“Land Scam” Appendix Figure 2

Online Appendix Contents

Appendix Figure 1: Flow Chart of the Andhra Pradesh “Land Scam”

Appendix Figure 2: Scatterplot of Monthly Cement Production, as Reported by CMA and the Government of India

Appendix Figure 3: Coefficient Plot of Main Result

Appendix Figure 4: Randomization Test Results

Appendix Table 1: Wooldridge Test for Auto-Correlation in Linear Panel Data

Appendix Table 2: AIC Test for Optimal Lag Length

Appendix Table 3: Im, Pesaran and Shin (IPS) Test for Unit Roots

Appendix Table 4: Cement Consumption and State Elections, Adding Lags and Leads

Appendix Table 5: Cement Consumption and National Elections, Adding Lags and Leads

Appendix Table 6: Cement Consumption and State Elections, Adding 12-month Lags and Leads

Appendix Table 7: Cement Consumption and State Elections, Four Lags of the Dependent Variable

Appendix Table 8: Cement Consumption and National Elections, Four Lags of the Dependent Variable

Appendix Table 9: Cement Consumption and State Elections, No Lags of the Dependent Variable

Appendix Table 10: Cement Consumption and State Elections, No Lags of the Dependent Variable

Appendix Table 11: Cement Production and State Elections, Adding Lags and Leads

Appendix Table 12: Cement Production and State Elections, Dropping Delhi and Haryana

Appendix Table 13: Cement Consumption and State Elections, Dropping Delhi and Haryana and Adding Lags and Leads

Appendix Table 14: Cement Consumption and National Elections, Controlling for IIP

Appendix Table 15: Randomization Test Statistics Appendix Table 16: Additional Examples of Land “Scams” from India’s States

Appendix Figure 1: Flow Chart of the Andhra Pradesh “Land Scam”

Appendix Figure 2: Scatterplot of Monthly Cement Production, as Reported by CMA and the Government of India

Note: The y-axis represents monthly cement production, as reported by CMA. The x-axis represents monthly cement production data, as reported by the Government of India. The data are at the all-India level and span the period between April 1999 and March 2010.

Appendix Figure 3: Coefficient Plot of Main Result

Note: Baseline model used to obtain original estimated t-statistic adapted from Column 4 of Table 2.

Appendix Figure 4: Randomization Test Results

To build confidence in our result, we make use of randomization inference (Fisher 1935; Rosenbaum 2002). Randomization inference is relevant to our case, as we are working with panel data where we are likely to have “clustering” among error terms within states. In the models described in the chapter, we addressed this issue using panel-corrected standard errors (PCSEs). Yet, we might be concerned about the robustness of these estimates as we have a relatively small number of states. Thus, we randomly shuffle the election month variable and re-estimate our model 1,000 times to create a reference distribution of t-statistics that would arise if the null hypothesis were true (that there is no statistically significant relationship between state elections and a decline in cement consumption).

We use the model in Column 4 of Table 2 as our baseline regression, but without using the correction for panel-corrected standard errors. The t-statistic on the election variable is 4.08 (see the Online Appendix for a graphic demonstration of the reference distribution of t- statistics obtained from the randomization test). The vertical reference line indicates the t- statistic on our baseline model. As the figure demonstrates, more than 95 percent of the time we obtain results that are of lesser statistical significance than in our baseline model. Thus, we are confident in the robustness of our finding.

Appendix Table 1: Wooldridge Test for Auto-Correlation in Linear Panel Data

Wooldridge test: H0 is no first order autocorrelation F(1,16) = 59.125 P>F = 0

Note: Test statistic obtained using the Stata command, -xtserial-.

Appendix Table 2: AIC Test for Optimal Lag Length

Optimal State AIC Lag Andhra Pradesh 0.916 4 Assam 0.504 4 Bihar -0.624 4 Delhi -1.532 3 Gujarat -0.660 1 Haryana -1.450 4 Jammu & Kashmir -0.388 4 Karnataka -1.698 3 Kerala -1.598 4 Madhya Pradesh -0.826 3 Maharashtra -1.499 3 Orissa -0.363 1 Punjab -1.592 3 Rajasthan -0.854 4 Tamil Nadu -1.312 3 Uttar Pradesh -1.352 3 West Bengal -0.923 2

Note: Akaike's information criterion (AIC) based on test statistics calculated using the Stata – varsoc- command.

Appendix Table 3: Im, Pesaran and Shin (IPS) Test for Unit Roots

Time Lags trend? W-t-bar p-value 3 Yes -16.8215 0 4 Yes -14.8001 0

Note: IPS test statistic obtained using the Stata command, -ipshin-. Based on the mean of the individual Dickey-Fuller t-statistics of each unit in the panel, the IPS test assumes that all series are non-stationary under the null hypothesis.

Appendix Table 4: Cement Consumption and State Elections, Adding Lags and Leads

(1) (2) (3) (4) (5) (6) Log cement consumpti Log cement Log cement Log cement Log cement Log cement DV: on consumption consumption consumption consumption consumption

