The study -
Reviews principal issues concerning foreign direct investment liberalisation and its relationship to trade liberalisation, development and economic growth.

Assesses the need for a multilateral agreement on policies towards FDI.

Discusses and proposes policy options for such an investment agreement.

The project has been commissioned by the Norwegian Ministry of Foreign Affairs, and has been carried out at Centre for International Economics and Shipping (SIØS), at Norwegian School of Economics and Business Administration (NHH) and Institute for Research in Economics and Business Administration (SNF).

Abstract

Do we need a multilateral agreement for policies towards international investments similar to the trade agreements that exist? Foreign direct investment (FDI) plays an important role in the international economy, and in many cases FDI is at least as important as trade is for international transactions between countries. FDI may have significant effects for the overall welfare of a country as well as for the income distribution within the country; hence, countries typically have incentives to use active policies towards FDI, similar to the incentives for trade policies. Nevertheless, when it comes to international policy regulation and co-ordination, trade and FDI are treated in a very asymmetric way, with extensive multilateral agreements for trade policies, but nothing similar for FDI. This report focuses on policies towards FDI and discusses the need for a multilateral agreement to regulate the use of such policies. There are good reasons to call for international co-ordination of FDI policies. The question is, however, whether a new multilateral investment agreement is the best solution, or if the need for international co-ordination could be better taken care of within existing agreements. After a brief review of the effects of FDI for host and home countries, and the incentives to use active policies towards such investments, the report discusses the pros and cons of a new multilateral investment agreement. In the final section of the report, the status for such an agreement after the Ministerial Declaration from Doha is sketched.

1. Introduction

One of the most striking features of the international economic development the last 10 – 15 years is the very strong growth in foreign direct investments (FDI). Figure 1 shows the average annual growth rates for trade and international investments over the period 1986 – 1999. While the growth in international trade has slowed down over the period, international investments, and in particular international mergers and acquisitions (M&A) have grown significantly. The growth continued throughout 2000, with almost 50 percent growth in M&A, while the overall growth in FDI was approximately 20 percent.

Figure 1

While growth rates indicate increased impact of FDI, they say little about how important such international transactions are. To get a picture of the importance of FDI, one can study the activities of foreign subsidiaries of multinational enterprises (MNEs), since FDI and the activities of subsidiaries are closely related. Data for trade versus sales from subsidiaries in an EU-US context give a very striking picture. For manufactures as an aggregate, subsidiaries of US multinationals in Europe sell approximately 3.8 times as much as total exports of manufactures from the US to EU, and the same is true for EU subsidiaries’ sales in the US (3.6 times larger than exports). Hence, at least for EU-US relations, the activities related to FDI dominate the international transactions. 1See Barba Navaretti, Haaland and Venables (2001) for data and discussion of the links between trade and FDI, with emphasis on the EU-US relations.

Foreign direct investments are considered important for host as well as for home countries. Host countries nowadays tend to emphasise potential benefits from inward FDI, e.g. employment effects, technological spillovers, linkages to domestic industries, and so on. However, there are also potential negative effects related to inward FDI. Foreign ownership implies loss of domestic influence and control; this becomes particularly clear in cases of foreign take-overs of domestic firms. Foreign ownership also means that profits and revenue are taken out of the countries, and in some cases that may be considered a problem. And there is often a fear that foreign-owned firms may be more likely to close down or scale down production and employment in difficult periods. For home countries, the assessment of outward FDI is also mixed. Some focus on the negative effects – through "export of jobs" – while others claim that foreign investments would secure efficiency and competitiveness, and in that way strengthen the home base as well.

There are important policy questions related to foreign direct investments. Traditionally, there were barriers to FDI in most countries; various measures made it more difficult for foreign firms than for domestic ones to establish production, and for foreign take-overs there have been strict rules in many countries. There were also restrictions related to moving income across borders, problems with taxation, and so on. More recently, the policies have changed significantly. Now many countries emphasise the benefits of foreign investments. As a consequence, very many countries have reduced or removed the former barriers to FDI; of the 150 policy changes towards FDI in 2000 reported by UNCTAD (2001), almost all (147) were towards more liberal regimes. In a lot of countries the current situation is actually one of promoting inward FDI through a number of direct and indirect measures. Due to the alleged benefits of inward FDI, countries try to be attractive as a host; and if the benefits are considered important enough, one may even be willing to offer special incentives or direct subsidies to ensure that the investments take place.

The focus of this report is on policy issues and the need for multilateral regulations of policies. However, in order to discuss such issues, we need to understand the motives for FDI and the effects of such investments for host and home countries. In section 2 motives for and effects of FDI will be sketched, followed by a discussion of the policy incentives towards FDI. Section 3 constitutes the core of the report; in the section arguments for and against a multilateral investment agreement are discussed and important prerequisites for a potential agreement are sketched.

