In India a real groundwater pumping revolution took place during the past 40 years, with a huge increase in the area irrigated by private wells: from 20% in the 1980s to 60% in 2007 (Kajisa et al., 2007). This rapid expansion in groundwater irrigation was possible through the individual investment of millions of farmers scattered throughout the countryside (Mukherji, 2007). It has been shown by several authors that groundwater irrigation has made a substantial contribution to the productivity and incomes of poor farmers (Bhattarai & Narayanamoorthy, 2003; Saleth et al.,2003; Manjunatha et al., 2011). However currently, the agricultural use of groundwater has turned into a matter of concern for policy makers and planners because effects of overdraft like premature failure of wells, decline in groundwater yield and lowering water tables are becoming more and more apparent (Chandrakanth et al., 2004; Diwakara & Chandrakanth, 2007; Nagaraj et al., 2005; Mukherji & Shah, 2005; Shah et al., 2008). In this perspective it is important to consider the decisions of farmers to invest in wells. In the past , by drilling a new well, farmers could expect to improve their access to groundwater. However, in situations of overdraft, this expectation might not be fulfilled due to cumulative interference of irrigation wells. The question then is why farmers keep on investing in additional or deeper wells (Varghese et al., 2013a; Palanisami et al., 2008.)?
In this study we use a balanced panel dataset (2001-2009) of 313 rural households of 6 villages in Andhra Pradesh and Maharashtra to determine the factors affecting the decision to invest in wells and the investment intensity. The data collection was conducted by the International Crops Research Institute for the Semi-Arid Tropics (ICRISAT). The model used is a "double-hurdle" model. This type of model is used because the decision whether to invest in wells and the decision about how much are not jointly made and might be determined by other factors. The model comprises a probit regression on adoption followed by a truncated regression on the nonzero observations (Cragg, 1971, Worku and Mekonnen 2012). As the name "double-hurdle" suggests, farm households must scale two hurdles in order to invest in wells. The first decision or hurdle for farm households in our setting is whether they will make any investment in wells at all, while their second decision is the intensity of investment, conditional on their first decision. In the double-hurdle model, both hurdles have equations associated with them, incorporating the effects of farmer characteristics and circumstances. Explanatory variables may appear in both equations or in either of them, and a variable appearing in both equations may have opposite effects in each of them.
The results of the double hurdle model (see table) confirm that the decision to invest in wells and the decision about how much to invest are explained by different processes. Past investments have a positive and significant impact on the decision whether to invest or not. This is in line with MarÃ©chal (2009) or Varghese et al. (2013) who attributed the well investment behaviour to a sunk cost fallacy (increasing the resources available to an unsuccessful venture in the hope of recovering past losses). Rainfall quantity in the previous year also significantly influenced the choice to invest. The probability to invest appears to be lower after a year in which rainfall quantity is higher. Also land tenure plays a role in this decision. The dummy for ownership of the irrigated land has a positive and significant coefficient. Meaning that farmers owning the land they irrigate or more likely to invest in wells. Finally the decision to invest is dependent on the district. The second hurdle considers the intensity of investment. In this model three variables come out as significant. The effect of the district clearly shows that both decisions are different. In the second hurdle model it is observed that farmers in Akola invest the highest amounts. Maybe aquifer conditions in this district force farmers to drill deeper (more costly) wells. Also the irrigated area has an effect, farmers irrigating more land tend to spend more to assure their water access. Finally there is a significant positive effect of total income. With famers with a higher income investing more. In contrast to the first hurdle the quantity of past investments does not significantly influence the amount currently invested. It is furthermore found that if water availability in the wells was good in the past year that this lowers the spending. Against expectations and in contrast with the results of Hadrich et al (2012) we did not observe a significant effect of the households financial status (indebtedness or savings).
