Why Did Canada Drop Its Retaliatory Tariffs? The Answer Ottawa Never Published

By Milad Ghobadibeygvand, BScN (Western University, 2014) · Published August 16, 2026 · Zeus eBikes Canada

Part two of Petitions to Fix Canada, a Zeus Media series that pairs the verified numbers behind Canada's hardest problems with ready-to-file petitions aimed at the people who legally own each lever. The series hub explains how to start a petition in Canada. Part one covered the cost of living.

1. The Question With No Published Answer

In three days, on August 19, 2026, an additional duty of 50 per cent lands on Canadian dairy, alcoholic beverages and motor vehicles entering the United States — signed July 20, published July 23, under a 1930 statute that had never before been used to impose a tariff (Proclamation 11047). What a Canadian household stands to lose here is not abstract: it is the price of a car, a block of cheese, and the job of whoever assembles or ships them. And the single most consequential Canadian decision of the past year — to withdraw most of Canada's counter-tariffs while keeping three — was made without publishing a word of the analysis behind it. This article is about that gap, and about the specific paperwork that closes it.

Set the two facts side by side. The decisions are meticulously documented: every order is registered, numbered, and published in the Canada Gazette exactly as the law requires. The reasoning is not. There is a difference between a government that tells you what it did and a government that tells you why, and Canadian law — as the next few sections show in the government's own words — requires far less of the second than most people assume.

A decision published is not a decision explained.The distinction this article runs on

A word on what this is not. It is not an argument that Canada should have kept its counter-tariffs, or dropped them, or signed something different at the CUSMA table. That judgment belongs to people with the file in front of them, and this series takes the deliberate position that a citizen's petition should not try to bind a negotiator's hands mid-negotiation. For the leverage side of that argument — what eighteen months of trade fights across nine countries actually show works — see our trade-war playbook. It is an argument that the analysis those decisions rested on — the modelling, the sectoral costing, the consultation responses already gathered and paid for — belongs to the public that funded it. Zeus sells electric bikes; a cheaper way to move is our lane, and we have written at length about why Canada is so expensive — including what it costs Canadians that two American companies clear almost every card payment in the country. This piece contains nothing for sale.

2. The 2026 Record — Every Number Sourced

As of August 16, 2026: Canada withdrew roughly $44.7 billion in counter-tariffs and kept three sectors at 25 per cent; the United States now stacks four separate tariff authorities against Canadian goods; and the CUSMA joint review ended without renewal, though the agreement stays in force to 2036. Each row below links to the instrument or report that produced it — including a Bank of Canada figure that predates the year's largest escalation.

Two rows deserve a second look, because they cut against the reflex to read all of this as collapse. Non-renewal is not expiry: the tariff schedules, rules of origin and dispute-settlement machinery of CUSMA are untouched, and the only thing non-extension changed is how often the three parties must sit down — annually instead of once. And the Supreme Court of the United States struck down the original tariff authority altogether, which is a reminder that these measures face real challenge inside the American system too. None of that makes the August 19 duties less real. It belongs in an honest ledger all the same.

3. What Was Decided, and What Was Explained

Why did Canada drop its retaliatory tariffs? The pattern only becomes visible when the decisions are lined up against their published reasoning. In every case below, the decision is fully documented and the analysis behind it is not — and in two of them the government was asked directly, by Parliament, and answered with a phrase instead of a number.

Parliament's own scrutiny of the tariff file is thinner than the stakes warrant. The Standing Committee on International Trade's only tariff-specific report of the period, adopted November 27, 2025 and presented December 8, 2025, is a single paragraph of political condemnation: no data, no cost estimate, no government response required or given. The House Standing Committee on Finance has never studied the counter-tariffs at all. The Parliamentary Budget Officer did cost them — twice — but never in a document with the word "tariff" in the title; the numbers sit inside the recurring Economic and Fiscal Outlook series, where almost nobody looking for them would think to check.

4. Which Powers Legally Owe You an Explanation

Most of Canada's sovereignty powers carry no legal duty to explain themselves. A surtax order must be tabled in Parliament within 15 sitting days, but no mechanism exists for Parliament to annul it. Sanctions orders must be tabled within five, but there is no annual report. Trade negotiations carry no statutory reporting requirement of any kind, and the rule requiring treaties to be tabled before ratification is a policy, not a law — one whose cited authority is a Throne Speech, and which a Prime Minister may waive by letter. Below is the ledger, statute by statute, quoted from the consolidated text.

The NSICOP revision power is not hypothetical, and the artifact is public. The Committee's Special Report on Foreign Interference in Canada's Democratic Processes and Institutions — submitted to the Prime Minister on March 22, 2024 and tabled June 3, 2024 — carries this line on its own title page: "(Revised version pursuant to subsection 21(5) of the NSICOP Act)." Inside, the Committee explains its own notation: "Where information could simply be removed without affecting the readability of the document, the Committee noted the removal with three asterisks (***) in the text." Section 21(5.1) worked exactly as written. A reader can see precisely where the report was cut, and never what was cut.

