Skip to content
NewsroomTrade

September 4, 2026, 4:11 PM · Data Story · 12 min read

Computers, Accessories and Chips Accounted for 123% of July's Goods-Import Increase

Three technology categories rose by $14.757 billion while all other goods collectively fell by $2.796 billion. Prices moved far less than import values, but the available monthly data cannot determine how much of the surge came from quantities, product mix, shipment timing or tariff treatment.

By Cumulant Research

Hover or tap an underlined term to see its definition.

Close-up of the reflective patterned surface of a 12-inch silicon wafer
A 12-inch silicon wafer illustrates one physical input behind the semiconductor category that helped drive July's import increase. Photo: 2x910, CC BY-SA 4.0, via Wikimedia Commons

The quick version

  • Seasonally adjusted Census-basis goods imports increased $11.961 billion from June to July. Computers added $6.895 billion, computer accessories $6.647 billion and semiconductors $1.215 billion. [BEA Exhibit 8](https://www.bea.gov/sites/default/files/2026-09/trad0726.pdf)
  • The three categories contributed $14.757 billion, or 123.4 percent of the net increase. The share exceeds 100 percent because every other goods category combined fell by $2.796 billion. [BEA Exhibit 8](https://www.bea.gov/sites/default/files/2026-09/trad0726.pdf)
  • Their combined import value increased about 24.7 percent, calculated from the rounded June and July levels, while the corresponding BLS import-price index increased about 1.8 percent. That weakens a uniform-price explanation but does not separate quantities from product mix. [BEA Exhibit 8](https://www.bea.gov/sites/default/files/2026-09/trad0726.pdf) [BLS price series](https://fred.stlouisfed.org/data/IR213)
  • Tariff documents exempt numerous computer and integrated-circuit classifications, but no published concordance shows what share of these broad end-use categories was exempt. July also blended tariff regimes before and after July 24. [USTR final action](https://ustr.gov/sites/default/files/files/Press/Releases/2026/FLIP%20301%20Investigation%20Final%20Action%20FRN%207-23-26%20FINAL.pdf)
  • The evidence establishes an unusually concentrated increase, not its cause. August data, detailed country and product records, quantities and a compatible calculated-duty analysis are needed to distinguish demand, inventory, timing, origin and tariff explanations. [BEA release](https://www.bea.gov/news/2026/us-international-trade-goods-and-services-july-2026)

Figure

Three categories contributed more than the entire July increase

Change in seasonally adjusted Census-basis goods imports

Computers
6.9
Accessories
6.65
Semiconductors
1.22
All other goods
-2.8
Total goods
11.96

The three highlighted categories sum to $14.757 billion. All other goods is the $11.961 billion total change minus those contributions. Signed bars must cross a prominent zero line.

Source: https://www.bea.gov/sites/default/files/2026-09/trad0726.pdf · $ billions · June to July 2026

Why it matters

July's wider US trade deficit was heavily shaped by a narrow group of technology imports rather than a broad increase across goods. That concentration matters to semiconductor and computer-hardware businesses, technology supply chains and investors interpreting trade data, but the evidence does not yet show whether it reflects lasting demand, temporary shipment timing or tariff-related behavior.

A $14.757 billion increase inside an $11.961 billion increase

The headline trade deficittrade deficitA trade deficit means the measured value of imports exceeds the value of exports during the period. [BEA](https://www.bea.gov/news/2026/us-international-trade-goods-and-services-july-2026) widened by $17.4 billion in July, from $71.2 billion to $88.6 billion. That change combined a $6.6 billion decline in exports with a $10.8 billion rise in imports on the broader balance-of-payments basis. [BEA](https://www.bea.gov/news/2026/us-international-trade-goods-and-services-july-2026)

