Introduction
The Standard Classification of Transported Goods (SCTG) consists of a blend of transportation characteristics, commodity similarities, and industry-of-origin considerations, designed to create statistically significant categories. It is a structured list that is defined at its less-detailed levels according to the Harmonized Commodity Description and Coding System (HS), and at more-detailed levels, according to patterns of industrial activity. Other factors in the definition of categories were transportation considerations such as volume, revenue, value, origin, and destination.
SCTG is a Canada-U.S. initiative, designed to provide categories for the 1997 U.S. Commodity Flow Survey (CFS) and to improve the integration of Canadian transportation data, particularly for marine, truck, and rail. The classification is also designed to permit comparison of Canadian and U.S. transportation data. In addition, because of its HS basis, SCTG can be used for other international comparisons. SCTG is an important development in the creation of an integrated system of classification that is used for economic analysis, covering production, shipments, and international trade.
Currently, Canadian transportation data are compiled according to three different classifications. The Standard Commodity Classification (SCC), which is the former Statistics Canada commodity standard, continues to be used for rail and truck data, whereas marine data for several years have been based on the Standard Classification of Goods (the SCG, which is Canada's extension of the HS). In addition, rail and truck data are not grouped in the same way, because truck data are based directly on the SCC, whereas rail data are obtained by converting to the SCC the Standard Transportation Commodity Code (STCC) of the Association of American Railroads (AAR). Truck data and most marine data are collected by Statistics Canada according to the detail provided on shipping documents and grouped into a few hundred categories. In contrast, Statistics Canada receives only aggregated rail data.
The U.S. used the STCC for the collection and publication of 1993 CFS data, but it was inadequate, because, among other weaknesses, STCC is primarily a rail-carrier-based classification. The CFS, however, is a shipper survey that collects information about commodities shipped by all modes, as well as intermodal movements. The SCTG has been designed to provide commodity groupings that better reflect goods transported by all modes.
Classification Structure and Data Significance
The structure of the SCTG is hierarchical, consisting of four levels that contain groupings based on HS or SCG "building blocks". These groupings are designed to create statistically significant transportation categories. The SCTG follows the classification principles that each level covers the universe of transported goods, and that each category in each level is mutually exclusive. These levels range from a minimum of 42 categories to a maximum of 512 categories.
| Level of Hierarchy | Number of Categories |
|---|---|
| First (2-digits) | 42 |
| Second (3-digits) | 137 |
| Third (4-digits) | 291 |
| Fourth (5-digits) | 512 |
The first, or two-digit, level consists of 42 HS-based categories, and is designed to provide analytically-useful commodity groupings for those users who are interested in an overview of groupings of transported goods. Where possible, these categories consist of industry-of-origin groupings within HS definitions.
The second, or three-digit, level consists of 137 HS-based categories. At this level, the categories consist of goods for which very significant product movements are expected to be recorded in both Canada and the U.S., thus providing the best basis for Canada-U.S. comparisons. There are two exceptions to the use of HS categories at the three-digit, or international, level. The first exception relates to non-food waste products (SCTG 41). In most cases, the HS identifies such waste, but where it does not, SCG detail was used to group these products. The other exception is for refined petroleum and coal products (SCTGs 17-19), which are defined according to SCG details in order to compensate for the lack of international agreement about useful sub-headings of HS 27.10.
The third, or four-digit, level consists of 291 HS- or SCG-based categories. This level of the SCTG is designed to create categories that reflect industry patterns and transportation characteristics that are often not provided for in HS categories. Data-wise, these categories are more important than those of the most-detailed level of the SCTG, and often consist of portions of HS categories as defined by SCG detail. It is likely that Canada will only use a selection of four-digit SCTG categories, because of confidentiality, insignificant data, or data-reliability issues.
The fourth, or five-digit, level consists of 512 HS- or SCG-based categories. This level, which often consists of SCG detail, is the collection level for the CFS, with each category designed to capture significant data that reflects industry patterns and transportation characteristics. For multimodal Canadian data, only some of the five-digit categories will be used, because many of these categories will yield only insignificant data. For data relating to single modes, however, more of the five-digit categories will be useful. In Canada, this level is not a collection level, but the first possible level of aggregation of the micro detail obtained from shipping documents for marine and truck transport. For rail data, the aim will be to continue to receive data obtained from the detailed STCC categories, but aggregated according to the SCTG. The richness of detail at this five-digit level is sufficient to accommodate those situations where a Canadian product is relatively more important to Canada than it is to the U.S., particularly where it is difficult to track shipments from one country to the other.
