FDI, AN ACRONYM OF YES, MAYBE AND UNCERTAINTIES
by Chakravarthi Raghavan
Geneva, 20 Sep 99 -- The current orthodoxy about Foreign Direct Investment as a NON-DEBT CREATING AND MORE STABLE FORM OF CAPITAL FLOW, LIKELY TO OFFER UNPRECEDENTED GROWTH OPPORTUNITIES FOR DEVELOPING COUNTRIES AND TECHNOLOGY AND SKILL TRANSFERS AND PRODUCTIVITY UPGRADING, is questioned in UNCTAD's Trade and Development Report 1999.
TDR-99 notes that with the decline in official financing and instability of private financial flows, FDI has come to be increasingly seen as a solution to the problems of resource gap and external financing. The significant upturn in FDI flows to developing countries in the 1990s, the success of some in attracting large amounts of FDI, and the 'relative stability' of FDI in the aftermath of the East Asia crisis, TDR-99 notes, have led to predictions that FDI is a more stable form of capital flow linked to an emerging international production system, and likely to offer new and, on some accounts, unprecedented growth opportunities to developing countries.
But on a closer examination of recent inflows, including their size, distribution, sources and utilization, and an assessment of their contribution to balance of payments, TDR-99 presents a more sceptic assessment.
The report notes that average annual growth of FDI inflows has risen from 15% in the 1980s to 28% in 1990s - reflecting in part structural and policy changes in potential southern hosts. And the share of world FDI received by developing countries has risen from around 16% in second half of 1980s to 28% in 1990s.
But, it notes, these flows are increasingly concentrated in a small number of locations - with 10 leading emerging market economies accounting for three-quarters of all FDI flows to the South; and with China, Brazil and Mexico alone accounting for almost one-half of the total flow. Even adjusting the flows to per capita basis does little to diminish their highly uneven distribution.
"The recent FDI-led integration" of developing countries is thus a highly selective process, TDR-99 says.
Data about FDI are not always up to date and reliable. FDI flows are usually distinguished under three different sources: equity capital, undistributed profits, and loans from parent companies to affiliates. Empirically, equity capital is defined as "investment made to acquire a lasting management interest (usually atleast 10 percent of voting stock) in an enterprise operating in a country other than that of the investor, whether in a new or existing firm. Acquisitions below 10% are considered portfolio investment.
TDR-99 points out these are somewhat arbitrary definitions and raise conceptual problems - nor do the definitions always reflect accurately national practices in classifying FDI.
Thus defined, while total FDI inflows into developing countries have been growing steadily, the total portfolio equity inflows have shown considerable instability since their emergence in mid-1980s.
Undistributed profits are a relatively important proportion of measurable FDI flows - though the share of re-invested earnings in total FDI in developing countries has been declining in favour of equity and loans during 1990s.
For e.g., more than half the total outflow of what is labelled FDI from the US consists of earnings of foreign subsidiaries retained by them and not remitted to the US. Some studies have questioned the validity of contending that such FDI is functionally indistinguishable from fresh capital inflows, as if these reinvested earnings are a flow of foreign resources crossing the borders of two countries.
TDR-99 says that clearly such reinvestments do not involve any cross-border flows and their treatment as a one-way inflow item in the capital account of the BOP is not admissible. But the difficulty is resolved by showing the profit as a payment abroad on current account, and the reinvested profits as a FDI inflow on capital account.
However, existing statistical measures cannot always distinguish between use of retained earnings for investment in equity capital, on the one hand, and acquisition of other financial assets such as government bonds, on the other.
Together with many other changes in global financial markets that have facilitated capital mobility such features of FDI make it difficult to evaluate its stability.
Another issue that has a bearing on stability and sustainability, and the impact on BOP of FDI relates to whether the inflows are utilized for mergers and acquisitions (M&A) or socalled greenfield investment.
While conventional analysis has usually treated FDI as if it consists essentially of additions to real capital stock of the host country, and though long-term considerations may play a role, M&A may also be greatly influenced by prospects of quick capital gains, particularly during periods of crisis.
In practice, TDR-99 points out, data do not allow independent estimation of these two components of FDI. But since data on cross-border M&A are available, greenfield investments can be estimated as a "residual" - difference between total FDI and M&A.
Cross-border M&A has accounted for between one-half and two-thirds of total world FDI flows - a figure that is higher for developed countries and lesser for developing, probably main because of the smaller role for M&A in China.
If China is excluded, the share of M&A in cumulative FDI in 1992-1997 turns out to be 72 percent, up from 22 during 1988-1991.
And when the residual is treated as greenfield component of FDI, its absolute annual level during 1992-1997 has been consistently below the level reached in 1991.
"Thus," says TDR-99, "the recent boom in FDI flows to developing countries has, with the exception of China, consisted predominantly of M&A, largely in the services sector."
The surge in M&A, especially in the services sector, is closely linked to privatization programmes adopted in 1990s in which TNCs have played a prominent role. According to World Bank data, during 1990-1997, privatization via foreign investors amounted to 12% of total FDI inflows into the developing world, except China.
