This paper estimates and compares the paid-out cost of cultivation of wheat in India, the most state-protected crop, during the input subsidy regime of the 1970s and 1980s and after its abolition in the 1990s, when economic reforms were initiated. The study uses the valuable time series information collected as part of the "comprehensive scheme" of the ministry of agriculture. After surveying the pattern of changes in inputs as well as costs of cultivation vis-à-vis the wholesale price index (a proxy for the general price level), the value of inputs which are exclusively market-purchased are analysed. A study of the weighted average of these costs establishes unequivocally that the costs of farm inputs increased very sharply in the post-reform period.
REVIEW OF AGRICULTUREEconomic & Political Weekly EPW june 28, 2008123Changing Pattern of Input Use and Cost of Cultivation M RaghavanThe agrarian crisis that has ravaged India’s rural country-side during the post-reform era has grown on a three-pronged set of symptoms: rising input costs, dwindling produce price realisation and the inability of farmers to abandon cultivation without alternative livelihood sources. It is a well known fact that the crisis started when the government decided to demolish, without giving adjustment time, the elaborate mechanism that was built up, in stages, in the post-independence period to the beginning of the 1990s to protect the peasantry from the vicissitudes of the market [Patnaik 2003]. These protec-tionist arrangements, consisting essentially of a system of input price subsidies and output price support, despite weaknesses in implementation, had enabled farmers to take up cultivation in a predictably stable price environment. During the post-reform period, the government not only slashed the subsidies on major inputs, but also absolved itself of the responsibility to produce or procure and distribute these inputs at farm gates. When prices of farm inputs thus went up, private operators seized the opportu-nity and pushed up prices further. The situation got worse when the rates of interests on institutional credits were raised, while the narrow window of such credits became narrower, dragging the cultivating households into the clutches of private money-lenders [Bagchi 2004]. There is incontrovertible statistical evidence in the vast litera-ture on the subject as to how declining farm prices aggravated the crisis. The views about rising costs of cultivation, on the other hand, though unquestionable, are based mostly on qualitative field level information gathered from areas where farmers’ suicides are reported. The costs of cultivation data that the Commission for Agricultural Costs and Prices (CACP) gives in its price policy reports, when put together, would show that expenses on farm inputs have registered a spectacular increase since the 1990s. These cost data were generated under a “comprehensive scheme” initiated in the late 1960s by the directorate of econo-mics and statistics (DES) of the ministry of agriculture, govern-ment of India and are available from 1970-71 to the present. The literature on India’s farm crisis has ignored this valuable source of information. Observers will recall that, when similar data under the farm management surveys (FMS) came out in the 1950s and 1960s, these led to some path-breaking studies on issues like production conditions and production relations in Indian agricul-ture. This was in spite of the fact that the FMS data were limited in sample size and related only to select districts for just three years at a stretch. As against this, the “comprehensive scheme” collects state-level time series data. Information-wise these were far wider in This paper estimates and compares the paid-out cost of cultivation of wheat in India, the most state-protected crop, during the input subsidy regime of the 1970s and 1980s and after its abolition in the 1990s, when economic reforms were initiated. The study uses the valuable time series information collected as part of the “comprehensive scheme” of the ministry of agriculture. After surveying the pattern of changes in inputs as well as costs of cultivation vis-à-vis the wholesale price index (a proxy for the general price level), the value of inputs which are exclusively market-purchased are analysed. A study of the weighted average of these costs establishes unequivocally that the costs of farm inputs increased very sharply in the post-reform period.I am immensely indebted to Prabhat Patnaik for encouraging me to work on costs of cultivation and enabling me to understand the implications of various cost concepts, to Abhijit Sen for useful discussions while writing this paper, and to V M Rao for reading through an earlier draft and suggesting several improvements. Needles to say I am alone responsible for the views expressed and any lacuna still remaining.M Raghavan (raghavan_m2005@rediffmail.com) teaches at the KMCT School of Business, Kozhikode, Kerala.
0 100 200 300 400 500 600 C
REVIEW OF AGRICULTUREEconomic & Political Weekly EPW june 28, 2008125early 1990s. Thereafter, the upward pace of the fixed cost curve tapers off and that of the operational cost spurts up, establishing beyond doubt that it is the expenses on variable inputs that have escalated the estimated average costs of cultivation. The gradual diminution of the upward movement of the fixed cost curve is not without reason. There are two items in fixed costs, namely, the rental value of owned land and interest imputed to owned fixed assets, which together had formed its near total in the earlier decades. The rental value of owned land is imputed at the prevailing rent for identical leased land in a village. If leasing of land does not exist in the village, the rental has to be calculated from the capitalised value of land. When farm prices face a prolonged depression, as has happened since the 1990s, ceteris paribus the value of land should also behave likewise. For the estimation of the second major item, i e, the interest on owned fixed capital, the first step is to work out the depreciated present value of capital invested by the sample households. The bank rate being declared by the Reserve Bank of India (RBI) was applied on this value to obtain the imputed interest on owned fixed capital. After the mid-1990s, it was replaced by the much larger commercial market rates of interest. Even then, if the imputed interest on owned fixed capital has not increased, it is a clear case of declining investments in fixed assets by farmers. To be precise, the fixed capital costs stagnated in the post-reform period because first, the prolonged depression in farm prices exerted downward pressure on the estimated value of land from which the rental is arrived at. Second, there has been a decline in the private investment in fixed assets, the present value of which is the basis for imputing the interest cost. 2 Trends in Variable Input CostsThe relentless increase in operational costs, as already stated, resulted essentially from the removal of input subsidies and the dependence of farmers on the open market for crucial items of inputs. This might also reflect, to a lesser extent, the changes effected in the methods of valuation of certain inputs and services, such as imputation of wages for family labour and calcu-lation of interest cost (about which more later). In what follows we will briefly survey the nature and trends in major variable physical inputs and their costs and then assess the extent of increase in paid-out costs during the post-reform period.2.1 Wages and EmploymentOn a casual perusal of these data in a different context, a recent study argues that “the cost of cultivation of crops has been increasing over the years because of increases in wage rate of labour, input prices and other