5. The Two Walls: Cabinet Confidence and the Minister's Office

Two structural features of Canadian law put certain records beyond the reach of every access mechanism in this article. These are not scandals. They are deliberate, longstanding features of the system, defended on real grounds. But a citizen who files without knowing where they sit will waste months, so they belong in the open.

Wall one: section 69 is an exclusion, not an exemption

Every other protection in the Access to Information Act — sections 13 through 26 — is drafted as "the head of a government institution may (or shall) refuse to disclose." That phrasing keeps the record inside the Act, which means the Information Commissioner can investigate the refusal and, since 2019, order disclosure. Section 69 uses different words entirely:

"This Part does not apply to confidences of the Queen's Privy Council for Canada…"Access to Information Act, s.69(1)

Because the Part "does not apply," the Commissioner's entire apparatus — complaint, investigation, the section 36.1 order power, Federal Court review under section 41 — never engages at all. Treasury Board's own December 2022 statutory review describes, in the government's own words, what happens instead: "if a requester complains about the application of Cabinet confidences to the Information Commissioner, the review process by the institution is for the ATIP coordinator to consult with their departmental legal services and to confirm to the Information Commissioner that the exclusion was correctly applied." The reviewer reviews a representation about the record, not the record.

The same government report records the breadth objection without dismissing it: "Public stakeholders and the Information Commissioner have singled out the exclusion of Cabinet confidences, which they believe to be overly broad. For instance, many stakeholders, including government institutions, noted that the wording in paragraph 69(1)(g) can be read to encompass any information put before a Minister as being 'related' to a Cabinet confidence, since a Minister may discuss it with colleagues."

There are real limits on the wall, and they matter. Section 69(3) sunsets the exclusion after twenty years, and applies it to discussion papers only until the related decisions are made public or four years have passed. In litigation the twin provision is harder still: Canada Evidence Act section 39 provides that on a minister's or the Clerk's written certificate, "disclosure of the information shall be refused without examination or hearing of the information by the court, person or body" — the judge is barred from even looking at the document to test the characterization.

Wall two: the minister's own office is not a "government institution"

In Canada (Information Commissioner) v. Canada (Minister of National Defence), 2011 SCC 25, the Supreme Court held that "the words in s. 4(1) of the Access to Information Act mean that the PMO and the relevant ministerial offices are not part of the 'government institution' for which they are responsible." A two-part control test can still reach departmental records physically sitting in a minister's office, but on the facts of that case none of the records at issue passed it. The practical consequence is a second, independent route by which a document can sit outside the Act — not because it is secret, but because of where it lives.

6. The Access System, Measured

If the formal route to a government explanation is an access request, the honest question is whether that route works. By the government's own measurement it is getting worse, year over year — and the institution that administers Cabinet confidences sits well adrift of the government-wide average.

Source: Treasury Board of Canada Secretariat, Access to Information and Privacy Statistical Report, 2024–25. Treasury Board itself excludes Immigration, Refugees and Citizenship Canada from this report, on the stated ground that it received 83.3 per cent of all requests and its request profile "differ[s] significantly" — an exclusion the government itself declares and labels.

Information Commissioner Caroline Maynard put the blended figure to the House Standing Committee on Government Operations and Estimates on February 5, 2026: "in 2024-2025, more than 20% of all access requests were not responded to within legislated timelines. When excluding Immigration, Refugees and Citizenship Canada… 35% of access requests were not responded to within legislated timelines." In her most recent annual report she described the machinery plainly: "The system that supports the administration of the Act remains stuck in the era of filing cabinets and typewriters in which it was born."

One more drafting detail explains a great deal of the 25.1 per cent that disclosed nothing. Section 21 — the advice-and-recommendations exemption, the one most likely to cover the analysis behind a tariff decision — contains no injury test. Sections 14, 15 and 16(1)(c) all require that disclosure "could reasonably be expected to be injurious" to something specific, and section 18 requires a showing of prejudice or material injury. Section 21 requires nothing of the kind: if a record is advice or recommendations and is under twenty years old, no harm need be shown at all. Treasury Board's own 2025 reform document concedes the consequence: the courts "have also said that factual information should generally be disclosed unless it is so tied to advice that revealing it would expose the nature of the advice. This interpretation is not reflected in the Act."

That reform document is live right now. Treasury Board launched the second statutory review of the Act on June 20, 2025; public engagement ran to June 15, 2026 and is closed, with the report to Parliament not yet published as of writing. Its published options include a public-interest override, narrowing section 21 so it cannot shield purely factual information, a statutory duty to document decisions, and a declassification regime. What is absent from the published options is any standalone proposal to make section 69 Cabinet confidences independently reviewable by the Information Commissioner — the single reform she has requested in two consecutive annual reports.

7. Two Reports That Were Promised and Cannot Be Found

Two documents in this file were publicly promised and appear not to have been delivered: the Hogue Commission's one-year progress report, and the independent review of the 2025 election protocol. Both checks were run by hand on August 16, 2026, against the government's own index pages. Both are stated at the strength the evidence supports: promised, and not where a citizen would look. Not concealed. Not refused. Because these are negatives, the method matters more than the conclusion — so here is exactly what was searched, and what came back.

