Debates about tariff evasion have focused on Chinese goods entering the US. This column examines the reverse flow and finds that China’s retaliatory tariffs on imports from the US also induced evasion, especially through the misreporting of US-origin goods as originating from other third countries. Trade data and customs enforcement cases help distinguish origin misreporting from genuine trade diversion. The estimates attribute about 35% of the tariff-induced reduction in recorded imports to underreporting, highlighting the need for stronger country-of-origin verification and showing how tariff evasion can make US–China decoupling appear greater than it actually is.

Tariff evasion has become a prominent concern in debates about the US–China trade war. Bloomberg has reported on the widening US–China trade-data gap, while the New York Times has documented offers to bring Chinese goods into the US at lower duties through fraudulent declarations and third-country routing (Curtis and Mayger 2026, Swanson and Gamio 2025). Wolfe (2025), writing in VoxEU, argues that avoidance of US tariffs helps explain why US-recorded imports from China fell more sharply than Chinese-recorded exports to the US. Auclair (2026) at the Peterson Institute for International Economics similarly cautions against treating the reported fall in US imports as a clean measure of decoupling. 

The recent White House report “The Great Transshipment Scam” has brought origin verification further into the policy spotlight. It also highlights a central challenge: distinguishing legitimate supply-chain adjustment from illegal pass-through trade and origin misreporting (White House 2026). 

However, most of the above discussions concern only Chinese goods entering the US. The reverse flow also deserves close attention. China’s retaliatory tariffs on US imports created similar incentives to conceal US imports or falsely declare another country of origin. Our research examines this overlooked side of the trade war and combines trade-data patterns with customs cases to distinguish genuine trade diversion from origin misreporting.

The overlooked side of the trade war

Beginning in 2018, the US and China imposed additional origin-specific tariffs on each other’s goods. As Figure 1 shows, China’s average tariff on US goods rose from around 8% to roughly 20%, while its tariffs on other origins remained below 10%. The resulting tariff wedge across origins made a false origin declaration potentially profitable. An importer could save duties by declaring a US product as originating in a lower-tariff country.

Figure 1 China’s retaliatory tariffs created tariff wedges across origins

Notes: Trade-weighted average tariff rates, 2018–2023, using 2017 exports to the world as weights. Blue lines show China’s tariffs while orange lines show US tariffs. Solid lines apply to the bilateral partner and dashed lines to other origins.
Source: Bown (2025).

In our paper, Liu et al. (2026), we detect tariff evasion associated with China’s retaliatory tariffs on US imports using the trade gap following Fisman and Wei (2004). Figure 2 plots US-reported exports to China, China-reported imports from the US, and the corresponding trade gap, defined as the former minus the latter, using data from UN Comtrade. 

After the outbreak of the US-China trade war, the trade flow from the US to China dropped dramatically. Conceptually, additional tariffs imposed by China should lead to a synchronised decline in trade value reported by both the US and China. However, as shown in Figure 2, a distinct discrepancy regarding the magnitude of reported trade contraction emerges. The total import value from the US reported by China declines by 16.3% between 2018 and 2019, which is in sharp contrast to the 7.4% decline in total exports to China reported by the US, resulting in an increase in the trade gap. 

Figure 2 US–China trade flows and the aggregate reporting gap

Source: Liu et al. (2026).

Figure 3 further compares products subject to China’s retaliatory tariffs with those never targeted. Between 2018 and 2019, the reporting gap for tariffed products increased towards zero, as China-reported imports fell more than US-reported exports. The gap for untariffed products remained broadly stable. This contrast is consistent with increased import underreporting following retaliation, pointing to potential tariff evasion. Building on this suggestive evidence, our difference-in-differences analysis shows that China’s retaliatory tariffs increased import underreporting, providing systematic evidence of tariff evasion.

Figure 3 US–China trade trends by retaliatory tariff status

Source: Liu et al. (2026).

Trade diversion or origin misreporting?

Traditionally, importers can evade duties by understating quantities or prices, or by misclassifying high-tariff goods into low-tariff categories. During the US-China trade war, we find significant evidence of quantity underreporting and product misclassification, but little evidence of price underreporting. More importantly, we find evidence that US-origin goods are misreported as originating in lower-tariff countries. Our analysis helps distinguish origin fraud from legitimate transshipment and supply-chain adjustment. 

Figure 4 illustrates the distinction. Genuine trade diversion changes sourcing or production. China may buy from a third-country producer, or goods may undergo sufficient processing there to acquire a new origin. As trade is genuinely redirected, both US-recorded exports to China and China-recorded imports from the US decline. Therefore, this adjustment does not widen their bilateral reporting gap.