Log cement consumptiont-1 0.34*** 0.33*** 0.33*** 0.33*** 0.33*** 0.33*** [10.73] [10.63] [10.54] [10.47] [10.28] [10.02] Log cement consumptiont-2 0.21*** 0.22*** 0.21*** 0.22*** 0.21*** 0.22*** [6.35] [6.46] [6.39] [6.39] [6.30] [6.33] Log cement consumptiont-3 0.16*** 0.16*** 0.16*** 0.16*** 0.16*** 0.16*** [5.04] [4.96] [5.06] [5.08] [5.09] [4.94] Electiont-6 0.06*** [2.69] Electiont-5 -0.00 -0.00 [-0.04] [-0.03] Electiont-4 -0.01 -0.01 -0.02 [-0.66] [-0.62] [-0.69] Electiont-3 -0.03 -0.03 -0.03 -0.03 [-1.39] [-1.43] [-1.55] [-1.55] Electiont-2 0.01 0.01 0.01 0.01 0.01 [0.46] [0.48] [0.53] [0.50] [0.55] Electiont-1 0.00 0.00 0.00 0.00 0.00 0.00 [0.10] [0.09] [0.13] [0.15] [0.20] [0.21] Election -0.12*** -0.12*** -0.12*** -0.12*** -0.12*** -0.12*** [-5.46] [-5.46] [-5.45] [-5.48] [-5.47] [-5.44] Electiont+1 0.03 0.03 0.03 0.03 0.03 0.03 [1.42] [1.43] [1.42] [1.49] [1.34] [1.29] Electiont+2 0.03 0.03 0.03 0.03 0.03 [1.39] [1.32] [1.26] [1.22] [1.17] Electiont+3 0.04 0.03 0.03 0.04 [1.56] [1.49] [1.54] [1.56] Electiont+4 0.01 0.01 0.01 [0.46] [0.53] [0.63] Electiont+5 0.04* 0.04* [1.72] [1.82] Electiont+6 0.00 [0.20]

Time & State- Time & Time & Time & Time & Time & Fixed effects Month State-Month State-Month State-Month State-Month State-Month Observations 2,992 2,975 2,958 2,924 2,890 2,856 R-squared 0.97 0.97 0.97 0.97 0.97 0.97 Number of states 17 17 17 17 17 17

Note: Z statistics in brackets. * significant at 10%; ** significant at 5%; *** significant at 1%. All models include three lags of the dependent variable, time fixed effects, and fixed effects for each state-month combination. Models are estimated using OLS with panel-corrected standard errors. Dependent variable is natural log of cement consumption.

Appendix Table 5: Cement Consumption and National Elections, Adding Lags and Leads

(1) (2) (3) (4) (5) (6) Log cement Log cement Log cement Log cement Log cement Log cement DV: consumption consumption consumption consumption consumption consumption

Log cement consumptiont-1 0.34*** 0.33*** 0.33*** 0.33*** 0.33*** 0.33*** [11.17] [11.06] [10.99] [10.93] [10.56] [10.31] Log cement consumptiont-2 0.19*** 0.19*** 0.19*** 0.20*** 0.20*** 0.20*** [5.87] [5.89] [5.88] [6.00] [5.97] [5.98] Log cement consumptiont-3 0.16*** 0.16*** 0.16*** 0.16*** 0.16*** 0.16*** [5.25] [5.21] [5.20] [5.21] [5.24] [5.05] Lok Sabha Electiont-6 -0.01 [-0.34] Lok Sabha Electiont-5 0.02 0.01 [0.71] [0.60] Lok Sabha Electiont-4 0.05** 0.05** 0.05** [2.01] [2.20] [2.01] Lok Sabha Electiont-3 0.01 0.02 0.02 0.02 [0.57] [1.00] [0.98] [0.88] Lok Sabha Electiont-2 -0.01 0.00 0.01 0.01 0.01 [-0.29] [0.04] [0.62] [0.58] [0.48] Lok Sabha Electiont-1 -0.05** -0.05** -0.05* -0.04 -0.04 -0.04 [-2.09] [-2.11] [-1.81] [-1.48] [-1.46] [-1.50] Lok Sabha Election -0.06** -0.06** -0.05** -0.05** -0.05** -0.05** [-2.43] [-2.47] [-2.19] [-1.98] [-1.96] [-2.01] Lok Sabha Electiont+1 -0.05** -0.05** -0.04* -0.03 -0.03 -0.04 [-2.10] [-2.13] [-1.76] [-1.37] [-1.41] [-1.48] Lok Sabha Electiont+2 -0.01 -0.00 0.01 0.01 0.01 [-0.30] [-0.01] [0.39] [0.35] [0.26] Lok Sabha Electiont+3 0.05* 0.07*** 0.07*** 0.06** [1.89] [2.81] [2.71] [2.56] Lok Sabha Electiont+4 0.03 0.04 0.03 [1.47] [1.53] [1.40] Lok Sabha Electiont+5 -0.00 -0.01

[-0.14] [-0.23] Lok Sabha Electiont+6 -0.00 [-0.11]

Year & State- Year & State- Year & State- Year & State- Year & State- Year & State- Fixed effects Month Month Month Month Month Month Observations 2,992 2,975 2,958 2,924 2,890 2,856 R-squared 0.97 0.97 0.97 0.97 0.97 0.97 Number of states 17 17 17 17 17 17

Note: Z statistics in brackets. * significant at 10%; ** significant at 5%; *** significant at 1%. All models include three lags of the dependent variable, year fixed effects, and fixed effects for each state-month combination. Models are estimated using OLS with panel-corrected standard errors. Dependent variable is natural log of cement consumption.

Appendix Table 6: Cement Consumption and State Elections, Adding 12-month Lags and Leads

(1) (2) (3) (4) (5) (6) Log cement Log cement Log cement Log cement Log cement Log cement DV: consumption consumption consumption consumption consumption consumption

Log cement consumptiont-1 0.33*** 0.33*** 0.31*** 0.28*** 0.40*** 0.39*** [9.91] [9.76] [9.24] [8.34] [12.04] [12.43] Log cement consumptiont-2 0.22*** 0.22*** 0.20*** 0.27*** 0.22*** 0.23*** [6.32] [6.24] [5.59] [7.79] [6.20] [6.81] Log cement consumptiont-3 0.16*** 0.16*** 0.22*** 0.18*** 0.14*** 0.15*** [4.85] [4.82] [6.67] [5.61] [4.38] [4.61]

Electiont-12 0.01 [0.27]

Electiont-11 0.02 0.02 [0.75] [1.08]

