## c2-goods-services-and-investment-income-accounts-by-enterprise-characteristics

## Source details

**Canonical URL:** [c2-goods-services-and-investment-income-accounts-by-enterprise-characteristics](https://www.imf.org/-/media/files/data/statistics/bpm6/approved-guidance-notes/c2-goods-services-and-investment-income-accounts-by-enterprise-characteristics.pdf)

## Other formats

- [Markdown version](/-/media/files/data/statistics/bpm6/approved-guidance-notes/c2-goods-services-and-investment-income-accounts-by-enterprise-characteristics.pdf.md)
- [Structured JSON version](/-/media/files/data/statistics/bpm6/approved-guidance-notes/c2-goods-services-and-investment-income-accounts-by-enterprise-characteristics.pdf.json)

---

### Background and motivation
- The IMF Committee on Balance of Payments Statistics endorsed creation of a Working Group on Balance of Payments Statistics Relevant to Global Value Chain (GVC) Analysis at its 2017 meeting to identify balance of payments components relevant for GVC indicators and better identify the role of multinational enterprises (MNEs) in current account transactions.
- The Working Group’s final report (GVC (BOPCOM 19/04)) proposed development of a compilation framework for GVC and a supplementary data collection for MNEs highlighting their role in the current account; the report drew on a stocktaking survey by the OECD and the IMF among 76 member countries.
- UNSD has started global data collection on goods trade by enterprise characteristics (TEC) and services trade by enterprise characteristics (STEC), taking into account existing initiatives by OECD and Eurostat.
- TEC and STEC aim to identify the types of enterprises engaged in cross-border trade to inform policy on growth, employment, and economic interdependency, complementing conventional trade statistics that show flows by goods and services types but not enterprise types.
- Microdata linking is recognized as offering scope to better understand production arrangements in a globalized economy; many countries have added ownership (foreign- or domestically-owned) to TEC and are developing STEC data.

### Quantitative size and existing evidence
- OECD TEC data: large enterprises continue to dominate international trade in goods and leading exporters are often responsible for the majority of imports.
- TEC data: only a small percentage of enterprises are directly engaged in international trade in goods, typically below 10 percent in OECD countries, with exceptions in small economies such as Slovenia and Estonia.
- SMEs generally export to neighboring markets but import from a much wider geographical base.
- Eurostat STEC releases indicate:
  - Domestically controlled enterprises account for around three quarters of services exports from Denmark, Iceland, and Lithuania.
  - Foreign-controlled enterprises are responsible for most services exports from Luxembourg, Hungary, and Ireland.
  - Eurostat publishes STEC figures as experimental statistics.
- Survey evidence cited:
  - Over 75 percent of respondents indicated they are able to identify MNEs in their economies.
  - About a half confirmed capability to break down trade statistics by foreign-controlled and domestic enterprises through data linking (noting trade in goods is more advanced than services).
  - Breaking down direct investment income by ownership appears feasible for many countries though more challenging than trade.

### Current international standards and classification recommendations
- Relevant manuals and guides referenced:
  - UN (2013) International Merchandise Trade Statistics (IMTS): Compilers Manual, Revision 1
  - UN (2016) Manual on Statistics of International Trade in Services 2010 (MSITS) Compiler’s Guide
  - EUROSTAT (2020) Compilers’ Guide on European Statistics on International Trade in Goods by Enterprise Characteristics (TEC)
  - EUROSTAT (2017) Compilers Guide for Statistics on Services Trade by Enterprise Characteristics (STEC)
- Economic activity classification recommendation: International Standard Industrial Classification of all Economic Activities (ISIC Rev. 4) at minimum at 2-digit level; OECD presents TEC based on ISIC Rev. 4.0; Eurostat produces TEC and STEC according to NACE Rev. 2.
- Product classification recommendations:
  - Goods: Central Product Classification 2.1 (CPC 2.1) at 2-digit level as a minimum (alternative: Harmonized System at 2-digit level (Chapter)).
  - Eurostat also uses Classification Products by Activity (CPA), the European version of CPC.
  - Services: EBOPS 2010, at least at the 12 items level (preferably with full disaggregation).
- Ownership concept (Eurostat 2020): refers to control and affiliation—indicates whether an enterprise is domestically or foreign controlled and, if domestically controlled, whether it has affiliates abroad.
  - Proposed ownership breakdown:
    - Domestically controlled
      - Multinational Enterprises (MNEs) (domestically controlled enterprises with own affiliates abroad)
      - Others (domestically controlled enterprises without own affiliates abroad)
    - Controlled from abroad (Foreign controlled enterprises)
    - Unknown
- Enterprise size thresholds (Eurostat 2020):
  - Small and Medium-sized Enterprises: 0 to 249 employees
  - Large Enterprises: 250 or more employees
  - Unknown: trade value of non-identified traders and of estimated trade
- Control definition note: consistent with Direct Investment (DI) statistics (BPM6, paragraph 6.12a)—direct or indirect ownership of 50 percent or more of voting power is evidence of such relationship. The Guidance Note (GN) uses “ownership” and “control” interchangeably consistent with DI definitions.

