A study by the Economic Policy Institute, using a different classification system, found that in nine representative school districts only 10 percent of spending went to actual administration and 16 percent to operations and maintenance. The rest went to programs for children. A nationwide survey by the Educational Research Service found that just over 9 percent of spending went to administration. That means school administration, central offices, school boards and central business services. How does NCES differ?

The NCES classifies “in the classroom” as instructional aides, teachers, textbooks and some athletics, as well as tuition paid for high-cost special needs students. The rest of the money is classified as support and administration. This includes professional development for teachers, curriculum, instructional leadership and school libraries. It also includes counselors, school nurses and truant officers. All of these, along with food services, maintenance, transportation, and actual administration such as payroll and data processing are counted in the category “administration.”

An analysis from Standard and Poor’s indicates that meeting the 65 percent target is not related to better student achievement. This seems counterintuitive; dollars to the classroom should help. But this only works when spending directly affects services that really are instructional, not just those defined as such by NCES.

Research shows the role that high-quality professional development can play in lifting test scores, yet professional development is considered an administrative expense. For too many poor children the school library is the main source of printed material, yet the library also is treated as an administrative expense.

A lot of support services have an impact on student learning. Guidance counselors can play a role by improving test preparation skills and encouraging persistence in schooling. Research even points to the importance of adequately maintained lighting and heating systems for classrooms. Clichés come to mind: You can’t expect to succeed by robbing Peter to pay Paul. And when something sounds too good to be true, it is.

The supporters of this proposal call themselves “First Class Education.” News that this proposal won’t help students is unlikely to deter them from pursuing it. An internal memo obtained by the Austin American-Statesman newspaper shows the purpose of this project is not to help kids. Instead the 65 percent rule is intended to create divisions within education unions; divert resources from the goals of the “education establishment”; and send a message to soccer moms that the Republican Party supports a more efficient public education, making Republicans seem more trustworthy when they support vouchers and charters.

You can see this strategy in action in Florida. For years, Gov. Jeb Bush has fought to prevent full implementation of the state’s class size reduction program. This year his supporters have crafted a constitutional amendment that would essentially replace full implementation of class size reduction with the 65 percent mandate. Voters choosing to support the 65 percent “solution” would be simultaneously voting for larger classes.

 In Louisiana, before Katrina struck, this proposal was used to fight off a cigarette tax increase that would have funded higher teacher pay. Similar battles can be expected in Arizona, Colorado, Georgia, Indiana, Missouri, Ohio and Washington, at the least.