Receipt — Canada's exports and the United States: the record break, and where the difference went

Basis: Statistics Canada's monthly trade record on its two published bases: the Balance of Payments by-partner series (seasonally adjusted, 1997 to the latest month) for the share and its arc, and the customs-basis domestic-export record (by product section, by province, with the detailed unwrought-precious-metals series) for where the movement sits. The 2025 United States tariff actions enter only as dated primary-document facts -- rates and effective dates read from the Federal Register texts. Every pinned value re-derives from archived source bytes recorded with content hashes; every computed figure is a named-formula Scenario over those pins. The transport companion uses Statistics Canada table 12-10-0177-01, with nominal, unadjusted customs-basis values by mode. It is separate from the balance-of-payments headline and compares the same month a year apart.

Confidence: The arithmetic is exact over pinned published figures, each re-derivable from archived source bytes recorded with content hashes; the open questions are compositional (price versus volume, gross versus value added, how durable the bullion channel proves) -- this receipt keeps them visible rather than resolving them. One published claim has been corrected: the first reading attached the record-low share and the record's fastest one-year fall to the wrong month. The correction, with the wording it replaces, is on the corrections record. No pinned value was wrong and the central reading did not change; regression coverage for the executive answer's record claims was added with the correction.

Next trigger at that reading: Each monthly international merchandise trade release, or an amendment to any pinned tariff document, whichever is published first. The June 2026 record was published August 4, 2026 and is the reading carried here; the July 2026 record follows at the issuer's next scheduled monthly release, a date no capture pins.

Next evidence at that reading: Next headline merchandise-trade release: November 4, 2026 for September, per the held October 6 issuer release. Other source families and legal documents have separate clocks.

Expiry at that reading: This Receipt describes the product and evidence dates stated above. Reassess it when the named trigger changes the evidence; a page rebuild alone is not a fresh assessment. Earlier readings and corrections remain in the product history.

Definitions

• The share is a ratio of twelve-month window sums (U.S.-bound over all-countries), not an average of monthly shares; windows are compared only to the immediately preceding twelve months.
• Balance-of-Payments exports and customs-basis domestic exports are different quantities: the first includes re-exports and balance-of-payments adjustments and is seasonally adjusted; the second excludes re-exports and is unadjusted. Levels are never compared across the two bases anywhere on the Brief.
• 'Rest of world' is all countries minus the United States inside one basis; the United Kingdom's movement is stated as a share of the rest-of-world change, a composition of changes rather than levels.
• The metals section is broader than precious metals; the unwrought gold, silver and platinum group metals series is an all-countries total, and no bilateral precious-metals flow is stated anywhere.
• Provincial exposure is read on four measures that are kept separate and never combined into a single score. Scale is a province's U.S.-bound domestic exports and that level over Statistics Canada's own Canada U.S.-bound total on the same customs table. Intensity is the province's U.S.-bound exports over that same province's all-countries exports. The shift is the difference between the two windows' intensities, in percentage points. Composition is the province's largest U.S.-bound summary section over that province's U.S.-bound total. All four are ratios of twelve-month window sums on one table and one basis, computed once before publication.
• A province's U.S. share is a share of that province's own goods exports -- not of its economy. It carries no gross-domestic-product, income or employment denominator, excludes services entirely, and says nothing about how much provincial output or employment depends on American demand. A measure built on a provincial GDP or employment denominator would answer a different question and is deliberately not attempted here.
• The ten provinces do not sum to the Canada total. Over the twelve months ending May 2026 they account for 99.97 per cent of national U.S.-bound domestic exports; the remainder is the territories, which the capture does not include, plus trade the table does not allocate to a province. Provincial shares are stated against a province's own total or against Statistics Canada's Canada total, never against a rescaled provincial sum, and the gap is published on the Brief rather than closed.
• Provinces are listed in the source table's own geography order throughout, and are never sorted by any of the four measures. The measures rank the provinces differently -- the largest exporter is not the most dependent, and neither is the province whose dependence moved most -- so any single ordering would assert a ranking the evidence does not support.
• Mode-of-transport comparisons retain exports and imports separately. Each U.S. share divides the U.S. value by the all-partner value for that same mode, flow and month; the modes are not ranked by economic importance or interpreted as tariff effects.

Limitations

• Every value is nominal: the energy fall sits beside lower oil prices across the same windows, and nothing on the Brief separates price from volume.
• Partner shares are gross shipment values, not domestic value added -- they measure where shipments go, not where income stays; value-added trade estimates are on a separate, slower official basis outside this record.
• No official source publishes a continuously updated statement of the tariff status of Canadian exports, so the Brief cannot state tariff coverage or effective rates; that absence is published as an evidence gap, and no movement is causally attributed to any named action.
• Provincial attribution follows the table's own province-of-origin convention for domestic exports. Goods one province contributes to another province's exports are counted where they cross the border, not where the value was added, so a province's figure understates its involvement in Canada-United States trade to the extent it supplies other provinces' exporters. Interprovincial and indirect supply-chain exposure is not measured anywhere on this Brief; no authoritative monthly source publishes it at this grain.
• The smaller provinces carry small export bases, and a single cargo can move a monthly value and therefore a twelve-month window. The largest provincial share move in this reading belongs to a province whose U.S.-bound exports are under one per cent of the national total; it is reported as what it is, on its own base, and is not evidence about the national trend.
• Each province's leading section is the largest by value in the current window only. A different window could name a different section for the same province, and the summary section is broader than any single commodity.
• The record low in the twelve-month share is a near-tie the record cannot separate: the windows ending March and April 2026 lie 0.004 percentage points apart, while between the July and August 2026 captures those same windows were revised by 0.014 and 0.016 percentage points -- three to four times the gap between them. April is the lower window in both captures, but by less than the issuer's own revision movement, so the Brief names a two-window low and credits no single month with the record.
• The June 2026 reading is computed from the issuer's data tables. The issuer's own June narrative release has not been captured -- the 2026-08-04 re-fetch of the archived Daily returned the May release unchanged -- so nothing in this reading quotes or relies on the issuer's June commentary.