Summary

This document includes the populated pre-analysis plan for “Why Discussion Rules Matter.” The black text is the original, unchanged text from the pre-analysis plan. Blue text indicates deviations and precisions.

First group of hypotheses: Institutional Centrality

This measures the degree of de facto centrality (or authority) of the de jure central official (the Sarpanch) in a typical decision-making process.

IC1 (H3.1): Oral participation to the meeting will be more likely for those sarpanches exposed to treatment requiring the sarpanch to speak (T1 and T2) versus those not exposed (Control).

Assessed by regressing “equal participation” on treatment status (T1 or T2 versus control). Two-tailed tests are used in all cases.
We check the robustness of results with bivariate regressions on treatment status and a multivariate regression including a standard set of demographic controls. We also control for the potential presence of a sarpanch relative during the interview (“the interview” being here understood as the group interview), both separately and in interaction with the gender of the sarpanch.
In order to measure equal participation, we make use of the following two survey items:
- Q29: Does the sarpanch respond first to the question (Y/N)
- Q30: Does the sarpanch speak at all? (Y/N)
In additional analyses, we disentangle between T1 and T2. We also include the series of individual and village level controls presented in G2.
Deviation 1: Living with in-laws and one’s own parents were planned to be included in the wealth index. We instead include them as separate controls, because they are not indicative of wealth.
Precision 1: Village-level controls were not presented in cell G2 of the PAP spreadsheet; instead, they were presented in cell H4. This was a typo on our part.

IC1 (H3.1) table, grouping T1/T2 together
	DV: Sarpanch responded first			DV: Sarpanch spoke at all
	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls
+ p < 0.1, * p < 0.05, p < 0.01, * p < 0.001
Each model is a linear regression of each outcome on an indicator for being assigned to discussion rules (1) versus not (0). Unadjusted and covariate-adjusted models presented. All regressions use robust standard errors.
(Intercept)	0.323***	0.406***	0.313**	0.802***	0.907***	0.938***
	(0.034)	(0.081)	(0.113)	(0.029)	(0.063)	(0.090)
Discussion rules	0.178***	0.138**	0.134**	0.055	0.039	0.036
	(0.042)	(0.042)	(0.043)	(0.034)	(0.034)	(0.035)
Sarpanch woman		-0.286*	-0.221		-0.019	0.044
		(0.126)	(0.139)		(0.119)	(0.108)
Sarpanch relative present		-0.367+	-0.392*		-0.096	-0.122
		(0.197)	(0.193)		(0.200)	(0.214)
Sarpanch SC		-0.101	-0.035		0.015	0.007
		(0.086)	(0.088)		(0.077)	(0.074)
Sarpanch OBC		-0.002	0.035		-0.027	-0.009
		(0.055)	(0.057)		(0.044)	(0.043)
Education above median		0.172***	0.176***		0.079**	0.073*
		(0.044)	(0.045)		(0.031)	(0.031)
Wealth above median		-0.022	-0.015		0.063*	0.054+
		(0.041)	(0.043)		(0.032)	(0.032)
Lives with in-laws		0.090	0.070		-0.080	-0.125
		(0.110)	(0.118)		(0.101)	(0.091)
Lives with own parents		0.076	0.077		-0.057	-0.053
		(0.048)	(0.051)		(0.038)	(0.039)
Years as sarpanch		-0.005	0.002		-0.013	-0.012
		(0.013)	(0.014)		(0.010)	(0.011)
Gender quota		-0.011	-0.052		-0.073	-0.098
		(0.069)	(0.075)		(0.065)	(0.068)
SC quota		0.014	-0.050		-0.052	-0.042
		(0.085)	(0.088)		(0.076)	(0.073)
OBC quota		0.048	0.010		0.088*	0.092*
		(0.053)	(0.055)		(0.042)	(0.040)
Sarpanch woman:Sarpanch relative present		0.218	0.165		-0.090	-0.126
		(0.210)	(0.204)		(0.217)	(0.233)
Percent of village land held by biggest landholding caste			-0.000			-0.001
			(0.001)			(0.001)
Distance (km) from block center			0.002			0.002
			(0.002)			(0.001)
Directly elected			0.087+			0.017
			(0.045)			(0.033)
Num.Obs.	605	549	506	605	549	506
R2	0.028	0.167	0.195	0.005	0.120	0.147

IC1 (H3.1) table, T1 versus control
	DV: Sarpanch responded first			DV: Sarpanch spoke at all
	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls
+ p < 0.1, * p < 0.05, p < 0.01, * p < 0.001
Each model is a linear regression of each outcome on an indicator for being assigned to T1 (1) versus control (0). Unadjusted and covariate-adjusted models presented. All regressions use robust standard errors.
(Intercept)	0.323***	0.471***	0.368**	0.826***	0.844***	0.921***
	(0.034)	(0.090)	(0.125)	(0.019)	(0.068)	(0.109)
T1	0.338***	0.301***	0.310***	0.039	0.048	0.050
	(0.047)	(0.047)	(0.048)	(0.030)	(0.038)	(0.039)
Sarpanch woman		-0.378**	-0.314*		0.037	0.021
		(0.130)	(0.135)		(0.142)	(0.139)
Sarpanch relative present		-0.671***	-0.641***		-0.303	-0.338
		(0.142)	(0.143)		(0.309)	(0.323)
Sarpanch SC		-0.214*	-0.161		0.011	-0.015
		(0.101)	(0.112)		(0.094)	(0.094)
Sarpanch OBC		-0.065	-0.030		-0.050	-0.027
		(0.059)	(0.060)		(0.050)	(0.050)
Education above median		0.194***	0.178***		0.121**	0.098**
		(0.051)	(0.052)		(0.037)	(0.037)
Wealth above median		0.014	0.017		0.067+	0.048
		(0.047)	(0.049)		(0.039)	(0.040)
Lives with in-laws		0.146	0.143		-0.113	-0.109
		(0.108)	(0.110)		(0.132)	(0.126)
Lives with own parents		0.087	0.109+		-0.033	-0.029
		(0.056)	(0.057)		(0.042)	(0.044)
Years as sarpanch		-0.021	-0.018		0.001	-0.000
		(0.015)	(0.016)		(0.012)	(0.013)
Gender quota		-0.026	-0.061		-0.101	-0.093
		(0.082)	(0.086)		(0.071)	(0.077)
SC quota		0.158	0.111		-0.042	-0.009
		(0.101)	(0.110)		(0.094)	(0.092)
OBC quota		0.035	0.004		0.086+	0.094*
		(0.057)	(0.059)		(0.048)	(0.047)
Sarpanch woman:Sarpanch relative present		0.543**	0.434*		0.172	0.134
		(0.170)	(0.169)		(0.324)	(0.339)
Percent of village land held by biggest landholding caste			0.001			-0.001
			(0.001)			(0.001)
Distance (km) from block center			-0.000			0.000
			(0.002)			(0.002)
Directly elected			0.123*			0.042
			(0.049)			(0.039)
Num.Obs.	407	370	340	605	370	340
R2	0.114	0.308	0.344	0.003	0.136	0.154

IC1 (H3.1) table, T2 versus control
	DV: Sarpanch responded first			DV: Sarpanch spoke at all
	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls
+ p < 0.1, * p < 0.05, p < 0.01, * p < 0.001
Each model is a linear regression of each outcome on an indicator for being assigned to T2 (1) versus control (0). Unadjusted and covariate-adjusted models presented. All regressions use robust standard errors.
(Intercept)	0.323***	0.340***	0.237	0.835***	0.944***	0.876***
	(0.034)	(0.101)	(0.144)	(0.018)	(0.084)	(0.118)
T2	0.005	-0.040	-0.047	0.013	0.014	0.015
	(0.048)	(0.049)	(0.051)	(0.031)	(0.039)	(0.041)
Sarpanch woman		-0.256	-0.209		-0.074	0.087
		(0.165)	(0.189)		(0.177)	(0.166)
Sarpanch relative present		-0.198	-0.229		-0.116	-0.126
		(0.274)	(0.272)		(0.264)	(0.282)
Sarpanch SC		-0.082	-0.013		-0.039	-0.012
		(0.100)	(0.103)		(0.109)	(0.106)
Sarpanch OBC		-0.009	0.021		-0.038	-0.006
		(0.066)	(0.069)		(0.060)	(0.059)
Education above median		0.165**	0.184**		0.050	0.060
		(0.055)	(0.057)		(0.042)	(0.043)
Wealth above median		-0.049	-0.045		0.064	0.063
		(0.050)	(0.053)		(0.041)	(0.041)
Lives with in-laws		0.050	0.062		-0.095	-0.168
		(0.151)	(0.164)		(0.142)	(0.135)
Lives with own parents		0.118+	0.094		-0.052	-0.049
		(0.060)	(0.063)		(0.052)	(0.054)
Years as sarpanch		-0.001	0.002		-0.022	-0.019
		(0.016)	(0.018)		(0.014)	(0.016)
Gender quota		0.061	0.002		-0.006	-0.104
		(0.078)	(0.090)		(0.099)	(0.099)
SC quota		0.020	-0.048		0.019	0.000
		(0.098)	(0.103)		(0.108)	(0.105)
OBC quota		0.077	0.044		0.122*	0.128*
		(0.065)	(0.069)		(0.055)	(0.053)
Sarpanch woman:Sarpanch relative present		0.082	0.047		-0.151	-0.183
		(0.281)	(0.274)		(0.286)	(0.303)
Percent of village land held by biggest landholding caste			0.000			-0.000
			(0.001)			(0.001)
Distance (km) from block center			0.002			0.002
			(0.002)			(0.002)
Directly elected			0.125*			0.012
			(0.055)			(0.045)
Num.Obs.	390	351	325	605	351	325
R2	0.000	0.136	0.169	0.000	0.127	0.158

IC2 (H3.2): Institutional centrality of the sarpanch will be higher for those exposed to treatment requiring the sarpanch to speak (T1 and T2) versus not exposed (Control).

We focus on differences in (i) means and (ii) the graphical distribution of values across treatment (clubbing T1 and T2 together) and control conditions for each dependent variable, with respect to our measures of institutional centrality (both perceptions and behaviors).
We also control for the potential presence of a sarpanch relative during the interview (“the interview” being here understood as the group interview), both separately and in interaction with the gender of the sarpanch. Cluster-corrected standard errors will be used at the taluka level. We check the robustness of results with bivariate regressions on treatment status and a multivariate regression including a standard set of demographic controls.
We draw on our group survey instrument to develop three primary measures of institutional centrality (IC), below. We include additional measures of IC in the Supplemental Appendix.
- Q35: “Who made the final decision?” (for the budget - enumerator codes, 1: sarpanch)
- Q39: Responded first to the question: What are the roles and responsibilities of the sarpanch? Which are most important to you? (1=Yes/0=anything else)
- Q91: “For what approximate percentage of the discussion did the sarpanch speak?” (0-100%, normalized to 0-1 scale)
In additional analyses, we disentangle the impact of T1 and T2. We also run bivariate regressions, clustering at the taluka level, and we also include the series of individual and village level demographic controls presented in G2.
Deviation 1: Living with in-laws and one’s own parents were planned to be included in the wealth index. We instead include them as separate controls, because they are not indicative of wealth.
Deviation 2: We originally planned to use clustered standard errors at the taluka level. This was an error on our part: because the treatment assignment occurs at the level of each village council, we do not cluster standard errors in these analyses.
Deviation 3: Because the first two variables are binary, comparing graphical distributions is not more informative than running regressions. However, we do still compare graphical distributions between treatment/control for the third outcome variable.
Precision 1: Village-level controls were not presented in cell G2 of the PAP spreadsheet; instead, they were presented in cell H4. This was a typo on our part.
Precision 2: We previously did not state how we would formally test whether the distributions of outcome variables under each treatment condition are statistically different from each other. We do so here using an asymptotic two-sample Kolmogorov-Smirnov test.

IC2 (H3.2) table, grouping T1/T2 together
	DV: Sarpanch made final budget decision			DV: Sarpanch responded first to roles/resp.			DV: Proportion of time sarpanch spoke
	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls
+ p < 0.1, * p < 0.05, p < 0.01, * p < 0.001
Each model is a linear regression of each outcome on an indicator for being assigned to discussion rules (1) versus not (0). Unadjusted and covariate-adjusted models presented. All regressions use robust standard errors.
(Intercept)	0.193***	0.345***	0.472***	0.786***	0.804***	0.727***	0.256***	0.279***	0.299***
	(0.029)	(0.077)	(0.107)	(0.030)	(0.069)	(0.093)	(0.011)	(0.024)	(0.035)
Discussion rules	0.122***	0.094*	0.101*	0.022	-0.001	-0.014	0.034*	0.015	0.015
	(0.037)	(0.038)	(0.040)	(0.035)	(0.037)	(0.036)	(0.014)	(0.012)	(0.013)
Sarpanch woman		-0.141	-0.106		-0.095	-0.029		-0.100**	-0.072+
		(0.119)	(0.135)		(0.130)	(0.127)		(0.038)	(0.040)
Sarpanch relative present		-0.001	-0.028		0.145***	0.126*		0.025	0.013
		(0.227)	(0.232)		(0.039)	(0.052)		(0.126)	(0.126)
Sarpanch SC		-0.168+	-0.180+		-0.039	0.009		0.016	0.025
		(0.095)	(0.098)		(0.075)	(0.076)		(0.021)	(0.022)
Sarpanch OBC		-0.055	-0.047		0.017	0.073		0.008	0.019
		(0.051)	(0.054)		(0.049)	(0.050)		(0.016)	(0.017)
Education above median		0.106**	0.121**		0.122***	0.129***		0.081***	0.079***
		(0.040)	(0.041)		(0.034)	(0.035)		(0.012)	(0.013)
Wealth above median		-0.023	-0.026		0.023	0.006		0.011	0.010
		(0.038)	(0.039)		(0.036)	(0.037)		(0.012)	(0.012)
Lives with in-laws		-0.109	-0.120		0.087	0.031		-0.006	-0.013
		(0.107)	(0.119)		(0.118)	(0.115)		(0.035)	(0.036)
Lives with own parents		-0.017	-0.041		0.012	0.003		0.026+	0.023
		(0.042)	(0.046)		(0.041)	(0.040)		(0.014)	(0.014)
Years as sarpanch		0.006	0.009		-0.015	-0.009		0.005	0.004
		(0.013)	(0.013)		(0.012)	(0.012)		(0.004)	(0.004)
Gender quota		0.006	-0.015		-0.110+	-0.156*		-0.018	-0.037+
		(0.057)	(0.064)		(0.063)	(0.065)		(0.018)	(0.019)
SC quota		0.115	0.122		0.079	0.051		-0.041*	-0.048*
		(0.093)	(0.096)		(0.072)	(0.072)		(0.020)	(0.022)
OBC quota		0.029	0.010		0.042	0.020		0.001	-0.007
		(0.050)	(0.053)		(0.048)	(0.047)		(0.016)	(0.016)
Sarpanch woman:Sarpanch relative present		-0.051	-0.030		-0.308**	-0.309**		-0.103	-0.104
		(0.234)	(0.240)		(0.099)	(0.108)		(0.128)	(0.128)
Percent of village land held by biggest landholding caste			-0.002*			0.000			-0.000
			(0.001)			(0.001)			(0.000)
Distance (km) from block center			0.000			0.002*			-0.000
			(0.002)			(0.001)			(0.001)
Directly elected			0.022			0.008			0.026*
			(0.042)			(0.036)			(0.013)
Num.Obs.	605	549	506	605	549	506	604	548	505
R2	0.016	0.119	0.128	0.001	0.087	0.119	0.010	0.366	0.376

IC2 (H3.2) table, comparing T1 to control
	DV: Sarpanch made final budget decision			DV: Sarpanch responded first to roles/resp.			DV: Proportion of time sarpanch spoke
	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls
+ p < 0.1, * p < 0.05, p < 0.01, * p < 0.001
Each model is a linear regression of each outcome on an indicator for being assigned to T1 (1) versus control (0). Unadjusted and covariate-adjusted models presented. All regressions use robust standard errors.
(Intercept)	0.193***	0.364***	0.445***	0.786***	0.795***	0.821***	0.256***	0.302***	0.332***
	(0.029)	(0.084)	(0.122)	(0.030)	(0.081)	(0.104)	(0.011)	(0.029)	(0.042)
T1	0.063	0.044	0.045	0.051	0.033	0.017	0.036*		0.015
	(0.041)	(0.043)	(0.046)	(0.039)	(0.040)	(0.039)	(0.016)		(0.014)
Sarpanch woman		-0.011	-0.020		0.036	0.052		-0.116**	-0.098**
		(0.128)	(0.138)		(0.135)	(0.124)		(0.038)	(0.037)
Sarpanch relative present		-0.079	-0.140		0.091	0.017		-0.120	-0.127
		(0.273)	(0.287)		(0.064)	(0.091)		(0.102)	(0.104)
Sarpanch SC		-0.147	-0.155		-0.010	0.026		-0.008	-0.005
		(0.097)	(0.102)		(0.091)	(0.093)		(0.027)	(0.030)
Sarpanch OBC		-0.067	-0.058		0.043	0.083		-0.000	0.009
		(0.057)	(0.060)		(0.057)	(0.058)		(0.020)	(0.020)
Education above median		0.120**	0.121*		0.129**	0.126**		0.101***	0.097***
		(0.046)	(0.047)		(0.041)	(0.042)		(0.015)	(0.015)
Wealth above median		-0.034	-0.024		0.011	-0.002		0.005	0.004
		(0.043)	(0.045)		(0.043)	(0.044)		(0.014)	(0.014)
Lives with in-laws		-0.217+	-0.209+		-0.071	-0.078		0.005	0.002
		(0.114)	(0.122)		(0.127)	(0.116)		(0.035)	(0.032)
Lives with own parents		-0.070	-0.087+		-0.042	-0.052		0.003	-0.001
		(0.048)	(0.053)		(0.050)	(0.049)		(0.017)	(0.018)
Years as sarpanch		0.016	0.022		-0.007	-0.004		0.006	0.005
		(0.015)	(0.016)		(0.014)	(0.014)		(0.005)	(0.005)
Gender quota		-0.057	-0.038		-0.099	-0.154*		-0.030	-0.043*
		(0.066)	(0.073)		(0.068)	(0.065)		(0.021)	(0.022)
SC quota		0.051	0.052		0.121	0.100		-0.026	-0.029
		(0.094)	(0.098)		(0.086)	(0.088)		(0.026)	(0.028)
OBC quota		0.072	0.056		0.030	0.028		0.001	-0.009
		(0.056)	(0.059)		(0.056)	(0.054)		(0.018)	(0.019)
Sarpanch woman:Sarpanch relative present		0.035	0.078		-0.253*	-0.202		0.040	0.031
		(0.281)	(0.295)		(0.125)	(0.142)		(0.104)	(0.107)
Percent of village land held by biggest landholding caste			-0.002*			-0.001			-0.000
			(0.001)			(0.001)			(0.000)
Distance (km) from block center			0.002			0.004*			-0.000
			(0.002)			(0.002)			(0.001)
Directly elected			0.011			0.024			0.024
			(0.049)			(0.042)			(0.016)
treatment_1_2_together								0.013
								(0.014)
Num.Obs.	407	370	340	407	370	340	406	369	339
R2	0.006	0.141	0.141	0.004	0.097	0.148	0.013	0.396	0.398

IC2 (H3.2) table, comparing T2 to control
	DV: Sarpanch made final budget decision			DV: Sarpanch responded first to roles/resp.			DV: Proportion of time sarpanch spoke
	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls
+ p < 0.1, * p < 0.05, p < 0.01, * p < 0.001
Each model is a linear regression of each outcome on an indicator for being assigned to T2 (1) versus control (0). Unadjusted and covariate-adjusted models presented. All regressions use robust standard errors.
(Intercept)	0.193***	0.313**	0.530***	0.786***	0.819***	0.740***	0.256***	0.256***	0.281***
	(0.029)	(0.096)	(0.133)	(0.030)	(0.089)	(0.126)	(0.011)	(0.031)	(0.048)
T2	0.186***	0.140**	0.152**	-0.009	-0.040	-0.056	0.031*	0.012	0.012
	(0.045)	(0.048)	(0.049)	(0.042)	(0.043)	(0.044)	(0.016)	(0.015)	(0.015)
Sarpanch woman		-0.215	-0.144		-0.187	-0.132		-0.098+	-0.066
		(0.157)	(0.190)		(0.158)	(0.166)		(0.055)	(0.059)
Sarpanch relative present		-0.136	-0.147		0.190**	0.147+		0.001	-0.016
		(0.254)	(0.259)		(0.070)	(0.086)		(0.166)	(0.161)
Sarpanch SC		-0.175	-0.182		-0.006	0.041		0.026	0.040
		(0.131)	(0.131)		(0.101)	(0.104)		(0.025)	(0.026)
Sarpanch OBC		-0.051	-0.059		-0.021	0.057		0.008	0.018
		(0.066)	(0.071)		(0.069)	(0.073)		(0.020)	(0.020)
Education above median		0.123*	0.153**		0.144**	0.163***		0.071***	0.068***
		(0.051)	(0.052)		(0.045)	(0.047)		(0.015)	(0.016)
Wealth above median		-0.010	-0.012		-0.000	-0.022		0.015	0.012
		(0.046)	(0.047)		(0.046)	(0.046)		(0.014)	(0.014)
Lives with in-laws		-0.003	-0.027		0.130	0.075		-0.013	-0.025
		(0.142)	(0.168)		(0.132)	(0.138)		(0.051)	(0.056)
Lives with own parents		-0.002	-0.031		0.027	0.010		0.048**	0.042*
		(0.056)	(0.058)		(0.054)	(0.051)		(0.017)	(0.017)
Years as sarpanch		0.001	-0.002		-0.026	-0.020		0.005	0.002
		(0.016)	(0.017)		(0.016)	(0.017)		(0.005)	(0.005)
Gender quota		0.004	-0.056		-0.034	-0.080		0.003	-0.020
		(0.078)	(0.091)		(0.092)	(0.096)		(0.022)	(0.023)
SC quota		0.171	0.152		0.029	0.015		-0.032	-0.047+
		(0.132)	(0.132)		(0.098)	(0.101)		(0.026)	(0.028)
OBC quota		-0.000	-0.028		0.097	0.064		0.011	0.002
		(0.065)	(0.068)		(0.067)	(0.069)		(0.020)	(0.021)
Sarpanch woman:Sarpanch relative present		0.110	0.126		-0.329*	-0.305*		-0.073	-0.064
		(0.266)	(0.271)		(0.136)	(0.148)		(0.167)	(0.163)
Percent of village land held by biggest landholding caste			-0.003*			0.001			-0.000
			(0.001)			(0.001)			(0.000)
Distance (km) from block center			-0.001			0.002			-0.000
			(0.002)			(0.002)			(0.001)
Directly elected			0.008			0.030			0.031+
			(0.052)			(0.048)			(0.016)
Num.Obs.	390	351	325	390	351	325	390	351	325
R2	0.042	0.135	0.159	0.000	0.093	0.129	0.010	0.366	0.376

## 
##  Asymptotic two-sample Kolmogorov-Smirnov test
## 
## data:  df$proportion_time_spoken[df$treatment_1_2_together == 1] and df$proportion_time_spoken[df$treatment_1_2_together == 0]
## D = 0.11625, p-value = 0.05803
## alternative hypothesis: two-sided

IC3 (H3.3): Assignment to any treatment (T1 or T2, considered jointly) will improve institutional centrality at a higher rate if the sarpanch is from a disadvantaged group.

IC3 is assessed by regressing our primary measures of centrality (as specified immediately above in IC2) on a binary indicator of treatment (bundling T1 and T2 versus control), an indicator of the relevant dimension of disadvantage (including one regression using the index of disadvantage) and the multiplicative interaction between the two. For each of our 3 dependent variables, we run a regression interacting treatment status with indicators of the sarpanch’s gender and caste, as defined above. We also consider the alternate measures of disadvantage defined above. Our focus is on the interaction effects for each of these specifications. Cluster-corrected standard errors will be used at the administrative block level. We check the robustness of results with bivariate regressions on treatment status and a multivariate regression including a standard set of demographic controls. We also control for the potential presence of a sarpanch relative during the interview (“the interview” being here understood as the group interview), both separately and in interaction with the gender of the sarpanch.
In additional analyses, we disentangle the impact of T1 and T2. We also run bivariate regressions, clustering at the taluka level, and we also include the series of individual and village level demographic controls presented in G2.
Deviation 1: Living with in-laws and one’s own parents were planned to be included in the wealth index. We instead include them as separate controls, because they are not indicative of wealth.
Deviation 2: We originally planned to use clustered standard errors at the taluka level. This was an error on our part: because the treatment assignment occurs at the level of each village council, we do not cluster standard errors in these analyses.
Precision 1: Village-level controls were not presented in cell G2 of the PAP spreadsheet; instead, they were presented in cell H4. This was a typo on our part.
Precision 2: We construct the index of disadvantage via all mentioned variables indicating disadvantage. We create an additive variable of disadvantage.

IC3 (H3.3) table, grouping T1/T2 together, Gender disadvantage
	DV: Sarpanch made final budget decision			DV: Sarpanch responded first to roles/resp.			DV: Proportion of time sarpanch spoke
	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls
+ p < 0.1, * p < 0.05, p < 0.01, * p < 0.001
Each model is a linear regression of each outcome on an indicator for being assigned to discussion rules (1) versus not (0). Unadjusted and covariate-adjusted models presented. All regressions use robust standard errors.
(Intercept)	0.333***	0.323***	0.435***	0.841***	0.762***	0.705***	0.354***	0.283***	0.304***
	(0.057)	(0.090)	(0.119)	(0.044)	(0.076)	(0.097)	(0.019)	(0.029)	(0.040)
Discussion rules	0.132+	0.123+	0.144+	0.057	0.053	0.012	0.022	0.010	0.009
	(0.068)	(0.072)	(0.073)	(0.050)	(0.051)	(0.044)	(0.022)	(0.023)	(0.023)
Sarpanch woman	-0.220***	-0.108	-0.058	-0.084	-0.032	-0.000	-0.153***	-0.105*	-0.079+
	(0.064)	(0.133)	(0.147)	(0.059)	(0.137)	(0.132)	(0.022)	(0.042)	(0.043)
Discussion rules:Sarpanch woman	-0.052	-0.046	-0.069	-0.076	-0.089	-0.042	-0.003	0.008	0.010
	(0.078)	(0.084)	(0.087)	(0.070)	(0.072)	(0.069)	(0.026)	(0.026)	(0.028)
Sarpanch relative present		0.002	-0.021		0.152***	0.130*		0.025	0.012
		(0.224)	(0.227)		(0.037)	(0.050)		(0.127)	(0.127)
Sarpanch SC		-0.170+	-0.181+		-0.041	0.009		0.016	0.025
		(0.095)	(0.099)		(0.074)	(0.076)		(0.021)	(0.022)
Sarpanch OBC		-0.054	-0.047		0.017	0.074		0.008	0.019
		(0.051)	(0.054)		(0.049)	(0.050)		(0.016)	(0.017)
Education above median		0.106**	0.121**		0.122***	0.129***		0.081***	0.079***
		(0.040)	(0.041)		(0.034)	(0.035)		(0.012)	(0.013)
Wealth above median		-0.025	-0.027		0.020	0.005		0.011	0.010
		(0.038)	(0.040)		(0.036)	(0.037)		(0.012)	(0.012)
Lives with in-laws		-0.109	-0.119		0.088	0.032		-0.006	-0.013
		(0.107)	(0.119)		(0.117)	(0.115)		(0.035)	(0.036)
Lives with own parents		-0.017	-0.041		0.013	0.003		0.026+	0.023
		(0.042)	(0.046)		(0.041)	(0.040)		(0.014)	(0.014)
Years as sarpanch		0.006	0.009		-0.015	-0.009		0.005	0.004
		(0.013)	(0.013)		(0.012)	(0.012)		(0.004)	(0.004)
Gender quota		0.006	-0.015		-0.111+	-0.156*		-0.018	-0.037+
		(0.057)	(0.064)		(0.063)	(0.065)		(0.018)	(0.019)
SC quota		0.118	0.125		0.084	0.053		-0.041*	-0.049*
		(0.094)	(0.097)		(0.071)	(0.072)		(0.020)	(0.022)
OBC quota		0.031	0.011		0.045	0.020		0.000	-0.008
		(0.050)	(0.053)		(0.048)	(0.047)		(0.016)	(0.016)
Sarpanch woman:Sarpanch relative present		-0.055	-0.037		-0.316**	-0.314**		-0.103	-0.103
		(0.231)	(0.235)		(0.098)	(0.107)		(0.129)	(0.129)
Percent of village land held by biggest landholding caste			-0.002*			0.000			-0.000
			(0.001)			(0.001)			(0.000)
Distance (km) from block center			0.001			0.002*			-0.000
			(0.002)			(0.001)			(0.001)
Directly elected			0.022			0.008			0.026*
			(0.042)			(0.036)			(0.013)
Num.Obs.	605	549	506	605	549	506	604	548	505
R2	0.096	0.119	0.129	0.031	0.089	0.119	0.242	0.366	0.376

IC3 (H3.3) table, comparing T1 to control, Gender disadvantage
	DV: Sarpanch made final budget decision			DV: Sarpanch responded first to roles/resp.			DV: Proportion of time sarpanch spoke
	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls
+ p < 0.1, * p < 0.05, p < 0.01, * p < 0.001
Each model is a linear regression of each outcome on an indicator for being assigned to T1 (1) versus control (0). Unadjusted and covariate-adjusted models presented. All regressions use robust standard errors.
(Intercept)	0.333***	0.341***	0.411**	0.841***	0.760***	0.806***	0.354***	0.302***	0.337***
	(0.057)	(0.095)	(0.134)	(0.044)	(0.085)	(0.108)	(0.019)	(0.032)	(0.046)
T1	0.073	0.077	0.088	0.093+	0.084	0.035	0.034	0.012	0.009
	(0.077)	(0.080)	(0.082)	(0.052)	(0.054)	(0.048)	(0.025)	(0.025)	(0.025)
Sarpanch woman	-0.220***	0.022	0.021	-0.084	0.085	0.070	-0.153***	-0.117**	-0.103*
	(0.064)	(0.143)	(0.154)	(0.059)	(0.140)	(0.129)	(0.022)	(0.042)	(0.040)
T1:Sarpanch woman	-0.042	-0.056	-0.071	-0.083	-0.085	-0.031	-0.013	0.002	0.009
	(0.088)	(0.095)	(0.099)	(0.075)	(0.079)	(0.076)	(0.030)	(0.030)	(0.031)
Sarpanch relative present		-0.071	-0.127		0.104	0.023		-0.120	-0.128
		(0.266)	(0.278)		(0.072)	(0.092)		(0.102)	(0.106)
Sarpanch SC		-0.149	-0.155		-0.013	0.026		-0.008	-0.005
		(0.098)	(0.103)		(0.090)	(0.093)		(0.027)	(0.030)
Sarpanch OBC		-0.066	-0.057		0.044	0.084		-0.000	0.009
		(0.057)	(0.060)		(0.057)	(0.058)		(0.020)	(0.021)
Education above median		0.121**	0.121*		0.131**	0.127**		0.101***	0.097***
		(0.046)	(0.047)		(0.042)	(0.042)		(0.015)	(0.015)
Wealth above median		-0.036	-0.026		0.008	-0.003		0.005	0.004
		(0.043)	(0.046)		(0.043)	(0.045)		(0.014)	(0.014)
Lives with in-laws		-0.217+	-0.208+		-0.070	-0.078		0.005	0.002
		(0.114)	(0.123)		(0.127)	(0.116)		(0.035)	(0.032)
Lives with own parents		-0.069	-0.086		-0.040	-0.051		0.003	-0.001
		(0.048)	(0.053)		(0.050)	(0.049)		(0.017)	(0.018)
Years as sarpanch		0.017	0.022		-0.007	-0.004		0.006	0.005
		(0.015)	(0.016)		(0.014)	(0.014)		(0.005)	(0.005)
Gender quota		-0.059	-0.040		-0.103	-0.155*		-0.030	-0.043*
		(0.066)	(0.073)		(0.069)	(0.066)		(0.021)	(0.021)
SC quota		0.056	0.057		0.128	0.102		-0.027	-0.029
		(0.095)	(0.099)		(0.085)	(0.088)		(0.026)	(0.028)
OBC quota		0.074	0.059		0.033	0.029		0.001	-0.009
		(0.056)	(0.059)		(0.056)	(0.054)		(0.018)	(0.019)
Sarpanch woman:Sarpanch relative present		0.028	0.067		-0.264*	-0.207		0.040	0.032
		(0.274)	(0.286)		(0.129)	(0.142)		(0.105)	(0.109)
Percent of village land held by biggest landholding caste			-0.002*			-0.001			-0.000
			(0.001)			(0.001)			(0.000)
Distance (km) from block center			0.003			0.004*			-0.000
			(0.002)			(0.002)			(0.001)
Directly elected			0.011			0.025			0.023
			(0.049)			(0.042)			(0.016)
Num.Obs.	407	370	340	407	370	340	406	369	339
R2	0.086	0.142	0.143	0.033	0.100	0.148	0.251	0.396	0.398

IC3 (H3.3) table, comparing T2 to control, Gender disadvantage
	DV: Sarpanch made final budget decision			DV: Sarpanch responded first to roles/resp.			DV: Proportion of time sarpanch spoke
	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls
+ p < 0.1, * p < 0.05, p < 0.01, * p < 0.001
Each model is a linear regression of each outcome on an indicator for being assigned to T2 (1) versus control (0). Unadjusted and covariate-adjusted models presented. All regressions use robust standard errors.
(Intercept)	0.333***	0.303**	0.504***	0.841***	0.782***	0.715***	0.354***	0.262***	0.288***
	(0.057)	(0.103)	(0.141)	(0.044)	(0.094)	(0.129)	(0.019)	(0.034)	(0.051)
t2	0.188*	0.157+	0.189*	0.021	0.024	-0.019	0.010	0.003	0.002
	(0.077)	(0.083)	(0.084)	(0.057)	(0.058)	(0.052)	(0.024)	(0.025)	(0.026)
Sarpanch woman	-0.220***	-0.201	-0.115	-0.084	-0.131	-0.103	-0.153***	-0.106+	-0.075
	(0.064)	(0.165)	(0.193)	(0.059)	(0.158)	(0.164)	(0.022)	(0.056)	(0.060)
t2:Sarpanch woman	-0.052	-0.029	-0.062	-0.075	-0.110	-0.062	0.007	0.016	0.018
	(0.093)	(0.100)	(0.102)	(0.082)	(0.085)	(0.083)	(0.030)	(0.030)	(0.031)
Sarpanch relative present		-0.135	-0.143		0.193**	0.150+		0.000	-0.016
		(0.253)	(0.255)		(0.060)	(0.081)		(0.168)	(0.164)
Sarpanch SC		-0.176	-0.184		-0.011	0.039		0.027	0.040
		(0.132)	(0.132)		(0.102)	(0.104)		(0.025)	(0.026)
Sarpanch OBC		-0.051	-0.058		-0.020	0.057		0.008	0.018
		(0.066)	(0.071)		(0.070)	(0.073)		(0.020)	(0.021)
Education above median		0.123*	0.153**		0.143**	0.162***		0.071***	0.069***
		(0.051)	(0.052)		(0.045)	(0.047)		(0.015)	(0.016)
Wealth above median		-0.010	-0.014		-0.003	-0.024		0.015	0.013
		(0.047)	(0.048)		(0.046)	(0.046)		(0.014)	(0.014)
Lives with in-laws		-0.003	-0.025		0.131	0.077		-0.013	-0.025
		(0.141)	(0.166)		(0.128)	(0.136)		(0.051)	(0.056)
Lives with own parents		-0.002	-0.031		0.027	0.010		0.048**	0.043*
		(0.056)	(0.058)		(0.054)	(0.051)		(0.017)	(0.017)
Years as sarpanch		0.001	-0.002		-0.026	-0.020		0.005	0.002
		(0.016)	(0.017)		(0.016)	(0.017)		(0.005)	(0.005)
Gender quota		0.005	-0.053		-0.032	-0.077		0.002	-0.020
		(0.078)	(0.090)		(0.092)	(0.096)		(0.022)	(0.023)
SC quota		0.172	0.155		0.036	0.019		-0.033	-0.048+
		(0.133)	(0.134)		(0.098)	(0.101)		(0.026)	(0.028)
OBC quota		0.000	-0.028		0.100	0.065		0.010	0.001
		(0.065)	(0.068)		(0.067)	(0.070)		(0.020)	(0.021)
Sarpanch woman:Sarpanch relative present		0.108	0.122		-0.335*	-0.310*		-0.072	-0.062
		(0.265)	(0.267)		(0.131)	(0.145)		(0.170)	(0.166)
Percent of village land held by biggest landholding caste			-0.003*			0.001			-0.000
			(0.001)			(0.001)			(0.000)
Distance (km) from block center			-0.001			0.002			-0.000
			(0.002)			(0.002)			(0.001)
Directly elected			0.007			0.029			0.031+
			(0.052)			(0.048)			(0.016)
Num.Obs.	390	351	325	390	351	325	390	351	325
R2	0.114	0.136	0.160	0.024	0.097	0.130	0.231	0.366	0.377

IC3 (H3.3) table, grouping T1/T2 together, Caste disadvantage
	DV: Sarpanch made final budget decision			DV: Sarpanch responded first to roles/resp.			DV: Proportion of time sarpanch spoke
	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls
+ p < 0.1, * p < 0.05, p < 0.01, * p < 0.001
Each model is a linear regression of each outcome on an indicator for being assigned to discussion rules (1) versus not (0). Unadjusted and covariate-adjusted models presented. All regressions use robust standard errors.
(Intercept)	0.203***	0.336***	0.468***	0.752***	0.779***	0.696***	0.261***	0.275***	0.295***
	(0.035)	(0.078)	(0.108)	(0.038)	(0.071)	(0.096)	(0.014)	(0.025)	(0.036)
Discussion rules	0.133**	0.108*	0.107*	0.062	0.035	0.025	0.041*	0.021	0.020
	(0.045)	(0.045)	(0.048)	(0.044)	(0.044)	(0.044)	(0.018)	(0.015)	(0.016)
Sarpanch SC	-0.034	-0.133	-0.167	0.113+	0.052	0.104	-0.017	0.032	0.035
	(0.060)	(0.104)	(0.107)	(0.059)	(0.090)	(0.089)	(0.022)	(0.027)	(0.028)
Discussion rules:Sarpanch SC	-0.025	-0.050	-0.019	-0.131+	-0.129+	-0.135+	-0.017	-0.023	-0.015
	(0.077)	(0.082)	(0.084)	(0.072)	(0.077)	(0.075)	(0.027)	(0.025)	(0.026)
Sarpanch woman		-0.141	-0.107		-0.096	-0.032		-0.100**	-0.072+
		(0.118)	(0.134)		(0.130)	(0.127)		(0.039)	(0.040)
Sarpanch relative present		-0.004	-0.029		0.137**	0.120+		0.024	0.012
		(0.227)	(0.232)		(0.051)	(0.067)		(0.127)	(0.126)
Sarpanch OBC		-0.055	-0.047		0.017	0.074		0.008	0.019
		(0.051)	(0.054)		(0.049)	(0.049)		(0.016)	(0.017)
Education above median		0.105**	0.121**		0.121***	0.128***		0.081***	0.079***
		(0.040)	(0.041)		(0.034)	(0.035)		(0.012)	(0.013)
Wealth above median		-0.023	-0.026		0.022	0.003		0.011	0.010
		(0.038)	(0.039)		(0.036)	(0.037)		(0.012)	(0.012)
Lives with in-laws		-0.109	-0.120		0.088	0.031		-0.006	-0.013
		(0.107)	(0.119)		(0.117)	(0.115)		(0.035)	(0.036)
Lives with own parents		-0.018	-0.041		0.012	0.002		0.026+	0.023
		(0.042)	(0.046)		(0.041)	(0.039)		(0.014)	(0.014)
Years as sarpanch		0.006	0.009		-0.015	-0.009		0.005	0.004
		(0.013)	(0.013)		(0.012)	(0.012)		(0.004)	(0.004)
Gender quota		0.007	-0.015		-0.109+	-0.153*		-0.018	-0.036+
		(0.056)	(0.064)		(0.063)	(0.065)		(0.018)	(0.019)
SC quota		0.115	0.122		0.077	0.049		-0.041*	-0.048*
		(0.093)	(0.096)		(0.071)	(0.071)		(0.020)	(0.022)
OBC quota		0.028	0.009		0.039	0.016		0.000	-0.008
		(0.050)	(0.053)		(0.048)	(0.047)		(0.016)	(0.016)
Sarpanch woman:Sarpanch relative present		-0.046	-0.028		-0.295**	-0.295*		-0.101	-0.102
		(0.234)	(0.240)		(0.104)	(0.116)		(0.129)	(0.128)
Percent of village land held by biggest landholding caste			-0.002*			0.001			-0.000
			(0.001)			(0.001)			(0.000)
Distance (km) from block center			0.000			0.003*			-0.000
			(0.002)			(0.001)			(0.001)
Directly elected			0.022			0.006			0.026*
			(0.042)			(0.036)			(0.013)
Num.Obs.	603	549	506	603	549	506	602	548	505
R2	0.020	0.119	0.128	0.006	0.091	0.124	0.018	0.366	0.376

