Per Ton of Rice in Local Currency 1. At the official Exchange Rate (US $1.00 = GH €1.15) $166.00 x 1.15 = €191. per ton 2. At the shadow Rate of Exchange (US $1.00 = CH €1.55) $166.00 x 1.55 = €257. per ton E. Domestic Economic Parity Value Per Ton of Paddy .At the milling rate of 55 percent, it requires 1.8 tons of paddy to obtain one ton of milled rice. Therefore, the domestic economic parity value of one ton of paddy is: €257. % 1.8 = €143. per ton or €11.50 per 180 1b. bag. In the economic analysis the import parity price or economic value of paddy is therefore assumed to be €12.00 per 180 1b. bag. BIBLIOGRAPHY BIBLIOGRAPHY Books Birmingham, Walter; Neustadt, I.; and Omaboe, E. N., eds. [1966]. Study of Contemporary Ghana, Vol. 1: The Economy of Ghana. Evanston: Northwestern University Press. [3’ . [1967]. A Study of Contemporary Ghana. V01. 2: Some Aspects of Social Structure. London: George Allen and Unwin, Ltd. Bradford, Lawrence A. and Johnson, Glenn L. [1953]. Farm Management Analysis. New York: John Wiley and Sons, Inc. Collinson, M. P. [1972]. Farm Management in Peasant Agriculture: A Handbook for Rural Development Planning in Africa. London: Praeger Publishers. Gittinger, J. Price. [1972]. Economic Analysis of Agricultural Projects. Baltimore: The Johns Hopkins University Press. Harberger, Arnold. [1970]. Project Evaluation. London: MacMillan. Hayami, Yujiro and Ruttan, Vernon W. [1971]. Agricultural Develgpment: An International Perspective. Baltimore: The Johns Hopkins Press. Johnston, Bruce F. and Kilby, Peter. [1975]. Agpiculture and Struc- tural Transformation: Economic Strategies in Late Developing Countries. New York: Oxford University Press. Killick, Tony. [1966]- "Agriculture and Forestry" in A Study_of Contemporary_Ghana, Vol. 1: The Economy of Ghana, Walter Birmingham, I. Neustadt and E. N. Omaboe, eds. Evanston: Northwestern University Press. Little, I. M. D. and Mirrlees, J. A. [1974]. Project Appraisal and Planning for Developing Countries. New York: Basic Books, Inc. Mishan, E. J. [1973]. Economics for Social Decisions: Elements of Cost-Benefit Analysis. New York: Praeger Publishers. Turnham, David and Jaeger, Ingelies. [1971]. The Employment Problem in Less Developed Countries: A Review of Evidence. Paris: OECD DevelopmenE‘Centre Studies, Employmentheries No. l. Upton, Martin. [1973]. Farm Management in Africa: The Principals of Production and Planning. London: Oxford University Press. 191 192 Wills, J. B., ed. [196]]. Agriculture and Land Use in Ghana. Oxford: Wills Ltd. Publications and Reports of the Government, Learned Societies and Other Organizations Agency for International Development, Development Assistance Program. [1975]. "Agricultural Sector Assessment." FY1976-FY1980, January, pp. 1-270. . Development Assistance Program. [1975]. "The Role of Women in Ghanaian Agriculture: A Preliminary Overview." FY1976-FY1980, January, pp. 271-280. Benneh, George. [1972]. "The Response of Farmers in Northern Ghana to the Introduction of Mixed Farming: A Case Study." Ann Arbor: Center for Research on Economic Development. Byerlee, Derek. [1973]. "Indirect Employment and Income Distribution Effects of Agricultural Development Strategies: A Simulation Approach Applied to Nigeria." African Rural Employment Paper No. 9. East Lansing: Michigan State University, Department of Agricultural Economics, African Rural Employment Research Network. Eicher, Carl; Zalla, Tom; Kocher, James; and Winch, Fred. [1971]. "Employment Generation in African Agriculture." East Lansing: Michigan State University, Institute of International Agriculture. Food and Agriculture Organization. [1972]. "Rural Employment Problems as Related to the Agricultural Development Programmes in Africa." Seventh FAO Regional Conference for Africa, Libreville, Gabon: September 25-30. Friedrich, K. H. [1971]. "Manual for Farm Management Investigations in Developing Countries." A Working Paper, Misc. 71/86, November 11., FAQ. Gemmill, Gordon and Eicher, Carl K. [1973]. "A Framework for Research on the Economics of Farm Mechanization in Developing Countries." African Rural Employment Paper No. 6. East Lansing: Michigan State University, Department of Agricultural Economics, African Rural Employment Research Network. Iftikhar, Ahmed. [Unknown]. "Technology and Employment Project, The Green Revolution, Mechanization and Employment." World Employment Programme Research Working Papers. Geneva: International Labour Office. Kline, C. K. et. a1. [1969]. Agricultural Mechanization in Equatorial Africa. Research Report No. 6. East Lansing: Michigan State University, Institute of International Agriculture. ll. [11' llllr .III‘ [1“ I l 1‘. l I I III I all llv.lll i l I] 193 Lipton, Michael and Moore, Mike. [1972]. "The Methodology of Village Studies in Less Developed Countries." Institute of Development Studies at the University of Sussex, Brighton, U.K., IDS Discus- sion Paper No. 10, June. Ministry of Agriculture, Economics and Marketing Division. [1972]. Report on Ghana Sample Census of Agriculture, Vol. 1. Accra: Republic of Ghana. . [1972]. Rice Development Prpject Feasibility StudyyMain Report and Appendices.Accra: Government of Ghana. E“‘ Miracle, Marvin P. and Serdman, Ann. [1968]. "State Farms in Ghana." University of Wisconsin, Land Tenure Center, March. \L' Norman, D. W. [1972]. "An Economic Survey of Three Villages in Zaria Province, Input—Output StUdy, V01. 