The striking result appears one layer down, in seasonally adjustedseasonally adjustedSeasonal adjustment estimates and removes recurring calendar patterns so adjacent months are more meaningfully comparable. [BEA explanatory notes](https://www.bea.gov/news/2026/us-international-trade-goods-and-services-july-2026) Census-basis goods imports. The total rose $11.961 billion, but computers rose $6.895 billion, computer accessories rose $6.647 billion and semiconductors rose $1.215 billion. Together they contributed $14.757 billion, equal to 123.4 percent of the net increase. [BEA Exhibit 8](https://www.bea.gov/sites/default/files/2026-09/trad0726.pdf)

A contribution can exceed 100 percent when other components move backward. Imagine five people carrying a table forward: three push it 14.757 feet, while the other two pull it back 2.796 feet. The table's net movement is 11.961 feet. Here, the three technology categories supplied the forward push and all other goods provided the offset.

Figure

Three categories contributed more than the entire July increase

Change in seasonally adjusted Census-basis goods imports

Computers
6.9
Accessories
6.65
Semiconductors
1.22
All other goods
-2.8
Total goods
11.96

The three highlighted categories sum to $14.757 billion. All other goods is the $11.961 billion total change minus those contributions. Signed bars must cross a prominent zero line.

Source: https://www.bea.gov/sites/default/files/2026-09/trad0726.pdf · $ billions · June to July 2026

What the category names do and do not reveal

The official end-use codes are M21300 for computers, M21301 for computer accessories, peripherals and parts, and M21320 for semiconductors and related devices. End-use classification describes principal economic use. It does not identify a single buyer, application or tarifftariffA tariff is a duty imposed on an imported good. [Census definitions](https://www.census.gov/foreign-trade/reference/definitions/index.html) rule. [BEA end-use codes](https://www.bea.gov/sites/default/files/2026-04/bea-end-use-codes.pdf)

No server or AI finding

The published table does not show that the increase consisted of servers, artificial-intelligence processors or data-center equipment. Those products may exist inside the categories, but Exhibit 8 does not quantify them. [BEA Exhibit 8](https://www.bea.gov/sites/default/files/2026-09/trad0726.pdf)

July levels were $34.603 billion for computers, $26.792 billion for accessories and $13.048 billion for semiconductors. Their combined $74.443 billion level was about 24.7 percent above the $59.687 billion sum of the displayed June levels. Because BEA cautions that displayed details may not equal totals after seasonal adjustment and rounding, the calculated percentage is appropriately reported to one decimal place. [BEA Exhibit 8](https://www.bea.gov/sites/default/files/2026-09/trad0726.pdf)

Prices rose, but not enough to tell the whole story

The Bureau of Labor Statistics import-price indexprice indexA price index tracks how prices for a defined basket change relative to a reference period. [BLS](https://www.bls.gov/news.release/ximpim.nr0.htm) for computers, peripherals and semiconductors increased from 54.8 in June to 55.8 in July, a rise of about 1.8 percent. The category's displayed import value increased about 24.7 percent. [BLS price series](https://fred.stlouisfed.org/data/IR213) [BEA Exhibit 8](https://www.bea.gov/sites/default/files/2026-09/trad0726.pdf)

Figure

The category's price index moved much less than its import value

BLS import-price index for computers, peripherals and semiconductors

June
54.8
July
55.8

The index is not seasonally adjusted and increased from 54.8 to 55.8, or about 1.8 percent. The axis should start at zero. This chart does not convert the seasonally adjusted import values into physical quantities.

Source: https://fred.stlouisfed.org/data/IR213 · Index, 2000=100 · June and July 2026

That gap is evidence against a simple story in which every imported item became roughly 25 percent more expensive while purchases were otherwise unchanged. It is not a quantity estimate. The price index is not seasonally adjusted, its product weights need not match the monthly import table, and a shift toward more expensive products can lift import value without appearing as an equal increase in a fixed-basket price index. [BLS price release](https://www.bls.gov/news.release/ximpim.nr0.htm) [BLS series metadata](https://fred.stlouisfed.org/data/IR213)

