This chapter presents the definitions and concepts underlying foreign direct investment statistics. It begins by defining economic territory, residence and institutional units. Then it defines foreign direct investment as establishing a lasting interest by a resident enterprise in one economy (direct investor) in an enterprise (direct investment enterprise) resident in an economy other than that of the direct investor. It explains that the direct or indirect ownership of 10% or more of the voting power of an enterprise resident in one economy by an investor resident in another economy is evidence of such a relationship. Next, it presents the Framework for Direct Investment Relationships used to determine and identify the extent and type of direct investment relationships. The chapter concludes with key accounting principles, including the time of recording of dividends, the valuation of transactions and positions, the debtor/creditor principle, and the asset/liability and directional presentations of FDI statistics.
OECD Benchmark Definition of Foreign Direct Investment (Fifth Edition)
Abstract
22. This chapter provides an overview of statistical units and the framework for direct investment relationships (FDIR). Rigorous application of these fundamental aspects is key to the implementation of the foreign direct investment (FDI) concepts and recommendations described in this Benchmark Definition. It is also matters both for the interpretation and use of FDI statistics. Definitions of statistical units underpinning the concepts and treatment of FDI are completely consistent with the general principles adopted in the 2025 System of National Accounts (2025 SNA, (United Nations et al., Forthcoming[1])) and the Integrated Balance of Payments and International Investment Position Manual, Seventh Edition (BPM7, (IMF, Forthcoming[2])) and, therefore, consistent with the concepts and definitions underlying macroeconomic statistics.
23. The first section, which describes the statistical units, has two main purposes. First, it identifies the economic territory to be covered by FDI statistics and highlights its main features. Second, it describes the economic agents that are involved in direct investment and stresses the importance of the institutional sectors making up the economy.
24. Thereafter, the statistical units that make up the FDI universe are described in more detail. The concept of direct investment, direct investor, direct investment enterprise (DIE) and the relationships between the various units based on the FDIR are explained. This section includes details of the definition and treatment of fellow enterprises.
25. The final section discusses accounting principles for macroeconomic statistics that are particularly relevant for FDI statistics. Specifically, it recommends that market value is the conceptually ideal basis for valuing direct investment transactions and positions and also provides guidance on the calculation of FDI at market value. While this might be relatively straightforward for equity transactions and positions involving companies where the equity securities are listed on an organised stock exchange, it is much less so for unlisted (or unquoted) shares. In this latter case, market value may have to be estimated from the data provided by these unlisted companies. This chapter lists recommended methods by which this can be achieved, as well as listing those methods that are not recommended.
2.2.1. Economic territory and the concept of residence
26. The concept of residence is a fundamental factor for direct investment statistics that measure cross-border investment between residents of two or more economic territories.1 Foreign direct investment includes transactions/positions between a resident and a non-resident institutional unit but excludes all transactions/positions between units that are residents of the same economy.2
27. The concept of residence in the Benchmark Definition is identical to that adopted by the SNA and the BPM. The residence of an economic entity (or an institutional unit) is attributed to the economic territory with which it has the strongest connection, in other words, its centre of predominant economic interest. Each institutional unit is a resident of one and only one economic territory. While some units, mostly households, may have connections with more than one economy, for statistical consistency, there is a need to attribute a single economic territory to each based on objective and comprehensive criteria.
Economic territory
28. The total economy is defined as the entire set of resident institutional units. Most entities have strong links with only one economy, so their residence is clear, but with increasing international economic openness, there is a growing number of institutional units that have connections to more than one economy.
29. Even though there is often close correspondence, the statistical definition of the economic territory (or the economy – see Box 2.1) is not identical to the concept of a country or to any other legal definition (such as nationality). In many cases, but not always, a country indeed constitutes the economic territory. The existence of an institutional unit and the economic territory in which it is resident are determined by a number of criteria including physical presence and being subject to the jurisdiction of the government of the territory.
The economic territory includes the following:
1. Land area, airspace, territorial waters, including jurisdiction over fishing rights and rights to fuel or minerals. In a maritime territory the economic territory includes islands that belong to the territory (BPM7, Chapter 4, Section A.3, paragraph 5.14).
2. Clearly demarcated territorial enclaves in the rest of the world such as embassies, consulates, military bases, scientific stations, information or immigration offices, aid agencies, central bank representative offices with diplomatic immunity, etc. They are established with the formal political agreement of governments of the territories where the land areas are physically located. Such enclaves are used by governments that own or rent them for diplomatic military, scientific or other purposes.
3. Territorial enclaves used by foreign governments and physically located within a territory’s geographical boundaries are not included in that economic territory but included in the territory of the governments that use them.
4. When a territory has a separate physical or legal zone that is under its control, but to which, to some degree, separate laws are applied (e.g., free trade zone or offshore financial centres), these special zones should be included in the economic statistics of the territory.
5. In cases of disputed zones, effective economic control may be unclear. In such cases, compilers make a decision on the inclusion or exclusion of the zone based on the particular circumstances and preferably include a description in the metadata.
6. In addition to economic territories under the effective control of a single government, the economic territory of a currency or economic union is also considered as a type of economic territory.
Note: The importance of economic territories described under (3) and (4) relates to the fact that they reflect the usual scope of macroeconomic policymaking.
Source: IMF (Forthcoming[2]), Integrated Balance of Payments and International Investment Position Manual, Seventh Edition (BPM7).
30. The most commonly used concept of economic territory is the area under the effective economic control of a single government. However, economic territory may be larger or smaller than this, as in a currency or economic union, which is discussed at more length in Annex 8.C.
Centre of predominant economic interest
31. Economic interests include current production, consumption, acquisition of assets and incurrence of liabilities, asset-holding, place of incorporation or registration, and the origin of applicable taxation and regulation.3
32. An institutional unit is considered to have a centre of predominant economic interest in an economic territory:
1. when there exists, within the economic territory, some location, dwelling, place of production, or other premises on which or from which the unit engages and intends to continue engaging, either indefinitely or over a finite but long period of time, in economic activities and transactions on a significant scale
2. if the location, even if it is not fixed, remains within the economic territory
3. if the unit has already engaged in economic activities and transactions on a significant scale in the territory for one year or more, or if the unit intends to do so. The choice of one year as a specific period is somewhat arbitrary and is used as an operational guideline to avoid uncertainty and to facilitate international statistical consistency.
33. While actual or intended physical presence for a year or more is the main criterion, other criteria apply in some special cases such as for corporations having little or no physical presence or individuals that move in such a way that that they do not stay in any territory for more than a year.
34. There are cases where the institutional unit has little or no physical presence in a territory. This arises with some categories of special purpose entities (SPEs) and similar units (see Section 6.2). Some cases of restructuring or outsourcing could also result in residual units with little or no physical presence. Moreover, these units might not undertake significant production. In such cases, the jurisdiction where the unit has its legal domicile and/or which regulates its activities is considered to be that unit’s predominant centre of economic interest while location and production criteria may not be meaningful.
2.2.2. Institutional units
35. An institutional unit is an economic entity that is capable, in its own right, of owning assets, typically able to incur liabilities, and engaging in economic activities and in transactions with other entities (2025 SNA, Chapter 5, Section A, paragraph 5.2).
36. Institutional units of an economy are classified by the SNA under two main categories: (i) households and (ii) legal and social entities (see Box 2.2). While the Benchmark Definition follows the general principles of the SNA to identify institutional units, the treatment of some specific issues concerning cross-border direct investment transactions/positions also need to be addressed.
37. Institutional units are defined by the SNA in the context of a single economy and to identify the residents of that economy. However, a legal entity may have to be split into separate institutional units that are resident in different economies to address the statistical treatment of entities that have strong links with more than one economy. As a result, some arrangements that are not legal entities in their own right may be recognised as being institutional units, i.e., quasi-corporations.
38. A quasi-corporation is an unincorporated business that operates as if it was a unit separate from its owner(s). Examples are branches, land ownership, partnerships (both of limited and unlimited liability), trusts, and resident portions of multi-territory enterprises. These quasi-corporations are treated as if they were corporations, i.e., as separate institutional units from the units to which they legally belong. For example, quasi-corporations owned by households or government units are grouped with corporations in the non-financial or financial corporate sectors. The purpose of this treatment is to separate from their owners those unincorporated enterprises that are sufficiently self-contained and independent, and hence they behave in the same way as corporations (2025 SNA, Chapter 5, Section B, paragraph 5.51).
39. The requirement for having (or the potential to produce) a set of accounts is essential in the definition of institutional units because it provides a recognised indication of the existence of an economic decision-making unit. The definitions do not require that the unit be autonomous. Therefore, wholly owned subsidiaries have legal identity and are separate institutional units from their parent units, even when all decisions are effectively made by another unit.
The definition of institutional units covered in this Benchmark Definition is fully in line with the concept of the SNA (1993 and subsequent updates) whereby an institutional unit is an economic entity that is capable, in its own right, of owning assets, typically able to incur liabilities, and engaging in economic activities and in transactions with other entities.
The main attributes of an institutional unit are:
it is entitled to own goods or assets in its own right; it is, therefore, able to exchange the ownership of goods or assets through transactions with other institutional units
it is able to take economic decisions and engage in economic activities for which it is itself held to be directly responsible and accountable at law
it is able to incur liabilities on its own behalf, to take on other obligations or future commitments and to enter into contracts
either a complete set of accounts, including a balance sheet, exists for the unit, or it would be possible and meaningful, from an economic viewpoint, to compile a complete set of accounts if they were to be required.
As classified by the SNA, there are two main types of institutional units:
1. Households are defined in the SNA as a single person or a group of persons who share the same living accommodation, who pool some, or all, of their income and wealth and who consume certain types of goods and services collectively, mainly housing and food. A person who pools income with the household in one economic territory, but is resident of another economic territory, is not classified as a member of that household. Households can be direct investors but not direct investment enterprises (DIEs).
