VII Trade Policy Issues Ahead
- Arvind Subramanian
- Published Date:
- September 2000
In coming years, ESA countries will be affected by the ongoing trend of trade liberalization at the regional level (Table 7.1). The EU-South Africa free trade agreement (EU-SA FTA) will also have significant trade and competitiveness effects for ESA countries with strong trade links with both the EU and South Africa; it will also have important revenue effects for several countries, notably the BLNS countries (Box 7.1). More fundamentally, it will impel a lowering of trade barriers in ESA countries, in part because their borders are porous. Direct pressures to lower trade barriers will also arise in the context of discussions about the trade arrangements that will succeed the Lomé Convention.
|EU South Africa/SACU free trade agreement (FTA)|
|Implementation by South Africa/SACU|
|Implementation by EU|
|Implementation by South Africa/SACU|
|Implementation by non SACU|
|COMESA FTA/customs union|
|Implementation of intraliberalization2|
|Implementation of extraliberalization3|
|Renegotiation of SACU|
|Implementation of intraliberalization2|
|Implementation of extraliberalization4|
|Renegotiation of Lomé Convention|
|Implementation by Eastern and Southern African (ESA) countries|
|Implementation by EU|
|East African Community customs union|
Each of these developments raises a set of important issues, namely, the unfinished agenda on unilateral liberalization, preferential integration within the region, preferential integration with industrial country partners, multilateral liberalization and the next round of trade negotiations in the WTO, the role of complementary policies, and the role of trading partners.
The Unfinished Agenda of Unilateral Liberalization
As discussed, many ESA countries made substantial progress on trade liberalization during the 1990s, but a large unfinished agenda lies ahead. First, 9 out of the group of 22 ESA countries, including Ethiopia, Kenya, and Zimbabwe, still have restrictive regimes, with pervasive quantitative restrictions and high tariffs, often in excess of 25 percent. Second, even in countries that have liberalized their trade regimes substantially, a number of problems relating to complexity and exemptions remain. South Africa is an example of a country with an overly complex tariff system, with 72 tariff lines and a high proportion of non-ad valorem tariffs. Mozambique protects its sugar and cashew industries with highly distortive trade taxes, subsidizing firms of doubtful viability. Furthermore, collection efficiency in ESA countries is very poor (often less than 50 percent), reflecting, among other things, the continuing problems that arise from generous exemptions. In Mauritius, for instance, forgone revenues from exemptions amount to about 75 percent of collections, and about 50 percent of collections when exemptions for exports are excluded.
Regional Preferential Integration
Many of the current RTAs in the region do not seem to possess the characteristics of a natural trading bloc in the sense of their products being usefully complementary. This is particularly the case of COMESA, where countries trade less with each other than what would be expected on the basis of their levels of economic development and geographical proximity (Subramanian and Tamirisa, 2000). This reinforces the principle that preferential trade arrangements in the ESA region should be outward-oriented and pursue MFN liberalization with as much vigor as intraregional liberalization. The CBI has embodied the principle of open regionalism, and COMESA is moving in that direction too. In the case of SADC, average external tariffs are lower than for CBI and COMESA countries, but reform needs to be extended to those selected labor-intensive products where restrictions remain high.
ESA is, perhaps uniquely, distinguished by the proliferation of a number of overlapping and eventually inconsistent regional arrangements (see Figure 7.1). This is in contrast to Asia and Latin America, where RTAs have less overlap and, hence, limited scope for conflict. Several problems arise as a result of such proliferation. First, some countries that are currently members of a customs union (Namibia and Swaziland) cannot offer different preferences to other groups of countries. This problem will be compounded if COMESA implements its common external tariff in 2004 as planned. There are also complications for the SADC countries that are members of COMESA, as it is not clear whether they would follow COMESA or SADC rates. Second, when countries are simultaneously members of several RTAs, implementation of the agreements can be difficult, as incentives for inefficient diversion of trade through areas with lower external tariffs or less onerous internal procedures emerge and customs officials are faced with the task of establishing the origin of goods coming from different groups of countries. In fact, ESA countries might find rules of origin impossible to implement, particularly if they differ among countries. Rules of origin will also complicate marketing and production decisions, as well as the rationalization of production (companies will have an incentive to differentiate between production for the region with local inputs and for the rest of the world with cheaper inputs). Third, the sheer administrative and political costs and distraction stemming from multiple initiatives create difficulties. Therefore, there is an urgent need to rationalize these initiatives in order to improve the chances for their success and the investment climate in the region.
