In this entry, I will recapitulate part of Heckman and Carneiro’s “Analysis of Specific Policies” section from their “Human Capital Policy” essay. In particular, I will address labor force improvement through improved pre-college schooling. The general consensus is that the return to schooling is over 10%, making it worthwhile to examine the means by which schools can be improved or additional schooling can be promoted in order to close the income gap.

To see how schooling can be improved, Heckman and Carneiro analyze the relationship between measurable student achievement metrics and lifetime incomes. At the high school level, it has been shown that in the early years, additional schooling can raise test scores for everyone yet fails to close the gap between the disadvantaged and the rest of the students. Moreover, the ability of additional schooling to raise scores diminishes as students progress through school. Across non-white male demographic groups, higher test scores correlate with higher wages but the improvement does not close income gaps. To this effect, it is arguably noteworthy to raise school quality, particularly for disadvantaged population segments. This is where things become tricky. The most common proposals to improve school quality, propositions supported by Professor Kruegman, are teacher-to-student ratio reductions, prolonged schooling, greater teacher salaries, and per-student expenditures. Heckman’s answer to all of the above is don’t bother. First, the evidence is greatly mixed on these interventions’ ability to raise test scores. Furthermore, even if there are slight positive effects on test scores, the test scores’ relationship with income is weak. The same is true for other interventions, as more recent studies increasingly suggest weak effects for all of the above interventions. To demonstrate more constructively, the authors take the most favorable estimates and lowest costs and model the net costs and benefits of these interventions. Even after incorporating additional beneficial parameters, such as increases in workforce productivity and taxes, the benefits are all negative. It is only with the highest, and most unrealistic, parameters that the model yields positive returns. Based upon this, they make two conclusions. First, America already overspends on their students, making additional investment unwise. Increased spending of the above kind promote will neither college-entrance nor college-readiness for minority students. This does not mean that schools cannot use additional funding but it does imply that lifetime earnings will benefit little and the benefits are likely to be low. One important omission is the efficacy of increased teacher salaries, which Heckman and Carneiro do not discuss, thereby making it difficult to understand whether teacher salaries are ineffective in raising students’ lifetime incomes.

Another proposition of improving schooling is not through investment into schools but rather, allowing the parents of pupils to choose better schools, thereby indirectly approaching the issue. The current consensus, given the dismal ranking of US students in international comparisons, that US public schools are failing the students. Heckman and Carneiro claim that this occurs because schools have very little incentive to maximize output. Under the current structure, schools hold local monopolies; therefore, for most principals and teachers there are few incentives to improve academic performance. Moreover, these “educational bureaucracies” are very opaque, therefore hinder accountability. School choice addresses many of the above issues. It increases parental involvement in the child’s schooling and promotes competition between schools, thus forcing them to improve the quality of their teaching in order to retain students. Allegedly, greater presence of parents improves transparency and accountability while competition introduces incentives for teachers to maximize output. Although feasible, Heckman mentions that “school choice systems would be most beneficial to those already able to exercise choice in the current system, the richer families”(p. 160) so it does little to promote equity. Furthermore, there is very little empirical evidence on these initiatives so most current insights stem from theoretical work that has not been tested against reality. Nevertheless, the few studies of voucher programs suggest that scholastic abilities do improve without higher costs. Overall, the conclusion is that competition has a greater ability in improving the quality of the labor force with a better cost-benefit tradeoff than an influx of money to schools. In the next entry, given the limitations of the above interventions and their respective empirical evidence, Heckman and Carneiro discuss interventions in earlier stages of life.



Heckman, J., Krueger, A. (2005). Human Capital Policy. Inequality in America: What Role for Human Capital Policies?  148-162