Charts show how much Meta, Google, and Amazon slashed H-1B petitions after Trump's visa crackdown

Amazon's certified H-1B filings fell from 4,647 to 3,057 in fiscal Q1 2026 (down ~34%), and Google and Meta saw roughly 50% declines year-over-year; Nvidia was the outlier, up ~17.6% (369 to 434). The drop appears driven by the Trump administration's policy changes (lottery skewed to highest-paid applicants and a $100,000 fee on new offshore petitions) plus ongoing Big Tech layoffs and hiring freezes, which raise hiring costs and scrutiny. Implication: tighter, costlier access to foreign tech talent could slow staffing for AI initiatives and weigh on growth outlooks in affected firms; monitor Q2 lottery filings and final visa approvals for confirmation.

Analysis

Firms are adjusting their labor strategies in a world where access to offshore technical talent is more expensive and slower — expect a sustained re-allocation of costs from recruiting to retention and tooling. That re-allocation will show up as higher per-head comp for specialist AI engineers, larger sign-on/retention equity pools, and faster deployment of internal retraining programs; pressure on gross margins will be concentrated in product groups that are still labor-heavy versus those driven by amortizable software or silicon.

Second-order winners are companies that already house large onshore engineering benches or own the IP stack that reduces the need for rapid external hiring: think hardware/software integrators and cloud vendors with sticky enterprise contracts. Conversely, businesses that historically relied on rapid, lower-cost global hiring or on rapidly scaling distributed R&D teams will see hiring velocity and feature cadence slow, creating acquisition windows and increasing M&A premiums for talent-rich startups.

Key catalysts that will change the picture are corporate commentary on talent pipelines during earnings (near-term) and any judicial or legislative moves that materially expand or constrain cross-border hiring (3–18 months). Monitor corporate line items for contractor spend, recruiting costs as a percentage of payroll, and product delivery slippage; a spike in contractor spend or a one-time ‘acquisition for talent’ will be the earliest signal that firms are substituting M&A for hiring.

Key Decisions for Investors

• Long NVDA 6-month call spread (buy near-ATM call / sell ~10–20% OTM call). Rationale: capital allocation priorities favor engineering-heavy AI winners; limited premium at entry with asymmetric upside if AI projects accelerate. Risk: supply/valuation shock; cap premium loss limited to paid premium.
• Pair trade — long ORCL (or MSFT) vs short AMZN, 3–9 month horizon. Rationale: ORCL/MSFT benefit from enterprise onshore demand and sticky software economics; AMZN carries higher operational leverage to headcount cycles. Position sizing: 1:1 dollar exposure; stop-loss at 8–10% adverse move; target 20–40% relative outperformance.
• Protective hedge for big-cap ad/consumer tech exposure (META/GOOGL): buy 6–9 month 3–5% OTM puts sized to cover 20–30% of equity exposure. Rationale: limits downside from slower product velocity or contract slippage while keeping upside intact. Cost is insurance premium; treat as tail-risk buy.
• Tactical short/underweight small-cap or talent-dependent AI startups into any rally (12 months). Rationale: elevated M&A premiums and contracting hiring channels favor incumbents buying teams rather than allowing organic scale; these names have the highest binary execution risk. Use options to keep risk finite (put spreads).