6. Country Practices in the Implementation of International Recommendations for FDI Statistics
- Neil Patterson, Marie Montanjees, Colleen Cardillo, and John Motala
- Published Date:
- September 2004
Key Findings of the 2001 SIMSDI Update
6.1 This chapter summarizes the findings from the 2001 update of the Survey of Implementation of Methodological Standards for Direct Investment (SIMSDI).1 Highlights of the results are provided in Box 6.1.
Box 6.1.Highlights of the 2001 SIMSDI Results
Areas where there have been marked improvements since 1997
Availability of FDI statistics, particularly:
—Income data (including reinvested earnings)
—Geographic and industrial sector breakdowns
Coverage of FDI statistics, particularly the inclusion of:
—Noncash acquisitions of equity
—Intercompany loans and financial leases
—Real estate owned by nonresidents
—Activities of special purpose entities (SPEs)
—Activities of offshore enterprises in the outward FDI statistics
—Expenditure on natural resource exploration
Areas where more than 75 percent of countries surveyed follow the international recommendations applicable to their economy
Use of the 10 percent ownership rule as the basic criterion for defining FDI relationships
Equity capital transactions between affiliated banks and between affiliated financial intermediaries
Recording of reverse investment equity transactions when two FDI relationships have been established
Inclusion of data on real estate owned by nonresidents
Inclusion of data on activities of SPEs
Inclusion of data on activities of offshore enterprises
Areas where, despite improvements, the majority of countries do not yet follow the international recommendations
Inclusion of activities of indirectly owned direct investment enterprises
Use of the Current Operating Performance Concept to measure direct investment earnings
Time of recording FDI income on equity (when payable or declared payable) and income on debt (when accrued)
Recording of reverse investment transactions when the FDI relationship is in one direction only
Inclusion of activities involving construction enterprises and mobile equipment
Valuation of FDI positions (assets and liabilities)
6.2 The results showed that there have been marked improvements in the availability of FDI statistics in the four years between the 1997 SIMSDI survey and the 2001 update—in particular, for inward and outward position data (an additional 13 and 14 countries, respectively) and for inward and outward FDI financial flows (an additional 9 and 10 countries, respectively). In addition, there have been improvements in the reporting of data on FDI income, with an additional 10 countries reporting data on inward FDI income on equity, an additional 11 reporting data on inward reinvested earnings, and an additional 10 reporting data on inward FDI income on debt.
6.3 During the same period, there was also a significant increase in the compilation of data showing geographic breakdowns, particularly for the inward and outward FDI financial flows data (an additional 11 and 13 countries, respectively) and for the position data (an additional 11 and 12 countries, respectively). Similar increases were seen in the number of countries compiling data showing breakdowns by industrial sectors.
6.4 The four years between the 1997 SIMSDI survey and the 2001 update also saw significant improvements in the items covered by the FDI statistics, with a marked increase in the number of countries now following the international recommendations in a number of areas. For the data on equity capital, increases were seen in the number of countries that include noncash acquisitions of equity (such as through the provision of capital equipment) and real estate owned by nonresidents; for the data on other capital, significantly more countries now include intercompany loans, both short-term and long-term, and financial leases. Other areas where the coverage of the data improved markedly were the inclusion of activities of SPEs in both the inward and outward FDI transactions data, the inclusion of relevant activities of offshore enterprises in the outward FDI transactions data, and the inclusion of expenditure on natural resources exploration in both the inward and outward FDI transactions data.
6.5 The results of the 2001 SIMSDI update also showed that there are now a number of areas where more than three-fourths of the 61 countries that participated follow the international recommendations applicable for their circumstances—namely, the use of the 10 percent ownership rule as the basic criterion for defining direct investment enterprises and direct investors, the inclusion of equity capital between affiliated banks and between affiliated financial intermediaries, the correct recording of reverse investment equity transactions when two FDI relationships have been established, the inclusion of purchases and sales of real estate by nonresidents, and the inclusion of data on activities of SPEs and offshore enterprises.
6.6 However, despite these improvements in the implementation of the international recommendations since the 1997 SIMSDI survey, the 2001 update indicated that there are still a number of aspects of these recommendations that are not yet followed by the majority of the 61 countries that participated in the update:
Only 11 countries fully apply the recommendations regarding the inclusion of indirectly owned direct investment enterprises for their inward FDI transactions data—the same number as in 1997.
Only 19 countries fully apply the recommended Current Operating Performance Concept for measuring their direct investment earnings in their inward FDI statistics.
Only 22 countries record income on equity (dividends and distributed branch profits) in their inward and outward FDI transactions data at the time they are payable, as recommended; only 25 record income on debt (interest) as it is accruing for their inward data, and 22 do so for their outward data.
