The answer
The anatomy of Canadian living standards
Real GDP per person rose 1.6% over the four quarters to 2026 Q2 and is now 0.1% below its 2022 Q3 record. A falling population did more of the work than output per worker, which also added.
A growing economy does not always mean more output per person. This page separates the roles of output per worker, employment, participation and population structure. It shows how the measured change adds up; it does not establish what caused it.
What adds to the change — and what subtracts
2025 Q2 → 2026 Q2 · percentage points
Economic output, adjusted for price changes, for each employed person.
The anatomy calls this the employment-rate term. It divides employment by the labour force, not by the working-age population.
The labour force as a share of the Labour Force Survey population aged fifteen and over.
The survey population aged fifteen and over divided by the total resident population. This mixes age structure and survey coverage.
The answer: +1.57 percentage points
These are parts of the measured change, not estimates of its causes. Every term and the total use log changes, not ordinary percentage growth rates. Each displayed value is rounded separately.
What share of the change in real GDP per person comes from output per worker, the employment rate, participation, and population structure?
Next: The business-sector labour-productivity release for 2026 Q2, which would complete this window's output-per-worker split; then the next quarterly GDP release or demographic estimate that moves any term materially. A…
The answer
Real GDP per person rose 1.6 per cent over the four quarters to 2026 Q2, to $60,944 in chained 2017 dollars. That is the largest four-quarter gain since the four quarters ending 2022 Q3, and it leaves the per-person level 0.1 per cent below the record set in that same quarter. Of the 262 quarters this record holds a per-person figure for, going back to 1961, only two stand at or above today's level: 2022 Q3 and 2026 Q2.
The gain is not mainly a production story. Real GDP grew 1.1 per cent over those four quarters while the population fell 0.5 per cent, so a smaller denominator supplied part of the rise. In the identity's own terms, population structure contributed +1.21 percentage points of the +1.57-point total, output per worker +0.80, the employment rate +0.02, and participation -0.46.
Output per worker adding is a change from the two windows before this one, in which it subtracted. It last added over the four quarters to 2025 Q3. What cannot yet be said is where that gain came from: Statistics Canada has not yet published business-sector labour productivity or hours worked for 2026 Q2, so the split of output per worker into output per hour and hours per worker is unavailable for this window and is reported as unavailable rather than estimated.
This is the Brief's second reading. Its first reading, on evidence through 2026 Q1, said per-person output had barely moved and that the whole small gain was compositional. Two things changed that: a new quarter, and Statistics Canada's restatement of 2026 Q1 on 28 August 2026, which raised that quarter's real GDP by 0.12 per cent and by itself moved the previously published four-quarter reading from +0.19 to +0.31 percentage points.
| Real GDP per person (total change) | +0.31 pp |
|---|---|
| Output per worker | -0.57 pp |
| Employment rate | +0.28 pp |
| Participation | -0.77 pp |
| Population structure | +1.37 pp |
| — of which output per hour (measured index) | -0.60 pp |
| Identity residual (always exactly zero) | +0.00 pp |
| — of which hours worked per employed person (measured) | -0.34 pp |
| — of which national-accounts scope reconciliation (derived) | +0.36 pp |
Exact log-change identity: the change in ln(real GDP per person) equals the sum of the four term changes; the output-per-worker term is further split into the measured productivity index, measured hours worked per employed person, and a narrower derived reconciliation. Contributions are reported in percentage points; the parts sum to the total exactly, and that arithmetic is re-checked every time the record is rebuilt.
What this means for the decision
The per-person turn is real but young, and one quarter of it rests on a population that fell rather than output that grew. Treat it as a single quarter of evidence, not a trend, until the productivity and hours release for 2026 Q2 shows whether output per worker gained from producing more per hour or from working more hours. Plans premised on rising average living standards should still be tested against the decomposition, not the aggregate.
What would settle it
The business-sector labour-productivity release for 2026 Q2, which would complete this window's output-per-worker split; then the next quarterly GDP release or demographic estimate that moves any term materially. A revision that changes a prior reading re-issues the Brief with its correction history intact -- as the 28 August 2026 restatement of 2026 Q1 did.
Explore the visual record
Anyone deciding against 'growth' — a budget, an investment, a workforce plan — needs to know which arithmetic is doing the work: more output per worker, more people working, or simply more people. The identity locates the arithmetic exactly; it does not attribute causes.
Evidence reviewed through 2026-09-02 · 4 cited sources · 0 member Resolutions · 8 chapters
Refreshed at each quarterly GDP release and each quarterly demographic estimate — the two releases that move the identity's numerator and denominator. Monthly labour releases update the intervening terms but the answer is refreshed on the quarterly cycle.
This Brief answers the Big Question Can Canada raise living standards, not just its population? · How this Brief was made — sources, method and limits →
Chapter 1§
The latest window
The latest four-quarter window of the identity, stated plainly: per-person growth is positive, with population structure the largest single contributor and output per worker also adding. Every number in this section is computed by the Brief's named scenario from already-held official series.
Chapter 2§
The four-term anatomy
The identity's four terms — output per worker, employment rate, participation, population structure — with each term's contribution in percentage points. A term's contribution says where the change shows up in measurement, not what caused it.
Chapter 3§
The long run
How the latest window compares with the record's full history: real GDP per person from the earliest comparable observation, split into an era-by-era ladder of population growth and per-person growth. States whether the composition of growth — how much comes from more people versus more output per person — has changed across the record, without asserting why.
This line covers 1961 Q1 to 2026 Q2 on one basis (real GDP over population, both held that far back). It is not the same as the four-term decomposition below: full decomposition, with the measured-hours subsplit, is available only from 1991 Q4 onward. The illustrative windows beneath this chart state exactly which measures cover each earlier period.
