Current IMF guidelines on balance of payments statistics can be found in the Balance of Payments Manual,fifth edition, (IMF, 1993). According to the Manual, "direct investment is the category of internationalinvestment that reflects the objective of a resident entity in one economy obtaining a lasting interest in anenterprise resident in another economy [paragraph 359]…the primary distinguishing feature of directinvestment [as compared to other balance of payments capital flows] is the significant influence that givesthe investor an effective voice in management [paragraph 360]." The general rule of thumb presented in theManual is that the direct investor owns (or controls) at least 10 percent of the ordinary shares, votingpower, or equivalent [paragraph 362]. The Manual, however, allows (although it does not recommend for the sakeof international comparability) countries to choose other criteria such that the direct investor has aneffective voice in management [paragraph 363].
Components of direct investment "are equity capital, reinvested earnings, and other capital associatedwith various intercompany debt transactions. Equity capital comprises equity in branches, all shares insubsidiaries and associates…, and other capital contributions. Reinvested earnings consist of the directinvestor’s share (in proportion to direct equity participation) of earnings not distributed as dividends bysubsidiaries or associates and earnings of branches not remitted to the direct investor. [paragraph 369]…Otherdirect investment capital (or intercompany debt transactions) covers the borrowing and lending offunds-including debt securities and supplier’s credits-between direct investors and subsidiaries, branches,and associates [paragraph 370]." Related to reinvested earnings, it should be noted that an offsetting entry(with opposite sign) should be included in current account flows, recorded under direct investment income.
Is this definition followed around the world as the standard international reporting practice for FDIinflows?
Our understanding is that the IMF guidelines are generally followed by industrial countries. Because ofincomplete information, however, the guidelines are not completely followed by many developing countries,particularly, related to reinvested earnings.
Is it true that India does not conform to IMF standards when it comes to FDI statistics?
Along with Mr. Pfeffermann of the IFC, the Department of Industrial Promotion and Policy (DIPP) of theMinistry of Commerce and Industry (India) has argued recently that FDI is significantly underestimated inIndia. DIPP estimates that actual FDI flows to India amount to $8 billion a year, compared to $2 to $3½billion in recent years in the official balance of payment statistics (from the Reserve Bank of India). Mr.Pfeffermann appears to be using this estimate. DIPP notes that FDI statistics in India do not includereinvested earnings (or undistributed profits) of foreign companies, subordinate debt, overseas commercialborrowings, and ADR/GDR issues. DIPP believes that ADR/GDR issues should be included as FDI instead ofportfolio flows in India because these inflows under Indian law are more akin to convertible instruments withequity transfer built in the instrument. DIPP contends that a substantial portion of the underestimate relatesto FDI statistics not including reinvested earnings. It estimates that the rate of return of foreign capitalis about 40 percent, and most of this return is reinvested in India.
The Reserve Bank of India (RBI) acknowledges that FDI may be underestimated in India’s balance ofpayments statistics. Data used in compiling the balance-of-payments are a by-product of reporting on foreignexchange transactions and are only partly complete. In particular, information on reinvested earnings flowsare obtained on a voluntary basis, and the data are slow in being sent to the RBI and only partial incoverage. The response rate to the voluntary survey is about 60 percent (of direct investment enterprises)after two to three years. The RBI staff, however, believes that reinvested earnings have declined in recentyears. The RBI does not fully agree with DIPP’s position concerning subordinate debt, overseas commercialborrowings, and ADR/GDR issues, although it does concede that some inter-company flows (between directinvestors and subsidiaries, branches, and associates) recorded as subordinate debt and overseas commercialborrowing should be recorded as FDI.
Do you agree with Pfeffermann's observations?
The IFC report claims that FDI flows to China (which amount to about $40 billion annually) are overstatedby 50 percent due to round tripping. Other studies, however, estimate that round tripping accounts for only 5to 25 percent of FDI flows to China. The source for the IFC claim is the World Bank’s 2002 GlobalDevelopment Finance report. The report suggests that a large part of FDI to China that originates in Hong Kongand the Virgin Islands is in fact round tripping. To reach the 50 percent estimate of round tripping, the IFCappears to have assumed all FDI from Hong Kong and the Virgin Islands is round tripping. This assumption isextreme, as, for example, many Taiwanese companies invest in Mainland China through Hong Kong. This is notround tripping, and in the Chinese statistics, these investments are recorded as originating in Hong Kong,instead of Taiwan.
What are the key weaknesses in the Chinese structure of governance that allows such "fudging"of figures?
China has recently started to participate in the IMF’s General Data Dissemination System (GDDS), and theauthorities are working to improve their economic statistics.
(A condensed form of the above appeared in the print magazine)