The national real-goods series offers broader context but not a technology-specific answer: total real goods imports increased 3.8 percent in July while nominal goods imports increased 3.9 percent. Because the release does not publish a monthly real series for these three exact five-digit categories, it cannot convert their dollar surge into a physical-volume increase. [BEA](https://www.bea.gov/news/2026/us-international-trade-goods-and-services-july-2026)

The tariff documents show openings, not a measured burden

The temporary Section 122Section 122Section 122 of the Trade Act authorizes temporary import restrictions for specified international-payments problems. [White House proclamation](https://www.whitehouse.gov/presidential-actions/2026/02/imposing-a-temporary-import-surcharge-to-address-fundamental-international-payments-problems/) surcharge imposed a 10 percent duty beginning February 24 and listed certain electronics among its exceptions. It remained in effect through 12:01 a.m. on July 24. [White House proclamation](https://www.whitehouse.gov/presidential-actions/2026/02/imposing-a-temporary-import-surcharge-to-address-fundamental-international-payments-problems/)

USTR's replacement Section 301Section 301Section 301 authorizes US trade action in response to foreign practices the government determines are actionable under the Trade Act. [USTR](https://ustr.gov/about/policy-offices/press-office/press-releases/2026/july/ustr-takes-action-forced-labor-section-301-investigations) action applied additional duties of 10 percent or 12.5 percent across 60 investigated economies, subject to product and other exemptions. It became effective for consumption entries on July 24, with an in-transit exceptionin-transit exceptionAn in-transit exception preserves earlier treatment for qualifying goods already on their final journey when a new tariff begins. [USTR final action](https://ustr.gov/sites/default/files/files/Press/Releases/2026/FLIP%20301%20Investigation%20Final%20Action%20FRN%207-23-26%20FINAL.pdf) for qualifying goods entered before July 28. USTR said the action responded to the covered economies' treatment of imports produced with forced labor. [USTR notice](https://ustr.gov/sites/default/files/files/Press/Releases/2026/FLIP%20301%20Investigation%20Final%20Action%20FRN%207-23-26%20FINAL.pdf) [USTR announcement](https://ustr.gov/about/policy-offices/press-office/press-releases/2026/july/ustr-takes-action-forced-labor-section-301-investigations)

The exemption schedules include numerous classifications in HTSHTSThe Harmonized Tariff Schedule classifies US imports under detailed product codes used to determine customs treatment. [Census definitions](https://www.census.gov/foreign-trade/reference/definitions/index.html) heading 8471 for computers and in heading 8542 for integrated circuits. But end-use categories and HTS classifications are different systems. End use is the aisle sign; HTS is the barcode on each product. Finding familiar barcodes on an exemption list does not show what share of the aisle's value received the exemption. [USTR exemption list](https://ustr.gov/sites/default/files/files/Press/Releases/2026/FLIP%20301%20Investigation%20Final%20Action%20FRN%207-23-26%20FINAL.pdf) [Census definitions](https://www.census.gov/foreign-trade/reference/definitions/index.html)

An exemption from one program also does not mean duty-free entry. Ordinary HTS duties and other trade measures can remain applicable. Certain advanced-computing chips were subject to a separate 25 percent Section 232Section 232Section 232 permits import restrictions when a Commerce Department investigation finds that imports threaten national security. [White House proclamation](https://www.whitehouse.gov/presidential-actions/2026/01/adjusting-imports-of-semiconductors-semiconductor-manufacturing-equipment-and-their-derivative-products-into-the-united-states/) tariff beginning January 15, although the proclamation created exceptions for US data-center use and several other domestic uses. [USTR notice](https://ustr.gov/sites/default/files/files/Press/Releases/2026/FLIP%20301%20Investigation%20Final%20Action%20FRN%207-23-26%20FINAL.pdf) [Section 232 proclamation](https://www.whitehouse.gov/presidential-actions/2026/01/adjusting-imports-of-semiconductors-semiconductor-manufacturing-equipment-and-their-derivative-products-into-the-united-states/)

