Ruchir Sharma on Why India Seems to Be on the Wrong Side of the AI Trade

India is being viewed as lagging the global AI trade as investors favor chips, compute and infrastructure, with low R&D spending and exposure to software and outsourcing jobs pressuring sentiment. Ruchir Sharma said India still has potential upside from strong nominal GDP growth, a valuation reset, and later-stage AI productivity gains. The piece is primarily commentary on positioning and medium-term relative performance rather than a near-term market catalyst.

Analysis

The market is treating AI as a capex cycle, and India is on the wrong side of that trade because its equity beta is still dominated by labor-arbitrage rather than compute monetization. That creates a real second-order vulnerability: even if domestic demand holds up, multiples can stay compressed as global allocators rotate toward beneficiaries of the hardware buildout and away from economies exposed to AI-driven substitution in services. The near-term loser set is broader than the obvious outsourcing names; enterprise software, IT services vendors, staffing, and firms with high offshored white-collar exposure all face a sentiment discount before actual earnings hit.

The underappreciated bull case is that India does not need to win the AI infrastructure race to participate in the productivity phase. If AI lowers service delivery costs globally, the benefit can flow to Indian firms that are already scaled, have large domestic distribution, and can translate cost savings into margin expansion faster than peers. That argues for a time horizon split: 0-6 months still favors de-rating risk, but 12-24 months could favor businesses with domestic pricing power and balance-sheet capacity to absorb a slower labor-growth regime.

The catalyst path is likely macro and narrative-driven rather than company-specific: a sharp improvement in real growth, a visible pickup in capex, or evidence that AI is being adopted to improve operating leverage without destroying demand. Conversely, another quarter of weak R&D intensity or visible margin pressure in global IT spending would reinforce the “structural laggard” label and keep foreign flows defensive. The consensus may be over-discounting the worst-case substitution story, but it is underpricing how long valuation penalties can persist when a market is absent from the highest-conviction global theme.

Key Decisions for Investors

• Maintain underweight India ex-domestic defensives for the next 3-6 months; fade rallies in India-heavy EM baskets if AI hardware leadership remains intact, since relative multiple compression can persist even with decent GDP prints.
• Pair trade: long infrastructure/AI beneficiaries in developed markets (e.g., SMH or SOXX) vs. short India software proxies over 1-2 quarters; use this as a sentiment trade until evidence emerges that Indian firms are converting AI into margin expansion.
• Selective long on high-quality Indian consumer/financial compounders with low offshore labor exposure over 12-24 months; these names should capture nominal GDP growth without the direct substitution risk facing IT services.
• Avoid initiating fresh longs in India IT services on valuation alone; only step in after a 15-20% drawdown or if management guides to AI-led productivity gains that can offset pricing pressure within 2-3 quarters.
• For higher-conviction contrarians, consider call spreads on India benchmark ETFs 12-18 months out, financed by short-dated puts, to express the view that the current AI underexposure discount is temporary but unlikely to reverse quickly.