managerial costs” [Narayana-moorthy 2007]. As a starting point, therefore, let us take the Table 1: Average Cost of Cultivation(Rs/hect) 1970-71 1970-81 1981-91 1991-2001 2001-05 2004-05Haryana 1,265.122,320.194,671.1514,556.29 23,271.11 24,179.23Madhya Pradesh 515.83 1,195.47 2,944.39 9,178.80 14,170.04 14,696.06Punjab 1,654.592,489.91 5,592.9215,964.32 23,135.18 24,197.48Rajasthan 2,311.124,302.0912,972.4620,122.7419,609.52Uttar Pradesh 1,408.62 2,180.95 4,800.97 12,426.80 18,730.03 20,812.56All states 1,417.05 2,100.57 4,291.48 12,669.42 19,269.48 19,910.52Table 2: Labour Use and Wage Cost 1970-71 1970-81 1981-91 1991-2001 2001-05 2004-05 (A) Human Labour Hours Per HectareHaryana 396.20484.61384.64338.60303.71293.37Madhya Pradesh 330.49 450.14 375.74 372.09 351.77 315.59Punjab 331.07440.01400.40326.84203.59163.56Rajasthan 686.62585.25567.14563.97504.02Uttar Pradesh 550.09 675.63 594.21 516.41 452.52 447.40All states 463.83 573.86 499.05 446.12 417.75 362.56 (B) Bullock Labour Hours Per HectareHaryana 134.19125.2150.0116.527.786.95Madhya Pradesh 157.29 186.82 127.51 55.74 43.28 34.55Punjab 104.7781.5821.783.711.831.53Rajasthan 209.16105.7449.81 19.82 16.84Uttar Pradesh 247.17 228.57 128.86 52.56 17.70 15.00All states 192.27 184.56 101.11 41.47 18.88 15.91 (C) Total Wage Cost (Rs/hectare)Haryana 232.31449.22699.892,618.123,971.973,866.63Madhya Pradesh 119.18 227.69 441.78 1,622.75 2,349.17 2,260.03Punjab 328.3430.48853.152,486.732,288.11,952.42Rajasthan 443.15792.853,363.174,969.334,618.55Uttar Pradesh 264.69 380.17 862.09 2,415.35 3,319.5 3,483.24All states 230.56 368.56 758.16 2,400.07 30,11.66 3,138.74 (D) Wage Cost as % of Operational CostHaryana 29.30 29.40 22.83 32.08 29.19 27.15Madhya Pradesh 41.38 33.19 26.53 30.14 27.41 25.35Punjab 35.42 28.23 22.64 29.07 20.1816.73Rajasthan 26.72 29.17 40.29 37.39 36.10Uttar Pradesh 29.72 26.15 26.55 31.59 28.39 26.13All states 30.72 27.69 25.57 32.05 26.37 25.66 (E) Casual Wage as % of Total WageHaryana 10.819.531.728.227.025.1Madhya Pradesh 27.4 32.1 41.1 34.7 32.6 34.3Punjab 25.938.752.455.848.551.2Rajasthan 21.617.813.416.718.9Uttar Pradesh 29.2 27.8 33.6 34.3 38.4 37.5All states 21.3 28.8 36.3 34.0 34.0 34.0wage-rate first. It is common knowledge that wage cost depends on the wage-rate and the labour-hours applied in cultivation. The labour use is a function inter alia of labour intensity in cultiva-tion, i e, the relative level of application of human labour to machine labour. Their relationship is generally negative, as machine-labour is a device that displaces human labour. The hours of human labour applied in cultivation depends also on the extent of bullock labour use. These two are positively correlated as more and more use of bullock labour would necessitate the requirement of more and more male farm labour. The employ-ment of manual workers in the cultivation of wheat, after record-ing some improvement until the mid-1970s, has been declining since. The decline was pervasive in all major wheat growing states though it was considerably steeper in Punjab, Haryana and Uttar Pradesh. In the mid-1970s, the wheat crop required on average 72 person-days of manual labour,2 ranging from 55-60 days in Punjab and Haryana to 85 days in Uttar Pradesh and Rajasthan. By 2004-05, this came down to 63 person-days in Rajasthan, 56 in Uttar Pradesh, 37 in Haryana and 20 in Punjab. Consistent with this, bullock labour use has almost disappeared from several parts of these states (Table 2).
REVIEW OF AGRICULTUREjune 28, 2008 EPW Economic & Political Weekly126Nonetheless, the estimated average wage cost in wheat culti-vation accounts for the largest segment of all items of operational costs. It remained more or less stagnant at 26-27 per cent during the 1970s and 1980s. In the 1990s as a whole, however, there has been a substantial improvement in the average share of wages to 32 per cent. The share of wages had been actually declining in the latter half of the 1990s, and presently it stands at 26 per cent of the operational costs. During 2001-05, the decline in the share of wages was around 3-4 percentage points in Haryana, Madhya Pradesh, Rajasthan and Uttar Pradesh and as large as 9 percentage points in Punjab. Wage costs, as reported in the comprehensive scheme, include wages actually paid to casual and permanently attached labour and those imputed to family labour. The casual workers are engaged to do certain specific peak-season jobs, such as digging, transplanting, weeding, manuring, threshing, etc. Wages paid to them are relatively higher. Although attached labourers also do similar work, it is an established fact that, money wages paid to them are generally sticky and lower than casual wages. The payments accounted against attached labourers, however, include the value of doles given to them in the form of second-hand dresses, slippers, tents to take rest, helps extended in times of marriages of family members, etc. As far as family labour is concerned, small peasants use it mostly because of its abundance in relation to their operational area. The big farmers may not use family labour to do the work that other wage workers do. However, they draw the services of family members during peak periods when sufficient casual workers are not available. The family labour is also used in certain tasks requir-ing special care and skill like handling of sophisticated machinery. The wages imputed for such work should be larger than the market wages for casual labour. In the 1970s and 1980s, wages forfamilylabour were imputed at wages for attached labour. Subsequently, casual wages have replaced the basis to impute family labour charges. The all-state average of casual wage as a proportion of total wage, which, after increasing from 28.8 per cent in the 1970s to 36.3 per cent in the 1980s, declined to remain stable at 34 per cent thereafter. The decline was more pronounced in Haryana, Madhya Pradesh and Rajasthan, while in Uttar Pradesh it was apparently rising. In Punjab, where a large number of male members of rural families migrate abroad, wages for casual labour are relatively high and have been distinctly on the rise. Even then, in recent years, the share of casual wages in total wage cost has marginally declined in these two states. Two points emerge clearly from these data. First, average wages as a ratio of operational costs have not increased in the same rate as its other components did since the 1990s. Secondly, wages paid to hired-labour as a proportion of operational costs have remained stagnant throughout the post-reform period. 2.2 Chemical Fertilisers and PesticidesFertiliser charges account for the second largest component, after wages, in the operational costs of cultivation. Being a vital input contributing to productivity, there is a tendency among farmers to go on increasing the application of chemical fertilisers (along with high-yielding seeds), unmindful of the accompany-ing problems. For example, chemical fertilisers induce weeds, pests and nematodes, requiring the application of pesticides and weedicides. Too much water, or too little water, adversely affects the fertiliser response. For controlled water supply, farmers have to adopt sophisticated technology. When high-yielding seeds, which are short duration in nature, are used in this system of cultivation, inter-cultures have to be carried out in quick succession. This will not be possible unless farmers can mechanise different operations like the land preparation, planting, watering, weeding, harvesting and threshing. In other words, it is the level of application of chemical fertilisers that determines the level of mechanisation and the paid out costs in cultivation. From the “green revolution” period of the 1960s, there has been a tremendous increase in the use of chemical fertilisers in the cultivation of