DHL Global Connectedness Tracker
Steven A. Altman
Caroline R. Bastian
March 2026
-
As seen in Figure 10, the share of U.S-reported imports coming from China has fallen sharply since the start of the U.S.–China trade war in 2018, with a dramatic plunge in early 2025 when tariffs temporarily surged above 100%. In 2017, China accounted for 22% of U.S. imports, falling to 13% in 2024 and further to 9% over the first nine months of 2025. Meanwhile, China’s share of imports to the rest of the world has continued to increase, a contrast that makes the declines in the share of U.S. imports coming from China even more striking. There is even a modest rising trend in the share of EU imports coming from China.
Despite the sharp declines in the U.S.-reported share of imports from China, we caution against concluding that U.S. reliance on made-in-China goods has significantly diminished. U.S. imports from other countries increasingly contain made-in-China components. While some “transshipment” of Chinese goods via third countries does occur, the main driver of this phenomenon appears to be the use of more Chinese inputs in manufacturing in other countries, especially in Southeast Asia. Available data through 2024 show no meaningful decline in the made-in-China share of overall U.S. consumption.[10]
-
China’s goods exports continued growing in 2025, despite a 20% drop in the value of China’s exports to the U.S. Strikingly, the drop in the value of China’s exports to the U.S. (USD 105 billion) was largely offset by a 13% ($79 billion) increase in China’s exports to the ASEAN (Association of Southeast Asian Nations) region. Coupled with gains in India, Japan, and Taiwan, China more than offset its reduced exports to the U.S. with larger exports to Asian markets. China also expanded exports to Africa by 46 billion U.S. dollars (+26%) and to the EU by 43 billion U.S. dollars (+8%).[11]
For a more granular view of the changing destinations of China’s exports, Figure 11 highlights the countries with the largest increases and decreases in shares of China’s exports comparing 2025 (Jan–Sep) versus 2024. The countries with the largest increases as export destinations for goods from China were Viet Nam, Hong Kong SAR, China, Thailand, India, and Germany, while the countries with the largest decreases were the United States, Russian Federation, Republic of Korea, Mexico, and Brazil.
-
As the geopolitical landscape continues to shift, summary profiles of the composition of countries’ international activity across blocs are especially useful to provide orientation for decision-makers and analysts. Figure 12 presents a template for profiling countries’ international flows across geopolitical categories (along with regions and country income levels). A sample profile for Korea highlights a notable shift in its activity toward U.S.-aligned countries in many areas since 2015, while also highlighting how Korea interacts far more with U.S. allies than with the U.S. itself.
-
The growing emphasis on de-risking international activity has drawn attention to the diversification of international flows across origin/destination countries. Policymakers and business leaders alike aim to avoid excessive reliance on any single partner—particularly those vulnerable to geopolitical instability. To track the diversification of international flows, we use two metrics: a diversification index (one minus the widely-used Herfindahl Hirschman Index of concentration) and the share of flows with countries outside a nation’s top five partners (origin/destination countries for a given flow) (See Figure 13).
The diversification of goods trade began increasing in 2016, with both measures on rising trends up to 2022. However, the diversification index began declining in 2023 and the share-based measure has not changed appreciably since then—suggesting no strong evidence of a sustained diversification trend for goods trade. The diversification measures show a recent decline for announced greenfield FDI and a small increase for announced M&A transactions.
_______
[1] Based on seasonally-adjusted monthly trade volumes from CPB World Trade Monitor.
[2] Figure 2 uses changes in trade value (measured in current U.S. dollars) instead of trade volume data due to more complete recent country-level trade value data.
[3] Composite forecast drawn from IMF World Economic Outlook, Economist Intelligence Unit, Oxford Economics, and S&P Global Market Intelligence, following methodology employed in Steven A. Altman and Caroline R. Bastian, DHL Trade Atlas 2025, DHL Group, 2025.
[4] We measure this using the ratio of trade in value added to world GDP, counting the value of traded goods only once regardless of how many borders they may cross in multi-country supply chains. Recent trends through 2024 were calculated based on data from the Asian Development Bank’s Multiregional Input-Output Tables at current prices (62-country version), and the 2025 projections are based on gross trade and GDP growth projections.
[5] UN Tourism Data Dashboard
[6] For a brief explanation of this scaling method and selected references, see Endnote 1 on p. 101 of the DHL Global Connectedness Report 2026. Additional details are provided in Section 8 of the same report.
[7] Calculated using simple averages across components within DHL Global Connectedness Index pillars, weighted averages across pillars (35% each for Trade and Capital, 15% each for Information and People).
[8] Refer to DHL Global Connectedness Report 2026, page 57, to see how specific countries were classified.
[9] Larger economies tend to trade less intensively than smaller economies, since more of their activity naturally takes place within their large domestic markets. As the world’s two largest economies, it is therefore unsurprising that the share of trade taking place between the U.S. and China is much lower than these two countries’ shares of both GDP and total trade.
[10] For additional discussion and results through 2024, refer to p. 52 of the DHL Global Connectedness Report 2026.
[11] This paragraph is based on data reported by China Customs (Monthly Bulletin released January 8, 2026)
[12] For evidence on why friendshoring could lead to nearshoring/regionalization, see DHL Global Connectedness Report 2026, p. 69.
[13] See DHL Global Connectedness Report 2026 p. 301 for a list of countries classified in each region.