Canada’s economy: national, provincial and metropolitan

The picture in one page

Canada's economy is growing, slowly and unevenly. Real GDP was 1.1% higher in Q2 2026 than a year earlier, with a quarter-on-quarter gain of 0.8% in that quarter, and real GDP per person was 0.6% higher; the population grew 0.5% over the four quarters to 1 July 2026, a quarter later. Over a longer span output per person has barely moved: real GDP per person is 1.5% above its end-2019 level. The labour market tells a similar story. Unemployment is 6.7%, but employment grew only 0.1% over the year and the participation rate is 64.8%, so the unemployment rate says little about hiring on its own; the labour section sets out how the two moved over the year.

The world reaches Canada through a small number of channels that the relationship explorer documents with their evidence: the United States took 70.2% of goods exports in September 2026 (balance-of-payments basis), energy products were 26.3% of all goods exports, Brent crude averaged US$125.00 a barrel in October 2026 and the US ten-year yield was 4.84% on 5 November 2026. Whether any of these moves Canadian jobs, prices or house prices is a matter of evidence the report grades one link at a time; none of this section is a forecast.

Six tensions

The report is organised as a sequence of sections, but the economy is not. Six tensions cut across the sections. Each is a pair of facts that look inconsistent until the relation between them is understood.

The data behind this graphic

What the rest of the report is about

Step	What happens
More people, not more per person	The economy grew largely by adding people. See growth and population.
Pay in dollars outruns productivity	Hourly pay in current dollars has risen far faster than real output per hour. See productivity.
Fewer looking, not more hired	Over the year, unemployment fell as participation fell. See labour.
Energy lifts income and costs	Higher prices raise export values and household fuel bills. See prices, trade and industry.
Borrowing costs lag the policy rate	In May the Bank said the last pandemic-era mortgages would renew within a year, at higher payments. See rates, housing and financial vulnerability.
One customer	Most exports go to one country, and trade policy is unsettled. See trade and outlook.

Download this graphic (SVG, with source)

More people, not more per person. Real GDP is 11.9% higher in Q2 2026 than at the end of 2019. Real GDP per person is 1.5% higher. The difference is population growth: by 1 July 2026, a quarter after the latest GDP figures, the population was 10.5% larger than on 1 October 2019. The growth section sets out the components of output; the population section explains where the new residents came from and why growth has now slowed to 0.5% a year.

Pay in dollars outruns productivity. Hourly compensation in current dollars is up 29.2% from the end of 2019 to Q2 2026, consumer prices 24.6% (to September 2026) and real output per hour 2.4%. The productivity section shows that output per hour has grown slowly for a generation and that Canada’s level of output per worker is 72% of the American. The investment section shows that business investment is no higher than in 2014.

Fewer looking, not more hired. Unemployment is 6.7%, close to the ten-year average, but participation is 64.8%, against 67.7% in November 2007, and employment is up only 0.1% over the year. The labour section separates the two.

How to use this report

Each section carries the same eight core headings, so that a reader can compare. A reader with a specific decision can start where it fits:

• Setting a budget or a price: prices, rates, labour and consumers.
• Deciding where to expand: regions, population, housing and industry.
• Lending or investing: credit and banks, financial vulnerability, investment and government.
• Selling abroad: trade, transportation, agriculture and the outlook.
• Judging the long run: productivity, long-run and technology.

Every figure in the text is drawn from a named source when the page is built. Where a source says something that the data cannot test, it is quoted as the source’s statement, and where the report has no evidence it says so. The report makes no forecast of its own.

Key findings at a glance

1. The economy grew 0.8% in Q2 2026, 1.1% over the year, but output per person is little changed since 2019. Real GDP is 11.9% higher than at the end of 2019 and real GDP per person only 1.5%. In arithmetic terms, population growth accounts for most of the rise in the economy's size. Growth

2. Business-sector productivity is up 2.4% from the end of 2019 to Q2 2026 while hourly pay in current dollars is up 29.2%. Unit labour costs, which combine the two, are up 26.2%, and consumer prices 24.6% (to September 2026). Productivity

3. Unemployment is 6.7%, but employment grew only 0.1% over the year and the participation rate is 64.8%. Over the year the unemployment rate fell while the participation rate also fell and employment growth was weak, which is consistent with fewer people looking for work rather than faster hiring; the reasons are not in these series. Labour

4. Inflation is 3.1%; Statistics Canada says gasoline and fuel oil drove the September acceleration, and the Bank of Canada says the Middle East conflict is keeping energy prices high. The CPI-trim core measure is 1.8%. Since January 2021 the real average hourly wage, which is not seasonally adjusted, has changed by -0.5%: pay has roughly kept pace with prices. Prices

5. The policy rate is 2.50% after the rise of 28 October, and the ten-year yield rose 0.52 percentage points over the past year. Longer-term borrowing costs depend on longer-term yields as well as on the policy rate. Rates

6. Housing starts, at 248,896 in September 2026, are near their decade average, new-home prices changed -2.0% over the year, and CMHC estimates a national supply gap of 187,000 to 238,000 homes a year. CMHC's estimates of the gap differ by city. Housing

7. Population grew only 0.5% in the year to 1 July 2026 as non-permanent residents fell by 155 thousand. Over the same four quarters Alberta's population grew 1.5%, Ontario's 0.3%, Quebec's 0.2% and British Columbia's 0.1%. Population

8. The United States takes 70.2% of goods exports and energy products are 26.3% of all goods exports (balance-of-payments basis, September 2026). The exposure differs by province: on a customs basis in calendar 2025, energy products were 76% of Alberta's exports but 1% of Ontario's. Trade

What happened

Real gross domestic product in the second quarter of 2026 was C$2.52 trillion at an annual rate (chained 2017 dollars), 0.8% higher than in the first quarter and 1.1% higher than a year earlier. The quarter followed 0.1%, -0.2%, 0.5% and -0.2% in the four quarters before it: output stalled from early 2025 to early 2026, then picked up. Over two years real GDP grew 3.2%, or 1.2% per person.

Within the quarter, exports contributed 4.7 percentage points to annualized growth, household spending 1.8 percentage points, business investment 1.7 percentage points and government spending 0.9 percentage points, while a drawdown of inventories subtracted 4.9 percentage points. These contributions are annualized, so they are about four times larger than the quarterly contributions that Statistics Canada discusses in prose.

By industry, the monthly series ends in August 2026, when total real output was essentially unchanged (0.0% from the previous month). Compared with a year earlier, services-producing industries grew 1.7% while goods-producing industries grew 1.0%; oil and gas extraction grew 1.5% and manufacturing 1.8%.

Why it happened

The data show which components moved, not why they moved. For the second quarter, the official explanations are the ones quoted above.

Why it matters

Aggregate output and living standards are different things. Real GDP is up 11.9% since the fourth quarter of 2019, but the population grew over the same period, so real GDP per person is up only 1.5%. Its peak over the past fifteen years was C$61,027 per person in Q3 2022, and the latest quarter is 0.9% below that peak.

Nominal GDP, which is measured in current dollars, grew 7.1% over the year, against 1.1% for real GDP. The gap reflects the change in economy-wide prices (the GDP deflator), and it matters because nominal income, government revenue and debt ratios respond to nominal GDP while living standards respond to real quantities.

Output, people and output per person

Three series tell the long-run story together: how big the economy is, how many people live in it, and the quotient. In 1990 real GDP was C$1.17 trillion a year for 27.5 million people. In the second quarter of 2026 it was C$2.52 trillion for 41.7 million.

Chained 2017 dollars. Real GDP and real GDP per person are annual rates.

Since 2000, real GDP is up 66.6%, population 36.9% (to 1 July 2026) and real GDP per person 21.9%. Between 2014 and the end of 2019 per-person output rose from C$57,193 per person to C$59,621 per person. In the six and a half years since it has risen to C$60,504 per person, which is a smaller gain in a longer time.

The data behind this graphic

Real GDP per person since 1990

Series	Latest
Real GDP per person	C$60,504 per person, Q2 2026

Download this graphic (SVG, with source)

The shape matters as much as the level. Per-person output fell in the early 1990s, then rose until 2008, dipped in 2008–09, grew slowly through the 2010s, fell sharply in the pandemic, recovered by late 2021 and has since been roughly flat. It reached C$61,027 per person in Q3 2022 and is below that now.

How that compares

Other economies have also faced slow growth in the 2020s, but over a decade Canada’s per-person growth stands out as low. At purchasing-power parity, GDP per person rose 4.2% in Canada over ten years, against 19.5% in the United States, 24.2% in South Korea and 12.7% in the euro area. Among the economies shown, only Mexico (2.3%) grew less. Germany (5.3%), which is above Canada in level, grew only slightly more.

The data behind this graphic

Growth in GDP per person over ten years

Item	Value
South Korea	24.2%
United States	19.5%
Euro area	12.7%
France	8.8%
Japan	8.3%
Australia	7.7%
United Kingdom	7.0%
Germany	5.3%
Canada	4.2%
Mexico	2.3%

Download this graphic (SVG, with source)

In the latest year, Canada’s real GDP grew 1.7% in 2025, a little above the euro area (1.4%) and the United Kingdom (1.4%), below the United States (2.2%) and well below the world (2.9%). A single year is a weak guide: Canada’s growth was 0.6% in 2015 and is higher now, while the United States’ has fallen from 2.9% to 2.2%. The ten-year comparison is the more informative one.

The data behind this graphic

Real GDP growth in 2025, selected economies

Item	Value
World	2.9%
United States	2.2%
Canada	1.7%
Euro area	1.4%
United Kingdom	1.4%
Australia	1.4%
Japan	1.2%
South Korea	1.0%
France	0.8%
Mexico	0.6%
Germany	0.2%

Download this graphic (SVG, with source)

History

The 2008–09 downturn took real GDP down 4.4% from its peak in Q3 2008. The pandemic quarter was far sharper: real GDP fell 12.7% from the fourth quarter of 2019 to the second quarter of 2020. Per-person output has had two further declines that are easy to miss in the total: a fall of 1.5% between Q4 2014 and Q2 2016, and of 2.4% between Q3 2022 and Q4 2023. These are measurements of the series, not statements about their causes. This report does not use an official dating of recessions, so it makes no claim about whether any period was a recession.

Across geographies

Provincial real GDP (at market prices, in 2017 constant prices) is published annually and ends in 2025. In that year Alberta's grew 3.3%, Ontario's 1.8%, Quebec's 1.1% and British Columbia's 2.2%. Over the ten years to then, real output rose 34.8% in British Columbia and 25.3% in Ontario but only 17.0% in Alberta and 8.1% in Newfoundland and Labrador. Alberta's ten-year figure includes the fall in 2020; its real output has risen 22.5% since then, and 13.0% since 2019.

The data behind this graphic

Real output growth over ten years

Item	Value
Prince Edward Island	39.2%
British Columbia	34.8%
Nova Scotia	26.8%
Ontario	25.3%
Quebec	23.9%
New Brunswick	17.2%
Alberta	17.0%
Saskatchewan	16.6%
Manitoba	16.5%
Newfoundland and Labrador	8.1%

Download this graphic (SVG, with source)

The chart shows the ten-year ranking in full. Prince Edward Island (39.2%) and British Columbia (34.8%) grew most. Newfoundland and Labrador (8.1%) grew least, then Manitoba (16.5%), Saskatchewan (16.6%) and Alberta (17.0%). The regions section looks at provincial output, jobs and population together.

Statistics Canada publishes GDP for metropolitan areas only in current dollars and several years late (to 2022), on a different basis from these constant-price provincial figures, so none is shown here and nothing here should be read as a city GDP. The city reports show the metropolitan series with those limits stated.

The Global report puts these numbers in context. Canada's real GDP growth in 2025, 1.7%, compares with 2.9% for the world, 2.2% for the United States and 1.4% for the euro area. Among thirteen major economies Canada ranks 5th of 13 on that measure and 4th of 13 on GDP per person at purchasing-power parity, 75% of the US level.

What it influences

Risks

What it means for decisions

For a business planning sales. National output is a poor guide to demand per customer, because per-person output has barely grown. A firm whose sales growth depended on a growing customer base may find that it slows with population growth, while a firm that depends on spending per customer faces near-flat output per person.

For a worker or household. Per-person output is a broad measure of the room for pay to rise in real terms. It has been flat for several years, and the productivity section shows what that means for pay.

For a policymaker. The two sources of growth, more people and more output per person, have different implications. A policy that adds people raises total output but need not raise output per person; a policy that raises investment or productivity raises output per person. The Bank of Canada, cited in the risks section, expects growth to slow in the fourth quarter, and the population section shows the first source slowing.

For an investor. The comparison with peers is more relevant than the national headline: a ten-year record of low per-person growth is part of the context for valuations and for the exchange rate, but this report does not estimate either.

What the data cannot say. The quarterly series are revised, and the international comparison uses a purchasing-power base that is itself revised. Growth contributions are annualised. Nothing here is a forecast.

What to watch

The next releases, both on 30 November 2026, are monthly GDP by industry for September and third-quarter GDP (dates are listed under the charts and in the release calendar).

What happened

Business-sector labour productivity, measured as real output per hour worked, was 105.3 (index, 2017 = 100) in Q2 2026, up 1.0% from the previous quarter and 1.2% from a year earlier. Since the fourth quarter of 2019 business-sector real output has risen 11.2%, but hours worked have risen 8.6%, so most of the growth in output came from working more hours and not from producing more per hour.

Compensation per hour, in current dollars, is up 29.2% since the end of 2019, far more than output per hour, which has pushed unit labour cost (compensation per unit of output) up 26.2% over the same period and 3.2% in the last year.

Why it happened

The accounting is straightforward: unit labour cost rises when hourly compensation grows faster than output per hour. Hourly compensation here is in current dollars, so it includes inflation; the productivity series is in real terms. Part of the 29.2% rise in compensation per hour is therefore price inflation, which is why unit labour cost is a better guide to cost pressure than compensation alone.

The data behind this graphic

Why pay that outruns productivity raises costs

Step	What happens
Compensation per hour	What employers pay for each hour, in current dollars, so it includes inflation.
Output per hour	Real output divided by hours worked: the amount produced in each hour.
Unit labour cost	Labour cost of each unit of output. It rises when the first moves faster than the second.
Margins or prices	Firms absorb the difference in margins, pass it into prices or cut hours. This report does not estimate which.

Download this graphic (SVG, with source)

This report does not separate the sources of the productivity slowdown: reallocation across industries, capital investment, technology adoption and measurement issues are all discussed in the literature and none is estimated here.

Why it matters

History

The business-sector productivity index stood at 63.7 in the first quarter of 1981 (2017 = 100). It is 105.3 now, so output per hour is up 65.3% in the 45 years of data this report holds. That is a rise, but a slow one, and the pace has not been even.

The data behind this graphic

Output has grown faster than hours, but not by much lately

Series	Latest
Real output	116.9, Q2 2026
Hours worked	111.1, Q2 2026
Output per hour	105.3, Q2 2026

Download this graphic (SVG, with source)

The decade-by-decade levels show the unevenness. Output per hour gained more in the 1990s than in either of the two decades after it.

Index, 2017 = 100. Compensation per hour is in current dollars; the other three are in real terms.

