IMPACT OF FARM MECHANIZATION ON RICE PRODUCTIVITY IN CAUVERY DELTA ZONE OF TAMIL NADU STATE AN ECONOMIC ANALYSIS

Transcription

1 IMPACT OF FARM MECHANIZATION ON RICE PRODUCTIVITY IN CAUVERY DELTA ZONE OF TAMIL NADU STATE AN ECONOMIC ANALYSIS Thesis submitted in part fulfillment of the requirement for the degree of DOCTOR OF PHILOSOPHY IN AGRICULTURAL ECONOMICS to the Tamil Nadu Agricultural University, Coimbatore. By M.CHIDAMBARAM (I.D.No ) DEPARTMENT OF AGRICULTURAL ECONOMICS TAMIL NADU AGRICULTURAL UNIVERSITY COIMBATORE

2 CERTIFICATE This is to certify that the thesis entitled IMPACT OF FARM MECHANIZATION ON RICE PRODUCTIVITY IN CAUVERY DELTA ZONE OF TAMIL NADU STATE AN ECONOMIC ANALYSIS submitted in part fulfillment of the requirement for the degree of DOCTOR OF PHILOSOPHY IN AGRICULTURAL ECONOMICS, to the Tamil Nadu Agricultural University, Coimbatore is a record of bonafide research work carried out by Mr.M.CHIDAMBARAM, under my supervision and guidance and that no part of this thesis has been submitted for the award of any other degree, diploma, fellowship or other similar titles and that the work has not been published in part or full in any scientific or popular journal or magazine. Place: Coimbatore Date : Dr.N.AJJAN (Chairman) Approved By Chairman : (Dr.N.AJJAN) Members : (Dr.C.SEKAR) (Dr. R.PALANISAMY) (Dr.D.ANANTHA KRISHNAN) EXTERNAL EXAMINER

3 ACKNOWLEDGEMENT I acknowledge with immense gratitude, the opportunity given by Dr.P.Murugesa Boopathi, Ph.D., former Vice-Chancellor, Tamil Nadu Agricultural University for doing the Doctoral Programme at the fag end of my career as Assistant professor. I place on record my heartfelt thanks and deep sense of gratitude to my chairman of Advisory Committee, Dr.N.Ajjan Ph.D., former Director, Centre for Agricultural and Rural Development Studies (CARDS), Professor, Department of Agricultural and Rural Management, Tamil Nadu Agricultural University, Coimbatore for providing me the best, and able guidance, concrete suggestions, constructive criticisms and learned counseling. I am grateful to Dr.C.Sekar Ph.D., Professor, Post Harvest Technology Centre, Dr.R.Palanisamy Ph.D., Professor (Mathematics) and Dr.D.Anantha Krishnan Ph.D., Professor, Department of Farm Machinery, Members of Advisory Committee, Tamil Nadu Agricultural University for their learned counsel, constructive criticisms and constant encouragement. I am extremely thankful to Dr.M.Chinnadurai Ph.D., Director, Centre for Agricultural and Rural Development Studies (CARDS) for his special interest, efforts and encouragement on my doctoral programme. I acknowledge with gratitude the support and encouragement given by Dr.R.Balasubramanian Ph.D., Professor and Head, Department of Agricultural Economics, Tamil Nadu Agricultural University. I owe immensely to Dr.M.R.Duraisamy Ph.D., Professor and Head, Department of Physical Sciences and Information Technology, Mr.M.Suresh, Assistant Professor and Er.A.P.Mohan Kumar, Tamil Nadu Agricultural University for their invaluable help extended for data analysis. I place on record the tireless help rendered by Mr.G.Sivaraj, Network specialist, in completing my thesis work throughout. I am thankful to my friends Mr.Vijayasarathy and Mr.Saikumar for their moral support extended during the period. M.CHIDAMBARAM

4 ABSTRACT IMPACT OF FARM MECHANIZATION ON RICE PRODUCTIVITY IN CAUVERY DELTA ZONE OF TAMIL NADU STATE AN ECONOMIC ANALYSIS By M.CHIDAMBARAM Degree Chairman : DOCTOR OF PHILOSOPHY IN AGRICULTURAL ECONOMICS : Dr.N.AJJAN Ph.D., Professor, Department of Agricultural and Rural Management, Centre for Agricultural and Rural Development Studies, (CARDS), Tamil Nadu Agricultural University, Coimbatore Rice is one of world s most favoured staple foods and more than 90 per cent of rice is produced and consumed in Asia. Rice being an important crop in India, there is a lot to focus on enhancing rice production and productivity. Rice is grown in 43.4 Million hectares in Kharif and rabi / summer season out of the total 142 Million hectares of land under cultivation. The area under rice is likely to reduce in future years due to diversification polices adopted by the government. Currently, the rice production in the country is passing through serious constraints like plateauing of yield, water scarcity and labour scarcity and inadequate institutional dynamics. It is estimated that demand for rice will be Million tons by the year In order to achieve this target, the productivity of rice has to be brought to the level of 2.7 tons per ha, which is 2.2 tons presently. The present rate of production growth (1.27 per cent) is below the population growth of 1.63 per cent. Therefore the present deceleration trend in production and yield is a cause of concern and has to be reversed. Farm mechanization is considered to be one of the several pathways of agricultural development. In modern agricultural practices, mechanization of farm is needed from the view point of the profitability of agriculture. Farmers, whether in the developed or developing economies, mechanize farm operations when the biological sources of energy, e.g., human and animal labour become more costly than the mechanical sources. There is a secular tendency everywhere for the biological sources to become costlier than the

5 mechanical sources. This is due, in part, to the increasing ease with which capital can be substituted for labour (rise in the elasticity of substitution) in agricultural and partly to the rise in the cost of human and animal labour relative to that of machines and fuel. With this end in view, the present study has been taken up with the overall objective of assessing quantitatively the impact of mechanization on the net farm income of the rice farmers in Cauvery Delta Zone (CDZ) of Tamil Nadu State. The specific objectives were to assess the extent of farm mechanization in rice farming; to estimate the costs, returns and profitability of rice cultivation for the different size groups of farmers and at different levels of mechanization; to study the resource use efficiency and technical efficiency of mechanized rice farms; to analyze the energy use pattern and efficiency in rice cultivation and to identify the constraints of mechanizations and to suggest policy options. The Cauvery Delta Zone (CDZ) of Tamil Nadu State formed the universe of the study since the zone was the major rice production environment which produces more than 40 per cent of the state rice production. The districts of Thanjavur, Thiruvarur and Nagapattinam were chosen purposively since these districts constitute around 70 per cent of the total ayacut area of Cauvery canal. A multi stage stratified random sampling procedure was used to with CDZ as universe, districts as the first stage, taluks as second stage, blocks as third stage, villages as fourth stage and the ultimate sampling units were the farmers. In total, six taluks, 12 blocks, 20 villages and 240 respondents were selected. The sample farms were post-stratified into four groups of farms, namely partially mechanized small farms, partially mechanized large farms, fully mechanized small farms and fully mechanized large farms. Farms with operations like ploughing, transplanting, irrigation, harvesting, threshing and cleaning mechanized were treated as fully mechanized farms while the farms with the above operations mechanized except transplanting were treated as partially mechanized farms. Weeding has been excluded in both the farms as this operation is not yet mechanized. Simple average, percentage, frequency analysis were used to study the demographic profile of farm families namely, family size, education, occupational pattern, cropping pattern, asset position, energy use pattern in rice cultivation, farm income and other socioeconomic characters. Cobb-Douglas production function and stochastic frontier production functions were employed to measure the technical efficiency of rice farms. Partial budgeting technique was used to ascertain the economic impact of rice transplanters and Garrett s ranking technique was used to study the constraints in farm mechanization.

6 The analysis indicated that the present extent of mechanization ranged from per cent for partially mechanized small farms to per cent for fully mechanized large farms. On an average, the extent of mechanization was per cent for all small farms and per cent for all large farms. As regards level of mechanization, partially mechanized farms had the level of mechanization upto per cent and the fully mechanized farms had per cent of mechanization. For partially mechanized farms (small and large), the total variable cost had worked out to Rs.46,174 per hactare for small farms and Rs.42,843 per hactare for large farms, proving that higher the farm size, lesser the cost of cultivation. The cost of production decreased with increased farm size and increased use of machineries. The impact of these factors had also reflected in the output of the crop. The yield obtained was Qtl per hactare for small farms and Qtl per hactare for large farms, resulting in a better performance on net farm income and the cost of production per unit of output. The net income over variable cost was Rs.15,506 per hactare for small farms and Rs.19,530 per hactare for large farms and the cost of production worked out to Rs.949 per Qtl and Rs.829 per Qtl for small and large farms, respectively. The benefit-cost ratio worked out to 1.35 for small farms and 1.49 for large farms. For fully mechanized farms (small and large), the cost of cultivation of rice had worked out to Rs.43,333 per hactare for small farms and Rs.40,028 for large farms. The significant difference in cost structure was the component of human labour. The production of rice for the small farms was Qtl per hectare and Qtl per hectare for large farms, revealing the efficiency of large farms. The net income realized by the small and large farms was of the order of Rs.22,081 per hectare and Rs.24,692 per hectare, respectively. The cost of production per quintal of rice worked out to Rs.795 for small farms and Rs.710 for large farms, indicating the advantage of economies of scale. The BCR was 1.54 for small farms and 1.65 for large farms. The resource use efficiency analysis based on significance of input variables (Cobb- Douglas and Stochastic frontier production function) indicated that the variables farm size and machineries were the important determinants of the efficiency of rice farms. Higher the farm size and higher the level of machineries increase the overall technical efficiency of the farms. The elasticity coefficients obtained for these two variables for the small and large farms and for partially and fully mechanized farms had indicated that the fully mechanized

7 large farms were the most efficient among the four different types of farms. Besides the study revealed that the component of machine labour was on the higher usage by the large farms while the small farms had used more of human labour. The energy use efficiency analysis indicated that fertilizer was the dominant source of energy contributing 12,399.7 MJha -1 which accounted for 51.1 per cent of the total energy utilized in rice cultivation. The total energy utilized for rice cultivation for mechanized small farms was 21, MJha -1 while mechanized large farms contributed 25, MJha -1. The operation wise energy use pattern in rice cultivation showed that among all the operations, manures and manuring consumed highest amount of energy (9, MJha -1 ) for mechanized small farms and 11, MJha -1 for mechanized large farms, followed by harvesting, threshing and cleaning which consumed 2, MJha -1 for small farms and 3, MJha -1 for large farms. Further, the total cost of input energy was found to be higher in the case of mechanized small farms with Rs.42,598 and that of mechanized large farms, was Rs.38,998. The total cost incurred per unit of input energy was Rs.1.98 and Rs.1.52 for the small and large farms, respectively. The large farms have consumed 25, MJha -1 of energy while the small farms used 21, MJha -1 indicating that higher the output energy (11, MJha -1 ) higher the level of input energy use ( MJha -1 ). The output-input energy ratio was the maximum (4.75) for fully mechanized large farms and the minimum (4.0) for partially mechanized small farms. The large farms were found to be more efficient than small farms with regard to specific energy ratio with 5.26 and 5.10 for small and large farms, respectively. The energy productivity ratio was the maximum for fully mechanized large farms and it was the minimum for partially mechanized small farms. The net energy return was the highest for fully mechanized large farms (96,037 MJha - 1 ) and the lowest for partially mechanized small farms confirming that higher the farm size higher the level of mechanization and the energy use efficiency was also higher. The energy intensity ratio obtained for the four types of farms also hold the same view. The ratio ranges from 0.47 to 0.72 as the size of farm and the level of mechanization increases. The constraints for mechanization as expressed by the respondents were (i) Nonavailability of certain types of major machineries like transplanters and combine harvesters in the peak season. (ii) Higher cost of machineries and higher rent charged by the firms were felt by the majority of the respondents as the major constraints. (iii) Lack of financial assistance and poor scope for custom hiring due to seasonal operations were preventing the sample farmers in owning certain machineries. (iv) the farm size and the soil type were also

8 expressed as major constraints for mechanization particularly by the small farmers and those in the tail end of CDZ. The study had succinctly revealed the importance of machineries on agricultural production. In order to make the farms self-dependent, efforts to be made to make the timely availability of the vital farm machineries needed by the farming community. The study has further confirmed that the farm size and the machineries are the crucial factors deciding the rice production. Since the size of farms is predominantly small, community organizations based on co-operative basis may be organized to make the farms potentially viable through enhanced level of mechanization. Weeding, one of the labour intensive operations in rice cultivation still being carried out manually or chemically for want of suitable machineries. The existing machineries were felt as either drudgery or uneconomical. The agricultural engineering department and the agricultural university should concentrate on the release of appropriate cost effective and ergonomically feasible devices for weeding, so that the rice farms of Cauvery Delta Zone (CDZ) would move towards full scale mechanization in the near future.

9 CONTENTS CHAPTER NO TITLE PAGE NO I INTRODUCTION 1 II CONCEPTS AND REVIEW 11 III DESIGN OF THE STUDY 36 IV DESCRIPTION OF THE STUDY AREA 53 V RESULTS AND DISCUSSION 72 VI SUMMARY AND CONCLUSION 118 BIBLIOGRAPHY APPENDICES

10 LIST OF TABLES Table No Title Page No 1.1 India s Rice Production in Global Context Area, Production and Productivity of Rice Trend in Area under Rice in Different Countries Trend in Production of Rice in Different Countries Trend in Area under Rice in Different States Area, Production and Productivity of Rice Availability of Farm Power in India Contribution of Different Power Sources in India Distribution of the Samples in the Study Area Details of Selected Sample Farms Analysis of Variance Distribution of Rainfall in Thanjavur District Land Utilization Pattern in Thanjavur District Land Holding Pattern in Thanjavur District Source wise Area Irrigated in Thanjavur District Cropping Pattern Changes in CDZ Annual Compound Growth Rates for Rice in CDZ Cropping Pattern in Thanjavur District Demography of Thanjavur District Farm Machinery Population in Thanjavur District Distribution of Rainfall in Thiruvarur District Land Use Pattern in Thiruvarur District Land Holding Pattern in Thiruvarur District Source Wise Area Irrigated in Thiruvarur District Cropping Pattern in Thiruvarur District Demography of Thiruvarur District Farm Machinery Population in Thiruvarur District Distribution of Rainfall in Nagapattinam District Land Use Pattern in Nagapattinam District

11 4.19 Land Holding Pattern in Nagapattinam District Source Wise Area Irrigated in Nagapattinam District Cropping Pattern in Nagapattinam District Demography Nagapattinam District Farm Machinery Population in Nagapattinam District Distribution of Sample Respondents across Different Age Groups Educational Status of Sample Farms Experience of Sample Farmers Family Size and Number of Members Engaged in Agriculture Household Annual Income of Sample Farmers Size of Holding of Sample Farms Average Investment Pattern in the Sample Farms (a) Cropping Pattern of the Sample Farms (b) Cropping Pattern of the Sample Farms (a) Extent of Mechanization in the Sample Farms (b) Extent of Mechanization in the Sample Farms Cost of Cultivation / Production of Rice per ha (Partially Mechanized Small and Large Farms) 5.11 Cost of Cultivation / Production of Rice per ha (Fully Mechanized Small and Large Farms) 5.12 Cost of Cultivation / Production of Rice per ha (Mechanized Small Farms) 5.13 Cost of Cultivation / Production of Rice per ha (Mechanized Large Farms) 5.14 Cost of Cultivation / Production of Rice per ha (Mechanized All Farms) 5.15 Cost of Cultivation / Production of Rice per ha (Small and Large Sized Farms) 5.15 (a) Cost of Cultivation / Production of Rice per ha (Summary) Partial Budgeting for Rice Transplanter (Fully Mechanized Small farms) 5.17 Partial Budgeting for Rice Transplanter (Fully Mechanized Large farms) 5.18 Estimated Cobb Douglas Production Function (Partially Mechanized Small Farms)

12 5.19 Stochastic Frontier Production Function (Partially Mechanized Small Farms) 5.20 Frequency Distribution Based on Technical Efficiency (Partially Mechanized Small Farms) 5.21 Estimated Cobb Douglas Production function (Partially Mechanized Large Farms) 5.22 Estimated Stochastic Frontier Production Function (Partially Mechanized Large Farms) 5.23 Frequency Distribution based on technical efficiency (Partially Mechanized Large Farms) 5.24 Estimated Cobb Douglas Production Function (Fully Mechanized Small Farms) 5.25 Estimated Stochastic Frontier Production Function (Fully Mechanized Small Farms) 5.26 Frequency Distribution based on technical efficiency (Fully Mechanized Small Farms) 5.27 Estimated Cobb Douglas production function (Fully Mechanized Large Farms) 5.28 Estimated Stochastic Frontier Production Function (Fully Mechanized Large Farms) 5.29 Frequency Distribution based on technical efficiency (Fully Mechanized Large Farms 5.30 Source Wise Energy Use Pattern in Rice Cultivation (Mechanized Small Farms) 5.31 Source Wise Energy Use Pattern in Rice Cultivation ( Mechanized Large Farms) 5.32 Operation Wise Energy Use Pattern in Rice Cultivation (Mechanized Small Farms) 5.33 Operation Wise Energy Use Pattern in Rice Cultivation ( Mechanized Large Farms) 5.34 (a) Costs of Different Sources of Input Energy Use (Mechanized Small Farms) 5.34 (b) Costs of Different Sources of Input Energy Use 112 ( Mechanized Large Farms) 5.35 Energy Use Efficiency in Rice Cultivation (All Size and Level 113 of Farms) 5.36 Mechanization Index (All Size and Level of Farms) (a) Constraints in Mechanization (Partially Mechanized Farms) (b) Constraints in Mechanization (Fully Mechanized Farms)

13 LIST OF FIGURES Figure No Title Page No 1.1 Area, Production and Productivity of Rice Map Showing the Study Area Number of Selected Sample Farms Frequency Distribution Based on Technical Efficiency (Partially Mechanized Small Farms) Frequency Distribution Based on Technical Efficiency (Partially Mechanized Large Farms) 5.3 Frequency Distribution Based on Technical Efficiency (Fully Mechanized Small Farms) 5.4 Frequency Distribution Based on Technical Efficiency (Fully Mechanized Large Farms ) Rice Transplanter in Operation (Slide) 89 LIST OF APPENDICES Appendix No Title 1. Land Use Pattern (Tamil Nadu and CDZ) Area, Production and Productivity of Rice in CDZ 3. Block Wise Area Under Rice in Thanjavur District Block Wise Area Under Rice in Thiruvarur District Block Wise Area Under Rice in Nagapattinam District Source Wise Area Under Irrigation in CDZ Farm Machinery Available in Cauvery Delta Zone 8. Distribution of Area Under Rice by Districts 9. Canal Water Release in Cauvery from Mettur Dam 10. Details of Major Farm Machinery in CDZ