Why no effective tariff rate appears here

The surge uses seasonally adjusted general importsgeneral importsGeneral imports include goods arriving from abroad whether they enter commerce immediately or first enter a bonded warehouse or foreign-trade zone. [Census definitions](https://www.census.gov/foreign-trade/reference/definitions/index.html), while Census calculated-duty fields describe unadjusted imports for consumptionimports for consumptionImports for consumption are goods cleared into US commerce immediately or after withdrawal from a bonded warehouse or foreign-trade zone. [Census definitions](https://www.census.gov/foreign-trade/reference/definitions/index.html). General imports include arrivals into warehouses and foreign-trade zones; consumption imports count goods when cleared into US commerce. A ratio from the second population cannot be presented as the tariff rate on the first population's increase. Census also warns that calculated dutycalculated dutyCalculated duty is a Census estimate that may overstate or understate duties and does not necessarily equal the amount paid. [Census definitions](https://www.census.gov/foreign-trade/reference/definitions/index.html) does not necessarily equal duty paid. [Census definitions](https://www.census.gov/foreign-trade/reference/definitions/index.html) [Census API variables](https://api.census.gov/data/timeseries/intltrade/imports/enduseimport/variables.html)

The concentration predates July

From January through July, computers were $110.857 billion above the comparable 2025 period, accessories were $53.355 billion higher and semiconductors were $31.281 billion higher. Their displayed changes sum to $195.493 billion. Total Census-basis goods imports increased only $23.633 billion, leaving a negative $171.860 billion residualresidualA residual is what remains after identified components are subtracted from a total. for all other goods. [BEA Exhibit 8](https://www.bea.gov/sites/default/files/2026-09/trad0726.pdf)

Figure

The same categories dominate the year-to-date increase

January through July 2026 compared with the same period of 2025

Technology bundle
195.49
All other goods
-171.86
Total goods
23.63

The displayed category changes total $195.493 billion. Subtracting them from the $23.633 billion total leaves a negative $171.860 billion residual for all other goods. BEA warns that displayed details may not equal totals because of seasonal adjustment and rounding. Signed bars must cross zero.

Source: https://www.bea.gov/sites/default/files/2026-09/trad0726.pdf · $ billions · January-July 2026 versus January-July 2025

The bundle therefore generated 827.2 percent of the net year-to-date increase using the displayed values. That percentage is mathematically accurate but should not be read as an economy-wide growth rate. It says that large technology gains were mostly offset by declines elsewhere. [BEA Exhibit 8](https://www.bea.gov/sites/default/files/2026-09/trad0726.pdf)

Why March 2025 is an imperfect precedent

The revised historical series shows consumer-goods imports rising $23.043 billion from February to March 2025 and then falling $32.503 billion in April. Capital-goods imports rose $3.422 billion in March and fell $2.638 billion in April. [Census historical imports](https://www.census.gov/foreign-trade/statistics/historical/SAIMP.pdf)

Figure

The March 2025 import rush had a different center

Revised seasonally adjusted Census-basis monthly changes

CategoryFebruary-MarchMarch-April
Consumer goods23.043-32.503
Capital goods3.422-2.638
Total goods18.61-65.415

March's increase was concentrated in consumer goods, while July 2026 was concentrated in capital-goods technology categories. The episodes therefore are not like-for-like evidence of a repeated mechanism.

Source: https://www.census.gov/foreign-trade/statistics/historical/SAIMP.pdf · $ billions · February-April 2025

That reversal demonstrates how a monthly import rush can unwind, but the analogy breaks at the category level. The March 2025 episode centered on consumer goods, while July 2026 centered on computers, accessories and semiconductors inside capital goods. Different products, buyers and policy dates prevent the earlier episode from establishing the cause or likely duration of the current increase. [Census historical imports](https://www.census.gov/foreign-trade/statistics/historical/SAIMP.pdf) [BEA Exhibit 8](https://www.bea.gov/sites/default/files/2026-09/trad0726.pdf)

Five explanations, none yet decisive

The published evidence leaves at least five competing explanations: a lower relative customs burden for some technology products, stronger final demand or inventoryinventoryInventory consists of goods a business owns but has not yet sold or used. accumulation, shipment timing around July 24, higher prices or a more expensive product mixproduct mixProduct mix means the combination of lower-priced and higher-priced items inside a broad statistical category., and a shift toward particular countries of origin. These mechanisms can operate together.