The SCTG is defined by HS or SCG codes only at the three-, four-, and five-digit levels, for presentation purposes. Many of the 42 categories at the two-digit analytical level would be defined by only one or a few codes. Others, however, would require extensive lists of codes, which are not as useful in defining the contents of these less-detailed categories as they are for the more-detailed distinctions.
Although the blend of the criteria used to create the SCTG resulted in four different levels of categories, many of the categories are identical at more than one level. One grouping--"Pharmaceutical Products" (SCTG 21)--is even identical at all four levels. In this case, four different criteria coincide at each level. The most-detailed, or collection, level corresponds to an industry-of-origin grouping, which is identical to an HS category that, in turn, is identified as an analytically useful product description.
For Canadian use only, the SCTG includes an additional category (42), which is subdivided into 4 three-digit categories that are further disaggregated into 16 categories. The first of the three-digit categories relates to mail and parcels. The contents of parcels cannot be identified, nor can the contents of mail other than items such as advertising flyers. It should be noted that, in the CFS, mailed parcels constitute one of the seven modes for which commodity information is collected. The contents of the second category, "Trailers on flat cars (TOFC), containers on flat cars (COFC), and other shipping containers, returned empty", are identified, but constitute transported goods for which there are no associated direct revenues, in that they are a cost of the provision of transportation services. The third category, which consists of unidentified freight or cargo, is associated with revenues, but the generally smaller size of the individual shipments results in their assembly with other small shipments in containers, thus forming a grouping that is impractical to identify by component. The fourth category relates to "Goods on company service". This covers significant movement of goods, which are owned by the transporting company and do not generate revenue. This might include the movement of service equipment to service rail cars in need of repair.
Classification According to Transportation Characteristics
The combination of the HS and SCG provides many thousands of categories to use as building blocks in the development of a standard classification of transported goods. Data significance was used to select appropriate categories for the SCTG, starting with the most aggregative categories and working down to the most detailed categories, according to the criteria of significance established for each of the four levels. The creation of SCTG categories was usually done by selecting from the structure of the HS or SCG (i.e., by using HS or SCG categories), or by grouping HS or SCG categories to form more significant categories. Both the individual categories or their groupings within the structure tend to follow industry lines, because each most-detailed category of the HS generally contains only the outputs of a single industry (although there are many exceptions), and transportation of these products is based on the characteristics of these products. Sometimes, however, SCTG categories were formed from HS or SCG categories that cut across the structure of the HS, in order to group products according to their shipping characteristics. In the SCTG, for example, raw agricultural products, such as tobacco leaves and sugar beets, were grouped because they are moved in bulk by similar kinds of transportation equipment.
Relationship to Industries-of-Origin
The more-detailed levels of the SCTG can be associated with four-digit industry classes for both the Canadian and American SICs, as well as NAICS. This is because, with the exception of residual categories, most four- and five-digit SCTG categories primarily contain the products of only one industry. This feature will facilitate comparison with industry data, as well as with other classifications based on the U.S. SIC, such as the U.S. Numerical List of Manufactured and Mineral Products (Product Codes) and the STCC.
Relationship to Other Commodity Classifications
In Canada, detailed HS or SCG categories are used to define the inputs and outputs of goods for industries as well as for classifying imports and exports. Comparison of these categories and those of the SCTG can thus be achieved by grouping such categories into the relatively few transportation categories.
In the U.S., imports and exports are HS-based, thus they can be compared to the SCTG, by grouping these much-more-detailed commodity categories into the relatively few transportation categories. For industrial commodity data, however, the U.S. uses an SIC-based Numerical List of Manufactured and Mineral Products. This list of products can only be compared to the SCTG by grouping the product codes to their SIC level, and then grouping them to their corresponding four-or five-digit SCTG categories.
