Trade, growth, and poverty reduction ¹

Economists have long observed that countries and regions that are linked by common institutions, currencies, or policies, and that enjoy relatively free access to each other’s markets, tend to converge to similar levels of income over time. Between 1960 and 1982, for example, the incomes of poorer regions or countries converged to those of richer ones at a rate of about 2 percent a year in the United States and different parts of Europe and among members of the Organization for Economic Cooperation and Development (OECD). Indeed, poorer countries and regions have, in general, grown faster than their richer neighbors with which they have close ties.

It is plausible that trade openness—the extent to which nationals and foreigners can trade with each other without artificial constraints (such as tariffs and quotas)—has played a role in the convergence process by facilitating specialization and promoting competition and the transfer of knowledge.

But it has also been observed that countries that trade widely tend to have good institutions and macroeconomic policies, triggering a chicken-and-egg debate about which comes first. Do the good institutions and policies that are associated with trade play a more important role than the latter in stimulating growth? Does trade stimulate growth or vice versa? Last, but not least important, does growth that is driven by trade raise the poor’s living standards or does it increase income inequality, making the poor poorer and the rich richer?

Trade is an important determinant of growth. A number of cross-country and panel regressions that have attempted to disentangle the effects of different factors on growth rates and to establish the direction of causality have found evidence that trade openness is strongly linked to faster economic growth. This holds true whether openness is measured in terms of a country’s trade policies (tariff and nontariff barriers) or as an outcome (the ratio of exports plus imports to GDP). The relationship is even stronger when purchasing-power-parity GDP is substituted for absolute GDP, thereby eliminating the effect of cross-country differences in the prices of nontraded goods.

Neither measure of openness is wholly satisfactory, however. Serious problems arise in the analysis of individual policies that restrict trade. Low average tariffs can mask restrictions targeted at key products. Nontariff barriers are particularly hard to measure. These can take many different forms—for example, discriminatory exchange rate policies that offer exporters a more appreciated exchange rate than importers, the uncertainty that stems from delays in customs clearance, and contingent protection (the threat that a country will restrict certain imports in the event of major import penetration, for instance under antidumping or safeguard provisions). Phytosanitary, sanitary, and technical standards may also serve protectionist purposes. The impact of such measures is extremely difficult to quantify.

Unfortunately, the second measure of openness—exports plus imports as a share of GDP—is equally imperfect. It reflects not only openness to trade but also the level of a country’s economic development, geographic factors such as distance from trading partners, and resource endowment.

Nonetheless, case studies support the argument that trade liberalization raises growth rates. Although opening up to trade does not guarantee faster growth, all of the countries that have taken off economically in the past 20 years have included trade opening in their reform packages. Two seminal studies in 1978 analyzed the phases through which liberalizing countries moved as they shifted from import substitution to outward-oriented trade policies (that is, policies without an anti-export bias).² The studies described how the distortions caused by various protectionist measures worked their way through the economy in mostly unplanned and undesirable ways and showed how exports and growth responded to substantial trade liberalization and appropriate macroeconomic policies. A large World Bank study that analyzed the design, implementation, and outcome of 36 trade liberalization episodes in 19 countries between 1946 and 1986 found that strong and sustained liberalization episodes resulted in rapid growth of exports and real GDP. ³

A study carried out in 1999 ⁴ found that, among relatively closed economies, the poorest in 1960 also grew the slowest between 1960 and 1985, but that low initial income was not correlated with slower subsequent growth in open economies. In closed economies, low initial income reduces potential benefits from scale economies, but trade openness, by allowing access to broader markets, overcomes this problem.

More recent microeconomic studies have documented several channels through which openness leads to higher productivity, including the import of machinery and equipment, which is usually accompanied by the transfer of know-how. Other studies have shown that import competition lowers margins and increases turnover and innovation.

Trade is a complement of other reforms. Much of the evidence about the effect of openness on growth is vulnerable to the criticism that the effect of openness has not been isolated from the effects of a good institutional environment or other reforms that were often implemented at the same time. In case studies and before-after comparisons, for example, the effects of trade liberalization are hard to disentangle from the effects of macroeconomic stabilization, internal price liberalization, changes in the foreign exchange system and the exchange rate, liberalization of the capital account, reform of social safety nets, and a host of other measures.

