TRADE ALONE IS NOT ENOUGH, SAYS UNCTAD; POVERTY REDUCTION REQUIRES DEVELOPMENT-LED TRADE

UNCTAD´s Least Developed Countries Report 2004, released today , provides the first comprehensive empirical analysis of the relationship between trade and poverty in the least developed countries (LDCs). The Report shows that export growth was rarely associated with sustained poverty reduction in the LDCs during the 1990s, even though trade contributes a high proportion of their GDP and most of them have undertaken far-reaching trade liberalization.

Persistent mass poverty in the LDCs is not due, according to the Report, to lack of integration into the global economy, nor to insufficient trade liberalization, but is rather the consequence of underdevelopment. In fact, the evidence indicates that the extensive trade liberalization has thus far not resulted in a form of integration that will support sustained and inclusive development. Moreover, trade liberalization has actually worsened the trade balance in LDCs.

The idea that the key problem for LDCs is lack of global integration does not conform to reality, the Report shows:

• The average level of trade integration of the LDCs, measured by the ratio of exports and imports of goods and services to GDP (which is often used as a measure of "openness"), was actually higher than that of high-income OECD countries in 1999-2001 (51%, as against 43%).
• Most of the LDCs now have more open trade regimes than other developing countries, and the trade regimes of the LDCs as a group are as open as those of high-income OECD countries.
• 76% of LDC exports also entered developed-country markets duty-free in 1999-2001, and this proportion will have increased following recent preferential market access initiatives in the Quad markets (Canada, the EU, Japan and the US).

But in the LDC context, trade liberalization plus preferential market access do not equal poverty reduction. The Report demonstrates that during the 1990s, the greatest improvement in both export growth and growth of average private consumption per capita was in those countries that opened moderately during the decade, rather than in those that opened the most.

The LDCs are caught in a complex position: they need the multilateral system more than other countries, but at the same time they find it most difficult to derive benefits from the application of its two central principles - trade liberalization and equal treatment for all members.

Making trade work for poverty reduction in the LDCs requires national development with global integration, not global integration without national development. But in the LDCs that established open trade regimes there was a pattern of post-liberalization aid fatigue in the 1990s. Countries received higher aid inflows as they undertook trade liberalization, but these inflows fell once an open trade regime was established. The upturn in aid to LDCs since 2000 (see UNCTAD/PRESS/2004/009) is likely to have reversed that pattern.

International trade cannot be a substitute for development aid in the LDCs, the Report says. Aid, debt relief and trade must all work hand-in-hand for poverty reduction. A development-led approach to trade will be more effective than a trade-led approach to development.

Trade should work …

The Report argues that the links between international trade and poverty vary between countries according to their production structure and level of development. In the 50 LDCs most of the population lives on incomes that are at or below a bare $1-a-day subsistence minimum, making extreme poverty a mass phenomenon.

In 1999-2001, the average GDP per capita of the LDCs was $0.78 per day. Adjusting for international purchasing power differences, in 2000 about 50% of the 668-million LDC population was living on less than $1 a day and about 80% on less than $2 a day.

Substantial poverty reduction will not be achieved in these circumstances through local anti-poverty projects, the Report warns. Rather, there is a need for sustained and inclusive economic development that substantially increases average household incomes and consumption.

International trade should play a central role in this process. The revenue from exports enables the acquisition, through imports, of the goods and services needed to develop productive capacities. Through exports, underutilized natural resources and surplus labour can be transformed into import capacity and increased investment.

…but in the LDCs, often it does not

However, by examining trends in average private consumption per capita and exports, the Report also demonstrates that export expansion has not generally been associated with poverty reduction. In low-income Asia and Africa over the long term, increases in average private consumption per capita are closely linked to reductions in the share of the population living on less than $1 or $2 a day. Trends in average private consumption per capita can thus be used as a proxy for poverty trends in the LDCs. This enables the identification in different countries and at different periods of three types of trade-poverty relationships (see table):

• A virtuous trade effect, where average private consumption per capita is rising along with export growth;
• An immiserizing trade effect, where average private consumption per capita is falling along with export growth; and
• An ambiguous trade effect, where there is no clear association between changes in average private consumption per capita and increased export growth.

Using available data for the LDCs during the periods 1990-1995 and 1995-2000, the Report finds that exports are growing in 51 of 66 cases. But among these 51 cases, the immiserizing trade effect is present in 18, and the virtuous trade effect in 22. Virtuous trade effects occur less frequently than ambiguous and immiserizing trade effects combined (29 cases). There are only three LDCs - Bangladesh, Guinea and Uganda - in which the virtuous trade effect is observed during both 1990-1995 and 1995-2000.

