The human capital and growth accounting approaches to measuring the benefits of education both have serious weaknesses. Like other goods and services, educational services have effects on the economic well-being of individuals and families. Because the economic well-being effects of education include private marketed and non-marketed impacts as well as external or public impacts, estimates of the aggregate value of educational services must encompass all of these. To obtain a true understanding of the effects of education on productivity, various non-marketed, private returns must be considered, including the health and fertility effects of education, effects on the value of home time of mothers, effects on criminal behavior, and effects deriving from the impact of education on the earnings distribution. A review of the existing literature on the effects of education in these areas supports the hypothesis that the provision of education services is likely to have a larger impact on economic well-being than is estimated by studies based upon the direct returns or growth accounting frameworks. Therefore, researchers must develop means to measure the full willingness to pay for educational services. (This study is one in a series on the relationship between education and productivity.) (MN)