14 CHAPTER I INTRODUCTION Rice (Oryza sativa) is the staple food and principal crop in humid and sub-humid Asia. From the Philippines in the east to eastern India in the west, from central and southern China in the north to Indonesia in the south, rice accounts for between 30 and 50 per cent of agricultural production and between 50 and 80 per cent of dietary intake (Hossain and Narciso, 2011). Because of its importance in providing national food security and generating employment and incomes for the low-income sectors of society, most Asian governments regard rice as a strategic commodity. Asia has done remarkably well in terms of meeting the food needs of its growing population over the last three decades. Since the mid-1960s, rice production has increased at a rate of 2.6 per cent per year, keeping pace with population growth and the income-growthinduced changes in per capita food consumption. Over four-fifths of the growth in production was due to the increase in yields, made possible through gradual replacement of traditional varieties with modern cultivars developed in rice research stations, supported by public investment for expansion of irrigation infrastructure, extension system and supply of credit facilities. Food grain production is the most important activity in India, which provides income and employment to a large section of the population. Among the food grain crops, rice is most important in terms of area coverage and supply of calories in the diet. Rice provides about 30 per cent of total calories in the Indian diet (Mclean et al; 2002). Given that the country still has about 37 per cent of its population below poverty line (Govt., of India, 2009), the growth in rice productivity is critical to the well being of millions of consumers and producers. Further the Indian rice production accounts for 21 per cent of global rice production thus contributing largely to global food security. Growth in population and economic prosperity are the two driving forces for increasing rice demand in India. According to the National Commission on Population, India s population will be 1340 millions in It is estimated that the demand for rice will be million tonnes by the year (Kumar et al; 2009). In order to achieve this target, the productivity of rice has to be brought to the level of 2.7 tonnes per ha, which is 2.2 tonnes presently. The present rate of production growth (1.27 per cent) is below the 1

15 population growth of 1.63 per cent. Therefore the present deceleration trend in production and yield is a cause of concern and has to be reversed. Indian share in global rice production has been hovering in the range of to per cent as shown in Table 1.1. Indian share dipped below 20 per cent only in Production of rice in India is expected to drop this year ( ) from million tonnes to million tonnes due to lower kharif output pegged at almost million MT. Year Table 1.1: India s Rice Production in Global Context World production in Million MT Indian production in Million MT India s Share (Percentage) * Source: MT-Metric Tonnes * Projected Rice crop in India occupying one third of area under cultivation and contributing about 44 per cent of calorie requirement to more than 70 per cent Indians. Rice production in India has transformed from 30 million tons in 1965 to a sustainable strong surplus level of million tons in Though there has been marked increase in productivity of rice from 1.10 to 3.03 MT/ha through , it is still much below the yields ranging from MT per hectare achieved by some of the countries like Egypt (9.43 MT per hectare), China (6.23 MT per hectare), Indonesia (4.43 MT per hectare), Vietnam (4.25 MT per hectare), Bangladesh (3.35 MT per hectare), Myanmar (3.24 MT per hectare) and Philippines (3.08 MT per hectare) (Table 1.2). Currently India is ranked last in terms of productivity of rice when compared with several countries of the world and the average productivity of rice in India is 3.01 Metric tonnes per hectare against the world average of 3.89 Metric tonnes per hectare. Further to sustain the present level keeping pace with the present level of population growth, yield improvement of not less than 2.5 to 3.5 per cent annual growth rate is required to add at least million Tonnes of additional rice per annum (Sheety et. al, 2013). 2

16 Table 1.2: Area, Production and Productivity of Rice (World Scenario) ( ) Area Production (Million Productivity Country (Million hectare) tonnes) (Metric Tonnes / ha) Egypt China India Indonesia Bangladesh Vietnam Myanmar Philippines World Source: Figure 1.1 Area, Production and Productivity of Rice (World Scenario) The global trend in area and production of rice given in Tables, 1.3 and 1.4 revealed that the trend in India s rice production has not been encouraging. While the area increased from million hectares in to million hectares in , the production has marginally increased from million tonnes to million tonnes for the same period. 3

17 Table 1.3: Trend in Area under Rice in Different Countries (MH) Country / Year Egypt China India Indonesia Bangladesh Vietnam Myanmar Philippines World Source: Rice Production in India MH Million Hectares The country's rice productivity increased from 2.21 tonnes per ha in to 3.01 tonnes per ha in , recording a maximum growth rate of 2.77 per cent during this period. The increase has been ascribed to the progressive increase in the contribution from all over the country, in general, and from eastern India, in particular, The compound growth rate of the northern region (Punjab, Haryana and western Uttar Pradesh), which was 6.46 and 4.24 per cent, respectively, in the two preceding decades before 2001, decreased to 2.57 per cent throughout the period to , while in the southern region it has been more or less static. Eastern India, which accounts for more than 60 per cent of the country's rice area, registered an impressive rate of 2.82 per cent in the period to , compared with 0.83 per cent during the preceding decade. However, analyses of the rice productivity at the state and district levels and according to season in high-productivity areas indicate rice yield declines, stagnates and decelerates. Table 1.4: Trend in Production of Rice in Different Countries (MT) Country / Year Egypt China India Indonesia Bangladesh Vietnam Myanmar Philippines World Source: MT Million Tonnes 4

18 Of the seven intensively cropped and highly productive rice states, Andhra Pradesh, Tamil Nadu and Karnataka represent irrigated rice-rice areas, Punjab and Haryana represent rice-wheat systems, while Orissa and West Bengal grow rainfed rice during kharif followed by dry season rice. There was an increase in productivity throughout the last decade, in Andhra Pradesh, Karnataka, Tamil Nadu, Punjab and Haryana of 2.51, 3.65, 4.01, 0.94 and 1.17 per cent, respectively. Compared with the preceding decade, growth rates were found to decelerate marginally in Andhra Pradesh, and drastically in Punjab and Haryana. As Orissa and West Bengal have picked up modern technologies, the yields in these states have been steadily increasing since Decelerating and declining trends in rice productivity in the irrigated ecosystems are a matter of great concern since, with 45 per cent of the rice area, these ecosystems contribute 61 per cent of the national rice production. To maintain the present level of sustainability and to be able to produce the targeted 113 million tonnes of grain paddy by the year 2020, efforts should be made to identify the appropriate technologies and the ecosystems to achieve the target. Table 1.5: Trend in Area under Rice in Different States (MH) States / Year Uttar Pradesh West Bengal Punjab Haryana Andhra Pradesh Orissa Tamil Nadu All India Source: Directorate of Economics & Statistics, New Delhi ( ). MH- Million Hectare The trend in area and production of rice for the seven intensively cropped states of the country is given in Table, 1.5 and 1.6. The data reveal that the area under rice has been more or less static in all the major rice growing states but the production increased from million tonnes in to million tonnes in

19 Table 1.6: Area, Production and Productivity of Rice in India ( ) State Area (MH) Production (MT) Productivity (MT/H) West Bengal Uttar Pradesh Orissa Andhra Pradesh Tamil Nadu Punjab Haryana All India Source: Problem Focus MH-Million Hectares, MT- Million Tonnes Indian agriculture is required to achieve food production of million tonnes by 2020 with the available land mass remaining at about 142 million hectares; it has to come through essentially a vertical expansion, gains through productivity in commodities. There is going to be demand for farm machineries that are ergonomically sound, economically affordable, economy in input use and quantum jumps in productivity. Unlike industry, where men, machines and materials are brought under one roof, agriculture requires these three being moved and various operations performed timely that gives the desired productivity. In this context, availability of appropriate farm power sources are imperative. However for the current level of intensity of farming and the required level of productivity, electro mechanical sources of farm power have to replace the human and animate sources of power. Farm mechanization is considered to be one of the several pathways of agricultural development. In modern agricultural practices, mechanization of farm is needed from the view point of the profitability of agriculture. Farmers, whether in the developed or developing economies, mechanize farm operations when the biological sources of energy, e.g., human and animal labour become more costly than the mechanical sources. There is a secular tendency everywhere for the biological sources to become costlier than the mechanical sources. This is due, in part, to the increasing ease with which capital can be substituted for labour (rise in the elasticity of substitution) in agricultural and partly to the rise in the cost of human and animal labour relative to that of machines and fuel. Further, as a result of globalization and liberalization, the mechanization of the farm becomes utmost necessary because to have a comparative cost advantage of the farming practices. With the implementation of the modern farming machinery, the cost of cultivation 6

20 may be reduced to a substantial level. Mechanization of farm is expected to generate enormous development opportunities for the agricultural sector. It will increase the marginal productivity of labour substantially and have a higher return per unit of land and labour. But the farm mechanization requires more initial capital, improved technical know-hows and quality support services. Lack of access to those services may constrain farmers to involve in farm mechanization. This study based on the primary field survey data tries to analyze the efficiency of mechanized farms with reference to the productivity of rice crop. The productivity of farms depends greatly on the availability and use of farm power by the farmers. Agricultural machines increase productivity of land and labour by meeting timeliness of farm operations and increase work output per unit time. Mechanization also enables efficient utilization of inputs such as seeds, fertilizers and irrigation water. During the last 50 years, the average farm power availability in India has increased from about 0.31 KW/ha in 1961 to about 1.35 KW per hectare in While in 1961 about per cent farm power was coming from animate sources, the contribution of animate sources of power reduced to about per cent in 2011 and that of mechanical and electrical sources of power increased from 5.10 per cent to per cent during the same period. The power productivity relationship shows that those states having higher farm power availability per hectare have higher productivity. The state of Punjab which is highly mechanized had about 98 per cent diesel engine power. The availability of farm power in India and the contribution of different power sources are presented in Tables, 1.7 and 1.8. Table 1.7: Availability of Farm Power in India Source wise (In Percentage) Total Power KW/ha Year Animal Mechanical Electrical Source: CIAE, BHOPAL The contribution of different power sources to the total power changed with time. The share of agricultural workers continuously declined since 1981 and to be only 5.09 per cent by and that of draught animal power from per cent to 6.37 per cent in the 7

21 same period. But the farm power from mechanical sources has increased from 7.5 per cent in 1971 to per cent in Table 1.8: Contribution of Different Power Sources in India Year Agrl. Worker Draught animal Machineries Total Power KW/ha * Source: CIAE, BHOBAL 8 *Estimated The ICAR in its vision 2020 document has projected the demand of food grains at about MT by The Planning Commission of Government of India has estimated the food grain requirement of 280 MT by the end of XII th five year plan ( ). Since the cultivated area cannot be increased, the additional production will be possible only by increased productivity and increased intensity of cropping. This will call for precision farming and timely farm operations which will require high capacity and precision equipment for which farm power availability to be increased substantially. To meet the demand for food grains production of about million tonnes by 2020 as projected by ICAR, the overall productivity of food grain production at National level will have to be increased from the present level of 3010 Kg per hectare (in 2011) to about 4300 Kg per hectare by 2020, for which besides other things, the average farm power availability will have to be increased from the present level of about 1.35 to 2.00 KW per hectare by There is close nexus between farm power availability and agricultural productivity. The level of farm power availabililty during 2001 was about 1.35 KW/ha and at the present level of sophistication in agriculture for taking two crops per year an average farm power availability of 2.0 KW/ha is considered essential (Anwar Alam 2011). Technological advancement is one of the important input coefficients that contributes more to the incremental production. The literature published during the last few decades on the impact of mechanization has produced conflicting evidences. In view of the lumpy nature of investment required for machineries, access to them by operators of various farm sizes may not be necessarily uniform (this is in contrast with chemical fertilizers and improved seeds, which are divisible in nature and generally all farmers can have access to them). Analyzing, therefore, mechanization of farms as a form of technical change on farm output is not known with a certain degree of accuracy. From the policy point of view it becomes necessary to have a complete knowledge of the various effects of a certain input. The farm

22 machineries, for example may be a labour substitute or complement. The use of machineries in agricultural production therefore should be appreciated if it increases the productivity. The focal point of these issues centers around four major issues. 1. Does the mechanization increase farm productivity? 2. To what degree is labour displaced? 3. With the rising prices of fuel energy, is it rationale and economical to mechanize? 4. What policies should the government follow to obtain the desirable benefits of mechanization while minimizing the undesirable effects? Given the four major issues of farm mechanization, this study will address itself only to the first one. 1.3 Objectives The major objective of the study is to quantitatively assess the impact of mechanization on the net farm income of the rice farmers in Cauvery Delta Zone (CDZ). The specific objectives are: 1. To assess the extent of farm mechanization in rice farming 2. To estimate the costs, returns and profitability of rice cultivation for the different size groups of farmers and at different levels of mechanization. 3. To study the resource use efficiency and technical efficiency of mechanized rice farms. 4. To analyze the energy use pattern in rice cultivation, and 5. To identify the constraints of mechanizations and to suggest policy options. 1.4 Hypotheses For the above, it has been hypothesized that 1. There exists no variation in the extent of mechanization among the different size of holdings. 2. There is no difference in cost of cultivation and yield of rice between the different levels of mechanization. 3. There exists technical inefficiency at different levels of mechanization. 4. Energy use pattern and efficiency do not vary among the farmers towards rice production. 9

23 1.5 Scope of the Study The results of the study could be used to have an appropriate model for mechanization of rice farms in the Cauvery delta zone of the state thereby to increase the production and productivity of the crop. The results of the study would be useful to the farmers in making rational investment on machineries. The study would throw light on the constraints being faced by the farmers of the Cauvery delta zone to achieve the full scale mechanization of rice farming. 1.6 Limitations of the Study The study was confined to CDZ, a limited geographical area. Besides, the study was mainly based on primary data collected by personal interview with the respondents using pretested questionnaires. In the absence of proper farm records and accounts, the response could be subject to recall bias. However, efforts were taken to minimize the recall bias through cross checks. The study was undertaken for the reference year Generalization of results could be made with caution. 1.7 Organization of the Thesis The results of the study have been reported under six chapters as detailed below. Chapter I: INTRODUCTION: Setting, problem focus, objectives, hypotheses, scope and limitations. Chapter II: CONCEPTS AND REVIEW: Review of concepts and past studies. Chapter III: DESIGN OF THE STUDY: Sampling design, methods of collection of data and tools of analysis. Chapter IV: DESCRIPTION OF THE STUDY AREA: Location, climate and rainfall, cropping pattern, farm power availability and other information relevant to the study. Chapter V: RESULTS AND DISCUSSION: General characteristics of the sample farms, resource use efficiency, technical efficiency, energy use pattern and constraints. Chapter VI: SUMMARY AND CONCLUSIONS: Brief summary of the findings, conclusions and policy implications. 10

24 CHAPTER II CONCEPTS AND REVIEW To design and conduct a research study successfully, the researcher must be conversant with the concepts and the results of the related past studies. Further, the results could be properly inferred, if the researcher is familiar with the earlier results. Therefore a review of the literature on the concepts and the past studies has been made in this chapter. 2.1 Concepts The following concepts have been discussed and defined for the study. 1. Mechanization and Mechanization index 2. Costs and returns 3. Productivity and Production Functions 4. Resource use efficiency and Technical Efficiency 5. Energy Use efficiency 6. Production constraints Mechanization and Mechanization index Tamil Nadu State Planning Commission (1984) described farm mechanization as the process of increasing farm productivity by investing on efficient tools and equipments. Palanisamy (1993) defined mechanization as an art of equipping agriculture with mechanical aids for increasing efficiency in the farm enterprises. Khalequzzaman and Karim (2007) defined mechanization as the process of injecting power and machinery between men and materials in a production system. According to Verma (2008), agricultural mechanization implies the use of various power sources and improved farm tools and equipment, with a view to reduce the drudgery of the human beings and draught animals, enhance cropping intensity, precision and timeliness of efficiency of utilization of various inputs and reduces the losses at different stages of crop production. 11

25 Rahman (2011) defined mechanized farms as those where the farmers generally used agricultural machineries such as power tillers, threshers for farm operation, on the other hand traditional farms are those who did not use farm machinery and carry out the activities by using animal power and human labour. For the present study, mechanization is defined as the use of machines like, Tractors, Power tillers, Transplanters, Pumpsets, Sprayers and combined harvesters by replacing the human power as well as animal power to reduce the drudgery and improve timeliness of agricultural operations which lead to increased production. There were two categories of farms, namely (i) those owning and using machineries and (ii) those not owning but using machineries on hire. Only the former category was called as mechanized farms, while the latter might qualify to be called as user of machine power but not mechanized farms. For improving farming efficiency, it was enough to use machineries and not necessarily by mechanized. For the present study all the farms that have used machineries whether owned or hired were treated as mechanized farms. Singh et al., (2006) presented a definition of mechanization index based on using living thing and machine input energy and calculated as: Mechanization Index ( IM) CEM / CEH CEA CEM where, CEM: Cost of using machine, CEH: Cost of Manpower and CEA: Cost of Animal Power. In the present study mechanization index has been redefined as the machine and fuel energy divided by the sum of fuel and machine energy as well as animal and manpower energy symbolized as MI Ed /( Ed Eh Ea ) where, MI: Machinery energy, E d : sum of machinery and fuel energy, E h : Human Energy and E a : Animal energy. It shows what contribution machine energy has allocated to itself in rice cultivation vis-a-vis the total energy consumed. The higher value of this index towards one show that most operations are done by machinery indicating that the higher level of mechanization has been utilized. This determination shows what level of mechanization is effective in energy consumption or how unreasonable it is to use machinery Costs and returns Johl and Kapur (1977) divided the fixed cost into two costs namely fixed cash cost and fixed non-cash cost. Fixed cash cost included land tax, insurance premium and annually 12