Electiont-10 0.01 0.01 0.01 [0.48] [0.34] [0.57]

Electiont-9 0.00 0.00 0.00 0.00 [0.19] [0.14] [0.09] [0.18]

Electiont-8 0.01 0.00 0.01 0.01 0.00 [0.26] [0.20] [0.26] [0.23] [0.11]

Electiont-7 0.02 0.02 0.02 0.02 0.01 0.02 [0.87] [0.86] [0.79] [0.76] [0.63] [0.81]

Electiont-6 0.06** 0.05** 0.05** 0.05** 0.05** 0.05** [2.54] [2.31] [2.29] [2.33] [2.32] [2.32]

Electiont-5 -0.00 -0.00 -0.00 -0.00 -0.01 -0.01 [-0.17] [-0.06] [-0.10] [-0.16] [-0.30] [-0.61]

Electiont-4 -0.02 -0.02 -0.01 -0.01 -0.01 -0.01 [-0.73] [-0.74] [-0.51] [-0.58] [-0.51] [-0.71]

Electiont-3 -0.03 -0.03 -0.04* -0.04 -0.04 -0.04* [-1.54] [-1.55] [-1.72] [-1.62] [-1.50] [-1.94]

Electiont-2 0.01 0.01 0.01 0.01 0.03 0.02 [0.58] [0.51] [0.46] [0.38] [1.17] [1.00]

Electiont-1 0.00 0.00 0.00 0.00 0.00 0.01 [0.13] [0.07] [0.05] [0.15] [0.06] [0.71] Election -0.12*** -0.12*** -0.12*** -0.12*** -0.13*** -0.04** [-5.39] [-5.39] [-5.36] [-5.44] [-5.57] [-2.19]

Electiont+1 0.03 0.03 0.03 0.02 0.01 0.01 [1.25] [1.24] [1.25] [1.08] [0.25] [0.37]

Electiont+2 0.03 0.03 0.02 0.01 0.01 0.01 [1.16] [1.17] [1.00] [0.49] [0.43] [0.68]

Electiont+3 0.04 0.04 0.02 0.02 0.02 0.02 [1.57] [1.55] [1.07] [1.05] [1.02] [1.19]

Electiont+4 0.01 0.01 0.01 0.01 0.01 0.01 [0.65] [0.46] [0.46] [0.43] [0.39] [0.54]

Electiont+5 0.04* 0.04* 0.04** 0.04* 0.04* 0.03 [1.96] [1.95] [1.96] [1.93] [1.92] [1.29]

Electiont+6 0.00 0.01 0.01 0.01 0.01 0.01 [0.19] [0.24] [0.28] [0.34] [0.47] [0.47]

Electiont+7 -0.02 -0.02 -0.02 -0.02 -0.02 -0.03 [-1.02] [-1.01] [-1.08] [-0.85] [-0.90] [-1.53]

Electiont+8 -0.01 -0.00 -0.00 0.00 -0.00 [-0.30] [-0.15] [-0.18] [0.07] [-0.08]

Electiont+9 -0.02 -0.01 -0.01 -0.01 [-0.67] [-0.56] [-0.52] [-0.57]

Electiont+10 -0.01 -0.01 [-0.45] [-0.46]

Electiont+11 -0.01 -0.01 [-0.52] [-0.69]

Electiont+12 -0.03 [-1.20]

Time & State- Time & State- Time & State- Time & State- Time & State- Time & State- Fixed effects Month Month Month Month Month Month Observations 2,822 2,788 2,754 2,720 2,686 2,652 R-squared 0.97 0.97 0.97 0.97 0.98 0.98 Number of states 17 17 17 17 17 17

Note: Z statistics in brackets. * significant at 10%; ** significant at 5%; *** significant at 1%. Models include time fixed effects and fixed effects for each state-month combination. Models are estimated using OLS with panel-corrected standard errors. Dependent variable is natural log of cement consumption.

Appendix Table 7: Cement Consumption and State Elections, Four Lags of the Dependent Variable

(1) (2) (3) (4) Log cement Log cement Log cement Log cement DV: consumption consumption consumption consumption Sample All All Urban Rural

Log cement consumptiont-1 0.31*** 0.31*** 0.29*** 0.34*** [9.34] [9.37] [5.77] [8.31]

Log cement consumptiont-2 0.19*** 0.19*** 0.16*** 0.23*** [5.49] [5.48] [3.04] [5.22]

Log cement consumptiont-3 0.12*** 0.12*** 0.14*** 0.10** [3.47] [3.49] [2.58] [2.25]

Log cement consumptiont-4 0.12*** 0.12*** 0.15*** 0.09** [3.69] [3.68] [3.07] [2.23]

Electiont-6 0.06*** 0.06*** 0.09*** 0.06 [2.71] [2.79] [3.14] [1.58]

Electiont-5 -0.00 -0.00 0.00 0.01 [-0.01] [-0.02] [0.16] [0.35]

Electiont-4 -0.01 -0.01 0.03 -0.06* [-0.64] [-0.62] [1.07] [-1.75]

Electiont-3 -0.03 -0.03 -0.04 -0.03 [-1.51] [-1.52] [-1.38] [-0.81]

Electiont-2 0.00 0.00 -0.01 0.03 [0.21] [0.20] [-0.36] [0.86]

Electiont-1 0.00 0.00 0.00 -0.01 [0.06] [0.05] [0.05] [-0.24] Election -0.13*** -0.16*** -0.11*** [-5.53] [-5.26] [-2.99] Scheduled Election -0.15*** [-5.63]

Electiont+1 0.03 0.03 0.05* -0.01 [1.28] [1.26] [1.85] [-0.16]

Electiont+2 0.02 0.02 0.05 -0.01 [0.97] [0.99] [1.62] [-0.25]

Electiont+3 0.03 0.03 0.06** 0.01 [1.32] [1.32] [1.98] [0.38]