2. FDI – Motives and Effects

2.1. FDI – Motives

The motives for FDI from the firms’ point of view can basically be grouped in two: market access or cost reductions. The former is often called horizontal FDI, while the latter is associated with so-called vertical FDI. Horizontal FDI means that the firm chooses to perform the same or similar production processes in different countries; the alternative would be to concentrate production in one location and export to all markets from there. When firms choose horizontal FDI it must be because they find it more advantageous to produce locally than to import to the market in question. One important reason for such FDI would thus be high trade costs or barriers to trade; hence, this is often called tariff-jumping FDI. There would normally be a trade-off between economies of scale in production and avoiding trade costs. Hence, such FDI is more likely the higher the trade costs are, and the larger the market in question is.

Vertical FDI, on the other hand, is not about duplicating the same processes in different locations; it is about splitting up the production process and localising different stages of the process different places. Hence, such FDI is often associated with production networking. By fragmenting the production process, the firms can take advantage of local factor market conditions and lower the overall costs of producing a product. A typical vertical FDI would e.g. localise skill-intensive parts of the process in countries well endowed with highly skilled labour, while the stages of the process that require less skills could be located in labour-abundant, lower-wage countries.

In either case – horizontal or vertical – the firms’ motives are of course to maximise profits. The choice depends on the overall conditions; production costs and trade costs matter, market size and the importance of economies of scale matter, and economic policies in both host and home countries matter. The focus of this report is the policy aspects, but it is important to keep in mind that it is the overall conditions that are decisive for the firms.

2.2. FDI – Host Country Effects

There are many possible channels through which inward FDI affects a country. We should at least distinguish between three or four types of effects: a rent effect, an output effect – which could be split into an employment effect and a competition effect – and a spillover effect.

Ownership itself may matter, as foreign ownership implies that rents may move out of the country. To the extent that this is normal return on the foreigners’ investment, it should not be considered a problem. If, on the other hand, there are extraordinary rents in the industry or special benefits given to firms in the industry, e.g. through domestic policies, rent shifting may be regarded as a problem.

Inward FDI also affects output and employment. If FDI replaces imports, it would imply a net increase in domestic employment; and if the initial situation is one with unemployment, this is a benefit for the host economy. Two caveats should, however, be added here. FDI may actually replace or take over activities of domestic firms, e.g. in the case of mergers and acquisitions (M&A), and then the net employment effect is not clear. Secondly, with full employment inward FDI may cause pressure in the local labour market, with wage growth as the effect. This may, of course, be a benefit to workers, but it represents increased costs and reduced employment for domestic producers. Hence, for the output effect to yield a welfare gain for the country, two conditions must be satisfied: first, the extra employment must not just crowd out domestic employment. And, secondly, there must be some kind of market failure such that the return from employment due to the FDI exceeds the return in the best alternative use of the labour.

Inward FDI may also affect the domestic firms in more direct ways. Firms competing with foreign suppliers will typically find that competition becomes tougher when the foreign firms choose FDI rather than trade. With lower marginal costs (i.e. no trade costs) the foreign firms can compete more fiercely and will typically gain market shares. How strong such competition effects are, depends on the type of competition in the market and the degree of substitutability between foreign and domestic products, but in any case the position of domestic competitors worsens.

Whereas domestic competitors may experience the tougher competition as a real burden, other local firms may benefit through spillover effects – the technological and pecuniary externalities that may be created by inward FDI. To the extent that the foreign firm controls technology or knowledge stock of higher quality than local firms, an inward investment may enable local firms to upgrade technology. Spillovers could be direct, from firm to firm, or indirect through the labour market. In either case, one would expect that the more there be of a technological gap at the outset, the more likely we are to see positive spillovers from inward FDI 2In a study of British manufacturing, Griffith and Simpson (2001) report higher levels of labour productivity and also higher productivity growth for foreign-owned firms relative to domestic firms. To a large extent, the productivity differences can be explained by higher investment per worker and different skill requirement, but the study nevertheless shows that there is a difference between foreign-owned and domestic firms. Hence, there may be something to learn from a different way of conducting business..

Pecuniary externalities arise if firms do not capture the entire surplus created by their market transactions. An obvious example is where the foreign firm buys intermediate goods from local suppliers. If there are scale economies or positive industry externalities in such intermediate industries, the increased demand from the FDI project could lead to output expansion and higher productivity or quality upgrading in the domestic industry. Another example is if the foreign firm supplies new products or new qualities that yield benefits for domestic consumers or domestic producers. In such cases FDI can act as a catalyst, initiating the development of a cluster of activities, with accompanying welfare gains in the host country (see e.g. Markusen and Venables, 1999, and Haaland and Wooton, 1999). There is considerable evidence of the role of FDI in the development of such clusters (for example, the Irish experience, Barry et al 2001). The World Investment Report 2001 (UNCTAD 2001) – subtitled "Promoting linkages" – emphasises the importance of spillovers and goes through a number of possible mechanisms through which FDI may affect local industry. A survey of spillover effects is contained in Blomström and Kokko (1998).