26 hired labour. Fixed non-cash cost included depreciation on building, machinery and equipments, interest on capital investment, cost of family labour and cost of management. According to Mauraya et al., (1994), the cost of production included the cost of production inputs like seeds, manures, fertilizers, irrigation, plant production chemicals, human and bullock labour, rental value of land at the prevailing market price and overhead cost comprising of interest on working capital and fixed capital, repairs and depreciation. According to Ahuja (1997) variable costs were those costs which increased on the employment of variable factors of production whose amount could be altered in the short-run. Maheswarappa et al., (1998) referred variable cost in terms of human labour, bullock labour, tractor power, seeds, manures, fertilizers, plant production chemicals, irrigation, repair and maintenance cost and interest on working capital. Gurjar and Varghese (2005) while studying cost of cultivation of Rabi crops in Rajasthan defined operational cost as the sum of cost of hired human labour, family labour, bullock labour, machine labour, seed, farm yard manures, fertilizer, insecticides, irrigation charges and interest on working capital. They defined fixed cost as the land revenue and taxes, depreciation on implements and buildings, rent paid for leased in land, rental value of owned land and interest on fixed investment and also defined total cost as the sum of operational cost and fixed cost. Varghese (2007) in his study viewed that the cost of cultivation covered both the paid out cost and imputed cost. The paid out cost included hired labour, maintenance expenses of owning animals and machinery, expenses on material inputs, depreciation on implements, farm building, land revenue, interest on working capital. The imputed costs consist of the value of family labour, rent of owning land and interest on owned fixed capital for which the farmers does not incur any cash expense. Nalini et al., (2008) reported that the cost of production in potato included cost on production inputs like seed tuber, manures and fertilizers, irrigation, owned and hired machinery, labour charges and interest on working capital. The Commission for Agricultural Cost and Prices (2010) defined the cost concepts as: Cost A1 = all variable costs on human labour, bullock labour, machine labour, seed, manures, fertilizers and chemicals + depreciation on building, machinery + land revenue, other taxes 13

27 and interest on working capital. Cost A2 = Cost A1 + rent paid for leased- in land. Cost B1= Cost A1 + interest on value of owned capital assets (excluding land), Cost B2 = Cost B1 + rental value of owned land and rent paid for leased- in land. C1= Cost B1 + imputed value of family labour, Cost C2 = Cost B2 + imputed value of family labour, Cost C3 = Cost C per cent of Cost C2. Johl and Kapur (1977) stated that the product of total production and price would give the gross returns. Singh and Asokan (1981) defined the net farm income as the difference between gross farm income and the total farm expenses (excluding rent paid for leased-in land) including overhead cost of depreciation and land revenue. Murugadas (1990) defined the gross income as the realization by the sale of the product, while net income was the residual gross income after deducting the cost of cultivation. Gurjar and Varghese (2005) defined gross returns as sum of value of main product and value of by-product. The net income is gross income minus total cost. Rohit et al., (2006) worked out a gross return by multiplying the total output with price received by farmers and net returns calculated by deducting the total costs from gross returns. Ram Singh and Abhey Singh (2008) calculated the gross return based on the actual prices received by the growers. Net returns obtained by deducting the respective cost from gross returns. Mathi (2009) calculated net income earned from elephant foot yam production by estimating gross income from yam production, minus cost C 3 by considering value of management input of the farmers as 10 per cent value of the total cost C Productivity and Production Functions Kerr and Swarup (1997), defined productivity as the quantity of output per unit of resource or input. Samuelson and Nordhans (1998), perceived productivity as a concept measuring the ratio of total output to a weighted average of inputs. 14

28 Allan Gowar (2004), considered productivity as the efficiency with which the input is transferred into output. Pandya and Kargaonkar (2007) defined agricultural productivity as an achievement of increased output for every unit of input and also indicated productivity as the ratio of input to output. Henderson and Quandt (1981) referred production function as the mathematical expression of the relationship between the quantities of input and output. Ferguson (1982) defined production function as a schedule showing the increase in the amount of output that can be produced from a specified set of inputs, given the existing technology. In short, production function had been referred as a catalogue of output possibility. According to Barthwal (1992), production function could be viewed as an embodiment of the technology which would yield increased output from the given set of inputs or specify the way in which input would cooperate together to produce a given level of output. Koutsoyiannis (2003) defined production function as a purely technical relation, which connotes factor inputs and outputs. It describes law of proportion: that is, transformation of factor inputs into products (out puts) at any particular time period. The production function includes all the technical efficiencies of production. Raju and Rao (2006) defined production function as a technical and mathematical relationship describing the manner and extent to which a particular product depended upon the quantities of inputs or services of inputs used at a given piece of land, with a given set of technologies and in a given period of time. They classified production function into four types, namely, continuous, discontinuous, short run and long run production functions. In the present study, productivity has been defined as the quantity of rice produced per unit area, i.e., yield in terms of kg per hectare and production function has been defined as the mathematical and technical relationship between the net farm income and the inputs (farm size, seeds, manures and fertilizer, machine labour, human labour and plant protection chemicals, etc) used in the cultivation of rice crop. 15

29 2.1.4 Resource Use Efficiency and Technical Efficiency Efficiency measurement started with Farrel (1957) who distinguished between technical, allocative, price and economic efficiencies. Technical efficiency shows the ability of a firm to obtain maximum output from the given inputs. It is the ratio of output to input and the greater the ratio, the more the magnitude of technical efficiency. Allocative efficiency shows the ability of a firm to utilize the inputs in its disposal at optimal proportion given their respective prices. A firm is efficiently allocative when its production takes place at a point where the marginal value product is equal to the marginal factor cost. Economic efficiency occurs where there are both technical and allocative efficiencies. Farrel s measure of efficiency depends on the existence of the efficient production function with which observed performance of the firm can be compared one of such changes is the development of stochastic frontier production model by Aigner et al., (1977) and Meeusen and Vanden Broeck (1977). These scholars view production function as a locus of maximum output level from a given input set and thus the output of each firm is bounded above a frontier. This frontier is believed to be stochastic in order to capture exogenous shocks beyond control of the firms. According to Heady (1957), efficiency is the capacity or ability of any person, process or thing to realize the specific goal. Economic efficiency is said to be achieved, when the resources are used in a manner to maximize the particular objective, which is relevant to the economic unit under consideration. Sund et al., (1980) classified frontier models into (i) Deterministic non-parametric frontiers (ii) Deterministic parametric frontiers (iii) Deterministic statistical frontiers, and (iv) Stochastic frontiers. Olson et al., (1980) found that there was little difference between Maximum Likelihood Estimates (MLE) and Corrected Ordinary Least Squares (COLS) estimates from their Monte Carlo study of estimation of Stochastic Frontier Production Function. According to Kalirajan (1990) economic efficiency enlists technical efficiency whereby the greatest output could be obtained from any given set of inputs in a technical production function and price efficiency yields equality between the marginal value product and opportunity cost. 16

30 Jayaram et al., (1992) referred technical efficiency as the maximum possible yield achievable with a given level of input use. Greene (1993) revealed that the level of technical efficiency of a particular firm is characterized by the relationship between observed production and some ideal or potential production. Kalirajan and Shand (1994) defined technical efficiency as the ratio of observed output to potential output. Although there is no apriori theoretical reasoning in the stochastic framework of measuring technical efficiency, the potential output is defined as the natural shift from the observed output. Kumbhakar (1994) defined the production frontier as the locus of maximum possible outputs for each level of input use. A producer is said to be technically efficient, if the observed output was maximized, given the input quantities and a failure on the part of the farm to produce the frontier level of output, given the input quantities is attributed to technical efficiency. According to Kumbhaker and Lovell (2000) technical efficiency is just one component of overall economic efficiency. However, in order to be economically efficient, a firm must first be technically efficient. Suresh et al., (2006) defined resource use efficiency as the farm s ability to obtain the maximum possible output from a given level of resources. Owombo et al., (2012) employed logistic regression model to identify the determinants of mechanization and revealed that education, extension visit and machine access were the significant determinants of adoption of mechanization practices. For the present study, technical efficiency is defined as the maximum possible yield achievement to the given level of input use at a given point of time Energy Use Efficiency Mohammadi and Omid (2010) have calculated the energy ratio, energy productivity and the specific energy based on the energy equivalents of inputs and outputs in agricultural production as follows: 17

31 Energy useefficiency Energy output ( MJ ha Energy input ( MJ ha 1 1 ) ) 1 Output ( kgha ) Energy productivity Energy input ( MJ ha 1 ) Energy input ( MJ ha Specific Energy 1 Output ( kgha ) 1 ) Energy 1 Energy input ( MJ ha ) Intensity Cost of cutivation ( Rs ha 1 ) Net Energy Energy output ( MJ ha ) Energy input ( MJ ha 1 1 ) Subramanian et al., (2003) studied the energy requirement in agricultural sector for different farming system in Tamil Nadu and calculated both the direct and indirect energy requirements for Paddy, Sorghum, Maize and Cotton using the energy coefficients adopted by Mittal et al., (1985) as indicated above. In the present study also the same energy equivalents as used by Mittal et al., and Subramanian et al., are followed Production Constraints A constraint is anything that prevents the system from achieving more of its goal. Constraints can be internal or external to the system. An internal constraint arises from the farm itself whereas external constraints arise from out of the farm. Nguyen Cong Thanh (2006) reported that the main constraints of the farmers are focused in three problems viz. agro-ecological constraints, technological constraints and socio-economic constraints for understanding the real situation in rice production and export of India and Vietnam, which are useful to find out the suitable solutions for overcoming the constraints and promoting rice production and exports. In the present study, the parameter namely higher input cost, timely unavailability of machineries, high cost of labours, and lack of technical information were identified as constraints. 18

32 2.2 Past Studies The past studies related to mechanization, mechanization index, concepts of cost and returns, resource use efficiency, technical efficiency, energy use in agriculture and the production constraints were reviewed and presented Mechanization and Mechanization index Mufti and Khan (1995) have carried out the performance evaluation of yanmar APR-8 paddy transplanter and have estimated that the cost of mechanized transplanting was 50 per cent higher than that of manual planting but the former method contributes 30 per cent increase in the output than the latter besides reducing the labour requirement by two-third. Garg, Mehal and Sharma (1997) have calculated that the use of six row paddy transplanter machine could give an increase of 250 kg per ha in the yield, saving of 45 per cent in cost and 60 per cent in manual labour as compared to manual transplanting. Behera et al., (1998) have concluded that in order to reduce the cost of cultivation and the time requirements, it is advisable to replace bullock by machines for operations such as ploughing, seeding, transplanting, weeding and harvesting which are both time consuming and labour intensive. The study of Shrivatsava and Shrivatsava (1998) on the comparative performance of bullock and tractor farms in Madhya Pradesh had demonstrated that the tractors helped to carry out timely operations thereby to realize higher cropping intensity and greater return per unit area for the crops wheat, rice and soybean. The use of tractor influenced the increased use of inputs, more employment opportunity through extensive and intensive use of land, expanded output and maximized net return. Prandhan et.al., (1998) have observed that rice harvesting is labour consumptive and expensive agricultural activity and calculated that cost of mechanical harvesting was just half of that of manual harvesting. Balasankari and Salokhe (1999) carried out a survey of 88 farmers using tractor in Coimbatore, of all the respondents 67 per cent of the respondents reported that the use of tractor helps to overcome the labour shortages during the peak season and that there is sparing of time in using tractor instead of manual labour. 19

33 Muhammad, Sivaswami and Jayan (1999) have examined that paddy cultivation in Kerala needs appropriate mechanization to cope with the increased cost of cultivation due to high wages and scaring of manual labour. Ploughing, transplanting, spraying, harvesting and threshing were done almost completely by using machines. Rane (2000) was of the view that the use of mechanization in sunflower cultivation is necessary to increase yield. They have quantified that the unit per unit of land increased by 20 per cent over the average yield using conventional methods of sunflower cultivation. Rai and Bezbaruah (2002) have reported that mechanization of the farm becomes utmost necessary to have a comparative cost advantage of the farming practices. With the implementation of the farm machinery, cost of cultivation may be reduced to a substantial level. They have further added that mechanization of farms is expected to increase the marginal productivity of labour substantially. Mahrouf and Rafeek (2002) studied the mechanization of paddy harvesting in Srilanka and concluded that the use of combine harvesters reduce the harvesting cost by Rs.3800 per hectare, increases the net returns by Rs.7850 per ha and that the cost of production of paddy was reduced by per cent. The study revealed that per cent of the total cost of production of paddy could be reduced by the use of combine harvesters which will also provide solutions for scarcity of labour during peak harvesting season. Chandrasekaran et al., (2008) analyzed the trend in labour, machinery and bullock labour power use in agriculture clearly indicated that there was a significant reduction in human labour use and bullock labour use in most of the crops and on the other hand, the machinery use on the increasing trend. Ganapathy and Karunanithi (2005) conducted a study in Lalgudi taluk of South India to assess the level of farm mechanization. The study revealed that, the farmers of Lalgudi taluk adopted a high level of mechanization and further revealed that the use of mechanical power was the highest for paddy and the lowest for cotton among other crops. Verma (2008) concluded that farm mechanization enhances the production and productivity of different crops due to timeliness of operations, better quality of operations and precision in the application of inputs. 20

34 Aurangazeb and Khan (2007) had studied the causes and effects of mechanization in Pakistan by classifying the farms as traditional and mechanized. Mechanized farms were those where the farmers generally use agricultural machinery and do not use the traditional methods of cultivation or use it but very rarely. Traditional farms are those where the farmers do not use machinery often or use it but sparsely. They had classified the farms as Small and Large. Farms with an area of upto 5 ha were treated as small and the farms above 5 ha as large. They had further reported that application of farm mechanization will affect the labour requirement, however at the same time the application of mechanization will boost up the overall productivity and production with the lowest cost of production. According the Olaoye and Rotini (2010) the level appropriate choice and subsequent proper use of mechanized input into agriculture has a direct and significant effect on achievable levels of land productivity, labour productivity, the profitability of farming, the sustainability, the environment and on the quality of the people engaged in agriculture. Ajeigbe et al., (2010) in their study asserted that mechanization enhanced productivity and income generating capacity of legume cereal cropping systems in Nigeria. Rahman et al., (2011) studied the effect of mechanization on labour use and profitability of wheat cultivation in Bangladesh. They have classified the farms into two groups such as mechanized and traditional farms to quantify the effect of farm mechanization. Mechanized farms are those where the farmers generally used agricultural machineries such as power tillers, threshers for farm operation on the other hand traditional farms are those who did not use farm machinery and carry out the activities by using animal power and human labour. Karunakaran (2011) had classified the different level of mechanization of farms as fully mechanized and partially mechanized farms for his study on economic evaluation of mechanization in paddy in Cauvery delta zone of Tamil Nadu. He considered the farms as fully mechanized if the major operations viz ploughing, transplanting and harvesting were done with machines. Singh (2006) estimated a mechanization index and its impact on production and economic factors in India and constructed a mechanization index. He concluded that higher the value of this index, higher the level of mechanization. 21

35 Naeimeh Samavateam et al., (2011) have estimated mechanization index for Garlic production in Iran and reported that Garlic production consumed a total of 40, MJ ha -1 energy and energy output was calculated as 26, MJ ha-1. The results further revealed that chemical fertilizer (41.7 per cent) was the major contributor of total energy use in garlic production and the output- input ratio was calculated as The mechanization index of 0.89 was the highest for the farms of between two to three hectares and the lowest index of 0.6 was obtained for the farms of one to two hectares. Owombo et al., (2012) studied the economic impact of mechanization in Nigeria on Maize crop. They have classified the farmers as adopters and non-adopters and concluded that the gross margins per hactare for adopters and non-adopters were Rs.42,575 and Rs. 28,302 respectively and that of mean net revenue were Rs.41,730 and Rs.73,727. The benefitcost ratio were 4.0 and 3.5 respectively for adopters and non-adopters. The logistic regression model revealed that education, extension visit and machine access were the significant determinants of adoption of mechanization practices. For the present study, two levels of farms namely partially mechanized and fully mechanized were considered. Farms which have used machineries for ploughing, transplanting, irrigation, plant protection, harvesting, threshing and winnowing were treated as fully mechanized farms. Farms which were using machineries for all the above operations except transplanting were considered as partially mechanized farms. Weeding has been excluded in both the farms since suitable machinery has not been introduced so far for weeding operations Costs and returns Singh and Asokan (1981) studied the concepts and methods used to estimate income in semi- Arid tropics by assessing the valuation of inputs and outputs. For purchased inputs the market price and for own inputs, the opportunity costs were considered. Seed was evaluated as the waited average annual price. Hired a family labour were estimated on the basis of hourly wage rate while the prevailing rental rates were used to reflect the opportunity cost of bullock labour. For machineries, the rental charges on hourly basis was accounted. Chinnappa (1998) classified the cost into two categories namely variable cost and fixed costs. The variable cost represented the out-of-pocket expenses namely cash expenses on human labour, manures and fertilizers and seed materials and the fixed cost of production 22

36 included mostly non-cash expenses such as depreciation, interest on fixed assets and land rent. He concluded in his study on the cost structure of sugarcane that the variable cost alone constituted per cent of the total cost and the rest per cent accounted by fixed cost. Niranjan et al., (2000) estimated the Gross Economic Return and Net Economic Return for the utilization of combine harvester. GER is the value of the reduced cost of harvesting paddy by the combined harvester. Net Economic return indicates the value of the expenditure incurred by the displaced labourers. He also employed partial budgeting technique and financial viability analysis to evaluate the field level performance of the machinery. Balappa and Hugar (2002) worked out cost and returns for production of tomato. The study revealed that farmers incurred a total cost of Rs.59, per hectare, of which variable cost accounted for per cent. Expenditure on human labour (29.49 per cent) and plant protection chemicals (18.81 per cent) were the major components of variable cost. Sundar and Kombai Raju (2004) computed cost and returns in gloriosa cultivation. The authors classified cost of production into two types namely establishment cost and maintenance cost. They also apportioned total establishment cost to different crop years and included under fixed cost. The study found that the average cost of production per kg of gloriosa seed was higher in small farmers. Narasimha et al., (2004) reported that 65 per cent of total expenditure on black gram was due to variable cost and concluded that the total cost per hectare increased with increase in farm size whereas returns showed an inverse relationship with the farm size. Jitendra Singh et al., (2006) used cost concepts in measuring cost and returns for paddy. The study found that the cost of cultivation for organic paddy over cost A 1 and cost C 3 as Rs.18,786/ha and Rs.31,651/ha and for inorganic paddy it was Rs.19,106/ha and 35,947/ha respectively. The yields from organic and inorganic paddy were q/ha and q/ha respectively. Net returns over cost A 1 and cost C 3 from organic and inorganic paddy were Rs /ha and Rs.7,279/ha; Rs.21,323/ha and Rs.4,483/ha respectively. Varghese (2007) in his study viewed that the cost of cultivation covered both the paid out cost and imputed cost. The paid out cost included hired labour, maintenance expenses of owning animals and machinery, expenses on material inputs, depreciation on implements, farm building, land revenue, interest on working capital. The imputed costs consist of the 23