View the data table for real GDP per person, quarterly (262 points)
| Period | Value |
|---|---|
| 1961 Q1 | $20,115 |
| 1961 Q2 | $20,534 |
| 1961 Q3 | $20,919 |
| 1961 Q4 | $21,098 |
| 1962 Q1 | $21,566 |
| 1962 Q2 | $21,647 |
| 1962 Q3 | $21,812 |
| 1962 Q4 | $22,129 |
| 1963 Q1 | $22,175 |
| 1963 Q2 | $22,406 |
| 1963 Q3 | $22,457 |
| 1963 Q4 | $23,042 |
| 1964 Q1 | $23,519 |
| 1964 Q2 | $23,509 |
| 1964 Q3 | $23,633 |
| 1964 Q4 | $23,635 |
| 1965 Q1 | $24,286 |
| 1965 Q2 | $24,412 |
| 1965 Q3 | $24,627 |
| 1965 Q4 | $25,104 |
| 1966 Q1 | $25,490 |
| 1966 Q2 | $25,933 |
| 1966 Q3 | $25,779 |
| 1966 Q4 | $25,972 |
| 1967 Q1 | $25,808 |
| 1967 Q2 | $26,245 |
| 1967 Q3 | $26,191 |
| 1967 Q4 | $26,181 |
| 1968 Q1 | $26,311 |
| 1968 Q2 | $26,936 |
| 1968 Q3 | $27,199 |
| 1968 Q4 | $27,463 |
| 1969 Q1 | $27,741 |
| 1969 Q2 | $27,772 |
| 1969 Q3 | $27,976 |
| 1969 Q4 | $28,260 |
| 1970 Q1 | $28,340 |
| 1970 Q2 | $28,259 |
| 1970 Q3 | $28,445 |
| 1970 Q4 | $28,387 |
| 1971 Q1 | $28,155 |
| 1971 Q2 | $28,899 |
| 1971 Q3 | $29,104 |
| 1971 Q4 | $29,319 |
| 1972 Q1 | $29,150 |
| 1972 Q2 | $29,900 |
| 1972 Q3 | $29,895 |
| 1972 Q4 | $30,501 |
| 1973 Q1 | $31,202 |
| 1973 Q2 | $31,418 |
| 1973 Q3 | $31,443 |
| 1973 Q4 | $31,911 |
| 1974 Q1 | $32,063 |
| 1974 Q2 | $32,287 |
| 1974 Q3 | $32,173 |
| 1974 Q4 | $31,866 |
| 1975 Q1 | $31,594 |
| 1975 Q2 | $31,902 |
| 1975 Q3 | $32,382 |
| 1975 Q4 | $32,619 |
| 1976 Q1 | $33,097 |
| 1976 Q2 | $33,716 |
| 1976 Q3 | $33,756 |
| 1976 Q4 | $33,690 |
| 1977 Q1 | $34,123 |
| 1977 Q2 | $34,203 |
| 1977 Q3 | $34,303 |
| 1977 Q4 | $34,736 |
| 1978 Q1 | $34,966 |
| 1978 Q2 | $35,124 |
| 1978 Q3 | $35,363 |
| 1978 Q4 | $35,606 |
| 1979 Q1 | $35,821 |
| 1979 Q2 | $36,212 |
| 1979 Q3 | $36,367 |
| 1979 Q4 | $36,494 |
| 1980 Q1 | $36,647 |
| 1980 Q2 | $36,522 |
| 1980 Q3 | $36,346 |
| 1980 Q4 | $36,718 |
| 1981 Q1 | $37,415 |
| 1981 Q2 | $37,720 |
| 1981 Q3 | $37,250 |
| 1981 Q4 | $36,930 |
| 1982 Q1 | $36,421 |
| 1982 Q2 | $35,911 |
| 1982 Q3 | $35,486 |
| 1982 Q4 | $35,052 |
| 1983 Q1 | $35,555 |
| 1983 Q2 | $36,174 |
| 1983 Q3 | $36,485 |
| 1983 Q4 | $36,841 |
| 1984 Q1 | $37,408 |
| 1984 Q2 | $38,048 |
| 1984 Q3 | $38,111 |
| 1984 Q4 | $38,600 |
| 1985 Q1 | $39,126 |
| 1985 Q2 | $39,171 |
| 1985 Q3 | $39,565 |
| 1985 Q4 | $40,055 |
| 1986 Q1 | $39,961 |
| 1986 Q2 | $40,103 |
| 1986 Q3 | $40,029 |
| 1986 Q4 | $39,607 |
| 1987 Q1 | $40,416 |
| 1987 Q2 | $40,794 |
| 1987 Q3 | $41,260 |
| 1987 Q4 | $41,643 |
| 1988 Q1 | $42,154 |
| 1988 Q2 | $42,406 |
| 1988 Q3 | $42,234 |
| 1988 Q4 | $42,301 |
| 1989 Q1 | $42,632 |
| 1989 Q2 | $42,616 |
| 1989 Q3 | $42,567 |
| 1989 Q4 | $42,288 |
| 1990 Q1 | $42,600 |
| 1990 Q2 | $42,269 |
| 1990 Q3 | $41,778 |
| 1990 Q4 | $41,233 |
| 1991 Q1 | $40,571 |
| 1991 Q2 | $40,658 |
| 1991 Q3 | $40,554 |
| 1991 Q4 | $40,495 |
| 1992 Q1 | $40,446 |
| 1992 Q2 | $40,366 |
| 1992 Q3 | $40,439 |
| 1992 Q4 | $40,523 |
| 1993 Q1 | $40,695 |
| 1993 Q2 | $40,969 |
| 1993 Q3 | $41,240 |
| 1993 Q4 | $41,270 |
| 1994 Q1 | $41,807 |
| 1994 Q2 | $42,312 |
| 1994 Q3 | $42,717 |
| 1994 Q4 | $42,885 |
| 1995 Q1 | $43,212 |
| 1995 Q2 | $43,127 |
| 1995 Q3 | $43,050 |
| 1995 Q4 | $43,086 |
| 1996 Q1 | $43,074 |
| 1996 Q2 | $43,282 |
| 1996 Q3 | $43,504 |
| 1996 Q4 | $43,704 |
| 1997 Q1 | $44,225 |
| 1997 Q2 | $44,623 |
| 1997 Q3 | $45,019 |
| 1997 Q4 | $45,320 |
| 1998 Q1 | $45,908 |
| 1998 Q2 | $45,854 |
| 1998 Q3 | $46,156 |
| 1998 Q4 | $46,663 |
| 1999 Q1 | $47,461 |
| 1999 Q2 | $47,766 |
| 1999 Q3 | $48,377 |
| 1999 Q4 | $48,909 |
| 2000 Q1 | $49,619 |
| 2000 Q2 | $50,092 |
| 2000 Q3 | $50,454 |
| 2000 Q4 | $50,392 |
| 2001 Q1 | $50,609 |
| 2001 Q2 | $50,626 |
| 2001 Q3 | $50,412 |
| 2001 Q4 | $50,538 |
| 2002 Q1 | $51,211 |
| 2002 Q2 | $51,376 |
| 2002 Q3 | $51,644 |
| 2002 Q4 | $51,776 |
| 2003 Q1 | $52,017 |
| 2003 Q2 | $51,826 |
| 2003 Q3 | $51,867 |
| 2003 Q4 | $52,069 |
| 2004 Q1 | $52,383 |
| 2004 Q2 | $52,889 |
| 2004 Q3 | $53,356 |
| 2004 Q4 | $53,575 |
| 2005 Q1 | $53,701 |
| 2005 Q2 | $53,978 |
| 2005 Q3 | $54,459 |
| 2005 Q4 | $54,812 |
| 2006 Q1 | $55,186 |
| 2006 Q2 | $55,086 |
| 2006 Q3 | $55,070 |
| 2006 Q4 | $55,103 |
| 2007 Q1 | $55,392 |
| 2007 Q2 | $55,811 |
| 2007 Q3 | $55,845 |
| 2007 Q4 | $55,727 |
| 2008 Q1 | $55,687 |
| 2008 Q2 | $55,760 |
| 2008 Q3 | $56,015 |
| 2008 Q4 | $55,157 |
| 2009 Q1 | $53,824 |
| 2009 Q2 | $53,109 |
| 2009 Q3 | $53,160 |
| 2009 Q4 | $53,573 |
| 2010 Q1 | $54,141 |
| 2010 Q2 | $54,293 |
| 2010 Q3 | $54,488 |
| 2010 Q4 | $54,895 |
| 2011 Q1 | $55,258 |
| 2011 Q2 | $55,255 |
| 2011 Q3 | $55,833 |
| 2011 Q4 | $56,084 |
| 2012 Q1 | $56,024 |
| 2012 Q2 | $56,083 |
| 2012 Q3 | $55,963 |
| 2012 Q4 | $55,883 |
| 2013 Q1 | $56,307 |
| 2013 Q2 | $56,510 |
| 2013 Q3 | $56,773 |
| 2013 Q4 | $57,158 |
| 2014 Q1 | $57,193 |
| 2014 Q2 | $57,595 |
| 2014 Q3 | $57,960 |
| 2014 Q4 | $58,162 |
| 2015 Q1 | $57,813 |
| 2015 Q2 | $57,597 |
| 2015 Q3 | $57,645 |
| 2015 Q4 | $57,491 |
| 2016 Q1 | $57,754 |
| 2016 Q2 | $57,309 |
| 2016 Q3 | $57,678 |
| 2016 Q4 | $57,767 |
| 2017 Q1 | $58,404 |
