Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Problems of size-tenure structure in Bangladesh agriculture and prospects of a land reform programme… Alam, Muhammad Mustafa 1980

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-UBC_1980_A1 A53.pdf [ 11.75MB ]
Metadata
JSON: 831-1.0094957.json
JSON-LD: 831-1.0094957-ld.json
RDF/XML (Pretty): 831-1.0094957-rdf.xml
RDF/JSON: 831-1.0094957-rdf.json
Turtle: 831-1.0094957-turtle.txt
N-Triples: 831-1.0094957-rdf-ntriples.txt
Original Record: 831-1.0094957-source.json
Full Text
831-1.0094957-fulltext.txt
Citation
831-1.0094957.ris

Full Text

PROBLEMS OF SIZE-TENURE STRUCTURE IN BANGLADESH AGRICULTURE AND PROSPECTS OF A LAND REFORM PROGRAMME IN DEVELOPING THE RURAL ECONOMY OF THE COUNTRY by MUHAMMAD MUSTAFA ALAM M.A. (Economics), The Univers ity of B r i t i s h Columbia, 1976 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES (Economics, P o l i t i c a l Science, Sociology) We accept th i s thesis as conforming to the required standard THE UNIVERSITY.OF BRITISH COLUMBIA Apr i l 1980 (c) Muhammad Mustafa Alam, 1980 In presenting th i s thesis in pa r t i a l fu l f i lment of the requirements for an advanced degree at the Univers ity of B r i t i s h Columbia, I agree that the Library sha l l make i t f ree ly avai lable for reference and study. I further agree that permission for extensive copying of th i s thesis for scholar ly purposes may be granted by the Head of my Department or by his representatives. It i s understood that copying or publ icat ion of th i s thesis for f inanc ia l gain shal l not be allowed without my written permission. Economics, P o l i t i c a l Science, Department nf Sociology  The Univers ity of B r i t i s h Columbia 2075 Wesbrook Place Vancouver, Canada V6T 1W5 Date May, 1980 i i ABSTRACT Exist ing s i ze -d i s t r i bu t i on of farms and land tenure arrangements in Bangladesh contribute to the perpetuation of inequal i ty and ine f f i c iency among agr icu l tura l households. A h i s t o r i c a l review of the land tenure system in Bangladesh shows that some character i s t i c s e s sent ia l l y r e f l e c t older trends while others are the outcome of more contemporary developments. Feudal i n te res t s , in the s t r i c t sense of the term, no longer ex i s t ; however, rent-receiv ing interests and tenant cu l t i v a t i on have survived, mainly.under sharecropping arrangements. With a rapid ly growing population, land shortage has become severe for most households. This has enhanced the economic socia l and p o l i t i c a l status of the r e l a t i v e l y large land owners who continue to hold s i gn i f i cant proportions of to ta l farmland. The patron-cl ient re lat ionship between them and the vast majority of landless and near-landless masses has helped maintain a semi-feudal agrarian structure. In the absence of an urban sector which can e f fec t i ve l y employ the 'superfluous ' labour from agr icu l ture, income-earning opportunities for a poor rural household are mainly related to i t s land-owning status. Dispar ity in land ownership being very high in Bangladesh agr i cu l ture , the d i s t r i bu t i on of income has understandably been ser iously skewed. i i i In considering the re lat ionsh ip between size-tenure structure and productive e f f i c i ency , t r ad i t i ona l theories are appl icable only to the micro - s tat ic context. M ic ro - s tat i c comparisons of agr i cu l tura l performance of d i f fe rent s ize and tenure groups of farms can, at best, help in formulating short run po l icy prescr ipt ions. For purposes of long-term development, however, po l i c ie s also need to be evaluated in terms of t he i r potential f o r meeting future needs of the country. The present study contends that neither of the two major cu l t i v a t i on arrangements which ex i s t in Bangladesh agr icu l ture (namely, small-scale owner cu l t i v a t i on and sharecropping cu l t i va t i on ) can meet adequate macro-dynamic standards in terms of improved equity and increased e f f i c i ency . In discussing the context of land reform for Bangladesh, i t i s noted that an ' e f f e c t i ve reform 1 for the country has to serve a number of goals, which include (a) attainment of equitable d i s t r i bu t i on of income, (b) ra i s ing the level of product iv i ty , (c) generating greater employment opportunities and (d) increasing the marketable surplus. A model of land reform which aims to.estab l i sh a group farming system based on.joint r e spon s i b i l i t i e s in ownership, organization and management, seems to o f fe r better prospects for removing the basic obstacles of s i z e - d i s a b i l i t y and tenur ia l d is incentives in Bangladesh agr icu l ture. It i s recognized, however, both during the process of i n i t i a t i o n and during the management phase of a group farming system, numerous pract ica l problems would be encountered. The present study iv suggests various measures for dealing with.these problems and, in doing so, seeks to draw upon the experiences of other countries. Whether an e f fec t i ve programme of land reform can be formulated and w i l l be implemented es sent ia l l y depends upon the nature of the p o l i t i c a l power-base and p r i o r i t y objectives of the ru l ing e l i t e . In Bangladesh, although po l i c i e s of government do not appear to have been dictated pr imar i ly by landed in te res t s , the power-elite has t r a d i t i o n a l l y been seen to cooperate with these interest s . The proportion of ag r i cu l tu ra l households without any land or other means of supporting themselves has, in the meanwhile, been increasing rapid ly. The impoverishment of a great majority of rural households, coupled with t he i r heightening demands for amelioration of t he i r s i t ua t i on , seems l i k e l y to create p o l i t i c a l circumstances whereby the allegiance of th i s section of peasantry could become essential f o r any e l i t e seeking to atta in or maintain p o l i t i c a l power in Bangladesh. Therefore, the prospects for carrying out an e f fec t i ve land reform in the country seem better for the future than they have been during the past decades. V TABLE OF CONTENTS Page ABSTRACT i i LIST OF TABLES v i i i LIST OF FIGURES x i LIST OF MAPS/CHARTS . xi i ACKNOWLEDGEMENTS . . . x i i i I. INTRODUCTION 1 Purpose of the Study 4 S ignif icance of the Study 5 Survey of L i terature 6 Studies on Farm Size, Land Tenure and Agr icu l tura l Product iv i ty 7 Studies on Land Reform 11 II. CONCEPTUAL FRAMEWORK AND PLAN OF. STUDY 20 Conceptual Framework 20 The Relevance of Size-Tenure Structure. . . . . . 20 Objectives of Land Reform 29 Dist r ibut ing Income More Equitably . . . . . . 30 Alternat ives to Land Reform?. . 30 Land Redistr ibut ion in Attaining Equity . . 34 Raising the Level of Agr icu l tura l Production 39 Adoption of HYVs 40 Generating More Employment Opportunities . . . 42 Increasing the Marketable Surplus 44 Def in i t ion of Land Reform 46 Models of Land Reform . . . 49 Red i s t r i bu t i v i s t Reform 49 vi Reform Aimed at Cooperative Farming 51 Reform Aimed at State Farming. . . . . . . . . 52 Reform Aimed at Group ( Co l l e c t i v i s t ) Farming 54 P o l i t i c a l Factors Af fect ing I n i t i a t i on and Implementation of Reform 55 Plan of Study 58 I I I. ORGANIZATION OF LAND IN BANGLADESH AGRICULTURE: PAST AND PRESENT 60 H i s to r i ca l Review 60 Elaboration of Size-Tenure Structure in Bangladesh Agr icu l ture 92 Regional Variations in*Bangladesh Agr icu l ture. . . . 112 Topographic and Cl imat ic Variat ions 113 Variations in Cropping Pattern. . 119 Variat ions in Size-Tenure Structure 125 Agr icu l tura l Systems 129 IV. PROBLEMS OF SIZE-TENURE STRUCTURE IN BANGLADESH AGRICULTURE 136 Problem of Unequal D i s t r ibut ion of Income. . . . . . 137 Land Tenure and Agr icu l tura l Product iv i ty. . . . . . 140 Micro-Stat ic Analysis . . . . . . . 140 Theoretical Premise 140 Conventional Theories 140 An Evaluation of Conventional Theories. . .. 152 Empirical Findings 159 Regression Analysis 160 Bogra and Rajshahi: An I n t e rd i s t r i c t Study 164 Physical-CI imatic Conditions 165 Land Tenure 169 Levels of Agr icu l tura l Product iv i ty and Possible Explanations 177 v i i Macro-Dynamic Analys is. 187 Farm Size and Agr icu l tura l Product iv i ty 194 Micro-Stat ic Analysis . 194 Theoretical Premise 194 Empirical Findings 197 Macro-Dynamic Analysis 202 Theoretical Premise 202 Empirical Findings 204 Conclusion 209 V. LAND REFORM FOR BANGLADESH 212 Why Land Reform in Bangladesh? . . . 212 Problems and Prospects of Various Models 215 Red i s t r i bu t i v i s t Reform 215 Reform Aimed at Cooperative Farming 224 Experience of Bangladesh in Cooperative Farming 224 Reform Aimed at State Farming 234 Reform Aimed at Group Farming 238 Problems and Prospects of I n i t i a t i on 238 Meeting the Objectives of Land Reform. . . . . 242. Management of Group Farms. . . . . 250 Popularizing Group Farms . 251. Conclusion • 255 VI. CONCLUSIONS 259 APPENDIX 267 BIBLIOGRAPHY 271. LIST OF TABLES Table Mi 1. Gross Rental Paid by Raiyats During Different Periods . 71 2. Raiyati and Under-raiyati Holdings in the Area that i s Now Bangladesh, 1938-40 76 3. D i s t r ibut ion of Farms Under.Different Size-Categories and the Area Covered by Them in Bangladesh, 1960 and 1968 93 4. D i s t r ibut ion of Farms Under Different Size-Categories and the Area Covered by Them in Selected Regions of Bangladesh, 1974. . 95 5. D i s t r ibut ion of Land Ownership Under Different Size-Categories and the Area Covered by Them in Bangladesh, 1977 97 6. Landlessness in Rural Bangladesh, 1977 99 7. D i s t r ibut ion of Farms Among the Main Tenure-Categories and the Area Covered by Them in Bangladesh, 1960, 1968 and 1977. . . 101 8. Owner-Operated Area and Tenant-Operated Area in Rural Bangladesh, 1960 and 1977 . . . . 105 9. Percentage of Tenant Households Paying Varying Amounts of Rent in Kind and.Areas Under Them in Bangladesh, 1977 106 10. Percentage of Tenant Households Making Additional Cash Payments Over and Above Rent-in-Kind in Bangladesh, 1977 107 11. Source of Agr icu l tura l Inputs Applied on Rented Land in Bangladesh, 1977 109 ix 12. Percentage of Tenancy Contracts Under Varying Durations of Tenancy and the Percentage of Total Area Covered by Them in Bangladesh, 1977 110 13. Yields of Different Rice Crops in Bangladesh in Different Years 122 14. Percentage D i s t r ibut ion of Agr icu l tura l Households by Their Land-owning Status in a Number of D i s t r i c t s in Bangladesh, 1977. . 127 15. Percentage of Farm Area Sharecropped in Different D i s t r i c t s of Bangladesh, 1960 . . . . 130 16. Temperature and Humidity in Bogra and Rajshahi (averages for the period 1966-70). . .168 17. General Information on Land Tenure in Bogra and Rajshahi, 1959-60 170 18. Percentage of Total Farm Area Operated Under Different Arrangements by Different Tenure-Categories in Bogra and Rajshahi, 1959-60 171 19. Percentage of Total Cult ivated Area Under Different Tenure-Categories in Bogra and Rajshahi, 1959-60 173 20. Farm Area and Cult ivated Area Under Different Categories of Tenure Arrangement in Bogra and Rajshahi, 1959-60. 175 21. Farm Implements and Work-Animals in Bogra and Rajshahi, 1959-60 181 22. Indebtedness and Sources of Loan in Bogra and Rajshahi, 1959-60 . 183 23. Farm S izes, Input-Use and Land Product iv i t ies of Owner-Cultivators and Tenants in Phulpur Thana, 1973-74 190 24. Average Cult ivated Area and Land Product iv i ty of Different Tenure-Categories in Dinajpur, Mymensingh and Rangpur, 1973-74 192 X 25. Value of Production Per Cult ivated Acre in Owner-Farms of Different Sizes in a Number of D i s t r i c t s of Bangladesh (various years). . . 198 26. Value of Production Per Cult ivated Acre in Owner Cult ivated Farms of Different Sizes in the Thanas of Phulpur and Thakurgaon, 1969-70 and 1973-74 200 27. Total Factor Product iv i ty of Farms of Different Sizes in the Thana of Phulpur, 1969-70 202 28. Acreage and Y ie ld of Tradit ional and High Yielding Var iet ies of Rice in Bangladesh, 1972-73 . . . 205 29. Percentage of Output Supplied to the Market by Different Sizes of Farms in Bangladesh, 1960. 207 30. Possible Cei l ings on Land and Consequences of D i s t r ibut ing the 'Surplus ' Land Among the Landless and Small Land Owners (owning 1 acre of less) in Bangladesh, 1977 217 31. Size of Farm and Per Capita Indebtedness in the Comilla Cooperatives, 1969 230 32. Percentage of the Members of Kotwali Thana KSSs Getting Loans (various years). . . . 230 A l . Selected Character i st ics of the D i s t r i c t (Appendix) of Bangladesh 268 xi LIST OF FIGURES Figure Page 1. Land Redistr ibut ion Among Small Farms and Its Effects on the Landless 36 2. Labour-intensive and Capital - Intens ive Technologies. 41 3. Tenant Cu l t i v a t i on , Cost-Sharing and Productive E f f i c iency 146 4. Marginal Value Product in Tenant Cu l t ivat ion and the Landlord's Par t i c ipat ion in an Ideal Lease 157 5. Micro-Stat ic E f f i c iency of Small Farms. . . . 195 x i i LIST OF MAPS/CHARTS Map Page 1. Bangladesh: Administrative Boundaries 114 2. So i l Composition of Bangladesh. . 115 3. Dry and Wet Regions of Bangladesh 117 4. Rice Crop Regions of Bangladesh 120 5. Regions of Some Important Crops Other than Rice in Bangladesh 124 6. Incidence of Sharecropping Cu l t i vat ion in the D i s t r i c t s of Bangladesh. . . 131 7. Soi l Composition of Rajshahi. 166 8. Soi l Composition of Bogra . 166 Chart 1. Degrees of Sharecropping Cu l t ivat ion in the Different D i s t r i c t s of Bangladesh, 1960 . . 132 2. Mult ip le Scalogram Analysis fo r the Regionalization of Bangladesh Agr iculture . . . 134 x i i i ACKNOWLEDGEMENTS I feel deeply indebted to my Research Supervisor, Dr. Geoffrey B. Hainsworth of the Department of Economics, for the valuable guidance he has provided me throughout the research. The unfa i l ing encouragement and the constructive c r i t i c i sms I received from him were enormously helpful in bringing the present work to i t s conclusion. I am thankful to Dr. Barrie M. Morrison for his keen interest in my work and his many advices on a number of issues of the thes i s . I fee l further obligated to him for l e t t i n g me borrow some important research material on Bangladesh. The numerous discussions I had with Dr. John R. Wood, Dr. Richard R. Bar ichel lo and Dr. Robert S. Anderson proved highly benef ic ia l to me. I would l i k e to take th i s opportunity to thank them a l l . I am pleased to express my gratitude to the Canadian Commonwealth Scholarship and Fellowship Committee for financing my programme. I must thank the Univers ity of B r i t i s h Columbia as well for of fer ing me a Graduate Fellowship. Thanks are also due to Hyacinth Wettasinghe for typing the contents of th i s thes i s . As always, my parents were eager to help me in every possible way. They painstakingly co l lected whatever research material I needed from xiv Bangladesh and sent them to me without loss of time. I fee l extremely grateful to them for the i r ever helpful att i tude toward me. This work would not have been possible without the constant support I received from my wi fe, Shahana. She has shared every stress and s t ra in of my academic pursuits which resulted in th i s thes i s . By singlehandedly managing the re spons ib i l i t y of looking a f te r our newborn son, Rishad, she allowed me ample time to devote to my research. I thank her most s incerely for a l l t h i s . 1 I. INTRODUCTION Bangladesh is pr imar i ly an ag r i cu l tu ra l country. More than 90% of i t s population l i v e in rural areas, and 77% of i t s labour force is engaged in agr icu l ture. The contr ibution of agr icu l ture to the gross domestic product (GDP) in Bangladesh varies from 55% to 58% from year to year (depending upon the to ta l ag r i cu l tu ra l output r e l a t i ve to the output of the non-agricultural sector) . Khan has pointed out that more than two-thirds of the GDP would be accounted for by the agr icu l tura l sector i f one considers such anc i l l a r y a c t i v i t i e s as transporting and marketing agr i cu l tu ra l products. This concentration on agr icu l ture can be part ly explained by the l imited resource base of /the country. Bangladesh has to depend ent i re l y on imports for such mineral resources as coa l , o i l and iron ore. There i s only a meagre supply of limestone from around the d i s t r i c t of Sylhet, so that much of the cement needed for construction jobs has to be imported as we l l . The l imited resource See Government of Bangladesh, The F i r s t Five Year Plan 1973-78, Dacca: Planning Commission, 1973, p. 83. A.R. Khan, The Economy of Bangladesh, London, The Macmillan Press, 1972, p. 38. 2 base that has resulted in such dependence on agr icu l ture prompted the fol lowing conclusion to a recent analysis of the country 's economic prospects: " . . . ce r ta in l y the short-term, and probably also the longer-term, future of Bangladesh seems almost inev i tab ly to be that of a pr imar i ly agr i cu l tura l country."3 Despite i t s concentration on ag r i cu l tu re , however, product iv i ty in th i s sector has remained very low. The main agr i cu l tu ra l produce of the country i s paddy, about 80% of the to ta l cropped acreage 4 being devoted to th i s crop. In terms of paddy y i e l d per unit of land, Bangladesh compares poorly with most other paddy-producing countries of Asia. Y ie ld of j u t e , which i s the second most important crop in the country (covering about 8% of a l l cropped acreage), has also been very low insp i te of favourable land and water condit ions.^ A. Robinson, Economic Prospects of Bangladesh, London: Overseas Development I n s t i tu te , 1973, p. 15. See N. Ahmed, A New Economic Geography of Bangladesh, New Delhi: Vikas Publishing House, 1976, p. 63. Bangladesh produced 1890 kg of paddy per hectare during 1978. The corresponding f igures for Burma, China, India, Japan, Malaysia and Thailand were 2019, 3534, 1975, 6250, 2524 and 2051 kg per hectare respect ively. See Food and Agr iculture Organization,. FAO Production Year Book, 1978, Rome, 1979, pp. 98-99. Per hectare jute y i e l d in Bangladesh was 1478 kg during 1978, whereas China produced 3039 kg per hectare during the same period (see ibid., p. 193). This shows that there i s a l o t of room fo r improving jute cu l t i va t i on in Bangladesh. 3 The low y i e l d , coupled with a high density of population (1,523 persons per square m i le ) , resulted in a meagre per capita income for the c i t i zens of Bangladesh (Tk. 1,208 during 1976-77; equivalent to approximately U.S. $ 80).^ It might seem paradoxical that product iv i ty per unit of land should be so low in Bangladesh where the s o i l i s of a high natural f e r t i l i t y typ ica l of a de l ta i c 8 region. Apart from the low product iv i ty per unit of land and the meagre per capita income, a severe skew in the d i s t r i bu t i on of income in Bangladesh agr iculture constitutes another major problem. This i s evident in a survey conducted in 1977 which found that about 33% of the rural households owned no cu l t i vab le land (some had just a piece of homestead land), whereas nearly 20% of a l l avai lable land was owned by less than 2% of the households, each 9 owning more than 10 acres. These data are reported by Government of Bangladesh, 'Population Change and Economic Development', Bangladesh, Vol. I l l , No. 5, September 1979, pp. 4-5. The high level of s o i l - f e r t i l i t y i s evidenced, fo r example, by the harvesting of as many as three crops within the year from the same piece of land with the appl icat ion of l i t t l e or no chemical f e r t i l i z e r . However, continuous cropping under such circumstances may ul t imately resu l t in decl in ing f e r t i l i t y . Government of Bangladesh, Summary Report of the 1977 Land Occupancy Survey of Rural Bangladesh, Dacca: Bangladesh Bureau of S t a t i s t i c s , 1977, Tables I and I I I. 4 Purpose of the Study Several a l ternat ive (and/or cumulative) explanations fo r the poor performance of agr iculture in Bangladesh might be suggested; for example, technological ba r r ie r s , population pressure, constraints set by nature, and i n s t i t u t i ona l bottlenecks. Exist ing l i t e r a tu re on the problems of ag r i cu l tu ra l development in underdeveloped countries indicates that s i ze -d i s t r i bu t i on of farms and tenure arrangements ( ' s ize-tenure s t r uc tu re ' , in short) in agr icu l ture are important variables in determining both the level of product iv i ty and the nature of income d i s t r i b u t i o n . ^ One of the main purposes of the present study is to invest igate how fa r and in what ways the prevai l ing size-tenure structure in Bangladesh agr iculture contributes to the problem of low product iv i ty. It also seeks to understand the re lat ionsh ip between size-tenure structure and the d i s t r i bu t i on of income among rural households in Bangladesh. On the basis of the f indings from th i s invest igat ion, the study then asks whether par t i cu la r a l terat ions in the size-tenure structure, under a land reform programme, can be expected to achieve greater equity and product iv i ty among other possible objectives of reform. See, for example, P. Dorner, 'Land Tenure, Income Dis t r ibut ion and Product iv i ty In teract ions ' , Land Economics,, Vol. XL, No. 3, August 1964; and A.K. Sen, ' S i ze of Holdings and P roduc t i v i t y ' , Economic Weekly, Annual Number, February 1964. 5 The central theme of the study can be formulated in terms of the fol lowing hypothesis: The ex i s t ing size-tenure structure in Bangladesh agr icu l ture imposes c r i t i c a l constraints upon the attainment of greater product iv i ty and a more equitable d i s t r i bu t i on of income, and an a l ternat ive set of arrangements ex i s t s which could a l l e v i a te these obstacles to rural development and improvement in l i v i n g standards fo r the majority of the people. Though the main thrust of the study w i l l be an analysis of economic behaviour, the thesis also proposes to look into some of the p o l i t i c a l and social dimensions of the problem of rural poverty and the prospects for land reform. S ignif icance of the Study The s ign i f icance of th i s study derives from the fol lowing two considerations: ( i ) The study examines a fundamental problem faced by the most important sector of the country and,, hence, has pa r t i cu la r s ign i f icance for future development and economic planning in Bangladesh, ( i i ) The absence of an up-to-date or comprehensive study of the problems of size-tenure structure and the prospects of a land reform in Bangladesh makes the present study add i t iona l l y useful in th i s context. 6 Any attempt to analyse the problems of agr iculture must s tar t with an invest igat ion of the re lat ionship between the avai lable arable land (the 'primary f ac to r 1 of production) and the ind iv iduals or groups who seek to sustain themselves and create a l i v i n g with what i s produced on that land. The nature of th i s basic re lat ionsh ip w i l l determine whether the agr icu l tura l sector i s able to play the optimum role i t should. Such important questions as the v i a b i l i t y of on-going technological change and the a v a i l a b i l i t y of a marketable surplus w i l l depend to a great extent on the relat ionships which are found to ex i s t between farmers and land as ref lected in the size-tenure structure. Evaluation of these relat ionships i s of pa r t i cu la r s ign i f icance to development planners in Bangladesh which, as we have noted, i s pr imar i ly an agr icu l tura l country. To substantiate the second point about the r e l a t i ve dearth of comprehensive studies on size-tenure structure and land reform in Bangladesh, i t i s necessary to provide a b r ie f survey of the ex i s t ing l i t e r a tu re on the subject. Survey of L i terature This section w i l l f i r s t survey the l i t e r a tu re on size-tenure problems in Bangladesh agr iculture and then give an account of the studies devoted to land reform and rural development in the country. 7 Studies on farm s i z e , land tenure and agr icu l tura l product iv i ty : The s i gn i f i cant contr ibution by A.K. Sen on th i s t o p i c , proposing that an inverse re lat ionsh ip ex i s t s between farm s ize and land product iv i ty in India, has been duplicated for Bangladesh 1 1 1 2 in a study conducted by Hossain. ' He estimates f igures for land product iv i ty (defined as output per unit of net cu l t i vated land) fo r d i f fe rent s izes of farms in a par t i cu la r l o c a l i t y of Bangladesh. He f inds that product iv i ty of land remains r e l a t i v e l y uniform up to 5 acres, but then decreases sharply as farm s ize increases. The major explanation provided for th i s i s that smaller farms, with an abundant supply of family labour, are able and w i l l i n g to u t i l i z e r e l a t i v e l y larger amounts of labour per unit of land. This study by Hossain i s e s sent ia l l y micro- s tat ic in nature. Although the basic approach i s one of fact f i nd ing , he does make a pol icy statement in the concluding paragraph of his paper: "The r e su l t s , thus, indicate that any (emphasis mine) agr icu l tura l po l icy that increases the percentage of land held by the small farms would increase product iv i ty and agr icu l tura l growth."13 A.K. Sen, 'An Aspect of Indian Ag r i c u l t u r e ' , Economic Weekly, Annual Number, February 1962. His basic idea has been elaborated in Chapter IV of the present study, pp. 194-197. 12 M. Hossain, 'Farm Size and Product iv i ty in Bangladesh Agr icu l ture: A Case Study of Phulpur Farms 1 , Bangladesh Economic Review, Vo l . I I , No. 1, January 1974. 1 3 Ibid.3 p. 493. 8 The above statement f a i l s to take into account several long-term implications of development under a l ternat ive agr icu l tura l systems. Advocacy of an extension of small-scale family farming needs to be analysed in greater deta i l before such pol icy measures can reasonably.be suggested. Given that the two prominent modes of production in Bangladesh agr icu l ture have long been owner cu l t i v a t i on and cu l t i v a t i on based on sharecropping tenancy, invest igations of the re lat ionships between land tenure and agr icu l tura l product iv i ty have been largely confined to considering the r e l a t i ve merits and demerits of these two d i f fe rent arrangements. Jabbar, for example, has undertaken a study of the r e l a t i ve productive e f f i c i ency of d i f fe rent tenure classes in selected areas 14 of Bangladesh. He f inds that part-operators (defined as those who cu l t i va te some of t he i r land and rent out the re s t ) , on average, produce more per unit of land than owner-operators (those who cu l t i va te only owned land) or part-tenants (those who cu l t i va te owned land as well as some rented land). Tradit ional Marshall ian arguments regarding lack of incentives among tenants to increase output or to invest in land improvements are offered by Jabbar as explanations for the poorest performance of the part-tenants. As for the better y ie lds obtained by part-operators than owner-operators, i t i s argued that the explanation l i e s M.A. Jabbar, An Investigation into the Effect of Farm Structure on Land Productivity in Selected Areas of Bangladesh. Unpublished Ph.D. D i ssertat ion, Univers ity of Wales, 1976. 9 in the greater access to inputs enjoyed by the former. Jabbar also f inds that output per unit of rented land has in general been less than that per unit of owner-cultivated land. As with the study by Hossain, c i ted e a r l i e r , Jabbar confines his analysis of d i f fe rent tenure arrangements within a s t a t i c framework. But from a development standpoint, i t i s not adequate to examine whether one tenure arrangement i s better than the other at a point in time. It also needs to be ca re fu l l y considered which one of them can best meet soc iety ' s future needs and explo i t i t s f u l l c apab i l i t i e s . Besides, i f both arrangements are found to be i n e f f i c i e n t from the point of .view, of u t i l i z a t i o n of social resources, i t may be necessary to seek a lternat ives beyond the arrangements already preva i l ing . Not a l l researchers have found results consistent with those obtained by Jabbar. Zaman, for example, found that in certain cases 11 sharecropping tenants have performed as well as the owner-cult ivators. However, he omitted to mention that the reason why a tenant may cu l t i va te rented land as intens ive ly may be related to considerations ar i s ing from his subsistence level of income. The suggestion made by Zaman that sharecropping arrangements should be encouraged in certa in instances to achieve more intensive c u l t i v a t i o n , may not, therefore, be ' M.R. Zaman, 'Sharecropping and Economic E f f i c iency in Bangladesh 1 , Bangladesh Economic Review, Vol. I, No. 2, Apr i l 1973. 10 able to achieve the stated goal i f the tenant households are found to operate above subsistence leve l s . Another study that makes more qua l i f i ed statements about the re l a t i ve e f f i c iency of tenant c u l t i v a t i o n , as compared to owner 1 g c u l t i v a t i o n , has also been published.by Hossain. He argues that one has to establ i sh that the farmers being compared in terms of the i r production performances have s imi la r access to other resources, in order that the i n t r i n s i c advantages or disadvantages of one arrangement over the other can be demonstrated. Thus, for example, he compares the performance of owner-cum-tenants (defined as those who cu l t i va te some owned land and some rented land) on t he i r owned land with t he i r performance on rented land. Since Hossain f inds that the owner-cum-tenants show greater e f f i c i ency on owned land than on rented land, he concludes that sharecropping has been i n f e r i o r to owner cu l t i v a t i on in the region he has studied. While his conclusion may be t rue, the methodology used for coming to i t has certa in problems. The overal l s ize of the owner-cum-tenant farms could evidently influence the resu l t s . When the owner-cum-tenants have T i t t l e land of the i r own and can get only a l i t t l e more on sharecropping contracts, they may not be able to afford half-hearted cu l t i va t i on of the rented land and may have to cu l t i va te a l l land with high intens i ty to meet family consumption needs. Besides, i t might so happen that an owner-cum-tenant cu l t i vates the rented land M. Hossain, 'Farm S ize, Tenancy and Land Product iv i ty: An Analysis of Farm Level Data in Bangladesh Ag r i c u l t u r e ' , Bangladesh Development Studies, Vo l . V, No. 3, July 1977. 11 less intens ive ly than he does his own land, but more intens ive ly than i t would be cu l t ivated by other owner-cult ivators. Studies on land reform: Exist ing studies on land reform in Bangladesh deal more with the background to land l e g i s l a t i on in the country than with an analysis of what might be included in a future land reform programme. Most of these studies have indicated the f a i l u re s of previous l e g i s l a t i v e attempts at reform. For example, Mukherji came to the fol lowing conclusion in this, regard: " . . . the provision of land ce i l i n g l e g i s l a t i on in Bangladesh does not indicate any radical s h i f t , from the East Bengal State Acquis i t ion and Tenancy, Act of 1950. Whatever boldness was introduced in the program i n i t i a l l y to prevent the circumvention of the land ce i l i n g law has been gradually eroded by subsequent amendments in response to demands from pressure groups."17 A major impl icat ion of Mukherj i ' s study i s that land red i s t r ibut ion in i t s e l f can const itute a desirable land reform programme for Bangladesh. However, he f a i l s to substantiate th i s by any theoret ica l or empirical invest igat ion into the relat ionships among size-tenure structure of farms, product iv i ty of land or labour, and d i s t r i bu t i on of income. I:N. Mukherji, 'Agrarian Reforms in Bangladesh', Asian Survey, Vol. XVI, No. 5, May 1976, p. 463. 12 In another study by A l im, i t has been concluded that in view of the shortage of land in Bangladesh, hardly anything can be 1 g achieved through a programme.of land red i s t r i bu t i on . He does not, however, explore any a l ternat ive means of transforming the size-tenure structure for atta in ing greater equity and increased productive e f f i c i ency . He argues that those with very l i t t l e or no land at a l l should be engaged in "various trades, professions and cottage indust r ies " so that they can "earn t he i r l i ve l i hood 1 g with d i gn i ty " . The whole study i s marred by general izations l i k e these. Besides, lack of ob jec t i v i t y and a s t a t i c nature of argumentation are severe l im i ta t ions to A l im ' s work. Abdullah has undertaken a.study of land reform in Bangladesh which i s quite valuable as far as h i s t o r i c a l descr ipt ion i s concerned. Like most other researchers in th i s area, he does not tackle the problem from the point of view of pol icy formulation. However, at the end of his study, he does of fer an opinion about the v i a b i l i t y of land 20 18 A. Al im, Land Reforms in Bangladesh: Social Changes, A g r i c u l t u r a l Development and Eradication of Poverty, Dacca: Samina (publ isher), 1979. 1 9 Ibid., pp. 137-138. 20 A. Abdullah, 'Land Reform and Agrarian Change in Bangladesh', Bangladesh Development Studies, Vo l . IV, No. 1, January 1976. 13 red i s t r ibut ion as a tool for atta in ing equity and growth, "A regime of small peasant proprietors w i l l be less propitious for accumulation and growth, less organical ly integrated with the market than even the ex i s t i ng : pet ty commodity producing semi-feudal structure ... It i s foo l i sh to expect, for example, that small farmers w i l l continue to grow as much cash crops as they now do when there are no longer any large farmers disposing of large surpluses of r ice."21 In Abdullah's opinion, cooperative measures patterned af ter the Comilla experience would be better able to foster growth and "weaken the la s t vestiges of semi-feudalism" in Bangladesh 22 agr icu l ture. However, he admits that these measures w i l l not be l i k e l y to mitigate i nequa l i t i e s , which means that one of the major objectives of land reform w i l l remain unrealized. Khan has undertaken a study of land reform for Bangladesh which includes some empirical invest igat ion into the problems of size-tenure 23 structure. He bases most of his analysis on data coming from the Agr icu l tura l Census of 1960, and some on the data from the Master 24 Survey of Agr iculture of 1968. In the context of the r e a l i t i e s of Ibid., p. 96. Ibid., p. 96. Also see in th i s regard A. Abdullah, M. Hossain and R. Nations, 'Agrarian Structure and the IRDP: Preliminary Considerations ' , . Bangladesh Development Studies, Vo l . IV, No. 2, Apr i l 1976. Khan, op.oit., Ch. 11, pp. 128-150. Government of Pakistan, Pakistan Census of A g r i c u l t u r e , 1960, Vol. I, Final Report - East Pakistan, Part I, October 1962; and Government of Bangladesh, Master Survey of Agriculture in Bangladesh, Seventh Round, Second Phase (1967-68), Dacca: Bangladesh Bureau of S t a t i s t i c s , (undated). 14 more recent times, one would have to revise many of Khan's contentions. Some of the relevant variables have undergone s i gn i f i cant changes over the years. For example, whereas in 1960 the tenant-operated area was 18% of a l l cu l t i vated land, and landless households formed about 7% of a l l agr icu l tura l households, the corresponding f igures were reported in the 1977 25 Land Occupancy Survey to be about 23% and 33% respect ively. Khan points out that the reform of 1950, which abolished zamindari, l e f t many farmers who had enough land to s t i l l be able to behave l i k e rural gentry. He also maintains that the i r r e l a t i v e l y large farms have been less e f f i c i e n t in the use of land and other resources and "have not shown any s i gn i f i can t tendency to modernize". It has been suggested by Khan that by putting a c e i l i n g on the land ownership of a family at 7.5 acres, and red i s t r ibut ing the ' surp lus ' land among the small owners and, i f poss ib le, the landless, one could achieve both appreciably greater equity and increased product iv i ty. However, as we sha l l see in Chapter V of the present study, under'the ex ist ing circumstances even with ce i l i ng s much lower than t h i s , not much can be achieved by th i s means to increase the welfare of the peasantry at large. Government of Bangladesh,..Summary Report of the 1977 Land Oocupanoy Survey of Rural Bangladesh, Dacca: Bangladesh Bureau of S t a t i s t i c s , 1977. Khan, op.oit., p. 134. 15 It would be f a i r to interpret Khan's comment on the poor performance of the r e l a t i v e l y large farms to mean that the ex i s t ing i n s t i t u t i ona l framework has not been conducive to improving the e f f i c i ency of larger farms. It would be inappropriate, however, to conclude from such a premise (and Khan does not do so) that large farms per se must inev i tab ly be detrimental for Bangladesh under a l l circumstances. Khan does not consider p o s s i b i l i t i e s of solving some of the problems of Bangladesh agr icu l ture by creating larger units under a d i f fe rent set of circumstances (for example, under group ownership arrangements). A more recent study addressing i t s e l f to the question of future 27 land reform in Bangladesh has been published by Zaman. He suggests that a c e i l i n g of 8 acres per family be put on landholding. The reason he does not advocate a lower c e i l i n g (for example, 5 acres) i s that "such a unit might not be v iable given the requirements of a minimum income, f u l l family employment, and optimum u t i l i z a t i o n of 28 farm implements, including bul locks " . Although the ce i l i n g he suggests can hardly be expected to have much impact in the present circumstances of Bangladesh agr i cu l tu re , in making th i s statement, he exposes some of the macro-deficiencies of small scale family farming in Bangladesh. M.A. Zaman, 'Bangladesh: The Case for Further Land Reform', South Asian Review, Vol. 8, No. 2, January 1975. Ibid., p. 99. 16 With regard to cooperation in Bangladesh agr i cu l ture , most studies dwell upon the r e l a t i v e merits and demerits of Comilla type cooperatives. Malek has made one such study, in which he f inds that the i n i t i a l broad enthusiasm for the Comilla movement 29 has apparently died down. He observes that the landless and the land-poor farmers have no incentive to j o i n the cooperatives because they are well aware that benefits avai lable accrue disproportionately with the s ize of landholding, although obl igations to the cooperatives more often remain s imi la r for a l l s izes of member farms. Malek's study f a i l s to make the point that the anomalies found in the Comilla cooperative movement germinated from the very conceptualization of the programme. It was thought that a s p i r i t of community development could be created through persuasion, assistance, and organizational guidance. The mistake was to disregard the fact that no such thing as a ' v i l l a g e community'1 was l i k e l y to emerge when the interests of d i f ferent s ize and tenure-categories of farms were so widely divergent. That the problems of the Comilla model are bound to reappear in other programmes patterned a f te r i t . ( f o r example, the Integrated Rural 30 Development Programme) has been demonstrated by Khan. He Q'.M.A. Malek, 'Rice Cul t ivat ion in Comilla Kotwali Thana: The Role of Cooperatives ' , Bangladesh Development Studies, Vol. IV, No. 3, July 1976. A.R. Khan, "The Comilla Model and the Rural Development Programme of Bangladesh: An Experiment in Cooperative Capital ism' , World Development, Vo l . 7, No. 4/5, April/May 1979. 17 concludes, "The consequence of intens i fy ing the implementation of the IRDP programme in a s i tuat ion of such i n i t i a l inequal i ty w i l l be to perpetuate, more l i k e l y to aggravate, such inequal ity."31 Although some authors have recognized the need to introduce group farming, which would be based on j o i n t r e spon s i b i l i t i e s in ownership, organization and management, no one has so far provided a comprehensive examination of the issue. S idd iqu i , for example, has expressed strong reservations against t reat ing red i s t r ibut ion 32 of land as a s t a t i c pol icy measure. He maintains, "In the long run, an ega l i ta r ian landholding structure provides the real basis for agr icu l tura l cooperation and c o l l e c t i v i s a t i o n which, in turn, f a c i l i t a t e s economies of scale, the large-scale introduction of modern technology, increased production and a larger agr icu l tura l surplus."33 Siddiqui has, however, stopped short of spe l l ing out the deta i l s of how the proposed t rans i t i on i s to be brought about. Several points stand out from the above survey of ex i s t ing l i t e r a t u r e on size-tenure structure and land reform in Bangladesh: 1. Most previous studies deal only with s t a t i c e f f i c i enc i e s . The long-term consequences of a l ternat ive pol icy measures hardly receive any Ibid., p. 417. K. S idd iqu i , 'Comment on Jannuzi and Peach: "A Note on Land Reform -in 'Bangladesh". 1 Journal of Peasant Studies, Vol. 6, No. 3, Ap r i l 1979. Ibid., p. 320. 18 attent ion. This has been the resu l t of confining the discussion to t r ad i t i ona l debates only. 2. The comparison of levels of performance of d i f fe rent categories of farms i s undertaken mainly at the micro l e v e l . Most researchers have f a i l e d to look into the p o s s i b i l i t y of a contradict ion ex i s t ing between micro-level e f f i c i ency in the given i n s t i t u t i ona l framework and the achievement of broader macro-efficiency. 3. Although some authors have argued that cost-sharing would make sharecropping an e f f i c i e n t arrangement, no one has sought to explain why cost-sharing has not been a more common arrangement between landlords and sharecropping tenants in Bangladesh. 4. A l l studies which have made po l icy suggestions re la t ing to land reform have been mainly preoccupied with a comparison of e f f i c i ency between sharecropping tenants and owner-cultivators. Following Herring, i t can be argued that th i s may not be the best way of approaching the problem, "We are comparing tenancy systems to owner-cultivator systems; that one is marginally superior to the other in product iv i ty does not se t t l e important pol icy questions because both systems f a l l pa thet i ca l l y short of the potential in terms of labour product iv i ty and land product iv i ty. "34 R.3. Herring, 'Share Tenancy and Economic E f f i c iency : The South Asian Case'. Peasant Studies, Vo l . 7, No. 4, Fa l l 1978, p. 244. 19 5. Land reform, in most instances, has been treated as synonymous with land red i s t r i bu t i on . The prospects of other models in atta in ing the objectives of a land reform programme have not been adequately investigated. 6. Most previous studies of land reform have been based on data which are almost two decades o ld. Conclusions drawn from these studies require re-examination in the context of more recent information on the Bangladesh s i tua t ion . 7. None of the studies have considered regional var iat ions in discussing the problems of size-tenure structure. The above l i t e r a t u r e review thus indicates that a researcher concerned with the problems of size-tenure structure and prospects of a land reform programme in Bangladesh has an opportunity to cover new and important ground. The present study hopes to f i l l some of the void that has been shown to ex ist in the ava i lable l i t e r a t u r e . 20 • II. CONCEPTUAL FRAMEWORK AND PLAN OF STUDY Conceptual Framework The relevance of size-tenure structure: In seeking to diagnose the prospects for ag r i cu l tu ra l development in any underdeveloped country, i t i s worthwhile to s tar t by examining the prevai l ing ' a g r i cu l tu ra l systems' within the country which determine the spec i f i c context of ' the problem'J Following Morrison, one can ident i f y four d i f fe rent c lusters of elements which can af fect the cu l t i v a t o r s ' production behaviour as 2 well as the nature of d i s t r i bu t i on of output and income among them. The natural cu l t i va t i ng environment constitutes the f i r s t of these c lusters . Included herein are the s o i l composition of an area, the timing and the quantity of r a i n f a l l during a year, drainage f a c i l i t i e s , access to i r r i g a t i o n water, and exposure to sun and wind. The second c lus ter represents the resources owned or under the control of the cu l t i v a t o r ; the important resources being land, labour, and cap i ta l stocks l i k e farm implements and draught animals. Not only the quantity of land held by a cu l t i v a t o r , but also i t s qua l i ty in terms For an elaboration of the concept of an agr icu l tura l system see B.M. Morrison, 'The Persistent Rural C r i s i s in As ia: A Sh i f t in Conception', Pacific Affairs, Vol. 52, No. 4, Winter 1979-80. 2 Ibid., pp. 633-637. 21 of the range of crops which can be produced on i t , the degree of fragmentation of the land he holds, and the proximity of the farmland to his res ident ia l unit are some of the other considerations which characterize the resource pos it ion of a cu l t i va to r in terms of land. The labour resource would include family labour as well as labour which would be made avai lable from the wage-labour market or through arrangements of labour exchange with other households. Again, the labour resource posit ion of a household would be determined by both the quantity of labour at i t s disposal as well as the qua l i ty of that labour. The th i rd c luster of elements covers the soc ieta l expectations which define the re lat ionsh ip between an indiv idual household's e f fo r t s toward acquiring the necess it ies of l i f e and enhancing i t s resources, on the one hand, and the demands made on i t by the "kindred and fe l low-c lass households" and the society at large, 3 on the other. The fourth c lus ter i s composed of the linkages of the agrarian society to outside agents and organizations. These linkages are established through the market re lat ions of that society with the 'outs ide ' world and through the actions of government and non-government agencies (national and foreign) with regard to the bui lding of i n f r a -st ructure, provision of extension services and implementation of various reform measures. 3 Ibid., p. 635. 22 The performance of the agr i cu l tu ra l sector, both in terms of e f f i c i ency in production and equity in d i s t r i bu t i on of income, of a par t i cu la r country or a region can be altered to the extent that the four components of an ag r i cu l tu ra l system discussed above are amenable to change. The f i r s t one points to the environmental determinism of the system. The s o i l composition, r a i n f a l l , surface topography and water table influence the methods of cu l t i v a t i on as well as the cropping pattern within a region. It i s true that a cu l t i va to r can attempt to make arrangements to create better drainage and i r r i g a t i o n f a c i l i t i e s within the given set of rainfal1-topography-water table conditions. However, these arrangements do not n u l l i f y the environment-specif ic ity of his operations. The second c luster of elements, which includes the c u l t i v a t o r ' s resources, would be more amenable to change. The d i s t r i bu t i on of farmland among the rural households can change with or without outside intervention. Such factors as population growth, h e r i t a b i l i t y r i gh t s , terms and conditions under which land mortgages are found to operate, and the att i tude of d i f fe rent groups toward agr i cu l tu ra l occupations as opposed to other occupations can be thought of as endogenous variables af fect ing the resource posit ion of a cu l t i v a to r in regard to the farmland he operates. His pos it ion can also be affected by l e g i s l a t i on introduced by the government. The qua l i t a t i ve content of his land, as characterized by the range of possible crops and the fragmentation of his holding, can be altered by the introduction of 23 newer crops through adaptive research work and by the adoption of voluntary or state-sponsored consolidation programmes. The labour resources of a cu l t i va t i ng household are best appreciated in re la t ion to i t s command over land resources. The p o s s i b i l i t y of a l terat ions in the land resources avai lable to the household obviously raises the prospect of altered factor proportions in production arrangements. The t i l t i n g of the balance of land resources can resu l t in a change in the or ig ina l labour resources of the households. For example, fewer labour units may be avai lable in the wage-labour market in cases where the households which were previously supplying these units have acquired so much land as to withdraw some of th i s labour to be used for cu l t i v a t i on of t he i r own farmland. Part of the th i rd c lus ter of elements, representing soc ieta l expectations, can be thought of as a force of cu l tura l determinism in the production and d i s t r i bu t i on behaviour of the cu l t i va tor s in the short run. Kinship t i e s and interdependence among various groups cannot change overnight. However, societal expectations and the i nd i v i dua l ' s reaction to these expectations are bound to change with changes brought about in the resource pos it ion of the household. Af ter a l l , these societal expectations are the re f lec t ions of the i nd i v idua l s ' perception of those secur i t ies which they feel they must have to guard against potential imbalances in t he i r resource posit ions and, hence, t he i r capacity to enhance family interests . 