Read across a row and the sources of growth separate. Between 1990 and 2000 the productivity index rose from 70.9 to 84.9. Between 2000 and 2010 it rose from 84.9 to 92.8, and between 2010 and the end of 2019 from 92.8 to 102.9. Since the end of 2019 it has moved from 102.9 to 105.3 in Q2 2026. The latest window is the shortest and the weakest of the five.

Over the past twenty years output per hour has grown 0.6% a year on average and over the past ten 0.7%. Quarter-to-quarter growth is noisy, and the averages smooth that, but the table shows the same thing without averaging.

The comparison with pay is the part of the history that has changed most. Over the 45 years, compensation per hour rose from 27.0 to 141.4 in current dollars, a figure dominated by inflation. The unit labour cost index tells the more useful story, because it nets productivity out.

The data behind this graphic

Pay per hour against output per hour

Series	Latest
Compensation per hour	141.4, Q2 2026
Unit labour cost	134.3, Q2 2026
Output per hour	105.3, Q2 2026

Download this graphic (SVG, with source)

Unit labour cost is up 83.8% since the start of 2000 and 26.2% since the end of 2019. A large part of the first rise is inflation, which a nominal series cannot separate, so the right comparison is with prices over the same period; the prices section carries that. The recent rise matters because it came quickly. Since 2000, hours worked have grown by more than output per hour: Canada’s business sector has grown more by adding hours than by making each hour more productive, and the same is true of the past six years.

The data behind this graphic

Since 2000, hours have grown faster than output per hour

Item	Value
Compensation per hour	128.0%
Unit labour cost	83.8%
Real output	61.7%
Hours worked	30.3%
Output per hour	24.0%

Download this graphic (SVG, with source)

Across geographies

Among the peer economies on this platform, Canada’s output per person employed (World Bank, constant 2021 purchasing-power-parity dollars) was $113,563 in 2025, which is 72% of the United States level of $156,983. This definition (per person employed, adjusted for purchasing power) differs from the Statistics Canada hours-based index above, so the two should be compared on direction, not level.

The data behind this graphic

Output per person employed, 2025

Item	Value
United States	$156,983
Australia	$114,582
Canada	$113,563
United Kingdom	$111,234
South Korea	$99,046
Japan	$87,888

Download this graphic (SVG, with source)

In 2000 a Canadian worker produced $98,476 against $109,946 in the United States. Canada was then ahead of the United Kingdom ($93,311) and Australia ($95,966). In 2025 the United Kingdom, at $111,234, is within a few per cent of Canada and Australia, at $114,582, is slightly ahead of it. South Korea, at $57,330 in 2000, is at $99,046 now: it has closed most of its distance from Canada in 25 years. Japan, at $87,888, is below all five.

The data behind this graphic

Growth in output per person employed over ten years

Item	Value
United States	16.2%
South Korea	15.0%
United Kingdom	6.4%
Canada	4.8%
Australia	1.7%
Japan	-1.2%

Download this graphic (SVG, with source)

Over the last ten years the United States and South Korea pulled ahead of the rest. Canada’s 4.8% sits fourth of the six, ahead of Australia and Japan and behind the United Kingdom. The United States rose 16.2%, so the gap that opened with it in the 2000s has grown rather than narrowed. This is a comparison of levels on a purchasing-power basis. It is not a measure of living standards, which also depend on hours worked per person, the terms of trade and how income is distributed.

Regional productivity differences are not assessed: Statistics Canada publishes industry-level productivity for provinces on a different, less timely schedule, and this report does not hold it. The provincial pattern of output per worker for energy-heavy provinces is described in the research note on Alberta’s energy economy.

What it influences

Productivity feeds into wages, unit labour costs and competitiveness. The Bank of Canada's rule of thumb, cited above, is the only statement here that links productivity and wages, and the relationship in Canadian data is not estimated in this report. That inflation feeds into wages is a hypothesis.

Risks

Real investment in machinery and equipment is up 2.8% since 2014 and in intellectual property products 50.7%, while investment in non-residential structures, which peaked in Q4 2014, is 17.3% lower.

The data behind this graphic

Business investment since 2014, by type of asset

Item	Value
Intellectual property products	50.7%
Machinery and equipment	2.8%
Non-residential structures	-17.3%

Download this graphic (SVG, with source)

Spending has grown in intellectual property such as software and research and fallen in non-residential structures such as buildings and pipelines. It is a fact about the asset mix, not by itself a cause of weak productivity, and this report does not estimate the link.

What it means for decisions

For an employer. The arithmetic of the unit labour cost chain gives three choices when pay rises faster than output per hour: accept lower margins, raise prices, or raise output per hour. The first two are limited by competition. The third is what investment in machinery, software and training aims to do, which is why the asset mix in the risks section matters to individual firms and not only to the national accounts. Exporters face the sharpest version, because they may not be able to pass costs on to foreign buyers.

For an investor or lender. A sector where unit labour cost has risen faster than prices has less room for further wage increases. This report does not show unit labour cost by sector; the business-sector aggregate here is only the backdrop.

For a household. Over long periods, living standards track output per person, which is output per hour multiplied by hours per person. Hours per person are limited by demographics and by choice; output per hour is the one that can keep rising. A decade of low growth in it is a decade in which pay gains above inflation are harder to fund.

What the data cannot say. The report cannot tell how much of the weakness is measurement, how much is the composition of what Canada produces, and how much is investment. The peers chart is a comparison of levels on one definition and is subject to revision. Nothing here is a forecast.

What to watch

The productivity release for the third quarter follows the third-quarter GDP release (see the release calendar). Watch whether hours worked and output keep diverging, whether unit labour cost keeps rising faster than prices, and whether business investment in machinery and equipment turns into faster output per hour.

What happened

The unemployment rate was 6.7% in October 2026; it rose by 0.2 percentage points from the previous month and fell by 0.2 percentage points from a year earlier. Employment fell by 4,700 in the month to 21.1 million and is 0.1% higher than a year earlier. The participation rate was 64.8%, which fell by 0.4 percentage points over the year, and the employment rate was 60.5% (-0.3 percentage points over the year).

By industry over the year to October 2026, employment rose 3.9% in health care and social assistance and 1.2% in professional services but fell in educational services (by 3.0%), public administration (1.8%), wholesale and retail trade (3.7%) and forestry, fishing, mining and oil and gas (3.0%).

Fewer of us at work

The unemployment rate counts people who are looking for work and cannot find it. It does not count those who are not looking, which is why two other rates matter more for the long run. The participation rate is the share of people aged 15 and over who are working or looking. The employment rate is the share who are working.

Both are lower than they were. The participation rate is 64.8%, down from 65.8% at the end of 2019 and well below the 67.7% of November 2007, its high of the past twenty years. The employment rate is 60.5%, against 62.2% at the end of 2019 and 63.6% in February 2008.

The data behind this graphic

Participation and employment rates since 1993

Series	Latest
Participation rate	64.80%, October 2026
Employment rate	60.50%, October 2026

Download this graphic (SVG, with source)

This is the sense in which the labour market is weaker than the unemployment rate suggests. Total employment is 21.1 million, 10.0% above December 2019, and employment has grown, but the working-age population has grown faster, so a smaller share of the population is employed. The gap between the two rates, 64.8% and 60.5%, is the unemployed. At the end of 2019 the same two rates were 65.8% and 62.2%.

Where the jobs are

Employment growth since 2019 has been concentrated.

The data behind this graphic

Employment by industry since December 2019

Item	Value
Professional services	29.1%
Public administration	24.8%
Health care	19.5%
Utilities	18.6%
Finance, insurance, real estate	18.6%
Education	15.7%
Information and culture	11.4%
Construction	9.1%
Transportation	7.3%
Manufacturing	6.6%
Resources	1.1%
Trade	-0.8%
Accommodation and food	-4.3%
Business support	-7.8%
Agriculture	-23.7%

Download this graphic (SVG, with source)

Professional services (29.1%), public administration (24.8%), health care (19.5%) and utilities (18.6%) lead. Employment in wholesale and retail trade is close to its end-2019 level (-0.8%), in accommodation and food it is lower (-4.3%), and in agriculture it is 23.7% lower. Manufacturing employment is 6.6% higher, even though its real output is lower, as the industry section notes.

Young and prime-age workers

Youth unemployment is 13.7% against 5.6% for those aged 25 to 54. Both are higher than in December 2019 (10.8% and 4.7%), and both are lower than the pandemic peaks. The gap between the two has widened.

The data behind this graphic

Unemployment by age group: end of 2019 and now

Item	December 2019	Latest
Ages 15 to 24	10.8%	13.7%
Ages 25 to 54	4.7%	5.6%

Download this graphic (SVG, with source)

Youth unemployment is a more cyclical measure than the headline: it rose more than the headline rate in 2008–09, in 2020 and after 2022, and it is lower than the 30.3% of May 2020. Today’s 13.7% is above the rate of 9.0% in July 2022, which suggests that young people have found it harder to find work than at the 2022 low.

Pay

The average hourly wage across all industries is C$37.89 an hour in October 2026, 32.3% above December 2019 (neither wage series is seasonally adjusted, so these comparisons set different months of the year against each other). In constant 2002 dollars it was C$22.14 an hour in September 2026, against C$21.00 an hour in December 2019: after inflation, pay per hour is 5.4% higher. It was C$22.95 an hour in April 2020, a peak reached as employment collapsed in the pandemic; a change in the mix of who is employed can raise an average without any individual’s pay rising.

The data behind this graphic

The average hourly wage, adjusted for inflation

Series	Latest
Real hourly wage	C$22.14 an hour, September 2026

Download this graphic (SVG, with source)

The overview shows that the same wage buys about what it bought in early 2021. The two statements are consistent: pay has risen since 2019 in real terms, but has roughly kept pace with prices since early 2021. The productivity section sets pay against output per hour, and the prices section covers inflation.

Why it happened

The unemployment rate fell over the year while employment grew only 0.1% and the employment rate did not rise. Arithmetically, that combination means the labour force shrank while the population grew, so the fall in the unemployment rate coincided with fewer people counted as unemployed rather than with faster job growth. This is a statement about the numbers; the reasons people left or did not enter the labour force are not in these series.

Why it matters

Unemployment alone can mislead. Over the long run, the employment rate and the participation rate say more about how much of the population is working. The youth unemployment rate was 13.7% against 5.6% for ages 25 to 54, a gap that shows the headline rate averages very different experiences.

Wages grew 2.2% over the year to October 2026, to an average of C$37.89 an hour, down from 3.5% a year earlier, while consumer prices rose 3.1% in the year to September 2026. Average hourly wages are therefore growing more slowly than consumer prices; the figure is for employees and is not adjusted for changes in the mix of jobs, which can move the average even if no individual's pay changes.

History

The unemployment rate peaked at 14.2% in May 2020, touched 4.8% in July 2022 and has averaged 6.6% over ten years, so the current rate is close to its decade average. Participation has fallen from 65.8% at the end of 2019. The job vacancy rate fell from a ten-year high of 6.0% in April 2022 to 2.9% in August 2026 (529 thousand vacancies; the series is not seasonally adjusted), which means far fewer unfilled jobs relative to employment than in 2022.

Across geographies

Provincial unemployment rates range from 5.0% in Manitoba to 8.7% in Newfoundland and Labrador; Alberta is at 7.3% and Ontario at 7.3%. Alberta's employment is 1.6% higher than a year ago while Quebec's is 2.3% lower. Average hourly wages are highest in British Columbia at C$39.54 an hour (1st of 10) and lowest in Manitoba at C$32.46 an hour (10th of 10); Alberta is at C$39.00 an hour (2nd of 10) and Ontario at C$38.96 an hour (3rd of 10).

Metropolitan unemployment rates (three-month averages) are shown separately and must not be read against the provincial monthly rates; see the city reports.

In the latest monthly data the unemployment rate is 6.7% in Canada (October 2026), 4.2% in the United States (September 2026), 6.5% in the euro area (September 2026) and 4.7% in the United Kingdom (July 2026) (national definitions differ; see the Global report).

What it influences

Employment and wages feed household income, spending and tax revenue. Any link from business investment to employment, or from employment to population growth, is a hypothesis: no source measures them here. That wages respond to inflation is also a hypothesis.

Risks

Employment fell by 115 thousand over three months while the unemployment rate stayed near its decade average.

What it means for decisions

For an employer. The headline unemployment rate near its ten-year average suggests that hiring should not be unusually difficult. The vacancy rate is 2.9%, down from 6.0% at its April 2022 peak, and growth in the average hourly wage is slower than inflation, so pressure on pay is lower than in 2022. The sectors that have added the most workers since 2019 are health care and professional services.

For a worker. Real pay is above its December 2019 level but fell over the year to September 2026. Young workers face a higher unemployment rate than at the 2022 low.

What the data cannot say. The Labour Force Survey is a sample, so monthly changes are noisy, and the series used in this section count people, not hours, and do not show why people leave the labour force.

What to watch

The November Labour Force Survey is due on 4 December 2026 (see the release calendar). Watch the participation rate (does it keep falling?), employment in health care and education, the youth unemployment rate and the wage growth figure relative to inflation.

What happened

Consumer prices were 3.1% higher in September 2026 than a year earlier, after 3.0% in August. The acceleration was in energy prices: gasoline prices were 27.8% higher than a year earlier (up from 22.8% a month before) and the energy group as a whole 19.0%. Food prices rose 2.2%, shelter 1.8% and rent 2.8%, while the mortgage interest cost index changed by -0.2%.

Measures of core inflation, on Bank of Canada definitions, were below the headline rate: CPI-trim 1.8%, CPI-median 1.9% and CPI-common 2.4%. The all-items index excluding food and energy rose 2.1%. Goods prices rose 3.7% against 2.6% for services, a year earlier the reverse (1.6% and 3.0%).

What it costs to live now

The headline figure measures how fast prices rose in a year. The price level is the cumulative result of many years. The consumer price index is 170.0 (2002 = 100), against 136.4 at the end of 2019 and 93.5 in January 2000. Prices are 24.6% higher than at the end of 2019, 36.8% higher than in January 2015 and 81.8% higher than in January 2000.

The data behind this graphic

The consumer price index since 2000

Series	Latest
Consumer price index	170.0, September 2026

Download this graphic (SVG, with source)

The line rose steadily until 2020, rose steeply in 2021 and 2022, when the annual rate peaked at 8.1% in June 2022, and has kept rising more slowly. In the fifteen months from March 2021 to June 2022 the index went from 139.6 to 152.9. Slower inflation lowers the rate at which the price level climbs but not the level itself.

The average household does not buy the average basket. Some components have risen much faster than others.

The data behind this graphic

Price changes since December 2019, by component

Item	Value
Gasoline	61.4%
Mortgage interest cost	54.2%
Energy	45.1%
Rent	33.9%
Food	32.9%
Shelter	30.9%
Transportation	30.2%
Services	25.4%
All items	24.6%
Goods	23.5%
Household operations and furnishings	9.6%
Clothing and footwear	-0.2%

Download this graphic (SVG, with source)

Gasoline is up 61.4%, the largest increase of the components shown, followed by mortgage interest cost (54.2%), energy as a whole (45.1%), rent (33.9%) and food (32.9%). At the other end, household operations, furnishings and equipment are up 9.6% and clothing and footwear -0.2%; clothing prices are lower in December than in September, and against the same month seven years earlier they are 3.5% lower. Fuel, housing finance, rent and food are hard to avoid, and they have risen fastest.

