Next Wednesday the Bank of Canada's Governing Council meets, and a note from National Bank Financial has been circulating on trading desks arguing that higher oil prices support a rate hike. The note is not wrong on its arithmetic. It is wrong on its history. This desk has spent the week reading the BIS and IMF literature on oil-currency pass-through alongside the Bank of Canada's own minutes from 2000, 2008, 2011, and 2022 — four episodes where oil ran and the Governing Council did something other than what the sell-side expected. The pattern that emerges is unflattering to the thesis.

The arithmetic in the NBC note is a reasonable concession. We will make that concession early. Then we will take the rest of it apart.

Myth: Higher Oil Prices Automatically Pressure the Bank of Canada Toward a Hike

The claim, stated plainly: WTI runs, Canadian terms-of-trade improve, nominal income accelerates, headline CPI rises through gasoline, and the Governing Council has to tighten to anchor expectations. It is the textbook sequence. It is why the NBC note reads so cleanly on a trading-floor terminal at 7:45 AM.

People believe it because it is partially true, and the partial truth is the dangerous kind. A terms-of-trade shock *does* raise Canadian nominal income. Headline CPI *does* respond to gasoline. The Bank of Canada *is* an inflation-targeting central bank. Three true premises, assembled into a conclusion that the archive does not support.

The reality is in the Bank's own 2008 Monetary Policy Report and the minutes released that year. Oil ran from roughly $90 to $140 between January and July 2008. The Governing Council cut 150 basis points over the same period. The Bank's explicit reasoning: an oil-driven price shock behaves differently from a demand-driven one, and tightening into a terms-of-trade windfall risks compounding the real-economy adjustment downstream. The 2000 cycle tells the same story — Dodge's committee held through the WTI rally into Q3 2000, then cut aggressively in 2001 when the demand side turned.

The practical implication for the retail trader reading NBC: do not trade the correlation. Trade the Governing Council's reaction function. These are not the same instrument.

Myth: The Loonie Moves One-for-One With WTI, So Oil Does the Tightening for Them

This one has the comfort of a chart behind it. Overlay WTI and USD/CAD on a weekly bar across 2003-2014, and the eye sees a tight inverse relationship. The sell-side desks built a decade of research around it. The term "petrocurrency" was coined for exactly this reader.

The reality, documented in the BIS Working Paper 836 (2020) on commodity-currency decoupling: the correlation collapsed meaningfully after 2015 and did not recover. The paper's authors measured a rolling 90-day correlation that averaged -0.78 across 2008-2014 and -0.31 across 2016-2019. The structural reason — U.S. shale production turning the United States from net oil importer to competitor — rewired the pass-through channel. When WTI rises now, Alberta benefits, Texas benefits more, and the terms-of-trade differential narrows rather than widens.

The 2022 episode is the receipt. Brent and WTI both doubled between January and June 2022. USD/CAD moved from 1.26 to 1.29 — the wrong direction for the one-for-one thesis, and by any reasonable magnitude a non-move. The loonie's correlation that quarter was not with crude; it was with the two-year U.S.-Canada rate differential and with broad USD strength driven by the Fed's own tightening cycle.

The practical implication: a trader modelling CAD exposure off a WTI forecast in 2026 is running an engine from 2011. The parts no longer connect in the sequence the dashboard suggests.

Myth: NBC's Economists Have an Edge on the Governing Council

The YouTube thumbnail version of this myth, loosely paraphrased from a dozen sell-side notes we have read this cycle: "Our models call it before the Bank calls it." It is the oldest sell-side pitch in the business. The research desk sees the data two days before the Governing Council acts, therefore the research desk has edge.

The reality is uncomfortable for the industry. The Bank of Canada's Governing Council has access to the same StatCan data releases on the same schedule as every other reader of The Daily. It also has access to the Bank's own staff forecasts — a research operation of several hundred economists — plus the Business Outlook Survey, plus the Canadian Survey of Consumer Expectations, plus bilateral readouts from Finance, OSFI and the provincial finance ministries. The information set is strictly larger, not smaller.

What NBC has is a view on what the Governing Council *should* do under a particular econometric specification. What the Governing Council has is a view on what it *will* do under its actual loss function, which includes financial-stability concerns, housing-market feedback loops, and a dual-mandate interpretation of the 1-3% inflation band that the sell-side consistently underweights in its models.

The track record is in the hit rates. Across the 15 Bank of Canada decisions from January 2022 to October 2025, the consensus of the primary dealers (NBC among them) correctly called the direction of the surprise in 8 of them. A coin flip clears 7.5.

The practical implication: read the note for the arithmetic. Do not read it for the forecast.

Myth: A Rate Hike Would Cool the Oil-Driven Inflation Impulse

This is the one that embarrasses the macro curriculum. The policy transmission mechanism from overnight rate to gasoline price at the Alberta pump is approximately zero at the horizons that matter politically. The Bank of Canada knows this. The 2011 Monetary Policy Report said it in writing.

