The US-China bilateral risk is priced. What is not priced is the transmission: the forty-plus economies that sit directly in the crossfire, dependent on China for supply and on the U.S. for revenue, with no hedge available on either side. This piece maps these economies, quantifies its exposure, and explains why the diversification narrative that is supposed to resolve it is making the underlying condition worse.
Economies with no hedge
Across 143 countries, 40 send more than 30% of their exports to the U.S. and China combined. Twenty-two send more than 40%. They include G20 members, major commodity exporters, and the primary beneficiaries of post-2018 supply chain reconfiguration.
Most lean heavily toward one power. Mexico's 84% export concentration toward the U.S. and Mongolia's 92% toward China are severe dependencies, but they are directionally clear. A portfolio manager can model a unipolar shock. The genuinely no-hedge cases, namely, countries with more than 15% of exports going to each simultaneously, number seven: Vietnam, South Korea, Japan, Chile, Ecuador, Taiwan, and Venezuela. For these economies, there is no preferred partner to retreat toward. Deterioration on either axis hits revenue directly, and the standard risk management response to rebalance toward the less hostile relationship is not available.
Bilateral Trade Exposure — 2023
143 countries · Merchandise trade shares to China and the United States · Bubble size = merchandise trade as % of GDP
The composition of that seven deserves attention. Japan and South Korea are G20 economies, U.S. treaty allies, and deeply embedded in Chinese intermediate goods networks. While Japan and South Korea are strategically realigning toward the U.S., their production systems remain deeply embedded in China-centered supply chains. Both economies continue to rely on Chinese intermediate inputs, meaning that China remains indirectly present in exports to the U.S. in ways bilateral trade figures do not capture. Taiwan's position requires no elaboration. Vietnam is the asset that every supply chain diversification thesis is currently long.
The import side: the number that is not in the model
The export dependency story is visible and partially priced. The import story is not.
Export share measures revenue dependency. Import share measures something harder to replace: supply dependency. The machinery, components, and intermediate inputs that a country's manufacturing base runs on. A revenue shock can be absorbed over time. A supply disruption is immediate and operational.
The median country in this dataset now sources 16.2% of its imports from China, up from 12.4%in 2015, a 30% increase in eight years, rising every single year except the 2022 post-COVID statistical noise. The MERICS Trade Dependency Database shows a clear asymmetry at the product level: EU and U.S. reliance on imports from China has increased over the past two decades, particularly in machinery and electronics where China is a dominant supplier, while China has reduced its dependence on manufactured imports from these economies. The asymmetry is structural. China’s industrial policy intensified following initiatives such as Made in China 2025 and the onset of the U.S.-China trade war, reinforcing cost advantages across supply chains. While Chinese exports are often price-competitive in many manufacturing sectors, this reflects a combination of scale, industrial policy, and ecosystem effects. These structural advantages help sustain China’s central role in global trade, even as policy efforts seek to redirect supply chains.
Global Median Trade Shares, 2015–2023
Median across all 143 countries · Both charts on the same 0–20% Y-axis · 9 annual data points per series
Source: merchandise trade dependence ratios. Median — not mean — used throughout to suppress outlier distortion (e.g. Mongolia at 92% exports to China). All four series confirmed from raw data: exp_chn range 3.99–5.51%, exp_usa range 4.73–6.52%, imp_chn range 12.42–16.20%, imp_usa range 4.21–4.92%.
What the data shows: Import from China (dashed red) is the only series with a meaningful trend — rising from 12.4% to 16.2% over 8 years, up in every year except 2022. All three other series — exports to China, exports to US, imports from US — are essentially flat, oscillating within a 2pp band. The world's median bilateral export profile barely changed. Its import dependency on China deepened by 30%.
Meanwhile U.S. import share has been flat to slightly declining. The divergence between the two lines is the core signal. One power is tightening its grip as a supplier. The other is not. The geopolitical pressure runs in one direction. The structural pull runs in the other.
Twelve countries in the dataset have a China import share that exceeds their China export share by more than 20 percentage points, meaning they purchase far more from China than they sell to it, with no export revenue to deploy as countervailing leverage. Liberia (49% imports from China, 4% exports), Kyrgyzstan (45%, 3%), Cambodia (42%, 6%), Russia (38%, 12%), Ghana (35%, 9%). The power asymmetry in these relationships is at its maximum. These are not countries negotiating with China. They are countries absorbing whatever terms China sets.
Quantifying the no-hedge position: the squeeze score
Bilateral share measures exposure. However, it does not capture how the exposure transmits into national decision-making. A 30% bilateral concentration in an economy where trade is half of GDP is uncomfortable, but bearable. The same concentration where trade is 160% of GDP is a different category of risk: there is no domestic economy large enough to act as a buffer. The shock passes straight through to output.
