BHARTIARTL The Distortion That Shrank Last Quarter's Headline Flipped and Inflated This One

The Same Distortion, Pointed the Other Way

Last quarter's post spent its opening explaining an unusual inversion: Bharti's headline net income growth (43.0%) understated the real, like-for-like growth (103.4%), because the year-ago quarter carried a one-time exceptional credit that inflated its base. This quarter, the same mechanism runs in the opposite direction. The audited results show zero exceptional items again - the second straight quarter with a clean nil, extending the streak first flagged three months ago - but this time the year-ago comparative (Sep-24) carried a Rs853.7 crore exceptional charge (a currency-devaluation writedown, previously covered here), not a credit. A charge in the base makes this year's growth look bigger than it really was on a clean basis, not smaller.

Owners' net income (the number most coverage will quote) grew 89.0% YoY, from Rs3,593.2 crore to Rs6,791.7 crore. On the company's own like-for-like basis - net income before exceptional items, which strips out that charge from the Sep-24 base - growth was a real but meaningfully smaller 69% (see Key Financial Metrics for the caveat on that comparative figure). EBITDA margin hit a fresh site-high of 58.1%, beating Jun-25's 57.3% and Mar-25's prior high of 57.4%. Revenue grew 25.7% to Rs52,145.4 crore, still inflated by Passive Infrastructure Services (Indus Towers) sitting in this year's base but not last year's - Indus only became a consolidated subsidiary on November 18, 2024, so Sep-24 is the last "clean" quarter before that changeover and every YoY comparison this quarter carries that asterisk.

Two threads from last quarter resolved differently. The finalized Bharti Airtel-to-Indus tower sale wasn't revisited this quarter - nothing left to disclose, since last quarter's reconciliation already closed it out. But the Bharti Hexacom tower-transfer pause and the Adani Data Networks spectrum deal both went unmentioned for a second consecutive quarter - two threads now sitting in the same silence (see Beyond the Usual). And a genuinely new capital-allocation signal emerged on the call: management disclosed, only in response to an analyst question, that the board has taken an "enabling provision" to raise Bharti's stake in Indus Towers from 51% to 56%.

The Prescription

Bharti should keep building out Xtelify as a real, separately branded software business, not just an internal tool dressed up for outside customers. This quarter the internal data/workflow engine that Gopal Vittal spent unusual time on last quarter got its own name - Xtelify - and its own product: a "built-in-India, telco-grade sovereign cloud platform" claiming up to 40% cloud-spend optimization for customers, with Singtel, Globe Telecom, and Airtel Africa named as signed partners. Xtelify also says it's in 70+ conversations with large customers and has already closed deals in BFSI and manufacturing. A named product with named external partners is a more credible diversification story than "we use this internally and some others might too" - Bharti should keep pushing Xtelify toward standalone disclosure (its own revenue line, its own customer count) rather than leaving it folded into Airtel Business commentary, since that's what would let the market actually price the optionality instead of taking it on faith.

What it should stop doing: using a Q&A aside as the disclosure channel for a material capital-allocation decision. The plan to raise Bharti's Indus Towers stake from 51% to 56% - a real move on a subsidiary already worth Rs28,944 crore in segment assets this quarter - surfaced only because an analyst happened to ask about it on the earnings call; it isn't in the quarterly results, the investor presentation, or any subsequent-events note in the company's own filed disclosures. Management's own reasoning for the move (Indus is "clearly undervalued," "a strong dividend paying asset") is a sound capital-allocation argument on its face, but a shareholder who skips the call transcript would have no way to know the board had authorized it. A company disciplined enough to run at 1.63x net debt/EBITDA and generate Rs18,556.7 crore of quarterly operating free cash flow doesn't need to let a Q&A answer carry disclosure weight that belongs in a press release.

