Overview

Industry definition
Output, supply and demand
Monthly projector
Data
Deflation

Industry definition

This description specifies which establishments constitute a particular industry. The definitions are given by an industrial classification called North American Industry Classification System (NAICS). NAICS defines industries as groups of establishments which are engaged in similar production processes. This means that establishments are grouped into industries according to similarity of input structures, labour skills or the production processes employed. If the necessary data are available, the most important activity or output is defined as the one which contributes most to the establishment's value added. Otherwise, the primary activity is determined using other variables such as employment or gross revenue figures. It is useful to note that the group of establishments comprising a particular industry may change over time as establishments emerge, go out of business, or change their production processes. In the CSNA, a distinction is also made at the establishment level as to whether it belongs to the business or the non-business sector. The business sector consists of establishments whose output is sold at prices which are economically significant (i.e., prices which have a significant influence on the amounts producers are willing to supply and purchasers wish to buy). Those establishments which provide goods and services at prices which are not economically significant are assigned to the non-business sector. For more detail on NAICS see North American Industry Classification System, Canada 1997, Statistics Canada, Catalogue no. 12-501-XPE, March 1998.

The definition of industries in this report is based on the 1997 version of the North American Industry Classification System (NAICS).

Top of Page

Output, supply and demand

This is a summary of the industry's economic activity, viewed from the perspectives of supply and demand. Values are derived from the annual input-output tables and presented here in millions of current dollars.

Output is the total value of the goods and services produced by establishments classified to the industry. The concept of output embraces all products that are sold, bartered or supplied to other establishments within the same enterprise as well as those products which are provided to other institutional units either free of charge or at prices which are not economically significant. Products which are retained by the producing establishment are also considered to be output if they are used for own final consumption, own gross fixed capital formation or are simply added to inventory prior to selling at a later date.

Normally, output is recorded when the production of a good or service is completed. In some cases, however, if the duration of the production process is longer than the accounting period, the concept of output also includes the value of unfinished products. For example, the completion of such large items as ships, cranes, buildings and bridges may take months or years. In such cases it is recognized that output is being produced continuously and partially completed products are recorded as work-in-progress.

Values of output can be expressed at prices including taxes, excluding taxes, including revenue received as subsidies, not including revenue received as subsidies. The SNA 1993 recommends expressing the value of output as equal to the amount purchasers pay to producers, minus any product tax payable, plus any product subsidy receivable. This is called valuation at basic prices. In Canada, industry surveys conducted by Statistics Canada ask producers to value their sales at prices before adding any product taxes and, in cases when producers receive product subsidies from governments, they are asked to value their products at prices that are actually paid by purchasers and retained by the producers.

These different valuations may be explained by the following example of a fictitious transport system. Suppose passengers of a transit system pay $0.90 for a bus ticket and this price includes $0.05 as tax. Further assume that the cost of such a ticket to the transit system is $1.30 but the city's government provides subsidy worth $0.45 in order to help the transit system to continue operating. The SNA 1993 will record the value of output by this transit system at basic prices, i.e., at $1.30 per ticket ($0.85 from the purchasers plus $0.45 as subsidy from the government). Statistics Canada, on the other hand, will record the value of output at $0.85 per ticket and call this valuation at modified basic prices.

As taxes on products are far more prevalent and significant than subsidies in Canada, the value of most products is reported at basic prices, except a handful of items which are at subsidized prices, not at prices plus subsidies. Most of the output in the Canadian input-output tables is therefore valued at basic prices and in the case of a handful of industries which receive product subsidies, the valuation of output uses modified basic prices. While this modification by the CSNA represents only a slight departure from the SNA 1993 guidelines, valuing output at modified basic prices has the advantage that the prices of products are observed prices as received by the producers and paid by the purchasers. Such prices are listed on the invoices, hence the valuation of transactions is verifiable from enterprise records. Values of output in this documentation are presented at modified basic prices.