Few countries correctly classify reverse investments when the FDI relationship is in one direction only (17 countries in the case of reverse investment involving the acquisition of equity by the direct investment enterprise in its direct investor and 25 in cases involving the provision of loans).
Only 23 countries include in their inward FDI transactions data the activities of quasi-corporations involving construction enterprises, and even fewer include the activities of quasi-corporations involving mobile equipment, such as ships, aircraft, and drilling rigs.
Few countries value their FDI asset and liability positions at market prices (only 21 countries value their inward equity capital positions at market prices, and even fewer value their inward other capital positions, and outward equity capital and other capital positions, at market prices).
6.7 The following section addresses in more detail some of these aspects, focusing in particular on the FDI items discussed earlier, which are primarily the same items that the Godeaux Report (IMF, 1992) identified as being important contributors to the discrepancies in the global data on FDI capital flows.
Selected Recording Issues
6.8 Both industrial and developing countries have difficulties in implementing fully the international recommendations for FDI statistics, which require adequate resources to conduct surveys, well-trained staff, and a high level of cooperation from private sector respondents. Private sector respondents, in turn, may also have difficulties in reporting data that conform to the international statistical recommendations because, for example, of countries’ accounting or taxation regulations and/or a misunderstanding of the reporting requirements.
Direct investment threshold
6.9 On a positive note, the 2001 SIMSDI update showed that 90 percent of the survey respondents (55 countries, including all OECD countries except Turkey) use the 10 percent ownership threshold as their basic criterion in recording inward FDI transactions. However, contrary to the international recommendations, about one-third of these countries, including an almost equal number of OECD and non-OECD countries, also employ additional criteria in making a final determination of an FDI transaction or relationship. This might include—as in Argentina, Belgium, Botswana, Israel, Korea, Mexico, the Netherlands, Nigeria, Norway, and Portugal—recognizing a direct investment relationship in cases where the nonresident direct investor owns less than 10 percent of the voting shares but has an effective voice in the management of the enterprise. Other deviations from the 10 percent ownership criterion can be the application of a value threshold or differing treatments of incorporated and unincorporated enterprises. For example, one G-7 country (Italy) treats all unincorporated enterprises with foreign ownership as FDI, regardless of the percentage owned by nonresidents. Moreover, in some cases the definitions employed by countries for inward and outward FDI may differ.
6.10 Such differences result in an inconsistent classification of FDI flows and stocks by countries. However, the Godeaux Report (IMF, 1992) concluded that the use of different ownership thresholds likely did not contribute significantly to global asymmetries, because a very high proportion of direct investment occurs in branches or in majority-owned subsidiaries, which both the investing and host countries would regard as direct investment enterprises.
Measurement of earnings
6.11 The BPM5 and the OECD Benchmark Definition recommend the use of the Current Operating Performance Concept to measure direct investment earnings, which defines the earnings of an enterprise as income from normal operations before nonrecurring items (for example, write-offs), and any realized or unrealized capital gains and losses and realized or unrealized foreign exchange rate gains and losses are accounted for. Operational earnings of the direct investment enterprise should be reported after provisions for depreciation and income and corporation tax charged on these earnings have been deducted. The statistical recommendations also call for the inclusion of reinvested earnings data of indirectly owned direct investment enterprises.
6.12 The 2001 SIMSDI update showed that 19 countries (31 percent of the 61 surveyed) fully apply the Current Operating Performance Concept in the measurement of inward direct investment earnings. This was more than double the number of countries that indicated in the 1997 survey that they followed this practice. Eight of the countries that reported implementing the BPM5 recommendation were OECD members, including two G-7 countries—the United Kingdom and the United States.
6.13 In many countries, enterprises use the “All-Inclusive Concept” of earnings measurement, under which earnings are the amount remaining after all items (including capital and exchange rate gains/losses and write-offs) that cause a change in shareholders’ or investors’ interests during the period are allowed for. According to the 2001 SIMSDI update, around half of the surveyed countries exclude exchange rate gains/losses, write-offs, and realized capital gains/losses, while approximately two-thirds exclude unrealized capital gains and losses. Also, approximately a quarter of the countries reported that the earnings measures do not include deductions for depreciation or make provision for host-country income/corporation taxes, as called for under the Current Operating Performance Concept for earnings measurement.
6.14 The different concepts that countries use in measuring operating earnings for determining rein-vested earnings, and the varying coverage of earnings of indirectly owned direct investment enterprises, are likely to be significant contributors to the asymmetries in bilateral and global data on direct investment. More generally, the 2001 SIMSDI update showed that some countries compile estimates of reinvested earnings on an annual basis and then divide this amount by four to derive quarterly estimates. The discussion of imbalances in global FDI income flows in Chapter 4 showed a persistent excess of credits of reinvested earnings from direct investment abroad in comparison with the debits recorded by the economies of the direct investment enterprises.