37 value of family labour, rent of owning land and interest in owing fixed capital for which the farmers does not incur any cash expense. Nalini et al., (2008) reported that the cost of production in potato included cost on production inputs like seed tuber, manures and fertilizers, irrigation, owned and hired machinery, labour charges and interest on working capital. Anwarul et al., (2010) calculated cost and returns in chillies production. The study showed that, on average total variable cost of production and total cost of production per hectare of land were Rs.71,950/- and Rs.78,950 respectively. The net return was Rs.73,164 and the Benefit-cost Ratio was The study concluded that cultivation of chillies was profitable, as BCR was greater than unity. Kudi et al., (2010) estimated on comparative analysis of profitability of nerica rice and local rice varieties production and examined the costs and returns; shows that labour and fertilizers inputs accounted for greater parts of the total variable costs incurred in both nerica rice and local rice varieties and were represented by 74 and 53 per cent respectively. The farm gate price of N80 per kg for nerica and N50 per kg for local rice was used in estimating the revenue and comparing with the total variable costs to obtain the gross margin which measured the economic performance of the two rice varieties. The gross margin analysis shows that from one hectare of land cultivated, the toal cost of production for nerica rice and local rice were N116,638 and N85,803 and gross revenue of N351,280 and N157, /ha respectively, thus making a gross margin of N234, 642 and N71, 699 per hectare respectively. Singh and Grover (2011) studied the economic viability of organic farming in wheat cultivation. They revealed that the total variable cost on per acre basis for the cultivation of organic wheat has been found less as compared to inorganic wheat. The net returns over variable cost of organic and inorganic wheat have been observed as Rs.21,895 per acre and Rs.16,700 per acre for organic growers. Rao (2011) had assessed the economics and sustainability of SRI (system of rice intensification) and traditional methods of paddy cultivation in North Coastal Zone of Andhra Pradesh for the period and shown that BCR was higher for SRI (1.76) than traditional (1.25) methods. Further, there was a 31 per cent yield gap between SRI and traditional methods, in which cultural practices (20.15%) had shown a stronger effect than input use (10.85%). 24

38 In the present study, the cost of production includes both variable costs and fixed costs. The total variable costs include labour costs such as human, animal and machinery usage cost, seed, manures and fertilizer, irrigation, plant protection chemicals and interest on working capital at 12.5 per cent. Johl and Kapur (1977) stated that the product of total production and price would give the gross returns. Ram Singh and Abhey Singh (2008) calculated the gross return based on the actual prices received by the growers. Net returns obtained by deducting the respective cost from gross returns. In the present study, gross income is calculated by multiplying the total output with the price received by the farmers during the time of harvest. Net farm income was calculated over the variable costs Productivity and Production functions Goyal et al., (2005) studied the technical efficiency of paddy farms in Haryana. A translog Stochastic Frontier Production Function was used for analyzing unbalanced panel data. The technical efficiency showed wide variation across sample farms ranging from 0.24 to The mean technical efficiency decline from 0.80 to 0.72 during the two study period. The study indicated the scope to improve the productivity of rice crops with the given level of inputs and technology. Singh (2007) attempted to examine the farm specific technical efficiency of wheat cultivation in Haryana using stochastic frontier approach. The estimates of technical efficiency indicated a high degree of inefficiency in production of wheat. The technical inefficiency worked out to be 27 per cent at the aggregate level and the same was 25,27 and 26 per cent for small and large size farms, respectively. Jyoti et al., (2008) estimated technical efficiency of paddy crop in Jammu District using Stochastic Frontier Production Function. The results showed that the minimum technical efficiency was 10 per cent and the mean technical efficiency was 37 per cent. The maximum number of farms falls under the category of per cent technical efficiency. Ingle et al., (2009) studied the resource use efficiency using Cobb Douglas production function in the cultivation of rose. The study revealed that the regression coefficient of nitrogen (0.018), manure (0.103) and pesticide (0.028) were highly significant at 25

39 one per cent level, while the regression co-efficients of phosphorus (0.025), potash (0.008) and irrigation (0.108) were positive and significant at five per cent level. Mohammed Asmatoddin et al., (2009) used Cobb-douglas production function to estimate resource productivity, resource use efficiency and optimum resource use in cereal crops on medium farms in Marathwada. The study revealed that in case of bajra area was positive and significant at 1 per cent level. Co-efficient of multiple determination was (R 2 ) which indicate per cent variation in independent variables. In regard to rabi jowar area and bullock were positive and highly significant at 1 and 5 per cent level, respectively. Co-efficient of multiple determination was (R 2 ) 0.93 which indicate that 93 per cent variation in explanatory variables. In case of wheat area and nitrogen were positive and significant at 1 and 5 per cent level, respectively. Co-efficient of multiple determination was (R 2 ) 0.90 which showed 90 per cent variation in explanatory variables. Mohammed Asmatoddin et al., (2009), used Cobb-douglas production function to determine resource productivity in cotton and sugarcane. The results revealed that the regression co-efficient of plant protection, nitrogen, phosphorus were positive but nonsignificant. Co-efficient of multiple determinations (R 2 ) was which indicated that per cent variation in all independent variables Resource Use Efficiency and Technical Efficiency Rajasekaran and Krishnamoorthy (1998) studied the technical efficiency in rice production in the kole lands of Kerala to assess the pesticide use behavior of farmers. The farm specific technical efficiencies ranged from 0.49 to 0.92 with a mean of They reported that the absence of proper scientific knowledge was the major determinant of technical inefficiency and over use of pesticides. Reddy and Sen (2004) while studying technical efficiency of rice farms of Andhra Pradesh using the Frontier Production Function found that technical efficiency of rice farms ranged between 6.67 and per cent with an average of per cent. Kumar et al., (2005) estimated efficiency level of irrigated rice farms of Uttaranchal using Data Envelopment Analysis (DEA) approach. The average level of the overall technical efficiency for irrigated rice farms growing local variety was 75 per cent and that for improved variety was 92 per cent. 26

40 Suresh et al., (2006) studied resource use efficiency of paddy cultivation in Peechi command area of Thrissur District of Kerala. The study has examined the resource productivity and allocative as well as the technical efficiency of 71 rice farmers of the command area. The allocative efficiency has indicated that marginal return per rupee increase in marginal, small and large farms would be Rs.2.83, Rs.1.57 and Rs.1.17 respectively. The average technical efficiency of the paddy farmers in the command area has been found as 66.8 per cent. Naidu and Sivasankar (2007) analyzed the influence of various inputs on the credit requirement using Cobb-Douglas function. The amount of credit was taken as the dependent variable while the independent variables were consumption expenditure, current farm expenditure and capital expenditure. The elasticities obtain were , and and significant at one per cent level for the three variables respectively. Dhanabalan (2009) used Cobb Douglas production function to determine resource use efficiency in milk production. The study found all the regression co-efficients were positive and statistically significant, indicating that the producers can increase the milk production by increasing the inputs. Rawlins (1985) evaluated the effects of the Jamaican Second Integrated Rural Developmental Project (IRDPII) on the level of technical efficiency for peasant farmers by using cross sectional frontier model. Although the results showed that non-contract farmers have higher average technical efficiency, it concluded that contract farming drives up the production frontier of contract farmers. Battese and Coelli (1995) used a stochastic frontier production model and proposed to estimate the levels of technical efficiency of small holder maize farmers in Tanzania and the results of efficiency analysis revealed that small holder farmers are not only producing at a lower level but are also operating relatively further from the production frontier. Xiaosang and Jeffrey (1998) used stochastic production function and cost frontier to derive technical, allocative economic efficiencies of Chinese conventional rice and they indicated that the technical efficiency varied widely (varying from 46.5 to 96.7 per cent) across the sample farms and it was time invariant. The mean technical efficiency was computed as 2.0 per cent, which indicated that on an average, the realized output could be increased by 18.0 per cent without resources. 27

41 Hazarika and Subramanian (1999) analyzed the technical efficiency of the Tea Industry in Assam using the stochastic frontier production function model. It was found that 29.4 per cent of the total farms that operated a large farm (estates) belonged to the most efficiency categories (96.0 to 99.0 per cent) and 8.8 per cent in the least efficient group (64.0 to 70.0 per cent). It was also observed that farm specific technical efficiency varied from 0.64 to 0.99 with mean technical efficiency of Mythily and Shanmugam (2000) studied the technical efficiency of rice growers in Tamil Nadu. They specified the Cobb Douglas functional form for estimating technical efficiency. The mean technical efficiency was computed as 82 per cent which indicated that on an average, the realized output can be increased by 18 per cent without any additional resources. The study also found that the technical efficiency varied widely, ranging from 46.5% to 96.7% across sample farms was is time invariant. The gap between realized and potential yield highlighted the need for improving farm level extension services. Alabi and Aruna (2005) analysed technical efficiency of family poultry production using Cobb-Douglas production function. The technical efficiency estimate shows that the technical efficiency of family poultry ranges between 0.09 and 0.63, with a mean of This indicates that on the average, the respondents are 22 per cent efficient in the use of a combination of their inputs. Tijani (2006) estimated technical efficiencies on rice farms in Osun State, Nigeria, and identified some socio-economic factors, which influenced the productive efficiency. These technical efficiencies were estimated using the stochastic frontier production function approach applied to primary data. A translog production function was used to represent the production frontier of the rice farms. He observed that the levels of technical efficiency ranged from 29.4 per cent to 98.2 per cent with a mean of 86.6 per cent, which suggests that average rice output falls 13.4 per cent short of the maximum possible level. Duraisamy (2007) employed Cobb-Douglas production method (OLS Method) and Stochastic Frontier Production Function (MLE Method) to study resource use efficiency and technical efficiency of major crops inclusive of paddy in Theni District, Tamilnadu. He reported that the variables FYM, chemicals, human labour and irrigation were having significant influence on the yield of paddy crop. The inefficiency factor namely, farm experience was positive and significant and this factor alone accounted for the discrepancies 28

42 between observed output and the frontier output arising out of technical inefficiency in paddy output. The other inefficiency factors like farm size, education and extension agency contact were also found to be non-significant factors. Yusuf and Maloma (2007) analyzed technical efficiency of poultry egg production. They found that majority of the farmers were technically efficient with 0.87 mean technical efficiency. Farmers with large farm size were most technically efficient (0.887) followed by medium farm size (0.867). Years of education and experience had a positive effect on technical efficiency while household size had a negative effect. Elibariki(2008) suggested that the technical efficiency is positively associated with level of education, the use of inorganic fertilizer, household size, engaging in small business, and usage of hand hoe, policies targeting these variables among others might have a positive impact on smallholders maize production and productivity by using stochastic frontier production model. Seidu Al-hassan (2008) examined the level of farm-specific technical efficiency of farmers growing irrigated and non-irrigated rice in Northern Ghana used a stochastic production frontier function. He concludes that rice farmers are technically inefficient. There is no significant difference in mean technical efficiencies for non-irrigators (53 per cent) and irrigators (51 per cent). The main determinants of technical efficiency in the study area are education, extension contact, age and family size. Providing farmers with both formal and informal education will be useful investment and a good mechanism for improving efficiency in rice farming. Kiatpathomchai et al., (2009) reported that the technical efficiency of rice farming in Southern Thailand could be improved through the reduction of inputs by 8-14 per cent and the current output of 3.5 tons of paddy rice per ha could be maintained. In order to improve technical efficiency of rice farms in Southern Thailand, our finding lead us to suggest advisory measures which focus on farm level under existing technology. The optimum rates of inputs in Southern Thailand are kg of seed, kg of N-fertilizer, and kg of P fertilizer per ha. Oluwatusin Femi Michael (2011) measured the level of technical efficiency in Nigerian yam farming using yam producers survey data. The primary data for the study were collected randomly from 240 yam farmers selected with multistage sampling technique. The 29

43 data collected were analyzed through Stochastic Frontier Production Function (SFPF) and the results revealed that the cost of yam sett used, labour and farm size were significantly different from zero and of importance in production of yam. Also, education, experience, access to credit were the main socio-economic characteristics affecting the technical inefficiency of yam farmers. Using parametric approaches to production, technical efficiency for rice growing farms were estimated for the sample farms specifically, a stochastic production function is employed. Technical efficiency obtained in this manner is a relative measure where the production frontier is defined by the farmers plots included in its estimation. Economic efficiency described by its component parts: technical efficiency and allocative (price) efficiency. A farmer is more technically efficient (TE) than his counterpart if he produces a higher output from a similar bundle of inputs. Allocative efficiency is reached when marginal cost of input is equal to the value of the marginal product of output. The concept of production frontiers and efficiency can be illustrated in Figure I, using output (Y axis) and inputs (X axis). The production frontier for a farm using best practice techniques is shown by frontier f, which is nothing but the stochastic production frontier. A farm operating at point B on the frontier where the price line P is tangent to its production frontier is economically efficient and there is neither technical nor allocative inefficiency. If on the other hand, the farm operates at point A on the frontier, it receives lower profits, arising due to allocative inefficiency (πa). In reality however, all farms do not operate at their best practices output curve f but below the frontier f due to various constraints such as inappropriate or outdated production technologies, organizational constraints and non-price factors such as information glitches etc and these factors can cause a farm to operate at a point such as C, using an input bundle L2 and receiving lower profit πc. At point C, the farm experiences both allocative and technical inefficiency. A movement to point of production at D would leave the farm allocatively efficient but still technically inefficient as output levels could be raised further to the levels at frontier f. In terms of output loss, a farm operating at C, experiences a shortfall in output given by Q1-Q3. Of this total shortfall, Q2-Q3 is attributable to technical inefficiency and Q1-Q2 is attributable to allocative inefficiency. 30

44 Figure 2.1: Production frontier, output oriented p B Q ( ) 1 b p f Q ( ) 2 a A D f 1 Output Q ( ) 3 c C L 2 L 1 Input There are various approaches to measuring efficiency, which can be categorized into parametric and non-parametric methods. Parametric techniques are further classified into deterministic and stochastic methods. In stochastic production frontiers, each firm s efficiency is measured relative to its own frontier rather than to some industry wide frontier. In essence, the difference between deterministic and stochastic methods lies in the treatment of the error term. In deterministic methods, the error is implicitly assumed and makes no distinction between unobserved variables that lie outside the control of the agent and those that lie within it. Stochastic models decompose the error term into purely statistical noise (that lies outside the control of the production agent), and inefficiency (a one-sided error term). Aigner et al., (1977) proposed the stochastic frontier production function with two independent error components. The one accounts for the presence of technical inefficiencies in production and other accounts for measurement errors in output and the combined effects of unobserved inputs in production. This methodology was used in number of studies to measure the technical efficiency. In this study the general production function (Battese and Coelli, 1995), with inefficiency effects defined as Yi f ( xi; )exp ( vi ui ) i 1,2,3,......, n 31

45 Where Y i denotes the output quantity of the i th farm, x i is a (1 x j) vector of input quantities and is a (j x 1) vector of unknown parameters to be estimated. The vi are two-sided random variables associated with measurement errors in output and other noise in the data which are beyond the control of firms. v i is assumed to be independently and identically distributed N 0, v ) and independent ofu i. In the absence of stochastic term u i, the model in ( 2 equation reduces to purely deterministic (mean) production function. The u i is defined as non-negative random variables which account for technical inefficiency effects in production and are independently distributed as truncations at zero of the N, ) distribution, k ( i u2 where: i 0 k ik i and i is a (I x K) vector of farm characteristics that affect k 1 efficiency and is an (K x 1) vector of parameters to be estimated. The s are random 2 variables generally defined by the normal distribution with zero mean and variance, with point of truncation as. i i Maximum likelihood estimation methods were used to simultaneously estimate the stochastic frontier and technical inefficiency effects models. For the likelihood function the variance term are parameterized as: u v and u / ( u v ), with 0 1(Battese and Coelli, 1995) The technical inefficiency for the i th firm is estimated as the expectation of u i conditional on the observed value v u ) : ( i i i TE i E[exp( u )[ v i i u ] E[exp( i 0 k k ik i k i v u )] i i In the present study resource use efficiency is measured in value terms by converting all inputs and output by their respective prices and estimated through stochastic frontier production function. 32

46 2.2.5 Production constraints Siju (2001) employed Garrett s ranking technique to rank the constraints faced by the cashew processing industry and revealed that marketing, labor unions, scarcity of capital, scarcity of labor, quality of raw material, procurement and storage were the main problems experienced by the farmers. In these problems, marketing ranks first and labour unions second. Jayachandran (2002) administered Garrett s ranking technique to rank the constraints involved in maize production. He found that inadequate transport facilities were the major problem faced by farmers, followed by the distant location of the regulated market, inadequate storage facilities, price fluctuation, etc. Balaji et al., (2003) used Garrett s ranking method to rank the constraints faced by the production and marketing of groundnut and it included incidence of pest and disease, erratic rainfall, water scarcity, forest animals, non-availability of good quality seeds, inadequate supply of labour coupled with the higher wage rate, low level of adoption of recommended technologies, lower marketed surplus, collusion among the traders in marketing, malpractices in weightment and delayed payment. Senthilnathan (2004) used Garret s ranking techniques to rank the benefits due to watershed implementation like soil and water conservation, soil fertility improvement, cropping pattern, increase in cropping intensity and ground water recharge. Sudha (2005) employed Garrett s ranking technique to find the constraints involved in adoption of Integrated Pest Management (IPM) Technology. She found that wage of labour as the major problem with the score of followed by non-availability of labour, time, lack of IPM inputs, lack of extension follow up practices, lack of proper training facilities, lack of confidence, complex practice, fragmented land holdings and lack of assured irrigation. Kumar and Kumar (2008) employed Garret ranking technique to identify the constraints faced by the farmers in contract farming. They found that delayed payment for crop production, lack of credit for crop production, scarcity of water for irrigation, erratic power supply and difficulty in meeting quality requirements were the major constraints faced by contract farmers. 33

47 Pradeep et al., (2009) studied that the constraints for production of small scale agro processing industries. By using the Garrett s ranking technique, the constraints were classified as serious and not serious. The serious problems were unfamiliar with export activities, lack of market intelligence. Very serious problems were cheaper/superior competitive substitute, inadequate supply of export information and high cost of packaging. Finally not so serious were high import and excise duties on packaging material, seasonal material, seasonal damage and high sales tax on packaging material. Sedaghat (2011) adopted Garret ranking technique to identify the constraints in production and marketing of Iran s Pistachio and results indicated that; inadequate irrigation, unsuitable domestic market structure accompanied with low received prices, price fluctuations and lack of appropriate chemical fertilizers were the major problems from the farmers point of view, while afflatoxin contamination standards, changing government policies toward export and irregular supply of produce to the market during the year were the sole hindrances from the traders/exports point view. In the present study also the Garrett s ranking technique has been used to rank the constraints as expressed by the respondents Energy Use Pattern Gyanendra Singh (2006) studied the contribution of the different sources of energy viz human, animate and mechanical in the production of various crops, state wise in India and also developed the mechanization indices for the crops, using secondary data on cost of cultivation of crops. The study revealed that 78.5 per cent farm power was contributed by the mechanical sources and the mechanization index based on cost of use of machinery was 14.5 per cent. In other words, the share of cost of human and animal energy in the total operational cost was 85.5 per cent. The crop wise mechanization index varied from the lowest value of 8.22 per cent for paddy to the highest value of 30 per cent in wheat. The study further revealed that higher the mechanization index lower the cost of cultivation. Islam et al., (2009) studied the energy utilization in unpuddled transplanting of wet season rice in Bangladesh. They studied the method of optimizing the energy consumption by different types of tillage operations of rice. The study compared the energy consumption pattern under puddled and unpuddled conditions. The operational energy input was found to be the highest (26-27 GJha -1 ) under puddled condition and the least (0.78 GJha -1 ) under 34

