8 claims.
2 survive.

Both numbers exist. They are not the same kind of number. ₹2 lakh is a slab threshold — the income above which tax begins. ₹12.75 lakh is an effective zero-tax ceiling reached by stacking a rebate and a standard deduction on top of a ₹4 lakh slab. Comparing them makes the gap look like 6.4× when the like-for-like slab comparison is ₹2L → ₹4L.

The relief is real. The multiplier is manufactured by switching units mid-row.

Secondary defect: the 2014 increase to ₹2.5 lakh was announced by Arun Jaitley in the July 2014 budget — the first NDA budget, not a Chidambaram measure.

This row is wrong at both ends, and wrong in a specific direction.

11% is not a 2014 number. It is the March 2018 peak — 11.18% by RBI’s domestic-operations series, 11.46% on the gross figure. That peak arrived four years into the current government, and it arrived because of it: the RBI’s Asset Quality Review, begun in 2015, forced banks to reclassify restructured loans that had been carried as standard. Recognition, not deterioration, is what moved the number from 4.1% to 11%.

The card takes the single worst asset-quality reading of the Modi era and files it under the previous government.

The right-hand cell has a different problem. 0.4% is not the gross NPA ratio — that is 2.1%. The nearest real figure to 0.4% is the net NPA ratio (0.5%), which subtracts provisioning. So the row compares a peak gross ratio against a current net ratio, five times apart in definition before the year error is even counted.

What is defensible: asset quality genuinely is at its best level in roughly two decades, and stressed assets have fallen from 9.8% of the loan book in March 2014 to 3.55% in March 2025. That is a real achievement. It does not need an invented baseline.

The government’s own tabled figures dismantle this row.

₹65 lakh crore is the Finance Ministry’s number for 2014–26 — a twelve-year window. The card relabels it 2019–26 and shortens it to seven. Nothing was invented; a period was quietly cut in half while the total stayed put.

The left cell is worse. Against the Ministry’s own ₹12.39 lakh crore for the full 2004–14 decade, a 2009–14 half would land somewhere near ₹7–8 lakh crore. ₹1.57 lakh crore is roughly a fifth of that — closer to a single year’s central capex in that period than to five years of it.

A 12-year total on one side, a figure a fifth of the real one on the other. The ratio does the arguing; neither number does the reporting.

The underlying trend survives the correction intact. Central capex ran near ₹2 lakh crore a year in FY2014-15 and is budgeted at ₹12.22 lakh crore for FY2026-27 — a genuine five-to-six-fold rise, and capex as a share of GDP roughly doubled from 1.6% to 3.2%. The honest version of this row is still a strong one.

RBI publishes end-March market capitalisation as a standing series. Checked against it, the left cell is understated by roughly three times and the right cell is overstated by around ₹45–50 lakh crore. Even measured to the all-time September 2024 peak of ~₹477 lakh crore, the NDA-period gain reaches about ₹326 lakh crore — not ₹370.

The choice of 2009 as a baseline is doing quiet work. March 2009 was the bottom of the global financial crisis, with the Sensex near 9,700 — the single most flattering starting point available for whoever comes next, and the least flattering for whoever is being measured from it.

Market capitalisation is a valuation, not an output. It rises when multiples expand, when new companies list, and when the rupee moves — none of which a finance minister creates.

Roughly ₹13 lakh crore of the “wealth created” under any government is simply new paper: IPOs and fresh listings entering the index, which add to market cap without any existing holder becoming richer.

This row holds. Both figures sit inside the tolerance you would expect from rounding and from a choice of measurement date. It is the most defensible cell on the card.

One qualification worth carrying: a meaningful share of the recent build is valuation, not accumulation. Gold has gone from roughly 6% of reserves in 2021 to about 14–17% by 2025–26, driven by both RBI purchases and a sharp run in the gold price. Reserves are also a two-way buffer — they were drawn down in 2022 to defend the rupee, and rose sharply in 2023–26 partly because inflows returned.

The starting point is sound. The end point is out of date and currently contested.

India was widely reported as fourth-largest in 2025 on the strength of IMF projections. Two things then moved against that: MoSPI shifted the GDP base year from 2011-12 to 2022-23 in early 2026, cutting nominal GDP by roughly 3–4% (FY26 revised from about ₹357 lakh crore to ₹345 lakh crore); and the rupee fell from about 84.6 to 88.5 to the dollar. The IMF’s April 2026 outlook places India sixth, behind both Japan and the UK.

Neither the fall nor the earlier claim says much about the economy. Nominal dollar rankings move on exchange rates and statistical revisions.

The substantive fact underneath — that India has climbed several places since 2013 and remains the fastest-growing major economy, third by purchasing-power parity — is not in dispute. “5th in 2026” simply is not what the current data says.

Accurate, and among the better-evidenced claims on the card. Retail inflation has not crossed 10% since 2012, and has averaged close to 5% over the last decade — inside the RBI’s statutory 2–6% band.

Two pieces of context the card omits. The inflation-targeting framework itself was built across both governments: the Urjit Patel Committee that designed it reported in January 2014 under the previous government, and the monetary policy committee was legislated in 2016. And the 2012–14 spike coincided with crude above $100/bbl, while much of the post-2014 period ran with far cheaper oil — a tailwind no finance minister arranged.

Note also that MoSPI moved the CPI base year from 2012=100 to 2024=100 in February 2026, so series spanning that break are not strictly continuous.

There is no institution in India that publishes a scam count. “78” traces to chain-forwarded op-ed content circulating from mid-2024, not to any audit body. It has no definition, no inclusion criteria and no source document — which is precisely why it is unfalsifiable and therefore useful in a graphic.

“0” is an assertion of the same type. Whatever one concludes about it, it has to sit alongside the Supreme Court striking down the electoral bonds scheme as unconstitutional in February 2024, and the SEBI proceedings that followed the Adani–Hindenburg allegations.

A cell with no measurable quantity on either side is not evidence. It is the punchline the other seven rows were arranged to set up.