Electiont+4 0.03 0.03 0.04 0.03 [1.30] [1.30] [1.26] [0.93]

Electiont+5 0.04* 0.04* 0.06* 0.02 [1.90] [1.95] [1.87] [0.43]

Electiont+6 0.01 0.01 0.02 -0.00 [0.29] [0.29] [0.72] [-0.11]

Fixed effects - Time State-Month Time & State-Month Observations 2,856 2,856 1,512 1,344 R-squared 0.97 0.97 0.95 0.98 Number of states 17 17 9 8 17

Note: Z statistics in brackets. * significant at 10%; ** significant at 5%; *** significant at 1%. All models include four lags of the dependent variable. Model (2) includes time fixed effects; Model (3) includes fixed effects for each state-month combination; and Model (4) includes time and state-month fixed effects. Models are estimated using OLS with panel-corrected standard errors. Dependent variable is natural log of cement consumption.

Appendix Table 8: Cement Consumption and National Elections, Four Lags of the Dependent Variable

(1) (2) DV: Log cement consumption Log cement consumption

Log cement consumptiont-1 0.31*** 0.31*** [9.63] [9.66]

Log cement consumptiont-2 0.18*** 0.18*** [5.26] [5.30]

Log cement consumptiont-3 0.12*** 0.12*** [3.62] [3.60]

Log cement consumptiont-4 0.11*** 0.11*** [3.53] [3.54]

Lok Sabha Electiont-6 -0.00 -0.00 [-0.08] [-0.03]

Lok Sabha Electiont-5 0.01 0.01 [0.35] [0.36]

Lok Sabha Electiont-4 0.05* 0.05** [1.95] [2.02]

Lok Sabha Electiont-3 0.02 0.02 [0.93] [0.87]

Lok Sabha Electiont-2 0.01 0.02 [0.57] [0.64]

Lok Sabha Electiont-1 -0.03 -0.04 [-1.36] [-1.40] Lok Sabha Election -0.05** [-2.10] Scheduled Lok Sabha Election -0.09*** [-2.65]

Lok Sabha Electiont+1 -0.04 -0.04 [-1.60] [-1.63]

Lok Sabha Electiont+2 0.00 -0.00 [0.04] [-0.06]

Lok Sabha Electiont+3 0.06** 0.06** [2.51] [2.54]

Lok Sabha Electiont+4 0.04 0.04 [1.59] [1.54]

Lok Sabha Electiont+5 0.00 0.00 [0.05] [0.14]

Lok Sabha Electiont+6 0.00 0.00 [0.13] [0.11]

Fixed effects Year & State-Month Year & State-Month Observations 2,856 2,856 R-squared 0.97 0.97 Number of states 17 17

Note: Z statistics in brackets. * significant at 10%; ** significant at 5%; *** significant at 1%. Model includes four lags of the dependent variable, year fixed effects, and fixed effects for each state-month combination. Models are estimated using OLS with panel-corrected standard errors. Dependent variable is natural log of cement consumption.

Appendix Table 9: Cement Consumption and State Elections, No Lags of the Dependent Variable

(1) (2) (3) (4) Log cement Log cement Log cement Log cement DV: consumption consumption consumption consumption Sample: All All Urban Rural

Electiont-6 0.08*** 0.08*** 0.11*** 0.08* [2.90] [2.96] [2.91] [1.92]

Electiont-5 0.04 0.04 0.04 0.06 [1.27] [1.27] [1.10] [1.31]

Electiont-4 0.02 0.02 0.06* -0.02 [0.64] [0.66] [1.71] [-0.40]

Electiont-3 -0.01 -0.01 0.00 -0.01 [-0.30] [-0.30] [0.11] [-0.28]

Electiont-2 0.02 0.02 0.02 0.04 [0.67] [0.67] [0.53] [0.87]

Electiont-1 0.01 0.01 0.02 0.00 [0.42] [0.41] [0.55] [0.10] Election -0.12*** -0.15*** -0.10** [-4.24] [-3.84] [-2.39] Scheduled Election -0.15*** [-4.71]

Electiont+1 -0.01 -0.01 0.01 -0.04 [-0.26] [-0.28] [0.30] [-0.90]

Electiont+2 -0.00 -0.00 0.03 -0.04 [-0.10] [-0.07] [0.72] [-1.03]

Electiont+3 0.01 0.01 0.05 -0.02 [0.45] [0.45] [1.25] [-0.48]

Electiont+4 0.02 0.02 0.04 0.00 [0.61] [0.62] [0.89] [0.08]

Electiont+5 0.05* 0.05* 0.08** 0.00 [1.75] [1.78] [2.09] [0.05]

Electiont+6 0.03 0.02 0.06 -0.02 [0.87] [0.86] [1.61] [-0.34]

Fixed effects - Time State-Month Time & State-Month Observations 2,856 2,856 1,512 1,344 R-squared 0.96 0.96 0.92 0.97 Number of states 17 17 9 8

Note: Z statistics in brackets. * significant at 10%; ** significant at 5%; *** significant at 1%. Model (2) includes time fixed effects; Model (3) includes fixed effects for each state-month combination; and Model (4) includes time and state-month fixed effects. Models are estimated

21 using OLS with panel-corrected standard errors. Dependent variable is natural log of cement consumption.

