26.4. Compilers of international merchandise trade statistics are advised to consider the following factors in designing and implementing data and metadata dissemination strategies. Each of these factors will be subsequently discussed:
(a) Variables to be made available;
(b) Timeliness of data and metadata dissemination;
(c) Coherence between disseminated data sets;
(d) Statistical confidentiality;
(e) Revision policy;
(f) Groups of users and their specific needs;
(g) Formats and means of dissemination.
1. Variables to be made available
26.5. Minimum set of data and structural metadata items to be disseminated. IMTS 2010 provides certain recommendations and encouragements in respect of what kinds of data and structural metadata items should be disseminated. They can be summarized as follows: countries should release international merchandise trade statistics for four trade flows (exports, re-exports, imports and reimports), detailed by six-digit HS codes and by categories of goods recommended for separate identification, with respect to each trading partner, using appropriate criteria for partner attribution, in terms of applicable statistical value and quantity units and by mode of transport. Table XXVI.1 lists a suggested set of data and structural metadata items to be disseminated, as well as their possible values.
26.6. Dissemination of new items recommended or encouraged by IMTS 2010. It is good practice to disseminate newly recommended and encouraged data items (e.g., imports valued FOB, in addition to imports valued CIF) in such a way as not to confuse users. Table XXVI.2 presents dissemination requirements for aggregate trade flows by partner country and mode of transport, while Table XXVI.3 presents requirements for detailed trade data in accordance with IMTS 2010 recommendations (R) and encouragements (E).
26.7. Timeliness of first data releases. Timeliness is defined in IMTS 2010 (para. 10.7) as the length of time between an event (the end of the reference period) and the availability of statistical information about this event. IMTS 2010 encourages countries to issue the first releases of data as follows: (a) monthly totals of exports and imports within 45 days after the end of the reference month, at least by major trading partners and basic commodity breakdown; (b) quarterly data within 60 days after the end of the reference quarter; (c) annual data within 90 days after the end of the reference year.
26.8. Advance release calendar. IMTS 2010 (para. 10.5) further recommends (a) that countries announce in advance the precise dates at which those statistics will be released and revised and (b) that this advance release calendar is posted before the beginning of each year on the website of the national agency responsible for the dissemination of the official trade statistics. While implementing those recommendations and encouragements, it is good practice to make clear the dates on which the provisional estimates and the final data (no longer subject to regular revision) will become available. Also, it is good practice to inform users about availability of such calendars using all appropriate means of communication.
26.9. Dealing with the trade-off between timeliness and reliability and accuracy. In producing data, there is usually a trade-off between the timeliness, on the one hand, and the reliability, accuracy and level of detail of the published data, on the other hand. Recognizing this trade-off, IMTS 2010 (paras. 10.5-10.6) encourages countries, while making relevant decisions, to take into consideration a number of factors such as user requirements, timing of the collection of initial data by the customs administrations and other source agencies, extent and timing of data revisions of the major data sources, and modes of data dissemination. It is good practice to discuss this trade-off explicitly with major user groups, in order that an understanding may be reached on the best solution, and to make this understanding publicly available.
26.10. Early dissemination of provisional estimates. To improve timeliness in the dissemination of international merchandise trade statistics, it good practice to publish, on a regular basis, the provisional estimates of total exports and imports, as well as of trade by major commodities and partners, soon after the end of the reference period (see above). Such estimates, by their very nature, would be based on relatively limited data content and replaced at a later date by more accurate, but less timely figures. However, compilers and users must be aware of the trade-off between quality (size of revisions) and timeliness (e.g., it is generally not a good practice to frequently publish large revisions (IMTS 2010, paras. 10.10-10.12)). Quality aspects need to be taken into account when deciding on the frequency of publication.
3. Coherence between disseminated data sets
26.11. Coherence of monthly, quarterly and annual data. Many countries use additional information for the compilation of annual trade statistics. In this connection, IMTS 2010 (para. 10.9) stresses that the data for the fourth quarter (or for the twelfth month) need to be compiled and disseminated in their own right and should not be derived as the difference between the annual totals and the sum for the first three quarters (or 11 months) in order to provide undistorted data for all months and quarters. It is good practice to provide in the reference metadata an appropriate explanation in this respect, so as to assist users in the correct interpretation and use of the data. It is also good practice to make users aware of particularly significant cases of non-additivity over time, and to provide the reasons for their existence.
