Journal Issue

Reaching the MDGs

Brad McDonald, Rob Gregory, and Katrin Elborgh-Woytek
Published Date:
September 2010
  • ShareShare
Show Summary Details

I. Introduction

1. Trade is an engine of development and can contribute strongly to reducing poverty.

Yet, due to access barriers abroad and obstacles at home, exports of the poorest countries are far below potential. Despite accounting for nearly 1 percent of global GDP, these “least-developed countries” (LDCs) account for under ½ percent of global non-oil exports—a figure that, in contrast to the dynamic trade growth of other developing countries, has been virtually unchanged over the past 15 years (Figure 1)1. This structural problem has become more urgent as the crisis reduced LDC exports (IMF, 2010).

Figure 1.LDC Exports and GDP, 1995–2008

(Share of world total, percent)

Source: IMF, Direction of Trade Statistics, 2010.

2. Following the financial crisis, a broad range of policy actions is needed to regain momentum toward achieving the MDGs. Low-income countries (LICs) need to sustain the good macroeconomic policies that helped them to weather the crisis relatively well. However, they also face massive financing needs if they are to close an “infrastructure gap” and address climate change—two major impediments to their development. Even faced with the need to address fiscal sustainability and repair and reform their financial systems, advanced market countries (AMs) should deliver on aid commitments. Against this backdrop, trade is one key area where action can spur growth and development and aid LICs’ progress toward the MDGs.2

3. There is scope for more and better trade preferences, along with several complementary actions, to bolster support for LDC exports. Prompted by the 2000 United Nations Millennium Declaration and the 2005 WTO Hong Kong Ministerial Declaration (Box 1), some countries have enhanced their trade preferences for LDCs—the share of LDC exports eligible for preferential access increased from 35 percent in the late 1990s to over 50 percent today. Nonetheless, with relatively high tariffs (“tariff peaks”) concentrated in agriculture and low-wage manufactures (sectors accounting for about half of LDC’s non-oil exports), market access barriers remain substantial. The AM and major Emerging Market (EM) countries can do more. Estimates suggest that expanding the coverage of their trade preferences could alone lead to increased annual LDC exports of some US$10 billion (Laborde, 2008, and Bouët and others, 2010), about 2 percent of LDC GDP, with other proposed reforms of preference schemes multiplying these benefits.3 Complementary steps are also needed to cut distortions in world agricultural markets, help LDCs to meet product standards, and to provide more effective “Aid for Trade.”

Box 1.What Have Countries Committed to on Market Access for LDCs?

Recognizing that LDCs face special difficulties in responding to the challenges of globalization, the 2000 United Nations Millennium Declaration set out a series of goals and commitments to help overcome these obstacles. In the area of trade, advanced countries committed to adopt “a policy of duty- and quota-free access for essentially all exports from the least developed countries.” The original intention was to have this policy implemented by May 2001.

WTO members committed to the objective of duty-free, quota-free market access for products from LDCs in the 2001 Doha WTO Ministerial Declaration, and elaborated on this in the 2005 Hong Kong WTO Ministerial Declaration. There, members agreed that developing countries “in a position to do so” should also provide duty-free and quota-free market access on a lasting basis for all products originating from all LDCs. In an important caveat, however, developed WTO members facing difficulties at that time to provide market access could instead provide duty-free and quota-free market access for products originating from LDCs under at least 97 percent of tariff lines.

4. To benefit fully from better trade opportunities, LDCs would need to tackle aggressively their own high tariff and nontariff barriers to trade. By restricting imports, these barriers effectively tax exports and hamper the ability of LDCs to gain from existing and future export opportunities. A reasonable goal within a few years might be for a maximum tariff rate of 25 percent and a simple average below 15 percent. Nontariff barriers, such as non-critical road blocks and checkpoints, and overly burdensome customs practices, significantly slow trade and should be removed. Service sector reforms would help to cut costs of key inputs such as transport and communications.4

5. In the following sections we translate these themes into specific actions.5 While our focus is on LDCs, many of the actions would also benefit other LICs. Allowing LDCs to source inputs from non-LDC LICs while remaining eligible for trade preferences would allow all LICs to share in the benefits. Cuts in domestic farm subsidies, particularly by AMs, would also help non-LDC LICs. Many “Aid for Trade” programs and initiatives to help countries meet product standards cover not only LDCs, but all LICs.

Action 1: Conclude the WTO Doha Round

6. Perhaps the single most important step that the international community can take to support the exports of LDCs and LICs—and, indeed, of all countries—is to conclude the WTO Doha Round. Broad multilateral trade reforms would spur global trade and growth and foster a global macroeconomic environment conducive to development. Most critically for all LICs, which need a secure global trading environment to export and to attract investment, concluding the Round would bring the added security of tightened WTO trade rules.

7. The LDCs and LICs stand to gain in several other ways from a WTO Doha Round conclusion. This would lock in certain levels of duty-free, quota-free treatment of LDC exports and cut allowed levels of domestic farm subsidies, although (as we stress below), in their actual policies the AMs and major EMs should go beyond these commitments. Trade facilitation negotiations would enhance the transparency of trade and customs regulations and improve border management systems, reducing trading costs (Martin and Mattoo, 2010).