wheat, mainly in Punjab, Haryana and Table 3: Fertiliser Use and Fertiliser Charges 1970-71 1970-81 1981-91 1991-2001 2001-05 2004-05 (A) Fertiliser Use (Qtl/hectare) Haryana 44.5467.52138.41176.81195.22194.56Madhya Pradesh 5.15 21.25 42.10 76.04 89.67 94.00Punjab 107.93123.75174.78209.56227.02236.79Rajasthan 47.61 51.18 83.78 108.80102.21Uttar Pradesh 48.22 61.86 104.30 132.73 149.92 155.82All states 46.06 62.19 102.36 132.81 152.51 156.22 (B) Fertiliser Charges (Rs/hectare)Haryana 111.09 225.52 668.961,687.082,331.26 2,396.26Madhya Pradesh 4.14 111.33 224.67 834.85 1,174.75 1,241.75Punjab 208.6414.04886.262,029.851,891.21,180.96Rajasthan 235.42283.38842.741,397.481,329.48Uttar Pradesh 93.08 213.25 558.86 1,361.43 1,882.22 2,053.69All states 100.75 227.89 533.85 1,339.3 1,754.82 1,716.84Table 4: Cost of Insecticides(Rs/hectare) 1970-71 1970-81 1981-91 1991-01 2001-05 2004-05Haryana 0.001.0458.71 170.65 706.66802.79Madhya Pradesh 0.00 0.00 0.50 1.41 9.90 7.02Punjab 0.414.14108.07417.851,129.041,180.04Rajasthan 0.885.3313.9923.0021.92Uttar Pradesh 0.02 0.07 1.60 13.05 21.76 33.66All states 0.22 0.89 25.01 90.92 278.84 303.21Table 5: Machine Labour Charges 1970-71 1970-81 1981-91 1991-2001 2001-05 2004-05 (A) Total Machine Labour Charge (Rs/hectare) Haryana 54.34152.08602.071,689.353,240.543,728.49Madhya Pradesh 1.79 17.42 133.53 760.75 1,334.03 1,533.38Punjab 80.43210.67685.921,683.143,466.433,866.43Rajasthan 116.72358.381,214.651,907.752,063.90Uttar Pradesh 38.05 81.37 449.42 1,291.22 2,619.46 2,969.46All states 35.77 98.95 434.38 1,281.87 2,520.07 2,832.10 (B) Hired as % of Total Machine Labour ChargeHaryana 30.03 48.73 59.23 75.43 71.23 67.91Madhya Pradesh 30.73 76.00 74.39 87.83 91.57 91.40Punjab 50.58 36.67 44.29 60.57 64.58 64.32Rajasthan 98.8380.6595.5691.0588.82Uttar Pradesh 75.14 71.97 81.90 93.65 89.18 85.51All states 69.42 60.45 68.52 83.92 81.44 79.00
REVIEW OF AGRICULTUREEconomic & Political Weekly EPW june 28, 2008127latter,whileescalating the prices of fertilisers made it inaccessi-ble to smaller farmers. Fertilisers induce attacks of insect-pests, weeds and plant parasites. Weeds compete with wheat crops for space, light, water and fertiliser nutrients. According to Indian Council of Agricul-tural Research, while insect-pests, like termites, stem borer, weevils and wheat-amphids render grain unfit for human consumption, plant parasites (nematodes) cause huge crop losses every year. Weeds in a hectare of wheat field are estimated to remove at least 20 kg of fertiliser nutrients and reduce a minimum of 12 kg in yields [ICAR 1997]. Farmers using high doses of chemi-cal fertilisers have to incur large expenses on insecticides, weedi-cides and herbicides. With the emergence of private operators, costs on this account have mounted several folds. The all-state average expenses per hectare on insecticides, which worked out to less than a rupee in the 1970s and just over Rs 25 in the 1980s, shot up by 365 per cent in the 1990s and by as much as 1,115 per cent in the first half of this decade (Table 4, p 126). 2.3 MachineLabour The machine labour charge, which was less than 4 per cent of the operational cost in 1970-71, rose tremendously to reach above 23 per cent in 2004-05 and relegated fertiliser cost from its second position in the operational cost. This could be attri-buted mainly to three factors: first, with the growing diffusion of technology in wheat, there has been widespread mechanisa-tion of almost all farming operations like ploughing, irrigat-ing, weeding, harvesting and threshing. Initially, only larger holders in prosperous states of Punjab, Haryana and Uttar Pradesh adopted mechanised means of cultivation. At present, even farmers in poorer states depend increasingly on machine labour as the statewide trends in its per hectare costs given in Table 5 (p 126) demonstrate. Table 8: Interest Costs(Rs/hectare) 1970-71 1970-81 1981-91 1991-2001 2001-05 2004-05 (A) Interest on Working Capital Haryana 15.6235.8679.63194.89328.26347.47Madhya Pradesh 6.30 16.20 43.32 133.90 212.59 226.15Punjab 18.6238.8286.80209.13316.26331.55Rajasthan 38.3964.06168.93280.43277.12Uttar Pradesh 16.78 35.19 81.66 185.20 292.84 338.47All states 15.12 31.32 73.29 178.12 284.49 310.26 (B) Interest on Fixed and Working CapitalHaryana 108.02202.48509.0312,45.721,878.751,904.29Madhya Pradesh 50.56 110.90 306.95 1,026.09 1,476.57 1,416.03Punjab 107.66197.81538.811,334.892,081.812,157.81Rajasthan 129.69481.011,111.712,047.741,715.56Uttar Pradesh 84.22 161.02 424.43 1,413.88 1,711.36 1,768.33All states 93.34 166.32 434.08 1,070.38 1,781.36 1,773.36Table 9: Cost of Cultivation of Selected Purchased Inputs (Rs/hectare) 1970-71 1970-81 1981-91 1991-2001 2001-05 2004-05Haryana 277.82620.491,756.465,151.238,544.448,955.10Madhya Pradesh 45.95 245.36 791.97 3,096.23 5,009.28 5,614.27Punjab 425.21810.732,011.775,345.946,244.716,641.56Rajasthan 745.861,235.123,990.126,899.576,725.72Uttar Pradesh 304.52 564.12 1,619.01 4,684.38 7,765.39 8,900.33All states 292.85 560.58 1,502.64 4,451.95 7,110.15 7,673.71Uttar Pradesh. Estimates made by Frankel (1971) show that the quantity of chemical fertilisers applied in cultivation in Punjab (including Haryana) in 1960-61 was 3.23 kg per hectare. As against this, the all-state average use of fertilisers was 62 kg in the 1970s, 102 kg in the 1980s, 133 kg in the 1990s and 153 kg in the first half of the current decade. On a closer look, these figures exhibit two significant features. First, there are sizeable inter-state differences in the application of fertilisers. Punjab reported the highest level of fertiliser use in wheat of almost twice as high as the average for the remaining states, followed by Haryana and Uttar Pradesh and the lowest by Madhya Pradesh and Rajasthan (Table 3, p 126). Secondly, despite continued increase in absolute terms, the trend rate of fertiliser use in the cultivation of wheat has noticeably decelerated in the post-reform era. The growth rates of fertiliser application, measured from its all-states averages, had registered a compound growth rate of 4 per cent per annum in the 1980s. During the entire reform period, the rate of growth was just 2.1 per cent per annum.The fertiliser charges, on the other hand, have been steeply accelerating since the 1990s. The all-state average rates of ferti-liser costs, estimated at Rs 228 per hectare in the 1970s, rose to Rs 534 in the 1980s, to Rs 1,339 in the 1990s and to Rs 1,755 per hectare in the first half of the current decade. In the 1980s, when the fertiliser application increased at 4 per cent per annum, the increase in the rate of growth of fertiliser charges was only 4.5 per cent. During the reform period, when fertiliser application growth decelerated to 2.1 per cent a year, fertiliser cost growth accelerated to 6.2 per cent per annum. The large increase in fertiliser application and a similar increase in the unit cost of fertiliser per hectare, witnessed in the 1970s and 1980s, were the result of a consciously devised policy that the government followed: (1) supply of fertlisers to cultivators at subsidised “farm-gate prices”, (2) protection of the domestic fertiliser industry with a “retention price scheme” that provided an assured return, and (3) encouragement of fertiliser distribu-tionbytheindustrywith the extension of an additional “freight equalisation” subsidy [Bhalla 2007]. In contrast to this, the 1990s and after marked the decontrol of fertiliser prices, the entry of private operators in its distribution and replacement of the “retention price scheme” with a “new fertiliser policy”. The Table 6: Cost of Seeds(Rs/hectare) 1970-71 1970-81 1981-91 1991-2001 2001-05 2004-05Haryana 96.74147.71346.48687.38984.401,013.50Madhya Pradesh 88.13 145.80 280.27 732.79 947.20 997.84Punjab 65.35110.57238.78568.69802.35865.35Rajasthan 173.25310.46849.231,310.621,323.06Uttar Pradesh 99.51 142.77 295.51 721.12 1,122.32 1,233.33All states 92.44 141.83 289.16 710.63 1,040.17 1,109.92Table 7: Irrigation Charges (Rs/hectare) 1970-71 1970-81 1981-91 1991-2001 2001-05 2004-05Haryana 109.60196.46370.92901.741,819.861,906.19Madhya Pradesh 2.33 31.40 242.48 797.61 1,660.08 1,961.51Punjab 71.95109.96179.56321.80431.33461.72Rajasthan 259.89452.341,145.422,647.532,389.05Uttar Pradesh 88.80 151.52 319.42 939.16 2,005.66 2,630.44All states 83.99 134.56 297.98 832.29 1,724.32 2,034.49
