Abstract

Stagnating agriculture and consequent failure of rapid economic growth to bring about poverty reduction as envisaged have been major constraints for India’s economic growth. Contrary to the view that slow-down in public investment for irrigation development is mainly responsible for the deceleration of agricultural growth, the paper argues that in spite of the Government initiatives and substantial investments in irrigation development, the area irrigated by public irrigation systems in India has stagnated or even declined. India’s irrigation economy has been undergoing a dramatic transformation with the control of irrigation shifting from the government to the individual farmers through millions of wells owned and operated by them. Though the booming tube well irrigation has generated substantial socio-ecological dividends in terms of flood mitigation and reduction in water logging and soil salinization, it has also been responsible for resource depletion and contamination of ground water in some parts of the country, leading to various adverse environmental and socio-economic consequences. There is need for achieving the right balance between supply and demand side measures for forging a sustainable ground water governance regime. Problems of groundwater overexploitation in India are bound to become more acute and widespread in the years to come unless corrective mechanisms are put in place before the problem becomes insolvable or not worth solving. Lack of information and absence of systematic monitoring of availability and withdrawal of ground water is a major barrier that prevents the transition from groundwater development to management mode. Further, unlike in the case of surface water irrigation systems, public agencies have only an indirect role to play in the national ground water sector due to its development mostly in the private, ‘informal’ sector and the quality and amount of application of science and management to this sector has been much less when compared to the former.

This paper attempts to trace the history of irrigation development from early 19th century tothe present to emphasize the shifting of focus from the government controlled major andmedium surface irrigation systems to farmer-controlled ground water irrigation systems.Various ideas adopted for creating demand-management regimes through direct regulations,economic instruments, tradable property rights and community resource management around the world have been reviewed to prove the point that ground water governance, throughout the world, is still ‘work in progress’. It also emphasizes the need for recognizing the importance of ground water irrigation systems in South Asia and for information systems and resource planning through establishing appropriate systems for regular ground water monitoring and for undertaking systematic scientific research on the occurrence, use and ways and means for augmenting and managing the resource. Need for initiating suitable demand and supply side management mechanisms and for undertaking ground water management in the river basin context have also been stressed.

Introduction

Stagnating agriculture has emerged, during recent years, as a speed breaker in India’s otherwise splendid and enviable growth story. The failure of rapid economic growth to bring about poverty reduction in commensurate manner is also another major concern linked with stagnant agriculture. It has been widely thought that the slow down in public investment in agriculture, mainly irrigation development, is the main culprit behind the deceleration in agricultural growth. Government of India’s Accelerated Irrigation Benefits Programme (AIBP) was conceived of as a response to the plea for increased public investment in irrigation. In recent budgets, the Union Finance Minister has been laying great stress on completing the “last mile irrigation projects” to step up the pace of irrigation development. Despite these initiatives, the area irrigated by public irrigation systems in India has stagnated, even declined (Shah 2009). In this paper, I want to argue that irrigation in India is in thethroes of a major transition. The irrigation business model that India has followed since early decades of 19th century has rapidly changed in recent years, and public policies based on colonial model of irrigation development are no longer in sync with new developments in Indian agriculture, which has come to depend heavily on groundwater irrigation by boreholes and pumps. Neither the goals of India’s irrigation policy nor our irrigation development strategy jives with the reality of our irrigation economy today. This transition has created a wholly new challenge of balancing food security and agrarian livelihoods on one hand and sustaining groundwater aquifers under stress. It brings into play a new socio-ecological dynamic that is best understood in the environmental economics framework.

Irrigation statistics compiled by the Government of India underestimate the scale of India’s irrigation economy which is booming like never before. Official estimates of the net irrigated area in India based on land use surveys is 57 M ha and the gross irrigated area is around 90 M ha. Other sources, however, suggest that there is great deal more irrigation going on in India. The most striking have been new estimates of global irrigated area based on remote sensing data published recently by the International Water Management Institute (IWMI). Based on the analysis of high resolution satellite imagery backed by extensive ground-truthing work, IWMI’s estimate suggests that in 2004, India had 99 M ha of netirrigated area and 132 M ha of gross irrigated area. Both these estimates are over 50 percent higher than the official estimates. In fact, IWMI’s estimates of irrigated area of today are nearest to what the government of India would like to achieve by 2020. Incredible as these new estimates may sound, recent rounds of national sample survey also suggests that India’s irrigation economy may be considerably larger than reflected in the official estimates.

