This paper makes several points based on a review of household survey evidence from Africa, Asia and Latin America. (i) In contrast to conventional wisdom, the evidence is very mixed as to the effect of non-farm employment on rural income inequality. The non-farm employment and microenterprise programmes now in vogue will not necessarily resolve rural income inequality problems and attendant social tensions nor automatically benefit the poor. (ii) Policymakers should be worried by substantial evidence of poor people's inability to overcome important entry barriers to many non-farm activities. (iii) The main determinants of unequal access to non-farm activities are the distribution of capacity to make investments in non-farm assets and the relative scarcity of low capital entry barrier activities. Therefore, it is crucial for public investments and policy to favour an increase in the access of the poor to assets that allow them to overcome non-farm employment entry barriers, (iv) It would be an error to assume that one can address asset-poverty and inequality in the non-farm sector without addressing farm-side problems and vice versa.