Pre-registered prediction P1Open
"No deal before the law" (a test of the foresight thought experiment)
This test belongs to the thought experiment about US intent: it states in advance what each reading expects, so the result can count for or against any of them, including the foresight reading.
- Registered
- Window
- No fixed dates
A = the later of two dates: the Royal Assent of Bill C-39 and the Royal Assent of the bill containing the Productivity Mega Deduction; window A to A+120 days - Status
- Open: The window has not closed; nothing has settled it yet.
What counts
The claim: the US avoids a real trade settlement with Canada until Canada's investment-attraction package becomes law. Soon after it does, a deal becomes possible, with no convincing new reason given.
The start date, which we call A: the package (reported to cut the marginal effective tax rate on new business investment from about 13% to about 6.4%) comes in two laws, Bill C-39 (permitting, no tax measures) and the bill containing the Productivity Mega Deduction (the tax cut). A is the later of their two Royal Assents. The window runs from A to A + 120 days. The "Tariff-action test around the new laws" uses the same start date. Also logged if it happens: any enactment of a capital-gains reinvestment rollover (a Conservative campaign proposal, not government policy, so it may never be part of the package).
What counts as a hit: within the window, the US publicly signals a deal is achievable, lowers or suspends tariffs on Canada, or a framework deal is announced; and the stated reason for the sudden shift is vague or out of proportion to anything Canada conceded. Extra support (not required): the deal includes energy, critical-mineral, investment or ownership terms. A hit supports the foresight reading in its minimal form. If a deal is announced, we record whether it was closed or only announced (for example, a framework with no tariff relief).
What counts against: this test is read in one direction only. Pressure that continues after A (new grievances, new tariffs on other goods, talks broken off), with raw goods still spared, does not count against the foresight reading: under that reading nobody has to ease on cue, and whoever applied the pressure may keep it on for other goals. It is no evidence either way, at most a slight tilt toward the plain self-interest reading. Only sustained escalation after A that damages the southbound flow (resources tariffed, southbound volumes cut) counts against the foresight reading; it favours the rival readings in which the US aims to weaken Canada (annexation or fragmentation; see the "Annexation rival" test). A deal or a collapse whose timing has nothing to do with the laws points to the accident reading or the plain self-interest reading (see the "No plan" test).
Scored together with the "Tariff-action test around the new laws": a deal inside the window with a vague reason, plus that test showing a drop in US restrictive actions after A, count more strongly for the foresight reading. Sustained escalation that damages the southbound flow after both laws pass counts against the foresight reading and its strongest version (a worked-out plan), and for the rival readings.
What each reading expects
| Reading | Expects | If seen |
|---|---|---|
| I0a Accident Tested hypothesis (intent) | Timing unrelated to the laws (see the "No plan" test) | Counts for it |
| I0b Plain self-interest; Canada's response not considered Tested hypothesis (intent) | Pressure simply continues, with raw goods spared: no evidence either way, at most a slight tilt toward this reading | No change |
| I2 Foreseen and wanted (this level or higher) Tested hypothesis (intent) | A deal or easing within 120 days of A, with a vague stated reason | Counts for it |
| R-weaken Weaken-Canada rivals (annexation or fragmentation) Rival reading | Sustained escalation after A that damages the southbound flow | Counts for it |
Observations
- Canadian counter-tariffs take effect on C$27.6B of US goods (about US$20B), matching the US Section 338 tariffs
US signs import bans on Canadian goods, effective 2026-09-29, the same day Canada's retaliation takes effect.
- Productivity Mega Deduction announced at the Canada Investment Summit: permanent immediate expensing covering more than 65% of business assets
Productivity Mega Deduction announced (draft legislative proposals). One of the two laws that set the start date A.
- Bill C-39, the Building Canada Strong Act, introduced by Steven MacKinnon
Bill C-39 (Building Canada Strong Act) introduced. One of the two laws that set the start date A.