Key Financial Metrics

Q2 FY26 (quarter ended Sep 30, 2025) vs. Q2 FY25, consolidated, audited

FX rates below are the company's own disclosed RBI reference rates for the quarter: INR 87.28 = USD 1 (average, for income-statement lines) and INR 88.76 = USD 1 (September 30, 2025 close, for balance-sheet and market-cap figures); the year-ago quarter used INR 83.73 (average) and INR 83.67 (close), both per this quarter's own disclosed comparative rates.

Metric	Q2 FY26 (INR)	Q2 FY26 (USD)	Q2 FY25 (INR)	YoY
Revenue	Rs52,145.4 crore	~$5.97B	Rs41,473.3 crore	✅ +25.7% (still inflated by Indus Towers sitting in this year's base but not last year's)
EBITDA» (audited results' own "profit before D&A, finance costs, share of JV, exceptional items and tax")	Rs30,289.1 crore	~$3.47B	Rs22,100.9 crore	✅ +37.1%, margin 58.1% - a fresh site-high, beating Jun-25's 57.3%
Operating Income» (EBIT, derived: audited EBITDA minus audited depreciation & amortisation)	Rs17,107.0 crore	~$1.96B	Rs11,100.9 crore	✅ +54.1%, margin 32.8% vs 26.8%
Profit before tax (audited, statutory)	Rs12,322.3 crore	~$1.41B	Rs5,897.4 crore	✅ +108.9% (this year has zero exceptional items; last year's PBT was reduced by an exceptional charge - see below)
Net Income» before exceptional items (owners, company's own comparative basis)	Rs6,791.6 crore	~$778M	Rs4,024.3 crore	⚠️ +68.8% - the company's own like-for-like comparison, but the Sep-24 base here is the deck's "re-instated" comparative, not the original as-reported figure (same caution flagged in prior posts)
Net Income (total profit for the period, after exceptional items, audited)	Rs8,650.8 crore	~$991M	Rs4,153.4 crore	✅ +108.3%
...of which, attributable to Bharti Airtel's own shareholders (audited)	Rs6,791.7 crore	~$778M	Rs3,593.2 crore	✅ +89.0% (overstates the like-for-like growth - see above)
...of which, attributable to non-controlling interests» (audited)	Rs1,859.1 crore	~$213M	Rs560.2 crore	✅ +231.9%
Basic EPS	Rs11.72	~$0.13	Rs6.21	✅ +88.7%
Capex (company-reported)	Rs11,362.3 crore	~$1.30B	Rs7,674.9 crore	⚠️ +48.1% - the comparative here is the 2024-09 post's own as-reported figure, not the investor deck's recast Sep-24 comparative (which is inflated by retroactively adding Indus's capex); part of the real growth is still genuinely Indus entering the base for the first time
Operating Free Cash Flow» (EBITDA minus capex, company-reported style)	Rs18,556.7 crore	~$2.13B	Rs14,346.0 crore	✅ +29.4% - same as-reported (non-recast) Sep-24 comparative as the capex row above
Total Cash (cash and cash equivalents, balance sheet, audited)	Rs6,298.2 crore	~$710M	Rs4,738.0 crore	✅ +32.9%

Note on the audited PBT and net-income rows: this quarter's Rs12,322.3 crore profit-before-tax figure is identical to "profit before exceptional items and tax," confirming a genuine nil for exceptional items - the second consecutive quarter this has happened. The Sep-24 comparative carried an exceptional charge of Rs853.7 crore (net), which pulled that quarter's PBT down from Rs6,751.1 crore (before exceptional) to Rs5,897.4 crore (after) - the opposite effect from Jun-24's exceptional credit, which pulled up the base used in last quarter's comparison. That's the whole mechanism behind this quarter's 89.0%-vs-69% gap.