7.1 The Hogue progress report

The Public Inquiry into Foreign Interference reported on January 28, 2025 with 51 recommendations across three lines of effort. Commissioner Marie-Josée Hogue found no evidence that Canada's institutions had been seriously compromised — she wrote that she had seen no evidence of "'traitors' in Parliament plotting with foreign states to act against Canada" — while calling information manipulation "an existential threat" to Canadian democracy. She judged that about half of the 51 could be implemented promptly, perhaps even before the next election. She declined to set deadlines, and asked for one thing instead:

"I have not set precise deadlines for implementing my recommendations, but I recommend the government report to Parliament within one year on its progress."Commissioner Marie-Josée Hogue, Final Report, Volume 5

One year from January 28, 2025 was January 28, 2026 — nearly seven months ago. Two of her recommendations bear directly on everything in this article: Recommendation 4 asks government to "prioritize developing a declassification system that allows the government to make certain information public where it is in the public interest and where it would not unduly prejudice national security," and Recommendation 8 asks for a whole-of-government Foreign Interference Strategy and, in her words, to "provide a public timeline for its completion."

What was searched. Both of Public Safety Canada's main public pages on foreign interference — the file's own hub and its "How Canada responds" page — were fetched directly on August 16, 2026. Both returned normally. Neither contains a single instance of the words "Hogue," "recommendation," "implementation," or "progress report." The department that owns the file does not mention the Commission's recommendations, or any status against them, anywhere on either page.

Individual actions certainly exist and are traceable: a National Counter Foreign Interference Coordinator's office now sits within Public Safety Canada, and in March 2026 the government replaced the Critical Election Incident Public Protocol with a new Cabinet directive that its own explainer says "incorporates recommendations from the Public Inquiry into Foreign Interference." What does not exist, anywhere a citizen would look, is the thing the Commissioner actually asked for: a single report showing where all 51 stand.

7.2 The 2025 election review

After each of the last two general elections, an independent reviewer assessed the election-protection protocol and the report was published: James Judd, a former CSIS Director, after the 2019 election; Morris Rosenberg, a former deputy minister of Justice, Health and Foreign Affairs, after 2021. The pattern was established, twice.

What was searched. The government's own Reports and publications index for democratic institutions was fetched on August 16, 2026 and every report link on it extracted. It lists the Judd assessment, the Rosenberg assessment, and a "Retrospective Report on the 45th General Election" — which is the Panel's own account of its work, a different document. There is no assessment of the 2025 protocol on the index.

The government's own retrospective report says, in the future tense, what is supposed to happen — and the fetched text reads:

That commitment was made by the government, in writing, in its own published report. The election was April 28, 2025. More than fifteen months later there is no named reviewer and no published assessment. And there is a second detail worth recording without over-reading it: the repealed protocol's section 9 required an "independent" post-election report; the replacement directive adopted in March 2026 requires only that "the Panel will produce a report." The word independent is not in the new text. Whether that change affects an obligation created under the old directive is a question this article cannot answer from any public source — which is itself a reason to ask it formally.

8. The Instrument Ladder — What Actually Binds

Before drafting anything, know what each instrument can and cannot make a government do. The uncomfortable finding is that the loudest instrument in Parliament has the weakest staying power, and the two that genuinely bind are the quiet administrative ones almost nobody reaches for.

A petition compels a response, not an answer

Standing Order 36(8) is unambiguous about the clock: every petition "shall forthwith be transmitted to the ministry, which shall, within 45 days, respond." If no response arrives, "the matter of the failure of the ministry to respond shall be deemed referred to the appropriate standing committee," which must meet within five sitting days. That is a real, automatic mechanism, and it is why a certified petition is worth filing.

What it does not do is set any standard for the content. The House's own Procedure and Practice, 4th edition, describes a tabling ritual built so that content is never even named (ch. 22, Government Response to Petitions): a government representative "informs the House that responses to a certain number of petitions are being tabled; no reference is made to specific petitions or the content of the responses." And on the closely related instrument of written questions, four Speakers across three governments have said the same thing:

"There are no provisions in the rules for the Speaker to review Government responses to questions."Speaker Andrew Scheer, April 3, 2012

In the same ruling Scheer quoted the long-standing authority that Speakers have never found a prima facie breach of privilege over the accuracy of a tabled document — and then added, in his own words, that to "accurate" he "might add the word 'complete.'" He also held that "it is acceptable for the government, in responding to a written question, to indicate to the House that it cannot supply an answer." Speaker Parent said in 1999 that "the Chair cannot comment on the quality or the factual content of the answers"; Speaker Milliken in 2005 that any dispute about accuracy "is a matter of debate… not something upon which the Speaker is permitted to pass judgment"; Speaker Scheer again in 2013 that "Speakers do not judge the quality of Government responses." Members have raised this as a question of privilege repeatedly, and it has never once been found to be a prima facie breach.