Value added, as noted above, is the newly created additional value by which an industry has augmented the value of material and service inputs it purchased from other industries. In other words, it is that portion of an industry's output that has been created by the industry itself. Value added is captured by subtracting from the value of output the value of intermediate inputs, whether purchased from other domestic industries or imported.

Though output is presented in Canada at modified basic prices for those handful of industries receiving product subsidies, value added is presented at basic prices, which is exactly the same concept as recommended by the SNA 1993. Historically, Canadian estimates of value added were published at factor cost (cost of labour and capital) and this cost did not include the net taxes on factors of production (taxes on factors of production minus subsidies on factors of production). With the historical revision of 2001, the CSNA adopted the SNA 1993 recommendation and since then value added is recorded at basic prices.

Value added at basic prices is higher than the traditional value added at factor cost by the amount of other taxes on factors of production (such as property and payroll taxes) less other subsidies provided to factors of production (such as labour training). Value added at basic prices of a producer is therefore equal to its output valued at modified basic prices less its intermediate consumption at purchasers' prices, plus any subsidies receivable by the producer on products. This is summarized in the following numeric example.

The preference for measuring value added at basic prices stems from the notion that this valuation better reflects the full cost of factors of production that is incurred by producers. International comparability also played an important role leading to this conversion and the shift to basic prices put Canada 's value added by industry on an equivalent basis with other member countries of the Organization for Economic Cooperation and Development (OECD).

Share of total economy indicates the industry's contribution to total GDP.

Share of sector shows the industry's relative importance among those industries which are grouped together according to a broad type of economic activity (for example mining, manufacturing, construction) as defined in NAICS.

Significant products of the industry represent the most important goods and services in terms of output value. Note that in many instances the products of one industry are also produced by some establishments which are classified to other industries. In some cases those establishments which are primarily engaged in producing an industry's significant products and thus constitute the particular industry represent only a small share of the total value of these products produced domestically. This occurs when other establishments which also produce relatively large amounts of the same products but only as a secondary activity are assigned to other industries. As noted earlier in the paragraph explaining industry definitions, the primary activity of an establishment is defined as the activity with the largest value added and the establishment is classified to the industry corresponding to this that activity. Activities performed by establishments in addition to the primary activity are called secondary activities.

Total supply of these products shows the total value of these significant products originating from all sources that are available for use in Canada. Domestic production shows the total value of these significant products produced by all establishments in Canada, whether classified to this industry or not. The term domestic indicates that production takes place within the boundaries of Canada, regardless whether the factors of production are owned by Canadians or non-residents. Imports are added in order to complete the supply picture. The purpose of showing the total supply of an industry's significant products is to give a measure of the market share an industry holds for its major products and to reveal the extent of the competition brought on by other industries, domestic and foreign.

Major users of these significant products is a list of purchasers who consume most of the industry's output. Some of the major purchasers may be other domestic industries who use up these products as intermediate inputs in their own production process. If a significant portion of the output of the producing establishment is supplied to another establishment belonging to the same industry for use as intermediate inputs into the latter's production, the industry itself is also listed among the major users. Other important purchasers may be households, governments, business enterprises and residents of other countries, who purchase these products for either final consumption or gross fixed capital formation. The purpose of showing the flow of an industry's primary products to the major users is to associate the industry's productive activity with the consuming industries and the needs (or willingness to buy) on the part of the various final demand categories.

Top of Page

Monthly projector

Under this heading is a description of the measurement of monthly constant price value added.

Before discussing the approach to measuring monthly value added, here is a brief description of the annual methodology within the framework of the input-output tables. For all but the most recent two years, annual estimates of GDP by industry are derived from data which are typically supplied by annual surveys or censuses, and from administrative (taxation) data. For the most recent two full years and for sub-annual periods, the lack of applicable data, particularly data on intermediate inputs, precludes a value-added calculation and estimates of GDP in these periods are projections, based on such proxy indicators as output or labour input, usually obtained from monthly surveys.

