We present in table 9 some regression analysis results in order to more formally establish the observations presented in the earlier section. The regressions employ the OLS technique with clustering on individual subjects and robust standard errors.Clustering procedures allow heteroskedasticity between and within clusters, as well as autocorrelation within clusters. Wooldridge (2003) reviews some issues in the use of clustering for panel effects. Table 8 specifies the variables used in the regression analysis.
Table 8: Description of Variables Used in the Regression Analysis
Variable Brief Description
Individual Contribution Dependent: Takes a value between 0 and 10
Age In years
Female Takes the value 1 if female, 0 otherwise
Scheduled Caste (SC) Takes the value 1 if Scheduled Caste (SC), 0 otherwise Scheduled Tribe (ST) Takes the value 1 if Scheduled Tribe (ST), 0 otherwise Farmer Takes the value 1 if occupation is farming, 0 otherwise Lagged Total Contribution Sum of all individual contribution in the last period, varies between
0 and 100.
Period 6 Takes the value 1 for the 6th period, 0 otherwise Period 11 Takes the value 1 for the 11th period, 0 otherwise High Income Takes the value 1 is high (>Rs. 200,000 per annum), 0 otherwise
Literate Takes the value 1 if literate, 0 otherwise
The regression equation is given as:
Individual Contribution =ß0+ ß1Age + ß2Female + ß3Scheduled caste + ß4Scheduled tribe + ß5Farmer +ß6Lagged total contribution+ ß7Period 6 + ß8Period 11 + ß9 High income + ß10Literate. (2)
Table 9 shows that the variables age, gender, the ST indicator, the indicator variable for literate participants (literate) and the first lags of average contribution are all significant at the 1 percent level.13The adjusted R2 indicates that the covariates together explain 14 percent of the variation in the dependent variable.
Women invest more than men; ST participants invest more than the scheduled castes and upper caste participants.14 Given that we control for women’s investment with the female dummy, the difference in contribution between ST and other participants is driven by the behaviour of men. Age in years though significant at the 5 percent level, leads to only a marginal increase (Rs 0.02) in contribution controlling for other variables. Furthermore, literate participants invest on average Re. 0.60 more than illiterate individuals controlling for other demographic characteristics. However the effect of communication is negligible. In the regressions, the dummies for period 6 (the period immediately following the second round of face-to-face communication after period 5) and period 11 (the period immediately following the second round of face-to-face communication after period 10) are not significant at the 5 percent level.
13We also ran regressions with the lags of individual contribution and found those to be positively and significantly related to individual contribution in the immediate next period at the 1 percent level.
These are not reported above.
14Following up on observation 6, we also ran regressions where we included as a covariate the interaction term between the gender/ST dummy and the proportion of women/ST in a particular session.
However the gender/ST dummy was highly correlated with this interaction term (the correlation for gender/ST was 0.92, significant at the 1 percent level). The regressions with these highly correlated regressors have not been reported.
Table 9: Determinants of Individual Contribution
The fact that the aggregate contribution in one period makes subjects contribute higher in the next may be driven by the idea of conditional cooperation (or reciprocity), i.e. - a subject contributes more if everyone in the group contributes more, and the group composition is unchanged over the course of the 15 periods.
Croson (1998) and Croson et al. (2005) use lagged total contribution in the same way we do and obtain the same reciprocity result, i.e. – a subject’s behaviour in the current period is positively and significantly related to the total contribution by the group in the last period.
Estimation 1 Estimation 2 Estimation 3
No. of Obs. 390 360 320
Adjusted R² 0.13 0.14 0.12
Variable Coefficient (Std. Err.) Coefficient (Std. Err.) Coefficient (Std. Err.)
Age 0.02*** (0.005) 0.01** (0.005) 0.01*** (0.005)
Note 1: Least squares, robust standard errors clustered by individual. ‘***’, ‘**’, ‘*’ = Significant at the 1, 5, 10 percent respectively.
Note 2: We run three different models, one of which uses all the covariates (Estimation 3).
Estimation 2 omits the variable Income and estimation 1 omits Income and Literacy. Since we don’t have literacy and income data for some villages, estimation 2 has 450 less observations and estimation 3 has 1050 less observations than estimation 1.
5 Conclusion
We run a field experiment using a linear Voluntary Contributions Mechanism (VCM) game with face-to-face communication and a subject pool comprising villagers from the Gori-Ganga basin in the Kumaon region of Uttarakhand in India. Our experiment uses a large group size and static repetition. The former is uncommon among laboratory VCM experiments and the latter is uncommon among field experiments.