REVIEW OF AGRICULTUREEconomic & Political Weekly EPW june 28, 2008129sector” credit to agriculture is no more available at the interest rate of 4.5 per cent, as in the 1970s and 1980s. Studies show that banks charge 15-18 per cent on agricultural credit granted to relatively poorer peasants. Further, once there is a default in repayments, these banks impose a penal interest of up to 24 percent on the compounded loan outstanding [George and Krishnaprasad 2006]. This is one reason why the interest cost on working capital has registered an unprecedented increase since the 1990s. Thirdly, with the institutional credit becoming difficult to obtain, farmers are forced to seek funds from outside the bound-aries of the formal financial systems. Conventionally, their main sources of borrowing were relatives and friends at zero interest, from traders against promise to sell their harvest, and from indig-enous moneylenders at exorbitant interest [Bardhan and Udry 2000]. When farm prices declined, it is only reasonable to believe that the possibilities of borrowing from friends and relatives in rural areas also disappeared. The other sources left are traders and indigenous moneylenders. The predominance of private moneylenders in credit transactions of farmers is yet another reason why the interest cost shot up by more than fourfolds during the post-reform era. 3 ConclusionsThe cost of cultivation data generated under the comprehensive scheme is unequivocal in establishing that coinciding with the economic reforms expenses on farming have soared to unprecedented heights. The foregoing analysis shows that not all items of costs have increased at the same pace. While fixed costs seemed to exhibit a gradual deceleration, operational costs have continued their relentless acceleration. The former reflects the prolonged depression in farm prices during the period under analysis and the resultant decline in private investments in agriculture. On the other hand, with the slash-ing of subsidies and the state agencies becoming redundant in distributing inputs, farmers naturally fell into the hands of unscrupulous private operators. This was instrumental in escalating operational costs. Our second major observation is that, during the post-reform period, there has been a steep decline in the labour hours applied in cultivation as also stagnation in casual wages. It is clear from this that the agrarian crisis in the post-reform period afflicted not only the cultivating households but also the entire agriculture-dependent population. During this period, there has been a deceleration in the rate of growth of chemical fertilisers applied in cultivation. Nonetheless, the corresponding rate of growth of fertiliser charges was three times higher than that of its physical application. The decontrol of fertiliser prices and the entry of private operators in distribution are evident from this. The break-up of variable costs available in the comprehensive scheme are averages of the owned and purchased inputs. It is difficult to ascertain from this the actual increase in paid-out costs of cultivation over time. We therefore desegregated the items of costs that the cultivators source exclusively or predomi-nantly from the market. These items are hired human labour, hired machine labour, chemical fertilisers, insecticides, irriga-tion charges and interests on short-term working capital. The average expenses of these six items in the cultivation of wheat were slightly above Rs 1,500 per hectare in the 1980s, varying from Rs 800 in Madhya Pradesh to Rs 2,012 in Punjab. It increased to Rs 4,500 per hectare in the 1990s and to Rs 7,700 per hectare in 2004-05, the last year for which we have the data, which is more than five times above the level in the 1980s. In Haryana and Uttar Pradesh, it was still higher (Table 9, p 127).A further exercise was carried out to see how these costs might have behaved in case the input subsidy regime of the 1980s continued as such during the post-reform period. Thus, we fitted a linear trend to the all-states average of the six items of costs for 1981-82 to 1990-91 and extended the same to 1991-92 to 2004-05. The extended trend line registers a linear growth rate of 8.2 per cent per annum. Compared to this, the actual increase in these cost items was 22.5 per cent and that of the WPI, a proxy for the general price level, 7 per cent. From this, two points become clear: First, even if the input subsidy regime continued the costs of cultivation of wheat, one of the most state protected crops, could have increased faster than the increase in the general price level. This shows the weakness of that regime. However, when that regime was in fact discontinued, the paid-out costs of crucial items of agricultural inputs increase at a rate that had not been seen earlier. Notes1 Afterwards, DES has published two more instal-ments, in 1996 and 2000, covering the data up to 1998-99. 2 The figures are estimated from Table 2 at the rate of eight human labour hours as one person-day. 3 For a long list of bank loans treated as priority sector credit for agriculture that the RBI has been enriching from 1994 to 2007, see Ramkumar and Chavan (2007).4 The comprehensive scheme introduces a new concept of “Cost-B1”, which is defined as Cost-A1 + “interest on the value ofowned fixed capital assets, excluding land”. The difference between Cost-B1 and Cost-A1 is the same as interest onfixed capital that it has been publishing separately. ReferencesBagchi, Amiya Kumar (2004): ‘The Common Minimum Programme’, Social Scientist, July-August, pp 3-16. Bardhan, Pranab and Christopher Udry (2000): ‘Fragmented Credit Markets’ inReadings in Devel-opment Economics, Oxford University Press, New Delhi, pp 76-93.Bhalla, G S (2006): Condition of Indian Peasantry, National Book Trust, India, New Delhi. – (2007): Indian Agriculture Since Independence, National Book Trust, India, New Delhi. George, J and P Krishnaprasad (2006): ‘Agrarian Distress and Farmers’ Suicides in the Tribal Districts of Wayanad’,Social Scientist, July-August, pp 70-85.Frankel, Francine R (1971):India’s Green Revolution: Economic Gains and Political Costs, Oxford Univer-sity Press, New Delhi. ICAR (1997): Hand Book of Agriculture, Indian Council of Agricultural Research, New Delhi.Kapre, B N (1974): Comprehensive Scheme for Studying the Cost of Cultivation of Principal Crops, Directo-rate of Economics and Statistics, Government of India, New Delhi. Narayanamoorthy, A (2007): ‘Deceleration in Agricultural Growth: Technology Fatigue or Policy Fatigue?’Economic & Political Weekly, June 23, pp 2375-79.Patnaik, Prabhat (2003): ‘Agricultural Production and Prices under Globalisation’ in his The Retreat to Unfreedom, Tulika Books, New Delhi, pp 198-210. Ram, G S (2001): ‘Data Relating to Cost of Cultivation Surveys’, in Chandrasekhar and Tilak (eds):India’s Socio-Economic Database – Surveys of Selected Areas, Tulika Books, New Delhi. Ramkumar, R and Pallavi Chavan (2007): ‘Revival of Agricultural Credit in the 2000s: An Explanation’, Economic & Political Weekly, December 29, pp 57-63. Sen, Abhijit and M S Bhatia (2004): Cost of Cultiva-tion and Farm Income, State of the Indian Farmer: A Millennium Study, Vol 14, Government of India), Academic Foundation, New Delhi.