In interpreting the role of trade reform as distinct from other aspects of policy, it is important to distinguish between preconditions, desirable complements, and beneficial reform spillovers. There are few preconditions—reforms in the absence of which trade openness is a poor idea—and there are a variety of reasons why trade openness might promote other reforms. Openness provides powerful channels for feedback on the effect of various policies on productivity and growth. For example, competition with foreign firms can expose inefficient industrial policies. It increases the marginal product of complementary reforms, in that better infrastructure, telephones, roads, and ports enable the export sector to perform better, and it raises the productivity of companies making products for the domestic market as well. In addition, trade liberalization may change the political reform dynamic by creating constituencies for further reform.

Growth reduces poverty. From 1978 to 1998, the proportion of extremely poor people in the world—those living on less than two 1985 dollars a day—fell sharply, from 38 percent to 19 percent. Because of population growth, the drop in the absolute number of poor was smaller but no less dramatic—from 1.4 billion to 1 billion. This decline is believed to be attributable almost entirely to growth, rather than to changes in the distribution of income.

The histories of China and India provide support for this finding. Between 1980 and 1992, per capita income in China grew 3.6 percent a year, while income disparities increased significantly (China’s Gini coefficient increased from 0.32 to 0.38). Nonetheless, the number of people living on less than two dollars a day decreased by some 250 million, as rapid income growth swamped the effect of the increase in inequality. Income inequality would have had to increase more than twice as fast as it did to undo the effects of strong growth. Similarly, the incidence of poverty in India fell dramatically, from 35 percent of the population in 1987/88 to 23 percent in 1999/2000, and it would have fallen to 21 percent in the 1990s had growth been entirely proportional to income; yet, as in China, the decline in absolute poverty was accompanied by an increase in income inequality.

On the question of whether growth fueled by trade benefits the poor more or less than it does other social groups, no clear pattern emerges from the numerous studies of individual liberalization episodes. This is not surprising: liberalization can change relative prices and incentives throughout an economy. A few generalizations can nonetheless be made. Trade liberalization tends to reduce monopoly rents and the value of connections to bureaucratic and political power. In developing countries, it may increase the relative wage of low-skilled workers. Poor consumers benefit from the lower prices that often follow trade liberalization. Liberalization of agricultural trade typically has the strongest effects on the poor—both positive and negative—since most poor people in developing countries are engaged in small-scale agricultural activities. But again, the overall benefits of growth for poverty reduction dominate the distributional impact. It is useful to recall that the same holds true for growth triggered by technical innovation—and that there has been little serious argument in favor of halting technical progress on the grounds that its benefits may initially be distributed unequally.

Concerns of developing countries

Despite the evidence that trade liberalization is likely to benefit even the poorest and least developed countries, many of these have concerns that need to be addressed before they can be persuaded to become full participants in multilateral trade negotiations. In addition to their concerns about the dislocation of farmers and workers, they are worried about the impact trade liberalization will have on their balance of payments, fiscal accounts, and terms of trade. Chief among these are a possible decline in export earnings if the erosion of preferences puts them in competition with exports from lower-cost producers; loss of government revenues if they eliminate tariffs on imports; and higher world prices for food imports if agricultural subsidies are eliminated.

Erosion of preferences. ⁵ A wide range of developing countries currently enjoy trade preferences in the form of very low or zero tariffs on their exports to richer countries. The Generalized System of Preferences provides preferential entry for a wide range of products from 144 countries and territories into OECD markets. In addition, members of the African, Caribbean, and Pacific group of countries receive greater preferential access to EU markets, and exports from the least developed countries (except for sugar, bananas, and rice) are receiving virtually duty- and quota-free access to the EU markets under the Everything-But-Arms Initiative and to several other OECD markets under similar schemes. African countries have preferential access to U.S. markets under the Africa Growth and Opportunity Act. As the OECD countries reduce tariffs on imports from all of their trading partners under the WTO’s most-favored-nation terms, the value of these trade preferences will erode, although this may be partly offset by expanding market size and higher world prices. Fearing the loss of their edge, some of the developing countries whose exports may suffer when they face more competition in key markets are resisting WTO efforts to lower tariffs and quotas generally—a conflict that pits the interests of different groups of developing countries against each other.

