This paper examines the argument that education systems in development countries should be financed more directly by private households. It finds that cost-recovery policies are likely to be harmful to efficiency and equity if significant resources were to be generated by these means. User fees at primary and secondary levels would increase enrollments only if the revenues so gained were spent on the provision of new school places, if there were excess demand for schooling, and if the negative enrollment response among the poor did not exceed the positive response from those who were willing and able to pay. Even so, many bright, poor students may withdraw from school. A scholarship policy would be needed, which may undermine the revenue-raising objectives of user charges. The case for user charges at the tertiary level is stronger; however, tertiary fees also offend equity principles and would make loans and scholarships necessary. The challenge of raising more public resources for education should mainly be addressed through increasing levels of direct and indirect taxation, in ways that move the balance of tax incidence in a more progressive direction. In addition, payroll taxes for graduates and other highly skilled workers trained at public expense provide a more effective and equitable means. In conclusion, other policies are available to improve equity and efficiency in education that are not substantially included in the neoliberal case. Two tables, one figure, and a list of other Economic Policy Series (EPS) papers that are available are included. Contains 78 references. (Author/LMI)