2. Legal and social entities are recognised by law or society independently of the natural persons, or other units, that may own or control them. Such entities include corporations, non-profit institutions and government units. Some unincorporated enterprises belonging to households or government units may behave in much the same way as corporations, and such enterprises are treated as quasi-corporations when they have complete sets of accounts, or such a set of accounts can be constructed. Only business enterprises can be both direct investors and DIEs; governments and non-profit enterprises can be direct investors but not DIEs.
Note: Only a summary of SNA and BPM concepts is included in this section. For a more detailed discussion of the items therein, the reader should refer to SNA and BPM.
Source: Based on United Nations et al. (Forthcoming[1]), 2025 System of National Accounts.
40. An ancillary corporation is a wholly-owned subsidiary whose productive activities are confined to providing services to the parent corporation and/or other affiliates owned by the same parent corporation. The kinds of services which may be produced by an ancillary unit are transportation, purchasing, sales and marketing, various kinds of financial or business services, computing and communications, security, maintenance, and cleaning. In some cases, the ancillary unit is located in a different economy from the companies it serves. An ancillary corporation is recognised as a separate institutional unit when it is resident in a different economy from that of any of its owners, even if it is not, in practice, autonomous.
Enterprises
41. An enterprise is a sub-group of the broad SNA category legal and social entities. It is defined as an institutional unit engaged in production of goods and/or services. An enterprise may be a corporation (including quasi-corporations), a non-profit institution, or an unincorporated enterprise. Incorporated enterprises and non-profit institutions are complete institutional units. An unincorporated enterprise, however, refers to an institutional unit – a household or government unit - only in its capacity as a producer of goods and services. It covers only those activities of the unit that are directed towards the production of goods or services.
42. Institutional units that have the sole function of holding financial assets on behalf of their owners provide a service to their owners, so they are enterprises, even though there may be no explicit service charges. For example, some mutual funds, holding companies, trusts, and special purpose entities may not receive payment from their owners for the services they provide. However, in these cases, a service is recognised as being provided to the owner, payable out of the property income or assets. Box 2.3 presents some structures related to enterprises.
Quasi-corporations
43. An unincorporated enterprise is a producer unit that is not a legal entity separate from the owner (household, government or foreign resident); the fixed and other assets used in unincorporated enterprises do not belong to the enterprises but to their owners, the enterprises as such cannot engage in transactions with other economic units nor can they enter into contractual relationships with other units nor incur liabilities on their own behalf. In addition, their owners are personally liable, often without limit,4 for any debts or obligations incurred in the course of production (2025 SNA, Chapter 5, Section B, paragraphs 5.51-5.56). An unincorporated enterprise attached to a non-resident is always considered as a quasi-corporation.
44. Even though a branch is not a separate legal entity, it is statistically identified as a quasi-corporation if it operates in an economy separate from that of its owner and undertakes production of goods and/or services on a significant scale in that economic territory.5
45. To avoid the multiplication of artificial units, the definition of branches requires several indicators of substantial economic activity and separate accounts.
46. The identification of branches has implications for the statistical reporting of both the branch and the legal entity from which the branch has been created. The operations of the branch should be excluded from the institutional unit of its head office in its home territory and its activities should be reported consistently in both of the economies to which it is connected. Each branch is a DIE if the FDI criteria are satisfied.
47. Construction of major specific projects (bridges, dams, power stations, etc.) may take several years to complete and be carried out and managed by non-resident enterprises through site offices in the territory where the project is being executed. In most instances, the operations managed from the site office satisfy the criteria to be a branch. In other cases, for example, for a short-term project (less than one year) or a project managed from the territory of the non-resident provider rather than a local office, the work provided to customers resident in the territory of those operations is classified as international trade in construction (i.e., an import of construction services by the territory of operations).
48. Some production processes may involve mobile equipment that operates in more than one economic territory, such as ships, aircraft, drilling rigs and platforms, and railway rolling stock. Some of these operations may take place outside any territory, such as ships operating on the high seas. Moreover, services such as consulting, maintenance, training, technical assistance and healthcare can also be delivered on-site by a branch or from a home base. The criteria for recognition of a branch apply in these cases. For example, the existence of a branch would be recognised if a shipping company has a secondary base for servicing its fleet that is substantial, permanent, and has its own accounts. However, in many instances, the activity can be seen as having been undertaken from a home base of operations, so that the operations are attributed to that unit, and the recognition of an additional unit for non-resident operations is not appropriate. In this case, the operation will give rise to international trade in services.
An enterprise group consists of all the enterprises under the control1 of the same owner or groups of owners. When a single owner or group of owners has control of more than one enterprise, the enterprises may act in a concerted way and the transactions between them may not be driven by the same concerns as “arm’s length” transactions. The framework for direct investment relationships can be used to determine which enterprises are under control or influence of the same owner.
There are two concepts of enterprise groups:
A multinational enterprise group consists of all the enterprises – regardless of their economies of residence – under the control of the same ultimate controlling parent.
A local (i.e., economy-specific) enterprise group consists of the investor and the legal entities controlled or influenced by that investor that are resident in the same economy. Ownership links that are external to the economy are not recognised in the formation of local enterprise groups.
A joint venture involves the establishment of a corporation, partnership or other institutional unit in which each party legally has joint control over activities of the unit. The units operate in the same way as other units except that a legal agreement between two or more parties establishes joint control over the unit. As an institutional unit, the joint venture may enter into contracts in its own name and raise financing for its own purposes. A joint venture maintains its own accounting records. Joint ventures are typically established for the purpose of executing a business undertaking in which the parties agree to share the profits and losses of the enterprise as well as the capital formation and contribution of operating inputs or costs. It is similar to a partnership (see Glossary), but typically differs in that there is generally no intention of a continuing relationship beyond the original purpose.
A joint venture may not involve the creation of a new legal entity. Whether a quasi-corporation is identified for the joint venture without a separate legal status depends on the arrangements of the parties and legal requirements. The joint venture is a quasi-corporation if it meets the requirements for an institutional unit, particularly by having its own records. Otherwise, if each of the operations is effectively undertaken by the partners individually, then the joint venture is not an institutional unit, and the operations would be seen as being undertaken by the joint venture partners separately (2025 SNA, Chapter 5). Because of the ambiguous status of joint ventures, there is a risk that they could be omitted or double-counted, so particular attention needs to be paid to them.
Note: 1. Control of a corporation refers to the ability to determine general corporate policy of a corporation, where general corporate policy is understood in a broad sense to mean the key financial and operating policies relating to the corporation’s strategic objectives as a market producer. In practice, control is determined to exist if an investor has more than 50 percent of the voting power in an enterprise. The control may be direct (through ownership of voting power or other arrangements) or indirect (through ownership of enterprises that in turn have voting power).
Source: Based on OECD (2009[3]), OECD Benchmark Definition of Foreign Direct Investment, Fourth Edition (BD4), https://doi.org/10.1787/9789264045743-en; and United Nations et al. (Forthcoming[1]), 2025 System of National Accounts.
49. Similarly to mobile equipment, a multi-territory pipeline that passes through a territory, but is not operated by a separate legal entity in that territory would be recognised as constituting a branch or not, depending on whether there is substantial presence, availability of separate accounts, etc. Where such operations are not separate entities, there may be a multi-territory enterprise as described below.
50. When a branch is identified, there are direct investment flows to and from the territory of its location, but the provision of goods or services to customers in that territory are resident-to-resident transactions. In contrast, if the operations are not substantial enough to qualify as a branch, the provision of goods or services to customers in that territory are imports of that territory.
51. Prior to its incorporation, a resident enterprise is imputed when preliminary expenses, such as those associated with mining rights, license fees, local office expenses and legal costs, are first incurred as steps prior to establishing a legal entity. As a result of identifying a quasi-corporation, the preparatory expenses are recorded in the economy of the future operations as being resident-to-resident transactions that are funded by a foreign direct investment inflow, rather than as the sale of non-produced assets to non-residents, exports of legal services, etc. Because of the limited scale of these activities, the assembly of data for these enterprises is feasible prior to incorporation. If the project does not subsequently go into operation, the value of the direct investment is eliminated by an entry for other changes in volume of assets and liabilities.
52. A resident notional unit (a kind of a quasi-corporation) is identified for statistical purposes for direct non-resident ownership of immovable assets such as land and buildings and other structures. The notional resident unit is recorded as owning the land, buildings and/or other structures and receiving the rent or rentals that accrue to that asset. The non-resident owner holds equity in the notional resident unit and then receives income from the notional resident unit in the form of property income. This treatment is designed so that the relevant non-financial assets are always assets of the economy in whose territory they are located; otherwise, the land or buildings would appear in another economy’s national balance sheet.
53. Consistent with this treatment, a notional unit is imputed for the ownership by non-residents of natural resources such as subsoil assets, non-cultivated biological resources, and water, and long-term rights to use these assets. The notional unit is nearly always a DIE.6 However, it is usually the case that ownership of land and other natural resources and rights to use these assets through a lease or permit over long periods of time are associated with a branch.
54. A multi-territory enterprise operates as a seamless operation over more than one economic territory. Such an enterprise, even though it has substantial activity in more than one economic territory, cannot be separated into a parent and branch(es) because it is run as a seamless operation and cannot supply separate accounts for each territory. Multi-territory enterprises are typically involved in cross-border activities and include shipping lines, airlines, hydroelectric schemes on border rivers, pipelines, bridges, tunnels, and undersea cables.7 Some non-profit institutions serving households (NPISH) may also operate in this way.
55. It is preferable that a parent and branch(es) be identified separately in the case of a multi-territory enterprise. If possible, enterprises should be identified in each territory according to the principles for identification of branches. If that is not feasible because the operation is so seamless that separate accounts could not be developed, it is necessary to prorate the total operations of the enterprise into the individual economic territories. The factor used for prorating should be based on available information that reflects the contributions to actual operations. The prorating of the enterprise implies that every transaction needs to be split into each component economic territory, a process which may be difficult to implement by compilers. Hence, for the economy of residence, each apparently domestic transaction would be split into resident and non-resident components. Equally, non-resident entities of those economies outside the territories of the multi-territory enterprises that have transactions and positions with such entities need to make the same split, so as to capture the counterparty claims in a consistent manner. Bilateral agreements between compilers will help to minimise possible asymmetries. This treatment has implications for other macroeconomic statistics and its implementation should always be coordinated with other statistical interests for consistency. Compilers in each of the territories involved are encouraged to cooperate in order to develop consistent data, avoid gaps, and minimise respondent and compilation burden.