For a household that rents, the rent increases since 2019 (33.9%) are a larger burden than the headline; for a homeowner who renewed a mortgage recently, the mortgage interest cost index (54.2%) is. The national average is 24.6% and describes neither. Services are up 25.4% and goods 23.5%: since December 2019 services prices have risen more than goods prices.

Do wages keep up?

Wages are the other side of the cost of living. Over three of the four horizons below, the real hourly wage has risen.

Percentage change to the latest month of each series. The wage series (October 2026) runs ahead of consumer prices (September 2026); the real hourly wage ends with prices, so it is the like-for-like comparison. Neither series is seasonally adjusted, so a row that starts in a different month of the year from the latest one also picks up seasonal movement.

Over the long term, pay has outpaced prices. Since January 2021 it has roughly kept pace: the real hourly wage has changed by -0.5%.

The comparison cuts two ways. A worker who has stayed in the same job will have seen pay rise more or less than the average, since the average wage is affected by who is employed. And output per hour has risen less than the real wage since 2019 (see the productivity section), so faster production does not account for all of the gain.

Why it happened

Crude oil, as measured by the West Texas Intermediate benchmark, averaged US$97.70 a barrel in October 2026, 62.3% more than a year earlier. Retail gasoline prices follow crude oil with refining margins, taxes and seasonal blends in between; over the whole history in the platform, year-over-year changes in the WTI price and in the gasoline CPI have moved closely together within the same month (correlation 0.87) and hardly at all a year apart (-0.03); the correlation edge in the explorer shows all four lags and is not a causal estimate.

Why it matters

Inflation changes what a given income can buy and what borrowers pay. The average hourly wage is up 32.3% since December 2019 (to October 2026) while consumer prices are up 24.6% (to September 2026). Deflating the average wage by the CPI gives a real hourly wage of C$22.14 an hour in September 2026 (in 2002 dollars): its change is 5.4% since December 2019, -0.7% over the past year and -0.5% since January 2021. Put plainly, the purchasing power of the average hourly wage is about where it was in early 2021. The real wage peaked over this decade at C$22.95 an hour in April 2020, coinciding with the pandemic lockdowns. Because the average wage is not adjusted for the mix of jobs, a change in who was employed can move it without any change in anyone's pay; this report does not test that explanation.

Inflation is the target of monetary policy, which affects the cost of borrowing (see the next section). Shelter inflation of 1.8% combines a mortgage interest cost index changing by -0.2% with rents rising 2.8%: renters and homeowners with mortgages are experiencing different inflation.

History

Over the past ten years inflation reached a maximum of 8.1% in June 2022, a minimum of -0.4% in May 2020 and averaged 2.7%. The mortgage interest cost index rose 60.8% between January 2021 and now and its annual growth peaked at 30.9% in August 2023. Prices overall are 24.6% higher than at the end of 2019, and slower inflation does not lower the price level, it only slows its rise.

Across geographies

Year-over-year inflation was highest of the ten provinces in Prince Edward Island (5.7%; 1st of 10 provinces) and Nova Scotia (5.6%) and lowest in Ontario (2.5%; 10th of 10); Alberta was at 3.6% and Quebec at 3.1%. Among the six largest metropolitan areas the range runs from 2.1% in Toronto to 3.5% in Edmonton (3.4% in Calgary, 3.0% in Vancouver and 3.0% in Montréal; the Ottawa–Gatineau series covers the Ontario part only, at 3.0%).

Inflation has risen in the other large advanced economies too: 3.7% in the United States, 4.0% in the euro area (October 2026) and 3.5% in the United Kingdom, against 3.1% in Canada, all September 2026 except the euro area. In the United States, the euro area and Canada the rise began in March, when the Brent crude price went from US$71.10 in February to US$103.70; the United Kingdom's rate was 3.0% in February and 2.6% in June before rising again (see the Global report). The definitions of the indexes differ, so compare direction and timing rather than decimals.

What it influences

Risks

What it means for decisions

For a household. The cost of living is a collection of different price increases. Fuel, rent, mortgage interest and food have risen the fastest, and they are the ones with the least room for substitution. Those who rent or who have recently renewed a mortgage have seen more inflation than the national figure.

For a business that sets prices. Wage growth (2.2% in October 2026) is below inflation (3.1% in September 2026), which reduces cost pressure from labour, but unit labour costs are 26.2% higher than at the end of 2019. Energy is the main source of the latest rise in headline inflation, and costs that depend on fuel are the exposed ones.

For an investor. The distinction between a price level and an inflation rate matters for anything with a fixed nominal income. A bond coupon or a lease with a fixed rent buys about a fifth less than it did at the end of 2019, even if inflation stopped today.

What the data cannot say. The CPI measures a fixed basket of goods and services bought by an average household. It does not include home purchase prices, only the cost of owning a home through mortgage interest and similar costs. Nothing here is a forecast.

What to watch

The next consumer price release is for October, on 16 November 2026 (see the release calendar), followed by the Bank of Canada's next policy announcement on 9 December 2026. Watch whether gasoline inflation eases with the winter-blend switch, whether the core measures stay close to the target, and whether services inflation continues to run below goods inflation.

What happened

The Bank of Canada's target for the overnight rate was 2.50% on 5 November 2026, after the Bank raised it on 28 October by 25 basis points; the net change over the past 250 business days is 0.25 percentage points, and the rate had not changed from the last cut, effective 30 October 2025, until that increase. Government of Canada benchmark yields are higher at longer maturities: 3.17% at two years, 3.39% at five, 3.67% at ten and 4.02% for the long bond. Over the past year the two-year yield rose 0.74 percentage points and the ten-year yield rose 0.52 percentage points, while the policy rate moved only at the end of the period.

The slope of the yield curve (ten-year minus two-year) is 0.5 percentage points, a normal upward slope. It reached -1.3 percentage points on 30 June 2023, when short rates were above long rates.

Where the rate cycle left borrowing costs

A rate cycle is easiest to see in a table of dates. The table gives the policy rate, the prime rate that banks charge their best customers, the five-year posted mortgage rate and the two government bond yields at five dates and at the latest reading. Each dated row is the first reading of that month.

The data behind this graphic

The policy rate, and the two- and ten-year yields since 2020

Series	Latest
Ten-year yield	3.67%, 5 November 2026
Two-year yield	3.17%, 5 November 2026
Policy rate	2.50%, 5 November 2026

Download this graphic (SVG, with source)

Three things follow from the table. First, the policy rate rose from 0.25% to 5.00% and was cut back and then raised by a quarter point on 28 October, and is now 2.50%, which is above where it began. Second, the five-year posted mortgage rate moved far less than the policy rate on the way down: it is 6.19%, against a peak of 7.04%, although the policy rate is half its peak. Third, the ten-year yield is 3.67%, higher than the 3.24% of 1 December 2025, so the market rates that fixed mortgages follow rose for most of the past year while the overnight rate stayed where it was. The weekly prime and posted-mortgage series end on 28 October 2026, while the policy rate and the yields run to 5 November 2026.

The data behind this graphic

Where each rate stands against its ten-year high

Item	Value
Five-year posted mortgage	-12.07%
Ten-year yield	-13.85%
One-year posted mortgage	-32.14%
Two-year yield	-36.22%
Prime rate	-38.19%
Policy rate	-50.00%

Download this graphic (SVG, with source)

The chart shows the same point as a distance from each rate’s ten-year high. The policy rate is furthest below its high. The prime rate, the two-year yield and the one-year mortgage rate come next, because they follow or anticipate the Bank’s decisions most closely; the prime rate shown is the weekly reading of 28 October 2026. The ten-year yield and the five-year mortgage rate are much closer to theirs. A household with a variable-rate mortgage tied to prime has seen relief. A household with a fixed-rate mortgage renewing now faces the five-year rate, and that has come down only a little.

The comparison with other central banks

Canada’s policy rate is below the US rate and level with the ECB’s deposit rate.

The data behind this graphic

Policy rates: January 2020 and now

Item	January 2020	Latest
Canada	1.75%	2.50%
United States	1.55%	3.88%
Euro area	-0.50%	2.50%

Download this graphic (SVG, with source)

Canada’s overnight rate is 2.50% against 1.75% in January 2020, while the US federal funds rate is 3.88% (from 1.55%) and the ECB’s deposit rate 2.50% (from -0.50%). All three central banks raised rates sharply in 2022 and 2023 and have since cut them, and all three have raised them again in 2026, the Bank of Canada last, on 28 October. The US rate is the highest; Canada’s and the ECB’s are level. The gap between the US and Canadian rates was small at the peak (5.33% against 5.00%) and has widened since, because the Bank of Canada has cut by more.

Why it happened

The Bank sets only the overnight rate. Longer yields are determined in bond markets rather than by the Bank, so they can move without any change in the policy rate (in standard terms they reflect expected future policy rates, inflation expectations and risk premiums, which this report does not measure). That was the pattern for most of the past year, until the Bank raised the overnight rate on 28 October.

On 28 October 2026, before any response to the increase, posted conventional mortgage rates were 5.49% for one year and 6.19% for five years, against a prime rate of 4.45%. The five-year posted rate changed by -0.85 percentage points over three years and remains well above its ten-year low of 4.64%.

Why it matters

The federal government pays too: federal interest expense is C$13.9 billion a quarter, up from C$5.5 billion in the first quarter of 2021.

History

The policy rate rose from 0.25% at the start of 2022 to a peak of 5.00% on 13 July 2023 and has since been lowered and, on 28 October 2026, raised again to 2.50%. The ten-year yield peaked at 4.26% on 3 October 2023 and is now 3.67%.

Across geographies

Interest rates are national, but their effects are local. Provincial and city housing, construction and household-debt data are not published on a basis that lets this report compare the rate sensitivity of regions; the housing section shows construction and prices by city. Not assessed: regional differences in mortgage structure, renewal timing and debt-service burdens, which require data (such as Bank of Canada or CMHC regional mortgage data) that this report does not hold.

Since January 2026 the US federal funds effective rate has gone from 3.64% to 3.88% (average of the latest week) and the ECB's deposit rate from 2.00% to 2.50%, while Canada's overnight rate went from 2.25% to 2.50%. The US ten-year Treasury yield, 4.84%, is 1.17 percentage points above Canada's; twelve-month changes in the two yields have a correlation of 0.94 (a statistical association, not a causal estimate). The Global report sets out the central banks and the bond markets.

What it influences

Interest rates reach housing demand and construction, household spending, business investment, the exchange rate and the federal interest bill. The Bank of Canada's own descriptions link the policy rate to mortgage rates, mortgage costs, household spending and inflation. The explorer records weaker links (rates to residential investment, rates to new housing prices) as observed correlations: -0.30 between rate changes and housing-start growth at a three-month lag, -0.47 with residential building investment at six months and -0.34 with new housing prices at twelve months. All four lags are shown on each edge and none of these is an estimated effect.

Risks

What it means for decisions

For a homeowner with a mortgage. The relevant rate is the one at renewal, not the policy rate. The five-year posted rate is 6.19%, against 4.79% at the end of 2021. A borrower who locked in in 2020 or 2021 will renew at a higher rate; the Bank of Canada’s estimate of the payment increase is in the financial-vulnerability section. A borrower with a variable-rate mortgage has seen interest costs fall as the policy rate came down, and may see them rise again with the 28 October increase.

For a business that borrows. Prime was 4.45% on 28 October 2026, before any response to the increase, down from 7.20% at the peak. Longer-term borrowing, which follows bond yields, has not eased by as much. A firm financing long-lived assets faces a ten-year yield of 3.67%.

For a saver or an investor. The yield curve slopes upward again (ten-year yield above two-year), which is the normal shape. Returns on short-term deposits and cash follow the policy rate, which the Bank raised on 28 October.

What the data cannot say. Posted rates are published rates, not the discounted rates most borrowers obtain. The series show market rates and not the distribution of mortgage rates across all outstanding loans, which depends on when each was written. Nothing here is a forecast of rates.

What to watch

The Bank of Canada's next announcement on 9 December 2026, the weekly prime and mortgage-rate postings (do they follow the 28 October increase?), the ten-year yield (how does it move after the increase?), the household debt service ratio for the third quarter, and mortgage-renewal commentary in the Bank's reports.

What happened

Real household final consumption expenditure was C$1.45 trillion at an annual rate in Q2 2026, 1.9% higher than a year earlier. Total final domestic demand, which adds business and government spending, was C$2.58 trillion at an annual rate.

The data behind this graphic

Real household spending, change on a year earlier

Series	Latest
Household spending	1.94%, Q2 2026

Download this graphic (SVG, with source)

The long run puts the latest figure in context. Real household spending has grown 2.3% a year on average over the past ten years. In the latest quarter the pace was 1.9%, below that average but not far from it. The extremes of the series show how unusual the pandemic was: spending fell 15.2% in the year to Q2 2020 and rose 13.9% in the year to Q2 2021.

Retail sales are published monthly by province. In August 2026 they changed from a year earlier by 7.6% in Alberta, 6.4% in Ontario, 2.2% in Quebec and 3.7% in British Columbia. By monthly sales, counting the three territories, Alberta ranks 4th of 13 and British Columbia 3rd of 13. These are current-dollar sales, so part of every figure is price change.

The data behind this graphic

Retail sales, latest month against a year earlier

Item	Value
Prince Edward Island	9.7%
Manitoba	8.5%
Nova Scotia	8.5%
Newfoundland & Labrador	8.1%
Alberta	7.6%
Ontario	6.4%
Saskatchewan	6.4%
New Brunswick	3.8%
British Columbia	3.7%
Quebec	2.2%

Download this graphic (SVG, with source)

In the latest month retail sales were higher than a year earlier in all ten provinces. The fastest growth was in Prince Edward Island (9.7%), Manitoba (8.5%) and Nova Scotia (8.5%); the slowest in Quebec (2.2%), British Columbia (3.7%) and New Brunswick (3.8%). Prince Edward Island is a small market where a few large stores or one price change can move the figure.

A single month against its year-earlier month can mislead, so it helps to take the longer view.

The data behind this graphic

Retail sales since December 2019

Item	Value
Prince Edward Island	49.4%
New Brunswick	42.6%
Newfoundland & Labrador	42.6%
Quebec	40.5%
Nova Scotia	40.2%
Manitoba	39.7%
Alberta	36.5%
Saskatchewan	33.4%
Ontario	32.2%
British Columbia	27.4%
Canada, consumer prices	24.6%

Download this graphic (SVG, with source)

From December 2019 to August 2026, retail sales in current dollars are up 49.4% in Prince Edward Island and 42.6% in New Brunswick, and up 27.4% in British Columbia, the lowest of the ten. Consumer prices rose 24.6% from December 2019 to September 2026. Every province’s sales are above that, but British Columbia’s only just: after allowing roughly for prices, British Columbia’s retail sales have grown the least. That is a statement about current-dollar sales and the national price index. Provincial price indexes differ and are not used here, and the data cannot separate more shoppers, higher prices and changed buying habits such as online purchasing.