IC3 (H3.3) table, comparing T1 to control, Caste disadvantage
	DV: Sarpanch made final budget decision			DV: Sarpanch responded first to roles/resp.			DV: Proportion of time sarpanch spoke
	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls
+ p < 0.1, * p < 0.05, p < 0.01, * p < 0.001
Each model is a linear regression of each outcome on an indicator for being assigned to T1 (1) versus control (0). Unadjusted and covariate-adjusted models presented. All regressions use robust standard errors.
(Intercept)	0.203***	0.340***	0.420***	0.752***	0.783***	0.809***	0.261***	0.295***	0.325***
	(0.035)	(0.084)	(0.123)	(0.038)	(0.083)	(0.108)	(0.014)	(0.029)	(0.043)
T1	0.108*	0.086+	0.081	0.072	0.055	0.035	0.049*	0.026	0.026
	(0.052)	(0.051)	(0.055)	(0.049)	(0.048)	(0.048)	(0.020)	(0.017)	(0.018)
Sarpanch SC	-0.034	-0.056	-0.080	0.113+	0.037	0.065	-0.017	0.020	0.019
	(0.060)	(0.107)	(0.111)	(0.059)	(0.100)	(0.103)	(0.022)	(0.030)	(0.033)
T1:Sarpanch SC	-0.141+	-0.154+	-0.127	-0.073	-0.080	-0.067	-0.036	-0.048+	-0.041
	(0.083)	(0.091)	(0.095)	(0.079)	(0.085)	(0.082)	(0.031)	(0.027)	(0.028)
Sarpanch woman		-0.003	-0.016		0.040	0.054		-0.113**	-0.096*
		(0.125)	(0.136)		(0.134)	(0.124)		(0.040)	(0.039)
Sarpanch relative present		-0.097	-0.151		0.082	0.012		-0.126	-0.130
		(0.273)	(0.286)		(0.075)	(0.101)		(0.104)	(0.107)
Sarpanch OBC		-0.067	-0.057		0.043	0.084		-0.000	0.009
		(0.057)	(0.060)		(0.057)	(0.058)		(0.020)	(0.020)
Education above median		0.119**	0.121*		0.129**	0.126**		0.101***	0.097***
		(0.046)	(0.047)		(0.041)	(0.042)		(0.015)	(0.015)
Wealth above median		-0.034	-0.026		0.011	-0.003		0.005	0.003
		(0.043)	(0.045)		(0.043)	(0.044)		(0.014)	(0.014)
Lives with in-laws		-0.218*	-0.209+		-0.071	-0.078		0.005	0.002
		(0.111)	(0.119)		(0.125)	(0.115)		(0.037)	(0.034)
Lives with own parents		-0.072	-0.090+		-0.043	-0.053		0.002	-0.002
		(0.048)	(0.053)		(0.050)	(0.049)		(0.017)	(0.018)
Years as sarpanch		0.017	0.023		-0.007	-0.004		0.006	0.006
		(0.015)	(0.016)		(0.014)	(0.014)		(0.005)	(0.005)
Gender quota		-0.063	-0.041		-0.103	-0.156*		-0.032	-0.044*
		(0.066)	(0.072)		(0.069)	(0.066)		(0.021)	(0.021)
SC quota		0.040	0.043		0.115	0.095		-0.030	-0.032
		(0.094)	(0.098)		(0.084)	(0.087)		(0.026)	(0.028)
OBC quota		0.068	0.052		0.028	0.026		-0.000	-0.010
		(0.057)	(0.059)		(0.056)	(0.054)		(0.018)	(0.019)
Sarpanch woman:Sarpanch relative present		0.057	0.094		-0.242+	-0.194		0.047	0.036
		(0.281)	(0.295)		(0.132)	(0.150)		(0.107)	(0.109)
Percent of village land held by biggest landholding caste			-0.002*			-0.001			-0.000
			(0.001)			(0.001)			(0.000)
Distance (km) from block center			0.002			0.004*			-0.000
			(0.002)			(0.002)			(0.001)
Directly elected			0.009			0.024			0.023
			(0.049)			(0.042)			(0.016)
Num.Obs.	406	370	340	406	370	340	405	369	339
R2	0.026	0.148	0.145	0.014	0.099	0.149	0.027	0.400	0.401

IC3 (H3.3) table, comparing T2 to control, Caste disadvantage
	DV: Sarpanch made final budget decision			DV: Sarpanch responded first to roles/resp.			DV: Proportion of time sarpanch spoke
	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls
+ p < 0.1, * p < 0.05, p < 0.01, * p < 0.001
Each model is a linear regression of each outcome on an indicator for being assigned to T2 (1) versus control (0). Unadjusted and covariate-adjusted models presented. All regressions use robust standard errors.
(Intercept)	0.203***	0.316**	0.544***	0.752***	0.791***	0.701***	0.261***	0.256***	0.282***
	(0.035)	(0.097)	(0.135)	(0.038)	(0.090)	(0.127)	(0.014)	(0.032)	(0.049)
t2	0.162**	0.134*	0.132*	0.050	0.011	-0.001	0.032	0.013	0.010
	(0.055)	(0.057)	(0.060)	(0.052)	(0.052)	(0.053)	(0.020)	(0.018)	(0.019)
Sarpanch SC	-0.034	-0.185	-0.213	0.113+	0.076	0.127	-0.017	0.027	0.037
	(0.060)	(0.130)	(0.130)	(0.059)	(0.109)	(0.108)	(0.022)	(0.029)	(0.031)
t2:Sarpanch SC	0.077	0.022	0.065	-0.182*	-0.173+	-0.182*	0.002	-0.003	0.006
	(0.095)	(0.101)	(0.104)	(0.087)	(0.092)	(0.092)	(0.031)	(0.029)	(0.030)
Sarpanch woman		-0.213	-0.139		-0.201	-0.146		-0.098+	-0.066
		(0.157)	(0.193)		(0.158)	(0.164)		(0.055)	(0.059)
Sarpanch relative present		-0.135	-0.145		0.185*	0.144		0.001	-0.015
		(0.256)	(0.262)		(0.088)	(0.109)		(0.166)	(0.161)
Sarpanch OBC		-0.051	-0.060		-0.019	0.059		0.008	0.018
		(0.066)	(0.071)		(0.069)	(0.072)		(0.020)	(0.021)
Education above median		0.123*	0.154**		0.144**	0.163***		0.071***	0.069***
		(0.051)	(0.052)		(0.045)	(0.047)		(0.015)	(0.016)
Wealth above median		-0.009	-0.011		-0.002	-0.028		0.015	0.012
		(0.046)	(0.048)		(0.046)	(0.046)		(0.014)	(0.014)
Lives with in-laws		-0.003	-0.027		0.133	0.076		-0.013	-0.025
		(0.143)	(0.171)		(0.131)	(0.136)		(0.052)	(0.056)
Lives with own parents		-0.002	-0.031		0.028	0.011		0.048**	0.042*
		(0.056)	(0.059)		(0.054)	(0.051)		(0.017)	(0.017)
Years as sarpanch		0.001	-0.002		-0.024	-0.019		0.005	0.002
		(0.016)	(0.017)		(0.016)	(0.017)		(0.005)	(0.005)
Gender quota		0.003	-0.061		-0.022	-0.065		0.003	-0.020
		(0.078)	(0.091)		(0.090)	(0.094)		(0.022)	(0.023)
SC quota		0.169	0.148		0.040	0.028		-0.032	-0.048+
		(0.133)	(0.133)		(0.099)	(0.101)		(0.026)	(0.028)
OBC quota		0.000	-0.026		0.091	0.057		0.010	0.002
		(0.065)	(0.069)		(0.066)	(0.069)		(0.020)	(0.021)
Sarpanch woman:Sarpanch relative present		0.107	0.118		-0.310*	-0.283+		-0.073	-0.064
		(0.267)	(0.274)		(0.146)	(0.163)		(0.167)	(0.162)
Percent of village land held by biggest landholding caste			-0.003*			0.001			-0.000
			(0.001)			(0.001)			(0.000)
Distance (km) from block center			-0.001			0.002			-0.000
			(0.002)			(0.002)			(0.001)
Directly elected			0.010			0.024			0.031+
			(0.052)			(0.047)			(0.016)
Num.Obs.	389	351	325	389	351	325	389	351	325
R2	0.045	0.135	0.160	0.011	0.102	0.139	0.013	0.366	0.376

IC3 (H3.3) table, grouping T1/T2 together, Gender disadvantage
	DV: Sarpanch made final budget decision			DV: Sarpanch responded first to roles/resp.			DV: Proportion of time sarpanch spoke
	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls
+ p < 0.1, * p < 0.05, p < 0.01, * p < 0.001
Each model is a linear regression of each outcome on an indicator for being assigned to discussion rules (1) versus not (0). Unadjusted and covariate-adjusted models presented. All regressions use robust standard errors.
(Intercept)	0.333***	0.323***	0.435***	0.841***	0.762***	0.705***	0.354***	0.283***	0.304***
	(0.057)	(0.090)	(0.119)	(0.044)	(0.076)	(0.097)	(0.019)	(0.029)	(0.040)
Discussion rules	0.132+	0.123+	0.144+	0.057	0.053	0.012	0.022	0.010	0.009
	(0.068)	(0.072)	(0.073)	(0.050)	(0.051)	(0.044)	(0.022)	(0.023)	(0.023)
Sarpanch woman	-0.220***	-0.108	-0.058	-0.084	-0.032	-0.000	-0.153***	-0.105*	-0.079+
	(0.064)	(0.133)	(0.147)	(0.059)	(0.137)	(0.132)	(0.022)	(0.042)	(0.043)
Discussion rules:Sarpanch woman	-0.052	-0.046	-0.069	-0.076	-0.089	-0.042	-0.003	0.008	0.010
	(0.078)	(0.084)	(0.087)	(0.070)	(0.072)	(0.069)	(0.026)	(0.026)	(0.028)
Sarpanch relative present		0.002	-0.021		0.152***	0.130*		0.025	0.012
		(0.224)	(0.227)		(0.037)	(0.050)		(0.127)	(0.127)
Sarpanch SC		-0.170+	-0.181+		-0.041	0.009		0.016	0.025
		(0.095)	(0.099)		(0.074)	(0.076)		(0.021)	(0.022)
Sarpanch OBC		-0.054	-0.047		0.017	0.074		0.008	0.019
		(0.051)	(0.054)		(0.049)	(0.050)		(0.016)	(0.017)
Education above median		0.106**	0.121**		0.122***	0.129***		0.081***	0.079***
		(0.040)	(0.041)		(0.034)	(0.035)		(0.012)	(0.013)
Wealth above median		-0.025	-0.027		0.020	0.005		0.011	0.010
		(0.038)	(0.040)		(0.036)	(0.037)		(0.012)	(0.012)
Lives with in-laws		-0.109	-0.119		0.088	0.032		-0.006	-0.013
		(0.107)	(0.119)		(0.117)	(0.115)		(0.035)	(0.036)
Lives with own parents		-0.017	-0.041		0.013	0.003		0.026+	0.023
		(0.042)	(0.046)		(0.041)	(0.040)		(0.014)	(0.014)
Years as sarpanch		0.006	0.009		-0.015	-0.009		0.005	0.004
		(0.013)	(0.013)		(0.012)	(0.012)		(0.004)	(0.004)
Gender quota		0.006	-0.015		-0.111+	-0.156*		-0.018	-0.037+
		(0.057)	(0.064)		(0.063)	(0.065)		(0.018)	(0.019)
SC quota		0.118	0.125		0.084	0.053		-0.041*	-0.049*
		(0.094)	(0.097)		(0.071)	(0.072)		(0.020)	(0.022)
OBC quota		0.031	0.011		0.045	0.020		0.000	-0.008
		(0.050)	(0.053)		(0.048)	(0.047)		(0.016)	(0.016)
Sarpanch woman:Sarpanch relative present		-0.055	-0.037		-0.316**	-0.314**		-0.103	-0.103
		(0.231)	(0.235)		(0.098)	(0.107)		(0.129)	(0.129)
Percent of village land held by biggest landholding caste			-0.002*			0.000			-0.000
			(0.001)			(0.001)			(0.000)
Distance (km) from block center			0.001			0.002*			-0.000
			(0.002)			(0.001)			(0.001)
Directly elected			0.022			0.008			0.026*
			(0.042)			(0.036)			(0.013)
Num.Obs.	605	549	506	605	549	506	604	548	505
R2	0.096	0.119	0.129	0.031	0.089	0.119	0.242	0.366	0.376

IC3 (H3.3) table, comparing T1 to control, Gender disadvantage
	DV: Sarpanch made final budget decision			DV: Sarpanch responded first to roles/resp.			DV: Proportion of time sarpanch spoke
	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls
+ p < 0.1, * p < 0.05, p < 0.01, * p < 0.001
Each model is a linear regression of each outcome on an indicator for being assigned to T1 (1) versus control (0). Unadjusted and covariate-adjusted models presented. All regressions use robust standard errors.
(Intercept)	0.333***	0.341***	0.411**	0.841***	0.760***	0.806***	0.354***	0.302***	0.337***
	(0.057)	(0.095)	(0.134)	(0.044)	(0.085)	(0.108)	(0.019)	(0.032)	(0.046)
T1	0.073	0.077	0.088	0.093+	0.084	0.035	0.034	0.012	0.009
	(0.077)	(0.080)	(0.082)	(0.052)	(0.054)	(0.048)	(0.025)	(0.025)	(0.025)
Sarpanch woman	-0.220***	0.022	0.021	-0.084	0.085	0.070	-0.153***	-0.117**	-0.103*
	(0.064)	(0.143)	(0.154)	(0.059)	(0.140)	(0.129)	(0.022)	(0.042)	(0.040)
T1:Sarpanch woman	-0.042	-0.056	-0.071	-0.083	-0.085	-0.031	-0.013	0.002	0.009
	(0.088)	(0.095)	(0.099)	(0.075)	(0.079)	(0.076)	(0.030)	(0.030)	(0.031)
Sarpanch relative present		-0.071	-0.127		0.104	0.023		-0.120	-0.128
		(0.266)	(0.278)		(0.072)	(0.092)		(0.102)	(0.106)
Sarpanch SC		-0.149	-0.155		-0.013	0.026		-0.008	-0.005
		(0.098)	(0.103)		(0.090)	(0.093)		(0.027)	(0.030)
Sarpanch OBC		-0.066	-0.057		0.044	0.084		-0.000	0.009
		(0.057)	(0.060)		(0.057)	(0.058)		(0.020)	(0.021)
Education above median		0.121**	0.121*		0.131**	0.127**		0.101***	0.097***
		(0.046)	(0.047)		(0.042)	(0.042)		(0.015)	(0.015)
Wealth above median		-0.036	-0.026		0.008	-0.003		0.005	0.004
		(0.043)	(0.046)		(0.043)	(0.045)		(0.014)	(0.014)
Lives with in-laws		-0.217+	-0.208+		-0.070	-0.078		0.005	0.002
		(0.114)	(0.123)		(0.127)	(0.116)		(0.035)	(0.032)
Lives with own parents		-0.069	-0.086		-0.040	-0.051		0.003	-0.001
		(0.048)	(0.053)		(0.050)	(0.049)		(0.017)	(0.018)
Years as sarpanch		0.017	0.022		-0.007	-0.004		0.006	0.005
		(0.015)	(0.016)		(0.014)	(0.014)		(0.005)	(0.005)
Gender quota		-0.059	-0.040		-0.103	-0.155*		-0.030	-0.043*
		(0.066)	(0.073)		(0.069)	(0.066)		(0.021)	(0.021)
SC quota		0.056	0.057		0.128	0.102		-0.027	-0.029
		(0.095)	(0.099)		(0.085)	(0.088)		(0.026)	(0.028)
OBC quota		0.074	0.059		0.033	0.029		0.001	-0.009
		(0.056)	(0.059)		(0.056)	(0.054)		(0.018)	(0.019)
Sarpanch woman:Sarpanch relative present		0.028	0.067		-0.264*	-0.207		0.040	0.032
		(0.274)	(0.286)		(0.129)	(0.142)		(0.105)	(0.109)
Percent of village land held by biggest landholding caste			-0.002*			-0.001			-0.000
			(0.001)			(0.001)			(0.000)
Distance (km) from block center			0.003			0.004*			-0.000
			(0.002)			(0.002)			(0.001)
Directly elected			0.011			0.025			0.023
			(0.049)			(0.042)			(0.016)
Num.Obs.	407	370	340	407	370	340	406	369	339
R2	0.086	0.142	0.143	0.033	0.100	0.148	0.251	0.396	0.398

IC3 (H3.3) table, comparing T2 to control, Gender disadvantage
	DV: Sarpanch made final budget decision			DV: Sarpanch responded first to roles/resp.			DV: Proportion of time sarpanch spoke
	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls
+ p < 0.1, * p < 0.05, p < 0.01, * p < 0.001
Each model is a linear regression of each outcome on an indicator for being assigned to T2 (1) versus control (0). Unadjusted and covariate-adjusted models presented. All regressions use robust standard errors.
(Intercept)	0.333***	0.303**	0.504***	0.841***	0.782***	0.715***	0.354***	0.262***	0.288***
	(0.057)	(0.103)	(0.141)	(0.044)	(0.094)	(0.129)	(0.019)	(0.034)	(0.051)
t2	0.188*	0.157+	0.189*	0.021	0.024	-0.019	0.010	0.003	0.002
	(0.077)	(0.083)	(0.084)	(0.057)	(0.058)	(0.052)	(0.024)	(0.025)	(0.026)
Sarpanch woman	-0.220***	-0.201	-0.115	-0.084	-0.131	-0.103	-0.153***	-0.106+	-0.075
	(0.064)	(0.165)	(0.193)	(0.059)	(0.158)	(0.164)	(0.022)	(0.056)	(0.060)
t2:Sarpanch woman	-0.052	-0.029	-0.062	-0.075	-0.110	-0.062	0.007	0.016	0.018
	(0.093)	(0.100)	(0.102)	(0.082)	(0.085)	(0.083)	(0.030)	(0.030)	(0.031)
Sarpanch relative present		-0.135	-0.143		0.193**	0.150+		0.000	-0.016
		(0.253)	(0.255)		(0.060)	(0.081)		(0.168)	(0.164)
Sarpanch SC		-0.176	-0.184		-0.011	0.039		0.027	0.040
		(0.132)	(0.132)		(0.102)	(0.104)		(0.025)	(0.026)
Sarpanch OBC		-0.051	-0.058		-0.020	0.057		0.008	0.018
		(0.066)	(0.071)		(0.070)	(0.073)		(0.020)	(0.021)
Education above median		0.123*	0.153**		0.143**	0.162***		0.071***	0.069***
		(0.051)	(0.052)		(0.045)	(0.047)		(0.015)	(0.016)
Wealth above median		-0.010	-0.014		-0.003	-0.024		0.015	0.013
		(0.047)	(0.048)		(0.046)	(0.046)		(0.014)	(0.014)
Lives with in-laws		-0.003	-0.025		0.131	0.077		-0.013	-0.025
		(0.141)	(0.166)		(0.128)	(0.136)		(0.051)	(0.056)
Lives with own parents		-0.002	-0.031		0.027	0.010		0.048**	0.043*
		(0.056)	(0.058)		(0.054)	(0.051)		(0.017)	(0.017)
Years as sarpanch		0.001	-0.002		-0.026	-0.020		0.005	0.002
		(0.016)	(0.017)		(0.016)	(0.017)		(0.005)	(0.005)
Gender quota		0.005	-0.053		-0.032	-0.077		0.002	-0.020
		(0.078)	(0.090)		(0.092)	(0.096)		(0.022)	(0.023)
SC quota		0.172	0.155		0.036	0.019		-0.033	-0.048+
		(0.133)	(0.134)		(0.098)	(0.101)		(0.026)	(0.028)
OBC quota		0.000	-0.028		0.100	0.065		0.010	0.001
		(0.065)	(0.068)		(0.067)	(0.070)		(0.020)	(0.021)
Sarpanch woman:Sarpanch relative present		0.108	0.122		-0.335*	-0.310*		-0.072	-0.062
		(0.265)	(0.267)		(0.131)	(0.145)		(0.170)	(0.166)
Percent of village land held by biggest landholding caste			-0.003*			0.001			-0.000
			(0.001)			(0.001)			(0.000)
Distance (km) from block center			-0.001			0.002			-0.000
			(0.002)			(0.002)			(0.001)
Directly elected			0.007			0.029			0.031+
			(0.052)			(0.048)			(0.016)
Num.Obs.	390	351	325	390	351	325	390	351	325
R2	0.114	0.136	0.160	0.024	0.097	0.130	0.231	0.366	0.377

IC3 (H3.3) table, grouping T1/T2 together, Disadvantage index
	DV: Sarpanch made final budget decision			DV: Sarpanch responded first to roles/resp.			DV: Proportion of time sarpanch spoke
	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls
+ p < 0.1, * p < 0.05, p < 0.01, * p < 0.001
Each model is a linear regression of each outcome on an indicator for being assigned to discussion rules (1) versus not (0). Unadjusted and covariate-adjusted models presented. All regressions use robust standard errors.
(Intercept)	0.412***	0.570***	0.681***	0.898***	0.987***	0.942***	0.434***	0.370***	0.368***
	(0.092)	(0.154)	(0.178)	(0.078)	(0.129)	(0.153)	(0.030)	(0.048)	(0.053)
Discussion rules	0.176	0.158	0.146	0.123	0.118	0.076	0.017	0.006	0.006
	(0.108)	(0.106)	(0.111)	(0.088)	(0.087)	(0.082)	(0.035)	(0.033)	(0.034)
Disadvantage index	-0.081**	-0.097+	-0.095+	-0.042	-0.084+	-0.094+	-0.064***	-0.038*	-0.028+
	(0.029)	(0.051)	(0.054)	(0.028)	(0.050)	(0.052)	(0.010)	(0.015)	(0.016)
Discussion rules:Disadvantage index	-0.023	-0.027	-0.019	-0.039	-0.047	-0.035	0.003	0.002	0.003
	(0.035)	(0.035)	(0.036)	(0.033)	(0.032)	(0.031)	(0.011)	(0.010)	(0.011)
Sarpanch woman		-0.025	-0.016		0.043	0.099		-0.061	-0.046
		(0.124)	(0.140)		(0.130)	(0.131)		(0.041)	(0.043)
Sarpanch relative present		-0.008	-0.030		0.137**	0.117+		0.023	0.011
		(0.224)	(0.228)		(0.048)	(0.060)		(0.124)	(0.125)
Sarpanch SC		-0.086	-0.104		0.109	0.161+		0.053+	0.053+
		(0.102)	(0.108)		(0.088)	(0.090)		(0.027)	(0.029)
Sarpanch OBC		-0.059	-0.052		0.018	0.075		0.007	0.018
		(0.051)	(0.054)		(0.048)	(0.049)		(0.016)	(0.017)
Education above median		0.057	0.078		0.070+	0.075+		0.064***	0.067***
		(0.046)	(0.047)		(0.038)	(0.039)		(0.014)	(0.015)
Wealth above median		-0.155*	-0.148*		-0.100+	-0.119*		-0.028	-0.019
		(0.063)	(0.067)		(0.056)	(0.058)		(0.019)	(0.020)
Lives with in-laws		-0.107	-0.113		0.081	0.031		-0.004	-0.011
		(0.103)	(0.116)		(0.114)	(0.115)		(0.034)	(0.036)
Lives with own parents		-0.020	-0.042		-0.002	-0.011		0.023+	0.021
		(0.042)	(0.046)		(0.041)	(0.040)		(0.014)	(0.014)
Years as sarpanch		0.010	0.013		-0.009	-0.003		0.007+	0.006
		(0.013)	(0.014)		(0.012)	(0.013)		(0.004)	(0.004)
Gender quota		0.014	0.005		-0.110+	-0.149*		-0.019	-0.036+
		(0.055)	(0.062)		(0.063)	(0.065)		(0.018)	(0.019)
SC quota		0.160+	0.163+		0.071	0.040		-0.037+	-0.046*
		(0.090)	(0.094)		(0.072)	(0.072)		(0.021)	(0.023)
OBC quota		0.037	0.017		0.038	0.012		0.002	-0.007
		(0.049)	(0.052)		(0.047)	(0.046)		(0.016)	(0.017)
Sarpanch woman:Sarpanch relative present		-0.035	-0.019		-0.307**	-0.307**		-0.101	-0.103
		(0.231)	(0.237)		(0.101)	(0.111)		(0.126)	(0.126)
Percent of village land held by biggest landholding caste			-0.002*			0.000			-0.000
			(0.001)			(0.001)			(0.000)
Distance (km) from block center			0.001			0.003*			-0.000
			(0.002)			(0.001)			(0.001)
Directly elected			0.012			0.008			0.025+
			(0.041)			(0.036)			(0.013)
Num.Obs.	546	546	503	546	546	503	545	545	502
R2	0.081	0.139	0.146	0.046	0.102	0.132	0.213	0.373	0.379

IC3 (H3.3) table, comparing T1 to control, Disadvantage index
	DV: Sarpanch made final budget decision			DV: Sarpanch responded first to roles/resp.			DV: Proportion of time sarpanch spoke
	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls
+ p < 0.1, * p < 0.05, p < 0.01, * p < 0.001
Each model is a linear regression of each outcome on an indicator for being assigned to T1 (1) versus control (0). Unadjusted and covariate-adjusted models presented. All regressions use robust standard errors.
(Intercept)	0.412***	0.557**	0.709***	0.898***	1.074***	1.174***	0.434***	0.365***	0.387***
	(0.092)	(0.173)	(0.201)	(0.078)	(0.145)	(0.170)	(0.030)	(0.057)	(0.065)
T1	0.149	0.150	0.144	0.115	0.132	0.061	0.027	0.022	0.021
	(0.121)	(0.118)	(0.124)	(0.094)	(0.092)	(0.086)	(0.039)	(0.037)	(0.038)
Disadvantage index	-0.081**	-0.084	-0.118+	-0.042	-0.122*	-0.140*	-0.064***	-0.027	-0.022
	(0.029)	(0.056)	(0.060)	(0.028)	(0.056)	(0.055)	(0.010)	(0.019)	(0.019)
T1:Disadvantage index	-0.031	-0.043	-0.041	-0.022	-0.038	-0.017	0.000	-0.004	-0.003
	(0.038)	(0.038)	(0.040)	(0.036)	(0.034)	(0.033)	(0.012)	(0.012)	(0.012)
Sarpanch woman		0.080	0.091		0.185	0.203		-0.088*	-0.076+
		(0.131)	(0.140)		(0.134)	(0.127)		(0.044)	(0.042)
Sarpanch relative present		-0.071	-0.125		0.105	0.028		-0.117	-0.126
		(0.259)	(0.271)		(0.100)	(0.111)		(0.102)	(0.104)
Sarpanch SC		-0.090	-0.056		0.179+	0.223*		0.020	0.019
		(0.101)	(0.109)		(0.101)	(0.100)		(0.035)	(0.038)
Sarpanch OBC		-0.073	-0.062		0.047	0.088		-0.000	0.009
		(0.057)	(0.060)		(0.056)	(0.057)		(0.020)	(0.021)
Education above median		0.075	0.061		0.059	0.053		0.088***	0.087***
		(0.055)	(0.057)		(0.047)	(0.045)		(0.018)	(0.019)
Wealth above median		-0.159*	-0.183*		-0.139*	-0.159*		-0.028	-0.022
		(0.071)	(0.076)		(0.062)	(0.061)		(0.025)	(0.025)
Lives with in-laws		-0.209+	-0.200+		-0.075	-0.081		0.006	0.003
		(0.109)	(0.117)		(0.126)	(0.116)		(0.035)	(0.032)
Lives with own parents		-0.077	-0.097+		-0.069	-0.079		-0.001	-0.005
		(0.049)	(0.053)		(0.050)	(0.049)		(0.018)	(0.018)
Years as sarpanch		0.023	0.030+		0.003	0.006		0.008	0.007
		(0.015)	(0.016)		(0.014)	(0.014)		(0.005)	(0.005)
Gender quota		-0.048	-0.017		-0.091	-0.141*		-0.030	-0.042+
		(0.065)	(0.071)		(0.069)	(0.067)		(0.021)	(0.022)
SC quota		0.113	0.111		0.112	0.087		-0.019	-0.023
		(0.081)	(0.086)		(0.085)	(0.088)		(0.027)	(0.030)
OBC quota		0.085	0.069		0.026	0.020		0.002	-0.008
		(0.055)	(0.058)		(0.055)	(0.053)		(0.019)	(0.019)
Sarpanch woman:Sarpanch relative present		0.039	0.076		-0.279+	-0.227		0.037	0.029
		(0.268)	(0.281)		(0.147)	(0.156)		(0.104)	(0.106)
Percent of village land held by biggest landholding caste			-0.002*			-0.002			-0.000
			(0.001)			(0.001)			(0.000)
Distance (km) from block center			0.003			0.003*			-0.000
			(0.002)			(0.001)			(0.001)
Directly elected			-0.004			0.025			0.022
			(0.048)			(0.041)			(0.016)
Num.Obs.	367	367	337	367	367	337	366	366	336
R2	0.082	0.168	0.173	0.034	0.122	0.173	0.217	0.402	0.402

IC3 (H3.3) table, comparing T2 to control, Disadvantage index
	DV: Sarpanch made final budget decision			DV: Sarpanch responded first to roles/resp.			DV: Proportion of time sarpanch spoke
	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls
+ p < 0.1, * p < 0.05, p < 0.01, * p < 0.001
Each model is a linear regression of each outcome on an indicator for being assigned to T2 (1) versus control (0). Unadjusted and covariate-adjusted models presented. All regressions use robust standard errors.
(Intercept)	0.412***	0.480**	0.623**	0.898***	1.033***	0.943***	0.434***	0.385***	0.384***
	(0.092)	(0.180)	(0.214)	(0.078)	(0.160)	(0.200)	(0.030)	(0.054)	(0.065)
T2	0.203	0.184	0.164	0.133	0.110	0.072	0.005	-0.012	-0.012
	(0.124)	(0.127)	(0.134)	(0.101)	(0.102)	(0.099)	(0.038)	(0.038)	(0.039)
Disadvantage index	-0.081**	-0.071	-0.042	-0.042	-0.096	-0.097	-0.064***	-0.053**	-0.042*
	(0.029)	(0.059)	(0.063)	(0.028)	(0.063)	(0.069)	(0.010)	(0.018)	(0.019)
T2:Disadvantage index	-0.013	-0.017	-0.004	-0.059	-0.061	-0.052	0.006	0.007	0.008
	(0.040)	(0.042)	(0.044)	(0.039)	(0.038)	(0.038)	(0.012)	(0.012)	(0.013)
Sarpanch woman		-0.144	-0.128		-0.031	-0.004		-0.045	-0.029
		(0.166)	(0.197)		(0.162)	(0.175)		(0.057)	(0.063)
Sarpanch relative present		-0.136	-0.145		0.187*	0.145		-0.003	-0.019
		(0.248)	(0.254)		(0.075)	(0.101)		(0.161)	(0.159)
Sarpanch SC		-0.100	-0.143		0.123	0.169		0.079*	0.081*
		(0.145)	(0.147)		(0.120)	(0.126)		(0.032)	(0.035)
Sarpanch OBC		-0.056	-0.065		-0.020	0.055		0.005	0.014
		(0.066)	(0.071)		(0.068)	(0.072)		(0.020)	(0.021)
Education above median		0.087	0.133*		0.087+	0.106*		0.046**	0.050**
		(0.057)	(0.058)		(0.050)	(0.052)		(0.017)	(0.018)
Wealth above median		-0.094	-0.059		-0.137+	-0.158+		-0.038+	-0.028
		(0.078)	(0.083)		(0.076)	(0.083)		(0.022)	(0.024)
Lives with in-laws		-0.007	-0.023		0.122	0.082		-0.010	-0.019
		(0.140)	(0.167)		(0.118)	(0.128)		(0.048)	(0.054)
Lives with own parents		0.004	-0.023		0.026	0.010		0.048**	0.043*
		(0.055)	(0.058)		(0.054)	(0.051)		(0.017)	(0.017)
Years as sarpanch		0.001	-0.003		-0.023	-0.018		0.006	0.003
		(0.016)	(0.017)		(0.016)	(0.017)		(0.005)	(0.005)
Gender quota		0.032	-0.020		-0.031	-0.065		0.002	-0.019
		(0.073)	(0.086)		(0.093)	(0.098)		(0.022)	(0.024)
SC quota		0.172	0.149		0.043	0.024		-0.033	-0.049+
		(0.136)	(0.136)		(0.102)	(0.104)		(0.027)	(0.029)
OBC quota		0.001	-0.026		0.093	0.059		0.013	0.004
		(0.064)	(0.068)		(0.066)	(0.069)		(0.020)	(0.021)
Sarpanch woman:Sarpanch relative present		0.119	0.133		-0.322*	-0.298+		-0.069	-0.061
		(0.260)	(0.266)		(0.137)	(0.157)		(0.163)	(0.160)
Percent of village land held by biggest landholding caste			-0.003*			0.001			-0.000
			(0.001)			(0.001)			(0.000)
Distance (km) from block center			-0.001			0.002			-0.000
			(0.002)			(0.002)			(0.001)
Directly elected			0.011			0.032			0.032*
			(0.052)			(0.048)			(0.016)
Num.Obs.	350	350	324	350	350	324	350	350	324
R2	0.090	0.147	0.165	0.048	0.116	0.148	0.206	0.381	0.385

IC4 (H3.4): Assignment to any treatment (T1 or T2, considered jointly) will improve institutional centrality at a higher rate if the gram sevak is from a disadvantaged group.

IC4 is assessed by regressing our two primary measures of centrality (as specified immediately above in IC2) on a binary indicator of treatment (bundling T1 and T2 versus control), an indicator of the gender of the gram sevak and the multiplicative interaction between the two. Our focus is on the interaction effects for each of these specifications. We also control for the potential presence of a sarpanch relative during the interview (“the interview” being here understood as the group interview), both separately and in interaction with the gender of the sarpanch. Cluster-corrected standard errors will be used at the administrative block level. We check the robustness of results with bivariate regressions on treatment status and a multivariate regression including a standard set of demographic controls.
In additional analyses, we disentangle between T1 and T2. We also include the series of individual and village level controls presented in G2.
Deviation 1: We originally planned to use clustered standard errors at the taluka level. This was an error on our part: because the treatment assignment occurs at the level of each village council, we do not cluster standard errors in these analyses.
Precision 1: We originally said there were two primary measures of centrality. This was on error on our part: this should be three.

IC4 (H3.4) table, grouping T1/T2 together
	DV: Sarpanch made final budget decision			DV: Sarpanch responded first to roles/resp.			DV: Proportion of time sarpanch spoke
	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls
+ p < 0.1, * p < 0.05, p < 0.01, * p < 0.001
Each model is a linear regression of each outcome on an indicator for being assigned to discussion rules (1) versus not (0). Unadjusted and covariate-adjusted models presented. All regressions use robust standard errors.
(Intercept)	0.195***	0.355***	0.473***	0.771***	0.774***	0.716***	0.253***	0.281***	0.302***
	(0.037)	(0.082)	(0.110)	(0.039)	(0.075)	(0.095)	(0.015)	(0.025)	(0.036)
Discussion rules	0.130**	0.097+	0.109*	0.025	0.021	-0.004	0.025	0.006	0.006
	(0.048)	(0.051)	(0.054)	(0.047)	(0.048)	(0.047)	(0.018)	(0.016)	(0.016)
Gram sevak woman	-0.006	-0.021	-0.008	0.040	0.073	0.034	0.006	-0.005	-0.009
	(0.059)	(0.059)	(0.061)	(0.060)	(0.063)	(0.064)	(0.023)	(0.020)	(0.021)
Discussion rules:Gram sevak woman	-0.017	-0.003	-0.018	-0.012	-0.059	-0.026	0.019	0.020	0.022
	(0.075)	(0.075)	(0.078)	(0.072)	(0.074)	(0.074)	(0.028)	(0.024)	(0.025)
Sarpanch woman		-0.138	-0.102		-0.099	-0.032		-0.101**	-0.073+
		(0.120)	(0.135)		(0.130)	(0.127)		(0.039)	(0.040)
Sarpanch relative present		-0.001	-0.026		0.154***	0.129*		0.023	0.009
		(0.228)	(0.233)		(0.038)	(0.051)		(0.126)	(0.126)
Sarpanch SC		-0.167+	-0.179+		-0.040	0.008		0.015	0.024
		(0.095)	(0.098)		(0.074)	(0.076)		(0.021)	(0.022)
Sarpanch OBC		-0.053	-0.045		0.016	0.072		0.007	0.018
		(0.051)	(0.054)		(0.049)	(0.050)		(0.016)	(0.017)
Education above median		0.106**	0.121**		0.122***	0.129***		0.081***	0.079***
		(0.040)	(0.041)		(0.034)	(0.035)		(0.012)	(0.013)
Wealth above median		-0.023	-0.026		0.022	0.006		0.011	0.010
		(0.038)	(0.039)		(0.036)	(0.037)		(0.012)	(0.012)
Lives with in-laws		-0.111	-0.123		0.088	0.032		-0.005	-0.012
		(0.107)	(0.120)		(0.118)	(0.116)		(0.035)	(0.037)
Lives with own parents		-0.018	-0.041		0.015	0.005		0.026+	0.022
		(0.042)	(0.046)		(0.041)	(0.040)		(0.014)	(0.014)
Years as sarpanch		0.006	0.009		-0.016	-0.009		0.005	0.004
		(0.013)	(0.013)		(0.012)	(0.012)		(0.004)	(0.004)
Gender quota		0.004	-0.017		-0.105+	-0.152*		-0.018	-0.037+
		(0.057)	(0.065)		(0.063)	(0.065)		(0.018)	(0.019)
SC quota		0.111	0.118		0.083	0.053		-0.039+	-0.047*
		(0.093)	(0.097)		(0.072)	(0.072)		(0.020)	(0.022)
OBC quota		0.028	0.008		0.043	0.020		0.002	-0.006
		(0.050)	(0.053)		(0.048)	(0.047)		(0.016)	(0.016)
Sarpanch woman:Sarpanch relative present		-0.053	-0.035		-0.316**	-0.313**		-0.099	-0.099
		(0.236)	(0.242)		(0.098)	(0.107)		(0.128)	(0.127)
Percent of village land held by biggest landholding caste			-0.002*			0.000			-0.000
			(0.001)			(0.001)			(0.000)
Distance (km) from block center			0.001			0.002*			-0.000
			(0.002)			(0.001)			(0.001)
Directly elected			0.022			0.007			0.026*
			(0.042)			(0.036)			(0.013)
Num.Obs.	605	549	506	605	549	506	604	548	505
R2	0.017	0.119	0.129	0.002	0.089	0.119	0.014	0.367	0.377

IC4 (H3.4) table, comparing T1 to control
	DV: Sarpanch made final budget decision			DV: Sarpanch responded first to roles/resp.			DV: Proportion of time sarpanch spoke
	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls
+ p < 0.1, * p < 0.05, p < 0.01, * p < 0.001
Each model is a linear regression of each outcome on an indicator for being assigned to T1 (1) versus control (0). Unadjusted and covariate-adjusted models presented. All regressions use robust standard errors.
(Intercept)	0.195***	0.381***	0.455***	0.771***	0.764***	0.810***	0.253***	0.307***	0.338***
	(0.037)	(0.089)	(0.125)	(0.039)	(0.087)	(0.105)	(0.015)	(0.030)	(0.043)
T1	0.057	0.024	0.031	0.066	0.067	0.034	0.026	0.002	0.002
	(0.054)	(0.057)	(0.061)	(0.051)	(0.051)	(0.050)	(0.022)	(0.018)	(0.019)
Gram sevak woman	-0.006	-0.041	-0.032	0.040	0.073	0.038	0.006	-0.011	-0.014
	(0.059)	(0.060)	(0.063)	(0.060)	(0.063)	(0.063)	(0.023)	(0.020)	(0.021)
T1:Gram sevak woman	0.015	0.049	0.036	-0.040	-0.085	-0.042	0.024	0.026	0.032
	(0.084)	(0.084)	(0.089)	(0.079)	(0.081)	(0.081)	(0.032)	(0.027)	(0.028)
Sarpanch woman		-0.011	-0.019		0.035	0.050		-0.120**	-0.102**
		(0.133)	(0.144)		(0.136)	(0.124)		(0.039)	(0.038)
Sarpanch relative present		-0.089	-0.146		0.109	0.026		-0.121	-0.128
		(0.279)	(0.292)		(0.071)	(0.090)		(0.105)	(0.108)
Sarpanch SC		-0.148	-0.154		-0.009	0.025		-0.009	-0.006
		(0.097)	(0.102)		(0.091)	(0.094)		(0.027)	(0.030)
Sarpanch OBC		-0.067	-0.057		0.044	0.083		-0.001	0.008
		(0.057)	(0.060)		(0.057)	(0.059)		(0.020)	(0.021)
Education above median		0.120**	0.121*		0.130**	0.126**		0.101***	0.096***
		(0.046)	(0.047)		(0.042)	(0.042)		(0.015)	(0.015)
Wealth above median		-0.033	-0.025		0.011	-0.001		0.005	0.004
		(0.043)	(0.046)		(0.043)	(0.044)		(0.014)	(0.014)
Lives with in-laws		-0.216+	-0.209+		-0.072	-0.079		0.008	0.005
		(0.116)	(0.125)		(0.127)	(0.116)		(0.035)	(0.032)
Lives with own parents		-0.073	-0.090+		-0.037	-0.049		0.002	-0.003
		(0.048)	(0.053)		(0.050)	(0.049)		(0.017)	(0.018)
Years as sarpanch		0.017	0.022		-0.008	-0.005		0.006	0.005
		(0.015)	(0.015)		(0.014)	(0.014)		(0.005)	(0.005)
Gender quota		-0.059	-0.041		-0.094	-0.150*		-0.029	-0.043+
		(0.068)	(0.075)		(0.069)	(0.066)		(0.021)	(0.022)
SC quota		0.049	0.051		0.124	0.102		-0.026	-0.028
		(0.094)	(0.099)		(0.086)	(0.089)		(0.026)	(0.028)
OBC quota		0.072	0.057		0.030	0.028		0.002	-0.007
		(0.057)	(0.059)		(0.057)	(0.055)		(0.019)	(0.019)
Sarpanch woman:Sarpanch relative present		0.044	0.086		-0.270*	-0.211		0.041	0.033
		(0.287)	(0.300)		(0.129)	(0.141)		(0.107)	(0.111)
Percent of village land held by biggest landholding caste			-0.002*			-0.001			-0.000
			(0.001)			(0.001)			(0.000)
Distance (km) from block center			0.002			0.004*			-0.000
			(0.002)			(0.001)			(0.001)
Directly elected			0.011			0.023			0.024
			(0.049)			(0.042)			(0.016)
Num.Obs.	407	370	340	407	370	340	406	369	339
R2	0.006	0.142	0.142	0.005	0.101	0.149	0.018	0.397	0.400

IC4 (H3.4) table, comparing T2 to control
	DV: Sarpanch made final budget decision			DV: Sarpanch responded first to roles/resp.			DV: Proportion of time sarpanch spoke
	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls
+ p < 0.1, * p < 0.05, p < 0.01, * p < 0.001
Each model is a linear regression of each outcome on an indicator for being assigned to T2 (1) versus control (0). Unadjusted and covariate-adjusted models presented. All regressions use robust standard errors.
(Intercept)	0.195***	0.328**	0.528***	0.771***	0.781***	0.728***	0.253***	0.254***	0.283***
	(0.037)	(0.101)	(0.134)	(0.039)	(0.094)	(0.128)	(0.015)	(0.032)	(0.048)
t2	0.212***	0.165*	0.181**	-0.021	-0.025	-0.051	0.025	0.003	0.004
	(0.060)	(0.065)	(0.068)	(0.057)	(0.058)	(0.059)	(0.021)	(0.018)	(0.019)
Gram sevak woman	-0.006	-0.015	0.002	0.040	0.080	0.041	0.006	-0.001	-0.005
	(0.059)	(0.060)	(0.063)	(0.060)	(0.064)	(0.065)	(0.023)	(0.021)	(0.021)
t2:Gram sevak woman	-0.057	-0.051	-0.064	0.021	-0.045	-0.017	0.013	0.019	0.018
	(0.091)	(0.094)	(0.097)	(0.084)	(0.089)	(0.091)	(0.032)	(0.028)	(0.029)
Sarpanch woman		-0.229	-0.157		-0.185	-0.130		-0.094+	-0.063
		(0.159)	(0.192)		(0.160)	(0.170)		(0.055)	(0.060)
Sarpanch relative present		-0.119	-0.128		0.194**	0.147+		-0.004	-0.020
		(0.259)	(0.264)		(0.068)	(0.085)		(0.164)	(0.160)
Sarpanch SC		-0.175	-0.182		-0.008	0.039		0.026	0.040
		(0.133)	(0.132)		(0.100)	(0.103)		(0.025)	(0.027)
Sarpanch OBC		-0.050	-0.057		-0.022	0.055		0.007	0.017
		(0.066)	(0.071)		(0.069)	(0.073)		(0.020)	(0.020)
Education above median		0.122*	0.153**		0.145**	0.163***		0.071***	0.069***
		(0.051)	(0.052)		(0.045)	(0.047)		(0.015)	(0.016)
Wealth above median		-0.009	-0.012		-0.001	-0.022		0.015	0.012
		(0.046)	(0.048)		(0.046)	(0.047)		(0.014)	(0.014)
Lives with in-laws		0.003	-0.022		0.126	0.073		-0.015	-0.026
		(0.142)	(0.168)		(0.134)	(0.140)		(0.052)	(0.056)
Lives with own parents		-0.005	-0.031		0.033	0.013		0.049**	0.042*
		(0.055)	(0.058)		(0.054)	(0.051)		(0.017)	(0.018)
Years as sarpanch		0.000	-0.002		-0.026+	-0.021		0.005	0.002
		(0.016)	(0.017)		(0.016)	(0.017)		(0.005)	(0.005)
Gender quota		0.009	-0.049		-0.028	-0.076		0.001	-0.022
		(0.081)	(0.094)		(0.092)	(0.098)		(0.022)	(0.023)
SC quota		0.161	0.144		0.039	0.021		-0.030	-0.046
		(0.133)	(0.133)		(0.098)	(0.101)		(0.027)	(0.029)
OBC quota		-0.004	-0.031		0.099	0.065		0.011	0.002
		(0.064)	(0.068)		(0.067)	(0.069)		(0.020)	(0.021)
Sarpanch woman:Sarpanch relative present		0.088	0.101		-0.329*	-0.303*		-0.066	-0.057
		(0.270)	(0.277)		(0.138)	(0.150)		(0.166)	(0.162)
Percent of village land held by biggest landholding caste			-0.003*			0.001			-0.000
			(0.001)			(0.001)			(0.000)
Distance (km) from block center			-0.001			0.001			-0.000
			(0.002)			(0.002)			(0.001)
Directly elected			0.007			0.029			0.031+
			(0.052)			(0.048)			(0.016)
Num.Obs.	390	351	325	390	351	325	390	351	325
R2	0.045	0.138	0.161	0.004	0.098	0.131	0.012	0.367	0.377

IC5 (H3.5): Assignment to any treatment (T1 or T2, considered jointly) will improve institutional centrality at a higher rate if the treatment also led to equal participation (IC1).

