Nick Kristof offers a perfectly plausible defense of pre-K programs but can't resist the allure of phony statistics:
That would be compelling if every child eligible for one year of pre-K was otherwise sure to spend one year in juvenile detention.
It would be a financial break-even if 9% of the kids eligible for pre-K would otherwise average one year in juvenile detention.
So what are the numbers? We glean this from the Children's Defense Fund:
On any given day, approximately 81,000 children (263 of every 100,000 youths ages 10 through the state’s upper age of origi¬nal juvenile court jurisdiction in the general population) are held in a juvenile justice residential placement. Additionally, 7,560 children are held in adult jails and 1,790 in adult prisons.
If my math is holding up that sums to roughly 300 kids per 100,000 youths "of the general population", or 0.3%. if this pre-K expansion targeted some subset of youths, e.g, the bottom 20% by income, and if all the detainees came from that subset, then roughly 1.5% of the targeted sub-group would end up in detention.
That is a long way from 9% noted above (And we are assuming that none of the targeted kids eventually end up in detention anyway. Realistic?) As to the duration of the detention, well, it would need to average six years for Mr. Kristof's green-eyeshade argument to make sense.
And as to how helpful this comparison might be, the CDF also tells us this:
With nearly one-half (48 percent) of youths in the juvenile justice system functioning below the grade level appropriate for their age and 30 percent reporting a learning disability diagnosis, compared to 28 and five percent of students in the general population, respectively, children in the juvenile justice system have critical learning needs that must be addressed if they are to get on a more positive track forward.
The kids who end up in detention are likely to have learning issues that are unmitigated by pre-K programs, although early detection might be helpful.
This doesn't mean that expanded pre-K programs are a bad idea. But it does suggest that there is no educational equivalent of the Laffer Curve; pretending these programs can pay for themselves in reduced detention costs is unrealistic.