Poverty reduction is an increasingly important consideration in the deliberations over multilateral trade liberalisation. Within this debate, the agricultural sector has assumed a prominent role, in part due to the potential for significantly higher agricultural prices which may translate into large real income losses for low income consumers. Competing with this consumption side effect is the income side. Since most of the world's poor live in rural areas, where agriculture is the dominant economic enterprise, higher agricultural prices following global trade liberalization serve to boost the incomes of many poor households. Which of these forces dominates depends on the particular country/household/trade policy combination in question. This paper aims to sort out the relative importance of these two competing effects for seven developing countries. In our analysis we use the combination of a global trade model and individual household models for 100 representative household groups in each of the focus countries to assess the short run poverty impacts of both national and international trade policy changes. Our analysis is short run in nature because we assume a limited degree of factor mobility in response to the trade policy shocks. In particular, self-employed workers are not able to find wage labour if prospects in their industry (e.g., fanning) turn sour. We also assume that capital cannot be reallocated to other activities over the time frame in question. Our findings suggest that trade liberalization reduces national poverty in six of the seven developing countries studied. However, these aggregate results mask substantial variation in poverty changes by individual household groups and by type of policy. When one focuses specifically on OECD trade policies, it becomes clear that agricultural policies are most important. This is due to three things. Firstly, this sector has relatively larger trade distortions than non-agriculture in OECD countries. This means that liberalisation of agriculture generates larger world price effects. Secondly, the budget share of the poor devoted to food products is very high, so they are vulnerable to food price shocks. Finally, in many of the focus economies, a number of the poor are highly dependent on agriculture for their earnings.