• Purpose of the classification
• Historical background
• The underlying concepts
• The classification structure and coding system
• Relationship to other classifications
• Summary of changes from NAPCS Canada 2012 to NAPCS Canada 2017
• Product data in the Canadian statistical system
• The methods for implementing the classification

Purpose of the classification

The North American Product Classification System (NAPCS) Canada is a classification of products (goods and services) designed primarily for use in statistical programs. It is Statistics Canada's official standard for the collection, processing and dissemination of product statistics in its economic, business and trade statistics programs. This includes, for example, statistics on the value of exports and imports by type of product, the value of industry production and consumption by type of product, and industrial product price indices.

Statistical classifications are comprehensive structured lists of mutually exclusive categories. In practice, this means that there is always a category in the classification if the object falls within the scope of the classification, and that the object can be classified in only one category. The section titled "The underlying concepts" further discusses the object and scope of NAPCS Canada.

The structure of NAPCS Canada is hierarchical. This type of classification system enables the collection, analysis and publication of data at different levels of detail, in a standardized way. The section titled "The classification structure and coding system" discusses the structure of NAPCS Canada in greater detail.

The purpose of standard classifications is to support the integration of data obtained from multiple sources by organizing the documentation, collection, processing, presentation and analysis of data in a systematic manner. Classifications are essential elements of a coherent and efficient statistical system.

NAPCS Canada has been developed to support the integration of product statistics. NAPCS Canada 2017 was approved as the departmental standard on September 19, 2016. The development of NAPCS Canada was a progressive process done over a period of time. The next section provides an overview of its development and implementation history.

Historical background

The first version of NAPCS Canada, known as version 0.1, was published in 2007. It was the first in a series of developments to re-organize the system used to classify product data in Statistics Canada's business and trade statistics programs.

The development of the classification started a few years earlier as a joint project of the national statistical agencies of Canada, Mexico and the United States. The NAPCS project followed the adoption by the three countries of a common industry classification, the North American Industry Classification System (NAICS) in 1997. The purpose of the NAICS project was to develop a standard that allowed comparisons of industry data among the three participating countries. The development of NAPCS was meant to serve the same purpose but for product statistics.

In the beginning, the trilateral project focused on developing a classification of service products. At that time, the national statistical systems produced much less information for services than for goods, and the development of a detailed standard classification for service products was seen as a necessary first step towards improving that component of the statistical system.

The first version of NAPCS Canada was largely based on unpublished trilateral work. The scope of the classification was limited to the products (outputs) of selected service-producing industries. Though presented as provisional, parts of the classification had been tested and used to collect product data in selected annual service industry surveys.^{Footnote 1} The classification also provided a basis to define service products in the Input-Output Commodity Classification (IOCC).^{Footnote 2} The simultaneous integration of NAPCS Canada in the collection system and the economic accounts made a significant contribution towards improving the coherence of product data.

At the time, NAPCS Canada (version 0.1) was complemented by several specific purpose classifications for the collection and publication of statistics concerning the supply and use of goods. Some of these classifications had the status of departmental standards, for instance the Annual Survey of Manufactures (ASM) - List of Goods, the Canadian Export Classification (CEC) and the Customs Tariff (CT). Others were not published as standard classification systems, yet were instrumental in the collection of product data. This included, for example, the classifications used to collect and organize data on the production of agricultural and mining products, and the classifications used to produce raw material price indices (RMPI) and industrial product price indices (IPPI).

The main weakness of the multi-classification system was the lack of integration between them. They had been developed at different points in time, sometimes on the basis of different criteria, to serve different purposes.

In particular, the lack of comparability between data on the domestic production of goods and the exports and imports of those same goods was seen as a significant shortcoming. The second phase in the development of NAPCS Canada addressed that issue.

NAPCS Canada 2007 was published in 2011. The scope of NAPCS Canada 2007 was limited to the universe of tradable goods, and it therefore complemented version 0.1 of NAPCS Canada whose scope was limited to selected services.

NAPCS Canada 2007 consisted of groups (3 digits) and classes (5 digits). The class level was developed to integrate the classifications used for production and international merchandise trade statistics. The approaches to integrate these classifications into NAPCS Canada had to differ however, due to constraints in their implementation.

In the case of international merchandise trade classifications, the approach consisted of using concordances. The classifications of exports and imports - the Canadian Export Classification (CEC) and the Customs Tariff (CT) - are derived from the Harmonized Commodity Description and Coding System (HS), a mandatory international classification maintained by the World Customs Organization (WCO). The concordance was developed between the classes of NAPCS Canada 2007 (5 digits) and those of the CEC and CT. This approach allowed for the presentation of import and export statistics on the basis of the HS as required by international agreement and on the basis of NAPCS as required for coherence of production and trade statistics within the Canadian statistical system.

The systems used to classify the domestic production of goods on the other hand are developed and maintained by Statistics Canada. The intent was to integrate those classifications as well as those used for industrial producer prices indices (IPPI) and raw material price indices (RMPI) at a level below the class level in a future version of NAPCS Canada.