The Groundwater Revolution

At the heart of the transformation that India’s irrigation economy has been undergoing is the wresting, by millions of small farmers, of the initiative for irrigation development from the hands of the State. Under the model of irrigation development that India followed since the 1830’s, the State has been the architect, entrepreneur, engineer and manager of irrigation systems. ‘Command area’ and ‘duty’ were the mantra of irrigation planning and management. The Government was the provider of irrigation and the farmer a passive recipient. In this model of unbalanced irrigation development, command areas were created near hydraulically opportune sites where reservoirs or weirs could be built and downstream areas could be ‘commanded’ by gravity flow. Farmers in the rest of the country were left to fend for themselves. Post-Independence, India followed much the same strategy for irrigation development that created pockets of prosperous command areas, leaving other parts to rain fed farming.

By 1970, the population pressure on farm lands in many parts of India had become so inexorable that farmers everywhere felt compelled to work their small farm holdings twice, or even thrice every year. Population pressure on farm lands then flagged off India’s tube well revolution. India—especially, in western and north-western parts-- had a centuries old tradition of irrigating with wells. Even in 1900, India had some 4 M ha under groundwater irrigation. At the time of independence, the areas irrigated by groundwater and surface water were evenly balanced. However, it was hardly expected by anybody that India would witness massive spread of tube well irrigation in the surface-water-abundant Ganga-Brahmaputra basin or hard rock peninsular India. Such a pattern of irrigation development appearedwholly inconsistent with the country’s hydro-geology. Equally inconsistent seemed to be large-scale groundwater irrigation in peninsular India with hard-rock aquifers that have poor infiltration and low storage; tanks have been considered ideal for capturing and storing rainwater for irrigation in these areas that comprise 65 percent of India’s land-mass.

At the onset of the 20th century, RC Dutt articulated the prevailing thinking about how irrigation should develop in different parts of India. “Every province in India has its distinct irrigation requirements. In the alluvial basins of the Ganges and the Indus the most suitable irrigation works are canals from these rivers; while away from the rivers, wells are the most suitable. In Bengal with its copious rainfall, shallow ponds are the most suitable works and these were the numerous in the olden times, sometimes of very large dimensions. In Madras and Southern India, where the soil is undulating and the underlying rock retains the water, the most suitable irrigation works are reservoirs made by putting up large embankments and thus impounding the water descending from hill slopes. Such were the old reservoirsof Madras.” (Dutt 1989, vol. II, p 119, footnote 1).

This thinking was endorsed 70 years later by the second Irrigation Commission. For millennia, irrigation in India had remained largely faithful to this dictum. Adaptive, minimalist, unobtrusive irrigation in India of 1800 was a reflection of this hydro-geologic make up of the sub-continental terrain. Constructive imperialism pioneered by Arthur Cotton in the south and Proby Cautley in the north took liberties with this ideal scheme. However, come 1970’s, and this age-old wisdom lay in tatters as a new era of atomistic irrigation unfolded and engulfed India—nay, all of South Asia-- with small-pump irrigation spreading everywhere like wildfire --in canal commands and outside, in arid, semi-arid and humid areas, upstream and downstream of river basins, in excellent alluvial aquifers as well as in poor,hard rock peninsular aquifers with limited storage potential. If the era of ‘constructive imperialism’ began tinkering with the hydrology of river basins, the recent era of atomistic irrigation with small wells and tube wells went about reconfiguring it totally.

