India's Cash-Flow Giants Are Beaten Down While the Market Pretends the Problem Is Bond Yields

The headlines say rising India bond yields are the problem. The cash flow says something else entirely.

India's 10-year government bond yield is hovering near 6.78 percent — up from 6.64 percent a year ago, yes, but nowhere near historically punishing levels. The all-time high was 14.76 percent in 1996. What's actually happened in 2026 is something bigger. Thirty constituents of the Nifty 50 have collectively erased ₹22.64 lakh crore — roughly $270 billion — in investor wealth. The index itself is down more than 9 percent year-to-date.

Bond yields make a convenient macro explanation. The real correction is sector-specific, narrative-driven, and in at least two cases, has pushed massive cash-flow generators to multi-year valuation lows. That is the setup worth paying attention to.

The IT Crash: 55 Percent Off, Still Printing Cash

TCS is down 55 percent from its August 2024 peak. Infosys is down 47 percent from its December 2024 high. Both hit multi-year lows in June after Accenture — a global proxy for offshore IT demand — slashed its FY26 revenue guidance to 3-4 percent. The Nifty IT index fell 5 percent that day to a three-year low.

The market narrative is AI disruption: clients are automating work that Indian IT firms used to staff with thousands of consultants. It sounds structural. It sounds terminal.

But India's largest mutual fund — Parag Parikh Flexicap, with AUM exceeding ₹1.1 lakh crore — has been buying into the crash. In June 2026, the fund added 54 lakh shares of Infosys (pushing the holding to 4.27 crore), 18.26 lakh shares of TCS (to 1.77 crore), and 31.15 lakh shares of HCL Technologies. Fund manager Rajeev Thakkar's reasoning is blunt: current cash flows, not AI narratives.

The financial proof path is straightforward. These companies earn in dollars and pay costs in rupees. The rupee has been weak, which means every dollar of revenue carries more rupee purchasing power than it did a year ago. Wage inflation remains subdued. And the core IT job — taking specifications, delivering code, maintaining systems — hasn't fundamentally changed regardless of whether more of it is assisted by machines.

Even at these battered prices, top large-cap IT names offer roughly a 6.7 percent free cash flow yield and a 5.7 percent shareholder yield, backed by net-cash balance sheets. A 6.7 percent FCF yield means an investor paying the current price gets back roughly 6.7 percent of that price in cash every year, before any growth or multiple expansion. That is not a distress signal. That is what a cash cow looks like when sentiment turns ugly.

The market is still pricing the old AI-disruption fear. But the operating path hasn't broken. Accenture itself is generating high free cash flow while it ramps M&A from $5 billion to $9 billion — a sign that cash generation in the broader IT-services ecosystem remains intact even as organic deal signing softens.

The inflection comes from the other side: AI implementation spending. Clients are moving from proof-of-concept to live AI use cases. The revenue uplift from that migration hasn't shown up in volumes yet, but brokerage estimates suggest it could begin offsetting discretionary spending weakness in the second half of FY27. That is a 12-month bridge, not a promise. But it is enough of a path to make the current discount to the 10-year average P/E — roughly 36 percent — look like overreaction rather than repricing.

HDFC Bank: Merger Friction, Not Business Failure

HDFC Bank's stock has fallen 25.85 percent year-to-date, dropping from ₹991 at year-end 2025 to ₹735 in early August. The market cap has shrunk by ₹3.9 trillion to ₹11.32 trillion. Christopher Wood at Jefferies sold his position. Foreign investors still hold 41.8 percent of the stock, but the selling pressure has been relentless.

The headline reason is the post-merger hangover. The July 2023 reverse merger with parent HDFC doubled the balance sheet but pushed the credit-deposit ratio (loans relative to deposits, a measure of funding stability) to elevated levels. The enlarged loan book outpaced cheap deposit growth, forcing reliance on costlier funding. Net interest margins — the spread between what the bank earns on loans and pays on deposits — slipped from pre-merger highs into the sub-3.3 percent range. Add a governance scare from the Chakrabarty resignation and an RBI card-issuance embargo, and the story felt like a compounder losing its edge.

Here is the number that tells a different story: Q1 FY27 standalone net profit was ₹19,060 crore, up 5 percent year-over-year, with net interest income growing 7 percent. Asset quality remains strong. The bank beat Nomura estimates.

Five percent earnings growth isn't exciting. But it matters because it came while the balance sheet was still normalizing post-merger. The merger integration is a drag, not a structural collapse. Management has initiated a leadership reset to address the overhang. The deposit rebuilding cycle — restoring the CASA ratio, which measures the share of low-cost current and savings deposits — will take quarters, not weeks. But the operating engine hasn't stopped turning.

A re-rating won't come until margins visibly bottom and CASA improves. But the stock is already priced for a company that's losing its franchise. The reality is a company that's integrating a massive merger while still growing profits. The gap between those two stories is what makes this a watchable setup rather than a dead-end value trap.

What Would Prove the Bear Right

I can be wrong again. The setup depends on things that haven't happened yet.

For IT, the bear case is real: if AI automation accelerates faster than new AI implementation spending, revenue per employee could decline structurally rather than temporarily. If global discretionary IT spending continues to soften through FY27 — especially in Europe, where the Middle East conflict has elongated decision cycles — the recovery bridge evaporates. The tripwire would be a consecutive quarter of negative constant-currency revenue growth at TCS or Infosys. That would signal the disruption thesis is materializing in earnings, not just in sentiment.

For HDFC Bank, the risk is that margin compression continues rather than stabilizing. If NIMs fall further below 3.3 percent while the credit-deposit ratio stays elevated, the post-merger hangover becomes a longer-term structural drag. The tripwire is a quarter where net interest income declines year-over-year while deposit growth stalls.

The Real Question

Bond yields are the excuse. The real question is whether cash-flow-rich businesses at multi-year lows deserve a rerating once the narrative selling exhausts itself.

The IT answer is cleaner. Dollar-earning, rupee-cost cash cows trading at a 36 percent discount to their 10-year P/E average, generating nearly 7 percent in free cash flow yield, with a credible — if not guaranteed — path to AI implementation spending offsetting headwinds within 12 months. The market bar is already low. The expectations have already reset. That is the setup.

HDFC Bank is a slower bridge. The merger friction is real, the margin story is unfinished, and the deposit cycle will take time. But the stock has been punished as if the franchise is broken, when the evidence points to a normalization period with intact asset quality and continuing profit growth.

Neither name is a quick trade. Both require patience. But the cash flow is the hard proof that separates a beaten-down quality business from a value trap. The market has done the work of resetting expectations. The question now is whether the numbers can deliver the surprise.