ECONOMIC OPTIMA IN RESOURCE ALIOOAIIOIILIZQJ‘E}? FOR SMALLHOLDER SUBSIS‘I’ENCE FARMING IIIIj’; GHANA I DIsserIatIon foI IIIO Degree Of EII D I MICHIGAN STATE UNIVERSITY YIADOM KWASI AITA KONADU 1974 i; ii LIBERIA” # Mmbxgan Univemry This is to certify that the I thesis entitled ECONOMIC OPTIMA IN RESOURCE ALLOCATION FOR SMALLHOLDER SUBSISTENCE FARMING IN GHANA presented by I YIADOM KWASI ATTA - KONADU has been accepted towards fulfillment of the requirements for AGRICULTURAL ECONOMICS 9%] WWW Major professor Lester V. Manderscheid Date September 20, 1974 0-7 639 gar ABSTRACT ECONOMIC OPTIMA IN RESOURCE ALLOCATION FOR SMALLHOLDER SUBSISTENCE FARMING IN GHANA By Yiadom Kwasi Atta—Konadu The primary objective of this study was to investi— gate optimal resource use for smallholder subsistence farmers--information needed to evaluate issues and emerging policies associated with smallholder producers of food crops in selected regions in Ghana. The major concern was to provide some insights into efforts necessary for expanding the productive potentials of the farms delineated in the study. The study was designed to interface with the maize improvement project of the Government of Ghana. Specifically, the issues were: 1) resource utilization and profit maximizing plans consistent with initial resource endowments and expanded resource use; 2) competitive position of crops produced using new technology and crops produced in nfixtures using indigenous technology; 3) dynamic inter- dependence between production, subsistence consumption and investment/disinvestment; 4) the use of on—farm storage of crops as additional means of increasing farm income; and 6) increased efficiency in labor utilization. If Yiadom Kwasi Atta-Konadu The methodology used included the use of static linear programming and poly-period linear programming to assess the income increasing possibilities for the repre- sentative farms by an optimum allocation of resources actually used by the farmers in the sample. The representa- tive farms were defined by the level of technology of production and by the ability to adopt agricultural innova- tions. The analysis was conducted in three empirical phases and two types of representative farms located in five regions in the country, viz. Brong-Ahafo, Ashanti, Central, Eastern and Volta regions. Phase I was designed to investi- gate the optimal allocation of currently available resource using currently utilized technology. The Phase II model incorporated on-farm storage activities and allowed borrowing up to optimum levels instead of putting a restriction on the amount of money that could be borrowed. Phase III, the Phase II model was expanded to include parallel cropping activities representing two alternative advanced technologies of producing crops in pure—stand. The data used were collected from a sample of 361 operating holders through intensive farm management survey carried out for a period of fifteen months during 1972-73. The holders were interviewed to obtain information regarding actual resource constraints facing them, the input-output relations encountered by them and food consumption. Several important policy implications emerge from the findings of this study. First, on all representative Yiadom Kwasi Atta-Konadu farms, the marginal value products (MVP's) of land and money capital were high, suggesting that increasing the use of these resources would lead to income gains. A large income increasing possibility was also indicated by large MVPs of agricultural inputs complementary to land such as labor, fertilizers, planting materials and farm implements. Regional variations in the magnitudes of the MVP's were indicated. Second, for all the representative farms mixed— cropping held a comparative advantage over pure—stand cropping, as shown by the magnitudes of the relevant shadow prices. The implication is that given the choice, the farmers would prefer growing crops in mixtures rather than in pure-stand—-a situation that would appear to militate against the introduction of new technology and/or enterprise specialization. Third, the results indicate that organiza- tion of an adequate credit supply is the starting point of any program to encourage the farmers to increase resource use. Credit policy should aim at providing credit to the farmers taking into account expected returns, production and household consumption requirements rather than using arbitrary rules. Fourth, significant income gains can be derived by removing the bottlenecks that lead to under— utilization of agricultural labor. One policy option discussed is the provision of a network of feeder roads and an organization of mass transit services to serve the farming communities. Fifth, the results provide the basis not only for direction in general product and input policy ion _.E is '5; ! Yiadom Kwasi Atta-Konadu formulations, but also indicate the magnitudes by which relevant policy variables such input subsidies and guaranteed . nunimum prices could be manipulated to achieve specified development goals. Major research needs highlighted by this study include: . 1) an incorporation of stochastic factors such as weather variability and risk and uncertainty associated with the adoption of new technology; 2) an expansion of the periods covered in the poly—period model to more rigorously a) investigate the dynamic interdependence of between production, consumption and investment/disinvestment; and b) account for the full production cycle of crops such as cassava and plan— tains often left in bush fallow and undergo continuous harvesting over an extended period of years; 3) economics of mixed-cropping vis—a—vis pure—stand cropping; and 4) benefit-cost analysis of feeder road construction and the building and location of storage facilities. The macro—effect of storage operations on prices will need an investigation also. ECONOMIC OPTIMA IN RESOURCE ALLOCATION FOR SMALLHOLDER SUBSISTENCE FARMING IN GHANA By Yiadom Kwasi Atta-Konadu A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Agricultural Economics 1974 C) Copyright by YIADOM KWASI ATTA—KONADU 1974 To beloved R, Morya, Lancllo and K—l7, my gratitude ii all and Lest grat are Stri Joh‘ ACKNOWLEDGMENTS The author would like to express his gratitude to all individuals who were very helpful in the development and completion of this thesis. Special thanks go to Dr. Lester Manderscheid (major professor and thesis supervisor) for his guidance and encouragement throughout the author's graduate work at Michigan State University. Sincere thanks are due to Dr. Stephen Harsh for his assistance in con- structing the models used in this study and to Drs. Glenn Johnson and Warren Vincent for their constructive criticism of an earlier draft of this thesis. The author is also grateful to the Ford Foundation for general financial support of his graduate work; the Harvard Development Advisory Service for providing the Opportunity for the graduate training at Michigan State University (in particular, individuals such as Drs. Joseph Stern, Ellon Gilbert and Roger Sellers exerted a composite intellectual influence); the Ministry of Agriculture, Ghana, for giving the author study leave and providing generous funding for the field work in Ghana; the University of Ghana, Legon, for the use of university facilities during the field work; the Bank of Ghana for providing staff support for the field work; Judith Stephenson, Kathy Ely and Teresa Owens for their aid in the computational work; and finally Theresa Abroaquah for her assistance in carrying out the field work. Finally, the author wishes to express his gratitude to Our Heavenly Father for the Gift of Life and all its accoutremen t S . iv TABLE OF CONTENTS Chapter Page I INTRODUCTION . . . . . . . . . . . . . . . 1 Nature of the Smallholder Problem . . . 2 Dimensions of the Smallholder Problem . . . . . . . . . . . . . . A The Objectives . . . . 7 Policy Issues Arising Out of the Problems of the Farmer . . . . . . . . 9 Scale of Operations . . . . . . . . 11 Money Capital . . . . . 12 Storage and Timing of Sales . . . . 1“ Accessibility to Farms . . . 14 Input Supply and Product Marketing . 15 Production Campaigns to Expand Food Output Capacity . . . . . . . 16 Specific Program Instruments . . . . 1? Concluding Remarks . . . . . . . . . . 18 II RESEARCH STRATEGY . . . . . . . . . . . . 19 Analytical Approaches . . . . . . . . . 19 The General Approach . . . . . . . . 20 Sources of Data . . . . . . . . . . . . 21 Data Collection . . . . . . . . . . 21 Analytical Models 24 Phase I . . . . . . . . . . . . . . 25 Phase II . . . . . . . . . . 26 Phase III 26 26 Concluding Remarks Chapter III AREAS STUDIES Similar Features of the Areas Studied Demographic Characteristics Climate . . . . . Contrasting Features Soils Vegetation Size of Farms Predominant Crops and Consumption Patterns . . . . . . . . Urban Influences Remarks Implications of These CharaCteristics for the Selection of Representative Farms for the LP Model . . . Introduction Classification of the Sample .Farms by Technological Category Representative Farm Characteristics Land Use Farm Labor Force Casual Labor Farm Capital . Cropping Pattern Crop Yields . i I I A l } Concluding Remarks I IV THE STRUCTURE OF THE LP MODELS FOR THE STUDY . . . Introduction . . . Phase I--The Intra— Firm Linear Lx Programming Model The Objective Function The Activity Set Crop Activities Purchasing Activities Labor Activities vi Page AS A5 A6 A6 118 50 59 Chapter Page Food Buying, Consumption and Sales Activities . . . . . . . . 61 Land and Financial Activities . 67 The Constraint Structure . . . . . 69 Agricultural Land . . . . . . . 69 Agricultural Labor . . . . . . . 70 On—Farm Labor Estimation . . 73 Off—Farm Labor . . . . . . . 74 Labor Overhead . . . . . . . 74 Money Capital Constraint . . 75 Output Balance . . . . . . . 75 Borrowing . . . . . . . . . . 76 Food Consumption Constraint . 76 Non—Negative Constraints . . . . 77 Phase II--The Poly—Period LP Model . . 77 The Objective Function . . . . . . 78 Additional Activities . . . . . . 79 Cash Flows . . . . . . . . . . . 79 Storage Activities . . . . . . . 81 Phase III—-New Technolog . . . . . . 85 Concluding Remarks . . . . . . . . . . 86 V OPTIMUM FARM ORGANIZATION WITH EXISTING RESOURCE AND RESPONSE COEFFICIENTS . . . 87 Programmed Solution of Phase I Results by Category and by Region . . . . . . 92 Comparison of Results of Phase I with Observed Sample Data: Category I and Category II Farms . 104 Category I Farms 104 Overview . . . . . . . . . . . . 108 Category II Farms . . . . . 108 Regional Comparison of Income and Farm Organizations by Technological Category Actual Versus Programmed . Comparison of Cropping Plans Under Programmed and Actual Conditions 110 113 Chapter Page Average Returns on Resources . . 114 Labor Use . . . . . . . . . . . 115 Optimal Solution of Phase II Model . . 116 Comparison of Optimal Organization and Income with Actual Organiza— tion and Income by Region and by Category . . . . . . . . . . . 120 Category I Farms . . . . . . . . 121 Category II Farms . . . . . . . 121 Regional Comparison of Farm Organizations by Category . . . . . 125 Comparison of Income, Marginal Value Products and Average Returns by Region and Category . . . . . 127 Programmed Income: Category II Farms, Phase III . . . . . . . . . . . . . . 130 Brong-Ahafo Region . . . . . . . 132 Ashanti Region . . . . . . . . . 133 Eastern Region . . . . . . . . . 134 Central Region . . . . . . . . . 134 Regional Comparison of Results of Phase III with Observed Sample Data: Category II Farms . . . 135 Regional Comparison of the Impact of Technology on Income, Employment and Farm Organization . . 137 Concluding Remarks 140 Summary of Phase I Results . . . . 141 Category II Farms 141 Category I Farms . . . . . . . 142 Cropping——General . . 143 Relevance of Subsistence FoOd Requirements . 143 Comparison of Optimum Plans with Actual . . 143 Summary of Phase II Results 144 Inter- -Area Comparison of MVPs i2; Labor Use viii Chapter Page Average Return Per Resource . . . 146 Crop Plans . . . . . . . 146 Subsistence or security Requirements . . . . . . . . . 146 Income Gains . . . . . . . . . . 147 Summary of Phase III Results . . . . 147 VI EFFECTS OF RESOURCE EXPANSION ON INCOME AND FARM ORGANIZATION . . . 148 Discussion of Category II Farms by Region: Phase I Category II Farms—- Central Region . . . . 151 Category II Farms——Eastern Region . . . 154 Category II Farms-—Ashanti Region . . . 158 Category II Farms——Brong-Ahafo Region . 161 Discussion of Category I Farms—— Phase I . . . . . . . . . . 164 Programmed Income Category II Farms——Phase II . . . . 171 Central Region . . . . . . . . . . . 172 Eastern Region . . . . . . . . . . . 173 Ashanti Region . . . . . . . . . . . 174 Brong-Ahafo Region . . . . . . . . . 174 Discussion of Category I Farms-- Phase II . . . . . . . . . . . 175 Concluding Remarks . . . . . . . . . . 176 VII POLICY ISSUES, SUMMARY AND CONCLUSIONS . . 179 Credit . . . . . . . . . . . . . . . . 180 Distribution . . . . . . . . . . . . 180 Rate of Interest . . . . . . . . . 183 Conditions for Credit . . . . 184 Farm—Household Interdependence . . . 185 Other Relevant Issues . . . . . . . 186 Labor Utilization . 186 Storage 188 Effects of varying Maize PriCe on Farm Income and Adjustment 189 Input Prices: Subsidies 193 Farm Size Factor 196 Summary . 197 Conclusion 200 ix Chapter .ouufiow goo . >07u5m EOHM 6099 .33 imp—own .aoawod 332 r e t p a h C .wumv zu>~=m Eouw vuusn—Eoo "wuuaom omww. A “hams u Smwu uumuw m o n H Al ngwtd w m coupon can: do ammo m o u H T T Emma w 0 Stan can: no :umu n o u .n HI HI mung v m town—mm a. 6 u u: no 5.413 c S o n a an Hi «mm—wfi. u q coupon van: :0 ammo m o n H T. Ht 953 m m vowuwn was: do :93 a o n H a- 7 255 u N BER. can: no ammo m o u H .7 HI Amy—ma m A v3.3a 3.8: do :mmu N o n H o H5027. o mmwoJI o wuamoJl o mw~oo.~| 0 3.3027 o 2.41 “I 20 u acuuugu o>auuanao .— HEP—m m5— = camcd 035—8 mamwcd mama—S #555 «35—8 mama mamas NAP—ma szmon H935 H358 ammo gmz “3:5 0:52 EH00 6: shown no: :3. Samoa cum—2193.3 .mluum HH AnomUuuu minim :wau Ha Hove: .63 0.33. .. PIE . .15.“. War... 3... flat £1. Chapter Page A Limitations and Suggestions for 1 Future Research . . . . . . . . 203 APPENDICES 206 207 229 230 "dr-VJUOUJP 236 BIBLIOGRAPHY 239 Table 2.1 3.2 3.3 3.4 4.3 4.4 4.5 4.6 4.7 4.8a LIST OF TABLES Selection of Holdings. Rainfall Profile in Project Areas in Inches. Percentage of Farms by Technological Category in Sample Areas . . Salient Features of the Farms in Five Regions: Brong—Ahafo, Ashanti, Eastern, Central and Volta (Ghana 1972) . . Average Acreage of Different Crop Enterprises by Technological Categories by Region. . Reasons for Mix- crOpping as Given by the Farmers: Number of Farmers Responding to~Specific Questions . . . CrOp Activities for Phase I Category II Brong— —Ahafo Region . Indut Purchasing Activities: Category II Fadms, Brong-Ahafo Region . . . . . . Labor Activities: Category II, Brong—Ahafo. Food Buying, Consumption and Sales Activities: Category II Farms, Brong-Ahafo Region . . . . . . . . . . . . . . . . . . Land and Input Purchasing Activities: Category II Farms, Brong—Ahafo Region Model II Cash Flows Category II Farms, Brong— —Ahafo Region . Page 23 31 36 37 41 42 51 52 53 54 55 56 Additional Activities and Constraints . . . 57 Marketing Ratios of Sample Maize Holdings in the Eastern Region. . . . . . 64 xi Table 4.8b 5.6 Page Marketing Ratios of Sample Maize Holdings in the central Region . . . . . 64 Marketing Ratio for Maize - Volta Region (Region 2) . . . . . . . . . . . . . . . . . 65 Marketing Ratios for Maize: Sample Maize Holdings, Brong-Ahafo Region . . . . . 65 Marketing Ratios for Maize: Sample Maize Holdings, Ashanti Region. . . . . . . . 65 Weights (C. ) for Conversion of Different Age— Sex Cohorts into Man— Equivalent Units . . 74 Relative Value of Maize Storage . . . . . . . 83 Characteristics of and Optimal Organization , for Category I Farms All Regions, Ghana, ~ 1972— 73 . . . . . . . . . 93 Characteristics of and Optimal Organization of Category II Farms, Ashanti, Brong-Ahafo, Central and Eastern Regions . . . . . . . . . 97 Comparison of MVPs, Salvage Values and Acquisition Cost of Labor by Region, Ghana, 1972- 73 (Category II Farms) . . . . . . . . 100 MVP of Resources: Category I Farms by Region (Phase I) . . . . . . . . . . . . . . 101 MVPs of Resources: Category II Farms by Region (Phase I) . . . . . . . . . . . . . . 103 Comparison of Phase I Results with Observed Sample Data Category I Farms by Region . . . 105 Comparison of Phase I Results with Observed Sample Category II Farms by Region . . . . . 106 Gross Income: Actual and Programmed Category I and Category II Farms All Regions . . . 111 Characteristics of and Optimal Organizations of Category I Farms Phase II—-The Poly- Period Model, Ghana, 1972— 73 . . . . . 117 Comparison of Phase II Results with Observed Sample Data Category I Farms by Region . . . 122 xii Table 5.7 5.9 5.10 6.1a 6.1b 6.2a 6.3a 6.3b 6.4a 6.5 Page Comparison of Phase II Results with Observed Sample Data Category II Farms by Region . . . 123 MVPs of Resource of Category I and Category II Farms-~Pahse II by Region . . . 128 Comparison of Phase III Results with Observed Sample Data Category II Farms . . . 136 Phase III Cropping Plan . . . . . . . . . . . 138 Net Income and Marginal Value Products with Resource Expansion on Transitional Farms: 24 Category II Farms, Central Region, Ghana, 1972— 73 . . . . . . . . . . . . . . 152 Summary of Economic Measures of Efficiency Category II Farms Central Region, Ghana 1972— 73 . . . . . . . 153 Net Income and Marginal Value Products with Resource Expansion on Transitional Farms, 22 Category II Farms, Eastern Region, Ghana, 1972—73 . . . . . . . . . 155 Summary of Measures of Economic Efficiency Under Varying Levels of Prices Category II Farms, Eastern Region, Ghana, 1972—73 . . . . 157 Net Income and Marginal Value Products with Resource Expansion on Transitional Farms: 22 Category II Farms, Ashanti Region, Ghana, 1972—73 . . . . . . . . . . . . . . . 159 Summary of Measure of Economic Efficiency Category II Farms Ashanti Region, Ghana, 1972- 73 . . . . . . . . . . . 160 Gross Income and Marginal Value Products with Resource Expansion on Transitional Farms: 22 Category II Farms, Brong-Ahafo Region, Ghana, 1972—73 . . . . . . . . . . . . 162 Summary of Measures of Economic Efficiency Category II Farms Brong— —Ahafo Region, Ghana, 1972— 73 . . . . . . 163 Net Income and Marginal Value Products with Resource Expansion on Traditional Farms: 50 Category I Farms, Central Region, Ghana, 1972- 73 . . . . . . . . . . . . . . . . 165 xiii Table 6.6 Net Income and Marginal Value Products with Resource Expansion on Traditional Farms: 72 Category I Farms, Volta Region, Ghana 1972— 73 . . . . . . . . Net Income and Marginal Value Products with Resource Expansion on Traditional Farms: 50 Category I Farms, Eastern Region, Ghana 1972- 73 . . . . . . . . . . . . Net Income and Marginal Value Products with Resource Expansion on Traditional Farms: 52 Category 1 Farms, Ashanti Region, Ghana, 1972— 73 . . . . . . . . . . . . Gross Income and Marginal Value Products with Resource Expansion on Traditional Farms: 52 Category I Farms, Brong—Ahafo Region, Ghana, 1972—73 . . . . . . . . . . . Summary of Economic Measures of Efficiency Under Varying Resource Level Category 1 Farms All Regions . . . Effects on Farm Organization and Income of Varying Maize Prices with Other Prices and Resources Held Constant . . . . . Price Ranges for Maize Sales, Consumption and Storage Activities for Maize . . . . . . Marginal Value Products and Resource Level of Category II Farms, Brong— Ahafo Region Ghana, 1972-73, Phases II and III . . Summary of Optimum Farm Plans Under Variable Resource Level (Phases II and 111), Category II Farms, Brong-Ahafo Region, Ghana, 1972—73 . . . . . . . . . . . Marginal Value Products and Resource Level of Category II Farms, Ashanti Region, Ghana, 1972— 73, Phase I and II . . . . . A Summary of Optimum Farm Plans Under Variable Resource Level (Phases II and 111), Category II Farms, Ashanti Region, Ghana 1972— 73 . . . . . . . . . . . Page 165 166 167 168 170 191 192 206 209 211 214 216 Table B. P P P P P H 3a 3b .4a .4b Ln-DLON Marginal Value Products and Resource Level of Category II Farms Eastern Region Ghana, 1972-73 (Phases II and III). A Summary of Optimum Farm Plans Under Variable Resource Level (Phases II and 111) Category II Farms, Eastern Region . Marginal Value Products and Resource Level of Category II Farms Central Region, Ghana, 1972- 73 . . Summary of Optimum Farm Plans Under Variable Resource Level (Phases II and 111), Category II Farms, Central Region, Ghana, 1972- 73 . . . . . . . . . . . . . . Alternative Resource Expansion Category 1 Farms, Phase II, All Region . Population Census 1970. Brong—Ahafo Region . . . . . . . . Population Census 1970: Ashanti Region . Population Census 1970: Volta Region . Population Census 1970: Eastern Region . Population Census 1970: Central Region . Marginal Value Products and Prices of Seasonal Product Inventories XV 219 221 224 226 230 231 233 234 235 LIST OF FIGURES Figure 4.1 Marketed surplus xvi Page 62 CHAPTER I INTRODUCTION Theories, policies and programs relating to agricul— tural development should be buttressed on empirical research. This study is an attempt in this direction. The lackluster record of agricultural development in Ghana, other than cocoa, could perhaps be blamed on its predominantly autochthonous system of farming that has over the years undergone very little change. Features of this system of farming are (l) subsistence production and subsistence food demand; (2) mixed cropping1 and (3) structural manyness of smallholder farming. Under these conditions the key to agricultural transformation seems to lie in (1) fuller commitment of the indigenous farmers to the money economy; and (2) an analysis and understanding of the conditions under which production by the smallholder subsistence farmer can be improved and made more profitable to organize. Since Ghana's independence in 1957, the general direc— tion of its agricultural policy has remained virtually the M 1The category of mixed—cropping mentioned here should be distinguished from mixed farming commonly practised in India, Pakistan and other areas. Mixed—cropping used in this context refers to the planting of two or more crops at random on a plot of arable land. 1 same: production of food to feed the people and raw materials for industries and the promotion of export crops to earn foreign exchange. However, we have witnessed kaleidoscopic changes in programs to give effect to the policy. From Nkrumah's socialist approaches to agricul— tural development to the turnabout and rhetoric about the efficacy of the private enterprise system under the National Liberation Council (NLC)1 and Busia's regimes and now, to self—reliance—-the epitome of the National Redemption Council (NRC)2 Operation Feed Yourself. In all the discussions, theories, concepts and schemes adopted to implement agricultural policies, there has been a con- siderable lack of information about the most important form of agriculture in the country-—i.e., smallholder farming—- information needed to evaluate issues and emerging policies relating to the smallholder producer. Nature of the Smallholder Problem Ghana's agriculture is predominantly composed of smallholders. In the 1970 census of agriculture it was estimated that out of 805,200 holdings, 81 percent were Smallholder operators.3 The 805,200 holdings had an average -——~—__._.____ lNational Liberation Council (NLC) (February 1966- September 1969). 197 ) 2National Redemption Council (NRC) (Since January 2. 3See Report on Ghana Sample Census of Agriculture 1970, Vol. 1, 1972. (i (x 11 ll 41 It .I. r.\ (x 3‘ Ti { (.1 .1 I (I. H .H.. .1 h ‘ CL 0» a p mf size of 5.6 people resulting in estimated farm population of 4,517,800 people or roughly 50 percent of the total population of the country.4 Of the holders, 30.6 percent cultivated less than 2.0 acres, 54.7 percent less than four acres and 81.9 percent less than ten acres. Only about 18.1 percent of the holders cultivated ten acres or more. The census report also provides information on the extent of commercial orientation of the farmers. Out of 805,200 holders, 11,110 (14 percent) were classified as producing for subsistence only, while 289,700 (36 percent) were classified as mainly subsistence and 404,400 (50 percent were operating mainly for sale.5 These two features-—smallholder farming and subsistence production—— interact with mixed cropping to produce unique problems related to low labor productivity, low land productivity, food shortages, rising food prices and economic environ— ment which has shown itself uncongenial to accelerated mechanized farming. The following questions can be raised in respect of the seemingly poor performance of the small- holder subsistence farming: (1) is the cropping system 4Estimated at 8.5 million. 