Appendix Table 10: Cement Consumption and State Elections, No Lags of the Dependent Variable

(1) (2) DV: Log cement consumption Log cement consumption

Lok Sabha Electiont-6 -0.00 -0.00 [-0.12] [-0.10]

Lok Sabha Electiont-5 0.01 0.01 [0.36] [0.36]

Lok Sabha Electiont-4 0.05** 0.05** [1.97] [2.02]

Lok Sabha Electiont-3 0.04 0.04 [1.60] [1.52]

Lok Sabha Electiont-2 0.04 0.04 [1.56] [1.60]

Lok Sabha Electiont-1 -0.01 -0.01 [-0.40] [-0.47] Lok Sabha Election -0.04 [-1.61] Scheduled Lok Sabha Election -0.07** [-2.30]

Lok Sabha Electiont+1 -0.04* -0.04* [-1.78] [-1.86]

Lok Sabha Electiont+2 -0.01 -0.02 [-0.55] [-0.69]

Lok Sabha Electiont+3 0.05** 0.05** [2.08] [2.06]

Lok Sabha Electiont+4 0.05** 0.05** [2.07] [2.00]

Lok Sabha Electiont+5 0.02 0.02 [0.90] [0.95]

Lok Sabha Electiont+6 0.02 0.02 [0.68] [0.63]

Fixed effects Year & State-Month Year & State-Month Observations 2,856 2,856 R-squared 0.96 0.96 Number of states 17 17

Note: Z statistics in brackets. * significant at 10%; ** significant at 5%; *** significant at 1%. Models include year fixed effects and fixed effects for each state-month combination. Models are estimated using OLS with panel-corrected standard errors. Dependent variable is natural log of cement consumption.

Appendix Table 11: Cement Production and State Elections, Adding Lags and Leads

(1) (2) (3) (4) (5) (6) Log cement Log cement Log cement Log cement Log cement Log cement DV: production production production production production production

Log cement productiont-1 0.59*** 0.59*** 0.60*** 0.58*** 0.57*** 0.57*** [13.00] [12.84] [12.73] [12.18] [11.55] [11.28] Log cement productiont-2 0.30*** 0.31*** 0.31*** 0.32*** 0.32*** 0.33*** [6.67] [6.72] [6.60] [6.68] [6.60] [6.49]

Electiont-6 -0.02 [-0.58]

Electiont-5 -0.03 -0.03 [-0.85] [-0.79]

Electiont-4 -0.04 -0.04 -0.04 [-1.01] [-1.03] [-1.00]

Electiont-3 -0.05 -0.05 -0.05 -0.05 [-1.23] [-1.25] [-1.23] [-1.25]

Electiont-2 0.04 0.04 0.04 0.04 0.04 [1.01] [0.99] [1.05] [0.99] [0.96]

Electiont-1 -0.05 -0.05 -0.05 -0.04 -0.04 -0.05 [-1.14] [-1.11] [-1.12] [-1.09] [-1.06] [-1.10] Election 0.03 0.03 0.03 0.03 0.03 0.03 [0.75] [0.75] [0.79] [0.75] [0.75] [0.74]

Electiont+1 0.07* 0.07* 0.07* 0.07* 0.07 0.07* [1.72] [1.71] [1.68] [1.67] [1.60] [1.67]

Electiont+2 -0.03 -0.03 -0.03 -0.03 -0.03 [-0.74] [-0.87] [-0.85] [-0.78] [-0.78]

Electiont+3 -0.02 -0.02 -0.02 -0.02 [-0.59] [-0.60] [-0.55] [-0.57]

Electiont+4 0.01 0.01 0.01 [0.26] [0.23] [0.20]

Electiont+5 0.01 0.00 [0.13] [0.07]

Electiont+6 0.00 [0.10]

Time & State- Time & State- Time & State- Time & State- Time & State- Time & State- Fixed effects Month Month Month Month Month Month Observations 2,727 2,711 2,678 2,645 2,612 2,579 R-squared 0.99 0.99 0.99 0.99 0.99 0.99 Number of states 17 17 17 17 17 17

Note: Z statistics in brackets. * significant at 10%; ** significant at 5%; *** significant at 1%. Models include two lags of the dependent variable, time fixed effects and fixed effects for each

24 state-month combination. Models are estimated using OLS with panel-corrected standard errors. Dependent variable is natural log of cement production.

Appendix Table 12: Cement Production and State Elections, Dropping Delhi and Haryana

(1) (2) (3) (4) Log cement Log cement Log cement Log cement DV: production production production production

Log cement productiont-1 0.53*** 0.53*** 0.59*** 0.57*** [4.41] [4.11] [12.13] [11.28] Log cement productiont-2 0.46*** 0.46*** 0.34*** 0.33*** [3.85] [3.57] [7.17] [6.49]

Electiont-6 -0.01 -0.02 -0.02 -0.02 [-0.15] [-0.43] [-0.50] [-0.58]

Electiont-5 -0.02 -0.03 -0.02 -0.03 [-0.33] [-0.54] [-0.43] [-0.79]

Electiont-4 -0.01 -0.02 -0.05 -0.04 [-0.16] [-0.31] [-1.30] [-1.00]

Electiont-3 -0.03 -0.00 -0.05 -0.05 [-0.36] [-0.03] [-1.16] [-1.25]

Electiont-2 0.09 0.07 0.06 0.04 [1.27] [1.05] [1.48] [0.96]

Electiont-1 0.01 -0.04 -0.03 -0.05 [0.10] [-0.62] [-0.76] [-1.10] Election -0.03 0.03 0.01 0.03 [-0.38] [0.48] [0.20] [0.74]

Electiont+1 0.21*** 0.20*** 0.05 0.07* [2.86] [2.69] [1.26] [1.67]

Electiont+2 -0.04 -0.05 -0.02 -0.03 [-0.98] [-0.96] [-0.40] [-0.78]

Electiont+3 -0.03 -0.05 0.01 -0.02 [-0.66] [-1.02] [0.21] [-0.57]

Electiont+4 0.00 0.01 0.02 0.01 [0.08] [0.15] [0.59] [0.20]

Electiont+5 -0.03 -0.01 0.01 0.00 [-0.69] [-0.10] [0.16] [0.07]

Electiont+6 0.01 0.01 0.00 0.00 [0.13] [0.12] [0.03] [0.10]

Fixed effects - Time State-Month Time & State-Month Observations 2,579 2,579 2,579 2,579 R-squared 0.96 0.96 0.99 0.99 Number of states 17 17 17 17

Note: Z statistics in brackets. * significant at 10%; ** significant at 5%; *** significant at 1%. All models include two lags of the dependent variable. Model (2) includes time fixed effects; Model (3) includes fixed effects for each state-month combination; and Model (4) includes time and state-month fixed effects. Models are estimated using OLS with panel-corrected standard

26 errors. Dependent variable is natural log of cement production. Observations for Delhi and Haryana are dropped due to missing data.

