Can India raise agricultural productivity while reducing groundwater and energy use?

Transcription

1 International Journal of Water Resources Development ifirst article, 1 17, 2013 Can India raise agricultural productivity while reducing groundwater and energy use? M. Dinesh Kumar a *, Christopher A. Scott b and O.P. Singh c a Institute for Resource Analysis and Policy, Hyderabad, India; b Center for Studies in Public Policy and School of Geography and Development, University of Arizona, Tucson, USA; c Department of Agricultural Economics, Institute of Agricultural Sciences, Banaras Hindu University, Varanasi, India (Received 23 February 2012; final version received 23 October 2012) This paper provides empirical evidence that power tariff reform with pro rata pricing and higher unit rates for electricity not only would promote equity, efficiency and sustainability in groundwater use, but also would be socio-economically viable for small-holder farmers. It shows that the arguments of high transaction cost and political infeasibility used against metering are valid only in specific regional contexts and under increasingly outmoded power-pricing and agricultural-production regimes, if one considers the recent advancements in remote sensing and the facts that overexploited regions have a low density of wells and are mostly owned by farmers who constitute a small segment of the farming community. Keywords: pro rata pricing; energy rationing; groundwater-electricity nexus; water productivity; prepaid meters Introduction For countries in the semi-arid and arid tropics, sustainability of agricultural production is largely dependent on irrigation water security (Grey & Sadoff, 2007; Shah & Kumar, 2008). Many such regions are primarily dependent on groundwater for irrigation (Schiffler, 1998). Developing countries, with a disproportionate share of the world s semi-arid and arid tropical regions, face problems of groundwater overdraft from excessive irrigation withdrawals, which threatens the sustainability of agricultural production and the livelihoods of millions of rural families who depend on it. India, China, Mexico, Oman, Iran, Pakistan and Morocco are just some of the countries trying to tackle groundwater overdraft through a variety of approaches (Garduño, van Steenbergen, & Foster, 2010; Giordano & Villholth, 2007; Kumar, 2007; Kumar, Scott, & Singh, 2011; Scott, 2011; Scott & Shah, 2004). In such regions, energy security and water security are inextricably linked. There cannot be a better example of this linkage than Indian agriculture. In , around 12.8 million electric pumps with a total of gigawatts (GW) of connected load consumed billion kilowatt-hours (kwh, also referred to as units in India) of electricity (India Energy Portal, n.d.). On the other hand, in regions with poor electrification but endowed with shallow aquifers, diesel pumps are put to service for pumping water. As irrigation becomes increasingly energy-intensive, energy security is critical for ensuring agricultural water security. Our ability to provide reliable and adequate energy supplies for other sectors of the * Corresponding author. dinesh@irapindia.org ISSN print/issn online q 2013 Taylor & Francis

2 2 M. D. Kumar et al. economy is heavily dependent on how efficiently water for crop production is managed. Unmanaged water demand for irrigated agriculture, which accounts for 21% of total electricity consumption, poses a serious challenge to energy security in India, which is the world s fourth-largest energy consumer (US Energy Information Administration, 2011). One of the key approaches debated and often tried in different parts of the world for the co-management of groundwater and electricity revolves around managing the groundwater electricity nexus through metering of groundwater and pro rata pricing of the electricity used to pump it (Kumar et al., 2011, for India; Scott & Shah, 2004, for Mexico; Shah, Giordano, & Wang, 2004a, for China; Zekri, 2008, for Oman), energy rationing to the farm sector using prepaid meters (Shah et al., 2004a for China; Zekri, 2008 for Oman), electricity supply rationing (Shah, Scott, Kishore, & Sharma, 2004b) and denial of electricity connections (Kumar, 2007; Malik, 2009) in regions which experience groundwater overdraft problems, tube well permits (Shah et al., 2004b, for China and Mexico) and management of groundwater quotas (Madani & Dinar, 2011; Scott & Shah, 2004, for Mexico). Particularly in the Indian context, electricity used for agriculture is heavily subsidized in India under both pro rata and flat-rate tariff regimes (Kumar et al., 2011; Scott & Sharma, 2009). Many are of the view that energy subsidies are essential in allowing smallholders to sustain irrigated crop production, which is becoming less profitable due to the rising cost of inputs, including irrigation. This idea, therefore, has high political currency (Kondepati, 2011). Such views are stronger now than ever before in the wake of the gradual removal of fertilizer subsidies and the increasing private cost of well irrigation resulting from declining water levels. But the long-term impacts of heavy energy subsidies on the economic cost of groundwater abstraction and agricultural productivity are not appreciated (Gulati, 2011). Some, however, do believe that free or heavily subsidized power offered to well irrigators leads to uncontrolled abstraction of groundwater and its inefficient use, resulting in overdraft and aquifer mining, with the undesirable consequence of a rising energy footprint for agriculture (Asad & Dinar, 2006; Madani & Dinar, 2011; Narayanamoorthy, 1997; Saleth, 1997; Scott & Shah, 2004). The reality is that energy subsidies in the groundwater-dependent and agriculturally prosperous states of India have been rising over the years (Government of India, 2002; Kumar et al., 2011). This has had a series of cascading, negative effects on the quality of power supply in the farm sector and overall power-sector viability. On the other hand, problems of groundwater mining are becoming more rampant. This is leading to serious erosion of both water and energy economies. A World Bank study in the state of Haryana pointed out that even farmers believe that the ability of the governments to offer improved power supply in the farm sector is heavily dependent on higher tariffs and metering (World Bank, 2001). Further, as noted by Birner and Sharma (2011) in the context of Andhra Pradesh and Tamil Nadu, electricity subsidies in the farm sector were never introduced in response to demand for concessions from farmers associations but were part of a larger populist welfare policy of the ruling parties. It is clear that the key to managing India s groundwater economy and power sector efficiently lies in managing the political economy of power subsidies. Impact of subsidized electricity on groundwater use and farm productivity Studies have empirically shown the negative impact of subsidized electricity on resource use efficiency (Kumar et al., 2011). Theoretically, since the marginal cost of using irrigation water is zero under a flat-rate tariff regime, farmers will keep applying water to