48 unpuddled condition of rice cultivation. Energy saving was reported to be of the order for per cent in the latter method due to lesser machinery use and reduced irrigation. Samavatean (2010) studied the energy balance between the input and output per unit area for garlic in Hamaden province of Iran. The study had revealed that the highest share of energy consumption belonged to chemical fertilizers (41.7 per cent) followed by diesel (13.94 per cent) and the total energy input of 40, MJ ha -1 was consumed for garlic production. Abubakar and Ahmad (2010) studied the pattern of energy consumption in millet production for selected farms in Jigawa and Nigeria for different size group of farmers and concluded that the small farms (less than 1.0 ha) consumed the highest total energy value of 6,078 MJ/ha while the large farms (more than 5.0 ha) expended the least amount of total energy value of 1,705 MJ/ha in millet production. Prasanna Kumar et al., (2013) studied the energy use pattern in cotton cultivation under irrigated situations of Raichur district and revealed that fertilizer was found to be the dominant source of energy contributing 3,206 MJ per acre, which accounted for per cent of the total energy utilized in cotton cultivation. The study further revealed that the energy utilized for cotton cultivation by small farmers (6,100 MJ/acre) was significantly higher than that of medium (5,890 MJ/acre) and large (5,621 MJ/ac) farmers. Future research on mechanization may be concentrated taking in to account the effects of different technologies of rice cultivation so that the impact of mechanization and the technology could be studied meaningfully. 35

49 CHAPTER III DESIGN OF THE STUDY Designing of proper methodology is needed to carry out a systematic analysis of any economic problem. This chapter explains the various techniques that have been used to meet the objectives and to test the hypotheses as stated earlier. Accordingly it highlights the selection of the study area, sampling procedure and the sample size. It also describes the variables, data collection and its limitation. The statistical tests, methodology, the model used in the study and the tools of analysis have been explained. 3.1 Selection of Study Area The universe of the study is the Cauvery Delta Zone (CDZ) of Tamil Nadu state. This zone is the major rice production environment which produces more than 40 per cent of the state rice production. In CDZ, Rice is being cultivated in 6.5 lakh hactares constituting more than 60 per cent of the gross cropped area and producing about 25 lakh tones. The zone spreads in 30 taluks of seven districts viz, Thanjavur, Thiruvarur, Nagapattinam in total and parts of Trichy, Perambalur, Cuddalore and Pudukkottai districts of which, the first three districts virtually form the Cauvery Delta Zone due to the intensity of rice cultivation. Cauvery Basin is again the biggest river basin of the state for all basin viz Periyar, Palar, Tamirabarani and Vaigai river basins. Rice has been traditionally cultivated in three seasons viz I season (Kuruvai) June-September; II Season (Samba) August November and III season (Thaladi) October February. Summer cropping of paddy has also been practised in this zone. For the present study, the districts of Thanjavur, Thiruvarur and Nagapattinam were chosen purposively since these districts constitute around 70 per cent of the total ayacut area of Cauvery canal. The map showing the selected area is given in Figure

50 Figure 3.1: Map Showing the Study Area 37

51 3.2 Sampling Procedure A multi stage stratified random sampling procedure was used to with CDZ as universe, districts as the first stage, taluks as second stage, blocks as third stage, villages as fourth stage and the ultimate sampling units were the farmers. In total, six taluks, 12 blocks, 24 villages and 240 respondents were selected. While the taluks, blocks and the villages were selected based on maximum area under paddy, the farmers were selected at random. The details about the selected taluks, blocks and the village wise sample size are presented in Table 3.1 and also in Figure 3.2. Table 3.1 Distribution of the Samples in the Study Area District Taluks Blocks Villages Thanjavur Thiruvarur Nagapattinam Thanjavur Thiruvaiyaru Thiruvarur Needamangalam Mayiladuthurai Kilvelur Ammapet Orathanad Budalur Thiruvaiyaru Thiruvarur Mannargudi Nannilam Needamangalam Mayiladuthurai Kuthalam Kilvelur Keelaiyur Sulaikottai Arundhavapuram Vandayar Irruppu Melaulur Thirukattupalli Cholagan patti Kadambankudi Gangaisamuthram Vaduvurvadapathi Pullavarayankudikadu Vadapathimangalam Vaduvurthenpadhi Anniyur Iluppur Kameswaram Keeranur Thippirajapuram Mekkirimangalam Manalmedu Marudhanallur Thevur Sendankadu Alankadu Radhanallur Sample size

52 3.3 Sources of Data The study was undertaken using both the primary and secondary data Primary Data The primary data on the socio-economic aspects of the respondents, land holding pattern, asset position, cropping pattern, cost and returns, awareness and adoption of mechanization and the constraints thereof were collected from the respondents by personal interview by administering well structured and pre-tested questionnaires at farm level. They were explained about the objectives and the scope of the study was made for research purpose. Since most of the farmers were literates and informative they could furnish the details without any difficulty even though they did not maintain any records. Cross checks were also done to test the validity of the information gathered Secondary Data Secondary information such as revenue details, land use pattern, cropping pattern, rainfall, sources of irrigation, area, production and productivity of Rice crop in different seasons and years, availability of farm machinery in study area and other relevant data required for the study were collected from the offices of the Joint Director of Agriculture of the respective districts, Assistant Director of Statistics, Office of PWD and the Agricultural Engineering, and also from season and crop reports, Agristat and other journals, Newsletters and websites, etc Post Stratification of Sample farms As explained in the preceding chapter, the farms were required to be classified as two groups viz small (upto 2 ha) and large farms (above 2 ha) and again each group had to be classified as partially mechanized and fully mechanized farms to suit the objectives of the study to study the efficiency of the farms at different levels of mechanization. For this purpose, the respondents were post stratified and the distribution of sample farms are given in Table

53 District Table 3.2 Details of Selected Sample Farms Small ( Up to 2 Ha) Partially Mechanized Fully Mechanized Partially Mechanized Large (>2 Ha) Fully Mechanized All farms Total Thanjavur Thiruvarur Nagapattinam CDZ Study Period The primary data collected pertained to the crop year (I,II and III seasons) (Fasli 1421), which was a very normal year with regard to timely release of canal water, normal rainfall and regular coverage of area under paddy in all seasons. Figure 3.2: Number of Selected Sample Farms 40

54 3.5 Tools of Analysis Simple average, percentage, frequency analysis were used to study the demographic profile of farm families namely, family size, education, occupational pattern, cropping pattern, asset position, energy use pattern in rice cultivation, farm income and other socioeconomic characters. Cobb-Douglas production function and stochastic frontier production function were employed to measure the technical efficiency of rice farms. Partial budgeting technique was used to ascertain the economic impact of rice transplanters and Garrett s ranking technique was used to study the constraints in farm mechanization Analysis of Variance One way analysis of variance technique was used to test whether there existed a significant difference in the average net income of the four groups of respondents viz. small farms partially mechanized, small farms fully mechanized, large farms partially mechanized, and Large farms fully mechanized and the results are presented in the Table 3.3. Net Income /Acre Table 3.3: Analysis of Variance Descriptives 95% Confidence Interval for Mean Std. Std. Lower Upper N Mean Minimum Maximum Deviation Error Bound Bound SP LP SF LF Total ONE WAY ANOVA Net Income/Acre ANOVA Sum of Squares df Mean Square F Sig. Between Groups Within Groups Total

55 Net Income/Acre Duncan Subset for alpha = 0.05 Farm code N SP LP SF LF Sig Means for groups in homogeneous subsets are displayed. The results of the ANOVA table are as above. The calculated F statistic was and it was found to be significant at 1 per cent level. This shows that there exists significant difference in net income per acre between the four groups. Among the groups, in order to find out which group was significantly different with regard to average net income, post hoc test was used. Though there are different methods of post hoc test, Duncan s Multiple Range Test (DMRT) is the most popular method. Hence the DMRT was applied to compare the mean values and the results are follows. DMRT results showed that there is significant difference in mean values of all the four groups. The highest average net income of of large farms fully mechanized was found to be significantly different from the next highest net income of of small farms fully mechanized group, followed by of large farms partially mechanized group and of Small farms partially mechanized group of farmers Descriptive Statistics The percentage and average analysis are used to analyze the socio-economic factors, land use pattern, asset positions, labour employment and machine power utilization, in the sample farms.analysis of variance is used to differentiate four categories of farms viz small farms partially mechanized, small farms fully mechanized, large farms partially mechanized and large farms fully mechanized Cost and Returns The per hactare cost of cultivation and the cost of production were worked out for the four different levels of mechanized farms, by average and percentage analyses. 42

56 Variables Included in the model are defined as below Farm size: Farms with the operative area of upto two hectares was considered as small farms and farms with an area of above two hectares was taken as large farms. The size of farm as measured by the operated area of the farm was considered in hectares. The operated area includes owned plus area leased in minus the area leased out. Seeds: The seeds obtained from the own storage or purchased at local or distant market. The market price is considered for both sources. For fully mechanized farms, the cost of seedlings as charged by the companies was considered, where as for partially mechanized farms the seed rate as followed by the farmers was accounted. Manures and fertilizers: The data of farm yard manure produced at farm level or neighboring farmers were valued at prevailing market price and chemical fertilizer of N, P, K was valued at actual price paid by the farmers. Machineries: The actual hire charges payable per hour or per hectare, whether owned or hired were taken as the cost of machineries for various operations of rice cultivation. Human Labour: The total number of male and female labourers measured interms of number of man days or women days. The labour cost of family labour and hired labour was converted to common physical unit (1 man day = 6 working hours). The family labour was considered separately and added to the hired labour to calculate the total labour requirement. Plant Protection Chemicals: It includes both insecticide and fungicide used for rice cultivation. The plant protection cost collected from farmers at actual price paid in the market. Interest on Working capital: Components included in working capital were cost of human labour, machine labour, manures and fertilizers, irrigation, plant protection materials and weeding. Interest on working capital was computed at the rate of 12.5 per cent per annum, the rate followed in the cost of cultivation scheme. Interest on Fixed capital: Interest rate for fixed capital was calculated at twelve per cent per annum, which is based on the rate of interest charged by nationalized banks for term-lending. Depreciation on Fixed capital (other than land): Depreciation on fixed capital was calculated at five per cent for farm machinery and ten per cent for irrigation structures. Rental 43

57 value of land is not accounted since it is very high in the study area and all the farms selected were owner operated. Yield: The total yield of rice obtained by farmers was expressed in terms of quintal per hectare. Gross Return: Per hectare gross returns was calculated based on the market prices for the produce in rupees. The returns from bye products were taken as reported by the respondents. Net farm income: It was calculated by taking into account gross returns subtracting the variable costs. This concept is used in the functional analysis so as to measure the impact of mechanization on cost reduction. Cost of Production per quintal of output: Obtained by dividing the total cost per hectare by the yield per hectare (Rs per Quintal) Production Function One of the objectives of the study is to evaluate resource use efficiency of farms and to analyze the differences, if any, between four categories of farms with respect to production efficiency to understand the impact of farm mechanization on production. For this purpose of analysis, Cobb - Douglas type of production function is the choice because it would show productivity of resources and returns to scale. When the production function is estimated by ordinary least squares method (OLS hereafter) and evaluated at the mean level, it would give the average production and not maximum production. To overcome this difficulty, the concept of frontier may be meaningfully applied. The distance by which the actual production point (level) lies below the production frontier implied by maximum production is considered as production inefficiency. This involves three components. They are technical, allocative and scale efficiencies. The present study is confined to overall production efficiency and not its components because only whole farm production function was estimated where aggregated production is measured in value terms, assumed at constant product prices Empirical Model In the present study, Cobb-Douglas type of production function and stochastic frontier production functions are employed to study the technical efficiency among rice farmers. 44

58 Ln( Y) Where Y=net farm income (Rs) 0 n Ln X i i i v u X 1=farm size (ha) X 2 =seeds (Rs) X 3 =manures and Fertilizers (Rs) X 4 =machineries (Rs) X 5 =human labour (Rs) X 6 =plant protection chemicals (Rs) Frontier Production Function The maximum feasible yield function is defined as one that corresponds to the best practiced technique among the given producers. The production function showing such maximum yield may be estimated empirically by means of frontier production which is defined by (Aigner et.al1977and Battese and Corra 1977) as follows y ) Where y is an (nx1) vector of observed output, w f ( x)exp( w) f ( x e (1) i i x is an (nxk) matrix of observed inputs and exp( w) e w is the error term. Suppose that a farm is observed at a production plan ( X 0,Y 0 ), then the plan is said to 0 0 technically efficient if Y f ( X ), and inefficient if 0 0 Y < f ( X ).Then Maximum feasible yield is defined as Y max( Y i, Hi) where Hi is the state of technology available for the i th farm. This maximum feasible yield is feasible for all, but realized by at least some sample farms. These farms are taken as the reference farms to define maximum feasible production. The production function Y f X, H ) in which Yi is the yield per hectare of the i th farm i i( ij i and X ij is per hectare level of j th input vector shows the actual production of the i th farm. The difference between the two is a measure of inefficiency in production and implies that there exists hope for those who have not realized maximum feasible yield, to raise production with given technology, bridging the gap in technology adoption Stochastic Frontier Production Function The major limitation of the frontier production function is its assumption of deterministic relationship which ignores the very real possibility that the farm s performance may be affected by factors entirely outside its control and factors under its control, the former is the collective effect of exogenous shocks both favorable and unfavorable and the latter is due to inefficiency of technology. Therefore the two sources of errors need to be separated to know the real effect. This is the idea behind the Stochastic Frontier Production Function and is referred as SFPF hereinafter. 45

59 Consider the Cobb Douglas function defined as Y i f w ( X ) e A (2) 5 j wi X ij e j 1 On logarithmic transformation it would be y i 5 x w (3) 0 j 1 j ij i Where lower case letters represent the log values of the corresponding variables in (2) with ln 0 A. disturbance Aigner et al., (1977) divided disturbance term v i and one sided real efficiency disturbance function based on the error term w u v for all i (4) i i i wi into two components a stochastic ui and construct a joint density They referred the model with this error specification as a stochastic frontier, since non positive component of the disturbance represents the short fall of the actual output from the frontier, while the frontier contains a normal disturbance term and is therefore known as stochastic. This specification avoids statistical difficulties as discussed by Greene (1980), that are encountered in the estimation of full frontiers ie., the presence of purely non positive error term. The model specified in (1) y f ( x)exp( w) f ( x) e y v u f ( x)exp( v u) f ( x) e f ( x) e w v e u (5) Here the systematic component, v, permit random variation of the frontier across the farms comprising the effect of statistical noise and random shocks outside the control of farms and the one sided component, u, captures the effect of inefficiency relative to the stochastic frontier. When u=o farm uses technology efficiently and maximum feasible yield is feasible also when u<0 there is technical inefficiency. Efficient use of technology and inefficient use technology is measured in interval (0, 1).It is equal to one on the production frontier (when u=0) and less than one beneath the frontier (when u<0).the stochastic frontier is v f ( x) e which cause the placement of the deterministic kernel f(x) to vary across the farms. This ensures that all observations to lie beneath the stochastic frontier. Within this frame work a measure of technical efficiency is given by e u y v f ( x) e (6) 46

60 The above definition of technical efficiency means that farms have their own unique potential production frontiers for a given technology and then frontiers may differ from one another based on socio-economic and physical environments. It is sensible to compare each farm s performance with its own potential maximum feasible performance rather than with some common notation of performance. Further Stochastic specification implies that frontier may change over time. Technical efficiency of farms is given by TE j n j y jt 1 yˆ (7) jt Direct estimates of the SFPF may be either obtained by Maximum likelihood estimation procedure (MLE hereafter) or using freely downloadable software of frontier 4.1 and by corrected ordinary least squares (COLS hereafter) method. The COLS estimates are easier to compute than MLE and any other methods although they are asymptotically less efficient. The estimation of COLS is as follows: Corrected Ordinary Least Squares The Corrected Ordinary Least Squares (COLS) estimators are similar to the estimators suggested by Richmond (1974) in the context of pure frontier. Begin with OLS, 1, estimator B X X X Y consistent. Its covariance matrix 2, X X 1, except for the constant term the OLS estimators are unbiased and 2 where is the variance of the error term. The bias in the constant term is 2 1 The variance 2 2 u and v can be 2, 2, 2 estimated consistently and are given by u 3 and v 2 v 3 2 Where, and are second and third moments of OLS residuals. The constant term can be corrected, 2 3 by adding to OLS estimated constant term the negative of the estimated bias. 2 1 To recapitulate the COLS estimator of all components of B except for the first term is the same as that of the OLS estimates. is N (0, Further u is assumed to be non-positive and has a truncated normal distribution and v 2 ) ) and the technique using the frontier 4.1 software. Alternatively to the above assumption of arbitrary but within the feasible range values of γ while using MLE technique. 47

61 The iterative procedure was suggested by Farrell 1957 which is a programming technique and is modified to probabilistic programming technique Probabilistic Frontier Programming Production Function The Cobb-Douglas function y i 0 5 j 1 x j ij w i, more generally may be rewritten as 5 y x w (8) where x i 1for all i and i=1, 2, 3,... n. i j 0 j ij i 0 Farrell used programming technique to estimate production function, an envelope of the production observations in input-output space. Farrell s method possesses limitations of (i) non considering non-constant return to scale and (ii) great sensitivity to extreme observations, possibly owing to measurement errors. In fact, extreme observations are used to estimate the frontier. To overcome the limitations inherent in Farrell s method, Aigner and Chu (1968) devised a method of estimating a frontier production function by constraining the error term to be non- positive. Assuming the maximum feasible yield (output) is i 5 y x (9) and each farm s output may be represented by j 0 j ij 5 yi j xij ui, where u i < (10) Here j 0 ui is the difference between yield obtained by each farmer and yield estimated by fitting the function to represent maximum feasible yield that ui can only be either negative or zero. The negative value of ui will vary among farms depending on their technical efficiency, according to how close they are to the best practised technique. Even without specifying the probability distribution of ui a function showing maximum feasible yield may be estimated by linear or quadratic programming technique. This technique minimizes the sum of absolute differences of sum of squared differences respectively under the constrained that all the difference be either negative or zero. The resulting function is either frontier or anti frontier function to an efficient frontier equation (8) should be estimated such that 5 j 0 x yˆ y an infinite number of β s will j ij i i satisfy the equation. To force the equation to lie as closely as possible to the observed set of points, minimizing constraints must be placed on some function of the resulting error terms. n u i i 1 Minimize 2 would be convenient for making comparisons with average functions. The quadratic constraints accentuates extreme observations. The alternate is to minimize the 48