Figure 7.1Main Regional Trade Arrangements
One suggestion to encourage wider regional cooperation and liberalization is to establish a forum comprising all ESA countries and to take a closer look at the costs of these multiple initiatives. The essential requirement is not necessarily to combine the different RTAs into one, but at least to ensure that their policies are designed and implemented in a mutually consistent fashion. In any case, it would be helpful if all the RTAs were to adopt similar and low external tariffs.
Is There a Need for Compensation?
A widely held view among ESA countries is that regional integration would entail the need for compensation, particularly when one of the partners is economically dominant. While this argument may hold in principle, the case for compensation that would be directed at South Africa would be conditioned to a large extent on how open the South African market becomes, particularly in the labor-intensive sectors in which partner countries have a comparative advantage (for example, clothing and textiles). Opening these sectors would engender the perception that partner countries could gain from the agreement and, hence, mute calls for other forms of compensation (transfers). Although the principle of asymmetric liberalization in the SADC is a form of transitional compensation, it will only have the desired effect if it eventually encompasses all the sectors that are of particular importance to South Africa’s regional trading partners, including those that are currently highly protected. In the case of the Common Market of the South (MERCOSUR) and North American Free Trade Agreement (NAFTA), the question of compensation never arose because of the perception that the smaller countries stood to gain from the liberalization unleashed by the agreements. In sum, as long as the regional integration scheme is appropriately designed, a compensation arrangement should not be necessary.
Box 7.1The Impact of Changes in the Trade Environment on the BLNS Countries
Important developments in the trade environment are expected to have an impact on Botswana, Lesotho, Namibia, and Swaziland (the BLNS countries), which, along with South Africa, constitute the South African Customs Union (SACU). The EU-South Africa free trade agreement or FTA, the renegotiation of the revenue-sharing arrangements among SACU countries, the post-Lomé Convention arrangements, the SADC FTA (to a lesser extent), and the future multilateral trade round under the auspices of the WTO are expected to have the following effects:
An adverse competitiveness impact, as the EU-SA FTA will reduce many of the preferential margins that BLNS suppliers enjoy vis-à-vis the EU under the Lomé Convention. This FTA and The SADC FTA will also erode BLNS suppliers’margin in the South African market relative to EU suppliers and non-SACU SADC suppliers. This could result not only in lost trade but also in investment diversion away from BLNS countries, as the erosion of preferential margins makes them a less attractive source of supply.
A beneficial pull effect will be felt from the growth impulse imparted to South Africa as a result of these developments.
An adverse impact on revenue transfers to the BLNS countries will result from a combination of reduced total receipts (as tariffs decline under the EU-SA FTA, SADC. and the next WTO round) and reduced shares stemming from the SACU renegotiations (for any given aggregate pool).
The competitiveness impact is likely to be limited since BLNS exports do not compete directly with EU or South African products. According to a detailed study (IDS/BIDPA, 1998), the value of BLNS products threatened by direct competition is relatively small: the quantitative impact will depend on how these products are treated in the EU-SA FTA and in the post-Lomé arrangements. The pull exerted on BLNS countries as a result of higher growth in South Africa—estimated at about 1 percent—is difficult to quantify. While small, it could nevertheless offset any adverse competitiveness impact in the BLNS countries.
The revenue effect, which is probably the most significant one, is summarized in the table below.
With respect to the revenue impact, three points are noteworthy. First, the trade developments are expected to affect Lesotho and Swaziland most, as they depend more on SACU revenues than Namibia and Botswana. Second, the full revenue impact will be seen only by the year 2010 when the EU-SA FTA will be fully implemented. Third, the magnitude of impact is subject to considerable uncertainty, depending on several factors, including the new revenue-sharing arrangement under SACU, the exemptions under the EU-SA FTA, and the new WTO round. For these countries, the policy implications are clear: efforts lo enhance the domestic tax base, especially for Lesotho and Swaziland, are imperative if the adverse revenue effects arising from changes in the trading environment are to be offset.
|SACU revenges (percent of local revenues)1||15.3||48.1||27.7||50.0|
|SACU revenues (percent of GDP)1||5.5||23.6||10.1||15.6|
|Reduction in SACU revenues2||-5 to -9||-13 to -21||-9 to -14||-14 to -28|
|Value of Lomé: preferences (millions of rand)3||9.9||0.2||6.1||20.0|
Liberalization of Investment and Services
As noted before, preferential integration for ESA countries in the area of trade in goods may not be the best approach to reap the benefits of globalization. A similar conclusion can be drawn when looking at trade in services and investment liberalization, because the most efficient service providers are likely to be in the industrial countries. This does not preclude greater cooperation in services and investment among ESA countries. One possible area of cooperation would be domestic regulation in the various services sectors, where ESA countries could pool resources and aim to achieve convergence in domestic regulatory regimes. This objective is already being pursued in the context of the financial sector and could be extended to sectors such as telecommunications, power, and transport. In the limit, there could even be common regulators for ESA countries in all these areas.
