James Heckman writes,

Although much public policy discussion focuses on the failings of schools, a major finding from the research literature is that schools and school quality contribute little to the emergence of test-score gaps among children. By the second grade, gaps in ranks of test scores across socioeconomic groups are stable, suggesting that later schooling has little effect in reducing or widening the gaps that appear before students enter school. In work with Pedro Carneiro, I performed a cost-benefit analysis of classroom-size reduction on adult earnings. While smaller classes raise the adult earnings of students, the earnings gains do not offset the costs of hiring additional teachers. The best way to improve schools is to improve the students sent to them. A substantial benefit of early interventions is improvement of the performance of disadvantaged children in schools.

…Important operational details of investment programs for disadvantaged children remain to be determined. Children from advantaged environments, by and large, receive substantial early investment, while children from disadvantaged environments more often do not. There is little basis for providing universal programs at zero cost, although some advocate such a policy. While there is a strong case for public support for funding interventions in the early childhood of disadvantaged children, there is no reason for the interventions to be conducted in public centers. Vouchers that can be used in privately run programs would promote competition and efficiency in the provision of early enrichment programs.

To summarize, Heckman argues that in order to be cost-effective, efforts to improve cognitive ability among the poor should

1. Focus on very young children
2. Target poor children, rather than all children
3. Allow the private sector competition to operate

Heckman does not discuss “No Child Left Behind,” but I would note that this expensive program violates all three of those ideas.