Sources

Public discrepancy register (18 findings, numbered within D1–D25)

Every internal mismatch found while validating the anchor figures, published in full — status, materiality and resolution. Nothing is reconciled silently.

• D1resolved

  FY26 OMB = 49,788 (narrative 48,788 is a typo; CL/OMB 111% reproduces only with table figure)

• D10open_material · material

  Project tables disagree: approved 1,00,340.61 (T4.4) vs 1,19,412.47 (T4.6); pipeline range 33-59k - referred to forensic audit

• D11explained

  Table 4.1 MVT 17,593.57 is CUMULATIVE; narrative 'in 2025-26 alone' wrong (annual ~3,300)

• D12resolved

  Accumulated PSE loss canonical series 42,930->72,851 (exec '31,571->78,851' = double transcription error)

• D13standing_rule

  PSE net loss has three bases (5.6 | 5.7 | 5.A/BPE) - name the base

• D15resolved_note

  Table 3.4 tax-shortfall 5,958.93 -> 5,758.93; interim column +/-0.10 rounding slip

• D16logged

  KIIFB premium computed 0.64-1.72 pp vs report's 1-1.5 claim - both adverse

• D18open

  Table 4.5 debt series narrower than RBI liabilities by aggregate scope

• D2standing_rule

  Three debt values circulate (35.5 BE | 33.22 RE | 37.84 FY24 consolidated) - every debt figure carries year + vintage + boundary

• D21resolved

  KWA FY24 '-6,153.36' = decimal misprint ~ -615.3

• D23noted

  Growth/buoyancy claims use FY26 BE endpoints; RE-basis SOTR growth ~5.2%

• D24minor

  FY25 capex 17,086 (BiB) vs 15,082 (AG outlay) - label the basis

• D25resolved

  Ch.7 Part A '33.32%' = transposition of 33.22

• D3standing_rule

  Table 2.6 WMA/OD days are CALENDAR years - never join to FY series on the label

• D4resolved

  XV RDG 53,137 vs 37,814 - boundary on the 2020-21 first-report year; both correct

• D5resolved

  XVI gain = +0.457 pp (1.925 -> 2.382)

• D6open

  5.07 L cr headline and 56,000 cr KIIFB obligation not yet boundary-classified

• D9open

  KIIFB inflow-outflow vs fund balance gap 91.5 cr (immaterial)