5The definition used for the classification were as follows: (a) operated for subsistence only—~no cash crop cultivated and little or no sale of food crOps, (b) oper- ated "mainly” for subsistence——more than 50 percent of produce intended for home consumption and, (c) operated mainly for sale——more than 50 percent of produce intended for sale. respon object availa object (A) to within famin insigh in Gha and al of goa responsible for the poor performance; (2) what are the objectives of the smallholder farmers; (3) are the resources available to them optimally organized when measured by the objectives of farming and prevailing state of arts' and (4) to what extent is the production potential utilized within the framework of existing pattern of smallholder farming? Answers to questions such as these provide insights into the conditions under which smallholder farming in Ghana can be restructured to make it more profitable and also point to the factors which constrain the attainment of goals defined by the farmers themselves. Dimensions of the Smallholder Problem In this thesis, we shall attempt to look at the present capacity of subsistence production and examine the production alternatives and the possibilities for expanding production capacity via increased efficiency in resource use. In Ghana today, not only food, but also fibre requirements seem to be outdistancing the capacity of the agricultural sector to produce them. To understand why this is so, one has only to examine the country's major constraints to agriculture. Constraints have dif— ferent ramifications which are pertinent to the discussion that follows. The root causes of insufficient supply could be attributed generally to such perennial constraints as a generally low level of technology, lack of price incentive, small 5 capital consid< ducti01 which 1 limit 1 limitii ist p0 the int and mi: 35am l'Uperh attent the gr 0f the SuPlSio small size of farm and insufficient resource base including capital, managerial knowehow,1abor and land. These can be considered as constraints of "nature” and could limit pro- duction capabilities. There is another type of constraint which can be considered as self-imposed and would needlessly limit production capacity also.6 The most outstanding and limiting of these self—imposed constraints are: (1) social- ist policies which in previous regimes in Ghana neglected the improvement of peasant farming; (2) subsistence farming and mixed-cropping; and (3) "Operation Feed Yourself” (OFY) as a mgdgg operandi. In a food shortage situation, ”self-reliance” or ”Operation Feed Yourself” can become an asset if adequate attention is initially paid to the crucial issue of laying the groundwork for long—term agricultural transformation of the economy, rather than relying on uncertain moral suasion to achieve quantitative targets in food production 6Following G. K. Helleiner [1969], self—imposed con— straints are used here to reflect certain political, social or economic rules or objectives which are to guide or con- strain one's development policies. A constraint on rules blocks off a range of possible alternative policies. The fact that the producers persist in adhering to their old rules of production——mixed-cropping and subsistence produc— tion—~they shut off other alternatives of production such as specialization and this behavior can be regarded as self—imposed. .. The socialist policies under Nkrumah's regime and the Operation Feed Yourself” are clearly self-imposed con- straints. The rules followed are political ones and they do block off a range of possible policies. irrespective of the cost. If development efforts echoed by the OFY are backed by research and experimentation, long-term gains can be expected. The critical issue facing smallholder farming in Ghana is that of carefully examining production alternatives and increasing the efficiency in the resource use in order to expand the present capacity of agricultural production. This point has forcibly been brought home by the present regime which has been reminding the people that their "survival depends largely on their ability to utilize fully the rich agricultural endowments of the country” [Ministry of Agriculture, 1972, p. 1]. Thus, honest appraisal of the abundant opportunities at hand for farming reaffirms the belief that the economy seems to have at its disposal the elements necessary and sufficient for solving its problem of rising food prices. The question of considering the possibilities of increasing farm returns through reorgan— ization of the available resources and enterprises appears to lie at the heart of the problem. Therefore, it would seem that a fuller utilization of the matrix of the present, through the commitment of the producers to commercial production, appears to be a sound approach to the solution of our farm problem and this disposition must dispel any tendency to put off until the morrow that which needs attention today if that morrow is to become a desirable 1.1 Calm] iight West that C alter reality.7 The popular presumption that the possibilities of such a reorganization are non—existent in a country like Ghana, is untenable. One can always point to Japan where, in the early days of its agricultural transformation, the farmers, even though lacking in capital resource (the most limiting resource), were still able to increase their efficiency and earnings by utilizing the surplus labor and other resources. These farmers used more surplus labor with very little capital, the principal increase in resource use emanating from a more complete utilization of unpaid family labor. This study seeks to explore the possibilities of such developments in Ghana. Operating within a modified framework of the self-imposed constraints referenced above,8 the study is designed to make some modest contributions to our understanding of the problems of smallholder farming in Ghana and also provide the missing links in the chain of knowledge and information needed to rationally formulate product and input policies at both the micro and macro levels. The Objectives The objectives of this study are: 1. To analyze the organization of subsistence M 8An example will be the incorporation of economic Calculations into the self-reliance policy. Such an approach might help prevent tying up scarce resources in unproductive Investments. In other words, the principle of comparative cost or advantage should not be needlessly sacrificed on the flier of self—reliance. farming in the major maize growing areas in Ghana so as to assess and appraise the economics of present resource use and the requisites for increasing agricultural output and farm incomes. 2. To determine the efficiency of resource utiliza— tion and profit maximizing plans consistent with the initial resource use, expanded resource use and technology of the categories of farming \ identified in the survey. i 3. To determine alternative technological potentials for producing farm output, which can be considered by the extension workers in their innovation diffusion efforts. 4. To evaluate the potential of the various policy instruments, such as product and factor prices, interest rates,on—farm storage, etc., which could be used to bridge the gap between actual and potential production and thus provide the frame- work for policy manipulations desired to achieve expanded food production and farm incomes in an optimal fashion. 5. To demonstrate the methodological reasonableness and efficacy in using linear programming techni— ques to examine the dynamics of on—farm storage of crop output with consideration given to con— I sumption withdrawals for family subsistence needs. L—1 ...._,__L A__. Policy Issues Arising Out of the Problems of the Farmer With the notable exception of maize, rice and beef, practically all the major food items consumed in Ghana are produced in the country. Expanded food production is needed to feed the expanding population adequately and nutritionally, both in the urban and rural sectors of the country. In the rural areas, food consumption patterns tend to be prescribed by availability, i.e., by what types of crops are grown in the vicinity. For instance, although maize growing occurs in all the study areas, it is only the main staple in three of the areas: Eastern, Central and Western regions. How- ever, with increasing trends towards urbanization near the food growing centers, it becomes necessary, not only to diversify to expand the scope of product mix in each region, but also to improve upon the distribution system to insure minimum delays in moving products to deficit areas. The market structure and the general pricing mechanism must be such that any long-term changes in food prices are quickly passed onto the producers and consumers without the ”mammy "10 truckers,"9 and "forestallers or the various intermediaries 9 . The expreSSLOn "mammy truckers” refers to the urban umrket 'mammies' or wholesalers who integrate backwards to run food transportation operations. 10The forestallers provide a link in the distribution chain between producers and urban wholesalers-~or the market mammies. Policy Issues Arising Out of the Problems of the Farmer With the notable exception of maize, rice and beef, practically all the major food items consumed in Ghana are produced in the country. Expanded food production is needed to feed the expanding population adequately and nutritionally, both in the urban and rural sectors of the country. In the rural areas, food consumption patterns tend to be prescribed by availability, i.e., by what types of crops are growu in the vicinity. For instance, although maize growing occurs in all the study areas, it is only the main staple in three of the areas: Eastern, Central and Western regions. How- ever, with increasing trends towards urbanization near the food growing centers, it becomes necessary, not only to diversify to expand the scope of product mix in each region, but also to improve upon the distribution system to insure Udnimum delays in moving products to deficit areas. The market structure and the general pricing mechanism must be such that any long-term changes in food prices are quickly passed onto the producers and consumers without the "mammy 9 "10 truckers,” and ”forestallers or the various intermediaries 9The expression ”mammy truckers" refers to the urban market 'mammies' or wholesalers who integrate backwards to run food transportation operations. 10The forestallers provide a link in the distribution chain between producers and urban wholesalers—~or the market mammies. absorb: benefit A ductim eertaix cocoyai increa: market: trend 1 the f0( Culturt and the Effect: attent; Effici, appropi increa: My 1 iSSUeS faI‘mer program Empim that mi lO absorbing a disproPOrtionate share of the accruing benefits. Agribusiness enterprises, which deal with the pro- duction of cassava chips, yam chips, plantain chips and certain convenient foods (i.e., instant yam "fufu,' instant cocoyam "fufu" and instant plantain ”fufu”ll ) are becoming increasingly important in the country. There are promising nmrkets, both domestic and foreign, for these items. This trend should encourage parallel expanded production in the food system. It would appear that the goal of agri- culture, looked at from the viewpoints of both the producers and the government, is expanded food production to match effective demand. In this study, we shall direct our attention to how this can be accomplished through more efficient utilization of resources and the selection of appropriate policy programs whose implementation can help increase the overall performance of the agriculture. In order to obtain a good perspective of the relevant policy issues, we shall examine here specific consequences of the farmer problems and match them with the corresponding policy programs. This will set the stage for directing our empirical analysis in this study and the policy implications that will emerge. 11"Fufu" a popular prepared food made from either plantain, cocoyams, yams or cassava or from a combination 0f yams and cassava, plantains and cassava and cocoyams and cassava in certain subjectively determined fixed proportions. 11 Scale of Operations In any economy where there is not much fixed invest— ments on farms in the form of buildings, land improvement structures, etc., farm size in the form of planted acreage is a reflection on the economic well—being of the farmer. Income flows from farming are typically allocated between family consumption and reinvestment in the farming opera— tions as a basis forthe generation of later income and 12 Thus, the ability of the small—holder consumption. farmers to generate these income flows is severely limited by the smallness of acreages farmed. With increasing pop— ulation pressures, the tendency to persist in cultivation of small acreages may lead to what Clifford Geertz [1963] "13 Alfred Dadson has termed ”agricultural involution. [1970] raises the question as to why the peasant farmers in Ghana cultivate such small areas if land is generally not scarce. Despite the popular views to the contrary, the tenurial arrangements as they exist in the project areas do not seem to constrain the ability of willing farmers to 12See Nakajma [1957, 1965], Mellor [1965a, 1965b], 1. J. Singh [1968], Wharton [1969] for a discussion of the farm-household interdependence. In a recent article by Lau and Yotopoulos 1973 , this interdependence has been referred to as ”non-block recursiveness". 13J. Dirck Stryker [1972] shows the reaction of peasant farmers to the increase in population density. In some cases, such as Java, he shows how food production was main- tained by continually increasing labor intensive techniques, while the marketed surplus was decreased——a process termed "agricultural involution." [Ho has Can tic 12 expand their acreages [Ollenu, 1971] and [Min and Fagger, 1971]. Nor is the notion of limited aspirations on the part of the farmers a good enough culprit since the literature documents several instances of peasant farmers responding to economic incentives [Schultz, 1964], [Chennareddy, 1967], [Hopper, 1965]. Rather the answer lies in what Dadson[1970] has termed " .the system of resource organization in indigenous farming:- The extensive and discontinuous pattern of land use, the heavy dependence on labor and the limited use of capital inputs." To help resolve the scale or resource proportionality problem the government has on—going programs to organize the farmers into cooperatives and to offer subsidized land clearing services to the farmers. In this context, an important policy consideration as we shall see below is that of providing money capital at low rate of interest to the farmers. Money Capital The production cycle of the crops covered in the study can be grouped into four Stages: (1) clearing and prepara— tion of land; (2) sowing, planting and fertilizing; (3) cultivation-—weeding and (4) harvesting. Each stage of the process requires capital, but the banks normally on their own will not provide credit to finance all the various stages (in practice, they choose the stage or stages at Whi( the hot: gov the ins 13 which the capital sum advanced would (a) ”be utilized to the optimum advantage to both parties, i.e., lender and borrower; and (b) be easily recovered with interest.” [J. E. Yeboa, 1968, p. 4]). However, both "lender and borrower may put different interpretations to what the”optimum advantage” implies, especially with respect to the timing of borrowing and repayment. The government has responded to the situation by providing generous, across—the—board, low-interest loans to the farmers. In spite of the availability of the government credit facility, one of the major complaints of the farmers interviewed was their lack of access to institutional loans. Where the loans were accessible, the rigidities in the bank's or government's lending require— ments——such as collateral and minimum acreage requirements (criterion)—-disqualify a majority of the farmers.14 The alternative to this type of loan is obviously noninstitu— tional money lending which invariably carries very high 15 interest rates. This study will attempt to investigate the effect of interest rates on farm organization and the 14To qualify for a loan, a farmer must be cultivating at least six acres. Referring to the census data in the Census Report, this implies that about 67.8 percent of the holdings in the country will be disqualified because they cultivate less than six acres. 15Loans for traditional money lenders are normally Carried for a short period. The high interest charges reflect the transaction cost and risk premium. exte prod I650 the IQVE orde in t has in] typr t0 1 and ensl the: One. qua SeVl Sta: [flay l4 extent to which the timing of borrowing during the crOp production year can change the character of other limiting resources . Storage and Timing of Sales Because of the extent of seasonal price movements, the farmers have an opportunity of increasing their gross revenue without necessarily increasing physical yields. This can be accomplished through storage activities in order to time marketing of their produce to the periods in which maximum gains can be obtained. The government has extension programs in operation to help the farmers in the techniques of storing their produce—-including the types of structures to use and the necessary steps needed to minimize storage losses. The dispersion of marketing and distribution facilities in the producing areas also ensures that the farmers always have a ready market for their produce. Accessibility to Farms The average farmer in the study areas resides about one-half hour walking distance from his land. The conse— quences of this are manifold: (l) in one production year, { several days are committed to walking which otherwise could be used working on the farm; (2) a heavy rainfall that starts early in the morning and continues to about 10 a.m. may put the farmers out of work for the whole of that day; high able that posi To c roac pro 1001 15 or (3) a mid-day heavy downpour, even for a brief period, ndght abort working efforts for the rest of the day. Since the hours spent in walking compete with avail- able hours for direct production efforts, it is conceivable that a reduction in the walking hours would contribute positively to expanding the scale of farm operations. To do this within the context of village living, feeder roads have to be constructed and mass transit services be provided at a cost less than the marginal value product of an hour of labor used in production. This study will 'look into this aspect of the farm problem. Input Supply and Product Marketing Insufficiency and nonavailability of inputs such as improved seeds, farm implements, fertilizers and other agricultural chemicals contribute to low crop yields. Related to the question of input supply is the diffusion of these new inputs and the techniques of their use to the farmers. Inadequate marketing facilities give rise to product price fluctuation, thereby affecting levels of production. The next section outlines the steps being taken 16 by the government and aid agencies to correct these difficiencies. 16The USAID Office in Ghana has been making substantial mnmxibution_through research and extension and direct financial aid in supporting programs in this area. 7": -o.._,..c-.~—r4" l6 ProductiOn Campaigns to Expand Food Output Capacity There are two outside agencies operating special pro— grams with the support of the Ministry of Agriculture: the USAID Focus and Concentrate program and the UNDP Project for ”Increased Farm Production through Fertilizer Use," usually known as the ”Ghana 20” Project or the "FAO Fertilizer Project." The aim of the Focus and Concentrate Project is to focus extension effort on a few farmers and to concentrate resources on them so that extension is not frustrated by the farmer's inability to acquire input. TWO of the study areas--Kpandu and Somanya in the Volta and Eastern Regions, respectively, are beneficiaries of this project. Participants in the project are chosen on the basis of their willingness to cooperate and their access to tillable land. For each selected participant, a farm plan is prepared Showing (a) farm layout, (b) CIOP rotation, (c) estimated labor requirements, (d) require- ments of seed, pesticide, fertilizer and equipment including custom-hired machinery services, a calendar of farm oper- ations, a farm budget and an achievement report. It is hoped that the perceived success of these farmers will be an effective way of diffusing agricultural innovations to their peers. The FAO Fertilizer Project includes not only all aspects of fertilizer use (through the laying out of demonstrations and trials in all parts of the country), but also tion of i practice: Tw< lmprovem Feed You: the draw contribu jects ca ahaize ture, 19 in Ghana Feed You Program SOlutior have Cor 'ong‘tei 17 but also deals with processing and marketing, the introduc— tion of improved seed varieties, storage and better cultural practices. Two other related production campaigns are the Crop Improvement Projects for maize and rice and the Operation Feed Yourself. The Crop Improvement Projects are still on the drawing board and this study is reckoned to make some contribution to the maize project. Details of those pro— jects can be seen in the following documents: ”Proposal for a Maize Development Project in Ghana" [Ministry of Agricul- ture, 1971] and "Proposal for a Rice Development Project in Ghana" [Ministry of Agriculture, 1971]. The Operation Feed Yourself campaign started in 1972 as an emergency prOgram or a rescue operation to quickly bring some temporary solution to the food deficit situation. The early successes have convinced the government that it can be used as a long-term development strategy. Specific Program Instruments The basic program instruments currently in use to further the objects of all these production campaigns are: (l) guaranteed minimum prices for maize to start with, but the government is presently considering expanding its scope to include other major food crops, such as yams, cassava, plantains, cocoyams and rice; (2) fertilizer subsidy, (3) subsidy on farm implements, such as machetes and (4) subsidizing the purchase of improved seeds and ensuring their timely distribution to the farmers. Bee resource study to the pro je economic l. Tl 3ha1y2e( 18 Concluding Remarks Besides enlarging the scope and content of optimum resource organization analysis, it is the purpose of this study to simulate the varied effect of policy decisions in the project areas by relating these decisions to the micro— economic aspects of decisions with regards to: l. Acreage expansion due to the provision of facili— ties for the farmers to clear more acreages for production. 2. Changes in infrastructure such as the provision of feeder roads and public transit services to cut down on the amount of time spent walking to farms. 3. Changes in the provision of credit-—in amount, rate of interest or timing of borrowing. 4. Changes in the subsidized prices of inputs such as fertilizers and other chemicals and, 5. Changes in output prices to evaluate the impact on farm income of different levels of guaranteed minimum prices. The effect of policies and other changes will be analyzed according to how they affect opportunities facing the representative farmer. The makes use and poly-‘ 5911319 su PIESented areas Stu t0 Which In Problem I resource the alloc reSpouses techniquE Problems. AIM Prommmj (3) aggr. ““1 <5) . each can CHAPTER II RESEARCH STRATEGY The general research approach for this thesis project takes use of a farm sample survey, static linear programming and poly-period linear programming. The details of the sample survey and the analytical techniques used will be )resented in this chapter. Chapter III will describe the ireas studied and the selection of representative farms :0 which the analytical procedures will be applied. Analytical Approaches In Chapter I, it was pointed out that the smallholder moblem receiving attention in this study centers around esource mobilization and allocation. The by-productsof he allocative problem are the issues associated with supply esponses and farm adjustment. There are several analytical echniques available to researchers seeking answers to such roblems. Among the several techniques used are: (1) linear rogramming in its multi—faceted forms; (2) budgeting; 3) aggregate time series analysis, (4) marginal analysis nd (5) simulation. These are not exclusively used, but ach can be used in combination of one or more of the others. 19 in generai depends, 1 the purpo of the st particula Lin Its most highly so farm adjt series d; tool in 5 most fea: Particul; P°nents J Budgetin its rele 0f farm “been; hing 6m Period l but the altas i' to defi 20 In general, however, the choice of an analytical technique depends, most importantly, upon the availability of data, the purpose for which the model is intended and the nature of the structural coefficients being sought to elucidate a particular problem. Linear programming is the approach used in this study. Its most important advantage lies in the fact that it is highly suitable for estimating supply functions and analyzing farm adjustment problems in an environment where no time series data exist. Simulation, generally, is a promising tool in such an environment and, perhaps, may provide the most feasible approach for analyzing the farm problem, particularly under imperfect knowledge as programming com- ponents and logical parts of simulation [Hart, 1967]. Budgeting is an alternative useful approach for assessing the relative profitability of different farm plans. However, its relevance lies in the availability of a sufficient number of farm plans to be evaluated. The General Approach The analytical techniques used to accomplish the objectives of this study involved the use of linear program- ming and cash-flow analysis within the framework of poly~ period programming. The procedures involved in carrying out the study included: (1) surveying Specified farming areas in five regions in Ghana; (2) using the sample data to define representative farm resource situations; (3) construe ming not the oper (A) prop and; (5) tiveness will be I] were de‘ and to . and ass interpr f°0d co by Skin in the Stands Dream ihcorp< type 0: 5919th "illist‘ 411er 21 constructing a structural framework for the linear program— ming models by determining the technological coefficients, the operational constraints and the activities or processes; (4) programming the representative farms in three phases and; (5) analyzing the factors which determine the effec— tiveness of given policy measures. The general approach will be elaborated in Chapter IV of this thesis. Sources of Data In Chapter I, the problem and objectives of this study were delineated. In order to answer the questions raised and to accomplish the objectives it was necessary to collect and assemble the data needed to analyze and meaningfully interpret the situation. Data Collection Data were required on farm organization, production, food consumption and resources. These data were collected y surveying farmers located in the study areas. The sample population was designed to include farmers in the specified areas who were producing maize in pure- tands or in mixtures with other crops. The population was restratified on the basis of geographic area so as to 'ncorporate differences in soils, vegetation, climate, ype of farming and urban influences. The five areas elected were the geographic areas earmarked by the inistry of Agriculture for the location of the Maize Crop provement Project. Th created sentativ loundati staff an the farm geneous 0f the i know-hon the com each are the are; ll Procedui used; the agr‘ (2) Pro' 8amPlin Samplin of Her the Cro ll Plete e drawn. Were to Each 22 The sampling rate was suggested by an ad hoc committee created by the Ministry of Agriculture comprising of repre- sentatives of (l) the USAID Office in Ghana; (2) the Ford Foundation staff; (3) Harvard Development Advisory Service staff and (4) the staff of Ministry of Agriculture. Because the farms in each geographic area were assessed to be homo- geneous in many characteristics such as the literacy levels of the farmers, farm size, cropping patterns, and technical know—how of the farmers, it was the general concensus of the committee members that approximately 50 holdings in each area would be fairly representative of farming in the areas. In the selection of farms and in the data collection procedure in each area a multi—stage sampling approach was used: (1) probability sampling of enumeration areas within the agricultural districts earmarked for the Maize Project; (2) probability sampling of holdings using nonuniform sampling plan, i.e., one that differed primarily as to the sampling fraction used in each area, and (3) random location of plots of prescribed dimensions in the fields for conducting the crop yield study. Within each selected enumeration area there was a com— lete enumeration of holdings from which the sample was drawn. The following table shows the number of holdings numerated by sample area, the sampling fraction applied 0 each region and the size of the sample. Table 2 Reg Brong-A Ashanti Volta F Easterr Central \ henty act as A St and grOUp lIOldings j in each a1 hocedure each area precision mllectio l. 23 Table 2.1. Selection of Holdings. Region Holdings Sample Minimum (Number) Fraction Sample (Percent) Size (Number) Brong-Ahafo 2,369 3.0 70 Ashanti 2,797 2.5 70 Volta Region 1,137 6.2 70 Eastern Region 356 19.7 70 Central Region 919 7.6 70 Twenty extra holdings were selected for each area to act as replacement for noncooperators. A statistically efficient procedure used was to rank and group the five areas on the basis of the number of loldings in each area. The larger the number of holdings in each area, the smaller the sampling fraction used. This rocedure was meant to equalize the size of the sample for ach area and thereby provide a comparable level of sampling recision among the areas. The following steps in the data ollection process were involved. 