There is also a trap in the written-question rules worth knowing: the 45-day clock under Standing Order 39(5)(a) exists only if the member requests it when filing. Absent that request, there is no deadline at all.

The power the House calls absolute — and what happened when it was used

Procedure and Practice is emphatic that Parliament's power to demand documents has no ceiling (ch. 20, Committee Powers): "The Standing Orders place no restriction on the power to order the production of papers and records. The result is a broad, absolute power… The House has never set a limit on its power to order the production of papers and records." Speaker Milliken's ruling of April 27, 2010 remains the strongest statement of it, including on the ground governments most often invoke:

"Accepting an unconditional authority of the executive to censor the information provided to Parliament would in fact jeopardize the very separation of powers… No exceptions are made for any category of Government documents, even those related to national security."Speaker Peter Milliken, April 27, 2010

Then there is the record of what that power actually produced.

Three modern tests of an "absolute" power, and in none of them did the order itself do the work. Once it produced documents after eighteen months of negotiation, in a different Parliament from the one that issued it. Twice it was extinguished by the calendar before it was ever enforced. An order for the production of papers does not survive prorogation or dissolution, and nothing carries into the next Parliament — which is why the instruments in the next subsection matter.

The two instruments that survive an election

Since 2019 the Information Commissioner has had the power to make binding orders, and the mechanism is genuinely different in kind. Under section 36.1 of the Access to Information Act, an order takes effect automatically "on the 31st business day" after the institution receives the Commissioner's report — or the 41st where a third party is involved — unless the institution itself applies to Federal Court to stop it. The burden of litigating sits with the government, not the citizen. Costs lean the same way: section 53(2) provides that where an application "has raised an important new principle," the Court "shall order that costs be awarded to the applicant even if the applicant has not been successful in the result."

Above that sits Federal Court review under sections 41 and 44, heard — per section 44.1 — "as a new proceeding," meaning the Court decides the question afresh rather than deferring to the institution's reasoning, and may order disclosure outright. Courts are not dissolved by an election.

9. The Six Petitions, Ready to File

Each petition below follows the House of Commons' published form: an addressee the rules permit, a preamble in which every factual clause is verifiable, and a prayer asking for concrete action the addressee unambiguously owns. Each is under the 250-word limit for e-petitions. They are free to copy, adapt and file at petitions.ourcommons.ca.

Petition 1 — Publish the analysis behind the tariffs Canada kept and the ones it dropped

Why this one first. This is the decision at the centre of the file, and the gap is documented rather than alleged. The Regulatory Impact Analysis Statement is the instrument Canadian regulatory practice uses to set out the costs and benefits of an order; across all four counter-tariff orders it contains qualitative reasoning and no dollar figures. Finance Canada separately ran a public consultation on the product lists, closed it with a promise the feedback would "help inform decisions," and never published what it heard. The Parliamentary Budget Officer's costing exists but sits inside a general fiscal outlook rather than any tariff-titled report. Everything requested below is a year old or older.

Where it goes and how hard it is. Federal e-petition. The surtax power under Customs Tariff section 53 is exercised by the Governor in Council on the joint recommendation of Finance and Foreign Affairs, so this cannot be deflected to the provinces. Difficulty: medium. Asks 2 and 4 are close to housekeeping. Ask 1 is the real one, and the predictable answer is that the analysis went to Cabinet — which is why the text says "prepared in support of," reaching departmental analysis rather than Cabinet documents (see section 5). Pair it with an access request to Finance Canada for the same departmental records; that is the copy that survives an election.

Petition 2 — Report the remission numbers Parliament has already asked for three times

Why it matters. Remission is where tariff policy meets individual firms: an order under Customs Tariff section 115 relieves a company of duties it would otherwise pay. Two published grounds govern it. But section 115 contains no reporting duty whatsoever, and outcomes surface only as lists of tariff codes in Gazette schedule amendments. Asked how many applications it had received, the government wrote "a very high volume of requests" — and then reused that exact phrase in filing after filing, at least five times across more than a year. Members of Parliament have put the question on the Order Paper three times. Nothing here requires naming a single company.

Where it goes and how hard it is. Federal e-petition; remission is exclusively federal. Difficulty: low. Every figure requested is administrative data the department must already hold to run the program. The anticipated deflection is commercial confidentiality, which the preamble removes in advance by asking only for aggregates. Ask 4 is deliberate: a government that declines is forced to say so in writing, on the record.

Petition 3 — Table the Hogue progress report, recommendation by recommendation

Why it matters. The Commissioner asked for exactly this, in her own report, and set the interval herself: "report to Parliament within one year on its progress." That year elapsed in January 2026. As set out in section 7, no consolidated status report could be found on either of Public Safety Canada's public pages on foreign interference as of August 16, 2026. The ask below is deliberately narrow — a status table, not an argument about whether any given recommendation is right.

Where it goes and how hard it is. Federal e-petition; the subject matter is squarely federal. Difficulty: low to medium. The anticipated deflection is national security, which the preamble neutralizes: a four-state status label for each numbered recommendation reveals nothing classified. If a consolidated report has in fact been tabled, this petition is answered simply by pointing to it — which is a good outcome, and a reason to prefer this framing to an accusation.

