I . '.~‘. I 'I IJJM‘” "'I‘I'I'IJ- ' . I ‘I? '1’"va "' "I f «’59—. .. - III‘V I 'I . ' V . I V {LIL VI‘; IV ‘01::IL‘ 1 “ " V: ‘: V I " " I I ‘ ' IV "I“. I. 'I . _I I'VE? ii" ' . I ”I : II'VV. IVI‘VI‘V'ILVVV VI -V I" IV ': "E 'V VI‘ IV V‘ IVV VI ::‘,IIVVII I VI . 'V'III 3‘ II<=J'V‘1I‘“IA" Mt VVV (HI III VVIVVVb V IVV R‘i VVVVVV IV E . V ”HIIIIOIVQHH' I’V V V VI VI: “131pr I I V ”I” 414..“ .‘fI V . I I‘V‘VI'I "‘VII‘IIifV ‘I' “"I‘Vl‘I‘ HIV VVVIVI' 399:” II VIVVI. ‘IV‘M‘ a! g, I dl “ 1I WWII!“ , .. ‘ "I”.IV EIVVIII I‘IV'I “‘IL‘JV ‘3‘} h! IIV"V‘:IV VI 5:1ij III. VVVNUW ":IfiIfiIVVI’III "’I‘hr' I“: ‘II ‘V: 589""?! 2“ \‘J‘ L 1‘2" ' ' "I -"3”'\I‘I1§1313I“I"IVVVI,VIV I VVVV lIVI‘iI‘IVI IIIII'IVIM m ' ' I15“. IIVII {‘1} V , II ' _ .l _ . .‘ ‘ V"IEI'IVIVA‘HV'I1VIII\;VI‘I 1.1‘. I" IIWI‘” ""4 II III" II “VI A, I“ "*‘I”:t , I ’1 In I: I III 3&3? “"“ I:II*"*1‘III'II"§ " V IIIIIV IIVV '~ “1‘1, ' " ‘ ' ' ' ‘II‘VVI 3331‘! 3% IV WI 1. ’L . 131‘: A. A‘: MI 4'37‘ may M {giggu AC. ‘ VI'I VI” " VIIIIIIII «J's.— r ‘f w? ' . 11:7,: L‘IMLFI‘Vhl {MN}. I . IVVV “hilt“! 3V «V '» —“ . I .. ‘ AA,’ 3—: 1:31 I 7’ :M' £29: '.—.. $fiq§§ P - LVN"- IN [IV- l‘, ‘V HI AV." . V IILL‘IMHI IVV V I ~ H; ' {1’1 :Vm . V' IV”: 'C :’;p":”.lI 3;!- ‘g‘: PM“; “'41; V“.II 1l = p14" a I III ‘ ' w 1““ “I'vwg ’rw‘IViV I ”Ii-Vin #312: ‘4 ”El kI A V . i ,: VII: 9" 'l'jV V “ V‘t‘ :’ ’ ‘ :' f“ My: ‘ 3‘ . 1 1‘33 _, ’IIIIIIM “31 I; II MI 3th I: 1’ “ ‘I -' It -.I r ‘I I “I? I, 1 "I54“ A ”I334"? f III ' i II My 3' " 3.59? : ‘5'!" 31'3“: ’ q 4* “ "sf-“II" i2 I .‘ IA'IIA .. :.I.. .?VII 5"“ h. I v ‘ V ‘ r I r“ J .,-V II I}. I V, ,1 I V 111.: :3' - ‘ 3 .73: HIV ; ‘ $9} ‘15 ‘l.. b h 51:“ $1 11“!“ $3qu 1336?. ' 4 "'3'? If‘ileV ‘F‘itfi 3.3“ 'l M3591! kiwi ‘I'I"‘II31‘[§V§I§I';5V” IIHIII QIKIIIILIV *iViI {$13 “V I "I ‘ ‘ ‘l tn H $5“! “1% «5% I' {Hi 1.“ It? :JVIIIQV IV V .V II III‘I‘I‘II‘I VViVVVIzI "IL I Hem“ ‘ I .V III ‘33?“ I: V .. V (’m‘mf" ,1. . Mistacortifythntthe dissertationéitifled INCREASING THE PRODUCTIVITY OF RICE BASED FARMING SYSTEMS OF THE PAKISTAN PUNJAB presented by Pervai z Ann‘r has been accepted towards fulfillment of the requirements for 1985 Agri fltural Economics __P_h-_D_-_ degtee in ajor professor RobertDale Stevens Date—W 535‘ MSUl: uAfl‘irmnttnAm‘oa/Eqml Opportunlq [Initiation 0-1277! IVIESI.J RETURNING MATERIALS: Place in book drop to LIBRARIES remove this checkout from :1-lllc-au-L your record. FINES will be charged if book is returned after the date stamped below. INCREASING THE PRODUCTIVITY OF RICE BASED FARMING SYSTEMS OF THE PAKISTAN PUNJAB BY Pervaiz Amlr A DISSERTATION Submitted to Michigan State University in Partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Agricultural Economics 1985 ABSTRACT INCREASING THE PRODUCTIVITY OF RICE BASED FARMING SYSTEMS OF THE PAKISTAN PUNJAB By Pervaiz Amir The two major cereal crops rice and wheat grown in the Punjab Rice areas are facing low net returns due to the increased input cost. With the gradual removal of input subsidies the present dominant rotation of Rice—Wheat is no longer profitable. The country is also facing severe balance of payment problems due to increased oilseeds imports. This calls for an investigation of constraints to the production of rice and other minor crops. The need for identifying alternative profitable crop technologies is also felt. The specific objectives of this study were (I) to study resource productivity in the rice based farming system in Gujranwala; Sialkot and Sheikhupura districts of Punjab. A particular objective is to provide a farmers View of the Kharif work situation; (2) to identify technical and socioeconomic factors which constrain increases in production of rice and competing crops in the Kharif season in the three districts; (3) to analyze through linear programming models the economic implications of increasing multiple cropping intensities with new crops such as sunflower and early maturing mungbean and changes in product price policies; and (1+) to draw implications from the study for research, extension and government policy. In 1984, 240 farmers were surveyed in 15 villages in districts Gujranwala, Sheikhupura and Sialkot. A. multistage sampling approach was used to select the respondents. The study was restricted to the heavy well drained soils to explore crop substitution. Two representaive farm models were constructed for small farms to analyze cropping patterns in the rice areas. The results show that farmers are less concerned with the technical constraints as perceived by researchers but face serious socioeconomic problems. Problems with canal water, public tubewells, impurity of inputs, high input costs etc were identified to be the major socio economic constraints. The Punjab farmer has started to adjust to cost increases by reducing area allocated to rice. The programming exercise revealed that sunflower and mungbean can easily be incorporated in the rice based system. Sunflower will compete with wheat and rice. The main limiting factors for these crops are seed, assured procurement, extension demonstrations. The new rice variety PK-258 shows promise. Given the government policy to discourage IRRI by price reductions we may see a gradual shift towards newer crops. This thesis is dedicated to my parents Dr.Amir Muhammed and Fazilat Amir. Both have borne great hardships to make this dream come true, I hope they realize some of its fruits. 1'1 ACKNOWLEDGEMENTS I would like to express my greatest gratitude to my major professor and thesis supervisor Dr. Robert Dale Stevens for guidance throughout all stages of the graduate program. I would also like to thank him for visiting me in Philippines and Pakistan during the field work stage of this dissertation. Professor Glenn L. Johnson has inspired me greatly during my graduate course work and as a member of my dissertation committee. 1am also greatful to Professors Garland Wood , Lester V. Manderscheid and Paul Strassman for serving on my guidance and thesis committees. Dr. Warren Vincent provided valuable input to the design of this dissertation, before retiring from the department in 1981+. Dr. Derek Beyerlee (CIMMYT Regional Economist) provided valuable guidance during my field work in Pakistan. 1 would like to express my appreciations to all of these persons. A number of agencies have supported this research. The International Rice Research Institute, Philippines provided a research fellowship for which I am greatful. I would like to thank Dr. John Flinn who supervised my work at IRRI and The US Agency for International Development in Islamabad for generously providing financial assistance to cover the cost of research credits back at Michigan State. ' The Pakistan Agricultural Research Council (PARC) in Islamabad provided the necessary logistic support in Pakistan. Dr.M .Yausaf (member plant sciences) coordinated my stay in Pakistan and made it a memorable one. Dr. H. Hasnain (Member Social Sciences) at PARC also helped in several ways. iii The PARC rice program staff deserves special mention for allowing me to participate in their on going research. Drs. M. Akbar (now with IRRI), M. Ashraf and M. Avesi introduced me to the research issues in rice. Similarly, Dr. M. Majeed (Director General) at the Punjab Rice Research Institute , Lahore, very kindly allowed me access to information on their cropping systems experiments. Dr. S. Mujtiba Naqvi at the Nuclear Institute for Agriculture and Biology, Faisalabad, provided me an excellent opportunity to review the work of the pulses program. Dr. M. Illays (pulses scientist) was kind enough to allow me use of uncompiled information on mung bean varietal trials. This is greatfully acknowledged. Finally, my wife, Faiza, and Daughter, Maria, for bearing all the hardships which go into the moulding of a Ph. D: Both have gracefully survived great neglect during this period. My parents and four lovely sisters sacrificed greatly to allow me this opportunity in life, I would like to thank them all. all. 1v TABLE OF CONTENTS {Egg LIST OF TABLES .................................................... vii LIST OF FIGURES .................................................. xvii GLOSSARY OF SPECIAL TERMS USED IN THE DISSERTATION .......... xviii CHAPTER I INTRODUCTION ......................................... 1 PERFORMANCE AND PROBLEMS IN THE PUNJAB AGRICULTURE. . . . . .......................................... 1 Conceptualizing the Yield Gap 6 Dynamics in Pakistan Agriculture ............................ 8 Multiple Cropping Systems in the Punjab ...................... 8 Rice-Wheat Rotation ..................................... '.10 ' PROBLEM STATEMENT ...................................... 11 OBJECTIVES OF THE STUDY .................................. 17 SCOPE OF THE DISSERTATION ............................... 18 ORGANIZATION OF THE DISSERTATION ....................... 18 CHAPTER 2 THE AGRICULTURAL DEVELOPMENT AND POLICY ENVIRONMENT OF THE PUNJAB RICE TRACT ................. 20 THE THEORETICAL FRAMEWORK ............................ 20 The Induced Innovation Hypothesis .......................... :21 Farmer Rationality and Response to Economic Incentives ................................................ 22 Implications of the Induced Innovation Hypothesis for the Present Study ............................ 23 THE CURRENT STATE OF AGRICULTURE IN THE PUNJAB PROVINCE .......................................... 25 i) The Kalar Rice Tract .................................... 26 ii) Cropping Patterns ...................................... 28 iii) Farm Size ............................................. 30 iv) Land Tenure ........................................... 30 v) Land Use .............................................. 30 vi) Irrigation Sources ...................................... 31 vii) Livestock ............ . ................................. 31 GOVERNMENT POLICY AND AGRICULTURAL DEVELOPMENT IN THE DEVELOPMENT IN THE PUNJAB ....................... 31 The General Policy Environment ............................ 31 Subsidies ................................................. 34 Marketing Policies ........................................ 37 Credit Policies ............................................ 39 Recent Trends in Product Prices ............................ 39 Estimates of Net Returns .................................. 1+1 SOURCES OF BACKGROUND INFORMATION FOR SURVEY DESIGN ............................................ 41 Pakistan Agricultural Research Council (PARC) ..... _ .......... 1+1 Punjab Government Institutions ............................. #3 Nuclear Institute for Agriculture and Biology ................. 41+ Other Sources ............................................ 44 Identification of the Important Issues ........................ 1+4 Technical Issues ........................................... 1+5 Socioeconomic Issues ...................................... 45 CHAPTER 3THE METHODOLOGY OF THE FARM SURVEY FOR IDENTIFICATION OF CONSTRAINTS LIMITING PRODUCTION INCREASE ..................................... 49 SURVEY DESIGN ............................................ 49 vi Objectives of the Survey ................................... 49 On Going Research ........................................ 49 1. Economic Evaluation of the Gujranwala Rice Maximization Project ............................. 5O 2. Rice Yield Optimization Program, Daska (Sialkot) ............................................. 50 Selection of the Survey Area ................................ 51 Sampling of Farms for the Survey ........................... 53 Sampling Frame and Selection of Respondents ................. 55 Design of Three Questionnaires ............................. 57 Questionnaire No.1 ........................................ 57 Questionnaire No.2 ........................................ 59 Questionnaire No.3 . . . .. ................................... 59 Pretesting and Validation of Questionnaires ................... 59 'Field Method ............................................. 59 Data Preparation and Analysis .............................. 61 CHAPTER 4 IDENTIFICATION OF CONSTRAINTS LIMITING INCREASED PRODUCTION OF RICE AND OTHER KHARIF CROPS -— FARMERS' VIEWPOINT .............................. 63 Overview SECTION I CHARACTERISTICS OF SAMPLE FARMS AND RESOURCE USE ........................................ 63 Farm Size .............. I ............... ' ................... 63 Tenure ................................................... 64 Fertilizer ................................................ 64 Water Availability .................................... , ..... 6 7 Tillage Operations ......................................... 67 Use of Insecticides, Pesticide, Herbicide ..................... 68 Credit ................................................... 68 vii Plant Density ............................................. 68 Cropping Patterns..............: ......................... 7O Planting Time for Rice ..................................... 7O Advantages of Growing Basmati vs IRRI Varieties ...... . ...... 72 Present Yield Levels ....................................... 72 SECTION II CONSTRAINTS TO INCREASING RICE PRODUCTION ............................................... 74 Technical Constraints as Perceived by the Researchers ........................................... 74 SECTION III KEY ISSUES RELATED TO LOW RICE PRODUCTIVITY IN THE PUNJAB ............................................. 78 Plant Density ............................................. 78 Varieties and Certified Seed ................................ 82 The Rice—Wheat Rotation and Soil Fertility Level .............. 82 Pest Management ......................................... 84 Price Structure ........................................... 86 Researchlssues.........................' .................. 86 Issues Related to Water and the Punjab Farmers ............... 87 Institutions ............................................... 90 SECTION IV SUMMARY OF FARMERS SUGGESTIONS FOR REMOVING CONSTRAINTS TO HIGHER PRODUCTIVITY ......... 90 SECTION V FACTORS AFFECTING TO THE PRODUCTION OF MINOR CROPS AND RELATED ISSUES ...................... 92 Reason for Leaving Land Fallow ............................. 92 Beginning of Change in The Punjab Rice Based Farming ............................................ 95 CONCLUSIONS .............................................. 95 A. Input Use ............................................. 95 viii B. Yields ................................................ 98 C. Other Findings ......................................... 99 CHAPTER 5 ECONOMIC ANALYSIS OF ALTERNATIVES FOR INCREASING THE INCOME OF RICE-WHEAT FARMERS IN THE PUNJAB ............................................ 102 The Study of Crop Substitution ............................. 102 Choice of the Analytical Technique ......................... 103 Approaches for Constructing Representative Farm Models ............................................ 108 The Choice of Two Representative Farms ................... 109 Sources of Input/Output Data and Quality ................... 111 BASIC INFORMATION ON TWO REPRESENTATIVE FARMS ...... 112 Farm One (Hafizabad) .................................... 112 Objective Function ....................................... 113 Activities ............................................... 114 Crop Yield Estimates ..................................... 114 Crop Prices ............................................. 116 Estimates of Unit Resource Requirements ................... 117 Estimates of Gross Margins ................................ 120 Structure of the Model .................................... 120 Minimum and Maximum Constraints ........................ 122 Farm Two (Daska) ........................................ 125 Objective Function ....................................... 125 Activities ............................................... 126 Crop Yield Estimates ..................................... 126 Estimates of Gross Margins ................................ 128 Crop Prices ............................................. 128 ix Estimates of Per Unit Resource Requirements ............... 128 Minimum and Maximum Constraints ........................ 133 Structure of the LP Model ...... . .......................... 133 CHAPTER 6 RESULTS OF FARM MODELS AND INVESTIGATION OF ALTERNATIVE CROP TECHNOLOGIES ..................... 136 SECTION I RESULTS OF THE BASE MODELS .................. 136 SECTION II ECONOMICS OF ALTERENATIVE CROP TECHNOLOGIES AND EFFECT ON THE FARM PLAN ........... 139 FARM ONE (HAFIZABAD) ................................... 141 FARM TWO (DASKA) ........................................ 145 SECTION III SENSITIVITY ANALYSIS WITH PRICES AND YIELDS ............................................... 147 IV POLICY IMPLICATIONS .................................. 157 Policy Implications of Findings Related for Rice-Wheat Farming ...................................... 157 Policy Implications of Research Findings Related to Production Constraints of Rice .......................... 158 Rice Price Policies and Effect on Farm Income ............... 160 Potential for New High Return Crops ....................... 161 The Effect of an Increased Support Price for Kharif Crops ............................................ 166, New Unexplored Technologies .............................. 167 New Variety of Rice ...................................... 168 A Hypothesis of Change ................................... 169 CHAPTER 7 SUMMARY AND RECOMMENDATIONS OF THE STUDY ..... 170 OBJECTIVES AND METHODS USED IN THE STUDY ............. 170 RESULTS .................................................. 172 RECOMMENDATIONS ....................................... 173 APPENDIX A ...................................................... 177 ' IIOOIOIIOICOIIICQOOOoICIIIOOOIlootuolnlltlozol xi Table 1.1 1.2 1.3 1.4 1.5 1.6 2.1 2.2 2.3 2.4 2.5 2.6 2.7 2.8 2.9 2.10 LIST OF TABLES Page Summary of Important Socioeconomic Indicators .............. 2 Yield of Major Crops in Pakistan (1975-84) .................... 4 Yield Gap of Various Crops in Pakistan (1982-83) .............. 9 Net Returns for Major Crops in Pakistan During 1974-81 (Rupees) ...................................... 12 Agricultural Export and Import of Major Crops (1972-84) ............................................. 12 Macro Level Efficienecy of Fertilizer Use on Rice in Punjab (1974—84) .................................... 14 Monthly (Rainfall Recorded in Gujranwala ................... 29 Distribution of Size of Farm Operating Units in Punjab in 1972 and 1980 (Percentage Farms) .............. 29 Number, Area and Average Size Of Private Farms Classified by Tenure in Pakistan (1972-1980) .............. 31 Land Use in Districts Gujranwala, Sialkot and Sheikhupura ........................................... 31 Area Irrigated by Different Sources in Punjab and Pakistan in 1981 ................................... 33 Provincial and Federal Government Subsidies on Selected Inputs in Pakistan 1975-1983 .................... 35 Aerial Spray on Paddy in Pakistan During (1972-1983) .......... 38 Changes in Price of Major Insecticide Used on Paddy Before and After Removal of Subsidies in 1980 ............................................... 38 Disbursement of Agriculutural Credit in Pakistan During 1973—84 by Major Credit Institutions (Million Rd) .......................................... 40 Changes in Nominal and Real Price of Wheat and Rice During 1976-84 Varieties ........................... 40 xii 2.11 2.12 3.1 3.2 3.3 3.4 4.1 4.2 4.3 4.4 4.5 4.6 4.7_ 4.8 4.9 4.10 4.11 Nominal and Real Procurement/Support Price of Basmati and IRRI Varieties (1975-83) .................... 42 Trends in Net Returns Per Acre 1979—80 to 1979-80 to 1982-83 in Rice Area of Punjab ....................... 42 Average Area Under Different Crops in 4 Major Rice Producting Districts of the Punjab During 1979— 82 (000 HA) ...................................... 52 Steps in the Selection of Villages for Survey (1984) ............ 52 Area Sown Under Important Crops by Tehsils and by Districts During Kharif 1981 in the Main Rice Growing Areas of the Punjab, 1981 (00 HA) ............... 54 Distribution of Sample Size According to Village Surveyed ............................................. 58 Distribution of Size of Farm Operated in the Three Tehsil (percentages) .............................. 65 Fertilizer Use on Basmati and IRRI Varieties in the Study Area (Kg/Acre) .............................. 65 Distribution of Basmati Area Fertilized in the Study Area in 1984 (percentages) (n:240) ................. 66 Plant Population for Rice in the Study Area .................. 69 Distribution of Transplant Time for Rice Basmati ' and IRRI in the Study Area (1984) ....................... 71 Average Yield of Basmati and IRRI on Sample Fields in the Study Area (1983—84) ............................. 73 Difference in Per Acre Yields Basmati and IRRI Rice Yields on the Basis of Land (Maunds) ................ 73 Farmers Perception of Major Constraints to Increasing Rice Production on Their Farms (Percentages) ............ 75 Farm Power Problems Faced by Farmers on Sam‘ple Rice Fields (n=76) ..................................... 75 Water Availability and Irrigation Problems Faced by Farmers on Sample Rice Fields ....................... 76 Labor Problems Faced by Farmers on Sample Fields ........... 76 xiii 4.12 11.13 4.14 4.15 4.16 4.17 4.18 4.19 4.20 4.21 4.22 4.23 5.1 5.2 5.3 5.4 5.5 5.6 5.7 Comments Repeatedly Recorded by the Enumerators at the End of the Interviews ............................. 77 Recommendations of the Punjab Extension Department of Rice Production .................................... 79 Problems With Hired Labor for Transplanting ..... I ........... 81 Summary of Major Differences in Production Practices in Wheat by Previous Crop Grown (1984) .................. 83 Summary of Differences in Production Practices of Rice by Previous Crop Grown 1983 .................... 85 Percentage Changes in Rice Area as a Result of Increases in Water Charges During (1980—1984) ............ 89 Percentage Change in Area Allocated to Rice as a Result of Tehsils of Sheikhupura, Afizabad and Daska During 1980-84 .................................. 89 Farmers Suggestions for Removing Constraints to the Production of Rice on Their Farms (Percentages) .......... 91 Crops Grown Instead of Rice during the Kharif Season ......... 93 Farmers Opinion About Crops Which Can Be Grown Instead of Rice ....................................... 93 Reasons for Not Planting Alternative Crops. ................. 94 Area Changes After Introduction of Sunflower in Three Villages in Tehsil Daska (1980-84) .................. 96 Characteristics of Ten Farmers Interviewed in Hafizabad and Daska for Background Information and Choice of Representativeness ...................... 110 Cropping Pattern of Farm One (Hafizabad) (1983-84) ......... 114 Gross Margin/Acre for Farm One (Hafizabad) Representative Farm ................................. 121 LP Tableau for Farm One (Hafizabad) (1984) ................ 123 Cropping Pattern of Farm Two (Daska) (1983—84) ............ 127 Gross Margins/Acre in Rupees for Daska Farm .............. 129 LP Tableau for Farm Two (Daska) (1984) Activity ........... 134 xiv 6.1 6.2 6.3 6.4 6.5 6.6 6.7 6.8 6.9 6.10 6.11 6.12 6.13 6.14 Comparision of Base Model for Farm One (Hafizabad) With 1983-84 Crop Plan ............................... 137 MVP and Cost of Forcing in Non Optimal Activities in the Base Solution .................................. 137 Comparision of Results of Base Model for Farm Two (Daska) With 1983-84 Crop Plan ........................ 140 MVP and Cost of Forcing in Non Optimal Activities in the Base Solution .................................. 140 Comparision of Basic Solution With Alternatives I and II for Farm One (Hafizabad) ....................... 142 MVP and Cost of Forcing Non Optimal Activity for Alternative 1 and II on Farm One (Hafizabad) ............. 144 Comparision of Basic Solution With Alternatives ' I and II for Farm Two (Daska) .......................... 146 MVP and Cost of Forcing Non Optimal Activity for Alternative I and II on Farm Two (Daska) ................ 148 Effect of a 25 Per Cent Increase in Sunflower and Mungbean-Prices at Three Different Yield Levels on Hafizabad Farm ................................... 149 MVP and Cost of forcing Non Optimal Activity in the Solution for a Three Different Yield Levels and a 25 Per Cent Increase in Price of Sunflower and Mungbean for the Hafizabad Farm ...................... 150 Comparison of Basic Solution With Increase in Price of Basmati Rice and Introduction of Sunflower and Mungbean Two (Daska) ................................ 152 MVP and Cost of forcing Non Optimal Activity With a Basmati Price Increase of 10 and 20 Per Cent on daska Farm ....................................... 153 Comparison of Basic solution With Introduction of PK-258, Sunflower and Mungbean for Farm Two (Daska) ............................................. 155 MVP and Cost of Forcing Non Optimal Activity With the Introduction of New Variety PK—258 Rice Variety, Mungbean and Sunflower ....................... 156 XV xvi LIST OF FIGURES Figure Page 1.1 Map of Pakistan ............................................ 3 1.2 Conceptual Model Explaining the Yield Gap Between Experiment Station Yield ‘ and Actual Farm Yield ...................................... 7 1.3 Major Oilseed Crops Growing Areas .......................... 16 2.1 Map of Punjab Province .................................... 27 3.1 Map Showing Location of Study Area ......................... 56 xvii 10. 11. 12. 13. 14. 15. 16. GLOSSARY OF SPECIAL TERMS USED IN THE DISSERTATION TERM Numberdar Major Crops Minor Crops Rabi ' Kharif Paddy Gur Desi Bahak Maund Kuddu IRRI Variety Basmati Kalar Dera Ratoon Crops MEANING The village head and government representative. Wheat, rice, sugarcane, cotton, maize. Also called non traditional crops. Includes sunflower, rapeseed, etc. Winter season crops (wheat, sugarcane, etc.). Summer season crops (rice, cotton etc. Unhusked rice. Local produce made from sugarcane on the farm. Used as a local substitute in the village. Refers to local. Keeping animals out in the field for direct manuring. A local unit equal to approximately 82.2 lbs. Puddling watered rice fields with plough. Refers to the varieties produced by the International Rice Research Institute (Philippines i.e, IR-6). A long grain aromatic variety’. A rice area in the Punjab famous for growing basmati rice. Farm shed. Sugarcane crop kept for more than 1 year in the field. xviii CHAPTER 1 INTRODUCTION PERFORMANCE AND PROBLEMS IN THE PUNJAB AGRICULTURE Agriculture usually plays an important role in the economies of Less Developed Countries (LDC's). In Pakistan agriculture adds 30 percent to the Gross Domestic Product (GDP), besides providing a living for more than 55 percent of the working population. For a summary of the important socioeconomic indicators see Table (1.1). Pakistan has a total area of 796,095 square kilometers and a population of 83.78 million (1). It comprises four provinces. Punjab province is the major agricultural region with 60 per cent of the country's agricultural land (Fig 1.1) and more than 65 per cent of national agricultural production. Sind is the second largest province with 30 per cent of the national agricultural land. The other two provinces, Baluchistan and North West Frontier Province rely heavily on Punjab and Sind for their agricultural needs. Pakistan has shown steady progressin agriculture. During the last 25 years it has recorded an average annual growth field of 3.66 percent (Govt.of Pakistan, 1983). The yields of major crops i.e. wheat, sugarcane, cotton and rice have increased in the last decade (Table 1.2). In particular Pakistan benefited greatly from the new wheat technology which produced a Green Revolution (Khan, 1975) (Eckert, 1970) Pakistan currently faces serious agriculture resource problems. Water logging and salinity cause large areas of productive land to be taken out of Table 1.1 Sunnary of Inportant Socioecononic Indicators Basic Indicators Growth of Structure of Agriculture & Production Production Food LAverage Annual Growth Rate‘%) 1. Pop (Million): 84.5 1. C29 1. CEP(million S) 1. Value added in 2. Area(000)Sq.Kn: 804 1) 1960—70: 6.7 i) 1960: 3500 i) 1970: 2281 3. (DP per capita: 11) 1970-80: 4.8 11) 1931: 25160 11) 1930: 2500 1) Dollar 1981:350 2. Agriculture 2. Agriculture 2. Volume of cereal ii).Rv.annual growth 1) 1960-70: 4.9 (GDP‘%) inports(000 nfltons) rate: 2.8 11) 1970—80: 2.6 i) 1960: 46 1) 1974: 177 4. Adult literacy(9@ 11) 1931: 30 11) 1931: 256 1980: 24 3. Food aid 5. Life expectancy at (000 m/tons) birth 1981: 50 i) 1974: 43 ii) 1981: 94 4. Fertilizer consump— tion (kg/ha of .\rable land) 1) 1970 : 9 ii) 1980 :13 . Average index of food prod. per capita(l969-7l: 100) 1) 1979-80: 74 U1 Source : world Bank ”Annual Report.” 1983. Asian Developrent Bank. ”Key Econonic Indicators.” 1984. Figure 1.]. Map of Pakistan PAKISTAN / n (DISTRICT-WISE) (fig, , ) II~ r I ' ‘ "I rm, \bbolubld . fmlfi Rhum. . 35m: ,_ V W.F.P S =- ium‘F—J V 01 Khin 7 211.41, ("new iiiiuium; Table 1.2 Yield of Nhjor Crops in Pakistan (1975—84) Y e a r Wheat R1ce Sugarcane Cotton ———————————————————————— KG/ACRE---------—-------—-----— 1975-76 574.9 169 8 14776 112 1976-77 579.4 633.6 15169 94 1977-78 532.8 620.6 14814 126 1978-79 602.43 653 85 14712 101 1979—80 634.8 640 15505 142 1980-81 665 654.3 15879.8 137 1981-82 633.6 703 15638 137 1982-83 679.4 705 14443 147 1983-84(P) 6324 680 15435 92 Source: 1) GovernTent of Pakistan, "Agricultural Statistics of Pakistan,” 1982. ii) Governnent of Pakistan "Pakistan Econanic Survey," 1983. 5 cultivation each year. It is estimated that 28000 — 40000 acres becomes unsuitable for cultivation annually (Afzal, 1980). The sixth plan (Government of Pakistan, 1983) stresses the need to tackle this problem through research and development. Despite these efforts the magnitude of the problem suggests that water logging and salinity will confront the country's agricultural scientists for many decades. Some success in reclaiming problem soil has been achieved through the use of gypsum. However, the high cost of this technology limits its use. The Government has encouraged the use of modern inputs through subsidies and sometimes through higher product support prices. The use of fertilizer, High Yielding Varieties (HYV's), water, tractors and credit has increased significantly over time. However, product support price policies in the past have not been so favorable. Wheat, sugarcane and cotton were often supported at lower prices, than in many other LDC's and also lower than world prices (FAO, 1984). To illustrate, the three year average (1979-81) support price for rice in Pakistan was $158/metric ton, while at the same time in Bangladesh, Malaysia and Philippines the support price was US $209, 247 and 188 per metric ton respectively. Similarly, the support price for wheat in 1979-81 and maize was lower compared to neighboring countries such as India and Bangladesh. Low product prices have led to great distortions in' Pakistan's agriculture. Summarizing the performance of Pakistan's agriculture for the past 20 years, Mcnerney (1982) states Much agricultural production in Pakistan relies on irrigation, the principal crops being wheat, sugarcane, rice and cotton - the later two providing the country's main exports. In general, prices received by producers of these products were held down by low prices set for their procurement by the government, fixed margins on their processing and by duties and excess profits taken by the government monopoly and export corporations. ' On the subsidizing of inputs and their implications for equity and distribution he adds ....... Many inputs attracted direct government subsidies, such as imported fertilizer, pesticides, seeds and tubewells. Others carried implicit subsidies, irrigation water was supplied through the public system at very low cost, domestically produced fertilizer was greatly under priced and credit was often available at low rates of interest. While offsetting to some extent the impact on farmers of low product prices, these subsidies further warped resource allocation away from an efficient pattern reflecting opportunity costs. A number of authors judge that these policies have skewed income distribution in favor of large farmers and urban. consumers who have benefited from the low cost inputs and subsidized food products. Landless farmers and laborers have been negatively impacted by these policies (Gotsh or Brown, 1980) (Mcnerney, 1982) and (Khan, 1975). Therefore there is a need to focus on the needs of the small and landless farmers. Conceptalizing the Yield Gap One of the key agricultural problems in LDC's is the large gap between yields obtained on farmer fields and those obtained on experimental stations (Chandler, 1976). Researchers at IRRI have developed a method (2) for studying different factors responsible for these yield differences (Gomez, 1976). They break the yield gap into two components (Fig 1.2). 