56. Analogous treatments can be applied to enterprises operating in zones of joint jurisdiction – the enterprise will need to be split into entities that are resident in each economy having jurisdiction over the zone, with flows into and out of the enterprise prorated between these entities. Positions and flows between the entities may also need to be developed.
57. Estates, other trusts, and partnerships are treated as separate institutional units if they are constituted in a different territory from that of their owners or beneficiaries.
58. Special corporate structures, such as those with little or no physical presence, holding companies, and nominees, are discussed in more detail in Chapter 6.
2.2.3. Institutional sectors
59. BPM7 recommends the standard presentation of direct investment in the balance of payments (BOP) and international investment position (IIP) be broken down by resident institutional sectors. The Benchmark Definition does not require as part of the standard data components a classification of direct investment statistics by institutional sectors: the emphasis is given instead to industry classification because it enables more in-depth analysis of globalisation and the impact of FDI on host economies than the sectoral breakdown. The institutional sectors of the SNA group together similar kinds of institutional units because the economic objectives, functions, and behaviours of corporations, non-profit institutions, government units and households are intrinsically different from each other. Classification principles for resident and non-resident institutional sectors are based on the nature of the economic activity undertaken by the institutional unit, which are production of goods and services, consumption to satisfy human wants, and accumulation of various forms of capital.
60. All residential institutional units are allocated to one and only one of the following institutional sectors: non-financial corporations, financial corporations, general government, non-profit institutions serving households (NPISH), and households. Any of the five sectors identified by the SNA to cover all the activities of an economic territory may include entities that are direct investors, because any resident unit may own or control a non-resident unit that qualifies as a DIE. On the other hand, all DIEs are classified either in the non-financial corporation sector or the financial corporation sector. This is because governments and households cannot be owned by non-resident units. Also, although NPISH can be established or owned by non-residents, financial transactions/positions involving NPISH are generally not driven by investment considerations and, therefore, they are generally not regarded as DIEs unless they qualify under the direct investment criteria.
61. The SNA also recommends subsectors of the institutional sectors to provide information on particular groups of units for policymaking purposes. Annex 2.C provides further information on institutional sectors and, in particular, the subsectors of the financial corporation sector.
62. For the BOP and IIP, the sectoral breakdown reflects the sectors and subsectors most relevant for cross-border transactions and positions and includes the following sectors:
Central bank
Deposit-taking corporations, except the central bank
General government
Other financial corporations
Non-financial corporations
Households and NPISH.
Notes: Items in italics are supplementary. n.a. Not applicable.
63. Table 2.1 presents the standard breakdown of direct investment financial transactions and income in the BOP and positions in the IIP. It includes the standard components, which are items that are fully part of the framework and contribute to the totals and balancing items; memorandum items, which are part of the standard presentation, but are not used in deriving totals and balancing items; and supplementary items, which are outside of the standard presentation but are compiled depending on circumstances in the particular economy. The standard presentation recommended in the BPM6, which was used to convert the asset/liability presentation to the directional presentation, is now a memorandum item (see Section 2.5.4 and Annex 2.B for additional information on the asset/liability and directional presentations of direct investment). For this conversion, a number of details on reverse investment and investment between fellows are needed. Regardless of the presentation used, these details are also helpful in shedding light on the general FDI trends of a given country. Therefore, both BPM7 and the Benchmark Definition continue to recommend that countries compile the details necessary to convert the aggregate FDI statistics on an asset/liability basis to the directional presentation. There is value in presenting aggregate FDI statistics on a directional basis as this facilitates the analysis of the most recent FDI trends, and, given that BOP FDI aggregate statistics are produced on a quarterly basis, it allows for the timely analysis of the nature and motivation of FDI, which is particularly beneficial where there are significant differences between the two presentations.
64. For FDI statistics, it can be useful to collect and compile data at the local enterprise group level; for example, if direct investment is initially channelled to a holding company and then to a manufacturing subsidiary, then it may clarify to classify the direct investment in manufacturing rather than a holding company operation. However, the implications of combining entities in different institutional sectors need to be carefully considered. In particular, it is important to collect and compile information for those parts of the local enterprise group that are classified to the financial intermediaries subsectors separately from those parts that are not classified to the financial intermediaries subsectors, not only for correct attribution by sector but also from the treatment of debt between affiliated financial intermediaries discussed in Annex 6.A. Box 2.3 provides further information for compilers on structures related to enterprises.
65. Foreign direct investment reflects the objective of establishing a lasting interest by a resident enterprise in one economy (direct investor) in an enterprise (direct investment enterprise) that is resident in an economy other than that of the direct investor. The lasting interest implies the existence of a long-term relationship between the direct investor and the direct investment enterprise and a significant degree of influence on the management of the enterprise. The direct or indirect ownership of 10% or more of the voting power8 of an enterprise resident in one economy by an investor resident in another economy is evidence of such a relationship. Some compilers may argue that in some cases an ownership of as little as 10% of the voting power may not lead to the exercise of any significant influence while on the other hand, an investor may own less than 10% but have an effective voice in the management. Nevertheless, the recommended methodology does not allow any qualification of the 10% threshold and recommends its strict application to ensure statistical consistency across countries.
66. The 10% threshold represents an investment of substantial magnitude that is considered sufficient evidence of the intended level of influence to distinguish between direct and portfolio investment. It also implies a long-term investment as it can be difficult to sell quickly. While financial accounting standards generally use a higher threshold of 20% to identify related parties, BPM7 and the Benchmark Definition retained the existing 10% threshold for several reasons. First, there are many ways in which economic accounting standards differ from financial accounting standards, such as what is considered capital or how to measure depreciation. There are other domains that do not follow financial accounting standards to identify influence relationships; for example, stock exchanges may have rules requiring shareholders to identify when they have a holding large enough to exert a certain level of influence, and those thresholds are often lower than 20%. Finally, there is little evidence that the 10% threshold is too difficult for compilers to implement, while changing the threshold would create breaks in time series and disruptions to compilation systems.
67. Direct investment includes the initial equity transaction that meets the 10% threshold and all subsequent financial transactions and positions between the direct investor and the DIE, as well as qualifying FDI financial transactions and positions between incorporated and unincorporated fellow enterprises included under the FDIR (see Section 2.4). Direct investment is not solely limited to equity investment but also relates to reinvested earnings and inter-company debt (see Chapter 3).
68. Direct investment includes inward and outward financial transactions/positions between directly and indirectly owned incorporated and unincorporated enterprises.9 The extent of the direct investment relationship is determined according to the framework for direct investment relationships (see Section 2.4.3).
69. Some relationships may exist between enterprises that may exhibit the characteristics of direct investment even though there are no links that qualify as direct investment. Such borderline cases should not be treated as direct investment (see Annex 6.A).
70. The classification of financial positions and transactions to direct investment requires that the two institutional units directly involved be resident in different economies and that they be in a direct investment relationship. This section presents the framework that has been developed to establish when two or more institutional units are in a direct investment relationship. While direct investment measures a wide variety of instruments as described in Chapter 3, direct investment relationships are determined using the voting power criterion only.
2.4.1. Foreign direct investor
71. A foreign direct investor is an entity (an institutional unit) resident in one economy that has acquired, either directly or indirectly, at least 10% of the voting power of a corporation (enterprise), or equivalent for an unincorporated enterprise, resident in another economy. A direct investor could be classified to any sector of the economy and could be any of the following:
an individual
a group of related individuals
an incorporated or unincorporated enterprise
a public or private enterprise
a group of related enterprises
a government body
an estate, trust or other societal organisation
any combination of the above.
72. For two or more individuals or other entities to be considered a combination, and thus be regarded as a single direct investor, they must be in a direct investment relationship or have a family relationship (in the case of individuals). The different individuals or other entities must be resident in the same economy as each other. They cannot include any investor that is a resident of the same economy as the DIE. Equity ownership in an enterprise held by a group of related investors acting in combination can be summed to establish either control or influence.
73. In the case where two enterprises each own 10% or more of each other’s voting power, each is a direct investor in the other.
2.4.2. Foreign direct investment enterprise
74. A direct investment enterprise is an enterprise resident in one economy and in which an investor resident in another economy owns, either directly or indirectly, 10% or more of its voting power if it is incorporated or the equivalent for an unincorporated enterprise.
75. The numerical threshold of ownership of 10% of the voting power determines the existence of a direct investment relationship between the direct investor and the direct investment enterprise. An ownership of at least 10% of the voting power of the enterprise is regarded as the necessary evidence that the investor has sufficient influence to have an effective voice in its management. In contrast to some other statistical measures such as those on the activities of multinational enterprises (AMNE), direct investment does not require control by the investor (i.e., more than 50% owned by the investor and/or its related enterprises). Direct investors may have direct investment enterprises in one economy or in several economies.
76. To facilitate international comparisons and to achieve global consistency of FDI statistics, the Benchmark Definition recommends a strict application of the 10% rule. Therefore, compilers should not qualify the 10% threshold further by applying other criteria. The Benchmark Definition does not recommend the use of other considerations such as representation on the board of directors; participation in policy-making processes; material inter-company transactions; interchange of managerial personnel; provision of technical information; and provision of long-term loans at lower than existing market rates.
2.4.3. Framework for direct investment relationships (FDIR)
77. The definitions of direct investors and DIEs above highlight the immediate relationship between two enterprises. As noted earlier, the legal structures of related enterprises can consist of many enterprises linked through complex ownership chains.
78. The FDIR is a generalised methodology for identifying and determining the extent and type of direct investment relationships. In other words, the FDIR allows compilers to determine the population of direct investors and DIEs to be included in FDI statistics.