Although the annual and monthly estimates of value added are based on the same concepts, definitions and classification schemes, the methodologies underlying the two sets of statistics use different data sources and techniques. As a result, annual input-output based value added and the yearly totals of the independently produced monthly estimates are not identical. The difference between the two is eliminated by integrating the monthly series into the annual input-output based annual values. This 'blending process', called 'benchmarking', involves distributing annual values across sub-annual periods using the growth rates in the monthly estimates. In the most current years when input-output tables are not yet available, estimates of monthly value added are projections which display monthly growth rates similar to those of the monthly indicators.

Annual estimates of value added are derived from the fully balanced input-output tables. The method of derivation depends on whether an industry belongs to the business or the non-business sector. As noted earlier, the business sector encompasses establishments whose output is sold at prices which are economically significant (i.e., prices which have a significant influence on the amounts producers are willing to supply and purchasers wish to buy). Because business sector industries both sell their products and purchase their intermediate inputs in the market, data on the full range of inputs and outputs are generally available. It is feasible therefore to measure value added of business sector industries residually as the value of the outputs less the value of intermediate inputs.

As noted earlier, the calculation starts with valuing output at modified basic prices. The valuation of intermediate inputs, on the other hand, uses purchasers' prices. Purchasers' prices represent amounts that are actually paid by the producers for their intermediate goods and services. Taking the difference between output at modified basic prices and intermediate inputs at purchasers' prices and adding subsidies on products to this residual amount gives value added at basic prices. Note that an alternative method of calculating current dollar annual value added by industry at basic prices is adding wages and salaries, supplementary labour income, mixed income and other operating surplus (the sum of these items constitutes GDP at factor cost), plus taxes less subsidies on the factors of production.

Non-business sector industries consist of non-market producers who produce goods and services for the benefit of persons, businesses or the community as a whole and supply their products either free of charge or sell them at prices which are not economically significant (i.e., prices which do not have a significant influence on the amounts producers are willing to supply or purchasers wish to buy).

The measurement of output of non-business sector industries cannot be based on direct observations of transactions between buyers and sellers on markets. Some of the non-business sector industries provide services to the community as a whole, for example defence, police, fire fighting, correctional or administrative services. These services are consumed by society collectively and are typically financed from taxation revenues received by governments. Since the consumption of collective services is not represented by money-exchange transactions on markets, it is not possible to determine their market value. Some non-business sector industries do produce goods and services which could be sold on markets, but for social, economic or political reasons, non-business industries choose to either supply their products and services free of charge or if they sell their products, they do so at prices which are not economically significant. Such prices are often intended to raise some revenue in order to reduce the cost of production, but do not necessarily reflect the total cost of production or the market value of the products. Because prices which are not economically significant do not reflect relative production costs or relative consumer preferences, they do not provide a suitable basis for valuing the outputs of non-business sector industries. Goods and services produced by non-business establishments are therefore valued based on their costs of production, including consumption of gross fixed capital formation.

Constant price annual estimates of value added of business sector industries are derived by a method known as double deflation. Double deflation involves deducting constant price intermediate inputs from constant price outputs. The constant price estimates of intermediate inputs and outputs are calculated by deflating each commodity separately with its own appropriate price index.

Since double deflation is not an available option for calculating constant price GDP for the non-business industries, these estimates are derived by dividing current values of labour income, depreciation of capital stock and taxes by their appropriate price indexes.

Though most industries comprise establishments entirely belonging either to the business or the non-business sector, a few include establishments from both sectors. Examples are radio and television broadcasting, education and health.

For an extensive description of the annual methodology readers should refer to The Input-Output Structure of the Canadian Economy, Statistics Canada, Catalogue no. 15-510, The Input-Output Structure of the Canadian Economy in Constant Prices 1961-1981, Statistics Canada, Catalogue no. 15-511, Service Industries in the Canadian Input-Output Accounts (Current Prices), Sources of Data and Methods of Estimation, Statistics Canada, Catalogue no. 15-601, No. 2 - occasional, and A Guide to Deflating the Input-Output Accounts, Sources and Methods, Statistics Canada, Catalogue no. 15F0077GIE, 2001.