REVIEW OF AGRICULTUREEconomic & Political Weekly EPW june 28, 2008123Changing Pattern of Input Use and Cost of Cultivation M RaghavanThe agrarian crisis that has ravaged India’s rural country-side during the post-reform era has grown on a three-pronged set of symptoms: rising input costs, dwindling produce price realisation and the inability of farmers to abandon cultivation without alternative livelihood sources. It is a well known fact that the crisis started when the government decided to demolish, without giving adjustment time, the elaborate mechanism that was built up, in stages, in the post-independence period to the beginning of the 1990s to protect the peasantry from the vicissitudes of the market [Patnaik 2003]. These protec-tionist arrangements, consisting essentially of a system of input price subsidies and output price support, despite weaknesses in implementation, had enabled farmers to take up cultivation in a predictably stable price environment. During the post-reform period, the government not only slashed the subsidies on major inputs, but also absolved itself of the responsibility to produce or procure and distribute these inputs at farm gates. When prices of farm inputs thus went up, private operators seized the opportu-nity and pushed up prices further. The situation got worse when the rates of interests on institutional credits were raised, while the narrow window of such credits became narrower, dragging the cultivating households into the clutches of private money-lenders [Bagchi 2004]. There is incontrovertible statistical evidence in the vast litera-ture on the subject as to how declining farm prices aggravated the crisis. The views about rising costs of cultivation, on the other hand, though unquestionable, are based mostly on qualitative field level information gathered from areas where farmers’ suicides are reported. The costs of cultivation data that the Commission for Agricultural Costs and Prices (CACP) gives in its price policy reports, when put together, would show that expenses on farm inputs have registered a spectacular increase since the 1990s. These cost data were generated under a “comprehensive scheme” initiated in the late 1960s by the directorate of econo-mics and statistics (DES) of the ministry of agriculture, govern-ment of India and are available from 1970-71 to the present. The literature on India’s farm crisis has ignored this valuable source of information. Observers will recall that, when similar data under the farm management surveys (FMS) came out in the 1950s and 1960s, these led to some path-breaking studies on issues like production conditions and production relations in Indian agricul-ture. This was in spite of the fact that the FMS data were limited in sample size and related only to select districts for just three years at a stretch. As against this, the “comprehensive scheme” collects state-level time series data. Information-wise these were far wider in This paper estimates and compares the paid-out cost of cultivation of wheat in India, the most state-protected crop, during the input subsidy regime of the 1970s and 1980s and after its abolition in the 1990s, when economic reforms were initiated. The study uses the valuable time series information collected as part of the “comprehensive scheme” of the ministry of agriculture. After surveying the pattern of changes in inputs as well as costs of cultivation vis-à-vis the wholesale price index (a proxy for the general price level), the value of inputs which are exclusively market-purchased are analysed. A study of the weighted average of these costs establishes unequivocally that the costs of farm inputs increased very sharply in the post-reform period.I am immensely indebted to Prabhat Patnaik for encouraging me to work on costs of cultivation and enabling me to understand the implications of various cost concepts, to Abhijit Sen for useful discussions while writing this paper, and to V M Rao for reading through an earlier draft and suggesting several improvements. Needles to say I am alone responsible for the views expressed and any lacuna still remaining.M Raghavan (raghavan_m2005@rediffmail.com) teaches at the KMCT School of Business, Kozhikode, Kerala.
0 100 200 300 400 500 600 C 
REVIEW OF AGRICULTUREEconomic & Political Weekly EPW june 28, 2008125early 1990s. Thereafter, the upward pace of the fixed cost curve tapers off and that of the operational cost spurts up, establishing beyond doubt that it is the expenses on variable inputs that have escalated the estimated average costs of cultivation. The gradual diminution of the upward movement of the fixed cost curve is not without reason. There are two items in fixed costs, namely, the rental value of owned land and interest imputed to owned fixed assets, which together had formed its near total in the earlier decades. The rental value of owned land is imputed at the prevailing rent for identical leased land in a village. If leasing of land does not exist in the village, the rental has to be calculated from the capitalised value of land. When farm prices face a prolonged depression, as has happened since the 1990s, ceteris paribus the value of land should also behave likewise. For the estimation of the second major item, i e, the interest on owned fixed capital, the first step is to work out the depreciated present value of capital invested by the sample households. The bank rate being declared by the Reserve Bank of India (RBI) was applied on this value to obtain the imputed interest on owned fixed capital. After the mid-1990s, it was replaced by the much larger commercial market rates of interest. Even then, if the imputed interest on owned fixed capital has not increased, it is a clear case of declining investments in fixed assets by farmers. To be precise, the fixed capital costs stagnated in the post-reform period because first, the prolonged depression in farm prices exerted downward pressure on the estimated value of land from which the rental is arrived at. Second, there has been a decline in the private investment in fixed assets, the present value of which is the basis for imputing the interest cost. 2 Trends in Variable Input CostsThe relentless increase in operational costs, as already stated, resulted essentially from the removal of input subsidies and the dependence of farmers on the open market for crucial items of inputs. This might also reflect, to a lesser extent, the changes effected in the methods of valuation of certain inputs and services, such as imputation of wages for family labour and calcu-lation of interest cost (about which more later). In what follows we will briefly survey the nature and trends in major variable physical inputs and their costs and then assess the extent of increase in paid-out costs during the post-reform period.2.1 Wages and EmploymentOn a casual perusal of these data in a different context, a recent study argues that “the cost of cultivation of crops has been increasing over the years because of increases in wage rate of labour, input prices and other managerial costs” [Narayana-moorthy 2007]. As a starting point, therefore, let us take the Table 1: Average Cost of Cultivation(Rs/hect) 1970-71 1970-81 1981-91 1991-2001 2001-05 2004-05Haryana 1,265.122,320.194,671.1514,556.29 23,271.11 24,179.23Madhya Pradesh 515.83 1,195.47 2,944.39 9,178.80 14,170.04 14,696.06Punjab 1,654.592,489.91 5,592.9215,964.32 23,135.18 24,197.48Rajasthan 2,311.124,302.0912,972.4620,122.7419,609.52Uttar Pradesh 1,408.62 2,180.95 4,800.97 12,426.80 18,730.03 20,812.56All states 1,417.05 2,100.57 4,291.48 12,669.42 19,269.48 19,910.52Table 2: Labour Use and Wage Cost 1970-71 1970-81 1981-91 1991-2001 2001-05 2004-05 (A) Human Labour Hours Per HectareHaryana 396.20484.61384.64338.60303.71293.37Madhya Pradesh 330.49 450.14 375.74 372.09 351.77 315.59Punjab 331.07440.01400.40326.84203.59163.56Rajasthan 686.62585.25567.14563.97504.02Uttar Pradesh 550.09 675.63 594.21 516.41 452.52 447.40All states 463.83 573.86 499.05 446.12 417.75 362.56 (B) Bullock Labour Hours Per HectareHaryana 134.19125.2150.0116.527.786.95Madhya Pradesh 157.29 186.82 127.51 55.74 43.28 34.55Punjab 104.7781.5821.783.711.831.53Rajasthan 209.16105.7449.81 19.82 16.84Uttar Pradesh 247.17 228.57 128.86 52.56 17.70 15.00All states 192.27 184.56 101.11 41.47 18.88 15.91 (C) Total Wage Cost (Rs/hectare)Haryana 232.31449.22699.892,618.123,971.973,866.63Madhya Pradesh 119.18 227.69 441.78 1,622.75 2,349.17 2,260.03Punjab 328.3430.48853.152,486.732,288.11,952.42Rajasthan 443.15792.853,363.174,969.334,618.55Uttar Pradesh 264.69 380.17 862.09 2,415.35 3,319.5 3,483.24All states 230.56 368.56 758.16 2,400.07 30,11.66 3,138.74 (D) Wage Cost as % of Operational CostHaryana 29.30 29.40 22.83 32.08 29.19 27.15Madhya Pradesh 41.38 33.19 26.53 30.14 27.41 25.35Punjab 35.42 28.23 22.64 29.07 20.1816.73Rajasthan 26.72 29.17 40.29 37.39 36.10Uttar Pradesh 29.72 26.15 26.55 31.59 28.39 26.13All states 30.72 27.69 25.57 32.05 26.37 25.66 (E) Casual Wage as % of Total WageHaryana 10.819.531.728.227.025.1Madhya Pradesh 27.4 32.1 41.1 34.7 32.6 34.3Punjab 25.938.752.455.848.551.2Rajasthan 21.617.813.416.718.9Uttar Pradesh 29.2 27.8 33.6 34.3 38.4 37.5All states 21.3 28.8 36.3 34.0 34.0 34.0wage-rate first. It is common knowledge that wage cost depends on the wage-rate and the labour-hours applied in cultivation. The labour use is a function inter alia of labour intensity in cultiva-tion, i e, the relative level of application of human labour to machine labour. Their relationship is generally negative, as machine-labour is a device that displaces human labour. The hours of human labour applied in cultivation depends also on the extent of bullock labour use. These two are positively correlated as more and more use of bullock labour would necessitate the requirement of more and more male farm labour. The employ-ment of manual workers in the cultivation of wheat, after record-ing some improvement until the mid-1970s, has been declining since. The decline was pervasive in all major wheat growing states though it was considerably steeper in Punjab, Haryana and Uttar Pradesh. In the mid-1970s, the wheat crop required on average 72 person-days of manual labour,2 ranging from 55-60 days in Punjab and Haryana to 85 days in Uttar Pradesh and Rajasthan. By 2004-05, this came down to 63 person-days in Rajasthan, 56 in Uttar Pradesh, 37 in Haryana and 20 in Punjab. Consistent with this, bullock labour use has almost disappeared from several parts of these states (Table 2). 
REVIEW OF AGRICULTUREjune 28, 2008 EPW Economic & Political Weekly126Nonetheless, the estimated average wage cost in wheat culti-vation accounts for the largest segment of all items of operational costs. It remained more or less stagnant at 26-27 per cent during the 1970s and 1980s. In the 1990s as a whole, however, there has been a substantial improvement in the average share of wages to 32 per cent. The share of wages had been actually declining in the latter half of the 1990s, and presently it stands at 26 per cent of the operational costs. During 2001-05, the decline in the share of wages was around 3-4 percentage points in Haryana, Madhya Pradesh, Rajasthan and Uttar Pradesh and as large as 9 percentage points in Punjab. Wage costs, as reported in the comprehensive scheme, include wages actually paid to casual and permanently attached labour and those imputed to family labour. The casual workers are engaged to do certain specific peak-season jobs, such as digging, transplanting, weeding, manuring, threshing, etc. Wages paid to them are relatively higher. Although attached labourers also do similar work, it is an established fact that, money wages paid to them are generally sticky and lower than casual wages. The payments accounted against attached labourers, however, include the value of doles given to them in the form of second-hand dresses, slippers, tents to take rest, helps extended in times of marriages of family members, etc. As far as family labour is concerned, small peasants use it mostly because of its abundance in relation to their operational area. The big farmers may not use family labour to do the work that other wage workers do. However, they draw the services of family members during peak periods when sufficient casual workers are not available. The family labour is also used in certain tasks requir-ing special care and skill like handling of sophisticated machinery. The wages imputed for such work should be larger than the market wages for casual labour. In the 1970s and 1980s, wages forfamilylabour were imputed at wages for attached labour. Subsequently, casual wages have replaced the basis to impute family labour charges. The all-state average of casual wage as a proportion of total wage, which, after increasing from 28.8 per cent in the 1970s to 36.3 per cent in the 1980s, declined to remain stable at 34 per cent thereafter. The decline was more pronounced in Haryana, Madhya Pradesh and Rajasthan, while in Uttar Pradesh it was apparently rising. In Punjab, where a large number of male members of rural families migrate abroad, wages for casual labour are relatively high and have been distinctly on the rise. Even then, in recent years, the share of casual wages in total wage cost has marginally declined in these two states. Two points emerge clearly from these data. First, average wages as a ratio of operational costs have not increased in the same rate as its other components did since the 1990s. Secondly, wages paid to hired-labour as a proportion of operational costs have remained stagnant throughout the post-reform period. 2.2 Chemical Fertilisers and PesticidesFertiliser charges account for the second largest component, after wages, in the operational costs of cultivation. Being a vital input contributing to productivity, there is a tendency among farmers to go on increasing the application of chemical fertilisers (along with high-yielding seeds), unmindful of the accompany-ing problems. For example, chemical fertilisers induce weeds, pests and nematodes, requiring the application of pesticides and weedicides. Too much water, or too little water, adversely affects the fertiliser response. For controlled water supply, farmers have to adopt sophisticated technology. When high-yielding seeds, which are short duration in nature, are used in this system of cultivation, inter-cultures have to be carried out in quick succession. This will not be possible unless farmers can mechanise different operations like the land preparation, planting, watering, weeding, harvesting and threshing. In other words, it is the level of application of chemical fertilisers that determines the level of mechanisation and the paid out costs in cultivation. From the “green revolution” period of the 1960s, there has been a tremendous increase in the use of chemical fertilisers in the cultivation of wheat, mainly in Punjab, Haryana and Table 3: Fertiliser Use and Fertiliser Charges 1970-71 1970-81 1981-91 1991-2001 2001-05 2004-05 (A) Fertiliser Use (Qtl/hectare) Haryana 44.5467.52138.41176.81195.22194.56Madhya Pradesh 5.15 21.25 42.10 76.04 89.67 94.00Punjab 107.93123.75174.78209.56227.02236.79Rajasthan 47.61 51.18 83.78 108.80102.21Uttar Pradesh 48.22 61.86 104.30 132.73 