62 n u i i 1 linear sum of errors. Minimize the extreme observations are not unduly weighted in this way. By setting ui 0 the equation can be written as 5 j 0 x u y then the estimation j ij i i n u i i 1 technique is to minimize subject to 5 j 0 x y and β s 0 in order to solve by linear j ij i n u i i 1 programming must be expressed as a linear function of ˆ i and X ij the shown we n u i i 1 sum the equations over all I and solve for n i 1 5 n n n 5 n n j xij ui yi Rewriting this equation as j xij ui yi j 0 i 1 i 1 i 1 j 0 i 1 i 1 However such a programming technique does not account for statistical errors. This was pointed out by Timmer who provided a simple technique to deal with these errors to some extent. He deleted a percentage of errors assuming that they are affected by statistical errors and estimated the frontier functions using the remaining observations by linear or quadratic programming techniques. Thus he gave probabilistic approach to the deterministic approach used by Aigner and Chu. But the selection of the certain percentage of observation was lack of economic and statistical justification. To overcome this, Aigner and Chu expressed the equation (10) in probabilitistic form P 5 i 0 x j ij y j p for all i=1,2,3..n where p is the specified probability within above statement holds. This approach consists of estimating the frontier using all observations and re-estimating the frontier after discarding 100 per cent efficient farms until predetermined level of p is obtained Logistic Regression Model Several studies have investigated the various socio-economic, cultural, institutional and political factors on the willingness of the farmers to use new technologies. The logit model rather than the linear regression model has been used as the dependent variable, i.e, the index of farm machineries is a binary of dummy variable. The variable is denoted by FM and it takes the value 1 for farms that have adopted mechanization to a certain minimum level (50 per cent of the maximum achievable scores) and 0 for those who have not done so. In most of the adoption studies, the dependent variable takes value between 0 and one and the models used 49

63 were multivariate logistic models. According to Adeogen et., al (2008), the logistic model which is based on cumulative logistic probability function was developed to analyse the adoption characteristics of farmers to farm mechanization because the responses recorded were discrete. In the empirical specification of the logistic model, the independent variables included are as follows: X 1 -Age of the farmer (years) X 2 - Farming Experience (years) X 3 - Education level of the farmer X 4 - Family size (Nos) X 5 -Access to credit {1 = yes and 0 otherwise} 3.6 Energy Use Nowadays agricultural sector has become more energy intensive in order to supply more food to the increasing population and provide sufficient and adequate nutrition. However, considering limited natural resources and the impact of using different energy resources on environment and human health, it is substantial to investigate energy use pattern in Agriculture. Modernization of these operations increases the energy consumption of agricultural production. Many experimental works have been conducted on energy use in Agriculture such as Wheat, Maize, Sugar beet, Grapes, Cotton and Apple. Energy requirements in agriculture are divided into two groups, direct and indirect. Direct energy includes human labour, diesel and water for irrigation and the indirect energy includes seeds, fertilizers, manures chemical and machinery. The energy output of these systems includes main yields. The economic inputs of these systems include costs of human labour, seeds and fertilizers, hired machinery, fixed costs and agricultural machinery. The economic output of these systems includes main yields and the economic analysis include ratio of total income to total expenses. Mittal et al., (1985) have given the equivalents for direct and indirect sources of energy as follows. 50

64 Particulars A. Inputs 1.Human Labour a. Adult man Man Hour 1.96 b. Woman Woman - Hour 1.57 c. Child Child Hour Animals Energy Coefficients of Inputs and Outputs Equivalent Units Energy Remarks (Mega Joules) 1 adult woman = 0.8 adult man 1 child = 0.5 adult man Bullocks : Large Pair Hour Body weight above 450 Kg Medium Pair- Hour Body weight above Kg Small Pair- Hour 8.07 Body weight less than 350 Kg 3. Diesel Litre It includes the cost of lubricant 4. Petrol Litre It includes the cost of lubricant 5. Electricity KW- Hour Machinery Distribute the weight of the a. Electric Motor Kg machinery equally over the total b. Prime movers Kg life span of the machinery (hours). (other than electric motor) c. Farm machinery Kg Find the use of machinery (hours) for the particular operation in a crop. 7.Chemical fertilizers Estimate the quantity of nitrogen, I. Nitrogenous Kg p205 and k20 in the chemical II. P2O5 Kg 11.1 fertilizer. Then compute the amount of energy input from III. K2O Kg 6.7 chemical fertilizer 8. Farm Yard Kg (Dry Mass) 0.3 Manure 9. Chemicals Chemicals requiring dilution at the I. Superior chemicals Kg 120 time of application. II. Zinc sulphate Kg 20.9 III. Inferior chemicals Kg 10.0 DDT, gypsum or any other chemicals not requiring dilution at the time of application. 10. Seed Output of crop production system and not processed. Same as that of output of crop production system. 51

65 B. Output 1. Main product a. Cereal crops, such as wheat, maize and Paddy (Not shelled rice) Kg (Dry Mass 14.7 The main output is grains b. Pulses Kg (Dry Mass) 14.7 The main output is grains 2. By Product a. Straw, vines etc., Kg (Dry Mass) 12.5 b. Stalks, vines etc., Kg (Dry Mass) Mechanization Index Mechanization index has been calculated as the machine and fuel energy divided by the sum of fuel and machine energy as well as animal and manpower energy symbolized as: MI E /( E E E ) a where, MI: Mechanization Index, E d : (sum of) Machinery and Fuel energy, E h : Human Energy E a : Animal energy. d d h 52

66 CHAPTER IV DESCRIPTION OF THE STUDY AREA A knowledge about the economic, social and physical characteristics of the study area would help the researcher to have a better understanding of the problem per se and interpretation of the results of the study. An attempt has been made in this chapter to present the various characteristics of the study area regarding the geographical features, climate and rainfall, demographic details, land use pattern, cropping pattern, sources of irrigation and infrastructural facilities of the three districts of the Cauvery delta zone. Geographical Features The geographical features like location, climate and rainfall, soil type, land use pattern, operational holdings, irrigation, cropping pattern, demography and the availability of farm machineries are needed to have an idea about the study area and the same are given below for the three districts of Cauvery Delta Zone (CDZ). 4.1 Thanjavur District Location Thanjavur district is located between and of Northern latitude and of the eastern longitude. It has an area of Sq Km. The district is bounded on the Northwest by Coleroon river which separates it from Trichy and Cuddalore districts. On the North and North-east, it is bounded by Nagapattinam district and palk strait and on the west by Pudukkottai and Trichy districts. The district consists of nine taluks, namely Thanjavur, Thiruvaiyaru, Orathanad, Pattukkottai, Peravurani, Kumbakonam, Papanasam, Valangaiman and Thiruvaidaimarudur. The district is located in the flat land of Cauvery delta with low gradient of only one meter drop for a distance of every Km Climate and Rainfall The season wise distribution of rainfall for the year 2012 is furnished in Table

67 SW Monsoon (June Sep) Table 4.1: Distribution of Rainfall (mm) NE Monsoon (Oct-Dec) Winter (Jan-Feb) Hot Weather (March May) Whole Year (Jan-Dec) Actual Normal Actual Normal Actual Normal Actual Normal Actual Normal (43.28) (49.78) (46.34) (39.66) (1.91) (3.14) (8.47) (7.42) (100.00) (100.00) Source: Office of the Assistant Director of Statistics, Thanjavur. The Thanjavur district has a humid tropical climate. The mean maximum monthly temperature of the district varies from C in February to 40 0 C in May. The mean minimum monthly temperature varies from C to C. The district is subjected to the influence of North-East monsoon from October to December. The rainfall decreases with increasing distance from the coast. The behavior of North-East monsoon has always been erratic in the recent past. Excess rainfall results in drought and the distribution of rainfall was uneven. The normal rainfall of the district is mm. The district is in receipt of major rainfall during southwest monsoon period accounting for about 50 per cent of the annual rainfall Soil Type The soils of Thanjavur district are classified under 23 series. The soils of the delta are generally alluvial varying in texture from heavy clay to light sandy type. The soil type of the old delta covering Cauvery and Vennar systems are clayey in texture with restricted drainage capacity, due to low infiltration and percolation rates. The uplands of the district have redferrugenous soil Land Use Pattern A study on the land utilization pattern in the study area would indicate the scope, if any, for better utilization of land and the land use statistics is furnished in Table 4.2. The total geographical area of Thanjavur districts is 3,60,160 Hectares, of which per cent is the total cropped area. The area under current fallows, non-agricultural use and other fallows constitute 1.31, and 8.66 per cent respectively. The total cropped area is 2,19,133 hectare. 54

68 Table 4.2 Land Utilization Pattern Sl. No Particulars Area (Ha) Percentage 1. Total Geographical Area 3,60, Forests 4, Barren and Uncultivable Land 2, Land put to Non agricultural use 78, Culturable wastes 11, Permanent pastures and other grazing Land 1, Land under miscellaneous crops and groves 6, Current fallows 4, Other fallow lands 31, Net area sown 1,27, Area sown more than once 91, Total cropped area 2,19, Source: G Returns register of the Assistant Director of Statistics, Thanjavur Land Holding Pattern The land holding pattern of Thanjavur district is furnished in Table 4.3. It could be observed from the table that about 90 per cent of the total farmers were small and marginal and 10 per cent constitute the medium and large category. The small and marginal farmers operate per cent of the total area cultivated while the medium and large farmers operate per cent of the cultivated area. Table 4.3: Land Holding Pattern of Farmers Sl. No Category Numbers Percentage Area (ha) Percentage 1. Marginal (< 1 ha) 2,13, , Small (1-2 ha) 38, , Semi Medium 18, , Medium(4-10 ha) 6, , Large (> 10 ha) , Total 2,77, Source: District Statistical Hand Book, Irrigation The sourcewise area irrigated in Thanjavur district is furnished in Table 4.4. The net area irrigated by all sources is 2,02,163 hectares. Over per cent of the irrigated area depends on Government canals. Tanks irrigate about 4.52 per cent of the net area irrigated. The rivers Cauvery and Vennar are irrigating old delta and the Cauvery Mettur project canal is feeding the new delta area. 55

69 Table: 4.4 Source wise Area Irrigated Source Number Area (Ha) Percentage Tube wells 19,496 22, Bore wells 9,153 18, Dug cum Bore wells 906 2, Filter points 2,769 2, Open wells 2,260 4, Total wells 34,584 49, a. Electric Motor 29,515 b. Diesel engine 4,093 Canals 25 (638 Km) 1,20, Gross irrigated area 1,70,266 Percentage to gross cropped Source: Season and Crop Report Cropping Pattern Changes in CDZ The performance of rice and the changes in area during the last four decades were analyzed, the data being collected from the secondary sources. The three year moving average time series smoothening was used to minimize the seasonal variation. The results are presented in Table 4.5. Table 4.5: Cropping Pattern Changes in CDZ Crops Triennium Ending Area (ha) Percent Triennium Ending Area (ha) Percent Triennium Ending Area (ha) Percent Triennium Ending Area (ha) Percent Paddy 6,21, ,75, ,16, ,55, Maize 2, , Pulses 1,52, ,92, ,55, ,25, Groundnut 43, , , , Gingelly 6, , , , Sugercane 6, , , , Cotton 1, , , , Coconut 19, , , , Others 42, ,11, , , GCA 8,96, ,90, ,67, ,25,

70 The data has revealed that the area under paddy was declining gradually from to The area under paddy which stood at 6.21 lakh hactare in had declined to 4.55 lakh hactares during the last four decades. i.e the percentage of paddy which accounted for per cent of the gross cropped area in had reduced to per cent in Conversely, the area under sugarcane had increased by four times and that the area under coconut had doubled during the period, indicating clearly that the paddy area had been diverted towards sugarcane and coconut. The reasons must be increasing labour scarcity, uncertainty of canal water and uneconomic cultivation of rice Performance of Rice in CDZ in Different Decades The area, production and productivity of rice were collected from the secondary sources and the performance analyzed through compound growth rates and the results are presented in Table 4.6. Table 4.6 Annual Compound Growth Rates Period Area Production Yield to to to to All Period It was observed that the productivity of rice had shown increasing trend up to and recorded a positive annual growth rate of 5.13 per cent. From to , production and productivity witnessed negative growth rate. Since all the three parameters had recorded a negative growth rate. The overall analysis had indicated that the area, production and productivity of rice in CDZ had been declining in the last two decades Cropping Pattern (Thanjavur District) The cropping pattern of Thanjavur district for the year is furnished in Table 4.7. Thanjavur district has the gross cropped area of 2,19,133 hectares of which rice accounted for per cent. The other major crops cultivated in the district are Pulses, Sugarcane, Groundnut, Maize and Gingelly. 57

71 Table 4.7: Cropping Pattern Sl. No Crop Area (ha) Percentage 1. Rice Kuruvai 24, Rice Samba 1,27, Rice Thaladi 10, Rice Summer 4,122 Total Rice 1,66, Black Gram 28, Green Gram 5,151 Total Pulses 33, Cotton Sugar cane 5, Gingelly 4, Groundnut 7, Maize 1, Total 2,19, Source: Office of the Joint Director of Agriculture, Thanjavur Demography The demographic details of Thanjavur district is furnished in Table 4.8. As per 2011 census, the total population of the district was 24,02,781. Rural population accounted for per cent and urbanites for per cent of the total population of the district. Literates constituted per cent of the total population. The total workers constituted per cent of the total population of which cultivators accounted for 6.10 per cent. Agricultural labourers constituted per cent of which 8.12 per cent were males and 4.28 per cent were females. 58

72 Table 4.8 Demography of Thanjavur District 2011 Particulars Total Percentage Total Households 5,44, Rural 3,38, Urban 2,05, Total Population 24,02, Rural 15,57, Urban 8,45, Total Male 11,83, Rural 7,55, Urban 4,27, Total Female 12,19, Rural 8,18, Urban 4,01, Sex Ratio 1,031 - Density Total Literates 18,02, Total Workers 1,02, Total Main Workers 86, Total Main Cultivators 1,46, Total Main Agricultural Labours 2,98, Male 1,95, Female 1,03, Source: Directorate of Census operations, Tamil Nadu Farm Machineries The details on the availability of farm machineries namely Tractors, Power Tillers, transplanters, Combine Harvesters, Power Weeders, Power Sprayers, etc, with the Government and Private sources are furnished in Table 4.9, along with the projected machinery requirement for The table clearly reveals that the machinery population has been steadily increasing over the years. It is evident there exists a substantial gap in the demand and supply of the major machineries required for rice cultivation. 59

73 Table 4.9: Details on Farm Machinery Population Year/ Machinery * Tractor with Rotavator Power Tiller Transplanter Combine Harvester Weeder Laser Leveler Power Sprayer Fully Automated seeder Source: Office of the AEE, Thanjavur * Requirement based on area under rice Thiruvarur District Location Thiruvarur district is located between and of the Northern latitude and between and of the Eastern longitude and 10 meters above MSL. The geographical area of the district is Sq. Km. The district is bounded by palk strait in the south, Thanjavur in the west, Nagapattinam district in the east and parts of Nagappattinam and Thanjavur districts in the North. The district consists of seven taluks namely, Thiruvarur, Mannargudi, Needamangalam, Thiruthuraipoondi, Nannilam, Kodavasal and Valangaiman Climate and Rainfall The district has a humid tropical climate with the mean maximum temperature of 39.7 o C and the minimum temperature of 22.6 o C. The district is in receipt of rainfall during South West and North East monsoons with the average rainfall of 303.8mm and 665.4mm in the respective seasons. During rainy seasons the water from Trichy and Thanjavur districts is drained into Thiruvarur district that causes flooding and heavy damages to the crops. The season wise rainfall data is given in Table

74 SW Monsoon (June Sep) Table 4.10: Distribution of Rainfall (mm) NE Monsoon (Oct-Dec) Winter (Jan-Feb) Hot Weather (March May) Whole Year (Jan-Dec) Actual Normal Actual Normal Actual Normal Actual Normal Actual Normal (29.72) (26.82) (62.45) (59.23) 55.5 (3.07) 60.1 (5.33) 83.5 (4.66) 97.7 (8.62) (100.00) (100.00) Source: Office of the Assistant Director of Statistics, Thiruvarur Soil Type Thiruvarur district is made up of tertiary and alluvial deposits. Similar to Thanjavur district the soils of the Thiruvarur district are generally alluvial varying in texture from heavy clay to light sandy type Land Use Pattern The land use pattern of Thiruvarur district is furnished in Table Table 4.11: Land Use Pattern Sl. No Details Area (ha) Percentage 1. Total geographical area 2,07, Forests 2, Barren and Uncultivable Land Land put to non Agricultural use 37, Cultivatable wastes 1, Permanent pastures and other grazing land Land under miscellaneous crops and groves not included in net area sown 2, Current fallows 2, Other Fallow Lands 17, Net area sown 1,43, Area sown more than once 53, Total cropped area 1,96, Source: G Returns register of the Assistant Director of Statistics, Thiruvarur. It could be seen from the table that the total geographical area of the district is 2,09,709 Ha. The share of the net area sown is per cent. The area under current fallows, non-agricultural use and other fallows constitute 1.02 per cent, per cent and 8.3 per cent, respectively. 61

75 4.2.5 Land Holding Pattern The land holding pattern of the farmers of Thiruvarur district is furnished in Table The data showed that about 90 per cent of the farmers constituted the small and marginal category, and 10 per cent of the farmers belong to medium and large size group. The small and marginal farmers operate per cent of the total area cultivated while the medium and large farmers operate per cent of the total cultivated area. Table 4.12: Land Holding Pattern of Farmers Sl. No Category Numbers Percentage Area (Ha) Percentage 1. Marginal (< 1 ha) 1,13, , Small (1-2 ha) 27, , Semi Medium (2-4 ha) 12, , Medium (4-10 ha) 3, , Large (> 10 ha) , Total 1,57, Source: District Statistical Hand Book ( ) Irrigation The sourcewise area under irrigation is given in Table Thiruvarur district has a net irrigated area of 1,56,338 hectares of which, per cent of the area has been irrigated by the Government canal and the remaining area has been irrigated by private tube wells (9.06 per cent). Table 4.13: Source Wise Area Irrigated Source Number Area (ha) Tube wells 10,461 18,418 Bore wells 4,912 12,730 Dug cum Bore wells 843 2,410 Filter points 1,754 3,430 Open wells 2,137 4,865 Total wells 20,107 41,853 a. Electric Motor 16,478 b. Diesel engine 2,155 Canals 13 (612Km) 1,06,863 Gross irrigated area 1,48,716 Percentage to gross cropped area Source: Season and Crop Report (