1) Difference between experimental yield and farmer yield. ii) Difference between potential farm yield and actual farm yield. Figure 1.2 Conceptual Model Explaining the Yield Gap Between Experiment Station Yield and Actual Farm Yield.._ 7 Nontransferable technology Yield gap 1 Environmental difference —‘— ——- BIOLOGICAL CONSTRAINTS - variety - weeds . — diseases and insects - problem soil Yield gap 11 ' water , l -— soil fertility i SOCIOLCONIIMIC (ZONERIIIAINIF; | — cosls and returns - credit — tradition and attitudes — knowledge ' — innu!availal)ili1y — institutions l rpmlmcnl Polenllol Actual Slolicn Yield Form Yield Form Yield Research institutions have traditionally been biased towards researching biological constraints, while laying less emphasis or often ignoring socio- economic and institutional factors. In Pakistan despite the increase in yields of major crops, the yield gap between the potential achievable yield compared to that obtained with existing technology remains large (Table 1.3). These large gaps are often attributed to socioeconomic and technical constraints such as credit, marketing, research and extension (Muhammed, l984). Dynamics in Pakistan Agriculture Farming in Pakistan is undergoing transformation from a subsistence agriculture to highly commercialized enterprises. Farmers are anxious to improve their economic well-being. Schultz (1964), Hopper (1965), and many others, have demonstrated farmers responsiveness to economic opportunities in developing countries. It has also been argued that transformation of agriculture requires that newer profitable opportunities be incorporated into the existing farming systems (Schultz, 1964L and Stevens, 1979). Unless, new alternative crops or production techniques are made available farmers continue growing traditional varieties and utilize traditional low production techniques. in order to improve farm income, crop diversification and cropping patterns changes are often required. Multiple Cropping Systems In the Punjab The complex multiple cropping systems practiced in the Punjab require careful evaluation to understand crop sequencing and their agronomic and economic merits. The growing of more than one crop on the same piece of land Table 1.3 Yield Gap of Various Crops in Pakistan (1982-83) Crop Yield ' National Yield gap Unachieved potential average yield under ex- yield potential perinental condition ---------------- KG/HA--------—--—----------—-per cent- Wheat 2,601 690 1,914 73 Paddy 3,842 705 3,137 82 Whize 2,811 515.4 2,274 82 Sugarcane 104,539 14,447 90,091 86 Cotton(Lint) 526 147 419 74 Rape/Nhstard 1,111 247 864 '78 Source: National coordinated prograns, Pakistan Agricultural Research Council, 1984. 10 in a year makes possible, both an increase in area cultivated per year as well as increase in the total yield per unit of area per year, Dalrymple (1971), Swaminathan, et a1. (1970), FAO (1983). Increasing multiple cropping intensities can help increase farm income in LDC's as well as aid in risk reduction. Research in Orissa, India (Sahu, 1979) and Philippines (Banta, 1973) has shown that marked improvements in small farm income can be realized by following this approach. The International Rice Research Institute has given considerable attention in improving and disseminating Multiple Cropping Systems methodology to other countries (3). In the Pakistan Punjab, the wheat-rice multiple cropping system has long been practiced. The predominant crop of the Punjab in the Kharif (summer) season is rice. It is the major export crop and contributes more than 20 percent to the country's export earnings. Pakistan grows and exports two types of rice. ‘The renowned 'Basmati' is a fine grained variety with excellent cooking quality, flavor and aroma and coarse grain varieties such as IR-6. The focus of research in Pakistan has mostly been on single crops. Very little is known about the interaction of biological and agronomic factors in different crop rotations and about the profitability of newer crop technologies. The Rice -Wheat Rotation The Kharif season in the Punjab rice tract is dominated by rice. The common rotations known to researchers are wheat rice, fallow-rice and fodder- rice (Amir and Chaudry, 1984). However, due to the great variability in international rice prices (Appendix Table A 1.1) and government monopoly on rice procurement there is a large difference between the export price and the 11 price paid to the farmers. Rice is an input intensive crop, more sensitive to increases in cost of labor, water and fertilizer. Because of a low support price and increases in the cost of inputs farmers since 1974 find net returns to rice declining relative to other crops (Table 1.4) as the government in its sixth year plan has announced a gradual removal of subsidies on inputs (Government of Pakistan, 1984). Unless appropriate adjustments in product prices are made or new more productive technologies are found farmers will face further reductions in income from rice. The government recognizes the need for diversification and plans to encourage minor crops (i.e., oilseeds, pulses) through different policies i.e. price support and credit (Govt. of Pakistan, 1984). A specific major area of concern over the past few years, has been the increasing import bill for edible oil (Table 1.5). The magnitude of the problem can be realized by the fact that in 1983 Pakistan earned Rs. 5978 million from exporting rice and cotton. While in the same year it spent Rs. 4224 million on the purchase of oilseeds. Guidelines provided in the sixth year plan (Government of Pakistan, 1984) state that " in domestic markets major emphasis will be on the rapid expansion of domestic oilseeds and protein based crops for human consumption." A critical question is if Pakistan is to increase production of oilseeds and protein based crops will this require major changes in present cropping patterns 7 or can these crops be incorporated within the existing crop rotations. PROBLEM STATEMENT The constraints to increased agricultural production in the major rice growing area of the Punjab known as the 'Kalar' tract are not well understood. 12 Table 1.4 Net Returns for Nhjor Crops in Pakistan During 1974-81 (Rupees) Year Wheat Rice Rice Cotton Sugarcane (Nbxi-Pak) (Basnati) (IRRI) (American) ————————————————————————————— RS/ACRE---—---——--——-----—---—-—-- 1974—75 342 707 660 496 629 1975-76 372 456 565 701 812 1976-77 398 560 539 632 842 1977-78 361 429 513 960 991 1978-79 570 591 497 1007 771 1979—80 529 459 530 449 726 1980-81 754 464 405 626 1170 Source: "Nfinistry of Agriculture, "Cbst Of Production Of ijor Crops," IslaTabad, 1982. Table 1.5 Agricultural Export and Inport of ijor Crops (1972-84) Y e a r Rice Export Cotton Export Oilseed Inport ----------------------- (Nfillion Rs.)----—--——---—————-—--—--—- 1972-73 1136 1167 224 1973-74 2098 367 819 1974-75 2304 1544 1360 1975-76 2479 981 1047 1976—77 2478 292 '1478 1977-78 2409 1102 1553 1978-79 3380 655 2953 1979-80 4179 3321 2295 1980-81 5602 5203 2625 1981-82 4128 2938 3450 1982-83 3683 3897 3670 1983-84 (1) 4448 1530 4224 Source: Governnent of Pakistan, ”Pakistan Econonic Survey," 1983. Data for July-Vhrch only. 13 Only recently has there been any attempt to understand the role of wheat in the area (Beyerlee et.a1., 1984). The possible substitution of new crops- such as sunflower for wheat and rice, has recently started receiving attention from researchers (Amir and Chaudry, 1984). Previous studies by (Saeed, 1980), (Rauf, 1979) and (Haider et a1, 1979) provide partial insight to the factors affecting the choice of cropping pattern in the rice areas. Haider (1979) has identified five factors which affect cropping patterns: 1. Change in government prices for rice and wheat and relationship of these prices to other crops. . Availability of marketing facilities. Irrigation and input availability. Duration of crop growth. Soil types, incidence of disease, tenure and date of sowing. However, this study does not identify changing input costs as a factor determining cropping pattern. The macro performance of the Punjab rice sector is viewed with concern by both government and research institutions. The enormous investment annually made for the provision of rice inputs such as fertilizer, credit, pesticides has not paid off in higher per acre yield of rice. A recent study (Amir, et.a1., 1984) challenges the promotion of higher fertilizer applications in the Punjab without complementing it with higher plant density. These authors conclude that despite the fact that fertilizer use on rice in the Punjab has increased two fold or more over the last 10 years, there has been virtually no change in the yields of the predominantly grown Basmati varieties (Table 1.6). An actual decline in the, yields of IRRI varieties has been observed during this period. Certain biological constraints such as zinc deficiency, plant density, varieties, insect pest etc (4) are thought to limit rice production in the Punjab. 14 Table 1.6 ' Macro-Level Efficiency of Fertilizer Use on Rice in Punjab (1974-82' Year Total Ferti- Area Per/ac* Kharif Yield Yield Total lizer off— (000 ac) Ferti— Water of Bas. of Produc- take Punjab lizer Availa— Ac. IRRI tion (000 NT) bility Kg/Ac (000 T) Milion/Ac 1974—75 288 2028 10.85 29.65 493 784 1115 1975-76 371 2166 15.63 31.61 985 771 1207 1976-77 414 2272 17.6 31.2 502 790 1332 1977-78 465 2589 17.4 34.05 441 801 1508 1978-79 588 2920 21. 33.3 524 754 1766 1979-80 719 2915 24.88 35.5 460 674 1513 1980-81 698 2623 27. 35.65 481 677 1362 1981—82 712 2690 — 36.18 — - * Calculated as 12 percent of total provincial consumption of fertilizer. This estimate is derived on the basis of surveys conducted by the Ministry of Food and Agriculture. For further details, see the first source cited below. Source: 1. "Statistical Bulletin on Rice in Pakistan," PARC, 1983. ' 2. "Punjab Development Statistics," Bureau of Statistics, Government of Punjab, Lahore, 1983. 3. "Pakistan Fertilizer Statistics," Planning 6( Development Division, Government of Pakistan, Islamabad, 1982. 15 Still the underlying socioeconomic factors responsible for the poor performance of this crop are not well understood. It is crucial to gain much better knowledge of the technical and socio—economic constraints which have caused Basmati to stagnate and IR-6 yields to decline in the Punjab. A review of various government studies (Government of Pakistan, 1983) and concerns expressed by farmers in different forums suggests low profitability of the major crops. With an upward trend in input cost, it becomes necessary that the government analyze the possible effects of changes in price and subsidy policies on cropping patterns. Very little attention has been given to explore the possibility of crop substitution in the rice based system of the Punjab. Given the fact that input costs are increasing and the relative profitability of rice to other crops is changing, we must improve our understanding of the micro level constraints facing the rice farmer. There is also a need to understand the dynamic impact of the new rice technologies (higher plant density, zinc application, variety etc.) on the farmers resource allocation decisions. In lother words. we should seek to understand the impact on cropping systems of new technology and modified cropping practices. The effect of changes in certain policy instruments i.e. input price, subsidy, price support in the rice area should also be analyzed. Pakistan is struggling hard to overcome its problem of huge foreign exchange costs for oilseed imports. At present the major focus of oilseed research and production has been in the 'barani‘ or dryland areas (Muhammed and Rehman, 1981). Virtually, no attention has been given to explore the possibility of crop substitution in the rice area (Districts Gujranwala, Sheikhupura and Sialkot) Fig 1.3. This area has never been seriously considered for minor crops such as sunflower and mungbean. It has always been maintained, that only rice 16 Figure 1.3 Major Oilseed Crops Growing Areas Axed )Immu & ; Kammlr 1,1 LOIIIII Kllalln Groundnut Raine-Mustard a :..tiI.-.~... b-iii'luwui Soybean I/ and wheat are suitable crops for the kalar tract. However with declining profitability for major crops the need to investigate alternative crops becomes apparent. A reasonable hypothesis is that if oilseeds are profitable and the present constraints on their adoption are removed, then oilseeds will substitute for rice or wheat in the‘ rice—wheat farming systems. Moreover, there is a vacant period of 2-3 months between rice and wheat. Are there any potential high value protien crops which can be incorporated within the dominant rice- wheat rotation? If so, what are the economic consequences of such alternative crops. In order to provide plausible answers to these questions research has to focus on investigating the technical and economic potential of high value cash crops such as sunflower, pulses etc. To conclude, as Pakistan is currently trying to set long term priorities for agricultural research, it becomes necessary to analyze more information about (1) multiple cropping systems (2) improved price polices, and (3) effect of introduction of new oilseed and protien based crop technologies. This information can help make clear research needs and aid in improved policy making. OBJECTIVES OF THE STUDY 1. To study resource productivity in the rice based farming systems in Gujranwala, Sialkot and Sheikhupura districts of Punjab. A particular objective is to provide a farmers view of the Kharif situation. 2. To identify technical and socioeconomic factors which constrain increases in production of rice and competing crops in the Kharif season in the three districts. l8 3. To analyze through linear programming the economic implications of increasing multiple cropping intensities with new crops such as sunflower and early maturing mungbean and changes in product price policies. 4. To draw implications from the study for research, extension and government policy. This study will complement the efforts of Pakistan Agricultural Research Council and IRRI to develop a multiple cropping systems research strategy for the rice-wheat based farming systems of the Punjab. ’ SCOPE OF THE DISSERTATION The research focds is restricted to the crop production sector on heavy 'soils in the rice based farming systems of districts Gujranwala, Sialkot and Sheikhupura of the Punjab. The investigation of the opportunity for substituting new crops such as sunflower and mungbean is included. The research is exploratory in nature and inferences drawn from the micro level data for national policies have to be interpreted and generalized with caution. ORGANIZATION OF THE DISSERTATION 1. Chapter 2 provides the theoretical framework and background information to the study area. It includes a description of the geophysical and agriculture environment of the Kalar rice tract. This is followed by a brief section on the current policy environment in the Pakistan Punjab and policies NOTES For further details on Pakistan's geography, see Rafique Akther, 'Pakistan Year Book,‘ East-West Publishing Company, Lahore, 1981. Throughout this dissertation, the word method and methodology are used interchangeably. Although from a philosophic viewpoint, they have two separate meanings. Methodology refers to the study of methods. While a method is a set of procedures, those interested in further commentary on the distinction may see Machlup, Fritz, "Methodology of Economics and Other Social Sciences," New York Academic Press, 1978 and Johnson, Glenn L., "Research Methodology for Economists,“ MacMillan Publishers (forthcoming). ' I I had the opportunity to visit IRRI, Philippines in February, 1984. This opportunity proved extremely helpful in reviewing literature and observing the experimental work of IRRI scientists. This dissertation research is funded by IRRI to support Multiple Cropping Systems work in Asia. The main purpose of this program is to introduce the Multiple Cropping Systems philosophy in developing countries and to provide support to national program in initiating such activities. Based on personal interviews with Dr. M. Akbar (former national coordinator rice and presently IRRI staff member) and Dr. M. Majeed, Director General, Punjab Rice Research Institute, Lahore. CHAPTER 2 THE AGRICULTURAL DEVELOPMENT AND POLICY ENVIRONMENT OF THE PUNJAB RICE TRACT The purpose of this chapter is to provide background information on the agriculture development environment in the Punjab rice tract. The chapter begins with a brief statement on the underlying theoretical framework of the thesis. This is followed by a description of the socioeconomic, and geophysical environment of the Punjab rice area, with special reference to the districts Gujranwala, Sialkot and Sheikhupura. In the next section a brief introduction to the general policy environment, policy issues related to subsidies, marketing, credit, trends in input and output prices, and estimates of net returns are discussed. In the final section different sources of secondary information and research issues are presented. THE THEORETICAL FRAMEWORK The process of agricultural development is one of gradual change. Each country follows its own peculiar path, while striving to improve rural [living and general economic health of its economy. Approaches to development have varied. The general focus has been on achieving higher growth rates, improving incomes, lowering malnutrition and improving literacy . In achieving one or more of these goals LDC's have undertaken massive investments often with the help of donor agencies (1). 20 21 The focus of this dissertation is on the determination of constraints to rice productivity and investigation of alternative profitable crops which can improve productivity in the Punjab rice areas. We are interested in a conceptual framework which may help guide investigation of technical change through the introduction of new crop technologies. Hayami and Ruttan (1980) have categorized agricultural development into the following models (1) resource exploitation, (2) conservation, (3) location, (4) diffusion, (5) high return input, and (6) the induced innovation model. Amongst these models the induced innovation model is reviewed to serve as a possible conceptual framework. The Induced Innovation Hypothesis One of the well accepted hypothesis of technical change has been forwarded by Hayami and Ruttan (1973 and 1980)(2). Their argument follows the classical theory of factor substitution. This theory states that changes or differences in the relative prices of factors of production influence the direction of invention or innovation (Hicks, 1932). Building on Hick '5 work Hayami and Ruttan have attempted to explain the inducement mechanism in the Public sector. According to these authors: Technical innovations that save the factors characterized by an inelastic supply, or by slower shifts in supply, become relatively more profitable for agricultural producers. Farmers are induced by shifts in relative prices to search for technical alternatives that save the increasingly scarce factors of production. They press the public research institutions to develop new technology and demand that agricultural firms supply modern technical inputs that substitute for the more scarce factor. Innovation is seen here as not only triggered by changes in factor prices as signaled to the private sector, but it also depends on how public sector researchers and administrators are "induced" to search for alternative profitable opportunities, because of higher expected profits, changes in 22 resource endowments and in response to pecuniary and non-pecuniary rewards." Hence, technical change can be regarded as an endogenous variable in the production process. The role of institutions is recognized in the model. The authors hypothesize that institutions which govern technology generation and production can be induced to change to take fuller advantage of new technologies. Farmer Rationality and Response to Economic Incentives Shultz (1964) and his students developed the fundamental thesis that farmers in developing countries behave rationally (in economic terms) and are efficient in the use of resources. In "Transforming Traditional Agriculture" Shultz explained the backwardness of farmers in developing countries by -asserting that traditional societies remained backward because of limited technical and economic opportunities. They respond to economic incentives in the same manner as farmers in the developed countries. His high payoff input model depended upon: 1) The capacity of the public sectorland private sector research institutes to. produce new technical knowledge. 2) The capacity of the private sector to develop, produce and market new technical inputs. 3) The capacity of farmers to acquire new knowledge and use new inputs effectively. Shultz also hypothesized that traditional societies remained backward because very little public investment had taken place in agriculture. This was partially due to the low returns in agriculture. These low returns were attributed to the low productivity of capital assets used by traditional farmers. 23 In later work Shultz (1979) emphasised the need to invest in human capital and agricultural research. This argument was based on evidence presented by (Griliches, 1958), and (Evenson, 1979) on the high payoffs which have been realized from research in different parts of the world. Returns to agriculture research in the Punjab have also been reported to be high at the rate of 50-60 per cent for wheat and 25-30 per cent for maize ( Pray, 1983) (3). Implications of the Induced Innovation Hypothesis for the Present Study The work of Shultz and Hayami 8t Ruttan recognizes farmer rationality, need for investment in research and responsiveness of the public sector to meet farmer needs. We are able to conceptualize and revitalise old and new innovative approaches to agricultural development. Current approaches such as farming systems (4) and multiple cropping systems have emerged as additional approaches in research. The Multiple Cropping Systems approach which has been developed at IRRI places emphaisis on the maximum use of land in a given year. It involves the analysis and introduction of new crops and intensive use of farm resources to achieve higher incomes, and improved soil productivity . In terms of the high payoff input model as newer more profitable alternatives are demanded by the private sector particularly due to adverse changes in relative factor prices the research community will be induced to search for technologies which meet farmer needs. The need to search for alternative technologies may be induced from within the public sector bureaucracy and also by factors such as a drastic shifts in import expenditures on a commodity. The search process for new technologies may take the form of design, material or capacity transfer from the international pool of technology (Hayami and Ruttan, 1980). This requires adaptive research to test and modify 24 technologies from different environments for use in a local setting. Sometimes if a technology is notiavailable or the problem is location specific, basic research may be required. When a new more profitable technology is available, it needs to be communicated to farmers. The more complex technologies often have to be effectively demonstrated to the farmers. Farmers need to learn the skills necessary to successfully utilize them. If the farmers find the new technologies profitable and are taught the skills to utilize them, then high adoption rates can be expected. Diffusion of a technology may be accelerated through the use of economic incentives. A higher product price or subsidy on specialized inputs i.e seed, pesticides can be used to stimulate adoption in the initial stages. Once the new technology is accepted these incentives should be reduced or withdrawn. However, providing a higher price alone may not lead to a high supply response (Krishna, 1983, Mellor, 1969). These authors find low supply elasticities for wheat, rice and maize in LDC’c. They argue for the need to draw a balance between price and other polices in order to successfully achieve higher production from new technologies. Based on the preceeding discussion of the induced innovation model, we can formulate a hypothesis of change for the Punjab rice areas “..if research efforts are concentrated on identifying new profitable rice and oilseeds technologies, which enable substitution for the scarce resources and reduce the burden of low net returns from the existing low return Kharif rice crop technologies, the new technologies will find rapid acceptability among farmers." If both the economic and technical consequences of new technologies can be estimated through research, then improved recommendations for different farm conditions can be made with greater confidence. 25 THE CURRENT STATE OF AGRICULTURE IN THE PUNJAB PROVINCE The decision was made to carry out this study in three major rice growing districts Sialkot, Gujranwala, and Sheikhupura of Punjab province. The Punjab, one of the four provinces, is the most heavily populated province in Pakistan. It contains about 58 per cent of the population.1n the Punjab more than 72 percent of the population of the Punjab resides in rural areas and relies on agriculture as a means of living (Government of Punjab, 1983). The total area of the province is approximately 40 million acres of which 40 percent lie in the Indus Basin plains. A detailed description of the agricultural situation in the Indus Basin was provided in a comprehensive study by WAPDA (1977) and the Indus Basin Study (Government of Pakistan, 1978). Pakistan has the largest canal irrigation system in the world (Afzal, 1980). Most of the network‘is in the Punjab. This vast natural resource allows for intensive agriculture and diversity in crop production i.e wheat, cotton, rice, maize, fruits. Agriculture in the Punjab is in a transitional phase. It is being transformed from a traditional agriculture to a commercial agriculture. Developments in agriculture in the Punjab have been documented'in studies by Falcon (1972), Gotsh (1971), Eckert (1970) and (Khan, 1978). These studies confirm the Punjab farmer's responsiveness to new opportunities, as they show high rates of adoption, for high yielding wheat and rice varieties experienced in the late sixties and early seventies which were documented by Dalrymple (1984). Many components of the new technology i.e fertilizer, pesticides, tractors, tubewells are well accepted by farmers. During the last 5-10 years, in the Kalar rice tract there have been no significant changes in new biological technologies. Newer wheat varieties 26 intoduced in late seventies (Yecora, SA-42) have been adoptedhbut no new rice variety has replaced the dominant basmati or IR-6 introduced in late sixties in the rice area. A rice variety developed by the Nuclear Institute for Agriculture and Biology locally called NIAB was rapidly adopted by farmers, however the government took strict measures to stop its cultivation, fearing it would be mixed with the traditionally grown basmati and cause a reduction in rice quality and hence in price for export. Investment in tractors and tubewells is continuing. The introduction of the stationary wheat thresher has had wide acceptance among farmers owning tractors. Tractors are also being used to pump water. The wheat reaper was introduced in the early 1980's. Its adoption has not been studied, nor is its presence much noticed in the rice area (based on extensive travel in the rice area in 1984). There are 'reports (personal communication) that almost 4000 reapers which can harvest both rice and wheat are operational in the Punjab. What implications do these technical innovations have on the cropping patterns and farm incomes are not yet understood. Fertilizer use on rice, wheat,sugarcane and cotton is well accepted by farmers. i) The Kalar Rice Tract The different agro-ecological zones in the Punjab have a specialized cropping pattern due to soil type, climate and irrigation availabilitiy (PARC, 1980). The focus of this dissertation is on the rice zone in the Punjab. The area known as the Punjab rice tract comprises the districts of Gujranwala Sheikhupura, Lahore, Gujerat, Sialkot, and Kasur (Fig. 2.1). Among these districts Gujranwala, Sheikhupura and Sialkot are the main rice producing areas. These three districts produce almost 52 per cent of the total rice in the 27 Figure 2.1 Map of Punjab Province 28 Punjab (Punjab Government, 1983). Unless otherwise indicated, the remaining discussion in this chapter will focus on these three districts. The geophysical environment of these districts is explained in the Pakistan Soil Survey studies of 1965 and 1968 (5). These three districts make up the 'Kalar Tract' which is world renowned for the production of fine grained aromatic "Basmati" rice. The Kalar tract has a continental climate with hot summers (temperature ranging 35 to 40 degree centigrade) and moderately cold winters. The average rainfall varies between 361-560 mm (Table 2.1). More than 80 percent of the rainfall is received in the months of July and August. High rainfall during these months enables rice production. The soils in the area are alluvial, varying from well drained and silty clay in nature to heavy. The best soils are suitable for the production of many different crops such as rice, wheat, sugarcane, fodder, vegetable, etc. (Chaudry, 1980). ii) Cropping Patterns The main Kharif (summer) season crops are rice, maize, fodder and cotton. In the Rabi (winter) season wheat, sugarcane and berseem (clover) are grown. The trends in cropped area for the different crops are shown in appendix (Tables A 2.1-2.3). The dominant rotation in the area is rice after wheat (6). Only recently have a number of different rotations less known to researchers been identified by Amir and Chaudry (1984) in the rice—wheat areas of Sialkot district. These rotations incorporate crops such as pulses, sunflower and oats. The cropping intensity in the rice tract varies from 130-160 percent (Wahid, 1981) and (Khan, 1975). 