Canada Added 75,100 Jobs in July (August 2026): Unemployment at a Two-Year Low of 6.4% — and Why Bank of Canada Rate Pricing Didn't Move

Statistics Canada reported employment up 75,000 (+0.4%) in July with the unemployment rate down to 6.4%, the lowest since July 2024. RBC had forecast 5,000 jobs and no change in the rate; the Reuters poll sat near 16,500. It is one of the largest labour-market surprises Canada has produced this cycle — and the Bank of Canada rate curve did almost nothing. Money markets held roughly 17 basis points of tightening priced by December, unchanged from before the release, and the Canadian two-year yield rose 2.3 basis points. The Canadian dollar nevertheless posted its best level since 11 June. Understanding how both of those things are true at once is the whole lesson of this release.

The preview flagged the unusual mechanic: US payrolls print at the same minute as the Canadian report, so USD/CAD receives new information on both legs simultaneously and the first move reflects the relative surprise. That is exactly what happened, and it happened in the most extreme available form — Canada beat by a factor of fifteen against one forecast while the United States unexpectedly shed jobs. Two labour markets moved in opposite directions on the same tick.

What actually happened

Employment rose 75,000 (+0.4%) in July, the unemployment rate fell 0.1 percentage point to 6.4% — described by Statistics Canada as the lowest rate since July 2024 — and the employment rate rose 0.1 point to 60.9%. The participation rate ticked up to 65.1% from 65.0%, which is the detail that makes the falling jobless rate meaningful: the rate fell while more people entered the labour force, not because they left it. Since April, total employment is up 181,000 (+0.9%). (Primary release: Statistics Canada.)

The industry detail was broad rather than concentrated. Wholesale and retail trade led with 21,000 (+0.7%), followed by finance, insurance, real estate, rental and leasing at 18,000 (+1.2%), professional, scientific and technical services at 17,000 (+0.8%) and construction at 16,000 (+1.0%). Public administration fell 15,000 (-1.2%) and agriculture 9,600 (-4.3%). Among core-aged women aged 25 to 54, unemployment fell 0.3 points to 5.2% on a 33,000 employment gain; the youth rate edged down to 12.6%.

Two lines cut against the headline, and both deserve equal billing. First, the gain was split almost exactly between full-time and part-time work — roughly 38,600 full-time against 36,600 part-time. The preview named the full-time share as the cross-check that decides whether a large print is the signal it appears to be, and on that test the answer is a shrug: the composition is neutral. Second, average hourly wages rose 2.8% year over year to $37.17, down from 3.3% and $37.20 in June. A labour market tightening fast does not usually decelerate its wage growth in the same month.

Why the rate curve ignored a fifteen-fold beat

A print this far above consensus would normally reprice a policy path. It did not, and the reason is that the Bank of Canada's hold has never been primarily a labour-market decision.

Money markets held about 17 basis points of tightening priced by December after the release — essentially where they sat before it. The economist reaction ran the same way. Ariane Curtis of Capital Economics said that while the Bank is likely to sound more hawkish following the fall in the unemployment rate, they are unlikely to rush into tightening policy. Andrew Hencic of TD Economics said they expect the Bank to stay on hold for the rest of the year. Douglas Porter of BMO Capital Markets framed a strengthening backdrop as something that could eventually push the Bank in a hawkish direction rather than something that does so now. David Rosenberg of Rosenberg Research said there is no smoking gun here for the Bank to shift back to a hawkish stance, pointing at the softening in wage growth. (Reaction and market pricing: The Globe and Mail.)

The constraint sits in the inflation data, not the jobs data. The Bank held at 2.25% on 15 July for the sixth consecutive decision while calling labour conditions soft and noting the unemployment rate has sat in a 6.5%-7% range since the end of 2024. Headline CPI eased to 2.8% year over year in June from 3.2% in May, while the Bank's preferred core measures averaged 1.9% — below the 2% target. (Statement: Bank of Canada; CPI: Statistics Canada.)

That configuration is the reason a strong jobs number changes so little. Good employment data removes the residual case for a cut — which was already faint — but it cannot manufacture a case for a hike while core inflation runs below target. The two ends of the policy argument are both blocked, and one month of hiring does not unblock either.

Where the loonie's move actually came from

The Canadian dollar traded 0.6% higher at 1.3935 per US dollar, or 71.76 US cents, its strongest intraday level since 11 June. The US dollar index fell nearly 0.5% to 99.47.

It is tempting to read that as the market rewarding Canadian hiring. The yield decomposition says otherwise. If Canadian rate expectations were doing the work, the Canadian two-year would have jumped and the spread would have narrowed from the Canadian end. Instead the Canadian two-year moved 2.3 basis points — a rounding error for a fifteen-fold surprise — and the bulk of the narrowing came from US yields falling on a genuinely bad American report.

This is the practical version of a point that sounds abstract until a day like this one demonstrates it: an exchange rate never prices one economy. A trader watching only the Canadian release would have concluded the loonie's gain was a Canadian story and drawn the wrong lesson about what Canadian data does to the pair. The structural version of that argument is in what drives the Canadian dollar.

The hold is still an oil story

Nothing in the jobs report touched the mechanism that actually governs the 2 September decision. The gap between Canada's 2.8% headline CPI and its 1.9% core average is almost entirely energy: gasoline was 20.5% higher than a year earlier in June even after falling 10.2% on the month, while CPI excluding gasoline ran at 2.2%.

That gap is closing without the Bank doing anything. Brent settled at $100.69 on 23 July, and by 5 August US benchmark crude was near $74.96 as the market priced a possible US-Iran arrangement on transit through the Strait of Hormuz. Last summer's prices are what this year's index is compared against, and a barrel in the mid-$70s cannot sustain a +20.5% gasoline line for long. We laid out the crude side, including why the loonie sat still through a full round trip in the barrel, in the OPEC and Hormuz read.

The consequence is that the commodity factor and the rate factor are pointing at the same place for once, and neither of them is pointing at the labour market. A cheaper barrel worsens Canada's terms of trade directly and pulls headline inflation toward a sub-target core — and July's employment strength, real as it is, does not offset either channel.

What would change the picture before 2 September

Two dates matter, and the jobs report is not one of them any more.

July CPI lands on 17 August. Because the entire hold case rests on the distance between headline and core, the size of the gasoline base-effect reversal in that print carries more weight for September than the labour data does. A headline that converges quickly toward 1.9% core reopens a conversation the strong jobs number just appeared to close.

The next Labour Force Survey is scheduled for 4 September — two days after the Bank meets. July's report was therefore the last full labour read the Governing Council receives before the decision, which is an argument for taking its composition seriously rather than its headline. On composition it said: hiring is broad, quality is mixed, wages are decelerating. Our July Bank of Canada decision note has the full statement read.

The proposition the preview set up was whether a labour market stuck near 6.5% unemployment could coexist with a policy rate held on the strength of an inflation number that energy is deflating. July answered half of it: the labour market is no longer stuck. The rate curve's response answered the other half by not responding at all. Read the methodology behind the five factors if you want to see how those inputs are separated rather than blended.

Educational macro context only — not investment advice.