Understanding Agrarian Impasse in

The key hypothesis of policy-makers during the 1980s was that raising tubewell density would trigger agrarian upsurge in Bihar as it did in Punjab, Haryana and western UP. The state did record high growth rates of cereal yields during the 1980s, higher than the national figures. However, this promising development could not be sustained in the 1990s, and cereal yields have stagnated since then. Based on fieldwork in eight villages of Bihar, the paper argues that, more than agrarian structure, the lack of adequate infrastructure and economic incentives has contributed to the agrarian stagnation in Bihar. The growth potential unleashed by the expansion of shallow tubewell irrigation has been constrained by (a) complete neglect of public sector investments in physical and institutional infrastructure and (b) unfavourable output to factor price ratios.

AVINASH KISHORE

I Gopalganj, and Madhubani in north Bihar. While Introduction Darbhanga and Madhubani are agriculturally least developed, Nalanda and Bhabhua are agriculturally the most developed fter a century-long stagnation, agriculture in eastern districts. , Saran and Gopalganj fall in between these experienced a turn around in 1980s with rapid ground- two extremes. This paper is an attempt to distill the lessons learned Awater development. Especially, and eastern during the fieldwork in these eight villages and put them in performed very well with growth rates in production perspective of the larger reality of agricultural development in becoming as high as 4-6 per cent during 1981-82 to 1991-92. Bihar. Agricultural production and productivity levels in the ‘middle state’ Bihar also witnessed a growth higher than the long-term III growth trend of the state [Pandey and Pal 2000]. Yet, agriculture Literature Review in the state grew at a slower pace, over a smaller base, for a shorter time after which the momentum was lost in spite of an impressive Till 1970s, the literature from agrarian studies were dominated expansion of tubewell irrigation. by discussion on issues related to agrarian structure and pro- This paper tries to explore the reasons for this relapse. The duction relations, groundwater development through expansion first section of the paper presents a review of literature available of tubewell irrigation and development of water markets became from the existing body of research on reasons for the persistent the thrust area after the pioneering RBI report on agricultural agrarian stagnation in eastern India. The second section explores productivity in eastern India [RBI 1984]. the latest trends in tubewell irrigation and discusses the structure Agriculture in Bihar has remained stagnant in spite of rich soil, of water markets and their productivity and equity implications abundance of easily accessible water and a rich peasant tradition.1 based on data from a primary survey. The third section of the This stagnation has been ascribed to several factors including paper discusses various macroeconomic factors which limit the state’s colonial legacy [Bharadwaj 1993; Banerjee 2002; farmers’ ability to leverage the newly created pump capital to Mearns 1999; Banerjee and Iyer 2002], ecological conditions increase crop productivity and raise incomes. This understanding [Ballabh and Pandey 1997; Ballabh and Sharma 1990], demo- is based on the knowledge of larger reality of agricultural scenario graphic pressure [Verma 1993] and most importantly, the land in Bihar and observations from field survey. The last and the tenure system and the agrarian structure it supports [Bhaduri concluding section discusses various alternatives for raising 1973; Prasad 1974]. These factors are believed to have impeded crops yield and farm incomes based on studies of well-performing the transition of Bihar’s agriculture from a semi-feudal to capi- farms in the region. talist production system – an essential condition for agricultural growth [Patnaik 1987]. II Inequitable land distribution and perverse land tenure relations Methodology Used in Study are considered to be the backbone of semi-feudal agrarian struc- ture in Bihar and therefore comprehensive land reforms is sug- The study involved an extensive review of literature available gested as the policy measure. Abolition of ‘zamindari’, redistri- on issues of agrarian structure and production relations, agricul- bution of surplus land, tenancy reforms and land consolidation tural growth, groundwater development and water markets in are four components of comprehensive land reforms [Ballabh eastern India and primary data collection using focused group and Walker 1992]. Among all Indian states, land reforms have discussions and interview schedules. been the worst failure in Bihar, even when it took the first The primary data were collected from field studies carried out initiative in this direction by abolishing the zamindari system in eight villages from six districts of Bihar in 2003. Two of our in 1953. While West Bengal could accomplish tenancy reforms villages are from Nalanda and Bhabhua districts of south Bihar through operation Barga and UP had considerable success in land while the rest six are from districts, viz, Muzaffarpur, Saran, consolidation [Pant 2004] both projects were abysmal failure in

3484 Economic and Political Weekly July 31, 2004 Figure 1: Shallow Tubewell Density in Bihar while land has become less important. This is reflected in the 2.00 9.83 declining share of land in the gross value of produce. But, eastern 8.22 9.19 0.00 7.02 5.61 India suffers from severe capital scarcity and existence of 8.00 4.26 6.00 2.32 3.16 monopolistic markets which exacerbate the existing inequality

NSA 4.00 2.00 and reinforce the outmoded structure. Machine-reforms, wherein 0.00 STWs/100 ha the government would help the resource poor farmers and land- 1986- 1987- 1988- 1989- 1990- 1991- 1992- 1993 87 88 89 90 91 92 93 94 less labourers to get ownership of agricultural implements so that the inequity in landholdings could be compensated by equity in Source: Minor Irrigation Census of India (1993). access to groundwater and technology, can bulldoze the ossified Figure 2: Density of Bores and Pumpsets (per 100 ha if NCA) agrarian structure and unfold a new era of rapid agricultural growth with relative equity and reduced poverty. They suggest 100 Bore Density Pump Density a government supported and subsidised capital formation in 80 private sector with special targeting on poor and landless to bring 60 about the change. 40 20 This was in line with the key recommendation of the widely 0 quoted RBI report on eastern India [RBI 1984] which also stressed mainly on a machine based approach of exploiting Jaale