1. The first stage consisted of designing the sample and the questionnaire; pretesting the questionnaire; effecting the necessary adjustments; selecting and training secondary school leavers for the enumera- tion work. 2. Interviewing using open-ended questions to collect information on major resources, enterprises, farms and their location. pr vi The out in thr may increa methods as 24 3. Area measurement using tape and two inches prismatic compass. Fields farmed during the pre— vious three years were identified and measured to constitute the stock of unused land. 4. Yield estimation using the classical method of crop yield estimation by means of density plots. 5. Visiting each farm—household twice a week to collect data on food consumption, purchases, sales, gifts, age, sex and size of the farm house— hold and income, for a period of 12 weeks. 6., Coding, tabulating and computer services: after the schedules were compiled and the area measure- ment, yield study and consumption survey completed, the enumerators were assembled for a period of two weeks to be trained to code the data. Later, the data was punched and put on a tape to be sent to Michigan State University. The linear program— ming computations were carried out on the CDC 6500 using a combination of Harsh/Black program and CDC Apex—I program at Michigan State University. Analytical Models The linear programming model in this study was carried t in three major phases. On an individual basis, a farmer 9 increase his income by (l) adopting modern production Lhods as opposed to the traditional technology; (2) seeking — and selectin determine th and (3) 3de by the size to explore t to farm reor critical res Includ were three t borrowing, f labor, opera fixed at the representati clearing of objective be representat' farms were clearing an was to dete acres would like planta 25 d selecting the most feasible combination of activities to termine the higher profit plan to be adopted by the farmer; Ld'(3) adjusting the size of the farm business as indexed r the size of cultivated acreage. The LP models were used ) explore the individual alternatives to suggest guidelines 3 farm reorganization and to point to the magnitude of the ritical resources required for the change. Phase I Included activities for different cropping enterprises ere three types of labor activities, input purchasing, orrowing, food buying and product selling activities. Land, abor, operating capital, borrowing and consumption were fixed at the levels indicated by the survey data for the epresentative farms. In the initial programming analysis, :learing of unused land was not an alternative, the first bjective being to determine optimum farm plans for each epresentative farm with the existing acreage. Later, the arms were reprogrammed with the added alternatives of land learing and resource expansion. The idea being pursued as to determine whether changes in resource use on existing cres would yield more or less profit than changes in resource se including additional acreage.l -‘—__._. 1The predominant cropping system found in the study :eas was that of shifting cultivation. Under this system aw land (mostly secondary forest) is brought into cultiva- cn each year while the farm land of the previous year is 5ft behind, though because of continuous cropping of crops ke plantain, cassava and cocoyams, the bush fallow land — In Ph. poly-period cash flows, ming horizo seven perio This ought to do expanded ve nologies fo model by ad technologic derived fro: of improved with the a stochastic The zation of accomplish ‘ sideration is periodi During the were ident recorded. cultivated stand befo cassava in 26 Phase II In Phase II, linear programming techniques within a -period framework of analysis was used to incorporate flows, storage and additional land clearing. A plan— horizon of one crop year was used and was divided into n periods to coincide with the major farm operations. Phase III This phase was used to determine what the farmers rt to do to maximize their incomes. This phase is an Lnded version of the Phase II model. Alternative tech— ;gies for continuous cropping were incorporated into the :1 by adding additional rows and column activities. The rnological coefficients for the continuous cropping were .ved from existing experimental data. In the selection mproved production methods, the problems associated l the adoption of innovations which incorporate hastic factors were not considered. Concluding Remarks The focus of this thesis is on optimum resource organi— on of smallholder subsistence farming in Ghana. To mplish the objective of the study by means of a con- ration of possible adjustment activities, data on the ariodically maintained and the crops are harvested. 1g the survey, farm land of the years 1971, 1970 and 1969 identified, measured and crops still found on the land rded. In isolated cases, a piece of land was observed .vated for two years in succession under maize in pure- lbefore introducing crops like plantain, cocoyams and va in mixtures. — present c of the fa were app] adjustmer process ( area. T1 descripti 27 nt organization, production, consumption and resources re farmers were Collected. Linear programming models applied on these data to depict the basis from which stments could be made. Chapter III describes the ess of selecting representative farms in each study This is followed by Chapter IV with a fuller ription of the LP models used. Th by the M integratr Ti and Volt. Ministry increasi: The samp‘ the folll in Brong region; CHAPTER III AREAS STUDIED The sample survey was carried out in the areas selected Ministry of Agriculture, Ghana, for the proposed 'ated crop improvement project for maize in the country. Five regions; Ashanti, Brong—Ahafo, Eastern, Central rlta, are involved in the project operated by the :ry of Agriculture to promote the use of a yield— rsing package of inputs among the maize producers. rmple enumeration areas for this study were drawn from illowing agricultural districts: Wenchi and Atebubu rug—Ahafo region: Mompong and Ejura, in Ashanti .; Kpandu in Volta Region; Asesewa in Eastern region 'edru in Central region. The Ministry of Agriculture Study reports the following ia for the selection of the project areas [Ministry iculture, April 1971] from which the sample farms elected: (I) existence of a sizable market—oriented aroduction;1 (2) suitable soil and climatic conditions 'A year round rural marketing activities are carried the project areas. Wholesales or "mammy" truckers re urban center converge at these marketing centers ‘twice a week to do business. The farm gate prices 28 i for the on-going mity to facilit: developr signifir that thr a signi: A1 profile regions tion of cohorts occupat' in the communi service 29 the growing of maize; (3) existence and performance of going programs dealing with maize production; (4) proxi— y to major consuming centers and general state of transport ilities; and (5) existence of/or good prospects for the elopment of a package of improved practices which will nificantly improve net returns per acre. (6) Evidence t the existing land tenure situation would not constitute ignificant barrier to the adoption of improved practices. Similar Features of the Areas Studied Demographic Characteristics Appendix Tables D.l to D.5 provide the demographic file of the population in the sample areas of the five ions studied. In all the areas, occupational distribu- n of the age—sex cohorts shows that for both sexes in the orts, age 15 and above, agriculture is the predominant upation. Nonagricultural occupations of employed persons the project areas include: (a) workers in transport and nunication; (b) craftsmen, production process workers, lice, sport and recreation workers; (0) professional, inical, administrative, executive and (d) managerial, :ommodities or the prices actually received by the farmers the rural wholesale prices on these markets less the cost :ransporting the commodities to the markets. Local whole— >rs also buy at these markets, to be sold later at retail es, i.e., prices at which the local people can buy the odities for direct household consumption. The rural markets of national significance are Asesewa, ru, Kpandu, Mampong/Ejura, and Wenchi/Atebubu in the ern, Central, Volta, Ashanti and Brong—Ahafo regions, ectively. clerical native em their ava selling a tions ide‘ including ments (su serving a provides The with resp the year. November , Bec of rainfa 3O rical workers, sales workers. These jobs provide alter- ive employment opportunities for the farm workers and ir availability enabled us to incorporate family labor ling activities in the LP model. Self—employed occupaa ns identified were: tailoring, petty—trading, crafts luding weaving and manufacturing of baskets, farm imple- ts (such as hoes) by blacksmith, goldsmithing and ing as a retainer to a local chief—-something that ides no direct pecuniary reward. Climate The sample areas have similar climatic conditions respect to the intensity and timing of rainfall during year. Two rainy seasons--April—Ju1y and September- :mber, are common to the selected sample areas. Because of the greater intensity and longer duration 'ainfall during the major season, March to August, the rr agriculture activities are concentrated in this period. short duration and lower level of rainfall during the r season (September—November) render this period less able for crop production. Maize is the only crop of rtance which is grown during the minor season. Its d is, however, lower than for major season yield. Contrasting Features Soils The real differences betWeen the sample areas are soils egetation, farm sizes, crops predominantly grown, ‘ Table 3.] Sam} Ejura—Asl Kintampo- Hohoe-Vol Swedru-Cr Koforiduz *Station: Source: although ates in . the pres and grou Volta ar Somanya Region) . maize pr but less varying between farms in soil cha the pro 31 a 3.1. Rainfall Profile in Project Areas in Inches. Sample Area Annual Average Average Minor (Major Season) Season April-July Sept.—Nov. 1—Ashanti 56.64 25.74 20.1 impo-Brong—Ahafo 61.68* 28.76 21.98 a—Volta 65.12 28.39 22.41 :u—Central 51.61 26.64 13.57 'idua—Eastern 57.12* 25.5 17.51 :ions adjoining or in fringes of sample areas. e: Ministry of Agriculture, 22- cit., 1971, Annex Table 2. rugh the major soil type, savanna ochrosols, predomin- ‘in all the project areas, there are varying amounts of resence of integrades of other soils, such as ochrosols round water laterites in Ashanti, Brong-Ahafo and the areas; and lethosols and ochrosols/lithosols in ya area, Eastern Region and in Swedru area (Central n). Ochrosols are considered quite satisfactory for production, while the integrades are satisfactory, ess desirable. The presence of these integrades in 1g amounts reveals sharp differences in soil fertility an the regions, thus justifying the aggregation of in each area where there are fairly homogeneous :haracteristics. Vegetation An added factor that provides a distinction between :oject areas is that of vegetation and associated topograph accessibi land clea imvestmen investmen one regio vegetatio provide a similar v Thr Eastern r centers. densities land, has the remit six acres bankers a‘ 0f Agricu tribution 1 Brong 2. Ashan m Easte .4: Centr Ln Volta Since far 32 aphy of the land. The latter has something to do with ibility and the amenability of the land to mechanized learing operations. Land clearing constitutes a major ment in the entire farming operations. The total ment involved in clearing an acre of land varies from gion to another and is a function of the type of tion and to some extent the topography. These features e another justification for aggregating farms within r vegetational categories. Size of Farms hree of the study areas, viz. Volta, Central and n regions are closer to important food consuming s. This factor, coupled with higher population ies in these areas and less availability of arable mas resulted in less land cultivated per acre than in naining two regions-—Ashanti and Brong—Ahafo. Using res as the cut—off point—-the level below which the c are unwilling to lend to the farmers-—the Ministry .culture census data report the following size dis~ .on of holding six acres or less: rng-Ahafo 54 percent of the overall holdings .anti 58 percent of the overall holdings tern 75.4 percent of the overall holdings tral 78.8 percent of the overall holdings ta 82.8 percent of the overall holdings rms of different size categories face different A maize, grown i crops, each or depende decisio in the are hea two cro systems importa and pla‘ plantai contras diet of Ashanti who gro conspic Pattern relevan crops g 33 itutional arrangements and market conditions, they may er in their objective functions (Sen, 1966]. Predominant Crops and Consumption Patterns Although the six crOps covered in the study-~viz. e, cassava, plantains, cocoyams, yams and pepper-—are n in all the regions, mainly in mixtures of two to six 3, the regions differ in the frequency of occurence of crop in each region. Because of farm-household inter- ndence, home food requirements tend to dominate the sion of the farmers as to what crops to grow. Farmers he study areas in the Volta, Central and Eastern regions neavy consumers of maize and cassava products. The crops consequently dominate the decisions regarding ems in the regions. In addition, yams are next in r‘tance in the study area in the Volta region; cocoyam blantains next in importance in the Eastern region and ain next in importance in the Central region. In ast, yams, and plantain feature predominantly in the of the farmers in the study areas in Brong—Ahafo and ti regions. While, with the exception of few farmers row maize in pure stand, crop specializaton per se is icuously absent, the historical food consumption ns in the study areas and the available market for the nt crOps, are important determinants of the types of grown in each study area. It is expected that the linear 1 the rele each of Ti regions those ir obvious which ti COIlsumir Tl historir iZat10n farms se area cor reprESer Vi of dual terized inqudir bodies. 34 programming model used in this study will capture lative comparative advantage of the crops to grow in f the study areas.» Urban Influences The study areas in the Central, Eastern and Volta s are much more influenced by urbanization than are in the areas in Ashanti and Brong—Ahafo regions. The 3 effect of the urbanization factor is higher prices the producers in the areas located closer to important Lng centers receive for their produce. Erasers The differences in soils, farm sizes, cropping patterns, Lcal consumption patterns and the influence of urban- 1 provide the necessary justification for analyzing eparately in each region. Accordingly, geographic nstitutes the first stage in the construction of the ntative farms. Impligations of These Characteristics for The Selection of Representative Farms for the LP Model Introduction 'thin each geographic area there are certain features agriculture present. At one extreme it is charac- by a highly mechanized commercial enterprise, g large holdings owned by individuals or corporate such as the State Farms Corporation, Food Production Corpo of Ag tiona holdi input agric as im cultu farms it is sente each as 3E incm varyf hate Onts Etc. 35 oration and Settlement Farms Division of the Ministry griculture. At the other extreme is found the tradi— al, subsistence, unmechanized operation, involving small ings that make little or no use of modern agronomic s. In between, there is a third category-~transitional ulture, which makes limited use of modern inputs, such porved seeds, fertilizers, insecticides and recommended ral practices. The survey excluded the highly mechanized and included the other two categories of farming, and 3 around these technological categories that the repre- 1tive farms are selected. In addition to the farmers in category being homogeneous in certain attributes, such ;e, cosmopoliteness,2 functional literacy and historical 1e levels, the two operational categories identify .ng abilities of the farmers to expand or innovate. Classification of the Sample Farms by Technological Category Farms which used fertilizers, improved seeds and other ing materials as well as other chemicals and had had sion worker contact within 12 months of the study were ified as Category II farms. Farms that used mainly tional technology of production with no extension agent t were classified as Category I farms. The number of within each category are shown in Table 3.2. Cosmopoliteness is a communication term used to desig- he degree of exposure of a traditional person to the e world through travels, readings, personal contacts, Table 3 Regic hemp! Ashanti Volta Fasten Centre; tradi makin deter the 1 are s falli Table samp] and r tVEir 0me 36 Percentage of Farms by Technological Category in Sample Areas. Category I Percent of Category II Percent of (Number) Sample Farms (Number) Sample Farms 0 54 76 19 24 52 72 20 28 71 100 0 O 58 78 16 22 53 75 18 25 Representative Farm Characteristics e farms in each area were subdivided into two groups-— hal farms and transitional farms. Data from farms 9 these two technological categories were used to a the initial resource restrictions which define asentative farm situations. The resource constraints Lstical averages of the resources used by the farms in the technological category classification. In , the initial resource situation is based on the rta as shown. The resource levels of Brong-Ahafo rti sampling unit farms appear to be higher than rnterparts in other regions as reflected by acres rm labor by man equivalent (ME) and operating capital. classification of the sample farms into traditional itional farms (i.e., Category I and Category II spectively) was a post enumeration exercise. All y farms in the Volta region were Category I. fl __ .Num Lu. ALF. if Al \..I. l. \.l r. Salient Features of the Farm in Five Regions: 37 Brutus—Ahafo, Ashanti, Eastern, Cmtral and Volta. a Region L Brag-Mo Ashanti Eastern Central Volta I Trad [Improved [ Trad TIrrproved I Trad Thrpraved Trad Improved 'l‘cad 54 19 52 20 58 i 16 53 18 71 m .y)/ 6.79 15.52 2.82 3.33 2.27 3.22 3.79 1.46 2.0 ed/ 6.22 23.23 6.03 5.82 3.43 3.21 3.38 3.51 L)/ 6.675 17.34 3.75 4.20 2.95 3.22 3.58 2.41 2.0 3.0 5.77 2.0 2.0 2.0 2.5 2.5 2.0 1.75 a pm 4.08 5.028 2.97 2 93 3.01 3.09 2.68 2.4 2.689 a . 61 . 2899 .792 .696 1.02 .96 . 75 .9987 3.458 a 1.636 3.4495 1.26 1.44 .98 1.04 1.3346 1.0013 .289 : 112.0 344.0 48.0 54.0 11.86 20.76 32.75 26.9 6.0 r/ 16.78 19.84 12.8 12.85 4.02 6.44 9.12 11.16 7.69 I 125.19 250.0 107.8 87.0 69.63 125.02 103.65 74.61 27.0 14. 42 28.75 20. 7 23. 6 38. 82 28. 87 30. 95 35.86 .8231 1.03 .9 1.95 2.39 1.91 1.65 5.1.3 .08 .07 .09 .2 .2 .2 .2 .52 3.03 3.56 3.4 4.5 4.67 5.55 4.8 7.02 58.52 84.0 116.36 47.0 54.0 34.96 32.5 45.2 4.33 4.9 5.6 6.9 7.3 1.76 2.4 1.9 .943 1.55 1.55 .62 .7 .26 .22 1.27 60.0 62.0 60.0 65.0 68.0 70.0 65.0 80.0 fled from Survey Data. @ tic 38 Usg The 17.34 acres owned per farm in Category II (transi- l) was the highest average acreage reported in the study e 3.3). The Volta region reported the smallest acres d. By definition, own land implies the farmer operated mily land over which he had a usufuctuary or possessory [Ollenu, 1971]. As the table indicates, with the tion of the Volta region, a sizeable portion of the cultivated acreage is rented. What this implies is given the opportunity in terms of the availability re production resources, the farmers may expand the vated acreage. Labor Force The family is the source of the bulk of farm labor In terms of man equivalents (M.E.) the average size :m family in the study areas ranged from 5.028 (for )ry II representative farm in the Brong—Ahafo region) .ow figure of 2.4 for Category II farm in the Central r. The differences between the areas emerge further .abor use was compared on the basis of M.E. per crop There were consistent differences between Category Category II representative farms when using this f Comparison. In all the areas, also, the comparison n the basis of "cultivated acre" per M.E. shows that ry II farms used more labor force. Casg casu pern casu tion the was labo (mak area clea labc the Savj Oper rep} VaS Rte Win er 39 ual Labor Additional to family labor available on the farm, some ual labor was needed periodically to supplement the nanent family labor force (Table 3 3). The amount of ual labor in man—days recorded in the areas was a reflec— n on the cropping pattern predominant in a given area. In study areas in Ashanti and Brong—Ahafo regions where yam the most important crop, the busy seasons during which or demand increased were land clearing, land preparation king yam mounds) and harvesting.3 In the remaining as where cassava was next in importance to maize, land aring and cultivation were the periods during which time or demand peaked. n Capital Capital appeared to be the most limiting resource in study areas. Two main sources of capital were observed: ings and noninstitutional credit. In the aggregate, rating capital was the highest for the Category II fesentative farm in the Study area in Brong-Ahafo. It least in the Volta region (Table 3.3). However, on per a basis, Category II farms in the Eastern region reported highest figure. 3Despite the frequent occurence of maizS in mixtures other crops in these areas, maize was a secondary in terms of its contribution to the gross income. 4O topping Pattern Table 3.3 summarizes the main crops and the different rep—mixtures recorded during the survey. Acreages for ach type of crop mixture are also shown by technological ategories (Table 3 4). Maize was the only crop that was rown in a sole stand. Mixed-cropping is a type of horizontal or lateral crop iversification. Two fundamental factors——physica1 and ocio-economic considerations——interact to determine the ypes of crops and mixtures found. Among the physical actors are rainfall, vegetation, soil, temperature. These ,nteract with socio-economic factors——tradition, food eating abits, accessibility (geographic), land-labor availability nd relative prices. In the survey, open—ended questions were put to the armers to ascertain the reasons for practicing mixed- rOPping. The answers were coded in binary units (Table 3.5). n addition to the reason of security which was assumed way during the questioning, the reasons of tradition were onsistent in Ashanti, Brong-Ahafo, Volta and Eastern egions for the major season cropping. The Central region Bported shortage of labor as the major reason. For the [nor season cropping, farmers expressed the need, in idition to security factors, to maximize returns on Lmiting factors—-land and labor. 41 Table 3.4. Average Acreage of Different Crop " ‘ by T b ‘ l ” t a ' by Region. Enterprise Region Ashanti [ Btong—Ahafo Central 1 Eastern l Volta Trad* I]1mproved** [ Trad [71mptoved L Trad I Improved l Trad 1 Improved ‘ Trad M 6.0 5.86 7.9 8.6 2.75 3.7 2.96 5.32 1.13 MC 4.5 9.19 6.62 1.28 .63 MP 4.6 MY 2.46 38.3 MV 6 7 7.35 3.63 MCP 1.02 1.51 5.16 3 84 M00 .69 4.55 1.25 3.11 2.1 1.34 MCY 5. 08 3.38 4.47 60.08 3.85 1.5 1.13 .34 MCV 3.4 3.68 1.69 1.28 4.32 .55 MPO 1.06 MPY 20.7 HOV 0.46 MYV 6.21 7.57 MCPO . 1.46 11 3.45 1.08 .98 .91 MCPY 1.25 1.12 3.6 .53 MCPV 1.16 11.14 1.82 .85 2.24 .45 MCOY 6.6 6.02 1.88 1.46 1.2 MCOV 2.4 ’11.19 2.55 3.4 3.83 MCYV 4.64 3.51 5.25 10.1 56 MPOY MPOV 1 1.87 4.13 MOYV 3.81 8.13 MCPOY 1.6 6.25 8. 1.56 2.08 .79 MCPOV 1.28 1.0 14.73 8.86 7.34 4.5 4.13 .74 MCPYV 7.61 11.31 .91 4.71 .98 MCOYV 1.22 3.76 .42 OYV 9.0 7.79 2.34 3. MCPOYV 1.12 16.63 .75 4.6 .39 (16) *Trad - Traditional or Category 1 PM **Improved = Transitional or Category rm2 Farms. f9 K_x: M = Maize C - Cassava P = Plantain 0 - ocoyam Y V - am - Vegetable Source: Compiled from survey data. .ncouuaUIO van-4006mm 0... WCHUEOnnU! mhflEhnk we Mun—512 .ILQEIh 1:1 .1. Eric I: ((~ 42 .33 zmfiam Eouw vwaamaoo "wousom I/ w H N Anonfiwuuww : 3 5 0H 1» Q m omwwuofid mmfismwA «N n NH N ma 0H m o muonnwwv: eunuwnH «N m 0H m CH w.” «a ma 2 aowumnwfinu: Z S R can: 5538 m 3 Nm 2 3 na 36.? «N 3 ma Hm we Emuuwo one: N w «m an 0 NH NH ma mm m.» .3an we on: uqumuw MN an 6 n «N S 3 we m m 33 mo $335 3 mm mm 3 mm on ma R 3 ow usage #5me 3“ mm mm me on om 3 mm Z on sowuwumuu 3050b J I [I f! comma commwm commwm comwwm :omwwm acmmwm cowwwm :Ommmm cmwwwm cmwmwm no mam “05.: Ho mm: .352 no .92 ~95: no mm: .352 . ommfilmnoum Human: :uuummm muao> Hanucwo nonwum comma“ if 653350 3:0me 3 mcavcoammm uuwfiuwm mo uwnfisz "muwanwh 2.3 .3 550 3 $5333?! i. . 1. 43 op Yields The average yields of crops are depressed when grown L mixtures rather than grown in pure stands. Reason often .ted for the depressed yield are: 1) lower plant density ? individual crops and 2) competition for nutrients, space 1d light [Norman, 1973]. However, the depressed yields of Ldividual crops are overcompensated by the aggregate yield :r acre of all the crops. ”Sample plot yield” estimation procedure was employed ‘derive the crop yields. There were considerable yield fferences between the two technological categories of presentative farms. Category II farms which benefited om fertilizers and superior planting materials and cultural actices produced higher crop yields. With the exception maize, the economic rate of fertilizer application for a crops had not been established for the country. Crop zld responses for the Category II representative farms erefore varied from locality to locality because of the ferent levels of fertilizer applications used. ConcludingrRemarks Two types of representative farms——indigenous or [itional and transitional farms—-were identified for the y. Throughout the study, those farms will be referred 8 Category I and Category II farms, respectively. Several variables such as the age of the holder, his 'acy level, size of farm labor force, net worth size lding as indexed by acreage cultivated, are important 44 en defining a representative farm. But, in general, the gor employed in defining a representative farm depends on the purpose of a particular study [Ogunforowa, 1972]. cording to Ogunforowa, if the objective of the study is it the derivation of aggregate supply functions, but rather lidentify the direction of farm adjustment or expansion .th and/or to estimate responses to varying resource and ice levels in an area, a less rigorous method of bench— rk farm construction may be used. Such was the purpose this study. Despite the usefulness of representative farm approach permitting limited statistical aggregation, and opera— onal flexibility, there are potential problems of aggre— tion associated with its use. Specifically, sampling ror, specification error and stratification error may nit the usefulness of the representative farms in predic— lg farm adjustment [Heady, 1961; Day, 1963; Lee, 1966; ler, 1961; Lard, 1963]. Although no attempt was made to imate the size of these errors in this study, attempts e made through two—stage probability sampling and careful itification procedures to minimize the likely impact of :e errors . CHAPTER IV THE STRUCTURE OF THE LP MODELS FOR THE STUDY Introduction This study uses three interrelated phases to discuss 3 a subsistence smallholder farming in Ghana. ilysis takes place in Phases 1 to III. 1. problems associated with optimum resource organization The empirical The static linear programming phase: this phase focuses on the determination of what the repre- sentative farmers are doing. It deals with land, initial operating capital (cash available), labor, 1t consumption levels and borrowing as fixed. includes activities for mixed—cropping, labor (hiring and selling), sales, purchases, credit and land clearing. The poly—period linear programming phase: this phase allows cash flows, varying farm size and associated assets with no restirction put on the amount of credit the farmers could obtain, inter— dependence of production——consumption and invest— ment, and on—farm storage of farm products with due allowance made for storage losses or attrition It is an expanded version of the Phase I model. 