Rural development projects are aimed at reducing: (1) the large differences in income between urban and rural areas; and (2) the income inequality that persists within rural areas because of the large differences in access to agricultural assets (particularly land), to social services (such as health, education, and housing), and to productive services (such as credit, marketing, and processing facilities). The lack of these facilities and services and the disparities in income and wealth have led some Latin American governments to promote projects which supply the deficient facilities and services to poor rural areas. This emphasis on integrated rural development has also been promoted by the international lending agencies--the World Bank and the Inter-American Development Bank in particular. While rural and agricultural development projects have been initiated in a number of countries, this paper emphasizes examples in Latin America in which the Inter-American Development Bank has been involved. Specific references to these projects do not represent an official viewpoint, however, and opinions expressed herein are exclusively mine. In executing agricultural and rural development projects in Latin America, most problems have arisen because of inattention to: (1) beneficiary participation in designing and implementing the projects; (2) dealing with changes in the existing institutional structure and public organization; (3) finding criteria for the selection of beneficiary persons, groups, and regions; and (4) resolving the debate among those espousing a product oriented development strategy and those supporting a social development oriented strategy. Two major project categories may be differentiated--the agricultural development project and the rural development project. Mosher has suggested a practical classification scheme for these projects. The former emphasizes supplying missing production services (marketing, credit, extension, and research) while the latter emphasizes public social services (education, health, and sanitation). Rural development projects should increase the participation of the rural poor in the planning stage of the projects. This is emphasized in the rural development programmes of Mexico and Colombia, but is absent from many other agricultural investment programmes in Latin American countries. For example, it has been common for governments to design and execute agricultural projects (such as irrigation) without a careful consideration of the receptiveness and the social characteristics of the beneficiary groups. This has given rise to serious problems such as frequent dissatisfaction of beneficiaries, underutilized investment works, high project unit costs, production failures, low rates of return, and heavy government subsidization of the beneficiaries. See, for example, the survey of Colombian agrarian reform projects by Howard. The candidate beneficiaries of rural development projects are the inhabitants who historically have been excluded-the small subsistence farmer, the landless worker, and the population in the small urban centre in predominantly rural regions. To the extent that the rural poor are concentrated in well defined regions, rural development projects may benefit this group, but if the rural poor are spread out through all rural and small urban areas, there is a reduced prospect that specific regional rural development projects will have the desired effects. It is clear that middle income rural inhabitants may benefit from traditional public programs because of the access they have to the programs. For example, school nutrition programs often do not reach the lowest income groups because the children of the poor attend school with much less frequency than other income classes (Offedal and Levinson).