Growth Without Accountability:
Measuring India's Institutional and Distributional Transformation, 2004-2026
This study examines the gap between official accounts and the empirical evidence on India's political economy from 2004 to 2026, setting the UPA decade (2004-2013) beside the NDA years (2014-2026). It analyses growth, employment, inequality, fiscal federalism, the integrity of official statistics and democratic institutions, and records where the growth figures and the other measures diverge.
| Indicator | UPA (2004-2014) | NDA (2014-2026) | Change |
|---|---|---|---|
| GDP Growth | 6.8% average | 6.2% average (to 2025-26) | -0.6pp |
| GDP Growth, excluding the worst year | 7.2% (excluding 2008-09) | 7.3% (excluding 2020-21) | +0.1pp |
| CPI Inflation | 8.0% average | 4.8% average | -3.2pp |
| Employment Elasticity | 0.01 (2004-05 to 2009-10) and 0.18 (2009-10 to 2011-12), RBI | 1.11 (2017-23), ORF* | Not comparable |
| Top 1% Income Share | 20.4% average | 21.9% average (to 2022) | +1.5pp |
| Tax Devolution (% of gross tax revenue) | 26.8% average | 32.7% average | +5.9pp |
| Democratic Rating | "Free" | "Partly Free" | Downgraded |
| Statistical Releases | Regular | Suppressed | Deteriorated |
*Employment elasticity is taken from published estimates, which use different data and methods. UPA period from Misra and Suresh, RBI Working Paper 06/2014, Table 4 (ratio of compound growth rates, NSSO employment). NDA period from Ghosh and Roy Bardhan, ORF Expert Speak, 2 October 2024 (the authors' linear model of employment on value added), who use the 2017-23 figure to argue that growth since 2017 has created jobs.
Every growth figure below is a GDP figure, and GDP counts work inside a boundary set by convention. Own the flat you live in and the accounts estimate a rent for it and add that to output. Cook the meal inside it and nothing is added. Women aged 6 and above who do unpaid domestic work spent 289 minutes a day on it in 2024 against 88 minutes for men (NSO Time Use Survey 2024), about 1,758 hours a year that appear in no output series. Published valuations of that work run from 7.5% to 36% of GDP, the spread coming from the choice of valuation method.
No figure in this summary is adjusted for it. They are the official numbers on the official boundary. Read the full treatment at someperspective.info/care.
The unemployment rate in 2017-18 was 6.1% for all ages (6.0% for those aged 15 and over), reported at the time as the highest since 1972-73 on a survey whose methods differ from the earlier ones, while the share of workers in regular jobs with social security fell from 11.5% in 2017-18 to 9.6% in 2022-23 and was 10.1% in 2023-24 (PLFS). PLFS records social-security cover only for regular employees, so the other nine in ten are workers with no recorded cover, most of them self-employed or casual. Published estimates put employment elasticity at 0.01 for 2004-05 to 2009-10 (RBI Working Paper 06/2014) and at 0.01 for 2011-16, a span that covers both governments, and 1.11 for 2017-23 (Ghosh and Roy Bardhan, ORF, 2024). The ILO and IHD India Employment Report 2024 finds that nearly two thirds of the employment added after 2019 was self-employment, mostly unpaid family work by women.
The top 1% income share stood at 22.6% in 2022-23, the latest year published, against 21.3% in 2014 and 18.4% in 2004. The World Inequality Lab finds that level higher than at any point since 1922, including the inter-war colonial period, and higher than in South Africa, Brazil and the US. At the other end of the distribution the bottom 50% share fell from 18.8% in 2004 to a low of 13.9% in 2017 and recovered to 15.0% in 2022-23. The top 1% held 40.1% of wealth in 2022-23, the highest share since the World Inequality Lab's wealth series begins in 1961. The wealth tax was abolished from 2015-16. Estate duty had been abolished in 1985.