Difficulties in Creating Data-Significant Categories
In developing the SCTG it was difficult to make reliable judgements about the significance of possible HS- or SCG-based categories, for several reasons. Firstly, the most-current U.S. CFS data (1993) were collected according to the STCC and only the most-aggregative-level data (48 categories) had been released at the time of the development of the SCTG. Secondly, these data could only be linked to HS categories through the use of a concordance developed by the AAR that links U.S. import categories to STCC, and the quality of this concordance had not been determined. Thirdly, Canada has good SCG-based transportation data, but only for marine. Rail data continue to be reported according to STCC categories, and truck data, according to a sample of commodities that are organized according to Canada's old commodity classification. Several years ago, SCG-based classifications had been developed for each of rail and trucking, but never used. At that time, related linkages were developed. These linkages were utilized to convert 1992 and 1994 data into proxy SCTG data. Finally, taking into account all of the relevant classifications involved in developing the SCTG, there was an enormous amount of detail to contend with. Despite all these challenges, the SCTG should provide much better data than is currently available.
Difficulties in Designing SCTG
Almost always the HS "building blocks" enabled the creation of categories that are useful for the classification of transported goods, but these categories had to recognize the limitations of the HS categories, which sometimes caused problems. For one group of products, for example, a small amount of adjustment had to be made. SCTG 32 covers metal basic shapes and 33 covers articles of metal. Many of the categories of HS Chapter 81 (the less-important base metals) have been assigned to either SCTG 32 or 33 according to SCG details. Of these SCG categories, a few consist of both basic shapes and articles, but they have been assigned to either SCTG 32 or 33, based on relative importance.
"Parts" are difficult to categorize, and it is difficult to distinguish between parts and accessories. The aim in categorizing parts was to use an approach that is as useful as possible from a reporting and identification perspective. In the SCTG, parts are normally associated with the machinery, equipment, or apparatus without which these products could not operate. This association works well for those shipments of new goods, which are often accompanied by additional parts. It also seems to work well for separate shipment of replacement parts, as their identification is usually in relation to the goods of which they form a part. Parts that are associated with a wide range of machines and equipment, however, are not easy to deal with. In such cases, the HS was used as a basis to define these parts as separate pieces of equipment in their own right (e.g., all pumps are together in SCTG, including motor vehicle fuel-oil pumps). The HS was also used to define motor vehicle parts. Despite these definitions, however, it is the respondents that ultimately have to be relied on to appropriately identify parts.
The HS generally does not make a distinction between paper and paperboard. In order to obtain industry-friendly groupings of these products, however, it was necessary to make these distinctions. SCG details were utilized, as was technical information about the products.
Agricultural products was a difficult grouping to compile, because its components are drawn from several different areas of the HS.
Both chemicals and textiles are organized differently in the HS than they are in industrial classifications. For SCTG purposes, these products were re-grouped.
Data Continuity
For users of CFS data, a concordance will be required to assist in linking the 1993 STCC-based survey results to those of the 1997 SCTG-based results. This could be done by directly comparing the two classifications, but would require many difficult-to-make decisions about the considerable number of instances where the two classifications are not compatible. It could also be done indirectly, by utilizing the AAR concordance of STCC to U.S. import tariffs, but only after the quality of this concordance has been determined. Because the AAR concordance uses the HS-extended tariff codes, it has the same basis as the SCTG categories. The immensity of this concordance, however, is a major problem, but it should be noted that its use will also be considered for data continuity purposes by Canada and for use in production of current data.
For Canadian data, the concordance that will link current data for marine transport to future SCTG-based data can utilize the fact that current data are based on the SCG. Truck data would have to be linked by a concordance that relates SCC-based truck categories to the SCTG, but it can be based on the recent work done by Transportation Division to provide proxy SCTG categories for truck data, as well as marine and rail data. This work utilized the concordances between SCC-based data for truck and rail and the SCG-based classifications for these two modes that had been prepared by Standards Division several years ago. With regard to rail data, it might be possible to link it to the SCTG by also utilizing the AAR concordance of STCC to U.S. import tariffs, rather than directly concording SCC-based rail categories to the SCTG. In any case, it will be difficult to link SCC-based rail categories because they have already been converted from STCC-based categories (in other words, a two-stage conversion of data will be required).
Changes to the HS
As of January 1988, forty-five countries first used the HS for international trade data. Canada also introduced the HS into its manufacturing data as of that year. The U.S., however, did not start using the HS for international trade data until January 1989 and is currently working on the future introduction of the HS into its manufacturing data.