45 —_v1 46 3. New technology phase: in addition to the histor- ical mixed-cropping enterprises embodied in the first two phases, this phase incorporates con— tinuous cropping (in pure—stands) of the six crops, viz. maize, cassava, plantain, cocoyam, yam and pepper. These enter the model as alternative cropping activities. Alternative technologies of producing these crops also feature in the model. The technological coefficients for the various cropping activities were derived from experimental data [Hudson, 1972]. This model expands the Phase 11 model to include alternative technologies of producing these crops. The policy analysis simulates several policy outcomes ed on varying resources and activity levels for Phases 1 II models. The linear programming models for the repre— Zative farms for the first three phases are discussed :he subsequent sections of this chapter. Under three lings: l) the objective function, 2) the constraint lcture and, 3) the activity set. A theoretical presenta— . of these will be given first to be followed by the vant empirical presentation. Phase I-—The Intra—Firm Linear Programming Model The Objective Function The linear objective function used in the study can afined as: 4—————————————----III-IllllllIl-Iiiilillr‘ 47 Max” = XV (Zisi‘) + z ZLL — Eu 21.313. - 2k ziI. j=1 J Y Y Y j=l J J j=l J J H v f - Z Z H — Z C.S. Y Y j=l J J subject to land, capital, labor and miscellaneous constraints, where: n = Net returns (profits) to fixed inputs. Cj = Average buying price of the jth commodity. S? = Actual level of jth food buying activity. Z; = Average selling price of jth output. S: = Actual level of jth selling activity. Z? = Opportunity cost of family labor per period y LY = Actual level of family labor hired out in hours per period y. Zj = Per unit cash cost of jth variable input. P. = Actual level of jth variable input purchasing J activity. Zi = The current cost of the jth quasi-fixed input J computedby the payback principle: i.e., for the jth investment Where T is the useable life, a, the rate of interest on loans to farmers by the banks, Cj is the acquisition cost. I- = The level of the jth quasi-fixed input. J Z? = Cost per hour of hired labor during period y. H = Actual amount of hired labor in hours in period y. deI 48 Maximization of net farm income subject to the satis— ction of household food consumption requirements is the erational goal of the farmers used in the model. It is "constrained” type of profit maximization [Day, 1962]. In bsistence farming of the type covered in this study, the ovision of food for the members of the farm household generally given top priority. Norman [1972] refers to is type of goal seeking as security and profit maximization. The Activity Set The activity set facing the representative farm is l noted by: i l’- . o,Pu, Ou+l,. ,Ok, HI, ,H’Y’ LI’ ’LY, “ m m —m —m I). o -,V_Y, QI" . .,Qm, Qm+l,. - .,Qp , C1,. . -,Cg ‘ I" . “’Sg’ FI’ ,Ff, Lm, LN, B} are: -- - "Pu = Activities involving the purchase of the variable inputs. -l" ,0k = Activities involving the purchase of quasi— fixed inputs. -,H = Activities to allocate hired labor to y farm Y operations or periods. ,LY = Activities to sell family labor in y period. .,V = Activities to allocate overhead labor in Y y periods. .,Q$ = Production activities involving pure-stand cropping in major season. 1:- - -,Q_m = Production activities involving mixed p cropping in major season. act and and and SCQ sec figs U89 1131 49 i, . .,C = Activities involving household consumption g of subsistence crops. 1,. ,3V = Activities involving the sale of V outputs for cash. ,. ,F = Activities involving the purchase of f I f . . outputs for domestic consumption. 4 = Activity associated with additional land clearing in major season. q = Activity associated with additional land clearing in minor season. i = Production of maize in the minor period. = The activity associated with the farm—firm's net borrowings. 1 4 . .each of the five regions is engaged in 1) production It is assumed in the study that the representative farm tivities; 2) input purchasing activities (both variable d quasi—fixed input); 3) labor activities (hiring, selling 1 overhead); 4) financial activities; 5) land activities; food purchasing activities; 7) consumption activities; I 8) supply of storage activities. The relevance and pe of these activities are discussed in detail in the tions that follow, using one region as an illustration. In addition, the following assumptions were made for model in this study: 1) input—output coefficients in— red are consistent with the farmer's cultural practices available technologies;2 2) the government input subsidy 1This refers to land additional to the piece of land ned already cleared for farming in the current season. 2The Phase III, however, the technological coefficients in the LP matrix were derived from recommended practices ; improved technology of production. and 0 their follo consi is El diffi by th banks 50 d output expansion programs will remain in essentially eir pre—l972 coup forms; 3) the problem of inflation llowing the 1971 devaluation of the currency is not nsidered; 4) the land tenure system in the study areas flexible enough to effect acreage expansion without fficulty; 5) the current methods of financing the farmers the Agricultural Development Bank and the commercial nks will continue. Phase I of the model is stirctly concerned with static near programming allocation. Category I and Category II resentative farms are considered representing traditional 1 transitional farms, respectively. The submatrices for 1 e Brong—Ahafo region are presented in Tables 4.1 to 4.7. 4 3 structures are the same for all the regions, so they a not repeated here. '2 Activities The crop and crop transaction activities included in Category II in Brong—Ahafo region, used as an example, maize, cassava, plantains, cocoyams, yams and vegetables )per). They enter the model in the various mixed—crop arprises. Maize is the only crop in pure-stand and is only minor season crop (MZN) (Table 4.1). In all the es, a negative sign in front of a coefficient indicates mg to the resource and positive signs indicates using resource . ...G!uw.v! ahfiiaxluwibhn INN NkOIMu-fid u. :II.?~ how Iiiuw>iuu< lahU .WIV Indira r .. 1111171114 .. u N m q a m n o o o.amm- Hon. 0 u o o.~Hl o o o u o o a.mHu m.mH- o u o omH- omH- o o n o H.~H- «H.aHI m.mHl o n «o. I NHm. I «Hm. I cm. I o n s.o| HH.mn H<.ml oH.n‘ o n H H H H o n N N N N O H N N N N mu 0 u 4 NH OH OH 0 n H H H H o n o m.H o.~ o.N o u o o.~ o o o u o o m.H ~.~ o n o m.H N o o u o H~.H o~.H w.H o M mH m.¢H nH mH came w mHN HHHH 4mm own mom N no mmH mMH oh m¢OH w. HNH cm as ooH HNm w. as oo om ow mHn w. o as «a «a stH w o mom mmw ”4H Hnn W o wmm «a ma oon w. o mm mm mm He.m w. H o o o am.HH v H H H H flulunlr o o o o Tullnullu||.||||:» mam amHm 2N: >smuz >om5: a. ma. wv. eHmI NHI mHI 0° 00¢ m.m~| oomwl a.mHl mm.ml OOHONN—l I-I o m.¢Hl OD mue. I m~.ol MCHONNH O CON .muwv sw>n=m Houw kusaaou "mopaom mymmh cmsum> wmsz<> omswso HmsNZ «ACNE mmx< mmom AH<2 HonHHHuumu ham mN oHnmuwwob unnuao mN Ema unnuao nu fimhooou usmuao ow oHMuomHm unnuso mu m>mmmmo uamuao «N A>v wnHma uanuso mN Any onwmfi unnuao Nu wxw ham HN .3: an 8 umfioudfl bum aH uuxmoa mom ma Huwmfinu ham NH manwuowo> mam ca 5% ham MH amhouoo ham «H .DHNUGMHQ bum NH m>wmmmo ham NH wNHmarvowm ham HH Honda Hmooa< 0H m powwow Henna me a uoHuom noan we m voHuma Hanna mo q vowuon unan co m vowuoa wonou mo N uoHnum uoamu so H VOHuom uonmq mo vomH common mafia: No vawH common mono: Ho o>Hu00nao H H H .-.04uw..~. Quin—(13:01.: IEhii MN (ALONEMCU ulbfiuw>qdfl< u«lii.~..uuln~ Mills . 000 52 GO GO 11 damn zw>u=m Eouw wouamaoo "wousom 1/1 . . . . . . mH 8. =20 H338 mm. mm. m. mm. am m N m o m o m no u onumHomo «H mm“? UHEH moon» hon MH H- hoe “HE. $2 as H H- H: .52 fins 32% name b3 HH his in: mem H- .36 as OH H: 553 ”in: muoxmmn .25 a H: mfimmm won .HoNHH IHuHom has w H I mug pH onnwu Iowa? .3; m H1 “Hz: 00H mam» man m H 1 “Hwoo 03.. mag» Iooou b.3— .m H 1 53m swan 53 IomHn .55 q H: mmad 00H 5838 55 m H: mm: H 3.35 room AHE— N Nmf 2.: mi an... qwr wdt o.m.. o.mv o.m: we... Wm: 00... was we coyofifl o>HuuonAo H Hmom mfim Mm mamxnewhououncHMuamHa m>mwmmo wNHmE 30¢ 5:34)) :1. >50. >50. hag .Aan has .35 h:£.1b.5 hflo 5).. . m:m _ mmdd_m_wH1H_ Vii—.1; rev; «(fin—.41— mrl. mm‘;3015;EAL:‘dawn—Ludwlmflx‘m:Tim—.73.}:Lm3wd1mmn$m\n¢‘mm\nav‘mwNnmw\nmm\nmm\ 30m..\ . 1 \ . Kick .l.:~.~<1usilhfl .wN \«LOuubuiU ”ZU#UN>NUU< 100.14 4.1:» hdéfih . 1 1111 -11111111J .mumu $5.5m Eouw wousnaoo "wouaom . . . . . :95 v moHumuomo - mm. mm. 3 8 S 8 E 323 .. Hausa. mm saw v H H H H H H H 5953 _. H Hanan HN HMHHCH m H omommfi .. o EH2 «N M3 m H £955 _. m 8H5 NH mm H H «mommj : .H @3qu HN S n H 28:3 .. m Horse ON on u H H Hmong ._ N SHEA 3 mm H H Hmong : H voice H 3 EH n .Honmu vow—+35 5 H H H H H H H fimomnj : Hounds H man w H E935 __ E H be ~0an oH mmN m H 3955 : om w Ho .Hoan mH mmN v H whommj : mm u «no .HoHHMH «H SH m H :83: _. E H to “23 HH ”MM w H «Home: .. 2 H to 83 NH oNH w H whommwuu< nuwnm Inc :uuuaaanuou Jilin vant 5.1 uwnuh 111.111.4111 1 111111.1 J 54 .mro zo>h=m Baum owuomaoo "condom o.osM H umsu mnH «Hauuumm> mH mam. M zHHo mnH oOH as» «H nn.H M soon mnH ONH amhouou mH Nn.oHM «Hen Human aHauaaHN NH no.n M mace mHH ONH apummuu HH mo. M >¢xo mnH oNN Hwa>v oNHm: OH HHNN. M H axe mpH ONN Hanev ouHIx mo “OH UNEMAHOU o I H H moum> mnH unnuao anmuowm> mo 0 I H- H moxNz mnH ONN Hsv unnuao oNHmz no 0 I H H moNz nnH oNN HOV unsung «NHmz No NH. o.nn m.m ma. o.n m.m OH. o.Nn H.n no. n.n m.a n.m o o H aoHuunse mum um>n ==n _HDJn_Huuunno Ho ufiuz .oz .EDkuJH aun...<1.m:3hn IWEMNE NH hueuwuflo mmfiue>wuu< NINEQ£LL.& 12::h 1:: Tin. v .\ 1.111 55 3.2% .mumv hm>nnw Eonw kuaaaoo "monnom H Boxwom u some woBOHHOQ n omNIVImm. mm. mm. 0. mm. m. n. NM. mm. m. mm. 1%. w.N m. o.m o.m co. m.~ 90. H1 ¢.HH w.m~ H.540 u Hmanm... wcHumuomo a cam H Nzazfi 6.5m Honda Ivde wanna: m a.m-Hw H Nzazfi 8% 8.35 IpflMH venous q 26W H1 Hing ouom HHHHM 35H so new m 3an M H1 H2924 whom HEMH mm.1mm.1mw.1 or 3.1 0.1 5.1 Nm.1 3.1 m.1 mmr qu w.N1 common .3me N m.1 o.m1 o.m1 wor mNI 8.1 00.1 m.HH1 w.mN1 u nOHuoaaw w>HuuoEo H 55 can 55 $5 95 Nmm Hum <15 44H: Sm 5m m 5?“.— Voom ¢IHanH mdv N uch mwfimz EEHHOU .oz 334 .nUEfiN-NUOEOU DEW DNNUu-KrHUUd‘ HWCONUNUDH‘ -h -V ”HAW-H I ... .. . .- .I.I.rI..Il... .1 1 1 I 57 dune hog gum Bung "meadow ENS? .meMOHoHEomu o>HuwFHouHm 95 on“. IE voodoonm wNHofi newoow Hood: some mg Ham N N I N mama; oUHoHoooB mam H H H zm 539.6 96 mHHn Houmao N H z: EoumHHo QC 9:: Ecumso H .1ng *NENSHH N H mg NHuocnoUHw 95 on“. .3 unease mNHmfi acumen HoaHB :moa mg was N233? o n N I - . . N N I H1 mHanmk mom wuHoHUowS ham m o u H . . . . . H H H H H H- zm 539.6 9.5 CC 3.3 839.6 N o u . . . . . H H H H H- z: Eoumvo 95 96 oan 5336 H Sign; 44.3."; . . . . . * will: _-:IIII- _ IIIIII _ 11 1 _ _ _ _ .Oz $2 $2.32 xNgaz III. .1t-4‘l:_.1€:‘t(:_ C ..... _ ..n5m naum vmydwid .meMOHocnumu w>HuMCHmuHm ozu wzu an uwusuoum wNHwE commww nocHa cums mz¢Nz van N mama: $3333 .3 zm BoumSQ z: Bouwao *mz H~ . . . . ._ mo mcmm. o.m N mm.w I o.w mmw. mN.m m mm.n I o.m HMnm. m.m N mm.o I 0.0 wmqm. mo.nH o mm.m I o.m wmmm. m.mm m mm.q I o.¢ coon. mm.HH w mm.m I o.m MMHm. no.m~ OH mm.~ I o.N mmoa. mm.@N mH mm.H I o.H camm. mn.¢~ mH mm.o I H.o l/L I/L OHH¢M Amousm Eouw vonafiou "wuuaom Has. one 8.: 09% N 8.0 HMso mmmm. 86 8.8 0.8 m 86 I 06 38. 3.2 8.3 ode e 36 I o.n 32. mag 2.2 3.: a $5 I 0..» 8:. 8.: S? 8.3 o 86 I o.m 23. 2.3 3.9“ of? w ma.~ I H; 38. «a.m.» 33 832 an 34 I 9H 3%. 3.2 3.8 0.? 2 mad I do SE 335 283528 63.3 33% Amu Em.» u N. 593000 I o afiauflmqm n m m>mmmmu u 0 Sam: n z alummfl N6 3.3 Nmé 3.: 0 N wanna.“ noon.“ H33 ~32 so»? we 033. mmh QMHH .3: oinmw 00m mu: mvu<\mhz hen—NH wuu4 >~r0m2 S.3 2.3 3.3 fie 3;: u Hauasmififié $95 o.~ S. 32.. :8: mm. 93. n. on. em. v “Eunice $9.6 ea 33 26.8: qémw Cdom Nfinm m.mo~ 36mm u whom\c.~=uwm mmono nm.~ wuu< women m :5: 36:: «magma a .32 m .3: u 2:85 mmouu 0..» 36¢ >8: mfm E3. s8: m4 flue 8: 0.0m 0.0m 0.mo 0&0 0.00 a: uwmemm o.~ mm...” NJ 95¢ Auommwm H0393 wNHNz RA 3. 3. mm; is. 03 a...» a; 3; as me Se 3 oi 3308 ming ”.3 34% 9.3 0;; 2.3 625 5.35: 8; Run. 0..» and 85 fl o8 2533 0.3 2: muEmuowg b5 26 24 mm; 84 2». fl 2N Sam: oéw 55m 5388 as a .2 m .3 m .5 m .8 II. a Eton aoflmasmaoo mama cam: .1 M32 mfl 5&3 Sum a.ma .1: 5: «an 2. 03 «as flow 3: 3.3 a.mm 3.3 3.3: u sanction TS $.Nn #3 a.mm 0.9: fl o3 wuaémsouoo Sum 93 3.8a 3.3 9:: ads u ammo can». an.» 0&3 55a 53%: :3 £4 1! o.~ NA o.~ Sue «€25 3.3 Tom 9% «an «.2 fl 8N machines :3 32 3mm 33 03m 3% ~23 135:. m; e. m. m4 a; fi EN 330 Sam: due 3s A? man Sm anon 32 “can; banqunhuefiumn ”.3 0.0 a; «.2 13 3 gm Abs «Nam: :3 2m mom En De gm 32 Roam.— .BzLBSoo o o o o o E: a €23 “35 SE n: N2 in new m3 93a :53 “233% o o o as o E: 0 135m “33 SE 8m NR «2 Sn mam 86a MES 3:92 an mom 8 m3 0 E: N Bah; H23 32 one 22 33 3m 33 32 Sad 33.23.32 0 o 9: SH SH 3m a Bios ~23 Saw a? «3 as” can a 32 ~23 H23 9 o 9: EN 3: 9 SEE ~23 :wm 0.2m a.mkw oémm 0.3m or: $2 p33 nonmalimismm S a: 2 m3 NS 3m n scram ~23 2% ed wma mad $6 3.0 3% 99:3 8 3 NE my: a Bios ~33 2% «3 oz 2N 8n NS mum m “3E3 ~33 2% 359538 m3 9: SN 23 my: a mafia» ~33 Sum NuHo> kuuawo Emummm Huamsg owmfiw ”is: Euummu Hmuuaoo muHo> Huang Own—3 nwaoum rmuoum 5:: 330.533 .MBINRH mamso .mcoawox .22. .mfihfim H rowwuau you mdoaumufiawwno Hun—Hugo run we mowumwuouumumg Jim p.33. 94 only sole crop raised. Brong—Ahafo had six crops with maizevcocoyam-cassava mixtures being allocated 1.5 acres and 4.6 acres for maize—plantain-cocoyam—pepper mixtures. All household food conSumption requirements were met from home production. The crop sales figures given in the "activities” section of Table 5.1 were the net after con— sumption withdrawals. Category I farms in the Ashanti region were diversified also with five crops-—maize in the minor season (1.2 acres) and maize—cassava-cocoyam-yam mixtures (3.75 acres). Since plantain did not appear in the cropping plan, 84 units of it were purchased to satisfy the consumption constraint. Without a food purchasing activity in the model, the plan would have forced in a cropping activity that included plantain. The competitive position of cropping activities not included in the optimum 2 plan is shown by the size of their shadow prices. The marginal costs associated with the nearest competitive 2Generally, the simplex procedure continually strives to find a better operating strategy than the one currently at hand. It does so by reviewing at each step of the sol— ution the marginal cost reduction or profit potential of all the activities that are not in the current solution. The simplex procedure determines what activities to include to improve solutions and ultimately, to find optimum solution. It is on the basis of this activity selection that the pro— cedures finally stops and determines that an optimum solu- tion has been found. The procedure also determines in each Stage the rate at which income will decrease if it were to introduce unprofitable activities into the solution. There are many names given for this unprofitableness of each of the activities that are not part of the solution: Marginal cost, shadow price, opportunity costs, multipliers, n values and dual variables. Reference here emphasizes the enter— prise's competitive position under different cropping patterns 95 enterprise involving plantain not in the solution space was ¢107.7. Therefore, to force one unit of plantain activity into the solution would have reduced the optimum income by ¢107.7 per unit. Similar reasoning applies to Category I farms in the Volta region which had diversified cropping with maize, cassava, cocoyam, yam and vegetables, but excluding plantain. The central region also had 3.6 acres allocated to maize—cassava—plantain-cocoyam—yam—vegetable mixtures. All food consumption requirements were satisfied. In the Eastern region, 2.0 acres were allocated for maize (sole crop) and 2.95 acres to maize—cassava—plantain- cocoyameyam mixtures. The shadow prices associated with maize as a pure- stand crop in the major season were ¢241.0 per unit (Ashanti region), ¢534.00 per unit (Eastern region), c516 per unit (Volta region), ¢462.0 per unit (Central region) and ¢58.0 per unit (Brong—Ahafo region). In addition to the several noneconomic arguments that could be used to justify crop— mixtures, the shadow prices quoted above show that from purely economic standpoints, it is "too expensive" for the farmers to raise maize as a pure-stand crop on a major season land. The quoted shadow prices show the degree of "unprofitableness” of maize as a pure-stand crop in the major season. and levels of resources. The higher the shadow price, the lower the competitive position in both the current and alternative optimum plan [Driebeek, 1969, pp. 103+]. - V 96 Table 5.2 shows for Category II or "transitional" farms in the study areas (Ashanti, Brong—Ahafo, Central and Eastern regions of Ghana) the gross income, family consumption withdrawals and the extent of disinvestment of family labor. In the Ashanti region, the crop plan was diversified with maize—cassava—plantain—cocoyam—yam enterprise allocated 3.8 acres. The shadow prices associated with the excluded cropping activities, maize (major season), maize—cassava— plantain-cocoyam—vegetables, maize—cassava-plantain—cocoyam and maize—yams were ¢70.4, ¢5-03, ¢ll.05 and ¢54.34, respectively. 0f the cropping activities mentioned above, maize (major season) appeared the least favored in an alternative crop plan, followed by maize-yam enterprise. Maize, as a pure-stand crop in the major season land, appeared to be ¢70.4 too expensive per unit to be forced into the optimal program. The programmed crop plan for Category II farms in Brong—Ahafo region selected all the six crops covered in the study. The marginal costs associated with the nearest competitive cropping enterprises were: ¢6.76 (minor season) maize in pure-stand; ¢20.09 (maize-cassava—plantain-yam vegetable mixtures); £34.06 (maize—yam mixtures). The shadow price for major season maize in pure-stand was £194.38 which shows that this cropping enterprise was the most expensive to be forced into the program. emuamaou "ouusom so maemo.u ‘ Ayudouu vummmwm n > any n v awhouoo u A 53.3: n m m>mmmmu u u ouflmz u S mmmmmm f}'/,/ 0.0 H . . I.I q «N w me 0 RH N muaaaw noan amuou >womoz «mm . . Hanna kuH: we oaumz : womb: m0.mH NM.MM@ M.MHOH .OHHH n: wuo<\mum Honda ~.m m.m : omoz Nm. mmw. m . mw.m s Hmufimm0\:u=uom macho m.m : >M02 mm.HHn w.anc W. me. u pm\nwauvm mmouo m.~ : >02 m.NeoN m.wN0N a.mwwm m~.cqm s onu¢\cu=uom mwouu N. : *0: .IIIIILIIIIMIIIIIvIIIIIIIILIIIIIIIIIIFIMIWWWWI a msounH mwouu o. m.m a. H nommom uoaazv exam: ”.m0 N .mw ouoo 0.00 mp4 moafimuwmw> N o< Mmmmmmmmmm m.n M.N mca. mmHH 00H Emw Bouuom o.sm m.~m NM.“ s.m saowa amsouou a.ms m.soa mo.eqm a.mNH u an unumm> aamm w.q m.¢ m . q om.oHH annum nwmuumflm a.me mmNN mnH an Edy Hamw mm.~ ms.a o.m a m afloaa m>mmmmu a.ma a.s~ q.~s e.sm ooa x mm m~m ca. naowu «New: s.ok m.wam N.os msHONH am coco Ha ~.mmm a.mmma Nam sauna namuamam Hamm mmmmmmmmmmm s.km N.smH s.oma a.ms naoou m>Mmmmo Hawm nm.mk mm.qoa mo. . q.s~ «.am a.moa N.o~ AHONN “was: wwam 7/ 0.m~a ac.¢n ocm 0.MNH wqwsouuom 0am 0a ooh : m venuum H nag m AH, o.~ onN o as s mmao was = o uoaumm “span mu.m -.m . Ix- . In: In- a poaazv can; maoa : n aoauwm gonna swam am a am he ma.m ea “cameo sums ma Nmm = a soanma gonna unam owcm mmNN cmmw Naqm : henna Hmnuam meme : m cease» “coma swam wok ems mam Han : Hanna sumaamnuowo mam mms smmN “Na : N aoauum roams swam own cum wqoa moo : Honda .>oz«umnouoo mma : A voaumm Honda muwm sea mma “mm saa = roams umaawuamm «ma om : a soauum “cams Hawm moa «RN mam NHN = roams um=m=< me am mma : s aoaumm gonna Hams com wee aqua mesa : nonwa sannlusselsmx Na Nea sma = m uoaumm poems Adam Nms smN Hmm «mm = “as“; Hanna oa mm = q eoauwm Henna Hawm mow 00m 000A Hmm mum nonmq nuuNZ|xum=~num mm 0N new : m wowgmm Honda Hfimm : N uowuwm wean; HHum muCHmnumcoo mm ma oHH uzom A vowhmm Hanna HHmm ommnm ommz< auwummm annuaou Iwaoum «uam:m< own: cuwummm Houucmu (mnoum fluqmnm< own: wwwuH>Huo< .mnowwwm cumummm cam amuucmu .ouwnelwaoum .Hacmsm< maumm N shamanwu we maoaumunndwno HmEauao was we muaumwumuuwumnu .N|m manna 98 In the Central region, the cropping plan was also diversified with_four crops~-maize, cassava, yams and vegetables. The program allocated 3.9 acres to the maize- cassava-yam mixtures and only 0.2 acres to the maize-cassava— vegetable mixtures. A comparison of the shadow prices of the excluded cropping activities shows maize in the major season as a pure-stand crop was the next favored enter- prise which could be considered. Its marginal cost was ¢60.66 as compared with ¢99.2-—minor season maize in pure- stand and c212.02--maizeecassava-p1antain-cocoyam-vegetable mixtures. In the Eastern region, 3.2 acres were allocated to the maize-cassava—plantain-cocoyam—yam mixture and 2.0 to the production of maize in the minor season as a pure- stand crop. Maize, as a pure-stand crop, was the most ~expensive cropping enterprise to be considered for inclusion in the optimal program. It had a shadow price of ¢664.7 associated with it. The next favored cropping enterprise which could be forced into the solution was maize—cassava— cocoyam—yam mixtures. It had a marginal cost of only £15.0 associated with it. The possibilities of varying family labor (as a fixed asset) were considered through the use of the asset—fixtiy theory [Johnson, 1970 and Clark Edwards, 1959]. The theory states that an asset or resource becomes fixed when the following condition is met: 99 ancquisition 2 MVPX 2 szalvage. where x is the resource, PX acquisition is the purchase price of acquiring one more unit of the resource, MVPx is the margional value product of x in production and PK salvage is the disposal or sale price. When the MVPX is less the salvage price, it is profitable to use less of x (or sell x by varying its quantity downward). When the MVP of x is greater than its acquisition price, it is profitable to acquire more of x (or the quantity of x varies upward).3 In Brong—Ahafo, for instance, no labor hiring activi- ties took place in any of the periods among Category II farms in that region. An examination of Table 5.3 shows that the MVPs were consistently higher than corresponding acquisition prices per unit of labor. However, among the Category I farms, there were substantial sales of family labor (Table 5.1), the MVPs of seasonal labor again indica— ting when it was profitable to sell family labor (Table 5.4a). In Ashanti, labor selling activities took place in all the seasons for both categories of farms, except period 2 in the case of Category II farms and periods 2 and 6 in the case of Category I farms. These were periods when the MVPs of seasonal labor were higher than its acquisition cost. Similarly, for both categories of farms in the Eastern and 3Here and in subsequent discussion, the MVP's should be interpreted with the caveat previously raised in mind. The holds for one unit of a particular resource and its behavior be erratic for further additional units of the resource. 100 Table 5.3. Comparison of MVPs, Salvage Values and Acquisition Cost of Labor by Region, Ghana, 1972—73, Category II Farms (Cedis (é))- - Region Periods P1 P2 1 P3 P4 P5 I P6 [ I’7 MVPs in Cedis Per Hour Brong—Ahafo .13 .114 .114 .123 .123 .102 .102 Ashanti 0.0 1.2 .06 0.0 0 0 .06 0.0 Eastern 0.0 .13 0.0 0.0 0.0 0 0 .14 Central 0.0 .1 0.0 .3 O 0 0.0 1.2 Salvage Values Per Hour Brong—Ahafo .06 .06 .06 .06 .06 .06 .06 Ashanti .06 .06 .06 .06 .06 .06 .06 Eastern 1 .0675 .0675 .0675 .0675 .0675 .0675 1.0 Central .0675 .0675 .0675 .0675 .0675 .0675 .0675 Acquisition Cost Per Hour Brong—Ahafo .0875 .075 .075 .0812 .0812 .0675 .0675 Ashanti .1125 .09 .106 .0938 .10 .0938 .0938 Eastern .106 .088 .125 .094 .106 .125 .094 Central .106 .088 .125 .094 .106 .125 .094 Source: The MVPs were derived from the LP solutions. The salvage values or the opportunity cost of family labor were the labor selling prices used as the objective coefficients in the LP The PX acquisition values given here refer to the cost of hiring a unit of labor used in the objective functions of the LP. __/4 lOl Table 5.43. MVP of Resources: Category I Farms by Region (Phase I). Resource Ashanti Brong— Eastern Volta .Central Ahafo l Marginal Value Products in Cedis Cash Expense .73 10.37 .009 .06 .09 Land (Major) 255.6 0 604.6 565.4 520.3 Land (Minor) 0.80 0 0 49.2 0 Planting Materials Maize .10 .68 .57 .57 .06 Cassava 4.3 28.4 23.66 3.01 2.5 Plantain 1.9 .68 .56 .11 .06 Cocoyam 5.2 34.11 28.4 47.58 3.0 Yam 8.67 51.96 47.3 4.0 5.0 Pepper 2.46 .93 4.7 4.8 .5 Implements Matchetes 1.45 7.30 7.95 8.00 .841 Hoes .67 4.44 3.69 3.7 .39 Axes 1.37 8.98 7.48 7.5 .79 Chissels .55 3.4 3.03 3.05 .32 Baskets .52 3.6 2.84 2.85 .30 Labor By Periods Period 1 .06 Period 2 .25 .35 .13 .13 .13 Period 3 .06 .06 Period 4 .06 .59 .06 Period 5 .06 Period 6 .25 Period 7 .06 Source: Computed 102 Central regions, disinvestment in family labor occurred in periods whose MVPs exceeded the acquisition cost per unit of labor. It is evident from Tables 5.2 that the gross returns per unit of individual resources were high. In the Central region, cash expense, as a resource, had a very low MVP (¢.0001) indicating that it was virtually not restricting. However, in Table 5.3, it is shown that the gross return per unit of capital (i.e., its average return) was 13 times as high as its MVP. As compared with Brong—Ahafo and Eastern regions, the MVP per unit of land in Central and Ashanti regions were relatively low. However, the MVPs of other farm inputs for Category II farms as demonstrated in Table 5.4b for Ashanti and Central regions would point to the high earning power of these inputs. In these two regions, the need for expanding the use of these inputs appears to be clearly demonstrated by the magnitudes of the respective MVPs which were consistently above their respective marginal factor cost. In Brong-Ahafo and Ashanti regions, the MVPs of land (major season) would indicate that greater income gain can be achieved by bringing more major season land into cultivation. Table 5.4a contains the MVPs of resources used on Category I farms in all the five regions. In Brong—Ahafo region, both major season land and minor season land are not a limitation as shown by zero marginal product. However, ____, 103 TABLE 5-4b. MVPs of Resources: Category 2 Farms by Region. Phase 1 Resource Unit Ashanti Brong— [Eastern Central Ahafo l l ----------- Marginal Value Produce (c) Cash Expense Cedi 7.78 1.81 .0001 7.32 Land (Major Season) Acre 2.41 201.35 744 2.01 Land (Minor Season) " 0 0 62.5 0 Planting Materials Maize Lb .52 .17 .06 .50 Cassava 100 21.96 7.01 2.5 20.81 Plantain Unit .53 .17 .06 .50 Cocoyam 100 26.35 8.42 3.0 24.97 Yam 100 43.92 14.03 5.0 41.6 Pepper Lb 20.0 5.6 .5 4.2 Other Inputs Fertilizers Lb 24.6 7.86 2.8 23.3 Matchetes Single 7.38 2.36 .84 6.99 Hoes " 3.43 1.09 .39 3.25 Axes " 6.94 2.21 .79 6.57 Chissels " 2.8 .897 .32 2.66 Baskets " 2.64 .84 .30 2.49 Labor By Periods Hour Period 1 " O .13 0 0 Period 2 " 1.2 .114 .13 1.1 Period 3 ” .06 .114 0 0 Period 4 " 0 .123 0 .3 ' Period 5 " o .123 0 1 Period 6 " .06 .102 0 0 I Period 7 " o .102 .11. 1.2 i Source: Computed . . 1 V.