### Proposed template, framework, and analytical scope
- The GN recommends development of a template introducing further disaggregation of imports and exports of goods and services and receipts/expenditures from investment income as a supplementary annual data collection highlighting enterprises’ role in the current account.
- Proposed breakdowns to examine: industry, product, ownership (foreign-owned/domestically-owned), and size to increase current account relevance for analysis of globalization, including GVCs and MNEs.
- Integration of TEC and STEC into the balance of payments conceptual framework is the first step toward possible future statistics by individual trading partner.
- The framework should apply at least at the total level of exports and imports and for total receipts and expenditures for investment income; partner, product, and industry dimensions are encouraged according to each country’s statistical capabilities.
- Template and framework considerations:
  - Precise definitions of MNE and control (Ultimate Investing Country vs Immediate Investor Country) should be elaborated in line with other Task Teams.
  - Consider further dividing SMEs into independent SMEs and SMEs that are part of a group.
  - Criteria to define "SMEs" and "large" enterprises should be clear and precise.
  - Clarification needed on partner country concept (origin/destination or country of consignment/dispatch) since BPM6 lacks a well-defined concept.
  - Alignment needed with Guidance Note C.4 on merchanting and with CATT’s work.

### Compilation methods, practical issues, and limitations
- TEC compilation methodology:
  - Break down IMTS by characteristics of the trading enterprise via linking microdata from the census of customs transactions with a centralized business register containing enterprise characteristics and reporting structure.
  - Linkage may require the level of individual business establishment for detailed industry or subnational geography.
  - Microdata linkage is facilitated by common identifiers (e.g., a tax identification number) between the trade register and the business register, allowing TEC compilation without additional respondent burden.
- STEC challenges:
  - International trade in services statistics are often compiled from multiple data sources, using models and estimations, unlike goods statistics based on customs data.
  - STEC compilation may be much more challenging for several countries, especially for compiling partner-country distributions for main types of services.
  - Some services cannot be linked directly to an enterprise; non-linkable data issues should be addressed in the GN.
- Practical reporting considerations:
  - The unit reporting customs transactions may differ from the unit producing the product; statistical agencies may need to reallocate trade to the unit most likely responsible for production, requiring substantial effort and resources.
- Compilation challenges and confidentiality:
  - Data availability and granularity at source level, data quality, compilation cost, confidentiality, and reporters’ burden are main impediments to data collection.
  - Accurate enterprise-characteristic information for total exports and imports may not be possible due to thresholds, unit non-response, illegal/informal activities, and transactions based on estimation models (e.g., CIF/FOB adjustment) and impacts on other indicators (e.g., transport and insurance services from CIF/FOB adjustment).
  - If underlying data are not collected/available at enterprise level (e.g., households/government purchases) or cannot be linked to business registers, they should be registered as “Unknown.”
  - Increasing share of goods and services purchased directly by individuals via e-commerce may be classified as “Unknown” unless distinction can be made; if made, it should be noted as an addendum item.
  - Confidentiality constraints may prevent granular disaggregation for some countries; the most economically relevant breakdown possible should be considered for dissemination.