As of January 1996, about 500 changes were made to the HS. These changes include new 6-digit categories, revisions to the coverage of some 4-digit categories as well as to 6-digit categories, and a chapter change for one 4-digit group of commodities. All of the 1996 changes have been incorporated into SCTG. In comparing the revised HS to its earlier version, it should be noted that sometimes the same codes continue to be used for categories that have been revised.
Hazardous Materials
The SCTG does not identify specific categories of products as being hazardous, nor does it include a special grouping of its categories under the title of "hazardous". This is because the HS does not include degree of hazardousness as a classification criterion. The STCC does, however, contain an additional listing of categories that are considered to be hazardous materials, and the Organization for Economic Co-operation and Development has produced a list of about 3,000 categories of hazardous products, but it is not based on the HS. There is, however, a class for hazardous chemical waste products. (See waste)
Implementation of, and Possible Future Changes to, SCTG Categories
Once the SCTG has been used by the U.S. for the 1997 CFS, information about the usefulness of the various categories, particularly the most-detailed ones, will become available. This information could be used to make some minor changes, but major changes would have a serious impact on data continuity.
In Canada, the SCTG can be incorporated into marine transport statistics with the least impact on data continuity, because these statistics are collected according to SCG-based classifications. This incorporation will require a re-grouping of data obtained from the "library" of reported descriptions that had been coded several years ago to the most-detailed level possible of the SCG. The process of determination of significant categories will result in a unique version of SCTG for marine data (i.e., the categories of the standard will be adjusted when required to reflect only categories important to marine transport). A similar re-grouping will be required for truck data. Rail data should be changed as soon as possible from the current STCC-converted-to-SCC basis to a STCC-converted-to-SCG basis, utilizing the AAR's concordance of STCC to U.S. import categories. Once data are obtained for these three series, consideration could be given to making minor changes to the SCTG.
In Canada, the SCTG will also be used to produce multimodal statistics. The comparison of marine, truck, and rail data will be based on those SCTG categories that are common to these three modes. For analytical purposes, all of the SCTG two-digit categories should be included in this comparison. For Canada-U.S. purposes, and for purposes of international comparison, ideally data for all of the three-digit categories should also be made available. At the four- and five-digit levels, however, there will only be some SCTG categories for which multimodal data are available.
For U.S. purposes, the five-digit level was designed to be the collection level of the CFS, with the less-detailed levels available for analysis or publication. Because each level of the SCTG covers all transported goods, however, the more-aggregated levels might also be used to assist in data collection--particularly, the three-digit level, which is the universally defined HS level.
It is hoped that this classification will prove useful for a variety of transportation applications and will be adopted by those who find it useful for collection and analysis of transportation data.
Procedures Used in SCTG Category Descriptions
In order to achieve consistency, a number of procedures were established:
- HS terminology is used, unless it was deemed necessary to adjust it to reflect North American (Canada-U.S.) usage. SCTG categories that are defined exclusively by a single HS heading or subheading usually are identical to such headings, so long as they were written from a "stand-alone" perspective. The advantage of using HS terminology is that it is often defined or described in the HS Notes.
- Two types of identification are used for residual categories. "Other..." is used in the title of residual categories whose content is defined by the immediately higher-level category minus specified categories. Categories beginning with "other" can be found at the 3-, 4-, or 5-digit levels and are identified by codes ending in "9" or "90". "N.e.c." is used in the title of residual categories that contain products, some of which are identified elsewhere in SCTG. Categories ending with "...n.e.c." appear only at the 2-digit level.
- "Including... " followed by a list of examples is used extensively in the description of 5-digit categories only. The list of products following "including" are not meant to be exhaustive of the universe of the category, but to include the more important examples as well as the less-obvious ones, with the intention of providing users with an understanding of the range of products covered by the category.
- "Except..." is used to identify those products that could be considered to be part of a particular category but that, for SCTG purposes, are not included in the category. "Except..." is used at all levels of the SCTG, but mostly at the 3-, 4-, or 5-digit levels.
SCTG Development Team
The Team consisted of, from Statistics Canada: Keith Hannett and Andreas Trau of Standards Division, under the direction of Shaila Nijhowne, and Louis Pierre, under the direction of Michel Cloutier and Tricia Trépanier (and, earlier, David Dodds) of Transportation Division; from the U.S. Department of Transportation's Volpe Centre: Nat Bottigheimer, under the direction of Mike Rossetti; and Walter Neece, on contract to the U.S. Bureau of the Census, under the direction of Jim Aanestad.