IC5 is assessed by regressing our primary measures of centrality on a binary indicator of treatment (bundling T1 and T2 versus control), indicators of equal participation (as described in IC 1 above) and the multiplicative interaction between the two. Our focus is on the interaction effects for each of these specifications. We also control for the potential presence of a sarpanch relative during the interview (“the interview” being here understood as the group interview), both separately and in interaction with the gender of the sarpanch. Cluster-corrected standard errors will be used at the administrative block level. We check the robustness of results with bivariate regressions on treatment status and a multivariate regression including a standard set of demographic controls.
In additional analyses, we disentangle between T1 and T2. We also include the series of individual and village level controls presented in G2.
Deviation 1: We originally planned to use clustered standard errors at the taluka level. This was an error on our part: because the treatment assignment occurs at the level of each village council, we do not cluster standard errors in these analyses.
Deviation 2: This analysis may induce post-treatment bias through conditioning on a post-treatment variable (treatment take-up). We conduct the analysis as specified but in the main paper will report the estimation of complier average causal effects instead.
Deviation 3: We initially specified using both indicators of equal participation, however we only include the indicator for whether or not the sarpanch spoke at all, as speaking first was not the prompt for T2.

IC5 (H3.5) table, grouping T1/T2 together
	DV: Sarpanch made final budget decision			DV: Sarpanch responded first to roles/resp.			DV: Proportion of time sarpanch spoke
	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls
+ p < 0.1, * p < 0.05, p < 0.01, * p < 0.001
Each model is a linear regression of each outcome on an indicator for being assigned to discussion rules (1) versus not (0). Unadjusted and covariate-adjusted models presented. All regressions use robust standard errors.
(Intercept)	0.026	0.233**	0.330**	0.711***	0.805***	0.729***	0.145***	0.221***	0.239***
	(0.026)	(0.077)	(0.100)	(0.075)	(0.097)	(0.114)	(0.020)	(0.028)	(0.037)
Discussion rules	0.075	0.034	0.109*	-0.100	-0.156	-0.179+	0.017	-0.006	-0.001
	(0.048)	(0.049)	(0.048)	(0.098)	(0.100)	(0.106)	(0.025)	(0.023)	(0.025)
Sarpanch spoke budget question	0.207***	0.116*	0.154***	0.095	-0.024	-0.040	0.138***	0.062**	0.060**
	(0.043)	(0.049)	(0.046)	(0.081)	(0.084)	(0.085)	(0.023)	(0.021)	(0.022)
Discussion rules:Sarpanch spoke budget question	0.041	0.066	-0.016	0.137	0.183+	0.196+	0.010	0.021	0.016
	(0.064)	(0.065)	(0.068)	(0.105)	(0.108)	(0.114)	(0.029)	(0.027)	(0.028)
Sarpanch woman		-0.136	-0.113		-0.088	-0.028		-0.097**	-0.074+
		(0.118)	(0.134)		(0.127)	(0.129)		(0.037)	(0.040)
Sarpanch relative present		0.006	-0.009		0.133**	0.116*		0.030	0.020
		(0.221)	(0.222)		(0.041)	(0.056)		(0.119)	(0.118)
Sarpanch SC		-0.172+	-0.181+		-0.045	0.007		0.014	0.024
		(0.094)	(0.098)		(0.075)	(0.076)		(0.020)	(0.022)
Sarpanch OBC		-0.051	-0.046		0.018	0.075		0.010	0.020
		(0.051)	(0.054)		(0.048)	(0.050)		(0.016)	(0.017)
Education above median		0.094*	0.111**		0.115***	0.123***		0.075***	0.074***
		(0.040)	(0.041)		(0.034)	(0.035)		(0.012)	(0.012)
Wealth above median		-0.033	-0.033		0.016	0.000		0.007	0.006
		(0.038)	(0.040)		(0.036)	(0.036)		(0.011)	(0.012)
Lives with in-laws		-0.098	-0.102		0.091	0.041		-0.000	-0.005
		(0.107)	(0.120)		(0.117)	(0.119)		(0.034)	(0.036)
Lives with own parents		-0.007	-0.034		0.022	0.011		0.031*	0.027+
		(0.043)	(0.046)		(0.041)	(0.039)		(0.013)	(0.014)
Years as sarpanch		0.008	0.010		-0.013	-0.007		0.006+	0.005
		(0.012)	(0.013)		(0.012)	(0.012)		(0.004)	(0.004)
Gender quota		0.018	-0.001		-0.102	-0.149*		-0.013	-0.030
		(0.055)	(0.063)		(0.063)	(0.064)		(0.018)	(0.020)
SC quota		0.127	0.128		0.094	0.064		-0.036+	-0.044*
		(0.092)	(0.096)		(0.072)	(0.073)		(0.019)	(0.021)
OBC quota		0.016	-0.004		0.036	0.014		-0.006	-0.014
		(0.050)	(0.053)		(0.047)	(0.047)		(0.016)	(0.016)
Sarpanch woman:Sarpanch relative present		-0.031	-0.013		-0.283**	-0.282*		-0.094	-0.093
		(0.228)	(0.230)		(0.099)	(0.109)		(0.121)	(0.120)
Percent of village land held by biggest landholding caste			-0.002*			0.001			-0.000
			(0.001)			(0.001)			(0.000)
Distance (km) from block center			0.000			0.002+			-0.000
			(0.002)			(0.001)			(0.001)
Directly elected			0.020			0.009			0.025*
			(0.042)			(0.036)			(0.013)
Num.Obs.	605	549	506	605	549	506	604	548	505
R2	0.053	0.134	0.140	0.032	0.099	0.130	0.121	0.392	0.397

IC4 (H3.4) table, comparing T1 to control
	DV: Sarpanch made final budget decision			DV: Sarpanch responded first to roles/resp.			DV: Proportion of time sarpanch spoke
	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls
+ p < 0.1, * p < 0.05, p < 0.01, * p < 0.001
Each model is a linear regression of each outcome on an indicator for being assigned to T1 (1) versus control (0). Unadjusted and covariate-adjusted models presented. All regressions use robust standard errors.
(Intercept)	0.026	0.267**	0.301*	0.711***	0.803***	0.816***	0.145***	0.255***	0.277***
	(0.026)	(0.084)	(0.117)	(0.075)	(0.106)	(0.121)	(0.020)	(0.032)	(0.043)
T1	-0.026	-0.030	0.024	-0.124	-0.160	-0.172	0.000	-0.018	-0.020
	(0.026)	(0.041)	(0.038)	(0.119)	(0.122)	(0.131)	(0.027)	(0.025)	(0.028)
Sarpanch spoke budget question	0.207***	0.109*	0.153**	0.095	-0.023	-0.035	0.138***	0.053**	0.053*
	(0.043)	(0.049)	(0.047)	(0.081)	(0.082)	(0.082)	(0.023)	(0.020)	(0.021)
T1:Sarpanch spoke budget question	0.088	0.080	0.016	0.195	0.226+	0.219	0.031	0.032	0.036
	(0.055)	(0.063)	(0.066)	(0.126)	(0.129)	(0.137)	(0.032)	(0.029)	(0.032)
Sarpanch woman		-0.017	-0.023		0.031	0.057		-0.118**	-0.098**
		(0.128)	(0.137)		(0.139)	(0.129)		(0.037)	(0.037)
Sarpanch relative present		-0.044	-0.087		0.089	0.024		-0.103	-0.106
		(0.264)	(0.271)		(0.074)	(0.105)		(0.091)	(0.093)
Sarpanch SC		-0.152	-0.152		-0.019	0.026		-0.010	-0.004
		(0.096)	(0.101)		(0.091)	(0.092)		(0.026)	(0.028)
Sarpanch OBC		-0.060	-0.053		0.046	0.087		0.004	0.012
		(0.057)	(0.060)		(0.057)	(0.058)		(0.019)	(0.020)
Education above median		0.102*	0.105*		0.118**	0.120**		0.093***	0.090***
		(0.046)	(0.048)		(0.042)	(0.043)		(0.015)	(0.015)
Wealth above median		-0.044	-0.032		0.005	-0.006		0.000	0.001
		(0.043)	(0.046)		(0.043)	(0.044)		(0.014)	(0.014)
Lives with in-laws		-0.201+	-0.192		-0.061	-0.072		0.012	0.009
		(0.114)	(0.121)		(0.131)	(0.121)		(0.033)	(0.031)
Lives with own parents		-0.064	-0.082		-0.035	-0.046		0.006	0.002
		(0.048)	(0.052)		(0.050)	(0.049)		(0.017)	(0.017)
Years as sarpanch		0.016	0.022		-0.007	-0.004		0.006	0.005
		(0.014)	(0.015)		(0.014)	(0.014)		(0.005)	(0.005)
Gender quota		-0.040	-0.023		-0.085	-0.150*		-0.023	-0.038+
		(0.065)	(0.072)		(0.070)	(0.065)		(0.021)	(0.022)
SC quota		0.064	0.055		0.142	0.111		-0.021	-0.027
		(0.092)	(0.097)		(0.087)	(0.087)		(0.024)	(0.026)
OBC quota		0.061	0.042		0.028	0.026		-0.004	-0.014
		(0.056)	(0.059)		(0.055)	(0.053)		(0.018)	(0.019)
Sarpanch woman:Sarpanch relative present		0.016	0.058		-0.249+	-0.201		0.031	0.024
		(0.272)	(0.280)		(0.131)	(0.152)		(0.094)	(0.096)
Percent of village land held by biggest landholding caste			-0.002+			-0.001			-0.000
			(0.001)			(0.001)			(0.000)
Distance (km) from block center			0.002			0.003*			-0.000
			(0.002)			(0.002)			(0.001)
Directly elected			0.004			0.023			0.021
			(0.048)			(0.041)			(0.015)
Num.Obs.	407	370	340	407	370	340	406	369	339
R2	0.055	0.156	0.156	0.043	0.112	0.160	0.136	0.417	0.418

IC4 (H3.4) table, comparing T2 to control
	DV: Sarpanch made final budget decision			DV: Sarpanch responded first to roles/resp.			DV: Proportion of time sarpanch spoke
	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls
+ p < 0.1, * p < 0.05, p < 0.01, * p < 0.001
Each model is a linear regression of each outcome on an indicator for being assigned to T2 (1) versus control (0). Unadjusted and covariate-adjusted models presented. All regressions use robust standard errors.
(Intercept)	0.026	0.191+	0.388**	0.711***	0.825***	0.765***	0.145***	0.195***	0.226***
	(0.026)	(0.098)	(0.125)	(0.075)	(0.111)	(0.141)	(0.020)	(0.034)	(0.048)
t2	0.174*	0.085	0.166*	-0.077	-0.153	-0.177	0.033	-0.002	0.007
	(0.079)	(0.074)	(0.076)	(0.116)	(0.115)	(0.126)	(0.032)	(0.030)	(0.030)
Sarpanch spoke budget question	0.207***	0.120*	0.165***	0.095	-0.027	-0.052	0.138***	0.063**	0.062**
	(0.043)	(0.052)	(0.049)	(0.081)	(0.084)	(0.087)	(0.023)	(0.022)	(0.023)
t2:Sarpanch spoke budget question	0.003	0.064	-0.020	0.076	0.136	0.146	-0.010	0.015	0.005
	(0.094)	(0.091)	(0.095)	(0.125)	(0.125)	(0.135)	(0.036)	(0.033)	(0.033)
Sarpanch woman		-0.199	-0.158		-0.174	-0.129		-0.092+	-0.072
		(0.151)	(0.185)		(0.155)	(0.167)		(0.053)	(0.061)
Sarpanch relative present		-0.126	-0.125		0.178*	0.131		0.007	-0.008
		(0.244)	(0.246)		(0.073)	(0.094)		(0.156)	(0.151)
Sarpanch SC		-0.170	-0.180		-0.007	0.040		0.028	0.041
		(0.132)	(0.130)		(0.102)	(0.104)		(0.026)	(0.027)
Sarpanch OBC		-0.046	-0.058		-0.021	0.057		0.010	0.018
		(0.066)	(0.070)		(0.069)	(0.073)		(0.019)	(0.020)
Education above median		0.117*	0.144**		0.145**	0.163***		0.068***	0.065***
		(0.050)	(0.051)		(0.046)	(0.048)		(0.015)	(0.016)
Wealth above median		-0.019	-0.022		-0.002	-0.024		0.011	0.008
		(0.047)	(0.048)		(0.046)	(0.046)		(0.014)	(0.014)
Lives with in-laws		0.009	-0.001		0.128	0.078		-0.007	-0.014
		(0.137)	(0.164)		(0.131)	(0.140)		(0.050)	(0.056)
Lives with own parents		0.008	-0.024		0.033	0.015		0.052**	0.046**
		(0.056)	(0.059)		(0.054)	(0.051)		(0.017)	(0.017)
Years as sarpanch		0.005	0.001		-0.024	-0.019		0.006	0.004
		(0.016)	(0.017)		(0.016)	(0.017)		(0.004)	(0.005)
Gender quota		0.004	-0.039		-0.037	-0.080		0.003	-0.013
		(0.075)	(0.087)		(0.091)	(0.095)		(0.023)	(0.025)
SC quota		0.171	0.151		0.035	0.023		-0.033	-0.047+
		(0.132)	(0.131)		(0.099)	(0.102)		(0.026)	(0.028)
OBC quota		-0.018	-0.048		0.093	0.063		0.002	-0.007
		(0.065)	(0.069)		(0.066)	(0.069)		(0.020)	(0.021)
Sarpanch woman:Sarpanch relative present		0.140	0.152		-0.306*	-0.283+		-0.060	-0.051
		(0.254)	(0.257)		(0.137)	(0.152)		(0.158)	(0.153)
Percent of village land held by biggest landholding caste			-0.003*			0.001			-0.000
			(0.001)			(0.001)			(0.000)
Distance (km) from block center			-0.002			0.001			-0.000
			(0.002)			(0.002)			(0.001)
Directly elected			0.005			0.033			0.030+
			(0.051)			(0.048)			(0.016)
Num.Obs.	390	351	325	390	351	325	390	351	325
R2	0.073	0.150	0.173	0.015	0.098	0.133	0.116	0.391	0.396

IC6 (H3.6a): The sarpanch will indicate greater comfort, confidence and engagement, and less fear in her/his comportment for those assigned to treatment (T1 or T2, considered jointly) versus not (Control).

IC6 is assessed by regressing indicators of the sarpanch’s comfort, confidence and engagement during the group meeting on a binary indicator of treatment (bundling T1 and T2 versus control), and on the set of demographic controls defined in our online appendix. We also control for the potential presence of a sarpanch relative during the interview (“the interview” being here understood as the group interview), both separately and in interaction with the gender of the sarpanch. Cluster-corrected standard errors will be used at the administrative block level in both cases.
In additional analyses, we disentangle between T1 and T2. We also include the series of individual and village level controls presented in G2.
To measure the sarpanch’s comfort, confidence and engagement we rely on the following items from the group survey:
- Q71: How loudly did the sarpanch speak, compared to others, during the interview? (Code volume as 1 if Sarpanch was either the loudest (1) or average, compared to everyone else (2). Code volume as 0 if the sarpanch was the quietest (3)). [Note: This is only coded if the sarpanch spoke at all during the entire interview, i.e. Q70 = 1.]
- Q72: IF YES to sarpanch speaks above: Did the sarpanch speak in a way that implied inferior status/deference towards other participants to the meeting, or did they seek approval from others when speaking? (Code equal status as 1 if answer is 0 (No), 0 if answer is 1 (Yes)).
- Q73: IF YES to sarpanch speaks above: Was the sarpanch’s speech hesitant (stuttering, slow, broken up, or interrogative) at times? (Code confidence as 1 if answer is 0 (No), 0 if answer is 1 (Yes)).
- Q75: IF YES TO sarpanch speaks above: Did she/he ever attempt to move the discussion toward his/her opinion? For example - did she/he ever try to convince or persuade others of her opinion? (Code: Engaged decision making as 1 if answer is 1 (Yes), and 0 if answer is 0 (No)).
- Q77: IF YES TO sarpanch speaks above: Did the sarpanch try to understand the opinion of others? For example, did the sarpanch actively ask the opinion or reasoning of other people in the room?(Code: Active listening as 1 if answer is 1 (Yes), and 0 if answer is 0 (No)).
- Q78: IF YES TO sarpanch speaks above, did the sarpanch engage in consensus building with others, trying to come to an agreement? (Code Consensus building as 1 if answer is 1 (Yes), and 0 if answer is 0 (No)).
- Q98.1-3: Overall, based on your own assessment and impressions, how would you characterize the body language of the sarpanch throughout the interview? Did her/his body language suggest…..Confidence/ Fear/Nervousness / Interest in the conversation? Code Confidence as 1 if Q98.1 answer is 1 (Definitely yes), 0 if answer is 2-3 (Maybe/ Definitely no). Code Fearless as 1 if Q98.2 answer is 2-3 (Maybe/ Definitely no), 0 if answer is 1 (Definitely yes). Code Interest as 1 if Q98.3 answer is 1 (Definitely yes), 0 if answer is 2-3 (Maybe/ Definitely no).
Deviation 1: We originally planned to use clustered standard errors at the taluka level. This was an error on our part: because the treatment assignment occurs at the level of each village council, we do not cluster standard errors in these analyses.
Precision 1: For the first six measures, we code sarpanches who did not speak at all as not exihibiting the measures of confidence.

IC6 (H3.6a) table, grouping T1/T2 together
	DV: Sarpanch loudest			DV: Sarpanch equal status.			DV: Sarpanch confidence speech			DV: Sarpanch engaged decision-making			DV: Sarpanch active listening			DV: Sarpanch consensus building			DV: Sarpanch body language confidence			DV: Sarpanch body language fearlessness			DV: Sarpanch body language interest
	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls
+ p < 0.1, * p < 0.05, p < 0.01, * p < 0.001
Each model is a linear regression of each outcome on an indicator for being assigned to discussion rules (1) versus not (0). Unadjusted and covariate-adjusted models presented. All regressions use robust standard errors.
(Intercept)	0.687***	0.699***	0.664***	0.458***	0.428***	0.467***	0.536***	0.537***	0.557***	0.417***	0.555***	0.606***	0.531***	0.619***	0.622***	0.734***	0.871***	0.849***	0.578***	0.628***	0.709***	0.771***	0.677***	0.698***	0.641***	0.685***	0.736***
	(0.034)	(0.075)	(0.102)	(0.036)	(0.087)	(0.120)	(0.036)	(0.077)	(0.110)	(0.036)	(0.080)	(0.116)	(0.036)	(0.087)	(0.120)	(0.032)	(0.081)	(0.107)	(0.036)	(0.078)	(0.110)	(0.030)	(0.068)	(0.090)	(0.035)	(0.076)	(0.110)
Discussion rules	0.029	-0.013	-0.002	-0.027	-0.034	-0.035	0.049	0.000	-0.015	-0.063	-0.098*	-0.102*	0.028	0.010	0.013	-0.008	-0.025	-0.031	0.008	-0.028	-0.009	0.014	0.004	-0.007	0.023	-0.006	-0.006
	(0.040)	(0.038)	(0.040)	(0.044)	(0.046)	(0.048)	(0.043)	(0.040)	(0.042)	(0.043)	(0.044)	(0.046)	(0.044)	(0.046)	(0.048)	(0.039)	(0.041)	(0.041)	(0.043)	(0.040)	(0.043)	(0.037)	(0.037)	(0.037)	(0.042)	(0.041)	(0.043)
Sarpanch woman		-0.160	-0.168		-0.213	-0.137		-0.219+	-0.260+		-0.166	-0.118		-0.154	-0.208		-0.157	-0.193		-0.381**	-0.286+		-0.097	-0.124		-0.174	-0.182
		(0.106)	(0.112)		(0.143)	(0.158)		(0.132)	(0.144)		(0.143)	(0.156)		(0.151)	(0.152)		(0.125)	(0.133)		(0.147)	(0.159)		(0.106)	(0.119)		(0.153)	(0.163)
Sarpanch relative present		0.135*	0.121+		-0.108	-0.123		0.202***	0.194***		0.116	0.079		0.218	0.221		0.243***	0.252***		0.017	-0.025		0.105***	0.096+		-0.002	-0.000
		(0.054)	(0.064)		(0.248)	(0.254)		(0.045)	(0.042)		(0.263)	(0.254)		(0.206)	(0.181)		(0.050)	(0.059)		(0.202)	(0.192)		(0.029)	(0.050)		(0.203)	(0.209)
Sarpanch SC		0.150*	0.153*		-0.012	-0.021		0.143+	0.139+		0.049	0.057		0.092	0.072		0.059	0.018		0.143+	0.138+		0.086	0.053		0.116	0.097
		(0.075)	(0.077)		(0.099)	(0.100)		(0.077)	(0.082)		(0.089)	(0.094)		(0.094)	(0.097)		(0.093)	(0.093)		(0.073)	(0.077)		(0.084)	(0.083)		(0.096)	(0.099)
Sarpanch OBC		-0.023	0.002		-0.042	-0.018		0.070	0.065		-0.032	-0.016		-0.070	-0.060		-0.001	-0.019		0.051	0.070		0.079+	0.056		0.002	-0.011
		(0.049)	(0.051)		(0.059)	(0.064)		(0.055)	(0.059)		(0.061)	(0.064)		(0.060)	(0.063)		(0.052)	(0.054)		(0.056)	(0.059)		(0.047)	(0.047)		(0.057)	(0.060)
Education above median		0.186***	0.212***		-0.009	-0.010		0.141***	0.137**		0.088*	0.080+		0.030	0.028		0.051	0.049		0.151***	0.159***		0.128***	0.143***		0.109**	0.134**
		(0.037)	(0.038)		(0.045)	(0.047)		(0.041)	(0.043)		(0.044)	(0.046)		(0.045)	(0.047)		(0.039)	(0.040)		(0.042)	(0.043)		(0.034)	(0.035)		(0.041)	(0.043)
Wealth above median		0.021	0.006		0.013	-0.024		0.036	0.037		0.045	0.052		0.128**	0.144**		0.073+	0.092*		0.094*	0.074+		0.038	0.049		0.045	0.060
		(0.038)	(0.038)		(0.044)	(0.046)		(0.040)	(0.042)		(0.043)	(0.045)		(0.044)	(0.046)		(0.039)	(0.041)		(0.041)	(0.043)		(0.035)	(0.036)		(0.041)	(0.043)
Lives with in-laws		-0.102	-0.056		0.137	0.138		-0.038	-0.005		-0.019	-0.038		0.041	0.083		0.023	0.050		0.060	0.050		0.038	0.054		-0.055	-0.026
		(0.084)	(0.085)		(0.130)	(0.142)		(0.119)	(0.125)		(0.126)	(0.133)		(0.134)	(0.135)		(0.102)	(0.109)		(0.133)	(0.141)		(0.092)	(0.101)		(0.141)	(0.148)
Lives with own parents		0.043	0.051		-0.027	-0.036		0.010	0.014		0.022	0.006		-0.008	0.004		-0.067	-0.040		0.046	0.041		0.070+	0.082+		0.005	-0.002
		(0.044)	(0.044)		(0.049)	(0.052)		(0.045)	(0.049)		(0.047)	(0.050)		(0.049)	(0.052)		(0.046)	(0.049)		(0.047)	(0.049)		(0.040)	(0.042)		(0.047)	(0.050)
Years as sarpanch		0.017	0.018		0.051***	0.046**		0.050***	0.045***		-0.016	-0.017		-0.026+	-0.027+		-0.024+	-0.020		0.005	0.003		0.020+	0.021+		0.010	0.009
		(0.013)	(0.013)		(0.014)	(0.016)		(0.013)	(0.013)		(0.014)	(0.015)		(0.014)	(0.015)		(0.013)	(0.013)		(0.013)	(0.014)		(0.011)	(0.012)		(0.014)	(0.015)
Gender quota		0.016	-0.008		0.015	-0.045		-0.135+	-0.114		-0.001	-0.053		-0.022	-0.004		0.062	0.071		0.037	-0.046		-0.087	-0.080		-0.004	-0.026
		(0.076)	(0.082)		(0.075)	(0.082)		(0.075)	(0.081)		(0.068)	(0.078)		(0.075)	(0.080)		(0.076)	(0.080)		(0.072)	(0.081)		(0.070)	(0.076)		(0.074)	(0.078)
SC quota		-0.118	-0.108		-0.023	-0.020		-0.127+	-0.126		-0.082	-0.067		-0.107	-0.091		-0.087	-0.065		-0.099	-0.106		-0.063	-0.061		-0.062	-0.043
		(0.073)	(0.075)		(0.096)	(0.097)		(0.076)	(0.081)		(0.087)	(0.093)		(0.092)	(0.095)		(0.091)	(0.091)		(0.071)	(0.076)		(0.081)	(0.080)		(0.093)	(0.097)
OBC quota		0.012	0.017		-0.029	-0.037		-0.034	-0.038		-0.034	-0.030		0.069	0.070		-0.060	-0.047		-0.080	-0.109+		-0.045	-0.033		0.007	-0.007
		(0.049)	(0.050)		(0.058)	(0.063)		(0.055)	(0.059)		(0.060)	(0.063)		(0.060)	(0.063)		(0.052)	(0.054)		(0.055)	(0.058)		(0.046)	(0.045)		(0.057)	(0.061)
Sarpanch woman:Sarpanch relative present		-0.285**	-0.317**		-0.084	-0.127		-0.348***	-0.379***		-0.139	-0.079		-0.092	-0.131		-0.374***	-0.429***		-0.251	-0.218		-0.280**	-0.293**		-0.205	-0.200
		(0.103)	(0.109)		(0.261)	(0.267)		(0.090)	(0.089)		(0.275)	(0.269)		(0.225)	(0.204)		(0.105)	(0.111)		(0.216)	(0.210)		(0.094)	(0.105)		(0.222)	(0.229)
Percent of village land held by biggest landholding caste			-0.001			-0.001			0.000			-0.000			0.001			0.001			-0.002+			-0.001			-0.001
			(0.001)			(0.001)			(0.001)			(0.001)			(0.001)			(0.001)			(0.001)			(0.001)			(0.001)
Distance (km) from block center			0.001			-0.001			-0.001			-0.002			-0.003+			-0.002			-0.000			0.002+			-0.000
			(0.001)			(0.002)			(0.002)			(0.002)			(0.002)			(0.002)			(0.002)			(0.001)			(0.002)
Directly elected			0.038			0.084+			0.043			0.071			-0.028			-0.066			0.079+			-0.007			-0.013
			(0.038)			(0.048)			(0.043)			(0.047)			(0.047)			(0.043)			(0.043)			(0.035)			(0.043)
Num.Obs.	605	549	506	605	549	506	605	549	506	605	549	506	605	549	506	605	549	506	605	549	506	605	549	506	605	549	506
R2	0.001	0.195	0.218	0.001	0.044	0.053	0.002	0.244	0.249	0.004	0.080	0.089	0.001	0.057	0.070	0.000	0.049	0.068	0.000	0.217	0.223	0.000	0.135	0.163	0.000	0.121	0.129

IC6 (H3.6a) table, effect of T1
	DV: Sarpanch loudest			DV: Sarpanch equal status.			DV: Sarpanch confidence speech			DV: Sarpanch engaged decision-making			DV: Sarpanch active listening			DV: Sarpanch consensus building			DV: Sarpanch body language confidence			DV: Sarpanch body language fearlessness			DV: Sarpanch body language interest
	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls
+ p < 0.1, * p < 0.05, p < 0.01, * p < 0.001
Each model is a linear regression of each outcome on an indicator for being assigned to T1 (1) versus not (0). Unadjusted and covariate-adjusted models presented. All regressions use robust standard errors.
(Intercept)	0.688***	0.728***	0.736***	0.458***	0.377***	0.403**	0.536***	0.578***	0.674***	0.417***	0.533***	0.491***	0.531***	0.650***	0.731***	0.734***	0.833***	0.826***	0.578***	0.667***	0.776***	0.771***	0.656***	0.781***	0.641***	0.645***	0.727***
	(0.034)	(0.089)	(0.123)	(0.036)	(0.102)	(0.147)	(0.036)	(0.088)	(0.130)	(0.036)	(0.096)	(0.141)	(0.036)	(0.101)	(0.144)	(0.032)	(0.094)	(0.126)	(0.036)	(0.089)	(0.131)	(0.030)	(0.078)	(0.106)	(0.035)	(0.084)	(0.128)
T1	0.033	-0.004	0.011	-0.035	-0.046	-0.043	0.031	-0.027	-0.034	-0.026	-0.058	-0.058	0.008	-0.010	-0.006	-0.018	-0.029	-0.039	-0.006	-0.046	-0.019	0.025	0.010	-0.003	0.052	0.017	0.017
	(0.045)	(0.043)	(0.046)	(0.049)	(0.052)	(0.055)	(0.049)	(0.045)	(0.048)	(0.049)	(0.051)	(0.054)	(0.050)	(0.053)	(0.055)	(0.044)	(0.047)	(0.048)	(0.049)	(0.046)	(0.049)	(0.041)	(0.042)	(0.043)	(0.047)	(0.045)	(0.048)
Sarpanch woman		-0.205	-0.173		-0.191	-0.180		-0.362*	-0.369*		-0.138	-0.076		-0.187	-0.150		-0.282+	-0.264+		-0.481**	-0.420*		-0.158	-0.165		-0.272	-0.243
		(0.125)	(0.131)		(0.190)	(0.196)		(0.165)	(0.169)		(0.198)	(0.197)		(0.176)	(0.181)		(0.147)	(0.152)		(0.173)	(0.175)		(0.139)	(0.140)		(0.175)	(0.180)
Sarpanch relative present		0.051	0.016		0.116	0.122		0.133***	0.114+		-0.215	-0.198		0.054	0.062		0.208**	0.217*		-0.205	-0.226		0.073+	0.002		-0.251	-0.276
		(0.084)	(0.108)		(0.352)	(0.325)		(0.039)	(0.060)		(0.290)	(0.293)		(0.332)	(0.257)		(0.080)	(0.098)		(0.295)	(0.300)		(0.044)	(0.058)		(0.300)	(0.301)
Sarpanch SC		0.096	0.095		0.069	0.056		0.055	0.034		0.125	0.135		0.032	-0.006		0.030	-0.019		0.056	0.029		0.038	-0.018		0.029	-0.001
		(0.089)	(0.092)		(0.115)	(0.119)		(0.093)	(0.100)		(0.110)	(0.120)		(0.114)	(0.118)		(0.113)	(0.113)		(0.086)	(0.095)		(0.102)	(0.099)		(0.110)	(0.113)
Sarpanch OBC		-0.054	-0.009		-0.071	-0.043		0.059	0.052		-0.019	-0.008		-0.076	-0.068		-0.025	-0.041		0.003	0.027		0.061	0.040		-0.041	-0.047
		(0.059)	(0.061)		(0.071)	(0.077)		(0.067)	(0.072)		(0.074)	(0.076)		(0.073)	(0.077)		(0.064)	(0.064)		(0.067)	(0.071)		(0.060)	(0.058)		(0.066)	(0.070)
Education above median		0.204***	0.210***		-0.020	-0.023		0.173***	0.168**		0.127*	0.121*		0.041	0.047		0.080+	0.085+		0.185***	0.189***		0.127**	0.148***		0.170***	0.182***
		(0.045)	(0.046)		(0.054)	(0.057)		(0.051)	(0.053)		(0.055)	(0.057)		(0.056)	(0.058)		(0.048)	(0.049)		(0.052)	(0.054)		(0.042)	(0.043)		(0.049)	(0.052)
Wealth above median		0.039	0.018		0.019	-0.041		0.011	0.005		0.045	0.063		0.121*	0.136*		0.095*	0.107*		0.091+	0.061		0.013	0.015		0.098*	0.108*
		(0.046)	(0.047)		(0.053)	(0.057)		(0.048)	(0.051)		(0.054)	(0.057)		(0.055)	(0.057)		(0.048)	(0.050)		(0.050)	(0.052)		(0.042)	(0.044)		(0.048)	(0.051)
Lives with in-laws		-0.092	-0.073		0.055	0.066		0.055	0.059		-0.038	-0.044		0.090	0.072		0.165	0.168		0.107	0.122		0.082	0.094		-0.017	-0.013
		(0.102)	(0.104)		(0.177)	(0.182)		(0.153)	(0.152)		(0.183)	(0.176)		(0.164)	(0.164)		(0.126)	(0.128)		(0.158)	(0.154)		(0.125)	(0.126)		(0.163)	(0.166)
Lives with own parents		0.029	0.041		0.007	-0.010		0.004	-0.001		0.024	0.003		-0.012	-0.002		-0.041	-0.028		0.023	0.028		0.085+	0.082		-0.027	-0.039
		(0.054)	(0.055)		(0.060)	(0.062)		(0.057)	(0.060)		(0.059)	(0.061)		(0.060)	(0.063)		(0.055)	(0.058)		(0.056)	(0.059)		(0.048)	(0.051)		(0.055)	(0.058)
Years as sarpanch		0.011	0.010		0.069***	0.061**		0.047**	0.042**		-0.032+	-0.032+		-0.036*	-0.029		-0.021	-0.015		0.005	0.000		0.026+	0.024		0.022	0.020
		(0.015)	(0.016)		(0.017)	(0.019)		(0.015)	(0.016)		(0.018)	(0.019)		(0.018)	(0.019)		(0.016)	(0.017)		(0.016)	(0.017)		(0.014)	(0.015)		(0.016)	(0.017)
Gender quota		0.025	-0.005		0.065	0.043		-0.112	-0.098		0.012	-0.057		-0.029	-0.041		0.073	0.056		0.038	-0.011		-0.048	-0.063		0.003	-0.026
		(0.089)	(0.095)		(0.087)	(0.094)		(0.087)	(0.092)		(0.079)	(0.090)		(0.086)	(0.094)		(0.089)	(0.093)		(0.083)	(0.095)		(0.083)	(0.088)		(0.083)	(0.087)
SC quota		-0.031	-0.038		-0.092	-0.093		-0.023	-0.020		-0.094	-0.064		-0.051	-0.025		-0.067	-0.048		-0.022	0.002		0.034	0.056		0.054	0.074
		(0.087)	(0.090)		(0.110)	(0.114)		(0.091)	(0.098)		(0.110)	(0.119)		(0.112)	(0.116)		(0.110)	(0.111)		(0.084)	(0.092)		(0.096)	(0.094)		(0.107)	(0.111)
OBC quota		0.002	-0.004		-0.038	-0.042		-0.035	-0.053		0.021	0.047		0.049	0.046		-0.090	-0.080		-0.026	-0.038		-0.037	-0.034		0.060	0.035
		(0.058)	(0.059)		(0.069)	(0.074)		(0.065)	(0.070)		(0.072)	(0.075)		(0.073)	(0.076)		(0.065)	(0.066)		(0.065)	(0.068)		(0.059)	(0.056)		(0.065)	(0.069)
Sarpanch woman:Sarpanch relative present		-0.175	-0.218		-0.227	-0.312		-0.327***	-0.391***		0.181	0.196		0.120	0.063		-0.301*	-0.374*		-0.023	-0.024		-0.320**	-0.290*		0.058	0.075
		(0.133)	(0.149)		(0.366)	(0.341)		(0.094)	(0.096)		(0.305)	(0.311)		(0.350)	(0.281)		(0.134)	(0.150)		(0.308)	(0.315)		(0.114)	(0.123)		(0.318)	(0.321)
Percent of village land held by biggest landholding caste			-0.002			-0.000			-0.001			0.001			0.000			0.001			-0.002			-0.002+			-0.001
			(0.001)			(0.001)			(0.001)			(0.001)			(0.001)			(0.001)			(0.001)			(0.001)			(0.001)
Distance (km) from block center			0.003			0.001			-0.001			-0.002			-0.006*			-0.002			-0.001			0.001			-0.000
			(0.002)			(0.002)			(0.002)			(0.002)			(0.002)			(0.002)			(0.002)			(0.002)			(0.002)
Directly elected			0.061			0.111+			0.038			0.072			-0.038			-0.026			0.066			0.022			0.000
			(0.047)			(0.059)			(0.052)			(0.059)			(0.059)			(0.052)			(0.052)			(0.043)			(0.051)
Num.Obs.	407	370	340	407	370	340	407	370	340	407	370	340	407	370	340	407	370	340	407	370	340	407	370	340	407	370	340
R2	0.001	0.217	0.235	0.001	0.071	0.083	0.001	0.269	0.280	0.001	0.086	0.090	0.000	0.058	0.073	0.000	0.067	0.083	0.000	0.255	0.242	0.001	0.150	0.182	0.003	0.191	0.192

