International trade in goods - aggregated and detailed data

Reference Metadata in Euro SDMX Metadata Structure (ESMS)

Compiling agency: Eurostat, the statistical office of the European Union


Eurostat metadata
Reference metadata
1. Contact
2. Metadata update
3. Statistical presentation
4. Unit of measure
5. Reference Period
6. Institutional Mandate
7. Confidentiality
8. Release policy
9. Frequency of dissemination
10. Accessibility and clarity
11. Quality management
12. Relevance
13. Accuracy
14. Timeliness and punctuality
15. Coherence and comparability
16. Cost and Burden
17. Data revision
18. Statistical processing
19. Comment
Related Metadata
Annexes
Footnotes



For any question on data and metadata, please contact: Eurostat user support

Download


1. Contact Top
1.1. Contact organisation

Eurostat, the statistical office of the European Union

1.2. Contact organisation unit

G5 - Trade in Goods

1.5. Contact mail address

2920 Luxembourg LUXEMBOURG


2. Metadata update Top
2.1. Metadata last certified 19/03/2021
2.2. Metadata last posted 15/11/2021
2.3. Metadata last update 15/11/2021


3. Statistical presentation Top
3.1. Data description

International trade in goods statistics (ITGS) published by Eurostat measure the value and quantity of goods traded between the EU Member States (intra-EU trade) and goods traded by the EU Member States with non-EU countries (extra-EU trade). ‘Goods’ means all movable property including electricity. ‘European’ means that the statistics are compiled on the basis of the concepts and definitions set out in EU legislation. ‘National’ statistics, i.e. statistics published at national level by the Member States, are compiled on the basis of national rules which may differ from EU rules. European ITGS are the official harmonised source of information about exports, imports and the trade balances of the EU, its Member States and the euro area.

 

Aggregated versus detailed data

International trade in goods statistics are published through different datasets grouped into two categories:

  • Aggregated data refer to macroeconomic indicators for the EU and euro area. Monthly (short-term indicators) and annual (long-term indicators) data are aggregated by broad product categories. Broad product categories are defined as one-digit codes of the Standard International Trade Classification (SITC) or aggregates of the Broad Economic Categories (BEC).
  • Detailed data refer to the most detailed level of the following product nomenclatures: the Combined Nomenclature (CN), the SITC, the BEC, the Classification of Products by Activity (CPA) and the Standard Goods Classification for Transport Statistics/Revised (NSTR). Detailed data also contain product aggregations at higher levels.

 

Statistical dimensions

Data periodicity (monthly or yearly) and product nomenclature (CN, SITC, BEC or CPA) differ depending on the dataset, but the following statistical fields are always available:

  • reporting country: country or geo-economic area such as EU or euro area;
  • partner country: EU Member State, non-EU country or geo-economic area.
  • reference period: month and/or year;
  • trade flows: import and export; and
  • product according to the relevant classification.

Besides the dimensions listed above, specific datasets contain information on the mode of transport (e.g. by sea, by air or by road) or the statistical procedure (normal trade versus trade for processing activities).

3.2. Classification system

Combined Nomenclature as primary product nomenclature

The Combined Nomenclature (CN) is the primary nomenclature as it is the one used by the EU Member States to collect detailed data on their trading of goods. It is based on the Harmonised Commodity Description and Coding System (managed by the World Customs Organisation (WCO)). The CN corresponds to the HS plus a further breakdown at eight-digit level defined to meet EU needs. It includes around 9 500 eight-digit codes and is subject to annual revisions that ensure it is kept up to date in the light of changes in technology or patterns of international trade in goods.

 

Other product nomenclatures

Eurostat manages correspondence tables enabling the transposition of detailed data collected according to the Combined Nomenclature into other classifications like the Standard International Trade Classification (SITC), the Classification of Products by Activity (CPA), the Broad Economic Activities (BEC) 

  • Standard International Trade Classification (SITC) managed by the United Nations) is correlated with the subheadings of the Harmonised System. SITC Rev. 4 comprises 2 970 basing headings which are aggregated into 262 groups, 67 divisions and 10 sections. The section level is the one used to disseminate aggregated data.
  • Classification of Products by Activity (CPA) – The CPA is a European version of the United Nations’ Central Product Classification (CPC), but arranged so that each product heading is assignable to a single heading of the European activity classification, the NACE Rev. 2. CPA enables trade statistics to be considered jointly with other sets of general economic statistics — such as national accounts, employment statistics and industrial statistics — for productivity analysis, links with input-output tables and the impact of trade on employment.
  • Broad Economic Categories (BEC) – Like the SITC, the BEC is managed by the United Nations. It permits the conversion of international trade data based on the SITC into end-use categories. The BEC has 19 basic categories that can be aggregated to approximate the three basic classes of goods (capital, intermediate and consumption goods).
  • NST-2007 – The Standard goods classification for transport statistics, is a statistical nomenclature for the goods transported by modes of transport. The NST 2007 considers the economic activity from which the goods originate. This classification is used since 1 January 2007. It comprises 20 chapter headings.

Country classification

An ISO alpha-2 coding applies, which means that each country is identified with a two-letter alphabetical code. See the publication Geonomenclature applicable to European statistics on international trade in goods for more information.

 

Statistical classifications can be accessed from EU Vocabularies.

3.3. Coverage - sector

Trade in goods statistics consist of both intra- and extra-EU trade flows. They cover all goods entering (imports) or leaving (exports) the statistical territories of the EU Member States. The statistical territory of a Member State corresponds to its customs territory with one exception: the statistical territory but not the customs territory of Germany includes Heligoland.

3.4. Statistical concepts and definitions

Reporting country – Except for some specific goods like vessels and aircraft, ITGS follow the physical movements of the goods. Member States should record an import when goods enter their statistical territory and an export when goods leave that territory except if those goods are in simple transit. Goods should be recorded only when adding to or subtracting from the stock of national material resources or, in the context of extra-EU trade, when customs formalities are applied.