| 2017 Q2 | $58,883 |
| 2017 Q3 | $58,663 |
| 2017 Q4 | $58,676 |
| 2018 Q1 | $59,230 |
| 2018 Q2 | $59,527 |
| 2018 Q3 | $59,555 |
| 2018 Q4 | $59,396 |
| 2019 Q1 | $59,415 |
| 2019 Q2 | $59,905 |
| 2019 Q3 | $59,783 |
| 2019 Q4 | $59,621 |
| 2020 Q1 | $58,304 |
| 2020 Q2 | $51,783 |
| 2020 Q3 | $56,455 |
| 2020 Q4 | $57,517 |
| 2021 Q1 | $58,450 |
| 2021 Q2 | $58,255 |
| 2021 Q3 | $59,274 |
| 2021 Q4 | $59,990 |
| 2022 Q1 | $60,447 |
| 2022 Q2 | $60,933 |
| 2022 Q3 | $61,020 |
| 2022 Q4 | $60,539 |
| 2023 Q1 | $60,840 |
| 2023 Q2 | $60,591 |
| 2023 Q3 | $60,134 |
| 2023 Q4 | $59,543 |
| 2024 Q1 | $59,598 |
| 2024 Q2 | $59,695 |
| 2024 Q3 | $59,787 |
| 2024 Q4 | $59,870 |
| 2025 Q1 | $60,182 |
| 2025 Q2 | $59,993 |
| 2025 Q3 | $60,208 |
| 2025 Q4 | $60,169 |
| 2026 Q1 | $60,368 |
| 2026 Q2 | $60,944 |
Before 1991 — illustrative context, not a fifth era. Full four-term decomposition, with the measured-hours subsplit, is available from 1991 (the fully aligned floor across all seven core series). Earlier periods provide long-run context using the longest comparable measures available: real GDP and population reach back to the start of the real GDP series; the four-term identity (without the hours subsplit) is additionally available for any window entirely within the labour force survey's own floor.
1961–1970 (1961 Q1–1970 Q4)Per person +3.53 pp/yr
- Period
- 1961–1970 (1961 Q1–1970 Q4)
- Real GDP
- +5.25 pp/yr
- Population
- +1.72 pp/yr
- Per person
- +3.53 pp/yr
- Coverage
- Real GDP and population only
1970–1980 (1970 Q4–1980 Q4)Per person +2.57 pp/yr
- Period
- 1970–1980 (1970 Q4–1980 Q4)
- Real GDP
- +3.97 pp/yr
- Population
- +1.40 pp/yr
- Per person
- +2.57 pp/yr
- Coverage
- Real GDP and population only
1980–1991 (1980 Q4–1991 Q4)Per person +0.89 pp/yr
- Period
- 1980–1991 (1980 Q4–1991 Q4)
- Real GDP
- +2.11 pp/yr
- Population
- +1.22 pp/yr
- Per person
- +0.89 pp/yr
- Coverage
- Also the four-term identity (no hours subsplit): output per worker +0.76 pp/yr, employment rate -0.30, participation +0.28, population structure +0.15
| Period (illustrative, before 1991) | Real GDP | Population | Real GDP per person | Coverage |
|---|---|---|---|---|
| 1961–1970 (1961 Q1–1970 Q4) | +5.25 pp/yr | +1.72 pp/yr | +3.53 pp/yr | Real GDP and population only |
| 1970–1980 (1970 Q4–1980 Q4) | +3.97 pp/yr | +1.40 pp/yr | +2.57 pp/yr | Real GDP and population only |
| 1980–1991 (1980 Q4–1991 Q4) | +2.11 pp/yr | +1.22 pp/yr | +0.89 pp/yr | Also the four-term identity (no hours subsplit): output per worker +0.76 pp/yr, employment rate -0.30, participation +0.28, population structure +0.15 |
Chapter 4§
The four eras
We repeat the same four-part calculation for four chosen periods, with the dates shown below. The final period runs to the latest available quarter. Showing the total and each contribution as annual rates lets us compare periods of different lengths.
The extra checks focus on total growth in GDP per person. One compares annual growth before and after the global financial crisis, over the periods shown below.
Another check moves each boundary between the four periods earlier and later, one at a time. This shows how much the GDP-per-person comparison depends on the chosen quarter. Results that fall within the pandemic shock are marked separately because the shock affects that comparison.
1991–2000 (1991 Q4–2000 Q4)Per person +2.43 pp/yr
- Era
- 1991–2000 (1991 Q4–2000 Q4)
- Per person
- +2.43 pp/yr
- Output per worker
- +1.84 pp/yr
- Employment rate
- +0.39 pp/yr
- Participation
- -0.05 pp/yr
- Population structure
- +0.25 pp/yr
2000–2010 (2000 Q4–2010 Q4)Per person +0.86 pp/yr
- Era
- 2000–2010 (2000 Q4–2010 Q4)
- Per person
- +0.86 pp/yr
- Output per worker
- +0.50 pp/yr
- Employment rate
- -0.11 pp/yr
- Participation
- +0.15 pp/yr
- Population structure
- +0.32 pp/yr
2010–2019 (2010 Q4–2019 Q4)Per person +0.92 pp/yr
- Era
- 2010–2019 (2010 Q4–2019 Q4)
- Per person
- +0.92 pp/yr
- Output per worker
- +0.78 pp/yr
- Employment rate
- +0.30 pp/yr
- Participation
- -0.20 pp/yr
- Population structure
- +0.04 pp/yr
2019–present (2019 Q4–2026 Q2)Per person +0.34 pp/yr
- Era
- 2019–present (2019 Q4–2026 Q2)
- Per person
- +0.34 pp/yr
- Output per worker
- +0.27 pp/yr
- Employment rate
- -0.22 pp/yr
- Participation
- -0.19 pp/yr
- Population structure
- +0.47 pp/yr
| Era | Per person (annualized) | Output per worker | Employment rate | Participation | Population structure |
|---|---|---|---|---|---|
| 1991–2000 (1991 Q4–2000 Q4) | +2.43 pp/yr | +1.84 pp/yr | +0.39 pp/yr | -0.05 pp/yr | +0.25 pp/yr |
| 2000–2010 (2000 Q4–2010 Q4) | +0.86 pp/yr | +0.50 pp/yr | -0.11 pp/yr | +0.15 pp/yr | +0.32 pp/yr |
| 2010–2019 (2010 Q4–2019 Q4) | +0.92 pp/yr | +0.78 pp/yr | +0.30 pp/yr | -0.20 pp/yr | +0.04 pp/yr |
| 2019–present (2019 Q4–2026 Q2) | +0.34 pp/yr | +0.27 pp/yr | -0.22 pp/yr | -0.19 pp/yr | +0.47 pp/yr |
Every column is the identity’s own annualized log change for that era, in percentage points per year; the four right-hand columns sum to the per-person column by construction.