Petition 4 — Publish the 2025 election review, and put "independent" back in the directive

Why it matters. Independent assessments were published after the 2019 and 2021 elections. For the 2025 election the government's own retrospective report states that a similar independent review "will be undertaken" and "will be made public." Fifteen months on, the government's own reports index lists no such assessment. Separately, the March 2026 directive that replaced the old protocol requires only that "the Panel will produce a report" — the word independent is absent from the new text.

Where it goes and how hard it is. Federal e-petition. A Cabinet directive is amended by Cabinet, so ask 3 is within the addressee's own authority. Difficulty: low. The government has already promised asks 1 and 2 in writing in its own published report, which is what makes a non-answer conspicuous. Ask 4 is the durable one: it converts a promise into a deadline.

Petition 5 — Publish Canada's objectives once a year, without publishing its positions

Why it matters, and why it is drawn narrowly. The CUSMA joint review of July 1, 2026 ended without renewal, and the agreement now faces an annual review cycle to 2036. Canada has published two rounds of "what we heard" from consultations — records of what stakeholders told the government rather than of what the government is seeking. The United States published nothing either, for a structural reason: the statute that required a published objectives document, Trade Promotion Authority, lapsed in 2021. This petition asks for objectives — the ends Canada is pursuing — and expressly not for positions, mandates, fallback levels or instructions. That distinction is the whole design: a negotiating team that must state what it is trying to achieve keeps every bit of its tactical freedom, and gains a public record to point at.

Where it goes and how hard it is. Federal e-petition; the treaty power and trade negotiations are federal. Difficulty: medium to high — this is the ask that touches an active file, and the honest expectation is a response citing the confidentiality of negotiations. That is why the preamble concedes the point in advance and confines the request to objectives, and why ask 2 is retrospective: reporting an outcome after a review has concluded costs Canada nothing at the table.

Petition 6 — Publish the withholding ledger: how often secrecy is claimed, and under which section

Why it matters. This is the only one of the six aimed at the machinery rather than a single file, and it would still matter after every other question here is answered. Section 69 is an exclusion, so the Information Commissioner cannot review it. Section 21 has no injury test, so nothing need be shown to invoke it. Treasury Board's own review recorded stakeholders and government institutions alike calling paragraph 69(1)(g) "overly broad." The Commissioner has asked for section 69 reviewability in two consecutive annual reports, and it does not appear among the published options of the review that closed in June 2026. The ask below is for counts, not records — no national-security objection can attach to a number.

Where it goes and how hard it is. Federal e-petition; the Act is federal. Difficulty: low for asks 1, 2 and 4; high for ask 3, which invites the government to state a position it has so far not stated. That is the point of framing it as "state its position, with reasons": a government is free to say no, and saying no in writing is itself an answer the public can weigh. Ask 4 is time-critical — the review report is the vehicle through which any of this changes.

10. How to File One in an Afternoon

Filing an e-petition takes four things: a text under 250 words, five to ten supporters to validate it, a Member of Parliament willing to sponsor it (30 days to decide), and 500 signatures to reach presentation. Presentation starts the 45-day response clock under Standing Order 36(8). The full mechanics are in the series hub, how to start a petition in Canada.

Two pieces of context from the hub's own dataset of every e-petition of the modern era are worth carrying into this: the median e-petition collects 791 signatures, and roughly 28 per cent never reach 500 at all. Five to ten committed people is not a formality — it is most of the work.

Live demand for this subject matter is not theoretical. As of August 16, 2026, e-petition e-7531, sponsored by Elizabeth May and addressing the conduct of the United States Ambassador to Canada, had passed 185,000 signatures and was still climbing; it remains open until November 18, 2026. We report the scale as a measurement of public engagement with the sovereignty file. The allegations it makes are ones this article has not investigated and takes no position on; it is not one of the six petitions above, and nothing here asks any reader to sign or not sign it.

11. Questions Readers Ask

Why did Canada drop its retaliatory tariffs?

The Government of Canada has never published a cost-benefit analysis of the decision. On September 1, 2025 it removed counter-tariffs covering roughly $44.7 billion in American goods, and the Regulatory Impact Analysis Statement's justification for doing so reduces to one claim: that the move reaffirms and helps to preserve the benefits of CUSMA's duty-free regime. Surtaxes of 25 per cent on steel, aluminum and motor vehicles were kept, with no stated end date. The dollar figures those statements do carry measure trade volume, never the cost or benefit of the choice itself, and no monetized comparison of keeping versus dropping them has been published by Finance Canada or by any parliamentary committee.

Which Canadian tariffs on the United States are still in force in 2026?

Three surtaxes remain, each at 25 per cent: steel and aluminum under SOR/2025-95, and motor vehicles under SOR/2025-118. Both orders are open-ended, framed as remaining until the United States eliminates its tariffs. Everything else, including the original $30.3 billion list under the United States Surtax Order (2025-1), SOR/2025-66, and a further $14.4 billion of goods, was removed effective September 1, 2025 by SOR/2025-181.