II. How Advanced Market Countries Can Help

8. Most trade preferences provided by AMs to LDCs are either incomplete or costly to utilize (Table 1). Moreover, for many products LDC preferences are no more generous than those offered to the larger and more diversified EM countries. The package of possible actions outlined in this section would expand preferences and better target benefits toward LDCs and LICs. At present, only 1 percent of AM imports come from LDCs.

Table 1.Preferential Duty-Free Access for LDC Exports to Developed Countries
SectorNumber of tariff linesaPreferential duty free importsb
In percent of total tariff linesIn percent of total imports

Percentage of tariff lines exempted of duty under preferential LDC schemes in relation to dutiable tariff lines under MFN regime (excluding all MFN duty-free treatment).

Percentage of import value exempted of duty under preferential LDC schemes. Data refer to the LDCs as a group.

Source: WTO, 2010a.

Percentage of tariff lines exempted of duty under preferential LDC schemes in relation to dutiable tariff lines under MFN regime (excluding all MFN duty-free treatment).

Percentage of import value exempted of duty under preferential LDC schemes. Data refer to the LDCs as a group.

Source: WTO, 2010a.

Action 2: Extend Duty-Free, Quota-Free Coverage to All Products from LDCs Without Delay

9. Preference schemes for LDCs vary widely in product and country coverage (Box 2). The EU “Everything but Arms” initiative provides duty-free, quota-free (DFQF) access to essentially all products from all LDCs. Most other major preference providers limit DFQF—to as little as 80 percent of the items in a country’s tariff schedule (“tariff lines”)—with broader coverage available only to selected LDC regions (Elliott, 2009). With 90 percent or more of LDC exports to each major advanced market being in less than 3 percent of tariff lines, even a few exclusions can matter greatly. Moreover, exclusions tend to be on labor-intensive items that face high regular (“MFN”) tariff rates.6

10. Providing full DFQF access could sharply increase LDC exports. Laborde (2008) estimates that broadening the coverage of preferences by major AMs would generate increased annual LDC exports of US$2.2 billion, equal to about 5 percent of LDC non-fuel exports to AMs and 6 percent of net official development assistance (ODA) from OECD countries to LDCs. As the increased exports would be equivalent to 0.02 percent of their total imports, they would lead to only very limited adjustment pressure in the AMs.

Action 3: Make Origin Rules More Flexible and Consistent

11. Restrictive Rules of Origin (RoO) limit the use of preference schemes (Box 3).7 Preference schemes with more liberal origin rules allow producers to source inputs flexibly. With more restrictive origin rules, in contrast, producers must limit input sourcing to suppliers at home or in the preference-providing country, even when doing so is more costly. This can be a particular problem for less-diversified LDCs, which depend on inputs (intermediate goods, processes, or patents) from other countries. Since origin rules are often complex, administrative burdens on customs administrations and potential exporters pose additional costs or outright obstacles. Altogether, because of these obstacles and costs, perhaps a quarter to a third of preference-eligible imports do not gain preference, while some trade that might have taken advantage of better-designed preferences is never undertaken.

Box 2.Preferential Access for LDC Exports in Key Advanced Markets

Advanced market countries provide duty-free treatment for many or all LDCs on 91 percent of dutiable MFN tariff lines (on a weighted average basis) (WTO, 2010a). The schemes for LDCs include broader product coverage or deeper preferences than the generalized system of preferences (“GSP”) schemes that these countries offer to a much broader group of developing countries.

Canada’s Least-Developed Countries’ Tariff (LDCT) program provides preferential market access to 49 LDCs. Since 2002, Canada has eliminated tariffs and quotas on 99 percent of tariff lines (excluding supply-managed agricultural goods), including on previously-excluded textile and apparel products. The 2004 liberalization of rules of origin (RoO) and cumulation provisions helped to promote South-South trade. Partly as a result of specific RoO for textiles and apparel, Canadian textile imports from LDC almost tripled in four years (WTO, 2007).

The European Union’s Everything But Arms arrangement provides full DFQF on all products (except armaments and munitions) from LDCs. This follows the elimination in October 2009 of the last remaining restrictions—for rice and sugar (EU, 2009).

Under Japan’s GSP scheme, preferential access is granted to imports from 105 developing countries, covering 98 percent of tariff lines, with restrictive rules of origin and cumulation provisions. Japan’s preference scheme for LDCs, in place through 2011, allows DFQF access on more than 98 percent of tariff lines. Cumulation is restricted to Japanese inputs (Elliott, 2010).

The United States provides duty free treatment on 6,500 tariff lines for imports from 43 African countries under the African Growth and Opportunity Act (AGOA). Benefits under AGOA are concentrated in a limited number of countries and products, with imports of mineral fuels accounting for 91 percent of U.S. imports under AGOA in 2009, and textiles and vehicles being other key sectors (WTO, 2010a). Other LDCs are eligible for the U.S. GSP, covering 4,650 tariff lines and available to about 130 developing countries. For most goods, AGOA RoO require a minimum local value added of 35 percent and allow cumulation across all AGOA beneficiaries.