Global financial crisis comparison: 2000 Q4–2008 Q3, +1.36 pp/yr; 2008 Q3–2019 Q4, +0.56 pp/yr.
Wording corrected September 7, 2026: the earlier label incorrectly said this comparison ended in 2010. The recorded periods and values are unchanged.
Boundary sensitivity: each era boundary shifted two quarters earlier and later, one at a time
| Boundary | Shift | Era before | Era after | Note |
|---|---|---|---|---|
| 2000 Q4 | -2 quarters | +2.50 pp/yr | +0.87 pp/yr | — |
| 2000 Q4 | +2 quarters | +2.35 pp/yr | +0.85 pp/yr | — |
| 2010 Q4 | -2 quarters | +0.79 pp/yr | +0.98 pp/yr | — |
| 2010 Q4 | +2 quarters | +0.88 pp/yr | +0.90 pp/yr | — |
| 2019 Q4 | -2 quarters | +1.03 pp/yr | +0.25 pp/yr | — |
| 2019 Q4 | +2 quarters | -0.61 pp/yr | +2.71 pp/yr | Lands inside the record's documented COVID-19 shock window -- not a fair boundary test, reported rather than smoothed over. |
Chapter 5§
The post-2019 break
The current era set against its predecessors: population growth running faster than in any earlier era while per-person growth runs slower than in any earlier era. Locates the record's per-person peak and states how the latest level compares with it, including how long the level has gone without a new peak. Decomposes the change from that peak to the latest quarter into the same four terms used throughout, so the peak-to-latest movement is read on the identical basis as every era above it.
| Record per-person peak | 2022 Q3 ($61,020 chained) |
|---|---|
| Latest quarter shown | 2026 Q2 ($60,944 chained, -0.12% against the peak) |
| Latest level first reached | 2022 Q3 |
| Quarters since a new peak | 14 quarters and counting |
| Term | Cumulative change since the peak |
|---|---|
| Per person, peak to latest (cumulative) | -0.12 pp |
| Output per worker | -0.09 pp |
| Employment rate | -2.21 pp |
| Participation | -0.46 pp |
| Population structure | +2.64 pp |
Cumulative log-change contributions from the peak quarter to the latest quarter, in percentage points; the four terms sum to the total by construction.
From the economy to the household · Q4 2025
An average can hide who is falling behind
Canada’s estimated average net saving was positive. 3 of the five income groups had negative averages.
In October–December 2025, estimated net saving averaged $1,913 per household, compared with $2,068 in the same quarter of 2024. The group averages below show why that national figure cannot describe everyone.
Current dollars per household, for one quarter; not seasonally adjusted. These are Statistics Canada estimates of group means, not medians or a count of households in difficulty. Comparing Q4 with Q4 reduces seasonal differences. Dollar changes have not been adjusted for inflation.
Net saving by income group
Each fifth contains 20% of households, ranked by disposable income adjusted for household size. A negative group average does not mean every household in that group spent more than it earned. Households can move between groups over time; this is not a fixed panel.
An estimation limit: quarterly spending distributions use annual spending patterns and assume the same proportional change within each spending category across household groups. Actual groups may have adjusted spending differently, which also affects estimated saving.
What goes into the national average?
Disposable income
$27,356− Consumption
$26,336+ Pension adjustment
$893= Net saving
$1,913Terms are rounded to the nearest dollar and may not sum exactly.
Net saving is an accounting measure, not cash left in a bank account. It adds the change in pension entitlements to disposable income minus consumption. Consumption also includes estimated services such as the housing services owners receive from their own homes. A negative number does not establish whether households borrowed, sold assets or drew on other resources.
Compare the exact income and saving figures
| Income group | Income, Q4 2025 ($) | Consumption ($) | Pension adjustment ($) | Net saving ($) | Net saving, Q4 2024 ($) | Change ($) |
|---|---|---|---|---|---|---|
| All households | $27,356 | $26,336 | $893 | $1,913 | $2,068 | −$155 |
| Lowest income fifth | $8,185 | $18,627 | −$374 | −$10,816 | −$10,095 | −$721 |
| Second income fifth | $16,858 | $21,461 | −$919 | −$5,523 | −$4,883 | −$640 |
| Third income fifth | $23,807 | $25,236 | −$73 | −$1,502 | −$1,463 | −$39 |
| Fourth income fifth | $32,211 | $27,612 | $2,380 | $6,978 | $6,410 | +$568 |
| Highest income fifth | $55,720 | $38,744 | $3,452 | $20,428 | $20,373 | +$55 |
Net saving = disposable income − consumption + change in pension entitlements. Source amounts are rounded to the nearest dollar, so the four displayed terms may differ by up to $2. We do not add or take a simple average of group means to construct Canada’s figure.
The wealth fifths below are not the income fifths above. The two sets of averages cannot be joined to describe the same households.
Wealth — a different grouping: the highest wealth fifth held 65.7% of net worth
At the end of 2025, the wealthiest fifth held 65.7% of net worth, compared with 65.2% a year earlier. Net worth means assets minus liabilities at a point in time. It is a stock of wealth, separate from income and saving during a quarter.
These are wealth groups, not the income groups above. We cannot join the two sets of averages to describe the same households. Negative net worth means liabilities exceed assets in that group on average; the negative share is retained in the table.
| Wealth group | Assets ($) | Liabilities ($) | Net worth, Q4 2025 ($) | Net worth, Q4 2024 ($) | Share of net worth (%) |
|---|---|---|---|---|---|
| All households | $1,268,248 | $189,606 | $1,078,642 | $1,024,750 | 100.0 |
| Lowest wealth fifth | $99,066 | $103,674 | −$4,608 | −$3,887 | −0.1 |
| Second wealth fifth | $311,454 | $143,547 | $167,907 | $163,842 | 3.1 |
| Third wealth fifth | $730,156 | $203,971 | $526,185 | $510,071 | 9.8 |
| Fourth wealth fifth | $1,377,774 | $217,628 | $1,160,145 | $1,111,522 | 21.5 |
| Highest wealth fifth | $3,822,787 | $279,209 | $3,543,579 | $3,342,201 | 65.7 |
Statistics Canada aligns survey wealth distributions with national balance-sheet totals and models periods without a wealth survey. Its method includes adjustments for populations the survey does not cover, including the territories. Asset values and distribution estimates can both be revised.
Statistics Canada: Household wealth distributions · 36-10-0660-01. Released 2026-04-13; captured 2026-09-08. Source ZIP SHA-256: af1182dc0b145b5571cf73bc4ee0324b4c1473e6eed65763a2adee0c9bbfa32d. Exact vectors and coordinates are retained with the source observations.
