Supply Chain Resilience, Geopolitical Risk Pricing, and the Economics of Friend-Shoring: A Quantitative Analysis of the Pax Silica Coalition Framework

The concentration of the world's most advanced semiconductor fabrication in a single economy, and of critical-mineral refining in another, has turned a triumph of comparative advantage into a strategic liability. When roughly 90% of the most advanced logic chips are made in one geography, and the refining of the rare earths and battery minerals that feed advanced manufacturing is dominated by a single state, the efficiency logic that built these supply chains becomes indistinguishable from concentrated risk. In response, policymakers have embraced "friend-shoring"—the deliberate reorganization of production toward trusted partners—yet the policy has run ahead of the evidence: advocates have not costed it, macroeconomic simulations remain too abstract to guide investment, and industry reports lack formal frameworks. This dissertation addresses that gap by asking how much resilience costs, who should pay for it, and how it should be priced against geopolitical risk. The empirical setting is the Pax Silica coalition, a 23-nation plurilateral framework established in December 2025 to secure artificial-intelligence, semiconductor, and critical-mineral supply chains, whose flagship is a 4,000-acre Economic Security Zone at New Clark City in the Philippines. The study integrates five quantitative methods—network analysis of trade topology, structural gravity estimation by Poisson pseudo-maximum likelihood, cooperative game-theoretic allocation via Shapley values, real options valuation with Monte Carlo simulation, and comparative policy analysis—applied entirely to public data (including UN Comtrade, the USGS, the World Bank, OECD Trade in Value Added, FRED, the USITC, SIA/BCG industry estimates, the World Economic Forum, SIPRI, and the Caldara-Iacoviello geopolitical risk index). This all-public-data design makes the analytical pipeline transparent and replicable, demonstrating that rigorous resilience assessment need not rely on proprietary or classified information. Four findings anchor the results. First, the gravity model recovers a coalition trade coefficient of −0.972 (p < .001), confirming a large, statistically significant friend-shoring cost premium. Second, re-weighting semiconductor trade toward coalition partners delivers a 43.3% average reduction in sourcing concentration measured by the Herfindahl-Hirschman Index. Third, real options analysis of New Clark City yields an option value of US$3.72 billion—37.2% of the investment—which lifts the probability of a value-creating outcome from near zero to roughly one in three and converts a deeply negative static valuation into a near-break-even proposition. Fourth, the Shapley allocation achieves the lowest core-violation rate among competing burden-sharing rules at 13.5%, and plurilateral coordination attains a 39-to-43-times efficiency advantage in diversification per dollar relative to unilateral industrial policy. All four hypotheses are supported. The contribution is both substantive and methodological. Substantively, the study supplies the missing price tags, allocation rules, and risk-adjusted valuations that friend-shoring policy has lacked, offering direct guidance to the United States Treasury and allied finance ministries. Methodologically, it shows how gravity models, cooperative games, and real options can be integrated within an artificial-intelligence-economics framework to price strategic uncertainty using only open data—reframing supply-chain resilience not as a cost to be minimized but as an option to be valued and a coalition good to be shared.