Show the working.

PRS reproduces Budget at a Glance, Receipts Budget and Expenditure Budget. Tables 1-4 are the primary numbers for MeraTax.

• t1 — Table 1: Budget at a Glance 2026-27
• t2 — Table 2: Break up of central government receipts in 2026-27
• t3 — Table 3: Break up of central government expenditure in 2026-27
• t4 — Table 4: Ministry-wise expenditure in 2026-27
• t9 — Table 9: Criteria for devolution

PRIMARY source that PRS reproduces. Not yet fetched directly. Phase 0 uses PRS as the transcription source; swap to this once the exact document URL is confirmed.

PRIMARY source for Table 2 figures. Not yet fetched directly.

PRIMARY source for Tables 3-4. Also the only source that carries the ministry x economic cross-tab needed to build a single unified tree (see expenditure-2026-27.json -> _modellingNote).

Denominator for every per-capita figure on the site. Observed series fetched directly from the World Bank API on 2026-08-18; the 2026 figure used for FY 2026-27 is a projection, not an observation - see data/reference/population.json.

Probed live on 2026-08-18. NOT usable for Phase 0: the most current Union Budget resource covers only 2021-22 to 2023-24. IS the source for the Phase 4 state layer - shortlisted resource ids below, all confirmed returning clean typed rows.

The catalogue has ~285,833 resources and NO free-text search. Unrecognised query parameters are silently ignored and return HTTP 200 with the entire unfiltered catalogue. Only filters[title] works, and it matches the title only. Always assert that the filtered total is lower than the unfiltered total before trusting a response.

Used for figures MeraTax computes rather than transcribes. Every node carrying this sourceId must also carry a 'formula' field stating exactly how it was derived.

Statement 4 — absolute and percentage break-up of MPCE by item group, all-India, rural and urban. Extracted directly from the PDF. Average MPCE Rs 4,122 rural / Rs 6,996 urban.

• s4 — Statement 4: Absolute and percentage break-up of MPCE by item groups in 2023-24, All-India

Page last updated 17 August 2026. Petrol Rs 11.90/litre total (basic 1.40 + SAED 3.00 + AIDC 2.50 + R&I cess 5.00); diesel Rs 7.80/litre (basic 1.80 + AIDC 4.00 + R&I cess 2.00). Changes far more often than the budget — re-check before any launch.

PRS confirms the Finance Bill 2026 made no change to slabs. Slab boundaries, the Rs 75,000 standard deduction, the Rs 60,000 section 87A rebate up to Rs 12 lakh, and the 4% cess should be re-verified against the Finance Act itself before launch.

12% and 28% slabs abolished; 5% merit and 18% standard retained; 40% introduced for demerit and luxury goods. Item-level rate assignments in indirect-tax-incidence.json are MeraTax's, not the Council's.

Primary source for the horizontal devolution formula and every state's inter se share for 2026-31, and for the Union-vs-States split of health and education spending. Extracted directly from the PDF by scripts/etl/parse-fc16.mjs. All 28 shares sum to exactly 100.000%; four were independently cross-checked against press reporting.

• t8.8 — Table 8.8: Formula for Horizontal Devolution
• t8.9 — Table 8.9: Inter se Share of States
• t5.7 — Table 5.7: Expenditures on Health and Education (percentage of GDP) — states only
• p4.27 — Paragraph 4.27 — Union expenditure on social services, health and education as a percentage of GDP

₹5,987.46 crore total between 2014-15 and 2024-25. Year-wise: ₹409.55 cr (2020-21), ₹317.48 cr (2021-22), ₹408.37 cr (2022-23), ₹656.6 cr (2023-24), ₹654.9 cr (2024-25). Reported via press coverage of Lok Sabha and Rajya Sabha replies — NOT yet read from the original answer. Advertising is not a line item in the budget documents, so there is no cleaner source; re-verify against the original reply before leaning on it.

Approximately 80 crore people covered. Widely cited; re-verify against the department's own current figure before publishing it as precise. Presented on the site as 'roughly 80 crore', not a precise count.

Pulled live and reconciled on 2026-08-18: 36 State/UT rows plus a Total row. The recomputed sum (51,96,597 sanctioned / 39,98,704 in position / 11,97,893 vacant) matches the source's own Total row exactly, and every row satisfies sanctioned - inPosition = vacant. The Total row is dropped on import so it cannot be double-counted.