Preferential Agreements with Industrial Countries
Two important imminent developments for the region are the EU-SA FTA and the renegotiation of the Lomé Convention. Since the share of non-EU countries in the trade of ESA countries is about 55 percent, there is considerable scope for inefficient trade diversion consequent upon the FTA. Unlike integration within the region, integration with the EU will have important revenue consequences, and if the efficiency gains are not forthcoming, ESA countries could be seriously disadvantaged. To counter these risks, ESA countries should continue to reduce their MFN tariffs while strengthening the domestic sources of tax revenue. Given the timetables for these agreements, their impact is likely to be felt only toward the end of this decade. This provides ESA countries with an opportunity to adjust to these developments.
Sub-Saharan African countries did not, in general, take advantage of the Uruguay Round in advancing their own liberalization (Harrold, 1995; and Wang and Winters, 1998) and failed to use the WTO as a mechanism lo lock in the reforms that were already under way. As described above, this was true in the area of both trade in goods and trade in services, as reflected in the large wedge between actual trade regimes and the regimes that were bound under the WTO. It must be recognized that in a mercantilist framework, such as the one underlying the WTO, ESA countries are disadvantaged to some extent. By virtue of their small markets, they have little to offer to their trading partners and, therefore, face little pressure from them to undertake trade liberalization.
In any new round of multilateral trade negotiations, ESA countries should at least use the WTO as a binding mechanism to lock in any reforms they undertake. In view of the history of reversals, including during the 1990s, and the attendant loss in policy credibility, it would be especially important to find a credible anchor for trade and other policies. This would apply not only to trade in goods, but also to trade in services. Developing countries have recently used the WTO services negotiations to lock in future regimes and to strengthen domestic regulations that can then facilitate further liberalization of key services sectors (Mattoo, 1999). ESA countries could also emulate this approach for trade in services.
The Further Decline of Preferences
One of the likely outcomes of the next round of multilateral trade negotiations will be the further erosion of preferential margins enjoyed by ESA countries. They will thus need to adjust to a trading environment in which reliance on preferences cannot be a basis for integration with the world economy. This erosion will stem from different sources, including further declines in industrial country tariffs on industrial products. For ESA countries, the elimination of the multifiber arrangement (the system of preferential access in textiles and clothing) and any further liberalization of the agricultural sector in OECD countries will make the biggest dent in their preferential benefits. The liberalization of agriculture and dismantling of the multifiber arrangement, by reducing world prices, will erode the rents they currently enjoy as a result of preferential quotas. For countries such as Mauritius, the loss of preferential access may have a significant impact on their growth strategy, and adjustment to the new environment acquires real significance. Similarly, the BLNS countries, which have guaranteed quotas in beef and sugar, will face declining export revenues.
Role of Complementary Policies
The Need for Fiscal Adjustment
Whether the cause is further unilateral liberalization or the implementation of regional agreements within Africa and with the EU, trade taxes will represent a steadily declining source of revenue for ESA countries in the coming years. Moreover, the magnitudes of this decline could be substantial. As the experience in the 1990s has shown, however, the phasing in of tariff reductions over a number of years leaves sufficient time for the countries in the region to adjust to the imminent changes. ESA countries need to strengthen further domestic sources of revenue—through a broadening of the revenue base, the introduction of new taxes where appropriate, and a strengthening of tax administration. Much of the improvements in the tax systems will need to come from the rationalization of exemptions and generous investment incentives. It is far from clear that these incentives really augment the total flow of investment to the region, and there is scope for cooperation among ESA countries to avoid wasteful tax competition for investment. This could mean an emphasis on greater convergence of tax regimes and on cost-effective, transparent, and well-administered incentives.
The Need for Complementary Domestic Policies
The individual case studies of trade reform reported in this paper point to the need for complementary domestic policies, in order to allow reforms to be implemented and sustained, and to unleash the favorable supply responses expected from increased integration. The determinants of growth are complex and include a number of policy, institutional, geographical, and exogenous variables (political developments, terms of trade, etc.). But experience shows that policies aimed at maintaining macroeconomic stability, removing structural bottlenecks, and creating an open and liberal investment regime are required lo promote rapid and sustainable economic growth.