Figure

Five explanations require five different tests

Evidence that would strengthen or weaken each account

ExplanationEvidence that would support itEvidence that would weaken it
Lower relative tariff burdenHigh exemption coverage and a persistently lower calculated-duty ratioSimilar duty ratios after controlling for origin and product
Demand or inventoryHigher quantities, inventories or investment, followed by sustained importsA one-month reversal without related demand or inventory growth
Shipment timingEntry-date clustering around July 24 and a subsequent August declineNo transition-date clustering and continued growth after July
Prices or product mixUnit values or product-level prices explain most of the dollar increaseQuantities rise strongly after comparable-product price adjustment
Country of originGrowth concentrates in origins receiving more favorable treatmentOrigin shares remain stable after compatible seasonal adjustment

These are decision rules for future evidence, not findings that a particular explanation caused July's increase.

Source: https://www.census.gov/data/developers/data-sets/international-trade.html

The tariff explanation would strengthen if a product-level HTS-to-end-use concordance showed high exemption coverage and if a calculated-duty-to-consumption-value ratio remained materially below that for other goods after controlling for origin. It would weaken if comparable products and origins carried similar total duty ratios. [Census API variables](https://api.census.gov/data/timeseries/intltrade/imports/enduseimport/variables.html) [USTR notice](https://ustr.gov/sites/default/files/files/Press/Releases/2026/FLIP%20301%20Investigation%20Final%20Action%20FRN%207-23-26%20FINAL.pdf)

A timing explanation would strengthen if entry-date records clustered immediately before the July 24 transition and imports subsequently fell. A demand or inventory explanation would strengthen if comparable-product quantities, business inventories or technology investment rose and imports remained elevated after the transition. Monthly nominal values alone cannot choose between them. [USTR effective-date provision](https://ustr.gov/sites/default/files/files/Press/Releases/2026/FLIP%20301%20Investigation%20Final%20Action%20FRN%207-23-26%20FINAL.pdf) [BEA explanatory notes](https://www.bea.gov/news/2026/us-international-trade-goods-and-services-july-2026)

Who has something at stake

  • Importers bear the customs obligation at entry, although later price changes determine how much cost is absorbed or passed onward. [Census definitions](https://www.census.gov/foreign-trade/reference/definitions/index.html)
  • Technology buyers could face different costs depending on the product's HTS code, origin, use and applicable tariff program. [USTR final action](https://ustr.gov/sites/default/files/files/Press/Releases/2026/FLIP%20301%20Investigation%20Final%20Action%20FRN%207-23-26%20FINAL.pdf)
  • Domestic competing producers could receive greater price protection on covered imports, but the trade table does not measure their output, employment or profits.
  • Foreign suppliers could gain or lose orders as relative tariff treatment changes by origin, but Exhibit 8 does not identify the origins behind the three category increases. [BEA Exhibit 8](https://www.bea.gov/sites/default/files/2026-09/trad0726.pdf)
  • Workers affected by forced labor are central to USTR's stated policy rationale, but aggregate import values do not measure working conditions or whether the tariffs changed foreign enforcement. [USTR](https://ustr.gov/about/policy-offices/press-office/press-releases/2026/july/ustr-takes-action-forced-labor-section-301-investigations)

No asset-price event study was conducted for this article. Any movement in technology shares, currencies or bond yields around the release would be a market reactionmarket reactionMarket reaction means a change in traded asset prices around news and is different from the news event's later effect on production, employment or spending., not proof of the trade figures' economic effect.

The finding

July's goods-import increase was not broad. Three technology categories rose by more than the total because the rest of the goods ledger contracted.