 

Partner country – This is the last known country of destination for intra- and extra-EU exports, the country of origin for extra-EU imports and the country of consignment for intra-EU imports

 

Product code – Goods are primarily classified by commodity code as set out in the EU Combined Nomenclature. Eurostat manages correspondence tables enabling the transposition of detailed data collected according to the Combined Nomenclature into other classifications like the Standard International Trade Classification (SITC), the Classification of Products by Activity (CPA), the Broad Economic Activities (BEC) or the standard goods classification for transport statistics (NST 2007). 

 

Statistical procedure (only in detailed data relating to extra-EU trade) This relates to customs procedures and distinguishes between different types of imports and exports: imports/exports covered by the customs inward processing procedure and imports/exports covered by the customs outward processing procedure are distinguished from normal transactions or transactions not recorded from customs declarations.

 

Mode of transport (only in detailed data)– This identifies the active means of transport (e.g. road, rail, sea) by which the goods leave/enter the statistical territory of a Member State for intra-EU trade and of the EU for extra-EU trade. Note that the collection of the mode of transport for intra-EU trade flows became optional in 2001. Therefore the information is not available for all the EU Member States since January 2001 as reference month.

 

Container (only in detailed data relating to extra-EU trade)

In addition to the mode of transport, for extra-EU trade, information is collected on whether or not goods are transported in containers (except when the mode of transport is postal consignment, fixed transport installation or the own propulsion category).

 

Nationality of the active means of transport (only in detailed data relating to extra-EU trade)

In addition to the mode of transport, for extra-EU trade, information is collected on the nationality of the active means of transport of the goods when they leave/enter the statistical territory of the EU (except when the mode of transport is rail, postal consignment, fixed transport installations and own propulsion). Note that the nationality of the means of transport became optional in 2010. Therefore the information is not available for all the EU Member States since January 2010 as reference month.

 

Trade value

This is the statistical value, i.e. the amount that would be invoiced in the event of sale or purchase at the national border of the reporting country. It is said to be a FOB (Free On Board) valuation for exports and a CIF (Cost Insurance Freight) valuation for imports. Hence, only incidental expenses (freight, insurance) are included and incurred for:

  • exports in the part of the journey located on the territory of the country where the goods are exported from;
  • imports in the part of the journey located outside the territory of the country where the goods are imported to.

 

Quantity (only in detailed data)

This is the weight of the goods in kilograms without packaging; this quantity is referred to as the ‘net mass’. For certain goods, a supplementary quantity is available. The supplementary unit varies according to the goods; it can be litres, number of pieces, carats, terajoules, square metres, etc. as detailed in the annual Commission regulation updating the Combined Nomenclature.

3.5. Statistical unit

Any natural and legal person lodging a customs declaration in a Member State is reporting to the extra-EU trade statistics on the condition that the customs procedure is of statistical relevance.

Within intra-EU trade statistics any taxable person carrying out an intra-EU trade transaction is responsible for providing the information. However small and medium trade operators are exempted from the obligation to provide Intrastat declarations. The Intrastat system is based on thresholds which allow intra-EU traders not to report on their transactions or provide less detailed information on condition that their total trade value does not exceed a certain amount during the previous or present calendar year. Those thresholds are fixed according to quality standards defined in the Intrastat legislation.

3.6. Statistical population

Trade in goods of the EU and its Member States includes all goods which add or subtract from the stock of material resources of the reporting Member State by entering (imports) or leaving (exports) its economic territory including goods for processing. Information on the goods is provided by legal or natural person.

3.7. Reference area
  • European Union (as aggregate and for each EU Member State)
  • Euro area (as aggregate and for each EA Member State)
3.8. Coverage - Time

For long term indicators (annual data):

  • European Union and Euro area as reporting entity since 2002
  • All Member States since 2002

For short term indicators (monthly data):

  • European Union as reporting entity since 2002, Euro area as reporting entity since 1999
  • All Member States since 1999 (except Croatia), Croatia since 2002

Detailed data  (monthly data)

  • European Union and Euro area as reporting entity since 2000;
  • Member States since 1988 or a more recent year, depending on the dataset and reporting country;
  • EFTA countries since 1995: Iceland, Liechtenstein, Norway and Switzerland;
  • United Kingdom until October 2020 as reference period (*);
  • United Kingdom (in respect of Northern Ireland) from January 2021 as reference period (**); and
  • Enlargement countries (candidate countries and potential candidates) since 2002: Albania, Bosnia and Herzegovina, Kosovo, Montenegro, North Macedonia, Serbia and Turkey.

(*) As a reporting country, the UK is no longer legally obliged to transmit any data to Eurostat since the end of the transition period (31 December 2020). This concerns new reference periods, as well as revisions of data previously transmitted to Eurostat. Currently the ITGS data transmitted by the UK to Eurostat cover the full year 2019, and months January to October 2020. Pending a new agreement on statistical cooperation between the EU and the UK, Eurostat will not publish any new data for reporter UK, whether via the database or other products, such as statistical publications.

 (**) The data transmitted by the United Kingdom, in the context of the Withdrawal Agreement Protocol on Ireland / Northern Ireland, do not include information on the trade in goods between Northern Ireland and the rest of the United Kingdom. The United Kingdom was requested to complete the data transmission with this missing information.

3.9. Base period

Starting from October 2019 as release date, trade in goods indices (IVU - unit value index and IVOL - volume index) are calculated for the reference year 2015 (2015 = 100) instead of 2010. This change has a limited impact on the calculation of indices, because IVU and IVOL are chained indices, that is for every period the base year is the previous year.


4. Unit of measure Top

Aggregated data - Long-term indicators (annual data)

  • trade value (in billion euro);
  • shares by reporting country or by main trading partners;
  • unit-value indices;
  • volume indices; and
  • shares by invoicing currency.

 

Aggregated data – Short-term indicators (monthly data)

  • gross and seasonally adjusted value (in million euro) for exports, imports and trade balances;
  • growth rates for exports and imports based on gross and seasonally adjusted values
  • unit-value indices;
  • gross and seasonally adjusted volume indices;
  • ratio of export and import indices; and
  • growth rates for gross and seasonally adjusted indices.

