What the data are
Every year the Canada Revenue Agency publishes T1 Final Statistics: totals for each line of the individual income tax return, by province and by range of total income ($0 to $5,000, $5,000 to $10,000, and so on up to $250,000 and over). Because those dollar ranges are fixed while incomes grow, the people in, say, the top range are a different slice of the population each year. This dataset re-expresses each line item by percentile group instead, so a group such as the top 1% means the same thing in every year. It covers line items for the ten provinces and all provinces combined, from to .
Income groups
Taxfilers in each province and year are ranked by total income assessed and split into twenty groups of 5% each, with the top group divided into the 95th–98th percentiles (4% of taxfilers) and the top 1%. Percentiles are based on the total number of returns, taxable or not. “All provinces” ranks taxfilers across the ten provinces together, so its groups are not the sum of the provincial groups. Broad groups add up percentile groups:
- Bottom 50%: 0–49th percentiles; Middle 40%: 50th–89th; Top 10%: 90th and up; Next 9%: 90th–98th; Top 5%: 95th and up; Top 1%: 99th and up.
Every line item is assigned to groups by the total income of the people reporting it. A deduction or credit shown for the top 1% is the amount claimed by the 1% of taxfilers with the highest total income, not by the 1% with the largest claims.
Measures
- Share of total: the group’s share of the item’s total in the province, in per cent.
- Dollars: the group’s total, in dollars, adjusted for inflation with the Canada-wide consumer price index (annual average).
- Per taxfiler: the group’s total divided by the number of taxfilers in the group (whether or not they reported the item), in dollars.
- Change compares the year with the comparison year: in percentage points for shares and in per cent for dollars. With per-taxfiler dollars and total income, this is sometimes called a growth incidence curve.
- The concentration curve plots the cumulative share of an item against the cumulative share of taxfilers, from lowest to highest income. For total income it is the Lorenz curve; the closer it bows below the diagonal, the more unequal the distribution. The concentration index in the legend is twice the area between the curve and the diagonal (the Gini coefficient, for total income), calculated from the groups shown, so it slightly understates inequality within groups.
How the groups are estimated
The CRA tables give, for each income range, the number of returns and the dollars of each item. The number of returns places each range on the percentile scale. Within a range, the cumulative share of each item is assumed to rise in a straight line, which gives its value at each percentile (linear interpolation). Negative amounts in an income range (for example, business losses) and suppressed cells are treated as zero.
Change in method from . The CRA’s top income range, $250,000 and over, held under 1% of taxfilers nationally until the late 2010s but about 2% by 2024. A straight line within that range spreads its income evenly across everyone in it and increasingly understates the top 1%. From , total income in the top range is instead assumed to follow a Pareto distribution fitted to that range’s average income, and every other item in the top range is divided in proportion to income. Shares of the top 1% and the 95th–98th percentiles therefore shift at –: nationally, the top 1% share of total income is 0.5 percentage points higher with the new method in the year it starts, and more in Alberta, Ontario and British Columbia, where the top range already held more than 1% of taxfilers. Groups below the top range are not affected. The top 1% share of income estimated this way is close to Statistics Canada’s estimates from its own tax data.
Some series are combined by Finances of the Nation where the CRA reports a line item in pieces in some years and as one total in others (marked “combined series”), and broader groupings such as Self-employment income and Pension income add several lines. Line items new since 2017 (for example, taxable scholarships, which were part of other income until 2021, and the deduction for enhanced CPP/QPP contributions) are included in the matching groupings. Many items exist only in some years, as tax rules change; the item list shows when each one is reported.
Sources
Canada Revenue Agency, T1 Final Statistics, Table 2 (all returns by total income class), for each province. Data for 1992 to 2008 are from CRA tabulations prepared for Kevin Milligan and Michael Smart, “Taxation and top incomes in Canada,” Canadian Journal of Economics 48(2), 2015. Consumer prices are from Statistics Canada table 18-10-0004, through the FON macroeconomic data.
Download the data
The full dataset (every item, province, year, income group and measure) is available as a CSV file, with a user guide to the original construction. The Download this item button in the panel gives one item for every province, year and income group.
How to cite
Questions or comments? Please email Trevor Tombe (University of Calgary) at .
Thanks: Alexander Hempel built the original version of this dataset.
Licence
This tool and its data are licensed under a Creative Commons Attribution 4.0 International licence (CC BY 4.0). You may share and adapt them for any purpose, including commercially, as long as you give appropriate credit (see How to cite), link to the licence and say if you made changes. The underlying CRA data are published under the Open Government Licence – Canada, and Statistics Canada data under the Statistics Canada Open Licence.