This study investigates the trends, patterns and prospects of youth involvement in agriculture in Ethiopia, by gender. It also analyses the determinants of youth labor supply in agriculture using household and youth sample survey data collected during 2010/11 and 2014/15 main agricultural seasons in Oromia, one of the designated high agricultural potential area of Ethiopia. Labor supply is measured as the total annual working days (in adult equivalent) of male and female youth members of the household allocated to on-farm and off-farm work. Based on this data the marginal products (shadow wages) of youth workers of each gender and net income (shadow income) are estimated, using a structural time-allocation models. Then the estimated shadow wages and shadow income are used as regressors in a structural model of youth labor supply. The results indicate that trends and patterns of youth involvement in agriculture vary across gender and work locations, and so do their marginal products. Whilst the on-farm participation of youth is declining across time irrespective of gender, the participation in off-farm agricultural activities is increasing for both. There is statistically no significant decrease in the total agricultural labor supply of both male and female youths. Further, we find that the effect of own shadow wage on labor supply is positive for male youth members, suggesting an upward sloping labor supply. However, the effect of own marginal product of female youth labor is negative, suggesting that female youth agricultural labor supply is backward bending. The findings challenge the presumption that youth are abandoning agriculture, at least in the survey areas. Policy implications of the results would be that changes in economic incentives such as shadow wages and shadow income matter for youth involvement in agriculture and off-farm agricultural employment opportunities could help to reduce youth underemployment.