Was CUSMA renewed at the July 2026 joint review?

No. United States Trade Representative Jamieson Greer stated that the United States did not agree to renew the USMCA in its current form, and that as a result the agreement is not renewed. Non-renewal is not termination. Under Article 34.7 the agreement stays fully in force until July 1, 2036, and the practical consequence of non-extension is that the Commission must now hold a joint review every year for the remainder of the term.

What are the new American tariffs taking effect on August 19, 2026?

Proclamations 11046, 11047 and 11048, signed July 20, 2026 and published July 23, 2026, impose an additional ad valorem duty of 50 per cent on certain Canadian alcoholic beverages, dairy and motor vehicles, effective 12:01 a.m. eastern time on August 19, 2026. They are made under section 338 of the Tariff Act of 1930, a 1930 authority that had not previously been used to impose tariffs.

Does a petition legally force the government to answer?

It forces a response, not an answer. Standing Order 36(8) requires the ministry to respond within 45 days, and failure to respond is automatically referred to committee. But there is no enforceable standard for what a response must contain. Speaker Scheer ruled in 2012 that there are no provisions in the rules for the Speaker to review government responses, added the word complete to that doctrine, and held it is acceptable for the government to indicate that it cannot supply an answer.

What is the strongest way to force the release of a government document in Canada?

On the record, the Access to Information track. An order by the Information Commissioner under section 36.1 takes legal effect automatically on the 31st or 41st business day unless the institution itself applies to Federal Court to stop it, and a Federal Court review under section 41 is heard as a new proceeding and can order disclosure outright. The House of Commons power to order papers is described as absolute, but in its three modern tests the order itself never compelled compliance: once documents followed after eighteen months of negotiation, and twice the order expired when Parliament was prorogued or dissolved.

Why can't the Information Commissioner review Cabinet confidences?

Because section 69 is drafted as an exclusion rather than an exemption. Every other protection in the Act says the head may or shall refuse to disclose. Section 69 says this Part does not apply, so the complaint, investigation and order powers never engage. Treasury Board's own 2022 review describes the resulting process as the institution's legal services confirming to the Commissioner that the exclusion was correctly applied, and records the criticism that paragraph 69(1)(g) is overly broad.

How well is Canada's access to information system actually working?

By the government's own measurement, worse each year. Treasury Board reports 64.5 per cent of requests closed within legislated timelines in 2024-25, down from 81.9 per cent in 2015-16, with 25.1 per cent of closed requests disclosing no information at all. Information Commissioner Caroline Maynard told a Commons committee in February 2026 that excluding one department, 35 per cent of requests missed their legislated timelines.

12. Limitations of This Analysis

1. Two central findings are negatives. Sections 7.1 and 7.2 establish that the Hogue progress report and the 2025 election assessment are not published on the owning department's pages or the government's own reports index, checked by hand on August 16, 2026. That is not proof they do not exist. A tabled document can sit in the parliamentary record without appearing on a departmental website, and an exhaustive Hansard and Order Paper sweep was not completed — our research budget was exhausted first. Both are reported as promised-and-not-found, not as concealed.
2. "No published cost-benefit analysis" is likewise a searched absence. It was checked against Finance Canada publications, the Regulatory Impact Analysis Statements of all four counter-tariff orders, the House Finance Committee's work list, and the Parliamentary Budget Officer's publication list. It remains possible that an analysis exists in a venue not searched.
3. Case law is cited at one remove in two places. CanLII was unreachable throughout this research (bot-detection, HTTP 403 on every attempt). The Ethyl and Babcock holdings are quoted from the Office of the Information Commissioner's own published investigator's guide rather than from the judgments — a strong regulator source, but secondary. The 2011 SCC 25 holding was retrieved from the Supreme Court's own text.
4. The Bank of Canada tariff figure is stale by the report's own admission. The 5.0 per cent average effective rate carries a stated data cutoff of July 10, 2026, before the 20 July proclamations. It is quoted as published and labelled accordingly; we have not adjusted it.
5. A live discrepancy between two government sources is unresolved. Public Safety Canada states the Foreign Influence Transparency and Accountability Act came into force August 4, 2026. The Justice Laws consolidated text, modified August 6, 2026, still marks every section "not in force." The likeliest explanation is a consolidation lag rather than a legal non-effect — but that is inference, not verification, and it is flagged rather than smoothed over.
6. Hogue's "about half" figure is her spoken estimate. It is quoted from contemporaneous reporting of her remarks rather than from a count in the report. Our own tally of the report's own markers produced a different number, so we publish her figure and discard ours.
7. The petition drafts are unfiled models, not legal advice. They follow the published admissibility rules and the 250-word limit, but the House's clerks determine admissibility, and wording may need adjusting. Nothing here is a substitute for advice from counsel or from a sponsoring Member's office.
8. Difficulty ratings are editorial judgment. The "low / medium / high" labels on each petition are our assessment of how easily each ask can be satisfied, not a measured probability of success.
9. Two claims were narrowed after review, and one could not be closed. An earlier draft said the Regulatory Impact Analysis Statements contained no dollar figures; they do — the figures measure trade volume, not the cost or benefit of the decision, and the claim now says so. An earlier draft also said the Winnipeg laboratory documents were never delivered; the primary record supports only that the order was never complied with. Access was reportedly negotiated later through a separate all-party mechanism, which we were unable to verify from a primary source before publication and therefore do not assert.
10. This is a snapshot of a fast-moving file. Duties take effect on August 19, 2026, the CUSMA review cycle is now annual, and the access-to-information review report is pending. Figures are current to August 16, 2026 and will need re-verification before reuse.