1 Text," Samaru Miscellaneous 9;” Paper #27, Institute for Agricultural Research, Samaru, Ahmadu Bello University. . [1973]. "Economic Analysis of Agricultural Production and Labour Utilization Among the Hausa in the North of Nigeria." African Rural Employment Paper No. 4. East Lansing: Michigan State University, Department of Agricultural Economics, African Rural Employment Research Network. . [1973]. "Methodology and Problems of Farm Management Investigations: Experiences from Northern Nigeria." African Rural Employment Paper No. 8. East Lansing: Michigan State University, Department of Agricultural Economics, African Rural Employment Research Network. Papanek, G. F. and Qureshi, M. A. [Unknown]. "The Use of Accounting Prices in Planning" in Organization, Planning and Programming for Economic Development. Washington, D. C. Rourke, B. E. [1971]. "Wages and Incomes of Agricultural Workers In Ghana." Technical Publication Series, No. 13. Legon: Institute of Statistical, Social and Economic ResearCh, University of Ghana. Spencer, Dunstan S. C. [1972]. "Micro-Level Farm Management and Production Economics Research Among Traditional African Farmers: Lessons from Sierra Leone." African Rural Employment Paper No. 3. East Lansing: Midhigan State University, Department of Agricul- tural Economics, African Rural Employment Research Network. Steel, William F. and Mabey, S. J. [1973]. "Unemployment and Income in Ghana Since 1960, An Annotated Bibliography." Montreal: McGill University, Centre for Developing Area Studies. 194 Tollens, Eric F. [1975]. "Problems of Micro-Economic Data Collection on Farms in Northern Zaire." African Rural Employment Working Paper No. 7. East Lansing: Michigan State University, Department of Agricultural Economics, African Rural Employment Research Network. United States Department of Agriculture/A.I.D. [1968]. Rice in West Africa. Washington, D. C.. Ward, William A. [1975]. "Incorporating Employment into Agricultural Project Appraisal: A Preliminary Report." African Rural Employment Working Paper No. 6. East Lansing: Michigan State University, Department of Agricultural Economics, African Rural Employment Research Network. (February). Periodicals Amoateng, A. M. [1971]. "Rice Production in the Northern Region," Ghana Farmer (December). Benneh, George [1972]. "The Response of Farmers in Northern Ghana to the Introduction of Mixed Farming: A Case Study," Georafiska Annaler, 54, Ser. B: 95-103. . [1972]. "Systems of Agriculture In Tropical Africa," Economic Geography, 48, No. 3 (July): 244-257. Due, Jean M. [1973]. "Development Without Growth--The Case of Ghana in the 1960s?" The Economic Bulletin of Ghana, 3, No. 1: 3-15. . [1971]. "Efficiency of Resource Use--The Case of the Ghanaian State Rice Farms," East African Journal of Rural Development, 4, No. 2: 77-94. Green, R. H. [1961]. "The Ghana Cocoa Industry," Economic Bulletin, 5, No. 1 (May). Knight, J. B. [1972]. "Rural-Urban Income Comparisons and Migration in Ghana," Bulletin Oxford University Institute of Economics and Statistics, 34, No. 2 (May): 119-228. Marsden, Keith. [1969]. "Towards a Synthesis of Economic Growth and Social Justice," International Labor Review (November). Steel, William F. [Unknown]. "Import Substitution and Excess Capacity in Ghana," Oxford Economic Papers, 24, No. 2: 212-240. 195 Unpublished Materials Franzel, Steven. [1974]. "Import Substitution of Food Products in Ghana." Cornell Agricultural Economics Staff Paper, No. 74-9, May (Mimeographed). Goodwin, Joseph Baxter. [1975]. "An Analysis of the Effect of Price Distortions on the Development of The Rice Milling Industry In Ghana," Ph. D. dissertation, University of Maryland. Ministry of Agriculture, University of Ghana, and Council for Scientific and Industrial Research (CSIR). [1970]. '"Proceedings of the Ghana Agricultural Conference on Rice Production, February 4-6, 1970." Accra: (August) (Mimeographed.) . [1971]. "Proposal for a Rice Development Project in Northern Ghana." Accra: Republic of Ghana, April (Mimeographed.) Roemer, Michael and Stern, Joseph J. [1972]. "Project Appraisal: Notes and Case Studies (Part I: Notes)," unpublished paper prepared for Project Appraisal Course, Accra: Ghana Institute of Management and Productivity, May (Mimeographed.) Rourke, B. E. and Hiadzi, H. H. S. [1970]. "The Use of Tractors in the Accra Agricultural District." Preliminary Report presented to the National Advisory Committee on Agricultural Mechanization. Accra: January 22. (Mimeographed.) Spencer, Dunstan S. C. [1973]. "The Efficient Use of Resources in the Production of Rice in Sierra Leone: A Linear Programming Study." Ph. D. dissertation, University of Illinois. . [1972]. "Enumerators Reference Manual," Rice Survey, 1971-72. Department of Agricultural Economics and Extension, Njala University College, University of Sierra Leone. (Mimeo- graphed.) United Nations, Economic and Social Council, Economic Commission for Africa. [1975]. "The Food Situation in Africa and a Programme of Action." (Mimeographed.) Winch, F. E. [1974]. "A Financial Evaluation of Combine Harvesters Operated by the Ministry of Agriculture in the Northern Region, 1973/74." Accra: A Report Prepared for the Director of Agricul- ture, Ministry of Agriculture. (Mimeographed.) Wylie, G. M. [1972]. "World Bank Rice Project Feasibility Report, Job 5--Farm Machinery Supply." Accra: Ministry of Agriculture, January. (Mimeographed.) —"'.‘€