3.5. The main types of non-customs data sources and their most typical uses are described below. However, it should be noted that the importance and the use of particular data sources may vary from country to country depending on national circumstances and needs.
1. Parcel post and letter post records
3.6. Universal Postal Union. The treatment of parcel post and letter post shipments by customs offices is governed by the acts of the Universal Postal Union (UPU), which is currently composed of 192 member States. The acts, which consist of the Constitution of the Universal Postal Union, the General Regulations of the Postal Union, and the Universal Postal Convention (UPC), are binding for all member States.
3.7. Forms CN 22 and CN 23. Among other matters, UPU deals with the issue of items (letter post, parcels) that are subject to customs control. It provides, for example, that items weighing less than two kilograms and with the value of their contents less than 300 special drawing rights (SDRs), should bear a special form (CN 22). All other items should be accompanied by form CN 23. Form CN 22 contains a description of content by separate articles, their net weight and value. Form CN 23, usually referred to as a customs declaration, requires additional information which should be provided by the sender and include such statistically important indicators as country of origin of goods, tariff number and customs value. The items and the respective forms are to be presented to customs, which then makes its decision regarding clearance based on the information provided on those forms.
3.8. Application of a threshold. If the values declared on the CN 22, CN 23 or other postal forms exceed the threshold value adopted for trade statistics purposes, compilers should then include those goods in trade statistics in full detail (commodity classification, value, quantity, partner country, etc.). If the value of the individual transaction does not exceed the threshold, then the treatment of the transaction should be consistent with the policy for compiling statistics from other low-valued customs records. IMTS 2010 (para. 1.3) encourages countries to estimate and include such flows if they are economically significant as determined by the statistical authorities of the compiling country. The agency responsible for compilation of trade statistics should contact postal authorities in order to ensure that the necessary information is collected and passed on to that agency on a regular basis. Most of this information is now recorded electronically and exchanged between postal operators in the form of EDI (electronic data interchange) messages (see box III.1 below).
3.9. Private parcel delivery services. The compilation of data with regard to items delivered by private parcel delivery services (e.g., couriers, express carriers, etc.) should follow a pattern similar to that for governmental postal services, utilizing all information available. The agency responsible for compiling the statistics should make special arrangements, either through customs or with the parcel carriers directly, to ensure that the necessary information is passed to them.
3.10. Increasing importance of parcel and letter post. With the rapid expansion of electronic commerce, the international movement of goods by post and parcel delivery services (both government and privately operated) is becoming more and more important; the compiling agency should devise a strategy with the aim of developing a compilation procedure that ensures that those merchandise flows are adequately reflected in trade statistics.
2. Aircraft and ships registers
3.11. Use of aircraft and ships registers. When aircraft and ships cross the borders of countries as items of trade and the appropriate customs records are created, those records should be used as the main source of information. However, in some countries, international trade in aircraft and ships may not be recorded by customs even if they cross borders; also, customs records may be incomplete or non-existent if those items do not cross customs borders. Under such circumstances, countries may use national (or international) ship and aircraft registers for evidence of a trade transaction, using change of ownership as it may be indicated in the register as the basis for the compilation of trade statistics. In addition to the use of registers, documented financial leasing agreements may indicate whether a change of economic ownership has occurred. Chapter XXIII discusses in detail the compilation of data on the trade in ships and aircrafts.
3. Enterprise surveys
3.12. Use of enterprise surveys. Obtaining information on international transactions in electricity, petroleum, gas and water through pipelines or on sales and purchases made by aircraft and ships in foreign ports or on the high seas, can be achieved by contacting the relevant enterprises. If such transactions are highly concentrated in just a few enterprises, the compiling agency can conduct a regular census (monthly or quarterly) of all involved enterprises to complete the international merchandise trade statistics. If for certain kinds of transactions, the number of involved enterprises is too large to make the holding a regular census feasible, then the compiling agency could conduct regular enterprise surveys of the specific sectors (i.e., airlines and/or shipping companies).
3.13. Surveys to capture border trade. Border trade is generally understood as trade between residents of adjacent areas of bordering countries. Such trade is typified by low quantities and low value but high frequencies. Given the relatively low value of the individual transactions, this kind of trade is frequently below the customs threshold and not recorded by the customs administration. However, for certain countries, such cross-border trade is economically significant and it is good practice for trade statisticians to capture the value, quantity and commodity detail of these transactions on a quarterly or annual basis. Chapter IV provides more detailed information about enterprise surveys and provides the example of the Informal Cross Border Trade Survey conducted by Uganda.
