There is a general consensus that most of the poor in developing countries are net food buyers and food price increases are bad for the poor. This could be expected of urban poor, but it is also often attributed to the rural poor. Recent food price increases have increased the importance of this issue, and the possible policy responses to these price increases. This paper examines the characteristics of net food sellers and buyers in nine low-income countries. Although the largest share of poor households is found to be net food buyers, almost 50% of net food buyers are marginal net food buyers who would not be significantly affected by food price increases. Only three of the nine countries examined exhibited a substantial proportion of vulnerable households. The average incomes (as measured by expenditure) of net food buyers were found to be higher than net food sellers in eight of the nine countries examined. Thus, food price increases, ceteris paribus, would transfer income from generally higher income net food buyers to poorer net food sellers. The analysis also finds that the occupations and income sources of net sellers and buyers in rural areas are significantly different. In rural areas where food production is the main activity and where there are limited non-food activities, the incomes of net buyers might depend on the incomes and farming activities of net food sellers. These results suggest the need for reevaluation of the consensus on the impact of food prices on food needs. Further work on the regional differences, and more important, on the second order effects, are necessary to answer these questions more precisely. Only on the basis of further analysis can we start generating better policy responses.