2.3. FDI – Home Country Effects

The home-country effects of outward FDI could similarly be classified as rent effects, output effects and possibly spillover effects. The rent effect captures the effect of outward FDI on the overall profits of the investing firm, or the share of these profits remaining in the home country. With profit-maximising firms we could normally expect the rent effect for the firm to be positive – otherwise they would not invest abroad. Whether the positive profit effect for the firm translates into a rent effect for the home country, may depend on a number of conditions. It is, for example, well known that multinational firms use transfer pricing to move taxable profits between countries, so even if there are overall benefits for the firm, they do not necessarily turn up as increased profits in the home country 3Although tax regimes are important for multinationals, tax policies will not be dealt with in this report. For a discussion of tax policies and multinationals with an emphasis on tax competition, see e.g. Schjelderup (2002) and Janeba and Schjelderup (2002)..

The output effect of the home country is often associated with the "export of jobs" argument, emphasising the direct reduction in home-country employment that follows from moving production abroad. While this may be true, the picture is not always as simple as that. The counter-argument is that by moving parts of the production process abroad, the overall profitability and competitiveness of the firm improve, which could secure and even strengthen the home base. However, the composition of home employment may be altered. Outward FDI could e.g. imply less low-skilled but increased high-skilled employment in the home country, as the home base often includes headquarters and R&D activities.

Finally, there may be spillover effects of importance for the home country as well. These are of at least two types. First, having production facilities in many locations may give the firm access to new insights, techniques or ways of organising the work that may be of benefit for the home base. And secondly, by concentrating the home activities to e.g. R&D and other skill-intensive processes, there may be room for more spillovers through clusters of high-skill activities in the home country.

2.4. FDI and Policies

The motives for FDI from the firms’ point of view as well as the effects of FDI from both the host and the home countries’ point of view could all be significantly influenced by economic policies. A wide range of policies affects FDI. General policies, like tax policies, trade policies, industrial policies and labour markets policies are all important for the overall conditions for FDI. In addition, there are policies that are specific to FDI, ranging from barriers to FDI, to incentive schemes, special tax regimes and even direct subsidies to attract inward FDI.

The fact that so many countries design specific policies towards FDI indicates the importance of such investments. Over time there has been a shift in policy stance in many countries, from barriers towards FDI to more liberal policies or even direct incentives to attract FDI. As noted above, UNCTAD (2000 and 2001) reports a significant number of explicit changes in policies towards FDI in recent years, and almost all of these are shifts to more liberal regimes. UNCTAD (2001) also discusses the very strong growth in the number of bilateral investment treaties (BITs); over the 1990s the number of BITs world-wide increased from around 400 to almost 1950. This is a clear indication of the importance that both host and home countries attach to FDI.

Why have we seen such significant shifts in the policy stance? Why are previous barriers to FDI removed? And why do countries use active policies to promote and attract FDI? In general, FDI policies – as trade policies – can normally be explained by one of two reasons: either the policies are there to try to secure overall domestic welfare gains, or they are introduced to give special benefits to some domestic interest groups at the expense of other domestic groups. In the latter case, the policy choice must follow from the fact that some groups are better organised and have stronger political influence than other groups (political-economy reasons). Barriers towards FDI – both in the host and in the home country – would typically be the outcome of a political-economy process, where those who feel their position threatened by FDI may have more political power than those who gain from FDI.

The shift in policy stance towards FDI can then either come from increased expected overall benefits from FDI, or from changes in the political-economy process. A combination of the two can probably explain many of the observed changes. In most countries, the potential benefits of inward FDI through spillovers and linkages to local industry have recently been emphasised to a much larger extent than before, and with such a shift in the assessment, the policies should change as well. In addition, groups that may be hurt by international investments both in host and home countries may have lost some of their political influence in many countries.

Based on economic theory, it is easy to argue for free and stable FDI conditions. The arguments are the same as for free trade; free and open markets ensure the most efficient use of resources in all countries. However, the fact that many countries now actively promote FDI through various policies is more difficult to explain. It is true that with externalities or other market failures the gains from FDI may be significant and self-reinforcing. In certain cases, FDI may work as a catalyst for domestic industry, as discussed above (see e.g. Markusen and Venables, 1999, or Haaland and Wooton, 1999), and that could in principle be a good reason to use active policies to attract FDI. However, it is not easy to identify the investment projects with the right characteristics, and without linkages or spillover effects, the costs of the subsidies or incentive schemes will typically more than outweigh the national benefits. So to fully understand the increased use of incentive schemes to attract FDI, political-economy reasons should be added to the potential overall welfare reasons. Some domestic interest groups clearly gain from FDI, and these may have gained political influence over the years. To take only one example, domestic small and medium sized firms would typically supply intermediates to the foreign firms establishing production in the country, and the political impact of such firms has increased significantly in many countries over the last decade.