That concentration is the publishable finding. The smaller movement in the corresponding price index makes a uniform-price story insufficient, while the year-to-date figures show the shift did not begin in July. Neither result proves a demand boom, an inventory build or a tariff effect. [BEA Exhibit 8](https://www.bea.gov/sites/default/files/2026-09/trad0726.pdf) [BLS price series](https://fred.stlouisfed.org/data/IR213)

August will be the first full month after the July 24 Section 301 transition, although qualifying in-transit entries and other tariff programs will still complicate interpretation. The next US trade release, covering August, is scheduled for October 6. [USTR final action](https://ustr.gov/sites/default/files/files/Press/Releases/2026/FLIP%20301%20Investigation%20Final%20Action%20FRN%207-23-26%20FINAL.pdf) [BEA release schedule](https://www.bea.gov/news/schedule)

What to watch

  • Whether August imports reverse or extend the July increases in computers, accessories and semiconductors.
  • Detailed product and country records that can identify which goods and origins drove the increase.
  • A compatible comparison of import values, quantities and calculated duties across the relevant tariff classifications.
  • Revisions to the seasonally adjusted monthly and year-to-date estimates.

How we did this

  • We used the September 3, 2026 BEA-Census trade release and Exhibit 8, retrieved September 4, 2026. Monthly values are seasonally adjusted, nominal, Census-basis general imports reported in millions of dollars. [BEA release](https://www.bea.gov/news/2026/us-international-trade-goods-and-services-july-2026) [Exhibit 8](https://www.bea.gov/sites/default/files/2026-09/trad0726.pdf)
  • The technology bundle contains the exact five-digit import end-use categories M21300, M21301 and M21320. [BEA end-use codes](https://www.bea.gov/sites/default/files/2026-04/bea-end-use-codes.pdf)
  • Monthly technology contribution equals $6.895 billion plus $6.647 billion plus $1.215 billion, or $14.757 billion. Dividing by the $11.961 billion total goods increase produces 123.4 percent. All other goods equals $11.961 billion minus $14.757 billion, or negative $2.796 billion. [BEA Exhibit 8](https://www.bea.gov/sites/default/files/2026-09/trad0726.pdf)
  • The technology bundle's displayed level equals $74.443 billion in July and $59.687 billion in June. The approximately 24.7 percent increase uses those rounded levels; the sum of BEA's separately displayed changes differs by $0.001 billion because details may not equal totals after seasonal adjustment and rounding. [BEA Exhibit 8](https://www.bea.gov/sites/default/files/2026-09/trad0726.pdf)
  • The price comparison uses the BLS import-price index for computers, peripherals and semiconductors, series IR213. Its increase from 54.8 to 55.8 is approximately 1.8 percent. The index is not seasonally adjusted, so it is used as a diagnostic comparison, not as a direct deflator for the seasonally adjusted import values. [BLS series](https://fred.stlouisfed.org/data/IR213)
  • Year-to-date technology growth equals $110.857 billion plus $53.355 billion plus $31.281 billion, or $195.493 billion. Subtracting this from total goods growth of $23.633 billion produces the negative $171.860 billion residual. [BEA Exhibit 8](https://www.bea.gov/sites/default/files/2026-09/trad0726.pdf)
  • The March 2025 comparison uses the currently published revised historical series, not the superseded numbers in the original March 2025 release. Monthly changes were calculated by subtracting adjacent seasonally adjusted levels. [Census historical imports](https://www.census.gov/foreign-trade/statistics/historical/SAIMP.pdf)
  • We reviewed the Section 122 proclamation, the Section 301 final action and the semiconductor Section 232 proclamation to identify dates, stated rates and listed exceptions. We did not infer exemption coverage from selected tariff lines. [Section 122](https://www.whitehouse.gov/presidential-actions/2026/02/imposing-a-temporary-import-surcharge-to-address-fundamental-international-payments-problems/) [Section 301](https://ustr.gov/sites/default/files/files/Press/Releases/2026/FLIP%20301%20Investigation%20Final%20Action%20FRN%207-23-26%20FINAL.pdf) [Section 232](https://www.whitehouse.gov/presidential-actions/2026/01/adjusting-imports-of-semiconductors-semiconductor-manufacturing-equipment-and-their-derivative-products-into-the-united-states/)
  • A defensible tariff analysis would separately sum Census CON_VAL_MO and CAL_DUT_MO for end-use codes 21300, 21301 and 21320, subtract those fields from compatible world consumption-import totals for the comparison group, and report both numerators, denominators and their percentage-point difference. It would describe a calculated-duty ratio, not duties actually paid. [Census API variables](https://api.census.gov/data/timeseries/intltrade/imports/enduseimport/variables.html) [Census definitions](https://www.census.gov/foreign-trade/reference/definitions/index.html)
  • Testing exemption coverage would additionally require an HTS-to-end-use concordance, origin-specific values and every applicable duty program. Testing the July 24 change itself would require entry-date data or an August comparison because the July monthly total blends two regimes. [USTR final action](https://ustr.gov/sites/default/files/files/Press/Releases/2026/FLIP%20301%20Investigation%20Final%20Action%20FRN%207-23-26%20FINAL.pdf)