3 International Journal of Water Resources Development 3 the crops until the yield maximizes, which correspond to the maximum gross (private) return. The net return also will be highest at this point. In other words, farmers will keep applying water until the net marginal return becomes zero. This is under ideal circumstances, where the farmer is well informed about the irrigation application corresponding to the maximum yield. But in practice, zero marginal cost tends to lead to excessive irrigation and may even result in yield losses. Nevertheless, in either case, irrigation does not correspond to the economically optimal level, because society incurs a huge cost for the energy used for crop production. The net marginal economic return thus becomes negative long before the net marginal (private) return approaches zero. Water use would therefore become highly inefficient (Kumar, 2007). Yet another consequence of flat-rate pricing is that farmers are tempted to grow crops with high water demand if these help maximize the return per unit of land. The reason is that there are no restrictions on the volume of water that farmers can pump from the aquifer underlying their piece of land (Kumar & Singh, 2001). The preference of well irrigators for high-water-demand crops like sugar-cane in semi-arid areas of Maharashtra, or paddy (rice) in semi-arid and arid areas of Andhra Pradesh, is a result of heavy electricity subsidies for groundwater pumping. Arguments against metering and pro rata pricing One dominant argument against the shift in power pricing is the higher marginal cost of supplying metered electricity owing to the high transaction cost of metering. This may reduce the net social welfare as a result of reductions in (1) demand for electricity and groundwater in irrigated agriculture and (2) net surpluses that individual farmers generate from farming (Shah, 1993). The second argument, by Saleth (1997), is that for the power tariff to be in the responsive (price-elastic) range of the power demand curve, prices have to be so high that they become socially unviable or politically untenable. The third argument is that under pro rata tariffs, the increased cost of pumping groundwater would be transferred to the water buyers. Mukherji et al. (2009) argued that under flat-rate tariffs, the water buyers would gain from low irrigation-water charges due to competitive water markets because well owners would have greater incentive to pump more water. The transaction-cost argument is one of the most pervasive arguments in the electricity groundwater management debate in India. This has mainly stemmed from the sheer number of groundwater abstraction structures in India, estimated to be around 25 million. But in the advancement of this argument, little attention has been paid to the number of wells in regions that, based on water scarcity, actually require co-management of electricity and groundwater. The fact is that the overwhelming numbers of wells in India are in the Indo-Gangetic belt (Scott & Sharma, 2009). This region, especially the eastern Gangetic plain, does not experience serious long-term problems of groundwater overdraft (Government of India, 2005; Sharma, 2009). Many of the areas facing problems of intensive use and overdraft of groundwater have relatively low spatial density of wells. These regions include north and central Gujarat, central Punjab, western Rajasthan and parts of peninsular India comprising parts of Andhra Pradesh, Karnataka and Tamil Nadu. As shown by Kumar (2007), there is no relation between intensity of groundwater use and well density. There is somewhere an inverse relationship between groundwater intensive use and well density. 1 The transaction cost of metering has almost become a non-issue today, with the advent of many advanced technologies for metering electricity consumption, including those based on remote