The rise of groundwater irrigation also transformed the organization of irrigation at the local level. In pre-Colonial India, co-operation at the community level was the dominant irrigation institution. Under the colonial rule, collaboration between the State and the engineering profession was at the centre stage of centralized, bureaucratic irrigation development and management. In this new era of atomistic irrigation, the State as well as science became onlookers in a ballgame whose rules and logic they did not understand, much less dictate. In an incipient atomistic irrigation economy of the 1980’s and later, neither the State nor the community was the entrepreneur, builder, or the manager of irrigation; it was the multitude of small-holders--Marx’s ‘millions of disconnected production units’-- each with his tiny, captive irrigation system, ostensibly unconnected with the rest. Until now, crops had to wait for water to be released and flow through a network of canals before getting irrigated; now, water was scavenged on-demand and applied just-in-time when crops needed it most.

than the British had invested during the entire 110 year period between 1830 and 1940. Yet, even according to the government of India’s figures, over 60 percent of irrigated areas are today served by groundwater. Other indicators suggest even this may be a serious underestimate. Remote sensing data as well as national sample survey suggest that as much as 75-80 percent of India’s irrigated area today is served by groundwater wells. Until 1960, Indian farmers owned just a few tens of thousands of mechanical pumps using diesel or electricity to pump water; today India has over 20 million modern water extraction structures. Every fourth cultivator household has a tube well; and two of theremaining three use purchased irrigation service supplied by tube well owners.

Socio-economic significance and impacts of the groundwater boom

The groundwater boom is a sub-continental phenomenon that has encompassed, besides India, arid regions of Pakistan Punjab and Sind—which boast of the world’s largest continuous surface irrigation system—and the humid Bangladesh and terai areas of Nepal. In these predominantly agrarian regions, the booming groundwater economies have assumed growing significance from viewpoints of livelihoods and food security; however, their significance as engines of rural and regional economic growth has remained under-studied. There are several ways to consider the scale of the groundwater economy;but one practical measure is the economic value of the groundwater production. An unpublished report for USAID in the early 1990’s placed the contribution of groundwater irrigation to India’s GDP at around 10 percent (Daines and Pawar, 1987); if that proportion held now, the size of the groundwater irrigation economy of India would be some US $ 75-80 billion. In table 1 below, we attempt a rough estimation of the market value of groundwater use in the Indian sub-continent. India, Pakistan, Bangladesh have active markets in pump irrigation service in which tube well owners sell groundwater irrigation to their neighbours at a price that exceeds their marginal cost of pumping. This price offers a market valuation of groundwater use in irrigation. We use available estimates of the number of irrigation wells and estimates from sample surveys on average yield of wells and annual hours ofoperation of irrigation tube wells in the countries covered. In India, for instance, a large number of farmers paid their neighbouring bore well owners US $ 0.04/m3 for purchased groundwater irrigation around 2000; applying this price to the annual groundwater use of say 200 billion m3 gives us US $ 8 billion as the economic value of groundwater used in Indian agriculture/year. For the Indian subcontinent, the corresponding estimate is around 10 billion US dollars. In many parts of water-scarce India, water buyers commonly enter into pump irrigation contracts offering as much as 1/3rd crop share to irrigation service provider; in water abundant areas, in contrast, purchased pump irrigation cost amounts generally to 15-18 percent of the gross value of output it supports. This can be used to draw the general inference that the agricultural output that groundwater irrigation supports is 4-5 times itsmarket value.

Explosive growth in shallow tube wells and small pumps has democratized Indian irrigation much like personal computers have democratized computing globally. By the same token, large canal irrigation systems are heading towards the future that mainframe computers are facing. Boreholes and small pumps took irrigation away from command areas to the nook and corner of the country. Among several things, the booming pump irrigation economy has: [a] offered some irrigation access to an overwhelming majority, rather than concentrating all irrigation benefits on small privileged groups in command areas; [b] thereby, helped soften growing farmer unrest in the region’s vast dry-land areas, which would have otherwise destabilized social and political structures; [c] has come to account for over 60 percent of irrigated areas, and 80 percent of irrigated farm output and resultant incomes; [d]drought-proofed the region’s agriculture against at least one monsoon failure and made large-scale famines history; [e] improved farm wages and increased demand for farm labor year-round; [f] demonstrated a strong pro-poor, inclusive bias in irrigated agriculture; [g] supported a new drive towards intensive diversification to high value products such as milk, fruit and vegetables, especially in dry land areas in a scale-neutral format. These impacts have benefited—directly and indirectly, to lesser or greater extent--around half a billion rural people in South Asia. One can not say that the South Asian peasant is much better off in 2000 compared to 1975; but one can confidently say that, other things being the same, he would have been immensely worse off but for the pump irrigation boom.