The impact of preference erosion outside the OECD is expected to be smaller and more gradual. Although the advantage of regional trade agreements that provide for lower (or zero) trade barriers between members would be eroded as regions reduced their common external tariffs in the context of multilateral trade liberalization, intraregional trade accounts for only a small share of most developing countries’ exports (with the exception of countries in Southeast Asia). Moreover, developing countries receive “special and differential treatment” in the WTO and are likely to negotiate a more gradual reduction in their trade barriers under any multilateral agreement.

Even under fairly conservative assumptions, the magnitude of the potential shock, in a realistic liberalization scenario, is small overall: between 0.5 percent and 1.2 percent of the total exports of 76 middle-income countries studied by Alexandraki and Lankes, depending on the elasticities of export supply. The effect is also likely to be spread over time in accordance with liberalization schedules established in the context of the Doha Round or other trade reforms. This should make it easier for countries to plan for adjustment.

Nonetheless, shocks could be significant for a small group of countries—those with undiversified export sectors, heavy dependence on EU and U.S. markets, and fragile macroeconomic frameworks (for example, some small island economies). However, the potential export loss must be considered in the context of a country’s broader macroeconomic framework, including other sources of current account receipts, such as tourism.

The middle-income developing countries that are the most exposed to preference erosion are exporters of sugar, bananas, and—to a lesser extent—textiles and clothing. Sugar and bananas account for three-fourths of the current preference margins (the difference between the most-favored-nation tariff and the preferential tariff for a product) of countries that have a total margin greater than 5 percent of the value of their exports.

The obvious implication is that the policy discussion—and any support for countries facing adjustment costs—should target these three products and the countries dependent on them. Another implication is that policy reforms in the sugar and banana regimes of the OECD countries, which take place mostly outside multilateral trade negotiations, matter more for preference erosion than does the Doha Round.

Higher prices for food imports. ⁶ World markets for agricultural commodities are highly distorted by the tariffs and subsidies—including export subsidies—used by many industrial countries to protect their agricultural producers. These policies are most costly to the OECD countries themselves—nearly all empirical studies demonstrate that the bulk of the cost of agricultural support falls on the country that adopts such policies. But by artificially lowering the cost of production, agricultural support can also make food produced in low-income countries uncompetitive with imports. Both the IMF and the World Bank have called on OECD countries to eliminate the trade-distorting support they provide to their agricultural sectors. At the World Summit on Sustainable Development in 2002, world leaders called for a reduction (and eventual removal) of agricultural support in rich countries, especially on products exported by developing countries. Indeed, a broad-based consensus has emerged that agricultural support policies in OECD countries are detrimental to the interests of developing countries.

There is no question that removal of agricultural subsidies could raise real income in the world and in the developing countries as a group. But it is sometimes overlooked that the subsidies actually benefit some poor developing countries by depressing world prices for certain agricultural products. As a consequence, consumers in developing countries—and the countries as a whole if they are net importers of these commodities—actually benefit from the subsidies and may suffer if they are removed. Countries that are net importers of grain and dairy products (the Middle East and Africa) are likely to be particularly hard hit by liberalization, but the increase for most of the 79 countries studied by Tokarick was small—less than 2 percent of affected imports.

What has been good for consumers has hurt domestic producers, however, who cannot compete with low-priced imports. With poverty concentrated in rural areas, low prices for agricultural products have hurt the poor disproportionately. They have also hurt developing countries that export products whose prices are artificially reduced by production subsidies in industrial countries. These developing countries may experience an increase in net export earnings if subsidies are removed.

Lost tax revenues. ⁷ Trade tax revenue has become less important over the past 20 years as countries have reduced tariffs, but it continues to be a major source of government finance in many low and middle-income countries, commonly accounting for one-fifth of total tax revenue, and often more. To the extent that trade liberalization cuts tariff revenues, these countries may have to develop other sources of government finance.

