OECD Benchmark Definition of Foreign Direct Investment (Fifth Edition)

11. Communicating and using FDI statistics

Abstract

Communicating their statistics to users is a core function of statistical departments and agencies. This chapter addresses some of the challenges of communicating foreign direct investment (FDI) statistics. The first section provides guidance on communicating with data users and helping them navigate the multiple measures of FDI that economies may publish in their statistics. It then recommends that statistical departments and agencies accompany the dissemination of the statistics with analysis that highlights recent developments. The chapter discusses some interpretations of FDI series and indicators that can be constructed from FDI statistics. It then discusses statistical confidentiality, which can significantly impact the dissemination of FDI statistics. It concludes by offering some advice on communicating with data suppliers and by advising a pragmatic approach to revisions.

11.1. Introduction

735. Communicating about the statistics that have been compiled is a key function of statistical departments and agencies. Effective communication policies can greatly enhance users understanding and utilisation of the statistics. Users benefit from comprehensive, consistent, accurate and reliable information communicated and disseminated on a timely basis in an accessible and understandable manner.

736. Both the Integrated Balance of Payments and International Investment Position Manual, Seventh Edition (BPM7, (IMF, Forthcoming[1])) and the 2025 System of National Accounts (2025 SNA, (United Nations et al., Forthcoming[2])) contain chapters on “Communicating and Disseminating Economic Statistics”. These chapters cover dissemination and communication, including communicating with data suppliers and users, statistical confidentiality, taxonomies, revisions and metadata. Furthermore, they include a framework for measuring alignment with the international macroeconomic statistical standards. Compilers are also encouraged to use language that makes it easier for users to understand specialist terminology.

737. The purpose of this chapter is to provide more detail on issues that are particularly relevant to foreign direct investment (FDI) statistics rather than to repeat the information included in BPM7 and the 2025 SNA. These include communicating with users about the multiple measures of FDI resulting from presenting the statistics according to the asset/liability and directional principles, helping them interpret FDI series, and presenting indicators that can be constructed from FDI statistics. The chapter then discusses statistical confidentiality and how it can impact the statistics to be released. It briefly offers advice on communicating with data suppliers and closes with a discussion of a pragmatic approach to data revisions. Compilers should refer to BPM7 and 2025 SNA for general principles on disseminating and communicating macroeconomic statistics.

11.2. Communicating with users

738. The goal of communicating statistics is to convey a message that tells users what happened, when and where it happened as well as contributing to understanding why and how it happened. An important part of communicating FDI statistics is to include interpretations of the statistics and include key indicators when the statistics are released. An additional complication for users is that economies may publish several different measures of FDI, including on the asset/liability and on the directional principles and including and excluding resident special purpose entities (SPEs). For some responses to common questions from users, see Annex 11.A.

11.2.1. Multiple measures of FDI statistics

739. FDI statistics are unusual in macroeconomic statistics in the sense that they have two principles for recording: the asset/liability principle and the directional principle. In addition, for economies that host significant numbers of SPEs, there can be significant differences in the statistics depending on whether resident SPEs are included or not. All of these different measures can make it difficult for users to determine which statistics best serve their needs. Therefore, it can be helpful to provide information on the uses of the different measures (for more information on the different uses, see Annex 2.B).

740. For FDI aggregates on the asset/liability basis, as a part of national macroeconomic statistics, it is useful to explain that they are consistent with balance of payments (BOP) and international investment position (IIP) statistics as well as the components of national accounts statistics. These data provide for an economy the aggregate totals of FDI assets and liabilities by type of instrument (equity, debt). Data are recorded regardless of the nature of the enterprise and the counterpart or of the direction of influence and control. These FDI statistics include all the funds that pass through resident SPEs as well as capital transiting through operating subsidiaries of multinational enterprises (MNEs) on behalf of the parent companies. In consequence, while data presented using the asset/liability principle provide an overall aggregate measure of FDI, they do not provide an appropriate basis for some analyses, such as exploring the factors that attract FDI to an economy or identifying the industries in an economy that are most attractive to foreign investors. These analyses benefit from data that reflect the direction of influence or control. Data on a directional basis may also be most appropriate for linking to other datasets that are interested in the residency of the parent company, such as developing statistics on the domestic operations of resident direct investors.