- Greer on CNBC: 'no urgency' for a deal with Canada; 'We're still getting what we need from them in terms of oil, gas, potash'
USTR Greer: "no urgency" on a Canada deal; "we're still getting what we need… oil, gas, potash".
- Bill C-39 at second reading in the House.
C-39 second-reading debate held 2026-09-23 to 09-25; no vote as of 09-27; the government moved to send it to a special committee. The Mega Deduction is still draft proposals with no tabling date (likely Budget 2026, in November; unverified). Start date A (the later Royal Assent) not reached. Much of C-39 Part 1 also needs orders in council after assent, so legal and in-force dates will be recorded separately. Confidence: medium.
- Trump on resuming trade talks with Canada before the midterms: 'I don't know... I'm in no rush'; says Canada charged 'tariffs that nobody knew about'
Before A: Trump says he is 'in no rush' to resume talks and gives no timing. Consistent with no deal before the laws, but a statement before A does not score. Logged, does not count.
- Trump signs a proclamation changing the list of Canadian goods under the 50% Section 338 tariffs, effective 30 October: 22 tariff lines out (thin hardwood plywood, some paper and plastic packaging, machinery), 100 in (carpets, valves, waste, cocoa powder, thin MDF and other goods)
Before A: a mixed change to the list of Canadian goods under the Section 338 tariffs (some goods removed, others added), signed 2026-10-23, effective 2026-10-30; no change to the 50% rate is reported. A partial change of scope, neither a deal nor a general easing, and before the start date: logged, does not count.
Corrections
Adds two points from the prediction as registered on 26 September 2026: the capital-gains rollover is a Conservative campaign proposal, not government policy; and an announced deal is recorded as closed or only announced.
- What counts
- was
The claim: the US avoids a real trade settlement with Canada until Canada's investment-attraction package becomes law. Soon after it does, a deal becomes possible, with no convincing new reason given. The start date, which we call A: the package (reported to cut the marginal effective tax rate on new business investment from about 13% to about 6.4%) comes in two laws, Bill C-39 (permitting, no tax measures) and the bill containing the Productivity Mega Deduction (the tax cut). A is the later of their two Royal Assents. The window runs from A to A + 120 days. The "Tariff-action test around the new laws" uses the same start date. Also logged if it happens: any enactment of a capital-gains reinvestment rollover. What counts as a hit: within the window, the US publicly signals a deal is achievable, lowers or suspends tariffs on Canada, or a framework deal is announced; and the stated reason for the sudden shift is vague or out of proportion to anything Canada conceded. Extra support (not required): the deal includes energy, critical-mineral, investment or ownership terms. A hit supports the foresight reading in its minimal form. What counts against: this test is read in one direction only. Pressure that continues after A (new grievances, new tariffs on other goods, talks broken off), with raw goods still spared, does not count against the foresight reading: under that reading nobody has to ease on cue, and whoever applied the pressure may keep it on for other goals. It is no evidence either way, at most a slight tilt toward the plain self-interest reading. Only sustained escalation after A that damages the southbound flow (resources tariffed, southbound volumes cut) counts against the foresight reading; it favours the rival readings in which the US aims to weaken Canada (annexation or fragmentation; see the "Annexation rival" test). A deal or a collapse whose timing has nothing to do with the laws points to the accident reading or the plain self-interest reading (see the "No plan" test). Scored together with the "Tariff-action test around the new laws": a deal inside the window with a vague reason, plus that test showing a drop in US restrictive actions after A, count more strongly for the foresight reading. Sustained escalation that damages the southbound flow after both laws pass counts against the foresight reading and its strongest version (a worked-out plan), and for the rival readings.nowThe claim: the US avoids a real trade settlement with Canada until Canada's investment-attraction package becomes law. Soon