Average weekly earnings were C$1,355 a week in August 2026, 3.8% higher than a year earlier; consumer prices in the same month were 3.0% higher than a year earlier. Earnings and the CPI are different surveys with different coverage, so the gap between them is a rough guide and not a measure of purchasing power.

Over the longer term, average weekly earnings, which are not seasonally adjusted, are up 31.8% on the same month seven years earlier, from C$1,028 a week to C$1,355 a week. The CPI rose 24.8% in the seven years to September 2026, so nominal pay has risen faster than prices over the period. Earnings include the effect of changes in the mix of jobs and hours, so they are not a measure of the pay of a typical worker.

Spending, people and output

The headline spending figures do not say who is doing the spending, and one of the largest changes in the past six years is how many people there are. Since the end of 2019 the population is up 10.5% to 41.8 million on 1 July 2026. Real household spending is up 14.5% to Q2 2026, so it has grown faster than the population, but not by much. Real GDP is up 11.9%, which is only just above population growth. Real GDP per person is therefore up 1.5% over six and a half years.

The data behind this graphic

Real GDP per person since 2000

Series	Latest
Real GDP per person	C$60,504 per person, Q2 2026

Download this graphic (SVG, with source)

The per-person line shows less growth than the totals do. Real GDP per person was C$61,027 per person in Q3 2022; it is C$60,504 per person now, a little lower than that peak. Since the first quarter of 2022 it has changed by 0.1% while the population grew 8.4% (to 1 July 2026) and real household spending 11.8%. Over that stretch the economy grew mainly by adding people, not by producing more per person. The population section covers where those people came from and the productivity section what that means for output per hour.

Real household spending has grown faster than real GDP. This report does not show how spending was financed: household debt is 176.4% of disposable income, down from 182.9% at the end of 2019 and a peak of 188.2% in Q3 2022, while credit card balances at chartered banks are C$115.5 billion, up 35.4% on the same month seven years earlier (the series is not seasonally adjusted).

The data behind this graphic

Credit card balances at chartered banks

Series	Latest
Credit card loans	C$115.5 billion, August 2026

Download this graphic (SVG, with source)

Card balances fell to C$67.9 billion in February 2021 in the pandemic and have since risen above the earlier level. They are current-dollar balances, so part of the rise is prices, and they show borrowing, not delinquency; the financial-vulnerability section covers arrears.

Why it happened

Why it matters

Household consumption is one of the main components of spending in the economy (see the growth section), so its pace shapes how much a slowdown in exports or investment is felt. Spending depends on income, borrowing costs and confidence; the interest-rates section shows the debt side, where household debt is 176.4% of disposable income.

History

Real household consumption has changed 14.5% since the last quarter of 2019. The series is quarterly and is revised.

Across geographies

The provincial spread in retail sales growth is wide: Prince Edward Island's rate is 9.7% and the Northwest Territories' 1.4%. Those two are small economies where a few stores or a price change can move the figure, and the data cannot say whether the spread reflects population change, prices, or tourism. No city retail series is held, so none of these figures describes a city.

What it influences

Risks

Not assessed: how much of monthly retail growth is price rather than volume by province, because monthly retail volumes are not held here.

What it means for decisions

For a retailer or consumer business. Demand is growing, but the growth is thinner than it looks per person. Real spending has outpaced the population only modestly since 2019, and real GDP per person is close to flat. Where population growth slows, as the population section describes, volume growth from a growing customer base slows with it.

For a lender. This report does not show how spending was financed. Credit card balances and household debt service are the indicators to follow, and the household-finances and financial-vulnerability sections follow them.

For a regional business. Retail growth differs widely by province; the data here do not show why. The provincial comparison is a guide to where nominal demand is growing, not to profitability.

For a household. Per-person output is the number behind incomes in the long run. When it is flat, real income growth for the average household has to come from somewhere else (for example a shift of income from capital to labour, or lower taxes), and none of those is in this section’s data.

What the data cannot say. Spending by category and by income group is not held here, so it is not possible to say who is spending more or on what.

What to watch

Statistics Canada's monthly retail trade and Labour Force Survey releases, the quarterly national accounts and the Bank of Canada's next announcement on 9 December 2026 (release calendar).

What happened

Household net worth, which is assets minus debt, was 1,021.9% of disposable income in Q2 2026. Its highest reading was 1,131.8% in Q1 2022. Real estate alone was worth 478.1% of disposable income, against a peak of 630.3% in Q1 2022.

The data behind this graphic

Household net worth and real estate as a share of disposable income

Series	Latest
Net worth	1,021.87%, Q2 2026
Real estate	478.07%, Q2 2026

Download this graphic (SVG, with source)

Debt was 14.8% of total assets, and homeowners' equity was 72.7% of the value of real estate. The effective interest cost on mortgage loans was 1.0% in the quarter, a quarterly rate, against 0.6% in early 2022; on non-mortgage loans it was 2.1% against 1.5%.

The distribution accounts divide households into fifths by disposable income. In 2025 the top fifth received 42.0% of all disposable income and the bottom fifth 6.3%. In 2020 the shares were 37.6% and 7.5%; in 1999 they were 40.1% and 6.3%.

Disposable income per household was C$32,802 in the bottom fifth and C$219,061 in the top fifth in 2025. Net saving per household was C$75,469 in the top fifth and -C$38,907 in the bottom fifth.

Five kinds of household

The distribution accounts divide households into five equal groups by disposable income. The average hides large differences. In the latest year, disposable income per household ranges from C$32,802 in the bottom fifth to C$219,061 in the top fifth, more than six and a half times as much.

The data behind this graphic

Disposable income per household, by fifth

Item	Value
Top fifth	C$219,061
Fourth fifth	C$120,182
Middle fifth	C$87,705
Second fifth	C$62,401
Bottom fifth	C$32,802

Download this graphic (SVG, with source)

The shares tell the same story as a proportion of the total. The top fifth of households receives 42.0% of all disposable income, the fourth 23.0%, the middle 16.8%, the second 12.0% and the bottom 6.3%. In 1999 the top fifth’s share was 40.1% and the bottom fifth’s 6.3%. The top share has risen by 1.9 percentage points since then, and the three middle fifths have each lost a little.

The data behind this graphic

Share of all disposable income, by fifth: 1999 and the latest year

Item	1999	Latest
Bottom fifth	6.3%	6.3%
Second fifth	12.6%	12.0%
Middle fifth	17.6%	16.8%
Fourth fifth	23.4%	23.0%
Top fifth	40.1%	42.0%

Download this graphic (SVG, with source)

The year 2020 is the exception. In that year the top fifth’s share was 37.6% and the bottom fifth’s 7.5%, a narrower distribution than in 2019 (40.4% and 6.3%) or in the latest year. The data do not say why, and no source cited here explains it. The distribution has since returned to its earlier pattern.

Saving

The saving numbers are the most unequal. Net saving per household, which is disposable income less consumption, is C$75,469 in the top fifth and C$18,912 in the fourth. In the middle fifth it is -C$10,092, in the second -C$19,944 and in the bottom fifth -C$38,907. The bottom three fifths, on average, spend more than they receive in disposable income.

That is not a statement about debt. Net saving in the national accounts is income less consumption, and it can be negative for households that draw on assets, for retirees who spend from savings, or for households that borrow. The tables do not separate these. They do show that the national saving rate is the product of a small group, and that an aggregate figure for household saving says little about the households in the lower fifths. The financial-vulnerability section describes the balance sheet from the lender’s side.

Who gained since 2019

Over the latest six years the top fifth gained most and the bottom fifth second most. Disposable income per household rose 32.8% in the top fifth and 28.4% in the bottom fifth, with 20.4%, 23.5% and 26.3% in between. Consumer prices rose by about a fifth from 2019 to 2025, slightly more than the second fifth's income, so the second fifth fell slightly behind prices and the others outpaced them. The comparison is in current dollars and does not adjust for each group’s own price index; the prices section shows that necessities have risen more than the average, which could affect lower-income households more.

The data behind this graphic

Growth in disposable income per household since 2019, by fifth

Item	Value
Top fifth	32.8%
Bottom fifth	28.4%
Fourth fifth	26.3%
Middle fifth	23.5%
Second fifth	20.4%

Download this graphic (SVG, with source)

Over the full twenty-six years from 1999 the differences are smaller in annual terms. Income per household is up 136.8% in the top fifth and 125.4% in the bottom fifth, both in current dollars. The top and bottom have grown at similar rates over the long run, which is why the shares have barely changed. What has changed in the last six years is that the three middle fifths have fallen behind both the top and bottom.

Why it happened

Not assessed. These tables show levels and shares; they do not separate the effects of asset prices, wages, government transfers, taxes, interest rates and household composition, and no source explaining the movements is cited here. The top fifth's share was 37.6% in 2020 and is 42.0% now; the data record both but do not say what drove the difference.

Why it matters

The balance-sheet ratios show how large household assets are relative to income and how exposed households are as a group to borrowing costs, which is the channel the interest-rates section describes. The distribution tables show that average figures hide very different circumstances: when the lowest fifth's net saving is -C$38,907 per household, an aggregate saving rate says little about it. A higher share of income at the top also matters for spending patterns, because the tables show households in different fifths saving very different amounts.

History

Net worth relative to income was 669.7% in 2000 and 1,021.9% now. Real estate relative to income was 272.6% in 2000. The top fifth's disposable income per household rose 136.8% between 1999 and 2025; the bottom fifth's rose 125.4%. These are current-dollar changes and include twenty-six years of price increases.

Across geographies

Not assessed. These tables are national; no provincial household balance sheet or income distribution is held here, so nothing here says how Alberta, Ontario or any city compares. The provincial and city sections cover earnings and employment, which are different measures.

What it influences

Risks

An aggregate balance sheet can look strong while particular households are exposed; these totals cannot show how assets and debts are spread. Real estate is worth 478.1% of disposable income, and a measure based on property values moves when prices move. Nothing here is a forecast.

What it means for decisions

For a business selling to consumers. Spending power is concentrated. The top fifth receives 42.0% of disposable income and saves most of what is saved. A product aimed at the top fifth faces a different market from one aimed at the bottom three, where spending exceeds income on average.

For a lender. The aggregate balance sheet is strong. Net worth is 1,021.9% of disposable income and real estate 478.1%. But the lower fifths’ negative net saving means that borrowing, or drawing on assets, is how consumption is financed there. The financial-vulnerability section follows the debt.

For a policymaker. Transfers and taxes shape disposable income, and the tables are after both. The 2020 pattern shows how large a change in the distribution is possible in a single year, though the data do not say what caused it.

For a household. The average fifth describes no one. A household’s position depends on its income group, its age, whether it owns a home and when it bought it, and none of these are in the tables.

What the data cannot say. The tables are national and annual. They are in current dollars, do not adjust for household size or age, and do not identify causes. Nothing here is a forecast.

What to watch

Statistics Canada's quarterly national balance sheet accounts and its annual distributions of household economic accounts (release calendar), and the Bank of Canada's household debt commentary.

What happened

Household credit market debt was 176.4% of disposable income in Q2 2026, and household net worth was 1,021.9% of disposable income.

A balance sheet in three parts

Household finances are easiest to understand as three separate things that are often run together: what households own, what they owe, and what they pay each period to service what they owe.

What they own. Household net worth is 1,021.9% of disposable income. That is above its level at the end of 2019 (929.1%) and well above the 638.4% of the first quarter of 2009, its low in the financial crisis, but below the 1,131.8% reached in Q1 2022. Housing accounts for the fall since that peak. Real estate holdings were 630.3% of income at the peak and are 478.1% now.

The data behind this graphic

Household net worth and real estate, as a share of disposable income

Series	Latest
Net worth	1,021.87%, Q2 2026
Real estate	478.07%, Q2 2026

Download this graphic (SVG, with source)

What they owe. Household credit market debt is 176.4% of disposable income, down from 188.2% in Q3 2022. The reduction is real, but the level is high by any longer comparison: it was 161.9% in the first quarter of 2010 and 114.0% in the first quarter of 2000. Debt is a smaller share of assets than in 2010 (14.8% against 18.6%), as asset values rose, and homeowners’ equity is 72.7% of home value.

The data behind this graphic

Household debt as a share of disposable income, five dates

Item	Value
Q3 2022 (peak)	188.2%
Q4 2019	182.9%
Latest	176.4%
Q1 2010	161.9%
Q1 2000	114.0%

Download this graphic (SVG, with source)

What they pay. The debt service ratio, which measures principal and interest payments as a share of disposable income, is 14.5%. That is below its 15.2% peak in Q1 2023 and above its pandemic low of 12.6%. The two components moved differently. The mortgage ratio is 7.7%, against 7.2% at the end of 2019. The non-mortgage ratio is 6.8%, lower than the 7.8% of 2019: households pay a larger share of income towards their mortgages than before the pandemic and a smaller share towards cards, car loans and lines of credit.

The data behind this graphic

What share of income goes to debt payments

Series	Latest
All debt	14.52%, Q2 2026
Mortgage	7.72%, Q2 2026
Other debt	6.80%, Q2 2026

Download this graphic (SVG, with source)

The effective interest cost on household mortgage loans shows why payments are higher while debt is lower relative to income: it is 1.0% a quarter now, against 0.6% at its low in Q1 2022 and 1.1% at its high in Q2 2024. The cost of non-mortgage borrowing is 2.1%.

Why it happened

Why it matters

History

Consumer insolvencies in the 12 months to June 30, 2026 numbered 145,762, 5.9 per cent more than the previous 12 months, while business insolvencies over the same period changed by -9.7 per cent.

Across geographies

The data behind this graphic

Consumer insolvencies, second quarter against a year earlier

Item	Value
British Columbia	10.9
Ontario	10.2
Canada	6.9
Quebec	3.9
Alberta	1.4

Download this graphic (SVG, with source)

The chart shows two provinces, British Columbia and Ontario, with filings rising faster than the national rate, and two, Quebec and Alberta, where they rose less. Ontario and British Columbia also account for most of the national increase in the number of filings. Among the smaller provinces, filings rose faster than the national rate in Prince Edward Island (14.7 per cent), Saskatchewan (11.1 per cent) and New Brunswick (8.8 per cent), on much smaller numbers. The Bank of Canada’s statement that stress is most acute among Toronto-area borrowers who took out a mortgage in 2022–23 is consistent with Ontario’s figure, but the filings are counts of insolvency proceedings, which depend on a province’s population and on how people choose to deal with debt, and they cannot be read as default rates.

What it influences

The data behind this graphic

How low-rate mortgages reach household budgets

Step	What happens
Pandemic-era mortgages	Borrowers who took five-year fixed-rate mortgages in 2020–21 locked in rates before the increases.
Rates rise	The policy rate was raised from the pandemic low; the posted five-year rate peaked in 2023.
Renewal at a higher rate	Borrowers renewing five-year fixed-payment mortgages taken out during the pandemic face payments the Bank puts at about 15% higher; the last of them renew in the 12 months after May 2026.
Payments absorb income	The mortgage debt service ratio is the measure that shows it.

Download this graphic (SVG, with source)

Risks

What it means for decisions

For a lender. The balance-sheet figures argue against alarm, and the payment figures argue against complacency. Households have large assets relative to income, and the mortgage debt service ratio, at 7.7%, is below its 8.2% peak. But the Bank of Canada’s own assessment is that stress has plateaued, not fallen, and that it is concentrated among households with large mortgages relative to income. The renewal wave is the main test. The relevant data are arrears, not insolvencies, and delinquency series are not held here.

