There is a pattern that emerges every time a Canadian dollar desk note carries the phrase "rate differential" in its opening line. The ceiling holds. ING's latest framing of the loonie — anchored on the Bank of Canada's policy gap with the Federal Reserve and unresolved USMCA review risk — fits the pattern exactly, and the interesting question is not whether ING is directionally right this quarter. It is why those two variables, named together, keep reappearing in roughly the same configuration across cycles. The pattern is older than this note, older than this cycle, and durable enough that it deserves to be reconstructed as a pattern rather than a forecast.
The Rate Gap Pattern That Refuses to Mean-Revert
The pattern: whenever the BoC sits below the Fed on policy rate by more than roughly seventy-five basis points, CAD strength against USD becomes a function of risk appetite alone, not of Canadian fundamentals. Carry does the talking. Domestic data goes mute.
Concede the strongest counter-argument first. Real-rate differentials should mean-revert. That is what every textbook teaches and what the long-run regressions in BIS working papers on G10 carry persistently show — at horizons of two to three years, nominal differentials tend to compress as the slower-growing economy catches its breath or the faster-growing one cools. The concession is real and we are not going to pretend otherwise. The argument that follows is about timing and the cost of waiting for the textbook to be right.
Here is the math, shown in prose so the reader can rebuild every step. Take a hypothetical policy gap of eighty basis points — Fed at five percent, BoC at four-point-two percent — sustained for one full year. On a notional position of one million USD funded in CAD at the cash leg, the carry cost before any spot movement is eight thousand dollars per year. Net of a typical broker financing markup of fifteen to twenty-five basis points on a major-pair overnight swap — which is the spread between interbank funding and what a retail or small-fund book actually pays — the effective carry burden lands closer to ninety-five to one hundred and five basis points annually. On the same one million notional, that is nine thousand five hundred to ten thousand five hundred dollars per year. Spread that across roughly two hundred and fifty-two trading days and you are paying somewhere between thirty-seven and forty-two dollars every day, every day, simply to be short USD against CAD when the gap is open. The trade has to earn that back in spot movement before it earns anything. At a typical daily realised volatility of forty to sixty pips in USD/CAD during a non-crisis quarter, the directional edge required to break even is non-trivial — and the longer you hold, the more compounding works against you rather than for you.
Now compound the second-order effect. When the gap persists for more than two quarters, the futures curve in CAD overnight index swaps starts pricing a one-year-forward differential that no longer matches the spot differential. Forward points widen. The implied cost of carry in any FX forward — which is what corporate hedgers and global macro funds actually trade — moves against CAD longs. At that point the trade stops being about whether the BoC will cut next or hold. It becomes about whether you can carry the position at the new forward cost without the daily mark eating the thesis alive. This is the mechanism by which a "fair value" argument can be analytically correct and economically losing for eighteen months in a row.
Operators built around carry sensitivity feel this first. The records of Saxo Bank and Interactive Brokers swap rates across G10 majors show how quickly retail-facing overnight rolls reflect institutional carry movements — the markup is small but the directional information embedded in it is fast. A persistent policy gap shows up in the swap rates before it shows up in the spot rate, and that is one of the reasons ING's framing keeps repeating across cycles. The desks reading the swap curve already know the cap exists. They wait for the spot to come and tap it.
The USMCA Renegotiation Cap and How It Shows Up in Basis Before It Shows Up in Spot
The second pattern is more interesting because it is harder to see in the headline price. USMCA renegotiation risk does not announce itself in USD/CAD spot. It announces itself in cross-currency basis, in the option skew, and in the bid-offer of one-year forwards weeks or months before the spot rate acknowledges anything has changed.
When trade policy uncertainty rises — and the periodic USMCA review windows are the canonical example — Canadian corporates with USD revenue streams and CAD cost bases face a hedging decision under tightened uncertainty. They tend to do two things in sequence. First, they lengthen the duration of existing USD-receivable hedges, locking in current forward rates rather than rolling shorter tenors. Second, the most sophisticated treasury desks buy USD calls against CAD as cheap insurance against a scenario where review outcomes shock the spot rate lower. Both actions push the same direction in the option market. One-year USD/CAD risk reversals — the price of USD calls over USD puts — drift positive even when spot is quiet. The skew telegraphs the concern before the spot reflects it.
The cap is built in the forward curve and the option skew weeks before it is visible in the price the headlines quote — by the time spot catches up, the desks that read the basis have already moved.
