Trade & Sourcing
The 2025 Steel Comparison That Shows Why Lower Import Charges Do Not Settle the Cost Question
In the complete 2025 flat-rolled steel country records, Canada’s import charges were 0.83% of customs value versus Japan’s 8.23%. Their combined calculated-duty and import-charge ratios were 25.81% and 18.10%. The records cannot rank total supplier costs because comparable country unit…
Editorial evidence cutoff: September 9, 2026. Published September 29, 2026. Observation periods are stated throughout; older figures are retrospective evidence.
A nearby steel source can have a much smaller transport-related charge ratio without having the smaller combined duty-and-charge burden. The completed 2025 flat-rolled steel record makes that distinction visible: Canada's import charges were 0.83% of customs value, compared with 8.23% for Japan. Add calculated duties, and the order reverses.
Canada's combined observed ratio was 25.81%; Japan's was 18.10%. Those figures do not establish that Japanese steel was cheaper. They establish that distance, transport charges and calculated duties cannot be collapsed into one assumption about total cost.
This retrospective is relevant to September 2026 sourcing decisions because it exposes a comparison problem that survives changes in tariff schedules. A percentage advantage in one cost layer can disappear when another layer is included. The actual purchase still requires comparable products, prices and entry facts.
THE TRANSPORT ADVANTAGE IS VISIBLE IN THE RECORD
In 2025, the tracked flat-rolled steel family recorded $1.97 billion of customs value from Canada and $16.39 million of import charges. Japan's corresponding values were $293.94 million and $24.20 million. Dividing charges by each country's customs value produces the 0.83% and 8.23% ratios.
Import charges are broader than a carrier's freight price. They include transport-related expenses such as insurance within the statistical definition. A lower ratio can reflect routing, shipment composition and the value of the merchandise, as well as transportation economics.
The Census trade definitions distinguish these charges from customs value. That separation lets the analysis compare the two layers without treating the percentage as a quoted rate per container, truck or ton. Neither country's ratio is a current transport offer.
CALCULATED DUTIES CHANGE THE COMPARISON
Canada's reported calculated duties on the tracked family were $491.70 million, equivalent to 24.97% of its customs value. Japan's were $28.99 million, equivalent to 9.86%. Adding the charge ratios yields the combined 25.81% and 18.10% burdens.
For Mexico, the same 2025 family record shows an 11.12% calculated-duty ratio and 4.40% import-charge ratio, for a combined 15.52%. The third observation reinforces the need to retain both layers rather than reducing the comparison to a near-versus-far distinction.
These are arithmetic ratios for recorded entries. They do not say that every Canadian steel item faced the same duty, that Japanese products were legally comparable, or that the three source countries supplied the same grades. A lower aggregate ratio can coexist with a higher purchase price for the particular material a factory needs.
THE MISSING UNIT PRICE PREVENTS A SUPPLIER RANKING
The country boards do not supply comparable country-specific unit prices for this analysis. That missing denominator is decisive. A burden expressed as a percentage cannot establish an absolute delivered cost when the underlying product values are different or unknown.
An illustrative comparison shows the issue without pretending to know the actual prices. If one product has a lower percentage add-on but a sufficiently higher base price, its combined dollar cost can still exceed the alternative. If specifications differ, even matching the currency and mass would not make the products interchangeable.
The records also omit several company-specific cost layers. Domestic onward delivery, inventory financing, quality requirements and processing needs can matter to the buyer. Their absence does not make the customs comparison useless. It defines the narrower question that the observed duty-and-charge ratios can answer.
A TOP-COUNTRY LIST CAN DISTORT THE DENOMINATOR
Canada represented 29.83% of the full tracked family's 2025 customs value, using the complete country board. That share must be calculated against all origins rather than only the leading countries shown in a compact table. Renormalizing a top-country display can make an origin look more dominant than it is in the full market represented by the data.
The family total was $6.60 billion. Canada's $1.97 billion divided by that total gives the full-panel share. The same denominator discipline is needed when comparing duties or charges. A visually tidy table of leading origins is not automatically a complete statistical universe.
For this investigation, the raw country boards reconcile to the family total, and the three named lanes have complete reported duty and charge coverage. This removes one possible source of an artificial comparison: a duty or transport numerator being divided by a customs value covering a different set of records.
WHAT THE 2025 RECORD CAN INFORM IN SEPTEMBER
The observed pattern is a reason to examine procurement assumptions, not a ready-made recommendation to change suppliers. If a sourcing model assumes that shorter distance guarantees the lowest import burden, the Canada-Japan comparison demonstrates why that assumption needs testing against the other layers.
A September 2026 quote comparison would start with matched specifications and quantities, identify the relevant customs treatment for each actual entry, and retain transport and other costs separately. The 2025 ratios can serve as a historical stress test for the structure of that model. They should not be inserted as if they were today's legal rates.
The same approach helps explain why two teams may disagree about whether a source is competitive. One may be comparing transport charges, another duty-inclusive imports, and another fully delivered production cost. Naming the boundary of the calculation can resolve an apparent factual disagreement before anyone changes a purchasing decision.
GEOGRAPHY IS ONE INPUT, NOT THE ANSWER
The annual evidence was available in the September 9, 2026 repository snapshot and carried forward from the September 6 fetch. It covers calendar 2025 entries in one defined manufacturing family. The USITC DataWeb source measures historical trade; it does not quote current suppliers or establish present eligibility under trade programs.
Calculated duties also differ from an audited record of cash duties paid. That limitation matters when translating a statistical ratio into a claim about what an individual importer ultimately settled. The defensible comparison is between the reported layers within the same annual data and coverage.
The finding is therefore both specific and transferable. Canada's lower charge ratio did not produce the lower combined duty-and-charge ratio in this family and year. Whether Canada, Japan or Mexico offered a better September purchase remained a question for matched products and actual terms. The historical record shows why the route length alone could never answer it.
Sources and evidence
Evidence period: 2025 annual same-family country lanes. The frozen evidence record lists the source files and verified hashes available September 9, 2026. Source revision: 5e4fb7726c3d40060c0151c086c903baae856cab. Later live-data updates do not alter the historical evidence in this article.
census.gov/foreign-trade/guide/sec2.html
Published 2026-09-29.