Beraiin region’s abundant groundwater resources using shallow tubewell Rampur Fariyani Nariyaar

Machhahi technology to intensify irrigation and increase agricultural pro- Khutwaara Kurmi bigha Chawai takki ductivity, thereby reducing poverty. Rapid tubewell capital Source: Ballabh et al (2002) for Machhahi and Fariyani and Primary Survey formation through government subsidised projects like free- (2003) for other seven village. boring scheme became the corner stone of public policy in Bihar Bihar. For all practical purposes, the state has almost given up as well as in other eastern states. This marked a clear shift in on this project of land reforms. After the act for compulsory land the policy focus from land-based strategies to water and machine consolidation was passed in 1956 in Bihar; only 3 per cent of capital-based strategy for agrarian change in the region. Unlike the total cultivated land was reported to have been consolidated in the past, the tubewell capital formation was encouraged mainly till 1976 and in fact land consolidation department has now in the private domain and special targeted schemes like the formally been abolished in the state [Wilson 1999]. million wells scheme were devised for small and marginal farmers Changing the structure of landownership proved to be difficult of scheduled castes and scheduled tribes. because the zamindari system was highly institutionalised and Evolution of rental markets in pumpsets (often called water the dominant groups were very strongly entrenched in the state market in irrigation literature) helped overcome the problem of machinery itself. The political difficulties apart, Dandekar and lumpiness of investment and scale bias of the tubewell techno- Rath [cited in Bardhan and Rudra 1979] have argued that in an logies, earlier thought to be the main problems in adoption of overpopulated state like Bihar, it is futile to try resolving the these technologies in Bihar and rest of eastern India having stamp problem of rural poverty by re-distribution of land which is in size plots and high land fragmentation [Dhawan1982]. Water short supply as this will have to be achieved by enforcing markets facilitated access to groundwater of even the small and extremely low land ceilings which cannot be justified on sound marginal farmers. Rapid groundwater development took place in economic considerations or rational calculations. Similarly, Bihar powered by both policy push and private initiative. Density Hayami (1981) argued that in conditions of high and increasing of borewells and pumping sets had registered an exponential population pressure on land, land reform is likely to be infeasible growth over last two decades. Between the two minor irrigation and futile without reversing the trend of increasing return from censuses in 1986-87 and 1992-93, the pump density increased land relative to labour. He posited that the trend would continue fourfold in Bihar (Figure 1). Primary surveys carried out in nine as long as increase in supply of labour outpaces the increase in villages in six districts of Bihar in 2002 and 2003 suggest that labour demand and hence the best cure to the growing poverty the pump density has increased even more rapidly in the last 10 and inequality was to augment land by investing in irrigation years after the latest minor irrigation census (Figure 2).2 The systems and developing land saving and labour-using techno- surveys suggest that the density of borewells is much higher than logies. He considered the technocratic approach to be the only the pump density. In fact in these nine survey villages there was practical approach for agrarian transformation in this region; one borewell for every 2.5 ha of cultivated area. Such high since land redistribution is extremely difficult both politically borewell density is a response to high and increasing land frag- and administratively [Hayami 1981]. mentation in the state and this has been made possible by low cost of boring [Ballabh and Choudhary 2002) and portability IV of diesel pumpsets. From Land to Groundwater: The Paradigm Shift Water Market as an Embedded Exchange: Examining This has been the viewpoint of development economists, namely, Polanyist Argument Tushaar Shah and Richard Palmer-Jones also who argued that in rainfed low productivity agricultural system control over land Even though the density of groundwater extraction measures and labour were the key and as a result inequitable land distri- (WEMs) has rapidly increased, the actual role and potential of bution has been at the core of the current outmoded agrarian this pump-led market based reform strategy to bring equitable structure. However, in the modern land-saving, input-intensive growth in the precapitalist agrarian economy of Bihar has been agriculture, access to capital holds the key to agricultural growth, questioned by many researchers (for instance Wood (1995),

Economic and Political Weekly July 31, 2004 3485 Figure 3: Land Ownership vs Water Selling marginal and small farmer categories together, they have 55 per 500 cent of all pumpsets while they cultivate only 25 per cent of the 400 total land-owned by the sample farmers. 300 The IWMI survey showed that a significant number of the sub- 200 marginal category of farmers, having less than 0.5 ha of land, also 100 Hours/year 0 owned pumpsets. From the sample of 263 farmers, 32 sub-marginal Water Selling in 0 2 4 6 8 10 12 farmers had nine pumpsets among them while in marginal category Land Ownership (0.5 – 1 ha landholding) 66 farmers had 25 pumpsets among them. Thus, a large number of pumps are now owned by small and very Figure 4: Hours of Pump-Renting vs Landholding Size small farmers, who are compelled to sell water as irrigation surplus 250 from their own fields is not enough to make the pump investment

200 viable. They pump for profit and not power as suggested by Woods (1995) and Dubash (2000). These farmers resort to renting 150 R2 = 0.009 out pumpsets to improve the capacity utilisation which ensures

Year 100 access to groundwater to even inadequately capitalised farmers.