4 4 M. D. Kumar et al. sensing and satellite technology, where the physical presence of the meter reader is not required (Kumar et al., 2011; Zekri, 2008). Remote sensing technology, for instance, is used in West Bengal to meter the electricity consumption of millions of shallow tube wells (Mukherji et al., 2009). Electronic meters with data-logging devices and facilities for transmitting the data through a satellite communications system are used on all 11-kilovolt distribution feeders in Andhra Pradesh for energy audit (Bhatia & Gulati, 2004). Regarding the second and third arguments, i.e. socio-economic viability and equity impacts of the shift to pro rata tariffs, their validity is examined in the next section. Attempts have consistently been made to link power-pricing policies to vote banks, arguing that for a ruling government any decision to raise power tariffs in agriculture would be nothing less than political suicide. While one can see the political interest in subsidizing farm power to please the millions of rural voters, the fact, as noted by Howes and Murugai (2003), Kumar and Singh (2001) and Vashishtha (2006a, 2006b), is that it is the large-holders, constituting a small fraction of the farming community, who actually appropriate the majority share of the subsidy benefits under the flat-rate system of charging electricity and under free power. Vashishtha (2006a) showed that in AP, large and very large farmers together received 73% of the subsidy benefits in , whereas small and marginal farmers received only 5.1%. The corresponding figures for Punjab were 73.8% and 6.3%, respectively (Vashishtha, 2006b). While a large majority of the farming community does not gain from such policies, the politicians are largely ignorant about this. Understanding these nuances would go a long way toward convincing political leaders of the need to do away with such policies, which have negative effects on productivity and equity, i.e. they neither benefit the larger section of the rural masses, nor do they help improve the water and energy economies of the states concerned. Impact of electricity pricing on energy and groundwater demand in agriculture in India Theoretically, the shift from flat-rate to pro rata tariffs would encourage farmers to use energy and groundwater efficiently by inducing a positive marginal cost of using electricity and groundwater (Asad & Dinar, 2006; Muñoz-Piña et al., n.d.; Narayanamoorthy, 1997). Higher water-use efficiency or water productivity would indicate higher energy-use efficiency. But the converse is not true. One of the responses to higher energy tariffs would be for farmers to improve the efficiency of water abstraction devices, including pump sets and suction pipes, if possible. But, beyond some point, this cannot help reduce the cost of irrigation, and therefore the subsequent response will be to make water use more efficient, in two ways. One is by reducing the cost of irrigation input and the other is by increasing the gross return. This can be done through three major steps: improving the technical efficiency of water use by optimizing irrigation water application; improving agronomic efficiency in water use (kg/m 3 of water) by optimizing agronomic inputs to crops; and shifting to crops with higher water productivity in economic terms (INR/m 3 ) (Kumar, 2007). Empirical studies showing the differential impacts of various energy pricing regimes on agricultural groundwater use and land and water productivity in irrigated agriculture were absent in India and elsewhere until recently. The findings of a recent empirical study are presented here to illustrate the impact of energy pricing on groundwater demand for crop production, the socio-economic viability of farming, and sustainability and equity in groundwater use. The study compared the farming enterprises of electric well owners who

5 International Journal of Water Resources Development 5 pay for electricity on the basis of connected load (flat-rate tariff), diesel well owners who pay for energy on the basis of consumption, and buyers of water from electric well owners and diesel well owners in eastern Uttar Pradesh and south Bihar; plus farmers who pay for electricity on a pro rata basis in north Gujarat. Here, the buyers of water from diesel well owners incur higher water charges as compared to the buyers of water from electric well owners. Diesel well owners who irrigate their own farms and buyers of water from electric and diesel well owners, are proxy for pro rata electricity pricing, along with the farmers in north Gujarat whose power consumption is metered. The primary data for the study covered 60 farmers for each of 10 categories of irrigators such as well owners and water buyers for electric and diesel well commands in eastern Uttar Pradesh and south Bihar, and farmers with metered pump connection and those with flat-rate connections from north Gujarat, making the total size of the sample from the three locations 600. The analysis included comparing the cost of irrigation water, irrigation dosage, physical productivity of water for crop production for individual crops (kg/m 3 ), water productivity in economic terms (INR/m 3 ) for the individual crops and farming system as a whole, and overall groundwater pumping rates per unit cropped for different categories of farmers. Comparison of irrigation applications and physical productivity of water in crop production for the same crop were used to analyze the impact of tariff change on groundwater-use efficiency. Comparison of net water productivity in crop production, dairy production and for the entire farming system in economic terms, as well as the net return per unit of land for the entire farm, were used to examine the impact of tariff change on the socio-economic viability of farming. The average pumping rates per unit area of irrigated land were used to analyze the impact of change in tariff regime on groundwater-use sustainability. The physical productivity of water in crop production (kg/m 3 ) for different irrigated crops was estimated by taking the ratio of the yield of the crop per unit area (kg/ha) and the volume of irrigated water applied per unit area (m 3 /ha). The economic productivity of water in crop production for different crops (INR/m 3 ) was estimated by taking the ratio of the net income return from crop production for the respective crops per unit area of land (INR/ha) and the volume of water applied per unit area (m 3 /ha). The net return for each crop was estimated by subtracting the input costs of seeds, farm labour, machinery, fertilizer, water and electricity (INR/ha) from the gross income (INR/ha), which was obtained by multiplying the yield of the crop (kg/ha) by its farm gate price (INR/kg) (based on Kijne, Barker, & Molden, 2003). The volume of water applied to each crop was estimated by multiplying the discharge of the well (m 3 /hour) from which irrigation was provided to the crop, measured in the field, and data on the total duration of irrigation applied to the respective crop over the entire crop season, obtained from the farmers during the survey. Cost of irrigation water for different categories of farmers Farmers who have metered power connections not only incur the marginal cost of using well water, but also pay a higher price for every unit of irrigation water (INR/m 3 )as compared to their counterparts with flat-rate connections. Similarly, farmers who are buyers of water from electric pump owners and diesel well owners in eastern UP and south Bihar also incur the positive marginal cost of using irrigation water and pay higher unit costs for irrigation water compared to their water-selling counterparts (Table 1).

