We derive a tractable nonlinear earnings function which we estimate separately for each individual in the NLSY79 data. These estimates yield five important parameters for each individual: three ability measures (two representing the ability to learn and one the ability to earn), a rate of skill...

This paper relaxes the assumption of homogeneous rates of return to schooling by employing nonparametric kernel regression. This approach allows us to examine the differences in rates of return to education both across and within groups. Similar to previous studies we find that on average blacks...

This paper proposes a new strategy to identify causal effects. Instead of finding a conventional instrumental variable correlated with the treatment but not with the confounding effects, we propose an approach which employs an instrument correlated with the confounders, but which itself is not...