The pattern of contribution in our study differs somewhat from laboratory experiments using similar designs such as Isaac and Walker (1988b) and Bochet et al.
(2006) and Bochet and Putterman (2009). Specifically, even with a relatively low marginal per capita return and a group size of 10, we find an average contribution rate to the common pool that starts around 55 percent which diminishes only slightly at the end of the session to around 45 percent. Thus our subject pool on average contributes close to half their endowment even in the very last period of a finitely repeated one-shot VCM game. We also delve into the demographic characteristics of our subject pool and find interestingly, that individual contribution to the common pool is determined by gender, age, caste, literacy and history of cooperation in the experiment. As we randomize among different demographics we do not create experimental groups that have just women, tribals, educated people or senior citizens.
In our mixed demographic groups women on average contribute more to the public good. Scheduled tribe (ST) men contribute more than men from the scheduled caste (SC) and upper caste. Some of this high contribution is driven by women and scheduled tribe participants who contribute significantly higher amounts when they are in groups with more than half of their own type. However even after controlling for group composition we find that these demographic groups contribute higher than average amounts to the public good. Broadly our results parallel those from the peasant societies explored by Andersen et al. (2008) where the higher contribution of Khasi men drives the difference in contribution to the public good between the tribal and the Hindu society. Face-to-face communication is not seen to increase average individual contribution. We conjecture that a high group size and a low marginal return on the public good may make it difficult for subjects to coordinate group contributions to the level of efficiency seen in laboratory studies on the VCM.
Differential rates of contribution among different demographic groups for certain public goods may help develop policies that reflect the development goals that are relevant to these groups. The constitution of India in its 73rd amendment (1992) has allowed for reservation of posts in gram panchayats (village councils) for women, scheduled castes and scheduled tribes in a bid to empower these groups in terms of community decision making. Chattopadhyay and Duflo (2003, 2004) analyse a natural experiment that compares the level of public good provision at the village level in two districts, Rajasthan and West Bengal and find that the level of provision of certain types of public goods was higher in areas where the post of pradhan (head of the council) was reserved. Differential provision arises from the preferences of women or scheduled castes/tribes (SC/ST) that are a higher priority in a reserved gram panchayat.15 Given that these groups have been traditionally disadvantaged in rural India, reservation could thus help empower them by supplying public goods that are more relevant to their needs. An artefactual game like ours highlights this demographic heterogeneity and subsequent field research could help frame contexts that allow us to study more sharply, the nature and extent of the differential provision of public goods.
References
Andersen, S., E. Bulte, U. Gneezy, and J. A. List, (2008), “Do Women Supply More Public Goods than Men? Preliminary Experimental Evidence from Matrilineal and Patriarchal Societies,” American Economic Review, 98, 376-381.
Binswanger, H. P., (1981), “Attitudes toward Risk: Theoretical Implications of an Experiment in Rural India, ”Economic Journal, 91 , 867-890.
Bochet, O., T. Page, and L. Putterman, (2006), “Communication and Punishment in Voluntary Contribution Experiments,”Journal of Economic Behavior and Organization, 60, 11-26.
Bochet, O. and L. Putterman, (2009), “Not Just Babble: Opening the Black Box of Communication in a Voluntary Contribution Experiment,”European Economic Review, 53, 309-326.
15As the Government of India randomly performed the reservation of posts, there were negligible differences between reserved and unreserved gram panchayats on average, using a large sample.
Bohnet, I. and F. Greig, (2008), “Is There Reciprocity in a Reciprocal-Exchange Economy? Evidence of Gendered Norms from a Slum in Nairobi, Kenya,” Economic Inquiry, 46, 77-83.
Brown-Kruse, J. and D. Hummels, (1993), “Gender Effects in Laboratory Public Goods Contribution: Do Individuals Put Their Money Where Their Mouth Is ?,” Journal of Economic Behavior and Organization, 22, 255-267.
Cason, T., T. Saijo, and T. Yamato, (2002), “Voluntary Contribution and Spite in Public Good Provision Experiments: An International Comparison,”
Experimental Economics, 5, 133-153.
Chakravarty, S., C. Sebi, E. Somanathan, and E. Theophilus, (2009), “Does Social Capital Exist? A Definition and a Test,” Working Paper, ISI, Delhi.
Chakravarty, S., D. Friedman, G. Gupta, N. Hatekar, S. Mitraand , and S. Sunder, (2011), “Experimental Economics: A Survey,” Economic and Political Weekly, 46 , 39-78.