The squeeze score operationalizes this. It takes a country's average bilateral exposure across both trade directions and both partners, then multiplies by merchandise trade as a share of GDP. The result is a single number that answers the question portfolio managers should be asking: if US-China pressure on third parties escalates, how much of this country's economic activity is directly in the firing line?
Vietnam scores 34, the third highest in the dataset, after Hong Kong and Mongolia. With merchandise trade at 164% of GDP, a sustained disruption in either key relationship does not register as a trade statistic but as a GDP shock. The IMF's fragmentation modelling illustrates why the trade openness multiplier matters: low-income and small open economies are most at risk from geoeconomic fragmentation precisely because of their heavy dependence on imports and exports of key products for which substitution is costly. The squeeze score is that risk, made country-specific and rankable.
Squeeze Score — Top 30
Squeeze = avg bilateral share (exports + imports to both China + US) × merchandise trade / GDP ÷ 100.
Score bar: relative squeeze magnitude · Stacked bar: China avg share (red) + US avg share (blue), total bilateral exposure · Bubble: merchandise trade as % of GDP, value shown inside
/GDP
The top of the distribution is dominated by small open economies with no domestic buffer: Vietnam (34), Cambodia (25), Singapore (22), Malaysia (20), Nicaragua (20). PIIE analysis shows that import concentration has increased across several Indo-Pacific economies since 2010, with China playing a central role as a dominant supplier. At the same time, many of these economies have expanded exports to the United States.
Squeeze Decomposition — Concentration × Trade Openness
X = average bilateral share (concentration toward US + China). Y = merchandise trade as % of GDP (amplifier). Background colour = squeeze score. Bubble colour: red = China-leaning, blue = US-leaning. Size = GDP.
However, there are caveats. A high squeeze score on a resource supplier with commodity leverage is a structurally different risk than the same score on an electronics assembler. The score surfaces the candidate. Sector composition and commodity substitutability determine whether the exposure is actually indefensible.
Mongolia will appear high in the distribution. Its 92% export share to China produces a large concentration figure, but Mongolia exports copper and coal, commodities for which China has limited short-run alternatives.
Chile presents the same issue: 40% of exports to China, predominantly copper, with genuine pricing power the share number does not reveal.
Another caveat is the apparent case, Hong Kong:
Hong Kong will appear near the top of any ranking that treats it as an independent economy: its trade volumes are massive relative to its size, and its merchandise trade as a share of GDP is among the highest in the world. But Hong Kong is part of China. Its exposure to the mainland is not crossfire risk in the way Vietnam's or South Korea's is.
What remains after both corrections is a distribution of genuinely exposed economies: concentrated, open, and positioned in the crossfire without the commodity leverage or structural insulation that would make the exposure manageable.
Why diversification is making this worse
The policy and investor response to crossfire exposure has been supply chain diversification, the China+1 framework, now a decade old and accelerating. The thesis is intuitive: move production out of China, reduce bilateral concentration, exit the no-hedge position. The data says the opposite is happening.
Average bilateral import exposure across all 143 countries rose from 23.3% in 2015 to 25.6% in 2023, every single year. CEPR's analysis of U.S. trade policy and supply chain restructuring is precise on this point: countries that exported more to the U.S. also increased or continued their linkages with China and in electronics, the industry contributing most to apparent decoupling, countries that increased U.S. exports also increased imports from China in the same sector, suggesting that displacing China on the export side often accompanied by embracing China-wide supply chains on the import side. Evidence from Rhodium Group reaches a similar conclusion from a different angle: diversification from China often involves Chinese firms, which are among the leading sources of manufacturing investment in Southeast Asia and remain key suppliers of intermediate inputs, blurring what constitutes a “Chinese” good in global supply chains.
That is what makes the structural assumption so consequential. The China+1 trade is being underwritten by a structural assumption that export diversification is reducing crossfire exposure. The import data falsifies that assumption. Countries gaining U.S. market share from China are not exiting the no-hedge position. They are deepening it on the supply side while appearing to reduce it on the revenue side. The Stanford FSI analysis, using global input-output data to trace both direct and indirect trade linkages, shows that declines in direct U.S.-China trade are often partially offset by rerouting through intermediary economies such as Vietnam and Mexico, underscoring how supply chains are reconfigured rather than unwound in response to policy pressures.
The number to carry forward
The median country now sources 16.2% of its imports from China, up 30% since 2015. That number rose through the trade war, through COVID, through the supply chain panic, through the geopolitical rupture. Every macro event that was supposed to trigger genuine diversification produced the opposite result on the import side.
That is the no-hedge position in one statistic. The export data tells a more comfortable story. The next instalment of this Geoeconomics series uses the divergence between the two as a test, and most of the celebrated diversification winners fail it.