Policies to promote and attract FDI may then be welfare improving for the country, or they may change the welfare distribution within the country. In the latter case, the overall welfare for the country would normally be higher without the policies. In the former case, the policies may increase overall welfare – if they work as planned. However, it is quite obvious that if one country sees benefits of attracting a certain investment, other countries would do as well. And if several countries use active policies to try to capture the same FDI project, the results may well be a harmful policy competition between the countries 4Fumagalli (1998) gives a number of examples of actual policy schemes to attract inward FDI..

2.5. Trade and investment

From the above discussion it should be clear that the importance of international investments has been growing significantly over the last 15-20 years, and that FDI in many cases is as important as trade in creating international or global markets. There are potential benefits from international factor movements – like FDI – in the same way as there are gains from trade. When market conditions differ between countries or regions, there are always potential benefits from arbitrage to "bridge the gap" between the markets. Through international economic interactions the markets become more efficient, and that may give rise to gains for all countries involved in the transaction. This is true both for international trade and for international factor flows; whether the interaction takes one form or the other, is often a question of relative transaction costs, without changing the basic principles.

There are close links between trade and investments, and between trade policies and investment policies. Both theoretical and empirical research emphasises these links. With what we called horizontal FDI, the purpose of the investments is to serve the foreign market from local plants rather than by trade, thus saving trade costs. In such cases FDI should be regarded a substitute to trade. With vertical FDI, foreign investments occur as part of a production-networking scheme, in which the production process is fragmented in order to take advantage of local cost and productivity advantages. In the latter case, the scope for cost reductions through more efficient production networks is better the more stable and free trade and investment conditions are, and trade and investments would typically appear as complements rather than substitutes.

Empirical analyses show that the bulk of FDI takes place between industrialised countries. This is true both for horizontal and vertical FDI. For the EU-countries for example, almost 50 percent (in 1998) of the stock of outward FDI is invested in other EU countries, more than 5 percent in EFTA countries, and some 25 percent in the US. For inward FDI the European dominance is even stronger 5See Barba Navaretti et al (2001) for more details.. For horizontal FDI market access and market position are the main motives; for such FDI it should not come as a surprise that the industrialised countries dominate, since that is where the larger markets are. However, as trade costs are emphasised as important reasons for such FDI, it is more of a surprise that there has been such a strong growth of FDI between countries that have very few remaining trade barriers. Two reasons help explain the development: first, even without formal trade barriers, there are trade and transaction costs that give foreign suppliers a disadvantage in a market, and secondly, there has been a strong liberalisation of investment policies over the period in question.

For vertical FDI – which is linked to production networking and cost reductions through international specialisation of the production process – one would expect the bulk of FDI flows to be between countries with fairly different factor market conditions. However, data show that even for vertical FDI interactions between industrialised countries dominate. For the EU countries, for example, more than 80 percent of trade in parts and components – which can be taken as an indication of production-networking activities – is with other industrialised countries 6It is difficult to distinguish different types of FDI in the investment data, hence, one must rely on indirect sources.. Given that such trade should be larger the more different relative costs between the countries involved are, it comes as a surprise that trade and FDI among industrialised countries dominate so much. The explanation may be that only these countries are sufficiently integrated to allow efficient fragmentation of production. If that is true, there should be ample opportunities for more network activities and hence much more efficient ways of production in the future. The fact that there has been a strong growth of trade in parts and components between the EU and the Central and Eastern European countries in the 1990s is a good indication of this scope for networking activities as trade and investment barriers go down and markets become more integrated.

In spite of the fact that there are gains from free capital movements as well as from free trade, most countries tend to use active policies to affect the markets. For trade policies it is well documented – both theoretically and empirically – that countries may have strong incentives to use active trade policies, even if the overall benefits would be higher with free trade. Trade policies could either be used to try to influence world market prices to the advantage of the country (at the expense of other countries) or to give special benefits to some groups at the expense of other domestic groups. The latter may be due to factor markets effects of protection – like e.g. labour gaining from the protection of labour-intensive industries – or it may be for strategic reasons, ensuring market power for domestic firms. In either case, the protection could only be understood in a political-economy setting, where political power and e.g. lobbying matter for the policy outcome.

Similarly, countries may have incentives to try to influence foreign investments, as discussed above. This is true both for inward and outward FDI, and the reasons for such policies are the same as for trade policies: national welfare gains, or political-economy reasons.

There are also close links between trade policies and investment policies. We have seen above how trade policies are among the key determinants of FDI flows and hence production activities abroad. For horizontal FDI, trade barriers induce more FDI, while for vertical FDI-projects the opposite is true. However, this is not a one-way relationship. The growth in FDI, and thus in the mobility of firms and production processes, has very significant implications for the use of trade policies. To take only one example, trade policies that are meant to protect domestic firms from foreign competition may well give the opposite result if the foreign firms end up choosing FDI. The purpose of a tariff or an anti-dumping duty is to increase the marginal cost of the foreign producers and hence improve the market position of the domestic ones. But if the foreign firms choose to produce locally, they would typically get lower marginal costs and gain market shares. Hence, the mobility of firms limits the scope for and the effects of active trade policies. The existence of FDI could also alter the political-economy process with regard to trade policies, due to the fact that the effects of trade barriers change when firms can move (see Ellingsen and Wärneryd, 1999).