12. Rules for trade preferences can be made more flexible, transparent, and consistent. Simple, flexible rules—such as a 1/3 value added threshold or a requirement for modest transformation of imported inputs (e.g., a change in the 6- or 4-digit tariff classification of the good)—would allow LDCs to use a wider variety of input sources and to reduce compliance costs, leading to the fuller utilization of preference schemes. Such rules would implicitly acknowledge LDCs’ low capital intensity and lack of horizontal or vertical integration. One good example—the U.S. Africa Growth and Opportunity Act (AGOA)—uses, for most goods, a minimum local value added threshold of 35 percent of the final price. Another positive reform example was Canada’s 2004 RoO modification, which set a minimum domestic value added threshold of 25 percent of value and was a key factor behind a 200 percent increase in textile and apparel imports from LDCs over the next four years (WTO, 2007). By harmonizing origin rules, preference providers can also give LDC exporters the flexibility to serve different markets without extra costs, providing a greater incentive for LDC producers to invest in exporting.

Box 3.Rules of Origin and Cumulation Provisions

Rules of origin (RoO) determine whether a good is considered to have “originated” in a preference-eligible country (Krueger, 1993). The restrictiveness of rules of origin is determined by two aspects: (i) the extent of activity that must be undertaken in the beneficiary country (the “origin rule”), and (ii) the extent, if any, to which imported inputs from either the preference provider or other preference beneficiaries counts toward meeting the origin rule (the “cumulation provision”). While both influence how easily preferences may be used, cumulation provisions also may encourage or discourage trade among preference beneficiaries. The WTO Agreement on Rules of Origin pertains primarily to non-preferential RoO (e.g., as used in antidumping actions) but includes transparency and other provisions for preferential RoO.

The origin rules differ widely across preference programs. They are frequently specified according to: (i) a minimum ratio of domestic value added to price, or (ii) sufficient transformation of an imported good, measured by a change in its tariff classification. Value added thresholds are more flexible if relatively low, such as at 25 to 35 percent. A change in tariff classification rule is more flexible if it requires only a change in the six-digit classification, rather than in a (broader) four-digit or two-digit classification. In some preference programs the RoO for many specific tariff lines are specified in an ad hoc manner, such as with reference to a particular input (e.g., the source of the textiles used in manufacturing apparel).

Cumulation provisions affect the overall restrictiveness of RoO by specifying the extent to which inputs from other countries may be used by the preference-eligible country. Cumulation rules therefore determine how easily preferential partners or preference beneficiaries can trade among themselves, using intermediate goods or processes originating in other countries. Cumulation may be bilateral (between the preference provider and preference beneficiary), diagonal (involving participants in a multi-partner preference program), regional (among members of a formal regional group), or full (among all beneficiary countries). By allowing inputs from two or more parties to be counted together, a cumulation rule can add flexibility to a preference scheme. In contrast, narrow or restrictive cumulation provisions rules exacerbate the effects of restrictive RoO and fragment production relationships.

The RoO influence the sourcing of inputs and thus the economic effects of a preference program. They can be a source of distortion when exporters turn to less efficient, more costly input sources to become eligible for preferences. In this way, restrictive RoO protect high-cost intermediate input suppliers from outside competition. Research into individual preference schemes indicates that complex RoO can limit LDC benefits to as little as a quarter of their potential (Mattoo and others, 2002). Ahmad (2007) discusses the role of RoO in the textiles and clothing sector. The costs of documenting and administering RoO can pose additional costs of 3 percent of value (Hoekman and Özden, 2005).

Action 4: Provide Extended Cumulation Provisions

13. How preferences affect third countries is influenced by cumulation provisions—an aspect of RoO. Under “bilateral cumulation,” only inputs sourced domestically or from the preference-providing country apply toward meeting the origin rules. Cumulation provisions may, however, be extended to allow sourcing from, for example, neighboring countries or from all preference beneficiaries. The AGOA, for example, allows cumulation across all AGOA beneficiaries, encouraging South-South trade. Similarly, the introduction of the Pan-European Cumulation System in 1997—under which then non-member central European countries could trade among themselves before exporting with preferences into the EU—spurred considerable increases in trade and investment in the region (Baldwin, 2006).

14. LDC preference schemes could extend cumulation to all LDCs and other non-LDC LICs.8 There would be three advantages: (i) LDCs could meet origin rules more easily and at lower cost; (ii) this would encourage South-South trade, and (iii) allowing LICs to share the benefits from LDC preferences would blunt the somewhat artificial distinction between LDCs and other LICs. Preference providers could extend cumulation even more broadly to include all countries to which they provide preferences, whether under unilateral arrangements or (reciprocal) regional trade agreements.

