Latest Developments in the Canadian Economic Accounts
The 2019 to 2021 Revisions of the Income and Expenditure Accounts

Introduction

This article presents revisions to the Gross Domestic Product (GDP) by Income and by Expenditure from the Income and Expenditure Accounts (IEA) for the 2019-to-2021 period. These revisions were reflected in the 2021 Provincial and Territorial Economic Accounts estimates, released on November 8, 2022.

The first and second quarters of 2022 have also been revised. In accordance with standard revision practice, these estimates will continue to be revised in upcoming cycles, and therefore, are excluded from the analysis in this article.

Revision process and phases

GDP estimates by IEA are subject to several types of revisions that take place with each quarterly release depending on the availability of data and the practicality of implementation. The first-quarter release includes revisions to all quarters of the previous year; the second-quarter release revises only the first quarter of the same year; the third-quarter release (the annual revision) includes revisions to all quarters back three years; and the fourth-quarter release includes revisions to all quarters of the same year.

Occasionally, substantial changes that are made to incorporate new concepts or methods (for example, integration of new SNA concepts) result in comprehensive revisions (back to 1961) and can occur with the annual revision. Appendix A illustrates the revision cycle for GDP, including an explanation of the open period and the vintages.

In this release, new benchmark values from the supply-use tables (SUTs) for the 2019 reference year have been incorporated into the GDP by IEA. SUTs integrate data from a wide range of sources into a balanced accounting framework. They represent the most detailed, coherent accounting system for the structure of the Canadian economy and are considered the most accurate benchmarks on which to base estimates.

Revisions for the 2020 and 2021 reference years incorporated up-to-date information from the Annual Survey of Manufacturing and Logging, the Annual Survey of Research and Development in Canadian Industry, the Annual Census of Mines, government public accounts, and updated administrative data. These revisions also reflect incorporation of T4 data for 2020 and preliminary T4 data for 2021 which is used to estimate wages and salaries.

Results from the Annual Capital and Repair Expenditure Survey, Actual, were incorporated for 2020, whereas 2021 was revised to incorporate major project updates from the upcoming 2021 Actual release (February 2023).

International trade in commercial services has been revised in 2020 and 2021 as the coverage of enterprises in the digital economy has been enhanced, as was estimation of imports of digital products and services. Revisions were mostly related to audio and video streaming and downloaded software activities. In addition, the coverage of imports of advertising services has been expanded significantly to better capture the extent of Canadian payments to foreign entities for online advertising made on social media, Internet browsers, etc.

Appendix B contains the timeline for data source integration into the IEA by major component of income-based GDP and expenditure-based GDP, using the first quarter of 2019 as an example.

Revisions for the 2020 and 2021 reference years also reflect updates to data sources and estimation methodologies due to pandemic-related disruptions. As a result, some initial estimates contained assumptions that evolved as more information became available, and revisions were necessary in subsequent releases.

This article summarizes the combined impact of revisions from all sources on annual and quarterly GDP, on a nominal and a real basis.

Revisions to growth of nominal gross domestic product



With incorporation of new benchmark data from the supply-use tables, the level of nominal GDP was revised up by $2.3 billion in 2019. The nominal GDP growth rate was revised up 0.10 percentage point, largely because of revised business capital investment, exports and household spending.

In 2020, nominal GDP was revised up by $2.9 billion, or 0.03 percentage points. Upward revisions in business capital investment and government final consumption expenditure were offset by a downward revision to net trade of -$4billion as the upward revision to exports couldn’t match the significant revision to imports.

In 2021, the revised nominal GDP growth rate (+13.6%) was stronger than the previously published 13.0%; which followed a 4.5% decline in 2020 owing to a widespread shutdown of economic activities to contain the spread of COVID-19 in 2020. Nominal GDP was revised up by $16.5 billion, or 0.59 percentage points. Upward revisions in business capital investment, government final consumption expenditure were again mitigated by a downward revision to net international trade.

The absolute revision to the growth rate—the difference between the revised average growth rate and the previously published average growth rate—was 0.24 percentage points for the 2019-to-2021 period.



On a quarterly basis, the absolute revision to the growth rate was 0.07 percentage points for the 2019-to-2021 period, with 2021 recording the largest upward revision of 0.14 percentage points. Throughout the period, the quarterly trend remained similar to the trend based on previously published estimates.