Thanks to its myriad and widespread benefits, pump irrigation revolution, aided by irrigation service markets, has been amongst the most powerful rural poverty alleviation phenomena without which the region would arguably have been in the throes of massive social and political instability. Pump irrigation boom in India since 1975 has created more irrigation in 30 years than public investments in canal irrigation did in 170. Pump irrigation has also brought about greater spatial equality in irrigation; it is spread all over the country unlike canal projects which have created concentrated pockets of agrarian prosperity in canal commands. Vibrant local, informal markets for pump irrigation service have helped India’s 20 odd million WEM owners to reach irrigation benefits to another 40-60 millionsmall holder families, covering a vast majority of the farming community with access to supplemental irrigation. Especially in north-western India, the rise of groundwater irrigation on private initiative has reduced water logging, which otherwise would have required massive public investment in drainage and salinity management. The pump irrigation economy has been the driving force behind national growth in food and agricultural economies, for example, transforming West Bengal (and Bangladesh) asthe region’s rice bowls. Pump irrigation farmers apply less water per hectare, achieve higher ratio of evapotranspiration to consumptive fraction, and obtain higher yields/ha compared to flow irrigators. Across rural economic classes, the distribution of pump ownership is more equal than land holdings. In dry-land areas, supplemental pump irrigation has had a dramatic impact of stabilizing rain-fed yields and promoted agrarian diversification. The impact of a widespread drought on agricultural and food production today is much more muted compared to 1960’s and before. Pump irrigation boom has been instrumental in all but banishing starvation deaths in the sub-continent. In effect, it has activated a sub-surface reservoir on a sub-continental scale—that always existed but remained largely unused—butwhich now captures and stores over 250-270 km3 of water in a normal year, creating on a massive scale space, time and form utility in agricultural water use, the object of any reservoir.

Sustaining the Groundwater Boom

Nothing is an unmixed blessing; and this is true about South Asia’s pump irrigation revolution since 1970’s which has been a prominent target of doomsday prophecies about an impending socio-ecological disaster (see, e.g., Seckler et al.1999; Postel 1999; Vaidyanathan 1996). There is much truth in this concern; however, tube well irrigation has generated substantial socio-ecological dividends as well. In flood prone eastern India, it has helped mitigate the rapacity of floods and water logging by reducing ‘rejected recharge’ by creating more storage in the aquifers. In the Indus basin too, tube well irrigation has reduced water logging and salinization, a task which would have taken hundreds of million dollars of investments in drainage.

Groundwater horror stories of India are however becoming increasingly frightening in arid alluvial and hard-rock aquifers. In some coastal plains along with arid alluvial plains facing overdraft, the central resource governance challenge is coping with salinization and depletion which, in a chronic form already visible in some parts, may seal the fate of agriculture, and of human settlement itself. Then, in hard rock areas of peninsular India, where tube well irrigation expansion is way out of proportion to the limited storage offered by aquifers, resource depletion is a serious issue in itself but has also aided growing concentration of fluoride and other salts in groundwater which is the main source of drinkingwater supply for rural as well as urban populations. Problems of geogenic contamination of groundwater—such as with arsenic in eastern Ganga basin and fluoride in much western and peninsular India are large and serious. The causal role of pump irrigation in mobilizing fluoride and other salts in groundwater is clearer than in arsenic contamination whose chemistry is still tenuous and disputed.