Action 5: Tilt Preference Benefits toward LDCs and LICs

15. Some AMs also provide preferences to the more advanced developing countries—including major EMs—under their standard “GSP” schemes. Together with tariff preferences under (reciprocal) regional trade agreements, these reduce the effective preference margins available to LDCs (Low and others, 2005).9 How rapidly to narrow or withdraw preferences for large, dynamic EM exporters is a complex issue that should consider the impact on the EM and the importing country. But the prospective benefit to LICs and LDCs warrants strong consideration. Graduation provisions should always be transparent and predictable, with ample notice of withdrawal. Another way to enhance benefits for LDCs is to renew their preference schemes well in advance, allowing time for investors to make appropriate decisions.

Action 6: Provide Government Procurement Preferences

16. Most AMs provide domestic suppliers some preference in government procurement. For certain types of procurement or for contracts above a threshold value, some countries that use procurement preferences waive the application of the preferences to permit imports from LDCs. This practice should become more common, allowing LDCs to compete on par with domestic suppliers whenever practicable.

III. How Emerging Market Countries Can Help

17. Major emerging markets countries (EMs) can potentially provide very valuable preference benefits to LDCs. Owing in part to the composition of the rapid economic growth in certain EMs, exports from the LDCs to Brazil, China, and India grew by an annual average of over 30 percent during 1999–2009, compared to around 12 percent a year for world exports to those countries. Those emerging markets now account for a third of all LDC exports.10 With substantial reforms since the 1990s, these EMs have reduced average MFN tariff rates to about 11 percent, but tariffs remain some 6 percentage points higher than those of the major AMs. The magnitude of trade flows and levels of MFN tariffs suggest large potential trade preference benefits for the LDCs.

Figure 2.LDC Exports, 1995–2008

(Share of total imports, percent)

Source: IMF, Direction of Trade Statistics, 2010.

18. Several major EMs have introduced and expanded LDC trade preferences, but coverage is selective. Progress since the 2005 WTO Hong Kong Ministerial Declaration has improved market access conditions for LDC exports.11 Amongst others, Brazil, China, India, and the Republic of Korea12 have announced or implemented LDC preference schemes. As they are at an earlier stage of implementation than those of the AMs, these preference programs have room to grow, albeit at a pace consistent with the remaining development needs of the new preference providers. The key directions for expansion and improvement are broadly similar to those for the AM schemes.

Action 7: Extend DFQF Coverage to All Products from LDCs by 2015

19. While recent steps are welcome, there are significant gaps in the coverage of preference schemes of large EMs (Box 4). Reflecting their relatively recent introduction and the incomplete development level of the preference providers themselves, these gaps are wider than those in the schemes of the industrialized countries. And, as with the industrialized countries, exclusions are prominent in those labor-intensive products for which LDCs might be expected to have a comparative advantage: agriculture, textiles, apparel, and footwear. Again, LDCs’ concentrated export structure suggests that even a small number of product exclusions can sharply limit the benefits of trade preference programs.

Box 4.LDC Preference Schemes of Major EMs

China announced duty-free access on 190 tariff lines beginning January 2005, and extended this to a total of 440 tariff lines in November 2007. The average margin of preference on these 440 lines is 10.4 percent. These preferences are available to 42 LDCs, including 31 in Africa and 11 in the Asia-Pacific. The scheme’s origin rules require a minimum domestic value added (40 percent of the total price), or a change in 4-digit tariff heading. There is no provision for LDC cumulation in the China scheme.

India announced an LDC preference scheme in April 2008, offering preferences on 94 percent of tariff lines comprising 92.5 percent of LDC global exports. Fifty LDCs are eligible to participate, but countries need to register. The origin rules require that domestic value added be at least 30 percent of the price and a change of 4-digit tariff category. The final process of manufacture must also be performed in the exporting country. Cumulation across LDC beneficiaries is not allowed.

Brazil announced in 2008 its intention to begin offering DFQF access. The Brazilian authorities expect that 80 percent of tariff lines will be covered initially, expanding subsequently to 100 percent.

The Republic of Korea introduced LDC preferences in 2008 and in April 2010 expanded product coverage to 85 percent of tariff lines. This is expected to increase to 95 percent of lines in 2012. Korea’s preferential rules of origin require products to be wholly obtained in the beneficiary country or that at least half of the final value of the product is produced domestically.

20. Research indicates that full DFQF product coverage for all LDCs in EM preference schemes could considerably raise LDC exports. For LDCs as a group, Bouët and others (2010) estimate the annual increase in LDC exports as US$7 billion.13 As with AM schemes, excluding even a few percent of tariff lines from DFQF would cut sharply into LDC benefits. With such large potential gains, there should be strong encouragement for major EMs to extend full DFQF. Model results also indicate that—given the respective sizes of the major EMs and of LDCs—the aggregate domestic impact on the EMs is tiny. Nevertheless, with some adjustment pressures likely within some narrow product categories, EMs may in some cases need several years to implement these LDC benefits.