29 Table 2.1 antth Rainfall Pattern Recorded in Gujranwala* \bnth Rainfallunn) June 3.3 July 295.6 August 206.0 Septetber 53.0 Cttober 12.2 November 4.6 Decenber 15.2 January 39.4 February 19.8 Nhrch 30.0 April 10.0 be 517.1 Total 747 O Source: (Personal connunication) Government Agriculture Fann, Gujranwala (1984) *Average of 10 years data, 1971-81. Table 2.2 Distribution of Size of Fann Cperating Units in The Punjab in 1972 and 1980 (Percentage Fanns) Size , ' 1972 1980 --Hacteres ---------------- Percent Fanns ------------ Below 2.5 12.72 15 1.02-3.04 28.57 34 3.05-5.06 23.76 22 5.07-10.12 23.12 20 (greater than) 11.03 9 10.12 Source: 1) Pakistan Canus of Agriculture ,Governrent of Punjab ,1972. ii) Pakistan Census of Agriculture, Govt. of Punjab, 1980. 30 iii) Farm Size The size of operational units in the rice area is distributed approximately in the same way as in the other parts of the Punjab as shown in Table (2.2). Between 1972 and 1980 there have been significant changes in farm size, the number of small land holders (less than 12.5 acres) increased by 6 percent to a total of 71 per cent. According to this classification small, medium (12.5-25 acres), large (greater than 25 acres) operational units can be categorized roughly in the proportion 70:20:10. A, similar distribution has been reported for Gujranwala in earlier studies by (Saeed, 1980) and (Khan, 1975). iv) Land Tenure Different tenurial arrangements are common in the rice tract. These include ownership, tenants, sharecroppers, etc. Khan's (1975) study showed that in Gujranwala district 84 percent farmers were owner operated, 15.2 percent tenants and 10.9 percent lessees in 1974. A similar distribution for the Punjab has been reported by Saeed (1980). These figures are similar to national figures for different tenure types (Table 2.3). I expect in the future to see more farms being categorized as "small farms.“ The increase in owner operated farms can partially be attributed to the fear of further land reforms among large land owners. v) Land Use Land use in Gujranwala, Sialkot, and Sheikhupura is classified into 12 categories by the soil survey department (Table 2.4). According to this classification the first four categories of high quality land (59.9 per cent) are irrigated and suitable for different types of crop production i.e rice, wheat, sugarcane, fodder. Land in the remaining categories is restricted to specific Table 2.3 31 By Tenure in Pakistan - 1972 and 1980 Nunber, Area and Average Size of Private Fanns Classified Fann Size(Ha) Type of Tenure No.0f FannsGMillion) %£hange 1972 ( Fann.Area(NLHa) 1980‘% Change 1971 1980 %Change +)(-) 1972 1980 CMner 1.57 2.23 ( cnmer/tenant 0.90 0.79 ( Tenants . 1.29 1.05 ( Source Table 2.4 Land Use in Gujranwala, Sialkot and Sheikhupura Districts 22. .32 .54 71 .27 .32 .50 S5. 1. Nbinly general 1482 cropping under irrigation fron perennial canals, seasonal canals and wells 2. General cropping 630 under irrigation fron wells,sote dry fanning 3. General cropping 442 and rice.Irrigation tron seasonal wells. 4. Nbinly rice under 1535 irrigation tron canals or wells 5. Nbinly flood-water 627 or winter cropping 6. Dry fanning and wheat 292 winter cropping 7. Dry fanning of 304 winter crops 8. Poor grazing land 741.5 fanning of wheatanillet 9. Unused land 521.5 10. Riverbed 126 11. Urban land 10 12. No photo coverage 49 Total 6760 Source: Government of Pakista, ”Soil Survey of Pakistan,” 1965, 19 32 crops such as rice, fodder, depending on soil type and topography. This study is limited to the first four land use associations. vi) Irrigation Sources The main source of irrigation in the rice areas is through private or public tubewells. Although Punjab has the largest canal irrigated system in the country, still the rice areas are deficient in canal water. In the rice areas the canals are usually operative on seasonal basis. Water is considered a limiting factor in many parts of the rice tract particularly where there are no canals. The growth of tubewells has increased rapidly during the last two decades (WAPDA; 1977). In 1980, 25 percent of the area in Punjab was irrigated by tubewells (Table 2.5). Although there is no data for the recent years,'large investments in installing tubewells in the rice areas continue. vii) Livestock Livestock is an integral part of the farming systems in the rice areas. Most of the livestock kept on farms is for family needs. However, near cities livestock is kept on commercial basis. The commonly found livestock in the three districts are buffaloes, sheep and goat. The population changes for the major livestock in the three districts are shown in (appendix Table A 2.4). GOVERNMENT POLICY AND AGRICULTURAL DEVELOPMENT IN THE DEVELOPMENT IN THE PUNJAB The General Policy Environment Policy in the Punjab is determined by the provincial and federal governments. The Punjab Planning Commission has the major responsibility for -’ fikea Irrigated by Different Sources in Punjab ,_and Pakistan in 1981. ---------------------- (Nfillion Hectares)---—---—------——------—- Chnal 17.78 26.67 (70) (73) Tubewells 6.47 6.89 (25) (19) .r Persian WElls .67 .84 Try (2.6) (2) ‘ Other sources .62 .96 (2.4) (6) Source: Agricultural Statistics of Pakistan, 1982. i - 1 Note: Figures in paraethesis are percentages. 34 overall planning, with guidelines from the federal government. The development approach is guided by the broad objectives outlined in the 5-year development plans . The emphasis in agricultural policies has been to encourage higher production. In past, this was accomplished by input subsidies, minimum support prices etc. At present a gradual policy of removing input subsidies is being followed. Almost all major crops have a regulated support price, the government also sets minimum procurement quotas for rice, wheat, cotton and sugarcane. Punjab is the major supplier of food grain to the other provinces and plays a dominant role in the country's agricultural exports, particularly rice and cotton. The aggravation of oilseeds import situation may require a major shift in agricultural policy to increase domestic oilseeds production. The Punjab government is also sensitive to this issue and is monitoring the situation for possible opportunities (7). Given the broad goals set out in the sixth plan (Government of Pakistan, 1983) to increase domestic production of oilseeds and protein crops one can expect a favorable response to new proposals targeted at increasing productivity of these crops. Subsidies Over the last two decades subsidies have been the main tool used by the government for accelerating crop production. By heavily subsidizing inputs especially, fertilizer, diesel fuel and credit the government popularized input use among farmers. To illustrate, government outlays on fertilizer rose by almost 166 per cent during (1975-1983) (Table 2.6). Despite this large increase in nominal expenditures on subsidies, the real value of the subsidies increased 35 Table 2.6 Provincial and Federal Government Subsidies on Selected Inputs in Pakistan (1975-1983) 1975—76 1976—77 1977/78 1978-79 1979-80 1980—81 1981—82 1982-83 ————————————————————————————————— (tuiiion Rs.)--—---------------------—-----—--- CPI 100 111.77 120.48 128.48 142.23 159.81 175.79 183.67 (1) Plant Protection 241 421 310 279 218 62 - - Real value* - 376.7 257.3 217.2 153.3 38.8 (2) Tube~ 43 48 20 27 22 48 24 24 wells Real value - 43 16.6 21 15.4 30 13.7 13 (3) Ferti- 600.8 131.52 614.56 1691.98 2544.6 2448.3 1819 1600 lizer Real Value — 117.67 510 1317 1725.8 1532 1034.8 871 Source : 1) Agricultural Statistics of Pakistan, 1983. ii) Government of Pakistan, "Fifth Plan," 1979. iii) Pakistan Fertilizer Statistics, 1982. Calcualted by the author * Real Value : Naninal Subsidy/CPI (1975—76:100) *100 Note: A Producer Price Index would be a preferred to CPI, however, this index was not available. 36 less. High rates of inflation and a weak rupee in the international market partially explains the drastic increases in nominal expenditures. These subsidies led to extreme distortions in the economy. The resource poor farmers and landless laborers were the main losers (Mcnereny, 1982), because credit and other production inputs (especially tractors and tubewells) were more accessable to the influential large land owners. Recently the government has initiated a program of gradual withdrawl of subsidies. The impact of reduced subsidies has not been fully documented. However, partial insight on the impact on fertilizer consumption can be infered from studies conducted by The National Fertilizer Corporation (NFC, 1980). Based on data from 15 districts in the Punjab it was found that fertilizer usage on rice was significantly affected by price changes. In 1980, with fertilizer price increases farmers reduced the application of fertilizer on basmati and IRRI varieties. The NFC study does not give the magnitude of changes specifically for rice, but indicates that fertilizer use on the Kharif-crop (1980) was reduced by almost 11 nt. lbs/acre compared to Kharif 1979, after the increase in fertilizer price in February 1980 in the study area. However, the study reported that fertilizer use on wheat and sugarcane did not significantly change, even after price increases. The government has attempted to keep input prices down by providing subsidies and exercising different controls i.e quotas, liscencing. Despite these efforts the real price of all inputs have increased in a greater proportion than the adjustments made in the product price (8). To illustrate, the whole sale price index (1975—76=100) for fertilizer rose to 163.95 in 1983-84, while the wholesale index for motor oil rose to 294.92 during 1983-84. Because of this domestic transportation cost rose rapidly. Between 1979-81 in terms of paddy 37 to fertilizer ratio 2.17 kg of basmati were required to purchase 1 kg of nitrogen (FAO, 1984). Similarly the removal of subsidies on plant protection affected its use. Table (2.7) shows that paddy area sprayed by air decreased from 10.4 per cent of the national paddy area in 1972/73 to 4.1 percent in 1982/83. The price of the commonly used insecticide for rice in the private market rose by about 400 per cent (Table 2.8). The use of insecticide on rice decreased significantly, while its use on cotton changed proportionately less (Personal communication). Marketing Polices There are two types of markets commonly found in Pakistan. Some markets are controlled by the public sector throughparastatals, corporations etc. These agencies serve as government agents for procurement, distribution and export of specialized commodities. Two major corporations are The Pakistan Rice Export Corporation and The Pakistan Cotton Export Corporation. These corporations exercise monopoly power over marketing and export of rice and cotton. Until their required quotas are met the private sector is prohibited from purchasing paddy or cotton from the farmers. Private markets are common in Punjab. Private traders participate in input supply and purchase of the surplus crops. In many instances private agents procure on behalf of the government. A general review of the marketing channels for paddy in the Punjab is provided by Rashid and Mohiyoddin (1980) (9). The current emphasis in the sixth plan (Government of Pakistan, 1983) is to strengthen marketing capabilities of all provinces. Special attention will be given to building storage facilities and improving 38 Table 2.7 Aerial Spray on Paddy in Pakistan During 1972—83 Year Area Sprayed Area Sprayed (000 Acre) (Per CEnt of Total) 1972-73 380 10.4 1973-74 810 21.7 1974-75 934 23.6 1975-76 1156 27.4 1976-77 1275 29.5 1977—78 795 17.0 1978-79 931 18.6 1979-80 148 2.9 1980—81 148 3.1 1981-82 198 4.1 1982—83 200 4 1 Source: Govt of Pakistan, "Pakistan Agriculture Statistics," 1984. Table 2.8 Changes in Price of Abjor Insecticide Used on Paddy Before and After Renoval Of Subsidy in 1980 Insecticide Subsidized List Price Cost After Subsidy Change Withdrawl ------------------ Rupees—--—--—---——-——----——-----Per cent-- 1. Diazinon 10 C 3.31 18.0 438 2. Padan C 3.73 19.0 409 3. Sevidal 4.17 _ 20.5 392 Source: Section Officer Plant Protection, Nunistry of Agriculture (personal conrunication). 39 transportation means. Greater investments are also expected in village to market roads. Credit Polices Credit is considered a major bottleneck in increasing crop production. This view is wide spread among government officials. Massive loans have been given by different national banks (Government of Pakistan, 1983). Disbursement of credit has increased sharply during the last five years (Table 2.9). Providing credit at cheap nominal rates (6-8 per cent) with inflation rising at 9—13 per cent has proven effective in increasing the purchase of capital intensive assets such as tractors, parts 'and tubewells. The government plans to allocate larger amounts through the newly organized farm level cooperative I societies (Government of Pakistan, 1984) Recent Trends in Product Prices For commercially important crops like wheat, cotton, rice and sugarcane a support price is assured by the government. These prices are formulated to provide an adequate incentive to the farmers. The recently established Agricultural Prices Commission in Islamabad has responsibility to help in formulating price polices at the federal level. Although, the nominal price of most major crops increased during the last decade, the real price actually declined due to high inflation (inflation rates ranged between 9—16 percent during the last 8-10 years). The real price of rice and wheat did not increase much during the period 1974-84 (Table 2.10). While in case of cotton the real price actually fell. According to Ashiq (1982) during 1956 to 1979 the index of rice price increased to 343.5 but the deflated price index declined to 89.6 40 Table 2.9 Disbursrent of Agricultural Credit in Pakistan During 173-84 by ijor Credit Institutions (NUllion Rs.) Year AEBP Taccavi Cooperatives Cknnercial Banks Total 1973/74 415.0 67 5 143.72 286.4 912.82 1974/75 395.0 12.5 81.54 520.9 1010 07 1975/76 532.2 25.67 91.84 808.1 1457.81 1976/77 637.9 13.14 95.45 970.1 1716.59 1977/78 429.8 9.0 138.04 1290.92 1867.76 1978/79 416.94 11.96 413.78 1381.11 2223.79 1979/80 711.55 8.2 708.64 1587.40 3015.79 1980/81 1066.6 8.3 1126.25 1862.77 4027.94 1981-82 1557.38 10.34 ‘ - 1100.8 2436.1 5104.62 1982-83 2310.44 2.69 1320.93 2680.89 6314.95 1983-84 3049 11.0 2287.0 3618.0 8965.0 Source : Nfinistry of Food, Agriculture and Cooperatives, State Bank of Pakistan and Planning and Developnent Division. Table 2.10 Changes in Noninal and Real Price of Wheat and Rice During 1976-84 CPI 100 111.77 120.48 128.47 142.23 159.81 175.79 183.67 193.54 Wheat Nominal 39.65 39.65 48.23 50 58 58 64 64 Support Price Real 35.47 32.91 37.54 35 36.3 33 34.85 33 Support Price Real 388.3 297.98 302.8 288.6 264 255 244 254 * Cotton price for AC—143 variety. 41 percent in 1978-79. Similarly, the real price of basmati (1975-76:100) reveals that the price of basmati remained almost constant, while the price of IRRI fell by Rs. 5.68 during 1975—1983 (Table 2.11). A recent FAO study (FAO,1984) reports that prices for major crops in Pakistan were well below the international prices, and in most cases even lower than its close neighboring India and Bangladesh. Estimates of Net Returns Based on the data presented above one would expect low net returns to many crops in the Punjab. In the Punjab the net returns have also shown decline (Chaudhry, 1981), (PERI, 1982). This is evident from Table (2.12), which shows an estimate of the net returns to rice in Punjab being as (10). These low returns are primarlily due to low price polices on the product side. SOURCES OF BACKGROUND INFORMATION FOR SURVEY DESIGN Different government institutions in the Punjab provided information which helped identify issues warranting attention in this research. A brief description of these institutions is given below. Pakistan Agricultural Research Council (PARC) This Federal body is responsible for coordinating, funding and conducting research on issues of national importance (11). Its headquater is in Islamabad. The Council has a National Coordinated Rice Program which monitors the developments in the area of rice research. Discussions with scientists from this program helped acquaint the author with details of the rice situation. The author reviewed current issues facing the rice sector and met officials in the 42 Table 2.11 Nahinal and Real Procurement/Support Price of Basnati and IRRI Varieties (1975—83) ------------------------------------- Rs. per 100 Kg-—--——-----——----—---- 66.98 76-77 139.32 124.65 80.38 72 77-78 ' 148.07 122.90 80.38 66.72 78-79 160.75 125.30 80.38 62.60 79-80 160.75 113.00 80.38 56.50 80-81 187.05 117.05 97.45 52.10 81-82 212.05 120.60 112.50 64.00 82-83 220.00 119.80 112.50 61.30 83—84 229.00 111.80 Note: Real price estimated as Support price/CPI (l975—76:100).*100. Table 2.12 Trends in Net Returns Per Hectare 1979-80 to 1982-83 In Rice Area of Punjab ________________________________________________________ .1........ Year Wheat Rice Cbtton Sugarcane (Basnati) (ATerican) ---------------------------- .NZQE—-------———-—--~--—---—-----——--— 1979—80 204.88* — 239.44* 765.5* — 639.68* 1980-81 1123.73** 762.86*+ -258.26** 3103.58** 1982-83 1117.00*** -1116.4*** - _ Note: Excluding land rent, artisan costs and interest on invesUTent etc. Sources: * ”Cost of production of major crops,” University of Agriculture, Faisalabad, Jan. 1982. ** ”Study on Cost of Production of Crops,” Govt. of Pakistan, VUnistry of Food, Agriculture and Cooperatives, Food & Agriculture Divn., (Planning Unit), Islanabad.1983 *** PARC, "A prelhhinary resource use analysis for Rice in Dist. Sialkot”, PARC, lslanabad, 1984 43 agriculture ministries (Federal and Provincial), Economists at the Pakistan Institute of Development Economics (Islamabad), the applied Economics Research Center (Karachi) and Pakistan Rice Export Corporation. Some of these professionals were responsible for policy making related to rice production, consumption and export. The PARC rice program has a small administrative unit at the Punjab Rice Research Institute (PRRI) at Kala Shah Kaku. This unit has been involved in a rice technology transfer project over the last 3—4 years. This field unit also executes the National Rice Varietal Trial Program to test the performance of new varieties under different environmental conditions. Discussions with this group helped identify key issues facing farmers in Gujranwala. PARC also arranges a rice seminar on a bi—annual basis. Based on this seminar a set of recommendations for the major rice growing areas are formulated. The reports of past seminars (for example, see ESSO, 1976 and CIBA-GEIGY, 1975) historical documents on development of rice technology and policy papers were reviewed at the PARC library to get a broad perspective of the situation (12). The PARC also has a National Coordinated Oilseeds and Pulses program. This research group had not conducted research in the rice tract, however different staff reports provided background on the status of oilseed and pulses research in the country (personal communication with national oilseeds and pulses coordinators at the National Agricultural Research Center, Islamabad). Punjab Government Institutions The Punjab Rice Research Institute is at Kala Shah Ka‘u, which is located about 5 miles outside Lahore. The author was stationed at this institute during 44 his field work and was able to interact with rice scientists on different aspects of rice production. The institute compiles an annual research report which summarizes research achievement and future plan of work. The series of reports date back to 1968. A review of this work helped avoid duplication of past efforts, and allowed the researcher to sharpen his focus on the key issues. The Institute had also conducted experiments on different crop rotations in the rice area. The results of these experiments are utilized in later chapters when the economics of alternative cropping patterns is explored. Nuclear Institute for Agriculture and Biology This institute is located at Fasialabad. The pulses group at the institute has reported success in the development of short duration mung bean varieties (NIAB, 1983). These researchers provided data on mung bean trials conducted on farmers fields in rice areas. This information is used in chapter 5 to analyze the economic feasibility of these varieties in rice-wheat based farming systems. Other Sources The Punjab soil survey department, Punjab Applied Economics Research Institute, University of Agriculture Fasialabad, The Ghee Corporation, extension departments and farmers were other sources of information. Identification of the Important Issues After a careful review of field notes, based on discussions with scientists, extension workers, government officials and with farmers a list of important issues relevant to the focus of this dissertation was developed. These issues are 45 briefly enummerated below. Technical Issues The rice scientists at the Punjab Rice Research Institute and Pakistan Agricultural Research Council considered the following factors to be important in limiting rice production in the Punjab. 1. Low plant density 2. Zinc deficiency 3. Low fertilizer use on basmati and IRRI varieties 4. Outdated varieties of rice 5. Pest problems These problems have been well documented in different studies (Majeed, 1979), (Ross, 1977) and the Indus Basin Study (1978). The results of the PARC applied research trials also support the importance of these factor in increasing rice yields (based on review of ten years of experimental data available with national coordinator for rice, NARC, Islamabad, also see PARC, 1981). Among the technical constraints, biological constraints have received greater attention from researchers, whereas less attention has been given to other areas such as improved farm equipment. Researchers have directed some efforts to test the feasibility of imported rice technologies i.e transplanters, fertilizer applicators. However, these technologies have not been adopted because of their poor performance under Pakistan conditions or due to high cost (Multan Institute, 1982). Socioeconomic Issues This area has been often neglected by researchers. During my preliminary visits to the study area most farmer complaints were directed at $951811. My fat-titers did not consider biological constraints as impeding yield, "Bit: str'egsed ”the importance of non technical factors. 5) 47 NOTES For a general survey of agricultural development theory see Eicher,Carl K and Staatz, John M. (eds). "Agricultural Development in the Third World." John Hopkins University Press, Baltimore, 1984. This section draws heavily on the induced innovation literature. The work by Hayami, ’Y. and Ruttan Vernon. 'Agricultural Development: An International Perspective.’ The John Hopkins University press 1980, gives a detailed exposition. Also see articles by same authors in Eicher and Staatz cited in footnote 1. Based on personal interview with Dr. Vernon Ruttan during his visit to Pakistan in September, 1984. Dr. Ruttan reported that one of his Ph.D student Joseph Nagy had found rates of return in the range of 50-60 Percent for Wheat research in Pakistan. A good overview to the approach is provided by Derek Beyerelee et a1 ”Farming Systems Perspective: Issues in Research and Technology." AJAE 64, No. 5 (1982). Also see the 1960 population census of these districts for further details on the cultural and socioeconomic background. At the time this research was conducted the results of the 1981-83 Census Survey were being compiled and were available only in draft form. A detailed description of the crop rotations, agronomic requirements and crop calenders for different regions of Pakistan can be found in: Agricultural Credit Division "Crops of Pakistan" Muslim Commercial Bank Limited Karachi, 1980. These views were expressed by the provincial Minister for Agriculture in his annual address to the Punjab Assembly in July, 1984. The minister also briefed the assembly on various development issues and argued for special attention to increasing oilseed production in the Province. Different sources of this information include the Agricultural Prices Commission, Punjab Applied Economics Research Institute (Lahore), Ministry of Agriculture (Islamabad) and University of Agriculture Fasialabald. 9) 10) 11) 12) 48 A good description of the marketing system in the Punjab is given in two studies: (i) Schinzel Hans U. "Determinants of Market Supply and Behavior of Agricultural Producers in Punjab" and (ii) Schmidt Gunter. "The Efficiency of Marketing System for Agricultural Products in Pakistan Punjab." Both studies are translations from earlier German editions (1976) and are published in their English version by Verlag Breitenbach Publishers Saarbrucken, Fort Lauderdale, 1982. The research for these two books was conducted in 1973, hence does not cover recent trends in marketing. Also see Chapter 1 For further details on the organizational setup, and working of this organization see Muhammed, Amir, 'Agricultural Research System of Pakistan,‘ PARC, Islamabad, 1982. A good source for reference on location of literature in Pakistan is "Agricultural Bibliography of Pakistan," PARC, Islamabad, 1983, Islamabad, 1983. CHAPTER 3 THE METHODOLOGY OF THE FARM SURVEY FOR IDENTIFICATION OF CONSTRAINTS LIMITING PRODUCTION INCREASES The purpose of this chapter is to present the survey method used to obtain information on rice and minor Kharif crop production in the Kalar tract. SURVEY DESIGN Objectives the Survey The survey of rice farmers had the general objective of increasing understanding of rice production in the Punjab. It was limited to the crop production activities and had the following specified objectives : 1. To obtain information on input levels, yields, cropping patterns, varieties and socio economic constraints in rice production. 2. To identify any existing multiple cropping systems unknown to researchers. 3. To gather information about factors limiting the allocation of land to minor Kharif crops with special reference to sunflower and pulses. 4. To obtain specific input/output information on sample fields to help construct the representative farm models. On Going Research At the time this research was conducted there were two other field research projects in the area. 49 50 1. Economic Evaluation of the Gujranwala Rice Maximization Project A team of researchers from PARC‘s Social Science Division was involved in evaluating a technology transfer project carried out in 1981 to 1983. The main objective of this research was to determine the economic impact of the new technology and to study this approach for wider applicability in other areas and for different crops. The author had a chance to participate in the design of the evaluation. 2. Rice Yield Optimization Program, Daska (Sialkot) This project started by PARC and the provincial government in 1984 had the objective of disseminating improved rice production technology in Daska. This project was designed to serve as a field laboratory where scientists get a chance to observe farm problems and work in multidisciplinary teams to solve them (Chaudry and Amir, 1984). The broader objectives of this pilot project are (l) to identify rice production constraints in the project area (2) involve the farmers in the design of a package suitable to the area needs (3) intitiate experimental work on multiple cropping systems and (4) experiment with alternative approaches of technology transfer. In the future it is expected that this project will involve scientists from different disciplines and also provide opportunity for students at Fasialabad University to do applied research. The whole orientation of this project is to encourage the researcher to observe and acquaint himself with farm level problems. By visually observing the performance of a technology package the researcher learns the broader socio-economic constraints which influence technology 51 adoption. Survey sites for my dissertation'field work were chosen in other areas to avoid possible contamination from these two research sites. Selection of the Survey Area Numerous reconnaissance field visits were made in the three districts of Gujranwala, Sialkot and Sheikhupura, prior to the design of the formal survey. This researcher had a chance to accompany members of the CIMMYT (Islamabad office) researchers during their preliminary visits to Sheikhupura district, where these researchers were planning to do wheat and maize work (1). This helped locate areas which could serve as potential survey sites The Soil Fertility Department in Lahore was approached for information on soil types, crop and land use in the Kalar tract. The area maps provided by the department were used to identify rice growing areas in the Punjab and in the three study districts (2). A decision was made early to restrict the formal survey to excellent and good type of soils which are suitable for general purpose cultivation, and to avoid problem soils i.e water—logged, saline, poor texture soils. This was justified on the basis that study of poor soils was not relevant to investigating multiple cropping systems as they only have remote chances for crop substitution. In order to represent the good and excellent soils of the Punjab rice tract, three districts; Gujranwala, Sheikhupura and Sialkot were purposively chosen. These districts represent more than 52 percent of the total rice area of the Punjab (Table 3.1). They are located near Lahore, the provincial capital. The agricultural areas are fairly accessable by road. 52 Table 3.1 Average Area Under Different Crops in 4 Major Rice Producing Districts of the Punjab during 1979-82 (000 HA) Crops Gujranwala, Sheikhupura, Gujerat Punjab Per cent and Sheikhupura districts. of Punjab Averages, 1979-80 to 1981-82 Wheat 2426.44 12864 _ l9 Sugarcane 186.00 1655 ll Maize 95.00 800 12 * Rice area excluding district Gujerat is approximately 52 per cent of Punjab Source: Goverrment of Punjab, 'Punjab Development Statistics,’ Lahore, 1983. I Table 3.2 Steps in the Selection of Villages for Survey (1984) Step Description 1. Select three rice districts from the Punjab purposively. Gujranwala, Sialkot and Sheikhpura were chosen. 2. Select three thesils purposively to represent excellent and good soils. All three thesils are to represent the cropping pattern of the district and are to be major rice growing tehsils. 3. Choose a union council purposively to represent general conditions in the tehsil. 4. Randomly chose five villages out of 8-12 from each union council for survey. Sample in a proportion of 60 (small), 20 (medium) and 20 (large).* * Where this proportion was difficult to maintain, it was adjusted based on the general distribution of a village. This problem was faced in Daska and Hafizabad tehsils. 