741. FDI statistics compiled according to the directional principle show outward and inward investment. Partner economy or industry disaggregation is recorded in the directional presentation of FDI statistics as that of the immediate counterparty involved in the investment chain for both inward and outward investment; inward positions are also to be presented according to the ultimate investing economy (UIE). Also, information for resident SPEs is shown separately in the standard presentation with the full geographic and industry disaggregation under the directional principle.

742. There are additional steps that statistical departments and agencies can take to help users understand the different measures. One is to explain the relationship between the asset/liability presentation and the directional presentation of FDI statistics using the example shown in Annex Table 2.B.9. Another strategy is to publish detailed FDI statistics by partner economy and by industry on the directional principle in a dedicated section separate from the BOP and IIP sections. A final strategy is to reserve the terms inward and outward FDI for the statistics on a directional basis and refrain from referring to direct investment assets as outward direct investment and direct investment liabilities as inward direct investment, which makes it harder for users to understand that there is a difference between the two recording principles.

11.2.2. Interpreting FDI series

743. FDI series are very informative and useful for both short and long-term analysis. Timely estimates of the direct investment activity allow the user to observe recent economic developments. On the other hand, longer series contribute to the measurement of the attractiveness of the reporting economy within the global market and the competitiveness of the economic agents, i.e., foreign direct investors and direct investment enterprises (DIEs).

744. An increase in inward investment by foreign direct investors implies additional capital injected into the economy (the domestic market) and, thus, is likely to have an impact on its economic performance. On the other hand, the size of outward investment transactions indicates the extent of penetration of resident direct investors in other markets. FDI financial transactions and income provide information for FDI activity within a given time period, while FDI positions indicate the levels of investment at a given point in time. The trends obtained from FDI financial transactions and FDI positions may be quite different. In addition, attention is drawn to the point concerning the interpretation of headline FDI statistics in the context of the impact of reverse investment and of investment between fellow enterprises. In this context, for example, a low headline figure for total inward FDI into an economy may result from a general low level of FDI investment under different instruments and across different FDI relationships. Alternatively, it could result from a high level of investment into the economy being partially or completely offset – or even exceeded – by outflows of investment by foreign-owned enterprises to their related direct investors or fellow enterprises abroad. Similar effects apply to outward direct investment.

745. Analysing inward and outward FDI by its sub-components namely equity, reinvestment of earnings (combined with equity for position data) and debt allows further refinement of FDI trends and the nature of investment. Even if the analysis of equity investment is at the core of direct investment, inter-company loans play an important role in financing FDI-related enterprises. However, trends in equity and debt components could be very different and may reflect different patterns. Reimbursements of inter-company loans need to be interpreted correctly and not as divestment. This interpretation is possible when we look at sub-components of FDI. Reinvestment of earnings tends to be the least volatile component. Changes in the reinvestment of earnings can reflect both changes in the level of earnings of affiliates and in the share of earnings that parents choose to reinvest in their affiliates. While reinvesting a higher share of earnings can be interpreted as reflecting a positive investment environment in the host economy, care should be taken as a number of factors can influence the share of earnings reinvested. For example, if the parent has a short-term need for funds, they may distribute a higher share of earnings than usual.

746. FDI by partner economy: the first classification specified by the Benchmark Definition requires that direct investment statistics be disaggregated geographically on a directional basis, i.e., for inward and outward investment. These series provide the geographical distribution of investment, i.e., economy of source and destination. This information is very useful in any comparative analysis on the attractiveness of an economy or a group of economies and for constructing analytical indicators.

747. FDI by industry: direct investment statistics should be disaggregated by industry of the enterprise receiving the direct investment, the DIE for both inward and outward investment. In other words, for outward investment, data are recorded according to the industry of the DIE abroad and for inward investment they reflect the industry of the DIE in the reporting economy. Disaggregation by industry should be based on the principal activity of the enterprise when it is involved in more than one activity. These data offer interesting indicators on the role of FDI in the host economy and the globalisation of markets.

748. The problems raised by SPEs are twofold: (i) the inclusion of resident SPEs may lead to an overstatement of FDI to and/or from the reporting economy and; (ii) investment via non-resident SPEs may lead to the distortion of information on the origin and destination of FDI. FDI series by immediate partner economy and by industry excluding resident SPEs resolve the first problem. The series of inward positions by UIE provides information on the ultimate origin, and the supplementary series on outward positions by ultimate host economy provides information on the ultimate destination.