Debt payments: 14.75% of income before interest, Q1 2026
National required principal and interest payments were 14.75% of household disposable income before interest payments in January–March 2026. That compares with 14.68% in Q4 2025, a change of +0.07 percentage points. Interest alone accounted for 8.61%.
A separate national comparison: these debt-service estimates are seasonally adjusted; the dollar flows are expressed at annual rates. They are newer than the household distributions above and cannot tell us the payment burden of any income or wealth group. The denominator adds interest back to disposable income; it is not the income amount in the saving panel.
| Measure | Q4 2025 | Q1 2026 |
|---|---|---|
| Disposable income ($ millions, annual rate) | 1,801,452 | 1,811,968 |
| Interest payments ($ millions, annual rate) | 169,044 | 170,760 |
| Income before interest ($ millions, annual rate) | 1,970,496 | 1,982,728 |
| Required principal ($ millions, annual rate) | 120,152 | 121,704 |
| Total debt payments ($ millions, annual rate) | 289,196 | 292,464 |
| Total debt service (%) | 14.68 | 14.75 |
| Interest-only debt service (%) | 8.58 | 8.61 |
Total debt service (%) = 100 × (required principal + interest) ÷ disposable income before interest. We check that the rounded official ratio reconciles to its own published flows. The dollar flows are rounded to millions, so totals may not sum exactly. This is different from debt-to-income, which compares a debt balance with income.
Statistics Canada: Debt service indicators · 11-10-0065-01. Released 2026-06-12; captured 2026-09-08. Source ZIP SHA-256: bdd027995f32de918157afdb0250c64b4ffea6492e81379f28ad09b8f0ecbbb3. Exact vectors and coordinates are retained with the source observations.
Sources, estimation limits and what changes next
Statistics Canada: Household income, consumption and saving distributions · 36-10-0662-01. Released 2026-04-13; captured 2026-09-08. Source ZIP SHA-256: 56c443cb277ef0411f490a9b2b91cd6b29484739f27f0a7e8aee1a8c6be2484f. Exact vectors and coordinates are retained with the source observations.
These institutional estimates combine macroeconomic totals with surveys, administrative inputs and modelling. Statistics Canada imputes missing elements and benchmarks the distributions to national accounts. The household-size adjustment used for income grouping counts the first adult as 1, other people aged 14 or older as 0.5, and younger children as 0.3. Amounts here remain means per household. Groups can change over time; this is not a fixed panel of households.
Read the household equivalence scale, income and consumption methods, quarterly consumption assumptions, group definitions, saving definition and wealth estimation report.
What changed in this page: this first household context compares Q4 2025 with Q4 2024 using one revised source edition. It adds a distributional view alongside the existing GDP-per-person explanation; it does not change that decomposition. A comparison year is not an earlier published vintage, and it does not measure the size of a revision.
The April 13, 2026 release revised income, consumption and saving from Q1 2025 and wealth back to 2010. It also announced a return to annual publication: reference year 2026 is scheduled for April 13, 2027. Quarterly observations do not imply quarterly releases.
Maintenance: check the issuer’s next release or correction, preserve the replacement tables and methods, compare revisions and recalculate before changing this reading. Debt-service observations are quarterly and follow their separate source release schedule. These manual source checks do not promise automatic publication.
Adapted from Statistics Canada, 2026. This does not constitute an endorsement by Statistics Canada of this product.
The same quarter, across Canada · Q4 2025
How net saving changed across Canada
Estimated average net saving fell in 9 provinces and territories, and rose in 4.
The national average changed by −$155 per household from Q4 2024 to Q4 2025. Showing all ten provinces and three territories makes the different directions visible.
Current dollars per household for one quarter; not seasonally adjusted. These are average accounting balances, including pension adjustments. Prices, household size and population composition differ across regions, so a larger dollar figure does not establish a higher standard of living.
Quarterly consumption is allocated using annual spending patterns; actual regional spending changes may differ from the model’s shared assumptions. That uncertainty also affects saving.
Coverage: all 13 provinces and territories, with none excluded from this panel. The territorial estimates use additional source information because the main income-model database covers the ten provinces. Comparable accounting concepts do not mean identical underlying survey coverage.
Solid vertical line: zero. Dashed reference: Canada’s Q4 2025 average, $1,913. Each row uses the same dollar scale; regions are in geographic order.
Read the regional income, consumption and pension components
| Region | Income ($) | Consumption ($) | Pension adjustment ($) | Saving, Q4 2025 ($) | Saving, Q4 2024 ($) | Change ($) |
|---|---|---|---|---|---|---|
| Canada | $27,356 | $26,336 | $893 | $1,913 | $2,068 | −$155 |
| Newfoundland and Labrador | $24,428 | $23,923 | $81 | $585 | $756 | −$171 |
| Prince Edward Island | $25,159 | $23,030 | $694 | $2,823 | $2,778 | +$45 |
| Nova Scotia | $23,184 | $24,047 | $975 | $112 | $202 | −$90 |
| New Brunswick | $23,363 | $23,423 | $826 | $767 | $878 | −$111 |
| Quebec | $22,966 | $22,336 | $1,826 | $2,456 | $2,220 | +$236 |
| Ontario | $28,445 | $27,750 | $785 | $1,480 | $1,991 | −$511 |
| Manitoba | $26,176 | $25,650 | $769 | $1,296 | $1,433 | −$137 |
| Saskatchewan | $28,929 | $26,110 | $440 | $3,259 | $3,091 | +$168 |
| Alberta | $31,939 | $28,499 | −$191 | $3,249 | $3,337 | −$88 |
| British Columbia | $29,867 | $29,160 | $655 | $1,363 | $1,410 | −$47 |
| Yukon | $34,434 | $26,084 | $437 | $8,788 | $8,802 | −$14 |
| Northwest Territories | $41,013 | $33,577 | $1,021 | $8,457 | $8,391 | +$66 |
| Nunavut | $34,576 | $26,876 | $3,326 | $11,025 | $11,330 | −$305 |
Disposable income − consumption + change in pension entitlements = net saving, subject to source rounding. The regional means cannot be summed or averaged equally to reproduce Canada. This panel describes differences; it does not identify their causes or individual households’ financial choices.
Regional sources, limits and comparison history
Statistics Canada: Regional household income, consumption and saving · 36-10-0663-01. Released 2026-04-13; captured 2026-09-08. Source ZIP SHA-256: 703b522fe24ce3c639f682b179265b4c3d0742f3f8b40090d3d6d62e3751ddab. Exact vectors and coordinates are retained with the source observations.
The income-input documentation explains the territorial supplements. The spending-input documentation describes differences in survey coverage, including exclusions from the underlying household-spending survey. These are national-accounts estimates, not a direct survey of every resident.
Both quarters come from the April 2026 edition. This is the first regional panel in this Brief; the older GDP and housing readings remain intact. A regional change in current dollars is not a measure of local inflation, service quality or the experience of a fixed household.
Maintenance follows the same annual household-distribution release and correction checks as the income panel. Before an update, retain the old reading, verify all 13 regional records and check that definitions and the national reconciliation still agree. No Bank of Canada survey is needed to describe this published accounting split.
Adapted from Statistics Canada, 2026. This does not constitute an endorsement by Statistics Canada of this product.