‘ IleI '(1 \.1 IIIIIIIIIIIII::::——————————————————————————————————————————————————————————~44i::: 104 when compared with the acquisition prices of the other farm inputs, the earning power of the restrictive inputs would appear high enough to warrant an increased use of an addi— tional unit of each of the resources. Comparison of Results of Phase I with Observed Sample Data: Category I and Category II Farms The programmed results of Phase I, Category I and Category II farms are shown in Tables 5.4c and 5.4d. They are discussed on a region by region basis. The actual crop plans are given in Table 3.3, Chapter III, which will be referred to often to facilitate comparison. Category I Farms In Brong-Ahafo region, the optimum gross income came to él798.3 as against ¢1206.4 from the actual average for the representative farm in the sample. This represents an increase of ¢591.9 or 49 percent. The actual crop plan had pure—stand maize in the minor season whereas the programmed crop plan did not. The program used 6.1 acres of major season land, .7 acres less than the amount available. In Ashanti region, the programmed crop plan for Category I farms in Ashanti allocated 3.75 acres for maize-cassava— cocoyam-yam enterprise in the major season and 1.2 acres for pure-stand maize in the minor season. The programmed income per acre came to t344.2 or 39 percent more income than the actual income. The program used all the cash available, including borrowing up to the limit. With a slack showing uuu< u o< mHUUU u u panama vausnEOUIloouaom o~ an mN an me uuou oaousu upon :H mwnm:u IIIIIIIIIIIIIIIIIIIIIIIIII o n.~ o.m ~.H o.m o o< new; common Meow: . lllllllllllllllllllll o.n o.m n¢.N mm.w mu.m mn.m “.0 H.@ o< snug nowwwm Mono: wswmmouu MW «H.Hm qm.- nN.qn a.mm m~.om n~.cm m.oom o on :mou vosouuom l mfiuuo IIIIIIIIIIIIIIIIIIIIIII mo.mo~ mo.mo~ aw.mo mo.mo w.noa w.noa mH.m~H mH.mNH AOV coma sumo maku IIIIIIIIIIIIIIIIIIIIIIII wnn www o.wq o.Hm can wNa «com o mum vow: uoamq rowan IIIIIIIIIIIII: name mums mnom moon mum pomp honed zflwsmm ~.¢¢H Illllllllllll oo.vaH n.voaq «.moqq N.nwma 4.0om m.uomno nufis wuaswom H woman no nomwnmmaoo .uq.m anmH 106 0mm comm mqmm mmwm wWDuo< wofifimuwoum q 0 K q— H m N v x w 0 m N I\ 00 I\ oo .wwawmaoo "wousom mEoocH Cw wwcmno opo< puma sowmwm Hoswz O .N wuu< Ucmq commom nemmz wsfimmouo ammo @m30huom ON :\ LA 00 \‘r 00 :\ some emmo know now: uonmq wwnflm N [\ C) H H usom vow: gonna hawfimm O 1\ m \o owom meow Hmsuo< onEmuwoum “.ammfi m.mmom Hmsuu< vanmpwoum mm.omoa m.wmom Hmsuo< wwEEmuwoum (I) ‘H '0 CD 0 QEOUGH mmOMU cmuowno suHB wuasmmm H mmmzm mo somHHmQEoo .wq.m mHnMH 107 for the minor season land, the MVP was zero. In the Eastern region, the optimum income was ¢1867.2 as against the sample average of ¢l403.4 representing a gain of 33 percent. The cash expense available was used up to its limit whereas only 27.4 percent or ¢53.9 of the borrowed money amounting to c74.2 was used up. All the major season land was devoted to maize—cassava-plantain—cocoyam—yam— vegetable enterprise. Thus, all the farm family's food consumption requirements were provided from its own resources. In the Volta region, the programmed income for the Category I farms was ¢1104.3. As Table 5.4c indicates, this repre- sented an increase of ¢277.4 or 34 percent. All the major season land and the minor season land were used to their maximum limits. The cropping plan did not include plantain which had to be purchased in order to satify the family consumption requirements. In the actual crop plan, shown in Table 3.3 all the family food requirements were met from home production. In the actual crop plan in this region, the cropping enterprises were so diversified as to produce at least all family food requirements. All the minor season land was allocated to maize. In the programmed crop plan, how- ever, minor season land was left idle, but the major season land was used to its limit and was allocated to maize— cassava—plantain—cocoyam-yam enterprise. The programmed income came to ¢1824.02 representing an increase of ¢382_04 or 26 percent. 108 Overview On the examination of Tables 3.3 and 5.4c, it is clear that the programming procedure has selected fewer crop enterprises than were actually observed in the sample. Pure—stand maize on major season land was completely elimin- ated from all the crop plans. It is evident that the pro— grammed crop plan did not find maize, as a pure—stand crop on major season land, competitive with the crop mix enterprises. The gross income increase by programmed allocation of resources in the Category I farms in the study areas were from 26 percent in the Central region to 49 percent in Brong—Ahafo region above the observed incomes. Category II Farms The programmed results of Category II--transitiona1—— farms are shown in Table 5.4d. The actual or sample crop plans for the representative farms can be seen on Table 3.3, Chapter III. It will be recalled that the sample farms in Volta region did not meet the main criterion for the speci- fication of Category II representative farms, viz., the adoption of some improved practice such as the use of fertilizers. Thus, both here and subsequent discussion of Category II or transitional farms exclude Volta region. In Brong—Ahafo region, the programmed income for Category II farms in Brong-Ahafo region was ¢5070.0 with the actual being ¢3255. An income gain of 57 percent was — fi—r 109 effected. On examination of labor inputs, it is evident that in the optimum plan, the amount of hired increased from 2,264 hours to 11,072 hours. Cropping activities, including yams, tend to require a great deal of labor, for instance, for preparing yam mounds. The 7.5 acres allocated to maize—cocoyam—yam and vegetable enterprise required a large increase in hired labor. The programmed plan in Ashanti region allocated 0.1 acres to pure—stand maize in the minor season and 3.8 acres to maize—cassava—cocoyam-yam enterprise. The resulting programmed income of é2068.3 represented an increase of 52 percent income over the actual income of ¢l354 observed for the sample. The hired labor inputs also increased from 432 hours to 609 hours in the programmed solution. In Eastern region, the programmed crop plan for the Category II farms in the Eastern region allocated 3.2 acres to maize-cassava—plantain-cocoyam-yam enterprise. Also, 2.0 acres were allocated to pure-stand maize in the minor season. As compared with the sample crop plan given in Table 3.4, it is obvious that the optimum plan selected fewer enterprises. The programmed income of ¢2662.3 was 37 percent more than the average income for the sample (¢1934.7). The total programmed income for the representative farm in Central region was ¢2028.7 as against ¢1636.53 in the actual plan. This represents a gain of 23 percent. 110 Both expenses and borrowed funds were used to their limits. The minor season land remained unused and there was also a slack of 0.87 of the major season land. Land obviously was not a constraining factor for this category of farm. Regional Comparison of Income and Farm Organizations by Technological Category Actual Versus Programmed The income increase by programmed allocation of resources for all representative farms is given in Table 5.4e. In the case of Category I farms, the increase ranged from 26 percent to 49 percent. The range for Category II farms was from 23 percent to 57 percent. These income increases are subject to a gamut of inter— pretations. Here, we shall attempt some hypothetical, but reasonable explanation for the observed divergencies in income. While no definite conclusions can be reached in the absence of the requisite data, this brief discussion will highlight areas that need research. The model used in the study discounted the question of risk and uncertainty. The technological coefficients used in the study reflected the weather conditions that actually existed. But, some measure of uncertainty might have been present influcencing the decision-making of the farmers. On the examination of the data in Table 5.4e, a definite pattern emerges. In both categories of farms, the income increase in Eastern, Central and Western regions 111 wouameoo "ouu=0m IIIIIIIIIIIIIIIIIIIIIIIIIIII: mm “.mm c.5m cm.m~ usoo oEoosH luom sH uwcmco is o N o m.~ o.~ .5 o 33. ES condom you“: III/L .1: «.2 N... ea SA 36 R... a... 32 2.3 commmm wenmz waammouo 3m 3m «3 S 2.3 .1: 3.2: 323 .38 :28 3.8.8 on” 02 8 S 86.2 No.2.“ 3;; 3.: .38 some fine quN whoaa ch moo oNH mom mau awe mflwoo vow: uopna pupa: mean onmw mane wnmm omeq omen mnam mmNN um com: nomad zflwsmm mmwm oncm amma wwow n.qmma m.Noc~ mm.emwa n.muom mfivoo wEoosH moose Hmsuu< voesmumonm Hosuu< noesenmoum Hmsuo< vosEmuwoum Monaco vmsamuwoum OE EmHmfim Egg om§<1uzomm 2m: z o H o m M .32: .GOHmom \5 9.5: H knowmumo mama macaw notwmno Law: muasmwm HH mmmnm mo comwummaoo .o.m man—mm. 123 J 33950 "3.58 l/J wad 0mm 00H wad N vacuuH CH owawzu O-oN XIII/III!!! . o.¢ . o N c N o o.~ o.~ : Auonazv use.“ vwmaflp . . . . . o a o o o a o s es Rs 35 .. case can.” vwmanD /1[1l[l]1.!/1ll/.l/ . . . . an m an m o N o N m.~ m.~ o.~ o.~ : Anommam uoawzv wwwuuu< lllllllllllllJlllllllllllJrlllllllll1llllllllllJllllllllllllllllllllL u o I o [Ill/ii] q “a m NH N o N o NN.n -.m ha.s Na.q : Anomuom “chase uwuouo< I'll! can mm.owma mag nq.mHNH mn.nw w.nnm nm.3 maumm HH muowmumo .Muma wamfimm vw>u~mno saw: mudammx HH wmwnm we domwummaoo .n.m «Haws 124 family labor and hired labor, the latter made available because of increased borrowings. The income gain ranged from 109 percent in the Eastern region to 336 percent in Ashanti region. A comparison of the cropping plans under the optimum (Appendix Tables B.lb, B.2b, B.3b and B.4b) and the actual situation (Table 3.4) show that the optimum plan allocated few profitable enterprises, suggesting some specialization enterprise-wise, not crop—wise as we will expect the situation to be. Yam is not only a high yield- ing crop, but is the most profitable crop to raise. In ,. fact, in the study areas in Ashanti and Brong-Ahafo, it is V the most important crop. Thus, we find 21.34 acres allo- ‘ cated to maize-cocoyam-yam—vegetable enterprise in Brong— r 1 Ahafo and 12.0 acres to maize—cassava-plantain—cocoyam-yam enterprise in Ashanti. The optimum plan in all regions excluded pure-stand maize in the major season, the shadow prices associated with it being ¢371.8, ¢4l9.9, ¢507.3 and ¢454.58 for Brong—Ahafo, Ashanti, Eastern and Central regions respectively. However, pure-stand maize in the minor season received the following allocations: 9.77 acres in Brong- Ahafo, 4.0 acres in Central region and 2.0 acres in Ashanti. It is rather surprising, the minor season land in the Eastern region was left unused in the optimum plan. Cash expense availability was not the reason since ¢349.1 cash was avail- able in period 5-—the period to commence operations on minor season land. The MVP per unit of family labor was zero in IIIIIIIIIIIIIIII::——————————————————————————————————————————————————————————————————————!==r 125 period 5 and labor hiring per unit in that period was ¢.lO too expensive. Thus, the program chose to salvage family labor instead of commit it to crop production. Regional Comparison of Farm Organizations by Category Table 5.5 presents the optimum incomes and farm organizations for Category I farms in all the study areas. The gross revenue per acre ranged from ¢310.66 in Volta region to £272.28 in Ashanti region. The gross income figures are net of the starting money capital and the principal of borrowed money. There was a marked increase in the acreage cultivated. All land, including hitherto idle land being brought into cultivation. On further examination of Table 5.5, it becomes obvious that the dependence of the farms on hired labor increased as compared with the Phase I situation. The ratio of hired labor to total labor inputs ranged from 45.5 percent in Central region to 53.47 in Volta region. The average return per unit of capital was above the cost of procuring one unit of it. In all the regions also, the average return per unit of labor input was high in compari- son with the cost of hiring one unit of labor. Enterprise specialization is indicated by the fewer enterprises in the optimum plan as compared with the initial Situation. Minor season maize in pure-stand received substantial acreage allocation in Brong-Ahafo region (five acres), in Eastern region (four acres) and in Volta region (3.74 acres). —+ 126 Appendix Tables B.1a to B.4b contain the optimum solutions for Category II farms. The gross monetary return per acre ranged from ¢415.55 in Ashanti to ¢300.23 in Brong-Ahafo region. The largest farms, however, were Brong-Ahafo (31.07 acres), followed by Ashanti (14.2 acres), Central region (12.48 acres) and Eastern region (9.224 acres). The dependence of the farms on hired labor is evidenced by the ratio of hired labor to total labor inputs-— 65.81 percent, 70.43 percent, 30 percent and 52.64 percent in Brong—Ahafo, Ashanti, Eastern and Central regions, respectively. The figures showing the amount of money borrowed per acre are high, far exceeding the institutional limit of ¢15.00 imposed by the Agricultural Development Bank for loans to small farmers. The average returns per unit of capital and per unit of labor are rather high, far exceeding the opportunity cost of borrowing (¢01.06) and the average wage rates which varied from region to region. Maize, as a pure—stand enterprise, did not appear in the optimum plan in all the regions. The shadow prices , associated with it were ¢37l.8, ¢419.9, ¢507.3 and ¢454.58, respectively, for Brong-Ahafo, Ashanti, Eastern and Central regions. The figures indicate the extent to which this crop enterprise is too expensive to be included in the optimum program. Generally, in the Phase II model, profitability of cropping enterprises, rather than subsistence requirements or security considerations, had a major impact on the cropping plans that emerged. —+' 127 Comparison of Income, Marginal Value Products and Average Returns by Region and Category On examination of Table 5.5 and Appendix Tables B.la, B.Za, B.3a and B.4a, it is evident that the returns per acre are higher for the Category II farms than the corresponding figures for Category I farms. For the two categories of farms in Brong—Ahafo, Ashanti, Eastern and Central regions, the difference in gross returns per acre were ¢l6.74, ¢l43.0, ¢156.94 and ¢63.98, respectively. The relatively lower gross return per acre for the Category II farm in Brong—Ahafo is quite understandable when viewed from the fact that an acreage of 31.07 perhaps is too much for efficient management under existing technology of production. This explanation is only conjectural; there may be other reasons to account for the phenomenon. However, it is noteworthy that the gross return figures for the Category II farms are higher than those for Category I farms. The ratio of hired labor to total labor inputs in Category II farmers in Brong-Ahafo and Ashanti were higher (65.81 percent and 70.43 percent, respectively) than the corresponding data for Category I farms in the regions, both sets of categories of farms, however, exhibit a heavy reliance on hired labor. There is also an appreciable reliance on hired labor in Eastern and Volta regions. The marginal value products (in cedis per unit of a resource) are given in Table 5.8 by region and for both categories. In Tables 5.4a and 5.4b, it was observed that 128 mafia... N .Cowouau “Hm .umo “Edna.— mfiuom a .Cowouwu "H .uuu vuusnaou “ooh—om : .2 . .2 . ma . no. .1 . o o... . in. «a . : n ".0.“qu : q.— . 3. 9". N: . mu. 2%. on. 3. am. : o weaken : o oo. o c c c c o o __ m uni—am .. co. ma. 3. we. 0 “NH. o in. ma. : a canon : 3. co. a... nee. :. 0.3.. Na. 2. 5.. : n vain—um .. n”. ma. 3.. v.2. ma. m3. NH. ma. ma. .. N venom 023.5, 8. S . o 2 . c «2 . c 2 . 8. so: a v3»: EVOHHUW m hon-NA om . Nn . mm . Nm . Na . um . Nn . um . «n . um . : 3.8—mun an . mm . mm . on . an . mm . mm . mm . mm . mm . : 3.2320 2. mm. 3. S. 3. 2. Q. S. Q. 8. ._ ".92 mm. .3. .3. .3. 3. 1‘. .3. .3. .3. as. : moon 3. mm . mm . am . mm . mm . mm . ma . mm . mm . cam—5m mouufiumz m.~ :2 «TN :2 «TN vmé .32 am.“ wan muuunauuvh $354240 m. 2. mm. R. R. an. 2. mm. mm. R. 4: 3&3 o.m o~.n w~.m mu.m mud mN.m RA :6 m~.m 36 03 an» .n min no.3? min 9..“ min mad min 3..” 21m 03 smmouou oo. oo. mo. 90. we. wc. ca. 38. we. we. HE: 535.: m.~ meé mw.~ no.~ nw.~ ne.~ a.mnl mo.~ $.N 3.5.1 o3 Ermmnwu we. co. 00. wo. wo. cc. cc. wo. we. 00. m5 aunt swan—Mum: can—swam ._ 22mm 3.9.. 2.3 o «.3 onflw N.m c muém : C055 25; vows—5 vmuuwau __ 353 1w": oénn Teen 5.93 cdmm To: méw.» 2.5mm : 3.325 was wanna: kuwuau .. 3.5 1%.. dam TS 36m 9% «.2 2.2 $.S 22:5 2.3 5:8 32 0.3.. 3.2m 9...: Sam 2... 12.. a .2; $.93 2 .Nsm 3.3. A833 2.3 we. wo. oo. 30. amo. cc. :0. #3. no. amo. macaw nmwu wsquumum : H HH H .3 H Hm H A shamaumu .umu .uau . one .uuu $3 .23 .uno .umo muHmm <.:o> 452.55 zzmhmfi— ou<=soo onu was HOANH wmmsuo>o mo N.H m. SOHuoHov cwaounu musmaH H. . Hoan .m H mm m o No.N m.wuom VGMH pumps: panama N.H m we n30 mo waHHmoHu H. . uan Hmcowunum .N H mm n o No.N n.w~o~ causeway HmHunH QH mmamao on .H m vao :H muuspoum wSHm> HmaH pm: usouumm wauoo n o m e m N H ammo uoaHE Hemma :Hmo waoocH coauwauam : w Humm A uoamq vGMH wcmH pwaéduwoum umEumm Hw:0HuHmcmHH co COHowLN .mNINNmH .mcwsc .COHmmM Hmuuuwo “maumm HH >Mowwumo «N . m monoommm cuHB muonpoum wsHm> Hmcwmumz new choCH umz .mH.o oHawH |||| 153 pmusdaoo "mouzom lll‘llj]\\ll ’j . . o MUCH 0U “OLGA OH mm m Nm an.em H.0m as.ma as.NN as mm a poem; “was: no enema .m . no sa< .m . a annumm . m o me.a ma.m N.oa wa.ma mm.Ha Nm.HH u Hmuaamo um . no. we. so. mas. aw. ammo. wmmo. mum pm\amz pas assume a n m o¢ Na ems ma.ows m.wsq m.Hms os.mme ma.Hme ma.aas a muo< ems cuss m omam seem “Ham anew swam mama mama mum poems masses cash ewes some seam anew wmmN mmmm same gonna Hmuoa 0 came Hams omen meoa «we mos mes mum Hosea sense w.o~s mo.qmm NN.mNm NH.H¢N mm.mna mm.mna ma.msa ammo ammo Hence .m o.ame n.5am mad N.¢NH mm.qoa mm.eoa mm.soa u emaouuom ua< .s w.m- mm.ewa NN.msH N¢.HHH e.ea e.ss e.qn a ammo mo ua< .m HN.OH o.m v.0 o.m n.q m.q m.q muo< v.mem mono< HNHOH .N e.aase a.mmme «.ammm o.amsm H.omMN a.mNON s.wNom a wacm>mm mmouo .H on an en on 0H ma «a Hw>mg monsommm. quD mudmmmz .maumnma .mcmsw .GOHwom Hmuuawo .maumm HH muowmumo hosmHonmm mo wmuswmmz UHEonoom wo humsasm .nH.o mHan 154 respectively. As the size level increased, labor hours per acre, amount of indebtedness per acre and the ratio of hired labor to the total labor inputs increased progressively. As far as labor use was concerned, except in situation 7, labor in period 1 was not limiting. Similarly, labor available was not limiting in period 3 till situation 6 was reached, and in periods 4 and 5, labor was not limiting till situation 5 was reached. In the periods in which labor was limiting, the MVP of labor continued to be greater than the wage rate (see also Table 5.4, Chapter V). This suggests that the farmers could expect a return to their labor which was not only equal to what hired labor could earn, but actually above it. In situations I to 5, when more than 50 percent of the labor inputs was supplied by the family itself, there were more periods in which the MVP of labor was zero (i.e., substantially below the wage rate), indicating that the farmers would be maximizing returns per acre rather than attempting to equate MVP with wage rate or the marginal factor cost. Category II Farms——Eastern Region It will be seen from Table 6.2a that in situation I both major season land and minor season land were limiting whereas cash expense was not. As expected in situation 2, as the land constraint was relaxed, the MVP per unit of land declined, whereas that of cash expense increased (i.e., from ¢.OO9 to c7.2). With increased labor availability in 1355 Ho. mm. «H. mo. $00. m.qo own H.H N.n $.5mu MH. ooo. m.~o q.q HmnH um: qu no mm 0N udmouwm .vuuamaoo “muusom . muum\oo.nm ad 0 oemo uom uHsHH maHBouuon can ammo whoa Neou .vaMH whoa Room .n H.smoe uuum\oo.ma um wow uHEHH deBouuon was ammo whoa NomH .wHMH whoa NONN .w s.emms wuum\oo.ON um umm uwaHH wdHBouuon use ammo muse NocH .pde when Now .m s.~mmm «HUM\oo.o~ us Henna duos Nom .q ~.ooh~ o.H ou we. Eouw uamHUfimwo loo HOQMH mo fiOHmHM? luou mfiu cam HOAmH wmmnuo>o mo aoHuonw dnu swsoufiu hHaasm HoamH pummmuonH .m o . Hme flaw." wwwdfifl woudmu uo sac mo wuHummHu vamH Hmdowunwm .N M. NOON mouflowwh 5H mwamau on .H mwuwo w m w m N H zmmo uoafie Honda puma puma :Hmu oaoudH aOHumDUHm woaamuwoum .MNINmmH .mamcu .uOHwom :uuummm .waumm HH zuowoumu NN "maumm HMdOHuHmcwuH :o aOHmcmaxm bouzowwm suHB muoavoum o=Hw> Hmchnmz cam wEoocH uwz .mN.w MHan 156 situation 3, the land became very limiting whereas cash expense was hardly a limiting factor. Examination of Table 6.2b shows that in that situation only 170 units of labor were hired, a figure representing 5.72 percent of the overall labor inputs. The income gain in situation 3 over situation 1 was 2 percent, hardly of significance. The only noticeable change in the two situations was the increase in labor use efficiency, i.e., 571 hours per acre of labor were used in situation 3 as compared with 994 hours in situation 1. As both land and cash expense resources expanded from situation 4 onward, the MVP of land declined and became zero in situations 6 and 7. However, the MVP of cash expense increased REE; passu. The erratic behavior of the MVPs of land and cash expense seems to suggest that an alternative resource expansion procedure would be preferable, particularly one that expanded the most limiting source (land) and kept cash expense, which was already adequate in the initial situation constant. In all the situations, more than 60 percent of labor requirements were supplied from family sources, though the percentage of hired labor to total labor inputs increased as the resource was expanded. The greatest income gain was in situation 7 (149 percent). However, the most feasible resource expansion would seem to lie between situations 5 and 6. If money is made more available, for instance, through credit, than according to nee-classical theory of the firm, 157 .wwusaaoo ”muuaom jig} am am no.Hm ow.oa aa.ea Na.m aw.aa o.m N poems Hmuoa 0“ nonmq wmuHm mo OHumm .o Illlltllllllllllllilllllllll Illllllllld MHm mew mom mwo Ham qu «mm mum wuu<\mum Honmq .m mm.qm om.m~ om.qH mm.mH ma.m mm.MH o.mH u muo<\uo30uuom uaa .q aa.a a.oa o.NH wN.NH em.aa ao.ma wo.ma a Hmuaamo\apsuaa .m mm. mm. Hm. mm. Hm. ow. Nm. u know cm2\GH5umm .N wo.mom mo.ama am.oam am.amm «.aHN am.HHm wa.aam a wuo<\ assume .H Illlllrllrllsllnut mane sane News amaa caaa swam aaam mum asacH poems Hmuoa oama mama Namm “mam coma aaam oaaa ape camp Hanan aaaaae amea aaaa oaoa ooa oaH «am mam ape name Hanan amuse 0.00m mm.NHm o.om~ mm.awa No.mNH No.mma No.maa a sum: ammo ua< Nam N.amm a.mHH a.mw m.oa a.mw q.wa a emzopuom baa awe ma.aom a.mom mm.HaN No.meH am.oaa Na.moa a camp ammo Hmuoa N.MH N.HH HH.w N.o N.m Nu.o N.m muo< beaumm mowed kuOH e.oqaa H.amoe a.awma o.~mmm N.ooaN e.HwHa a.masa a aaooaa amend on ma ma an UH ma mA mousomwm uHcD wuswmoz HH muommumo .mmoHum we mHm>wH wcqum> amps: mocwHonwm oHEocoom mo wousmmmz mo humEESm .mnnmmmH .mamnu .uowwmm cumummm .mahmm .nN.o mHan 158 it is better, from a profit maximizing point of view, to borrow money up to the point where the MVP of additional unit of it equals interest plus the principal to be paid. This condition may be satisfied with a resource expansion in between situation 5 and 6. Category II Farms——Ashanti Region Table 6.3a and Table 6.3b give the details of the results of the seven alternative resource situations in Ashanti region. It can be seen from Table 6.3a that because land was already adequate in the initial situation, its expansion alongside with capital, left its MVP unchanged at zero. A more feasible expansion policy was to expand cash expense. As expected, the MVP of cash expense decreased successively with the expansion of this resource. In Phase II, when an optimum amOunt of borrowing took place (¢1219.43), the MVP per unit of land in this representative farm rose to ¢446.76 with the MVP per unit of money capital——¢.06--just equal to the interest rate. The income gain from resource expansion in situation 3 was 11 percent while that of situation 7 was 185 percent. The maximum optimum income in Phase II for this category of farms was ¢5900. It appears, therefore, that if the level of operating capital used in situation 7--¢975 had been used in situation 2 (i.e., to make it somewhat comparable to Phase II situation), the resulting optimum income would have approximated the level indicated in situation 7. 159 .OousnEoo "condom HO. m~.m O O me a.mowm whom\oo.mm um uwm uHEHH wcHsoHuon was Ammo whoa NOON AOGmH wuoa Room .5 no. no. mm. «O. HO. mn.m o o mMH 0.0qu wuum\oo.mm um uwm uHaHH maHBOHHon paw ammo whoa NomH .OamH whoa NONN .O co. co. no. mO. w.m O O OO H.momm ouow\o0.0N um umw uHEHH wcHSOHHon can ammo muoa NOOH .OSMH whoa Now .m mm. OO. OO. OO.H w.o O O mm a.mmom wuom\oo.ow um uwm uHEHH waHaouuon mam ammo whoa Non .pumH common Henna whoa NOm .q OO. mo. mm. ¢.O O O HH 0.00NN o.H co co. Bouw uawHUmeooo HOAMH mo GOHmnw>aou mnu mam Hoan paws Iuw>o mo cOHuoHou wnu nwsounu kamsm HOAMH OmmmwuodH .m 00. co. N.H w.n O O o m.wOON pde pawns: pwuawu no use we wdHumeu pnMH HmaOHquwm .N OO. 00. N.H w.n O O m.wOON muusomwu HmHanH :H wwumco o: .H unmoumm mHuwu Ammo ROCHE Henna CHNU waouaH nOHumsuHm UGMH wcmq pwafimumonm 2:23 225 33mg Eczema. .mEmm HH knowwuwu NN "mfiumm HNQOHuHmGMpH co COchmme oUMSOmmm :uHB mausvoum wsHm> Hmchuwz was waoocH umz .wm.o mHan 160 .pmusaaou “condom mH.am Hm.ws HH.am am.ma ma.mH HH HH H u83 Hmuoa on Hoan emHHm no aOHumm .s smm aHm sma aNa mam HNHH OHHH ape muo<\mpm HopmH .m aN.ss sa.sm sH.om o.Hm wN.aN sm.am sm.am a aHo<\easoHHom Hes .s sm.m sa.s Na.m NH.a ma.oH aw.a ma.a a HauHamo\:H=uam .m as. as. He. am. so. as. as. a mum cmz\¢H=Hme .N aa.oam H.amm s.sss m.HHm sm.asm ma.ssm wN.ssm a wuo<\cp=umm .H aaam NHoa mmsm Haws mama Hams aHNs mHm usaaH HonH Hmuoa smsm swam waam Hon mama smmm Nasm ape Hoan aHHamm has mama mama HmoH omHH sea Has Has mum HonH emHHm “as ssa sas a.mNN N.HsH mmH mNH mNH a emsopuom ammo Has Hem HHN saH m.omH am am am a same game Has mOOH mHH N.Ham H.HaN OHN OHN OHN a same name Hmuoa m.sH m.MH s.a N.m N.s m.m w.m auo< e.HHme wanes Hmuoa a.mamm a.omws H.momm a.ame~ a.maNN m.msoN m.wso~ a esooaH mwouo UH HH HH oH 0H mH w1h 00H50w0M_ UHHHD wHDmmwz .mHINnOH .mcmzw .GOHmom Hummcm< menmm HH %Howwumo monoHonmm uHanoom mo whammmz mo kaEESm .nm.o anmH 161 It will be seen in Table 6.3b that as the resource levels expanded, the ratio of hired labor to total labor input increased. The amount of borrowed money used per acre also increased from situatons 4 to 7.5 The return per unit of capital again declined in that range, but in general, the returns per acre and per man hour of labor displayed an erratic behavior. Category II Farms—-Brong—Ahafo Region In Table 6.4a, it will be seen that in situaton 3, there was an income gain of 7.4 percent. As expected, the MVP per unit of labor by period declined as compared with the initial situation. However, the MVPs per unit of money capital and per unit of land increased, also as expected.‘ From situations 4 to 7, the ratios of MVP per unit of labor by period remained constant and above the wage rate, except in periods 7 in situations 6 and 7 when the MVP just equated its salvage value. The values of the MVP would indicate that the farmers throughout the resource expansion sequence could increase their earnings if they were prepared to put in extra hours of work. In Table 6.4b, the labor use per acre was the lowest in situation 3, indicating labor use efficiency in relation to other situations. \— 5Situations 4 to 7 more clearly depict size sequence eXpansion of resources of land and capital. 