Action 8: Make Origin Rules More Flexible and Consistent

21. Benefits for LDCs from EM preference schemes are handicapped by RoO similar to those employed by the advanced market countries (Box 4). Despite complexities, the preference schemes of EMs have certain positive attributes that can be replicated in other schemes. Under China’s scheme, for example, origin (and thus preference benefits) can be conferred on a product based either on a minimum local value added threshold or a change in tariff heading. India’s low 30 percent value added threshold gives potential LDC exporters flexibility in sourcing their inputs.

Action 9: Provide Extended Cumulation Provisions

22. Quite limited cumulation provisions under most EM schemes make them more costly and difficult to use (Box 3). There is a risk that this could promote “hub and spoke” trade between the preference provider and the individual beneficiary without also promoting trade among the beneficiaries. As with AM schemes, this can be overcome by extending cumulation provisions. LDC beneficiaries should be able to source (or “cumulate”) inputs from all LDCs and LICs while remaining eligible for preferences. This would provide the added flexibility needed for effective use of preferences, provide some (albeit more limited) benefits to non-LDC LICs, and promote South-South trade.

Action 10: Provide Government Procurement Preferences

23. Several recent WTO Trade Policy Reviews highlight the scope to improve the efficiency of government procurement within the EMs. With EMs facing similar or greater challenges than the advanced countries in opening their procurement schemes to greater competition, they should consider ways to allow greater LDC involvement. Doing so would be a valuable complement to what the EMs can do with conventional trade preferences.

IV. Making the Most of Preferences: Complementary Measures

24. Even with enhanced trade preferences, complementary actions are needed for exports of LDCs and non-LDC LICs to meet their potential. First, to take full advantage of increased market access opportunities, LICs need help to meet often challenging product standards of major importers. Second, reducing trade-distorting farm subsidies around the world would facilitate higher and more stable agricultural commodity prices, which would in turn support greater agricultural investment in LICs. Third, LICs need to reform their own policies to enhance flexibility and reduce the anti-trade bias often present in their trade and tax regimes. And, fourth, more effective regional integration—policies, institutions, and infrastructure—in LIC regions would expand the size and scope of their markets and bring economies of scale. These steps should all be anchored in a secure, open multilateral trading system with strong rules and enforcement that protect the interests of the less powerful LDCs and LICs.

25. Initiatives are under way in these areas. These complementary actions will take time to develop and implement. Some require legislative changes or budget resources, while others require coordination on complex issues by many countries. Whatever the obstacles, we underscore their importance to promoting LIC exports and to poverty reduction.

Action 11: Reduce Trade-Distorting Domestic Farm Subsidies

26. Beyond protective trade policies, OECD member countries provide annual farm support of about US$130 billion.14 While many have redesigned farm support mechanisms to reduce the distortions on production and trade, these subsidies tend to stimulate output and depress market prices. And, because payments under many schemes rise when world prices fall, they also exacerbate market volatility. Major EMs provide lower levels of overall producer support than OECD countries, but the bulk of that support is provided through highly distortionary trade measures. In both AMs and EMs, support varies across commodities and is often greatest for such commodities as cotton and sugar, which are of export interest to LDCs.

27. Cotton is perhaps the example that best illustrates the issues at stake. It accounts for about 20 percent of LDCs’ agricultural exports and for about a quarter of total merchandise exports in some West African LDCs. Cotton farmers receive substantial domestic subsidies in some key industrialized countries and are protected with high tariffs in some major emerging markets. Calculations by Anderson and Valenzuela (2007) suggest that reforming domestic cotton subsidies could increase annual cotton exports of West African LDCs by some US$200 million, as higher market prices would encourage greater output.15

28. Even where LDCs and other LICs are not currently net exporters, reducing trade-distorting domestic subsidies will often benefit them over time. Many LDCs and other LICs are already net food exporters. Where they are not, it is often in part because depressed, volatile world prices have inhibited agricultural investment. With the additional investment that could result from higher and more stable prices, in time many more LICs would benefit from the reduced subsidies. Nonetheless, as subsidies are withdrawn those LICs that remain net importers could face a gradual reduction of their terms of trade. And even in the bulk of countries that stand to benefit overall, it is important to protect vulnerable groups such as the urban poor.

Action 12: Assist LDCs and Non-LDC LICs in Meeting Standards of Importing Countries

29. Exporters from LICs face particular challenges in meeting standards in the markets of major importers.16 These standards are often necessary for a smoothly functioning market or to avoid the spread of pests and diseases and protect the food supply. However, LICs seem to be disproportionately affected (Disdier and others, 2008). Reasons may include producers being reluctant to invest in the capacity to meet the standards because of their small scale, as a result of which they may frequently forego export opportunities.

30. Help can be provided in at least two ways. First, AMs and major EMs can provide additional financial and technical assistance to help LICs to identify priorities, to plan and execute strategies to meet product standards, and to participate more actively in international standard setting activities (Jaffee, 2006). Efforts such as the Standards and Trade Development Facility17 (a partnership of international organizations) and those of international agencies and bilateral providers could be expanded. Second, when formulating standards, AM and EM governments should consider the effect on LICs. Although individual countries sometimes need to diverge from an existing international standard, much scope exists to harmonize standards or, in many cases, to recognize the equivalence of others’ standards or testing. With support in these areas, more LIC and LDC producers will also find it worthwhile to invest in meeting standards.

