Catalog

	Table of Contents
p. 65 of 69 BACK | NEXT Foreword, Acknowledgments, Acronyms and Abbreviations, Definitions I. Developing Asia and the World II. Economic Trends and Prospects in Developing Asia III. Foreign Direct Investments in Developing Asia Trends Impact of Foreign Direct Investment Importance of the Policy Context International Investment Agreements Six Asian Case Studies Conclusions and Policy Implications Endnotes References

Asian Development Outlook 2004 : III. Foreign Direct Investments in Developing Asia

---

International Investment Agreements

In addition to national policies, bilateral agreements also shape FDI policy frameworks. Over 2,100 BITs and 2,200 double taxation treaties (DTTs) were in effect by end-2002 (UNCTAD 2003a). These BITs vary across countries but generally contain binding commitments on expropriation, transfer of funds, and compensation due to armed conflict or political instability. In the context of bilateral and multilateral trade and investment negotiations, most-favored-nation treatment obliges the host country to offer equally advantageous investment conditions to potential investors from all treaty signatories. National treatment (nondiscrimination) requires the same treatment of both foreign and domestic investors. Some BITs include provisions for extension of national and most-favored-nation treatment to FDI. Disagreements between foreign investors and the host government are usually referred to private arbitration centers of the International Chamber of Commerce or the International Centre for Settlement of Investment Disputes.

BITs are intended to protect foreign investors against unpredictable host country actions that would negatively affect the profitability of their investments. In this sense, they guard against problems of dynamic inconsistency (Box 3.4). However, their implementation may not always be effective in developing countries. Furthermore, their benefits for the host countries are not clear. Empirical evidence has not found a strong link between the existence of a BIT and an increase in FDI flows to BIT signatories. Furthermore, BITs complement, rather than substitute for, institutional quality, including strength of property rights (Hallward-Dreimeier 2003). DTTs may even reduce FDI where potential FDI includes an element of tax evasion. At the same time, BITs may reduce policy options for host country governments and leave them open to being sued for substantial amounts. While using private sector arbitration mechanisms has generally worked satisfactorily so far, it raises the potential for political disagreements in that a sovereign judicial system can be overruled by a foreign arbitration panel that is unelected and usually operates with little transparency. Decisions handed down behind closed doors by the arbitration panels, which have no public accountability, cannot be amended by the domestic legal system.

Despite the asymmetry of BIT benefits for foreign investors and host economies, these agreements have proliferated. Since 1959, the number of BITs concluded with countries in developing Asia has grown, reaching 815 by end-2002, with the surge occurring mostly in the 1990s (Figure 3.8). While a larger number of DTTs was also concluded in the 1990s, the increase in DTTs was relatively more gradual than for BITs (Figure 3.9). The PRC has signed bilateral agreements with 107 countries on protection of investment, and with 75 countries on taxes (Figure 3.10). By December 2002, India had signed 46 bilateral investment promotion and protection agreements.

As some developing economies mature and their outward FDI flows increase, as in the case of Korea, they become more interested in protecting the rights of their own investors. They may also reach regional agreements to avoid race-to-the-bottom offerings of competitive incentives to attract FDI (Box 3.5). In the broadest international context, this has led to calls for a multilateral framework on investment. Currently, the prime example of a multilateral framework is the Agreement on Trade-Related Investment Measures (the TRIMs Agreement), discussed in the next section.