Primary document downloaded and read on 2026-08-18. The 6%-of-GDP commitment is paragraph 26.2, quoted verbatim in data/reference/targets.json. Paragraph 26.1 is the one that matters most: it records the same 6% figure as having been envisaged by the 1968 Policy, reiterated in 1986 and reaffirmed in 1992, and cites 4.43% of GDP for 2017-18 on the Ministry's own broader basis.

• p26.1 — Paragraph 26.1 — history of the 6% target and the 4.43% baseline
• p26.2 — Paragraph 26.2 — the 6% of GDP commitment

NOT yet read at paragraph level. The 2.5%-of-GDP-by-2025 commitment is corroborated by PIB releases and repeated ministerial replies in Parliament, but the primary PDF was unreachable on 2026-08-18 (mohfw.gov.in returned 404 and main.mohfw.gov.in did not resolve). Treat the clause wording as paraphrase until the document itself is re-read; the 2.5% figure and the 2025 date are not in doubt.

Table 3.1.1 (continuation page), extracted by scripts/etl/parse-police.mjs and reconciled against the PDF's own All India row on every run: 27,22,669 sanctioned / 21,41,305 in position / 5,81,364 vacant across 36 States/UTs. The most recent edition BPR&D has published as of 2026-08-19.

• t3.1.1 — Table 3.1.1 — Sanctioned (S) and Actual (A) Strength of Police Forces in States/UTs

Only the 'B. High Court' rows are imported; the Supreme Court row and the source's own 'Total (B. High Court)' row are excluded so nothing is double counted. Reconciled 2026-08-19: 25 High Courts, 1,122 sanctioned / 766 working / 356 vacant, and every row satisfies sanctioned - working = vacant.

Source archive SHA-256 b8325e5d9dd0f04a…, shapefile SHA-256 602505cbaa149c3a…. Projected Lambert Conformal Conic (metres) in the original; converted to WGS 84 lon/lat by scripts/etl/build-boundaries.mjs and simplified at 0.008 degrees, matched to the 1:1M source scale. Output bounding box 68.178-97.413E, 6.753-37.088N; centroids spot-checked against known positions; the three contested regions (Jammu & Kashmir, Ladakh, Arunachal Pradesh) visually verified against the Government of India depiction before the flag was enabled.

The Ministry's own hosts were unreachable when this was retrieved (mohfw.gov.in 404s the historical path; main.mohfw.gov.in does not resolve from this network). Recovered via the Internet Archive Wayback Machine's CDX index, which serves an unmodified snapshot of the exact file the Ministry published — verified by content type (application/pdf) and file size (8,927,323 bytes) at capture, and by SHA-256 2de925050f2b3298… of the downloaded copy, not by trusting the page alone.

Seven editions (2019-20 through 2025-26), fetched by scripts/etl/fetch-prs-archive.mjs and parsed by scripts/etl/parse-prs.mjs, which auto-detects each document's own fiscal year from its text rather than assuming the 2026-27 layout. PRS does not publish a full Union Budget Analysis before 2019-20 — earlier years carry only a two-page summary and the Railway Budget, confirmed by checking the PRS budget-year filter directly, which itself bottoms out at 2014-15.

Five editions. 2014-15 to 2016-17 are 'Expenditure Budget Vol. I, Statement 2' (ministry-wise, Plan/Non-Plan split) plus 'Statement 1' (Summary of Expenditure, carries Total Expenditure) — both fetched per year. 2017-18 and 2018-19 are 'Expenditure Profile, Statement 1' (Revenue/Capital split) which is economic AGGREGATES ONLY: it has no ministry-wise rows at all, confirmed by reading the document directly. As a result FY2015-16 and FY2016-17 have no ministry-wise actuals available from any source this site could reach, and data/union/ministry-series-2012-2026.json records those two years explicitly as gaps rather than omitting or interpolating them.

The DEFAULT index for real-terms figures on this site: a GDP deflator covers the whole economy's output, which is the right basis for deflating government expenditure aggregates. World Bank labels India by the calendar year in which the fiscal year BEGINS, so 2012 = FY2012-13 — verified against NY.GDP.MKTP.CN rather than assumed. Series ends at FY2025-26, so the 2026-27 Budget Estimate has no real-terms figure by design.

• NY.GDP.DEFL.ZS — GDP deflator, annual index, India

An ALTERNATE view only, never a headline on its own. CPI measures a household consumption basket, not government output, and over FY2012-13 to FY2025-26 it rises 1.955x against the GDP deflator's 1.603x. Education crosses the 'did it shrink' line depending which index is chosen, so this site publishes both and headlines the GDP deflator.

• FP.CPI.TOTL — Consumer price index, annual, India