Irrigation and cooperation: An empirical analysis of 48 irrigation communities in South India

Economic Development and Cultural Change

Volumn 48, Issue 4, 2000, Pages 847-865

  Bardhan, Pranab ^a  

^a UNIVERSITY OF CALIFORNIA   (United States)

Author keywords

[No Author keywords available]

Indexed keywords

CANAL; COMMUNITY RESOURCE MANAGEMENT; IRRIGATION; WATER MANAGEMENT;

INDIA; TAMIL NADU;

EID: 0033810992     PISSN: 00130079     EISSN: None     Source Type: Journal    
DOI: 10.1086/452480     Document Type: Article

References (47)

1. I have elaborated on this theme in P. Bardhan, Land, Labor, and Rural Poverty (New York: Columbia University Press, 1984), chap. 16.
2. The better-known accounts are in E. Ostrom, Governing the Commons: The Evolution of Institutions for Collective Action (Cambridge: Cambridge University Press, 1990); R. Wade, Village Republics: Economic Conditions for Collective Action in South India (Cambridge: Cambridge University Press, 1987); and A. Vaidyanathan, "Water Control Institutions and Agriculture: A Comparative Perspective," Indian Economic Review 20, no. 1 (1984): 25-84. See also T. K. Jayaraman, "Farmers' Organizations in Surface Irrigation Projects: Two Empirical Studies from Gujarat," Economic and Political Weekly 16, no. 39 (1981): A89-A98; K. W. Easter and K. Palanisami, "Tank Irrigation in India and Thailand: An Example of Common Property Resource Management," staff paper (University of Minnesota, Department of Agricultural and Applied Economics, August 1986); N. Sengupta, Managing Common Property: Irrigation in India and the Philippines (New Delhi: Sage Publications, 1991); and A. Vaidyanathan, "Transferring Irrigation Management to Farmers," Economic and Political Weekly 29, no. 47 (1994): 2925-27.
3. The better-known accounts are in E. Ostrom, Governing the Commons: The Evolution of Institutions for Collective Action (Cambridge: Cambridge University Press, 1990); R. Wade, Village Republics: Economic Conditions for Collective Action in South India (Cambridge: Cambridge University Press, 1987); and A. Vaidyanathan, "Water Control Institutions and Agriculture: A Comparative Perspective," Indian Economic Review 20, no. 1 (1984): 25-84. See also T. K. Jayaraman, "Farmers' Organizations in Surface Irrigation Projects: Two Empirical Studies from Gujarat," Economic and Political Weekly 16, no. 39 (1981): A89-A98; K. W. Easter and K. Palanisami, "Tank Irrigation in India and Thailand: An Example of Common Property Resource Management," staff paper (University of Minnesota, Department of Agricultural and Applied Economics, August 1986); N. Sengupta, Managing Common Property: Irrigation in India and the Philippines (New Delhi: Sage Publications, 1991); and A. Vaidyanathan, "Transferring Irrigation Management to Farmers," Economic and Political Weekly 29, no. 47 (1994): 2925-27.
4. The better-known accounts are in E. Ostrom, Governing the Commons: The Evolution of Institutions for Collective Action (Cambridge: Cambridge University Press, 1990); R. Wade, Village Republics: Economic Conditions for Collective Action in South India (Cambridge: Cambridge University Press, 1987); and A. Vaidyanathan, "Water Control Institutions and Agriculture: A Comparative Perspective," Indian Economic Review 20, no. 1 (1984): 25-84. See also T. K. Jayaraman, "Farmers' Organizations in Surface Irrigation Projects: Two Empirical Studies from Gujarat," Economic and Political Weekly 16, no. 39 (1981): A89-A98; K. W. Easter and K. Palanisami, "Tank Irrigation in India and Thailand: An Example of Common Property Resource Management," staff paper (University of Minnesota, Department of Agricultural and Applied Economics, August 1986); N. Sengupta, Managing Common Property: Irrigation in India and the Philippines (New Delhi: Sage Publications, 1991); and A. Vaidyanathan, "Transferring Irrigation Management to Farmers," Economic and Political Weekly 29, no. 47 (1994): 2925-27.
5. The better-known accounts are in E. Ostrom, Governing the Commons: The Evolution of Institutions for Collective Action (Cambridge: Cambridge University Press, 1990); R. Wade, Village Republics: Economic Conditions for Collective Action in South India (Cambridge: Cambridge University Press, 1987); and A. Vaidyanathan, "Water Control Institutions and Agriculture: A Comparative Perspective," Indian Economic Review 20, no. 1 (1984): 25-84. See also T. K. Jayaraman, "Farmers' Organizations in Surface Irrigation Projects: Two Empirical Studies from Gujarat," Economic and Political Weekly 16, no. 39 (1981): A89-A98; K. W. Easter and K. Palanisami, "Tank Irrigation in India and Thailand: An Example of Common Property Resource Management," staff paper (University of Minnesota, Department of Agricultural and Applied Economics, August 1986); N. Sengupta, Managing Common Property: Irrigation in India and the Philippines (New Delhi: Sage Publications, 1991); and A. Vaidyanathan, "Transferring Irrigation Management to Farmers," Economic and Political Weekly 29, no. 47 (1994): 2925-27.
