The Food for Education (FFE) program was introduced to Bangladesh in 1993 and hasbeen operating for more than 8 years. This paper evaluates the effect of this program onschool participation and duration of schooling using household sample survey data collectedin 2000. Various evaluation methodologies are employed. We found that the program issuccessful in that the participating children on average have 20 to 30 per cent higher schoolparticipation rates, relative to their counterfactuals who did not participate in the program.Conditionalonschoolparticipation,participantsalsostayatschool0.5ofayearto2yearslonger than their counterfactuals. Using estimated earnings functions from the BangladeshHousehold Income and Expenditure survey, these combined education effects of the FFEprogram would represent an increase in lifetime earnings of between 7 and 16 per cent if theparticipant is going to work in the rural sector, and 13 to 25 per cent if in the urban sector.These increases would bring large numbers of households above the poverty line.