Trade Policy and Pro-Poor Growth

This paper analyzes empirically the impact of trade policy and sector specific openness on pro-poor growth in a cross-country approach to answer the question, whether the poorest 20 and 20 to 40 percent benefit from trade openness. To capture this issue, we estimate the distribution effect of eight different openness indicators, six adjusted trade sector indicators (agricultural raw materials exports and imports, food exports and imports, manufactures exports and imports) and two tariff indicators (export duties and imports duties). In addition, we estimate the total effect, i.e. the distribution and growth effect, to analyze potential trade-offs between the impact of trade liberalization on poverty via overall economic growth and distribution. To test the poverty effects, we collect an irregular and unbalanced panel of time-series cross-country data on the first and second quintile share in 72 countries for the period 1971 to 1999 and apply two econometric specifications, a growth equation and a system GMM equation. We estimate the poverty effects of trade policy for all countries and, separately, for developing/transitional and industrial countries due to considerable differences in economic structure. Finally, we estimate poverty effects of trade liberalization with respect to the level of the countries’ development. Combining empirical findings of the system GMM estimation for both the distribution and total effect, estimation results suggest the importance of sector specific trade policy for the poorest 20 and 20 to 40 percent. First, liberalization in agricultural raw material exports is very important for the poorest 40 percent of low income developing countries due to both the distribution and total effect. In addition, liberalizing imports in agricultural raw materals is highly positively related to the mean income of the poor without changing the distribution. Second, trade reforms in food exports affect negatively the mean income of the poorest 40 percent in low income developing countries through the growth effect. However, higher food imports are associated with positive distribution effects, but without total effects on the poorest 20 percent in low income developing countries. Third, promotion of manufactures exports lead to a positive total effect on the poorest 40 percent in developing countries via the growth effect, while trade reforms in manufactures imports are never relevant. Finally, reduced export and import duties affect positively the mean income of the poorest 40 percent in low income developing countries, an effect primarily driven by the growth effect. Findings for agriculture exports, food exports, export and import duties, however, are only relevant if we exploit information on both the cross-country and within-country variation of the income of the poor in a system GMM estimator. In addition, results of the growth equation suggest positive total effects of agriculture imports on the poorest 20 and 20 to 40 percent in development countries driven by the growth effect alone. Thus, empirical findings suggest the following policy recommendations with respect to poverty-reducing trade reforms in low-income developing countries. While results are not always consistent between the growth equation and the system GMM estimation, liberalization of agricultural raw material exports and imports seems to be the most promising approach. On the other hand, liberalization in food markets and manufactures imports are not associated with poverty alleviation in low- income developing countries. Finally, a promotion of manufactures exports and a reduction of export and import duties seem to increase mean income of the poorest 40 percent in low-income developing countries only via the growth effect.