### Key recommendations (paragraph 26) and template summary (Annex I)
- Key recommendations:
  - Include enterprise characteristics as part of the annual reporting of the current account, as supplementary items, using the supplementary framework for the balance of payments building on existing TEC/STEC statistics (EU/OECD) and investment income by enterprise characteristics.
  - Provide in the Manual some general criteria regarding thresholds (on product, industry, size, and ownership classifications) but leave countries to decide on their own thresholds.
  - Provide further clarification around classifications and collections, and include definitions of foreign-owned/domestically owned enterprises in the updated Manual; the initiative may fit appropriately as part of the foreign affiliates trade statistics (FATS).
- Annex I — Proposed template structure (summary):
  - Encouraged supplementary breakdowns by:
    - Trading partner (Top 5 Partners; Rest of the World)
    - Product / Service (Top 5 Products; The Other Products)
    - Industry (Top 5 Industries; The Other Industries)
  - Enterprise characteristics to be reported for goods and services (exports and imports) and receipts/expenditures of investment income:
    - By enterprise's ownership:
      - Domestically controlled
      - MNE
      - Other
      - Controlled from abroad
      - Unknown
    - By enterprise's size:
      - SME
      - Independent
      - Part of a group
      - Large enterprises
      - Unknown
  - Note: "Receipts of investment income" and "Expenditures of investment income" marked "*To be discussed with the Direct Investment Task Team".

### Implementation considerations, coordination, and manuals to update (paragraphs 27–29; Annex III)
- Implementation considerations (paragraph 27):
  - Countries with advanced statistical systems already producing these statistics could incorporate them into annual balance of payments data submission at least at the most aggregated level.
  - The Compilation Guide (CG) accompanying the BPM should offer clear advice and techniques for linking Business Registries with trade and investment income data, with emphasis on linking services trade statistics and investment income data.
  - Staged implementation might be considered.
- Impact on statistical standards and coordination (paragraph 28):
  - The BPM should be updated and the accompanying CG should expand the topic drawing on Eurostat’s work.
  - Coordination should ensure full consistency with updates of the IMTS, MSITS, and the SNA.
  - Annex III provides a comprehensive overview of international manuals that need updating.
- Outcome of the October 2020 IMF Committee on Balance of Payments Statistics meeting (paragraph 29):
  - The Committee supported the GN’s proposal to include enterprise characteristics as part of the annual reporting of the current account, as supplementary items.
  - The Committee agreed the updated Manual could provide general criteria regarding thresholds, but countries should decide their own thresholds.
  - The Committee noted the need for further clarification around classifications and collections, and for including definitions of foreign-owned/domestically owned enterprises in the updated Manual.
  - Following the Committee’s discussion, the CATT liaised with the DITT to further develop details on investment income reporting.
  - The GN has been revised to reflect the discussions and recommendations.
- Annex III — Manuals and guidance to update (summary list):
  - IMF (2009) Balance of Payments and International Investment Position Manual, sixth edition
  - IMF (2009) Balance of Payments and International Investment Position Manual, sixth edition, Compilation Guide
  - UN (2013) International Merchandise Trade Statistics: Compilers Manual, Revision 1
  - UN (2010) Manual on Statistics of International Trade in Services 2010
  - UN (2016) Manual on Statistics of International Trade in Services 2010 Compiler’s Guide
  - EUROSTAT (2020) Compilers’ Guide on European Statistics on International Trade in Goods by Enterprise Characteristics (TEC)
  - EUROSTAT (2017) Compilers Guide for Statistics on Services Trade by Enterprise Characteristics (STEC)
  - OECD (2008) OECD Benchmark Definition of Foreign Direct Investment, fourth edition

*Source: IMF Guidance Note — SECTION I: THE ISSUE; Outcome of the October 2020 IMF Committee on Balance of Payments Statistics meeting; annexes and compilation guidance excerpts.*