1632 .kusaaoo "mQHDOm MMH. oq.H o.OHN a.qu O.mmq.NH whom\oo.om um uwm uHEHH wanouuon pom sumo whoa NOON .vamH whoa Noon .n NOH. NOH. MNH. mNH. sHH. sHH. mMH. ss.H s.sHN N.aa O.HOH.oH muum\oo.m~ um uwm uHEHH wcH3ounon van ammo whoa NomH .vamH apes NONN .0 «OH. NOH. MNH. maH. sHH. sHH. mMH. Hw.H q.HON 0.0s m.omm.n mHum\oo.ON um uom uHEHH MGHBOHHOO cam :wwo whoa NOOH .van muofi Now .n NOH. NOH. MNH. mNH. sHH. sHH. MMH. HO.H O.HON m.m~ O.MHm.O whom\oo.ow um umm HHEHH mawsou upon cam ammo whoa Now .OaMH common Memos whoa Nom .s QHO. «HO. woo. woo. No. mm.H mq.H m.wNN O.H o.ssq.m O.H ou ow. How uaUHonwooo .3an mo uOHmuo>uoo can Hoan one: |Ho>o Mo GOHuwHoO or» nwdousu quOGH Hoan wwwmouoaH .m NO. NO. HO. HO. mo. q.q m.ONH «.0 o.~mm.m UGMH woman: wooden no use we maHumeo waH HmGOHquom .N NOH. NOH. MNH. MNH. sHH. sHH. MH. Hw.H O.HON m.OnO.m mouzommu HmHuHcH CH mwamno on .H uamuumm wapwo HoaHE Henna cho waoocH HOHumsun OGNH UGmH Omaamumoum :o :OHmcmmxm mu mnuwmmH mamno .conmm owm5< NN "mahmm HMCOHuHmcwHH unomwm :uHB muusponm wsHm> choum .mfihmm HH zuowwumu HmanHmz use maoucH mwouo .ms.w anmy 163 .Omusafioo ”mousom Hoan HmHOH on Hoan ewHHm Ho cOHHmm .s Illllllllllllllllllllllll mHo<\mHm Hoan .m muo<\meOHHom ua< .s HmuHawo\uH3umm .m Hm amz\eusuwm .N muo<\:u=uwm .H IIIIIIIIIIIIIIIIIIIIIIIIII musacH HopmH HMHOH Hoan aHHEmH Hes HoHaH emHHm Has wwzouuom ammo Hmuoe IIIIIIIIIIIIIIIIIIIIIIIIIII pom: mmHo< HmHOH vamp ammo Hes mmBouHom “as mEOUfiH mmOHU wm.ow mq.mw mo.on Om.mm wo.qm m.wo w.mo N «mm «MOH MOOH NQOH NHO ONOH MHOH H «a.mH mm.mw 0.0m mm.mH No.mH om.NH a.mH u w.o m.o Nq.m O.H NH.O so.m m.w v ON. ON. mm. mm. mm. mm. mm. m O.QNN wO.HnN OH.HON mm.mwN wm.omm wH.mmN q.HmN u qunq qmqwm Nmmmm qumm quqH NHNON NNOmH mum Ommo Ommo Ommo Ommc Ommo ommm Onmo mum NOHHq soomm NOOON MQOnH wmnn NQwMH NmNHH mum mN.Hme oqu OHOH qu mo.omm mo.omm mw.omm v Nw.mm wH.mm mm.mN mq.NN mo.mw cm.mH q.mH muo< omn mmo oom mum 0mm 0mm 0mm u mN.HwOH mom mHm mqq mo.oqm mo.oqm mo.oqm u mquH HOHOH m.ommm MHmO seem O.Nmmm m.o~om u I! {fill/[III]; UH MH QH UH mH ¢H I’ll‘l L[ Ill. Hw>®d OUHDOmwm C mmwz . mu: .mcmso .cOHmom omwsnoo oau mam Hoan womauo>o mo aowuoa loo wnu :msoufiu hammsm momma commouoaw .m coma oq.w m¢.mHm o. m.mmnu voufion Ho aBo «0 wawumoao coma Hmaowufivvm .N Ln n.5wwa wuusommu aH owamnu on .H unwouwm wflwoo ammo Hedge momma Gama mfiooaH coaumsufim wcmu mama woaamuwoum // .mNINmmH .mswso .¢0Hmwx auoummm .mfiumm H huomoumo om ”wagon HMGOHunmuH so EOHwammxm wouaomom nuws muusvoum oDHm> Hosflwh «2 com oEooaH umz .m.o oaamfi 167 .vousmaoo "ouuaom co. Na. 0 o co. «Na. 00. am. o.aa Nam m.m w.mnma o.a ou wo. Bouw uaoaoawmooo gonna mo aoamno>noo wnu woo Monma wmoauo>o mo doauoa luv onu nwsou5u haamnm uonma uwmmwuoua .m co. co. 0 co. we. no. we. mw.m o Noaa n.~ n.oqma mama woman: common no czo we maaumoau coma amooauawuo .N mo. mm. o oo. oo. mm. 00. Mn. 0 wmm m.qoma oopsomwp awauaca ma wmaMSU o: .a nocaa MmeE dame ofiouaa coaumduam mama coma uoaamuwoum .anlmsma .mcmco .coamom aucmaw< .waumm H %uomwumu Nm ”wagon amcoauavmuy do :oamsoaxm wousowom nuaz muusvoum opam> amaawnmz cam waooca uoz .m.e canny 168 .kusmfioo ”wounom mm. o0. «am. 0 ca Axomamv Axomamv m.Nmma o.a ou ow. Eouw unmaoawwoou uonma mo coamHo>uou com momma vmosuo>o mo noauoaou mam swdouau musmfia uonma commouoaa .m 00. mm. 00. mam. q.oa Axomamv Axumawv m.wmna mama woman: wouawu Mo :30 mo wnaumoau mama amnoauawum .N 00. mm. 00. mam. q.oa Axomawv Axomamv m.wmna oousomou amauaaa Ca ownmao o: .a unwouom mawwu a m N a ammo moaaa Hemma :amu mwoauom Na Honma osoocH mama mama coasmuwonm nHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHH coaumsuam .mmlmnma mcmno .coamom owm£Mom: m o.~ ~.N o.~ ouud >500: n o.n o.a o.m muu< >womo: w . . o.m o.m muud womuz m m n e a m.v whoa >omz a . . n.m m.n w.m m. uuu¢ woo: m S m a m4 m5 9m 95¢ 82 N o.~ NA o.~ o4 ea a; M: m4 0855 muu< unama a Qme mnammouo o o o 0.3 mg: «.5 Na o a; ma 2 ~.m .12 a...“ a.m. N :53 ~83 cu uonma m vouac canon w wn com cam mac maca aww o~o now can ohm ao~ awn wmm muw wmaa um whom you mono: wanna m o o o n.wa «.nm m.a~ v.n m.w m.oa m.m a.- «.aa a.ma q.m u when you vmsouuon . . . . . unseen a o 3 a Z a 3 a a m o To TD saw a.ma Tow oaa a.ma a.ma Néa 53 a HERE a. a. a. m use assume m . . . m. m. o. a. n. n. q. a. o. N. a. u u: uwm ahauuu N c “on a mmu a mam a.mmm mlaqm m.mw~ a.owm ~.m~a m.mnm m.mom m.~m~ «.cow N.Nam a.mmq w.oom u you Guano“ a Nmoq «mom nm0m mace mama aomq ohom «mom amen amqm moon oqmm maom «mac wmoq um muamqa Momma amuop Nmoq anon nmom wwwm mwmm wmon mnwu «mom oamm ooqw waN mmnm NmmN mmoq cams um Manna zaaaom . . . N ama mom naa oma am «an aNa mmm um wanna nauam N mNa N mNa N mNa a.qma a.¢ma a.ama N.woa o.m~a o.m~a m.nw o.-a m.ow o.qm m.qma o.waa u now: . . . nmmv annoy m Noa m sea m sea m.om a.mm a.mm n.om a.mm a.mm N.am m.~a u mozouuop . . . . . o . nah—=05 N mwa N nma “.mma M.Ww M mm M mm M mo n.oo n.ow o.mN a.mm o.n~ o.¢m n.mca a.moa u Ammo mo unaoaa . . . . .m o.m o.m m.< n.¢ m.m o.m 0.0 o.m u mmuum asuoa m Nana m mama m mmna o.omma n.oon wmouu ua ma ma oa pa wa oa na sa ua na ma ua pa ma :oamwm owmn :oamom amuucou uaaz anuH .mcoamom aa< .msumm H muomoumo ao>ma wousomom wcazuo> nova: hodwauawum mo mwuammw: vasomoom mo humaadm .oa.w canny 171 corresponding figures in Category II farms Phase I (Table 6.10). In situation 3 in all the regions, labor use effi- ciency in relation to other situations is indicated by the magnitudes of labor hours used per acre. A picture that emerges is that by making more credit available to the farmers, by removing the constraint that brings about underutilization of farm labor (situation 3) and by making it possible to supply farm inputs such as planting materials and simple farm implements to the farmers in sufficient numbers, there will be some income gains, as indicated in Table 6.5 to 6.9. The latter policy option would be supported by the magnitudes of the MVPs per unit of inputs shown in Table 5.4a, Chapter V. Programmed Income Category II Farms--Phase II The discussion in this section is related to Appendix Tables B.la to Table B.4b. The alternative resource situa- tions to be considered here are: 1) initial resource with unlimited amount of borrowing, column 2A; increased labor use through the conversion of the labor hiring coefficient from .66 to 1.0, column 2B; allowing land clearing up to a limit of 40 acres (20 acres in the major season and 20 acres in the minor season) while at the same time putting maximum and minimum acreage constraint on pure—stand maize, but maintaining land renting activities in both the major season (RENTM) and minor season (RENTN) with their corresponding constraints (RENTLIMT and RENTLINT), column 2D. 172 In the programming results presented in Chapter V, it was observed that major season maize in pure—stand was in a very weak competitive position as compared with the crop mixtures. A justification for imposing maize maximum and minimum constraints was to find out the effect on income and farm organization if pure—stand maize cropping was forced into the optimum cropping plan in order to meet some of the requirements of the Maize Crop Improvement Project. Central Region One effect of converting the labor hiring coefficient from the.previous level (.66 to (1.0) was to ensure that by eliminating the average of an hour a day spent on walking, hired labor could be made to contribute at least eight hours of service for the same pay. Column 2A in Table B.4a presents 1 the programming results. The gross income was ¢4835.94, 1 representing a gain of 3.4 percent over the initial situation [ (Column 2A). The average return per acre increased slightly, but a sharp reduction in labor inputs per acre was achieved.6 The programmed returns per unit of capital and labor were c9.l3 and £1.06, respectively——a marked improvement over the corresponding figures in the initial situation. The ratio of hired labor to total labor inputs was 28.95 percent as against 52.64 percent indicating a fall in employment, but 6By increasing the number of hours worked per day from 5.66 hours to 8 hours, say 1 cedi spent on hired labor will get more work done than before. IIII||||IIIIIIIIIIIIIlIll[T____________________________—___"__—____“I""I”II 173 high income gain. The programmed cropping plan that emerged was unchanged. The programmed income that emerged was é4647.28, a slight decrease from the income in the initial period. The programmed cropping plan showed some specialization with 21.43 acres devoted to major season maize in pure-stand, 14.0 acres to minor season maize in pure-stand and 7.05 acreSIxymaize- cassava-yam mixtures. The gross income per acre, however, decreased to ¢109.4O from the previous level of c374.64. By removing the maize acreage constraint (situation 2D), the gross income jumped to ¢14,380.65, leading to greater return per unit of capital, per unit of labor and per acre. Eastern Region In.situation 2B, it is evident from Appendix Table B.3a that there was an income gain of ¢63.63, as compared with situation 2A. The returns per acre, per unit of capital and per hour of labor increased slightly. As expected, labor hours used per acre declined from 689 to 618. With the imposition of maize acreage limits, the gross income declined from ¢4051.77 to t3855.21. The average returns per acre of land, per unit of money capital and per unit of labor declined rather drastically to ¢l33.02, ¢2.22 and ¢0.34, respectively. With the removal of the maize acreage limits in situation 2D, the gross income — I 7 174 increased to c13,476.66 with the other related average measures correspondingly increasing also.7 Ashanti Region By increasing the coefficient on labor hours worked per day (situation 2B), the gross income increased to ¢6228.87, representing an increase of ¢328.03 or 5.6 percent over the gross income in situation 1. The labor hours used per acre declined as was expected, but the average returns per unit of capital, per unit of labor and per acre increased (Table B.2a). With the imposition of the maize acreage limits in situation 2C, the gross income was c4088.12 as compared with a gross income of ¢l4,225.41 when the constraints were removed in situation 2D. The average return per acre in situation 2C was ¢1l9.54 as against ¢4lS.95 in situation 2D. Brong—Ahafo Region It can be seen in Appendix Table B.la that by increasing coefficient on labor hours worked per day, there was an increase in income of ¢386.83 or a gain of 4.2 percent over the income in situation 2A. Again, as expected, there was a reduction in labor hours used per acre, i.e., 545 hours of labor were used in situation 2A as against 722 hours in situation 2A. 7It did not return to the previous level (situation 2A) because the land renting activity was retained, so that the overall acreage expansion was 29.244 acres as compared With 9.224 acres in situation 2A. 175 With the imposition of maize acreage limits, 22.8 acres were allocated to maize in the major season, 19.77 acres to maize in the minor season (Appendix Table B.lb), and 18.54 acres to maize—cocoyam-yam-pepper enterprise. With the removal of the constraints, pure-stand maize in the major season did not come into the optimum solution. It is shown in the Appendix Table B.lb to be ¢37l.79, too expensive to be forced into the plan. The resulting gross income in situation 2C was c8,285.42, as against ¢16,762.27 in situation 2D, a difference of ¢8,476.85. 'Discussion of Catggpry I Farms-—Phase II For Category I farms in Phase 11, two resource situa- tions were examined: 1) the initial situation (2A) and increasing coefficient on labor hours worked per day (situation 2B). The results are summarized in Appendix Table D.2. The gain in income ranged from 3 percent in Brong-Ahafo region to 10.9 percent in Volta region as a resulting of shifting from situation 2A to situation 2B. With the exception of the Volta region, the optimum cropping organization remained the same in both situations. The overall picture that emerges is not only income gain in situation 2B over situation 2A, but 1) returns per unit of labor, per unit of capital and per acre increased; 2) there was a reduction in labor used per acre in situation 23 as compared with situation 2A and 3) with the exception of Category I farm in Ashanti region, the amount of money 176 borrowed per acre declined in situation 28 as compared with situation 2A. With increased availability of labor and with borrowing allowed to its optimum point where the MVP of additional unit of capital was equal to its marginal factor cost, land became the more limiting factor. Thus, for all the repre— sentative farms in the regions, the marginal value product per unit of land increased in situation 2B as compared with situation 2A. Concluding Remarks The preceding discussion in this chapter, aimed at three broad policy issues: 1) making more cash available through credit expansion; 2) increasing size of farms; and 3) eliminating walking time for labor. In connection with the Phase I static linear programming model, seven alterna— tive resource situations were examined to throw some light on the most feasible path of resource expansion for Category II farms. Three alternative situations were examined for Category I farms. In connection with the Phase II poly- period model, four alternative situations were examined for Category II farms and two situations for Category I farms. According to neo-classical production economics, maximum output from agriculture is forthcoming from given resources only as mobile resources such as money capital and labor are applied to immobile resources, such as land in a manner that the ratio, %¥%, is approximately equal in f7: 177 all its uses. The principle of factor proportionality also suggests a liberal application of the resource in plentiful supply, in order to economize on the relatively scarce resource. Empirical studies, such as this one, which attempt to operationalize these economic principles, are of value in suggesting the path of economic adjustments and policies designed to promote not only increased agricultural produc- tion, but also efficient agricultural production. However, the earlier caveat regarding extrapolating the MVP beyond one unit should be remembered when policy actions are contemplated. The inter—area and inter—situation comparison of marginal productivity in the Phase I model in this chapter has provided some insight into the policy issues posed. Relatively high marginal value product per unit of land in Eastern, Volta and Brong—Ahafo (for Category II farms) indicates that farm expansion in these regions should receive special attention. In Central and Ashanti regions, the emphasis would be on capital. Situation 2A in Phase II model clearly demonstrates J the weak competitive position of maize. In order to force major season pure—stand maize into the crop plans, it became necessary to impose minimum acreages. The analysis indicates that this is an expensive thing to do. Again, the analysis also shows that permiting labor to work a full eight hours a day will prove profitable for the farmer. What this study 178 did not investigate is the cost of making such a situation possible. In the next chapter, these policy issues will be explored further. Some alternatives not empirically tested in this study will also be discussed. CHAPTER VII POLICY ISSUES, SUMMARY AND CONCLUSIONS The study has employed static linear programming and poly—period programming to investigate the most profitable farm organizations for representative farms identified in the study areas in Ghana. It is the purpose of this chapter to relate the programmed results to the micro-economic aspects of decisions with regard to policies or programs affecting 1) changes in the provision of credit; 2) changes in infrastructure such as the provision of a network of feeder roads and public transit services to reduce the time farmers spent walking to farms; 3) on—farm storage organiza— tion as a contributing factor in the profit maximizing efforts of the farmers; 4) changes in the subsidized prices Of inputs such as fertilizers and other chemicals; 5) changes in guaranteed minimum price for maize; and 6) acreage expan- sion or the size factor of smallholder subsistence production. The first five programs are specific development programs subsumed under the Maize Crop Improvement project for the agricultural areas included in this study. The strategy used in this section of the chapter is to focus on policy- related empirical findings in the study. Other issues which 179 180 were not empirically tested in our study will be introduced inasmuch as they have a bearing on the relevant policy issue being discussed. Credit It is the policy of the Ghanaian government to provide cheap credit to ”small” farmers as well as ”big” farmers. The central issue revolves around this question: How can credit he made an effective instrument in developing agri~ culture in Ghana? This question has wide ramifications encompassing both micro and macro aspects of decision-making. We shall.main1y address ourselves to the former in our discussion. Four sub-issues immediately are inferrable from the main question posed. They are distribution, interest rate, loan conditions and firm-household interdependence. Distribution On the average, 90 percent of the Category I farmers interviewed in the survey reported that they faced the problem of inadequate credit from institutional sources. The corresponding figure for Category II farmers was slightly lower—-73 percent——but it is still substantial. Given the Size of holding (a range of 2.41 to 17.34 acres for Category II farms and 2.0 to 6.68 acres for Category I farms——Tab1e 3-3) and the interdependence of production, consumption and Savings or investment, the traditional-cum—"transitional” farming methods have kept production per farm household at 181 a level which barely meets consumption requirements with little left over for savings.1 The implication is that capital needed to purchase additional inputs or hire labor to clear more land must be borrowed. The timing of borrowings and the amounts thatcan be borrowed are two major considera- tions.2 In the absence of making institutional sources of loanable funds easily available (in the sense that the farmers do have knowledge about where to go for credit, how far they have to travel to get the loans and whether the attitudes of the bank officials do not frustrate the efforts of the farmers to obtain loans),3 farmers needing loans inevitably resort to traditional money lenders, who charge high interest rates. 1This situation was not tested empirically in the study. It is merely a description of the situation as observed, based mainly on the smallness of the operating capital the sample farmers had to cope with. 2The situation is different in the United States, for instance. The farmers generally are in a position to deter— mine how much money capital they need, whether to borrow or not and how much to borrow. Heady and Swanson, for instance, report in their study that 61.5 percent of the farmers refused to use additional credit because of risk factors. [Heady and Swanson, 1952]. 3During the survey, questions were put to the farmers to determine the main impediments preventing them from getting credit. In addition to collateral requirements, which proved to be the major hindrance, 90 percent of the farmers reported that they had no knowledge as to where to go for credit; 85 percent reported that the bank offices were too far away from them; and 62 percent reported that they were often frustrated in their efforts by bureaucratic delays suggestive of indirect kick—back demands by the officials. 182 With the restrictions on borrowings removed in the Phase II, the model determined not only the optimum amount to be borrowed, but also the timing of borrowings in response to production and consumption requirements. As an illustra- tion, the amount borrowed for Category II farms in Brong-Ahafo was c346.0 in Phase I. The total acreage cultivated was 17.4 acres and the gross income was ¢5,070.00. With the removal of the borrowing constraint in Phase II, the aggre- gate amount borrowed was ¢l,580.89 and the gross income increased to c9,310.01. According to nee—classical theory of the firm, farmers wanting to maximize profit should borrow money up to the point where the marginal value product per unit of additional money invested in the farm business equals the interest rate. Table 5.8, Chapter V, gives the details of the marginal value productivity of capital used for cash expenses for representative farms in Phase II. In comparison with the return per unit of capital given for Phase I model shown in Tables 5.4a and 5.4b, it is evident that in Phase II the marginal condition for allocative efficiency postulated above is satisfied. Under existing credit arrangements, it is evident from Table 5.4a that with respect to Category II farms, farmers in Ashanti, Brong-Ahafo and Central region suffer more from inadequacy of capital than farmers in Eastern region. For Category I farms, farmers in all the regions except Brong—Ahafo suffer from capital inadequacy (Table 5.4b). 183 Rate of Interest While formulating a credit policy, inter-region and inter—category difference should be taken into consid- eration. A general approach for determining the credit needs of the farmers is to use the technique of resource- variable programming. This procedure will help determine the maximum amount of money that can be borrowed to maximize income. But, the cost—range reports provided along side the linear programming solutions showed that the optimum solutions were highly insensitive to charges in interest rates. For all the categories of farms and in all regions, four levels of interest rates were tried to determine their effect on income and farm organization: 6 percent, 9 percent, 12 percent and 15 percent per annum. No basic change ensued, thus substantiating the observation that the optimum solu— tions were insensitive to changes in interest rates. The Phase II model which eliminated the borrowing constraint emerged as a better guide to determining the optimum amount of capital needed to maximize income on individual represen— tative farms. Presently, the credit needs of the farmers are determined by the official credit institutions, which are required by law not to charge more than 6 percent interest on the loans to the small farmers. At the same time, it must pay 7% percent on savings deposits it receives. Thus, the cost of securing funds and making a loan to the small farmer is higher than the expected return at current interest rates. 184 The question then is——what interest rate to charge. Any rate of interest used in the Phase II model (i.e., whether 9 percent, 12 percent or 15 percent) would have left the basis unchanged.4 An important guide is to have estimate of the MVP per unit of capital as reflected in the Phase model. The returns differ between regions and between categories. This would suggest a multi—interest rate structure which would be difficult to implement. A useful guide may be to strike a compromise between the rate of interest of 14 percent which the commerical banks charge and the 7% percent which the official lending institution must pay on savings deposits it receives. The evidence is that it is the adequacy of the credit not the rate of interest which is of concern to the farmers. Farmers in the sample who received loans from money lenders paid between 100 to 200 percent interest. They probably did this because the MVP per unit of capital was very high for some farmers in some regions as implied by the Phase I results. It is the conclusion of this study that the adequacy of credit is what claims immediate attention. The question of determination of the appropriate rate of interest must be further investigated in another study. Conditions for Credit Three major requirements are embodied in the granting Of loans to farmers in the study areas: collateral, pure- stand cropping and a maximum of ¢15.0 an acre loan for each 185 farmer. Since the farmers interviewed have little or no fixed investments to act as collateral, a credit policy devised on the basis of farm planning will be more effective than one based on security of loans. The average returns per acre, per unit of capital and per man—hour of labor are substantial (Tables 5.1, 5.2, B.la to B.4a) and would be helpful in guidng the loan—granting policies. For Category II farms, for instance the amount borrowed per acre in the model solution ranged from a low of ¢40.88 in the Eastern region to a high of c85.0 in Ashanti region. These are a marked departure from the bank's limit of ¢15.00 per acre. Similarly, the study has concluded that mixed-cropping is in a stronger competitive position than pure—stand cropping. The implication is that, a lending policy based on overall productivity rather than pure-stand requirement will contribute more to farm expansion and higher income gains. Farm—Household Interdependence The Phase 11 model confirms that household consump- tion requirements feature in the profit maximizing decisions of the representative farms. What this implies is that in addition to estimating credit needs based on production requirements, estimate should be made of family expenditure on the basis of family budget approach. The banks can separate the two types of credit needs by giving credit for productive purposes in kind in the form of implements, fertilizers, needs, etc. —_——— 186 Other Relevant Issues There are other relevant issues connected with credit which are not covered in this study. Some of these are: the impact of credit on employment and income distribution; the encouragement of thrift as a condition of receiving loans and the establishment of rural credit institutions. These issues are important and merit consideration in another study. Labor Utilization The survey results show that the farmer walks an average of three miles to and from the farm. By Ghanaian standards, walking on bush paths with several obstacles in the form of fallen trees and rivers without bridges, this would take about 60 minutes per day. After adjusting for travel time and labor works only about two—thirds (66 percent) of the normal average of eight hours a day.4 In Chapter VI, the effect of removing this bottleneck on farm income and farm organization was determined by eliminating; l) labor overhead on the part of the family labor, and 2) changing labor hiring coefficient from .66 to 1.0. The effects of these changes on income, labor utiliza~ tion, amount of money borrowed and farm organizations are summarized in Tables 5.6 and 5.7 for the Phase II model and Tables 6.1 to 6.6 for the Phase 1 model. ' aThe implication is that 1 unit of hired labor supplies .66 hours of labor per day. The coefficients of labor hiring activities by period given in Table 4.3 reflect this observa- tion. 187 Using Category II farms, Brong—Ahafo region as an illustration, there was an income gain of 7.4 percent with 1.43 acres of minor season land which was in slack in the initial phase being brought into cultivation. The amount of labor hired per acre was reduced from 1,013 hours in the initial situation to 612 hours suggesting labor use efficiency. The returns per unit of capital increased from c8.5 to ¢9.12; that per man hour of labor from ¢O.29 to ¢O.38 and gross return from é5,070.3 to ¢5,444.0. As the figures in the tables indicate, there are minor inter-region and inter—category variations. But, the general tendency was an overall improvement: greater income and less labor used per acre. The implication is that any measure than can help remove the bottleneck that leads to this type of labor utilization will prove profitable to the farm business. A possible solution, as both private and public measure, is a network of feeder roads to open up the geographic areas. Farmers, for instance, can be bussed to their farms. The results of this aspect of the study has implications for other issues not empirically verified in this study. For instance, 1) the benefiEcost ratio of constructing feeder roads needs investigation, 2) market—related benefits and cost, i.e., location of storage facilities at points near consuming centers; prompt evacuation of perishables as soon as they are harvested, etc.; 3) the possibility of feeder road construction paving the way for settled farming so that 188 the farmers need not walk long distances to their farms; and 4) associated with the last issue is the feasibility or infeasibility of establishing neighborhood schools up to ' secondary school level, so that school children could stay longer periods with their parents and help in the farm work. All these issues warrant consideration. Storage A storage sub-model was incorporated into the Phases II and III of the poly-period model in order to provide some guidelines to farmers as to the most feasible timing of the sale of commodities such as maize, yamsand pepper. In arriving at the programmed solution, marginal value products associated with the output transfer or inventory carry-over from period to period were computed. These are shown in Appendix E. The MVPs indicate the shadow prices associated with the balance equations and also indicate the rate of change in the objective value or the optimum income if slightly more or slightly less of that particular commodity were made available. According to Driebeck, I' .if these shadow prices had been used in the objective function, each stage of the storage process would have been optimized by itself and a solution identical to the overall solution would have been arrived at." [Driebeck, 1969, p. 96] The market prices of the commodities in the various periods are given in row II of Appendix E in each region. 