Monthly estimates of value added are projections, derived from a collection of output or input indicators. This projector-based approach is necessary because complete monthly information on outputs and intermediate inputs is not available. Constructing value added as the difference between intermediate inputs and outputs requires data that are collected only by annual surveys, thus, the derivation of monthly value added must rely on a less comprehensive data base, usually provided by monthly surveys. Monthly surveys do not cover all commodities produced by a certain industry nor all producers of a certain commodity, and usually collect only scarce information on intermediate inputs. Nevertheless, monthly surveys do provide sufficient data on each industry to serve as suitable indicators of the monthly movement in outputs or inputs.

The basic assumption underlying the projector method is that changes in outputs or inputs acceptably reflect growth rates in value added. In constant prices, this assumption is reasonable since technological advances which permit a different amount of output to be produced from the same amount of inputs normally occur slowly. Changes in the production process, for example shifts in production from goods that require a high degree of fabrication to goods that require less, or substitution of an input by another, may cause some deviations in the relationship between value added and outputs. Thus, indicators based on broad categories of products or materials may not always reveal changes in value added accurately. The size of the divergence depends on the degree of homogeneity of the industry. The more diversified the production of an industry is, the more sensitive value added is to the changes in the product mix.

The most frequently used proxy indicator is output. Generally, estimating changes in value added at constant prices using output as indicator yields a close approximation to the change in constant price value added. Although the use of output as a single indicator may lead to skewed results if intermediate inputs measured at constant prices do not change in the same proportion as output at constant prices, in the absence of monthly information on intermediate inputs it has proven to be an acceptable method of estimating constant dollar value added. Approximately two thirds of total GDP is derived based on indicators of output.

In instances when data on output are not available, labour input is chosen as an alternative indicator. For the most part, approximating changes in value added by changes in labour input is subject to the same sort of assumptions and limitations that apply to output as an indicator. Even though the labour series usually relate fairly closely to actual work done, they fail to take account of changes in productivity. If, for example, the production process becomes more efficient because of an increase in labour skills or the use of more or better capital equipment, the employment-based indicator will underestimate the increase in value added. Because employers tend not to adjust their skilled labour force in response to short term fluctuations in demand, labour input may be steady whereas value added is in fact changing. Despite the potential pitfalls, changes in employee compensation at constant wage rates, or simply changes in the number of employees, yield a fairly close estimate of changes in constant price value added over a short-term periods. Approximately one quarter of total GDP is estimated based on labour input.

For a few industries, both output and labour input indicators are used in conjunction, whereas in some cases, an indirect approach is applied instead of a more direct measurement of either outputs or inputs. An example is the truck transportation industry. Output by this industry is estimated based on gross revenues received from direct purchases of freight services by industries for intermediate use. Such expenditure by businesses is projected based on the movement in output of the consuming industries. The assumption is that the amount paid for truck transport is a fixed proportion of an industry's output. Another case in point is the residential construction industry. Output of this industry is estimated based on the value of work-put-in-place each month. This monthly value is derived from data on housing starts, completions and average values of building permits for each type of dwelling, such as single dwellings, semi-detached dwellings, row housing and apartments.

For a very small number of industries relevant sub-annual information is entirely lacking. In such cases, monthly estimates of value added are distributions of the input-output (IO) based annual benchmarks across sub-annual periods. In the total absence of sub-annual data the simplest method of calculating a monthly series whose level is dictated by the annual benchmarks could be dividing the annual values by twelve. However, this approach would introduce breaks between the months of December and January. In order to avoid such artificial discontinuities the method of quadratic minimization is used to find a suitable monthly series with a smooth transition from one year to the next. In the most current period when IO benchmarks are not yet available, projected monthly estimates of value added are fitted on a more or less smooth line which continues the general direction of the overall course of production beyond the IO years.