11.2.3. FDI indicators

749. This section describes indicators that can be constructed from FDI statistics to communicate the role of FDI in the economy.

Extent of globalisation through FDI

750. This set of indicators (Table 11.1) is the most commonly used measurement of the globalisation of an economy. It measures the extent of cross-border investment realised with the objective of a lasting interest to and from an economy. It also allows comparisons across economies based on the relative importance of FDI, i.e., (i) financial transactions; (ii) income transactions; or (iii) FDI positions expressed as a percentage of gross domestic product (GDP). Each ratio indicates the relative importance of globalisation for the reporting economy either for total investment or for investment by industry sector. An increase of the ratio implies a greater share of foreign investment, thus an increase in the relative impact of globalisation. It is, however, to be noted that FDI transactions and FDI positions are not fully compatible with GDP, which measures the total value added from production. Nevertheless, in the absence of other more meaningful international comparisons of the relative size of globalisation, GDP remains the best common reference. In addition, comprehensive and timely GDP statistics are available in most countries.

751. FDI financial transactions (inward and outward) expressed as a percentage of GDP indicate the degree of globalisation of an economy based on the economic climate for a given time period, i.e., the changes over that time period. This indicator provides early information on the relative attractiveness of economies (both domestic and foreign) and industries for new investment after allowing for the withdrawal of investment (disinvestment) during that same time period. FDI transactions are compiled in many economies on a quarterly basis or on a monthly basis, with a view to allowing a timely monitoring of their FDI activities.

752. FDI income transactions (inward and outward) as a percentage of GDP provide information on the relative importance of the earnings of DIEs both in the reporting economy and abroad.

Table 11.1. Extent of globalisation through FDI

Total FDI or by industry

Inward FDI financial transactions as a percentage of GDP

Outward FDI financial transactions as a percentage of GDP

Inward FDI income transactions as a percentage of GDP

Outward FDI income transactions as a percentage of GDP

Inward FDI positions as a percentage of GDP

Outward FDI positions as a percentage of GDP

Source: Based on OECD (2009[3]), OECD Benchmark Definition of Foreign Direct Investment, Fourth Edition (BD4), https://doi.org/10.1787/9789264045743-en.

753.  FDI positions (inward and outward) as a percentage of GDP indicate the extent of globalisation at a given point in time. The ratio for inward FDI positions indicates the extent of foreign ownership (or foreign presence) in an economy. The ratio for outward investment indicates the degree of ownership (or presence abroad) of economic agents in other markets. They also illustrate, respectively, the level of dependence of the domestic economy on foreign economies and its level of penetration in foreign markets. A comparison of the results obtained for inward and outward FDI will indicate the comparative importance of the economy as a source or recipient of FDI.

FDI in host and from investing economies or by industries

754. Expressed as ratios (as percentage of totals for both inward and outward FDI), the results reflect the process of globalisation of the economies by their level of financial expansion abroad and their dependence on financing from abroad (Table 11.2). The indicator related to investing (home) and host economies shows the evolution of the share of individual economies as origins of direct investment (outward investment by reporting economy) or as hosts of direct investment (inward investment by non-residents to the reporting economy). The ratio based on FDI financial transactions allows analysis of the changes between two periods while the ratio based on positions indicates the structural developments over time. The geographical analysis can be further deepened by measuring the share of the sub-component FDI equity capital in such investment by partner economy. Changes in these ratios will indicate a greater/lesser contribution to globalisation by the host or investing economy.

755. The indicator based on industry sectors is similar to the previous indicator (by partner economy) but the focus this time is on the industry for both inward and outward investment. This indicator describes the relative contribution of different economic sectors to the international economic system by the measurement of the share of FDI positions by economic sector abroad or by the measurement of the dependence of the home economic sectors on investment from abroad. Changes in these ratios indicate a greater/lesser contribution to globalisation by individual economic sectors.