3. The need for Policy Co-ordination and Investment Agreement

In spite of the close relationship between international trade and foreign direct investments, there is a strong asymmetry when it comes to international policy regulations. While trade policies are regulated through multilateral agreements and the WTO system, policies towards foreign investments are either purely national or governed by bilateral or regional agreements. 7According to UNCTAD (2001) there were 1941 bilateral investment treaties (BITs) in the world by the end of 2000; five times more than in 1990. A natural question is therefore whether a more symmetric treatment is called for. Do we need a multilateral investment agreement as a complement to the trade agreements in the WTO system? And if so, what should such an agreement be like?

These are not new questions. There have been previous attempts to create a multilateral investment agreement, and after the Ministerial Conference in Doha, November 2001, the issue is actually put on the agenda for the next WTO round. We will come back to the implications of the Doha Ministerial Declaration in the final section of this report. However, in the present section we will try to address the question of a multilateral investment agreement in a more analytical way. To do so, we should realise that we are really dealing with three or four key questions:

• Is there a need for international co-ordination of policies towards FDI?
• If so, is this best done through a new multilateral agreement?
• What should such an agreement include?
• Is the WTO the right institutional setting for such policy co-ordination and/or agreement?

There is not a unique, correct answer to questions like these. There are arguments for and against each one, and there may be several possible solutions to the problems in question. Hence, we cannot come up with the right solution; but we can sketch the most relevant arguments and considerations.

3.1. A need for International Policy Co-ordination?

It is not difficult to come up with good reasons why there may be a need for international co-ordination of investment policies. Some of these are similar to the case for trade agreements, while others are more specific to investments. Hoekman and Saggi (1999) list a number of reasons in favour of international policy co-ordination (see Box 1). First, active policies may actually reduce overall welfare for the country itself. If e.g. barriers to inward FDI appear for political-economy reasons, the policy need not improve overall welfare. FDI barriers to protect domestic producers would for example imply that consumers lose due to higher costs and less competition, and in the same way as for trade barriers, the consumer losses will typically exceed the producers’ gains. In a political-economy context protection may nevertheless be the outcome, as long as producers have stronger political influence than consumers do 8This political-economy story does not only apply to FDI barriers; it is equally important for FDI incentives. The key element is that the policies are based on the influence of special interest groups, rather than overall national welfare, as discussed in section 2.4.. Hence, an international policy agreement – limiting the possibilities to protect local producers – could be welfare improving.

Secondly, policies in one country affect other countries. If one country is successful in attracting a foreign investment, other potential hosts may lose. In general, that is not a problem; if the country with the best overall conditions for a particular type of activity gets the investment, that would enhance overall efficiency. However, if the firm’s choice of location is determined by active policies and incentive schemes rather than by the underlying market conditions in the successful country, the outcome may be both inefficient and unreasonable. And if several countries try to use active policies to attract the same inward FDI, the outcome may well be a harmful policy competition between countries.

If two or more countries use subsidies, tax regimes or other incentives to attract the same investment, the competition between them may lead to too generous incentive schemes. Haaland and Wooton (1999) for example, show that even in a setting where inward FDI gives overall gains to the economy, the successful host country may end up with such expensive incentive schemes that there are no net benefits left in the country. In such a setting, all the benefits will be transferred to the foreign multinational. Hanson (2001) concludes that the appropriate response to such policy competition is "to seek international co-operation between governments to prevent multinationals from extracting all gains associated with their presence in the economy."

Thirdly, Hoekman and Saggi (1999) – focussing on developing countries – emphasise the importance of political stability and credible policy commitments. Such issues are, of course, generally important for international interactions. They are, however, of special significance when it comes to investments. A foreign direct investment is a long-term commitment to engage in economic activities in a country, and among the key factors in deciding whether or where to invest would be the expectations about future market conditions and political conditions. International co-ordination and agreements may be one way of reducing the uncertainty; in particular for countries with a history of unstable political conditions.

A further argument for policy co-ordination is that it could reduce transaction costs and make all policies more transparent and coherent. The wide set of national and bilateral policy regimes that exists today could in itself be a barrier to investments that would otherwise be efficient and beneficial for the parties involved.

All the points discussed so far are good reasons to promote free, stable and transparent investment conditions. However, while some of these problems genuinely require international co-ordination to be solved, others do not. The political-economy argument, for example, is about domestic priorities and political processes within a country. There is no need for international co-ordination to solve the problem, should one wish to do so. Nevertheless, an international agreement could facilitate domestic policy changes as well.