Chapter 6§
Inside output per worker
This chapter separates the change in output per worker into three parts. Output per hour uses the measured labour-productivity index, which covers the business sector and uses National Accounts hours definitions. The measured hours term tracks changes in the business-sector hours-worked index relative to employment measured by the Labour Force Survey. Hours are measured separately, rather than folded into a residual.
The remaining part is calculated. It reconciles the gap between whole-economy real GDP and the business-sector measures. It is not an hours effect. The three parts use different coverage and definitions, which must be kept visible.
At this Brief's second reading, the productivity and hours records end one quarter before the GDP record. The split is unavailable for any period ending at that latest GDP quarter, including the latest era. We do not estimate the missing quarter. Earlier windows are unchanged. The companion Brief, Capital Behind Each Worker, covers investment.
Chapter 7§
The denominator is doing the work
Why the population-structure term dominates: structural evidence already in the Record. The old-age dependency break is composed here as structure — a durable shift in who the denominator counts — never as news. The working-age share of the population is also read era by era, showing whether the structural shift is a recent development or a longer-running trend.
Old-age dependency ratio (65+ per 100 aged 15-64), July 1: level shift detected around 2014
A two-segment mean-shift scan over the last 40 years places the largest level shift in old-age dependency ratio (65+ per 100 aged 15-64), july 1 around 2014: mean 18.35 before vs 26.58 after (shift = 4.4× pooled variability). The detector locates the shift; it does not attribute a cause.
Technical details
Chapter 8§
What the identity can and cannot say
The calculation cannot tell us whether changes in output per worker reflect investment, composition, measurement, or demand. It shows where the change appears in the four-part calculation, not what caused it. What remains derived in the sub-split is now a narrower national-accounts scope reconciliation, not an hours effect. Era boundaries are conventions, not natural breakpoints, and are shown with a sensitivity check rather than presented as the only possible reading. Chained-dollar levels are not additive across periods; every era comparison here uses the identity's own log changes, never a difference of dollar levels across widely separated periods; a Canada–United States comparison is deferred until the two countries' measures can be matched and the source licences allow commercial use.
Wording corrected September 7, 2026: an older sentence incorrectly ruled out a contribution from output per worker. The current finding and calculation are unchanged.
A guide to the numbers · Fixed example: January–March 2026
How revisions change an economic reading
An economic estimate can change after the quarter has ended. The useful question is whether the new estimate changes what we can say about the economy.
A small revision can cross zero
Statistics Canada’s August 28, 2026 release raised its estimate of real GDP for January–March 2026 by 0.12%. That small revision changed the sign of growth compared with the same quarter a year earlier.
Two estimates of the same quarter
Real GDP compared with January–March 2025 · Close-up around zero
Earlier estimate
August 28 estimate
The earlier estimate put output just below its level a year before; the revised estimate put it just above. Both readings were close to zero. Crossing that line does not, by itself, establish a strong expansion.
Follow the calculation
Real GDP measures the volume of output with price changes accounted for. These levels are in millions of chained 2017 dollars, seasonally adjusted at annual rates: recurring seasonal patterns are adjusted for, and the quarterly level is expressed at a yearly pace. It is not the dollar total produced during those three months.
| Quarter measured | Earlier edition | August release |
|---|---|---|
| January–March 2025 | 2,502,038 | 2,502,038 |
| January–March 2026 | 2,500,680 | 2,503,604 |
The comparison quarter stayed the same. Divide each estimate for 2026 by the 2025 level, subtract one, then multiply by 100.
Earlier: (2,500,680 ÷ 2,502,038 − 1) × 100 = −0.05%
Revised: (2,503,604 ÷ 2,502,038 − 1) × 100 = +0.06%
The revision to the level is a different calculation: (2,503,604 ÷ 2,500,680 − 1) × 100 = +0.12%. It compares the two estimates with each other.
A changed figure can leave a conclusion intact
The same release also revised GDP at current prices, before adjustment for price changes. That changed the denominator in Countability’s historical defence-spending comparison for fiscal 2025–26. Holding its NATO-reported spending estimate fixed, the recomputed share still rounded to 1.92% before and after.
This was a comparison with the former 2% guideline, not a new assessment of NATO’s current commitments or an audited spending result. Read the current defence-spending assessment.
Why does changing GDP affect a spending share?
A share divides spending by GDP. With the same spending amount above the line, a larger GDP figure below it produces a smaller share. The two shares can still look the same when rounded.
The retained comparison holds NATO’s reported $62,713 million estimate fixed. It averages four quarterly GDP levels at annual rates to match April 2025–March 2026. Adding four annual-rate levels would overstate that denominator.
| Quarter | Earlier edition | August release |
|---|---|---|
| April–June 2025 | 3,211,260 | 3,211,260 |
| July–September 2025 | 3,260,736 | 3,260,736 |
| October–December 2025 | 3,284,580 | 3,284,580 |
| January–March 2026 | 3,321,588 | 3,328,492 |
| Four-quarter average | 3,269,541 | 3,271,267 |
Averages are shown to the nearest million dollars. The shares below use the unrounded averages.
Earlier: 62,713 ÷ 3,269,541 × 100 = 1.92%
Revised: 62,713 ÷ 3,271,267 × 100 = 1.92%
This reuses a stated estimate and a matched GDP comparison. It does not replace NATO’s own published ratio or make the estimate an audited total.
Name the kind of change
- New period: a new quarter is added. April–June 2026 was new in this release.
- Routine revision: the issuer updates an estimate for a period already reported. January–March was the earlier quarter revised here.
- Correction: an identified error is repaired. Say whether the error was the issuer’s or Countability’s.
- Method change: a definition, classification or estimation method changes. Check whether the earlier and later numbers remain comparable.
The published reading treats this GDP restatement as routine, based on the issuer’s revision practice and the absence of a correction footnote in the retained table metadata. That does not identify the particular cause, or rule out every possible method question. Statistics Canada explains why GDP estimates are revised.
Try it: read a revised headline
Allow five minutes. A headline says, “GDP revision shows the economy was growing, so the earlier reading was an error.” Use the two estimates above to explain what changed and which part of that headline needs more evidence.
Then explain why a spending share can decrease with no change in spending, and what you would need before saying GDP estimates usually revise upward.
Compare your answer
The revised estimate changes the sign of the year-over-year comparison for the same quarter. Both estimates are close to zero. A later estimate does not itself establish that the earlier publication contained an error; that requires evidence about what was wrong.
A larger denominator lowers a ratio when its numerator is held fixed. To establish a usual direction of GDP revisions, compare many matched release editions under compatible definitions. One restatement cannot answer that question.
Return to the current Living Standards reading ↑ · Link to this lesson
Sources, retained editions and responsibility
Fixed example: January–March 2026, as revised in the August 28, 2026 release. Statistics Canada table 36-10-0104-01, Canada, gross domestic product at market prices; chained 2017 dollars for the growth example and current prices for the spending denominator; seasonally adjusted at annual rates in both cases.
The earlier normalized source records were retained on August 10, 2026; the later ones on September 2 in Toronto (September 3 UTC). Countability retains both editions. The figures printed above stay fixed when the current reading advances. The source’s latest table may contain later revisions and cannot, by itself, reproduce the earlier estimate.