3.14. Integrated approach to data collection. In the case of the use of non-customs data sources, such as enterprise surveys, it is recommended that countries follow an integrated approach to data collection and make use of business registers and enterprise identification numbers in order to obtain the required information with minimal costs and burden on enterprises. The integrated approach to data collection is of particular importance for the fulfilment of additional information requirements, such as for goods for processing (including obtaining information on change of ownership), intra-firm trade, etc., which often cannot be satisfied by the use of customs declarations only. For further information on the integrated approach to economic statistics, see chapter XI.
4. Foreign shipping manifests
3.15. Use of foreign shipping manifests. Foreign shipping manifests may contain some of the same information that is relevant to trade statistics as is found on customs declarations; they may be of use as a source in cross-checking and/or supplementing information gathered from customs declarations. Quantities in weight and number are usually available, as is information on freight costs, general description of the commodities (although the commodity code may be missing), names and addresses of the parties to the transaction and country of shipment. Such other information as labour charges for packing, value of packages and fees for docking of cartage, marine insurance, inland freight and some other commissions may also be available. The main deficiency of foreign shipping manifests is that the value of the goods is frequently missing and if the value is provided, it may summarize a number of tariff lines, may be entered in the currency of the exporting country or may be on a free-on-board (FOB) basis.
3.16. Cooperation with port administrations. In some countries, port administrations produce certain statistics from shipping manifests for port management purposes. Those statistics may also be used to cross-check the data collected from customs declarations. Ideally, there should be collaborative agreements between the statistical office, customs and port administrations, aimed at mutual assistance in the compilation of trade-related statistics.
5. Currency exchange records and records of monetary authorities
3.17. Use of information from International Transactions Reporting Systems. Under an international transactions reporting system (ITRS), banks and other financial institutions are required to collect information on all transactions between residents and non-residents that have a corresponding financial flow and which are settled through them. That information is then supplied to the central bank for regulatory and/or statistical purposes. Those records may provide a supplementary source of information and information for cross-checking the customs-based trade statistics, as applicable.
3.18. Limitations in the use of data from an ITRS. An international transactions reporting system is set up to reflect financial flows that do not necessarily correspond to the physical movements of goods relevant to international merchandise trade statistics. Therefore, due care should be exercised to separate cross-border merchandise flows from service, income, transfers, and financial flows. Also, the partner-country attribution in ITRS would be based on the residence of the transactors rather than on the country of origin of the goods and their country of last known destination. Although ITRS may be used to provide an early broad estimate of total merchandise trade, commodity or country detail is invariably less detailed. Quantity data may not be covered at all.
3.19. Further limitations. There may be potential biases in the data if there are exchange controls that may encourage understatement of exports and overstatement of imports; those may be harder to identify because an ITRS does not provide the possibility of inspection. Timing issues also arise for an ITRS, since a financial transaction is measured at the time it is handled through the banking system. This may result in a recorded time different from that when the goods changed ownership (as required for balance-of-payments and national accounts statistics) or that when the goods crossed borders (as used in international merchandise trade statistics). Another disadvantage of ITRS is that it can combine two or more transactions in a single bank settlement. Especially in those cases where these transactions concern both imports and exports, the ITRS becomes less useful as a source of information.
3.20. Advantages in the use of data from an ITRS. The advantage of an ITRS is that sometimes it can provide more timely total trade data than a survey or customs system. For example, provision of data from an ITRS may be faster because customs declarations from some border posts may take longer to arrive, or the central bank and/or commercial banks may have fully computerized systems which operate faster than the systems used in the customs and/or statistical offices.
6. Reports of commodity boards
3.21. Use of reports from commodity boards. Commodity boards are quasi-governmental or commercial organizations established in some countries that monitor the production and shipment of goods considered economically important for a country; they may also market the products internationally on behalf of the producers. Those boards often issue reports that show the volume of commodities exported during a particular period: monthly, quarterly or yearly. The reports may include details, such as quantity of products sold (e.g., in metric tons), the value of the sales, country of destination of the commodities, and (probably) the administrative costs expended. If commodity board reports are consistently available, they may serve as supplementary sources or for cross-checking of customs records, especially regarding the information on quantities. In such a case, compilers are advised to analyse data from these reports and to use them as appropriate. For an example of the use of information from commodity boards, see chapter VII.