Table 11.2. FDI in host and from investing economies or by industries

Relative share of inward FDI financial transactions by partner economy as a percentage of total inward FDI transactions

Relative share of outward FDI financial transactions by partner economy as a percentage of total outward FDI transactions

Relative share of inward FDI positions by partner economy as a percentage of total inward FDI positions

Relative share of outward FDI positions by partner economy as a percentage of total outward FDI positions

Relative share of inward FDI financial transactions by industry as a percentage of total inward FDI transactions

Relative share of outward FDI financial transactions by industry as a percentage of total outward FDI transactions

Relative share of inward FDI positions by industry as a percentage of total inward FDI position

Relative share of outward FDI positions by industry as a percentage of total outward FDI position

Source: Based on OECD (2009[3]), OECD Benchmark Definition of Foreign Direct Investment, Fourth Edition (BD4), https://doi.org/10.1787/9789264045743-en.

Return on FDI

756. This set of indicators provides information regarding the profitability of DIEs (Table 11.3). For example, when the rate of return of inward FDI (inward FDI income as a percentage of total inward FDI positions) increases, it implies that the resident DIEs are more profitable and are more competitive for investors. However, observations based purely on the results of the statistical ratios are not sufficient to draw conclusions on the competitiveness of enterprises (or an economy). Many other factors should also be taken into account, such as cyclical or structural factors, developments in that sector of economic activity as well as other factors related to the global strategy of the investing enterprise(s).   

Table 11.3. Return on FDI

Inward FDI income debits/expenditures as a percentage of total inward FDI position [rate of return for total inward FDI or by industry or investing economy]

Outward FDI income credits/revenues as a percentage of total outward FDI position [rate of return for total outward FDI or by industry or investing economy]

Source: Based on OECD (2009[3]), OECD Benchmark Definition of Foreign Direct Investment, Fourth Edition (BD4), https://doi.org/10.1787/9789264045743-en.

Dynamics of FDI in reporting economy

757. A first set of indicators under this group measures the share of foreign capital in the reporting economy (Table 11.4). First of all, it compares, at a specific point in time, the direct investment liability positions to the overall liabilities (balance sheet totals) of the domestic sector as reported in the national balance sheet accounts. Likewise, it also compares the direct investment asset positions to total assets (balance sheet totals) of the reporting economy. In the first case, an increase of the ratio implies a higher foreign presence through FDI. In the latter indicator, an increase implies the opposite, i.e. an increasing presence by the home economy in foreign markets.

758. Another set of indicators provides detailed information on the relative share of components of FDI financial transactions and positions. The information is based on aggregate data by sub-components of FDI (equity capital, reinvested earnings and other capital) on a directional basis. The focus is on examining the dynamics of inward investment in the reporting economy or outward investment by the reporting economy. The indicators would show the role of equity capital, reinvestment of earnings and other capital (e.g., inter-company debt) in financing FDI into and from the economy and the share of equity and other capital in the inward and outward FDI positions. The shares of the different components will vary across economies, depending on a number of factors, such as the competitiveness of economies, maturity of investment, and institutional setting. It could also be of interest to calculate these indicators with data disaggregated by partner economy or by industry to understand how the dynamics of FDI may vary across those dimensions.

Table 11.4. Dynamics of FDI in reporting economy

FDI liability positions as a percentage of total liabilities of the domestic economy (balance sheet totals)

FDI asset positions as a percentage of total assets of the domestic economy (balance sheet totals)

Inward equity capital transactions as a percentage of inward FDI transactions

Inward reinvested earnings as a percentage of inward FDI transactions

Inward other capital transactions as a percentage of inward FDI transactions

Outward equity capital transactions as a percentage of outward FDI transactions

Outward reinvested earnings as a percentage of outward FDI transactions

Outward other capital transactions as a percentage of outward FDI transactions

Inward position of equity capital and reinvested earnings as a percentage of inward FDI positions

Inward other capital positions as a percentage of inward FDI positions

Outward positions of equity capital and reinvested earnings as a percentage of outward FDI positions

Outward other capital positions as a percentage of outward FDI positions

Source: Based on OECD (2009[3]), OECD Benchmark Definition of Foreign Direct Investment, Fourth Edition (BD4), https://doi.org/10.1787/9789264045743-en.

Share of FDI by type

759. Analysing the share of FDI by type is part of the information needed to analyse the impact of FDI particularly in the host economy (Table 11.5). Spill-over effects of the investment will vary depending on the type of investment (e.g., greenfield investment and extensions of capacity compared to mergers and acquisitions) and its relative share over total investment of an economy.