 

Detailed data (monthly and annual data)

  • trade values (in euros or national currency
  • quantities in 100 kg (net mass)
  • wherever relevant, quantities in the supplementary unit, according to the Combined Nomenclature (e.g. litres, square metres, number of items)


5. Reference Period Top

Theoretically, the reference period for the information on international trade in goods transactions should be the calendar month of export or import of the goods. However, in practice the reference period for extra-EU trade is generally the calendar month during which the customs declaration is accepted by customs authorities. The reference period for intra-EU trade may be adapted in case of sales or purchases to the calendar month during which the VAT on the intra-EU supplies or acquisitions becomes chargeable. The chargeable event relates to the issue date of the invoice.


6. Institutional Mandate Top
6.1. Institutional Mandate - legal acts and other agreements

General statistical legislation

Regulation (EC) No 223/2009 of the European Parliament and of the Council on European statistics

 

Intra-EU trade legislation (or Intrastat)

  • Regulation (EC) No 638/2004 of the European Parliament and of the Council
  • Implementing Commission Regulation (EC) No 1982/2004

 

Extra-EU trade legislation (or Extrastat)

  • Regulation (EC) No 471/2009 of the European Parliament and of the Council
  • Implementing Commission Regulation (EC) No 92/2010
  • Implementing Commission Regulation (EC) No 113/2010

 

All regulations relevant for the European statistics on international trade in goods can be found in the publication Legislation on European statistics on international trade in goods or consulted from the Legislation page of the International trade in goods section on Eurostat website. All legal texts of the EU are accessible on Eur-Lex.

6.2. Institutional Mandate - data sharing

Not applicable


7. Confidentiality Top
7.1. Confidentiality - policy

Regulation (EC) No 223/2009 on European statistics (recital 24 and Article 20(4)) of 11 March 2009 (OJ L 87, p. 164), stipulates the need to establish common principles and guidelines ensuring the confidentiality of data used for the production of European statistics and the access to those confidential data with due account for technical developments and the requirements of users in a democratic society.

As a general definition, data used by national and EU authorities for producing statistics are considered confidential if statistical units can be identified, either directly or indirectly, and information about individuals or businesses is disclosed as a result.

Note that confidential information can be found only in detailed data or in data by enterprise characteristics. The aggregated data or the data by invoicing currency are not detailed enough to make it possible to identify a specific trader.

7.2. Confidentiality - data treatment

Principle of passive confidentiality — The National Statistical Authority has to take appropriate measures only if requested to do so by companies which feel that their interests would be harmed by publication of the data. This principle is recommended by the United Nations in its IMTS 2010 publication and set out in the EU legislation for detailed statistics on intra- and extra-EU trade. It has the great advantage of limiting the loss of information for users and thus making the data more useful.

 

Criteria for granting confidentiality — The choice of the criteria to be met by the company is left to the National Statistical Authority. The most common criteria are the following:

  • the number of companies on the market is limited, e.g. fewer than three companies contribute to a single data cell; or
  • the company is in a dominant position on the market (by representing, for instance, at least 75 % of the total trade) or is at least an important stakeholder for a specific data cell.

 

Data elements possibly hidden – In detailed data, confidentiality can apply to the product code and/or the partner country. Information about a product or partner may be considered commercially sensitive by the provider of the statistical information for the trade value, the quantity or the unit value (value/quantity) which can be considered a proxy for the price. When confidentiality is granted, data are hidden at the detailed level, which means that the trade is not allocated to the real product code and/or the real partner. It is nevertheless included in the total trade of the reporting country (EU Member State or EFTA country) and even, wherever possible, allocated to the real chapter (i.e. to the real HS2 code). This minimum level of provision is set out in the EU legislation: Member States shall transmit data declared confidential to Eurostat so that they may be published at least at chapter level of the CN provided confidentiality is thereby ensured.


8. Release policy Top
8.1. Release calendar

The precise date of the monthly release for international trade data is disseminated on the website. News releases are issued and Euro indicators are updated on the web at 11:00 am CET on the day indicated.

8.2. Release calendar access

Release calendar - Eurostat (europa.eu)

8.3. Release policy - user access

In line with the EU legal framework and the European Statistics Code of Practice Eurostat disseminates European statistics on Eurostat's website (see item 10 'Accessibility and clarity') respecting professional independence and in an objective, professional and transparent manner in which all users are treated equitably. The detailed arrangements are governed by the Eurostat protocol on impartial access to Eurostat data for users.


9. Frequency of dissemination Top

Aggregated data: Long-term and short-term indicators are updated on a monthly basis.

Detailed data: Most datasets are updated on a monthly or daily basis. A few contain only annual data and are updated on a yearly basis.


10. Accessibility and clarity Top
10.1. Dissemination format - News release

News releases on-line

10.2. Dissemination format - Publications

Statistics Explained articles – Statistics Explained is an official Eurostat website presenting all statistical topics in an easily understandable way. Together, the articles make up an encyclopedia of European statistics, completed by a statistical glossary clarifying all terms used and numerous links to further information and the very latest data and metadata. The website is a portal for occasional and regular users alike. All Statistics Explained relating to ITGS can be accessed under this link. A series of articles are available, among them:

The articles in Statistics Explained are regularly updated (usually once a year) and can be included in virtual publications, such as International trade in goods - a statistical picture.

 

News Releases  – First results on euro area and EU trade flows are published in the Euro indicators News Release on international trade in goods. The monthly news release is published online around 46 days after the reference month. The precise publication date is announced on the Eurostat website. Ad-hoc news releases are also published in particular when detailed data or data by invoicing currency are available for a new reference year.

 

All publications are accessible free of charges via the following link: https://ec.europa.eu/eurostat/web/international-trade-in-goods/publications 

10.3. Dissemination format - online database

The Eurostat website is the only place the European ITGS are published. The data are accessible through different paths, the main one being the ‘Database’ page of the ‘International trade in goods’ dedicated section.

 

The so-called ‘Selected dataset’ are predefined tables focusing on key indicators. They are refreshed automatically when data are loaded into the databases. The ITGS Main tables include key indicators for the EU, its Member States and the euro area both annually (long-term indicators) and monthly (short-term indicators). Data are available only for a limited range of products (e.g. SITC aggregates).

