The PERM Freeze Is Not a Crisis. It Is a Corrective.

For thirty years, a quiet bargain underwrote India's technology miracle. Indian engineers, educated at state expense, would cross the Pacific, obtain H-1B visas, wait years, sometimes decades, in the green card backlog, and staff the back offices of corporate America. The firms that sent them, Tata Consultancy Services, Infosys, and their peers, built revenue empires on this arbitrage. The arrangement suited everyone: American corporations got cheap, skilled labour; Indian IT giants got billable heads; and the Indian government could point to software export figures as proof of economic modernisation.

Washington has now disrupted that arrangement. The US suspension of the PERM labour certification programme, the gateway through which employers sponsor foreign workers for permanent residency, has frozen green card pathways for Indian IT professionals embedded across the American technology economy. The freeze has landed with particular weight on large Indian IT services firms whose onsite delivery model depends on the PERM pipeline as a retention mechanism, the implicit promise that years of faithful service in America would eventually yield a green card. That promise has now been suspended.

The industry's first instinct will be to treat this as a visa spat, to lobby through NASSCOM, to dispatch delegations, and to wait for the policy to reverse. That instinct is understandable. It is also the wrong response.

The Conveyor Belt and Its Costs

The green card conveyor belt was never just a human resources mechanism. It was the scaffolding of a business model. Indian IT firms grew to their current scale by placing engineers at client sites in the United States, billing by the hour or the project, and earning in dollars while paying salaries in rupees. The PERM programme was the instrument that made long-term onsite staffing viable, without the prospect of permanent residency, retaining engineers in America for multi-year client engagements would have been structurally impossible.

What this model did not build, in thirty years of operation, was intellectual property that India owned. It did not produce a SAP, an Oracle, or a Salesforce. India became the world's back office at precisely the moment it had the talent and the capital to become the world's product laboratory. The green card conveyor belt was comfortable enough that the harder, riskier work of building software products for global markets, work that iSPIRT's founders have been urging for over a decade, never acquired the urgency it deserved.

The PERM freeze has introduced exactly that urgency, at a moment when AI is dismantling the labour-arbitrage model anyway. The two forces together, an American immigration crackdown and an AI productivity wave that reduces headcount requirements, make the existing IT services model structurally precarious in ways that a policy reversal in Washington would not fix.

What AI Is Already Doing to the Argument

Strip away the immigration dimension and the underlying shift is still happening. AI coding assistants, automated testing frameworks, and agentic workflow tools are compressing the unit economics of software services. A task that once required ten engineers on a three-month engagement can now be scoped differently. American client firms are asking those questions already. Infosys's strategic collaboration with OpenAI and TCS's pivot toward AI-led services signal that the large Indian firms understand, at the level of partnership announcements, that the old model is under stress. The question is whether the organisational transformation matches the branding.

There is a version of the AI transition that Indian IT navigates on American terms: retrain engineers to work with AI tools, keep them billable, keep the services model alive, just with fewer heads and more automation. That version preserves the dependency. There is another version, harder, slower, more capital-intensive, where Indian firms build AI-native products that sell globally, generating revenue from software rather than from labour hours. The PERM freeze makes the second path not merely advisable but, over any planning horizon of five years or more, necessary.

The Domestic Market India Has Not Yet Taken Seriously

India's own digital infrastructure has matured enough to anchor a genuine product economy. The stack built over the past decade, payments rails, identity infrastructure, data-sharing frameworks, gives Indian product companies a home market of scale that did not exist when the IT services model was being designed in the 1990s. A firm that builds enterprise software for Indian banking, logistics, or healthcare now addresses a market of genuine size, with the additional advantage that solutions built for Indian conditions, cost sensitivity, regulatory complexity, infrastructure variability, travel well to other emerging markets.

The NASSCOM ecosystem has articulated this case for years, that India's TECHADE aspiration requires a shift from IT services to software products, from body-shopping to brand-building. The PERM freeze is the first external shock large enough to give that argument political and boardroom salience.

What has been missing is not the argument but the incentive structure. Senior engineers at TCS or Infosys who were riding the green card queue had a rational reason to stay on the conveyor belt. Remove the destination, and the calculus changes. Some of those engineers will choose to return to India. Some will choose to stay in America regardless, building startups or joining American product firms. The ones who return carry capital, networks, and the experience of building at scale inside complex American enterprises. If India's policy environment can absorb that cohort through startup incentives, R&D tax treatment, and access to patient capital, the PERM freeze could seed a product-company wave that the services model never generated.

The Statecraft Dimension

None of this diminishes the legitimate diplomatic dimension. India's government should engage Washington on the PERM suspension, both as a matter of principle, these are workers whose careers have been built in good faith around a legal pathway, and as a matter of the bilateral relationship's texture. The India-US technology partnership, which has deepened across multiple ministerial tracks, depends on reciprocal confidence that legal immigration pathways are stable. A freeze that disrupts that confidence is a legitimate bilateral concern, not merely an industry grievance.

But statecraft and strategy are not the same thing. India can press Washington on PERM while simultaneously deciding, at the level of industrial policy, that the services export model needs a different architecture for the decade ahead. The two positions are not contradictory; they are sequential. Represent the immediate interests of Indian professionals and firms caught in the freeze; use the freeze as the occasion for a structural rethink that should have happened five years ago.

S. Jaishankar has written that India's destiny is too large to be merely a part of the future of others. For three decades, India's most talented engineers and the firms that employed them arranged themselves as components inside an American technology stack. The PERM freeze has interrupted that arrangement. The question India's tech establishment must now answer, in boardrooms in Bengaluru, in policy rooms in Delhi, and in the startup corridors that have grown up between them, is whether to restore the old arrangement as soon as Washington allows it, or to use this interval to build something that belongs to Bharat. The first option is a return to comfort. The second is the one that 2047 requires.