50 Interestingly, the analysis of sample survey data showed that even large farmers were selling water as much as the small and marginal 0 Hours of Pump Rent per farmers. The correlation between landholding size and hours of 0 5 10 15 20 25 water selling per year was positive, though weak (r = 0.18) and Landholding in Acres significant only at 90 per cent confidence level (Figure 3). This Figure 5: Hours of Pump-Renting vs Landholding Size probably indicates increasing acceptance of water selling by even the well off sections of the rural society in Bihar, a change from 200 the early days of groundwater development when it was a social 150 taboo [Clay 1974 as cited in Mukherjee 2004 and Wood 1995]. Wood argues that quite like in canal irrigation, there are tail- 100 enders in groundwater irrigation also. The pumpset and boring Year combinations irrigate the land of pumpset owner first; close non- 50 competitive kin second; favoured clients third; and then a non- 0 random group of non-intimate others. These non-intimate others, Hours of Pump Rent per 024681012who are generally the marginal and sub-marginal farmers of Landholding in Acres deprived castes, like tail enders in canal irrigation, have by design, more precarious access to pump rental services in availability, Chakravarty (2001) and Wilson (2002). They have used detailed timeliness and price of water. This constrains their ability to village studies to show how machine distribution has in fact realise the full potential of groundwater irrigation resulting in increased the inequity and strengthened the vice grip (rent extracting sub-optimal productivity outcomes. capacity) of already dominant landlords. They argue that the Wood’s analysis was based on a detailed study of one village rental markets for equipments are far from impersonal and context- in Purnea: one of the most underdeveloped districts of Bihar. neutral as assumed in neoclassical economics. Wood suggests We made an attempt to test if it was true across different regions that exchanges in rural water markets have non-economic motives of the state and if the picture has changed over last 10 years since and they are essentially linked to the social fabric and embedded Wood did his study. We correlated total hours of pumping/acre/ in social relationships. The imperfect, socially embedded ground- year of water buyers against their landhoding size and found the water markets have functioned to reinforce rather than relax the correlation between the two variables to be very weak and dependence, constraining the opportunity for many small farmers insignificant. The scatter plot (Figure 4) revealed that both very with non-favoured client status to increase their cropping inten- large farmers and very small farmers use much less hours of sity. As a productivity outcome, land is either increasingly coming irrigation per acre of their net sown area than the small and medium into the hands of adequately capitalised farmers or its potential farmers. The correlation was run again after excluding the large with the new technology is under-realised. Thus, these techno- farmers (> 10 acres of land) from the sample and found that the logical opportunities are socially constrained to serve non- correlation coefficient for this sample became moderately strong productive purposes. (0.525); positive and significant even at 99 per cent confidence This, however, does not seem to be true everywhere. In our level (Figure 5). This suggests that irrigation use intensity follows survey, we found access to groundwater to be more equitable Table 1: Distribution of Cultivable Land and Pumpsets among than the land distribution in Bihar. According to the last minor Different Land Class Sizes in Seven Villages of Bihar irrigation census (1992), 76.67 per cent of all mechanised shallow Farmer Size Marginal Small Medium Large tubewells (STWs) in the state are owned by marginal and small <2.5 acres 2.5-5 acres 5-10 acres >10 acres Total farmers. This is much more than their share in the total cultivable Pump-owners (PO) 34 39 31 30 134 land in the state. This was corroborated in a recent primary survey Land (acres) 55.19 157.94 239.03 613.02 1065.18 of 263 farmers carried out by IWMI in seven villages in six Land/PO 1.62 4.05 7.71 20.43 7.95 Non-owners 64 29 22 14 129 districts of both north and south Bihar (Table 1). As evident from Land (acres) 66.96 118.66 158.57 187.8 531.99 the table, marginal farmers in the sample had only 7 per cent Land/PO 1.05 4.09 7.21 13.41 4.12 of the total land cultivated by the sample but they accounted for Pumps/100 acre 27.83 14.10 7.80 3.75 8.39 25 per cent of all pump-owners in the sample. If we take both Source: Primary Survey, 2003.

3486 Economic and Political Weekly July 31, 2004 Figure 6: Installation of Pumpsets over Last Two Decades over last decade. It would be interesting to point out here that 76 more than half of all pumpsets the samples selected (76 out of 80 137) were purchased after 1995, the year when Shah and Ballabh carried their study in Muzaffarpur (Figure 6). 60 42 40 Groundwater Irrigation Intensity in Rice-Wheat Systems 19 20 The irrigation pattern of pump-owners was compared with that of renters for paddy and wheat. The purpose of this comparison 0 was to see if the differential pumping patterns had any significant Till 1985 1985-95 After 1995 impacts on the productivity outcomes. In paddy, the most im- portant crop for Bihar, there was no significant difference between Figure 7: Annual Pumpage Range (Rice + Wheat) ha pump-owners and renters in terms of hours of pumping/irrigation per acre and crop yield (Table 3). Also, the correlation between 32.87%32.87 35.00 per cent hours of irrigation/acre of paddy and the yield obtained was found 30.00 21.30 to be weak and insignificant at the state level. This is under- 25.00 per21.30% cent 15.74 14.35 standable as rainfall in the survey year was timely and adequate. 20.00 14.35% per15.74% cent per cent 7.87 A recent collaborative study by Rajendra Agricultural University 15.00 7.87% Farmers 2.78 per cent 4.63 (RAU) and International Rice Research Institute (IRRI) in eight 10.00 2.78% per4.63% cent 0.46