Chattopadhyay, R. and E. Duflo, (2003), “The Impact of Reservation in the Panchayati Raj: Evidence From a Nationwide Randomized Experiment,”
Working Paper, MIT, Cambridge MA.
Chattopadhyay, R. and E. Duflo, (2004), “Women as Policy Makers: Evidence From A Randomized Policy Experiment in India,” Econometrica, 72 , 1409 -1443.
Croson, R. T. A., (1998), “Theories of Altruism and Reciprocity: Evidence from Linear Public Goods Games,” Working Paper, The Wharton School, University of Pennsylvania.
Croson, R. T. A., E. Fatas, and T. Neugebauer, (2005), “Reciprocity, Matching and Conditional Cooperation in Two Public Goods Games,” Economics Letters, 87, 95–101.
Croson, R. T. A. and U. Gneezy, (2004), Gender Differences in Preferences, Mimeo, University of San Diego.
Croson, R. and J. Shang, (2008), “The Impact of Downward Social Information on Contribution Decisions,” Experimental Economics, 11, 221-233.
Eckel, C., and P. J. Grossman, (2008), “Differences in the Economic Decisions of Men and Women: Experimental Evidence.” In: Plott, C. and V. Smith, (eds.), Handbook of Experimental Results, Elsevier, New York, 1, 509-519.
Fehr, E. and S. Gachter, (2000), “Cooperation and Punishment in Public Goods Experiments,” American Economic Review, 90, 980-994.
Frey, B. S. and S. Meier, (2004), “Social Comparisons and Pro-Social Behaviour:
Testing "Conditional Cooperation" in a Field Experiment,” American Economic Review, 94, 1717–1722.
Hardin, G., (1968), “The Tragedy of the Commons,”Science, 162, 1243-1248.
Harrison, G. W. and J. A. List, (2004), “Field Experiments, ” Journal of Economic Literature, 42, 1009–1055.
Henrich, J., R. Boyd, S. Bowles, C. Camerer, H. Gintis, R. McElreath, and E. Fehr, (2001), “In Search of Homo Economicus: Experiments in 15 Small-Scale Societies,” American Economic Review, 91, 73-78.
Henrich, J., R. Boyd, S. Bowles, H. Gintis, E. Fehr, C. Camerer, R. McElreath, M.
Gurven, K. Hill, A. Barr, A. J. Ensminger, D. Tracer, F. Marlow, J. Patton, M.
Alvard, F. Gil-White, and N. Smith, (2005), “Economic Man in Cross-Cultural Perspective: Behavioural Experiments from 15 Small-Scale Societies,”
Behavioral and Brain Sciences, 28, 795-855.
Henrich, J., S. Heine, and A. Norenzayan, (2010), “The Weirdest People in the World?,” Behavioral and Brain Sciences, 33, 61-83.
Isaac, R. M. and J. Walker, (1988a), “Group Size Effects of Public Goods Provision:
An Experimental Examination,” Quarterly Journal of Economics, 103, 179-199.
Isaac, R. M. and J. Walker, (1988b), “Communication and Free Riding Behaviour:
The Voluntary Contribution Mechanism,” Economic Inquiry, 26, 585-608.
Keefer, P. and S. Knack, (2005), “Social Capital, Social Norms and the New Institutional Economics,” In: Ménard, C. and M. M. Shirley, (eds.), Handbook of New Institutional Economics, Springer, Netherlands, 701-725.
List, J. A., (2004), “Young, Selfish and Male: Field Evidence of Social Preferences,” Economic Journal, 114, 121-149.
Marwell, G. and R. Ames, (1980), “Experiments on the Provision of Public Goods, II. Provision Points, Stakes, Experience and the Free Rider Problem,” American Journal of Sociology, 85, 926-937.
Olson, M., (1965), The Logic of Collective Action: Public Goods and the Theory of Groups, Cambridge, Harward.
Ostrom, E., J. Walker, and R. Gardner, (1992), “Covenants With and Without a Sword: Self-Governance is Possible,” American Political Science Review, 86, 404-417.
Sobel, J., (2005), “Interdependent Preferences and Reciprocity,” Journal of Economic Literature, 43, 392-436.
Sugden, R., (1984), “Reciprocity: The Supply of Public Goods through Voluntary Contributions,” Economic Journal, 94, 772-787.
Von Furer-Haimendorf, C., (1983), “Tribal Problems in India,” In : Thapar , R. (ed.), Tribe, Caste, and Religion in India, Macmillan , New Delhi, 1-6.
Wooldridge, J., (1993), “Cluster-Sample Methods in Applied Econometrics,”
American Economic Review, 93, 133-138.