53 Sampling of Farms for the Survey Each of the three districts chosen for this study is reasonably homogeneous with respect to type of agriculture, ethnic groups, cropping patterns. etc. However, there is considerable heterogenety in soil types within each district. In order to represent the excellent and good soils it was necessary to chose a cluster of villages which not only represent soil conditions but are similar in other respects i.e tenure, ethnic groups. A multi stage sampling approach was used in the three districts. A schematic presentation of the steps followed in choosing the 15 surveyed villages is shown in Table (3.2)., Each district is composed of a number of sub—districts called tehsils. The cropping pattern in each tehsil was studied carefully prior to arriving at a final decision Table (3.3). The major criteria used in selecting thesils was the acreage. under rice and the representativeness of the cropping pattern in relation to the district . After detailed discussions with the district extension officers and researchers of the Punjab Rice Research Institute, one tehsil in each district was chosen purposively, Daska (Sialkot), Hafizabad (Gujranwala) and tehsil Sheikhupura (Sheikhupura). Within each tehsil, there are a number of local government units called Union Councils. Each Union Council comprises a cluster of 8-12 villages. One Union Council was purposively chosen from each tehsil to represent the district. This choice was based on discussions with the local agricultural officer and visits to 3-4 different Union Councils. The accessability by jeep and representativeness of the general development situation were also taken into account. From each Union Council, five villages were chosen randomly. If a village was located within two miles of 54 Table 3.3 Area Sown Under Inportant Crops by Tehsils and By Districts during Kharif 1981 in the Nhin Rice Growing Areas of the Punjab, 1981 (00 HA).l ----------------------------------------------------------------------------- S.No Tehsil/District Total Rice Jowar Bajra Nbize Cotton Sugarcane Others Area Sown Sialkot 2613 1538 5 241 105 66 122 536 l. Sialkot2 526 312 l 28 26 3 17 139 2. Daska 736 582 3 17 8 13 37 76 3. Narowal 492 306 - 44 l4 19 22 87 4. Pasroor 439 214 - 42 l5 18 27 123 5. Shakargarh 420 124 l 110 42 l3 19 111 Gujranwala 3084 2411 39 155 51 49 166 215 1. Gujranwala 1566 1285 I4 55 24 15 55 108 2. Hafizabad* 871 605 16 72 21 20 62 75 3. thirabad 647 521 9 28 6 14 39 30 Sheikhupura 2764 1914 12 52 121 47 218 398 l. Sheikhupura* 1162 657 8 43 65 33 138 218 2. Nankana Sahib 748 493 l 4 39 13 66 132 3. Ferozewala 852 764 3 5 l7 1 14 48 Lahore - 1. Lahore 713 376 l 3 45 17 26 247 Kasur 1861 569 l 13 52 242 332 652 l. Kasur 802 293 - 9 23 65 111 301 2. Chunian 1059 276 1 4 29 177 221 351 1 To get estkmates in acres, nultiple hectares by 2.47. Source: Government of Punjab, 'Kharif Crop Statistics, 1981,' Lahore, 1931. * Chosen Tehsil. 55 the main road it was dropped from the sample and replaced with another randomly chosen village. This approach allowed good representation of the more remote areas and reduced urban bias which can resultjfrom being located near highways or small market towns. etc. Location of the 14 villages chosen for the survey are shown in fig 3.1. Originally it was planned to survey 15 villages, however one village had to be dropped because of non cooperation. The author tried to find a replacement, but due to heavy rains and transport difficulties it became virtually impossible to reach the village chosen as a replacement. A decision was made to drop the fifth village in Sheikhupura. The sample size was increased in equal proportion among the other four villages. Sampling Frame and Selection of Respondents An initial contact with the ‘Numberdar' (village head) and Agricultural Officer of the area was made. The numberdar served as the contact person in the village. The purpose of the survey was explained to him to elicit his cooperation. The numberdar was helpful in identifying the land of each farmer in the village. The local agricultural officer was contacted for a list of names of farmers in each village. These lists did not prove useful as they contained only the names of the few farmers where the field assistants had applied insecticide. In some sample villages a listing of all the farmers was available, while in others the records maintained were either outdated or were not appropriate for formulating a sampling frame. In the later case it was decided that the 'Numberdar' along with a few older farmers would be asked to provide a list by recalling the names and land holding of all farmers in their village. From these listings, a 56 Figure 3.1 Map Showing Location of Study Area S tudy Area \y Peshawar I '~ ‘ 1' ”l - ' x I A _ I Ananlni, ,. ‘ l ‘ l r “ Mullhn o . KM Simmq / 3 Bulk It'd'u'll ' '(Ihnkwnl Din-Ml KlIN 1.1mm“ x Dnmlt ll .‘ \ - . Mu'inwfih | . SI. _, 'i ‘ / .hushé (‘loo MI I , (éugoilhn I 1 f l T. I.bw‘aia‘{ 3an I \ I. Chlmot ' i ,\ ' l Jhumru % l , f . : . I ‘_ ll / ’ , , l 607 V / \-‘ / mfir 3’ Vt'm‘ l i Rhiwmo 1511.1.y .11.” / " , ' / _ m2 . KRSU . i / N ,/ . [N'// ,Gnndn Son iwfilu -(\'{()/(‘[‘}()IIY I,Snluwul' ’ - . l. 'I 4 Survey Area 57 sample of 15 farmers from each village in the proportion: 70 percent small (less than 12.5 acres), 20 percent medium (12.6-25 acres) and 10 percent of large (greater than 25 acres) farmers was selected (3). In the cases where this proportion was difficult to maintain an adjustment was made. This gave a total of 240 farmers in 14 villages in the 3 districts. A distribution of the sample farmers in each village is shown in Table (3.4). This sample allows one to draw statistical inference at the tehsil level. A larger sample covering more tehsils would be desirable. But financial and time considerations constrained the researcher from carrying out a larger study. Design of the Three Questionnaires Based on the issues identified during the informal meetings with farmers, researchers, and extension officials, three formal questionnaires were designed, one to gather information on constraints to the production of rice and minor crops,and the second to obtain input-output data on one sample rice field on 76 farms, and a third for information on sunflower. Questionnaire No. 1 This farm questionnaire was designed to collect detailed information on a variety of variables including water, fertilizer, pesticide and credit use; factors limiting production of minor crops, and reasons for growing particular rice varieties. The data was collected for two seasons, 1983 and 1984 from 240 respondents (Table 3.4). As the field work was carried out during the rice planting season the output information for 1984 is based on the farmers expected yield in 1984. A copy of questionnaire is attached in the appendix (B-I). Table 3.4 Distribution of Sample Size According to Village Surveyed 58 No. of Farms Where Qiestionnaire 2 Was UsedM Dhi lwan Haf izabad Ud owa 1 I Hutboki Buddoki Budha Chak Chatta Chak Shamsa Bhullar Mah Thakra Chak Pussya Kukay Chak Quesnonnane‘No.l wasusedforall249respondents These 76 farmers were randomly chosen amongst the 240 respondents. One village had to be dropped because of extreme political rivilary with the others. The sample size was a justed according to these 4 leages.'Because offeW'farntslocated nearthe leage htthe other leages,in Nlaki-46O the field santple observations were increased to 9 for Questionnaire No. 2. 59 Questionnaire No. 2 This short questionnaire was designed to obtain specific information on input use and constraints on sample fields (Appendix B-II). This information was utilized in discussing the farmers viewpoint about his constraints and also helped to provide background in selecting representative farms for modeling. This field questionnaire was used to obtain information from only 76 farms (4). These respondents were chosen randomly from the 240 farmers and they were equally distributed in the three tehsils Table (3.4). However, in making this choice the nearness of the field to the farm was taken into account. Questionnaire No. 3 A'short questionnaire was used outside the survey area for to obtain information on sunflower grown on 11 farms in tehsil Daska (see appendix B—III). This was a side sample to obtain specific information on this crop. It was executed in the Daska Rice Yield Optimization Project area where the author was able to locate some sunflower growers. The sample was chosen purposively because of the small number of farmers growing sunflower. Pretestin and Validation of uestionnaires The questionnaires were pre-tested. The pre—test involved 15-20 farmers, they were also asked to comment on the difficulty of certain questions. Those identified as irrelevant or ambigous were reworded or dropped prior to finalizing the questionnaires. 60 Field Method The survey was carried out during June—July, 1984. Six enummerators were hired and intensively trained in data collection procedures and role playing, before proceeding with the formal data collection. They were provided handouts, explaining details of the local terminology i.e, local names for varieties, common fertilizers used. Two teams of three were formulated each with a well qualified supervisor as the team leader. The supervisors had M.Sc degrees in agriculture from Fasialabad Agriculture University, while the enummerators were B.Sc students at Fasialabad. Five members of the two survey teams came from the rice area. The supervisors and enummerators were thoroughly briefed on the objectives and design of the research. Each team was provided a field assistant to assist in measuring plant densities on sample fields. The author supervised this data collection operation to ensure that quality control was maintained. This was accomplished by randomly chooseing 8-10 farmers from each Union Council and re-interviewing them to cross check the consistency of responses and quality of the information recorded. Each enummerator had been assigned a fixed number of questionnaires to be filled based on his ability. The cross check showed that the information had been recorded with care. Heavy rains during the survey work posed severe problems of communication. Because the research sites were in remote areas with earth or "Kacha" roads, these villages had to be accessed on foot. The researcher initially faced other problems as the Government had recently introduced the Islamic Tax called ‘Ushr.’ This tax had been implemented for almost one year at the time of this study. It is levied on the total 61 produce in Kharif and Rabi season. In some cases, farmers were hesitant in revealing information on crop yields . However, after assurance that the information would only be used for research purpose and was not being collected for the government the farmers willingly cooperated. Data Preparation and Analysis As soon as the field work was completed the data was edited and rechecked in the field before bringing it to Islamabad for final processing. An Apple Interactive Data Analysis (AIDA) survey program was used on an Apple 111 micro computer. The data was cleaned and coded on the questionnaires and entered into the computer. This procedure avoided first transforming data to analysis sheets and later feeding it to the computer. The data was reentered for verification to detect possible errors prior to the final analysis. 62 NOTES These researchers were conducting diagnostic surveys to identify researchable issues on wheat. For further details of this research, see Beyerlee, D. et.al. "Wheat in the Rice-BasedFarming Systems of the Punjab — Implications for research and extension." CIMMYT, Islamabad, Pakistan, 1984. These maps are for restricted use and their source cannot be revealed. This proportion is adopted on the basis of survey work done by the Agricultural University Fasialabad in Gujranwala district. The same questionnaire was utilized by PARC researchers in their evaluation study in Gujranwala. This research is refered to under the heading "Ongoing Research" in this chapter. CHAPTER 4 IDENTIFICATION OF CONSTRAINTS LIMITING INCREASED PRODUCTION OF RICE AND OTHER MINOR KHARIF CROPS — FARMERS' VIEWPOINT Overview This chapter reports the results of the survey work described in chapter 3. The discussion is presented under four sections. Section I describes the characteristics of the farms i.e farm size, tenure, cropping patterns, level of input use and rice yields. Section II summarizes rice production constraints as perceived by the farmers. It highlights the institutional problems which are partially responsible for these bottlenecks. The researchers perspective on these production constraints is also presented. Section III focuses on key issues related to resource productivity in the rice based system. Factors responsible" for its poor performance are identified. The section differs from the previous one in its macro focus. Certain questions raised by the farmers are analyzed in the light of the existing policy environment. Section IV includes a brief summary of farmer suggestions for removing constraints on rice production. Section V briefly discusses on constraint to the production of alternative Kharif crops. SECTION I CHARACTERISTICS OF SAMPLE FARMS AND RESOURCE USE Farm Size In designing the survey a proportion of 70 small (less than 12.5 acres): 20 medium, (12.5—25 acres) and 10 large, (greater than 25 acres) was planned based 63 64 on previous studies. However in Daska and Hafizabad due to small operational holdings it was difficult to find medium and large farmers. In Daska particularly an overwhelming majority of farmers are small holders, category 44.6 per cent farms had less than 5 acres (Table 4.1). This Table shows that small farmers are the bulk of the farming community in the study area. Tenure Tenure status was only inquired from farmers who were interviewed for information on sample rice fields. Among 76 respondents, which were randomly chosen from each village a majority (65.3 percent) were owners, 20 per cent owner—cum-tenants and 14.7 per cent tenants. Fertilizer Farmers generally used two types of fertilizer, nitrogenous and phosphatic. The most commonly used fertilizer on basmati rice is Di- Ammoniurn Phosphate (DAP) which was used by 35.4 per cent farmers (n:240), while 4.2 per cent used nitrophos. Interestingly, 22 per cent farmers did not use any fertilizer on basmati rice. Average amount of fertilizer use on basmati rice is lower (35.5 kg/acre) as compared to IRRI rice varieties (Table 4.2). This is partly due to the lodging problem with basmati rice. The average amount of fertilizer use on basmati rice was higher on farms which .allocate less area to rice. As illustrated in Table (4.3) 65.5 per cent farmers applied fertilizer on less than 5 acres, while 87.2 percent farms applied fertilizer to less than 10 acres. This suggests that as the area allocated to basmati rice increases, fertilizer usage decreases. 65 Table 4.1 Distribution of Size of Farns Operated in Three Tehsils (percentages) Size Sheikhpura Hafizabad Daska ------------------------ (acres)------—--—----—---—----—-———-—---—-—- n: 83 83 74 Stall 96 cf(96) 96 cf (96) 96 cf(96) les than) 5 12 12 6 6 44.6 44.6 5.1-7.5 7 2 19.2 13 3 l9 3 12.2 56.8 7.6—10 2O 5 39.7 22 9 42 2 9.5 66.3 10.1-12.5 16 9 56.6 13 3 55 5 10.8 77.1 Nbdium 12 51-25 22 9 79.5 34 9 90 4 16 2 93 3 Lar e greater than) 25 20.5 100 9.6 100 6.7 100 Note: cf = cummulative frequency. Table 4.2 Fertilizer Use on Basmati and IRRI Varieties in the Study Area (Kg/Acre) Variety Sheikhpura Hafizabad Daska All Basmati NEAN 38 82 36.52 28 93 35 491 S) 17 36 15.63 14 53 16 44 CV 06 USERS 71 7O 47 188 IRRI NEAN 45.15 44.42 34.65 41.7 so . 15.37 17.42 15.45 16.94 cv %) Users 13 36 20 69 '. II . "f ' ;rII.,:...'. WHWHflfi-A'I mum unmet. aamtimaamimam'm . Study Area in 1984 (percentages) :(n=2.40) \l'fi cu! -u. u :'- .tfif.»fi "-1-“ ------------------------------------------------------- M1335 Interval Frequency Cumulative Frequency (Acres) Do not Use 21.7 21 7 .01-5 43.8 65.5 5. 01-10 21.7 87.2 10. 01-15 7.1 94.3 ' 15. 01- 20 2.9 97.2 , 20.01-25 1.7 98.9 .25.01-30 .4 99.3 30.01-50 .7 100.0 67 Farm Yard manure on rice was used by 73 per cent farmers, if they use FYM, 54 per cent do not use any chemical fertilizer on rice. The 'bahak' system is also very common. This involves temporarily housing animals in the field, so that the manure generated is deposited directly in the field. Zinc has been identified to be deficient in many rice soils (Saleem and Bertilsson, 1979). However only 4 per cent (n:76) farmers reported using zinc. Many were not aware of its use or knew how to diagnose its deficiency. Water Availapility Water is perhaps the most costly and limiting resource in the rice based system of the Punjab. There are three main sources of irrigation during the Kharif season. Canal water is limited in the Kharif Season and is seldom sufficient for growing rice. Supplemental irrigation from private or public tubewells is usually necessary. The number of irrigations applied depends on the frequency of rainfall during the monsoon season (July to August). The cost of water is an important determinant of the variety sown and number of irrigations applied. The cost of irrigation for basmati rice is higher than for IRRI rice. This is due to the longer time it takes for basmati rice to ripe. Farmers reported that on an average the cost per acre for water would run to approximately 972 rupees with a diesel tubewell, assuming no canal water is used. Tillage Operations Land preparation varies by soil type and crop sown prior to rice. Fallow land usually receives more tillage than if a previous crop was sown. On the average basmati rice receives 11 ploughing and plankings while IRRI rice 68 receives 10. IRRI rice receives less land preparation because it is normally grown on average land and planted earlier than basmati rice. Use of Insecticides, Pesticides and Herbicide The use of these inputs was virtually non existent in the study area. Insecticide was only applied on the rice nursury, if provided by the extension department at subsidized rates or free of cost. There was no use of herbicide among the sample farmers. However, 70 per cent of the farmers practiced water flooding and/or manual weed control while 30 per cent did not weed at all. Weed control also differs by farm size. It was more common on small than larger farms. Credit Among 240 respondents, 92 per cent did not use any formal credit i.e bank, arthi (middleman). Only 7 per cent used credit from the commercial sources. Farmers were asked if more credit were made available, which inputs would they use it for. The general response was that fertilizer would be used in higher quantities. Formal credit from commercial banks is more Commonly used to finance the purchase of tractors, tubewells etc. Plant Density A sample measurement of 75 rice fields was taken to determine the average plant density. Three counts were taken, one in the middle and two on the sides. A one meter rectangular rod was used for obtaining these parameters. The results are summarized in Table (4.4). The average density was the highest in Daska averaging 16.18 plant per square meter or 161,800 plants/hectare or 65,506 plants/acre. Average density levels did not vary 69 Table 4.4 Plant Population for Rice* in the Study Area* Plant Density** Sheikhpura Hafizabad Sialkot l. Sanple I (side) Mean 14.23 13.08 16.28 SD 3.69 3.39 5.24 CV (%) 26 26 32 2. Sample 11 hmiddle) Mean 12.92 12.96 16.16 SD 3.96 4.64 4.18 CV (%) 31 36 26 3. Sanple III (side) Nban 13.64 13.25 16.12 SD . 3.96 4.64 4.18 OI (%) 29 35 26 4. Average of sarrple I, II and III Mean 13.64 13.25 16.18 SD 3.45 3.32 4.04 CV (%) - 25 25 - 25 * Plant density for all types of rice (plants per square meter). For a cmparision of basrrati and IRRI densities, see Appendix A, Table 4.3. ** To get plants per acre miltiply by 4050.0. 7O greatly from tehsil to tehsil, although there was considerable variation within each tehsil as indicated by the high coefficient of variation. Cropping Patterns The cropping pattern in the three tehsils was dominantly rice-wheat. Minor variations with fodder, cotton and sugarcane were also common. Almost 90 percent farmers reported rice-wheat as the dominant rotation on their three _ largest parcels. In the Kharif season rice is the main crop in all three districts (For a summary of the cropping pattern on three largest parcels see Appendix A Table 4.1). In tehsil Hafizabad sugarcane is gaining importance. This is due to the relatively high returns for this crop. Planting Time for Rice The planting time for rice is shown in Table (4.5). Most basmati rice is transplanted during the first week of July and completed by the beginning of August. IRRI rice is transplanted early in June. The nursery transplanting takes place in the end of May for IRRI and basmati rice varieties. Based on experimental data provided by the Punjab Rice Research Institute, there is a significant relationship between time of planting and yield (Appendix A Table 4.2). Rice planted late in July and early August yields less than if transplanted between June and July. Farmers also realize this and try to transplant rice at the optimum time. The survey data shows that farmers follow an optimum time of transplanting. The variations are usually due to delayed rains and problems with finding transplanter labor. 71 Table 4.5 Distribution of Transplanting Tine for Rice Basmati and IRRI in the Study Area (1984) anth Week Percentage Nay 4 .42 June 1 .83 June 2 1.67 June 3 5.42 June 4 17.50 July 1 19.20 July 2 30.58 July 3 19.58 July 4 4.20 August 1 .42 August 2 .42 72 Advantages of Growing Basmati Rice vs IRRI Rice Varieties Farmers were asked the advantages of growing basmati rice vs IRRI rice. Basmati rice was preferred by more than 80 per cent respondents (n=240) for its high price and quality of grain and straw. IRRI rice varieties had the advantage of higher yields and early maturity. Wheat can be more easily grown after IRRI rice than basmati rice varieties. Those farmers who can afford to purchase fertilizer and other inputs diversify by growing both IRRI and basmati rice. The main problem reported with the IRRI rice varieties was pest attack. The new high yielding variety KS-282 has been reported more resistant to some of the major rice pests. Present Yield Levels Rice yields on sample fields for the years 1983 and expected yield for 1984 were obtained from 76 farmers (Table 4.6). Average oasmati rice yield in the study area in 1983 was 16.81 maunds/acre, while expected yield in 1984 was 19.91 maund (one maund is approximately 82.2 pounds) per acre. These yields are comparable to those reported by (Saeed, 1980), (PERI, 1982) and (Chaudhry, 1980) for Gujranwala district, which range from 18-22 maunds/acre. Farmers reported lower yields in 1983 due to pest attacks. There was less deviation in expected yield in 1984 (33.23/mds/acre) compared to actual yield in 1983 (33.85/mds/acre) for IRRI varieties. Besides asking the yield question on specific rice fields farmers were asked past yield trends on their best, average and poor land (Table 4.7). On best land average yield of basmati rice was reported as 20.35/mds/acre while IRRI rice is 37.34/mds/acre. The yields on poor land for basmati rice are extremely low 9.5/mds/acre. The coefficient of 73 Table 4.6 Average Yield of Basrati and IRRI on Sanple Fields in the Study Area (1983/1984) (n:76) Variety 1983 1984 (Expected Yield) ------------ nfis/acre--—-—------ Basrati Nban 16.81 19 91 33 7.58 6 2 CN(%9 45 31 n 49 55 IRRI Nban 33.23 33 85 SD 13.6 11 74 CV06 41 35 n 9 16 Table 4.7 Difference in Per Acre Yields Basmati and IRRI Rice Yields on the Basis of Land Type (Nhunds) ---------- Land Type on Fanm ------—---- Statistic Best Average Poor Basmati Nban 20.35 14.71 9 5 SD 7 32 6.34 4 98 n 214 195 176 IRRI Nban 37 34 26.48 20 95 SD 10 61 9.82 10 66 74 variation in case of yields on poor land for basmati rice and IRRI rice is more than 50 per cent indicating high variability in responses. SECTION II CONSTRAINTS TO INCREASING RICE PRODUCTION A summary of constraints to rice production as percieved by the farmers is given in table (4.8). The most critical factors limiting higher productivity in the farmers opinion were the high cost of water and fertilizer, and pest problems. In Sheikhupura shortage of canal water was considered the most important factor affecting productivity. Farmers also reported dissatisfaction with the quality of diesel oil and fertilizer sold locally. Lack of credit was only considered a minor constraint in one district, Sheikhupura. The other categories include problems of uncertain electricity, erroneous electricity bills and water logging . Specific problems related to irrigation labor and farm power noted on sample rice fields are summarized in Tables (4.9—4.11). A listing of farmer comments noted at the end of each interview are summarized in Table (4.12). This table gives a general picture of the factors farmers considered needed to be addressed. This evidence indicates that farmers are more concerned about the socioeconomic factors than technical constraints. Technical Constraints as Perceived by the Researchers A review of literature presented in previous chapters and the series of visits with technical scientists revealed that very significant limitations in increasing yields on farmers fields are also technical. These factors fall in the following categories: need for improved varieties, low plant densities, poor 75 Table 4.8 Fanmers Perception of ijor Constraints to Increasing Rice Production on Their Fanms (Percentages) 1. High cost of fertilizer 35 35 48 and water. 2. l + insect attack 15.67 18 28 3. Shortage of canal water 20.5 6 7 and high cost of inputs 4. Purity of inputs and 4.8 7 5 narketing issue 5. Lack of credit 6 - — 6. Miscellaneous 18 33.7 15 . n: 83 83 75 * Note: The question was asked in such a nanner that it is not possible to separate responses. I Table 4.9 Fanm Power Problems Faced by Fanmers on Sample Rice Fields (n:76) Problem* Sheikhpura Hafizabad Daska ------------------------------------- percentages-----4--' 1. No problem ' 57.7 32 60 2. Difficulty in getting hired services 26.7 12 4 3. Poor perfonmance of hired services in Kaddu — 8 4 4. Shortage of repair services 8 28 12 5. Expensive diesel oil 7.7 20 20 76 Table 4.10 Water Availability and Irrigation Problems Faced by Fanmers on Sanple Rice Fields Problem Sheikhpura Hafizabad Daska --------------------------------- percentages-—-—--—--------—- 1. Shortage of canal water 65.4 32 8 2. Non availability of canal water 7.7 8 16 3. Expensive diesel 7.7 16 12 4. High electricity charges and load shedding 7.7 16 16 5. No problem 7.7 20 44 6. 2+3 3 8 8 4 Table 4.11 Labor Problems Faced by Fanmers on Sanple Rice Fields n=76 Problem Sheikhupura Hafizabad Daska 1. No problem 30.8 40 36 2. Shortage of Labor* 53.8 48 56 3. Higher wages demanded 7.7 4 4 4. Poor Perfonmance 3 8 - _ 5. 2+3 3 8 8 4 * Usually for transplanting. 77 Table 4.12 Comments Repeatedly Recorded by the Enumerators at the End of the Interview (n=240) 1. Bribary demands by the electricity department and banks. 2. Lack of funds. 3. Electricity supply problems. 4. Road problem. 5. Inpurity of diesel and fertilizer supplied by local sellers. 6. Water logging. 7. Canal water shortage. 8. Problems with public tubewells. 9. No arial spray. 10. High price of inputs. 11. Differences in output prices. * The order of this list does not indicate inportance or number of responses. Note: Crnnents were noted at end of each interview, cannot separate and give exact frequency. 78 fertilization, zinc deficiency, post harvest losses, and pest attacks. Although researchers have known about these problems, little has been achieved in solving them. The current package of technology designed by the Agricultural Research Council and the Punjab Rice Research Institute makes the following recommendations (Table 4.13). In this brochure which has been in existence for over 15 years, recommendations have not changed much despite the fact that new technologies have emerged in the area (rice reaper, new varieties like NIAB-9) and input and output price relationships have changed. The economic analysis for these recommendations is not available. The bi annual seminars held at the federal level have tended to be repetitive. Fertilizer experiments and agronomic trials carried out on the field are repeated year after year by a number of institutions (Soil Survey, PARC, PRRI etc). In this way these institutes have refined recommendation for the dosage and type of fertilizer for different crops. However, merely continuing the same trials (often repeated for 10—15 years) will not result in further improvements in these recommendations. SECTION 111 KEY ISSUES RELATED TO LOW RICE PRODUCTIVITY IN THE PUNJAB This section provides a discussion on the various factors which are affecting rice productivity. Plant Density Plant density data presented earlier supports the view that plant densities in rice are far below the recommended levels . Seventy per cent of the 79 Table 4.13 Recommendations of the Punjab Extension Department of Rice Production 1 Practice weed control by several puddling operations. 2 Grow only basmati, IR-6 and KS-282. 3. Use zinc sulphate. 4. Spray Nursery. 5 Use good proportion of fertilizer. 6 Use insecticide. 7 Keep plant to plant distance at 9 inches. Source: i) Department of Agriculture Infonmation, 'Rice Production Bulletin (Urdu edition), Why 1983, Lahore. ii) Shmilar recommendation in annual reports of the Punjab Rice Research Institute. 80 .‘v respondents (n:240) reported that they were not getting the desired level of plant density, even though 89 percent (n:240) were aware of the relationship between plant density and yield. Amir et al (1984) have shown with experimental data that at low plant densities yields remains unaffected by increasing fertilizer application on basmati rice (see appendix C for further discussion). The highest response to fertilizer is observed at high plant population (100,000/ac) and high fertilizer. The main issue is how to increase densities. These authors have argued that densities can be improved if the farmer is willing to supervise the operation himself. Farmers were asked what problems they faced with hired labor (Table 4.14). Poor performance of labor in transplanting was considered a major reason (16.67 per cent) for low plant densities (16.67 per cent). Those farmers who were willing to pay higher wages to the transplanters and provide other benefits in kind i.e food, tobacco, fodder were able to get higher densities. This was especially true where the transplanters were school children working over summer vacations. One future concern is that if more area is brought under rice cultivation further lowering of plant densities can be expected as per acre labor input will decrease. Given the inefficient performance of fertilizer at low plant densities we can partially explain the declining yield trend in case of IRRI rice varieties and yield stagnation in basmati rice. One hypothesis explaining why more than 22 per cent farmers do not use fertilizer on basmati is that at low plant densities it does not pay to use fertilizer. . I _ ‘Wkpt'mfi‘t 'i'. “so has mi -.u--__ 1'; bag! .5159” W . liable-4.14 Problems With Hired Labor for Transplanting “‘5'”. ”H i.“ '-~ '1 (“-=ng) __ ,' ~31 ._. .. 35.1". JVv'h' ‘.