4. Statistical confidentiality
26.12. Statistical confidentiality versus user needs. Statistical confidentiality refers to the protection of information of individual statistical units and should be distinguished from other forms of confidentiality under which information is not disseminated based on considerations such as national security concerns. It is good practice for compilers of international merchandise trade statistics to always strive for a full coverage of all trade transactions that are within the scope of IMTS while applying appropriate methods to keep certain information confidential. IMTS 2010 (para. 10.2) recognizes the need both for statistical confidentiality and for balancing it against the demand for public information in cases where the application of statistical confidentiality would limit or make it impossible to provide sufficient or meaningful information. It is also good practice to disseminate, along with the data, a quantitative indicator of the quantity of goods subject to confidentiality.
26.13. Development and implementation of confidentiality rules. IMTS 2010 (para. 10.3) recommends that passive confidentiality be applied as much as possible, i.e., that data be treated as confidential only at the trader’s request and when the statistical authority finds the request justified based on the confidentiality rules, unless the use of active confidentiality is already the established, desired and accepted practice. It is further recommended that in suppressing data due to confidentiality, any information deemed confidential (suppressed) be reported in full detail at the next higher level of commodity and/ or partner aggregation that adequately protects confidentiality. However, the implementation of these recommendations on statistical confidentiality also depends to a large extent on each country’s legislation and the general confidentiality policy adopted by its statistical system. An important challenge in the implementation of confidentiality rules is to ensure that confidentiality is applied across all the different classifications in which data are disseminated while preserving the goal of maximizing information. An example of good practices in the development and implementation of confidentiality rules is provided in box XXVI.1.
26.14. Informing about confidentiality rules. It is good practice for all countries to develop and publish an overview of their confidentiality rules with respect to international merchandise trade data in order that data reporters may be assured that their right to confidentiality is guaranteed, while data users may be informed about certain data limitations, thereby enabling them to use the data more appropriately. It is also good practice to provide users with details on what part of the data are affected most by the application of confidentiality rules and on the magnitude of this effect.
5. Revision policy
26.15. Features of a good revision policy. Recognizing that data revisions are an essential part of country practices, IMTS 2010 (paras. 10.10–10.11) encourages countries to develop a revision policy that is well designed, carefully managed, transparent and well coordinated with other areas of statistics and hence allows users to cope with revisions in a systematic manner. The following are some good practices in relation to the revision policy (see IMTS 2010, para. 10.12):
(a) Availability of a detailed description of the revision policy on the responsible agency’s website;
(b) Reasonable stability of timing of the revisions from year to year;
(c) Predetermined timing of revisions (clearly reflected in the data release calendar);
(d) Prior notification to users whenever a revision requires changes in the time series going as far back as the beginning of the series to retain methodological consistency, explanation for the reasons for the revision and provision of information on its possible impact on the data;
(e) Easy access to sufficiently long time-series of revised data;
(f) Dissemination of all revised monthly, quarterly and annual data so as to ensure consistency of all data available to users, including seasonally adjusted data and indices;
(g) Adequate documentation of revisions in the statistical publications and databases;
(h) Coordination of the revision policy with non-customs data providers, which might be the origin of large revisions;
(i) Establishment of a vintage database to measure the size of revisions and generate quality indicators.
26.16. Country experiences in setting up and implementing a good revision policy are described in box XXVI.3 and in box XXVI.4.
Revision policy of the Philippines
The figures for the previous month are revised to include documents that were submitted late and were not included before the cut-off date, i.e., for exports 10 days after the reference month and for imports 25 days after the reference month. The revised monthly figures are reflected in the next month’s press release.
Processing, cleaning, and updating of data files to produce final tabulations are carried out for exports 5 to 7 days after the press release date, and for imports 5 to 10 days after the press release date.