For a household. The figures describe the average. A household with a recent mortgage, a high loan-to-income ratio and little savings is in a different position from one with a paid-off home. The Bank notes that households without sufficient savings are the ones that could fall behind if unemployment rises sharply.

For a policymaker. Wealth in housing is not liquid, and it does not pay a mortgage. The household-finances section shows that the bottom fifth of households spend more than they receive; the balance-sheet figures above are national averages and do not speak to them.

For a business. Consumer-facing sectors are exposed to the same budgets: the consumers section shows that real household spending has grown faster than real GDP since 2019 while output per person has barely moved, and that card balances are higher than before the pandemic.

What the data cannot say. Arrears, loan-to-income distributions by borrower and the renewal schedule are in the Bank’s reports but not in the series this report holds. The insolvency figures are counts, and the second-quarter figures postdate the Bank’s May 2026 report; they are not rates and are not seasonally adjusted. Nothing here is a forecast.

What to watch

The Office of the Superintendent of Bankruptcy's monthly and quarterly insolvency statistics, the Bank of Canada's next Financial Stability Report and Statistics Canada's quarterly national balance sheet accounts (release calendar) show whether the plateau in stress persists. Delinquency series are not held here (bank lending is in the credit-and-banks section), and no source cited here gives a forecast of insolvencies.

What happened

Chartered banks held C$1.74 trillion of residential mortgages at the end of August 2026, C$1.07 trillion of business loans and C$723.3 billion of personal loans, a category that includes C$115.5 billion of credit card loans.

Over the year to August 2026 mortgage balances changed by 1.7%, business loans by 8.4%, personal loans by 9.9% and credit card loans by 6.5%.

Households' debt payments, principal and interest together, took 14.5% of their disposable income in Q2 2026; 7.7% of income went on mortgage debt and 6.8% on other debt. A year earlier the total was 14.8%.

Banks and other depository institutions earned a quarterly return on equity of 0.032 in Q2 2026, against 0.021 a year before and 0.008 in the first quarter of 2023, the weakest reading in the series that began in 2020.

Where the lending went

The balance sheets of the chartered banks hold three big categories. Residential mortgages are the largest, at C$1.74 trillion. Business loans are second at C$1.07 trillion and personal loans, which include C$115.5 billion of credit card balances, are C$723.3 billion.

The data behind this graphic

What chartered banks lend, by category

Item	Value
Residential mortgages	C$1.74 trillion
Business loans	C$1.07 trillion
Personal loans, including cards	C$723.3 billion

Download this graphic (SVG, with source)

Their growth has been very different. Since December 2019 business loans are up 71.8%, mortgages 42.6%, personal loans 35.3% and credit cards 30.9%. The balances are not seasonally adjusted, and business and card balances are usually higher in December than in the months around it, so those two figures understate the rise: against the same month seven years earlier, business loans are up 81.8% and card loans 35.4%. Business borrowing has grown much faster than the mortgage book. Over a longer period the gap is wider still: since August 2015 business loans have grown 155.1% and mortgages 76.4%.

The data behind this graphic

Loan balances in 2015 and now

Item	August 2015	Latest
Residential mortgages	C$987.3 billion	C$1.74 trillion
Business loans	C$418.1 billion	C$1.07 trillion
Personal loans	C$459.2 billion	C$723.3 billion
Credit cards	C$76.1 billion	C$115.5 billion

Download this graphic (SVG, with source)

The most recent year looks different. Mortgage balances rose only 1.7% over the latest twelve months, while business loans rose 8.4% and personal loans 9.9%.

The data behind this graphic

Loan growth over twelve months

Item	Value
Personal loans	9.9%
Business loans	8.4%
Credit card loans	6.5%
Residential mortgages	1.7%

Download this graphic (SVG, with source)

The data behind this graphic

Loan balances since 2010

Series	Latest
Residential mortgages	C$1.74 trillion, August 2026
Business loans	C$1.07 trillion, August 2026
Personal loans	C$723.3 billion, August 2026

Download this graphic (SVG, with source)

What households pay

On the household side the burden of payments matters more than the balance. The debt service ratio, principal plus interest as a share of disposable income, is 14.5%: 7.7% on mortgages and 6.8% on other debt. It was 12.6% in the pandemic quarter, 12.2% in the first quarter of 2000 and 15.2% at its peak in Q1 2023. The financial-vulnerability section describes the components, the renewal of low-rate mortgages and the insolvency figures.

Bank earnings

Banks and other depository institutions earned a quarterly return on equity of 0.032 in the latest quarter, from 0.008 in the first quarter of 2023, the weakest reading in the series, which begins in 2020. The table does not say whether the quarterly ratio is annualised.

The data behind this graphic

Return on equity, banks and other depository institutions

Series	Latest
Quarterly return on equity	0.03, Q2 2026

Download this graphic (SVG, with source)

Why it happened

Why it matters

History

Residential mortgages held by chartered banks were C$987.3 billion in August 2015 and C$1.27 trillion in August 2020; business loans were C$418.1 billion and C$642.3 billion on the same dates, and credit card loans C$76.1 billion and C$73.5 billion. These are current-dollar balances and include inflation.

The debt service ratio was 12.2% in the first quarter of 2000 and 12.6% in the second quarter of 2020, and its highest reading since 1990 was 15.2% in Q1 2023. The mortgage ratio was 6.2% in the second quarter of 2020 and is 7.7% now.

Across geographies

Not assessed. All three tables are national. Chartered-bank lending by province exists in a Statistics Canada table that is not held here, and no provincial debt service ratio is held, so nothing here says where borrowing or payment burdens are highest. The Office of the Superintendent of Bankruptcy's provincial insolvency counts in the financial-vulnerability section are the only regional evidence of stress in this report.

What it influences

This links lending to the interest-rates and household-finances sections: payments on debt reach consumer spending through the share of income left after debt service, a mechanism the Bank of Canada describes for mortgage costs and that these tables do not test.

Risks

What it means for decisions

For a small or medium-sized business. Business loans at chartered banks have grown fastest since 2019, but the Bank of Canada notes that impairments on loans to small businesses have continued to increase in recent months while they have decreased for loans to large businesses. Small firms rely mainly on banks and credit unions for finance, so they are more exposed if credit tightens.

For a household. Mortgage balances at chartered banks are almost flat. This does not mean any single household’s payments are flat: the renewal wave described in the financial-vulnerability section changes payments for those whose mortgages renew, whatever the system-wide balance.

For an investor in financial institutions. Earnings have recovered and the large banks hold capital well above minimums, according to the Bank of Canada. The Bank also says that small and medium-sized banks and credit unions with concentrated exposure to regions or industries affected by tariffs are more vulnerable. The tables here are national aggregates and do not distinguish between institutions.

What the data cannot say. The three tables are national and in current dollars. There is no information on delinquency, loan-to-income distribution or lending by province, and the series are for chartered banks, not credit unions or non-bank lenders. Nothing here is a forecast.

What to watch

Statistics Canada's monthly release of the Bank of Canada chartered-bank series, its quarterly national balance sheet accounts and its quarterly financial statistics show whether balances, payment burdens and bank earnings are changing; the Bank of Canada's next Financial Stability Report states whether arrears and past-due loans have moved. No open table of arrears or delinquency rates is held here, and no source cited here gives a forecast of loan growth or bank earnings.

What happened

Housing starts were 248,896 at a seasonally adjusted annual rate in September 2026, a pace that fell 11.3% from a year earlier and is close to the ten-year average of 237,592 (the latest month is a single monthly reading). The peak of the decade was 319,843 in March 2021. Real residential investment is 22.9% below its Q1 2021 peak and the value of new residential building construction in August 2026 changed -1.1% from a year earlier. The New Housing Price Index changed -2.0% over the year, is -2.4% compared with February 2022 and peaked in August 2022. The CPI rent index rose 2.8%.

How much is being built, per person

National starts hide large differences in what each province is doing relative to its size. Dividing each province’s average annual starts over the last twelve months by its population shows how much it is building per resident.

The data behind this graphic

Housing starts per 1,000 residents, by province

Item	Value
Alberta	9.7
Prince Edward Island	9.6
New Brunswick	8.6
British Columbia	7.6
Nova Scotia	7.0
Quebec	6.6
Manitoba	5.8
Saskatchewan	5.0
Ontario	3.8
Newfoundland and Labrador	3.3

Download this graphic (SVG, with source)

Alberta is building 9.7 homes a year for every 1,000 residents, the most of any province, followed by Prince Edward Island (9.6) and New Brunswick (8.6). British Columbia is at 7.6, Quebec at 6.6 and Ontario at 3.8, lower than all but Newfoundland and Labrador (3.3). Ontario has 16.3 million people, the largest population, and builds at less than half of Alberta’s rate.

The data behind this graphic

Starts per 1,000 residents: five-year average and the latest year

Item	Last five years	Last twelve months
Alberta	8.7	9.7
British Columbia	8.1	7.6
Quebec	5.9	6.6
Ontario	4.9	3.8
Saskatchewan	3.9	5.0
Manitoba	5.1	5.8
Prince Edward Island	7.2	9.6
New Brunswick	6.7	8.6

Download this graphic (SVG, with source)

The comparison with the last five years shows who has changed. The ratio rose in eight provinces and fell in Ontario and British Columbia. Ontario’s rate per resident was 4.9 over five years and is 3.8 over the last twelve months. British Columbia’s is down too (8.1 to 7.6). Alberta’s has risen, from 8.7 to 9.7, and so have Prince Edward Island’s and New Brunswick’s. Alberta’s rise coincides with the fastest population growth among the provinces, as the regions section shows; Ontario’s decline comes as its population growth has slowed sharply, to 0.3% in the four quarters to 1 July 2026. The ratio uses the latest population for both periods, so it is a comparison of building, not of building per resident at the time.

Prices, rents and pay

The three measures of housing cost have moved differently since 2019. The New Housing Price Index measures what builders charge for new homes and does not include resale prices, which are not in this report’s series. It is 16.2% above December 2019. Rents, by contrast, are up 33.9%, and the mortgage interest cost index, which reflects the interest that owners pay, is up 54.2%. Average hourly wages are up 32.3% (to October 2026, a month later than the price series).

The data behind this graphic

Housing costs and pay since December 2019

Item	Value
Mortgage interest cost	54.2%
Rent	33.9%
Average hourly wage	32.3%
Shelter, all	30.9%
New-home prices	16.2%

Download this graphic (SVG, with source)

Put against pay, a renter’s housing cost has risen a little faster than the average wage (33.9% against 32.3%), and the mortgage interest cost index, which reflects the interest that owners with a mortgage pay, has risen by far more. Of the three measures, the price of a new home is the one that has risen by less than pay. That fact is less reassuring than it looks. The index has fallen from its high (4.7% below the August 2022 peak), but financing a purchase costs more than in 2021: the posted five-year mortgage rate was 6.19% on 28 October 2026, against 4.79% at the end of 2021. Affordability depends on the price, the interest rate and income together, and the three series here do not combine them.

City by city

New-home prices have moved in two directions since February 2022.

The data behind this graphic

New-home prices since February 2022, by city

Item	Value
Calgary	11.0%
Montréal	9.0%
Edmonton	-1.5%
Ottawa–Gatineau (Ontario part)	-2.9%
Vancouver	-3.6%
Toronto	-8.5%

Download this graphic (SVG, with source)

Prices are higher in Calgary (11.0%) and Montréal (9.0%) and lower in Toronto (-8.5%), Vancouver (-3.6%), the Ontario part of Ottawa–Gatineau (-2.9%) and Edmonton (-1.5%). Montréal’s index is at its highest of the decade now, while Toronto’s is 9.2% below its March 2022 peak. Since the end of 2019, though, prices are up in every city shown: 42.9% in Montréal, 33.4% in Calgary, 16.2% in Vancouver and 2.4% in Toronto.

Why it happened

Starts reflect developers' decisions, which depend on financing costs, construction costs, approvals and expected sales or rents. None of these is decomposed here. One observation from the data stands out: over the twelve months to September 2026 the number of units under construction changed by 9.2% in Montréal, 4.1% in Ottawa–Gatineau, 4.0% in Edmonton, 2.1% in Calgary, -2.3% in Vancouver and -4.2% in Toronto.

Why it matters

Over the same period since February 2022, new-home prices for the country as a whole changed by -2.4% and the rent index by 27.5%; since the end of 2019 the rent index has risen 33.9%. The mortgage interest cost component of the CPI, which captures the cost of owning with a mortgage, is changing -0.2% a year, so renters and owners face different inflation (see the prices section).

History

Starts surged to 319,843 in March 2021 and have since settled near the decade average. New-home prices peaked in August 2022 and are lower now; the rent index's annual growth peaked at 8.9% in May 2024 and has slowed to 2.8%. Policy-rate changes and housing indicators are associated in the data, as the explorer shows (for example, a correlation of -0.30 between 12-month changes in the policy rate and in starts at a three-month lag). That is an association between overlapping 12-month changes, not an estimated effect.

Across geographies

New-home prices over the year changed -4.9% in Toronto, -2.1% in Vancouver, 0.0% in Calgary, -0.8% in Edmonton, 1.2% in Montréal and 1.0% in the Ontario part of Ottawa–Gatineau. Since February 2022 they changed 11.0% in Calgary and 9.0% in Montréal but -8.5% in Toronto and -3.6% in Vancouver. Starts by province (12-month averages) show Ontario at 62,390 units a year, Quebec at 60,190, Alberta at 49,529 and British Columbia at 43,325.

What it influences

Residential construction supports employment in construction and related trades and feeds household spending through the wealth and moving effects that this report does not measure. Any link from housing costs to business investment and population growth is only a hypothesis (no source measures them here). The relationships that carry evidence are the policy-rate and mortgage-rate links (Bank of Canada) and the supply-affordability link (CMHC).

Risks

What it means for decisions

For a developer or lender. Building per resident is highest in Alberta, where population is growing fastest among the provinces. The ratio rose in eight provinces and fell in Ontario and British Columbia. New-home price changes do not line up with population growth across these cities.

For a renter. Rents are up by about as much as average pay since 2019 and rose 2.8% over the last year, after a peak of 8.9% in May 2024. That is a slowing rise, not a fall.

For a buyer. The price of a new home is lower than at its peak in most cities. The cost of financing it is higher than in 2021, and the interest-rates section shows by how much.

What the data cannot say. The per-resident starts ratio uses the latest population for both periods. Starts are units begun, not completed, and are not split by tenure. The New Housing Price Index does not cover resale prices. Nothing here is a forecast.

What to watch

The monthly CMHC housing starts release (mid-month), Statistics Canada's New Housing Price Index and investment in building construction (see the release calendar), the Bank of Canada's next announcement on 9 December 2026, and CMHC's next housing supply report.

What happened

Canada's population was 41.8 million on 1 July 2026, 0.5% more than a year earlier and 0.2% more than at the start of the previous quarter. Over the last four quarters Canada received 368 thousand immigrants (new permanent residents). The number of non-permanent residents (temporary workers, international students and asylum claimants) changed by -155 thousand over the same four quarters, to 2.8 million, or 6.7% of the population. The peak was 3.0 million in Q4 2024. Net emigration was 62 thousand.