Some important conclusions for policy making can be derived from these results. First the hypothesis of the occurrence of a sunk cost fallacy is confirmed. To avoid the negative consequences of this phenomenon governments could more closely regulate the credit markets which currently allow farmers to build up more and more debts. Further it was observed that the intensity of investment increases under scarcity conditions. This seems to point to appropriative competition behavior as described by Varghese et al. (2013b). As a solution for this problem governments could install clearly defined and secure property rights over groundwater. Finally the fact that water access seems to influence the decision to invest supports current efforts of the Indian government to revitalize the traditional irrigation tanks. Bekele, G.& Mekonnen, A., 2012. Investments in Land Conservation in the Ethiopian Highlands: A Household Plot-level Analysis of the Roles of Poverty, Tenure Security, and Market Incentives. International Journal of Economics and Finance, 4(6), 32-50. Bhattarai, M. & Narayanamoorthy, A., 2003. Impact of Irrigation on Rural Poverty in India: An Aggregate Panel-Data Analysis. Water Policy, 5, 443-458. Chandrakanth, M.G., et al., 2004. Combating negative externalities of drought: a study of groundwater recharge through watershed. Economic and Political Weekly, 39(11), 1164-1170. Cragg, J., 1971. Some Statistical Models for Limited Dependent Variables with Application to the Demand for Durable Goods. Econometrica, 39, 829-44. Diwakara, H. & Chandrakanth, M.G., 2007. Beating negative externality through groundwater recharge in India: a resource economic analysis. Environment and Development Economics, 12, 271-296. Hadrich, J.C., et al., 2012. Incentives for Machinery Investment. Proceedings of the Agricultural and Applied Economics Association Annual Meeting, Seattle, Washington. Kajisa, K., et al., 2007. Effects on Poverty and Equity of the Decline in Collective Tank Irrigation Management in Tamil Nadu, India. Agricultural Economics, 36(3), 347-362. Manjunatha, A.V., et al., 2011. Impact of Groundwater Markets in India on Water Use Efficiency: A Data Envelopment Analysis Approach. Journal of Environmental Management, 92, 2924- 2929. MarÃ©chal, J.-C., 2009. EditorÂ’s Message: The Sunk Cost Fallacy of Deep Drilling. Hydrogeology Journal, 18, 287-289. Mukherji, A., 2007. The energy-irrigation nexus and its impact on groundwater markets in eastern Indo-Gangetic basin: Evidence from West Bengal, India. Energy Policy, 35(12), 6413Â–6430. Nagaraj, N., et al., 2005. Economic Analysis of Groundwater Markets in Central Dry Zone of Karnataka. Paper Presented at 5th Annual IWMI-TATA Partnerships Meet, Gujarat, India, 7p. Narayanamoorthy, A., 2007. Does groundwater irrigation reduce rural poverty? Evidence from Indian states. Irrigation and Drainage, 56(2-3), 349Â–362. Palanisami, K., C. Umetsu, and C.R. Ranganathan, 2008a. Why Farmers Still Invest in Wells in Hard-Rock Regions When the Water-Table Is Fast Declining? In Taniguchi, M., Burnett, W.C., Fukushima, Y., Haigh, M. & Umezawa, Y. (eds.): From Headwaters to the Ocean: Hydrological Changes and Watershed Management. Taylor and Francis, London, United Kingdom. pp. 503-508. Shah, T., et al., 2008. Groundwater governance through Electricity supply management: assessing an innovative intervention in Gujarat, western India. Agricultural Water Management, 95, 1233-1242. Saleth, M.R., et al., 2003. Dynamics of Irrigation- Poverty Linkages in Rural India: Analytical Framework and Empirical Analysis. Water Policy, 5, 459-473. Taylor, M. 2013. Liquid Debts: credit, groundwater and the social ecology of agrarian distress in Andhra Pradesh, India. Third World Quarterly, 34:4, 691-709 Vadivelu, A.G., 2009. Evolution of Property Rights Regimes in the Groundwater Economy of India Â–Constraints on moving towards a common property regime. Proceedings of the Development Studies Association Conference on "Current Crisis and New Opportunities", University of Ulster, England. Varghese, S.K., et al., 2013a. Are investments in groundwater irrigation profitable? A case of rice farms from South India. Journal Of The American Water Resources Association, 49(1), 52-66. Varghese, S.K., et al., 2013b. Estimating the causal effect of water scarcity on the groundwater use efficiency of rice farming in South India. Ecological Economics, 86, 55Â–64.