79. For a compiling economy, the FDIR identifies all enterprises related to a particular enterprise whether it is a direct investor or a DIE or both. For example, within a group, it is possible that a DIE itself owns 10% or more of the voting power of another non-resident enterprise, in which case the DIE is itself a direct investor in a further DIE. The question is therefore whether there is a direct investment relationship between the further enterprise and the original enterprise.
80. In Figure 2.1, enterprises A, B and C are in different economies. Enterprise A owns 80% of the voting power in enterprise B and is a direct investor in B. Enterprise B, in turn, owns 80% of the voting power in enterprise C and is a direct investor in C. A has control over B, and through its control over B, has indirect control over C. As a result, financial transactions between A and C are also relevant to FDI even though A directly holds no equity in C, and A and C are therefore considered to be in the direct investment relationship that embraces A, B and C. Financial transactions and positions as well as associated income transactions between A, B and C should be included in direct investment statistics.
81. In relatively simple cases such as that in Figure 2.1, where each link in the ownership chain is a single equity holding and there is majority ownership (control) at each stage, it is clear that the direct investment relationship continues down the chain of ownership. However, when some links are non-controlling links and voting power of an enterprise is held by more than one member of a direct investment relationship, the extent of the relationship may be less obvious.
Source: OECD (2009[3]), OECD Benchmark Definition of Foreign Direct Investment, Fourth Edition (BD4), https://doi.org/10.1787/9789264045743-en.
82. It is not uncommon for an entity to be a direct investor in more than one DIE. In Figure 2.2, enterprises A, B and C are each in different economies. Enterprise A owns 80% of the voting power in enterprise B and is a direct investor in B. Enterprise A also owns 20% of the voting power in enterprise C and is a direct investor in C. Enterprise A controls B and has significant influence over C. As a result, financial transactions and positions between B and C are also relevant to FDI statistics even though there is no equity participation between them. For example, B may raise capital that it then lends to C at a concessional rate due to the control by A. It is reasonable to consider A, B and C to be in the same direct investment relationship in which B and C are considered ‘fellow enterprises’ of one another.
Source: OECD (2009[3]), OECD Benchmark Definition of Foreign Direct Investment, Fourth Edition (BD4), https://doi.org/10.1787/9789264045743-en.
83. In Figure 2.3, there are two overlapping direct investment relationships, one with enterprise A as the direct investor and the other with enterprise B as the direct investor. The DIE, C, is in a direct investment relationship with both A and B. Enterprise C is controlled by direct investor A, which owns 70% of the voting power of C, and C is significantly influenced by direct investor B, which owns 20% of the voting power of C. Despite their common ownership of C, enterprises A and B are not in a direct investment relationship with each other.
Source: OECD (2009[3]), OECD Benchmark Definition of Foreign Direct Investment, Fourth Edition (BD4), https://doi.org/10.1787/9789264045743-en.
84. The FDIR aims to identify all enterprises over which the investor has significant influence under the 10% voting power criterion for FDI. In this determination, it is necessary to establish whether each enterprise under consideration is a subsidiary, an associate, or is not relevant in FDI – all three categories when combined being exhaustive and individually mutually exclusive. Those enterprises that are subsidiaries or associates are included in the direct investment relationship, while those categorised as not influenced are not included.
Subsidiaries in FDI
85. Subsidiaries in FDI are described as follows:
1. A subsidiary is an enterprise in which an investor owns more than 50% of its voting power, i.e., it is controlled by the investor.
2. Where an investor and its subsidiaries combined own more than 50% of the voting power of another enterprise, this enterprise is also regarded as a subsidiary of the investor for FDI purposes.
3. In determining the scope of a direct investment relationship, the degree of influence that may be exercised through controlling links (more than 50% of voting power) is not diminished by the existence of multiple links in an ownership chain.
a. An enterprise controlled by a subsidiary of an investor or by a group of subsidiaries (which may also include the investor) is itself regarded as a subsidiary for FDI purposes.
b. However, for clarification, it should be stressed that an enterprise controlled by an associate or by any group including an associate is regarded as an associate in the context of FDI.
Associates in FDI
86. Associates in FDI are described as follows:
1. An associate is an enterprise in which an investor owns directly at least 10% of the voting power and no more than 50%.
2. Where an investor and its subsidiaries combined own at least 10% of the voting power of an enterprise but no more than 50%, the enterprise is regarded as an associate of the investor for FDI purposes.
3. Where an associate, either as an individual or in combination with its subsidiaries, own more than 50% of an enterprise, this enterprise is regarded for FDI purposes as an associate of the higher-level investor.
4. In determining the scope of a direct investment relationship involving associates, the degree of influence that may be exercised through a single or cumulative influencing link (from 10% to 50%) along an ownership chain is diminished by one degree. Thus, an enterprise that is an associate of a subsidiary - or of a group of the investor’s subsidiaries (which may include the investor) - is regarded as an associate of the investor for FDI purposes.
Not relevant in FDI
87. The following are not relevant in FDI:
1. An enterprise in which an investor owns less than 10% of its voting power is regarded as not influenced by the investor and is therefore out of the scope of the FDIR. However, it should be stressed that a particular investor in a chain of ownership within the FDIR may indirectly hold less than 10% (but more than 0%) of an enterprise’s voting power; this enterprise should be included in the FDIR as a subsidiary or an associate of the investor if the relevant criteria described earlier are fulfilled; that is, influence is maintained through a chain of influence and control as in the description of associates in the preceding section.
2. It should be also noted that an enterprise that is an associate of an enterprise associated to an investor is not influenced by this investor in the context of FDI, i.e., it is not regarded as an associate of the investor within the FDIR.
Fellow enterprises in FDI
88. The coverage of subsidiaries and associates within the FDIR along what might be termed ‘vertical’ FDI ownership chains is generally relatively straightforward. However, an enterprise in one economy may be related through the FDIR to another enterprise in the same economy, or in a different economy, without either being a direct investor in the other, but through both being directly or indirectly influenced by the same enterprise in the ownership hierarchy. This 'common parent' must be a direct investor in at least one of the enterprises in question. Such enterprises can be considered to be related through a ‘horizontal’ linkage within the FDIR – not involving FDI equity of 10% or more – and are called fellow enterprises. It should be noted, however, that for FDI statistics, only cross-border transactions and positions between FDI-related enterprises should be recorded. Where there is no voting power acquired in a fellow enterprise, or where less than 10% of the enterprise’s voting power is obtained, all the investment made by one fellow enterprise in one economy in its fellow in another economy is included in FDI statistics but needs to be distinguished in compiling the data to facilitate the preparation of FDI data on both the asset/liability and directional presentational bases referred to in Section 2.5.4. Therefore, for a particular compiler, fellow enterprises may be considered as those resident and non-resident enterprises related within the FDIR other than those that have a direct investor-direct investment enterprise relationship (i.e., under the “10% or more” voting power criterion) with one another.
89. It is likely that many fellow enterprises will be DIEs in their own right but, clearly, some will not. For example, if a resident direct investor, A, has a DIE abroad, B, as well as a resident subsidiary, C, then C, by definition, is not a DIE of A, but B and C are fellow enterprises within the FDIR covering A, B and C.
90. The direct investment relationship extends in both directions along an ownership chain from an enterprise to cover all of its subsidiaries and associates and all of the investors within the FDIR for which the enterprise is a subsidiary or an associate. It also covers ‘across chain’ relationships as described below and in Figure 2.4 (see also Annex 2.A).
91. It should be noted that identifying the extent of a direct investment relationship from an investor, B, which is an associate of another investor, A, may identify enterprises in a direct investment relationship with enterprise B that are not in a direct investment relationship with enterprise A. Determining the existence of one or more direct investment relationships under the FDIR depends largely on whether a common parent exists for enterprises in different ownership chains.
92. It must be stressed that direct investment is only recorded when there is a financial transaction or position between entities in different economies that are in a direct investment relationship (including fellow enterprises). However, it should also be noted that the residence of units is not a feature of the definition of subsidiaries and associates for FDI purposes. The FDIR may include within the relationship enterprises that are resident in the same economy. For example, a domestic subsidiary of a direct investor is a fellow enterprise of the non-resident DIEs of that direct investor, and any transactions and positions between them are relevant to FDI.
93. Figure 2.4 also highlights that equity positions within one economy may be important in determining the extent of a direct investment relationship. Despite being resident-to-resident positions, the equity positions between enterprises B and E and between enterprises C and F are part of the equity determining the relationship between enterprises A and E and between enterprises A and F.
Source: OECD (2009[3]), OECD Benchmark Definition of Foreign Direct Investment, Fourth Edition (BD4), https://doi.org/10.1787/9789264045743-en.
94. Box 2.4 summarises the principles and norms that determine the extent of a direct investment relationship.
95. Recognising practical difficulties compilers may encounter in fully applying the FDIR, two alternative methods may be applied: the “participation multiplication” method, and the “direct influence/indirect control” method. Should compilers choose to apply either of these alternative methods due to practical difficulties in implementing the FDIR, they should include this information in their metadata. However, such countries should endeavour to move towards applying the FDIR over time.
96. Annex 2.A provides more information and further examples concerning the determination of the extent of the direct investment relationship under the FDIR along with the two alternative methods. It also describes the practical implementation of the FDIR for measuring investment income transactions, and for classifying financial account transactions and international investment positions.
Basic types of enterprises:
A subsidiary is an enterprise in which the investor has control of more than 50% of the voting power.
An associate is an enterprise in which the investor has control of at least 10% of the voting power and no more than 50%.
Fellow enterprises are enterprises which do not have enough (or any) voting power in each other to constitute FDI influence but have a common parent.
Principles for extending the relationship through indirect ownership:
A series of subsidiaries can continue as long as control exists at each stage in the ownership chain – a chain such as that in Figure 2.1 can continue indefinitely.
Any subsidiary can extend the FDI relationship to an associate by owning from 10% to 50% of the voting power of that enterprise.
An associate can extend the FDI relationship to another associate of the higher-level investor only by owning more than 50% of the voting power of that enterprise. Such a chain of associates can be extended as long as majority ownership of voting power exists at each stage.