Balance sheet metric	Sep 2025 (INR)	Sep 2025 (USD)	Sep 2024 (INR)	YoY
Total Assets	Rs523,862.9 crore	~$59.02B	Rs460,982.1 crore	⚠️ +13.6% (Indus Towers not yet a consolidated subsidiary a year ago)
Total Liabilities	Rs363,508.9 crore	~$40.96B	Rs350,679.1 crore	⚠️ +3.7%
Equity attributable to owners of the parent	Rs117,796.0 crore	~$13.27B	see note	✅ up from Mar-25's Rs113,671.9 crore - QoQ comparison, see note
Book value per share (company-disclosed, adjusts for partly-paid shares)	Rs203.10	~$2.29	n/a	⚠️ down slightly from Jun-25's Rs204.6
Net Debt (company-reported, incl. leases)	Rs194,713.1 crore	~$21.94B	see note	⚠️ see note below
Net Debt excluding lease obligations	Rs126,651.3 crore	~$14.27B	see note	⚠️ see note below
Net Debt/EBITDA (annualized, company-reported)	1.63x		n/a - see note	✅ improved from Jun-25's 1.70x
Interest Coverage Ratio (company-reported, ISCR)	6.69x		4.93x	✅ improved from this quarter's own Jun-25 comparative of 6.35x

Total Assets and Total Liabilities YoY come directly from this quarter's audited segment note, which discloses a genuine Sep-24 comparative (unlike the investor deck's recast figures) - a real, usable YoY for once. Owners' equity is shown QoQ instead, since the segment note only splits total equity (owners + non-controlling interests) for Sep-24, not the owners-only portion. Net Debt's YoY row stays "see note" for the same reason flagged in the last two posts: the deck's own Sep-24 comparative (Rs205,620.4 crore incl. leases) predates Q3 FY25's Indus consolidation and is explicitly a re-instated, not originally-reported, figure. The reliable comparison is QoQ: net debt including leases rose slightly to Rs194,713.1 crore from Jun-25's Rs191,579.4 crore, a real, non-recast figure both times.

Key Operational Metrics

An investor presentation and earnings call transcript were available for this quarter; no separate press release was sourced.

Metric	Sep-25	Sep-24	YoY
Total customer base (consolidated)	~623,554,000	~563,170,000	✅ +10.7%
India customer base	~449,738,000	~406,528,000	✅ +10.6%
Africa customer base	~173,816,000	~156,642,000	✅ +11.0%
Mobile Services India ARPU»	Rs256	Rs233	✅ +9.9% (benefited from continued prepaid-to-postpaid and feature-to-smartphone mix improvement, per management)
Postpaid base (Mobile Services India)	~27,523,000	~24,667,000	✅ +11.6%
5G customer base (India, per management commentary)	~167,000,000	~152,000,000 (QoQ comparison, per Jun-25 post; no Sep-24 base was disclosed this quarter)	✅ +9.9% QoQ
Homes Services net additions (quarter)	~951,000		✅ another record pace, ahead of Jun-25's 939,000
Homes Services customer base	~11,900,000		✅ +39.5% YoY per management
Digital TV (DTH») customer losses (quarter)	~340,000 lost		⚠️ worse than Jun-25's ~200,000 lost, management attributes it to "pronounced seasonality"
Airtel Payments Bank monthly transacting users	~104,802,000	not directly comparable	✅ crossed 100 million this quarter

Management framed ARPU growth in the same terms as the last three calls - "featurephone to smartphone upgrades, prepaid to postpaid upgrades, data monetization, and international roaming" - with no fresh mention of a tariff increase ask. A genuinely new regulatory item on AGR^» came up in the call's opening remarks too - see Beyond the Usual.

Segment Comparison

Bharti Airtel still reports seven consolidated segments, with Mobile Services South Asia disposed since June 2024. This is the fourth consecutive quarter with a genuine, audited Passive Infrastructure Services (Indus Towers) consolidation, but the third straight quarter where its YoY comparison isn't available - Indus wasn't a subsidiary in Sep-24 - while every other segment's YoY is now clean.