Indirectly owned direct investment enterprises
6.15 While compilers are generally able to record and correctly classify transactions between the direct investor and the directly owned direct investment enterprise, extending the coverage to include indirectly owned direct investment enterprises, as discussed in Box 5.1, is difficult to implement. Only 11 (18 percent) of the surveyed countries could fully apply the recommended recording treatment for their inward FDI transactions data (no change from 1997),2 while another 28 (46 percent) countries indicated that they were able to partially apply the treatment. The findings were similar for outward FDI transactions and for the reporting of FDI stock data, except that fewer countries partially apply the recommended recording treatment for stock data.
6.16 Canada was the only G-7 country that indicated that it was able to fully apply the recommended treatment of including indirectly owned direct investment enterprises for inward and outward FDI transactions. The other countries that indicated that they fully apply the treatment to both their inward and outward FDI transactions were Argentina, Australia, Botswana, Denmark, Estonia, Iceland, Ireland, Norway, South Africa, and Sweden.
6.17 The issue of indirectly owned enterprises is being reassessed in the upcoming revision of BPM5, with the aim of possibly simplifying the present international recommendations, which many countries have found difficult to apply or explain to survey respondents. Options being considered are to limit the indirectly owned enterprises included in the direct investment relationship to either those in which the direct investor has an indirect ownership of 10 percent or more (the so-called U.S. methodology) or to subsidiaries of the directly owned enterprise, regardless of the percentage indirectly owned by the direct investor (a proposal favored by a number of European countries).
6.18 Reverse investment occurs when a direct investment enterprise has acquired a financial claim on its direct investor, through the acquisition of shares in its direct investor or through the provision of a loan to its direct investor (see Box 5.1). Only 17 (35 percent) of the 49 surveyed countries for which reverse investment was known to be applicable record an increase in equity claims on direct investors of less than 10 percent of the total equity of the direct investor as a reduction in FDI in the reporting economy (an increase in equity claims on direct investors), in accordance with the international statistical recommendations. Approximately equal numbers of OECD and non-OECD countries reported that they apply the recommended treatment; Japan was the only G-7 country that did so. Twenty-three of the countries that do not implement the BPM5 treatment record the equity investment in the foreign direct investor as part of portfolio investment assets, and a number of others record it as FDI abroad.
6.19 In the case of reverse investment involving the provision of a loan by the direct investment enterprise to its direct investor, when the direct investment enterprise owns less than 10 percent of the total equity of its direct investor, the situation was somewhat better: 25 countries reported that they record a reduction in FDI in the reporting economy (an increase in loan claims on direct investors). These countries include three G-7 countries—Germany, Japan, and the United States—and 13 other OECD countries. Thirteen of the 23 countries that do not apply the BPM5 treatment, including one G-7 country (the United Kingdom) and two other OECD countries (Hungary and Korea), recorded the loan transactions under other investment assets—loans (that is, not as FDI).
Special purpose entities
6.20 The 2001 SIMSDI update showed that most of the 40 countries for which SPEs were known to be applicable include SPEs in their inward direct investment statistics; only five of the surveyed countries for which SPEs were applicable (including Bolivia, Malaysia, the Netherlands, and Tunisia) do not include the activities of SPEs in their data on inward FDI transactions. Six exclude them from their inward position data, including Bolivia, Luxembourg, Malaysia, the Netherlands, and Tunisia.
6.21 In the case of SPEs that have the sole purpose of financial intermediation, the BPM5 and the OECD Benchmark Definition recommend that transactions with affiliated banks and affiliated financial intermediaries, except for permanent debt and equity capital, be excluded from the FDI data.3 The 2001 SIMSDI update indicated that five countries—four OECD countries (Belgium, Finland, France, and Germany) and one non-OECD country (Argentina) did not follow this recommendation for their inward and outward FDI transactions and position data. In addition, two OECD countries (Canada and the Netherlands) did not apply the recommended treatment for their outward transactions and position data, and one non-OECD country (Chile) did not apply it for its inward and outward transactions data.
6.22 The international statistical methodology recommends that, for transactions between affiliated banks and between affiliated financial intermediaries, only those transactions associated with permanent debt and equity capital should be recorded as direct investment. Deposits, loans, and other claims and liabilities related to the usual banking activities between affiliated banks, and claims and liabilities related to usual financial intermediation activities between financial intermediaries, should be excluded from FDI capital and instead be classified under portfolio investment or other investment in the balance of payments statistics, as appropriate. The 2001 SIMSDI update indicated that only two countries—Costa Rica and Guatemala—did not exclude usual banking activities between affiliated banks from their inward FDI, and nine countries did not exclude claims and liabilities related to usual financial intermediation activities between affiliated financial intermediaries from their inward FDI data (Belgium, Costa Rica, Finland, Germany, Guatemala, Hungary, Indonesia, Kazakhstan, and Korea—for Korea, long-term loans only).