Percentage of per cent 5.00 per0.46% cent villages of north Bihar also showed similar results with irrigation 0.00 coefficient being weak and insignificant in the production func- 0-25 25-50 50-75 75-100 100- 125- 150- 230 tion plotted for paddy [Singh, Paris and Luis undated]. 125 150 175 In the case of wheat, a weak but positive and significant Hours of Pumping/Year/ha correlation was found between pump-ownership and hours of an inverted ‘U’ shaped pattern vis-à-vis landholding size. Small irrigation/acre provided to the crop. Hours of irrigation and yield and medium farmers (cultivating 2.5 to 10 acres of land) practice also had weak but positive and significant correlation between the most input intensive agriculture while the sub-marginal and them. But the correlation between pump-ownership and yield was large farmers have diversified away from agriculture giving insignificant. A closer look at the data suggested that though the relatively lower attention to farming on their own land. incidence of pump-ownership was higher among the farmers in Wood’s analysis of pumpset as an abettor of inequality is not south Bihar, wheat yields were lower there. acceptable for two reasons: first, it was found that diesel pumps This dichotomy was noticed during IWMI’s fieldwork also. were the most widely distributed mechanical input across various Farmers in villages of south Bihar invest greater amount of resource landholding classes and castes in all the villages and second, no and effort in kharif crop while rabi crops were more important significant and systematic difference was noticed in the pump rentals paid by different categories of farmers. Also, none of the Table 2: Increase in Cost of Agricultural Inputs in Nariyaar, 129 water purchasers in the sample complained about any dif- Muzaffarpur between 1995 and 2003 ficulty in accessing pumpsets on rent. This is the reason why Item 1995 (1) 2003 (2) Ratio (2/1) except the severely waterlogged areas, almost all land cultivated Diesel (Rs/liter) 7.85 24 3.05 by the sample farmers was irrigated. Irrigation (Rs/hr) 22-25/10-12 50-60 2.5 Over the years as pump densities have increased, water markets Tractor (Rs/ha) 400 825-1100 2.06 have become broader, deeper and more efficient and monopoly Team (Rs/day) 40 60 1.5 Thressor 16th share 10th share 1.6 rents have gone down significantly as is evident from the declining Sowing Rs 20 + Meal Rs 40 + Meal 2 water to diesel price ratio in the area. For example, when Shah Spade-work Rs 20.00 Rs 40 2 and Ballabh studied the six villages in 1995, diesel price was Weeding Rs 8-10/half day Rs 15/half day 1.5 around Rs 7.50 – 9.05 while water price ranged from Rs 20 to Storage (Rs/qtl) 65 140 2.15 30 (Table 4 in Shah and Ballabh 1997:A-186). Seven years later, Source: Fieldwork Report of IRMA Students involved in the Six Village Study when one of the six villages (Nariyar) was revisited, it was found Coordinated by Shah and Ballabh (1997) for the data of 1995 and that while diesel price had increased more than threefold to authors’ primary survey for data of 2003. around Rs 25 per litre, water price in the village had increased only twofold from Rs 22-25 to Rs 50 per hour. The water to diesel Table 3: Crop-Economics of Farmers Obtaining Average price ratio had declined from 2.80 – 3.18 to less than 2 now. and High Yield of Wheat The timeline data on water to diesel price ratio in Table 4 of Wheat Modal Yield (N = 23) High Yield (N = 5) Shah and Ballabh (1997) study reveals that in 1990s, one rupee Yield (Q/ha) 20 34 increase in price of diesel used to cause several times higher Gross return 11,000 18,700 increase in pump-rent in the market. IWMI’s field studies in Bihar Seeds 1,339.2 1,573.75 suggest that the trend has now reversed. Several water buyers Fertilisers 2,616.25 2,780 Tractor 1,965 2,600 admitted to us that pump rentals had not increased as frequently Irrigation 1,756.25 2,250 as diesel prices after the recent enforcement of open market price Labour cost 1,375 2,337.5 regime on diesel. This indicates a significant increase in the buyer Total cost 9,051.7 11,541.25 power in the market which might have occurred due to increased Cost/quintal 452.58 339.44 competition in the market induced by increase in pump density Source: Primary Survey, 2003.

Economic and Political Weekly July 31, 2004 3487 Figure 8: Pumping/ha of Wheat (Hours) of six villages in Muzaffarpur. Water buyers in IWMI’s sample also use lesser hours of groundwater irrigation but use it more 50.00 44.29%44.29 per cent productively to reap same levels of yields as the pump owners. 40.00 30.00 Thus, it can be inferred that water markets have become more per30.00% cent 30.00 widespread and dense over the years facilitating almost universal access to groundwater in Bihar. Markets have also become more 20.00 15.24%15.24 per cent efficient as evident from declining monopoly rents over the years. 5.71 4.76 10.00 per5.71% cent per4.76% cent Even though groundwater irrigation has become widespread and

Percentage of Farmers market exchanges have become more equitable, the intensity of 0.00 0-25 25-50 50-70 70-90 100+ groundwater irrigation has remained quite low. Farmers continue to economise on their irrigation cost which has risen sharply due Figure 9: Yield Range (Rice + Wheat) in Tonnes/ha to rise in diesel prices. The need for upfront cash payment for