after it does, a deal becomes possible, with no convincing new reason given. The start date, which we call A: the package (reported to cut the marginal effective tax rate on new business investment from about 13% to about 6.4%) comes in two laws, Bill C-39 (permitting, no tax measures) and the bill containing the Productivity Mega Deduction (the tax cut). A is the later of their two Royal Assents. The window runs from A to A + 120 days. The "Tariff-action test around the new laws" uses the same start date. Also logged if it happens: any enactment of a capital-gains reinvestment rollover (a Conservative campaign proposal, not government policy, so it may never be part of the package). What counts as a hit: within the window, the US publicly signals a deal is achievable, lowers or suspends tariffs on Canada, or a framework deal is announced; and the stated reason for the sudden shift is vague or out of proportion to anything Canada conceded. Extra support (not required): the deal includes energy, critical-mineral, investment or ownership terms. A hit supports the foresight reading in its minimal form. If a deal is announced, we record whether it was closed or only announced (for example, a framework with no tariff relief). What counts against: this test is read in one direction only. Pressure that continues after A (new grievances, new tariffs on other goods, talks broken off), with raw goods still spared, does not count against the foresight reading: under that reading nobody has to ease on cue, and whoever applied the pressure may keep it on for other goals. It is no evidence either way, at most a slight tilt toward the plain self-interest reading. Only sustained escalation after A that damages the southbound flow (resources tariffed, southbound volumes cut) counts against the foresight reading; it favours the rival readings in which the US aims to weaken Canada (annexation or fragmentation; see the "Annexation rival" test). A deal or a collapse whose timing has nothing to do with the laws points to the accident reading or the plain self-interest reading (see the "No plan" test). Scored together with the "Tariff-action test around the new laws": a deal inside the window with a vague reason, plus that test showing a drop in US restrictive actions after A, count more strongly for the foresight reading. Sustained escalation that damages the southbound flow after both laws pass counts against the foresight reading and its strongest version (a worked-out plan), and for the rival readings. - What counts fr
- was
Affirmation : les États-Unis évitent un véritable règlement commercial avec le Canada jusqu'à ce que l'ensemble de mesures canadiennes pour attirer l'investissement ait force de loi. Peu après, une entente devient possible, sans nouvelle raison convaincante. La date de départ, que nous appelons A : l'ensemble de mesures (qui ferait passer, selon les comptes rendus, le taux effectif marginal d'imposition sur les nouveaux investissements des entreprises d'environ 13 % à environ 6,4 %) tient en deux lois, le projet de loi C-39 (permis, aucune mesure fiscale) et le projet de loi contenant la Mégadéduction à la productivité (la baisse d'impôt). A est la plus tardive de leurs deux sanctions royales. La fenêtre va de A à A + 120 jours. Le « Test des mesures tarifaires autour des nouvelles lois » utilise la même date de départ. Est aussi consignée, si elle survient, toute adoption d'un report d'impôt sur les gains en capital réinvestis. Ce qui compte comme un succès : dans la fenêtre, les États-Unis signalent publiquement qu'une entente est possible, abaissent ou suspendent des tarifs visant le Canada, ou un accord-cadre est annoncé; et la raison invoquée pour ce revirement soudain est vague ou sans commune mesure avec ce que le Canada a concédé. Appui supplémentaire (non requis) : l'entente comprend des conditions sur l'énergie, les minéraux critiques, l'investissement ou la propriété. Un succès appuie la lecture de la prévoyance dans sa forme minimale. Ce qui compte contre : ce test ne se lit que dans un sens. Une pression qui se poursuit après A (nouveaux griefs, nouveaux tarifs sur d'autres produits, négociations rompues), les produits bruts étant toujours épargnés, ne compte pas contre la lecture de la prévoyance : selon cette lecture, personne n'est tenu d'assouplir au bon moment, et ceux qui ont exercé la pression peuvent la maintenir pour d'autres objectifs. Ce