Most notable example for estimating monthly value added from annual output measures is Crop production. Agriculture production is a special type of activity as the process of producing grain lasts several months. Additionally, final output is only known after the harvest is completed in the fall. Similar to other industries whose production process stretches over several months or even years, the concept of monthly agricultural production is based on the notion that output is being produced continuously and monthly output of field crop is therefore viewed as partially completed production. Monthly output is therefore considered as an addition to work-in-progress and is measured by distributing annual crop values across months. The distribution is carried out in two stages and relies on a quadratic minimization method. First, a quarterly series with a smooth transition from one year to the next is obtained from the annual values for each type of crop. Special treatment is given to years with bumper crops or droughts and as a result, most of the production excess or shortfall in those years is reflected in the harvest quarter. Then from each of the quarterly series a monthly series is calculated that has a smooth transition from one quarter to the next and these monthly grain estimates serve as proxies in the monthly projection system.

As annual grain crop data become available only after the crops are harvested in the fall, using such data as indicators of monthly output poses the special problem of estimating value added in the months prior to the harvest period of the most current year. The CSNA's answer to this question is to project an estimated annual value for each type of grain in the current year using information from a series of probability surveys of farmers collecting data on seeding intentions, seeded and harvested area, yield, production and stocks for the principal field crops, etc.

The initial estimates of crop production for the current year are made with the January estimates. They are based on data found in the Agriculture and Agri-Food Canada publications Canada: Grain and Oilseeds Outlook and Canada: Pulse and Special Crops Outlook which are published in February. The harvested area in hectares for each crop is forecast as well as the yield in metric tons per hectare, from which the total production in thousands of metric tons is derived, province by province. The forecasts reflect trends in harvested areas and yields, adjusted using special conditions in particular provinces. For example, in Saskatchewan and Alberta yields were forecast to be below trend in 2004/05 due to low soil moisture reserves. New forecasts are made in May which incorporate the results of Statistics Canada's seeding intentions survey Field Crop Reporting Series, Catalogue no. 22-002. Afterwards, these forecasts continue to undergo a series of revisions incorporating updates collected by subsequent runs of this survey in July, September and November.

As noted earlier, although a wide range of input data is used to measure value added by industry, data coverage is generally incomplete. In addition, the concepts, classifications, accounting conventions and valuations methods at the input data level may not always coincide with those of the CSNA. Therefore, methodologies using projectors, however complex, are merely compromises from the theoretically appropriate model. The use of a projector system for calculating value added by industry requires not only the firm implementation of national accounting concepts and definitions of the CSNA, but also the ongoing review of projectors and methodologies to detect biases, breaks in continuity, or any other changes in consistency and suitability. Whenever quality evaluation of the projector system deems it necessary, improvements are made to existing methodologies and occasionally entirely new projectors are implemented.

One of the most important sources of feedback on the quality and consistency of the value added estimates is provided by economic analysis. A projector is considered reliable if economic intelligence is supported by the value added estimates. In other words, the underlying data consistently reflect any factors that affect output of a particular industry, such as strikes, large contracts, government policies and programs, etc.

Given that the quality of the monthly estimates of value added by industry depends not only on the source data but also on the quality of transformations such as deflation, seasonal adjustment and benchmarking, projectors are regularly evaluated from these perspectives as well. The analysis uses both quantitative and non-quantitative assessment methods. Quality dimensions that can be expressed in quantitative terms are appraised using statistical tools. An example is revision analysis. Since revisions usually relate to better data due to higher response rates, studying the frequency and the size of the revisions provides important measures of quality. The quality of seasonal adjustment is appraised by measuring how well identifiable is the seasonal variation in the projector and how much residual seasonality remains in the final seasonally adjusted estimates. Test values showing highly identifiable seasonality and the absence of residual seasonality are signs of good quality seasonal adjustment. Regression analysis is another statistical tool frequently used for examining the relationship between the monthly projector and the input-output based annual benchmarks. Non-quantitative assessments of how well a proxy works are largely subjective evaluations of the conceptual adequacy of the monthly source data.

Top of Page

Data

This field lists the data sources that are feeding into the monthly calculation of value added. Since GDP by industry measures production of all goods and services in the Canadian economy, a great quantity of data are used from literally hundreds of different sources. While the majority of the data sources are monthly surveys conducted either by Statistics Canada or by other government agencies, business associations, individual companies, trade magazines, stock exchanges and various Internet web sites of business organizations also serve as valuable data providers.

