Information identified as archived is provided for reference, research or recordkeeping purposes. It is not subject to the Government of Canada Web Standards and has not been altered or updated since it was archived. Please "contact us" to request a format other than those available.
|The input-output structure of the Canadian economy
Data quality, concepts and methodology
Data quality, concepts and methodology
Presentation of data
A. Contents of the input-output tables
In the Canadian System of National Accounts, the Input-Output Accounts have two sets of interrelated accounts:
The Industry Accounts contain the commodity composition of industry output and its complete costs of production including profits (surplus). The Commodity Accounts, on the other hand, show the supply and disposition of goods and services. The framework of the Canadian Input-Output Tables is closely related to the one described in the United Nations report, System of National Accounts, 1993. The inputs and outputs of industries, classified by commodity, are presented in separate tables. The Canadian Input-Output Tables are rectangular rather than square since the number of goods and services exceeds the number of industries. Square tables have been an international standard for most countries from the beginning of Input-Output concepts. Square means having the same dimensions, either product-by-product tables or industry by industry. The product by product table shows which products are used in the production of which other products; the industry by industry table which is by far the most widely used shows which industry uses the output of which other industry. From our own rectangular framework we can build square tables: industry by industry and industry by categories of final demand.
For each year, three basic tables are prepared:
The Input-Output Accounts are prepared and balanced at the most detailed level known as the "Worksheet" (W) level, with 300 industries for the year 2000 and 2001. The number of goods and services and primary inputs add to 727 in 2000 and 2001. In the final demand, there are 170 categories for the years 2000 and 2001. The Worksheet (W) level data are then aggregated into Link (L), Medium (M), and Small (S) levels for consistency during the entire period as shown in the following table.
At the W, L and M levels of detail, some of the entries in various national matrices are confidential under the provisions of the Statistics Act. Consequently, data are provided to users after suppressing the confidential cells where applicable. The provincial tables are available at the “S” level of aggregation only and cell suppression is performed at that level.
The tables in the present publication are at the S level of aggregation, covering 51 goods and services and 8 primary inputs, (a total of 59), 25 industries and 16 final demand categories 1. The concordances among the various levels of aggregation for industries, goods and services and primary inputs are provided in the Appendix I .
Text table 1
B. Understanding the content of the national input-output tables
This section presented here highlights the descriptive as well as the analytical usefulness of input-output data, using a series of examples. Observed relationships between industries producing several goods and services and various users of these goods and services allow establishing meaningful economic relationships. A number of possible applications of the data are also offered. The examples are meant to familiarize the reader with the richness of the basic data as well as their potential for various analytical uses. For simplicity we present the data content using the National Input-Output Tables . The same demonstration applies to the provincial tables as well.
The illustration uses one industry “manufacturing”, and one commodity, “petroleum & coal products”, at the S level of aggregation. Utilizing data for 2001 (tables 7, 8 and 9), the industry can be identified as column 8 of the Make and Use Matrices and the commodity as row 26 in the Make, Use and Final Demand Matrices.
The simple question, “who produces what?” is easily answered by referring to the Make Matrix where each row shows the distribution of output of each commodity by industry and where each column lists the distribution of output of each industry by commodity. For example, one may find that in 2001 the manufacturing industry (column 8 of the Make Matrix) produced $35,230 million of petroleum and coal products (row 26) whereas all domestic industries combined (row 26 in the last column of the Make Matrix) produced $39,705 million of that commodity. Thus, the manufacturing industry accounted for 88.7 percent of the total domestic production of this commodity in 2001 (35,230 / 39,705) x 100 = 88.7 percent.
The Input-Output Tables display the demand for various goods and services which may be used as intermediate inputs (goods and services that are purchased by an industry from other industries to produce its outputs) or for final consumption by persons, governments, etc. So, the question, “who buys what?” is answered by referring to the Use (Input) and Final Demand Matrices. Elements in the row of the Use Matrix show the use of a particular commodity by various industries whereas elements of the same row appearing in the Final Demand Matrix indicate its use by various categories of final demand. For example, in 2001, altogether, industries used petroleum and coal products worth $25,560 million – the sum of row 26 across the columns of the Use Matrix. Individual entries across row 26 represent individual uses by the industry shown as a column.
(iii) Composition of outputs and inputs
The detailed presentation of data on outputs and inputs presented in this publication permits analysis of the composition of outputs and inputs of each industry over time. Once again, using the Make Matrix, one is able to observe, say for 2001, that the total output of the manufacturing industry (column 8) amounted to $583,799 million. Of this, petroleum and coal products accounted for $35,230 million which represents 6.0 percent of the total output of that industry. In a similar manner, inputs that are needed to produce outputs of industries can be identified in the Use Matrix. For example, column 8 of the Use Matrix lists various inputs needed to produce the output of the manufacturing industry. The magnitude of changes in the value of inputs over time can be identified and analyzed.
(iv) Gross domestic product (GDP) by industry
Generally speaking, the gross output of an industry consists of the value of the goods and services produced by that industry: the value of steel produced by the steel industry; the value of cars and trucks produced by the motor vehicles manufacturing industry; etc. A brief reflection tells us that, for example, the production of steel requires coal and iron ore and other inputs. These goods and services are not the products of steel mills. Rather, they are produced by other industries and used by the steel mills as their inputs. Thus, a measure of the unduplicated production of an industry should exclude all intermediate inputs purchased from other industries. This measure is called value added or Gross Domestic Product.