189 The results show what income gain per unit of commodity would have been if the farmers had the know-how and resources to organize year-round storage operations. Thus, the potential income gains from storage suggest a close look at the benefit-cost ratios of encouraging on—farm storage. Three policy options that warrant considerations are: 1. Provision of the requisites including credit to the farmers to build the storage facilities themselves. 2. Encouragement of cooperatives to establish the storage facilities. 3. Public provision of these facilities to be operated on the behalf of the farmers at a cost. Plantains and cassava in their natural form cannot be stored for a long-period. However, when processed into chips or flour, these can be stored for a considerable length of time. Thus, an important linkage of agricultural output increasing efforts with rural industries may make substantial contribu— tion toward improving rural welfare. Effects of Varying Maize Price on Farm Income and Adjustment The linear programming solution involved in this study used the actual selling and buying prices of maize to reflect regional differences. Because the government's guaranteed Udnimum price for maize was fixed at a level common to all the farming areas, four levels of the same price for all 190 regions were examined for their effect on the profit maximizing plans that had already been obtained. The concern here was to find out whether in view of the weak competitive position of maize as a pure~stand crop in the major season, the changes in price would lead to basic changes in the cropping plans. The minimum selling prices used were ¢8.0, ¢l0.0, ¢12.0, and ¢15.0 with other prices and resource levels held constant Phase I levels.5 Table 7.1 summarizes the programmed results for Category I farm in Volta region and Category II farms in Brong—Ahafo, Ashanti, Eastern and Central region. Table 7.2 presents the range report to show the sensitivity of the optimum plans of the respective representative farms to changes in the price of maize. The range report shows a lack of sensitivity of the optimum solution to maize price changes in Brong-Ahafo (O to ¢8.0) and Volta region (0 to ¢12.85). However, in Bronge Ahafo at the price of ¢10.0 there was a change of the cropping plan with 5.8 acres allocated to minor season maize in pure-stand. Between ¢l0.0 and c16.0 price changes did not affect the basis. In Ashanti, although there was a basic change in the optimum solution at the price of ¢l0.0 , it was not until the price of maize had risen to tlS.O before further basic change occurred with 2.63 acres allocated to 5In Phase II, seasonal selling and buying prices were used. The parametric price variations, therefore, relied On the Phase I model where annual average prices were used. Table 7.1. Effects on Farm Org Prices with Other 191 anization and Inco Prices and Resourc me of Varying Maize es Held Constant. M“.— Region Activity/ Unit Maize Price in Cedis (d) Income } ¢" 15 .' 0‘ 0“ Income 5935.7 Activity: _ MCPO 13.83 213325 MOW 3.5 Maize (Minor) 5.8 Income 2058.2 2098.6 2139 2208 Activity: Ashanti MCPOY 3.0 3.8 3.8 1.75 MCPO Acre 2.63 Income ¢ 1925 1968 2011 2074 Activity: Ea t MCOP Acre 3.2 3.2 3.2 3.2 S em Maize (Minor) Acre 2.0 2.0 4.0 4.0 Income ¢ 2529 2584 2639. 8 2721 Activity: MCY Acre 3.9 3.9 3.9 3.9 central MCO Acre .02 .02 .02 .5 Income c 1220 1224 1228 1251 Activity: V 1t MCOYV Acre 2.39 2.39 2.39 2.24 O a Maize (Minor) Acre 1.75 legend M = Maize C = Cassava P = Plantain 0 = Cocoyam Y=Yam V = Pepper = Cedi Source : ( currency) Computed j” 192 maize. In the Central region, Eastern region and the Volta region, basic changes in the optimum solutions occurred beyond the price of ¢12.85 causing 2.0 acres more to be allocated to maize. However, in none of the situations did major season maize in pure—stand come into the solution space, a clear manifestation of the weak competitive posi— tion which maize in pure—stand holds in the major season. If the objective of using the guaranteed minimum price scheme is not only to ensure ready market for sellers of maize, but also to expand the output of maize relative to the other crops, a minimum price of t12.0 a bag would appear necessary. Table 7.2. Price Ranges for Maize. Category II Farms. Region Price Ranges Brong-Ahafo 0.0 to 8.0 Ashanti 3.46 to 9.69 Eastern 1.87 to 12.85 Central 1.71 to 12.85 Volta+ 0.0 to 12.85 +For Category I Farms Source: Computed 193 Input Prices: Subsidies6 In this section, we examine three inputs which during the time of the survey were being subsidized by the govern- ment, viz. fertilizers, weedicides and matchetes. In this year, the subsidy on matchetes has been withdrawn. We shall be guided in our discussion by cost range reports which were part of the linear programming output of Phase II.7 In the Central region, for instance, the program used 33.4 bags of fertilizers at an initial cost of ¢2.8 a bag-—i.e., the subsidized price. The results show that the linear programming solution would remain optimal so long as the price of fertilizer stays between ¢0.0 and £17.18. In Brong-Ahafo, the program used 104.9 bags of fertilizers at an initial cost of ¢2.8 a bag. The range R 6The LP models in this study used variable inputs in fixed ratio to land. Theoretically, farmers will adjust the rate of inputs used as prices change. The type of adjustment of input use consequent upon price change would be different in the theoretical case from the actual situa- tion modeled in this study. Therefore, care should be exercised in the interpretation of the range report. 7For a discussion of cost ranges, see Driebeck, 1969. The cost range shows the stability of the LP solution for changes in the cost of a single activity, keeping all other costs, technical coefficients, resources, etc., constant. The range report also shows what other or new activities Would be selected at either the minimum or max1mum cost. The range normally include the objective coefficient value Of the relevant activity. Altering cost or price Within the range can cause changes in the objective value even though the optimum plan or the operating strategy remains unchanged. .4._-...n IIIIIIIIIII:444444___________________________________—____-___—__—az____—tinfimw' 194 report further indicates that the linear programmed solution would also remain optimal so long as the price of fertilizer stays between é0.0 and ¢9.5. In Ashanti region, an amount of 54.8 bags was used with the price ranging between ¢0.0 and ¢25.6. Similarly, in the Eastern region, 32.9 bags were used with the price ranging from ¢0.0 to ¢28.0. To verify the stability of the LP solutions with regards to changes in fertilizer prices, three levels of fertilizer prices were used to determine their individual effects on the optimum solution: ¢3.6, ¢4.5 and ¢S.7. In each case, there was no change in the basis indicating that if fer- tilizer prices were raised to ¢5.6 a bag, the programming solution would remain optimal. The range reports indicate further that with the exception of Brong-Ahafo region, the optimum, various optimum plans were insensitive to fertilizer price changes. Increasing the cost of fertilizer to say ¢5.6 does not seem to suggest a different operating strategy even though the total income decreases with increases in fertilizer price. As to the level of fertilizer price to suggest, the MVP per unit of fertilizer in the Phase II model is not a useful guide. With the elimination of the borrowing constraints in Phase 11, all MVPs per unit of the inputs approximate their respective marginal cost. It follows that, if a fertilizer price of say ¢5.6 a bag had been used, the MVP per unit would have approximated t5.6. For the three levels of fertilizer prices tested for Category II farms in BrongrAhafo, viz. ¢3.6, ¢4.5 and ¢5.7 a bag each, the corresponding MVPs per unit of fertilizers were £3.75, ¢4.7 and £5.9, respectively. Furthermore, as indicated in a preceding footnote, the model used a fixed ratio of fertilizer to land planted to a given crop enterprise. The implication is that if the farmers are loaned all the money needed and charged the full cost of fertilizer of about él3.0 a bag, they will still equate MVP with MFC. But, since the objective of the subsidy is to encourage fertilizer consumption, the government might as well leave the subsidy at its present level. However, there are other distributional effects which the subsidy imposes that this study has not investigated. An example would be the income redistribution effect of across the board subsidized fertilizer prices. An alternative guide to input pricing is to examine the MVPs per unit given in Table 5.4a, since the Phase I model depicts the actual constraints facing the farmers. Of the four regions, the MVPs of fertilizer, machetes and other inputs approximate their opportunity costs. In other regions, however, they are high. With respect to herbicide, the current price of ¢6.0 a bag would have to fall to c3.6 a bag before it would be possible for crop enterprises, alternative technology 2, to come into the optimal solutions. The farmers interviewed 196 cited the price of herbicide as the major reason why they were not applying it. Thus both the model and the survey suggest a downward revision of the price of herbicide if the government wants to encourage its use. Weeding, using manual labor, is a labor intensive farm operation whereas the use of herbicide to achieve the same purpose is rather a labor saving device. In Appendix Table B.2a the use of herbicide reduced labor hours used per acre in Phase 11 (column 2A) from 965 hours to 626.2 hours (column 3A). It is clear from the results that the application of this technology will reduce farm employment. Farm Size Factor As mentioned earlier, Ghanaian agriculture is composed predominantly of smallholders. An analysis of the 1970 census data, for instance, reveals that approximately 65 percent of the farmers operate less than 10 acres of which about 20 percent produce only for subsistence consumption, 50 percent produce a surplus for sale and 30 percent produce mainly for sale. With the exception of the representative farms in the Brong—Ahafo region, the sample data in this study are a true reflection of the conditions portrayed in the census survey. However, the earning power of the resources used by all the categories of farms in this study (i.e., their respective marginal value products) point to the great scope for enlarging the productive capacity of the farmers through resource expansion. According to IIIIIIIIIIIIIT_____________________________________________—_______________—'——_T" 197 Johnson [1968], however, productive capacity is only a measure of potential ability to produce and is of little help in forecasting supply or predicting the amount actually produced and released to the market. He writes: ” . .forthcoming supplies depend on the degree to which actual price relationships permit producers to attain or exceed productive capacity“ [Johnson, 1968]. The size sequence expansion options empirically verified in Chapter VI point to the income gains that can be attained (Tables 6.1 to 6.7 and Table 5.6, Chapter V). The income gains reported are only indicative of the potential that can be attained. They assume, for instance, that all the produce will be harvested. However, there are other aspects which this study has not investigated, viz., prices have a lot to do with how much of the crop the farmers havest and carry to the market for sale. This issue also needs further investigation as it is integrated with product, feeder roads, storage and distribution policies. Summar The agricultural economy of Ghana possesses vast potential for increasing agricultural output and associated employment. Presently, however, productivity is low and the state of agricultural technology has been relatively static even though the Ministry of Agriculture has invested in efforts to modernize farming in the country. Given this state of affairs, 198 a diagnostic study was needed to identify the small farmer problem and to provide some insights into efforts necessary for expanding the productive potentials of farms. Accordingly, this study was desinged to focus attention on the following objectives: 1. Analysis of the organization of subsistence farming in the major maize growing areas so as to assess and appraise the economics of present resource use and the requisites for increasing agricultural output and farm incomes. 2. Determination of the efficiency of resource utiliza- tion and profit maximizing plans consistent with initial resource use and expanded resource use and technology of the categories of farming identified in the survey. 3. Evaluation of the potentials of the various policy instruments such as product and factor prices, rate of interest, on—farm storage, etc., which could be used to bridge the gap between actual and potential production and thus provide the framework for policy manipulations desired to achieve expanded food production and farm incomes in an optional fashion. 4. To determine alternative technological potentials of producing farm output, which can be considered by the extension workers in their innovation diffusion efforts. 199 5. To demonstrate the methodological reasonableness and efficiency of using linear programming techniques to examine the dynamics of on—farm storage of crop output with consideration given to consumption withdrawals for family subsistence needs. Static linear programming and poly-period linear program- ming were used to assess the income increasing possibilities for the representative farms by an optimum allocation of the resources actually used by the farms in the sample. The representative farms were defined by the level of technology of production and by the ability to adopt agricultural inno- vations. Thus, two representative farms-~traditional and transitional were defined for each geographic area. The analysis was repeated for three empirical phases and for all the five geographic areas located in five regions in the country, viz. Brong—Ahafo, Ashanti, Central, Eastern and Volta regions. In the static linear programming model in Phase I, seven alternative resource situations were analyzed to determine the most feasible resource expansion. With the results of the Phase I model pointing to the large earning power not only of the most restrictive resource—-money capital and 1and——but also the complementary inputs such as seeds, fertilizers, and simple farm implements, the model in Phase II allowed borrowing up to optimum levels instead of putting a restriction on the amount of money that could be 200 borrowed at the going rate of interest of 6 percent per annum. On—farm storage activities were also incorporated into the Phase II model. In Phases I and II, only those activities were included which were actually undertaken by the farmers of the sample of 1972—1973. In Phase III, however, parallel cropping activities representing two alternative advanced technologies of producing crops in pure—stand were introduced. The remainder of this chapter will concentrate on the major findings, implications of the conclusions and suggestions for further research. W . ~. 1. On both the transitional and indigenous farms, the marginal value products of land and capital were high, suggesting that increasing the use of these resources would lead to income gains. A great income raising possibility was also indicated by the marginal value products of agricul- tural inputs such as labor, fertilizers, planting materials, and farm implements used by the farmers. Generally, the pressure for increase in farm size is shown by the high MVP per acre of land in the study areas. 2. For all the categories of farms in the study areas, mixed-cropping had a comparative advantage over pure—stand cropping, as indicated by the shadow prices. The implication is that given the choice, the farmers would prefer growing crops in mixtures rather than in pure—stand, a fact that 201 militates against the introduction of new technology or enterprise specialization in the study areas. The advantage which mixed cropping held over pure—stand cropping was found to rest on the fact that new fertile land was continously being brought into cultivation. When there is no frontier of land, this advantage may disappear. 3.. The analysis also points out that the starting point in a program to encourage farmers to increase resource use in the study area is the organization of adequate credit supply. This conclusion immediately follows: If the marginal value product per unit of capital is high, the formulation of credit policy should aim at providing credit to farmers taking into account expected returns, prodcution requirements and household consumption requirements as well. A credit policy based on productivity would be more effective than a policy based on security of loans. Farm planning, developed into an effective extension tool, would provide guidance to the institutional loaners. 4. Both labor use efficiency and income gains could be derived if thebottlenecks that to give way to under- utilization of both family and hired labor are removed. One policy option considered is the provision of a network of feeder roads. The cost and benefit aspects of this policy option, however, needs to be studied separately. 5. In view of the limited resources of the government to help farmers, additional efforts should be made to help —: r 202 farmers maximize their incomes by organizing on—farm storage operations. Again, the benefit—cost aspect needs to be looked into as well as the macro—effect of storage on total village prices. 6. The cost range report indicated that the programmed solutions were insensitive to fertilizer and herbicide price changes. Theoretically, farmers will adjust the rate of input use as prices change, but since the model used inputs in fixed ratio to land (e.g., 4 bags of fertilizer per acre for a given enterprise) rate could be determined rather indirectly through changes in enterprise. The sensi— tivity analysis is, therefore, of little help in offering guidelines as to the price of fertilizer or herbicide to recommend. However, within the framework of the current practice of recommending fertilizer use in fixed quantities to land, the sensitivity analysis has proved helpful in determining the enterprise or a combination of enterprises that would help maximize income. Despite the insensitivity of programmed solution to fertilizer price changes, it is recommended that the subsidy be maintained to encourage increased consumption of the input. The income distributional consequences of across the board fertilizer subsidy program needs investigation. 7. The parametric maize price analysis identified two levels of maize price which when applied to the study areas would encourage the farmers to plant pure—stand maize for 203 market, viz. ¢12.85 a bag in the Central, Eastern and Volta regions and ¢9-50 a bag in Ashanti and Brong-Ahafo regions. But, since it may not be administratively or politically feasible to maintain two levels of guaranteed minimum prices for maize in the country, a compromise price, fixed at ¢12.0 a bag applicable to all regions in the country might be tried. Limitations and Suggestions For Future Research Some limitations of this study must be noted. Linear programming models were used to assess the income increasing potentials of resources. The extent of income increase determined by the programmed results can be overstated because of the survey results used to derive the yield and price assumptions used in the models. The models also ignored stochastic factors such as weather variability which can affect the farmer's decision—making. Another limitation of the study is the use of bio- logical yields instead of actual yields in the model. This limitation can be avoided by extending the period of study to about four years to cover one full production cycle. However, we need to consider the benefit—cost of the two alternative approaches. The models can be further used to investigate factor proportionality and derive more detailed responses to input price Changes. The poly—period model which covered only one year divided into periods will need further extension so as to adequately investigate the dynamic interdependence between fin 204 production, consumption and investment over a period of say five years. The results also indicate what further activities and restrictions will need to be incorporated into the models. These activities will include activities that allow changes in technologies, and changes in the support prices for land clearing activities. Additional restrictions should be considered for seasonal labor to permit the incorporation into the model or work specializa— tion by age and by sex. Crop mixtures have been demonstrated in the study to hold comparative advantage over pure-stand crop enterprises using advanced technology. It is the belief of the author that the advantage which mixed-cropping now holds over pure- Stand cropping is due to the availability of new frontier lands. In the long run, when shifting cultivation has pushed land to the extensive margin, the advantage alluded to will disappear. Both agronomic and socio—economic research is needed to investigate this long-run soil exhaustion argument. The conclusion was reached that removing the bottle- necks that lead to labor utilization will increase labor use efficiency and farm income. It was suggested that a network of feeder roads to facilitate this is a feasible Policy option. The cost and benefit analysis of the suggested program is needed. This study also highlights the importance of collecting 205 input-output data in farm management research. Without a continuous supply of these basic data, it is impossible to formulate any programs dealing with farm planning in the dynamic agricultural economy of Ghana. The Phase II model needs to be extended for a period of four years to adequately analyze storage delays for cocoyam, plantain and cassava-~crops that undergo continuous harvesting. However, such a model will still be deficient in the absence of agronomic data dealing with the following special features of the crops: 1. The varying maturity dates of plantains of the same variety planted at the same time. 2. The time it takes for cassava and cocoyam to reach maturity beyond which date the root crops start undergoing deterioration. It is suggested here that a combined LP model and simulaton of distributed delays of the type used in the Korean Sector Study [Johnson, et.a1., 1972] will be appropriate in analyzing the continuous harvesting of plantains, cocoyam and cassava CI'OpS. APPENDIX A 206 Table A.l. Sales, Consumption and Storage Activities for Maize. Column Activities Column Interacting Coefficient Number Name Row 1 Buy maize Period 1 BMZl MZOP1* -1 2 Consume maize Period 1 COMZI MZOPl l 3 COMZl CMZl l 4 Buy maize Period 2 BMZZ MZOPZ —l 5 Consume maize Period 2 COMZZ MZOPZ l 6 COMZZ CM22 1 7 Buy maize Period 3 BMZ3 MZOP3 -l 8 Consume maize Period 3 COMZ3 MZOP3 1 9 COMZ3 CMZ3 l 10 Sell maize Period 4 SMZ4 MZO —l 11 SMZ4 CASHP4** -5 . 4 12 Consume maize Period 4 COMZ4 MZOP4 1 l3 COMZ4 CMZ4 l 14 Buy maize Period 4 BMZ4 MZOP4 -l 15 BMZ4 CASHP4 6.0 16 Store/transfer maize Period 4 MZTR4 MZOP4 l 17 MZTR4 MZOPS - .973 18 Sell maize Period 5 SMZS NZ 1 l9 SMZS CASHPS —5.04 20 Consume maize Period 5 COMZS MZOPS 1 21 COMZS CMZS l 22 Buy maize Period 5 BMZS MZOPS -l 23 BMZS CASHPS 5. 64 24 Store/transfer maize Period 5 MZTRS MZOPS l 25 MZTR5 MZOP6 - .97 26 Sell maize Period 6 SMZG MZOP6 1 27 SMZ6 CASHP6 -6.4 28 Consume maize Period 6 COMZ6 MZOP6 l 29 COMZ6 CMZG l 30 Buy maize Period 6 BMZfi M2 P6 —1 31 BMZ6 CASHP6 6.77 32 Store/transfer maize Period 6 MZTR6 MZOP6 l 33 TR6 MZOP7 - .989 34 Sell maize Period 7 SMZ7 MZOP7 1 35 SMZ7 CASHP7 -8. 7 36 Consume maize Period 7 COMZ7 MZOP7 l 37 (301427 CM27 1 38 Buy maize Period 7 BMZ7 MZOP7 —l 39 BMZ7 CASHP7 9. 7 40 Sell maize Period 7 SMZ77 MZOP77 1 41 SMZ77 CASHP7 —10.06 42 Consume maize Period 7 COMZ77 MZOP77 l 43 COMZ77 CMZ7 1 44 Buy maize Period 7 BMZ77 MZOP77 -l 45 BMZ77 CASHP7 10 06 *MZOP1,. . . ,MZOP7/MXOP77 : Maize inventory, Period 1,. . .,Period 7. **CASHP1,. . .,CASHP7 : Cash at hand, Period 1,. . .,Period 7. APPENDIX B APPENDIX B EXPLANATIONS OF TERMS USED IN TABLES B.la TO B.4b Table B.la LANDMI (Acre) Major season land in acres LANDNI (Acre) Minor season land LANDMZ (Acre) Unused land, major season in acres LANDNZ (Acre) Unused land, minor season in acres RENTLIMT (Acre) Maximum constraint for land renting-—20 acres major season RENTLINT (Acre) Maximum constraint for land renting minor season up to 10 acres MZLIMT (Acre) Maximum maize acreage 1imit-—20 acres MAZLIMT (Acre) Minimum maize acreage limit, equal to initial acreage CUSTOMH (Acre) Custom—hired machine services in major season CUSTONH (Acre) Custom-hired machine services in minor season CASHPI - - - CASHPZ (Cedi) Cash at hand period 1 to 7 CASH START (Cedi) Level of starting money capital LABRPI — — — LABRP7 (Hour) Family labor in periods 1 to 7 LABRAM (Hour) Annual family labor LABROFI — - — LABROF7 (Hour) Off—farm labor periods 1 to 7 LABROFAM (Hour) Annual off-farm labor availability Table B.2a STACHASHI (Cedi) Start cash in period 1 BORRWI — » - BORRW6 (Cedi) Borrow cash periods 1 to 6 CASDI — - — CASD6 (Cedi) Cash at hand periods 1 to 6 CLEARM (Acre) Clear unused major season land CLEARN (Acre) Clear minor season land REMTM (Acre) REnt major season land REMTM (Acre) Rent minor season land BWEED (Bag) Purchase weedicide CUSTHIRM (Acre) Custom—hire machine in major season CUSTHIRM (Acre) Custom—hire machine in minor season BLABRI - - — BLABR7 (Hour) Hire labor periods 1 to 7 SLABRI - - - SLABR7 (Hour) Sell labor periods 1 to 7 SMZ4 (220 lbs) Sell maize period 4 MATR4 (220 lbs) Store maize period 4 SMZS (220 lbs) Sell maize period 5 MZTR5 (220 lbs) Store maize period 5 207 208 SMZ6 (220 lbs) Sell maize period 6 MZTR6 (220 lbs) Store maize period 6 SMZ7 (220 lbs) Sell maize period 7 SMZ77 (220 lbs) Sell minor season maize period 7 SMZVP4 (220 lbs) Sell maize as vegetable period 4 SMZVP7 (220 lbs) Sell maize as vegetable period 7 SCAS7 (200 lbs) Sell cassava period 7 SPLA7 (Bunch) Sell plantain period 7 SCOY6 (120 lbs) Sell cocoyam period 6 TCOYP6 (120 lbs) Store cocoyam period 6 SCOY7 (120 lbs) Sell cocyam period 7 SYAM6 (100) Sell yam period 6 TYAM6 (100) Store yam period 6 SVEG4 - - - SVEG7 (lbs) Sell pepper period 4, 5 and 6 TVEGH — - — TVEG6 (lbs) Store pepper period 4, 5 and 6 Maize (Acre) Maize output (Acre) Maize (Acre) Cassava (Acre) Plantain (Acre) Cocoyam (Acre) Yam (Acre) Pepper MZA2 (Acre) Maize output using recommended practices MAZ3 (Acre) Maize output using recommended practices with weedicide CASAI (Acre) Cassava output using recommended practices CASAZ (Acre) Cassava output using recommended practices with weedicide PLATAI (Acre) Plantain output using recommended practices PLATA2 (Acre) Plantain output using recommended practices with weedicide COYOAI (Acre) Cocoyam output using recommended practices COYOAZ (Acre) Cocoyam output using recommended practices with weedicide YAMAI (Acre) Yam output using recommended practices YAMAZ (Acre) Yam output using recommended practices with weedicide VEGAI (Acre) Pepper output using recommended practices VEGA2 (Acre) Pepper output using recommended practices with weedicide 0 H A .3: .354 "3: ago." E H33 E H33 3:: wounommm .HHH as E 33,5 .mTNRH 5%. . m I :0 no.