Basis for extending the relationship through joint ownership:
Where the investor and its subsidiaries combined own more than 50% of the voting power of an enterprise, this enterprise is a subsidiary of the investor.
Where the investor and its subsidiaries combined own at least 10% of the voting power of an enterprise but no more than 50%, this enterprise is an associate of the investor.
Where an investor’s associate and the associate’s subsidiaries combined own more than 50% of the voting power of an enterprise, the enterprise is an associate of the investor.
Source: OECD (2009[3]), OECD Benchmark Definition of Foreign Direct Investment, Fourth Edition (BD4), https://doi.org/10.1787/9789264045743-en.
2.5.1. Time of recording
97. In principle, FDI financial transactions are to be recorded when a change of equity ownership occurs between a resident and a non-resident and while a FDI relationship exists between the two. The creation, change or extinguishing of foreign direct investment positions results from these transactions. As stated in the System of National Accounts (2025 SNA, Chapter 4, Section A, paragraph 4.7), “a transaction is… an interaction between two institutional units by mutual agreement…” It follows that transactions should be recorded on an accrual basis (see 2025 SNA, Chapter 4), that is when the economic event giving rise to them occurs rather than when they are settled. While the accrual basis is close to business accounting, it may not always be possible to obtain data on an accrual basis. Often the sources of information determine the time at which the transactions may be recorded.
2.5.2. Valuation
98. Market prices are the appropriate principle for valuation of transactions and positions data. Market prices reflect the command that entities have currently over resources (for assets) and the current charge on their resources required for the liquidation of their liabilities. The use of market prices also serves as the only basis under which all parties can calculate their assets and liabilities consistently, and thereby helping to reduce asymmetries between the creditor and debtor. Market prices are used for transactions as they best reflect the mutual exchange of value between two unrelated entities.
99. Holdings included in the direct investment position may be denominated in a foreign currency. When these positions are translated into the compilation currency of the reporting economy, the closing mid-point between the buy and sell spot exchange rate relevant at the date to which balances relate should be used. Similarly, transactions should be converted at the mid-point of the buy and sell spot exchange rates at the time of the transaction.
100. Details on the market valuation of FDI positions is provided in Chapter 4. Market valuation of FDI equity positions is often not available because the equity is unlisted; in these cases, the compiler may need to approximate the market value. Chapter 4 and Annex 4.A discuss methods to make these approximations.
2.5.3. The debtor/creditor principle
101. In compiling and presenting FDI statistics, account must be taken that information on the funding received or provided or on the associated FDI income transactions may be collected by different means and from different sources. The geographical allocation of the amounts involved, firstly determining whether they are between residents and non-residents and then, for the latter, determining the specific economy involved, depends on the allocation method used. This Benchmark Definition strongly recommends the use of the debtor/creditor principle. A debtor is a person or an entity that has a financial obligation to another person or entity. Conversely, a creditor is a person or entity that has a financial claim on another person or entity. Therefore, a debtor has a financial liability to a creditor and a creditor has a financial claim (an asset) on a debtor. For FDI statistical purposes, under the debtor/creditor principle, the FDI assets (both transactions and positions) of the compiling economy are allocated to the economies of residence of the non-resident debtors; its FDI liabilities are allocated to the economies of residence of the non-resident creditors allocated on the basis of the debtor/creditor principle. Therefore, adherence to this principle – although in some cases it may be difficult to implement – results in a complete and conceptually consistent set of transaction and position data at an economy or regional level. Use of other allocation approaches (such as the economy of the transactor) can lead to incorrect results between transactions and positions at bilateral (economy or regional) level.
2.5.4. The asset/liability principle and the directional principle
102. This Benchmark Definition recommends that FDI data be presented in two ways: on a straightforward asset/liability basis (i.e., under the asset/liability principle) and reflecting the direction of direct investment influence within the context of the FDIR (i.e., under the directional principle). The two principles are described as follows:
Asset/liability principle: Under this approach, a compiling economy reports for resident institutional units classified under the FDIR as direct investors, DIEs or as fellow enterprises, all FDI financial claims on and obligations to non-residents using the normal balance sheet data showing gross assets and liabilities for positions, and net transactions for each category. The data presented on this basis, while compiled distinguishing the nature of the relationship between the FDIR counterparts, do not incorporate any offsetting of reverse direct investment transactions or positions in equity or debt between a DIE and its direct investor. Similarly, the asset/liability presentation does not incorporate any offsetting of any transactions or positions between fellow enterprises.
Directional principle: Presentation of the FDI data on a directional basis is designed to provide users with information reflecting the direction of influence underlying the direct investment. This requires a compiling economy to determine whether the investment was inward foreign direct investment, i.e., that the influence giving rise to it originated abroad, and that it resulted in the establishment by a non-resident direct investor of a DIE resident in that economy; or that it was outward foreign direct investment, i.e., that the influence giving rise to it originated within the compiling economy, and that it resulted in the establishment by a resident direct investor of a DIE abroad. Under the directional principle, the determination by a compiling economy of the direction of the investment (inward or outward) transactions and positions between a resident fellow enterprise and a non-resident fellow enterprise is made by reference to whether the ultimate controlling parent of the fellow enterprises is resident or non-resident. If the ultimate controlling parent is a resident of the compiling economy, then the transactions and positions between the two fellow enterprises are categorised as outward foreign direct investment; if the ultimate controlling parent is not a resident of the compiling economy, then the transactions and positions between the two fellow enterprises are categorised as inward foreign direct investment.
Difference of the two presentations. The two presentations differ in their treatment of reverse investment. Reverse investment occurs when a DIE invests in its parent. Under the asset/liability presentation, the asset side includes assets of both resident parent companies and of resident DIEs, and the liability side includes all liabilities of resident parents and resident DIEs. Under the directional presentation, reverse investment is subtracted to derive the amount of total outward or inward investment of the reporting economy.
Annex 2.B provides more information on the asset/liability and directional presentations of direct investment, including detailed conversion matrices between the two presentations for FDI positions, financial transactions and income.
103. The asset/liability basis facilitates macroeconomic analyses, such as examining the composition and size of an economy’s liabilities and assets to assess its vulnerability to crises and examining the cross-border links between resident companies and non-resident companies. By providing consistent information on the composition and size of assets and liabilities by functional category of investment (e.g., direct investment or portfolio investment) and by instrument (e.g., equity or debt), an economy’s IIP provides important insights into how vulnerable its economy is to external market conditions. For example, assessing the share of total debt liabilities in direct investment is important because the returns to creditors in direct investment depend on the performance of the debtor. In contrast, the returns to creditors on debt liabilities in portfolio investment do not depend on the performance of the debtor but are required even if the debtor is in difficulty, and, hence, pose a greater risk to the economy. That is, the creditor in a direct investment relationship is likely to be more patient because of their relationship with the debtor.
104. The statistics on a directional basis are used when presenting the detailed statistics, such as those by partner economy or by sector of economic activity. They are useful for studying the nature and motivation for foreign direct investment because they differentiate between those making the investment decision and those receiving the investment, and, so, they can be used to identify the most important source economies for direct investment as well as assessing market access in recipient economies. Directional statistics can answer questions about the economic activities attracting the greatest amounts of direct investment in an economy. Identifying the direction of investment is key to analysing the impact of inward and outward direct investment on the economy, such as analysing the role of foreign-owned enterprises in the host economy and the foreign investment of resident parent companies on the home economy. It is also needed when using data-linking between FDI and other statistics to measure the activities of multinational enterprises or greenfield investment (using the capital approach).
References
[4] IMF (2009), Balance of Payments and International Investment Position Manual, Sixth Edition (BPM6), International Monetary Fund, Washington D.C, https://doi.org/10.5089/9781589068124.069.
[2] IMF (Forthcoming), Integrated Balance of Payments and International Investment Position Manual, Seventh Edition (BPM7), International Monetary Fund, Washington D.C.
[3] OECD (2009), OECD Benchmark Definition of Foreign Direct Investment, Fourth Edition (BD4), OECD Publishing, Paris, https://doi.org/10.1787/9789264045743-en.
[1] United Nations et al. (Forthcoming), 2025 System of National Accounts, United Nations et al., New York.
105. This annex presents the framework for direct investment relationships (FDIR), the preferred method for identifying the extent and type of direct investment relationships. It also provides information on two alternatives to the FDIR – the participation multiplication method (PMM) and the direct influence/indirect control method (DIIC). The methods are compared in terms of inclusions and exclusions from direct investment. Practical implementation of the FDIR for measuring investment income and classifying financial transactions and positions is also discussed.
106. The FDIR includes in direct investment all subsidiaries and associates of an investor.
A subsidiary is an enterprise in which an investor owns more than 50% of the voting power. In Annex Figure 2.A.1, B is a subsidiary of A.
Source: OECD (2009[3]), OECD Benchmark Definition of Foreign Direct Investment, Fourth Edition (BD4), https://doi.org/10.1787/9789264045743-en.
Where an investor and its subsidiary(s) combined own more than 50% of the voting power of an enterprise, the owned enterprise is also regarded as a subsidiary of the investor. In Annex Figure 2.A.2, C is a subsidiary of A.
Source: OECD (2009[3]), OECD Benchmark Definition of Foreign Direct Investment, Fourth Edition (BD4), https://doi.org/10.1787/9789264045743-en.
An associate is an enterprise in which an investor owns at least 10% and no more than 50% of the voting power. In Annex Figure 2.A.3, B is an associate of A.
Source: OECD (2009[3]), OECD Benchmark Definition of Foreign Direct Investment, Fourth Edition (BD4), https://doi.org/10.1787/9789264045743-en.
Where an investor and its subsidiary(s) combined own at least 10% but no more than 50% of the voting power of an enterprise, the owned enterprise is regarded as an associate of the investor. In Annex Figure 2.A.4, C is an associate of A.
Source: OECD (2009[3]), OECD Benchmark Definition of Foreign Direct Investment, Fourth Edition (BD4), https://doi.org/10.1787/9789264045743-en.