Similar arguments apply for the question of political stability and credibility; this is mainly about domestic policies, but commitment to an international agreement may help. The fact that the bulk of the bilateral investment agreements (BITs) involve developing countries, is a clear indication that international agreements may be called for. Important elements of these BITs have to do with ensuring transparency, stability and discipline concerning expropriation, rights to transfer and so on. So for these problems the question is really whether international co-ordination through a multilateral investment agreement would be more efficient than the large set of bilateral agreements that exist today.

When it comes to international externalities and international policy games, on the other hand, non-co-operative solutions are not sufficient; there is a need for international co-operation to ensure efficient outcomes. Whether we talk about trade policies or investment policies, it is well known that if each country acts individually, and sets its policy strategically in reaction to what it expects other countries to do, all countries may end up with inefficient levels of protection or subsidies. Only binding international agreements can ensure a more efficient result. For trade policies, the GATT system (and now the WTO) plays this role. For investment policies we do not so far have a similar, multilateral system.

3.2. Is a multilateral investment agreement the best solution?

The above discussion shows that there may be good reasons to argue for international policy co-ordination and regulation. It is, however, not obvious what the right solution to these needs is. A well-functioning multilateral investment agreement could clearly solve most of the problems pointed out above. But do we need such an agreement, or are there other solutions that could achieve the same goals more efficiently or at a lower cost?

A strong argument in favour of a new multilateral investment agreement is that a successful agreement would represent a comprehensive and consistent way of dealing with investment policies. The credibility that committing to such an agreement would give, could be of great importance – in particular for developing countries in which political instability and uncertainty may work as a deterrent to inward FDI. An agreement that in addition limits the use of investment incentives and the danger of subsidy games would be even more important for developing countries. Against these advantages one should weight the difficulties of actually reaching an agreement.

In spite of the good reasons for international policy co-ordination, Hoekman and Saggi (1999) concluded that the time for a multilateral agreement was not yet there. Their arguments are twofold: first, they claim that a lot of the co-ordination failure could be fixed within the existing agreements, and secondly, they emphasise the potential difficulties in reaching a new agreement. Even if things may have changed since they wrote their article in 1999, these are important arguments that should be discussed before a conclusion is reached regarding negotiations for a new agreement.

Existing agreements in which investment issues are – or could be – dealt with, comprise the following (see Brewer and Young, 1998, for a review). For services sectors the General Agreement on Trade in Services (GATS) defines FDI as a mode of supply; hence policies towards FDI could be treated in the same way as trade policies within the GATS. When it comes to policy competition issues, it would be natural to build on and extend the Agreement on Subsidies and Countervailing Measures (SCM) such that investment incentives and subsidies are also covered. And for questions related to credibility and stability of policy and market conditions – issues of particular importance for many developing countries – the agreements on trade related investment measures (TRIMs) and on trade related intellectual property rights (TRIPs) could play a key role. The TRIMs agreement should in principle ensure national treatment for foreign-owned facilities, while the protection of intellectual property rights is of vital importance for example when it comes to technology transfers. Hence, these agreements are important in ensuring stable conditions for FDI; however, they may need further improvement and extensions to cover all the relevant aspects of investment-related policies.

This brief list of existing agreements in which policies towards FDI are or can be covered, shows that it is possible to achieve a lot without a new investment agreement. However, the list also reveals that significant changes and extensions will have to be made to the existing agreements if FDI is to be covered in a satisfactory way. Hence, to the extent that investment policies should be dealt with in a multilateral framework, it is really a question of political feasibility whether one chooses to mend existing agreements or to aim at a new comprehensive investment agreement. The alternative is, of course, to continue to let national policies and bilateral investment treaties govern FDI policies.

A possible argument against launching negotiations for a new agreement is that previous attempts to agree on multilateral investment agreements have not succeeded. Turrini and Urban (2001) study the OECD initiative for a multilateral agreement on investments (MAI), and explain why the initiative failed. According to Turrini and Urban (2001) two main considerations are at the core of these problems of reaching agreements on foreign investment. First, committing to a multilateral agreement would limit a country’s political control and freedom of action vis-à-vis multinational firms. While this may generate more FDI, it may also be considered a serious political problem in many countries. Linked to this is the question of dispute settlement mechanisms in an investment agreement; in the OECD MAI proposal dispute settlement could be state-to-state or investor-to-state relations. The latter was subject to significant opposition.

Secondly, countries are in very different positions when it comes to FDI, in particular if we include developing countries. While industrialised countries typically have more or less balanced inflows and outflows of FDI, and hence would be affected both as host and home country by new regulations, most developing countries are net receivers of FDI. To the extent that host and home countries are affected in different ways from an investment agreement, the imbalance may be an obstacle to reaching an agreement. Turrini and Urban also point to the fact that an international investment agreement that does not include all countries could lead to investment diversion.

The most important lesson to learn from previous experience is probably the need to have a process that is broad and inclusive, such that the interests and views of all involved countries are taken seriously. In the question of FDI, countries are in very different positions, and the relevant policy issues may differ significantly among countries.

