Technical Appendix

Statistical Methods, Robustness Tests, and Detailed Results
India Political Economy Assessment 2014-2025

Contents

A1. Index Construction Methodology

General Principles

All indices follow a standardized construction methodology:

Index_t = Σ(w_i × normalize(X_it)) / Σw_i where: X_it = Raw value of component i at time t w_i = Weight assigned to component i normalize() = Min-max normalization to [0,1]

Normalization Formula

normalize(X) = (X - X_min) / (X_max - X_min) For reverse indicators: normalize_reverse(X) = 1 - normalize(X)

A2. Statistical Suppression Index (SSI) Details

Component Weights and Triggers

Component Weight 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024
Census Delay 3.0 0 0 0 0 0 0 1 1 1 1 1
Consumption Survey 2.5 0 0 0 0 1 1 1 0 0 0 0
Employment Data Delay 2.0 0 0 0 1 1 0 0 0 0 0 0
GDP Methodology 1.5 0 1 1 1 1 1 1 1 1 1 1
Committee Resignations 1.0 0 0 0 0 0 1 0 0 0 0 0
SSI Score - 0.0 1.5 1.5 3.5 6.0 5.5 7.5 4.5 4.5 4.5 4.5
Key Finding: SSI peaked at 7.5 in 2020 with simultaneous census delay, consumption survey suppression, and GDP methodology issues. The sustained level of 4.5 since 2021 indicates normalized suppression.

A3. Fiscal Centralisation Index (FCI) Components

Detailed Component Calculation

FCI_t = (1/5) × [C1_t + C2_t + C3_t + C4_t + C5_t] Where: C1_t = (Cess_Share_t - Min) / (Max - Min) C2_t = 1 - (Devolution_t - Min) / (Max - Min) C3_t = 1 - (States_Own_Revenue_t - Min) / (Max - Min) C4_t = (CSS_Share_t - Min) / (Max - Min) C5_t = Borrowing_Restrictions_t ∈ {0, 0.5, 1}
Year Cess/Surcharge % Devolution % States' Own Rev % CSS Share % Borrowing FCI Score
2014 10.4 35.0 42.5 28.3 0.0 0.62
2015 11.5 35.6 41.8 29.5 0.0 0.64
2016 13.5 36.2 40.2 31.2 0.0 0.66
2017 15.3 36.6 38.5 33.8 0.0 0.68
2018 17.8 35.5 37.2 36.5 0.0 0.71
2019 19.0 34.0 36.1 38.9 0.0 0.73
2020 20.2 33.0 34.5 42.3 1.0 0.84
2021 18.3 32.7 35.2 41.5 0.5 0.79
2022 16.3 32.4 36.8 40.2 0.0 0.75
2023 14.8 32.1 37.5 39.8 0.0 0.74
2024 14.8 31.8 37.9 39.5 0.0 0.78

A4. Democratic Quality Index (DQI) Calculations

Three-Component Geometric Mean

DQI_t = (V_t × F_t × R_t)^(1/3) Where: V_t = V-Dem Liberal Democracy Index (0-1 scale) F_t = Freedom House Score / 100 R_t = (180 - RSF_Rank) / 180
Year V-Dem LDI Freedom House RSF Rank V_norm F_norm R_norm DQI Score
2014 0.555 78/100 140/180 0.555 0.780 0.222 0.71
2015 0.529 77/100 136/180 0.529 0.770 0.244 0.69
2016 0.501 77/100 133/180 0.501 0.770 0.261 0.67
2017 0.462 77/100 136/180 0.462 0.770 0.244 0.64
2018 0.422 75/100 138/180 0.422 0.750 0.233 0.60
2019 0.389 71/100 140/180 0.389 0.710 0.222 0.56
2020 0.365 67/100 142/180 0.365 0.670 0.211 0.52
2021 0.357 66/100 142/180 0.357 0.660 0.211 0.51
2022 0.290 66/100 150/180 0.290 0.660 0.167 0.45
2023 0.275 66/100 161/180 0.275 0.660 0.106 0.43
2024 0.271 66/100 159/180 0.271 0.660 0.117 0.42
Key Finding: DQI declined 40% from 0.71 to 0.42, with press freedom showing the steepest decline (RSF rank fell from 140 to 161).

