This publication has already covered three faces of a fragmenting world — nations hedging between blocs, central banks fleeing a shared currency, companies rebuilding the supply buffers they spent decades eliminating. None of those letters put a number on what fragmentation actually costs. The IMF has. This letter reads that number, and the more interesting one hiding just underneath it.
Since the global financial crisis, the ratio of world goods trade to global GDP has held remarkably steady — fluctuating between 41% and 48%, according to the IMF's own tracking. Read that number alone and globalization looks essentially intact, fragmentation talk overstated. Read what's happening underneath it, and a different picture appears: the same total volume of trade, increasingly rerouted along geopolitical lines rather than pure economic efficiency. This letter is about the gap between those two readings, and what the IMF's own modelling says it actually costs.
IMF First Deputy Managing Director Gita Gopinath put the paradox directly in a 2024 address that remains the clearest framing available: private-sector concern about fragmentation — measured by mentions in corporate earnings calls — has surged, even as there are not yet clear signs of deglobalization at the aggregate level. The trade-to-GDP ratio hasn't moved. What has moved, underneath that stable aggregate, is composition: trade and investment flows are being redirected along geopolitical lines, with China's share of specific US import categories falling even as China's overall global trade volume holds up by finding other buyers. Banque de France has built a dedicated analytical tool specifically because policymakers need to see through the stable headline number to the redirection happening beneath it — a genuine, purpose-built instrument for measuring a phenomenon the aggregate statistics are structurally bad at revealing.
The IMF's own Measure of Aggregate Trade Restrictions (MATR) dataset is the closest thing to a hard fragmentation ledger that exists: 145 countries, 99% of global GDP, 24 aggregated sectors and 136 disaggregated commodities, tracking tariffs, non-tariff barriers, and foreign-exchange restrictions back seventy years. What it shows is a genuine trend reversal — trade barriers fell steadily through most of the twentieth century, and that decline has reversed over the past decade specifically, concentrated in commodities and semiconductors flagged as national-security-relevant.
IMF staff modelling, published as a formal Staff Discussion Note, puts real numbers on the cost of this redirection: global output losses ranging from 0.2% of GDP in a limited-fragmentation, low-adjustment-cost scenario up to 7% of GDP in a severe-fragmentation, high-cost scenario — and if technological decoupling is added to trade fragmentation specifically, the loss in some countries could reach 8-12% of GDP. The distributional finding is the part worth sitting with directly: emerging-market and low-income countries are identified as most at risk, specifically because they lose access to the knowledge spillovers that historically came bundled with open trade relationships — a cost that doesn't show up in a simple tariff-revenue calculation.
Separate IMF modelling (Okuda & Tsuruga) quantifies a second, less-discussed cost: the loss of international diversification benefits for large advanced economies if they become unable to trade financial assets with geopolitically distant countries — a genuine portfolio-construction cost sitting on top of the trade-flow cost, largely invisible in headline GDP figures because it shows up as reduced risk-adjusted returns rather than a line-item loss.
The headline trade-to-GDP ratio has barely moved since the financial crisis. That stability is exactly what makes the composition shift underneath it easy to miss — and expensive to ignore.
Middle Powers documented the diplomatic hedging (sixteen nations refusing to pick a bloc). GeoFinance documented the reserve-currency response (68% of central banks adding gold). Resilience Economy documented the corporate response (dual sourcing, just-in-case inventory replacing just-in-time). All three are real, all three are already published, and none of them put a number on the aggregate cost of the phenomenon they're each describing a symptom of. This letter is the ledger entry underneath those three trajectories — not a fourth description of fragmentation, but the actual price tag IMF economists have attached to it.
The Data Layer Nobody Else Is Watching. Trade-restriction and geopolitical-distance data — the raw material behind the IMF's MATR dataset and Banque de France's tracking tool — is increasingly a genuine input into institutional risk models. Bloomberg (part of Bloomberg L.P., not separately listed) and similar data-and-analytics providers monetise exactly this kind of geopolitical-exposure scoring directly, a picks-and-shovels position on fragmentation-measurement itself becoming institutional infrastructure.
Diversification as the Direct Hedge. Given the IMF's own finding that fragmentation specifically erodes cross-border diversification benefits for large advanced economies, broad multi-region equity exposure (the same instrument this publication named in Letter 152 — a diversified emerging-market position) is a direct, literal hedge against the exact portfolio cost this letter documents, not merely a generic "spread your bets" recommendation.
Global trade's headline stability (41-48% of GDP since the financial crisis) is real and also misleading on its own — it obscures a genuine, accelerating redirection of that trade along geopolitical lines, a redirection the IMF's own MATR dataset confirms has reversed a seventy-year trend of falling trade barriers. The cost is not hypothetical: IMF staff modelling prices it at 0.2% to 7% of global GDP depending on severity, rising to 8-12% in specific countries once technological decoupling is included, with emerging markets and low-income countries bearing the disproportionate share through lost knowledge spillovers. This publication's own Middle Powers, GeoFinance, and Resilience Economy trajectories each describe a symptom of this same phenomenon; this letter is the number underneath all three. The risk to this thesis: IMF modelling of this kind carries genuine, acknowledged uncertainty — Gopinath's own framing is that fragmentation hasn't yet shown up at the aggregate level, meaning the more severe scenarios in this letter remain projections, not observed outcomes.
Founder, NextGen Economics · Bangalore, India · July 2026
Sources: IMF Staff Discussion Note, "Geoeconomic Fragmentation and the Future of Multilateralism" (Jan 2023, updated context 2026) · IMF Working Paper, "Geoeconomic Fragmentation and International Diversification Benefits" (Okuda & Tsuruga) · IMF, "The Costs of Geoeconomic Fragmentation" (Finance & Development, Bolhuis, Chen & Kett) · IMF (Gita Gopinath), "Geopolitics and Its Impact on Global Trade and the Dollar" (2024 address) · Banque de France, "Trade War and Geoeconomic Fragmentation" · ScienceDirect, "Geopolitical Fragmentation and Trade" (MATR dataset). Builds on this publication's own Middle Powers, GeoFinance, and Resilience Economy trajectories.
Not investment advice. This letter evaluates a geoeconomic policy trend; it does not constitute a recommendation regarding any security.