__._,_-____-_,,,__‘__________.._____________________-________’ _____ Item Percentage 1. No problem 27.0 2. Poor performance 16.67 3. High wage rate demanded 13.0 4. Shortage of labor 14.2 5 Not available on time 2.5 6. Miltiple responses 27.0 82 Varieties and Certified Seed Varietal development in rice has not kept pace with other technological changes in the country. New varieties of fine and coarse rice have. been available, but the government policy has been to protect basmati variety. This is often justified as protecting national foreign exchange earnings. Over the last few years the Punjab Rice Research institute has increased work on varietal development. However, with only one Ph.D in plant breeding the program moves slowly. Despite this shortage in manpower the institute has recently released KS-282 a short stature coarse grain variety which is reported to be more resistant to some of the common pests and gives about 10 percent higher yield than the existing IRRI rice varieties. Virtually none of the farmers interviewed use certified seed. All seed comes from the farmers carry over stocks from previous years. No study was found during the review which documents the affect of seed saving practices on rice production in the Punjab. There is ample general evidence which establishes the relationship between yield and age of seed (1). The effect of quality of seed needs to be further investigated in relation to the farmers conditions. The Rice-Wheat Rotation and Soil Fertility Level CIMMYT researchers (1984) have shown that wheat yields are affected by the crop grown before wheat i.e rice or maize. Table (4.15) shows the effect of previous crop sown on wheat yields. Wheat grown after basmati rice gives a yield of 1.63 tons/hectare which is an estimated reduction of 33 per cent compared to wheat grown after maize or other less exhaustive crops. 83 Table 4.15 Summary of ijor Differences in Production Practices in Wheat by Previous Crop Grown (1984). Basmati IRr6 Nbize and Fallow Rice Rice other crops bhnber of fields 73 37 18 24 Percent clay soils 49 51 44 33 Average number tillage 5.6 5.7 7.4 7.6 operations Percent planted after 49 22 47 46 December lst. Percent pre—irrigated 25 38 78 91 Percent fields with Phalaris weed problem 41 16 17 12 Yield (T/ha) 1.63 1.84 2.17 1.94 Source: Personal Communication, Derek Byerlee (Regional Economist), CINWNT Office, Islamabad, 1984. 84 Difference in the cultural practices which depend on the rotation are indicated. The yields of basmati rice are significantly affected if the previous crop was wheat (17 per cent), from an average basmati yield of 19 maunds (Table 4.16). This data indicates that the traditional rice-wheat rotation leads to lower yields than other rotations in the system. Pest Management This is a controversial area which has not received much attention. from the research community. The sharp reduction of subsidies on aerial plant' protection and free extension service spraying have reduced the use of pesticide on rice. The government since 1982 has decided to allow the private sector to handle the marketing of pesticides (personal communication with section officer, plant protection). Numerous reports of outdated and mixed pesticides sold to farmers were recorded during the survey. This has led to great mistrust and probably further reduction in use of pesticides by farmers. During a number of group visits and as commented by farmers, the abandoning of aerial spray has lead to decreases in the rice crop over the last few years. The governments argument (1) is that in the past aerial spray used to cause heavy loses to livestock. It is said some influential farmers complained about this problem, hence the government decided to discontinue aerial spraying. The small farmers point of view is that this decision was not made on a majority basis and that very rarely did any loss occur, because they are informed ahead of spray time by the local extension agents and took the necessary precautionary measures. Some government officials think that aerial spraying would be ineffective on the new pests. Farmers however stated that the aerial 85 Table 4.16. Summary of Differences in Production Practices of Rice by Previous Crop Grown 1984 (n:76) Number of Wheat Berseem Fallow Fields .(45) (25) (4) Average number of 11 10 14 tillage operations Average fertilizer(kg)/ac 21.4 23.48 - Percent using fertilizer 53.3 60 75 Percent practicing weed 89 88 50 control. Percent using nanure 56 68 75 Average plant density 14.3 14.42 15.33 (per sq. neter) Expected Yield nds/acre 22.79 28.32 19.83 Source: Cum survey data. 86 spraying not only saved rice from insects but also proved effective in killing mosquitoes and other unwanted pests which pose hazard for the farmer (3). At present we do not know how aerial spray effects the ecology of the rice area. Based on these preliminary findings it is time to reevaluate the current policy on aerial spray by considering the social benefits vs the potential costs. Farmers appeared agreeable to a fee to cover the cost of this service. We should also explore the alternative of adoption of manual Sprayers. Although this technology is well adopted for cotton, the factors influencing its use on rice need to be studied. Price Structure Given the general complaint of low net returns from rice, particularly basmati it is desirable to evaluate pricing alternatives in the present policy making environment. Apparently the government does not plan to increase the price of rice (4). Recently (5) the government announced a 3 rupees increase in the price of basmati rice and a 2 rupee reduction in the price of IRRI rice. Pakistan enjoys a strong international position in the high quality basmati rice market. The export of IRRI rice varieties has not been very successful, primarily because the international price has declined due to a reduction in demand from countries like Indonesia, Philippines and Sri Lanka has declined as they are now producing more rice. It is unlikely that major shifts in demand for Pakistani rice will occur. If Pakistan wants to maintain its competitive position in export of basmati rice a higher rice price or input subsidy may have to be offered to increase current production. The data presented later in this study show that acreage allocated to rice is likely to decrease if current trends 87 in production continue. It is also important that the government pay attention to reduce the current wide marketing margins. Reductions in these margins could enable either a lower export price or a higher price to farmers. Research Issues The general investment level in research on rice has been low as compared to other crops. The Punjab Rice Research Institute was formed before separation from India in 1937. More recently, a number of scientists in Rice Technology have been trained abroad particularly at The International Rice Research Institute in the Philippines and Japan. Despite these efforts, the Punjab government still remains short of trained technical manpower in agriculture especially at the Ph.D level. This issue was probed with the agricultural training officials in Islamabad. Their reply was that a number of times they have requested the Punjab government to nominate candidates for advanced training abroad in rice, but the response has generally been poor. Although in other crops the training programs are progressing on schedule. Issues Related to Water and the Punjab Farmers Response to Changes in Costs Water is the lifeblood of rice. Most canals in the rice areas are bi -annual and usually run during the summer rice growing season but there is shortage of canal water at this time. Almost all farmers who relied only on canal irrigation reported water a major constraint. Those farmers who owned tubewells felt that water was adequate on their farms. Farmers who relied on public tubewells were the most seriously in need of more water. Unnecessary bribery demands by staff from the electricity department and lack of maintenance of facilities were considered main issues by those relying solely on Public tubewells. 88 The cost of water is another controversial issue. The general consensus among farmers during the survey was that water costs are the major costs in rice production; especially for those relying on tubewell water . However, a review of several studies shows that in calculating the cost of production for rice (one of the factors taken into account in formulating price policy) water costs were underestimated, for it is assumed that farmers only pay government water taxes for canal water and incur no other water costs. The tubewell owners unanimously considered water as the main cost item in producing rice. An indirect estimating procedure is the share of the produce paid for water. The share of the water was reported at 1/3 of the yield. The tubewell owner does not share any other expenses i.e fertilizer, land preparation etc with the person who purchases water. This also indicates the high proportion of water costs in rice production. Given the increasing costs of water utilized in the production of rice through tubewell expansion one would expect some response from the farming community. Farmers are rapidly adjusting to these cost increase by reducing rice acreage (Table 4.17). Almost 48 per cent farmers reported reducing rice area during 1980—1984. Only 8 per cent reported increase in rice area on their farms . Among those reporting reduction in area 43 per cent had reduced area by 25 per cent during 1980—84, while 73.7 per cent reported decrease in area of less than 25 per cent during the same period. The tehsil wise response is summarized in (Table 4.18). This rapid change in area signals a drastic reduction of incomes from rice; and opportunities for alternative crops. While this discussion has assumed farmers responses were due to increases in water 89 Table 4.17 Percentage Changes in Rice Area as a Result of Increases in Water Charges During 1980—84 Area Decreased Total (percentage) less than 25 42.7 25.1—50 40.8 50.1-75 12.6 greater than 75 3.9 n: .103. Area Increased greater than 25 73.7 25.1-50 15.8 50.1-75 10.5 n: 19 No Change n: 118 Table 4.18 Percentage Change in Area Allocated to Rice in the the Tehsils of Sheikhupura, Afizabad and Daska During 1980-84 Per cent Change Sheikhupura Hafizabad Daska Total in Area n: 83 83 74 240 'less than 5 44.6 38 77 52.5 5.1-25 24 26.5 9.5 20.4 25.1-50 24 25.3 5.4 18.75 8 50 7.2 9.6 8 l 8.3 90 cost it is quite possible that some farmers may have taken other cost increases as fertilizer, labor into account in revealing rice area changes. Institutions The main groups involved in the production of rice are the input suppliers, marketing agencies, extension personnel and farmers. Government attempts to influence production at each stage. In case of wheat the government has shown keen interest, which partially explains why the country achieved self sufficiency in 1982. However, because rice is grown only in specialized areas and has less political significance as compared to wheat and cotton the federal government effort has been minimal. Such a lack of interest is partially represented by low investment in rice research. SECTION IV . . , SUMMARY OF FARMERS SUGGESTIONS FOR REMOVING CONSTRAINTS TO HIGHER PRODUCTIVITY Farmers in the study area suggested that reestablishing aerial spraying and a subsidy on diesel fuel and fertilizer as ways to increase rice production (Table 4.19). Aerial spraying was the most frequent suggestion and it appears in many multiple responses. In Sheikhupura however the water constraint was the most serious problem. Farmers demanded more repair services and repair of tubewells. Farmers depending on public electric’ tubewells suggested the electricity turn off (load shedding) period should be rotated among villages. At present electric supply to a given village is shut off at a fixed time. Any farmer whose turn falls in the particular period is seriously affected, often unable to continue growing rice. Rotating the supply period would partially solve the problem. There were also problems of operators not being on duty or leaving 91 Table 4.19 Fanmers Suggestions for Removing Constraints to the Production of Rice on Their Fanms (Percentages) 1. Provision of areial 12 59 56 spray and subsidy on Diesel fuel and Fertilizer. 2. Provide nere canal water, repair 46 20.5 16 services for government tubewells and tractors 3. Areial spray and increase size of 10.8 7.3 5.6 local canal outlet (nogha) 4. Provision of interest free loans 9 6 3.6 6 5. Others* 21 6 9.6 16 4 n: 83 83 74 * Includes suggestions like regularity in supply of electricity, free provision of insecticides/weedicides by agricultural department, and the establishment of rice procurement center. 92 the tubewell unattended. This resulted in breakdown and burning of the electric motor. Farmers suggested that the government make arrangements to train public tubewell operators from the village. SECTION V FACTORS AFFECTING TO THE PRODUCTION OF MINOR CROPS AND RELATED ISSUES As discussed earlier the dominant annual rotations in the rice area are wheat—rice and wheat—fodder (maize, sorhgum)—rice. Rarely do farmers grow other crops. Table (4.20) shows that 24 farmers out of 240 reported growing pulses instead of rice, while only 4 farmers grew sunflower. When asked whether. a minor crop could be added before or after rice most respondents replied no (Table 4.21). In the farmers opinion fodder (maize, sorhgum,oats) was the only real alternative. As to why alternative minor crops were not planted, farmers felt water shortage, unsuitable land, pest problems, marketing problems, risk and small size of land holding were the main reasons. Reasons for Leaving Lahd Fallow In response to this question the major problem was water shortage (Table 4.22). It is important to note that this set of data refutes the long held view that farmers only keep land fallow to improve fertility. The farmers did not consider this a very important reason. In other words there are other genuine constraints i.e water shortage which force farmers to leave land fallow. 93 Table 4.20 Crops.Grown Instead of Rice During " the Kharif Season Crop Sheikhpura Hafizabad Sialkot Total Did not respond 49 73 67 190 Pulses 12 9 3 24 Oi lseeds 3 1 — 4 Sugarcane 10 - - 3 Vegetables 6 - - 6 Multiple Response 10 - 4 4 n: 83 83 74 240 Table 4.21 Farmers Opinion About Crops Which Can Be Grown Instead of Rice (Number of Respondents) Crop Sheikhpura Hafizabad Sialkot Total Did not respond . 27 18 25 70 Maize (fodder and grain ) - 23 31 15 69 Sugar cane 1 6 14 21 Cotton - 7 2 9 Sorghum fodder 10 9 4 23 Others and multiple responses 12 22 14 48 94 Table 4.22 Reasons for Not Planting Alternative Crops (Number of Respondents) Reason Sheikhupura Hafizabad Sialkot Total n: 83 83 74 240 No Response 27 28 67 122 Insect Problems 14 22 2 38 Risky 6 l 1 8 Water Shortage 21 5 - 26 Marketing problems 1 5 - 6 Others/multiple responses 14 22 4 40- 95 Beginning of Change in the Punjab Rice Based Farming In the previous sections farmers response to increased cost of water was discussed. Farmers have responded by reducing rice acreage. The question is what alternative adjustments are taking place (or could take place) on the areas going out of rice. One interesting innovation has been the rapid shift to sunflower in parts of the Daska Rice Optimization Project Area. The Ghee Corporation delivered some sunflower seed to selective farmers and established a marketing center. The 7 villages in this Daska project are located 7-8 miles away from the villages surveyed in this study. Table (4.23) shows that the number of farmers increased from 6 to 40 during 1980-84. Area in the three villages increased from 18 acres to 165 acres or approximately 800 percent. Most of this increase in area has come from substituting for wheat while some of it has been taken away from rice. Farmers have reported that they tend to get 15—20 per cent higher yields for rice if the previous crop is sunflower. This observation is only based on interviews with 15—20 farmers, there is no concrete evidence to support this claim. CONCLUSIONS The following conclusions are drawn from the survey: A. Input Use 1. In Kharif 1984 on user farms an average of 35.5 kg/acre of fertilizer was used on Basmati and 42 kg/acre was used on IRRI varieties. A more important finding was that 22 per cent farmers did not use any fertilizer 96 Table 4.23 Area Changes After Introduction of Sunflower in Three Villages in Tehsil Daska* (1980—84) Year No. of Farmers 1Vlusa1rrainian Bubkanwala Othina Total growing sunflower. ---------- area in acres ----------- in the three villages 1980 6 11 3 4 18 1981 10 15 - 11 26 1982 15 40 2 15 57 1983 34 80 10 40 130 I 984 40 90 15 60 165 Notes: * These three villages are amongst the 7 chosen for improving yields of rice and studying multiple cropping systems. Source: Data provided by Dr. M. Asharf in charge Daska Rice Yield Optimization Program, Pakistan Agricultural Research Council, 1984. 97 on basmati rice. This is contrary to the wide held view among researchers that a high per centage of rice farmers use fertilizer. Most farmers were unaware of zinc deficiency and its effect on yield. Only 4 per cent of the , farmers reported using zinc sulphate to treat zinc deficiency. Previous studies by the Punjab Rice Research Institute and The Pakistan Research Council shows that deficiency of this important micronutrient can affect rice yields by 20—30 per cent. Water was the major cost item for rice production on farms depending on tubewell water from diesel tubewells. This cost was estimated at Rs.972/acre for the rice season. Farmers depending on Public tubewells also reported high cost and difficulty with water availability. An estimate of water cost is 1/3 of the gross income from rice. The use of insecticides, pesticides and weedicides for rice was almost non existent in the study area in 1984. Farmers only utilized insecticide on the rice nursury provided it was given free of cost or at highly subsidized rates by the extension department. Manual weed control was practiced by 73 per cent of the farmers. Formal credit use was extremely low among the farmers surveyed. Ninety two per cent farmers reported that they do not use any formal source of credit. Only 7 per cent farmers obtained credit from commercial banks, Arthis etc. '. i 98 6. Plant densities were found to be the highest in Tehsil Daska. A high coefficient of variation was found for all three Tehsils. Plant densities were found to differ from tehsil to tehsil. In Sheikhupura (55223), Hafizabad (53644) and in Daska 65789 plants per Hectare respectively. Compared to the recommended density of 101000 plants/acre). B. Yields 7. An average yield of 16.18 mds/acre in 1983 and expected yield of 19.91 for 1984 basmati rice was reported. In case of IRRI varieties, a yield of 33.85 maunds for 1983 and an expected yield of 33.23 mds/acre for 1984 was reported. These estimates are consistent with those reported by earlier studies conducted in the rice tract. C. Other Findings 8. Farmers considered socio economic constraints to be the major factors limiting increased production of rice. These factors included , discontinuation of aerial spray, , high cost of water and fertilizer, load shedding of electricity, impure inputs, marketing services and low product price. - Continusly following a rice- wheat rotation was found to reduce soil fertility and yields of rice and wheat. Farmers estimate a 17 per cent reduction in yield of basmati rice, from an average yield of 19 maunds, if the previous crop was wheat. Similarly, wheat yields were reduced if the 10. 11. .12. 99 previous crop is basmati rice. Rice cultural practices were found to differ, depending on the previous crop sown. Farmers reported receiving a 8-10 per cent less than official price for rice in the local market. Farmers in the Punjab have started to respond to increases in water costs. Forty eight per cent farmers reported reducing area allocated to rice during 1980—84. Seventy three per cent farmers reported reducing area by less than 25 per cent , while 43 per cent reported reducing area by more than 25 per cent area during the 1980-1984. Only 8 per cent of the farmers reported increase in area allocated to rice. With respect to questions about increasing cropping intensity most farmers did not judge that a minor crop i.e sunflower or mungbean could be grown before or after rice. Out of 240 farmers, only 24 grew mungbean and 4 grew sunflower. The main reasons for not growing minor crops were water shortage, unsuitable land, pest problems, risk, and small size of land holding. The evidence from this survey does not support the long held view among researchers that a major reason why farmers leave land fallow was to improve soil fertility. Farmers reported that there were many other constraints like water shortage , credit and marketing which forced farmers to leave land fallow. 14. 100 Farmers in the Daska area have started to shift to sunflower in place of both rice and wheat. It was shown that in three villages of the Daska Rice Yield Optimization Project sunflower acreage increased by over 800 per cent during 1980-84. This evidence strongly supports the possibility that sunflower might substitute in the rice areas. 1. 101 NOTES Based on a series of mimeographed material at the Punjab Rice Research Institute, some of these relationships are reported in IRRI annual reports (1977—83). Based on interviews with several policy makers in the ministry of agriculture. The Minister of Agriculture also thought that discontinuation of areial spray was justified because he had received reports of livestock losses. However, there was no indiciation that any formal investigation into the nature of the problem had been carried out. In a memo to Dr. M. Yausaf (September, 1984) member plant Sciences, PARC, I strongly argued that this issue needs to be investigated and we should be careful in drawing any conclusions based on a few reports. During the field work numerous farmers complained that since the discontinuation of aerial spray, the local pest population during Kharif season has increased so much that they were unable to sleep at night. Many farmers thought that this drastically effected their work efficiency. As stated by the Federal Minister for Planning, Dr. Mahbubul Haq and other government officials at various forums. Also evident from the recent (september) announcement of procurement price of basmati and IRRI, the price of basmati paddy was increased by 2 rupees while the price of IRRI was reduced by three r'upees. 102 CHAPTER 5 ECONOMIC ANALYSIS OF ALTERNATIVES FOR INCREASING THE INCOME OF RICE-WHEAT FARMERS IN THE PUNJAB The purpose of this chapter is to describe the technique used to study the economic consequences of introducing new more profitable technologies and changing government factor/product price policies at the farm level. This Iobjective was achieved by modeling two representative farms. The choice of analytic technique, description of the model farms and structure of the programming models are discussed in this chapter. The Study of Crop Substitution In the preceeding chapter different constraints to rice production were reviewed based on the farmer survey. A number of factors i.e high input cost , low plant density, government plant protection policy, water shortage, low product price were identified as probable constraints limiting rice yields in the study area. In the field study, farmers reported low returns for rice as compared with other crops. There was evidence also that farmers had reduced area allocated to rice over the past 4 years, due to increases in input cost, particularly water. As the government continues its policy of removing subsidies on inputs, the profit margin will further decrease, assuming continued government price control. It is therefore important to obtain better information about the impact of (l) introducing new cost reducing technologies '1 103 generated by research and (2) of changes in the relative prices of different crops on farm incomes and cropping patterns. One additional questions to those identified in previous chapters will be explored. It is that of incorporating pulses during the fallow period between wheat and rice. As alluded to earlier, the main stress in the sixth year plan (Government of Pakistan, 1983) was on finding ways to increase domestic production of oilseeds and protein based crops. The need arises due to the large ' increase in foreign exchange payments for oilseed imports. Similarly protein deficiency is a common problem of LDC's and Pakistan is no exception. The possible roles of oilseeds and pulses in rice based farming systems is not well understood. From a policy point of view, it is imperative that we gain knowledge of the possible effects of these crops on existing cropping patterns. With increasing water charges, it is also important to identify profitable cropping patterns which utilize less water than rice which could contribute to greater farm income. Choice of the Analytical Technique There are a number of techniques which can be used to study small farm constraints and determine the optimal allocation of resources. Budgeting and linear programming have been widely used to address this question. Both have their merits and demerits which are well stated in texts by Heady (1958), Dillion and Hardark‘ar (1980), Harsh et.al. (1981). LP is a mathematical procedure used by decision makers to solve resource allocation problems (Barker, 1964). It is used to minimize or maximize a linear objective function subject to a set of constraints. Budgeting has similar uses. According to Bradford and Johnson (1953), "a budget is the summation of all 104 expected products (or increased inventories) times their respective price less the summation of all items used in production. The budgeting approach has been widely applied in simplified programming to prepare optimal farm_ plans. Budgeting allows the comparison and weighing of different plans against each other and provides a rationale for choosing the one that appears the best (Agarwal and Sharma, no date). Because budgeting is simpler and less costly than LP in terms of computation requirements and professional expertise, it has often been prefered by farm managers. However recent advances in high speed computation has reduced the cost of linear programming and made it more practical for farm analysis. LP provides more information for whole farm analysis. It also has the capacity to incorporate numerous activities and constraints which are often difficult to.include in budgeting. More recent advances in the methodology allow the incorporation of risk (i.e MOTAD) and optimization over time "dynamic programming" (Dillion et a1, 1977). LP and Budgeting have similar assumptions which are listed below: 1) Divisibility. All factors are assumed to be perfectly divisable. Hence labor can be used for a fraction of an hour and draft power can be utilized in fractional units . 2) Linearity or Additivity. All inputs are assumed to have a linear relationship with constant marginal physical product. Constant returns to scale and fixed factor proportions are assumed. 3) Perfect Knowledge. .It is assumed that the decision maker has perfect foresight of the resources required in the production of a given activity, inputs and output are known a priori. All coefficients are known with certainity. In other words the model solution is deterministic. 105 4) Non Negativity. None of the activities can be produced in negative amounts nor can resources be used in negative quantities. Besides the assumptions listed above, LP has several other limitations. LP does not take into account the risk factor . In some cases this may be a very important consideration . Even the modifications in more advanced techniques i.e MOTAD and stochastic programming the risk factor is not fully accounted for. An LP program maximizes income from a set of activities which gives the highest returns. In many situations the farmer may have certain preferences which change his planned course of action. Moreover LP is a prescriptive tool. It tells us what the farmer should do given a set of constraints and a prespecified decision rule (objective function). Hence the solution obtained from an LP analysis may-not be exactly similar to the plan of action a farmer follows. In reality analytic techniques oversimplify the real world. Therefore it is important that the analyst be cautious in interpretation of the results for policy analysis. Perhaps the direction of change is more important and meaningful than the absolute values obtained from a model. Both LP and budgeting require the same information for analysis i.e, prices, yields, per unit cost. Farm plans formulated by either approach give similar results. Although those obtained with LP are more accurate (Agarwal and Sharma, no date). This may not be true if an LP analysis ignores important constraints or activities. The main advantage of LP over budgeting is that it provides more information than the budgeting approach. A typical computer LP algorithm will provide the following information. 106 1. Value of The Objective Function. This information tells us which activities are included in the optimal solution, at what levels and the net income from a typical farm plan. 2. Marginal Value Product. This information distinguishes LP from simple budgeting . Marginal value product or shadow price of a resource tells the contribution of forcing an additional unit into the optimal solution. MVP or scarcity value of a resource is only possible if it is a constraint in the model. Surplus resources have a zero MVP. 3. Cost of Forcing in a Non Optimal Activity. LP also tells us the cost of incorporating a non optimal activity into the solution. Given the above considerations, LP is more appropriate for this study than simple budgeting. It not only provides more information but can effectively be used to investigate alternative plans through sensitivity analysis precisely and at low cost. If LP is to be utilized efficiently in farm planning the analyst has to weigh the cost of constructing large complex models against the expected benefits. Often simpler models can adequately handle the most important variables and resource constraints (Agarwal and Sharma, no date). LP has been effectively employed to study small farm problems in both developed and developing countries. An excellent bibliographic source for older LP studies is provided by (Riley and Gass, 1958). Some recent examples from LDC'c include the following. Heyer (1971) used LP and addressed 3 questions related to constraints faced by peasant farmers in Kenya. The questions include: 107 a) Optimal product mixes and optimal production strategy. b) The effect of innovations. c) Problems that need research and solutions. d) Shadow prices of critical resources. Sektheera, (1979), employed linear programming to find the possibility of incorporating new, multiple cropping patterns into the existing farming systems in Thailand . She found that the proposed multiple cropping system must be rice based. The dominant role of exchange labor implied that it would be difficult for one farmer to accept an innovation, if it was not generally accepted by the whole commmunity. Kamuanga, (1982) used LP'to test the profitability of improved practices in the rice production system of Mali. His results suggest that optimum farm size was 3.67 hectares and farm income could be increased if row seeding and mechanical weeding techniques were adopted. Eponou (1983) studied the impact of increasing the price of rice, subsidizing rice inputs and providing credit to producers in the Ivory Coast through LP . He found that with current constraints providing further credit would greatly enhance yields in the Ivory Coast. Minde (forthcoming, 1985) studied the non-farm employment potential in Tanzania. The main objective of this study was to‘determine what alternative production activities could be undertaken to improve income in rural areas of lowland Morogoro. This study shows that mandatory requirement to produce cotton in the Morogoro area reduced farm incomes. 