India's data-center suppliers added $47B in market value while the Nifty 500 lost $300B. Track backlog growth and foreign ownership to gauge the AI capex trade's durability.
India lacks homegrown AI giants and semiconductor leaders. Yet a cluster of industrial suppliers to data centers has added about $47 billion in combined market value this year. The benchmark NSE Nifty 500 has lost over $300 billion in the same period.
The mechanism is not about AI software or chip design. It is about the physical infrastructure that makes AI queries possible. Every GPT prompt, every cloud inference run, and every model training cycle passes through a power-hungry data center. Those centers need transformers, switchgear, fiber optic cable, precision cooling, and backup power. Indian industrial firms that manufacture these components are booking multi-year backlogs while the headline indexes bleed.
In Mumbai dealing rooms, this is called the "AI capex trade."
A data center consumes 10 to 50 times the electricity of a comparable commercial building. That power must be delivered, conditioned, cooled, and backed up. The components are not off-the-shelf; they require custom engineering, long lead times, and factory certification.
Nomura Holdings Inc. analysts led by Akash Gupta wrote in a June 2 report that two-to-four year lead times for some components have "created an enviable seller's market with multi-year backlogs." Orders secured now will generate revenue between 2027 and 2029.
The practical implication for a trader: the revenue visibility for these suppliers extends beyond the current capex cycle. The backlog itself becomes a valuation anchor, not a speculative guess.
Total investments in global hyperscale data centers are likely to exceed $1.2 trillion between 2025 and 2027, according to Angel One. This is not a marginal increase. It is a step-change in industrial demand that dwarfs the wireless 4G rollout, the post-2008 LNG build-out, and the early-2010s shale boom.
Sterlite Technologies Ltd., the optical-fiber maker owned by the Vedanta Group, has surged more than 530% this year. The catalyst was a $1.1 billion multi-year contract from a US-based hyperscaler last month.
The stock now trades at about 70 times its 12-month forward earnings. That is roughly 3.7 times the NSE 500's forward multiple of 19 times. The valuation demands execution perfection.
HFCL Ltd., which competes with Sterlite in fiber and telecom infrastructure, has jumped 191% year-to-date.
MTAR Technologies Ltd. makes precision cooling and power components for data centers. Its shares have more than trebled in 2026.
Both companies remain excluded from most broad domestic indexes. That means passive flows, which dominate institutional equity allocation, are not buying these names. The rally is entirely active money and retail conviction.
Anant Raj Ltd. is the only listed pure-play data center operator in India. Its stock has gained just 8% this year.
The divergence between Anant Raj and the equipment makers reveals a market preference for revenue visibility over thematic exposure. Anant Raj's earnings depend on leasing utilization rates, which are lumpy and opaque. Sterlite and MTAR book contracts with specific dollar values and delivery schedules. The market is pricing the clearer cash flow path.
Shareholding of foreign funds in Indian industrials rose to 14% as of end-March, the highest in two years, according to Elara Capital (India) Pvt.
This is striking because global funds remain record sellers of Indian equities overall. The allocation shift is sector-specific, not market-wide. Foreign investors are rotating out of consumer, financial, and tech services stocks into the industrial supply chain.
Practical rule: Track the foreign ownership percentage in these industrials each quarter. A plateau or decline would signal that the marginal buyer has finished accumulating, even if the stock price continues rising on domestic momentum.
Amazon.com Inc. plans to invest $12.7 billion in cloud infrastructure in India through 2030. Alphabet Inc. is spending about $15 billion on an AI infrastructure hub in Visakhapatnam.
A Reliance Industries Ltd. joint venture signed an $11 billion pact to build local data centers last year. AdaniConnex Pvt. has partnerships with Google and Uber Technologies Inc. to construct their facilities.
These commitments are not speculative. They are capital allocation decisions by the world's largest companies, backed by balance sheets that can absorb cost overruns.
Angel One noted that the market is now rewarding companies with visible AI-linked earnings rather than just thematic exposure. The brokerage said the biggest near-term risk is valuation, as share rallies have left "no room for execution disappointments."
This is a concrete trading rule. When the stock is pricing in perfection, any miss – a delayed shipment, a margin compression, a customer renegotiation – becomes a 20-30% drawdown event.
The data center supply chain includes an equal-weighted Bloomberg index of 28 Indian companies covering transformers, switchgear, wires, cables, cooling systems, and power components. That index has added $47 billion in market value this year, a rise of nearly 50%.
Hitachi Energy India Ltd., ABB India Ltd., and Cummins India Ltd. are the large-cap beneficiaries. These companies have diversified industrial businesses, the data center vertical is becoming an outsized growth driver.
Finolex Cables Ltd., which makes electrical cables used in data center construction, has surged nearly 36% this year. CEO Mahesh Viswanathan said in an earnings call last month that this was "the right time to be in this industry."
Both AMZN and GOOGL have direct India commitments that feed this supply chain. Amazon's $12.7 billion cloud infrastructure plan and Alphabet's $15 billion Visakhapatnam hub are the demand-side anchors that give the equipment makers their multi-year visibility.
For a trader tracking the GOOGL stock page or the AMZN stock page, the India data center capex cycle is a downstream demand signal that is not yet priced into those stocks' core AI infrastructure narratives.
Nomura analysts wrote that data center capex "has emerged as the single largest contemporary industrial investment cycle" – larger than the 4G rollout, the post-2008 LNG build-out, or the early-2010s shale boom.
This comparison matters because each of those prior cycles created multi-year compounders among industrial suppliers. The shale boom drove decades of returns for pressure pumping and pipeline companies. The LNG build-out created a generation of engineering and construction winners. Data center capex, by Nomura's framing, is that same type of structural demand shift.
R. Sivakumar, chief investment officer at Axis Mutual Fund, captured the distinction: "We may be on the wrong end of the AI trade, we could be on the right side of the AI capex trade."
The "wrong end" refers to India's lack of AI software and semiconductor stocks. The "right side" is the physical infrastructure that must be built regardless of which company wins the AI model race.
These stocks have already repriced. The easy money – the discovery of the theme and the initial re-rating – is done. What remains is the execution phase, where quarterly results will either validate or challenge the multiples.
The market cap added ($47 billion) is large enough to be visible, the individual stocks remain too small for index-heavy institutional mandates. That creates a liquidity risk: when the theme cools, selling pressure will concentrate on the most liquid names (ABB India, Cummins India, Hitachi Energy) while the smaller names (Sterlite, MTAR, HFCL) may gap down on thin volume.
The next concrete catalyst is the Q1 FY27 earnings season, when order books from the 28 index constituents become visible. The single most important number will not be revenue growth. It will be backlog growth. A sequential increase in backlog confirms the multi-year seller's market. A flat or declining backlog, even with strong revenue, would signal that new orders are slowing.
Prepared with AlphaScala research tooling and grounded in primary market data: live prices, fundamentals, SEC filings, hedge-fund holdings, and insider activity. Each story is checked against AlphaScala publishing rules before release. Educational coverage, not personalized advice.