This is the structural reason ING and similar desk notes keep naming USMCA renegotiation as a cap rather than a binary risk event. A binary risk event would imply that CAD trades freely until a deadline and then re-prices around the outcome. The actual mechanism is gradient. Each headline that pushes the renegotiation calendar later, or introduces a new working-group dispute on rules-of-origin or labour content, widens the basis a touch more. The spot does not crater; it simply cannot extend higher. The ceiling is hydraulic — the more pressure builds underneath, the more the air at the top compresses.
The historical analogue worth holding in mind is the pattern around earlier North American trade negotiations through the late nineteen-eighties and the original NAFTA discussions in the early nineteen-nineties. Across those periods the BIS quarterly review papers documented a persistent feature: spot CAD/USD volatility remained well-behaved by major-pair standards while the implied volatility surface and the forward basis carried the entire signal. Traders who read only spot were systematically late. The pattern then is the pattern now. ING is naming the cap because the cap mechanism — basis-first, spot-second — is structurally identical across decades of North American trade-policy episodes.
The Commodity Offset That Almost Always Underperforms Its Headlines
Every cycle, somewhere in the second or third paragraph of an analyst note on CAD, there appears a sentence to the effect that "higher oil prices provide an offset". Every cycle, the offset disappoints. The pattern is so consistent it deserves a name.
The mechanism people imagine is straightforward. Canada exports crude. Higher WTI means more USD revenue flowing into a CAD-cost economy. That should support CAD. The mechanism the market actually prices is messier. The correlation between WTI and USD/CAD on a daily basis has weakened materially since the broader development of US shale capacity in the twenty-tens, because every move higher in WTI now triggers a partially offsetting US production and trade-balance response. The bilateral oil-revenue advantage Canada once enjoyed against the US no longer flows through the FX cross with the same coefficient. Decade-old beta estimates are still being applied to a relationship the underlying economics have already remapped.
Compound that with the carry math from the first section. Even when oil rallies twenty percent in a quarter and CAD responds with a two to three percent appreciation against USD, a trader long CAD for the oil thesis has spent roughly a quarter of those gains paying daily swap costs if the BoC–Fed gap remained open through the trade. The net P&L on the "oil offset" trade in a wide-rate-gap environment is routinely a third to a half of the headline number. This is not a controversial finding inside macro-fund books that have run the back-test; it is the reason commodity-overlay desks at Pepperstone, IC Markets and similar broker-facing flow venues see speculative CAD long positioning roll off faster than directional analyst notes would predict. The flow data and the narrative do not match because the carry tax is invisible in the narrative.
There is a further wrinkle worth naming. The oil-offset thesis assumes that Canadian oil prices clear at the WTI benchmark. In practice Western Canadian Select trades at a discount that widens precisely when North American pipeline politics tighten — which is to say, in the same quarters that USMCA renegotiation noise spikes. The two caps reinforce each other. The commodity that is supposed to provide the offset prices at a worse local benchmark precisely when the policy uncertainty is highest. This is not bad luck. It is the same gradient of trade-policy uncertainty showing up in two market-clearing prices simultaneously.
Put the three patterns together and ING's framing stops looking like a quarter-by-quarter forecast and starts looking like a description of a steady-state. Rate gap caps appreciation through carry. USMCA renegotiation builds the cap quietly in the forwards and skew. The commodity offset that should counterbalance both keeps falling short of its headline coefficient. The loonie does not break; it simply cannot break out.
So What Do You Actually Do
If you are a directional trader holding CAD longs because a desk note named the rate gap as a known cap, the immediate question is whether you are being paid to wait. Run the carry math on your own book. If the all-in financing cost of your CAD long is north of roughly four hundred to four hundred and fifty basis points annualised against the funding leg you have chosen, the position needs to deliver more than one and a half percent spot appreciation per quarter to break even after costs. That is a meaningful bar. In a capped environment where the upper bound is enforced by basis and skew, the realistic upside is often half of what is required. The asymmetric position is short volatility through the cap — selling the wing the carry is paying for — rather than long CAD outright.
If you are a corporate treasury with USD receivables and CAD costs, the calendar matters more than the headline. The patterns above suggest the cost of insurance against a USMCA shock rises silently through the basis and skew well before any specific event window. Layering hedges incrementally across quarters captures cheaper insurance than reacting to a single news event. The desks that consistently come out ahead of these review cycles are not the ones with the best forecast. They are the ones who started buying optionality before the forward curve made the cap visible to anyone reading only spot.