### SECTION I: THE ISSUE

### c2-goods-services-and-investment-income-accounts-by-enterprise-characteristics - SECTION I: THE ISSUE

### Background and motivation
- At its 2017 meeting, the IMF Committee on Balance of Payments Statistics (Committee) endorsed creation of a Working Group on Balance of Payments Statistics Relevant to Global Value Chain (GVC) Analysis to identify components in the current balance of payments framework relevant for GVC indicators and to better identify the role of multinational enterprises (MNEs) in current account transactions.
- The Working Group’s final report—GVC (BOPCOM 19/04)—proposed development of a compilation framework for GVC and a supplementary data collection for MNEs highlighting their role in the current account; the report drew on a stocktaking survey by the OECD and the IMF among 76 member countries.
- UNSD has started global data collection on goods trade by enterprise characteristics (TEC) and services trade by enterprise characteristics (STEC), taking into account existing initiatives by OECD and Eurostat.
- Conventional international trade statistics show flows by goods and services types but do not identify the types of enterprises engaged in cross-border trade; TEC and STEC aim to fill that gap to inform policy on growth, employment, and economic interdependency.
- Microdata linking is increasingly recognized as offering scope to better understand production arrangements in a globalized economy; many countries have added ownership (foreign- or domestically-owned) to TEC and are developing STEC data.
- TEC statistics now break down trade in goods by economic activity, size-class of enterprises, ownership, trade concentration, geographical diversification, and products traded to evaluate enterprises’ role in globalization and impacts on employment, production and value added.
- STEC complements TEC given increasing tradability of services (accelerated by digitalization and ICT), providing a fuller picture of traders’ activity across goods and services markets.

### Quantitative size and existing evidence
- OECD TEC data highlight that large enterprises continue to dominate international trade in goods and that leading exporters are often responsible for the majority of imports.
- TEC data show that only a small percentage of enterprises are actually engaged directly in international trade in goods, typically below 10 percent in OECD countries, with exceptions in small economies such as Slovenia and Estonia.
- Data indicate that SMEs generally export to neighboring markets but import from a much wider geographical base.
- Eurostat STEC releases indicate: Domestically controlled enterprises account for around three quarters of services exports from Denmark, Iceland, and Lithuania. Foreign-controlled enterprises are responsible for most services exports from Luxembourg, Hungary, and Ireland. Eurostat publishes STEC figures as experimental statistics.
- Survey evidence cited: Over 75 percent of respondents indicated they are able to identify MNEs in their economies, and about a half confirmed capability to break down trade statistics by foreign-controlled and domestic enterprises through data linking (noting trade in goods is more advanced than services). Breaking down direct investment income by ownership appears feasible for many countries though more challenging than trade.

### Current international standards and classification recommendations
- Relevant existing manuals and guides:
  - UN (2013) International Merchandise Trade Statistics (IMTS): Compilers Manual, Revision 1
  - UN (2016) Manual on Statistics of International Trade in Services 2010 (MSITS) Compiler’s Guide
  - EUROSTAT (2020) Compilers’ Guide on European Statistics on International Trade in Goods by Enterprise Characteristics (TEC)
  - EUROSTAT (2017) Compilers Guide for Statistics on Services Trade by Enterprise Characteristics (STEC)
- Suggested economic activity classification: International Standard Industrial Classification of all Economic Activities (ISIC Rev. 4) at minimum at 2-digit level; OECD presents TEC based on ISIC Rev. 4.0; Eurostat produces TEC and STEC according to NACE Rev. 2 (European version of ISIC Rev.4.0).
- Suggested product classifications:
  - Goods: Central Product Classification 2.1 (CPC 2.1) at 2-digit level as a minimum (alternative possibility: Harmonized System at 2-digit level (Chapter) for goods).
  - Eurostat also disaggregates by Classification Products by Activity (CPA), the European version of CPC.
  - Services: EBOPS 2010, at least at the 12 items level (preferably with full disaggregation).
- Ownership concept: Eurostat (2020) concept of ownership refers to control and affiliation—indicates whether an enterprise is domestically or foreign controlled and, if domestically controlled, whether it has affiliates abroad.
  - Proposed ownership breakdown (based on Eurostat concept):
    - Domestically controlled
      - Multinational Enterprises (MNEs) (domestically controlled enterprises with own affiliates abroad)
      - Others (domestically controlled enterprises without own affiliates abroad)
    - Controlled from abroad (Foreign controlled enterprises)
    - Unknown
- Enterprise size measurement and thresholds (Eurostat 2020):
  - Small and Medium-sized Enterprises: 0 to 249 employees
  - Large Enterprises: 250 or more employees
  - Unknown: trade value of non-identified traders and of estimated trade
- Note on control definition: Consistent with Direct Investment (DI) statistics (BPM6, paragraph 6.12a)—direct or indirect ownership of 50 percent or more of voting power is evidence of such relationship. The GN uses “ownership” and “control” interchangeably consistent with DI definitions.