108 In the Punjab a number of authors have used LP to_ study different problems. The most noteworthy are those by Gotsch, (1975). These studies by Gotsch and his associates focused on mechanization, credit and income distribution. One conclusion of these studies was that future research in Punjab should focus on water management at the farm level. Approaches for Constructing Representative Farm Models There are two approaches normally followed in obtaining data for LP (Young and Richards, 1978). First, one can take average values of inputs and outputs from survey data and use them to construct representative farm models. The second approach is to select a representative farm based on the researcher's judgement and derive parameters from that farm. The later approach has certain distinct advantages. The analyst is at least certain that the farm chosen is a representative farm based on the his/her best judgement, utilize actual resources and the objective function, is specific to a farmer. In sensitivity analysis this approach can be supplementedwith cross sectional data from surveys. The draw back of this approach is that it is very difficult to define a representative farm and requires considerable familiarity with the farming systems and socioeconomic setting of the study area. The average approach uses average values and can represent large data sets. However many assumptions have to be made in constructing a hypothetical farm. And it is very unlikely that one can effectively model a farm which represents all conditions on a real farm. 109 The Choice of Two Representative Farms Given the relative advantages of the representative farm approach, a decision was made to model two farms. The justification of restricting the modeling exercise is that to carefully obtain information on a representative farm involves considerable time and cost. Both these resources are limited in a dissertation study. Because this study seeks to analyze alterative crop technologies and is restricted to heavy soils it was judged that modeling two farms will provide an adequate description of the situation. The analysis is carried out for small farm conditions. The survey results reported in chapter 4 show that more than 60 percent of the farmers in the rice study area cultivate less than 12.5 acres of land. Thus, it is this group that needs to be studied. Considerable time was spent on making sure that the farms selected were representative of farming conditions. In this process 10 farmers were chosen randomly from Hafizabad and Daska tehsil. Sheikhupura was not included for modeling as it appears to be similar to Gujranwala with respect to tenure, land holdings, cropping patterns and constraints. The 10 farmers were intensively interviewed to gain insight to their peculiar constraints , resource base and dominant crop production activities. Information on prices, input use, cropping pattern , sources of farm power,constraints levels, objective functions, etc. was obtained (table 5.1). To make the final selection information, obtained through the survey work reported in chapter 4 was supplemented with information from other macro level studies. Consultation with other researchers and extension officers helped make this choice. Perhaps the most helpful was the researchers familiarity with the situation as an observer of rice farming in the Kalar tract 110 Table 5.1 Characteristics of Ten Fanmers Interviewed in Hafizabad and Daska for Background Infonmation and Choice of Representativeness Fanmer Fanm Soil Source Source Crops Adult Wheat Rice No. Size Type of of Grown Wbrkers Yields Yeilds (acres) Power Irrigation Mas/Ac NUS/Ac Hafizabad 1 8 Excellent Hired TubeWell Wheat,Rice 2 21 21 Tractor Sugarcane, Fodder, vegetables 2 7 Good Chm Tubewell As Above 2 15 17 Tractor + Cotton 3 13 Good and Bullock Tubewell As Above 3 17 19 Excellent + Canal No Cotton 4 8.7 Good Bullock Tubewell Same Above 2 18 20 + Senji* 5 6.5 Excellent Bullock Tubewell Same Above 2 22 23 No Senji Daska 1 4 Good Bullock Tubewell Rice, Wheat, 3 16 18 Fodder 2 6.3 Excellent Bullock Tubewell Same Above 2 19 24 3 12 vMixed Bullock Tubewell As Above 4 15 19 ' + IRRI 4 2.4 Excellent Bullock Tubewell .As Above + 2 19 26 watenmelon 5 5 Excellent Bullock Tubewell As Above 2 18 19 No Melons * Senji is a fibrous crop used to nake string. Note: Mixed neans average, excellent and good type of land. 111 over a period of 8—10 years. The model farms were chosen to represent as much as possible the average small farm in the tehsils of Hafizabad and Daska._ Source of Input/Output Data and Quality of Data Data for the representative farm analysis was obtained from three sources; interviews with representative farmers, data from the farm survey, experimental results and government agencies involved in policy formulation. 1. Farmers' Interviews. A series of intensive interviews were held by the author with the two representative farmers to obtain information on their constraints, resource base, objective functions, etc. No formal questionnaires were used. The procedure followed was interactive and iterative, allowing the farmer and researcher to communicate in an informal manner. The input output information obtained was verified against different farm management studies (PERI, 1980), (Khan,l975), (Saeed, no date) and (Chaudrey, 1980). Special emphasis was placed in these farm visits on the quality of information. by including numerous cross checks during the questioning sessions. 2. Experimental Result . The second source of data was experimental work by the Rice Research Institute and research institutions at Faisalabad. The information obtained from these sources is used to study the economic consequences of new cropping patterns which incorporate sunflower, mungbean and new rice varieties. The PARC rice program which in collaboration with the PARC pulses program carried out a number of agronomic trials on short duration mungbean varieties (55—60 days maturity) provided by NIAB, Faisalabad. The objective of these trials was to test the performance of these varieties in rice areas. They were carried out in late May 1984, whereas the recommended date is immediately after wheat harvest in April. The delay 112 resulted because the seed was received late from NIAB scientists. The trials were carried out by farmers under the supervision of an agronomist in .5-1 acre plots. I visited the four trial sites 4-5 times to observe the performance of these varieties in the rice areas. No pest problems were encountered in Sheikhupura and Daska sites. In Hafizabad the crop showed virus attack (Report of NARC plant pathologist and entomolgist visiting the area). The Nuclear Institute for Agriculture and Biology at Fasialabad provided two years on—farm data from the rice areas for short duration mungbean (personal communication). This information was supplemented with some Cropping Systems Experiments conducted by the Punjab Rice Research Institute, (personal communication). This is used to derive input/output coefficients for senstivity analysis in the next chapter. 3. Other Government Agencies. Information on various long term policy measures was obtained from the Agricultural Prices Commission and other government agencies. The researcher also had an opportunity to discuss future policy trends with a number of senior officials in the agriculture ministry. BASIC INFORMATION ON THE TWO REPRESENTATIVE FARMS Farm One ( Hafizabad) This farm is in village Udowali in tehsil Hafizabad. The village is located 4-5 miles away from the main road. While the farm is located approximately 1- 1/2 miles away from the village. The farmer and his wife share their house in the village with their son who is married and has 2 children, an 8 years old son and a 2-1/2 year old daughter. The total land holding of the farmer is 10 acres. No land is rented out or rent in. Most of the time the father and son remain on the farm as their livestock is sheltered there. The farmer has recently 113 purchased a tubewell with a loan from another son working in Islamabad. 1n the past he has experimented with a different crops including sugarcane, rape seed, gram, etc. However, at present the major crops grown on his farm are wheat, basmati rice, Kharif fodder and berseem. His cropping pattern for 1983-84 is shown in table (5.2). The farmer owns a pair of bullocks, 2 buffaloes, one donkey and 2 calves. When I inquired from the villagers how they would rate this farmer, the villagers including the numberdar considered him an average representative farmer of the area in terms of land holding, cropping pattern, labor availability, source of irrigation etc. Objective Function The objective function is a summation of all crop production activities multiplied by their respective prices minus the variable cost .The following activities have been included in the programming analysis based on the farmers subsistence and income goals. 1. Basmati rice (July-Nov) 2. Wheat after basmati rice (Dec-April) 3. Sugarcane (Jan—Dec) 4. Wheat after fallow (August-April) 5. Kharif fodder (July-Nov) 6. Rabi fodder (Dec—April) 7. Leasing out land (yearly basis) The farmer was requested to comment on his objective function, whether it is meeting subsistence requirements, maximizing leisure, or maximizing income. The farmer replied he wanted to assure that his minimum subsistence for wheat and rice was fulfilled. Only after his subsistence is guaranted would Table 5.2 Cropping Pattern of Farm One (Hafizabad) (1983-84) -———--_---—------_—-——-———‘—_———————————----—_ Basmati Kharif Fallow Kharif fodder Sugarcane Kharif Total Wheat fallow Wheat after rice Rabi Fodder Sugarcane Rabi Rabi Rabi Rabi and Khar if 115 he allocate land to any other crops. The farmer reported that he could lease out land at Rs. 500/acre to his neighbors. Activities The farmer traditionally grows 4-5 acres of basmati rice and has experience with growing IR—6, but no longer grows IR—6. He grows 5-6 acres of wheat for home consumption and utilizes the straw for his animals. Both these requirements are explicitly incorporated in the model as constraints. The farmer requires winter and summer fodder for his livestock. Minimum constraints for these requirements are discussed later in this chapter. As the farmer owned his tubewell he was asked whether there were any opportunities to sell water to other farmers. The farmer thought that because all his neighbors had their own tubewells it would be difficult for him to sell any water. Crop Yield Estimates Crop yield estimates for a normal year were obtained from the farmer, by taking him to each of his fields and requesting him to comment on the yield obtained from that particular patch of land. As most farmers in the area harvest and thresh the produce of 3-4 acres together and cannot explicitly formulate any opinion about per acre yield an indirect measure was adopted. This involved asking the farmer his total production from his different fields, and dividing this by the area. Validity of these estimates were confirmed by cross checking the data with the survey results as well other studies conducted by Saeed (no date) and Chaudry (1980) which also provides yield estimates for the area. The farmer is progressive in outlook, hardworking and desirous of 116 improving his economic well-being. He reported that his yields would be relatively higher than those obtained by others in the area as he spends most of his time on the farm and tries to ensure that required operations are performed at the right time. After discussing yield estimates with the farmer the researcher also visited his nearby neighbors, for a discussion and to get some idea about the representative farmer's estimates of his yields. In general these were confirmed by the neighboring farmers who gave further assurance of the validity of these estimates. There are generally no good estimates available for yields of 'Rabi' and 'Kharif' fodder. However, as these 2 types of fodders are sold on a per acre basis, prices prevalent at different periods . for both these fodders were obtained. The price received by the farmer at the peak selling time in Rabi 1983 and Kharif 1984 are used in this model. This is justified because the decision to sell is made when the crop matures. In case of winter fodder the decision to sell is made before the first cutting. Similarly for Kharif fodder (i.e ' maize, sorhgum) it is sold on basis of the whole crop and not by single cuttings. Hence the price which prevails before the first cutting is a reasonable price to use. Crop Prices The prices used are those obtained by the farmer at harvest time in 1983—84 season. The farmer reported that after meeting his subsistence requirements, he immediately disposes of all surplus. The prices received by farmers from the local buyer 'Arhti' are usually 8-10 per cent less than those reported as the official procurement price. The farmer reported that last year he was able to sell basmati paddy at a price of Rs. 85/-per 40 kg. even though 117 j the official rate was Rs. 90/—per 40 kg. These prices were also cross checked with the survey data reported earlier and were found to be reasonably representative. Estimates of Unit Resource Requirements The LP model incorporates parameters representing per unit requirement for each resource, and in this model 4 resource constraints are included: 1. Land The farmer owns 10 acres of land. Amongst this area 9.7 acres are cultivable, whereas the remainder is used for animal sheds. The model structure discussed later has 26 land constraints. It has two seasonal land constraints for Kharif and Rabi land and 24 bi monthly constraints to capture land use for different crops. The land is quiet homogeneous and of excellent quality. 2. Labor The farmer and his son work on the farm. They felt that their labor was adequate to handle most crop operations and manage livestock during the regular season. Labor is hired only during peak periods, especially during rice transplanting, wheat harvesting ,threshing, sugarcane seeding/weeding and ‘Gur' making, etc. The farmer reported that during summer (May to October) both father and son work 13 hours per day, while in winter they work only 10 hours. Six total hours were deducted from each day for livestock maintenance. Thus it is assumed that the two farmers work 10 hours during summer and 7 hours during winter on his crops for approximately 12.5 days in each 15 day period. A 12.5 days work routine in each 15 day period allows for incidental visiting, 118 holidays and sick leave.The farmer was asked to elaborate on this matter, he felt that the assumption was reasonable although he was not certain about the exact timing of the holidays he takes.The model includes only the months of January, March, May, June, July, August, November when labor is a constraint. Similarly hiring during peak labor demand, during rice transplanting, wheat harvesting/threshing and sugarcane weeding and 'Gur' making have been. taken into account in calculating gross margins. Usually labor is hired on task basis i.e transplanting. The rate during 1984 Kharif an 1983 Rabi were used in to estimation of gross margins. Detailed labor requirements for each crop were derived through indepth interviews followed by a series of follow-up visits to determine the labor requirements for each operation during different periods. Labor requirements for each activity are measured in terms of man/hours per acre. This profile was then re-examined in light of different studies on labor utilization (Aslam, 1980 and Saeed, no date). As labor requirements are broken down on bi-monthly basis, this helped the farmer to identify different operations performed in each period. Most of the responses were given on the basis of a 'Desi' or local calender. The 12 'Desi' months were converted into the seasonal calender. Based on this calender shown in (appendix table A 5.1), the different requirements for hired labor during peak seasons were estimated. Wage rates paid by the farmer differ by activity. Both the cash payment and the value of the payment in kind were incorporated'in determining the wage rate for labor. Because availability of hired labor was not currently a problem possible shortage are only taken into account in the sensitivity analysis. 119 2. Water Water requirements for each crop activity are based on the average number of irrigations required to irrigate a one acre crop field at different periods. This is measured in number of hours it takes to water a 1 acre crop field. The farmer owns a tubewell which has a capacity to run continously for 15 hours a day during the peak season. This farmer has a 2 hour allocation of canal water each week during May and October. The farmer did not feel that water was a major constraint on his farm; however, due to the high cost of diesel oil, he felt that his main limitation was the cost of pumping water and not the capacity of the tubewell. 3. Draft Power The farmer owns a pair of bullocks which can work 6 hours a day for approximately 24 days a month. A six day rest period is assumed. Each bullock can plough and plank an acre of dry field in 6 hours and a watered rice field in 8 hours. However, during the rice season this animal work load can be increased to 12 hours. The farmer also reported that at peak demand periods there was sharing/exchange or "mangi" practiced among the farmers. This helps partially remove the draft power constraint during rice land preparation, especially for rice puddling. There is no payment for mangi service as long as 2—3 farmers make prior arrangements. The work rates for the bullocks reported by farmers are normal and have been verified with earlier studies by Aslam (1980) and Ahmed (1972). The farmers can also hire tractor services at the rate of Rs 35/acre per ploughing and Rs. lO/acre for planking when required. However he prefers not to hire tractor services. The farmer felt that he had sufficient 120 draft power. Therefore in the model tractor services were not included. The farmer also owned 2 milking buffaloes, which were not used for draft purposes. Estimates of Gross Margins Gross margins on the basis of activities are estimated by first multiplying crop yield with prices received at the farmgate at harvest time. To this is added the value of any by-products such as straw. From this is deducted the variable cost which includes items such as hired labor, fertilizer, farm yard manure (FYM), seed, fuel, land revenue and tax, threshing, marketing, etc. Detailed derivation of the gross margins for the Gujranwala representative farmer are shown in Table (5.3). Structure of the Model Based on the information presented above a linear programming model of the farm was constructed. The mathematical format of the generalized model used in this study is given below. MAX Z = f(x) = C'X Subject to x 0, A X S Where 2 is the objective function to be maximized m x 1 vector of gross returns and prices n x 1 vector of activity levels > X 0 11 mx matrix of input - output coefficients 5 = m x 1 vector of resource constraints Table 5.3 Gross Whrgins/Acre for Hafizabad Representative Fanm ———-__--------__--______..-----_-----_-_..__----....-—-.--_--___--—-----_-----—----—— Pricefimd Total Straw Income 250 100 97 Gross Seed/ Income Nursery FM“ 2052 960 2400 4000 Trans- planting Others* Crops Yield/ NH Acre Rabi 1. Wheat 25 nds (Fallow) 2. Wheat 10 nds after rice Kharif 3. Rice(Bas) 23 nds 4. Kharif - fodder 5. Rabi - fodder 6. Sugarcane 40 ads (Rabi & Kharif) Cost (rs.) Fe't water Harvest- ing* 1. 126 112 - 2. 126 112 - 3 130 969 - 4. 32.9 - 5. - 165.34 - 6. - 366.23 520 Income 65 1625 65 650 85 1955 @ 120/ - kenal @ 300/ - kenal 100 4000 Thres- Weed- ing ing 24.76 - 24.76 - 221 60 82.9 - 320 - 70 50 70 50 50 36 120 182.22 100 TW: Gross xhrgin 709.76 1455.24 539.76 330 1364 391 273.92 816. 520.84 2089.16 1888.45 2461.55 ‘ Other than sugarcane, ail other harvesting operations with family labor. 122 The algorithm used to for the LP analysis takes into account the slack variable for the less than resource constraints, the surplus for the greater than constraint. The model as specified above is in a generalized form as opposed to 'standard form' which also shows slack variables (Anderson and Swenny, 1983) The tabular presentation of the model shown in table (5.4) is based on the information presented in the preceeding sections. Objective Function (CJ's). Six crop production and 1 land leasing activity. Constraints. Six types of constraints are incorporated. 1) Land availability during Kharif and Rabi Season during critical 15 day periods in each month. 2) Family labor availability of labor in each 15 day period. 3) Restriction on hired labor during each 15 period. 4) Irrigation constraints in different 15 day periods 5) Draft power constraints in each 15 day period . 6) Minimum and maximum constraints for each crop production activity. Minimum and Maximum Constraints The farmer has the following minimum constraints to meet: 1. He requires at least i) 40 maunds of wheat ii) 30 maunds of rice 123 Table 5.4 LP Tableau for Farm One (Hafizabad) (1984) Activity Row Resource Uni t Bas theat Baswheat KF RF SC Lease Sign RHS No. ---------- Objective Function (Rs.)----- - Cut -------------- 361 1455 330 816 2089 2461 500 Land Constraint 1 Rbland acre 1 l 1 l 1 LT 9.7 2 Khland acre 1 1 l 1 1T 9.7 3 Janl land acre 1 l 1 l 1 LT 9.7 4 JanZ land acre 1 l l l 1 LT 9.7 5 Febl land acre 1 1 l 1 1 LT 9.7 6 FebZ land acre 1 l l l 1 LT 9.7 7 Marl land acre 1 l 1 l 1 LT 9.7 8 Ner land acre 1 1 1 l 1 LT 9.7 9 Aprl land acre 1 l l l 1 LT 9.7 10 AprZ land acre 1 l l 1 1 LT 9.7 11 Mayl land acre 1 l 1 LT 9.7 12 May2 land acre 1 l 1 LT 9.7 13 Junl land acre 1 l 1 LT 9.7 14 JunZ land acre .03 l 1 1 LT 9.7 15 Jull land acre 1 1 l 1 LT 9.7 16 JulZ land acre 1 l l 1 LT 9.7 17 Augl land acre 1 l l l 1 LT 9.7 18 Aug2 land acre 1 l l 1 1 LT 9.7 19 Septl land acre 1 l 1 l 1 LT 9.7 20 SeptZ land acre 1 l 1 l 1 LT 9.7 21 Cctl land acre 1 l l l 1 LT 9.7 22 Oth land acre 1 l l 1 1 LT 9.7 23 Novl land acre 1 1 l 1 LT 9.7 24 Nov2 land acre 1 l l 1 LT 9.7 25 Decl land acre 1 l l 1 LT 9.7 26 Dec2 land acre 1 l l 1 LT 9.7 Labor Constraint 27 Janl Labor Man/days 43 4O LT 200 28 JanZ Labor Man/days 7 7 43 4O LT 200 29 \hrZ Labor Nan/days 25 4O LT 200 30 Mayl Labor Man/days 6 6 18 43 3 LT 200 31 MayZ Labor Van/days 6 LT 250 32 Junl Labor Man/days 3O 6 3 4 LT 250 33 JulZ Labor Man/days 35 6 32 LT 250 34 Augl Labor Man/days 15 25 LT 250 35 Nov2 Labor Man/days 22 12 LT 200 36 Decl Labor - Man/days 6 21 LT 200 37 Jan2 Hlabor Van/days 252 LT 500 38 DecZ Hlabor Man/days 252, LT 500 39 MayZ Hlabor Elan/days 3O LT 500 40 JulZ Hlabor \bn/days 4O LT 500 41 Apr2 Hlabor Man/days 4O 4O LT 500 Note: 1 : First 15 days of a mnth and 2 : second 15 days of a month. 124 Table 5.4 (Cbntinued) Bas 361 KP RF 3: ------------ ADbjective Function (Rs.)-------- 816 2089 2461 500 Lease Sign RHS CUT ------------- Irrigation Constraint hours hours hours hours hours hours pair/hrs pair/hrs pair/hrs pair/hrs pair/hrs pair/hrs pair/hrs pair/hrs pair/hrs pair/hrs pair/hrs pair/hrs pair/hrs acres acres acres acres naunds rnaunds acres acres acres 8 15 15 35 23 theat Baswheat 1455 330 6 6 6 6 6 6 6 12 12 18 25 15 18 18 18 60 6O 24 12 12 LT LT LT LT LT LT LT LT LT LT LT LT LT LT LT LT LT LT 92 192 192 192 188 188 96 96 96 96 96 96 196 96 96 96 96 96 \IOLI'IKII W4? r—‘NOowr—NN .. .... WHO 42 Jull water 43 JulZ water 44 Augl Whter 45 Aug2 water 46 Nov2 water 47 Decl water Bullock Constraint 48 Janl Power 49 Janl Power 50 Febl Power 51 Feb2 Power 52 NhrZ Power 53 be2 Power 54 Jull Power 55 Augl Power 56 AugZ Power 57 Sepl Power 58 SepZ Power 59 Nov2 Power 60 Decl Power Max lrm and Min irra 61 KH Nbx 62 RF th 63 SC th 64 Lease Nbx 65 Wheat Nfin 66 Bastati Nhn 67 KF Nfin 68 RF NHn 69 SC Nun Note: Nfinz nuninum requireTent of fanTer, max: Nbxinlntrequiretent or willingness to grow. l25 iii) 1.5 acres of Kharif fodder iv) 2 acres of Rabi fodder v) .5 acres of sugarcane crop which he kept as ratoon. 2. Maximum amount of Kharif fodder is 2.5 acres 3. Maximum amount of Rabi fodder is 3 acres 4. Maximum acreage of sugarcane he is willing to grow is 1 acre Farm Two (Daska) This farm is located in Daska, District Sialkot. It is situated almost 4 miles off the metaled (pucca) road. The farmer and his brother jointly own the farm. The older farmer is the main decision maker. Their own land holding is 8 acres and they have rented in 5 acres at the fixed rate of 7.5 maunds of Basmati paddy per acre annually. Among the total 13 acres, 4 acres are excellent 'mera' land suitable for cultivating rice, vegetables and sugarcane. This land, however, has minor patches of salinity. The five acres hired are good to average land mainly suitable for rice and wheat. His farm buildings are spread out on 1 kannal of land and major assets include : a 16 HP tubewell, a fodder cutter, 12 trees and 5 animals. Objective Function The farmers objective functions were thoroughly investigated and can be summarized as a summation of the following comments:— 1. Growing basmati rice with farm yard manure on excellent and good land. 2. Growing basmati rice without farm yard manure on average and good land. 126 3. Growing IRRI rice with farm yard manure on excellent and good land. 4. Growing IRRI rice without farm yard manure on average and good land. 5. Growing wheat on excellent land. 7. Growing wheat on average land. 8. Kharif fodder. 9. Rabi Fodder. The farmers main objective is to meet his subsistence requirements of wheat and rice and also to grow enough rice to cover the cost of rented land. Cash is a constraint in the sense that it is not freely available from banks. However, the farmer reported that he could obtain as much credit as desired from the informal sources especially the 'arhti.‘ Hence credit is not considered a constraint in the model. Activities The farmer grows both basmati rice and IRRI rice. He grows wheat for home consumption. Straw or ‘Bhosa’ from wheat and rice are utilized by his animals. His last year's cropping pattern is shown in Table (5.5). His minimum requirements for wheat and rice are explicitly incorporated in the model. As the farmer owns his own tubewell, water is not a constraint for him. The farmer does not have access to canal water. Crop Yield Estimates Crop yield estimates were obtained from the farmer. He was asked to comment on yield differences based on land types. His estimates of yields were then reconfirmed from 3 other'farmers located near his farm. From these 127 Table 5.5 Cropping Pattern of Farm Two (Daska) (1983-84) Crop Season Acres Basrmti Kharif 4 IRRI .~ Kharif 4 Kharif fodder Kharif 2.1 Kharif total 10.1 Wheat Rabi 11 Rabi Fodder Rabi 1.5 Sugarcane Rabi and Kharif .2 Rabi total 12.5 Note: Cropping intensity is 175 per cent. 128 estimates it appears that the farmer's reported yields were in conformity with those provided by the other neighboring farmers. Estimates of Gross Margins Gross margins are estimated in the same manner as for the first farmer reported earlier. The detailed calculations are shown in Table ( 5.6). Crop Prices The crop prices used in the programming model are those obtained by the farmer at the time of harvest in 1983 (for Rabi crops) and 1984 for Kharif Crops. This information was compared to that obtained from the survey data reported in chapter 4 and found reasonably representative of the general prices received by farmers. This farmer sells most of his surplus to the local 'Arhti' and gets a relatively low price (8—10 per cent) for his produce. The discount may partially be interpreted as a shadow price for interest charged on the credit extended‘to the farmer. The farmer did not consider this as an interest payment and thought that the price was lower because of the supply demand situation, immediately following crop harvest. Because he purchases inputs on credit with a promise to sell his produce to the same arthi, he can not sell in the open market. If he was to sell elsewhere he might get a slightly higher price. The farmer did not consider this very important compared to his access to unlimited inputs. Estimates of Per Unit Resource Requirements 1. Labor All three male members of the family, the farmer, his brother and son work on the farm. The females in the house do not work on the farm. The 129 Table 5.6 Gross Nhrgins/Acre in Rupees for Daska Fanm. Crops Yield/ Price/NH Grain Straw Gross Seed/ NH Acre lncoTe + Income IncoTe Nursery ------------------------ Rs.--------—------------------ Wheat ( Rabi) 1. Wheat on excellent 32 65 2080 300 2380 100 land * 2. Wheat average 16 65 1040 150 1190 100 land Rice Basmait (Kb) 3. Excellent land 19.5 85 1657.5 50 1707.5 40 with FYNL 4. Excellent land 18 85 1530 50 1580 40 w/o FYNI 5. Good land w/Fth 13.5 85 1147.5 35 1182.5 40 6 Good land w/o FY81 12 85 1020 35 1055 40 Rice IRRI (Kharif) 7. Excellent land 35.5 52 1846 30 1876 40 with FYAL 8. Excellent land 33 52 1716 30 1746 30 w/o FYM 9. Good land 20.5 52 1066 30 1066 30 with FYM 10. Good land 18 52 936 30 966 30 w/ 0 RM 11. Kharif Fodder — - - - 800 28 12. Rabi Fodder Wheat — - — - 1600 100 t The fanter did not report yield differences due to fallow land. growing rice after wheat, etc. However, the dominant rotation followed on the land is rice after wheat. 130 Table 5.6 (Continued) Crops Fert Water Harvest Thresh- Whed Trans- Other* TVC Gross lizer ing ing ing plant Nhrgin ------------------------------- Rs.-----~-------——---—-----—-—------~—-—- Wheat (Rabi) 1. Wheat on excellent 187 124.38 195. 49.56 - - 17 792.9 1707 land 2. Wheat average 187 124.38 195 49.56 — - 17 792.9 517 Land Rice Basmait (Kb) 3. Excellent land 128 972 100 - 100 90 35 1465 242 with FY“. 4. Excellent land 259 972 100 - 100 90 35 1596 -16 w/o Ffifll 5. Good land w/FYNI 128 972 100 — 100 90 35 1465 -282. 