24 The fourth c l u s te r , which includes the linkages of the agrarian society to the outside world, both national and in te rnat iona l , i s of extreme importance in inf luencing the product ion-distr ibut ion or ientat ion of the society. The nature of these external linkages are determined by a var iety of factors which are largely outside of the indiv idual peasant's autonomy. The market r e l a t i on s , which shape many of these l inkages, would be.dependent upon the pol icy of the State with regard to extraction of a marketable surplus and i t s use. Leg i s lat ive actions of the government are l i k e l y to be designed in accordance with such po l i c i e s . For example, i f the government feels that the d i s integrat ion of ex i s t ing large farms within the country would run counter i t i t s po l icy with regard to increased marketing of the products, as opposed to self-consumption by the cu l t i v a to r s , i t would be hesitant to l eg i s l a te low land ce i l i ng s . The degree of hes itat ion would vary with the ca lcu lat ion of p o l i t i c a l costs and benefits. The p o l i t i c a l cost of not implementing low land ce i l i ngs may l i e in the a l ienat ion of the small land owners and landless households who could have benefited by subsequent red i s t r ibut ion of land. On the other hand, supporting the r e l a t i v e l y large farmers by various means w i l l understandably ensure the i r p o l i t i c a l a l legiance. The government may also f ind newer p o l i t i c a l a l l i e s in the growing urban areas i f i t can keep the urban dwellers happy with an adequate supply of food items at cheap pr ices. Again, the p o l i t i c a l cost-benefit analysis w i l l become relevant to the government as con f l i c t s of interests become acute between those 25 (both for consumption and manufacturing) and rural suppliers of these products. If p o l i t i c a l gain is perceived to be higher in making a l l iances with the former group, l e g i s l a t i on might be forthcoming to a l t e r the balance of resource ownership and control among d i f fe rent rural groups. In th i s way, the resource posit ions of d i f fe rent groups can both influence and be influenced by linkages established by the government. Non-government organizations can also r a l l y various interest groups to press for a l terat ions in re l a t i ve and absolute resource s i tuat ions. The above discussion underscores the importance of the resource posit ion of cu l t i va tor s in determining the agr icu l tura l system within which they operate. Many factors can a f fect product iv i ty and income d i s t r i bu t i on which may be independent of size-tenure structure. However, land i s the primary factor of production and attains par t i cu la r importance in the whole resource package avai lable to the cu l t i va tor s espec ia l ly in resource-poor densely populated countries. This i s so because (1) land occupies a unique status as a resource c r i t i c a l to basic subsistence and productive a c t i v i t i e s , and (2) ownership and u t i l i z a t i o n of other resources l i k e seeds, f e r t i l i z e r s and draught animals i s often e i ther internal to or dependent upon the ownership of land. It follows from this, that agr icu l tura l systems in.an underdeveloped country would be greatly influenced by the land tenure system, which can be taken to include the customs and organizational arrangements by which land i s owned and cu l t ivated by the members of an agrarian society. 26 The concept of land tenure, comprising both the status of land ownership enjoyed by the cu l t i va tor s and the organizational arrangement within which land i s cu l t i va ted , provides a basis for examining the decisions of the cu l t i vator s with regard to such important variables in the development matrix as the in tens i ty of input-use, cropping pattern, choice of technology, level of employment and the generation and use of marketable surpluses. As a simple example of a contrasting agr i cu l tura l system, one could look at the land tenure arrangement in a plantation economy and examine i t s ramif icat ions upon various economic, social or p o l i t i c a l outcomes. Ownership of land in a typ ica l plantation i s in the hands of 4 corporations of which there may be a large number of shareholders. These corporations are mostly foreign-based and the majority of the i r shareholders are resident abroad. The plantations are wage labour-based, export-oriented enterprises. The export or ientat ion often necessitates large investments in processing of the products and in transportation. To minimize the i d l e capacity of the whole capita l outlay, crops which can be harvested cont inual ly are favoured. The in tens i ty of input-use is high in the plantations due to s t r i c t control over the use of various inputs within a framework of p r o f i t motive. Although most plantations are established in sparsely For an elaboration of the plantation economy see P.P. Courtenay, Plantation Agriculture, London: C. Bel l and Sons L td . , 1965. 27 populated l o c a l i t i e s (these l o c a l i t i e s w i l l lend themselves to the large-scale farming typ ica l of the plantations without creating major social and p o l i t i c a l imbalances) cheap labour i s made avai lable by encouraging immigration from more populous parts within or outside of the country. The labour-intensive technology of the plantations 5 i s often a resu l t of the nature of crop involved as w e l l . The rel iance on labour-intensive technology does not, however, mean non- res t r i c t i ve labour employment since the p ro f i t maximization motive would i dea l l y require that;employment be l im i ted at the point where marginal product of labour equals i t s wage. The owners of plantations are often able to exercise some degree of monopsony power over the levels of remuneration paid to workers since the host countries are mostly unable to of fer these workers a d i ve r s i f i ed set of income-earning opportunit ies. To combat the leverage exercised by the owners in sett ing wages, the workers would be tempted to unionize and start a process of c o l l e c t i v e bargaining. If the government cannot or w i l l not resort to coercion fo r suppressing such developments then the plantations can soon become non-viable as high-p r o f i t enterprises due to r i s i n g labour costs and the increased uncertainties in business operations. For example, rubber tapping is done most e f f i c i e n t l y by an expert hand with' nothing more than a kn i fe, plucking of qua l i ty tea leaves i s best done by bare hands, and the harvesting of banana crops requires a kind of se lect ive judgement which can be provided best by human labour. 28 The b r ie f discussion of the plantation economy carr ied out above shows how the ownership pattern of land and the organization of i t s production af fect such factors as the cropping pattern, choice of technology, intens i ty of input-use, nature of market structure, d i s t r i bu t i on of income, and the long-term v i a b i l i t y of the arrangement as a whole. The land tenure system under other arrangements ( l i k e sharecropping c u l t i v a t i o n , small-scale owner c u l t i v a t i o n , state farming, or group farming with j o i n t ownership and management) can also be expected to a f f e c t , admittedly in varying fashions and degrees, a l l of these factors . In other words, whatever the arrangement, the land tenure system plays a central ro le in shaping production and d i s t r i bu t i on behaviour. The question of farm s i ze , which i s int imately related to the land tenure system, i s also of great importance in determining the nature of such cruc ia l variables in the development matrix as the d i s t r i bu t i on of income, technology in use, level of employment and the ex t rac t ib i 1 i.ty of surpluses from the market. In short, the size-tenure structure would greatly influence the nature of the various c lusters of elements which have been i den t i f i ed as determining the agr icu l tura l system. In t ry ing to ident i f y the problems of ag r i cu l tu ra l development within a region, i t i s important to understand the process through which the ex i s t ing s i tuat ion has emerged, espec ia l ly with regard to i t s s ize-tenure structure. This can help one d ist inguish between the forces which tend to perpetuate the prevai l ing socio-economic system, 29 and those which push for change. The success of any programme of ag r i cu l tu ra l development in an area which suffers from numerous problems related to i t s s ize-tenure structure w i l l depend upon the a b i l i t y to properly i dent i f y and mobil ize appropriate forces of change. Finding a viable structure which seeks to better sa t i s f y basic equity and e f f i c iency c r i t e r i a is what constitutes the essence of a programme of land reform. The fol lowing section: w i l l examine some of the spec i f i c objectives which may be set for such reform programmes and which govern the se lect ion of a su i table de f i n i t i on for the term ' land reform'. Various models of land reform w i l l then be considered, which w i l l be followed by a discussion of the p o l i t i c a l factors which may af fect the i n i t i a t i o n and implementation of a reform. Objectives of land reform: Depending upon the preva i l ing size-tenure structure, one or more of the fol lowing may const itute the set of basic development-related objectives of a land reform programme: ( i ) d i s t r i bu t i ng income more equitably, ( i i ) r a i s i n g the level of production, ( i i i ) generating more employment opportunit ies, ( iv) increasing the marketable surplus. These objectives have to be discussed in some deta i l before one can come to terms with the de f i n i t i ona l n icet ies of the concept of 30 land reform since the various forms which a ' land reform 1 can take w i l l obviously be affected by the objectives being set for i t . ( i ) D i s t r ibut ing income more equitably. A lternat ives to land reform? Raising the levels of income of small farmers and the landless constitutes a major objective of many land reform programmes. Measures other than land reform have often proven unsuccessful in e f f e c t i v e l y red i s t r ibut ing income among the agr i cu l tu ra l population. A b r ie f review of the problems and prospects of these other measures w i l l enable one to appreciate the importance of incorporating the objective of a more equitable d i s t r i bu t i on of income within a programme.of land reform.^ ' A l te rnat i ve s ' to land reform in red i s t r ibut ing income may include (a) tenancy reform, (b) small-farmer a i d , and (c) progressive land taxat ion. (a) Tenancy reform: Tenancy reforms are pr imar i ly concerned with contro l l ing rents and establ ishing security of tenure for the tenants. The basic shortcoming of a l l tenancy reforms is that these are not d i r e c t l y concerned with the problems of landlessness. But l e t us examine b r i e f l y the role that tenancy reform can play in bettering the lo t of those for whom i t i s designed, that i s , the tenants. For a detai led analysis of these a lternat ives see M. L ipton, 'Towards a Theory of Land Reform', in D. Lehmann (ed.), Agrarian Reform and Agrarian Reformism: Studies of Peru, Chile, China and India, London: Faber and Faber, 1974, pp. 274-281. Most tenancy reforms are involved in tack l ing the problems of sharecropping tenancy. 31 The tenants are dependent on the i r respective landlords not only in gaining access to land, but also in a var iety of other ways. The landlords may help the tenants in providing personal protect ion, extending loans, making contacts with government o f f i c i a l s , f inding marketing channels, arranging part-time employment, and the l i k e . The tenants often rea l i ze that the State may not provide them with a l l of these services i f tenancy reform causes the landlords to terminate such patronage. This i s why they may not i n s i s t on the i r r ights even i f laws are made on one or more of the po l icy measures of tenancy reform. And, without the cooperation of the tenants, none of these measures can achieve any great degree of success. The strength of the landlord l i e s in his stock of the scarce resource that land represents. Tenancy l e g i s l a t i on cannot do much to c u r t a i l such strength and to el iminate the socio-economic obstacles imposed by the nature of rural power structure. Tenancy reforms shy away from the fundamental issue in agrarian reform and, therefore, the red i s t r ibut ion of income that can be effected through such measures i s usually l i t t l e . In the words of L ipton, "In general, tenancy reform in the ' so f t s t a t e s " of the Third World breaks upon the rock of landlord power, and the effects of evasion can include insecur i ty that worsens both rural income d i s t r i bu t i on and the standards of capita l and „ land acquis i t ion and maintenance on tenant farms." L ipton, op.oit., p. 277. 32 (b) Small-farmer a i d : The programme of small-farmer aid rests on the be l i e f that by offer ing cap i ta l assistance to small farmers one can solve not only the problem of d i spar i ty among the r ich and poor agr i cu l tura l households, but also boost agr i cu l tura l production. The t rad i t i ona l argument regarding the e f f i c i ency of small farms i s resorted to in suggesting that these w i l l perform even better with more c a p i t a l . Whether the government exchequer can bear the burden of an extensive programme of small-farmer aid i s a fundamental question in th i s regard. Besides, such aid again does not d i r e c t l y deal with landlessness. The pos it ion of the landless can in fact worsen i f the transfer of capi ta l i s made from those land owners who hired most of the rural wage-labour insofar as the demand for wage-labour i s reduced by the move. Moreover, small-farmer aid does not aim at solving the problems of tenant cu l t i v a t i on . In an agrarian structure dominated by r e l a t i v e l y large farmers, cred i t intended for small farmers i s l i k e l y to f ind i t s way to the pr iv i leged few who have the p o l i t i c a l strength to influence the 9 di rect ion of aid d i s t r i bu t i on by local author i t ie s . And once the cred i t ends up in the 'wrong' hands, i t may not be used for ag r i cu l tu ra l purposes, or i t may be cycled by them to the small farmers at higher rates. The experience.of Bangladesh in th i s regard, discussed in Chapter V, offers evidence of such happenings. 33 (c) Progressive land taxat ion: There i s at least the theoret ica l p o s s i b i l i t y that through the imposition of a progressive land tax the government can at ta in a considerable level of equity in the rural sector. The idea i s that a high tax on the possession of land beyond certa in l im i t s may discourage the large land owners from keeping a l l that land at the i r disposal. This, i t i s hoped by the proponents of th i s measure, w i l l help in an automatic ega l i ta r ian red i s t r ibut ion of land. Problems of administering a progressive land taxation pol icy can, however, be overwhelming. Proper information on land r eg i s t r a t i on , which i s a prerequis ite for determining the levels of taxes to be paid by d i f fe rent households i s t y p i c a l l y absent in most underdeveloped countries of today. Therefore, evasions are l i k e l y to be very common. The government cannot expect as enthusiast ic a par t i c ipat ion from the land-poor masses in implementing a tax pol icy as i t could in attempting to implement a land reform programme. The reason i s that the land-poor households do not perceive any immediate substantial gain from taxation po l i c ie s carr ied out by the 'd i s tant S ta te ' . And without the i r wholehearted cooperation no such po l i c i e s of the government can meet with success. The b r i e f discussion above brings us to the conclusion that the measures that have usually been suggested as a l ternat ives to land reform in atta in ing equity can only have l imited success and cannot be a subst itute for land reform. 34 Land red i s t r ibut ion in atta in ing equity: When i t comes to the question of attaining a more equitable d i s t r i bu t i on of income through a land reform programme, the usual suggestion is to red i s t r ibute part of the land belonging to the r e l a t i v e l y land-r ich among those who are r e l a t i v e l y land-poor. However, as Berry has shown, i t i s a mistake to assume that a l l such red i s t r ibut ion w i l l resu l t in general w e l f a r e . ^ The argument i s summarized in the fo l lowing. Under conditions of perfect markets for the factors of production and the product, and constant returns to sca le, factor proportions on farms would be s i ze -neut ra l . That i s , a l l farms, small and large, would use various factors of production in the same proportion. Within such circumstances, transference of land from one farm to the other would be equivalent to a t ransfer of capi ta l and income. Therefore, land reform, through a red i s t r ibut ion of land, would be able to achieve a more equitable d i s t r ibut ion, of income in a region where the i n i t i a l d i s t r i bu t i on pattern was unequal. However, perfect markets are not a r e a l i t y . Under conditions of imperfect markets, red i s t r ibut ion of land may not necessari ly mean a more equitable d i s t r i bu t i on of income. One can consider the fol lowing R.A. Berry, 'Land Reform and the Agr icu l tura l Income D i s t r i b u t i o n 1 , Pakistan Development Review, Vol. XI, No. 1, Spring 1971. 35 two types of farms for the sake of theoret ical exposit ion: (a) large farms, where landless wage-labour and some labour from smaller farms are employed, and (b) small farms, where cu l t i va t i on i s based on family labour. Now, i f land reform takes land from the large farms and d i s t r ibutes i t to the small farms, i t might hurt the landless wage-labour. The important thing to consider in th i s regard i s the nature of the fol lowing re lat ionsh ip : h W <; f where, R, = the extra labour hired by rec ip ients of land a f te r the red i s t r i bu t i on . R = the labour that i s withdrawn from the wage-labour market by the rec ip ients of land so that they may apply th i s labour to the i r own land (which increases in amount a f te r the red i s t r ibut ion of land takes place). L f = the labour that gets f i r e d by the large farms due to the i r having to surrender some land they had o r i g i n a l l y owned. If Rh + Rw i s greater than lf, the wage-earner w i l l benefit since i t means that the e f fec t i ve demand for t he i r labour has, in f a c t , increased after the red i s t r i bu t i on . If R, + R i s equal..'to> h w ^ the landless wage-earner maintains his or ig ina l status. If R. + R f 3 3 h w i s less than L^, the wage-earner faces a decrease in the level of wage, which i s brought about by a decrease in the e f fect i ve demand for wage labour. 36 The ef fect that red i s t r ibut ion of land to the small farms may have on the landless i s i l l u s t r a t e d in Figure 1, which was suggested by Be r r y . ^ t o 3 O <D o > \ \ \ \ N \ \ " \ V -J rv: V »| ^ \ •— P' L, Labour Figure 1. Land Redistr ibution Among Small Farms and Its Effects on the Landless Ibid., p. 34. 37 Total family labour avai lable to a typ ica l small farm i s represented by OL in Figure 1. MP represents the marginal product of labour for the farm. ST i s the supply pr ice of labour curve which gives the re lat ionship between the wage at which the marginal indiv idual would work off-the-farm and the units of labour on the farm. Once the costs of transportation and being away from home are considered, the supply pr ice of labour curve corresponding to MP sh i f t s up to S T . The units of labour offered as wage-labour by the farm i s measured to the l e f t along the X-axis with L as the o r i g i n . I f , for example, the wage offered i s OW, the farm would be w i l l i n g to supply LL.! amount of Tabour to the wage market. Let i t now be supposed that a land reform programme i s implemented which gives some extra land to the farm, and that the marginal product of labour curve for the farm goes up to M 'P ' . The supply price of labour curve (considering costs of transportation and being away from home) corresponding to M'P 1 i s S"T". Here, i f the wage offered i s OW, the typ i ca l farm that i s being considered would o f f e r LL 2 amount of labour to the wage market. Redistr ibut ion of land, in the context of Figure 1, has reduced the amount of labour offered to the wage market by the typ ica l small farm by the amount, LLi. - LL 2 = L 2 L i 38 Whether the landless wage-labour w i l l benefit or not a f te r the kind of land red i s t r ibut ion that has been considered here, would depend part ly on. whether l2l\is greater or smaller than the labour previously hired on the transferred land. If i t i s greater, the landless wage earner gains; i f smaller, he loses. It i s possible to express the change in the demand for wage-labour a f ter the red i s t r ibut ion of land takes place, say V, in the fol lowing way: V to ta l labour applied on small farms a f ter red i s t r ibut ion + [ family labour removed [from wage market family labour applied on small farms after red i s t r ibut ion [labour applied on [large farms If V attains a pos i t ive value then one could say that the red i s t r ibut ion of land would increase the welfare of the wage-labourer. If V equals zero, his status remains unchanged. When V i s negative, his welfare decreases. What the above exercise t r i e s to demonstrate i s that a pol icy which attempts to d i s t r ibute income more equitably through a programme of land red i s t r ibut ion may do more harm than good i f a l l the consequences are not ca re fu l l y considered in advance. Another point to note about the effectiveness of a programme of land red i s t r ibut ion in attaining equity in an area of high population density re lates to long run considerations. Even i f the i n i t i a l 39 red i s t r ibut ion of land succeeds in removing the wide d i s p a r i t i e s , the inequal i ty may tend to reappear over time. I f the rec ip ients of new land are l e f t with very small farms even af ter the red i s t r ibut ion takes place, many of them may need to se l l part or a l l of the i r land to the r e l a t i v e l y r i ch land owners (who are there insofar as the red i s t r ibut ion of land does not achieve absolute equal ity) to meet unforeseen expenditures necessitated by emergencies l i k e a crop f a i l u r e . It can be expected that these se l l e r s of land w i l l f i nd i t d i f f i c u l t to gather enough money for buying land to regain t he i r previous land-owning status. Therefore, although one can achieve some equity in the short run through red i s t r ibut ion of land, much of i t can be los t due to the dynamics of demand-supply conditions in the land market. ( i i ) Raising the level of agr icu l tura l production. If a land reform programme does not incorporate the objective of ra i s ing the general level of agr i cu l tura l production without much delay then most other objectives set for i t become less meaningful. Without an ultimate increase in the level of production, red i s t r ibut ion of income in most underdeveloped countries would resu l t in nothing better than shared poverty. There are certa in re lat ionships which ex i s t between the size-tenure structure and product iv i ty in the 12 agr i cu l tu ra l sector. Therefore, a plan intended for ra i s ing the These relat ionships are discussed at length in Chapter IV, pp. 140-209. 40 level of product iv i ty has to take into account the nature of spec i f i c re lat ionships obtaining there in th i s regard. The necessity of a f r u i t f u l adoption of the technology of the high y ie ld ing va r ie t ie s (HYVs) or some range of comparable technologies for increasing product iv i ty per unit of land and labour cannot be overemphasized. Circumstances more conducive to the adoption of HYVs (or some range of comparable technology) have to be created through a land reform programme i f the objective of ra i s ing the general level of ag r i cu l tu ra l production i s to be attained. This brings one to an examination of how certa in arrangements in agr iculture may hinder th i s adoption. Adoption of HYVs: Although the HYVs are s ize-neutral in terms of crop-spec i f ic y ie lds per unit of land, the bias of "green revo lut ion" technology nevertheless favours the r e l a t i v e l y large farmers. The 13 point i s i l l u s t r a t e d in Figure 2. Relative factor prices faced by the small farmer are represented by AB in Figure 2, and the ones faced by the Targe farmer are represented by CD. The small farmer uses a technology depicted by the levels of input-use corresponding to E. This technology i s labour-intensive "in nature. The technology expected to be used by large farms can be represented by F, at which cap i ta l i s used more intens ive ly than labour. The i l l u s t r a t i o n was proposed by K. G r i f f i n , The Political Economy of Agrarian Change: An Essay on the Green Revolution, London: The Macmillan Press, 1974, p. 50. 41 Figure 2. Labour-intensive and Capital- Intensive Technologies. A new technology that f a l l s within the area BGD represents an u l t ra-super ior s i tuat ion for the large farms which face a r e l a t i v e l y low price of capita l and material inputs. The bias in th i s case i s in favour of the large farms. To i l l u s t r a t e , isoquant qq may p a r t i a l l y s h i f t to q q 1 , thereby enabling the large farms to produce prof i tab ly in a more cap i ta l - in tens ive way at F '. The sh i f t i s i r re levant for the small farms since the only feas ib le combination 42 of outputs to produce the given level of output (represented by the isoquant) s t i l l corresponds to the previously attained point E. Movement from F to F' by large farmers was made possible by the introduction of a more cap i ta l - i n tens i ve technology. HYVs are cap i ta l intensive. Hence, the introduction of HYVs in a s i tuat ion of d i spar i ty in land ownership w i l l go in favour of the large farms. The b io log ica l character i s t i c s of HYVs are such that they need heavy investments in contro l led i r r i g a t i o n , chemical f e r t i l i z e r s and protective pest ic ides. Smaller farms cannot gather enough resources to make these investments. Besides, r i s k aversion, which is character i s t i c of small subsistence farming, requires that lumpy investments in s ingle projects are avoided. The only way of atta in ing large scale adoption of HYVs conducive to general welfare i s , therefore, to make sure that the access to land and the various inputs enjoyed by most farmers i s such that the adoption becomes both feas ib le and pro f i tab le for a l l of them. ( i i i ) Generating more employment opportunit ies. The level to which labour i s u t i l i z e d on a farm i s c losely related to i t s s ize-tenure structure. Re lat ive ly large land owners, who depend on wage-labour for cu l t i va t i ng t h e i r farmland, may end up employing fewer units of labour per unit of land than would be possible 14 under owner cu l t i v a t i on of small farms. Therefore, the objective of For an elaboration of th i s point see Chapter IV, pp. 194-195. 43 generating employment opportunities in the short run may require that dependence on wage-labour be reduced and that the farm area of those endowed with under-ut i l i zed family labour be increased. In the long run, however, small farms are l i k e l y to get smaller i f the v i r g in lands are exhausted and the arrangements for family farming operations are continued under laws of inheritance. Under -ut i l i zat ion of family labour i s l i k e l y to reappear in these farms since they would be l e f t with a high man-land r a t i o . In view of the shortage of land, i t i s necessary to create opportunities of labour absorption at a larger scale not only on farmland, but of f the farm as we l l . As Dorner puts i t : "Labour u t i l i z a t i o n in agr iculture should be evaluated in a context wider than that pertaining only to work on the farm. The broader issue concerns the mobi l izat ion of labour and cap i ta l for the development of rural industry and of soc ia l overhead cap i ta l (or infrastructure) in rura l areas."15 Therefore, a land reform programme, to be able to cater to the needs of long-term employment generation in the agr icu l tura l sector of a densely populated area, has to have the v is ion of that kind of an agr iculture which can provide the appropriate atmosphere for the generation of an increased level of savings, with which investments can be made to expand operations, both on and off the farm, within the rural areas. What type of a land reform programme w i l l best serve th i s purpose cannot, however, be i den t i f i ed without reference P. Dorner, Land Reform and Economic Development, Middlesex: Penguin Books, 1972, p. 107. 44 to the objective conditions of the land-man re lat ionsh ip obtaining in the country or region where reform i s to be carr ied out. ( iv) Increasing the Marketable Surplus. It i s of great interest to the urban sector that the marketable 1 g surplus from agr iculture be kept at high leve l s . Urban consumers want an assured supply of agr icu l tura l food items at prices they can af ford. Processors of raw materials l i k e to be assured of a steady supply of the inputs they need from agr icu l ture. Se l lers of consumer items are interested in seeing the rural sector monetized so that the market for t he i r products i s extended. Although generation of marketable surplus for meeting such urban demands may, under certa in circumstances, create unfavourable terms of trade for agr i cu l ture , creation of a surplus can also be important in developing the agr i cu l tu ra l sector i t s e l f . Increased savings to enable greater investments (for example, in the i ndu s t r i a l l y produced ag r i cu l tu ra l inputs) would not be possible without the monetization of agr icu l ture. The portion of the marketable surplus that can produce th i s money income has been considered by Sanghvi as part of the "obl igatory surplus" for the development of a g r i c u l t u r e . ^ Therefore, although the agr i cu l tura l sector may have to guard against the i n c l i na t i on of See B.F. Johnston and J.W. Mel l o r , 'The Role of Agr iculture in Economic Deve lopmentAmerican Economic Review, Vo l . L I , No. 4, September 1961, pp. 571-581. See P. Sanghvi, Surplus Manpower in Agriculture and Economic Development, Bombay: Asia Publishing House, 1969, pp. 186-198. 45 the urban sector to procure i t s surplus at unfavourable terms, a marketable surplus needs to be generated for the sake of agr icu l ture i t s e l f . How favourable w i l l the terms offered be to ag r i cu l tu re , and whether the proceeds of marketing w i l l be invested in ag r i cu l tu re , w i l l largely depend upon the relat ionships of cu l t i va t i ng peasants with the marketing agencies. The most su itable arrangement would usually involve some form of producer-cooperative for marketing, so that dependence on middlemen, who can reduce both the control of peasants over the market and the flow of proceeds into investments in agr i cu l ture , can be minimized. The goal of increased marketable surplus may well be better served in 'laissez-faire' economies where concentration of land ownership i s quite high. However, marketable surplus coming from a few large land owners is not consistent with the goal of better d i s t r i bu t i on of income since the l i o n ' s share of such a commercialization would be pocketed by those few land owners. In countries where the concentration of land ownership i s ' n o t so acute, and where most cu l t i vab le land i s under very small family farms, there may hardly be a surplus after self-consumption by the f am i l i e s . Price incentives would be inappropriate under these circumstances. Since farmers would employ .most of the i r land to produce food grains for the i r own consumption, the pr ice e l a s t i c i t y of acreage devoted to these crops 46 18 would be low. Price incentives, which are based on the assumption that, within the given i n s t i t u t i ona l and technological circumstances, there ex ists enough excess capacity for production, are thus l i k e l y to meet with f a i l u r e . Therefore, i t i s important in the designing of land reform that an e f fect i ve way be found to increase marketable surplus. The objectives of land.reform discussed above are not % independent of each other. For example, the achievement of a more equitable d i s t r i bu t i on of income may, under certain circumstances, require that more employment opportunities are generated as a precondition. In some cases these objectives may be in c o n f l i c t with each other. For example, certain.means of increasing the level of product iv i ty may m i l i t a t e against the objective of achieving greater levels of employment. In th i s way, a programme of land reform becomes involved in solving a much more complex agrarian problem than would be evident by merely looking at the l i s t of major object ives. Def in i t ion of land reform: Although the necessity for implementing a land reform programme to achieve objectives such as those l i s t e d above has been stressed by numerous scholars and administrators in many underdeveloped countr ies, they have often f a i l e d to agree on what exactly should const itute a On the basis of data covering 1948-63 which were col lected from 9 d i f fe rent d i s t r i c t s of Bangladesh, Hussain has worked out the pr ice e l a s t i c i t y of r i ce acreage in Bangladesh to be only .05. See S.M. Hussain, "A Note on Farmer Response to Price in East Pak i s tan ' , Pakistan Development Review, Vo l . IV, No. 1, Spring 1964. 47 land reform programme. The fol lowing statement made by Lehmann i s worth noting in th i s context: "When doctrines command widespread agreement, the time has usually come for a re-examination; when men who otherwise disagree on fundamental p o l i t i c a l values agree on an issue of importance, they are probably using cruc ia l terms in widely d i f f e r i ng senses; ..."19 The problem of supplying a widely acceptable de f i n i t i on of land reform i s not rooted only in such subjective d i f ferences, but also i s the resu l t of wide var iat ions in agrarian s i tuat ions and problems, from country to country and region to region. It i s thus not very useful to l i m i t the meaning of the term to : " . . . red i s t r ibut ion of property or r ights in land for the benefit of small farmers and 20 agr icu l tura l labourers". While such a de f i n i t i on may su i t the purpose of a pa r t i cu la r country in i t s given s o c i a l , economic and p o l i t i c a l circumstances at a point in time, i t would undoubtedly narrow down the scope of the term rather d r a s t i c a l l y . On the other hand, to produce a de f i n i t i on of land reform which would cover most of the possible s i tuat ions would be to d i l u te the essence of the term 21 to an extent which may render i t v i r t u a l l y useless. 19 See the 1 Introduction 1 by Lehmann in D. Lehmann (ed.), Agrarian Reform and Agrarian Reformism: Studies of Peru, Chile, China and India, London: Faber and Faber, 1974, p. 13. 20 D. Warriner, Land Reform in Principle and Practice, Oxford: Clarendon Press, 1969, p. x i v . 21 The integral de f i n i t i on used by the United Nations (see United Nations, Progress in Land Reform, New York, 1954, pp. IX-XI) i s one such de f i n i t i o n . It suggests the use of the term to cover a l l changes in agr i cu l tura l or rural i n s t i t u t i on s that can accelerate economic, s o c i a l , and p o l i t i c a l development. 48 Hirsch has pointed out that one needs to understand the spec i f i c character i s t i c s of the pre-reform agrarian s i tuat ion of a country quite int imately before being able to come up with a su itable 22 de f i n i t i on of land reform for that country. Information on the fol lowing aspects, at a minimum, are necessary: 1. Nature of development of the country concerned. Some of the relevant questions are: Is the country pr imar i ly agr i cu l tura l ? What i s the per capita real income of the rural population and of the whole population? What i s the level of commercialization of agr iculture? 2. Exist ing patterns of ownership of land and arrangements for c u l t i v a t i o n . One may ask the fol lowing questions in th i s regard: What i s the nature of ownership d i s t r i bu t i on of land among the r u r a l households? Is the system based on wage-labour c u l t i v a t i o n , sharecropping, owner-cult ivat ion, or communal cu l t i vat ion? 3. Nature of the most pressing problems in the man-land re lat ionship of the country. Some important questions are: How important i s land in the s o c i a l , economic and p o l i t i c a l l i ve s of the people? Does the land ownership pattern represent an unequal d i s t r i bu t i on of G.P. Hirsch, 'Some Fundamentals of Land Reform', Oxford Agrarian Studies, Vol. I, No. 2 (New Ser ies ) , 1972, pp. 136-137. 49 property ownership and income-earning opportunities even a f te r income from other sources i s taken into account? Is the size-tenure structure creating obstacles to ag r i cu l tu ra l e f f i c i ency in terms of higher y ie lds per unit of land and labour? Is the structure preventing the adoption of an improved and viable technology? Answers to these questions are expected to vary s i g n i f i c an t l y from one country to another and also from one region to another. Therefore, one i s well advised to wait t i l l the questions posed above (and other related questions) are s a t i s f a c t o r i l y answered with regard to the country or region one intends to study before developing a de f i n i t i on of land reform which can prove useful in the given context. Models of land reform: Depending upon the nature of the pre-reform agrarian society and the v i s ion of the needed change in th i s soc iety, i t i s possible to conceive a number of 'models' of land reform. A b r ie f discussion of four such models i s undertaken below. Although not representative of a l l possible kinds of reform, these models are often used as central guidelines for reforms, ( i ) Red i s t r i bu t i v i s t reform. This kind of a reform is commonly suggested in areas where the ownership d i s t r i bu t i on of land i s extremely skewed or where there i s a large number of very small farms and landless households who are found 50 to barely maintain subsistence from avai lable land (or other income producing sources), and who could be benefited by a programme of red i s t r ibut ion of land from the r e l a t i v e l y large land owners. Thus, the r e d i s t r i b u t i v i s t reform i s based pr imar i ly on immediate equity considerations. Proponents of th i s kind of reform have also argued, however, that t he i r programmes would be able to achieve both greater equity and increased e f f i c iency in ag r i cu l tu ra l production since the 23 smaller farms would tend to u t i l i z e land more intens ive ly . The i n i t i a l requirement of the reform i s to impose, through l e g i s l a t i o n , a c e i l i n g on land s u f f i c i e n t l y low to provide a ' surp lus ' of land which can be d i s t r ibuted among the small land-owning households and the landless households to raise the i r farm sizes to ' des i rab le ' l eve l s . The question that immediately comes to mind i s what these ' des i rab le ' and 'maximum' farm sizes should be. The answer would depend on whether the goal i s to a l l e v i a te only the burden of absolute poverty suffered by the poorest without worrying about the question of re l a t i ve poverty among various groups, or to substant ia l l y reduce ex i s t ing intergroup d i spa r i t i e s in land ownership. The d i s t r i b u t i v i s t reform generally suggests that the dispossessed land owners be given compensations, which are to be u lt imately borne by the recipients of land. A programme of input-subs idizat ion coupled with extension work i s suggested to ensure the economic v i a b i l i t y See Hossain (1974),op.oit., p. 493, and Khan (1972), op.cit., pp. 133-134. 51 of the newly created small farms. The essence of the d i s t r i b u t i v i s t reform can be summarized in the fol lowing way: Imposition of a low ce i l i n g on land -red i s t r ibut ion of ' surp lus ' land among the small farmers and the landless - . family farm operations with subsidized inputs and extension services for a minimum period subsequent to the reform. ( i i ) Reform aimed at cooperative farming. Advocates of reform based on cooperative (as d i s t i n c t from c o l l e c t i v i s t ) farming generally accept the v i s ion of an agrarian society based on indiv idual family farming. They maintain, however, that cooperation among the farm households i s necessary for i n f ra s t ructura l developments, technological s h i f t s and to reap economies of scale in certain aspects of production and marketing of output. They have suggested that indiv idual farms should j o i n in cooperative soc iet ies which could encourage or require members to save and pool some of the i r income out of which members can, when appropriate, seek loans from the cooperatives. These loans could then be used to invest in buying qua l i t y inputs which would be supplied through the cooperatives. The savings could also finance some overhead and organizational f a c i l i t i e s . The argument i s that when the small savings made by the ind iv idual members are pooled together, a large sum can be gathered which, when used for extending loans and financing overheads, can take care of the basic credit requirements of agr icu l tura l development. It i s hoped that the process would be self-perpetuating in that the i n i t i a l 52 operation, i f successful, is l i k e l y to increase the incentive to save and invest and, therefore, to raise product iv i ty , marketable surpluses and savings potent ia l . Though i n i t i a l success may depend upon substantial government subsidies to provide members with su f f i c i en t i n i t i a l credit and extension serv ices, i t i s argued that the cooperatives would o f fe r the government an e f fec t i ve medium through which to help smaller farmers without wasting much time or resource in bureaucratic processes, and that the cooperatives would soon become se l f - f i nanc ing , having l i t t l e need for publ ic subsidy. An important consideration with regard to the a b i l i t y of the cooperatives to achieve better l i v i n g conditions for a l l i s whether or not a programme of land red i s t r ibut ion has to precede cooperat iv i -zat ion. It may be found that membership in the cooperatives w i l l be d i r e c t l y related to the equal i ty achieved in ownership of land. In an inega l i ta r ian s i tuat ion i t would be d i f f i c u l t to develop community interests in development projects. . The r e a l i z a t i o n of th i s factor has influenced some of the proponents of cooperative farming to suggest a land red i s t r ibut ion programme before cooperativ izat ion of the anc i l l a r y functions i s achieved, ( i i i ) Reform aimed at state farming. In a state farming system the ownership r ights of land are vested in the s tate, and the cu l t i va t i on of land i s carr ied out with wage-labour managed by sa lar ied personnel. The purpose is to create for government greater f l e x i b i l i t y and control over production choices 53 ( involving such factors as technology and cropping patterns), marketing decis ions, and also d i s t r i bu t i on of income among rural households. The argument for state farming can be made more strongly in certa in contexts than in others. If the government feels that addit ional foreign exchange earnings are c r i t i c a l for the nat ion ' s further development, i t may f i nd entrepreneurial responses in producing exportable items to be inadequate under ex i s t ing arrangements. If the government feels i t has the bureaucratic and technical expertise to launch a comprehensive programme of publ ic involvement in agr i cu l ture , then, under the stated circumstances, i t may consider the case of state farming to be a strong one. When i t comes to the question of d i s t r i bu t i on of income in an equitable manner, the c r i t i c a l considerations become the effectiveness of the government machineries of monetary and f i s c a l po l i c ie s and how extensive a pol icy of income control w i l l be imposed throughout the society. It i s understood that conversion from a family farming system (or any other system) to a state farming system cannot be achieved overnight. It w i l l involve taking over of the management of farming, t r a d i t i o n a l l y in the hands of indiv idual households, by the state. If i t i s to be achieved without undue coercion, th i s w i l l require that these households get 'adequate' compensation. This i n i t i a l cost has to be supplemented with various other costs of reorganizing agr iculture through bureaucratic procedures. Therefore, the government has to be on a strong f inanc ia l 54 as well as bureaucratic foot ing. The cost-benefit analysis has to be carr ied out very ca re fu l l y since agr icu l ture i s too large an enterprise to permit the government to subsidize a continued loss in i t s operations. ( iv) Reform aimed at group ( c o l l e c t i v i s t ) farming. In th i s kind of reform the emphasis i s on the attempt to equalize income-earning opportunities through the achievement of equal access to j o i n t l y owned land f o r . a l l households. In a group farming unit the avai lable land and capita l assets are owned c o l l e c t i v e l y and cu l t i va t i on i s also a j o i n t re spons ib i l i t y . Income accruing to an indiv idual member generally depends upon the 'work points ' he accumu-lates . The decisions on production, income d i s t r i bu t i on and investment are a l l made c o l l e c t i v e l y or by representative committees of the membership. The usual strategy suggested for the transformation of a family farming system into a group farming.system i s a gradual one. The i n i t i a l phase involves the removal of any large skews in the d i s t r ibu t ion of land ownership. This c a l l s for c e i l i n g l eg i s l a t i ons and the i r implementation. The next phase i s to encourage group ownership and cu l t i va t i on among sections of the farm households whose decisions can be more eas i l y influenced through moral suasion and material incentives. It becomes very important to demonstrate e f f i c i e n t operation of these farms i f other sections of the rura l households are to be induced to j o i n the group farming system. This 55 can be associated with heavy government subsidizat ion of the f i r s t ' p i l o t ' group farms. Such d i rect state supports have to be withdrawn t a c t f u l l y i f the group farms are to be s e l f - p rope l l i n g , and th i s const itutes a de l icate problem for the i n i t i a t i o n and v i a b i l i t y of the system. It i s often suggested that the group farming system would also increase long-term productive e f f i c i ency in agr icu l ture. The arguments are much the same as extended by the proponents of cooperative farming. However, i t i s emphasized here that to ta l group commitment to the process of ag r i cu l tu ra l development requires a convergence of group interests which can be best attained by group ownership and group cu l t i v a t i on . The scenario of land reform aimed at group farming can be summarized in the fol lowing way: Imposition of a low c e i l i n g on land -red i s t r ibut ion of ' surp lus ' land among small farmers and the landless -gradual conversion of ownership r ights into usufructuary r ights conducive to farming by group ventures. P o l i t i c a l factors af fect ing the i n i t i a t i o n and implementation of reform: The economic promise of a land reform programme usually cannot in i t s e l f create enough incentive among the ru l ing e l i t e to i n i t i a t e a land reform. The e l i t e w i l l always weigh the p o l i t i c a l costs and benefits before embarking on them. A fu l l - f l edged land reform w i l l , in a l l l i ke l i hood , create favourable responses among some groups 56 within the rural society, and unfavourable reactions among others. If the e l i t e in power feels that the p o l i t i c a l benefits of making a l l i e s with the favoured groups are outweighed by the costs of a l ienat ing the dispossessed then i t w i l l not i n i t i a t e or support the programme. Insofar as land reform can benefit a majority of the impoverished rural households, i t may be able to earn new legitimacy for a power-el ite. It w i l l obviously be a valuable p o l i t i c a l asset with which to earn greater confidence of the peasantry at large. However, in s i tuat ions where the r e l a t i v e l y large land owners are represented heavily in the p o l i t i c a l power structure, one cannot hope that such legitimacy w i l l be sought from the peasantry. But, in circumstances where a ru l ing e l i t e (for example, an urban-based or military e l i t e ) i s expected to benefit by :cur ta i l ing the p o l i t i c a l power of the r e l a t i v e l y large land, owners, a quick decision might well be forthcoming to seek such popular legit imacy. For example, i f a manufacturing sector i s expected to develop very rap id l y , the ru l ing e l i t e may f ind i t necessary to strengthen i t s indust r ia l power base, which may, in certain circumstances, be possible only through the curtailment of some of the powers of the landed c lass. The resu l t might be a land reform which would favourably af fect the low-income groups within the rural community. The nature of the e l i t e in power w i l l determine not only whether a land reform programme w i l l be i n i t a t e d , but also w i l l d ictate the content of the programme that may be undertaken. In th i s context i t 57 i s useful to fol low the c l a s s i f i c a t i o n of e l i t e s suggested by 24 Ta i . He i den t i f i e s the fol lowing two categories of e l i t e s : ( i ) e l i t e s who are separated from the landed c la s s , ( i i ) e l i t e s who are cooperative with the landed class. Which of the above two kinds of e l i t e s i s in power can be expected to be related to the nature of the agrarian structure and to other h i s t o r i c a l circumstances. However, the obvious point here is that any land reform programme that ser iously aims to cu r t a i l the pr iv i leged posit ion of the r e l a t i v e l y land-r ich households cannot be expected to come from those e l i t e s who are cooperative with the landed c lass. Tai goes on to subdivide the cooperative e l i t e s into dominant y and conc i l i a to ry e l i t e s , the former with dominant representation of non-land in te res t s , and the l a t t e r with greater representation of landed interests . Among these two sub-groups of e l i t e s , the dominant e l i t e s can be expected to implement programmes of land reform which are r e l a t i v e l y more benef ic ia l to the poor. However, as a precondition for any meaningful land reform, i t i s often necessary that the potential benef ic iar ies of the reform be organized to provide a p o l i t i c a l base from which systematic pressure for change can be exerted. In other words, the strength of peasant mobi l izat ion by organized p o l i t i c a l H-C. Ta i , Land Reform and Politics, Berkeley: Univers ity of Ca l i f o rn i a Press, 1974, pp. 90-93. 58 parties plays an important role in formulating and implementing land reforms. Plan of Study With the above conceptual framework in mind, the purpose of the present study i s to focus upon the nature of problems which emerge from the size-tenure structure in Bangladesh agr i cu l tu re , and to investigate prospects of a l ternat ive land reform programmes in solving 25 these problems. The next chapter (Chapter I I I) w i l l f i r s t present a h i s t o r i c a l review of the man-land re lat ionsh ip in Bangladesh. Concurrently with an economic analysis of the past, an account w i l l be presented of the main p o l i t i c a l developments which have affected the man-land re lat ionsh ip in the country over time. This w i l l be fol lowed, in the same chapter, by an elaboration of the d i f fe rent aspects of the size-tenure structure in Bangladesh agr icu l ture. Regional var iat ions within the country are also to be pointed out in th i s context. Chapter IV w i l l further invest igate the role of the size-tenure arrangements in determining the problem of pers istent poverty and inequal i ty in Bangladesh agr icu l ture. In discussing the nature of some of these problems, necessary theoret ica l expositions w i l l be attempted pr ior to making relevant empirical invest igat ions. The purpose of the present study has been discussed in greater deta i l in Chapter I, pp. 4-5. 