The monthly projector system relies on two main types of data on goods and services: physical quantities and nominal values. Quantities of goods may be described by the number of items and physical characteristics such as the product's weight, volume or length. These units are normally used to describe quantities of homogeneous goods, for example tons of grain, litres of motor gasoline, and so on. In case of a complex commodity, the description of the quantity produced is a bundle of attributes and in addition to size, colour, material composition and method of production, the physical properties of a commodity may also include qualitative features such as the geographical location or even the purpose of production. A quantity indicator of a service may use quantities of goods or the number of people who were affected by the service. An example is the number of passengers or quantities of goods that were transported for a particular distance. Nominal values of goods entering the value added calculation are mostly current dollar values of shipments and inventory accumulations, and in the case of services, revenues and sales by the service provider. Input type projectors generally rely on the number of employees who are engaged in the production process.

Top of Page

Deflation

Due to the absence of monthly inputs and outputs, the monthly deflation methodology cannot replicate the double deflation technique that is used in the input-output system. Instead, monthly constant price estimates are calculated by either valuing quantity data using their prices (or unit values) in the base year or by removing the effects of price fluctuation from nominal values with their appropriate price indexes. The choice of which deflation technique to use is determined by the nature of the projector.

Base year valuation involves the summing up of current quantities valued at the prices of a base year. The algebraic expression of this method of building a constant price series is:

where q is the quantity and p is the price, t stands for the current period and 0 for the base period and the summation covers the entire range of goods and services produced by a particular industry. Generally, in an attempt to minimize biases caused by shifts in a mix of commodities, the basic unit of output is defined at as fine a level of detail as possible. Commodities with different physical characteristics and products of varying quality are generally taken as different products and are given their own suitable prices.

An alternative method for valuing current production in the prices of a fixed period in the past is price deflation. Price deflation removes price changes simply by dividing current values of goods and services by their appropriate price indexes. This method is widely used for many service producing industries whose output is either not measurable in terms of quantities or for which quantity data are not available. It is also used for a large number of manufacturing industries whose output projectors are current price shipments and inventories.

Generally, industry deflators in the monthly GDP system are composite indexes which describe the price movements for various groups of goods and services produced by a given industry. The majority of industry deflators are weighted averages of price indexes of commodities and Consumer Price Indexes (CPI). For example, deflators for the total commodity output of individual manufacturing industries are computed by weighting the Principal Commodity Group (PCG) price indexes by the production values of those PCGs in each industry. The weights are taken from the IO tables. For the period without IO tables, that is the most current 32 to 43 months, the weights remain based on the last available IO table. The calculation of an aggregate price index I_t describing the price change for the entire output of an industry is represented by the following algebraic expression:

where

is the current value of commodity i derived from the IO tables,

is the price index of commodity i, t is the current month and i is the base period.

This type of deflator is used most extensively for deflating output of those manufacturing industries for which the movement in constant price output is assumed to be represented by the month-to-month growth rates in constant price manufacturing shipments adjusted by the change in constant price inventories. Deflators for shipments are weighted averages of PCG price indexes as described above, while deflators of inventories are calculated as moving averages of the shipment deflator. For inventories of goods in process, the number of terms in the moving averages corresponds to the duration of the production process whereas those for finished products reflect the length of time a finished product remains in stock. This number is referred to as inventory turnover period and is calculated as the average stock to sales ratio by dividing end-of-month inventory levels by monthly sales. Since opening inventories are equal to closing inventories of the previous month, deflators for closing inventories of the preceding month are used to deflate opening inventories of the current month.

Constant price estimates of output by many service producing industries are calculated using weighted aggregations of the various appropriate Consumer Price Indexes (CPI) as deflators. In some isolated cases such as Radio and Television Broadcasting industry, or Computer and Peripheral Equipment Manufacturing industry, specific price indexes are used which are obtained from price observations such as advertising rates and import prices.