34.42 diesel pump irrigation also forces cash-starved farmers to under- 40.00 per34.42% cent irrigate. The economisation of irrigation cost is clearly evident from the annual pumpage data of all categories of farmers in our sample 30.00 21.86 per21.86% cent 17.21 (Figure 7). Three-fourths of sample farmers in Bihar provide less 16.28 17.21% 20.00 per16.28% cent per cent than hundred hours of irrigation per hectare of their land under 6.51 per6.51% cent 3.26 intensive rice-wheat cropping system. Assuming an average Percentage 10.00 per3.26% cent 0.47 discharge of 35 m3/hour, the depth of irrigation works out to per0.47% cent 0.00 be less than 0.35 metres for 71.3 per cent farmers. This is quite 0-5 5 to 6 6 to 7 7 to 8 8 to 9 9 to 10 10+ low for a double cropped land even in a high rainfall area like Bihar. If we look at wheat, the main rabi crop in the state, it can be found Tonnes/ha that half of the farmers provide less than 50 hours of irrigation Figure 10: Gross Value of Produce/ha per hectare of their cropped area (Figure 8). This translates to an average irrigation depth of just 0.175 metres which is much lower 40.00 35.19 per35.19% cent than the irrigation level recommended for the crop in the state. 35.00 This over-economy of groundwater use affects the yields of 30.00 23.15 21.76 23.15% the crops received in the state and precludes further intensi- 25.00 per21.76% cent per cent 20.00 fication of cultivation on land in spite of high population pressure. 15.00 9.25 Figure 6 shows that 40 per cent of the farmers in Bihar receive 9.25% 7.87%7.87 Percentage per cent 10.00 0.00 per cent 3.70 per3.70% cent less than 6 tonnes of cereal yield (rice + wheat) per hectare of 5.00 per0.00% cent net sown area. Almost three-fourths of farmers (72. 56 per cent) 0.00 receives less than 7 tonnes of cereal yield per hectare of net sown 0-10'000 10-20000 20-25000 25-30000 30-35000 35-40000 40-45000 area (Figure 9). Combined yield of rice and wheat of one-fifth of farmers in our sample was less than a typical Punjabi farmer’s Figure 11: Comparative Performance of Rice Yields rice yield alone. Accordingly, gross returns from agriculture are in Bihar, West Bengal and UP also extremely low. While one-third of farmers in the sample earned Rs 25 to Rs 30,000 per hectare of their net sown area 2500 under rice-wheat system, another one-third realised less than 2000 Rs 25,000. Combined per hectare gross value of produce at farm- 1500 gate prices was less than Rs 35,000 for 90 per cent of the farmers 1000 in the state (Figure 10). The average figure for the whole sample

500 is around Rs 28,000/ha/year. Considering the land-man ratio of 12.78 persons/ha in rural Bihar, the gross value of produce/capita/ 0 year amounts to just around Rs 2,000. This is one of the important reasons for high incidence of rural 1980-811981-821982-831983-841984-851985-861986-871987-881988-891989-901990-911991-921992-931993-941994-951995-961996-971997-981998-99 poverty in Bihar, the only state in India where the absolute number Bihar UP West Bengal Bengal of poor people below poverty line has registered an increase Source: Agriculture, CMIE, November 2000: 69 between the last two rounds of sample surveys of NSSO [Ahluwalia 2001]. to the farmers in the villages of north Bihar. Separate analysis carried out for the four villages of north Bihar showed that yield V difference between pump-owners and water buyers was small Why Tubewell Irrigation Development Is Not but significant at 95 per cent confidence level while the difference Enough to Sustain Growth? in hours of irrigation given per unit area of wheat was significant even at 99 per cent confidence level. Correlation coefficients Agricultural growth remains low in Bihar in spite of increased between ownership and yield and ownership and hours of ir- pump density and access to irrigation. Growth fuelled by ground- rigation were also found to be positive and significant though water development has been short-lived and much less significant quite weak (0.176 and 0.281 respectively). Except in case of than that in the neighbouring West Bengal, Eastern UP and even wheat in the four villages of north Bihar, the results of our survey Bangladesh with which it shares its history and ecology (Figure 11). are similar to what Shah and Ballabh (1997) found in their study Whatever production increase has been realised had been offset

3488 Economic and Political Weekly July 31, 2004 Figure 12: Net State Domestic Product of Agriculture of Bihar of Agricultural Cost and Prices (CACP), let alone the minimum at 1993-94 Constant Prices support price offered by . Post-1991 increase in oil prices, reduction in subsidy on diesel 1800000 and fertiliser and slackening food prices in both national and 1600000 international markets have together created a situation where 1400000 terms of trade have progressively moved against agriculture

1200000 and the profitability of food production has gone down in spite of a modest increase in crop productivity in Bihar. 1000000 According to Harriss (1993), it was favourable factor price 800000 ratio (in conjunction with availability of a new high yielding 600000 variety of rice (IR 36)) that facilitated agricultural take-off in 400000 West Bengal. Now, depressed foodgrain prices and increased cost

NSDP Agriculture in Rupees 200000 of cultivation are producing a reverse effect, suppressing farmers’ margins and motivation to intensify inputs in agriculture, result- 0 ing in relapse of stagnation in state in late 1990s after a short

