Fewer unemployed, not clearly more employed
Between September 2025 and September 2026, the number of unemployed people in Canada fell by 121,200, to 1,474,900, and the unemployment rate fell from 7.1% to 6.5%. Statistics Canada published the figures on 9 October 2026. The standard errors of the two changes are 40,600 and 0.2 point, so both falls are well beyond the survey’s margin.
Employment rose by 94,900 over the same year, but its standard error is 67,500: the rise cannot be told apart from no change. The labour force, everyone in work or looking for it, was 26,300 lower, against a standard error of 63,800, so it was essentially unchanged. The participation rate slipped from 65.2% to 64.8%, a change at the edge of the margin.
In September alone, employment fell by 68,300, more than twice its standard error of 33,200. The unemployment rate edged up from 6.4% to 6.5% in the month, a change within its margin of 0.1 point.
| Measure | Sep 2025 | Sep 2026 | Change | Beyond the margin? |
|---|---|---|---|---|
| Population, 15 and over | 34,671,900 | 34,854,900 | +183,000 | No sampling error |
| Labour force | 22,606,000 | 22,579,700 | -26,300 | No |
| Employed | 21,009,900 | 21,104,800 | +94,900 | No |
| Unemployed | 1,596,100 | 1,474,900 | -121,200 | Yes |
| Unemployment rate | 7.1% | 6.5% | -0.6 | Yes |
| Participation rate | 65.2% | 64.8% | -0.4 | No |
What the survey counts
Unemployed means the ILO definition: without work, available and looking. It is measured by the Labour Force Survey, not by the number of people claiming Employment Insurance, which is a different figure.
The survey is a sample, so every change comes with a margin. Statistics Canada publishes the standard error of each change; this edition treats a change as established only when it is more than twice that size.
The population is the survey’s own, projected by Statistics Canada from its demographic estimates. It carries no sampling error, and it is the scale against which every other figure is drawn.
- THE MEASURE: unemployed under the ILO definition, people without work who are available and have looked for work in the past four weeks. It is not the number of people claiming Employment Insurance.
- THE MARGIN: the Labour Force Survey is a sample. Statistics Canada publishes a standard error for each change; a change more than twice that size is unlikely to be noise, a smaller one cannot be told apart from no change.
- THE POPULATION: the people aged 15 and over the survey represents, projected by Statistics Canada from its demographic estimates. It has no sampling error, and it sets the scale of every other figure.
So 121,200 fewer unemployed does not say that 121,200 more people found work: employment rose by 94,900, within the margin, and the rest of the fall came from a labour force that did not grow.
One measure, one margin and one population, and a reading that forgets any of them says what the source does not.
A population that stopped growing fast
In the year to September 2026, the population aged 15 and over grew by 183,000, the smallest 12-month rise in a series that allows the comparison from October 1977. It had grown by 1,193,800 in the year to September 2024 and by 582,100 in the year to September 2025.
With the population adding fewer people, the labour force did not grow, and unemployment fell even though employment barely moved. The arithmetic holds exactly: the 121,200 fall in unemployment is the 94,900 rise in employment plus the 26,300 change in the labour force.
Over three years the rate has risen and fallen back: 5.5% in September 2023, 6.5% in 2024, 7.1% in 2025 and 6.5% in 2026. Published employment grew each year, by 322,900, 213,900 and then 94,900, the last within the margin.
| September | Population | Labour force | Employed | Unemployed | Rate |
|---|---|---|---|---|---|
| 2023 | 32,896,000 | 21,654,500 | 20,473,100 | 1,181,400 | 5.5% |
| 2024 | 34,089,800 | 22,231,800 | 20,796,000 | 1,435,900 | 6.5% |
| 2025 | 34,671,900 | 22,606,000 | 21,009,900 | 1,596,100 | 7.1% |
| 2026 | 34,854,900 | 22,579,700 | 21,104,800 | 1,474,900 | 6.5% |
What these figures do not say
They do not say why. Nothing here measures immigration policy, tariffs, interest rates or the ageing of the population, and no policy is assessed.
They do not say that more Canadians found work: the rise in employment is within the survey’s margin. They say nothing about pay, hours or Employment Insurance.
They cover Canada as a whole, not any province, and none of this is a forecast. One month of survey data is noisier than a year, and the next release may revise the picture.