Where an investor’s associate (and its subsidiaries combined) own more than 50% of an enterprise, the owned enterprise is regarded as an associate of the investor. In Annex Figure 2.A.5, D is an associate of A.
Source: OECD (2009[3]), OECD Benchmark Definition of Foreign Direct Investment, Fourth Edition (BD4), https://doi.org/10.1787/9789264045743-en.
107. The FDIR aims to identify all enterprises over which the investor has significant influence following the 10% voting power criterion for FDI. The degree of FDI influence that may be exercised through controlling links (where more than 50% of the voting power is held) is not diminished by the existence of multiple links. Thus, an enterprise controlled by a subsidiary or by a group of related subsidiaries (which may also include the investor) is itself regarded as a subsidiary. An enterprise controlled by an associate is also regarded as an associate. The degree of FDI influence that may be exercised through a single or cumulative non-controlling link (where from 10% to 50% of the voting power is held) is diminished by one degree. Thus, an associate of a subsidiary or a group of subsidiaries (which may include the investor) is regarded as an associate. An enterprise that is an associate of another associated enterprise is not, in any FDI sense, influenced by the investor in question, and, therefore, is not included in the FDIR. A chain of ownership is followed until the degree of influence that may be exercised by the investor is diminished to the point where an enterprise can be categorised as not influenced. Under the FDIR, the direct investment relationship extends from an investor to cover all of its subsidiaries and associates.
108. In summary, under the FDIR the direct investment relationship extends in both directions along an ownership chain from an enterprise to cover all of its subsidiaries and associates and all of the investors within the FDIR for which the enterprise is a subsidiary or an associate. It also covers ‘across chain’ relationships. Where more than one ownership chain originates from the same direct investor, all entities in all chains are in a direct investment relationship with one another. Note that some of the links in a chain may be within one economy.
109. The participation multiplication method (PMM) for direct investment includes all enterprises in which an investor has voting equity participation of at least 10%. The calculation of the participation percentage is based on a straight multiplication and summation of direct and indirect participation percentages.
110. More specifically, an indirect participation in a given enterprise at the bottom of a chain of ownership is calculated by taking the investor’s participation in the first enterprise, multiplied by the first enterprise’s participation in the next enterprise, multiplied by the corresponding percentages for all other intervening enterprises in the chain, multiplied by the last intervening enterprise’s participation in the given enterprise. In Annex Figure 2.A.6 according to this method, A has a 16% participation in C (20% of 80%). Because this participation is at least 10%, A and C are regarded as being in a direct investment relationship.
Source: OECD (2009[3]), OECD Benchmark Definition of Foreign Direct Investment, Fourth Edition (BD4), https://doi.org/10.1787/9789264045743-en.
111. If the investor’s interest is held through more than one participation chain, then the percentages of direct and indirect participation in all chains are summed to determine the investor’s total participation percentage. If the combined direct and indirect participation percentage is less than 10% in an enterprise in another economy, then that enterprise is not considered to be in a direct investment relationship with the investor. In Annex Figure 2.A.7, A holds 3% of D indirectly through B and 9% of D indirectly through C, so A holds a total of 12% of D from its combined holdings through B and C and, as a result, D is a DIE of A.
Source: OECD (2009[3]), OECD Benchmark Definition of Foreign Direct Investment, Fourth Edition (BD4), https://doi.org/10.1787/9789264045743-en.
112. It should be noted that under the FDIR, D would not be considered to be in a direct investment relationship with A.
113. The direct influence/indirect control method (DIIC) for direct investment includes all enterprises of which the voting power is 10% or more directly owned, plus all enterprises that are controlled by them (ownership of more than 50% of the voting power), plus all other enterprises in a continuous chain of majority ownership.
114. This method allows the first link in an ownership chain to be a non-controlling link, but all subsequent links must be controlling links. Thus, the DIIC breaks the ownership chain at the second non-controlling link (as in the FDIR) where the first link from the investor is a non-controlling link. On the other hand (and contrary to the FDIR), the DIIC breaks the ownership chain at the first non-controlling link where the first link from the investor is a controlling link. As such, the DIIC will always identify the enterprises in a direct investment relationship as a subset of those identified by the FDIR. In Annex Figure 2.A.8 according to this method, B and C are in a direct investment relationship with A and with each other; E and G are also in a direct investment relationship with A (indirectly), B and C and with each other, while enterprise D is only in a direct investment relationship with B and E, and enterprise F is only in a direct investment relationship with C and G.
Source: OECD (2009[3]), OECD Benchmark Definition of Foreign Direct Investment, Fourth Edition (BD4), https://doi.org/10.1787/9789264045743-en.
115. It should be noted that under the FDIR, A would be in a direct investment relationship with all of the other enterprises except enterprise F.
116. Consistent with the indirect control principles of the FDIR, the DIIC includes in the direct investment relationship cases where combined control is exercised by a number of affiliated enterprises when their first links are controlling links.
117. Annex Figure 2.A.9 presents six different relationships between an indirectly held enterprise and an ultimate investor. These relationships are analysed in Annex Table 2.A.1 in terms of whether the indirectly held enterprise is in a direct investment relationship with enterprise A.
Source: OECD (2009[3]), OECD Benchmark Definition of Foreign Direct Investment, Fourth Edition (BD4), https://doi.org/10.1787/9789264045743-en.
Source: OECD (2009[3]), OECD Benchmark Definition of Foreign Direct Investment, Fourth Edition (BD4), https://doi.org/10.1787/9789264045743-en.
118. From the above examples, it can be seen that the DIIC will always identify a subset of the enterprises in a direct investment relationship as being in that relationship. It does not recognise the relationship between enterprise A and enterprises F and G as being a direct investment relationship. The DIIC will never identify two enterprises as being in a direct investment relationship where none exists according to the FDIR.
119. The PMM may also fail to recognise a direct investment relationship where one exists according to the FDIR; for example, the PMM does not recognise the direct investment relationship between enterprise A and enterprises G and I because its voting power falls below the 10% threshold. Even where a chain of control exists, the PMM may fail to recognise a direct investment relationship between enterprises sufficiently separated in the chain. Unlike the DIIC, the PMM may also identify enterprises as being in a direct investment relationship where none exists according to the FDIR; for example, the PMM considers enterprise H to be in a direct investment relationship with enterprise A because its voting power is 16% and, thus, exceeds the 10% threshold.
120. Given that other related statistics are using the enterprise group as a basic unit, it is interesting to show how direct investment relationships may relate to enterprise groups. The example in Annex Figure 2.A.10 is offered to help understand these relationships.
121. In this example, there are three FDI relationships under the FDIR, one global enterprise group and one local enterprise group comprised of the following units:
FDI Relationship based on A: A, C, D, E, F, G, I (= for PMM and DIIC)
FDI Relationship based on B: B, C, D, E, F, G, I (= for PMM and DIIC)
FDI Relationship based on C: C, D, E, F, G, H, I (= for PMM, DIIC excludes G)
Global Enterprise Group: C, D, E, F, G, I
Local Enterprise Group: D, E, F, G
122. The following points may be noted:
The purpose of the diagram and the FDI relationships described above is to show direct investment relationships based around different direct investors and how they overlap, but only looking down the investment chain (the 'up-the-chain' relationship, e.g., concerning the inclusion of enterprises A and B in the full FDIR based on enterprise C is ignored for this particular purpose);
Enterprise H is not in a direct investment relationship with either A or B because, under the FDIR, C is an associate of A, and therefore D is an associate of A, and E and F are associates of A, and G is an associate of A, but H, as it is an associate of an associate, is therefore not influenced by A and is not covered by the FDIR based on A (and similarly for the FDIR based on B).
Source: OECD (2009[3]), OECD Benchmark Definition of Foreign Direct Investment, Fourth Edition (BD4), https://doi.org/10.1787/9789264045743-en.
123. Economies 2 and 3 have to identify all three FDI relationships to fully articulate the potential FDI counterparties that need to be considered in measuring FDI. In the case of economy 3, it is important not to stop at enterprise C, but to also identify the two direct investors, A and B.
124. One could also note that:
C is the ultimate controlling parent for D, E, F, G and I
C is the parent of D, E, F and G
The local enterprise group (D, E, F and G) is the parent of I.
125. Income on direct investment equity is measured on the basis of current operating performance and accrues to the direct investor as it is earned. The operating performance of a DIE includes the earnings accruing to it from any enterprises in which it holds equity which are also in a direct investment relationship with the direct investor, be it through direct or indirect ownership.
126. When measuring income credits/revenues, the entity resident in the reporting economy that has a non-resident DIE is treated as the direct investor in ownership chains. Any income accruing to the resident entity includes the proportional accrued income of all entities in chains of ownership with the resident as the direct investor. All income should be 'equity-accounted' in accordance with accrual principles.
127. When measuring income debits/expenditures, the degree of influence recognised between the non-resident direct investor and the resident DIE is the critical factor:
If the resident DIE is a subsidiary of the non-resident direct investor, ownership chains from the resident DIE are followed past the first associate encountered in a chain and broken when the second associate is reached; while
If the resident DIE is an associate of the non-resident direct investor, ownership chains from the resident DIE are broken when the first associate in a chain is reached.
128. In some cases, it may be necessary to go further up the chain of ownership than the immediate investor to identify cases of cumulative ownership.
129. The earnings accruing to the non-resident direct investor are its proportion of the accrued consolidated earnings of the resulting ownership chain.
130. In compiling FDI statistics, any financial transaction or position between entities in a direct investment relationship is classified as a direct investment transaction or position (with the exception of the exclusion from FDI statistics of debt transactions and positions between financial intermediaries in a direct investment relationship and of financial derivatives). Only transactions and positions between FDI-related enterprises which are resident in different economies are included in FDI statistics.
