The Great Trade Rewiring: China's Share of US Imports Fell From 21% to 7%, and the Global Supply Chain Will Never Look the Same

In 2017, China accounted for 21% of everything the United States imported — the culmination of two decades in which Chinese manufacturing reshaped the global economy. By mid-2026, that figure has fallen to roughly 7%. The decline is not gradual. It is not a rounding error. It is the largest reallocation of supply chain trade in the modern era, and it is happening in real time: in the first four months of 2026, China's merchandise exports to the US fell 10.2% year-on-year to $133.4 billion, while US imports from China dropped to levels not seen since China joined the World Trade Organization in 2001.

Economists Laura Alfaro and Davin Chor, in a study published through the National Bureau of Economic Research and presented at the IMF, have called this the “Great Reallocation.” The term is precise: what is happening is not deglobalization — the US is not importing less overall — but a fundamental rewiring of where those imports come from. The beneficiaries are identifiable: Vietnam, Mexico, Taiwan, and increasingly India. The losers are equally clear. And the implications for the global GDP rankings are only beginning to be understood.

The Three Phases of Decoupling

The decline did not happen in a single event. It unfolded in three distinct phases, each driven by a different policy mechanism.

Phase 1 (2018–2020): The tariff opening.Section 301 tariffs imposed during the first Trump administration targeted an initial $50 billion of Chinese goods at 25%, eventually expanding to cover approximately $370 billion. China's import share dropped from 21% to roughly 18%. The decline was real but moderate — many businesses absorbed the tariffs or rerouted through third countries rather than fundamentally restructuring supply chains.

Phase 2 (2021–2024): Technology controls and strategic decoupling.The Biden administration maintained Trump-era tariffs and added semiconductor export controls that went beyond trade policy into industrial strategy. The CHIPS Act subsidized domestic semiconductor production. The restrictions on advanced AI chips (A100, H100) and EUV lithography equipment to China signaled that the decoupling was not merely about trade balances but about the architecture of technological competition. China's share fell from 18% to 13%.

Phase 3 (2025–present): Liberation Day and the acceleration. The “Liberation Day” tariff announcements in April 2025 imposed cumulative tariff rates exceeding 100% on many categories of Chinese goods. The share of US imports from China dropped four percentage points in a matter of months — from 13% to 9% by July 2025 — and has continued declining to approximately 7% by mid-2026. This phase reversed two full decades of trade integration in under 18 months.

The Winners: Vietnam, Mexico, Taiwan, India

The reallocation has not dispersed evenly across the global economy. The NBER study found that US imports diversified mainly among its top-20 existing trade partners, not to new source countries. The four clearest beneficiaries each represent a different type of supply chain substitution.

Mexico is the nearshoring story. Bilateral trade with the US grew 6.8% in the first two months of 2026 to $147.3 billion, even as overall US trade with the rest of the world contracted 4.5%. Mexico's USMCA utilization rate — the share of trade conducted under the preferential North American trade agreement — jumped from 44.8% to 88.7% in 2025, as companies restructured supply chains to qualify for tariff-free treatment. Mexico has overtaken China as the US's largest goods trading partner, a reversal that would have seemed improbable a decade ago. Mexico's economy in 2026 is increasingly defined by this shift.

Vietnam is the friendshoring exemplar. Its share of US imports rose from 2% in 2017 to approximately 4–5% by 2025 — a doubling that reflects Vietnam's role as a manufacturing platform for electronics, textiles, and furniture previously sourced from China. Samsung, Intel, and Apple have all expanded Vietnamese operations. Vietnam's GDP growth of 6.5–7% in recent years has been substantially powered by this trade reallocation.

Taiwan is the strategic technology story. US imports from Taiwan surged nearly 60% in a single year, driven almost entirely by semiconductors. TSMC's Arizona fabs are under construction, but in the interim, the US has deepened — not reduced — its dependence on Taiwanese chip production. South Korea's semiconductor exports have benefited similarly, though to a lesser degree.

India is the emerging-market dark horse. Taiwanese firms are increasingly relocating manufacturing to India, with FDI surging to over $665 million between 2018 and 2023. India's Union Budget 2026 introduced comprehensive semiconductor tax exemptions to capture a larger share of the technology supply chain. While India's $4.19 trillion economy is already large, its share of US imports remains small — under 3% — suggesting significant room for growth as the reallocation continues.

The Connector Economy Problem

The headline — China's share of US imports fell from 21% to 7% — tells a cleaner story than the reality warrants. The Alfaro-Chor research reveals that much of the reallocation has not severed supply chain connections to China but rerouted them. Vietnam imports significantly more intermediate goods from China than it did in 2017, assembles or lightly processes them, and then exports finished products to the United States. Mexico's manufacturing sector has deepened its Chinese component sourcing even as bilateral US-China trade fell.

These “connector economies” — countries that serve as intermediaries between China's manufacturing base and the US market — are the defining feature of the new trade landscape. The Stanford SCCEI characterizes this as “selective decoupling from China rather than a full US retreat from globalization.” The pattern is visible in the data: trade between hypothetical geopolitical blocs has grown 4% slower than trade within blocs since the Ukraine war, according to the WTO. But the total volume of global trade continues to grow.

The implication is that the US has reduced its direct exposure to Chinese manufacturing but not its systemicdependence on it. A disruption in Chinese intermediate goods production — whether from a Taiwan Strait crisis, a domestic economic downturn, or further policy restrictions — would cascade through Vietnam, Mexico, and other connector economies into US supply chains, potentially with longer lead times and less visibility than the old direct-from-China model.