Revisions are also due to:
(a) Consistency between FOB values and volume, both in gross weight in kilos and quantity.
(b) Validation of values to ensure credibility and accuracy of the final figures.
(c) Adjustments due to scope and coverage.
26.17. Diversity of user groups and needs. A key to the usefulness of trade statistics is its broad dissemination. Trade statistics are compiled to serve the needs of many users, including Governments, the business community, the mass media, non-governmental organizations (NGOs), compilers of other economic statistics such as balance of payments and national accounts, various regional, supranational and international organizations, researchers and the public at large. The uses of trade data are numerous and range from the development of national, regional and international trade and general economic policies to market analysis, determination of the economic characteristics of traders, infrastructure planning and provision of input into the system of national accounts and balance-of-payments statistics.
26.18. Equal treatment of users. IMTS 2010 (para. 10.13) recommends that all users be treated equally and that data be disseminated without preference to any national or international user group. To ensure that such treatment is upheld, it is good practice to make all kinds of trade data available to all users at the same predetermined time.
26.19. Monitoring of data needs of users. It is good practice to systematically monitor changing user needs in order to ensure the relevance of the compiled data and their adequate dissemination. Such monitoring, as well as subsequent actions taken, should be part of inter-agency cooperation efforts within the established institutional arrangements. It is good practice to establish close and long-term relationships with representatives of major user groups in order to identify user needs and the most effective means of disseminating data and metadata. This might be achieved through standing advisory committees as well as ad hoc promotional events. National and world statistics days can be celebrated in this connection.
26.20. Surveys of user satisfaction regarding data dissemination. In order to ensure the most effective dissemination, it is good practice to conduct user satisfaction surveys. Such surveys might identify user groups that might need to be given more attention, owing to a lack of certain technical means of accessing data, or more detailed explanations with respect to using the data properly. A user satisfaction survey focus on the following aspects of data dissemination:
(a) User-friendliness of the trade statistics database interface;
(b) Clarity and completeness of available metadata;
(c) Desirability of continued issuance of traditional paper publications;
(d) Means of improving data and metadata presentation.
26.21. Outreach activities. It is good practice to conduct regular outreach activities aimed at helping users to better understand data and put them to the most effective use, including efforts to improve the statistical literacy of users and to prevent misinterpretation, for example, within the context of a broad public relations strategy designed to deepen the general public’s understanding of the importance of statistics. The following outreach activities can be encouraged: conduct of seminars focused on specific user groups; provision of tutorials and user guides explaining how to find data on the dissemination website; organization of press conferences and including contact information in press releases to assist users in the correct interpretation of the statistics; enabling user groups to participate in annual conferences, book fairs and other suitable events; and launching of awareness campaigns, e.g., a “national statistics day/week/month”.
26.22. User support to ensure correct interpretation of data. While statistics can be used acceptably and interpreted in many different ways, it is important to maintain trust in, and the credibility of, official trade statistics. Hence, it is good practice for the responsible statistical agency to prevent obviously erroneous interpretation of the data, and to undertake the necessary corrective actions if such faulty interpretations are detected (for instance, by conducting press conferences, providing press releases and writing letters to the editors of publications where misinterpretations have been detected). One good practice for preventing misinterpretation of data is to give special attention to establishing direct contacts with other government agencies, international organizations and universities, as these are users of foreign trade statistics whose analyses have a major impact on public policy and public opinion.
7. Formats and means of dissemination
26.23. Use of different means of dissemination. Both data and metadata can be disseminated in various formats and by various means. IMTS 2010 (para. 10.13) recommends that countries choose the dissemination format and means of dissemination that best suits their users’ needs. In view of the diversity of user groups it is good practice to adopt several formats and means of dissemination to ensure that data and metadata are effectively delivered. For example, press releases of international merchandise trade statistics aimed at the general public have to be disseminated in ways that facilitate re-dissemination by mass media, while more comprehensive or detailed statistics intended for researchers need to be disseminated through online databases, with hard-copy publications used as reference material.