A few years ago, David Seckler wrote alarmingly that a quarter of India’s food harvest is at risk if she fails to manage her groundwater properly. Many people today think that Seckler’s may well have been an underestimate; and that if India does not take charge of her groundwater, its agricultural economy may crash. Sandra Postel (1999) has suggested that some 10 percent of the world’s food production depends on overdraft of groundwater to the extent of 200 km3; most likely, 100 km3 out of this occurs in Western India. Conditions in North China plains they are no better. In the lower Indus basin in Pakistan and the Bhakra system in Northern India, groundwater depletion is not a problem but soil and groundwater salinization is. IWMI’s past research to understand the dynamics of groundwater socioecologies indicates some recurring patterns. In much of South Asia, for example, the rise and fall of local groundwater economies follow a 4-stage progression outlined in Figure 1 below, which is self explanatory. It underpins the typical progression of a socio-ecology from a stage where unutilized groundwater resource potential becomes the instrument of unleashing an agrarian boom to one in which, unable to apply brakes in time, it goes overboard in exploiting its groundwater.

The 4-stage framework outlined in Figure 1 shows the transition that South Asian policymakers and managers need to make from a resource development mindset to a resource management mode. 40 years of Green Revolution and mechanized tube well technology have nudged many regions of South Asia into stage 2-4. However, even today, there are pockets that exhibit characteristics of stage 1. But the areas of South Asia that are at stage 1 or 2 are shrinking by the day. Many parts of Western India were in this stage in 1950’s or earlier, but have advanced into stage 3 or 4. An oft cited case is North Gujarat where groundwater depletion has set off a long term decline in the booming agrarian economy; here, the foresightful well-off farmers—who foresaw the impending doom-forged a generational response and made a planned transition to a non-farm, urban livelihood. The resource poor have been left behind to pick up the pieces of what was a booming economy barely a decade ago. This drama is being re-enacted in ecology after groundwater socio-ecology with frightful regularity (Moench 1994; Shah 1993; Barry and Issoufaly 2002).

In stage 1 and early times of stage 2, the prime concern is to promote profitable use of a valuable, renewable resource for generating wealth and economic surplus; however, in stage 2 itself, the thinking needs to change towards careful management of the resource. Yet, the policy regime ideal for stage 1 and 2 have tended to become ‘sticky’ and to persist long after a region moves into stage 3 or even 4. IWMI’s recent work in North China plains suggests that the story is much the same there as well. The critical issue to address is: does stage 4 always have to play out the way it has in the past? Or, are there adaptive policy and management responses in stage 2 that can generate a steady-state equilibrium, which sustains the groundwater-induced agrarian boom without degrading the resourceitself? In the remainder of this paper, we review the prospects and opportunities for forging such a steady-state equilibrium.

Environmental Economics of Aquifers and Institutional Response

Groundwater modeling is the playing field for hydro-geologists. These have developed a rather formidable repertoire of models that analyze the complex behavior of aquifers in response to development. However, in a region like South Asia where millions of smallholders directly interfere with the aquifer processes without let or hindrance, we have little understanding of how users respond to its development, and in due course, its depletion or deterioration. Developing such understanding is an important area of work for environmental economists.

How do India’s groundwater users relate to aquifer development? How do they respond, as individuals and as a collectivity sharing a portion of an aquifer, to groundwater depletion or quality deterioration? When do they choose to co-operate and when to compete? Do they actually choose? Or are they impelled to behave in a certain way by natural processes they are confronted with? Are there situations in which they find it easier than in others to co-operate for the greater common good? These, and many other such questions, are crucial for us to explore but require a marriage of hydrogeology and social sciences such as economics, political science, and sociology.

Much hydrogeology is about the impact of human intervention on aquifer behavior. But environmental economics also needs to explore the impact of aquifer conditions on human behavior, especially, the behavioral response of people living off it. By institutional response, I mean the central behavioral tendencies of groundwater irrigators and the social dynamic that results from different aquifer conditions. In keeping with Veblen (1934), the original institutionalist, I treat institutions as ‘settled habits of thought common to the generality of men’.