IC6 (H3.6a) table, effect of T2
	DV: Sarpanch loudest			DV: Sarpanch equal status.			DV: Sarpanch confidence speech			DV: Sarpanch engaged decision-making			DV: Sarpanch active listening			DV: Sarpanch consensus building			DV: Sarpanch body language confidence			DV: Sarpanch body language fearlessness			DV: Sarpanch body language interest
	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls
+ p < 0.1, * p < 0.05, p < 0.01, * p < 0.001
Each model is a linear regression of each outcome on an indicator for being assigned to T2 (1) versus not (0). Unadjusted and covariate-adjusted models presented. All regressions use robust standard errors.
(Intercept)	0.687***	0.654***	0.585***	0.458***	0.461***	0.465**	0.536***	0.604***	0.649***	0.417***	0.568***	0.742***	0.531***	0.624***	0.704***	0.734***	0.867***	0.893***	0.578***	0.631***	0.784***	0.771***	0.722***	0.721***	0.641***	0.782***	0.897***
	(0.034)	(0.090)	(0.131)	(0.036)	(0.109)	(0.155)	(0.036)	(0.094)	(0.141)	(0.036)	(0.097)	(0.144)	(0.036)	(0.106)	(0.154)	(0.032)	(0.098)	(0.131)	(0.036)	(0.097)	(0.142)	(0.030)	(0.087)	(0.115)	(0.035)	(0.095)	(0.139)
T2	0.025	-0.024	-0.020	-0.019	-0.023	-0.028	0.070	0.030	0.005	-0.104*	-0.156**	-0.160**	0.050	0.038	0.044	0.003	-0.015	-0.015	0.023	-0.015	-0.005	0.002	-0.002	-0.014	-0.009	-0.041	-0.048
	(0.047)	(0.045)	(0.047)	(0.050)	(0.054)	(0.056)	(0.050)	(0.048)	(0.050)	(0.049)	(0.050)	(0.053)	(0.050)	(0.054)	(0.055)	(0.045)	(0.048)	(0.049)	(0.050)	(0.048)	(0.050)	(0.043)	(0.043)	(0.043)	(0.049)	(0.049)	(0.052)
Sarpanch woman		-0.056	-0.065		-0.170	-0.020		-0.245	-0.304		-0.212	-0.211		-0.052	-0.130		-0.058	-0.095		-0.515**	-0.433+		-0.192	-0.246		-0.192	-0.226
		(0.127)	(0.141)		(0.188)	(0.212)		(0.183)	(0.200)		(0.157)	(0.175)		(0.209)	(0.218)		(0.159)	(0.171)		(0.198)	(0.224)		(0.152)	(0.172)		(0.226)	(0.256)
Sarpanch relative present		0.153*	0.148		0.037	0.030		0.213***	0.181***		0.281	0.216		0.214	0.214		0.312***	0.321***		-0.039	-0.091		0.135**	0.116+		-0.053	-0.056
		(0.077)	(0.096)		(0.286)	(0.300)		(0.049)	(0.054)		(0.281)	(0.282)		(0.272)	(0.233)		(0.055)	(0.075)		(0.238)	(0.226)		(0.044)	(0.069)		(0.247)	(0.257)
Sarpanch SC		0.156	0.186+		-0.136	-0.101		0.211*	0.228*		-0.044	-0.030		0.192+	0.153		0.053	-0.011		0.197*	0.234*		0.075	0.069		0.151	0.150
		(0.100)	(0.103)		(0.131)	(0.131)		(0.101)	(0.108)		(0.107)	(0.114)		(0.115)	(0.117)		(0.119)	(0.117)		(0.093)	(0.094)		(0.099)	(0.098)		(0.136)	(0.140)
Sarpanch OBC		-0.025	-0.010		-0.080	-0.040		0.012	0.004		-0.061	-0.037		-0.115	-0.120		-0.050	-0.067		0.054	0.055		0.074	0.053		0.014	0.001
		(0.063)	(0.067)		(0.078)	(0.082)		(0.071)	(0.076)		(0.077)	(0.081)		(0.077)	(0.081)		(0.064)	(0.065)		(0.073)	(0.078)		(0.063)	(0.062)		(0.072)	(0.079)
Education above median		0.195***	0.233***		0.010	0.006		0.138**	0.122*		0.072	0.068		0.053	0.047		0.075	0.061		0.136**	0.154**		0.121**	0.144**		0.074	0.111*
		(0.047)	(0.048)		(0.058)	(0.060)		(0.051)	(0.054)		(0.055)	(0.057)		(0.057)	(0.058)		(0.048)	(0.049)		(0.052)	(0.055)		(0.044)	(0.046)		(0.053)	(0.056)
Wealth above median		-0.011	-0.019		-0.042	-0.053		-0.017	-0.017		0.069	0.069		0.133*	0.149**		0.064	0.087+		0.056	0.049		0.064	0.079+		-0.011	-0.006
		(0.047)	(0.048)		(0.056)	(0.058)		(0.048)	(0.050)		(0.052)	(0.055)		(0.054)	(0.057)		(0.049)	(0.051)		(0.051)	(0.053)		(0.045)	(0.046)		(0.051)	(0.054)
Lives with in-laws		-0.164+	-0.114		0.200	0.175		-0.021	0.029		-0.094	-0.115		-0.094	-0.066		-0.121	-0.124		0.135	0.122		0.001	0.048		-0.129	-0.080
		(0.094)	(0.098)		(0.168)	(0.185)		(0.164)	(0.169)		(0.142)	(0.152)		(0.183)	(0.190)		(0.125)	(0.136)		(0.186)	(0.203)		(0.123)	(0.132)		(0.209)	(0.231)
Lives with own parents		0.057	0.040		-0.052	-0.071		0.020	0.016		-0.000	-0.015		-0.020	-0.017		-0.083	-0.044		0.045	0.009		0.008	0.032		-0.001	-0.018
		(0.056)	(0.056)		(0.063)	(0.067)		(0.058)	(0.061)		(0.059)	(0.063)		(0.063)	(0.067)		(0.059)	(0.062)		(0.061)	(0.063)		(0.050)	(0.053)		(0.061)	(0.066)
Years as sarpanch		0.026	0.025		0.045*	0.043*		0.047**	0.035*		-0.008	-0.013		-0.028	-0.033+		-0.022	-0.014		0.011	0.006		0.021	0.020		0.002	-0.004
		(0.016)	(0.017)		(0.019)	(0.020)		(0.016)	(0.018)		(0.017)	(0.018)		(0.018)	(0.019)		(0.016)	(0.017)		(0.017)	(0.018)		(0.015)	(0.016)		(0.018)	(0.019)
Gender quota		-0.001	-0.042		-0.066	-0.176+		-0.121	-0.111		0.087	0.074		0.020	0.033		0.124	0.129		0.121	0.028		0.040	0.040		0.071	0.049
		(0.098)	(0.104)		(0.100)	(0.105)		(0.099)	(0.108)		(0.079)	(0.084)		(0.103)	(0.107)		(0.102)	(0.104)		(0.092)	(0.102)		(0.104)	(0.113)		(0.101)	(0.107)
SC quota		-0.097	-0.092		0.086	0.076		-0.255*	-0.261*		0.054	0.040		-0.227*	-0.185		-0.120	-0.069		-0.112	-0.175+		-0.063	-0.073		-0.108	-0.105
		(0.099)	(0.102)		(0.130)	(0.131)		(0.100)	(0.107)		(0.106)	(0.114)		(0.114)	(0.117)		(0.118)	(0.115)		(0.093)	(0.095)		(0.094)	(0.095)		(0.134)	(0.138)
OBC quota		0.030	0.055		0.018	0.020		-0.025	-0.012		-0.028	-0.048		0.103	0.123		0.002	0.032		-0.121+	-0.158*		-0.066	-0.032		-0.029	-0.038
		(0.063)	(0.067)		(0.076)	(0.081)		(0.071)	(0.077)		(0.075)	(0.080)		(0.075)	(0.080)		(0.063)	(0.064)		(0.073)	(0.077)		(0.062)	(0.061)		(0.073)	(0.080)
Sarpanch woman:Sarpanch relative present		-0.377**	-0.399**		-0.258	-0.307		-0.330**	-0.284*		-0.385	-0.297		-0.052	-0.059		-0.383**	-0.401**		-0.191	-0.132		-0.361**	-0.312*		-0.170	-0.130
		(0.133)	(0.144)		(0.304)	(0.317)		(0.110)	(0.119)		(0.295)	(0.297)		(0.293)	(0.256)		(0.125)	(0.132)		(0.260)	(0.250)		(0.129)	(0.140)		(0.272)	(0.284)
Percent of village land held by biggest landholding caste			0.000			-0.000			0.000			-0.001			0.001			0.001			-0.002*			-0.001			-0.001
			(0.001)			(0.001)			(0.001)			(0.001)			(0.001)			(0.001)			(0.001)			(0.001)			(0.001)
Distance (km) from block center			0.000			-0.001			-0.002			-0.004*			-0.004			-0.003			-0.000			0.002			-0.001
			(0.002)			(0.002)			(0.002)			(0.002)			(0.003)			(0.002)			(0.002)			(0.002)			(0.002)
Directly elected			0.050			0.054			0.098+			0.070			-0.069			-0.136*			0.111*			0.026			0.022
			(0.049)			(0.062)			(0.054)			(0.059)			(0.058)			(0.053)			(0.054)			(0.044)			(0.056)
Num.Obs.	390	351	325	390	351	325	390	351	325	390	351	325	390	351	325	390	351	325	390	351	325	390	351	325	390	351	325
R2	0.001	0.206	0.237	0.000	0.044	0.057	0.005	0.250	0.254	0.012	0.123	0.150	0.002	0.069	0.098	0.000	0.048	0.085	0.001	0.207	0.233	0.000	0.132	0.166	0.000	0.117	0.127

IC6b (H3.6b): We additionally expect this effect to be driven by sarpanches from disadvantaged groups.

IC6b is assessed by running the same model, plus an interaction between treatment (bundling T1 and T2 versus control) and a binary indicator of the disadvantaged status of sarpanch (with sarpanches from disadvantaged groups primarily either understood as women, Scheduled castes or both in separate models). We also control for the potential presence of a sarpanch relative during the interview (“the interview” being here understood as the group interview), both separately and in interaction with the gender of the sarpanch. Cluster-corrected standard errors will be used at the administrative block level in both cases.
In additional analyses, we disentangle between T1 and T2. We also include the series of individual and village level controls presented in G2.
The same measures are used as in IC6a.
Deviation 1: We originally planned to use clustered standard errors at the taluka level. This was an error on our part: because the treatment assignment occurs at the level of each village council, we do not cluster standard errors in these analyses.
Precision 1: For the first six measures, we code sarpanches who did not speak at all as not exihibiting the measures of confidence.
Precision 2: For each outcome, we run one regression interacting a female indicator with treatment status, as well as a regression interacting an indicator for SC with treatment status.

IC6b (H3.6b) table, grouping T1/T2 together, Gender disadvantage
	DV: Sarpanch loudest			DV: Sarpanch equal status.			DV: Sarpanch confidence speech			DV: Sarpanch engaged decision-making			DV: Sarpanch active listening			DV: Sarpanch consensus building			DV: Sarpanch body language confidence			DV: Sarpanch body language fearlessness			DV: Sarpanch body language interest
	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls
+ p < 0.1, * p < 0.05, p < 0.01, * p < 0.001
Each model is a linear regression of each outcome on an indicator for being assigned to discussion rules (1) versus not (0). Unadjusted and covariate-adjusted models presented. All regressions use robust standard errors.
(Intercept)	0.899***	0.711***	0.672***	0.478***	0.405***	0.464***	0.841***	0.594***	0.614***	0.580***	0.590***	0.637***	0.580***	0.578***	0.596***	0.754***	0.838***	0.832***	0.841***	0.671***	0.750***	0.899***	0.665***	0.694***	0.884***	0.768***	0.823***
	(0.037)	(0.077)	(0.109)	(0.061)	(0.099)	(0.130)	(0.044)	(0.079)	(0.110)	(0.060)	(0.094)	(0.129)	(0.060)	(0.098)	(0.130)	(0.052)	(0.086)	(0.112)	(0.044)	(0.082)	(0.117)	(0.037)	(0.072)	(0.095)	(0.039)	(0.077)	(0.115)
Discussion rules	-0.001	-0.028	-0.011	0.003	-0.004	-0.031	-0.041	-0.073	-0.081	-0.109	-0.143+	-0.138+	0.074	0.063	0.043	0.025	0.018	-0.012	-0.051	-0.084	-0.056	0.020	0.019	-0.002	-0.100*	-0.113*	-0.106+
	(0.043)	(0.045)	(0.049)	(0.071)	(0.075)	(0.077)	(0.053)	(0.052)	(0.054)	(0.070)	(0.075)	(0.079)	(0.069)	(0.073)	(0.074)	(0.061)	(0.061)	(0.059)	(0.054)	(0.055)	(0.058)	(0.042)	(0.044)	(0.041)	(0.049)	(0.051)	(0.056)
Sarpanch woman	-0.329***	-0.178	-0.178	-0.031	-0.178	-0.133	-0.475***	-0.304*	-0.335*	-0.255***	-0.218	-0.159	-0.076	-0.092	-0.174	-0.030	-0.107	-0.171	-0.410***	-0.446**	-0.339*	-0.199***	-0.079	-0.118	-0.380***	-0.298+	-0.295+
	(0.058)	(0.115)	(0.122)	(0.075)	(0.159)	(0.172)	(0.062)	(0.141)	(0.151)	(0.073)	(0.159)	(0.171)	(0.075)	(0.165)	(0.167)	(0.066)	(0.139)	(0.145)	(0.063)	(0.153)	(0.165)	(0.055)	(0.122)	(0.133)	(0.060)	(0.166)	(0.179)
Discussion rules:Sarpanch woman	0.002	0.025	0.014	-0.060	-0.049	-0.006	0.087	0.120	0.108	0.043	0.074	0.059	-0.096	-0.087	-0.049	-0.064	-0.070	-0.031	0.042	0.091	0.076	-0.044	-0.025	-0.008	0.162*	0.175*	0.163*
	(0.070)	(0.071)	(0.076)	(0.090)	(0.095)	(0.098)	(0.076)	(0.077)	(0.080)	(0.087)	(0.093)	(0.097)	(0.089)	(0.094)	(0.096)	(0.079)	(0.081)	(0.081)	(0.077)	(0.079)	(0.083)	(0.067)	(0.070)	(0.069)	(0.075)	(0.078)	(0.082)
Sarpanch relative present		0.133*	0.120+		-0.104	-0.122		0.192***	0.184***		0.110	0.073		0.225	0.226		0.248***	0.255***		0.010	-0.032		0.107***	0.097+		-0.015	-0.016
		(0.055)	(0.065)		(0.255)	(0.255)		(0.053)	(0.048)		(0.267)	(0.258)		(0.198)	(0.177)		(0.052)	(0.060)		(0.211)	(0.201)		(0.028)	(0.050)		(0.221)	(0.226)
Sarpanch SC		0.150*	0.153*		-0.013	-0.021		0.146+	0.140+		0.051	0.058		0.090	0.072		0.057	0.018		0.145*	0.139+		0.085	0.053		0.121	0.099
		(0.075)	(0.077)		(0.099)	(0.100)		(0.076)	(0.082)		(0.089)	(0.094)		(0.095)	(0.097)		(0.093)	(0.093)		(0.073)	(0.077)		(0.084)	(0.083)		(0.097)	(0.100)
Sarpanch OBC		-0.024	0.002		-0.042	-0.018		0.070	0.064		-0.033	-0.017		-0.070	-0.060		-0.001	-0.019		0.051	0.069		0.079+	0.056		0.001	-0.013
		(0.049)	(0.051)		(0.059)	(0.064)		(0.055)	(0.059)		(0.061)	(0.064)		(0.060)	(0.063)		(0.052)	(0.054)		(0.056)	(0.059)		(0.047)	(0.047)		(0.056)	(0.060)
Education above median		0.186***	0.212***		-0.009	-0.010		0.141***	0.137**		0.087*	0.080+		0.030	0.028		0.052	0.049		0.151***	0.158***		0.128***	0.143***		0.108**	0.134**
		(0.037)	(0.038)		(0.045)	(0.047)		(0.041)	(0.043)		(0.044)	(0.046)		(0.045)	(0.047)		(0.039)	(0.040)		(0.042)	(0.043)		(0.034)	(0.035)		(0.041)	(0.043)
Wealth above median		0.022	0.006		0.012	-0.025		0.040	0.040		0.047	0.054		0.125**	0.143**		0.071+	0.091*		0.096*	0.076+		0.038	0.049		0.050	0.064
		(0.038)	(0.038)		(0.044)	(0.046)		(0.040)	(0.042)		(0.043)	(0.045)		(0.044)	(0.046)		(0.039)	(0.041)		(0.041)	(0.043)		(0.035)	(0.036)		(0.041)	(0.043)
Lives with in-laws		-0.103	-0.056		0.138	0.138		-0.039	-0.006		-0.020	-0.039		0.042	0.084		0.024	0.051		0.059	0.049		0.038	0.054		-0.057	-0.028
		(0.084)	(0.086)		(0.130)	(0.142)		(0.117)	(0.124)		(0.128)	(0.134)		(0.134)	(0.134)		(0.102)	(0.109)		(0.133)	(0.140)		(0.092)	(0.101)		(0.143)	(0.150)
Lives with own parents		0.043	0.051		-0.026	-0.036		0.009	0.014		0.021	0.005		-0.006	0.004		-0.066	-0.040		0.045	0.040		0.070+	0.082+		0.002	-0.003
		(0.044)	(0.044)		(0.049)	(0.052)		(0.045)	(0.048)		(0.047)	(0.050)		(0.049)	(0.052)		(0.046)	(0.049)		(0.047)	(0.049)		(0.040)	(0.042)		(0.046)	(0.049)
Years as sarpanch		0.016	0.018		0.051***	0.046**		0.050***	0.045**		-0.017	-0.017		-0.026+	-0.027+		-0.023+	-0.020		0.005	0.003		0.020+	0.021+		0.010	0.008
		(0.013)	(0.013)		(0.014)	(0.016)		(0.013)	(0.013)		(0.014)	(0.015)		(0.014)	(0.015)		(0.013)	(0.014)		(0.013)	(0.014)		(0.011)	(0.012)		(0.013)	(0.015)
Gender quota		0.017	-0.008		0.014	-0.045		-0.133+	-0.114		0.000	-0.053		-0.024	-0.004		0.061	0.071		0.038	-0.046		-0.087	-0.080		-0.001	-0.025
		(0.076)	(0.082)		(0.075)	(0.082)		(0.074)	(0.081)		(0.067)	(0.077)		(0.075)	(0.080)		(0.076)	(0.081)		(0.072)	(0.080)		(0.070)	(0.076)		(0.073)	(0.078)
SC quota		-0.119	-0.109		-0.020	-0.020		-0.134+	-0.131		-0.086	-0.069		-0.102	-0.088		-0.083	-0.064		-0.104	-0.109		-0.062	-0.061		-0.072	-0.050
		(0.073)	(0.075)		(0.096)	(0.097)		(0.075)	(0.080)		(0.088)	(0.093)		(0.093)	(0.095)		(0.091)	(0.091)		(0.072)	(0.076)		(0.080)	(0.080)		(0.095)	(0.098)
OBC quota		0.012	0.017		-0.028	-0.037		-0.037	-0.040		-0.037	-0.031		0.071	0.071		-0.058	-0.047		-0.083	-0.110+		-0.044	-0.032		0.002	-0.010
		(0.049)	(0.050)		(0.058)	(0.063)		(0.055)	(0.059)		(0.060)	(0.063)		(0.060)	(0.063)		(0.052)	(0.054)		(0.056)	(0.058)		(0.046)	(0.045)		(0.057)	(0.061)
Sarpanch woman:Sarpanch relative present		-0.282**	-0.315**		-0.088	-0.127		-0.338***	-0.368***		-0.133	-0.073		-0.100	-0.136		-0.380***	-0.432***		-0.243	-0.210		-0.282**	-0.294**		-0.190	-0.183
		(0.104)	(0.110)		(0.267)	(0.267)		(0.094)	(0.092)		(0.279)	(0.272)		(0.218)	(0.201)		(0.105)	(0.112)		(0.225)	(0.218)		(0.094)	(0.105)		(0.239)	(0.245)
Percent of village land held by biggest landholding caste			-0.001			-0.001			0.000			-0.000			0.001			0.001			-0.002+			-0.001			-0.001
			(0.001)			(0.001)			(0.001)			(0.001)			(0.001)			(0.001)			(0.001)			(0.001)			(0.001)
Distance (km) from block center			0.001			-0.001			-0.001			-0.002			-0.003+			-0.001			-0.000			0.002+			-0.001
			(0.002)			(0.002)			(0.002)			(0.002)			(0.002)			(0.002)			(0.002)			(0.001)			(0.002)
Directly elected			0.038			0.084+			0.043			0.071			-0.028			-0.066			0.079+			-0.007			-0.013
			(0.038)			(0.048)			(0.043)			(0.047)			(0.047)			(0.043)			(0.043)			(0.035)			(0.043)
Num.Obs.	605	549	506	605	549	506	605	549	506	605	549	506	605	549	506	605	549	506	605	549	506	605	549	506	605	549	506
R2	0.126	0.195	0.218	0.007	0.045	0.053	0.173	0.246	0.251	0.056	0.081	0.090	0.022	0.059	0.070	0.008	0.050	0.068	0.145	0.219	0.224	0.076	0.135	0.163	0.083	0.127	0.134

IC6b (H3.6b) table, effect of T1, Gender disadvantage
	DV: Sarpanch loudest			DV: Sarpanch equal status.			DV: Sarpanch confidence speech			DV: Sarpanch engaged decision-making			DV: Sarpanch active listening			DV: Sarpanch consensus building			DV: Sarpanch body language confidence			DV: Sarpanch body language fearlessness			DV: Sarpanch body language interest
	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls
+ p < 0.1, * p < 0.05, p < 0.01, * p < 0.001
Each model is a linear regression of each outcome on an indicator for being assigned to T1 (1) versus not (0). Unadjusted and covariate-adjusted models presented. All regressions use robust standard errors.
(Intercept)	0.899***	0.725***	0.731***	0.478***	0.329**	0.348*	0.841***	0.609***	0.710***	0.580***	0.586***	0.551***	0.580***	0.618***	0.732***	0.754***	0.798***	0.816***	0.841***	0.682***	0.802***	0.899***	0.641***	0.782***	0.884***	0.698***	0.794***
	(0.037)	(0.091)	(0.130)	(0.061)	(0.112)	(0.156)	(0.044)	(0.089)	(0.130)	(0.060)	(0.106)	(0.151)	(0.060)	(0.111)	(0.154)	(0.052)	(0.099)	(0.133)	(0.044)	(0.094)	(0.138)	(0.037)	(0.082)	(0.112)	(0.039)	(0.087)	(0.133)
T1	0.014	0.001	0.017	0.027	0.025	0.025	-0.038	-0.072	-0.079	-0.085	-0.135	-0.133	0.058	0.036	-0.008	0.027	0.022	-0.026	-0.027	-0.068	-0.052	0.036	0.033	-0.004	-0.038	-0.061	-0.065
	(0.047)	(0.049)	(0.052)	(0.080)	(0.084)	(0.086)	(0.061)	(0.060)	(0.061)	(0.080)	(0.085)	(0.089)	(0.078)	(0.082)	(0.084)	(0.068)	(0.067)	(0.067)	(0.060)	(0.062)	(0.065)	(0.045)	(0.047)	(0.046)	(0.054)	(0.058)	(0.063)
Sarpanch woman	-0.329***	-0.200	-0.168	-0.031	-0.123	-0.114	-0.475***	-0.406*	-0.413*	-0.255***	-0.214	-0.148	-0.076	-0.142	-0.152	-0.030	-0.232	-0.252	-0.410***	-0.502**	-0.452*	-0.199***	-0.135	-0.167	-0.380***	-0.347+	-0.322
	(0.058)	(0.131)	(0.135)	(0.075)	(0.200)	(0.206)	(0.062)	(0.172)	(0.176)	(0.073)	(0.209)	(0.208)	(0.075)	(0.187)	(0.193)	(0.066)	(0.159)	(0.163)	(0.063)	(0.178)	(0.178)	(0.055)	(0.152)	(0.154)	(0.060)	(0.189)	(0.196)
T1:Sarpanch woman	-0.002	-0.009	-0.010	-0.111	-0.117	-0.116	0.068	0.075	0.076	0.075	0.129	0.126	-0.094	-0.077	0.002	-0.081	-0.085	-0.021	-0.008	0.036	0.055	-0.041	-0.038	0.003	0.114	0.129	0.138
	(0.079)	(0.081)	(0.086)	(0.102)	(0.108)	(0.112)	(0.087)	(0.088)	(0.091)	(0.100)	(0.106)	(0.111)	(0.101)	(0.107)	(0.111)	(0.090)	(0.092)	(0.094)	(0.087)	(0.090)	(0.096)	(0.074)	(0.077)	(0.078)	(0.084)	(0.088)	(0.093)
Sarpanch relative present		0.052	0.018		0.134	0.142		0.122**	0.101+		-0.234	-0.220		0.066	0.061		0.221*	0.221*		-0.211	-0.236		0.079+	0.002		-0.270	-0.300
		(0.085)	(0.109)		(0.375)	(0.347)		(0.037)	(0.058)		(0.280)	(0.287)		(0.328)	(0.257)		(0.093)	(0.102)		(0.299)	(0.306)		(0.044)	(0.059)		(0.311)	(0.314)
Sarpanch SC		0.096	0.095		0.065	0.056		0.058	0.034		0.129	0.136		0.030	-0.006		0.027	-0.019		0.057	0.029		0.036	-0.018		0.033	-0.001
		(0.089)	(0.092)		(0.114)	(0.118)		(0.093)	(0.100)		(0.110)	(0.119)		(0.115)	(0.118)		(0.112)	(0.113)		(0.086)	(0.095)		(0.102)	(0.099)		(0.112)	(0.115)
Sarpanch OBC		-0.054	-0.009		-0.071	-0.042		0.058	0.051		-0.019	-0.009		-0.076	-0.068		-0.025	-0.041		0.003	0.027		0.061	0.040		-0.042	-0.048
		(0.059)	(0.061)		(0.071)	(0.077)		(0.067)	(0.072)		(0.073)	(0.076)		(0.073)	(0.077)		(0.064)	(0.064)		(0.067)	(0.071)		(0.060)	(0.058)		(0.066)	(0.070)
Education above median		0.204***	0.210***		-0.018	-0.022		0.172***	0.167**		0.125*	0.120*		0.043	0.047		0.082+	0.085+		0.184***	0.188***		0.128**	0.148***		0.168***	0.181***
		(0.045)	(0.046)		(0.054)	(0.057)		(0.051)	(0.053)		(0.055)	(0.057)		(0.056)	(0.058)		(0.047)	(0.049)		(0.052)	(0.054)		(0.042)	(0.043)		(0.050)	(0.052)
Wealth above median		0.039	0.017		0.014	-0.044		0.014	0.007		0.049	0.067		0.118*	0.136*		0.091+	0.106*		0.093+	0.063		0.012	0.015		0.103*	0.112*
		(0.046)	(0.047)		(0.053)	(0.057)		(0.048)	(0.051)		(0.054)	(0.057)		(0.055)	(0.058)		(0.048)	(0.050)		(0.050)	(0.052)		(0.042)	(0.044)		(0.048)	(0.051)
Lives with in-laws		-0.092	-0.073		0.057	0.067		0.054	0.058		-0.040	-0.046		0.090	0.072		0.166	0.168		0.107	0.121		0.083	0.094		-0.018	-0.015
		(0.102)	(0.104)		(0.178)	(0.183)		(0.152)	(0.151)		(0.187)	(0.179)		(0.162)	(0.164)		(0.125)	(0.128)		(0.158)	(0.153)		(0.125)	(0.126)		(0.166)	(0.168)
Lives with own parents		0.029	0.041		0.010	-0.008		0.002	-0.002		0.021	0.001		-0.010	-0.002		-0.039	-0.027		0.022	0.027		0.086+	0.082		-0.030	-0.041
		(0.054)	(0.055)		(0.060)	(0.063)		(0.057)	(0.060)		(0.059)	(0.062)		(0.060)	(0.063)		(0.055)	(0.059)		(0.056)	(0.059)		(0.049)	(0.051)		(0.055)	(0.058)
Years as sarpanch		0.011	0.010		0.070***	0.062**		0.046**	0.041*		-0.033+	-0.032+		-0.035+	-0.029		-0.020	-0.015		0.005	0.000		0.026+	0.024		0.020	0.019
		(0.015)	(0.016)		(0.017)	(0.019)		(0.015)	(0.016)		(0.018)	(0.019)		(0.018)	(0.019)		(0.016)	(0.017)		(0.016)	(0.017)		(0.014)	(0.015)		(0.016)	(0.017)
Gender quota		0.024	-0.005		0.060	0.040		-0.109	-0.095		0.018	-0.053		-0.032	-0.041		0.069	0.055		0.039	-0.009		-0.050	-0.063		0.008	-0.022
		(0.089)	(0.095)		(0.087)	(0.093)		(0.087)	(0.093)		(0.078)	(0.088)		(0.086)	(0.094)		(0.089)	(0.094)		(0.083)	(0.094)		(0.084)	(0.089)		(0.082)	(0.086)
SC quota		-0.030	-0.037		-0.082	-0.086		-0.030	-0.024		-0.105	-0.072		-0.044	-0.025		-0.059	-0.046		-0.025	-0.002		0.037	0.055		0.043	0.066
		(0.087)	(0.089)		(0.110)	(0.114)		(0.091)	(0.098)		(0.110)	(0.119)		(0.113)	(0.116)		(0.108)	(0.110)		(0.085)	(0.093)		(0.096)	(0.094)		(0.109)	(0.113)
OBC quota		0.003	-0.004		-0.033	-0.039		-0.038	-0.056		0.016	0.043		0.052	0.046		-0.086	-0.080		-0.027	-0.040		-0.036	-0.034		0.055	0.031
		(0.058)	(0.059)		(0.069)	(0.073)		(0.065)	(0.069)		(0.072)	(0.075)		(0.073)	(0.077)		(0.065)	(0.066)		(0.065)	(0.068)		(0.059)	(0.056)		(0.065)	(0.069)
Sarpanch woman:Sarpanch relative present		-0.176	-0.219		-0.243	-0.330		-0.317***	-0.379***		0.198	0.215		0.110	0.064		-0.312*	-0.377*		-0.019	-0.016		-0.325**	-0.289*		0.075	0.096
		(0.134)	(0.150)		(0.388)	(0.361)		(0.093)	(0.094)		(0.295)	(0.305)		(0.345)	(0.282)		(0.142)	(0.152)		(0.311)	(0.321)		(0.115)	(0.124)		(0.328)	(0.334)
Percent of village land held by biggest landholding caste			-0.002			-0.000			-0.001			0.000			0.000			0.001			-0.002			-0.002+			-0.001
			(0.001)			(0.001)			(0.001)			(0.001)			(0.001)			(0.001)			(0.001)			(0.001)			(0.001)
Distance (km) from block center			0.003			0.001			-0.001			-0.002			-0.006*			-0.002			-0.001			0.001			-0.000
			(0.002)			(0.002)			(0.002)			(0.002)			(0.002)			(0.002)			(0.002)			(0.002)			(0.002)
Directly elected			0.061			0.112+			0.037			0.071			-0.038			-0.026			0.066			0.022			-0.001
			(0.047)			(0.059)			(0.052)			(0.059)			(0.059)			(0.052)			(0.053)			(0.043)			(0.052)
Num.Obs.	407	370	340	407	370	340	407	370	340	407	370	340	407	370	340	407	370	340	407	370	340	407	370	340	407	370	340
R2	0.126	0.218	0.235	0.012	0.074	0.086	0.186	0.271	0.281	0.047	0.090	0.093	0.017	0.060	0.073	0.009	0.070	0.083	0.167	0.256	0.242	0.070	0.150	0.182	0.115	0.195	0.197

IC6b (H3.6b) table, effect of T2, gender disadvantage
	DV: Sarpanch loudest			DV: Sarpanch equal status.			DV: Sarpanch confidence speech			DV: Sarpanch engaged decision-making			DV: Sarpanch active listening			DV: Sarpanch consensus building			DV: Sarpanch body language confidence			DV: Sarpanch body language fearlessness			DV: Sarpanch body language interest
	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls
+ p < 0.1, * p < 0.05, p < 0.01, * p < 0.001
Each model is a linear regression of each outcome on an indicator for being assigned to T2 (1) versus not (0). Unadjusted and covariate-adjusted models presented. All regressions use robust standard errors.
(Intercept)	0.899***	0.672***	0.600***	0.478***	0.463***	0.502**	0.841***	0.660***	0.710***	0.580***	0.571***	0.734***	0.580***	0.595***	0.671***	0.754***	0.850***	0.881***	0.841***	0.681***	0.828***	0.899***	0.714***	0.712***	0.884***	0.855***	0.978***
	(0.037)	(0.093)	(0.137)	(0.061)	(0.115)	(0.161)	(0.044)	(0.094)	(0.140)	(0.060)	(0.106)	(0.156)	(0.060)	(0.112)	(0.160)	(0.052)	(0.100)	(0.133)	(0.044)	(0.097)	(0.145)	(0.037)	(0.089)	(0.119)	(0.039)	(0.091)	(0.142)
T2	-0.016	-0.056	-0.041	-0.021	-0.026	-0.081	-0.043	-0.068	-0.084	-0.133+	-0.161+	-0.148+	0.091	0.088	0.092	0.023	0.014	0.002	-0.075	-0.105	-0.070	0.006	0.012	-0.001	-0.161**	-0.169**	-0.165*
	(0.050)	(0.054)	(0.059)	(0.080)	(0.085)	(0.088)	(0.061)	(0.060)	(0.063)	(0.079)	(0.084)	(0.089)	(0.077)	(0.082)	(0.083)	(0.068)	(0.071)	(0.068)	(0.062)	(0.065)	(0.068)	(0.048)	(0.049)	(0.046)	(0.060)	(0.063)	(0.068)
Sarpanch woman	-0.329***	-0.085	-0.081	-0.031	-0.173	-0.061	-0.475***	-0.331+	-0.373+	-0.255***	-0.216	-0.202	-0.076	-0.008	-0.093	-0.030	-0.033	-0.081	-0.410***	-0.593**	-0.483*	-0.199***	-0.179	-0.235	-0.380***	-0.303	-0.318
	(0.058)	(0.132)	(0.145)	(0.075)	(0.199)	(0.220)	(0.062)	(0.186)	(0.201)	(0.073)	(0.170)	(0.184)	(0.075)	(0.215)	(0.222)	(0.066)	(0.170)	(0.178)	(0.063)	(0.199)	(0.223)	(0.055)	(0.160)	(0.177)	(0.060)	(0.233)	(0.266)
T2:Sarpanch woman	0.004	0.056	0.035	-0.003	0.005	0.090	0.110	0.168+	0.150	0.000	0.008	-0.020	-0.095	-0.086	-0.081	-0.045	-0.050	-0.030	0.096	0.153	0.108	-0.051	-0.024	-0.023	0.205*	0.218*	0.198+
	(0.083)	(0.085)	(0.090)	(0.104)	(0.110)	(0.113)	(0.089)	(0.090)	(0.094)	(0.098)	(0.104)	(0.109)	(0.102)	(0.107)	(0.110)	(0.091)	(0.094)	(0.094)	(0.091)	(0.094)	(0.098)	(0.079)	(0.082)	(0.082)	(0.090)	(0.095)	(0.101)
Sarpanch relative present		0.151+	0.146		0.037	0.025		0.208**	0.174**		0.281	0.217		0.216	0.218		0.313***	0.323***		-0.044	-0.097		0.136**	0.117+		-0.060	-0.066
		(0.082)	(0.099)		(0.286)	(0.285)		(0.064)	(0.054)		(0.283)	(0.280)		(0.263)	(0.225)		(0.052)	(0.073)		(0.257)	(0.242)		(0.042)	(0.066)		(0.274)	(0.284)
Sarpanch SC		0.158	0.187+		-0.136	-0.098		0.217*	0.233*		-0.043	-0.031		0.189	0.151		0.051	-0.012		0.203*	0.237*		0.074	0.068		0.159	0.156
		(0.100)	(0.103)		(0.131)	(0.132)		(0.099)	(0.107)		(0.107)	(0.114)		(0.116)	(0.118)		(0.119)	(0.117)		(0.092)	(0.093)		(0.099)	(0.098)		(0.139)	(0.142)
Sarpanch OBC		-0.026	-0.011		-0.080	-0.041		0.010	0.002		-0.061	-0.036		-0.114	-0.119		-0.050	-0.067		0.053	0.053		0.074	0.054		0.012	-0.002
		(0.063)	(0.067)		(0.078)	(0.082)		(0.070)	(0.076)		(0.077)	(0.081)		(0.077)	(0.081)		(0.065)	(0.066)		(0.072)	(0.078)		(0.063)	(0.063)		(0.072)	(0.078)
Education above median		0.196***	0.234***		0.010	0.007		0.140**	0.124*		0.072	0.068		0.052	0.046		0.074	0.061		0.138**	0.156**		0.121**	0.144**		0.076	0.113*
		(0.047)	(0.048)		(0.058)	(0.060)		(0.051)	(0.054)		(0.055)	(0.057)		(0.057)	(0.058)		(0.048)	(0.049)		(0.052)	(0.055)		(0.044)	(0.046)		(0.053)	(0.056)
Wealth above median		-0.009	-0.018		-0.042	-0.051		-0.012	-0.014		0.069	0.069		0.131*	0.147**		0.063	0.086+		0.060	0.051		0.063	0.078+		-0.005	-0.002
		(0.047)	(0.048)		(0.056)	(0.058)		(0.048)	(0.051)		(0.052)	(0.055)		(0.054)	(0.057)		(0.049)	(0.051)		(0.051)	(0.053)		(0.045)	(0.046)		(0.051)	(0.054)
Lives with in-laws		-0.164+	-0.116		0.200	0.172		-0.022	0.024		-0.095	-0.114		-0.094	-0.064		-0.121	-0.123		0.134	0.118		0.001	0.049		-0.131	-0.087
		(0.096)	(0.100)		(0.168)	(0.185)		(0.162)	(0.168)		(0.142)	(0.151)		(0.184)	(0.190)		(0.126)	(0.137)		(0.184)	(0.201)		(0.123)	(0.132)		(0.212)	(0.235)
Lives with own parents		0.057	0.040		-0.052	-0.071		0.020	0.016		-0.000	-0.015		-0.019	-0.017		-0.083	-0.044		0.045	0.010		0.008	0.032		-0.002	-0.017
		(0.056)	(0.056)		(0.063)	(0.067)		(0.058)	(0.061)		(0.059)	(0.063)		(0.063)	(0.067)		(0.059)	(0.062)		(0.061)	(0.063)		(0.050)	(0.053)		(0.061)	(0.065)
Years as sarpanch		0.026	0.025		0.045*	0.043*		0.047**	0.035*		-0.008	-0.013		-0.029	-0.033+		-0.022	-0.014		0.012	0.006		0.021	0.020		0.003	-0.003
		(0.016)	(0.017)		(0.019)	(0.020)		(0.016)	(0.018)		(0.017)	(0.019)		(0.018)	(0.019)		(0.016)	(0.017)		(0.017)	(0.018)		(0.015)	(0.016)		(0.017)	(0.019)
Gender quota		-0.002	-0.043		-0.066	-0.180+		-0.123	-0.117		0.086	0.074		0.021	0.037		0.124	0.130		0.119	0.023		0.040	0.041		0.068	0.040
		(0.098)	(0.104)		(0.100)	(0.104)		(0.098)	(0.108)		(0.079)	(0.085)		(0.103)	(0.107)		(0.103)	(0.104)		(0.091)	(0.101)		(0.105)	(0.113)		(0.099)	(0.106)
SC quota		-0.100	-0.094		0.086	0.071		-0.265**	-0.269*		0.053	0.041		-0.222+	-0.181		-0.117	-0.067		-0.121	-0.181+		-0.062	-0.071		-0.121	-0.115
		(0.099)	(0.102)		(0.130)	(0.131)		(0.098)	(0.105)		(0.107)	(0.115)		(0.116)	(0.118)		(0.118)	(0.116)		(0.092)	(0.094)		(0.094)	(0.095)		(0.136)	(0.140)
OBC quota		0.029	0.055		0.018	0.019		-0.029	-0.013		-0.028	-0.048		0.106	0.123		0.003	0.032		-0.125+	-0.159*		-0.065	-0.032		-0.035	-0.040
		(0.063)	(0.067)		(0.077)	(0.081)		(0.071)	(0.077)		(0.075)	(0.080)		(0.075)	(0.080)		(0.063)	(0.064)		(0.073)	(0.077)		(0.062)	(0.061)		(0.073)	(0.079)
Sarpanch woman:Sarpanch relative present		-0.374**	-0.397**		-0.257	-0.301		-0.321**	-0.273*		-0.384	-0.298		-0.057	-0.064		-0.386**	-0.403**		-0.182	-0.125		-0.363**	-0.313*		-0.157	-0.116
		(0.137)	(0.146)		(0.304)	(0.303)		(0.116)	(0.118)		(0.297)	(0.296)		(0.285)	(0.249)		(0.123)	(0.131)		(0.277)	(0.265)		(0.129)	(0.140)		(0.297)	(0.309)
Percent of village land held by biggest landholding caste			0.000			-0.000			0.000			-0.001			0.001			0.001			-0.002*			-0.001			-0.002
			(0.001)			(0.001)			(0.001)			(0.001)			(0.001)			(0.001)			(0.001)			(0.001)			(0.001)
Distance (km) from block center			0.000			-0.001			-0.003			-0.004*			-0.004			-0.003			-0.000			0.002			-0.001
			(0.002)			(0.002)			(0.002)			(0.002)			(0.002)			(0.002)			(0.002)			(0.002)			(0.002)
Directly elected			0.050			0.054			0.099+			0.070			-0.069			-0.136*			0.112*			0.026			0.023
			(0.050)			(0.061)			(0.054)			(0.059)			(0.059)			(0.053)			(0.054)			(0.044)			(0.056)
Num.Obs.	390	351	325	390	351	325	390	351	325	390	351	325	390	351	325	390	351	325	390	351	325	390	351	325	390	351	325
R2	0.123	0.207	0.237	0.001	0.044	0.059	0.178	0.257	0.259	0.079	0.123	0.150	0.020	0.071	0.099	0.004	0.049	0.085	0.131	0.213	0.236	0.071	0.132	0.166	0.088	0.129	0.137