59 Chapter V w i l l concern i t s e l f with the context of land reform for Bangladesh and w i l l s tart with a discussion of the various objectives which may be set for the reform. The economic prospects of the various land reform models (developed in Chapter II) w i l l be analysed in reference to the context, of Bangladesh agr icu l ture. This chapter w i l l also attempt to set out the p o l i t i c a l factors conditioning the reform. The previous experiences of Bangladesh and also those of certain other countries with regard to the various 'models' of land reform w i l l be noted to help elucidate the case for a l ternat ive land reform programmes in Bangladesh. Chapter VI w i l l conclude the thesis with a summary of f ind ings, followed by a consideration of the pol icy implications ar i s ing from the study. 60 II I. ORGANIZATION OF LAND IN BANGLADESH AGRICULTURE: PAST AND PRESENT H i s to r i ca l Review Recorded accounts of the history of land tenure in Bengal (part of which i s now Bangladesh) show that as fa r back as the times of the early Hindu period, which existed about three thousand years ago, the ownership of land was in the hands of indiv iduals rather than communities J However, i t was not uncommon for the v i l l a ge r s to group together in communities to defend the i r t e r r i t o r i e s from outside threats. These communities operated by exacting a share of the produce from indiv idual cu l t i va to r s . The headman or the ch ie f ta in in the v i l l a ge community at times became extremely powerful and attained the posit ion of a v i r t ua l ru ler of the land of a number of neighbouring v i l l a ge s . When Bengal came under the rule of the Mughals during the sixteenth century, the local ch iefta ins were allowed to keep on cont ro l l ing the areas under the i r influence on condition that they would co l l e c t revenue from the land and forward i t to the State. This was a convenient arrangement for the Mughals who did not have adequate See Government of Bengal , Report of the Land Revenue Commission, Bengal, Vol. I, A l ipore: Bengal Government Press, 1940, p. 7. 61 access to the i n t e r i o r of the country during the early days of t he i r ru le . Thus, the intermediate 'government1 of the zamindavs (landlords) f lour ished during Mughal ru le . In addit ion to revenue c o l l e c t i o n , the task of maintenance and extension of the basic inf rastructure of the v i l l ages was bestowed upon them. They maintained certa in armed agencies of enforcement (for example, paiks and bavkandazes) which were essential in preserving the authority they enjoyed. The i n s t i t u t i on of zamindavi f lour i shed further during the B r i t i s h ru le. The zamindavs under B r i t i s h administration were given proprietary r ights to the land, and the revenue which was to be paid to the State by them were f ixed in perpetuity through the enactment of the Permanent Settlement in 1.793. It was the hope of the B r i t i s h rulers that , under such arrangements, the zamindavs would not only help in the co l l ec t i on of land revenue, but would also make substantial investments in agr icu l ture. The argument was that since rents demanded by the State were f i x ed , and not proportional to the to ta l ag r i cu l tu ra l produce of the land, the zamindavs would f i nd i t p ro f i tab le to invest in agr iculture so that the rent charged by them from the vaiyats (cu l t ivators ) could be increased, which would mean an equivalent increase in the incomes of the zamindavs. In the words of Lord Cornwall i s , under whose auspices the Permanent Settlement came into 62 being, the Settlement would be j u s t i f i e d because, " . . . the large cap i ta l s possessed by many of the natives in Calcutta which are now employed in usury or monopolizing sa l t and other necessaries of l i f e w i l l be appropriated to the more useful purposes of purchasing and improving land."2 The optimism conveyed by the above statement was further stressed by Cornwallis on the eve of the proclamation of the Permanent Settlement, " . . . there i s every ground to expect that the large cap i ta l s possessed by the natives ... w i l l be applied to the purchase of landed property as soon as the tenure i s declared to be secure, and they are capable of estimating what p ro f i t they w i l l be certa in of deriving from i t , by the public tax upon i t being unalterably f i xed. "3 Although the Permanent Settlement was successful in at t ract ing large volumes of cap i ta l in the purchase of land, very l i t t l e of i t was spent on ' improving ' land, as Cornwallis had hoped. As a resu l t 4 the product iv i ty of agr iculture remained extremely low. It would be in s t ruct i ve to know why the zamindars in Bengal f a i l e d to act in the same way as did the i r counterparts, the landlords in England who demonstrated such good husbandry that Cornwallis f e l t i t worthwhile to rep l i cate the arrangement in India. c Quoted in R. Guha, A Rule of Property for Bengal: An Essay on the Idea of Permanent Settlement, Par i s : Monton & Co., 1968, p. 172. 3 Ibid., p. 172. ^ See M.M. Islam, Bengal Agriculture, 1920-1946: A Quantitative Study, Cambridge: Cambridge Univers ity Press, 1978, p. 185. 63 The performance of the landlords in England can be explained to a considerable extent by the process of urbanization set in motion by the Industrial Revolution which had the ef fect of ra i s ing the prices of agr icu l tura l produce. Under such circumstances "the landlords ' gross return on the i r investment was between 15 to 20 per cent as against a return of only 5 or 6 per cent from land 5 purchase". The Industr ial Revolution also had an impact on the labour market which favoured investments by landlords in land improvement. Because of the increasing opportunities in the indust r ia l sector, the landlords found i t imperative to make these investments i f they were to a t t ract s u f f i c i en t numbers of tenants to cu l t i va te t he i r land. The s i tuat ion in Bengal was very d i f fe rent from that in England. Because of the existence of a.high demand for land in the land-rent market (brought about by the increasing man-land r a t i o in agr iculture) the zamindavs found i t exceedingly prof i tab le to invest in the purchase of land. There are several reasons why they did not perceive as large a p ro f i t in investing capita l for agr icu l tura l production. In the event of taking up agr iculture as a business enterprise in the t r ad i t i ona l sense, the zamindavs would have to gear the i r production for meeting foreign demands, since the domestic market was l imi ted in ^ Ibid., p. 186. 64 s i ze. The associated uncertainties of demands for the i r produce would act as one of the deterrents to making such investments by the zamindavs. Also, there would be uncertainties in the levels of competition to be faced from other countries producing s im i la r agr icu l tura l commodities for the purpose of exporting. Since the export trade in Bengal and elsewhere in India was monopolized by foreigners, the zamindavs would.find addit ional reasons to expect a less than adequate share of the proceeds from export of the i r produce. On top of a l l t h i s , the uncertainties of natural factors inf luencing agr icu l tura l product iv i ty in Bengal; for example, the -timing of the onset and the duration of monsoons, as well as the lengthy gestation gap involved in obtaining ' s a t i s f a c to r y ' returns from investments in agr icu l tura l production would further prompt the zamindavs to be less enthusiast ic about making these investments. Thus, they were found to concentrate on the purchase of land for almost the exclusive purpose of renting i t out. One important aspect of the zamindavi system in India was that the re spons ib i l i t y of revenue co l l ec t i on was often passed on to intermediate agencies. These intermediaries were to co l l e c t revenues from the land of the zamindavs and extend to them (the zamindavs) a previously f ixed amount of revenue. These new tenure-holders, who were ca l led talukdavs, made sure that they co l lected enough revenue The l imi ted s ize of the domestic market was an inev i tab le resu l t of the B r i t i s h pol icy of turning India into an economy based pr imar i ly on the export of primary products. 65 from the raiyats to be l e f t with a 'proper ' income a f te r paying the amounts due to the zamindars. The talukdars themselves were frequently responsible for the creation of further intermediate tenures. Thus, they handed over part of the i r re spons ib i l i t y of revenue co l l ec t i on to patni-talukdars. The patni-talukdars, in the i r turn, would create further intermediaries known as dar-patni talukdars. The process of subinfeudation would thus go on. Huq reports that in 1911 there were as many as 20 grades of subinfeudation in the d i s t r i c t of Bakerganj (which is now divided into the d i s t r i c t s of Barisal and PatuakhaTi).^ The subinfeudation of Bengal agr icu l ture has often been blamed for the poor investments made in the agr icu l tura l production process, and the subsequent low product iv i ty of land and labour. For example, the Land Revenue Commission of.Bengal reported that, "The system has severed the connection between the zamindars and ra iyats in estates where subinfeudation ex i s t s , and has defeated the intent ion of Lord Cornwallis to establ i sh a landlord and'tenant system in Bengal on the English model. It has prevented the zamindars from f u l f i l l i n g the functions which provide the economic j u s t i f i c a t i o n for a landlord and tenant system, because with few exceptions the tenure-holders immediately above the ra iyats have neither the incentive nor the capita l to ef fect agr icu l tura l improvement."8 M. Huq, Final Report on the Revisional Survey and Settlement Operations in'the district of Bakerganj, 1940-42 and 1945-54, Dacca: Government of East Pakistan, East Pakistan Government Press, 1957, p. 53; (Huq's f igure obtained from Abdullah, op:cit., p. 69). Government of Bengal, op.cit., p. 34. 66 The problem of subinfeudation was not, however, evenly d i s t r ibuted throughout Bengal. For example, whereas only 34% of the to ta l land under the zamindavis of Faridpur were under the d i rect control of the zamindavs, the corresponding f igures for 9 Dacca and Mymensingh were 64% and 66% respect ively. This shows that there were large areas of Bengal which were free of the problems of subinfeudation. However, these areas were not characterized by any appreciable amount of capi ta l investment for ra i s ing ag r i cu l tu ra l p roduct iv i ty .e i ther . Therefore, one cannot be convinced that the root cause of the problem was the sub in feudat ion.^ As argued e a r l i e r , the explanation has to be sought by invest igat ing whether the expected rate of.return:from investing in enhancing agr icu l tura l product iv i ty by the zamindavs was greater or smaller than what could be obtained by purchasing, and then renting out, land. It has been hypothesized here that the p ro f i t s from the l a t t e r were perceived to be higher by the zamindavs, and hence the low investment in ra i s ing p r o d u c t i v i t y . ^ Islam, op.ait.,p. 192. It i s to be noted here that the tenure-holders were to pay a fixed part of the i r revenue earnings to t he i r superiors so that the nature of the i r incentives were s imi la r to that of the zamindavs. One might argue that the very existence of intermediary tenure categories would seem to contradict th i s hypothesis that the zamindavs perceived a high rate of return from renting out land and were interested in enjoying such returns. However, one must not f a i l to see the poss ib i1 i ty that the zamindavs might have properties scattered over d i f fe rent d i s t r i c t s which necessitated the creation of intermediaries. Besides, those zamindavs busy with d i f fe rent urban occupations would also be expected to re ly on such intermediaries. 67 Whether numerous intermediaries were involved in the co l l ec t i on of land revenue or not, the burden of revenue on the vaiyat was usually a heavy one. Exactions of the Permanent Settlement have been described in the fol lowing words by Wheeler, "The truth cannot be doubted that the poor and industrious tenant i s taxed by his Zamindar, or Col lector for every extravagance that avar ice, ambition, pr ide, vanity or intemperance may lead him in to , over and above what i s generally deemed the established rent of his lands. If he i s to be married, a ch i l d born, honours conferred, luxury indulged, and sufferances and f ines exacted even for his own misconduct, a l l must be paid by the ryot. And.what heightens the d i s t re s s fu l scene, the more opulent, who can better obtain redress for imposit ion, escape, while the weak are obliged to s ubm i t . " ^ When the raiyats started showing signs of serious unrest over the exercise of arb i t rary power by the zamindars, the government f e l t that i t was necessary to take certa in steps to appease them. Consequently, the Rent Acts of 1859 and 1869 and the Tenancy Act of 1885 were put through. These Acts were directed toward spe l l ing out Quoted in B.N. Datta,. Dialectics of Land Economics of India, Calcutta: Mohendra Publishing Committee, 1952, pp. 127-28, and the source mentioned i s Wheeler, Early Records of British India, p. 373. 68 the r ights of d i f ferent categories of tenure. Three classes of 13 raiyats were defined: 1. Raiyats at f ixed rent: The raiyats who could prove that from the time of the Permanent Settlement they themselves or the i r ancestors were holding the land at the same rent belonged to th i s category. They were v i r t ua l owners of the land they operated upon. Their interests in land were transferable and her i tab le. Also they could mortgage and sublease t he i r land. In short, the raiyats could use t he i r land in any way they pleased so long as the revenue was paid in due time. If revenue payments were not made in time, the raiyats at f ixed rent could not be ejected outr ight. However, part of the i r holdings could be sold.to gather unpaid revenue i f decreed by court. 2. Occupancy raiyats: The land occupied by these raiyats was her itable but not transferable. The Rent Acts of 1859 and 1869 had incorporated provisions which allowed transfer of land by occupancy raiyats under two circumstances - a) in cases where the raiyat obtained permission from his zamindar to make the Much of the information on these classes of raiyats has been obtained from Abdullah, op.cit. 69 transfer, b) where local custom authorized such transfers. The occupancy raiyats were not permitted to use the i r land in any way they pleased. Even af ter the Rent Acts and the Tenancy Act they could be evicted i f they did not use the land under them in the way that was prescribed by the State. With regard to payments of revenues on time, they shared the same r ights with the raiyats.at f ixed rent. 3. Non-occupancy raiyats: These raiyats were denied the r ights of transfer or inheritance of t he i r land unless these were allowed by custom. Unlike the f i r s t two categories of raiyats, the non-occupancy raiyats could be ejected for non-payment of revenues in time. The rent paid by them could be increased and depended on market forces. Thus, while the f i r s t two categories of raiyats were given, by the law of the book, some protect ion, the non-occupancy raiyats were s t i l l insecure a f te r the Rent Acts and the Tenancy Act. The raiyati area in the part of Bengal that now forms Bangladesh had a predominant number of occupancy raiyats. Bel l reports that in the d i s t r i c t of Dinajpur 14 85% of the area under raiyati interest was held by occupancy raiyats. F.O. B e l l , Final Report on the Survey and Settlement Operations in the District of Dinajpur, 1934-40, A l ipore: Bengal Government Press, 1942, p. 86; (th is f igure provided by Bel l has been obtained from Abdullah, op.cit., p. 78). 70 The Tenancy Acts had provisions for increasing the rents payable by these vaiyats to the revenue co l l ec to r under the fol lowing circumstances: (1) when the rate i s below the prevai l ing rate in the v i l l a ge or the neighbouring v i l l a g e , (2) when the average pr ice of food crops r i s e s , (3) when the productive qua l i ty of land is improved by the action of the landlords, (4) when the productive e f f i c i ency of land i s improved by f l u v i a l act ion. It can be eas i l y understood from the above that the Acts provided the zamindavs and the intermediary tenure-holders with s u f f i c i en t opportunity to increase rent at rates which were l i k e l y to be a rb i t ra ry by the very nature of the grounds on which.the increase was permitted by law. Islam has provided some data ( o r i g i na l l y from the reports on Land Revenue Administration of the Bengal Presidency) which show the increases in the legal rents paid by the vaiyats over a stretch of 15 25 years. These data are presented in Table 1. Islam, op.oit., p. 190. 71 Table 1. Gross Rental Paid by Raiyats During Different Periods Year Total Legal Rental ( in thousand Rupees) Index 1914/1.5 - 1918/19 122,388 100 1919/20 - 1923/24 139,556 113 1924/25 - 1928/29 148,842 120 1929/30 - 1933/34 161,466 131 1934/35 - 1938/39 169,996 138 Source: M.M. Islam, Bengal Agriculture, 1920-1946: A Quantitative Study, Cambridge: Cambridge Univers ity Press, 1978, p. 190. (Or ig ina l l y from Government of Bengal, Report on Land Revenue Administration of the Bengal Presidency. ) As Table 1 shows, the increase in the legal rent over the period of time under consideration (1914/15 - 1938/39) was nearly 40%. It i s to be noted here that the tota l rentals reported in Table 1 do not 16 include the abwabs ( i l l e g a l cesses). That the zamindars were not Mukerjee reports that the incidence of abwdb. in certa in instances was as high as 120% of the legal raiyati r en ta l . See R. Mukerjee, Land Problems of India, London: Longmans, Green & Co., 1933, p. 133. 72 always respectful of the l e g a l i t i e s of the r ights of the raiyats has been well expressed by Abdullah, "The zemindars undoubtedly got away with a great deal that was i l l e g a l . The paiks and barkandazes of the zemindars maintained a law and order in the v i l l ages that did not always coincide with the noble ideals of benevolent l eg i s l a to r s . A raiyat could be beaten up, his standing crop d i s t ra ined, and sometimes his house set on f i r e , and i f a su i t ensued, pol ice o f f i ce r s as well as witnesses could be bribed or ( in the case of witnesses) bu l l i ed to ensure a favourable verd ict . This was a l l the more probable as lawyers and judges were c lose ly connected, by b i r t h , marriage or purchase, with the landed a r i s t o c r a c y . 7 It was increasingly being f e l t that the Tenancy Act of Bengal, which was enacted in 1885, required some amendments i f the exp lo i tat ion of the raiyats were to be reduced. A committee was appointed in 1921, of which S i r John Kerr was made the Chairman, for suggesting appropriate amendments to the Act. Among other things, the Committee suggested that occupancy r ights should be extended to cover sharecroppers. However, in the face of strong opposition from the zamindars and other landed interest groups, the amendments were not o f f i c i a l l y adopted. When the Tenancy Act of 1885 was f i n a l l y amended in 1928, nothing more than the f i x a t i on of the transfer fee (payable by the raiyat i f he decided to sublet his land) was achieved which could be considered 18 important. The sharecroppers remained, as before, without any legal status. The d i f fe rent Tenancy and Rent Acts, and the subsequent Abdullah, op.ait.3 pp. 78-79. The transfer fee was set at 20% of the sale price of land, or f i ve times the rent. 73 amendments to these Acts, were designed in the s p i r i t of a compromise between the zamindavs and the other intermediate tenure-holders on the one hand, and the vaiyats on the other. Thus, d i f fe rent kinds of secur i t ies were extended to the vaiyats only so far as these were not contrary to the basic interests of the landed class. It was not un t i l 1932 that any serious proposal for the abo l i t ion of the zamindavi system was made. This proposal came from the Bengal Provincia l Praja Samity (which l a te r came to be known as the Kvishak Pvaja Party). Under the leadership of A.K. Fazlul Huq, the Party pushed for a replacement of the Permanent Settlement by "a more equitable system and laws su itable to the needs and requirements 19 of the people". In 1938 the Government of Bengal appointed a Land Revenue Commission to investigate the problems of land tenure as created by the operation of the Permanent Settlement. The Commission presented i t s report in 1940 which has been popularly known as the Floud Commission Report a f te r the name of the Chairman of,the Commission, S i r Francis Floud. The Report recommended that the Permanent Settlement be declared nu l l and void and,that the rent-receiv ing interests of the zamindavs as intermediaries between the raiyats and the State be abolished. The fol lowing reasons were shown for the recommendation. Cited in Islam, op.ait. The source mentioned i s The Indian Annual Register (1938), Vol. I I, p. 219. 74 1. Loss of revenue: It was argued that although prices of many agr icu l tura l goods increased over the years, the government could not get an increased revenue due to the f i x i t y of revenue introduced through the Permanent Settlement. However, the raiyats were not the ones to reap the benefits of low revenue-intake by the State. The zamindars raised the revenue payable by the raiyats from time to time without having to pass i t on to government. Thus, in e f f e c t , both the raiyat and the government were being weakened through the operation of the Permanent Settlement. 2. Administrative disadvantages: According to the Floud Commission Report, "the Permanent Settlement became synonymous with a pol icy of non-interference in the zamindari estates and in consequence Government o f f i ce r s were much less in touch with the tenantry ..."20 The administration was thus l e f t without any v i l l a ge maps, record of r i ght s , knowledge of local conditions and customs; a l l of which created numerous problems. 3. Effects of subinfeudation: The Floud Commission Report re i terated the f inding of the Simon Commission in .emphasizing the extent of subinfeudation in Bengal caused by the Permanent Settlement. Government of Bengal, op.ait. 3 p. 33. 75 It maintained that the 50 or more grades of tenure-holders who were quite frequently found between the vaiyat and the zamindav reduced the chance that any tenant-landlord re lat ionsh ip might emerge which would be conducive to higher agr i cu l tura l production. Noting the high degree of subinfeudation in d i f fe rent parts of Bengal and the d i rect re lat ionsh ip between the number of i n t e r -mediaries and the burden of revenue experienced by the vaiyats, the Floud Commission reported, " I t i s not too much to say that the extent of subinfeudation has become an incubus on the working agr icu l tura l .popu lat ion, which f inds no j u s t i f i c a t i o n in the performance of any material service so far as agr i cu l tura l improvements are concerned, and f a i l s to provide any e f fec t i ve means for the development of the resources of land ..."21 The Commission also found that there was a substantial number of undev-vaiyati holdings (holdings created by leasing in land from the vaiyats) during the late 19301 s which lacked in incentives to improve agr iculture due to imposition of high rents. Table 2 has some relevant information on the vaiyati and undev-vaiyati holdings which existed during 1938-40. Ibid., p. 37. 76 Table 2. Raiyati and Under-raiyati Holdings in the Area that i s Now Bangladesh, 1938-40 Raiyats Under-raiyats Number of Holdings 9,372,000 3,108,000 Area Covered (acres) 17,926,000 2,172,000 Average Size (acres) 1.91 0.70 Source: Government of Pakistan, Economy of Pakistan, 1950, Karachi: Off ice of the Economic Advisor, Min i s t ry of Economic A f f a i r s . (Quoted in Abdullah, op.cit., p. 70.) 4. Loss of occupancy r i ght s : The Floud Commission found that the number of actual cu l t i va tor s who possessed occupancy r ights declined over the years. Between 1921 and 1931 households without occupancy r ights increased by 49%. These were the households who comprised most of the under-raiyats. The rent that they were required to pay was "excess ive" in the opinion of the Report. Some under-raiyats were cu l t i va t ing the land under the barga (sharecropping) system whereby the agr i cu l tura l produce was divided into halves between the raiyats and under-raiyats. The increase in the number of bargadars (sharecroppers) was considered to be harmful fo r the society, which i s 77 evident in the fol lowing quote from the Report, "The rapid increase in the number of bargadars i s one of the major disquiet ing features of the present times; and i t i s an ind icat ion of the extent to which the hereditary ra iyats are losing the i r status and being depressed to a lower standard of l i v ing . "22 Although the f i n a l recommendation of the Floud Commission to abolish the Permanent Settlement was made as ear ly as 1940, nothing was done in th i s regard for another decade. It was only after the B r i t i s h rule came to an end that a change was f i n a l l y made. The Bengal zemindars under B r i t i s h rule were mostly Hindus. This was because the B r i t i s h administrat ion, which was established by replacing the Muslim ru le r s , sought the support of the Hindus at large for consolidating the i r p o l i t i c a l power in an atmosphere of general antagonism among the dethroned Muslims. The resu l t of th i s a l l i ance was the economic and p o l i t i c a l domination by the Hindus over the 23 Muslims. During the nat iona l i s t movement in India the vast majority of the Muslim population of Eastern Bengal supported the c a l l by the Muslim leaders for a separate State for the Muslim minor i t ies of India. They saw in the creation of th i s new State, Pakistan, a chance to rid: themselves of economic and p o l i t i c a l domination by the Hindus. The nat iona l i s t movement in India culminated in 1947 in the Ibid., p. 38. See K.B. Sayeed, The P o l i t i c a l System of Pakistan, Boston: Houghton M i f f l i n Company, 1967, p. 10. 78 creation of two separate States, India and Pakistan, the former with a majority of Hindu population and the l a t t e r with a majority of Muslims. The s i tuat ion obtaining in East Pakistan (formerly East Bengal) immediately after the pa r t i t i on of India was not, therefore, conducive to the continual operation of the Hindu zamindars. Most of them found i t necessary to f l ee the country and 24 take refuge in India. This created a s i tuat ion where i t was r e l a t i v e l y easy for the Government of Pakistan to respond to the • public sentiment by abolishing the zamindari system in East Pakistan. This was done through the enactment of the East Pakistan State Acquis i t ion and Tenancy Act of 1950 which contained the fol lowing main provisions V. The government i s to acquire a l l rent-receiv ing interests in land and the actual t i l l e r s of the s o i l are to become d i rect ' tenants ' under the government, thus gaining the status of maliks .(proprietors). Their r ights in land would be her itable and transferable. See R. Jahan, Pakistan: Failure in National Integration, London: Oxford Univers ity Press, 1973, p. 18. For further discussion on these provisions see Government of East Pakistan, Report of the Land Revenue Commission, East Pakistan, Dacca: East Pakistan Government Press, 1969, pp. 3-4. 79 Subletting of land i s to become i l l e g a l in future. However, cu l t i va t i on under a sharecropping arrangement i s not to be considered a form of sublett ing. A c e i l i n g of 100 bighas (33.3 acres) per fami ly, or 10 bighas (3.3 acres) per member of the fami ly, whichever i s larger, i s to be put on the ownership of a l l cu l t i vab le neej-jote (owned land). A maximum of 10 bighas (3.3 acres) are to be allowed as homestead land. Lands in excess of th i s c e i l i n g are to be acquired by the government. This c e i l i n g i s to be relaxable in special uses of land (for example, in orchards and dairy farming). 'Excess ' land acquired through the imposition of the above ce i l i n g i s to be d i s t r ibuted among those cu l t i vator s who hold less than 3 acres of land. A graduated scale is to be followed in paying compensation for the acquis i t ion of rent-receiv ing interest s . The scale would range from 10 times the net annual income in cases of persons with net incomes of Rs. 500 or less to 2 times the net annual income in case of persons with net incomes of Rs. 100,000 or more. 80 6. Compensation for the acquis i t ion of excess lands would be paid at the rate of f i ve times the net annual p r o f i t from the land that i s acquired. I f one uses the terminology of Krishna then the 1950 Act was d i rected, at least on paper, towards achieving some of the " l i b e r a t i v e " 26 and " d i s t r i b u t i v e " object ives. A l l measures which abolish i n t e r -mediary r i gh t s , regulate tenancy, free labourers from feudal bonds, and formalize the r ights and duties of the part ies in tenancy or labour contracts have been c l a s s i f i ed as " l i b e r a t i v e " by Krishna.. The d i s t r i bu t i ve measures are those which involve the imposition of some ce i l i n g on landholding, the conversion of tenants and landlords into guaranteed owners, the acquis i t ion of ' surp lus ' land, and the transfer of the ' surp lus ' by g i f t or lease or sale to the landless and/or the small holders, and the consolidation of fragmented holdings. The l i be ra t i ve measures of the 1950 Act were successful to the extent that the cu l t i va tor s were freed from d i rect feudal bondage. However, the abo l i t ion of rent-receiv ing interests of the zamindavs cannot by i t s e l f suggest that the cu l t i va to r s would be mater ia l l y we l l - o f f a f te r the implementation of the measure. As the provisions of the Act show, there was no emphasis put on regulating tenancy or formaliz ing the r ights and duties of the parties involved in tenancy R. Krishna, ' land Reform and Development in South A s i a ' , in W. Froehlich (ed.), Land Tenuve, I n d u s t v i a l i z a t i o n and Social Stability, Milwaukee: The Marquette Univers ity Press, 1961. 81 or other labour contracts. Thus, avenues were l e f t open to the r e l a t i v e l y r i ch land owners to replace the abwabs ( i l l e g a l exactions) 27 of the zemindars. This inev i tab ly reduced the element of actual l i be ra t ion brought about by the Act. The d i s t r i bu t i v e measures of the Act of 1950 were not implemented for a long time. During the f i r s t decade after the creation of Pakistan, the country experienced severe p o l i t i c a l i n s t a b i l i t y . The p o l i t i c a l feud among r i v a l interest groups within the ru l ing Muslim League, which were es sent ia l l y r i v a l landlord fact ions or ig inat ing from West Pakistan, was responsible for frequent changes in the government. Thus, during 1950-58, Pakistan had as many as seven 28 Prime Min isters . In the midst of such changing circumstances in the p o l i t i c a l arena, the d i s t r i bu t i ve measures of the Act of 1950 remained -largely unimplemented. After the coup of 1958, which brought General Ayub Khan to power, the ce i l i n g provisions of the Act were revised. Many of the r i ch land owners were rest ive under the prospect of having the provisions of the 1950 Act implemented. Ayub Khan found that i t was necessary for his p o l i t i c a l survival to ra ise the ce i l i n g to such 29 a level that only the very large land owners were affected by i t . Thus, in 1961, the amount of retainable land owned by a family was raised from 100 to 375 bighas (that i s , from 33.3 acres to 125 acres). 27 Many of these r i ch land owners were intermediate tenure holders under the zamindars who occupied the land l e f t by the zamindars in the wake of the pa r t i t i on of India. See Mukherji, op.ait., p. 453. 28 Sayeed, op.ait., p. 92. 29 Jahan, op.ait. p. 57. 82 Besides, options were created whereby the land owners could make g i f t s out of t he i r land. This allowed the r i c h land owners an excel lent opportunity of d i s t r ibut ing whatever land they had in excess of the ce i l i n g to the i r fr iends and r e l a t i ve s , and s t i l l remain in e f fec t i ve control of the land. Under such circumstances, very l i t t l e ' surplus ' land was made avai lable for r ed i s t r i bu t i on . According to the figures supplied by the Commissioner of Land Reform, which were reported by Abdullah, the to ta l amount of ' surp lus ' land 30 was only 163,741 acres. This constituted less than 1% of the to ta l cu l t i vab le land of East Pakistan. There were over 350,000 landless 31 households in the province during the ear ly 1960's. I f the surplus land was to be d i s t r ibuted equally among these households then the average s ize of the i r holdings would become only .47 acre. To make the s i tuat ion even more f r u s t r a t i n g , much of the ' surp lus ' land was not 32 pa r t i cu l a r l y . su i t ab le for cu l t i v a t i on . J U Abdullah, op.ait., p. 83. 31 This f igure for Tandlessness has been calculated from the percentage of landless holdings reported by the Government of East Pakistan, Master Survey of Agriculture in East Pakistan, S ixth Round (1964-65), Dacca: East Pakistan Bureau of S t a t i s t i c s , 1966, p. 17. 32 Abdullah, op.ait., p. 83. 83 The salami (price) that was to be paid by the rec ip ients of land was f ixed mainly with a view to ra i s ing adequate compensation for the dispossessed. The Land Revenue Commission of 1959 had found that the sett ing of the salami at 50% of the market value of 33 the land was too high from the point of view of the rec ip ients . Even then the government went so far as to increase the price to the f u l l market value. Thus, only the r e l a t i v e l y we l l - o f f households among the e l i g i b l e rec ip ients of the surplus land could benefit by the r ed i s t r i bu t i on . A l l of th i s points out the inadequacy of the red i s t r i bu t i ve measures adopted by the Ayub regime. There were no other attempts at land reform un t i l a f te r the eastern wing of Pakistan emerged as an independent nation, Bangladesh, through a nat iona l i s t struggle in the year 1971. During the rule of Ayub, the c i v i l - m i l i t a r y bureaucracy which ran the country had very l i t t l e representation from East Pakistan 34 (roughly 30% in the higher c i v i l bureaucracy and 5% in the army). The Bengali counter -e l i te of East Pakistan t r i e d in vain to gain a share of power i n ' t he central decision-making apparatus of the country 35 through the elect ions of 1959 and 1964. Subsequently, in 1966 the East Pakistan-based nat iona l i s t party, the Awami League, launched a Government of East Pakistan, Report of the Land Revenue Commission, East Pakistan, Dacca: East Pakistan Government Press, 1969, pp. 45-46. R. Jahan, 'East Pakistan During the Decade of Ayub', in P.J. Bertocci (ed.), Prelude to Crisis: Bengal and Bengal. Studies in 1970, East Lansing: Asian Studies Centre, Michigan State Univers i ty, August 1972, p. 22. Ibid., p. 22. 84 programme which strongly demanded provincia l autonomy. Ayub took resort to a po l icy of t o ta l suppression of the autonomy movement. This had the resu l t of further increasing the fee l ing of discontent among the people of East Pakistan who, through a mass movement under the leadership of the Awami League, toppled the Ayub regime. General Yahya Khan, the successor of Ayub, was found to pursue the pol icy of suppression with more b ru ta l i t y than ever.. When the Awami League won a landsl ide v ic tory in the general elect ions held in December 1970, the Yahya regime responded with an attempt at an armed repression of the forces of autonomy. This turned the movement for autonomy into a na t iona l i s t struggle for the independence of East Pakistan which culminated in the b i r th of Bangladesh in December 1971. After coming to power, the Awami League government made several pronouncements with regard to.changes to.be brought about in the land tenure system of the country which raised the expectations of the land-poor masses in the agr i cu l tura l sector. The F i r s t Five Year Plan of Bangladesh, which was launched.in 1973, maintained the fo l lowing: "Radical land reform measures w i l l have to be implemented because (land) d i s t r i bu t i on and tenure systems are the fundamental factors determining rural employment and income d i s t r i bu t i on in a predominantly agr icu l tura l society."36 Government of Bangladesh, the First Five Year Plan, 1973-78, Dacca: Planning Commission, 1973, p. 89. 85 However, the actions of the government hardly matched the i r words. Although the average s ize of farm in Bangladesh agr icu l ture was only about 2.5 acres during the aftermath of the independence of the country, the F i r s t Five Year Plan re i terated i t s e a r l i e r decision to put the ce i l i n g on land ownership at 100 bighas (33.3 acres) per family which was o r i g i n a l l y incorporated in the Bangladesh Land Holding (Limitation) Order 1972 (vide Pres ident ia l Order 98 of 1972). It may be noted here that th i s c e i l i n g was the same as suggested by the Act of 1950. According to the 1968 Master Survey of Agr icu l ture, less than 5% of the farmland was held by 37 farms which had some land in excess of th i s c e i l i n g . Therefore, i t i s easy to understand that the red i s t r i bu t i ve ef fect of th i s c e i l i n g , i f implemented, would be rather l im i ted . However, the Bangladesh Land Holding (L imitat ion) Order of 1972 was d i f fe rent than the Act of 1950 in defining a family. The 1972 Order defined a family to include a "person and his w i fe , son,/ unmarried daughter, son's w i fe, son's son, and son's unmarried 38 daughter". The 1950 Act, however, had not included the married son of a person l i v i n g separately in i t s de f i n i t i on of a family. Thus, the concept of family was broadened by the 1972 Order. This was done to reduce the chances of transfer of land among members of a family Government of Bangladesh, Master Survey of Agriculture in Bangladesh, Seventh Round, Second Phase (1967-68), Dacca: Bangladesh Bureau of S t a t i s t i c s , (undated), p. 9. Mukherji, op.cit., p. 456. 86 aimed at avoiding the handover of the surplus land to the government. However, the government yielded to the pressures exerted on i t by the r e l a t i v e l y r i ch land owners to change the de f i n i t i on of a ' f am i l y ' in the i r favour. Thus, the fol lowing clause was added to the or ig ina l d e f i n i t i o n : "Provided further that an adult and married son who has been l i v i n g in a separate mess independently of his parents continuously since f i ve years before the 16th day of December, 1971 (the Independence Day of Bangladesh), and his wi fe, son and unmarried daughter shal l be deemed to const itute a separate f a m i l y . " 3 9 Subsequently, the de f i n i t i on of a ' f am i l y ' was further revised by l i f t i n g the condition that a married son l i v i n g separately had to be l i v i n g independently for f i v e years to be considered outside the family of his father. Instead, i t was mentioned in the second amendment that "the Revenue Off icer would accept the son's claim to be the head of a separate family while his father was a l i ve i f he found, on v e r i f i c a t i o n , that the son was an adult and married with independent means of l i ve l ihood since before the 20th February, 40 1972". It i s to be noted that not only was the de f i n i t i on of a ' f ami l y ' made more amenable to the evasion of the c e i l i n g , but the date for the a p p l i c a b i l i t y of the land c e i l i n g was also moved to the 20th February, 1972 from the i n i t i a l l y planned 16th December, 1971. The i n i t i a l plan of giving the ce i l i n g l eg i s l a t i on a retrospective Cited in ibid., p. 457, and the source mentioned i s Bangladesh Gazette (Extraordinary), November 4, 1972. Mukherji, op.ait., p. 457. 87 ef fect was designed to n u l l i f y the ef fects of the nominal land transfers made during the interim period by a large number of persons who antic ipated a new and lower land ce i l i n g from the government than what was in existence before independence. The above discussion indicates that only a half-hearted attempt was made at changing the pattern of land ownership in Bangladesh by the Awami League government despite i t s many pronouncements favouring drast ic changes. The reason why the Awami League regime was not interested in bringing about real i n s t i t u t i ona l reorientations was that i t had a p o l i t i c a l interest in keeping the r e l a t i v e l y large land owners pleased. Jahan reports that over 75% of the Members of Parliament in 1973 had landholdings 41 of 6.5 acres or more, and over 50% of them had 15.5 acres or more. This indicates the considerable land-richness of the Members of Parliament when one notes the fact that average farm s ize in the country was about 2.5 acres during the aftermath of independence. The Awami League leadership at the centre could not, therefore, be expected to make a move that would al ienate th i s class of r e l a t i v e l y large land owners. This largely explains why the radical ism announced in the Plan was not matched by a low land ce i l i n g l eg i s l a t i on in the Parliament. See R. Jahan, 'Members of Parliament in Bangladesh', Legislative Studies Quarterly, Summer 1976, p. 361. 88 The f a i l u r e of the Awami League Party to de l i ver an honest and e f f i c i e n t government, which could bring about the necessary changes to reduce interpersonal d i spa r i t y , combat the gal loping i n f l a t i o n , and restore law and order, resulted in widespread d i s sa t i s fac t ion among the Mukti. Bahini (freedom f ighters) leading 42 to a po lar izat ion between them and the cent r i s t Awami Leaguers. As the po lar izat ion became more acute, Sheikh Mujibur Rahman, the Prime Minister of Bangladesh and the leader of the Awami League Party, decided to abandon "the facade of parliamentary government 43 and resorted to a one-party d ictatorsh ip and t o t a l i t a r i a n con t ro l " . On January 25, 1975, the Constitution was amended in favour of a pres idential form of government. Through th i s amendment, Sheikh Mujibur Rahman was made the President of the country for f i ve years. Further, the President was empowered with authority to organize one " "National Party" and suspend the a c t i v i t i e s of a l l p o l i t i c a l groups 44 that refused to j o i n the "new" party". Thus came into existence the Bangladesh Krishak Sramik Awami League (Bangladesh Peasants, Workers and People's League, BKSL). The t o t a l i t a r i a n control that Mujib sought to exercise was to be f a c i l i t a t e d by strengthening the p o l i t i c a l See T. Maniruzzaman, 'Bangladesh in 1975: The Fa l l of the Mujib Regime and i t s Aftermath ' , Asian Survey, Vo l . XVI, No. 2, February 1976, p. 119. Ibid., p. 119. Ibid., p. 120. 89 submissive m i l i t i a - Jatiyo Rakkhi Bdhini - at the cost of the expansion of the regular armed forces. This, coupled with the d i s sa t i s f ac t i on of the army regarding the influence of India over the Mujib government, sparked off a coup on August 15, 1975, in which Mujib was k i l l e d along with his ent i re family. After suppressing a countercoup (of November 3, 1975) staged by the pro-Mujib elements, Major-General Ziaur Rahman emerged as the new 45 leader of Bangladesh. Ziaur Rahman's government, which started off as a Mart ia l Law government, has evolved into a c i v i l i a n government through the general e lect ions of February 1979, in which the newly formed party led by Ziaur Rahman (Bangladesh National Party) gained a majority of seats in the Parliament. President Ziaur Rahman (his i s a mixture of pres ident ia l and parliamentary forms of government) has since been consolidating his p o l i t i c a l power. Zia has, on several occasions, announced his government's intention to ef fect agr icultural .reforms. For example, ta lk ing about the redemption of his e lect ion promises of bettering agr icu l tura l product iv i ty and achieving d i s t r i bu t i ve j u s t i c e , he mentioned that these objectives w i l l be reached through a revolution which he defined in the fol lowing words: "Our revolution w i l l be peaceful and come in democratic process through parliament ... Our revolution means a d i s t r i bu t i on system Ibid., p. 125. 90 based on social j u s t i ce ... It w i l l establ i sh p o l i t i c s of production and soc ia l j u s t i c e . We shal l bring about agr icu l tura l reforms."46 Zia has elsewhere expressed the opinion that the rural poor would not be benefited to any appreciable extent by po l i c ie s of agr i cu l tura l development so long as they are not involved in the planning and implementation of these p o l i c i e s , and that t he i r involvement would be possible only in a s i tuat ion where the t rad i t i ona l l ines of authority are done away with. This i s evident from his speech at the FAO Conference on Agrarian Reform and Rural Development, " . . . planning from the top has in pract ice usually meant planning for the top. Poverty-focussed rural development, on the other hand, emphasises increasing the largely untapped productive capacity of the majority of the rural population. ... The present awareness that development should be pr imar i ly for the poor i s a welcome departure from the past, but i t i s incomplete in not recognizing that development planning and implementation should be increasingly by the poor themselves. Impl ic i t in th i s i s a necessary awareness and organization by the poor which could appear to represent a threat to the establishment ... Rural development involves or i s perceived to involve r ed i s t r i bu t i on , or in less po l i t e terms, the giving up of power and resources by those who now control them."47 Cited by S. Kamaluddin. See S. Kamaluddin, 'Bangladesh: Revolution and Confusion ' , Far Eastern Economic Review, Vol. 106, No. 44, November 2, 1979, p. 32. Government of Bangladesh, 'Text of Z i a ' s Speech at FAO Conference on Agrarian Reform and Rural Development', Bangladesh, Vo l . I l l , No. 2, August 1979, pp. 4-5. 91 In spite of the sentiments expressed in the above statement, nothing has so far been done to change the land ownership pattern, although i t was admitted by the government that the lowering of the c e i l i n g to 100 bighas (33.3 acres) by the Awami League government was 48 expected to a f fect only 3.2% of the net cropped area. One wonders whether the problems of s h i f t i n g "power and resources" from those who control them now to the impoverished class are unsurmountable within the context of a " revo lut ion " to be carr ied out through the workings of the Parliament. In th i s context, the observations of Haque are worth noting, " D r a s t i c ^ and reforms would adversely a f fect a l l MPs, whatever the i r party a f f i l i a t i o n s , and Zia appears poised to give them time to think i t over. Z i a ' s "Revolution" at the moment seems to be another broader and more vigorous version of the "Grow More Food" program or "Green Revolution" of e a r l i e r governments, benefit ing the large peasants more than the small farmers (or the landless who const itute a substantial element of the population)."49 Under a succession of p o l i t i c a l regimes influenced by the interests of r e l a t i v e l y large land owners, the conditions of land-poor households have worsened. This can be evidenced by the fol lowing elaboration of size-tenure structure in Bangladesh agr icu l ture. Ibid., p. 6. A. Haque, 'Bangladesh 1979: Cry for a Sovereign Parl iament ' , Asian Survey, Vol. XX, No. 2, February 1980, p. 223. 92 Elaboration of Size-Tenure Structure in Bangladesh Agr iculture 50 Bangladesh agr iculture i s characterized by the predominance of small-s ized farms. The proportion of to ta l farmland covered by the smaller farms is increasing with time. A comparison of the relevant data for 1960 and 1968, presented in Table 3, i s i nd icat i ve 51 of the trend. As Table 3 shows, the number of farms below 5 acres increased by about 5% between 1960 and 1968. Farms in a l l reported s i ze -categories above 5 acres, however, decreased in number during the same period. Population growth and the operation of the law of inheritance over the years were the primary reasons for the decl ine in the s ize of a large number of farms. 50 The size-tenure structure is discussed in th i s section at a national l e v e l . Although relevant disaggregated data are very sparse, some regional var iat ions w i l l be considered in a subsequent section in th i s Chapter. See pp. 112-135. 51 The data are from Government of Pakistan, Pakistan Census of Agriculture, 1960, Vol.. I, Final Report - East Pakistan, Part I, October 1962; and Government of Bangladesh, Master Survey of Agriculture in Bangladesh, Seventh Round, Second Phase (1967-68), Dacca: Bangladesh Bureau of S t a t i s t i c s , (undated). Both the surveys reported data on the nation as a whole. Although the samples studied for estimating the population f igures were d i s s im i l a r , s i m i l a r i t i e s in def in i t ions used and the random nature of the se lect ion of the samples make the two surveys comparable for our purpose. Table 3 Distribution of Farms Under Different Size-Categories and the Area Covered by Them in Bangladesh, 1960 and 1968 Size-Category Farms Farm Area 1960 Number of Percentage • Farms of Farms 1968 Number of Percentage Farms of Farms 1960 Area Covered Percentage (acres) of Total Area 1968 Area Covered Percentage (acres) of Total Area Less than 5 acres 5.0 - 7.49 acres 7.5 - 12.49 acres 12.5 acres or more 4,784,900 77.94 698,450 11.38 442,360 7.22 213,770 3.48 5,697,000 82.95 632,000 9.20 360,000 5.25 179,000 2.60 9,264,734 42.64. 4,192,948 19.30 4,158,797 19.14 4,109,348 18.92 11,059,000 51.29 3,831 ,000 17.77 3,347,000 15.52 3,325,000 15.42 Total: 6,139,480 100.00 6,868,000 100.00 21,725,827 100.00 21,562,000 100.00 Note: Farms include a l l land area under a farming household. Sources: 1. Government of Pakistan, Pakistan Census of Agriculture, 1960, Vol. I, Final Report - East Pakistan, Part I, October 1962. 2. Government of Bangladesh, Master Survey of Agriculture in Bangladesh, Seventh Round, Second Phase (1967-68), Dacca: Bangladesh Bureau of S t a t i s t i c s , (undated). 