The 5-digit classes of NAPCS Canada 2007 were aggregated into 3-digit groups. The group level was designed to serve two purposes:

• Provide analytical groupings to publish selected statistics.
• Provide a basis to define higher level categories of alternate structures or variants. The first variant was for the presentation of export and import statistics. The new higher level categories replaced the summary import groups (SIG) and the summary export groups (SEG) that had been in use for several decades.

International merchandise trade statistics were first published on the basis of NAPCS in October 2012.

NAPCS 2007 represented an important step towards the greater harmonization of product statistics. In addition to providing a framework to integrate existing production and international trade statistics, it established an approach to further develop and complete the system. The approach was fully implemented in NAPCS Canada 2012 (version 1.0), published in 2013.

The main accomplishment of NAPCS Canada 2012 was to merge previous versions of NAPCS Canada and program-specific classifications^{Footnote 3} into a single system covering all goods and services. The classification system was designed to accommodate the needs and constraints of several business and trade statistics programs. The design of the classification is discussed in more detail in the section titled “The classification structure and coding system”.

In addition to a traditional standard hierarchical structure, NAPCS Canada 2012 embedded a system to create variants of the standard classification for programs that required a more detailed classification (extension variants), or a different aggregation structure to meet different analytical needs (regrouping variants). The design of variants is discussed in more detail in the section titled “Classification variants”.

Two updates to NAPCS Canada 2012 (versions 1.1 and 1.2) have been released, both in 2015. The updates were meant to improve the relevance and coherence of the classification. The most significant change was the addition of definitions for all categories, introduced in version 1.2. At the lowest level of the classification, definitions include a descriptive text, as well as illustrative examples, inclusions and exclusions where necessary. This finalized the development of NAPCS Canada into a complete standard statistical classification.

The release of NAPCS Canada 2017 constitutes the first revision of the classification coordinated with the revision of NAICS. Future revisions are planned every five years. The revision process encourages the active participation of users of the classifications.

The underlying concepts

Statistical classifications are built around three basic concepts: the object classified or statistical unit, the scope or universe of the classification, and the criteria used to group statistical units in standard categories. In the case of NAPCS Canada, two of these concepts – the statistical unit and the scope of the classification - are modelled on concepts of the 2008 System of National Accounts (SNA).^{Footnote 4}

The SNA provides the set of definitions, classifications and accounting rules to support the production of internationally comparable economic accounts. The concepts underlying the production account are particularly relevant for product classifications. They are described in detail in Chapter 6 (The production account) of the 2008 System of National Accounts (SNA) manual. The sections that follow provide a brief overview and relevant excerpts from it.

The statistical unit

The following are relevant excerpts from the 2008 SNA manual concerning the concept of product in economic accounts:

"Products are goods and services (including knowledge-capturing products) that result from a process of production." (paragraph 6.14)

"Goods are physical, produced objects for which a demand exists, over which ownership rights can be established and whose ownership can be transferred from one institutional unit to another by engaging in transactions on markets. They are in demand because they may be used to satisfy the needs or wants of households or the community or used to produce other goods or services." (paragraph 6.15)

"Services are the result of a production activity that changes the conditions of the consuming units, or facilitates the exchange of products or financial assets. These types of services may be described as change effecting services and margin services respectively. Change-effecting services are outputs produced to order and typically consist of changes in the conditions of the consuming units realized by the activities of producers at the demand of the consumers. Change-effecting services are not separate entities over which ownership rights can be established. They cannot be traded separately from their production. By the time their production is completed, they must have been provided to the consumers.... Margin services result when one institutional unit facilitates the change of ownership of goods, knowledge-capturing products, some services or financial assets between two other institutional units." (paragraphs 6.17 and 6.21).

"Knowledge-capturing products concern the provision, storage, communication and dissemination of information, advice and entertainment in such a way that the consuming unit can access the knowledge repeatedly ... Whether characterized as goods or services, these products possess the essential common characteristic that they can be produced by one unit and supplied to another, thus making possible division of labour and the emergence of markets." (paragraph 6.22)

From these excerpts, the following elements emerge as fundamental to identify products:

• they are goods and services that result from production processes;
• they are in demand to satisfy the needs of households or the community, or used to produce other goods or services;
• they are produced by one unit and supplied to another; and
• they are transacted (bought, sold, transferred or placed in inventory).

The use of these criteria in combination excludes services produced for own use by households.

As is often the case, there are exceptions to the rules. NAPCS Canada includes a few products that do not result from production processes, namely recovered products such as recycled paper and plastics and used merchandise. These products result from a consumption process rather than a production process. They are however in demand and their inclusion allows for the measurement of significant transactions in selected surveys that use NAPCS Canada.

The scope of the classification

The SNA concepts underlying the production accounts were also used to determine the scope of NAPCS Canada. It is modelled on one of the five elements that define the production boundaries of the SNA.

"The production of all goods or services that are supplied to units other than their producers, or intended to be so supplied, including the production of goods or services used up in the process of producing such goods or services." (paragraph 6.27a)

This concept embodies both market and non-market goods and services.

In addition to the production of goods or services supplied to units other than their producers, the production boundary of the SNA includes the production of selected products for final consumption or capital formation by their producers, as well as the production of housing services by owner occupiers. This type of production is generally known as own account production.