Table 11.5. Share of FDI by type

Inward greenfield investment and extensions of capacity as percentage of total inward FDI transactions

Outward greenfield investment and extensions of capacity as percentage of total outward FDI transactions

Inward M&A equity transactions as a percentage of total inward FDI equity transactions

Outward M&A equity transactions as a percentage of total outward FDI equity transactions

Note: ‘M&A’ refers to mergers and acquisitions.

Source: Based on OECD (2009[3]), OECD Benchmark Definition of Foreign Direct Investment, Fourth Edition (BD4), https://doi.org/10.1787/9789264045743-en.

11.3. Statistical confidentiality

760. By law, most FDI statistics producers that use surveys as a data source do so for statistical purposes only and often under legislation that requires the statistics not be disseminated or published in a way that permits the identification of data referring to a particular entity. Thus, it is important to ensure appropriate data confidentiality policies, anonymisation techniques and disclosure checking procedures are in place as part of the process before publication of any statistics.

761. As a result, compilers in many economies may encounter the possibility of confidential data occurring in the FDI statistics to be disseminated. Such information generally does not directly identify the entity, e.g., the name and address of an enterprise. However, the particular context in which data are presented e.g., classification by partner economy, by industry of enterprise, by type of financial instrument as well as cross-classifications of these attributes, may allow users to determine the identity of the entity in question and the value of its activity from the information provided (as well as other information they may have independently available to them). In general, the greater the level of detail in the analyses, as well as the degree of concentration of particular types of enterprises and their activities in a compiling economy, the greater the likelihood of identifying confidential data.

762. FDI information can be regarded as being confidential in a primary sense for a number of reasons:

• if a compiler declares it to be confidential

• if there is only one or at most two entities giving rise to the information

• if the contribution of a particular enterprise (or even two enterprises) dominates the contributions of all other entities.

763. In the last two cases, the usual confidentiality rules specify that where there are fewer than three contributors to a particular statistical cell then this information is regarded as being confidential. The reason for this is that, for example, where there are two contributors then either of them can potentially determine the contribution of the other. Clearly, where there is only one contributor then any user with sufficient knowledge can potentially identify the entity involved as well as the value of its activity. In the case of statistical dominance, the confidentiality disclosure rules applied may vary between compiling economies. For example, in some economies, if an entity contributes at least 75% of the value of the information to be disseminated then it is declared confidential, but other economies may use a dominance threshold of 80%. Alternatively, where two entities contribute, say, 90% or more of the value of the information being disseminated, it is declared confidential.

764. Whatever disclosure rules individual economies adopt, compilers in these economies have to suppress this confidential information in their dissemination of statistical results. However, suppression of primary confidential data is not sufficient on its own to prevent identification and disclosure of such confidential data. It is necessary for compilers to suppress additional data – secondary confidential data – in the results disseminated to prevent derivation of the primary confidential data. This operation is necessary within a single dissemination item (e.g., a single table of data), but it often has to be extended where multiple dissemination items exist (e.g., multi-tabular data that include data overlaps). As a result, a significant amount of data may need to be suppressed by compilers to cover both primary and secondary confidential data. It is important to stress that such suppression generally applies on a 'bottom-up' basis, i.e., at the greatest level of detail first and then proceeding to increasing levels of aggregation where necessary. It is important to stress that, in general, suppressions become less common in higher aggregates, provided the contribution of the entities whose data are suppressed in more detailed results then pass disclosure processes in higher aggregates.

765. As much as statistical confidentiality is very important, it should not be used in itself as a reason not to release information. Instead, the goal should be to maximise the dissemination of information as a public good for the wide range of users while still ensuring confidentiality obligations are met.

766. For researchers, alternative approaches allowing access to micro data for statistical purposes or research purposes should be considered, for example:

• Secure data labs to allow researchers to access and analyse micro data, whereby the research published does not reveal any confidential data.

• Signed data access agreements, which allow access to secure online data areas for specific research or analyses and with limited time.

11.4. Communication with suppliers

767. Statistical departments and agencies need to engage with data suppliers to ensure the timely and complete provision of data. Given that surveys are the primary data source for the compilation of FDI statistics in many economies, this section will discuss a few key elements of an effective engagement strategy with survey respondents.