 

The ‘Database’ entry point gives access to several datasets classified under the folders ‘International trade in goods – aggregated data’ (under which can be found the short- and long-term indicators) and ‘International trade in goods – detailed data’. The last heading under the latter folder concerns access to the Easy Comext database via a tailor-made application for ITGS. Easy Comext can also be accessed directly at http://epp.eurostat.ec.europa.eu/newxtweb/ or through an internet search for ‘Easy Comext’. Users needing to extract huge amounts of data should use Comext Analytical which is accessible through the Easy Comext interface but requires a login (‘Login’ button). User manuals are available under this link.

10.4. Dissemination format - microdata access

Not applicable

10.5. Dissemination format - other

The Bulk Download facility allows users to download Eurostat datasets in a format which can easily be imported into a chosen tool for further analysis. International trade in goods statistics are available in csv format from the Comext domain.

10.6. Documentation on methodology

User Guide on European statistics on international trade in goods – The purpose of this Guide is to explain to a wide range of users how the statistics relating to trade in goods, both between EU Member States and with non-EU countries, are collected, compiled, processed and published at European level. The different issues are tackled in a question and answer format.

Compilers Guide on European Statistics on international trade in goods – This Guide provides clarifications, often through concrete examples, to help the compilers to better interpret and apply the complex legislation on ITGS.

10.7. Quality management - documentation

Quality Report on European statistics on international trade in goods — This Report provides users with a tool to assess the quality of the international trade in goods statistics published by Eurostat. The data quality can be assessed against indicators covering the following components: relevance, accuracy, timeliness and punctuality, accessibility and clarity, comparability and coherence.


11. Quality management Top
11.1. Quality assurance

Quality reporting and assessment

Both Intrastat and Extrastat basic acts (in force since 2009 for Intrastat and since 2010 for Extrastat) include standardised quality articles. These articles are in line with the ESS Quality definition, European Statistics Code of Practice and the Regulation of European Statistics. In particular, they require the Member States to provide Eurostat with annual quality reports within a fixed deadline.

Data quality checks prior to any dissemination

The prime responsibility for ensuring the data completeness and accuracy rests with the National Statistical Authorities. Further checks are carried out by Eurostat, essentially to ensure that the transmission of the requested data has been carried out satisfactorily, that datasets are complete, error-free, and there are no extreme values (outliers).

11.2. Quality management - assessment

Relevance

The key users of European ITGS and their respective needs are very well known. The data relevance can be assessed as good thanks to regular satisfaction surveys and frequent contacts with users.

 

Accuracy

The European ITGS benefit from well-established data collection systems supported by effective validation and compilation tools. Nevertheless, the intra-EU trade statistics may suffer from late or non-response from the trade operators liable to statistical reporting. The issue is however limited to the detailed levels of the product classification as the Intrastat legislation makes mandatory the compilation of estimates for any missing data at least at the 2-digit level of the Combined Nomenclature (HS2) and by partner Member States. Over the last years, actions were also carried out to push the Member States facing high issues of non- or late-response to implement any necessary measures to ensure the data exhaustiveness.

Additionally, confidentiality has an impact on data accuracy at very detailed (i.e. CN eight-digit) level. Aggregated levels are in general much less impacted thanks to the legal obligation for reporting countries to ensure dissemination at least at two-digit level of the Combined Nomenclature. The EU legislation however foresees derogation to that principle when the dissemination of real trade values at chapter level would disclose confidentiality information and would then harm the interest of the economic operator. In such a case, it is allowed to use the HS chapter 99 instead of the real chapter.

That said, it should be kept in mind that basic data consist of millions of records to be produced every month, which means that it is impossible to achieve complete accuracy. As in all statistical work, a balance has to be struck between the resources devoted to checking and the likely benefit. Therefore the users should be aware of the margin of inaccuracy in the data used, at least at the most detailed level of data.

 

Timeliness and punctuality

The current timeliness is assessed as good for all data flows and no further acceleration is requested by the users except for data by enterprise characteristics which should be available only within 18 months after the reference year. This late availability is not due to the standard trade in goods data (first annual results available within 3 months after the reference year) but to the late update of the Business Register which is the information source for the enterprise characteristics.

Thanks to well-established data collection and compilation procedures, the punctuality of data transmissions to Eurostat is very good. Issues are very rare and generally minor in terms of delay.

 

Accessibility and clarity

The accessibility of trade in goods data is very good thanks to the use of all Eurostat standard dissemination channels: news releases, pdf and on-line publications, visualisation tools, predefined tables, databases and bulk download facility allowing users to download Eurostat datasets in a format which can easily be imported into a chosen tool for further analysis.

The dissemination of the European ITGS is supported by a complete set of structural metadata that make it easy to identify, retrieve and browse the data. The reference metadata describing the contents and the data quality are also quite exhaustive. In particular, it is worth mentioning the User Guide on European statistics on international trade in goods and the Quality Report on European statistics on international trade in goods.

 

Coherence

The coherence is first ensured by the Intrastat and Extrastat regulations which set up the concepts and definitions applicable to the compilation of European ITGS and secondly by a set of recommendations promoting the best methods and practices among the reporting countries.

The coherence is further strengthened by Eurostat’s harmonised approach to data production and dissemination regardless of the type of data and the reporting country.

 

Comparability

  • Comparability over time — The European ITGS benefit from a high level of comparability over time thanks to the stability of the concepts, definitions and classifications. As far as possible, this comparability is also maintained when methodological changes occur, thanks to the retroactive recalculation of the time series.
  • Comparability across countries — The European ITGS benefit from a high level of comparability across countries thanks to the implementation of harmonised rules for data collection and compilation. The comparability across countries could be however improved through further harmonisation in the national practices as regard some specific goods or movements.
  • Comparability across domains — The European ITGS constitute an essential source of information for the compilation of the balance of payments statistics and national accounts. Nevertheless, comparability across domains is affected by differences in concepts and definitions like, for instance, the application of the principle of physical movements through the national frontier for trade in goods statistics versus the change of economic ownership between residents and non-residents for the balance of payments.


12. Relevance Top
12.1. Relevance - User Needs

As international trade forms a major part of the world economy, statistics on trade in goods are an instrument of primary importance for numerous public and private sector decision makers. For example, international trade statistics

  • enable EU authorities to prepare multilateral and bilateral negotiations under the common commercial policy;
  • enable EU authorities to evaluate the progress of the Single Market and the integration of EU economies;
  • constitute an essential source of information for balance of payments statistics, national accounts and economic studies; and
  • help EU businesses conduct market research and define their commercial strategy.