Box 3.4 Bilateral Investment Treaties With the collapse of the Cancun trade talks due to seemingly unbridgeable differences on the so-called Singapore issues and agriculture, bilateral investment treaties (BITs) have now become a more important instrument in the protection of foreign investment. BITs normally deal with the following: scope and definition of foreign investment, access to specific sectors and industries, nondiscrimination, most-favored-nation treatment, guarantees and compensation in respect of expropriation, compensation for war and civil disturbance, guarantees of free transfer of funds and repatriation of capital and profits, and dispute-settlement provisions. BIT provisions vary considerably, even in BITs signed by the same country, implying different bargaining positions and approaches (UNCTAD 2000). The first BIT was concluded between Germany and Pakistan in 1959. Since then, the number has grown, reaching 2,181 by the end of 2002 (Box Figure 1). In their early years, BITs were often concluded between industrial (usually capital-exporting) and developing (usually capital-importing) countries, at the initiative of the former to ensure legal protection for its firms in the latter. The developing country usually agreed to the BIT to encourage foreign investors. Things changed from the late 1980s, when developing countries started to sign BITs among themselves. The distinction between capital-exporting industrial countries and capital-importing developing countries began to blur. Countries entered into BITs to protect their outward investment to, as well as inward investment from, the other BIT partner. Through the years, it is not only the number of BITs that has increased. The number of countries that are party to BITs likewise increased significantly. From only two at the end of the 1950s, the number has increased to 172-24 industrial countries, 53 countries from Africa, 31 from Latin America and the Caribbean, 15 from the Middle East, 31 from developing Asia, and 18 from central and eastern Europe-by the end of 2002. Since the 1960s, countries in Africa have been actively involved in BITs. Starting from only three BITs in 1960, African countries are currently party to 533 BITs, most of which were concluded with members of the European Union. Countries in developing Asia have also concluded a large number of BITs, the majority of which were signed after the dissolution of the former Soviet Union in 1991. In recent years, the Central Asian republics have been the most prolific BIT parties, signing 144 BITs between 1992 and 2002. Up until the late 1970s, only a handful of countries in Latin America and the Caribbean were involved in BITs. It was not until the late 1980s that the rest of the countries in that region began concluding their first BITs. The number of BITs involving Latin American and Caribbean countries soon increased rapidly, reaching 413 at the end of 2002, mostly with Western European countries. Meanwhile, Central and Eastern European countries had concluded 716 BITs by end-2002, mainly among themselves and with other developing countries. Western European countries have concluded the most number of BITs, reaching 1,038 in 2002. More than half of these BITs were signed with developing countries. Non-European developed countries became involved with BITs relatively late in the 1970s and early 1980s. But they started to catch up in the 1990s, such that by the end of 2002, the US had 45 BITs, Canada 24, Australia 21, Japan 10, and New Zealand 2. Box Figure 2 shows the top 10 countries in terms of number of BITs signed as of end-2002. These trends indicate the increasing importance that countries have placed on the role of FDI in their development strategies. Sources: UNCTAD (2000, 2003b).

Agreement on Trade-Related Investment Measures

After a failed attempt to formulate an MAI among mostly industrial countries in the Organisation for Economic Co-operation and Development (OECD), the TRIMs Agreement was formulated in the Uruguay Round of the General Agreement on Tariffs and Trade (GATT) (1986-1994), the forerunner of WTO (Box 3.6). TRIMs are a subset of the incentives and regulations designed to influence FDI. Broadly speaking, they consist of incentives and regulations deemed to have a direct impact on international trade. A combination of factors led to the inclusion of foreign investment in the work program of the Uruguay Round negotiations. First, there had been the changing perception of the role of FDI in development. Foreign investment had become increasingly important and the flows far more sizable. Perceptions had also moved from initial anxiety about FDI to a more welcoming stance. A dispute between the US and Canada over the latter’s application of performance measures on foreign firms around this time also facilitated debate on the linkages between GATT rules and foreign investment policy (Bora 2001).

The TRIMs Agreement recognizes that certain investment measures distort trade and that these distortions are inconsistent with GATT principles. Export subsidies, import entitlements, minimum export requirements, and local content requirements directly affect volumes and prices of imports and exports, and in some cases the composition of trade. Local content requirements mean that imports are treated less favorably than domestic inputs, violating the national treatment principle of GATT. A trade-balancing requirement that limits the quantity of imported products that can be used if an MNE does not meet its export target also violates national treatment obligations. Some foreign exchange balancing requirements impose a similar scheme whereby a corporation’s permitted imports are tied to the value of its exports to ensure a net foreign exchange earning.

Despite this recognition, only a brief TRIMs Agreement was put into effect on 1 January 1995 after a lengthy debate.

All WTO developing member countries were to have implemented the TRIMs Agreement and eliminated their relevant regulations by 1 January 2000. Twenty‑six developing country members with widely varying economic characteristics gave notice that at that time they still had a variety of policies in existence, however. Most of the policies related to the auto industry or the agro-food industry. The policies overwhelmingly adopted by these countries were local content schemes. The second most frequently notified type of TRIM was foreign exchange balancing requirements (Bora 2001).