The Use Matrix displays all the costs incurred in production by an industry: the goods and services used as intermediate inputs; indirect taxes and subsidies; and returns to the factors of production (rows 56 to 59 of the Use Matrix), namely: wages and salaries, supplementary labour income, mixed income and other operating surplus.
(v) Gross domestic product (GDP) for the total economy
There are two measures of the Gross Domestic Product which provide valuation of the production of goods and services at different levels. They are:
Basic Price valuation represents the earnings of the factors of production as measured by the costs of labour (wages and salaries, supplementary labour income); capital inputs (mixed income and other operating surplus); plus indirect taxes on factors of production less subsidies on production. Indirect taxes on products(row 52) are a cost to the industry and are reflected in the market price of goods and services produced by the industry. Subsidies on products (row 53), where applicable, have the effect of lowering cost to the firms receiving them, and thus reduce the market prices of goods and services produced by the industry. They are shown as negative (-) entries in the Use (Input) Matrix. Both indirect taxes and subsidies on products and on production affect the market price valuation of goods and services.
It should be noted that indirect taxes are paid not only by the business sector but also by other sectors of the economy which are covered in the Final Demand Matrix of this publication. Subsidies, by definition, apply only to the business sector of the economy.
The GDP at Market Prices as calculated from a revenue approach is called income-based GDP and from an expenditure approach is called expenditure based GDP. The Input-Output Accounts also incorporate components of expenditure-based GDP and is obtained by summing up personal expenditure, capital formation, government current expenditure, change in the value of inventories and net exports (exports less imports) appearing in the Final Demand Matrix. In the Final Demand Matrix, these categories are shown at a finer level of detail.
(vi) Domestic availability of goods and services
It may be of interest to know the domestic availability of a particular commodity (or all goods and services) for one or many years. The domestic availability of a commodity is defined as total production less exports plus imports (assuming no inventory change). These values can be read directly from the Input-Output Tables. In 2001, the domestic production of petroleum and coal products was $39,705 million (last column of the Make (Output) Matrix) while total exports were $9,361 million (column 14 of the Final Demand Matrix) and imports were $4,854 million (column 16 of the Final Demand Matrix). Thus, the domestic availability of this product can be derived as: $39,705 million domestic production - $9,361 million exports + $4,854 million imports = $35,198 million.
(vii) Import share
The data used in the preceding section also allow the calculation of the import share of a commodity, that is, what portion of the domestic availability of a commodity comes from foreign countries. From these data, the import share for the same commodity can be calculated as follows:
(Total imports / Total domestic availability) x 100
that is for 2001:
(4,854 / 35,198) x 100 = 13.8 percent.
(viii) Export share
For countries with considerable foreign trade like Canada, there is a keen interest in measures of export intensity. A simple indicator of such an intensity is the proportion of total domestic production that is exported. Using the same commodity as in the previous illustrations, 23.6 percent of the domestic production of petroleum and coal products was exported in 2001 as follows:
(9,361 / 39,705) x 100 = 23.6 percent.
(ix) Indirect taxes and subsidies
In compiling Input Output Accounts at producers’ prices and purchasers’ prices, the costs or margins that arise between the value received by the producer (i.e. the value that covers all costs of production) and the value paid by the purchaser must be identified. One of these margins is the value of commodity taxes paid by the purchaser, be it an industry or a category of final demand. All commodity indirect taxes and all other indirect taxes are classified to the industry or final demand category paying them. The total of such taxes agrees with values published in the National Economic and Financial Accounts (Catalogue 13-001-XPB) but the Input Output Accounts provide additional information such as the industries and final demand categories which pay these taxes. Similarly, in the Input-Output Tables, subsidies are allocated to the industries receiving them. Also, from 1986 onwards, subsidies are broken down into two categories: commodity subsidies and non-commodity subsidies. These details and their routing in the Input-Output Accounts make additional analyses possible, e.g., to determine the effects of taxes and subsidies, to build certain policy simulation models, etc 2.
Data Accuracy Measures
Data from Input-Output Accounts presented in this publication are assessed for relative statistical reliability and assigned a rating that indicates the level of confidence with which they may be used. The ratings refer to data for the latest benchmark year for which estimates are published. The ratings are ‘A’ for the most reliable data, ‘B’ for reliable data, and ‘C’ for data that is less reliable but still acceptable. The ratings are determined for outputs, intermediate inputs and GDP components of each Link or ‘L’ level input-output industry. The ratings apply to entire vectors, rather than individual data points. For instance, a rating of ‘A’ or most reliable applies to all outputs of the Crop and animal production industry, whereas the industry’s intermediate input vector as a whole is assigned a ‘B’ rating. The industry’s GDP components are also assigned a ‘B” rating.
Data reliability ratings are a product of data integration and analysis inherent in the compilation of input-output tables. They rely both on the quantitative attributes of the survey and administrative data sources that are used, such as sample size, response rate and coefficient of variation, and on the expert judgement of analysts who undertake data integration of various source data. In general, the highest quality rating is assigned to a data set that originates from a survey or administrative source with the largest sample size and smallest undercoverage that requires no indirect estimation of missing detail. A reliable or ‘B’ rating is assigned to a data set estimated from source data with some but not all of the above attributes. Finally, data sets with a ‘C’ rating involve significant application of indirect estimation techniques and rely on source data with small samples, undercoverage, or both.
Text table 2