IC6b (H3.6b) table, grouping T1/T2 together, Caste disadvantage
	DV: Sarpanch loudest			DV: Sarpanch equal status.			DV: Sarpanch confidence speech			DV: Sarpanch engaged decision-making			DV: Sarpanch active listening			DV: Sarpanch consensus building			DV: Sarpanch body language confidence			DV: Sarpanch body language fearlessness			DV: Sarpanch body language interest
	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls
+ p < 0.1, * p < 0.05, p < 0.01, * p < 0.001
Each model is a linear regression of each outcome on an indicator for being assigned to discussion rules (1) versus not (0). Unadjusted and covariate-adjusted models presented. All regressions use robust standard errors.
(Intercept)	0.647***	0.667***	0.630***	0.459***	0.417***	0.453***	0.549***	0.535***	0.551***	0.376***	0.510***	0.559***	0.534***	0.614***	0.612***	0.759***	0.887***	0.876***	0.534***	0.602***	0.678***	0.752***	0.656***	0.669***	0.624***	0.682***	0.734***
	(0.042)	(0.077)	(0.103)	(0.043)	(0.089)	(0.123)	(0.043)	(0.079)	(0.113)	(0.042)	(0.083)	(0.119)	(0.043)	(0.089)	(0.122)	(0.037)	(0.082)	(0.109)	(0.043)	(0.080)	(0.113)	(0.038)	(0.071)	(0.091)	(0.042)	(0.079)	(0.113)
Discussion rules	0.076	0.034	0.040	-0.013	-0.018	-0.017	0.042	0.004	-0.008	-0.007	-0.034	-0.043	0.021	0.016	0.025	-0.033	-0.047	-0.064	0.057	0.009	0.029	0.051	0.033	0.030	0.026	-0.002	-0.003
	(0.050)	(0.045)	(0.047)	(0.053)	(0.053)	(0.056)	(0.053)	(0.047)	(0.051)	(0.051)	(0.052)	(0.055)	(0.053)	(0.054)	(0.056)	(0.046)	(0.047)	(0.048)	(0.053)	(0.048)	(0.052)	(0.045)	(0.045)	(0.045)	(0.051)	(0.048)	(0.051)
Sarpanch SC	0.133+	0.268**	0.257**	-0.001	0.030	0.022	-0.040	0.152	0.156	0.133+	0.211+	0.199+	-0.008	0.109	0.101	-0.081	0.001	-0.062	0.144+	0.235*	0.232*	0.062	0.159	0.143	0.054	0.127	0.103
	(0.069)	(0.096)	(0.098)	(0.078)	(0.125)	(0.126)	(0.079)	(0.099)	(0.103)	(0.078)	(0.113)	(0.118)	(0.079)	(0.115)	(0.118)	(0.072)	(0.112)	(0.113)	(0.075)	(0.093)	(0.098)	(0.063)	(0.099)	(0.099)	(0.074)	(0.117)	(0.122)
Discussion rules:Sarpanch SC	-0.140+	-0.168*	-0.148+	-0.043	-0.059	-0.062	0.033	-0.013	-0.024	-0.173+	-0.230*	-0.203*	0.016	-0.024	-0.042	0.085	0.082	0.115	-0.151+	-0.131	-0.134	-0.105	-0.104	-0.128	-0.003	-0.015	-0.008
	(0.083)	(0.085)	(0.087)	(0.094)	(0.104)	(0.107)	(0.094)	(0.089)	(0.092)	(0.093)	(0.097)	(0.101)	(0.094)	(0.101)	(0.104)	(0.086)	(0.091)	(0.092)	(0.091)	(0.089)	(0.092)	(0.077)	(0.078)	(0.079)	(0.089)	(0.093)	(0.097)
Sarpanch woman		-0.162	-0.170		-0.214	-0.138		-0.219+	-0.261+		-0.168	-0.122		-0.154	-0.209		-0.157	-0.191		-0.382*	-0.289+		-0.097	-0.126		-0.174	-0.182
		(0.106)	(0.114)		(0.143)	(0.159)		(0.132)	(0.144)		(0.148)	(0.160)		(0.151)	(0.152)		(0.124)	(0.132)		(0.150)	(0.163)		(0.108)	(0.123)		(0.154)	(0.164)
Sarpanch relative present		0.125+	0.114		-0.111	-0.125		0.201***	0.193***		0.102	0.070		0.216	0.219		0.248***	0.257***		0.009	-0.031		0.099**	0.091		-0.003	-0.001
		(0.069)	(0.079)		(0.246)	(0.252)		(0.045)	(0.043)		(0.259)	(0.248)		(0.209)	(0.186)		(0.047)	(0.061)		(0.197)	(0.187)		(0.036)	(0.059)		(0.203)	(0.209)
Sarpanch OBC		-0.023	0.003		-0.042	-0.018		0.070	0.065		-0.032	-0.016		-0.070	-0.060		-0.001	-0.019		0.051	0.070		0.079+	0.057		0.002	-0.011
		(0.049)	(0.051)		(0.059)	(0.064)		(0.055)	(0.059)		(0.061)	(0.064)		(0.060)	(0.063)		(0.052)	(0.054)		(0.056)	(0.059)		(0.047)	(0.047)		(0.057)	(0.060)
Education above median		0.186***	0.212***		-0.009	-0.011		0.141***	0.137**		0.087*	0.079+		0.030	0.028		0.052	0.049		0.150***	0.158***		0.128***	0.143***		0.109**	0.134**
		(0.037)	(0.038)		(0.045)	(0.047)		(0.041)	(0.043)		(0.044)	(0.045)		(0.045)	(0.047)		(0.039)	(0.040)		(0.042)	(0.043)		(0.034)	(0.035)		(0.041)	(0.043)
Wealth above median		0.020	0.003		0.013	-0.025		0.036	0.037		0.044	0.049		0.128**	0.143**		0.073+	0.094*		0.093*	0.072+		0.038	0.047		0.045	0.060
		(0.038)	(0.038)		(0.044)	(0.046)		(0.040)	(0.042)		(0.043)	(0.045)		(0.044)	(0.046)		(0.039)	(0.041)		(0.041)	(0.043)		(0.035)	(0.036)		(0.041)	(0.043)
Lives with in-laws		-0.101	-0.056		0.138	0.138		-0.037	-0.005		-0.018	-0.039		0.042	0.083		0.022	0.051		0.061	0.050		0.038	0.054		-0.055	-0.026
		(0.086)	(0.088)		(0.131)	(0.142)		(0.119)	(0.126)		(0.133)	(0.138)		(0.135)	(0.135)		(0.102)	(0.108)		(0.138)	(0.146)		(0.095)	(0.106)		(0.142)	(0.148)
Lives with own parents		0.043	0.050		-0.027	-0.036		0.010	0.014		0.021	0.004		-0.008	0.003		-0.067	-0.039		0.046	0.040		0.069+	0.081+		0.005	-0.002
		(0.044)	(0.044)		(0.049)	(0.052)		(0.045)	(0.049)		(0.047)	(0.049)		(0.049)	(0.052)		(0.046)	(0.049)		(0.047)	(0.049)		(0.040)	(0.042)		(0.047)	(0.050)
Years as sarpanch		0.018	0.019		0.052***	0.046**		0.050***	0.045***		-0.015	-0.016		-0.026+	-0.027+		-0.024+	-0.021		0.006	0.004		0.021+	0.021+		0.010	0.009
		(0.013)	(0.013)		(0.014)	(0.016)		(0.013)	(0.013)		(0.014)	(0.015)		(0.014)	(0.015)		(0.013)	(0.014)		(0.013)	(0.014)		(0.011)	(0.012)		(0.014)	(0.015)
Gender quota		0.018	-0.005		0.015	-0.043		-0.135+	-0.113		0.000	-0.048		-0.022	-0.003		0.062	0.068		0.038	-0.043		-0.086	-0.077		-0.004	-0.026
		(0.076)	(0.081)		(0.075)	(0.082)		(0.075)	(0.081)		(0.067)	(0.077)		(0.075)	(0.080)		(0.076)	(0.080)		(0.072)	(0.080)		(0.070)	(0.076)		(0.074)	(0.078)
SC quota		-0.120	-0.110		-0.024	-0.021		-0.127+	-0.127		-0.084	-0.069		-0.107	-0.091		-0.086	-0.064		-0.100	-0.108		-0.064	-0.063		-0.062	-0.043
		(0.074)	(0.076)		(0.096)	(0.097)		(0.076)	(0.081)		(0.088)	(0.093)		(0.092)	(0.095)		(0.092)	(0.092)		(0.072)	(0.076)		(0.081)	(0.081)		(0.093)	(0.097)
OBC quota		0.008	0.013		-0.030	-0.039		-0.034	-0.039		-0.040	-0.036		0.068	0.069		-0.058	-0.044		-0.083	-0.113+		-0.047	-0.036		0.007	-0.007
		(0.049)	(0.050)		(0.058)	(0.063)		(0.055)	(0.059)		(0.060)	(0.063)		(0.060)	(0.063)		(0.052)	(0.054)		(0.055)	(0.058)		(0.046)	(0.045)		(0.057)	(0.061)
Sarpanch woman:Sarpanch relative present		-0.268*	-0.301*		-0.078	-0.120		-0.347***	-0.376***		-0.116	-0.058		-0.090	-0.127		-0.382***	-0.441***		-0.238	-0.203		-0.270**	-0.280*		-0.203	-0.199
		(0.110)	(0.118)		(0.259)	(0.265)		(0.091)	(0.090)		(0.272)	(0.264)		(0.227)	(0.208)		(0.104)	(0.113)		(0.212)	(0.205)		(0.097)	(0.110)		(0.223)	(0.230)
Percent of village land held by biggest landholding caste			-0.001			-0.001			0.000			-0.000			0.001			0.001			-0.001			-0.001			-0.001
			(0.001)			(0.001)			(0.001)			(0.001)			(0.001)			(0.001)			(0.001)			(0.001)			(0.001)
Distance (km) from block center			0.001			-0.001			-0.001			-0.002			-0.003+			-0.002			-0.000			0.002+			-0.000
			(0.001)			(0.002)			(0.002)			(0.002)			(0.002)			(0.002)			(0.002)			(0.001)			(0.002)
Directly elected			0.036			0.084+			0.043			0.068			-0.028			-0.065			0.077+			-0.009			-0.013
			(0.038)			(0.048)			(0.043)			(0.046)			(0.047)			(0.043)			(0.043)			(0.035)			(0.043)
Num.Obs.	603	549	506	603	549	506	603	549	506	603	549	506	603	549	506	603	549	506	603	549	506	603	549	506	603	549	506
R2	0.007	0.201	0.222	0.002	0.045	0.053	0.003	0.244	0.249	0.010	0.090	0.097	0.001	0.057	0.070	0.002	0.050	0.071	0.006	0.220	0.226	0.004	0.137	0.168	0.003	0.121	0.129

IC6b (H3.6b) table, effect of T1, Caste disadvantage
	DV: Sarpanch loudest			DV: Sarpanch equal status.			DV: Sarpanch confidence speech			DV: Sarpanch engaged decision-making			DV: Sarpanch active listening			DV: Sarpanch consensus building			DV: Sarpanch body language confidence			DV: Sarpanch body language fearlessness			DV: Sarpanch body language interest
	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls
+ p < 0.1, * p < 0.05, p < 0.01, * p < 0.001
Each model is a linear regression of each outcome on an indicator for being assigned to T1 (1) versus not (0). Unadjusted and covariate-adjusted models presented. All regressions use robust standard errors.
(Intercept)	0.647***	0.706***	0.707***	0.459***	0.375***	0.397**	0.549***	0.579***	0.673***	0.376***	0.499***	0.455**	0.534***	0.645***	0.719***	0.759***	0.851***	0.853***	0.534***	0.636***	0.736***	0.752***	0.648***	0.768***	0.624***	0.642***	0.724***
	(0.042)	(0.090)	(0.122)	(0.043)	(0.103)	(0.148)	(0.043)	(0.089)	(0.131)	(0.042)	(0.098)	(0.142)	(0.043)	(0.103)	(0.146)	(0.037)	(0.095)	(0.128)	(0.043)	(0.090)	(0.133)	(0.038)	(0.080)	(0.107)	(0.042)	(0.086)	(0.131)
T1	0.076	0.035	0.053	-0.040	-0.042	-0.034	0.025	-0.030	-0.033	0.029	0.003	-0.006	-0.000	-0.001	0.011	-0.050	-0.060	-0.079	0.061	0.009	0.039	0.045	0.024	0.016	0.058	0.022	0.023
	(0.056)	(0.050)	(0.052)	(0.059)	(0.060)	(0.063)	(0.059)	(0.053)	(0.057)	(0.058)	(0.060)	(0.064)	(0.060)	(0.062)	(0.066)	(0.053)	(0.054)	(0.055)	(0.059)	(0.054)	(0.058)	(0.050)	(0.049)	(0.050)	(0.057)	(0.052)	(0.056)
Sarpanch SC	0.133+	0.183+	0.184+	-0.001	0.078	0.076	-0.040	0.050	0.037	0.133+	0.257+	0.246+	-0.008	0.051	0.031	-0.081	-0.038	-0.103	0.144+	0.175+	0.151	0.062	0.069	0.022	0.054	0.041	0.010
	(0.069)	(0.104)	(0.107)	(0.078)	(0.138)	(0.143)	(0.079)	(0.109)	(0.115)	(0.078)	(0.131)	(0.139)	(0.079)	(0.129)	(0.134)	(0.072)	(0.127)	(0.129)	(0.075)	(0.101)	(0.110)	(0.063)	(0.111)	(0.108)	(0.074)	(0.131)	(0.135)
T1:Sarpanch SC	-0.129	-0.146	-0.151	0.006	-0.015	-0.033	0.027	0.010	-0.004	-0.174	-0.223*	-0.186	0.020	-0.032	-0.064	0.114	0.116	0.144	-0.208*	-0.202*	-0.207+	-0.056	-0.053	-0.068	-0.009	-0.021	-0.019
	(0.095)	(0.098)	(0.101)	(0.108)	(0.119)	(0.122)	(0.108)	(0.101)	(0.104)	(0.106)	(0.112)	(0.119)	(0.108)	(0.116)	(0.119)	(0.097)	(0.103)	(0.106)	(0.105)	(0.101)	(0.106)	(0.087)	(0.088)	(0.090)	(0.100)	(0.107)	(0.112)
Sarpanch woman		-0.197	-0.169		-0.190	-0.179		-0.362*	-0.369*		-0.127	-0.070		-0.185	-0.148		-0.288*	-0.269+		-0.470**	-0.414*		-0.155	-0.163		-0.271	-0.243
		(0.127)	(0.134)		(0.191)	(0.198)		(0.165)	(0.170)		(0.201)	(0.199)		(0.176)	(0.182)		(0.145)	(0.149)		(0.178)	(0.180)		(0.142)	(0.144)		(0.176)	(0.181)
Sarpanch relative present		0.034	0.003		0.115	0.119		0.134**	0.114+		-0.241	-0.214		0.050	0.056		0.222**	0.229*		-0.229	-0.244		0.067	-0.004		-0.253	-0.277
		(0.106)	(0.133)		(0.352)	(0.327)		(0.041)	(0.061)		(0.265)	(0.270)		(0.337)	(0.267)		(0.074)	(0.100)		(0.269)	(0.270)		(0.050)	(0.064)		(0.298)	(0.300)
Sarpanch OBC		-0.054	-0.008		-0.071	-0.043		0.059	0.052		-0.019	-0.007		-0.076	-0.068		-0.025	-0.042		0.003	0.028		0.061	0.040		-0.041	-0.047
		(0.059)	(0.061)		(0.071)	(0.077)		(0.067)	(0.072)		(0.074)	(0.076)		(0.073)	(0.077)		(0.064)	(0.064)		(0.066)	(0.071)		(0.060)	(0.058)		(0.066)	(0.070)
Education above median		0.203***	0.210***		-0.020	-0.023		0.173***	0.168**		0.127*	0.121*		0.041	0.047		0.081+	0.085+		0.184***	0.189***		0.127**	0.148***		0.170***	0.182***
		(0.045)	(0.046)		(0.054)	(0.057)		(0.051)	(0.053)		(0.055)	(0.057)		(0.056)	(0.058)		(0.048)	(0.049)		(0.051)	(0.053)		(0.042)	(0.043)		(0.049)	(0.052)
Wealth above median		0.039	0.015		0.019	-0.041		0.011	0.005		0.044	0.060		0.120*	0.135*		0.095*	0.109*		0.091+	0.058		0.013	0.014		0.098*	0.107*
		(0.046)	(0.047)		(0.053)	(0.057)		(0.048)	(0.051)		(0.054)	(0.057)		(0.055)	(0.058)		(0.048)	(0.050)		(0.050)	(0.052)		(0.042)	(0.044)		(0.048)	(0.051)
Lives with in-laws		-0.092	-0.074		0.055	0.066		0.055	0.059		-0.039	-0.044		0.090	0.072		0.165	0.168		0.106	0.121		0.082	0.094		-0.017	-0.013
		(0.105)	(0.108)		(0.178)	(0.183)		(0.153)	(0.153)		(0.188)	(0.179)		(0.164)	(0.165)		(0.124)	(0.125)		(0.167)	(0.163)		(0.127)	(0.129)		(0.165)	(0.167)
Lives with own parents		0.027	0.038		0.007	-0.010		0.004	-0.001		0.021	-0.000		-0.012	-0.003		-0.040	-0.025		0.021	0.024		0.085+	0.081		-0.027	-0.039
		(0.054)	(0.055)		(0.060)	(0.063)		(0.057)	(0.060)		(0.058)	(0.061)		(0.060)	(0.063)		(0.055)	(0.059)		(0.056)	(0.059)		(0.048)	(0.051)		(0.055)	(0.058)
Years as sarpanch		0.012	0.010		0.069***	0.062**		0.047**	0.042**		-0.031+	-0.031+		-0.036+	-0.029		-0.022	-0.016		0.007	0.001		0.026+	0.024+		0.022	0.020
		(0.015)	(0.016)		(0.017)	(0.019)		(0.015)	(0.016)		(0.018)	(0.019)		(0.018)	(0.019)		(0.016)	(0.017)		(0.016)	(0.017)		(0.014)	(0.015)		(0.016)	(0.017)
Gender quota		0.018	-0.008		0.064	0.043		-0.112	-0.098		0.003	-0.061		-0.030	-0.043		0.078	0.059		0.029	-0.015		-0.051	-0.064		0.002	-0.027
		(0.088)	(0.094)		(0.087)	(0.094)		(0.088)	(0.092)		(0.079)	(0.089)		(0.086)	(0.094)		(0.089)	(0.093)		(0.082)	(0.093)		(0.083)	(0.089)		(0.083)	(0.087)
SC quota		-0.041	-0.049		-0.093	-0.095		-0.023	-0.020		-0.110	-0.078		-0.053	-0.030		-0.058	-0.037		-0.036	-0.014		0.030	0.050		0.053	0.073
		(0.088)	(0.090)		(0.111)	(0.115)		(0.092)	(0.098)		(0.112)	(0.120)		(0.111)	(0.115)		(0.111)	(0.112)		(0.085)	(0.091)		(0.097)	(0.094)		(0.108)	(0.112)
OBC quota		-0.001	-0.009		-0.038	-0.043		-0.035	-0.053		0.015	0.041		0.048	0.043		-0.087	-0.075		-0.031	-0.045		-0.039	-0.036		0.060	0.035
		(0.058)	(0.059)		(0.069)	(0.074)		(0.065)	(0.070)		(0.073)	(0.076)		(0.073)	(0.077)		(0.066)	(0.066)		(0.064)	(0.067)		(0.059)	(0.056)		(0.066)	(0.069)
Sarpanch woman:Sarpanch relative present		-0.154	-0.199		-0.225	-0.308		-0.328***	-0.391***		0.213	0.219		0.125	0.072		-0.317*	-0.392*		0.005	0.002		-0.313**	-0.281*		0.061	0.077
		(0.147)	(0.168)		(0.367)	(0.342)		(0.095)	(0.097)		(0.282)	(0.290)		(0.354)	(0.290)		(0.132)	(0.152)		(0.284)	(0.287)		(0.117)	(0.127)		(0.317)	(0.320)
Percent of village land held by biggest landholding caste			-0.002			-0.000			-0.001			0.001			0.000			0.001			-0.002			-0.002+			-0.001
			(0.001)			(0.001)			(0.001)			(0.001)			(0.001)			(0.001)			(0.001)			(0.001)			(0.001)
Distance (km) from block center			0.003			0.001			-0.001			-0.002			-0.006*			-0.002			-0.001			0.001			-0.000
			(0.002)			(0.002)			(0.002)			(0.002)			(0.002)			(0.002)			(0.002)			(0.002)			(0.002)
Directly elected			0.059			0.111+			0.038			0.071			-0.038			-0.025			0.065			0.022			0.000
			(0.047)			(0.059)			(0.052)			(0.058)			(0.059)			(0.052)			(0.052)			(0.043)			(0.051)
Num.Obs.	406	370	340	406	370	340	406	370	340	406	370	340	406	370	340	406	370	340	406	370	340	406	370	340	406	370	340
R2	0.010	0.222	0.240	0.001	0.071	0.083	0.002	0.269	0.280	0.009	0.096	0.097	0.000	0.058	0.073	0.004	0.071	0.089	0.010	0.263	0.251	0.003	0.151	0.184	0.006	0.191	0.192

IC6b (H3.6b) table, effect of T2, Caste disadvantage
	DV: Sarpanch loudest			DV: Sarpanch equal status.			DV: Sarpanch confidence speech			DV: Sarpanch engaged decision-making			DV: Sarpanch active listening			DV: Sarpanch consensus building			DV: Sarpanch body language confidence			DV: Sarpanch body language fearlessness			DV: Sarpanch body language interest
	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls
+ p < 0.1, * p < 0.05, p < 0.01, * p < 0.001
Each model is a linear regression of each outcome on an indicator for being assigned to T2 (1) versus not (0). Unadjusted and covariate-adjusted models presented. All regressions use robust standard errors.
(Intercept)	0.647***	0.625***	0.558***	0.459***	0.445***	0.447**	0.549***	0.601***	0.643***	0.376***	0.531***	0.696***	0.534***	0.621***	0.699***	0.759***	0.874***	0.911***	0.534***	0.619***	0.771***	0.752***	0.697***	0.683***	0.624***	0.783***	0.903***
	(0.042)	(0.093)	(0.132)	(0.043)	(0.111)	(0.158)	(0.043)	(0.096)	(0.144)	(0.042)	(0.099)	(0.148)	(0.043)	(0.107)	(0.156)	(0.037)	(0.098)	(0.132)	(0.043)	(0.098)	(0.146)	(0.038)	(0.088)	(0.116)	(0.042)	(0.098)	(0.142)
T2	0.076	0.030	0.018	0.018	0.007	-0.003	0.062	0.038	0.013	-0.051	-0.085	-0.095	0.046	0.043	0.051	-0.013	-0.029	-0.040	0.053	0.007	0.013	0.058	0.045	0.039	-0.013	-0.042	-0.056
	(0.058)	(0.053)	(0.056)	(0.062)	(0.065)	(0.068)	(0.061)	(0.057)	(0.060)	(0.059)	(0.060)	(0.064)	(0.062)	(0.064)	(0.066)	(0.054)	(0.057)	(0.058)	(0.062)	(0.058)	(0.061)	(0.051)	(0.052)	(0.053)	(0.061)	(0.060)	(0.063)
Sarpanch SC	0.133+	0.243*	0.246*	-0.001	-0.087	-0.063	-0.040	0.222+	0.241*	0.133+	0.071	0.070	-0.008	0.201	0.164	-0.081	0.031	-0.050	0.144+	0.232*	0.263*	0.062	0.150	0.152	0.054	0.149	0.138
	(0.069)	(0.114)	(0.116)	(0.078)	(0.143)	(0.145)	(0.079)	(0.116)	(0.122)	(0.078)	(0.120)	(0.129)	(0.079)	(0.128)	(0.130)	(0.072)	(0.130)	(0.128)	(0.075)	(0.105)	(0.108)	(0.063)	(0.111)	(0.111)	(0.074)	(0.146)	(0.151)
T2:Sarpanch SC	-0.151	-0.181+	-0.126	-0.095	-0.103	-0.082	0.035	-0.024	-0.027	-0.162	-0.240*	-0.214+	0.007	-0.018	-0.024	0.054	0.045	0.082	-0.098	-0.074	-0.062	-0.153+	-0.158+	-0.176+	0.011	0.004	0.027
	(0.096)	(0.100)	(0.101)	(0.107)	(0.119)	(0.124)	(0.107)	(0.104)	(0.107)	(0.104)	(0.107)	(0.111)	(0.108)	(0.116)	(0.119)	(0.098)	(0.103)	(0.103)	(0.104)	(0.105)	(0.108)	(0.090)	(0.091)	(0.093)	(0.103)	(0.110)	(0.115)
Sarpanch woman		-0.071	-0.075		-0.178	-0.026		-0.247	-0.306		-0.231	-0.227		-0.053	-0.132		-0.054	-0.089		-0.521*	-0.437+		-0.204	-0.259		-0.191	-0.224
		(0.127)	(0.143)		(0.188)	(0.211)		(0.184)	(0.201)		(0.171)	(0.188)		(0.209)	(0.218)		(0.159)	(0.170)		(0.202)	(0.228)		(0.154)	(0.177)		(0.227)	(0.255)
Sarpanch relative present		0.147	0.146		0.034	0.028		0.212***	0.181**		0.273	0.212		0.213	0.214		0.313***	0.323***		-0.041	-0.092		0.130*	0.112		-0.053	-0.055
		(0.092)	(0.108)		(0.280)	(0.296)		(0.050)	(0.055)		(0.275)	(0.272)		(0.275)	(0.237)		(0.053)	(0.080)		(0.235)	(0.224)		(0.056)	(0.085)		(0.248)	(0.259)
Sarpanch OBC		-0.023	-0.008		-0.079	-0.039		0.012	0.005		-0.058	-0.033		-0.115	-0.120		-0.051	-0.069		0.055	0.056		0.076	0.056		0.014	0.000
		(0.063)	(0.067)		(0.078)	(0.083)		(0.071)	(0.076)		(0.077)	(0.081)		(0.077)	(0.081)		(0.064)	(0.065)		(0.073)	(0.078)		(0.063)	(0.062)		(0.073)	(0.079)
Education above median		0.195***	0.233***		0.010	0.006		0.138**	0.122*		0.072	0.068		0.053	0.047		0.075	0.061		0.137**	0.154**		0.121**	0.144**		0.074	0.111*
		(0.047)	(0.048)		(0.058)	(0.060)		(0.051)	(0.054)		(0.054)	(0.056)		(0.057)	(0.059)		(0.049)	(0.049)		(0.052)	(0.054)		(0.043)	(0.045)		(0.053)	(0.056)
Wealth above median		-0.013	-0.022		-0.044	-0.055		-0.017	-0.018		0.066	0.063		0.133*	0.148**		0.065	0.089+		0.055	0.047		0.062	0.074		-0.011	-0.005
		(0.047)	(0.048)		(0.056)	(0.059)		(0.048)	(0.050)		(0.052)	(0.055)		(0.054)	(0.057)		(0.049)	(0.051)		(0.051)	(0.053)		(0.045)	(0.046)		(0.051)	(0.054)
Lives with in-laws		-0.160	-0.114		0.203	0.175		-0.021	0.029		-0.090	-0.114		-0.094	-0.066		-0.122	-0.124		0.137	0.122		0.004	0.049		-0.129	-0.080
		(0.097)	(0.102)		(0.167)	(0.184)		(0.165)	(0.170)		(0.158)	(0.166)		(0.183)	(0.191)		(0.125)	(0.135)		(0.190)	(0.207)		(0.129)	(0.142)		(0.210)	(0.230)
Lives with own parents		0.058	0.041		-0.052	-0.070		0.020	0.016		0.001	-0.013		-0.020	-0.016		-0.083	-0.045		0.046	0.010		0.009	0.033		-0.001	-0.018
		(0.056)	(0.056)		(0.063)	(0.067)		(0.058)	(0.061)		(0.058)	(0.062)		(0.063)	(0.067)		(0.059)	(0.062)		(0.061)	(0.063)		(0.050)	(0.053)		(0.061)	(0.066)
Years as sarpanch		0.027+	0.026		0.046*	0.044*		0.047**	0.035*		-0.006	-0.011		-0.028	-0.033+		-0.022	-0.015		0.012	0.006		0.023	0.021		0.002	-0.004
		(0.016)	(0.017)		(0.019)	(0.020)		(0.016)	(0.018)		(0.017)	(0.018)		(0.018)	(0.019)		(0.016)	(0.017)		(0.017)	(0.018)		(0.015)	(0.016)		(0.018)	(0.019)
Gender quota		0.011	-0.032		-0.059	-0.169		-0.119	-0.109		0.103	0.090		0.021	0.035		0.121	0.122		0.126	0.032		0.050	0.054		0.071	0.047
		(0.097)	(0.103)		(0.101)	(0.106)		(0.099)	(0.108)		(0.077)	(0.082)		(0.103)	(0.107)		(0.103)	(0.104)		(0.092)	(0.101)		(0.103)	(0.111)		(0.102)	(0.108)
SC quota		-0.086	-0.083		0.093	0.082		-0.253*	-0.259*		0.069	0.055		-0.226*	-0.184		-0.123	-0.074		-0.107	-0.171+		-0.054	-0.060		-0.108	-0.107
		(0.103)	(0.105)		(0.129)	(0.130)		(0.101)	(0.108)		(0.105)	(0.112)		(0.115)	(0.117)		(0.118)	(0.116)		(0.095)	(0.096)		(0.099)	(0.100)		(0.134)	(0.137)
OBC quota		0.024	0.050		0.014	0.016		-0.025	-0.013		-0.036	-0.056		0.103	0.122		0.003	0.035		-0.124+	-0.160*		-0.071	-0.039		-0.029	-0.037
		(0.063)	(0.067)		(0.076)	(0.082)		(0.071)	(0.077)		(0.075)	(0.081)		(0.076)	(0.080)		(0.064)	(0.065)		(0.073)	(0.077)		(0.062)	(0.061)		(0.074)	(0.080)
Sarpanch woman:Sarpanch relative present		-0.357*	-0.384*		-0.246	-0.297		-0.328**	-0.281*		-0.359	-0.271		-0.050	-0.056		-0.388**	-0.411**		-0.183	-0.125		-0.344*	-0.290+		-0.170	-0.134
		(0.140)	(0.152)		(0.299)	(0.314)		(0.111)	(0.121)		(0.290)	(0.290)		(0.295)	(0.258)		(0.124)	(0.136)		(0.258)	(0.250)		(0.134)	(0.149)		(0.274)	(0.287)
Percent of village land held by biggest landholding caste			0.000			-0.000			0.000			-0.001			0.001			0.001			-0.002+			-0.001			-0.002
			(0.001)			(0.001)			(0.001)			(0.001)			(0.001)			(0.001)			(0.001)			(0.001)			(0.001)
Distance (km) from block center			0.001			-0.001			-0.002			-0.004+			-0.004			-0.003			0.000			0.002			-0.001
			(0.002)			(0.002)			(0.002)			(0.002)			(0.003)			(0.002)			(0.002)			(0.002)			(0.002)
Directly elected			0.047			0.051			0.097+			0.064			-0.069			-0.133*			0.109*			0.021			0.023
			(0.049)			(0.062)			(0.054)			(0.058)			(0.059)			(0.053)			(0.054)			(0.044)			(0.056)
Num.Obs.	389	351	325	389	351	325	389	351	325	389	351	325	389	351	325	389	351	325	389	351	325	389	351	325	389	351	325
R2	0.010	0.214	0.240	0.005	0.046	0.058	0.006	0.250	0.254	0.020	0.136	0.160	0.002	0.069	0.098	0.004	0.049	0.086	0.011	0.208	0.234	0.008	0.139	0.175	0.003	0.117	0.127

IC7a (H3.7a): Public indicators of respect for the sarpanch will be higher for those assigned to any treatment (T1 or T2, considered jointly) versus not (Control).

IC7a is assessed by regressing indicators of the upa sarpanch and gram sevak behavior during the group meeting on a binary indicator of treatment (bundling T1 and T2 versus control), and on the set of demographic controls defined in our online appendix. We also control for the potential presence of a sarpanch relative during the interview (“the interview” being here understood as the group interview), both separately and in interaction with the gender of the sarpanch. Cluster-corrected standard errors will be used at the administrative block level in all models.
In additional analyses, we disentangle between T1 and T2. We also include the series of individual and village level controls presented in G2.
We use the following outcome measure:
- Q76: IF YES to sarpanch speaks above: Was the sarpanch interrupted at any point while she/he spoke? Code centrality as 1 if answer is 0 (No), and 0 if answer is 1 (Yes).
Deviation 1: We originally planned to use clustered standard errors at the taluka level. This was an error on our part: because the treatment assignment occurs at the level of each village council, we do not cluster standard errors in these analyses.

IC7a (H3.7a) table
	Impact of T1/T2 combined			Impact of T1			Impact of t2
	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls
+ p < 0.1, * p < 0.05, p < 0.01, * p < 0.001
Each model is a linear regression of each outcome on an indicator for being assigned to the relevant treatment condition(s) (1) versus not (0). Unadjusted and covariate-adjusted models presented. All regressions use robust standard errors.
(Intercept)	0.751***	0.667***	0.713***	0.751***	0.688***	0.806***	0.751***	0.695***	0.686***
	(0.033)	(0.076)	(0.109)	(0.033)	(0.088)	(0.126)	(0.033)	(0.099)	(0.149)
Discussion rules	0.063	0.040	0.042
	(0.039)	(0.040)	(0.042)
Sarpanch woman		-0.286*	-0.310*		-0.190	-0.143		-0.377*	-0.434*
		(0.122)	(0.129)		(0.145)	(0.129)		(0.174)	(0.190)
Sarpanch relative present		-0.069	-0.071		0.120**	0.140		-0.135	-0.146
		(0.211)	(0.222)		(0.041)	(0.095)		(0.257)	(0.262)
Sarpanch SC		0.129	0.120		0.046	0.018		0.180	0.192
		(0.090)	(0.090)		(0.076)	(0.078)		(0.135)	(0.137)
Sarpanch OBC		0.013	0.000		0.012	0.005		0.040	0.029
		(0.050)	(0.053)		(0.055)	(0.061)		(0.068)	(0.072)
Education above median		0.008	0.001		0.028	0.032		-0.011	-0.018
		(0.036)	(0.038)		(0.043)	(0.046)		(0.048)	(0.051)
Wealth above median		-0.043	-0.036		-0.039	-0.040		-0.064	-0.058
		(0.036)	(0.038)		(0.044)	(0.046)		(0.048)	(0.051)
Lives with in-laws		0.155	0.175		0.069	0.054		0.234	0.291+
		(0.112)	(0.112)		(0.135)	(0.115)		(0.154)	(0.159)
Lives with own parents		0.071+	0.069+		0.068	0.070		0.058	0.060
		(0.037)	(0.041)		(0.045)	(0.049)		(0.051)	(0.054)
Years as sarpanch		0.047***	0.039**		0.052***	0.046**		0.045**	0.033+
		(0.012)	(0.013)		(0.015)	(0.016)		(0.016)	(0.017)
Gender quota		0.062	0.064		0.005	-0.029		0.057	0.071
		(0.069)	(0.077)		(0.078)	(0.085)		(0.096)	(0.103)
SC quota		-0.160+	-0.153+		-0.096	-0.081		-0.210	-0.225
		(0.090)	(0.090)		(0.075)	(0.078)		(0.137)	(0.141)
OBC quota		-0.005	0.024		-0.072	-0.060		0.023	0.057
		(0.049)	(0.052)		(0.057)	(0.061)		(0.066)	(0.070)
Sarpanch woman:Sarpanch relative present		-0.084	-0.085		-0.234*	-0.256+		-0.051	-0.077
		(0.234)	(0.246)		(0.119)	(0.147)		(0.296)	(0.305)
Percent of village land held by biggest landholding caste			0.000			-0.000			-0.000
			(0.001)			(0.001)			(0.001)
Distance (km) from block center			-0.003+			-0.005**			-0.000
			(0.002)			(0.002)			(0.002)
Directly elected			0.042			0.029			0.085+
			(0.037)			(0.047)			(0.049)
T1				0.086*	0.060	0.071
				(0.042)	(0.043)	(0.046)
T2							0.039	0.014	0.008
							(0.045)	(0.048)	(0.051)
Num.Obs.	552	504	466	366	336	310	355	322	299
R2	0.005	0.087	0.096	0.011	0.114	0.134	0.002	0.094	0.108

IC7b (H3.7b): We additionally expect this effect to be driven by sarpanches from disadvantaged groups.

IC7b is assessed by running the same model (as above), plus an interaction between treatment (bundling T1 and T2 versus control) and a binary indicator of the disadvantaged status of sarpanch (with sarpanches from disadvantaged groups primarily either understood as women, Scheduled castes or both in separate models). We also control for the potential presence of a sarpanch relative during the interview (“the interview” being here understood as the group interview), both separately and in interaction with the gender of the sarpanch. Cluster-corrected standard errors will be used at the administrative block level in all models.
In additional analyses, we disentangle between T1 and T2. We also include the series of individual and village level controls presented in G2.
The same measures are used as in IC7a.
Deviation 1: We originally planned to use clustered standard errors at the taluka level. This was an error on our part: because the treatment assignment occurs at the level of each village council, we do not cluster standard errors in these analyses.
Precision 1: For each outcome, we run one regression interacting a female indicator with treatment status, as well as a regression interacting an indicator for SC with treatment status.

