Poverty reduction is an increasingly important consideration in the deliberations over multilateral trade liberalization. However, the analytical procedures used to assess the impacts of multilateral trade liberalization on poverty are rudimentary, at best. Most poverty studies have focused on a single country using detailed household survey data. When it comes to multi-country, global trade liberalization analyses, researchers are forced to resort to a discussion of average, or per capita effects, suggesting that if per capita real income rises, then poverty will fall. As we show in this paper, such an inference can be misleading. Our paper combines results from a newly available international, cross-section consumption analysis for 1996, with earnings data from household surveys from seven focus countries, to analyze the implications of multilateral trade liberalization for poverty in several developing countries in Asia, Africa and Latin America. By emphasizing the earnings-side of the poverty story, we complement earlier studies of poverty that have tended to emphasize consumption determinants, often the exclusion of earnings impacts. The analysis begins by focusing on the impact of trade liberalization on households at the edge of poverty the marginal households in our terminology. We stratify households according to their primary source of income, identifying those that are specialized (95% or more of their income) in agriculture enterprises, on-agriculture enterprises, wage/salary labor, and transfers. All other households are considered to be diversified, and therefore likely to be less vulnerable to trade shocks. Since previous multi-region analyses have focused on the per capita effects, and since the technology for explaining the per capita welfare impacts of multilateral trade liberalization is well understood, we decompose the departures of marginal household welfare from these per capita effects. These differences are explained in terms of deviations in consumption and earnings shares. We find that the differences in earnings shares are generally more important in explaining the changes in marginal households' welfare than the differences in their consumption profiles. The multilateral trade liberalization scenario that we examine involves complete elimination of merchandise tariff barriers, agricultural export subsidies and textile and apparel quotas in place in 1997. This ignores the potential impact of other non-tariff barriers as well as the significant barriers to trade and investment in services and trade distorting domestic farm policies. While this liberalization scenario is accordingly stylized, it does offer a useful benchmark for assessing the potential poverty impacts of multilateral measures. Of particular interest is our partitioning of the effects of countries own policies versus those of other countries on poverty. We measure poverty using both the headcount measure and the Foster-Greer-Thorbecke transfer measure (the total transfer required to lift all households out of poverty, as a portion of the poverty level of income). We find that the aggregate measure of poverty is reduced in Indonesia, Philippines, Thailand, Uganda, and Zambia, while it is increased in Brazil and Chile. The largest percentage reduction in poverty occurs among agriculture-specialized households in Brazil and Chile (more than 30% reductions). Poverty also falls between 7% and 9% for the agriculture-specialized households in Philippines and Thailand and for the wage labor households in Indonesia and Philippines. On the other hand, there are increases in poverty, ranging from 5% to 11% among the self employed, non-agricultural households in Indonesia and the labor-specialized households in Brazil and Chile. When we decompose the poverty changes associated with different countries' trade policies, we find that developed countries' policies play and important role in these poverty changes. Agricultural trade policies in the developed countries are particularly important. Their liberalization leads to a reduction in poverty among the agriculture-specialized households in all the focus economies. However, the impact on other poor households is mixed. So the aggregate, short run, poverty impact depends on the relative importance of the different strata in the total population of poor. In general, trade policies tend to have a diverse impact on poor households, depending on their primary income source. For this reason, we believe that stratification is very important if one is to capture this diversity of impacts. As with any ambitious research undertaking, there are many limitations associated with this study. More work is needed to deal with the problem of underreporting of income. We have made crude adjustments in the most extreme cases, based on comparisons with national control totals, but further efforts in this direction would be useful. A second limitation has to do with the short run nature of our analysis, dictated by the absence of formation in the household surveys regarding the factor composition of profit-type income. We would like to draw on additional information in the surveys in order to disaggregate profits into returns to land, capital, skilled and unskilled labor, as well as determining how the mix of profit-type income varies across the income spectrum. These limitations notwithstanding, we believe that there is considerable value in this kind of rigorous, post-simulation assessment of the poverty impacts of trade liberalization. We plan to continue to push forward in systematically addressing these limitations as time, data, and resources permit.