n'est une preuve ni dans un sens ni dans l'autre, tout au plus une légère inclinaison vers la lecture du simple intérêt propre. Seule une escalade soutenue après A qui nuit au flux vers le sud (ressources frappées de tarifs, volumes vers le sud réduits) compte contre la lecture de la prévoyance; elle favorise les lectures rivales selon lesquelles les États-Unis cherchent à affaiblir le Canada (annexion ou fragmentation; voir le test « Lecture rivale de l'annexion »). Une entente ou une rupture dont le moment n'a rien à voir avec les lois pointe vers la lecture de l'accident ou celle du simple intérêt propre (voir le test « Aucun plan »). Évaluation conjointe avec le « Test des mesures tarifaires autour des nouvelles lois » : une entente dans la fenêtre avec une raison vague, et ce test montrant une baisse des mesures restrictives américaines après A, comptent ensemble plus nettement pour la lecture de la prévoyance. Une escalade soutenue qui nuit au flux vers le sud après l'adoption des deux lois compte contre la lecture de la prévoyance et sa version la plus forte (un plan élaboré), et pour les lectures rivales.nowAffirmation : les États-Unis évitent un véritable règlement commercial avec le Canada jusqu'à ce que l'ensemble de mesures canadiennes pour attirer l'investissement ait force de loi. Peu après, une entente devient possible, sans nouvelle raison convaincante. La date de départ, que nous appelons A : l'ensemble de mesures (qui ferait passer, selon les comptes rendus, le taux effectif marginal d'imposition sur les nouveaux investissements des entreprises d'environ 13 % à environ 6,4 %) tient en deux lois, le projet de loi C-39 (permis, aucune mesure fiscale) et le projet de loi contenant la Mégadéduction à la productivité (la baisse d'impôt). A est la plus tardive de leurs deux sanctions royales. La fenêtre va de A à A + 120 jours. Le « Test des mesures tarifaires autour des nouvelles lois » utilise la même date de départ. Est aussi consignée, si elle survient, toute adoption d'un report d'impôt sur les gains en capital réinvestis (une promesse électorale conservatrice, pas une politique du gouvernement; elle pourrait donc ne jamais faire partie de l'ensemble de mesures). Ce qui compte comme un succès : dans la fenêtre, les États-Unis signalent publiquement qu'une entente est possible, abaissent ou suspendent des tarifs visant le Canada, ou un accord-cadre est annoncé; et la raison invoquée pour ce revirement soudain est vague ou sans commune mesure avec ce que le Canada a concédé. Appui supplémentaire (non requis) : l'entente comprend des conditions sur l'énergie, les minéraux critiques, l'investissement ou la propriété. Un succès appuie la lecture de la prévoyance dans sa forme minimale. Si une entente est annoncée, nous consignons si elle a été conclue ou seulement annoncée (par exemple un accord-cadre sans allègement tarifaire). Ce qui compte contre : ce test ne se lit que dans un sens. Une pression qui se poursuit après A (nouveaux griefs, nouveaux tarifs sur d'autres produits, négociations rompues), les produits bruts étant toujours épargnés, ne compte pas contre la lecture de la prévoyance : selon cette lecture, personne n'est tenu d'assouplir au bon moment, et ceux qui ont exercé la pression peuvent la maintenir pour d'autres objectifs. Ce n'est une preuve ni dans un sens ni dans l'autre, tout au plus une légère inclinaison vers la lecture du simple intérêt propre. Seule une escalade soutenue après A qui nuit au flux vers le sud (ressources frappées de tarifs, volumes vers le sud réduits) compte contre la lecture de la prévoyance; elle favorise les lectures rivales selon lesquelles les États-Unis cherchent à affaiblir le Canada (annexion ou fragmentation; voir le test « Lecture rivale de l'annexion »). Une entente ou une rupture dont le moment n'a rien à voir avec les lois pointe vers la lecture de l'accident ou celle du simple intérêt propre (voir le test « Aucun plan »). Évaluation conjointe avec le « Test des mesures tarifaires autour des nouvelles lois » : une entente dans la fenêtre avec une raison vague, et ce test montrant une baisse des mesures restrictives américaines après A, comptent ensemble plus nettement pour la lecture de la prévoyance. Une escalade soutenue qui nuit au flux vers le sud après l'adoption des deux lois compte contre la lecture de la prévoyance et sa version la plus forte (un plan élaboré), et pour les lectures rivales.