IC7b (H3.7b) table, Gender disadvantage
	Impact of T1/T2 combined			Impact of T1			Impact of t2
	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls
+ p < 0.1, * p < 0.05, p < 0.01, * p < 0.001
Each model is a linear regression of each outcome on an indicator for being assigned to the relevant treatment condition(s) (1) versus not (0). Unadjusted and covariate-adjusted models presented. All regressions use robust standard errors.
(Intercept)	0.894***	0.723***	0.776***	0.894***	0.724***	0.861***	0.894***	0.739***	0.736***
	(0.038)	(0.084)	(0.118)	(0.038)	(0.097)	(0.138)	(0.038)	(0.103)	(0.154)
Discussion rules	-0.041	-0.031	-0.030
	(0.046)	(0.052)	(0.057)
Sarpanch woman	-0.234***	-0.374**	-0.395**	-0.234***	-0.242	-0.209	-0.234***	-0.450*	-0.495*
	(0.060)	(0.136)	(0.143)	(0.060)	(0.156)	(0.141)	(0.060)	(0.183)	(0.198)
Discussion rules:Sarpanch woman	0.161*	0.123	0.123
	(0.072)	(0.077)	(0.082)
Sarpanch relative present		-0.078	-0.083		0.107**	0.121		-0.138	-0.152
		(0.205)	(0.216)		(0.038)	(0.092)		(0.245)	(0.251)
Sarpanch SC		0.130	0.120		0.047	0.017		0.181	0.193
		(0.089)	(0.090)		(0.074)	(0.076)		(0.134)	(0.136)
Sarpanch OBC		0.010	-0.003		0.010	0.002		0.036	0.024
		(0.049)	(0.053)		(0.055)	(0.061)		(0.068)	(0.072)
Education above median		0.008	0.002		0.028	0.032		-0.008	-0.015
		(0.036)	(0.038)		(0.043)	(0.045)		(0.048)	(0.051)
Wealth above median		-0.040	-0.034		-0.036	-0.038		-0.061	-0.056
		(0.036)	(0.038)		(0.044)	(0.046)		(0.048)	(0.051)
Lives with in-laws		0.153	0.173		0.067	0.051		0.233	0.286+
		(0.112)	(0.112)		(0.133)	(0.112)		(0.156)	(0.163)
Lives with own parents		0.069+	0.068+		0.065	0.068		0.059	0.061
		(0.037)	(0.040)		(0.044)	(0.049)		(0.051)	(0.054)
Years as sarpanch		0.047***	0.039**		0.051**	0.045**		0.045**	0.033+
		(0.012)	(0.013)		(0.015)	(0.016)		(0.016)	(0.017)
Gender quota		0.066	0.065		0.011	-0.023		0.057	0.065
		(0.069)	(0.078)		(0.079)	(0.086)		(0.098)	(0.106)
SC quota		-0.165+	-0.157+		-0.102	-0.086		-0.215	-0.229
		(0.089)	(0.089)		(0.073)	(0.075)		(0.136)	(0.139)
OBC quota		-0.007	0.022		-0.075	-0.063		0.020	0.056
		(0.049)	(0.052)		(0.057)	(0.061)		(0.066)	(0.069)
Sarpanch woman:Sarpanch relative present		-0.072	-0.072		-0.223+	-0.240+		-0.035	-0.063
		(0.229)	(0.241)		(0.118)	(0.145)		(0.287)	(0.296)
Percent of village land held by biggest landholding caste			0.000			-0.000			-0.000
			(0.001)			(0.001)			(0.001)
Distance (km) from block center			-0.003+			-0.006**			-0.000
			(0.002)			(0.002)			(0.002)
Directly elected			0.042			0.027			0.087+
			(0.037)			(0.047)			(0.048)
T1				-0.006	0.009	0.005
				(0.051)	(0.055)	(0.061)
T1:Sarpanch woman				0.142+	0.090	0.116
				(0.079)	(0.084)	(0.090)
T2							-0.075	-0.064	-0.066
							(0.055)	(0.062)	(0.068)
T2:Sarpanch woman							0.175*	0.140	0.131
							(0.085)	(0.092)	(0.098)
Num.Obs.	552	504	466	366	336	310	355	322	299
R2	0.035	0.092	0.100	0.056	0.117	0.138	0.040	0.101	0.113

IC7b (H3.7b) table, Caste disadvantage
	Impact of T1/T2 combined			Impact of T1			Impact of t2
	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls
+ p < 0.1, * p < 0.05, p < 0.01, * p < 0.001
Each model is a linear regression of each outcome on an indicator for being assigned to the relevant treatment condition(s) (1) versus not (0). Unadjusted and covariate-adjusted models presented. All regressions use robust standard errors.
(Intercept)	0.780***	0.680***	0.733***	0.780***	0.701***	0.826***	0.780***	0.705***	0.705***
	(0.038)	(0.078)	(0.111)	(0.038)	(0.089)	(0.127)	(0.038)	(0.102)	(0.151)
Discussion rules	0.037	0.020	0.015
	(0.045)	(0.045)	(0.049)
Sarpanch SC	-0.093	0.081	0.056	-0.093	-0.004	-0.045	-0.093	0.148	0.144
	(0.076)	(0.110)	(0.112)	(0.076)	(0.101)	(0.107)	(0.076)	(0.139)	(0.142)
Discussion rules:Sarpanch SC	0.085	0.070	0.093
	(0.087)	(0.093)	(0.097)
Sarpanch woman		-0.286*	-0.309*		-0.194	-0.146		-0.372*	-0.429*
		(0.120)	(0.127)		(0.141)	(0.123)		(0.173)	(0.186)
Sarpanch relative present		-0.065	-0.067		0.130**	0.150		-0.133	-0.145
		(0.217)	(0.231)		(0.050)	(0.108)		(0.266)	(0.276)
Sarpanch OBC		0.014	0.001		0.013	0.006		0.041	0.029
		(0.049)	(0.053)		(0.055)	(0.061)		(0.068)	(0.072)
Education above median		0.008	0.002		0.029	0.032		-0.011	-0.018
		(0.036)	(0.038)		(0.043)	(0.045)		(0.048)	(0.051)
Wealth above median		-0.043	-0.035		-0.039	-0.039		-0.063	-0.056
		(0.036)	(0.038)		(0.044)	(0.046)		(0.048)	(0.051)
Lives with in-laws		0.155	0.176		0.069	0.055		0.233	0.291+
		(0.110)	(0.109)		(0.129)	(0.107)		(0.152)	(0.155)
Lives with own parents		0.071+	0.070+		0.069	0.072		0.059	0.060
		(0.037)	(0.040)		(0.044)	(0.049)		(0.051)	(0.054)
Years as sarpanch		0.047***	0.039**		0.051***	0.046**		0.044**	0.032+
		(0.012)	(0.013)		(0.015)	(0.016)		(0.016)	(0.017)
Gender quota		0.062	0.062		0.010	-0.026		0.053	0.064
		(0.069)	(0.077)		(0.079)	(0.085)		(0.096)	(0.103)
SC quota		-0.160+	-0.153+		-0.092	-0.074		-0.214	-0.232
		(0.090)	(0.091)		(0.076)	(0.080)		(0.139)	(0.142)
OBC quota		-0.004	0.026		-0.071	-0.057		0.025	0.060
		(0.049)	(0.052)		(0.057)	(0.062)		(0.067)	(0.070)
Sarpanch woman:Sarpanch relative present		-0.089	-0.093		-0.247*	-0.272+		-0.053	-0.084
		(0.240)	(0.255)		(0.123)	(0.156)		(0.304)	(0.317)
Percent of village land held by biggest landholding caste			0.000			-0.000			-0.000
			(0.001)			(0.001)			(0.001)
Distance (km) from block center			-0.003+			-0.005**			-0.000
			(0.002)			(0.002)			(0.002)
Directly elected			0.044			0.030			0.089+
			(0.038)			(0.047)			(0.049)
T1				0.060	0.036	0.040
				(0.050)	(0.049)	(0.053)
T1:Sarpanch SC				0.084	0.089	0.111
				(0.096)	(0.102)	(0.106)
T2							0.010	-0.006	-0.024
							(0.054)	(0.055)	(0.059)
T2:Sarpanch SC							0.091	0.066	0.102
							(0.099)	(0.108)	(0.113)
Num.Obs.	550	504	466	365	336	310	354	322	299
R2	0.009	0.089	0.098	0.016	0.116	0.137	0.007	0.095	0.111

Second group of hypotheses: Private Preferences

This measures the extent to which the private preferences of other actors have changed.

P1 (H3.8): The distance between post-treatment, private preferences of sarpanch and post-treatment, private preferences of other group members should be lower for groups assigned to any treatment.

Assessed by regressing each of these four metrics (in this case, comparing the sarpanch’s post-treatment preferences to the post-treatment preferences of other actors) on treatment status (T1 or T2 versus Control). We also control for the potential presence of a sarpanch relative during the interview (“the interview” being here understood as the group interview), both separately and in interaction with the gender of the sarpanch. We do this exercise twice. P1a tests the difference between the sarpanch’s post-treatment allocation and the upa-sarpanch’s post-treatment allocation. P2b tests the difference between the sarpanch’s post-treatment allocation and the gram sevak’s post-treatment allocation.
In additional analyses, we disentangle between T1 and T2.
We run bivariate regressions, clustering at the taluka level.
We also include the series of individual and village level demographic controls presented in G2.
We use of the following four metrics:
- Metric 1: Earth’s Mover distance’s measure of budgetary congruence, as defined in Lupu et al 2017.
- Metric 2: Average distance in allocation across the 5 priorities. Practically, we calculate the difference in %age allocated to each item across the two distributions; we subsequently average this difference (that is, we compute the total difference across the five items and divide this total difference by 5).
- Metric 3: Binary measure coding whether the two distributions identify the same priority as the “top” priority (that is, the priority allocated the largest %age.)
- Metric 4: Continuous measure identifying the distance between the %age of funds allocated to the “top” priority across the two allocations.
Deviation 1: We originally planned to use clustered standard errors at the taluka level. This was an error on our part: because the treatment assignment occurs at the level of each village council, we do not cluster standard errors in these analyses.
Deviation 2: We originally planed to use the Earth Mover’s Distance (Lupu et al 2017) as our primary outcome. After discussion with the developers of the measure, we decided not to use this measure for two reasons: 1) it does not provide additional information relative to simpler measures, like metrics 2-4, and 2) the package to implement the measure is also not operational.
Deviation 3: We are unable to implement metric 3 and 4 as planned because we did not ask respondents what their top priority was post-budget discussion. We thus cannot identify top priorities among the respondents who did not have one budget category with the highest amount of allocation.

P1a (H3.8a) table, Sarpanch and upa sarpanch post-private preferences, average difference
	Impact of T1/T2 combined			Impact of T1			Impact of T2
	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls
+ p < 0.1, * p < 0.05, p < 0.01, * p < 0.001
Each model is a linear regression of each outcome on an indicator for being assigned to the relevant treatment condition(s) (1) versus not (0). Unadjusted and covariate-adjusted models presented. All regressions use robust standard errors.
(Intercept)	9.594***	8.519***	11.592***	9.594***	7.275***	10.149***	9.594***	8.372***	10.936***
	(0.610)	(1.510)	(2.255)	(0.610)	(1.800)	(2.638)	(0.610)	(1.784)	(2.810)
Discussion rules	-0.131	0.000	-0.050
	(0.748)	(0.801)	(0.818)
Sarpanch woman		2.452	1.729		0.207	-0.370		4.178	3.371
		(1.979)	(2.196)		(1.984)	(2.072)		(2.834)	(3.163)
Sarpanch relative present		-2.858	-3.678		-6.440*	-6.974*		-3.824	-4.740
		(2.991)	(3.106)		(2.855)	(3.258)		(3.567)	(3.889)
Sarpanch SC		0.435	0.239		0.896	0.498		0.373	0.449
		(1.585)	(1.654)		(1.672)	(1.786)		(2.338)	(2.366)
Sarpanch OBC		-1.155	-0.896		-1.617	-1.476		0.158	0.750
		(1.002)	(1.043)		(1.149)	(1.137)		(1.311)	(1.378)
Education above median		0.861	0.620		2.146*	2.071*		0.172	-0.130
		(0.798)	(0.823)		(0.968)	(0.980)		(1.007)	(1.047)
Wealth above median		0.332	-0.363		0.546	-0.514		0.606	0.028
		(0.773)	(0.819)		(0.918)	(0.987)		(0.942)	(0.994)
Lives with in-laws		-3.223+	-3.286+		-1.485	-1.575		-3.212	-3.490
		(1.695)	(1.813)		(1.625)	(1.678)		(2.401)	(2.586)
Lives with own parents		-0.578	-0.415		-0.679	-0.521		0.277	0.567
		(0.864)	(0.930)		(1.062)	(1.135)		(1.051)	(1.103)
Years as sarpanch		0.148	-0.019		0.166	0.033		0.094	-0.065
		(0.264)	(0.267)		(0.319)	(0.312)		(0.323)	(0.342)
Gender quota		0.391	0.958		1.548	1.835		-1.612	-0.567
		(1.288)	(1.388)		(1.425)	(1.522)		(1.790)	(1.886)
SC quota		1.259	1.590		1.633	2.260		1.457	1.441
		(1.572)	(1.623)		(1.654)	(1.797)		(2.372)	(2.355)
OBC quota		2.281*	2.237*		2.872**	2.863**		1.023	0.725
		(0.976)	(1.021)		(1.081)	(1.089)		(1.328)	(1.395)
Sarpanch woman:Sarpanch relative present		1.380	2.057		6.718+	7.155+		0.204	1.051
		(3.402)	(3.569)		(3.431)	(3.822)		(3.899)	(4.220)
Percent of village land held by biggest landholding caste			-0.038+			-0.027			-0.042
			(0.021)			(0.026)			(0.026)
Distance (km) from block center			-0.006			-0.023			0.026
			(0.030)			(0.034)			(0.042)
Directly elected			0.580			0.353			0.305
			(0.851)			(1.019)			(1.094)
T1				0.004	0.039	0.086
				(0.849)	(0.911)	(0.939)
T2							-0.279	-0.097	-0.218
							(0.884)	(0.967)	(1.002)
Num.Obs.	604	548	505	406	369	339	389	350	324
R2	0.000	0.024	0.036	0.000	0.048	0.058	0.000	0.028	0.043

P1b (H3.8b) table, Sarpanch and gram sevak post-private preferences, average difference
	Impact of T1/T2 combined			Impact of T1			Impact of T2
	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls
+ p < 0.1, * p < 0.05, p < 0.01, * p < 0.001
Each model is a linear regression of each outcome on an indicator for being assigned to the relevant treatment condition(s) (1) versus not (0). Unadjusted and covariate-adjusted models presented. All regressions use robust standard errors.
(Intercept)	10.080***	10.628***	13.035***	10.080***	10.061***	12.896***	10.080***	11.181***	12.658***
	(0.578)	(1.454)	(2.194)	(0.578)	(1.723)	(2.564)	(0.578)	(1.843)	(2.866)
Discussion rules	-0.053	-0.103	-0.254
	(0.704)	(0.719)	(0.760)
Sarpanch woman		0.141	-0.811		0.350	-0.191		-0.848	-2.407
		(2.096)	(2.245)		(1.654)	(1.822)		(3.135)	(3.425)
Sarpanch relative present		-2.005	-2.368		-4.612	-6.021+		-3.510	-3.600
		(3.260)	(3.318)		(3.753)	(3.512)		(3.743)	(3.627)
Sarpanch SC		1.058	0.682		0.845	0.218		0.695	0.594
		(1.307)	(1.454)		(1.483)	(1.671)		(1.985)	(2.118)
Sarpanch OBC		0.727	0.505		0.710	0.393		1.159	1.048
		(0.897)	(0.979)		(1.054)	(1.150)		(1.214)	(1.314)
Education above median		-0.571	-0.431		-0.107	0.016		-0.883	-0.784
		(0.716)	(0.746)		(0.857)	(0.887)		(0.931)	(0.969)
Wealth above median		0.338	0.553		0.907	1.282		0.502	0.609
		(0.722)	(0.766)		(0.851)	(0.892)		(0.923)	(0.985)
Lives with in-laws		-5.342**	-4.864*		-5.770***	-5.822***		-4.501	-3.810
		(1.890)	(2.014)		(1.292)	(1.392)		(2.846)	(3.149)
Lives with own parents		-1.708*	-1.799*		-2.648**	-2.559*		-1.174	-1.454
		(0.756)	(0.816)		(0.935)	(0.998)		(0.952)	(1.020)
Years as sarpanch		0.579*	0.697**		0.776*	1.042***		0.375	0.314
		(0.246)	(0.259)		(0.304)	(0.309)		(0.305)	(0.319)
Gender quota		2.223*	2.776*		2.521*	3.264**		2.702*	3.334*
		(1.002)	(1.083)		(1.158)	(1.257)		(1.343)	(1.352)
SC quota		-0.297	0.048		-0.502	-0.160		-0.465	-0.018
		(1.273)	(1.378)		(1.389)	(1.526)		(1.992)	(2.082)
OBC quota		0.682	0.681		0.650	0.690		-0.687	-0.707
		(0.921)	(0.992)		(1.070)	(1.132)		(1.225)	(1.321)
Sarpanch woman:Sarpanch relative present		1.952	1.924		6.079	6.860+		4.163	4.596
		(3.620)	(3.738)		(4.206)	(4.101)		(4.351)	(4.318)
Percent of village land held by biggest landholding caste			-0.029			-0.040+			-0.003
			(0.018)			(0.024)			(0.022)
Distance (km) from block center			-0.029			-0.022			-0.047
			(0.030)			(0.033)			(0.041)
Directly elected			-0.622			-1.449+			0.001
			(0.746)			(0.875)			(0.935)
T1				0.104	-0.060	-0.273
				(0.793)	(0.802)	(0.843)
T2							-0.224	-0.069	-0.270
							(0.833)	(0.874)	(0.925)
Num.Obs.	605	549	506	407	370	340	390	351	325
R2	0.000	0.057	0.069	0.000	0.083	0.111	0.000	0.044	0.052

P2 (H3.9): Assignment to any treatment (T1 or T2, considered jointly) should differentially reduce inequality between sarpanches from disadvantaged groups in the distance between post-treatment, private preferences of sarpanch versus those of other group members.

P2 is assessed by regressing each of these three metrics (in this case, comparing the sarpanch’s post-treatment preferences to the post-treatment preferences of other actors) on treatment status (T1 or T2 versus Control), as well as on sarpanches’ indicators of disadvantage (gender and caste), and on the interaction between these variables. We focus here on this interaction. We also control for the potential presence of a sarpanch relative during the interview (“the interview” being here understood as the group interview), both separately and in interaction with the gender of the sarpanch. As above, we repeat this exercise twice. P2a tests the difference between the sarpanch’s post-treatment allocation and the upa-sarpanch’s post-treatment allocation. P2b tests the difference between the sarpanch’s post-treatment allocation and the gram sevak’s post-treatment allocation.
In additional analyses, we disentangle between T1 and T2.
We run bivariate regressions, clustering at the taluka level.
We also include the series of individual and village level demographic controls presented in G2.
Deviation 1: We originally planned to use clustered standard errors at the taluka level. This was an error on our part: because the treatment assignment occurs at the level of each village council, we do not cluster standard errors in these analyses.
Deviation 2: We originally planed to use the Earth Mover’s Distance (Lupu et al 2017) as our primary outcome. After discussion with the developers of the measure, we decided not to use this measure for two reasons: 1) it does not provide additional information relative to simpler measures, and 2) the package to implement the measure is also not operational.
Deviation 3: We are unable to implement metric 3 and 4 as planned because we did not ask respondents what their top priority was post-budget discussion. We thus cannot identify top priorities among the respondents who did not have one budget category with the highest amount of allocation.

P2a (H3.9a) table, Sarpanch and upa sarpanch post-private preferences, average difference; Gender disadvantage
	Impact of T1/T2 combined			Impact of T1			Impact of T2
	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls
+ p < 0.1, * p < 0.05, p < 0.01, * p < 0.001
Each model is a linear regression of each outcome on an indicator for being assigned to the relevant treatment condition(s) (1) versus not (0). Unadjusted and covariate-adjusted models presented. All regressions use robust standard errors.
(Intercept)	10.122***	8.582***	11.694***	10.122***	7.642***	10.528***	10.122***	8.158***	10.730***
	(1.032)	(1.730)	(2.379)	(1.032)	(1.984)	(2.726)	(1.032)	(1.924)	(2.890)
Discussion rules	-0.388	-0.081	-0.167
	(1.202)	(1.272)	(1.263)
Sarpanch woman	-0.827	2.358	1.597	-0.827	-0.313	-0.823	-0.827	4.507	3.606
	(1.280)	(2.292)	(2.463)	(1.280)	(2.306)	(2.395)	(1.280)	(2.948)	(3.238)
Discussion rules:Sarpanch woman	0.335	0.133	0.191
	(1.545)	(1.633)	(1.637)
Sarpanch relative present		-2.869	-3.697		-6.574*	-7.112*		-3.805	-4.714
		(3.013)	(3.132)		(2.979)	(3.349)		(3.487)	(3.827)
Sarpanch SC		0.439	0.242		0.925	0.501		0.347	0.433
		(1.587)	(1.656)		(1.666)	(1.784)		(2.329)	(2.360)
Sarpanch OBC		-1.156	-0.897		-1.620	-1.483		0.162	0.757
		(1.003)	(1.043)		(1.147)	(1.137)		(1.315)	(1.380)
Education above median		0.861	0.620		2.133*	2.064*		0.164	-0.137
		(0.799)	(0.824)		(0.971)	(0.984)		(1.005)	(1.044)
Wealth above median		0.336	-0.359		0.580	-0.490		0.587	0.017
		(0.776)	(0.821)		(0.926)	(0.992)		(0.943)	(0.996)
Lives with in-laws		-3.225+	-3.289+		-1.495	-1.586		-3.206	-3.472
		(1.694)	(1.812)		(1.607)	(1.667)		(2.423)	(2.617)
Lives with own parents		-0.580	-0.417		-0.703	-0.535		0.281	0.567
		(0.865)	(0.930)		(1.063)	(1.135)		(1.051)	(1.103)
Years as sarpanch		0.148	-0.019		0.158	0.031		0.093	-0.066
		(0.265)	(0.268)		(0.321)	(0.313)		(0.323)	(0.341)
Gender quota		0.393	0.958		1.584	1.857		-1.602	-0.544
		(1.292)	(1.390)		(1.439)	(1.535)		(1.793)	(1.885)
SC quota		1.252	1.581		1.558	2.212		1.495	1.469
		(1.576)	(1.624)		(1.646)	(1.788)		(2.365)	(2.349)
OBC quota		2.277*	2.233*		2.840**	2.837**		1.040	0.729
		(0.972)	(1.019)		(1.076)	(1.086)		(1.323)	(1.395)
Sarpanch woman:Sarpanch relative present		1.391	2.076		6.836+	7.279+		0.165	1.015
		(3.423)	(3.592)		(3.537)	(3.904)		(3.819)	(4.153)
Percent of village land held by biggest landholding caste			-0.038+			-0.027			-0.042
			(0.021)			(0.026)			(0.026)
Distance (km) from block center			-0.007			-0.025			0.027
			(0.030)			(0.034)			(0.042)
Directly elected			0.580			0.346			0.301
			(0.852)			(1.026)			(1.094)
T1				-0.733	-0.495	-0.385
				(1.367)	(1.431)	(1.434)
T1:Sarpanch woman				1.189	0.890	0.791
				(1.751)	(1.879)	(1.889)
T2							-0.054	0.282	0.084
							(1.338)	(1.441)	(1.459)
T2:Sarpanch woman							-0.607	-0.646	-0.508
							(1.804)	(1.887)	(1.922)
Num.Obs.	604	548	505	406	369	339	389	350	324
R2	0.001	0.024	0.036	0.001	0.048	0.059	0.005	0.029	0.044

P2b (H3.9b) table, Sarpanch and gram sevak post-private preferences, average difference; Gender disadvantage
	Impact of T1/T2 combined			Impact of T1			Impact of T2
	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls
+ p < 0.1, * p < 0.05, p < 0.01, * p < 0.001
Each model is a linear regression of each outcome on an indicator for being assigned to the relevant treatment condition(s) (1) versus not (0). Unadjusted and covariate-adjusted models presented. All regressions use robust standard errors.
(Intercept)	10.687***	9.569***	12.113***	10.687***	9.381***	12.577***	10.687***	10.338***	11.559***
	(0.916)	(1.595)	(2.287)	(0.916)	(1.810)	(2.642)	(0.916)	(1.940)	(2.880)
Discussion rules	0.548	1.259	0.806
	(1.114)	(1.182)	(1.245)
Sarpanch woman	-0.947	1.725	0.385	-0.947	1.314	0.190	-0.947	0.450	-1.161
	(1.179)	(2.280)	(2.418)	(1.179)	(1.902)	(2.071)	(1.179)	(3.161)	(3.391)
Discussion rules:Sarpanch woman	-1.240	-2.226	-1.721
	(1.432)	(1.484)	(1.562)
Sarpanch relative present		-1.829	-2.197		-4.364	-5.905+		-3.430	-3.461
		(2.992)	(3.102)		(3.508)	(3.441)		(3.428)	(3.291)
Sarpanch SC		0.999	0.662		0.793	0.216		0.595	0.513
		(1.330)	(1.469)		(1.506)	(1.680)		(2.033)	(2.153)
Sarpanch OBC		0.741	0.520		0.719	0.400		1.181	1.090
		(0.901)	(0.982)		(1.061)	(1.154)		(1.216)	(1.309)
Education above median		-0.566	-0.428		-0.083	0.023		-0.913	-0.820
		(0.715)	(0.746)		(0.854)	(0.885)		(0.931)	(0.968)
Wealth above median		0.274	0.516		0.844	1.262		0.432	0.553
		(0.721)	(0.765)		(0.847)	(0.889)		(0.923)	(0.985)
Lives with in-laws		-5.319**	-4.837*		-5.753***	-5.814***		-4.481	-3.720
		(1.899)	(2.018)		(1.345)	(1.410)		(2.844)	(3.126)
Lives with own parents		-1.679*	-1.788*		-2.607**	-2.548*		-1.164	-1.460
		(0.758)	(0.818)		(0.937)	(1.000)		(0.958)	(1.025)
Years as sarpanch		0.585*	0.698**		0.790**	1.044***		0.368	0.308
		(0.245)	(0.258)		(0.301)	(0.308)		(0.304)	(0.318)
Gender quota		2.183*	2.772*		2.452*	3.244*		2.735*	3.450*
		(1.010)	(1.093)		(1.165)	(1.262)		(1.356)	(1.363)
SC quota		-0.167	0.124		-0.361	-0.120		-0.315	0.127
		(1.291)	(1.390)		(1.409)	(1.535)		(2.032)	(2.102)
OBC quota		0.747	0.714		0.710	0.712		-0.618	-0.683
		(0.927)	(0.995)		(1.086)	(1.143)		(1.227)	(1.320)
Sarpanch woman:Sarpanch relative present		1.765	1.747		5.862	6.756+		4.012	4.405
		(3.364)	(3.526)		(3.974)	(4.023)		(4.050)	(3.999)
Percent of village land held by biggest landholding caste			-0.028			-0.039+			-0.000
			(0.018)			(0.024)			(0.022)
Distance (km) from block center			-0.027			-0.021			-0.044
			(0.029)			(0.033)			(0.041)
Directly elected			-0.622			-1.443			-0.019
			(0.745)			(0.876)			(0.928)
T1				0.462	0.931	0.123
				(1.308)	(1.364)	(1.400)
T1:Sarpanch woman				-0.727	-1.649	-0.663
				(1.635)	(1.647)	(1.721)
T2							0.630	1.423	1.329
							(1.259)	(1.364)	(1.444)
T2:Sarpanch woman							-1.833	-2.543	-2.693
							(1.674)	(1.759)	(1.852)
Num.Obs.	605	549	506	407	370	340	390	351	325
R2	0.013	0.061	0.071	0.007	0.085	0.112	0.016	0.050	0.059

P2a (H3.9a) table, Sarpanch and upa sarpanch post-private preferences, average difference; Caste disadvantage
	Impact of T1/T2 combined			Impact of T1			Impact of T2
	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls
+ p < 0.1, * p < 0.05, p < 0.01, * p < 0.001
Each model is a linear regression of each outcome on an indicator for being assigned to the relevant treatment condition(s) (1) versus not (0). Unadjusted and covariate-adjusted models presented. All regressions use robust standard errors.
(Intercept)	8.755***	8.227***	11.167***	8.755***	7.143***	9.908***	8.755***	8.061***	10.397***
	(0.661)	(1.513)	(2.236)	(0.661)	(1.796)	(2.605)	(0.661)	(1.789)	(2.805)
Discussion rules	0.589	0.421	0.477
	(0.864)	(0.917)	(0.932)
Sarpanch SC	2.717+	1.493	1.518	2.717+	1.406	1.238	2.717+	1.318	1.629
	(1.443)	(1.992)	(2.048)	(1.443)	(2.070)	(2.165)	(1.443)	(2.421)	(2.447)
Discussion rules:Sarpanch SC	-2.332	-1.499	-1.825
	(1.693)	(1.803)	(1.850)
Sarpanch woman		2.441	1.701		0.253	-0.329		4.022	3.182
		(1.992)	(2.203)		(2.001)	(2.078)		(2.897)	(3.230)
Sarpanch relative present		-2.946	-3.759		-6.541*	-7.081*		-3.886	-4.783
		(3.064)	(3.193)		(2.952)	(3.392)		(3.644)	(4.015)
Sarpanch OBC		-1.155	-0.890		-1.617	-1.469		0.182	0.792
		(1.003)	(1.043)		(1.151)	(1.139)		(1.312)	(1.376)
Education above median		0.858	0.617		2.144*	2.072*		0.175	-0.131
		(0.799)	(0.823)		(0.969)	(0.981)		(1.008)	(1.047)
Wealth above median		0.326	-0.394		0.545	-0.533		0.582	-0.041
		(0.773)	(0.819)		(0.918)	(0.988)		(0.941)	(0.992)
Lives with in-laws		-3.215+	-3.289+		-1.488	-1.578		-3.174	-3.485
		(1.723)	(1.845)		(1.645)	(1.699)		(2.484)	(2.698)
Lives with own parents		-0.585	-0.427		-0.689	-0.545		0.285	0.585
		(0.862)	(0.926)		(1.061)	(1.132)		(1.045)	(1.093)
Years as sarpanch		0.157	-0.012		0.172	0.037		0.108	-0.045
		(0.264)	(0.267)		(0.319)	(0.312)		(0.323)	(0.340)
Gender quota		0.400	0.995		1.509	1.807		-1.481	-0.374
		(1.270)	(1.355)		(1.428)	(1.506)		(1.755)	(1.820)
SC quota		1.244	1.570		1.573	2.164		1.580	1.617
		(1.574)	(1.624)		(1.663)	(1.808)		(2.396)	(2.380)
OBC quota		2.248*	2.184*		2.851**	2.820*		0.955	0.630
		(0.979)	(1.024)		(1.083)	(1.091)		(1.334)	(1.401)
Sarpanch woman:Sarpanch relative present		1.530	2.250		6.840+	7.313+		0.415	1.358
		(3.459)	(3.636)		(3.493)	(3.921)		(3.958)	(4.316)
Percent of village land held by biggest landholding caste			-0.037+			-0.026			-0.040
			(0.021)			(0.026)			(0.026)
Distance (km) from block center			-0.005			-0.023			0.029
			(0.030)			(0.034)			(0.042)
Directly elected			0.557			0.344			0.234
			(0.849)			(1.018)			(1.094)
T1				0.429	0.274	0.435
				(0.985)	(1.038)	(1.057)
T1:Sarpanch SC				-1.349	-0.863	-1.249
				(1.907)	(2.086)	(2.148)
T2							0.777	0.489	0.544
							(1.083)	(1.157)	(1.198)
T2:Sarpanch SC							-3.285+	-1.975	-2.506
							(1.923)	(2.058)	(2.127)
Num.Obs.	602	548	505	405	369	339	388	350	324
R2	0.007	0.025	0.038	0.013	0.048	0.059	0.011	0.031	0.048

P2b (H3.9b) table, Sarpanch and gram sevak post-private preferences, average difference; Caste disadvantage
	Impact of T1/T2 combined			Impact of T1			Impact of T2
	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls
+ p < 0.1, * p < 0.05, p < 0.01, * p < 0.001
Each model is a linear regression of each outcome on an indicator for being assigned to the relevant treatment condition(s) (1) versus not (0). Unadjusted and covariate-adjusted models presented. All regressions use robust standard errors.
(Intercept)	10.704***	11.087***	13.635***	10.704***	10.322***	13.205***	10.704***	11.604***	13.327***
	(0.679)	(1.476)	(2.202)	(0.679)	(1.719)	(2.558)	(0.679)	(1.864)	(2.852)
Discussion rules	-0.780	-0.764	-1.001
	(0.850)	(0.874)	(0.930)
Sarpanch SC	-2.029	-0.605	-1.134	-2.029	-0.162	-0.740	-2.029	-0.592	-0.873
	(1.274)	(1.717)	(1.849)	(1.274)	(1.833)	(2.006)	(1.274)	(2.177)	(2.298)
Discussion rules:Sarpanch SC	2.399	2.360	2.595+
	(1.521)	(1.499)	(1.554)
Sarpanch woman		0.159	-0.769		0.260	-0.244		-0.634	-2.169
		(2.098)	(2.249)		(1.624)	(1.805)		(3.190)	(3.489)
Sarpanch relative present		-1.865	-2.252		-4.412	-5.882		-3.425	-3.545
		(3.122)	(3.207)		(3.750)	(3.615)		(3.457)	(3.311)
Sarpanch OBC		0.728	0.499		0.712	0.386		1.127	0.999
		(0.896)	(0.977)		(1.052)	(1.148)		(1.213)	(1.313)
Education above median		-0.565	-0.426		-0.103	0.015		-0.885	-0.780
		(0.717)	(0.747)		(0.858)	(0.888)		(0.931)	(0.969)
Wealth above median		0.348	0.598		0.909	1.307		0.537	0.697
		(0.723)	(0.766)		(0.854)	(0.894)		(0.922)	(0.985)
Lives with in-laws		-5.356**	-4.861*		-5.764***	-5.819***		-4.556	-3.817
		(1.894)	(2.020)		(1.256)	(1.368)		(2.904)	(3.219)
Lives with own parents		-1.700*	-1.785*		-2.629**	-2.530*		-1.188	-1.482
		(0.756)	(0.818)		(0.936)	(1.000)		(0.953)	(1.021)
Years as sarpanch		0.565*	0.687**		0.763*	1.037***		0.355	0.289
		(0.245)	(0.258)		(0.303)	(0.308)		(0.304)	(0.317)
Gender quota		2.207*	2.719*		2.596*	3.298**		2.520+	3.089*
		(1.001)	(1.084)		(1.159)	(1.256)		(1.351)	(1.367)
SC quota		-0.272	0.077		-0.382	-0.035		-0.632	-0.237
		(1.275)	(1.381)		(1.399)	(1.536)		(2.003)	(2.100)
OBC quota		0.736	0.756		0.693	0.746		-0.593	-0.587
		(0.921)	(0.991)		(1.067)	(1.130)		(1.226)	(1.325)
Sarpanch woman:Sarpanch relative present		1.715	1.648		5.840	6.655		3.874	4.213
		(3.484)	(3.630)		(4.191)	(4.181)		(4.103)	(4.061)
Percent of village land held by biggest landholding caste			-0.031+			-0.041+			-0.005
			(0.018)			(0.024)			(0.022)
Distance (km) from block center			-0.030			-0.023			-0.050
			(0.029)			(0.033)			(0.040)
Directly elected			-0.587			-1.436			0.093
			(0.746)			(0.877)			(0.937)
T1				-0.490	-0.522	-0.724
				(0.951)	(0.969)	(1.026)
T1:Sarpanch SC				2.088	1.703	1.620
				(1.719)	(1.666)	(1.720)
T2							-1.121	-0.867	-1.218
							(1.042)	(1.085)	(1.152)
T2:Sarpanch SC							2.759	2.695	3.124+
							(1.759)	(1.784)	(1.881)
Num.Obs.	603	549	506	406	370	340	389	351	325
R2	0.005	0.061	0.073	0.007	0.085	0.113	0.008	0.050	0.060

P3 (H3.10): Distance between the sarpanch’s pre-treatment private preferences and other actors’ post-treatment private preferences should be smaller in treated groups (T1 + T2, considered jointly) than control groups.

P3 is assessed by regressing each of these four metrics (in this case, comparing the sarpanch’s pre-treatment preferences to the post-treatment preferences of other actors) on treatment status (T1 or T2 versus Control). We also control for the potential presence of a sarpanch relative during the interview (“the interview” being here understood as the group interview), both separately and in interaction with the gender of the sarpanch. We repeat this exercise twice. P3a tests the difference between the sarpanch’s pre-treatment allocation and the upa-sarpanch’s post-treatment allocation. P3b tests the difference between the sarpanch’s pre-treatment allocation and the gram sevak’s post-treatment allocation.
In additional analyses, we disentangle between T1 and T2.
We run bivariate regressions, clustering at the taluka level.
We also include the series of individual and village level demographic controls presented in G2.
Deviation 1: We originally planned to use clustered standard errors at the taluka level. This was an error on our part: because the treatment assignment occurs at the level of each village council, we do not cluster standard errors in these analyses.
Deviation 2: We originally planed to use the Earth Mover’s Distance (Lupu et al 2017) as our primary outcome. After discussion with the developers of the measure, we decided not to use this measure for two reasons: 1) it does not provide additional information relative to simpler measures, and 2) the package to implement the measure is also not operational.
Deviation 3: We are unable to implement metric 3 as planned because we did not ask respondents what their top priority was post-budget discussion. We thus cannot identify top priorities among the respondents who did not have one budget category with the highest amount of allocation.