149.92 155.82All states 46.06 62.19 102.36 132.81 152.51 156.22 (B) Fertiliser Charges (Rs/hectare)Haryana 111.09 225.52 668.961,687.082,331.26 2,396.26Madhya Pradesh 4.14 111.33 224.67 834.85 1,174.75 1,241.75Punjab 208.6414.04886.262,029.851,891.21,180.96Rajasthan 235.42283.38842.741,397.481,329.48Uttar Pradesh 93.08 213.25 558.86 1,361.43 1,882.22 2,053.69All states 100.75 227.89 533.85 1,339.3 1,754.82 1,716.84Table 4: Cost of Insecticides(Rs/hectare) 1970-71 1970-81 1981-91 1991-01 2001-05 2004-05Haryana 0.001.0458.71 170.65 706.66802.79Madhya Pradesh 0.00 0.00 0.50 1.41 9.90 7.02Punjab 0.414.14108.07417.851,129.041,180.04Rajasthan 0.885.3313.9923.0021.92Uttar Pradesh 0.02 0.07 1.60 13.05 21.76 33.66All states 0.22 0.89 25.01 90.92 278.84 303.21Table 5: Machine Labour Charges 1970-71 1970-81 1981-91 1991-2001 2001-05 2004-05 (A) Total Machine Labour Charge (Rs/hectare) Haryana 54.34152.08602.071,689.353,240.543,728.49Madhya Pradesh 1.79 17.42 133.53 760.75 1,334.03 1,533.38Punjab 80.43210.67685.921,683.143,466.433,866.43Rajasthan 116.72358.381,214.651,907.752,063.90Uttar Pradesh 38.05 81.37 449.42 1,291.22 2,619.46 2,969.46All states 35.77 98.95 434.38 1,281.87 2,520.07 2,832.10 (B) Hired as % of Total Machine Labour ChargeHaryana 30.03 48.73 59.23 75.43 71.23 67.91Madhya Pradesh 30.73 76.00 74.39 87.83 91.57 91.40Punjab 50.58 36.67 44.29 60.57 64.58 64.32Rajasthan 98.8380.6595.5691.0588.82Uttar Pradesh 75.14 71.97 81.90 93.65 89.18 85.51All states 69.42 60.45 68.52 83.92 81.44 79.00
REVIEW OF AGRICULTUREEconomic & Political Weekly EPW june 28, 2008127latter,whileescalating the prices of fertilisers made it inaccessi-ble to smaller farmers. Fertilisers induce attacks of insect-pests, weeds and plant parasites. Weeds compete with wheat crops for space, light, water and fertiliser nutrients. According to Indian Council of Agricul-tural Research, while insect-pests, like termites, stem borer, weevils and wheat-amphids render grain unfit for human consumption, plant parasites (nematodes) cause huge crop losses every year. Weeds in a hectare of wheat field are estimated to remove at least 20 kg of fertiliser nutrients and reduce a minimum of 12 kg in yields [ICAR 1997]. Farmers using high doses of chemi-cal fertilisers have to incur large expenses on insecticides, weedi-cides and herbicides. With the emergence of private operators, costs on this account have mounted several folds. The all-state average expenses per hectare on insecticides, which worked out to less than a rupee in the 1970s and just over Rs 25 in the 1980s, shot up by 365 per cent in the 1990s and by as much as 1,115 per cent in the first half of this decade (Table 4, p 126). 2.3 MachineLabour The machine labour charge, which was less than 4 per cent of the operational cost in 1970-71, rose tremendously to reach above 23 per cent in 2004-05 and relegated fertiliser cost from its second position in the operational cost. This could be attri-buted mainly to three factors: first, with the growing diffusion of technology in wheat, there has been widespread mechanisa-tion of almost all farming operations like ploughing, irrigat-ing, weeding, harvesting and threshing. Initially, only larger holders in prosperous states of Punjab, Haryana and Uttar Pradesh adopted mechanised means of cultivation. At present, even farmers in poorer states depend increasingly on machine labour as the statewide trends in its per hectare costs given in Table 5 (p 126) demonstrate. Table 8: Interest Costs(Rs/hectare) 1970-71 1970-81 1981-91 1991-2001 2001-05 2004-05 (A) Interest on Working Capital Haryana 15.6235.8679.63194.89328.26347.47Madhya Pradesh 6.30 16.20 43.32 133.90 212.59 226.15Punjab 18.6238.8286.80209.13316.26331.55Rajasthan 38.3964.06168.93280.43277.12Uttar Pradesh 16.78 35.19 81.66 185.20 292.84 338.47All states 15.12 31.32 73.29 178.12 284.49 310.26 (B) Interest on Fixed and Working CapitalHaryana 108.02202.48509.0312,45.721,878.751,904.29Madhya Pradesh 50.56 110.90 306.95 1,026.09 1,476.57 1,416.03Punjab 107.66197.81538.811,334.892,081.812,157.81Rajasthan 129.69481.011,111.712,047.741,715.56Uttar Pradesh 84.22 161.02 424.43 1,413.88 1,711.36 1,768.33All states 93.34 166.32 434.08 1,070.38 1,781.36 1,773.36Table 9: Cost of Cultivation of Selected Purchased Inputs (Rs/hectare) 1970-71 1970-81 1981-91 1991-2001 2001-05 2004-05Haryana 277.82620.491,756.465,151.238,544.448,955.10Madhya Pradesh 45.95 245.36 791.97 3,096.23 5,009.28 5,614.27Punjab 425.21810.732,011.775,345.946,244.716,641.56Rajasthan 745.861,235.123,990.126,899.576,725.72Uttar Pradesh 304.52 564.12 1,619.01 4,684.38 7,765.39 8,900.33All states 292.85 560.58 1,502.64 4,451.95 7,110.15 7,673.71Uttar Pradesh. Estimates made by Frankel (1971) show that the quantity of chemical fertilisers applied in cultivation in Punjab (including Haryana) in 1960-61 was 3.23 kg per hectare. As against this, the all-state average use of fertilisers was 62 kg in the 1970s, 102 kg in the 1980s, 133 kg in the 1990s and 153 kg in the first half of the current decade. On a closer look, these figures exhibit two significant features. First, there are sizeable inter-state differences in the application of fertilisers. Punjab reported the highest level of fertiliser use in wheat of almost twice as high as the average for the remaining states, followed by Haryana and Uttar Pradesh and the lowest by Madhya Pradesh and Rajasthan (Table 3, p 126). Secondly, despite continued increase in absolute terms, the trend rate of fertiliser use in the cultivation of wheat has noticeably decelerated in the post-reform era. The growth rates of fertiliser application, measured from its all-states averages, had registered a compound growth rate of 4 per cent per annum in the 1980s. During the entire reform period, the rate of growth was just 2.1 per cent per annum.The fertiliser charges, on the other hand, have been steeply accelerating since the 1990s. The all-state average rates of ferti-liser costs, estimated at Rs 228 per hectare in the 1970s, rose to Rs 534 in the 1980s, to Rs 1,339 in the 1990s and to Rs 1,755 per hectare in the first half of the current decade. In the 1980s, when the fertiliser application increased at 4 per cent per annum, the increase in the rate of growth of fertiliser charges was only 4.5 per cent. During the reform period, when fertiliser application growth decelerated to 2.1 per cent a year, fertiliser cost growth accelerated to 6.2 per cent per annum. The large increase in fertiliser application and a similar increase in the unit cost of fertiliser per hectare, witnessed in the 1970s and 1980s, were the result of a consciously devised policy that the government followed: (1) supply of fertlisers to cultivators at subsidised “farm-gate prices”, (2) protection of the domestic fertiliser industry with a “retention price scheme” that provided an assured return, and (3) encouragement of fertiliser distribu-tionbytheindustrywith the extension of an additional “freight equalisation” subsidy [Bhalla 2007]. In contrast to this, the 1990s and after marked the decontrol of fertiliser prices, the entry of private operators in its distribution and replacement of the “retention price scheme” with a “new fertiliser policy”. The Table 6: Cost of Seeds(Rs/hectare) 1970-71 1970-81 1981-91 1991-2001 2001-05 2004-05Haryana 96.74147.71346.48687.38984.401,013.50Madhya Pradesh 88.13 145.80 280.27 732.79 947.20 997.84Punjab 65.35110.57238.78568.69802.35865.35Rajasthan 173.25310.46849.231,310.621,323.06Uttar Pradesh 99.51 142.77 295.51 721.12 1,122.32 1,233.33All states 92.44 141.83 289.16 710.63 1,040.17 1,109.92Table 7: Irrigation Charges (Rs/hectare) 1970-71 1970-81 1981-91 1991-2001 2001-05 2004-05Haryana 109.60196.46370.92901.741,819.861,906.19Madhya Pradesh 2.33 31.40 242.48 797.61 1,660.08 1,961.51Punjab 71.95109.96179.56321.80431.33461.72Rajasthan 259.89452.341,145.422,647.532,389.05Uttar Pradesh 88.80 151.52 319.42 939.16 2,005.66 2,630.44All states 83.99 134.56 297.98 832.29 1,724.32 2,034.49