26.24. Redesigning paper publications. It is good practice to periodically redesign paper publications in order to make use of innovative means of data and metadata presentation and better reflect user demands. In this connection, countries are advised that it is no longer necessary to issue paper publications in an old fashioned way containing a set of tables or providing very detailed data on trade in respect of particular commodity groups and partners. A better practice is to focus such publications on the main features of a country’s external trade, and present data in a more user-friendly way by resorting to enhanced visual elements, such as color charts and by adding more analytical information.
26.25. Central role of electronic databases. IMTS 2010 (para. 10.13) recommends that the official country trade statistics be made available to users through the electronic databases maintained by the responsible agency (see box XXVI.5 for the experience of Brazil). It is good practice to ensure that such databases:
(a) Allow free and equal access to all users to any data record considered part of official trade statistics;
(b) Contain an extensive metadata and knowledge base;
(c) Allow users to query the entire database through a user-friendly interface, and to download query results in the commonly used electronic data formats (such as comma-delimited text files), thereby reducing the need for personalized handling of most data requests and greatly enhancing the efficiency of data dissemination;
26.26. Use of social media. It is also good practice to use social media as an additional tool with which to reach trade statistics users, in particular journalists.
26.27. Special data requests. Regular data dissemination should satisfy most, if not all, user needs. However, some users might have special needs which would require highly complex data extraction, which they themselves might not be able to perform. It is a good practice to offer such users premium data extraction services on a fee basis. Countries should ensure that users are made fully aware of all available options for obtaining the required data.
8. Dissemination of metadata and information on data quality
26.28. Metadata requirements. The provision of adequate metadata and quality reports on international merchandise trade statistics are as important as the provision of the statistical data itself. IMTS 2010 (para. 9.22-9.25) takes into account a broad spectrum of metadata requirements and recognizes that different levels of detail of metadata can be considered by countries for dissemination. In this context, IMTS 2010 (para. 9.22) recommends as a minimum the segmentation of metadata into two levels (reference and structural metadata) and developing appropriate means of their dissemination. It should be recalled that reference metadata are intended for the detailed description of the scope, coverage and quality of data and can be presented separately from data, while structural metadata items are an integral part of the statistics database and should be extractable together with any given data item (see chap. XXV).
26.29. Providing reference and structural metadata. It is good practice to make reference metadata available to users in a separate document posted on the website of the responsible agency and linked to the data query window, so that users can be immediately informed about the existence and importance of such metadata. Reference metadata can be made available in a separate paper publication as well and used in various outreach activities. It is important that reference metadata be compiled following the recommendations contained in IMTS 2010 (para. 9.23). With respect to structural metadata, it is good practice for data query options to include all relevant metadata variables and for those variables to be extracted by default unless explicitly “unclicked” by the user (e.g., quantity data should be extracted together with the respective quantity unit).
26.30. Raising awareness. Users often do not notice or use the available metadata and additional efforts are required to raise their awareness. It is good practice to include explanations of the importance of metadata for the correct interpretation of data and of the effective use of the data in all relevant outreach activities. Even if metadata goes unused, the very fact that they are compiled and made available is reassuring to those who wish to see high standards of credibility upheld.
26.31. Dissemination of trade data to international, supranational and regional organizations. IMTS 2010 (para. 10.15) encourages countries to cooperate with the international, supranational and regional organizations in identifying and applying the most efficient means of dissemination of their trade statistics and related metadata. It is good practice in this regard to review the Statistical Data and Metadata Exchange (SDMX) format for possible use in the exchange and sharing of their data.
 The SDMX technical standards and content-oriented guidelines can provide common formats and nomenclatures for exchange and sharing of statistical data and metadata using modern technology. The dissemination of national data and metadata using web technology and SDMX standards is encouraged as a means to reduce the international reporting burden and to increase the efficiency of the international data exchange. For additional information on SDMX, see http://www.sdmx.org/.
 See the recommendations contained in decision 39/112, adopted by the United Nations Statistical Commission at its thirty-ninth session (Official Records of the Economic and Social Council, 2008, Supplement No. 4 (E/2008/34),chap. I.B.