It is odd that people don't seem to recognize the flaw of only looking at benefits when doing a cost/benefit analysis.

From Preschool Really Is the Magic Bullet by Peter D. Salins.

An admirably succinct statement of the problem.

America's most intractable educational failure is the woeful school performance of its most disadvantaged children. The educationally "left behind" include not only racial minorities but millions of lower-income white children, especially boys. No other accomplishment would so dramatically transform American society as significantly reducing the school-achievement disparities between these children and the rest. In pursuit of such a breakthrough, political leaders and educators have, for over 50 years, experimented with an extensive repertoire of reforms: racial integration, increased spending, smaller class sizes, and most recently charter schools and teacher merit pay. The problem is that, by and large, the reforms haven't succeeded -- at least with respect to their stated goal of enabling the "left behind" to catch up with the rest.

All that is left out is any reference to the cost. Not only have we tried but we have spent a lot and stirred up a lot of civil unrest in pursuit of the goals. High costs and low results tend to discourage future efforts.

Most people would go along with the more general root cause statement.

Countless studies have put their finger on what causes disadvantaged children to do so badly in school. The families in which they are raised -- often headed by single mothers -- are unable to give them the cognitive stimulation that is both essential for early-childhood development and common in the homes of their middle-class peers. These early-age disparities are then magnified throughout the later years of formal schooling.

What can be done? I agree regarding Hirsch. It is amazing we can invest so much time, money and energy and so avoid one of the most important practitioners.

Given this diagnosis, what is the remedy? E.D. Hirsch, for decades an insufficiently heeded voice in the education-reform wilderness, argues that differences in cultural literacy, more than any other single factor, separate socioeconomically disadvantaged children from the rest -- and that for them to catch up, this deficit must be erased before they enter first grade by enrolling them in preschool, with continued reinforcement of cultural literacy in the elementary-school grades.

But it has to be the right kind of preschool: where children spend enough time each week, in small classes, taught by trained, well-paid professionals that give them a foundation in the kinds of cognitive skills and social behaviors on which later school success depends.

"Aye, there's the rub." The right kind of preschool. The expensive kind. So how are we doing? Well, we've got the expensive part down pat.

Most existing preschool programs, especially those attended by poor children, fail to meet these criteria. The most conspicuous case in point is Head Start, which enrolls over 1 million poor, primarily minority children in centers with limited weekly hours and minimal educational content, overseen by poorly trained and poorly compensated staff. Numerous evaluations, including the most recent one (conducted by the Brookings Institution in 2010), document that Head Start students make minimal academic gains while they are in the program, and that whatever little benefit there is vanishes by first grade.

I'd like to know more about the French experience. Of course France has a population density four times that of the US which is problematic in terms of emulation.

Further validation of expanding preschool access lies outside the United States. Moved by the enormous disparities in K-12 academic achievement among French schoolchildren, largely correlated -- as in the U.S. -- with social class and ethnicity (children of France's large North African immigrant community tend to perform less well than their peers), France in the 1980s launched its universal, public, and free preschool system, open to (but not compulsory for) all children age two to five. Strongly influenced by Hirsch's work, the program's French designers made sure their preschools had rigorous cultural-literacy content, well-trained teachers, reasonable staff ratios, and good facilities. All evaluations since then show that this initiative has sharply raised the academic achievement and high-school graduation rates of French schoolchildren -- across the board, but most notably among the disadvantaged.

The rest of the article explores how to bring about expensive universal pre-K education.

A nice clear articulation of the benefits and the size of the prize. It would be nice to achieve this. Costs are mostly left out of the analysis. Just how much would it cost to achieve this desirable outcome. How confident, after the decades-long Head Start failure, could we be in achieving the expensive results? These critical questions are left unaddressed. Any proposition that only looks at the benefits and not the costs is likely to come out positively in terms of a cost/benefit analysis. It is odd that people don't seem to recognize the flaw of only looking at benefits when doing a cost/benefit analysis.

The even larger issue unaddressed in the article is the old economic issue of incentives. If you provide free universal quality pre-K education programs that are demonstrably effective, what are the incentives and disincentives that you are creating for individuals, families, and groups and what are the long term implications of those incentives? That is by far and away the most critical part of the conversation and it is not addressed at all - likely because the implications are troubling.