Appendix Table 13: Cement Consumption and State Elections, Dropping Delhi and Haryana and Adding Lags and Leads

(1) (2) (3) (4) (5) (6) Log cement Log cement Log cement Log cement Log cement Log cement DV: production production production production production production

Log cement productiont-1 0.64*** 0.63*** 0.64*** 0.63*** 0.61*** 0.62*** [22.11] [21.84] [22.28] [21.56] [21.22] [21.33] Log cement productiont-2 0.28*** 0.28*** 0.29*** 0.30*** 0.30*** 0.31*** [9.81] [9.89] [10.03] [10.27] [10.67] [10.75]

Electiont-6 -0.02 [-0.73]

Electiont-5 -0.04 -0.04 [-1.42] [-1.31]

Electiont-4 -0.04 -0.04 -0.04 [-1.33] [-1.41] [-1.35]

Electiont-3 -0.06** -0.06** -0.06** -0.06** [-2.02] [-2.08] [-2.17] [-2.18]

Electiont-2 0.05 0.05 0.05* 0.05* 0.05* [1.64] [1.64] [1.73] [1.76] [1.71]

Electiont-1 -0.05 -0.04 -0.05 -0.04 -0.04 -0.04 [-1.61] [-1.56] [-1.61] [-1.54] [-1.54] [-1.56] Election 0.03 0.03 0.03 0.03 0.03 0.03 [0.93] [0.93] [1.00] [0.95] [0.98] [1.04]

Electiont+1 0.02 0.01 0.01 0.01 0.01 0.01 [0.55] [0.53] [0.51] [0.47] [0.41] [0.47]

Electiont+2 -0.03 -0.03 -0.03 -0.03 -0.03 [-0.99] [-1.16] [-1.21] [-1.14] [-1.12]

Electiont+3 -0.02 -0.02 -0.02 -0.02 [-0.62] [-0.64] [-0.60] [-0.60]

Electiont+4 0.02 0.02 0.02 [0.83] [0.87] [0.83]

Electiont+5 0.01 0.01 [0.43] [0.37]

Electiont+6 -0.00 [-0.07]

Time & State- Time & State- Time & State- Time & State- Time & State- Time & State- Fxed effects Month Month Month Month Month Month Observations 2,649 2,634 2,604 2,574 2,544 2,514 R-squared 0.99 0.99 0.99 0.99 0.99 0.99 Number of states 15 15 15 15 15 15

(3) includes fixed effects for each state-month combination; and Model (4) includes time and state-month fixed effects. Models are estimated using OLS with panel-corrected standard errors. Dependent variable is natural log of cement production. Observations for Delhi and Haryana are dropped due to missing data.

Appendix Table 14: Cement Consumption and National Elections, Controlling for IIP

-1 -2 -3 -4 Log cement Log cement Log cement Log cement DV: production production production production

Log cement consumptiont-1 0.52*** 0.52*** 0.34*** 0.32*** [16.79] [16.76] [10.89] [10.27] Log cement consumptiont-2 0.27*** 0.27*** 0.22*** 0.20*** [7.96] [7.98] [6.41] [6.01] Log cement consumptiont-3 0.19*** 0.19*** 0.17*** 0.16*** [6.13] [6.15] [5.51] [5.08] IIP 0.00 -0.00 0.00*** 0.00** [0.79] [-0.11] [10.55] [2.46]

Lok Sabha Electiont-6 0.03 0.04 -0.02 -0.00 [0.73] [0.90] [-0.83] [-0.03]

Lok Sabha Electiont-5 0.02 0.03 0.00 0.02 [0.45] [0.66] [0.06] [0.67]

Lok Sabha Electiont-4 0.08** 0.09** 0.04 0.05** [1.98] [2.07] [1.52] [2.16]

Lok Sabha Electiont-3 0.05 0.05 0.02 0.03 [1.31] [1.12] [0.75] [1.04]

Lok Sabha Electiont-2 -0.02 -0.02 0.01 0.01 [-0.41] [-0.41] [0.28] [0.57]

Lok Sabha Electiont-1 0.04 0.03 -0.04* -0.04 [1.01] [0.75] [-1.71] [-1.44] Lok Sabha Election -0.10** -0.10** -0.05** -0.05** [-2.53] [-2.36] [-2.18] [-2.00]

Lok Sabha Electiont+1 0.03 0.03 -0.03 -0.03 [0.86] [0.62] [-1.43] [-1.22]

Lok Sabha Electiont+2 0.00 -0.00 0.01 0.00 [0.07] [-0.06] [0.24] [0.17]

Lok Sabha Electiont+3 0.03 0.02 0.06** 0.06** [0.67] [0.49] [2.56] [2.39]

Lok Sabha Electiont+4 -0.04 -0.05 0.02 0.03 [-1.11] [-1.06] [0.90] [1.09]

Lok Sabha Electiont+5 -0.05 -0.05 -0.02 -0.01 [-1.32] [-1.25] [-0.79] [-0.55]

Lok Sabha Electiont+6 0.03 0.02 -0.01 -0.00 [0.64] [0.52] [-0.45] [-0.19]

Year & Fixed effects - Year State-Month State-Month Observations 2,856 2,856 2,856 2,856 R-squared 0.95 0.95 0.97 0.97 Number of states 17 17 17 17

Note: Z statistics in brackets. * significant at 10%; ** significant at 5%; *** significant at 1%. All models include three lags of the dependent variable. Model (2) includes year fixed effects; Model (3) includes fixed effects for each state-month combination; and Model (4) includes year and state-month fixed effects. Models are estimated using OLS with panel-corrected standard errors. Dependent variable is natural log of cement consumption.