Segment	Revenue (Q2 FY26)	Revenue (Q2 FY25)	YoY	Segment Result (Q2 FY26)	Margin (Q2 FY26)	Margin (Q2 FY25)
Mobile Services India	Rs28,116.7 crore	Rs24,837.1 crore	✅ +13.2%	Rs8,815.0 crore	✅ 31.3%	24.9%
Mobile Services Africa	Rs13,679.5 crore	Rs10,163.1 crore	✅ +34.6%	Rs4,489.2 crore	✅ 32.8%	30.6%
Airtel Business	Rs5,276.0 crore	Rs5,655.5 crore	⚠️ -6.7%	Rs1,566.7 crore	✅ 29.7%	25.4%
Passive Infrastructure Services (Indus Towers, full quarter)	Rs8,222.2 crore	n/a - not a subsidiary yet		Rs2,894.4 crore	35.2%
Homes Services	Rs1,864.6 crore	Rs1,432.1 crore	✅ +30.2%	Rs301.4 crore	⚠️ 16.2%	23.6%
Digital TV Services	Rs753.2 crore	Rs758.6 crore	⚠️ -0.7%	Rs(60.6) crore	⚠️ -8.0%	0.2%
Others	Rs93.4 crore	Rs89.8 crore	✅ +4.0%	Rs65.7 crore
Total segment revenue	Rs58,005.6 crore	Rs42,936.2 crore	✅ +35.1%	Rs16,817.8 crore

("Segment Result" is profit before finance costs, charity/donation, exceptional items and tax - an EBIT-level margin - per the audited segment note, consistent with prior posts.)

Mobile Services India's margin expansion continued, up another 6.4 percentage points YoY to 31.3%, ahead of Jun-25's 30.8% and extending the reversal tracked since Dec-24 to a new high across every quarter covered here. Africa's margin also expanded, to 32.8% from 30.6%, on both constant-currency growth and a favorable currency swing this quarter. Airtel Business kept the same pattern as last quarter - revenue down 6.7% on the deliberate commodity exit, margin up to 29.7% from 25.4% - confirming the exited low-margin business barely mattered to profit. Homes Services margin fell further still, to 16.2% from 23.6%, extending the multi-quarter margin compression flagged in each of the last several posts even as customer net-adds hit another record, as FWA and fiber capex continues to outrun what it's converting to segment profit. Digital TV's segment result went meaningfully negative (-8.0% margin), its steepest loss yet - Mar-25 dipped fractionally negative first (a Rs0.2 crore loss, rounding to roughly 0%), and Jun-25 recovered to a thin positive 0.3%, but this quarter's loss is an order of magnitude larger than either - as the subsidy-elimination transition management flagged two quarters ago continues to bite segment profitability even as it helps cash flow.

Airtel Business and Homes: capex stepped back up, in different directions

Two quarters ago Airtel Business capex normalized down after a one-time Cloud-platform buildout spike. This quarter, consolidated capex rose to Rs11,362.3 crore, up 48.1% against the real, as-reported Sep-24 figure - though a meaningful chunk of that growth is simply Indus's own Rs2,584.5 crore of capex entering the base for the first time, the same caveat that applies to this quarter's revenue and EBITDA growth. Even stripping Indus out, capex still rose to roughly Rs8,777.8 crore - spread mainly across Homes (Rs1,860.6 crore, funding the FWA/fiber land-grab management described on the call). Management's own framing on the call was that 5G standalone itself needs "really no meaningful capex" since it's software-led, but flagged a possible "bump-up" from the new Nxtra/Google data-center partnership in Vishakhapatnam - explicitly not yet sized, with management calling it "premature to talk about" a guidance change. Worth watching whether that data-center investment shows up as a fresh line item in the next post.

Mobile Services India and Homes: still the two engines, still diverging on capital efficiency

Mobile Services India ARPU crossed Rs256, continuing the steady climb from Jun-25's Rs250 and Mar-25's Rs245. Homes Services posted its fourth consecutive record net-add quarter (~951,000, edging past Jun-25's 939,000), but margin compression continued in lockstep - the same tension flagged in each of the last three posts, still with no inflection.