We would reverse this framing if two conditions held simultaneously. The first: a sustained closure of the BoC–Fed policy gap to inside twenty-five basis points for two consecutive quarters, with futures curves pricing the convergence as durable rather than transitional. The second: a published USMCA review outcome that explicitly removes rules-of-origin and labour-content disputes from the open-question column rather than deferring them to a subsequent working group. Until both conditions are met, the cap that ING is naming is the cap that will hold. The pattern will keep reappearing in roughly the same configuration because the structural mechanism has not changed.
FAQ
What does ING mean when it calls the BoC–Fed rate gap a "cap" on the loonie?
A cap, in this framing, means an upper bound on CAD appreciation against USD that holds for as long as the Bank of Canada's policy rate sits meaningfully below the Federal Reserve's. The mechanism is carry: holding CAD long against USD when the BoC pays less than the Fed costs roughly the size of the gap, annualised, plus broker swap markups. That cost compounds daily and forces any directional CAD long thesis to overcome a structural drag before earning anything from spot movement.
Why does USMCA renegotiation risk show up in forwards before it shows up in spot?
Because the institutional players who hedge against the risk — Canadian and US corporates with cross-border revenue and costs — act on uncertainty before they act on resolution. They extend hedge durations, buy USD calls against CAD, and lengthen forward exposure. All of these flows widen cross-currency basis and push one-year risk reversals positive. The spot rate is the last surface to reflect the build-up because spot trades on flow that includes speculative positioning, not just hedging.
How much daily swap cost am I paying to hold a CAD long against USD?
On a one-million USD notional position with the BoC roughly eighty basis points below the Fed and a typical broker overnight markup of fifteen to twenty-five basis points, the all-in carry cost lands between roughly thirty-seven and forty-two dollars per trading day. Over a full year that compounds to between nine and a half and ten and a half thousand dollars in financing alone, before any spot move. Use your broker's published swap rates to recompute for your own size and tenor.
Does a rally in oil prices reliably push CAD higher against USD?
Less reliably than headline coverage suggests. Since the build-out of US shale production capacity through the twenty-tens, the bilateral oil-revenue advantage Canada once held against the United States has compressed. Every WTI rally now triggers a partially offsetting US production response, weakening the historical beta between WTI and USD/CAD. Even when oil rallies move CAD, a long position held through a wide rate-gap environment surrenders a meaningful portion of the gain to carry costs.
What is Western Canadian Select and why does it matter for the FX thesis?
Western Canadian Select is the benchmark price for the heavy crude blend produced in Alberta. It trades at a discount to WTI that widens when North American pipeline politics tighten — which historically coincides with the same quarters that USMCA renegotiation noise spikes. The result is that the Canadian oil revenue that is supposed to offset trade-policy pressure on CAD prices at a worse local benchmark precisely when policy uncertainty is highest, weakening the offset mechanism.
What signal would tell me the cap is breaking rather than holding?
Two coincident signals. First, a sustained narrowing of the BoC–Fed policy gap to inside twenty-five basis points for at least two quarters, with overnight index swap futures pricing the convergence as durable. Second, a published USMCA review outcome that removes specific contested questions — rules-of-origin and labour-content provisions — from the open-issue column rather than deferring them to follow-up working groups. Either signal alone is insufficient; the cap mechanism requires both legs.
Are forex brokers like Saxo Bank, Interactive Brokers, IC Markets, Pepperstone, Exness, XM and FXCM useful for reading carry signals?
The overnight swap rates each broker publishes for major pairs reflect institutional funding markets with a small retail markup layered on top. Monitoring how those swap rates evolve across providers — particularly when policy-rate gaps shift — can give a faster read on how the funding leg is pricing carry than waiting for analyst commentary. The absolute level differs broker to broker; the direction of change is the more useful signal.
Is this analysis a forecast or a description of a pattern?
A description of a pattern. The argument is that the rate-gap cap, the basis-first transmission of trade-policy uncertainty, and the underperforming commodity offset are structural features of the USD/CAD cross that have recurred across multiple cycles. ING's note fits the pattern; the pattern is older than the note. A forecast would name a specific level and date. The pattern analysis names the mechanism and the conditions under which it holds.