NAPCS Canada does not separately identify products produced on own account. The inclusion of own account production in the production boundary of the SNA is more a matter of accounting than a matter of classification. From a classification point of view, products produced on own account can be classified in the same manner as those produced for others. For example, housing services produced by owner occupiers could be classified with housing services produced for market. The relevant category of NAPCS is class 76411 – Residential rents.

The inclusion or exclusion of assets is an issue often raised in discussions of the scope of product classifications. NAPCS Canada is a product classification, not an asset classification. As is the case in the SNA, assets and products are classified using different classifications in the Canadian system.

That said, most assets begin their life cycle as products resulting from a production process. These products later become assets by virtue of their use as capital in a production process. These are known in the SNA as non-financial produced assets, and they are products in NAPCS Canada. For example, this would include machinery and software.

So called non-financial non-produced assets however are out of scope for NAPCS Canada. The SNA describes these assets as “consisting of three categories: natural resources; contracts, leases and licences; and purchased goodwill and marketing assets”. Though the assets are not in scope for NAPCS because they are not produced, the services of these assets are in scope since they are produced. For example, the services of natural resources assets are covered by NAPCS 6511121 - Licensing of rights to explore for or exploit renewable and non-renewable resources and the services of franchise assets are covered by NAPCS 6511114 - Franchising agreements.

Financial assets are also out of scope for NAPCS Canada.

The universe of NAPCS Canada can be summarized as follows:

Out of scope

• Financial assets
• Non-financial, non-produced assets (except the services related to them)

In scope

• Non-financial, produced assets (as products, not as assets)
• Market goods, intangible goods and services
• Non-market public goods and services
• Recovered and used products (even if not the result of production)

The classification criteria

The classification criteria refer to the attribute(s) of the statistical unit used to create the most detailed categories of the classification and to group them into analytical aggregates. The attribute used to create the most detailed categories of a classification must be observable and verifiable in the context of a statistical operation, or it must be possible to derive the information from a set of observed characteristics.

Some classifications are built by the systematic application of one or more criterion. The National Occupational Classification (NOC) is a good example of such a system. The NOC uses two attributes of jobs: the ten broad occupational categories are based on skill type and the categories within are largely based on skill level. The North American Industry Classification System (NAICS) also defines its most detailed categories on the basis of a single criterion whenever feasible; establishments are grouped into industries according to the similarity in the production processes used to produce goods and services.

Other classifications use mixed criteria to create detailed categories and groupings. NAPCS Canada falls into this category, in part, as a result of the nature of the classification, and in part as a result of characteristics inherited from classification systems embedded into it.

The universe of products is very diverse. NAPCS Canada recognizes close to 3,000 categories of goods and services at its most detailed level (7-digit) to represent that diversity. In theory though, it could include several times that number of detailed categories, but such a classification would not be sustainable in a statistical system. One of the challenges in building such a system is to develop useful criteria to identify a manageable number of detailed categories. Because the universe of products is so diverse, it is not possible to use a single criterion.

The criteria most commonly used to identify products in NAPCS Canada, separately or in combination, are:

• Physical characteristic of the product – For example, live animals are classified by species, ores and concentrates by type of metal or mineral deposit, and crude oil by density.
• Stage of processing – For example, unwrought metal products are distinguished from basic and semi-finished metal products, and pulp and paper products are distinguished from converted paper products.
• Technology or process– For example, cold rolled steel products are distinguished from hot-rolled steel products, transportation services are distinguished by mode, and advertising space and time are classified by mode of delivery.
• Purpose or intended use – For example, architectural and engineering services are classified according to the type of project into which they are embedded, education services by the nature of the program, and public administration services according to purpose served.
• Function – For example, fixed telecommunication are distinguished from mobile telecommunications, management services are classified according to the type of advice provided, and personal and personal care services are classified according to the need being fulfilled.

The number of categories at the most detailed level of NAPCS Canada reflects the accumulated experience with the implementation of product classifications.

The organization of detailed categories into higher level groupings is not based on a unique criterion either, but there is a dominant principle of organization nested in NAPCS Canada, the industry of origin.

This characteristic is inherited from classifications embedded into NAPCS Canada; many of these classifications had been developed to collect product data (outputs) for specific industries and the Supply and Use Classification of the Input-Output framework that underlies the class level of NAPCS Canada is very much based on an industry of origin model.

As a result, the presentation and organization of NAPCS Canada is similar to what is traditionally found used in industry classifications. The outputs of primary industries (agriculture, forestry and mining) appear at the beginning of the structure, followed by those of manufacturing industries, transportation services industries, trade industries, and other services producing industries.

The approach used in NAPCS Canada to create detailed categories and group them into analytical groupings is not unique. The United Nations Central Product Classification (CPC) and the Eurostat Standard Classification of Product by Activity (CPA) use a similar approach. The CPC and CPA are comparable to NAPCS Canada in purpose and in scope.

The classification structure and coding system

NAPCS Canada contains a standard classification structure and standard variants of that structure. The standard structure is intended for broad use, whereas each variant is designed to meet a specific user need.