Beyond the Usual

A material Indus Towers stake increase surfaced only in a Q&A answer, not in any filed document

An analyst on this quarter's call asked about "the announcement to raise the equity stake in Indus Towers by 5%." Management confirmed the board has taken "an enabling provision to climb by 5%" - from the current ~51% to potentially 56% - describing Indus as "a strong dividend paying asset" and "a highly undervalued asset" worth allocating capital toward. Whether or when that provision gets activated is, in Vittal's own words, "a speculative question because that, we have not come to that bridge." Nothing about this appears in the quarter's audited results, the investor presentation, or any subsequent-events note in the company's own filed disclosures - the only record of it is the transcript of an analyst's question and the answer that followed. A capital-allocation decision material enough for a board resolution deserves a disclosure channel more durable than a Q&A transcript.

Two threads from last quarter went quiet for a second straight time

The Bharti Hexacom tower-transfer pause - already unmentioned last quarter after being disclosed as "in abeyance" the quarter before that - stayed silent again: nothing in the results, the presentation, or either half of this quarter's earnings call. The Adani Data Networks 26 GHz spectrum deal, announced last quarter as a definitive agreement pending regulatory approval, also got zero mentions anywhere in this quarter's documents - no update on whether approval has been sought, granted, or is still pending. Two separate threads, both now two quarters deep into silence, is a pattern worth naming rather than two isolated non-events: neither is necessarily stalled, but a reader following this site has no way to tell the difference between "quietly progressing" and "quietly shelved" for either one.

Four smaller, genuinely interesting items from this quarter, none a criticism of the company:

The Supreme Court's recent order on the telecom industry's long-running AGR dispute permits the government to undertake "a comprehensive assessment, reassessment, and reconciliation" of AGR dues, including interest and penalties, through FY2016-17 - a meaningfully more favorable posture than the outright rejection of the industry's curative petition covered here two years ago. The order was made in Vodafone Idea's petition specifically, but Bharti said on the call it is "now planning to take up our matter with the government" as well.

Airtel and Google announced they will jointly build a data center in Visakhapatnam under Nxtra, Bharti's data-center subsidiary - management's stated ambition is to reach "over 1 gigawatt" of capacity within a few years, roughly four times Nxtra's current operating capacity, though the capex commitment for this specific facility wasn't sized on the call.

The auditor's report carries the same "Emphasis of Matter" this quarter as in prior quarters covered on this site: material uncertainty about the ability of one of Indus Towers' largest customers to continue as a going concern, and the consequential exposure to Bharti's own consolidated receivables, property, plant and equipment, and intangible assets tied to that customer. This is a recurring disclosure, not a new development, but it's a live reminder of the counterparty risk flagged here as far back as the 2023-03 post - now sitting directly on Bharti's own consolidated balance sheet rather than one step removed via the equity method.

Airtel Africa's ongoing share buyback continued into a further tranche - $34.7 million returned via 14.2 million shares as of September 30, 2025, out of the $55 million second tranche - and the company arranged with Barclays to keep the buyback running autonomously through closed periods if needed. Combined with mechanical creep from the buyback shrinking the share count, Bharti's effective economic stake in Airtel Africa rose to 62.60% (from 62.47% at the end of last quarter), continuing the steady pattern tracked across the last several posts.

Management's Framing: A Branded Platform, a New Capital-Allocation Lever, and the Same Silences

Vittal's opening remarks introduced "Xtelify" by name for the first time on this site - branding the internal data/workflow engine discussed but unnamed last quarter as a distinct digital arm with its own sovereign-cloud product and the same Singtel/Globe Telecom/Airtel Africa partnerships cited before. This belongs at the top of the read because it's the clearest sign yet that the platform pivot is graduating from an internal capability to a named, externally-facing business (see The Prescription). On capital allocation, the quarter's real news wasn't dividend policy - for the first time across four quarters of this thread, no analyst asked about it at all, and management didn't volunteer anything either. Instead, capital-allocation questions centered entirely on the new Indus Towers stake increase (see Beyond the Usual), a genuinely different capital-allocation conversation than the "when will the dividend framework arrive" question that dominated the last four calls. Whether that's a permanent shift in what the market wants explained, or just what came up given time constraints on this particular call, is worth watching into the next post.