An average groundwater user in India has little or no formal knowledge of hydro-geology. But s/he certainly has ideas and even theories about how it all works underneath the earth’s crust (Rosin 1993; Shah 2000). A lot of these popular theories will not withstand scientific scrutiny; yet, farmers’ decisions and actions are guided by their theories more than by formal science. One way to think about how farmers form their theories is by referring to what economist John Muth (1961) called rational expectations which help people formulate their view of the future state of things. Rational expectations are to be distinguished from adaptive expectations, which see the future as little more than a mechanical reproduction of the past. The rational expectations model suggests that people take into account all the information available to them—including the expectations of others they regard highly--to arrive at an expectation which differs from the actual only by a random error (Muth 1961; Sargent 2002). When the behavior of most or all agents is shaped by such rational expectations, selffulfillingprophecies abound. If majority customers expect a bank to fail, and begin a run on it, a smallbank may actually fail. If most traders expect stock prices to rise, and start buying in that expectation, the market will actually skyrocket even when fundamentals suggest no reason for it to. Likewise, the expectations people living on or off an aquifer have about where it is headed in response to development or conservation shape their individual or collective behavior towards it and towards the ‘aquifer community’.

An ‘aquifer community’ can be viewed as a collectivity of aquifer users in a locality who are aware of their interdependence in their use of a common aquifer or a portion thereof. Researchers from the British Geological Survey (2004) put it elegantly when they define it as a group of groundwater users who are ‘mutually vulnerable and mutually dependent because of the centrality of resource use in supporting livelihoods’. The level of awareness of this inter-dependence is a measure of the strength or weakness of the aquifer community. In understanding the institutional dynamic in an aquifer, important are the rational expectations that a representative farmer has about the impact of another farmer’s withdrawal on own water availability (s), and of the whole community’s withdrawals on her groundwater availability (S); individual farmer’s water conservation effort on her water availability (h)and the community’s conservation effort on her water availability (H). Five situations outlined in table 2 represent the types of institutional dynamic that aquifer conditions generate in response to development in South Asia.

[Situation 1] atomistic individualism (s=0; S=0; h=0; H=0): occurs when each farmer is an insignificant user in an abundantly recharged water table aquifer; his abstraction has little impact on himself or other users; likewise, aquifer development has little discernible impact on the individual user; here, interdependence amongst users goes unnoticed; ‘aquifer community’ is non-existent, and rational expectations fail to generate institutional dynamic of the kind we observe in the remaining four situations;

[Situation 2] collusive opportunism (s=0; S occurs when aquifer development sharply raises the cost of groundwater abstraction without greatly reducing water supply or quality; here, wealthy farmers establish de facto control over the resource, and collude against the resource poor but spearhead political mobilization to defend their access to and control over the resource; irrigators display limited inter-dependence and are a weak aquifer community;

[Situation 3] rivalrous gaming (s0): occurs when aquifer development sharply raises the cost of water production and also limits available groundwater supply that users actively compete for; this condition promotes intense and destructive rivalry among competing users; irrigators display a strong sense of interdependence but are a dysfunctional aquifer community; sporadic evidence of beneficial effects of community conservation fail to metamorphose into organized collective action;

[Situation 4] co-operative gaming(s0; H>>0): under certain catalytic conditions, rivalrous game metamorphoses into a co-operative game that reduces the cost and risk of water production and augments water availability to the entire community; positive expectations that so result foster a strong sense of benign interdependence and a highly functional aquifer community; such aquifer communities are ripe for proactive local groundwater self-governance;

[Situation 5] exit (s This state occurs when groundwater development results in rapid quality deterioration without affecting supply. Costs and risks of groundwater use become prohibitive; and users begin giving up irrigated farming or farming itself. Pervasive negative expectations inspire fatalism, hopelessness and despair that overwhelm the strong sense of interdependence; aquifer community takes a downward spin and eventually withers away.

The framework set out above is helpful in making sense out of how millions of farmers have responded to the ecological consequences of rapid groundwater development in different parts of India. In alluvial aquifers of arid western Rajasthan and North Gujarat, groundwater irrigators are running a race to the ‘pump house’; competitive deepening of tube wells is the name of the game here. In these regions, we never hear about spontaneous efforts by farming communities to harvest rainwater and recharge aquifers on a large scale; the predominant institutional response takes the form of mobilizing to maximize and preserve energy subsidies. In humid alluvial plains of the Ganga-Brahmaputra-Meghana basin, groundwater irrigation here is a major poverty-alleviator and poses no environmental