The standard classification structure

The standard classification structure is hierarchical, that is, a structure where categories at the lower levels are aggregated into the next higher level. It comprises four levels; group, class, subclass and detail. The table below provides the number of categories within each of these levels.

Nomenclature and number of categories within each level of NAPCS Canada 2017

Level	Coding	Number of categories
Group	3-digit code	157
Class	5-digit code	506
Subclass	6-digit code	1,409
Detail	7-digit code	2,737

The advantage of hierarchical classifications is that they enable the collection, dissemination and analysis of data at different levels of detail, in a standardized way. For example, a survey program may collect data using the most detailed level of the classification but publish data at a higher level to protect confidentiality.

NAPCS Canada is unique in that each level of the hierarchy has been designed with a particular use in mind. This approach was adopted in recognition that different product statistics programs can support a more or less detailed classification, and to facilitate the integration of different types of product data. The main purpose of each level of the classification is described below, from the most detailed to the most aggregated level.

The detail level (7-digit) of the classification was designed to be the most precise for which business statistics programs would collect and publish data on the outputs of industries. This level of the classification is most commonly used in annual industry surveys to collect data on revenues by type of goods or services produced.

The primary purpose of the subclass level (6-digit) is to support the production of price indices for the products defined in NAPCS. At the time of publishing NAPCS Canada 2017, this use was limited to the production of industrial product price indices (IPPI), raw material price indices (RMPI) and international merchandise trade price indices (IMTPI). The IPPI program measures price changes at the factory gate for products sold by Canadian manufacturers and the RMPI program measures price changes for raw materials purchased by Canadian industries for further processing. The universe of industrial products and raw materials represents about half of the subclasses of NAPCS. The IMTPI measures price changes for imported and exported goods.

The class level (5-digit) is the target level to produce coherent statistics in current and constant dollars on the supply and use of products, including the production, imports, exports and consumption of products. It is at the core level of the classification. As mentioned earlier in the introduction, the input-output accounts provide the framework to integrate product statistics into a coherent system that describes the supply and use of goods and services in our economy. For that reason, the Supply and Use Product Classification (SUPC) used for the production of input-output tables is largely embedded at the class level of NAPCS.

Finally, the group level (3-digit) provides higher level aggregates primarily for presentation and analytical purposes. This level is also the basis to define alternative aggregation structures, known as regrouping variants (discussed in the next section).

NAPCS Canada 2017 uses a traditional hierarchical coding system where the code of a child adds a digit to the code of the parent. The classification of crude oil and bitumen (group 141) shown below illustrates how the coding system works.

The classification of crude oil and bitumen (group 141)

Code	Title
141	Crude oil and bitumen
14111	Conventional crude oil
141111	Conventional crude oil
1411111	Light and medium crude oil
1411112	Heavy crude oil
14112	Crude and diluted bitumen
141121	Crude and diluted bitumen
1411211	Crude bitumen
1411212	Diluted bitumen
14113	Synthetic crude oil
141131	Synthetic crude oil
1411311	Synthetic crude oil

The reader will note that a category at a lower level can be identical to a category at the next higher level; in the example above, synthetic crude oil is found at the detail (1411311), subclass (141131) and class (14113) levels. This approach ensures that the classification is comprehensive at every level.

In addition to codes and titles, NAPCS provides definitions to help users understand the intended scope of each category and facilitate implementation. The definitions are constructed based on a set of guidelines developed by the Neuchâtel group of the United Nations Economic Commission for Europe (UNECE);^{Footnote 5} and guidelines from the Generic Statistical Information Model (GSIM) – Statistical Classifications Model. The definition at the most detailed level includes:

• A general description of the category
• A list of illustrative examples
• Where necessary, a list of borderline cases that belong to the category (inclusions)
• Where necessary, a list of borderline cases that do not belong to the category with a reference to the classification codes to which the excluded cases belong (exclusions).

Classification variants

Standard classifications are essential components in a coherent statistical system. That said, it is impossible for a single classification to serve all analytical needs. This limitation of standard classifications is well recognized in the field and has been addressed by the development of standard classification variants, of which there are two types:

• Extension variants add one or several levels below the most detailed level of the standard classification by splitting categories of the standard classification.
• Regrouping variants add one or several levels above a level of the standard classification by regrouping categories of the standard classification.

Typically, classification variants are subject-specific and are narrower in scope than the complete standard classification. They are not meant to replace the standard, but rather to complement it by adding new categories where needed (extension variants), or enhancing the analytical usefulness of the classification by changing its organization (regrouping variants). Ideally, statistical programs that adopt a classification variant can also present data on the basis of the standard classification.

