Madagascar's textile and apparel industry has been among the fastest growing in Sub-Saharan Africa. Fueled by low labor costs, a fairly productive labor force, and preferential access to industrial countries, Madagascar's exports of textile and apparel products grew from about US$45 million in 1990 to almost half a billion in 2001. The impact of this export surge has been large in terms of employment and wages, but less so in terms of poverty reduction. To address the concern of whether the poor benefit and to what extent, the author follows a new approach to identify the beneficiaries of globalization and to quantify the benefits at the household level, so as to understand which segments of the population benefit most and which, if any, are marginalized. The analysis focuses on the labor market channel which has been recognized as the main transmission between economic growth and poverty. The methodology uses household level data and combines the wage premium literature with matching methods. The results point to a strong variation in the distribution of the benefits from export growth with skilled workers and urban areas benefiting most. From a poverty perspective, export-led growth in the textile and apparel sector has only a small effect on overall poverty. This study points to two reasons for this. First, a large majority of the poor are unable to enjoy the new employment opportunities, given their lack of skills sought by the expanding textile and apparel export industry. Second, most of the poor reside in rural areas where the employment effect is small. The results indicate that the effects of an increase in exports of textiles for poverty reduction are felt only in urban areas, mostly through job creation. Some of the urban poor are good candidates for finding employment in the expanding sector. But the urban poor are likely to find employment only in unskilled jobs. Given that unskilled wages are kept low by a large reserve labor sector, the gains are limited, and the overall impact on poverty is small. More generally, the results of this study suggest that two factors are required if export-led economic growth is to significantly reduce poverty. First, growth and job creation must not be restricted to a few geographic areas but need to reach areas where the majority of the poor live. Second, poor people must be assisted in obtaining the skills sought by expanding industries.