REVIEW OF AGRICULTUREEconomic & Political Weekly EPW june 28, 2008129sector” credit to agriculture is no more available at the interest rate of 4.5 per cent, as in the 1970s and 1980s. Studies show that banks charge 15-18 per cent on agricultural credit granted to relatively poorer peasants. Further, once there is a default in repayments, these banks impose a penal interest of up to 24 percent on the compounded loan outstanding [George and Krishnaprasad 2006]. This is one reason why the interest cost on working capital has registered an unprecedented increase since the 1990s. Thirdly, with the institutional credit becoming difficult to obtain, farmers are forced to seek funds from outside the bound-aries of the formal financial systems. Conventionally, their main sources of borrowing were relatives and friends at zero interest, from traders against promise to sell their harvest, and from indig-enous moneylenders at exorbitant interest [Bardhan and Udry 2000]. When farm prices declined, it is only reasonable to believe that the possibilities of borrowing from friends and relatives in rural areas also disappeared. The other sources left are traders and indigenous moneylenders. The predominance of private moneylenders in credit transactions of farmers is yet another reason why the interest cost shot up by more than fourfolds during the post-reform era. 3 ConclusionsThe cost of cultivation data generated under the comprehensive scheme is unequivocal in establishing that coinciding with the economic reforms expenses on farming have soared to unprecedented heights. The foregoing analysis shows that not all items of costs have increased at the same pace. While fixed costs seemed to exhibit a gradual deceleration, operational costs have continued their relentless acceleration. The former reflects the prolonged depression in farm prices during the period under analysis and the resultant decline in private investments in agriculture. On the other hand, with the slash-ing of subsidies and the state agencies becoming redundant in distributing inputs, farmers naturally fell into the hands of unscrupulous private operators. This was instrumental in escalating operational costs. Our second major observation is that, during the post-reform period, there has been a steep decline in the labour hours applied in cultivation as also stagnation in casual wages. It is clear from this that the agrarian crisis in the post-reform period afflicted not only the cultivating households but also the entire agriculture-dependent population. During this period, there has been a deceleration in the rate of growth of chemical fertilisers applied in cultivation. Nonetheless, the corresponding rate of growth of fertiliser charges was three times higher than that of its physical application. The decontrol of fertiliser prices and the entry of private operators in distribution are evident from this. The break-up of variable costs available in the comprehensive scheme are averages of the owned and purchased inputs. It is difficult to ascertain from this the actual increase in paid-out costs of cultivation over time. We therefore desegregated the items of costs that the cultivators source exclusively or predomi-nantly from the market. These items are hired human labour, hired machine labour, chemical fertilisers, insecticides, irriga-tion charges and interests on short-term working capital. The average expenses of these six items in the cultivation of wheat were slightly above Rs 1,500 per hectare in the 1980s, varying from Rs 800 in Madhya Pradesh to Rs 2,012 in Punjab. It increased to Rs 4,500 per hectare in the 1990s and to Rs 7,700 per hectare in 2004-05, the last year for which we have the data, which is more than five times above the level in the 1980s. In Haryana and Uttar Pradesh, it was still higher (Table 9, p 127).A further exercise was carried out to see how these costs might have behaved in case the input subsidy regime of the 1980s continued as such during the post-reform period. Thus, we fitted a linear trend to the all-states average of the six items of costs for 1981-82 to 1990-91 and extended the same to 1991-92 to 2004-05. The extended trend line registers a linear growth rate of 8.2 per cent per annum. Compared to this, the actual increase in these cost items was 22.5 per cent and that of the WPI, a proxy for the general price level, 7 per cent. From this, two points become clear: First, even if the input subsidy regime continued the costs of cultivation of wheat, one of the most state protected crops, could have increased faster than the increase in the general price level. This shows the weakness of that regime. However, when that regime was in fact discontinued, the paid-out costs of crucial items of agricultural inputs increase at a rate that had not been seen earlier. Notes1 Afterwards, DES has published two more instal-ments, in 1996 and 2000, covering the data up to 1998-99. 2 The figures are estimated from Table 2 at the rate of eight human labour hours as one person-day. 3 For a long list of bank loans treated as priority sector credit for agriculture that the RBI has been enriching from 1994 to 2007, see Ramkumar and Chavan (2007).4 The comprehensive scheme introduces a new concept of “Cost-B1”, which is defined as Cost-A1 + “interest on the value ofowned fixed capital assets, excluding land”. The difference between Cost-B1 and Cost-A1 is the same as interest onfixed capital that it has been publishing separately. ReferencesBagchi, Amiya Kumar (2004): ‘The Common Minimum Programme’, Social Scientist, July-August, pp 3-16. Bardhan, Pranab and Christopher Udry (2000): ‘Fragmented Credit Markets’ inReadings in Devel-opment Economics, Oxford University Press, New Delhi, pp 76-93.Bhalla, G S (2006): Condition of Indian Peasantry, National Book Trust, India, New Delhi. – (2007): Indian Agriculture Since Independence, National Book Trust, India, New Delhi. George, J and P Krishnaprasad (2006): ‘Agrarian Distress and Farmers’ Suicides in the Tribal Districts of Wayanad’,Social Scientist, July-August, pp 70-85.Frankel, Francine R (1971):India’s Green Revolution: Economic Gains and Political Costs, Oxford Univer-sity Press, New Delhi. ICAR (1997): Hand Book of Agriculture, Indian Council of Agricultural Research, New Delhi.Kapre, B N (1974): Comprehensive Scheme for Studying the Cost of Cultivation of Principal Crops, Directo-rate of Economics and Statistics, Government of India, New Delhi. Narayanamoorthy, A (2007): ‘Deceleration in Agricultural Growth: Technology Fatigue or Policy Fatigue?’Economic & Political Weekly, June 23, pp 2375-79.Patnaik, Prabhat (2003): ‘Agricultural Production and Prices under Globalisation’ in his The Retreat to Unfreedom, Tulika Books, New Delhi, pp 198-210. Ram, G S (2001): ‘Data Relating to Cost of Cultivation Surveys’, in Chandrasekhar and Tilak (eds):India’s Socio-Economic Database – Surveys of Selected Areas, Tulika Books, New Delhi. Ramkumar, R and Pallavi Chavan (2007): ‘Revival of Agricultural Credit in the 2000s: An Explanation’, Economic & Political Weekly, December 29, pp 57-63. Sen, Abhijit and M S Bhatia (2004): Cost of Cultiva-tion and Farm Income, State of the Indian Farmer: A Millennium Study, Vol 14, Government of India), Academic Foundation, New Delhi. 