There are now six variants of NAPCS Canada 2017:

• Capital expenditures on non-residential construction (regrouping variant) - This variant defines two new high level aggregates (sections and divisions). The divisions result from the regrouping of standard classes (5-digit) of NAPCS 2012 covering non-residential buildings, infrastructures, selected mining and oil and gas support services, and selected remediation services. The codes for the added sections and divisions are alphanumeric. The variant replaced a program specific classification in use since 1965.
• Agriculture goods (extension variant) - This variant adds two additional levels (8 and 9 digits) below the detailed categories of NAPCS covering farm and unprocessed fish products.
• Industrial product price index (regrouping variant) - This variant defines a new aggregate level (sections) by regrouping 3-digit categories of NAPCS 2012 covering products sold by Canadian manufacturers. The coding for the new sections is alphanumeric. The variant replaced a program specific classification (Principal Commodity Groups or PCG) in use since the early 1980s.
• Raw Materials Price Index (regrouping variant) - This variant defines a new aggregate (sections) by regrouping 3-digit categories of NAPCS 2012 covering raw materials processed by Canadian manufacturers. The coding for the new sections is alphanumeric. The variant replaced a program specific classification (Principal Commodity Groups or PCG) in use since the early 1980s.
• Merchandise import and export accounts (regrouping variant) - This variant defines two new high level aggregates (sections and divisions). The divisions result from the regrouping of standard groups (3-digit) of NAPCS 2012 covering imported and exported merchandise. The codes for the added sections and divisions are alphanumeric. The variant replaced the summary import groups (SIG) and summary export groups (SEG) that had been in use for several decades.
• Manufacturing and logging (extension variant) - This variant adds an extra 8-digit variant code (level 5), under the 7-digit standard classification's detail code (level 4).

Relationship to other classifications

North American Industry Classification System (NAICS) Canada

NAICS Canada is the official standard to group Canadian establishments into industries. It supports the collection, processing and dissemination of industry-based statistics.

The introduction of the NAICS manual describes the classification as "...an industry classification system developed by the statistical agencies of Canada, Mexico and the United States. Created against the background of the North American Free Trade Agreement, it is designed to provide common definitions of the industrial structure of the three countries and a common statistical framework to facilitate the analysis of the three economies."

NAICS Canada and NAPCS Canada are fundamentally different in that they have their own purpose, underlying concepts and scope. The table below outlines the basic characteristics of the two systems to illustrate those differences.

Basic characteristics of NAICS Canada and NAPCS Canada

Characteristics of the classification	NAICS Canada	NAPCS Canada
Statistical unit (or object classified)	Establishments (producers)	Products
Classification criteria	Similarity of production process(es)	Similarity in a physical characteristic, stage of processing, technology, purpose or function
Scope or universe	Establishments operating within Canada	Products produced in Canada and consumed in Canada or elsewhere, or produced elsewhere and consumed in Canada
Statistical domain	Industry statistics (e.g., industrial outputs, and intermediate, labour and capital consumption)	Product statistics (e.g., industry outputs and inputs by type, exports and imports by type, price indices)
Examples of analysis supported by related data	Changes in industrial structure or productivity over time	Total supply and demand, market share analysis, shifts in consumption behaviour

Though different, these systems are complementary in that they provide the underlying classification frameworks for the country's production accounts.

Though it is the case that many products are entirely or mostly produced by one industry, it is not always so. For example, meals are mostly produced and consumed in food services establishments, but are also produced in hotels, cinemas, supermarkets and schools. It is also worth noting that product classifications are used outside the scope of industry statistics; for some uses, knowing the industrial origin of products is neither useful, nor desirable.

Supply and Use Product Classification (SUPC)

In the first section of the introduction, the purpose of NAPCS Canada is described as supporting the integration of product data obtained from multiple sources by organizing the documentation, collection, processing, presentation and analysis of data.

In practice, the source for integrated product data is the Input-Output Account (IO) of the Canadian System of Macroeconomic Accounts (CSMEA). The IO provides the set of definitions, classifications and accounting rules necessary to produce balanced industry and commodity accounts which highlight the sources of supply and demand in the economy. The Supply and Use Product Classification (SUPC) is the basis for the commodity accounts. Making the SUPC an integral part of NAPCS Canada was a necessary condition to achieve the stated objective.

The SUPC has been integrated into NAPCS Canada at the class level (5-digit) where conceptually and practically feasible, and the SUPC has been adapted to conform to NAPCS Canada where possible. There are, however, a small number of conceptual and practical constraints that prevent embedding the SUPC into NAPCS entirely. The most significant differences are conceptual in nature.

The first conceptual difference concerns the scope of the two classifications.

The SUPC includes a number of imputed, non-market, fictive and primary input (e.g., labour and capital) commodities that are out-of-scope for NAPCS. The inclusion of these commodities in the SUPC is required to implement SNA accounting concepts.

NAPCS includes a number of intellectual property products that are out of scope for the SUPC, namely intellectual property produced on own account for sale with all-attendant rights. The inclusion of these commodities in NAPCS results from trilateral agreement. The rationale for inclusion is that these types of intellectual properties are produced by one unit and supplied to another. The data to fully implement this part of the classification are not available.

The second conceptual difference concerns the use of different classification criteria.

In finance and education services in particular, NAPCS Canada uses criteria agreed to in trilateral development work. These criteria are thought to be more representative of the specific services than reflecting the institutional organization delivering the services, as found in the existing classification. To date, the data to fully implement NAPCS are not available. As well some of the NAPCS detail in finance services is not consistent with CSMEA concepts.