In Malaysia, TRIMs Agreement compliance was completed with the phaseout at end-December 2003 of the local content policy on motor vehicles for both new and existing firms. Local content requirements are also being phased out in Viet Nam as part of its moves toward WTO accession. Compliance with the TRIMs Agreement was also instrumental in overcoming resistance to investment policy reform in Thailand, as TRIMs were gradually abolished.

A couple of dozen countries requested extensions of the transition period. Among them, Argentina, Malaysia, Philippines, and Thailand cited financial crises that added to their structural adjustment problems as a major factor behind their extension requests. Colombia and Pakistan cited specific development reasons for their extension requests. Colombia detailed difficulties in transforming its economic model, especially in terms of developing substitutes for illegal crops, which may also be relevant in a number of Asian economies. They argued this would require domestic absorption, or local content policy, to ensure that farmers are able to sell their produce. Pakistan argued that opening its economy to import competition rapidly would not allow it to exploit domestic resources optimally, to promote the transfer of technology, or to promote employment and domestic linkages. Another reason cited for an extension request was inconsistency between preferential trade agreements and multilateral obligations (Bora 2001).19

The current TRIMs Agreement resulted from a compromise. During the Uruguay Round discussions, developed economies, including the European Union (EU), Japan, and US initially proposed to establish a comprehensive agreement on investment. Their proposed framework covered a wide range of areas such as technology transfer requirements, restrictions on the transfer of profits overseas, controls on foreign exchange flows, government reviews of foreign investment performance, and nationalization. This plan faced strong resistance from governments of developing countries. Brazil and India maintained that investment was outside GATT’s competence, while other developing countries tended to take a defensive position with regard to any agreement on TRIMs (Ariff 1989). In particular, many developing countries resisted the extent to which market access for foreign firms would be included. The result was that negotiations focused on policies that applied to the operations of foreign firms. Even then, negotiations still proved difficult.

Since the enactment of the Agreement, the main focus has been on the trade effect of local content and export performance requirements. Since 1995, TRIMs Agreement obligations of new members on accession to WTO have depended on the terms of their accession. So far, all acceding countries have agreed to implement the TRIMs Agreement upon accession regardless of whether they are developing countries or not.

The TRIMs Agreement, in relation to the complicated issues it covers, is very short. Its brevity is a manifestation of the divisions that emerged between developed and developing countries. This lack of consensus also explains the vague nature of elements of the agreement. The current TRIMs Agreement does not contain a basic definition of investment, even though the definition of investment has profound implications for the scope and coverage of the Agreement. The definition of investment in the draft OECD MAI, for example, went far beyond the traditional notion of FDI to include portfolio investment, debt capital, intellectual property rights, and various forms of tangible or intangible assets (Ganesan 1998). Korea has suggested using an enterprise-based definition of investment for investment liberalization, and an asset-based definition for investment protection. This approach could help developing countries avoid negative effects of volatile short-term portfolio investment.

There is also no well-defined phase‑in period in place to bring laws into conformity for members that notify WTO of relevant policies under the TRIMs Agreement. Members are under no obligation to respond to notification requirements in detail. This has caused some implementation difficulties. None of the notifying countries has either developed an implementation plan or identified alternative policies that could be used to achieve conformity.

TRIMs are typically used in conjunction with a number of other policies. One aspect, which was not taken into account during the Uruguay Round negotiations, was how the removal of certain TRIMs without addressing companion policies would affect trade. For example, local content schemes are usually combined with a subsidy. The TRIMs Agreement disciplines trade policy, but not incentives. Views are still divided on how to deal with both incentives and regulations.

Ambiguity in the wording of the TRIMs Agreement has made the interpretation of obligations difficult. Some developing countries’ lack of capacity to fully understand the scope and implications of these obligations has exacerbated this problem. These problems have also created a tension between the generally accepted notion of efficiency and the broader definition of development. Some work to clarify these issues has been done recently by the WTO Council for Trade in Goods, but solving these problems will require much time and effort (Bora 2001).

The current TRIMs Agreement relies on the state-to-state mechanisms of WTO for dispute settlement and arbitration under which, for example, a dispute settlement panel is established and makes its judgment. Some argue that it is necessary to establish investor-to-state mechanisms to ensure that investors receive a hearing (Moran 2002). Others believe that inclusion of an investor-to-state dispute settlement mechanism would add an excessive burden on developing countries’ legal machinery and impose a threat to their national sovereignty (Tangkitvanich et al. 2003). Korea’s preliminary position was that the dispute settlement mechanism of a multilateral framework on investment should not cover investor-to-state disputes.