P3a (H3.10a) table, grouping T1/T2 together, sarpanch pre- upa sarpanch post-
	DV: Sarpanch pre- upa sarpanch post- average distance			DV: Sarpanch pre- upa sarpanch post- distance from sarpanch's top priority
	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls
+ p < 0.1, * p < 0.05, p < 0.01, * p < 0.001
Each model is a linear regression of each outcome on an indicator for being assigned to discussion rules (1) versus not (0). Unadjusted and covariate-adjusted models presented. All regressions use robust standard errors.
(Intercept)	21.637***	19.840***	26.025***	47.176***	41.822***	62.436***
	(0.792)	(2.042)	(2.948)	(2.271)	(5.886)	(8.493)
Discussion rules	-0.052	-0.408	-0.492	0.369	-0.950	-2.038
	(0.981)	(1.049)	(1.099)	(2.818)	(3.005)	(3.132)
Sarpanch woman		7.122*	5.611+		19.305*	13.986
		(2.757)	(2.988)		(8.229)	(8.831)
Sarpanch relative present		-0.018	-1.438		-0.410	-5.028
		(5.097)	(5.813)		(12.233)	(15.260)
Sarpanch SC		1.762	1.322		-2.035	-3.650
		(2.214)	(2.281)		(6.451)	(6.683)
Sarpanch OBC		0.042	-0.043		-1.621	-1.437
		(1.473)	(1.558)		(4.224)	(4.473)
Education above median		0.720	0.127		1.587	-0.104
		(1.035)	(1.070)		(3.011)	(3.096)
Wealth above median		-0.441	-0.737		-2.017	-2.847
		(1.020)	(1.070)		(2.972)	(3.121)
Lives with in-laws		-3.442	-3.359		-6.804	-6.191
		(2.511)	(2.584)		(7.577)	(7.673)
Lives with own parents		0.664	1.640		2.774	5.305
		(1.107)	(1.153)		(3.253)	(3.389)
Years as sarpanch		0.188	0.144		0.329	0.206
		(0.340)	(0.357)		(0.981)	(1.027)
Gender quota		-2.113	-0.180		-6.578	-0.913
		(1.589)	(1.753)		(4.675)	(5.181)
SC quota		-0.172	-0.834		6.817	5.075
		(2.208)	(2.277)		(6.363)	(6.584)
OBC quota		-1.231	-1.151		-0.781	-0.688
		(1.436)	(1.519)		(4.182)	(4.427)
Sarpanch woman:Sarpanch relative present		-3.505	-3.337		-7.642	-7.042
		(5.560)	(6.263)		(13.822)	(16.651)
Percent of village land held by biggest landholding caste			-0.085**			-0.281***
			(0.027)			(0.073)
Distance (km) from block center			-0.014			-0.002
			(0.042)			(0.118)
Directly elected			-0.212			-1.531
			(1.091)			(3.192)
Num.Obs.	592	536	494	594	538	497
R2	0.000	0.027	0.051	0.000	0.028	0.056

P3a (H3.10a) table, T1 versus control, sarpanch pre- upa sarpanch post-
	DV: Sarpanch pre- upa sarpanch post- average distance			DV: Sarpanch pre- upa sarpanch post- distance from sarpanch's top priority
	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls
+ p < 0.1, * p < 0.05, p < 0.01, * p < 0.001
Each model is a linear regression of each outcome on an indicator for being assigned to T1 (1) versus control (0). Unadjusted and covariate-adjusted models presented. All regressions use robust standard errors.
(Intercept)	21.637***	20.770***	26.266***	47.990***	44.309***	63.296***
	(0.792)	(2.366)	(3.415)	(1.654)	(6.929)	(10.186)
T1	0.061	0.076	0.057	-1.605	-0.944	-2.013
	(1.117)	(1.210)	(1.284)	(2.850)	(3.482)	(3.669)
Sarpanch woman		6.555*	4.998+		16.922+	11.880
		(2.959)	(2.993)		(9.005)	(8.847)
Sarpanch relative present		-1.012	-3.208		-8.759	-17.198
		(9.573)	(10.559)		(20.210)	(23.944)
Sarpanch SC		0.718	0.642		-7.973	-8.532
		(2.602)	(2.583)		(7.648)	(7.745)
Sarpanch OBC		-2.602	-2.376		-10.446*	-9.240+
		(1.712)	(1.769)		(4.932)	(5.096)
Education above median		-0.196	-0.621		-1.115	-2.431
		(1.250)	(1.279)		(3.696)	(3.779)
Wealth above median		0.995	0.487		1.762	0.538
		(1.223)	(1.305)		(3.584)	(3.832)
Lives with in-laws		-4.102	-4.226+		-6.841	-7.633
		(2.603)	(2.454)		(8.168)	(7.466)
Lives with own parents		0.890	1.377		3.396	4.470
		(1.357)	(1.401)		(3.938)	(4.087)
Years as sarpanch		-0.097	-0.050		-0.539	-0.310
		(0.420)	(0.439)		(1.220)	(1.269)
Gender quota		-0.757	1.316		-2.797	2.965
		(1.766)	(1.930)		(5.226)	(5.707)
SC quota		0.305	-0.458		10.474	8.291
		(2.593)	(2.601)		(7.478)	(7.619)
OBC quota		-0.182	-0.375		4.504	3.699
		(1.621)	(1.673)		(4.759)	(4.895)
Sarpanch woman:Sarpanch relative present		1.485	1.827		12.007	14.718
		(9.854)	(10.798)		(21.382)	(24.924)
Percent of village land held by biggest landholding caste			-0.079*			-0.265**
			(0.033)			(0.094)
Distance (km) from block center			0.003			0.040
			(0.049)			(0.144)
Directly elected			-0.181			-1.206
			(1.318)			(3.855)
Num.Obs.	395	358	329	594	359	331
R2	0.000	0.034	0.052	0.001	0.043	0.061

P3a (H3.10a) table, T2 versus control, sarpanch pre- upa sarpanch post-
	DV: Sarpanch pre- upa sarpanch post- average distance			DV: Sarpanch pre- upa sarpanch post- distance from sarpanch's top priority
	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls
+ p < 0.1, * p < 0.05, p < 0.01, * p < 0.001
Each model is a linear regression of each outcome on an indicator for being assigned to T2 (1) versus control (0). Unadjusted and covariate-adjusted models presented. All regressions use robust standard errors.
(Intercept)	21.637***	17.841***	23.268***	46.760***	35.826***	53.676***
	(0.792)	(2.472)	(3.861)	(1.626)	(7.040)	(10.799)
T2	-0.172	-0.467	-0.733	2.003	0.265	-1.146
	(1.164)	(1.252)	(1.307)	(2.900)	(3.550)	(3.693)
Sarpanch woman		7.871+	7.169		24.035*	21.440
		(4.119)	(4.697)		(11.967)	(13.500)
Sarpanch relative present		2.201	0.903		1.542	-2.238
		(6.106)	(6.895)		(14.467)	(17.918)
Sarpanch SC		2.871	2.058		5.348	2.443
		(3.175)	(3.290)		(8.492)	(8.805)
Sarpanch OBC		2.429	2.478		7.168	7.767
		(1.923)	(2.072)		(5.356)	(5.810)
Education above median		1.706	0.979		6.294+	4.626
		(1.315)	(1.365)		(3.717)	(3.845)
Wealth above median		-0.690	-1.030		-2.902	-4.010
		(1.254)	(1.315)		(3.597)	(3.775)
Lives with in-laws		-3.115	-2.950		-8.810	-7.844
		(3.771)	(4.101)		(11.094)	(11.928)
Lives with own parents		1.377	2.651+		4.024	7.250+
		(1.380)	(1.441)		(4.025)	(4.175)
Years as sarpanch		0.126	-0.140		0.135	-0.354
		(0.429)	(0.470)		(1.206)	(1.311)
Gender quota		-2.030	-1.024		-7.021	-4.310
		(2.226)	(2.492)		(6.157)	(6.784)
SC quota		-0.939	-1.128		2.235	2.013
		(3.238)	(3.357)		(8.618)	(8.944)
OBC quota		-1.968	-1.917		-5.590	-5.489
		(1.889)	(2.014)		(5.333)	(5.717)
Sarpanch woman:Sarpanch relative present		-6.625	-5.239		-10.034	-6.791
		(6.680)	(7.444)		(16.562)	(19.751)
Percent of village land held by biggest landholding caste			-0.072*			-0.244**
			(0.034)			(0.092)
Distance (km) from block center			-0.018			-0.013
			(0.059)			(0.162)
Directly elected			0.346			-0.606
			(1.397)			(4.010)
Num.Obs.	383	344	319	594	347	322
R2	0.000	0.042	0.062	0.001	0.047	0.073

P3b (H3.10b) table, grouping T1/T2 together, sarpanch pre- gram sevak post-
	DV: Sarpanch pre- gram sevak post- average distance			DV: Sarpanch pre- gram sevak post- distance from sarpanch's top priority
	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls
+ p < 0.1, * p < 0.05, p < 0.01, * p < 0.001
Each model is a linear regression of each outcome on an indicator for being assigned to discussion rules (1) versus not (0). Unadjusted and covariate-adjusted models presented. All regressions use robust standard errors.
(Intercept)	23.412***	23.006***	28.183***	50.974***	45.132***	63.050***
	(0.763)	(1.958)	(2.753)	(2.266)	(5.810)	(8.093)
Discussion rules	-1.089	-1.655	-2.416*	-1.560	-3.210	-5.883+
	(0.947)	(1.007)	(1.043)	(2.782)	(2.963)	(3.061)
Sarpanch woman		6.758*	4.776		18.288*	11.050
		(2.819)	(3.069)		(8.618)	(9.064)
Sarpanch relative present		1.834	0.557		-3.180	-7.708
		(5.518)	(5.830)		(15.530)	(18.053)
Sarpanch SC		2.220	1.886		0.813	-0.842
		(1.925)	(2.101)		(5.608)	(6.038)
Sarpanch OBC		-0.094	0.008		-1.028	-0.891
		(1.409)	(1.518)		(3.990)	(4.271)
Education above median		-0.454	-1.181		0.285	-1.988
		(0.987)	(1.004)		(2.881)	(2.917)
Wealth above median		-0.222	-0.477		-0.086	-1.215
		(0.990)	(1.031)		(2.901)	(3.003)
Lives with in-laws		-3.110	-2.244		-4.163	-1.614
		(2.448)	(2.557)		(7.629)	(7.651)
Lives with own parents		0.279	0.965		2.023	4.178
		(1.098)	(1.152)		(3.242)	(3.388)
Years as sarpanch		0.081	0.086		0.347	0.471
		(0.330)	(0.347)		(0.948)	(0.998)
Gender quota		-2.973+	-1.527		-9.457*	-4.626
		(1.649)	(1.828)		(4.667)	(5.164)
SC quota		-1.212	-1.710		5.090	4.279
		(1.901)	(2.055)		(5.510)	(5.880)
OBC quota		-0.066	0.113		1.330	1.881
		(1.393)	(1.481)		(3.988)	(4.236)
Sarpanch woman:Sarpanch relative present		-5.443	-5.256		-5.835	-3.737
		(6.018)	(6.343)		(16.946)	(19.368)
Percent of village land held by biggest landholding caste			-0.093***			-0.288***
			(0.024)			(0.066)
Distance (km) from block center			0.072+			0.187
			(0.042)			(0.118)
Directly elected			-0.064			-1.370
			(1.045)			(3.055)
Num.Obs.	593	537	495	595	539	498
R2	0.002	0.025	0.061	0.001	0.028	0.065

P3b (H3.10b) table, T1 versus control, sarpanch pre- gram sevak post-
	DV: Sarpanch pre- gram sevak post- average distance			DV: Sarpanch pre- gram sevak post- distance from sarpanch's top priority
	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls
+ p < 0.1, * p < 0.05, p < 0.01, * p < 0.001
Each model is a linear regression of each outcome on an indicator for being assigned to T1 (1) versus control (0). Unadjusted and covariate-adjusted models presented. All regressions use robust standard errors.
(Intercept)	23.412***	24.177***	29.307***	50.470***	47.487***	64.731***
	(0.763)	(2.223)	(3.008)	(1.631)	(6.608)	(9.183)
T1	-1.042	-1.604	-2.587*	-1.605	-3.618	-7.064*
	(1.076)	(1.152)	(1.206)	(2.758)	(3.403)	(3.554)
Sarpanch woman		7.693**	6.150+		19.820*	14.587
		(2.927)	(3.218)		(9.276)	(9.719)
Sarpanch relative present		1.975	-1.270		-13.862	-24.255
		(8.265)	(8.612)		(22.556)	(24.926)
Sarpanch SC		-0.552	-0.264		-8.633	-8.348
		(2.065)	(2.164)		(6.358)	(6.601)
Sarpanch OBC		-2.208	-1.429		-7.956+	-5.761
		(1.616)	(1.699)		(4.667)	(4.923)
Education above median		-1.344	-2.315+		-1.329	-4.263
		(1.192)	(1.199)		(3.527)	(3.571)
Wealth above median		0.612	0.325		3.034	2.036
		(1.175)	(1.228)		(3.487)	(3.653)
Lives with in-laws		-4.743+	-4.675+		-7.496	-7.884
		(2.497)	(2.647)		(8.264)	(8.184)
Lives with own parents		0.500	0.606		1.468	1.980
		(1.285)	(1.328)		(3.874)	(4.007)
Years as sarpanch		-0.005	0.132		0.328	0.837
		(0.402)	(0.418)		(1.173)	(1.231)
Gender quota		-2.719	-1.255		-8.736+	-4.426
		(1.827)	(1.962)		(5.168)	(5.606)
SC quota		-0.268	-0.920		9.706	8.689
		(2.012)	(2.100)		(6.148)	(6.310)
OBC quota		0.687	0.479		5.022	4.486
		(1.582)	(1.630)		(4.637)	(4.817)
Sarpanch woman:Sarpanch relative present		-1.666	0.064		15.641	22.849
		(8.647)	(8.982)		(23.744)	(26.016)
Percent of village land held by biggest landholding caste			-0.103***			-0.302***
			(0.028)			(0.079)
Distance (km) from block center			0.106*			0.255+
			(0.044)			(0.131)
Directly elected			0.141			-0.355
			(1.207)			(3.615)
Num.Obs.	396	359	330	595	360	332
R2	0.002	0.034	0.085	0.001	0.039	0.082

P3b (H3.10b) table, T2 versus control, sarpanch pre- gram sevak post-
	DV: Sarpanch pre- gram sevak post- average distance			DV: Sarpanch pre- gram sevak post- distance from sarpanch's top priority
	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls
+ p < 0.1, * p < 0.05, p < 0.01, * p < 0.001
Each model is a linear regression of each outcome on an indicator for being assigned to T2 (1) versus control (0). Unadjusted and covariate-adjusted models presented. All regressions use robust standard errors.
(Intercept)	23.412***	21.865***	25.731***	49.869***	41.526***	57.215***
	(0.763)	(2.445)	(3.625)	(1.587)	(7.300)	(10.571)
T2	-1.138	-1.413	-2.141+	0.121	-2.176	-4.507
	(1.130)	(1.211)	(1.249)	(2.834)	(3.501)	(3.589)
Sarpanch woman		4.259	1.690		14.370	5.480
		(4.139)	(4.232)		(12.605)	(12.876)
Sarpanch relative present		2.702	1.681		-1.124	-5.107
		(7.142)	(7.267)		(19.233)	(21.834)
Sarpanch SC		3.813	3.322		7.593	4.623
		(2.682)	(2.849)		(7.165)	(7.682)
Sarpanch OBC		2.246	2.167		7.086	6.854
		(1.813)	(1.994)		(5.125)	(5.563)
Education above median		0.540	0.238		4.240	3.107
		(1.244)	(1.283)		(3.620)	(3.683)
Wealth above median		-0.566	-0.839		-1.651	-3.412
		(1.240)	(1.300)		(3.605)	(3.739)
Lives with in-laws		-0.915	1.403		-1.058	5.114
		(3.584)	(3.496)		(11.249)	(11.119)
Lives with own parents		-0.036	0.727		1.498	4.076
		(1.407)	(1.481)		(4.166)	(4.362)
Years as sarpanch		-0.022	-0.196		-0.086	-0.344
		(0.416)	(0.450)		(1.185)	(1.272)
Gender quota		-1.617	-0.577		-6.353	-2.528
		(2.190)	(2.399)		(6.037)	(6.479)
SC quota		-1.861	-2.444		2.158	1.719
		(2.644)	(2.799)		(7.078)	(7.573)
OBC quota		-1.406	-1.269		-3.494	-2.950
		(1.797)	(1.960)		(5.121)	(5.534)
Sarpanch woman:Sarpanch relative present		-6.050	-5.248		-5.039	-0.570
		(7.808)	(7.935)		(21.249)	(23.618)
Percent of village land held by biggest landholding caste			-0.076*			-0.248**
			(0.031)			(0.083)
Distance (km) from block center			0.064			0.139
			(0.062)			(0.170)
Directly elected			0.909			0.261
			(1.335)			(3.881)
Num.Obs.	384	345	320	595	348	323
R2	0.003	0.028	0.055	0.000	0.035	0.061

Third group of hypotheses: Collective Preferences

This measures the extent to which the collective preference reported by the group (the outcome of their collective deliberation) has changed.

CP1 (H3.12): The distance between (post-treatment) private preferences of sarpanch and outcome of group decision making should be lower for treated (T1, T2) groups than for those not treated (Control).

CP1 is assessed using differences between the private (post-treatment) preferences of the sarpanch versus the group decision, comparing the mean “preference difference” across treatment status (T1 or T2 versus Control). We use each of the three metrics above to run this comparison.
- Metric 1: average distance in allocation across the 5 priorities. This metric is by far the most sophisticated of the three. Practically, we calculate the difference in %age allocated to each item across the two distributions; we subsequently average this difference (that is, we compute the total difference across the five items and divide this total difference by 5). Proceeding in this manner, we obtain a number between 0 and 100, with 100 being a most different allocation across the five items.
- Metric 2. Binary measure coding whether the two distributions identify the same priority as the “top” priority (that is, the priority allocated the largest %age.
- Metric 3: Continuous measure identifying the distance between the %age of funds allocated to the “top” priority across the two allocations.
In additional analyses, we disentangle between T1 and T2.
We also include the series of individual and village level demographic controls presented in G2.
We also include a control for the potential presence of a sarpanch relative during the interview, both separately and in interaction with the gender of the sarpanch.
Deviation 1: We originally planed to use the Earth Mover’s Distance (Lupu et al 2017) as our primary outcome. After discussion with the developers of the measure, we decided not to use this measure for two reasons: 1) it does not provide additional information relative to simpler measures, and 2) the package to implement the measure is also not operational.
Deviation 2: We are unable to implement metric 2 as planned because we did not ask respondents what their top priority was post-budget discussion. We thus cannot identify top priorities among the respondents who did not have one budget category with the highest amount of allocation.
Deviation 3: We are unable to implement metric 3 as planned because we did not ask respondents what their top priority was post-budget discussion. We thus cannot identify top priorities among the respondents who did not have one budget category with the highest amount of allocation.

CP1 (H3.12) table, Sarpanch post- and collective group preferences, average difference
	Impact of T1/T2 combined			Impact of T1			Impact of T2
	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls
+ p < 0.1, * p < 0.05, p < 0.01, * p < 0.001
Each model is a linear regression of each outcome on an indicator for being assigned to the relevant treatment condition(s) versus not. Unadjusted and covariate-adjusted models presented. All regressions use robust standard errors.
(Intercept)	12.152***	13.649***	12.474***	12.152***	12.865***	11.052***	12.152***	13.525***	12.718***
	(0.660)	(1.652)	(2.355)	(0.660)	(1.930)	(2.890)	(0.660)	(2.057)	(2.872)
Discussion rules	-0.014	0.030	-0.026
	(0.809)	(0.854)	(0.882)
Sarpanch woman		0.256	-0.619		-1.920	-2.417		2.646	1.332
		(2.532)	(2.758)		(2.556)	(2.672)		(3.502)	(4.062)
Sarpanch relative present		-3.610	-3.460		-8.067***	-7.580*		-3.271	-3.238
		(3.477)	(3.459)		(2.322)	(3.100)		(4.438)	(4.263)
Sarpanch SC		0.526	0.590		0.762	1.092		-1.617	-1.449
		(1.774)	(1.894)		(1.996)	(2.176)		(2.472)	(2.626)
Sarpanch OBC		-0.141	0.255		-0.657	-0.077		0.307	0.820
		(1.108)	(1.187)		(1.380)	(1.451)		(1.340)	(1.445)
Education above median		-1.468+	-1.687+		-1.259	-1.877+		-0.663	-0.818
		(0.859)	(0.902)		(1.071)	(1.101)		(1.078)	(1.139)
Wealth above median		0.309	-0.249		0.526	-0.101		0.037	-0.494
		(0.840)	(0.888)		(1.009)	(1.055)		(1.055)	(1.119)
Lives with in-laws		-1.649	-0.948		1.036	1.289		-3.489	-2.992
		(2.361)	(2.506)		(2.337)	(2.337)		(3.336)	(3.758)
Lives with own parents		-2.216*	-2.060*		-2.221*	-1.625		-2.932**	-3.423**
		(0.882)	(0.954)		(1.092)	(1.176)		(1.064)	(1.115)
Years as sarpanch		0.105	0.175		0.151	0.161		0.124	0.240
		(0.298)	(0.313)		(0.353)	(0.359)		(0.377)	(0.395)
Gender quota		1.205	1.441		1.493	1.831		0.839	1.394
		(1.143)	(1.271)		(1.345)	(1.490)		(1.382)	(1.525)
SC quota		0.471	0.977		0.088	0.182		2.987	3.691
		(1.746)	(1.865)		(1.925)	(2.110)		(2.453)	(2.605)
OBC quota		-0.221	-0.134		0.610	0.739		-0.684	-0.796
		(1.138)	(1.229)		(1.405)	(1.487)		(1.369)	(1.489)
Sarpanch woman:Sarpanch relative present		-0.234	-1.014		5.039+	3.716		-1.842	-2.279
		(3.824)	(3.861)		(2.947)	(3.652)		(4.918)	(4.787)
Percent of village land held by biggest landholding caste			0.014			0.020			0.020
			(0.020)			(0.025)			(0.025)
Distance (km) from block center			0.013			0.043			-0.010
			(0.035)			(0.043)			(0.045)
Directly elected			-0.503			-1.004			-0.508
			(0.886)			(1.063)			(1.108)
T1				-0.682	-0.591	-0.650
				(0.903)	(0.963)	(1.002)
T2							0.716	0.598	0.531
							(0.968)	(1.033)	(1.083)
Num.Obs.	595	540	498	400	364	334	382	344	319
R2	0.000	0.032	0.040	0.001	0.041	0.054	0.001	0.052	0.068

CP2 (H3.13): [Heterogeneous treatment effects] the effect of the treatment, as implied by CP1, should be larger if the sarpanch belongs to a disadvantaged category (CP2a) or if the gram sevak is female (CP2b).

CP2 is assessed using a regression interacting sarpanch and/or gram sevak characteristics with treatment status for each relevant dependent variable. Our focus is on the interaction term; we cluster on administrative block level. We use the same three measures of preference distance outlined in CP1. Two-tailed tests are used in all cases. We use each of the three metrics above to run this comparison.
In additional analyses, we disentangle between T1 and T2.
We also include the series of individual and village level demographic controls presented in G2.
We also include a control for the potential presence of a sarpanch relative during the interview, both separately and in interaction with the gender of the sarpanch.
Deviation 1: We originally planed to use the Earth Mover’s Distance (Lupu et al 2017) as our primary outcome. After discussion with the developers of the measure, we decided not to use this measure for two reasons: 1) it does not provide additional information relative to simpler measures, and 2) the package to implement the measure is also not operational.
Deviation 2: We are unable to implement metric 2 as planned because we did not ask respondents what their top priority was post-budget discussion. We thus cannot identify top priorities among the respondents who did not have one budget category with the highest amount of allocation.
Deviation 3: We are unable to implement metric 3 as planned because we did not ask respondents what their top priority was post-budget discussion. We thus cannot identify top priorities among the respondents who did not have one budget category with the highest amount of allocation.
Deviation 4: We originally planned to use clustered standard errors at the taluka level. This was an error on our part: because the treatment assignment occurs at the level of each village council, we do not cluster standard errors in these analyses.

CP2a (H3.13a) table, Sarpanch post- and collective group preferences, average difference; Gender disadvantage
	Impact of T1/T2 combined			Impact of T1			Impact of T2
	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls
+ p < 0.1, * p < 0.05, p < 0.01, * p < 0.001
Each model is a linear regression of each outcome on an indicator for being assigned to the relevant treatment condition(s) versus not. Unadjusted and covariate-adjusted models presented. All regressions use robust standard errors.
(Intercept)	10.650***	12.051***	10.636***	10.650***	11.660***	9.492**	10.650***	12.287***	11.179***
	(1.155)	(1.856)	(2.518)	(1.155)	(2.073)	(2.950)	(1.155)	(2.201)	(3.060)
Discussion rules	1.708	2.097	2.090
	(1.340)	(1.369)	(1.400)
Sarpanch woman	2.360+	2.663	1.774	2.360+	-0.186	-0.548	2.360+	4.545	3.091
	(1.402)	(2.801)	(2.966)	(1.402)	(2.778)	(2.857)	(1.402)	(3.713)	(4.182)
Discussion rules:Sarpanch woman	-2.758	-3.379+	-3.443+
	(1.685)	(1.741)	(1.784)
Sarpanch relative present		-3.343	-3.110		-7.618***	-7.005**		-3.158	-3.029
		(3.081)	(3.049)		(1.885)	(2.650)		(3.863)	(3.683)
Sarpanch SC		0.426	0.538		0.663	1.082		-1.779	-1.586
		(1.730)	(1.852)		(1.946)	(2.126)		(2.372)	(2.518)
Sarpanch OBC		-0.131	0.274		-0.647	-0.050		0.322	0.861
		(1.101)	(1.178)		(1.379)	(1.446)		(1.319)	(1.419)
Education above median		-1.471+	-1.690+		-1.223	-1.852+		-0.727	-0.886
		(0.855)	(0.898)		(1.067)	(1.095)		(1.070)	(1.133)
Wealth above median		0.220	-0.310		0.417	-0.191		-0.051	-0.547
		(0.836)	(0.884)		(1.006)	(1.051)		(1.049)	(1.112)
Lives with in-laws		-1.615	-0.897		1.064	1.329		-3.456	-2.870
		(2.357)	(2.497)		(2.276)	(2.266)		(3.403)	(3.813)
Lives with own parents		-2.172*	-2.040*		-2.155*	-1.579		-2.910**	-3.425**
		(0.882)	(0.955)		(1.094)	(1.178)		(1.065)	(1.118)
Years as sarpanch		0.117	0.180		0.178	0.172		0.120	0.239
		(0.297)	(0.312)		(0.351)	(0.356)		(0.375)	(0.393)
Gender quota		1.149	1.444		1.368	1.742		0.902	1.574
		(1.142)	(1.271)		(1.345)	(1.492)		(1.383)	(1.517)
SC quota		0.654	1.116		0.326	0.359		3.180	3.877
		(1.698)	(1.817)		(1.873)	(2.056)		(2.344)	(2.487)
OBC quota		-0.127	-0.071		0.709	0.836		-0.584	-0.758
		(1.137)	(1.220)		(1.415)	(1.489)		(1.351)	(1.461)
Sarpanch woman:Sarpanch relative present		-0.539	-1.406		4.615+	3.159		-2.075	-2.583
		(3.469)	(3.498)		(2.619)	(3.280)		(4.396)	(4.270)
Percent of village land held by biggest landholding caste			0.016			0.022			0.023
			(0.020)			(0.025)			(0.025)
Distance (km) from block center			0.017			0.049			-0.006
			(0.035)			(0.042)			(0.045)
Directly elected			-0.530			-1.002			-0.569
			(0.882)			(1.058)			(1.098)
T1				0.862	1.177	1.281
				(1.509)	(1.549)	(1.583)
T1:Sarpanch woman				-2.434	-2.952	-3.253
				(1.883)	(1.982)	(2.030)
T2							2.554+	2.809+	2.800+
							(1.495)	(1.578)	(1.646)
T2:Sarpanch woman							-2.991	-3.744+	-3.801+
							(1.986)	(2.033)	(2.122)
Num.Obs.	595	540	498	400	364	334	382	344	319
R2	0.005	0.038	0.046	0.009	0.048	0.061	0.009	0.061	0.078

CP2a (H3.13a) table, Sarpanch post- and collective group preferences, average difference; Caste disadvantage
	Impact of T1/T2 combined			Impact of T1			Impact of T2
	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls
+ p < 0.1, * p < 0.05, p < 0.01, * p < 0.001
Each model is a linear regression of each outcome on an indicator for being assigned to the relevant treatment condition(s) versus not. Unadjusted and covariate-adjusted models presented. All regressions use robust standard errors.
(Intercept)	11.871***	13.732***	12.557***	11.871***	12.871***	10.996***	11.871***	13.592***	12.831***
	(0.790)	(1.678)	(2.372)	(0.790)	(1.941)	(2.899)	(0.790)	(2.075)	(2.865)
Discussion rules	-0.062	-0.091	-0.131
	(0.988)	(1.016)	(1.059)
Sarpanch SC	0.905	0.223	0.334	0.905	0.739	1.267	0.905	-1.827	-1.709
	(1.444)	(2.146)	(2.232)	(1.444)	(2.343)	(2.507)	(1.444)	(2.506)	(2.619)
Discussion rules:Sarpanch SC	0.099	0.430	0.365
	(1.737)	(1.857)	(1.914)
Sarpanch woman		0.261	-0.613		-1.922	-2.408		2.682	1.375
		(2.533)	(2.759)		(2.560)	(2.683)		(3.523)	(4.085)
Sarpanch relative present		-3.583	-3.442		-8.062***	-7.606*		-3.254	-3.224
		(3.463)	(3.449)		(2.337)	(3.113)		(4.417)	(4.232)
Sarpanch OBC		-0.141	0.254		-0.657	-0.075		0.302	0.811
		(1.108)	(1.188)		(1.381)	(1.452)		(1.343)	(1.448)
Education above median		-1.468+	-1.686+		-1.259	-1.876+		-0.663	-0.818
		(0.860)	(0.903)		(1.073)	(1.101)		(1.080)	(1.140)
Wealth above median		0.312	-0.241		0.526	-0.107		0.044	-0.477
		(0.840)	(0.887)		(1.010)	(1.054)		(1.056)	(1.119)
Lives with in-laws		-1.651	-0.947		1.036	1.289		-3.497	-2.993
		(2.363)	(2.505)		(2.341)	(2.350)		(3.346)	(3.766)
Lives with own parents		-2.215*	-2.059*		-2.221*	-1.630		-2.935**	-3.428**
		(0.883)	(0.955)		(1.095)	(1.180)		(1.068)	(1.121)
Years as sarpanch		0.102	0.173		0.151	0.162		0.121	0.237
		(0.297)	(0.313)		(0.353)	(0.359)		(0.376)	(0.394)
Gender quota		1.203	1.434		1.494	1.824		0.810	1.353
		(1.145)	(1.273)		(1.352)	(1.494)		(1.389)	(1.539)
SC quota		0.474	0.980		0.090	0.160		2.959	3.652
		(1.743)	(1.863)		(1.926)	(2.122)		(2.482)	(2.638)
OBC quota		-0.212	-0.124		0.611	0.729		-0.668	-0.774
		(1.140)	(1.232)		(1.407)	(1.492)		(1.376)	(1.499)
Sarpanch woman:Sarpanch relative present		-0.281	-1.057		5.033+	3.756		-1.894	-2.353
		(3.816)	(3.855)		(2.957)	(3.664)		(4.909)	(4.767)
Percent of village land held by biggest landholding caste			0.014			0.020			0.020
			(0.020)			(0.025)			(0.025)
Distance (km) from block center			0.013			0.043			-0.011
			(0.036)			(0.043)			(0.046)
Directly elected			-0.500			-1.004			-0.495
			(0.889)			(1.065)			(1.118)
T1				-0.819	-0.601	-0.567
				(1.127)	(1.167)	(1.218)
T1:Sarpanch SC				0.519	0.039	-0.296
				(1.875)	(2.049)	(2.134)
T2							0.838	0.467	0.364
							(1.180)	(1.221)	(1.295)
T2:Sarpanch SC							-0.487	0.439	0.550
							(2.084)	(2.238)	(2.330)
Num.Obs.	593	540	498	399	364	334	381	344	319
R2	0.002	0.032	0.040	0.005	0.041	0.054	0.003	0.052	0.068

CP2b (H3.13b) table, Sarpanch post- and collective group preferences, average difference; Gram sevak female
	Impact of T1/T2 combined			Impact of T1			Impact of T2
	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls
+ p < 0.1, * p < 0.05, p < 0.01, * p < 0.001
Each model is a linear regression of each outcome on an indicator for being assigned to the relevant treatment condition(s) versus not. Unadjusted and covariate-adjusted models presented. All regressions use robust standard errors.
(Intercept)	12.631***	14.254***	13.154***	12.631***	13.619***	11.717***	12.631***	14.031***	13.294***
	(0.869)	(1.775)	(2.421)	(0.869)	(2.049)	(2.948)	(0.869)	(2.169)	(2.931)
Discussion rules	-0.511	-0.911	-1.077
	(1.065)	(1.110)	(1.154)
gramsevak_female	-1.245	-1.544	-1.951	-1.245	-1.732	-2.208	-1.245	-1.544	-1.986
	(1.331)	(1.389)	(1.431)	(1.331)	(1.393)	(1.450)	(1.331)	(1.385)	(1.439)
Discussion rules:gramsevak_female	1.284	2.331	2.623
	(1.634)	(1.711)	(1.788)
Sarpanch woman		0.239	-0.605		-2.020	-2.339		2.991	1.774
		(2.574)	(2.800)		(2.685)	(2.796)		(3.543)	(4.110)
Sarpanch relative present		-3.940	-3.829		-8.419**	-8.024*		-3.836	-3.919
		(3.514)	(3.533)		(2.596)	(3.402)		(4.478)	(4.298)
Sarpanch SC		0.493	0.582		0.689	1.115		-1.577	-1.379
		(1.764)	(1.885)		(2.005)	(2.186)		(2.452)	(2.602)
Sarpanch OBC		-0.202	0.216		-0.719	-0.070		0.268	0.810
		(1.115)	(1.198)		(1.399)	(1.479)		(1.341)	(1.445)
Education above median		-1.491+	-1.704+		-1.284	-1.903+		-0.693	-0.842
		(0.861)	(0.905)		(1.072)	(1.103)		(1.083)	(1.141)
Wealth above median		0.326	-0.261		0.558	-0.113		0.047	-0.541
		(0.840)	(0.888)		(1.009)	(1.056)		(1.057)	(1.123)
Lives with in-laws		-1.600	-0.889		1.153	1.316		-3.590	-3.116
		(2.401)	(2.551)		(2.467)	(2.473)		(3.367)	(3.793)
Lives with own parents		-2.286*	-2.169*		-2.353*	-1.809		-3.007**	-3.568**
		(0.886)	(0.961)		(1.106)	(1.199)		(1.070)	(1.121)
Years as sarpanch		0.123	0.195		0.170	0.187		0.167	0.289
		(0.298)	(0.314)		(0.353)	(0.361)		(0.377)	(0.397)
Gender quota		1.163	1.313		1.441	1.624		0.583	0.989
		(1.145)	(1.274)		(1.341)	(1.488)		(1.394)	(1.548)
SC quota		0.493	0.981		0.037	0.048		2.985	3.696
		(1.738)	(1.862)		(1.928)	(2.117)		(2.418)	(2.576)
OBC quota		-0.190	-0.101		0.625	0.729		-0.660	-0.776
		(1.144)	(1.238)		(1.416)	(1.507)		(1.373)	(1.495)
Sarpanch woman:Sarpanch relative present		0.129	-0.573		5.373+	4.199		-1.213	-1.452
		(3.854)	(3.930)		(3.143)	(3.897)		(4.968)	(4.842)
Percent of village land held by biggest landholding caste			0.016			0.024			0.023
			(0.020)			(0.025)			(0.025)
Distance (km) from block center			0.012			0.046			-0.012
			(0.035)			(0.043)			(0.045)
Directly elected			-0.472			-0.957			-0.462
			(0.890)			(1.069)			(1.112)
T1				-1.200	-1.562	-1.656
				(1.206)	(1.282)	(1.329)
T1:gramsevak_female				1.335	2.425	2.519
				(1.818)	(1.941)	(2.040)
T2							0.275	-0.476	-0.777
							(1.255)	(1.322)	(1.413)
T2:gramsevak_female							1.159	2.632	3.181
							(1.963)	(2.080)	(2.197)
Num.Obs.	595	540	498	400	364	334	382	344	319
R2	0.001	0.035	0.044	0.004	0.046	0.060	0.003	0.057	0.075

CP3 (H3.14): The distance between (pre-treatment) private preferences of sarpanch and outcome of group decision making should be lower for treated (T1, T2) groups than for those not treated (Control).

CP3 is assessed using differences in the mean distance between the private (pre-treatment) preferences of the sarpanch versus the group decision, comparing the mean “preference difference” across treatment status (T1 or T2 versus Control). We use each of the three metrics above to run this comparison.
In additional analyses, we disentangle between T1 and T2.
We also include the series of individual and village level demographic controls presented in G2.
We also include a control for the potential presence of a sarpanch relative during the interview, both separately and in interaction with the gender of the sarpanch.
Deviation 1: We originally planed to use the Earth Mover’s Distance (Lupu et al 2017) as our primary outcome. After discussion with the developers of the measure, we decided not to use this measure for two reasons: 1) it does not provide additional information relative to simpler measures, and 2) the package to implement the measure is also not operational.
Deviation 2: We only collected group top priorities for 4/5 districts, hence the smaller sample size for the binary measure.
Deviation 3: We originally planned to use clustered standard errors at the taluka level. This was an error on our part: because the treatment assignment occurs at the level of each village council, we do not cluster standard errors in these analyses.

CP3 (H3.14) table, Sarpanch pre- and collective group preferences, average difference
	Impact of T1/T2 combined			Impact of T1			Impact of T2
	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls
+ p < 0.1, * p < 0.05, p < 0.01, * p < 0.001
Each model is a linear regression of each outcome on an indicator for being assigned to the relevant treatment condition(s) versus not. Unadjusted and covariate-adjusted models presented. All regressions use robust standard errors.
(Intercept)	22.513***	21.502***	26.707***	22.513***	22.069***	28.601***	22.513***	21.564***	25.437***
	(0.846)	(2.131)	(2.979)	(0.846)	(2.391)	(3.309)	(0.846)	(2.642)	(3.879)
Discussion rules	-2.200*	-2.353*	-2.722*
	(1.052)	(1.123)	(1.157)
Sarpanch woman		7.699*	5.423		6.493+	4.367		7.858+	5.941
		(3.238)	(3.321)		(3.360)	(3.362)		(4.383)	(4.653)
Sarpanch relative present		1.007	-0.247		-0.042	-3.308		3.117	1.979
		(6.250)	(6.581)		(10.013)	(10.754)		(7.245)	(7.392)
Sarpanch SC		2.254	1.714		-1.114	-1.064		4.790	4.128
		(2.415)	(2.485)		(2.974)	(3.071)		(3.261)	(3.186)
Sarpanch OBC		-0.628	-0.561		-4.049*	-2.987		1.499	1.164
		(1.528)	(1.606)		(1.753)	(1.827)		(1.891)	(2.053)
Education above median		-0.652	-1.555		-2.702*	-3.624**		1.210	0.598
		(1.116)	(1.141)		(1.333)	(1.352)		(1.377)	(1.422)
Wealth above median		-0.984	-1.456		1.201	0.553		-2.460+	-2.925*
		(1.079)	(1.122)		(1.286)	(1.338)		(1.307)	(1.377)
Lives with in-laws		-4.010	-2.966		-2.156	-2.246		-4.506	-2.453
		(2.929)	(2.878)		(2.971)	(2.757)		(3.999)	(4.123)
Lives with own parents		0.332	1.276		1.767	2.238		-0.936	0.105
		(1.213)	(1.272)		(1.434)	(1.503)		(1.479)	(1.571)
Years as sarpanch		0.030	-0.169		-0.363	-0.388		-0.128	-0.422
		(0.369)	(0.393)		(0.452)	(0.474)		(0.450)	(0.491)
Gender quota		-1.170	0.458		-1.039	1.042		-0.145	0.409
		(1.686)	(1.899)		(1.932)	(2.115)		(2.049)	(2.383)
SC quota		-0.530	-1.104		1.481	0.929		-2.438	-2.964
		(2.405)	(2.473)		(2.946)	(3.052)		(3.266)	(3.204)
OBC quota		-0.385	-0.079		1.567	1.394		-1.831	-1.348
		(1.526)	(1.588)		(1.754)	(1.803)		(1.894)	(2.027)
Sarpanch woman:Sarpanch relative present		-4.738	-4.280		-0.661	1.308		-6.754	-5.482
		(6.612)	(6.964)		(10.293)	(11.012)		(7.820)	(7.974)
Percent of village land held by biggest landholding caste			-0.081**			-0.103***			-0.072*
			(0.027)			(0.030)			(0.035)
Distance (km) from block center			0.058			0.072			0.072
			(0.045)			(0.054)			(0.064)
Directly elected			0.019			-0.701			0.751
			(1.203)			(1.437)			(1.495)
T1				-3.287**	-3.092*	-3.696**
				(1.204)	(1.268)	(1.327)
T2							-1.041	-1.042	-1.398
							(1.240)	(1.339)	(1.391)
Num.Obs.	583	528	487	389	353	324	376	338	314
R2	0.007	0.043	0.066	0.019	0.082	0.120	0.002	0.061	0.078

CP3 (H3.14) table, Sarpanch pre- priority and collective group preference percentage difference
	Impact of T1/T2 combined			Impact of T1			Impact of T2
	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls
+ p < 0.1, * p < 0.05, p < 0.01, * p < 0.001
Each model is a linear regression of each outcome on an indicator for being assigned to the relevant treatment condition(s) versus not. Unadjusted and covariate-adjusted models presented. All regressions use robust standard errors.
(Intercept)	48.190***	41.858***	60.679***	48.190***	45.627***	65.552***	48.190***	38.257***	54.233***
	(2.344)	(5.948)	(8.518)	(2.344)	(6.778)	(9.652)	(2.344)	(7.335)	(10.808)
Discussion rules	-2.924	-3.871	-5.683+
	(2.906)	(3.067)	(3.132)
Sarpanch woman		24.554**	18.437+		22.601*	17.431		26.155*	20.082
		(9.293)	(9.588)		(10.659)	(10.726)		(12.181)	(12.618)
Sarpanch relative present		2.664	-2.129		-4.990	-14.939		1.303	-3.094
		(14.835)	(16.907)		(21.548)	(24.704)		(17.683)	(19.278)
Sarpanch SC		0.952	-0.484		-11.077	-11.207		12.719	11.056
		(6.287)	(6.421)		(7.443)	(7.590)		(8.510)	(8.287)
Sarpanch OBC		-1.361	-1.061		-11.910*	-9.480+		8.271	7.758
		(4.096)	(4.351)		(4.801)	(5.013)		(5.080)	(5.561)
Education above median		-0.500	-3.005		-3.430	-6.278+		4.085	2.945
		(3.032)	(3.092)		(3.670)	(3.725)		(3.684)	(3.734)
Wealth above median		-2.271	-3.777		2.849	1.185		-6.390+	-8.703*
		(3.010)	(3.122)		(3.596)	(3.737)		(3.609)	(3.719)
Lives with in-laws		-8.753	-6.097		-4.960	-5.641		-10.370	-4.156
		(8.525)	(8.295)		(9.755)	(9.142)		(11.180)	(11.029)
Lives with own parents		1.393	3.359		2.615	3.590		0.240	2.185
		(3.307)	(3.495)		(3.954)	(4.165)		(4.086)	(4.353)
Years as sarpanch		0.500	0.182		-0.530	-0.368		0.210	-0.435
		(0.996)	(1.060)		(1.225)	(1.294)		(1.197)	(1.306)
Gender quota		-8.906+	-5.056		-9.888+	-5.233		-6.969	-5.300
		(4.754)	(5.418)		(5.466)	(6.117)		(5.846)	(6.685)
SC quota		5.818	4.347		12.231+	11.441		-0.035	-2.575
		(6.178)	(6.295)		(7.315)	(7.487)		(8.446)	(8.217)
OBC quota		0.524	0.832		5.837	5.186		-4.542	-4.585
		(4.068)	(4.318)		(4.630)	(4.825)		(5.060)	(5.488)
Sarpanch woman:Sarpanch relative present		-11.141	-8.376		4.877	12.399		-6.845	-2.023
		(16.056)	(18.038)		(22.644)	(25.641)		(19.749)	(21.064)
Percent of village land held by biggest landholding caste			-0.297***			-0.311***			-0.293**
			(0.072)			(0.086)			(0.089)
Distance (km) from block center			0.181			0.167			0.266
			(0.127)			(0.154)			(0.174)
Directly elected			0.784			-0.099			3.353
			(3.295)			(3.939)			(4.102)
T1				-5.460	-5.792+	-8.128*
				(3.345)	(3.500)	(3.638)
T2							-0.246	-0.874	-2.485
							(3.403)	(3.626)	(3.709)
Num.Obs.	594	538	498	397	360	332	386	347	323
R2	0.002	0.042	0.073	0.007	0.071	0.103	0.000	0.067	0.105

CP3 (H3.14) table, Sarpanch pre- priority and collective priority same
	Impact of T1/T2 combined			Impact of T1			Impact of T2
	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls
+ p < 0.1, * p < 0.05, p < 0.01, * p < 0.001
Each model is a linear regression of each outcome on an indicator for being assigned to the relevant treatment condition(s) versus not. Unadjusted and covariate-adjusted models presented. All regressions use robust standard errors.
(Intercept)	0.369***	0.383***	0.258+	0.369***	0.410***	0.141	0.369***	0.354**	0.261
	(0.039)	(0.096)	(0.133)	(0.039)	(0.109)	(0.158)	(0.039)	(0.122)	(0.179)
Discussion rules	0.135**	0.152**	0.158**
	(0.047)	(0.050)	(0.051)
Sarpanch woman		0.191	0.195		0.186	0.213		0.227	0.213
		(0.165)	(0.167)		(0.189)	(0.182)		(0.224)	(0.231)
Sarpanch relative present		-0.206	-0.200		-0.346***	-0.242		-0.221	-0.225
		(0.227)	(0.224)		(0.085)	(0.195)		(0.244)	(0.231)
Sarpanch SC		-0.071	-0.053		-0.191	-0.166		0.006	0.031
		(0.106)	(0.108)		(0.121)	(0.121)		(0.157)	(0.157)
Sarpanch OBC		0.087	0.099		0.099	0.093		0.081	0.091
		(0.067)	(0.069)		(0.080)	(0.081)		(0.087)	(0.093)
Education above median		-0.001	0.016		0.075	0.091		-0.045	-0.032
		(0.049)	(0.050)		(0.060)	(0.061)		(0.061)	(0.063)
Wealth above median		0.082+	0.068		0.062	0.063		0.118*	0.101
		(0.049)	(0.050)		(0.059)	(0.061)		(0.060)	(0.062)
Lives with in-laws		-0.245+	-0.242+		-0.184	-0.185		-0.177	-0.178
		(0.140)	(0.139)		(0.167)	(0.156)		(0.183)	(0.185)
Lives with own parents		-0.032	-0.047		-0.056	-0.069		-0.037	-0.048
		(0.058)	(0.059)		(0.069)	(0.069)		(0.075)	(0.077)
Years as sarpanch		-0.019	-0.020		-0.025	-0.030		-0.010	-0.013
		(0.016)	(0.018)		(0.020)	(0.021)		(0.021)	(0.023)
Gender quota		-0.061	-0.069		-0.123	-0.143		-0.182	-0.167
		(0.097)	(0.101)		(0.107)	(0.110)		(0.130)	(0.138)
SC quota		0.136	0.134		0.209+	0.202+		0.111	0.112
		(0.102)	(0.103)		(0.115)	(0.114)		(0.153)	(0.155)
OBC quota		-0.048	-0.054		-0.109	-0.099		-0.030	-0.042
		(0.066)	(0.067)		(0.077)	(0.078)		(0.086)	(0.091)
Sarpanch woman:Sarpanch relative present		0.340	0.367		0.521**	0.483*		0.274	0.301
		(0.254)	(0.251)		(0.158)	(0.234)		(0.289)	(0.275)
Percent of village land held by biggest landholding caste			0.001			0.003*			0.002
			(0.001)			(0.002)			(0.001)
Distance (km) from block center			0.001			0.001			-0.002
			(0.002)			(0.002)			(0.003)
Directly elected			0.057			0.060			0.061
			(0.053)			(0.063)			(0.065)
T1				0.146**	0.157**	0.165**
				(0.055)	(0.058)	(0.059)
T2							0.125*	0.147*	0.158**
							(0.055)	(0.060)	(0.061)
Num.Obs.	490	447	433	324	295	287	323	294	284
R2	0.016	0.056	0.065	0.021	0.092	0.116	0.016	0.066	0.074

CP4 (H3.15): [Heterogeneous treatment effects] the effect of the treatment, as implied by CP3, should be larger if the sarpanch belongs to a disadvantaged category (CP4a) or if the gram sevak is female (CP4b).