Valuation of FDI stocks
6.23 Although the BPM5 recommends that market prices be used as the basis of valuation for stocks, it recognizes that in practice there may be some departures from the market price principle in the recording of direct investment, where book values from the balance sheets of direct investment enterprises (or of direct investors) are often used to determine the value of the stock of FDI. These balance sheet values, if recorded on the basis of current market value, would be in general accordance with the principle. If based on historical cost or on an interim but not current revaluation, such balance sheet values would not conform to the recommended valuation principle.
6.24 The 2001 SIMSDI update showed that, of the 51 reporting countries that compiled data on inward FDI equity positions, 21 countries (41 percent) reported using market values as the primary method of valuation of their inward FDI equity capital positions; 36 countries used book valuations; and some countries used a mixture of market and book valuations. (Only eight countries reported using market valuations in the 1997 SIMSDI.) Approximately equal numbers of OECD and non-OECD economies used market valuations for their inward FDI equity capital positions. The OECD economies were Australia, Austria, Belgium, France, Italy, Mexico, New Zealand, the Slovak Republic, Sweden, and the United States; the non-OECD economies were Botswana, Croatia, Estonia, Hong Kong SAR, Israel, Kazakhstan, Malaysia, Russia, Singapore, South Africa, and Thailand.4 In the case of Israel and Singapore, only listed companies with direct investment were recorded at market values, and unlisted companies were recorded at book values. South Africa was also able to record only part of the equity capital at market values. Valuation practices for outward equity positions were broadly similar.
6.25 Notwithstanding the increase in the number of countries employing market prices to value direct investment, this is an area that is difficult to measure when company shares are not listed on stock exchanges.5
6.26 In view of the distortions in statistics created by the wide range of valuation practices applied by European Union member states for the compilation of FDI equity stocks in IIP statistics, a joint ECB/Eurostat Task Force on Foreign Direct Investment recently made a proposal to value euro-area inward and outward FDI equity stocks by valuing (i) FDI in listed companies’ shares on the basis of stock exchange prices and (ii) FDI in nonlisted companies’ shares on the basis of net asset values, assuming a lack of any appropriate market reference for these companies.6
6.27 In most cases, it is not possible to assess the impact of the above recording practices on the data of individual reporting countries or on the global FDI statistics published by the IMF and UNCTAD. Moreover, it is also difficult to assess whether countries that indicated adherence to a certain recording treatment are indeed able to fully implement the treatment in practice. For example, in the case of reverse investment involving loans to direct investors, one-third of the 25 countries that indicated that they implement the recommended treatment did not report such data to the IMF in their balance of payments submissions. The 2001 SIMSDI update indicated that the data sources used by many countries are inadequate for the compilation of FDI statistics that are fully consistent with the recommended methodology. Moreover, because of the dominance of the industrial countries in FDI, it would seem that methodological improvements in these countries would be needed to narrow the sizable discrepancies that exist in the global FDI data.
See Foreign Direct Investment Statistics: How Countries Measure FDI, 2001 (IMF and OECD, 2003).
To be considered in full compliance, a country needed to include in its FDI statistics (i) the earnings data of indirectly owned direct investment enterprises and (ii) all equity capital and other capital transactions within the group of related enterprises, regardless of the percentage ownership held by the related enterprises in each other.
In 2002, the international methodology was changed to cover SPEs with the “primary” function of financial intermediation, rather than just those with the “sole” function of financial intermediation as stated in BPM5.
Croatia and Malaysia compile, but do not disseminate, FDI data on market values. Two other countries—Ecuador and Ireland—that use book valuations as the primary method for valuing inward FDI equity capital reported that they also compile market value data on inward FDI but do not disseminate them.
The issue of how countries compile market values for unlisted enterprises will be examined in the 2003 SIMSDI survey, which will be undertaken in 2004, with the results expected to be published in 2005.
The task force also recommended that inward and outward FDI equity stocks be reported to the ECB and Eurostat with a split between listed and nonlisted FDI companies, and FDI stocks in listed companies’ shares be reported on the basis of both market values and book values. The task force also expressed the opinion that the compilation of FDI stocks should be based on information collected via FDI surveys and that the provision of annual FDI stocks based on the accumulation of balance of payments flows should be discontinued as soon as possible. The proposal calls for data for FDI stocks with the recommended valuation to be sent to the ECB and Eurostat for the first time in September 2006 in reference to end-2004 and end-2005 stocks; this timetable will be reviewed.