94 Although small farms were the most common in Bangladesh during 1960, 10.70% of a l l farms were 7.5 acres or more in s i ze. These farms could be considered large in the context of Bangladesh agr iculture where the average farm s ize stood at 3.5 acres in 1960. In 1968 about 8% of a l l farms were found to be in th i s larger s ize-category, the average farm size at that time being 3.1 acres. The area covered by these large farms continued to be substantial as can be seen in Table 3. They accounted for about 31% of to ta l farmland in 1968. Farms below 5 acres in s ize in 1968 constituted about 83% of a l l farms, the area covered by them being only about 51% of to ta l farmland. This shows the inequal i ty of landholding. Some area studies made in the seventies are avai lable which o f fe r more up-to-date, i f p a r t i a l , information on the percentage of farms and farm area under d i f fe rent s ize-categor ies. Data reported in Table 4 are from one such area study. Table 4 indicates that the inequal i ty in landholding was quite considerable in both the d i s t r i c t s of Mymensingh and Rangpur. The inequal i ty was found to be more acute in Rangpur where 57% of farms were less than 2.5 acres i n . s i z e , the area covered by these farms being only 18.6% of to ta l farm area. Twelve per cent of the farms in that d i s t r i c t were 7.5 acres or more, and the area covered by these r e l a t i v e l y large farms was 43.6% of to ta l farm area. Table 4. D i s t r ibut ion of Farms Under Different Size-Categories and the Area Covered by Them in Selected Regions of Bangladesh, 1974 Size-Category Mymensingh Rangpur Percentage of Farms Percentage of Total Farm Area Percentage of Farms Percentage of Total Farm Area Less than 1 acre 12 2.8 25 3.4 1 - 2.49 acres 30 13.7 32 15.2 2.5 - 4.99 acres 33 36.8 24 25.0 5.0 - 7.49 acres 21 36.4 7 12.8 7.5 - 9.99 acres 4 10.3 4 10.0 10.0 acres or more - - 8 33.6 Notes: ( i ) Nature of data did not allow calculat ion of relevant information f o r a l l s ize-categories reported in Table 3. ( i i ) The data cover 100 farms in each of the d i s t r i c t s of Mymensingh and Rangpur. The data, therefore, represent both absolute and percentage f igures. ( i i i ) The source of data for Table 4 covered one more d i s t r i c t , v i z . Dinajpur. However, there i s an error in the data reported for that d i s t r i c t which prevented i t s inclus ion in Table 4. The error is detected by adding up the percentage figures for areas covered under a l l s ize-categories of farms, While the percentages should add up to 100, i t i s found that these add up to only 82. Source: M.A. J&bbar, An Investigation into the Effect of Farm Structure on Resource Pro-ductivity in Selected Areas of Bangladesh. Unpublished Ph.D. Dissertat ion, Univers ity of Wales, 1976, pp. 76-78. cn 96 The overal l s i tuat ion at the national level i s avai lable from 52 the 1977 Land Occupancy Survey. This survey does not provide the kind of data that would help in comparing the relevant f igures for the s ize-categories used in the previous two tables. However, the detai led s i ze -d i s t r i bu t i on of ownership units and the areas covered by them as reported in the 1977 Land Occupancy Survey are presented in Table 5. The average s ize of holding in 1977, according to the Land Occupancy Survey, was only 1.84 acres. As Table 5 shows, more than a quarter of the avai lable farmland was owned by those who had more than 8 acres of land, that i s , more than four times the average s i ze. And these farms constituted only 3.43% of a l l farms. One thing to be noted about Tables 3 to 5 i s that none of these considered the landless households. The inega l i ta r ian pattern of landholding becomes even more pronounced i f one looks at the number of rural households without land. Sat i s factory information on the level of landlessness in Bangladesh was not avai lable pr io r to the 1977 Land Occupancy Survey. Neither the 1960 Census of Agr iculture nor the 1968 Master Survey reports the number of landless households. However, according to the 1964 Master Survey of Agr icu l ture, 3.6% of 53 households at that time were t o t a l l y landless. This f igure for Government of Bangladesh, Summary Report of the 1977 Land Occupancy Survey of Rural Bangladesh, Dacca: Bangladesh Bureau of S t a t i s t i c s , 1977. Government of East Pakistan, Master Survey of Agriculture in East Pakistan, S ixth Round (1964-65), Dacca: East Pakistan Bureau of S t a t i s t i c s , 1966, p. 17. Table 5. D i s t r ibut ion of Land-Ownership Under Different Size-Categories and the Area Covered by Them in Bangladesh, 1977. Size-Category (acres) Number of Households Percentage of A l l Households Area Covered (acres) Percentage of Total Area .01 _ 1.00 5,621,303 53.35 1 ,800,165 9.30 1.01 - 2.00 1,946,450 18.47 2,792,855 14.43 2.01 - 3.00 1 ,055,543 10.02 2,551 ,605 13.18 3.01 - 4.00 624,412 5.93 2,153,332 11.13 4.01 - 5.00 389,454 3.70 1,741,880 9.00 5.01 - 6.00 247,534 2.35 1,334,433 6.90 6.01 - 7.00 169,817 1 ..61 1 ,102,116 5.96 7.01 - 8.00 120,339 1.14 899,196 4.65 8.01 - 9.00 82,365 0.78 696,996 3.60 9.01 - 10.00 49,589 0.47 476,443 2.46 10.01 - n .oo 39,931 0.38 415,482 2.15 11.01 - 12.00 34,086 0.32 392,750 2.03 12.01 - 13.00 18,408 0.17 228,625 1.18 13.01 - 14.00 26,356 0.25 351,372 1.82 14.01 - 15.00 15,711 0.15 230,535 1.19 Over 15 95,790 0.91 2,183,991 11.29 Total 10,537,088 100.00 19,351,776 100.00 Source: Government of Bangladesh, Summary Report of the 1977 Land Oooupanoy Survey of Rural Bangladesh, Dacca: Bangladesh Bureau of S t a t i s t i c s , 1977. 98 landlessness seems to be unusually low when one considers the high man-land r a t i o in Bangladesh. The extent of 'near-landlessness 1 was, according to the 1964 Master Survey of Agr icu l ture, however, quite considerable: 28.6% of households had less than .25 acres of 54 land, most of which would be required for homestead. Other f igures for landless fami l ies are ava i lable from a survey conducted by the Co-operative S t a t i s t i c s and Research Organization of the Government of East Pakistan. The survey, conducted in 1966 and based on a sample study of 12 v i l l ages in four d i f fe rent d i s t r i c t s , puts the 55 f igure of landlessness at 6.97%. However, the sample base i s much too small in th i s case to put much rel iance on the f igure provided. The 1977 Land Occupancy Survey documents landlessness more adequately for our purpose. The relevant information i s summarized in Table 6. Turning now to the question of access to or control of productive land, one can d ist inguish the fol lowing main categories: ( i ) owner-farmers: Those who own t he i r farms and use family labour and/or hired labour for cu l t i va t i ng the i r farms, ( i i ) owner-cum-tenants: those who cu l t i va te some owned land and some rented land mainly by family labour, The average area covered by homestead in rural Bangladesh has been found to be .18 acre. Government.of East Pakistan, Agricultural Credit in East Pakistan, Dacca: The Registrar of Cooperative Soc ie t ie s , 1967, p. 13. Table 6. Landlessness in Rural Bangladesh, 1977 Number of Households Percentage of A l l Households Number of Persons Percentage of A l l Persons Households with no land 1,311,570 11.07 5,884,927 8.13 Households with only homestead land 2,574,163 21.72 12,818,545 19.03 Total of households having no or ju s t homestead land 3,885,733 32.79 18,703,472 27.10 Source: Government of Bangladesh, Summary Report of the 1977 Land Oecupanoy Survey of Rural Bangladesh, Dacca: Bangladesh Bureau of S t a t i s t i c s , 1977. 100 ( i i i ) pure tenants jbargadars): those who have no land of t he i r own and rent land to cu l t i va te mainly by family labour, ( iv) wage-labourers: those who o f fe r family labour for h i re. One should note that the fourth.category, although consist ing mainly of landless persons, i s not mutually exclusive from the other categories. Some land-owning farm fami l ies do of fer some labour to the wage market. Information on the r e l a t i ve importance of the f i r s t three tenure categories as found in each of the three surveys of 1960, 1968 and 1977 has been summarized in Table 7. The data presented in Table 7 show that owner-farmers have been, and s t i l l are, dominant in numbers as well as in the area held by them. Between 1960 and 1968 the number of owner-farmers increased by more than 5 percentage points, whereas the number of owner-cum-tenants decreased by more than 1%. The number of pure tenant farms increased, but only by less than 2 percentage, points. The explanation fo r these changes may l i e in the fact that many of .the farms which were previously renting out land became smaller over the years due to subdivision among he i r s , and were l e f t with l i t t l e land to rent out 56 af ter employing family labour on t he i r lands. Data on the areas It i s i n s t ruct i ve to note here that average cu l t ivated area per farm came down from 3.12 acres in 1960 to 2.86 acres in 1968. This was associated with an increase in the number of farms by 730,000 during th i s period. 101 Table 7. D i s t r i b u t i o n of Farms Among the Main Tenure-Categories and the Area Covered by Them in Bangladesh, 1960, 1968 and 1977 Year and Tenure Category Farms Area Number of Farms Percentage of Farms Total Area (acres) Percentage of Area 1960 Owner-farmers 3,731,110 61.00 11 ,653,910 54.00 Owner-cum-tenants 2,308,330 37.00 9,829,813 45.00 Pure tenants 100,040 2.00 242,104 1.00 A l l farms 6,139,480 100.00 21,725,827 100.00 1968 Owner-farmers 4,567,180 66.48 12,592,790 58.40 Owner-cum-tenants 2,058,250 29.96 8,252,160 38.27 Pure tenants 244,570 3.56 718,050 3.30 A l l farms 6,870,000 100.00 21 ,563,000 100.00 1977 Owner-farmers 5,005,800 61.17 10,164,900 53.95 Owner-cum-tenants 2,618,300 31.99 7,846,900 41.66 • Pure tenants 559,500 6.84 827,900 4.39 A l l farms 8,183,600 100.00 18,839,700 100.00 Note: The 1977 survey d id not report data on owner-farmers as such, but presented f i gu re s f o r the fo l l ow ing two groups: ( i ) owner - cu l t i va to r s : those who c u l t i v a t e d t h e i r land with fami l y labour on ly , ( i i ) owner-managers: those who employed some wage-labour i n c u l t i v a t i n g t h e i r land. When we add up the f i gures f o r the above two groups we end up wi th the re levant informat ion f o r owner-farmers. Sources: 1. Government of Pak i s tan , Pakistan Census of Agriculture, I960, Vo l . I, F ina l Report - East Pak i s tan, Part I, October 1962. 2. Government of Bangladesh, Master Survey of Agriculture in Bangladesh, Seventh Round, Second Phase (1967-68), Dacca: Bangladesh Bureau of S t a t i s t i c s , (undated). 3. Government of Bangladesh, Summary Report of the 1977 Land Occupancy Survey of Rural Bangladesh, Dacca: Bangladesh Bureau of S t a t i s t i c s , 1977. 102 covered by d i f fe rent tenure-categories during 1960 and 1968 lend support to th i s hypothesis. When one compares the s i tuat ion in 1977 with that of 1968, one f inds , however, that the number of owner-farmers decreased over these years by more than 5%. The number of owner-cum-tenant and pure tenant farms increased by over 2% and 3% respect ively during the same period. This indicates that with the passage of time some of the owner-farmers started renting land to sa t i s f y family needs. Data on areas covered by the tenure-categories during the relevant periods lend support to t h i s . The picture of land-hunger that one gets from the discussion above i s further reinforced by the rapidly increasing number of pure . 57 tenants. Data obtained from the 1977 Land Occupancy Survey can be used to produce the fol lowing information on the d i s t r i bu t i on of rented land among land-owning and landless households: percentage of rented land under land-owning households 80.78 percentage of rented land under landless households 19.22. Evidence that indiv idual tenants found i t increasingly d i f f i c u l t to obtain land in the rent market i s provided by M.G. Sattar, 'Family Case Study of an Owner-cum-Tenant Farmer', in Government Of Bangladesh, Proceedings of the Bangladesh-FAO Workshop on the Problems of Small and Subsistence Farmers and A g r i c u l t u r a l Labourers, Dacca: Bangladesh Agr i cu l tu ra l Research Count i l , 1974. 103 The households with no cu l t i vab le land obtained less than a f i f t h of the to ta l area rented out although they constituted over 32% of rural households. Explanation fo r th i s may l i e in the fol lowing three main reasons: 1. In making a rental contract, the supplier of land, or the landlord, i s l i k e l y to prefer the potential user of his land to supply his own work animals and farm implements. Although we have no data on t h i s , i t i s probably safe to argue that the landless, who are the poorest among the rural c lasses, have fewer of these assets than the i r land-owning neighbours. 2. The households without any land, when offered land for sharecropping, may develop too deep an attachment to the land, to be eas i l y evicted by the landlord at l a te r dates. 3. The landlords want to create a larger c i r c l e of influence for atta in ing greater p o l i t i c a l power. They can earn the al legiance of the landless by simply giving them, employment as wage-labour on portions of the i r ( landlords ' ) land, whereas, to gain the favour of some of the households owning small amounts of land, they (the small land owners) have to be given land on rent. 104 The importance of the land-rent market in rural Bangladesh i s depicted in Table 8. Table 8 indicates that the area that was rented has increased over the years. In Bangladesh most of the rented land i s taken in under sharecropping contracts. Thus, according to the 1960 Census, 88.98% of to ta l rented land was reported to have been taken on under sharecropping contracts. The rest was given out on a cash-rent basis. Neither the 1968 Master Survey nor the 1977 Land Occupancy Survey reports separate data for the sharecroppers and the cash-renters. 58 However, the survey conducted by Jabbar provides some recent data. According to th i s survey, in 1974, out of a l l farms renting in land, 92.26% did so on a sharecropping basis. The usual pract ice under sharecropping i s to l e t the tenant keep half of the produce from rented land and l e t the landlord have the other half as rent for his land. The s i tuat ion in 1977 i s depicted in Table 9. Both in terms of percentage of households and percentage of area, the 50-50 sharing of the produce between the tenant and the landlord i s the most common in Bangladesh. The rest of the contracts are mainly based on terms whereby more than 50% of the produce i s demanded as rent. It has also been reported that over arid above the payments in k ind, in some cases, payments were also required in cash as part of the rent. 59 Table 10 presents some of the deta i l s in th i s regard. Jabbar, op.ait. Although i t i s common knowledge in Bangladesh that the landlords often demand extra labour from the i r tenants for numerous jobs unrelated to the cu l t i v a t i on of the rented land, lack of data prevented i t s inclus ion in the Table. Table 8 Owner-Operated Area and Tenant-Operated Area in Rural Bangladesh, 1960 and 1977 Owner-operated area: Number of acres (covering a l l farmland of owner-farmers and the owned farmland of owner-cum-tenants) Percentage of to ta l farmland 1960 1977 17,779,567 14,531,100 82 77 Tenant-operated area: Number of acres (covering a l l farmland of tenants and the rented farmland of owner-cum-tenants) Percentage of to ta l farmland 3,946,260 4,308,600 18 23 Notes: ( i ) Owner-operated area includes a l l area of the owner-farmers and the area which i s owned by the owner-cum-tenants in the i r respective farms. ( i i ) Tenant-operated area includes a l l area of the tenants and the area which i s rented by the owner-cum-tenants. Sources: 1. Government of Pakistan, Pakistan Census of Agriculture, I960, Vol. I, Final Report - East Pakistan, Part I, October 1962. 2. Government of Bangladesh, Summary Report:of the 1977 Land Occupancy Survey of Rural Bangladesh, Dacca: Bangladesh Bureau of S t a t i s t i c s , 1977. 106 Table 9. Percentage of Tenant Households Paying Varying Amounts of Rent in Kind and Areas Under Them in Bangladesh, 1977 Amount of Rent Percentage of Households Percentage of Area Less than 50% of produce 1.26 1.54 50% of produce 93.27 90.42 More than 50% of produce 5.47 8.04 Total 100.00 100.00 Source: Government of Bangladesh, Summary Report of the 1977 Land Occupancy Survey of Rural Bangladesh, Dacca: Bangladesh Bureau of S t a t i s t i c s , 1977. 107 Table 10. Percentage of Tenant Households Making Additional Cash Payments Over and Above Rent-in-Kind in Bangladesh, 1977 Categories of Payers of Rent-in-Kind Percentage of House-holds Paying Addi-t ional Cash Over Payments in Kind Percentage of Area for which Additional Cash Payments Made Less than 50% of produce .04 .04 50% of produce 8.26 10.43 Over 50% of produce .67 .90 Total 8.97 11.37 Source: Government of Bangladesh, Summary Report of the 1977 Land Occupancy Survey of Rural Bangladesh, Dacca: Bangladesh Bureau of S t a t i s t i c s , 1977. 108 The prevalence of cash payment was widespread not so much among those contracts which required less than 50% of the produce as rent, but among the other two kinds of contracts shown in Table 10 (more so among those paying 50% of produce as rent) . Therefore, one cannot argue that the primary explanation for the cash payment i s that those who paid less than 50% of the produce as rent had to be charged something extra so that the e f fec t i ve rent could be brought nearer to the standard 50% share. The re spons ib i l i t y of supplying the basic ag r i cu l tu ra l inputs in a sharecropping contract l i e s mainly with the tenant, which is 60 evidenced by the data presented in Table 11. Duration of tenancy contracts is another factor that has to be considered to throw more l i gh t on the nature of tenant-landlord re lat ionsh ip. As Table 12 shows, there i s an ind i rect re lat ionship between the duration of tenancy and the percentage of contracts. Over half the contracts were of a duration of e i ther 2 years or.up 61 to a year. The landlords prefer shorter duration probably because they fear that i f the rented lands are cu l t ivated by the same tenants for a long time then these tenants may ult imately s tar t demanding r ights s imi la r to those enjoyed by owners. Although i t may be true that the landlords, in some cases, help t h e i r tenants to obtain permits to buy the inputs at f i xed p r i ces , the cost is usually incurred by the tenants. One may note in th i s regard that President Zia has ca l l ed for a minimum tenure of 3 years for sharecropping contracts in Bangladesh. 109 Table 11. Source of Agr icu l tura l Inputs Applied on Rented Land in Bangladesh, 1977 Inputs Percentage Provided by Landlord Percentage Provided by Operator Seed .59 99.41 Fe r t i 1 i ze r .36 99.64 Pest ic ide .22 99.78 I r r i gat ion f a c i l i t y .03 99.47 Source: Government of Bangladesh, Summary Report of the 1977 Land Oooupanoy Survey of Rural Bangladesh, Dacca: Bangladesh Bureau of S t a t i s t i c s , 1977. 110 Table 12. Percentage of Tenancy Contracts Under Varying Durations of Tenancy and the Percentage of Total Area Covered by Them in Bangladesh, 1977 Duration of Tenancy Percentage of Contracts Percentage of Total Rented Area Up to 1 year 30.85 21.89 2 years 24.48 25.98 3 years 15.52 18.41 5 years 5.55 6.07 Source: Government of Bangladesh, Summary Report of the 1977 Land Occupancy,Survey of Rural Bangladesh, Dacca: Bangladesh Bureau of S t a t i s t i c s , 1977. The owner-cum-tenant farms in Bangladesh are t y p i c a l l y small. The 1977 Land Occupancy Survey puts the average s ize of owned land of the owner-cum-tenants at 1.67 acres. The amount that they were able to rent was also small — the average household renting only 1.33 acres. More than 80% of the owner-cum-tenants rented less than 2 acres of land. i n To summarize the discussion so far in th i s section we may note the fol lowing fact s : 1. There i s an overwhelming number of very small farms. 2. At the same time, the area covered by a r e l a t i v e l y few large farms i s quite s i gn i f i can t . 3. Families without any farmland of t he i r own const itute about a t h i r d of a l l rural households. 4. Although owner cu l t i va t i on i s s t i l l the dominant form of c u l t i v a t i o n , tenancy i s gaining in importance. 5. About 39% of a l l farm households rent in land; the percentage of farmland that i s rented amounts to 22.87% (these are 1977 f igures ) . A majority of the households who get land on rent are themselves owners of some land. 6. Sharecropping i s by far the most dominant arrangement under which land i s rented. The average s ize of land rented by a household i s very small. The usual rent i s 50% of to ta l produce on rented land. Additional cash payments over and above rent in kind are present in about, a tenth of the contracts. The prime r e spon s i b i l i t y for bearing input costs in most of the sharecropping contracts in Bangladesh l i e s with the tenant. Relat ive ly short - l i ved tenure contracts have been found to be more numerous. 112 So f a r , the discussion on the size-tenure structure in Bangladesh agr iculture has not considered the regional var iat ions within the country. We now turn to a discussion of the regional va r ia t ions , not only in the size-tenure st ructure, but also in some other related variables which shape the nature of the tenure pattern and i t s domain of problems and prospects. Regional Variations in Bangladesh Agr iculture In the preceding chapter i t was maintained that there may ex i s t d i f fe rent ' ag r i cu l tu ra l systems' in d i f ferent l o c a l i t i e s of the same country. To investigate the nature of such systems within Bangladesh one has to look for regional var iat ions in the variables which were CO taken to define such a system. Since data on a l l of these variables are not ava i lable for Bangladesh, an attempt w i l l be made in the fol lowing to demonstrate, with whatever data are ava i lab le , that there are indeed d i f fe rent systems in the country which cannot read i ly be i den t i f i ed by looking at a s ingle var iable such as tenur ia l arrangements. By incorporating the concept of varying agr icu l tura l systems into the discussion on the size-tenure structure contained in the preceding sect ion, one is better able to appreciate the spec i f i c nature of the problem of size-tenure structure v i s -a -v i s ag r i cu l tu ra l development in d i f ferent regional contexts. See the discussion in Chapter I I, pp. 20-21. 113 Topographic and c l imat ic var iat ions: Although the de l ta i c p la in of Bangladesh has a r e l a t i v e l y broad degree of homogeneity in topographic and c l imat i c condit ions, closer examination reveals s i gn i f i cant differences. The land of the country i s comprised of d i f ferent kinds of s o i l as summarized 63 in Map 2. Some of the more extensive so i l s are s i l t s , sandy and clayey loams, red s o i l s , h i l l s o i l s and h i l l wash, and swampy s o i l s . The s i l t s are mainly confined to the central region covering large parts of the d i s t r i c t s of Tangai l , Mymensingh, Dacca, Comil la, Faridpur and Pabna. Sandy loams are the most extensive s o i l in the d i s t r i c t of Sylhet. The northeastern d i s t r i c t s (Dinajpur, Rangpur, Rajshahi and Bogra) also have large proportions of th i s var iety of s o i l . In the southwestern part of the country the dominant s o i l i s clayey loam. The southern-most d i s t r i c t s of Khulna and Patuakhali have s i gn i f i cant proportions of swampy, sal ine and sandy s o i l s . The southeastern part of the country (the d i s t r i c t of Chittagong H i l l Tracts in par t i cu la r ) i s mainly comprised of h i l l s o i l s and h i l l wash. These s o i l s are based on sandstones, shales or c lays. Parts of the d i s t r i c t s of Dacca, Tangai l , Mymensingh, Dinajpur, Rangpur, Bogra and Rajshahi are composed of red s o i l s . The d i f fe rent kinds of s o i l which make up the ag r i cu l tu ra l land of Bangladesh have d i f fe rent degrees of f e r t i l i t y . The ease with which these s o i l s can be worked also var ies. The s i l t s are very Information on the regional differences in s o i l composition has been obtained from Ahmed, op.oit., pp. 21-26. Map 1. Bangladesh: Administrative Boundaries 114 INDIA BAY OF B E N G A L 0 4 0 8 0 i i i i i i M I L E S International boundary _ . . . . II II Divisional District Map 2. So i l Composition of Bangladesh 115 116 f e r t i l e and can be worked eas i l y . The loams are intermediate in qua l i ty between s i l t s and c lays, the l a t t e r being a heavy s o i l which turns s t icky during the rainy season and hard during the dry season. The red s o i l s , which have developed on the old a l l u v i a , are of r e l a t i v e l y low f e r t i l i t y . This kind of s o i l i s very d i f f i c u l t for ploughing during the dry season since i t becomes extremely hard and porous. The level of f e r t i l i t y of the h i l l y s o i l s i s general ly low. The swampy s o i l s are often covered by peat due to the accumulation of organic matter on the surface. This can raise the level of f e r t i l i t y of th i s kind of s o i l . However, the swampy s o i l often receives seasonal s i l t deposits which turns i t into a s t i f f , black clay. This can make the working of th i s s o i l d i f f i c u l t . The swampy s o i l s , which are often formed near coastal regions, may also suffer from some degree of s a l i n i t y . There are also var iat ions in r a i n f a l l and the water-holding capacity of the land among d i f ferent regions of the country. Although the whole country benefits from the monsoonal r a i n f a l l during July to September, the average annual r a i n f a l l i s not evenly d i s t r ibuted throughout. It i s possible to div ide the country into three broad categories of dry land, moderately wet land and very wet land on the basis of the a v a i l a b i l i t y of surface water which depends upon the amount of r a i n fa l l . and the water-holding capacity of the land. These three regions have been shown in Map 3. The northeastern d i s t r i c t s of Dinajpur, Bogra, Rajshahi, Pabna, Kushtia and Jessore const itute the Map 3. Dry and Wet Regions of Bangladesh Dry Region • •••I (average annual r a i n f a l l 60 inches) Moderately Wet Reqion (average annual r a i n f a l l 85 inches) Very Wet Region (average annual r a i n f a l l 150 inches) 118 dry region. The average annual r a i n f a l l for the region i s about 64 60 inches. The moderately wet region i s comprised of the d i s t r i c t s of Rangpur, Mymensingh, Tangai1, Dacca, Comil la, Faridpur, Patuakhal i, Bar isal and Khulna. The average annual r a i n f a l l in th i s region i s 65 around 85 inches. The very wet region of the country i s mainly confined to the d i s t r i c t s of Sylhet, Noakhali, Chittagong and Chittagong H i l l Tracts. The average annual r a i n f a l l in th i s region i s about 150 inches. Among these three d i s t r i c t s , Sylhet gets the most r a i n f a l l (annual average of 175 inches). The differences in r a i n f a l l , coupled with topographic d i f ferences, account for d i f fe rent 66 levels of f looding in d i f ferent parts of the country. Thus, in the r e l a t i v e l y high dry region, f looding i s mainly shallow. In the wet regions, almost the whole of Sylhet and parts of Mymensingh and Dacca suffer from deep f looding. Comilla and Faridpur are areas of moderate f looding. Flooding in the d i s t r i c t s of Bar isal and Khulna i s usually shallow but widespread. Apart from rain water, t i d a l waves occasionally 64 This f igure for the annual r a i n f a l l was calculated from data provided by Government of Pakistan, Population Census of Pakistan, .1961: District Census Report, Kushtia, Karachi: Ministry of Home A f f a i r s , (undated), p. 1-19; and Government of Bangladesh, S a t i s t i c a l Digest of Bangladesh, No. 8, Dacca: Bangladesh Bureau of S t a t i s t i c s , 1972, p. 6. 65 See Government of Bangladesh, S t a t i s t i c a l Digest of Bangladesh, No. 8, Dacca: Bangladesh Bureau of S t a t i s t i c s , 1972, p. 6. 66 See B.L.C. Johnson, Bangladesh, New York: Harper and Row Publishers, 1975, p. 28. 119 cause temporary flooding in the southern parts of these d i s t r i c t s . The other area of mainly shallow flooding in the wet regions i s the d i s t r i c t of Chittagong H i l l Tracts. Variations in cropping pattern: Differences in s o i l conditions and a v a i l a b i l i t y of water have resulted in d i f fe rent cropping patterns in various parts of the country. Cu l t ivat ion of r i c e , which i s by far the major crop of Bangladesh (78% of the cropped acreage is devoted to r i ce ) and of each of the d i s t r i c t s , involves d i f fe rent combinations of a number of r i ce crops in d i f fe rent regions (see Map 4 ) . ^ 7 The northwestern d i s t r i c t s of Dinajpur, Rajshahi and Bogra are found to devote most of the r i ce land to transplanted aman. The clayey s o i l s of these d i s t r i c t s have to be softened adequately by ra in water before land preparation and sowing can be possible with the avai lable technology; and the cu l t i va t i on ofaman allows t h i s . Although the seed beds are to be prepared ear ly in the monsoon (March/April), transplantation can wait t i l l June or Ju l y . The cu l t i va to r s f ind i t possible to take care of the seed beds with re l a t i ve ease because of the smallness of the plots involved. Aman i s preferred against any other ra in-fed crop during the period of the wet season in which the land becomes workable because i t i s good qua l i ty r i ce with reasonably high y ie ld s . The Information on cropping pattern was avai lable from Government of Pakistan, Pakistan Census of Agriculture, I960, Vol. I, Final Report - East Pakistan, Part I, October 1962, pp. 162-244. Map 4. Rice Crop Regions of Bangladesh 111 [IH Mainly Transplanted Aman (about 85% of 11111 tl r i c e acreage) 1+ + 1 Mainly Broadcast Aman (about 70% of ' ~*"^ r i ce acreage) j ^ ^ j Mainly Aus (about 70% of r i ce acreage) 11/11 Mainly -4MS and Transplanted Aman (about 90% of L - — 1 r i ce acreage) •Mainly Aus and Broadcast Aman (about 90% of r i ce acreage) Mainly Boro (about 90% of r i ce acreage) 121 southwestern region of the country i s also found to devote most of i t s r i ce land to the transplanted aman. Again, the reason i s that, with given technology, the preparation of the land for cu l t i va t i ng ra in- fed crops cannot be undertaken immediately a f ter the onset of monsoonal r a i n . The rain has to wash down the sa l t on the surface of the land deep enough to allow the roots to develop in a s a l t - f r ee s o i l . The moderately flooded s i l t s are best suited to broadcast aman. Therefore, one can d ist inguish a be l t of land area s tart ing from southwestern Sylhet, bordering the d i s t r i c t s of Noakhali and Comil la, covering parts of Dacca and most of Faridpur. and Pabna which i s characterized by the cu l t i va t i on of broadcast aman in most of the r i ce f i e l d s . The aus paddy thr ives best on higher loamy clay s o i l s . The d i s t r i c t s of Jessore and parts of Chittagong H i l l Tracts are i dea l l y suited for th i s crop. The cu l t i va t i on of aus requires that the f i e l d i s ready for broadcasting by A p r i l , which means that th i s crop i s l imited to only those s o i l types which are l i g h t enough to be worked on with l i t t l e r a in . The cropping period of aus allows some regions to cu l t i va te both aus and transplanted or broadcast aman depending upon the so i l and water conditions of the regions. Those areas which are found to devote comparable portions of to ta l r i ce acreage to two var ie t ie s can be seen in Map 4. The bovo paddy i s the other dominant r i ce var iety in Bangladesh which i s cu l t ivated during the dry season. Those areas within the country which are heavily flooded during the rainy season 122 can o f fe r enough ground water for i r r i g a t i o n during the dry season for the cu l t i v a t i on of boro. Thus, Sylhet and parts of Noakhali and Chittagong are the main boro producing d i s t r i c t s . Boro i s r e l a t i v e l y coarse when compared with the other r i ce crops. However, the y i e l d i s the highest among the d i f fe rent r i c e crops (see Table 13). This has been so even before the introduction of the high y ie ld ing var iety of boro, IR-8. Since i t s int roduct ion, per acre y i e l d has gone up even higher. The data presented in Table 13 on per acre y ie lds of d i f fe rent r i ce va r ie t ie s pr ior to the introduction of IR-8, and af ter i t was introduced, w i l l provide evidence to th i s e f fec t . Table 13. Yields of Different Rice Crops in Bangladesh in Different Years Period Under Consideration Y ie ld (maunds/acre)* Aus Aman . Boro Pre IR-8 5 years ' average covering 1955-56 to 1959-60 9.0 10.6 11.8 Post IR-8 5 years 1 average covering 1968-69 to 1972-73 9.2 11.9 22.9 Source: Government of Bangladesh, Bangladesh Agriculture in Statistics, Dacca: Min istry of Agr icu l ture, November, 1973. * 1 maund = 82.29 lbs. 123 Although r i ce crop cu l t i va t i on i s the major occupation of the cu l t i vator s in Bangladesh agr i cu l tu re , there are some important cash crops as w e l l . The most important among these i s j u te . About 10% of the to ta l cropped acreage i s devoted to j u te . Abundant r a i n f a l l and s o i l s r i ch in s i l t s are essential for the cu l t i va t i on of j u te . However, i f the plants are l e f t to grow in submerged condit ions, then the qua l i ty of jute deter iorates. Therefore, areas of deep flooding are not wel l suited for th i s crop. The primary jute be l t , which i s located around the centre of the country i s shown on Map 5. One may note here that the cropping period of jute (February to June) overlaps with that of aus paddy, so that these two crops compete with each other for cu l t i vab le land. Sugar cane i s the second important cash crop which i s produced mainly in the d i s t r i c t s of Dinajpur, Rajshahi and Kushtia. Other important crops include tea along the southeastern border of Sylhet, tobacco in Rangpur, wheat and pulses in the d i s t r i c t s of Rajshahi, Kushtia and Jessore. Areas within which these d i f fe rent crops are important as subsidiary crops to r i ce have been shown in Map 5. Map 5. Regions of Some Important Crops Other than Rice in Bangladesh 124 125 Variations in size-tenure s t ructure: The r e l a t i v e l y dry region of the country has t r a d i t i o n a l l y been found to have a lower population density than most other parts of the country. The re l a t i ve ease with which most of the wet land could be cu l t i vated resulted in dense habitation in those areas. However, some wet areas have not been easy to cu l t i v a t e ; for example some areas of longstanding deep flooding (for example, in Sy lhet) , areas with swampy and sa l ine s o i l s (for example, in Khulna), areas which are h i l l y and rugged (for example, in Chittagong H i l l Tracts) . Therefore, even within the wet regions, population settlements have not been equally dense everywhere. Not only the ease with which the land could be cu l t i va ted , but also the nature of crop that could be produced within a region affected i t s population density. During the e a r l i e r days, when the products of the farm were almost exc lus ive ly meant to s a t i s f y domestic consumption needs, the regions which could assure the cu l t i vator s of a steady supply of food crops were most favoured for purposes of settlement. Thus, Dacca d i v i s i on which is the smallest of a l l the d i v i s i ons , had the largest populati as ear ly as 1901, one reason for which was the s u i t a b i l i t y of i t s climate and agr icu l tura l land for purposes of producing various 68 foodcrops with r e l a t i ve ease. Some other parts of the country which are not c l i m a t i c a l l y or topographically as we l l - su i ted for See Government of Pakistan, Census of Pakistan Population, 1961, Vol. 2, East Pakistan, Karachi, June 1964, p. 11-7. 126 agr icu l tura l operations have gradually attracted more and more people because of increasing levels of urbanization of those areas. The Chittagong d iv i s ion had a.population in 1901 which was smaller than that of the d iv i s ion of Rajshahi (Chittagong d iv i s ion 6.8 m i l l i o n , Rajshahi d iv i s ion 7.5 m i l l i o n ) , although i t had a larger area (Chittagong d iv i s ion 16,510 square mi les, Rajshahi d i v i s i on 69 12,677 square mi les) . With increasing urbanization of some parts of the Chittagong d i v i s i on , i t s population in 1961 surpassed that of the d i v i s i on of Rajshahi by about 2 m i l l i o n people (Chittagong d iv i s ion 13.6 m i l l i o n , Rajshahi d iv i s ion 11.8 mi 11 i o n ) . ^ This increase in population was not due only to an increase in the number of people l i v i n g in urban areas. Rural population also increased s i g n i f i c a n t l y . ^ The opening up of newer areas of cu l t i va t i on by creating better communication systems and by of fer ing the cu l t i va tor s some opportunities of d i ver s i f y ing the i r source of income, the urbanization process was able to a l t e r the pattern of density of population in d i f fe rent rural areas. These d i f f e r e n t i a l densit ies of settlement have been ref lected in the differences in the average sizes of farms in d i f ferent parts of the country. The average farm s ize cannot, however, provide one with an idea about the skew in the 6 9 Tbid.3 pp. II-6, 11-11. 7 0 Ibid., p. II-6. ^ For deta i l s see ibid., p. 11-20. 127 d i s t r i bu t i on of land among the various farm households. The nature of th i s skew i s l i k e l y to be an important factor in determining the pattern of land tenure within a region. It i s beyond the scope of the present study to undertake a deta i led h i s t o r i c a l invest igat ion into the causes of the various degrees of th i s skew in d i f fe rent parts of the country. Nevertheless, i t may be worthwhile to remember here the e a r l i e r discussion in th i s chapter on the process in which remnants of feudalism have perpetuated wide interpersonal d i spa r i t i e s in land ownership, which could have attained d i f fe rent degrees in d i f ferent parts of the.country. Table 14 presents some data on the d i spar i t y of land ownership among the rural households of a number of d i s t r i c t s in Bangladesh. Table 14. Percentage D i s t r ibut ion of Agr icu l tura l Households by Their Land-owning Status in a Number of D i s t r i c t s in Bangladesh, 1977 D i s t r i c t Percentage of Landless Households Percentage of Households with More than 5 acres Dinajpur 31.1 14.9 Rajshahi 30.0 13.3 Kushtia 36.6 13.3 Khulna 29.2 10.6 Tangail 27.3 8.3 Comi11 a 26.1 6.3 Sylhet 31.2 10.7 Noakhali 33.2 5.4 Source: Bangladesh Agr icu l tura l Research Counci l , Incidence of Landlessness and Major Landholding and Cultivating Groups in Rural Bangladesh, Dacca: Directorate of Agr iculture (Extension and Management), May 1978. 128 It can be seen in Table 14 that the d i spar i ty in land ownership i s most acute in the northwestern region of the country. It was mentioned above that sharecropping cu l t i va t i on i s the most extensive tenure arrangement next to owner cu l t i v a t i on . However, the extent of sharecropping cu l t i va t i on d i f f e r s among d i f ferent parts of the country. Table 15 shows the percentage of farm area sharecropped in the d i f ferent d i s t r i c t s of the country (see page 130). To ident i f y the factors which were s i gn i f i cant in determining the levels of sharecropping cu l t i va t i on in the d i s t r i c t s of Bangladesh, a mult ip le regression was run with the incidence of sharecropping cu l t i va t i on as the dependent var iable and population density, average annual r a i n f a l l , farm area as percentage of to ta l land area, average farm s i ze , percentage of farm area under large (12.5 acres or more) farms, and percentage of households w.ith small farms (below 2.5 acres) 72 as the independent var iables. The regression results provided the fol lowing information on the re lat ionship between the dependent variable and those independent variables which were significant in explaining var iat ions in the incidence of sharecropping cu l t i v a t i on : S '=15.83168 - .13039 R + .59907 L (-3.26) (2.91) R2 = .53 Data used for th i s regression analysis are put in the Appendix. See Table A l , pp. 268-270. 129 where, S = percentage of farm area sharecropped, R = average annual r a i n f a l l , and L = percentage of farm area under large (12.5 acres or more) farms. The coe f f i c i en t of corre lat ion between S and R was - .50 and that between S and L was .45. The coef f i c ient s of R and L were found to be s i g n i f i c an t l y d i f fe rent from zero allowing for 1% and .5% errors respect ively. Therefore, both R and L were s i gn i f i cant in determining S. Looking at the f igures provided in Table 15 i t i s possible to ident i f y d i f fe rent regions within Bangladesh which correspond to d i f ferent degrees of sharecropping cu l t i v a t i on . These regions are i den t i f i ed in Map 6, and also in Chart 1. Agr icu l tura l Systems The reg iona l izat ion of Bangladesh agr icu l ture which has been attempted above has been done separately for separate variables l i k e s o i l composition, a v a i l a b i l i t y of water, cropping pattern and tenure status. It would be worthwhile to see, however, the nature of the reg ional izat ion which emerges when a l l of these variables (and other related ones) are considered simultaneously. This would give us an approximation to the var iat ions in agr i cu l tura l systems within the country. With the help of the Guttman-Lingoe 1s Mult ip le Scalogram 130 Table 15. Percentage of Farm Area Sharecropped in Different D i s t r i c t s of Bangladesh, 1960 D i s t r i c t Percentage of Farm Area Sharecropped Dinajpur 24.39 Rajshahi 24.81 Kushtia 21.18 Jessore 18.01 Pabna 19.18 Bogra 13.18 Rangpur 14.25 Bar isa l^ 17.80 Khula 23.47 Dacca 12.53 2 Mymensingh 14.75 Faridpur 15.53 Comilla 3.84 Chittagong H i l l Tracts 1.45 Sylhet 8.90 Noakhali 13.10 Chittagong 19.03 Includes the present d i s t r i c t s of Barisal and Patuakhali. Includes the present d i s t r i c t s of Mymensingh and Tangai l. Source: Government of Pakistan, Pakistan Census of Agriculture, I960, Vo l . I, Final Report - East Pakistan, Part I, October 1962. 131 Map 6. Incidence of Sharecropping Cu l t ivat ion in the D i s t r i c t s of Bangladesh, 1960 Willi 5-10% of Cropped Acreage I 1 Below 5% of Cropped Acreage 132 Chart 1. Degrees of Sharecropping Cu l t ivat ion in the Different D i s t r i c t s of Bangladesh, 1960 Incidence of Sharecropping D i s t r i c t Very High Sharecropping (above 20% of cropped acreage sharecropped) Khulna Rajshahi Dinajpur Kushtia High Sharecropping (15-20% of cropped acreage sharecropped) Bar isal Pabna Jessore Chittagong Faridpur Moderate Sharecropping (10-15% of cropped acreage sharecropped) Bogra Noakhali Rangpur Mymensingh Dacca Little Sharecropping (5-10% of cropped acreage sharecropped) Sylhet Very Little Sharecropping (Less than 5% of cropped acreage sharecropped) Comi11 a Chittagong H i l l Tracts 133 Analysis i t was possible to ident i f y the d i s t r i c t s which could be grouped together on the basis of the values of 19 d i f fe rent 73 variables in the d i s t r i c t s . The variables were the percentage of land covered by each of the nine d i f fe rent s o i l types (see Map 2), average annual r a i n f a l l , population density, farm area as percentage of tota l land area, average farm s i ze , percentage of farm area covered by farms sized 12.5 acres or more, percentage of households with farms below 2.5 acres, percentage of cropped acreage devoted to r i c e , percentage of cropped acreage devoted to the important secondary crops ( jute, sugar cane, wheat, and pulses), percentage of cropped acreage devoted to non-reported crops (that i s , other than r i c e , j u t e , sugar cane, wheat, or pulses), and percentage of farm 74 area sharecropped. The groupings of d i s t r i c t s suggested by th i s analysis are shown in Chart II. It can be seen that none of these groups correspond to any of the groups i den t i f i ed in Chart I on the basis of the degrees of sharecropping cu l t i v a t i on in the d i s t r i c t s . This goes to show that although two d i s t r i c t s within the country may be quite s imi la r in terms of the pattern of land tenure, there may be considerable var iat ions in terms of other important var iables. Any pol icy which attempts to a l t e r the size-tenure structure for achieving spec i f i c goals of development in a country has to be sens it ive to the regional var iat ions within the country. The above. L. Guttman and J.C. Lingoes, The Guttman-Lingoes Nonmetrio Program Series, Ann Arbor: Mathesis Press, 1973. The relevant data have been put in the Appendix. See Table A l , pp. 268-270. 134 Chart 2. . Mult ip le Scalogram Analysis for the Regionalization of Bangladesh Agr iculture .CHITTAGONG HILL TRACTS .COMILLA •KUSHTIA, SYLHET| I .BANGPUB | .NOAKHALI | BAJSHAHI. *DINAJPUR, | JESSORE, ! BOGRA DACCA, + .PABNA MY KENSINGH •FARIDPUR j .CHITTAGONG .KHULNA .BARISAL Note: ' Represents locat ion of a s ingle d i s t r i c t . + Represents locat ion of' two d i s t r i c t s . * Represents location of three d i s t r i c t s . 135 exercise in regional iz ing Bangladesh agr i cu l ture , although less than comprehensive in nature (which was inev i table given the lack of a v a i l a b i l i t y of necessary information), can be of use in formulating developmental po l i c i e s which are sens i t ive to the 75 relevant local character i s t i c s . One may note here that for a f u l l e r treatment of the reg iona l izat ion of Bangladesh agr icu l ture, which would help micro- level implementation of development p o l i c i e s , one would require more detai led information on topographic and c l imat ic variables as well as on factors l i k e i r r i g a t i on f a c i l i t i e s , communication systems, marketing serv ices, and rural-urban interactions in trade, research and t ra in ing . 136 IV. PROBLEMS OF SIZE-TENURE STRUCTURE IN BANGLADESH AGRICULTURE It was noted in Chapter II that the problem of agr icu l tura l development within a region can be analysed in terms of the character i s t i c s of four d i f fe rent c lusters of elements which define the ag r i cu l tu ra l system of that r e g i o n J To recap i tu la te , these clusters include the natural cu l t i va t i ng environment, resources owned or contro l led by the cu l t i v a t o r , soc ieta l values and expectations, and linkages of the agrarian society to the outside market. It was argued that the resource posit ion of the cu l t i va tor s was of major importance in inf luencing the other c lu s te r s , both in the short and the long run. In what follows in th i s chapter, an attempt i s made to understand how the d i s t r i bu t i on of the ownership and control of land resources (as ref lected in the size-tenure structure) among the agr icu l tura l households of Bangladesh affects the performance of the primary sector in terms of equity in the d i s t r i bu t i on of income and e f f i c i ency in the process of production. In Bangladesh, where land has occupied a unique status as a resource c r i t i c a l to basic productive and subsistence a c t i v i t i e s , the major determinant of the d i s t r i bu t i on of income-earning See pp. 20-21. 137 opportunities has been the nature of the ex i s t ing size-tenure structure. The size-tenure structure has also affected the productive e f f i c i ency of farms by inf luencing such factors as the level of employment, nature of technology, and the in tens i ty of input-use, a l l of which can be considered from the fol lowing two points of view: ( i ) the micro-static point of view dealing with the operation of indiv idual households at a point in time, and ( i i ) the macro-dynamic point of view dealing with the operation of the agr i cu l tu ra l sector as a whole over time. The present chapter w i l l begin by re la t ing the problem of unequal d i s t r i bu t i on of income to size-tenure structure in Bangladesh agr icu l ture. This w i l l be followed by an analysis of the micro - s tat ic as well as macro-dynamic e f f i c i enc i e s of d i f fe rent tenure and s ize groupings of farms, which create several problems for potential in ra i s ing the level of product iv i ty , increasing the marketable surplus, creating more employment opportunit ies, and adopting newer technologies. Problem of Unequal D i s t r ibut ion of Income The extent of inequal i ty in the ownership of land among rural households in Bangladesh has been noted in the previous chapter. To recap i tu la te , the data presented in Table 5 (page 97 ) showed that in 1977 about 53% of the households, each with an ownership unit of 138 not more than one acre, covered only about 9% of the to ta l land area under a l l households. On the other hand, ownership units of more than 10 acres, which were owned by only about 2% of the households, covered about 20% of the to ta l land. This i s the picture of inequal i ty that one gets when the comparison i s l imited to land-owning households. The fact that in 1977 about 1.3 m i l l i o n households had no land at a l l adds to the severity of the inequal i ty (see Table 6, page 99 ). A l so, when one considers the fact that about a th i rd of a l l rural households had no land avai lable for cu l t i va t i on (some of them with jus t the homestead land), whereas about a f i f t h of the to ta l farmland was owned by less than 2% of the households, one can better appreciate the problem of inequal i ty in 2 the d i s t r i bu t i on of ownership of farmland. Although the pract ice of renting land has become widespread in Bangladesh ( in 1977 as many as 38.83% of the households rented some land), th i s has not reduced to any appreciable extent the d i f f e r e n t i a l in income-earning opportunities between the r i c h and the poor land owners. The households which rented land, according to the 1977 Land Occupancy Survey, managed only to end up with an average farm s ize ( including owned and rented farmland) of 2.72 acres. Although about a quarter of a l l farmland was leased out, there were too many households looking for a portion of i t . Therefore, the average farm s ize of the rec ip ients Calculated from figures provided by Government of Bangladesh, Summary Report of the 1977 Land Occupancy Survey of Rural Bangladesh, Dacca: Bangladesh Bureau of S t a t i s t i c s , 1977, table II. 139 of rented land could not be raised to any appreciable degree. One addit ional point to be noted here i s that only about 19% of a l l rental contracts were obtained by the households without any farmland of t he i r own. As reported in Table 9 (page 106), in about 99% of the rental contracts the rent to be paid to the landlord was f ixed at 50% or more of the output obtained from the rented land. In add i t ion, rental contracts covering more than 11% of the tota l rented land were found to require extra cash payments to the landlords over and above the 50% or higher rental rate in terms of physical outputs. A l l of these indicate that the inequal i ty in income-earning opportunity on farmland, as ref lected in the pattern of ownership of farmland, was hardly a l tered by the operations of the rental market. The problem posed by the inequal i ty of land ownership in rural Bangladesh i s compounded by the very l imi ted employment opportunities avai lable outside agr icu l ture. Only about 25% of the to ta l labour force i s non-agr icu l tura l , and a s izable part of th i s percentage i s accounted for by the r e l a t i v e l y educated ind iv iduals who come mostly from land-r ich households. Thus, the inequal i ty in land ownership in rural Bangladesh remains a pers istent obstacle in the way of a broader d i s t r i bu t i on of income opportunit ies. The possible reasons for the landlord ' s preference to rent out land to the land-owning households rather than the landless have already been discussed in the preceding chapter. See p. 103. The numerous government and semi-government i n s t i t u t i on s and business organizations in the urban sector of Bangladesh hire a considerable amount of r e l a t i v e l y s k i l l e d labour. 