6 6 M. D. Kumar et al. Table 1. Cost of irrigation water for different categories of farmers. Area Water source Average (INR/m 3 ) 1 Range (INR/m 3 ) Eastern UP Electric pump owners Electric pump buyers Diesel pump owners Diesel pump buyers North Gujarat Metered connections Non-metered connections South Bihar Electric pump owners Electric pump buyers Diesel pump owners Diesel pump buyers Source: Calculated from authors primary data. 1 USD1 ¼ INR50. Cropping patterns of different categories of farmers Analysis of cropping patterns of well owners and water buyers under different modes of energy pricing, i.e. connected load (electric well owners) and unit consumption (diesel well owners, and water buyers in both diesel and electric well commands) in eastern UP shows that in diesel well commands, pump owners allocate about 26% of the gross cropped area to paddy cultivation, whereas in the case of water buyers, it is only 22%. In electric well commands, pump owners allocate 12% to paddy and water buyers allocate about 15% to paddy. Electric pump owners also grow groundnut. Water buyers in both electric and diesel well commands allocate larger portions of their cropped area under green fodder and vegetables during kharif season (summer rains) as compared to pump owners. Water buyers in diesel well commands grow lentils of the Arhar variety. Water buyers in electric well commands grow lady s finger. Major crops grown during winter season are wheat and barley, potato, pea, gram, mustard, linseed and alfalfa. The percentage areas allocated for some crops (wheat, pea, potato and alfalfa) are smaller for well owners than for water buyers; for others (mustard, gram, barley and linseed), they are larger for pump owners than for water buyers. In diesel well commands, pump owners allocate a larger share of their cropped area to winter crops as compared to water buyers. Such sharp differences are not seen in the case of electric well commands. During the summer season, crops such as green fodder, sunflower and vegetables are grown by well owners in electric well commands, but their water-purchasing counterparts grow only green fodder. In diesel well commands, the crops grown during summer season are green fodder and vegetables. Both diesel well owners and water buyers are found to be growing some green fodder. In the case of north Gujarat, the major crops grown by the tube well owners under both tariff regimes are green fodder, food grain crops, pulses, groundnut and cash crops such as cluster bean, cotton and castor. The farmers of this region allocate small areas to green fodder. During kharif, tube well owners under a pro rata tariff regime allocate slightly larger percentages of the cropped area to cotton, castor and fodder bajra. During winter, tube well owners under a flat-rate tariff regime allocate more area to green fodder, wheat and mustard. The tube well owners under the pro rata tariff regime allocate slightly larger areas to cumin, a high-value cash crop with low water consumption. The major crops grown during summer season are green fodder and bajra.

7 International Journal of Water Resources Development 7 In south Bihar, no significant difference was seen in kharif cropping patterns between well owners and water buyers in electric well commands or diesel well commands (Table 2). During winter, water buyers in electric well commands cultivate gram and carrot. Diesel pump owners and water buyers in both diesel and electric well commands keep a larger area for growing potato. During summer, only diesel pump owners and water buyers in their commands cultivate green fodder. In general, electric pump owners allocate larger areas to different crops as compared to electric pump water buyers. There is a similar trend in case of diesel commands. Overall, the water buyers (in eastern UP and south Bihar) and farmers who have metered electricity connections (in north Gujarat) allocate some amount of land to highly water-efficient crops, which are also less water-consuming. Irrigation application and water productivity in crop production Higher physical productivity of water use for a given crop indicates more efficient use of irrigation water through on-farm water management or better farm management. Higher water productivity in economic terms means better economic viability of irrigated production, if land is available in plenty (Kumar et al., 2011). Analysis of crop water application and water productivity of various crops in the three seasons for well owners and water buyers in electric well commands in eastern UP (Table 3a) shows that the total amount of irrigation water applied for crop production is higher for pump owners as compared to water buyers, and the differences are statistically significant. For instance, in the case of kharif paddy, average depth of irrigation was 7.1 cm for well owners against 2.9 cm for water buyers in electric well commands. Further, for most crops, both physical and economic productivity of water are higher for water buyers than their water-selling counterparts. Equally important is the fact that water buyers do not grow crops during summer when crop water requirements are generally high, whereas well owners grow vegetable crops with high water demand. Regarding diesel well commands, though the well owners as well as the water buyers are confronted with a marginal cost of using water, the water buyers incur a higher cost for irrigation water. But there is not much difference between the cropping patterns of pump owners and water buyers, except that water buyers do not grow sugar-cane or maize. To economize on irrigation water, water buyers cultivate water-efficient crops such as Arhar, black gram and green gram during kharif season. The cropping pattern during winter is the same for diesel pump owners and water buyers. During the summer season, only pump owners grow vegetables. The estimates of irrigation-water application and water productivity in physical and economic terms for different crops show that the water buyers in diesel well commands apply less water to their crops as compared to their water-selling counterparts. Further, the physical productivity of water (kg/m 3 ) and water productivity in economic terms (INR/m 3 ) is higher for water buyers as compared to diesel pump owners for all crops. This could be owing to the higher marginal cost of irrigation water in the case of diesel well commands. In north Gujarat, electric pump owners, who pay a marginal cost for electricity, maintain higher water productivity in both physical and economic terms for all crops as compared to those who are paying for electricity on the basis of connected load (pump horsepower). Further, they do not keep high-water-demand alfalfa, which is a fodder, in their fields during summer. Comparison of the irrigation-water application and water productivity of crops raised by the two categories of farmers in diesel well commands of the south Bihar plains (Table 3b)

8 8 M. D. Kumar et al. Table 2. Cropping patterns of well owners and water buyers under different energy regimes, south Bihar. Electric pumps Diesel pumps Crop Owners Water buyers Owners Water buyers Area (ha) % of GCA Area (ha) % of GCA Area (ha) % of GCA Area (ha) % of GCA Kharif (rainy) season 1. Paddy Masureya (green fodder) Maize (green fodder) Rabi (winter) season 1. Wheat Potato Alfalfa (green fodder) Mustard Gram Radish Carrot Coriander Summer season 1. Onion Maize NP chary (green fodder) Gross cropped area (GCA) Source: Calculated from authors primary data.