Owing to consequent, frequently drastic, falls of exchange rate and declines in asset prices, financial crises in emerging markets have also created opportunities for highly profitable M&A. In 1998, for e.g., while total FDI flows to the five Asian countries affected by the crisis declined by $1.5 billion, cross-border M&A in those countries is estimated to have risen to more than $3 billion.
The shift of FDI towards acquisition of ownership rights over existing assets is associated with changes in sectoral composition. FDI in the primary sector has been declining for some time, though still a relatively more important component of total FDI in developing than in developed. The secondary sector still accounts for the largest share of total FDI in many developing countries, but the pattern in the 1990s has definitely involved a shift away from manufacturing to tertiary sectors where much M&A activity has taken place.
There has been a rising share of non-tradable sectors in total FDI flows.
But clearly stocks of assets of host country, in public and private sectors, for M&A is a finite process, and its limits are even tighter in poorer countries. FDI attracted by privatization of public assets is a one-off nature. As for FDI involving acquisition of existing private equity, the presence of non-residents in stock markets of many emerging markets is already at high levels, and for foreign acquisitions to grow, the stock market equities have to grow too.
"These considerations suggest that any simple projection of recent FDI trends may not be a sound basis to assess the sustainability of such flows."
As for the BOP impact, the issue could be viewed in terms of net transfers (comparing FDI inflows with associated payments abroad, including profit remittances, royalties, licence fees and wage remittances as well as interest paid on net loans of parent company to subsidiary.) This was a prominent issue of early literature on FDI, but is no longer so. Countries with long history of TNC involvement - hence a large stock of foreign capital - usually had negative net transfers, and they remained negative until 1988, when China was excluded.
"But it is the boom in FDI, rather than stagnation in profit remittances, that has reversed the situation in favour of developing countries in the 1990s. In fact, profit remittances continued to increase at an average rate of 10 percent per annum from 1988 to 1998."
Data limitations prevent any comprehensive analysis covering all developing countries of the broader concept of net transfers - taking account of royalties, licence fees and interest payments to parent companies.
BUT UNCTAD ESTIMATIONS OF ROYALTIES AND LICENCE FEES FOR ARGENTINA (1986-196), MEXICO AND THAILAND (1987-1996) GIVE A TOTAL OUTFLOW OF $9.9 BILLION FOR THESE ITEMS OR OVER 10 PERCENT OF FDI INFLOW FOR THESE COUNTRIES.
Viewing it in a broader approach, and including the effect of FDI on exports and imports, in addition to net transfers, and the net outcome in terms of BOP, UNCTAD has some conclusions from a few country studies available. In the case of Malaysia the total impact of trade balance of foreign firms and their income flows on Malaysia's current account has been negative in every year during 1980-1992. The impact on the trade balance became positive during the late 1980s owing to a strong export expansion. But as exports became more import-intensive, this effect diminished.
And the net foreign exchange outflows on current account were offset by new FDI inflows on capital account only after the late 1980s, but the cumulative payments impact during the whole period was negative.
A similar picture emerges for THAILAND where FDI surged after 1986, and FDI came in particularly from Japan. It HAD A POSITIVE IMPACT ON INVESTMENT AND GROWTH, and partly explained the rise in export/GDP ratio from 29% in 1987 to 36% in 1992. BUT IT WAS ALSO THE CAUSE OF A STRONGER RISE IN IMPORTS, BECAUSE THE ASSOCIATED INVESTMENT AND PRODUCTION WAS IMPORT-INTENSIVE. It has been estimated THAT 90% OF ALL MACHINERY AND EQUIPMENT USED FOR FDI PROJECTS AND 50% OF RAW MATERIALS WERE IMPORTED AND THUS FDI HAD A NEGATIVE IMPACT ON TRADE BALANCE IN THE LATE 1980S AND EARLY 1990S. This was reinforced by the rising payments abroad for royalty and licence fees and rising profit remittances. "THESE FEATURES APPEAR TO HAVE CONTRIBUTED TO THE EXTERNAL IMBALANCES THAT PLAYED AN IMPORTANT ROLE IN THE SUBSEQUENT CRISIS".
As for the recent surge in FDI into Brazil, TDR-99 cites ECLAC studies that because of upward trend in associated remittances, increased concentration of FDI in non-tradeable sectors and gradual exhaustion of privatization-linked FDI, "in the near future there will be a significant deterioration in BOP (due to activities) of TNCs in the Brazilian economy. According to ECLAC, the TNC-led restructuring of the automotive sector has been an important factor behind the renewed FDI flows into the region, and has worsened the trade balance in both Argentina and Brazil, because of the import dependence of such FDI - whereas in Mexico the impact has been strongly positive.
And even when FDI-linked activities may incur foreign-exchange deficits, such investments may still improve BOP if it creates significant externalities enhancing the export potential of the economy, TDR-99 says. There may be net benefits still if there are significant technological spillovers from FDI and the presence of TNCs.
