This study analyzes the distribution (by income class and race ) of the costs and benefits of Boston's public schools. Costs are the taxes supporting public education; benefits are measured by the increase in expected lifetime earnings due to education , in an effort to . improve upon the previous use of expenditure as a benefit measure. results indicate that public education redistributes income from rich to poor and from non-whites to whites. In absolute terms, education benefits children from upper-income families more than poor children, and whites considerably more than non-whites. DUsLIG education has traditionally been thought of as having important distributional consequences, and as being a means of upward economic mobility. controversy over public ediication of the last decade or so has cast doubt on the latter assumption, and has questioned again whether public education lives up to its egalitarian ideals. While some educational questions must remain controversial, the distributional impact of education can be readily measured. major purpose of this paper, then, is to determine how the * This paper summarizes the results of my senior honors thesis (Harvard College, 1969). For the sake of brevity, tedious methodological details and many intermediate results have been omitted. They are contained in the original thesis, available upon request. I wish to thank Stephan Michelson for his continued assistance. comments of Professors Dale Jorgenson, Daniel Holland, and a referee of this journal were also helpful. * * Graduate student, Department of Economics, Harvard University; Center for Educational Policy Research, Harvard University. costs and benefits of a public school system are distributed by income and race. In so doing, it will be possible (1) to determine the biases in both the costs and the benefits separately, (2) to judge whether public schools benefit all children equally, and (3) to determine how education redistributes income among income groups and between races. In the absence of markets, costs and benefits may be difficult to measure. A secondary purpose of this paper is to develop a more meaningful index of educational benefit than has been previously used. public costs of public education are the taxes which support it, and the distribution of costs is the incidence of these taxes. Private costs are external to the redistributive effects of public education and will be ignored here, even though they may partly determine who benefits from public education. Measuring the benefits of education is considerably more difficult. Previous costbenefit studies of government-provided goods and services measured benefits by expenditure per person,1 largely out of 1E.g., Oswald Brownlee, Estimated Distribution of Minnesota Taxes and Public Expenditure Benefits , University of Minneapolis Studies in Economics and Business, No. 21 (Minneapolis: 1960); Irwin Gillespie, The Effect of Public Expenditure on the Distribution of Income, in Richard Musgrave (ed.) Essays in Fiscal Federalism (Washington, D.C.: Brookings Institution, 1965); Richard Musgrave and Darwin Daicoff, Who Pays the Michigan Taxes?, Michigan Tax Study Staff Papers (Lansing, Michigan: 1959), Ch. 4; Tax Foundation, Tax Burdens and Benefits of Government Expenditure by Income Class, 1961 and 1965, Research Publication No. 9 (New York: 1967).