Statistics satisfy these needs in a variety of ways. Users may need either annual aggregated or detailed monthly data on products or partner countries. They may be interested in trade values in current prices or at constant prices. Alternatively, their interest may be in quantities rather than in values. These examples, which are far from exhaustive, show the diversity of users and their requirements. Eurostat tries to meet these various needs and toadapt toachanging environment, such as changes due to globalisation.

12.2. Relevance - User Satisfaction

User needs are monitored regularly. At EU level, there are regular contacts with key institutional users (e.g. Commission services or the European Central Bank) and with other main user groups such as trade associations. In daily work, users can easily communicate their requests and needs to Eurostat by using dedicated tools integrated in the data dissemination. Many countries also conduct regular user satisfaction surveys.

Eurostat undertook its last large-scale user satisfaction survey focusing on ITGS in 2007. The survey found a very high level of satisfaction among users: 85 % were very or fairly satisfied with the data. This high level of satisfaction has been confirmed by every general User Satisfaction Survey carried out by Eurostat on an annual basis since then 2009. In the 2017 general survey, ‘International trade’ received the second most positive evaluation of all the statistical domains, with 62.3 % of respondents rating the data quality as very good or good and 20.4 % as adequate. ‘Economy and finance’ statistics received the highest evaluation, with 64.8 % (very good or good) and 16.8 % (adequate) respectively. These two domains outperformed the average rates

12.3. Completeness

The EU trade statistics are based on the EU legislation which is directly applicable in the Member States. In particular, the legislation includes a clear and precise list of all the statistical variables to be provided by the Member States to Eurostat. All the mandatory variables are provided by all the Member States.


13. Accuracy Top
13.1. Accuracy - overall

The European ITGS benefit from well-established data collection systems supported by effective validation and compilation tools. Nevertheless, the intra-EU trade statistics may suffer from late or non-response from the trade operators liable to statistical reporting. The issue is however limited to the detailed levels of the product classification as the Intrastat legislation makes mandatory the compilation of estimates for any missing data at least at the 2-digit level of the Combined Nomenclature (HS2) and by partner Member States. Over the last years, actions were also carried out to push the Member States facing high issues of non- or late-response to implement any necessary measures to ensure the data exhaustiveness.

Additionally, confidentiality has an impact on data accuracy at very detailed (i.e. CN eight-digit) level. Aggregated levels are in general much less impacted thanks to the legal obligation for reporting countries to ensure dissemination at least at two-digit level of the Combined Nomenclature. The EU legislation however foresees derogation to that principle when the dissemination of real trade values at chapter level would disclose confidentiality information and would then harm the interest of the economic operator. In such a case, it is allowed to use the HS chapter 99 instead of the real chapter.

That said, it should be kept in mind that basic data consist of millions of records to be produced every month, which means that it is impossible to achieve complete accuracy. As in all statistical work, a balance has to be struck between the resources devoted to checking and the likely benefit. Therefore the users should be aware of the margin of inaccuracy in the data used, at least at the most detailed level of data.

13.2. Sampling error

International trade in goods statistics are not collected via samples and so are not impacted by sampling errors. Intra-EU trade data are collected via the Intrastat system which is considered as a cut-off census. Extra-EU trade data are based on the records of trade transactions in customs declarations and are so considered to be of administrative source.

13.3. Non-sampling error

Errors often arise when trying to capture movements of goods or trade activities about which little information is available or which are inherently complex to measure, such as:

  • Goods that by their nature are difficult to classify — Errors may arise in the product code allocated due to the complexity of the classification (around 9 500 subheadings in the CN), for instance for chemical or pharmaceutical products.
  • Processing trade valuation — Errors arise when the processing costs are reported rather than the trade value corresponding to the total amount which would be invoiced if the goods were sold or bought.
  • Product with embedded services — Errors arise when the total amount of the contract (transaction including the supply of goods and services) is reported rather than the value of the goods alone.
  • Repairs versus processing — Repairs should be excluded from trade in goods statistics but processing should be included. Errors may arise when the distinction between repairs and processing is difficult to make.
  • Quasi transit — Intrastat declarations may be missing for goods imported from a non-EU country for immediate dispatch to another Member State or goods imported from another Member State for immediate dispatch to a non-EU country.
  • Triangular trade — trade involving three Member States through sales/purchase contracts but with physical movement of goods between only two of them. A problem arises when the trade is reported according to the sales/purchase contract rather than the physical movement.

 

Besides errors, the accuracy of trade in goods statistics at detailed level is impacted by estimates for non-collected data.

  • Estimates for missing intra-EU trade – In order to reduce the statistical burden on businesses, intra-EU trade data are collected only from the biggest intra-EU traders. Only traders whose annual intra-EU trade exceeds a certain threshold have to submit Intrastat declarations. This means that the Intrastat data collection does not cover 100 % of the intra-EU trade. However the EU legislation requires complete coverage, which means that the loss caused by the thresholds and by non-response from non-exempted traders must be compensated with estimates.
  • Estimates for missing extra-EU trade – Theoretically, the problem of late or non-response should not exist for extra-EU trade data as Extrastat, the data collection system, is based on customs declarations. Nevertheless, estimates may be still necessary to compensate for delayed or incomplete customs records. It should also be noted that, for simplification purposes, the reporting countries are allowed to compile less detailed information for transactions below the statistical threshold of EUR 1 000 and 1 000 kilograms.
  • Estimates for non-collected statistical value – In intra-EU trade, only the invoice value — the amount agreed on the sales agreement — is systematically collected from the providers of statistical information (PSIs). When not provided by the PSIs, the statistical value shall be estimated by the National Statistical Authority. Note that, in extra-EU trade, the statistical value is based on the value determined for customs purposes; therefore, there is generally no need to estimate it.
  • Estimates for non-collected net mass – Under the legal provisions on intra-EU trade statistics, where there is a supplementary unit laid down for a specific CN product code (e.g. litres, m², pieces), it is not mandatory to request the specification of net mass from the PSIs. However the net mass shall be transmitted to Eurostat whatever the CN8 code. It should be so estimated by the National Statistical Authorities if not collected.