The arithmetic of a slowdown

The population changes by four things. People are born and die; permanent residents arrive; temporary residents come and go; and some residents emigrate. In the last four quarters, the population grew by 189 thousand, to 41.8 million.

The data behind this graphic

What moved Canada’s population in the last four quarters

Item	Value
Immigrants (adds)	368 thousand
Net emigration (subtracts)	62 thousand
Non-permanent residents, net change	-155 thousand

Download this graphic (SVG, with source)

Immigrants added 368 thousand. Net emigration took away 62 thousand. The net change in non-permanent residents, who are temporary workers, students and asylum claimants, was -155 thousand. Natural increase (births less deaths) is not in these series; it makes up the difference between the sum of the three and the total change. The point is that the temporary-resident number moved from adding to subtracting. At its peak the net change was 308 thousand in a single quarter, in Q3 2023; its lowest was -80 thousand in Q4 2025.

The data behind this graphic

Immigrants and the net change in temporary residents

Series	Latest
Immigrants	99 thousand, Q2 2026
Non-permanent residents, net change	-19 thousand, Q2 2026

Download this graphic (SVG, with source)

Temporary residents: in, then out

The stock of non-permanent residents, 2.8 million now, was 3.0 million at its peak in Q4 2024 and 1.3 million when the series begins in 2021. The fall has not been even. Study-permit holders have gone from 670 thousand in Q1 2024 to 399 thousand, a change of -40.5%. Work-permit holders peaked later, at 1.6 million in Q4 2025, and are 1.5 million now.

The data behind this graphic

Non-permanent residents by type of permit

Series	Latest
All non-permanent residents	2.8 million, Q3 2026
Work permits	1.5 million, Q3 2026
Study permits	399 thousand, Q3 2026

Download this graphic (SVG, with source)

The two groups are different populations with different consequences. Work-permit holders are permitted to work, so a fall in their number can reduce labour supply as well as the number of households. The labour section shows that employment is only 0.1% higher than a year ago.

Who absorbed the change

The change was not spread evenly. Over four quarters the net change in non-permanent residents was -86 thousand in Ontario, -32 thousand in British Columbia and -22 thousand in Quebec, together about nine tenths of the national -155 thousand. Alberta’s was -7,957 and Manitoba’s -6,814; in New Brunswick it was 1,521.

The data behind this graphic

Net change in non-permanent residents by province, last four quarters

Item	Value
New Brunswick	1,521
Prince Edward Island	-264
Nova Scotia	-357
Newfoundland and Labrador	-560
Saskatchewan	-849
Manitoba	-6,814
Alberta	-7,957
Quebec	-22 thousand
British Columbia	-32 thousand
Ontario	-86 thousand

Download this graphic (SVG, with source)

That distribution matches the pattern of population growth in the regions section: the three provinces that lost the most temporary residents are the three whose population grew least (0.3% in Ontario, 0.2% in Quebec and 0.1% in British Columbia). Alberta’s population grew 1.5% while it lost 7,957 temporary residents, and interprovincial migration added 22 thousand people. The research note on population flows by province covers all four flows, and the one on six cities covers the cities.

The long view

Population has grown in each decade, but the path is not uniform.

Since 1990 the population is up 52.2%, since 2000 36.9% and since 2010 23.6%. The recent years are a large part of those totals. Since the start of 2022 the population is up 8.4%, and since the start of 2024 only 2.8%: a surge, and then a slowdown to the pace Statistics Canada describes as the lowest since the mid-1940s.

Why it happened

The population changes by natural increase, immigration, the net movement of non-permanent residents and emigration. In the last four quarters, immigrants added 368 thousand, the net change in non-permanent residents subtracted 155 thousand and net emigration subtracted 62 thousand. The remaining difference is natural increase (births minus deaths) and statistical adjustments, which are not held here. The fall in study-permit holders is the largest piece of the non-permanent resident decline: over the year they changed by -26.1%, while work-permit holders changed by 2.7%.

Why it matters

Population growth adds workers, consumers and households. When it slows sharply, so does the growth of the labour force, the demand for housing and the base for sales, tax revenue and public services.

History

The population has risen 10.5% since the fourth quarter of 2019, most of it before the past year. The number of immigrants peaked at 145 thousand in Q1 2023 and the net inflow of non-permanent residents peaked at 308 thousand in Q3 2023 before turning negative, reaching -80 thousand in Q4 2025.

Across geographies

Growth over the past year was 1.5% in Alberta, 1.0% in Saskatchewan and 1.0% in New Brunswick but only 0.3% in Ontario, 0.2% in Quebec and 0.1% in British Columbia.

Interprovincial migration is part of the Alberta story: net interprovincial migration added 22 thousand to Alberta's population over four quarters (1st of 10 among provinces) and removed 18 thousand from Ontario (10th of 10). The net change in non-permanent residents over four quarters was -86 thousand in Ontario, -32 thousand in British Columbia, -22 thousand in Quebec and -7,957 in Alberta.

City populations are annual estimates on 2021 boundaries and end in 2025, a year before the national series. They changed by 3.1% in Edmonton, 2.9% in Calgary, 2.3% in Ottawa–Gatineau, 0.5% in Montréal, 0.2% in Vancouver and 0.0% in Toronto in that year. Because they end in 2025, they may not show the 2026 fall in temporary residents.

What it influences

Population growth feeds labour supply, housing demand, consumer spending and tax revenue. Immigration policy feeds into population growth (a strongly evidenced accounting relationship). Population growth is linked to housing and rents by an observed association in Calgary and a hypothesis nationally.

Risks

Estimates of non-permanent residents are the most uncertain component and were revised by more than usual in 2026. Further revisions to the stock of temporary residents would change the population and per-capita figures in this report.

What it means for decisions

For a landlord or developer. The fall in temporary residents is concentrated in Ontario, British Columbia and Quebec, the three most populous provinces. Whether rent growth falls as a result is a hypothesis. Rents rose 2.8% over the year to September 2026, after a peak annual rise of 8.9% in May 2024.

For an employer. The loss of work-permit holders from their peak is a loss of workers, not only of households, and it arrives while employment is flat. Sectors that rely on temporary workers may feel it first. The labour and industry sections show where employment is concentrated.

For a government. A slower-growing population could lower growth in tax revenue and in demand for services, and the ageing trend has resumed; the long-run section shows how far ageing has gone. The levels plan is the policy variable.

What the data cannot say. Non-permanent resident estimates were revised by more than usual, as Statistics Canada notes. Natural increase and statistical adjustments are not in the series. City figures end a year before the national ones.

What to watch

The next quarterly population estimates, what the 2027–2029 levels plan released on 5 November 2026 does to arrivals, and whether the net outflow of non-permanent residents continues.

What happened

Canada's merchandise exports were C$78.6 billion in September 2026 and imports C$73.2 billion, a balance-of-payments surplus of C$5.4 billion. Exports to the United States were C$55.2 billion, 70.2% of the total, up from 69.9% in the previous month; the United States supplied 58.5% of imports. Against the same month a year earlier, exports to the United States changed by 20.4%, to China by 67.3%, to the United Kingdom by 52.9% and to the European Union by 3.9%. Most of these changes are large; this report does not separate timing effects, price effects and underlying demand.

A balance built from opposites

The sum of Canada’s goods trade is small. Over the twelve months to September 2026 the balance is C$10.7 billion on a balance-of-payments basis, near zero against about C$865.7 billion of exports. The number is small because it is the sum of large numbers with opposite signs.

The data behind this graphic

Goods trade balance by product group, twelve months to August 2026

Item	Value
Energy	C$156.6 billion
Metal products	C$72.3 billion
Farm and food	C$27.2 billion
Forestry	C$8.3 billion
Aircraft	C$4.2 billion
Minerals	-C$0.5 billion
Chemicals	-C$24.4 billion
Motor vehicles and parts	-C$59.8 billion
Electronics	-C$63.7 billion
Consumer goods	-C$71.8 billion

Download this graphic (SVG, with source)

Energy has a surplus of C$156.6 billion over the twelve months to August 2026, of which crude oil alone is C$120.7 billion. Metal products have a surplus of C$72.3 billion, which in Statistics Canada’s classification includes precious metals such as gold (the gold price was US$4,411 an ounce in August 2026 against US$3,368 a year earlier), and farm and food products C$27.2 billion. Against that, consumer goods have a deficit of C$71.8 billion, electronics C$63.7 billion, motor vehicles and parts C$59.8 billion and chemicals C$24.4 billion.

The shape is the shape of a resource exporter: it sells raw and lightly processed materials and buys manufactured goods. It is also a reason that the trade balance is sensitive to commodity prices. A fall in energy prices would shrink the largest surplus, and the products Canada buys would not become cheaper by the same amount. The product groups overlap (crude oil is part of energy) and do not cover all goods, so they do not sum to the total. Motor vehicles and parts are a deficit because Canada imported C$146.1 billion of them over the twelve months to September 2026 and exported C$86.3 billion.

The data behind this graphic

The monthly goods trade balance since 2015

Series	Latest
Goods trade balance	C$5.4 billion, September 2026

Download this graphic (SVG, with source)

The monthly balance swung from a surplus of C$7.0 billion in May 2022 to a deficit of C$7.2 billion in April 2025. Statistics Canada says in the release cited above that tariff announcements can change the timing of shipments, so monthly balances swing for that reason as well as for prices and volumes. It is C$5.4 billion in the latest month.

The exchange rate

One US dollar cost an average of C$1.41 in October 2026. It cost C$1.21 in May 2021, the low of the series, and C$1.44 in January 2025, the high. A weaker Canadian dollar raises the Canadian-dollar value of exports sold in US dollars, so part of the rise in dollar exports since 2019 is currency. The exchange-rate series begins in 2017 here, so a longer comparison is not available in this report.

The data behind this graphic

Canadian dollars per US dollar

Series	Latest
C$ per US$	C$1.41, October 2026

Download this graphic (SVG, with source)

Dependence, province by province

The United States is the dominant customer for almost every province, but the dependence differs. In 2025, 90% of New Brunswick’s domestic exports went to the United States, 85% of Alberta’s and 80% of Prince Edward Island’s. British Columbia (51%), Saskatchewan (54%) and Newfoundland and Labrador (35%) are the least dependent.

The data behind this graphic

Share of each province’s exports sold to the United States, 2025

Item	Value
New Brunswick	90%
Alberta	85%
Prince Edward Island	80%
Quebec	72%
Ontario	72%
Nova Scotia	69%
Manitoba	66%
Saskatchewan	54%
British Columbia	51%
Newfoundland and Labrador	35%

Download this graphic (SVG, with source)

The research notes on Canada–United States trade, on Canada’s other trading partners and on provincial energy exports develop each of these in turn, with historical context and the structure of the agreements.

Why it happened

Exports of energy products are C$20.6 billion, 50.8% more than a year earlier, and crude oil exports are C$14.7 billion, 46.9% higher. The WTI price went from US$63.70 a barrel in September 2025 to US$96.90 in September 2026, so part of the change in export value is price rather than volume; this report does not split the two.

Why it matters

Concentration is the key exposure. In calendar 2025, the United States took 72.3% of Canada's goods exports and supplied 45.9% of its imports on a customs basis; the United Kingdom, the next largest single partner in the table, took 6.0% of exports. Energy products make up 26.3% of exports now, against a low of 9.4% in April 2020 and a high of 29.2% in April 2022: when energy prices rise, the energy share of exports rises with them, and so does the economy's exposure to those prices.

In September 2026, motor vehicles and parts exports were C$7.3 billion and consumer goods exports C$7.7 billion.

History

The United States share of exports ranged from 64.7% in April 2020 to 79.6% in February 2025 over twenty years and averaged 74.0% over ten. Energy's share of exports has ranged from 9.4% in April 2020 to 29.2% in April 2022. One US dollar cost C$1.41 in October 2026, a change of 1.0% from a year earlier and of 13.7% from five years earlier, having peaked at C$1.44 in January 2025.

Across geographies

Exposure differs sharply by province. In 2025, 90% of New Brunswick's domestic goods exports and 85% of Alberta's went to the United States, against 35% for Newfoundland and Labrador and 51% for British Columbia; Ontario and Quebec, whose exports are mainly manufactured, were at 72% and 72%. Energy products made up 76% of Alberta's domestic exports, 54% of Newfoundland and Labrador's, 38% of New Brunswick's and 33% of Saskatchewan's but only 1% of Ontario's.

These are measured shares of export value by province of origin. They show where a change in US demand, US tariffs or energy prices would first appear in export values; they do not measure employment or income effects, and the explorer says so on each of the exposure edges. City exposure is not measured here. Statistics Canada does publish the value of exports by census metropolitan area (Table 12-10-0137-01, trade in goods by exporter characteristics), but on a different basis from these province-of-origin figures, and those figures are not held in the data library.

What it influences

Exports drive manufacturing and resource output, transportation and port activity, and provincial revenue. The relationship explorer records the documented exposures (US share by province, energy share by province, motor vehicles in exports). The effects of US trade policy on manufacturing and logistics are hypotheses, and the effect of the exchange rate on exporters is not measured here.

For the world trade backdrop, tariff events documented so far and the other economies' trade openness, see the Global report and its tariffs section; the United States took 72% of Canada's goods exports in 2025.

Risks

What it means for decisions

For an exporter. A balanced total conceals strong segments and weak ones. The sectors with surpluses (energy, metals, farm and food) sell to a world market whose prices move; the sectors with deficits (consumer goods, electronics, vehicles) compete with imports at home. Which side a business is on determines its exposure to prices and to tariffs.

For a provincial government. A province selling nine tenths of its exports to one country is more exposed to its trade policy than one selling half. In 2025 the largest dependences were in New Brunswick, Alberta and Prince Edward Island.

For a policymaker. The target of doubling non-US exports in a decade, cited in the transportation section, would be a large change from the present shares.

What the data cannot say. The product groups are broad. The monthly figures are volatile and affected by the timing of tariff announcements. Value is not volume, and no tariff schedules are held here.

What to watch

The October trade release on 8 December 2026, monthly US share of exports (does it stay near its recent level?), the effect of tariffs on autos and electrical products, and energy prices. The release calendar lists the next dates.

What happened

Real business gross fixed capital formation was C$428.0 billion at an annual rate in Q2 2026, 0.2% higher than a year earlier, or 17.0% of GDP. Within it, machinery and equipment rose 6.3% over the year and intellectual property products 3.3%, while non-residential structures changed by -2.1% and residential structures by -1.7%. Government investment is C$103.4 billion, 3.6% higher over the year and 23.5% higher than at the end of 2019. Business investment contributed 1.7 percentage points to annualized GDP growth in the quarter.

Twelve years without growth

The sharpest fact in the investment accounts is a comparison across more than a decade. Real business investment is C$428.0 billion at an annual rate, against C$431.2 billion in the first quarter of 2014: a change of -0.7%. Over those twelve years real GDP rose 25.2%, household spending 31.7%, exports 28.6% and government investment 34.2%. Business investment is the one large component that did not grow.