140 Land Tenure and Agr icu l tura l Product iv i ty It has been shown e a r l i e r (see Table 7, page 101) that in Bangladesh the two major tenure arrangements are owner cu l t i v a t i on by family labour and cu l t i va t i on under sharecropping arrangements. One can discuss the re l a t i ve productive e f f i c iency of these arrangements e i ther in a micro-s tat ic context or in a macro-dynamic context. In what follows in th i s section a micro - s tat ic analysis i s f i r s t attempted. Some theoret ica l considerations are put forward, which are analysed in the l i g h t of quant i tat ive information. The micro - s ta t i c analysis i s then followed by a macro-dynamic evaluation of the land tenure pattern in Bangladesh. M ic ro- s tat ic analysis Theoretical Premise Conventional Theories: If productive e f f i c i ency i s measured by output per unit of land (which makes sense in a land-scarce s i tuat ion) then t r ad i t i ona l theory would lead one to believe that sharecropping tenancy (referred to as tenancy from now on) would be expected to resu l t in lower productive e f f i c i ency than would be the case under owner cu l t i v a t i on . A l f red Marshall made a s i gn i f i cant contribution in extending th i s hypothesis. A. Marshal l , Principles of -Economies, New York: The Macmillan Company, 1948. 141 Considering the usual 50% share of the tota l output that a tenant pays to the landlord as rent, he observes: " . . . when the cu l t i v a to r has to give to his landlord half of his returns to each dose of cap i ta l and labour that he applies to the land, i t w i l l not be to his interest to apply any doses the to ta l return to which is less than twice enough to reward him."6 The Marshall ian argument is thus that sharecropping arrangements are destined to resu l t in less output per unit of land than would be produced under owner cu l t i v a t i on . This argument has subsequently been extended and supported by several other economists; l e t us bracket them together and refer to them as School I J The Marshall ian concept has not, however, gone unchallenged. Cheung, for example, has argued that the intens i ty of input-use under sharecropping arrangements can be expected to be the same as under owner cu l t i v a t i on . 6 Ibid., p. 644. 7 These economists include Bardhan, B e l l , Srinivasan and Zusman. See P.K. Bardhan and T.N. Sr inivasan, 'Crop-sharing Tenancy in Agr icu l ture: A Theoretical and Empirical Ana l y s i s ' , American Economic Review, Vo l . 61, No. 1, March 1971; and C. Bel l and P. Zusman, 'A Bargaining Theoretic Approach to Cropsharing Contracts ' , American Economic Review, Vol. 66, No. 4, September 1976. See S.N.S. Cheung, The Theory of Share Tenancy, Chicago: The Univers ity of Chicago Press, 1969. 142 We shal l refer to Cheung and to those who have supported his l i ne 9 of reasoning as School II. Before one can judge the r e l a t i ve merits of the conclusions reached by these two d i f fe rent schools of thought outl ined above, i t i s important to look more c lose ly at the basic arguments being made by each one of them. These are as fol lows. School I: For the purpose of exposition one can wr ite a simple production function of a tenant farm in the fol lowing way: Q = f(L,N) (1) where, Q = output, L = land, and N = labour. Now consider a tenant farm-family which i s t ry ing to maximize i t s income. If the opportunity cost of family,labour employed by the tenant i s taken to equal the going wage rate then the net income equation for.the tenant can be written as: Y = Qp - wN - rQp (2) Apart from Cheung, Newberry and S t i g l i t z pursued an ant i -Marshal l ian l i n e . See D.M.G. Newberry, 'The Choice of Rental Contract in Peasant A g r i c u l t u r e ' , in L.G. Reynolds (ed.), Agriculture in Development Theory, New Haven: Yale Univers ity Press, 1975; and J.E. S t i g l i t z , ' Incentive and Risk Sharing in Sharecropping', Review of Economic Studies, Vol. 41, No. 2, Ap r i l 1974. 143 where, Y = tenant 's net income, p = pr ice per unit of output, w = wage ra te , and r = proportional rent ( in kind) to be paid to the landlord (0 < r < 1). The tenant seeks to maximize Y subject to the production function given in equation (1). Assuming a f ixed supply of land to the tenant in the short run, the condition to be s a t i s f i ed for maximization of income for the tenant requires that the par t i a l der ivat ive of the income equation, that i s , equation (2), with respect to N be set equal to zero. 3N = 0 or 9 (Qp - wN - rQp) = 0 or w = (1 - r) p (3) Equation (3) can be interpreted as fo l lows: If 50% of the produce is to be given away as rent to the landlord (that i s , when r =1/2) then the tenant applies labour to the rented 144 land up to the point where value of the marginal product of labour i s twice the wage rate. Under owner cu l t i v a t i on the income equation takes the fol lowing form: Y 1 = Q p^ - Nnw (4) The condition to be s a t i s f i ed for maximization of income here i s , 8 Y1 o r w = p ( 5 ) , . wage _ value of the marginal a 1 S ' rate product of labour Therefore, under owner c u l t i v a t i o n , labour i s p ro f i tab ly applied up to the point where marginal product of labour i s jus t equal to the wage rate. Comparing equation (5) with equation (3) one f inds that labour input is less intens ive ly applied on tenant cu l t i vated land than on owner cu l t ivated land. The Marshallians use s imi la r arguments to show that u t i l i z a t i o n of a l l other inputs per unit of sharecropped land would be less intensive when compared with the corresponding f igures for owner 145 cu l t ivated land. The r e su l t , they suggest, would be a lower productive e f f i c iency under a tenancy arrangement. Some authors who accepted the basic argument of the Marshall ian school (that sharecropping resu lts in r e l a t i v e l y lower levels of productive e f f i c iency when compared to owner c u l t i v a t i on ) , have suggested that i f the var iable costs of production were shared between landlord and tenant in the same proportion as output i s shared then the 'Marshall ian d i s incent i ve ' of tenant cu l t i v a t i on would disappear, and tenancy would resu l t in productive e f f i c iency comparable to that under owner c u l t i v a t i o n . ^ The argument can be 11 aided with the help of a diagram. In Figure 3, PQ represents the marginal value product (MVP^) of a tenant farm. The input in question i s measured along the X-axis. Prices of the product and the input are taken to be constant and exogenously given. The marginal factor cost (MFC) incurred in using the input i s given by AB, which i s pa ra l l e l to the X-axis. OA i s the unit pr ice of the factor and since th i s pr ice remains the same a l l along, the MFC curve stays pa ra l l e l to the X-axis. See, for example, E.O. Heady, Economics of Agricultural Production and Resource Use, Englewood C l i f f s : Prent ice-Hal l Inc., 1952. * We have t r i e d to;elaborate upon the analysis done by Adams and Rask. See D.W. Adams and N. Rask, 'Economics of Cost-Share Leases in Less Developed Countr ies ' , American Journal of A g r i c u l t u r a l Economics, Vol. 50, No.4, November 1968. 146 Figure 3 . Tenant Cu l t i va t ion , Cost-Sharing and Productive E f f i c iency Under owner c u l t i v a t i o n , production w i l l take place at a level where an amount of the variable input i s used jus t s u f f i c i en t to equalize the marginal value product and the marginal f ixed cost. In Figure 3 , OR i s the level of variable input-use at which both marginal value product and marginal factor cost are equal to OA. 147 Now consider the operation of a tenant who does not share any of the cost of production with his land-owning partner. Following Adams and Rask, leasing of land without cost-sharing 12 can be ca l led " t r ad i t i ona l lease". If the tenant has to pay a 50% share of the to ta l output from sharecropped land as rent, then his marginal value product (MVP^) would be represented by NQ where N would be the mid-point on OP. Under t r ad i t i ona l lease, the marginal factor cost facing the tenant farms would be AB. Therefore, the level of var iable input-use would be given ,by OS, where both marginal value product and marginal factor cost would be equal to OA. The level of var iable input-use in th i s case would, thus, be lower than the level attained under owner cu l t i v a t i on by an amount equal to SR. What i f the cost of the input i s shared between the landlord and the tenant in the same proportion as the output is shared (th is 13 has been ca l led the " ideal lease" by Adams and Rask)? In Figure 3, CD, which i s equidistant from OQ and AB, represents the marginal factor cost facing the tenant under ideal lease who pays for only half of the var iable input in use and gets half of the produce as his share. Under such circumstances the var iable input w i l l be used at the level OR (since the tenant 's marginal value product curve and his 2 Ibid., p. 935. 3 ibid.j p. 935. 148 marginal f ixed cost curve intersect at H, which by the very assumptions underlying the drawing of the curves,wi l l l i e on FR). Therefore, i f ideal lease could be adopted, then the appl icat ion of variable inputs could be carr ied to the point where the owner-cu l t i va to r would set his use of those inputs. The tota l net return (to the tenant and landlord combined) under t r ad i t i ona l and ideal leases would be: Net return under . = ( , t t ) _ ( t o t a l c o s t ) t r ad i t i ona l lease OPTS - OAES APTE Net return under ideal lease . 4.x ["cost incurred"] ( tota l output) - L 5 y t e n a n t J T c o s t i n c u r r e d ! |_ by l a n d l o r d J OPFR - OCHR - CAFH OPFR - OAFR APF APF i s . c l e a r l y greater than APTE. That i s , to ta l net return to both the landlord and the tenant i s higher when ideal lease i s adopted as compared to the results obtained under t r ad i t i ona l lease. This i s so because the marginal costs faced by the tenant and the 149 owner-cultivator are equal under an ideal lease so that the levels of input-use and product iv i ty of the tenant reach levels equivalent to those of the owner-cultivator. tenant farm given in equation (1), Q = f (L ,N) ; with Q, L and N representing output, land, and labour respect ively. L i s now considered to be a var iab le, the value of which i s given by, where, H = tota l landholding of a l l tenant farms, and M = number of tenant farms. If the value of H is given, as would be the case in the short run, L i s determined by the value of M. It i s postulated that the number of farms can become a var iable even in the short run since each landlord is capable of deciding upon the number of tenant farms he would create, thereby inf luencing the value of M. Value of to ta l rent earned by a l l landlords, R, i s given by, School II: Assume the same simple production function of a L H/M (6 ) R = MrQp (7) (as before, r and p stand for proportional rend-in-kind and pr ice per unit of output respect ive ly) . 150 The fol lowing re lat ionsh ip would hold true under competitive assumptions: wN = (1 - r) Qp (8) It i s assumed that in making the sharecropping contract the landlord has the f i n a l say about the values of r and N. The landlord, therefore, is seeking to maximize R through the choice of M, r, and N, subject to equation (8). To f i nd out the conditions for th i s maximization one can construct the fol lowing Lagrangean, Z = MrQp - A |wN - (1 - r) Qpj (9) Taking pa r t i a l der ivat ives of the expression Z with respect to M, r, N, and x, and sett ing them a l l equal to zero ( for obtaining the necessary conditions for maximization) one gets the fo l lowing: -§T= rQp + rMp -fg-+ A (1-r) ^ = 0 (10) -|^- = MQp - A Qp = 0 (11) „ r p _ j j | _ _ , w + x ( 1_. r ) .^-o (12) 151 _|L = - JwN - (1-r) Qp] =0 (13) Noting that L = H/M so that 8L _ H , and that x = M (from equation 11), i t i s possible to s impl i fy equation (10) in the fol lowing manner: -4"] - < ^ > [ - | f ^ - ] p - 0 or, or, rQp = pL or, rQp_ = _dQ_ ( 1 4 ) L 8 L value of to ta l rent from a tenant farm = value of the margi-That i s , to ta l land area of the farm nal product of land, or that, value of rent per unit of land equals the value of the marginal product of land. 152 Again, noting that A = M, one can s impl i fy equation (12) to get, p-i- <i5> value of the marginal = . o r ' production of tenant labour wage rate. Going back to the condition for maximization of income by an owner-cult ivator, which was given in equation (5), i t i s found that the condition there is the same as the one established in equation (15) for maximization of income under tenant farming. Therefore, as Cheung and his followers maintain, the intens i ty of input-use under sharecropping would be the same as under owner cu l t i v a t i on . An Evaluation of Conventional Theories: One can best appreciate the merit of the Marshall ian argument in a s i tuat ion where the tenants are above the marginal level of subsistence. In such circumstances arguments of poor incentive on the part of the tenant may be v a l i d . However, the Marshallian argument (of r e l a t i ve i ne f f i c i ency of sharecropping cu l t i v a t i on as compared to owner cu l t i va t ion ) may not necessari ly hold true in s i tuat ions of high population density where the pressure on land results in very small owner cu l t i vated farms as well as very small 153 sharecropped units. These sharecroppers, who are faced with very few a l ternat ive employment opportunit ies, have to put a l l t he i r e f for t s into the cu l t i va t i on of sharecropped land to meet basic family needs from the i r share of the tota l produce. It may be true that impoverished tenants w i l l not be in a posit ion to invest much cap i ta l in the various stages of ag r i cu l tu ra l production, but they w i l l have to use avai lable family labour as intens ive ly as possible i f basic needs are to.be met. In other words, the existence of subsistence requirements which pers i s t i r respect ive of production-organization, coupled with very low opportunity cost of labour, may cause cu l t i v a t i on under sharecropping tenancy to be as labour- intensive 14 as under owner cu l t i v a t i on . Insofar as i t i s true that small owner-cu l t i va to r s , l i k e small tenants., are handicapped by a shortage of resources to. invest in land improvement projects or in qua l i ty inputs, i t i s conceivable that the productive e f f i c i enc i e s (measured in terms of output per unit of land) of the owner-cultivators and tenants w i l l show l i t t l e d i f ference. The argument of some Marshall ian scholars that to ta l net returns to both landlord and tenant could be increased i f an i d ea l , instead of a t r a d i t i o n a l , lease were adopted does not necessari ly hold true under 15 a l l circumstances. An exploration of the conditions under which both That th i s i s so in Bangladesh has been the observation of Hossain. See Hossain (1977) op.dit., p. 313. Note here that the t r ad i t i ona l lease has been defined as a 50-50 sharing of output between the landlord and the tenant without any cost-shar ing, and the ideal lease as 50-50 sharing of both costs and output. 154 parties can gain by a s h i f t from t rad i t i ona l to ideal lease i s in order. If 'economic rent ' for the tenant i s defined as an output or payment in excess of the minimum supply pr ice needed to keep the use of the input at the given level then, by Figure 3 (page 146) economic rent for the tenant , under t rad i t i ona l lease A I N t ' a n a economic rent for the tenant •_ „ N H under ideal lease Since CNH i s greater than ANE, the tenant w i l l benefit from the s h i f t from t rad i t i ona l to ideal lease. In order to f ind out how the landlord i s affected by the s h i f t from t rad i t i ona l to ideal lease, one has to estimate both the cost that would be incurred by him in the process and the amount by which his share of the proceeds would increase. If the l a t t e r i s greater in value than the former, he would benefit more by cost-sharing, otherwise not. Going back to Figure 3, the landlord has to bear a to ta l cost equal to CAFH i f he makes a contract for ideal lease. Out of CAFH, the area CAEG i s equivalent to an income transfer from the landlord to the tenant. The explanation i s as fol lows. The s h i f t from t rad i t i ona l to ideal lease creates the addit ional output STFR. The extra cost for producing th i s addit ional output i s SEFR. Ideal lease requires that out of th i s extra cost the landlord bears an amount given by GEFH. Under t r ad i t i ona l lease the tenant was already prepared to produce OPTS. Therefore, at that time the tenant was incurr ing a cost of OAES. By bearing a cost amounting to CAFH the 155 landlord in an ideal lease i s , in e f f e c t , t ransferr ing CAEG amount of his income to the tenant. The addit ional output that the landlord receives due to changing over to ideal lease from the t r ad i t i ona l kind i s given by the fo l lowing: addit ional output going to landlord due to s h i f t from = t rad i t i ona l to ideal lease increase in output due to adoption of ideal lease amount taken by tenant from addit ional output = STFR - SEHR = ETFH If the area ETFH is not greater than CAFH then the landlord would not be interested in transferr ing from t rad i t i ona l to ideal lease. In other words, the condition for involving the landlord in ideal lease i s (ETFH > CAFH). The area ETFH i s given by .£(1/2 OA.SR) + (1/2 CA.SR)j , and the area CAFH equals (OS.CA + CA.SR). By s impl i fy ing these expressions for the two d i f fe rent areas one can write the condition (ETFH > CAFH) as (SR>2 0S). That i s , Condition necessary for landlord to benefit : SR>2 OS from ideal lease 156 What the above condition means i s the fo l lowing, Change in the level of input-use due to transfer from t r ad i t i ona l to ideal lease has to be greater than twice the level of input-use under t r ad i t i ona l lease. Figure 4 t r i e s to show that th i s condition i s more l i k e l y to be s a t i s f i ed by r e l a t i v e l y f l a t marginal value product curves. In th i s f i gu re , when PQ represents the marginal value product curve for the farm, the condition to be s a t i s f i ed for involving the landlord in cost-sharing i s (SR>2 OS). This has not been f u l f i l l e d in Figure 4. When the marginal value product curve i s represented in the same f igure by the r e l a t i v e l y f l a t l i ne P nQ, the condition takes the form (S-|R.j>2 OS^). Even here the condition i s not quite s a t i s f i e d . But when the marginal value product curve becomes as f l a t as P^ Q the condition for landlord ' s par t i c ipat ion in ideal lease gets f u l f i l l e d . In other words, we get ^R , , >2 OS^). The question now i s —when would we encounter r e l a t i v e l y f l a t marginal value product curves so that the landlords may gain from a switch from t rad i t i ona l to ideal lease? This would occur when a par t i cu la r level of input-use i s associated with a r e l a t i v e l y low marginal value product. Given an exogenously determined pr ice level for the output, th i s would mean that the chances of gain for the landlord in making the switch would be greater in those cases where the marginal product does not vary so appreciably with changes in the level of inputs-use. 157 Figure 4. Marginal Value Product in Tenant Cu l t ivat ion and the Landlord's Par t i c ipat ion in an Ideal Lease The arguments extended by the ant i -Marshal l ian school on the other hand are based on assumptions which may not hold true in many cases. One of the basic assumptions made by the scholars associated with th i s school i s that under any sharecropping contract the amount of labour applied by the tenant on rented land i s determined by the landlord. This would require that the tenant 's operations are 158 s t r i c t l y supervised by the landlord, which may not be the usual practice in par t i cu la r instances. One cannot deny the fact that the landlord may influence the amount of labour applied per unit of the rented land by specifying the crops to be sown, seed var ie t ie s to be selected, and agro-chemicals to be used. However, even in cases where such spec i f icat ions are found to be a part of the rental contract, i t would be un rea l i s t i c to assume that the landlord, through the very act of making these spec i f i ca t ions , controls the amount of labour applied on rented land since each of these conditions imposed on the tenant can be f u l f i l l e d with the appl icat ion of varying levels of labour input. Besides, the f inanc ia l status of the potential tenant in many cases may not be such that conditions which imply higher capita l investments by the tenant ( for example, the use of more agro-chemicals to augment the amount of labour used per unit of land) would be pract icable. The anti-Marshal l ian analysis rests also on the contention that r, the proportional rent, i s determined exclusively by the landlord. This assumption can be accepted in those circumstances where the landlord 1 has monopolistic control over the land-rent market. However, there may be s i tuat ions in which the landlords are more numerous than would enable them to exercise monopolistic or o l i g o p o l i s t i c control over the market. In these s i tuat ions i t w i l l be a mistake to argue that the landlord can set the value of r as he pleases. 159 Empirical Findings Returning to the par t i cu la r circumstances of land tenure in Bangladesh reviewed in the preceding chapter, one can see that the assumption made by the anti-MarshalTian school with regard to the control exercised over r by the landlord i s not generally v a l i d , given that there are numerous land owners renting out land. Also, in the context of Bangladesh ag r i cu l tu re , the sharecropper i s found to make his production decisions on the rented land largely independent of the d i rect ion or supervision, of the landlord. In Bangladesh the average farm s ize of the owner-cum-tenants who obtained more than 80% of the to ta l rented land in 1977 was 3 acres according to the 1977 Land Occupancy Survey. This indicates that many sharecroppers in Bangladesh are operating above the level of subsistence. Cost-sharing has not been a common pract ice in the sharecropping arrangements found in Bangladesh (see Table 11, page 109). Under these circumstances, i t w i l l be worthwhile to investigate in the Bangladesh context the v a l i d i t y of the Marshall ian argument that sharecroppers would make less intensive use of the various inputs and would end up with lower output per unit of land than would have occurred under owner cu l t i v a t i on . In proceeding to make th i s The 'subsistence farm s i z e ' can be found out by d iv id ing the product of the number of family members and per capita subsistence needs ( in agr i cu l tura l output) by average y i e l d per acre. Per capita subsistence need has been taken as .170 ton of r i ce fol lowing B l a i r (see H.W. B l a i r , The Elusiveness of Equity: I n s t i t u t i o n a l Approaches to Rural Development in Bangladesh, Ithaca: Rural Development Committee, Cornell Univers i ty, 1974, p. 12). The subsistence s ize works out to be about 2.5 acres. 160 invest igat ion, f i r s t a regression analysis w i l l be attempted and then an i n t e r d i s t r i c t study undertaken. Regression analys is : Data avai lable allowed the estimation of the fol lowing simple regression equations to test the re lat ionship between sharecropping cu l t i v a t i on and intens i ty of input-use:^ 7 a 1 + b 1 T + e.j ( i ) = a 2 + b 2 T + e 2 ( i i ) where, M = farmyard manure used per unit of net sown area, T = percentage of t o ta l cu l t i vated area under sharecropping tenancy, F = chemical f e r t i l i z e r used per unit of net sown area, and e-| e 2 = error terms. These regression equations were estimated by using data on the use of farmyard manure and chemical f e r t i l i z e r s per unit of net down area in seven d i f fe rent d i s t r i c t s of Bangladesh (Rajshahi, Kushtia, Bogra, Ba r i s a l , Dacca, Mymensingh and Comilla) which were ava i lab le in Government of Bangladesh, Farm'Management Research in Retrospect, Dacca: Min ist ry of Agr icu l ture, 1973. Information on the incidence of tenancy in these d i s t r i c t s has already been reported in Table 15, (see p. 130). 161 Estimation of regression equation ( i ) y ielded the fol lowing resu l t s : M 22.68 - 0.8198 T (-3.16) R 2 .6657 The sign of the coe f f i c i en t of T has come out negative ind icat ing an inverse re lat ionsh ip between the level of tenancy and the in tens i ty 2 of manure-use. The value of R shows that about 67% of the var iat ion in manure-use has been explained by var iat ions in levels of share-cropping tenancy. This represents a good f i t of the data. Probabi l i ty error for the estimated equation works out to be .025, which means that the coe f f i c ient of T i s s i g n i f i c an t l y d i f fe rent from zero allowing 2.5% error. A l l th i s allows us to conclude that in estimating regression equation ( i ) the extent of sharecropping tenancy has been found to be a s i gn i f i cant var iable in determining the use of farmyard manure. Intensity of manure-use has been low in areas of high levels of share-cropping tenancy as compared to the in tens i ty demonstrated in areas of low levels of tenancy. The re l a t i ve i n tens i t i e s with which chemical f e r t i l i z e r s were used in areas of d i f f e r i n g levels of sharecropping tenancy are revealed by the fol lowing results obtained in estimating regression equation ( i i ) : F = 21.58 - 0.8765 T (-3.24) R2 = .6770 162 Again, the coe f f i c i en t of T has come out negative suggesting an inverse re lat ionsh ip between the extent of sharecropping 2 tenancy and the use of chemical f e r t i l i z e r s . Value of R here shows that about 68% of the var iat ion in f e r t i l i z e r use can be explained by var iat ions in levels of sharecropping tenancy. Probabi l i ty error in th i s case i s the same as i t was in the previous estimated equation, that i s , .025. Therefore, extent of share-cropping tenancy can be taken as a s i gn i f i cant variable in determining the in tens i ty of the use of chemical f e r t i l i z e r s . The obvious simple regression equation that can help in test ing whether there i s an inverse re lat ionsh ip between the extent of tenancy and product iv i ty per unit of land can be written as, C a 0 + b-T + e ( i i i ) where, C product iv i ty per unit of land, T percentage of t o ta l cu l t i vated area under sharecropping tenancy, and e 3 error term. Data for estimating regression equation ( i i i ) came from the same source as used for estimating equations ( i ) and ( i i ) . 163 Results obtained in estimating regression equation ( i i i ) were the fo l lowing: C = 702.9 - 13.00 T (-1.65) R2 = .3517 Although the negative sign of the coe f f i c i en t for T conforms 2 with theoret ica l expectations, the value of R i s not sat i s factory . Besides, the probab i l i ty error i s found to be .10, which means that the re lat ionsh ip would be va l i d i f one allows for an error as high as 10%. While the results obtained in estimating the above simple regression equations are indicative of the underlying re lat ionships among the variables involved, one should note that they do not provide conclusive support. It would have been more desirable to use data on larger sample sizes in estimating these equations. Such data were, unfortunately, not ava i lab le. Estimation of mu l t ip le , rather than simple, regression equations might also have increased the explanatory power of the regression models. However, problems of muTt ico l l inear i ty would l i k e l y af fect the qua l i ty of the results in that case. An a l ternat ive way to proceed in test ing the effects of tenancy on input-use and product iv i ty i s presented in the fol lowing i n t e r -d i s t r i c t study. 164 Bogra and Rajshahi: An i n t e r d i s t r i c t study. Bogra and Rajshahi are two neighbouring d i s t r i c t s in the northwestern region of Bangladesh. Of the two, Bogra i s smaller in s ize (with a land area of 1 ,502 square miles as opposed to 3,654 square miles for Rajshahi). The land tenure patterns in these d i s t r i c t s are quite d i s s im i l a r in that the percentage of farmland under sharecropping tenancy in Rajshahi i s almost twice as large as in Bogra. Thus, a comparison of the agr i cu l tura l production behaviour in these two d i s t r i c t s would suggest an opportunity for judging the effects of varying degrees of sharecropping cu l t i va t i on on land use and p roduct i v i t i e s . The d i s t r i c t s of Bogra and Rajshahi were chosen as sample areas for th i s study also because they were quite s imi la r to each other in respect to topography, s o i l composition, and c l imat ic conditions. I f one or more of these factors were s i g n i f i c an t l y d i f fe rent in the two d i s t r i c t s then i t would be d i f f i c u l t to segregate the effects of the d i f f e r e n t i a l impacts of land tenure patterns on production behaviour. The time period of data for th i s study i s 1959-60. Selection of th i s pa r t i cu la r time period was necessitated pr imar i ly by the 18 a v a i l a b i l i t y of data. Data used in th i s study came mainly from the fol lowing sources: i . Government of Pakistan, Population Census of Pakistan, 1961: District Census Report, Bogra, Karachi: Min istry of Home A f f a i r s , (undated), i i . Government of Pakistan, Population Census of Pakistan, 1961: D i s t r i c t Census Report, Rajshahi, Karachi: Min istry of Home A f f a i r s , (undated). i i i . Government of Pakistan, Pakistan Census of Agriculture, 1960, Vol. I, Final Report - East Pakistan, Part I, October 1962. 165 In th is study a discussion of the phys ica l -c l imat ic conditions as well as the land tenure patterns in the two d i s t r i c t s w i l l be followed by a comparison of the production behaviour and an analysis of the reasons for the demonstrated d i f f e r en t i a l in agr icu l tura l production performances in these areas. Phys ica l -c l imat ic conditions Soi l composition: Both Bogra and Rajshahi are part of the great a l l u v i a l plains of Bangladesh. Information on s o i l composition, which i s avai lable in Map 2 (page 115), has been magnified for Bogra. and Rajshahi in Maps 7 and 8 on the next page for getting a c learer picture of the comparability of s o i l in the two d i s t r i c t s . As can be seen, s o i l s c l a s s i f i e d as sandy loams (doash) are the dominant so i l - type and cover s imi lar proportions of to ta l land in both d i s t r i c t s . These s o i l s are quite f e r t i l e , being of an intermediate qua l i t y between more f e r t i l e s i l t s and less f e r t i l e c lays. Red and yellow so i l s {khiar or lalmati) are the next major s o i l constituents of both Bogra and Rajshahi. These so i l s are part of the old a l l u v i a of the Barind t rac t and are of a r e l a t i v e l y low level of f e r t i l i t y . Red and yellow so i l s are very hard when dry and hence become d i f f i c u l t to plough. Maps 7 and 8.would show s imi la r proportions of tota l land in Bogra and Rajshahi composed of these s o i l s . S i l t loams {palimati) are the next important constituent of the s o i l s of the sample areas, which again Map 7. So i l Composition of Rajshahi Map. 8. Soi l Composition of Bogra 167 const itute comparable proportions of tota l land area in the two d i s t r i c t s . These so i l s are composed of fresh s i l t which i s deposited after the water in an area inundated by a r i ve r subsides. The r i ve r Jamuna inundates the eastern border of Bogra, thereby accounting for the s i l t loams in i t s s o i l . The s i l t loams in Rajshahi along the southwestern border of the d i s t r i c t are the result of the inundations of the r i ve r Padma. These s o i l s are highly f e r t i l e and can be ploughed with r e l a t i ve ease. A very small proportion of land in Rajshahi i s composed of clays {matial). These so i l s become hard during the dry season and st icky during the monsoon. Consequently, ploughing these so i l s i s a d i f f i c u l t task. To quantify the areas covered by so i l s of broad f e r t i l i t y groups in the two d i s t r i c t s , the proportion of so i l s that are r e l a t i v e l y more f e r t i l e (that i s , s i l t s and sandy loams) const itute about 60% of the land of each d i s t r i c t , while the other 40% i s composed of so i l s of r e l a t i v e l y low f e r t i l i t y (that i s , clays and red/yellow s o i l s ) . Cl imatic condit ions: As shown in Map 3 (page 117), both Bogra and Rajshahi belong to the dry region of the country. According to a study conducted by Manalo, the average annual r a i n f a l l fo r a seventy-year period (from 1902 to 1972) was 65 inches for Bogra and 58 inches 19 for Rajshahi. During 1959-60, which i s the time period under E.B. Manalo, Agvo-Climatio Survey of Bangladesh, Bangladesh Rice Research In s t i tu te , (undated), pp. 48, 50. 168 consideration, these f igures were 65 inches and 56 inches respectively. Data on the temperature and humidity experienced by the two d i s t r i c t s during 1959-60 were not ava i lab le. However, information on these important c l imat ic factors were avai lable for the period 1966-70 and are presented in Table 16. The data on c l imat ic conditions show that Bogra and Rajshahi have very s imi la r cl imates. Bogra enjoys a r e l a t i v e l y greater amount of r a i n f a l l than Rajshahi.; but the difference (about .75 inch per month) does not appear very s i gn i f i can t . Table 16. . Temperature and Humidity in Bogra and Rajshahi (averages for the period 1966-70) Bogra Rajshahi Average annual temperature ( F) - maximum 106 108 - minimum 44 45 Average annual humidity 78 (percentage) 78 Source: Government of Bangladesh, Statistical Digest of Bangladesh, No. 8, Dacca: Bangladesh Bureau of S t a t i s t i c s , 1972. 169 Land Tenure The basic facts on land tenure in the two d i s t r i c t s are summarized in Table 17. To determine the extent of tenant cu l t i va t i on in the two d i s t r i c t s , one needs to estimate the amount of land that i s rented in both by tenants and by owner-cum-tenants. Although the or ig ina l data did not provide information on the percentage of to ta l farm area rented by owner-cum-tenants, these were calculated with the help of data on, ( i ) tenant operated farm area: covering a l l area under pure tenants and area which i s rented by owner-cum-tenants, ( i i ) to ta l farm area of pure tenants, ( i i i ) to ta l farm area of owner-cum-tenants. . Areas owned and rented by the owner-cum-tenants can now be calculated as fo l lows: Area rented by owner-cum-tenants = ( i ) - ( i i ) Area owned by owner-cum-tenants = ( i i i ) - [ ( i ) - ( i i ) ] The to ta l farm area under tenancy i s obtained by simply adding the area rented by the owner-cum-tenants to the area covered by the pure tenants. The relevant f igures for Bogra and Rajshahi are presented in Table 18. This table shows that Rajshahi has a much higher proportion of farmland under tenancy than Bogra. The reason for th i s may l i e in the disparate h i s to r i ca l experiences of the two d i s t r i c t s . Table 17. General Information on Land Tenure in Bogra and Rajshahi, 1959-60 D i s t r i c t and Tenure Category Percentage of A l l Farms Percentage of Total Farm Area Average Size of Farm (acre) Bogra Owner-farmers 56 58 4.01 Owner-cum-tenants 40 40 3.82 Pure tenants 4 2 1.79 Rajshahi Owner-farmers 43 37 4.34 Owner-cum-tenants 55 62 5.65 Pure tenants 2 1 2.92 Source: Government of Pakistan, Pakistan Census of Agriculture, 1960, Vol. I, Final Report - East Pakistan, Part I, October 1962. 171 Table 18. Percentage of Total Farm Area Operated Under Different Arrangements by Different Tenure-Categories in Bogra and Rajshahi, 1959-60 Bogra Rajshahi Area owned by owner-cum-tenants as percentage of to ta l farm area 25.31 37.62 Area rented by owner-cum-tenants as percentage of to ta l farm area 14.09 24.38 Percentage of farm area under tenancy 16.09 25.38 Source: Government of Pakistan, Pakistan Census of Agriculture, I960, Vol. I, Final Report -East Pakistan, Part I, October 1962. Rajshahi was a t r ad i t i ona l stronghold of the rent-receiv ing 20 Hindu zamindars even before the B r i t i s h came into the picture. The B r i t i s h rule strengthened the i r posit ion further by incorporating the 21 d i s t r i c t within the "great zamindari of Rajshahi". In contrast, Bogra, which came into existence as a d i s t r i c t as recently as 1821, 22 was a hinterland of the great zamindari estates. As i t has been See Government of Pakistan, Population Census of Pakistan, 1961: District Census Report, Rajshahi, Karachi: Min ist ry of Home A f f a i r s , (undated), p. 6. Ibid., p. 7. 22 Ibid., p. 7. 172 argued ea r l i e r in Chapter I I, even after the zamindari system was abolished short ly af ter Pakistan 's independence, the old d i spar i ty in land ownership pers i sted, though at a somewhat lower l e v e l . This was so because many of the underproprietors of those Hindu zamindars who f l ed to India were able to occupy the land l e f t behind by the i r superiors. As a consequence Rajshahi was l e f t with r e l a t i v e l y more large 1 and owners, than Bogra; land owners who would be in a pos it ion to rent out land. The most common rental arrangement in both d i s t r i c t s was share-cropping. Eighty-two per cent of a l l rented land in Bogra was under sharecropping contracts, the corresponding f igure for Rajshahi being about 98%. One can now determine the percentages of farm area under sharecropping tenancy in the two d i s t r i c t s for the time period under consideration (1959-60): D i s t r i c t Percentage of Total Farm Area Sharecropped Bogra 13.19 Rajshahi 24.87 It might be noted that the percentage of to ta l farm area under sharecropping tenancy i s not necessari ly ident ica l with the percentage of to ta l cu l t ivated area under sharecropping tenancy. It i s the l a t t e r of these two that one should be concerned with i f the goal i s to determine the extent 'of rel iance on sharecropping tenancy for the actual cu l t i va t i on of farmland. Information on how the tota l cu l t i vated 173 area in the two d i s t r i c t s was d i s t r ibuted among d i f fe rent tenure-categories has been summarized in Table 19 below. Table 19. Percentage of Total Cult ivated Area Under Different Tenure-Categories in Bogra and Rajshahi, 1959-60 Tenure-Categories Bogra Rajshahi Owner-farmers Owner-cum-tenants Pure tenants 58.07 35.36 40.05 63.62 1.88 1.02 Source: Government of Pakistan, Pakistan Census of Agriculture, iSfiOj-Vol. I, Final Report - East Pakistan, Part I, October 1962. Again, one i s faced with the task of f inding out the percentage of to ta l cu l t i vated area under tenancy, part of which would be under the owner-cum-tenants and part under the pure tenants. Given the nature of avai lable data, th i s can be done only with the assumption that the owner-cum-tenant cu l t i vates equal proportions of farmland he owns and what he rents i n . This assumption can be j u s t i f i e d by noting 174 the fact that those land owners who rent extra land do so because they f i nd owned land to be less than what they require to meet family needs. They cu l t i va te the i r own land to the greatest possible extent before renting land; and, of course, rented land would not be kept underut i l ized e i ther. The fol lowing formulations have been used to obtain f igures on how the cu l t i vated area under the owner-cum-tenants was divided between the land they own and land they hold under tenancy: where, y-, = portion of the cu l t ivated area under the owner-cum-tenants owned by them, y„ = portion of the cu l t ivated area under the owner-cum-tenants obtained under tenancy, x-j = farm area owned by owner-cum-tenants, x 2 = farm area rented by owner-cum-tenants, X = to ta l farm area under the owner-cum-tenants (owned and rented), Y = tota l cu l t ivated area under the owner-cum-tenants Y, and Y (owned and rented). Since values of x^, x 2 , X and Y were already known, y-j and y 2 could be calculated eas i l y . Information on a l l of these has been presented in Table 20. 175 Table 20. Farm Area and Cult ivated Area Under Different Categories of Tenure Arrangement in Bogra and Rajshahi, 1959-60 Bogra Rajshahi (acres) (x i ) Farm area owned by owner-cum-tenants 189,927 694,882 (x 2 ) Farm area rented by owner-cum-tenants 108,028 451 ,647 (X) Total farm area under the owner-cum-tenants (owned and rented) 297,955 1 ,146,529 (Y) Total cu l t i vated area under the owner-cum-tenants (owned and rented) 276,122 1,028,478 (Yl) Portion of the cu l t i vated area under the owner-cum-tenants owned by,them 176,010 623,334 (y 2 ) Portion of the cu l t ivated area under the owner-cum-tenants obtained under tenancy 100,1:12 405,144 Source: Government of Pakistan, Pakistan Census of Agvioultuve, I960, Vol. I, Final Report -East Pakistan, Part I, October 1962. 176 By adding the value of with the area cu l t ivated by pure tenants one can f ind the to ta l cu l t i vated area under tenancy. With the information gathered so far one can also calculate the fol lowing f igures for the percentages of to ta l cu l t i vated area under share-cropping tenancy in the two d i s t r i c t s . . D i s t r i c t Percentages of Total Cult ivated Area Under Sharecropping Tenancy, 1959-60  Bogra 13.39 Rajshahi 25.51 As compared to Bogra, one f inds Rajshahi to be the d i s t r i c t with a much greater rel iance on sharecropping tenancy for the cu l t i va t i on of i t s farmland. Data on the degree of landlessness in the sample areas of th i s study were, unfortunately, not ava i lab le. However, data were avai lable on hired labour as a percentage of to ta l ag r i cu l tura l workers. These data may o f fe r some indicat ion as.to the level of landlessness since agr icu l tura l wage-earners are drawn mostly from the landless households. In Bogra, 8.92% of a l l ag r i cu l tura l workers were hired labour, the corresponding f igure for Rajshahi being 11.82%. It i s f a i r to conclude from the above discussion that Bogra and Rajshahi are two d i s t r i c t s in Bangladesh where considerable s i m i l a r i t i e s in s o i l composition and c l imat ic conditions p r eva i l , but where the extents of sharecropping tenancy are s i gn i f i c an t l y d i f fe rent . 177 Therefore, large differences in product iv i t ies demonstrated by the two d i s t r i c t s would be explained less by the phys ica l -c l imat ic differences than by organizational arrangements including the d i f f e r e n t i a l tenure arrangement. Whether the levels of product iv i ty have been s i g n i f i c an t l y d i f fe rent in Bogra and Rajshahi, and how fa r such differences can be convincingly associated with differences in the incidence of sharecropping tenancy i s discussed in the fol lowing section. Levels of agr icu l tura l product iv i ty and possible explanations For the sake of s imp l i c i t y the levels of product iv i ty w i l l be compared here only in terms of the y ie lds per acre of the most 23 important crop for both the d i s t r i c t s — r i c e . Average product iv ity f igures in r i ce y ie lds per acre for Bogra and Rajshahi for the time period under consideration are as fo l lows: D i s t r i c t Rice Y ie ld (maunds per acre, 1 maund = 82.29 lbs. ) Bogra 9.5 Rajshahi 7.5 Percentage of to ta l cropped acreage under r i ce was about 80% in Bogra and 70% in Rajshahi. 178 Rajshahi, the d i s t r i c t with the greater percentage of cu l t ivated land under sharecropping tenancy, was thus producing less per unit of land than Bogra. To judge whether th i s difference in product iv i ty can be attr ibuted to a var iat ion in the i n ten s i t i e s of input-use in the d i s t r i c t s resu l t ing from the i r indiv idual tenure patterns, one has to investigate the ef fects on product iv i ty of factors independent of the in tens i ty of input-use. It has already been shown that the s o i l composition and the c l imat ic conditions in the two d i s t r i c t s are quite s im i la r . The possible contr ibution of each of the fol lowing factors in creating the product iv i ty d i f f e r e n t i a l w i l l now be considered: ( i ) nature of the r i ce crops produced, ( i i ) a v a i l a b i l i t y of f e r t i l i z e r s , ( i i i ) a v a i l a b i l i t y of i r r i g a t i o n f a c i l i t i e s , ( iv) farm implements and work-animals in possession, (v) a v a i l a b i l i t y of c r ed i t , (v i ) percentage of labour force in non-agricultural a c t i v i t i e s , ( v i i ) internal demands for r i c e . Going back to the study on the reg ional izat ion of Bangladesh agr iculture presented in the previous chapter, i t can be seen in Map 4 (page 120) that aus and aman are the two major r i c e crops produced in these d i s t r i c t s . Per acre average y i e l d of the aman crop i s about 1.2 24 maunds more than the per acre y i e l d of the aus crop. Therefore, See Ahmed, op.ait., pp. 67-69. 179 one may argue that i f Bogra had a r e l a t i v e l y greater proportion of i t s r i ce - land under the more productive aman crop then at least part of the difference in y i e l d could be explained by the var iat ion in the pattern of r i c e cropping. However, the facts do not support th i s argument. The percentages of r i ce lands devoted to aus and aman in the two d i s t r i c t s during the period 1959-60 show that Rajshahi had a s l i g h t l y greater proportion of the r ice- lands under 25 the more productive aman: Percentage of Percentage of Rice-Land Under Rice-Land Under Aman Aus Bogra 68 30 Rajshahi 71 25 Bogra was reported to have used farmyard manure and/or chemical f e r t i l i z e r s on 49% of i t s cropped acreage. The corresponding f igure for Rajshahi was 38%. How much farmyard manure i s applied to the cropped acreage depends upon the extent to which the common by-products of a farm are u t i l i z e d . Although no concrete data are ava i l ab le , i t could be expected that, because of the geographical cont iguity of Bogra and Rajshahi (which would allow easy mobi l i ty between the d i s t r i c t s ) there would not be wide var iat ions in the levels of supply of chemical Rajshahi 'also had a greater proportion of i t s r ice- lands under the less important, but h ighest-y ie ld ing, r i ce crop — b o r o . 180 f e r t i l i z e r s sold by dealers in re la t ion to demands ex i s t ing in the d i s t r i c t s . One, therefore, has reason to conclude that the greater use of manure and f e r t i l i z e r s in Bogra was caused not by d i f f e r en t i a l a v a i l a b i l i t y of these inputs, but rather by a decision in favour of more intensive u t i l i z a t i o n of avai lable resources. I r r i gat ion by power-pumps covered .17% and .12% of to ta l cu l t i vated areas of Bogra and Rajshahi respect ively during 1959-60. These areas were so i n s i gn i f i can t in s ize that the s l i gh t edge enjoyed by Bogra in power-pump i r r i g a t i o n appears neg l ig ib le in the search for possible explanations for the difference in product iv i ty. Most of the i r r i g a t i o n in both d i s t r i c t s resulted from small-scale indiv idual i n i t i a t i v e , the scope for which was s imi la r in both 27 d i s t r i c t s . However, Bogra had 14% of i t s cu l t i vated land i r r i g a t e d , while in Rajshahi only 8% of cu l t ivated land was i r r i g a ted . The benefit that Bogra received from more widespread i r r i g a t i o n seems to have been the resu l t of a more intensive use o f .ava i lab le resources. The a v a i l a b i l i t y of farm implements and work-animals in the two d i s t r i c t s has been presented in Table 21. Neither of the d i s t r i c t s received any appreciable amount of chemical f e r t i l i z e r s through government agencies during 1959-60. In his "Water D i s t r i c t Grouping" Tabors has put Bogra and Rajshahi within the same group. See R. Tabors, The definition of Multifunctional Planning Regions: A Case Study of East Pakistan, Cambridge (Mass.): Center for Population Studies, Harvard Univers i ty, May 1971, pp. 62-63. 181 Table 21. Farm Implements and Work-Animals in Bogra and Rajshahi, 1959-60 Bogra Rajshahi Number of wooden ploughs 189,730 395,710 Cult ivated area per plough (acres) 3.7 4.1 Cult ivated area per work-animal (acres) 2.0 2.2 Source: Government of Pakistan, Pakistan Census of Agriculture, 1960, Vol. I, Final Report -East Pakistan, Part I, October 1962. Khusro has calculated that in West Bengal the minimum s ize of farm below which a pair of bullocks and a plough would be under-28 employed (the plough unit) i s 5 acres. Conditions of agr icu l ture in both Bogra and Rajshahi, which share common borders with West Bangal, have been s imi lar to those ex i s t ing in West Bengal. Therefore, i t would not be unwarranted to accept Khusro's f igure of plough unit for Bogra and Rajshahi as we l l . The average area covered by a pair of bullocks in each of the d i s t r i c t s was less than th i s plough unit of A.M. Khusro, Economies of Land Reform and Farm Size in India, Delhi: The Macmillan Company of India, 1973, p. 51. 182 5 acres, and about 90% of the to ta l farm area in each d i s t r i c t was covered by those households which possessed ploughs (and work-animals to go with them). Hence, neither Bogra nor Rajshahi appear l imi ted by a v a i l a b i l i t y of ploughs and work-animal constraints in 29 cu l t i va t ing farmland. It i s important to enquire whether Bogra enjoyed a better posit ion than Rajshahi in terms of a v a i l a b i l i t y of c red i t which would enable i t to invest more money in agr iculture to produce more per unit of land. To examine t h i s , f igures on indebtedness and sources of loans are presented in Table 22. As Table 22 shows, a greater percentage of farms in Rajshahi were able to obtain loans from government as well as non-government sources. It i s probable that Rajshahi being the d i s t r i c t with a higher percentage of r e l a t i v e l y large farms, a greater percentage of households there could provide the co l l a te ra l needed for obtaining government loans. Greater proportions of the loans in both the d i s t r i c t s were obtained from non-government sources which included r e l a t i v e s , fr iends and r i ch land owners. The percentage of farms taking loans from non-government sources was higher in Rajshahi One may note here that although there does not ex ist a w e l l -developed market for the h i r ing of farm implements and work-animals in Bangladesh, some sharing of these items does take place, pa r t i cu l a r l y among k i th and k in. This lends extra support to the present argument. 