6. The better-known accounts are in E. Ostrom, Governing the Commons: The Evolution of Institutions for Collective Action (Cambridge: Cambridge University Press, 1990); R. Wade, Village Republics: Economic Conditions for Collective Action in South India (Cambridge: Cambridge University Press, 1987); and A. Vaidyanathan, "Water Control Institutions and Agriculture: A Comparative Perspective," Indian Economic Review 20, no. 1 (1984): 25-84. See also T. K. Jayaraman, "Farmers' Organizations in Surface Irrigation Projects: Two Empirical Studies from Gujarat," Economic and Political Weekly 16, no. 39 (1981): A89-A98; K. W. Easter and K. Palanisami, "Tank Irrigation in India and Thailand: An Example of Common Property Resource Management," staff paper (University of Minnesota, Department of Agricultural and Applied Economics, August 1986); N. Sengupta, Managing Common Property: Irrigation in India and the Philippines (New Delhi: Sage Publications, 1991); and A. Vaidyanathan, "Transferring Irrigation Management to Farmers," Economic and Political Weekly 29, no. 47 (1994): 2925-27.
7. The better-known accounts are in E. Ostrom, Governing the Commons: The Evolution of Institutions for Collective Action (Cambridge: Cambridge University Press, 1990); R. Wade, Village Republics: Economic Conditions for Collective Action in South India (Cambridge: Cambridge University Press, 1987); and A. Vaidyanathan, "Water Control Institutions and Agriculture: A Comparative Perspective," Indian Economic Review 20, no. 1 (1984): 25-84. See also T. K. Jayaraman, "Farmers' Organizations in Surface Irrigation Projects: Two Empirical Studies from Gujarat," Economic and Political Weekly 16, no. 39 (1981): A89-A98; K. W. Easter and K. Palanisami, "Tank Irrigation in India and Thailand: An Example of Common Property Resource Management," staff paper (University of Minnesota, Department of Agricultural and Applied Economics, August 1986); N. Sengupta, Managing Common Property: Irrigation in India and the Philippines (New Delhi: Sage Publications, 1991); and A. Vaidyanathan, "Transferring Irrigation Management to Farmers," Economic and Political Weekly 29, no. 47 (1994): 2925-27.
8. The better-known accounts are in E. Ostrom, Governing the Commons: The Evolution of Institutions for Collective Action (Cambridge: Cambridge University Press, 1990); R. Wade, Village Republics: Economic Conditions for Collective Action in South India (Cambridge: Cambridge University Press, 1987); and A. Vaidyanathan, "Water Control Institutions and Agriculture: A Comparative Perspective," Indian Economic Review 20, no. 1 (1984): 25-84. See also T. K. Jayaraman, "Farmers' Organizations in Surface Irrigation Projects: Two Empirical Studies from Gujarat," Economic and Political Weekly 16, no. 39 (1981): A89-A98; K. W. Easter and K. Palanisami, "Tank Irrigation in India and Thailand: An Example of Common Property Resource Management," staff paper (University of Minnesota, Department of Agricultural and Applied Economics, August 1986); N. Sengupta, Managing Common Property: Irrigation in India and the Philippines (New Delhi: Sage Publications, 1991); and A. Vaidyanathan, "Transferring Irrigation Management to Farmers," Economic and Political Weekly 29, no. 47 (1994): 2925-27.
9. For a similar analysis of determinants of the adoption of distributive rules and of cooperative maintenance efforts with data from 54 farmer-managed surface irrigation systems in central Mexico, see J. Dayton-Johnson, Rules and Cooperation on the Local Commons: Theory with Evidence from Mexico (Ph.D. diss., University of California, Berkeley, 1998). I also compare some of the results with those for irrigation systems in Nepal by W. F. Lam, Governing Irrigation Systems in Nepal: Institutions, Infrastructure, and Collective Action (Oakland, Calif.: ICS Press, 1998).
10. For a similar analysis of determinants of the adoption of distributive rules and of cooperative maintenance efforts with data from 54 farmer-managed surface irrigation systems in central Mexico, see J. Dayton-Johnson, Rules and Cooperation on the Local Commons: Theory with Evidence from Mexico (Ph.D. diss., University of California, Berkeley, 1998). I also compare some of the results with those for irrigation systems in Nepal by W. F. Lam, Governing Irrigation Systems in Nepal: Institutions, Infrastructure, and Collective Action (Oakland, Calif.: ICS Press, 1998).
11. Unfortunately, the data and descriptions regarding some of these organizational variables were not systematic or complete enough to allow reliable coding and inclusion in the subsequent regression analysis.