76 4.2.7 Cropping Pattern The cropping pattern of Thiruvarur district is furnished in Table Thiruvarur district has the gross cropped area of the 1,96,597 hectares of which paddy accounted for per cent. The other crops cultivated in the district were pulses, Gingelly, Groundnut, Sugarcane and Cotton. Table 4.14: Cropping Pattern Sl. No Crop Area (Ha) Percentage 1. Rice Kuruvai 35, Rice Samba 68, Rice Thaladi 27, Rice Summer 1,500 Total Rice 1,33, Black Gram 51, Green Gram 19,432 Total Pulses 70, Cotton 1, Sugar cane 2, Gingelly 4, Groundnut 3, Total 1,96, Source: Office of the Joint Director of Agriculture, Thiruvarur. 63

77 4.2.8 Demography The demographic details of Thiruvarur district is furnished in Table As per 2011 census, the total population of the district was 12,68,094. Rural population accounted for per cent and urbanites for per cent of the total population of the district. Literates constituted per cent of the total population. The total working population in the district is per cent. Cultivators accounted for 4.89 per cent of the total population and per cent of the workers depend on agriculture. Table 4.15: Demography of Thiruvarur District 2011 Particulars Total Percentage Total Households 2,98, Rural 2,35, Urban 63, Total Population 12,68, Rural 9,65, Urban 3,02, Total Male 6,27, Rural 4,80, Urban 1,46, Total Female 6,40, Rural 5,03, Urban 1,37, Sex Ratio 1,020 - Density Total Literates 9,60, Total Workers 5,52, Total Main Workers 4,62, Total Main Cultivators 62, Total Main Agricultural Labours 1,94, Male Female Source: Directorate of Census operations, Tamil Nadu. 64

78 4.2.9 Farm Machineries The details on the availability of farm machineries available with Government and Private Sources are furnished in Table 4.16, along with the projected machinery requirement for Akin to Thanjavur district, the machinery population in Thiruvarur district has been increasing over the years and it is evident there exists a substantial gap in the demand and supply of the major machineries required for rice cultivation. Table 4.16: Details on Farm Machinery Population Year/ Machinery * Tractor with Rotavator Power Tiller Transplanter Combine Harvester Weeder Laser Leveler Power Sprayer Fully Automated seeder Source: Office of the AEE, Thiruvarur * Requirement based on area under rice Nagapattinam District Location Nagapattinam district was carved out from erstwhile Thanjavur district in 1991, also called as east Thanjavur, the paddy granary of South India. It is a coastal district situated in the state of Tamil Nadu with a long coastline of 187 km. It is a peninsular deltaic district surrounded by Bay of Bengal in the east, Palk Strait in the south, Thiruvarur and Thanjavur districts in the west and the Cuddalore district in the north. It lies on the shores of the Bay of Bengal between north latitude 10 o 10 and 11 o 20 east longitude 79 o 50. The geographical area is km 2 and has a population of 14,87,005. The district headquarters is Nagapattinam. The district has two main revenue divisions namely Mayiladuthurai and Nagapattinam for administrative purpose. It comprises of seven taluks namely Keelavellur, Mayiladuthurai, Nagapattinam, Sirkazhi, Tharangambadi, Tirukuvalai and Vedaranyam with 11 development blocks and 11 panchayat unions, covering 433 village panchayats. 65

79 4.3.2 Climate and Rainfall The climate in the district ranges from semi-arid to sub humid. It experiences hot dry summer from March to May when temperature is fairly high going up to an average of 35 o c and cold weather in December and January when a minimum of 24.6 o c has been recorded. In general, the district has high relative humidity during October to March, when winds blow from north easterly and easterly directions. Average annual rainfall of the district is mm. The season wise distribution of rainfall of the district over years is presented in Table The district receives bimodal rainfall. The district receives bimodal rainfall. The southwest and northeast monsoon account for per cent and per cent respectively of the average annual rainfall in the region. Winter rains account for 6.5 per cent while summer showers account for 6.1 per cent of the annual rainfall in the district. In Nagapattinam district, the distribution of the rainfall for the study period did not show major variations when compared to the average rainfall for the district. SW Monsoon (June Sep) Table 4.17: Distribution of Rainfall (mm) NE Monsoon (Oct-Dec) Winter (Jan-Feb) Hot Weather (March May) Whole Year (Jan-Dec) Actual Normal Actual Normal Actual Normal Actual Normal Actual Normal (24.9) (20.6) (67.1) (66.8) 45.3 (2.9) 85.7 (6.5) 78.9 (5.1) 80.5 (6.1) (100.00) (100.00) Source: Office of the Assistant Director of Statistics, Nagapattinam. (Figures in parentheses indicate percentage) Soil Type Sandy coastal alluvium is the predominant soil type of the district Land Use Pattern A knowledge on the allocation of land among various uses would throw light on the potential for developing farming and allied economic activities. Details on the land utilization pattern in Nagapattinam district is presented in Table

80 Table 4.18: Land Use Pattern Sl. No Details Area (ha) Percentage 1. Total geographical area 2,71, Forests 4, Barren and Uncultivable Land 33, Land put to non Agricultural use 47, Cultivatable wastes 3, Permanent pastures and other grazing land Land under miscellaneous crops and groves not included in net area sown 10, Current fallows 18, Other Fallow Lands 20, Net area sown 1,31, Area sown more than once 52, Total cropped area 1,83, Source: G Returns register of the Assistant Director of Statistics, Nagapattinam. The geographical area of Nagapattinam district is 2,71,583 ha. It could be discerned from the table that gross cropped area accounted for 67.7 per cent of the total geographical area in the district. The net sown area occupied per cent and land put to nonagricultural uses for per cent. Barren and uncultivatable land, total fallows and cultivable waste accounted for 12.31, and 1.41 per cent respectively of geographical area of the district. The land under miscellaneous tree crops and forest area accounted for 3.91 and 1.71 per cent respectively of the total geographical area of the district Land Holding Pattern The details on the distribution of operational land holdings of Nagapattinam district are furnished in Table It could be observed from the table that about 60 per cent of the cultivated area is being owned by about 90 per cent of the small and marginal farmers and 40 per cent of the area being operated by the medium and large size group of farmers. It shows there exists wide disparity in the ownership and the operatorship of the farm holdings. 67

81 Table 4.19: Land Holding Pattern of Farmers Sl. No Category Numbers Percentage Area (Ha) Percentage 1. Marginal (< 1 ha) 1,39, , Small (1-2 ha) 27, , Semi Medium (2-4 ha) 11, , Medium (4-10 ha) 3, , Large (> 10 ha) , Total Source: District Statistical Office, Nagapattinam Irrigation The major determination of the production performance of agriculture is the availability and intensity of irrigation. Nagapattinam district is predominantly irrigated by Cauvery and Vennar river basin system and are called the old delta region. River Coleroon also acts as an irrigation source for this district. Canal water is the major source of irrigation. The water supply is dependent upon the water release of Cauvery. Besides, there are six streamlets and all of them depend on monsoon rains for their replenishment. The area irrigated and the major sources and their relative contribution to net irrigated area are shown in table Table 4.20: Source Wise Area Irrigated Source Number Area (Ha) Tube wells 8,430 10,215 Bore wells 6,435 13,168 Filter points 1,743 2,115 Open wells 2,118 2,736 Total wells 18,726 28,234 a. Electric Motor 12,321 b. Diesel engine 5,428 Canals 9 (548 Km) 1,23,906 Gross irrigated area - 1,52,140 Percentage to gross cropped area Source: Season and Crop Report ( ) 68

82 Nagapattinam district, as a whole, has nine canals and 18,726 wells in total. Canals form the principal source of irrigation accounting for per cent of the gross irrigated area indicating the predominance of canal irrigation in the district. Overall, the district has per cent of the gross cropped area being irrigated through canals and wells Cropping Pattern A study of the crop pattern would provide an idea of the decision behavior of the farmers of the crop-mix prevalent in the region. The distribution of area under major crops in Nagapattinam district is furnished in Table Table 4.21: Cropping Pattern Sl. No Crop Area (ha) Percentage 1. Rice Kuruvai 26, Rice Samba 79, Rice Thaladi 14, Rice Summer 1,385 Total Rice 1,21, Black Gram 51, Green Gram 28,761 Total Pulses 80, Cotton Sugar cane 3, Gingelly Groundnut 2, Maize Total 1,83, Source: Office of the Joint Director of Agriculture, Nagapattinam. Paddy was the dominant crop in Nagapattinam district accounting for about 66 per cent of the gross cropped area followed by pulses with per cent and sugarcane with 1.18 per cent. The oilseeds accounted for about 1.30 per cent of the gross cropped area. Maize crop is picking up in the district and presently cultivated in 24 hectares. 69

83 4.3.8 Demography The demographic details of Nagapattinam district is furnished in Table As per 2011 census, the total population of district was 16, 14,069 with a population density of 617 per Km 2. Rural population accounted for per cent and urbanites for per cent of the total population of the district. Literates constituted per cent of the total population. The total working population in the district is per cent. Cultivators accounted for 4.08 per cent of the total population and per cent of the workers depend on agriculture. Table 4.22: Demography Nagapattinam District 2011 Particulars Total Percentage Total Households 3,72, Rural 2,83, Urban 89, Total Population 16,14, Rural 12,23, Urban 3,90, Total Male 7,97, Rural 5,67, Urban 2,30, Total Female 8,16, Rural 6,09, Urban 2,07, Sex Ratio 1,025 - Density Total Literates 12,27, Total Workers 7,18, Total Main Workers 5,43, Total Main Cultivators 65, Total Main Agricultural Labours 2,20, Male 1,33, Female 87, Source: Directorate of Census operations, Tamil Nadu. 70

84 4.3.9 Farm Machineries The details on the availability of farm machineries with the Government and Private sources are furnished in Table 4.23, along with the projected machinery requirement for Table 4.23: Details on Farm Machinery Population Year/ Machinery * Tractor with Rotavator Power Tiller Transplanter Combine Harvester Weeder Laser Leveler Power Sprayer Fully Automated seeder Source: Office of the AEE, Nagapattinam * Requirement based on area under rice The data on availability and the requirement of major machineries for rice cultivation are very similar to the other two districts of CDZ and proportionate to the area under rice. In this district also the gap between the availability and requirement of the machineries is wide. 71

85 CHAPTER V RESULTS AND DISCUSSION The data collected through the survey were subjected to statistical analysis, taking into consideration the objectives of the study. The results of the analysis are presented and discussed under the following sections. 5.1 Profile characteristics of the sample farms 5.2 Cropping pattern of sample farms 5.3 Extent of Mechanization in Rice farming 5.4 Cost and Returns of Rice cultivation 5.5 Resources use efficiency and Technical efficiency 5.6 Energy use efficiency and Mechanization index 5.7 Constraints of Mechanization 5.1 Profile Characteristics of the Sample Farms A brief description of the sample farmers would provide the necessary setting for the discussion. Hence the demographic features, occupational status, farm size, cropping pattern, asset position and investment pattern were analyzed and the results are presented and discussed in this section Age Distribution of Head of the Sample Farmers The age of the head of the farm households is important since he or she is the decision maker in the family. Age generally correlates the mental maturity in farming and the decision making in the various operations of the farmers. The age distribution of the farm households in the sample is presented in Table

86 Table 5.1: Distribution of Sample Respondents across Different Age Groups Number of Farmers Sl. No Age group years Small farm Large Farms All Farms Percentage 1 Up to 30 Years Above Total Average The cross section of the age group clearly reveals that 78.0 per cent of the farmers were more than 40 years old and among them about 56.6 per cent of them fall under the age group of years. The average age was 46.85, and for small, large and all farms. The percentage of youngsters i.e upto 40 years constitute only 22.0 per cent of the total sample indicating that younger lot are gradually moving away from farming. This would affect the eagerness and awareness in the adoption of new technologies in agriculture Educational Status The level of education of the sample farmers is an important factor influencing the decision making behavior to a great extent. The details on the educational category of the sample households under different literacy levels are presented in Table 5.2. Sl. No Literacy Level Table 5.2: Educational Status of Sample Farms Number of farmers Small farms Large Farms All Farms Percentage 1. Illiterates Primary (Up to 5 th Std) Secondary (6-12 th Std) Collegiate (above 12 th Std) Total It could be seen from the table that about 90 per cent of the sample farmers were literates of which 80.0 per cent of them studied upto school level and about 10.0 per cent were degree holders. The literacy rate had improved a lot, may be due to several educational programmes. This factor may offset the disadvantage of age factor in the improvement of farm practices. 73

87 5.1.3 Experience of Farming The experience of farmers also plays a vital role alongside the age and education in the execution of the farm practices. The experience of the sample farmers is furnished in Table 5.3. Sl. No Experience (Years) Table 5.3: Experience of Sample Farmers Number of farmers Small farms Large Farms All Farms Percentage 1. Less than More than Total It could be observed from the table that majority of the farmers i.e per cent of the sample respondents had more than 20 years of experience of which 25 per cent of them had more than 30 years of experience. Roughly about 40.0 per cent of the sample respondents had less than 20 years of experience which is in conformity with the age group distribution of the farmers Family Composition of the Sample Respondents The family size and the number of members engaged in agriculture of the selected farms are furnished in Table 5.4. Sl. No Table 5.4: Family Size and Number of Members Engaged in Agriculture Number of Members engaged in Family Size Members Agriculture Size Small Large All Percentage Small Large All Percentage Farms farms 1. Up to Above Total Average It could be deduced from the table that about 80 per cent of the families are small sized one having family members upto 5 and per cent of the families are large one with more than five members over all. Conversely, as the family size increases the number of 74

88 persons engaged in agriculture decreases. This shows that more persons were found to move towards non-agricultural activities or professions. The data revealed that about 95 per cent of the members of small sized families are engaged in agriculture while only about five per cent of the large sized family members are sticking to agriculture, again in tune with the age group distribution of the sample farmers The Annual Income of Sample Farmers The sample households were post stratified in to three different groups based on the annual income from all the sources. Households with annual income of upto Rs.1,00,000 were categorized as low income group, households with annual income group between Rs. 1,00,00 and Rs. 2,00,000 as medium income group and those with annual income exceeding Rs. 2,00,000 as high income group. The income group wise distribution of sample farms are furnished in Table 5.5. Sl. No Table 5.5: Household Annual Income of Sample Farmers Number of farmers Annual income (Rs ) Small Large All Percentage 1. Upto Rs. 1,00, Rs.1,00,000 2,00, Above Rs. 2,00, Total The income pattern of the households clearly reveal that the selected farmers were reasonably better off. The income included both the agricultural and non-agricultural income and hence the income slab was reasonably higher since most of the farmers had sufficient share of off-farm and non-farm income. The average size of small farms was 1.77 ha and that of large farms was 4.36 ha which had a direct bearing on the income level of the farmers Size of Operational Holdings The size of farm is the structural parameter influencing the level and pattern of farm production. It would also decide the level and pattern of farm mechanization. The details on the size of holding are given in Table

89 Table 5.6: Size of Holding of Sample Farms Sl. No Number of farm Total area (ha) Average size Percentage 1. Small (up to 2 ha) Large (above 2 ha) Total The data collected were post-stratified based on regarding the size of holding. The farms with an area of upto two hactares were categorized as small farms and the farms with an area of more than 2 hactare were taken as large farms. Accordingly, out of the total sample size of 240, small farms were 130 and large farms were 110. The average size of small farms was 1.77 hactare and that of large farms was 4.36 hactare. The total area of operation of the sample farms was hactares, of which the small farms constituted per cent and the share of large farms was per cent Investment Pattern Land, buildings, livestock, machinery, equipments, tools and implements were the fixed assets owned by the sample farmers. The details on the average distribution of total farm investment per hactare are given the Table 5.7. Table 5.7: Average Investment Pattern in the Sample Farms (Rs Per Ha) Sl. Value per hactare (Rs) Assets No Small Large All Percentage 1 Land (Ha) Farm Buildings Live stock Irrigation structures (Borewell, Electric Motor, Diesel Engine) 5 Machineries (Tractor, Power Tiller, Transplanter, Weeder) Tools and Implements Total It could be evidenced from the table that the major investment by both small and large farms was on machineries. While the per hactare investment on the above capital item was Rs in small farms and Rs on large farms. The next in importance was investment on irrigation structures, namely borewells, electric motors and diesel engines which accounted for Rs per hactare in small farms and Rs in large farms. 76

90 Percentage wise, machineries and implements accounted for the major share of investment with per cent of the total investment. The second major investment was on irrigation structures with per cent for both the farms taken together. Livestock accounted for per cent while the investment on farm buildings was 9.02 per cent. 5.2 Cropping Pattern The cropping pattern in general would indicate the economic significance of different crops of the sample farms and hence the details on the same are furnished in Tables 5.8(a) and 5.8(b). It could be observed from the table that the cropping pattern in both the small and large farms was dominated by rice accounting for per cent of the gross cropped area followed by pulses (Rice fallow) with per cent. The other crops cultivated are ground nut, gingelly and maize, together aggregating to less than ten per cent of the gross cropped area of the sample farms. Out of the total cultivated area of hactare under paddy by the sample respondents, hactare has been cultivated by small farms and hactare by the large farms. This means that the large farms accounted for twice the area cultivated by the small farms under paddy. Seasonwise, the area under paddy accounted for per cent, per cent and per cent under Kuruvai (Season I), Samba (Season II) and Thaladi/ Summer (Season III) respectively, for all the selected farms. The cropping intensity worked out to 170 per cent for small farms, per cent for large farms and per cent, overall. Table 5.8 (a): Cropping Pattern of the Sample Farms Paddy Total Total Crop / Farms I II III paddy pulses Season Season Season Small NO.OF FARMS Large AREA (Ha) ALL FARMS (100.00) Small Large (25.72) (29.20) (10.90) (65.82)) (18.92) 77