There have been significant export take-offs in a large number of LDCs since the late 1980s. But average private consumption per capita has not taken off in the same way. This is not a country-specific phenomenon. Instead, there are multiple cases, regardless of their export specialization, where a surge in export growth has been accompanied by stagnation or a decrease in average private consumption per capita (see chart).

Why trade expansion is rarely associated with poverty reduction

The Report identifies seven major reasons why trade expansion is so rarely associated with sustained poverty reduction in the LDCs. These are underpinned by both national and international factors:

• Slow export growth. Below a threshold annual export growth rate of 5%, there is greater probability that export growth will be associated with declining GDP per capita than with increasing GDP per capita in the LDCs. Many of the non-oil commodity-exporting LDCs (27 of the 50 countries in this group) have found it difficult to achieve and sustain export growth rates above this threshold. This is due to (i) falling commodity prices, (ii) increasing competition in saturated international commodity markets and inability to keep up with rising product and logistical standards, and (iii) specialization in commodities for which there is sluggish or declining world demand.
• Decreased aid inflows. Almost all the LDCs have large trade deficits which have been financed by aid inflows. In these circumstances the positive import supply effects of export growth can be offset by decreased aid inflows (or increased debt service obligations). Import growth rates were slower than export growth rates in three quarters of the LDCs for which data are available in 1990-1995 and two thirds of the countries in 1995-2000.
• Low level of investment and low investment efficiency. Sustained economic growth requires not simply export expansion but also a strong investment-export nexus in which imported equipment, raw materials and production inputs are put to good use and lead to the continuous expansion of employment and improvement of labour productivity in the economy as a whole. But investment is low and inefficient in the LDCs, the latter because of (i) the weakness of the domestic entrepreneurial class and (ii) the central importance of aid for capital formation. During the period 1990-2001, aid accounted for 50% of annual capital formation in the LDCs as a group, as compared with 5% in other developing countries. Unless aid is effective, the trade-poverty relationship will break down.
• High rates of population growth and the consequent need for high rates of employment generation. Employment generation is the key to poverty reduction, and the higher the population growth rate, the greater the effort required to ensure that the working-age population is fully and productively employed. The median age of the LDC population was 18.1 years in 2000, and the working-age population will increase by 30% between 2000 and 2010.
• The trade-off between domestic resource mobilization and poverty reduction. It is very difficult for "belt-tightening" to occur in very poor countries with mass poverty without a rising incidence of poverty. In two thirds of the cases of immiserizing or ambiguous trade effects in LDCs in the 1990s, there was an increasing domestic resource mobilization effort (indicated by a falling share of private consumption in GDP).
• Weak linkages between export sectors and the rest of the economy. Given the structure of production, employment and enterprise, export-led growth is often enclave-led growth in an LDC context. This is a form of economic growth that is concentrated in a small part of the economy, either geographically or sectorally. It is exemplified by manufacturing activities in export processing zones, tourist developments that rely on imports and also mining developments. Some 71% of the working-age population was employed in agriculture in 2000, and export growth is unlikely to be poverty-reducing unless it strengthens the development complementarities between agriculture and emerging non-agricultural activities.
• Civil conflict. This is a major cause of poverty in the LDCs, but the Report - which has a separate chapter on the way civil conflict affects trade and poverty - shows that there has been a tendency for exports to be positive during conflict episodes, and sometimes even higher than in the period preceding conflict. This is particularly evident in cases of conflict recurrence, as some economic actors increasingly just get on with their business regardless of, and even adjusting to, an environment of repeated conflict. Whatever the cause, conflict episodes are usually characterized by immiserizing trade.

Trade liberalization and poverty

Given the relationship between trends in exports and average private consumption per capita, it is hardly surprising that trade liberalization is not an unambiguously pro-poor trade policy in the LDC context.

The usual view of the relationship between trade liberalization and poverty is that trade liberalization is likely to have adverse effects in the short run - particularly as social groups that formerly benefited from a protectionist tariff regime are exposed to international competition - but that in the long run, the effects will be favourable because trade liberalization will increase the growth potential of the economy.

The Report finds just the opposite.

Poverty trends during and immediately after trade liberalization in the LDCs are in fact very mixed, and not invariably negative as some claim. But there are many grounds for concern about the long-term effects in terms of both the sustainability of economic growth and its inclusiveness.