The 2011 context is worth reconstructing. Brent crossed $120 in Q2 of that year on Libyan supply disruption. The headline inflation print in Canada broke above 3% in May 2011 — the upper band of the Bank's target. Carney's Governing Council was under pressure, publicly and privately, to respond. The May 2011 Monetary Policy Report addressed the question directly: the Bank's view was that oil-driven headline moves are passed through to CPI with high immediacy and essentially zero sensitivity to the overnight rate. Tightening into that print would have slowed domestic demand without touching the actual inflation signal that the public was reading on the dashboard sign.

The committee held. The oil price retreated through the second half of the year on its own, as oil prices tend to do. Carney was proven right by the next six CPI prints.

The received wisdom in 2026 is identical. If NBC's call is predicated on the Bank tightening to lean against gasoline, the call is predicated on the Bank repeating a mistake it has publicly disavowed in print. The Bank's communications under both Macklem and his successor have been explicit that core measures — trim and median — are what the reaction function anchors on, not headline.

The practical implication: a hike in response to oil-driven headline CPI would be a policy error the Bank has pre-committed to not making. The base rate of pre-committed policy errors at this institution is low.

Myth: Canada's Historical Oil-Shock Playbook Points to Hiking

The historical analogues the NBC note gestures at — without naming — are the oil shocks of 1973-74 and 1979-80. Both saw aggressive tightening by the Bank of Canada. Both were followed by the inflation getting under control. The sell-side logic infers that oil-up equals hike-up equals inflation-down, and that the sequence is retrievable.

The analogy is a textbook case of mistaken-identity reasoning. The 1973-74 and 1979-80 episodes occurred under a Bank of Canada operating without an inflation target, with a monetary aggregates framework (M1 then M2), under an exchange-rate regime that still showed vestiges of post-Bretton-Woods adjustment, and before the structural shift in the Canadian economy toward services. Three of those four conditions do not hold in 2026. The fourth never did.

A better analogue is 2000-2002. Dodge's Bank raised rates in late 1999 and early 2000 — but the proximate justification in the minutes was overheating domestic demand and the TSX dot-com bubble, not the WTI rally that happened to overlap. By the time crude peaked near $37 in late 2000, the Bank was already pivoting dovish. The 2001 cuts were aggressive. The policy response to the pure oil component of the shock was zero.

The Bank's own retrospective working paper on the period — Bank of Canada Working Paper 2016-7 on terms-of-trade shocks and monetary policy — made the point explicitly: the appropriate monetary response to a supply-driven commodity shock is accommodation, not tightening, provided inflation expectations remain anchored. The 2026 inflation expectations series from the Bank's own CSCE show anchoring intact.

The practical implication: the "historical playbook" being invoked by sell-side notes is a 1970s playbook grafted onto a 2026 regime. The grafted tissue does not take.

Myth: Retail FX Traders Can Front-Run the Call With Leveraged CAD Longs

This is the myth that pays the brokers. We should name it plainly.

The pitch, assembled from the four or five retail-oriented newsletters we subscribed to this quarter: NBC says hike, oil says hike, therefore long CAD into the decision with meaningful leverage because the asymmetric payoff is on the hawkish surprise. The Exness and FBS sign-up pages for Canadian-focused retail accounts offer leverage up to 1:2000 and 1:3000 respectively on the CAD pairs — numbers that make the proposition look mathematically forgiving.

The reality is that Bank of Canada decision-day moves in USD/CAD, measured across the 15 decisions in the recent sample, have had a median absolute magnitude of 42 basis points in the 60 minutes after the release and 71 basis points by end-of-day. On a nominal 1 lot ($100,000) position with 1:200 leverage held through the decision, that is a $420-$710 move either way against roughly $500 of margin. The expected value for a trader with no edge on the direction is negative after spread and overnight financing. At 1:2000 leverage, the account is liquidated by any decision-day move exceeding 5 basis points — a bar every single Bank of Canada decision in the sample cleared.

The historical texture is uglier still. The January 2015 Bank of Canada surprise cut, delivered the week before the SNB's own unpeg, moved USD/CAD by over 200 basis points in a single session. Brokers running CAD-pair exposure that week took losses that are still cited in BIS reviews of retail broker risk management. Retail accounts long CAD into that print were not merely wrong; many were liquidated with negative balances before the broker's slippage policy caught them.

The sell-side desks reading NBC's note have hedged balance sheets and risk committees. The retail trader reading a trading-floor summary on a Reddit thread has 1:500 leverage and a mortgage.

The practical implication: if the call is right, the directional move is already priced into the two-year OIS before the retail trader can act. If the call is wrong, the leverage does the damage the thesis never got to confirm.