14. Timeliness and punctuality Top
14.1. Timeliness

According to the EU legislation, Member States shall provide Eurostat with monthly aggregated data within 40 calendar days after the reference month.

Member States shall provide Eurostat with monthly detailed data within 40 calendar days after the reference month for extra-EU trade (*) and within 70 calendar days after the reference month for intra-EU trade.

First results (including estimates) on Euro area and EU trade balances are published on line around 46 days after the reference month in the Euro-indicators news release on international trade in goods.

The latest supplied detailed data (new reference periods and revisions) are published at the date of the monthly news release. Long-term and short-term indicators are updated at the same date.

 

(*) Data on trade with UK result from the aggregation of figures collected via the Intrastat system (trade with UK (in respect of Northern Ireland)) and via customs declarations (trade with UK (excluding Northern Ireland)). Intrastat data are generally available with a 1-month time lag. This means that data on trade with UK might be available later than data on trade with the other non-EU countries.

14.2. Punctuality

Thanks to well-established data collection and compilation procedures, the punctuality of data transmissions to Eurostat is very good. Issues are very rare and generally minor in terms of delay.


15. Coherence and comparability Top
15.1. Comparability - geographical

European figures versus national figures

EU legislation serves as a basis for compiling the intra- and extra-EU trade statistics published by Eurostat. However, European statistics, which cover the EU as a whole, and the statistics published by the Member States, are not always directly comparable. Member States may apply a different concept at national level but they have to provide Eurostat with harmonised data according to the Community concept. The most common differences between the Community concept and the national concepts are as follows:

  • Use of the general trade system at national level while the Community statistics are compiled according to the special trade system;
  • Exclusion from national statistics of ‘quasi-transit’, which means of
    • goods imported from a non-EU country, cleared through customs and immediately dispatched to another Member State (the Member State of final destination); or
    • goods imported from another Member State (the Member State of actual export), cleared through customs and immediately dispatched to a non-EU country.

The customs formalities distinguish between simple transit, which is not recorded in Community statistics, and quasi-transit.

  • Inclusion of repairs in national statistics — these are excluded from the scope of Community statistics from 2006; and
  • Country of origin vs. Member State of consignment — for Community statistics for intra-EU imports, the partner country is the Member State of consignment but for national statistics it may be the country of origin.

 

Asymmetries in intra-EU trade statistics

In theory, intra-EU trade statistics should be less affected by asymmetries than extra-EU trade statistics as issued from more harmonised rules. Exports from Member State A to Member State B, as reported by A, should be almost equal to imports into B from A, as reported by B. Due to a different valuation principle (CIF > FOB), imports should be slightly higher than exports.

However, since the Intrastat system came into operation, bilateral comparisons have revealed major and persistent discrepancies in the intra-EU trade statistics. Therefore, comparisons based on intra-EU trade statistics must be handled with caution and should take these into account. The main reasons for the discrepancies are known and are partly the same as in the case of extra-EU trade. There are also factors that are specific to intra-EU trade, such as estimates for non-collected data.

  

Asymmetries in extra-EU trade statistics

There are two main approaches for measuring international trade in goods: the general trade system and the special trade system. EU ITGS use the latter, which means that goods from a non-EU country that are received into customs warehouses are not recorded unless they subsequently go into free circulation in the Member State of receipt (or are placed under the customs procedures for inward processing). Similarly, outgoing goods from customs warehouses are not recorded as extra-EU exports. The general trade system, which is used by most of the EU’s main partner countries, is broader, including all goods entering or leaving the country.

Since intra-EU trade statistics are not directly linked to customs procedures, they are not compiled on a general or special trade basis.

Comparing extra-EU trade statistics with the figures published by non-EU countries for the same trade flows inevitably highlights some discrepancies. Besides the trade system and errors such as product or partner misclassification, the most common reasons for asymmetries are:

  • Methodological differences: trade coverage (e.g. data collection thresholds, treatment of specific goods or movements of goods), definition of partner country (e.g. country of re-export vs. country of origin), definition of statistical territory, different valuation principles (e.g. FOB valuation for exports and CIF valuation for imports);
  • Time lag: the same operation is recorded for a different reference period;
  • Statistical confidentiality: the goods movement is made confidential by one of the partners;
  • Different practices in the treatment of revisions; and
  • Problems of currency conversion.
15.2. Comparability - over time

Changes in the product nomenclature

A particular issue of comparability over time concerns the product classification used for trade in goods detailed data. The most detailed data are collected and published by eight-digit codes of the Combined Nomenclature (CN). Some changes are made to the CN every year. Eurostat maintains conversion tables between successive versions of the CN in order to improve comparison over time. 

Year Creations Deletions Total Net change CN Codes
2020 54 104 158 -50 9 483
2019 9 9 18 +0 9 533
2018 18 13 31 +5 9 533
2017 687 573 1 260 +114 9 528
2016 55 27 82 +28 9 414
2015 21 14 35 +7 9 386
2014 43 40 83 +3 9 379
2013 35 42 77 -7 9 376
2012 907 818 1 725 +89 9 383
2011 132 281 413 -149 9 294
2010 180 306 486 -126 9 443
2009 127 257 384 -130 9 569
2008 75 96 171 -21 9 699
2007 917 1 039 1 956 -122 9 720
2006 486 740 1 226 -254 9 842
2005 97 175 272 -78 10 096
2004 273 503 776 -230 10 174
2003 19 15 34 +4 10 404
2002 780 654 1 434 +126 10 400
2001 50 90 140 -40 10 274

 

Methodological changes

The most important methodological changes or other events affecting ITGS in the past years are listed in the table below.