Action 13: Fill Key Gaps in Providing Aid for Trade

31. More and better-focused “Aid for Trade” would help the LICs to take advantage of available opportunities. Donor commitments under “Aid for Trade” rose in 2008 to US$42 billion, 62 percent above the 2002–05 baseline, accounting for 37 percent of sector allocable ODA.18 The Enhanced Integrated Framework ( is now fully operational in assisting LDCs to integrate trade into their national development plans. While the emphasis on trade-related assistance as a separate initiative is rather new, some early research concludes that—when well-targeted (e.g., at “trade facilitation” activities that are often directly related to trade)—“Aid for Trade” cuts trading costs and is indeed reflected in increased exports (Cali and te Velde, 2009).

32. Despite tentative successes in the “Aid for Trade” initiative, important gaps remain. Much attention has been given to improving hard infrastructure and productive capacity, but Hoekman and Wilson (2010) argue that “the efficiency, variety and costs of services inputs are critical for the competitiveness of firms and farmers.…” Improving services-related policies and regulations requires more support. A second gap lies in regional cooperation. For land-locked and many other countries, more effective and supportive regional institutions, policies, and infrastructure are a necessary condition for benefitting fully from export opportunities. Overcoming the inadequate support for regional trade-related projects requires broadening the range of support instruments (now rigidly focused at the national level) and improving the capacity of local partners, such as regional secretariats. Enhancing support for regional integration strategies is a specific objective in the 2010–11 Aid for Trade Work Program of the WTO Committee on Trade and Development (WTO, 2009).

Action 14: LICs Should Reform Their Own Trade and Tax Regimes

33. LIC trade regimes have improved, but a considerable anti-trade bias remains. This bias reflects moderate to high average levels of protection, and uneven and unstable tariffs, trade-related taxes, and other trade-related policies. Behind trade restrictions, highly protected sectors draw scarce resources (such as capital and skilled labor) from other sectors in which the country has a comparative advantage, leaving them less competitive.

34. Redoubling LIC trade-related policy reforms would help them to take the best advantage of export opportunities. The focus should be on reducing tariff dispersion and average tariff rates. For the next few years, a reasonable goal might be for a maximum tariff rate of 25 percent and a simple average MFN tariff below 15 percent—implying somewhat lower, but much more even, levels of protection, with little revenue impact. Further tariff rate reductions would be even more beneficial.19 The IMF and others can help LICs to improve tax and customs administration capacity to avoid resort to discriminatory measures beyond the tariff.

Action 15: Take Steps to Guard against Disruptions in the Global Trade System

35. Disruptions to the global trading system could impact LICs most strongly. Prompt action has helped to avert a widespread resort to protectionism in the wake of the global financial crisis. But unemployment remains high in many countries, and fiscal and monetary stimulus policies—which helped to contain protectionist sentiment—will gradually be withdrawn. Further enhancing the monitoring of protectionist measures and concluding the Doha Round to reduce the scope for backsliding are needed (Gregory and others, 2010). Those protectionist measures that were implemented as a result of the financial crisis now need to be rolled back. Climate change and related policy responses may pose even more complex challenges. Work is needed to ensure that any related trade measures (e.g., border tax adjustments) are genuinely equivalent to domestic policy measures and do not discriminate against trade partners.

V. Conclusions: Moving Ahead

36. An open and stable global trading system is critical for the development prospects of poor countries. Concluding the WTO Doha Round is perhaps the single most important step that can be taken to support LICs. Tightened trade rules would guard against backsliding from present levels of openness. A more prosperous world economy, with more trade opportunities for LICs, would set the best possible global environment for development.

37. Trade preferences help to raise exports of the poorest countries, supporting poverty reduction through economic growth. Existing preferences already help to offset some of the high trade barriers that exist in areas such as agriculture and light manufacturing. But coverage is incomplete and preferences often prove difficult to use in practice. The possible actions we identify for AMs and EMs would substantially increase the value of these preferences for LDCs, in a manner that could also benefit non-LDC LICs.

38. Complementary measures are needed to benefit all LICs. Trade capacity should be increased by: (i) filling gaps in Aid for Trade to bring more effective regional integration in LIC areas and to build more efficient trade-critical services, such as in transport and communications; (ii) helping the LICs to meet standards in importing countries and to participate actively in standard setting; (iii) encouraging agricultural investment in LICs by reducing support for farmers in AMs and EMs; and (iv) reforming domestic trade regimes to reduce their anti-trade bias.


    AhmadM.2007“Impact of Origin Rules for Textiles and Clothing on Developing Countries,”Issue Paper 3 (Geneva: International Centre for Trade and Sustainable Development).