Read Countability’s September 2 revision reading, which also records the effects on GDP per person and net debt relative to GDP. The defence numerator is the same NATO Table 1 estimate used in the existing denominator comparison; this lesson does not substitute a newer spending estimate.
This is one worked restatement, not a revision history or a measure of Statistics Canada’s usual revision size, direction or reliability. Countability prepared the explanation and teaching exercise with AI assistance and is responsible for them. No issuer endorsement or external specialist review is claimed. About Countability and contact.
Adapted from Statistics Canada, table 36-10-0104-01, 2026. This does not constitute an endorsement by Statistics Canada of this product.
Scenarios
The identity, latest four quarters
How does the change in real GDP per person decompose over the latest four-quarter window?
Exact log-change identity: the change in ln(real GDP per person) equals the sum of the four term changes; the output-per-worker term is further split into the measured productivity index, measured hours worked per employed person, and a narrower derived reconciliation. Contributions are reported in percentage points; the parts sum to the total exactly, and that arithmetic is re-checked every time the record is rebuilt.
| Real GDP per person (total change) | +0.31 pp |
| Output per worker | -0.57 pp |
| Employment rate | +0.28 pp |
| Participation | -0.77 pp |
| Population structure | +1.37 pp |
| — of which output per hour (measured index) | -0.60 pp |
| Identity residual (always exactly zero) | +0.00 pp |
| — of which hours worked per employed person (measured) | -0.34 pp |
| — of which national-accounts scope reconciliation (derived) | +0.36 pp |
Assumptions:
- Monthly labour series are sampled at quarter starts to align with quarterly GDP and population (a convention reviewed and adopted in July 2026).
- The labour force is participation times the LFS population aged fifteen and over — both on the same LFS basis.
- The hours-plus-reconciliation term is a derived residual, not a measured component; it absorbs the LFS-versus-SNA basis difference.
The identity, 1991–2000
How did the change in real GDP per person decompose across the first era of the record?
Exact log-change identity computed over one era window instead of the latest four quarters: the change in ln(real GDP per person) equals the sum of the four term changes; output per worker is further split into measured output per hour, measured hours worked per employed person, and a narrower derived reconciliation. Contributions are the era's total change in percentage points, not yet annualized; the identity residual is zero by construction.
| Real GDP per person (total change) | +21.86 pp |
| Output per worker | +16.61 pp |
| Employment rate | +3.48 pp |
| Participation | -0.45 pp |
| Population structure | +2.23 pp |
| — of which output per hour (measured index) | +19.74 pp |
| Identity residual (always exactly zero) | +0.00 pp |
| — of which hours worked per employed person (measured) | +3.63 pp |
| — of which national-accounts scope reconciliation (derived) | -6.76 pp |
Assumptions:
- Monthly labour series are sampled at quarter starts to align with quarterly GDP and population, the same convention the latest-window scenario uses.
- The labour force is participation times the LFS population aged fifteen and over, both on the same LFS basis.
- Era boundaries are quarter-exact decade ends; results are not annualized in this scenario (see the companion window_growth_split scenarios for the annualized total-growth split).
The identity, 2000–2010
How did the change in real GDP per person decompose across the second era of the record?
Exact log-change identity computed over one era window instead of the latest four quarters: the change in ln(real GDP per person) equals the sum of the four term changes; output per worker is further split into measured output per hour, measured hours worked per employed person, and a narrower derived reconciliation. Contributions are the era's total change in percentage points, not yet annualized; the identity residual is zero by construction.
| Real GDP per person (total change) | +8.56 pp |
| Output per worker | +5.00 pp |
| Employment rate | -1.14 pp |
| Participation | +1.50 pp |
| Population structure | +3.19 pp |
| — of which output per hour (measured index) | +7.68 pp |
| Identity residual (always exactly zero) | +0.00 pp |
| — of which hours worked per employed person (measured) | -5.48 pp |
| — of which national-accounts scope reconciliation (derived) | +2.81 pp |
Assumptions:
- Monthly labour series are sampled at quarter starts to align with quarterly GDP and population, the same convention the latest-window scenario uses.
- The labour force is participation times the LFS population aged fifteen and over, both on the same LFS basis.
- Era boundaries are quarter-exact decade ends; results are not annualized in this scenario (see the companion window_growth_split scenarios for the annualized total-growth split).
The identity, 2010–2019
How did the change in real GDP per person decompose across the third era of the record?
Exact log-change identity computed over one era window instead of the latest four quarters: the change in ln(real GDP per person) equals the sum of the four term changes; output per worker is further split into measured output per hour, measured hours worked per employed person, and a narrower derived reconciliation. Contributions are the era's total change in percentage points, not yet annualized; the identity residual is zero by construction.
| Real GDP per person (total change) | +8.26 pp |
| Output per worker | +6.99 pp |
| Employment rate | +2.74 pp |
| Participation | -1.81 pp |
| Population structure | +0.34 pp |
| — of which output per hour (measured index) | +10.10 pp |
| Identity residual (always exactly zero) | +0.00 pp |
| — of which hours worked per employed person (measured) | -2.74 pp |
| — of which national-accounts scope reconciliation (derived) | -0.37 pp |
Assumptions:
- Monthly labour series are sampled at quarter starts to align with quarterly GDP and population, the same convention the latest-window scenario uses.
- The labour force is participation times the LFS population aged fifteen and over, both on the same LFS basis.
- Era boundaries are quarter-exact decade ends; results are not annualized in this scenario (see the companion window_growth_split scenarios for the annualized total-growth split).
The identity, 2019–present
How did the change in real GDP per person decompose across the current era?
Exact log-change identity computed over one era window instead of the latest four quarters: the change in ln(real GDP per person) equals the sum of the four term changes; output per worker is further split into measured output per hour, measured hours worked per employed person, and a narrower derived reconciliation. Contributions are the era's total change in percentage points, not yet annualized; the identity residual is zero by construction.
| Real GDP per person (total change) | +1.25 pp |
| Output per worker | +0.55 pp |
| Employment rate | -0.91 pp |
| Participation | -1.22 pp |
| Population structure | +2.83 pp |
| — of which output per hour (measured index) | +1.18 pp |
| Identity residual (always exactly zero) | +0.00 pp |
| — of which hours worked per employed person (measured) | -1.50 pp |
| — of which national-accounts scope reconciliation (derived) | +0.87 pp |
Assumptions:
- Monthly labour series are sampled at quarter starts to align with quarterly GDP and population, the same convention the latest-window scenario uses.
- The labour force is participation times the LFS population aged fifteen and over, both on the same LFS basis.
- Era boundaries are quarter-exact decade ends; results are not annualized in this scenario (see the companion window_growth_split scenarios for the annualized total-growth split).
The identity, per-capita peak to latest
How did the change in real GDP per person decompose from its record peak to the latest quarter?