  Type of trade Event
2020 Intra- and extra-EU Change in the definition of intra- and extra-EU trade due to the United Kingdom withdrawal from the European Union from 31 January 2020.
2014 Intra-EU

For intra-EU imports, minimum coverage from collected data reduced from 95 % to 93 % of total trade value

2013 Intra- and extra-EU Change in the definition of intra- and extra-EU trade due to Croatia’s EU accession on 1 July 2013
2012 Intra- and extra-EU Combined Nomenclature impacted by the revision of the Harmonised System
2010 Intra- and extra-EU Introduction of the concept of change in economic ownership to record the trade in ships and aircraft and to determine the partner country allocation for deliveries to ships and aircraft and sea products (based on the economic ownership of the ships/aircraft)
Intra- and extra-EU Reporting Member State and partner country for goods delivered to and from offshore installations determined by the exclusive rights of a country to exploit seabed or subsoil of the area (exclusive economic zone) where the offshore installation is established
Intra-EU Estimates for non-collected net mass mandatory
2009 Intra-EU For intra-EU imports, minimum coverage from collected data reduced from 97 % % to 95 % % of total trade value
2007 Intra- and extra-EU Change in the definition of intra- and extra-EU trade due to EU enlargement (accession of Bulgaria and Romania on 1 January 2007)
Intra- and extra-EU Combined Nomenclature impacted by the revision of the Harmonised System
2006 Extra-EU Repairs excluded from the scope of extra-EU trade
Intra-EU Collection of net mass in Intrastat no longer mandatory if a supplementary quantity is collected
2005 Intra-EU Repairs excluded from the scope of intra-EU trade
Intra-EU Collection of net mass in Intrastat no longer mandatory for a specific list of CN8 codes for which a supplementary quantity is collected
2004 Intra- and extra-EU Change in the definition of intra- and extra-EU trade due to EU enlargement (accession of Czechia, Estonia, Cyprus, Latvia, Lithuania, Hungary, Malta, Poland, Slovenia and Slovakia on 1 July 2004)
15.3. Coherence - cross domain

Apart from the ITGS, information on trade flows can be found in national accounts, business statistics and balance of payments data. These are compiled and produced according to the recommendations (sources and methods) of various international organisations, e.g. Eurostat, International Monetary Fund, United Nations. The table below gives an overall idea of the main differences between these sources.

  International Trade Statistics National Accounts Balance of Payments Business Statistics
Concepts & Definitions Community concept and definitions Aggregation of national statistics based on ESA 2010 Aggregation of national statistics based upon IMF Balance of Payments Manual (Sixth edition) (BPM6). It implies some methodological discrepancies (FOB/FOB, difference of coverage). BoP statistics in Member States are most of the time derived from Member States’ ITGS. BoP statistics sent to Eurostat are compiled according to the Community concept. Statistics are based on activity sector
Statistical unit/ object/ population Cross-border movement of goods   International transaction with change of ownership Enterprise, local unit
Classifications (nomenclature) Combined Nomenclature (CN) Institutional sectors IMF classification NACE
Geographical breakdown Detailed geographical breakdown Intra-EU, Extra-EU, World Detailed geographical breakdown All partners
Reference period Monthly Annual Quarterly Annual
Correction methods National corrections National corrections National corrections National corrections
15.4. Coherence - internal

A key feature of the EU ITGS is their coherence. Aggregated data are constructed from detailed data across Member States and product classifications with the help of official correspondence tables. This ensures internal coherence. In addition, in order to maintain coherence, Eurostat calculates derived indicators such as unit value indices or seasonally adjusted series from the detailed data provided by Member States.


16. Cost and Burden Top

Performance and cost

In order to improve performance and reduce the costs of data collection and processing, the use of IT tools in data collection has been identified as a priority. Consequently, the proportion of paper declarations has decreased over time. 99% of the total intra-EU trade and 100% of the extra-EU trade are collected electronically. Paper declarations are no longer used to collect intra-EU detailed data in half of the EU Member States. Also several measures have been taken to develop and promote common tools for data validation.

 

Burden assessment

Intrastat is considered to be the largest business survey in the EU. In 2015 it was estimated that there were around 3.7 million European businesses involved in trade in goods between the Member States, out of which only about 13.1 % – around 485 000 – were liable to Intrastat reporting. Out of these 485 000 reporting businesses, 207 000 (43 %) reported only imports, 129 000 (27 %) only exports and 150 000 (31 %) reported both flows.

From the beginning of Intrastat, it became clear that the reporting burden imposed by this system on businesses was quite heavy. The Intrastat system has so undergone a number of changes and amendments of legislation in its 25-year history. These revisions have however not changed its basic characteristics but have rather enabled Member States to exempt more of their smallest businesses from Intrastat reporting obligations without making too drastic compromises in quality. While the objective of these revisions has always been clear, there has been recurring discussion on how the Intrastat system should be further simplified. Two alternative approaches were often put forward: raising the exemption thresholds and a single flow system.

In November 2011, the European Council called upon the European Statistical System (ESS) to take effective measures ensuring a substantial reduction of the response burden by redeveloping Intrastat, while maintaining at the same time, a sound level of quality. In response to this call, the ESS adopted an innovative statistical approach consisting in exchanging micro-data on intra-EU exports between the EU Member States allowing them to use those mirror data for the compiling of their own intra-EU imports statistics. Such an approach follows the principle that data collected and available within the ESS need not to be collected more than once. Thus, each trade transaction collected in one Member State may serve as a data source for two Member States: first, for compiling the intra-EU exports of the exporting Member State and, second, for compiling and/or verifying the intra-EU imports of the partner Member State. The modernised Intrastat system is being implemented on the basis of new legal provisions laid down by Regulation (EU) 2019/2152 on European Business Statistics (EBS). The EBS legislation will become applicable for international trade in goods statistics (and hence the modernised Intrastat system) starting from 1 January 2022.


17. Data revision Top
17.1. Data revision - policy

International trade in goods statistics, like many published statistics, must balance the need for timely information with the need for accuracy. Inevitably, the detailed data sent for a given month are subject to the possibility of later revision as a consequence of errors, omissions or — particularly with the Intrastat system — late declarations by the providers of the statistical information.

Data are revised frequently according to national needs and practices. However, it is recommended that the countries provide Eurostat with final detailed data at the latest by October following the reference year. At that time data become ‘final’ and should not be revised further except in exceptional and well-justified cases.

When monthly results already transmitted to Eurostat are subject to revision at national level, revised results must be transmitted no later than the month following their availability. NB: In the context of the withdrawal of the UK from the European Union, revisions of data where the UK was a reporting country (reference period up to and including October 2020) are no longer transmitted to Eurostat.