9 International Journal of Water Resources Development 9 Table 3a. Irrigation water application and water productivity in electric well commands in Eastern UP. Electric pump owners Electric pump water buyers Crop Depth of irrigation water use (cm) Water productivity (kg/m 3 ) Net water productivity (INR/m 3 ) Depth of irrigation water use (cm) Water productivity (kg/m 3 ) Net water productivity (INR/m 3 ) Kharif season 1. Paddy Vegetable Lady s finger Maize Sesamum Sugarcane Bajra Black gram Groundnut Rabi season 1. Wheat Potato Pea Alfalfa Gram Mustard Barley Summer season 2. Sunflower Vegetables Source: Calculated from authors primary data.

10 10 M. D. Kumar et al. Table 3b. Irrigation water application and water productivity in physical and economic terms in diesel well commands in south Bihar. Crop Depth of irrigation (mm) Diesel well owners Physical water productivity (kg/m 3 ) Net water productivity (INR/m 3 ) Depth of irrigation (mm) Diesel pump water buyers Physical water productivity (kg/m 3 ) Net water productivity (INR/m 3 ) Kharif season 1. Paddy Rabi season 1. Wheat Potato Alfalfa Mustard Coriander Summer season 1. Onion Maize Source: Calculated from authors primary data. shows that the average depth of irrigation is much higher for diesel well owners as compared to their water-buying counterparts. For instance, in the case of kharif paddy, diesel well owners apply 196 mm of irrigation water, against 156 mm for water buyers. Similarly, in the case of winter wheat, diesel well owners apply 211 mm of irrigation water compared to 169 mm for water buyers. Regarding water productivity in crop production, for all crops except onion and summer green fodder, water buyers in diesel well commands secure higher physical water productivity as compared to pump owners. Again, for all crops except onion, the water buyers secure higher water productivity in economic terms as compared to pump owners. Similarly, comparison of estimated mean values of irrigation-water application and water productivity in physical and economic terms for both pump owners and water buyers in electric pump command areas in the south Bihar plain for all crops shows that water buyers apply less water to their crops, and maintain higher physical water productivity for many crops, in comparison to electric well owners. However, they secure lower water productivity in economic terms for most crops, except radish and onion. This could be due to the higher cost of irrigation water that many water buyers are paying (as seen in Column 4 of Table 1 and in Figures 1 and 2), which eventually reduces the net return from crop production (the value of the numerator of water productivity) (Kumar et al., 2011). The trends emerging from the foregoing analysis are: (1) net water productivity of water buyers from electric pumps is greater than from diesel pumps both in east UP and south Bihar; (2) net water productivity of electric pump owners under flat-rate provision is comparatively less than that under pro rata tariffs; (3) water productivity of electric pump owners in economic terms is less than that of diesel pump owners; and (4) economic water productivity of buyers of water from electric pumps is less than those buying water from diesel well owners. The analysis shows that water buyers in diesel and electric well commands, and the farmers who have metered connections, secure higher water productivity in physical terms

11 International Journal of Water Resources Development 11 Figure 1. Cost of pumping groundwater and selling price, electric pumps, south Bihar plain. (kg/m 3 ) for most crops as compared to water-selling well owners. This means that when confronted with a positive marginal cost for irrigation water, farmers are encouraged to use water more efficiently. Farm-level water productivity Although water is not a scarce resource in the eastern UP and south Bihar regions, farmers should try and economize the use of water because using more water means paying more for pump rental services. The farm is the unit for many investment decisions by farmers in agriculture, including water-allocation decisions. Hence, farmers try to optimize water allocation over the entire farm, rather than individual crops, to maximize their returns. The analysis for diesel well commands in eastern UP and south Bihar clearly shows that water productivity in overall farm operation is much higher for water buyers than their water-selling counterparts. It was INR12.89/m 3 for water buyers against INR8.67/m 3 for well owners in eastern UP, and INR12.43/m 3 against INR11.97/m 3 in south Bihar. Farmlevel water productivity for water buyers in diesel well commands in eastern UP is nearly 50% higher ([ ] [100/8.67]) than that of their water-selling counterparts, and therefore is statistically significant. In electric well commands also, differences exist in favour of water buyers in spite of the very low marginal cost of using water. The overall water productivity in farming operations was INR11.18/m 3 against INR10.98/m 3 in eastern UP, and INR10.13/m 3 against INR9.28/m 3 in south Bihar; whereas the marginal cost of irrigation water for water buyers was only INR0.65/m 3 for water-buying farmers of electric well commands in eastern UP and INR0.70/m 3 for south Bihar. Farm-level water productivity is much higher Figure 2. Cost of pumping groundwater and selling price, diesel pumps, south Bihar plain.