"Nevertheless, such benefits are not spontaneous and may not compensate for additional foreign exchange deficits if FDI predominantly takes the form of M&A in non-traded sectors.
"IN ANY CASE IF THE PAYMENTS OUTCOME OF TNC-RELATED ACTIVITIES IS CONSTANTLY A DEFICIT, THE ECONOMY WOULD NEED TO GENERATE NET FOREIGN EXCHANGE ELSEWHERE, SINCE MEETING SUCH A DEFICIT BY SIMPLY RELYING ON A NEW INFLOW OF FDI WOULD MEAN ENGAGING IN AN UNSUSTAINABLE PROCESS OF PONZI FINANCING."
Examining and comparing relationship between export growth and FDI inflows during 1980s and 1990s in South and East Asian countries, the Bank for International Settlements has singled out "significant weakening of the relationship between FDI and the growth of exports in the 1990s" as a factor contributing to the factor payment problems and the crisis in East Asia. The BIS also notes that prospects had become "dimmer that the initial deterioration of the current account, brought about by the imports of capital goods associated with foreign direct investment, would eventually be corrected by new export activity generated by the increase in capacity."
UNCTAD notes that the same comparison for a larger number of developing countries, including several in Latin America, shows that this weakening of the link between FDI and exports is widespread in the developing world. The same inflows of FDI were associated with less rapid expansion in exports during 1991-1996 than in 1985-1990 in both Asia and Latin America.
"Notwithstanding other possible influences originating from global economic conditions, such as increased competition in world markets, slow growth and adverse price movements, the increasing concentration of FDI in services sectors seems likely to have played an important role in the weakening of the link between FDI and export growth."
While TDR-99 refrains from drawing the necessary conclusion, THIS LINE OF REASONING BASED ON EMPIRICAL EVIDENCE RAISES THE QUESTION WHETHER DEVELOPING COUNTRIES, IN FURTHER NEGOTIATIONS ON SERVICES, SHOULD CALL A HALT TO OPENING UP THEIR MARKETS FOR SERVICE EXPORTS FROM THE NORTH THROUGH CAPITAL AND INVESTMENT AS A MODE OF SUPPLY OR IN FURTHER FINANCIAL SERVICES LIBERALISATION OF THEIR BANKING, INSURANCE AND STOCK-MARKETS.
In a final chapter, devoted to Rethinking Policies for Development, TDR says that attracting FDI to obtain foreign technologies and secure other advantages associated with international production network of TNCs can offer a faster route to establishment of competitive industries.
"HOWEVER, THE BENEFITS FROM HOSTING TNCS ARE NOT AUTOMATIC, AND THE POLICY OBJECTIVES OF THE HOST COUNTRY IN SUCH MATTERS AS LOCAL CONTENT, TECHNOLOGICAL UPGRADING AND BOP STABILITY MAY CLASH WITH THE COMMERCIAL INTERESTS OF CORPORATIONS.... REPLACING THE HIGH IMPORT CONTENT OF TNC ACTIVITIES IN MANUFACTURING WITH DOMESTIC PRODUCTION REMAINS AN IMPORTANT OBJECTIVE IN MANY COUNTRIES. EQUALLY, THE POTENTIAL TECHNOLOGICAL AND OTHER SPILLOVERS, PARTICULARLY FOR MIDDLE-INCOME ECONOMIES AND IN SECTORS WHERE SPECIFIC KNOWLEDGE AND CAPITAL EQUIPMENT ARE CLOSELY KNIT TOGETHER, STILL REQUIRE THAT HOST GOVERNMENTS PRESERVE THEIR ABILITY TO BARGAIN EFFECTIVELY WITH TNCS."
THE TDR ANALYSIS PRESENTS PERHAPS THE MOST EFFECTIVE CASE AGAINST ANY NEGOTIATIONS FOR MULTILATERAL FRAMEWORK OF RULES ON INVESTMENT AT THE WTO, EVEN ONE PROVIDING FOR POST-ESTABLISHMENT MFN AND NATIONAL TREATMENT AND OTHER PRIVILEGES FOR FOREIGN INVESTORS.
Perhaps the UNCTAD analysis of the FDI could be summed up as FDI IS AN ACRONYM FOR FOREIGN DIRECT INVESTMENT THAT MAY BE FOREIGN BUT COULD ALSO BE ROUND-TRIPPING DOMESTIC CAPITAL (THAT OUTFLOWS LEGITIMATELY OR ILLEGALLY, AND THEN BROUGHT BACK), NOT ALWAYS DIRECT INFLOWS AS IT COULD BE REINVESTMENT AND LOAN-CAPITAL THAT IS DEBT CREATING, AND NOT NECESSARILY INVESTMENT AND CAPITAL ACCUMULATION, BUT CONSUMPTION. (SUNS4513)
This document was published in the South-North Development Monitor(SUNS), edited by Mr C. Raghavan. It is being circulated for the benefit of the NGO community.
For recirculation please obtain permission from Mr Raghavan at email@example.com