Though these differences are real, the goal of integrating the SUPC into NAPCS has largely been achieved. More than 80% of the remaining categories of NAPCS classes coincide with those of SUPC, and the remaining 20% or so are splits or groupings of SUPC categories.

Central Product Classification (CPC)

The CPC is an international standard to classify all goods and services. It is published by the Statistics Division of the Department of Economic and Social Affairs of the United Nations.

The CPC is very similar to NAPCS Canada in concept, purpose and scope. The introduction of the CPC manual describes it as follows:

"The Central Product Classification (CPC) constitutes a complete product classification covering all goods and services. It serves as an international standard for assembling and tabulating all kinds of data requiring product detail, including statistics on industrial production, domestic and foreign commodity trade, international trade in services, balance of payments, consumption and price statistics and other data used within the national accounts. It provides a framework for international comparison and promotes harmonization of various types of statistics related to goods and services."^{Footnote 6}

The similarity extends to the general organization of the classification. Like NAPCS Canada, the CPC is a four level hierarchical classification that contains a similar number of detailed categories.

NAPCS Canada is not fully compatible with the CPC. In order to do so, it would have been necessary for the most detailed categories of NAPCS Canada to coincide with, be grouped or be subset of, the most detailed categories of the CPC. This would have required changes to the national classification not possible at this time.

It is the intent to develop and publish a concordance between the two systems that will assist users in understanding the similarities and differences between them.

Harmonized Commodity Description and Coding System (HS)

The HS is the international standard used by more than 200 countries, including Canada, to classify and measure international merchandise trade. Like NAPCS Canada, the HS is a product-based system, but its universe is limited to transportable goods. These HS goods are mostly covered by groups 111 to 482 of NAPCS Canada.

The HS is a more detailed classification than NAPCS Canada; at its most detailed level (sub-heading or 6-digit), it recognizes more than 5,000 goods. The HS generally classifies detailed goods on the basis of the following attributes: component material, degree or stage of processing, and use or function.

The structure of the HS is different than the structure of NAPCS Canada and no attempt was made to integrate it into NAPCS Canada. Doing so would have made it more difficult to achieve the main objective of NAPCS Canada, that of integrating several Canadian classifications into a single standard classification. That said, the class level (5-digit) of NAPCS Canada is by design a bridge level between various classifications, and the most detailed categories of the HS map well to the class level of NAPCS Canada.^{Footnote 7}

Extended Balance of Payments Services Classification (EBOPS) 2010

The EBOPS is the international standard to classify and measure international trade in services. It can be seen as a complement to HS for the classification and measurement of international trade in goods and services.^{Footnote 8} A version of EBOPS is used in the Canadian Balance of Payments Accounts.

At its highest level, the 2010 version of EBOPS recognizes twelve broad service categories, which are further sub-divided into a small number of detailed categories.

For the most part, the EBOPS is a product based classification system. For those categories that are product based, it is theoretically possible to create a concordance between EBOPS and more detailed product classifications like NAPCS Canada. Such a concordance has already been developed between the Central Product Classification (CPC) of the United Nations and EBOPS,^{Footnote 9} and the CPC is similar to NAPCS Canada in concept, purpose and scope.

However, there are three broad categories that are defined on the basis of the entity engaged in the trade activity or the mode of consumption, rather than on the basis of the type of service traded. These categories are travel, construction and government goods and services. The use of different classification criteria in these cases makes it difficult to develop a concordance between the classifications systems. This is especially true for the travel category, which includes traveller expenses on a range of goods and services. There is a need for convergence of product classifications of industry and trade for comparative purposes.

Trilateral North American Product Classification System (NAPCS)

The trilateral NAPCS 2017 is a six-level hierarchical structure consisting of 24 sections, 61 subsections, 172 divisions, 276 groups, 497 subgroups, and 1,167 trilateral products. At this time, the trilateral NAPCS structure is a reference classification system for beta testing. The development of the trilateral version of NAPCS^{Footnote 10} has had a significant influence on the development of NAPCS Canada.

As part of their common classification development work, the three national statistical agencies reconsidered the traditional approaches in the organization of industry and product classifications. In particular, there were numerous discussions regarding the relevant classification criteria for each type of classification.^{Footnote 11} In the case of the product classification, the conclusion was that the analytical needs of users of product data would be better served if the aggregation structure emphasized demand-based attributes of products rather than supply-based attributes of products. This is a departure from the existing practice of emphasizing the industry of origin of products, clearly a supply-based attribute. The new structure emphasizes attributes such as the substitutability of products, the complementary nature of products, or the similarity in markets being served by the products.

NAPCS Canada does not embed the new approach. Doing so would have been too disruptive and would have complicated the task of moving from a multi-to-single product classification system. Instead, NAPCS Canada uses a more traditional aggregation structure, more or less based on the industry of origin of products.

That said, the approach adopted for the trilateral NAPCS is a useful exercise. The most detailed categories of NAPCS Canada have been defined so as to permit mapping into the most detailed trilateral categories. This means that the Canadian detailed categories can be re-organized using the trilateral aggregation structure. In effect, the trilateral aggregation structure becomes a regrouping variant of the Canadian aggregation structure, with few exceptions. The availability of data at the most detailed level of NAPCS Canada will influence the extent of the trilateral work.

