HDFC Bank Integrated Annual Report 2025-26: A Review

Official report: Get the HDFC Bank Integrated Annual Report 2025-26 from the company's website. This page is The Footnotes' independent review of it.

Verdict: HDFC Bank has the most rigorous materiality and climate accounting of any Indian bank report we have read, and it buries the year's biggest story. The outcome of the review into its Chairman's resignation gets one sentence, on page 15 of a 617-page report. Score: 26 of 35.

Everything is in one PDF: about 250 pages of integrated narrative, then the Directors' Report, the corporate governance report, the accounts and the BRSR, with shareholder information at page 617. Page numbers below are the printed ones. Screenshots are small excerpts reproduced for review and remain the copyright of HDFC Bank Limited; the full report is on the Bank's investor relations pages.

The resignation, and where the answer is

On 18 March 2026, Atanu Chakraborty resigned as Part-time Chairman and Independent Director with a statement that, in the interim Chairman's own words, "led to speculation on the Bank's governance standards". Keki Mistry became Interim Part-time Chairman the same day. The Board engaged domestic and international law firms to review the concerns.

Here is where the report tells you what happened:

• The Chairman's message (pp. 8-11) opens with IMF growth estimates, inflation, the West Asia conflict and India's GDP. Only under a later heading, "Corporate Event", does it describe the resignation and the legal review, and say the findings were received on 26 June 2026 and sent to the stock exchanges. It doesn't say what they were.
• The MD & CEO's message (p. 15) does, in one sentence: the law firms' findings were "in essence" that Mr Chakraborty's statement "and its implications were not substantiated by the record reviewed and witness interviews".
• The corporate governance report (p. 325) records that the Board set up a Special Committee of four independent directors on 26 March 2026 (M. D. Ranganath, Lily Vadera, Harsh Kumar Bhanwala and Sandeep Parekh) to oversee the legal review. It does not record the conclusion.

So the answer is in the report, but a reader has to know to look in the CEO's letter rather than the Chairman's or the governance section. For a document built on integrated thinking, the year's most material governance event deserved the first page and a paragraph in the Board's own voice.

The leadership change has continued since. The report notes that Rajiv Kumar was appointed Part-time Chairman on 29 June 2026, subject to approvals. RBI approved the appointment for three years from 15 July 2026. RBI has also approved Anup Bagchi as MD & CEO for three years from 27 October 2026, when Sashidhar Jagdishan's tenure ends, subject to shareholder approval.

The best materiality chapter in Indian banking

The Bank refreshed its double materiality assessment in FY 2025-26 under GRI Universal Standards 2021. The chapter sets out an eight-step method, explains that the 14 topics from the previous cycle were revisited, that survey responses were weighted by stakeholder group, and that scores reflect scale, likelihood and breadth. The result is plotted (p. 68).

Each topic then gets a page: a definition, impact and financial materiality described separately, the capitals involved, SDGs, and a performance figure. This is the most complete materiality disclosure we have seen in an Indian report, and it is dated. Reliance's, by contrast, points readers back to its 2022-23 report.

Climate: low financial materiality, 35 million tonnes financed

One placement on the matrix needs explaining. Climate change sits in the quadrant for high impact but low financial materiality.

Eighty pages later (p. 151), the Bank reports financed emissions using the PCAF method: 34.96 million tCO2e across ₹2.67 lakh crore of business and project loans to 313 listed borrowers in 39 sectors, including metals, mining, petroleum, automotive and energy supply. The average data-quality score is 1.11, and there is an honest paragraph on data limitations.

For a lender, that is what the financial materiality of climate looks like. Its own operations emit about 0.33 million tCO2e, so the loan book is roughly a hundred times larger. Either the matrix is scoring operational climate impact only, in which case it should say so, or it is understating a transition risk the Bank has itself measured. There is a target for carbon neutrality in operations by FY 2031-32, and none for financed emissions.

The same page repeats the paragraph beginning "The Bank acknowledges the evolving nature of financed emissions accounting" in full, twice in a row. A small slip, and in a report this long, a telling one.

Assurance is at the top of the range: SGS India gives reasonable assurance on BRSR Core KPIs and limited assurance on other indicators.

A Board that owns the report

The "About the Report" page (p. 4) is a model. It names the frameworks, states the boundary, confirms no restatements, names the assurance provider, and describes how the report itself is governed. It is prepared under the Group CFO, reviewed by senior management, and "formally reviewed and approved by the Board of Directors".

That last line is what the <IR> Framework means by a statement of responsibility. Infosys and Tata Steel both stop at management review.

The capitals, 176 pages apart

The value creation model (pp. 60-61) tabulates, for each of six capitals, inputs, outcomes, and the risks and SDGs that apply. Every cell has a figure: 2,11,178 employees and 23.12% attrition under human capital; ₹1,316 crore of CSR and 10.1 crore customers under social; about 28% sustainable finance under natural.

A separate "Connecting the Capitals" chapter (p. 236) does something we haven't seen elsewhere. It writes out, in prose, how each capital depends on the others: financial to manufactured through branches and business correspondents, intellectual to customer through digital journeys.

The problem is distance. The model is on page 60 and the connections are on page 236. TCS puts the same content on one spread. The model is also a table of inputs and outcomes rather than a flow through business activities.

What the year's numbers say

The highlights (p. 7) and the MD & CEO's message give the results cleanly. Advances grew 12.1%, against 5.4% the year before. Deposits grew 14.4%. Profit after tax was ₹74,671 crore, return on assets 1.94%, and gross NPA 1.15%.

Some of the more telling numbers are further back:

• Deposit mix (p. 260). CASA deposits were 34.1% of the total. Time deposits grew fastest, up 15.5% to ₹20,44,953 crore, against 11.9% for savings and 12.9% for current accounts. Funding growth is coming from the most expensive kind of deposit.
• Lending mix (p. 260). Gold loans grew 34%. Retail mortgages grew 6.3%.
• Shareholder returns (p. 14). The Bank issued its first-ever bonus shares, 1:1, and paid ₹15.50 of dividend including a ₹2.50 special interim.
• Reach (p. 16). 234 branches were added, taking the total to 9,689.

The ten-year financial table is at page 238. What is missing is a short section on how the merger with HDFC Ltd still affects the balance sheet and margins, which a new reader needs.

617 pages

The design is clean, with section colour bars and a consistent capital icon set. The problem is volume: about 250 pages of narrative before the statutory sections, with each chapter running to dozens of pages. The value-creation story is there, but the reader has to find it. This report would be stronger at half the length, with detail moved to the online version.

The contents page links to an online edition, and the value creation model invites readers to "click the number beside each element". The PDF mixes single pages and spreads, which works better on a phone than all-spread files.

What other companies can copy

• A Board-approved "About the Report" page (p. 4), including how the report is governed.
• A dated double materiality matrix, with a page per topic (p. 68 onwards), impact and financial materiality kept separate.
• A financed-emissions disclosure with the method, coverage, data-quality score and limitations (p. 151), if you are a lender.
• A chapter that writes out how the capitals connect (p. 236), ideally next to the model rather than 176 pages away.

What not to copy: the length, a Chairman's letter that leads with macroeconomics in a year with a governance crisis, and a review outcome stated only in passing.

The score

How it compares

Every review on The Footnotes uses the same seven-point card, so scores can be compared. Scores are editorial opinion of the published document.