What this cannot establish

  • The monthly surge is measured with seasonally adjusted general imports, not imports for consumption. Goods entering warehouses or foreign-trade zones can appear in the former before appearing in the latter. [Census definitions](https://www.census.gov/foreign-trade/reference/definitions/index.html)
  • The figures are nominal values. They combine prices, quantities and product mix, and no monthly real series for the exact three five-digit categories is published in Exhibit 8. [BEA release](https://www.bea.gov/news/2026/us-international-trade-goods-and-services-july-2026)
  • The BLS price index is not seasonally adjusted and does not supply a direct quantity measure for the seasonally adjusted import values. [BLS series](https://fred.stlouisfed.org/data/IR213)
  • No HTS-to-end-use concordance was constructed, so the share of each category's customs value covered by tariff exemptions is unknown.
  • No origin-by-program table was available for the three categories. Applicable treatment can depend on origin, ordinary duties, Section 232, Section 301, product exemptions and declared use. [USTR final action](https://ustr.gov/sites/default/files/files/Press/Releases/2026/FLIP%20301%20Investigation%20Final%20Action%20FRN%207-23-26%20FINAL.pdf) [Section 232 proclamation](https://www.whitehouse.gov/presidential-actions/2026/01/adjusting-imports-of-semiconductors-semiconductor-manufacturing-equipment-and-their-derivative-products-into-the-united-states/)
  • July combines entries under the Section 122 regime, the Section 301 regime and an in-transit exception. A full-month statistic cannot isolate the July 24 policy change. [Section 122](https://www.whitehouse.gov/presidential-actions/2026/02/imposing-a-temporary-import-surcharge-to-address-fundamental-international-payments-problems/) [Section 301](https://ustr.gov/sites/default/files/files/Press/Releases/2026/FLIP%20301%20Investigation%20Final%20Action%20FRN%207-23-26%20FINAL.pdf)
  • Calculated-duty estimates may be understated or overstated and do not necessarily equal duties paid. [Census definitions](https://www.census.gov/foreign-trade/reference/definitions/index.html)
  • The end-use categories do not isolate servers, artificial-intelligence processors, data-center hardware or any particular importer. [BEA end-use codes](https://www.bea.gov/sites/default/files/2026-04/bea-end-use-codes.pdf)
  • The historical March 2025 comparison involves different product categories and policy circumstances, so it illustrates a possible reversal pattern rather than predicting one.
  • Monthly trade data are revised, and detailed components may not equal totals because of seasonal adjustment and rounding. [BEA Exhibit 8](https://www.bea.gov/sites/default/files/2026-09/trad0726.pdf)

This is AI-assisted analysis under stated assumptions; it is not investment advice or a price target. Figures are as of the publication date and trace to the cited sources; markets and disclosures change.

