Financial Shenanigans
The Forensic Verdict
Yes Bank is a reconstructed bank carrying a long forensic tail. Reported accounts today come from joint statutory auditors appointed under the post-2020 RBI rotation regime with unqualified opinions, asset quality has genuinely improved (gross NPA 1.3%, net NPA 0.2% at FY26-TTM), and capital adequacy is comfortable (CET1 13.5%). The risk is not that the current numbers are fabricated — it is that (i) live litigation from the 2020 collapse could still hit equity, (ii) reported profit growth leans on investment-provision write-backs and trading gains rather than core spread expansion, and (iii) a 7% ROE on a balance sheet that has been recapitalised four times means accounting profit and economic profit are still not the same number. Forensic risk score is 45 / 100 (Elevated). The single data point that would most move this grade: a Supreme Court ruling on the ₹8,415 crore AT-1 bond write-down dispute, which is sub judice and reserved for judgment.
Forensic Risk Score
Red Flags
Yellow Flags
CFO / NI (5y)
Accrual Ratio FY26 (%)
Elevated, not Critical. The bank has already had its accounting failure event (FY2020 RBI moratorium and reconstruction). Current statements are audit-clean. But the historic event continues to generate live legal claims, and reported earnings growth depends materially on non-spread items.
Shenanigans scorecard — 13-category coverage
Breeding Ground
Yes Bank's breeding ground reads "post-failure, still under guard." There is no promoter group. Statutory auditors rotate under RBI rules. The audit committee is chaired by an independent director and the chairman is an ex-RBI Deputy Governor. The 2020 governance disaster has been answered with structural fixes — but the legacy continues to generate enforcement action.
Net read: The breeding-ground risk is bimodal — current vintage is monitored more tightly than almost any other Indian private bank because of the failure history, but the failure history itself is the breeding-ground story. Investors should accept that residual legal and reputational risk from the Rana Kapoor era will continue to surface for several years.
Earnings Quality
Reported profit growth in FY2025 (+92% YoY net income) overstates the underlying improvement. Operating profit grew 25.6%, which is the cleaner number. The 92% reported net income jump is amplified by a 42% drop in provisions and contingencies — and that drop is driven entirely by a swing in investment-related provisions, not NPA provisions.
NPA provisions actually rose 18% to ₹2,879 cr in FY25 — that is the slippage and ageing line continuing to flow through. The headline collapse in "provisions and contingencies" comes from booking a ₹1,737 cr write-back on investment provisions (up from a ₹543 cr write-back in FY24). Stripping that out, the underlying credit cost trend is flat to up — not the deflation the net-profit line suggests.
Other income now contributes 40%+ of operating revenue — high for a domestic retail bank. Within other income, FY25 saw profit on sale of investments jump 81% to ₹411 cr and miscellaneous income at ₹1,059 cr. Treasury gains in a falling-rate environment are real but not durable; the durable component is fee income (commission/exchange/brokerage at ₹3,713 cr, up 21%).
ROE has averaged 4% over FY22–FY26 against an India bank cost-of-equity around 13–14%. The bank is earning back its capital base, not exceeding it. From a forensic lens, this matters because profit growth optics are real but the absolute return profile leaves no margin for an accounting surprise.
Cash Flow Quality
Standard CFO/NI tests do not work for a bank — deposit growth dominates the line. For Yes Bank, the relevant cash test is net interest income plus stable fee income, not statement-of-cash-flows operating cash flow.
The CFO line swings on deposit gathering — FY21 (+₹55,396 cr CFO) and FY23 (–₹25,816 cr CFO) reflect deposit flows, not earnings sustainability. For Yes Bank specifically, the more honest test is whether the deposit franchise is rebuilding without paying premium rates and whether NII is growing.
NII growth is in line with balance-sheet growth (advances +8.1% in FY25, deposits +6.8%) and NIM is stuck at 2.4%. This is the cleanest indicator that the headline net-income growth is partly cosmetic — core spread economics are flat. The lift between operating profit growth (25.6%) and net-profit growth (92.3%) is non-operating.
CASA ratio FY25
CASA improvement from 30.9% to 34.3% in FY25 is the genuine franchise progress, supported by 32% growth in savings deposits. Term deposits grew only 1.6% — i.e. the bank is shedding price-sensitive bulk and adding granular liabilities. This part of the story is real.
Metric Hygiene
Three management-favoured metrics deserve forensic adjustment.
Largest off-balance-sheet exposure: AT-1 bonds. In March 2020 the Reconstruction Scheme wrote down ₹8,415 crore of Additional Tier-1 bonds. Bondholders sued; the Bombay High Court initially ruled in their favour in January 2023; appeals went to the Supreme Court, which has heard arguments and reserved judgment. An adverse ruling could create a new liability that the disclosed CET-1 and balance sheet do not yet reflect. Management language on the Q4 FY26 call was deliberately non-committal ("matter is sub judice", "we respect the proceedings of the court").
What to Underwrite Next
The accounting risk here is a position-sizing limiter, not a thesis breaker. The current franchise is being rebuilt under unusually strong external oversight; the open question is whether legacy contingencies materially impair equity before the new vintage of earnings compounds enough to absorb them.
Highest-priority diligence items for the next 12 months:
Supreme Court AT-1 bond ruling. Track docket filings and the Court's calendar. An adverse ruling would force recognition of up to ₹8,415 cr (or a settled portion thereof) as a liability — a 14–18% hit to book equity at FY26 levels. A favourable ruling would compress this risk to zero.
ED / SEBI cases on the 2018-19 Anil Ambani exposures. The criminal/quasi-criminal proceedings target former management and counterparties, not the current bank. But monitor for any direct order against the bank for restitution, regulatory penalty, or claw-back of recoveries already booked. The August 2025 SEBI rejection of the settlement plea kept the matter live.
Investment provision write-back recurrence. If FY26 results again show provisions dropping because of investment write-back, that confirms a non-durable earnings lever. Look for separate disclosure of NPA provisions, standard-asset provisions, investment provisions, and "other" provisions in the Q4 FY26 standalone results.
Other income disaggregation. Demand a clean split of: (a) commission and fee income, (b) FX and trade income, (c) profit on sale of investments, (d) revaluation gains/losses, (e) recoveries from written-off accounts (the ARC channel), (f) miscellaneous income. Item (e) is non-repeatable; items (c)/(d) are mark-to-market.
SMBC integration and any "fresh-start" accounting. SMBC took a roughly 20% stake in Q2 FY26. Watch FY26 annual report for purchase-accounting adjustments, related-party transactions in trade-finance flows to Japanese corporates in India, and any change in audit-committee composition.
Signal that would downgrade the grade to "Watch" (21–40):
Supreme Court rules in the bank's favour on the AT-1 matter AND FY27 net-profit growth comes with NPA provisions stable or rising. That combination would confirm that the legacy file is closing and earnings quality is genuine.
Signal that would upgrade the grade to "High" (61–80):
Adverse AT-1 ruling AND fresh disclosure of any sec-143(12) fraud above ₹50 cr AND another provision write-back propping net profit by more than 30%. Any two of those three would flip the file.
Effect on valuation and position sizing: The forensic findings argue for a 10–15% holding-company-style discount to book value relative to better-vintage private banks (HDFC Bank, ICICI Bank), a tighter position-size cap (single-name limit) until the AT-1 docket resolves, and refusal to model FY25's 92% net-profit growth as a forward run-rate. The accounting risk is not a footnote and it is not yet a thesis breaker — it is a margin-of-safety requirement.