### Issues for discussion and proposed template
- The Guidance Note (GN) recommends development of a template introducing further disaggregation of imports and exports of goods and services and receipts/expenditures from investment income, summarized as a supplementary table. This would be an annual data collection highlighting enterprises’ role in the current account.
- Proposed breakdowns to examine: industry, product, ownership (foreign-owned/domestically-owned), and size to increase current account relevance for analysis of globalization, including GVCs and MNEs.
- Integration of TEC and STEC into the balance of payments conceptual framework is the first step toward possible future statistics by individual trading partner.
- The framework should apply at least at the total level of exports and imports and for total receipts and expenditures for investment income; partner, product, and industry dimensions are encouraged according to each country’s statistical capabilities.
- Compilation should build on ongoing collections (Eurostat-OECD TEC and STEC) and existing national TEC/STEC statistics based on IMTS and ITSS.
- The GN proposes incorporating TEC and STEC data in trade accounts by introducing disaggregation by nationality and size.
- Template and framework considerations:
  - Precise definitions of MNE and control (Ultimate Investing Country vs Immediate Investor Country) should be elaborated in line with other Task Teams (TTs).
  - Consider further dividing SMEs into independent SMEs (not part of a group) and SMEs that are part of a group to identify SMEs that may benefit from affiliates’ capital inputs.
  - Criteria to define "SMEs" and "large" enterprises should be clear and precise.
  - Clarification needed on partner country concept (origin/destination or country of consignment/dispatch) since BPM6 lacks a well-defined concept.
  - Alignment needed with GN C.4 on merchanting and with CATT’s work.

### Compilation challenges, limitations, and confidentiality
- Data availability and granularity at source level, data quality, compilation cost, confidentiality, and reporters’ burden are main impediments to data collection.
- It is not possible to obtain accurate enterprise-characteristic information for total exports and imports due to thresholds, unit non-response, illegal/informal activities, and transactions based on estimation models (e.g., CIF/FOB adjustment) and impacts on other indicators (e.g., transport and insurance services from CIF/FOB adjustment).
- If underlying data are not collected/available at enterprise level (e.g., households/government purchases) or cannot be linked to business registers, they should be registered as “Unknown.”
- Increasing share of goods and services purchased directly by individuals via e-commerce may be classified as “Unknown” unless distinction can be made; if made, it should be noted as an addendum item.
- Some countries may be unable to include disaggregation due to confidentiality criteria, especially at granular levels; the most economically relevant breakdown possible should be considered for dissemination.

### Outcomes and recommendations
- IMTS and MSITS Compilers’ guides should include clear guidance for compilation of TEC and STEC statistics, drawing on Eurostat’s previous efforts.
- Recommendations should be coordinated with relevant CATT GNs (e.g., C.1 on recording transactor-based components of services and C.4), other Task Teams involved in the BPM (e.g., DITT—D.5), and updates to the 2008 SNA, notably the Advisory Expert Group (AEG) Task Team on Globalization Statistics (GZTT).
- The GN’s proposed framework is largely based on Eurostat’s recommendations and aims to improve trade in value added (TiVA) estimates and international integrated economic accounts.

*Source: IMF Guidance Note — SECTION I: THE ISSUE*

### 26.      Specifically, the GN recommends:

### c2-goods-services-and-investment-income-accounts-by-enterprise-characteristics - 26. Specifically, the GN recommends:

### Key recommendations (paragraph 26)
- To include enterprise characteristics as part of the annual reporting of the current account, as supplementary items, as presented in Annex I. The balance of payments is the overarching framework and could build on existing TEC/STEC statistics (EU/OECD) and investment income by enterprise characteristics to develop this supplementary framework for the balance of payments.
- Regarding the thresholds (on product, industry, size, and ownership classifications), to provide in the Manual some general criteria (as in paragraphs 12–15) but leaving the countries to decide on their own thresholds.
- To provide further clarification around classifications and collections, as well as for including the definitions of foreign-owned/domestically owned enterprises in the updated Manual. This initiative may fit appropriately as part of the foreign affiliates trade statistics (FATS).

### Implementation considerations and practical guidance (paragraph 27)
- Some countries with advanced statistical systems already produce these statistics; inclusion into annual balance of payments data submission should be straightforward for them, at least at the most aggregated level.
- For countries that will face difficulties developing the proposed breakdown, the Compilation Guide (CG) that accompanies the BPM should offer clear advice and techniques for linking the Business Registry with trade and investment income data.
- Particular emphasis should be placed on linking services trade statistics and linking with investment income data.
- Staged implementation might be considered.

