Geoeconomics is the new geopolitics: Playing offense in the new economy

The piece warns that a new economic order is being shaped by geoeconomic fragmentation and rapid technological innovation, with governments reasserting economic roles via industrial policy, tariffs and strategic investments. Key data points: global goods trade growth ~2.4% and services exports ~4.6%, digital trade expanding ~12% annually, AI-related semiconductor goods drove ~43% of merchandise trade growth in H1 2025, AI sector capex reached at least $400 billion in 2025 and is projected to exceed $750 billion by 2029, and AI could add up to $15 trillion to global GDP by 2030. Implications for allocators include prioritizing exposure to AI infrastructure, semiconductors, energy and critical minerals, while factoring geopolitical risk, potential increased regulatory oversight of tech firms, and the need for supply‑chain diversification.

Analysis

Market structure: The emerging AI Super System concentrates durable economic value in compute, advanced semiconductors, data-center real estate, grid-capex and critical-minerals suppliers. Direct beneficiaries: NVDA, TSM, ASML, AMZN, GOOGL, EQIX/DLR, ALB/MP and utilities/transformer manufacturers (e.g., NEE, ABB exposure) as demand for compute and power outpaces near-term supply; losers include legacy CPU vendors (INTC) and export-dependent OEMs facing tariff/friction risk. Expect pricing power uplift in advanced-node semiconductors and lithium/copper markets with visible tightness through 2026–2029 (AI capex projected $750bn by 2029), pushing commodity and industrial input inflation and upward pressure on yields.

Risk assessment: Tail risks include abrupt export controls or asset nationalization (China/US) that can remove key suppliers overnight, energy-grid failures that cap compute growth regionally, or a rapid AI regulatory clampdown that curbs monetization; probability medium, impact high. Immediate (days) risk is policy announcements; short-term (weeks–months) is capex allocation and supply-chain shifts; long-term (3–5 years) is structural re-shoring and minerals geopolitics. Hidden dependencies: grid capacity, financing for long‑lived infrastructure, and rare-earth/refining concentration in a few jurisdictions; catalysts are CHIPS-style subsidies, large model launches, or a major outage in a hyperscaler region.

Trade implications: Favor concentrated exposure to leaders of compute and supply (1–2% positions in NVDA, TSM, ASML) and infrastructure plays (1% EQIX/DLR, 1% NEE) with 12–36 month horizons; hedge country/regulatory risk via diversified miners ETF (LIT) or 1% exposure to MP/ALB. Use relative-value: long TSM (+1.5%) / short INTC (-1.5%) to express node leadership; buy longer-dated call spreads on NVDA or SOXX (6–12 months) to capture upside while capping premium. Rotate out of small-cap export-oriented industrials and certain global logistics names into semis, energy infra and critical-minerals names over the next 4–12 weeks as policy clarity emerges.

Contrarian angles: Consensus underweights grid, transformer, and EPC suppliers — these can be 2nd-order winners if AI capex forces local generation and storage builds; consider select small-cap power-equipment names before they re-rate. The market may be overpricing hyperscalers’ ability to self-provide at scale; data‑center REITs could see asymmetric returns if hyperscalers outsource to manage regulatory/geopolitical risk. Historical parallel: semiconductor strategic races (1970s–80s) show governments can rapidly change industry economics via subsidies — expect episodic re-rating, not linear appreciation.

Key Decisions for Investors

• Establish a 1.5% long position in NVIDIA (NVDA) with a 6–12 month horizon; if volatility is elevated, implement a defined-risk call spread (buy 6–9 month calls ~30–40% OTM, sell calls ~50–60% OTM) to capture expected AI-driven revenue re-rating while limiting premium decay.
• Initiate 1.0–1.5% long positions in data-center REITs Equinix (EQIX) and Digital Realty (DLR) (split exposure) for 12–24 months to capture wholesale lease growth from hyperscalers; add 0.5% more if occupancy/utilization rises >200 basis points or wholesale pricing increases >5% QoQ.
• Allocate 2.0% to critical-minerals exposure via 1% MP Materials (MP) + 1% Albemarle (ALB) or a 2% position in the LIT ETF, horizon 12–36 months; buy on dips >15% or accumulate if lithium/copper futures rise >25% signaling supply squeeze.
• Execute a pair trade: long 1.5% Taiwan Semiconductor (TSM) / short 1.5% Intel (INTC) for 12 months to express advanced-node share gains; unwind if TSM underperforms by >20% relative to INTC or if export-control headlines force cross-border delisting risk.
• Monitor (actively, 0–30 days) for export-control or subsidy announcements (US/China/EU); if restrictive export measures expand, hedge equity delta by buying 3–6 month put protection on affected semis (SOXX) equal to 1–2% portfolio notional and reprice within 48 hours of announcement.