Believing the existing TRIMs Agreement to be inadequate, industrial country parties such as the EU, Japan, and US largely view it as a mechanism for removing performance requirements on foreign investment (Greenfield 2001). Still, the Agreement represents a step forward in ensuring that countries are all subject to the same rules respecting the use of certain investment-related performance requirements. Although most FDI has occurred between industrial countries, the Agreement allowed investment issues to be discussed in the context of multilateral negotiations. These discussions continued through the WTO Working Group on Trade and Investment where members have further assessed the linkages between trade, FDI, and development. It has also allowed disputes between member states to be settled in the WTO context, and has enforced GATT provisions.

The Ongoing Divide Regarding a Multilateral Framework on Investment

Investment was again put on the agenda for the Doha round of WTO negotiations. Investment, competition policy, transparency in government procurement, and trade facilitation were labeled the “Singapore issues,” following the WTO work program in the 1996 Singapore Ministerial Declaration. The Doha Declaration continued to attach the usual operational qualifications of “trade-related aspects only” for investment and competition policy. However, as evidenced at the Fifth WTO Ministerial Conference in Cancun, Mexico, in 2003, members are still far from an agreement on investment issues.

In the WTO discussions, negotiations take place in what is known as a “single undertaking” in which concessions in one sector can be traded off against concessions in other sectors. In Cancun, resistance by more developed economies to make further concessions on agriculture prompted some developing countries to refuse to negotiate on the Singapore issues, and vice versa. In the end, the discussions broke down on the issue of investment, although disagreement on investment alone could probably have been overcome.

In a joint submission with Brazil to the WTO’s Committee on TRIMs in October 2002, India argued that the TRIMs Agreement should be amended to incorporate stipulations that provide developing countries with the flexibility to implement development policies. In particular, it proposed that developing countries should be allowed to use investment measures or performance requirements to promote domestic manufacturing capabilities in high value-added sectors, to stimulate transfer and indigenous development of technology, to promote domestic competition, and to correct restrictive business practices (Kumar 2003). Despite its increasing trend of investment abroad, Malaysia also views multilateral rules on investment as impinging on development policy options and has called for clarification of the issues before negotiations begin (Tham 2003).

Some developing countries take the view that TRIMs and other investment measures are domestic investment issues that should therefore not involve WTO. This point was emphasized during the WTO Ministerial meeting in Cancun. India and others have also asserted that the mandate of WTO is confined to trade and does not extend to investment. They fear they would be deprived of a major means of exercising control over foreign firms operating locally if their right to impose TRIMs or other investment measures were removed. Some developing countries, including India and Malaysia, also consider that policies such as domestic content requirements are essential policy tools for industrialization. At the WTO Doha Ministerial Conference in November 2001, a number of countries stated that joint-venture requirements encourage indigenization. They believed developing countries should be allowed to use TRIMs and other investment measures flexibly in pursuit of developmental objectives because each country’s unique needs and circumstances require sufficient freedom and flexibility to pursue one’s own policies. They propounded the view that, although the TRIMs Agreement established uniform obligations for all members, it does not take account of structural inequalities and disparities in levels of development; technological capabilities; or social, regional, and environmental conditions; and does not incorporate a meaningful development dimension. A legally binding treaty on foreign investment would further reduce the degree of flexibility available (Ganesan 1998).

In a broader context, Panagariya (2001) has pointed out that trade is generally easier to liberalize than investment, which is easier to liberalize than labor flows. Within trade, goods trade is easier to liberalize than services trade. Within investment, FDI is easier to liberalize than portfolio investment. And within labor, opening up to immigration of skilled labor is easier than opening up to immigration of unskilled labor.

To date, it is still unclear whether WTO will negotiate investment-related issues further, and what form and scope the negotiations will take if they do proceed. Furthermore, it is not clear whether any new talks would center around extension of the current TRIMs Agreement or a new comprehensive multilateral framework on investment. In large part, this reflects the continuing difference of opinions between developing and industrial countries. The next subsection explores some issues that might arise in future negotiations.

