Investment in human capital has long stirred an interest geared towards investigatingrnwhether or not it is profitable. Among such human capital variables, education has beenrnrecognized as the most important investment in human capital for a long time. The mainrnobjective of this study is to examine and estimate the returns to schooling in connectionrnwith self-employment in selected urban centers in Ethiopia. Using a Mincerian earningrnfunction and correcting for sample selection bias, the paper shows that (a) the privaternaverage returns to a one-year increase in schooling is about 5.6%. This figure is smallerrnthan the global average (as estimate by Psacharopolous, 1993) by 5.4% and even by 9.4% from the average of some developing countries. This could partially be explained by the overall low quality of education in Ethiopia; (b) the Mincerian rates of return to primary, secondary and higher education are 6.9%, 7.9% and 17.2%, respectively. As is apparent from these coefficients, the highest return is derived from higher education and the lowest return for primary education. This indicates the existence of substantial productivity differential between college graduates and those who comes from the lower level of education; (c) educating females is marginally more profitable than educating males. This implies that expanding the provision of school places to cover women is not only equitable but also economically efficient as well; (d) the coefficient of start-up capital in the earning equation is positive and significant. It indicates that those household heads who put higher start-up capital earn more than others. The coefficient of dummy for service sector is also positive and significant. This implies that those household heads who run their business in the service sector earn more than others