TRADE IN GOODS
259
differential treatment and that their implementation concerns be addressed. Developing countries sought more time to comply with other countries' new SPS measures, to ensure a longer 'reasonable interval' between publication of a country's new SPS measure and its entry into force, and put into practice the principle that governments should accept that different measures used by other governments can be equivalent to their own measures for providing the same level of health protection for food, animals and plants. Calls were also made to reinforce the review of the SPS Agreement, encourage developing country participation in setting international SPS standards and improve financial and technical assistance concerning SPS issues.
In April 2003 the SPS committee adopted a principle of applying special and differential treatment for developing countries. This was based on a Canadian proposal whereby members agreed to consultations whenever a developing country identifies a problem with a SPS measure. In 2004 the committee also dealt with the equivalence issue. Equivalence is the mutual acceptance of another member's standards that although different in process have the same effect. The objective is to help developing nations prove that their products are as safe as those in developed nations and to speed up recognition of equivalence of SPS measures for products previously traded or those for which information already exists. In 2002, the World Bank initiated a programme to enhance the capacity of developing WTO members to participate in negotiations and implement standards. The Standards and Trade Development Facility (STDF) included the WTO, the World Health Organization (WHO), FAO, the World Organization for Animal Health (OIE) and the Codex Alimentarius. The principal objectives of the STDF are to increase participation of developing countries in forming international standards and facilitate the implementation of existing requirements.
5.9. Trade-Related Investment
Measures
The value of sales by foreign affiliates of multinational firms now exceeds global exports of goods and services. The observed growth in foreign direct investment (FDI) is a consequence of many changes in the world economy, including the decline in communication and transportation costs, and, importantly, liberalization of FDI regimes in many countries. Perceptions about multinational firms and their effects on host countries have undergone a transformation. Most countries are now quite eager to attract FDI; many offer financial incentives to attract FDI and have concluded bilateral investment treaties (BITs). There were close to
260
THE POLITICAL ECONOMY OF THE WORLD TRADING SYSTEM
2,500 BITs in force in 2006, compared to some 400 at the beginning of 1990 (Dolzer and Schreuer, 2008). On the other hand, many countries continue to subject multinationals to performance requirements. For example, multinationals may have to comply with trade-related investment measures (TRIMs) such as local content, export or technology transfer requirements. In fact, it is not unusual to find investment incentives being offered in conjunction with performance requirements and other restrictions on FDI, perhaps to partially offset the negative impact of the latter on the likelihood of investment by multinationals. The specific type of policy used often depends on whether FDI is resource-seeking, domestic market-oriented or export-oriented (Caves, 2007). The schizophrenic nature of policy stances reflects the guarded optimism with which many countries continue to view the entry of multinational firms into their markets.
Trade-related investment measures are policies used by governments with a view to forcing foreign investors to meet certain performance standards. Trade-related investment measures often involve discrimination against imports by creating incentives (additional to tariffs imposed at the border) to source from domestic producers. The most prevalent TRIMs are local content requirements—a condition that a minimum proportion of inputs used by an investor be of domestic origin. In most circumstances such measures are inefficient. This is because they either act like a tariff on intermediate goods (this is the case for a local content requirement, where manufacturers are forced to use high cost local inputs) or as a QR (this is the case with a so-called trade-balancing requirement, which acts to restrict imports to a certain quantity). A local content requirement, although equivalent to a tariff, is inferior in welfare terms because the government does not collect any tariff revenue.
An economic case for TRIMs requires there to be domestic distortions or externalities from FDI. Absent such market failures, the optimal FDI policy is no policy at all—governments should allow for unfettered market transactions. Thus, under perfect competition, domestic content rules lower welfare by raising the price of domestic inputs: the resulting benefits to input suppliers are outweighed by the costs incurred by final goods producers (Grossman, 1981). As multinational firms typically arise in oligopolistic industries, the presence of imperfect competition in the host economy is an obvious potential rationale for intervention. Analyses of content protection and export performance requirements under conditions of imperfect competition illustrate that the welfare effects of such policies may be positive (Rodrik, 1987). However, the standard normative prescription applies: more efficient instruments can be identified to address the underlying distortions. For example, in the case of welfare-reducing anticompetitive practices resulting from market power or collusion, vigorous competition policies are called for, whereas domestic policy distortions such as tariffs should be removed at the source. This approach is implicit in the WTO, which not only aims at progressive liberalization of trade, but also prohibits the use of most TRIMs.