183 Table 22. Indebtedness and Sources of Loan in Bogra and Rajshahi, 1959-60 ( i ) Percentage of farms reporting debt ( i i ) Percentage of farms reporting debt from government sources * ( i i i ) Percentage of farms reporting debt from non-government sources * ( iv ) Percentage of farms reporting debt from both government and non-government sources ( i i i ) + ( i i ) - ( i ) (v) Percentage of farms reporting debt from only non-government sources ( i i i ) - ( iv) (v i ) Percentage of farms reporting debt from only government sources ( i ) - ( i i i ) * These farms may or may not have obtained loans from other sources. Source: Government of Pakistan, Pakistan Census of Agriculture, I960, Vol. I, Final Report -East Pakistan, Part I, October 1962. 184 probably because of the considerat ion, among others, that the proportion of farms having r e l a t i v e l y large landholdings and hence the capacity to extend loans was higher there. A greater proportion of the r e l a t i v e l y small farmers, who would need the cred i t most, were found to have borrowed money from one source or another in Rajshahi. Thus, in Rajshahi, 49% of those farms which were below 5 acres in s ize obtained loans from some source, whereas in Bogra 43% of the farms in the same size-category were found to secure loans. A l l of th i s indicates that i t was not so much a cred i t handicap that caused Rajshahi to perform less productively than Bogra. Another source of cap i ta l f o r the farm households could be the income earned by family members employed in non-agricultural occupations. Since most of the non-agricultural workers maintain close contact with the rural, areas of the i r o r i g i n , and since they t r y to help t he i r rural re la t i ves with money i f they can afford i t , i t i s worthwhile to enquire i f one or the other d i s t r i c t appeared more l i k e l y to benefit from such advantages. The only information that could be gathered in th is,respect was the percentage of the to ta l labour force engaged in non-agricultural a c t i v i t i e s in each of the d i s t r i c t s . Since Bangladesh had no major indust r ia l centre during 1959-60 that could at t ract any s i gn i f i cant flow of labour from the various d i s t r i c t s , one can assume that most of the non-agricultural 185 labour force remained within the i r d i s t r i c t s of o r i g i n . Therefore, the fact that 10.79% of the to ta l labour force in Bogra and 11.81% of that in Rajshahi were employed in non-agricultural occupations would indicate that the agr i cu l tu ra l sectors of the two d i s t r i c t s 30 were in comparable posit ions to obtain capita l from th i s source. Therefore, the difference in the levels of product iv i ty in the two d i s t r i c t s would again seem not to be explainable in these terms. It has been shown that Rajshahi has performed r e l a t i v e l y poorly compared to Bogra with respect, to per acre y i e l d of r i c e . What has not been considered yet i s the re lat ionship between internal demands fo r r i c e grain and i t s production. Is i t possible that Rajshahi, with a lower population density (769 per square mi le, compared to 1,048 per square mile in Bogra), could generate an internal supply of per capita r i ce grain equal to , or even higher than, the corresponding This, of course, i s based on the assumption that non-agricultural occupations with d i f ferent levels of remuneration had comparable openings for the labour force of both d i s t r i c t s . This i s not an un rea l i s t i c assumption given that neither had a well developed urban indus t r i a l i zed or t rad ing ' sector . Tabors {op.ait., p. 50) has put these d i s t r i c t s in the same group in his "Industry D i s t r i c t Groupings", which lends extra support to the assumption. If one considers the fact pointed but by Tabors, that Rajshahi had an edge over Bogra in terms of cottage industry employments then- one only helps in strengthening the conclusion that agr iculture in Rajshahi did not have a narrower scope for obtaining capita l from the source under consideration. 186 supply made in Bogra even with a Tower per acre y ie ld ? This question gains in importance when one enquires i f ' s a t i a t i o n ' of basic needs for food and shelter imposes any l im i t s on work ef for t s and product iv i ty. Such doubts can be put to rest by noting the figures on per capita r i ce production along with the figures for per acre y i e l d in the two d i s t r i c t s during the period under consideration, 1959-60. D i s t r i c t Per Capita Rice Per Acre Y ie ld Production (maunds)  (maunds) Bogra 4.25 9.5 Rajshahi 3.36 7.5 The important point to note about the above figures is that the d i s t r i c t with the lower y i e l d produced less per capita as we l l . Therefore, lower y i e l d per unit of land in Rajshahi could not be explained by lower demands on i t s lands. From the above discussion, i t appears that Rajshahi, which had a much greater rel iance on sharecropping cu l t i va t i on than Bogra, achieved a lower level of product iv i ty mainly due to a less intensive u t i l i z a t i o n of avai lable resources. Although one would require more data than are ava i lable to be conclusive about i t , the i n t e r d i s t r i c t study strongly implies that when sharecroppers farm tenancies greater in s ize than necessary to meet basic subsistence needs, the intens i ty of input-use by them tends to be lower than 187 31 would l i k e l y be the case under owner cu l t i v a t i on . Macro-dynamic analysis:. Although the preceding discussion evinces some evidence for the r e l a t i ve e f f i c i ency of owner cu l t i va t i on over sharecropping cu l t i va t i on in certa in parts of Bangladesh within a s t a t i c set of circumstances, i t does not fo l low that the perpetuation of owner cu l t i v a t i on in i t s ex i s t ing form would make good sense for Bangladesh. Two important considerations in th i s regard are: 1. whether the r e l a t i ve e f f i c iency of owner cu l t i va t i on can continue to ex ist with the passage of time, 2. whether the e f f i c iency of owner c u l t i v a t i o n , as demonstrated within a region or a country, i s sat i s factory r e l a t i ve to other possible forms of agrarian organization. It was pointed out in evaluating the conventional theories on the re l a t i ve productive e f f i c iency of d i f fe rent tenure groups that the 'marginal ' tenants are l i k e l y to. behave more l i k e small owner-cu l t i vator s in making use of the various inputs. As sharecropped farms become smaller due to increased pressure on land brought about by population growth, so w i l l owner cu l t ivated farms. Eventually, 31 The average farm-size under tenants in Rajshahi, the d i s t r i c t with the lower product iv i ty , was over 5 acres, of which over 3 acres were owned by them. This shows that the tenants in Rajshahi had farms well over the subsistence s ize of 2.5 acres (subsistence s ize discussed on p.159). In contrast, the average farm s ize of tenants in Bogra was about 4 acres, of which about 2 acres were owned by them. 188 one may end up with very small sharecropped as well as owner cu l t ivated farms. If one accepts the argument that neither of these groups of farms would l i k e l y be in a posit ion to afford s i gn i f i can t investments in securing better qua l i ty inputs or improving the qua l i ty of land then the levels of product iv i ty achieved would increasingly be a function of the intens i ty with which they use t r ad i t i ona l inputs mainly generated from within the farm i t s e l f . Therefore, as the sizes of sharecropped and owner cu l t ivated farms become smaller in s ize one would expect a narrowing down of the difference between the levels of product iv i ty achieved by them. Unfortunately, no recent data on the farm sizes of d i f fe rent tenure groups in Bogra and Rajshahi are ava i lab le which could have been used, along with the product iv i ty f igures for the relevant period, to show how changes in farm sizes of tenure groups in these d i s t r i c t s since 1959-60 may have affected the i r p roduct i v i t ie s . However, one can look at some.recent national f igures to get some idea of the change in the average s ize of farms which rented land in the two d i s t r i c t s during that time. The 1977 Land Occupancy Survey shows that the average s ize of farms renting land during that year was 2.7 acres. During 1959-60 the corresponding f igure was 4.2 acres. It i s to be expected that the average s ize of farms renting in land in Bogra and Rajshahi would also come down appreciably over 189 the years. One would, therefore, expect that the levels of product iv ity achieved by the two d i s t r i c t s at a more recent date would become narrower, given that Bogra did not gain dispropor-t ionate ly from any public po l i c i e s with regard to agr icu l tura l development. Figures on r i ce y ie ld s in the two d i s t r i c t s fo r 1972-73, when compared with those of 1959-60 (reported e a r l i e r in the i n t e r d i s t r i c t study, and repeated below), support the above 32 contention. D i s t r i c t Rice Y ie ld (maunds/acre) 1959-60 1972-73 Bogra 9.5 10.2 Rajshahi 7.5 9.3 That the product iv i ty per unit of land achieved by small tenants may not lag fa r behind those of small owner-cultivators can also be demonstrated with the data avai lable from an area study 33 done by Hossain. Table 23 contains some of the results he obtained. One can see in th i s table that the tenants, who had farms sized only about 2.5 acres, surpassed the owner-cultivators (again, with about 2.5 acres of land) in product iv i ty per unit of land. Since the qual i ty of land under the owner-cultivators and the sharecroppers was The 1972-73 figures have been obtained from Ahmed, op.cit., p. 66. Hossain (1977), op.ait. 190 Table 23. Farm Sizes, Input-Use and Land Product iv i t ies of Owner-Cultivators and Tenants in Phulpur Thana, 1973-74 Owner- Tenants Cult ivators Average s ize of farm (acres) 2.54 2.58 Use of Labour (days/acre) 91 100 Intensity of land-use (tota l cropped acreage as percentage of cu l t ivated acreage) 192 200 Land product iv i ty (taka per acre at 1969/70 prices) 672 704 Source: M. Hossain, 'Farm S ize, Tenancy and Land Product iv i ty: An Analysis of Farm Level Data in Bangladesh Ag r i cu l t u re 1 , Bangladesh Development Studies, Vol. V, No. 3, Ju ly 1977, pp. 314, 316, 321, 345. reported to be comparable, and since the small tenants were not expected to have greater access to the resources used than the owner-cult ivators, one possible conclusion in th i s regard would be that the tenants, being closer to the subsistence minimum af ter paying rents to the landlords, had to cu l t i va te the land under them, with greater in tens i ty to meet 34 family needs with the i r share of the to ta l product. That the qua l i ty of land under the owners and the tenants were comparable was mentioned by Hossain. See ibid., p. 319. 191 The pol icy implications ar i s ing out of the above analysis are c lear . To ra ise the level, of product iv i ty in Bangladesh agr icu l ture in the short run i t would be desirable to do away with the present dependence on sharecropping cu l t i v a t i on . However, the longer run requirements would not be met by. merely replacing sharecropping cu l t i va t i on by owner cu l t i v a t i on . In the context of Bangladesh agr icu l ture, both of these arrangements would resu l t in the long run in low levels of output per unit of land and labour. A study by Jabbar provides some useful information on the product iv i t ies of owner-cum-tenants, as compared to those of owner-cu l t i v a to r s , in s i tuat ions where the farmers operated d i f fe rent amounts 35 of farmland. Table 24 presents his f indings re la t ing to three d i f fe rent d i s t r i c t s of Bangladesh — Dinajpur, Mymensingh and Rangpur. This table shows that owner-cum-tenants produced more per acre of cropped acreage than did owner-cultivators in each of the three d i s t r i c t s . However, the only s i gn i f i cant difference was observed in Rangpur where the owner-cum-tenants had the lowest farm area (2.46 acres) as compared to other d i s t r i c t s . Data reported above indicate that the owner-cultivators in Bangladesh do not always demonstrate better productive e f f i c i ency over the tenants. More importantly, whether owner cu l t i v a t i on resu lts in more or less e f f i c i ency than does tenant c u l t i v a t i o n , the very poor Jabbar, op .'ait. Jabbar uses d i f fe rent names for these tenure-categories (part-tenants and owner-operators). Table 24. Average Cult ivated Area and Land Productiv ity of Different Tenure Categories in Dinajpur, Mymensingh and Rangpur, 1973-74 D i s t r i c t and Tenure Category Average Cult ivated Area (acres) Value of Crops and By-products per Cropped Acreage (Taka) Dinajpur Owner-cultivators 6.31 1,157 Owner-cum-tenants 4.77 1,179 Mymensingh Owner-cultivators 3.57 1,059 Owner-cum-tenants 2.64 1,063 Rangpur Owner-cultivators 3.60 1,470 Owner-cum-tenants 2.46 1,590 Source: M.A. Jabbar, An Investigation into the Effect of Farm Structure on Resource Productivity in Selected Areas of Bangladesh. Unpublished Ph.D. Dissertat ion, Univers ity of Wales, 1976, pp. 83, 101. 193 levels of product iv i ty achieved by e ither of these categories of farmers would suggest that ne i ther- in i t s present form can help the process of ag r i cu l tu ra l development in Bangladesh. The regional survey of Bangladesh agr iculture in the previous chapter showed that the d i s t r i c t s of Sylhet, Comilla and Chittagong H i l l Tracts had below 10% of t he i r to ta l cropped acreage under share-cropping c u l t i v a t i o n , whereas in the d i s t r i c t s of Khulna, Rajshahi, Dinajpur and Kushtia the corresponding f igure was over 20%. However, the product iv i t ies achieved in both of these categories of d i s t r i c t s were far from sat i s factory (the average r i ce y i e l d per acre during 1972-73 being 10.36 maunds for the f i r s t category and 9.02 for the 36 37 second category). ' Thus, although the analysis carr ied out so fa r does indicate that, mainly because of equity considerations and also because of some e f f i c iency considerations, i t would be worthwhile in the short run to replace sharecropping cu l t i va t i on in Bangladesh by owner c u l t i v a t i o n , neither of these a lternat ives in the i r present forms would be appropriate for the country in the long run. A macro-dynamic analysis of the re lat ionsh ip between farm s ize and product iv i ty casts doubt on the v i a b i l i t y of owner cu l t i v a t i on in These r i ce y i e l d f igures are calculated from data provided by Ahmed. See Ahmed, op. oit.3 p. 66. 37 These product iv i ty f igures compare w/ith about 40 maunds per acre in China and 65 maunds per acre in Japan. See Food and Agr iculture Organization, FAO Production Year Book, 1978, Rome, 1979, p. 98, (or ig ina l f igures in kg/ha). 194 Bangladesh in the long run. This w i l l be discussed in th i s chapter af ter some addit ional aspects of micro- s tat ic relat ionships between farm size and product iv i ty have been examined. Farm Size and Agr icu l tura l Product iv i ty M ic ro - s tat i c analysis Theoretical premise It has already been stated in Chapter I (page 7 ) that the work of A.K. Sen i s considered a landmark in the analysis of the micro-38 s t a t i c e f f i c i enc i e s of d i f ferent sizes of farms. Sen builds his argument on the contention that in the family-based agr icu l ture of densely populated less developed countr ies, farms which are r e l a t i v e l y small cu l t i va te the i r land as intens ive ly as possible to produce as much as they can per unit of land. The opportunity cost of labour for them i s very low. Small farms with surplus family labour cannot eas i ly f i nd a l ternat ive employment opportunities for th i s surplus outside of the i r farms. For th i s reason they keep on applying labour . to t he i r land even when i t s marginal product f a l l s below the market wage rate. These farms maximize output per unit of land in order to meet family consumption needs. In view of the low opportunity cost of the i r family labour and the subsistence nature of t he i r farming, the p ro f i t maximizing condition (equalizing marginal' product of labour with going wage rate) i s not relevant for them. Sen, op. ait. 195 The r e l a t i v e l y large farms which employ wage labour to cu l t i va te the i r land would, however, hire labour only to the point where i t s marginal product equals the wage rate. Otherwise, they would be net losers. Re lat ive ly large farms are, therefore, associated with lower i n tens i t i e s of labour-use than the smaller farms. The resu l t i s that output per unit of land i s higher on smaller farms. The point can be i l l u s t r a t e d with the help of Figure 5. 0 D B Labour input »-Figure 5. Micro-Stat ic Ef f ic iency of Small Farms 196 In Figure 5, AB represents the value of the marginal product of labour, and OC the level of wage which i s exogenously determined. Re lat ive ly large farms based on wage labour cu l t i va t i on would employ OD amount of labour per unit of land since the corresponding point on AB, that i s , E, would sa t i s f y the p ro f i t maximizing condit ion, marginal value product of labour (DE) = wage rate (OC). Value of the to ta l product i s represented by area OAED and the p ro f i t would equal area CAE (OCED being the to ta l wage b i l l ) . Farms based on family labour w i l l not, however, hesitate to use family labour up to B. Here the value of the tota l product would equal the area OAB. Thus, family-based farming of the smaller farms (where labour supply per unit of land i s abundant), when compared with the wage labour-based farming undertaken by the r e l a t i v e l y large land owners, would produce a greater output. The value of the output produced by the smaller farms would be greater in value by the amount given by, OAB - OAED = DEB. If th i s to ta l value i s divided by the unit pr ice of output one gets the amount of addit ional output that family-farming of the smaller 197 farms would resu l t i n . It i s suggested that not only land, but also the cap i ta l avai lable in the agr icu l tura l sectors of most underdeveloped countries would tend to be u t i l i z e d more intens ive ly by the smaller farms. The reason, i t i s argued, i s that avai lable capita l (draught power, manure, and the l i ke ) are most complementary to the use of human labour. Empirical f indings Data on the levels of land product iv i ty achieved by d i f fe rent s ize-categories of owner-farmers in Bangladesh were avai lable for a number of d i s t r i c t s in the country. These data are presented in Table 25. In a l l of the f i v e d i s t r i c t s reported in th i s tab le , the value of production per cu l t ivated acre was highest amongst farms in the smallest size-category. This i s supportive of an inverse re lat ionship between s ize of farm and product iv i ty per unit of land, which conforms to the hypothesis proposed by Sen's theory. Several other researchers have also reported inverse r e l a t i o n -ships between farm-size and product iv i ty per unit of land. Hossain, for example, f inds that in two d i f ferent ihanas (administrative units within a d i s t r i c t ) of Bangladesh.—-Phulpur and Thakurgaon ( in the d i s t r i c t s of Mymensingh and Dinajpur, respect ive ly ) , product iv i ty per unit of land was higher in owner cu l t i vated farms which were smaller 39 in s i ze. His data, which re late to two d i f fe rent periods of time Hossain (1977), op.ait. 198 Table 25. Value of Production Per Cult ivated Acre in Owner-Farms of Different Sizes in a Number of D i s t r i c t s of Bangladesh (various years) D i s t r i c t and Year of Study Size Group of Value of Production Farms' Per Cult ivated Acre (Taka) Bogra, 1966-69 Low 576 Medium 492 High 413 Dinajpur, 1962-63 Low • 465 Medium 360 High 309 Kushtia, 1966-69 Low 531 Medium 487 High 433 Mymensingh, 1962-65 Low 399 Medium 389 High 329 Rajshahi 1963-66 Low 404 Medium 339 High 275 Low groups were defined as having less than 6 acres, medium groups as having between 6.01 to 12 acres, and high groups as having more than 12 acres. Source: Government of Bangladesh, Farm Management Research in Retrospect, Dacca: Min istry of Agr icu l ture, 1973. 199 (1969-70 and 1973-74) for each thana, .can be seen in Table 26. It would be worthwhile to examine how farms of d i f fe rent sizes in the country have performed in terms of product iv i ty , not only per unit of land, but also in terms of. product iv i ty of d i f fe rent factor-rinputs taken together; that i s , in terms of to ta l factor product iv i ty. Although i t i s reasonable to concentrate on the product iv i ty of land as a measure of e f f i c i ency in s i tuat ions of extreme land sca rc i t y , one could gain further ins ights into the performance of various operating groups by looking at to ta l factor product iv i ty f igures in the country as a whole or in various regions thereof. Unfortunately, data are not avai lable to allow th i s at the national l e v e l . Some information on the regional level i s , however, ava i lable from a separate study undertaken 40 by Hossain on the thana of Phulpur. He employed the fol lowing expression to determine the r e l a t i ve tota l factor product iv i ty of a par t i cu la r s ize group of farms: L. MPL. + N. MPN. + C MPC H i L.- MPL + N. MPN H T ~ C ^ K P C where, A. = r e l a t i ve to ta l factor product iv i ty of s ize group i , L. =• units of land under a l l farms in s ize group i , N.. = units of labour employed by a l l farms in s ize group i , C . - = units of capi ta l used by a l l farms in s ize group i , MPL., MPN., MPC. = marginal products of land, labour and capita l respect ively of a l l farms in s ize group i , MPL, MPN, MPC = marginal products of land, labour and capita l respect ively of a l l farms in a l l s ize groups. 4 0 Hossain (1974), op.ait. Table 26. Value of Production Per Cultivated Acre in Owner Cult ivated Farms of Different Sizes in the Thanas of Phulpur and Thakurgaon, 1969-70 and 1973-74 Thana and Farm Size Value of Production Per Cult ivated Acre (Taka at 1969/70 Prices) 1969-70 1973-74 Phulpur Farms with 6.5 acres or less 461 672 Farms with more than 6.5 acres 294 528 Thakurgaon Farms with 6.5 acres or less 331 353 Farms with more than 6.5 acres 301 303 Source: M. Hossain, 'Farm Size, Tenancy and Land Productiv ity: An Analysis of Farm Level Data in Bangladesh Ag r i cu l tu re ' , Bangladesh Development Studies, Vol. V, No. 3, July 1977, p. 314. ro o o 201 A., in Hossain's terms, is therefore a r a t i o of the actual output of the s ize group i to the output that one would have expected from i t were i t to demonstrate the average e f f i c i ency of a l l farms in u t i l i z i n g the avai lable inputs. The to ta l factor product iv i t ies of d i f fe rent s ize groups of farms, as reported by Hossain, are presented in Table 27. By d e f i n i t i o n , large A^ f igures indicate greater levels of productive e f f i c i ency . It can be seen from Table 27 that the s ize group of farms between 2.5 and 4.99 acres were the most e f f i c i e n t of 41 the various s ize groups reported. These farms did not, however, const itute the smallest s ize group.. Farms which had less than 2.5 acres of land constituted the smallest.group, and these were found to be less e f f i c i e n t , in the to ta l factor product iv i ty sense, than the s ize group immediately above. This shows that one cannot be assured that an inverse re lat ionsh ip between farm-size and productive e f f i c iency w i l l continue to be found below certa in levels of farm-size when tota l factor product iv i ty i s used as the measuring s t i ck . A p laus ib le reason fo r th i s would be that very small farmers (such as those in Hossain's study below 2.5 acres) may not be able to use a l l ava i lable resources to the f u l l e s t extent on the i r lands due to the smallness of the i r holdings. It should be noted that the s ize groups of farms reported in Table 27 were much smaller than those in Tables 25 and 26. 202 Table 27. Total Factor Product iv i ty of Farms of Different Sizes in the Thana of Phulpur, 1969-70 Size-Group of Farms Total Factor Product iv i ty (acres) Index ( A . ) Up to 2.49 0.99 2.5 - 4.99 1.08 5.0 - 7.49 1.05 7.5 and over 0.82 Source: M. Hossain, 'Farm Size and Product iv i ty in Bangladesh Agr icu l ture: A Case Study of Phulpur Farms', Bangladesh Economic Review, Vol. I I , No. 1, January 1974, p. 490. Macro-dynamic analysis Theoretical premise Farm e f f i c iency considered in a dynamic context refers e s sent ia l l y to i t s a b i l i t y to generate savings for making necessary investments and for adopting newer technologies to f a c i l i t a t e further growth. With regard to product iv i ty , the emphasis s h i f t s from what gets produced per acre of farms under given circumstances to what could be produced under 203 d i f fe rent sets of circumstances which might be introduced. In the words of Bhagwati and Chakravarty, " . . . we should emphasize that the ranking by private and social p r o f i t a b i l i t y of the farms by s i z e -classes may diverge from the i r ranking by acreage product iv i ty. Also the s t a t i c e f f i c iency of the smaller farms, i f demonstrated, may be consistent with t he i r dynamic i ne f f i c i ency from the viewpoint of savings, investment and innovation."42 The question of "savings, investment and innovation" i s highly important, for example, in evaluating prospects for the successful adoption of the technology involved with. HYVs, which require high expenditures on inputs such as chemical . f e r t i 1 i z e r s , contro l led water supply and pest ic ides. Although i t i s true that these inputs are d i v i s i b l e to a certain extent, smaller farmers may not be able to generate enough income to be able to make adequate investments in 43 these r e l a t i v e l y cost ly inputs. The poor economic condition of these farmers would l i k e l y be ref lected in stronger tendencies toward r i s k aversion. Since the newer plant var ie t ie s are more prone to diseases, smaller farms often prefer to avoid the r i sks involved. Under such circumstances, adoption of HYVs (or some range of comparable technology) and hence marketable surplus, can be expected to remain low. J . Bhagwati, and S. Chakravarty, 'Contributions to Indian Economic Analys is: A Survey', American Economic Review, Vol. LIX, No. 4, September 1969, p. 44. In th i s context i t i s useful to make note of the biases of technology discussed in Chapter II, 204 Two other considerations are important in the macro-dynamic context. F i r s t i s the level of u t i l i z a t i o n of ava i lable resources from the point of view of the society as a whole. Underemployment of manpower in small farms can be taken as a source of output loss in th i s context, espec ia l ly since the r e l a t i v e l y larger farms could have been cu l t i vated more intens ive ly with the help of that labour. If the f u l l 'market value ' of inputs supplied by the household i s included in the cost of operation then the net income generated by small farms would decline sharply and the micro-s tat ic analysis of the previous section would not hold t o t a l l y true. The second consideration relates to the question of the long-term v i a b i l i t y of the system as a whole. In places where land scarc i ty is coupled with the custom of subdividing holdings among he i r s , a s i tuat ion i s bound to emerge over time where the extent of parce l i zat ion of land in small farms becomes so great that a typ ica l household cannot generate enough income from i t s land to sa t i s f y family consumption needs and s t i l l be l e f t with enough resources to invest in measures aimed at improving conditions of farming. Empirical f indings In Bangladesh most cu l t i vab le land remains under low y ie ld ing t r ad i t i ona l va r ie t ie s of crops, although a switch to the HYVs could more than double the product iv i ty achieved with t r ad i t i ona l va r i e t i e s . Table 28 depicts the s i tuat ion as i t existed during 1972-73. 205 Table 28. Acreage and Y ie ld of Tradit ional and High Yielding Var iet ies of Rice in Bangladesh, 1972-73 Rice Variety Acreage Y ie ld (maunds per acre) Tradit ional var iety High y ie ld ing var iety 21,165,000 10.87 2,632,000 26.83 Source: Government of Bangladesh, Bangladesh Agriculture in Statistics, Dacca: Min ist ry of Agr icu l ture, November, 1973. According to a study done in 1971 the gross cost of producing HYVs per acre of land was Rs. 644.49, whereas the gross cost of 44 producing local var ie t ie s per acre worked out to be Rs. 446.39. In the face of such a cost d i f f e r e n t i a l the r e l a t i v e l y land-r ich farmers could.adoptl'the HV.Ys more read i ly than others. Seventy-three pe rcent of farms which adopted HYVs had more than 2.5 acres of land. These farms constituted only about a quarter of a l l farms in Bangladesh. Two facts are evident from the reported data: ( i ) a vast majority of farms have not adopted the HYVs, See : M.I. Khan, 'The Role of IR-20 in Solving the Food Problem of East Pak i s tan 1 , Pakistan Development Review, Spring 1971. 206 ( i i ) r e l a t i v e l y small farms cu l t ivated the HYVs less than the r e l a t i v e l y large farms. The i n a b i l i t y of so many farmers to adopt the new technology of HYVs has been one of the major, reasons for low levels of 45 product iv i ty in Bangladesh agr icu l ture. Since adoption was mainly re s t r i c ted to the r e l a t i v e l y large farms, the d i spar i ty of income-earning opportunities between small and large farms has s i g n i f i c an t l y widened. Government pronouncements l i k e the fol lowing have admitted that the introduction of HYVs .in Bangladesh has been associated with these trends, fo r example, " . . . r isk-perception of the small farmers for the recurring f loods, poor and marginal resource-base of the majority farmers and the i r fragmented holdings, resulted in slow adoption rates of the new technology. The early adopters, the more aff luent farmers, took advantage of the s i tuat ion and the process of concentration of land in a few hands started in 1966 . . . " 4 6 The conclusion seems inescapable that the macro-dynamic e f f i c i ency in the agr i cu l tu ra l sector of Bangladesh has suffered because of the i n a b i l i t y of the vast majority of small farms to adopt the new technology. It i s true that the HYVs would be less preferable on grounds of taste.considerations as we l l . However, given the acute poverty of so many farmers, th i s would not be a major variable determining the rate of adoptionrof the HYVs which had such a promise of ra i s ing the level of output. Government of Bangladesh, 'Text of Z i a ' s Speech at FAO Conference on Agrarian Reform'and Rural Development', Bangladesh, Vo l . I l l , No. 2, Aug. 1979, p. 5. 207 Table 29. Percentage of Output Supplied to the Market by Different Sizes of Farms in Bangladesh, 1960 Size-Group of Farms (acres) Marketed Surplus as Percentage of Quantity Produced Below 2.5 6.2 2.5 - 4.9 8.2 5.0 - 7.4 10.5 7.5 - 12.5 11.4 Above 12.5 12.7 Source: Government of Pakistan, Family Expenditure Schedule: Karachi: Central S t a t i s t i c a l O f f i ce , 1960. Many small farms did not have the means to produce enough to sa t i s f y family consumption needs, l e t alone supply substantial amounts of the i r product to the market. A c lear inverse re lat ionship between farm size and marketed surplus in Bangladesh agr iculture would be expected, and can be observed in Table 29. Loss of marketable surplus has, therefore, been another source of macro-dynamic i ne f f i c i ency of ex i s t ing s i ze -d i s t r i bu t i on of farms. An important concept in the analysis of i n p u t - u t i l i z a t i o n in a macro-dynamic context i s the plough un i t , which, as noted e a r l i e r (page 181), was defined by Khusro as the area which can adequately 208 47 employ a pair of bul locks. In the absence of a well developed market for h ir ing animal power there w i l l be under-ut i1 izat ion 48 of th i s source of power in farms which are below th i s plough un i t . If one accepts Khusro's f igure of plough unit for West Bengal (5 acres) as appropriate for Bangladesh then, according to the 1977 Land Occupancy Survey,more than 57% of the to ta l farmland at that 49 time was held in units smaller than the plough unit . This would cer ta in l y enta i l a great loss of e f f i c i ency at the macro level since the market for h i r ing animal power in Bangladesh, though not absent, is f a i r l y l imi ted and unre l iab le. In Bangladesh the growth of population and the pract ice of subdividing agr icu l tura l holdings among heirs has s i g n i f i c an t l y reduced 50 average farm-size over the years. That the parce l i zat ion of farm units w i l l become more serious i f . t he present system i s l e f t unattended i s obvious. The very t iny farm households would then be l e f t with A.M. Khusro, 'Farm Size and Land Tenure in I nd i a ' , Indian Economic Review, Vol. IV, (New Ser ies ) , 1969. Lower productive e f f i c i ency of very small farms as evidenced in the to ta l factor product iv i ty analysis of the preceding section i s worth noting in th i s regard. For the plough unit in West Bengal see Khusro (1973) op.cit., p. 51. Thus the average farm-size came down from 3.1 acres in 1968 to 2.3 acres in 1977. See (1) Government of Bangladesh, Master Survey of Agriculture in Bangladesh, Seventh Round, Second Phase (1967-68), Dacca: Bangladesh Bureau of S t a t i s t i c s , (undated); and (2) Government of Bangladesh, Summary Report of the 1977 Land Occupancy Survey of Rural Bangladesh, Dacca: Bangladesh Bureau of S t a t i s t i c s , 1977. 209 i n su f f i c i en t land to effectively.employ a l l family labour on i t and would be unable to obtain an income su f f i c i en t to meet consumption needs in addit ion to investment requirements for improving and d ivers i fy ing farming a c t i v i t i e s . Therefore, i t w i l l be d i f f i c u l t fo r them to keep up with the demands of the growing number of dependents per unit of land. This long run ' n on - v i ab i l i t y ' of small-s ized family farming in Bangladesh represents an ominous portent of the dynamic i ne f f i c i ency of small farming in the country. Conclusion Although a more deta i led.set of information than was ava i lab le would be required to f u l l y understand a l l the problems emerging out of the size-tenure structure in Bangladesh agr i cu l tu re , the discussion in th i s chapter has, nevertheless, outl ined the major obstacles to agr i cu l tura l development result ing from the ex i s t ing structure. The size-tenure structure in Bangladesh agr icu l ture has been found to be responsible for much of the d i spa r i t i e s in income-earning opportunities of d i f fe rent households., With regard to productive e f f i c i ency , i t has been demonstrated that sharecropping tenancy, when compared with owner cu l t i v a t i on in a micro - s tat ic context, i s l i k e l y to resu l t in a less intensive u t i l i z a t i o n of avai lable resources, and a lower leve l of output per unit of land i f the farms renting in land 210 operate above the level of ' subs i stence ' . As these tenant farms get smaller, they f i n d i t necessary to cu l t i va te a l l land (owned and rented) with more vigour to meet family needs. Consequently, rented land under very small farms has been observed to be cu l t i vated with an e f f i c i ency that did not lag behind the e f f i c i ency demonstrated by owner-cultivators who had s imi la r resource endowments. However, since the macro-dynamic e f f i c i ency remains rather low for both sharecropping and owner cu l t i v a t i on in Bangladesh, and seems l i k e l y to deter iorate over time, the more important observation in th i s regard would be that neither of these arrangements can cater to the long-term needs of the country. As regards the question of farm-size, the apparent r e l a t i ve micro-s t a t i c e f f i c i ency of farms in the smal ler : categor ies, measured in product iv i ty per unit of land at a point in time, does not meet the long-term requirements of (a) creating su f f i c i en t e f fect i ve employment opportunities for the growing labour force, (b) encouraging adoption of newer technologies l i k e the HYVs and (c) generating larger marketable surpluses for purposes of savings and investments. The pol icy impl ications of these f indings can be viewed both in terms of short run and long run requirements. Equity considerations point to the necessity of adopting, as ear ly as poss ible, po l i c ie s which can reduce d i spa r i t i e s in land ownership in the areas of highly 51 skewed ownership patterns. ' In the short run, i t i s also necessary As Table 14 (page 127) indicated, the problem of d i spar i ty in land ownership was most acute in the northwestern d i s t r i c t s of Dinajpur, Rajshahi and Kushtia. 211 to adopt po l i c i e s which can replace sharecropping cu l t i va t i on by owner cu l t i va t i on in the areas of high tenancy for atta in ing 52 greater equ i tyand ra i s ing the general level of product iv i ty. However, long run considerations require that a l ternat ive arrangements of size-tenure structure (as d i s t i n c t from the t rad i t i ona l small-scale owner cu l t i va t i on or tenant cu l t i va t ion ) be found to create a v iable ag r i cu l tu ra l sector in the country. What a lternat ives can best serve the purpose of reconci l ing these short run goals with the long run goals i s the subject of the chapter that fol lows. Looking at Chart 1 (p. 132), Dinajpur, Rajshahi, Kushtia and Khulna being the d i s t r i c t s with the highest levels of share-cropping tenancy in the country, such a pol icy would mainly be concentrated in those areas. 212 V. LAND REFORM FOR BANGLADESH Why Land Reform in Bangladesh? As stated in the introductory chapter, Bangladesh i s pr imari ly an agr i cu l tura l country. The nature of income d i s t r i bu t i on and level of product iv i ty per unit of land or labour in the agr i cu l tu ra l sector determine whether the country w i l l be capable of f inding a development path which has the characte r i s t i c features of " red i s t r i bu t ion with growth" needed for long run economic and p o l i t i c a l v i a b i l i t y . ^ The discussion on the problems of size-tenure structure contained in the previous chapter points to the necessity for a l te r ing the agrarian structure i f the dual goals of equalizing income-earning opportunities and ra i s ing the level of product iv i ty are to be achieved. Almost a th i rd of rural households in Bangladesh are landless. Most of them have to earn the i r l i v i n g by searching for opportunities to s e l l t he i r labour, often on a day-to-day basis, in the wage market. Very few of them manage to obtain access to the lands of others as The term " red i s t r ibut ion with growth" i s borrowed from the t i t l e of a book which marked a reor ientat ion in conceptualizing the requirements of development. See H. Chenery et al., Redistribution With Growth, London: Oxford Univers ity Press, 1974.. 213 2 sharecroppers for reasons discussed e a r l i e r in the study. When one looks only at land-owning households, the interhousehold d i spa r i t y in ownership of land also i s very high. Of a l l the households which owned some farmland during 1977, 67% belonged 3 to the s ize group with 2 acres or less. They accounted for only about 25% of a l l avai lable farmland. Ownership units with more than 10 acres of farmland, on the other hand, although representing only 2.6% of a l l households which owned some farmland, accounted for almost a f i f t h of a l l avai lable farmland. Income-earning opportunities in a heavily populated agr i cu l tu ra l society based on t r ad i t i ona l technology are strongly influenced by the land ownership status of the contending households. Thus, a highly skewed d i s t r i bu t i on of income has resulted among the various households in the rural sector of Bangladesh. In terms of the conceptual framework suggested in Chapter II, one can readi ly argue that s i gn i f i cant reductions in income inequal i ty in Bangladesh can only be achieved through changes in the patterns of land ownership and tenure arrangements. With regard to the r e l a t i ve productive e f f i c iency of d i f fe rent farm sizes and tenure-categories, the present study has led to the conclusion that one cannot hope to raise levels of product iv i ty per See p. 103. These, and the subsequent f igures for 1977 used in th i s chapter, are taken from Government of Bangladesh, Summary Report of the 197? Land Occupancy Survey of Rural Bangladesh, Dacca: Bangladesh Bureau of S t a t i s t i c s , 1977. 214 unit of land or labour in e ither the micro - s tat ic or macro-dynamic context without making modifications in ex i s t ing size-tenure 4 structures in Bangladesh. Neither the employment s i tuat ion nor the technological choices emerging out of ex i s t ing size-tenure structures are conducive to sustained long-term growth or improved equity outcomes-. To achieve these ends i t i s necessary to adopt a pol icy of more e f fec t i ve u t i l i z a t i o n . o f ex i s t ing resources which i s conducive to both employment expansion and technological progress. The f indings of Chapters III and IV ( b r i e f l y recapitulated above) can help answer three sets of questions concerning the requ i s i te components (or de f in i t i on ) of land reform in the context of Bangladesh. These questions related to the stage of development of the country, the patterns of size-tenure s t ructure, and the nature of the most pressing problem in the man-land re lat ionship in the country. The pattern of size-tenure structure in Bangladesh poses some c r i t i c a l constraints upon simultaneously atta in ing more equitable d i s t r i bu t i on of income and increased product iv i ty per unit of land and labour. The components of land reform for Bangladesh, as suggested by the f indings so f a r , can be put together in the fol lowing formulation: A land reform programme for Bangladesh which seeks to atta in broader equity in access to land, and increased product iv i ty (per unit of land and labour) See Chapter IV, pp. 209-211. 215 needs to incorporate, at a minimum, po l i c i e s aimed at removing tenur ia l d is incentives and s i z e - d i s a b i l i t i e s of farm units under ex i s t ing arrangements.^ The l i k e l y effectiveness of each of the series of a l ternat ive proposals, or 'models' of land reform developed in Chapter II w i l l now be considered in turn in the context of Bangladesh. Problems and Prospects of Various Models ( i ) Red i s t r i bu t i v i s t Reform. Proponents of th i s type of land reform suggest a low ce i l i n g be leg i s la ted on the permissible s ize of holding of a family. 'Surplus ' land generated by implementing the ce i l i n g i s to be d i s t r ibuted among the land-poor to enhance the i r economic opportunit ies. It i s suggested that by making more land avai lable to the small farmers and landless households, where there i s an abundance of family labour, one can hope for more intensive cu l t i va t i on of a greater proportion of The term 'access to land ' rather than 'ownership of land 1 i s used here to cover a l ternat ive p o s s i b i l i t i e s in atta in ing equity in the use of ag r i cu l tu ra l land: An equitable d i s t r i bu t i on of land ownership among indiv idual households i s only one of the p o s s i b i l i t i e s . It has been considered important to look into other p o s s i b i l i t i e s in achieving the desired equity. 216 tota l land. It i s argued that, when coupled with appropriate extension serv ices, th i s would help achieve the important objective of increasing the level of production. An examination of what c e i l i n g on land might be necessary to achieve an appreciable measure of equity in rural Bangladesh, s tart ing from the ex ist ing context of land ownership patterns, i s in order. Table 30 contains information on the ' surp lus ' land that could have been generated in 1977 by imposing d i f fe rent levels of ce i l i ng s on the land ownership of a family. As th i s table shows, i f a ce i l i n g as low as 4 acres were set on land, and the ' surp lus ' generated was d i s t r ibuted among the landless households and'those with 1 acre of land or le s s , the average s ize of farm for the rec ip ients of land would s t i l l remain below 1 acre.^ If a l l of them received the same amount of land then the landless would be l e f t with only .71 acres. If they used part of th i s land for homestead (average area of homestead in rura l Bangladesh being about .18 acres), they would be l e f t with a l i t t l e over half an acre each for purposes A 4-acre c e i l i n g can be considered a low one in the l i g h t of the previous discussion on. a plough unit in Bangladesh agr icu l ture (see p. 2 ). Also, i t i s much lower than what some other scholars have recommended. Khan, for example, suggested a c e i l i n g of 7.5 acres, while Zaman ca l led for an 8-acre c e i l i n g ; see Khan (1972), op.eit., p. 142, and Zaman (1975), op.ait.3 p. 103. Table 30 Possible Cei l ings on Land and Consequences of D i s t r ibut ing the 'Surplus ' Land Among the Landless and Small Land Owners (owning 1 acre or less) in Bangladesh, 1977 Ce i l i ng (acres) Percentage of Families Having to Surrender Some Land ' Surplus ' Land (acres) Average Size of the Ownership Units of the Recipients After D i s t r ibut ion Amount Received by Each Landless.Family (acres) Post-Distr ibut ion Average Size of Ownership Units of Those Who Previously Owned Small (1 acre or Less) Land Areas (acres) 2 25.06 8,820,086 1.46 1.27 1.50 3 16.15 6,465,775 1.12 0.93 1.16 4 10.88 4,896,299 0.89 0.71 0.94 5 7.59 3,812,309 0.74 0.55 0.78 6 4.07 3,063,154 0.63 0.44 0.67 7 3.64 2,497,365 0.55 0.36 0.59 Source: Government of Bangladesh, Summary Report of the 1977 Land Oaoupany Survey of Rural Bangladesh, Dacca: Bangladesh Bureau of S t a t i s t i c s , 1977. 218 of cu l t i v a t i on . If a reform programme such as th i s i s implemented, the country w i l l be l e f t with a s i tuat ion where about half of i t s cu l t i vab le land would be held in farms sized 2 acres or less. It has been shown in Chapter II that i f land i s taken from the r e l a t i v e l y large land owners, and d i s t r ibuted among r e l a t i v e l y small land owners, the welfare of the landless wage labour may decline as demand for the i r services i s cur ta i led by sh i f t i ng land from those who depend heavily on wage-labour to those with family labour s u f f i c i en t to cu l t i va te most or a l l of the newly received land. Even where the landless gain ownership of land through the red i s t r ibut ion programme, the i r overal l s i tuat ion may s t i l l deteriorate i f t he i r income from the land i s less than the amount they lose from .. sources which were avai lable before the red i s t r i bu t i on . Before the red i s t r i bu t i on , members of landless households could have been earning the i r l i ve l ihoods by working, as pure tenants, or wage labourers, or both. Implementation of a low ce i l i n g on land may prompt many of those farm fami l ies which previously had land in excess of the c e i l i n g , and which used to rent out some of the owned land, to bring part or a l l of that land under the i r own cu l t i va t i on since family labour would become more abundant r e l a t i ve to ava i lab le farm land under the changed circumstances. This w i l l adversely a f fect the previous tenants who might have been better o f f as tenants than as rec ip ients of very small fragments of land. 219 One should rea l i ze that the taking away of some land from households owning more than the ce i l i n g w i l l normally involve payment of compensation to the dispossessed. Nothing short of a to ta l agrarian revolution can resu l t in zero compensation being required. Compensation, whatever i t s amount may be, has to be provided by the new recipients of land in one way or another. To ease the burden of payment of t h i s compensation two suggestions are generally made.7 ( i ) that the compensation be paid in terms of a f ixed percentage of y i e l d (an average y i e l d f igure i s to be used for th i s purpose) of the land, ( i i ) that the payment be staggered over a period.of years. The f i r s t suggestion i s based on the recognition of the fact that i f percentages of actual y i e l d (rather than average y ie ld ) are demanded as instalments by which the to ta l pr ice of land i s to be paid, one would probably create dis incentives s imi la r to those associated with sharecropping.tenancy. The suggestion of staggered payment i s made to further l ighten the burden on new occupants. Two factors w i l l play important roles in determining the period of time See in th i s regard Khan (1972), op.cit., pp. 140-141. 220 over which the payment can be spread out: ( i ) the surplus income that the rec ip ients of land can earn over and above family requirements within.the period of each instalment and the ease with which th i s can be 'captured' for compensation purposes, ( i i ) the a b i l i t y and intent ion of the government to secure wi11ingness of the rec ip ients of compensation to accept delayed payments. Even a ' r a d i c a l ' programme of land red i s t r ibut ion in Bangladesh, however, i s not l i k e l y to create a s i tuat ion where most of the farms would be able to generate s i gn i f i can t surpluses over and above family needs. If the ce i l i n g i s set at the low level of 4 acres, and the ' surp lus ' land i s d i s t r ibuted equally among a l l the landless and owners of less than one acre of land then * as Table 30 shows, the rec ip ients of land would be l e f t with an average ownership unit of only .89 acre. These small farms would const itute over half (58.51%) of a l l farms in Bangladesh. Owners of such farms can hardly be expected to come up with much of a.marketable surplus to make compensation payments. Table 30 shows that 10.88% of rural households would have to surrender some amount of land i f the 4-acre c e i l i n g i s imposed. Although apparently small in proportion to tota l population, these 221 households would be able to create a powerful opposition to any l e g i s l a t i on suggesting such a c e i l i n g . Even i f the ce i l i n g were to be l e g i s l a ted , the experiences of other countries suggest that i t s implementation would be very d i f f i c u l t . It i s i n s t ruct i ve to note Frankel ' s comments on the Indian experience: "In retrospect, i t was probably inev i tab le that a development strategy requir ing extensive land reform and i n s t i t u t i ona l change as preconditions for success should meet with powerful opposition from landed groups; and that in a p o l i t i c a l democracy, where landowning interests are heavi ly represented in the l eg i s l a tu re s , th i s resistance should manifest i t s e l f in a go-slow approach toward agrarian reform. ... most l e g i s l a t i on on tenancy reform and ce i l i ng s on 1andownership had not been e f f ec t i ve l y implemented."9 Another point to consider is that compensation demanded by the disaffected land owners for any surrendered land would l i k e l y be very high i f they are in a posit ion to influence government decision's on th i s matter. Consequently, so would be the price that rec ip ients of the ' surp lus ' land in a r e d i s t r i b u t i v i s t programme would have to pay. In th i s regard, one may refer back to the information on the landholding status of the p o l i t i c a l l y dominant in Bangladesh, which has been reported in Chapter I I I , pp. 87, 91. F.R. Frankel, India's Green Revolution: Economic Gains and Political Costs, Princeton: Princeton Univers ity Press, 1972, p. 4. Note here that in the aftermath of the zamindari system, c e i l i n g l eg i s l a t ions in what was then East Pakistan were associated with high compensation. Therefore, the subsequent red i s t r i bu t i on meant higher prices for land. See Chapter I I I , p. 83. 