The Cost of Rewiring

Trade reallocation is not free. The US Joint Economic Committee reported a total US trade balance of −$55.9 billion for April 2026, reflecting the continued high cost of imports from alternative sources. Multiple studies have documented that prices for goods previously sourced from China are higher when sourced from Vietnam, Mexico, or domestic production — a finding that should surprise no one familiar with why supply chains concentrated in China in the first place.

The tariff regime itself imposes direct costs on US businesses and consumers. Cumulative tariff rates on Chinese goods now exceed 100% for many product categories. The economic rationale is that short-term consumer costs are worth the long-term strategic benefit of supply chain diversification — but the distributional impact is uneven. Industries that can relocate manufacturing (electronics, textiles) have done so. Industries that cannot easily substitute Chinese inputs (rare earths, certain pharmaceutical precursors, specialized chemicals) face permanent cost increases.

The impact on China has been substantial but not catastrophic. China's total exports grew 2.4% in 2025 despite the US decline, partly because Chinese manufacturers redirected output to Southeast Asia, the Middle East, and Latin America. The US share of China's exports has fallen even as total Chinese exports remained stable. China's trade surplus with the rest of the world is near record highs. The decoupling, in other words, is bilateral, not global.

The Geopolitical Dimension: Trade Follows Alliances

The WTO data is unambiguous: trade is fragmenting along geopolitical lines. Trade in goods between hypothetical “Eastern” and “Western” blocs has grown 4% more slowly than intra-bloc trade since the start of the Ukraine war. This is not yet a full bifurcation of the global trading system, but the pattern is accelerating rather than stabilizing.

The ITIF's 2026 study on “The Global Trade Battleground” documents US-China competition for trade influence in the Global South — the countries that have so far benefited from the reallocation. Vietnam, Mexico, Indonesia, and India are being courted by both sides. The countries that position themselves as reliable “connector economies” — open to investment from both blocs — stand to gain the most. Those that are forced to choose may find their growth constrained.

The BRICS expansion is partly a response to this dynamic. By building alternative trade settlement mechanisms and financial institutions, BRICS members are constructing the infrastructure for a parallel trading system that is less dollar-dependent and less susceptible to US tariff policy. Whether that system achieves meaningful scale depends on whether member states can overcome the coordination problems that have historically limited South-South trade integration.

What Comes Next: Three Scenarios

Managed decoupling (base case).US-China direct trade stabilizes at 5–8% of US imports. Connector economies absorb the difference. Global trade volumes continue growing at 2–3% annually, but the geography shifts. Costs are higher than the pre-tariff baseline but manageable. This is broadly the current trajectory.

Escalation.US tariffs extend to connector economies (particularly Vietnam, which already faces scrutiny for Chinese transshipment). Trade fragmentation deepens. Global trade growth slows to 1–2% or less. The WTO's March 2026 forecast of 1.9% global trade growth already reflects this risk. Emerging economies face the worst of both worlds: disrupted supply chains and reduced foreign demand.

Strategic reconciliation.A US-China trade reset reduces tariffs on non-strategic goods while maintaining restrictions on semiconductors, AI, and dual-use technology. This would halt the reallocation in consumer goods while deepening it in strategic sectors. The probability is low in the current political environment, but the economic pressure — from inflation, from consumer costs, from the inefficiency of rerouted supply chains — builds with every quarter of maintained tariffs.

The Implications for GDP Rankings

The Great Reallocation is already reshaping global GDP rankings. Vietnam's sustained 6.5–7% growth is partly a function of absorbed supply chains. Mexico's trade surplus with the US is at record levels. Taiwan's semiconductor exports have pushed its economy toward the $1 trillion mark. India's manufacturing sector, while still small relative to its economy, is growing faster than at any point in the past decade.

China's GDP growth, meanwhile, has slowed to 4.5–5.0% — still high by advanced-economy standards, but the lowest sustained rate in decades. The trade reallocation is not the primary cause (the property crisis and demographic decline are larger factors), but the loss of US market share adds a structural headwind that was absent as recently as 2017. The gap between China's GDP and the US's, which had been narrowing since 2001, has stabilized or begun to widen depending on the metric.

The most consequential shift may be the one that is least visible in the aggregate data: the emergence of a new tier of “connector economies” whose growth models are built not on endogenous innovation or resource extraction, but on their position in the geography of great-power competition. These economies are growing faster than they otherwise would, but their growth is contingent on a geopolitical configuration that could shift again. The supply chains that moved from China to Vietnam in three years could, in theory, move from Vietnam to somewhere else in three more. The Great Reallocation is not a destination. It is a process — and one that, based on the data, is still accelerating.

Sources: Laura Alfaro & Davin Chor, “An Anatomy of the Great Reallocation in US Supply Chain Trade,” NBER Working Paper 34490 (2026); CEPR VoxEU update (2026); Stanford SCCEI, “Friendshoring? Nearshoring? Reshoring?” (2026); ITIF, “The Global Trade Battleground: US-China Competition in the Global South” (April 2026); WTO Global Trade Outlook, March 2026; US Census Bureau; US Joint Economic Committee Monthly Trade Update, April 2026; Mexico Business News; Fortune; South China Morning Post. All trade share figures based on goods imports unless otherwise noted. Data as of June 16, 2026.