TRADE IN GOODS
26l
Trade-related investment measures were initially one of the more controversial topics on the agenda of the Uruguay Round negotiations. Many developing countries were of the view that attempting to agree to broad-ranging multilateral disciplines on policies affecting investment went far beyond the scope of the GATT, and that the GATT was not necessarily the appropriate forum to address investment-related policies. Certain OECD countries, the US in particular, were of the view that policies distorting investment flows could have a significant impact on trade flows, and should be subject to multilateral trade disciplines. At the start of the Uruguay Round, the US sought to negotiate rules for a long list of TRIMs, including investment-related measures such as remittance policies, ownership limitations and investment incentives. In the end, the TRIMs agreement that emerged was not very ambitious. It basically prohibits measures that are inconsistent with the GATT national treatment principle (Article III) and the ban on the use of QRs (Article XI). The agreement includes a list of prohibited measures (including local content, trade-balancing, foreign exchange-balancing and domestic sales requirements) and requires that all policies not in conformity with the agreement be notified within 90 days of entry into force of the agreement. All such measures must be eliminated within two, five or seven years, for industrialized, developing and least developed countries respectively. However, Article 5.3 of the TRIMs Agreement provides for extension of such transition periods, based on specific requests. In such cases individual members need to provide the Council for Trade in Goods with justification based on their specific trade, financial and development needs (see below).
The listed prohibited measures were already illegal under the GATT. What the TRIMs agreement essentially does is to reaffirm that GATT rules apply in this area. Although this was a point of view that was long held and defended by most OECD countries, it had been resisted by developing countries. The agreement prohibits both mandatory measures and those 'with which compliance is necessary to obtain an advantage' (such as a tax concession or subsidy). Noteworthy is that export performance requirements were not included in the illustrative list. This is somewhat inconsistent with the GATT's prohibition on the use of export subsidies, as the two instruments are very similar in effect.
Although the TRIMs agreement does not go beyond existing GATT rules, these disciplines are quite powerful. The most important TRIMs-related dispute settlement at the time of writing, the 1996 case brought by the EU, Japan and the US against provisions of Indonesia's National Car Programme, may be indicative of the future. Under the contested programme, the government granted 'National Car' company status to Indonesian companies that met specified criteria as to ownership of facilities, use of trademarks and technology. National Car companies were required to meet increasing local content requirements over a three-year period; if requirements were met, they benefitted from exemption from the prevailing luxury tax on sales of cars and exemption from import duties on parts and
2Ö2
THE POLITICAL ECONOMY OF THE WORLD TRADING SYSTEM
components. National Cars manufactured in a foreign country that fulfilled the local content requirements were also exempt from import duties and luxury tax. Such imported National Cars were deemed to comply with the 20 per cent local content requirement for the end of the first production year if the value of 'counterpurchased' Indonesian parts and components accounted for at least 25 per cent of the value of the imported cars (WTO, 1998k). The panel found that this programme violated the national treatment rule. A major reason Indonesia was targeted was that the policy measures were introduced after the entry into force of the TRIMs agreement. A number of countries apply similar policies were sheltered by the transition period agreed in the Uruguay Round, a number of which have been extended on a case-by-case basis.
In many cases, surveys show that TRIMs require firms to take actions that they would have taken anyway. For example, a policy that requires firms to export is inconsequential if firms were going to export even in the absence of such a requirement. Surveys by the US Department of Commerce for 1977 and 1982 indicated that only 6 per cent of all the overseas affiliates of US firms felt constrained by TRIMs such as local content requirements, although a far greater percentage operated in sectors where TRIMs existed. In other words, TRIMs often failed to bind (UNCTC, 1991). However, the surveys did not take into account that TRIMs may carry efficiency consequences for the world by discouraging FDI in the first place.
The available empirical evidence suggests that local content and related policies are costly to the economy. A compelling discussion of the evidence illustrating how counterproductive and damaging domestic content requirements and joint venture requirements can be for host country development is provided by Moran (2002). Moreover, domestic content requirements often do not achieve the desired backward and forward linkages, encourage inefficient foreign entry, and create potential problems for future liberalization if those who enter lobby against a change in regime. Governments constrained in eliminating costly status quo trade-related policies that aim at industrial development because protected industries are able to prevent their abolition may be assisted by an international agreement to overcome this resistance. In practice, transition periods will be important in phasing out WTO illegal programmes, as investment decisions will have been taken in the past on the basis of prevailing policies. One example of a phase-out policy is described in Box 5.8.