    • Search Google Scholar
    • Export Citation

    AndersonK. and E.Valenzuela2007“The World Trade Organization’s Doha Cotton Initiative: A Tale of Two Issues,”The World EconomyVol. 30No. 8 pp. 12811304.

    • Crossref
    • Search Google Scholar
    • Export Citation

    AndersonK. and A.Winters2008“The Challenge of Reducing International Trade and Migration Barriers,”Policy Research Working Paper 4598 (Washington: World Bank).

    • Search Google Scholar
    • Export Citation

    BaldwinR.2006“Multilateralising Regionalism: Spaghetti Bowls As Building Blocs on the Path to Global Free Trade,”CEPR Discussion Paper 5775 (London: Centre for Economic Policy Research) August.

    • Search Google Scholar
    • Export Citation

    BouëtA.D. LabordeDebucquetE.Dienesch and K.Elliott2010“The Costs and Benefits of Duty-Free, Quota-Free Market Access for Poor Countries: Who and What Matters,”Working Paper 206 (Washington: Center for Global Development).

    • Search Google Scholar
    • Export Citation

    CaliM. and S.Page2010“Reforming Trade Preferences for LDCs” (London: Overseas Development Institute).

    CaliM. and D. teVelde2009“Does Aid for Trade Really Improve Trade Performance?” (unpublished; London School of Economics and Overseas Development Institute).

    • Search Google Scholar
    • Export Citation

    DisdierA.B.FekaduC.Murillo and S.Wong2008“Trade Effects of SPS and TBT Measures on Tropical and Diversification Products,”Issue Paper 12 (Geneva: International Centre for Trade and Sustainable Development).

    • Search Google Scholar
    • Export Citation

    ElliottK.2010“Open Markets for the Poorest Countries: Trade Preferences That Work” (Washington: Center for Global Development).

    • Search Google Scholar
    • Export Citation

    ElliottK.2009“Opening Markets for Poor Countries: Are We There Yet?,”Working Paper 184 (Washington: Center for Global Development).

    • Search Google Scholar
    • Export Citation

    ElliottK.2007“Trade Policy for Development: Reforming U.S. Trade Preferences,” CGD Brief (Washington: Center for Global Development).

    • Search Google Scholar
    • Export Citation

    EngelJ.2009“Assessing the Chinese and Indian LDC Preference Schemes: Initial Observations” (London: Department for International Development).

    • Search Google Scholar
    • Export Citation

    European Union2009“EC Generalised Scheme of Tariff Preferences (GSP) 2009–11,”GSP Fact SheetBrussels.

    European Union2004“Communication from the Commission to the Council, the European Parliament and the Economic and Social Committee on the Function of the Community’s Generalised System of Preferences for the Ten Year Period from 2006–15,”Brussels.

    • Search Google Scholar
    • Export Citation

    GregoryR.C.HennB.McDonald and M.Saito2010“Trade and the Crisis: Protect or Recover,”IMF Staff Position Note 10/07 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

    HoekmanB. and J.Wilson2010“Aid for Trade: An Action Agenda Looking Forward,”World Bank PREM Economic Premise 25Washington.

    HoekmanB. and C.Özden2005“Trade Preferences and Differential Treatment: A Selective Survey,”World Bank Policy Research Working PaperWashington.

    • Search Google Scholar
    • Export Citation

    IMF2010“Reaching the MDGs: Macroeconomic Prospects and Challenges in Low-Income Countries,”Background Note by IMF Staff for the United Nations MDG Summit (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

    IMF2008“Structural Reforms and Economic Performance in Advanced and Developing Countries,”available via the Internet at

    • Search Google Scholar
    • Export Citation

    JaffeeS.2006“Sanitary and Phytosanitary Regulation: Overcoming Constraints,” in Trade Doha and Development: A Window into the Issuesed. by R.Newfarmer (Washington: World Bank).

    • Search Google Scholar
    • Export Citation

    JalesM.2010“How Would a Trade Deal on Cotton Affect Exporting and Importing Countries?”Issue Paper 26 (Geneva: International Centre for Trade and Sustainable Development).

    • Search Google Scholar
    • Export Citation

    KruegerA.1993“Free Trade Agreements as Protectionist Devices: Rules of Origin,”NBER Working Paper No. 4352Cambridge, MA: National Bureau of Economic Research.

    • Search Google Scholar
    • Export Citation

    LabordeD.2008“Looking for a Meaningful Duty Free Quota Free Market Access Initiative in the Doha Development Agenda,”Issue Paper 4 (Geneva: International Centre for Trade and Sustainable Development).

    • Search Google Scholar
    • Export Citation

    LowP.R.Piermartini and J.Richtering2006“Non-Reciprocal Preference Erosion Arising from MFN Liberalization in Agriculture: What Are the Risks?”WP ERSD-2006-02 (Geneva: World Trade Organization).

    • Search Google Scholar
    • Export Citation

    LowP.R.Piermartini and J.Richtering2005“Multilateral Solutions to the Erosion of Non-Reciprocal Preferences in NAMA,”WP ERSD-2005-05 (Geneva: World Trade Organization).