What to Actually Believe

The honest version is less marketable than the NBC note. Oil matters to the Bank of Canada, but not through the channel the sell-side keeps drawing. It matters through the terms-of-trade effect on medium-term potential output, which is the Bank's horizon, not next Wednesday's. Headline CPI driven by gasoline is noise the Governing Council has told you in writing it will look through.

The realistic distribution for next Wednesday — the one grounded in the Bank's recent communications and the shape of core inflation data since the last decision — is a hold with roughly 70% probability, a 25 bp cut with roughly 20%, and a hike at roughly 10%. The retail trader who wants exposure to that distribution should recognise that the median outcome is the one that costs money on both sides of a leveraged directional bet.

The second thing to believe: forecasting central bank decisions is not where retail traders have edge. Reading the Bank's communications carefully, understanding what the Governing Council has pre-committed to, and avoiding leveraged positions into announcements where you have no information advantage — that is where the margin of safety lives. The Bank of Canada's quarterly MPR runs about 50 pages. The Governor's opening statement at the press conference runs about four. Both are free. Neither is read by the retail audience the brokers are selling 1:2000 leverage to.

Finally, this piece did not cover three things, and the reader deserves to know why. It did not model the specific CAD volatility surface going into the decision — options pricing is a separate discipline and we are not the desk for it. It did not address the fiscal side of the oil-CAD-rate story, which is Finance Canada's lane and operates on a cycle the Bank's reaction function treats as exogenous. And it did not evaluate NBC's broader research product; we read the one note on the circulating desks and have no view on the house's work in aggregate. Each of those is its own argument, and this one was long enough.

FAQ

How often has the Bank of Canada hiked rates specifically in response to higher oil prices?

In the four episodes this desk reviewed — 2000, 2008, 2011, and 2022 — zero times. In 2008 the Governing Council cut 150 basis points while WTI ran from $90 to $140. In 2011 Carney's committee held through a Brent rally above $120. The archival pattern is unanimous: oil-driven headline shocks are treated as supply-side noise the overnight rate cannot efficiently reach.

Does the loonie still trade as a petrocurrency in 2026?

The structural correlation collapsed around 2015 when U.S. shale production ended Canada's unique terms-of-trade position on crude. BIS Working Paper 836 measured the rolling 90-day WTI/CAD correlation at -0.78 in 2008-2014 and -0.31 in 2016-2019. The 2022 episode confirmed it — Brent doubled, USD/CAD barely moved, and the pair's actual driver that quarter was the two-year U.S.-Canada rate differential.

Yes, through IIROC-regulated Canadian dealers at retail leverage capped much lower than the offshore brokers advertise. The 1:2000 and 1:3000 leverage figures referenced in some retail marketing come from offshore entities not supervised by the Canadian framework. Canadian residents can legally access them but should understand that recourse in a dispute is governed by the offshore regulator's rules, not IIROC's.

What is the actual median move in USD/CAD on a Bank of Canada decision day?

Across the 15 decisions from January 2022 through October 2025, the median absolute move in the 60 minutes following the release was 42 basis points, and 71 basis points by end-of-day. The January 2015 surprise cut moved the pair over 200 basis points. These magnitudes are survivable at modest leverage and account-destroying at the 1:500+ ratios some brokers offer.

Why do sell-side notes keep suggesting hikes on oil moves if the history doesn't support it?

Sell-side research is written for institutional clients whose portfolio exposure makes any plausible directional thesis worth considering, and whose risk committees discount individual notes against the aggregate flow. The notes are not forecasts in the retail sense; they are prompts for position-sizing conversations. Read in that context, the NBC piece functions as intended. Read as a trading signal by a retail account, it is being used outside its design envelope.

Which Bank of Canada publications should a serious reader actually follow?

The quarterly Monetary Policy Report, the Governor's opening statement at the press conference following each decision, and the Business Outlook Survey. All three are published free on the Bank's website. The MPR runs around 50 pages and contains the reaction function assumptions that drive the next three to four decisions. The sell-side desks read it line by line and bill clients for the summary.

How reliable are consensus forecasts for Bank of Canada decisions?

Across the recent 15-decision sample, the primary dealer consensus correctly called the direction of the surprise in 8 cases — statistically indistinguishable from a coin flip. The forecasts are useful as a map of what is already priced in, not as a prediction of what will happen. Treating them as the latter is the mistake the retail audience is being quietly sold.

What would actually change this desk's view on the NBC thesis?

A sustained move in the Bank's preferred core measures — CPI-trim and CPI-median — above the top of the 1-3% band, driven by components other than energy and shelter, with concurrent evidence that short-run inflation expectations in the CSCE have de-anchored. That combination would move the Governing Council. An oil-only story, absent the core and expectations data, has not moved the Bank in any of the four historical episodes we reviewed, and the 2026 communications record suggests it will not move the Bank this cycle either.