Sources

  1. 01U.S. International Trade in Goods and Services, July 2026, U.S. Bureau of Economic Analysis and U.S. Census BureauPrimary
  2. 02U.S. International Trade in Goods and Services, July 2026, Full Release and Exhibits, U.S. Bureau of Economic Analysis and U.S. Census BureauData
  3. 03Imports of Goods by Principal End-Use Category, Seasonally Adjusted Historical Series, U.S. Census BureauData
  4. 04U.S. Bureau of Economic Analysis End-Use Codes, U.S. Bureau of Economic AnalysisData
  5. 05Import Price Index for Computers, Peripherals and Semiconductors, Federal Reserve Bank of St. Louis, sourced from the U.S. Bureau of Labor StatisticsData
  6. 06U.S. Import and Export Price Indexes, July 2026, U.S. Bureau of Labor StatisticsPrimary
  7. 07Imposing a Temporary Import Surcharge to Address Fundamental International Payments Problems, The White HousePrimary
  8. 08Notice of Actions in Forced Labor Section 301 Investigations, Office of the United States Trade RepresentativePrimary
  9. 09USTR Takes Action in Forced Labor Section 301 Investigations, Office of the United States Trade RepresentativePrimary
  10. 10Adjusting Imports of Semiconductors, Semiconductor Manufacturing Equipment, and Their Derivative Products, The White HousePrimary
  11. 11International Trade Definitions, U.S. Census BureauPrimary
  12. 12International Trade Imports by End-Use API Variables, U.S. Census BureauData
  13. 13International Trade Data API, U.S. Census BureauData
  14. 14BEA Release Schedule, U.S. Bureau of Economic AnalysisPrimary
US tradegoods importscomputerssemiconductorstariffsforced labor policyus-importstrade-deficitUnited States

Related

Trade

June imports fell, but the data cannot isolate the temporary surcharge

US imports fell $7.3 billion in June and the trade deficit narrowed $4.4 billion, but the published totals combine surcharge-covered goods with exemptions and goods under other tariff rules. Price-adjusted imports rose in the second quarter, and an abnormal 2025 comparison leaves lower demand, shipment timing and supplier switching unresolved. [June FT900](https://www.census.gov/foreign-trade/Press-Release/ft900/ft900_2606.pdf) [Surcharge proclamation](https://www.whitehouse.gov/presidential-actions/2026/02/imposing-a-temporary-import-surcharge-to-address-fundamental-international-payments-problems/)

Container ships, cranes and stacked cargo containers at the Port of Los Angeles, with the Vincent Thomas Bridge behind them.
Trade

Two baselines put Los Angeles June imports 68,490 to 93,874 TEUs above trend, but do not identify why

The Port of Los Angeles recorded 530,557.5 loaded-import TEUs in June. Two trend-and-seasonality checks put that result 68,490 to 93,874 TEUs above their point estimates, but one prediction interval still contains the observation and the available aggregate data cannot assign the gap to early shipping, gateway shifts, stronger demand or model error.

Container ships, cargo cranes and stacked containers at the Port of Los Angeles with the Vincent Thomas Bridge in the background.
Trade

The Trade Deficit Shrunk $189.3 Billion as Four Tech Import Lines Rose $183.5 Billion

The first-half U.S. trade deficit narrowed because exports rose $198.3 billion, not because imports fell. Four technology-hardware import lines rose by $183.5 billion, but the available data cannot show how much represented artificial-intelligence investment or the unwinding of tariff front-loading.

Rows of server racks inside the CyrusOne data center in Carrollton, Texas
Trade

For continuously taxed imports, 80% to 100% of Friday's new duty matches the old surcharge

At 12:01 a.m. Eastern on July 24, the temporary 10% Section 122 surcharge reached its stated endpoint as new Section 301 duties of up to 12.5% took effect. For an otherwise identical import taxed under both policies, 80% to 100% of the new duty numerically matches the old charge, but the available public data cannot support that range for the entire import basket.

Container ships and cargo cranes at the Port of Los Angeles, with the Vincent Thomas Bridge in the background.