91 Groundnut Table 5.8 (b): Cropping Pattern of the Sample Farms Gingelly Other crops Gross cropped Area Cropping Intensity Small Large All (3.38) (1.83) (2.37) Figures in Parentheses indicate percentage 5.3 Extent of Mechanization in CDZ The operation wise and size group/ level wise usage of machineries was analyzed through frequency distribution of farms and the same is presented in Tables 5.9(a) and 5.9(b). Table 5.9 (a): Extent of Mechanization in the Sample Farms (in percentage) Sl. No Operations 1. Preparatory cultivation Small Farms (years) Large Farms (years) > >10 3 (2) 2. Transplanting 51 (39) 3. Manures & Manuring 4. Weeding 5. Plant protection 7 (5) 36 (28) 53 (41) 28 (22) (42) - 29 (26) 42 (38) 37 (34) (2) 6. Irrigation - 18 (14) 7. Harvesting 21 (16) 8. Threshing and Cleaning 68 (52) (15) (100) 75 (58) 38 (29) 30 (23) 23 (18) 12 (9) (Figures in parentheses indicate percentage) 78 (60) 3 (2) 7 (6) - 7 (6) 12 (11) (100) - 10 (9) 18 (16) 78 (71) 18 (16) 68 (62) 18 (16) 82 (74) 17 (15) - 78

92 Table 5.9 (b): Extent of Mechanization in the Sample Farms (in percentage) Types of Farms Extent of Mechanization Types of Farms Extent of Mechanization Small farms partially All small farms mechanized Small farms fully All large farms mechanized Large farms partially All Partially mechanized mechanized farms Large farms fully mechanized All fully mechanized farms The frequency distribution of farms had revealed that 63 per cent of small farms and 72 per cent of large farms were using either tractor or power tiller for plouging and levelling operations for more than ten years. Overall, the preparatory cultivation works were mechanized by 93 per cent of small farms and 98 per cent of large farms. The major operations mechanized were plant protection and irrigation. While 100 per cent of the plant protection operation was carried out using power sprayers by both the size groups of farms, about 60 per cent of small farms and 75 per cent of large farms were using electric motors and diesel engines for lifting water for more than ten years. It could be revealed from that the table that irrigation has been mechanized to the extent of 97 per cent in small farms and 100 per cent in large farms. As regards transplanting operation, the transplanters have been introduced only five years back and became successful only in the last two to three years. About 40 per cent of small farms and 42 per cent of the large farms have resorted to mechanical means while the rest of the farmers were adopting manual transplanting. In the last five years, weeding has been undertaken by kono weeders and power weeders by 7 per cent of small farms and 26 per cent of large farms. However, farmers prefer to go for either manual weeding or chemical weeding due to drudgery or uneconomical means of mechanical weeding. Weeding is the only operation failing to take up through machines for want of appropriate machinery. Ever since the combine harvesters were introduced from farmers have resorted to mechanical harvesting. About 90 per cent of small farmers and 93 per cent of large farmers have been deploying combine harvesters for harvesting, threshing and cleaning. Before the introduction of combine harvester, farmers were harvesting the rice crop manually and used mechanical threshers for threshing and winnowing. 79

93 The overall extent of mechanization ranged from per cent for partially mechanized small farms to per cent for fully mechanized large farms. On an average, the extent of mechanization was per cent for small all small farms and per cent for all large farms. As regards level of mechanization, partially mechanized farms had the level of mechanization upto per cent and the fully mechanized farms had per cent of mechanization. 5.4 Cost and Returns The cost and returns structure of rice farming in the study area have been estimated and presented in Tables 5.10 to The results are so presented in such a way to know the comparative performance of different levels of mechanization and for different size group of farmers over the economics of rice cultivation Partially Mechanized Small and Large Farms Partially mechanized farms were those farms where all the operations except transplanting were mechanized. While comparing the partially mechanized farms, the total variable cost had worked out to Rs.46,174 per hactare for small farms and Rs.42,843 per hactare for large farms, proving that higher the farm size, lesser the cost of cultivation. While the cost of inputs remained more or less the same for both the farms, there was a significant difference in the cost structure towards the cost of three types of labour viz human, animal and machine labour. The cost of human labour was Rs.23,395 per hactare (40.51 per cent) for small farms and Rs.19,894 per hactare (35.93 per cent) for large farms, while the reverse in trend is true with regard to the use of machineries. The cost of machine labour was Rs.9,240 per hactare (16.0 per cent) for small farms and Rs.10,432 per hactare (18.84 per cent) for large farms. Similarly, the use of animal labour was also declining as the farm size increases which worked out to Rs.1,582 per hactare for small farms and Rs.769 per hactare for large farms. 80

94 Sl. No 1. Table 5.10: Cost of Cultivation / Production of Rice per ha Particulars Partially mechanized Small farms Large farms I. VARIABLE COST Human Labour 23,395 (40.51) 19,894 (35.93) 2. Animal Labour 1,582 (2.74) 769(1.38) 3. Machine Labour 9,240 (16.0) 10,432 (18.84) Total Labour 34,217 (59.26) (56.17) 4. Seeds/Seedlings 1,482 (2.56) 1,876 (3.38) 5. Manures and Fertilizer 5,182 (8.97) 4,335 (7.83) 6. Plant Protection 932 (1.61) 1,235(2.23) 7. Irrigation 1,645 (2.84) 1,782 (3.22) Working Capital 9,241 (16.0) 92,28(16.66) 8. Interest on Working 12.5 Per cent 2,716 (4.70) 2,520 (4.55) 9. Total variable cost 46,174 (99.97) 42,843 (77.39) II. FIXED COST 6,315 (10.93) 7,482 (13.51) SUB TOTAL I + II 52,439 (90.91) 50,325 (90.90) 10. Managerial 10 per cent 52,48(9.08) 5,032(9.09) III. TOTAL COST 57,737 (100.0) 55,357(100.0) 11. Main product (Quintal) Value of Main Product 58,840 59, Value of By Product 2,840 3, Gross Income 61,680 62, Net Income Over variable cost 15,506 19, Cost of Production per Quintal BCR The fixed cost structure between the two size of farms was also significant. While the share of fixed cost for the small farms was Rs.6,315 per hactare (10.93 per cent) the same for the large farms was Rs.7,482 per hactare (13.51 per cent). The total cost of cultivation had worked out to Rs. 57,737 per hactare for small farms and Rs. 55,357 per hactare for large farms. 81

95 Sl. No Table 5.11: Cost of Cultivation / Production of Rice per ha Particulars I. VARIABLE COST Human Labour Animal Labour Machine Labour Small Farms Fully mechanized 14,430 (25.57) 845 (1.5) 16,342 (28.95) Large Farms 7,565 (13.90) 534 (0.98) 19,396 (35.64) Total Labour 31,617 (56.02) 27,495 (50.53) 4. Seeds/Seedlings 1,235 (2.18) 1,185 (2.18) 5. Manures and Fertilizer 5,361 (9.5) 5,963 (10.95) 6. Plant Protection 1,140 (2.02) 1,365 (2.51) 7. Irrigation 1,431(2.53) 1,666(3.06) Working Capital 9,167 (16.24) 10,179 (18.70) 8. Interest on Working 2,544 (4.51) 2,354 (4.32) 9. Total variable cost 43,333 (76.78) 40,028 (73.56) II. FIXED COST 7,968 (14.11) 9,435 (17.34) SUB TOTAL I + II 51,301 (90.90) 49,463 (90.91) 10. Managerial 10 per cent 5,130 (9.09) 4,946 (9.09) III. TOTAL COST 56,431 (100.0) 54,409 (100.0) 11. Main product (Quintal) Value of Main Product 62,799 62, Value of By Product 2,615 2, Gross Income 65,414 64, Net Income Over variable cost 22,081 24, Cost of Production per Quintal BCR While the per hectare cost of production was able to be decreased with increased farm size and increased use of machineries, the impact of these factors had also reflected in the output of the crop. The yield obtained was Qtl per hactare for small farms and Qtl per hactare for large farms, resulting in a significant performance on net farm income and the cost of production per unit of output. The net income over variable cost was Rs.15,506 per hactare for small farms and Rs.19,530 per hactare for large farms and the cost of 82

96 production worked out to Rs.949 per Qtl and Rs.829 per Qtl for small and large farms respectively. The benefit cost ratio worked out to 1.35 for small farms and 1.49 for large farms. The results of the cost and return analysis have revealed that higher the farm size and higher the use of machineries would have a positive impact on the overall performance of rice production Fully Mechanized Small and Large Farms The analysis of costs and returns for fully mechanized small and large farms had shown almost similar comparative performance as that of the partially mechanized small and large farms. However, the per hactare cost of cultivation and cost of production had significantly reduced in fully mechanized farms. The cost of cultivation of rice had worked out to Rs.43,333 per hactare for small farms and Rs.40,028 for large farms. The significant difference in cost structure was the component of human labour. The small farms have incurred Rs.14,430 per hactare towards the cost of human labour and it was only Rs.7,565 per hactare for large farms. Conversely, large farms had more of machine labour replacing human labour substantially. The cost of machine labour for large farms worked out to Rs.19,396 per hactare while it was Rs.16,342 per hactare for the small farms. The share of total labour cost accounted for 56 per cent for small farms and 50 per cent for large farms, which made the main difference in the cost of cultivation of the two size groups. The next major component, namely the cost of inputs worked out to Rs.9,167 per hactare for small farms and Rs.10,179 per hactare for large farms. Akin to the partially mechanized farms, the share of fixed cost was higher for fully mechanized large farms which stood at Rs.9,435 per hactare, whereas the fixed cost for small farms worked out to Rs.7968 per hactare. The total cost of cultivation worked out to Rs.54,409 per hactare for large farms and Rs.56,431 per hactare for small farms. As regards the returns, the production of rice for the small farms was Qtl per hectare and Qtl per hectare for large farms, revealing the efficiency of large farms. The net income realized by the small and large farms was of the order of Rs.22,081 per hectare and Rs.24,692 per hectare respectively. The cost of production per quintal of rice worked out to Rs.795 for small farms and Rs.710 for large farms, indicating the advantage of economies of scale. The BCR was 1.54 for small farms and 1.65 for large farms. 83

97 Sl. No Table 5.12: Cost of Cultivation / Production of Rice per ha Particulars I. VARIABLE COST Partially mechanized Small farms Fully mechanized 1. Human Labour 23,395 (40.51) 14,430 (25.57) 2. Animal Labour 1,582 (2.74) 845 (1.5) 3. Machine Labour 9,240 (16.0) 16,342 (28.95) Total Labour 34,217 (59.26) 31,617 (56.02) 4. Seeds/Seedlings 1,482 (2.56) 1,235 (2.18) 5. Manures and Fertilizer 5,182 (8.97) 5,361 (9.5) 6. Plant Protection 932 (1.61) 1,140 (2.02) 7. Irrigation 1,645 (2.84) 1,431(2.53) Working Capital 9,241 (16.0) 9,167 (16.24) 8. Interest on Working ,716 (4.70) 2,544 (4.51) Per cent 9. Total variable cost 46,174 (99.97) 43,333 (76.78) II. FIXED COST 63,15 (10.93) 7,968 (14.11) SUB TOTAL I + II 52,439 (90.91) 51,301 (90.91) 10. Managerial 10 per cent 5,248(9.08) 5,130(9.09) III. TOTAL COST 57,737 (100.0) 56,431 (100.0) 11. Main product (Quintal) Value of Main Product 58,840 62, Value of By Product 2,840 2, Gross Income 61,680 65, Net Income Over variable cost 15,506 22, Cost of Production per Quintal BCR The overall analysis of the cost structure has revealed that large sized farms had the cost advantage over the small farms. The average cost of cultivation per hectare of paddy was Rs.41,435 for large farms, while the small farms had incurred the average cost of cultivation at Rs.44,753 per hactare. Similarly, the average cost of cultivation for fully mechanized farms was Rs.41,680 per hactare, while it was Rs.44,508 per hactare for partially mechanized 84

98 farms, revealing that the cost of cultivation will decrease as the level of mechanization increases. Table 5.13: Cost of Cultivation / Production of Rice per ha Sl. No 1. Particulars I. VARIABLE COST Human Labour Partially mechanized 19,894 (35.93) Large farms Fully mechanized 7,565 (13.90) 2. Animal Labour 769(1.38) 534 (0.98) 3. Machine Labour 10,432 (18.84) 19,396 (35.64) Total Labour 31,095 (56.17) 27,495 (50.53) 4. Seeds/Seedlings 1,876 (3.38) 1,185 (2.18) 5. Manures and Fertilizer 4,335 (7.83) 5,963 (10.95) 6. Plant Protection 1,235(2.23) 1,365 (2.51) 7. Irrigation 1,782 (3.22) 1,666(3.06) Working Capital 9,228(16.66) 10,179 (18.70) 8. Interest on Working 12.5 Per cent 2,520 (4.55) 2,354 (4.32) 9. Total variable cost 42,843 (77.39) 40,028 (73.56) II. FIXED COST 7,482 (13.51) 9,435 (17.34) SUB TOTAL I + II 50,325 (90.90) 49,463 (90.91) 10. Managerial 10 per cent 5,032(9.09) 4,946 (9.09) III. TOTAL COST 55,357(100.0) 54,409 (100.0) 11. Main product (Quintal) Value of Main Product 59,168 62, Value of By Product 3,205 2, Gross Income 62,373 64, Net Income Over variable cost 19,530 24, Cost of Production per Quintal BCR On production front, the average productivity for small farms was quintal per hactare, while it was Qtl for large farms. Comparing the two levels of mechanization, it was apparent that mechanization would lead to increased productivity. The average 85

99 productivity of fully mechanized farms was quintal per hactare, while the partially mechanized farms were able to produce only Qtl per hactare. This factor might be mainly attributed to the advantage of mechanical transplanting which help in uniform spacing, uniform number of seedlings per hill, shallow planting, profuse tillering and rooting anchorage. Table 5.14: Cost of Cultivation / Production of Rice per ha Sl. No Particulars Partially mechanized Fully mechanized I. VARIABLE COST 1. Human Labour 21,644.0 (38.27) 10,997.5 (19.84) 2. Animal Labour 1,175.5 (2.07) (1.24) 3. Machine Labour 9,836.0 (17.39) 17,869.0 (32.24) Total Labour 32,656.0 (57.73) 29,556.0 (53.33) 4. Seeds/Seedlings 1,679.0 (2.96) 1,210.0 (2.18) 5. Manures and Fertilizer 4,758.5 (8.41) 5,662.0 (10.22) 6. Plant Protection 1,083.5 (1.92) 1,252.5 (2.26) 7. Irrigation 1,713 (3.02) 1,548.5 (2.27) Working Capital 9,234.5 (16.32) 9,673.0 (17.45) 8. Interest on Working 12.5 Per cent 2,618.0 (4.62) 2,449.0 (4.41) 9. Total variable cost 44,508.0 (78.71) 41,680.5 (75.20) II. FIXED COST 6,898.5 (12.19) 8,701.0 (15.70) SUB TOTAL I + II 51,382.0 (90.90) 50,382.0 (90.90) 10. Managerial 10 per cent 5,140.0 (9.10) 5,038.0 (9.10) III. TOTAL COST 56,547.0 (100.0) 55,420.0 (100.0) 11. Main product (Quintal) Value of Main Product 59, , Value of By Product 3, , Gross Income 62, , Net Income Over variable cost 17, , Cost of Production per Quintal BCR

100 On the income side, the average net income of the small farms worked out to Rs.18,793 per hactare while the large farms realized an average net income of Rs.22,113 per hactare capturing the difference in the cost structure of the respective size of farms. Similarly, the partially mechanized farms had a net income of Rs.17,518 per hactare while the fully mechanized farms generated a net income of Rs. 23,389 per hactare. Table 5.15: Cost of Cultivation / Production of Rice per ha Sl. No Particulars Small farms Large farms I. VARIABLE COST 1. Human Labour 18,912.5 (13.13) 13,729.5 (25.02) 2. Animal Labour 1,213.5 (2.12) (11.86) 3. Machine Labour 12,791.0 (22.40) 14,914.0 (27.17) Total Labour 32,917.0 (57.66) 29,295.0 (53.37) 4. Seeds/Seedlings 1,358.5 (2.37) 1,530.5 (2.78) 5. Manures and Fertilizer 5,271.5 (9.23) 5,149.0 (9.38) 6. Plant Protection 1,036.0 (1.81) 1,300 (2.36) 7. Irrigation 1,538.0 (2.69) 1,724.0 (3.14) Working Capital 9,204.0 (16.12) 9,703.5 (17.67) 8. Interest on Working 12.5 Per cent 2,630.0 (4.60) 2,437.0 (4.44) 9. Total variable cost 44,751.0 (78.39) 41,435 (75.49) II. FIXED COST 7,141.5 (12.5) 8,458.5 (15.41) SUB TOTAL I + II 51,870.0 (90.86) 49,894.0 (90.90) 10. Managerial 10 per cent 5,189.0 (9.14) 4,989.0 (9.10) III. TOTAL COST 57,084.0 (100.0) 54,883.0 (100.0) 11. Main product (Quintal) Value of Main Product 60, , Value of By Product 2, , Gross Income 63, , Net Income Over variable cost 18, , Cost of Production per Quintal BCR

101 The average cost of production of rice worked out to Rs.872 per quintal for small farms and Rs.766 per quintal for large farms, proving the efficiency of large sized farms. Similarly, the average cost of production for partially mechanized farms was Rs per Qtl, while the same worked out to Rs per quintal for fully mechanized farms. From the analysis of cost and returns of rice production it could be concluded that the large sized and fully mechanized farms were comparatively more efficient than the other category of size and level. The comparative cost and returns estimates for different size group and for different level of mechanization is summarized and presented in Table 5.15 (a) Table 5.15 (a): Cost of Cultivation / Production of Rice per ha Sl. No Particulars Partially mechanized Fully mechanized Small farms Large Small Large farms 1 Total cost (Rs) 57,737 55,357 56,431 54,409 2 Main product (Qtl) Gross Income (Rs) 61,680 62,373 65,414 64,725 4 Net Income (Rs) 15,506 19,530 22,081 24,697 5 Cost of Production / Qtl BCR Partial Budgeting Analysis As set out earlier in the design of study, the transplanting operation was the deciding factor for the classification of farms as partially mechanized and fully mechanized farms. The farms with all operations mechanized except transplanting were classified as partially mechanized farms and the farms with all operations mechanized including transplanting were classified as fully mechanized farms. It is to be mentioned that the weeding operation had been excluded in both the type of farms since no suitable machineries have been introduced so far for weeding. Transplanting has been one of the major labour intensive operations in rice farming. Transplanter have been introduced only very recently in CDZ and successful in the operation. The farmers in the study area have been joyous after the introduction of the transplanters on two aspects. Firstly, its arrival had been handy at a time when there was severe labour problem during the planting season. Secondly, they could get an additional yield of per cent due to meticulous planting, uniform spacing, shallow planting, profuse tillering and other agronomic factors. Therefore, ascertaining the economic impact of rice transplanters is of great importance in rice cultivation. Hence, partial budgeting analysis was performed and the results are furnished in Tables 5.16 and

102 Figure 5.5: Rice Transplanter in Operation 89