1960 1962 1964 1966 1968 1970 1972 1974 1977 1979 1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001 period of growth in the 1980s. Similar problems have been reported in Bangladesh also when rise in diesel price following Source: Directorate of Economics and Statistics, Government of Bihar the Gulf war resulted in a fall in boro production in subsequent by population growth and as a result, per capita food availability years [Palmer-Jones 1999]. Decline in rice price-fertiliser and labour productivity have remained at the same level as in price ratio has been associated with slowdown in agricultural the 1960s (Figure 12). growth in Bangladesh [Shahabuddin 1999] and Assam [Hussain What explains the persistence of agrarian stagnation in Bihar and Kalita 2004]. In an analysis of slowdown of agricultural in spite of rapid pump capital formation? Why have farmers in growth in Bangladesh, Palmer-Jones (1999) posits that the trends Bihar failed to leverage access to groundwater to increase pro- of rice prices in relation to agricultural input prices including ductivity, intensity and income from agriculture? agricultural wages have been an important determinant of agricultural growth. Declining Output-Input Price Ratio Decapitalisation of the Rural Areas This issue was raised repeatedly in focus group discussions with farmers during the fieldwork and in all the eight villages Reduced rate of public capital formation and virtual collapse covered in the study the response was unanimous: sharp increase of infrastructure (such as rural roads, power supply, major and in diesel prices is the principal factor affecting agricultural medium irrigation systems, sugar mills, etc) has further worsened growth. Today diesel has replaced muscle as the main motive the situation. Empirical results for a period 1980-98 of an all power in agriculture and its price has increased more than three India level study suggest that gross domestic product of agri- times in last eight years (from 1995 to 2003) resulting in a culture (GDPA) is strongly influenced by: (a) capital formation corresponding increase in costs of irrigation, land preparation in agriculture; and (b) terms of trade [Gulati and Bathila 2001]. and threshing (Table 3). Since diesel is available only on cash Traditionally, public capital formation in has payment, there has been a substantial increase in the cash been mainly in form of major and medium irrigation systems, outlay of agriculture with increased dieselisation and rising whereas in the 1970s and 1980s the focus shifted towards diesel prices, resulting in a resource squeeze in these credit- expanding well irrigation by providing increasing amount of starved villages. electricity to agricultural sector [Dhawan 1996]. As subsidies were reduced, there has been a substantial increase In the last two decades, public capital investment in agriculture in price of chemical fertilisers, For example, the retail price of in Bihar has concentrated mainly in increasing the density of urea remained stable at Rs 2.50/kg. throughout the 1980s while shallow tubewells and pumpsets through credit and subsidy in last 10 years the price has increased more than two times to schemes like Free Boring Scheme, Million Shallow Tubewell about Rs 5.30/kg. There has been a much steeper rise in prices Project, On-farm Water Management Project, IRDP loans, etc. of DAP and potash. Agricultural wages have also doubled up However, performance in major and medium irrigation systems and between 1995 and 2003 and they have become progressively and rural electrification which would have given long-term cost more monetised. There has been a sharper increase in peak-season advantage to farmers in the state has been miserable. The actual wage rates than the non-peak season wage rates. Increased out- irrigated area by major and medium schemes has gone down from migration in peak seasons is cited as the reason for this increase 2.15 million hectares in 1990 to 1.58 million hectares in 1997 even during a time when land owners’ returns from agriculture [Sharma 1998]. Nearly 21,000 villages are yet to be electrified have taken a beating. in the state, while by Bihar State Electricity Board’s own ad- Depressed foodgrain prices and absence of effective price mission, 18,000 of the electrified 47,000 villages have become support were also considered to be equally responsible for de-electrified due to various reasons.3 The per capita annual agricultural stagnation. It was pointed out that while cost of all power consumption is as low as 22 kWh in north Bihar and inputs have increased manifold, food prices have gone down in 87 kWh in south Bihar [Sharma 1998]. Thus, when private cost real terms. The average farm-gate price of paddy and wheat has of irrigation was falling in other states of India on account hovered around Rs 350/quintal and Rs 525/quintal respectively of free and subsidised power supply in 1980s and 1990s, in the last three seasons. The harvest price of paddy is lower farmers in Bihar incurred increasingly high cost for irrigation. than even its C2 cost of cultivation as estimated by Commission Poor infrastructure means inefficiencies and higher costs of