13. The Bottom Line

Canada's sovereignty file in 2026 is not short of decisions. It is short of published reasons. Roughly $44.7 billion in retaliatory tariffs were withdrawn and three sectors kept, with one sentence of rationale and no dollar figures in the document that exists to carry them. A trade agreement went through a joint review that produced no renewal, with neither party publishing what it was trying to achieve. A commission of inquiry made 51 recommendations and asked for a progress report within a year; the year passed. A government promised in writing that an independent review of the last election "will be made public"; fifteen months passed.

None of that requires a theory of bad faith, and this article advances none. The far more mundane explanation is written into the statutes quoted in section 4: Canadian law is built to require disclosure that a decision was made, and almost never to require disclosure of why. A surtax order is tabled but cannot be annulled. A sanctions regime generates no report until it ends. Trade negotiations carry no reporting duty at all. The rule requiring treaties to be tabled before ratification is a policy resting on a Throne Speech, waivable by a letter. Silence is not the exception in this system. It is the default the drafting produces.

Canadian law requires governments to disclose that they acted. It rarely requires them to say why.The finding this article rests on

What follows from that is not despair, and it is not a demand that negotiators show their hand. It is narrower and more useful. The analysis is already written. The consultation submissions are already collected. The recommendations are already numbered. The status of each is already known inside government. Every one of the six petitions in section 9 asks only that something already produced with public money be produced to the public — and each is paired with an access request that keeps working after Parliament rises.

The analysis is already written. The consultation submissions are already collected. The status of each recommendation is already known inside government.What the six petitions actually ask for

The instruments are real, and their limits are real. A petition buys 45 days and a dated entry in the public record; it does not buy a complete answer, and four Speakers have said so plainly. The House's power to demand documents is called absolute, and it has been extinguished twice this decade by nothing more dramatic than a calendar. The two instruments that actually bind — a Commissioner's order that self-executes in 31 days, and a Federal Court review heard afresh — are the quiet administrative ones almost nobody reaches for. Now you know which is which.

The 50 per cent duties land on August 19, 2026. Whatever Canada decides to do about them, the decision will be published. Whether the reasoning is published too is, for once, a question with a filing number.

More from this series: the hub, How to Start a Petition in Canada That the Government Must Answer · Arm one, How to Save Money in Canada (2026) · on how vague law creates accountability gaps elsewhere, Generally Accepted Practice · on payment sovereignty, What If Canada Owned Its Own Money?