CP4 is assessed using a regression interacting leader and/or gram sevak characteristics with treatment status for each relevant dependent variable. Our focus is on the interaction term; we cluster on administrative block level. We use the same three measures of preference distance outlined in CP3. Two-tailed tests are used in all cases. We use each of the three metrics above to run this comparison.
In additional analyses, we disentangle between T1 and T2.
We also include the series of individual and village level demographic controls presented in G2.
We also include a control for the potential presence of a sarpanch relative during the interview, both separately and in interaction with the gender of the sarpanch.
Deviation 1: We originally planed to use the Earth Mover’s Distance (Lupu et al 2017) as our primary outcome. After discussion with the developers of the measure, we decided not to use this measure for two reasons: 1) it does not provide additional information relative to simpler measures, and 2) the package to implement the measure is also not operational.
Deviation 2: We only collected group top priorities for 4/5 districts, hence the smaller sample size for the binary measure.
Deviation 3: We originally planned to use clustered standard errors at the taluka level. This was an error on our part: because the treatment assignment occurs at the level of each village council, we do not cluster standard errors in these analyses.

CP4a (H3.15a) table, Sarpanch pre- and collective group preferences, average difference; Gender disadvantage
	Impact of T1/T2 combined			Impact of T1			Impact of T2
	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls
+ p < 0.1, * p < 0.05, p < 0.01, * p < 0.001
Each model is a linear regression of each outcome on an indicator for being assigned to the relevant treatment condition(s) versus not. Unadjusted and covariate-adjusted models presented. All regressions use robust standard errors.
(Intercept)	20.176***	19.639***	24.346***	20.176***	20.364***	26.144***	20.176***	20.466***	23.990***
	(1.505)	(2.469)	(3.275)	(1.505)	(2.702)	(3.563)	(1.505)	(2.879)	(4.123)
Discussion rules	-0.602	0.081	0.038
	(1.768)	(1.864)	(1.830)
Sarpanch woman	3.667*	10.548**	8.576*	3.667*	8.988*	7.359+	3.667*	9.593*	7.682
	(1.806)	(3.585)	(3.632)	(1.806)	(3.738)	(3.756)	(1.806)	(4.545)	(4.748)
Discussion rules:Sarpanch woman	-2.326	-3.978+	-4.519+
	(2.199)	(2.321)	(2.334)
Sarpanch relative present		1.328	0.202		0.616	-2.410		3.218	2.163
		(6.315)	(6.546)		(10.216)	(10.757)		(7.312)	(7.357)
Sarpanch SC		2.105	1.606		-1.324	-1.166		4.644	3.996
		(2.385)	(2.448)		(2.905)	(2.993)		(3.228)	(3.150)
Sarpanch OBC		-0.628	-0.548		-4.060*	-2.970		1.505	1.196
		(1.525)	(1.598)		(1.751)	(1.821)		(1.892)	(2.044)
Education above median		-0.669	-1.567		-2.652*	-3.573**		1.123	0.508
		(1.116)	(1.141)		(1.333)	(1.351)		(1.388)	(1.433)
Wealth above median		-1.098	-1.543		1.032	0.402		-2.547+	-2.979*
		(1.078)	(1.120)		(1.294)	(1.347)		(1.298)	(1.367)
Lives with in-laws		-3.961	-2.888		-2.094	-2.155		-4.469	-2.318
		(2.929)	(2.888)		(2.943)	(2.783)		(4.014)	(4.128)
Lives with own parents		0.397	1.322		1.886	2.349		-0.907	0.116
		(1.210)	(1.270)		(1.435)	(1.499)		(1.470)	(1.565)
Years as sarpanch		0.038	-0.167		-0.331	-0.376		-0.137	-0.428
		(0.370)	(0.393)		(0.455)	(0.476)		(0.451)	(0.491)
Gender quota		-1.249	0.445		-1.246	0.867		-0.097	0.568
		(1.692)	(1.904)		(1.950)	(2.129)		(2.055)	(2.399)
SC quota		-0.291	-0.894		1.880	1.285		-2.269	-2.800
		(2.376)	(2.438)		(2.876)	(2.965)		(3.236)	(3.177)
OBC quota		-0.285	0.001		1.694	1.547		-1.748	-1.314
		(1.521)	(1.577)		(1.741)	(1.777)		(1.894)	(2.019)
Sarpanch woman:Sarpanch relative present		-5.105	-4.785		-1.259	0.473		-6.982	-5.783
		(6.653)	(6.906)		(10.477)	(10.993)		(7.851)	(7.907)
Percent of village land held by biggest landholding caste			-0.079**			-0.101**			-0.070*
			(0.027)			(0.030)			(0.035)
Distance (km) from block center			0.064			0.084			0.076
			(0.045)			(0.054)			(0.064)
Directly elected			-0.002			-0.689			0.712
			(1.202)			(1.432)			(1.495)
T1				-1.942	-0.578	-0.646
				(1.954)	(2.016)	(1.983)
T1:Sarpanch woman				-1.911	-4.222	-5.207*
				(2.488)	(2.621)	(2.639)
T2							0.738	0.965	0.826
							(2.033)	(2.170)	(2.170)
T2:Sarpanch woman							-2.625	-3.381	-3.726
							(2.568)	(2.694)	(2.761)
Num.Obs.	583	528	487	389	353	324	376	338	314
R2	0.016	0.048	0.073	0.032	0.089	0.130	0.013	0.066	0.084

CP4a (H3.15a) table, Sarpanch pre- priority and collective group preference percentage difference; Gender disadvantage
	Impact of T1/T2 combined			Impact of T1			Impact of T2
	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls
+ p < 0.1, * p < 0.05, p < 0.01, * p < 0.001
Each model is a linear regression of each outcome on an indicator for being assigned to the relevant treatment condition(s) versus not. Unadjusted and covariate-adjusted models presented. All regressions use robust standard errors.
(Intercept)	41.750***	37.301***	54.675***	41.750***	41.763***	59.340***	41.750***	35.356***	50.527***
	(4.147)	(6.744)	(9.188)	(4.147)	(7.500)	(10.241)	(4.147)	(7.802)	(11.174)
Discussion rules	0.712	2.043	1.217
	(4.850)	(5.096)	(5.046)
Sarpanch woman	10.060*	31.470**	26.290*	10.060*	28.157*	24.791*	10.060*	30.749*	24.416+
	(4.994)	(10.246)	(10.398)	(4.994)	(11.596)	(11.601)	(4.994)	(12.641)	(12.898)
Discussion rules:Sarpanch woman	-4.920	-9.710	-11.333+
	(6.063)	(6.385)	(6.411)
Sarpanch relative present		3.430	-1.026		-3.558	-12.747		1.569	-2.640
		(14.384)	(16.092)		(21.337)	(23.838)		(17.252)	(18.488)
Sarpanch SC		0.630	-0.682		-11.484	-11.345		12.362	10.787
		(6.245)	(6.365)		(7.362)	(7.494)		(8.444)	(8.229)
Sarpanch OBC		-1.298	-0.936		-11.854*	-9.303+		8.333	7.909
		(4.103)	(4.346)		(4.815)	(5.018)		(5.100)	(5.554)
Education above median		-0.490	-2.976		-3.277	-6.113		3.936	2.813
		(3.033)	(3.090)		(3.674)	(3.726)		(3.706)	(3.756)
Wealth above median		-2.574	-4.032		2.473	0.797		-6.680+	-8.908*
		(3.018)	(3.127)		(3.624)	(3.770)		(3.604)	(3.708)
Lives with in-laws		-8.648	-5.911		-4.850	-5.450		-10.283	-3.831
		(8.505)	(8.282)		(9.666)	(9.124)		(11.175)	(10.974)
Lives with own parents		1.517	3.439		2.853	3.851		0.256	2.144
		(3.305)	(3.494)		(3.960)	(4.162)		(4.077)	(4.353)
Years as sarpanch		0.518	0.180		-0.458	-0.344		0.185	-0.456
		(0.998)	(1.061)		(1.236)	(1.300)		(1.198)	(1.307)
Gender quota		-9.109+	-5.096		-10.347+	-5.638		-6.886	-4.958
		(4.759)	(5.418)		(5.494)	(6.132)		(5.834)	(6.680)
SC quota		6.460	4.944		13.177+	12.390+		0.488	-2.071
		(6.130)	(6.238)		(7.217)	(7.366)		(8.372)	(8.170)
OBC quota		0.780	1.051		6.150	5.613		-4.323	-4.496
		(4.065)	(4.304)		(4.626)	(4.797)		(5.064)	(5.472)
Sarpanch woman:Sarpanch relative present		-11.942	-9.501		3.686	10.536		-7.396	-2.685
		(15.570)	(17.189)		(22.386)	(24.722)		(19.261)	(20.241)
Percent of village land held by biggest landholding caste			-0.291***			-0.305***			-0.286**
			(0.072)			(0.087)			(0.089)
Distance (km) from block center			0.196			0.195			0.275
			(0.127)			(0.153)			(0.174)
Directly elected			0.787			0.013			3.299
			(3.290)			(3.924)			(4.106)
T1				-2.904	-0.157	-0.553
				(5.373)	(5.588)	(5.594)
T1:Sarpanch woman				-3.154	-9.497	-12.958+
				(6.888)	(7.263)	(7.366)
T2							4.250	4.391	3.006
							(5.534)	(5.849)	(5.806)
T2:Sarpanch woman							-6.377	-8.960	-9.283
							(7.037)	(7.368)	(7.435)
Num.Obs.	594	538	498	397	360	332	386	347	323
R2	0.012	0.046	0.079	0.022	0.075	0.112	0.012	0.071	0.110

CP4a (H3.15a) table, Sarpanch pre- priority and collective priority same; Gender disadvantage
	Impact of T1/T2 combined			Impact of T1			Impact of T2
	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls
+ p < 0.1, * p < 0.05, p < 0.01, * p < 0.001
Each model is a linear regression of each outcome on an indicator for being assigned to the relevant treatment condition(s) versus not. Unadjusted and covariate-adjusted models presented. All regressions use robust standard errors.
(Intercept)	0.370***	0.384***	0.262+	0.370***	0.419***	0.151	0.370***	0.351**	0.260
	(0.066)	(0.107)	(0.142)	(0.066)	(0.118)	(0.166)	(0.066)	(0.126)	(0.182)
Discussion rules	0.151+	0.152+	0.153+
	(0.077)	(0.081)	(0.080)
Sarpanch woman	-0.001	0.190	0.190	-0.001	0.173	0.201	-0.001	0.231	0.215
	(0.082)	(0.183)	(0.184)	(0.082)	(0.202)	(0.195)	(0.082)	(0.232)	(0.239)
Discussion rules:Sarpanch woman	-0.032	0.001	0.008
	(0.098)	(0.103)	(0.103)
Sarpanch relative present		-0.206	-0.201		-0.354***	-0.249		-0.220	-0.225
		(0.230)	(0.228)		(0.093)	(0.197)		(0.246)	(0.232)
Sarpanch SC		-0.071	-0.053		-0.190	-0.166		0.006	0.031
		(0.107)	(0.108)		(0.121)	(0.121)		(0.158)	(0.158)
Sarpanch OBC		0.087	0.099		0.099	0.093		0.081	0.091
		(0.068)	(0.070)		(0.080)	(0.082)		(0.087)	(0.094)
Education above median		-0.001	0.016		0.075	0.091		-0.045	-0.032
		(0.049)	(0.050)		(0.060)	(0.061)		(0.061)	(0.063)
Wealth above median		0.082+	0.068		0.062	0.063		0.118*	0.101
		(0.049)	(0.050)		(0.059)	(0.061)		(0.060)	(0.062)
Lives with in-laws		-0.245+	-0.242+		-0.184	-0.184		-0.178	-0.178
		(0.140)	(0.139)		(0.167)	(0.156)		(0.183)	(0.186)
Lives with own parents		-0.032	-0.047		-0.056	-0.070		-0.037	-0.048
		(0.058)	(0.059)		(0.069)	(0.069)		(0.075)	(0.077)
Years as sarpanch		-0.019	-0.020		-0.025	-0.030		-0.010	-0.013
		(0.016)	(0.018)		(0.020)	(0.021)		(0.021)	(0.024)
Gender quota		-0.061	-0.069		-0.123	-0.143		-0.181	-0.167
		(0.097)	(0.101)		(0.107)	(0.111)		(0.130)	(0.138)
SC quota		0.136	0.134		0.207+	0.200+		0.111	0.112
		(0.102)	(0.103)		(0.116)	(0.114)		(0.154)	(0.156)
OBC quota		-0.048	-0.054		-0.109	-0.100		-0.030	-0.042
		(0.066)	(0.067)		(0.077)	(0.078)		(0.087)	(0.091)
Sarpanch woman:Sarpanch relative present		0.340	0.368		0.528**	0.490*		0.273	0.301
		(0.257)	(0.254)		(0.162)	(0.236)		(0.291)	(0.277)
Percent of village land held by biggest landholding caste			0.001			0.003*			0.002
			(0.001)			(0.002)			(0.001)
Distance (km) from block center			0.001			0.001			-0.002
			(0.002)			(0.002)			(0.003)
Directly elected			0.057			0.060			0.061
			(0.053)			(0.064)			(0.065)
T1				0.156+	0.144	0.153+
				(0.088)	(0.092)	(0.089)
T1:Sarpanch woman				-0.019	0.022	0.020
				(0.113)	(0.118)	(0.118)
T2							0.147+	0.152+	0.160+
							(0.085)	(0.091)	(0.090)
T2:Sarpanch woman							-0.047	-0.008	-0.003
							(0.113)	(0.120)	(0.122)
Num.Obs.	490	447	433	324	295	287	323	294	284
R2	0.017	0.056	0.065	0.022	0.093	0.116	0.017	0.066	0.074

CP4a (H3.15a) table, Sarpanch pre- and collective group preferences, average difference; Caste disadvantage
	Impact of T1/T2 combined			Impact of T1			Impact of T2
	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls
+ p < 0.1, * p < 0.05, p < 0.01, * p < 0.001
Each model is a linear regression of each outcome on an indicator for being assigned to the relevant treatment condition(s) versus not. Unadjusted and covariate-adjusted models presented. All regressions use robust standard errors.
(Intercept)	22.008***	21.513***	26.620***	22.008***	21.954***	28.472***	22.008***	21.668***	25.385***
	(1.022)	(2.198)	(3.041)	(1.022)	(2.452)	(3.382)	(1.022)	(2.684)	(3.907)
Discussion rules	-2.318+	-2.369+	-2.614+
	(1.293)	(1.365)	(1.414)
Sarpanch SC	1.613	2.214	1.974	1.613	-0.675	-0.687	1.613	4.469	4.243
	(1.826)	(2.878)	(2.923)	(1.826)	(3.328)	(3.408)	(1.826)	(3.461)	(3.433)
Discussion rules:Sarpanch SC	0.143	0.057	-0.376
	(2.233)	(2.395)	(2.445)
Sarpanch woman		7.700*	5.418		6.531+	4.386		7.905+	5.926
		(3.242)	(3.324)		(3.356)	(3.361)		(4.408)	(4.672)
Sarpanch relative present		1.010	-0.266		-0.132	-3.363		3.144	1.972
		(6.268)	(6.584)		(9.916)	(10.679)		(7.283)	(7.414)
Sarpanch OBC		-0.628	-0.560		-4.051*	-2.985		1.491	1.168
		(1.528)	(1.607)		(1.756)	(1.829)		(1.893)	(2.053)
Education above median		-0.652	-1.555		-2.702*	-3.622**		1.211	0.598
		(1.117)	(1.142)		(1.334)	(1.354)		(1.380)	(1.424)
Wealth above median		-0.983	-1.465		1.194	0.537		-2.449+	-2.933*
		(1.080)	(1.125)		(1.289)	(1.344)		(1.311)	(1.385)
Lives with in-laws		-4.010	-2.966		-2.155	-2.245		-4.519	-2.453
		(2.933)	(2.880)		(2.960)	(2.749)		(4.013)	(4.133)
Lives with own parents		0.332	1.275		1.761	2.228		-0.942	0.108
		(1.214)	(1.274)		(1.436)	(1.508)		(1.481)	(1.573)
Years as sarpanch		0.029	-0.168		-0.358	-0.387		-0.133	-0.420
		(0.370)	(0.393)		(0.454)	(0.474)		(0.453)	(0.492)
Gender quota		-1.170	0.462		-1.076	1.025		-0.180	0.423
		(1.687)	(1.903)		(1.949)	(2.130)		(2.059)	(2.397)
SC quota		-0.530	-1.104		1.442	0.892		-2.480	-2.947
		(2.407)	(2.477)		(2.956)	(3.061)		(3.269)	(3.207)
OBC quota		-0.384	-0.090		1.550	1.373		-1.806	-1.358
		(1.532)	(1.596)		(1.760)	(1.810)		(1.911)	(2.047)
Sarpanch woman:Sarpanch relative present		-4.744	-4.235		-0.547	1.397		-6.833	-5.449
		(6.636)	(6.981)		(10.219)	(10.955)		(7.860)	(8.023)
Percent of village land held by biggest landholding caste			-0.081**			-0.102***			-0.072*
			(0.027)			(0.030)			(0.034)
Distance (km) from block center			0.058			0.072			0.072
			(0.045)			(0.054)			(0.064)
Directly elected			0.015			-0.703			0.745
			(1.203)			(1.439)			(1.487)
T1				-3.523*	-2.885+	-3.512*
				(1.479)	(1.515)	(1.590)
T1:Sarpanch SC				0.574	-0.760	-0.656
				(2.542)	(2.792)	(2.908)
T2							-0.915	-1.242	-1.324
							(1.553)	(1.671)	(1.745)
T2:Sarpanch SC							-0.563	0.666	-0.242
							(2.609)	(2.794)	(2.850)
Num.Obs.	581	528	487	388	353	324	375	338	314
R2	0.012	0.043	0.066	0.025	0.082	0.120	0.005	0.061	0.078

CP4a (H3.15a) table, Sarpanch pre- priority and collective group preference percentage difference; Caste disadvantage
	Impact of T1/T2 combined			Impact of T1			Impact of T2
	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls
+ p < 0.1, * p < 0.05, p < 0.01, * p < 0.001
Each model is a linear regression of each outcome on an indicator for being assigned to the relevant treatment condition(s) versus not. Unadjusted and covariate-adjusted models presented. All regressions use robust standard errors.
(Intercept)	46.229***	41.213***	59.909***	46.229***	44.124***	63.868***	46.229***	38.594***	54.320***
	(2.739)	(6.090)	(8.613)	(2.739)	(6.885)	(9.722)	(2.739)	(7.436)	(10.830)
Discussion rules	-2.546	-2.941	-4.758
	(3.471)	(3.643)	(3.766)
Sarpanch SC	6.392	3.274	1.752	6.392	-5.418	-6.428	6.392	11.687	10.872
	(5.235)	(7.751)	(7.766)	(5.235)	(8.787)	(8.938)	(5.235)	(9.162)	(8.964)
Discussion rules:Sarpanch SC	-1.956	-3.332	-3.235
	(6.364)	(6.749)	(6.732)
Sarpanch woman		24.541**	18.397+		23.148*	17.700+		26.304*	20.106
		(9.261)	(9.569)		(10.285)	(10.424)		(12.267)	(12.660)
Sarpanch relative present		2.465	-2.270		-6.152	-15.638		1.375	-3.088
		(14.789)	(16.879)		(20.130)	(23.431)		(17.755)	(19.352)
Sarpanch OBC		-1.370	-1.059		-11.966*	-9.485+		8.247	7.752
		(4.097)	(4.351)		(4.817)	(5.014)		(5.090)	(5.570)
Education above median		-0.510	-3.007		-3.436	-6.247+		4.087	2.945
		(3.032)	(3.092)		(3.656)	(3.718)		(3.691)	(3.740)
Wealth above median		-2.297	-3.840		2.772	1.019		-6.359+	-8.692*
		(3.015)	(3.132)		(3.606)	(3.762)		(3.623)	(3.743)
Lives with in-laws		-8.724	-6.095		-4.935	-5.612		-10.415	-4.156
		(8.484)	(8.258)		(9.338)	(8.773)		(11.222)	(11.046)
Lives with own parents		1.379	3.340		2.492	3.435		0.233	2.183
		(3.309)	(3.502)		(3.948)	(4.176)		(4.091)	(4.357)
Years as sarpanch		0.523	0.200		-0.443	-0.328		0.191	-0.439
		(0.997)	(1.059)		(1.223)	(1.289)		(1.203)	(1.308)
Gender quota		-8.907+	-5.007		-10.401+	-5.463		-7.085	-5.324
		(4.773)	(5.445)		(5.531)	(6.187)		(5.874)	(6.704)
SC quota		5.809	4.344		11.696	10.949		-0.166	-2.603
		(6.188)	(6.307)		(7.346)	(7.544)		(8.451)	(8.223)
OBC quota		0.450	0.745		5.635	4.947		-4.459	-4.569
		(4.081)	(4.336)		(4.665)	(4.848)		(5.097)	(5.533)
Sarpanch woman:Sarpanch relative present		-10.810	-8.036		6.251	13.462		-7.074	-2.069
		(16.089)	(18.103)		(21.445)	(24.525)		(19.880)	(21.272)
Percent of village land held by biggest landholding caste			-0.294***			-0.303***			-0.293**
			(0.072)			(0.087)			(0.088)
Distance (km) from block center			0.183			0.169			0.266
			(0.127)			(0.153)			(0.175)
Directly elected			0.726			-0.222			3.366
			(3.290)			(3.933)			(4.088)
T1				-4.376	-3.124	-5.813
				(4.009)	(4.083)	(4.274)
T1:Sarpanch SC				-4.135	-9.863	-8.358
				(7.295)	(7.822)	(8.036)
T2							-0.453	-1.511	-2.602
							(4.147)	(4.418)	(4.585)
T2:Sarpanch SC							-0.435	2.145	0.388
							(7.336)	(7.811)	(7.734)
Num.Obs.	592	538	498	396	360	332	385	347	323
R2	0.007	0.042	0.074	0.011	0.075	0.106	0.008	0.067	0.105

CP4a (H3.15a) table, Sarpanch pre- priority and collective priority same; Caste disadvantage
	Impact of T1/T2 combined			Impact of T1			Impact of T2
	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls
+ p < 0.1, * p < 0.05, p < 0.01, * p < 0.001
Each model is a linear regression of each outcome on an indicator for being assigned to the relevant treatment condition(s) versus not. Unadjusted and covariate-adjusted models presented. All regressions use robust standard errors.
(Intercept)	0.352***	0.385***	0.257+	0.352***	0.402***	0.120	0.352***	0.367**	0.283
	(0.046)	(0.098)	(0.135)	(0.046)	(0.110)	(0.159)	(0.046)	(0.122)	(0.182)
Discussion rules	0.146*	0.150*	0.159**
	(0.057)	(0.059)	(0.060)
Sarpanch SC	0.056	-0.078	-0.050	0.056	-0.147	-0.092	0.056	-0.052	-0.032
	(0.085)	(0.130)	(0.132)	(0.085)	(0.140)	(0.142)	(0.085)	(0.165)	(0.164)
Discussion rules:Sarpanch SC	-0.040	0.009	-0.004
	(0.103)	(0.110)	(0.112)
Sarpanch woman		0.191	0.195		0.182	0.205		0.241	0.233
		(0.166)	(0.168)		(0.187)	(0.178)		(0.225)	(0.233)
Sarpanch relative present		-0.206	-0.200		-0.353***	-0.250		-0.218	-0.224
		(0.228)	(0.225)		(0.100)	(0.227)		(0.252)	(0.239)
Sarpanch OBC		0.087	0.099		0.099	0.093		0.081	0.090
		(0.067)	(0.069)		(0.080)	(0.082)		(0.087)	(0.093)
Education above median		-0.001	0.016		0.075	0.091		-0.045	-0.032
		(0.049)	(0.050)		(0.060)	(0.062)		(0.061)	(0.063)
Wealth above median		0.082+	0.068		0.060	0.060		0.122*	0.105+
		(0.049)	(0.050)		(0.059)	(0.061)		(0.060)	(0.061)
Lives with in-laws		-0.245+	-0.242+		-0.180	-0.177		-0.183	-0.184
		(0.140)	(0.139)		(0.165)	(0.152)		(0.184)	(0.187)
Lives with own parents		-0.032	-0.047		-0.058	-0.072		-0.035	-0.047
		(0.058)	(0.059)		(0.069)	(0.069)		(0.075)	(0.077)
Years as sarpanch		-0.019	-0.020		-0.024	-0.029		-0.011	-0.015
		(0.016)	(0.018)		(0.020)	(0.021)		(0.021)	(0.023)
Gender quota		-0.061	-0.069		-0.123	-0.141		-0.189	-0.180
		(0.097)	(0.101)		(0.107)	(0.109)		(0.130)	(0.139)
SC quota		0.137	0.134		0.200+	0.187		0.112	0.112
		(0.102)	(0.103)		(0.117)	(0.116)		(0.150)	(0.152)
OBC quota		-0.048	-0.054		-0.111	-0.103		-0.027	-0.039
		(0.066)	(0.067)		(0.077)	(0.079)		(0.086)	(0.091)
Sarpanch woman:Sarpanch relative present		0.339	0.367		0.532**	0.498+		0.260	0.288
		(0.255)	(0.252)		(0.168)	(0.262)		(0.296)	(0.283)
Percent of village land held by biggest landholding caste			0.001			0.003*			0.002
			(0.001)			(0.002)			(0.001)
Distance (km) from block center			0.001			0.001			-0.002
			(0.002)			(0.002)			(0.003)
Directly elected			0.057			0.057			0.068
			(0.053)			(0.064)			(0.066)
T1				0.175**	0.177**	0.197**
				(0.066)	(0.068)	(0.069)
T1:Sarpanch SC				-0.102	-0.071	-0.117
				(0.119)	(0.129)	(0.130)
T2							0.117+	0.117+	0.124+
							(0.066)	(0.070)	(0.073)
T2:Sarpanch SC							0.021	0.105	0.119
							(0.118)	(0.126)	(0.128)
Num.Obs.	489	447	433	323	295	287	323	294	284
R2	0.017	0.056	0.065	0.023	0.093	0.119	0.020	0.068	0.077

CP4 (H3.15b) table, Sarpanch pre- and collective group preferences, average difference; Gram sevak female
	Impact of T1/T2 combined			Impact of T1			Impact of T2
	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls
+ p < 0.1, * p < 0.05, p < 0.01, * p < 0.001
Each model is a linear regression of each outcome on an indicator for being assigned to the relevant treatment condition(s) versus not. Unadjusted and covariate-adjusted models presented. All regressions use robust standard errors.
(Intercept)	23.484***	22.827***	27.750***	23.484***	23.329***	29.464***	23.484***	22.731***	26.482***
	(1.118)	(2.293)	(3.044)	(1.118)	(2.530)	(3.315)	(1.118)	(2.811)	(3.912)
Discussion rules	-2.423+	-3.137*	-3.564*
	(1.396)	(1.476)	(1.534)
gramsevak_female	-2.524	-3.113+	-3.227+	-2.524	-3.236+	-3.151+	-2.524	-3.045+	-3.401+
	(1.692)	(1.802)	(1.842)	(1.692)	(1.816)	(1.855)	(1.692)	(1.827)	(1.881)
Discussion rules:gramsevak_female	0.820	2.180	2.243
	(2.106)	(2.224)	(2.289)
Sarpanch woman		7.947*	5.746+		7.649*	5.608+		8.344+	6.538
		(3.240)	(3.352)		(3.319)	(3.336)		(4.361)	(4.674)
Sarpanch relative present		0.689	-0.551		-1.206	-4.459		2.336	1.105
		(6.367)	(6.679)		(10.188)	(10.727)		(7.318)	(7.449)
Sarpanch SC		2.301	1.822		-0.891	-0.724		4.863	4.273
		(2.405)	(2.474)		(2.945)	(3.038)		(3.232)	(3.160)
Sarpanch OBC		-0.606	-0.479		-3.895*	-2.722		1.431	1.177
		(1.536)	(1.625)		(1.773)	(1.864)		(1.909)	(2.089)
Education above median		-0.706	-1.602		-2.759*	-3.676**		1.111	0.522
		(1.113)	(1.139)		(1.326)	(1.341)		(1.374)	(1.416)
Wealth above median		-0.968	-1.491		1.208	0.493		-2.445+	-3.002*
		(1.075)	(1.117)		(1.273)	(1.322)		(1.302)	(1.373)
Lives with in-laws		-4.068	-3.083		-2.770	-2.933		-4.577	-2.542
		(2.942)	(2.932)		(2.927)	(2.736)		(3.953)	(4.105)
Lives with own parents		0.207	1.123		1.628	2.075		-1.117	-0.160
		(1.208)	(1.267)		(1.413)	(1.473)		(1.480)	(1.579)
Years as sarpanch		0.049	-0.143		-0.296	-0.324		-0.066	-0.349
		(0.370)	(0.394)		(0.448)	(0.473)		(0.451)	(0.491)
Gender quota		-1.395	0.155		-1.646	0.313		-0.584	-0.235
		(1.694)	(1.918)		(1.975)	(2.175)		(2.096)	(2.484)
SC quota		-0.821	-1.426		0.915	0.288		-2.581	-3.117
		(2.414)	(2.485)		(2.953)	(3.061)		(3.255)	(3.184)
OBC quota		-0.491	-0.203		1.303	1.055		-1.853	-1.369
		(1.540)	(1.618)		(1.778)	(1.858)		(1.912)	(2.070)
Sarpanch woman:Sarpanch relative present		-4.530	-4.067		0.393	2.440		-5.977	-4.525
		(6.745)	(7.077)		(10.472)	(10.989)		(7.951)	(8.094)
Percent of village land held by biggest landholding caste			-0.078**			-0.100***			-0.068*
			(0.027)			(0.030)			(0.034)
Distance (km) from block center			0.065			0.084			0.074
			(0.046)			(0.055)			(0.065)
Directly elected			0.063			-0.636			0.838
			(1.202)			(1.421)			(1.504)
T1				-2.877+	-3.220+	-3.742*
				(1.585)	(1.672)	(1.756)
T1:gramsevak_female				-0.687	0.554	0.131
				(2.413)	(2.516)	(2.585)
T2							-1.923	-2.643	-3.159+
							(1.679)	(1.789)	(1.898)
T2:gramsevak_female							2.326	4.060	4.378
							(2.481)	(2.652)	(2.790)
Num.Obs.	583	528	487	389	353	324	376	338	314
R2	0.013	0.048	0.072	0.033	0.096	0.134	0.007	0.069	0.088

CP4 (H3.15b) table, Sarpanch pre- priority and collective group preference percentage difference; Gram sevak female
	Impact of T1/T2 combined			Impact of T1			Impact of T2
	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls
+ p < 0.1, * p < 0.05, p < 0.01, * p < 0.001
Each model is a linear regression of each outcome on an indicator for being assigned to the relevant treatment condition(s) versus not. Unadjusted and covariate-adjusted models presented. All regressions use robust standard errors.
(Intercept)	48.905***	43.223***	61.661***	48.905***	47.094***	66.422***	48.905***	39.293***	55.313***
	(3.067)	(6.399)	(8.727)	(3.067)	(7.215)	(9.804)	(3.067)	(7.816)	(10.978)
Discussion rules	-2.134	-4.827	-6.772
	(3.844)	(4.036)	(4.127)
gramsevak_female	-1.850	-3.206	-2.801	-1.850	-3.576	-2.830	-1.850	-2.859	-3.339
	(4.758)	(5.042)	(5.090)	(4.758)	(5.141)	(5.193)	(4.758)	(5.049)	(5.102)
Discussion rules:gramsevak_female	-1.572	2.537	2.732
	(5.878)	(6.196)	(6.335)
Sarpanch woman		24.774**	18.611+		23.409*	18.008+		26.703*	20.733
		(9.297)	(9.623)		(10.658)	(10.793)		(12.127)	(12.639)
Sarpanch relative present		2.284	-2.531		-6.135	-15.767		0.428	-4.098
		(15.038)	(17.034)		(21.711)	(24.726)		(17.821)	(19.423)
Sarpanch SC		0.999	-0.429		-10.945	-11.047		12.811	11.220
		(6.284)	(6.425)		(7.433)	(7.610)		(8.461)	(8.251)
Sarpanch OBC		-1.333	-1.023		-11.813*	-9.347+		8.240	7.802
		(4.120)	(4.398)		(4.850)	(5.085)		(5.111)	(5.623)
Education above median		-0.519	-3.011		-3.414	-6.264+		4.059	2.948
		(3.033)	(3.096)		(3.666)	(3.723)		(3.688)	(3.742)
Wealth above median		-2.258	-3.813		2.848	1.126		-6.390+	-8.796*
		(3.013)	(3.126)		(3.600)	(3.747)		(3.618)	(3.734)
Lives with in-laws		-8.794	-6.108		-5.324	-5.885		-10.484	-4.270
		(8.534)	(8.340)		(9.727)	(9.176)		(11.065)	(10.932)
Lives with own parents		1.239	3.200		2.366	3.362		0.053	1.896
		(3.325)	(3.516)		(3.983)	(4.189)		(4.132)	(4.400)
Years as sarpanch		0.517	0.201		-0.474	-0.329		0.267	-0.373
		(1.001)	(1.065)		(1.229)	(1.302)		(1.205)	(1.313)
Gender quota		-9.154+	-5.325		-10.449+	-5.739		-7.451	-6.013
		(4.770)	(5.460)		(5.529)	(6.213)		(5.910)	(6.891)
SC quota		5.593	4.228		11.835	11.140		-0.063	-2.618
		(6.212)	(6.353)		(7.369)	(7.582)		(8.431)	(8.201)
OBC quota		0.468	0.823		5.707	5.095		-4.498	-4.535
		(4.100)	(4.383)		(4.678)	(4.913)		(5.101)	(5.561)
Sarpanch woman:Sarpanch relative present		-10.857	-7.972		5.903	13.228		-5.914	-0.864
		(16.272)	(18.176)		(22.801)	(25.649)		(19.973)	(21.296)
Percent of village land held by biggest landholding caste			-0.295***			-0.308***			-0.290**
			(0.072)			(0.087)			(0.089)
Distance (km) from block center			0.184			0.173			0.266
			(0.129)			(0.155)			(0.178)
Directly elected			0.845			-0.014			3.477
			(3.306)			(3.945)			(4.135)
T1				-4.703	-6.557	-8.870+
				(4.369)	(4.547)	(4.713)
T1:gramsevak_female				-1.587	2.018	1.801
				(6.796)	(7.103)	(7.277)
T2							0.697	-2.610	-4.426
							(4.622)	(4.926)	(5.098)
T2:gramsevak_female							-1.819	4.262	4.688
							(6.847)	(7.255)	(7.546)
Num.Obs.	594	538	498	397	360	332	386	347	323
R2	0.004	0.042	0.074	0.008	0.072	0.104	0.002	0.068	0.107

CP4 (H3.15b) table, Sarpanch pre- priority and collective priority same; Gram sevak female
	Impact of T1/T2 combined			Impact of T1			Impact of T2
	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls	Unadjusted	Adjusted, demographic controls	Adjusted, demographic and village controls
+ p < 0.1, * p < 0.05, p < 0.01, * p < 0.001
Each model is a linear regression of each outcome on an indicator for being assigned to the relevant treatment condition(s) versus not. Unadjusted and covariate-adjusted models presented. All regressions use robust standard errors.
(Intercept)	0.326***	0.320**	0.218	0.326***	0.348**	0.103	0.326***	0.277*	0.220
	(0.049)	(0.101)	(0.133)	(0.049)	(0.114)	(0.160)	(0.049)	(0.125)	(0.176)
Discussion rules	0.149*	0.168*	0.163*
	(0.062)	(0.065)	(0.066)
gramsevak_female	0.105	0.133	0.123	0.105	0.129	0.108	0.105	0.125	0.124
	(0.079)	(0.082)	(0.083)	(0.079)	(0.083)	(0.083)	(0.079)	(0.084)	(0.085)
Discussion rules:gramsevak_female	-0.040	-0.043	-0.017
	(0.096)	(0.101)	(0.102)
Sarpanch woman		0.178	0.182		0.182	0.200		0.253	0.248
		(0.167)	(0.169)		(0.191)	(0.186)		(0.223)	(0.229)
Sarpanch relative present		-0.203	-0.206		-0.290**	-0.197		-0.241	-0.257
		(0.221)	(0.219)		(0.098)	(0.201)		(0.238)	(0.228)
Sarpanch SC		-0.080	-0.064		-0.198	-0.177		-0.009	0.014
		(0.106)	(0.107)		(0.120)	(0.121)		(0.156)	(0.154)
Sarpanch OBC		0.080	0.089		0.098	0.089		0.074	0.080
		(0.067)	(0.069)		(0.079)	(0.082)		(0.085)	(0.091)
Education above median		0.001	0.016		0.077	0.092		-0.041	-0.031
		(0.049)	(0.050)		(0.060)	(0.061)		(0.061)	(0.063)
Wealth above median		0.085+	0.070		0.064	0.065		0.125*	0.107+
		(0.049)	(0.050)		(0.058)	(0.061)		(0.059)	(0.060)
Lives with in-laws		-0.241+	-0.236+		-0.186	-0.181		-0.202	-0.211
		(0.141)	(0.141)		(0.169)	(0.160)		(0.184)	(0.187)
Lives with own parents		-0.024	-0.040		-0.046	-0.061		-0.020	-0.034
		(0.058)	(0.059)		(0.069)	(0.069)		(0.074)	(0.076)
Years as sarpanch		-0.022	-0.023		-0.028	-0.033		-0.013	-0.018
		(0.016)	(0.018)		(0.020)	(0.022)		(0.021)	(0.024)
Gender quota		-0.042	-0.052		-0.106	-0.123		-0.168	-0.159
		(0.097)	(0.100)		(0.106)	(0.110)		(0.127)	(0.134)
SC quota		0.161	0.162		0.225+	0.219+		0.159	0.162
		(0.102)	(0.102)		(0.115)	(0.114)		(0.151)	(0.152)
OBC quota		-0.040	-0.045		-0.108	-0.097		-0.017	-0.033
		(0.066)	(0.067)		(0.077)	(0.078)		(0.084)	(0.088)
Sarpanch woman:Sarpanch relative present		0.335	0.375		0.460**	0.430+		0.307	0.352
		(0.248)	(0.245)		(0.165)	(0.239)		(0.278)	(0.267)
Percent of village land held by biggest landholding caste			0.001			0.003*			0.002
			(0.001)			(0.002)			(0.001)
Distance (km) from block center			0.000			0.001			-0.003
			(0.002)			(0.002)			(0.003)
Directly elected			0.060			0.060			0.066
			(0.053)			(0.064)			(0.066)
T1				0.184**	0.212**	0.201**
				(0.071)	(0.076)	(0.076)
T1:gramsevak_female				-0.094	-0.123	-0.078
				(0.112)	(0.116)	(0.117)
T2							0.111	0.111	0.112
							(0.073)	(0.078)	(0.081)
T2:gramsevak_female							0.015	0.056	0.082
							(0.111)	(0.118)	(0.119)
Num.Obs.	490	447	433	324	295	287	323	294	284
R2	0.022	0.066	0.076	0.027	0.100	0.122	0.029	0.089	0.101

Why Discussion Rules Matter: Populated pre-analysis plan

Rachel Brulé, Alyssa Heinze, Simon Chauchard