12 12 M. D. Kumar et al. for farmers who are confronted with a marginal cost of electricity in north Gujarat as compared to those who pay for electricity based on connected load (INR7.90/m 3 of water against INR6.30/m 3 of water). Here, the difference in water productivity is 25%. The water productivity improvement is highest in eastern UP in the diesel well commands, where water buyers incur the highest marginal cost of irrigation. Further, comparison between electric well owners and diesel well owners in both locations substantiates the earlier point that positive marginal cost promotes efficient use of water at the farm level. Therefore, when confronted with a positive marginal cost of irrigation water, farmers are encouraged to use water more efficiently over the entire farm from the economic point of view. The higher net water productivity in economic terms (INR/m 3 ), which farmers obtain even at a high cost of irrigation water, is suggestive of the fact that it is possible to keep irrigation costs high enough to induce improved efficiency in water use in both physical and economic terms without compromising on farming prospects (Kumar et al., 2011). Equity impacts A dominant hypothesis in electricity pricing policy is that under a flat-rate system of pricing, well owners have a strong incentive to pump out more water and, as a result, the price at which water is traded in the market would approach the cost of production of water, while under pro rata pricing the well owners would pass the additional cost burden on to the buyers, with the result that prices would go up. Thus, according to the proponents of this hypothesis, pro rata pricing would adversely impact equity in access to groundwater (Mukherji et al., 2009). As Table 4 indicates, in eastern UP, the monopoly price ratio 2 (MPR) was higher in the case of electric well commands than in diesel well commands. While the price charged by electric pump owners, who pay for electricity on the basis of connected load, is 3.6 times their cost of pumping, the price charged by diesel pump owners is only 1.8 times their cost of pumping. In south Bihar, however, the trend is opposite. The average price charged by electric well owners is lower than the implicit cost of pumping water (INR0.70/m 3 against INR0.77/m 3 ), whereas the average price charged by diesel well owners (INR2.15/m 3 )is higher than the cost they incur for pumping groundwater (INR1.87/m 3 ). However, this is based on average figures of cost and price. A look at the cost and price figures for individual wells brings out a different picture. A few electric well owners incur very high implicit pumping costs higher than the average selling price (Figure 1). The monopoly price ratio for many others is higher than even the average monopoly price ratio of diesel well owners (1.15) and much higher than that of many individual diesel well owners, who incur very high production costs (Figure 2). More importantly, the monopoly price ratio for some diesel well owners is less than 1. Another interesting phenomenon found in both electric (Figure 1) and diesel (Figure 2) well commands is that the selling price of water is more or less same across the farmers, though the unit cost of pumping water varies. The selling price is decided by the market conditions, irrespective of the cost farmers incur for pumping water (Kumar et al., 2011). Fewer potential sellers against a large number of potential buyers would increase the monopoly power of well owners. Perhaps this is what is happening in the village with electric pumps in eastern UP. On the other hand, the presence of large number of sellers against few buyers would reduce the monopoly power of well owners. They would be

13 International Journal of Water Resources Development 13 forced to sell water at prices conditioned by the market (Kumar, 2007). Perhaps this is what is happening in the village with electric pumps in south Bihar. In sum, the mode of pricing of electricity does not influence the monopoly power of well owners in the market. On the other hand, flat-rate pricing puts large well owners in a very advantageous position because they can bring down their implicit unit cost of pumping groundwater. Therefore, pro rata pricing of electricity would promote equity in access to groundwater, if many farmers from within the same area have access to electricity connections. Sustainability impacts on groundwater Pricing would introduce efficiency, but may not ensure sustainability of resource use (Kumar et al., 2011). The total pumpage per unit of cultivated area could be a good indicator of the sustainability impacts of change in the mode of pricing on groundwater (Kumar et al., 2011). However, because aquifer characteristics including recharge processes and rates are not reliably documented, it is beyond the scope of this paper to undertake water-balance assessments. Instead, this study focuses on the pumping behaviour of farmers. The results of analysis carried out for eastern UP and south Bihar show that the pumpage of groundwater per unit area of cultivated land is lower for water buyers, though their holdings are of smaller size. The data for north Gujarat show that the pump owners with metered connections, in spite of having smaller-sized land holdings (2.95 ha against 3.45 ha for those paying on the basis of connected load) use much less water per hectare of land (304 hours per year) as compared to their flat-rate counterparts (444.0 hours per year). This represents a 30% reduction in pumpage per unit area. The difference in aggregate pumping is much greater between farmers with meters and those without meters. Such a high reduction is water usage per unit of cultivated land, which is disproportionately higher than the reduction in net return per unit of land, is made possible through high improvements in water productivity in economic terms (Kumar et al., 2011). In spite of the slight reduction in pumping, the net return from unit area of land was found to be higher for water buyers in eastern UP and the south Bihar plains. For instance, in the case of water buyers of electric well commands in eastern UP, the net income was INR27,570/ha, while it was INR24,880/ha for well owners. Similarly, in the case of water buyers in diesel well commands, the net return was INR18,075/ha against INR14,528/ha for diesel well owners. This is achieved through high improvement in water productivity through selection of less water-consuming and higher-valued crops 3 (Kumar et al., 2011). This indicates that introducing a marginal cost for water and electricity promotes not only efficient use of water, as manifested by higher farm-level water productivity, but also more sustainable use of water. Table 4. regimes. Selling price of well water and monopoly price ratio (MPR) under different pricing Region District Electric well command Selling price (INR/m 3 ) MPR Diesel well command Selling price (INR/m 3 ) MPR North Gujarat Banaskantha Eastern UP Varanasi and Mirzapur South Bihar plains Patna Source: calculated from authors primary data

