Returns to schooling and ability in a British cohort: control functions with multiple educational outcomes

IZA Logo

IZA Seminar

Place: Schaumburg-Lippe-Str. 9, 53113 Bonn

Date: 15.07.2003, 12:15 - 13:30


Presentation by 

Fernando Galindo-Rueda (Office of National Statistics)


This paper extends Garen's control function method to estimate the returns to educational outcomes when two educational outcomes, cognitive development and schooling, are jointly considered. Estimates based on data from men in a British cohort born in 1970 find evidence of self selection based on indiosyncratic gains. In particular, cognitive development -not schooling- appears to be the human capital variable that needs to be controlled for endogeneity: Children with unusually large ability for their observed characteristics tend to have lower baseline earnings but higher returns to schooling. The results also suggest that there is a significant divergence between expected and actual marginal benefits of human capital. Most of the variation in the rate of return to ability and schooling that is due to observable attributes is driven by differences in the marginal costs. These findings justify further research on the returns to different types of educational outcomes, their timing and the intra-household decision making process, as opposed to the more standard, narrow focus on schooling.

For more information, please contact