7. Administrative records associated with taxation
3.22. Use of administrative records associated with taxation. When customs records are not available or are incomplete, it may be possible for administrative records associated with taxation to be used in deriving trade statistics. This is particularly true where value added tax (VAT) systems are in place, as is the case in the European Union (see chapter X for more information).
8. Data exchanges between countries
3.23. Use of data exchanges between countries. In some circumstances, particularly where non-reporting or errors in the collected data are prevalent, a data exchange between partners can improve data quality and reduce the burden on traders and statistical compilers; exchanges could cover all transactions or only a subset of transactions thought to involve special problems. The exchange may be a permanent arrangement or may be limited to a specific time frame for dealing with a temporary situation. Given the confidentiality aspect of the data exchange, it may be necessary to have a signed agreement between the partners. Examples of country data exchanges are provided in chapter IX, section C, and chapter X, section D, discusses further possibilities for such exchanges. Chapter XXVI covers the issue of confidentiality in more detail.
3.24. Reconciliation studies. Before undertaking a data exchange, it is important to conduct a detailed trade data reconciliation study (see chap. IX) to fully understand the differences between the two partners' statistics and the adjustments that will be needed, for example, to derive each partner's export data from counterpart import statistics (see chap. XVI on issues related to partner-country attribution). Because of the greater customs scrutiny to which imports are subjected by customs in most countries, it is usually more feasible to derive estimates of exports from counterpart imports.
3.25. Advantages and disadvantages. Data exchange can substantially reduce reporter burden and improve data quality, particularly if certain trade flows with a partner country are underreported. It can also foster greater communication and cooperation between the customs and statistical agencies in the two countries. In the case where the partner’s imports information is used as the country’s exports information, the reporting burden on importers may increase if they have to report additional data elements to meet the exporting country's needs. Further, such exchange may reduce each partner's flexibility in modifying its classifications and processes. Because of the need to align classification and processing schedules, it would be difficult to implement data exchanges with multiple trading partners. It may also be difficult to implement a data exchange when there are significant amounts of trade transiting one partner en route from the other to a third country, or with distant partners where there may be large differences the time of recording.
 A similar form of trade is shuttle trade. IMTS 2010 defines “shuttle trade” broadly as “goods acquired by all categories of travellers, including non-resident workers, to a significant scale as defined by national law” (see IMTS 2010, para. 1.16). However, para. 10.19 of the Balance of Payments and International Investment Position Manual, Sixth Edition (BPM6) (Washington D.C. 2009) defines shuttle trade as follows:
a.) Shuttle trade (sometimes called informal crossborder trade) covers transactions involving the purchase of goods in an economy by travelers (nonresidents) who then transport these goods back to their economy of residence where they are to be sold; goods purchased by travelers in their home country for resale abroad; and goods purchased by travellers abroad in one economy and sold abroad in a second economy.
b. The OECD glossary of statistical terms defines shuttle trade as “the activity in which individual entrepreneurs buy goods abroad and import them for resale in street markets or small shops”. Often the goods are imported without a full declaration so that import duties can be avoided (see Organization for Economic Cooperation and Development (OECD), “Glossary of statistical terms” (Paris, 2007), available from: http:// stats.oecd.org/glossary/). Earlier discussions stress the informal character of this type of trade and its statistical underreporting (see “Shuttle trade”, a paper presented at the Eleventh Meeting of the IMF Committee on Balance of Payments Statistics, Washington D.C., 21-23 October 1998 (BOPCOM98/1/3).
 A manifest is a transport document that serves as a tally-sheet, and gives a detailed summary of all bills of lading (or air waybills) issued by a carrier (or its agent) for a particular voyage of a particular vessel or vehicle. For cargo carrying vessels or vehicles, a manifest lists its consignor, consignee, number, origin, destination, value, and other such information primarily for use by the customs authorities (see http://www.businessdictionary.com).
 Data exchanges between countries are usually exchanges of customs data. In the context of this Manual, non-domestic customs data is considered a non-customs data source, as the term “customs data” is used to refer to domestic customs data only. Within a customs union, the term “customs data” refers to customs data of all member States.
 One form of data exchange entails the use of one country's imports data as a substitute or input into another country's exports data.