222 This would resu l t in a s i tuat ion where the r e l a t i v e l y wel l -to-do among e l i g i b l e rec ip ients of land would obtain the bulk of the ' surp lus ' leaving very l i t t l e for poorer households. The problem of high compensation and the subsequent high pr ice for the ' surp lus ' land might be reduced i f the power-elite f inds an alternate power-base to the land-r ich c lass. Such was the case during the implementation of the land reform programme in Japan during the late 1940'sJV By that reform, land owners having land in excess of 2.5 acres were made to s e l l the ' surp lus ' to the government. The pr ice that they received for the i r land was extremely low compared to standard market rates of that time. This was possible because the p o l i t i c a l leverage exercised by the r e l a t i v e l y land-r ich farmers declined s i gn i f i c an t l y over the years a f ter the Japanese peasantry was freed from feudal bondage by the reform of 1868. The rapid pace of indus t r ia l development replaced the predominance of r i ch land owners in p o l i t i c s by a r i ch class within the urban-industr ial sector. This allowed the government to get away with paying only nominal compensation to the disaffected farm fami l i e s . The power-elite in Bangladesh, however, seem not to have any such alternate power-base and, hence, no dras t i c measures which adversely a f fect the land-r ich can be expected from them. See World Bank, Land Reform: Sector Policy Paper, Washington, D.C., May 1975, p. 65. 223 Even i f the 4-acre c e i l i n g on land i s duly implemented and the proposed red i s t r ibut ion of land i s carr ied out, there would s t i l l remain an appreciable degree of d i s pa r i t y . i n the pattern of land ownership. About 59% of a l l farms would be l e f t with less than 1 acre of land, whereas about 11% of households would continue to reta in 4 acres of land. The smallest of the land owners may f ind i t very d i f f i c u l t to hold onto the i r land for very long. Various emergencies may force them to s e l l of f t he i r land to those 12 who can afford to buy i t . Therefore, in the course of time, one i s l i k e l y to f ind that the to ta l e f fect of the kind of reform d i s -cussed above has been rather neg l i g ib le . If the red i s t r ibut ion of land i s made only among very small land owners, and not the landless, the problem of too th in a d i s t r i bu t i on may be reduced. But i t has already been noted that th i s kind of red i s t r ibut ion may worsen the condition of the landless — the poorest among the rural population. Such an approach would defeat Some evidence of such emergency sales has been provided by Karim, who, in a family case-study of a landless labourer, found that the family had some land in the past which had to be sold for meeting medical expenses of a member of the family. See M.R. Karim, 'Family Case Study of Md. Jo ina l Abedin ' , Family Case Study No. 7, in Government of Bangladesh, Proceedings of the Bangladesh FAO Workshop on the Problems of Small and Subsistence Farms and Agricultural Labourers, Dacca: Bangladesh Agr icu l tura l Research Counci l , 1974. 224 the attainment of equity which must be a major objective of any land reform programme in Bangladesh. If the objectives of higher product iv i ty and increased equity are to be reconciled in Bangladesh agr icu l ture, one thus has to look to a l ternat ive programmes of land reform than those suggested by th i s par t i cu la r d i s t r i b u t i v i s t model. ( i i ) Reform Aimed at Cooperative Farming. Is i t possible to achieve the dual goals of equity and product iv i ty by red i s t r ibut ing ' surplus ' land (obtained by imposing an ownership ce i l i n g ) not quite so t h i n l y , and then forming cooperatives to provide anc i l l a r y functions so that none of the farms suffers a serious handicap of c r ed i t , farm machinery or other inputs? A review of the past experience of Bangladesh in th i s regard i s s u f f i c i en t to answer th i s important question. Experience of Bangladesh in cooperative farming A widely publ ic ized cooperative venture was launched in the d i s t r i c t of Comilla in the early 1960's through the i n i t i a t i v e of the Pakistan Academy for Rural Development (PARD; now the Bangladesh Academy for Rural Development., BARD). As a. p i l o t project, the programme was i n i t i a l l y confined to Kotwali .thana. A two-t ier system of cooperation was planned. Numerous primary cooperative soc iet ies were created within v i l l a ge s , and a' thana level central 225 cooperative society was to coordinate the a c t i v i t i e s of the v i l l a ge - l e ve l soc iet ies . The fol lowing four aspects were emphasized 13 as the charac te r i s t i c feature of the whole programme: 1. Formation of Agr icu l tura l Cooperative Societies in V i l l ages . The farmers were considered too numerous to be i nd i v idua l l y attended to by the Academy in providing them with information, assessing c r ed i t -worthiness, and a l locat ing necessary inputs. It was thought that group re spons ib i l i t y should be created to f a c i l i t a t e such operations. The resu l t was the creation of the .v i l l age level Krishi Samabaya Samiti (KSS - ' Agr icu l tura l Cooperative Soc ie ty ' ) . 2. Savings and Credit. The Academy f e l t that the farmers needed inputs l i k e better seeds, chemical f e r t i l i z e r s and.pesticides to produce higher levels of output; and to afford these they would require access to c red i t . Credit avai lable from sources such as f r iends , r e l a t i ve s , and money lenders were considered inadequate. Loans For a detai led descr ipt ion of these aspects see B l a i r , op. oit. 3 pp. 29-32. 226 extended by the t r ad i t i ona l money lender were found by the Academy to be exp lo i ta t i ve in nature. Funds from outside sources were thus to be provided through the KSS for some time so that i t could cope with the i n i t i a l demands for c red i t . But i t was hoped that the members would begin pooling the i r indiv idual savings to at ta in s e l f - s u f f i c i ency . To ensure the generation of savings in the KSS, every member was required to put some money, however small in amount, into an account with the society every week. The thana-level central society would extend loans to the members to a level TO times the i r to ta l savings. Input Subsidies. Subsidization of inputs such as better seeds, f e r t i l i z e r s , pest ic ides, and small power-pumps was thought to be necessary to popularize t he i r use. It was hoped that subsequently, when the users of these inputs became aware of t h e i r value in ra i s ing product iv i ty , the subsidy could be removed. 227 4. Training. Each KSS was to elect a manager and a "model farmer", both of whom were to report weekly to the t fcma- leve l Agr icu l tura l Cooperative Federation (ACF) fo r necessary t ra in ing and advice. The idea was to avoid the problems that an extension worker, sent from the outside, would, face in getting due recognition among the members of a KSS. The manager of the KSS was given t ra in ing in accounting, administrat ion, and management. The "model farmer" was given advice on how to adopt new approaches to farming and newer and better technologies. The cooperative experiment in Comilla was able to raise the level 14 of output among the members. The basic explanation for th i s was the huge subsidies paid to the members. In the words of B l a i r , "Massive and seemingly unremovable subsidies have been at the very heart of the Comilla program and after the f i r s t few years have become more and more a drain rather than a channel for ag r i cu l tu ra l development in Bangladesh. To begin w i th , the major enticement to get farmers to form a KSS i s the chance to get a.tubewell (or low-lift.pump i f t he i r v i l l a ge i s next to a r i v e r ) ; the cap i ta l cost of a tubewell, current ly reckoned at about Tk. 42,000, i s completely subsidized, and the operating cost (about Tk. 3,000 per year) i s almost 90 percent underwritten ..."IS See Q.M.A. Malek, 'Rice Cul t ivat ion in Comilla Kotwali Thana: the ro le of cooperat ives ' , Bangladesh Development Studies, Vol. IV, No. 3, Ju ly 1976, p. 363. B l a i r , op.oit., p. 45. 228 It was found in 1969 that the eight years of operation of the cooperatives required an investment of Tk. 9.3 m i l l i o n for ju s t the i r r i g a t i o n and mechanization projects , whereas the cumulative savings of the members amounted to only about Tk. 1.6 • I T 16 m i l l i o n . Members of the KSS were not representative of the lowest level ( in terms of land ownership) of farms in the d i s t r i c t . In 1964, 46% of the farms in Comilla had less than one acre of land, whereas, within the cooperatives, only about 15% of the members were found to belong to th i s low land ownership category.^ 7 Only those who had at least half an acre of land were able to seek membership in the cooperatives. Thus, the poorest, who were landless or near landless, had no d i rect ro le in the cooperatives. Akhtar Hameed Khan, the founder-director of the cooperative movement in Comilla himself stated that, " . . . i t was by no means a panacea for the misery of the landless. Nor was i t an attempt at red i s t r ibut ion of incomes ... i t could neither furnish f u l l employment nor lessen the d i spar i ty between owners of land and hired hands. In f ac t , better drainage, l i n k roads and These data are from S.A. Rahim, Rural Cooperatives and Economic Development of Subsistence Agriculture, Comil la: Pakistan Academy for Rural Development, 1972, (reported in B l a i r , op.cit., p. 46). Data on the land ownership of member and non-member farms were reported in B l a i r , op.cit., Table II (p. 15) and were o r i g i n a l l y contained in F. Akhter, Characteristics of the Members of the Comilla Cooperatives, Comil la: Pakistan Academy for Rural Development, 1964. 229 i r r i g a t i o n substant ia l l y enhanced the value of land and i t s rent. The unearned increment of the landowners was a hundred times more than the wages earned by the 1abourers."18 The managers of the cooperatives were owners of much larger holdings than were the members. In 1964, over 50% of the managers had at least 3 acres of land, the corresponding f igure for the 19 members being only 24.1%. The r e l a t i v e l y large farmers within the cooperatives were found to become gradually more powerful in getting the advantages of the cooperative programme channelled in the i r own personal favour. Data presented in Tables 31 and 32 show that, with the passage of time, more and more of the money avai lable as loans was. being taken up by these larger farms, thereby reducing the percentage number of the benef ic iar ies of the cred i t programme. A.H. Khan, 'The Comilla Project, A Personal Account ', International Development Review, Vol. XVI, No. 3, 1974, p. 5. Data contained in Akhter, op.oit., and reported in B l a i r , op.ait., p. 15. B l a i r further reports that in 1970 Mannan found over 53% of the managers to have ownership units sized over 5 acres. See B l a i r , op.oit., p. 15; o r i g ina l data in M.A. Mannan, Rural Leadership and Its Emerging Pattern in Bangladesh (With Special Reference to Comilla Kotwali Thana), Comilla: Bangladesh Academy for Rural Development, 1972. 230 Table 31. Size of Farm and Per Capita Indebtedness in the Comilla Cooperatives, 1969 Size of Farm Per Capita Indebtedness (Taka) Up to 1 acre 85 1 - 3 acres 1391 3 acres or more . 3112 Source: A.A. Khan, Rural Credit Programme of A g r i c u l t u r a l Cooperative Federation, Comil la: Bangladesh Academy for Rural Development, 1971, (reported in A.R. Khan, 'The Comilla Model and the Integrated Rural Development Programme of Bangladesh: An Experiment in . 'Cooperative Cap i ta l i sm ' , World Development, Vol': 7, No. 4/5, April/May 1979. Table, 14, p.. 412). Table 32. Percentage of the Members of Kotwali Thana KSSs Getting Loans (various years) Year Percentage of Members Getting Loan 1966-67 84.0 1968-69 70.1 1969-70 34.1 1970-71 33.5 Source: Pakistan Academy for Rural Development, Twelfth Annual Report, 1970-71, Comil la, (reported in H.W. B l a i r , The Elusiveness of Equity: I n s t i t u t i o n a l Approaches to Rural Development in Bangladesh, Ithaca: Rural Development Committee, Cornell Un ivers i ty , T974,. p. 56). 231 Defaults in repayment of loans was greater among the ones who borrowed r e l a t i v e l y larger sums of money. While 15% of the loans of up to Tk. 500 were reported to have been overdue in 1971, loans which were over Tk. 2,500 each had an overdue percentage of 27.5. This shows that a large number of the r e l a t i v e l y r i ch members of the cooperatives were able to serve the i r personal interests at the cost of those of the cooperatives. The managers of the cooperatives were also found to show more interest in u t i l i z i n g the i r pos it ion for personal gain than for the betterment of the members. The manager's 'domain of misdeeds' has •been outl ined in a comment made by Hussain, "How many things can a manager do to damage his society? He can steal money. He can juggle the accounts. He can hide the soc iety ' s records from the members. He can take the l i o n ' s share of the loan in his own name and in the names of his sons. He can cause the society to f a l l into arrears. He can show incompetence in organizing j o i n t projects. And he can carry out a l l these misdemeanors with imperturbable disdain for law . . . The manager of Durgapur was doing a l l all of these things."20 (Emphasis mine.) The inev i tab le conclusion of the discussion on Comilla cooperatives carr ied out above is that the programme suffered from a basic contra-d ic t ion between the existence of unequal pr ivate land ownership among the members and the supposition of equal access to opportunities made avai lable by the cooperatives. M.Z. Hussain, A Field Investigation into, the Management of Village Cooperatives in Comilla Experimental Area, Comilla: Pakistan Academy for Rural Development, 1966, (quoted in B l a i r , op.oit., pp. 60-61). 232 The Integrated Rural Development Programme (IRDP), which has been introduced in several other d i s t r i c t s since Bangladesh came into being, i s designed after the Comilla model. It has combined the Comilla type cooperatives with the new production technology of the HYVs. A document published by the IRDP outl ines the objective of the programme in the fol lowing words, "The cooperatives have been envisaged as a vehicle for economic development enabling the farmers to r a l l y together for protecting, themselves from domination by landlords, money-lenders of a semi-feudal society to develop a new leadership for challenging the t r ad i t i ona l vested interests."21 However, th i s document i t s e l f states that the Integrated Rural Development Programme " w i l l not d i r e c t l y help improve the l o t of the 22 landless farmers". Under these circumstances, the IRDP w i l l hardly be able to "develop a new leadership for challenging the t r ad i t i ona l vested i n te re s t s " . If the experience of Comilla cooperatives i s any ind i ca t ion , the more l i k e l y outcome of the programme would be to strengthen the hands of those t r ad i t i ona l vested interests . The reasons why one would feel skeptical about the p o s s i b i l i t y of combining Integrated Rural Development Programme, Proposals for the First Five'Year Plan, Part I I I , Dacca, January 1973, (quoted in Abdullah et al., op. ait., p. 246). See Abdullah et al., op.cit., p. 247. 233 growth and equity by launching programmes based on the Comilla model are worth recounting. Four important considerations in evaluating such programmes are: ( i ) the nature of access to the cooperatives enjoyed by d i f fe rent income groups, ( i i ) the nature of access to supplies of inputs enjoyed by these groups, ( i i i ) the nature of increments in income of the various groups as determined by ( i ) and ( i i ) above, ( iv) the nature of disposal of the increment by various groups. It has already been discussed how the f i r s t three of the above go against the r e l a t i v e l y small farms and the landless households. With regard to the disposal of the increments of income, the r e l a t i v e l y large farmers/may be expected to spend more on the education of t he i r chi ldren and invest more in trade and business. However, they may also be expected to buy more land. This land would mostly come from the impoverished non-member households who may f i nd i t necessary to s e l l of f part or a l l of t he i r land for meeting various family needs. According to a study done by the Harvard Centre for Population Studies, " . . . a process of land agglomeration s imi la r to that now being witnessed in the Punjab (India) 234 appears to be the most l i k e l y outcome: the dynamic implications of th i s are c lear. "23 What the experience of cooperative farming in Bangladesh should teach one i s that, without major a l terat ions in the d i s t r i bu t i on of access to land, cooperation among local households for solving local problems by the i r own i n i t i a t i v e would not be fea s ib le , and that the two main goals of increased product iv i ty and better d i s t r i bu t i on of income would remain mutually inconsistent. Success of cooperation would be possible only i f under-ut i l i zed labour and other resources are more e f f e c t i ve l y u t i l i z e d in production a c t i v i t i e s through a pooling of ava i lable resources, and i f the income i s d i s t r ibuted more equitably. If cooperation i s l imited only to anc i l l a r y functions then even an i n i t i a l ega l i ta r ian d i s t r i bu t i on of land ownership cannot at ta in the goals of dynamic e f f i c i ency . Most farms would s t i l l be unable or unwi l l ing to take the r i sks involved in making investments necessary for the agr icu l tura l sector as a whole to move up to a higher production funct ion. In short, the problems faced by the r e d i s t r i b u t i v i s t model in atta in ing dynamic e f f i c iency would continue to ex i s t . ( i i i ) Reform Aimed at State Farming. If i t were possible for the state to abolish ownership of land in pr ivate hands then problems of d i spar i ty ar i s ing out of unequal d i s t r ibu t ion ,o f land ownership could obviously be done away with. Center for Population Studies, Bangladesh Land, Water and Power Study, Cambridge (Mass.): Harvard Univers i ty, 1974, (quoted in Abdullah et al., op.cit., p. 251). 235 However, state farming with wage labour would share many of the problems encountered in wage labour cu l t i v a t i on of pr ivate holdings. State farming c a l l s fo r a very high degree of bureaucratic involvement by the government. This would es sent ia l l y create a gap between those who are to undertake the actual process of cu l t i va t i on and those who perform managerial functions. It w i l l be d i f f i c u l t t o . i n s t i l l a fee l ing of true par t i c ipat ion among the cu l t i vator s in a s i tuat ion where the ' d i s t an t ' state apparatus i s the primary entrepreneur. Even i f the government wanted to secure par t i c ipat ion of l o ca l peop l e in management, the majority of cu l t i vator s would s t i l l feel remote from.the management since only previously land-r ich households can be expected to have members in t he i r fami l ies educated enough to perform the managerial functions. In f a c t , t h i s could operate as a new source of intergroup d i spar i ty within the rural areas. Although the purpose of the state farms may be to exercise greater government control over farming a c t i v i t i e s in order to at ta in equity and product iv i ty , the vast ly scattered farm-households in Bangladesh w i l l not lend themselves eas i l y to bureaucratic management by outside agents. When the government outl ines i t s plan to nat ional ize a l l land into state farms, opposition can be expected to come not only from the r e l a t i ve l y . l a r ge farms but also from small ones. The large farmers do not want to give up the pr iv i leged posit ion 236 they enjoy; they would stand to lose the basis of t he i r economic, p o l i t i c a l and social domination in the event of a state takeover, as t he i r ro le as patrons would no more be in existence. Opposition from the r e l a t i v e l y smaller farms, on the other hand, would ar i se from in-bred suspicions towards government actions resu l t ing from the i r abuse by publ ic bodies in the past, and the i r observation that r e l a t i v e l y large land owners have been able to influence public projects in ways which have taken away some of the benefits o r i g i n a l l y 24 meant for the poor. The sense of security enjoyed by a land owner, however small the ownership unit may be, cannot be adequately 25 compensated for by turning him into a wage labourer of a state farm. State farming w i l l tend to re ly heavily on public borrowing to t ide over various problems of farming. It has been discussed in the previous section how government involvement in the Comilla cooperative movement resulted in a high level of government subsidizat ion of agr i cu l tura l operations there. Subsidization of state farms can be expected to reach even higher levels since the ent i re management of these farms would be the d i rect re spons ib i l i t y of government. State farming, in the context of Bangladesh agr icu l ture, would be un l i ke ly Works Programme under the Basic Democrats during the 19601s provides one such example. For de t a i l s , see R. Sobhan, Basic Democracies, Works Programme and Rural Development in East Pakistan, Dacca: Bureau of Economic Research, Univers ity of Dacca, 1968, pp. 118-124. Although in Bangladesh a government job i s highly cherished as a secure source of income, i t cannot replace the sense of security which i s provided by ownership of land. 237 to create work or se l f -he lp incentives replacing or enlarging upon those present under t r ad i t i ona l family farming. In such circumstances, i t would be d i f f i c u l t to at ta in s e l f - s u f f i c i ency within the state farms; so a perpetual dependence on government would appear inev i tab le. B l a i r points out that r i sk-avers ion may be one of the characte r i s t i c features of bureaucrat ical ly managed agr icu l tura l 26 development programmes in Bangladesh. He maintains, "In general, r i s k i s a subject treated by economists in the context of economic decision-making, but i t also has a bureaucratic dimension. Just as small farmers tend to be deterred from innovation by the i r ca lcu lat ion of r i s k , so the administrator in a rural development programme is v i t a l l y concerned to minimize r i s k to his own career by pleasing his superiors."27 It i s , therefore, probable that the bureaucracy in charge of state farms, when faced with a choice between investing in an undertaking which has great promise for agr i cu l tura l growth but also high r i sk s ( for example, the construction of earthen embankments to hold back f loods, which w i l l pay o f f by increasing the level of production only i f i t does not rain so much as to overflow or wash away the embankments) and one that i s r e l a t i v e l y r i s k - f r ee , though much less promising (for example, a programme of more e f fec t i ve H.W. B l a i r , 'Rural Development, Class Structure and Bureaucracy in Bangladesh', World Development, Vo l . 6, No. 1, 1978. Ibid., p. 73. 238 weeding of agr icu l tura l land), would se t t l e for the l a t t e r . And in the absence of an att i tude to bear possible r i sks i t would be hard to move Bangladesh agr iculture away from i t s present stagnation. ( iv) Reform Aimed at Group Farming. In the context of the present analysis i t i s useful to fol low Reed in l i s t i n g the fol lowing basic character i s t i c s of a group 28 farming un i t : ( i ) land and capita l assets are owned j o i n t l y by the members of the un i t , ( i i ) the bulk of the land i s cu l t i vated j o i n t l y by members of the un i t , ( i i i ) a l l members part ic ipate in making decisions on production, d i s t r i bu t i on and investment, ( iv) the products and p ro f i t s of the unit are shared among i t s members. Problems and Prospects of I n i t i a t i on Although the proposal of j o i n t ownership of land and cap i ta l assets has the theoret ica l promise of doing away with the inequa l i t ie s in income ar i s ing out of the unequal d i s t r i bu t i on of land ownership among d i f fe rent households, the very introduction of group farming practices into the rural society of Bangladesh i s l i k e l y to be a formidable task. The t r ad i t i on of family farming in the country w i l l E.P. Reed, ' Introducing Group Farming in Less Developed Countries: Some Issues ' , in P. Dorner (ed.), Cooperative and Commune: Group Farming in the Development of Agriculture, Madison: The Univers ity of Wisconsin Press, 1977, p. 360. 239 not readi ly accept the idea of group ownership and management of assets. Opposition to any suggestion for an immediate adoption of group farming operations i s l i k e l y to be very strong among land-owning households for reasons s imi la r to those already discussed under the model of state farming. Thus one cannot expect to see a widespread and spontaneous enthusiasm for group farming among these rural households. It would be naive to believe that group farming arrangements can accomplish j o i n t ownership of land and other cap i ta l assets overnight. Even under the post-revolutionary circumstances, in China the introduction of a group farming system 29 was possible only after a number of intervening phases. Thus, although the post-revolution Chinese government was committed to the abo l i t ion of pr ivate property in land, agr icu l ture was managed, by and large, through household decisions on production and marketing of the produce unt i l about 1954. The gradual conversion of the family farming system in China into communes where ownership, management and d i s t r i bu t i ona l aspects of farming are t o t a l l y c o l l e c t i v i z e d , went through an intermediate phase in which mutual-aid teams popularized the idea of cooperation. In Bangladesh agr icu l ture, where the d i s t r i bu t i on of land ownership i s very unequal, one cannot expect a convergence of interests of the d i f fe rent households in matters of cooperation in See United Nations, Progress in Land Reform, Sixth Report, New York, 1976, pp. 57-58. 240 ag r i cu l tu ra l production. Therefore, the i n i t i a l phase toward the establishment of group farming in Bangladesh may well require that the present skew in the d i s t r i bu t i on of land ownership be reduced through the implementation of land ce i l i n g l eg i s l a t i ons and the subsequent d i s t r i bu t i on of the surplus land among the landless and the small land owners. There has to be a simultaneous programme for popularizing group farming a c t i v i t i e s among the peasantry so that the rec ip ients of land can themselves perceive the red i s t r ibut ion process as the f i r s t step towards group farming. It w i l l , however, be very d i f f i c u l t to get the i r voluntary support for such programmes unless the f i r s t few group farming units can achieve s i gn i f i can t success in meeting the objectives of land reform. The required ce i l i n g l e g i s l a t i on would prove to be a d i f f i c u l t task for an administration which r e l i e s heavily on the r e l a t i v e l y r i ch land owners for p o l i t i c a l su rv iva l . However, with the increasing pauperization of the households in lower, economic classes, pressures for land red i s t r ibut ion w i l l l i k e l y be increasing and, as a consequence, only a government which earns legitimacy from the peasantry at large through the implementation of a programme of land red i s t r ibut ion can expect to hold power for any length of time. It has been seen that even a ' r a d i c a l ' land red i s t r ibut ion programme in Bangladesh would leave most of the rec ip ient households very small in s i ze. Cooperation in a range of a c t i v i t i e s s imi la r to 241 what was t r i e d in Comilla could r e l a t i v e l y eas i ly be accomplished among them. However, when i t comes to pooling of land for j o i n t cu l t i v a t i on under group ownership, cooperation may not be so readi ly forthcoming. This would be quite natural in an environment where land ownership has t r a d i t i o n a l l y been associated with socia l and economic secur i ty. To make a dent in th i s t r ad i t i on i t w i l l be necessary to f i r s t f i nd out i f there are groups within the agr icu l tura l sector which would lend themselves to group farming operations more eas i l y . It would probably be easier to inculcate the idea of group farming among the smallest of the small farmers who, with t he i r meagre land and other inves t ib le resources, f ind i t d i f f i c u l t to earn enough from the i r indiv idual farms to meet family 30 needs. Those who were previously landless could, for example, be . considered f i r s t . Making land d i s t r i bu t i on to them conditional upon the i r jo in ing the group farms may also help i n i t i a t e the programme. The landless would understandably be less opposed to such an idea than those who are 'addicted ' by the i r indiv idual ownership of land. Whether i t would be possible to induce the other land-owning .households into spontaneously jo in ing the group farming project would, again, depend to a great, measure upon the success of the ' p i l o t ' group farming units in meeting the various objectives of a After a ' r a d i c a l ' land red i s t r ibut ion programme i t would be harder for them to f i nd employment as sharecroppers or wage labourers to subst itute income from the i r own farms. See discussion in Chapter II, (pp. 34-38). 242 land reform programme in Bangladesh. In what fo l lows, an attempt i s made to evaluate the prospects of group farming in meeting these objectives for the country. Whether there are ways and means in which the problems of group farming, as revealed by the experiences of other countr ies, can be successful ly coped with are also discussed. Meeting the Objectives of Land Reform Equality in access to land enjoyed by the members of a group farm can help achieve the objective of a more equitable d i s t r i bu t i on of income only i f a number of conditions are met. There has to be a norm for the sharing of output based on work-performance. In assigning jobs to the members, indiv idual differences have to be taken into account so that these differences do not create undue d i spa r i t i e s in the incomes earned by them. Also, rotat ion of tasks among indiv iduals have to be encouraged so that no one member feels confined to a ' d i f f i c u l t ta sk ' . To make sure that those households which have more minors and other dependents r e l a t i ve to able-bodied adults do not suf fer disproportionately from the norm of 'sharing income in accordance with work^performance 1, there have to be certa in socia l services l i k e r e l i e f for ch i l d care (which, for purposes of population cont ro l , could be made a decreasing function of the number of chi ldren 31 in the household), old-age pension and health-insurance programmes. That labour components of d i f fe rent households within the communes in China was a source of d i spar i ty in income-earning opportunities for them has been reported by Stavis. See B. Stav is , People's Communes and Rural Development in China, Special Series on Rural Local Government, Ithaca: Rural Development Committee, Cornell Univers i ty, November 1974, p. 164. 243 Po l i c ie s have to be adopted which can e f f ec t i ve l y control d i spa r i t i e s in income-earning opportunities among group farms in d i f fe rent areas. This w i l l c a l l for a l locat ion of agr icu l tura l supplies and technical aid to the group.farms in such a way that the t r a d i t i o n a l l y backward areas can move up. Also, i t may.require that , as fa r as poss ible, no area gets a monopoly of supplying one or two agr icu l tura l commodities which are valued highly in the 32 urban markets. It i s true in Bangladesh, as in most other less developed countries of the world, that the r e l a t i v e l y r i ch rural households have a d i ve r s i f i ed . se t of assets, not a l l of which are rural-based. Thus, even after access to land i s equalized, income-earning opportunities of d i f fe rent members may vary s i g n i f i c an t l y i f certain conditions are not st ipulated for a household to be able to j o i n the group farms. If a household, which i s disproport ionately we l l - o f f by v i r tue of i t s off-farm assets, i s allowed to obtain membership in a group farm then the basic p r i nc ip le of equal i ty of a l l members w i l l be v io lated. Also, not a l l members in that case w i l l perceive a s i m i l a r i t y of interest in agr icu lture. As a possible means of reducing the chances of such consequences, i t may be required In China, areas close to c i t i e s were not permitted to spec ia l i ze in commodities highly priced in the urban markets so that these areas could not become disproportionately wealthy. The pol icy in th i s regard was "take grain as the base and have a l l round development"; see B. Stav is , Making Green Revolution: The Politics of A g r i c u l t u r a l Development in China, Rural Development Monograph No. 1, Ithaca: Rural Development Committee, Cornell Univers i ty, 1974, p. 253. 244 to r e s t r i c t membership to those households which o f fe r t he i r f u l l - t ime services to the group farms. A common c r i t i c i s m l a i d against j o i n t ventures in agr icu l ture is that they can resu l t in low incentives among the workers, making the object ive of ra i s ing product iv i ty per unit of land or labour hard to achieve. This c r i t i c i s m i s quite va l i d in those cases where the members f ind i t d i f f i c u l t to ident i f y with the organization to which they belong, and where they do not perceive a d i rect re lat ionsh ip between the work they perform and the remuneration they receive. Group farming can reduce the d i s incent ive problem ar i s ing out of i s o l a t i on of the members insofar as i t can create opportunities for a l l to part ic ipate in the decision-making process. Further, an e f fec t i ve incentive mechanism within the group farms might involve remunerating each member according to the work-points he earns. This w i l l establ i sh a d i rect re lat ionsh ip between indiv idual work-efforts and subsequent rewards, thereby reducing the problem of d i s incent ive. The po l icy of remunerating the members according to work-points earned need not require that a l l payments to them be deferred t i l l t he i r points have been adequately evaluated, which may take as long as a f u l l cropping season. Each member could be paid an 'advance 1, perhaps each month, from the.tota l income that he would merit at the end of the season on the basis of his overal l work-performance. 245 Inspite of various forms of incentives that may be b u i l t into the system, i t i s quite l i k e l y that i f group farming were to be introduced in Bangladesh, productive e f f i c i enc i e s would suffer in the immediate short run due to the confusions of t r an s i t i on . To minimize th i s loss in e f f i c i ency , the group farms would have to be formed in such a way that each member feels at ease with the land he works on and the other members he works with. Therefore, f a m i l i a r i t y with the land and i t s people has to be taken as one of the basic c r i t e r i a for determining the e l i g i b i l i t y of an indiv idual to j o i n a par t i cu la r group farming unit . In Bangladesh a para ( l o ca l i t y ) has usually been the approximate ent i ty around which the i nd i v i dua l ' s s o c i a l , economic and p o l i t i c a l l i f e has revolved. These can act as i n i t i a l cohesive units of farming under the changed 33 circumstances as we l l . However, the age-old s t r a t i f i c a t i o n of the households in terms of t he i r s o c i a l , economic and p o l i t i c a l status cannot change simultaneously with a change in t he i r land-owning status. The creation of a community consist ing of a l l the members of a l o c a l i t y , who w i l l have convergent interests in agr i cu l ture , w i l l require careful 34 planning involving a var iety of t rans i t i ona l arrangements. These units may ult imately need to be expanded beyond such a local level so that they can d i ve r s i f y t he i r farming a c t i v i t i e s . This, however, has to be done gradually so that unfami l i a r i t y among members does not create problems. Some of these arrangements are discussed in a subsequent section where various p o s s i b i l i t i e s of popularizing group farms are considered in some d e t a i l . See pp. 252-256. 246 To benefit from the commonality of experience of members in achieving product iv i ty increases, the small teams of workers within the group farms, who are to be assigned spec i f i c jobs, may be conveniently formed by getting the neighbours in the d i f fe rent l o c a l i t i e s together. They are f ami l i a r with the land, they know each other, and they may also.have a past of organizing l imi ted neighbourly se l f -he lp projects. However, th i s should be viewed as a t rans i t i ona l arrangement only. The creation of work-teams based on geographical proximity of the members would tend.to bunch together members of the same kinship who usually l i v e close to each other. If these work-teams are allowed to maintain t he i r o r i g ina l composition for very long, past kinship r i v a l r i e s among them are bound to re-emerge which can affect overal l product iv i ty of the farm. Therefore, through frequent meetings of a l l members and through regular exchanges of members of d i f fe rent work-teams, which would create an opportunity for them to further interact with each other, an atmosphere may be created where these teams need no longer be formed on the basis of neighbourhoods:or any other c r i t e r i a which may ult imately r e f l e c t previous po la r i zat ions , and may, in e f f ec t , put obstacles in the way of achieving e f f i c i ency . It was ea r l i e r pointed out that generation of e f fec t i ve employment opportunities for an increasing labour force would const itute another major objective for a land reform programme in 247 Bangladesh. Group farming, insofar as i t can involve a l l members in the decision-making process, would help in ensuring e f fec t i ve labour u t i l i z a t i o n . Thiswould be so because the labour supply of a l l the households put together would const itute the tota l labour supply made avai lable to the group farm, and i t would be uneconomical not to make the f u l l e s t use of th i s labour. It may also be argued that workers with varying interests and spec ia l i zat ions can hope to be accommodated within a group farm since i t would have a wider scope of operations. Fu l le r u t i l i z a t i o n of labour (and the complementary ag r i cu l tu ra l inputs) within the framework of a group farming system can, a f ter the short run confusion has disappeared, ra ise the prospects of achieving greater product iv i ty per unit of the scarce resource — 35 land. This would have the ef fect of ra i s ing the level of marketable surplus. Therefore, .the group farms would be in a better pos it ion to systematical ly bui ld up adequate f inanc ia l reserves to,be able to make investments in in f ras t ructura l and v iable technological projects. This, in tu rn , can increase product iv i ty and create more employment opportunit ies. Fu l le r u t i l i z a t i o n of labour can also f a c i l i t a t e a more productive exp lo i tat ion of resources other than land. For example, the vast water resources of the country can be used for f i s h farming, duck ra i s ing and cu l t i va t i on of water-borne crops. 248 There i s a great deal of interdependence among the various objectives of land reform within a group farming model. The e f fec t i ve employment of labour and u t i l i z a t i o n of a l l other ava i lable resources depends upon the equal i ty in income-earning opportunities of members. And i t i s only through the e f fec t i ve u t i l i z a t i o n of labour and other ava i lable resources that the c r i t e r i a of dynamic e f f i c iency can be f u l f i l l e d . The i n i t i a l task, however, i s to ensure that there i s an equal ity in the access to land and other resources enjoyed by the d i f fe rent households. It i s apparent from the above discussion that there has to be a considerable degree of complementarity between the objectives of the government and of the group farms for the success of group farming to be achieved. When i t comes to the question of ra i s ing agr icu l tura l product iv i ty , one can perceive a convergence of interests between these part ies . However, the interests of the group farms may not converge with those of the government in such matters as consumption on the farms and the surplus to be extracted from 36 agr iculture (for exporting or for meeting urban consumption needs). Such problems cannot be resolved through coercion. Rather, by adopting measures such as of fer ing the agr i cu l tura l sector better See G. Etienne, 'China 's Agr icu l ture: Present S i tuat ion and Prospects ' , in P. Dorner (ed.), Cooperative and Commune:^ Group Farming in the Economic Development of Agriculture, Madison: The Univers ity of Wisconsin Press, 1977, pp. 151-152. 2'49 terms of trade for crops required for the urban consumption/export sector, the government can encourage the production and marketing of those agr icu l tura l commodities. This would, however, involve a gestation gap which has to be tolerated by the government i f the group farms are to be established on a strong foot ing. Whether the government would do th i s or not would, again, depend upon the nature of i t s power-base. Only i f th i s base i s mainly provided by the peasantry at large can one hope for such an att i tude from the government. Although the ex i s t ing s i tuat ion in Bangladesh does not seem ideal i n . th i s regard, one can contemplate a not-too-distant future where the p o l i t i c a l al legiance of the peasantry w i l l become 37 a necessity for keeping p o l i t i c a l power in Bangladesh. This, however, i s not to say that con f l i c t s of interests between the government and the group farms w i l l cease to ex i s t in such an eventual ity. The broadly defined interests of the nation as a whole can s t i l l , at times, become divergent from the r e l a t i v e l y narrow range of interests of the group farms. However, the important point to note i s that there would be greater chances of reconc i l i a t i on between the contenders when they have basic interests in cooperating with each other. The above discussion indicates that there are considerable prospects of meeting the objectives of land reform through group farming. However, management of group farms can be a d i f f i c u l t task. See p. 240. 250 The problems of management under such arrangements and some possible approaches toward solving them are discussed below. Management of Group Farms The management of group farms in matters of organization of work and of bookkeeping i s going to be a complicated task. During the i n i t i a l period the members of these farms cannot be expected to display the necessary expertise for managing the i r farms. As a consequence, these farms w i l l inev i tab ly require help in th i s regard 38 from outside sources. This can ra i se a problem of suspicion among the members toward the management since there i s , among the v i l l a g e r s , a t r ad i t i ona l d i s t rust toward ' ou t s ide r s ' . There may be a number of 39 ways in which th i s problem can be reduced. 1. The members should have a large say in the h i r ing of outside managers. 2. The members should have the r ight to remove any of the managers through consensus o"r majority dec is ion, i f need be. 3. Rather than keeping the ent i re decision-making function confined to the outside managers, there should be attempts to incorporate into i t the experiences of older peasants. Even when the h i r ing is done from 'outs ide ' sources, the supply of s k i l l e d managers would be short in Bangladesh. Therefore, the government would have to i n i t i a t e programmes which can provide t ra in ing to those who already have the basic qua l i f i c a t i on s . For a f u l l e r discussion of th i s subject see Reed, op.cit., pp. 370-371. 25:1 During the phase that the group farms continue to depend upon outside help for management purposes, members of these farms w i l l themselves have to be trained to gradually take up the managerial r e spon s i b i l i t i e s . Since the managers would not have powers to make decisions on production and d i s t r i bu t i on within the group farms, one would not expect to be faced with a s i tuat ion where the 'outside i n te re s t s ' would not be wi11ing to re l inqu i sh the i r 'power 1. After the management has been taken over by the members themselves, further cautions may be necessary to ensure that the managers do not evolve 40 into small pr iv i leged groups within the farms. This w i l l require that there be a large number of members trained in managerial functions so that the r e spon s i b i l i t i e s of management can be alternated among groups of these members. Also i t i s necessary that the control of management functions by some of the members does not c u r t a i l the sense of involvement of a l l members in the decision-making process. Popularizing Group Farms. It was pointed out e a r l i e r that i t would probably be r e l a t i v e l y easy to form the i n i t i a l group farms among those households which were previously landless. These households with t he i r meagre resources would not, however, be able to establ i sh the i r farms on a strong footing In his study on the co l l e c t i ve agr iculture in Soviet Central As ia , Khan reports: "The spread between real earnings of the top management and the earnings of the f i e l d worker would appear to be quite large. " See A.R. Khan and D. Ghai, ' Co l l ec t i ve Agr iculture in Soviet Central A s i a ' , World Development, Vol. 7, No. 4/5, April/May 1979, p. 83. 252 without government support. However, i f the government i s not committed to the basic pr inc ip les of group farming then one may expect th i s support to i n te r fe re , on occasions, with the group involvement in the decision-making process. Under such circumstances many households outside the group farms w i l l not feel inc l ined toward jo in ing these schemes. To at t ract addit ional pa r t i c i pa t i on , the pioneering group farms, with se lect ive support provided by government, w i l l have to set an example for others. To reduce uneasiness due to an abrupt.end of t he i r ownership status as ind iv idua l s , members may be allowed to keep a certa in percentage of the i r land for personal c u l t i v a t i on . This, however, has to be monitored very ca re fu l l y by the government i f newer d i spa r i t i e s and con f l i c t s of interest .are to be avoided in future.. The experience of Mexico in th i s regard shows that, in the absence of pooling a l l operations and decision-making functions in the associat ive groups (the ejidos), cu l t i va t i on could, not be made group-oriented, and the success of the land red i s t r ibut ion programme of the 19301s in bringing about equal i ty in the landholding status of rural 41 households was ser iously threatened by 1960. One way in which indiv idual households may be encouraged to c o n t r i -bute t he i r land to the group farms would be to provide the members World Bank, Land Reform: Sector Policy Paper, Washington, D.C, May 1975, p. 72. 253 42 with credi ts on easier terms. The government can also make a pol icy of red i s t r ibut ing part of the surplus land among the group farms when they are formed. This w i l l resu l t in a higher par t i c ipat ion rate in group farming a c t i v i t i e s . Group farms also need to organize the i r own insurance schemes (for example, for health and education) to cover a l l members which may act as an incentive to households to work together. Cooperatives of group farms at the thana and subdivision levels can also hedge 43 against other r i sks and uncertainties of agr icu l tura l operations. Further incentives may be provided by favouring group farms in matters of cred i t and government procurement. But, again, a de l icate balance has to be maintained between too much and too l i t t l e government involvement. Excessive autonomy within the group farms during i n i t i a l phases w i l l be unwarranted since these farms need ' p ro tect ion ' much the same way as the ' i n fant indus t r ie s ' in the urban areas do. Excessive government support and cont ro l , on the other hand, w i l l d e f i n i t e l y defeat the purpose of group involvement in a l l phases of farming. 42 The government must take care not to restage the Comilla drama. Evaluation of potential product iv i ty and savings of a farm (with which standing loans can be repaid) has to precede the extension of c red i t . 43 . . . For the ro le of interfarm cooperation in group farming a c t i v i t i e s see O.M. Schi 11 er, Cooperation and Integration in A g r i c u l t u r a l Production, Concepts and P r a c t i c a l Application: An International Synopsis, Bombay: Asia Publishing House, 1969, p. 194-195. 254 In implementing a group farming arrangement, one must also 44 take account of var iat ions within the country. In Bangladesh certa in areas of the country suffer heavily from f looding. These areas could be cu l t ivated more e f f ec t i ve l y i f the problem of f looding could be reduced. Although f lood protection in the country as a whole would require massive investments, some l o c a l i t i e s can be made r e l a t i v e l y f lood-f ree with modest investments. For example, parts of the d i s t r i c t s of Comilla and Faridpur, which are moderately flooded (50% of land under 1 m of water), would be able to u t i l i z e the affected land better i f th i s flooding is control led by a project involving the construction of earthen embankments and digging of 45 canals. On top of the other incentives given to the agr icu l tura l households to j o i n group farming, the government can arrange to support such projects on the condition that group farming be practiced in these areas. S im i la r l y the sinking of deep tubewells in the r e l a t i v e l y dry regions of the country (see Map 3, p. 117) can be arranged under s imi la r conditions. The government might also encourage group farming a c t i v i t i e s in various l o c a l i t i e s by helping to select those crops which are best suited to the given phys ica l -b io log ica l condit ions, and by sett ing up interregional exchange f a c i l i t i e s through which each Variat ions in s o i l composition, c l imat ic condit ions, cropping pattern and size-tenure structure are discussed in Chapter III (pp. 112-129). A f u l l e r account of the regional var iat ions would have to include considerations of numerous other variables (see footnote 75, p. 135). The flooding f igures are provided by Johnson, op.ait., p. 28. 