6. Good land w/o FY81 259 972 100 - 100 90 35 1596 Rice IRRI (Kharif) 7. Excellent land 193 650 100 - 100 90 35 1168 678 with FYNL 8. Excellent land 259 650 100 - 100 90 35 1234 512 w/o FYNl _ 9. Good Land 193 650 100 - 100 90 35 1168 —72 \vith FYM 10. Good land 259 650 100 - 190 90 35 1234 -268 w/ 0 FYM 11. Kharif Fodder 105 131 - - - - 30 266 534 12. Rabi Fodder - — 175 - - - 35 210 1390 Notes: 1. TVC is total variable cost. 2. No difference was reported on costs of rice production on excellent land, good land, although the differences in yield are noteworthy. l3l farmer felt that labor was adequate to handle most crop operations and maintenance of livestock. Labor is hired only during transplanting of rice and harvesting of wheat. The farmer reported that they worked daily approximately 10 hours on the farm irrespective of winter or summer season. Out of these 10 hours, the three farmers worked 1—2 hours together on maintenance of livestock and preparation. of fodder. For the purpose of this study, it is assumed that the farmer works approximately 12.5 days in each 15 days period. This matter was discussed with the farmer and he was asked to comment on how many holidays does he and the other two farmers take in a year. The farmer estimated roughly 1-1/2 months holidays which are utilized for visit to relatives, Eid festivals, sickness, etc. No labor shortage in the vilage was reported, which is consistent with the land holding per family which was low on the average of 2-3 acres. Details of labor requirements for each crop activity were discussed with the farmer. In a series of visits, these parameters were estimated and cross checked with other studies and with the data obtained from-the 5 farmers interviewed for background information. Labor requirements are broken down on a bi-monthly basis and only the critical periods are incorporated. As there are 3 members of the family working on the farm, labor is only likely to be a constraint during peak season periods, especially during the transplanting seasons. However, customarily this operation is performed by landless workers, or nomads. Students on vacation from school also participate in transplanting. There is no distinct labor market in the village and there are not many opportunities in the nearby town to sell labor. Landowners generally are hesitant to work for other farmers for wages. This does not imply that a zero shadow wage prevails in the villages. There is a social opportunity cost of work. Exchange labor is quite common among farmers. The farmer did not consider 132 hired labor to be a problem in the village, because of the small land holding per family in the village. Hence this constraint is not incorporated in the model. 2. Water Water requirements were estimated by asking the farmer the number of hours required to irrigate a 1 acre field during a 15 day period for each crop. The farmer thought that he'could sell water for three acres at the rate of 1/3 share of the rice produce and 1/4 share of the wheat produce. But because his neighbors had recently installed a tubewell there was no demand. Again it is not the pumping capacity which is the prime constraint but it is the actual cost of water which bars the farmer from utilizing more water specially in the production of rice. As noted earlier the farmer does not have access to canal water. 3. Draft Power The farmer owns a pair of buffaloes with which he works 6 hours a day for approximately 24 days a month. The work rate for dry and wet land is the same as for the Gujranwala farmer. The farmer reported that he is able to hire a tractor if required but prefers to utilize. his own draft power. Tractor services are available at the rate of Rs. 35/acre for plouging and Rs.10/acre for planking. However, the farmer reported that he does not purchase tractor services. The present draft power on the farm was considered sufficient. It may be noted that traditionally bullock power is not sold in the area. Farmers supplement draft power needs especially during rice puddling on exchange basis. Based on this information, it is assumed that the farmer does not hire tractor services. The farmer maintains two milking buffaloes to meet family requirements. l33 Minimum and Maximum Constraints The following are the minimum and maximum constraints:- 1. Requires at least i) 64 maunds of wheat. ii) 93 maunds of paddy. iii) Rabi fodder 1.5 acres of berseem. vi) 2.1 acres of Kharif fodder. v) .5 acres of sugarcane 2. Maximum Rabi fodder 2.7 acres 3. Maximum Kharif fodder 3 acres Structure of the LP Model The matrix for the Daska farm is shown in table (5.7). The objective function comprises of 12 crop production activities. The different constraints are listed below. 1. Constraint on maximum available type of land during the Kharif and Rabi season. The farmer has two types of land excellent and good land. 2. Maximum water availability in each 15 day period. The most critical months are included in the model. 3. .Maximum Family labor available on the farm during each 15 day period. 4. Draft power availability in each 15 day period. 5. Maximum and minimum acreage requirement for each production activity. 6. A non negetive constraint for each activity. Table 5.7 LP Tableau Fanm The (Eeska) (1984) Activity* 134 q 5 6 7 wmeM‘PWNHI’" 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 ERabland EKhland CRabland Gkhland TJanllad TJan21ad TFebllad TFebZlan Tharllan TNerlan TAprllan TApr21an R“ayllan TbeZlan TJunllan T3un21an TJulllan TJu121an TAugllan TAugZIan TSepllan TSelean TCttllan TCttZlan TNovllan EVOVZlan UDecllan 03ecllan and Constraint: acre acre acre acre acre acre acre acre acre acre acre acre acre acre acre acre acre acre acre acre acre acre acre acre acre acre acre acre .05 ,_.,__,__,_._,_._,_.,_..__._ 2 3 1 l .05 .05 ,_...,_..._,..,_,_,_,_,_ HF_,_._,_,_.._._._..—. .05 ,__.,_...,_,__.,_.,_.._.,_.._,_. . H —._._.-»—_-»-»-H-0 .05 .—p—-—-.——.—..._.._,_p...._ .05 v—‘v—‘b—Hr—«r—D—r— .05 ,_.,_.,__,_.L_.._.._.._.._..._. P-D—r—b—P-D—b—ny—I— *.%ctivities are as follows: BaSTBtl on excellent land w/o FM“ Baatati on good land with FM“ Bastati good land w/o FM“ _ Basnati rice on excellent land with FM“ IRRI rice on excellent land with FM“ [RRI on excellent land w/o FYM IRRI on good land with FY81 IRRI on good land w/O FY81 Wheat on excellent Wheat on good land Kharif Fodder Rabi Fodder land 135 Table 5.7 (Continued) Row Resource Unit 1' 2 3 4 5 6 7 8 9 10 11 12 Sn RHS N0. -------------------- Objective Function (Rs.)** ----------------------- Labor Constraint 29 1Apr i 1 Nan/hr 2 2 24 23 29 LT 338 30 LJuneZ 18 18 18 18 18 18 18 18 32 LT 338 31 LJulyl 22 22 22 22 33 33 33 33 29 LT 338 32 LJulyZ Man/hr l5 l5 15 15 15 8 8 8 29 LT 338 33 LAugZ Man/hr 20 20 20 20 21 21 21 21 17 LT 338 34 I.Nov2 Man/hr 8 8 8 8 23 4 LT 338 Irrigation Constraint 35 [Julyl Hours 4 4 4 4 10 10 10 10 4 LT 324 36 lJuly2 Hours 4 4 4 4 10 10 10 10 4 LT 324 37 lAugl Hours 10 10 10 10 12 12 12 12 2 L T324 38 lAugZ Hours 10 10 10 10 10 10 10 10 LT 324 39 ISeptl Hours 9 9 9 9 10 10 10 10 LT 324 Bullock Power Constraint 40 BJuneZ Pair/Hr 18 18 18 18 22 LT 220 41 BNovl Pair/Hr 12 12 12 12 14 14 14 14 LT 220 42 BNOVZ Pair/Hr 10 10 LT 120 Maxirra and .‘vlinin‘e Constraints 43 Max RF Acres 1 LT 2.7 44 Max KF Acres 1 LT 3.0 45 Min Wheat Nbunds 32 16 GT 64 46 Min Bas Maunds 19.5 18 13.5 12 ‘ GT 93 47 Min KF Acres GT 2.1 48 Min RF Acres (71' 1.5 CHAPTER 6 RESULTS OF FARM MODELS AND INVESTIGATION OF ALTERNATIVE CROP TECHNOLOGIES This chapter is presented in three parts: In the first section the base plans for the Gujranwala and Sialkot farm are presented. The basic solutions are compared with the crop plan followed during Rabi 83 and Kharif 84. The second section explores the possibility of introducing sunflower and mungbean into the models. Different assumptions are made and tested in this section. In the third section possible affects of product price changes on the farm plan are explored and the effect of a technological change is simulated. The last section draws policy implications from the results of the analysis of representative farms and the survey results reported in chapter 4. SECTION I RESULTS OF THE BASE MODELS Base Run Of Farm One (Hafizabad). The base model for the Hafizabad farm in district Gujranwala brings six activities into the optimal solution and has an objective function value of Rs. 17750. Activities include basmati rice, Kharif fodder, wheat after rice, wheat planted on fallow land, berseem and sugarcane. The model solution is compared with the farm plan followed during 1983—84 (Table 6.1). The cropping intensity is higher (144 per cent) in the model solution. The model proposes only 1.3 acres of basmati rice whereas the farmer currently grows 4 acres. This difference is explained by the fact, that the model allocates more area to fallow land during Kharif which is utilized in 136 137 Table 6.1 Comparision of Base Model for Farm One (Hafizabad) With 1983-84 Crop Plan 1983-84 Plan Base Model Acres Value of Objective Acres Function Activity Rs. 17748.9 KHARIF Basrrati Rice 4 1.3 Kharif After Fodder 2.0 2.5 RABI Wheat Fallow 2 4.9 Wheat After Basmati » 3.25 1.3 Rabi Fodder 1.5 2.5 Sugarcane* . 5 l. 0 Cropping Intensity (%) 142 149 * Covers both Rabi and Kharif season. Table 6.2 MVP and Cost of Forcing in Non Optimal Activities in the Base Solution Constraint MVP Activity Cost KF fodder Max 1450 Lease Land 955.24 Sugarcane Max 1006 Basmti Min —33 Land Septl 816 Land Dec 1 309 Land Dec2 330 138 Rabi for growing wheat (wheat/fallow). The summer fodder (maize, sorghum) acreage is almost identical to that allocated under the crop plan for 1983—84. Wheat is an important crop for small farmers, especially if a farmer has to feed a large family. Being the main staple, the farmer would ensure that his subsistence is met before allocating wheat acreage to other crops. The model incorporates a minimum subsistence requirement for wheat, which was specified as a constraint. The basic solution recommends 6.2 acres of wheat, whereas last year the farmer allocated 5.25 acres to wheat. Less Rabi fodder (berseem) is grown by the farmer in comparison to what the model 'suggests. The model recommends, 2.5 acres the maximum the farmer reported he was willing to grow, while last year the farmer allocated 1.5 acres. Marginal Value Product of Resources. The marginal value product or shadow price of each scarce resource is shown in (Table 6.2). Farm labor ,bullock power and pumped water have no shadow price, because in each period the farmer has a surplus of these resources. Allocating one more acre to Kharif fodder will increase the value of the objective function by Rs. 1449, similarly if an additional acre for sugarcane were available it would increase the value of the objective function by Rs. 1006. However, an additional unit of basmati would reduce the value of the objective function by Rs. 33. The competition for land is the highest during the the first period of September and December. These are months when the farmer makes decision on implementing his Rabi crop plan. Availability of one more unit of land during the second period of December will increase the value of the objective function by Rs.330. If one unit of land were leased out it would reduce the value of the objective function by Rs. 955. The farmer can only earn Rs.500 by leasing out land. 139 Basic Solution of Farm Two (Daska). The model for the Daska farm in district Sialkot suggests 6 production activities. The optimal solution gives an objective function value of Rs. 16374. The base plan is compared with the present cropping pattern followed on the farm (Table 6.3). The model output is comparable to the current farm plan. According to the model the farmer should allocate 5.7 acres to the production of basmati rice and 1.3 acres to the production of IRRI. Similarly, the model suggests allocating 3.9 acres to wheat production on excellent land, and 5.9 acres on average land. The maximum amount permissible for Kharif and Rabi fodder appears in the optimal solution. Shadow Price of Unused Resources and Cost of Forcing Non Optimal Activities. The shadow price of the unused resources and cost of‘forcing non optimal activities is shown in Table (6.4). If additional Basmati is grown on excellent land it will reduce farm income by Rs. 278 and 279, depending on whether it_ was grown with or without manure. Similarly growing IRRI will reduce the objective function, if included in the farm plan. SECTION II ECONOMICS OF ALTERNATIVE CROP TECHNOLOGIES AND EFFECT ON THE FARM PLAN In this section we investigate the economic potential of sunflower and mungbean on the two representative farms. The effect of these new crop technologies on the cropping pattern and farm income is studied. 140 Table 6.3 Catparision of Results of Base Nbdel for Fanh Two (Daska) With 1983—84 Crop Plan x 1983-84 Plan Base Model Activity Acres Value of Objective Acres Function Rs. 16374 Basmati Rice 4 5.72* IRRI Rice 2 l.3** Kharif Fodder 2.1 3 *** Wheat 11 10.2 Rabi Fodder l 5 2.7 Cropping Intensity (%) 175 178 Notes: * Basrrati excellent land with FYM (2.62 acres) + Basmati on average land with HM (3.1 acres) = 5.72. ** IRRI grown on excellent land with FYM. *** Wheat on Excellent Land (3.9 acres) + on average land (6.3 acres) = 10.2. Table 6.4 MVP and Cost of Forcing in Non Optimal Activites in the Base Solution Constraint MVP Activity Cost Excellent Rabi Land 1190 Bas on excellent land w/o FYM 278.35 Excellent Kharif Land 655 Bas on average land w/O FYI“ 278.84 Kharif Fodder Max 610.7 IRRI on excellent land w/O FYM 183.58 Rabi Fodder Max 633.0 IRRI on average with FYM 124.42 Max FYM 12.42 IRRI on average land w/o FYM 308 Dec2 Land 517 Notes: Dec: Decenbe, 1: first fifteen days of a month, 2: second 15 days of a rmnth, Bas: Basnati rice. 141 FARM ONE (HAFIZABAD) Introduction of Sunflower (Alternative 1). Two types of sunflower have been identified as potential substitute crops for wheat and rice. Sunflower can be grown in spring (Feb-May) and as a Kharif crop (July-Oct). The inclusion of this new technology at an average yield of 16 maunds/acre (see appendix A Table 6.1 for details of gross margins) could significantly change the base plan. A 16 md/acre average yield is based on information obtained during the survey work in Daska. The prices received by farmers at the Market Committee in Daska are used. This support price is announced prior to the planting season and is assured by the procurement center at Daska. This information was obtained from the survey work described in chapter three. Sunflower appears in the optimal solution both as a Rabi and Kharif crop. Table (6.5) shows a comparison of the farmers existing cropping pattern, the pattern proposed in the basic solution and a new cropping pattern with sunflower. The model suggests allocating 3.26 acres to Rabi and 4.62 acres to Kharif sunflower, which would increase the value of the objective function by 29 per cent. Two types of substitution take place in this model. First, Rabi sunflower competes directly with wheat . Both wheat after fallow and wheat grown after basmati are reduced with the incorporation of sunflower. The base model brings in more wheat grown on Kharif fallow land than a higher acreage of basmati rice. Implying that on the Hafizabad farm a fallow wheat rotation is more profitable than a Basmati rice—wheat annual rotation. However, with the incorporation of sunflower, the area for basmati remains unchanged while area allocated to wheat has decreased. Only the minimum requirement specified (30 maunds which can be obtained from appoximately 1.3 acres) is being proposed by the model. Kharif and Rabi fodder is also reduced by .5 acres in both cases 142 Table 6.5 CoTparision of Basic Solution With Alternatives 1 & II for Fanm Che (Hafizabad) Base Nbdel Alternative 1 Alternative 11 Value of ijective Function --------------- (Rs.) ---------------------------- 17749 22952 29342 Activity Acres Acres Acres Bastati l 3 1.3 1.3 Wheat/Kharif Fallow 4 9 1.28 1.6 Wheat/After Basrati 1.3 .53 - Kharif Fodder 2.5 2.0 2.0 Rabi Fodder 2 5 2.0 2.0 Sugarcane 1 O .5 .5 Rabi Sunflower - 3.26 .6 Kharif Sunflower - 4.62 4.3 ang Bean - - 5.0 Cropping Intensity 149 165 184 143 and so is sugarcane, which is reduced to .5 acres compared to the base plan of 1984. The cropping intensity with the new alternative of sunflower will increase by 16 per cent. The shadow prices and cost of forcing in non optimal activities are summarized in Table (6.6). Introduction of Short Duration Mungbean With Sunflower (Alternative (I_I) As stated earlier in chapter 5 the Nuclear Institute for Agriculture and Biology has evolved new short duration mungbean varieties . The agronomy of these varieties is better understood than their economics. Here we explore the simultaneous impact of introducing mungbean and sunflower into the farm plan. Table (6.5) shows that 5 acres, the maximum acreage constrained in the model appears in the optimal solution. However, incorporating mungbean reduces the acreage allocated to spring sunflower as suggested in alternative I. This implies that there would be direct competition for land, if not for other resources if these two technologies are simultaneously introduced. The other crops proposed in this scheme are not significantly different from those resulting from alternative I discussed in the preceeding section. Similarly, if mungbean is introduced alone without sunflower, the maximum amount allowed in the model (5 acres) appears in the solution. A 65 per cent increase in income is observed over the base model and a 28 per cent increase over the above sunflower model (alternative I). The MVP and cost of forcing in non Optimal Activities for alternative 11 into the plan are summarized in Table (6.6). The the MVP of Rabi fodder (Rs.-498) and Kharif fodder (Rs.—942) has decreased over the alternative of only incorporating 144 Table 6.6 NNP and Cost of Forcing Non Optinal Activity for Alternative 1 and II on Fanh Che (Hafizabad) Jan 2 labor Augl bullock CttZ lan CBCl land Apr112 lan Nun basrati Min Rabi Fodder Nfin Kharif Fodder Nun wheat NWP Activity Cost of Forcing Non Cptinal Activity Alternative Alternative 1 ll 1 11 47.28 - Basmati Rice - 202.74 44.32 - Lease Land out 625.9 2904 816.08 816 309.0 885 - 1702 ~33.26 -58 -200.5 -498 -555.43 -942 — -77 - 532 th nungbean 145 sunflower. In other words allocating an additional unit to Rabi and Kharif fodder has a high opportunity cost in the presence of'a sunflower and mungbean crops. FARM TWO (DASKA) Introduction of Sunflower (Alter—natiie_l_). When both spring and autumn crops of sunflower are introduced into the representative farm in Tehsil Daska we observe a sharp increase in farm income from the crop production activities. The introduction of sunflower increases the objective function by 105 per cent.The new farm plan (Table 6.7) includes 7.1 acres of Rabi sunflower and 5.24 acres of Kharif sunflower. The area allocated to basmati rice increases from 2.62 acres of excellent land to 3.9 acres, while that under basmati grown on good/average land decreases from 3.1 acres in the basic solution to 1.26 acres with the incorporation of sunflower. Alternative 1 takes out wheat grown on average land and the farmer's needs are met by allocating all available excellent land in R'abi (3.9 acres) to wheat production. This model also reduces the area allocated to fodders. Mungbean and Sunflower (Alternative 11). The value of the objective function is higher by more than 300 per cent compared to the base model. A maximum of 8 acres of mungbean assumed in the model are brought into the solution. The maximum amount of mungbean acreage assumed is based on minimum acreage requirements for fodder. The major changes are in Rabi sunflower which is 4.5 acres, which is less than proposed in alternative I. Rabi fodder is also higher from alternative I (1.5 acres) but less than suggested in the 146 Table 6.7 Corparision of Basic Solution With Alternatives I <3: 11 for Fanm Two (Daska) Base Model Alternative I Alternative 11 Value of Objective Function ----------------- (Rs.) --------------------- 15256.74 31264 46434 Activ1ty Acres Acres Acres Basmati/excellent 2.6 3.9 3.9 land with FYM Basmati/good 3.1 1.26 1.26 land with FYM IRRI/excellent land 1.3 — - with FYM Wheat on excellent 3.9 3.9 3.9 land Wheat on average land 6.3 - 2.0 Kharif Fodder 3.0 2 1 2.1 Rabi Fodder 2.7 1.5 2.1 Spring Sunflower 7.5 4.9 Kharif Sunflower — 5.6 5.6 Ming Bean - - 8.0 Cropping Intensity 178 200 261 147 base model (2.7 acre). The MVP and cost of forcing non optimal activities information for alternatives 1 and II for the Daska farm is summarized in Table (6.8). SECTION III SENSITIVITY ANALYSIS WITH PRICES AND YIELDS A. The Effect of a Higher Price Support for Sunflower__and Mungbean at Different Yield Levels Given the Government's interest in increasing the production of oilseeds and pulses one may realistically assume that a higher support price may be used in the future to popularize these crops. The effect of an increased price support with three different yield scenarios for the Hafizabad Farm is analyzed. A 25 per cent increase in the price of sunflower and mungbean from September, 1984 prices prevailing in the Daska market is assumed. We observe a significant improvement in farm income even at the low yield scenario. Farm income will increase by 32 per cent by moving from a low yield high price to medium yield high price scenario (Table 6.9). The yield data are based on the survey of 11 sunflower growers in the Daska area. The yield data for mungbean is derived from on farm trials 'conducted by scientists from the Nuclear Institute for Agriculture and Biology in the rice areas. It may be noted that Kharif sunflower does not appear in the solution at low yield, high support price scenario. This is because spring sunflower and mungbean compete with each other. Allocating one unit of land to the production of sunflower will reduce the value of the objective function by Rs. 107.5. A marked increase in the MVP for October land is observed. The value of October and December land increases significantly, from a low yield to a high yield scenario (Table 6.10). 148 Table 6.8 MVP and Cost of Forcing Non Optimal Activity for Alternative I and II on Farm Two (Daska) Constraint MVP Activity Cost of Forcing Non Optimal Activity Alternative Alternative 1 II I 11 Excellent Bas excellent Kh land 1406.5 1406.5 w/o FYM 478.5 478.5 Excellent Bas average Rabi land — 1119.0 w/o FYM 479 479 Aprill lan 1702 517.0 IRRI excellent Apr112 Ian - 1185.0 with FYM 2240 2240 Mayl lan - 1069.0 IRRI excellent Aug2 lan 1702 — . land w/o FYM 2636 2636 Oct2 Ian - 633.0 IRRI on average Bas Nbximm -147 -147 land w/o FYI“ 1814 1814 Kharif Max -1091 -458 Wheat average ' Rabifodder on average land 1185 — Max -552 - Max Mingbean - 1155 Notes: The following abbreviations are used -- Bas=basmati rice, 1: first 15 days of a month, 2: second 15 days of a month, lan=land. 149 Table 6.9 Effect of a 25 Per Cent Increase in Sunflower and Nhngbean Prices at Three Different Yield Levels on Hafizabad Fanm ------- ADbjective Function in Rs.--——-—---- 27058 35655. 43380 Activity Acres Acres Acres 1. Basnati 1.3 1.3 1.3 2. Wheat Fallow. 1.6 1.6 1.28 3. Wheat after Bas - - .53' 4. Kb Fodder 2.5 2.0 ‘ 2.0 5. Rb Fodder 2.5 2.0 2.0 6.5ugarcane -.6 .5 .5 8. Kb Sunflower - .6 .39 9. Rabi Sunflower 4.6 4.3 3.7 10. Nhngbean 5.0 5.0 5.0 * A 25 per cent increase in over September, 1984 price of Rupees 150 for sunflower and 270 for mungbean. Notes: Low Wbdium High 10 nds/acre sunflower and 8 nds/acre nungbean. 16 nds/acre sunflower and 10 nds/acre nungbean. 20 nos/acre sunflower and 12 nos/acre nungbean. || 11 ll 150 Table 6.10 NW? and Cost of Forcing Non Cptinal Activity in the Solution for a Three Different Yield Levels and a 25 Per Cent Increase in Price of Sunflower and angbean for the Hafizabad Fanm. Cbnstraint ------ ~VNP (Rs) ----- Activity ---------- Cost(Rs ) -------- L M H L M H Kh Fodder 443 - - Wheat/Bas* 350.8 82.74 - Wheat Min -40 ~125 -182 Lease Land 1961. 4104 5491.91 Bas Nun -35 -84 -112 Kh Sunflower 107.5 - - RB Fod Min - ~1698 -3242 Sugarcane Min ~ -2142 -3530 AprZIan 1285 2302 3052 Cctllan 373 816 816 Decllan 805 1485 2124 Mungbean th 950 607 661 * Wheat after Bastati. Notes: L=low,-Nhnedium, and H=high. Sate definitions of yield as in Table 6.8. Lan=land, 1=first 15 days of the nonth and 2=second 15 days of the nonth. 151 B. The Effect of 10 or 20 Per Cent Increas§_1_n_Price of Basmati Rice Accompanied With 25 Per Cent Increase in the Price of Sunflower and @ngbean on Daska Farm Given the low returns to rice it is worthwhile to explain the effect of increase in basmati price by' 10 per cent and 20 per cent when the government supports sunflower and mungbean at a 25 per cent higher price.-This effect is analyzed on the Daska farm. With a 10 per cent price increase 6.2 acres of Basmati rice appear in the solution. More interestingly, IRRI rice drops out of the optimal solution. The solution proposes 4.9 acres of Kharif sunflower and 5.64 acres of Rabi sunflower (6.11). The maximum amount of mungbean (8 acres) allowed in the solution appears in the solution. The set of crop activities remains unchanged with a 20 percent increase in rice price, although, the value of the objective function is changed. This exercise shows that a higher support price for basmati rice in the presence of new crops such as sunflower and mungbean" will not lead to increased production of basmati. However, it will probably shift resources away from IRRI varieties. A three rupees increase in the price of Basmati and a reduction of 2 rupees in the price of IRRI announced in September, 1984 signals that this may well be the policy objective the government is interested to achieve. The MVP and cost of forcing in non optimal activities is summarized in Table (6.12) C. Sensitivity of a Potential New Basmati Varietv PK—258. Based on discussions with scientists at the Punjab Rice Research Institute and as reported by other senior officials in the Agricultural Ministry there is encouraging news about PK-258, a potential high yielding basmati variety. The researchers (personal communication) have claimed that this variety yields 60- 152 Table 6.11 Comparision of Basic Solution With Increase in Price of Bastati Rice and Introduction of Sunflower and Mungbean The (Daska) Base Nbdel Bastati Rice Price Increase ------- (per cent)--——---——— 10 20 Value of ijective Function --------------- (Rs.) ------- . ----- 15256.74 59609 60595 Activity Acres Acres Acres Bastati/excellent 2.6 3.9 3.9 land with Fth Bastati/good 3.1 - 1.26 land with FYNI IRRI/excellent land 1.3 - - with FW“ . Wheat on excellent 3.9 3.9 3.9 land ' Wheat on average land 6.3 2 0 2.0 Kharif Fodder ' 3.0 2 l 2.1 Rabi Fodder 2.7 , 2.1 2.1 Spring Sunflower 4.9 4.9 Kharif Sunflower - 5 6 5.6 ang Bean - 8 0 8.0 Cropping Intensity 178 252 262 153 Table 6.12 NW? and Cbst of Forcing Non Cptinal Activity With a Bastati Price Increase of 10 and 20 Per cent on Daska Fanm Constraint NWP Activity Cost of Forcing hbn CpthTal Activity -------------------- BaSTati Price Increase (Rs.)-—---———--—<—---—----—------- (per cent) 10 20 10 20 Kb Fodder Min -1085 -1058 Bas excellent Bas Min -183 -174 1542 w/o FYM 464 515 Bas average Rabi land 1190 1190 w/o FYM 545 546 Aprill lan 517 517 IRRI excellent Apr112 Ian 1785 1785 with l-YM 3126 3176 IRRI excellent Aug2 lan 1699 1699 land w/o FM“ 3322 3372 Cth lan 633 633 IRRI on average Bas Nbxinun -147 -147 land with FYNI 2414 2414 Kharif Nbx -1091 -458 IRRI on average Rabi Fodder land w/o FYNI 2610 2610 Nbx -552 - th Nhngbean 1124 1124 154 70 per cent higher than the existing basmati rice. Even though it is premature to speculate the success of the variety as it is still under experimental trials, it is interesting to simulate its affect on the cropping pattern and farm income. . This exercise is carried out on the Daska farm. The 1984 prices for basmati, sunflower and mungbean are assumed. As detailed input coefficients are not yet known, only an adjustment of Rs. 60 is made to capture the increase in fertilizer use. This is justified as the variety can stand higher doses of nitrogen than the traditionally grown basmati. A yield increase of 50 per cent from existing yields on different land types on Daska farm is assumed. A less than 60-70 per cent increase in yield is assumed to reflect experimental and management conditions. The results show that- all excellent land (3.9 acres) available on the Daska farm during Kharif season is allocated to basmati rice with farm yard manure (6.13). The optimal solution brings in 5.9 acres of wheat. The cropping intensity changes only minimally (less than .5 per cent) over a crop plan with traditional basmati rice, sunflower and mungbean (alternative 11 for Daska discussed earlier). The objective function increases by 17 per cent. Hence a new variety in the presence of sunflower and mungbean can lead to improvements in farm income. The scarcity value and cost of forcing non optimal activity information is summarized in Table (6.14). The relative cost of growing IRRI variety on any type of land on Daska farm will increase substantially. 156 Table 6.14 MVP and Cost of Forcing Non Optirral Activity With the Introduction of New Variety I-‘K-258 Rice Variety, Mingbean and Sunflower Constraint MVP Activity Cost of Forcing Non Optinal Activity Bas Excellent w/o FYM 175 Bas Average Rabi Ex land 1190 with FYI“ 1417 Bas Average w/o FYM 1888 Aprill lan 517 IRRI Excellent Apr112 Ian 1185 with FYM 336 IRRI Excellent Aug2 lan 371 land w/o I-YM 532 Oct2 lan 633 IRRI on Average - land with FYM 1814 Kh Fodder Min —458 IRRI on Average Kharif average land 698 land w/o FYM 2010 ‘ Nbx Mingbean 1155 157 SECTION IV POLICY IMPLICATIONS Policy Implications of Findings Related for Rice—Wheat Farming The results of the linear programming analysis and survey work reported in chapter 4 are now combined to draw preliminary conclusions for policy analysis. A key finding of this research, reported in chapter 4 is that the Punjab rice farmers have started responding to the changes in input cost, particularly the cost of water. Among 240 farmers surveyed in tehsils, Hafizabad, Sheikhupura and Daska, 43 per cent reported a reduction in the area allocated to rice during 1980-1984, while only 8 per cent reported an increase in rice acreage during this period. The LP analysis shows that wheat and forage crops are comparatively more profitable than rice. One clear indication of this is given in the base solution of the Hafizabad farm where more land is kept fallow during kharif. This land is used in the Rabi season to grow wheat. In other words, early wheat is more profitable than growing wheat after basmati. This finding has several implications. Firstly, it establishes the role of wheat in the rice based system. Secondly, we are now clear that rice price policies have to be formulated keeping in mind the rotational relationships in the Kalar rice tract. Thirdly, policies which discourage the production of IRRI varieties such as lowering its price need to be formulated with caution. Wheat planted after IRRI rice gives a higher yield than if planted after basmati rice. Assuming that all areas taken out from IRRI varieties due (due to discouraging IRRI price policies) is substituted with basmati rice, we can 158 hypothesize further reduction in yields of wheat, because most wheat will be planted late after harvesting basmati rice. The results reported in chapter 4 support the above argument. The CIMMYT data and results of this study show that there is a significant difference in cultural practices for both rice and wheat depending on rotation followed. Wheat grown after 'IRRI rice tends to give 10—15 per cent higher yei-ld than if it is grown after basmati. Hence, it is important that in the future we carefully weigh the effect of any policy instrument not only on a single crop, but on the total crop rotation. This requires that future research efforts be focused on developing some wheat varieties which are especially suitable for the rice areas. At present no wheat variety has been developed keeping in mind the specialized ,agro—biological and socioeconomic environment of the rice areas. One reason has been that the wheat research stations are located in Fasialabad, and concentrate on breeding varieties which are generally suitable for all irrigated and dryland areas. Policy Implications of Resear_c_1_i Findings Related to Produc’gon Constraints of Rice The farmers view of constraints to rice production is significantly different than those perceived by the researchers. Farmers consider socioeconomic factors to be more important than technical constraints perceived by researchers. Farmers mentioned that the high cost of water and fertilizer, discontinuation of aerial spray, low production price, bribery demands by electricity and irrigation departments, impure inputs are limiting rice production in the Punjab. The researchers have identified low plant density, use of non recommended varieties, and zinc deficiency as the major 159 technical factors limiting‘production. Farmers were unaware of the zinc deficiency problem and generally agreed that plant densities were low. But were not always clear about the relationship between plant density and yield. This divergence in view points is partly a lack of communication between the researcher and farmer. Many researchers showed and discussed their research work with the author. However, they were often unaware of farm problems or did not know how to relate their findings to the field situation. This calls for closer researcher farmer contact. Farmers stressed the need to improve marketing infrastructure and procurement arrangements. Most government procurement agencies adhere to regulations (i.e. quality, grading) during the purchase of paddy from the farmers. There is also the need to ensure that suppliers of production inputs provide farmers inputs in exact amounts and of the right quality. Farmers reported low use of credit. This highlights the fact that the formal credit institutions (banks, cooperative societies) are not functioning well. Numerous farmers complained-duuring the field visits that influential members of the cooperative societies were the only ones benefiting from the cooperative schemes. Usually these were the large and educated farmers. However, farmers reported that if more credit was made available, they would buy more fertilizer. This study reports low plant densities for rice in tehsils of Hafizabad, Daska and Sheikhupura. 