3.3. The scope of a multilateral investment agreement

The ultimate aim must be a multilateral agreement that includes all relevant policies towards FDI – both barriers to FDI and incentive schemes to attract FDI. It should include policies both in the host and the home country. The agreement should be based on general principles of non-discrimination, such as most favoured nation and national treatment. And it should include a dispute settlement system, similar to what we have for trade policy disputes.

To make such an agreement based on general non-discrimination principles work, it is necessary to ensure transparency and comparability of all relevant policies. A large number of policy areas will be affected, some of which many countries would consider as within the national rather than international domain. Barriers to FDI include issues like rules for ownership structure and requirements to e.g. local content or export performance of the foreign-owned firms. Incentives to attract FDI include a wide range of policies, from direct subsidies, via the provision of infrastructure to the layout of the tax regime. To try to reach agreement on a consistent and transparent way of treating such a wide range of policies in a multilateral setting would be a great challenge. The aim must, nevertheless, be an agreement that is wide enough and good enough to ensure that most countries find it advantageous to participate.

Such an all-inclusive agreement is normally not achieved in one go. Previous experience – e.g. with trade policy agreements – indicates that it is necessary to proceed stepwise. The first step has often been to define the issues, agree on the general principles and aims, and set the agenda and the schedule for further negotiations. This may seem like a small step. It is, however, a vital step in the sense that by putting the issues – in this case policies for foreign direct investments – on the agenda for multilateral negotiations, it is made clear that general non-discriminatory rules should in principle apply. It is also an important step in the sense that the sequencing of events may play a key role for the future success. To give only one example, the success of the early GATT rounds followed to a large extent from the fact that important areas were left out of the negotiations, and that focus was on tariffs only. One consequence of the success with respect to tariff reductions was the flourishing of non-tariff barriers. In subsequent rounds the scope has successfully been extended both when it comes to policy measures to be regulated and with respect to areas that are covered. Without this stepwise procedure, it is not obvious that one would have succeeded.

In terms of a possible investment agreement, it is important to make sure that the first step includes measures that are regarded as relevant and beneficial to all parties, and that the sequencing of events is such that it seems beneficial for countries with quite different FDI positions.

3.4. The institutional setting

Finally, it is necessary to look at the institutional setting of a multilateral investment agreement, should such an agreement be realised. There are obvious reasons why WTO would be the right place, but we should also discuss whether there are arguments against such a solution.

In this report the close links between trade and FDI and between trade policies and FDI policies have been emphasised. Given these close links, it makes sense to have a common institutional setting for the multilateral rules and regulations of investments and trade. The principles governing an investment agreement should to a large extent match those of the trade agreements, and that also applies for dispute settlement mechanisms. That clearly indicates that an investment agreement should be part of the WTO system.

In terms of negotiations to reach an investment agreement there could also be advantages of linking investment and trade issues. Such a linkage gives a wider set of policies on the negotiation table, and hence more possibilities of mutually beneficial trade-off. While an investment agreement in isolation may give little scope for compensation for countries that may be in doubt of their benefits from committing to such an agreement, a "grand bargain" within the WTO system would open up for other types of trade-offs (see e.g. Hoekman and Saggi, 1999).

A third reason in favour of dealing with investment policies in the WTO is that WTO has a large number of member countries, including many developing countries for which credible commitment to investment rules could be of significant value. Although – as discussed in section 2 – industrialised countries have dominated both as host and home countries for FDI so far, there is scope for significant growth in mutually beneficial FDI flows between industrialised and developing countries in the future. This may be particularly true for production networking through vertical FDI. It is well-known that inward FDI may work as a catalyst for industrial development, through technological spillovers as well as through forward and backward linkages to local firms and industries (see e.g. Markusen and Venables, 1999). Such effects may be particularly important for developing countries. However, for such FDI to take place, the interaction between trade and investment regimes is a key determinant. Both barriers to FDI in the developing countries and international subsidy games to attract investments may prevent otherwise beneficial FDI in developing countries from taking place. Free and stable conditions for investments as well as for trade are thus important prerequisites if these potential benefits are to be realised.

One danger of trying to include an investment agreement in the WTO system is that it could put too much pressure on the system. The experience from the Uruguay-round and from the problems in connection with the start of the new round may be taken as a warning that the system may not be ready for big new issues. However, the fact that the topic is now on the agenda for the new round indicates that such questions of political feasibility may be solvable. In any case, it seems difficult to find other suitable institutional settings for a multilateral investment agreement. So if such an agreement does not appear within the WTO system, the alternative is probably to continue with a set of bilateral and regional investment agreements rather than a more comprehensive, multilateral agreement.