14 14 M. D. Kumar et al. Overall impacts of pro rata pricing of electricity in the farm sector Thus, pro rata pricing for electricity does promote efficiency and sustainability in the use of groundwater. But, more importantly, the price level at which the irrigation demand starts responding to tariff hikes is socio-economically viable. Pro rata pricing is unlikely to create negative impacts on access equity in groundwater because selling prices for irrigation water are determined by the monopoly power of the well owners, on which the mode of pricing has no influence. The positive efficiency impact of pro rata pricing is evident from the lower quantities of irrigation that farmers apply to crops and higher physical productivity of water in crop production. The sustainability impact is evident from the lower volume of groundwater used per unit of cropped area. The socio-economic viability is evident from the higher economic productivity of water at the farm level and the higher net return per unit area of irrigated land. The empirical evidence raises further questions about the validity of arguments against such pricing. Technological advancements to reduce transaction cost of metering Loss of electricity in transmission and distribution systems in India was as high as 31.2% of the total electricity available for supply to consumers in (Central Electricity Authority, 2006), and it is alleged that a large part (30 35%) of such losses is an unaccounted-for result of pilferage (Tiwari & Shah, 2003). Metering and pro rata tariff fixation are crucial for reducing the unaccounted-for losses in electricity distribution, improving the financial working of the state electricity boards (SEBs) and reducing overall power deficits, while ensuring equity, efficiency and sustainability in groundwater use. Today, technologies exist for metering and also for controlling electricity consumption by farmers (Kumar & Amarasinghe, 2009; Kumar et al., 2011; Zekri, 2008). Prepaid electronic meters, which are operated through scratch cards and can work on satellite and internet technology, are ideal for use in remote areas to monitor energy use and control groundwater use online from a centralized station (Zekri, 2008). Over the past few years, there has been a remarkable improvement in the quality of services provided by internet and mobile (satellite) phone services, especially in rural areas, with a phenomenal increase in the number of consumers (Kumar et al., 2011). Such technologies are particularly important when there are large numbers of consumers and the transaction cost of visiting wells and taking meter readings is likely to be very high. It is inevitable that they will be adopted in rural India (Kumar et al., 2011; Tiwari & Shah, 2003; Zekri, 2008). Prepaid meters prevent electricity pilferage (Kumar et al., 2011; Zekri, 2008). They can be operated through tokens, scratch cards or magnetic cards, and can be recharged digitally through the Internet and text messaging. They are extensively used to meter the electricity consumption of agricultural wells in the North China Plains (Shah et al., 2004a) and general electricity consumers in South Africa (Tiwari & Shah, 2003). They help electricity companies restrict the use of electricity. The company can decide on the energy quota for each farmer on the basis of reported connected load and total hours of power supply, or sustainable abstraction levels per unit of irrigated land (Zekri, 2008). Kumar et al. (2011) analyzed five policy options for regulating groundwater use in agriculture, weighing their potential benefits in terms of groundwater and energy demand management against the practical problems each one poses in implementation. Four of these five options take advantage of prepaid electronic meters: (1) pro rata pricing with energy rationing based on groundwater sustainability considerations; (2) pro rata pricing with energy rationing based on connected load (in horsepower) and supply hours; (3) fixed energy quota based on supply hours and connected load with pricing based on connected

15 International Journal of Water Resources Development 15 load; (4) pro rata pricing and unrestricted use of electricity; and (5) connected load based tariff with fixed supply hours. This is vastly different from the current policy of restricting supply hours alone, as tried under the Jyotigram Yojna programme in Gujarat State of western India. While option 3 is easily implementable to manage the groundwater energy nexus in agriculture, those who have large-capacity pump sets would be entitled to a large quantity of electricity. Here again, only farmers with large holdings will have an incentive to use groundwater efficiently. This is because there is no marginal cost of using water, and the incentive to conserve electricity and groundwater has to come from the opportunity costs of not being able to irrigate one s entire land, which will be less for farmers with small holdings and large pump sets. Further, because farmers with large-capacity pump sets would corner the lion s share of the subsidy benefits, equity in access to groundwater could still be an issue. Also, water-use efficiency in small farms could be low. One way to overcome this issue is to fix higher charges per horsepower of connected load for farmers who enjoy large quotas. Option 2 is slightly difficult as it would involve introducing consumption-based pricing along with energy rationing, but it would conserve both groundwater and electricity by all categories of farmers, because in this case, farmers access to groundwater in volumetric terms would be defined by the connected load and supply hours. As is evident, option 1 is best for co-management of groundwater and electricity. Because it is possible to limit the total abstraction of groundwater from a regional aquifer using sustainability considerations (such as long-term sustainable yield of the aquifer), and then ration volumetric abstraction by individual well owners, on considerations of their land holdings etc., it would address the issues of equity, efficiency and sustainability. But implementing this would require great political will because traditionally farmers have enjoyed unrestricted rights to use groundwater and these rights would be challenged by this policy intervention. Keeping in view the social benefits of reducing electricity consumption and conserving groundwater, the government can offer subsidies for meters if farmers are willing to go for option 1 or 2. In implementing any of these options it is necessary that SEBs set up a computerized database of all agro wells, comprising their latitude and longitude, physical characteristics and land use details. Conclusions There is mounting empirical evidence that power tariff reform with pro rata pricing and higher unit rates for electricity not only would promote equity, efficiency and sustainability in groundwater use, but also would be socio-economically viable for small-holders. While metering is essential in introducing the suggested tariff reforms, the arguments of high transaction cost used against metering are becoming outdated. On the other hand, managing the political economy for electricity tariff reforms would depend on three important steps. First, inform the political class and bureaucracy that free or heavily subsidized power for the farm sector not only makes little sense economically, but also has high political costs when compared to metered pricing, provided power supply reliability is ensured. Second, convince farmers that good-quality, unrestricted power supply at higher unit cost would enhance their ability to maximize revenue from irrigation and transform farming through higher land and water productivity as well as efficiency and flexibility in managing farming. Third, demonstrate to the state electricity bureaucracies that it is possible to introduce metering with acceptable costs through the use of information technology (Gulati, 2011) and satellite technology. Thus it would be possible to introduce electricity tariff reforms in agriculture, which would reduce power demand and increase agricultural productivity, while simultaneously permitting State Electricity Boards to meet the rapidly increasing non-farm demand for power.

