Another way to spend money on education

The educational achievement of a child involves many inputs: parents, household situations, neighborhoods, nourishment, and of course schools. As a result of the education reform movement, all non-school inputs are currently sidelined. If a child performs poorly, we are made to believe that school-side problems are to blame. It must be failing teachers, failing school models, and failing school cultures. Therefore we must fire teachers and administrators, and close down schools altogether to make room for better charter schools.

For the most part, I have focused on the economic issues involved in education on this blog. Mountains of social science research show that living in poverty — as more than 1 in 5 American children do — has extremely negative effects on educational achievement. Given that, I am generally of the opinion that eliminating or reducing poverty is probably one of the best education policies to pursue. This would involve transferring money to poor people until they are not poor anymore.

One of the other proposed ways to improve education — often favored by liberals — is to spend more money on it. That is, put more money into the schools. How effectively that improves education is heavily disputed. It seems obvious that at an initial point, more spending is sure to improve educational achievement. A school spending $1/pupil will surely perform worse than a school spending $5000/pupil, all else equal. It also seems obvious that at a certain point, a marginal increase in spending is not likely to have that much of an impact. It is hard to imagine a school spending $1 million/pupil will perform much better than one spending $995,000/pupil. There is a lot of room in between those two extremes, and that tends to be where the debate is fought.

In general then, it would seem that school spending has diminishing marginal utility. Every additional dollar you spend on schooling probably delivers less educational improvement than the dollar before it. When you combine this realization with the fact that giving dollars to poor people to make them less poor should also improve educational achievement, you reach an interesting conclusion. There is almost certainly some school funding level at which it would be more beneficial to education to reduce school spending and give the savings directly to poor kids’ families. That is, there must be some funding point at which giving an additional dollar to a school would generate less marginal educational achievement than giving that same dollar to the poor.

I do not know where that point is, but it may have already been passed in some school districts. For instance, the DC school system, which is notoriously full of poor kids and notoriously low-performing, spends more per pupil than any other state in the country. Its per-pupil current spending for students is $18,667 per year. Its total expenditures per pupil (which includes things like capital expenditures as well) is over $29,000 per year. I cannot say for sure, but it’s conceivable that cutting down per-pupil spending by $1,000/year and cutting a check for that amount to each parent of a child attending a DC public school would actually generate more educational achievement.