Appendix Table 15: Randomization Test Statistics

Randomization test T(obs) c n p=c/n SE(p) [95% Conf. Interval] -4.08 1 1000 0.001 0.001 0.0000253 0.0055589

Appendix Table 16: Additional Examples of Land “Scams” from India’s States

State Scandal Time period URL Andhra Pradesh Chairman and Managing Director of Andhra 2011-2012 http://www.indianexpress.com/news/apiicemaar Pradesh Industrial Infrastructure -land-deal-nbw-against-emmar-mgf-md-shravan- Corporation (APIIC) B.P. Acharya allegedly gupta/956273/ conspired with Emaar Group and unnamed public officials to benefit Emaar at APIIC's expense in a transfer of public land. The case implicated several Congress Party politicians, including former Chief Minister YS Reddy and his son, Jagan Reddy. Andhra Pradesh The private business activities of Jagan Mid-2000s to http://www.firstpost.com/india/cbi-court-issues- Reddy, son of former Chief Minister YS present summons-to-andhra-home-minister-in-jagan- Reddy, were allegedly given undue case- assistance from his father's government, 778777.html?utm_source=fwire&utm_medium= which is said to have granted favors to his hp investors. One prominent example was the transfer of government land in Kadapa district to a cement company that invested in Jagan's own firm, Bharati Cements. Bihar Bihar chief minister Nitish Kumar has 2012 http://www.hindustantimes.com/India- ordered an investigation into an alleged land news/Bihar/Nitish-orders-probe-into-Bihar-land- scam, in which children of leaders of the scam/Article1- ruling JD(U)-BJP alliance were shown 723012.aspx#sthash.WALveOk5.dpuf favoritism in the allotment of land meant for industrial use.

Delhi The revenue department of the Delhi 2012 http://www.hindustantimes.com/India- government uncovered a land scam in east news/NewDelhi/Revenue-department-suspends- Delhi and has suspended three officers for 3-officials-in-east-Delhi-land-scam/Article1- colluding with private individuals and 856361.aspx#sthash.QK1fky3t.dpuf registering government land in private names. Gujarat The CAG has accused the government of 2011-2012, http://articles.timesofindia.indiatimes.com/2013 Gujarat of chronically mismanaging its court -04-04/ahmedabad/38277773_1_cag-report- allotment of public land by failing to adhere proceedings government-land-slvc to standards of uniformity and transparency. ongoing The report estimates that these practices have cost taxpayers Rs 747 crore in one year. The case is currently in front of the Gujarat High Court. Gujarat The Gujarat High Court is currently involved Favors http://www.rediff.com/news/slide-show/slide- in legal proceedings surrounding CAG throughout show-1-cag-slams-modi-govt-for-huge-financial- allegations of improper land allocation, Modi tenure, irregularities/20120330.htm#3 during the tenure of chief minister Narendra report Modi, to the Gujarat International Finance released in Tec (GIFT) City Company. The court has 2011, news sought replies from, among others, elected report from and bureaucratic officials at the state and 2012 central government levels. Haryana The CAG accused the Haryana State 2008-2009 http://www.indianexpress.com/news/cag- Industrial and Infrastructure Development unveils-rs-400-cr-dlf-hsiidc-land-deal- Corporation Ltd. (HSIIDC) of accepting an scam/1086958/1 undervalued rate for land that was passed on to a large, private real estate company, DLF Ltd.

Haryana Congress Party President Sonia Gandhi's Ongoing http://articles.economictimes.indiatimes.com/20 son-in-law Robert Vadra has been investigation 13-04-04/news/38278372_1_haryana-ias-robert- implicated in several dubious land transfer vadra-dlf-ashok-khemka deals, including one that involved the sale of Haryana land to Vadra's company at below- market price. Jammu & "Senior government officials" were accused Ongoing http://www.indlaw.com/guest/DisplayNews.aspx Kashmir of an illegal land grab by exceeding the investigation ?A1E2D324-6911-49EE-95EB-C8A9CE161CF7 authority of the Roshni Act, which dealt with land distribution in certain regions. The case is being investigated by the State Vigilance Office (SVO). The Jammu and Kashmir High Court recently allowed the SVO investigation to proceed unimpeded. Karnataka A Karnataka government-appointed 2012 http://timesofindia.indiatimes.com/india/Politici committee which surveyed wakf (Islamic ans-middlemen-mafia-in-Rs-2-lakh-crore- charity) board lands established that major Karnataka-wakf-land- irregularities took place between 2001 and scam/articleshow/12422183.cms 2012 in the transfer of government land to private parties through mutation (the process through which the government land is transferred or sold to individuals, instituttions, persons, or firms after a proper verification of documents). According to the committee, this process was violated while selling or transferring wakf land to private individuals. Karnataka The state Lokayukta (anti-corruption 2010-2011 http://newindianexpress.com/thesundaystandar ombudsman) has lodged several cases d/article603249.ece against former chief minister, BS