Is a Multilateral Framework on Investment Necessary?

Investment measures would largely become a nonissue if trade liberalization succeeded in dismantling tariff and nontariff barriers to trade. For example, local content requirements tend to raise production costs and render final products uncompetitive. A local content program can only be sustained behind protectionist walls, as in India. Similarly, elimination of protection will diminish the need for export incentives, as in Thailand. Without tariffs, quotas, and other import barriers, there is less rent to extract and thus less scope for performance requirements. Thus, trade liberalization can also induce more liberal investment regimes. The more successful is trade liberalization, the less will there be need to worry about investment agreements. This raises two questions: should liberalization focus on trade, and is an agreement on investment needed?

The reality is that trade and FDI coexist. Impediments to trade are a factor in the growth of FDI, but other market imperfections also have important influences on the decisions of firms to invest abroad. Real market conditions seldom approximate the free trade model. Oligopoly rather than perfect competition is a characteristic of many market structures in which foreign firms operate, and these firms have considerable discretion over the choice of market in which they operate (Balasubramanyam 1991). It is unrealistic to assume that all trade barriers will disappear soon. In these circumstances, restrictions on foreign investment, as well as incentives to promote it, may exist for a long time.

An agreement on investment might strengthen the investment climate of host countries and con­­tribute to trade liberalization. Foreign investment and trade are not necessarily substitutes for each other. Often they have a complementary relationship. Effects of restrictions on trade or investment are empirically indistinguishable from one another. This argues for reducing barriers to investment under multilateral disciplines, just as barriers to trade have been reduced under GATT/WTO rules.

Is an Investment Agreement a North-South Divide?

Industrial countries represent both major sources of, and hosts for, FDI. The increased extent of intra-industry FDI among the industrial nations blurs the distinction, at least among those nations, between host and source countries. The investment issue is thus of interest to industrial countries as both suppliers and recipients of FDI. Developing countries are mainly recipients of FDI, but a number of them-most prominently Hong Kong, China; Korea; Singapore; and Taipei,China-have undertaken significant investment abroad. Thus for some developing countries, their stake or interest in the TRIMs Agreement and other investment issues may be more similar to that of their industrial counterparts than to other developing countries.

There are different views between, and within, industrial and developing nations. For example, given their generally open capital markets, relatively higher income levels, and preoccupation with agricultural liberalization, countries in Latin America were not particularly opposed to negotiation of the TRIMs Agreement. Much of the opposition derived from countries in Africa and Asia (Panagariya 2001). Still, in its submission to the Working Group on the Relationship between Trade and Investment at WTO, the Government of Korea supported the EU position on banning technology transfer requirements for foreign investment (Greenfield 2001).

Conventional wisdom holds that developing countries engage in trade-distorting investment measures while industrial countries do not. However, trade and investment figures clearly show that developed countries also use investment measures. Most industrial countries make available location-based incentive packages for both domestic and international investors. Ireland reports that its special incentive packages have attracted more than 1,200 foreign firms to its economy, and these account for 70% of the country’s industrial output and three quarters of its manufactured exports (O’Donovan 2000). OECD (and others) found that almost 90% of all domestic support programs in the EU were available to foreign investors (OECD 1996, Moran 2002). Any multilateral effort to create a level playing field for national and international companies among source and host countries around the world would be seriously deficient if it ignored the proliferation and escalation of location-based incentives by industrial countries.

Overall, it appears that, with or without a multilateral framework for investment, many countries have carried out liberalization of investment regimes. This has two possible implications: (i) a multilateral framework is redundant; or (ii) it is more necessary as policies converge, and a more comprehensive international agreement on investment becomes increasingly possible and necessary for facilitating the liberalization process and governing investment measures. Which outcome emerges will depend on the bargaining positions adopted by different countries and the attitudes they hold toward the process. Even a small group of Asian economies hold widely diverging views on negotiating a multilateral framework for investment, from strongly in favor (Korea) to strongly opposed (India), from viewing it as a helpful spur to domestic liberalization (Thailand) to a constraint on development policy options (Malaysia), and from acceptance if implementation is gradual (PRC) to concern over capability to address the challenges of achieving compliance (Viet Nam).

Back Importance of the Policy Context	Next Six Asian Case Studies