12. Below I will comment on how these may not always be good indicators of cooperation among the farmers.
13. I have also used an alternative variable, NUM-ACRE, the number of households using the irrigation source per acre of the ayacut area, but usually this variable is less significant than NUMUSERS in the regressions.
14. M. Olson, The Logic of Collective Action: Public Goods and the Theory of Groups (Cambridge, Mass: Harvard University Press, 1965).
15. F. Weissing and E. Ostrom, "Irrigation Institutions and the Games Irrigators Play: Rule Enforcement without Guards," in Game Equilibrium Models, vol. 2, ed. R. Selten (Berlin: Springer, 1990).
16. Of course, other things do not remain the same. We shall see in the data that the existence of guards is positively (though not very significantly) correlated with NUMUSERS, and guards have a positive influence on cooperation.
17. For a theoretical analysis of some of the complexities, see J. M. Baland and J. P. Platteau, "Wealth Inequality and Efficiency in the Commons: The Unregulated Case," Oxford Economic Papers 49 (1997): 451-82, and " Wealth Inequality and Efficiency in the Commons: The Regulated Case," Oxford Economic Papers 50 (1998): 1-22; and J. Dayton-Johnson and P. Bardhan, "Inequality and Conservation on the Local Commons: A Theoretical Exercise," CIDER Working Paper no. 096-071 (University of California, Berkeley, 1998).
18. For a theoretical analysis of some of the complexities, see J. M. Baland and J. P. Platteau, "Wealth Inequality and Efficiency in the Commons: The Unregulated Case," Oxford Economic Papers 49 (1997): 451-82, and " Wealth Inequality and Efficiency in the Commons: The Regulated Case," Oxford Economic Papers 50 (1998): 1-22; and J. Dayton-Johnson and P. Bardhan, "Inequality and Conservation on the Local Commons: A Theoretical Exercise," CIDER Working Paper no. 096-071 (University of California, Berkeley, 1998).
19. For a theoretical analysis of some of the complexities, see J. M. Baland and J. P. Platteau, "Wealth Inequality and Efficiency in the Commons: The Unregulated Case," Oxford Economic Papers 49 (1997): 451-82, and " Wealth Inequality and Efficiency in the Commons: The Regulated Case," Oxford Economic Papers 50 (1998): 1-22; and J. Dayton-Johnson and P. Bardhan, "Inequality and Conservation on the Local Commons: A Theoretical Exercise," CIDER Working Paper no. 096-071 (University of California, Berkeley, 1998).
20. Olson.
21. For a theoretical analysis of a U-shaped relationship between wealth inequality and cooperation in the context of a two-player noncooperative model of conservation of a common-pool resource, see Dayton-Johnson and Bardhan.
22. It is well documented in the irrigation literature that no single-dimensional variable can adequately capture the quality of water supply, which depends on many factors (e.g., volume, timing, crop produced, and cultivation practice). As an alternative to the variable AVGACCES I have also tried a variable indicating whether water is sufficient for the paddy crop in the primary season, but this variable turned out to be less significant in the statistical analysis than AVGACCES.
23. Alternatively, the farmers may get involved in cooperation, when the PWD fails to do its job, but this seems unlikely from the details of the survey's description of the irrigation units.
24. I have tried another variable to indicate access of farmers to alternative, private, sources of water within the village (e.g., pump sets and open wells). However, in the data set this variable is highly correlated with variables like CANAL. One possible reason for this may be that government canals increase the subsoil water for private wells.
25. This is true except in the case of MARKET, where the negative effect may be modified by the market opportunities for produce raising the return to cooperation on water.
26. P. Seabright, "Is Cooperation Habit-Forming?" in The Environment and Emerging Development Issues, ed. P. Dasgupta and K. G. Maler (Oxford: Clarendon, 1997).
27. Since this is a logit analysis, the signs are to be appropriately interpreted as indicating directions of changes in the probability of the dependent variable taking on a particular value with changes in the values of an independent variable, even though in the statement of the results I sometimes do not spell it out fully.
28. Dayton-Johnson; Lam (both cited in n. 3 above).
29. From the data of the 480 sampled farmers it is clear that when average farmers believe that the water rules have been crafted jointly (rather than by the elite or by the government), they are more likely to have positive comments about the water-allocation system and about rule compliance by other farmers.
