This activity provides information for current economic analysis, from an industry point of view.
Data release – November 29, 2013
This activity provides information for current economic analysis, from an industry point of view. Gross Domestic Product (GDP) by industry at basic prices is a measure of the economic production which takes place within the geographical boundaries of Canada. The term "gross" in GDP means that capital consumption costs, that is the costs associated with the depreciation of capital assets (buildings, machinery and equipment), are included. The production estimates are prepared for 214 separate industries using the North American Industrial Classification System (NAICS 2007).
The GDP by industry measures provide an alternate measure of total economic activity that supplements the income and expenditure-based estimates prepared by Income and Expenditure Accounts Division, and constitute an extension (on a monthly basis) of the System of National Accounts Input-Output (IO) Tables.
The target population consists of all establishments in Canada. The establishment is the level at which the accounting data required to measure production is available. The establishment, as a statistical unit, is defined as the most homogeneous unit of production for which the business maintains accounting records from which it is possible to assemble all the data elements required to compile the full structure of the gross value of production (total sales or shipments, and inventories), the cost of materials and services, and labour and capital used in production.
Data are collected from other Statistics Canada surveys and/or other sources.
The GDP measures rely heavily on a wealth of information from various areas of Statistics Canada, from other federal departments and agencies, from provincial government departments, and from private industry sources. This large amount of information is compiled, integrated and analysed as part of the complex process of arriving at GDP by industry.
For example, data from the Monthly Survey of Manufacturing are used for most (but not all) manufacturing industries. Data from the Survey of Employment, Payrolls and Hours (SEPH) are used for many service industries.
Data at the working level industry, the lowest level of industry detail for which GDP estimates are compiled directly, are verified for large month-to-month percentage changes and potential issues arising from source data as well as analyzed for time series consistency, links to current economic events and with respect to coherence with related economic indicators not used in the derivation of the GDP estimates.
The current price estimates of Gross Domestic Product (GDP) at basic prices (or value added) by industry can be measured directly by summing the factor incomes and depreciation or indirectly by deducting the cost of the intermediate goods and services used in the production process from the value of gross production or output; see Catalogue no. 15-201-X, The Input-Output structure of the Canadian Economy.
Constant price estimates of GDP by industry measure economic growth of industries with the effect of price variations removed. The constant price estimates of GDP are obtained by the double-deflation method, as described in publication Catalogue no. 15F0077G, A guide to deflating the input-output accounts: sources and methods.
Both the current and constant price estimates of GDP by industry however can only be derived annually within the framework of the Input-Output tables, and for all but the most recent two or three years. For the years following the most recent IO tables, and for the monthly estimates, real GDP by industry can only be obtained from projecting the relationship between real gross output and real valued added, which holds over short periods of time. That is, the volume of value added generated from a given volume of output for a specific industry is generally constant over short periods of time, as major technological changes are required to change this relationship significantly.
To estimate on a monthly basis the real value added by an industry, indicators of real output, employment, or real inputs are used to project the relationship between these characteristics and value added, as determined from the deflated Input-Output tables. See the document Catalogue no. 15-547-X, Gross Domestic Product by Industry: Sources and Methods for an overview of the general approach and of the various statistical techniques used in the derivation of monthly GDP. The document Catalogue no. 15-548-X, Gross Domestic Product by Industry: Sources and Methods with Industry Details provides a detailed description of the data sources used for each industry. A document containing an up-to-date summary of the data sources and methods used for compiling monthly national estimates of GDP by industry is included below in the documentation section.
Real measures of GDP by industry are calculated for individual industries and then aggregated to arrive at total economic growth. The monthly gross domestic product (GDP) by industry data are chained volume estimates with 2007 as their reference year. This means that the estimates for each industry and aggregate are obtained from a chained volume index multiplied by the industry's value added in 2007. For the period 2007 to 2009, the monthly estimates are benchmarked to annually chained Fisher volume indexes of GDP obtained from the constant-price input-output tables. For the period starting with January 2010, the estimates are derived by chaining a Laspeyres volume index at 2009 prices to the prior period. This makes the monthly GDP by industry estimates more comparable with the expenditure-based GDP data, chained quarterly.
