The ineffectiveness of school inputs: a product of misspecification?

Two-thirds of education production function studies relating learning to school and parental inputs also include parental income. This confounds demand and production functions, since in demand functions income determines the school inputs used in the production function. In a comprehensive review of the literature, we show that with this misspecification significantly positive school input coefficients are 39% less common. Then, with Project TALENT student-level data from 1960 and pooled state data for 1987–1992, we examine the impact of including income with no other change in specification. This causes most school inputs to become less significant. Hausman tests suggest that in OLS regressions there is a correlation between the school input measures and the error term, perhaps due to the omission of a good measure of parental time with the student. This appears to bias the school input coefficients toward zero, but can be corrected with IV methods. [JEL I21, H52, H42]  1999 Elsevier Science Ltd. All rights reserved.