131. When classifying positions and transactions, all ownership chains of which the resident enterprise is a member must be considered. The ownership chains are those for which the resident enterprise is treated as the direct investor, and any ownership chains in which the resident enterprise is an associate or a subsidiary. The latter chains are identified by moving “up” the chain from the resident enterprise, through the first non-controlling link, and stopping at the highest enterprise before the next non-controlling link above that point is encountered. The resident enterprise is then in a direct investment relationship with all enterprises that are in a direct investment relationship with any enterprise above it in an ownership chain. Note that there may be multiple chains and multiple direct investors identified in this process. This may also result in a number of fellow enterprises being identified through going up a chain and coming down another chain originating at the same investor, i.e., ‘horizontal’ or across-chain links between such fellow enterprises (which may invest in one another but for which such investment involves no voting power or, where it does, it is limited to less than 10% ownership).
132. For example, considering Annex Figure 2.A.8 above, direct investment positions and transactions for each economy would be recorded according to FDIR as shown in Annex Table 2.A.2.
Source: OECD (2009[3]), OECD Benchmark Definition of Foreign Direct Investment, Fourth Edition (BD4), https://doi.org/10.1787/9789264045743-en.
133. There are two presentations of FDI statistics: the asset/liability presentation and the directional presentation. Historically, FDI was shown according to the directional principle, which organises the transactions and positions according to the direction of investment for the reporting economy – either outward or inward. Therefore, for a particular economy, all transactions and positions of direct investors resident in that economy are shown as outward investment, and all transactions and positions for direct investment enterprises (DIEs) resident in that economy are shown as inward investment. The presentation of aggregate FDI statistics changed with the Balance of Payments and International Investment Position Manual, Sixth Edition (BPM6, (IMF, 2009[4])) and the fourth edition of the Benchmark Definition, both of which recommended that aggregate FDI statistics be shown on an asset/liability basis, which organises FDI statistics according to whether the investment relates to an asset or a liability for the reporting economy. For example, an economy’s assets include equity investment by direct investors resident in that economy in their non-resident DIEs because that investment corresponds to a claim that they have on assets in other economies. Similarly, an economy’s liabilities include non-resident direct investors’ equity investment in DIEs resident in that economy because that investment represents a claim that non-resident investors have on assets in the reporting economy. The change to recording aggregate FDI statistics on the asset/liability basis was recommended to make FDI statistics consistent with other macroeconomic statistics in general and with the statistics for other functional categories of investment in the balance of payments and international investment position.
134. Data presented on a directional basis are better suited to examine the nature and motivations of FDI. The directional principle is the preferred presentation for FDI data broken down by industry and by partner economy because it allows for the identification of the main sources and main destinations of FDI. However, under the asset/liability measures, this is not possible because, for example, the acquisition of assets combines both investment by resident direct investors in their non-resident DIEs and loans by resident DIEs to their foreign direct investors. The directional presentation is also helpful for FDI data on an aggregate level as it sheds light on the motivations behind FDI trends in a timely manner (as aggregate data are released on a quarterly basis). For these reasons, the Benchmark Definition continues to recommend that countries compile the details necessary to convert aggregate FDI statistics from the asset/liability presentation to the directional presentation (see Section 2.5.4Annex 2.B and Annex B).
Recording under the asset/liability and directional presentations
135. The two presentations differ in their treatment of reverse investment and investment between fellow enterprises, which can occur in both equity and debt. Reverse investment occurs when a DIE invests in its direct investor. Under the asset/liability presentation, the asset side includes assets of both resident direct investors and of resident DIEs, and the liability side includes all liabilities of resident direct investors and resident DIEs. Under the directional presentation, reverse investment is subtracted to derive the amount of total outward or inward investment of the reporting economy. Therefore, if a resident direct investor borrows money from one of its non-resident DIEs, this is subtracted in calculating the reporting economy’s outward investment because it reduces the amount of money that that economy’s direct investors have invested in their non-resident DIEs. Similarly, if a resident DIE lends money to its non-resident direct investor, this is subtracted when calculating inward investment because it reduces the amount of money that the non-resident direct investor has invested in that economy. While reverse equity investment is to be treated the same way as reverse debt investment, it is so rare that most of the difference between the two presentations is due to differences in the treatment of reverse debt investment.
136. The treatment of transactions and positions between fellow enterprises deserve special attention under the directional presentation. To reflect the direction and degree of influence exerted by resident and non-resident direct investors in the reporting economy, it is important to recognise that a resident fellow does not achieve any influence over a non-resident fellow if it made a loan to that non-resident fellow—the influence remains with the direct investor common to both fellows. Similarly, a non-resident fellow did not achieve any influence over a resident fellow by extending a loan to it—the influence remained with the direct investor common to the fellows.
137. Under the directional principle, the recording of transactions and positions between fellow enterprises in a reporting economy depends on the residence of the ultimate controlling parent (UCP) of the fellow enterprises because it is the UCP that ultimately controls the transactions of the fellows. While this treatment applies to both equity and debt investment between fellows, equity investment is rare so it is debt that has the biggest impact on the statistics. If the UCP of the fellow enterprise is resident in the economy, then loans by and to the fellow enterprise are treated as outward investment. Any loan from a fellow enterprise to a fellow enterprise resident in another economy is treated as an increase in outward investment by the reporting economy because it represents an increase in the influence that a resident direct investor (the UCP) has on the DIE in another economy. Similarly, if the fellow enterprise receives a loan, it reduces outward direct investment just as it would if the UCP had received a loan because such investment reduces the total amount the resident direct investor – the UCP – has invested abroad.
138. If the UCP is not resident in the economy, then transactions and positions are treated as inward investment. If the fellow enterprise makes a loan to a fellow in another economy, that is treated as a reduction in inward investment to the reporting economy as funds that flowed into the reporting economy from the non-resident UCP have now flowed to another economy, reducing the amount of foreign direct investment in the reporting economy. It should not be treated as outward investment as making a loan to a fellow enterprise in another economy does not give the resident fellow any influence over the operations of the fellow in the other economy; instead, it is the UCP that still has the influence. If the fellow resident in the reporting economy receives a loan from a fellow in another economy, it increases inward investment as the non-resident UCP’s influence in the reporting economy has increased.
139. Six classes of direct investment assets and six classes of direct investment liabilities can be defined that will serve as the building blocks for the presentations of FDI statistics described in this section:
Direct investment assets: (FDI Assets = A1+A2+A3+A4+A5+A6)
A1: Equity assets of direct investors in direct investment enterprises
A2: Debt instrument assets of direct investors in direct investment enterprises
A3: Equity assets of direct investment enterprises in their direct investors (reverse investment)
A4: Debt instrument assets of direct investment enterprises in their direct investors (reverse investment)
A5: Equity assets in fellow enterprises abroad
A5.1: If ultimate controlling parent is resident
A5.2: If ultimate controlling parent is non-resident
A6: Debt instrument assets in fellow enterprises abroad
A6.1: If ultimate controlling parent is resident
A6.2: If ultimate controlling parent is non-resident
Direct investment liabilities: (FDI Liabilities = L1+L2+L3+L4+L5+L6)
L1: Equity liabilities of direct investment enterprises to direct investors
L2: Debt instrument liabilities of direct investment enterprises to direct investors
L3: Equity liabilities of direct investor to their direct investment enterprises (reverse investment)
L4: Debt instrument liabilities of direct investor to their direct investment enterprises (reverse investment)
L5: Equity liabilities to fellow enterprises abroad
L5.1: If ultimate controlling parent is non-resident
L5.2: If ultimate controlling parent is resident
L6: Debt instrument liabilities to fellow enterprises abroad
L6.1: If ultimate controlling parent is non-resident
L6.2: If ultimate controlling parent is resident
140. In the first category, the financing flows in the same direction as the influence or control, from the direct investor to the DIE (A1 and A2; L1 and L2). This is in line with the core principle for direct investment that a direct investor acquires influence or control through the provision of financing.
141. In the second category of reverse investment, financing flows in the opposite direction as the influence or control (A3 and A4; L3 and L4). This category reflects the reverse of the standard FDI financing flows as the direct investor uses its influence to have its DIEs provide equity or debt financing for its own operations. Such investment is limited to cases where the acquisition of voting power by a DIE in its direct investor does not meet the 10% criterion for establishing a separate direct investment relationship in its own right.
142. The last category shows financing involving fellow enterprises that have no direct investment influence upon one another (i.e., the 10% voting power criterion is not met as there is not sufficient equity ownership in each other) but have a common parent (A5 and A6; L5 and L6). The classes of assets and liabilities in or to fellow enterprises are further subdivided into two sub-classes according to the residence of the ultimate controlling parent (UCP), i.e., the direct investor at the top of the control chain (A5.1, A5.2, A6.1 and A6.2; L5.1, L5.2, L6.1 and L6.2).
143. These building blocks can be arranged in different presentations suitable for a variety of analytical requirements. Macroeconomic accounts, balance of payments and the national accounts present financial data primarily on an asset and liability basis. Thus, FDI data on this basis are useful when analysing direct investment in relation to other macroeconomic variables of the external sector of the economy. In addition, most FDI data are compiled as part of the balance of payments programmes in each country, bridging macroeconomic accounts and other presentations of FDI statistics.
FDI positions
144. To illustrate the asset/liability approach for FDI positions, the six classes presented above are to be grouped as shown in Annex Table 2.B.1 below.
Source: OECD (2009[3]), OECD Benchmark Definition of Foreign Direct Investment, Fourth Edition, https://doi.org/10.1787/9789264045743-en.
145. The Benchmark Definition stresses the use of two terms when presenting FDI statistics according to the directional principle: outward FDI and inward FDI. Outward FDI includes the net assets of resident enterprises exerting control or influence on non-resident enterprises (net assets of resident direct investors and net assets in fellow enterprises abroad when the UCP is resident). Inward FDI includes the net liabilities of resident enterprises controlled or influenced by non-resident enterprises (net liabilities of resident DIEs to direct investors and net liabilities to fellow enterprises abroad when the UCP is non-resident). The components of outward and inward FDI are schematically shown as in Annex Table 2.B.2 below.
Note: * entered as a deduction in outward or inward FDI.
Source: OECD (2009[3]), OECD Benchmark Definition of Foreign Direct Investment, Fourth Edition (BD4), https://doi.org/10.1787/9789264045743-en.