“ om cum—E maoum .meumm HH zuowoumu we Hoe/mg auusowwm pew muoapoum 65.25 Hmcfiwumz .maé 0.3m“. don—HS, sum? A v 633300 "muuaom Illlllllilill it‘ll Nice «:18 ~32 3.3 RES N BEE eon: .58 8. 8.2 an «omfi :55 mo 3%” . .2 See 3.: mean 3 ueufifiaéom o R... was seen a.m.; o.~2 use excise n5 . [Illlllllxllllu 8 . .lxllll me...» wn.n mné mnd wed A~A C Hw>UA E HU>0A E .HU>0~H E H0>QH UMCD NUHDOMwM .:0Hwom ommsoq ounaomom oaamwum> Home: mamam .mnINmmH .mqmnu sham Edaaumo we humafiam .LH.m QHnMH 21.2 So. So. So. So. So. 3 «Guam no.mo~. mméo.‘ 2.03 mc.mo~ modem : 2.25 mm.nnu :63 3.qu ww.m- mwdmu : 825. 3.3. and.“ 3.3 3.3 3.3 OOH $25 no. «w. No. no. No. : 98m 3. we. we. we. we. : groom. 25.3 mmécm wméon 363 :53 fl oz 308 wmo. wmo. wmo. mmo. wmo. £055 55mm on . on . mm . on . an . e5 CON hm¢ow mwé 3.3 3.3 36 Hod : :sz amt: oo.m~ amKN war: mea : «SE 3.3 owéca 8&8. 3.0m m .9... : £56 34.: $63 8an 34: H .3." : firm mad: 36: 362 wed: win: : SEN: 002 w :13 «n23 :4: .373 MJHN Pro: m mm.m.n 50.3 min.» ciao .. emu: q on.~o <53 : .6: m mmKo NmKo ~m.$ 5.3 m.$ : r: N ant—Km 2.3% 8.2 86mm wénm u< 03w: H nqumm woo. «co. woo. woo. woo. >058 no.3 3.3 3.3 no.3 3.3 : coma mHKmHo omfimmv: “3.3mm 313$ mHKnHo : 08:5 36mg 3.85..” oménwm 3.33 3530 : mama. one. 30. 3c. 30. was. : QO>m @430 $43: «0625 mémoo wrnmmm n3 coma. «m an UN mm oq = Hu>mu : Hw>uq : fiw>mq : Hw>ug awn: vuuaomox .Av mfivoov HH can NH wmwzm .mnINan .mdmzo .:0kum fiuamnm< .maumm HH zuomwumu mo Hw>mq wuuaowwm tam muuawoum 05Hm> Hmcwwwmz .m~.w wmeH 215 moan> xume munumwunwu A V ¢ vuuanaou "wuuaom Newww Nwwww Nwwé NHH.Hw uwwé N H58 8. «85 ummHm mo 0HHuu : Ho>uq : Hw>oq = H0>uA : Ho>wq uHc: wuuusomux .vmaaauaoo w~.m wanmw 21.6 «we . wwww use «we wwwH = wmewam eH. eH. eH. eH. eH. = wmweHe ewe. Hwe , eHwH mm N em = weeeqe Hew HwHe wwHw wHwM wwwN : weeeHw wwww waew ewew new wwww : «mque wHH. wwwH wweH we we muaom HeeeHw w~.~w __ zmHeympo «e.w uuu< zmHmemeo w~.H ewe emmsw e.w e.w : zazmm e.e~ e.e~ : zyzmw wH.we H.we e.wwH H.weH H.5w : xxemqu e.w e.ww .w~ e.w e.w uuue zmwuHu wH.www e~.Heew w.wweH e.e~wH “.new.H = weweo “He. wHe. “He. “He. “He. : wzwmee ww.~H e.ww e.ww e.ww H.ww wowwn : Hw>wa : Hw>wA z Hw>ou c Hubwq awn: wuusomum .mNINan .mamau .aowmwm Haem4m< .mauem HH whomwumu .HHHH eam HH mummeee Hm>wH wouaowwm memem> “wee: mamHe aume assHueo we eumaasw e .n~.w emeH 217 no. no. no. 5. «NK: 3.me Q: e.ww: O.H”: : Sawm 3.0m.— héum Ton ~3ku «HHNH : 8.2:. n.n m.m n.n m.m m. COH 92km mm. mm. mm. mm. mm. : “Meow mm. em. 3. mm. mm. : @mwoou. a.m.: mc.nwo mmH a.m.wN 1an n5 ONH owoom mmofi 0.3mm Ncmw oumn ounm 5E5 55mm ow. ming .102 .7an e.w”: n5 ecu 55% na. no. mm. no. no. : n>Nxm e.w méfi TON emmé omod : «warm .35.“ NA: Q3 «.3 N6.” : 25$ 3.: mména m6: a.mn mKn : nNZm Ho.mN $33 imwa ~.No «.mo : SEN: ~.m N.m ~.n ~.m N.n : osz $.eu e.w: HIBN ow .73 : EN: e.w e.w o.m e.w e.w : mNZm nKN e.w: Tcou we me n..— ONN SHHNZ n3 e.w e.w e.w e.w «sz woo. So. «we. wmo. «mo. .. 533m «no. go. «we. wmo. «we. : 333m mm." ww..“ co." 03 as : 935m mm «we. go. «no. «mo. : «mean mOH. ms. med. mwc. m3. : namjm mm wHH. m3. mmc. mg. : Numjm 03 :H. :H. 2.8. :a. : 555m 02:6. mace mum meH has 9:5: 535m wq : H035 : Hw>wA c Hw>0A HE: ouusomom 60:93:00 .nué 3”st 218 .vuuanaou "ouuaom «NJ N mN ww..wN.n : .25“; «N N539 MN 56 : flaw NN on .noa : N952 HH noéq N.Nm e.w N.N.n N.NH : Mono: OH m¢.n e.w w.m a.ma o.o Mom: m .n.nn H.nm a.mn a.m.». : you: w mm.mo m.nHH a.m.: med: a.m.: : E: n $.2me «ON «ON fifiHN «e.wON : . omvz o No.03 mdma mdma N.Nm:n m.wma : ~62 m mien a.mm a.mw wood” ohdw : w: w omKnn Non Non n.wmm .an : >2 m ow.mNm QKNM mKnm «.mow mfinm : u: N HNéow ma.» NJNN 9.13 05.3 u< wnwmz .n "oszmo omw onm No. 0mm 0.96 : 5326 No. 0N0. mco. ch. wNo. : 083. So. So. nmo. nmc. nno. : owmgm coda «0.3” wad." coda +34: : mum—>9 co. we. we. cc. we. : mum>w «.3 «.3 «.0H v.3 «.3 mod cum—PH uu = Hw>wu : H255 : awash : MESA yuan wousomwm .waaaflaoe we .w 3.3. 219 we. new wH. new NwH. new ewe. wen we. wen .. wemweH wwe. wNwH NH. wNwH an. wNwH wwe. new wwe. wNwH .. wmmwwH wH. New wH. new wH. New ewe. New ewH. New ._ Nemweq ewe. wNw eH. er eH. eew we. wew we. er “no: Hmmw0A «HE Hw>ufl as»: HU>04 mg H0>NA A; HU>01~ uHGd mehflomvm .AHHH wee HH wwmmwev wnuNNwH .mcmeu .aowwwm cumummm maumm HH muowwumu mo Hm>wA wuusomwm can muuswoum m=Hm> Hm:Hmum: .mm.m wanma 220 wand; Juana muaumounou A 0 03.3500. "won—How Nn.mN www.mn NHO 5.6 Non N mHDAZH momj AdHOH OH. amHm ho OHHg 0.0q 00.00 00.nm 5.0m ww.?w A3 mmo<\Qm—zox¢om 92¢ .20 0mm No.00». 0H0 owe usom ”me—HERVE a: HN.m 00.0 NN.N 00.0 No.0 : A0ng 00. New N3. N0» 00. Nos 00. N05 00. N05 : e.w—mi N11. 0: Nw. 0: N3. 0.: wec. own NwH. own : 05:23 3va nNH SNNV noN AHNNV moN QNNV MON QNNV MON .50: P555 oa : .335 : H95..— = .335 m .325 uwca muuHSmwm. 822:8 .3 .w 033.. 221 I: . 311‘ IIWII|I| 2m No8 Rm can 048 .. gain 3 . S . S . S . .. 2mg «8. 22 «m3 «8. «no. ._ cumin 8. 33 :2 3o. 8. .. 2min 82 :3 we: a 82 .. Swim to. Bo Bo mmc. «B. 26: 2min e.w __ zfifimsu 33 5:596 03 wan fimzm e.w e.w .. 2.25 ON 98 : 2.5mm W: 3.: can 0;: ._ 5230 e.w «N 93 oé oé 98¢ 53.5 mdou 5 .3m QSS icon ~48 .. 896 So. :0. So. :8. SS. ._ £38m «Na «.22 .232 Nag a.m.: 836 c8. 80. So. one. as. .. 2:28 0. 8m 82 N .33 «.68 €08 __ «85 So. 3c. So. 3c. So. _. «Egon 3o. 30. So. So. So. 826 2&8 8. ow.» NZ mdw 93 _. Exxon 88. 88. 88. £8. 80. _. ~85 0&3 #8.: mag 32% «no? ._ 358 92 88. £8. <2 #2 __ 3mg 88. N . 2:. N . £3 £8. 88. .. Exxon o. m3 a.mfi a.mNH c.m2 a.mfl $30 Ewéfim : H0>~A = Ho>oa : Ho>mn : H954 : Hm>ou awn: wuuaomom 6723 .255 .EOHMUK Eflummm . WEE—NW HH town-HMO . AHHH HEM HH mwmflfimv Hu>0w~ OUHDOmwm anmHhm> “Over—D mfimHm Emmy ESEHUQO NO Aha—aw < .n—m om mHDNH 222 me. me. me. me. no. a «056 wm¢.n< DJ: 5.3 www.mq www.mq : swarm 0.3 e.ww." v.3 a.m.» 0.3 : EASE. no.5 no.5 mofi moé no.“ 03 crawm méma qcéwn 2.62" m.HmH méma : nwoom «Ken NA: céou «.now «.53 : @558. $4 Nwé um...” Nwé ww..“ n..— CNH cwcum moo. mNH. H3. mcc. moo. £25m 55mm hmémm NKE minnm naAnN noémm ”5 com nmwa = .335 : H954 : H255 : Ho>wA SE: uun=omwx wwsuwuaoo and wanna. 223 .uuunnfloo "wunaow N56 N53: «N e.w H<2Nz mu N6: _. «<9: Nu wing : H HN min: : ~52» o~ n53 13.2% mm TN: ~oo «H 5mg : .2950 : 5.3m : «5.5m 3 m.mwn .2513 ma a.mmu : m00: q BK: TEN «.mun «42 5.5:“ : 258: m «.cNN 902.. N5: m.mm~ ~.<~N : our. N m.~mq ego: No.8 mficm 9N3 uuu< IN: H uqummozu No. we. No. NC. NC. : mom—a Ho. Ho. Ho. Ho. Ho. : wow—Z. no. me. no. no. no. __ coma woo. wee. moc. ace. woe. : mom: 3e. 3e. «cc. «cc. go. : mum>m and Nm.m wad ~m.m $8 an; coma. wA : H26; = 13.3 : H95; 3:: 3.30m; .82:qu an .m 2an 224 3. EN 30. «5 2o. 3.. 8. in 3. «Z .. Emmi 8. m; 8. m5 8. mg we. 2» 8. m3 _. 25:: 2. :3 :6. RN :8. 23 mo. RN 3. RN .. «5:3 8. 3n 8. 3m .3. can we. 9% 8. 3n ham 355 8. 3.: 8. 3.: 8. 3;: 8. 8:: 8. 3.: __ $2558 8. 8. co. .. Snug No. 8. No. No. 8. __ 858 8. no. 8. 8. S. 2:93 no. no. no. 8. no. 2:25 .5. .8. .5. .8. .3. . 255 no. mo. 3. 3. mo. ._ 258 8. 8. 8. we. 8. 2.8 :55 3.2 _. 58.58 2 . i 32 $858 mm . M £3 mafia—ME S . 2. mm. mm. mm. .. 5:84: qm.~ .2: 3d .8.“ «mg 3 55555 3.5 Na __ 05.2% 8 . S N .m __ Edfiz 8 $3 3 . q .. ofiqfi: Sod: SJ __ bags? 3.8 0.2 9.8 93 ._ H.255“; 353 98 262 c.c~ ._ saga 3.? o.~ 3.2 o.~ ww.? c.~ 3.3 o.” 3.0m e.w _. 225 :3 EA 353 SA $.an Ga :33 RA 3:; SA ._ $823 3.8 8d 3.: 8g 3.3 oc.~ 3.3 c.~ afiwm o.~ _. 283 £53 3..» $63 3..» $.23 :3 8.9; 3; 3.2m 33 86¢ Epzfi «a 5 8 m: «N .32 H33 :2 H23 .3: 3:: gr. H33 .5: H23 5.5 8&6QO .AAuv wfiwwov mmlmnma .mnmso .aoamwm awnucmo mauam HH zuomwumu no Hw>wq monsomwx can muostHm m=Hm> Hmdfiwumx .mq.m manmh 225 .vuuamsoo "wousom mn.mn NoH.~m No.m~ mm.m~ New.~m .. msamzu mom¢g gm on mxo<\zm=pmx a~.o~o Ho.mqn.n Hm.mm~.~ MH.mmq em.~mm on amzommom Hzpoz< mnq.m omm.a~ om~.HH “HnH Nmmm muse: xmux mwom .. 05:05.8 ANm.ooev we.nwm ANnmv Ausmv ANm.~wv mnam .. >m<¢mv o~¢ A>v wu. we. owe «a. owq .. hmmnv won «H. men .. omxm¢q Aqmav Na mac. HmN mac. HmN we. HnN co. HmN upon mammUA m; HU>UA E HU>0A m; H0>0A m; HU>UH urn: NON—uomwm uwzafiuaoo mq.m wafima 226 0.0mm mo.mmo.n £43; imam 3.5m __ 2512 MS. $63 $63 3o. moo. .. 9545 No.8” 3.3m; 2.2m; «.mma 24m” .n 355» moo. 25?; no.5“; moo. moo. _. 23.5 $5an $43; mmauné ogoHJ oiomfifi . 25.5 to. 3684 no.8”; one. to. Pie: Ego oé _. zfifimoo 32 zfifimou om.~ mum BEE 8.2 8.3 .. 2.2mm oo.o~ oo.o~ _. 2.2mm o.~ oo.fi 8.2 o.~ o.~ .. 55.8 m.... 3.3 3.2 mg fin 32 $330 .. 895 So. So. So. So. So. .. 8mg 2&0 mm.omm RA: m~.~ no.8 .. £228 moo. moo. moo. moo. moo. _. 2.0.5 3.8 363 3:8” 8... 3.2 .. 958m 3.3 moo. moo. 3.: 5.3 ._ Ema moo. 3.8 3.2 moo. moo. ._ 3:28 So. moo. moo. So. So. _. 896 8.2 moéon :45 mm.2 mm.2 .. mime» moo. moo. moo. moo. moo. _. 398 3.3.. 2.2%; “.mmo 26$ 3&3 __ Sodom moo. moo. moo. moo. moo. .. 896 2.2 $52 2.2: .12 $3 __ 358 330 9: 3.3 9.: $18 9.: flooo 9.: 3.3 9.: «2.8 $335 on 5 UN on H.— HU>UA E Ham/04 E HU>N1~ fl H0>0A = H05..— UHGD Noun—006M .omummma .mamno .zowmwm Hmuuawo .wEmh HH .Comwuwo .3: van HH mummsmv #055 mousomom uanmfiwmfi yawn: mcaam Sham Esaunumo we bgm 23d 933. 227 mo. no. mo. . mo. no. 3 «85 2.3 one: omom 2.2 3.2 :25 3.: 363 3.8 3.: 8.2 o3 o5: om. 2. E. on. om. _. 28m $4 $4 $4 $4 $4 .. 8.63 no... 8... S... 8... 2... fl o2 8.8”“ 3. 3. i. Z. Z. 555 23mm 362 3.2:; 35% 26mm on.mmm .5 go 5.3 2. m: mo; mod mo.~ .. Burn mm... mmfi 8.: mm; mm... : SEE 2.2 93 No.2 2.: 9...: __ tag 2.? a.m.: £33 9...? 2.2 _. Em NT: 3.2: $52 3:: 2:3 _. SE; 1: H: G.“ $.N H: ._ 3% 2.? 3:5 34: 8.? 2.3 DEN: on... 33 on... 8... 85 H 3% 2.: 2m: mod: 2.? 2.: fl oz 35: oh... mo... mom 2... E... _. 3% Non. So. woo. «no. go. ._ Ego mom. zoo. «mo. .33 8o. _. 95.5 8.8 «mo. «mo. o.$ oi. .. 2.3m coo. moo. moo. moo. go. _. 2.3m «:2 m2. m3. ZAS $.52 .. 224m nooo. SH. :1 mooo. mooo. _. 2.2.5 SH. SH. 3;: if: ._ 2.3... 3.5 mom: 335 in... 230 3.2%; 326 odomn 38o mammo #5.. 2.2.; wu : H0>0A : Hw>wa : Ho>ou : aa>oq awn: wouaowwm .vmncwunoo p<.m wanna 228 DUUflhm Evan—m " E vegans—cu ”consom . = ~ ma 3.an : 28> m." «e.w: : «Sky 5 :4: : gr ea No.oo = Na 89 o8. 8o. o8. 2o. = ofia moo. oHo. HHo. moo. moo. = mompa moo. ooo. moo. moo. moo. : moo>m «Houo mm.n moouo mm.n moouo mn.m moouo 5m.m umouu mm.m moo mommy wd E Hv>wA E HU>QA E H0>0A E H0>fld uHCD ”UHSOMUE .ouaofiuaoo om.o wanna APPENDIX C 229 vuuanaoo “00»:Om EMNIEMN ,Iouooncunuuuam Iu>wmmaulonawz uuu< umufluw Iaowlfiuhcuou Im>nmmuoluuwuz EMMHEWN .ocou.:ouunu~m u>ummuununauz mmmmmmmw nn.wn nm.mn m.mq oa.nn w~.n< na.o< n.0m «a.mn mm.w< uamau Manna aquH cu uopuq curd: uo oval: eq.0q A¢.0n No.0m mo.n¢ h~.nm no.0n ao.mq nw.n~ ~n.on mauuu Iuom unsoa< Nun Nan o.mo~ muse: ouu< yum muaoz uoawa mwvou mundane hum ayauox muvmu use: an: ham unauum o.@~m w.o¢¢ a.maq o.uum o.mn¢ oo.Ho¢ n.5mn cm.~om N.<~¢ n.mmn muoomzo on“; we m>= m¢.¢qm oo.o~n eo.mmu nn.nm~ ~.oo~ m.aw~ nm.oo~ N.oo~ m~.~m~ ~m.«m¢ mn.nco wa.~mn nm.n~a «no n.mcm w.~qm n.4na o. i no.ncqn 5.5mnm mm < 22¢ mzoqu mmomo .mcowmum HH< .HH ammnm .msumm H zuommumu GOHmawmxm wuusomwz w>HumcthH< .H.Q mHan APPENDIX D M .muuu< .mUHumHumum we smmusm Hmuuamo .ONmH dOHumHaaom mo mamamo mnmsu "mounom 0 mm H NmH mnH HmH H om co <5 mm ow mm om Nmm m mmou¢ nOHumHmadam mm N m mwm omm «No wH No NHN NNH me em on nH Now 2 Ho ummm 0H I‘I H mm mm om m «H @H HH m 0H NH H mm m AuoHHumHQ sooomu Hugo: H o H H o mm um>o moumo moumm om-mH oHIoH m-m ouH “mum H ammo uwsuo maom vQHOHQBmcD vw%0Haam vmw< a mo BOme HH< muH>Huo< UHanoum kuOH Awummw cHV wmw< xmm %uHHw00H .Amuwnaaz GHV COHmmM owmnHuu< UHEououw H33. Amuwww a5 3? ‘1'!“ .N .9 QHDNH no .Amuunazz sHv :onum Huowomo NommH mamnmu :oHumHs m 2132 .muou< .mUHuwHumuw mo amwusm Hmuuamo .oan .dOHuMHdmom we mamcwo mange "muuaom H ¢ H Hon mmNN wwwm HQN Hmm nQMH HmNH NHw mNm non omH SUENQM cmH ohm mNMN Nan «NH own “QMH HMON ooh QNw can moH 9:5: o onummuom mcHuasm . um : m H < H H uom $8 :5 monmo 3-2 oNIoH «HIS mum 7H 83H uw%0Hnauas vszHmam a mH a mw BOHoMI [I 39.. .3383 muH>Huu< UHeoaoom Hmuoa Hmummw aHv mmm< Illilllllllltll .Amum naaz aHV :0mem muHo> "oan mamdwo aOHuMHnmom .n.n mHamH 2133 .muuu< .moHumHuwum mo =mmn=m Hmuuamu .Oan .aOHuMHdmom mo mamawu manna "wousom HH II II w ON .3 H OH NH OH OH mH mH q mm m 93:. counuwuwasfim OH m m Om Nq mm w m HN NH pH HH O m mm 2 we umwm o It H mm cm we OH OH mN HH OH ON MH q NHH .m aomuprm< mH m H m6 me no q mH ON ON OH OH ON O NH 2 Imamoxww cm I1 m ow mmN me Hm ow OHH Hm we Hm R mH mmm m mdmwxwm mo ON o omH OON mmm mm mm. mOH Om we mm cw NH mHo z 95%.: o humwyom 9:355 .32.: .2»... 5 H38 “um” .26 «on? .273. 3.2 «HIS mym .7H :3. H 8? Hmnuo wan: vaOHafimaD @93ng cwm< a mo 30.15 Hi» zuH>Huu< UHEocoum Hauow “mummy. a: wwwHw xmw zuHHmUOA .Amuwfifisz GHV msmwxwm .COHmmM :uwuwmm "Oan msmnmo :OHuMHamom .<.Q anMH 234 .muuu< .mUHumHumum mo smuusm Hmuucwo .ommH .EHumHsmom Ho wH.—mummy MEN—#9 "QUHHHOm on mN II «mm NOm NHH Hm OO nOH OOH On NNH NHH RN «mm m wEOH an N O OQN OON «on Na mm NmH NHH ow OHH ma AN mOO E IamoEow H m I: Hm «m Om «I OH mm OH HH OH «N c OHH m moHHw¢ ummm OH H N Hm mm on m NH OH OH mH ON mN N mHH 2 mo umwm «H m 1» mOH nNH NmH OH «O OO Nm «m me mm O «mm H OH H O «m OHH an OH «H mq on me mm om N mHm z =amzx< 95%: o znumwuom muHua=m Hafiz Homo 5 H33. New.” ~25 3-3 3-2 3-2 3.3 mum .1 :3. H 8?. Hmnuo mac: OwNOHmamup owHOHmam me< O OO BOme HH¢ qu>Huo< oHEocoum Hmuoe Hmumm» :HV mmm< xmm zuHHmuoq .Amuwnasz :HV =Ed33< moEou .GOwax Hauudwo HONmH mam:wo coHumHjaom .m.o MHONH _—:.. ,‘J.’ __..,_, .,,_HH APPENDIX E 235 Table 5.1. Marginal Value Products and Prices of Seasonal Product Inventories (By Region). Eastern Volta Central Brong-Ahafo Ashanti MVP Prlce MVP Prlce MVP Price MVP Price MVP Price MZOPl 12.83 12.2 13.82 12.2 12.93 12.2 10.6 10.0 10.55 9. 4 MZOPZ 15.19 14.45 15.72 14.4 15.19 14.5 13.4 12.75 14. 13 33 MZOP3 12.79 12.26 14.46 12.25 12 79 12 26 9.6 9.25 10.77 10 33 MZOP4 8 39 8 39 11.66 .40 13 37 8 39 8 14 5.4 10. 2 0 5 l4 3 14.98 9.70 13 74 4 8 37 5.04 11.18 5 61 MZOP6 14 7 11 32 13.11 12.2 14 12 11 32 8 4 11.86 8 71 MZOP7 l4 8 14 8 14.0 4.8 14 28 14 28 8 3 8 73 12.7 12 7 PLAOPl 46 44 65 .52 46 .44 75 71 45 42 ‘ PLAOPZ 47 45 62 .55 47 .45 69 66 45 43 PLAOP3 71 68 79 .69 71 .68 74 71 71 68 PLAOP4 95 95 1 06 1.0 97 .95 81 78 77 77 PLAOPS 1 01 1.0 1 01 .96 1 03 1.0 87 85 55 55 PLAOP6 .75 .75 .98 .89 .76 .75 .62 .62 .47 .47 PLAOP7 ' .48 .56 .45 .45 .56 .42 .58 .52 .32 .32 CASOPl 4.47 4.25 6.14 5.2 4 5 4.25 .4 .25 3.39 3 2 CASOPZ 3.47 3.3 5.42 4.6 3 47 3.3 .5 .37 3.79 3 6 CASOP3 4.55 4.36 6.3 5.1 4 55 4.36 .7 .6 3.93 3 77 CASOP4 4.38 4.38 5.3 4.4 4 5 4.38 .6 .5 2.5 2 5 CASOPS 3.52 3.52 5.9 5.1 3 6 3.52 .8 ..75 2.0 2 0 CASOP6 3.85 3.85 5.5 5.2 3 9 3.85 4.0 .8 2.2 2 22 CASOP7 3.64 3.66 5.7 5.7 3 64 3.64 3.9 .51 2.1 2 1 COYOPl 9.64 9.17 5.6 4.6 9 72 9.17 .49 .96 7.95 .75 COYOPZ 4.1 3.9 5.26 4.1 4.1 3.9 .28 .83 10.51 10.0 COYOP3 6.15 5.9 6.23 5.3 6 15 5.9 1.8 1 .32 10.42 10.0 COYOP4 6.5 6.5 6.58 5.4 6.67 6.5 1 .3 1.0 9.83 9. 3 COYOPS 6.4 6.4 6.88 5.6 6 56 6.4 .46 .25 10.33 10.33 COYOP6 5.81 4.2 6.03 5.9 8 13 4.0 .68 .68 6.9 6.9 COYOP7 6.25 6.95 6.46 6.46 6 95 6.25 .2 .58 6.33 6.33 YAMOP1 44 46 42.29 36.7 32.4 44 83 42.29 3 .06 3 .25 34 87 32 69 YAMOPZ 51.2 48.7 9.4 34.2 5 .2 48.7 3 .63 3 .75 34.69 33 YAMOP3 58.4 56.0 6.2 41.3 5 .4 56 3 .7 2 .49 36.49 YAMOP4 37.75 37.75 37.11 33.2 38.72 37.75 2 .85 2 15.3 15 33 YAMOPS 39.75 39.0 34.66 33.1 40.0 39.0 2 .79 2 .25 7.0 YAMOP6 42.0 35.0 38.59 38.4 42.0 35.0 2 .87 1 .34 27.16 21 7 YAMOP7 45.0 45.21 41.5 40.6 45.2 45.21 3 .0 3 .0 9.2 29 2 VEGOPl .14 .13 .13 .09 .14 .13 .13 .13 .13 .13 VEGOPZ .12 .12 .12 .11 .13 .12 .12 .12 .12 .12 VEGOP3 .11 .11 .11 .09 .15 .11 .11 11 11 11 VEGOP4 15 .10 .12 12 .15 .10 106 O9 14 O9 VEGOPS 15 .11 .12 ll 16 .15 09 14 1O VEGOP6 16 . 13 .13 11 16 11 15 12 VEGOP7 16 14 . 14 09 16 15 129 126 14 14 Source : Computed. APPENDIX F APPENDIX F The APPEX-l Reference Manual supplied by the Control Data Corporation (CDC) to be used on CDC 6500 computer and the Harsh-Black control program were the primary source for deriving the linear programming solutions for this study. Basically, the APEX-l is an optimization system, its main function being to optimize MPS formatted linear models to either maximize gains or minimize losses. In this study, the data generated by the Harsh—Black optimiza~ tion system were using an appropriate subroutine, trans- formed into a linear model that the APEX—l system could analyze. The following files, representing the matrices for the representative farms in the three phases of analysis are contained on Tape 64, APLIB or Applications Programming Library, Computer Center, Michigan State University, East Lansing, Michigan. 236 ‘ .4“..-_‘._.. a 237 Appendix F (Continued) Library File Name Position (LFN) R3F2M08 Random 01 R1F2M8 ” 02 R5F1M9 " O3 R1F2D8 ” O4 R4F2MO8 (Abandoned) ” 05 RlFZJl ” 06 R3F1R1 ” 07 R2F1Rl ” O8 R3F2Jl " O9 R4F2Jl ” lO RlFlJl ” ll R4FlRl ” 12 R5F1Jl " 13 R5F2Jl " l4 R5F2Ml ” 15 R5FlMl ” l6 R2FlMl ” l7 R4FlMl " l8 RlFlMl ” l9 R1F2Ml " 20 R2FlJl " 21 R3F2Ml " 22 R3FlJ1 " 23 R4F1Jl " 24 R4F2Ml ” 25 R1F1D2 ” 26 R5F2G2 " 27 R5F2M8 " 28 R3F2M8 " 29 R4F2M8 ” 30 R5F1M8 " 31 R1F2MO8 " 32 Where Rl. . .RS represent region 1. .region 5 and F1 and F2 represent Category I and Category II To obtain files off the tape will control cards and procedures: farms, respectively. require the following 238 Appendix F (Continued) II. Obtaining Files Off the Tape and Range Report PNC (or program name card) JOB Card PW or Password ATTACH, APLIB, APLIB. APLIB, TT64, R*(Insert LFN here) = OLDPL. UPDATE, C = Tape l,N. CATALOG, NEWPL, (PFN), ID = ATTA- HAL, CONTROL, 0 = OUTPUT, CC = 9. HAL, *APEXI. AUTORFL, PART. RFL, 42000. APEX, SOLVE, MAX, RANGE, L, TER = 300, LOG = 50. 6 7 8 9 Obtaining Listing of Matrix and Linear Equations and Range Report PNC JOB Card PW DMPX (OFF) ATTACH, OLDPL, _(PFN) , UPDATE, C = TAPE 1, F. LISTTY, I = TAPE 1, W95, HAL, CONTROL, 0 = OUTPUT, CC = 9. HAL, 7"APEXI. AUTORFL, PART. RFL, 42000. APEX, SOLVE, MAX, EQ, RANGE, L, TER = 300, LOG = 500. 6 7 8 9 BIBLIOGRAPHY BIBLIOGRAPHY Adams, S. N. 1962. HSoils and Manuring," in J. Brian Wills (ed.), Agriculture and Land Use in Ghana. London: Oxford University Press. Behrman, J. R. 1968. Supply Response in Traditional Agriculture: A Studygof Four Major Crops in Thailand in l937-63. Amsterdam: North-Holland Publishing Co. Bentsi—Enchillak. 1964. Customary Land Law in Ghana. London: Seveet and Maxwell. Channareddy, V. 1967. ”Production Efficiency in South Indian Agriculture,” Journal of Farm Economics. Vol. 49, November. Connor, L. J. 1954. ”Long—run Adjustments for Farm Opera- tors in a Sparsely Populated, High Risk Area of the i Great Plains.” Unpublished Ph.D. dissertation, Oklahoma State University, Stillwater, Oklahoma. Day, Richard H. 1963. "An Approach to Production Response,” Agricultural Economic Research. Economic Rsearch Service, l4:No. 4, USDA. 1963. "Use of Representative Farms in Studies of Inter—regional Competition and Production Response,” Journal of Farm Economics. 45-14 38:1445. Driebeck, Norman J. 1969. Applied Linear Programming. Cambridge, Mass.: Little, Inc. Edwards, Clark. 1958. "Resource Fixity, Credit Availability 3 and Agricultural Organization.” Unpublished Ph.D. “ dissertation, Michigan State University. Enchill. Ghana Land Law: An Exposition, Analysis and j Critigue. University of Ghana, Law Department, 109 et seg. Hart, Neil E. 1967. ”Organization and Structure of Producer 3 Units of Farm Products," in Implications of Changes on - Farm Management and Marketing Research. CAED Report 29, Iowa State University, Ames, Iowa. ‘c-‘AJ. - 239 240 Heady, E. O. 1961. "Uses of and Concepts in Supply Analysis,” in Agricultural Supply Functions. Edited by E. O. Heady, et a1. Ames, Iowa: Iowa State University Press. and Swanson, Earl R. 1952. Resource Productivity in Iowa Farming. Research Bulletin, Agricultural Experiment Station, Iowa State College, June. Heyer, Judith. 1971. ”A Linear Programming Analysis of Constraint on Peasant Farms in Kenya," Food Research Institute Studies in Agricultural Economics, Trade gpd Development, Vol. 10, No. I. Heidhues, Theodore. 1966. "A Recursive Programming Model of Farm Growth in Northern German," Journal of Farm Economics, Vol. 48, No. 3, Part 1, August. Helleiner, G. K. 1969. "Economic Constraints on Socialist Planning in Tanzania.” Unpublished Seminar Paper, January. Hill, Polly. 1963. Migrant Cocoa Farmer of Southern Ghana: A Study of Rural Capitalism. Cambridge: Cambridge University Press. Hopper, David M. 1965. ”Allocative Efficiency in Traditional Indian Agriculture," Journal of Farm Economics, 47 611- 624, August. Hudson, Estel H. 1972. Farm P1anning_Manual. Legon Extension Bulletin No. 16, University of Ghana. Jones, William O. 1960. ”Economic Man in Africa," Food Research Institute Studies in Agricultural Economics, Trade and Development, Vol. 1, No. 2. Johnson, Glenn. 1968. Food Supply, Agricultural and Eponomic Development. Consortium for the Nigerian Rural Development, Working Paper No. 8, Department of Agricultural Economics, Michigan State University East Lansing, September. . 1971. "The Quest for Relevance in Agricultural Economics," American Journal of Agricultural Economics. Vol. 53, No. 5, December. ,et a1. 1972. Korea: Agricultural Sector Analysis and Recommended Development Strategies, 1971-1985. Department of AgricuItural Economics, Michigan State University, East Lansing, Michigan. ifl -~-——=~.__ —4,,_ my -. . . .. . 241 Lard, Curtis F. 1963. "Profitable Organization of Repre- sentative Farms in Lower Michigan and Northeastern Indiana with Special Emphasis on Feed Grains and Livestock," Unpublished Ph.D. dissertation, Department of Agricultural Economics, Michigan State University. Lee, John E. 1966. "Exact Aggregation--A discussion of Miller's Theorem," Agricultural Economics Research, N0. 2, Vol. 18, USDA. Martin, Rod J. and Plaxico, James S. 1965. "Poly—Period Analysis of Growth and Capital Accumulation of Farms in the Rolling Plains of Oklahoma and Texas," Economic Research Service, USDA, Bulletin No. 1381. Mellor, John W. 1965a. "The Subsistence Farmer in Traditional Economies," ADC Seminar, East-West Center, Honolulu, February-March. 1965b. "Towards a Theory of Agricultural Develop- ‘ment," Paper prepared for SSRC symposium on Agricultural Development, University of Chicago, May. Miller, T. A. 1961. ”Aggregation Error in Representative Farm Linear Programming Functions," in Agricultural Sppply Functions. Edited by E. O. Heady, et a1. Ames, Iowa: Iowa State University Press. Ministry of Agriculture, Ghana. 1971. Proposal for a Maize Development Project in Ghana. Accra. . 1972. Operation Feed Yourself, Instruction No. II/1/73. Accra. . Report on Ghana Sample Census of Agriculture, 1970. Vol..1. Accra. Nakajima, Chihiro. 1957. ”Equilibrium Theory of Family Farm," Oska Daigaku Keisaigaku. July. "The Subsistence Farmer in Commercial Economics," Paper presented at the ADC Seminar on Subsistence and Peasant Economics, East—West Center, Honolulu, Hawaii, February—March. Nathan Consortium for Sector Studies. 1970. Grain Marketing, Trangport and Storage. Vol. 1, Principal Report. Prepared fOr Ministry of Finance and Economic Planning, Accra, Ghana, April. Norman, D. W. 1973. "Economic Analysis of Agricultural Production and Labor Utilization Among the Hause in the North of Nigeria," African Rural Employment, Paper No. 4, Department of Agricultural Economics. Michigan State University, East Lansing. 242 Nyanteng, V. K. 1972. The Storage of Foodstuffs in Ghana. Technical Publications Series No. 18, Institute of Statistical, Social and Economic Research, Legon. , and Van Apeldoorn, G. J. 1971. The Farmer and Marketing of Foodcrops, Technical Publication No. 19, Institute of Statistical, Social and Economic Research, Legon Ogunforowa, Olahisi. 1972. ”Derived Resource Demand, Pro— duct Supply and Farm Policy in the North Central State of Nigeria." Unpublished Ph D. dissertation, Iowa State University, Ames, Iowa. Ollenu, Nii Amaa (Justice). 1962. Customary Law in Ghana. London: Seveet and Maxwell. 1971. ”Problems in Land Administration and Development in Ghana." Paper presented at a Seminar on Land Problems and Development in Ghana, Department of Land Economics, University of Science and Technology, Kumasi, October. Schaller, William N. 1962. "A Recursive Programming Analysis of Regional Production Response," Unpublished Ph.D. dissertation, University of California. Schultz, Theodore, W. 1964. Transforming Traditional Agriculture. New Haven: Yale University Press. Sen, A. K. 1966. ”Peasant and Dualism with or without Surplus Labor," Journal of Political Economics. 74, October. Singh, 1. J. 1968. "Recursive Programming Models of Agricultural Development,” Social Systems Research Ipstitute, SMF 6836, University of Wisconsin, October. Stern, Joseph J., and Roemer, Michael. 1972. Pgoject Appraisal, Notes and Case Studies (Part 1: Notes). Ghana: Ministry of Economic Affairs. Stryker, Dirck J. 1972. "Vent for Surplus Growth in a Colonial Economy: A Tentative Model.” Discussion Paper No. 153, Economic Growth Center, Yale University. Tax, Sol. 1953. ”Penny Capitalism,” Smithsonian Institute of Social Anthropology, Publication No. 16, Washington, D.C.: U.S. Government Printing Office. Tyner, Fred H. and Tweeten, L.G. 1966. ”Optimum Resource Allocation in U.S. Agriculture,” Journal of Farm Egonomics, Vol. 48, No. 3., Part I, August. 1 243 United Nations Development Programme. 1969. Crop Storage. PL SF/GHA 7, Technical Report 1. Welsch, Delane E. 1965. ”Response to Economic Incentive by Abakiliki Farmers in Eastern Nigeria," gournal of Farm Economics, 47:900-14, November. Wharton, Clifton R. 1969. Subsistence Agriculture and Economic Development. Chicago: Aldine. Wheeler, Richard G. 1967. ”Research Orientation in Economic Development with Special Reference to Brazil,” Agrippi; tural Economics Research, Vol. XIX, No. 3, July. Wise, J. and Yotopoulos, P.A. 1969. "The Empirical Content of Economic Rationality: A Test for a Less Developed Economy,” Journal of Political Economics, 77, November. Yeboah, J. E. 1968. "Credit for Production,” an unpublished Seminar Paper at the Agricultural Conference organized by USAID, Ghana. Yotopoulos, P.A. 1967. Allocative Efficiency in Economic Development, A Cross Section Analysis of Epirus Farming. Center for Planning and Economic Research, Athens, Greece. , and Lau, L. J. 1973. "On Modelling the Agricultural Sector in Developimg Economies: An Integrated Approach of Micro and Macro—economics,” Food Research Institute Mgmorandum, No. 148., June.