During the 1990s, the best trade-poverty relationships are found in countries that opened moderately during the decade. The Report also finds that GDP growth rates, export growth rates and investment growth rates are all higher in LDCs in the post-liberalization economic environment than in the pre-liberalization era. But imports have grown faster than exports after liberalization. Although new export goods and services are certainly emerging, they have had a limited impact on the overall composition of exports. This is not yet shifting towards greater specialization in dynamic products and increased competitiveness. In addition, there has been a repeated tendency for aid inflows to peak during trade liberalization and then fall.

Trade liberalization has been undertaken in underdeveloped countries with weak internal transport and communications infrastructure, weak levels of domestic market integration and high levels of subsistence production. This is exacerbating the problem of enclave-led growth in countries that export manufactures, minerals and oil, and is also leading to a repeated pattern in low-population-density agricultural-exporting LDCs where poor regions are being left out. In agriculture-exporting LDCs with a high population density, a different structural problem is emerging. The development of non-agricultural employment is necessary to relieve pressure on land. But this has proved difficult to achieve. Rapid and deep trade liberalization has been associated with de-industrialization, as import substitution industries have collapsed when they are exposed to international competition without adequate preparation.

The opportunity of market access preferences and multilateral trade liberalization

Past experience shows that countries must have at least a minimum base of production and supply capabilities to take advantage of market access preferences. Those LDCs that have been able to take advantage of the preferences provided under the Multifibre Arrangement (from 1974 to 1995) and then the Agreement on Textiles and Clothing by developing simple manufactures exports have achieved high and steady export growth. But the commercial benefits of special preferential market access provided by the Quad countries have been reduced by limits on product coverage, rules of origin, problems of stability and predictability, and the adverse effects of non-tariff barriers. With regard to the latter, it is notable that no less than 42% of LDC exports faced environment-related trade barriers (such as sanitary and phytosanitary standards and technical barriers to trade) in 1999-2001.

Similarly, the LDCs cannot be expected to gain much from further multilateral trade liberalization unless improvements are made in their productive capacities to enable them to benefit from any subsequent growth of global trade. For example, if multilateral trade liberalization had been undertaken in 2000 and all the gains had been instantaneous and equivalent to the 5% increase in developing countries´ income which most simulation models suggest, a country like Ethiopia would have had a real per capita income in 2001 of $127 with multilateral trade liberalization, rather than $121 without it.

Policy implications

Policies to make international trade work for poverty reduction in the LDCs assume that (i) the best way to reduce poverty is not through promoting economic development, but rather through promoting international trade and (ii) international trade is best promoted through trade liberalization and preferential market access. But the evidence suggests that within the LDC context, this trade-led approach to development will not generate substantial and sustained poverty reduction.

The findings have important implications for both the LDCs and their development partners. Their development partners should not imagine that preferential market access or multilateral trade liberalization will substitute for international aid as a central mechanism for supporting poverty reduction. These can be complementary processes. But the LDCs still require increased and better aid not just to meet urgent basic needs but also to build productive capacities. International measures to enhance the positive impact of trade on poverty reduction also need to go beyond multilateral trade liberalization. The LDCs themselves can maximize the poverty-reducing effects of international trade by pursuing a development-led approach to trade rather than a trade-led approach to development, the Report concludes (see UNCTAD/PRESS/2004/011).

Chart. Indices of real exports and real private consumption per capitain selected LDCs, 1980-2001 (Base year 1990 = 100)There is a repeated pattern in which sharp rises in exports are associated with little change in private consumption per capita or even a decline.

Source: UNCTAD, The Least Developed Countries Report 2004.

Table. Classification of LDCs on the basis of their real growth rates ofprivate consumption per capita and of exports of goods and services,1990-1995 and 1995-2000

Based on the close relationship between the level of private consumption per capita and the incidence of poverty in conditions of generalized poverty which prevail in the LDCs, trends in private consumption per capita (in 1985 purchasing power parity $) are used as proxy for poverty trends. The threshold of a 1% growth rate in private consumption per capita used here is highly conservative: according to this threshold, it would take 70 years for countries to double the level of their private consumption per capita.The table shows that in the 1990s, the occurrence of export expansion without poverty reduction (situations of immiserizing trade effect and of ambiguous trade effect) was more frequent than that of export expansion with poverty reduction in the LDCs. One third of the cases in the 1990s were situations of immiserizing trade. Positive export growth rates are a necessary condition for poverty reduction. But export expansion is no guarantee of poverty reduction.

Source: UNCTAD, The Least Developed Countries Report 2004.