Summary of changes from NAPCS Canada 2012 to NAPCS Canada 2017

Based on a 5-year revision cycle, NAPCS Canada 2017 is a new classification series after NAPCS Canada 2012 (Version 1.2). Some categories were split, and others were merged. New categories were incorporated, and some were deleted or transferred, for a total net increase of 123 product categories at different levels, providing better relevancy to statistical programs and users. Most of the changes resulted in the change of scope of the category affected (real changes) and they include: combinations (mergers or take-overs), decompositions (breakdowns or splits), and transfers. Other changes were editorial (virtual changes), as they relate to editing of category titles, definitions and codes, without affecting the scope. The detailed list of changes can be obtained from Standards Division at statcan.standardsproduct-normesproduit.statcan@statcan.gc.ca. A concordance table between NAPCS Canada 2012 and NAPCS Canada 2017 will also be published.

Changes in NAPCS Canada 2017 relative to NAPCS Canada 2012 Version 1.2

Net count of categories by level

Level	NAPCS Canada 2017	NAPCS Canada 2012 Version 1.2	Added	Eliminated	Net difference
3-digit level (Group)	157	156	3	2	+1
5-digit level (Class)	506	506	12	12	0
6-digit level (Subclass)	1,409	1,389	51	31	+20
7-digit level (Detail)	2,737	2,635	153	51	+102
Total	4,809	4,686	217	96	+123

Product data in the Canadian statistical system

Product statistics can be described as those that inform on different aspects of the supply and use of goods and services in the economy, such as the value and quantity of goods and services produced by industries, the value and quantity of goods and services consumed by industries and households, the value and quantity of imports and exports of goods and services, and price changes at different points in the distribution chain. These statistics are collected by various surveys or obtained from administrative sources, and integrated into the country's economic accounts.

Product statistics are ubiquitous in the statistical system. The few examples below are by no means exhaustive; the purpose is to provide the reader a sense of the variety of statistical programs collecting and producing such statistics, and identify those that are using NAPCS Canada or are planning to do so at the time of publishing NAPCS Canada 2017.

Statistics Canada's business and trade statistics programs are the main sources of product statistics. They administer hundreds of monthly, quarterly, annual and ad hoc programs, most of which collect product data.

Among these many programs are so-called industry surveys; they are designed to measure the contribution of specific industries to the Canadian economy. Industry surveys typically collect data on the revenues, expenses and employment characteristics of establishments, and the revenue and expense components of these surveys generally gather data by type of goods or services produced and used. The majority of industry surveys are part of the agency's Integrated Business Statistics Program (IBSP) and use, or plan to use, NAPCS to classify revenues by type of goods or services produced. The following are a few examples of surveys that fall in this broad category:

• The Annual Survey of Manufacturing and Logging Industries (2103)
• The Annual Wholesale Trade Survey (2445)
• Annual Survey of Service Industries: Software Development and Computer Services (2410)
• Annual Survey of Service Industries: Amusement and Recreation (2425)

The business and trade statistics programs also comprise several programs designed specifically to gather product data. Two of these programs are large in scope and based on NAPCS, albeit in very different ways:

• The Retail Commodity Survey (2008)
• Canadian International Merchandise Trade (2201 and 2202)

The retail commodity survey collects data on retail sales for more than 100 commodities from Canadian retailers. The classification of commodities used for collection is compatible with NAPCS.

The Canadian international merchandise trade program is based primarily on administrative data obtained from the Canadian Border Services Agency (CBSA) and the U.S. Customs Border Protection (via the U.S. Census Bureau). By virtue of an international agreement, the coding of Canadian exports and imports must be based on the Harmonized Commodity Description and Coding System (HS) maintained by the World Customs Organization (WCO). The relationship between the HS and NAPCS is defined by concordances; a more detailed description of this approach is provided in the section titled "The methods for implementing the classification".

In addition to the retail commodity survey and the Canadian international merchandise trade statistics program, business and trade statistics programs include several surveys that collect data (value and/or quantity) on the production, disposition, consumption or sales of specific goods. Though these surveys were not designed on the basis of NAPCS, they often are NAPCS compatible, at least in part. The following are a few examples of surveys that fall in this category:

• Fruits and Vegetables Survey (3407)
• Maple Products (3414)
• Monthly Refined Petroleum Products (2150)
• Crude Oil and Natural Gas (2198)
• Annual Industrial Consumption of Energy Survey (5047)

Price indices are core product statistics. Statistics Canada tracks the movements of prices in a variety of markets with more than thirty statistical programs. Three major price programs have adopted NAPCS to produce their indices:

• Industrial Product Price Index Program (IPPI) (2318). The IPPI measures price changes at the factory gate for commodities sold by manufacturers in Canada.
• Raw Material Price Index Program (RMPI) (2306). The RMPI measures price changes for raw materials purchased by industries in Canada for further processing.
• International Merchandise Trade Price Index (IMTPI) (2203). The IMTPI measures price changes for imported and exported goods.