Outward direct investment = A1-L3+A5.1-L5.2+A2-L4+A6.1-L6.2
Inward direct investment = L1-A3+L5.1-A5.2+L2-A4+L6.1-A6.2
146. FDI asset positions under the asset/liability principle are (generally) greater than FDI outward positions calculated according to the directional principle. In the same manner, FDI liability positions under the asset/liability principle are (generally) greater than the FDI inward positions calculated according to the directional principle.
147. The difference between the asset position and the outward FDI position is equal to:
which simplifies to:
The difference between the liability position and the inward FDI position is equal to:
which simplifies to:
148. The net position (outward FDI less inward FDI and assets less liabilities) is the same for both presentations, i.e., the difference between FDI assets and FDI liabilities is equal to the difference between outward FDI and inward FDI.
149. Annex Table 2.B.3 and Annex Table 2.B.4 illustrate the differences in FDI presentations moving from the asset liability/principle to the directional principle based on the example of a multinational enterprise group given in Annex Figure 2.B.1.
Note: In this example each of the enterprises A through E is in a different economy.
Source: OECD (2009[3]), OECD Benchmark Definition of Foreign Direct Investment, Fourth Edition (BD4), https://doi.org/10.1787/9789264045743-en.
Taking B as the reporting economy. Net FDI liability position: USD 72.
Source: OECD (2009[3]), OECD Benchmark Definition of Foreign Direct Investment, Fourth Edition (BD4), https://doi.org/10.1787/9789264045743-en.
Taking B as the reporting economy. Net inward FDI position: USD 72.
Source: OECD (2009[3]), OECD Benchmark Definition of Foreign Direct Investment, Fourth Edition (BD4), https://doi.org/10.1787/9789264045743-en.
The same component entries appear in both accounts but the entries shown as negative amounts in the directional table correspond to the relevant positive entries on the opposite side of the asset/liability table.
Gross values of assets and liabilities are presented
In both presentations net FDI is the same (USD 72)
Under the directional principle:
The reverse investment asset position with A effectively reduces the aggregate headline inward FDI position of B (it is netted out)
There is a negative inward FDI debt liability position into E resulting from the recording of loans with fellow enterprises (as the ultimate controlling parent of B is non-resident). This also reduces the aggregate headline inward FDI position of B.
The outward FDI position is effectively reduced as a result of the recording of the reverse investment loans from C and D.
Under the asset/ liability principle:
There is an asset position with A that contributes to the gross total.
There is a liability position with C that contributes to the gross total.
The notion of inward and outward FDI is replaced by the notion of FDI asset and FDI liability.
150. Where reverse investment exists or where investment occurs between fellow enterprises, the overall inward direct investment position for an economy will generally be lower in magnitude than the overall direct investment liability position presented on an asset/liability basis. The equivalent outcome to that for overall inward direct investment will also result for the overall outward direct investment position data compared to the overall direct investment assets.
FDI financial transactions
151. For transactions, the asset/liability principle is schematically shown as in Annex Table 2.B.5 below.
Source: OECD (2009[3]), OECD Benchmark Definition of Foreign Direct Investment, Fourth Edition (BD4), https://doi.org/10.1787/9789264045743-en.
152. To illustrate the directional principle for transactions, the FDI elements are schematically shown as in Annex Table 2.B.6 below.
Note: * entered as a deduction in outward or inward FDI.
Source: OECD (2009[3]), OECD Benchmark Definition of Foreign Direct Investment, Fourth Edition (BD4), https://doi.org/10.1787/9789264045743-en.
153. These differences essentially involve transposing reverse transactions and transactions between fellow enterprises in the two presentations.
FDI income
154. For direct investment income under the asset/liability principle, elements are schematically shown as in Annex Table 2.B.7 below.
Source: OECD (2009[3]), OECD Benchmark Definition of Foreign Direct Investment, Fourth Edition (BD4), https://doi.org/10.1787/9789264045743-en.
155. To illustrate the directional principle for direct investment income, elements are schematically shown as in Annex Table 2.B.8 below.
Note: * entered as a deduction in outward or inward FDI.
Source: OECD (2009[3]), OECD Benchmark Definition of Foreign Direct Investment, Fourth Edition (BD4), https://doi.org/10.1787/9789264045743-en.
156. To help communicate with users on the two presentations of FDI statistics, the OECD’s Working Group on International Investment Statistics developed the template shown in Annex Table 2.B.9. The table begins with the asset/liability figures and shows the net direct investment calculated as assets less liabilities. It then shows the figures on a directional basis and includes several footnotes explaining the relationships between the two presentations. It ends with the net direct investment on a directional basis, calculated as outward less inward, to emphasise that the balances are the same under the two presentations.
By principle and type of capital
Notes:
1. Sum of the following four components: Reporting country direct investment abroad, equity assets of resident direct investor in direct investment enterprise (DIE) + Reporting country direct investment abroad, lending and debt securities (net), assets + foreign direct investment in the reporting country, equity assets of resident DIE in direct investor + foreign direct investment in reporting country, lending and debt securities (net), assets.
2. Sum of the following four components: foreign direct investment in reporting country, equity liabilities of resident DIE to direct investor + Foreign direct investment in reporting country, lending and debt securities (net), liabilities + Reporting country direct investment abroad, equity liabilities of resident direct investor to DIE + Reporting country direct investment abroad, lending and debt securities (net), liabilities.
3. Net lending of reporting country direct investor to DIE and other related companies abroad. Other related companies are those companies which are in the same multinational enterprise group as the reporting country direct investor, but which are neither its directly or indirectly owned affiliates nor its direct or indirect investors (i.e., fellow companies).
4. Assets minus liabilities.
5. Net lending of DIE in the reporting country to foreign direct investor and other related companies abroad. Other related companies abroad are those companies which are in the same multinational enterprise group as the resident direct invest enterprise, but which are neither its direct or indirect investors nor its direct or indirectly owned affiliates (i.e. fellow companies).
6. Liabilities minus assets.
157. The System of National Accounts (SNA) recommends subsectors of the institutional sectors to provide information on particular groups of units for policymaking purposes. This annex provides further information on institutional sectors and, in particular, the subsectors of the financial corporation sector.
158. The central bank is the national financial institution (or institutions) that exercises control over key aspects of the financial system.
159. Deposit-taking corporations except the central bank have financial intermediation as their principal activity. To this end, they have liabilities in the form of deposits or financial instruments (such as short-term certificates of deposit) that are close substitutes for deposits.
160. Money market funds (MMFs) are collective investment schemes that raise funds by issuing shares or units to the public. The proceeds are invested primarily in money market instruments, MMF shares/units, transferable debt instruments with a residual maturity of less than one year, bank deposits and instruments that pursue a rate of return that approaches the interest rates of money market instruments. MMF shares can often be transferred by cheque or other means of direct third-party payments.
161. Non-money market investment funds are collective investment schemes that raise funds by issuing shares or units to the public and investing predominantly in longer-term financial assets, such as equity shares, bonds, mortgage loans, and non-financial assets.
162. Other financial intermediaries except insurance corporations and pension funds (ICPFs) consist of financial corporations that are engaged in providing financial services by incurring liabilities, in forms other than currency, deposits or close substitutes for deposits, on their own account for the purpose of acquiring financial assets by engaging in financial transactions on the market. It is a feature of a financial intermediary that transactions on both sides of the balance sheet are carried out in open markets.
163. Financial auxiliaries consist of financial corporations that are principally engaged in activities associated with transactions in financial assets and liabilities or with providing the regulatory context for these transactions but in circumstances that do not involve the auxiliary taking ownership of the financial assets and liabilities being transacted.
164. This sub-sector consists of institutional units providing financial services, where most of either their assets or liabilities are not transacted on open financial markets. It includes entities transacting within only a limited group of units such as with subsidiaries or subsidiaries of the same holding corporation or entities that provide loans from their own funds provided by only one sponsor. Other financial intermediaries except insurance corporations and pension funds are distinguished from captive financial institutions and money lenders in that the latter serve a limited group only for at least one side of the balance sheet.
165. Insurance corporations consist of incorporated, mutual or other entities whose principal function is to provide life, accident, sickness, fire or other forms of insurance to individual institutional units or group of units or reinsurance services to other insurance corporations. Captive insurance is also included, i.e., an insurance company that serves only its owner.
166. Pension funds included here are only those social insurance funds that are institutional units separate from the units that create them. They are established for purposes of providing benefits on retirement for specific groups of employees. They have their own assets and liabilities, and they engage in financial transactions in the market on their own account.
← 1. For a more detailed discussion of these broad statistical concepts, see BPM7 (IMF, Forthcoming[2]).
← 2. FDI transactions consist of transactions in equity, debt, and income. Throughout the manual, FDI financial transactions is used to refer to transactions in equity, covering equity capital transactions and reinvestment of earnings, and debt transactions. FDI income transactions is used to refer to income transactions, including dividends and withdrawals from income of quasi-corporations, reinvested earnings, and income on debt.
← 3. Citizenship and plans for future location are non-economic connections to an economic territory.
← 4. However, some unincorporated enterprises may have limited liability such as a limited liability partnership.
← 5. The common usage of the term branch is broader, where a “branch” may also mean establishments, incorporated subsidiaries, or industrial classification group.
← 6. Instances may occur when land and buildings are owned (wholly or in part) by unrelated non-residents, wherein some (or all) of the non-resident owners hold less than 10% of the equity in the (notional) unit. In those cases, the 10% threshold is still applicable so that some (or all) of the owners may not be direct investors.
← 7. Similar issues may arise for a Societas Europaea, that is, a company created under European Union law that is able to operate in any member state.
← 8. In general, ordinary shares are the same as voting power. However, there may be instances that the voting power is not represented by ordinary shares. In such cases, compilers must determine the voting power.
← 9. DIEs are also referred to as “foreign affiliates” (subsidiaries, associates, unincorporated businesses) that are either directly or indirectly owned by the direct investor or their non-resident branches. See Glossary for a definition of affiliated entities.