Revisions to growth of real gross domestic product



The mean absolute percentage-point revision to the annual average growth rate of real GDP for the 2019-to-2021 period was 0.21. The revision was lowest in 2019 (+0.01 percentage points) and highest in 2021 (+0.47 percentage points). The growth rate was revised up (+0.16 percentage points) in 2020.

Data table for Chart 1



Data table for Chart 1
Table summary
This table displays the results of Data table for Chart 1. The information is grouped by year (appearing as row headers), Previous estimate and Revised estimate, calculated using % change units of measure (appearing as column headers).

year	Previous estimate	Revised estimate
	% change
2019	1.9	1.9
2020	-5.2	-5.1
2021	4.5	5.0
Source: Statistics Canada, Latest Developments in the Canadian Economic Accounts, catalogue number 13-605-X.

Based on chained dollars, the growth rate of real GDP was unchanged in 2019; up 0.10 percentage points in 2020 and up by 0.50 percentage points in 2021.



The mean absolute percentage-point revision to the annual average growth rate of quarterly real GDP for the 2019-to-2021 period was 0.07. The largest revision was in 2021 (+0.16 percentage points) and the smallest in 2019 (+0.01 percentage point). The growth rate was revised up (+0.03 percentage point) in 2020. The extent of these revisions was minimal and well within historical limits, ranging from -0.33 to 0.43 percentage points over the last decade.

Data table for Chart 2



Data table for Chart 2
Table summary
This table displays the results of Data table for Chart 2. The information is grouped by Quarter (appearing as row headers), Zero , Previous estimate and Revised estimate, calculated using % change units of measure (appearing as column headers).

Quarter	Previous estimate	Revised estimate
	% change
2019
I	0.1	0.1
II	1.0	1.0
III	0.3	0.4
IV	0.3	0.3
2020
I	-2.2	-2.1
II	-11.0	-10.9
III	9.0	9.0
IV	2.2	2.1
2021
I	1.1	1.3
II	-0.8	-0.6
III	1.3	1.4
IV	1.6	1.7
Source: Statistics Canada, Latest Developments in the Canadian Economic Accounts, catalogue number 13-605-X.

Based on chained dollars, the movement of real GDP recorded the largest upward revision in 2021, with revisions to each quarter. In 2021, the first and second quarters each recorded an upward revision of 0.2 percentage points, while the third and fourth quarters each recorded a 0.1 percentage point upward revision.

In 2019, real GDP quarterly growth rates underwent limited revisions, with the only revision occurring in the third quarter (+0.1 percentage point). In 2020, real GDP recorded a minor upward revision in the first and second quarter (+0.1 percentage point) and a downward revision in the fourth quarter (-0.1 percentage point). The growth rate was unchanged in the third quarter. At this point in the revision cycle for 2021, revisions are within range of the historical average.

Appendix A

GDP revision cycle

Description for Figure 1

The GDP revisions cycle is a set cycle year-to-year. Revisions occur to previously released quarters based on the reference period open with each release. For the first quarter of GDP by income and by expenditure, the release includes revisions to all quarters of the previous year; a second quarter release includes revisions to only the first quarter of the same year; a third quarter release is the annual revision and includes revisions to all quarters back three years (due to incorporation of the Supply-Use Table (SUT) for three years prior); and a fourth quarter release includes revisions to all quarters of the same year.

Comprehensive revisions (quarterly, back to 1961) can occur with the annual revision when there are significant changes to concepts or methods, such as integration of new System of National Accounts concepts.

The table illustrates the main production period for each quarter, the corresponding reference quarter/year and the release date. Using the first quarter of a year as an example: the first vintage of those estimates is released in May of the same reference year with revisions to all quarters previous year. The second vintage of the first quarter is released in August with the second quarter release. The third vintage of the first quarter is released with the third quarter release (annual revision) in November. The fourth vintage of the first quarter is released in February of the following year with the fourth quarter release. The fifth vintage of the first quarter is released in May with the first quarter of the following year. The sixth vintage of the first quarter is released with the third quarter release of the following year (annual revision) in November. The seventh vintage of the first quarter is released with the third quarter of two years after (annual revision) in November. The eighth vintage of the first quarter and the last revision vintage (excluding comprehensive revisions) is released in November with the third quarter release three years after, and reflects the incorporation of the SUT benchmark for that year.

Appendix B

Revisions due to different vintages of source data