Yeddyurappa, involving illegal land denotifications benefitting firms owned, or paritially owned, by the chief minister's family members. Kerala Kerala Congress (Joseph) party member and 2007 http://www.indianexpress.com/news/kerala- then-state Minister of Public Works T.U. land-scam-cm-faces-pressure-to-sack- Kuruvila's three children had tried to sell minister/212303/ some 50.53 acres of land in Idukki that they had bought from 19 different people, to businessman K.G. Abraham, in return for which they received a generous Rs 6.75 crore kickback. The deal went south after the businessman sent a legal notice to Kuruvila and went public after he got back neither the title documents to register the sale, nor his money. Kerala Former chief minister and CPM leader V.S. Ongoing; land http://www.ndtv.com/article/south/kerala-land- Achuthanandan has been accused in a land sale in 2011 scam-chargesheet-against-former-chief-minister- scam case implicating his relative, with a achuthanandan-soon-294921 chargesheet forthcoming. Madhya Pradesh Former minister Suresh Seth alleged that Scam took http://ibnlive.in.com/news/mp-charge-sheet- when Madhya Pradesh Industry Minister place in 2004; filed-in-land-scam-case-involving-bjp- Kailash Vijaywargiya was Indore mayor (in chargesheete mla/392935-3.html 2004), three acres of land was allotted to d in 2013 Nanda Nagar Sahkari Samiti, headed by his right-hand man Mendola. Lokayukta Police submitted a charge sheet against BJP MLA Ramesh Mendola and 16 others. A charge sheet has been submitted and Mendola and others were released on bail.

Maharashtra Former Maharshtra chief minister Manohar 1999 http://www.rediff.com/news/2002/aug/19land.h Joshi surrendered his membership in the tm state legislative assembly in the wake of the Bombay high court's ruling that he had improperly allocated land in Pune to his son- in-law. Maharashtra Deputy chief minister of Maharashtra, Ajit Early 2000s http://www.bbc.co.uk/news/world-asia-india- Pawar, who is farm minister Sharad Pawar's 20000561 nephew, is accused of misusing his powers to give away 348 acres of land "at throw away prices" to the company which developed the Lavasa hill city in 2002. Orissa The CAG accused the government of Orissa Last 20 years http://www.financialexpress.com/news/cag- of a long-term, ongoing failure to smells-rs.4000cr-land-scam-in-orissa/1061008/1 appropriately discipline the sale of land, which has resulted in the collusion of "several politicians, senior bureaucrats, judges, and journalists" and government ministers with their "discretionary quotas" over about 20 years. Orissa A senior BJD minister, S.N. Patro, was 2012-2013 http://zeenews.india.com/news/odisha/odisha- implicated by the CAG in a violation of the govt-cancels-land-allotment-in-favour-of- state's land-use policy in the lease of a plot minister-s-son_842041.html to his son. The lease was cancelled by the government following the CAG report. Punjab Former Patiala deputy commissioner Vikas 2011-present http://timesofindia.indiatimes.com/india/Land- Garg was arrested following his alleged scam-tainted-former-deputy-commissioner-of- involvement in a controversial land sale Patiala-arrested/articleshow/18998369.cms deed worth Rs 250 crore to private buyers. The vigilance bureau (VB) filed a

first information report (FIR) against Garg and nine others for selling 6,000 square yards of government land worth Rs 250 crore in 2011. Punjab A joint Punjab-Haryana court tribunal found Ongoing http://spaceviewtimes.com/index.php?option=c several instances of public land illegally held investigation om_content&view=article&id=14222:land-grab- by private entities, at least some of which scam-involving-punjab-vvips-unearthed-tribunal- included politicians, bureaucrats, police, and indicts-dgp&catid=108:headlines revenue officials from Punjab. Rajasthan The state's anti-corruption bureau 2010 http://www.financialexpress.com/news/exposed registered a case against the former minister -rajasthan-s-biggest-land-scams/718392 of state for urban development and housing Pratap Singh Singhvi and others, alleging that the minister alloted government land to private firms in a fraudulent manner. Tamil Nadu Top officials of the DMK party have been 2008-2010 http://www.indianexpress.com/news/new-land- charged with misuse of discretionary power scam-surfaces-in-tamil-nadu/721555/ in the distribution of land to, among others, MLAs, politicians, and high-ranking bureaucrats. Tamil Nadu India's Chemical and Fertilizer Minister M.K. 2011-2012 http://indiatoday.intoday.in/story/fir-registered- Alagiri, a DMK MP from Tamil Nadu, and his against-mk-alagiri-son-in-connection-with-illegal- son, Durai Dayanidhi, are alleged to have mining-case/1/212221.html been involved in a mining scam that involved encroachment on public land, according to an FIR. Uttar Pradesh The former UP government headed by the 2011 http://articles.timesofindia.indiatimes.com/2012 BSP has allegedly distributed land in Noida -12-28/noida/36035564_1_noida-authority- (in western UP) to favored individuals and abadi-land-sanjeev-saran other friendly private entities. The Noida 38

Authority removed at least two officials from their posts in the wake of the scandal.

Uttar Pradesh Former UP Chief Minister Mayawati and one 2003, Court http://articles.timesofindia.indiatimes.com/2013 of her ministers, Nasimuddin Siddiqui, were Case in 2012 -01-29/india/36614922_1_taj-heritage-corridor- charged with corruption in conjunction with scam-accord-sanction-review-petition the Taj Heritage Corridor improvement project in the early 2000s. The case is currently before the Supreme Court. West Bengal West Bengal Chief Minister Mamata 2009 http://www.indianexpress.com/news/mamata- Banerjee expelled three of her Trinamool expels-three-partymen/514789/ Congress party members over allegations that they were complicit in a landgrab in the Rajarhat-Bhangar areas. (This was before TMC took over the government of WB in 2011.) West Bengal A judge conspired with the West Bengal Early 2000s http://articles.timesofindia.indiatimes.com/2006 government to green-light a land sale of -02-04/india/27794611_1_high-court-judge- which he himself was a beneficiary. The land bidding-process-supreme-court was distributed out of the Chief Minister's quota. The Supreme Court upheld the Calcutta High Court order that found fault with the retired judge.