Cesses and surcharges, which states do not share in, rose from 13.5% of the Union's gross tax revenue in 2014-15 to a peak of 24.4% in 2020-21, and were 18.0% in 2024-25 and 14.3% in the 2026-27 budget estimate (15th Finance Commission; Union Receipt Budgets). The GST in 2017 ended states' autonomy over most indirect taxation, and extra borrowing in 2020-21 was tied in part to reforms. Devolution moved the other way. After the 14th Finance Commission raised the states' share of the divisible pool from 32% to 42%, devolution rose from 27.1% of gross tax revenue in 2014-15 to 33.9% in 2024-25 (16th Finance Commission, Annexure 7.1; Budget at a Glance 2026-27). On the published fiscal ratios the two eras are close, and the NDA's higher Fiscal Centralisation Index comes from the GST and the 2020 borrowing conditions.
The Statistical Suppression Index (SSI), a 0-10 weighted sum of graded suppression severities across six documented streams, was 0 across the UPA decade, peaked at 9.0 in 2021-22, and was 4.65 in 2026. The largest single component is the postponed census (2.5 of the 9.0 at the peak), followed by the COVID-mortality undercount (2.0; official deaths of about 0.5 million against estimates of 3 to 5 million excess deaths). The others include the 2021 census, postponed in 2020 and now notified for enumeration in February 2027, the withheld 2017-18 consumption survey, the shelved GDP back-series, and the permanent NSSO→NSO merger.
Several international indices document institutional erosion. Freedom House downgraded India from "Free" to "Partly Free", V-Dem reclassified it as an "electoral autocracy" in its March 2021 Democracy Report, and press freedom ranking stood at 157 of 180 in 2026. The Democratic Quality Index (DQI), the geometric mean of the V-Dem Liberal Democracy Index, Freedom House, the RSF press-freedom rank and the V-Dem Core Civil Society Index, fell from a UPA-era average of 0.57 to 0.28 in 2026 (0.49 → 0.28 within the NDA period), driven hardest by the collapse in civil-society space (0.669 → 0.302).
A 0-10 weighted sum of graded suppression severities across six streams (census, consumption, employment, GDP back-series, institutional, COVID-mortality undercount). It was 0 across the UPA decade, peaked at 9.0 (2021-22), and was 4.65 in 2026, when the census counts as a scheduled delay.
Combines four published fiscal ratios (cess share, devolution, states' own revenue and non-Finance-Commission grants) with two coded components (the GST and the 2020 borrowing conditions). On a relative 0-1 scale it averaged 0.30 under UPA and 0.43 under NDA, peaked at 0.87 in 2020 and was 0.35 in 2026. On the four measured ratios alone the eras are close (0.49 under UPA, 0.45 under NDA), so the NDA's higher score rests on the GST, which took most independent indirect taxation away from the states in 2017.
Geometric mean of the V-Dem Liberal Democracy Index, Freedom House, the RSF press-freedom rank and the V-Dem Core Civil Society Index. It fell from a UPA-era average of 0.57 to 0.28 in 2026 (0.49 → 0.28 within the NDA period).
The study identifies three mechanisms that have kept the official account dominant despite the evidence against it:
Comparison with Turkey, Hungary, and Brazil shows India following its own model of "populist growth without accountability". In it, economic dynamism coexists with institutional decay, electoral democracy facilitates authoritarian practices, and technological inclusion masks material exclusion. This contradicts modernization theory's assumption that growth promotes democratization.
These findings extend the study of competitive authoritarianism to the digital age and show how electoral legitimacy can coexist with systematic institutional capture. The study provides evidence for policymakers, researchers, journalists, and citizens concerned about India's democratic and economic direction.
Democracy depends on information, debate, and accountability. This research contributes to the democratic task of holding power to account through evidence, and every figure in it can be checked against its published source. India's future is open, and the choice between continued concentration and inclusive renewal will be made by what is done now.
Full research available at someperspective.info
© 2025-2026 Dr. Varna Sri Raman | Educational use encouraged with attribution
Last updated: 10 October 2026