### Impact on statistical standards and coordination (paragraph 28)
- Besides updating the corresponding paragraphs in the BPM, the accompanying CG should include and expand the topic drawing on the work undertaken by Eurostat on these topics.
- Coordination should ensure full consistency with updates of the IMTS, MSITS, and the SNA.
- Annex III provides a comprehensive overview of all international manuals that need to be updated.

### Outcome of the October 2020 IMF Committee on Balance of Payments Statistics meeting (paragraph 29)
- The IMF Committee on Balance of Payments Statistics supported the GN’s proposal to include enterprise characteristics as part of the annual reporting of the current account, as supplementary items.
- The Committee agreed that the updated Manual could provide some general criteria regarding the thresholds (on product, industry, size, and ownership classifications), but countries should decide their own thresholds.
- The Committee noted the need for further clarification around classifications and collections, and for including definitions of foreign-owned/domestically owned enterprises in the updated Manual.
- Following the Committee’s discussion, the CATT liaised with the DITT to further develop details on investment income reporting.
- The GN has been revised to reflect the discussions and recommendations.

### Annex I — Proposed template structure (summary)
- Encouraged supplementary breakdowns for the current account items by:
  - Trading partner (Top 5 Partners; Rest of the World)
  - Product / Service (Top 5 Products; The Other Products)
  - Industry (Top 5 Industries; The Other Industries)
- Enterprise characteristics to be reported for goods and services (exports and imports) and receipts/expenditures of investment income:
  - By enterprise's ownership:
    - Domestically controlled
    - MNE
    - Other
    - Controlled from abroad
    - Unknown
  - By enterprise's size:
    - SME
    - Independent
    - Part of a group
    - Large enterprises
    - Unknown
- Note: "Receipts of investment income" and "Expenditures of investment income" marked "*To be discussed with the Direct Investment Task Team".

### Compilation methods and issues (Annex II; COMPILATION METHODS AND ISSUES)
- TEC statistics methodology:
  - Break down IMTS by characteristics of the trading enterprise via linking microdata from the census of customs transactions with a centralized business register containing enterprise characteristics and reporting structure.
  - May require linkage at the level of individual business establishment for detailed industry or subnational geography.
  - Microdata linkage is facilitated by common identifiers (e.g., a tax identification number) between the trade register and the business register, allowing TEC compilation without additional respondent burden.
- STEC challenges:
  - International trade in services statistics are often compiled from multiple data sources, using models and estimations, unlike goods statistics based on customs data.
  - STEC compilation may be much more challenging for several countries, especially for compiling partner-country distributions for main types of services.
  - Some services cannot be linked directly to an enterprise; issues with non-linkable data should be addressed in the GN.
  - Despite challenges, these indicators are considered invaluable for policy given globalization and integration.
- Practical reporting considerations:
  - Inside the same enterprise, the unit reporting customs transactions may differ from the unit producing the product; statistical agencies may need to reallocate trade to the unit most likely responsible for production.
  - This reallocation requires substantial effort and resources and could inform staged implementation decisions.

### Annex III — Manuals and guidance to update (list)
- IMF (2009) Balance of Payments and International Investment Position Manual, sixth edition
- IMF (2009) Balance of Payments and International Investment Position Manual, sixth edition, Compilation Guide
- UN (2013) International Merchandise Trade Statistics: Compilers Manual, Revision 1
- UN (2010) Manual on Statistics of International Trade in Services 2010
- UN (2016) Manual on Statistics of International Trade in Services 2010 Compiler’s Guide
- EUROSTAT (2020) Compilers’ Guide on European Statistics on International Trade in Goods by Enterprise Characteristics (TEC)
- EUROSTAT (2017) Compilers Guide for Statistics on Services Trade by Enterprise Characteristics (STEC)
- OECD (2008) OECD Benchmark Definition of Foreign Direct Investment, fourth edition

---


_Source: https://www.imf.org/-/media/files/data/statistics/bpm6/approved-guidance-notes/c2-goods-services-and-investment-income-accounts-by-enterprise-characteristics.pdf_