Data at the working level industry is analyzed for time series consistency, links to current economic events, issues arising from the source data and with respect with coherence. As well, the growth rates of total GDP by industry are compared with the growth rates of total expenditure based GDP (record no. 1901) on a quarterly basis, and divergences are reconciled and minimized.
Statistics Canada is prohibited by law from releasing any information it collects which could identify any person, business, or organization, unless consent has been given by the respondent or as permitted by the Statistics Act. Various confidentiality rules are applied to all data that are released or published to prevent the publication or disclosure of any information deemed confidential. If necessary, data are suppressed to prevent direct or residual disclosure of identifiable data.
Revisions arise from updates to benchmark data, projectors, methodologies and seasonal adjustment. The revision policy is to revise back to previous year for the January to June reference months; for the July reference month, back to January of the forth previous year; and for the August to December reference months, back to January of the current year. Historical revisions are carried out usually once per decade.
Statistical revisions are carried out in order to incorporate the most recent information from surveys, taxation statistics, public accounts, censuses, etc., as well as new methodologies, data sources, concepts or definitions and the annual benchmarking process to the Input-Output Accounts.
The monthly data are adjusted by a factor that reflects variations in the number of trading days within each month to derive the seasonally unadjusted data and by a seasonal adjustment factor to derive the seasonally adjusted data. The seasonal adjustment factors are derived using the X-12-ARIMA method. Any trading day adjustments generated by this program are based not only on the number of days in the month but also on the relative importance of each day of the week.
The monthly measures are projections of the annual estimates of constant price Gross Domestic Product at basic prices by industry. It is generally true that there are fewer data available to prepare the monthly estimates than to prepare the annual measures and these monthly data are usually not as reliable as the annual data. The relationship between annual and monthly measures in the period for which both are available is used to estimate necessary adjustments to the monthly data in the current period.
GDP by industry depends on the Input-Output Tables (record no. 1401) to which it is anchored. The quality of the Input-Output Tables is matched where there are high quality projectors of industrial production. When necessary, the GDP by industry program may be required to use projectors of industrial production which are of lower quality than the underlying GDP from the Input-Output tables, in these cases quality of the measure is lessened but still acceptable. As well, the availability of appropriate prices affects the reliability of the real GDP estimates. In general, the higher the level of aggregation, the more reliable are the estimates. There is a trade-off between timeliness and accuracy. As more robust data becomes available, estimates are revised and become more accurate until the benchmark Input-Output Tables are published, approximately two and a half years after the monthly estimates of GDP by industry were first published.
In general, weaknesses in source data arise mainly from the following: a) undercoverage; b) inappropriate concepts and definitions. These are briefly discussed below:
a) Undercoverage - This weakness is normally corrected by inflating reported data by a factor that allows the data to represent the universe concerned.
b) Concepts and definitions not suitable for the SNA - The data used in the derivation of monthly GDP are quite varied in coverage, details, definitions and concepts and often these factors do not coincide with those required. They must be thoroughly examined and adjusted for consistency and coverage using carefully designed estimating procedures.
No direct measures of the margin of error in the estimates can be calculated. Data reliability ratings are a product of data integration and analysis inherent in the compilation of GDP by industry. They rely both on the quantitative attributes of the data sources used, such as sample size, response rate and coefficient of variation, and on the expert judgement of analysts who undertake data integration of these various sources.
In addition, as the monthly GDP by industry program relies on the annual estimates of real GDP obtained from the Input-Output tables, the quality of these estimates has a direct influence on the quality of the monthly GDP measures. The quality ratings of the current dollar GDP estimates obtained from the Input-Output tables for the latest benchmark year are published in Statistics Canada catalogue 15-201, "The Input-Output Structure of the Canadian Economy" Text table 2.
At the moment, there are no data reliability ratings for the GDP by industry estimates in constant prices.
For further information please refer to "Latest developments in the Canadian economic accounts", Statistics Canada, Catalogue no. 13-605-X.
The side bar menu of this electronic publication includes: chronology of events, conceptual changes, classification changes, and data revisions.