255 46 region can acquire agr icu l tura l inputs and consumption items. This w i l l help to make agr icu l ture more productive and is l i k e l y to increase the attractiveness of group farming. Given that the government i s committed to the establishment of s e l f - r e l i ance among the members of the group farms, i t would be able to reduce i t s part in the exchange programme by l e t t i n g the thana and the d i s t r i c t level cooperatives of group farms take over more of t he i r own r e spon s i b i l i t i e s . However, government involvement has to continue in certa in aspects. For example, when two d i f fe rent cooperatives earn s i g n i f i c an t l y d i f fe rent levels of income due to factors related to cropping pattern, c l imate, or l ocat ion , i t i s the government who has to take up various measures to equalize the i r income for ascertaining balanced regional.development. Thus, the "second generation problems" c a l l fo r continued v ig i lance from the government. Conclusion The above discussion has pointed out that , while a l l the models of land reform considered in the context of Bangladesh would be d i f f i c u l t to implement (though in varying degree), the only model See Maps 4 and 5, (pp. 120, 124) for the reg iona l izat ion of the country in terms of various important crops. 256 which seems capable of creating a long-term viable solut ion for the agr icu l tura l sector of the country i s that which involves group farming. It may be noted that th i s model r e l i e s on a continuous process of change in man-land relat ionships for atta in ing the goals of increased equity and onr-going product iv i ty enhancement. It s tarts of f with a red i s t r ibut ion of land which, however, i s not seen as an end in i t s e l f ; rather i t i s considered as an i n i t i a l step which cu r t a i l s the incidence of sharecropping and which establishes a more equitable d i s t r i bu t i on of land ownership as the basis fo r f u l l e r cooperation among neighbouring households in d i f fe rent spheres of agr icu l tura l production. Group farming requires that the cooperation be extended not only to anc i l l a r y functions (such as c r ed i t , marketing and extension) but also to the ownership, decision-making and management aspects of farming. It also c a l l s fo r stepped-up government involvement during the i n i t i a l phases, though not to the extent that would be involved under state farming. Group farming, therefore, borrows from the other models in seeking to traverse the path from an inega l i ta r ian small-scale farming to a j o i n t l y owned and managed farming system. Although the problems of landlessness and 'mini^farming 1 are a common feature of a l l regions within Bangladesh agr icu l ture (which suggest that pooling of land and other resources is a necessary condition a l l over the country for establ ishing a v iable ag r i cu l tu ra l 257 sector) , some regions have the character i s t i c s which would more read i ly lend themselves to group farming a c t i v i t i e s than others. These regions are characterized by a highly skewed land ownership d i s t r i bu t i on . Here, the search for a ' surp lus ' land over leg i s la ted ce i l i ng s and also the search for households which could r e l a t i v e l y eas i l y be induced to j o i n in group farms can hope to meet with greater success. After the i n i t i a l group farming units have been establ ished, they have to demonstrate considerable success in achieving the various goals of land reform i f the households in other areas are to j o i n in s imi la r un i ts . The discussion in Chapter III suggested that in re la t ion to the landed interests the ru l ing e l i t e s in Bangladesh have t r a d i t i o n a l l y been "cooperative e l i t e of the dominant k ind " , which means that these e l i t e s , though not dictated wholly by the landed interests in formulating agr icu l tura l p o l i c i e s , nevertheless, were inc l ined to pursue po l i c ie s which would keep these r e l a t i v e l y r i ch households 47 s a t i s f i ed . The present.government has emphasized that "development planning and implementation should be increasingly by the poor 48 themselves". However, in p ract i ce , no signs are yet v i s i b l e to suggest a s h i f t in the ru l ing e l i t e away from the i r t r ad i t i ona l al legiance to the r e l a t i v e l y land-r ich households. Meanwhile, the demands of the r e l a t i v e l y poor within the rural areas are being shaped not only by the i r growing poverty, and See pp. 87, 91. See quote on page 90. 258 p o l i t i c a l pronouncements such as that c i ted above, but also by the f a i l u r e of a number of 'models' of land reform already attempted. The inadequacies of r e d i s t r i b u t i v i s t programmes in Bangladesh have been discussed in Chapter III.as well as in the present chapter. The f a i l u r e of the widely publ ic ized programme of l imited cooperation which found expression in the Comilla Cooperatives has also been noted. A l l th i s helps shape the perception of the poorer classes in weighing the d i f fe rent a l ternat ives so that the objective conditions necessary for the i n i t i a t i o n of a group farming system may not be as far away in Bangladesh as i t may have appeared in e a r l i e r decades. 259 VI. CONCLUSIONS Bangladesh agr iculture has been characterized not only by low levels of product iv i ty per unit of land and labour but also by a considerable skew in the d i s t r ibu t ion of income among the agr icu l tura l households. The basic source of d i spar i t y i s i den t i f i ed as the unequal d i s t r ibu t ion in ownership of land. This d i spar i t y i s only moderately improved by the access to land which some of the land-poor households get as wage labourers or tenant -cu l t ivators . About 33% of households in rural Bangladesh have no farmland of the i r own. Non-farm employment opportunities being very scarce, most adult members of these households seek access to the land of others. This constitutes a high demand on the land of those r e l a t i v e l y few households who do not have enough family labour for cu l t i va t i ng a l l of the owned land without outside help (or who are not inc l ined to do so even i f they have the necessary labour). Since a l ternat ive employment opportunities are very l im i t ed , the wage labourer or the tenant-cu l t ivator cannot bargain to any great extent with his potential employer. The economic dominance of r e l a t i v e l y large land owners (along with the i r higher education, access to information and urban connections) serves to reinforce the i r stronger p o l i t i c a l and social s i tuat ions. A l l of th i s goes to perpetuate the ex i s t ing s t r a t i f i c a t i o n of the rural society. 260 With the passage of time Bangladesh agr iculture i s becoming more and more characterized by farms of very small s izes. It i s f u t i l e to argue that , since these farms are able to cu l t i va te land more intens ive ly than the larger farms due to r e l a t i v e l y abundant family labour, one should advocate a perpetuation or extension of such small family farming units. As Herring has noted, " . . . the poverty of the c u l t i v a t o r , and the consequent state of i l l - h e a l t h and malnourishment, not only exact unacceptable human costs, but also almost cer ta in ly depress the prospects for a dynamic and progressive agr i cu l tura l sector in the long run." < As the l i t e r a t u r e review in the introductory chapter revealed, a number of authors, in addressing themselves to the problems of rural development in Bangladesh, have been engrossed with the task of establ i sh ing the r e l a t i ve productive e f f i c i ency of small-scale family farming of owner-cultivators over that of tenant cu l t i v a t i on . The present study has emphasized that such an approach may not be adequate, or even appropriate, in determining pol icy options fo r the future. The inadequacy of such exercises becomes apparent as one observes that there are large areas of the country which re ly mainly on owner c u l t i v a t i o n , and yet which produce very l i t t l e per unit of land and labour. 2 Herring, op.cit.3 p. 245. See Chapter IV, p. 193. 261 It has also been shown that , over time, tenant cu l t i v a t i on may create farms so small as to resu l t in cu l t i va t i on as intensive as under owner cu l t i va t i on and, hence, in tenant-farm product iv i t ie s not lagging behind the owner-cultivators with s imi la r resource endowments. However, i t should be recognized that th i s cannot'be used as a premise to suggest that Bangladesh agr iculture might be as well or better served by tenant cu l t i va t i on as by owner cu l t i v a t i on . The antecedent immiserization of the cu l t i vator s involved, and the d i spar i ty of income between the landlords and the tenants have to be recognized and the i r p o l i t i c a l implications taken into account. The assessment of th i s study in comparing owner cu l t i va t i on with sharecropping cu l t i va t i on i s that.neither i s capable of meeting the long-term requirements of agr icu l tura l development in Bangladesh. It i s necessary to do away with the inega l i ta r ian system of share-cropping cu l t i v a t i on ; however, small-scale owner cu l t i v a t i on also cannot be the answer in the long run. .The evaluation of d i s t r i b u t i -v i s t reform p o s s i b i l i t i e s fo r Bangladesh brings one inev i tab ly to th i s conclusion. D i s t r i b u t i v i s t reforms in Bangladesh cannot at ta in the simultaneous goals of creating s e l f - s u f f i c i e n t small-scale family farms which are market oriented as we l l . It has been shown that a ce i l i n g on land even as low as 4 acres, with the subsequent red i s t r ibut ion of ' surp lus ' land among the landless and owners of 1 acre or less of land, would leave the recipients (who const itute more than half of a l l rural 262 households) with an average farm s ize s t i l l less than 1 acre. Although one can argue that such a land c e i l i n g , i f implemented, would cu r t a i l ex i s t ing d i spa r i t i e s in land ownership, i t does not necessari ly fol low that the general welfare of a l l impoverished groups would be enhanced. Erstwhile tenant or wage-labour households may in fact be adversely affected by too th in a red i s t r ibut ion of land. This w i l l happen i f demands for t he i r services from the hitherto r e l a t i v e l y large land owners decline so much that the resu l t ing loss in income exceeds the extra income they can earn from the newly received land. Besides, evidence has been put forward to suggest that the rec ip ients of very small pieces of land are l i k e l y over time to be pressured to s e l l t he i r land in the face of emergencies such as crop f a i l u r e , the sudden i l l ne s s of a family member, and the l i k e . In th i s way, a land red i s t r ibut ion programme in Bangladesh can gradually lose much of i t s effect iveness. The problems which the dynamics of small-scale farming, with the associated laws of inheritence, would create for Bangladesh agr icu l ture in the long run cannot be solved by cooperation in anc i l l a ry , functions. The subsistence nature of small farms would make them ' r i s k - a ve r t e r s 1 . Risk aversion w i l l be manifested in t he i r resistance toward essential changes in technology and production organization. L i t t l e change can be expected i f the basic responsibi l for making investments and bearing r i sks continue to rest with such small farms i n i s o l a t i on . 263 It has been argued in th i s study that, for equity to be achieved in the agr i cu l tura l sector of Bangladesh, access to land enjoyed by d i f fe rent households has to be somehow equalized. There i s a theoret ica l p o s s i b i l i t y that a state farming system, by abolishing private ownership in land and rely ing on wage-labour cu l t i va t i on under state ownership, could remove d i spa r i t i e s evolving out of any pr io r unequal d i s t r i bu t i on of land ownership among d i f fe rent households. However, i t has been argued that the management of state farms would tend to be highly bureaucratic in nature. This would not only l i m i t attainment of equity by the p o s s i b i l i t y of co l lus ion between hitherto large land owners and members of the bureaucracy (who often have kinship and other t i e s with each other), but would also f a i l to meet the condition of dynamic e f f i c i ency . Incentives among members of a state farm would be rather low in the context of Bangladesh. Past experience has made them skeptical about the merits of government action re la t ing to rura l development programmes. It would, therefore, not be appealing to them to work under the 'd i s tant entrepreneur' embodied in the state. The e f f i c i ency of the state farms would also suf fer because bureaucrats avoid taking r i sk s which can earn the displeasure of superiors in the event of the i r not paying of f . Administrators of state farms may further t ry to avoid r i sks by adopting a cap i ta l - i n tens i ve technology instead of one that i s labour-intens ive, since a large labour force can become p o l i t i c a l l y powerful and may cause erosion in t he i r power to control outcomes. 264 Publ ic subsidy for state farms would also assume large proportions, espec ia l l y during the i n i t i a l phases, since the state would have the re spons ib i l i t y for strengthening a l l of these farming operations. In the absence of dynamic e f f i c iency being attained in the state farms, th i s subsidization is l i k e l y to become a perpetual character i s t i c of th i s mode of agrarian operations. Any model suggested for equalizing access to land must also meet other requirements of Bangladesh agr iculture by sat i s factory responses to the fol lowing questions: ( i ) can i t guard against future d i spar i t ie s ? ( i i ) can i t create e f fec t i ve employment opportunities? ( i i i ) can i t incorporate w i t h i n . i t s e l f an e f fec t i ve incentive mechanism? ( iv) can i t sustain i t s e l f in the long run by achieving greater productiv ity? (v) can i t be expected to bear the r i sks involved in undertaking investments which are necessary for dynamic ef f i c iency? It ' has been maintained in the present study that group farming, as a theoret ica l concept, can provide sat i s factory answers to the above questions by vesting ownership of assets, r e spon s i b i l i t i e s for a l locat ing jobs and sharing of p ro f i t s in the hands of the group. Joint decisions in planning and implementation which r e f l e c t majority 265 opinion would e f f ec t i ve l y guard against future d i s pa r i t i e s . Group farms would use avai lable labour.more extensively, as well as more intens ive ly ; this-would be in accordance with the need of employment-generation. By.relat ing remuneration to work-performance i t would maintain indiv idual incentive. More e f fec t i ve u t i l i z a t i o n of labour within th i s incentive mechanism would also promise greater product iv i ty once the t rans i t i on phase has been overcome. Group farms would gradually bu i ld up in capacity to invest in newer technologies l i k e that of the .HYVs-. A l l th i s would further the cause of ag r i cu l tu ra l growth. However, group farming w i l l encounter.many pract ica l problems both in implementation and in subsequent organization and management. Transit ion from t rad i t i ona l family farming to group farming would be a painstaking process. Implementation,of the programme of group farming w i l l have to be ca re fu l l y phased. During the t ran s i t i ona l phase any c o n f l i c t between the interests of group farms and the rest of agr icu l ture would have-to be resolved so as to further the cause of group farming. It i s not easy to reorient family farms toward group consciousness. A great deal of interact ion among the members w i l l be necessaryto achieve th i s end. Member-involvement in managerial functions may also be l imited in the short run due to lack of s k i l l . Hir ing of managers from the outside w i l l have to be engineered and monitored ca re fu l l y so that problems of bureaucratic management do not emerge. 266 It has been argued in th i s study that any e f fec t i ve land reform programme in Bangladesh w i l l have to adopt group farming as i t s long-term goal for achieving equity and el iminating s i z e - d i s a b i l i t i e s and tenur ia l d i s incent ives. A b r i e f review of the p o l i t i c a l factors which can influence the i n i t i a t i o n and implementation of such a land reform programme in Bangladesh shows that the power-el ite.of the country, though vocal about the necessity of group-action in agr icu l ture, have not come out with any spec i f i c proposals for group farming. Presumably th i s i s because they see implementation of such a reform to be contrary to the interests of the i r power-base. However, only pa r t i a l success of various projects based on l imited cooperation, coupled with the increasing p o l i t i c a l consciousness among the lower income groups in agr iculture and the i r heightening expectations, i s gradually creating an atmosphere whereby the necessity for group farming i s being f e l t more and more. It must be given serious consideration by those responsible for devising a development programme which s a t i s f a c t o r i l y meets the c r i t e r i a of economic growth and equity, as well as sat i s fy ing the p o l i t i c a l requirements for long run s t a b i l i t y and survival in Bangladesh. 267 APPENDIX Table A l . Selected Character is t ics of the D i s t r i c t s of Bangladesh D i s t r i c t Dinajpur Rajshahi Kushtia Jessore Pabna Bogra Rangpur Ba r i s a l 1 Khulna Dacca 2 Mymensingh Faridpur Comilla Chittagong H i l l Tracts Syl het Noakhali Chittagong Soi l Composition (%) Sandy Loam 85 50 45 50 70 15 55 75 60 S i l t s 10 10 10 45 10 20 75 65 55 40 40 20 Clayey Loam 90 90 60 50 30 H i l l and H i l l Wash 95 15 55 Saline and Alkal ine 10 10 20 Average Annual Rainfal l (inches) Population Density per Square Mile Red Clayey Swampy Sandy 15 _ _ - 64 659 35 5 - - 52 788 _ _ - - 58 882 _ _ - 61 877 _ 10 - - 58 1,157 40 - - - 58 1,075 10 _ - - 67 1,130 _ _ 25 5 78 1,187 _ _ 30 10 65 600 25 - - - 73 1,909 15 3 - - 75 1,141 5 10 - - 70 1,311 5 - - - 88 1,794 _ 93 75 _ 10 - - 169 737 _ . - 123 1,468 - - - 5 105 1,139 of Barisal and Patuakhali. 2 Includes the present d i s t r i c t s of Mymensingh and Tangail. Includes the present d i s t r i c t s Conti nued ro cr> 00 D i s t r i c t Farm Area as Percentage of Total Area Average Farm Size (acres) Percentage of Farm Area Owned by Farms Sized 12.5 acres or More Percentage of Households with Farms Below 2.5 acres Percentage of Total Cropped Acreage Devoted to Rice Percentage of Total Cropped Acreage Devoted to Jute, Sugar Cane, Wheat and Pulses Percentage of Total Cropped Acreage Devoted to Other Crops Dinajpur 76 50 5.5 22 80 22 81 89 6 58 11 53 Rajshahi 80 86 5.0 22 43 34 70 45 17 66 11 89 Kushtia 73 04 5.8 25 36 24 53 22 32 80 13 98 Jessore 84 78 4.4 13 92 34 67 99 19 31 12 70 Pabna 80 82 3.9 19 61 47 65 03 18 44 16 53 Bogra 80 10 3.8 13 90 15 80 62 9 71 9 67 Rangpur 92 73 4.0 16 78 42 78 76 11 18 10 06 Barisal 79 93 3.6 17 00 52 84 08 6 15 9 77 Khulna 43 88 3.9 16 96 47 86 35 4 46 9 19 Dacca 78 93 2.9 11 30 59 67 54 16 84 15 62 Mymensingh 81 77 3.5 14 07 50 75 60 15 79 8 61 Faridpur 81 34 3.2 12 62 45 64 43 17 84 17 73 Comilla 69 73 1.8 4 00 77 77 42 11 35 11 23 Chittagong H i l l Tracts 5 30 4.1 17 98 55 57 33 34 42 33 Syl het 53 56 4.0 27 44 51 92 26 1 04 6 70 Noakhali 63 66 2.0 10 81 76 87 23 5 28 7 49 Chittagong 41 76 2.6 22 51 69 85 90 6 81 7 29 Continued CTl D i s t r i c t Percentage of Farm Area Sharecropped Dinajpur 24.39 Rajshahi 24.81 Kushtia 21.18 Jessore 18.01 . Pabna 19.18 Bogra 13.18 Rangpur 14.25 Barisal 17.80 Khulna 23.47 Dacca 12.53 Mymensingh 14.75 Faridpur 15.53 Comi11 a 3.84 Chittagong H i l l Tracts 1.45 Sylhet 8.90 Noakhali 13.10 Chittagong 19.03 For Columns 1 - 9 : N. Ahmed, A New Economic Geography of Bangladesh, New Delhi: Vikas Publishing House, 1976, p. 25. For Column 10: ( i ) Government of Pakistan, Population Census of Pakistan, 1961: District Census Report, Kushtia, Karachi: Ministry of Home A f f a i r s , (undated), p. 1-19. ( i i ) Government of Bangladesh, Statistical Digest of Bangladesh, No. 8, Dacca: Bangladesh Bureau of S t a t i s t i c s , 1972, p. 6. For Column 11: Government of Pakistan, Census of Pakistan Population, 1961, Vol . 2, East Pakistan, Karachi, June 1964, p. 11-11. For Column 12: ( i ) Government of Pakistan, Pakistan Census of Agriculture, 1960, Vol .1, Final Report - East Pakistan, Part I, October 1962, pp. 26-28. ( i i ) Government of Pakistan, Census of Pakistan Population, 1961, Vol.2, East Pakistan, Karachi, June 1964, p. 11-11. For Columns 13 - 19: Government of Pakistan, Pakistan Census of Agriculture, 1960, Vo l . I, Final Report - East Pakistan, Part I, October 1962. pp. 26-34, 75-86, and 162-244. 271 BIBLIOGRAPHY Abdullah, A., 'The Class Bases of Nationalism: Pakistan and Bangladesh', in B. Thomas and S. Lavan (eds.), West Bengal and Bangladesh: Perspectives from 1972, East Lansing: Asian Studies Center, Michigan State Univers i ty, 1972. , 'Land Reform and Agrarian Change in Bangladesh', Bangladesh Development Studies, Vol. IV, No. 1, January 1976. , Hossain, M. and Nations, R., 'Agrarian Structure and the IRDP: Preliminary Considerations ' , Bangladesh Development Studies, Vol. IV, No. 2, Apr i l 1976. Adams, D.W. and Rask, N., 'Economics of Cost-Share Leases in Less Developed Countr ies ' , American. Journal of A g r i c u l t u r a l Economics, Vol. 50, No. 4, November 1968. Ahmed, I., 'Reduction in Rural Income Inequality through Land Redist r ibut ion: A Quantitative Est imate ' , Bangladesh Development Studies, Vol. IV, No. 4, October 1976. and Timmons, F., 'Current Land Reforms in East Pak i s tan ' , Land Economics, Vol. XLVII, No. 1, February 1971. Ahmed, N. , A New Economic'Geography of Bangladesh, New Delhi : Vikas Publishing House, 1976. Akhter, F., Characteristics of the Members of The Comilla Cooperatives, Comil la: Pakistan Academy for Rural Development, 1964. Alam, M.M., Characteristics of Newly Elected Representatives (Chairman, Vice-Chairman and Members) of Union Parishads of Kotwali Thana, Comilla, Comil la: Bangladesh Academy for Rural Development, September 1974. Alamgir, M., 'Some Aspects of Bangladesh Agr icu l ture: Review of Performance and Evaluation of P o l i c i e s ' , Bangladesh Development Studies, Vol. I l l , No. 3, Ju ly 1975. Alim, A., Land Reforms in Bangladesh: Social Changes, A g r i c u l t u r a l Development and Eradication of Poverty, Dacca: Samina (publ isher), 1979. 272: A l l an , W., 'Land Tenure and P roduc t i v i t y ' , in J . Hutchinson (ed.), Population, and Food Supply, Cambridge: Cambridge Univers i ty Press, 1969. Anderson, R.S., 'Impressions of Bangladesh: The Rule of Arms and the P o l i t i c s of Exhortat ion ' , Pacific Affairs, Vol. 49, No. 3, Fa l l 1976. Authtah, C , 'The Green Revolution — P a r t i c i p a t i o n by Small Versus Large Farmers' Indian Journal of A g r i c u l t u r a l Economics, Vol. XXVI, No. 1, January-March 1971. Bangladesh Academy for Rural Development, Twelfth Annual Report: July 1970 - June 1971, Comil la, May 1973. Bangladesh Agr icu l tura l Research Counci l, Proceedings of the Workshop on Appropriate Technology, February 6-8, 1975, Dacca, 1975. , Incidence of-Landlessness and Major Landholding and: Cultivating Groups in Rural Bangladesh, Dacca: Directorate of Agr iculture (Extension and Management), May 1978. Bardhan, P.K. and Sr inivasan, T.N., 'Crop-Sharing Tenancy in Agr icu l ture: A Theoretical and Empirical Ana l y s i s ' , American Economic.Review, Vol. LXI, No. 1, March 1971. Barua, T.K., P o l i t i c a l Elite in Bangladesh: A Socio-Anthropological and H i s t o r i c a l Analysis of the Processes of their Formation, Bern: Peter Lang Publishers, 1978. B e l l , C., ' A l ternat ive Theories of Sharecropping: Some Tests Using Evidence from Northeast I nd i a ' , Journal of Development Studies, Vol. 13, No. 4, July 1977. : —and Zusman, P., 'A Bargaining Theoretic Approach to Cropsharing Contracts ' , American Economic Review, Vol. 66, No. 4, September 1976. B e l l , F.O., Final Report on the Survey .and Settlement Operations in the District of Dinajpur, 1934-40, A l ipore: Bengal Government Press, 1942. Berry, R.A., 'Land Reform and the Agr icu l tura l Income D i s t r i b u t i o n 1 , Pakistan Development Review, Vol. XI, No. 1, Spring 1971. 273 Bertocc i , P . J . , Elusive Villages: Social Structure and Community Organization in Rural East Pakistan, Unpublished Ph.D. Dissertat ion, Michigan State Univers i ty, 1970. Bhaduri, A., 'A Study in Agr icu l tura l Backwardness Under Semi-Feudal ism 1, Economic Journal, Vol. 83, No. 329, March 1973. Bhaduri, P.R., 'Progress of Land Tenure System in Unpartitioned Bengal Since the Days of Permanent Settlement (1793-1947)', Indian Journal of Economics, Vol. XXXV, Part I I I, January 1955. Bhagwati, J.N. and Chakravarty, S., 'Contributions to Indian Economic Analys is: A Survey', American Economic Review, Vol. LIX, No. 4, .Part IX, September,,1969. B l a i r , H.W.The Elusiveness of Equity: Institutional Approaches to Rural Development in Bangladesh, Ithaca: Rural Development Committee, Cornell Univers i ty, 1974. , 'Rural Development, Class Structure and Bureaucracy in Bangladesh', World Development, Vol. 6, No. 1, January 1978. Boesen, J . , Madsen, B.S. and Moody., T., :Ujamaa ^--Sociali'sm.From Above, Uppsala: Africana Publishing Company, 1977. Bose, S., 'The Strategy of Agr icu l tura l Development in Bangladesh', in E.A.G. Robinson and K. G r i f f i n (eds.), The Economic Development of Bangladesh Within a Socialist Framework, New York: Hoisted Press, 1974. Byres, T . J . , 'Land Reform, Indus t r ia l i za t ion and the Marketed Surplus in India: An Essay on the Power of Rural B i a s 1 , in D. Lehmann (ed.), Agrarian Reform and Agrarian Reformism: Studies of Peru, Chile, China and India, London: Faber and Faber, 1974. Center for Population Studies, Bangladesh Land, Water and Power Study, Cambridge (Mass.): Harvard Univers i ty, 1974. Chaudhuri, D.P., 'New Technologies and Income Dis t r ibut ion in Ag r i cu l t u re ' , in D. Lehmann (ed.), Agrarian Reform and Agrarian Reformism: Studies of Peru, Chile, China and India, London: Faber and Faber, 1974. Chayanov, A.V., The Theory of Peasant Economy, I l l i n o i s : The American Economic Associat ion, 1966. 274 Chenery, H. et a l . 3 Redistribution With Growth, London: Oxford Univers ity Press, 1974. Cheung, S.N.S., The Theory of Share Tenancy, Chicago: The Univers ity of Chicago Press, 1969. Chowdhury, A., ' Social S t r a t i f i c a t i o n in a Bangladesh V i l l a g e ' , Journal of the A s i a t i c Society of Bangladesh, Vol. XX, No. 3, December 1975. Courtenay, P.P., Plantation Agriculture, London: C. Bel l and Sons L td . , 1965. Dal ton, G., 'Is Economic Anthropology of Interest to Economists?', American Economic Review, Vo l . 68, No. 2, May 1978. Das, B.B., 'Lessons of Land Movement', Indian Left Review, Vol. I, No. 1, February 1971. Datta, B.N., Dialectics of Land-Economics of India, Ca lcutta: Mohendra Publishing Committee, 1952. Dorner, P., 'Land Tenure, Income D i s t r ibut ion and Product iv i ty In teract ions ' , Land Economics, Vol. XL, No. 3, August 1964. , Land Reform and Economic Development, Middlesex: Penguin Books, 1972. Dovring, F., 'Land Reform: A Key to Change in Ag r i cu l t u re ' , in N. Islam (ed.), A g r i c u l t u r a l Policy in Developing Countries, London: Macmillan, 1974. Dutt, K., 'A Cr i t ique of Land Red i s t r i bu t i on ' , Mainstream, Vo l . IX, No. 5, October 3, 1970. Etienne, G., 'China 's Agr icu l ture: Present S ituat ion and Prospects ' , in P. Dorner (ed.), Cooperative and Commune: Group Farming in the Economic Development of Agriculture, Madison: The Univers ity of Wisconsin Press, 1977. F ink le r , K., 'From Sharecroppers to Entrepreneurs: Peasant Household Production Strategies Under the Ejido System of Mexico ' , Economic Development and Cultural Change, Vo l . 27, No. 1, October 1978. 275 Food and Agr iculture Organization, Report of the Agricultural Team for Bangladesh, Rome, 1973. , FAO Production Year Book, 1978, Rome, 1979. Frank, A.G., 'The Var iet ies of Land Reform', Monthly Review, Vol. 14, No. 11, March 1963. Frankel, F.R., India's Green Revolution: Economic Gains and P o l i t i c a l Costs, Princeton: Princeton Univers ity Press, 1972. Galesk i , B., 'The Models of Co l lec t i ve Farming', in P. Dorner (ed.), Cooperative and Commune: Group Farming in the Economic Development of Agriculture, Madison: The Univers ity of Wisconsin Press, 1977. Georgescu-Roegen, N., 'Economic Theory and Agrarian Economies', Oxford Economic Papers, New Series, Vo l . -XII, No. 1, February 1960. Gersovitz, M., 'Land Reform: Some Theoretical Considerat ions ' , Journal of Development Studies, Vol. 13, No. 1, October 1976. Government of Bangladesh, Master Survey of Agriculture in Bangladesh, Seventh Round, Second Phase (1967-68), Dacca: Bangladesh Bureau of S t a t i s t i c s , (undated). S t a t i s t i c a l Digest of Bangladesh, No. 8, Dacca: Bangladesh Bureau of S t a t i s t i c s . — ! , Bangladesh Agriculture in Statistics, Dacca: Min ist ry of Agr icu l ture, November 1973. , Farm Management Research in Retrospect, Dacca: Min ist ry of Agr icu l ture, 1973. , The F i r s t Five Year Plan, 1973-78, Dacca: Planning Commission, 1973. , Bangladesh Census of Population - 1974, Dacca: Census Commission, Min ist ry of Home A f f a i r s , 1975. , Summary Report of the 1977 Land Occupancy Survey of Rural Bangladesh, Dacca: Bangladesh Bureau of S t a t i s t i c s , 1977. , 'Text of Z i a ' s Speech at FAO Conference on Agrarian Reform and Rural Development', Bangladesh, Vo l . I l l , No. 2, August 1979. 276 , 'Population Changes and Economic Development', Bangladesh, Vol. I l l , No. 5, September 1979. Government of Bengal, Report of the Land Revenue Commission, Bengal, Vol. I, A l ipore: Bengal Government Press, 1940. Government of East Pakistan, Master Survey of Agriculture in East Pakistan, S ixth Round (1964-65), Dacca: East Pakistan Bureau of S t a t i s t i c s , 1966. , A g r i c u l t u r a l Credit in East Pakistan, Dacca: The Registrar of Cooperative Soc iet ies , 1967. , Report of the Land Revenue Commission, East Pakistan, Dacca: East Pakistan Government Press, 1969. Government of Pakistan, Economy of Pakistan, 1950, Karachi: Off ice of the Economic Advisor, Min istry of Economic A f f a i r s , 1951. , Family Expenditure Schedules, Karachi: Central S t a t i s t i c a l Of f i ce , 1960. , Population Census of Pakistan, 1961: District Census Report, Bogra, Karachi: Min ist ry of Home A f f a i r s , (undated). , Population Census of Pakistan, 1961: District Census Report, Kushtia, Karachi: Min istry of Home A f f a i r s , (undated). , Population Census of Pakistan, 1961: District Census Report, Rajshahi, Karachi: Min istry of Home A f f a i r s , (undated). , Pakistan Census of Agriculture, 1960, Vol. I, Final Report -East Pakistan, Part I, October 1962. , Census of Pakistan Population, 1961, Vol. 2, East Pakistan, Karachi, June 1964. G r i f f i n , K., The P o l i t i c a l Economy of Agrarian Change: An Essay on the Green Revolution, London: The Macmillan Press, 1974. Guha, R., A Rule of Property for Bengal: An Essay on the Idea of Permanent Settlement, Par i s : Monton & Co., 1963. Guttman, L. and Lingoes, J .C . , The Guttman-Lingoes Nonmetric Program Series, km Arbor: Mathesis Press, 1973. 277 Hainsworth, G.B., 'Economic Growth and Poverty in Southeast As ia: Malaysia,. Indonesia and the Ph i l i pp i ne s ' , Pacific Affairs, Vol. 52, No. 1, Spring 1979. Hamid, M.A., 'Ganamilan Cooperative Farming in Bangladesh 1,in J . Wong (ed.), Group Farming in Asia: Experiences and Potentials, Singapore: Singapore Univers ity Press, 1979. Haq, M . N . ,.Village Development in Bangladesh (A Study of Monogram Village), Comil la: Bangladesh Academy for Rural Development, 1973. Haque, A., 1 Bangladesh 1979: Cry for a Sovereign Parl iament 1 , Asian Survey, Vol. XX, No. 2, February 1980. Heady, E.O., Economics of Agricultural Production and Resource Use, Englewood C l i f f s : Prentice-Hal l Inc., 1952. Herring, R.J., 'Share Tenancy and Economic E f f i c iency: the South Asian Case', Peasant Studies, Vol. 7, No. 4, Fa l l 1978. Hirsch, G.P., 'Some Fundamentals of Land Reform', Oxford Agrarian Studies, Vol. I, No. 2 (New Ser ies) , 1972. Hossain, M., 'Farm Size and Product iv i ty in Bangladesh Agr icu l ture: A Case Study of Phulpur Farms', Bangladesh Economic Review, Vol. I I, No. 1, January 1974. , 'Farm S ize, Tenancy and Land Product iv i ty: An.Analysis of Farm Level Data in Bangladesh Ag r i c u l t u r e ' , Bangladesh Development Studies, Vol. V, No. 3, July 1977. , 'Factors Affect ing Tenancy: the Case of Bangladesh Ag r i cu l t u re 1 , Bangladesh Development Studies, Vol. VI, No. 2, Summer 1978. Hossain, M.Z., A Field Investigation into the Management of Village Cooperatives in Comilla Experimental Area,. Comilla: Pakistan Academy for Rural Development, 1966. Huq, M., Final Report on the Revisional Survey and Settlement Operations '• in' the. District of Bakerganj, 1940-42 and 1945-54, Dacca: Government of East Pakistan, East Pakistan Government Press, 1957. Husain, A.M.M., 'The Shimla Cooperative Farming Project in Bangladesh 1, in J . Wong {ed.), Group Farming in Asia: Experiences and Potentials, Singapore: Singapore Univers ity Press, 1979. 278 Hussain, S.M., 'A Note on Farmer Response to Price in East Pak i s tan 1 , Pakistan Development Review, Vol. IV., No. 1, Spring 1964. Ilchman, W.F. and Uphoff, N.T., The P o l i t i c a l Economy of Change, Berkeley: Univers ity of Ca l i f o rn ia Press, 1969. Integrated Rural Development Programme, Proposals for the First Five Year Plan, Part I I I , Dacca, January 1973. Ishaque, H.S.M., A g r i c u l t u r a l Statistics by Plot to Plot Enumeration in Bengal, 1944-45, A l ipore: Government of Bangal, Bengal Government Press, 1946. Islam, A.K.M.A., A Bangladesh Village Conflict and Cohesion: An Anthropological Study of Politics, Cambridge (Mass.): Shenkman Publishing Company, 1974. Islam, M.M., Bengal Agriculture, 1920-1946: A Quantitative Study, Cambridge: Cambridge Univers ity Press, 1978. Islam, N., Development Planning in Bangladesh: A Study in P o l i t i c a l Economy, London: C. Hurst and Company, 1977. 'Development.Strategy of Bangladesh, Oxford: Pergamon Press, 1978. Jabbar, M.A., An Investigation into the Effect of Farm Structure on Land Productivity in Selected Areas of Bangladesh, Unpublished Ph.D. D i ssertat ion, Univers ity of Wales, 1976. Jack, J .C. , The Economic .Life of a Bengal District, Oxford: The Clarendon Press, 1916. Jacoby, E.H., Man and Land:- The Essential Revolution, New York: Alf red A. Knopff Inc., 1971. Jahan, R., 'East Pakistan During the Decate of Ayub', in P.J. Bertocci (ed.), Prelude to Crisis: Bengal and Bengal Studies in 1970, East Lansing: Asian Studies Center, Michigan State Univers i ty, August 1972. , Pakistan: F a i l u r e in National Integration, London: Oxford Univers ity Press, 1973. 279 , 'Members of Parliament in Bangladesh 1, Legislative Studies Quarterly, Summer 1976. Jahangir, B.K., 'Local Action for Se l f -Re l iant Development in Bangladesh', in International Foundation for Development A l ternat ives , IFDA Dossier 15, January/February 1980. Jannuzi, F.T. and Peach, J .T. , 'Note on Land Reform in Bangladesh: The Eff icacy of C e i l i n g s ' , Journal of Peasant Studies, Vol. 6, No. 3, Ap r i l 1979. Johnson, B.L.C., Bangladesh, New York: Harper and Row Publishers, 1975. Johnston, B.F. and Mel lor, J.W., 'The Role of Agr iculture in Economic Development', American Economic Review, Vo l . L I , No. 4, September 1961. Johnston, J . , Econometric Methods, New York: McGraw H i l l Book Co. Inc., 1972. Josh i , P.C., 'Marx and the Agrarian Problem', in P.C. Joshi (ed.), Homage to Marx, Delh i : People's Publishing House, 1970. •—, 'Land Reform and Agrarian Change in India and Pakistan Since 1947', in R. Datta and P.C. Joshi (eds.), Studies in Asian Social.Development, No. 1, Bombay: Tata McGraw-Hill Publishing Company L td . , 1971. — : — , 'Land Reform: An Urgent Problem in Bangladesh', in V.K.R.V. Rao (ed.), Bangladesh Economy: Problems and Prospects, Delhi : Vikas Publ icat ions, 1972. Kahnert, F. et al., Agriculture and Related Industries in Pakistan, Par i s : Organization for Economic Cooperation and Development, 1970. Kamaluddin, S.., 'Bangladesh: Revolution and Confusion'-, Far'..Eastern Economic. Review, Vol. 106, No. 44, November 2, 1979. Karim, M.R., 'Family Case Study of Md. Joinal Abedin ' , Family Case Study No. 7, in Government of Bangladesh, Proceedings of the Bangladesh - FAO Workshop on the Problems of Small and Subsistence Farmers and-Agricultural Labourers, Dacca: Bangladesh Ag r i -cu l tura l Research Counci l, 1974. 280 Khan, A.A., Rural Credit Programme of A g r i c u l t u r a l Cooperative Federation, Comil la: Bangladesh Academy for Rural Development, 1971. Khan, A.H., 'The Comilla Project A Personal Account', International Development Review, Vo l . XVI, No. 3, 1974. Khan, A.R., The Economy of Bangladesh, London: The Macmillan Press, 1972. , 'The Comilla Model and the Rural Development Programme of Bangladesh: An Experiment in Cooperative Capital i sm ' , World Development, Vol. 7, Nos. 4/5, April/May, 1979. and Ghai, D., ' Co l l e c t i ve Agr icu l ture in Soviet Central A s i a ' , World Development, Vol. 7, Nos. 4/5, April/May 1979. Khan, M.I., 'The Role of IR-20 in Solving the Food Problem of East Pak i s tan ' , Pakistan Development Review, Spring 1971. Khusro, A.M., 'Farm Size and Land Tenure in I nd ia ' , Indian Economic Review, Vol. IV (New. Ser ies ) , 1969. , The Economics of Land Reform and Farm Size in India, Delhi: The Macmillan Company of India, 1973. King, R., Land Reform: A World Survey, London: G. Bel l and Sons Ltd. , 1977. K l a t t , W., 'Agrarian Issues in Asia ( I. Land as a Source of C o n f l i c t ) ' , International A f f a i r s , Vol. 48, No. 2, Ap r i l 1972. , 'Agrarian Issues in Asia ( I I. Reform and Insurgency) 1, International Affairs, Vo l . 48, No. 2, July-1972. Koo, A.Y.C., 'An Economic J u s t i f i c a t i o n for Land Reformism', Economic Development and Cultural Change, Vo l . 25, No. 3, Apr i l 1977. Krishna, R., 'Agrarian Reform in India: The Debate on C e i l i n g s ' , Economic Development and Cultural Change, Vol. VII, No. 3, Apr i l 1959. , 'Land Reform and Development in South A s i a ' , in W. Froehlich (ed.), Land Tenure, I n d u s t r i a l i z a t i o n and Social Stability, Milwaukee: The Marquette Univers ity Press, 1961. 281 Landsberger, H.A. (ed.), Rural- Protest: Peasant Movement and Social. Change, London: The Macmillan Press, 1974. Le, C. andTen i s c i , T., Triangular Regression Package, Vancouver: Computing Centre, The Univers ity of B r i t i s h Columbia, October 1978. L ipton, M., ' I n te rd i s c ip l i na ry Studies in Less Developed Countr ies ' , Journal of Development Studies, Vol. 7, No. 1, October 1970. , 'Towards a Theory of Land Reform', in D. Lehmann (ed.), Agrarian Reform and Agrarian Reformism: Studies of Peru, Chile, China and India, London: Faber and Faber, 1974. Long, E.J., 'The Economic Basis of Land Reform in Underdeveloped Economies', Land Economics, Vol. XXXVII, No. 2, May 1961. Luykx, N., 'A,Publ ic Pol icy P r o f i l e on Rural Development in Bengal ' , in R.V.M. Baumer (ed.), Aspects of Bengali History and Society, Honolulu: The Univers ity Press of Hawaii, 1975. Malek, Q.M.A., 'Rice Cu l t ivat ion in Comilla Kotwali Thana: The Role of Cooperatives ' , Bangladesh Development Studies, Vo l . IV, No. 3, July 1976. Manalo, E.B., Agro-Climatic Survey of Bangladesh, Bangladesh Rice Research In s t i tu te , (undated). Mandel, E., Marxist Economic Theory, London: Merl in Press, 1974. Maniruzzaman, T., 'Bangladesh: An Unfinished Revolution? ' , Journal of Asian Studies, Vo l . XXXIV, No. 4, August 1975. , 'Bangladesh in 1975: The Fa l l of the Mujib Regime and Its Aftermath ' , Asian Survey, Vol. XVI, No. 2, February 1976. Mannan, M.A., Rural Leadership and Its Emerging Pattern in Bangladesh (With Special Reference to Comilla Kotwali Thana), Comilla: Bangladesh Academy for Rural Development, 1972. Marshal l , A., Principles of Economics, New York: The Macmillan Company, 1948. 282 Migdal , J.S., Peasants, Politics and Revolution: Pressures toward P o l i t i c a l and Social Change in the Third World, Princeton: Princeton Univers ity Press, 1974. Minnas, B.S., 'Rural Poverty, Land Redistr ibution and Development Strategy: Facts and P o l i c y ' , Indian Economic Review, Vol. V (New Ser ies) , No. 1, Ap r i l 1970. Moore, B. , Social Origins of Dictatorship and Democracy: Land and Peasant in, the Making of the Modern. World, Boston: Beacon Press, 1966. Moore, M.P., 'The Logic of In terd i sc ip l inary Stud ies ' , Journal of Development Studies, Vol. 11, No. 1, October 1974. Morrison, B.M., 'The Persistent Rural C r i s i s in As ia: A Sh i f t in Conception', Pacific Affairs, Vol. 52, No. 4, Winter 1979-80. Mukerjee, R. , Land Problems of India, London: Longmans, Green & Co., 1933. Mukherjee, R., The Dynamics of a Rural Society A Study of the Economic Structure in a Bengal Village, Be r l i n : Akademie-Verlag, 1957. Mukherji, I.N., 'Agrarian Reforms in Bangladesh', Asian Survey, Vol. XVI, No. 5, May 1976. Myrdal, G., Asian Drama: An Enquiry into the Poverty of Nations, Vols. I, II and II I, London: The Penguin Press, 1968. , 'Land Reform in i t s Broader Economic and Social S e t t i n g ' , in United Nations, Food and Agr iculture Organization of the United Nations and International Labour Organization, World Land Reform Conference, (Rome, 20 June - 2 July), New York: United Nations, 1968. Newberry, D.M.G., 'The Choice of Rental Contract in Peasant Ag r i cu l t u re ' , in L.G. Reynolds (ed.), Agriculture in Development Theory, New Haven: Yale Univers ity Press, 1975. Parthasarathy, G., 'The Fa i lu re of Tenancy L e g i s l a t i o n ' , in Indian Society of Agr icu l tura l Economics, Seminar on Problems of Small Farms, Seminar Series - VII, Bombay, 1968. 283 , 'Land Hunger and Rural Power S t ructure ' , Seminar (Symposium on Land Hunger), No. 140, Ap r i l 1971. Ping, H.K., 'Revolt of the Landless Peasants' Far Eastern Economic Review, Vol. 103, No. 2, January 12, 1979. Rahim, S.A., Rural Cooperatives and Economic Development of Subsistence Agriculture, Comilla: Pakistan Academy for Rural Development, 1972. Rahman, A.T.R., 'Administration and i t s P o l i t i c a l Environment in Bangladesh', Pacific Affairs, Vol. 47, No. 2, Summer 1974. Rahman, M., A New Cooperative System for. Comilla Thana, Rural Cooperation P i l o t Experiment, Tenth Annual Report (1969-70), Comilla: Bangladesh Academy for Rural Development, January 1972. Rahman, M.A., 'Land Reform and Agrarian Revolution in Bangladesh', (or ig ina l in Bengal i), Paper presented at the Bangladesh 'Economic Association Conference on the F i r s t Five Year Plan (March 17-20, 1974), Dacca, 1974. , ' P r i o r i t i e s and Methods for Soc i a l i s t .Development .of•Bangladesh 1, in E.A.G. Robinson and K. G r i f f i n (eds.), The Economic Development of Bangladesh ..Within a S o c i a l i s t Framework, New York: Hal sted Press, 1974. , 'Farm S ize, E f f ic iency and the Socio-Economics of Land D i s t r i b u t i o n ' , Bangladesh Development Studies, Vol. I l l , No. 3, July 1975. Rahman, M.M., 'Land D i s t r ibut ion: i t s impact on rural development', Bangladesh Times, March 3, 1978. Raj, K.N., ' Ag r i cu l tu ra l Development and D i s t r ibut ion of Land Holdings' Indian Journal of A g r i c u l t u r a l Economics, Vol. XXX, No. 1, January-March, 1975. Rajman, A.N., ' Ag r i cu l tura l Holdings: Consolidation and C e i l i n g ' , Asian Economic Review, Vol. IV, No. 2, February 1962. Rashid, H.E., Geography of Bangladesh, Boulder: Westview Press, 1977. 284 Raychaudhuri, T., 'Permanent Settlement in Operation: Bakarganj D i s t r i c t , East Bengal ' , in R.E. Frykenberg (ed.), Land Control and Social Structure in Indian History, Madison: The Univers ity of Wisconsin Press, 1969. Reed, E.P., ' Introducing Group Farming in Less Developed Countries: Some Issues ', in P. Dorner (ed.), Cooperative and Commune: Group Farming in the Economic Development of Agriculture, Madison: The Univers ity of Wisconsin Press, 1977. Robinson, A., Economic Prospects of Bangladesh, London: Overseas Development I n s t i tu te , 1973. Robinson, J . , 'Chinese Agr icu l tura l Communes', in C.K. Wilber (ed.), The P o l i t i c a l Economy of Development and Underdevelopment, New York: Random House, 1973. Roy, A., 'Green Revolution An Assessment', The Marxist Review, Vol. VI, No. 6, December 1972. Rutton, V.W., 'The Green Revolution: Second Generation Problems', American Journal of A g r i c u l t u r a l Economics, Vol. LI I, No. 5, December 1.970. Sanghvi, P., Surplus Manpower, in Agriculture and Economic Development, Bombay: Asia Publishing House, 1969. Sarkar, N., ' Con f l i c t of Social and Individual Objectives of Peasant Farming in Asia and A l ternat ive Strategies of Agr icu l tura l Development1, Economic and P o l i t i c a l Weekly, Vol. 9, No. 5, February 2, 1974. Sattar, M.G., 'Family Case Study of an Owner-cum-Tenant Farmer', in Government of Bangladesh, Proceedings of .the Bangladesh - FAO Workshop on the Problems of Small and Subsistence Farmers and Agricultural Labourers, Dacca: Bangladesh Agr icu l tura l Research Counci l , 1974. Sayeed, K.B., The Political System of Pakistan, Boston: Houghton M i f f l i n Company, 1967. S c h i l l e r , O.M., Cooperation, and Integration in A g r i c u l t u r a l Production, Concepts and P r a c t i c a l Application: An International Synopsis, Bombay: Asia Publishing House, 1969. 285 Sen, A.K., 'The Choice of Agr icu l tura l Techniques in Underdeveloped Countr ies ' , Economic Development and Cultural Change, Vol. VI I, No. 3, Apr i l 1959. , 'An Aspect of Indian Ag r i c u l t u r e ' , Economic Weekly, Annual Number, February 1962. , ' S ize of Holdings and P roduc t i v i t y ' , Economic Weekly, Annual Number, February 1964. and Varghese, T.C., 'Tenancy and Resource A l l o c a t i o n ' , Seminar (a symposium on Land Relations and Agr icu l tura l Growth), No. 81, May 1966. S idd iqu i , K., 'Comment on Jannuzi and Peach: 'A Note on Land Reform' ' , Journal of Peasant Studies, Vol. 6, No. 3, Ap r i l 1979. Singh', R.P., Decision Making and Rural E l i t e s : A Study of Rural-Urban E l i t e s . I n t e r a c t i o n in Decision Making Process, Delhi : Research Publ icat ions, 1975. Sinha, S.N., 'Marketable Surplus in Agr iculture in Underdeveloped Countr ies ' , Economic Review, Vol. XI I I, No. 22, Apr i l 7, 1962. Sivaswami, K.G., 'Sub-tenancy in Zamindari Areas ' , Indian-Journal of Economies, Vol. XXVIII, 1947-48. Smith, D.V., 'New Seeds and Income Di s t r ibut ion in Bangladesh 1, Journal, of Development Studies, (Special issue on Employment, Income Dis t r ibut ion and Development), Journal of Development Studies, Vo l . 11, No. 2, January 1975. Sobhan, R., Basic Democracies, Works Programme and Rural Development in East Pakistan, Dacca: Bureau of Economic Research, Univers ity of Dacca, 1968. Solaiman, M., Land Holding and Cooperatives in Five Comilla Villages, Comilla: Bangladesh Academy for Rural Development, May 1974. Stavis , B., People's Communes and Rural Development in China, Special Series on Rural Local Government, Ithaca: Rural Development . Committee, Cornell Univers i ty, November 1974. 286 , Making Green Revolution: .The Polities of A g r i c u l t u r a l Development in China, Rural Development Monograph No. 1, Ithaca: Rural Development Committee, Cornell Un ivers i ty , 1974. S t i g l i t z , J .E . , ' ' Incentives and Risk Sharing in Sharecropping 1, Review of Economic Studies, Vol. 41, No. 2, Ap r i l 1974. Stuart, R.C., The Collective Farm in. Soviet Agriculture, London: Lexington Books, 1972. Sultan, K.M.T., Problems of Rural Administation in Bangladesh, Comilla: Bangladesh Academy for Rural Development, October.1974. Tabors, R. , The D e f i n i t i o n of Multifunctional Planning Regions: A Case Study of East Pakistan, Cambridge (Mass.): Center for Population Studies, Harvard Univers i ty, May 1971. Ta i , H-C., Land Reform and Politics, Berkeley: Univers ity of Ca l i f o rn ia Press, 1974. Tepper, E.L., 'The Administration of Rural Reform: Structural Constraints and P o l i t i c a l Dilemmas', in R.D. Stevens, H. Alavi and P.J. Bertocci (eds.),. Rural Development in Bangladesh and Pakistan, Honolulu: The Univers ity Press of Hawaii, 1976. Thiesenhusen, W.C., 'What Changing Technology Implies for Agrarian Reform', Land Economics, Vol. L, No. 1, February 1974. Thomas, J.W., The Rural Public Works Program and East Pakistan's Development, Unpublished Ph.D. D i ssertat ion, Harvard Univers i ty, November 1968. Thorner, D. and Thorner, A., Land and.Labour in India, Bombay: Asia Publishing House, 1962. Tuma, E.H., Twenty-Six Centuries of Agrarian Reform: A Comparative Analysis, Berkeley: Univers ity of Ca l i f o rn ia Press, 1965. — , 'Agrarian Reform in H i s to r i ca l Perspective Rev i s i t ed ' , Comparative Studies in Society and History, Vol. 21, No. 1, January 1979. Tyers, R. , Optimal Resource A l l o c a t i o n in Transitional A g r i c u l t u r e : Case Studies in Bangladesh, Unpublished Ph.D. D i s sertat ion, Harvard Univers i ty , January 1978. 287 Umar, B., Politics and Society in East Pakistan and Bangladesh, Dacca: Mowla Brothers, 1973. United Nations, Progress in Land Reform, New York, 1954. , Progress in Land Reform, S ixth Report, New York, 1976. Warriner, D., Land Reform in P r i n c i p l e and Practice, Oxford: Clarendon Press, 1969. Wong, J . , 1Communization of Peasant Agr icu l ture: China's Organizational Strategy for Agr icu l tura l Development', in P. Dorner (ed.), Cooperative and Commune: Group Farming in the Economic Development of Agriculture, Madison: The Univers ity of Wisconsin Press, 1977. Wood, G., 'From Riyat to Rich Peasant', South Asian Review, Vol. 7, No. 1, October 1973. Wood, J.R., 'Land and P o l i t i c s : A Comparison of Agrarian Reform in Japan and I nd ia ' , in A.W. Cordier (ed.), Columbia Essays in International Affairs, Vol. I l l , New York: Columbia Univers ity Press, 1968. World Bank, Land Reform:. Sector Policy Paper, Washington, D.C, May 1975. Za i d i , S.M.H., The Village Culture in Transition: A Study of East Pakistan Rural Society, Honolulu: East-West Center Press, 1970. Zaman, M.A., 'Bangladesh: The Case for Further Land Reform', South Asian Review, Vol. 8, No. 2, January 1975. Zaman, M.R., 'Sharecropping and Economic E f f i c iency in Bangladesh', Bangladesh Economic Review, Vol. I, No. 2, Ap r i l 1973. 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0094957/manifest

Comment

Related Items