The data behind this graphic

Business investment since 2010

Series	Latest
Business investment	C$428.0 billion, Q2 2026

Download this graphic (SVG, with source)

It did not move in a straight line. It fell to C$345.8 billion in Q2 2020 (the pandemic quarter), rose to C$450.1 billion in Q1 2022 and has since eased to C$428.0 billion. It stands 2.5% above its end-2019 level, a small gain over six and a half years. Investment is lumpy, and a single quarter proves little; the twelve-year comparison is what stands out.

Because investment adds to the capital stock, the cost of a flat line accumulates. A smaller addition to plant, equipment and software each year leaves each worker with less to work with than they would otherwise have had. The productivity section shows the output side of the same picture. The two are consistent, but the national accounts do not establish that one caused the other.

Why it happened

The national accounts show what was invested in, not why firms invested or held back. The pattern across assets is itself informative. Investment in intellectual property products is 50.7% higher than in 2014 and machinery and equipment 2.8%, while non-residential structures (which include pipelines, plants and commercial buildings) have changed by -17.3% and total business investment by -0.7% since the first quarter of 2014. The weakness is concentrated in non-residential structures.

The data behind this graphic

Real investment by type of asset, latest quarter

Item	Value
Residential structures	C$154.6 billion
Non-residential structures	C$130.7 billion
Machinery and equipment	C$85.6 billion
Intellectual property products	C$63.9 billion

Download this graphic (SVG, with source)

The data behind this graphic

Investment since the end of 2019, by type

Item	Value
Intellectual property products	39.7%
Government investment	23.5%
Machinery and equipment	9.1%
Business investment, total	2.5%
Non-residential structures	2.1%
Residential structures	-6.6%

Download this graphic (SVG, with source)

The composition has changed more than the total. Spending on intellectual property products, which means software, research and development and similar assets, has grown 39.7% since the end of 2019, and machinery and equipment 9.1%. Non-residential structures, which are the second-largest category of business investment at C$130.7 billion, have grown only 2.1%, and housing investment is down 6.6%.

Non-residential building construction put in place was 10.3% higher in the latest month than a year earlier, with industrial buildings changing by 22.1%, institutional buildings by 13.5% and commercial buildings by 1.2%.

The data behind this graphic

Building construction by type, latest month against a year earlier

Item	Value
Industrial	22.1%
Institutional	13.5%
All building	2.2%
Commercial	1.2%
Residential	-1.1%

Download this graphic (SVG, with source)

By type, institutional and industrial building are up most on a year earlier, and commercial building is close to flat. Institutional construction (hospitals, schools, universities) is influenced by public budgets. Industrial construction (C$1.0 billion in the latest month) is small in value but rising quickly from a low base. Residential construction, at C$8.6 billion, is by far the largest category and is the one the housing section covers.

The oil and gas cycle inside the total

Statistics Canada’s capital expenditure survey makes it possible to look inside the structures figure at one industry, oil and gas extraction. It is measured in current dollars and covers all non-residential tangible capital, not only structures, but it shows the size of the swing.

The data behind this graphic

Capital spending: all industries, and oil and gas extraction

Series	Latest
All industries	C$401.2 billion, 2026
Oil and gas extraction	C$43.6 billion, 2026

Download this graphic (SVG, with source)

Spending by oil and gas extraction was C$76.1 billion in 2014, C$37.6 billion in 2016 and C$21.8 billion in 2020, and is C$42.0 billion in 2025 on a preliminary basis. The industry spent less in 2025 than in 2014 in current dollars, so after inflation the fall is larger. Spending by all industries together went the other way. It was C$272.1 billion in 2014 and C$386.8 billion in 2025, and the 2026 intentions of C$401.2 billion are 47.4% above 2014, all in current dollars.

The data behind this graphic

Oil and gas as a share of all capital spending

Item	Value
2014	28.0%
2016	16.4%
2019	12.4%
2022	11.2%
2025	10.9%
2020	8.8%

Download this graphic (SVG, with source)

Oil and gas extraction accounted for 28.0% of all capital spending in 2014 and 10.9% in 2025. Spending by every other industry combined rose from C$196.0 billion to C$344.7 billion. In current dollars, oil and gas accounted for most of the 2014-16 fall in capital spending, and the rise since has come from the other industries. The comparison with real investment needs care, because the survey measures current dollars and the national accounts measure real chained dollars, so the two should be compared on direction, not size.

The same dataset shows the exposure of one province. Alberta’s share of energy capital spending and the dependence of its public revenue on it are described in the research note on Alberta’s energy economy.

Why it matters

Investment builds the capital stock that determines future output per worker. Weak investment in structures and a long run of weak output per hour (see the productivity section) are consistent with one another, but the national accounts do not show that one caused the other. That business investment supports employment is a hypothesis with no evidence here; the Bank of Canada said in 2015 that the fall in oil prices would weaken oil and gas investment. Public investment has grown faster than business investment since 2019, which matters for infrastructure and for the fiscal position.

History

Real business investment peaked at C$450.1 billion in Q1 2022; it is 2.5% above its level at the end of 2019 and -0.7% relative to the first quarter of 2014. Foreign direct investment into Canada totalled C$80.3 billion over the last four quarters, with a ten-year quarterly high of C$36.5 billion in Q2 2024; Canadian direct investment abroad over the same period was C$93.3 billion. Direct investment flows are volatile and include mergers, acquisitions and reinvested earnings, so they are not a measure of new plants.

The data behind this graphic

Direct investment into Canada and out of it

Series	Latest
Foreign direct investment in Canada	C$23.2 billion, Q2 2026
Canadian direct investment abroad	C$14.4 billion, Q2 2026

Download this graphic (SVG, with source)

Foreign direct investment into Canada and Canadian direct investment abroad are each large and volatile. Over the last four quarters inflows were C$80.3 billion and outflows C$93.3 billion. In the last twenty quarters the totals were C$390.3 billion in and C$595.3 billion out. Canadian companies have invested more abroad than foreigners have invested in Canada over that span.

Across geographies

Investment in new building construction (current dollars, not seasonally adjusted) in the latest month changed from a year earlier by 3.6% in Ontario, 2.4% in Quebec, 1.5% in British Columbia and -1.7% in Alberta; the monthly value was C$4.4 billion in Ontario and C$1.9 billion in Alberta. City-level construction investment is not held here.

The data behind this graphic

Building construction investment, latest month against a year earlier

Item	Value
Saskatchewan	31.3%
Newfoundland and Labrador	26.9%
Prince Edward Island	25.3%
Manitoba	6.4%
Ontario	3.6%
New Brunswick	2.8%
Quebec	2.4%
British Columbia	1.5%
Alberta	-1.7%
Nova Scotia	-22.7%

Download this graphic (SVG, with source)

In the latest month, building investment was 31.3% higher than a year earlier in Saskatchewan, 26.9% in Newfoundland and Labrador and 25.3% in Prince Edward Island, and lower in Nova Scotia (-22.7%) and Alberta (-1.7%). Monthly provincial building values are volatile and the small provinces swing most, so this chart should be read as a snapshot of the latest month, not as a trend. Over the longer window from August 2019, Saskatchewan (148.9%) and New Brunswick (144.8%) show the largest increases and British Columbia (33.7%) among the smallest, though these are current-dollar values and construction costs have risen.

Investment by industry and province comes from Statistics Canada's capital and repair expenditures survey, which publishes intentions and actuals annually. This report holds it only for oil and gas, mining and support activities, nationally and for Alberta, in current dollars, so this section cannot say which other provinces or industries are investing more.

What it influences

Investment affects employment and construction activity in the short run and productive capacity in the long run. Any link between business investment and employment or housing costs is only a hypothesis; the link from oil prices to energy-sector investment has the Bank of Canada as its source.

Risks

What it means for decisions

For a business owner. If investment has been flat in real terms for twelve years, a firm that has raised its investment has gone against the national trend. Where firms are choosing between structures and intangible assets, the national data show where the money went: software and research have grown, buildings have not.

For a lender or investor. The investment story is less concentrated than it was. The decline of oil and gas as a share of capital spending, from 28.0% to 10.9%, means that national investment figures now depend less on one commodity price than they once did, and more on the combined behaviour of manufacturing, services, utilities and others. The data in this report do not split that group further.

For a resident. The effect appears later, in productivity and in wages: a smaller addition to the capital stock means less to work with, and the productivity section shows what has happened to output per hour.

What the data cannot say. Investment by industry is available here only for oil and gas, mining and support activities, and in current dollars. Investment in data centres, clean power and manufacturing is not separated. Building values are nominal and include price changes.

What to watch

The third-quarter national accounts (business investment by asset), monthly building construction investment, and the annual capital and repair expenditures intentions.

What happened

On a national-accounts basis, governments together borrowed C$53.8 billion over the four quarters to Q2 2026 (net lending of -C$53.8 billion). The federal government's net lending was -C$50.5 billion over that period and the provincial and territorial governments' was -C$53.6 billion. In the latest quarter alone the consolidated sector's balance was C$4.8 billion, the federal government's -C$4.5 billion and the provinces' -C$12.5 billion. Consolidated revenue over the year was C$1.40 trillion and expense C$1.42 trillion.

Federal net debt was C$1.47 trillion in March 2026, 4.9% more than a year earlier and 86.8% more than at the end of 2019, when it was C$785.6 billion.

The interest bill: a dollar story and a share story

The federal government paid C$13.9 billion in interest in the latest quarter, the highest in a series that begins in 1990. In the second quarter of 2019, the same quarter seven years earlier (the series is not seasonally adjusted), the figure was C$7.0 billion, and its low was C$5.0 billion in Q2 2020. Over the past year interest expense totalled C$52.2 billion.

The data behind this graphic

Federal interest expense as a share of federal revenue

Item	Value
1995 Q2	34.0%
2000 Q2	21.5%
2005 Q2	14.8%
2010 Q2	12.3%
Latest (2026 Q2)	10.5%
2023 Q2	9.0%
2015 Q2	9.0%
2019 Q2	8.3%

Download this graphic (SVG, with source)

Revenue has grown too, so the burden is better measured as a share of revenue. At 10.5% it is higher than in the same quarter of 2019 (8.3%) and higher than its low of 5.0% in the first quarter of 2021. But it is far lower than in the 1990s, when it was 34.0% in the second quarter of 1995, and lower than in 2000 (21.5%). The dollar figure is at a record because the debt is large and rates are higher than in the 2010s. The share is moderate because revenue is also much larger in nominal terms, C$132.4 billion a quarter now against C$51.5 billion in the same quarter of 2000.

Two readings are possible, and the data do not decide between them. One is that the burden has returned to a level that is manageable by the standard of the past thirty years. The other is that it has risen steeply since its low in early 2021, and that it may keep rising as maturing debt is refinanced at higher rates. The Department of Finance statement later in the section attributes the rise in debt charges mainly to higher effective interest rates on a larger stock of marketable bonds. This report holds no projection of that path.

Who is borrowing

Governments together borrowed C$53.8 billion over the year, on the national-accounts basis. That is more than the gap between revenue and expense, because net borrowing also counts governments' net investment in non-financial assets such as buildings and infrastructure. Federal borrowing was C$50.5 billion and provincial and territorial borrowing C$53.6 billion. They do not add to the consolidated figure because the consolidated total also includes local governments and the public pension plans; the difference between the sum of the two and the total is their combined net lending.

The data behind this graphic

Net lending (+) or borrowing (−) by level of government

Series	Latest
Federal	-C$4.5 billion, Q2 2026
Provincial and territorial	-C$12.5 billion, Q2 2026

Download this graphic (SVG, with source)

The pandemic quarter dominates the long chart: federal net lending reached -C$98.0 billion in Q2 2020. The recent change is that the provinces have caught up with the federal government. Their worst quarter in ten years was -C$23.9 billion in Q1 2025, and over the past year they borrowed slightly more than Ottawa did. Over five years the consolidated total is -C$121.8 billion.

The data behind this graphic

What has grown over seven years

Item	Value
Federal interest expense	98.3%
Government expense	56.1%
Government revenue	55.3%
Nominal GDP	48.6%

Download this graphic (SVG, with source)

The chart for the whole period makes the arithmetic plain. Federal interest expense has grown far faster than anything else. Because the government series are not seasonally adjusted, the chart compares the latest quarter with the same quarter seven years earlier: consolidated expense is up 56.1% and revenue 55.3%, against 48.6% for nominal GDP. Federal net debt rose because the federal government borrowed in almost every quarter of the period. Net debt itself is C$1.47 trillion, from C$785.6 billion at the end of 2019 and C$685.1 billion at the end of 2014, so it has grown by more than nominal GDP over both periods.

The data behind this graphic

Federal net debt

Series	Latest
Net debt	C$1.47 trillion, March 2026

Download this graphic (SVG, with source)

Why it happened

The deficit depends on both spending and revenue. Against the same quarter seven years earlier, consolidated expense is up 56.1% and revenue 55.3%, against nominal GDP 48.6%. Spending has therefore grown faster than the economy, and faster than revenue. The largest deficit of the period was -C$105.0 billion in Q2 2020. This report does not decompose the changes in revenue and expense into policy changes, inflation and growth.

Interest costs have risen with borrowing and rates: federal interest expense over the year was C$52.2 billion. The latest quarter's bill was 98.3% higher than in the same quarter seven years earlier and 10.5% of that quarter's federal revenue.

Why it matters

Debt service competes with other spending and with tax cuts, and a rising interest bill makes governments more sensitive to interest rates and to the growth of nominal GDP. Whether the position is sustainable depends on interest rates relative to nominal growth and on projected primary balances, which require forecasts that this report does not hold. It is therefore not assessed here and nothing in this section should be read as a judgement on sustainability.

History

Federal net lending's worst quarter in the past twenty years was -C$98.0 billion in Q2 2020. The provinces' worst quarter in the past ten years was -C$23.9 billion in Q1 2025. Five-year consolidated net lending totals -C$121.8 billion.

Across geographies

The data behind this graphic

Net operating balance as a share of revenue, fiscal year that began in April 2024

Item	Value
Alberta	8.5%
New Brunswick	4.1%
Nova Scotia	3.0%
Newfoundland and Labrador	-3.3%
Manitoba	-7.7%
Prince Edward Island	-7.8%
Ontario	-8.2%
Quebec	-9.2%
Saskatchewan	-9.3%
British Columbia	-16.5%

Download this graphic (SVG, with source)

The stock of liabilities tells a different story. Alberta’s liabilities are 1.48 times its revenue, among the lowest of the ten, while Manitoba’s are 3.43 times, Newfoundland and Labrador’s 2.95 and Quebec’s 2.88. British Columbia has the largest deficit and a relatively low multiple (1.82): so the large deficit sits on a comparatively small stock. These are gross liabilities, which include pension and other obligations and are not offset by financial assets, so they overstate net debt.

The data behind this graphic

Liabilities as a multiple of annual revenue, fiscal year that began in April 2024

Item	Value
Manitoba	3.43
Newfoundland and Labrador	2.95
Quebec	2.88
Ontario	2.39
New Brunswick	2.14
Saskatchewan	2.12
British Columbia	1.82
Prince Edward Island	1.51
Alberta	1.48
Nova Scotia	1.30

Download this graphic (SVG, with source)

Provincial public finances are available as a recent aggregate: the provincial and territorial governments' net lending was -C$53.6 billion over the four quarters to Q2 2026, on Statistics Canada's national-accounts basis. Province-by-province accounts are available only for past fiscal years. In the fiscal year that began in April 2024, the net operating balance of the provincial government was C$6.9 billion in Alberta (8.5% of its revenue), -C$16.9 billion in Ontario, -C$15.9 billion in Quebec and -C$13.2 billion in British Columbia; liabilities were C$492.0 billion in Ontario, C$495.5 billion in Quebec, C$144.6 billion in British Columbia and C$120.6 billion in Alberta. These are accounts for a past year, not the current budget. One sourced sensitivity, for Alberta's oil revenue, is given below.