Exact log-change identity computed over one era window instead of the latest four quarters: the change in ln(real GDP per person) equals the sum of the four term changes; output per worker is further split into measured output per hour, measured hours worked per employed person, and a narrower derived reconciliation. Contributions are the era's total change in percentage points, not yet annualized; the identity residual is zero by construction.
| Real GDP per person (total change) | -1.07 pp |
| Output per worker | -1.32 pp |
| Employment rate | -1.72 pp |
| Participation | -0.46 pp |
| Population structure | +2.43 pp |
| — of which output per hour (measured index) | -0.84 pp |
| Identity residual (always exactly zero) | +0.00 pp |
| — of which hours worked per employed person (measured) | -1.15 pp |
| — of which national-accounts scope reconciliation (derived) | +0.67 pp |
Assumptions:
- Monthly labour series are sampled at quarter starts to align with quarterly GDP and population, the same convention the latest-window scenario uses.
- The labour force is participation times the LFS population aged fifteen and over, both on the same LFS basis.
- Era boundaries are quarter-exact decade ends; results are not annualized in this scenario (see the companion window_growth_split scenarios for the annualized total-growth split).
Growth composition, 1991–2000
In the first era, how much of total real GDP growth was population growth, and how much was per-person growth?
Splits one era's total real GDP growth into its numerator (real GDP) and denominator (population) own annualized log-change, restating the same era window's per-person growth as what it is arithmetically made of.
| Numerator, annualized | +3.43 pp |
| Denominator, annualized | +1.00 pp |
| Ratio, annualized | +2.43 pp |
| Window length (years) | 9.0 years |
Assumptions:
- Annualization divides the window's total log change by its exact length in years (quarter-counted), the same convention the era-ladder scenarios use.
- Real GDP and population are both quarterly, chained/postcensal series already held in the record; nothing here is projected.
Growth composition, 2000–2010
In the second era, how much of total real GDP growth was population growth, and how much was per-person growth?
Splits one era's total real GDP growth into its numerator (real GDP) and denominator (population) own annualized log-change, restating the same era window's per-person growth as what it is arithmetically made of.
| Numerator, annualized | +1.89 pp |
| Denominator, annualized | +1.03 pp |
| Ratio, annualized | +0.86 pp |
| Window length (years) | 10.0 years |
Assumptions:
- Annualization divides the window's total log change by its exact length in years (quarter-counted), the same convention the era-ladder scenarios use.
- Real GDP and population are both quarterly, chained/postcensal series already held in the record; nothing here is projected.
Growth composition, 2010–2019
In the third era, how much of total real GDP growth was population growth, and how much was per-person growth?
Splits one era's total real GDP growth into its numerator (real GDP) and denominator (population) own annualized log-change, restating the same era window's per-person growth as what it is arithmetically made of.
| Numerator, annualized | +2.06 pp |
| Denominator, annualized | +1.14 pp |
| Ratio, annualized | +0.92 pp |
| Window length (years) | 9.0 years |
Assumptions:
- Annualization divides the window's total log change by its exact length in years (quarter-counted), the same convention the era-ladder scenarios use.
- Real GDP and population are both quarterly, chained/postcensal series already held in the record; nothing here is projected.
Growth composition, 2019–present
In the current era, how much of total real GDP growth was population growth, and how much was per-person growth?
Splits one era's total real GDP growth into its numerator (real GDP) and denominator (population) own annualized log-change, restating the same era window's per-person growth as what it is arithmetically made of.
| Numerator, annualized | +1.67 pp |
| Denominator, annualized | +1.47 pp |
| Ratio, annualized | +0.20 pp |
| Window length (years) | 6.2 years |
Assumptions:
- Annualization divides the window's total log change by its exact length in years (quarter-counted), the same convention the era-ladder scenarios use.
- Real GDP and population are both quarterly, chained/postcensal series already held in the record; nothing here is projected.
Directly observed
Method and limitations
- An exact accounting identity in log changes: the change in real GDP per person equals the sum of output per worker, the employment rate, participation, and population structure — with output per worker further split into output per hour (the measured Statistics Canada labour-productivity index), hours worked per employed person (measured directly from the business-sector hours-worked index), and a narrower derived reconciliation for the remaining national-accounts scope gap. Monthly labour series are sampled at quarter starts to align with quarterly GDP and population, a convention reviewed and adopted in July 2026. The same identity is computed for the latest window and for each era in the long-run record, on one basis throughout. The terms sum to the total exactly; that arithmetic is re-checked automatically every time the record is rebuilt. The identity locates where change shows up in measurement; it never asserts why.
- Monthly labour series are sampled at quarter starts to align with quarterly GDP and population (a convention reviewed and adopted in July 2026).
- The labour force is participation times the LFS population aged fifteen and over — both on the same LFS basis.
- The hours-plus-reconciliation term is a derived residual, not a measured component; it absorbs the LFS-versus-SNA basis difference.
- Monthly labour series are sampled at quarter starts to align with quarterly GDP and population, the same convention the latest-window scenario uses.
- The labour force is participation times the LFS population aged fifteen and over, both on the same LFS basis.
- Era boundaries are quarter-exact decade ends; results are not annualized in this scenario (see the companion window_growth_split scenarios for the annualized total-growth split).
- Monthly labour series are sampled at quarter starts to align with quarterly GDP and population, the same convention the latest-window scenario uses.
- The labour force is participation times the LFS population aged fifteen and over, both on the same LFS basis.
- Era boundaries are quarter-exact decade ends; results are not annualized in this scenario (see the companion window_growth_split scenarios for the annualized total-growth split).
- Monthly labour series are sampled at quarter starts to align with quarterly GDP and population, the same convention the latest-window scenario uses.
- The labour force is participation times the LFS population aged fifteen and over, both on the same LFS basis.
- Era boundaries are quarter-exact decade ends; results are not annualized in this scenario (see the companion window_growth_split scenarios for the annualized total-growth split).
- Monthly labour series are sampled at quarter starts to align with quarterly GDP and population, the same convention the latest-window scenario uses.
- The labour force is participation times the LFS population aged fifteen and over, both on the same LFS basis.
- Era boundaries are quarter-exact decade ends; results are not annualized in this scenario (see the companion window_growth_split scenarios for the annualized total-growth split).
- Monthly labour series are sampled at quarter starts to align with quarterly GDP and population, the same convention the latest-window scenario uses.
- The labour force is participation times the LFS population aged fifteen and over, both on the same LFS basis.
- Era boundaries are quarter-exact decade ends; results are not annualized in this scenario (see the companion window_growth_split scenarios for the annualized total-growth split).
- Annualization divides the window's total log change by its exact length in years (quarter-counted), the same convention the era-ladder scenarios use.
- Real GDP and population are both quarterly, chained/postcensal series already held in the record; nothing here is projected.
- Annualization divides the window's total log change by its exact length in years (quarter-counted), the same convention the era-ladder scenarios use.
- Real GDP and population are both quarterly, chained/postcensal series already held in the record; nothing here is projected.
- Annualization divides the window's total log change by its exact length in years (quarter-counted), the same convention the era-ladder scenarios use.
- Real GDP and population are both quarterly, chained/postcensal series already held in the record; nothing here is projected.
- Annualization divides the window's total log change by its exact length in years (quarter-counted), the same convention the era-ladder scenarios use.
- Real GDP and population are both quarterly, chained/postcensal series already held in the record; nothing here is projected.
This Brief is published in full: the complete cited record behind its headline question. Countability can commission a related Brief with custom scenarios, a recurring refresh, or a different decision question.
Commission a related Brief.