IMF research on the revenue effects of trade liberalization in 125 countries from 1975 to 2000 has shed some light on the nature and magnitude of the issue. The trend toward trade liberalization over the past two decades contributed to a loss of trade tax revenue for low-income countries on the order of 2½ percent of GDP, or about one-sixth of their total tax revenue, while the decline in high- and middle-income countries was less marked but still significant. Although domestic taxes generally replaced lost trade tax revenue in high- and middle-income countries, domestic tax yields as a share of GDP did not increase, on average, in low-income countries during the period in question.

These results certainly do not imply that countries were mistaken to introduce trade reforms that cost them tax revenue, but they suggest strongly that too little coordination has occurred between trade liberalization and the strengthening of domestic tax systems, and that the revenue problem deserves more serious attention than it has received.

Nearly half of the low-income countries that have cut their tariff rates and consequently lost tariff revenue over the past twenty years have recovered less than 70 percent of this lost revenue from other sources. Yet some low-income countries—including Malawi, Senegal, and Uganda—have succeeded in recovering their lost trade tax revenues. In all of these cases, a significant part of the revenue recovery came from strengthening domestic consumption taxes—excises and, typically, a VAT (value-added tax). The performance of income tax revenue was also stronger in these three countries than in those low-income countries that failed to recoup their losses. In many cases, revenue recovery was part of a broad reform effort. For example, Senegal’s IMF-supported program in 1999–2000 included a major simplification of the country’s tariff structure, together with the unification of the multiple VAT rates into one single rate.

These success stories demonstrate that the difficulties are not so much technical as political: policymakers need to have a strong commitment to reforming domestic tax systems. Their experience offers useful lessons. Liberalization itself can limit revenue loss and even increase net revenue to the extent that it spurs growth and imports—especially if nontariff barriers, whose elimination raises revenues, are cut. But with deeper tariff reform, revenue recovery also requires a committed and continuous effort, over several years, to broaden tax bases, purge exemptions, simplify rate structures, and improve revenue administration. Strengthening the domestic consumption tax system through excises and, especially, through a simple, broad-based VAT, can have a crucial role to play in this regard, and improving income tax collection can also make an important contribution to revenue recovery.

Smoothing the path to liberalization ⁸

Although developing countries stand to gain over the longer term from their own trade liberalization and that of other countries, some may initially experience adjustment costs. They may also have difficulty meeting certain obligations, such as the establishment of systems for monitoring and ensuring the protection of patents and other intellectual property rights; restrictive sanitary standards such as the fumigation of exports of fruits and vegetables; certification that products are in compliance with international standardization codes; and the modernization of customs procedures. And opening their markets to cross-border financial services may stress their domestic regulatory and supervisory capacities.

Their wealthier trading partners can help the developing countries meet some of the most onerous obligations or can lighten them, while the international financial institutions and other donors can provide “aid for trade”—technical assistance with capacity building and institutional reform and financial support to mitigate adjustment costs.

Technical assistance. The August 2004 Doha Round “framework agreement” called on the IMF and other international agencies to provide technical assistance for trade facilitation. The IMF provides assistance on customs administration modernization and tariff reform. The IMF’s technical assistance in customs administration is strategic in nature, aimed at providing the overall framework for reform and continuing oversight, with other donors providing support on specific aspects. Technical assistance in trade policy usually has a wider tax focus than tariffs alone, to help countries strengthen domestic tax collection to compensate for the loss of trade revenues stemming from tariff reduction. Building on work already done in recent years, the IMF is examining the fiscal impact of tariff reductions in poor countries under possible scenarios for the Doha Round, with the aim of assisting countries early and in a proactive manner.