The Stock Fell 6.5% This Quarter, and Rose 102.8% Over Two Years

Bharti Airtel's shares closed at Rs1,878 on the BSE on September 30, 2025, per the company's own disclosed closing price - down 6.5% from the Rs2,009 close on June 30, 2025 covered in the last post, the first quarterly decline on this site since the 7.1% drop covered in the Dec-24 post. Zooming out, the stock is still up roughly 102.8% over the two years to September 30, 2025 (from Rs926.40 on September 29, 2023, independently sourced) - a smaller two-year gain than Jun-25's roughly 128.6%, reflecting both this quarter's price decline and a somewhat higher two-years-ago base. The company has not split its stock since 2009, so no split adjustment is needed for either price. As always: the stock's mood swings quarter to quarter regardless of what the underlying business is actually doing - this quarter, the business kept growing while the stock pulled back, echoing the same divergence seen in the Dec-24 quarter.

Target Valuation Range

Enterprise value Rs1,320,900 crore, implying a perpetual FCF growth rate of roughly 5.5% - down from ~6.0% last quarter, reversing that quarter's uptick. The stock got cheaper on every headline multiple this quarter, reversing last quarter's re-rating, and the reverse DCF confirms this is a genuine cooling in the market's growth expectations, not just an artifact of the earnings base.

Market cap → enterprise value	Q1 FY26	Q2 FY26
Market capitalization	Rs1,205,900 crore	Rs1,126,200 crore (~$126.9 billion)
Book value/share	Rs204.6	Rs203.10
Enterprise value	Rs1,397,500 crore	Rs1,320,900 crore (~$148.8 billion)

Peer-multiple sanity check	Q1 FY26	Q2 FY26
Trailing P/E	32.42x	27.77x (company-reported; reversing last quarter's sequential increase - trailing EPS now benefits from two consecutive exceptional-item-free quarters, so this fall on a strengthening earnings base is a real re-pricing)
P/B	9.67x	9.10x (company-reported, Market Price/Book Value; tracking the share-price decline)
EV/EBITDA	12.40x	11.04x (company-reported; the first sequential decline in three posts)

Reverse DCF: Solving for the perpetual free-cash-flow growth rate implied by enterprise value, using the trailing four quarters' company-reported Operating Free Cash Flow (Dec-24 through Sep-25: Rs19,895.9 crore + Rs13,003.5 crore + Rs19,859.8 crore + Rs18,556.7 crore = Rs71,315.9 crore) and a WACC of approximately 11.2% (updating the methodology used in prior posts: an ~11.9% cost of equity off a ~7.1% risk-free rate, 0.8 beta, and 6% India market risk premium, blended against an estimated post-tax cost of debt of roughly 7.2%, weighted at the ~85.3%/14.7% equity/net-debt split implied by this quarter's enterprise value - a slightly lower equity weighting than last quarter's 86.3%/13.7% split, since enterprise value fell while net debt held roughly flat), the market's Rs1,320,900 crore enterprise value implies a perpetual FCF growth rate of roughly 5.5% - down from Jun-25's implied 6.0%, reversing that quarter's uptick. Enterprise value fell roughly 5.5% (from Rs1,397,500 crore) while trailing free cash flow actually grew roughly 3.0% (from Rs69,271.8 crore) - the market paid less this quarter for more trailing cash flow, the cleanest signal in this post that the cooling is real and not just a base effect.

The honest read this quarter: nothing here says the stock is cheap outright - a ~5.5% perpetual growth assumption is still not a demanding one for a business growing EBITDA at 37% - but the direction matters more than the level. Last quarter this site flagged that "the market moved ahead of the business rather than the other way around" for the first time; this quarter, that gap closed again, with the business's cash generation outpacing what the market is now willing to pay for it.

---