4. Concluding remarks – status after Doha

The Ministerial Conference of WTO in Doha November 2001 decided that negotiations on a multilateral investment agreement would take place after the next session of the Ministerial Conference. The declaration states that there is a "…case for a multilateral framework to secure transparent, stable and predictable conditions for long-term cross-border investment, particularly foreign direct investment……". The declaration emphasises the interrelationship between trade and investments, and recognises that special attention is needed to ensure that an agreement reflects the interests of both host and home countries, and in particular the special situation for developing countries. The need for technical assistance and capacity building in developing countries is emphasised. In the period till the next session of the Ministerial Conference work will be done to clarify questions having to do with scope and definition, transparency, non-discrimination, modalities of commitments based on a positive list à la GATS, safeguards and exceptions, dispute settlement mechanisms, and so on.

The Doha declaration seems to a large extent to reflect many of the topics and concerns that have been discussed in this report. The close links between trade and investments make it natural to include an investment agreement in the WTO framework. However, there are a number of obstacles that need to be dealt with before an agreement – or even negotiations for an agreement – can be launched. Many of these obstacles have to do with the fact the FDI flows are very imbalanced, and that countries thus have different interests when it comes to policies vis-à-vis FDI.

In terms of what an agreement should include, very little is said in the declaration. Transparency and non-discrimination are points that are explicitly mentioned; but how broad the negotiations should be and whether they should include all types of FDI policies is not yet clear. Based on the discussion above, it follows that the ultimate aim of an agreement should be to include all relevant policies – both barriers to FDI and incentive schemes to attract FDI. While regulating the use of barriers to FDI may have been the main motivation for introducing talks about a possible multilateral investment agreement in many countries, the growing use of policies to attract FDI may be equally distorting. If policies like special tax regimes or direct or indirect subsidies play a decisive role in determining where an FDI takes place, then the outcome need not be efficient and the potential gains from FDI need not be realised. Hence, there is a need to regulate the use of incentives as well as barriers to FDI. With incentive schemes, the danger of policy competition seems even more pronounced than for barriers, and if that is true, international co-ordination and commitment to common rules are prerequisites for an efficient outcome.

So to conclude, the Doha meeting made significant progress towards launching negotiations for a multilateral investment agreement, both in terms of actually including such an agreement on the agenda for the new round, and in recognising a number of challenges and problems. However, even if the first important step of putting the topic on the agenda has been taken, there is still a long way to go before a satisfactory agreement can be reached. And to avoid that this part of the new round should jeopardise the whole round, it is crucial to proceed stepwise and to take all the challenges and problems mentioned above seriously when preparing the negotiations.

References

Barba Navaretti, G., J.I. Haaland and A. Venables (2001), "Multinational Corporations and Global Production Networks: The Implications for Trade Policy", Report prepared for the European Commission, Directorate General for Trade.

Barry, F. H. Görg and E. Strobl (2001), "Foreign Direct Investments, Agglomeration and Demonstration Effects: an Empirical Investigation", CEPR Discussion Paper 2907, London, Centre for Economic Policy Research.

Blomström, M. and H. Kokko (1998), "Multinational corporations and spillovers", Journal of Economic Surveys, 12, 247-277.

Brewer, Thomas L. and Stephen Young (1998), "Investment issues at the WTO: The architecture of rules and the settlement of disputes." Journal of International Economic Law, 1998, 457-470.

Ellingsen, T. and K. Wärneryd (1999), "Foreign direct investment and the political economy of protection", International Economic Review, vol.40, 357-379.

Fumagalli, C. (1998), "Competition for FDI and Asymmetric Countries", mimeo., Universitat Pompeu Fabra.

Griffith, R. and H. Simpson (2001), "Characteristics of foreign-owned firms in British manufacturing", IFS Working paper 01/10.

Haaland, J. I. and I. Wooton (1999), "International competition for multinational capital", Scandinavian Journal of Economics 101, 631-649.

Hanson, Gordon H. (2001), "Should countries promote foreign direct investment?", G-24 discussion paper series, UNCTAD, 2001.

Hoekman, B. and K. Saggi (1999), "Multilateral Discipline for Investment-Related Policies?", paper presented at the conference Global Regionalism, Rome.

Markusen, J. and A. Venables (1999), "Foreign Direct Investments as a Catalyst for Industrial Development", European Economic Review 43, 335-356.

Janeba, E. and G. Schjelderup (2002), "Tax Competition: A Review of the Theory", report to the Norwegian Ministry of Foreign Affairs on the programme "Global Markets and Governance".

Schjelderup, G. (2002), "International Tax Competition: Is it harmful?", report to the Norwegian Ministry of Foreign Affairs on the programme "Global Markets and Governance".

Turrini, A. and D. Urban (2001), "For whom the MAI? A theoretical perspective on multilateral agreements on investments", CEPR Discussion paper no. 2774, London, Centre for Economic Policy Research.

UNCTAD (2000), World Investment Report 2000. Cross-border Mergers and Acquisitions and Development, United Nations, New York and Geneva.

UNCTAD (2001), World Investment Report 2000. Promoting Linkages, United Nations, New York and Geneva.