The remaining price programs, including the Consumer Price Index (CPI), use program specific classifications not based on NAPCS, though some are NAPCS compatible.

The business and trade statistics programs also comprise a number of programs that do not, strictly speaking, measure an aspect of the supply and use of goods and services, but that do have a product dimension. The two best known ones are:

• Annual Survey of Research and Development in Canadian Industry (RDCI) (4201)
• Annual Capital and Repair Expenditures Survey (CAPEX) (2803)

Though NAPCS Canada was not primarily designed for this type of application, it is possible in some instances to develop NAPCS compatible classifications for use in such programs. For instance, the component of the RDCI that measures research and development expenditures by field of science is compatible with NAPCS, as is the component of the CAPEX that measures capital expenditures on non-residential construction.

While product statistics are common in business and trade statistics programs, they are rare in household statistics programs. The one major exception is the Survey of Household Spending (3508). It collects household expenditures on the basis of a detailed survey specific classification that was designed prior to the introduction of NAPCS.

The various product statistics described above are of interest by themselves, but their full analytical value resides in their integration into a coherent accounting framework. That is the role of the Canadian System of National Accounts (CSNA).

The CSNA has several accounts designed to "measure activities associated with the production of goods and services, the sales of goods and services in final markets, the supporting financial transactions, and the resulting wealth positions".^{Footnote 12} Among these, the input-output accounts provide the framework to integrate product statistics from a variety of sources into a system that describes the supply and use of goods, services and production factors in our economy. The supply and use product classification (SUPC)^{Footnote 13} underlies that framework; several versions have been developed over time. In order to achieve the objective of improving the coherence of product-level data in the Canadian statistical system, the SUPC has been incorporated into NAPCS Canada where feasible, starting with the 2009 version of the SUPC. It is the intent to maintain this integration in the future.

There are a number of conceptual and practical constraints that prevent embedding the SUPC into NAPCS entirely. That said, the integration of the classifications used for the collection and assembly of product data and the classification used to create input-output tables in a single classification is a key feature of NAPCS Canada, the feature that will contribute most to improving coherence.

The methods for implementing the classification

There are three main methods for implementing NAPCS Canada: by integrating the relevant components of the classification in survey instruments, by coding on the basis of descriptions supplied by respondents, and by forced concordances.

The integration of the classification in survey instruments is a method commonly used in industry surveys to collect revenue by type of good or service produced (sold) as well as in surveys dedicated to the collection of commodity data. Since these two types of surveys target homogeneous populations, it is possible to identify the relevant components of NAPCS at the time of their design. For example, a survey targeting accounting firms would embed the products of NAPCS classes 77121 - Accounting, tax preparation, bookkeeping and payroll services and 77511 - Management consulting services since the products of these classes represent the most likely sources of revenues of these firms. For the same reason, a survey targeting retailers would embed the products of NAPCS group 561 - Retail services (except commissions) and selected products from groups 571 - Repair and maintenance services (except for buildings and motor vehicles) and 581 - Rental and leasing (except rental of real estate).

This is the most commonly used method for implementing NAPCS. The main challenge with this method is to develop reporting guides that support a consistent interpretation of the classification by respondents.

Coding on the basis of descriptions supplied by respondents is very common for industry and occupation statistics, but not so for product statistics. This is very much a function of the source of information for coding.

Industry coding is largely based on business activity descriptions from administrative records and occupation coding is largely based on descriptions supplied by individuals participating in broad scope surveys such as the Labour Force Survey or the Census of Population. In both cases, the coding exercise requires choosing among all the available codes of the relevant classification. In the product statistics domain, coding is mostly done by respondents to industry and commodity surveys; coding based on descriptions supplied by respondents is generally restricted to unallocated revenues in those surveys.

Automated and computer assisted coding are commonly used for this type of coding exercise. It consists of comparing respondents' supplied descriptions to a bank of coded descriptions. The main challenge with this method is to obtain sufficiently rich information from respondents to assign codes at the most detailed level of the classification.

As discussed earlier, the implementation of NAPCS Canada in the international merchandise trade statistics program is achieved by forced concordances. Each of the most detailed categories the Canadian Export Classification (CEC) and the Customs Tariff (CT) is linked to one of the five-digit classes of the standard classification. The CEC and CT are derived from the Harmonized Commodity Description and Coding System (HS), the mandatory classification for the coding and declaration of goods imported into Canada or exported from Canada.

The forced concordance approach is only viable if the systems being linked are based on similar concepts and are sufficiently comparable. The HS and NAPCS have been developed independently to serve different purposes and not always compatible. That said, because the concordances are done at the relatively aggregated class level of NAPCS, most of the inconsistencies between the HS based systems and NAPCS are eliminated.

At the time of publishing NAPCS Canada 2017, this was the only application of the forced concordance method.

Notes

Status

This standard was approved as a recommended standard on May 29, 2012.

2007 version of NAPCS Canada

• NAPCS  Canada 2007 - Classification structure

CSV format

• NAPCS Canada 2007 - Classification structure  (CSV, 16.9KB)

Variants of the NAPCS Canada 2007

• Merchandise import and export accounts