768. One key element is minimising reporting burden. There are several different aspects to minimising reporting burden. One aspect is to make it as easy as possible for respondents to reply to the survey. This can mean providing more choice for respondents on how they can respond (e.g., paper, telephone data entry, secure electronic file transfer, etc.). It also means only asking for the information that is needed to compile the statistics and to minimise any duplication (i.e., not asking for the same information on several different surveys) unless it is absolutely necessary. Another element is to make it easier for respondents to get help in replying to the survey questionnaire, including such things as standard responses to common queries posted on the website, providing a phone number or email to respondents for assistance, and ensuring timely responses.

769. Another element of minimising the burden on reporters is to use language that respondents can understand. Given that most respondents to FDI surveys are businesses that are pulling data from their record-keeping systems, it is important to use financial accounting, rather than economic accounting, terms when possible. Including references to relevant financial accounting standards can also be helpful. When it is necessary to use economic accounting terms, it is important to define them in plain language. Illustrations can also be helpful in conveying the meaning; for example, in defining fellow enterprises, it can be useful to include an illustration of a simple MNE group with fellows identified. Data collectors also need to be aware of any changes to the business accounting rules, including changes to terminology, to ensure data collected remain valid.

770. It is also necessary to have effective communication with survey respondents. It is important to explain to respondents why their participation is required and the steps taken to reduce the burden on them. In addition, compilers should explain how they will keep suppliers’ information secure and confidential and value their time by making it as easy as possible to reply to the FDI questionnaires. Finally, it is important to listen and act upon feedback from survey respondents.

11.5. Revision policy

771. Revisions are an essential part of data compilation. Revisions occur as a consequence of the trade-off between the timeliness of published statistics and their reliability, accuracy and comprehensiveness. To address this trade-off, statistical departments and agencies often compile and disseminate provisional estimates that are then revised when new and more accurate source data become available. Although, in general, repeated revisions may be perceived as reflecting negatively on the reliability of official statistics, attempts to avoid them by producing accurate but very delayed estimates will result in failure to meet users’ needs for timely information. Regular communication with users, an established revision policy, and communicating well in advance of such updates or improvements will develop greater understanding why revisions occur and help to ameliorate negative reaction.

772. Economies are encouraged to develop a revision policy that includes a predetermined schedule (including not only when revised estimates will be released but the periods to be revised), reasonable stability from year to year, openness, advance notice of methodological changes, easy access for users to sufficiently long time series of revised statistics, and the adequate documentation of revisions in statistical publications and databases.

773. It is recommended that compilers take a pragmatic approach to revisions that fall outside the normal revision policy. For example, tax restitution or tax amnesties may affect FDI income in periods that are not subject to revisions. Economies should revise periods inside the revision policy time span. For periods that fall outside the normal revision policy, compilers should weigh the costs and benefits of revising FDI income for earlier periods in deciding whether to revise the earlier periods.

References

[1] 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.

[2] United Nations et al. (Forthcoming), 2025 System of National Accounts, United Nations et al., New York.

Annex 11.A. Responses to common user questions

774. There are some common questions that users have about FDI statistics. The following are some examples of these questions with possible responses that compilers might consider when replying to their users.

What is included in foreign direct investment?

775. Direct investment is a category of cross-border investment made by a resident in one economy (the direct investor or parent) with the objective of establishing a lasting interest in an enterprise (the direct investment enterprise or affiliate) that is resident in an economy other than that of the direct investor.

776. The motivation of the direct investor is a strategic long-term relationship with the direct investment enterprise to ensure a significant degree of influence by the direct investor in the management of the direct investment enterprise. The “lasting interest” and significant degree of influence is evidenced when the direct investor owns at least 10% of the voting power of the direct investment enterprise.

777. Direct investment may also allow the direct investor to gain access to the economy of the direct investment enterprise which it might otherwise be unable to do. The objectives of direct investment are different from those of portfolio investment whereby investors do not generally expect to influence the management of the enterprise.

What does FDI income measure?

778. Direct investment income is part of the return on the direct investment position; that is, the return on equity and debt investment. FDI income payments measure the total return within a specific period on direct investment positions paid by enterprises in the reporting economy to their foreign direct investors. FDI income receipts measure the total return within a specific period on direct investment positions received by investors in the reporting economy from their direct investment enterprises abroad.

779. The return on equity consists of the direct investor’s share of the earnings of the direct investment enterprise. The earnings are measured according to the current operating performance concept, which measures earnings from normal business operations and excludes non-recurring items (such as write-offs) and holding gains and losses. Income on debt measures interest on inter-company debt between direct investors and direct investment enterprises and between fellow enterprises.