'Based on this information it is important that the extension service convey the message that merely increasing fertilizer dose will not lead to higher yields, unless supplemented with higher densities. In order to improve plant densities alternatives such as direct seeding should be explored more carefully. The Punjab Rice Research Institute (personal 160 communication) had conducted some experiments in the early 70's, however the results were inconclusive. The author observed a few farmers outside the survey area who were direct seeding rice instead of manual transplanting. Perhaps it is time that direct seeding be experimented on farmers fields. Three factors should be considered in analyzing this technology (1) the saving in cost of labor as a result of direct seeding; (2) the trade off between yield obtained by direct seeding vs manual transplanting and (3) benefits of improved plant densities and fertilizer efficiency realized from direct seeding. In summary, the major constraints to the production of rice as reported by the farmer lie in the institutional domain. It was also noted that problems differed from tehsil to tehsil. A single solution is not only inappropriate but also not workable. The current extension recommendations formulated do not take into account the socioeconomic variables. Nor is the technology scrutinized for its economic profitability. This suggests that the design of technology involve all the participants, farmers being one of them. Research conducted on isolated experiment stations without relevance to the farm situation, and policies formulated without considering the major participant, cannot serve the farmer's needs. Rice Price Policies and Effect on Farm Income The government has been following a policy where the price of rice paddy is increased by 3—4 per cent every few years. Such a price increase is usually ineffective in shifting resources in favor or away from rice due to high inflation. For example, a three rupee increase in the price per maund of basmati, translates to only an average increase of 3 per cent in gross income 161 from rice assuming a yield of 20 maunds/acre. Such a price increase does not significantly effect the farmers income or decision strategy. On small farms the price perception (the way farmers perceive change in price) and price reaction (how and at what level they react) threshold levels are often different than on large farms which have greater access to credit and other inputs. Usually price policies formulated at the federal level underestimate the cost of inputs. The real cost of water, the major cost in the production of rice is perhaps the most underestimated input of them all. Hence, price formulation agencies (The Agricultural Price Commission) may need to take the necessary steps to strengthen their micro data base to aid in formulating price recommendations. Potential for New High Return Crops The programming exercise indicates that there is great potential for the incorporation of sunflower and mungbean in the rice areas. These crops have been grown in other parts of the country for sometime. Why has the government not realized their potential in the rice areas? A plausible answer is that only recently has the rice situation become so unfavorable, that earlier these crops were not worth exploring for these areas. If this is true, we have an opportunity to provide rice farmers alternative crops which can significantly improve farm incomes. Earlier research work has already reported the agronomic feasibility of these minor crops. This study shows the economic implications of incorporating these crops into the present cropping patterns. The cropping intensities can be expected to increase by almost 11 per cent with the inclusion of sunflower and mungbean while incomes can be expected to 162 increase significantly. Mungbean, being a legumenous crop can also help improve soil condition. The process of change has already begun in the rice based farming systems. The preliminary efforts of the Pakistan Ghee Corporation suggest that sunflower has great potential in the area. The micro level evidence from the three villages in the Daska Rice Optimization Project area illustrates that high rates of adoption of sunflower and mungbean can be expected if certain constraints are removed. The major constraints limiting the production of oilseeds in the rice area as observed during the field work and based upon discussion with scientists and administrators are listed below. 1. Lack ,of Sunflower and Mungbean Seed and Varieties. This is perhaps the major complaint of farmers. At present there are both insufficient availability of seed and few hybrid varieties. This was also identified to be a major problem in a Federal Government Review of the Rabi - 1984 planning meeting held in august, 1984 in Islamabad. 2. Marketing Problems. Farmers are facing severe problems in marketing sunflower. The government has not been able to channel the payments effectively. This delay in payment often discourages small growers. Another problem has been establishing appropriate market agencies to procure sunflower on behalf of the government. The main reason why sunflower is being rapidly adopted by Daska farmers is a guaranteed market and because a senior official of the Ghee Corporation resides in the area. 3. Threshing Eguipment. Farmers in the Daska area reported difficulty in hiring a sunflower thresher. On inquiry, the author found that there was only one manufacturer of sunflower threshers in Daska (one of the main tubewell and thresher manufacturing cities in Punjab). This manufacturer was charging 163 an extremely high price because there were no competitors. For small cultivators (1—5 acres) threshing was not reported to be a major problem, as the operation can be performed manually. But those with higher acreage under sunflower required threshing equipment. More general problems of increasing sunflower production have been described by Muhammed and Rehman (1981) and in the proceedings of the national sunflower workshop (PARC, 1984). These reports outline a long term operational stretegy to improve the national oilseeds situation. The government has also shown keen interest in increasing production of oilseeds, however, this has not been backed with any price support, or technology transfer campaigns. Unless, a comprehensive program to effectively distribute the technology is designed and implemented, a major increase in production will only occur slowly. An important question is what is the expected impact of introducing sunflower on a large scale in the Kalar tract. Sunflower can be grown as either . a Kharif or Rabi crop. The programming exercise suggests that on small farms (less than 12.5 acres), sunflower can replace rice and compete with wheat. Preliminary evidence suggests that growing sunflower before rice can enhance the yields of rice. If this relationship can be verified through experimentation, then growing sunflower instead of wheat (especially late wheat planted after basmati rice) should result in increased farm incomes. The entomology and pathology of sunflower pests has not been explored in the rice areas. During the field work farmers did not report any serious pest problems. However, if new varieties or different crop production techniques are introduced, this may become a problem in the future. Similarly, some experts are concerned that high temperatures during the months of May and 164 June can affect the Rabi sunflower recently, however none of the above problems were mentioned during the field work. At present we have no scientific data on which to base our inquiries. These stresses have to be scientifically analyzed. The success of sunflower in Daska can be attributed to the efforts of local. farmers and the Pakistan Ghee Corporation extension section. A broader success in the whole of rice farming area will require major effort on part of the Punjab goverment accompanied with federal government assistance. The Daska area can serve as an extension model for promoting sunflower and mungbean in other areas. Given the comparatively low returns for rice (both basmati and IRRI), we can expect that in the future some rice area being replaced by sunflower if the constraints mentioned earlier are removed. Assuming a 25 per cent reduction in the rice area of the Kalar tract, the gross export revenue lost from basmati rice will be approximately Rs. 454 million/annum. We have assumed only a 25 per cent reduction in basmati rice area because of export considerations. If this area is substituted with sunflower even at a low yield scenario of 15 maunds/acre, we can expect a gross revenue import substitution gain of Rs. 913 million/annum or almost a 100 per cent increase in foreign exchange position from this substitution (Table 6.15). Similarly, if sunflower is substituted for wheat we can expect considerable increase in gross income. The substitution of sunflower for wheat is a sensitive issue, because the government for the last 5 years has been laying special emphasis on achieving self sufficiency in wheat. The districts Gujranwala, Gujerat, Sialkot and Sheikhupura produce about 11 per cent of the total wheat produced in the Punjab. Hence, in order to analyze the substitution 165 Table 6.15 Vbcro Economic Inpact of the Substitution of Sunflower for Wheat and Rice in the Rice-Based Areas of Punjab Wheat Rice.Area .Area Substituted Gross Revenue by Gross Area* in the Shifting to Sun— Revenue in the Four dis- ---------------- Flower (NL Rs.) , Reduced Four tricts Rice Wheat ------- Yield —————— From Dist- 15 18 24 Rice ricts nd/ac nd/ac nd/ac (WLRs.) --------------- Thousand Hactere------- Gross Revenue-—---------—---———-- 2426.33 1625 - - Scenorios & Assumptions 1. 50% reduction in - 1214 2729.6 3275.5 4367.4 — wheat acreage 30% reduction in - 728 1637.8 1965.3 2620.1 — wheat acreage. ll. 25% reduction in 406 - 913.3 1095.9 1461.2 454.03 rice acreage * Gujranwala, Sialkot, Sheikhupura and Gujrat. Note: 1. Three different yield levels based on survey conducted during June—July, 1984. 2. Prices: Rs. 150 (Sunflower), Rs.85 (Bastati rice) and Rs. 64 (Wheat) per 40 kgs. 3. Yields for rice and wheat are taken as district averages for the last three years. 4. Sunflower yields are based on survey data. Gross Revenue Reduced From Wheat (M.Rs . ) 1201.6 720.25 166 effect a more comprehensive study which takes into account the domestic resource cost and changes in social benefits would be required. The profitability of mungbean, at 1984 prices, suggests that it has great potential to improve farm incomes. There is an urgent need to popularize its cultivation especially on those farms which will specialize on rice production. The main constraints are lack of sufficient seed and know how. The need for exerted extension efforts in this areais recognized. The government should also explore the potential demand for mungbean. An assured market will be necessary to increase production of this crop. The Effect of an Increased Support Price for Kharif Crops The linear programming analysis shows that,an increase in price of sunflower and mungbean will lead to substantial increase in farm income. However, a higher price does not change the basic solution obtained with 1984 prices. In other words, the 1984 price of sunflower and mungbean was quite favorable in comparison to rice and wheat prices. Some studies have argued for a higher support price (USAID, 1983) for oilseeds. Initially greater attention should be placed on other constraints like marketing, seed and lack of information. A few selective incentives such as increased price of sunflower, assurance of marketing facilities, subsidizing seed and sunflower thresher will definitely help popularize sunflower production. The change in the rice based system should be gradual. The government should carefully monitor the situation for assisting farmers with new technology and extension support to achieve higher production. Simply making partial provision of some of these facilities will not lead to increased oilseed production. The government has to mobilize the necessary resources and make it a national priority. The 16.7 commitment to increase oilseed production has to come from the highest authorities, perhaps the president. A fundamental problem of the Punjab Agricultural Research System (as well as the national research system) is the dichotomy of research and extension. This lack of communication has been documented in numerous reports. The issue now lies in the leadership for taking up the challenge to increase productivity in the rice and other areas. Chaudhry (1984) found that the extension departments in the Punjab are ill equipped to handle such a task, not in number of workers but also the quality of leadership at different levels. Most of the scientifically trained manpower at the federal and provincial levels is concentrated in the research institutions and other government departments. The extension department is looked down on by farmers and other officials, due to poor training, facilities and a traditional stigma that "this section of agriculture is incompetent both professionally and impotent in the power structure." This lack of credibility has lead to frustration among the extension staff. Leadership for transferring new technologies has to come from research. However, the arrangement must be mutually acceptable. The professional planning has to be syncronized if results are to be expected. New Unexplored Technologies Only recently has there been any testing of the manual weeder in the Daska project area. The farmer response appears to be very encouraging. The question one may ask is why is the testing of such innovative technologies delayed. We have not fully exploited the technological base in rice production. Traditionally, the emphasis has been on two areas HYV's and fertilizer. We must further explore cost reducing technologies, especially in 168 the area of farm machinery. There is no study which documents the effect of the new rice/wheat reaper and thresher on labor, cropping patterns and farm income. Despite the fact that approximately 4,000 reapers are currently in the field (personal communication). Future research should address this and similar questions to determine the implications for modifying multiple cropping systems. New Variety of Rice The programming exercise shows that the new variety PK-258 has potential to increase farm income from rice. Unfortunately, the government has never seriously considered the economic potential of new varieties. The barrier has always been that 'Pakistan has a comparative advantage and monopoly position in the production of traditional basmati.‘ Perhaps it is now time to seriously analyze the potential of alternative varieties, not only from the point of view of rice but also with respect to its effect on other crops; particularly wheat. The new variety KS-282 has not found wide acceptance among farmers, despite this variety's good resistance to some of the common rice pests its yields are only 8-10 per cent higher than lR-6. Moreover, as the government is following a policy of discouraging IRRI varieties, one can expect further reduction in area allocated to IRRI varieties. This whole argument signifies the fact that the government has not been able to provide broad guidelines to research institutions for developing varieties. Years of hard work go into the breeding of a variety. However, once it is developed, only then the government realizes that it is not suitable from an export point of view. The 169 banning of lRRi—9 a recently released short stature variety in December, 1984, illustrates the point. A Hypothesis of Change Given the manner in which farmers are responding to increase in input costs in the rice area, one may hypothesize a certain degree of specialization will take place in the future. Three possibilities are envisaged based on the analysis of this dissertation. 1. Rice farming wil be concentrated on those farms with low rice input costs, especially canal water, and on farms which can heavily supplement canal water with tubewell water. 2. As the area allocated to rice is reduced over time, higher per acre yields can be expected, due to increase in per acre availability of inputs (water, plant density, labor, credit, fertilizer). The real question is whether this increase in per acre yield will compensate for the decline in production from the reduced area. 3. If the government decides to promote oilseeds and mungbean on heavy soils in the rice areas, we can expect a major shift in area to take place in favor of rice and wheat. This increase in oilseeds production can lead to substantial saving in foreign exchange which is currently being spent on oilseed imports. CHAPTER 7 SUMMARY AND RECOMMENDATIONS OF THE STUDY Punjab rice farmers have been experiencing declining net returns for rice, their most important Kharif crop. With constant rice prices, increasing input costs and declining IRRI and stagnant basmati rice yields have lead to great frustration among farmers in the Kalar rice tract. The government plans to gradually remove input subsidies, which will further lower net returns for major crops. A low support price for rice, demand fluctuations in the international market for coarse rice, high overhead and marketing costs and a government monopoly on rice procurement have resulted in a great difference between the export and procurement price of basmati rice. Despite the fact that the social profit margin has been rising over time the average farmer has not shared these benefits. Another area of concern has been the soaring import bill for oilseeds. Rice exports account for almost 20 per cent of the country's gross earnings from trade, however, in l983 almost the same amount of foreign exchange was spent to import oilseeds. Realizing this, the government has recently shown keen interest in promoting domestic production of oilseeds and pulses. OBJECTIVES AND METHODS USED IN THE STUDY The broader objective of this study was to investigate the constraints responsible for low productivity of rice and the identification of potential oilseed and pulses crop technologies to increase farm income. The study focuses on crop substitution in a multiple cropping systems framework. 171 The specific objectives of this study were (1) To study Resource Productivity in the Rice Based Farming System in Gujranwala, Sialkot and Sheikhupura districts of the Punjab. A particular objective is to provide a farmer view of the Kharif situation (2) To identify technical and socioeconomic factors which constrain increases in production of rice and competing Kharif crops in the Kharif Season in the three districts (3) To analyze through linear programming models the economic implications of incorporating within existing cropping systems new crops such as sunflower and early maturing mungbean and (4) to draw implications from the study for research,extension and government policy. The method used included a survey of 240 farmers in three tehsils Hafizabad, Sheikhupura and Daska to obtain information on constraints to the production of rice and minor Kharif crops. The 240 respondents were randomly chosen from 14 villages in the three tehsils. The survey of rice farmers had the general objective of increasing understanding of rice production in the Punjab. It was limited to the crop production activities and had the following specific objectives: 1. To obtain information on input levels, yields,cropping patterns varieties and socio economic constraints in rice production. 2. To identify any existing multiple cropping systems unknown to researchers. 3. To gather information about factors limiting the allocation of land to minor kharif crops with special reference to sunflower and pulses. 4. To obtain specific input/output information on sample fields to help construct the representative farm models I72 Cross tabulations, percentages, basic statistics (mean, variance, coefficient of variation) were estimated from the survey data using a microcomputer. RESULTS The results of the study show that there are both technical and socio- economic constraints which limit rice production in the Punjab. The technical constraints are low plant densities, use of non recommended varieties, low fertilizer use, and zinc deficiency. The socioeconomic constraints were considered more important by farmers. These include discontinuation of aerial spraying, uncertain availability of electricity supply for water pumping, high cost of pumped water and fertilizer, bribery demands by the electricity and irrigation departments and low product price. Continous rice-wheat rotation were found to significantly reduce the productivity of the farming system. The potential effect on farm income of changes in crops was analyzed by employing linear program models for small rice farms to analyze cropping patterns and the affect of new crop technologies. These two farms were located in tehsils Hafizabad and Daska. The potential of oilseeds and mungbean crops was analysed. The results show that on small farms both oilseed and mungbean have great potential to improve farm income and increase cropping intensities. On the Hafizabad farm the introduction of sunflower increased the cropping intensity by 11 per cent and farm income by 29 per cent , compared to the base model. Similarly, the simultaneous introduction of sunflower and mungbean resulted in a 65 per cent higher income and a 23 per cent 173 increase in cropping intensity over the base model. A more than 100 per cent increase in,farm income was observed on the Daska farm when sunflower was introduced. Cropping intensity on this farm increased by 15 per cent. In another analysis a higher support price for sunflower and mungbean was found to further increase income. However, a minor 5-10 per cent increases in price of rice was found to have no affect on the cropping pattern. Analysis of new rice varieties such as PK-258 showed great potential for increasing farm income, however it is likely to reduce area allocated to IRRI varieties. , RECOMMENDATIONS A major purpose of this dissertation was to provide information valuable to federal and provincial agricultural research agencies including particularly research managers, administrators and scientists to help focus multiple cropping systems research on important. issues in Punjab rice areas. Nine recommendations follow. 1. Based on the discussion presented in chapter 6 there appears to be great potential for incorporating sunflower and mungbean into the existing rice based farming systems in heavy soil areas at existing yield and price levels. Because of decreasing returns to rice the government may seriously consider promoting high value crops such as sunflower and mungbean in the rice areas. This recommendation supports the governments broader objective of increasing oilseed production inPakistan. 2. Current research on oilseeds is concentrated in Barani areas. This work has been in progress for the last 5 years with little change in barani 174 areas. The findings from‘ this study suggest that work on testing new oilseed and pulses technologies to refine a package of technology suitable for rice areas might pay off very rapidly. Biological ( especially pathological and entomological) and socio economic factors related to oilseed production and marketing need to be given attention. The spread of sunflower in the Daska area demonstrates that farmers will rapidly respond if constraints such as market, seed and threshing equipment are removed. There is a need for a multipurpose thresher which can be used to harvest wheat, rice and sunflower. 3. As discussed in'chapter 4, the research facilities available at the Punjab Rice Research Institute need to be further strengthened. Additional manpower training at the Ph. D level is needed. The Punjab government should encourage, Ph. D level training'in rice technology. The federal government has repeatedly announced training opportunities for rice scientist abroad. Immediate attention needs to be given to these opportunities. Also students at the University of Agriculture, Fasialabad, need to be encouraged to work on thesis problems related to rice-wheat based farming systems. 4. It is important to take a holistic view of the rice-wheat farming system. In chapter 4 it was shown that these two crops interact strongly and influence each others yield if planted in sequence. Most wheat research in the Punjab is carried on in Fasialabad. It is time that serious considerations be given to upgrade the Punjab Rice Research Institute to include other crop sections such as wheat, oilseeds, pulse,fodder. This recommendation is justified on grounds, that in order to conduct meaningful mission oriented research, multiple systems or cropping systems 175 research is needed and scientists from the relevant disciplines have to work closely and be in contact with the farm situation in the problem area. The rice research institute has the necessary physical infrastructure (buildings, land ). This institute could not only work on broader problems of the Punjab rice but could also meet other research needs of the rice system. 5. As argued in chapter 6, the linkage between the Rice Export Corporation and Research Agencies working on rice is virtually non existent. Scientists work almost independently from the export needs of the Rice Export Corporation. In order to effectively develop technology which can serve the needs of the farmers and also take into account the export requirements the communication between the relevant participants (scientists, farmers, officials of rice corporation ) may be improved. 6. The need to improve marketing facilities cannot be over stressed. By reducing overhead and other costs the profit margin of the government can be shifted to the farmers. This increase in farmer income may be reinvested to increase production. Marketing research should examine how marketing costs can be reduced. 7. In chapters 4 and 6, it was suggested that agencies involved in price policy formulations need to strengthen the micro level data base on which decisions are made. Changes in input cost such as the cost of water, needs to be estimated with greater caution and taken into consideration in formulating su'pport prices for rice. This study reports a decline in net income from rice and in the area allocated to rice in the three major rice growing tehsils of the Punjab. Punjab government statistics show an 176 increase in rice area. Policy makers are cautioned to weigh carefully both the micro and macro evidence. 8. The lack of knowledge about farm problems is common among researchers who remain isolated from the field. Pakistan is determined to increase production of all major crops. If the research community is to participate effectively, scientists from all domains have to be familiar with farm problems. Each discipline has to relate to the farm situation. The ultimate test of research in developing countries is its contribution to the understanding and solving of farm problems'. Such a Farm Oriented Research and Development (FORD) approach requires a set of incentives. which encourage scientists and administrators to exert energies to improve farm productivity. In countries where most scientific manpower is concentrated in research institutes, participation in the technology diffusion process becomes a major responsibility. 9. Based on the field visits and meetings with extension personal it appears that the extension service in the rice area appears to be adequately staffed . Further investments in the World Bank funded training and visit approach are taking place. However, there is the lack of a selective incentive scheme for extension agencies, which can mobilize efforts to improve farm productivity. Often the extension workers are poorly trained and do not possess new knowledge. It is common observation that the farmer usually knows more than the extension agents. Attention should be given to improving the performance of the extension departments. A detailed study should be conducted to determine alternative institutional arrangements which can improve communication between research and extension. APPENDICES APPENDIX A . TABLES FOR CHAPTERS 1-7 Table 1.1 177 International Prices of Rice (1977—81) Short Grain No. 1, USA (California Long Grain White 5% Broken 2nd Grade FCB Bangkok Thailand (Husked) —————————————————————————————— LB $/Tonne—--—-----------------—----- 1977 1978 1979 '1980 1981 450 454 425 517 629 F0?) Mi l ls F03 Bangkok USA (Texas) Thialand (Nfilled) 478 276 454 369 447 334 503 433 565 484 256 365 334 428 474 Source: Agricultural Statistic of Pakistan, 1982. 178 . .mw®H .muaumaumum Housmazowum< CmImwxmm "we“:ow .smmE use muss mo soauuzvoua :H :owuuspmu mumzm m we: wbwcu onImn nwuwm xzz nomad no: ma UH \H .nq.m we hOuUMM m >3 wfimwufise .opuo bum CH ”a I I I I I meoawmmm .aH H.o « I I I I I I I I bozoawcsm .mH I I I I I I I I :mmnzom .NH N.H 0.0 e.o N.o H.o a H.o N.o N.o q.o Ezmmmom .0H 4 e a « a « « x a « uscwzzobo .mH tumumzz m.m H.o v.9 . ~.q e.m q.m w.o m.m q.oa m.ma @ vmommmmm .qa s.o H.o H.o H.a . . . - I - mmmssa aaam;x .ma m.H o.H ¢.N o.m o.m N.N N.N m.~ q.m m.N puma: .NH H.o N.o H.o * e I m.ow mq.oe Nq.Oq em.om \M :mflz .HH I I I I I I m.oq nq.oe mq.oq mq.oq was: .OH m.a o.oa o.NH o.nH N.qH q.qH m.mH Om.~H qq.mH MH.HH Embo .a w.< n.m m.o m.n H.w A.“ m.mH 0N.OH mm.m Hm.m couuoo .m w.mH m.ma o.m N.HH m.NH «.ma «.ma ma.oH ma.qa mm.HH wcmopmmsm .m «.mH o.mH m.mH w.mH N.@H o.