16 16 M. D. Kumar et al. Notes 1. One major reason for this is that the hard-rock areas, where well density is high, have extremely low groundwater potential due to low rainfall and low infiltration of rainwater. But the cost of drilling wells is low, and therefore most farmers own wells, though irrigating only small patches of land. In the deep alluvial areas, where well density is very low, the yield of aquifers is very high and the irrigation potential of deep tube wells is very high. But because the cost of drilling tube wells is prohibitively high, fewer farmers can afford it; the farmers with deep tube wells sell water to the well owners who do not own deep tube wells. 2. The monopoly price ratio, which is used to express the monopoly power of sellers, is defined as the ratio of the selling price to the cost of production and supply. 3. Though the net returns per unit of land were marginally lower for farmers who paid on a pro rata basis in north Gujarat, this is not a concern, because in water-scarce regions like north Gujarat farmers would not have land constraints in maximizing returns. Even if the farmers attempt to expand the area to maintain the net farm return at the previous level, the aggregate water usage would still be lower than the previous level. References Asad, M., & Dinar, A. (2006). The role of water policy in Mexico: Sustainability, equity and economic growth considerations (Latin America and the Caribbean Region Sustainable Development Working Paper No.27). Washington, DC: World Bank. Bhatia, B., & Gulati, M. (2004). Reforming the power sector: Controlling electricity thefts and improving revenue. Washington, DC: World Bank. Birner, R., & Sharma, N. (2011). Electricity supply to agriculture in Andhra Pradesh and Punjab: Evolution and reform initiatives. In R. Birner et al. (Eds.), The political economy of agricultural policy reform in India: Fertilizer and electricity for irrigation (Research Monograph) (pp ). Washington, DC: International Food Policy Research Institute. Central Electricity Authority (2006). All India electricity statistics general review New Delhi: Central Electricity Authority, Ministry of Power, Government of India. Garduño, H., & van Steenbergen, F., & Foster, S. (2010). Stakeholder participation in groundwater management. GW Mate Briefing Note Series, Note 6. Giordano, M., & Villholth, K. (2007). The agricultural groundwater revolution: Setting the stage. In M. Giordano & K. G. Villholth (Eds.), The agricultural groundwater revolution: Opportunities and threats to development (Comprehensive Assessment of Water Management in Agriculture) (pp. 1 4). UK: CABI. Government of India (2002). Annual report on the working of state electricity boards and electricity department: New Delhi: Planning Commission (Power and Energy Division), Government of India. Retrieved from Government of India (2005). Dynamic ground water resources of India. New Delhi: Central Ground Water Board, Ministry of Water Resources, Government of India. Grey, D., & Sadoff, C. (2007). Sink or swim: water security for growth and development. Water Policy, 9(6), Gulati, M. (2011, April 2). 50% savings, 100% free power. The Times of India. Retrieved from AW¼ Howes, S., & Murugai, R. (2003). Incidence of agricultural power subsidies: An estimate. Economic and Political Weekly, 38(16), India Energy Portal (n.d.). Energy sectors, agriculture. Retrieved from: org/subthemes_link.php?text¼agriculture&themeid¼15 Kijne, J., Barker, R., & Molden, D. (2003). Improving water productivity in agriculture: Editors overview. In J. Kijne, et al. (Eds.), Water productivity in agriculture: Limits and opportunities for improvement (Comprehensive Assessment of Water Management in Agriculture) (pp. xi xix). UK: CABI, International Water Management Institute. Kondepati, R. (2011). Agricultural groundwater management in Andhra Pradesh, India: A focus on free electricity and its reform. International Journal of Water Resources Development, 27, Kumar, M. D. (2007). Groundwater management in India: Physical, institutional and policy alternatives. New Delhi: Sage.