What is the relationship between FDI transactions, positions (stocks) and income?

780. FDI financial transactions refer to those cross-border transactions which qualify as direct investment recorded during the reference period (year, quarter, or month). FDI positions represent the value of the stock of direct investment held at the end of the reference period (year, quarter, or month). Direct investment positions are affected not only by financial transactions recorded prior to and during the period but also by other changes in price, exchange rates, and volume. FDI income data, closely linked to the stocks of investment, are used for analysis of the productivity of the investment and are also used as part of the calculation of the rate of return on the total funds invested.

How to interpret negative values for FDI transactions and positions?

781. Negative values in transactions may indicate disinvestment in assets or discharges of liabilities. In the case of equity, the direct investor may sell all or part of the equity held in the direct investment enterprise to a third party; or the direct investment enterprise may buy back its shares from the direct investor thereby reducing or eliminating its associated liability. If the financial transaction is in debt instruments between the direct investor and the direct investment enterprise, it may be due to the advance and redemption of inter-company loans or movements in short term trade credit. Negative reinvested earnings indicate that, for the reference period under review, the dividends paid out by the direct investment enterprise are higher than current income recorded (if that is the decision of the board of managers) or that the direct investment enterprise is operating at a loss.

782. Negative outward investment resulting from the directional principle applied to fellow enterprises relates to disinvestment by the resident ultimate controlling parent who receives funds from non-resident subsidiaries. Negative inward investment represents net claims of a resident subsidiary (who has a non-resident ultimate controlling parent) on non-resident fellow enterprises.

783. The changes in FDI positions are affected by the accumulated flows and hence may also result in negative values for the aggregate position, but mainly for other capital (e.g., when the loans from the direct investment enterprise to the parent exceed the loans – or even the original capital – given by the parent to the direct investment enterprise; it could be the case where conduits or treasury companies are involved). This is particularly important when FDI statistics are presented on a directional basis given the significance of inter-affiliate transactions and positions in debt. A negative equity position could also result if there were negative reinvested earnings in the direct investment enterprise such that the accumulated negative reinvested earnings exceeded the accumulated other equity.

Why are there such large revisions in FDI statistics?

784. To serve users requiring timely data, the compilers of FDI statistics make initial estimates for some components on preliminary information available at that time. It is especially difficult to have timely and at the same time complete data on the accrued income when the results are first published, as these values are only known once the direct investment enterprise has closed its books and the current operating income and earnings distributions have been determined for the reference period.

785. Consequently, when more complete data are provided for the reference period, data compilers revise the estimates for the previous periods. However, compilers are recommended to provide to the public at large a description of the revisions.

Why are large FDI transactions concentrated in particular countries – i.e., what is the role of pass-through funds and/or passing through special purpose entities (SPEs)?

786. Multinational enterprises frequently use SPEs for their inward and outward direct investment, including the large amounts of pass-through funds, passing through entities in jurisdictions which may offer advantages, e.g., for tax purposes. Even though transactions/positions with SPEs are included in assets and liabilities of a direct investment enterprise, they are separately identified in aggregate FDI statistics and in the statistics disaggregated by partner economy or by industry. In other words, users can exclude transactions and positions with resident SPEs in FDI statistics. The purpose of excluding resident SPEs is to reduce the overstatement of FDI statistics and to provide more realistic analysis and hence estimates of real FDI in an economy.

How complementary are FDI statistics and the statistics on the activities of multinational enterprises (AMNE) and foreign affiliate statistics (FATS)?

787. The statistics on direct investment include cross-border investment made with the objective of establishing a lasting interest to exercise an influence in the management of the direct investment enterprise (the target entity) which is evidenced by the ownership of at least 10% of the voting power by the direct investor. Control by the direct investor of the direct investment enterprise is not necessary to qualify as FDI. The framework for direct investment relationship (FDIR) establishes the chains of relationships between direct investors and their direct investment enterprises and includes qualifying cross-chain relationships between direct investment enterprises. As such, FDI statistics compiled using the FDIR include certain transactions and positions between fellow enterprises which are not linked through a direct 10% or more ownership of voting power. FDI statistics only record the financial value of the investment. The geographical allocation is based on the debtor/creditor principle excluding resident special purpose entities.