Economic and Political Weekly July 31, 2004 3489 production, marketing, storage and processing making agricul- production and increasing profit margins. IWMI research in north ture uncompetitive in the state. China shows that it is through relentless intensification and raising crop yields that Chinese farmers have retained the Absence of ‘the New Agrarianism’4 in Bihar viability of their farming in the face of declining global foodgrain prices and rising input costs and local taxes [Shah, Giordano Some researchers argue that phenomena like rapid decline in and Wang 2004]. rural infrastructure, absence of procurement support and poor This strategy, however, will catch on only if production and performance of rural credit schemes may well be the problems price risks are low and farmers have access to suitable crop of demand rather than supply [Wilson 1999 and Shah, personal varieties which respond adequately to input intensification. Paddy communication]. Kalpana Wilson argues that de-electrification crop faces risk from flood in large parts of north Bihar. Lack and state apathy by themselves cannot explain the stunted growth of suitable varieties for poorly drained low lands called ‘chaunrs’ of capitalism in Bihar. Faced with similar problems, farmers in which form a significant proportion of the total cultivable land other parts of country have responded very differently. Remu- in the region is also a big stumbling block. This is the reason nerative prices, reliable power supply, free electricity and writing why south Bihar where problems of flood and drainage are not off of farm loans have been the rallying points of peasant very serious (except in Tal) level of input use in paddy movements in the capitalistically developed rural areas of the and its yields are much higher when compared to flood-prone country like western UP, Punjab, Haryana, , Maharashtra north Bihar. and Karnataka where peasants of varying economic status came In case of wheat, the main rabi crop in Bihar, almost all our together to extract concessions from the state [Gail Omvedt respondents complained that the crop was not very responsive quoted in Harriss 1992]. This marked a clear shift from the to input intensification and seed up-gradation. Agronomists historical pattern of peasant movements which had almost always agree to this point. The problem lies in the late sowing of wheat targeted at intra-rural class relationships [Hariss 1992]. But there in Bihar which forces early maturity, resulting in lower yields. has been complete absence of any such movements in Bihar Field experiments carried out by the ICAR scientists in Direc- which shows a lack of demand for such services even if the need torate of Water Management Research, have shown that is not lacking. Wilson (1999) argues that this is mainly because one month delay in wheat sowing from its recommended time the peasant capitalist development which started in parts of Bihar of November 19-25, results in about 53 per cent reduction in under favourable conditions of 1970s and 1980s could not be yield [Singh et al 2002]. Such delay is common in Bihar and supported by the underlying agrarian structure and the nature of in fact in many areas the sowing goes on till as late as second the state. week of January. The delay is caused by late sowing (and hence late harvest) of paddy crop. At present farmers wait for monsoon VI rains to sow paddy and transplantation is carried out generally Towards Breaking Agrarian Impasse in the third week of July. ICAR studies show that advancing sowing of rice by a month will help to realise higher yields of Input Intensification for Improving Efficiency and both rice and wheat. However, this requires irrigation for the crop Income: A Possible Option for Farmers in pre-monsoon period. Therefore, increasing groundwater use is still the key to agrarian development in Bihar [Singh et al 2001]. With all these structural and macroeconomic constraints facing Prohibitively high cost of irrigation is the biggest hurdle in large- agriculture in Bihar, what are the options available for an in- scale adoption of this proven strategy of yield enhancement in dividual farmer to expand his production and profit margins? rice-paddy systems of Bihar. To understand this, an analysis of best practices in agriculture was carried out in one of our study villages, in which the VII agricultural practices and crop-economics of a small group of Policy Choices for State farmers who achieved the highest productivity levels were studied and compared with that of the modal group of farmers. The Rural electrification is one state initiative which can provide comparison revealed two important points. First, the yield ob- major boost to agriculture in Bihar. It can bring down the cost tained by most of the farmers was just enough to cover their input of irrigation and improve the working capital situation of farmers. costs and there was hardly any surplus left even for consumption, It can also trigger growth in storage and processing infrastructure let alone capital formation. This resulted in continued poverty which will permit value addition and the much needed crop and stagnation of agrarian economy. Also, with just 28 per cent diversification in the state.5 Cheaper access to irrigation with increase in cost of cultivation, the productivity increased by 85 electricity will also encourage farmers to bring larger areas under per cent and the net returns increased by more than 250 per cent. summer crops where higher yields can be obtained under more This means that most of the farmers in Bihar are practising controlled conditions. what Tushaar Shah calls ‘cost-covering agriculture’ and from this Behaviour of producer prices has been another important point if the crop yields are raised even by 20-25 per cent, the impediment to agricultural growth in Bihar. The high level net incomes of farm households would increase by 60-70 per Committee Report on Food Policy notes that the producer prices cent making a significant dent in rural poverty in the state. Second, were higher in this region than Punjab-Haryana in the early 1980s farmers who made more intensive use of inputs and incurred but became lower in the early 1990s as surpluses appeared (http:/ higher costs towards cultivation were more efficient and /fcamin.nic.in/hlc_contents.htm). The combined pressure from competitive producers with much lower per quintal cost of subsidised sales by Food Corporation of India (FCI) under its cultivation. This shows that input intensification to increase Open Market Sales Scheme (OMSS) and local seasonal gluts yields is one possible way of reducing per quintal cost of results in producers getting prices well below the actual economic