17.2. Data revision - practice

Metadata reports on revisions to trade in goods data are published monthly measuring the revision both in terms of the value change and the percentage change. Both are useful complementary measures for different users of trade in goods data. Measuring the percentage change can inform users on the effect of changes on individual Member States data. Measuring the value change allows users to be informed on the impact of the changes in trade figures from both large and small economies on EU and EA aggregate figures.

The definition a ‘large’ revision will depend very much on the data user and how they use the data. To aid both revisions monitoring and user understanding of the impact of revisions, Eurostat and Member States have agreed on thresholds to define different levels of revision and these definitions are included with the published metadata reports mentioned above. Thresholds have been defined for both changes between subsequent data publication and between when the data were first published and the most recently available data.

The metadata reports (revisions between subsequent data deliveries and revisions between first and last data deliveries) can be found on the Information on data page of the International trade in goods section on Eurostat website.


18. Statistical processing Top
18.1. Source data

Extrastat and Intrastat: two data collection systems

Traditionally ITGS are based on the data collected by customs authorities on trade transactions between countries. Customs declarations are used for statistical purposes as the basic data source which provides detailed information on exports and imports of goods with a geographical breakdown.

The first piece of EU legislation on ITGS was adopted in 1975; it provided general guidelines on data collection and obliged Member States to send their data to Eurostat. The advent of the Single Market on 1 January 1993, with its removal of customs formalities between Member States and subsequent loss of trade statistics data sources, required the establishment of a new data collection system: Intrastat. Since then ITGS are based on two data collection systems: Extrastat and Intrastat.

  • Extrastat data on trade in goods with non-EU countries are collected by customs authorities and are based on the records of trade transactions in customs declarations.
  • Intrastat data are directly collected from intra-EU trade operators once a month. Alternative data sources may be used for some specific goods and movements like for among with ships, aircraft, gas and electricity.
18.2. Frequency of data collection

Monthly

18.3. Data collection

Extra-EU trade statistics are collected via customs declarations. Trade operators fulfilling their reporting obligations to the Customs Authority in a Member State are providing statistical data on the same occasion. The statistical information depends, therefore, very much on customs practices, definitions and policies and only few dimensions are collected purely for statistical purposes. The dependence on customs procedures entails to a high quality and nearly total coverage of data on trade with non-EU countries.

 

Intra-EU trade statistics are collected directly from trade operators as a consequence of the abolishment of customs formalities at the borders between the EU Member States. Less statistical data elements are collected compared to extra-EU trade. In addition, private individual and small- and medium-scale traders are excluded from any statistical reporting obligations. The providers of the statistical information (PSI) are any taxable persons in a Member State carrying out intra-EU trade and being above a certain threshold of annual trade. The National Statistical Authority use the taxable amounts of intra-EU acquisitions and deliveries provided by the National Fiscal Authority to identify the target population and ensure that the PSIs comply with their statistical obligations.

18.4. Data validation

The prime responsibility for ensuring the accuracy of international trade data rests with national authorities. Further checks are carried out by Eurostat, essentially to ensure that the transmission of the requested data has been carried out according to the requirements, that datasets are complete, error-free, and that the correctness of extreme values (outliers) is confirmed.

18.5. Data compilation

EU and euro area aggregates are calculated on the basis of the harmonised figures provided by the Member States according to the Community concept.

 

Seasonally and working days adjusted figures – As with most economic data, interpretation of monthly trade in goods data is made more difficult by regular seasonal fluctuations. For instance, trade in many agricultural products is highly seasonal and month to month fluctuations are often dominated by these seasonal factors. Monthly data are also affected by the varying lengths of months (and in particular the number of working days they contain) and the effects of public holidays whose dates may change from year to year: Easter is the prime example. One partial solution to the effects of seasonality on comparisons over time is to make comparisons only with the same month of the previous year but this is not a completely satisfactory procedure since it cannot provide a proper measure of the fluctuations through a year. To deal with this problem, Eurostat adjusts the trade in goods data provided by Member States for seasonal components and working days. Raw data are processed by JDemetra+, a tool developed by Eurostat. JDemetra+ carries out corrections for working days and applies a method of seasonal adjustment known as SEATS. The parameters and the models are revised once a year. ARIMA and regression coefficients are updated for each monthly release. Under special circumstances, such as a sudden economic slowdown, seasonal factors may be frozen until changes can be properly incorporated into the model. A full technical discussion of the methods available and the criteria for the choice is available on the Eurostat web page dedicated to seasonally adjusted data.

 

Compilation of indices – Data provided by Member States are in current prices, which are prices relevant to the reference period concerned. For several analyses it is satisfactory to work in those values and, if needed, to convert the series of values into an index form following a simple arithmetic operation. However, the development over time of the value of trade flows is determined by both the quantities sold and price variations. For a number of analytical purposes it is then necessary to distinguish between these two elements and in particular to measure movements in the volume of trade estimated in the constant prices of some previous base years. It is therefore desirable to have some measure of how price movements in international trade have developed. Information on values and related quantities is taken from ITGS to estimate a set of deflators which are then used to calculate unit value indices and derive estimates of volumes. Intra‑ and extra‑EU trade statistics are used at their most detailed level — eight-digit CN subheadings by partner country — for calculating indices. The fluctuations in ‘unit values’, which are derived from current price values divided by quantities for each flow, are used as indicators of price variations. Although the calculation system uses weights which are changed annually, for the convenience of users the published indices are based on reference years that are normally changed every five years. The current reference year for the unit value and volume indices is 2015. Data are available monthly as an index (2015=100) showing the percentage change from both the previous month and the same month of the previous year. Unit value and volume indices are calculated for the various aggregates. Volume indices are also adjusted for working days and seasonal variations. Unit value indices are available in euro, national currency for Member States not belonging to the euro area and US dollars. The first compilation step consists in compiling unit-value indices from trade values all expressed in euros. The indices in euro are then converted in national currencies and US dollars.

18.6. Adjustment

No adjustment applied by Eurostat


19. Comment Top

Not applicable


Related metadata Top


Annexes Top


Footnotes Top