    • Search Google Scholar
    • Export Citation

    MartinW. and A.Mattoo2010“The Doha Development Agenda: What’s on the Table?”Journal of International Trade and Economic DevelopmentVol. 19No. 1 pp. 81107.

    • Crossref
    • Search Google Scholar
    • Export Citation

    MattooA.D.Roy and A.Subramanian2002“The Africa Growth and Opportunity Act and Its Rules of Origin: Generosity Undermined?”IMF Working Paper 02/158 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation

    OECD2010“Agricultural Policies in OECD Countries at a Glance, 2010” (Paris: Organisation for Economic Co-Operation and Development).

    • Search Google Scholar
    • Export Citation

    OECD and WTO2009Aid for Trade at a Glance 2009: Maintaining Momentum (Paris and Geneva: Organisation for Economic Co-Operation and Development and World Trade Organization).

    • Search Google Scholar
    • Export Citation

    WTO2010a“Market Access for Products and Services of Export Interest to Least-Developed Countries” (Geneva: World Trade Organization).

    • Search Google Scholar
    • Export Citation

    WTO2010b“Progress in Implementation of the 2010–11 Aid-For-Trade Work Programme: Background Note” (Geneva: World Trade Organization).

    • Search Google Scholar
    • Export Citation

    WTO2009“Aid-for-Trade Work Programme: 2010–11” (Geneva: World Trade Organization).

    WTO2007“Canada’s Least-Developed Countries’ Market Access Initiative” (Geneva: World Trade Organization).


The United Nations identifies 49 LDCs according to criteria such as per capita GDP and structural impediments to growth. LDCs are typically the poorer and more vulnerable countries within the broader LIC group.


Anderson and Winters (2008) discuss the merits for developing countries of open markets abroad and at home. IMF (2008) finds that trade liberalization has helped to boost economic growth and stability in both developed and developing countries, and that open trade policies contributed to the success of subsequent financial sector reform.


While the focus here is on trade in goods, AMs and EMs could also look for ways to provide effective trade preferences for services. Examples could cover: (i) particular modes of supply, such as through the limited-term presence of foreign workers (“Mode 4” of the WTO General Agreement on Trade in Services), or (ii) particular services sub-sectors, such as where there are limits on transport service providers (e.g., in maritime shipping). Discussions at the WTO to formulate a waiver under which countries could extend non-reciprocal services trade preferences are a welcome development in this context.


Some of the actions we propose here reflect proposals made by others, such as the Center for Global Development Working Group on Global Trade Preference Reform (Elliott, 2010) or the LDCs themselves. Farm subsidy reforms and trade preferences for the LDCs are of course topics covered in the Doha negotiations, though each can be implemented prior to a Doha conclusion or beyond any Doha commitments.


For example, while the United States has a low average most favored nation (MFN) tariff rate—about 4 percent—duties collected on imports from Bangladesh and Cambodia are equivalent to over 15 percent of import value.


We use the term “rules of origin” to encompass provisions that determine whether origin is conferred and the good becomes eligible for trade preferences. For the discussion here, we separate these into “origin rules” and “cumulation provisions” (see Box 3).


Minimal provisions may be needed to ensure that goods from non-LDC LICS are not transshipped via LDCs.


Using 2003 data and weighting by bilateral trade with five preference providers (Australia, Canada, EU, Japan, and the United States), Low and others calculate that, for non-agricultural products, LDCs had on average a 6.4 percentage point preference against the MFN tariff. After adjusting for preferences provided to other (non-LDC) countries, the effective margin of preference was only 1.6 percent. In a subsequent paper on agriculture (Low and others, 2006), the authors calculated an effective margin of preference of only 0.1 percent.


In 2008, China overtook the European Union (EU) as the largest single importer of LDC products, accounting for 23 percent of LDC exports.


For details of the tariff treatment of LDC exports in selected emerging markets, see WTO 2010a.


The Republic of Korea avails itself of WTO developing country status.


Cali and Page (2010) suggest that the model results may overstate the potential gains because of the high level of sectoral and geographic aggregation and because RoO and costs of compliance have in the past limited the use of preferences. However, the Action Plan here would go far toward addressing the latter issue.


OECD (2010), Table 1.1. The figure excludes consumer subsidies and government support of general services.


Jales (2009) also explores this issue.


We use the term “standards” here in a broad sense to encompass (mandatory) technical regulations, (voluntary) product standards, conformity assessment procedures, as well as sanitary and phytosanitary (SPS) measures. The latter refer to measures to protect human, animal, and plant life or health. The WTO Agreement on Technical Barriers to Trade (TBT) and Agreement on the Application of Sanitary and Phytosanitary Measures establish multilateral disciplines in these areas.


OECD and WTO (2009) reviews the Aid for Trade initiative. The 2008 figures are reported in WTO (2010b).


Trade tax revenues have become less important as countries have turned to less distortionary revenue sources; we expect this to continue. In addition, trade taxes will fall as more regional trade agreements are implemented.

Other Resources Citing This Publication