What it influences

Government deficits and debt feed into interest costs, tax policy and the scope for public investment. The Department of Finance links interest rates on its marketable debt to the federal interest bill, and the sensitivity in Alberta's budget is the only quantified estimate of the oil price's effect in this section.

Risks

Higher interest rates on renewed debt, slower nominal growth and lower commodity-related revenue raise deficits; the opposite eases them.

What it means for decisions

For a taxpayer or voter. The share of federal revenue that goes to interest is the number that matters, not the dollar figure. At 10.5% it is about a tenth of every revenue dollar. It was about a third in 1995.

For a bondholder or lender. Federal net debt has grown faster than nominal GDP. The ratio of interest to revenue is the one to watch, because it shows whether the interest bill is crowding out other spending. This report does not hold fiscal projections and does not assess sustainability.

For a business that sells to government or depends on it. Provincial positions differ widely. A province with a surplus and low liabilities relative to revenue (Alberta) has more fiscal room than one with a large deficit (British Columbia), but Alberta’s revenue depends on oil prices, as the sensitivity in the Alberta budget shows, and a fall in prices could change the picture quickly.

For a policymaker. A larger interest bill reduces room for other spending or tax cuts. The levers are the growth of nominal GDP, the primary balance and interest rates, and they interact; the productivity section is relevant because faster real growth raises nominal GDP and, through revenue, the primary balance.

What the data cannot say. The statistical-basis provincial figures are for a past year and are not each province’s current budget. The national-accounts figures differ from the Fiscal Monitor’s budgetary figures. Nothing here is a judgement on whether debt is sustainable.

What to watch

Quarterly government finance statistics for the third quarter, the monthly Fiscal Monitor, federal and provincial budget and fiscal updates, and the yield on long-term government bonds.

What happened

Services-producing industries account for roughly 75.0% of real output (an approximation, since chained-dollar components are not additive). Among the largest individual industries, real estate and leasing is 13.2%, manufacturing 8.6%, health care 8.1%, finance and insurance 7.7%, construction 7.3%, public administration 7.2%, mining and oil and gas 5.2% and agriculture and forestry 1.9%. Over the year to August 2026, real output changed by 3.4% in information and cultural industries, 5.5% in transportation and warehousing, 2.6% in finance and insurance, 1.5% in oil and gas extraction, 1.8% in manufacturing, -0.1% in professional services and -7.7% in agriculture, forestry, fishing and hunting.

The shape of the economy

Canadian industry is easier to understand by size, then by change. At the latest reading the largest single industry is real estate and leasing, at C$313.2 billion a year in real output, which includes the imputed rent that homeowners are deemed to pay themselves. Manufacturing is second at C$202.9 billion, followed by health care (C$193.2 billion), finance and insurance (C$183.1 billion) and construction (C$172.4 billion). Mining, quarrying and oil and gas is C$123.4 billion, of which oil and gas extraction is C$79.3 billion.

The data behind this graphic

Real output by industry, latest month

Item	Value
Real estate and leasing	C$313.2 billion
Manufacturing	C$202.9 billion
Health care	C$193.2 billion
Finance	C$183.1 billion
Construction	C$172.4 billion
Public administration	C$170.1 billion
Professional services	C$168.3 billion
Wholesale trade	C$128.5 billion
Retail trade	C$127.5 billion
Mining, oil, gas	C$123.4 billion
Transport	C$108.2 billion
Information	C$83.4 billion
Hospitality	C$53.7 billion
Utilities	C$48.6 billion
Agriculture, forestry, fishing	C$44.2 billion

Download this graphic (SVG, with source)

Growth since 2019 sorted industries into two groups, as the opening chart shows. Finance (25.9%), professional services (25.5%) and health (21.3%) have led. Manufacturing is the only large industry smaller than it was (-4.2%). Transportation and warehousing (6.2%) and utilities (4.6%) have grown least among the industries that grew.

Twenty-six years of structural change

The 2019 starting point hides a longer pattern. Back to January 2000, real output across the economy is up 70.7%. The services-producing side is up 90.6% and the goods-producing side 33.5%.

The data behind this graphic

Goods-producing and services-producing output

Series	Latest
Services-producing	C$1.78 trillion, August 2026
Goods-producing	C$600.3 billion, August 2026

Download this graphic (SVG, with source)

The data behind this graphic

Real output by industry since January 2000

Item	Value
Finance and insurance	135.8%
Information and culture	122.0%
Professional services	119.6%
Real estate and leasing	104.7%
Construction	96.7%
Retail trade	96.0%
Wholesale trade	94.5%
Oil and gas extraction	87.5%
Health care	85.3%
Public administration	76.3%
Transportation and warehousing	73.4%
Accommodation and food	58.8%
Agriculture and forestry	43.3%
Utilities	40.5%
Manufacturing	-8.2%

Download this graphic (SVG, with source)

Three things stand out. First, the fastest growth has been in finance and insurance (135.8%), information and culture (122.0%) and professional services (119.6%), all more than doubling in real terms. Second, oil and gas extraction (87.5%) has outgrown the economy as a whole, a reminder that after the price slump of 2014–16 the industry did not shrink in volume terms, even though its capital spending did (see the investment section). Third, manufacturing is the only large industry that produces less than in January 2000 (-8.2%).

The jobs data add a second dimension. Total employment is up 44.0% over the same period, less than output, so output per job has risen.

The data behind this graphic

Where output and jobs have grown together, and where they have not

Item	Real output	Jobs
Professional services	119.6%	119.6%
Finance and insurance	135.8%	73.0%
Health care	85.3%	102.5%
Construction	96.7%	107.5%
Information and culture	122.0%	36.8%
Transportation	73.4%	44.0%
Accommodation and food	58.8%	26.6%
Utilities	40.5%	32.0%
Manufacturing	-8.2%	-18.1%
Agriculture	43.3%	-44.0%

Download this graphic (SVG, with source)

The pairs show how different industries have grown. In professional services output and jobs rose by about the same proportion (119.6% and 119.6%), so growth came from adding people. In finance and insurance, output rose 135.8%; jobs, on a wider measure that includes real estate and leasing, rose 73.0%, which suggests each job supports more output. In agriculture the contrast is the sharpest: output, which here includes forestry, fishing and hunting, is up 43.3% while agricultural jobs are down 44.0%. In manufacturing, output is -8.2% and jobs -18.1%: fewer people making slightly less. Health care and construction employ more than twice as many people as in 2000 (jobs up 102.5% and 107.5%), and their output has grown by less than that, which in these measures means less output per job than before. Whether that reflects measurement, mix or a real change in efficiency is not something these series can settle.

Since the end of 2019, the factory story is different again. Manufacturing output is 4.2% lower (to August 2026) while manufacturing employment is 6.6% higher (to October 2026). Real output has fallen and employment has risen, which is the opposite of the 2000s pattern. The two are measured in different surveys and the monthly employment series is noisy, so this is a caution, not a conclusion.

The data behind this graphic

Manufacturing output and manufacturing sales

Series	Latest
Sales, current dollars	C$79.9 billion, August 2026
Output, real dollars	C$202.9 billion, August 2026

Download this graphic (SVG, with source)

The chart makes a point about nominal figures. Manufacturing sales in current dollars have grown much faster than real output because prices have risen; a reader who sees only the sales figure would conclude that the sector has grown, while the real series shows it has not.

Energy and mining

Crude oil production was 154.0 million barrels in August 2026, 1.9% higher than a year earlier and close to the high of the available history since 2016 (156.5 million barrels in December 2025). Alberta produced 84.8% of it. Oil and gas extraction output is 15.8% above its end-2019 level. Energy products were 26.3% of goods exports in September 2026 (see the trade section).

Alberta's pattern differs from the headline numbers: the resource group (forestry, fishing, mining, quarrying and oil and gas) employed 5.6% of Alberta's workers in 2025, but energy products were 76% of its domestic exports that year. In Newfoundland and Labrador the employment share was 6.4% and the export share 54%. Measured by jobs, resource industries are small; measured by export value (and, for governments, by royalties), they are large.

Output of metals and minerals, potash, critical minerals and coal is not separately measured in the series held here; the value of shipments by mineral is shown below.

Manufacturing

Manufacturing sales (shipments) were C$79.9 billion in August 2026, 14.2% higher than a year earlier, with new orders 12.1% higher and unfilled orders 17.0% higher. Manufacturing employment changed by -0.9% over the year to October 2026 and real manufacturing output by 1.8% over the year to August 2026. Nominal sales are rising much faster than real output because prices are rising:

Manufacturing's share of employment in 2025 was 10.8% in Quebec and 10.0% in Ontario, the provinces most exposed to US demand for manufactured goods (the exposure edges in the explorer record the export shares). Motor vehicles and parts exports were C$7.3 billion in September 2026, a change of -1.4% from a year earlier.

Agriculture and food

Agriculture, forestry, fishing and hunting output changed by -7.7% over the year to August 2026 and employment in agriculture by -3.9% over the year to October 2026. Exports of farm, fishing and intermediate food products were C$5.5 billion in September 2026, a change of 18.9% from a year earlier. In 2025 agriculture was 4.6% of Saskatchewan's employment, 2.0% of Manitoba's and 1.5% of Alberta's. Crop yields and Canadian input costs are not held here; the agriculture section covers world grain prices.

Construction, transportation, finance, technology and tourism

Construction output is 7.5% above its end-2019 level and employment 9.1% higher; the value of non-residential building construction in the latest month changed by 10.3% from a year earlier (see the investment section). Transportation and warehousing output changed by 5.5%; the Transportation Services Index changed by 2.0% and rail carloadings by 8.0% from a year earlier, to 32.8 million tonnes in August 2026.

The information and communication technology sector produces 6.0% of real output; its output changed by 3.0% over the year and is 30.5% higher than at the end of 2019. Professional, scientific and technical services output changed by -0.1% and employment by 1.2%. Investment in data centres, artificial intelligence and clean technology is not measured separately and is not assessed here.

Finance and insurance output changed by 2.6% over the year, and finance, insurance, real estate and leasing employment changed by -1.8%. Finance, insurance, real estate and leasing was 8.7% of Ontario's employment in 2025.

Tourism is not a single industry in these series. A partial indicator is non-resident visitors entering Canada: 4.2 million in August 2026, 4.6% more than a year earlier, of which residents of the United States are 77.8%. Seasonal patterns make comparisons with other months misleading, so only same-month comparisons are shown.

Electricity generation was 51.6 TWh in August 2026, 8.5% higher than a year earlier: hydro is 54.8%, combustible fuels 26.3%, nuclear 12.4%, wind 5.1% and solar 1.3%. Wind generation changed by 1.0% and solar by -7.0% from the same month a year earlier.

Statistics Canada publishes the value of mineral shipments by mineral. The value of shipments (current dollars, so prices are included) was C$773 million for copper in August 2026, C$992 million for potash, C$309 million for nickel, C$38 million for cobalt, C$21 million for zinc and C$3.1 billion for gold. Copper's value of shipments is 122.8% higher than a year earlier and nickel's 38.3%; the copper price went from US$9,670 a tonne in August 2025 to US$14,326 in August 2026, so part of the change is price, and the volume effect cannot be separated here. This report does not say which of these minerals the federal government classes as critical, because it does not hold that list, and it has no data on mine output volumes, processing capacity or exploration spending.

Why it happened

Industry outputs can respond to different drivers: oil and gas to commodity prices and capital spending, manufacturing to export demand and inputs, construction to financing and approvals, and services to household incomes. The data here record that each industry moved differently over the year; they do not establish causes. The explorer records documented exposures between industries and provinces.

Why it matters

Industry structure determines how regions respond to shocks. The employment shares above show that provinces differ in what they produce, and the export shares show how differently the same shock would reach them. A shock to oil prices would reach Alberta, Saskatchewan and Newfoundland and Labrador through export values and public revenue much more than through employment, while a shock to US demand for manufactured goods would reach Ontario and Quebec through factory jobs.

History

Oil and gas extraction is 15.8% above its end-2019 level and manufacturing -4.2% relative to it. Manufacturing sales are 39.6% higher in current dollars than at the end of 2019, which is far more than real output, again reflecting prices. Crude oil production has risen 31.1% since January 2016.

Across geographies

Resource-heavy economies (Alberta, Saskatchewan, Newfoundland and Labrador) and manufacturing-heavy economies (Ontario, Quebec) face different drivers, as the employment and export shares above show. City-level industry structure is not published on a basis comparable with these provincial figures and is addressed, where possible, in the city reports.

What it influences

Industry output feeds employment, exports, investment and public revenue. The oil price has sourced links to energy investment, national income, consumer energy prices and exports, and the explorer connects industries to provinces through measured exposures.

Energy-related output is exposed to world prices: Brent crude was US$125.00 a barrel in October 2026 and the European gas price US$28.50 per million BTU; the Global report's energy section covers the world market and the commodities section metals, potash and grains.

Risks

What it means for decisions

For an employer or investor. The sectors that have grown fastest in real terms since 2000 are finance, information and professional services. The one large sector that has shrunk, manufacturing, is exposed to trade and exchange-rate risk. Both facts matter for hiring and location decisions.

For a policymaker. Output growth and job growth are different goals. Agriculture, manufacturing and finance have increased output per job; health care, construction and public services have added many jobs. Policy that targets one of these does not automatically produce the other. The shift away from goods production and towards services also changes how shocks arrive: services are less traded, so a drop in US demand matters less for them than for factories, and they depend more on domestic demand.

For a region. The industry shares in this section show how much provinces differ in what they produce. These series do not measure how much of a province's growth comes from its mix of industries. The provincial comparisons later in the report use these shares.

What the data cannot say. The series are real output and employment, not profits, wages or investment. They do not say whether the shift into services has been good or bad for living standards; they say that it has happened. The chained-dollar shares are approximate, and the latest employment month is later than the latest output month, so the two sides of each pair do not end on the same date.

What to watch

Monthly GDP by industry (next release 30 November 2026), the manufacturing survey, crude oil production, auto sector exports and the effect of tariff changes on manufacturing orders.

What happened

Real output of agriculture, forestry, fishing and hunting was C$44.2 billion at an annual rate in August 2026, -7.7% on a year earlier and 7.7% above its level at the end of 2019. Employment in agriculture was 222 thousand in October 2026, -3.9% over the year.

Merchandise exports of farm, fishing and intermediate food products were C$5.5 billion in September 2026, 7.0% of all merchandise exports and 18.9% on a year earlier.

More from fewer

The most striking thing about Canadian agriculture is how much has changed in how it is done. Real output in January 2000 was C$30.9 billion a year; it is C$44.2 billion now, up 43.3%. Employment in agriculture was 396 thousand in January 2000 and is 222 thousand, down 44.0%. On these two measures, which differ in scope, output per job is therefore more than two and a half times what it was.