It was agreed in the Uruguay Round that the agreement be reviewed in the year 2000, at which time it might be complemented by provisions on competition and investment policy. In the course of this review process, the Council for Trade in Goods was to consider whether the agreement should be complemented with provisions on investment policy and competition policy. As mentioned earlier, the Singapore ministerial conference established working groups to study the relationship between Trade and Investment on the one hand and Trade and Competition Policy on the other. If negotiations had been launched in the Doha Round, this would have opened the prospect of stronger multilateral disciplines for
TRADE IN GOODS
263
Box 5.8. South African TRIMs in the automotive sector
South Africa has long had a trade policy designed to support the development of a local car industry/Eight production facilities operated in the country as of the late 1990s, producing 38 models, with an average production run of 37,000 units, a very low amount compared to international best practice. Until the mid-1980s, tariffs on cars were very . high, ranging up to 100 per cent or more, and car assemblers were subject to requirements and incentives to source locally. Local content requirements aimed to reduce screwdriver assembly operations that would otherwise be profitable because tariffs on parts and components were well below those on cars. Starting in 1989, a decision was taken to increase competition in the sector. A tariff reduction programme was designed and announced, and an export incentive scheme was created in order to encourage plants to attain greater scale economies. This involved the granting of (tradable) import credits on the basis of realized export volumes—in effect, net foreign exchange earnings counted towards the minimum (50 per cent) local content requirement. In 1995 all local content requirements were abolished and further reductions in tariffs were announced (going beyond the nation's WTO commitments), with an ultimate aim to attain a maximum rate of 40 per cent for vehicles and 30 per cent for parts in 2002.
The South African phase-out strategy of TRIMs in the car sector is interesting in that it sought to balance economic and social policies. The negative impact on the components sector of elimination of local content requirements and the concurrent gradual reduction of tariffs on components was offset in part by the incentive programme to encourage exports of automotive products. This increased demand for high-quality local output. Although the programme distorts incentives—for example, car companies have an incentive to procure high-value components in which the country may not have the greatest comparative advantage to maximize import credits—the programme led to a significant expansion of automotive exports such as leather seat covers, tyres, and exhaust systems, in the process facilitating adjustment to a policy environment without TRIMs.
Source. Black (1999).
TRIMs. However, no progress proved possible on competition and investment policies, and the TRIMs review did not result in additional provisions, decision by the members. However, it became clear very rapidly that not much could be accomplished in that respect. The working groups' work and the TRIMs review dragged on and not much happened in terms of additional provisions. Developing countries resisted any attempts to extend the scope of the TRIMs Agreement to include a broader definition of investment, re-estabiishment rights or additional restrictions on nontrade-related performance requirements. Developing countries took the opportunity of the TRIMs review to press for amendment of the treaty to reinforce its development dimension. For example, a joint Brazil-India submission called for expanding the 'policy space' to use certain TRIMs. The OECD countries generally considered that a 'watering-down' of TRIMs disciplines would set a bad precedent and argued that extensions of Uruguay Round transition periods offered
2Ö4      THE POLITICAL ECONOMY OF THE WORLD TRADING SYSTEM
enough flexibility. At the same time, however, these developing countries continued to sign bilateral investment treaties.
Developing countries were to have implemented the TRIMs Agreement and eliminated their relevant regulations by 1 January 2000. However, 26 such countries gave notice that at that time they still had a range of TRIMs-incompatible policies in existence and that they intended to maintain many of them. Most of the policies related to the auto industry or the food industry and involved local content requirements. The second most frequently notified type of TRIMs was foreign exchange balancing requirements (Bora, 2001). Developing countries have argued that the process for negotiating extensions to the duration of transition periods should be undertaken through a multilateral framework. In contrast, the EU, Japan and the US held that requests for deadline extensions should only be considered on a 'case-by-case' basis. The bilateral nature of the process caused concern among developing countries that high-income members could use the threat of rejecting requests for an extension as bargaining leverage. Disagreements over extension procedures were partly resolved in July 2000 by a decision that the council chair would oversee multilateral negotiations.
Developing countries did not represent a common front on all TRIMs issues. Given their rather open capital markets, higher income levels, and interest in agricultural trade liberalization, countries in Latin America were not particularly opposed to negotiation of TRIMs. Opposition came from countries in Asia and Africa that sought to maintain the freedom to limit the extent of foreign ownership and production within their economies. Other factors behind objections to a multilateral agreement on investment included asymmetries in the obligations to be undertaken and in the distribution of benefits, limited capacity to negotiate, and limited resources for implementation.
The positions on TRIMs taken by many developing countries in the Doha Round included a push for: (1) unlimited extensions of transitional periods under TRIMs Article 5.2; (2) an exemption from disciplines on the two performance requirements listed in the TRIMs Annex (local content and trade balancing); (3) a ban on extending the list of restricted policies; and (4) agreement that the Council for Trade in Goods automatically would grant extensions of transitional periods under TRIMs Article 5.3 to all developing and least developed countries that request them.
5.10. Conclusion
Despite the complexities of the various agreements, the GATT is basically a simple agreement. The key disciplines are nondiscrimination (national treatment and MFN),