3490 Economic and Political Weekly July 31, 2004 cost of production in Bihar. Poor infrastructure and under- Gulati, Ashok and Seema Bathila (2001): ‘Capital Formation in Indian developed primary agricultural marketing network further worsen Agriculture: Revisiting the Debate’, Economic and Political Weekly, 36(20), pp 1697-1708. the problem of farmers by increasing the cost of marketing and Hagen, James R (2000): The Raw and the Simmered: Environmental Contexts price fluctuations. Unless this is addressed by extending procure- of Food and Agrarian Relations in the Gangetic Plain, Power, Agrarian ment operations more effectively in the state, there will be Structure and Peasant Mobilisation in Modern India: A Symposium, May insufficient incentive for the sustained yield increase. EPW 23-25, University of Virginia. Harriss, John (1992): ‘Does the ‘Depressor’ Still Work? Agrarian Structure and Development in India: A Review of Evidence and Argument’, Address for correspondence Journal of Peasant Studies, 19 (2), pp 189-227. [email protected] – (1993): ‘What Is Happening in Rural West Bengal? Agrarian Reform, Growth and Distribution’, Economic and Political Weekly, 28(24), pp 1237-47. Notes Hayami,Yujiro (1981): ‘Agrarian Problems of India from an Eastern and South Eastern Asia Perspective’, Institute of Economic Growth, New . 1 Very few people would know that transplanted, wet rice was first Hussain, T A and R R Kalita (2004): ‘Shallow Tubewell Irrigation in Assam: developed and used in the world in the amply watered fields below the Bountiful Production but Forlorn Farmers’, unpublished paper. Rajgir Hills in during the sixth century BC [Hagen 2000]. Mearns, Robin (1999): Access to Land in Rural India: Policy Issues and 2 Data for two of these nine villages (Machhahi and Fariyani) have been Options, World Bank Policy Research Working Paper 2123. obtained from Ballabh et al(2003). Palmer-Jones, Richard (1999): ‘Slowdown in Agricultural Growth in 3 http://www.bihar.nic.in/Depts/Energy/rural_electrification.htm Bangladesh: Neither a Good Description nor a Description Good to Give’ 4 The word New Agrarianism was used first by L I Rudolph and S H in Ben Rogaly, Barbara Harriss-White (eds), Sonar Bangla: Agricultural Rudolph in their famous book: In the Pursuit of Lakshmi: The Political Growth and Agrarian Change in West Bengal and Bangladesh, Sage Economy of the Indian State’ first published in 1987 by Chicago Publications, New Delhi. University Press. Pandey, Sushil and S Pal (2000): ‘The Nature and Causes of Changes in 5 Among all states in India, Bihar has the highest share of its gross cultivated Variability of Rice Production in Eastern India: A District Level Analysis’ area under cereals (~ 90 per cent) and it has increased in last 10 years. in S Pandey, B C Barah, R A Villano and S Pal (eds), Risk Analysis and Management in Rainfed Rice Systems, NCAP/IRRI Workshop on References Risk Analysis and Management in Rainfed Rice Systems, September 21-23, 1998, New Delhi. Ahluwalia, Montek S (2000): ‘Economic Performance of States in Post- Prasad, Pradhan H (1974): ‘Reactionary Role of Usurer’s Capital in Rural Reforms Period’, Economic and Political Weekly, 35 (19), pp 1637-48. India’, Economic and Political Weekly, 9 (32-34), pp 1305-08. Ballabh, Vishwa and B M Sharma (1992): HYV Adoption, Productivity Gaps (1984): Agricultural Productivity in Eastern India, and Production Adjustment Mechanisms in Flood-prone Areas of Uttar Reserve Bank of India, , India. Pradesh: A Synthesis, Research Report 5, IRMA, Anand Shah, Tushaar and Vishwa Ballabh (1997): ‘Water Markets in North Bihar: Ballabh, Vishwa and Kameshwar Choudhary (2002): ‘Groundwater Six Village Studies in ’, Economic and Political Development and Agriculture Production: A Comparative Study of Weekly, 32 (52), pp A-183-A-190. Eastern Uttar Pradesh, Bihar and West Bengal’, Draft Discussion Paper Shah, Tushaar, Mark Giordano and Jinxia Wang (2004): ‘Water Institutions Presented in 2nd IWMI-Tata Annual Partners’ Meet, January 27-29. in a Dynamic Economy: What Is China Doing Differently from India?’ Ballabh, Vishwa and Sushil Pandey (1999): ‘Transitions in Rice Production Paper presented in 3rd IWMI-Tata Annual Partners’ Meet, Anand, Systems in Eastern India: Evidence from Two Villages in Uttar Pradesh’, February 17-19. Economic and Political Weekly, 34(13), pp A111-A116. Sharma, I D (1998): Report on Concurrent Evaluation of Integrated Rural Ballabh, Vishwa and T S Walker (1992): ‘Land Fragmentation and Consolidation Development Programme: 5th Round: July 1995-June 1996, A N Sinha in Dry Semi-arid Tropics of India’, Arth Vijnana, 34(4), pp 363-87. Institute of Social Studies. Banerjee, Abhijeet V (2002): ‘Who Is Getting the Public Goods in India: Singh, R K P, Thelma R Paris and Joyce S Luis (2002): Impact of Labour Some Evidence and Some Speculation’, unpublished. Out-Migration on Rice-Based Household Economy and Gender Roles Banerjee, Abhijit and Lakshmi Iyer (2002): ‘History, Institutions and Economic in Bihar, India, Draft Report of a Collaborative Project of Rajendra Performance: The Legacy of Colonial Land Tenure Systems in India’, Agricultural University, Bihar and International Rice Research Institute, http://info.worldbank.org/etools/docs/voddocs/183/367/ Philippines. land_tenure_india.pdf Singh, Sita Ram, U S Gautam, Ujjwal Kumar, A Rahman and S K Sinha Bardhan, Pranab and Ashok Rudra (1979): ‘Labour Employment and Wages (2001): Groundwater Development to Enhance Surface and Rain Water in Agriculture: Results of a Survey in West Bengal’, Economic and Utilisation and Agricultural Productivity in Southern Bihar, ICAR- Political Weekly, 15 (45 and 46), pp 943-49. IWMI Ground Water Policy Initiative, Policy Dialogue on Groundwater Bhaduri, Amit (1973): ‘A Study in Agricultural Backwardness under Semi- Management, CSSRI. Feudalism’, Economic Journal, 83, pp 120-37. Shahabuddin, Quazi (1999): ‘Agricultural Growth Performance in Bangladesh: Bharadwaj, Krishna (1993): ‘Regional Differentiation in India’ in Binoy A Note on the Recent Slowdown’ in Ben Rogaly, Barbara Harriss-White N Verma (ed), Agrarian Relation Institutional Change and Colonial (eds), Sonar Bangla: Agricultural Growth and Agrarian Change in West Legacy, Volume (III), Ashish Publishing House, New Delhi. Bengal and Bangladesh, Sage Publications, New Delhi. Chakravarty, Anand (2001): ‘Caste and Agrarian Class: A View from Bihar’, Stiglitz, Joseph E (1989): ‘Rational Peasants, Efficient Institutions and a Economic and Political Weekly, April 28, pp 1449-62. Theory of Rural Organisation: Methodological Remarks for Development Clay, E J (1974): ‘Innovation, Inequality and Rural Planning: The Economics Economics’ in Pranab Bardhan (eds), Economic Theory of Agrarian of Tubewell Irrigation in Kosi Region, Bihar, India’, PhD Thesis, University Institutions, Clarendon Publishers, Oxford. of Sussex cited Mukherji, Aditi (2004), Groundwater Markets in Swain, Mamata (1999): ‘Agricultural Tenancy and Interlinked Transactions- Ganga-Meghana-Brahmputra (GMB) Basin: A Review of Literature. I: Neoclassical and Marxist Approaches’, Economic and Political Weekly, Paper presented in 3rd IWMI-Tata Annual Partners’ Meet, Anand, 34 (37), pp 2657-66. February 17-19. Wilson, Kalpana (1999): ‘Patterns of Accumulation and Struggles of Rural Dhawan, B D (1982): Development of Tubewell , Agricole Labour: Some Aspects of Agrarian Change in Central Bihar’, Journal Publishing Academy, New Delhi. of Peasant Studies, 26 (2-3), pp 316-54. – (1987): ‘Impact of Irrigation on Farm Economy in High Rainfall Areas: – (2002): ‘Small Cultivators in Bihar and New Technology: Choice or Kal Project’, Economic and Political Weekly, 23(53). Compulsion?’ Economic and Political Weekly, March 30, pp 1229-38. – (1996): ‘Relationship in Public and Private Investments in Indian Agriculture Wood, Geoff (1995): ‘Private Provision after Public Neglect: Bending with Special reference to Public Canals’, Indian Journal of Agricultural Irrigation Markets in North Bihar’, Development and Change, 30 (4); Economics, 51(1-2). pp 775-94.

Economic and Political Weekly July 31, 2004 3491