American importers and Chinese exporters have spent the past two years gaming the tariff system, and a new Federal Reserve study puts a dollar figure on the deception: US $67 billion in manipulated trade flows during 2020 alone.
The paper, released through the Fed’s Board of Governors by economists Hunter Clark and Anna Wong, tracks a bizarre statistical flip. Before 2018, goods arriving from China carried higher U.S. customs values than the export prices China reported, because the American figure includes freight and insurance. That relationship inverted almost overnight in March 2020 and stayed flipped.
China’s own data showed higher export prices for the same containers that U.S. ports logged at lower import prices. The reversal was too sudden for transport-cost swings to explain. “The incentive to mis-report rose in lock-step with the duty rate,” the authors wrote.
Their calculation: US $55 billion of the apparent 2020 deficit reduction came from U.S. firms understating invoice values to dodge duties. Another US $12 billion came from Chinese exporters overstating shipments to pocket larger value-added-tax rebates. That leaves US $20 billion genuinely unexplained.
The Trump administration slapped duties on roughly US $370 billion of Chinese products, arguing the levies would force Beijing to end forced-technology transfers and subsidised over-capacity. Instead, the study shows, both sides found creative compliance.
For American companies, the arithmetic was simple. A lower invoice meant a lower duty bill. For Chinese exporters, a higher export declaration meant a bigger VAT rebate from Beijing. The two incentives aligned perfectly, and the trade data broke.
The consequences ripple beyond customs ledgers. Trade negotiators on both sides now face a data fog. If US $55 billion of the deficit reduction is tax avoidance, not real export gains, then the trade balance never improved as much as the headline numbers suggested. That matters for future tariff negotiations — both sides have been bargaining over a phantom.
U.S. importers who played by the rules got undercut. Honest firms paid full duties while competitors shaved billions off their invoices. The study does not name companies, but the incentive structure it describes would have rewarded the least scrupulous operators in any industry importing from China.
Chinese tax authorities face a different problem. If exporters systematically overstated shipments, Beijing paid VAT rebates on phantom goods. The US $12 billion figure is the Fed economists’ estimate of that overpayment. It may be low.
The paper also raises questions about enforcement. U.S. Customs and Border Protection examines a fraction of container manifests. The invoice values Clark and Wong analyzed passed through the system undetected for months. The reversal started in March 2020 and continued through the end of the year.
For the broader economy, the study suggests tariff policy produced unintended side effects that dwarf the primary effects. The stated goal was to reduce the trade deficit and pressure China on intellectual property. What actually happened was a US $88 billion drop in stable two-way commerce compared with 2017, plus a parallel market in mispriced goods.
More than half of that deficit shrinkage was bookkeeping games. The real trade picture is worse than the official numbers show.
Clark and Wong do not offer policy prescriptions. Their paper is descriptive — a forensic accounting of what the tariff war did to the data. But the implications are clear. Any future administration that wants to use tariffs as a negotiating tool will have to account for the fact that the numbers on the customs forms may not match the numbers on the ships.