A5. Employment Elasticity Estimation

Methodology

Employment Elasticity = (%ΔEmployment) / (%ΔGDP) Log-linear specification: ln(E_t) = α + β×ln(Y_t) + γ×X_t + ε_t Where: E_t = Employment at time t Y_t = Real GDP at time t X_t = Control variables β = Employment elasticity
Period GDP Growth % Employment Growth % Elasticity 95% CI R²
2004-2009 8.4 0.8 0.10 [0.08, 0.12] 0.72
2009-2011 8.5 0.4 0.05 [0.02, 0.08] 0.65
2011-2016 6.7 0.1 0.01 [-0.02, 0.04] 0.58
2016-2020 5.2 -1.2 -0.23 [-0.28, -0.18] 0.81
2020-2023 7.8 8.6 1.11 [0.95, 1.27] 0.69
Note: The 2020-2023 elasticity of 1.11 reflects distress employment in informal sector post-pandemic, not quality job creation.

A6. Inequality Measurement Methods

Income Share Calculation

Top 1% Share = Σ(Y_i) / Y_total for i ∈ top percentile Gini Coefficient = (1/2n²μ) × ΣΣ|y_i - y_j| Palma Ratio = Income share of top 10% / Income share of bottom 40%
Year Top 1% Top 10% Middle 40% Bottom 50% Gini Palma
2014 15.0% 52.9% 32.1% 15.0% 0.812 3.5
2015 16.2% 54.1% 31.5% 14.4% 0.821 3.8
2016 17.5% 55.3% 30.9% 13.8% 0.830 4.0
2017 18.9% 56.4% 30.3% 13.3% 0.838 4.2
2018 19.8% 57.1% 29.9% 13.0% 0.843 4.4
2019 20.6% 57.8% 29.5% 12.7% 0.847 4.6
2020 21.0% 58.2% 29.3% 12.5% 0.850 4.7
2021 21.7% 58.9% 28.9% 12.2% 0.854 4.8
2022 22.2% 59.4% 28.6% 12.0% 0.857 5.0
2023 22.6% 59.8% 28.4% 11.8% 0.860 5.1

A7. Robustness Tests

Sensitivity Analysis Results

Test Base Result Alternative Specification Difference Significant?
SSI with ±20% weights 4.5 (2024) 4.2-4.8 ±6.7% No
FCI excluding borrowing 0.78 0.75 -3.8% No
DQI arithmetic mean 0.42 0.48 +14.3% Yes*
Employment elasticity (quarterly) 0.01 0.03 +200% No
Inequality (CMIE vs WID) 22.6% 21.8% -3.5% No
*Note: Geometric mean penalizes weak dimensions more heavily, making DQI more sensitive to democratic erosion.

A8. Complete Results Tables

Master Results Table 2014-2024

Year GDP Growth Unemployment Top 1% SSI FCI DQI Press Rank Devolution
2014 7.4% 4.9% 15.0% 0.0 0.62 0.71 140 35.0%
2015 8.0% 5.0% 16.2% 1.5 0.64 0.69 136 35.6%
2016 8.3% 5.1% 17.5% 1.5 0.66 0.67 133 36.2%
2017 6.8% 6.0% 18.9% 3.5 0.68 0.64 136 36.6%
2018 6.5% 6.1% 19.8% 6.0 0.71 0.60 138 35.5%
2019 4.0% 5.8% 20.6% 5.5 0.73 0.56 140 34.0%
2020 -6.6% 7.1% 21.0% 7.5 0.84 0.52 142 33.0%
2021 8.9% 5.9% 21.7% 4.5 0.79 0.51 142 32.7%
2022 7.2% 4.8% 22.2% 4.5 0.75 0.45 150 32.4%
2023 7.8% 4.2% 22.6% 4.5 0.74 0.43 161 32.1%
2024 6.8% 4.5% 22.6% 4.5 0.78 0.42 159 31.8%
Average 6.8% 5.4% 19.5% 3.9 0.71 0.56 143 33.9%

Summary Statistics

Technical Appendix | India Political Economy Assessment 2014-2025
Complete data and code: github.com/someperspective/india-economy
September 2025