30. This is also empirically corroborated by Lam.
31. In the corresponding OLS regressions neither NUMUSERS nor NUM-ACRE is at all significant.
32. There is reason to believe that access to irrigation refers more to general accessibility to water sources and structures than to the flow of water through the field channels, so the reverse causation (i.e., better-maintained channels, implying better access to irrigation water) is an unlikely explanation of the positive coefficient. In fact, in general in the canal areas AVGACCES is very low, and yet DISTFDCH is high.
33. Lam finds a positive coefficient for his explanatory variable on lining but in a different context comments on the design of irrigation infrastructure constructed by the Department of Irrigation (DOI) in Nepal, which may be relevant for this case: "Although the unlined canal was prone to damage, it was easy to maintain. Whenever there was leakage, farmers could easily locate the leaks and seal the leaks with mud. Now that the DOI has lined the canals with bricks, farmers find it difficult to locate the leaks at all. Even if they can locate them, it is difficult to seal them. Farmers complain that there is little that they can do to keep the canal in good condition" (p. 180).
34. Following a suggestion of Gershon Feder, I ran a separate regression for the 30 units that are lined (partially or fully). The results are similar to those reported in table 2, except that the significance of some of the variables diminishes (partly because of the reduced number of observations).
35. Dayton-Johnson.
36. In four of these six units the PWD, rather than a traditional village organization, looks after the field-channel maintenance as well (though the farmers make some cash contributions). This means good maintenance may not always be interpreted simply as an indicator of cooperation among the farmers.
37. In some alternative regressions (not reported here) I have found that the interactive effect of PGUARD and NUMUSERS has a significantly negative coefficient, indicating that the effectiveness of the guard system in encouraging cooperation decreases as the number of users increases.
38. Of course, a two-stage regression is somewhat problematic when the second stage is a nonlinear logit model. One can use OLS as a check to confirm the results. But the problem with using OLS is that with dependent variables that are grouped, the error terms cannot be assumed to be normally distributed. The assumption is even worse, and the results more biased, if OLS is used when the dependent variable takes on only 0-1 values, as in the case of H2OCOOP and VIOLH2O. In the case of DISTFDCH I have used OLS as a check, and the results are largely confirmed (except for the variable NUMUSERS, as explained in n. 22).
39. It is also the case that in canal areas the water available to the ayacut is from a flow (rather than from a storage reservoir as in the case of tanks), and guarding against water theft may therefore be more serious in the case of tanks than in the case of canals. In any case the TRADLORG variable is not statistically significant as a direct determinant in the cooperation regressions, and it therefore can be legitimately used as an instrumental variable.
40. Only about one in five villages has the proportional cost-sharing rule in the data set.
41. In one-third of the villages in the data set formal water rights, as opposed to customary rights, exist. Again these are likely to be determined by historical-conventional factors, and therefore this variable can be used as a legitimate instrument.
42. In some alternative regressions (not reported here) I have found that the interactive effect of SHRPROP and GINI is positive in the DISTFDCH regression, which suggests that the higher the level of inequality the more positive the influence of proportional cost sharing on the quality of maintenance.
43. Exactly the same pattern is observed in the other 12 canal-based units.
44. For example, in 29 out of 48 villages rule violations are mainly by the rich farmers or upper castes, as reported in the ayacut-level survey. Also at the level of the sampled farmers, the average farmer's perception is that the rich (with respect to landholding, the top four out of the 10 sampled farmers in the village) comply with rules significantly less than do the poor (the bottom four out of the 10 sampled farmers), particularly in water allocation and financial cost sharing.
45. About 44% of the villages in the data set have a market for well water.
46. There is also evidence in the data set that the farmers contribute to the village collective fund much less when the government is involved, presumably because the government is then assumed to be the financier of last resort.
47. Lam (n. 3 above).