788. The statistics on AMNE/FATS cover affiliates which are controlled by an enterprise resident in another economy. In principle, AMNE data cover a sub-set of the entities involved in FDI. The AMNE/FATS present detailed data on the foreign affiliates, e.g., on the employment, turnover, value added, etc. Nevertheless, the geographical attribution of the units is based on the ultimate controlling (investing) economy (for inward investment) or the ultimate hosting economy (for the outward data).

789. In theory, FDI and AMNE/FATS statistics are very closely related and, therefore, complementary. However, methodologies applied to FDI and AMNE/FATS statistics are not consistent and render difficult the analysis of financial and income transactions and positions along with the economic impact of the multinational enterprises covered by AMNE/FATS. Inward FDI positions allocated to the ultimate investing economy reduces these discrepancies to the extent that samples of enterprises are also harmonised.

Why are there discrepancies (sometimes significant) in the data provided by international organisations?

790. Several International organisations compile and disseminate FDI data. These include the OECD, Eurostat, the European Central Bank (ECB), the International Monetary Fund (IMF), and the United Nations Conference on Trade and Development (UNCTAD). While international organisations have taken steps to align their reporting frameworks, differences persist. FDI statistics of the OECD and Eurostat are essentially based on a common framework focusing on the directional principle and detailed FDI statistics. IMF and ECB compile and disseminate FDI as a functional category of balance of payments and international investment position. Data regarding the euro area are consistent between Eurostat and the ECB. OECD and IMF research demonstrated that the main differences between their aggregate FDI statistics are largely due to the timing of revisions. UNCTAD data are in a number of cases different due to adjustments and different data sources.

Why are there such large asymmetries in bilateral FDI statistics? That is, why does the outward FDI to country B reported by country A not equal the inward investment from country A reported by country B?

791. These asymmetries arise from the general difficulty in capturing data from very large and complicated multinational enterprises (MNEs). Other issues, such as differences in valuation, definitions, and methods can also contribute. Bilateral FDI statistics are classified by the immediate investing economy. The immediate investing economy is the economy that directly holds the direct investment enterprise. MNEs often have very complicated ownership structures with several layers of direct investment enterprises. As a result, there can be one directly held direct investment enterprise with dozens of indirectly held direct investment enterprises below it and these indirectly held direct investment enterprises can be in many different economies. These complicated structures make it very difficult for compilers to correctly identify the directly held direct investment enterprise and correctly attribute the data to an economy. This is made more complicated by the fact that MNEs can easily change their structures.

792. These issues are particularly problematic for economies that are hosts to a large number of special purpose entities (SPEs). SPEs are entities that have little employment, operations, or physical presence in an economy but that are used to raise capital or hold assets or liabilities. Examples of SPEs include shell companies, holding companies, and conduits. These entities often are the recipients of large FDI financial transactions that are then invested in other economies where the MNE operates. These entities can have large FDI positions, but those positions are in turn held outside of the economy. While compilers strive to cover resident SPEs in their FDI statistics, it can be difficult, leading to potentially large differences in the bilateral figures.

793. Differences in valuation can also cause asymmetries, especially for positions. Because there is often no market price for the direct investment asset, an estimate has to be made to value the FDI position at current period, or market, prices. Different methods, data sources, and assumptions can lead to differences in these estimates of FDI positions at market value.

794. Another source of asymmetries can be differences in the samples of the populations that are surveyed. If one partner economy does not survey the enterprise, it will have to estimate the values for that enterprise. This estimate will likely differ from the data reported by that enterprise to the other partner economy.

795. Finally, bilateral FDI statistics are often presented according to the directional principle, and there is a built-in asymmetry in the directional principle. Fellow enterprises are entities that have no direct investment relationship themselves but are relevant to FDI statistics because they have a direct investor in common. Transactions and positions between fellow enterprises are recorded according to the residence of the ultimate controlling parent of the fellow enterprises. For example, assume enterprises A in economy 1 and B in economy 2 are fellow enterprises with an ultimate controlling parent C in economy 3. If A makes a loan to B, this is recorded as positive inward direct investment in B from A. However, rather than recording this as outward direct investment from A to B as would be required for symmetry, it is instead recorded as negative inward direct investment in A from B because the provision of the loan from A to B did not give A any influence over B as would be required for it to be outward direct investment. Rather, the influence remains with the ultimate controlling parent, C.