Average effective tariff rate in the U.S.

Updated 17/06/2026

This page was launched on 09/04/2025. It presents historical data on the U.S. average effective tariff rate together with pre- and post-substitution estimates of this rate by The Budget Lab at Yale, reflecting enacted changes to tariff policy. The U.S. customs duty revenue time series obtained from Monthly Treasury Statement releases and seasonally adjusted by me using X-13ARIMA-SEATS is also presented on this page. This customs duty revenue data, together with monthly goods imports data, allow the calculation of the average effective tariff rate, measured as the ratio of customs duty revenue to the value of goods imports. The resulting time series enables tracking this rate on a monthly basis. The spreadsheet is available for download. This page is updated regularly following the release of underlying data.

Additional information

• The pre-substitution estimate of the average effective tariff rate is derived under the assumption that there are no shifts in import shares of different countries in response to tariffs. The post-substitution estimate reflects the rate after imports shift in response to tariffs. According to the analysis published by The Budget Lab at Yale on April 8, 2026, implemented tariffs are equivalent to the average effective tariff rate of 11.8% measured pre-substitution. This estimate takes into account the 10% Section 122 tariffs that were imposed immediately following the SCOTUS ruling on February 20, 2026, which struck down the IEEPA tariffs. The Section 122 tariffs became effective on February 24, 2026, and will expire on July 24, 2026. If these tariffs are not extended, the pre- and post-substitution average effective tariff rates are estimated at 9.7% and 8.2%, respectively. These estimates also include Section 232 pharmaceutical tariffs that will go into effect on September 29, 2026. However, if Section 122 tariffs are made permanent, the pre- and post-substitution average effective tariff rates are estimated at 12.2% and 10.5%, respectively.

• Prior to the release of the Monthly Treasury Statement for May 2026 on June 10, 2026, calculations of the average effective tariff rate presented on this page were based on customs duty revenue net of refunds. In May 2026, tariff refunds ($21.972 billion, NSA) exceeded gross customs duty collections ($21.930 billion, NSA), resulting in negative net customs duty revenue (-$42 million, NSA) – the first negative monthly observation in the available data set, which dates back to January 1980. Before May 2026, the net customs duty revenue time series was seasonally adjusted using a multiplicative specification. This approach requires strictly positive values and therefore cannot accommodate negative observations. The unprecedented negative value recorded in May 2026 rendered this approach infeasible. Historically, tariff refunds were small relative to customs duty collections. As a result, gross and net customs duty revenue moved closely together, and the choice between the two measures had little effect on the measurement of the average effective tariff rate. The invalidation of the IEEPA tariffs and the resulting surge in tariff refunds constituted a structural break in this historical relationship. Accordingly, the use of net customs duty revenue ceased to provide a stable measure of the tariff burden on imports. Figures presenting the underlying not seasonally adjusted time series for monthly gross and net customs duty revenue, as well as tariff refunds, since January 2017 are provided at the end of this page to illustrate this structural break. To maintain a seasonally adjusted measure of the average effective tariff rate, the calculations presented on this page from June 2026 onward are based on gross rather than net customs duty revenue. This change also preserves the objective of measuring the taxation of imports through an average effective tariff rate – a drop in tariff revenue net of refunds to zero does not mean that imports were not subject to tariffs. It is also worth noting that, because gross and net customs duty revenue were historically very similar, the methodological change introduced in June 2026 has only a negligible effect on previously published estimates.

• The seasonal adjustment procedure is performed every time a new monthly observation of customs duty revenue becomes available. As a result, the seasonally adjusted time series of the customs duty revenue, as well as the seasonally adjusted time series of the average effective tariff rate, are revised over time.

• The data from Daily Treasury Statement releases are used occasionally to approximate customs duty revenue in a given month before the scheduled Monthly Treasury Statement is published. Relevant information is provided below the data table if this is the case.

Source: own calculations based on U.S. Treasury, Census Bureau and BEA data.