References

1. Proclamation 11047 of July 20, 2026, "Imposing Additional Duties To Offset Canadian Discrimination Against the Commerce of the United States With Respect to Dairy," 91 FR 46653, published July 23, 2026. Federal Register. Companion proclamations: 11046 (alcoholic beverages); 11048 (motor vehicles). Retrieved and verified against the Federal Register document API, August 16, 2026.
2. United States Surtax Order (2025-1), SOR/2025-66 — the original $30.3 billion countermeasure list, made under Customs Tariff s. 53(2). laws-lois.justice.gc.ca.
3. Order Amending and Repealing Certain Orders Made Under the Customs Tariff (United States Surtax), SOR/2025-181, P.C. 2025-637, registered August 29, 2025, Canada Gazette Part II, Vol. 159, No. 19. gazette.gc.ca.
4. United States Surtax Order (Steel and Aluminum 2025), SOR/2025-95, made under Customs Tariff s. 53(2), as amended by SOR/2025-181. Consolidation: laws-lois.justice.gc.ca; Regulatory Impact Analysis Statement: Canada Gazette.
5. United States Surtax Order (Motor Vehicles 2025), SOR/2025-118, made under Customs Tariff s. 53(2). Vehicles entitled to the United States Tariff are subject to a content-based reduction under the CUSMA Rules of Origin Regulations. Consolidation: laws-lois.justice.gc.ca; Regulatory Impact Analysis Statement: Canada Gazette.
6. Global Affairs Canada, "Statement by Minister LeBlanc following trilateral CUSMA joint review meeting," July 1, 2026. canada.ca.
7. Office of the United States Trade Representative, "Ambassador Greer Issues Statement on the USMCA Joint Review," July 2026. ustr.gov.
8. CUSMA, Chapter 34 (Final Provisions), Article 34.7. international.gc.ca. The published consolidation notes it is "for information purposes only"; the CUSMA Secretariat holds the authoritative text.
9. Bank of Canada, Monetary Policy Report, July 2026, tariff assumptions (data cutoff July 10, 2026). bankofcanada.ca.
10. Office of the Parliamentary Budget Officer, Economic and Fiscal Outlook, September 2025. pbo-dpb.ca.
11. Customs Tariff, S.C. 1997, c. 36, s. 53 and s. 115.
12. Special Economic Measures Act, S.C. 1992, c. 17, ss. 4 and 7. laws-lois.justice.gc.ca.
13. Investment Canada Act, R.S.C. 1985, c. 28 (1st Supp.), ss. 25.4, 25.6, 25.9, 36, 38.1. laws-lois.justice.gc.ca.
14. Department of Foreign Affairs, Trade and Development Act, S.C. 2013, c. 33, s. 174. laws-lois.justice.gc.ca.
15. Global Affairs Canada, Policy on Tabling of Treaties in Parliament (2008). treaty-accord.gc.ca.
16. National Security and Intelligence Committee of Parliamentarians Act, S.C. 2017, c. 15, s. 21 (incl. ss. 21(5) and 21(5.1)). laws-lois.justice.gc.ca.
17. NSICOP, Special Report on Foreign Interference in Canada's Democratic Processes and Institutions (revised version pursuant to s. 21(5)), submitted March 22, 2024, tabled June 3, 2024. nsicop-cpsnr.ca (PDF).
18. Access to Information Act, R.S.C. 1985, c. A-1: s. 21, s. 69, s. 69.1, and ss. 36.1, 41, 44, 44.1, 53(2), 93. Full text.
19. Canada Evidence Act, R.S.C. 1985, c. C-5, s. 39 and s. 38.13.
20. Canada (Information Commissioner) v. Canada (Minister of National Defence), 2011 SCC 25.
21. Learning Resources, Inc. v. Trump, Supreme Court of the United States, decided February 20, 2026 (October Term 2025). Case identity, court and decision date verified via the CourtListener opinion database, August 16, 2026; the slip opinion text itself was not retrievable at time of writing, so the holding is stated in summary rather than quoted.
22. Treasury Board of Canada Secretariat, Access to Information Review: Report to Parliament, December 2022. canada.ca.
23. Treasury Board of Canada Secretariat, Access to Information and Privacy Statistical Report, 2024–2025. canada.ca.
24. Treasury Board of Canada Secretariat, 2025 Review of the Access to Information Act: Policy Approaches. canada.ca.
25. Office of the Information Commissioner of Canada, Annual Report 2025–2026. oic-ci.gc.ca. Also Annual Report 2024–2025. oic-ci.gc.ca.
26. Office of the Information Commissioner of Canada, "Key messages — appearance before the Standing Committee on Government Operations and Estimates, February 5, 2026." oic-ci.gc.ca.
27. House of Commons, Standing Orders, S.O. 36(8), 39(5) and 108(1)(a). ourcommons.ca.
28. House of Commons, Procedure and Practice, 4th ed. (2025): ch. 3.4, Rights of the House as a Collectivity · ch. 20.5, Committee Powers · ch. 22, Government Response to Petitions.
29. Speakers' rulings, from the House's published decisions: Speaker Peter Milliken, April 27, 2010 (production of documents) and February 8, 2005; Speaker Andrew Scheer, April 3, 2012 (insufficient response to a written question); Speaker Andrew Scheer, January 29, 2013 (relevance of a government response); Speaker Gilbert Parent, February 8, 1999. Index.
30. Public Inquiry into Foreign Interference in Federal Electoral Processes and Democratic Institutions, Final Report, January 28, 2025, Volumes 1 and 5. foreigninterferencecommission.ca. Catalogue CP32-169/2-2025E-PDF.
31. Privy Council Office / Democratic Institutions, Retrospective Report on the 45th General Election. canada.ca (URL spelling as published). Reports index: canada.ca. Both fetched and checked August 16, 2026.
32. Democratic Institutions, Cabinet Directive on the Coordinated Response to Threats to Elections, March 2026. canada.ca.
33. Public Safety Canada, "Government of Canada establishes foreign influence registry," August 4, 2026. canada.ca. Foreign-interference hub and "How Canada responds" pages fetched and checked August 16, 2026.
34. Foreign Influence Transparency and Accountability Act, S.C. 2024, c. 16 (Part 4), ss. 28–32. laws-lois.justice.gc.ca. Regulations: SOR/2026-152, Canada Gazette Part II, July 1, 2026. gazette.gc.ca.
35. Parliament of Canada Act, R.S.C. 1985, c. P-1, ss. 79.4, 79.41, 79.42. laws-lois.justice.gc.ca.
36. House of Commons, e-petition e-7531 (sponsor: Elizabeth May), signature count and status as retrieved August 16, 2026. ourcommons.ca.
37. Research files behind this article: .claude/research/canada-us-trade-tariffs-2026-08-15.md, disclosure-machinery-2026-08-15.md, foreign-interference-file-2026-08-15.md, and the demand harvest at sovereignty-demand-2026-08-15/. Available on request.