Role of Major Trading Partners
Achieving integration in the world economy is predominantly determined by domestic policy actions, although a favorable external trading environment can also play a supportive role. What can ESA countries’ major trading partners do to foster such an environment? The first-best approach would be the elimination of all industrial countries’ market access barriers on African exports on a MFN basis, combined with unilateral, nondiscriminatory trade liberalization by African countries. A second approach would be for the main industrial countries to enter into reciprocal free trade agreements with ESA countries. Under this approach, ESA countries would suffer relatively little trade diversion from such comprehensive agreements because nearly all their major suppliers (the United States, EU, Canada, and Japan) would be included. A third approach would be for the industrial countries to eliminate on an MFN and nonreciprocal basis all market access barriers on products of export interest to African countries and to bind these in the WTO (Wang and Winters. 1998). Another possibility would be for trading partners to reduce market access barriers to African exports on a preferential and nonreciprocal basis and bind these also in the WTO.29 It is far from clear that this approach, involving a continuation and extension of existing nonreciprocal preferential arrangements, such as the Lomé Convention, would be beneficial. The African experience (under the Lomé Convention and Generalized System of Preferences) demonstrates that preferences have not led to successful integration and may actually have encouraged inefficient specialization.
The first two approaches discussed above would be preferable but difficult to achieve in the near future. The third one would be nondiscriminatory and, hence, consistent with WTO rules, but its quantitative impact would be uncertain, given that MFN tariffs on African exports are generally low. There are, however, some specific actions industrial countries might undertake in the short run to promote the integration of African countries into the world economy.
MFN bound and applied tariffs on agricultural goods of interest to ESA countries could be eliminated, market access commitments could be implemented transparently, and exemption from special safeguards measures could be granted. Particular attention should be paid to tariff peaks and also to tariffs on processed goods, so that tariff escalation might be reduced (on wood and leather products, for example). If, as a second best, access is granted to ESA countries preferentially, both new and existing preferences should be tariff free, unconditional, and free of ceiling quotas. In addition, market access would be improved if industrial countries simplified their rules of origin requirements.
The phasing in of Uruguay Round liberalization on products of export interest to Africa could be accelerated. Tariff cuts on textiles and garments and the abolition of quotas and special safeguards for these products, for example, will encourage exports and, if offered preferentially, assist the ESA countries to gain market experience before the multifiber arrangement is phased out and a much more competitive environment is ushered in.
More generally, tariff escalation should be reduced by extending tariff cuts to all stages of production to remove obstacles to the processing of primary commodities and to encourage export diversification.30 Even though when preferences are taken into account sub-Saharan African countries face low average tariffs in industrial countries,31 these low tariffs do not extend to processed and temperate agricultural products or to textiles and garments. Outside the Lomé Convention, African countries face the same high tariffs and nontariff barrier regimes as the industrial countries’ other non-FTA partners, even though their low level of industrialization means that their exports attain neither the level nor the composition necessary to threaten sectors in industrial countries.
Although the Uruguay Round agreements allow contingent protection actions in particular circumstances, the industrial countries could exempt African countries from antidumping, countervailing, and safeguard actions. Starting from very low bases, African exports can sometimes record very rapid growth rates and be hard hit by contingent protection measures. The absolute volumes of African exports are, however, not large enough to be a threat to industrialized countries, even if they approach the thresholds set out in the Uruguay Round agreements.
This discussion of the main trade policy issues facing ESA countries leads to the following conclusions. First, for a number of ESA countries, particularly those with restricted trade regimes, there is an urgent need to push ahead with unilateral liberalization so that the benefits of reduced rent-seeking activities, increased efficiency, and improved export performance can be reaped. Second, liberalization should extend beyond trade in goods to cover services and investment regimes. Third, in any new round of multilateral trade negotiations, ESA countries should use the WTO to lock in any reforms they undertake and reap the attendant benefits of signaling the durability of reform efforts. Fourth, there is a need to rationalize the multiplicity of regional initiatives and to embrace the principle of open regionalism by reducing barriers to trade with all countries, thereby minimizing the risks associated with preferential integration. Fifth, industrial countries should assist the trade liberalization efforts of ESA countries by reducing market access barriers and eliminating restrictions on African exports. An attractive option in this regard would be for the major industrial countries lo negotiate reciprocal free trade agreements with ESA countries. This would improve market access for ESA countries while also furthering their own liberalization efforts. Lastly, it is important to emphasize that the success of these efforts will depend critically on a comprehensive set of complementary domestic policies aimed at ensuring macroeconomic stability, strengthening domestic institutions, and providing a more secure and predictable environment for investment.