In contrast, origin misreporting changes the declaration. Goods may pass through a third country, be repackaged, or receive false documentation to wash away the US origin. If the US continues to record them as exports destined for China but Chinese customs attributes them to another origin, China’s recorded imports from the US fall relative to US-recorded exports to China. The bilateral reporting gap widens. A corresponding discrepancy emerges for the countries falsely declared as the origin: China records more imports from them than they report exporting to China. Their trade gaps therefore move in the opposite direction to the US–China trade gap.

Figure 4 Genuine trade diversion and origin misreporting leave different reporting patterns

Notes: Stylised illustration of the mechanisms in Liu et al. (2026), Figure 6. 

By examining these corresponding changes in trade gaps, our analysis helps distinguish legitimate trade diversion from origin misreporting. Our work finds that for products subject to retaliatory tariffs, the US–China trade gap — exporter-reported exports minus China-reported imports — increases relative to untariffed products. More importantly, origin misreporting is more pronounced where alternative origins offer plausible maritime connections and substantial trade with China before the trade war. Shipping connections make the claimed route credible while established trade provides statistical cover, allowing relabelled US goods to blend into existing flows. Together, these patterns link the reporting discrepancies to incentives for false origin declarations.

Evidence from the origins receiving shifted declarations

Origin misreporting should also leave a trace in the records of the country named as the alternative origin. China would record imports from that country that the country itself does not report exporting to China. We find precisely this recipient-side pattern: origins offering stronger incentives for misreporting show relatively more China-recorded imports than their own recorded exports to China. 

This recipient-side evidence strengthens the interpretation that declarations move away from the targeted origin towards lower-tariff alternatives. The discrepancy appears on both sides of the reporting shift. It also connects with Wolfe’s (2025) suggestion that persistent tariffs could alter trade-data gaps with other major partners: examining only the targeted bilateral relationship can miss where the shifted declarations appear.

Evidence from customs enforcement

Customs enforcement records provide direct corroboration. We assemble 5,179 disclosed trade-violation cases from six major Chinese coastal customs districts over 2022–2026. Of 1,414 import tariff-evasion cases, 50 explicitly mention the trade war or retaliatory tariffs, or identify the US as the actual origin.

Origin misreporting appears in 64% of these US-related cases, compared with 5.1% of the other cases (Figure 5). This striking difference supports our previous findings: disguising US origin is a common way to evade China’s retaliatory tariffs.

Figure 5 Origin misreporting is prominent in US-related customs cases

Notes: A case may involve multiple evasion methods. 
Source: Liu et al. (2026), Figure 8.

Implications for measuring decoupling

The reporting distortion is economically substantial. We estimate cumulative imports after correcting the additional underreporting and compare them with a no-retaliation counterfactual. Within our estimation sample, recorded imports over 2018–2023 are an estimated $477.5 billion below that counterfactual. Approximately $310.4 billion reflects underlying trade contraction caused by retaliatory tariffs, while $167.1 billion reflects additional underreporting associated with tariff evasion (Figure 6).

Thus, about 35% of the estimated tariff-induced reduction in recorded imports reflects reporting distortions. This changes the interpretation of economic separation: part of the apparent reduction in bilateral trade comes from changing declarations rather than real decoupling. A shift towards third-country origins in customs records can therefore make supply-chain relocation appear larger than it is.

Figure 6 Reporting distortions account for about 35% of the estimated reduction in recorded imports

Notes: Baseline decomposition of the cumulative tariff-induced reduction in China-reported imports from the US in the estimation sample, 2018–2023, relative to the no-retaliation counterfactual. 
Source: Liu et al. (2026).

Conclusion

Our findings have two main policy implications. First, trade data should be used cautiously to measure US–China decoupling and supply-chain relocation, as failing to account for origin misreporting can overstate both. Second, customs enforcement should strengthen origin verification by combining origin documentation, shipping routes, and production-capacity data. AI tools can help detect inconsistencies across these sources and flag suspicious declarations for further investigation.

The mechanism extends beyond China’s retaliation. As tariffs become more differentiated across trading partners, larger cross-origin wedges strengthen incentives to shift declarations, relabel goods, and use third-country transit to conceal origin. Antidumping and countervailing duties can generate similar incentives. Reliable measurement of supply-chain adjustment and effective customs enforcement therefore depend on distinguishing a change in a good’s declared origin from a change in where it was produced.

Source : VOXeu