With other international partners, the IMF has been engaged in a joint effort to promote the reform of trade regimes as part of poverty reduction strategies and trade-related technical assistance and capacity building. At the core of this agenda has been the IMF’s involvement in the Integrated Framework (IF), a cooperative effort of six agencies (the IMF, the International Trade Center, the United Nations Conference on Trade and Development, the United Nations Development Program, the World Bank, and the World Trade Organization, which is the chair), with the participation of bilateral donors and developing countries. The IF coordinates the preparation of Diagnostic Trade Integration Studies (see box on page 14) in developing countries, often prepared under the leadership of the World Bank with contributions from the IMF and other agencies.

The studies identify policy and assistance priorities (“action matrix”), which are reviewed in national workshops that include government, the private sector, and civil society, with the objective of integrating them into national development and poverty reduction strategies. The action matrices are presented to donors for funding, as appropriate, but the IF also has a small funding capacity of its own for capacity-building projects that require rapid follow-up.

Trade Integration Mechanism. In 2004, the IMF introduced its Trade Integration Mechanism (TIM) to help countries experiencing temporary shortfalls in their export earnings because of trade liberalization by other countries—for example, the erosion of trade preferences or the expiration of quotas in 2005 under the WTO’s Agreement on Textiles and Clothing. The IMF’s members can also receive financial support under the TIM if they suffer balance of payments problems caused by higher bills for food imports as a result of a reduction in agricultural subsidies in industrial countries. While such balance of payments problems are likely to be small for most countries and would eventually be offset by the positive effects of more open trade, they could be significant for some countries in the short run. The first two countries to receive financial assistance under the TIM were Bangladesh and the Dominican Republic in 2004 and 2005, respectively. Trade-related financial support can also be provided through other existing IMF lending facilities.

Trade Facilitation Initiative. Under this initiative, the World Bank has increased both its analytical work and its lending to help countries take advantage of trade integration. New lending will be made available for investment in ports, roads, and other necessary infrastructure, as well as for reform of customs procedures and other trade-related institutions. The World Bank supports countries’ efforts to increase their international competitiveness, such as retraining workers and making payments to help them maintain their incomes while they shift into export-oriented activities. The World Bank is also building a Global Trade Facilitation Partnership for Transportation and Trade.

Expanding aid for trade. In 2005, the Development Committee, an advisory body that represents the member countries of the IMF and the World Bank, endorsed proposals to increase assistance to poor countries in overcoming infrastructural and other supply constraints that could prevent them from benefiting from open trade and in mitigating and managing adjustment costs.

One proposal is to enhance the Integrated Framework, including by providing predictable, multiyear financing of about $200 million–$400 million over an initial five-year period. Another is to examine the adequacy of existing mechanisms for addressing regional and cross-country aid for trade. A third is for the IMF and the World Bank to make a firm commitment to assist countries with adjustment needs by providing analysis, advice, and, when necessary, financial support. The proposal to enhance the Integrated Framework is being examined by a task force of donor countries and least developed countries that will report in April 2006 on suggested organizational and governance reforms. The December 2005 Hong Kong Ministerial Meeting of the WTO established a work program aimed at determining ways in which aid for trade can be “operationalized” and at identifying financing mechanisms that would ensure increased levels of funding.

Why multilateral trade liberalization is important

Trade is a driving force for economic growth in developed and developing countries alike. The Doha Round of trade negotiations, if successful, can therefore be a powerful tool for development in the poorest countries, affording them the chance to raise incomes and living standards and achieve the UN Millennium Development Goals.

Rich countries must play their part in promoting free trade by eliminating trade restrictions and providing financial and technical assistance to poor countries that may have difficulty becoming fully integrated into the world trade system, whether because of short-term adjustment costs, supply side constraints, or other factors. Middle-income countries must reduce tariffs that affect not only their own citizens but other developing countries. And the poorest countries, even as they receive more aid for trade and win time to implement some WTO rules, must reform their own trade regimes and improve governance and institutions to reap the benefits of multilateral trade liberalization. With the assistance of donors and the international financial institutions, they will need to incorporate trade policy reforms in national strategies for development and poverty reduction.

Responsibility for the success of the Doha Round of trade negotiations does not rest with the rich countries alone. Indeed, the developing countries, given their voice in the WTO, will play the leading role in shaping their own future. Driving the Doha Round to a satisfactory conclusion will require determined political leadership from all countries.